ISSN: 2346-3775 Vol. 9 No. 3 September 2021 Published by: Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia In association with: Chinese Academy of Tropical Agricultural Sciences (CATAS), Haikou, Hainan, P.R. China International Center for Tropical Agriculture (CIAT) retains copyright of articles with the work simultaneously licensed under the Creative Commons Attribution 4.0 International License (to view a copy of this license, visit creativecommons.org/licenses/by/4.0/). Accordingly, users/readers are free to share (to copy, distribute and transmit) and to remix (to adapt) the work under the condition of giving the proper attribution. i Editors Jean Hanson, Danilo Pezo, International Livestock Research Institute (ILRI), Tropical Agriculture Research and Higher Education Ethiopia Center (CATIE), Costa Rica Management Committee Robert J. Clements, Danilo Pezo, Agricultural Consultant, Tropical Agriculture Research and Higher Education Australia Center (CATIE) Lui Guodao, Rainer Schultze-Kraft, Chinese Academy of Tropical Agricultural Sciences The Alliance of Bioversity International and CIAT, (CATAS), P.R. China Colombia Jean Hanson, Cacilda B. do Valle, International Livestock Research Institute (ILRI), Empresa Brasileira de Pesquisa Agropecuária (Embrapa), Ethiopia Brazil Asamoah Larbi, Lyle Winks, Agricultural Consultant, Former editor of “Tropical Grasslands”, Ghana Australia Michael Peters, The Alliance of Bioversity International and CIAT, Kenya Editorial Board Caterina Batello, Michael David Hare, Food and Agriculture Organization of the United Nations Ubon Ratchathani University, (FAO), Italy Thailand Robert J. Clements, Huan Hengfu, Agricultural Consultant, Chinese Academy of Tropical Agricultural Sciences Australia (CATAS), P.R. China Albrecht Glatzle, Mario Herrero, Iniciativa para la Investigación y Transferencia de Commonwealth Scientific and Industrial Research Tecnología Agraria Sostenible (INTTAS), Paraguay Organisation (CSIRO), Australia Orlando Guenni, Masahiko Hirata, Universidad Central de Venezuela (UCV), University of Miyazaki, Venezuela Japan Jean Hanson, Peter Horne, International Livestock Research Institute (ILRI), Australian Centre for International Agricultural Research Ethiopia (ACIAR), Australia ii Johann Huguenin, T. Reginald Preston, Centre de Coopération Internationale en Recherche University of Tropical Agriculture Foundation (UTA), Agronomique pour le Développement (CIRAD), France Colombia Muhammad Ibrahim, Kenneth Quesenberry, Tropical Agriculture Research and Higher Education University of Florida, Center (CATIE) USA Asamoah Larbi, H. Max Shelton, Agricultural Consultant, The University of Queensland, Ghana Australia Carlos E. Lascano, Werner Stür, Universidad Nacional de Colombia - Sede Bogotá, Australian Centre for International Agricultural Research Colombia (ACIAR), Australia Robert Paterson, Cacilda B. do Valle, Agricultural Consultant, Empresa Brasileira de Pesquisa Agropecuária (Embrapa), Spain Brazil Bruce Pengelly, Agricultural Consultant, Australia Principal Contacts Michael Peters The Alliance of Bioversity International and CIAT Kenya Phone: +254 709 134 130 Email: CIAT-TGFT-Journal@cgiar.org Technical Support José Luis Urrea Benítez The Alliance of Bioversity International and CIAT Colombia Phone: +57 2 4450100 Ext. 3354 Email: CIAT-TGFT-Journal@cgiar.org iii Table of Contents Preamble 50 years publishing on tropical forages - from Tropical Grasslands and Pasturas Tropicales to Tropical Grasslands- Forrajes Tropicales vi Michael Peters, Robert Clements, José Luis Urrea-Benitez, Liu Guodao Research Papers Canopy characteristics of ‘Mavuno’ hybrid brachiariagrass and ‘Marandu’ palisadegrass harvested at different harvest intensities 249 Luan F. Rodrigues, Joao M.B. Vendramini, Antonio C. dos Santos, Jose C.B. Dubeux Jr, Fabricia R.C. Miotto, Luciano F. Sousa, Nayara M. Alencar Effects of stubble height and season of the year on morphogenetic, structural and quantitative traits of Tanzania grass 256 Nauara Moura Lage Filho, Aline da Rosa Lopes, Aníbal Coutinho do Rêgo, Felipe Nogueira Domingues, Cristian Faturi, Thiago Carvalho da Silva, Ebson Pereira Cândido, Wilton Ladeira da Silva Effect of sowing rate and date on establishment and growth of Trichloris crinita, a native American pasture grass from arid environments, in the Arid Chaco of Argentina 268 Deolindo L.E. Domínguez, Pedro R. Namur, Pablo F. Cavagnaro Biomass production and nutritional properties of promising genotypes of Tithonia diversifolia (Hemsl.) A. Gray under different environments 280 Julián Esteban Rivera, Tomás E. Ruíz, Julian Chará, Juan Florencio Gómez-Leyva, Rolando Barahona Evaluation of ten perennial forage grasses for biomass and nutritional quality 292 Mulisa Faji, Gezahagn Kebede, Fekede Feyissa, Kedir Mohammed, Muluneh Minta, Solomon Mengistu, Aschelew Tsegahun Is organic fertilizer application a viable alternative to synthetic fertilizer for Piatã grass? 300 Sírio Douglas da Silva dos Reis, Marco Antonio Previdelli Orrico Junior, Michely Tomazi, Stéfane Souza Cunha, Ana Carolina Amorim Orrico, Joyce Pereira Alves, Edgar Salvador Jara Galeano Evaluation of corn-soybean inter-cropping systems in southwestern Japan 307 Ahmad Seyar Azizi, Ikuo Kobayashi, Jonathan Chuuka, Genki Ishigaki Quality properties of sunn hemp (Crotalaria juncea L.) and maize (Zea mays L.) silages 315 Gülcan Demi̇roğlu Topçu, Şükrü Sezgi̇ Özkan Criterios de uso y conservación de árboles en potreros basados en el conocimiento local de los ganaderos en una zona de bosque seco tropical en Colombia 321 Nelson Pérez-Almario, Eliana Lizeth Medina-Rios, Jairo Mora-Delgado, Dagoberto Criollo-Cruz, Julian Roberto Mejía Physiological responses of Bajra-Napier hybrids and a tri-specific hybrid to salinity stress 337 Seva Nayak Dheeravathu, Kajal Singh, Pramod W. Ramteke, Reetu, Nilamani Dikshit, Mahendra Prasad, Dibyendu Deb, Thulasi Bai Vadithe iv Genetic Resources Communications Clearing confusion in Stylosanthes taxonomy. 3. S. hamata sensu stricto vs. S. hamata sensu lato 348 Bruce G. Cook, Rainer Schultze-Kraft Genetic diversity and population structure of Heteropogon contortus L. germplasm collected from diverse agro-climatic regions in India and development of a core germplasm set 359 Ajoy Kumar Roy, Devendra Ram Malaviya, Pankaj Kaushal, Sanat Kumar Mahanta, Rupali Tewari, Roopali Chauhan, Amaresh Chandra Short Communications What drives the adoption of fodder innovation(s) in a smallholder dairy production system? Evidence from a cross- sectional study of dairy farmers in India 371 D. Thirunavukkarasu, N. Narmatha, S. Alagudurai Uso de sensores remotos en la determinación del forraje disponible de Urochloa humidicola cv. Llanero bajo pastoreo en la Altillanura colombiana 376 Raúl Alejandro Díaz Giraldo, Mauricio Álvarez de León, Otoniel Pérez López The effects of increasing concentrations of Trichanthera gigantea leaves in pellets on the nutritive value and short-term intake of diets of grass plus pellets offered to lambs reared under tropical conditions in the Caribbean 383 H.A. Jack, L.M. Cranston, J.L. Burke, M. Knights, P.C.H. Morel Effects of plant spacing and fertilizer level on forage yield and chemical composition of hybrid Urochloa cv. Mulato II grass during the first 150 days of growth under irrigation supplementation, in Chagni Ranch, Awi Zone, Ethiopia 391 Wondimagegn Tadesse, Berhanu Alemu, Mesganaw Addis v Preamble 50 years publishing on tropical forages - from Tropical Grasslands and Pasturas Tropicales to Tropical Grasslands-Forrajes Tropicales MICHAEL PETERS1, ROBERT CLEMENTS2, JOSÉ LUIS URREA-BENITEZ3 AND LIU GUODAO4 1International Center for Tropical Agriculture (CIAT), Nairobi, Kenya. 2Consultant. 3International Center for Tropical Agriculture (CIAT), Cali, Colombia. 4Chinese Academy of Tropical Agricultural Sciences (CATAS), Hainan, PR China. Creation of Tropical Grasslands-Forrajes Tropicales sole sponsor and co-publisher. An official foundation ceremony was held at CATAS on 13 December 2012 in The online journal Tropical Grasslands-Forrajes Danzhou, Hainan, PR China. CATAS has continuously Tropicales (ISSN official abbreviation Trop. Grassl.- afforded grant support since 2013 and the journal is Forrajes Trop.) was created in 2012 as a successor to currently published in association with CATAS. This the former journals, Tropical Grasslands, published allows publication without any article submission during 1967‒2010 by the Tropical Grassland Society and processing charges. Dr. Liu Guodao, CATAS of Australia Inc., and Pasturas Tropicales, published Vice president serves as co-chair of the Management during 1979‒2007 by the International Center for Committee, and and Dr. Huan Hengfu, CATAS Senior Tropical Agriculture (CIAT). Tropical Grasslands was researcher, serve as member of the Journal Editorial an important vehicle for disseminating the work of Board. authors from all countries on research and development The inaugural issue of the journal was presented in the assessment, management and utilization of at the 22nd International Grassland Congress, 15‒19 pastures and fodder crops in tropical agriculture, as well September 2013 in Sydney, Australia (Schultze-Kraft as reviews in these topics (Pulsford 2010). During the et al. 2013), and was the result of a co-publication 44 years of production, 160 issues were produced, 18 agreement with the Organizing Committee. It contained of which were proceedings of conferences on tropical those papers relevant to tropical pastures and forages pastures (Tropical Grasslands 2010). On the other hand, that were presented at the Congress, including keynote Pasturas Tropicales, directed at Spanish speakers and papers, presented papers and poster papers. focusing initially for Latin American audiences was first published as Pastos Tropicales, being essentially a Characteristics of the journal newsletter. Scientific publication became increasingly important for the journal and from 1986, its 8th volume, The main features of the journal are that it is international, the name was changed to Pasturas Tropicales, which published online only, open access (no charges for was published as an imprint journal until 2007. subscription or publication fees), bilingual (English The name of the new journal Tropical Grasslands- and Spanish), peer reviewed and guided by an Editorial Forrajes Tropicales indicates both its bilingualism Board composed of the world's leading tropical forage (English or Spanish, with abstracts in both languages) scientists. Further information on the journal is available and the desire to continue the tradition of both former at its website (www.tropicalgrasslands.info). All journals (Schultze-Kraft et al. 2013). issues of the former journals Tropical Grasslands and The initial steps to establish the journal were made Pasturas Tropicales can also be accessed there. Tropical possible through a seed money grant received from an Grasslands-Forrajes Tropicales follows the publication anonymous donor in memory of Dr. José M. Toledo, series of the Australian CSIRO Tropical Agriculture leader of the former CIAT Tropical Pastures Program Genetic Resources Communication (ISSN 0159-6071) in the 1980s. During the 5-year period 2013‒2017, published during 1980‒2000. By kind permission of the journal was sponsored by grants from the Chinese CSIRO, the issues that deal with (sub)tropical forages Academy of Tropical Agricultural Sciences (CATAS) can be accessed on the website as well. and the Australian Centre for International Agricultural The journal is published every four months by CIAT Research (ACIAR). Since 2019, CATAS has been the in Cali, Colombia, with three issues: the first published vi on January 30th, the second published on May 31st and (5-yr IF). The journal is also indexed in Scopus, and the third published on September 30th. the CiteScore for 2020 was 1.50, making considerable progress since 2017 (0.90), 2018 (1.00) and 2019 (1.30). The editors Lyle Winks and Rainer Schultze-Kraft CiteScore is the way Scopus measures the impact of its indexed journals. The SCImago Journal & Country Rank Since its inception Tropical Grasslands-Forrajes is a portal that includes the journals and country scientific Tropicales has counted on two dedicated editors, Rainer indicators developed from the information contained in Schultze-Kraft and Lyle Winks. Without their dedication, the Scopus® database (Elsevier B.V.). These indicators ideas and work ethics for almost 10 years the creation can be used to assess and analyze scientific domains. and continuity of the journal, with increasing standards For 2020, the Journal had a score of 0.26, and in 2019 and distribution would not have been possible. was ranked in the second highest value amongst the set Prof. Dr. Rainer Schultze-Kraft is a forage scientist of agronomy journals ranked by Scimago Journal Rank with 50 years of experience. Receiving his PhD on (Q2). forages from the University of Giessen, Germany in the Since 2018, the Journal is indexed in SciELO, a 70s, he moved on to CIAT working on forages until the bibliographic database, digital library, and cooperative late 80s. After that he received a call from the University electronic publishing model of open access journals. of Hohenheim, Germany, to continue his academic work SciELO was created to meet the scientific communi- on tropical forages there until his retirement in the early cation needs of developing countries (with a focus on 2000s, when he became a CIAT Emeritus Scientist. His Latin America) and provides an efficient way to increase work on tropical forages - with a focus on tropical forage visibility and access to scientific literature. In 2019, legumes - has received worldwide scientific recognition the Journal was included in the National Bibliographic across the tropics in Asia, Africa and the Americas. He Index (Publindex) of the Colombian Ministry of Science acted as Managing and Spanish Editor of the journal. Technology and Innovation, based on indicators and Lyle Winks is an agricultural scientist, who specialised standards of scientific quality at the international level. in ruminant nutrition and worked as a researcher in the In the 2020 classification the journal was ranked as “A2”, Queensland Department of Primary Industries (QDPI) in accordance with the fulfillment of the internationally for almost 30 years. He published numerous papers on recognized evaluation criteria for scientific publications pasture-based beef production in north Queensland, and related to the processes of editorial management, became Director of the Beef Cattle Husbandry Branch evaluation, visibility and impact. In 2019, the Journal of QDPI (1984-92), supervising its research throughout was also included in the Regional Online Information the State. From 1992-2010 he was the Editor of Tropical System for Scientific Journals of Latin America, the Grasslands, publishing 18 volumes (74 issues) during Caribbean, Spain and Portugal (Latindex), a network of that time. He currently runs a beef production enterprise institutions that work jointly to gather and disseminate in SE Queensland. He acted as English Editor of the information of serial scientific publications produced in journal. Latin America. SHERPA RoMEO (S/R) is an online resource that Current status of the journal aggregates and analyzes publisher open access policies from around the world, and uses different colors to help By June 2021, Tropical Grasslands-Forrajes Tropicales highlight publisher's archiving policies. In 2018, the had published 334 papers. These included 177 in special journal reached the classification as a RoMEO Green issues (115 contributions to the International Grassland Journal, the highest open access category. Updated Congress 2013 and 62 contributions to the International annually, the Information Matrix for Journal Analysis Leucaena Conference 2018) and 157 in regular issues. (MIAR in Spanish) database gathers key information The journal is indexed in all major abstract and citation for the identification and analysis of journals. The databases of peer-reviewed literature. The Journal is system creates a correspondence matrix between the indexed in the core collection of Science Citation Index journals and the databases and repositories that index or Expanded and two additional indexes of Clarivate Web include them. The ICDS (Composite Index of Secondary of Science, which provides the best-known impact factor Dissemination in Spanish) is an indicator that shows the indicator (formerly: ISI Journal Impact Factor). The visibility of the journal in different scientific databases of current Journal Impact Factor is 0.611 (2020) and 0.897 international scope. A high ICDS means that the journal vii is present in different information sources of international 2020, the back end (the part that connects to the database relevance. The ICDS 2020 was 10.3. The Journal is also and the server that uses the website) of the website accepted/registered in several major databases/networks was upgraded, to a newer and safer server. This also (EBSCO, ROAD, AGRIS, CABI, DOAJ and Google included the purchase of HTTPS certificates to establish Scholar), by which international visibility is increased. secure connections. However, this also made the former The Journal´s current Google Scholar h-index is 20; the statistics provided by AWStats/CIAT IT become obsolete i10-index is 50. and from this year the website is keeping track of the user statistics via Google Analytics. Recent statistics indicate Authors 69,552 pageviews for the year 2020, while for the first six months of 2021 (i.e. until June 2021) the number of With the mission of providing high-quality tropical visits increased to 56,186 pageviews. forage knowledge to a wide group of stakeholders in the Users of the journal are worldwide, with currently tropics and making it accessible to the less developed half from the Americas and 25% from Asia (Figure 2). countries (Urrea-Benítez et al. 2020), 78% of the 1,837 The number of users from Africa is steadily increasing authors who have published in the journal so far are from as internet access is becoming more widespread, critical the global tropics (Africa, Asia, Latin America). The for an online journal. Open Access allows unrestricted authors came from 52 countries (Figure 1). use of the journal. Of interest is also the analysis of the operating system from where people are browsing the Reception Web metrics website, as an indicator of the main language of the users. While English, Spanish, Portuguese, Spanish and The number of total visits and the number of unique Chinese speakers are well represented, francophone visitors to its website have been increasing steadily. In users are so far under-represented (Figure 3). Figure 1. Countries of the authors published in the journal by June 2021. Source: Google Analytics. Disclaimer: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Tropical Grasslands-Forrajes Tropicales concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. viii Oceania 5.30% the International Livestock Research Institute (ILRI) for many years and have long and broad experience in tropical forages. Jean Hanson is an Emeritus fellow with Africa ILRI in Ethiopia and Danilo Pezo is currently working 10.68% with CATIE in Costa Rica. The new editorial team will continue to be well supported by Jose Luis Urrea, Europe Communications Specialist at The Alliance of Bioversity 8.22% International and CIAT, Colombia. Americas In the coming year the new editorial team will work 51.22% on further modernizing the journal including a fully Asia automatized submission and review process. Further 24.39% effort will be placed on increasing the quality and number of contributions and the reach of the journal to the global south, in particular Asia and Africa. In addition, an assessment of how to increase utilization from Francophone countries will be initiated aiming to Figure 2. Location of users of the journal by continent. Source: Google Analytics. maintain Tropical Grasslands-Forrajes Tropicales as the foremost journal on tropical forages at the global level. Chinese 4.56% Others References 9.05% (Note of the editors: All hyperlinks were verified 20 September 2021). Pulsford J. 2010. The Tropical Grassland Society of Australia Spanish Inc. - a great achiever. Tropical Grasslands 44:217−220. 19.08% English bit.ly/3yKYYgG 52.35% Schultze-Kraft R; Winks L; Bai C; Clements RJ; Larbi A; Peters M; Valle CB do. 2013. Empowering the next generation of tropical forage researchers: A new e-journal Portuguese for the 21st Century. In: Michalk DL; Millar GD; Badgery 14.96% WB; Broadfoot KM, Eds. Revitalising grasslands to sustain our communities: Proceedings of the 22nd International Grasslands Congress, Sydney, Australia, 15−19 September 2013. Australia: New South Wales Department of Primary Figure 3. Main language of the users of the journal based on Industry. p. 1950−1952. uknowledge.uky.edu/igc/22/3-9/3 the browser/operating system. Source: Google Analytics. Tropical Grasslands. 2010. Foreword. Tropical Grasslands 44:213−214. bit.ly/2WA79OW Outlook Urrea-Benítez JL; Peters M; Burkart S. 2020. ICTs in agriculture: State of the art tools for broader access to tropical The journal is in a transition period to the new editors, forage knowledge. Poster prepared for the Tropentag 2020: Dr. Jean Hanson, English and Managing Editor and Dr. Food and Nutrition Security and its Resilience to Global Crises, Virtual Conference, 9–11 September 2020. CIAT, Danilo Pezo, Spanish Editor. Both have worked with Cali, Colombia. hdl.handle.net/10568/111074 ix Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):249–255 249 doi: 10.17138/TGFT(9)249-255 Research Paper Canopy characteristics of ‘Mavuno’ hybrid brachiariagrass and ‘Marandu’ palisadegrass harvested at different harvest intensities Características del dosel de la pastura brachiaria híbrida 'Mavuno' y la pastura 'Marandu' recolectadas a diferentes intensidades de cosecha LUAN F. RODRIGUES1, JOAO M.B. VENDRAMINI2, ANTONIO C. DOS SANTOS1, JOSE C.B. DUBEUX JR3, FABRICIA R.C. MIOTTO1, LUCIANO F. SOUSA1 AND NAYARA M. ALENCAR1 1Departamento de Zootecnia, Universidade Federal do Tocantins, Araguaína, TO, Brazil. uft.edu.br 2Range Cattle Research and Education Center, University of Florida, Ona, FL, USA. rcrec-ona.ifas.ufl.edu 3North Florida Research and Education Center, University of Florida, Marianna, FL, USA. nfrec.ifas.ufl.edu Abstract ‘Mavuno’ is a newly released brachiariagrass (Urochloa hybrid) cultivar with limited information available in the literature. The objective of this study was to compare forage characteristics of this cultivar and ‘Marandu’ palisadegrass [Urochloa brizantha (Hochst. ex A. Rich.) R.D. Webster cv. Marandu] harvested at 2 different stubble heights during 2 growing seasons (January‒April). The study was conducted in Araguaína, TO, Brazil in 2017 and 2018. Treatments were the factorial arrangement of 2 brachiariagrass cultivars, Mavuno and Marandu, harvested at 2 harvest intensities, 5 and 15 cm stubble height, distributed in a randomized complete block design with 4 replicates. Response variables were canopy height, forage accumulation, proportion of leaf, stem and dead material, and concentration of crude protein (CP) and in vitro digestible organic matter (IVDOM). Mavuno and Marandu did not differ (P>0.05) in forage accumulation (mean = 3,800 kg DM/ha/harvest) and IVDOM concentration (mean = 637 g/kg); however, Mavuno had lower CP concentration (101 vs. 110 g/kg), greater proportion of stems (16 vs. 13%) and less dead material (4 vs. 6%) than Marandu (P<0.05). Harvesting at 5 cm stubble height rather than 15 cm increased herbage accumulation per harvest (4,100 vs. 3,500 kg DM/ha) with decreased proportion of leaves (77 vs. 84%) and CP concentration (101 vs. 115 g/ kg) (P<0.05). Our data suggest that Mavuno is a useful addition to the range of brachiariagrass cultivars for sowing in tropical regions and further studies are needed to evaluate the long-term persistence of Mavuno under different management practices in a range of environmental situations. While harvesting at 5 cm stubble height rather than 15 cm increased forage accumulation but reduced CP concentration, regardless of cultivar, longer-term effects on the stability of these pastures with these harvest frequencies and heights are open to question and studies should be continued for longer periods to assess longevity of stands under the 2 management strategies. Applying maintenance fertilizer during the growing season might have prevented the marked decline in dry matter accumulation as the season advanced and this hypothesis should be tested. Keywords: Harvest severity, nutritive value, tropical pastures, Urochloa spp. Resumen "Mavuno" es un cultivar de pasto brachiaria (híbrido de Urochloa) recientemente liberado con información limitada disponible en la literatura. El objetivo de este estudio fue comparar las características del forraje de este cultivar y el pasto 'Marandu' [Urochloa brizantha (Hochst. Ex A. Rich.) R.D. Webster cv. Marandu] cosechado a 2 alturas diferentes de rastrojo durante 2 temporadas de crecimiento (enero‒abril). El estudio se realizó en Araguaína, TO, Brasil en 2017 y 2018. Los tratamientos fueron el arreglo factorial de 2 cultivares de Brachiaria, Mavuno y Marandu, cosechados a Correspondence: Joao M.B. Vendramini, Range Cattle Research and Education Center, University of Florida, Ona, FL 33865, USA. Email: jv@ufl.edu Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 250 L.F. Rodrigues, J.M.B. Vendramini, A.C. dos Santos, J.C.B. Dubeux Jr., F.R.C. Miotto, L.F. Sousa and N.M. Alencar 2 intensidades de cosecha, 5 y 15 cm de altura de rastrojo, distribuidos en un diseño de bloques completos al azar con 4 repeticiones. Las variables de respuesta fueron altura del dosel, acumulación de forraje, proporción de hoja, tallo y material muerto, y concentración de proteína cruda (PC) y materia orgánica digestible in vitro (IVDOM). Mavuno y Marandu no difirieron (P> 0.05) en la acumulación de forraje (media = 3,800 kg MS / ha / cosecha) y la concentración de IVDOM (media = 637 g / kg); sin embargo, Mavuno tuvo menor concentración de PC (101 vs 110 g / kg), mayor proporción de tallos (16 vs 13%) y menos material muerto (4 vs 6%) que Marandu (P <0.05). La cosecha a 5 cm de altura de rastrojo en lugar de 15 cm aumentó la acumulación de forraje por cosecha (4.100 vs 3.500 kg MS / ha) con una proporción menor de hojas (77 vs 84%) y concentración de PC (101 vs 115 g / kg) ( P <0,05). Nuestros datos sugieren que Mavuno es una adición útil a la gama de cultivares de Brachiaria para la siembra en regiones tropicales y se necesitan más estudios para evaluar la persistencia a largo plazo de Mavuno bajo diferentes prácticas de manejo en una variedad de situaciones ambientales. Si bien la cosecha a 5 cm de altura de rastrojo en lugar de 15 cm aumentó la acumulación de forraje pero redujo la concentración de PC, independientemente del cultivar, los efectos a largo plazo sobre la estabilidad de estos pastos con estas frecuencias y alturas de cosecha están abiertos a cuestionamientos y los estudios deben continuar por más tiempo para evaluar la longevidad de las plantaciones bajo las 2 estrategias de manejo. La aplicación de fertilizantes de mantenimiento durante la temporada de crecimiento podría haber evitado la marcada disminución en la acumulación de materia seca a medida que avanzaba la temporada y esta hipótesis debería ser probada. Palabras clave: Pastos tropicales, severidad de la cosecha, Urochloa spp., valor nutritivo. Introduction grass herbage accumulation, nutritive value and persistence (Sollenberger and Burns 2001). In general, Brachiariagrasses (Urochloa spp.) are among the most forage harvested at shorter stubble height could have commonly planted forage species in tropical regions, and decreased residual leaf area and root growth, which may Marandu palisadegrass [Urochloa brizantha (Hochst. ex limit the regrowth rate of the pasture (Inyang et al. 2010). A. Rich.) R.D. Webster cv. Marandu] has been one of Depending on only limited numbers of brachiariagrass the most frequently used cultivars in Brazil, representing species and cultivars makes the livestock industry in approximately 35% of the total forage seed production in tropical regions vulnerable to infestations by pests the country (Jank et al. 2011). Palisadegrass is primarily and diseases. Therefore, it is important to diversify the used in extensive grazing systems that are subjected to genetic sources of brachiariagrass to create more resilient relatively low levels of inputs such as commercial fertilizer grazing systems in different regions of the world. The and liming (Miles et al. 2004). Despite the widespread objective of this study was to compare Marandu with use of palisadegrass, areas of Marandu have declined the new Mavuno brachiariagrass under different harvest for unknown reasons in some regions of Brazil (Barbosa intensities in a tropical region. We hypothesized that 2006). Therefore, new brachiariagrass cultivars need to be Mavuno and Marandu would have similar forage tested to potentially replace Marandu in those areas. accumulation and nutritive value. Mavuno is a hybrid brachiariagrass registered in Brazil (MAPA nº 30488) and was released as a Material and Methods commercial cultivar in April 2013. It originated from a cross between ruzigrass [Urochloa ruziziensis (R. Germ. The study was conducted at the Federal University of & C.M. Evrard) Crins] and U. brizantha, that has been Tocantins, Araguaína, Brazil (07º5' S, 48º12' W; 277 used for forage systems in tropical regions; however, masl), from January to April 2017 (Year 1) and December there is limited scientific information available about 2017 to April 2018 (Year 2). The experimental period Mavuno. Da Silva et al. (2020) observed that Mavuno chosen covered the growing season at the experimental had greater herbage accumulation and higher nutritive location, where only infrequent and scarce rainfall occurs value than ‘Tifton 85’ bermudagrass (Cynodon spp.) in the spring months (September‒December). The soil and Jiggs bermudagrass [Cynodon dactylon (L.) Pers.] type was Entisol (psamments, quartzipsamments). in Florida, USA, but had similar forage characteristics Initial soil characterization (0‒20 cm horizon) indicated to ‘Mulato II’ brachiariagrass, which is also a Urochloa that mean pH was 5.3 and Mehlich-1 extractable P, K, hybrid cultivar. Mg and Ca concentrations were 5, 20, 340 and 145 Harvest frequency and intensity are the most mg/kg, respectively. According to the Köppen climatic influential factors in terms of warm-season perennial classification, the region has a tropical humid summer Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Forage characteristics of Mavuno brachiariagrass 251 with well-defined rainy and dry seasons, with average Measurements annual rainfall of 1,828 mm. The rainfall, minimum and maximum temperatures during the experimental period Before each harvest, canopy height was measured using are presented in Table 1. a calibrated stick at 5 random points per plot from ground Treatments were the factorial arrangement of 2 level to the highest point reached by leaves or stems brachiariagrass cultivars, Mavuno and Marandu, with no disturbance of the sward. An area of 0.75 m2 and 2 harvest intensities, 5 and 15 cm stubble height, was harvested manually and subsamples were dried at distributed in a randomized complete block design with 55 ºC for 72 h and used to assess herbage accumulation, 4 replicates. The harvest intensity treatments provided a morphological composition and nutritive value. For comparison between a moderate harvest stubble height determining morphological composition, a subsample (15 cm) and a short stubble height, which may modify was taken and manually separated into leaf, stem and production of the pasture due to limited residual leaf area dead material. The remaining forage on each plot was (Giacomini et al. 2009). clipped at the same stubble height and removed from the Plot size was 3 × 3 m with 0.5 m alleys between plots after each harvest. plots and 1.0 m between blocks. On 16 January 2016, Tiller density and tiller mass were evaluated before the existing vegetation in the experimental area was the forage was harvested. The tillers in one 0.25 m 2 metal sprayed with glyphosate [N- (phosphonomethyl) ring per experimental unit were counted and the data glycine; Roundup Ultra 2, Monsanto Company, St used to estimate tiller density/m 2. Tillers were harvested, Louis, MO, USA] at a level of 0.8 kg/ha, following dried at 55 ºC for 72 h and tiller dry mass was calculated. which the seedbed was disked with a tandem disk until A further subsample for nutritive value determination there was no remaining vegetation on the soil surface. was taken, dried in the same way and ground to pass a Approximately 14 d after the soil preparation, seed 1 mm stainless steel screen in a Wiley mill (Model 4, was sown into the plots manually in rows 30 cm apart Thomas-Wiley Laboratory Mill, Thomas Scientific, at a depth of 2 cm. The seeding rate for both grasses Swedesboro, NJ, USA). The nutritive value analyses were conducted on whole-plant samples (leaf + stem). In vitro was 10 kg/ha, following the recommendation of the digestibility of organic matter (IVDOM) was determined seed company (Wolf Seeds) for those cultivars and using the two-stage technique described by Tilley and seed lots. Plots received 30 kg N, 13 kg P and 25 kg Terry (1963) and modified by Moore and Mott (1974). The K/ha approximately 14 d after germination. Plots were micro-Kjeldahl technique was used with a modification of clipped in January 2017 (Year 1) and December 2017 the aluminum block digestion described by Gallaher et al. (Year 2) at the respective treatment stubble height, (1975) for N determination. Crude protein was estimated fertilized with 60 kg N, 6 kg P and 50 kg K/ha and by multiplying N concentration by 6.25. evaluated every 28 d thereafter until April each year. The fertilizer sources were urea, simple superphosphate Statistical analysis and potassium chloride and rates chosen were used to represent the limited fertilizer strategies used by The data were analyzed using the PROC MIXED producers in tropical and subtropical regions. technique of SAS (SAS Institute Inc. 1996). Response Table 1. Monthly rainfall, minimum and maximum temperatures during the experimental period in Year 1 and Year 2 in Araguaína, TO, Brazil (INMET 2018) and long-term average rainfall (1984‒2018). Month, year Rainfall (mm) Average rainfall (mm) Max temperature (°C) Min temperature (°C) Year 1 January 2017 292 257 30.7 21.4 February 2017 345 265 30.6 21.7 March 2017 252 286 31.0 22.0 April 2017 208 221 31.4 22.2 Year 2 December 2017 256 227 30.5 22.4 January 2018 256 257 30.7 21.7 February 2018 345 265 30.3 22.3 March 2018 315 286 30.7 22.5 April 2018 124 221 32.0 22.0 Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 252 L.F. Rodrigues, J.M.B. Vendramini, A.C. dos Santos, J.C.B. Dubeux Jr., F.R.C. Miotto, L.F. Sousa and N.M. Alencar variables were canopy height, forage accumulation, CP, There were no differences in forage accumulation IVDOM and leaf, stem and dead material proportions. between cultivars (P>0.05), but pasture harvested at 5 Cultivar, harvest intensity, months and their interactions cm accumulated more forage than that at 15 cm (P<0.05; were considered fixed effects. Blocks and year were Table 2). As reflected by differences in canopy height, considered random effects. Month was analyzed as a forage accumulation declined progressively from January repeated measurement and the covariance structure to April (P<0.05; Table 3). selected based on the least Akaike information criterion While leaf proportion in harvested forage was similar value. Normality of residues and homogeneity of for the 2 cultivars (mean = 80.5%), Marandu had less variances were tested using conditional studentized stem (13 vs. 16%) and more dead material (6 vs. 4%) residual plots. Treatments were considered different than Mavuno (P<0.05) (Table 2). when P≤0.05 by LSD test. Main effects and interactions There was a cultivar × harvest intensity interaction not discussed in the Results and Discussion sections for tiller density (Table 4; P<0.05). Tiller density of both were not significant (P>0.05). Main effects were not cultivars did not differ (370 tillers/m2) when harvested at discussed if there was a significant (P<0.05) interaction 5 cm, but Mavuno had greater (P<0.05) tiller density than with the respective independent variable. Marandu when harvested at 15 cm (475 vs. 375 tillers/ m2). Tiller density declined as the season progressed Results for both harvest intensities but the differences were significant (P<0.05) only when harvested at 5 cm. Tiller Canopy height differed between cultivars and harvest mass was greater when harvested at 5 cm than when intensities, with no cultivar × harvest intensity interaction harvested at 15 cm (Table 2) and tiller mass declined (Table 2). Mavuno was taller than Marandu at harvest and progressively from January to April (P<0.05) (Table 3). forage harvested at 15 cm was taller than that harvested There were cultivar, harvest intensity (Table 2) and at 5 cm. In addition, month of harvest had an effect on month effects on CP concentrations (Table 3) but no canopy height (Table 3), which declined progressively significant interactions among the variables. Mavuno from January to April (P<0.05). had lower CP concentration than Marandu and forage Table 2. Effects of brachiariagrass cultivars, Mavuno (Urochloa hybrid) and Marandu [Urochloa brizantha (Hochst. ex A. Rich.) R.D. Webster], and harvest intensity (5 and 15 cm stubble height) on canopy height at harvest, forage accumulation/harvest, proportion of leaf, stem and dead material, tiller mass and crude protein (CP) and in vitro digestible organic matter (IVDOM) concentrations in 2017 and 2018 (means for two years). Parameter Cultivar Harvest consistency s.e P value Mavuno Marandu 5 cm 15 cm cv. Harvest height cv. × Harvest height Canopy height (cm) 31A1 28B 27b2 33a 0.71 <0.01 <0.01 0.34 Forage accumulation (kg DM/ha) 3,874 3,744 4,100a 3,500b 277 0.62 0.04 0.30 Leaf (%) 80 81 77b 84a 1.41 0.36 <0.01 0.83 Stem (%) 16A 13B 17a 12b 1.06 <0.01 <0.01 0.56 Dead material (%) 4B 6A 6a 4b 0.54 <0.01 <0.01 0.22 Tiller mass (g DM/m2) 1.0 1.0 1.2a 0.9b 0.07 0.98 <0.01 0.96 CP (g/kg) 101B 110A 101b 115a 3.1 0.03 0.05 0.16 IVDOM (g/kg) 634 640 640 634 7.0 0.19 0.27 0.09 1Cultivar means followed by the same upper-case letters are not different (P>0.05). 2Harvest consistency means followed by the same lower-case letters are not different (P>0.05). Table 3. Effects of month of harvest on canopy height and forage accumulation, plus crude protein (CP) and in vitro digestible organic matter concentrations (IVDOM) of brachiariagrass cultivars, Mavuno (Urochloa hybrid) and Marandu [Urochloa brizantha (Hochst. ex A. Rich.) R.D. Webster], harvested at 2 harvest intensities (5 and 15 cm stubble height) in 2017 and 2018 (means for two years). Parameter January February March April s.e Canopy height (cm) 43a1 31b 24c 22c 1.05 Herbage accumulation (kg DM/ha) 5,660a 4,330b 2,980c 2,260c 438 Tiller mass (g/m2) 1.4a 1.1b 0.8bc 0.6c 0.12 CP (g/kg) 110ab 101b 119a 82c 0.53 IVDOM (g/kg) 620b 650a 638a 640a 0.83 1Means within rows followed by the same lower-case letter are not different (P>0.05). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Forage characteristics of Mavuno brachiariagrass 253 Table 4. Harvest intensity × month effects on proportion of leaf, stem and dead material and tiller density in brachiariagrass cultivars, Mavuno (Urochloa hybrid) and Marandu [Urochloa brizantha (Hochst. ex A. Rich.) R.D. Webster], harvested in 2017 and 2018 (means for two years). Harvest intensity (stubble height) January February March April s.e Leaf (%) 5 cm 76bA 75bA 74bB 85aB 15 cm 76cA 79cA 87bA 94aA 1.4 s.e. 1.8 Stem (%) 5 cm 22aA 21aA 15bA 12bA 15 cm 22aA 18aA 6bB 4bB 1.0 s.e. 1.7 Dead material (%) 5 cm 3bA 4bA 11aA 5bA 15 cm 2bA 3bA 6aB 3bA 0.5 s.e. 0.8 No. of tillers/m2 5 cm 392aA 383aA 349bB 330cB 15 cm 401aA 411aA 406aA 383aA 9 s.e. 9 Within parameters, means within columns followed by the same upper-case letters and means within rows followed by the same lower-case letters are not different (P>0.05). harvested at 15 cm had higher CP concentration than and 51 min from December to May, which has potential that harvested at 5 cm. Crude protein concentration to decrease forage accumulation as well. was highest in March and lowest in April (P<0.05). Harvesting at 5 cm resulted in greater forage Conversely, IVDOM was not affected by either cultivar accumulation than harvesting at 15 cm despite displaying or harvest height (Table 2) but was lower in January than lesser height at all stages. This reinforces the findings in the remaining months (P<0.05) (Table 3). of Rodrigues et al. (2014), who harvested ‘Xaraes’ palisadegrass at different stubble heights (10, 20, 30, Discussion 40 and 50 cm) and observed an increase in herbage accumulation with decreasing stubble height. The greater canopy height for Mavuno than for Marandu The similar proportion of leaves in Mavuno and was likely due to its greater proportion of stems and Marandu canopies (mean 80.5%) was likely the main stem elongation resulted in forage with greater height. factor leading to similar IVDOM for forage from the While Da Silva et al. (2020) observed that Mavuno had 2 grasses. While Mavuno had a greater proportion greater canopy height than Mulato II, Jiggs and Tifton of stems than Marandu, it also had less dead material 85 bermudagrass, the correlation coefficients of canopy and the effects of these traits on IVDOM may have height with forage accumulation and light interception tended to cancel each other out. It is well reported in were only r = 0.60 and r = 0.56, respectively, indicating that the literature that stems have greater concentrations canopy height may not be an accurate indicator of forage of structural carbohydrates than leaves, and structural accumulation. This finding was supported by results from carbohydrates are generally less digestible than non- our study because the 2 cultivars did not differ in forage structural carbohydrates, which are commonly present at accumulation despite Mavuno being taller than Marandu. higher proportions in leaves (Chapman et al. 2014). It is The decrease in canopy height and forage also well documented that senescent material may have accumulation from January to April may be related to lost a significant proportion of cell contents, leading to the fact that fertilizer was applied only at the beginning decreased digestibility (Dubeux Jr et al. 2006). of the experimental period, considering that rainfall The greater proportion of stems in forage harvested at and temperature during the experimental period were 5 than at 15 cm was expected because there is a greater relatively uniform (Table 1), except for April 2017, when proportion of stems in the bottom layers of warm-season rainfall was lower than in other months. In addition, there forage canopies as reported by Vendramini et al. (2019) is a decrease in daylength from 12 h and 31 min to 11 h and Pontes et al. (2017). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 254 L.F. Rodrigues, J.M.B. Vendramini, A.C. dos Santos, J.C.B. Dubeux Jr., F.R.C. Miotto, L.F. Sousa and N.M. Alencar Mavuno harvested at 15 cm had greater tiller density during the growing season might have prevented the than Marandu, but there was no difference in tiller marked decline in dry matter accumulation as the season mass, indicating that Mavuno had less weight per tiller. advanced and this hypothesis should be tested. The same trend was observed with harvest intensity Further studies are warranted to evaluate the effects because forage harvested at 5 cm had lower tiller of additional abiotic and biotic factors on production density but greater tiller mass than forage harvested at and survival of Mavuno in a range of environmental 15 cm. Euclides et al. (2019) observed that increasing situations. grazing intensity, i.e. grazing to a shorter stubble height, decreased persistence of palisadegrass. The greater Acknowledgments decline in tiller density during the growth period when forage was cut at 5 cm rather than 15 cm in our study We thank the CAPES foundation for sponsoring the suggests that the more severe cutting height could result assistantship of the first author’s Ph.D. program. in reduced persistence of these pastures over time. Mavuno has lower CP concentration than Mulato References II under similar management systems and Da Silva et al. (2020) suggested that Mavuno may have an (Note of the editors: All hyperlinks were verified 28 May 2021). intrinsically lower CP concentration than other Barbosa RA. 2006. Morte de pastos de braquiarias. 1st Edn. selected brachiariagrasses. The marked reduction in CP Embrapa Gado de Corte, Campo Grande, MS, Brazil. bit. concentration observed in April was likely due to the ly/3jY1Twi run-down in soil N levels since no fertilizer was applied Barnard C. 1969. Herbage plant species. Australian Herbage after commencement (~90 d) plus limited rainfall during Plant Registration Authority, CSIRO, Canberra, ACT. April (Table 1). Australia. The slight increase in stem proportion in the forage Chapman DF; Lee JM; Waghorn GC. 2014. Interaction between harvested at 15 cm failed to decrease IVDOM of forage plant physiology and pasture feeding value: a review. Crop as the stems produced were relatively immature (28 and Pasture Science 65:721–734. doi: 10.1071/CP13379 days). While we expected Mavuno to have greater Da Silva HMS; Vendramini JMB; de Oliveira FCL; Soares Filho CV; Kaneko M; Silveira ML; Sanchez JMD; IVDOM than Marandu, there was no difference in Yarborough JK. 2020. Harvest frequency effects on IVDOM between the cultivars. Both Mavuno and Mulato herbage characteristics of ‘Mavuno’ brachiariagrass. Crop II are hybrids of ruzigrass and palisadegrass, and among Science 60:1113–1122. doi: 10.1002/csc2.20046 the brachiariagrass species ruzigrass is known to have Dubeux Jr JCB; Sollenberger LE; Vendramini JMB; Stewart consistently highest digestibility (Barnard 1969; Rosa et RL; Interrante SM. 2006. Litter mass, deposition rate, and al. 1983). The harvest intervals employed and fertilizer chemical composition in bahiagrass pastures managed at management used in this study may have negated the different intensities. Crop Science 46:1299–1304. doi: potential differences in IVDOM between Mavuno and 10.2135/cropsci2005.08-0262 Marandu. Euclides VPB; Montagner DB; Macedo MCM; de Araujo In summary, the absence of marked differences in AR; Difante GS; Barbosa RA. 2019. Grazing intensity affects herbage accumulation and persistence of Marandu performance between Mavuno and Marandu during palisadegrass in the Brazilian savannah. Grass and Forage the study suggests that Mavuno is a suitable option for Science 74:450–462. doi: 10.1111/gfs.12422 broadening the range of brachiariagrass genetic resources Gallaher RN; Weldon CO; Futral JG. 1975. An aluminum in tropical regions. However, Mavuno had greater block digester for plant and soil analysis. Soil Science canopy height and proportion of stem in the canopy, and Society of America Journal 39:803–806. doi: 10.2136/ lower CP concentration than Marandu, when harvested sssaj1975.03615995003900040052x at a fixed harvest frequency of 28 d. While harvesting at Giacomini AA; da Silva SC; Sarmento DOL; Zeferino CV; 5 cm stubble height rather than 15 cm increased herbage Souza Junior SJ; da Trindade JK; Guarda VA; Nascimento accumulation but reduced CP concentration, regardless Junior D. 2009. Growth of marandu palisadegrass subjected of cultivar, longer-term effects on the stability of these to strategies of intermittent stocking. Scientia Agricola 66:733–741. doi: 10.1590/S0103-90162009000600003 pastures with these harvest frequencies and heights are INMET (Instituto Nacional de Meteorologia). 2018. September open to question and studies should be continued for 25. inmet.gov.br/portal longer periods to assess longevity of stands under the 2 Inyang U; Vendramini JMB; Sellers B; Silveira MLA; Lunpha management strategies. Applying maintenance fertilizer A; Sollenberger LE; Adesogan A; Paiva LM. 2010. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Forage characteristics of Mavuno brachiariagrass 255 Harvest frequency and stubble height affects herbage of Xaraés grass subjected to intensity of cuts. Revista accumulation, nutritive value, and persistence of ‘Mulato Brasileira de Saúde e Produção Animal 15:815–826. (In II’ brachiariagrass. Forage and Grazinglands 8:1–7. doi: Portuguese) doi: 10.1590/S1519-99402014000400002 10.1094/FG-2010-0923-01-RS Rosa B; Rocha GP; Silva HL. 1983. Consumo voluntario e Jank L; de Valle CB; Resende RMS. 2011. Breeding tropical digestibilidade aparente de feno Brachiaria decumbens forages. Crop Breeding and Applied Biotechnology 11:27– Stapf e Brachiaria ruziziensis Germain et Evrard em 34. doi: 10.1590/S1984-70332011000500005 diferentes idades de corte. Anais das Escolas de Agronomia Miles JW; do Valle CB; Rao IM; Euclides VPB. 2004. e de Veterinaria 13:5–27. bit.ly/37Puiis Brachiariagrass. In: Moser LE; Burson BL; Sollenberger SAS Institute. 1996. SAS user’s guide. Release version 6. SAS LE, eds. Warm-Season (C4) grasses. Agronomy Monograph Institute, Cary, NC, USA. 45. ASA, CSSA, SSSA, Madison, WI. p. 745–784. doi: Sollenberger LE; Burns JC. 2001. Canopy characteristics, 10.2134/agronmonogr45.c22 ingestive behaviour and herbage intake in cultivated Moore JE; Mott GO. 1974. Recovery of residual organic matter tropical grasslands. In: Proceedings of the 19th International from in vitro digestion of forages. Journal of Dairy Science Grassland Congress, São Pedro, SP, Brazil. p. 321–327. 57:1258–1259. doi: 10.3168/jds.S0022-0302(74)85048-4 Tilley JMA; Terry RA. 1963. A two-stage technique for the in Pontes LS; Baldissera TC; Giostri AF; Stafin G; dos Santos vitro digestion of forage crops. Grass and Forage Science BRC; Carvalho PCF. 2017. Effects of nitrogen fertilization 18:104–111. doi: 10.1111/j.1365-2494.1963.tb00335.x and cutting intensity on the agronomic performance of Vendramini JMB; Sollenberger LE; Leite de Oliveira FC; warm-season grasses. Grass and Forage Science 72:663– Herling VR; Gomes VC; Sanchez JMD; Yarborough JK. 675. doi: 10.1111/gfs.12267 2019. Herbage characteristics of continuously stocked Rodrigues RC; Lana RP; Cutrim Jr. JAA; Sanchês SSC; limpograss cultivars under stockpiling management. Crop Galvão CML; de Sousa TVR; Amorim SEP; de Jesus Science 59:2886–2892. doi: 10.2135/cropsci2019.05.0299 APR. 2014. Accumulation of forage and sward structure (Received for publication 30 April 2020; accepted 8 May 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):256–267 256 doi: 10.17138/TGFT(9)256-267 Research Paper Effects of stubble height and season of the year on morphogenetic, structural and quantitative traits of Tanzania grass Efectos de la altura residual y de la estación del año en las características morfogénicas, estructurales y cuantitativas del pasto Tanzania NAUARA MOURA LAGE FILHO1, ALINE DA ROSA LOPES1, ANÍBAL COUTINHO DO RÊGO2, FELIPE NOGUEIRA DOMINGUES3, CRISTIAN FATURI2, THIAGO CARVALHO DA SILVA2, EBSON PEREIRA CÂNDIDO4 AND WILTON LADEIRA DA SILVA5 1Animal Science, Federal University of Pará, Castanhal, PA, Brazil. ufpa.br 2Animal Science, Federal Rural University of the Amazon, Belém, PA, Brazil. novo.ufra.edu.br 3Universidade Federal dos Vales do Jequitinhonha e Mucuri, Unaí, MG, Brazil. ufvjm.edu.br 4Federal Rural University of the Amazon, Capanema, PA, Brazil. novo.ufra.edu.br 5Federal University of Goiás, Goiânia, GO, Brazil. ufg.br Abstract The objective of this study was to evaluate regrowth period (RP), morphogenetic, structural and productive characteristics of the guinea grass cultivar Tanzania [Megathyrsus maximus (syn. Panicum maximum)] under different stubble heights (SH) during dry (DS) and rainy (RS) seasons in the eastern Amazon region. The treatments were: 5, 15, 25, 35, 45 and 55 cm SH, distributed in a randomized complete block design with 6 replicates. In the 2 seasons, RP decreased linearly with increase in SH, and was considerably shorter in the RS (47 d). Leaf appearance rate decreased linearly from 0.071 to 0.051 leaves/tiller/d with increasing SH, and it was higher during the RS. Increase in SH increased leaf elongation rate, stem elongation rate and leaf area index. In the RS, climatic conditions favored the morphogenesis, resulting in higher herbage accumulation (8,693 kg DM/ha) than in the DS (2,597 kg DM/ha). In associating seasons with SH, we recommend that Tanzania grass be managed at SH between 35 and 45 cm in the DS, resulting in RP from 61 to 64 days, and at SH of 35 cm in the RS, resulting in RP of 41 days. Studies to test this management strategy seem warranted. Keywords: Amazon biome, dry season, herbage accumulation, Megathyrsus maximus, rainy season. Resumen En el estudio se evaluaron el período de rebrote (RP), las características morfogénicas, estructurales y productivas de pasturas de Tanzania [Megathyrsus maximus (syn. Panicum maximum)] en diferentes alturas de rastrojo (SH) y estaciones del año, estación seca (DS) y estación lluviosa (RS) en la región Amazónica oriental. Los tratamientos incluyeron cinco SH: 15; 25; 35; 45 y 55 cm en un diseño de bloques completamente al azar con seis repeticiones. En las dos estaciones el RP disminuyó linealmente con el aumento de SH, y fue considerablemente menor en la RS (47 d). La tasa de aparición de hojas disminuyó linealmente de 0.071 a 0.051 hojas/macolla/d con el aumento de SH, y fue mayor durante la RS. El aumento en SH proporcionó aumento en la tasa de alargamiento de hoja, en la tasa de alargamiento del tallo, y en el índice de área foliar. En la RS, las condiciones climáticas favorecieron la morfogénesis del cultivar Tanzania, lo que resultó en mayor acumulación de forraje (8,693 kg DM/ha) que DS (2,597 kg DM/ha). En la asociación de estaciones con SH, recomendamos que pasturas de Tanzania se maneje en SH entre 35 y 45 cm en DS, correspondiente a RP de 61 a 64 días, y en el SH de 35 cm en la RS, correspondiente a RP de 41 días. Los estudios para probar esta estrategia de gestión parecen justificados. Correspondence: W.L. da Silva, Animal Science Department, Federal University of Goiás (UFG), Goiânia, 74690-900, GO, Brazil. E-mail: wiltonladeira@ufg.br Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stubble height and season effects on Tanzania grass 257 Palabras clave: Acumulación de biomasa, Amazonia oriental, época lluviosa, época seca, Megathyrsus maximus. Introduction a view to optimizing pasture use and, consequently, reducing environmental impacts in the region. In aiming to intensify pasture management, producers in The hypothesis tested in this experiment was that, Brazil largely use cultivars of guinea grass [Megathyrsus when associated with differing SH following harvesting, maximus (syn. Panicum maximum)], a species of African climatic conditions typical of the Amazon region origin well adapted to tropical conditions. A cultivar (precipitation regime, mainly) affect both the structure widely used is Tanzania, due to its favorable traits and production of Tanzania grass. Therefore, the objective such as high production potential and nutritional value of this study was to examine the morphogenetic and (Paciullo et al. 2016). However, this cultivar is also very structural traits and herbage production of Megathyrsus demanding in terms of soil fertility (Pezzopane et al. maximus cv. Tanzania managed under 5 different SHs 2016) and management owing to its high stem elongation throughout an experimental year in the eastern Amazon rates, notably near flowering time. region. Tanzania grass pastures have gained prominence in the Amazon biome thanks to the expansion of Brazilian Material and Methods livestock to this region, despite very strict environmental laws (Nascimento et al. 2019). Thus, many concerns Experimental site arise regarding the management of this forage plant due to the different soil-climatic conditions found in the The experiment was conducted on the experimental region, such as highly acidic, phosphorus-poor and high- farm at the Federal Rural University of the Amazon water-table soils, annual precipitation above 2,000 mm, (UFRA), located in Igarapé-Açu, Pará, Brazil (01º07'21” plus minimum temperature above 20 °C and availability S, 47º36'27” W; 50 masl), from August 2017 to August of light during most of the year. 2018, in plots of Tanzania grass established in 2014. In addition to the soil-climatic peculiarities of the The soil is classified as a Yellow Latosol (Oxisol) region that influence herbage production, it is important with a sandy-loam texture and a low slope gradient. to understand how grasses respond to the intensity/ Soil analysis performed in the 0 20 cm layer revealed severity (Silva et al. 2016, 2019; Gomide et al. 2019) and the following chemical characteristics: pH (CaCl2) =⁠ frequency of defoliation (Moura et al. 2017; Pedreira et 4.7; organic matter = 7.98 g/kg; P (ion-exchange resin al. 2017) under the specific conditions mentioned above. extraction method) = 1.54 mg/dm3; K = 3.0 mmolc/dm3; Defoliation frequency can be determined based on light Ca = 28.0 mmolc/dm3; Mg = 28.0 mmolc/dm3; H + Al interception (LI) by the canopy, which, according to = 47.2 mmolc/dm3; cation-exchange capacity = 106.2 recent research (Pedreira et al. 2017; Tesk et al. 2018; mmolc/dm3; and base saturation = 55.7%. Silva et al. 2019), has shown potential for use when The local climate is classified as a tropical monsoon the level of 95% is adopted. At this stage pastures are (Am) type according to the Köppen classification, with considered to have high proportions of leaves and low a short dry season and heavy rains during the rest of the proportions of stem and dead material in the herbage year. Total precipitation during the experimental period mass, in addition to better quality (Brougham 1956; was 2,270 mm, consisting of 130 mm from September Korte et al. 1982; Parsons et al. 1988). to December 2017 and negative soil water balance (96 Defoliation intensity/severity can be predefined based to 175 mm monthly deficit) (characterized as the dry on stubble height (SH) and is widely used in research on season, DS), and 2,140 mm from January to August 2018 forage plants in tropical (Silva et al. 2016; Pereira et al. (rainy season, RS) with positive soil water balance from 2018; Gomide et al. 2019; Tesk et al. 2020) and temperate February to July (15‒380 mm monthly surplus) (Table regions (Kohmann et al. 2017; Insua et al. 2020). As 1). Mean temperature during the experimental period such, it has become a management tool on many farms. was 27.5 °C. However, studies on effects of SH on morphology and forage production in pastures in Brazil have been located Experimental design and management in regions where soil-climatic characteristics are very different from those encountered in the Amazon biome. Treatments consisted of 5 Tanzania grass stubble heights This reinforces the need for studies like the present one, (SH: 15, 25, 35, 45 and 55 cm) evaluated for a full year which may assist in decision-making in the field with with data separated into dry (DS) and rainy (RS) seasons. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 258 N.M. Lage Filho, A.R. Lopes, A.C. do Rêgo, F.N. Domingues, C. Faturi, T.C. da Silva, E.P. Cândido and W.L. da Silva The experiment was laid out as a randomized complete Morphogenetic and structural traits block design with 6 replicates, in a total of thirty 3 × 4 m experimental units spaced 1 m apart. Leaf blades and stems were measured on 5 tillers per plot, once weekly, during the regrowth period. After Table 1. Monthly rainfall and water balance in Igarapé-Açu, each harvest, 5 new tillers were selected. PA, Brazil. All leaves of each tiller were numbered and classified Month Precipitation (mm) Soil water balance (mm) as expanded (with the ligule visible), expanding (no visible Climate mean (1994‒2017) Experimental period (2017‒2018) ligule) or senescent (when the end of the leaf blade showed some sign of senescence). Leaves with more than 50% of the Sep 44.7 34.2 -98 blade length compromised by senescence were considered Oct 22.0 17.1 -152 Nov 18.5 0.0 -175 dead. In the expanded leaves, length was measured from Dec 96.6 78.9 -96 the tip of the blade to its ligule. For expanding leaves, Jan 236.1 153.3 0 length was measured from the tip of the blade to the ligule Feb 308.0 514.4 380 of the youngest fully expanded leaf. In the case of senescent Mar 456.7 303.4 125 leaves, length of the green leaf blade was measured from Apr 389.4 408.8 255 the ligule to the point where senescent tissue was visible. May 258.5 381.8 245 The length of stem plus sheath was measured from ground Jun 199.6 146.3 20 level to the ligule of the last fully expanded leaf. Jul 155.0 134.8 15 The data were used to estimate the rates of leaf Aug 62.4 97.2 1 appearance (LAR, leaves/tiller/d), leaf elongation (LER, cm/tiller/d), stem elongation (SER, cm/tiller/d) and leaf In June 2017, still in the RS, the various plots senescence (LSR, cm/tiller/d) for each tiller. The number were harvested at the appropriate stubble heights for of live leaves per tiller (NLL) was also determined by implementation of the study. Subsequently, plots received direct counting. Leaf lifespan (LLS, days) was determined the equivalent K dose of 100 kg/ha in the form of KCl using the values of LAR and NLL per tiller (Lemaire plus 200 kg N/ha in the form of urea divided into 3 equal and Chapman 1996). Phyllochron (PHY, days/leaf) was applications (January, March and May 2018). After this calculated as the inverse of LAR. Final leaf size (FLS, harvest, light interception (LI) by the canopy in each cm) was determined as the length of expanded leaf blades. plot was monitored using the AccuPAR LP-80 canopy Leaf and stem elongation rates and LSR were obtained by analyzer (Decagon®) throughout the regrowth period dividing the difference between the final and initial lengths of the grass, with readings taken daily at 3 points per of the green or senescent leaf blades or stems, by the number plot, until the canopy reached 95% LI. Upon reaching of days in the regrowth period. Leaf appearance rate was 95% LI, the plots were harvested at the appropriate SH calculated by dividing the number of expanded leaves per (treatments) using a mower. tiller by the number of days in the regrowth period. The harvest intervals in the plots and the number of Height of forage in plots in pre-harvest condition was harvest events, according to the SH and the evaluation measured whenever the plot reached 95% LI, at 5 points periods, are shown in Table 2. The total number of harvest per plot, using centimeter-graded rules. The indirect leaf events was determined by the length of the regrowth area index (LAI) was also determined in the pre-harvest period (RP) in each treatment, i.e. time to reach 95% condition using the AccuPAR PAR/LAI LP-80 canopy LI. Regrowth period was calculated as the time (days) analyzer (Decagon®). Readings were taken at 3 points between one harvest and the subsequent one. per plot along with the measurement of LI. Table 2. Duration of dry and rainy seasons and number of harvest events (in parentheses) of Tanzania grass for different stubble height (SH) treatments in Igarapé-Açu, PA, Brazil. SH (cm) Season of the year Total harvest events Dry Rainy 15 10 Sep ‒ 31 Dec 2017 (1) 01 Jan ‒ 04 Aug 2018 (3) 4 25 20 Sep ‒ 31 Dec 2017 (1) 01 Jan ‒ 03 Aug 2018 (4) 5 35 01 Sep ‒ 31 Dec 2017 (2) 01 Jan ‒ 10 Aug 2018 (5) 7 45 01 Sep ‒ 31 Dec 2017 (2) 01 Jan ‒ 05 Aug 2018 (5) 7 55 01 Sep ‒ 31 Dec 2017 (3) 01 Jan ‒ 06 Aug 2018 (7) 10 Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stubble height and season effects on Tanzania grass 259 Tiller population density (TPD) was evaluated oven at 60 °C for 72 h, and the proportions of each whenever the plots intercepted 95% of light. For this morphological component were calculated based on the assessment, the total number of live tillers within a 1 weight of the components. Herbage accumulation was m × 0.5 m frame per plot was counted and the result calculated for each harvest cycle and, at the end of each expressed as number of tillers per square meter. period, all accumulations were summed, generating the Tiller demography was evaluated in a single total herbage in each period (THA). tussock per plot. Immediately after the grass was cut to implement the SH, the tillers present in the tussock were Statistical analysis counted. These tillers were marked and termed ‘zero- generation tillers’ (G0). At the subsequent harvest, live Data were analyzed using the PROC MIXED procedure of tillers from G0 were counted, and new emerged tillers, SAS software (SAS Institute Inc. 2008). The model used termed ‘generation-one tillers’ (G1), were marked. The for all studied variables contained the effects of SH, season total number of dead tillers was always counted in each of the year and their interactions, which were considered evaluation cycle until the end of the experimental period. fixed effects, whereas the blocks (replicates) and their Tiller counts were made successively, totaling 5, 6, 8, interactions were considered random effects. The procedure 8 and 10 generations for SHs of 15, 25, 35, 45 and 55 of repeated measures over time was used with the variance cm, respectively. From the obtained data, the following components as a covariance structure. Treatment means variables were calculated and expressed as tillers/100 were considered different and interactions significant when tillers per regrowth period, following Bahmani et al. P≤0.05. Polynomial orthogonal contrasts were used when (2003): tiller appearance rate (TAR) = number of emerged SH effects were observed. Means as a function of seasons tillers divided by the number of existing tillers at the were compared by the F-test. previous marking event; tiller mortality rate (TMR) = number of dead tillers divided by the number of existing Results tillers at the previous marking event; and tiller survival rate (TSR) = number of remaining tillers divided by the There was no significant interaction effect (P>0.05) number of tillers existing at the previous marking event. between SH and season for the length of the regrowth period (RP) (Table 3). The RP decreased linearly with Herbage accumulation and morphological composition increase in SH and was considerably shorter in the RS than in the DS (P≤0.05). Herbage accumulation above the SH was measured by collecting the herbage within a 1 m × 0.5 m quadrat per Morphogenetic and structural traits plot whenever the canopy reached 95% LI. The collected samples were halved: one half being used to determine There was a significant interaction effect between SH DM concentration, and the other for a manual separation and season (P≤0.05) for the morphogenetic variables of the morphological components, namely: leaf blade LLS and SER. LAR, PHY and LER variables were (LB), stem + sheath (ST) and dead material (DeM). influenced by both SH and season (P≤0.05), whereas Samples were then weighed and dried in a forced-air LSR was influenced by SH only (Table 4). Table 3. Effects of stubble height and season of the year on the duration of the regrowth period (RP) of Tanzania grass (Megathyrsus maximus) in Igarapé-Açu, PA, Brazil. Season Stubble height (cm) Effect Mean s.e. 15 25 35 45 55 Regrowth period (d) Dry 115.6 96.3 64.1 61.4 59.6 79.3a 1.15 Rainy 71.1 53.7 41.2 40.7 29.4 47.3b 0.68 Mean 93.1 75.0 52.6 51.2 44.6 L (<0.0001) s.e. 1.32 1.21 1.21 1.21 1.28 Within stubble heights seasonal means followed by different letters are different (P≤0.05). L: observed significance level for linear effects of SH. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 260 N.M. Lage Filho, A.R. Lopes, A.C. do Rêgo, F.N. Domingues, C. Faturi, T.C. da Silva, E.P. Cândido and W.L. da Silva Table 4. Effects of stubble height and season of the year on leaf appearance rate (LAR), leaf elongation rate (LER), phyllochron (PHY), leaf senescence rate (LSR), leaf lifespan (LLS) and stem elongation rate (SER) of Tanzania grass (Megathyrsus maximus) in Igarapé-Açu, PA, Brazil. Season Stubble height (cm) Effect Mean s.e. 15 25 35 45 55 LAR (leaves/tiller/d) Dry 0.063 0.051 0.048 0.058 0.053 0.050b 0.003 Rainy 0.080 0.068 0.060 0.058 0.050 0.064a 0.002 Mean 0.071 0.059 0.054 0.058 0.051 L (<0.0001) s.e. 0.003 0.003 0.004 0.003 0.004 LER (cm/tiller/d) Dry 1.45 1.60 1.67 2.22 2.38 1.86b 0.239 Rainy 2.92 3.29 3.60 3.37 3.97 3.43a 0.239 Mean 2.18 2.44 2.64 2.79 3.18 L (<0.0001) s.e. 0.234 0.234 0.247 0.234 0.247 PHY (d/leaf) Dry 15.87 19.61 20.83 17.24 18.87 18.48a 1.60 Rainy 12.50 14.71 16.67 17.24 18.87 15.99b 1.50 Mean 14.09 16.95 18.52 17.24 18.87 L (<0.0001) s.e. 1.58 1.58 1.50 1.58 1.50 LSR (cm/tiller/d) Dry 0.138 0.152 0.217 0.269 0.353 0.226 0.025 Rainy 0.114 0.193 0. 239 0.307 0.363 0.243 0.026 Mean 0.126 0.172 0.228 0.288 0.358 Q (<0.0001) s.e. 0.060 0.060 0.036 0.036 0.025 LLS (d) Dry 69.49a 58.72a 45.01a 42.78a 41.85a Q (0.0201) 51.56 1.44 Rainy 44.44b 37.93b 35.69b 35.18b 24.73b L (<0.0001) 35.59 0.74 Mean 56.94 48.32 40.35 38.98 33.29 s.e. 2.062 1.618 1.618 1.618 1.618 SER (cm/tiller/d) Dry 0.060a 0.069a 0.074b 0.075b 0.052b NS 0.065 0.007 Rainy 0.071a 0.091a 0.104a 0.117a 0.138a L (0.0079) 0.100 0.006 Mean 0.065 0.080 0.089 0.096 0.095 s.e. 0.011 0.011 0.011 0.011 0.011 Within stubble heights means for seasons followed by different letters are different (P≤0.05). L, Q: observed significance level for linear and quadratic effects of SH, respectively. Leaf appearance rate decreased linearly with Stem elongation rate was not significantly affected by SH increasing SH (P≤0.05) and overall was greater during during the DS (P>0.05) but increased linearly (P≤0.05) the RS than during the DS (0.064 vs. 0.050 leaves/ with increase in SH during the RS. For SH of 35, 45 and tiller/d) (Table 4). In contrast, average LER increased 55 cm, SER was higher during the RS than during the linearly as SH increased but was greater during the RS DS (P≤0.05) (Table 4). than during the DS (3.43 vs. 1.86 cm/leaf/d). Phyllochron In the analysis of structural traits (Table 5), a increased linearly with increase in SH (P≤0.05), since significant interaction effect between SH and season LAR declined linearly, and was longer during the DS (P≤0.05) was observed for FLS and NLL. Final leaf than in the RS (18.48 vs. 15.99 d/leaf). size increased linearly with increasing SH during the Leaf senescence rate increased quadratically DS (P≤0.05) with no effect of SH during the RS. Leaves (P≤0.0001) with increase in SH, with no significant effect were larger (P≤0.05) during the RS at all SHs except 55 of season. Leaf lifespan decreased with increasing SH, cm. Number of live leaves decreased with increasing the response being quadratic during the DS (P≤0.05) and SH in both seasons, linearly (P≤0.05) during the DS and linear during the RS (P≤0.0001). On average, LLS was quadratically (P≤0.05) during the RS. At SHs of 35, 45 greater during the DS of the year at all SHs (P≤0.05). and 55 cm, NLL was higher in the RS than in the DS Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stubble height and season effects on Tanzania grass 261 (Table 5). Tiller population density was affected only by Herbage accumulation and morphological composition SH (P≤0.05), decreasing quadratically as SH increased (P≤0.05). In the pre-harvest condition, i.e. upon reaching There was a significant interaction effect between 95% LI, canopy height was not altered (P>0.05) by SH SH and season for proportions of leaf blade and dead or season. Average height of Tanzania grass at 95% LI material (P≤0.05). Isolated effects of SH (P≤0.05) and was 75.0 ± 0.40 cm in the DS and 76.5 ± 0.46 cm in season of the year (P≤0.05) were detected for THA and the RS. Leaf area index rose linearly with increasing SH stem + sheath (Table 7). (P≤0.05), ranging from 2.78 in the plots managed at 15 Total herbage accumulation decreased linearly with cm to 3.97 in those managed at SH of 55 cm (Table 5). increase in SH in both seasons (P≤0.05). Only about As regards tiller demographic variables, a significant 22% of THA occurred in the DS (P≤0.05) (Table 7). interaction effect between SH and season (P≤0.05) was Leaf blade proportion increased linearly (P≤0.05) detected for tiller mortality rate (TMR) (Table 6). A with increase in SH in both seasons (Table 7). While leaf linear reduction in TMR was observed as SH increased proportion was greater (P≤0.05) in the DS than the RS at 15 in both seasons (P≤0.05). For most SHs TMR was higher cm SH, at 55 cm SH the reverse was the case. During the in the RS but differences were significant (P≤0.05) RS, DeM proportion decreased linearly with increasing SH only for 15 and 25 cm SH. Tiller appearance rate also but showed a quadratic response during the DS (P≤0.05). decreased linearly with increasing SH (P≤0.05) and was ST proportion responded quadratically to increasing SH higher during the RS than in the DS. TSR also increased (P≤0.05), and at the height of 55 cm this component was linearly with increase in SH (P≤0.05) in both seasons not found in the samples of herbage accumulated above the and was higher during the DS than in the RS (67.5 vs. residual stubble. Stem + sheath percentage was higher in 60.3%) (Table 6). the RS (6.21%) than in the DS (5.38%). Table 5. Effects of stubble height and season of the year on final leaf size (FLS), number of live leaves (NLL), tiller population density (TPD) and leaf area index (LAI) of Tanzania grass (Megathyrsus maximus) in Igarapé-Açu, PA, Brazil. Season Stubble height (cm) Effect Mean s.e. 15 25 35 45 55 FLS (cm) Dry 24.1b 25.6b 28.3b 28.3b 31.7a L (0.0003) 27.6 1.16 Rainy 35.0a 30.9a 33.4a 34.6a 33.9a NS 33.6 1.18 Mean 29.6 28.2 30.8 31.4 32.8 s.e. 1.22 1.15 1.15 1.15 1.15 NLL Dry 3.82a 3.64a 3.33b 3.02b 2.67b L (<0.0001) 3.29 0.08 Rainy 3.81a 3.85a 3.97a 3.78a 3.19a Q (0.0235) 3.72 0.08 Mean 3.81 3.74 3.65 3.40 2.93 s.e. 0.11 0.10 0.09 0.09 0.09 TPD (tillers/m²) Dry 264 258 260 234 224 228 3.60 Rainy 272 274 276 220 222 232 3.22 Mean 268 266 268 228 224 Q (0.0295) s.e. 6.34 5.36 5.08 5.08 5.08 LAI Dry 2.63 3.27 3.64 3.96 4.02 3.50 0.08 Rainy 2.93 3.27 3.65 3.67 3.91 3.55 0.03 Mean 2.78 3.43 3.64 3.86 3.97 L (<0.0001) s.e. 0.09 0.09 0.09 0.09 0.09 Within stubble heights means followed by different letters are different (P≤0.05). L, Q: observed significance level for linear and quadratic effects of SH, respectively. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 262 N.M. Lage Filho, A.R. Lopes, A.C. do Rêgo, F.N. Domingues, C. Faturi, T.C. da Silva, E.P. Cândido and W.L. da Silva Table 6. Effects of stubble height and season of the year on tiller mortality rate (TMR), tiller appearance rate (TAR) and tiller survival rate (TSR) of Tanzania grass (Megathyrsus maximus) in Igarapé-Açu, PA, Brazil. Season Stubble height (cm) Effect Mean s.e. 15 25 35 45 55 TMR (tillers/100 tillers) Dry 42.4b 31.0b 27.2a 26.0a 23.8a L (0.0005) 30.3 1.14 Rainy 58.9a 44.5a 36.4a 35.3a 23.4a L (<0.0001) 39.7 1.03 Mean 50.7 37.7 31.8 30.6 23.6 s.e. 2.04 1.63 1.63 1.63 1.63 TAR (tillers/100 tillers) Dry 37.9 31.0 30.9 31.9 22.1 30.5b 1.73 Rainy 58.7 49.0 39.1 39.1 28.8 42.9a 1.70 Mean 48.3 40.0 35.0 35.5 25.4 L (<0.0001) s.e. 2.69 2.83 2.69 2.69 2.69 TSR (tillers/100 tillers) Dry 46.6 68.1 72.8 74.0 76.2 67.5a 1.69 Rainy 41.1 55.5 63.6 64.7 76.6 60.3b 0.87 Mean 43.8 61.8 68.2 69.4 76.4 L (<0.0001) s.e. 2.41 2.22 2.22 2.22 2.22 Means followed by different letters comparing the effect of seasons are different (P≤0.05). L: observed significance level for linear effects of SH Table 7. Effects of stubble height and season of the year on total herbage accumulation (THA) and leaf blade (LB), dead material (DeM) and stem (ST) proportions of Tanzania grass (Megathyrsus maximus) in Igarapé-Açu, PA, Brazil. Season Stubble height (cm) Effect Mean s.e. 15 25 35 45 55 THA (kg DM/ha) Dry 3,503 2,671 2,510 2,304 1,647 2,527b 125 Rainy 10,142 9,262 8,664 7,913 7,482 8,693a 187 Mean 6,822 5,966 5,587 5,109 4,565 L (<0.0001) s.e. 197 179 197 197 197 LB (g/kg DM) Dry 611a 762a 780a 777a 793b L (0.0004) 745 16.0 Rainy 561b 756a 798a 833a 857a L (<0.0001) 761 12.0 Mean 587 759 789 805 825 s.e. 17.5 16.2 16.2 16.2 16.2 DeM (g/kg DM) Dry 201a 198a 163a 163a 209a Q (<0.0001) 187 10.0 Rainy 228a 167a 166a 154a 142b L (<0.0001) 172 10.0 Mean 215 183 165 159 176 s.e. 15.8 15.8 15.8 15.8 15.8 ST (g/kg DM) Dry 182 45.9 32.5 9.8 0.0 53.8b 3.1 Rainy 198 59.5 39.7 12.8 0.0 62.1a 3.1 Mean 190 52.7 36.1 11.3 0.0 Q (<0.0001) s.e. 8.2 8.2 8.2 8.2 0.0 Means followed by different letters comparing the effect of seasons are different (P≤0.05). L: observed significance level for linear effects of SH Discussion the canopy (Paciullo et al. 2016), as well as by changes in temperature and water availability (Tilley et al. 2019), Shorter regrowth periods in pastures managed during which was highly contrasting between the two seasons. the RS, as compared with the DS, are mainly due to Environmental conditions adverse to tiller development contrasting climatic conditions between the seasons. lead to less herbage accumulation after harvest (Table 7), Tillering is directly affected by the light intercepted by causing plants to expend more time and larger amounts of Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stubble height and season effects on Tanzania grass 263 reserves to re-intercept 95% LI (Silva et al. 2019). In this al. (2016) in Tifton-85 pastures, reinforcing the inverse study, we observed that differences in regrowth period relationship between these variables. between the two seasons declined as SH was increased Higher LER occurs at higher SH (Table 4), due to up to 45 cm. For instance, at SH of 15 cm the difference the elongation of pseudostems. Shorter pseudostems between seasons in regrowth period was 44.5 days, favor a rapid leaf emergence, resulting in lower LER. whereas at 45 cm this difference was only 20.7 days. The opposite is also true, as leaves take longer to emerge In grass species, the regrowth period can also be from longer pseudostems, resulting in higher LER. In influenced by harvesting or grazing, especially as a this respect, Zanine et al. (2018) observed an increase in function of SH, which is directly related to the remaining LER from 11.54 to 15.21 cm/tiller/d as SH was increased LAR. Pastures managed under lower SH need longer from 30 to 50 cm in Tanzania grass pastures managed intervals to recover leaf area to be able to achieve 95% LI. at 95% LI in the pre-grazing condition during the RS. As a consequence, they show a longer regrowth period, The higher LER (3.43 cm/tiller/d) seen in the RS was which, in practice, is not recommended. The management likely due to the greater water availability during this of Tanzania grass in an intermittent grazing system has season than in the DS (2,087 vs. 169 mm). Barbosa et been recommended in Brazil with a pre-grazing height al. (2011) studied the effects of defoliation intensities of 70 cm (LI = 95%) (Euclides et al. 2014; Zanine et al. and frequencies in Tanzania grass and observed LER of 2018) and a post-grazing SH close to 50% of the pre- 4.16 and 1.16 cm/tiller/d in the rainy and dry seasons, grazing height, i.e. 35 cm. However, what differs in the respectively, and associated this difference with climatic studies are the varying regrowth periods required by the differences between the two seasons. grass to start from SH of 35 cm and reach 95% LI (70 cm). The increase in LSR in response to increasing These differences in regrowth period are usually related to SH observed in both seasons (Table 4) is related to factors such as time of year, soil management and fertility intraspecific competition for light, which reduces the and the climate of the region, warranting the development quantity and quality of the light that penetrates the pasture of studies in different parts of the country. On average, as the grass becomes taller. As an adaptation response to the regrowth period of Tanzania grass in the RS in central this competition, the plant starts to invest in elongating Brazil has ranged between 21 and 32 days, at pre- and its internodes to elevate leaves to the top of the pasture, post-grazing heights of 70 and 35 cm, respectively. In our where light is more abundant. Simultaneously, the leaves study, this regrowth period was 41.2 days, as the grass located at the base of the tussocks become more shaded, reached 95% LI at the greatest average height (76.5 cm). which accelerates leaf senescence (Duchini et al. 2013). This was probably due to the typical excess rainfall that For this reason, during the regrowth period, both SER occurs in the region during the RS (2,086 mm in 2018), and LSR exhibit the same response pattern. These results leaving part of the months with a cloudy sky. corroborate those described by Silva et al. (2016) in Tifton-85 grass. Morphogenetic and structural traits Leaf lifespan expresses the tissue flow occurring in the plant, which is normally higher during the DS Decreasing LAR in tropical grasses in response to (Table 4). The longer LLS in the DS may be the result increases in SH (Table 4) are usually associated with the of low precipitation as well as lower day-length, which longer pseudostems that occur at greater stubble heights, are typical of this season in tropical conditions. In this which, in turn, result in a greater distance to be traveled scenario, grasses extend the lifespan of green leaves by the leaf until its exposure above the tube (Lemaire and reduce leaf tissue turnover, which results in lower and Chapman 1996). This is better understood when we LAR and LER as well as a longer PHY, as previously analyze the increase in PHY, i.e. the time taken for two discussed. It is thus clear that these traits are influenced consecutive leaves in the tiller to appear, following an by seasons of the year, which are directly related to increase in SH. Phyllochron follows an inverse response environmental conditions. pattern to that of LAR; at lower SH, PHY tends to decrease The shortened LLS in response to increasing SH and LAR tends to increase, possibly related to the plant's in both seasons (Table 4) can be better understood need to restore the photosynthetic apparatus shortly after when considered together with the leaf senescence harvest or a more intense grazing event. Similar results process, which increased along with increasing SH. showing a decrease in LAR and an increase in PHY Once senescence is established, nutrients are redirected following increases in SH were described by Silva et to younger leaves, which reduces the photosynthetic Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 264 N.M. Lage Filho, A.R. Lopes, A.C. do Rêgo, F.N. Domingues, C. Faturi, T.C. da Silva, E.P. Cândido and W.L. da Silva activity of older leaves and leads to a reduction in LLS shorter pastures (lower SH) and vice-versa (Lima et al. (Oliveira et al. 2007), as observed in our study. 2017; Santana et al. 2017). This fact can be explained From SH of 35 cm upwards, higher SER was as a response to the tiller size/density compensation observed during the RS (Table 4), which may be mechanism existing in higher plant communities associated with a larger amount of leaves remaining after (Matthew et al. 1995). By using this mechanism, grasses harvest at these greater SHs. These remaining leaves regulate pasture leaf area and, consequently, the ability to usually cause shading on the tillers, prompting them to intercept incoming light. This can cause greater shading elongate their stem to capture light in the upper layers at the base of the canopy under higher LAI, which can of the canopy, which also explains the linear increase reduce the stimulation of the basal and axillary buds for in SER along with increasing SH. Another explanation the production of new tillers. Indeed, LAI rose linearly for the difference in SER between the seasons may be with increasing SH, which reinforces the premise related to the reproductive stage of the cultivar, which described. This increase in LAI may be associated with produces inflorescences during the summer (RS) the increase in LER as SH was raised. under tropical conditions. As inflorescence-emergence When managed under lower SH, Tanzania grass may approaches, tropical grasses elongate their stems so that exhibit a higher TMR (Table 6), which is likely due to the inflorescences reach and remain in the upper strata the removal of the apical meristem at harvest, as it is of the pasture (Pedreira et al. 2017), in an attempt to considered a tall grass, reaching 1.2 m in height in free facilitate seed dispersal. Zanine et al. (2018), working growth. On the other hand, lower SH also promotes with the same guinea grass (cv. Tanzania) also observed higher TAR, which indicates that there was a balance a higher mean SER in summer than in winter (13.38 vs. between deaths and the appearance of tillers at the lower 4.87 mm/tiller/d) and in pastures under greater SH (50 SH, which contributed to the perenniality of the cultivar. vs. 30 cm) during summer (15.21 vs. 11.54 mm/tiller/d). The inverse relationship between SH and TAR observed Leaf blade length, represented by FLS, is a structural in the study can be explained by the greater light variable that responds to the intensity of defoliation. intensity that reaches the base of the canopies managed Higher values for this variable are associated with under lower SH, in addition to the higher LAR (Sbrissia greater SH, agreeing with the greater leaf sheath length et al. 2010) (Table 4). In theory, the appearance of a new (Volaire et al. 2014). Thus, the distance to be traveled leaf allows the development of a new tiller (Skinner and by the leaf blade inside the pseudostem is greater, which Nelson 1992). results in an extended elongation time and, consequently, Knowledge about seasonal variations in the rates of a longer new leaf (Duru and Ducrocq 2000). However, tiller appearance, mortality and, consequently, survival, this response pattern was observed only during the DS, is important for understanding the mechanisms involved which still provided the lowest FLS as compared with in perenniality and tiller turnover in pastures. During the RS for stubble heights up to 45 cm (Table 5). This the RS, although the pastures had a higher TAR (42.9 reduction in FLS during the DS may be associated with tillers/100 tillers), it was not high enough to compensate the lower average LER observed in this season (Table 4), for lower tiller survival (60.29 tillers/100 tillers) than as these two traits are known to be correlated (Lemaire in the DS (67.53 tillers/100 tillers), and this condition and Chapman 1996). could negatively affect plant persistence and pasture Since it is inversely related to LSR, the number of productivity. On the other hand, during the DS, despite live leaves per tiller decreased in both seasons with the low rate of tiller appearance, their survival was high, increase in SH (Table 5), whereas average LSR increased in an effort to maintain pasture persistence under these (Table 4). This is likely because, as they contained larger climatic conditions. proportions of senescent leaves at the base of the canopy, pastures with greater SH had a higher percentage of dead Herbage accumulation and morphological composition leaves, which resulted in a lower NLL per tiller, as well as the greater SH displaying lower LAR (Table 4). These With increase in SH the plants needed a shorter time effects were also described by De Carvalho et al. (2016) interval and, probably, smaller amounts of reserves to and Silva et al. (2016) in Tifton-85 grass managed under reach 95% LI, i.e. a shorter regrowth period (Silva et different grazing intensities. al. 2019). This led to a decrease in THA as compared Stubble height is a factor that affects tiller density, with the pastures with lower SH, which required a with higher densities being commonly observed in longer interval to intercept 95% of incident light again Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stubble height and season effects on Tanzania grass 265 and thus accumulated more herbage. Other evaluated RS may be related mainly to the effect of herbage harvest factors that also help to explain the decreasing THA in the upper stratum, as highlighted above, as well as the in response to the increase in SH were the reductions greater precipitation occurring during this season. in leaf and tiller appearance rates and tiller density as Long regrowth periods, resulting from low SH, well as increasing leaf senescence, as SH was increased. culminated in undesirable changes in the structure Similarly, Hamilton et al. (2013) reported a reduction of the forage canopy, characterized by an increased in the accumulation of ryegrass and tall-fescue herbage proportion of ST and DeM in the herbage accumulated when SH was increased from 2 to 15 cm. above the stubble. Under grazing conditions this can Only about 22% of THA occurred in the DS, which result in herbage losses due to the amount of material would be due to climatic conditions being favorable for largely rejected by grazing animals, thereby negatively regrowth of the grass in the RS (higher precipitation influencing harvest efficiency and the nutritional value rates, mainly), favoring greater number of harvests in of the produced material. that period, which is consistent with the premise that The observed decrease in the proportion of ST in the tropical pastures in Brazil produce 60‒80% of the total accumulated herbage with increasing SH was due to the herbage mass in the RS and the remaining 20‒40% in the harvest intensities themselves. In tropical pastures, it DS (Pedreira et al. 2005; Fernandes et al. 2014; Oliveira is known that, from a certain point (strata closer to the et al. 2020). This more favorable environment reduced ground level), the stem component starts to represent a the average regrowth period from 80.1 days in the DS much more significant percentage of the stratum. This to 51.2 days in the RS, resulting in more harvests being can affect herbage intake, since intake can be affected possible during the RS. In addition, variables directly by components associated with the architecture and the related to herbage accumulation, such as LAR, LER morphological and botanical composition of the pasture, and SER, were higher in the RS. Our findings confirm which define its structure. Stem is the component that the existence of production seasonality in grasses of the most restricts intake due to the physical barrier it imposes genus Megathyrsus, corroborating the results reported on the grazing process (Laca and Lemaire 2000). Fontes by Luna et al. (2016) and Santos et al. (2016). One must et al. (2014) examined the effect of defoliation intensities also remember that the RS was twice as long as the DS. in Brachiaria grasses and also observed that higher The proportion of leaf blade (LB) in herbage defoliation intensities resulted in a higher proportion of accumulated above the stubble increased with increasing stems in the samples. SH in both seasons, whereas the average proportion This study made it possible to confirm the hypothesis of ST decreased (Table 7). When we increased SH, that interactions between seasons of the year in the we harvested the herbage in the upper strata of the eastern Amazon region and SH affect the structure and canopy, where there is a higher proportion of leaves and production of Tanzania grass. In the DS, Tanzania grass a lower percentage of stems. In practice, to maximize changed its morphogenetic and structural traits to result herbage accumulation, Tanzania grass pastures must in only 22.6% of annual total herbage accumulation be managed at lower SH. However, this accumulated occurring during the DS. The remaining 77% of herbage herbage contains a lower proportion of green leaves and accumulation was due to the beneficial changes in the a higher proportion of stems than forage from a greater morphogenetic and structural traits of the grass provided SH, where there is less THA with a higher proportion of by the favorable climatic conditions occurring in the RS. green leaves and a lower proportion of stems. Therefore, Results from this study lead us to conclude that Tanzania in ideal terms, a compromise should be made in choosing grass should be managed at SH between 35 and 45 cm the SH to balance the amount of accumulated herbage in the DS, corresponding to RP from 61 to 64 days, and with its nutritional value. at SH of 35 cm in the RS, corresponding to RP of 41 The higher proportion of LB observed at the SH of days. This should result in a compromise between yield 15 cm during the DS (Table 7) may be due to the greater and quality of forage produced. Studies to assess the removal of morphological components, which put the outcome of such a strategy seem warranted. plants in an even more adverse condition than is ‘normal’ in the DS. Thus, the plants needed to rebuild their leaf Acknowledgments area and maintain more leaves alive (NLL) to ensure the perenniality of the cultivar. The higher proportion of LB To Coordenação de Aperfeiçoamento de Pessoa de Nível in the herbage accumulated at SH of 55 cm during the Superior (CAPES) for the first author's scholarship. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 266 N.M. Lage Filho, A.R. Lopes, A.C. do Rêgo, F.N. Domingues, C. Faturi, T.C. da Silva, E.P. Cândido and W.L. da Silva References 2013. Stubble height management changes the productivity of perennial ryegrass and tall fescue pastures. Agronomy (Note of the editors: All hyperlinks were verified 9 July 2021). Journal 105:557–562. doi: 10.2134/agronj2012.0293 Insua JR; Garcia SC; Berone GD; Basso B; Utsumi SA. Bahmani I; Thom ER; Matthew C; Hooper RJ; Lemaire G. 2020. Field indicators of leaf nutritive value for perennial 2003. Tiller dynamics of perennial ryegrass cultivars ryegrass and tall fescue pastures under different growing derived from different New Zealand ecotypes: effects of and management conditions. Grass and Forage Science cultivar, season, nitrogen fertiliser, and irrigation. Australian 75:159–168. doi: 10.1111/gfs.12470 Journal of Agricultural Research 54:803–817. doi: 10.1071/ Kohmann MM; Sollenberger LE; Vendramini JMB; Silveira AR02135 ML; Moreno LSB. 2017. Harvest stubble height and Barbosa RA; do Nascimento Jr. D; Vilela HH; da Silva K fertilization affect performance of jiggs and ‘Tifton SC; Euclides VPB; Sbrissia AF; Sousa BML. 2011. 85’ bermudagrasses. Crop Science 57:3352–3359. doi: Morphogenic and structural characteristics of guinea grass 10.2135/cropsci2017.05.0308 pastures submitted to three frequencies and two defoliation Korte CJ; Watkin BR; Harris W. 1982. Use of residual leaf severities. Revista Brasileira de Zootecnia 40:947–954. area index and light interception as criteria for spring- doi: 10.1590/S1516-35982011000500002 grazing management of a ryegrass-dominant pasture. New Brougham RW. 1956. Effects of intensity of defoliation on Zealand Journal of Agricultural Research 25:309–319. doi: regrowth of pasture. Australian Journal of Agricultural 10.1080/00288233.1982.10417892 Research 7:377–387. doi: 10.1071/ar9560377 Laca EA; Lemaire G. 2000. Measuring Sward Structure. In: De Carvalho RM; Santos MER; Carvalho BHR; Carvalho Mannetje Lt; Jones RM, eds. 2000. Field and laboratory CRA; da Silveira JPF; da Costa LKP. 2016. Morphogenesis methods for grassland and animal production research. of marandu palisade grass closer to or farther from cattle CABI Publishing, Wallingford, UK. p. 103–121. doi: feces varies by season. Semina: Ciências Agrárias 37:3231– 10.1079/9780851993515.0103 3242. doi: 10.5433/1679-0359.2016v37n5p3231 Lemaire G; Chapman D. 1996. Tissue flows in grazed plant Duchini PG; Guzatti GC; Ribeiro Filho HMN; Sbrissia AF. communities. In: Hodgson J; Illius AW, eds. The ecology 2013. Tiller size/density compensation in temperate climate and management of grazing systems. CABI Publishing, grasses grown in monoculture or in intercropping systems Wallingford, UK. p. 3–36. doi: 10.2307/3802360 under intermittent grazing. Grass and Forage Science Lima CLD; Difante GS; Basso KC; Emereciano Neto JV; 69:655–665. doi: 10.1111/gfs.12095 Montagner DB; Vasconcelos RIG; Trindade TFM; Veras Duru M; Ducrocq H. 2000. Growth and senescence of the ELL. 2017. Canopy structure and tillering of piatã and successive grass leaves on a tiller. Ontogenic development marandu grasses under two grazing intensities with and effect of temperature. Annals of Botany 85:635–643. sheep. Bioscience Journal 33:135–142. doi: 10.14393/BJ- doi: 10.1006/anbo.2000.1116 v33n1a2017-33943 Euclides VPB; Montagner DB; Difante GS; Barbosa RA; Luna AA; Difante GS; Montagner DB; Emereciano Neto JV; Fernandes WS. 2014. Sward structure and livestock de Araújo IMM; Fernandes LS. 2016. Tillering dynamic performance in guinea grass cv. Tanzania pastures managed and structural characteristics of tropical grasses under by rotational stocking strategies. Scientia Agricola 71:451– cutting management. Bioscience Journal 32:1008–1017. 457. doi: 10.1590/0103-9016-2013-0272 doi: 10.14393/BJ-v32n4a2016-32944 Fernandes FD; Ramos AKB; Jank L; Carvalho MA; Martha Matthew C; Lemaire G; Hamilton NRS; Hernandez-Garay Jr. GB; Braga GJ. 2014. Forage yield and nutritive value A. 1995. A modified field-thinning equation to describe of Panicum maximum genotypes in the Brazilian savannah. size/density relationships for defoliated swards. Annals of Scientia Agricola 71:23–29. doi: 10.1590/S0103- Botany 76:579–587. doi: 10.1006/anbo.1995.1135 90162014000100003 Moura AM; Tomich TR; Pereira LGR; Teixeira AM; Paciullo Fontes JGG; Fagundes JL; Backes AA; Barbosa LT; Cerqueira DSC; Jayme DG; Machado FS; Gomide CAM; Campos ESA; da Silva LM; Morais JAS; Vieira JS. 2014. Herbage MM; Chaves AV; Gonçalves LC. 2017. Pasture productivity accumulation in Brachiaria brizantha cultivars submitted to and quality of Urochloa brizantha cultivar Marandu defoliation intensities. Semina: Ciências Agrárias 35:1425– evaluated at two grazing intervals and their impact on milk 1438. (In Portuguese) doi: 10.5433/1679-0359.2014v35n production. Animal Production Science 57:1384–1391. 3p1425 doi: 10.1071/AN16715 Gomide CAM; Paciullo DSC; Morenz MJF; Costa IA; Lanzoni Nascimento N; West TAP; Börner J; Ometto J. 2019. What CL. 2019. Productive and morphophysiological responses drives intensification of land use at agricultural frontiers of Panicum maximum Jacq. cv. BRS Zuri to timing and in the Brazilian Amazon? Evidence from a decision game. doses of nitrogen application and defoliation intensity. Forests 10:464. doi: 10.3390/f10060464 Grassland Science 65:93–100. doi: 10.1111/grs.12226 Oliveira AB; Pires ASV; Matos Neto U; de Carvalho GGP; Hamilton SA; Kallenbach RL; Bishop-Hurley GJ; Roberts CA. Veloso CM; da Silva FF. 2007. Morphogenesis of Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stubble height and season effects on Tanzania grass 267 tanzaniagrass submitted to fertilization and cut intensities. structural characteristics of Piatã grass in silvopastoral systems Revista Brasileira de Zootecnia 36:1006–1013. (In in the Brazilian savannah. Agriculture, Ecosystems and Portuguese) doi: 10.1590/S1516-35982007000500004 Environment 233:16–24. doi: 10.1016/j.agee.2016.08.026 Oliveira JKS; Corrêa DCC; Cunha AMQ; do Rêgo AC; SAS. 2008. SAS Users’ guide. Version 9.1. Statistical Analysis Faturi C; da Silva WL; Domingues FN. 2020. Effect of System (SAS) Institute Inc., Cary, NC, USA. nitrogen fertilization on production, chemical composition Sbrissia AF; da Silva SC; Sarmento DOL; Molan LK; Andrade and morphogenesis of guinea grass in the humid tropics. FME; Gonçalves AC; Lupinacci AV. 2010. Tillering Agronomy 10:1840. doi: 10.3390/agronomy10111840 dynamics in palisadegrass swards continuously stocked by Paciullo DSC; Gomide CAM; Castro CRT; Mauricio RM; cattle. Plant Ecology 206:349–359. doi: 10.1007/s11258- Fernandes PB; Morenz MJF. 2016. Morphogenesis, 009-9647-7 biomass and nutritive value of Panicum maximum under Skinner RH; Nelson CJ. 1995. Elongation of the grass leaf and different shade levels and fertilizer nitrogen rates. Grass its relationship to the phyllochron. Crop Science 35:4–10. and Forage Science 72:590–600. doi: 10.1111/gfs.12264 doi: 10.2135/cropsci1995.0011183X003500010002x Parsons AJ; Johnson IR; Williams JHH. 1988. Leaf age Silva WL; Ruggieri AC; Reis RA; Silva Sobrinho AG; structure and canopy photosynthesis in rotationally and Malheiros EB. 2016. Effects of postgrazing residue on continuously grazed swards. Grass and Forage Science morphogenetic and structural characteristics of Tifton 85 43:1–14. doi: 10.1111/j.1365-2494.1988.tb02136.x pastures. Semina: Ciências Agrárias 37:2043–2052. doi: Pedreira CGS; Braga GJ; Portela JN. 2017. Herbage 10.5433/1679-0359.2016v37n4p2043 accumulation, plant-part composition and nutritive value Silva WL; Costa JPR; Caputti GP; Lage Filho NM; Ruggieri on grazed signal grass (Brachiaria decumbens) pastures in AC; Reis RA. 2019. Effects of grazing intensity and response to stubble height and rest period based on canopy supplementation strategies on Tifton 85 production and on light interception. Crop and Pasture Science 68:62–73. doi: sheep performance. Small Ruminant Research 174:118– 10.1071/CP16333 124. doi: 10.1016/j.smallrumres.2019.03.015 Pedreira CGS; Rosseto FAA; da Silva SC; Nussio LG; Moreno Tesk CRM; Cavalli J; Pina DS; Pereira DH; Pedreira CGS; LSB; Lima MLP; Leme PR. 2005. Forage yield and grazing Jank L; Sollenberger LE; Pedreira BC. 2020. Herbage efficiency on rotationally stocked pastures of ‘Tanzania-1’ responses of Tamani and Quênia guineagrasses to grazing guineagrass and ‘Guaçu’ elephantgrass. Scientia Agricola intensity. Agronomy Journal 112:2081–2091. doi: 10.1002/ 62:433–439. doi: 10.1590/S0103-90162005000500004 agj2.20189 Pereira LET; Herling VR; Avanzi JC; da Silva SC. 2018. Tesk CRM; Pedreira BC; Pereira DH; Pina DS; Ramos TA; Morphogenetic and structural characteristics of signal grass Mombach MA. 2018. Impact of grazing management in response to liming and defoliation severity. Pesquisa on forage qualitative characteristics: a review. Scientific Agropecuária Tropical 48:1–11. doi: 10.1590/1983- Electronic Archives 11:188–197. doi: 10.36560/1152018667 40632018v4849212 Tilley MS; Heininger RW; Crozier CR. 2019. Measuring tiller Pezzopane JRM; Santos PM; Evangelista SRM; Bosi C; development and mortality in winter wheat under field Cavalcante ACR; Bettiol GM; Gomide CAM; Pellegrino conditions. Agrosystems, Geosciences & Environment GQ. 2016. Panicum maximum cv. Tanzânia: climate trends 2:1–11. doi: 10.2134/age2018.12.0066 and regional pasture production in Brazil. Grass and Forage Volaire F; Barkaoui K; Norton M. 2014. Designing resilient and Science 72:104–117. doi: 10.1111/gfs.12229 sustainable grasslands for a drier future: Adaptive strategies, Santana SS; Brito LF; Azenha MV; Oliveira AA; Malheiros functional traits and biotic interactions. European Journal EB; Ruggieri AC; Reis RA. 2017. Canopy characteristics of Agronomy 52:81–89. doi: 10.1016/j.eja.2013.10.002 and tillering dynamics of Marandu palisade grass pastures Zanine AM; Nascimento Jr. D; Silva WL; Sousa BML; Ferreira in the rainy–dry transition season. Grass and Forage Science DJ; da Silveira MCT; Parente HN; Santos MER. 2018. 72:261–270. doi: 10.1111/gfs.12234 Morphogenetic and structural characteristics of guinea grass Santos DC; Guimarães Jr. R; Vilela L; Pulrolnik K; Bufon pastures under rotational stocking strategies. Experimental VB; França AFS. 2016. Forage dry mass accumulation and Agriculture 54:243–256. doi: 10.1017/S001447971600022 (Received for publication 31 July 2020; accepted 24 June 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):268–279 268 doi: 10.17138/TGFT(9)268-279 Research Paper Effect of sowing rate and date on establishment and growth of Trichloris crinita, a native American pasture grass from arid environments, in the Arid Chaco of Argentina Efecto de la densidad y fecha de siembra en el establecimiento y crecimiento de Trichloris crinita, una gramínea forrajera nativa de ambientes áridos de las Américas, en el Chaco Árido de Argentina DEOLINDO L.E. DOMÍNGUEZ1, PEDRO R. NAMUR2 AND PABLO F. CAVAGNARO3,4 1Instituto de Biología Agrícola Mendoza (IBAM), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Universidad Nacional de Cuyo (UNCuyo), Mendoza, Argentina. mendoza.conicet.gov.ar/portal 2Instituto Nacional de Tecnología Agropecuaria (INTA) E.E.A La Rioja, La Rioja, Argentina. inta.gob.ar/larioja 3Instituto Nacional de Tecnología Agropecuaria (INTA) E.E.A La Consulta, Argentina. inta.gob.ar/laconsulta 4Facultad de Ciencias Agrarias, Universidad Nacional de Cuyo, Mendoza, Argentina. fca.uncu.edu.ar Abstract In arid regions, revegetation with locally adapted native species can improve forage production and help ameliorate soil degradation. We investigated the effects of 3 sowing dates and 3 sowing rates of Trichloris crinita cv. Chamical-INTA, a perennial forage grass native to arid and semi-arid regions, on pasture establishment parameters in the Argentinian Arid Chaco phytogeographical region. Sowing date significantly influenced plant density and soil coverage at the end of the growing season, with the latest sowing date increasing mean plant density and soil coverage by 42‒66% and 16‒38%, respectively, relative to the 1st and 2nd dates. Conversely, the later sowing dates (2nd and 3rd dates) exhibited significantly lower mean values for all plant growth-related traits, i.e. tillers per plant, plant height and percentage of flowering plants. Sowing rate had a strong effect on plant density at the end of the growing season but not on plant growth parameters. Under the conditions of this study, using intermediate sowing densities (7.5 kg seed/ha) and sowing early in the season, when temperatures were still mild, delivered the best results in terms of pasture density and establishment efficacy. Early sowing resulted in a greater percentage of flowering plants and seed set prior to the first winter frosts, which should ensure ongoing establishment of plants in the next wet season. Longer-term studies to examine the survival of plants and possible increase in plant density over time are necessary to determine if this procedure has sustainable benefits for pastures in the area. Keywords: Degraded areas, forage productivity, pasture management, plant density, vegetation recovery. Resumen En regiones áridas y semiáridas, la siembra de especies nativas adaptadas localmente puede contribuir a mejorar la productividad forrajera y atenuar la degradación del suelo. Este trabajo investigó los efectos de tres densidades y tres fechas de siembra en la implantación y el establecimiento de Trichloris crinita cv. Chamical-INTA, una gramínea forrajera nativa de zonas áridas, en la región fitogeográfica del Chaco Árido, Argentina. La fecha de siembra influyó significativamente en la densidad de plantas establecidas y cobertura del suelo al final de la época de crecimiento, con siembras en la tercera fecha mostrando incrementos para estas variables de 42‒66% y 16‒38%, respectivamente, Correspondence: .F. Cavagnaro, Instituto Nacional de Tecnología Agropecuaria (INTA) E.E.A La Consulta, Ex Ruta 40 Km 96, San Carlos, 5567 Mendoza, Argentina. E-mail: cavagnaro.pablo@inta.gob.ar Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Establishment of Trichloris crinita in the Argentinian Arid Chaco 269 comparado con las siembras en la primera y segunda fechas. Contrariamente, las fechas de siembra tardías (segunda y tercera fechas) mostraron valores significativamente menores para todos los parámetros de crecimiento evaluados (p.ej. brotes por planta, altura de planta y porcentaje de floración). La densidad de siembra tuvo un fuerte efecto sobre la densidad final de plantas establecidas, pero no en los parámetros de crecimiento. Las siembras tempranas con densidad intermedia (7.5 kg/ha) tuvieron los mejores resultados en términos de eficacia del establecimiento de la pastura. La siembra temprana resultó en un mayor porcentaje de plantas florecidas al final de la temporada, y las semillas resultantes deberían favorecer el establecimiento de la pastura en la temporada siguiente. A futuro, serán necesarios estudios de largo plazo que monitoreen la supervivencia y densidad de plantas en el tiempo, a fin de determinar si estas prácticas poseen beneficios sostenibles para las pasturas de T. crinita en esta región. Palabras clave: Áreas degradadas, densidad de plantas, manejo de pasturas, productividad forrajera, recuperación de vegetación. Introduction grass native to arid and semi-arid regions of North and South America (Peterson et al. 2007). Under natural Drylands, i.e. arid, semi-arid and dry subhumid regions conditions, it behaves as a typical aestival species, combined, cover nearly 41% of the Earth’s land surface growing whenever soil water is available and the and are home to more than 38% of the total global temperature is above 10 °C (Seligman et al. 1992). In population (Global Land Project 2005; Millennium these dry lands, the species is widely recognized for Ecosystem Assessment 2005). Severe land degradation is its good forage quality, drought tolerance, resistance present on 10‒20% of these lands, affecting ~250 million to trampling and grazing, rapid growth and aggressive people, mainly in developing countries (Millennium competition with other native species (Kozub et al. Ecosystem Assessment 2005), and current expectations 2017). In environments with low water availability, it is are that these estimates will increase over time due to used as forage for range grazing and for restoration of climate change and population growth. degraded rangelands (Passera et al. 1992; Cavagnaro and In arid and semi-arid regions, land desertification, Trione 2007; Guevara et al. 2009). characterized by low fertility and organic matter These arid environments are typical in the north- and concentration in the soil, is widespread, and this situation central-west part of Argentina. The ‘Arid Chaco’ region, is often aggravated by overgrazing by domesticated located in the ‘Chaco’ phytogeographical province, animals (Papanastasis 2009). Different studies have is one of these drylands, presenting an east-to-west estimated that 20‒73% (with a mean of 60%, considering annual rainfall gradient of 250‒550 mm, with 80% of estimates from all studies) of the world’s grazing areas the rainfall occurring in mid-summer (November‒ are moderately to severely degraded (Lund 2007). January) (Morello et al. 1985). In this region, extensive Moreover, loss of perennial grasses in rangelands, often rearing of beef cattle and, to a lesser extent, goats, is accompanied by severe soil erosion and salinity, is a the main productive activity (Rueda et al. 2013), with frequent component of desertification processes in arid pasture grasses being the main feed source for livestock. regions (Waters and Shaw 2003). However, in recent decades, forage production has The traditional strategy to address these degradation decreased steadily as a consequence of land degradation issues has been to reseed the degraded areas with due to overgrazing, thereby altering the landscape to introduced perennial species. However, in recent decades shrublands with extensive areas of bare soil (Blanco et the trend has moved towards the use of native species, al. 2005; Karlin 2013). Under this scenario, applying with a clear recognition of their intrinsic adaptive and revegetation strategies, for example by sowing seeds of ecological value (Waters and Shaw 2003). Selection of locally-adapted native grasses, such as T. crinita, may drought-tolerant species, with adequate seed available, be effective in increasing plant numbers and therefore and the utilization of appropriate and sustainable forage production, as reported in previous studies with management practices, especially with regard to pasture other grass species (Passera et al. 1992; Blanco et al. establishment, are critical for a successful revegetation 2005; Quiroga et al. 2009; Mora et al. 2013). program (Quiroga et al. 2013). Initial seedling growth and establishment are stages Trichloris crinita (Lag.) Parodi [syn. Leptochloa of extreme susceptibility to a wide range of stresses, due crinita (Lag.) P.M. Peterson & N. Snow (Peterson et al. to the small seedling size, and these stages are critical 2012; 2015)] (Chloridoideae, Poaceae) is a perennial for achieving productive pastures (Praat 1995; Skinner Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 270 D.L.E. Domínguez, P.R. Namur, and P.F. Cavagnaro 2005; Bertram 2008). To date, we are unaware of any of flowering plants was estimated on 21 May at the studies to evaluate the effects of sowing density and end of the growing season for all treatments, i.e. 180, date on seedling survival and pasture establishment 150 and 120 days after 1st, 2nd and 3rd sowing dates, using T. crinita. We hypothesized that early sowing of respectively. The percentage of the land surface covered T. crinita using high sowing densities would improve with plants was estimated using photo images taken the implantation and establishment of the pasture in an from above on 6 April 2015, i.e. 135, 105 and 75 days arid environment, such as the Argentinian ‘Arid Chaco’ after the 1st, 2nd and 3rd sowing dates, respectively. phytogeographical region. The objective of this work This date was arbitrarily selected as a point in time, was to investigate how different sowing rates and dates when the pasture appeared to have reached its maximum affect the establishment and growth of T. crinita pasture vegetative growth and the leaves of some plants began to in the year of implantation. senesce. All measurements were performed in selected and standardized areas by using a quadrat of 1 m2 in the Material and Methods center of each experimental unit. Mean daytime temperature (T) and the distri-bution of A field trial was carried out in the Agricultural rainfall were monitored during the experiment. Temperature Experimental Field of the ‘Universidad Nacional de La data were obtained from the agrometeorological station of Rioja Sede Chepes’ (31°20' S, 66°38' W). The experiment the National Institute of Agricultural Technology (INTA), began with the sowing of the grass in the spring of 2014 located at El Portezuelo (EEA INTA La Rioja AER El and finished with the first frost of autumn of 2015. Portezuelo), La Rioja, Argentina, located ~50 km from Three sowing rates were used: 0.25 g/m2, now referred the experimental site. The amount and distribution of to as low sowing rate (LR); 0.75 g/m2, intermediate precipitation were recorded with a rain gauge located on sowing rate (IR); and 1.25 g/m2, high sowing rate (HR); the site of the experiment. they are equivalent to 2.5, 7.5 and 12.5 kg seed/ha, The data were analyzed using mixed linear models respectively, or 300, 900 and 1,500 caryopses/m2. An with a factorial structure, treating sowing date, sowing unseeded plot was used as a Control, with the objective rate and their interactions as fixed effects, while treating to obtain complementary information on the dynamics the 3 blocks as random effects. Different structures of of the natural vegetation, as determined by the seed residual variance were considered, and the best models bank already on site. Three sowing dates were used: 22 were selected using the Akaike (AIC) and Schwarz November 2014 (sowing date 1; after the first rains); (BIC) information criteria (Di Rienzo et al. 2017). All 22 December 2014 (sowing date 2); and 21 January statistical and graphic analyses were performed with 2015 (sowing date 3). Spikelets of Trichloris crinita cv. InfoStat version 2018 software (Di Rienzo et al. 2018). Chamical-INTA with a mean germination rate of 85% The data were expressed as mean ± standard error, and P were sown into a soil with superficial tillage (5 cm-deep values <0.05 were considered significant, using the DGC soil movement with a hand rake). After depositing seeds test (Di Rienzo et al. 2002). on the soil surface, the soil was raked to incorporate the The photographs taken to determine the percentage of seeds and to avoid random dispersal by the wind. The soil the soil surface covered with plants, using canopy area as was classified as typic torriorthent, with the following the criterion, were analyzed with CobCal® 2.0 software characteristics: silt loam texture, low organic matter (Ferrari et al. 2009), which estimates the area or percentage content (1.5% of soil mass), pH of 8.2, conductivity of vegetation coverage based on colorimetric analysis. of 2.12 mS/cm, 0.1% total nitrogen and a C:N ratio of 9.7. A complete randomized block design was used. The Results total trial area was 144 m2 and consisted of 12 treatments of sowing rate × sowing date combinations, with 3 Weather conditions during the experiment replicates, totaling 36 experimental units (plots) of 4 m2 (2 × 2 m) each. Mean day temperature and rainfall throughout the Every 15 days for the following 180 days (until 21 growing season are shown in Figure 1A. May 2015), the following parameters were measured: Mean daily temperature for each vegetative period, density of T. crinita plants and other narrow-leaf i.e. from the relevant sowing date to 21 May 2015, was (grasses) and broad-leaf plants; number of tillers per 23.6 °C ± 4.4; 23.5 °C ± 4.6; and 22.6 °C ± 4.3, for the plant; and plant height of T. crinita. The percentage 1st, 2nd and 3rd sowing dates, respectively. In general, Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Establishment of Trichloris crinita in the Argentinian Arid Chaco 271 A 35 120 30 100 25 80 20 60 15 40 10 5 20 0 0 B 30 180 160 25 140 20 120 100 15 80 10 60 40 5 20 0 0 Nov Dec Jan Feb Mar Apr May Figure 1. A. Time-course variation of mean day temperature (upper graph part with adjusted trend line) and rainfall (histogram in the lower graph part) in the growing season at the site of the experiment. B. Time-course variation of mean day temperature (black line) and mean precipitation (histogram) during the growing season (November‒May) at the site of the experiment for the decade 2011‒2021. Dispersion bars indicate standard deviations. mean temperature increased steadily from the 1st Plant density of T. crinita and other species sowing date (22 November 2014) until early February (6 February 2015), reaching a maximum of 33.6 °C on 10 Both sowing date (P=0.0051) and sowing rate (P<0.0001) January 2015. Additionally, Figure 1B presents weather of T. crinita seed had significant impacts on final plant conditions at the site of the experiment for the last decade. density (FPD) at the end of the experiment, i.e. 135, Total precipitation during the experiment was 548 105 and 75 days after the 1st, 2nd and 3rd sowing dates, mm, with amounts for the 3 growing periods being 450 respectively, but there was a significant (P=0.0142) mm for the 1st sowing date, 338 mm for the 2nd sowing interaction between sowing date and sowing rate (Table date and 289 mm for the 3rd date. In November 2014, 1; Figure 2). The highest values for FPD occurred for the 98 mm of rainfall were registered, with 64 mm received intermediate sowing rate (IR; 0.75 g/m2) treatment at the before the 1st sowing date (21 November 2014). 1st and 3rd sowing dates (P<0.05), with means of 135 and Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Mean day temperature (°C) Mean day temperature (°C) Rainfall (mm) Rainfall (mm) 272 D.L.E. Domínguez, P.R. Namur, and P.F. Cavagnaro 131 plants/m2, respectively. The lowest FPD (22 plants/ days, reaching a maximum plant density (MPD) 45‒60 m2) was for the low sowing rate (LR; 0.25 g/m2) at the 1st days after sowing (DAS), with mean values of 48, 195 sowing date (Figure 2). Among the Control plots, only and 199 plants/m2 for LR, IR and HR, respectively those for the 2nd sowing date presented T. crinita plants, (P<0.05). Plant density then declined steadily until April with a mean FPD of 2 plants/m2. Presumably, these with final populations of 22, 135 and 81 plants/m2 for plants developed from T. crinita seeds that were part of LR, IR and HR, respectively (P<0.05). the natural seed bank in the soil, as this species is native For the 2nd sowing date, there was little emergence at to arid and semi-arid regions of Argentina, including the observation dates before 5 February (45 DAS), when peak Arid Chaco, where the experiment was conducted. plant numbers were recorded (58, 145 and 164 plants/m2 Figure 3 depicts the time-course variation of T. crinita for LR, IR and HR, respectively; P<0.05). A rapid decline plant density throughout the study. At all sowing rates in plant numbers occurred in the following 2 weeks, after for the 1st sowing date, minimal emergence of T. crinita which plant density remained relatively stable until the seedlings was observed at 15 days after sowing, but end of the experiment, with FPDs of 43, 99 and 60 plants/ emergence increased significantly in the following 15 m2 for LR, IR and HR, respectively (P<0.05). Table 1. Effects of sowing date and sowing rate, and their interaction, on final plant density of Trichloris crinita and other narrow- and broad-leaf species, mean number of tillers per plant, plant height and percentage of flowering plants of T.crinita, plus percentage of soil covered with vegetation at the end of the study (last measurement). Plant density at end of study Tillers Plant height % soil % flowering Trichloris crinita Narrow-leaf species Broad-leaf species per plant coverage plants Sowing date 6.8** ns ns 6.1** 55.5*** 4.6* 16.8*** Sowing rate 37.7*** 6.1** 5.1** ns ns 19.5*** ns Date × rate 3.5* ns ns ns ns 3.8** ns Numbers are the F value from ANOVA. Vegetation cover was determined with the software CobCal 2.0 (Ferrari et al. 2009) using digital photographs taken from above. 210 LR IR HR All sowing rates combined 180 150 a a b b A 120 bB b b 90 B 60 b 30 c 0 Sowing date 1 Sowing date 2 Sowing date 3 Figure 2. Plant density at the end of the study (FPD) for Trichloris crinita for each sowing date and sowing rate, i.e. at 180, 150 and 120 days after the 1st, 2nd and 3rd sowing dates, respectively. Low (LR), intermediate (IR) and high sowing rates (HR) correspond to sowing densities of 0.25, 0.75 and 1.25 g seed/m2, respectively. Columns represent means and bars are the standard errors. Lower- case letters indicate significant differences among all different combinations of sowing dates and rates and upper-case letters indicate comparisons among sowing dates considering all sowing rates for a given sowing date combined. Columns with the same letter are not significantly different at P<0.05. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Plant density (plants/m2) Establishment of Trichloris crinita in the Argentinian Arid Chaco 273 280 SD1 / LR SD1 / IR SD1 / HR 240 SD2 / LR SD2 / IR SD2 / HR SD3 / LR SD3 / IR SD3 / HR 200 160 120 80 40 SD1 SD2 SD3 0 Figure 3. Time-course variation of mean density (plants/m2) of Trichloris crinita during the growing season. Arrows for SD1, SD2 and SD3 indicate the 1st (22/11/2014), 2nd (22/12/2014) and 3rd (21/01/2015) sowing dates, respectively. For each sowing date, 3 sowing rates were used: low (LR; 0.25 g seed/m2), intermediate (IR; 0.75 g seed/m2) and high (HR; 1.25 g seed/m2). For the 3rd sowing date, peak plant emergence was sowing rates were combined within sowing dates, tiller detected 15 DAS, with plant densities of 157, 197 and numbers were greater for the first sowing date than for 210 plants/m2 for LR, IR and HR, respectively (P<0.05). the subsequent sowing dates (9.8 vs. 7.3 and 7.1 tillers/ During the next 15 days, plant density decreased rapidly plant for progressive sowings) (Table 1; Figure 5; and then remained relatively stable until the end of the P=0.105). Production of tillers throughout the growing experiment, when plant numbers were 108, 131 and 99 plants/m2 for LR, IR and HR, respectively. 25 While all plots contained some non-T. crinita a All non-L. crinita species plants, with monocots more abundant than dicots 20 (Figure 4), Control plots had higher frequency of these Narrow-leaf species plants (P<0.05) than plots where T. crinita was sown 15 Broad-leaf species (Table 1; Figure 4). The most frequent grasses were: b Aristida adscensionis, Digitaria californica, Cenchrus 10 b ciliaris, Pappophorum caespitosum, Chloris virgata, 5 b Sporobolus pyramidatus and Setaria leucopila, whereas the main broad-leaf species were: Flaveria bidentis, 0 Gomphrena tomentosa, Allionia incarnata and Solanum Control LR IR HR elaeagnifolium. Figure 4. Plant density (plants/m2) for non-Trichloris crinita species at the end of the growing season for Control (0 g Production of tillers seed/m2) and the sowing rates low (LR; 0.25 g seed/m2), intermediate (IR; 0.75 g seed/m2) and high (HR; 1.25 g seed/ For T. crinita, while tiller density at the end of the m2). Columns represent mean value ± standard error. Columns study was not significantly affected by sowing rate for all non-T. crinita species with the same letter are not on any sowing date (P>0.05), when data for different significantly different at P<0.05. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Plant density (plants/m2) Plant density (plants/m2) 274 D.L.E. Domínguez, P.R. Namur, and P.F. Cavagnaro season revealed that, despite the lack of significant was evident for the 1st sowing date, with IR presenting variation among sowing rate treatments at the end of the tallest plants, followed by HR and LR (Supplementary the experiment, there was a tendency for higher tillers Figure S2 A), whereas no clear differences among the per plant (TPP) in the IR treatment than in LR and HR sowing rate treatments were observed for the 2nd and treatments for the 3 sowing dates (Supplementary Figure 3rd sowing dates (Supplementary Figures S2 B and C). S1). For the 1st sowing date, mean TPP remained at 1 during the first 60 DAS, then increased steadily to reach 80 an overall mean of 9.8 TPP at the end of the experiment. a LR TPP for the IR treatment was higher than for LR and 70 a A IR HR treatments during most of the growing season 60 HR (Supplementary Figure S1 A). In general, similar TPP a All sowing rates combined variation patterns were observed for all sowing dates, 50 with different sowing rate treatments not varying much b b b B from each other, except for a tendency in favor of IR 40 b b b B for the early part of the growing season (Supplementary 30 Figures S1 B and C). 20 14 10 a LR 12 a a IR 0 A HR Sowing date 1 Sowing date 2 Sowing date 3a 10 All sowing rates combined a Figure 6. Height of Trichloris crinita plants at the end of the a B growing season for each sowing date and rate. Low (LR; 0.25 8 a a a B g seed/m2), intermediate (IR; 0.75 g seed/m2) and high sowing rates (HR; 1.25 g seed/m2) were used on the1st (22/11/2014), 6 2nd (22/12/2014) and 3rd (21/01/2015) sowing dates. Columns represent mean value ± standard error. Lower-case 4 letters indicate differences among the different combinations of sowing dates and rates, while upper-case letters indicate 2 overall differences between sowing dates. Columns with the same letter are not significantly different (P>0.05). 0 Sowing date 1 Sowing date 2 Sowing date 3 Vegetation cover Figure 5. Tillers per plant in T. crinita at the end of the growing season, as influenced by sowing date and sowing rate. Low (LR; The percentage of soil coverage was significantly affected 0.25 g seed/m2), intermediate (IR; 0.75 g seed/m2) and high sowing by sowing rate (P<0.0001), sowing date (P=0.0214) and rates (HR; 1.25 g seed/m2) were used on the 1st (22/11/2014), 2nd (22/12/2014) and 3rd (21/01/2015) sowing dates. Columns their interaction (P=0.0094) (Table 1; Figure 7). Soil represent mean value ± standard error. Lower-case letters indicate coverage was highest in the IR treatment at all sowing differences among the different combinations of sowing dates and dates but differences were significant for only the 1st rates, while upper-case letters indicate differences among sowing (84.6% coverage) and 3rd (77.3% coverage) sowing dates combining all sowing rates. Columns with the same letter dates. The lowest soil coverage was observed in the LR are not significantly different (P>0.05). plots sown early in the season. Plant height Percent of flowering plants Plant height was significantly influenced by sowing date The percentage of flowering plants of T. crinita at the (P<0.0001) but not by sowing rate (Table 1). Overall, end of the growing season was affected by sowing date T. crinita plants sown earliest (1st date) were 62‒75% (P=0.0001) but not sowing rate (Table 1). Early-sown taller than those sown on the 2nd and 3rd dates (Figure plots (1st date) had significantly higher percentage of 6). The time-course variation for this trait, as influenced flowering plants (75‒88%) than plots sown on the 2nd by sowing date and rate, is presented in Supplementary (32‒37%) and 3rd (40‒48%) dates, regardless of the Figure S2. Variation among the sowing rate treatments sowing rates used (Figure 8). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tillers per plant Plant height (cm) Establishment of Trichloris crinita in the Argentinian Arid Chaco 275 100 Discussion a LR IR a 80 HR b To the best of our knowledge the present study examined, b All sowing rates combined A for the first time, the effects of sowing rate and date on b A establishment of T. crinita pastures. Although T. crinita b b B 60 has been used for decades, as forage and for revegetation b purposes, it was not until recently that cultivars of this c species were developed and seed was made available 40 in sufficient quantities for widespread use. To date, 6 T. crinita cultivars can be found at the Argentinian Cultivar 20 Registry of the National Institute of Seeds (INASE). Thus, the use of a reliable and genetically-certified seed source of T. crinita cv. Chamical-INTA, the cultivar used in this 0 study, provides confidence in the genetic homogeneity of Sowing date 1 Sowing date 2 Sowing date 3 the plant material used in this study. Figure 7. Soil coverage (%) of established Trichloris crinita plants at the end of the growing season for each sowing date and Effects of sowing rate rate. Low (LR; 0.25 g seed/m2), intermediate (IR; 0.75 g seed/ m2) and high sowing rates (HR; 1.25 g seed/m2) were used on The plots in which intermediate sowing rates (IR) were the 1st (22/11/2014), 2nd (22/12/2014) and 3rd (21/01/2015) used had higher plant density and soil coverage at the sowing dates. Columns represent mean value ± standard error. end of the growing season than plots of low (LR) and Lower-case letters indicate differences among the different high sowing rates (HR) and this was consistent for all combinations of sowing dates and rates, and upper-case letters sowing dates (Figures 2 and 7). Competition among indicate overall differences between sowing dates. Columns individuals may explain the effects of sowing rate on with the same letter are not significantly different (P>0.05). final plant density in the experimental plots. According to Harper (1977), Burton et al. (2006) and Brooker et 100 a LR al. (2008), the density of established plants increases a a A IR with sowing rate up to a maximum, which depends on 80 HR the characteristics of the crop and its interaction with All sowing densities combined the environment, and then decreases due to excessive 60 b competition for resources among the plants. This b b behavior was clearly observed throughout the growing b b Bb season for all sowing rates and dates analyzed (Figure 40 B 3). In other words, all treatments revealed an initial peak in plant density followed by a decrease, presumably due 20 to the high competition among the plantlets, but the IR treatment showed a much higher initial peak than the LR treatment, and a more gradual decline after the peak than 0 the HR treatment, resulting in the IR treatment having Sowing date 1 Sowing date 2 Sowing date 3 the highest density of established plants and highest soil coverage at the end of the experiment. Thus, in terms Figure 8. Percentage of Trichloris crinita plants flowering of density of a T. crinita pasture under agroecological at the end of the growing season for each sowing date and conditions similar to those used in the present study, our rate. Low (LR; 0.25 g seed/m2), intermediate (IR; 0.75 g seed/ results suggest an optimum sowing rate of 0.75 g seed/ m2) and high sowing rates (HR; 1.25 g seed/m2) were used m2, which translates to 7.5 kg seed/ha. on 1st (22/11/2014), 2nd (22/12/2014) and 3rd (21/01/2015) sowing dates. Columns represent mean value ± standard error. The present study was conducted during the growing Lower-case letters indicate differences among the different season (from spring to autumn), in which implantation combinations of sowing dates and rates, and upper-case letters and establishment of the pasture takes place. After the indicate overall differences between sowing dates. Columns non-vegetative winter period, and re-sprouting of the with the same letter are not significantly different (P>0.05). plants in spring, we evaluated plant density, considering Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Flowering plants (%) Soil coverage (%) 276 D.L.E. Domínguez, P.R. Namur, and P.F. Cavagnaro only those plants that were visually sprouted and had growth period of these early-sown plants, as compared resumed their vegetative growth; these densities were with those sown later in the season, as evidenced from similar to those recorded at the end of the previous season comparisons of the time-course variation for this trait (data not presented). This suggests that at the end of the among different sowing dates (Supplementary Figure first growing season the pasture was fully established. S2). Similarly, overall production of tillers per plant A significant negative relationship was found was significantly higher for plots sown on the 1st date between the density of established T. crinita plants than for plots sown on the 2nd and 3rd dates (Figure 5), and the density of all other (non-T. crinita) plants [the and monitoring of this trait revealed similar variation Pearson’s correlation coefficient (r) value was -0.66; in tillering as for plant height (Supplementary Figure P<0.0001], which probably resulted from the natural S1). Thus, a significant positive correlation (r=0.66; seed bank in the soil and included other grasses and P=0.0002) was observed between T. crinita plant height broad-leaf species. IR and Control plots had the lowest and TPP. This positive association is not surprising, as and highest density of non-T. crinita plants, respectively both traits are considered sub-components of the same (Figure 4). The number of non-T. crinita plants in Control overall vegetative growth. plots increased slowly and gradually during the growing The strong effect of sowing date on percentage of season, whereas in plots sown with T. crinita the number plants of T. crinita flowering at the end of the growing of plants of other species did not vary throughout the season, with the early planting presenting, on average, experiment (data not presented). Altogether, these data a 0.9‒1.4-fold increase compared with later sowings, clearly reflect the competitive pressure of T. crinita was likely a function of plants from the earlier sowing plants on other native grasses and dicots in determining being more mature following a longer growth period, the final pasture composition of plant species. This is thereby allowing a larger number of plants to reach the in agreement with Quiroga et al. (2009) and Blanco et reproductive stage, as compared with plants sown later. al. (2013), who proposed that the sowing of seeds in Flowering is affected by a combination of physiological semi-arid environments can effectively increase the age and day length, and older plants experienced a plant density of desirable species in the short term. In longer growth period and more hours of day-length than the particular case of T. crinita, sowing of this species in younger plants. dry regions of South America, where much of the land is extensively degraded, may contribute to the recovery of Influence of climatic factors the ground cover and increase forage grass availability. Mean annual rainfall in the Arid Chaco region varies Effects of sowing date from 450 mm in the east to 200 mm in the west, with 80% occurring between November and March (Morello Sowing date had significant effects, although smaller et al. 1985). Thus, the climatic conditions in the present than those of sowing rate (Table 1), on plant density study were rather atypical for this region, given that 548 of T. crinita at the end of the growing season, with the mm of rainfall occurred during the experiment (Figure latest sowing date presenting higher overall mean plant 1). For this reason, the results of the present study are numbers (for all densities combined) than the earlier suitable for seasons and/or regions with this amount of sowing dates (Figure 2). These results are likely due rainfall. Therefore, there is a need to verify the results to the more favorable (lower) temperatures and higher of this experiment in other seasons with different water rainfall received during the establishment phase of the conditions. plantlets after the 3rd sowing date, as compared with the From the beginning of the experiment until 6 January warmer and drier conditions affecting plantlets from the 2015, when the highest plant density for the 1st sowing 1st and 2nd dates (Figure 1). date was reached, a total of 219 mm of rainfall was As was to be expected, vegetative growth parameters received, representing 40% of the total rainfall received. of individual plants, such as plant height and number This provided plants from the 1st sowing date with of tillers per plant (TPP), as well as the percentage of adequate soil moisture in the early stages and throughout flowering plants at the end of the growing season, were the growing season. strongly influenced by sowing date. The greater plant January was the warmest month, presenting the heights in plots from the earliest sowing date, regardless highest monthly mean temperature (27.4 °C) (Figure 1). of sowing rate, were likely due to the longer vegetative This may explain why seedlings from the 2nd sowing Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Establishment of Trichloris crinita in the Argentinian Arid Chaco 277 date (22 December 2014) did not emerge until the first is the measurement of basal area of the stand and this week of February, i.e. 45 DAS, which coincided with method should be utilized as well in future studies. In the emergence of seedlings from the 3rd sowing date addition, longer-term studies are warranted to determine (Figure 3). January also coincided with the beginning of the survival of the species over time and its contribution the decline in plant density for the 1st sowing date. This to the pasture in terms of plant numbers and dry matter suggests that the abundant rainfall received before January yields. Since the rainfall received during the study was at favored the initial establishment of seedlings from the the top end of the range of annual precipitation expected first sowing, whereas the subsequent high temperatures in the region, further studies are needed to determine the in January were less favorable for seedling establishment outcome of sowings where more typical rainfall patterns and general plant growth, i.e. plant height and TPP, in and amounts were received. plots from the 2nd and 3rd sowing dates. Nonetheless, comparably higher plant density and soil coverage were Acknowledgments observed in plots from the 3rd sowing date (Figures 2 and 7), presumably due to the lower temperatures and rainfall The authors acknowledge the National University received in this period (Figure 1). of La Rioja (UNLaR) ‘Sede Chepes’ for providing the necessary infrastructure for the development of Conclusions the experimental work, and the National Institute of Agricultural Technology (INTA) E.E.A. La Rioja, for To the best of our knowledge, this study is the first to providing seed of Trichloris crinita cv. Chamical-INTA. document the effects of sowing date and sowing rate on the establishment and plant growth parameters References of Trichloris crinita, a good-quality pasture grass native to arid regions. Results showed that, under the (Note of the editors: All hyperlinks were verified 14 July 2021). environmental and management conditions of the present work, sowing date significantly influenced plant density Bertram N. 2008. Respuesta de agropiro alargado (Thinopyrum and soil coverage at the end of the growing season, with ponticum) a la densidad de siembra y al arreglo espacial: the latest (3rd) sowing date increasing both parameters. Desarrollo del área foliar, estructura poblacional e intercepción lumínica en el año de implantación. M.Sc. On the other hand, sowing date had the opposite effect Thesis. Universidad Nacional de Mar del Plata, Balcarce, on individual plant growth-related parameters evaluated, Argentina. hdl.handle.net/20.500.12123/7145 with the earliest (1st) sowing resulting in greater Blanco L; Biurrrun F; Ferrando C. 2005. Niveles de production of tillers, plant height and percentage of degradación de la vegetación del Chaco Árido, una flowering plants at the end of the season. Since the study aproximación cuantitativa a partir de imágenes satelitales. terminated with the first frost, earlier flowering from INTA EEA La Rioja, Argentina. the first sowing should ensure more seed was produced Blanco L; Namur PR; Ferrando C; Rettore A; Namur P; Ávila to support plant populations over the subsequent year. R; Molina J; Orionte E. 2013. Vegetation trend after roller Sowing rate had a strong significant effect on the final chopping and native grass seeding in La Rioja. Revista de plant density of the pasture (and therefore on soil la Facultad de Agronomía UNLPam 22(Sp2):17–22. (In Spanish) bit.ly/3mN3aJB coverage) but not on individual plant growth parameters, Brooker RW; Maestre FT; Callaway RM; Lortie CL; Cavieres with intermediate sowing densities yielding the densest LA; Kunstler G; Liancourt P; Tielbörger K; Travis JMJ; plant stands at the end of the growing season. Thus, using Anthelme F; Armas C; Coll L; Corcket E; Delzon S; ~0.75 g seed per m2 (7.5 kg/ha) and sowing early in the Forey E; Kikvidze Z; Olofsson J; Pugnaire F; Quiroz CL; season, when temperatures were still mild, delivered the Saccone P; Schiffers K; Seifan M; Touzard B; Michalet best results in terms of pasture density and establishment R. 2008. Facilitation in plant communities: the past, the efficacy, which is likely due to the more favorable present, and the future. Journal of Ecology 96:18–34. initial growing conditions in which the plants were doi:10.1111/j.1365-2745.2007.01295.x able to take advantage of available water, nutrients and Burton CM; Burton PJ; Hebda R; Turner NJ. 2006. light resources and develop sufficiently to combat later Determining the optimal sowing density for a mixture of native plants used to revegetate degraded ecosystems. periods of high temperatures. We used aerial photos to Restoration Ecology 14:379–390. doi: 10.1111/j.1526- assess ground cover, i.e. area of canopy was the criterion. 100X.2006.00146.x A more effective assessment of amount of soil coverage Cavagnaro JB; Trione SO. 2007. Physiological, morpholog- Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 278 D.L.E. Domínguez, P.R. Namur, and P.F. Cavagnaro ical and biochemical responses to shade of Trichloris Administración de Parques Nacionales. Buenos Aires, crinita, a forage grass from the Arid Zone of Argentina. Argentina. Journal of Arid Environments 68:337–347. doi: 10.1016/j. Papanastasis VP. 2009. Restoration of degraded grazing lands jaridenv.2006.06.004 through grazing management: Can it work? Restoration Di Rienzo JA; Casanoves F; Balzarini MG; Gonzalez L; Ecology 17:441–445. doi: 10.1111/j.1526-100X.2009.00 Tablada M; Robledo CW. 2018. InfoStat versión 2018. 567.x Centro de Transferencia InfoStat, FCA, Universidad Passera CB; Borsetto O; Candia RJ; Stasi CR. 1992. Shrub Nacional de Córdoba, Argentina. infostat.com.ar control and seeding influences on grazing capacity in Di Rienzo JA; Guzmán AW; Casanoves F. 2002. A multiple- Argentina. Journal of Range Management 45:480–482. comparisons method based on the distribution of the root doi: 10.2307/4002906 node distance of a binary tree. Journal of Agricultural, Peterson PM; Columbus JT; Pennington SJ. 2007. Biological, and Environmental Statistics 7:129–142. doi: Classification and biogeography of new world grasses: 10.1198/10857110260141193 Chloridoideae. Aliso: A Journal of Systematic and Di Rienzo JA; Macchiavelli RE; Casanoves F. 2017. Modelos Evolutionary Botany 23:580–594. doi: 10.5642/aliso. lineales generales mixtos: Aplicaciones en InfoStat. Grupo 20072301.43 InfoStat, Córdoba, Argentina. hdl.handle.net/11554/8691 Peterson PM; Romaschenko K; Arrieta YH. 2015. A molecular Ferrari DM; Pozzolo OR; Ferrari HJ. 2009. CobCal, software phylogeny and classification of the Eleusininae with a new para estimación de cobertura vegetal. Instituto Nacional de genus, Micrachne (Poaceae: Chloridoideae: Cynodonteae). Tecnología Agropecuaria, EEA Concepción del Uruguay, Taxon 64:445–467. doi: 10.12705/643.5 Entre Ríos, Argentina. Peterson PM; Romaschenko K; Snow N; Johnson G. 2012. Global Land Project. 2005. Science plan and implementation A molecular phylogeny and classification of Leptochloa Strategy. IGBP (International Geosphere Biosphere (Poaceae: Chloridoideae: Chlorideae) sensu lato and related Program) Report No. 53/International Human Dimensions genera. Annals of Botany 109:1317–1329. doi: 10.1093/ Programme Report No. 19, IGBP Secretariat, Stockholm. aob/mcs077 Guevara JC; Grünwaldt E; Estevez OR; Bisigato AJ; Blanco Praat JP. 1995. Row spacing and seeding rate interactions LJ; Biurrun FN; Ferrando CA; Chirino CC; Morici EF; in perennial ryegrass and tall fescue swards established Fernández B; Allegretti LI; Passera CB. 2009. Range by direct drilling (No-Tillage). Ph.D. Thesis. Massey and livestock production in the Monte Desert, Argentina. University, Palmerston North, New Zealand. hdl.handle. Journal of Arid Environments 73:228–237. doi: 10.1016/j. net/10179/3076 jaridenv.2008.02.001 Quiroga RE; Blanco LJ; Orionte EL. 2009. Evaluación de Harper JL. 1977. Population Biology of Plants. Academic estrategias de rehabilitatión de pastizales áridos. Ecologia Press, London, UK. Austral 19:107–117. bit.ly/2UKRQBQ Karlin MS. 2013. Historia de ocupación y uso de recursos. Quiroga RE; Fernandez RJ; Golluscio RA; Blanco LJ. 2013. In: Karlin MS; Karlin UO; Coirini RO; Reati GJ; Zapata Differential water-use strategies and drought resistance in RM, eds. El Chaco Árido. Editorial Encuentro, Argentina. Trichloris crinita plants from contrasting aridity origins. p. 155–174. Plant Ecology 214:1027–1035. doi: 10.1007/s11258-013- Kozub PC; Cavagnaro JB; Cavagnaro PF. 2017. Exploiting 0228-4 genetic and physiological variation of the native forage Rueda CV; Baldi G; Verón SR; Jobbágy EG. 2013. Human grass Trichloris crinita for revegetation in arid and semi- appropriation of primary production in the Dry Chaco. arid regions: An Integrative Review. Grass and Forage Ecologia Austral 23:44–54. (In Spanish). hdl.handle. Science 73:257–271. doi: 10.1111/gfs.12337 net/11336/4329 Lund HG. 2007. Accounting for the world’s rangelands. Seligman NG; Cavagnaro JB; Horno ME. 1992. Simulation Rangelands 29:3–10. doi: 10.2111/1551-501X(2007)29 of defoliation effects on primary production of a warm- [3:AFTWR]2.0.CO;2 season, semiarid perennial-species grassland. Ecological MEA (Millennium Ecosystem Assessment). 2005. Ecosystems Modelling 60:45–61. doi: 10.1016/0304-3800(92)90012-4 and human well-being: Desertification synthesis. World Skinner RH. 2005. Emergence and survival of pasture species Resources Institute, Washington, DC. sown in monocultures or mixtures. Agronomy Journal Mora S; Cabral DR; Rosales Mercado IA. 2013. Establishment 97:799–805. doi: 10.2134/agronj2004.0211 of pasto plumerito (Trichloris crinita Parodi) in the Waters CM; Shaw NL. 2003. Developing native grass seed sowing year. Revista de la Facultad de Agronomía industries for revegetation in Australia and the Western UNLPam 22:119–122. (In Spanish) hdl.handle.net/20. United States: a contrast in production and adoption. 500.12123/4089 In: Proceedings of the VIIth International Rangelands Morello JH; Protomastro J; Sancholuz L; Blanco C. 1985. Congress, Durban, South Africa. bit.ly/2UbYmS3 Estudio macroecológico de los Llanos de La Rioja. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Establishment of Trichloris crinita in the Argentinian Arid Chaco 279 Supplementary Data 12 12 12 A B LR IR HR C 10 10 10 8 8 8 6 6 6 4 4 4 2 2 2 0 0 0 Supplementary Figure S1. Time-course var8i0ation of the mean number of tillers per p8la0nt produced by Trichloris crinita plants during the growing season, for the 1st (22 November 2014) (A), 2nd (22 December 2014) (B), and 3rd (21 January 2015) (C) sowing dates. Data points for low sowing rate (LR, 0. 25 g seed/m2) are indicated by clear squares, intermediate sowing rate (IR, 0.75 g seed/ m2) by gray circles and high sowing rate (HR, 1.25 g seed/m2) by black triangles. 80 80 80 A B LR IR HR C 70 70 70 60 60 60 50 50 50 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 Supplementary Figure S2. Time-course variation for mean plant height of Trichloris crinita during the growing season for the 1st (22 November 2014) (A), 2nd (22 December 2014) (B), and 3rd (21 January 2015) (C) sowing dates. Data points for low sowing rate (LR, 0. 25 g seed/m2) are indicated by clear squares, intermediate sowing rate (IR, 0.75 g seed/m2) by gray circles and high sowing rate (HR, 1.25 g seed/m2) by black triangles. (Received for publication 19 October 2020; accepted 6 July 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Plant height (cm) Tillers per plant 22-Nov-14 22-Nov-14 7-Dec-14 07-Dec-14 22-Dec-14 22-Dec-14 6-Jan-15 06-Jan-15 21-Jan-15 21-Jan-15 5-Feb-15 05-Feb-15 20-Feb-15 20-Feb-15 7-Mar-15 07-Mar-15 22-Mar-15 22-Mar-15 6-Apr-15 06-Apr-15 21-Apr-15 21-Apr-15 6-May-15 06-May-15 21-May-15 21-May-15 22-Dec-14 22-Dec-14 6-Jan-15 06-Jan-15 21-Jan-15 21-Jan-15 5-Feb-15 05-Feb-15 20-Feb-15 20-Feb-15 7-Mar-15 07-Mar-15 22-Mar-15 22-Mar-15 6-Apr-15 06-Apr-15 21-Apr-15 21-Apr-15 6-May-15 06-May-15 21-May-15 21-May-15 21-Jan-15 21-Jan-15 5-Feb-15 5-Feb-15 20-Feb-15 20-Feb-15 7-Mar-15 7-Mar-15 22-Mar-15 22-Mar-15 6-Apr-15 6-Apr-15 21-Apr-15 21-Apr-15 6-May-15 6-May-15 21-May-15 21-May-15 Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):280–291 280 doi: 10.17138/TGFT(9)280-291 Research Paper Biomass production and nutritional properties of promising genotypes of Tithonia diversifolia (Hemsl.) A. Gray under different environments Producción de biomasa y propiedades nutricionales de genotipos destacados de Tithonia diversifolia (Hemsl.) A. Gray bajo diferentes condiciones ambientales JULIÁN ESTEBAN RIVERA1*, TOMÁS E. RUÍZ2, JULIAN CHARÁ1, JUAN FLORENCIO GÓMEZ-LEYVA3 AND ROLANDO BARAHONA4 1Centro para la Investigación en Sistemas Sostenibles de Producción Agropecuaria – CIPAV, Cali, Colombia. cipav.org.co 2Instituto de Ciencia Animal, San José de las Lajas, La Habana. Cuba. ica.edu.cu 3Laboratorio de Biología Molecular, TecNM-Instituto Tecnológico de Tlajomulco, México. ittlajomulco.edu.mx 4Universidad Nacional de Colombia, Sede Medellín, Colombia. medellin.unal.edu.co Abstract Tithonia diversifolia is a shrub with excellent forage characteristics that has shown a wide genetic and phenotypic diversity. The objective of this study was to determine the biomass production and nutritional quality of seven genotypes of T. diversifolia with outstanding characteristics for ruminant nutrition, to analyze the Genotype x Environment (GxE) interaction of biomass production and to compare the performance of these genotypes with grasses offered normally in tropical conditions. For the GxE interaction the AMMI and SREG models were used, and evaluations were made in three environments. In the GxE analysis, the interaction was significant and effects of the environment on biomass productivity were observed with differences among genotypes. In the three environments, the high content of crude protein (28.89 g/100 g of DM), the low fiber content (30.95 g of neutral detergent fiber - NDF/100 g of DM) and the high percentages of in vitro degradation of DM for all the genotypes was adequate to be offered to ruminants. This study identified superior genotypes of T. diversifolia with good productive and adaptive performance for high-altitude and low-altitude zones with low fertility soils. Keywords: Forage productivity, genetic diversity, GxE interaction, multivariate analysis, nutrient supply, SREG model. Resumen Tithonia diversifolia es un arbusto con excelentes características forrajeras que ha mostrado una amplia diversidad genética y fenotípica. El objetivo de este estudio fue determinar la producción de biomasa y la calidad nutricional de siete genotipos de T. diversifolia con características sobresalientes para la nutrición de rumiantes, analizar la interacción Genotipo x Ambiente (GxE) de la producción de biomasa y comparar el desempeño de estos genotipos con gramíneas ofrecidas normalmente en condiciones tropicales. Para la interacción GxE se utilizaron los modelos AMMI y SREG, y se realizaron evaluaciones en tres ambientes. En el análisis GxE, la interacción fue significativa y se observaron efectos del ambiente sobre la productividad de la biomasa con diferencias entre genotipos. En los tres ambientes, la composición química fue adecuada para ser ofrecida a los rumiantes. Cabe destacar el alto contenido de proteína bruta (28.89 g/100 g de MS), el bajo contenido de fibra (30.95 g de fibra detergente neutra - FDN/100 g de MS) y los altos porcentajes de degradación in vitro de la MS para todos los genotipos. Se puede concluir que existen genotipos superiores de T. Correspondence: Julian Esteban Rivera, Centro para la Investigación en Sistemas Sostenibles de Producción Agropecuaria – CIPAV. Carrera 25 # 6 – 62 Cali, Colombia Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Biomass production and nutritional properties of Tithonia diversifolia genotypes 281 diversifolia con capacidad de tener un buen rendimiento productivo y adaptativo para zonas de alta y baja altitud con suelos de baja fertilidad. Palabras clave: Análisis multivariado, diversidad genética, interacción GxE, modelo SREG, oferta de nutrientes, productividad de forraje. Introduction environment interaction (GxE) in agricultural crops, as well as their stability and adaptation to different Silvopastoral systems (SPS) have proven to be a suitable environments (Bhartiya et al. 2017). alternative to increase production efficiency and reduce This study was carried out to measure biomass the environmental impact of livestock systems (Jose et production and nutritional quality of different al. 2019). One of the shrub species used in Colombia and T. diversifolia provenances under different environment Mexico as a component of the SPS is Tithonia diversifolia and management conditions and compare their chemical (Hemsl.) A. Gray. T. diversifolia has excellent forage composition with the nutritional quality of two grasses characteristics with high biomass productivity and usually offered in tropical conditions to identify stable nutritional value, and also wide phenotypic variation, genotypes with potential as feed. Variables measured which provides an opportunity to identify and select included those associated with nutrient content and with outstanding genotypes capable of achieving higher agronomic performance. productivity (Ruiz et al. 2013). It has been grown under different edaphoclimatic conditions and exhibits Materials and Methods a high degree of genetic diversity and variability in its agronomic properties, nutrient content, and adaptability Genotypes evaluated (Holguín et al. 2015). Although good productive responses are frequently Seven genotypes of T. diversifolia previously identified reported in grazing ruminants receiving T. diversifolia- by Rivera et al. (2017) were included in this study (Table supplemented diets, greater benefits could be possible 1). These genotypes were previously selected based on by carrying out evaluation and identification of different the Dice dissimilarity index and the weighted forage genotypes to select cultivars with desirable characteristics potential index (WFPI) (Holguín et al. 2015) from a (Holguín et al. 2015; Rivera et al. 2018). One area group of 30 populations collected in Colombia and of interest is to identify elite germplasm adapted to Mexico. The selected genotypes presented outstanding marginal conditions such as acid and low-fertility soils performance in biomass production, number of stems of the tropics and subtropics. and overall growth (Rivera et al. 2017). In successful forage selection programs, the influence of environmental factors on plant productivity and Location of experiment nutritional quality is the basis for identifying more efficient cultivars for animal nutrition and economic The study was carried out in three environments performance of farms (Schultze-Kraft et al. 2018). In (environment 1: Tropical lowlands with fertilization; recent years, the AMMI (additive main effects and environment 2: Tropical lowlands without fertilization; multiplicative interaction) and SREG (sites regression) environment 3: Tropical highlands without fertilization) models have been used to determine the genotype- during 2018 and 2019. Environments 1 and 2 were Table 1. Location of the collection sites of the genotypes evaluated Identification Municipality Department masl Precipitation Temperature Coordinates(m) (mm/year) (ºC) N W Genotype 1 Granada Meta 326 2410 27.2 3°53'42.06'' -74°11'48.72'' Genotype 2 Belén de los Andaquíes Caquetá 232 2840 23.5 01°14'49.2" -75°46'28.3" Genotype 3 La Paz Cesar 623 1220 27.2 10°14'18.66'' -73°6'21.539'' Genotype 4 Santa Rosa de Cabal Risaralda 1870 2610 16.2 4°52'39.430" -75°34'58.563" Genotype 5 Encino Santander 1608 870 22.3 6°11'26.52'' -73°8' 49.139'' Genotype 6 Charalá Santander 1383 2130 23.4 6°16'46.8'' -73°9'49.499'' Genotype 7 Manizales Caldas 2159 2545 16.3 5°0'51.538" -75°33'58.302" masl: meters above sea level Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 282 J.E. Rivera, T.E. Ruíz, J. Chará, J.F. Gómez-Leyva, R. Barahona located in Meta, Colombia (3°47'21"N, 73°49'16"W) m2), dew point (ºC), wind speed (m/s) and THSW index at 530 masl in a region classified as belonging to the (Thermal sensation due to wind), relative humidity and tropical humid forest life zone (bh-T) (Holdridge irradiance (instantaneous solar radiation) were recorded 1978). Environment 3 was located in Caldas, Colombia using a Vantage Pro 2TM (Davis ®) weather station. (5°0'45"N, 75°25'47"W) at an altitude of 2300 masl, which corresponds to a lower montane moist forest Nutritional and agronomic variables (bh-MB) (Holdridge 1978). Morphological variables were measured during four Experimental design harvests on five plants per plot, taking 2 measurements during the rainy and 2 during the dry season in each An experimental area of 642 m2 was established in each environment. A uniformity cut at 10 cm height was environment. In order to ensure genetic homogeneity, made 4 months after planting on the whole plot. For the planting material of all genotypes were produced in environment 3, harvests were made every 60 days by the laboratory using explant clonal reproduction. Each cutting 5 randomly selected plants per plot at 10 cm one of the 642 m2 areas consisted of 21 plots (4 x 5.5 height when plants had reached an average plant height m) with three replicates of 36 plants for each genotype of 109.3 cm, and in environments 1 and 2, harvests of T. diversifolia planted in a randomized complete were made every 40 days by cutting 5 randomly block design. Neighboring pastures of Urochloa selected plants per plot at 10 cm height when plants brizantha cv. Marandú in environment 1 and 2 and had reached an average plant height of 95.5 and 69.1 Cenchrus clandestinus in environment 3 were used as cm, respectively. The cutting regimes were established the reference for comparison with local feed supply. based on the harvesting times usually used in each zone The level of fertilization used in environment 1 was in for the predominant forage species (Urochloa brizantha accordance with the extraction of nutrients of 40 days cv. Marandú and Cenchrus clandestinus respectively). old T. diversifolia plants (Botero et al. 2019). These Nutritional traits were determined at the Animal nutrients were applied by fertilizing with urea (46% N), Nutrition Laboratory, the Colombian Corporation for ammonium phosphate (DAP) [(NH4)2HPO4; 46% P2O5, Agricultural Research (AGROSAVIA) by near-infrared 18% N] and potassium chloride (KCl, 60% K2O) at a spectroscopy (NIRS) using two chemometric tools fertilizer rate of 16.22 g/plant (324 kg/ha), 2.15 g/plant (GLOBAL and LOCAL) using a scanning VIS/NIR (43 kg/ha) and 4.89 g/plant (98 kg/ha) respectively. spectrometer (Foss NIRSystems model 6500) and the WinISI 4.7.0 software (Ariza-Nieto et al. 2018). The Soil analysis nutritional variables were determined using samples from one harvest in the dry season and one in the rainy Three soil samples were taken from 20–30 cm depth season (Table 2). in each block at the beginning of the experiment. The following chemical and physical variables were Genotype-by-environment interaction and data analysis measured: pH, electrical conductivity (E.C.) (dS/m), bulk density (g/cc), organic matter (%), texture, For the GxE interaction analysis, AMMI (Mandel 1971) exchangeable acidity (mg/kg), exchangeable calcium and SREG site regression analysis (Yan et al. 2000) (mg/kg), Iron (mg/kg), Manganese (mg/kg), Copper models were used. Material stability was measured (mg/kg), Zinc (mg/kg), Boron (mg/kg), Phosphorus using the Shukla's Stability Variance. The analyses (mg/kg) and Cation exchange capacity (meq/100g). The were performed in RStudio using the "ggbiplot", different determinations were carried out at AGRILAB "GGEBiplotGUI" (GGEplot) and "agricolae" soil laboratory (Bogotá, Colombia). libraries (R Core Team 2019). Tukey's contrast test (0.05 significance level) was used when significant Environmental conditions differences between means were detected, and when the data groups did not meet the conditions for a During the experimental period, precipitation (mm), parametric analysis, the Kruskal-Wallis and Mann- temperature (ºC), humidity (%), solar radiation (W/ Whitney tests were applied. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Biomass production and nutritional properties of Tithonia diversifolia genotypes 283 Table 2. Morphological and nutritional variables Variables Measurement method Morphological variables Plant height (PlantH) Measured using a tape measure from the base of the main stem to the flag leaf. Stem diameter (StemDiam) Measured using a vernier caliper at a height of 15 cm. The average of two randomly selected stems was used. Leaf-stem ratio (Leaf:Stem) Calculated from the fresh weight of stems and green leaves at harvest. Number of branches (Bran, stems Measured by manual counting per plant at harvest. with leaves) Leaf area (LeafAre) Mean of ten randomly collected leaves from each plant collected and analyzed in ImageJ® 1.47v software. Green forage per plant (GreenF) Fresh weight (g) of green leaves and small stems with diameters of less than 5 mm taken as a mean of 5 plants. Dry forage per plant (DW) Determined after drying the green forage for 72 hours in a forced-air oven at 65 °C (g/100 g). Survival (Surv) Calculated by the difference between the number of plants planted and the final number at harvest. Presence of pests or diseases Scored in each of the plots by observing for one minute the presence of pests causing evident damage to the plant material. If pests were present, damage was rated from 1 to 3, with 1 being low damage and 3 being severe damage. Nutritional variables Dry Matter (DM) Crude protein (CP) Ether extract (EE) Neutral detergent fiber (NDF) Acid detergent fiber (ADF) Determined using NIRS FOSS 6500 LOCAL and GLOBAL models WinISI 4.7.0 software Total digestible nutrients (TDN) (Ariza-Nieto et al. 2018). in vitro DM degradability (IVDMD) Gross energy (GE) Net energy for lactation (NEL) Calcium (Ca) Determined using AA and UV-VIS spectrophotometry. Based on methods NTC 5151 Phosphorus (P) (ICONTEC 2003) and NTC 4981 (ICONTEC 2001) respectively. Results m2, and accumulated precipitation was 905.9 mm. The cumulative rainfall during the rainy season was 673.6 Soil analysis mm and 232.3 mm during the dry season. The soils in all sites were acidic with different levels of Nutritional and agronomic variables fertility (Table 3). Measurements of morphological and agronomic Environmental conditions variables found in the three environments are presented in Table 4. During the evaluation period, the pest or For environments 1 and 2, average temperature was 25.1 disease damage was minimum (level 1) and occurred in ± 1.3 ºC, relative humidity was 77.8 ± 9.4%, average environments 1 and 2 due to the presence of Acromyrmex dew point was 20.6 ± 1.17 ºC, wind speed was 0.68 spp. and Atta spp. The incidence of these ant attacks was ± 0.2 m/s, average THSW index was 27.7 ± 1.7 ºC, not associated to a specific genotype of T. diversifolia. solar radiation was 478 ± 48.3 W/m2, and accumulated The variables LeafAre, Leaf:Stem ratio, and Bran precipitation was 1119 mm. During the rainy season the presented a positive significant correlation with the cumulative rainfall was 922.6 mm and during the dry production of DW with Pearson coefficients of 0.86, 0.89 season it was 195.4 mm. For environment 3, average and 0.78, respectively. Significant differences between temperature was 15.3 ± 0.85 ºC, relative humidity was genotypes were found in each environment and there was 87.4 ± 5.4%, average dew point was 13.2 ± 0.77 ºC, wind also an effect of season in most variables. In environments speed was 0.42 ± 0.13 m/s, average THSW index was 1 and 2, genotypes 7 and 5 had the highest growth. In 15.5 ± 1.24 ºC, solar radiation was 249.9 ± 40.54 W/ environment 3, genotypes 4 and 7 had the highest growth Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 284 J.E. Rivera, T.E. Ruíz, J. Chará, J.F. Gómez-Leyva, R. Barahona rates. In addition, there were significant differences in rainy season was 1.4 times that of the low precipitation variables such as plant height, leaf size, stems per plant season, and the use of fertilizer increased DW production and leaf-stem ratio that were 1.5, 1.38, 1.67 and 1.63 times 1.9 times on average, and 2.3 times during the rainy higher in fertilized plants, respectively. season. The genotypes with best tolerance to the dry Table 5 presents the results of the chemical analyses season represented by the smallest decrease in biomass of T. diversifolia forage samples and evaluated grasses. production compared to the rainy season were 3, 1 and 2. In In environments 1 and 2 some differences were observed environment 3, the genotypes with the highest production among T. diversifolia genotypes (CP, IVDMD and P), but were 4 and 7 with DW yield of 152.6 and 128.9 g/plant compared with the U. brizantha pasture, all genotypes respectively, despite showing greater yield variability in show higher nutrient supply than this grass. In addition, this environment. The lowest performing genotypes were despite not relevant finding a difference in the nutritional 1, 3 and 6. In this site, the genotypes decreased their yield traits between environments 1 and 2, the greater growth on average by 13.5% as an effect of the dry season with observed in genotypes 5 and 7 (Table 4) allowed them genotypes 5 and 6 having the least reduction. to offer significantly more nutrients in terms of g per In the analysis of variance of the AMMI model plant, compared to the other genotypes. The season had (Table 6), genotypes, environments and GxE interaction an effect on all parameters except NDF and GE content. presented significant differences for DW production per In environment 3 there were differences between plant (Table 6). T. diversifolia genotypes and with the C. clandestinus In Figure 2 (left) the genotypes that were collected grass commonly used in the highland tropics of from similar environments are associated with better Colombia. CP, EE, NDF, ADF, TDN, IVDMD and performance in that environment. In addition, the NEL had differences between genotypes. The season genotype ranking graph (right) shows that the genotypes also influenced the nutritional traits of the genotypes closest to the center point are the best in all environments evaluated (Table 5). (7, 5 and 6). Genotype-by-environment interaction Performance stability throughout environments In environments 1 and 2, the genotypes with the highest According to Shukla's stability index, the most DW yield were 7 (106.5 g per plant) and 5 (89.7 g per plant), stable genotypes were 7 and 5, due to their relatively and the genotypes with lowest DW were 1, 4 and 3 with an high productive performance across environments. average of 65.8, 68.7 and 73.4 g/plant, respectively. Figure GxE interaction was also found in the survival of 1 shows the graphic representation of all genotypes in genotypes during the experimentation period (Figure each environment according to its DW production. The 3). The average survival at the end of the experiment genotypes at the most extreme points and close to the blue in environments 1 and 2 was 82.3% with the effect of arrows are the best performing (Figure 1). fertilization (p=0.0068). In environment 3, the average In environments 1 and 2, the production of DW in the survival rate was 65.8%. Table 3. Chemical and physical characteristics of the soils Characteristic Environment 1 Environment 2 Environment 3 pH 4.72 (±0.03) 4.68 (±0.03) 5.45 (±0.08) Electrical conductivity (dS/m) 0.06 (±0.01) 0.06 (±0.01) 0.17 (±0.02) Bulk density (g/cc) 1.49 (±0.04) 1.47 (±0.05) 0.99 (±0.01) Organic matter (%) 1.68 (±0.35) 1.45 (±0.30) 8.16 (±0.1) Texture Loam-Clay-Sandy Loam-Clay-Sandy Loam Interchangeable acidity (mg/kg) 202 (±26.8) 203 (±18.2) 41.4 (±10.4) Exchangeable calcium (mg/kg) 209 (±82.6) 178 (±65.4) 426 (±98.2) Iron (mg/kg) 374 (±78.2) 374 (±75.1) 202 (±69.2) Manganese (mg/kg) 6.97 (±3.16) 6.30 (±2.31) 18.2 (±8.55) Copper (mg/kg) 0.86 (±0.24) 0.76 (±0.14) 2.80 (±0.70) Zinc (mg/kg) 0.63 (±0.17) 0.50 (±0.15) 18.00 (±5.2) Boron (mg/kg) 0.13 (±0.03) 0.14 (±0.01) 0.06 (±0.01) Phosphorus (mg/kg) 6.03 (±1.95) 5.60 (±1.67) 13.7 (±2.08) Cation exchange capacity (meq/100g) 3.40 (±0.42) 3.27 (±0.22) 3.46 (±0.59) Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 285 J.E. Rivera, T.E. Ruíz, J. Chará, J.F. Gómez-Leyva, R. Barahona Table 4. Morphological and agronomic variables of the genotypes of T. diversifolia Environment / Genotype PlantH StemDiam Bran Leaf:Stem LeafAre Environment 1 Gen1 62.1bc 8.75bc 9.64ab 0.89 46.2c Gen2 58.3c 8.04c 8.03b 0.98 44.9c Gen3 63.8bc 8.93bc 8.72ab 0.96 54.2abc Gen4 59.7c 8.43c 8.42ab 0.93 48.7bc Gen5 75.1ab 10.1ab 9.75ab 0.96 59.4ab Gen6 66.1bc 9.99ab 8.75ab 0.91 49.6bc Gen7 84.5a 10.9a 11.1a 0.92 66.2a p- value <0.001* <0.001* 0.013* 0.929 <0.001* SEM 2.07 0.28 0.48 0.04 2.95 Season effect <0.001* 0.004* <0.001* <0.001* <0.001* Dry season 60.6 9.71 6.55 0.71 36.73 Rainy season 72.6 8.72 11.6 1.14 67.92 Environment 2 Gen1 98.6c 11.3b 11.4ab 0.78 74.01b Gen2 97.9c 11.4b 11.1ab 0.83 71.3b Gen3 90.4c 12.1ab 10.1b 0.86 75.3b Gen4 92.3c 11.1b 12.6ab 0.83 74.9b Gen5 117.1ab 13.2a 12.4ab 0.81 84.4b Gen6 103.3bc 12.3ab 10.4ab 0.8 79.8b Gen7 122.4a 12.5ab 13.9a 0.8 102.6a p- value <0.001* 0.002* 0.023* 0.578 0.001* SEM 3.89 0.18 0.64 0.01 4.8 Season effect <0.001* 0.029* <0.001* 0.494 <0.001* Dry season 82.3 11.7 8.23 0.808 52.8 Rainy season 123.5 12.3 15.2 0.824 107.4 Environment 3 Gen1 75.5d 8.59c 13.84ab 0.75a 69.1c Gen2 100abc 10.35ab 16.3ab 0.68ab 84.1a Gen3 86.2c 9.85b 12.3b 0.74ab 72.3ab Gen4 111a 10.98a 18.9a 0.65b 82.8ab Gen5 101.5ab 9.97b 16.3ab 0.7ab 59.8c Gen6 95bc 9.71b 15.2ab 0.72ab 71bc Gen7 111.7a 10.23ab 18.3ab 0.66ab 77.9ab p- value 0.003* <0.001* <0.001* 0.0105* <0.001* SEM 2.62 0.13 0.62 0.01 2.01 Season effect <0.001* 0.014* <0.001* <0.001* <0.001* Dry season 89.1 9.74 12.7 0.65 65.9 Rainy season 104.2 10.2 18.9 0.74 78.9 Figure 1. Dry biomass productivity GGEplot representation of the genotypes of T. diversifolia in the three environments. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Biomass production and nutritional properties of Tithonia diversifolia genotypes 286 Table 5. Nutritional traits (g/100 g of DM; Mcal/kg of DM) of the genotypes Environment 1 Genotypes Season U. brizantha p- value SEM Characteristic Gen1 Gen2 Gen3 Gen4 Gen5 Gen6 Gen7 Rainy Dry Genotype Season DM 15.0 16.0 15.5 15.9 15.5 15.6 15.3 14.3 16.8 22.7 0.252 <0.001* 0.223 CP 33.9ab 32.8abc 32.8abc 31.8bc 34.3a 31.4c 34.1ab 35.2 30.8 10.9 0.004* <0.001* 0.488 Ash 15.0 15.0 15.4 14.9 15.0 15.0 14.7 15.4 14.6 7.9 0.657 0.001* 0.126 EE 1.98 2.00 2.10 2.19 2.10 2.20 1.93 1.49 2.65 1.59 0.662 <0.001* 0.102 NDF 31.2 31.7 30.8 32.2 31.3 31.8 31.3 30.6 32.3 65.3 0.951 0.016* 0.349 ADF 15.4 15.3 14.9 15.1 16.4 15.0 15.9 16.2 14.7 48.5 0.922 0.042* 0.371 Ca 1.60 1.41 1.56 1.47 1.33 1.68 1.49 1.27 1.74 0.41 0.503 <0.001* 0.061 P 0.44 0.46 0.45 0.44 0.45 0.42 0.44 0.51 0.38 0.20 0.933 <0.001* 0.012 TDN 76.0 75.1 75.2 74.4 76.0 74.1 75.9 76.4 74.1 51.7 0.265 <0.001* 0.349 IVDMD 82.8ab 81.9ab 82.1ab 81.2ab 82.9a 80.9b 82.7ab 83.6 80.5 57.7 0.013* <0.001* 0.375 GE 4.29 4.28 4.28 4.27 4.30 4.26 4.29 4.29 4.26 4.09 0.849 0.053 0.008 NEL 1.75 1.73 1.73 1.71 1.75 1.70 1.74 1.75 1.69 1.15 0.284 <0.001* 0.009 Environment 2 Genotypes Season U. brizantha p- value SEM Characteristic Gen1 Gen2 Gen3 Gen4 Gen5 Gen6 Gen7 Rainy Dry Genotype Season DM 16.7 16.3 16.0 16.5 16.2 16.2 16.2 14.9 17.7 22.7 0.705 <0.001* 0.238 CP 29.6 27.8 30.3 27.3 28.9 30.5 29.8 31.1 27.2 10.9 0.118 <0.001* 0.470 Ash 14.8 14.2 14.9 14.3 15.2 15.3 14.8 15.3 14.3 7.94 0.222 0.002 0.165 EE 2.13 2.27 2.05 2.31 1.99 2.08 1.97 1.46 2.76 1.59 0.221 <0.001* 0.112 NDF 30.4 31.0 30.2 30.7 31.9 30.8 31.1 30.9 30.7 65.3 0.882 0.708 0.331 ADF 14.0 12.8 13.6 11.2 13.9 13.4 13.2 15.3 13.6 48.5 0.226 <0.001* 0.456 Ca 1.96 1.99 1.90 1.96 1.83 1.88 1.80 1.79 2.1 0.41 0.651 0.009* 0.039 P 0.37c 0.37bc 0.41abc 0.38bc 0.44a 0.44a 0.42ab 0.47 0.34 0.20 0.009* <0.001* 0.012 TDN 73.6 72.0 73.6 72.0 72.5 73.9 73.4 73.9 71.9 51.7 0.181 <0.001* 0.298 IVDMD 80.3 78.5 80.3 78.5 79.1 80.6 80.0 80.7 78.5 57.7 0.181 <0.001* 0.320 GE 4.24 4.20 4.21 4.17 4.20 4.23 4.25 4.21 4.23 4.09 0.271 0.234 0.010 NEL 1.69 1.65 1.69 1.65 1.66 1.70 1.68 1.7 1.64 1.15 0.138 <0.001* 0.007 Environment 3 Genotypes Season C. clandestinus p- value SEM Characteristic Gen1 Gen2 Gen3 Gen4 Gen5 Gen6 Gen7 Rainy Dry Genotype Season DM 16.3 16.6 17.3 17.1 17.3 17.5 17 16.7 17.3 18.1 0.058 0.005* 0.127 CP 27.1ab 28.7a 27.1ab 29.5a 28.9a 25.4b 26.2b 27.5 27.1 20.9 0.005* 0.879 0.292 Ash 14.7 15.1 14.4 15.1 14.6 15.1 14.4 14.5 15 12.1 0.246 0.029* 0.106 EE 1.84c 2.34a 1.93bc 2.19ab 2.15abc 1.93bc 2.04abc 1.59 1.89 2.31 0.002* 0.095 0.029 NDF 32.1a 30.9ab 29.8bc 28.5c 29.7bc 31.6a 31.1ab 28.8 32.2 44.5 0.006* 0.008* 0.377 ADF 2.52 2.16 2.42 2.23 2.32 2.27 2.22 2.08 2.53 0.65 0.055 <0.001* 0.051 Ca 0.38 0.4 0.36 0.41 0.38 0.36 0.38 0.35 0.41 0.35 0.382 <0.001* 0.006 P 71.1bc 71.9abc 71.4abc 73.2a 72.5ab 70.1c 70.9bc 71.6 71.5 63.6 0.001* 0.802 0.202 TDN 77.6bc 78.5abc 77.9abc 79.8a 79.1ab 76.5c 77.4bc 78.2 78.1 67.6 0.001* 0.797 0.218 IVDMD 4.15 4.17 4.13 4.21 4.19 4.14 4.16 4.12 4.2 4.09 0.429 0.001* 0.012 GE 1.62bc 1.64abc 1.63abc 1.67a 1.65ab 1.60c 1.62bc 1.64 1.63 1.34 0.002* 0.801 0.005 NEL 1.62bc 1.64abc 1.63abc 1.67a 1.65ab 1.60c 1.62bc 1.64 1.63 1.34 0.002* 0.801 0.005 DM: dry matter; CP: crude protein; Ash: ashes; EE: ether extract; NDF: neutral detergent fiber; ADF: acid detergent fiber; Ca: calcium; P: phosphorus; TDN: total digestible nutrients; IVDMD: in vitro DM degradability; GE: gross energy; NEL: net energy of lactation; SEM: standard error of the mean; * Different letters in the same row denotes statistical difference according to the Tukey test (p <0.05). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Biomass production and nutritional properties of Tithonia diversifolia genotypes 287 Table 6. GxE interaction AMMI model. Analysis of variance for the dry matter production Df Sum Sq Mean Sq F value Variance (%) Pr(>F) Environment 2 72305 36152 98.5 53.3 0.0002 *** Rep (Environment) 6 2201 367 1.23 1.62 0.294 Genotype 6 32820 5470 18.4 24.3 2.77E-14 *** Environment x Genotype 12 28418 2368 7.98 20.9 2.97E-10 *** Residuals 99 29360 297 Figure 2. GGEplot dry matter yield of the T. diversifolia genotypes (left). Genotypes ranking with respect to the ideal genotype (right) Gen7 Gen6 Gen5 Gen4 Gen3 Gen2 Gen1 0 10 20 30 40 50 60 70 80 90 100 Environment 1 Environment 2 Environment 3 Figure 3. Tithonia diversifolia survival rate (%) in three environments in two experimental sites. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Genotypes 288 J.E. Rivera, T.E. Ruíz, J. Chará, J.F. Gómez-Leyva, R. Barahona Discussion 2018). For example, genotypes 5 and 7 and genotypes 4 and 7 could be used in low-altitude and high-altitude, According to Rivera et al. (2017), there is wide genetic respectively in conditions similar to those evaluated in diversity in the populations of T. diversifolia evaluated. this study. A total of 105 fragments were amplified, of which 5% Ribeiro et al. (2016), reported high nutrient content were monomorphic and 95% polymorphic. In addition, in T. diversifolia, which can then be employed either the analysis based on the genome proportion (genetic as a supplement to diets based on tropical pastures, structure) of each population showed seven well-defined or as a forage source capable of partially replacing groups. In Mexico, Del Val et al. (2017) evaluated commercial concentrates in ruminant diets. The most 20 materials for feed purposes from eight localities outstanding chemical fractions in T. diversifolia are and obtained a total of 157 bands of which 33 were the high percentages of CP in leaves (>25%), the low monomorphic and 124 polymorphic, indicating 79% fiber contents (NDF and ADF), the acceptable mineral polymorphism. Thus, in the dendrogram performed with contents (Ca and P) and the good degradability values of the Dice coefficient and using the UPGMA classification DM and energy. The results of this study are consistent method, no significant relationship was found among with those reported by researchers such as Ribeiro et 10 samples and only two were similar (at a similarity al. (2016), who identified its use in the feeding of high level of 1.47, 2 groups were observed). Yang et al. production dairy cattle, its nutritional value, as well as its (2012) found great genetic variability in collections of fermentation dynamics. this species obtained from four regions in China and The seven genotypes evaluated presented higher two in Laos. In this research, the mean values of Nei of amounts of CP and lower percentages of fiber (NDF genetic diversity (H) and the Shannon index of diversity and ADF) than that reported by La O et al. (2012), who (I) were 0.2937 and 0.432, respectively, and 84.62% evaluated nine genotypes of T. diversifolia in Cuba and of polymorphism was observed, demonstrating wide found protein and NDF values from 18.3 to 26.4 and from diversity of T. diversifolia materials and conferring it 14.8 to 25.7%, respectively. Likewise, in terms of CP, great adaptation to diverse environments. contents reported in this study are as high or even higher T. diversifolia exhibited adaptation to the climatic than those found in tropical legumes such as Stylosanthes and edaphic environments evaluated since outstanding guianensis (18.2%, Morgado et al. 2009) and Arachis genotypes were observed in each environment. The pintoi (19.7%, Khan et al. 2013). Its degradability, characteristics that could contribute to this adaptability energy, Ca and P content do not limit voluntary intake are the large root volume, that improves the efficiency and nutrient availability at the ruminal level, despite the to obtain nutrients from the soil (Jama et al. 2000), the 60 days regrowth age used in environment 3. Although possibility to associate with different microorganisms, the genotypes at this site had significant differences, they further favoring this property, especially in low fertility all presented a high supply of nutrients. As a result of soils (Rivera et al. 2018) and its genetic diversity (Yang evaluations in the three environments, and two seasons, et al. 2012). The soils in the evaluation sites were diverse it was also possible to determine that the nutritional and could represent a large area of the tropics where soils value of T. diversifolia is maintained under different are characterized by acidity and a range of fertility levels. environmental conditions, presenting a superior nutrient Morphological and agronomic characteristics showed offer than that of tropical pastures. According to Rivera great variability among genotypes and environments, et al. (2015) the inclusion of T. diversifolia in Brachiaria- and several of them (LeafAre, Leaf:Stem ratio and based systems can support an increased number of Bran) had a significant and direct correlation with DW animals per hectare and increase milk production as well production. Some of these traits related to leaf growth as milk quality. could be used to predict the growth of T. diversifolia and Identifying stable genotypes adapted to different therefore employed for the selection of genotypes with conditions and with the ability to achieve high greater productivity and adaptation (Ruiz et al. 2013). In performance in variable environments has been an addition, the variability found can be used strategically ongoing challenge in the study of forage species (Liang et in selection programs and future varietal improvement. al. 2015). The characterization of stable genotypes across This could be carried out comprehensively with multi- different environments is an important task, although criteria evaluations based on adaptability, productivity difficult to achieve due to the frequent influence of GxE and nutritional quality (Holguín et al. 2015, Rivera et al. interactions (Senger et al. 2016). Yan (2002) indicated Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Biomass production and nutritional properties of Tithonia diversifolia genotypes 289 that typically, the environment explains most of the total et al. (2015), found survival rates above 90% for plants yield variation (up to 70% or more), while genotypes and from three establishment methods under highland GxE interaction are generally small. This is specifically conditions after one month of sowing. The differences true for traits like plant yield, as for example found in between the two studies are probably due to the adverse this study. environmental conditions during seedling establishment The SREG chart (Figure 2) identifies the ideal in this research. genotype as the one with a high score on the first axis of the principal component (CP1) that is associated Conclusions with high yields and scores close to zero on the second axis of the principal component (CP2). This is related T. diversifolia has the ability to adapt to different to good stability (Figure 1 and Figure 2) as shown by edaphoclimatic conditions and offer a high amount genotypes 5 and 7. Furthermore, in the GGE Biplot, of nutrients for ruminants. The high percentage of the genotypes located towards the center of the figure CP, the low fiber values and the high percentages of are less representative than those located at the corners energy and degradability of DM are outlined as the or vertices of the polygon, which are considered more most remarkable characteristics in this species. Despite responsive (positively or negatively, genotypes 6, 3 and its wide plasticity, environmental conditions modify 2) to the environmental conditions. Genotypes located in the yield of T. diversifolia genotypes showing GxE sectors of the SREG chart where there are no sites, are interaction and favoring the possibility of identifying considered to have poor performance behavior in most and selecting genotypes with greater productive potential of the sites evaluated (Yan et al. 2001) (Genotypes 1, 2 that are better adapted to specific sites. In this research, and 3). genotypes 5 and 7 were the most outstanding in site 1 In the genotype classification, the graph of the so- and genotype 4 was the most outstanding material in site called "ideal" genotype is shown (Figure 2). An "ideal" 2, while genotypes 5 and 6 showed stability across sites. genotype is one with the highest performance in test environments and stable performance (Yan and Kang Acknowledgements 2002). Although such an "ideal" genotype may not exist in reality, it can be used as a reference for the evaluation The authors wish to thank the Colombian Sustainable of different genotypes. A genotype is more desirable if it Cattle Ranching project, funded by the Government is closer to the "ideal" genotype (Yan and Kang 2002) as of the United Kingdom and the Global Environment was the case of genotypes 7 and 5. Facility, for supporting the field and laboratory work, The production of biomass found in this study was and to El Porvenir and Los Alpes farms for their help lower than those reported by Alonso Lazo et al. (2015), during the experiments and for providing access to their who evaluated four grazing frequencies and different facilities. Thanks also to Minciencias for supporting planting distances in Cuba and found weights between J. Rivera’s Ph.D. studies (National Doctorates 727- 1,400 and 2,300 g of green weight/plant and from 200 2015) and to CIPAV (contract 80740-006-2020) and the to 600 g of dry weight of the entire plant. These weights Colombian Autonomous Fund for Science, Technology were similar to those reported by Gallego et al. (2015) and Innovation Francisco José de Caldas. in Colombia under conditions similar to those given in environment 3, and where Ruiz et al. (2013) recorded References weights of 100 green leaves between 110 and 190 g at 42 days, and between 150 and 240 g at 60 days. Botero et al. (Note of the editors: All hyperlinks were verified 27 August 2021). (2019) found a positive response of T. diversifolia when it was fertilized. The response found by these authors Alonso Lazo J; Achang Fagra G; Tuffi Santos LD; Arruda was greater than that reported in this study (2.5 times Sampaio R. 2015. Productivity performance of Tithonia more DW when T. diversifolia was fertilized). diversifolia in grazing with different rest times during both seasons of year. Livestock Research for Rural Significant differences were found in the survival Development Volume 27, Article#115. (In Spanish). lrrd. rate of the genotypes (Figure 3) for both montane rain org/lrrd27/6/alon27115.html forest and tropical rain forest conditions, evidencing that Ariza-Nieto C; Mayorga OL; Mojica B; Parra D; Afanador- high rainfall in clay loam soils negatively affects the Tellez G. 2018. Use of LOCAL algorithm with near infrared survival of T. diversifolia. For this parameter, Gallego spectroscopy in forage resources for grazing systems in Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 290 J.E. Rivera, T.E. Ruíz, J. Chará, J.F. Gómez-Leyva, R. Barahona Colombia. Journal of Near Infrared Spectroscopy 26(1): hypogaea L.) as an alternate forage source for sheep. 44–52. doi: 10.1177/0967033517746900 Tropical Animal Health and Production 45:849–853. doi: Bhartiya A; Aditya JP; Singh K; Pushpendra; Purwar JP; 10.1007/s11250-012-0297-8 PJ; Agarwal A. 2017. AMMI & GGE biplot analysis of La O O; González H; Orozco A; Castillo Y; Ruiz O; Estrada multi environment yield trial of soybean in North Western A; Ríos F; Gutiérrez E; Bernal H; Valenciaga D; Castro BI; Himalayan state Uttarakhand of India. Legume Research Hernánez Y. 2012. Composición química, degradabilidad 40(2):306–312. doi: 10.18805/lr.v0iOF.3548 ruminal in situ y digestibilidad in vitro de ecotipos de Botero J; Gómez A; Botero M. 2019. Yield, agronomic Tithonia diversifolia de interés para la alimentación de parameters and nutritional quality of Tithonia diversifolia in rumiantes. Revista Cubana de Ciencia Agrícola 46(1):47– response to different fertilization levels. Revista Mexicana 53. redalyc.org/articulo.oa?id=193024313008 de Ciencias Pecuarias 10(3):789–800. doi: 10.22319/rmcp. Liang S; Ren G; Liu, J; Zhao X; Zhou M; McNeil D; Ye G. v10i3.4667 2015. Genotype-by-environment interaction is important Del Val R; Miranda JM; Flores MX; Solorio B; Solorio for grain yield in irrigated lowland rice. Field Crops FJ; Gómez-Leyva JF; González-Palomares S. 2017. Research 180:90–99. doi: 10.1016/j.fcr.2015.05.014 Diversidad genética de Tithonia diversifolia (Hemsl) Gray Mandel J. 1971. A new analysis of variance model for non- de Michoacán: Análisis con marcadores de ADN-SSR. additive data. Technometrics 13(1):1–18. doi: 10.1080/ Reaxion Ciencia y Tecnología Universitaria 4(3):10–14. 00401706.1971.10488751 bit.ly/3zpWhBs Morgado ES; Almeida FQ de; Silva VP; Gomes AVC; Gallego L; Mahecha-Ledesma L; Angulo-Arizala J. 2015. Galzerano L; Torres H; Rodrigues LM. 2009. Digestion Crecimiento y desarrollo de Tithonia diversifolia Hemsl. of carbohydrates of forages in horses. Revista Brasileira A Gray en condiciones de trópico alto. Actas 3° Congreso de Zootecnia 38(1):75–81. (In Portuguese). doi: 10.1590/ Nacional de Sistemas Silvopastoriles - VIII Congreso S1516-35982009000100010 Internacional de Sistemas Agroforestales, Puerto Iguazú, R Core Team. 2019. RStudio: Integrated Development for R. Argentina, 7‒9 Mayo 2015. p. 53–58. RStudio, Inc., Boston, MA, USA. rstudio.com Holdridge LR, eds. 1978. Ecología basada en zonas de Ribeiro RS; Terry SA; Sacramento JP; Silveira SR; Bento CBP; vida. Instituto Interamericano de Cooperación para la Silva EF da; Montovani HC; Gama MAS da; Pereira LGR; Agricultura (IICA), San José, Costa Rica. repositorio.iica. Tomich TR; Mauricio RM; Chaves AV. 2016. Tithonia int/handle/11324/7936 diversifolia as a supplementary feed for dairy cows. Plos Holguín V; Ortíz S; Velasco A; Mora JR. 2015. Evaluación One 11:e0165751. doi: 10.1371/journal.pone.0165751 multicriterio de 44 introducciones de Tithonia diversifolia Rivera JE; Chará J; Gómez-Leyva JF; Ruíz T; Barahona (Hemsl.) A. Gray en Candelaria, Valle del Cauca. Revista R. 2018. Phenotypic variability and phytochemical de la Facultad de Medicina Veterinaria y de Zootecnia composition of Tithonia diversifolia A. Gray for 62(2):57–72. doi: 10.15446/rfmvz.v62n2.51995 sustainable animal production. Livestock Research for ICONTEC (Instituto Colombiano de Normas Técnicas Rural Development 30, Article#200. (In Spanish). lrrd. y Certificación). 2001. Alimentos para animales. org/lrrd30/12/rive30200.html Determinación del contenido de fosforo. Método Rivera JE; Cuartas CA; Naranjo JF; Tafur O; Hurtado EA; espectrofotométrico. NTC 4981. ICONTEC, Bogotá, Arenas FA; Chará J; Murgueitio E. 2015. Effect of an Colombia. bit.ly/2WxHOoo intensive silvopastoral system (iSPS) with Tithonia ICONTEC (Instituto Colombiano de Normas Técnicas diversifolia on the production and quality of milk in the y Certificación). 2003. Alimento para animales. Amazon foothills, Colombia. Livestock Research for Determinación de los contenidos de calcio, cobre, hierro, Rural Development. 27, Article#189. (In Spanish). lrrd. magnesio, manganeso, potasio, sodio y zinc. Método org/lrrd27/10/rive27189.html de espectrometría de absorción atómica. NTC 5151. Rivera JE; Gómez-Leyva JF; Castaño K; Morales JG; ICONTEC, Bogotá, Colombia. bit.ly/2V30RXo Barahona R; Del Val R. 2017. Molecular, chemical and Jama B; Palm CA; Buresh RJ; Niang A; Gachengo C; morphological diversity in Tithonia diversifolia (hemsl.) Nziguehbh G; Amadalo B. 2000. Tithonia diversifolia as Gray for animal feed in Colombia and Mexico. Libro a green manure for soil fertility improvement in western de Actas IX Congreso Internacional Sobre Sistemas Kenya. A review. Agroforestry Systems 49(2):201–221. Silvopastoriles. Manizales, Colombia, 6‒8 September doi: 10.1023/A:1006339025728 2017. p. 249–255. (In Spanish). Jose S; Walter D; Mohan Kumar B. 2019. Ecological Ruiz TE; Torres V; Febles G; Díaz H; González J. 2013. Growth considerations in sustainable silvopasture design and performance of ecotypes of Tithonia diversifolia according management. Agroforestry Systems 93:317–331. doi: to morphological components. Livestock Research for 10.1007/s10457-016-0065-2 Rural Development 25, Article#154. (In Spanish). lrrd.org/ Khan MT; Khan NA; Bezabih M; Qureshi MS; Rahman lrrd25/9/ruiz25154.html A. 2013. The nutritional value of peanut hay (Arachis Schultze-Kraft R; Rao IM; Peters M; Clements RJ; Bai C; Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Biomass production and nutritional properties of Tithonia diversifolia genotypes 291 Liu G. 2018. Tropical forage legumes for environmental trial data. Crop Science 41(3):656–663. doi: 10.2135/ benefits: An overview. Tropical Grasslands-Forrajes cropsci2001.413656x Tropicales 6(1):1–14. doi: 10.17138/tgft(6)1-14 Yan W; Hunt LA; Sheng Q; Szlavnics Z. 2000. Cultivar Senger E; Martin M; Dongmeza E; Montes JM. 2016. Genetic evaluation and mega environment investigation based variation and genotype by environment interaction in on the GGE biplot. Crop Science 40(3):597–605. doi: Jatropha curcas L. germplasm evaluated in different 10.2135/cropsci2000.403597x environments of Cameroon. Biomass and Bioenergy Yan W; Kang MS. 2002. GGE Bi-plot Analysis: A graphical 91:10–16. doi: 10.1016/j.biombioe.2016.04.017 tool for breeders, geneticists, and agronomists. CRC Press. Yan W. 2002. Singular value partitioning in bi-plot analysis Boca Raton, FL, USA. doi: 10.1201/9781420040371 of multi-environment trial data. Agronomy Journal Yang J; Tang L; Guan Y; Sun W. 2012. Genetic diversity of 94(5):990–996. doi: 10.2134/agronj2002.0990 an alien invasive plant Mexican Sunflower (Tithonia Yan W; Cornelius PL; Crossa J; Hunt LA. 2001. Two diversifolia) in China. Weed Science, 60(4):552–557. doi: types of GGE biplots for analyzing multi-environment 10.1614/WS-D-11-00175.1 (Received for publication 11 February 2021; accepted 18 August 2021; published 30 September 2021) © 202# Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):292–299 292 doi: 10.17138/TGFT(9)292-299 Research Paper Evaluation of ten perennial forage grasses for biomass and nutritional quality Evaluación de biomasa y calidad nutricional en diez gramíneas forrajeras perennes MULISA FAJI1, GEZAHAGN KEBEDE1, FEKEDE FEYISSA2, KEDIR MOHAMMED1, MULUNEH MINTA1, SOLOMON MENGISTU1 AND ASCHELEW TSEGAHUN1 1Ethiopian Institute of Agricultural Research, Holetta Agricultural Research Center, Holetta, Ethiopia. eiar.gov.et 2Ethiopian Institute of Agricultural Research, Addis Ababa, Ethiopia. eiar.gov.et Abstract A study was carried out to evaluate 10 perennial forage grass accessions from 4 species for herbage dry matter yield and nutritional quality at Holetta Agricultural Research Center. The evaluated grass species and varieties were one Desho grass (Pennisetum) variety Kulumsa, four Urochloa decumbens (ILRI-14721, ILRI-14720, ILRI-13205 and ILRI-10871), three Urochloa ruziziensis (ILRI-14813, ILRI-14774 and ILRI-13332) and two Setaria sphacelata (ILRI-143 and ILRI- 6543) accessions. Plant height and forage dry matter yield were significantly affected by accession over years, during the establishment and production phases. Combined analysis indicated that the tested accessions varied significantly for plant height with the Setaria accessions taller than the other tested species. Combined data analysis revealed that forage dry matter yield significantly varied according to species and Desho grass (variety Kulumsa) was higher in dry matter yield than the other grasses tested. Fiber contents (NDF, ADF and ADL) were significantly influenced by accession. Crude protein yield differed among the accessions and Desho grass had higher crude protein, followed by U. decumbens 13205, U. decumbens 14721 and S. sphacelata 6543. Based on dry matter yield and crude protein U. decumbens 13205, U. ruziziensis 13332, S. sphacelata 6543 and Desho grass (var. Kulumsa) are recommended as alternative forage grasses for the study area and similar agro-ecologies. Keywords: Desho grass, forage yield, Urochloa, Setaria, crude protein. Resumen Se llevó a cabo un estudio para evaluar 10 accesiones de gramíneas forrajeras perennes de 4 especies para determinar el rendimiento de materia seca y la calidad nutricional del forraje en el Centro de Investigación Agrícola de Holetta. Las especies y variedades de gramíneas evaluadas fueron una pasto Desho (Pennisetum) variedad Kulumsa, cuatro accesiones de Urochloa decumbens (ILRI-14721, ILRI-14720, ILRI-13205 e ILRI-10871), tres de Urochloa ruziziensis (ILRI-14813, ILRI-14774 e ILRI -13332) y dos de Setaria sphacelata (ILRI-143 e ILRI-6543). La altura de la planta y el rendimiento de materia seca del forraje se vieron afectados significativamente por la accesión a lo largo de los años, durante las fases de establecimiento y producción. El análisis combinado indicó que las accesiones probadas variaron significativamente la altura de la planta en las accesiones de Setaria, siendo más altas que las otras especies probadas. El análisis de datos combinados reveló que el rendimiento de materia seca del forraje varió significativamente según la especie y el pasto Desho (variedad Kulumsa) fue mayor en rendimiento de materia seca que los otros pastos evaluados. El contenido de fibra (NDF, ADF y ADL) se vio significativamente influenciado en cada accesión. En cuanto a rendimiento de proteína cruda el pasto Desho fue el mayor, seguido por U. decumbens 13205, U. decumbens 14721 y S. sphacelata 6543. Basado en el rendimiento de materia seca y proteína cruda U. decumbens 13205, U. ruziziensis Correspondence: Mulisa Faji, Ethiopian Institute of Agricultural Research, Holetta Agricultural Research Center, P. O. Box 31 Holetta, Ethiopia. E-mail: mulisa.faji2016@gmail.com Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Evaluation of perennial grasses in Ethiopia 293 13332, S. sphacelata 6543) y pasto Desho (var. Kulumsa) se recomiendan como pastos forrajeros alternativos para el área de estudio y condiciones agroecológicas similares. Palabras clave: proteína cruda, Pennisetum, rendimiento forrajero, Setaria, Urochloa. Introduction yield potential (Rodrigues et al. 2014). U. ruziziensis also produces abundant roots which contribute to The central highland of Ethiopia is characterized by a the collection of water, soil aggregation and aeration crop-livestock mixed farming system. Livestock is an (Kluthcouski et al. 2004). Recent studies indicate that integral component of most of the agricultural activities adoption of U. brizantha (Hochst. ex A. Rich.) R. D. in the country. The share of the livestock subsector Webster [previously named as Brachiaria brizantha in the national economy is estimated to be 12–16% of (Hochst. ex A. Rich.)] cultivars as cut-and-carry fodder the total Gross Domestic Product (GDP), 30–35% of for dairy cattle have increased milk production on the agricultural GDP (LMA 1999), 19% of the export participating farms in Kenya by 15‒40% (Schiek et al. earnings (FAO 2004) and 31% of the total employment 2018). Similarly, use of the Urochloa hybrid Mulato II (Feleke 2003). Although Ethiopia has a large livestock fodder in dairy and beef enterprises in Rwanda enabled population (CSA 2016), the productivity of livestock is a 30% increase in milk production and a 20% increase low with the major hindrances being shortage of feed in meat production (CSB 2016). resources in terms of quantity and quality (Demeke et Setaria sphacelata (Schumach.) Stapf & C. E. Hubb. al. 2017). To combat these nutritional constraints, the is a perennial C4 grass, which can produce more than 20 use of locally available and introduced forage species t DM/ha annually (Taylor et al. 1976; Sithamparanathan adapted to the local agro-ecological conditions is 1979). It has been recommended for use in tropical and recommended. The cultivation of high quality forages subtropical countries with a minimum yearly rainfall with high herbage yield and adaptability to biotic and of 750 mm or 580 mm on very fertile soils (Botha abiotic environmental stresses is one of the options 1948). However, it grows better in wetter areas with no to increase livestock production under smallholder prolonged dry season (Rattray 1960). S. sphacelata has farmer conditions (Tessema 2005). The introduction of the desirable forage attributes of high yield (Singh et promising improved forage crops like Urochloa, Setaria al. 1995), high crude protein concentration (de Almeida and Desho grass is an advocated strategy to alleviate the and Flaresso 1991) and good animal performance prevailing feed crisis in the country. (Jones and Evans 1989). Most of the Urochloa (previously named as Desho [Cenchrus glaucifolius (Hochst. ex A. Rich.) Brachiaria) species and varieties that have been Rudov & Akhani] formally known as Pennisteum developed are resistant to Napier grass stunt and smut glaucifolium Hochst. ex A. Rich. is a perennial grass and disease affecting Napier grass varieties in Eastern is palatable to cattle, sheep and other herbivores (FAO Africa (Ghimire et al. 2015; Maass et al.2015). 2010). Desho grass serves as a business opportunity Urochloa is well adapted to low-fertility soils and for farmers in Ethiopia (Shiferaw et al. 2011; Tilahun diseases. It withstands heavy grazing and sequesters et al. 2017). According to Lukuyu et al. (2011), it is carbon through its large root system with enhanced very important to have chemical composition and nitrogen use efficiency and minimized greenhouse utilization information of locally available feeds for gas emissions (Arango et al. 2014; Moreta et al. 2014). their inclusion into livestock feeding programs. Despite Urochloa decumbens (Stapf) R. D. Webster (previously their significant potential for forage production, there named as Brachiaria decumbens Stapf) is reported is little research on the comparative advantage of to be drought resistant and resilient when grown on producing Desho, Urochloa and Setaria species in the infertile soils, producing high herbage yields with central highlands of Ethiopia. The present study aimed relatively low levels of fertilizer inputs. U. ruziziensis to evaluate the performance of Urochloa, Setaria and (R. Germ. & C. M. Evrard) Crins [previously named Desho grass species and varieties and recommend the as Brachiaria ruziziensis (R. Germ. & C. M. Evrard)] best ones with combined attributes of high herbage plays an important role in soil erosion control and yield and quality for wider distribution among livestock ecological restoration. The grass has high dry matter producer communities in the Ethiopian highlands. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 294 M. Fajil, G. Kebede, F. Feyissa, K. Mohammed, M. Minta, S. Mengistu, A. Tsegahun Materials and Methods varieties (Table 2). Seeds of the Urochloa and Setaria species were obtained from the International Livestock Description of the study area Research Institute (ILRI), while Desho grass was obtained from the Debre Zeit Research Center (DZARC). The experiment was conducted at the Holetta The experiment was conducted as a Randomized Agricultural Research Center (HARC), Ethiopia, during Complete Block Design (RCBD) with three replications. the main cropping seasons of 2014 to 2019 under rain- The experimental fields were ploughed and harrowed to fed conditions. HARC is located at 9°00'N latitude and a fine seedbed. The seeds were grown in a nursery and 38°30'E longitude at an altitude of 2,400 m.a.s.l. It is vegetative parts in the form of root splits from mature characterized by a long term (30 years) average annual plants were used for planting which was accomplished at rainfall of 1,055 mm, average relative humidity of 60.6%, the beginning of the main rainy season (in mid-June). Plot and average maximum and minimum air temperature of size was 7.2 m2 (3x 2.4m). The root splits were planted 22.2°C and 6.1°C, respectively. Rainfall is bimodal and with the intra and inter row spacing of 0.25 m and 0.5 m about 70% of the precipitation falls in the period from respectively. The spacing between plots and blocks was June to September while the remaining thirty percent 1.5 m. Phosphorus fertilizer was uniformly applied to all falls in the period from March to May (EIAR 2005). plots at planting in the form of diammonium phosphate The soil type of the area (Table 1) is predominantly red (DAP, 18% N, 20% P, 1.5% S) at the rate of 100 kg/ha. Nitosol (Keneni 2007). After every harvest, the plots were top dressed with 100 kg urea (46% N)/ha of which one-third was applied at the Experimental treatments and design first shower of rain (in May) and the remaining two-thirds applied during the active growth stage of the plant, during The study involved ten perennial forage grass species and the mid-growing season (July–August). Table 1. Properties of soils in the study area Parameter Values Method of Analysis pH (1:2.5 H2O) 4.94 Potentiometric method Organic carbon (%) 1.79 Dichromate oxidation method (Walkley and Black 1934) Total nitrogen (%) 0.20 Kjeldhal method (Jackson 1958) Available P (ppm) 5.60 Olsen method (Olsen et al. 1954) CEC (meq/100 g) 18.24 NH4OAc method (pH=7) Na+ (meq/100 g) 0.16 NH4OAc method (Okalebo et al. 1993) K+ (meq/100 g) 5.03 NH4OAc method (Okalebo et al. 1993) Ca2+ (meq/100 g) 29.50 NH4OAc method (Okalebo et al. 1993) Mg2+ (meq/100 g) 13.75 NH4OAc method (Okalebo et al. 1993) P(mg kg-1) 5.6 NH4OAc method (Okalebo et al. 1993) Texture Sand (%) 18 Bouyoucos hydrometric method Silt (%) 15 Bouyoucos hydrometric method Clay (%) 67 Bouyoucos hydrometric method Source: Holetta Agricultural Research Center meteorological data report Table 2. Evaluated grass species Species Common name Accession Country of origin Cenchrus glaucifolius (Hochst. ex A. Rich.) Rudov & Akhani) Desho grass Kulumsa Ethiopia Setaria sphacelata (Schumach.) Stapf & C. E. Hubb. common setaria ILRI-143= cv. Kazungula Zambia Setaria sphacelata (Schumach.) Stapf & C. E. Hubb. common setaria ILRI-6543= cv. Narok Kenya Urochloa decumbens (Stapf) R. D. Webster signal grass ILRI-10871 = cv. Basilisk Uganda Urochloa decumbens (Stapf) R. D. Webster signal grass ILRI-13205 Kenya Urochloa decumbens (Stapf) R. D. Webster signal grass ILRI-14720 Rwanda Urochloa decumbens (Stapf) R. D. Webster signal grass ILRI-14721 Rwanda Urochloa ruziziensis (R. Germ. & C. M. Evrard) Crins ruzi grass ILRI-13332 unknown Urochloa ruziziensis (R. Germ. & C. M. Evrard) Crins ruzi grass ILRI-14774 Burundi Urochloa ruziziensis (R. Germ. & C. M. Evrard) Crins ruzi grass ILRI-14813 Burundi Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Evaluation of perennial grasses in Ethiopia 295 Data collection forage dry matter yield potential. The result of a combined analysis during the This experiment involved two phases, namely production phase (2015–2019) showed that plant height establishment (in mid June 2014) and productive phases at harvesting was significantly (P<0.001) influenced (2015–2019). Data were collected on vigor, plant height by accession (Table 3). S. sphacelata accessions were at harvesting and forage dry matter yield. Plant vigor significantly (P<0.001) taller than the other evaluated was recorded during the establishment phase (mid June perennial forage grass species in 2014, 2015, 2018 and 2014–June 2015) on a scale from 1–5 and converted to a 2019 experimental years but in 2018 the plant height percentage. Plant height was measured from the ground recorded for S. sphacelata accessions was non-significant to the highest leaf at the time of forage harvesting stage. (P>0.05) with Desho grass (variety Kulumsa). Plant height and number of tillers per plant were recorded from 6 randomly selected plants from the whole plot. For Dry matter yield the determination of biomass yield, Setaria accessions were harvested at 10% flowering stage using a quadrat Forage dry matter yield was significantly (P<0.001) measuring 3 * 2.4 m2 (7.2 m2) areas. Desho and Urochloa different for accessions over the production years (Table were harvested at >40cm before flowering, the height 4). Desho grass had significantly (P<0.001) higher forage of cutting determined by previous studies. The plot was dry matter yield than other evaluated grasses in 2014, 2015, cut twice per year in May–June and October–November. 2016 and 2017, excluding U. decumbens 13205 that was Weight of the total fresh biomass yield was recorded from not significantly (P>0.05) different from Desho grass in each plot in the field and a 500 g sub-sample was taken the 2016 production phase. In 2018 U. decumbens 10871, from each plot to the laboratory to determine dry matter U. decumbens 13205 and U. decumbens 14721 had higher yield. Sub-samples were oven dried at 65°C for 72 hours. (P<0.001) forage dry matter yield than the other grasses. The oven dried samples were ground to pass through a The forage dry matter yield increased with production 1 mm sieve for laboratory analysis. Before scanning, years for the first three consecutive years (2014–2016) for the samples were dried at 60 °C overnight in an oven each evaluated grass species, exclusive of S. sphacelata 143 to standardize the moisture and then 3 g of each sample which showed a slight decrease from the first (2015) year of was scanned by Near Infra-Red Spectroscopy (NIRS). production to fourth year (2018) of production. However, in Percentage dry matter (DM), ash, crude protein (CP), the third and fourth years of production all accessions showed neutral detergent fiber (NDF), acid detergent fiber (ADF), a decrease in forage dry matter yield except U. ruziziensis acid detergent lignin (ADL) and in-vitro dry matter 13332 and U. decumbens 10871. During the fifth (2018) year digestibility (IVDMD) were predicted using a calibrated of production to the end of this experiment, all evaluated NIRS (Foss 5000 apparatus and WinISI II software). grasses showed biomass yield increase. Desho grass had higher forage dry matter yield during the establishment Statistical analysis phase (2014) than the other grasses. The analysis of variance (ANOVA) procedure of the Forage chemical composition SAS general linear model (GLM) (ersion 9.4 was used to analyse the quantitative data (SAS 2002). The LSD test Nutritional qualities of the perennial forage grass species at 5% significance was used for comparison of means. evaluated at Holetta are presented in Table 5. NDF and ADF content were significantly (P<0.001) different among Results the accessions and U. ruziziensis had lower ADF and NDF content than the other grasses. Plant vigor and height ADL was significantly (P<0.05) different among the accessions. IVDMD and crude protein percentage were The result of the analysis indicated that vigor was not significantly (P>0.05) influenced either by species or significantly (P<0.001) affected by species (Table accession. 3) with a rating of more than 50% plant vigor except Crude protein yield (CPY) was significantly (P<0.001) for U. decumbens 14720. The plant vigor percentage different among the accessions. Despite having the lowest performance of the species was positively associated CP percentage, Desho grass had higher (P<0.001) CPY than with the forage dry matter yield during the establishment U. ruziziensis, S. sphacelata and U. decumbens accessions, year showing plant vigor can be a good indicator of the except U. decumbens 14721 and S. sphacelata 6543. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 296 M. Fajil, G. Kebede, F. Feyissa, K. Mohammed, M. Minta, S. Mengistu, A. Tsegahun Table 3. Vigor and plant height (cm) of perennial forage grass species at harvest. S. Species Plant height (cm) in year Productive Combined Vigor (%) No 2014 2015 2016 2017 2018 2019 phase analysis 1 U. ruziziensis ILRI-14813 44.13cde 42.23d 76.20d 38.87e 54.19bc 39.20d 50.14d 49.14e 60.00cde 2 U. ruziziensis ILRI-14774 34.67de 49.45cd 50.87e 50.00e 45.00c 37.80d 46.62d 44.63e 56.20def 3 U. ruziziensis ILRI-13332 52.20c 61.67c 81.40d 42.77e 65.00bc 39.73d 58.11d 57.13de 70.00bc 4 U. decumbens ILRI-14721 40.67cde 59.93cd 103.03bc 112.77abc 67.88b 56.13bc 79.95bc 73.40c 53.40def 5 U. decumbens ILRI-10871 45.23cd 66.67c 92.37cd 102.77abc 68.09b 60.30bc 78.04c 72.57c 80.00ab 6 U. decumbens ILRI-14720 29.23e 54.72cd 94.80cd 101.13bcd 59.58bc 61.97bc 74.44c 66.90cd 46.60f 7 U. decumbens ILRI-13205 41.33cde 60.55c 91.77cd 92.77d 64.55bc 54.73c 72.87c 67.62cd 63.40cd 8 S. sphacelata ILRI-143 119.47a 130.83a 119.37ab 114.47ab 113.22a 82.53a 112.08a 113.31a 76.60b 9 S. sphacelata ILRI-6543 121.67a 130.00a 127.00a 117.74a 116.22a 85.00a 115.19a 116.27a 50.00ef 10 Desho grass (var. Kulumsa) 89.43b 91.67b 105.30bc 100.00cd 95.47a 65.57b 91.60b 91.24b 90.00a Mean 61.80 74.77 94.21 88.33 74.93 58.30 77.91 75.22 64.62 CV 15.18 14.13 12.54 8.97 16.99 10.87 23.20 25.48 9.27 Significance level *** *** *** *** *** *** *** *** *** ***=P<0.001; Means with the same letter are not significantly different. Table 4. Dry matter yield (t/ha) per year of perennial forage grass species tested per year at Holetta from 2014 to 2019. S. Species Dry matter yield (t/ha) in each year Productive Combined No 2014 2015 2016 2017 2018 2019 phase analysis 1 U. ruziziensis ILRI-14813 2.43cd 13.04bc 19.11cde 8.37d 7.64de 9.75cde 11.58ef 10.06de 2 U. ruziziensis ILRI-14774 1.30de 5.07e 12.23e 8.16d 4.33f 8.08de 7.56f 6.52e 3 U. ruziziensis ILRI-13332 3.45c 17.37bc 24.12bcd 8.70d 7.62de 11.53cde 13.87de 12.13cd 4 U. decumbens ILRI-14721 0.96e 10.92cde 30.07ab 23.24b 17.04a 19.78ab 20.21bc 17.00bc 5 U. decumbens ILRI-10871 0.76e 7.39de 25.03bc 14.93cd 18.62a 20.92a 17.38bcd 14.61bcd 6 U. decumbens ILRI-14720 0.64e 10.03de 23.32bcd 16.70bc 12.40bc 13.57bcd 15.20cde 12.78cd 7 U. decumbens ILRI-13205 1.19de 14.23bcd 36.53a 22.77b 18.36a 18.30ab 22.04ab 18.56b 8 S. sphacelata ILRI-143 5.27b 18.62b 23.26bcd 14.70cd 9.86cd 8.41de 14.97de 13.35bcd 9 S. sphacelata ILRI-6543 5.74b 18.41b 17.19de 12.20cd 5.88ef 7.22e 12.18ef 11.11de 10 Desho grass (var. Kulumsa) 13.14a 33.41a 36.55a 34.48a 13.54b 15.69abc 26.75a 24.27a Mean 3.49 14.84 24.74 16.43 11.53 13.33 44.15 14.06 CV 24.39 27.75 18.45 27.41 16.54 27.72 16.17 59.18 Significance level *** *** *** *** *** *** *** *** ***=P < 0.001; Means with the same letter are not significantly different. Table 5. Nutrient content of perennial forage grasses S. No Species DM% Ash% NDF% ADF% ADL% CP% CPY (t/ha) IVDMD% 1 U. ruziziensis ILRI-14813 91.35de 17.72a 61.95b 29.92b 4.44bc 6.33 0.64cd 59.81 2 U. ruziziensis ILRI-14774 91.37de 16.72abc 63.55b 31.27 b 4.74abc 6.72 0.44d 59.56 3 U. ruziziensis ILRI-13332 91.16e 17.14ab 62.68b 30.68 b 4.27c 6.22 0.75bcd 60.92 4 U. decumbens ILRI-14721 91.60bcd 16.52abcd 67.14a 34.72 a 4.86ab 5.57 0.95abc 55.35 5 U. decumbens ILRI-10871 91.83bc 15.25cd 69.48a 37.50 a 5.25a 5.58 0.81bc 55.68 6 U. decumbens ILRI-14720 91.47cde 15.56bcd 67.55a 35.32 a 5.19 a 6.87 0.88bc 57.16 7 U. decumbens ILRI-13205 91.72bcd 16.26abcd 68.03a 35.34 a 4.84ab 5.57 1.04ab 56.26 8 S. sphacelata ILRI-6543 91.70bcd 14.94d 67.55a 35.74 a 4.80abc 6.98 0.92abc 54.14 9 S. sphacelata ILRI-143 92.26a 15.69bcd 66.99a 36.32 a 4.59bc 6.96 0.77bc 54.51 10 Desho grass (variety Kulumsa) 91.92ab 12.92e 69.29a 37.64 a 4.48bc 5.04 1.23a 56.44 Mean 91.64 15.87 66.42 34.45 4.75 6.18 0.84 56.98 CV 0.27 5.83 2.27 4.97 7.06 13.22 59.77 4.37 Significance level ** *** *** *** * ns *** ns DM% = Dry matter percentage; Ash% = Ash percentage; NDF% = Neutral detergent fiber percentage; ADF% = Acid detergent fiber percentage; ADL% = Acid detergent lignin percentage; CP% = Crude protein percentage; CPY = Crude protein yield; IVDMD = In-vitro dry matter digestibility; ** = P<0.01; * = P<0.05; *** = P<0.001; ns = non-significant; Means with the same letter are not significantly different. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Evaluation of perennial grasses in Ethiopia 297 Discussion itself a measure of cell-wall content; thus there is a negative relationship between the NDF content of feeds Desho and Setaria showed better vigor than Urochloa, and the rate at which they are digested. Schroeder et al. suggesting that Desho and Setaria were faster to establish (2012) also reported that as NDF percentages increase, and had superior competition against the weeds than dry-matter intake generally will decrease. U. ruziziensis Urochloa species especially in the establishment phase. accessions had ADF above the minimum recommended This can be an important characteristic to establish value (17–21 percent) for NRC (2001). This result these forages on soil bunds for soil conservation in the suggests that U. ruziziensis species will have moderate livestock-crop mixed area. Soil bunds are available digestibility compared to the other grasses evaluated in for free grazing during the non-cropping season and this experiment. Nussio et al. (1998) reported that forage these grasses can tolerate the grazing due to their fast with ADF content around 40% or more, shows low establishment characteristics. S. sphacelata accessions intake and digestibility. In this study forage materials and Desho grass were taller during the establishment from all the grass species had low CP below the 7% CP year, possibly due to the morphological vertical growth required for microbial protein synthesis in the rumen characteristics of the species and plant vigor. Plant that can support at least the maintenance requirement of height differences can be attributed to the morphological ruminants (Van Soest, 1994). IVDMD levels were low and physiological differences among the cultivars and Mugeriwa et al. (1973) reported that the IVDMD (Nguku 2015), Setaria having different morphology to values greater than 65% indicate good feeding value and the Urochloa accessions. values below this threshold level result in reduced intake U. ruziziensis accessions provided the highest forage due to lowered digestibility. dry matter yield in the establishment phase, suggesting that this species is fast growing and more easily established Conclusions than U. decumbens accessions. During the production phase, Desho grass produced significantly more forage Based on dry matter yield and crude protein yield dry matter yield than other evaluated grass species. This data, U. decumbens 13205, U. ruziziensis 13332, implies Desho grass is more adaptable to Nitosol and S. sphacelata 6543 and Desho grass (variety Kulumsa) cold air conditions than U. ruziziensis, U. decumbens are recommended for the study area and similar agro- and S. sphacelata grasses. The Urochloa accessions are ecologies as alternative forage grasses. true tropical plants and their growth almost stops when temperatures drop below about 20 °C. Setaria is more Acknowledgments subtropical and will continue to grow at much lower temperatures than the Urochloa accessions. Forage dry The funds for this study were granted by the Ethiopian matter yield increased with production years for the first Institute of Agricultural Research. We are grateful to the three consecutive years due to climatic variation (rainfall technical and field assistants of the forage and pasture pattern, temperature, frost). Desho grass had the highest research program, Holetta Agricultural Research Center forage dry matter yield and more vigorous growth that for data collection. We thank also the Holetta laboratory resulted in the well-established root system that enabled technicians and researchers working in animal nutrition the grass to extract better growth resources from the soil. for the laboratory analysis. Although differences were seen in nutrient content, all grasses studied were low quality. This may be the References result of harvesting when over mature with only two harvests per year. Farmer practices of harvesting were (Note of the editors: All hyperlinks were verified 8 September 2021). followed in the experiment to reflect the local feeding situation. Grasses are usually harvested after 6 to 8 Arango J; Moreta D; Núñez J; Hartmann K; Domínguez M; weeks of growth to obtain better quality feed. In this Ishitani M; Miles J; Subbarao G; Peters M; Rao I. 2014. Developing methods to evaluate phenotypic variability study, forage materials from all the grass species had in biological nitrification inhibition (BNI) capacity greater than 60% NDF which may result in low intake of Brachiaria grasses. Tropical Grasslands-Forrajes and digestibility in livestock. McDonald et al. (2002) Tropicales 2:6–8. doi: 10.17138/TGFT(2)6-8 reported that the primary chemical composition of feeds Botha JP. 1948. Setaria grasses. Farming in South Africa that determines the rate of digestion is NDF, which is 23:729–735. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 298 M. Fajil, G. Kebede, F. Feyissa, K. Mohammed, M. Minta, S. Mengistu, A. Tsegahun CSA (Central Statistical Agency). 2016. Agricultural sample pastures with annual crops. In: Guimarães EP; Sanz JI; survey. Volume II, Report on livestock and livestock Rao IM; Amézquita MC; Amézquita E; Thomas RJ, Eds. characteristics (private peasant holdings). Statistical Agropastoral systems for the tropical savannas of Latin Bulletin 585(2):33–35 Addis Ababa, Ethiopia. America. CIAT Publication No. 338. Centro International CSB (Climate-Smart Brachiaria Program). 2016. CSB de Agricultura Tropical (CIAT), Cali, Colombia. hdl. annual review meeting. KALRO (Kenya Agricultural and handle.net/10568/54057 Livestock Research Organization), Embu, Kenya. goo.gl/ LMA (Livestock Marketing Authority). 1999. LMA Annual VbzJ4D report, Addis Ababa, Ethiopia. de Almeida EX; Flaresso JA. 1991. Introdução e avaliação de Lukuyu B; Franzel S; Ongadi PM; Duncan AJ. 2011. Livestock forrageiras tropicais no Alto Vale do ltajaí, Santa Catarina, feed resources: Current production and management Brasil. Pasturas Tropicales 13(3):23–30. bit.ly/3CxhpH2 practices in central and northern rift valley provinces of Demeke S; Mekuriaw Y; Asmare B. 2017. Assessment of Kenya. Livestock Research for Rural Development 23, livestock production system and feed balance in watersheds Article#112. lrrd.org/lrrd23/5/luku23112.htm of North Achefer district, Ethiopia. Journal of Agriculture Maass BL; Midega CAO: Mutimura M; Rahetlah VB; and Environment for International Development Salgado P; Kabirizi JM; Khan ZR; Ghimire SR; Rao 111(1):175–190. IM. 2015. Homecoming of Brachiaria: Improved EIAR (Ethiopian Institute of Agricultural Research). 2005. hybrids prove useful for african animal agriculture. East Holetta Agricultural Research Center/HARC/ progress African Agricultural and Forestry Journal 81(1):71–78. report 2005/06. EIAR/HARC, Holetta, Ethiopia. doi: 10.1080/00128325.2015.1041263 FAO (Food and Agricultural Organization of the United McDonald P; Edwards RA; Greenhalgh JFD; Morgan CA. Nations). 2004. Livestock sector brief, Ethiopia. FAO, 2002. Animal nutrition 6th edition. Pearson Educational Rome, Italy. bit.ly/39xMttz Limited. Edinburgh, UK. p. 255–657. FAO (Food and Agriculture Organization of the United Moreta DE; Arango J; Sotelo M; Vergara D; Rincón A; Ishitani Nations). 2010. Grassland Index. A searchable catalogue M; Castro A; Miles J; Peters M; Tohme J; Subbarao GV; of grass and forage legumes. FAO, Rome, Italy. Rao IM. 2014. Biological nitrification inhibition (BNI) Feleke G. 2003. A review of the small scale dairy sector- in Brachiaria pastures: A novel strategy to improve eco- Ethiopia. FAO prevention of food losses program: Milk efficiency of crop-livestock systems and to mitigate climate and dairy products, post-harvest, Losses and Food Safety change. Tropical Grasslands-Forrajes Tropicales 2:88–91. in Sub-Saharan Africa and the Near East. Food and doi: 10.17138/tgft(2)88-91 Agriculture Organization of the United Nations (FAO), Mugeriwa JS; Christensen DA; Ochetim S. 1973. Grazing Rome, Italy. bit.ly/3nROdXb behavior of exotic dairy cattle in Uganda. East African Ghimire SR; Njarui DMG; Mutimura M; Cardoso JA; Johnson Agricultural and Forestry Journal 39(1):1–11. doi: 10.1080/ L; Gichangi E; Teasdale S; Odokonyero K; Caradus JR; 00128325.1973.11662610 Rao IM; Djikeng A. 2015. Climate-smart Brachiaria for NRC (National Research Council). 2001. Nutrient improving livestock production in East Africa: Emerging requirements of dairy cattle, 7th revised ed. National opportunities. In: Vijay D; Srivastava MK; Gupta CK; Academy Press, Washington, DC, USA. bit.ly/3jSkbA4 Malaviya DR; Roy MM; Mahanta SK; Singh JB; Maity Nguku SA. 2015. An evaluation of Brachiaria grass cultivars A; Ghosh PK, Eds. Proceedings of the XXIII International productivity in semi arid Kenya. M.Sc. thesis, South Grassland Congress, New Delhi, India, 20–24 November Eastern Kenya University, Kenya. repository.seku.ac.ke/ 2015. p. 361–370. hdl.handle.net/10568/69364 handle/123456789/1380 Jackson ML. 1958. Soil Chemical Analysis. Prentice-Hall, Nussio LG; Manzano RP; Pedreira CGS. 1998. Valor Hoboken, NJ, USA. alimentício em plantas do gênero Cynodon. Simpósio Jones RM; Evans TR. 1989. Live weight gain from four sobre manejo da pasagem. FEALQ/ESALQ, Piracicaba, nitrogen fertilized grasses grazed over the growing season Brazil. p. 203–242. in coastal south-east Queensland. Tropical Grasslands Okalebo JR; Gathua KW; Woomer PL. 1993. Laboratory 23:75–79. bit.ly/3AwVFdO methods for soil and plant analysis: A working manual. Keneni G. 2007. Phenotypic diversity for biological nitrogen Tropical Soil Biology and Fertility Programme, Nairobi, fixation in Abyssinian field pea (Pisum sativum var. Kenya. abyssinicum) germplasm accession. Report on independent Olsen SR; Cole CV; Watanabe FS; Dean LA. 1954. Estimation study for PhD. Addis Ababa University, Addis Ababa, of available phosphorus in soils by extraction with sodium Ethiopia. bicarbonate. Circular 939. US Department of Agriculture, Kluthcouski J; de Oliveira IP; Yokoyama LP; Dutra LG; Washington DC, USA. bit.ly/3COQUgz Portes TA; da Silva AE; Pinheiro BS; Ferreira E; de Castro Rattray JM. 1960. The grass covers of Africa. FAO Agricultural EM; Guimarães CM; Gomide JC; Balbino LC. 2004. The Series No. 49. Food and Agriculture Organization of the Barreirão system: recovering and renewing degraded United Nations (FAO), Rome, Italy. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Evaluation of perennial grasses in Ethiopia 299 Rodrigues RC; Sousa TVR; Melo MAA; Araújo JS; Lana Range Management and Agroforestry 16(1):39–45. RP; Costa CS; Oliveira ME; Parente MOM; Sampaio Sithamparanathan, J. 1979. Seasonal growth patterns of IBM. 2014. Agronomic, morphogenic and structural herbage species on high rainfall hill country in northern characteristics of tropical forage grasses in northeast North Island: I. Temperate grasses. New Zealand Brazil. Tropical Grasslands-Forrajes Tropicales Journal of Experimental Agriculture 7(2):157–162. doi: 2:214–222. doi: 10.17138/tgft(2)214-222 10.1080/03015521.1979.10426183 SAS. 2002. SAS User's Guide: Statistics Released 6.12. SAS Taylor AO; Rolley JA; Hunt BJ. 1976. Potential of new Inc., Cary NC., USA. summer grasses in Northland. New Zealand Journal Schiek B; González C; Mwendia SW; Prager SD. 2018. Got of Agricultural Research 19:477–481. doi: 10.1080/ forages? Understanding potential returns on investment 00288233.1976.10420978 in Brachiaria spp. for dairy producers in Eastern Africa. Tessema Z. 2005. Variation in growth, yield, chemical Tropical Grasslands-Forrajes Tropicales 6:117–133. doi: composition and in vitro dry matter digestibility of Napier 10.17138/tgft(6)117-133 grass accessions (Pennisetum purpureum). Tropical Schroeder R; Snijders PJM; Wouters AP; Steg A; Kariuki Science 45(2)67–73. doi: 10.1002/ts.51 JN. 2012. Variation in DM digestibility, CP, yield and Tilahun G; Asmare B; Mekuriaw Y. 2017. Effects of harvesting ash content of Napier grass (Pennisetum purpureum) and age and spacing on plant characteristics, chemical their production from chemical and environmental factors. composition and yield of desho grass (Pennisetum National Animal Husbandry Research Station, KARI, pedicellatum Trin.) in the highlands of Ethiopia. Tropical Naivasha, Kenya. Grasslands-Forrajes Tropicales 5:77-84. doi: 10.17138/ Shiferaw A; Puskur R; Tegegne A; Hoekstra D. 2011. TGFT(5)77-84 Innovation in forage development: Empirical Van Soest PJ. 1994. Nutritional ecology of the ruminant, 2nd evidence from Alaba Special District, Southern edition. Cornell university press, Ithaca, NY, USA. jstor. Ethiopia. Development in Practice 21(8):1138–1152. org/stable/10.7591/j.ctv5rf668 doi: 10.1080/09614524.2011.591186 Walkley AJ; Black IA. 1934. Estimation of soil organic Singh KA; Rai RN; Balaraman N; Singh LN. 1995. Evaluation carbon by the chromic acid titration method. Soil Science of forage grasses for mid hills of Sikkim (Eastern Himalaya). 37, 29-38. (Received for publication 23 January 2021; accepted 2 September 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):300–306 300 doi: 10.17138/TGFT(9)300-306 Research Paper Is organic fertilizer application a viable alternative to synthetic fertilizer for Piatã grass? ¿Son los fertilizantes orgánicos una alternativa viable a fertilizantes sintéticos para el pasto Piatã? SÍRIO DOUGLAS DA SILVA DOS REIS1, MARCO ANTONIO PREVIDELLI ORRICO JUNIOR1, MICHELY TOMAZI2, STÉFANE SOUZA CUNHA1, ANA CAROLINA AMORIM ORRICO1, JOYCE PEREIRA ALVES1 AND EDGAR SALVADOR JARA GALEANO1 1Universidade Federal da Grande Dourados, Faculdade de Ciências Agrárias, Dourados, MS, Brazil. ufgd.edu.br/ 2Embrapa Agropecuária Oeste, Dourados, MS, Brazil. embrapa.br/agropecuaria-oeste Abstract Organic fertilizer in many cases can replace mineral fertilizers and in consequence reduce production costs and improve soil quality. Thus, the aim of this work was to evaluate productive, morphogenic and structural characteristics of Piatã grass (Urochloa brizantha) fertilized with urea, organic compost and biofertilizer throughout a year. The trial design was a block split-plot in time (seasons) design with 4 treatments (fertilizing with urea, organic compost, biofertilizer and Control) and 6 repetitions. The evaluated parameters were: dry matter production (DMP), leaf elongation rate (LER), leaf appearance rate (LAR), phyllochron (PHYL), leaf lifespan (LLS), pseudostem elongation rate (SER), final leaf length (FLL), number of live leaves (NLL) and number of tillers (NT). The highest LAR values were observed during summer and spring for the treatment with urea, which also produced the highest LER values. No difference was found in SER among the fertilizer treatments but all fertilized treatments were superior to Control. NT and DMP values were highest (P<0.05) in the treatment with urea, followed by biofertilizer, organic compost and Control. In conclusion, while the use of urea provided greatest forage production, applying biofertilizer gave superior yields to organic compost. Other benefits of organic fertilizers should be assessed as well as combinations of organic and inorganic fertilizers. Keywords: Biofertilizer, nitrogen, organic compost, season, urea, Urochloa brizantha. Resumen La fertilización orgánica, en muchos casos, puede reemplazar a los fertilizantes minerales y, en consecuencia, reducir los costos de producción y mejorar la calidad del suelo. Así, el objetivo de este trabajo fue evaluar las características productivas, morfológicas y estructurales del pasto Piatã (Urochloa brizantha) fertilizado con urea, compuesto orgánico y biofertilizante durante un año. Para eso, se utilizó un diseño de bloques con parcelas divididas en el tiempo (estaciones), compuesto por cuatro tratamientos (fertilización con urea, compuesto orgánico, biofertilizante y control) y seis repeticiones. Los parámetros evaluados fueron: producción de materia seca (DMP), tasa de elongación de hojas (LER), tasa de aparición de hojas (LAR), filocrón (PHYL), vida útil de las hojas (LLS), tasa de elongación de pseudotallo (SER), longitud final de la hoja (FLL), número de hojas vivas (NLL) y número de macollas (NT). Los valores de LAR más altos se observaron durante el verano y la primavera para el tratamiento con urea, que también producjo los valores más altos de LER. No se encontró diferencia en el SER entre los fertilizantes probados, sin embargo, hubo una diferencia entre estos tratamientos y el control. Los valores de NT y DMP fueron mayores en el tratamiento con urea, seguido de biofertilizante, compuesto Correspondence: Prof. M. A. P. Orrico Jr, Curso de Zootecnia da Faculdade de Ciências Agrárias da Universidade Federal da Grande Dourados (UFGD), Rodovia Dourados - Itahum km 12. CEP 79804- 970 - Dourados, MS, Brazil. E-mail: marcoorrico@yahoo.com.br Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Organic fertilization of Piatã grass 301 orgánico y control. Se puede concluir que el uso de urea brindó mayor rendimiento forrajero, sin embargo, la fertilización orgánica con biofertilizante resultó ser más ventajosa en comparación con el compuesto orgánico. Palabras clave: Biofertilizante, compost orgánico, estaciones del año, nitrógeno, urea, Urochloa brizantha. Introduction producing solid fertilizer) and biodigestion (anaerobic process producing liquid fertilizer), which result in The use of synthetic fertilizer is the most common way the production of organic compost and biofertilizer, to restore soil nutrients; however, its use is becoming respectively. Even when the same raw material (poultry impractical due to high prices on the international litter for example) is used, the final chemical composition market. According to data published by ANDA (2019), and its value as fertilizer may be markedly different 36 million tonnes of fertilizer were applied in Brazil between these two types of fertilizer, which could result in 2019, and approximately 80% of this material was in distinctly different responses (Bowden et al. 2007). imported. It is possible that use of an alternative form Therefore, this research aimed to determine if organic of nutrients, e.g. organic fertilizer made from organic compost and biofertilizer produced different responses waste, could greatly reduce agricultural fertilizer costs in growth and productivity of the pasture grass Piatã in the country. (Urochloa brizantha), and to assess differences in The utilization of organic fertilizer may be a viable responses produced between these organic fertilizers and alternative for fertilizing tropical pastures (Orrico Jr. et urea when applied to the grass. al. 2012). Organic wastes are cheaper than conventional inorganic fertilizers and contain additional nutrients Material and Methods important for forage growth. In the literature, there are several studies that prove the efficiency of using organic The trial was carried out in a greenhouse at the experimental fertilizers on pastures (Orrico Jr. et al. 2013; Ryals area of Embrapa Agropecuária Oeste, Dourados, Mato et al. 2016; Grave et al. 2018; Orrico Jr. et al. 2018). Grosso do Sul, Brazil (22°16’30” S, 54°49’00” W). The However, there is a very wide variety of organic wastes climate in the region according to the Köppen classification with different chemical compositions, which may lead is type Cwa (humid mesothermal, with hot summers and to a great variety of responses. Among the main waste dry winters). The meteorological data recorded during the treatment systems are composting (aerobic process experiment are presented in Figure 1. 14 40 12 35 10 30 25 8 20 6 15 4 10 2 5 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Summer Autumn Winter Spring Summer Sunlight Min T Average T Max T Radiation Figure 1. Average (average T), maximum (Max T) and minimum (Min T) air temperatures, hours of sunlight and radiation measured during the experimental period (2017‒2018) at Dourados-MS, Brazil. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Sunlight (h) Temperature (oC) Radiation (MJ/m2/day) 302 S.D.S. dos Reis, M.A.P. Orrico Jr, M. Tomazi, S.S. Cunha, A.C.A. Orrico, J.P. Alves and E.S.J. Galeano. The soil used was an Oxisol of clay texture with and initial lengths of leaf blades divided by the number of the following characteristics: sand, 12.8%; silt, 10.7%; days of the evaluation period; (4) pseudostem elongation clay, 76.5%; pH in CaCl2, 4.78; P, 5.14 mg/dm 3; K, 1.00 rate (SER, cm/tiller/day) - the difference between initial cmolc/dm3; Ca2+, 2.86 cmolc/dm3; Mg2+, 1.29 cmolc/ and final stem lengths divided by the number of days dm3; Al3+, 0.15 cmolc/dm3; H+Al, 6.08 cmolc/dm3; of evaluation; (5) leaf lifespan (LLS) - the number of cation-exchange capacity (CEC), 11.22 cmolc/dm3; OM, live leaves multiplied by the phyllochron; (6) final leaf 27.24 g/kg; and base saturation, 45.85%. length (FLL, cm/tiller) - the mean leaf blade length of all A block split-plot in time (seasons) design with 4 expanded leaves present on a tiller; (7) number of live treatments, i.e. Control (= no fertilizer) and fertilization leaves (NLL) - the total number of green leaves on each with urea, organic compost and biofertilizer, and 6 tiller; and (8) number of tillers (NT) - the total number of replicates, was used giving a total of 24 experimental green tillers in each pot. units (40 L pots). All fertilized pots received 400 kg The parameters were submitted to analysis of N/ha/year (0.8 g N/pot), applied in ten 40 kg N/ha variance using the split-plot in time scheme (using the applications. The concentrations of N, P, K, Mg, Ca and PROC MIXED procedure) to assess the effect of the Na were 2.13, 1.77, 2.6, 0.41, 0.86 and 0.71 g/100 g in main treatments (fertilizer types), secondary treatments compost and 0.23, 0.19, 0.31, 0.05, 0.09 and 0.071 g/100 (seasons) and their interaction (fertilizer type × season). mL in biofertilizer, respectively. The means of the treatments were compared by Tukey’s Soil moisture in the pots was maintained at around test at 5% probability. The statistical analysis was 70% of field capacity throughout using an irrigation performed through the software SAS 6.1. system. On 6 December 2016, 30 seeds of Piatã grass (Urochloa brizantha; syn. Brachiaria brizantha) were Results sown in each pot. Seven days after emergence, seedlings were thinned to retain the most vigorous 9 plants in PHYL, LAR, LLS and NLL showed interactions each pot. A standardization cut was made 50 days after (P<0.01) between season and type of fertilizer applied sowing (25 January 2017) at 20 cm from the soil surface (Figure 2). LAR on treatments fertilized with urea during (beginning of the experimental period). Subsequently, summer and spring was greater (P<0.05) than those for evaluation harvests were performed every 35 days at 20 the remaining treatments. However, during autumn and cm from the soil. After each harvest, the next application winter differences in LAR between fertilized treatments of fertilizer was applied and a new data collection cycle were small and LAR on all fertilized treatments exceeded began. Pots were irrigated when fertilizer was applied, to (P<0.05) those of Control. Overall highest LAR values reduce nitrogen loss by urea volatilization. Ten harvests occurred in summer and the lowest in winter (P<0.05). were made between 1 March 2017 and February 2018. PHYL values followed the inverse behavior of LAR, The total weight of green forage contained in the pots with highest values being recorded in winter, the absolute above a height of 20 cm from the soil was recorded at each highest value (29.9 days) for Control in winter and the harvest. The material collected was taken to the laboratory lowest values for all treatments in summer. and placed in a forced-air oven at 65 °C for at least 72 h Control had longer (P<0.05) LLS than fertilized to determine the dry matter concentration according to the treatments in autumn and winter but there were no methodology described by AOAC (2005). differences between treatments in summer and spring In order to assess forage morphogenic and structural (P>0.05). The highest LLS value was obtained for the characteristics, 3 tillers per pot were tagged with colored Control during winter (176 days) with about 40 days string after the standardization cut. The leaves and living in summer and spring. NLL values varied from 5 to and senescent parts were measured every 3 days with a 6.5 leaves/tiller between treatments tested, with no rule and the data were used to calculate the following consistent difference between treatments. morphogenic and structural characteristics: (1) leaf There were no significant interactions between season appearance rate (LAR, leaves/tiller/day) - the number and fertilizer type for other parameters, so main effect of leaves that appeared divided by the number of days responses only are shown in Table 1. LER values were of cycle evaluation; (2) phyllochron (PHYL, days) - the significantly affected by fertilizer type, being highest for interval between the appearance of 2 consecutive leaves the urea treatment (3.99 cm/d) and lowest for Control on a tiller, the opposite of LAR; (3) leaf elongation rate (2.49 cm/d) with compost and biofertilizer intermediate (LER, cm/tiller/day) - the difference between the final (mean 3.5 cm/d) (P<0.01). Pseudostem elongation rate Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Organic fertilization of Piatã grass 303 35 200Aa Aa 30 160 25 20 Ba 120 BaBa Aa Ba Ba CaAb 15 Bb ABbBb Bb 80 Bb Ac AcAc Cb BbBbBb Bb Ac Ac Ac10 Bc AcAcAcAc Ac 40 5 0 0 Summer Autumn Winter Spring Summer Autumn Winter Spring 0.21 8 Aa Aa 0.18 7Ba AaBaBa Ab Aa ABa Ba Ab ABab Aa 6 BbBb ABa BaBa Bb ABb0.15 Ab Bb Bb Bb Bb Bb 5 0.12 ABb Cb 4 0.09 AcAcAc 3 0.06 Bc 2 0.03 1 0.00 0 Summer Autumn Winter Spring Summer Autumn Winter Spring Control Urea Compost Biofertilizer Figure 2. Effects of fertilizer type and season on phyllochron (PHYL), leaf appearance rate (LAR), leaf lifespan (LLS) and number of live leaves (NLL) of Piatã grass. PHYL: effects of type of fertilizer (P<0.01), season (P<0.01) and interaction between type of fertilizer and season (P<0.01) (s.e.m. = 0.634). LAR: effects of type of fertilizer (P<0.01), season (P<0.01) and interaction between type of fertilizer and season (P<0.01) (s.e.m. = 0.003). LLS: effects of type of fertilizer (P<0.01), season (P<0.01) and interaction between type of fertilizer and season (P<0.01) (s.e.m. = 3.548). NLL: effects of type of fertilizer (P<0.01), season (P<0.01) and interaction between type of fertilizer and season (P=0.03) (s.e.m. = 0.049). Means for season with different lower-case letters differ by Tukey’s test (P<0.05); and means for fertilizer type with different upper- case letters differ by Tukey’s test (P<0.05). was greater for all fertilizer treatments than for Control with the following order: summer > spring > autumn (P<0.01), resulting in final leaf length following the > winter (Table 1; P<0.01). Leaf growth in winter was same pattern (P<0.01). Parameters with the greatest less than half of that in summer, while pseudostem fertilizer effects were number of tillers/pot and DM growth virtually ceased in winter. Final leaf length yield/pot. The urea treatment produced the greatest in summer and spring exceeded those in autumn and number of tillers/pot followed by biofertilizer, compost winter (P<0.01). However, number of tillers per pot was and Control with significant differences between all greatest in summer with no difference between other treatments. This resulted in significant differences in seasons (P<0.01). As might be expected for tropical DM yields/pot for all treatments, with the highest yield grasses, DM production was greatest in summer (9.16 g for urea (8.47 g DM/pot) and the lowest for Control DM/pot) and lowest in autumn and winter (mean 4.59g (3.88 g DM/pot). DM/pot), with spring intermediate (mean 6.40 g DM/ Both LER and SER were strongly affected by season pot). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) LAR (leaf/day) PHYL (days) NLL (leaves/tiller) LLS (days) 304 S.D.S. dos Reis, M.A.P. Orrico Jr, M. Tomazi, S.S. Cunha, A.C.A. Orrico, J.P. Alves and E.S.J. Galeano. Table 1. Effects of fertilizer type and season on productive, morphogenic and structural characteristics of Piatã grass. Parameter Fertilizer Season s.e.m. P value Control Urea Compost Biofertilizer Summer Autumn Winter Spring F S F*S LER (cm/d) 2.49C 3.99A 3.43B 3.56B 4.81a 2.95c 1.84d 3.88b 0.135 <0.01 <0.01 ns SR (cm/d) 0.03A 0.03A 0.03A 0.02A 0.01a 0.02a 0.02a 0.01a 0.004 ns ns ns SER (cm/d) 0.44B 0.58A 0.54A 0.58A 0.78a 0.61c 0.05d 0.70b 0.031 <0.01 <0.01 ns FLL (cm) 14.6B 19.0A 17.4A 18.5A 18.2a 17.2b 16.3b 18.2a 0.288 <0.01 <0.01 ns NT (no./pot) 75D 122A 89C 101B 117a 94b 95b 96b 1.859 <0.01 <0.01 ns DMP (g DM/pot) 3.88D 8.47A 5.88C 6.50B 9.16a 4.47c 4.70c 6.40b 0.561 <0.01 <0.01 ns LER = leaf elongation rate; SR = senescence rate; SER = pseudostem elongation rate; FLL = final leaf length; NT = number of tillers; DMP = dry matter production. Means for season with different lower-case letters differ by Tukey’s test (P<0.05) and means for fertilizer type with different upper-case letters differ by Tukey’s test (P<0.05). Discussion Piatã grass that were measured supported the growth responses obtained, with leaf appearance rate (LAR) This glass-house study has provided valuable information being greater for urea treatments than for other treatments on the variation in growth responses in Piatã grass to in spring and summer in particular and LER and SER inorganic and organic fertilizers throughout the year. on all fertilized treatments exceeding those of Control. While urea produced a 118% increase in DM yield over The superiority of urea over compost and biofertilizer the unfertilized Control, responses with biofertilizer in producing rapid growth responses was reflected were only 67% and with compost were 52%. These in the higher LAR values for urea combined with the differences are probably a function of differences in the greater values for LER. Higher values of LAR and LER ready availability of the N in the various fertilizers. The result, in most cases, in forages with a high proportion published literature suggests there is a strong positive of leaves. Fertilizers that promote a high proportion of correlation between availability of N in the fertilizer leaves also produce forage with high levels of crude applied and growth responses in tropical forages (Al- protein. Mihret et al. (2018) also observed higher values Solaimani et al. 2017; McRoberts et al. 2018). With of LAR, LER and CP in grasses fertilized with synthetic synthetic fertilizers like urea, a high proportion of the N fertilizers (NPK) when compared with organic fertilizer. is readily available to plants and can be rapidly absorbed Unfortunately, we did not measure the CP concentration to produce rapid growth responses (Bowden et al. 2007). in the forage in this study. On the other hand, the availability of N in the organic Higher numbers of tillers per pot with urea were an fertilizers depends on a range of factors, including C:N important component of the superior responses produced ratio in the fertilizer, origin of the material and the with this fertilizer. According to Fagundes et al. (2006), treatment to which the waste has been subjected (Gutser ready availability of N leads to more rapid formation et al. 2005). When organic residues are submitted to of axillary buds (due to greater LAR values), which, anaerobic biodigestion, NH4+-N concentration in the consequently, contributes to the emergence of new tillers. effluents increases, which leads to a higher proportion This greater density of young tillers in the pasture results of the N being available for plants (Gutser et al. 2005). in the improvement of forage productivity (Caminha et Organic compost presents significant levels of N in al. 2010). organic form, which becomes more slowly available in Another variable that deserves to be highlighted in the soil. This would help to explain why organic compost this work is the marked seasonality of production of had the lowest values of dry matter production between forage even under controlled conditions in a pot trial. the types of fertilizer tested. According to Araujo et al. (2018), one of the main Orrico Jr. et al. (2018) required a 4.8-fold higher characteristics of tropical forages is seasonality of dose of organic compost to reach the same forage production, with growth being greatest during summer production obtained by Orrico Jr. et al. (2013) with the and spring seasons. Although this study was carried out use of biofertilizer. Although this comparison refers to in pots, the forage was exposed to the same variations separate studies, these were both pot studies performed of regional photoperiod and temperatures as would be by the same authors under very similar environmental the case in a field study. Of the total annual production conditions, i.e. type of forage, soil type and season. obtained (24.73 g DM/pot or equivalent to 12,376 kg The morphogenic and structural characteristics of DM/ha), proportions produced in different seasons were: Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Organic fertilization of Piatã grass 305 26, 37, 18 and 19% for spring, summer, autumn and References winter, respectively. This highlights the important effects of both temperature and hours of daylight on growth of (Note of the editors: All hyperlinks were verified 28 July 2021). tropical pastures, in this case Piatã grass, since the study was conducted under conditions where soil moisture Al-Solaimani SG; Alghabari F; Ihsan MZ; Fahad S. 2017. levels were maintained at 70% field capacity throughout. Water deficit irrigation and nitrogen response of Sudan grass under arid land drip irrigation conditions. Irrigation When seasonality of rainfall is taken into account, one and Drainage 66:365–376. doi: 10.1002/ird.2110 might question whether or not N application in autumn ANDA (Associação Nacional Para a Difusão de Adubos). and winter in the absence of irrigation is warranted and 2019. Pesquisa setorial de macro indicadores. bit. this aspect should be investigated in field studies. ly/3zM7f44 (accessed 24 July 2021). These results are important because, on many farms, AOAC (Association of Official Analytical Chemists). 2005. fertilizing of pastures with organic fertilizer is frequently Official methods of analysis. AOAC International, practiced during all months of the year. Waste is Arlington, VA. produced daily and converted to compost or biofertilizer Araujo LC de; Santos PM; Rodriguez D; Pezzopane JRM. and farmers do not store the waste for lengthy periods 2018. Key factors that influence for seasonal production (mainly biofertilizers, that are very diluted). Nitrogen of guinea grass. Scientia Agricola 75:191–196. doi: 10.1590/1678-992x-2016-0413 applied during seasons when the grass cannot absorb Blum J; Melfi AJ; Montes CR; Gomes TM. 2013. Nitrogen it can lead to an excess of N in the soil, with possible and phosphorus leaching in a tropical Brazilian soil contamination of groundwater (Blum et al. 2013); further cropped with sugarcane and irrigated with treated sewage investigations of this system seem warranted. effluent. Agricultural Water Management 117:115–122. doi: 10.1016/j.agwat.2012.11.010 Conclusions Bowden C; Spargo J; Evanylo G. 2007. Mineralization and N fertilizer equivalent value of composts as assessed Application of synthetic fertilizer (urea) resulted in by tall fescue (Festuca arundinacea). Compost Science greater forage production than application of organic & Utilization 15:111–118. doi: 10.1080/1065657X.2007. fertilizers. However, there are other benefits from 10702320Caminha FO; Carneiro SC da; Paiva AJ; Pereira LET; Mesquita applying organic fertilizer, such as increase in soil P de; Guarda VD. 2010. Stability of tiller population of organic matter, improvement in soil structure, etc. While continuously stocked marandu palisade grass fertilized with the fertilizer N in urea was readily available for plants, nitrogen. Pesquisa Agropecuaria Brasileira 45:213–220. the slow release of N from the biofertilizers does not (In Portuguese) doi: 10.1590/S0100-204X2010000200013 necessarily mean that the remaining N is lost from the Fagundes JL; Fonseca DM da; Morais RV de; Mistura C; system. This N could become available for plant use Vitor CMT; Gomide JA; Nascimento Jr D do; Santos subsequently. Since waste is a by-product of agricultural MER; Lambertucci DM. 2006. Evaluation of structural systems and must be disposed of in an environmentally characteristics of the signalgrass in a nitrogen fertilized safe manner, application to fields to reduce the levels of pasture over the seasons of the year. Revista Brasileira de inorganic fertilizer to be applied is a beneficial practice. Zootecnia 35:30–37. (In Portuguese) doi: 10.1590/s1516-35982006000100004 These aspects warrant further investigation. Grave RA; Nicoloso RS; Cassol PC; da Silva MLB; Mezzari MP; Aita C; Wuaden CR. 2018. Determining the effects Acknowledgments of tillage and nitrogen sources on soil N2O emission. Soil and Tillage Research 175:1–12. doi: 10.1016/j.still.2017. This research was carried out with the financial support 08.011 of the Foundation for the Support to the Development of Gutser R; Ebertseder T; Weber A; Schraml M; Schmidhalter Teaching, Science and Technology of the State of Mato U. 2005. Short-term and residual availability of nitrogen Grosso do Sul (Fundect grant number -0239/2014), after long-term application of organic fertilizers on arable and the Master’s degree scholarship was provided by land. Journal of Plant Nutrition and Soil Science 168:439– the Coordination for the Improvement of Higher Level 446. doi: 10.1002/jpln.200520510McRoberts KC; Parsons D; Ketterings QM; Hai TT; Quan Personnel (CAPES). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 306 S.D.S. dos Reis, M.A.P. Orrico Jr, M. Tomazi, S.S. Cunha, A.C.A. Orrico, J.P. Alves and E.S.J. Galeano. NH; Ba NX; Nicholson CF; Cherney DJR. 2018. Urea and Vargas Jr. FM de. 2013. Morphogenetic characteristics composted cattle manure affect forage yield and nutritive of palisadegrass Piata fertilized with effluent from value in sandy soils of south-central Vietnam. Grass and poultry processing plant. Ciência Rural 43:158–163. (In Forage Science 73:132–145. doi: 10.1111/gfs.12289 Portuguese). doi: 10.1590/S0103-84782012005000125 Mihret B; Asmare B; Mekuriaw Y. 2018. Effect of fertilizer type Orrico Jr. MAP; Silveira AP da; Orrico ACA; Schwingel AW; and plant spacing on plant morphological characteristics, Carnavali PL; Alves DC. 2018. Use of organic compost yield and chemical composition of desho grass (Pennisetum for the fertilization of Piatã and Paiaguás grasses: effects pedicellatum Trin.) in Northwestern Ethiopia. Agricultural of dose on morphogenetic, structural, nutritional, and Science and Technology 10:107–114. bit.ly/3jz9PVE productive characteristics. Compost Science & Utilization Orrico Jr. MAP; Centurion SR; Orrico ACA; Sunada NS. 26:201–208. doi: 10.1080/1065657X.2018.1457998 2012. Effects of biofertilizer rates on the structural, Ryals R; Eviner VT; Stein C; Suding KN; Silver WL. 2016. morphogenetic and productive characteristics of Piatã Grassland compost amendments increase plant production grass. Revista Brasileira de Zootecnia 41:1378–1384. doi: without changing plant communities. Ecosphere 7:e01270. 10.1590/S1516-35982012000600009 doi: 10.1002/ecs2.1270 Orrico Jr. MAP; Orrico ACA; Centurion SR; Sunada NS; (Received for publication 6 October 2020; accepted 24 June 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):307–314 307 doi: 10.17138/TGFT(9)307-314 Research Paper Evaluation of corn-soybean inter-cropping systems in southwestern Japan Evaluación de sistemas de cultivos intercalados de maíz con soya en el suroeste de Japón AHMAD SEYAR AZIZI*1, IKUO KOBAYASHI2, JONATHAN CHUUKA1 AND GENKI ISHIGAKI2 1Graduate School of Agriculture, University of Miyazaki, Miyazaki, Japan. miyazaki-u.ac.jp/english 2Sumiyoshi Livestock Science Station, Field Science Education Research Center, Faculty of Agriculture, University of Miyazaki, Miyazaki, Japan. miyazaki-u.ac.jp/sfield *Ministry of Agriculture, Irrigation and Livestock (MAIL), Agricultural Research Institute of Afghanistan (ARIA), Badam Bagh, Kabul, Afghanistan. mail.gov.af/en Abstract To assess the effects of inter-cropping corn and soybean under southwestern Japan’s climatic conditions, 5 different treatments were compared, namely: CW (mono-cropped corn - weeded); CTW (corn + soybean cv. Tachinagaha - weeded); CT (corn + soybean cv. Tachinagaha - unweeded); CSW (corn + soybean cv. Suzukaren - weeded); and CS (corn + soybean cv. Suzukaren - unweeded). Parameters measured were plant height, yield, nutrient composition of corn and soybean and the numbers of Japanese beetles (Popillia japonica). Plant height of mono-cropped corn was significantly (P<0.05) greater than that of corn in most of the inter-cropped treatments. The number of Japanese beetles had increased dramatically, especially on unweeded inter-cropped treatments, at 55 DAS (days after sowing). Fresh and dry matter yields (FMY and DMY) of corn did not differ among treatments (P>0.05), while CTW treatment produced higher FMY and DMY for soybean (P<0.05) than in CSW and CS. Weeding tended to reduce the number of Japanese beetles on soybean plants, but it did not affect yield of soybean in this study. Neutral detergent fiber (NDF) and acid detergent fiber (ADF) concentrations in corn cobs, whole corn plants and whole soybean plants did not differ among treatments (P>0.05), while crude protein (CP) concentration in whole corn plants in CTW exceeded (P<0.05) those for mono-cropped corn and CSW treatments. These results indicated that soybean can be successfully inter-cropped with corn in southwestern Japan. Soybean plants may be infested with Japanese beetles. It is advisable to control weeds in the stands to reduce the level of beetle infestation and to minimize competition for the planted crops. Keywords: Corn-legume intercropping, crude protein, forage yield, nutrient composition, Popillia japonica. Resumen Para evaluar los efectos del cultivo intercalado de maíz y soja en las condiciones climáticas del suroeste de Japón, se compararon 5 tratamientos diferentes, a saber: CW (maíz monocultivo - desyerbado); CTW (maíz + soja cv. Tachinagaha - desyerbado); CT (maíz + soja cv. Tachinagaha - sin desyerbar); CSW (maíz + soja cv. Suzukaren - desyerbado); y CS (maíz + soja cv. Suzukaren - sin desyerbar). Los parámetros medidos fueron la altura, el rendimiento, la composición de nutrientes del maíz y la soja y el número de escarabajos japoneses (Popillia japonica). La altura de la planta de maíz monocultivo fue significativamente (P<0.05) mayor que la del maíz en la mayoría de los tratamientos entre cultivos. El número de escarabajos japoneses había aumentado drásticamente, especialmente en tratamientos de cultivos intercalados sin deshierbe, a los 55 DAS (días después de la siembra). Los rendimientos de materia fresca y seca (FMY y DMY) del Correspondence: Genki Ishigaki, Sumiyoshi Livestock Science Station, Field Science Education Research Center, Faculty of Agriculture, University of Miyazaki, 10100-1 Shimanouchi, Miyazaki 880-0121, Japan. E-mail: gishigaki@cc.miyazaki-u.ac.jp Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 308 A.S. Azizi, I. Kobayashi, J. Chuuka and G. Ishigaki maíz no difirieron entre los tratamientos (P>0.05), mientras que el tratamiento CTW produjo FMY y DMY más altos para la soja (P<0.05) que en CSW y CS. El deshierbe tendió a reducir el número de escarabajos japoneses en las plantas de soja, pero no afectó el rendimiento de la soja en este estudio. Las concentraciones de fibra detergente neutra (NDF) y fibra detergente ácida (ADF) en mazorcas de maíz, plantas enteras de maíz y plantas enteras de soja no difirieron entre tratamientos (P>0.05), mientras que se excedió la concentración de proteína cruda (CP) en plantas enteras de maíz en CTW (P<0.05) los de los tratamientos de maíz monocultivo y CSW. Estos resultados indicaron que la soja se puede intercalar con el maíz en el suroeste de Japón. Las plantas de soja pueden estar infestadas de escarabajos japoneses. Es aconsejable controlar las malas hierbas en los rodales para reducir el nivel de infestación de escarabajos y minimizar la competencia por los cultivos plantados. Palabras clave: Composición nutricional, cultivos intercalados maíz-leguminosa, Popillia japonica, proteína cruda, rendimiento de forraje. Introduction of corn with soybean in southwestern Japan has not been reported, despite the corn-soybean inter-cropping system Maize (Zea mays L.), also known as corn, is one of the becoming popular worldwide. most important cereal crops in the world, and is grown Ishigaki et al. (2017) investigated the agronomic in a wide range of environments worldwide. It is being characteristics and yield of several soybean varieties as increasingly used for a range of purposes, such as human forage crops, and reported on the risk of insect damage food, feed for livestock and a source of raw material for during the cultivation period. The authors confirmed industrial products. The demand for corn as animal feed that leaf damage caused by scarabs, especially Japanese will continue to grow faster than the demand for its use beetles (Popillia japonica; Coleoptera: Rutelidae), as human food, particularly in Asia, where a doubling was remarkable. Since the cultivation period of forage of production is expected to almost 400 million tonnes soybeans and the time when scarabs are most prolific in 2030 (Paliwal et al. 2000). Although corn is high overlap, it is important to determine the impact of in digestible starch and water-soluble carbohydrates beetle infestation on production of forage soybeans. In (WSC), making it a high-energy feed for ruminants addition, weeds can hamper soybean growth. However, (Masoero et al. 2006), it does not provide sufficient crude there are no registered pesticides or herbicides for use in protein (8.8%) for high levels of production (National forage soybeans in Japan. Therefore, there is a need to Research Council 2001). Therefore, supplementation document the occurrence of Japanese beetles under the with protein feeds is needed to fulfill the requirements corn-soybean inter-cropping system and the effects on of high-producing ruminants. In order to improve yields forage yield and quality of corn in southwestern Japan, and forage quality of corn, alternatives are being sought. with the aim of developing a pest management strategy One suitable alternative may be inter-cropping of for forage soybeans that does not rely on the use of corn with legumes such as soybean [Glycine max (L.) pesticides. Merr.] (Ofori and Stern 1987; Carruthers et al. 1998). The present studies were conducted: (i) to assess the Legumes have long been recognized as a good source occurrence of Japanese beetle on soybean plants under of crude protein (CP) (Anil et al. 2000). However, mono-cropped and inter-cropped cultivation; and (ii) to leguminous forage is highly difficult to ensile because evaluate the effects on growth, forage yield and nutrient of its high buffering capacity and low level of WSC composition of inter-cropping corn with 2 forage soybean (Maasdorp and Titterton 1997), but it is possible to varieties relative to mono-cropped corn. ensile high-energy corn silage with protein-rich forages to obtain a better nutrient composition (Anil et al. 2000). Material and Methods Recently, several studies have found that inter-cropping of corn with legumes is a feasible option to increase CP Experimental site concentration in forage produced (Prasad and Brook 2005; Contreras-Govea et al. 2009; Zhu et al. 2011; The field experiment was carried out at the Sumiyoshi Costa et al. 2012). Herbert et al. (1984) and Putnam Livestock Science Station, Field Science Education et al. (1986) reported that inter-cropping of corn with Research Center, Faculty of Agriculture, University of soybean increased CP concentration by 19‒36 and 11‒15 Miyazaki, Japan (31°55′ N, 131°28′ E; 10 masl), from percentage units, respectively. However, inter-cropping April to July 2018. Meteorological conditions are shown Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Corn-soybean inter-cropping systems in Japan 309 in Table 1. Air temperature and precipitation during the insects (Popillia japonica) were first observed feeding experimental period were obtained from the database on soybean leaves in late May (24‒30 May). The CW, of the Geospatial Information Authority of Japan (jma- CTW and CSW treatments were weeded by hand during net.go.jp/miyazaki), and were recorded 16 km from the the 2nd and 4th weeks after sowing. experimental site. The soil type of the experimental field was characterized as sand-dune Regosol with moderate Table 2. Description of experimental treatments and cropping organic matter (4.0%), 0.09% N, 0.3% P, 0.2% K and a systems. soil pH of 6.2. Treatment Cropping system Description CW Mono-crop Corn - weeded Table 1. Mean temperature and precipitation during April– CTW Inter-crop Corn + soybean cv. July 2018 in the area of the field experiment in Miyazaki, Tachinagaha - weeded southwestern Japan. CT Inter-crop Corn + soybean cv. Month Temperature (°C) Precipitation (mm) Tachinagaha - unweeded April 17.6 60 CSW Inter-crop Corn + soybean cv. Suzukaren May 20.5 409 - weeded June 23.6 480 CS Inter-crop Corn + soybean cv. Suzukaren July 27.5 580 - unweeded Plant material Data collection and analysis In this experiment, a corn cultivar, NS118 Super At 24, 45, 68 and 80 DAS (days after sowing), 3 plants (KANEKO seed company), which matures in a period (each of corn and soybean) were selected at random from of 118 days, and 2 soybean cultivars, Tachinagaha and each treatment for measuring plant height. All Japanese Suzukaren, released by National Agriculture and Food beetles (Popillia japonica) on all plants in each plot (4 × Research Organization (NARO), were used. Tachinagaha 3.2 m) were collected and counted at 49, 55, 57, 66, 70 and is medium-late maturing and Suzukaren is late maturing. 78 DAS. Corn and soybean inter-crops were harvested by using sickles on 13 July 2018. All plants were cut at Experimental design and treatments about 10 cm above the ground in the net plot area (3 × 1.7 m), excluding border rows. Fresh matter yield (FMY) The experimental plots were laid out in a randomized was measured in each experimental plot. To determine complete block design (RCBD) with 5 treatments and dry matter (DM), 10 samples of corn and soybean plants 3 replications for each treatment. The description of were selected from each plot, weighed fresh and dried treatments is summarized in Table 2. Individual treatment in an oven at 70 °C for 48 hours to calculate dry matter plots were 12.8 m2 (4 × 3.2 m) and there were 15 plots yield (DMY). The dried samples were ground to 1 mm in all. The cropping systems were mono-crop corn and for chemical analysis. N concentration was measured corn-soybean inter-cropping. Prior to planting, lime with using the NC-Analyzer (model Sumigraph NC-220F, 10% MgO at 800 kg/ha and compost (2.5% N, 4.0% P Sumika Chemical Analysis Service Ltd, Japan), allowing and 2.1% K) at 1,114 t/ha were added to the soil in March. calculation of crude protein concentration (CP%) and After field preparation, corn seed was sown on 16 April crude protein yield (CPY) for each species and cultivar as 2018 with 5 rows per plot, at an inter-row spacing of 75 well as total CPY/ha. Neutral detergent fiber (NDF) was cm, and intra-row spacing of 25 cm. Sowing rate of corn measured using a modified Ankom Filter bag technique was 68,571 viable seeds/ha and sowing depth was 4–5 (Ankom Technology, ANKOM200 Fiber Analyzer, NY, cm. In inter-cropped treatments, a single row of soybean USA) and acid detergent fiber (ADF) was run sequentially was sown 20 cm from a corn row, with an intra-row after NDF using the Ankom Filter bag technique. spacing of 6 cm, giving a sowing rate of 285,714 viable seeds/ha; sowing depth was 5‒6 cm. No rhizobium Statistical analysis was applied to the soybean seeds and no herbicides or insecticides were used. Basal N:P:K fertilizer (14:12:10) One-way ANOVA was conducted to assess the effects was applied at a rate of 400 kg/ha during corn planting of cropping system and treatment on growth traits and and 80 kg N/ha in the form of urea (46% N) was applied yield traits. Additionally, Tukey’s test was conducted to on 17 May 2018. The crops were inspected manually and assess the means of these traits. Percentage data were Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 310 A.S. Azizi, I. Kobayashi, J. Chuuka and G. Ishigaki transformed into angular figures (Claringbold et al. first observed feeding on soybean leaves in late-May 1953). All statistical analyses were carried out with the (24‒30 May) and was present throughout June, with program R software (version 3.1.1, R Core Team 2014). a few individuals still observed in July. At 49 DAS, only low numbers of beetles were collected (0.7‒7.1 Results individuals/plot, i.e. 4 × 3.2 m) with higher, but not significantly so (P>0.05), numbers on unweeded Plant height treatments. By 55 DAS, insect numbers had increased dramatically, especially on unweeded inter-cropped Plant height of mono-cropped corn (CW) at 24 DAS treatments. There were significant differences (P<0.05) (days after sowing) was significantly (P<0.05) greater between unweeded and weeded treatments for cv. than that for CT treatment (Table 3) and by 45 DAS Tachinagaha but not for cv. Suzukaren. The situation heights of corn for CW and CSW were greater (P<0.05) remained similar at 66 DAS. By 70 DAS and 78 DAS than those for CTW and CT. At 68 and 80 DAS, corn in insect numbers had dropped to fewer than 3 individuals CW was taller (P<0.05) than corn in most inter-cropped per plot on all treatments. Corn plants suffered only treatments (Table 3). Weeding had no significant (P>0.05) minimal damage from insect pests. effect on plant height of corn. For soybean, plant heights at 24 and 45 DAS for CTW and CT were significantly (P<0.05) greater than those for CSW and CS (Table 3). Differences between cultivars declined with time and by 68 DAS only soybean in CTW was taller than soybean in CSW and CS, while by 80 DAS only CT was taller than CS. As for corn, weeding had no effect on height of soybean. Table 3. Plant height (cm) of corn and soybean at different growth stages for the inter-crop and mono-crop treatments. Treatment 24 DAS 45 DAS 68 DAS 80 DAS Corn CW 36±1.3a 233±1.3a 282±1.2a 293±2.0a CTW 34±1.4ab 217±2.9b 268±2.9b 282±2.4b CT 30±0.9b 222±3.8b 271±2.2b 284±2.5b CSW 31±1.5ab 234±2.0a 273±2.4b 287±1.9ab CS 31±1.1ab 225±2.3ab 272±1.5b 284±1.8b Soybean CTW 13±0.4a 53±2.5a 95±2.9a 92±1.7ab CT 12±0.5a 53±1.2a 88±3.4ab 94±2.0a Figure 1. Adult Japanese beetles (Popillia japonica) feeding CSW 10±0.5b 39±1.9b 81±4.1b 87±3.0ab CS 10±0.5b 39±2.6b 76±3.1b 86±1.8b on soybean leaves. Data are presented as mean ± s.e.; means within columns and crop type with different letters are significantly different Yield traits (P<0.05) by Tukey-test. DAS: days after sowing. CW: mono- crop corn - weeded; CTW: corn + soybean cv. Tachinagaha Treatments applied had no significant effects (P>0.05) - weeded; CT: corn + soybean cv. Tachinagaha - unweeded; on fresh (FMY) or dry matter (DMY) yields of corn CSW: corn + soybean cv. Suzukaren - weeded; CS: corn + with FMY ranging from 63.3 to 59.8 t/ha and DMY soybean cv. Suzukaren – unweeded. from 21 to 17.5 t/ha (Table 5). However, failing to weed the inter-cropped treatments produced a non-significant Insect occurrence decrease in both FMY and DMY of corn (mean 9.2%). Soybean yields varied from 4.5 to 6.6 t FM/ha and 1.0 No signs of disease were observed. However, a number to 1.5 t DM/ha, with weeded Tachinagaha outyielding of insect pests were observed during the growing other treatments in terms of DMY (P<0.05). Total DMY season. The most significant one was Japanese beetle ranged from 18.7 to 22.5 t DM/ha with no significant (Popillia japonica) (Figure 1; Table 4), which was differences between treatments (P>0.05; Table 5). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Corn-soybean inter-cropping systems in Japan 311 Table 4. Numbers of Japanese beetle (Popillia japonica) per plot (4 × 3.2 m) collected on soybean plants. Treatment 49 DAS 55 DAS 57 DAS 66 DAS 70 DAS 78 DAS CW - - - - - - CTW 0.7±0.3a 1.8±0.4b 4.0±0.8ab 8.4±1.6b 2.9±1.0a 0.1±0.1a CT 7.1±3.4a 23.2±7.6a 21.8±7.2a 22.0±6.3a 1.2±0.7ab 1.9±1.0a CSW 0.7±0.4a 0.1±0.1b 1.0±0.8b 2.8±0.9b 0.2±0.1b 0.0±0.0a CS 4.1±1.5a 16.2±5.5ab 19.0±6.5ab 12.1±2.6ab 1.1±0.4ab 0.4±0.2a Data are presented as mean ± s.e.; means within columns with different letters are significantly different (P<0.05) by Tukey test. DAS: days after sowing. CW: mono-crop corn - weeded; CTW: corn + soybean cv. Tachinagaha - weeded; CT: corn + soybean cv. Tachinagaha - unweeded; CSW: corn + soybean cv. Suzukaren - weeded; CS: corn + soybean cv. Suzukaren - unweeded. Table 5. Fresh and dry matter yields (t/ha) of corn and soybean. Treatment FMY DMY Total DMY Corn Soybean Corn Soybean CW 62.7±3.5 19.6±1 19.6 CTW 63.3±1.1 6.6±0.1a 21.0±1.2 1.5±0.1a 22.5 CT 59.8±4.1 5.4±0.3ab 17.5±0.8 1.2±0.1b 18.7 CSW 67.3±2.6 4.7±0.3b 20.0±0.9 1.1±0.1b 21.1 CS 60.7±0.8 4.5±0.3b 18.7±0.3 1.0±0.1b 19.7 Data are presented as mean ± s.e.; means within columns with different letters are significantly different (P<0.05) by Tukey test. CW: mono-crop corn - weeded; CTW: corn + soybean cv. Tachinagaha - weeded; CT: corn + soybean cv. Tachinagaha - unweeded; CSW - corn + soybean cv. Suzukaren – weeded; CS: corn + soybean cv. Suzukaren – unweeded. FMY: fresh matter yield; DMY: dry matter yield. Nutrient composition all other treatments except inter-cropped weeded corn + cv. Suzukaren, while CP yields of soybean of both Treatments applied had very little effect on chemical Tachinagaha treatments exceeded those of Suzukaren composition of forage produced (Table 6). The only (P<0.05; Table 7). Similarly, total CP yield of the inter- significant effect was for CP concentration, where CP% cropped weeded Tachinagaha treatment exceeded those of corn forage from the Tachinagaha inter-cropped of all other treatments (P<0.05). NDF concentration in weeded treatment exceeded that of whole corn forage whole corn ranged from 54 to 57.9%, while concentration grown as a sole crop (7.0 vs. 5.3%; P<0.05). As a in soybean ranged from 44.4 to 48.5%. On the other result, CP yield from corn in the Tachinagaha inter- hand, ADF concentrations for corn ranged from 35.1 to cropped weeded treatment exceeded (P<0.05) those of 37.2% and for soybean from 36.9 to 42.1%. Table 6. Nutrient composition (% DM basis) of corn and soybean. Nutrient Treatment CW CTW CT CSW CS NDF Corn - cobs 35.0±1.6 36.6±0.9 38.7±1.7 38.3±0.5 38.7±0.9 Corn – whole plant 56.1±0.4 54.0±1.5 57.9±2.1 54.0±1.1 55.2±0.4 Soybean - whole plant - 44.4±1.8 47.0±0.4 48.5±0.4 47.6±0.8 ADF Corn - cobs 18.4±1.0 20.0±1.0 19.3±1.4 19.5±0.5 19.4±0.2 Corn – whole plant 37.2±0.3 35.1±0.3 37.1±1.4 36.9±0.8 36.3±0.9 Soybean – whole plant plant - 36.9±2.3 39.6±1.0 42.1±0.7 40.0±1.7 CP Corn - cobs 6.3±0.2 5.4±0.2 5.6±0.3 5.9±0.2 6.1±0.2 Corn – whole plant 5.3±0.1b 7.0±0.3a 6.0±0.4ab 5.6±0.1b 5.8±0.2ab Soybean – whole plant - 17.8±0.7 18.9±0.9 17.6±0.9 16.6±0.9 Data are presented as mean ± s.e.; means within columns with different letters are significantly different (P<0.05) by Tukey test. CW: mono-crop corn - weeded; CTW: corn + soybean cv. Tachinagaha - weeded; CT: corn + soybean cv. Tachinagaha - unweeded; CSW: corn + soybean cv. Suzukaren - weeded; CS: corn + soybean cv. Suzukaren - unweeded. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 312 A.S. Azizi, I. Kobayashi, J. Chuuka and G. Ishigaki Table 7. Crude protein yield (t/ha) of corn and soybean. under corn-soybean inter-cropping systems in the study Treatment Corn Soybean Total location. The higher populations of insects in unweeded CW 1.04±0.1b - 1.04±0.1b treatments indicated that weeding has more advantages CTW 1.48±0.3a 0.27±0.03a 1.75±0.2a than merely reducing competition for resources. Our CT 1.04±0.1b 0.23±0.03ab 1.27±0.1b 0.19±0.03b results suggest that weeding and physical insect control CSW 1.13±0.1ab 1.32±0.1b 0.17±0.03b under corn-soybean inter-cropping systems should be CS 1.08±0.1b 1.25±0.1b Data are presented as mean ± s.e.; means within rows with a carried out to prevent possible yield loss. However, different letter are significantly different (P<0.05) by Tukey in Japan, there are no registered pesticides for control test. CW: mono-crop corn - weeded; CTW: corn + soybean cv. of scarab beetles, including Japanese beetle. Thus, Tachinagaha - weeded; CT: corn + soybean cv. Tachinagaha - pesticide-free control methods, such as the use of light unweeded; CSW: corn + soybean cv. Suzukaren - weeded; CS: traps, should possibly be tested for control of this insect corn + soybean cv. Suzukaren - unweeded. under southwestern Japan's climatic conditions. Discussion Yield traits Plant height The absence of significant differences in total dry matter yield (TDMY) between mono-cropped corn and The reduction in plant height of corn in inter-cropped corn-soybean inter-cropped treatments suggests that treatments relative to the mono-cropped corn treatment competition for resources prevented any marked increase would have been a response to increased competition in dry matter production from a given area regardless of for light, moisture and nutrients, where corn was sown species involved. However, absolute DMYs for the inter- with the legumes. Baker (1979) and Mbah et al. (2007) cropped weeded treatments were greater than for mono- indicated that inter-cropped treatments have reduced cropped corn, suggesting that more and larger studies growth relative to mono-crops because of competition for should be conducted to verify these findings. Similarly, resources. This hypothesis is reinforced by the fact that the consistent, though non-significant, reductions in differences were not significant during the early stages yields of forage in unweeded treatments also suggest of growth but increased as plants matured and available that more and larger studies seem warranted to verify resources were consumed. Surprisingly, weeding of these findings. Reta Sánchez et al. (2010) and Baghdadi inter-cropped treatments had no significant effect on et al. (2016) reported that total DMY of corn-soybean plant height of corn, despite the fact that competition inter-cropped treatments was similar to that of mono- for resources in the unweeded treatments might have cropped corn. The higher DMYs for Tachinagaha than been expected to be greater than in weeded treatments. for Suzukaren suggest that the former cultivar might be While Tachinagaha plants were significantly taller than more suitable for growing in this environment. Different Suzukaren plants in the early growth stages, differences competitive ability of the soybean cultivars could be a were no longer significant at the end of the study, despite factor in the different yields displayed (Callaway 1992), absolute values favoring Tachinagaha. These differences although we did not have any pure legume treatments appear to be merely varietal differences as opposed to in this study. Gutu et al. (2015) in their experiment on any superiority of Tachinagaha in terms of competitive corn-soybean inter-cropping also stated that forage and ability. Unfortunately, we failed to record biomass grain yields of soybean were significantly different for yields of weeds in unweeded treatments, as Rose et al. different soybean varieties. (1984) reported up to 45% differences in weed biomass production when in competition with various soybean Nutrient composition genotypes. The absence of significant differences in NDF and ADF Occurrence of insects of corn in the different treatments was not surprising, as all crops were harvested at the same maturity stage The significant insect injury to soybean caused by of corn plants as suggested by Mugweni et al. (2001). Japanese beetle (Popillia japonica) was not surprising While CP concentration in corn forage in inter-cropped as this insect is a voracious feeder on soybean leaves treatments was generally not significantly greater than and could pose problems for soybean growth and yield that of mono-cropped corn, absolute values consistently Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Corn-soybean inter-cropping systems in Japan 313 favored the inter-cropped treatments. Added to this was and forages: weed control by intercrops combined with the much higher CP concentration in the soybean forage interrow cultivation. European Journal of Agronomy than in the corn forage, so overall forage produced in 8:225–238. doi: 10.1016/S1161-0301(97)00062-2 inter-cropped treatments was of higher quality than the Claringbold PJ; Biggers JD; Emmens CW. 1953. The angular mono-cropped corn. As a result, animals fed the mixed transformation in quantal analysis. Biometrics 9:467–484. forage would be expected to perform at a higher level doi: 10.2307/3001438Contreras-Govea F; Muck RE; Armstrong KL; Albrecht than those fed mono-cropped corn. Numerous studies, KA. 2009. Nutritive value of corn silage in mixture with e.g. Lithourgidis et al. (2006), and Eskandari et al. climbing beans. Animal Feed Science and Technology (2009), have reported similar results indicating that 150:1–8. doi: 10.1016/j.anifeedsci.2008.07.001 inter-cropping cereal crops with legumes significantly Costa PM; Villela SDJ; Leonel FP; Araújo SAC; Araújo KG; increased the CP yield per hectare. Ruas JRM; Coelho FS; Andrade VR. 2012. Intercropping In conclusion, the findings of this study clearly of corn, brachiaria grass and leguminous plants: showed that corn and soybean can be inter-cropped with productivity, quality and composition of silages. Revista simultaneous sowing under southwestern Japan's climatic Brasileira de Zootecnia 41:2144–2149. doi: 10.1590/ conditions without any deleterious effects on the corn. S1516-35982012001000002 While dry matter yields of forage produced might not be Eskandari H; Ghanbari A; Galavi M; Salari M. 2009. Forage quality of cow pea (Vigna sinensis) intercropped with significantly greater than that for mono-cropped corn, the corn (Zea mays) as affected by nutrient uptake and light overall quality of the forage would be superior in the inter- interception. Notulae Botanicae Hort Agrobotanici Cluj- cropped system, especially in terms of CP concentration. Napoca 37:171–174. bit.ly/37Ri6h5 It appears that cv. Tachinagaha might be superior to Gutu T; Tamado T; Negash G. 2015. Effect of varieties and cv. Suzukaren for growing in this environment. Again, population of intercropped soybean with maize on yield it appears that weeding may reduce the level of insect and yield components at Haro Sabu, Western Ethiopia. infestation on the soybeans and possibly increase yields. Science, Technology and Arts Research Journal 4:31–39. These findings need verification on a larger scale and in a doi: 10.4314/star.v4i4.5 range of seasonal conditions to ensure recommendations Herbert SJ; Putnam DH; Poss-Floyd MI; Vargas A; Creighton are soundly based. Further investigations are required JF. 1984. Forage yield of intercropped corn and soybean in various planting patterns. Agronomy Journal 76:507–510. to determine the most effective non-chemical control doi: 10.2134/agronj1984.00021962007600040001x methods for insects (Japanese beetle) and for control of Ishigaki G; Arai M; Fukuyama K. 2017. Development of invasive weeds in corn-soybean inter-cropping systems soybean production technique by living multi method with under southwestern Japan's climatic conditions. tropical grasses in southwestern Japan. Final Reports for Research Grants for Meat and Meat Products 36:400–405. References (In Japanese). Lithourgidis A; Vasilakoglou IV; Dhima K; Dordas C; (Note of the editors: All hyperlinks were verified 13 July 2021). Yiakoulaki M. 2006. Forage yield and quality of common vetch mixtures with oat and triticale in two seeding Anil L; Park J; Phipps RH. 2000. The potential of forage– ratios. Field Crops Research 99:106–113. doi: 10.1016/j. maize intercrops in ruminant nutrition. Animal Feed fcr.2006.03.008 Science and Technology 86:157–164. doi: 10.1016/S0377- Maasdorp BV; Titterton M. 1997. Nutritional improvement 8401(00)00176-0 of maize silage for dairying: mixed-crop silages from sole Baghdadi A; Halim RA; Ghasemzadeh A; Ebrahimi M; and intercropped legumes and a long-season variety of Othman R; Yusof MM. 2016. Effect of intercropping of maize. 1. Biomass yield and nutritive value. Animal Feed corn and soybean on dry matter yield and nutritive value of Science and Technology 69:241–261. doi: 10.1016/S0377- forage corn. Legume Research 39:976–981. doi: 10.18805/ 8401(97)81639-2 lr.v39i6.6643 Masoero F; Rossi F; Pulimeno AM. 2006. Chemical Baker EFI. 1979. Mixed cropping in northern Nigeria. III. composition and in vitro digestibility of stalks, leaves and Mixtures of cereals. Experimental Agriculture 15:41–48. cobs of four corn hybrids at different phenological stages. doi: 10.1017/S0014479700009182 Italian Journal of Animal Science 5:215–227. doi: 10.4081/ Callaway M. 1992. A compendium of crop varietal tolerance ijas.2006.215 to weeds. American Journal of Alternative Agriculture Mbah E; Muoneke C; Okpara D. 2007. Effect of compound 7:169–180. doi: 10.1017/S088918930000477X fertilizer on the yield and productivity of soybean and maize Carruthers K; Prithivirj B; Fe Q; Cloutier D; Martin RC; in soybean/maize intercrop in southeastern Nigeria. Tropical Smith DL. 1998. Intercropping corn with soybean, lupin and Subtropical Agroecosystems 7:87–95. bit.ly/3r9rK7B Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 314 A.S. Azizi, I. Kobayashi, J. Chuuka and G. Ishigaki Mugweni BZ; Titterton M; Maasdorp BV; Gandiya F. 2001. soybean density studies. II. Yield composition and protein. Effect of mixed cereal-legume silages on milk production Experimental Agriculture 22:373–381. doi: 10.1017/ from lactating holstein dairy cows (R7010). In: Smith T; S0014479700014629 Godfrey SH, eds. Sustaining livestock in challenging dry R Core Team. 2014. R: A language and environment for season environments. Proceedings of the 3rd Workshop statistical computing. R Foundation for Statistical on Livestock Production Programme Projects. Matobo, Computing, Vienna, Austria. R-project.org/ Zimbabwe. 26–28 September 2000. p. 82–89. bit.ly/37WVpId Reta Sánchez DG; Espinosa Silva JT; Palomo Gil A; Serrato Ofori F; Stern W. 1987. Cereal–legume intercropping systems. Corona JS; Cueto-Wong JA; Gaytán Mascorro A. 2010. Advances in Agronomy 41:41–90. doi: 10.1016/S0065- Forage yield and quality of intercropped corn and soybean 2113(08)60802-0 in narrow strips. Spanish Journal of Agricultural Research Paliwal RL; Granados G; Lafitte HR; Violic AD. 2000. 3:713-721. doi: 10.5424/sjar/2010083-1269 Tropical maize: improvement and production. Food and Rose JS; Burnside OC; Specht JE; Swisher BA. 1984. Agriculture Organization (FAO), Rome, Italy. Competition and allelopathy between soybeans and weeds. Prasad RB; Brook RM. 2005. Effect of varying maize Agronomy Journal 76:523–528. 10.2134/agronj1984.0002 densities on intercropped maize and soybean in Nepal. 1962007600040005x Experimental Agriculture 41:365–382. doi: 10.1017/ Zhu Y; Bai CS; Guo XS; Xue YL; Ataku K. 2011. Nutritive S0014479705002693 value of corn silage in mixture with vine peas. Animal Putnam DH; Herbert SJ; Vargas A. 1986. Intercropped corn- Production Science 51:1117–1122. doi: 10.1071/AN11125 (Received for publication 14 August 2020; accepted 12 July 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):315–320 315 doi: 10.17138/TGFT(9)315-320 Research Paper Quality properties of sunn hemp (Crotalaria juncea L.) and maize (Zea mays L.) silages Propiedades de calidad de ensilaje de Crotalaria (Crotalaria juncea L.) y maíz (Zea mays L.) GÜLCAN DEMİROĞLU TOPÇU AND ŞÜKRÜ SEZGİ ÖZKAN Ege University, Faculty of Agriculture, Department of Field Crops, Izmir, Turkey. agr.ege.edu.tr Abstract Maize is an ideal forage crop for ensilage because of its high levels of fermentable carbohydrates, although it is low in protein. Sunn hemp is a legume with a high crude protein content with potential to be used in combination with maize to provide a silage with a higher protein content. Different percentages of sunn hemp-maize mixtures of 80-20, 60-40, 40-60 and 20-80 respectively were compared to silages of sole maize and sunn hemp. In the laboratory study, DLG classifications (color, smell, structure, total score and quality class), silage loss (%), silage pH, dry matter content, flieg score, crude protein content, crude ash content, NDF, ADF, metabolic energy (MJ kg/DM), dry matter intake, percent digestible dry matter and relative feed value were determined at the end of 60 days ensilage. The crude protein contents of silages increased as the sunn hemp ratio in the mixtures increased. In addition, pure sunn hemp silage and mixtures, especially 80% sunn hemp mixed with 20% maize, were found suitable for silage and it was concluded that sunn hemp and sunn hemp-maize silage mixtures could be used in animal husbandry. Keywords: Ensilage, feed value, forage crop, legume mixture, protein supply. Resumen El maíz es un cultivo forrajero ideal para ensilaje por sus altos niveles de carbohidratos fermentables, aunque es bajo en proteínas. La crotalaria es una leguminosa con un alto contenido de proteína cruda con potencial para ser utilizada en combinación con el maíz para proporcionar un ensilaje con un mayor contenido de proteína. Se compararon diferentes porcentajes de mezclas de crotalaria-maíz de 80-20, 60-40, 40-60 y 20-80 respectivamente con ensilajes de sólo maíz y crotalaria. En el estudio de laboratorio, clasificaciones DLG (color, olor, estructura, puntaje total y clase de calidad), pérdida de ensilado (%), pH del ensilaje, contenido de materia seca, puntaje de flieg, contenido de proteína bruta, contenido de ceniza bruta, NDF, ADF, energía metabólica (MJ kg/MS), consumo de materia seca, porcentaje de materia seca digestible y el valor relativo del alimento se determinaron al final de los 60 días de ensilaje. El contenido de proteína cruda de los ensilajes aumentó a medida que aumentaba la proporción de crotalaria en las mezclas. Además, el ensilaje de crotalaria pura y las mezclas, especialmente el 80% de crotalaria mezclada con el 20% de maíz, resultaron adecuados para el ensilaje y se concluyó que la crotalaria y las mezclas de ensilaje de maíz y crotalaria se podrían utilizar en la cría de animales. Palabras clave: Cultivo forrajero, ensilaje, mezcla de legumbres, suministro de proteínas, valor alimenticio. Correspondence: Gulcan Demiroglu Topcu, Ege University, Agriculture Faculty, Field Crops Department, 35100 Bornova, Izmir/Turkey. E-mail: gulcan.demiroglu.topcu@ege.edu.tr Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 316 G. Demiroğlu Topçu and Ş.S. Özkan Introduction available potassium (400 ppm). The climate and soil properties at Izmir, Turkey are favorable for sunn hemp The constant growth of the world population requires cultivation. the development of productive and efficient agricultural The Tillage Sun cultivar of sunn hemp and C-955 practices to meet food demand. Turkey is very suitable cultivar of silage maize were used for the study. The plants for animal husbandry with the required natural resources were grown separately in the field on approximately and ecological conditions and the livestock sector is 0.004 ha each. The plants were simultaneously sown at important for the general economy (Seydosoglu 2019). the beginning of July 2019 under second crop planting Roughages, one of the indispensable feed sources of conditions. The sunn hemp was seeded in rows with 40 livestock, are mainly obtained from meadow pasture cm row spacing (50 kg/ha) and maize was sown in rows areas and forage crop production in field agriculture with 70 cm row spacing (95.238 plants/ha). Conventional (Moore et al. 2020). According to 2018 data, total hay agriculture practices were carried out during the growing production of these two sources in Turkey was 31 million season. Since there were no significant problems of tons, which is not sufficient for livestock needs (Acar et pests, diseases or weeds in the study site, no chemical al. 2020). was applied. Turkey's ecological conditions and topography allows the cultivation of many forage plants (Tan and Plant harvesting and ensiling process Yolcu 2021). In addition to traditional forage crops, growing alternative forage crops will help to prevent Sunn hemp was harvested at the beginning of feed shortages. Many studies reported that sunn hemp flowering and maize was harvested at the dough stage (Crotalaria juncea L.), which was previously not simultaneously from the middle rows of the plots. Plants cultivated in Turkey, can be successfully grown in regions were harvested by hand by cutting at soil level. Plants with Mediterranean climatic conditions (Demiroğlu were chopped to about 2-3 cm in size and the chopped Topçu and Özkan 2019). materials were thoroughly mixed to attain homogeneity. Maize silage is one of the most important forage Different mixture ratios (sunn hemp %-maize %; 100- crops worldwide thanks to its high biomass yield, high 0, 80-20, 60-40, 40-60, 20-80, 0-100; on the basis of energy values and the high content of non-structural fresh weight) were used for preparing the silage in four carbohydrates that favor the fermentation process. replications with a randomized design (6 treatments x 4 Protein contribution to the ruminal system is low in replications = 24 silage samples). maize silage (Colombini et al. 2010). Legume forage Samples of 500±20 g of each silage mixture were crops such as sunn hemp and soybean can be ensiled by placed in separate vacuum bags (thickness 110 microns mixing up to 50% with maize in order to increase the or more), and after 99.9% of the air was removed by protein content in maize silage (Zavala et al. 2011; Sulas vacuum, bags were glued and closed (Johnson et al. et al. 2012). 2005). The bagged silage samples were stored in a dark The aim of this study was to determine the sensory and cool environment at 24±4°C for 60 days. and chemical silage quality properties of silages prepared with different ratios of maize and sunn hemp under Assessment of silage samples Mediterranean climate conditions. Physical quality analysis for DLG classifications (DLG Materials and Methods 1987) and chemical properties were examined after 60 days of fermentation. The silage loss was calculated Site description and agronomic details according to Danley et al. (1973). The pH of the silage juice was measured by the HANNA HI 2211 pH/ORP This research was carried out in the experimental fields pH meter (Hanna Instruments Ltd., USA). The silage and silage laboratories of the Field Crops Department, samples were dried in an oven at 65°C for 48 hours, Faculty of Agriculture, Ege University in Izmir, Turkey ground and passed through a 2 mm sieve to prepare for during the 2019 growth season. Izmir has typical chemical analysis. Nitrogen content was determined by Mediterranean climate conditions and a silty-clay loam the Kjeldahl method and multiplied by the coefficient of soil with pH 7.8, organic matter (1.13%), salt (0.075%), 6.25 to obtain crude protein concentrations (AOAC 1990). total N (0.11%), available phosphorus (40 ppm) and Crude ash content was determined at 550ºC (Bulgurlu Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Determination of quality properties in sunn hemp-maize silages 317 and Ergul 1978). The neutral detergent fibre (NDF) and DLG classifications of the silage samples are given in acid detergent fibre (ADF) concentrations were analysed Table 1. The total DLG score obtained by summing the by the sequential detergent analysis method (Van color, smell and structure scores is an indicator of the Soest et al. 1991). Flieg score, metabolizable energy quality class of silage (20-18:very good, 17-14:good, (ME), estimated dry matter intake (DMI), digestible 13-10:medium, 9-5:low, 4-0:deteriorated) (DLG 1987). dry matter (DDM) and relative feed value (RFV) were In terms of sensory properties, all silage alternatives determined by using the following commonly used (sole crop and mixtures) were classed as high quality. formulas (Kirchgessner and Kellner 1977; Van Dyke and There were significant differences (P<0.05) in silage Anderson 2000; Morrison 2003). losses of fresh matter and silage pH among pure and Flieg score = 220 + (2 x % Dry Matter - 15) - 40 x pH mixed silages (Table 2). Silage pH values increased Dry Matter Intake (% of BW) = 120 / (% NDF) as the ratio of sunn hemp in the silage was increased. Digestible Dry Matter (%) = 88.9 - (0.779 x % ADF) Significant differences (P<0.05) were found among Relative Feed Value = (DMI x DDM) / 1.29 silage types for dry matter content and Flieg score Metabolizable Energy, MJ kg/DM = 14.70 - 0.150 x (Table 2). Dry matter content and Flieg score values of ADF silages decreased as the ratio of sunn hemp increased. Crude protein and crude ash values were found to be Statistical analysis significantly different (P<0.05) among the sunn hemp and maize silages (Table 2). Statistical analyses were conducted using ANOVA, The silage NDF, ADF and metabolizable energy Statistical Analysis System version 7.0 (SAS Institute values were found to be statistically significantly 1998) for a completely randomized design. The treatment (P<0.05) different in sole crop and different ratios of means were compared by the LSD test described by Steel sunn hemp and maize mixtures (Table 3). and Torrie (1980). Significant differences (P<0.05) in dry matter intake, digestible dry matter content and relative feed value Results were observed among the silage mixtures (Table 3). The relative feed value of the silage decreased as the ratio of Physical observation values (color, smell, structure) and maize in the mixture decreased. Table 1. Physical observation values and DLG classification of sunn hemp-maize silage. Mixtures Color (point) Smell (point) Structure (point) DLG (point) Quality Class 100% S 2.00 12.75 4.00 18.75 Very Good 100% M 2.00 14.00 4.00 20.00 Very Good 80% S + 20% M 2.00 13.25 4.00 19.25 Very Good 60% S + 40% M 2.00 13.75 4.00 19.75 Very Good 40% S + 60% M 2.00 14.00 4.00 20.00 Very Good 20% S + 80% M 2.00 14.00 4.00 20.00 Very Good S=sunn hemp, M=maize Table 2. Silage loss, pH, dry matter, flieg score, crude protein and crude ash contents of sunn hemp-maize silage Mixtures Silage Loss Silage Dry Matter Flieg Score Crude Protein Crude Ash(%) pH (g/kg) (score) (g/kg DM) (g/kg DM) 100% S 3.06 a 4.44 a 273.4 e 81.95 d 163.8 a 70.7 a 100% M 2.47 d 3.93 d 339.0 a 115.59 a 82.2 f 62.1 d 80% S + 20% M 2.83 b 4.30 b 288.3 d 90.53 c 148.8 b 69.4 a 60% S + 40% M 2.60 c 4.13 c 306.4 c 101.09 b 130.1 c 66.9 b 40% S + 60% M 2.62 c 4.10 c 313.2 c 103.51 b 115.8 d 65.1 c 20% S + 80% M 2.52 cd 3.98 d 325.8 b 111.09 a 100.2 e 63.1 d Mean 2.68 4.15 307.7 100.63 123.5 66.2 CV (%) 2.97 1.43 2.24 3.02 2.47 1.75 LSD (0.05) 0.12 0.09 10.4 4.58 4.6 1.7 S=sunn hemp, M=maize. Means followed by different letters are significantly different (P<0.05). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 318 G. Demiroğlu Topçu and Ş.S. Özkan Table 3. Silage quality features of sunn hemp-maize silage. Mixtures NDF ADF ME DMI DDM(g/kg DM) (g/kg DM) (MJ kg-1 DM) (% of BW) (%) RFV 100% S 603.8 a 422.7 a 8.36 f 1.99 d 55.97 d 86.23 e 100% M 520.8 d 318.7 f 9.92 a 2.30 a 64.08 a 114.46 a 80% S + 20% M 578.1 b 402.7 b 8.66 e 2.08 c 57.53 c 92.58 d 60% S + 40% M 572.6 b 387.6 c 8.89 d 2.10 bc 58.71 c 95.38 d 40% S + 60% M 555.3 c 356.6 d 9.35 c 2.16 b 61.12 b 102.44 c 20% S + 80% M 529.7 d 340.2 e 9.60 b 2.27 a 62.40 b 109.59 b Mean 560.1 371.4 9.13 2.15 59.97 100.11 CV (%) 1.72 1.94 1.27 2.04 1.56 3.04 LSD (0.05) 14.5 1.08 0.17 0.07 1.41 4.58 S=sunn hemp, M=maize, DMI=Dry Matter Intake, DDM=Digestable Dry Matter, RFV=Relative Feed Value. Means followed by different letters are significantly different (P<0.05). Discussion negatively correlated with NDF and the digestible dry matter of silage is inversely related with ADF (Yucel et The highest quality silage was determined as the mixture al. 2018). The metabolizable energy values of all silage of 80% sunn hemp with 20% maize. The findings of samples examined in the study were at an acceptable this study are similar with other studies on different level (Boguhn et al. 2003) and similar to the 8.2 MJ legumes with maize (Budakli Carpici 2016; Titterton and kg/DM reported for sunn hemp silage (Titterton and Maasdorp 1997). Loss of silage dry matter and decrease Maasdorp 1997). Relative feed value has been used to in the feed value of silage of 3-5% could occur due to compare the quality of legume and legume/grass hays or respiration or fermentation (Buxton et al. 2003). Wang et silages (Jeranyama and Garcia 2004) and was positively al. (2009) reported that pure sunn hemp silage has a high correlated with dry matter intake and digestible dry pH value. The results of our study were in accordance matter contents of silage (Yucel et al. 2018). According with the previous research and pH values increased as to the quality classification of Rohweder et al. (1978), the ratio of sunn hemp increased (Zavala et al. 2011). the sole sunn hemp and mixtures of sunn hemp-maize The fermentation of silages may be adversely affected silage studied were of acceptable quality for use as feed. by dry matter. Panyasak and Tumwasorn (2015) reported that the dry matter content of well fermented silage Conclusions should be between 25-40%. The high dry matter values obtained indicate that the soluble carbohydrate content Maize silage is used extensively in Turkish dairy rations per unit dry matter of silage was high and the lactic acid to address the inadequacy of quality feed supply. The fermentation was also successful. results of this study showed that sunn hemp could make The Flieg score value provides a practical assessment good silage and improve the nutritive value of maize of the chemical properties of silage. All ratios of sunn silage. Sole sunn hemp silage and the mixture of 80% hemp with maize in this study were included in the "very sunn hemp and 20% maize were found suitable for good" quality class and Flieg scores were inversely making good quality silage and it was concluded that proportional to silage pH as expected (Woolford 1984). they could be used in animal husbandry. Crude protein contents increased as the ratio of sunn hemp was increased. Budakli Carpici (2016) found that References crude protein contents of mixtures of different silages varied between 7.08–17.43% and Martínez-García (Note of the editors: All hyperlinks were verified 6 September 2021). (2015) reported increased crude protein content as a result of increases in legume ratio in silage mixtures Acar Z; Tan M; Ayan I; Onal Asci O; Mut H; Basaran U; similar to that reported in this study. The NDF content is Gulumser E; Can M; Kaymak G. 2020. Forage crops situation and development possibilities of agriculture in lower than found by Titterton and Maasdorp (1997) who Turkey. Turkey Agricultural Engineering IX. Technical reported a NDF value of 675 g kg/DM for sunn hemp Congress, Ankara, Turkey, 13–17 January 2020, silage. The results of the study indicated that as the sunn Proceedings Book, 1:529–553. hemp ratio is increased in the mixtures, the NDF and ADF AOAC (Association of Official Analytical Chemists). 1990. ratios decreased. The dry matter intake ratio of silage is Official methods of analysis. 15th Edn. AOAC Inc., Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Determination of quality properties in sunn hemp-maize silages 319 Arlington, VA, USA. performance of Pelibuey lambs. Tropical Animal Health Boguhn J; Kluth H; Steinhöfel O; Peterhänsel M; Rodehutscord and Production, 47(8):1561‒1566. doi: 10.1007/s11250- M. 2003. Nutrient digestibility and prediction of 015-0898-0 metabolizable energy in total mixed rations for ruminants. Moore, KJ; Collins M; Nelson CJ; Redfearn DD. 2020. Forages, Archives of Animal Nutrition 57(4):253–266. doi: Volume 2: The science of grassland agriculture. John Wiley 10.1080/00039420310001594405 & Sons., Croydon, UK. doi: 10.1002/9781119436669 Budakli Carpici E. 2016. Nutritive values of soybean silages Morrison JA. 2003. Hay and pasture management. In Illinois ensiled with maize at different rates. Legume Research Agronomy Handbook, Urbana IL, USA. bit.ly/3hqrVrn 39(5):810–813. doi: 10.18805/lr.v0iOF.3772 Panyasak A; Tumwasorn S. 2015. Effect of moisture content and Bulgurlu S; Ergul M. 1978. Physical, chemical and biological storage time on sweet corn waste silage quality. Walailak analyses methods of feeds. Ege University Agriculture Journal of Science and Technology 12(3):237‒243. bit.ly/ Faculty Press, Izmir, Turkey. (In Turkish). 3nkZfE5 Buxton DR; Muck RE; Harrison JH. 2003. Silage science Rohweder DA; Barnes RF; Jorgensen N. 1978. Proposed and technology. Agronomy. 42. ASA, CSSA, and SSSA, hay grading standards based on laboratory analyses Madison, WI, USA. doi: 10.2134/agronmonogr42 for evaluating quality. Journal of Animal Science Colombini S; Reis RB; Broderick GA; Combs DK. 2010. 47(3):747‒759. doi: 10.2527/jas1978.473747x Rumen protein degradability in tropical grasses: SAS Institute. 1998. INC SAS/STAT user’s guide release 7.0, Comparison of results obtained using fluorimetric and Cary, NC, USA. colorimetric O-phthalaldehyde assays for degradation Seydosoglu S. 2019. Effects of different mixture ratios of grass products. In: Crovetto GM, ed. Energy and protein pea (Lathyrus sativus L.) and barley (Hordeum vulgare) metabolism and nutrition. EAAP Publication No. on quality of silage. Legume Research 42(5):666‒670. doi: 127. Wageningen Academic Publishers, Wageningen, 10.18805/LR-468 Netherlands. p. 727‒728. Steel RGD; Torrie JH. 1980. Principles and procedures of Danley MM; Vetter RL; Wedin WF. 1973. Modified laboratory statistics. A biometrical approach. 2nd Edition. McGraw- silo unit for studying the fermentation of maize (Zea mays Hill Book Company, New York, USA. L.) grain. Agronomy Journal 65:621-624. 10.2134/agronj1 Sulas C; Ruda P; Salis M; Atzori AS; Correddu F; Cannas A; 973.00021962006500040028x Carroni AM. 2012. Legume-cereal mixtures ensiling in Demiroğlu Topçu G; Özkan SS. 2019. Effects of different Sardinia. Options Meiterraneennes 102:489‒492. bit.ly/ growing stages on the yield and some forage quality 3oTTg74 characteristics of sunn hemp (Crotalaria juncea L.) grown Tan M; Yolcu H. 2021. Current status of forage crops as second crop in mediterranean climate condition. COMU cultivation and strategies for the future in Turkey: A Journal of Agriculture Faculty 7(1):119‒126. (In Turkish) review. Journal of Agricultural Sciences, 27(2):114‒121. doi: 10.33202/comuagri.492881 doi: 10.15832/ankutbd.903732 DLG (Deutsche Landwirtschafts-Gesellschaft). 1987. Bewertung Titterton M; Maasdorp BV. 1997. Nutritional improvement of von Grünfutter, Silage und Heu, Merkblatt No. 224. DLG, maize silage for dairying: mixed crop silages from sole and Frankfurt am Main, Deutschland. intercropped legumes and a long season variety of maize. Jeranyama P; Garcia AD. 2004. Understanding relative feed 2. Ensilage. Animal Feed Science and Technology, 69(1- value (RFV) and relative forage quality (RFQ). Extension 3):263-270. doi: 10.1016/S0377-8401(97)81640-9 Extra Paper 352. SDSU Extension Service, Brookings, Van Dyke NJ; Anderson PM. 2000. Interpreting a forage SD, USA. openprairie.sdstate.edu/extension_extra/352 analysis. Circular ANR-890. Alabama Cooperative Johnson HE; Merry RJ; Davies DR; Kell DB; Theodorou MK; Extension System, Auburn, AL, USA. bit.ly/3nohuIB Griffith GW. 2005. Vacuum packing: a model system for Van Soest PJ; Robertson JB; Lewis BA. 1991. Method laboratory‐scale silage fermentations. Journal of Applied for dietary fiber, neutral detergent fiber, and nonstarch Microbiology 98(1):106‒113. doi: 10.1111/j.1365-2672. polysaccharides in relation to animal nutrition. Journal of 2004.02444.x Dairy Science 74(10):3583‒3597. doi: 10.3168/jds.s0022- Kirchgessner M; Kellner RJ. 1977. Zur schätzung der 0302(91)78551-2 umsetzbaren energie von grün‐und rauhfutter mit einfachen Wang S; Chen C; Yu T; Liu H. 2009. Study on ensiling of kenndaten. Zeitschrift für Tierphysiologie Tierernährung Crotalaria juncea L. Journal of Taiwan Livestock Research und Futtermittelkunde 38(1‐6):297‒301. doi: 10.1111/ 42(4):309‒318. j.1439-0396.1977.tb00240.x Woolford MK. 1984. The Silage Fermentation. Grassland Martínez-García CG; Valencia-Núñez K; Bastida-López Research Inst Press, Hurley, England. J; Estrada-Flores JG; Miranda-de la Lama GC; Cruz- Yucel C; Inal I; Yucel D; Hatipoglu R. 2018. Effects of mixture Monterrosa RG; Rayas-Amor AA. 2015. Effect of ratio and cutting time on forage yield and silage quality of different combinations of soybean-maize silage on its intercropped berseem clover and italian ryegrass. Legume chemical composition, nutrient intake, degradability, and Research 41(6):846‒853. doi: 10.18805/LR-400 Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 320 G. Demiroğlu Topçu and Ş.S. Özkan Zavala D; Valencia E; Randel PF; Ramos-Santana R. L.) for silage production. The Journal of Agriculture of the 2011. Botanical composition, yield and fermentative University of Puerto Rico 95(3-4):133‒149. doi: 10.46429/ characteristics of lablab (Lablab purpureus L.) and sunn jaupr.v95i3-4.2571 hemp (Crotalaría juncea L.) with yellow corn (Zea mays (Received for publication 19 January 2021; accepted 2 September 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):321–336 321 doi: 10.17138/TGFT(9)321-336 Artículo científico Criterios de uso y conservación de árboles en potreros basados en el conocimiento local de los ganaderos en una zona de bosque seco tropical en Colombia Criteria for use and conservation of trees in pastures based on farmers’ local knowledge in a tropical dry forest zone in Colombia NELSON PÉREZ-ALMARIO1,2, ELIANA LIZETH MEDINA-RIOS2, JAIRO MORA-DELGADO2, DAGOBERTO CRIOLLO-CRUZ1 Y JULIAN ROBERTO MEJÍA1 1Corporación Colombiana de Investigación Agropecuaria – Agrosavia, Colombia. agrosavia.co 2Grupo Sistemas Agroforestales Pecuarios, Universidad del Tolima, Ibagué, Colombia. ut.edu.co Resumen Se atribuye a los árboles un papel importante en las fincas ganaderas, cumpliendo diversas funciones. El estudio documenta la conservación de especies leñosas en fincas ganaderas con base en el conocimiento local y técnico en una región seca de la parte alta de la cuenca del rio Magdalena, Colombia. Se aplicaron 195 cuestionarios semiestructurados para identificar la percepción de los productores sobre la clasificación y usos de especies arbóreas, con base en criterios físicos, nutricionales, fenológicos y ambientales, como indicadores de conservación y uso de las especies en potreros. Con estos indicadores se construyeron índices que identificaron características importantes de las especies mencionadas por los ganaderos. Los datos se analizaron con estadística descriptiva, comparaciones de media y técnicas de análisis multivariados. Los productores aprecian a las especies con base en criterios de uso y funciones relacionadas con su actividad productiva. Seis especies altamente valoradas: Gliricidia sepium, Guazuma ulmifolia, Pithecellobium dulce, Albizia guachapele, Acacia farnesiana y Albizia saman coinciden con otros estudios de conocimiento local. Adicionalmente, el trabajo aporta información relevante de otras doce especies que no han sido reportadas en estudios previos. Se sugiere que el valor y uso potencial de estas especies para fincas ganaderas debe investigarse más a fondo. Palabras clave: Árboles multipropósito, forraje, investigación participativa, reconocimiento, sistemas silvopastoriles, zonas secas. Abstract Trees are attributed an important role in livestock farms, fulfilling various functions. The study documents the retention of woody species in cattle farms from local and technical knowledge in the upper part of the Magdalena river basin, Colombia. 195 semi-structured questionnaires were applied to identify the perception of producers about the classification and uses of tree forage species, based on physical, nutritional, phenological and environmental criteria, as indicators of conservation and use of species in pastures. With these indicators, indices were constructed that identified important characteristics of the species mentioned by the ranchers. Data were analyzed with descriptive statistics, mean comparisons, and multivariate analysis techniques. It is recognized that producers appreciate species based on criteria of use and functions related to their productive activity. Six highly valued species Gliricidia sepium, Guazuma ulmifolia, Pithecellobium dulce, Albizia guachapele, Acacia farnesiana and Albizia saman coincide with other studies of local knowledge. The study provides relevant information on twelve species associated with livestock, which have not been reported in previous studies, so it suggests deepening and complementing with scientific knowledge to recognize and Correspondencia: Nelson Pérez Almario, Corporación Colombiana de Investigación Agropecuaria (Agrosavia), Km 10 vía Espinal - Ibagué, Espinal, Tolima, Colombia. Correo electrónico: neperez3@yahoo.com Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 322 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía assess the use of these potential species for livestock production, allowing interaction between knowledge in a concerted technological system. Keywords: Dry zones, forage, multipurpose trees, participatory research, silvopastoral systems. Introducción algún tipo de beneficio inmediato (Sirrine et al. 2010). El conocimiento de especies leñosas con múltiples funciones Uno de los principales problemas que enfrenta la es más significativo en áreas secas, dada la función de ganadería del trópico es la limitada producción de sombra y oferta de alimento que estas especies pueden forrajes en cuanto a cantidad y calidad de las gramíneas aportar a los animales en épocas de sequía prolongada en épocas de sequía. Es el resultado de la explotación (Serrano et al. 2014; Sierra et al. 2017). tradicional extensiva que ha llevado a bajas producciones Por tanto, el objetivo del estudio es documentar la y bajos ingresos de los productores, mostrando ser la percepción de los productores ganaderos sobre el uso y alternativa menos viable de producción ganadera (López la clasificación de las especies leñosas presentes en los et al. 2009). potreros de sus fincas en la zona de estudio, con base en Villanueva et al. (2009) resaltan la importancia de las características de los árboles, y relacionarlas con los conocer las especies forrajeras que puedan ser incluidas aportes de uso y funciones percibidos. en sistemas silvopastoriles para asegurar una producción adecuada durante épocas de sequía. Por tanto, diferentes Materiales y Métodos autores destacan la importancia de los árboles en los sistemas de producción ganadera, más aún si se parte Zona de estudio del conocimiento que los productores han acumulado a través del tiempo (Pezo 2009). Dicho conocimiento El estudio se realizó en la parte alta de la cuenca del no debe restringirse a la identificación de especies, sino río Magdalena y cubre siete municipios en el norte del también considerar las características, usos, limitantes y Departamento del Huila y once en el centro y sur del potencialidades de los árboles en sus fincas (Stokes 2001). Departamento del Tolima. Su área seca es influenciada Diversas características de los árboles y arbustos por una zona desértica (‘La Tatacoa’) y tiene una han sido reconocidas por los productores al valorar extensión aproximada de 1,200,000 ha, en el denominado tanto su potencial forrajero como su importancia Valle Cálido del Alto Magdalena (Corpoica-Cortolima como generadores de servicios ecosistémicos. Esto 2011) (Figura 1). Ecológicamente, la zona de estudio los clasifica como componentes multipropósito que corresponde al bosque seco tropical (bs-T) y bosque han hecho contribuciones importantes en los sistemas muy seco tropical (bms-T), con áreas subhúmedas y de producción ganadera, al tiempo que contribuyen al semiáridas, respectivamente (Holdridge 1978). El rango equilibrio del medio ambiente y aportan a la economía de los promedios mínimo y máximo de precipitación familiar (Harvey et al. 2008; Pezo 2009). anual es de 1,270 a 1,880 mm con distribución bimodal Pérez-Almario et al. (2017) han demostrado que en (abril‒mayo y octubre‒noviembre); es importante Colombia existe un gran número de especies leñosas con resaltar que en el año se presentan entre 240 y 265 días alto potencial para la alimentación bovina, con contenidos sin precipitación. La temperatura promedio oscila entre nutricionales más altos que las gramíneas, y que a su vez 26 y 30 °C y la humedad relativa entre 56 y 79% (Pérez- contribuyen con servicios ambientales. Lo anterior ha Almario et al. 2017). Parte de los suelos en la zona llevado a que los productores encuentren en los árboles son bastante erosionados con afloramientos rocosos y una opción como fuente forrajera, sobre todo en épocas fertilidad mediana a baja. Se han reportado Ultisoles, críticas, y que hayan desarrollado criterios que les permiten Alfisoles e Inceptisoles (Mantilla et al. 1998). seleccionar, adoptar y conservar las especies adaptadas a las condiciones de sus fincas (Mosquera 2010). Muestreo Varias investigaciones han identificado diferentes características de los árboles en las fincas (Sierra et al. Se seleccionó una muestra de 195 fincas con ganado 2017; Holguín et al. 2018). Sin embargo, los criterios bovino en los 18 municipios de la zona de estudio con más utilizados por los productores al momento de decidir un rango altitudinal entre 300 y 1,000 msnm (Cuadro sobre la conservación, es decir, la no tala, de los árboles en 1), aplicando criterios cualitativos y cuantitativos de las sus potreros son aquellos que de una u otra forma generan teorías de muestreo (Teddlie y Yu 2007). Las fincas se Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Conocimiento local sobre árboles en fincas ganaderas 323 eligieron a partir de un muestreo estratificado donde los En las 195 fincas se realizaron entrevistas a los estratos correspondieron a las áreas de las fincas para productores y se aplicó un cuestionario semiestructurado. La definir los conglomerados del estudio. La importancia indagación correspondió a la caracterización de las fincas y del método de muestreo consistió en lograr que el mayor a la identificación de especies leñosas, con las preferencias número de fincas elegidas proporcionara la mayor de uso de los ganaderos y la valoración de las especies. información posible para profundizar sobre la pregunta Estas se encuentran sobre todo como árboles o arbustos de investigación (Martínez-Salgado 2012). dispersos en potreros, producto de la regeneración natural (Figura 1B). Algunas especies también se encuentran como árboles en cercas vivas, sembrados por los ganaderos para tal fin o nacidos por regeneración natural. Cuadro 1. Distribución de las fincas encuestadas por municipio en el sur del Tolima y norte del Huila, Colombia. No.consecutivo Municipio Departamento No. de fincas 1 Coyaima Tolima 12 2 Coello Tolima 1 3 Espinal Tolima 9 4 Guamo Tolima 13 5 Ibagué Tolima 23 6 Natagaima Tolima 8 7 Ortega Tolima 20 8 Piedras Tolima 6 Figura 1. A) Zona de estudio en 18 municipios (amarillo) 9 Prado Tolima 10 de los departamentos del Huila y Tolima, entre latitudes N 10 Saldaña Tolima 4 2°37' y 4°33', y longitudes O 74°53' y 75°24'. B) Árboles en 11 San Luis Tolima 7 fincas ganaderas de la zona de estudio. El punto rojo indica la 12 Campoalegre Huila 8 ubicación del desierto La Tatacoa. 13 Hobo Huila 1014 Neiva Huila 11 15 Palermo Huila 24 Los criterios para seleccionar la muestra fueron 16 Rivera Huila 7 tres: fincas ganaderas con árboles; accesibilidad a las 17 Villavieja Huila 16 fincas; y saturación de la información. El primero fue un 18 Yaguará Huila 6 criterio ‘sine qua non’ dada la naturaleza del objetivo de Total de fincas 195 la investigación, mientras que el segundo dependió de la infraestructura vial y el costo de desplazamientos. El Características del cuestionario estructurado tercer criterio, el nivel de saturación de la información, es entendido como el punto en el cual se ha escuchado Con un cuestionario semiestructurado, diseñado por cierta diversidad de información, y con cada entrevista Ospina y Pérez-Almario (2013), se indagó sobre el u observación adicional no aparecen nuevos elementos tamaño de las fincas, distribución del uso de la tierra, las (Mayan 2009). Mientras siga apareciendo nueva características productivas de las fincas y los usos de los información, la búsqueda no para, por lo cual este criterio árboles (véase Anexo). es tanto punto de partida del diseño de muestreo como Para clasificar los usos de las especies se establecieron resultado: el criterio de suficiencia (saturación) en la tres categorías: 1) uso forrajero (ramoneo directo, cercas información solo puede determinarse en el proceso y no vivas, corte y acarreo); 2) confort animal (sombrío); y a priori, dejando con el investigador la responsabilidad 3) otros usos (madera y mejoramiento del suelo). Las de determinar cuándo el nivel de saturación es lo variables de cada categoría fueron calificadas como de uso suficientemente alto para declarar apropiado el alto (A), medio (M) y bajo (B), para cada especie. muestreo (Martínez-Salgado 2012). La información Para valorar las especies, se registraron las frecuencias obtenida con cierto nivel de repetibilidad o con nuevas (porcentajes) con que los ganaderos usaron los juicios especies o características que describen el nivel de uso de valor previamente identificados en las diferentes y conservación de estas en las fincas por parte de los categorías de criterios, para decidir si una especie debe ganaderos, indica el nivel de saturación. conservarse en la finca. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 324 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía Los cuatro grupos de criterios evaluados fueron: índices de los cuatro criterios para diferenciar las especies Criterios físicos (9): 1. Hojas suaves para el ganado; 2. No de mayor importancia se obtuvo un índice consolidado. tiene espinas; 3. Hojas pequeñas; 4. Tiene hojas en la punta La distribución de las especies en función de los criterios de la rama; 5. Hojas duras; 6. Hojas dispersas en las ramas; se representó mediante un gráfico Biplot; para ello se usó 7. Tienen espinas; 8. Hojas grandes; 9. Tiene raíz profunda. el paquete estadístico InfoStat (Di Rienzo et al. 2018). Criterios nutricionales (8): 1. Altamente nutritivo; 2. El ganado se engorda; 3. Aumenta la leche; 4. Es rico Resultados en calcio y fósforo; 5. Controla parásitos externos; 6. Muy digestible; 7. Controla enfermedades y/o parásitos Características productivas de las fincas internos; 8. Le gusta al ganado y no es amargo. La mayor proporción de las fincas analizadas son Criterios fenológicos (5): 1. Retiene parte de las hojas; pequeñas y medianas. El área difiere (P>0.05) entre los 2. No se caen las hojas; 3. Se caen todas las hojas; 4. Se tres grupos resultantes del análisis de conglomerados. caen los frutos; 5. No se caen todos los frutos. Mientras que el conglomerado C2 presenta mayor Criterios ambientales (5): 1. Tolera sequía; 2. Tolera número de fincas de tamaño mediano, el conglomerado encharcamiento y sequía; 3. Se encuentra en varias C1 (fincas más pequeñas) ocupa un segundo lugar en alturas (m.s.n.m.); 4. Se encuentra en varios tipos de cuanto a número de fincas, y el conglomerado C3, con suelo; 5. Produce alta sombra y confort. el menor número de fincas, muestra las propiedades Para cuantificar la información obtenida mediante el de mayor superficie. Las fincas grandes poseen mayor cuestionario se construyó una base de datos en Excel y número de potreros (pasturas) cuya área promedio (32 se procedió a estandarizar los datos usando el método de ha) es además muy superior al de los potreros en las estandarización (z-score) según el protocolo de Schuschny fincas medianas y pequeñas (Cuadro 2). y Soto (2009). Este consiste en ajustar la variable original, La distribución y los usos de la tierra en las fincas restándole a cada valor la media y dividiendo este entre difieren entre conglomerados. La proporción del área la desviación estándar. El resultado para cada cálculo dedicada a potreros es mayor en las fincas pequeñas y fue denominado Índice de Diferencia Ajustada (I), cuyos medianas (69 y 70%, respectivamente) comparadas con resultados se obtuvieron mediante la fórmula: la observada en fincas grandes (47%). Sin embargo, las fincas grandes tienen una importante proporción del área (xi − X x ) dedicada a conservar los bosques ribereños y áreas de I = ∑ i σ barbecho (27.7 y 25.5 ha, respectivamente). De otro lado, x , donde: i la conservación de los árboles difiere con el tamaño de I es el valor del índice; las fincas, pues las pequeñas conservan en promedio 98 x i es el valor de cada unidad de la variable del árboles, las medianas 515 y las grandes 2,223 (Cuadro 3). criterio; X xi es el valor promedio de la variable usado para el Descripción de usos de los árboles ajuste; σ xi es el valor de la desviación estándar de cada Se reportaron 31 especies leñosas que los ganaderos variable usado para el ajuste. relacionaron con diferentes usos en sus fincas. Los resultados sugieren que hay una alta proporción de especies Análisis estadístico con usos múltiples diferentes al forrajero. Se reportaron: 16 especies usadas para el consumo por ramoneo directo, Se aplicaron análisis descriptivos, comparaciones de las cuales 7 se consideran de uso alto, 2 de uso medio y de medias usando la diferencia mínima significativa 7 de uso bajo; 10 especies para corte y acarreo en bancos (LSD de Fisher) y análisis multivariados [componentes forrajeros (8 de uso alto y 2 de uso medio); 31 especies principales, conglomerados (método de Ward)], mediante para el confort de los animales (13 de uso alto, 5 de uso los cuales se agruparon las fincas según su área, uso de la medio y 13 de uso bajo); 8 especies para uso en cercas tierra, número y área de potreros, grupos de animales y vivas (6 de uso alto y 2 de uso medio); y 26 especies para número de estos. Las variables de las especies incluidas madera usada en la finca (10 de uso alto, 4 de uso medio en cada criterio de uso y conservación se describieron y 12 de uso bajo). Todas las especies fueron mencionadas con el índice (I) anteriormente mencionado. Sumando los con alto uso como mejoradores de suelos (Cuadro 4). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Conocimiento local sobre árboles en fincas ganaderas 325 Cuadro 2. Número y área de las fincas por conglomerado y número y área media de potreros en las fincas (medias ± error estándar). Conglomerado No. de fincas Área promedio (ha) No. de potreros1/finca Área promedio de potrero (ha) C1 76c 39.4c ± 8.29 5.0b ± 0.37 5.4b C2 99b 61.0b ± 7.11 9.5b ± 0.75 4.5b C3 20a 400.0a ± 10.90 17.6a ± 3.61 31.5a Medias en una misma columna seguidas por una letra común no difieren significativamente (P>0.05). 1Incluye potreros con y sin árboles. Cuadro 3. Distribución de áreas de uso de la tierra en las fincas por conglomerado (medias + error estándar). Uso de la tierra C1 C2 C3 Cultivo anual/transitorio (ha) 7.7b ± 3.4 5.8b ± 1.5 143.4a ± 23.4 Pasturas sin árboles (ha) 21.7b ± 7.1 27.8b ± 3.9 103.6a ± 15.9 Pasturas arboladas (ha) 5.4b ± 0.8 14.7b ± 2.4 82.0a ± 32.5 Pastos corte (ha) 0.2a ± 0.05 0.9b ± 0.2 1.0b ± 0.3 Cultivos permanentes (ha) 0.7b ± 0.3 1.8b ± 1.6 11.9a ± 6.5 Barbecho (ha) 1.3b ± 0.5 4.4b ± 0.9 25.5a ± 10.6 Bosque ribereño (ha) 2.1b ± 0.4 3.2b ± 0.5 27.7a ± 12.04 Bosque/parche bosque (ha) 0.5a ± 0.2 1.0b ± 0.3 1.5b ± 0.7 Sistema silvopastoril intensivo (ha) 0.1b ± 0.04 0.4a ± 0.4 0.6a ± 0.5 Huertos familiares (ha) 0.3b ± 0.06 1.04b ± 0.6 2.8a ± 2.25 Número de árboles por finca 98.5c ± 21.6 515.0b ± 27.6 2,223.4a ± 69.7 Medias en una misma fila seguidas por una letra común no difieren significativamente (P>0.05). Índices para el uso y la conservación de las especies Ciruelo, Dinde, Gomo, Palma real, Tachuelo, Tamarindo y Vainillo. Los criterios individuales representados en los índices o Criterios físicos. Las variables de mayor importancia valores de importancia mostraron seis especies con índice para los ganaderos fueron ‘hojas suaves para el ganado’ más alto, pero con diferente orden en los criterios físicos, (25.8%); ‘no tiene espinas’ (23.8%); ‘hojas pequeñas’ nutricionales, fenológicos y ambientales: Matarratón (13.2%); ‘hojas dispersas en las ramas’ (12.5%); y ‘hojas (G. sepium), Guácimo (G. ulmifolia), Payandé (P. duras’ (9.6%), las cuales acumularon el 84.9% de la dulce), Iguá (A. guachapele), Pelá (A. farnesiana) y información suministrada. Samán (A. saman) (Cuadro 5). Para estas especies, los índices (número de veces que fueron mencionadas) de Criterios nutricionales. Las variables nutricionales más los respectivos criterios oscilaron entre 11 y 19 (criterios importantes fueron ‘altamente nutritivo’ (21.7%); ‘el físicos), 12 y 23 (nutricionales); 8 y 16 (fenológicos); y ganado se engorda’ (19.5%); ‘aumenta la leche’ (19.4%); 9 y 17 (ambientales), respectivamente. Sin embargo, al ‘es rico en calcio y fósforo’ (18.9%); y ‘controla parásitos consolidar (sumar) los índices de cada criterio para las externos’ (15.5%), las cuales acumularon el 95% de la mismas especies se encontraron valores de importancia información. alta (75.3; 65.2; 52.5; 48.1; 48.0; y 42.8) para Matarratón Criterios fenológicos. Para el grupo fenológico, las (G. sepium), Guácimo (G. ulmifolia), Payandé (P. dulce), variables importantes fueron ‘retiene parte de las hojas’ Iguá (A. guachapele), Pelá (A. farnesiana) y Samán (33.4%); ‘no se caen las hojas’ (28.3%); y ‘se caen (A. saman), respectivamente (Cuadro 5). todas las hojas’ (16.7%), acumulando el 78.5% de la información. Aquí la percepción de los productores Características de uso y conservación de mayor sobre la importancia de la conservación de las hojas en importancia para los ganaderos los árboles es entendida como forraje disponible para el consumo animal en la época seca. El nivel de saturación para este estudio es alto, debido Criterios ambientales. Las variables de este grupo a que la información obtenida en los cuatro criterios que mostraron importancia alta para los ganaderos incluye un importante número de especies leñosas con fueron ‘muy tolerantes a sequía’ (25.7%); ‘tolerantes información ‘nueva’ (distintos usos) que aún no ha a encharcamientos y sequías’ (23.1%); ‘se encuentran sido reportada para el sector ganadero en zonas secas en diferentes alturas sobre el nivel del mar’ (18.9%); de Colombia; entre ellas: Acacia amarilla, Ambuco, y ‘en diferentes tipos de suelos’ (18.4%). Entre estas, Angarillo, Bayo, Cachingo, Carbonero, Chaparro, acumularon el 86% de la información registrada. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 326 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía Cuadro 4. Nombres, usos e intensidad de uso de las especies reportadas por los productores. Nombre común Nombre científico1 Uso forrajero Confort animal Otros usos No. productores que (local) Ramoneo Cerca viva2 Corte y acarreo Sombrío Madera Mejora suelo mencionan la especie Acacia amarilla Senna siamea (Lam.) H.S. Irwin & Barneby(Fabaceae) A B A 1 Ambuco Acacia canescens (Britton & Killip) García-Barr. B A A 10 (Fabaceae) Angarillo Chloroleucon mangense var. vincentis (Benth.) B B A A 50 Barneby & J.W. Grimes(Fabaceae) Bayo Albizia niopoides (Benth.) Burkart (Fabaceae) A A M A 10 Botón de oro Tithonia diversifolia (Hemsl.) A. Gray (Compositae) B A B A 3 Cachingo Erythrina fusca Lour. (Fabaceae) M A A M B A 15 Carbonero Calliandra riparia Pittier (Fabaceae) A M B A 38 Chaparro Curatella americana L.(Dilleniaceae) B B A 14 Ciruelo Spondias purpurea L. (Anacardiaceae) A A B A 16 Cují Prosopis juliflora (Sw.) DC.(Fabaceae) B A A 25 Dinde Maclura tinctoria (L.) D. Don ex Steud. (Moraceae) A A A 27 Gomo Cordia alba (Jacq.) Roem. & Schult. (Boraginaceae) A M A M B A 14 Guácimo Guazuma ulmifolia Lam. (Malvaceae) A A M B A 163 Gualanday Jacaranda caucana Pittier (Bignoniaceae) A A A 5 Guayaba Psidium guajava L. (Myrtaceae) M B A A 5 Iguá Albizia guachapele (Kunth) Dugand (Fabaceae) A A A A 128 Leucaena Leucaena leucocephala (Lam.) de Wit (Fabaceae) A A M B A 56 Matarratón Gliricidia sepium (Jacq.) Walp. (Fabaceae) B A A B B A 179 Moringa Moringa oleifera Lam. (Moringaceae) A A B A 1 Nacedero Trichanthera gigantea (Humb. & Bonpl.) Nees A M B A 15 (Acanthaceae) Neem Azadirachta indica (A. Juss.) (Meliaceae) A M A 22 Orejero Enterolobium cyclocarpum (Jacq.) Griseb. (Fabaceae) A B A 4 Palma real Attalea butyracea (Mutis ex L. f.) Wess. B A 10 Boer(Arecaceae) Patevaca Bauhinia variegata L.(Fabaceae) A A B A 7 Payandé Pithecellobium dulce (Roxb.) Benth. (Fabaceae) B A A M A 117 Pelá Acacia farnesiana (L.) Willd. (Fabaceae) B B A A 109 Samán Albizia saman (Jacq.) Merr.(Fabaceae) B A B A 94 Tachuelo Zanthoxylum rhoifolium Lam. (Rutaceae) B A A 1 Tamarindo Tamarindus indica L. (Fabaceae) A M A 16 Totumo Crescentia cujete L. (Bignoneaceae) B M A A A 20 Vainillo Senna spectabilis (DC.) H.S. Irwin & A B A 3 Barneby(Fabaceae) A = Uso alto; M = Uso medio; B = Uso bajo. 1Nomenclatura científica según The World Flora Online (worldfloraonline.org); 2Uso forrajero por ramoneo y corte y acarreo. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Conocimiento local sobre árboles en fincas ganaderas 327 Cuadro 5. Valor de los índices de las especies según el orden de importancia para los productores en cada criterio Especie Físico Especie Nutricional Especie Fenológico Especie Ambiental Guácimo 19.1 Matarratón 23.1 Matarratón 16.6 Matarratón 17.0 Payandé 19.1 Guácimo 21.5 Guácimo 9.9 Guácimo 14.7 Matarratón 18.6 Iguá 14.4 Pelá 8.7 Iguá 12.0 Pelá 15.3 Pelá 14.4 Samán 8.7 Payandé 11.0 Iguá 13.7 Payandé 13.9 Payandé 8.5 Samán 10.7 Samán 11.4 Samán 12.1 Iguá 8.0 Pelá 9.6 Angarillo 8.7 Ciruelo 8.6 Leucaena 4.9 Carbonero 5.0 Cují 8.2 Totumo 5.1 Angarillo 4.0 Ciruelo 4.4 Carbonero 7.1 Carbonero 4.8 Ciruelo 3.5 Leucaena 2.9 Ciruelo 6.6 Angarillo 4.2 Carbonero 3.4 Angarillo 2.4 Cachingo 6.0 Neem 4.0 Cují 2.6 Totumo 2.4 Totumo 6.0 Chaparro 3.3 Neem 2.3 Nacedero 2.0 Leucaena 5.0 Leucaena 3.2 Dinde 1.8 Cachingo 2.0 Dinde 4.4 Cují 2.6 Gomo 1.7 Dinde 1.8 Neem 3.7 Cachingo 2.3 Totumo 1.3 Cují 1.8 Nacedero 3.5 Gomo 2.3 Tamarindo 1.3 Chaparro 1.3 Chaparro 3.3 Palma real 2.2 Ambuco 1.1 Bayo 1.3 Ambuco 3.1 Dinde 2.1 Cachingo 1.1 Palma real 1.2 Gomo 2.9 Nacedero 1.3 Nacedero 1.0 Neem 0.9 Palma Real 2.1 Bayo 1.3 Chaparro 0.8 Gomo 0.6 Bayo 1.7 Ambuco 1.3 Palma real 0.8 Tamarindo 0.6 Tamarindo 1.2 Orejero 1.0 Bayo 0.6 Patevaca 0.5 Patevaca 1.1 Botón de oro 0.8 Patevaca 0.4 Botón de oro 0.5 Gualanday 0.9 Guayaba 0.7 Gualanday 0.3 Orejero 0.4 Botón de oro 0.6 Gualanday 0.6 Moringa 0.3 Gualanday 0.3 Orejero 0.6 Patevaca 0.6 Botón de Oro 0.2 Ambuco 0.3 Acacia amarilla 0.1 Tamarindo 0.5 Orejero 0.2 Acacia amarilla 0.1 Tachuelo 0.1 Moringa 0.3 Vainillo 0.1 Vainillo 0.09 Guayabo 0.1 Acacia amarilla 0.3 Acacia amarilla 0.05 Moringa 0.05 Vainillo 0.04 Vainillo 0.07 Tachuelo 0.05 Guayaba 0.04 Moringa 0.02 Tachuelo 0.03 Guayaba 0.02 Tachuelo 0.01 Grupos de especies con similitud entre criterios sí y se asocian con los criterios nutricional y ambiental al encontrarse cerca al eje que los divide. Sin embargo, En la Figura 2 se presenta una gráfica biplot como una especies como Angarillo (C. mangense var. vincentis), herramienta estadística del análisis de componentes Carbonero (C. riparia) y Ciruelo (S. purpurea) están en principales, para ilustrar la forma cómo se relacionan y los mismos cuadrantes, lo cual sugiere similitud entre agrupan características multivariables. Se observa que ellas, pero no se asocian claramente con los criterios de los cuatro grupos, los criterios físicos constituyen físico, nutricional y ambiental. el vector de la mayor distancia desde el origen y por En los Cuadrantes II y III se encuentran las especies tanto representa la variable que más contribuye a la Orejero (E. cyclocarpum), Dinde (M. tinctoria), Totumo variabilidad de las observaciones. En los Cuadrantes (C. cujete), Vainillo (S. spectabilis), Neem (A. indica), I y IV se encuentran las especies que más reconocen Gomo (C. alba), Nacedero (T. gigantea), Bayo (A. los ganaderos (las mencionadas con más frecuencia) niopoides), Moringa (M. oleifera), Ambuco (A. canescens), mientras que los Cuadrantes II y III incluyen las especies Cují (P. juliflora), Palma real (A. butyracea), Cachingo (E. mencionadas con menor frecuencia y valoración. fusca), Tachuelo (Z. rhoifolium), Patevaca (B. variegata), Las especies observadas en el Cuadrante I incluyen a Tamarindo (T. indica), Chaparro (C. americana), Pelá (A. farnesiana), Guácimo (G. ulmifolia) y Payandé Leucaena (L. leucocephala) y Gualanday (J. caucana). (P. dulce) mientras que el Cuadrante IV muestra a Iguá Estas especies muestran similitud entre sí al encontrarse (A. guachapele), Samán (A. saman) y Matarratón (G. ligadas al eje que divide sus cuadrantes, pero no se asocian sepium). Estas seis especies presentan similitud entre a ningún criterio de selección. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 328 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía Figura 2. Biplot de criterios y especies arbóreas según menciones de los ganaderos de las fincas en la zona de estudio. Discusión pequeños manifiestan que sus opciones para establecer actividades agrícolas son limitadas, dada la evidente Caracterización de las fincas escasez de agua en la zona del estudio. Olaya et al. (2000) reportan un déficit hídrico durante gran parte del año, La muestra de fincas analizada constituye un reflejo de lo cual es limitante para la agricultura si no se cuenta la estructura agraria de la región Alto Magdalena, en la con riego, siendo la ganadería extensiva la actividad cual es evidente una concentración de la propiedad en más rentable, por su baja inversión. Por tanto, la mayor pocos latifundios y un fraccionamiento de las fincas de parte de una finca pequeña es destinada a pasturas y las áreas medianas y pequeñas que representan la mayor actividades agrícolas se limitan a pequeñas áreas para el proporción de la muestra (hasta 90%). Esto coincide autoconsumo. Hallazgos similares fueron reportados por con otros estudios (Mora-Delgado et al. 2014; Medina- Rodríguez (2020). Ríos 2019) que describen una alta concentración de las Una característica interesante en esta tipología fincas ganaderas en el Tolima al tiempo que observan un de predios es que, durante la apertura de la frontera fraccionamiento de propiedades pequeñas, caracterizadas agrícola para el establecimiento de pasturas, no arrasaron por problemas de sobrepastoreo. Este estudio evidencia totalmente con los árboles; aún se conservan especies la presencia de amplias áreas de conservación para usos de árboles dispersos en potreros, cercas vivas y bosques de bosque en las fincas grandes (13.7%) en comparación ribereños. Varios autores, que han estudiado las formas de con 10.4 y 6.8% en las fincas pequeñas y medianas, conservación de especies leñosas en fincas ganaderas de la respectivamente; esto es concordante con lo reportado región, dan cuenta de la retención de los árboles (Serrano por Medina-Ríos et al. (2016). Las fincas grandes tienen et al. 2014; Sierra et al. 2017; Herrera 2020), una evidencia una mayor posibilidad de diversificar la producción del potencial de estos sistemas para la conservación de la agropecuaria, mientras que los propietarios de predios diversidad florística (Harvey et al. 2008). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Conocimiento local sobre árboles en fincas ganaderas 329 Descripción de usos de los árboles según la percepción la fertilidad. Esto es atribuido a la caída de las hojas de los ganaderos de los árboles por ser deciduos, los que contribuyen a incrementar la materia orgánica y, en el caso de Gran parte de los criterios utilizados por los ganaderos leguminosas, además aportes importantes de nitrógeno. para identificar las especies arbóreas se relacionan con La percepción de los productores respecto a la la identificación de las especies más conocidas por su tolerancia de los árboles a factores de clima, y la calidad nutritiva; de hecho, a esta relación se llega por plasticidad de estos para permanecer en diferentes tipos indicadores empíricos como el aumento en la cantidad de suelo, también ha sido reportada por Muñoz (2004) de leche y o ganancia de peso observada. Por otra parte, y Mora-Delgado et al. (2014) quienes asociaron la las percepciones sugieren una apreciación de las especies presencia de árboles con indicadores de “suelos buenos” leñosas por su tolerancia a variables del clima como o “suelos malos”. Conceptos como el que el follaje de las sequías prolongadas. Estas percepciones ratifican un árbol “es rico en calcio y fosforo”, mencionados por la capacidad de los productores para usar modelos los ganaderos, son valoraciones que hace el productor causales para la identificación y conservación de árboles sin base experimental, pero son usados para tomar importantes frente a la estacionalidad forrajera, el valor decisiones. Al respecto hay que reconocer que estudios de alimenticio y la tolerancia a sequías. Similitudes con estos etnobotánica han validado algunas plantas como fuentes criterios fueron reportados por Muñoz (2004) y Esquivel ricas en calcio, fósforo y otros nutrientes (Earle 2001), lo et al. (2011), quienes advierten que tales relaciones cual sugiere la necesidad de ampliar la investigación en empíricas no constituyen evidencia científica, sino estas relaciones para darles soporte científico. expresiones de la experiencia basada en la observación. Los resultados sugieren que el aprecio hacia los árboles Sin embargo, sus juicios van más allá de la simple en los predios ganaderos no se da de forma aleatoria, sino percepción o la sola identificación de especies (Mora- que depende de características funcionales que facilitan Delgado et al. 2014; Vásquez et al. 2014) en la medida la provisión de bienes y servicios para el productor, que que los criterios y relaciones son establecidos con base pueden resultar en ingresos adicionales, convirtiéndose en la comprobación empírica repetida en él tiempo. Los en un factor determinante para el mantenimiento de criterios están relacionados con el aprovechamiento de algunas especies en las fincas. recursos forrajeros no convencionales, pero también son usados en arreglos espaciales no planeados ofreciendo Características de las especies según los criterios productos como madera y servicios ecosistémicos como definidos el confort animal y mejoramiento del suelo (Pérez- Almario et al. 2017). Entre las características importantes de las especies Varias especies registradas en este estudio son sobresalen la calidad nutritiva del forraje asociada al reportadas también en trabajos fuera de Colombia, incremento de la producción ganadera; la textura y el respecto a sus características de consumo y calidad tamaño de las hojas; y el control de parásitos externos. nutritiva. Por ejemplo, en México (Pinto-Ruiz et al. Otras características determinantes en la decisión de los 2010; Olivares-Pérez et al. 2016) y Costa Rica (Esquivel ganaderos sobre la conservación o no de las especies en los et al. 2011), destacaron a G. ulmifolia, C. alba (sin. potreros son la percepción de la permanencia de las hojas C. dentata), P. dulce, Erythrina sp. y G. sepium. Para en los árboles y la tolerancia a diferentes tipos de estrés. Costa Rica, Stokes (2001) y Muñoz (2004) reportaron El proyecto FRAGMENT (p.ej., Restrepo-Sáenz et que G. sepium, G. ulmifolia, Erythrina sp. A. saman, al. 2004) coincide con nuestros resultados al reportar L. leucocephala, E. cyclocarpum y Crescentia sp. que G. ulmifolia, G. sepium, y A. saman son las especies fueron utilizados por el 87% de los ganaderos para forrajeras mejor calificadas, según la percepción de los suplementación animal durante las épocas de sequía. ganaderos en estudios desarrollados en Centroamérica. En Una de las características mencionadas por todos los dicho estudio además incluyeron L. leucocephala, pero no ganaderos respecto a las especies estudiadas corresponde obtuvo una buena calificación en este estudio. Lo anterior a los aportes de estas al mejoramiento del suelo, entre es debido a altos consumos y aportes significativos a la las que se destacan las raíces profundas que ayudan a producción, calidad nutricional y palatabilidad del forraje “descompactar” el suelo y los nutrientes que mejoran (Pérez-Almario 2011; Joya et al. 2004). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 330 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía Conclusiones Universidad del Tolima, Ibagué, Colombia. repository. ut.edu.co/handle/001/3276 Se confirma el conocimiento sobre el valor multiuso Holguín VA; Garcia II; Mora-Delgado J. 2018. Árboles y de árboles en fincas ganaderas, especialmente en el arbustos para silvopasturas: uso, calidad y alometría. arreglo de árboles dispersos en potreros. Esto ratifica la Universidad del Tolima, Ibagué, Colombia. importancia de indagar a los ganaderos sobre sus saberes Joya M; López M; Gómez R; Harvey CA. 2004. Conocimiento local sobre el uso y manejo de los árboles en las fincas y prácticas como una premisa esencial para que el ganaderas del municipio de Belén, Rivas. Encuentro ejercicio de la investigación de leñosas sea relevante para 68:44–59. doi: 10.5377/encuentro.v0i68.4256 sus necesidades, y por ende para sus sistemas ganaderos. López M; Solís G; Murrieta J; López R. 2009. Percepción de Este estudio sugiere la importancia de diferentes los ganaderos respecto a la sequía: viabilidad de un manejo funciones y roles de los árboles que ocupan diferentes de los agostaderos que prevenga sus efectos negativos. estratos en fincas ganaderas. Se requiere continuar Estudios sociales (Hermosillo, Son.) 17:221–241.bit.ly/ procesos experimentales agronómicos y zootécnicos que 3ATyTwh reconozcan el potencial productivo y el valor ecológico Mantilla G; De la Torre LE; Gómez CE; Ordóñez N; Ceballos de estas especies. JL; Euscátegui C; Pérez P; Pérez S; Martínez N; Sánchez R; Maldonado N; Gaitán J; Chávez L; Chamorro C; Flórez A. 1998. Los suelos: estabilidad, producción y degradación. Agradecimientos En: Leyva P, ed. El medio ambiente en Colombia. IDEAM, Bogotá, Colombia. hdl.handle.net/20.500.12324/18777 A la Corporación Colombiana de Investigación Martínez-Salgado C. 2012. El muestreo en investigación Agropecuaria – Agrosavia; Ministerio de Agricultura y cualitativa. Principios básicos y algunas controversias. Desarrollo Rural – MADR; Universidad del Tolima. Ciência & Saúde Coletiva 17(3):613–619. doi: 10.1590/ S1413-81232012000300006 Referencias Mayan M. 2009. Essentials of qualitative inquiry. Walnut Creek: Left Coast Press, New York, USA. doi: 10.4324/ (Nota de los editores: Los enlaces se verificaron el 19 de agosto de 2021). 9781315429250 Medina-Ríos EL. 2019. Conocimiento local de ganaderos Corpoica-Cortolima. 2011. Caracterización socioambiental de sobre el componente leñoso para la implementación de las zonas secas del departamento del Tolima. En informe innovaciones tecnológicas en la zona norte del Tolima. Tesis técnico (Ed. Autor), (pp. 160). de maestría. Universidad del Tolima, Ibagué, Colombia. Di Rienzo J; Casanoves F; Balzarini M; Gonzalez L; Tablada repository.ut.edu.co/handle/001/3277 M; Robledo C. 2018. Software estadístico Infostat. Medina-Ríos EL; Mora-Delgado J; Rodríguez-Márquez M. Argentina. www.infostat.com.ar 2016. Indicadores bioeconómicos de fincas ganaderas del Earle, JS. 2001. Plantas medicinales en el trópico húmedo. trópico bajo del norte del departamento del Tolima. Revista Editorial Guayacán, Guápiles, Costa Rica. Tumbaga 1(11):81–96. bit.ly/3h3BPz9 Esquivel H; Ibrahim M; Harvey AC; Benjamin T; Sinclair Mora-Delgado J; Castañeda R; Piñeros R. 2014. Paisajes, FL. 2011 Dispersed trees in pasturelands of cattle farms pasturas y pastos. Métodos para determinar cantidad in a tropical dry ecosystem. Tropical and Subtropical y calidad. Universidad del Tolima. Ibagué, Colombia. Agroecosystems 14: 933-941. bit.ly/3j0fyn9 repository.ut.edu.co/handle/001/3245 Harvey CA; Villanueva C; Ibrahim M; Gómez R; López Mosquera DH. 2010. Conocimiento local sobre bienes y M; Kunth S; Sinclair F. 2008. Productores, árboles y servicios de especies arbóreas y arbustivas en sistemas producción ganadera en paisajes de América Central: de producción ganadera de Rivas, Nicaragua. Tesis de implicaciones para la conservación de la biodiversidad. En: maestría. CATIE, Turrialba, Costa Rica. hdl.handle. Harvey CA; Saenz J, eds. Evaluación y conservación de net/11554/907 biodiversidad en paisajes fragmentados de Mesoamérica. Muñoz D. 2004. Conocimiento local de la cobertura arbórea Instituto Nacional de Biodiversidad (INBIO), Santo en sistemas de producción ganadera en dos localidades de Domingo de Heredia, Costa Rica. p. 197–224. Costa Rica. Tesis de maestría. CATIE, Turrialba, Costa Holdridge L. 1978. Ecología, basada en zonas de vida. Rica. hdl.handle.net/11554/2894 Instituto Interamericano de Cooperación para la Olaya A; Sánchez M; Tovar A. 2000. Directrices para Agricultura (IICA), San José, Costa Rica. repositorio.iica. zonificación, uso y manejo del desierto la Tatacoa. int/handle/11324/7936 Entornos 1(12):27–46. bit.ly/3n3lo9Q Herrera C. 2020. Caracterización y tipificación de los Olivares-Pérez P; Rojas-Hernández S; Avilés-Nova F; sistemas productivos de ovinos de pelo en el municipio Camacho-Díaz LM; Cipriano-Salazar M; Jiménez-Guillén de Villavieja, departamento del Huila. Tesis de maestría. R; Quiroz-Cardozo F. 2016. Use of non-leguminous trees Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Conocimiento local sobre árboles en fincas ganaderas 331 in silvopastoral systems in the south of the State of Mexico. del Tolima, Ibagué, Colombia. repository.ut.edu.co/handle/ Ecosistemas y Recursos Agropecuarios 3(8):193–202. bit.ly/ 001/3278 3j1rF3u Serrano JR; Andrade HJ; Mora-Delgado J. 2014. Ospina SD; Pérez-Almario N. 2013. Selección de diez Caracterización de la cobertura arbórea en una pastura especies arbóreas y/o arbustivas para sistemas ganaderos del trópico seco en Tolima, Colombia. Agronomía del Tolima, Huila, Cesar y Meta. Informe técnico. Mesoamericana 25(1):99–110. bit.ly/3kcNsob Corporación colombiana de investigación agropecuaria Sierra E; Andrade HJ; Segura MA. 2017. Percepción local (Agrosavia), Bogotá, Colombia. del componente arbóreo en fincas agropecuarias de la Pérez-Almario N. 2011. Rasgos funcionales nutricionales zona seca del norte del Tolima, Colombia. Revista de de especies leñosas en sistemas silvopastoriles y su Investigación Agraria y Ambiental 8(2):17–28. doi: contribución a la sostenibilidad de la ganadería bovina 10.22490/21456453.2027 en la época seca en el departamento de Rivas, Nicaragua. Sirrine D; Shennan C; Sirrine J. 2010. Comparing Tesis de maestría. CATIE, Turrialba, Costa Rica. hdl. agroforestry systems ex ante adoption potential and ex handle.net/11554/11027 post adoption: on-farmer participatory research from Pérez-Almario N; Ospina S; Mora J; Criollo D; Medina E. 2017. southern Malawi. Agroforestry Systems 79:253–266. Atributos funcionales para seleccionar especies de árboles doi: 10.1007/s10457-010-9304-0 y diseñar sistemas silvopastoriles o agrosilvopastoriles Schuschny A; Soto H. 2009. Guía metodológica: Diseño de en zonas secas. IX congreso internacional de sistemas indicadores compuestos de desarrollo sostenible. CEPAL; silvopastoriles. Manizales, Colombia, Septiembre 6–8 GTZ, Santiago, Chile. bivica.org/file/view/id/2485 2017. Stokes L. 2001. Farmer´s knowledge about the management Pezo D. 2009. Los pastizales seminaturales de América Central: and use of trees on livestock farms in the Cañas area of un recurso forrajero poco estudiado. Agroforestería en las Costa Rica. Bangor University, Bangor, Wales. Américas 47:4–5. hdl.handle.net/11554/6513 Teddlie Ch; Yu F. 2007. Mixed methods sampling: A typology Pinto-Ruiz R; Hernández D; Gómez H; Cobos MA; Quiroga R; with examples. Journal of Mixed Methods Research Pezo D. 2010. Árboles forrajeros de tres regiones ganaderas 1(1):77–100. doi: 10.1177/1558689806292430 de Chiapas, México: usos y características nutricionales. Vásquez F; Mora-Delgado J; Aguilar Støen M. 2014. Saber Universidad y Ciencia 26(1):19–31. bit.ly/3iZLXdq popular de especies forrajeras en la zona central de Restrepo-Sáenz C; Muhammad I; Harvey CA; Harmand JM; Nicaragua: un estudio en grupos focales. Revista de Ciencias Morales JL. 2004. Relaciones entre la cobertura arbórea en Agrícolas 31(1):27–41. doi: 10.22267/rcia.143101.43 potreros y la producción bovina en fincas ganaderas en el Villanueva C; Ibrahim M; Casasola F; Ríos N; Sepúlveda C. trópico seco en Cañas, Costa Rica. Agroforestería en las 2009. Sistemas silvopastoriles: una herramienta para la Américas 41–42: 29–36. http://hdl.handle.net/11554/10053 adaptación al cambio climático de las fincas ganaderas Rodríguez C. 2020. Transferencia de tecnología en comunidades en América Central. Informe técnico CATIE No. 377. ganaderas del sur del Tolima: una sistematización CATIE, Turrialba, Costa Rica. p. 103–125. hdl.handle. de experiencias. Tesis de maestría. Universidad net/11554/10149 (Recibido para publicación 27 agosto 2020; aceptado 5 agosto 2021; publicado 30 septiembre 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 332 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía Anexo. Criterios de selección de especies forrajeras con potencial para sistemas silvopastoriles Guía de Encuesta Fecha Municipio No. Encuesta Nombre del encuestador Nombre del productor(a) Coordenadas geográficas Departamento Revisado por: I. Datos generales de la unidad de produccion A. Residencia del productor ¿Ud. vive en la finca? 1. De forma permanente 3. Ocasionalmente 2. Por temporada 4. Otro B. Información productiva del hato Número de bovinos en el hato Vacas en producción: ____ Vaca horra/escotera: ____ Novilla levante: ____ Novilla vientre: ____ Novillos: ____ Crías hembras: ____ Crías machos: ____ Toros: ____ Sistema de monta: Natural ____ Artificial ____ Cuál ____ Producción de leche diaria: ____ Por vaca día: ____ Vende la leche: Si ____ No ____ Litros de leche vendidos ____ Autoconsumo ____ Procesa la leche: Si ____ No ____ En cuajada ____ Queso ____ Otros ____ Litros de leche transformada ____ % de la producción que comercializa ____ Autoconsumo ____ Costo de transporte para la comercialización: ____ Donde comercializa el producto: en la finca ____ pasteurizadora ____ otro ____ Cuál ____ Precio por kilo o litro ____ Fuente de agua: Las fuentes de agua de la finca son propias: Si ____ No ____ El agua de uso en la finca procede de quebradas ____ Ríos ____ Nacimientos ____ # de nacimientos ____ II. Sistema de produccion y limitaciones tecnológicas A. Uso de la tierra en la finca Registre la tierra ocupada físicamente durante el año pasado en cada finca de su propiedad Conceptos Área total (ha) Conceptos Área total (ha) Cultivos anuales/transitorios (ha) Rastrojos/barbecho (ha) Pasturas sin árboles (ha) Bosque ribereño/galería (ha) Pasturas con árboles (ha) Sistemas silvopastoriles intensivos (ha) Pastos de corte (ha) Huertos familiares (ha) Plantaciones permanentes (ha) Número divisiones potreros Total área de la finca (ha) ¿Qué labores o actividades realiza en el año para mantener a sus animales? Labores Frecuencia de la labor (mes) Insumos unidad (mes) Cantidad (unidad) 1. alimentación 2. vitaminación 3. vacunación 4. desparasitación. Externa 5. desparasitación. Interna 6. otro (especifique) Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Conocimiento local sobre árboles en fincas ganaderas 333 B. Árboles en potreros ¿Como tiene distribuidos los potreros en la finca? Especies leñosas en el sistema de pastoreo Nombre del Densidad especies de árboles ¿Cuántos árboles/ha? ¿Qué tipo de sombra ¿Árboles de regeneración pasto y/o arbustos predominantes (Clave 2) tiene? (Clave 3) natural o establecidos? (Clave 1) (Clave 4) 1 Pasturas naturales sin árboles 2 Pastos naturales con alta densidad de árboles 3 Pastos naturales con mediana densidad de árboles 4 Pastos naturales con baja densidad de árboles 5 Pastos mejorados con alta densidad de árboles 6 Pastos mejorados con mediana densidad de árboles 7 Pastos mejorados con baja densidad de árboles Total Clave 1 Clave 2 Clave 3 Clave 4 Pasto natural 1. Alta densidad 1. mayor a 100 árboles/ha 1. Copa no definida 1. Naturales 2. Mediana densidad 2. 50 a 99 árboles/ha 2. Copa ancha 2. Establecidos 3. Baja densidad 3. 1 a 49 árboles/ha 3. Copa cerrada 3. Ninguno Pasto mejorado 4. Alta densidad 4. mayor a 100 árboles/ha 4. Copa cónica 5. Mediana densidad 5. 50 a 99 árboles/ha 5. Copa abierta 6. Baja densidad 6. 1 a 49 árboles/ha 6. Ninguna Especies leñosas forrajeras encontradas en la finca Nombre especies Densidad/ha Orden de preferencia Criterios físicos Criterios nutricionales Criterios fenológicos Criterios ambientales y otros 1. 2. 3. ~ 20. Clave de densidad Clave 1 Clave 2 Clave 3 Clave 4 1. Alta 1. Hojas suaves para el ganado 1. Altamente nutritivo 1. Retiene parte de las hojas 1. Tolerantes a sequía 2. Media 2. No tiene espinas 2. El ganado se engorda 2. No se caen las hojas 2. Tolera encharcamiento y sequía 3. Baja 3. Hojas pequeñas 3. Aumenta la leche 3. Se caen todas las hojas 3. Se encuentra en varias alturas (m.s.n.m.) 4. Tiene hojas dispuestas en la punta de la rama 4. Es rico en calcio y fósforo 4. Se caen los frutos 4. Se encuentra en varios tipos de suelo 5. Hojas duras 5. Controla parásitos externos 5. No se caen todos los frutos 5. Produce alta sombra y confort 6. Hojas dispersas en la rama 6. Muy digestible 7. Tienen espinas 7. Controla enfermedades y/o parásitos internos 8. Hojas grandes 8. Le gusta al ganado y no es amargo 9. Tiene raíz profunda Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 334 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía Uso de las especies forrajeras Nombre especies Uso forrajero Confort animal Otros usos Importancia Ramoneo Cerca viva Corte-acarreo Árbol sombrío Madera Mejora suelos Descompacta suelo Aporta M.O Reduce temperatura dosel (Alto, medio, bajo) 1. 2. 3. ~ 20. C. Sistema de manejo de los forrajes Clave Aplicaciones # de veces o pases Tipo insumos Cosecha Preparación del suelo Clave 1 Siembra semilla Clave 2 Fertilización Clave 3 Control malezas Clave 4 Poda Clave 5 Cosecha de forraje Clave 6 Riego Clave 7 Clave 1 Clave 2 Clave 3 Clave 4 Clave 5 Clave 6 Clave 7 1. Roza y quema 1. Al voleo 1. química 1. química 1. cada 6 meses 1. pastoreo directo 1. gravedad 2. Control químico 2. Manual en surcos 2. orgánica 2. biológica 2. cada año 2. rotacional 2. aspersión 3. Mecanización 3. Siembra directa con estolones 3. orgánica y química 3. manual 3. otro especifique 3. alterno 3. gravedad y aspersión 4. Otro (especifique) 4. Mecanizada 4. ninguna 4. mecánica 4. nunca 4. corte y acarreo 4. ninguno 5. Renovación 5. ninguna 5. ninguno 5. otro (especifique) 6. Otro (cuál) Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Conocimiento local sobre árboles en fincas ganaderas 335 D. Principales limitaciones tecnicas de la ganaderia (rumiantes) ¿Sobre cada uno de los parámetros siguientes, está Ud. satisfecho, o piensa que algo le impide lograr mejores resultados? Parámetros ¿Satisfecho? 1. Si __ 2. No __ ¿De dónde proviene el problema? Clave 1 ¿Cómo podría mejorar? Clave 2 ¿Por qué no lo hace? Clave 3 Área forrajera Carga animal Incremento de peso Edad del primer parto Intervalo entre partos Duración de la lactación Producción diaria de leche de una vaca en invierno Producción diaria de leche de una vaca en verano Mortalidad Costos de producción Mejoramiento genético de su hato Otro: Cuál__________________ Clave 1: Clave 2: Clave 3: 1. Manejo alimentación 1. Mejorando la sanidad 1. No sabe hacerlo 2. Manejo reproducción 2. Cambiando el tamaño de potreros 2. No sería rentable 3. Condiciones de manejo del ganado 3. Cambiando el tipo de pasto 3. Falta de recursos financieros 4. Precios de los productos 4. Utilizando pastos de corte 4. Otro (especifique) 5. Manejo de la sanidad 5. Mejorando la raza 6. Genética 6. Mejorando las condiciones de vida del ganado 7. Otro (especifique) 7. Otro (especifique) E. Manejo tecnico en otras actividades pecuarias Mencione: ____________________________ no. Animales ____ F. Principales limitaciones tecnicas de las actividades pecuarias (rumiantes) ¿Sobre cada uno de los parámetros siguientes, está Ud. satisfecho, o piensa que algo le impide tener mejores resultados? Parámetros ¿Satisfecho? 1. Si _2. No_ ¿De dónde proviene el problema? Clave 1 ¿Cómo podría mejorar? Clave 2 ¿Por qué no lo hace? Clave 3 Incremento de peso Edad del primer parto Intervalo entre partos Producción semanal de huevos Mortalidad Costos de producción Mejoramiento genético de sus animales Otro: Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 336 N. Pérez-Almario, E.L. Medina-Rios, J. Mora-Delgado, D. Criollo-Cruz y J.R. Mejía Clave 1 Clave 2 Clave 3 1. Manejo alimentación 1. Mejorando la sanidad 1. No sabe hacerlo 2. Manejo reproducción 2. Mejorando la alimentación 2. No sería rentable 3. Condiciones de vida del ganado 3. Mejorando el potencial genético 3. Falta de recursos financieros 4. Precios de los productos 4. Mejorando las condiciones de vida 4. Otro (especifique) 5. Manejo de la sanidad 5. Otro (especifique) 6. Genética 7. Otro (especifique) G. Cambios y limitaciones en cuanto a comercializacion y almacenamiento de los productos pecuarios COMERCIALIZACION A. ¿Ha habido cambios en las condiciones de comercialización de sus productos pecuarios en el último año? B. ¿Hay problemas en cuanto a ello? 1. Si (llenar cuadro) 2. No. (Pasar a columna 8 del cuadro) A B Nombre del ¿Cuáles han sido Fuente del ¿Qué resultado Cuantifique este ¿Está satisfecho ¿Está satisfecho de ¿Cuál es el ¿Qué habría que ¿Por qué no se producto los cambios? cambio ha tenido este resultado% de este cambio? las condiciones de problema? (si hacer para resolver hace? Clave 6 Clave 1 Clave 2 cambio? Clave 3 Clave 3 (1 a 5) 1. si; 2. no comercialización? existe) Clave 4 este problema? 1. si; 2. no Clave 5 Leche cruda Leche transformada Carne Crías Clave 1 Clave 2: Clave 3: Clave 4: Clave 5 Clave 6: 1. El agente comercializador 1. modificación de las condiciones 1. Ha mejorado el precio 1. Precio muy bajo (en general) 1. Mejorando la calidad 1. No sabe hacerlo socioeconómica. 2. El lugar de venta del producto 2. innovación propia 2. Ha disminuido el tiempo de venta 2. Se vende a una época en que el 2. Organizarse varios productores 2. No sería rentable precio es bajo 3. La forma de llevar el producto 3. por imitación directa 3. Disminuido perdidas de producto 3. El costo de acceso al mercado es 3. Que haya más comerciantes 3. Falta de recursos financieros muy alto 4. La presentación del producto 4.inducida por Consejo técnico, sin 4. Ha aumentado las ventas 4. Variabilidad de precio de un año 4. Que se mejoren las vías 4. Otro (especifique) que haya una demanda al otro 5. La calidad del producto 5.Cambio técnico p/respuesta a una 5. Otro (especifique) 5. Otro (especifique) 5. Disminuyendo costos transporte demanda 6. Otro producto transformado 6.demanda canalizada por una OP 6. No sabe 7. El momento de venta 7. inducido por razones no técnicas 7. Otro (especifique) (especifique) 8. Otro (especifique) 8= otro (especifique) Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):337–347 337 doi: 10.17138/TGFT(9)337-347 Research Paper Physiological responses of Bajra-Napier hybrids and a tri-specific hybrid to salinity stress Respuestas fisiológicas de los híbridos Bajra-Napier y de un híbrido tri- específico al estrés por salinidad SEVA NAYAK DHEERAVATHU1, KAJAL SINGH2, PRAMOD W. RAMTEKE2, REETU1, NILAMANI DIKSHIT1, MAHENDRA PRASAD1, DIBYENDU DEB1 AND THULASI BAI VADITHE3 1ICAR-Indian Grassland and Fodder Research Institute, Jhansi, Uttar Pradesh, India. igfri.res.in 2Department of Biological Sciences, Sam Higginbottom University of Agriculture, Technology and Sciences, Allahabad, Uttar Pradesh, India. shiats.edu.in 3Department of Microbiology, Acharya Nagarjuna University, Guntur, Andhra Pradesh, India. nagarjunauniversity.ac.in Abstract Physiological responses of 3 Bajra-Napier (Cenchrus spp., syn. Pennisetum spp.) hybrid varieties, viz. BNH-3, BNH-6, BNH-10, and 1 tri-specific hybrid (TSH) were tested under different gradients of soil salinity, i.e. Control, 4, 6 and 8 dS/m electric conductivity (ECe), in a pot trial. The experiment was laid out in a factorial completely randomized design with 3 replications. Shoot dry weight, root dry weight, root:shoot ratio and chlorophyll a, chlorophyll b, total chlorophyll and carotenoid concentrations were reduced with increasing salinity level as compared with Control. However, the concentration of Na+ in leaves increased and K+ concentration decreased with increasing salinity level. Physiological parameters, i.e. relative water content (RWC), membrane stability index (MSI), chlorophyll stability index, carotenoid stability index and K+: Na+ ratio, in leaves tended to be higher in the BNH-3 variety than in other varieties. Shoot dry weight showed highly positive significant correlation with RWC, MSI, K+ concentration and K+:Na+ ratio, while it was negatively correlated with Na+ concentration (P<0.01). All BN hybrid varieties and the tri-specific hybrid studied were susceptible to salinity stress, showing marked reductions in growth as the level of salinity increased above 4 dS/m. However, even at salinity levels producing EC of 8 dS/m these varieties still produced 25‒44% DM yields. There are prospects for improving forage yields from saline soils by planting these hybrids but further breeding studies are warranted to identify germplasm with greater tolerance of saline conditions if these soils are to be utilized effectively to contribute more to supplying forage to support the world’s ruminant population. Keywords: Cenchrus americanus, Cenchrus purpureus, Cenchrus squamulatus, dry matter yields, Pennisetum hybrids, salt-tolerance, tropical grasses. Resumen Se examinaron las respuestas fisiológicas de 3 variedades híbridas de Bajra-Napier (Cenchrus spp., syn. Pennisetum spp.), a saber, BNH-3, BNH-6, BNH-10, y 1 híbrido ttri-específico (TSH) bajo diferentes gradientes de salinidad del suelo: Control, 4, 6 y 8 dS/m de conductividad eléctrica (EC), en un ensayo en macetas. El experimento se realizó en un diseño factorial completamente al azar con 3 repeticiones. El peso seco del brote, el peso seco de la raíz, la relación raíz:brote y las concentraciones de clorofila a, clorofila b, clorofila total y carotenoides se redujeron con el aumento del nivel de salinidad en comparación con el Control. Sin embargo, la concentración de Na+ en las hojas aumentó y la de K+ disminuyó con el aumento del nivel de salinidad. Los parámetros fisiológicos: contenido relativo de agua (RWC), índice de estabilidad de la membrana (MSI), índice de estabilidad de la clorofila, índice de estabilidad de los carotenoides y la Correspondence: S.N. Dheeravathu, ICAR-Indian Grassland and Fodder Research Institute, Jhansi, Uttar Pradesh, India. E-mail: sevanayak2005@gmail.com Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 338 S.N. Dheeravathu, K. Singh, P. W. Ramteke, Reetu, N. Dikshit, M. Prasad, D. Deb, and T.B. Vadithe relación K+: Na+, en las hojas tendieron a ser más altos en la variedad BNH-3 que en otras variedades. El peso seco de los brotes mostró una correlación significativa altamente positiva con el RWC, el MSI, la concentración de K+ y la relación K+:Na+, mientras que se correlacionó negativamente con la concentración de Na+ (P<0.01) Todas las variedades híbridas BN y el híbrido tri-específico estudiado fueron susceptibles al estrés por salinidad, mostrando marcadas reducciones en el crecimiento a medida que el nivel de salinidad aumentaba por encima de 4 dS/m. Sin embargo, incluso a niveles de salinidad que producían una EC de 8 dS/m, estas variedades seguían produciendo un rendimiento de 25‒44% de materia seca. Hay perspectivas de mejorar los rendimientos de forraje de los suelos salinos mediante la siembra de estos híbridos, pero se justifica la realización de más estudios de mejoramiento para identificar el germoplasma con mayor tolerancia a las condiciones de salinidad si se quiere utilizar estos suelos de manera eficaz para contribuir más al suministro de forraje para mantener a la población mundial de rumiantes. Palabras clave: Cenchrus americanus, Cenchrus purpureus, Cenchrus squamulatus, gramíneas tropicales, híbridos de Pennisetum, rendimiento de materia seca, tolerancia a la sal. Introduction the major constraint in green fodder production. There is a need to use degraded lands, particularly saline soils, Salinity is one of the major abiotic stresses of arid and by identifying salt-tolerant crops and grasses, which semi-arid regions that affect crop growth, development and could be used as fodder for grazing livestock (Kumar productivity (Pons et al. 2011). About 20% of the world’s and Sharma 2020). cultivated area and about half of the world’s irrigated lands Bajra-Napier (BN) hybrid is an interspecific hybrid are affected by salinity stress (Sairam and Tyagi 2004). between bajra [Cenchrus americanus (L.) Morrone, More than 800 million hectares of land throughout the the name currently accepted by the GRIN taxonomy world are adversely affected by high salinity (Munns and (npgsweb.ars-grin.gov/gringlobal/taxon/taxonomysearch) Tester 2008). In India, salt-affected soils occupy an area of for Pennisetum glaucum) and Napier grass (Cenchrus about 6.73 Mha of which saline and sodic soils constitute purpureus (Schumach.) Morrone, syn. Pennisetum about 40 and 60%, respectively (Singh et al. 2010). purpureum). Bajra-Napier hybrid and tri-specific hybrid The physiological responses of a plant to salinity (Cenchrus americanus × C. purpureus × C. squamulatus; are often complex and multi-faceted, which makes syn. Pennisetum glaucum × P. purpureum × P. squamulatum) experiments difficult to design and interpret (Negrão are perennial, multi-cut forage grasses with high biomass et al. 2017). Salinity poses two major threats to plant and high nutritional quality coupled with high palatability growth, i.e. osmotic stress and ionic stress (Flowers and (Singh et al. 2018). BN hybrids can withstand drought for Colmer 2008). The responses to these changes are often a short spell and currently about one hundred thousand accompanied by a variety of symptoms, such as a decrease hectares are grown in India. Considering the adverse in leaf area, an increase in leaf thickness and succulence, effects of salt stress on crop growth and productivity, abscission of leaves, necrosis of roots and shoots and a the development of salt-tolerant genotypes and more decrease in internode lengths (Parida and Das 2005). particularly salt-tolerant BN hybrids and tri-specific Roots, being a primary organ, are directly exposed to hybrids could play a major role in sustaining livestock saline environments, but their growth is less vulnerable production in the salt-affected lands and would also be to salinity than that of shoots (Picchioni et al. 1990). The helpful in future breeding programs. We hypothesize that accumulation of Na+ in roots is an adaptive response used these hybrids are salt-tolerant and should produce well in by various woody species to avoid its toxicity in shoots saline soils. Keeping in view the above facts, the present (Picchioni et al. 1990; Gucci and Tattini 1997). experiment was conducted to evaluate the physiological Livestock production is the backbone of Indian responses in 3 BN hybrids and 1 tri-specific hybrid (TSH) agriculture and it has been projected that the livestock grown under saline conditions in a glasshouse. population will increase to around 286.5 million adult cattle units by 2050 (IGFRI Vision 2050). The major Materials and Methods concern is to ensure sufficient green fodder is available throughout the year, as there is a deficiency of green fodder Experimental design and concentrate feed (Semple et al. 2003). Cultivation of cereals and cash crops has resulted in the reduction in the This pot study was conducted at Crop Improvement area of land for fodder production for livestock, which is Division of ICAR - Indian Grassland and Fodder Research Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Salinity stress in Bajra-Napier hybrids 339 Institute, Jhansi (25°45’ N, 78°58’ E; 243 masl), during Reduction in performance relative to Controls Rabi (winter season, October‒March) 2018 in a complete (%ROC) was calculated as follows: randomized block design. Root slips of 4 varieties, viz. BNH-3, BNH-6, BNH-10 and TSH were collected from %ROC = Value for Control-Value for stressed plants × 100 ICAR-IGFRI Technology Demonstration Block and Value for Control planted in pots containing 6 kg of soil at 4 different (Control, 4, 6 and 8 dS/m) levels of salinity and 3 replications. The Statistical analysis initial properties of the collected soil were: slightly alkaline with pH 7.62; electrical conductivity (ECe) 1.12 dS/m; The study was conducted as a factorial experiment based and low in organic carbon (0.49%). The total nitrogen, on a completely random design with 3 replications. available phosphorus and potassium concentrations in the The data were analyzed by Microsoft Excel and SAS soil were 213, 13.8 and 191 kg/ha, respectively. Saline 9.3 statistical analytical tool and the significance of conditions were created by adding a mixture of NaCl, differences between treatment means was checked with Na2SO4, MgCl2 and CaSO4 (in ratio 13:7:1:4) to pots to Duncan's multiple range test at P<0.05. provide electrical conductivity of treated soils of 4, 6 and 8 dS/m at 30 days after transplanting with a Control (1.12 Results dS/m) for comparison. Plants were harvested at 30 and 55 days after stress was imposed. Significant to highly significant interactions were found between variety and level of salinity for SDW and RWC Shoot dry weight and root dry weight at the first harvest and for MSI and carotenoids at the second harvest, whereas highly significant interactions At each harvest, i.e. at 60 and 85 days of age, above- were found for K+ and Na+ concentrations and K+:Na+ ground material was removed, placed in paper bags ratio at the first harvest and for SDW, RDW, RWC, Chl and oven-dried at 45 °C until a constant weight was a, Chl b, Total Chl, K+ and Na+ concentrations at the reached after about 72 hours to determine shoot dry second harvest (Table 1). weight (SDW). At the 85-day harvest, roots were also collected and dried (RDW). Root:shoot ratio (RSR) was Effects of salt stress on shoot dry weight, root dry determined based on the shoot and root values measured. weight and root:shoot ratio Physiological parameters Shoot dry weight (SDW), root dry weight (RDW) and root:shoot ratio (RSR) declined for all varieties as level The acetone method was applied to green leaf samples of salinity increased (Table 2). While an ECe level of (200 mg fresh weight) from the 3rd leaf from top portion 4 dS/m had no significant effect on growth at the first to extract chlorophyll a (Chl a), chlorophyll b (Chl b), harvest, at the highest salinity level reduction in SDW total chlorophyll (Total Chl) and carotenoid (Car) and over Controls ranged from 56% for BNH-3 to 75% for 100 mg leaf (green leaf) samples were used to determine BNH-6, and at the second harvest from 61% for BNH-3 to membrane stability index (MSI) (3rd leaf from top 72% for BNH-10. Reductions in RDW over the Controls portion) according to the method of Premachandra et at the second harvest were more pronounced than for al. (1990). The relative water content (RWC) of 100 mg SDW with reductions of 19‒33% at 4 dS/m and 71‒78% leaf samples (3rd leaf from top portion) was analyzed at 8 dS/m. As a result, RSR declined from 0.42‒0.54:1 by the method of Weatherley (1950). Sodium (Na+) and for Controls to 0.33‒0.39:1 at the highest salinity level potassium (K+) concentrations in 1 g dry leaf (sampled (Table 2). At the first harvest, SDW showed positive from young shoot leaves) samples were determined significant correlations (P<0.01) with RWC, MSI, K+ by the flame photometer method of Jackson (1973). and K+:Na+ ratio and negative correlations with Chl a, Chlorophyll stability index (CHSI) was calculated Chl b, Total Chl, carotenoid and Na+ concentrations. At by following the method described by Sairam et al. the second harvest, SDW indicated positive significant (1997) using the formula: (Total chlorophyll in salt- correlations with RDW, RSR, RWC, MSI and Chl b, stressed plants/Total chlorophyll in Control plants) × Total Chl, carotenoid and K+ concentrations, while Na+ 100; a similar formula was used to determine carotenoid concentrations showed negative significant correlations stability index (CARSI). (P<0.01) with RSR, MSI and K+:Na+ ratio (Table 6). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 340 S.N. Dheeravathu, K. Singh, P. W. Ramteke, Reetu, N. Dikshit, M. Prasad, D. Deb, and T.B. Vadithe Table 1. ANOVA results of the effects of salt stress on SDW, RDW, RWC, MSI, Chl a, Chl b, Total Chl, Car, K+, Na+ and K+:Na+ ratio of Cenchrus hybrid varieties. First Harvest Mean square Variable df SDW RWC MSI Chl a Chl b Total Chl Carotenoids Chl a+b K+ Na+ K+:Na+ ratio V 3 4.43* 90* 301** 0.18** 0.10** 0.55** 0.008** 0.004NS 0.069** 0.268** 0.081** ECe 3 164.45** 569** 363** 0.02* 0.02* 0.09* 0.002NS 0.001NS 1.859** 0.085** 0.765** V×ECe 9 1.59* 7* 2NS 0.004NS 0.003NS 0.004NS 0.001NS 0.08NS 0.027** 0.075** 0.010** Error 30 0.726 74.564 46.657 0.008 0.005 0.021 0.001 0.045 0.004 0.004 0.002 Second Harvest Mean square SDW RDW RWC MSI Chl a Chl b Total Chl Carotenoids Chl a+b K+ Na+ K+/Na+ ratio V 3 4.14* 2.99** 85** 110* 0.15** 0.07** 0.4** 0.02** 0.03NS 0.183** 0.035** 4.123 NS ECe 3 146.75** 41.60** 4669** 159* 0.40** 0.21** 1** 0.01** 0.03NS 17.843** 0.575** 1237.96 NS V×ECe 9 1.50** 0.14** 48** 14* 0.01** 0.01** 0.04** 0.0005* 0.01NS 0.170** 0.014** 0.860 NS Error 30 0.155 0.005 0.826 36.355 0.002 0.0012 0.006 0.0001 0.018 0.004 0.003 1.779 SDW - shoot dry weight RDW - root dry weight RWC - relative water content MSI - membrane stability index Chl a - chlorophyll a Chl b - chlorophyll b Total Chl -total chlorophyll Car - carotenoid K+ - potassium Na+ - sodium K+:Na+ ratio - potassium to sodium ratio V - variety ECe - electrical conductivity of the extract of a saturated soil-paste. Table 2. Effects of salt stress on shoot dry weight (g/pot), root dry weight (g/pot) and root:shoot ratio of Cenchrus hybrid varieties at 60 days (1st harvest) and 85 days (2nd harvest) of age. Variety/ Shoot dry weight (1st harvest) Shoot dry weight (2nd harvest) Root dry weight (2nd harvest) Root:shoot ratio Treatment TSH BNH-6 BNH-3 BNH-10 TSH BNH-6 BNH-3 BNH-10 TSH BNH-6 BNH-3 BNH-10 TSH BNH-6 BNH-3 BNH-10 Control 11.3±0.6 12.0±0.9 11.2±0.5 12.4±0.6 10.8±0.1 10.0±0.5 10.1±0.2 12.4±0.2 5.5±0.1 4.8±0.0 6.0±0.1 5.2±0.0 0.51 0.48 0.54 0.42 ECe4 11.0±0.5 11.5±0.2 10.9±0.3 12.0±0.3 9.6±0.2 9.5±0.1 8.7±0.1 11.2±0.4 4.5±0.0 3.2±0.0 4.7±0.0 4.1±0.0 0.47 0.34 0.43 0.36 ROC% 2 4 2 3 11 5 14 10 19 33 22 21 ECe6 6.2±0.1 5.7±0.0 7.8±0.1 6.8±0.2 5.1±0.1 4.2±0.1 5.8±0.0 5.6±0.1 1.9±0.0 1.4±0.0 2.9±0.0 1.9±0.0 0.37 0.33 0.37 0.34 ROC% 45 53 30 45 52 58 42 55 65 71 52 63 ECe8 4.5±0.1 3.0±0.1 4.9±0.0 4.0±0.1 3.7±0.1 3.3±0.0 3.9±0.0 3.4±0.0 1.4±0.0 1.1±0.0 1.7±0.0 1.1±0.0 0.39 0.33 0.35 0.33 ROC% 60 75 56 68 66 68 61 72 74 78 71 79 Mean (n = 3) ROC% - per cent reduction over Control for ECe of 4, 6 and 8 dS/m. Salinity stress in Bajra-Napier hybrids 341 Effects of salt stress on Relative water content and dS/m. Chlorophyll stability index (CHSI) and carotenoid Membrane stability index stability index (CARSI) also declined as salinity level increased, with the major reduction occurring between 6 Relative water content (RWC; %) and Membrane stability and 8 dS/m (Table 5). Photosynthetic pigments (Chl a, Chl index (MSI; %) were considered reliable parameters to b and Total Chl) showed significant positive correlations assess the salt stress and tolerance of crop species. RWC of with each other and carotenoid concentrations at both leaf declined in all varieties with increasing salinity, with first and second harvests (Table 6). percentage reduction relative to Controls at the highest salinity level ranging from 48 to 63% for the different Effects of salt stress on K+ and Na+ concentrations and varieties at the first harvest and from 50 to 69% at the K+:Na+ ratio in leaves second harvest (Table 3). Membrane stability index (MSI) for all varieties also declined with increasing salinity at Potassium concentrations in leaves at the first and second first (P<0.01) and second (P<0.05) harvests. RWC showed harvests declined as salinity levels increased (Table 7) but highly significant positive correlations with SDW, MSI, K+ differences failed to reach significance (P>0.05) despite and K+:Na+ ratio at the first harvest, and highly significant reductions in concentrations at 8 dS/m ECe being about 52 positive correlations with SDW, RDW, MSI and K+ and and 70%, respectively. In contrast, sodium concentrations moderately significant correlation with K+:Na+ ratio at the showed little consistent response at the first harvest second harvest (Table 6). MSI showed highly significant (P>0.05) but increased markedly for TSH, BNH-3 and positive correlations with SDW, RWC, K+ and K+:Na+ ratio BNH-6 and decreased for BNH-10 at the second harvest at the first harvest, and with SDW, RDW, RSR, RWC, K+ with again no significant responses (P>0.05). In general and K+:Na+ ratio at the second harvest. K+:Na+ ratio declined as level of salinity increased at both harvests with the effect being much more pronounced Effects of salt stress on photosynthetic pigments and at the second harvest (except for BNH-10) but again Chlorophyll stability index and Carotenoid stability index differences were not significant (P>0.05). In addition to correlations mentioned earlier, K+ Data in Table 4 show that chlorophyll a, chlorophyll concentrations showed significant positive correlations b, total chlorophyll and carotenoid concentrations with K+:Na+ ratio at the first and second harvests, while decreased as salinity increased at both harvests with Na+ concentration showed significant negative correlations the main part of the decline occurring between 6 and 8 with K+:Na+ ratio in first and second harvests (Table 6). Table 3. Effects of salt stress on Relative water content and Membrane stability index in Cenchrus hybrid varieties. Variety/ Relative Water Content (%) Treatment 1st Harvest 2nd Harvest TSH BNH-6 BNH-3 BNH-10 TSH BNH-6 BNH-3 BNH-10 Control 82.7±1.21a 78.4±2.40a 74.6±1.70a 78.7±2.04a 77.1±2.23a 70.0±2.31bc 68.0±1.73cd 72.0±1.79b 4 dS/m 78.7±1.91ab 70.0±2.31b 73.0±1.15a 74.0±1.73a 67.0±1.73d 61.0±2.31e 60.0±1.15e 66.0±0.58d (5) (11) (2) (6) (13) (13) (12) (8) 6 dS/m 52.0±2.19c 49.0±2.48c 60.0±2.19a 56.0±1.50c 44.0±1.15g 38.0±1.73h 48.0±1.04f 48.0±1.44f (37) (37) (20) (29) (43) (46) (29) (33) 8 dS/m 38.0±1.44d 29.0±2.31de 39.0±0.92d 36.0±1.73d 24.0±1.33k 23.0±2.31k 34.0±1.73i 30.0±2.19j (54) (63) (48) (54) (69) (67) (50) (58) Variety/ Membrane Stability Index (%) Treatment 1st Harvest 2nd Harvest TSH BNH-6 BNH-3 BNH-10 TSH BNH-6 BNH-3 BNH-10 Control 65.0±1.44a 54.2±1.54ab 64.7±1.20a 58.3±0.96a 53.0±3.18a 48.0±2.6a 50.0±2.89a 49.0±3.76a 4 dS/m 57.7±1.50a 46.2±2.17abc 61.5±1.17a 53.3±1.80ab 33.09±1.8b 32.0±2.5b 37.0±1.73b 35.0±1.2b (11) (15) (5) (9) (26) (33) (26) (28) 6 dS/m 42.73±1.92abc 31.72±2.26c 42.11±0.87abc 38.16±1.25bc 34±1.62c 30±2.48bc 35±1.45bc 32±2.37bc (34) (41) (35) (35) (37) (38) (30) (35) 8 dS/m 24.21±1.45c 19.31±2.13c 28.32±1.97c 24.71±1.84c 26±1.82d 14±2.23d 26±1.51cd 25±1.62d (63) (64) (56) (58) (51) (70) (48) (49) Means within column(s) followed by the same letter(s) are not significantly different (P>0.05). N=3. Values in parenthesis depict per cent reduction over control (ROC%). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 342 S.N. Dheeravathu, K. Singh, P. W. Ramteke, Reetu, N. Dikshit, M. Prasad, D. Deb, and T.B. Vadithe Table 4. Effects of salt stress on chlorophyll and carotenoid concentrations (mg/g fresh weight) in Cenchrus hybrid varieties Variety/Treatment First harvest Second harvest Chl a Chl b Total Chl Car Chl a Chl b Total Chl Car TSH Control 0.60+0.0 0.47+0.048 1.06+0.004 0.22+0.03 0.56+0.02 0.43+0.017 0.99+0.038 0.16+0.009 4abc abc cbd a bcd edf ed bcd BNH-6 0.81+0.05 0.65+0.028 1.46+0.083 0.19+0.02 0.75+0.06 0.59+0.021 1.34+0.076 0.11+0.028 ab ab ab ab a ab ab fhig BNH-3 0.88+0.05 0.69+0.038 1.57+0.008 0.25+0.02 0.78+0.06 0.61+0.035 1.39+0.052 0.22+0.030 a a a a a a a a BNH-10 0.62+0.03 0.49+0.042 1.11+0.076 0.19+0.02 0.57+0.04 0.45+0.015 1.03+0.023 0.15+0.007 abc abc abcd ab bcd cde cde cdef TSH 4 dS/m 0.56+0.04 0.42+0.0 0.99+0.00 0.20+0.018 0.51+0.02 0.37+0.015 0.88+0.03 0.15+0.008 bc (7) 4bc (9) 4cbd (7) ab (7) cde (10) efg (14) efg (11) cde (8) BNH-6 0.73+0.05 0.57+0.02 1.33+0.078 0.17+0.016 0.66+0.05 0.49+0.017 1.14+0.07 0.10+0.006 ab (9) ab (12) ab (9) ab (9) abc (13) cd (22) bcd (15) hifg (15) BNH-3 0.84+0.04 0.65+0.03 1.50+0.007 0.23+0.023 0.71+0.03 0.54+0.018 1.25+0.05 0.20+0.016 a (5) a (6) a (5) a (6) ab (8) abc (12) ab (10) ab (9) BNH-10 0.58+0.03 0.46+0.03 1.04+0.072 0.18+0.016 0.51+0.03 0.40+0.013 0.91+0.02 0.12+0.006 abc (6) abc (7) bcd (6) ab (7) cde (11) defg (14) edf (12) b (18) TSH 6 dS/m 0.46+0.05 0.39+0.016 0.85+0.031 0.18+0.01 0.35+0.03 0.30+0.023 0.65+0.052 0.13+0.012 c (23) bc (17) cd (17) ab (16) fg (38) g (29) g (34) cdefg (20) BNH-6 0.61+0.04 0.49+0.038 1.10+0.006 0.15+0.01 0.47+0.01 0.41+0.023 0.88+0.037 0.08+0.005 abc (25) abc (25) abcd (25) b (20) def (38) defg (30) efg (34) fhig (28) BNH-3 0.72+0.05 0.58+0.029 1.30+0.076 0.23+0.02 0.65+0.03 0.49+0.040 1.14+0.068 0.17+0.004 a (18) ba (17) ab (17) a (8) abc (16) bcd (19) bcd (18) bc (22) BNH-10 0.49+0.03 0.41+0.017 0.90+0.018 0.18+0.01 0.40+0.02 0.34+0.012 0.74+0.029 0.13+0.012 bc (21) bc (16) cd (16) ab (8) ef (30) fg (24) fg (28) defg (10) TSH 8 dS/m 0.30+0.03 0.26+0.016 0.56+0.019 0.14+0.01 0.21+0.01 0.20+0.007 0.41+0.020 0.08+0.003 c (50) c (44) d (44) ab (36) gh (62) h (53) h (59) hij (51) BNH-6 0.32+0.03 0.28+0.029 0.60+0.054 0.1+0.01 0.12+0.02 0.12+0.030 0.24+0.029 0.04+0.020 c (60) c (57) d (57) b (37) h (84) h (80) h (82) j (64) BNH-3 0.54+0.02 0.43+0.019 0.97+0.005 0.2+0.02 0.46+0.00 0.35+0.023 0.81+0.027 0.12+0.014 bc (39) bc (38) bcd (38) b (35) def (41) fg (43) efg (41) defg (46) BNH-10 0.31+0.03 0.25+0.019 0.56+0.010 0.1+0.02 0.21+0.02 0.19+0.004 0.43+0.026 0.09+0.004 c (50) c (49) d (49) b (37) gh (63) h (58) h (58) hig (39) Means in column (s) followed by the same letter (s) are not significantly different (P>0.05). N=3. Values in parenthesis depict per cent reduction over control (ROC%). Table 5. Effects of salt stress on chlorophyll stability index and carotenoid stability index in Cenchrus hybrid varieties. Variety/ Chlorophyll stability index (%) Treatment 1st harvest 2nd harvest EC4 EC6 EC8 EC4 EC6 EC8 TSH 93 80 53 88 66 41 BNH-6 91 75 41 85 66 18 BNH-3 95 82 61 91 83 87 BNH-10 94 81 50 94 79 21 Variety/ Carotenoid stability index (%) Treatment 1st harvest 2nd harvest EC4 EC6 EC8 EC4 EC6 EC8 TSH 93 84 64 92 80 49 BNH-6 91 80 63 85 72 36 BNH-3 94 92 65 91 78 54 BNH-10 93 92 63 82 90 61 Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Salinity stress in Bajra-Napier hybrids 343 Table 6. Correlations among different parameters in Cenchrus hybrid varieties subjected to salinity stress. PM 1st Harvest SDW RWC MSI Chl a Chl b Total_Chl Car Chl a:b % K+ % Na+ K+:Na+ ratio SDW — RWC 0.977*** — MSI 0.930*** 0.964*** — Chl a -0.139 -0.041 -0.059 — Chl b -0.130 -0.033 -0.042 0.996*** — Total Chl -0.130 -0.036 -0.054 0.999*** 0.998*** — Car -0.173 -0.034 -0.027 0.801*** 0.792*** 0.791*** — Chl a:b -0.261 -0.179 -0.243 0.682** 0.622* 0.660** 0.633** — % K+ 0.796*** 0.817*** 0.841*** -0.312 -0.310 -0.314 -0.004 -0.251 — %Na+ -0.178 -0.223 -0.300 0.337 0.331 0.341 0.056 0.269 -0.368 — K+:Na+ ratio 0.720** 0.751*** 0.799*** -0.351 -0.349 -0.355 0.007 -0.258 0.976*** -0.548* — PM 2nd Harvest SDW RDW RSR RWC MSI Chl a Chl b Total Chl Car Chl a:b % K+ % Na+ K+:Na+ ratio SDW — RDW 0.927*** — RSR 0.651** 0.858*** — RWC 0.963*** 0.936*** 0.710** — MSI 0.808*** 0.880*** 0.802*** 0.874*** — Chl a -0.258 -0.194 0.016 -0.123 -0.139 — Chl b -0.264 -0.184 0.032 -0.133 -0.150 0.993*** — Total Chl -0.274 -0.201 0.018 -0.140 -0.152 0.998*** 0.997*** — Car -0.424 -0.287 0.040 -0.264 0.020 0.734** 0.717** 0.732** — Chl a:b -0.260 -0.194 0.039 -0.120 -0.074 0.839*** 0.782*** 0.815*** 0.695** — % K+ 0.779*** 0.838*** 0.764*** 0.786*** 0.816*** -0.199 -0.189 -0.203 -0.151 -0.158 — % Na+ -0.596* -0.679** -0.598* -0.714** -0.614* 0.132 0.161 0.148 0.243 0.001 -0.379 — K+:Na+ ratio 0.572* 0.732** 0.857*** 0.663** 0.778*** -0.041 -0.076 -0.060 0.115 0.200 0.654** -0.711** — PM - parameters; SDW - shoot dry weight; RDW - root dry weight; RSR - root:shoot ratio; RWC - relative water content; MSI - membrane stability index; Chl a - chlorophyll a; Chl b - chlorophyll b; Total Chl - total chlorophyll; Chl a:b - Chl a:Chl b ratio; Car - carotenoid; K+ - potassium; Na+ - sodium; K+:Na+ ratio - potassium: sodium ratio. 344 S.N. Dheeravathu, K. Singh, P. W. Ramteke, Reetu, N. Dikshit, M. Prasad, D. Deb, and T.B. Vadithe Table 7. Effects of salt stress on Na+ and K+ concentrations in leaves of Cenchrus hybrid varieties over 2 harvests. Variety Treatment 1st Harvest 2nd Harvest % K+ %Na+ K+:Na+ ratio % K+ % Na+ K+:Na+ ratio TSH Control 1.81±0.06bd 1.67±0.03ca 1.09±0.02bd 1.93±0.03cd 0.08+0.00a 24.08+0.40bd ECe4 1.12±0.02bc 1.90±0.05c 0.59±0.01bc 1.53±0.02c 0.23±0.01ab 6.71±0.19bc ECe6 0.95±0.03b 1.80±0.06cb 0.53±0.03b 0.51±0.02bc 0.56±0.01ac 0.92±0.06b ECe8 0.88±0.03ab 1.83±0.03cb 0.48±0.01ab 0.26±0.01ac 0.67±0.01ad 0.39±0.01ab BNH-6 Control 1.67±0.05ad 1.65±0.03ad 1.01±0.05ad 1.22±0.01ad 0.05±0.005ab 24.48±2.18ad ECe4 1.00±0.07ac 1.98±0.01cd 0.50±0.03ac 0.65±0.01ac 0.23+0.01b 2.78+0.10ac ECe6 0.82±0.01ab 2.00±0.06bd 0.41±0.02ab 0.59±0.01ab 0.59±0.02bc 1.01±0.02ab ECe8 0.76±0.02a 2.27±0.04bd 0.34±0.00a 0.49±0.02a 0.70±0.01bd 0.70±0.04a BNH-3 Control 1.80±0.04ad 1.64±0.03a 1.10±0.04d 1.82±0.01bd 0.09±0.00a 20.90±0.78abd ECe4 1.13±0.01ac 1.64±0.04ac 0.69±0.01cd 0.76±0.02bc 0.20±0.01ab 3.70±0.36abc ECe6 1.12±0.01ab 1.70±0.02ab 0.66±0.00bd 0.67±0.01b 0.30+0.01ac 2.21+0.17ab ECe8 0.93±0.08a 1.50±0.03ab 0.62±0.04ad 0.58±0.01ab 0.46±0.01ad 1.27±0.03aab BNH-10 Control 1.62±0.03bd 1.70±0.02ab 0.95±0.03bd 2.31±0.01ac 0.59±0.01ab 3.92±0.05abd ECe4 1.33±0.03bc 1.95±0.03bc 0.68±0.02bc 0.68±0.02c 0.23±0.01b 3.00±0.05abc ECe6 0.99±0.01b 1.67±0.04b 0.59±0.01b 0.61±0.01bc 0.29±0.02bc 2.13±0.18ab ECe8 0.75±0.02ab 1.60±0.04b 0.47±0.004ab 0.48±0.02ac 0.30±0.02bd 1.60±0.04ab Means in columns followed by the same letter (s) are not significantly different (P>0.05), where letter “a” represents the least value. N = 3. Discussion or no effect on growth and dry fodder yield at the low level of salinity (4,000 ppm NaCl) in perennial ryegrass, Salinity stress affects growth and productivity in plants by tall fescue and orchard grass. altering physiological mechanisms like water relations, As a macronutrient, potassium (K+) mostly metabolism, ion accumulation, nutrient imbalance and contributes to a plant’s survival when exposed to various Reactive Oxygen Species (ROS) generation. While environmental stresses such as drought, salinity and cold salinity tolerance in annual forages and plants is well (Wang and Wu 2013). The positive role of K+ in the defined (Roy and Chakraborty 2014; Munns et al. response to salinity is due to: (1) its competitiveness with 2020a, 2020b; Rahimi et al. 2021), this is not the case sodium (Na+) for binding sites and maintaining relative for perennial grasses and plants. Salts are common water content (RWC) in plants (Capula-Rodríguez et al. and necessary components of soil and many salts (e.g. 2016); and (2) its ability to regulate the balance between sodium nitrate, potassium carbonate, bicarbonate and ROS and antioxidants to adjust protein synthesis potassium chloride) are essential plant nutrients at low and stomatal function, thereby improving a plant’s concentrations. photosynthetic status (Wang et al. 2013). Moreover, Grasses are quite variable in their tolerance of salinity foliar spraying of perennial ryegrass with KNO3 (10 mM) in terms of growth (Khan et al. 1999; Hester et al. enhanced growth, chlorophyll concentration and K:Na 2001; Muscolo et al. 2003; Joshi et al. 2004). Muscolo ratio when grown under saline conditions. The decrease et al. (2003) reported that the biomass of kikuyu in RWC under saline conditions can be attributed to a grass (Cenchrus clandestinus formerly Pennisetum reduction of soil water potential in the root zone (Munns clandestinum) leaves and roots was affected by 150 mM et al. 2006). Sairam and Tyagi (2004) and Singh et al. NaCl and extensively reduced at high concentration of (2020) suggested that reduced shoot height, leaf area NaCl (200 mM) compared with Control, while growth and number of leaves in sensitive genotypes under saline was little affected at lower concentrations of NaCl (50 conditions may be due to their leaves having lower mM). relative water content and membrane stability index. In Our results showed that shoot dry weight, root dry addition, the accumulation of Na+ and Cl− ions can lead weight and root:shoot ratio declined for all varieties as to the production of ROS which, in turn, increases the the level of salinity increased, while the low level of permeability of the cell membrane and decreases MSI 4 dS/m had very little or no effect on growth and dry (Nazar et al. 2011). RWC and MSI are good indicators matter yield. These results agreed with Al-Ghumaiz et of leaf water status and stability of membranes and al. (2017), who reported that dry fodder yield declined at are successfully used to determine stress resistance high levels of salinity (8,000 ppm NaCl) with very little or tolerance in many crop plants (Bangar et al. 2019; Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Salinity stress in Bajra-Napier hybrids 345 Rahimi et al. 2021). Many reports reveal that RWC and other studies that indicate that plants subjected to MSI are reduced under drought and salinity (Bangar et al. increased salinity levels show decreased photosynthetic 2019; Rahimi et al. 2021) and those plants that maintain pigments (Aghaleh et al. 2009; Jampeetong and Brix high RWC and MSI under extreme stress are regarded as 2009; Al-humaiz et al. 2017). being more stress-tolerant (Bangar et al. 2019; Rahimi Numerous studies have shown that salt tolerance et al. 2021). In our study, the reductions in RWC at the is ultimately manifested in plants through several first harvest at the highest salinity level ranged from physiological processes including Na+ uptake and 48 to 63% and at the second harvest from 50 to 69%, exclusion, in homeostasis, especially between K+:Na+ while reductions in MSI ranged from 56 to 64% at the ratio and partitioning (Ren et al. 2005). Various studies first harvest and from 48 to 71% at the second harvest. have shown that plants increase Na+ uptake and reduce This indicates that, while these varieties can tolerate low K+ uptake under salt stress (Horie et al. 2001; Zhu 2003). salinity levels, impacts on these parameters at higher The K+ ions are beneficial to plants and by increasing K+ levels of salinity are quite significant. In our study, concentration, plants can reduce the absorption of Na+ shoot dry weight (SDW) was positively correlated with ions to a certain extent, thus improving the K+: Na+ ratio. RWC and MSI at both harvests. The highest reductions Generally, the data in Table 7 indicate that the mean in SDW relative to Controls at both first and second percentages of Na+ in leaves of all varieties increased harvests occurred at the highest salinity level and ranged with increase in salinity, while K+ concentration from 56 to 75% at the first harvest and from 61 to 72% at declined because Na+ effectively competes with K+ the second harvest, which are of comparable magnitude for uptake in a common transport system, i.e. the Na+ to the reductions in RWC and MSI. Our results are in concentration in saline environments is usually greater conformity with Rahimi et al. (2021), who reported than that of K+ (Gorham et al. 1990). In other words, the that RWC and MSI were significantly and positively decrease in K+ resulted from the presence of excessive correlated with K+: Na+ ratio and K+ concentration in Na+ in the growth medium because high external Na+ shoots and roots of rye grass under salinity stress. concentrations are known to have an antagonistic effect Chlorophyll has been proposed as a useful biochemical on K+ uptake in plants (Sarwar et al. 2003). Interestingly indicator of salt tolerance in different plants (Akram and K+ concentration in leaf tissue of Controls was relatively Ashraf 2011) as chlorophyll and carotenoids are involved similar for both harvests, while Na+ concentration was in the primary step concerning energy production during much lower at the second than the first harvest. photosynthesis. Since salinity affects chlorophyll and carotenoid levels, it is not surprising that the growth Conclusions of plants is inhibited when grown in saline situations. Salt stress increases the activity of chlorophyllase, This study has shown that the varieties of BN hybrids which promotes degradation of chlorophyll and reduces and the tri-specific hybrid studied were all susceptible to chlorophyll concentration in plants (Yang et al. 2011). salinity stress, showing marked reductions in growth as Although salt stress can reduce chlorophyll concentration, the level of salinity increased above 4 dS/m. However, the extent of the reduction depends on the salt tolerance dry matter yields obtained at high salinity level (ECe of the particular plant species. Differences in reductions of 8 dS/m) were still at the range of 25‒44%. There are in chlorophyll concentrations between the different prospects for improving forage yields from saline soils varieties in our study suggest that the degree of tolerance by planting these hybrids but further breeding studies are of salinity by the various varieties was relatively similar, warranted to identify germplasm with greater tolerance although BNH-3 did display lower reductions relative to of saline conditions if these soils are to be utilized Control than other varieties as salinity level increased. effectively to contribute more to supplying forage to Carotenoids play an important role as a precursor in support the world’s ruminant population. signalling during plant development under abiotic stress as they protect the membranes from oxidative Acknowledgments damage (Verma and Mishra 2005). While all varieties demonstrated reductions in carotenoid concentrations The authors acknowledge Indian Council of Agricultural relative to Controls with increasing salinity, at the higher Research (ICAR) and ICAR–Indian Grassland and salinity levels BNH-10 showed a tendency to suffer less Fodder Research Institute, Jhansi for financial support reduction than other varieties. These results corroborate for conducting the experiment. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 346 S.N. Dheeravathu, K. Singh, P. W. Ramteke, Reetu, N. Dikshit, M. Prasad, D. Deb, and T.B. Vadithe References Jackson ML. 1973. Soil chemical analysis. 2nd Edn. Prentice Hall of India Private Limited, New Delhi, India. (Note of the editors: All hyperlinks were verified 1 September 2021). Jampeetong A; Brix H. 2009. Effects of NaCl salinity on growth, morphology, photosynthesis and proline accumulation Aghaleh M; Niknam V; Ebrahimzadeh H; Razavi K. 2009. of Salvinia natans. Aquatic Botany 9(3):181–186. doi: Salt stress effects on growth, pigments, proteins and 10.1016/j.aquabot.2009.05.003 lipid peroxidation in Salicornia persica and S. europaea. Joshi AJ; Mali BS; Hinglajia H. 2004. Salt tolerance at Biologia Plantarum 53:243–248. doi: 10.1007/s10535- germination and early growth of two forage grasses 009-0046-7 growing in marshy habitats. Environmental and Akram SM; Ashraf M. 2011. Exogenous application of Experimental Botany 54(3):267–274. doi: 10.1016/j. potassium dihydrogen phosphate can alleviate the adverse envexpbot.2004.09.005 effects of salt stress on sunflower. Journal of Plant Nutrition Khan MA; Ungar IA; Showalter AM. 1999. Effects of salinity 34:1041–1057. doi: 10.1080/01904167.2011.555585 on growth, ion content and osmotic relations in Halopyrum Al-Ghumaiz NS; Abd-Elmoniem EM; Motawei MI. 2017. mucronatum (L.) Stapf. Journal of Plant Nutrition 22:191– Salt tolerance and K/Na ratio of some introduced forage 204. doi: 10.1080/01904169909365617 grass species under salinity stress in irrigated areas. Kumar P; Sharma PK. 2020. Soil Salinity and Food Security Communications in Soil Science and Plant Analysis in India. Frontiers in Sustainable Food Systems. 4:533781. 48(12):1494–1502. doi: 10.1080/00103624.2017.1374398 doi: 10.3389/fsufs.2020.533781 Bangar P; Chaudhury A; Tiwari B; Kumar S; Kumari R; Bhat Munns R; James RA; Läuchli A. 2006. Approaches to KV. 2019. Morphophysiological and biochemical response increasing the salt tolerance of wheat and other cereals. of mungbean [Vigna radiata (L.) Wilczek] varieties at Journal of Experimental Botany 57(5):1025–1043. doi: different developmental stages under drought stress. 10.1093/jxb/erj100 Turkish Journal of Biology 43:58–69. doi: 10.3906/biy- Munns R; Day DA; Fricke W; Watt M; Arsova B; Barkla 1801-64 BJ; Bose J; Byrt CS; Chen ZH; Foster KJ. 2020a. Energy Capula-Rodríguez R; Valdez-Aguilar LA; Cartmill DL; costs of salt tolerance in crop plants. New Phytologist Cartmill AD; Alia-Tejacal I. 2016. Supplementary 225(3):1072–1090. doi: 10.1111/nph.15864 calcium and potassium improve the response of Munns R; Passioura JB; Colmer TD; Byrt CS. 2020b. Osmotic tomato (Solanum lycopersicum L.) to simultaneous adjustment and energy limitations to plant growth in saline alkalinity, salinity, and boron stress. Communications soil. New Phytologist 225(3):1091–1096. doi: 10.1111/ in Soil Science and Plant Analysis 47(4):505–511. doi: nph.15862 10.1080/00103624.2016.1141924 Munns R; Tester M. 2008. Mechanisms of salinity tolerance. Flowers TJ; Colmer TD. 2008. Salinity tolerance in halophytes. Annual Review of Plant Biology 59(1):651–681. doi: New Phytologist 179:945–963. doi: 10.1111/j.1469- 10.1146/annurev.arplant.59.032607.092911 8137.2008.02531.x Muscolo A; Panuccio MR; Sidari M. 2003. Effects of salinity Gorham J; Wyn Jones RG; Bristol A. 1990. Partial on growth, carbohydrate metabolism and nutritive characterization of the trait for enhanced K+‐Na+ properties of kikuyu grass (Pennisetum clandestinum discrimination in the D genome of wheat. Planta 180:590– Hochst). Plant Science 164(6):1103–1110. doi: 10.1016/ 597. doi: 10.1007/BF02411458 S0168-9452(03)00119-5 Gucci R; Tattini M. 1997. Salinity tolerance in olive. In: Janik Nazar R; Iqbal N; Syeed S; Khan NA. 2011. Salicylic acid J, ed. Horticultural Reviews 21:177–214. John Wiley & alleviates decreases in photosynthesis under salt stress by Sons Inc., USA. bit.ly/3jB8PQZ enhancing nitrogen and sulfur assimilation and antioxidant Hester MW; Mendelssohn IA; Mckee KL. 2001. Species metabolism differentially in two mungbean cultivars. and population variation to salinity stress in Panicum Journal of Plant Physiology 168(8):807–815. doi: hemitomon, Spartina patens and Spartina alterniflora: 10.1016/j.jplph.2010.11.001 morphological and physiological constraints. Negrão S; Schmöcke SM; Tester M. 2017. Evaluating Environmental and Experimental Botany 46(3):277–297. physiological responses of plants to salinity stress. Annals doi: 10.1016/S0098-8472(01)00100-9 of Botany 119(1):1–11. doi: 10.1093/aob/mcw191 Horie T; Yoshida K; Nakayama H; Yamada K; Oiki S; Parida AK; Das AB. 2005. Salt tolerance and salinity effects Shinmyo A. 2001. Two types of HKT transporters with on plants: a review. Ecotoxicology and Environmental different properties of Na+ and K+ transport in Oryza Safety 60:324–349. doi: 10.1016/j.ecoenv.2004.06.010 sativa. Plant Journal 27(2):129–138. doi: 10.1046/j.1365- Picchioni GA; Miyamoto S; Storey JB. 1990. Salt effects on 313x.2001.01077.x growth and ion uptake of pistachio rootstock seedlings. IGFRI (Indian Grassland and Fodder Research Institute). 2015. American Society for Horticultural Science 115(4):647– Vision 2050. IGFRI, New Delhi, India. bit.ly/3l8QL06 653. doi: 10.21273/JASHS.115.4.647 Pons R; Cornejo MJ; Sanz A. 2011. Differential salinity- Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Salinity stress in Bajra-Napier hybrids 347 induced variations in the activity of H+-pumps and Na+/H+ of Experimental Agriculture 43:357–371. doi: 10.1071/ antiporters that are involved in cytoplasm ion homeostasis EA02081 as a function of genotype and tolerance level in rice cell Singh GB; Bundela DS; Sethi M; Lal K; Kamra SK. 2010. lines. Plant Physiology and Biochemistry 49(12):1399– Remote sensing and geographical information system 1409. doi: 10.1016/j.plaphy.2011.09.011 for appraisal of salt-affected soils in India. Journal Premachandra GS; Saneoka H; Ogata. 1990. Cell membrane of Environment Quality 39(1):5–15. doi: 10.2134/ stability an indicator of drought tolerance as affected jeq2009.0032 by applied nitrogen in soybean. Journal of Agricultural Singh D; Garg AK; Chauhan A. 2018. Fodder yield and quality Science 115(1):63–66. doi: 10.1017/S0021859600073925 assessment of different bajra napier hybrids in central Rahimi E; Nazari F; Javadi T; Saadi S; Silva JAT da. 2021. Gujarat of India. Range Management and Agroforestry Potassium-enriched clinoptilolite zeolite mitigates 39(2):269–273. https://bit.ly/3nrEs1x the adverse impacts of salinity stress in perennial Singh K; Dheeravathu SN; Ramteke PW; Reetu; Dikshit ryegrass (Lolium perenne L.) by increasing silicon N; Vadithe TB. 2020. Effect of salt stress on morpho- absorption and improving the K/Na ratio. Journal of physiological and green fodder yield of Bajra napier Environmental Management 285:112–142. doi: 10.1016/j. Hybrids and TriSpecific Hybrid. Forage Research jenvman.2021.112142 46(3):241–247. bit.ly/3BvyVuI Ren ZH; Gao JP; Li LG; Cai XL; Huang W; Chao DY; Zhu Verma S; Mishra SN. 2005. Putrescine alleviation of growth MZ; Wang ZY; Luan S; Lin HX. 2005. A rice quantitative in salt stressed Brassica juncea, by inducing antioxidative trait locus for salt tolerance encodes a sodium transporter. defense system. Journal of Plant Physiology 162(6):669– Nature Genetics 37(10):1141–1146. doi: 10.1038/ng1643 677. doi: 10.1016/j.jplph.2004.08.008 Roy S; Chakraborty U. 2014. Salt tolerance mechanisms Wang M; Zheng Q; Shen Q; Guo S. 2013. The critical role in Salt Tolerant Grasses (STGs) and their prospects in of potassium in plant stress response. International Journal cereal crop improvement. Botanical Studies 55:31. doi: of Molecular Sciences 14(4):7370–7390. doi: 10.3390/ 10.1186/1999-3110-55-31 ijms14047370 Sairam RK; Dashmukh PS; Shukla DC. 1997. Tolerance of Wang Y; Wu WH. 2013. Potassium transport and signaling in drought and temperature stress in relation to increased higher plants. Annual Review of Plant Biology 64:451– antioxidant enzyme activity in wheat. Agronomy and Crop 476. doi: 10.1146/annurev-arplant-050312-120153 Science 178(3):171–178. doi: 10.1111/j.1439-037X.1997. Weatherley PE. 1950. Studies in water relations of cotton tb00486.x plants. I. The field measurement of water deficit in Sairam RK; Tyagi A. 2004. Physiology and molecular biology leaves. New Phytologist 49:81–97. doi: 10.1111/j.1469- of salinity stress tolerance in plants. Current Science 8137.1950.tb05146.x 86:407–421. jstor.org/stable/24108735 Yang JY; Zheng W; Tian Y; Wu Y; Zhou DW. 2011. Effects Sarwar G; Ashraf MY; Naeem M. 2003. Genetic variability of various mixed salt-alkaline stresses on growth, of some primitive bread wheat varieties to salt tolerance. photosynthesis, and photosynthetic pigment concentrations Pakistan Journal of Botany 35(5):771–777. pakbs.org/ of Medicago ruthenica seedlings. Photosynthetica pjbot/PDFs/35(5)/PJB35(5)11.pdf 49(2):275–284. doi: 10.1007/s11099-011-0037-8 Semple WS; Cole IA; Koen TB. 2003. Performance of Zhu JK. 2003. Regulation of ion homeostasis under salt stress. some perennial grasses on severely salinized sites on the Current Opinion in Plant Biology 6(5):441–445. doi: inland slopes of New South Wales. Australian Journal 10.1016/S1369-5266(03)00085-2 (Received for publication 20 October 2020; accepted 24 August 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):348–358 348 doi: 10.17138/TGFT(9)348-358 Genetic Resources Communication Clearing confusion in Stylosanthes taxonomy. 3. S. hamata sensu stricto vs. S. hamata sensu lato Aclarando confusiones en la taxonomía de Stylosanthes. 3. S. hamata sensu stricto vs. S. hamata sensu lato BRUCE G. COOK1 AND RAINER SCHULTZE-KRAFT2 1Formerly Queensland Department of Agriculture and Fisheries, Brisbane, QLD, Australia. daf.qld.gov.au 2Formerly International Center for Tropical Agriculture (CIAT), Cali, Colombia. ciat.cgiar.org/ Abstract Stylosanthes hamata (L.) Taub., a suffruticose leguminous species with spreading prostrate or ascending stems, is widely distributed in the Caribbean region. It was originally described as Hedysarum hamatum by Linnaeus and later transferred to Stylosanthes by Taubert. To date, chromosome analysis of accessions of S. hamata originating from the Caribbean islands has revealed all to be diploids (2n=20). An accession of a morphologically similar Stylosanthes species, collected near Maracaibo in Venezuela in 1965 and subsequently misidentified as S. hamata, has found application as sown forage on low fertility soils in the subhumid to dry tropics since its registration as cultivar Verano in Australia in 1975. This morphotype has been shown to be tetraploid, and has been referred to in the literature as “tetraploid S. hamata” or “S. hamata sensu lato”. More recent work has demonstrated that the tetraploid is in fact an allotetraploid with S. hamata sensu stricto and S. humilis Kunth as the putative diploid progenitors. Various authors have recommended that the allotetraploid be treated as a separate species. We support this recommendation and suggest that, based on the information provided in this paper, the new species be described and validly published following examination of a more exhaustive range of specimens. Keywords: Cytology, Fabaceae, molecular markers, morphology, phylogeny. Resumen Stylosanthes hamata (L.) Taub. es una leguminosa subarbustiva con tallos postrados a ascendentes. Es ampliamente distribuida en la región del Caribe y fue originalmente descrita por Linnaeus como Hedysarum hamatum y después transferida por Taubert a Stylosanthes. Con base en análisis de cromosomas quedó evidente que todas las accesiones de S. hamata originarias de las islas del Caribe son diploides (2n=20). Una accesión de una especie de Stylosanthes morfológicamente similar, colectada en 1965 cerca de Maracaibo, Venezuela y erróneamente identificada como S. hamata, llegó a ser ampliamente usada como forraje sembrado en suelos de baja fertilidad en regiones tropicales secas a subhúmedas, después de su registro como cultivar Verano en Australia en 1975. Se estableció que este morfotipo es tetraploide y en la literatura se le encuentra denominado “S. hamata tetraploide” o “S. hamata sensu lato”. En un estudio más reciente se demostró que el tetraploide es en realidad un alotetraploide con S. hamata sensu stricto y S. humilis Kunth como supuestos progenitores diploides. Varios autores han recomendado que el alotetraploide sea tratado como una especie separada. Apoyamos esta recomendación y sugerimos que, con base en la información recopilada en este documento, la nueva especie sea descrita y válidamente publicada. Palabras clave: Citología, Fabaceae, filogenética, marcadores moleculares, morfología. Correspondence: B.G. Cook, Brisbane, Queensland, Australia. Email: brucecook@aapt.net.au Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stylosanthes hamata – sensu stricto vs. sensu lato 349 Introduction strictly appropriate. The former is incorrect because there is published scientific evidence showing it to be As with previous papers in this series (Cook and an allotetraploid with S. hamata as one of the putative Schultze-Kraft 2019; Schultze-Kraft et al. 2020), genome donors, and an allopolyploid should not be we draw attention to an issue involving two related assigned to the taxon of one of the putative parents (pers. chromosomal races of Stylosanthes, one diploid and one comm. M. Schori, USDA ARS). The latter term should not tetraploid. In this case the two broad karyotypic groups be used because it is imprecise, implicitly embracing both are mostly referred to under the same species epithet, types. Stace and Cameron (1984) addressed the issue by hamata. The taxonomic confusion arose following the referring to the tetraploid as Stylosanthes sp. nov. (2n=40), decline of large areas of naturalized stands of S. humilis which also lacks specificity. Stace (pers. comm. to R. Kunth in northern Australia in the early 1970s due to Schultze-Kraft, June 1984) suggested the allotetraploid the spread of anthracnose, a serious disease caused by be called “Stylosanthes maracaibensis” in reference to the the fungus, Colletotrichum gloeosporioides. Research geographical origin of the species, or “S. hemihamata” in identified an accession of Stylosanthes originating reference to its alloploid origin. However, neither of the from near Maracaibo in Venezuela as having potential proposed epithets has been validly published as prescribed to replace S. humilis. This accession, catalogued as by the International Code of Nomenclature for algae, S. hamata CPI 38842, was released as cultivar ‘Verano’ fungi, and plants (Turland et al. 2018). We will therefore in Australia (McKay 1975), with a similar accession, refer to the Maracaibo allotetraploid as Stylosanthes sp. CPI 55822, from the same Maracaibo region, released as nov. throughout the remainder of this paper. ‘Amiga’ (Edye 1997), both appearing in the registration The genus Stylosanthes can be divided into two statement as Stylosanthes hamata (L.) Taub., with a taxonomic sections on the basis of presence or absence of tetraploid chromosome complement of 2n=40. Previous a small feathery appendage at the base of the flower and work had shown S. hamata to be diploid, 2n=20 loment, possibly a small rudimentary secondary floral (Cameron 1967). Although plants of ‘Verano’ and axis often referred to as the axis rudiment. Stylosanthes ‘Amiga’ are similar to diploid morphotypes, there is now hamata possesses the axis rudiment and is accordingly strong evidence that the cultivars more correctly belong placed in section Stylosanthes, while the other genome to a new tetraploid species. donor of Stylosanthes sp. nov., S. humilis, which lacks The recommendation to revise S. hamata, providing the axis rudiment, is assigned to section Astyposanthes clear taxonomic distinction between the diploid and (Hert.) Mohl. Stylosanthes sp. nov., which possesses the tetraploid types, is not novel, having already been axis rudiment in the lower flowers only, may best remain raised over a number of years by Stace and Cameron unassigned by virtue of its intersectional origins. (1987), Maass and Sawkins (2004), and Calles and Schultze-Kraft (2016). This paper serves to reiterate Cytology and morphology the urgency for taxonomic revision of S. hamata sensu lato by presenting current cytological evidence The existence of both diploid (Cameron 1967) and supported by morphological, geographical, genetic, tetraploid (Brolmann 1979; Stace and Cameron 1984, and rhizobiological evidence that S. hamata sensu lato 1987) accessions assigned to S. hamata in various genetic actually comprises two distinct species. resource collections has been long recognized. In his pioneer chromosome work, Cameron (1967) showed two Taxonomy Caribbean island accessions, CPI 33205 from Guadeloupe and CPI 33231 from Puerto Rico, to be diploid (2n=20). Stylosanthes hamata was described by Linnaeus (1758) The Jamaican specimen from which the lectotype Sloane as Hedysarum hamatum based on Sloane’s illustration of illustration was prepared was most probably also diploid a specimen from Jamaica and Burman’s illustration of since current understanding suggests that the diploid race a specimen from Sri Lanka. Taubert (1891) provided a only is native in Jamaica and other Caribbean islands. more detailed description in a monograph of Stylosanthes Recognition of a tetraploid race within the northern South and transferred the species to Stylosanthes. American populations of S. hamata was first mentioned The Maracaibo tetraploid has been referred to by Stace and Cameron (1984). During 1986, the collection in the literature as “tetraploid Stylosanthes hamata” of Stylosanthes sp. nov. was substantially increased, or “Stylosanthes hamata sensu lato”, neither being following collecting expeditions, primarily to Venezuela Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 350 B.G. Cook and R. Schultze-Kraft but also to Colombia (Edye 1986), targeting collection of (Edye and Topark-Ngarm 1992) nominates a number of “tetraploid S. hamata”. features that could separate this species from S. hamata S. hamata has been described by a number of authors (L.) Taub., but does not identify the specimens observed since Taubert (1891): Mohlenbrock (1957); Costa in compiling the description. A similar description (2006); Costa et al. (2008); Calles and Schultze-Kraft prepared by Stace (unpublished 1987) was based on (2010); and Vanni (2017). However, none of these the examination of only two specimens of Stylosanthes provides any morphological distinction between diploids sp. nov. from each of Venezuela and Colombia. While and tetraploids, even though several may unknowingly both nominate a number of relative morphological have included tetraploid specimens from within the differences between the two species, the most consistent possible geographic distributional range of Stylosanthes field differences in both descriptions are the presence in sp. nov. Similarly, Burt’s (1983) observation that there Stylosanthes sp. nov. of the axis rudiment in the lower part was considerable variability among the many accessions of the inflorescence only and of a long terminal bristle on held as S. hamata in the Australian tropical forages the tips of the stipules and bracts. We believe that these collection may also be confounded, since by 1983 differences need to be confirmed through examination there were already more than 20 tetraploid accessions of a wider range of identified herbarium specimens for a in the CSIRO collection. It is therefore conceivable that comprehensive description of a new species. some characters or dimensional range extremes may be Table 1 below highlights some of the currently attributable to tetraploid specimens in those descriptions, recognized key differences and similarities among the but any morphological differences were considered to three species, drawing on information from Edye and fall within the species circumscription. Topark-Ngarm (1992) and various sources relating to The only published description of Stylosanthes sp. nov. S. hamata and S. humilis. Table 1. Comparison of key features of Stylosanthes sp. nov. and its putative parent species. Feature S. hamata Stylosanthes sp. nov. S. humilis Ploidy 2n=20 2n=40 2n=20 Life cycle Short-lived perennial Short-lived perennial Obligate annual Stem hairs Line of fine hairs along Line of fine hairs along Line of fine hairs along alternating sides of internodes alternating sides of internodes alternating sides of internodes Stem bristles Absent Absent Abundant Stipule bristles Absent On tips of teeth On sheath and teeth Bract bristles Absent On tips of teeth Abundant Axis rudiment Present In lower flowers only Absent Loment beak Uncinate; beak ≤ upper articulation Uncinate, slightly coiled; beak Uncinate to coiled; longer than other ≥ upper articulation two spp., beak ≥ upper articulation Seed color Mostly cream, yellow to light brown Frequently tan to dark maroon, Mostly brown to black ± mottled Geographic distribution S. hamata. This is the group that has been the target of forage collection expeditions and provided two forage The diploids are geographically more widespread cultivars. It is also the group that has contributed to than the tetraploids, being found from about 28° N in taxonomic confusion. Florida, USA, through much of the Caribbean island • Guatemala population represented in the Australian region to about 8° N in Venezuela, with adventive Pastures Genebank by two accessions, APG 57426 populations between about 3° and 9° S in the north- (=CPI 46587) and APG 57837 (=CPI 46588), eastern Brazilian states of Ceará and Pernambuco occurring around 16° N. (Edye and Maass 1997). • USA population at four separate sites along the Three distinct tetraploid populations have been southeast coast of Florida between about 26° N and identified (Edye and Maass 1997): 27° N; distinguished by short curved beak on the upper • Venezuela-Colombia population occurring between articulation; possibly a separate species; sympatric 9° N and 11°30' N found sympatrically with diploid with the more widespread diploids (Brolmann 1979). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stylosanthes hamata – sensu stricto vs. sensu lato 351 A disjunct population of a species identified as alcohol dehydrogenase (ADH) isozyme analysis, first S. hamata but of undetermined ploidy occurs at about demonstrated that Stylosanthes sp. nov. (2n=40) comprises 21° S in the landlocked state of Mato Grosso do Sul, S. hamata (2n=20) and S. humilis (2n=20) genomes, Brazil (Costa et al. 2008). implying that Stylosanthes sp. nov. is an allotetraploid Reference to tabulated collection data for a large product of the two diploid species. They also noted range of diploid and tetraploid accessions held in the that Stylosanthes sp. nov. should not be confused with Australian forage germplasm collection as S. hamata another taxon known at the time as Stylosanthes sp. aff. (Date 2010), shows that average annual rainfall was hamata that has since been identified as S. scabra Vogel. (300–)500–1,000(–1,600) mm in areas where diploid Curtis et al. (1995), using restriction fragment length accessions were collected and (250–)500–800(–2,200) polymorphism (RFLP) analysis of genomic DNA from mm for tetraploids. Soil pH at collection sites was mostly representative accessions of S. humilis, S. hamata and in the range of (6.2–)6.5–7.5(–8.5) for diploid and (5.4–) Stylosanthes sp. nov. (‘Verano’), presented molecular 6.0–7.5 for tetraploid accessions. Both diploids and evidence that the two diploids are progenitors of ‘Verano’. tetraploids are commonly found at lower elevations, but Gillies and Abbott (1996) undertook detailed analysis of collections of both have been made at elevations >1,000 chloroplast DNA restriction fragment length variation masl. Lists of diploid and tetraploid accessions with the to reconstruct the maternal phylogeny of a range of Australian CPI and CIAT equivalent accession numbers Stylosanthes species. They concluded that S. humilis are shown in the Appendix. Stace and Cameron (1987) is the likely maternal parent of Stylosanthes sp. nov., also included genomic structure along with ploidy in the and S. hamata, by inference from previously published list of CPI accessions held as S. hamata by CSIRO in 1981. findings, the likely paternal progenitor. Further evidence This work revealed that the two Guatemalan tetraploids on the origin and individuality of Stylosanthes sp. nov. is have different genomic structure from that of Stylosanthes provided by Vander Stappen et al. (1999a, 1999b, 2002). sp. nov. that they refer to as the “Maracaibo tetraploid”. Conclusion Rhizobiology This paper presents clear evidence that the tetraploid Date (2010) noted that with few exceptions, S. hamata taxon to which the widely used forage cultivars and Stylosanthes sp. nov. fell into different pairs of ‘Verano’ and ‘Amiga’ belong is not only cytologically groups produced from analysis of extensive accession and to some extent morphologically different from the × Bradyrhizobium effectiveness experiments. The diploid S. hamata sensu stricto, but can conclusively be former showed a high level of specificity in respect to separated from that species on the basis of its phylogeny effectiveness of nodulation by bradyrhizobia, and the as determined from molecular studies. It is clearly not latter showed the typical rhizobial response patterns of an autotetraploid derived solely from S. hamata (L.) promiscuity for tetraploid accessions. In this screening, Taub., but an allotetraploid derived from S. hamata (L.) Bradyrhizobium strains CB2126 and CB3050 were Taub. and S. humilis Kunth. We strongly and respectfully selected as suitable for S. hamata and the wide-spectrum suggest that future authors desist from using taxon names strains CB756 and CB1650 for Stylosanthes sp. nov. such as “tetraploid S. hamata” and “S. hamata sensu CB2126 and/or CB3050 were also effective on many of lato” in reference to the above cultivars and conspecific the more promiscuous Stylosanthes sp. nov. accessions accessions, but in the absence of a validly published (Eagles and Date 1999). name, the allotetraploid be referred to in the first instance as Stylosanthes sp. nov. (Maracaibo allotetraploid). Molecular biology and phylogeny Acknowledgments The dearth of stable morphological characters means that classification of Stylosanthes at the species level is We are extremely grateful to Arsenio Ciprián for his extremely difficult. However, various cytological and help in the arduous task of creating the consolidated molecular-level procedures have facilitated phylogenetic accessions list with CPI and CIAT identifiers, and to Dr analysis that irrefutably separates Stylosanthes sp. nov. Melanie Schori for valuable advice in preparation of this from S. hamata. Stace and Cameron (1984, 1987), using paper. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 352 B.G. Cook and R. Schultze-Kraft References Edye LA; Topark-Ngarm A. 1992. Stylosanthes hamata (L.) Taub. In: Mannetje L't; Jones RM, eds. Plant resources (Note of the editors: All hyperlinks were verified 24 August 2021). of South-East Asia No. 4. Forages. Pudoc Scientific Publishers, Wageningen, the Netherlands. p. 213–216. Brolmann JB. 1979. Distribution and significance of Gillies ACM; Abbott RJ. 1996. Phylogenetic relationships Stylosanthes hamata (L.) Taub. in south Florida. Florida in the genus Stylosanthes (Leguminosae) based upon Scientist 42:63–64. jstor.org/stable/24319551 chloroplast DNA variation. Plant Systematics and Burt RL. 1983. Observation, classification and description. Evolution 200:193–211. doi: 10.1007/BF00984935 In: McIvor JG; Bray RA, eds. Genetic resources of forage Linnaeus C. 1758. Systema naturae per regna tria naturae, plants. CSIRO, Melbourne, Australia. p. 169–181. secundum classes, ordines, genera, species, cum Calles T; Schultze-Kraft R. 2010. Stylosanthes (Leguminosae, characteribus, differentiis, synonymis, locis. 10th Edn. Dalbergieae) of Venezuela. Willdenowia 40:305–329. doi: Laurentius Salvius, Stockholm, Sweden. doi: 10.5962/bhl. 10.3372/wi.40.40211 title.542 Calles T; Schultze-Kraft R. 2016. New species, nomenclatural Maass BL; Sawkins M. 2004. History, relationships and changes and recent taxonomic studies in the genus diversity among Stylosanthes species of commercial Stylosanthes (Leguminosae): An update. Tropical significance. In: Chakraborty S, ed. High-yielding Grasslands-Forrajes Tropicales 4:122–128. doi: 10.17138/ anthracnose-resistant Stylosanthes for agricultural systems. TGFT(4)122-128 Australian Centre for International Agricultural Research, Cameron DF. 1967. Chromosome number and morphology of Canberra, Australia. p. 9–26. bit.ly/3DvaBLl some introduced Stylosanthes species. Australian Journal McKay THE. 1975. Register of Australian herbage plant of Agricultural Research 18(3):375–379. doi: 10.1071/ cultivars. B. Legumes 12. Stylo c. Stylosanthes hamata (L.) AR9670375 Taub. (Caribbean stylo) cv. Verano. (Reg. No. B-12c-1). Cook BG; Schultze-Kraft R. 2019. Clearing confusion in Journal of the Australian Institute of Agricultural Science Stylosanthes taxonomy: 1. S. seabrana B.L. Maass & 41:271–272. 't Mannetje. Tropical Grasslands-Forrajes Tropicales 8:40– Mohlenbrock RH. 1957. A revision of the genus Stylosanthes. 47. doi: 10.17138/TGFT(8)40-47 Annals of the Missouri Botanical Garden 44:299–351. doi: Costa LC; Sartori ALB; Pott A. 2008. Taxonomic study of 10.2307/2394648 Stylosanthes (Leguminosae – Papilionoideae – Dalbergieae) Schultze-Kraft R; Cook BG; Ciprián A. 2020. Clearing in Mato Grosso do Sul, Brazil. Rodriguésia 59:547–572. confusion in Stylosanthes taxonomy. 2. S. macrocephala (In Portuguese.) doi: 10.1590/2175-7860200859310 M.B. Ferreira & Sousa Costa vs. S. capitata Vogel Costa NMS. 2006. Revisão do género Stylosanthes Sw. Ph.D. and S. bracteata Vogel. Tropical Grasslands-Forrajes Thesis. Universidade Técnica de Lisboa, Instituto Superior Tropicales 8:250–262. doi: 10.17138/TGFT(8)250-262 de Agronomia, Lisbon, Portugal. Stace HM; Cameron DF. 1984. Cytogenetics and the evolution Curtis MD; Cameron DF; Manners JM. 1995. Molecular of Stylosanthes. In: Stace HM; Edye LA, eds. The biology evidence that diploid Stylosanthes humilis and diploid and agronomy of Stylosanthes. Academic Press Australia, S. hamata are progenitors of allotetraploid S. hamata cv. North Ryde, NSW, Australia. p. 49–72. doi: 10.1016/ Verano. Genome 38:344–348. doi: 10.1139/g95-044 B978-0-12-661680-4.50008-1 Date RA. 2010. Bradyrhizobium effectiveness responses in Stace HM; Cameron DF. 1987. Cytogenetic review of taxa Stylosanthes hamata and S. seabrana. Tropical Grasslands in Stylosanthes hamata sensu lato. Tropical Grasslands 44:141–157. goo.gl/5rETlX 21:182–188. bit.ly/3mwmHxI Eagles DA; Date RA. 1999. The CB Rhizobium/Bradyrhizobium Taubert PHW. 1891. Monographie der Gattung Stylosanthes. strain collection. CSIRO Tropical Agriculture, Genetic Verhandlungen des Botanischen Vereins Berlin Resources Communication No. 30. bit.ly/3BoyaUf Brandenburg 32(1):1–34. bit.ly/3DvJRKx Edye LA. 1997. Register of Australian herbage plant cultivars. Turland NJ; Wiersema JH; Barrie FR; Greuter W; Hawksworth B. Legumes 12. Stylo (c) Stylosanthes hamata (L.) DL; Herendeen PS; Knapp S; Kusber W-H; Li D-Z; Taub. (Caribbean stylo) cv. Amiga. (Reg. No. B-12c-2). Marhold K; May TW; McNeill J; Monro AM; Prado J; Australian Journal of Experimental Agriculture 37:841– Price MJ; Smith GF, eds. 2018. International Code of 842. doi: 10.1071/EA97079_CU Nomenclature for algae, fungi, and plants (Shenzhen Edye LA. 1986. Stylosanthes collecting in Colombia and Code) adopted by the Nineteenth International Botanical Venezuela, 16.01.‒05.09.1986. Interim Report. AGPG: Congress Shenzhen, China, July 2017. Regnum Vegetabile IBPGR 86 178. International Board for Plant Genetic 159. Koeltz Botanical Books, Glashütten, Germany. Resources, Rome, Italy. hdl.handle.net/10568/81391 Vander Stappen J; De Laet J; Gama-López S; Van Campenhout Edye LA; Maass BL. 1997. The biogeography of Stylosanthes S; Volckaert G. 2002. Phylogenetic analysis of Stylosanthes hamata, S. scabra and “Stylosanthes seabrana”. Tropical (Fabaceae) based on the internal transcribed spacer region Grasslands 31:417–423 bit.ly/38an5JT (ITS) of nuclear ribosomal DNA. Plant Systematics and Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stylosanthes hamata – sensu stricto vs. sensu lato 353 Evolution 234:27–51. doi: 10.1007/s00606-002-0193-1 G. 1999b. Genetic relationships among Stylosanthes Vander Stappen J; Weltjens I; Munaut F; Volckaert G. 1999a. species as revealed by sequence-tagged site markers. Inter-specific and progeny relationships in the genus Theoretical and Applied Genetics 98:1054–1062. Stylosanthes inferred from chloroplast DNA sequence doi: 10.1007/s001220051167 variation. Comptes Rendus de l'Académie des Sciences, Vanni RO. 2017. The genus Stylosanthes (Fabaceae, Paris. III. Sciences de la Vie 322:481–490. doi: 10.1016/ Papilionoideae, Dalbergieae) in South America. Boletín S0764-4469(99)80098-5 de la Sociedad Argentina de Botánica 52:549–585. Vander Stappen J; Weltjens I; Van Campenhout S; Volckaert doi: 10.31055/1851.2372.v52.n3.18033 (Received for publication 21 March 2021; accepted 13 August 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 354 B.G. Cook and R. Schultze-Kraft Appendix: Range of Stylosanthes hamata and related tetraploid species germplasm, as far as ploidy levels have been assessed, with Australian (CPI) and CIAT forage genebank identifiers. A: Diploid accessions [Stylosanthes hamata (L.) Taub.] Source2,3 CPI No. CIAT No. Country1 State/Department 33205 1010 GLP Port Louis 1, 2, 3, 4(*) 33231 12299 PRI San Juan 1, 2, 3, 4(*) 36046 12316 USA Florida 2, 3, 4(*) 37037 12318 DOM 2, 3 37038 12319 DOM 2, 3, 4 38843 12322 USA Florida 2, 3 40264 12339 BRA Pernambuco 2, 3, 4 40268 12332 BRA Ceará 2, 3 40275 12333 BRA Ceará 2, 3 49080 12346 COL Atlántico 2, 3 50997 12356 VEN 2, 3, 4 50998 not reg. VEN 4 56211 not reg. GLP 2, 3 57247 12389 VEN Falcón 2 57248 12390 VEN Falcón 2, 3 61623 12404 VEN Nueva Esparta 2 61624 not reg. VEN Nueva Esparta 2 61669 not reg. VEN 2, 3 61670 12406 VEN 2, 3, 4 61671 12407 VEN 2, 3 61672A 12408 VEN 3 62160 87 COL Atlántico 2 62162 not reg. VEN Nueva Esparta 4 65361 58 COL Atlántico 2 65363 87 COL Atlántico 2, 4 65364 88 COL Atlántico 4 65369 141 COL Atlántico 2 65370 142 COL Atlántico 2 70358 not reg. ATG St. George 2 70359 12426 ATG St. John 2 70360 12427 ATG St. John 2, 4(*) 70361 12428 ATG St. John 2 70362 12429 ATG St. John 2 70363 not reg. LCA 2 70364 12430 ATG St. George 2 70365 12431 ATG St. Phillip 2 70366 12432 ATG St. George 2, 4(*) 70367 12433 ATG St. George 2 70368 not reg. ATG St. George 2 70369 12434 CUB Matanzas 2 70370 12435 ATG St. George 2, 4(*) 70371 12436 ATG 2, 4(*) 70372 not reg. ATG 2, 4(*) 70373 not reg. ATG 2 70374 12437 ATG 2, 4(*) 70375 not reg. ATG 2 70376 12438 ATG St. George 2 70377 not reg. ATG 2 70520 12439 USA Florida 2 70523 12440 USA Florida 4(*) Continued Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stylosanthes hamata – sensu stricto vs. sensu lato 355 A: Diploid accessions [Stylosanthes hamata (L.) Taub.] Source2,3 CPI No. CIAT No. Country1 State/Department 70524 12441 USA Florida 4 70525 12442 USA Florida 2, 4 70526 12443 USA Florida 2 70527 12444 USA Florida 2 72850 12446 USA Florida 4 72852 12447 USA Florida 4 72854 12448 USA Florida 4 72959 12449 USA Florida 2, 4 73484 12450 ATG St. George 4(*) 73485 12451 ATG St. George 2 73486 12452 ATG St. George 4(*) 73487 12453 ATG St. George 4(*) 73488 12454 ATG St. John 4(*) 73490 not reg. ATG St. Phillip 2 73491 not reg. ATG St. George 4(*) 73497 not reg. KNA St. Kitts 4(*) 73498 not reg. ATG St. Paul 4(*) 73499 not reg. ATG St. Paul 4(*) 73501 12457 ATG St. George 4(*) 73505 not reg. ATG 4(*) 73506 not reg. ANT Curaçao 4(*) 73507 not reg. ANT Curaçao 4(*) 73509 not reg. ANT Curaçao 4(*) 73511 1475 CUB Matanzas 4 73513 12459 KNA Nevis 4(*) 73514 12460 KNA Nevis 4(*) 73515 12461 KNA Nevis 4(*) 73517 1465 KNA Nevis 4(*) 73519 1466 KNA Nevis 4(*) 73523 12462 ANT Curaçao 4(*) 82313 not reg. CUB Santiago de Cuba 2 94130 12666 USA 4 94443 12674 USA 4 99670 12680 USA Florida 4 99675 12685 PRI Corozal 4 105678 not reg. BRA Bahia 4(*) 109305 11194 COL Atlántico 4 109307 11196 COL Atlántico 4 109308 11197 COL Atlántico 4 109310 11199 COL Atlántico 4 109312 11201 COL Atlántico 4 109314 11203 COL Atlántico 4 109315 11204 COL Atlántico 4 109316 11205 COL Magdalena 4 109346 11237 COL Guajira 4 110066 12534 VEN Zulia 4 110067 12535 VEN Zulia 4 110077 12539 VEN Falcón 4 110083 12542 VEN Falcón 4 110084 12543 VEN Falcón 4 110087 12544 VEN Falcón 4 110090 12547 VEN Falcón 4 Continued Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 356 B.G. Cook and R. Schultze-Kraft A: Diploid accessions [Stylosanthes hamata (L.) Taub.] Source2,3 CPI No. CIAT No. Country1 State/Department 110099 12553 VEN Lara 4 110108 12558 VEN Lara 4 110110 11779 VEN Cojedes 4 110114 11781 VEN Nueva Esparta 4 110119 not reg. VEN Sucre 4 110125 not reg. VEN Aragua 4 110171 not reg. VEN Lara 4 110173 12586 VEN Lara 4 110174 11793 VEN Lara 4 110176 12587 VEN Lara 4 110179 12588 VEN Trujillo 4 110181 11795 VEN Trujillo 4 110185 11796 VEN Lara 4 110186 12590 VEN Lara 4 110190 12593 VEN Mérida 4 110207 not reg. VEN Distrito Capital 4 110311 124 COL Atlántico 4 B: Tetraploid accessions [Stylosanthes sp. nov. (Maracaibo allotetraploid)] Source2,3 CPI No. CIAT No. Country1 State/Department 38842 1 VEN Zulia 2, 3 38842 1953 VEN Zulia 2 55812 12371 VEN Zulia 2, 3 55820 12372 VEN Zulia 2, 3 55821 12373 VEN Zulia 2, 3, 4 55822 12374 VEN Zulia 2, 3, 4 55823 12375 VEN Zulia 2, 3 55824 12376 VEN Zulia 2, 3 55825 12377 VEN Zulia 2 55826 12378 VEN Zulia 2, 3, 4 55827 12379 VEN Zulia 2, 3 55828 12380 VEN Zulia 2, 3 55830 12381 VEN Zulia 2, 3 55831 12382 VEN Zulia 2, 3 61672B 12408 VEN 2, 3 61672BB 12408 VEN 4 65365 114 VEN Zulia 2, 4 65367 120 VEN Zulia 2 65368 122 VEN Zulia 2, 4 65371 147 VEN Guárico 2, 4 65962 12412 COL Magdalena 4 65965 12415 COL Magdalena 2, 4 68837 167 COL Guajira 2, 4 68838 174 COL Magdalena 2, 4 68840 1039 COL Magdalena 2, 4 109320 11209 COL Magdalena 4 109325 11214 COL Magdalena 4 109326 11215 COL Magdalena 4 109331 11221 COL Magdalena 4 109332 11222 COL Magdalena 4 109344 11235 COL Guajira 4 109347 11238 COL Guajira 4 Continued Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Stylosanthes hamata – sensu stricto vs. sensu lato 357 B: Tetraploid accessions [Stylosanthes sp. nov. (Maracaibo allotetraploid)] Source2,3 CPI No. CIAT No. Country1 State/Department 109349 11240 COL Cesar 4 109350 11241 COL Cesar 4 110024 12409 VEN Zulia 4 110025 12509 VEN Zulia 4 110026 12510 VEN Zulia 4 110027 12511 VEN Zulia 4 110028 12512 VEN Zulia 4 110029 12513 VEN Zulia 4 110030 12514 VEN Zulia 4 110033 12515 VEN Zulia 4 110035 12516 VEN Zulia 4 110036 11761 VEN Zulia 4 110037 11762 VEN Zulia 4 110038 12517 VEN Zulia 4 110039 12518 VEN Zulia 4 110040 12519 VEN Zulia 4 110041 12520 VEN Zulia 4 110042 12521 VEN Zulia 4 110043 11763 VEN Zulia 4 110044 12522 VEN Zulia 4 110045 12523 VEN Zulia 4 110046 11764 VEN Zulia 4 110048 12525 VEN Zulia 4 110049 12526 VEN Zulia 4 110050 11765 VEN Zulia 4 110051 11766 VEN Lara 4 110057 12529 VEN Zulia 4 110068 12536 VEN Zulia 4 110069 12537 VEN Zulia 4 110070 11770 VEN Zulia 4 110095 11778 VEN Lara 4 110098 12552 VEN Lara 4 110104 12555 VEN Trujillo 4 110109 12559 VEN Lara 4 110116 11782 VEN Táchira 4 110134 12568 VEN Zulia 4 110135 12569 VEN Zulia 4 110138 11787 VEN Zulia 4 110162 12580 VEN Zulia 4 110166 12582 VEN Aragua 4 110168 12584 VEN Yaracuy 4 110205 12596 VEN Miranda 4 110206 12597 VEN Distrito Capital 4 110209 12598 VEN Aragua 4 110316 179 COL Magdalena 4 110317 182 VEN Zulia 4 Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 358 B.G. Cook and R. Schultze-Kraft C: Tetraploid accessions (Guatemala and Florida) Source2,3 CPI No. CIAT No. Country1 State/Department 46587 12343 GTM Alta Verapaz 2, 3, 4 46588 12344 GTM 2, 3 94444 12675 USA Florida 4 1Country abbreviations: ANT = Netherlands Antilles; ATG = Antigua and Barbuda; BRA = Brazil; BHS = Bahamas; CUB = Cuba; COL = Colombia; DOM = Dominican Republic; GLP = Guadeloupe; GTM = Guatemala; LCA = Saint Lucia; KNA = Saint Kitts and Nevis; PRI = Puerto Rico; USA = United States of America; VEN = Venezuela. 2Sources: 1 - Cameron (1967); 2 - H. Stace pers. comm. (1984); 3 - Stace and Cameron (1987); 4 - Date (2010). 3Source with (*): accession is mentioned as “presumed diploid”. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):359–370 359 doi: 10.17138/TGFT(9)359-370 Genetic Resources Communication Genetic diversity and population structure of Heteropogon contortus L. germplasm collected from diverse agro-climatic regions in India and development of a core germplasm set Diversidad genética y estructura poblacional de germoplasma de Heteropogon contortus L. provenientes de regiones agro-climáticas diversas en India y desarrollo de una colección núcleo AJOY KUMAR ROY1, DEVENDRA RAM MALAVIYA1,2, PANKAJ KAUSHAL1,3, SANAT KUMAR MAHANTA1, RUPALI TEWARI1, ROOPALI CHAUHAN1 AND AMARESH CHANDRA1,2 1ICAR-Indian Grassland and Fodder Research Institute, Jhansi, India. igfri.res.in 2Present Address: ICAR - Indian Institute of Sugarcane Research, Lucknow, India. 3Present Address: ICAR - National Institute of Biotic Stress Management, Raipur, India. Abstract Heteropogon contortus, an important constituent of major grasslands of India, Australia and many countries in Africa, Asia and the Americas, is important for pasture and grassland productivity. Hence genetic improvement of the grass needs attention. A genetic variability study, including development of a core subset, was carried out by evaluating 235 accessions collected from different agro-ecological zones of India. The study, based on 16 metric and 14 non-metric traits along with 8 nutritional parameters, indicated that considerable genetic variability existed among the germplasm and selection could result in identification of suitable types for target environments. Clustering and subclustering was performed to select 35 accessions to form a core subset. The statistical analysis indicated that the core subset captured almost all the variability present in the entire germplasm. The study will help researchers to focus future studies on this core subset in developing genetic improvement programs. Keywords: Black spear grass, forage grass, grassland, plant genetic resources, rangeland. Resumen Heteropogon contortus es una pastura altamente utilizada en la India, Australia y muchos países de África, Asia y América; muy importante para la productividad forrajera. Por tanto, es necesario prestar atención al mejoramiento genético de la hierba. Se llevó a cabo un estudio de variabilidad genética, incluido el desarrollo de un subconjunto central, mediante la evaluación de 235 accesiones recolectadas de diferentes zonas agroecológicas de la India. El estudio, basado en 16 rasgos métricos y 14 no métricos junto con 8 parámetros nutricionales, indicó que existía una variabilidad genética considerable entre el germoplasma y que la selección podría resultar en la identificación de tipos adecuados para los ambientes objetivo. Se realizaron agrupaciones y subgrupos para seleccionar 35 accesiones para formar un subconjunto central. El análisis estadístico indicó que el subconjunto central capturó casi toda la variabilidad presente en todo el germoplasma. El estudio ayudará a los investigadores a centrar los estudios futuros en este subconjunto central en el desarrollo de programas de mejora genética. Palabras clave: Barba negra, gramínea forrajera, pasturas, recursos genéticos, variabilidad. Correspondence: Devendra Ram Malaviya, ICAR - Indian Institute of Sugarcane Research, Lucknow, India. Email: drmalaviya47@rediffmail.com Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 360 A.K. Roy, D.R. Malaviya, P. Kaushal, S.K. Mahanta, R. Tewari, R. Chauhan and A. Chandra Introduction and often dominates the under-storey of woodlands in tropical and subtropical Australia. Heteropogon contortus (L.) P. Beauv. ex Roem. & Schult. The grass is a nutritious fodder which can be easily (Poaceae: Andropogoneae) is an important perennial conserved as hay (Bor 1960). It is highly palatable during pasture grass. It is commonly known as black spear grass the vegetative phase. However, its robust awns shed easily or bunch spear grass (Australia), tangle head (United and are considered a negative factor since at maturity States), pilli grass (Hawaii), assegai grass (Zimbabwe) they are pointed sharp and intermingle, causing injury and Lampa ghas (India). The grass is native to Africa, in the mouth and stomach of grazing animals. It spreads southern Asia, northern Australia and parts of Oceania naturally and is considered an aggressive invasive species and is naturalized in tropical and subtropical regions of the because its seeds survive even after burning of rangelands/ Americas, East Asia and Oceania. Commonly found in grasslands by burying themselves (Roy et al. 2019a). tropical Africa (Soromessa 2011), it is also very common In India, livestock production is primarily based in other tropical, subtropical and warm temperate regions on rangeland and grassland grazing. About 40% of the of the world, particularly in the Indian subcontinent, country is available for grazing of livestock as pasture Burma, North Africa, Australia and the Pacific as a lands, forest lands, cultivated wastelands, fallow lands, perennial range grass. The grass is frequently present in non-agricultural lands, miscellaneous tree crops and major grasslands of India such as Sehima-Dichanthium, groves etc. In such grazing lands in tropical, subtropical, Dichanthium-Cenchrus-Lasiurus, Themeda-Arundinella arid, semi-arid and lower hills of India and isoclimatic and Peninsular India grasslands (Dabadghao and conditions in the world, H. contortus is widely adapted. Shankarnarayan 1973; Malaviya et al. 2018; Malaviya However, to increase pasture productivity it is essential and Roy 2021) and establishes well around tropical and to introduce high-yielding accessions in such grazing subtropical grasslands of the world (Carino 1999). lands, especially in dry areas. There is a deficit in H. contortus is an important shrub-layer grass of moist demand and supply of green and dry forages for livestock deciduous forests (Rhind 2010) dominated by Shorea in India (Roy et al. 2019b) and perennial grasses like robusta in the upper Gangetic plains (Singh 2012). Heteropogon are important in bridging the gap. Historically, Hawaiians used fire to manage grasslands At an international level, genetic diversity of the grass dominated by H. contortus (Hoffmann 2008; Daehler and is conserved in germplasm collections at the Southern Goergen 2005), although Goergen and Daehler (2001) Regional Plant Introduction Station, Griffin, Georgia of reported the grass to be only lowly tolerant of clipping and the USDA/ARS National Genetic Resources Program, burning. It has also been found to be an effective live mulch and in Indonesia, Ethiopia and Zimbabwe; however, in tropical agricultural systems (Moata 2009). H. contortus, there is negligible representation of germplasm from the the dominant grass in northern Australia savanna, is a valued Indian subcontinent. ICAR-Indian Grassland and Fodder species for cattle, but noxious for sheep because of its awns Research Institute, Jhansi, India (ICAR-IGFRI) has and a weed in sown grass-legume pastures (FAO 2007). collected a wide diversity of this grass through a series of H. contortus prefers areas with rainfall less than explorations from various agro-climatic conditions in the 800 mm, is adapted to coarse-textured soils with pH country. The germplasm is maintained at ICAR-IGFRI 6‒8 and has moderate to low tolerance of drought, high and ICAR-National Bureau of Plant Genetic Resources, temperatures, soil salinity and low soil fertility (Moata New Delhi, India. Knowledge about the existing genetic 2009). Fang and Xiong (2015) and Wang et al. (2016) variability is a prerequisite for any genetic improvement indicated that, although the grass is drought-tolerant, program, and studies of available natural variation are severe water stress inhibits its growth. Due to its tolerance helpful in developing a genetic improvement strategy. of soil nutrient deficiencies and limited soil moisture, it Hence, the present study was undertaken to critically is a valuable species for vegetation restoration (Goergen evaluate available germplasm, collected from diverse parts and Daehler 2001; 2002). Owing to its high nutritional of India, to enable breeders to utilize the genetic divergence and medicinal value, Daehler and Goergen (2005) for isolating promising types. Additionally, development considered ethnobotanical and ecological research was of a core germplasm subset was envisaged, to allow important for restoration of H. contortus-dominated researchers to focus on a limited range of germplasm for grasslands. H. contortus is tolerant of limited shading further studies and also to accelerate breeding programs. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Genetic diversity and core subset in Heteropogon contortus 361 Material and Methods accessions in terms of: ash (%); organic matter (%); neutral detergent fiber (NDF) (%); acid detergent fiber (ADF) (%); A study of variability was conducted on 235 accessions hemicellulose (%); cellulose (%); lignin (%); and crude of H. contortus collected from different agro-ecological protein concentrations (%). AOAC (1980) methods were zones of India and conserved at ICAR-IGFRI, Jhansi, followed for estimation of dry matter, crude protein and India. A list of accessions along with their place of ash, whereas the Goering and Van Soest (1970) method collection is given in Table 1. The soil and climatic was followed for NDF and ADF. In vitro dry matter conditions of the place of collection following agro- digestibility (IVDMD) % was estimated following Tilley ecological zones defined by Ahmad et al. (2017) and and Terry (1963) for 81 accessions. For these analyses, ICAR (2018), are presented in Table 2. samples were collected on different dates, so as to match a Seeds of the accessions were taken from the gene bank uniform 50% flowering stage (i.e. 50% of the inflorescence of IGFRI and planted in a nursery. Six-week-old seedlings in the row having spikelets at the anthesis stage). The were transplanted in 3 m-long paired rows, spaced at 75 whole plant was harvested from 10 cm above ground and cm, accommodating 6 tussocks in each row. Rows of the samples were oven dried at 60 °C and ground. plants were 1 m apart. Data on morphological traits were recorded at 50% flowering stage on 3 randomly-selected Statistical analysis plants, excluding border plants at the ends of the rows. Seed-related observations were recorded at maturity. The metric traits data (excluding ligule length) were Data on 16 quantitative metric traits were recorded, averaged over replications and analyzed statistically i.e. length of main tiller (cm); number of tillers/tussock; using Non–Hierarchical Euclidian Cluster Analysis of fresh weight (g) of single tussock (average of 3) harvested grouping of accessions (Sparks 1973). The computer at 50% flowering; main tiller diameter (cm); internodal software ‘Statistical Tool for Agricultural Research length (cm); number of nodes on main tiller; leaf blade (STAR)’ (IRRI 2020) was used for computation. length (cm); leaf blade width (cm); leaf sheath length The accessions per cluster for the core subset were (cm); leaf sheath width (cm); flag leaf blade length (cm); decided upon following Brown (1989a). The number of flag leaf blade width (cm); flag leaf sheath length (cm); accessions for a core subset was kept at 15% of the total flag leaf sheath width (cm); ligule length (mm); and germplasm following Brown (1989a). Accessions from inflorescence length (cm). Length of the main tiller was each cluster were selected as per the formula below, measured from ground-level to the tip of the last emerged following Roy et al. (2020) and the logarithmic strategy leaf. The fourth leaf from the top was considered for suggested by Brown (1989b). Accessions for the core measuring leaf width and length. Internodal length was subset from each cluster were randomly selected. measured between 3rd and 4th nodes. s = (log pi /mƩt=1 log p) × n Data were also recorded for 14 non-numeric traits, i.e. where: growth (overall vegetative growth by visual observation s = number of accessions selected in a character; as very poor, poor, medium, good, very good), habit p = size of cluster; (prostrate, decumbent, erect), leaf hairiness (glabrous, pi = proportion of ith cluster; light hairy, hairy), leaf blade color (light green, green, n = number of accessions to be selected for core (15% bluish green, greenish blue, greenish violet), leaf of base collection); and sheath color (light green, green, greenish violet), tiller m = total number of clusters. internode color (light green, greenish violet, light violet, For larger clusters, a second level of clustering was violet), node anthocyanin coloration (light violet, violet, done for further assortment of accessions, to constitute dark violet), node color (green, greenish violet, violet, the core subset. dark violet), node hairiness (light hairy, non-hairy), Correlation among various quantitative traits was studied anthocyanin coloration on leaf sheath (absent, weak, using Microsoft Excel program (MS-Excel). Key characters strong), anthocyanin coloration on leaf blade (absent, contributing to diversity were identified using Principal weak, medium, strong), spike color (light green, green, Component Analysis. Scree plot analysis was performed greenish violet), awn stature (hard, soft) and awn to determine the number of principal components to be pubescence distribution (pubescence up to 50% awn retained as per Cattell (1966). The mean value for different length, pubescence on full awn length i.e. 100%). traits, as obtained in the core value, was compared with that Nutritional quality parameters were analyzed for 167 of total germplasm by student’s t-test using MS-Excel. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 362 A.K. Roy, D.R. Malaviya, P. Kaushal, S.K. Mahanta, R. Tewari, R. Chauhan and A. Chandra Table 1. List of accessions of Heteropogon contortus and their place of collection (state in India). SN Accession State SN Accession State SN Accession State SN Accession State 1 Bundel var-1 MP 60 IG02-635 Cg 119 IG95-13 UP 178 IG97-167 MP 2 Hc-13 UP 61 IG02-636 Cg 120 IG95-15 UP 179 IG97-181 MP 3 Hc-15 UP 62 IG02-637 Cg 121 IG95-17 UP 180 IG97-182 MP 4 Hc-17 UP 63 IG02-639 Cg 122 IG95-21 UP 181 IG97-183 MP 5 Hc-18 UP 64 IG02-640 Cg 123 IG95-23 UP 182 IG97-183 MP 6 Hc-23 UP 65 IG02-641 Mh 124 IG95-242 MP 183 IG97-183 MP 7 Hc-5/18 UP 66 IG02-641 Mh 125 IG95-242 MP 184 IG97-209 MP 8 IG01-513 MP 67 IG02-642 Mh 126 IG95-25 UP 185 IG97-209 MP 9 IG01-514 MP 68 IG02-643 Mh 127 IG95-258 MP 186 IG97-224 MP 10 IG01-516 MP 69 IG02-644 Mh 128 IG95-26 UP 187 IG97-253 MP 11 IG01-517 MP 70 IG02-645 Mh 129 IG95-270 MP 188 IG99-210 Rj 12 IG01-519 MP 71 IG02-646 Mh 130 IG95-271 MP 189 IG99-210 Rj 13 IG01-520 MP 72 IG02-647 Mh 131 IG95-274 MP 190 IG99-219 Rj 14 IG01-520 MP 73 IG02-649 Mh 132 IG95-277 MP 191 IG99-311 Rj 15 IG01-522 MP 74 IG02-650 Mh 133 IG95-279 MP 192 IG99-312 Rj 16 IG01-522 MP 75 IG02-651 Mh 134 IG95-280 MP 193 IG99-313 Rj 17 IG02-129 UP 76 IG02-652 Mh 135 IG95-283 MP 194 IG99-314 Rj 18 IG02-185 UP 77 IG02-653 Mh 136 IG95-284 MP 195 IG99-314 Rj 19 IG02-190 UP 78 IG02-655 Mh 137 IG95-284 MP 196 IG99-315 Rj 20 IG02-191 UP 79 IG02-655 Mh 138 IG95-286 MP 197 IG99-316 Rj 21 IG02-195 UP 80 IG02-657 Mh 139 IG95-287 MP 198 IG99-317 Rj 22 IG02-204 UP 81 IG02-657 Mh 140 IG95-289 MP 199 IG99-318 Rj 23 IG02-205 UP 82 IG02-658 MP 141 IG95-290 MP 200 IG99-319 Rj 24 IG02-209 UP 83 IG02-658A MP 142 IG95-292 MP 201 IG99-319 Rj 25 IG02-291 HP 84 IG02-659 MP 143 IG95-293 MP 202 IG99-320A Rj 26 IG02-293 HP 85 IG02-660 MP 144 IG95-328 MP 203 IG99-321 Rj 27 IG02-342 MP 86 IG02-661 MP 145 IG95-340 MP 204 IG99-322 Rj 28 IG02-343 MP 87 IG02-663 MP 146 IG95-341 UP 205 IG99-323 Rj 29 IG02-344 MP 88 IG02-665 MP 147 IG95-341 UP 206 IG99-325 Rj 30 IG02-345 MP 89 IG02-666 MP 148 IG95-343 UP 207 IG99-326 Rj 31 IG02-346 MP 90 IG02-668 MP 149 IG95-344 UP 208 IG99-327 Rj 32 IG02-347 MP 91 IG02-670 MP 150 IG95-344 UP 209 IG99-329 Rj 33 IG02-348 MP 92 IG02-671 MP 151 IG95-345 UP 210 IG99-330 Rj 34 IG02-349 MP 93 IG02-679 MP 152 IG95-346 UP 211 IG99-333 Rj 35 IG02-350 MP 94 IG03-361 TN 153 IG95-346 UP 212 IG99-335 Rj 36 IG02-351 MP 95 IG03-371 Kl 154 IG95-347 UP 213 IG99-336 Rj 37 IG02-352 MP 96 IG03-371 Kl 155 IG95-348 UP 214 IG99-337 Rj 38 IG02-353 MP 97 IG03-376 Kl 156 IG95-349 UP 215 IG99-338 Rj 39 IG02-354 MP 98 IG03-377 Kl 157 IG95-350 UP 216 IG99-338 Rj 40 IG02-355 MP 99 IG2000-101 UP 158 IG95-352 UP 217 IG99-345 MP 41 IG02-356 MP 100 IG2000-73 UP 159 IG95-352 UP 218 IG99-346 MP 42 IG02-357 UP 101 IG2000-73A UP 160 IG95-363 UP 219 IG99-349 MP 43 IG02-358 UP 102 IG2000-74 MP 161 IG95-366 UP 220 IG99-50 Cg 44 IG02-359 UP 103 IG2000-93 MP 162 IG95-367 UP 221 IG99-51 Cg 45 IG02-362 UP 104 IG2000-98 UP 163 IG95-368 UP 222 IGO2-184 UP 46 IG02-363 J&K 105 IG95-101 MP 164 IG95-369 UP 223 IGO2-186 UP 47 IG02-364 J&K 106 IG95-103 MP 165 IG95-369 UP 224 IGO2-187 UP 48 IG02-371 MP 107 IG95-104 MP 166 IG95-371 UP 225 IGO2-188 UP 49 IG02-375 MP 108 IG95-104-1 MP 167 IG95-374 UP 226 IGO2-189 UP 50 IG02-487 MP 109 IG95-104-2 MP 168 IG95-374 UP 227 IGO2-191 UP 51 IG02-624 MP 110 IG95-104A MP 169 IG95-7 UP 228 IGO2-192 UP 52 IG02-625 MP 111 IG95-104B MP 170 IG95-99 MP 229 IGO2-193 UP 53 IG02-626 Mh 112 IG95-105 MP 171 IG96-164 Kt 230 IGO2-200 UP 54 IG02-627 Mh 113 IG95-105A MP 172 IG96-167 Kt 231 IGO2-201 UP 55 IG02-629 Mh 114 IG95-106 MP 173 IG96-21 TN 232 IGO2-281 HP 56 IG02-630 Mh 115 IG95-108 MP 174 IG96-97 TN 233 IGO2-281 A HP 57 IG02-631 Mh 116 IG95-109 MP 175 IG97-163 MP 234 IGO2-288 HP 58 IG02-632 Mh 117 IG95-110 MP 176 IG97-165 MP 235 IGO2-294 HP 59 IG02-633 Mh 118 IG95-111 MP 177 IG97-166 MP Cg = Chhattisgarh; HP = Himachal Pradesh; J&K = Jammu & Kashmir; Kl = Kerala; Kt = Karnataka; MP = Madhya Pradesh; Mh = Maharashtra; Rj = Rajasthan; TN = Tamil Nadu; UP = Uttar Pradesh. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Genetic diversity and core subset in Heteropogon contortus 363 Table 2. Soil and climate conditions in the Indian states where accessions of Heteropogon contortus were collected [Source: Ahmad et al. (2017); ICAR (2018)]. SN State Climate Precipitation PET (mm) CGP Soil (mm/yr) 1 Rajasthan Hot arid/semi-arid <300‒800 1,500‒2,000 <90 Saline, alluvium-derived with desert 2 Maharashtra Hot semi-arid to hot 600‒1,000 1,600-1,800 90‒180 Shallow and medium (with inclusion of subhumid deep) black; black & red 3 Madhya Pradesh Hot semi-arid to hot 500‒1,000 1,400‒2,000 90‒180 Alluvium-derived soils or medium and subhumid deep black soils 4 Karnataka Hot subhumid to 600‒1,000 1,300‒1,600 90‒150 Coastal alluvium-derived soils;shallow semi-arid and medium (with inclusion of deep) black, red loamy 5 Himachal Pradesh Warm subhumid to 1,600‒2,000 800‒1,300 180‒210+ Brown forest and podzolic soils humid, also with some perhumid zone 6 Jammu & Kashmir Warm subhumid to 1,600‒2,000 800‒1,300 180‒210+ Brown forest and podzolic soils humid also with some perhumid zones 7 Chhattisgarh Hot subhumid 1,200‒1,600 1,400‒1,500 150‒180 Red and yellow soils 8 Tamil Nadu Hot semi-arid 600-1,000 1.300‒1,600 90‒150 Red loamy soils 9 Uttar Pradesh Hot subhumid 1,000‒1,500 1,300‒1,500 90‒180 Red and black soils, alluvium-derived 10 Kerala Hot humid, perhumid 2,000‒3,200 1,400‒1,600 90‒210+ Red, lateritic and alluvium-derived soils PET = Potential Evapotranspiration; CGP = Crop growing period (no. of days) Results temperate environments, i.e. Himachal Pradesh and Jammu and Kashmir, grouped in Cluster 3; however, Observations were recorded on 235 accessions of this cluster included also accessions from the hot arid H. contortus for 16 numeric and 14 non-numeric climates of central, western and northern India. morphological traits and on 167 accessions for 9 Principal Component Analysis revealed that the first nutritional quality parameters in order to characterize the 6 principal components accounted for more than 80% of germplasm being maintained at the IGFRI gene bank, and the cumulative variability (Table 4). Scree plot analysis to further develop a core subset of germplasm. Clustering revealed that the first 2 principal coordinates or up to of accessions using 15 metric traits (excluding ligule 6 principal components can be retained for explaining length) resulted in formation of 6 distinct clusters (Figure most of the diversity (Figure 2). 1, Table 3). Clusters 3 and 4 were big clusters consisting Height of the plants ranged from 36 to 110 cm (mean 74 of 101 and 94 accessions, respectively, whereas Cluster 1 cm) (Table 4). Accessions with poor tillering and shorter was of moderate size comprising 30 accessions. Clusters plant heights were generally annual types. Robust 2, 5 and 6 were small clusters, comprising 3, 5 and 2 accessions possessed as many as 265 tillers, whereas the accessions, respectively. In Cluster 6, the 2 accessions minimum tiller number per tussock was 9 only. These were vigorous with high values for fresh weight, tiller 2 traits contribute significantly to total biomass, which height, tiller diameter, leaf length and leaf width. These was well reflected in fresh biomass, which ranged from two accessions originated from north and central India, 20 to 435 g. The highest fresh biomass per tussock representing tropical climate conditions. The accessions was noted for Cluster 6 (263 g) with 2 accessions only. of Cluster 2 also showed high values for agronomic The second highest average fresh biomass per tussock traits. Other clusters showed moderate values for various occurred in accessions of Cluster 3 (143 g), followed by agronomic traits. Collections from the State of Uttar that of Cluster 4 (125 g). Morphological traits, which Pradesh were from the semi-arid districts near Jhansi and showed highly significant (P<0.01) positive correlation the majority of these 63 accessions grouped in Clusters with fresh biomass, were plant height (0.299), number 3 and 4, with a few in Clusters 1, 5 and 6. Similarly, 29 of tillers (0.758) and leaf length (0.289). Tiller diameter accessions from the arid climate of Rajasthan (western with 0.130 and leaf width with 0.140 correlation Indian state) grouped mostly in Clusters 3 and 4. Barring coefficients were also positively significant (P<0.05). 2 accessions, the remaining accessions were from However, internodal length was not correlated with fresh Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 364 A.K. Roy, D.R. Malaviya, P. Kaushal, S.K. Mahanta, R. Tewari, R. Chauhan and A. Chandra 0 5 10 15 H166 Clusters H218 H122 H17 H124 5 H53 H214 H46 H90 H192 H197 H136 H1 H152 H50 H202 H162 H13 H176 H109 H153 H190 1 H115 H30 H201 H174 H185 H40 H103 H24 H8 H182 H142 H5 H19 H73 6 H123 H231 H2 H74 H23 2 H98 H155 H54 H71 H134 H55 H114 H220 H48 H79 H221 H60 H113 H167 H188 H125 H130 H154 H144 H145 H117 H129 H81 H224 H43 H110 H121 H223 H226 H227 H86 H106 H143 H157 H158 H59 H141 H146 H139 H169 H184 H160 H149 H171 H133 H57 4 H66 H68 H7 H208 H211 H191 H199 H200 H196 H204 H101 H230 H21 H219 H44 H65 H159 H88 H203 H229 H28 H76 H64 H107 H83 H228 H18 H212 H72 H189 H120 H45 H70 H42 H175 H58 H177 H131 H127 H205 H170 H100 H26 H34 H156 H20 H118 H10 H168 H222 H235 H225 H80 H16 H102 H35 H51 H206 H29 H75 H82 H97 H180 H111 H161 H63 H112 H39 H165 H38 H232 H95 H179 H78 H137 H77 H186 H25 H234 H87 H172 H91 H195 H138 H140 H89 H107 H41 H198 H96 H15 H9 H151 H99 H173 H49 H92 H207 H135 H213 H3 H11 H94 3 H61 H126 H128 H85 H217 H93 H108 H233 H52 H183 H22 H215 H47 H132 H194 H36 H37 H181 H187 H33 H67 H12 H84 H105 H6 H193 H116 H209 H150 H163 H31 H148 H56 H164 H178 H4 H62 H69 H32 H147 H119 H14 H210 H27 H216 Figure 1. Clustering of Heteropogon contortus accessions (dendrogram using agglomerative clustering method). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Genetic diversity and core subset in Heteropogon contortus 365 Table 3. Accessions of Heteropogon contortus belonging to different clusters. Cluster Accessions (as per SN in Table 1) 1 1, 5, 8, 13, 19, 24, 30, 40, 46, 50, 53, 90, 103, 109, 115, 136, 142, 152, 153, 162, 174, 176, 182, 185, 190, 192, 197, 201, 202, 214 2 2, 74, 231 3 3, 4, 6, 9, 10, 11, 12, 14, 15, 16, 22, 25, 27, 29, 31, 32, 33, 35, 36, 37, 38, 39, 41, 47, 49, 51, 52, 56, 61, 62, 63, 67, 69, 75, 77, 78, 80, 82, 84, 85, 87, 89, 91, 92, 93, 94, 95, 96, 97, 99, 102, 105, 107, 108, 111, 112, 116, 119, 126, 128, 132, 135, 137, 138, 140, 147, 148, 150, 151, 161, 163, 164, 165, 168, 172, 173, 178, 179, 180, 181, 183, 186, 187, 193, 194, 195, 198, 206, 207, 209, 210, 213, 215, 216, 217, 222, 225, 232, 233, 234, 235 4 7, 18, 20, 21, 23, 26, 28, 34, 42, 43, 44, 45, 48, 54, 55, 57, 58, 59, 60, 64, 65, 66, 68, 70, 71, 72, 76, 79, 81, 83, 86, 88, 98, 100, 101, 104, 106, 110, 113, 114, 117, 118, 120, 121, 125, 127, 129, 130, 131, 133, 134, 139, 141, 143, 144, 145, 146, 149, 154, 155, 156, 157, 158, 159, 160, 167, 169, 170, 171, 175, 177, 184, 188, 189, 191, 196, 199, 200, 203, 204, 205, 208, 211, 212, 219, 220, 221, 223, 224, 226, 227, 228, 229, 230 5 17, 122, 124, 166, 218 6 73, 123 Table 4. Descriptive statistics of 15 quantitative metric traits among Heteropogon contortus accessions, cluster mean performances and principal components. C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 Descriptive statistics of traits among total accessions Min 36.2 9.00 20.0 0.10 3.73 3.00 3.50 0.30 3.27 0.30 1.07 0.10 3.90 0.10 2.67 Max 110.7 265.0 435.0 0.34 14.1 13.0 29.0 1.03 9.00 1.03 10.7 0.63 12.0 0.50 12.0 average 74.3 74.3 125.6 0.19 8.32 5.39 14.1 0.60 5.96 0.60 6.50 0.35 7.16 0.35 5.64 SD 12.61 45.38 72.93 0.04 1.91 1.43 3.64 0.11 0.92 0.11 1.51 0.06 1.14 0.06 0.98 Kurtosis 0.23 2.32 2.51 2.13 -0.09 5.34 1.69 1.25 1.10 1.25 0.82 2.17 1.49 1.02 12.78 Skewness 0.22 1.28 1.26 0.44 -0.13 1.83 0.72 0.25 0.56 0.26 -0.26 -0.16 0.36 -0.53 2.11 Cluster means of traits I 61.0 51.8 69.8 0.16 8.17 4.67 9.63 0.48 5.16 0.48 5.15 0.30 6.51 0.30 5.06 II 95.3 66.1 109.7 0.22 9.17 5.89 18.9 0.70 7.27 0.70 6.88 0.36 8.36 0.37 11.10 III 72.6 90.1 143.1 0.18 8.73 4.93 13.7 0.57 5.87 0.57 6.26 0.34 7.02 0.34 5.48 IV 79.1 65.8 124.9 0.21 7.89 6.00 15.5 0.67 6.17 0.67 7.38 0.38 7.35 0.38 5.85 V 72.2 56.5 77.1 0.17 9.22 5.00 14.6 0.51 6.92 0.51 2.39 0.18 9.67 0.21 5.49 VI 107.0 72.5 262.5 0.34 7.00 11.5 28.7 0.90 8.35 0.90 6.65 0.35 6.85 0.35 4.8 Principal Components SD 2.23 1.46 1.35 1.17 0.99 0.93 0.83 0.73 0.71 0.63 0.49 0.44 0.42 0.31 0.07 VP 0.33 0.14 0.12 0.09 0.07 0.06 0.05 0.04 0.03 0.03 0.02 0.01 0.01 0.01 0.00 CP 0.33 0.48 0.60 0.69 0.75 0.81 0.86 0.89 0.93 0.95 0.97 0.98 0.99 1.00 1.00 EV 4.99 2.14 1.83 1.38 0.97 0.86 0.69 0.53 0.51 0.40 0.24 0.19 0.18 0.10 0.01 Descriptive statistics of traits among core subset Min 42.4 11.0 20.0 0.10 3.73 3.00 3.50 0.30 3.27 0.30 1.07 0.10 3.90 0.10 2.67 Max 110.0 265.0 435.0 0.34 14.07 13.00 29.00 1.03 9.00 1.03 9.57 0.45 12.00 0.45 11.95 average 75.0 83.6 136.9 0.19 8.47 5.71 14.31 0.59 6.04 0.59 5.90 0.32 7.27 0.32 5.52 SD 16.03 59.24 100.1 0.05 2.52 2.06 5.52 0.17 1.28 0.17 1.94 0.08 1.58 0.07 1.72 Kurtosis 0.13 1.69 1.97 2.82 -0.41 3.53 1.20 0.75 0.49 0.58 0.23 0.76 1.75 1.44 6.42 Skewness 0.297991 1.28 1.26 0.44 -0.13 1.83 0.72 0.25 0.56 0.26 -0.26 -0.16 0.36 -0.53 2.11 t test* 0.406 0.190 0.262 0.425 0.375 0.195 0.427 0.337 0.369 0.381 0.045 0.021 0.341 0.025 0.349 C1 = length of main tiller (cm); C2 = number of tillers/tussock; C3 = fresh weight/tussock (g); C4 = tiller internode diameter (cm); C5 = 4th inter-nodal length (cm); C6 = number of nodes on main tiller; C7 = 4th leaf blade length (cm); C8 = 4th leaf blade width (cm); C9 = 4th leaf sheath length (cm); C10 = 4th leaf sheath width (cm); C11 = flag leaf blade length (cm); C12 = flag leaf blade width (cm); C13 = flag leaf sheath length (cm); C14 = flag leaf sheath width (cm); C15 = inflorescence length (cm); EV = Eigen Values; VP = variance proportion; CP = cumulative proportion; * t test core subset vs. all accessions (shows P values). yield (-0.076). A difference of almost 3 times was noted varied from 3 to 14 cm with an average of 8.32 cm. for tiller internode diameter, which ranged from 1 to 3 Number of nodes per tiller varied from 3 to 13. Leaf mm. Internodal length between the 3rd and 4th nodes blade and leaf sheath lengths varied from 3 to 29 cm Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 366 A.K. Roy, D.R. Malaviya, P. Kaushal, S.K. Mahanta, R. Tewari, R. Chauhan and A. Chandra and 3 to 9 cm, respectively. Leaf blade and leaf sheath regards plant habit, 157 accessions were decumbent, 76 widths both varied from 0.3 to 1 cm. Such variation were erect and only 2 were prostrate. Tiller internode was also noted for flag leaf length and width. Flag leaf color varied from violet to green with some mixed blade length differed among accessions from 1 to 10 cm, shades. Greenish-violet tiller was predominant with whereas flag leaf sheath length differed from 3 to 12 cm. 114 accessions, whereas violet and light violet figured Width of flag leaf blade and sheath varied from 0.1 to 0.6 in 54 accessions. Sixty-seven accessions had light green cm and 0.1 to 0.35 cm, respectively. Inflorescence length tillers. Nodes were mostly greenish-violet to dark violet, also varied significantly, i.e. 2 to 11 cm. except in 16 accessions, which possessed green nodes. Nodes were predominantly non-hairy and only 60 accessions possessed scant hair on nodes. Leaf surface of 164 accessions was glabrous, while 71 accessions had medium or scant hairs. Light green and green were the most dominant leaf blade colors with 170 and 53 accessions, respectively. Two accessions were blue-green and 6 accessions had greenish-blue leaf blades. Green leaf sheath color was also dominant with 171 accessions being green or light green. Sixty-four accessions were noted with greenish-violet leaf sheaths. Anthocyanin coloration on leaf surfaces was less common than on leaf sheath surfaces and nodes. Spike color was green among all accessions but varied in intensity. Except for 2 accessions which possessed soft awns, accessions had hard awns. The soft-awned accessions were short-lived annual types. Ligule length ranged from 0.5 to 1 mm among accessions, with the majority being 1 mm. Figure 2. Scree plot of Heteropogon contortus accessions. One of the important nutritional parameters, crude protein concentration, varied from 2 to 10% with an Visual observation of non-metric traits revealed that average of 5.7%, whereas IVDMD varied from 31 to 97 accessions were poor to very poor in growth, whereas 59% with an average of 46.2% (Table 6). Variation in 72 accessions were good to very good (Table 5). As fiber concentrations ranged from 74 to 87% (average Table 5. Variation for non-metric traits and the ligule length in 235 Heteropogon contortus accessions. Growth Habit Leaf hairiness Leaf blade color Leaf sheath color Good 39 Decumbent 157 Glabrous 164 BG 2 Green 101 Medium 66 Erect 76 Light hairy 65 G 53 GV 64 Poor 61 Prostrate 2 Hairy 6 GV 4 LG 70 Very Good 33 GB 6 Very Poor 36 LG 170 Node color Node hairiness Tiller internode color Anthocyanin coloration on Anthocyanin coloration leaf blade on leaf sheath DV 60 LH 60 GV 114 Absent 151 Absent 88 G 16 NH 175 LG 67 Medium 46 Strong 30 GV 104 LV 14 Strong 29 Weak 117 V 55 V 40 Weak 9 Node anthocyanin Spike color Awn pubescence Awn stature Ligule length (mm) coloration distribution Absent 2 G 185 50% 203 Hard 233 0.05 42 DV 72 GV 30 100% 32 Soft 2 0.1 193 LV 32 LG 20 V 129 BG = bluish-green; G = green; GV = greenish-violet; LG = light green; LV = light violet; V = violet; GB = greenish-blue; DV = dark violet; LH = Light hairy; NH = Non-hairy; Number against traits are number of accessions. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Genetic diversity and core subset in Heteropogon contortus 367 Table 6. Nutritional parameters of 167 accessions of Heteropogon contortus. Ash % OM % CP % NDF % ADF % Hemi-cellulose % Cellulose % Lignin % IVDMD1 % Among accessions Min 5.35 87.9 2.14 74.4 38.2 23.72 22.9 2.70 31.5 Max 12.10 94.7 10.10 87.1 59.0 44.89 51.1 21.41 59.9 Average 8.58 91.4 5.67 81.7 48.0 33.70 36.9 7.39 46.2 Mean values among clusters Cluster 1 8.60 91.4 5.93 81.8 47.1 34.7 36.2 7.18 46.3 Cluster 2 7.67 92.3 5.06 81.7 47.1 34.6 36.5 6.49 44.9 Cluster 3 8.69 91.3 5.52 81.9 47.6 34.3 36.2 7.76 46.0 Cluster 4 8.48 91.5 5.76 81.6 48.7 32.8 37.9 7.20 46.4 Cluster 5 9.00 91.0 5.57 81.4 47.0 34.4 35.7 6.48 51.8 Cluster 6 9.56 90.4 5.37 78.9 47.6 31.3 33.9 7.57 48.6 Among core subset accessions Min 5.87 89.45 3.81 78.6 41.8 28.0 31.3 2.70 36.0 Max 10.55 94.13 7.79 85.7 52.8 39.1 41.7 8.04 52.5 Average 8.67 91.33 5.69 81.7 47.6 34.1 36.4 6.58 46.6 t test2 0.375 0.375 0.462 0.470 0.230 0.264 0.214 0.004 0.381 OM = organic matter; CP = crude protein concentration; NDF = neutral detergent fiber concentration; ADF = acid detergent fiber concentration; Cell = cellulose; Lig = lignin; IVDMD = in vitro dry matter digestibility. 181 accessions analyzed. 2t test core subset vs. all accessions (shows P values). Table 7. Core subset of Heteropogon contortus germplasm. Cluster Original Log Proportion No. of accessions No. of second-level No. of accessions selected Accession number size value of log value to be selected cluster formed from each cluster (as per SN in Table 1) 1 30 1.477 0.21 8 8 8 5, 13, 46, 53, 90, 115, 153, 214 2 3 0.477 0.07 2 2 2 74, 231 3 101 2.004 0.29 10 10 10 9, 22, 35, 85, 91, 105, 126, 147, 207, 222 4 94 1.973 0.28 10 10 10 20, 23, 54, 66, 101, 121, 158, 160, 184, 229 5 5 0.699 0.10 3 3 3 17, 122, 218 6 2 0.301 0.04 2 1 2 73, 123 Total 235 6.931 35 35 81.7%) for NDF and 38 to 58% (average 48.0%) showed non-significant variation, except for lignin for ADF. Differences for lignin were much greater and concentration (Table 6). ranged from 2.7 to 21.4%. Ash concentration ranged from 5 to 12% and organic matter from 87 to 94%. Accessions Discussion showed variation for hemicellulose and cellulose, which varied from 23 to 44% and 22 to 51%, respectively. Analysis of data recorded on various morphological A subset of 35 accessions was identified as a quantitative metric and non-metric traits, as well as core germplasm set, representing 15% of the total nutritive parameters established high variability among germplasm evaluated (Table 7). This subset represented Heteropogon contortus accessions. The germplasm 8, 2, 10, 10, 3 and 1 accessions from Clusters 1 to 6, represented tropical semi-arid climates (126 accessions respectively. Almost all variability was captured in the from Chhattisgarh, Madhya Pradesh and Maharashtra), core subset (Tables 4 and 6). Student’s t-test for various tropical arid climate (29 accessions from Rajasthan), morphological traits of the core subset against the total tropical subhumid climate (9 accessions from Tamil Nadu, accessions showed non-significant variation, except for Karnataka and Kerala), subtropical semi-arid climate flag leaf blade length and width and flag leaf sheath (63 accessions from Uttar Pradesh) and subtemperate width (Table 4). Similarly, t-test for various nutritional to temperate climate (8 accessions from Himachal parameters of the core subset against the total accessions Pradesh, Jammu and Kashmir). A high degree of genetic Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 368 A.K. Roy, D.R. Malaviya, P. Kaushal, S.K. Mahanta, R. Tewari, R. Chauhan and A. Chandra variation was also noted for non-metric traits, growth The presence of genetic variability in the germplasm and habit along with nutritional parameters. Clustering collection of this apomictic grass can be attributed to of accessions collected from different places indicated recombinations taking place owing to residual sexuality that these accessions either originated from a common among some plants. The grass is an aposporous apomict source or moved from one place to other, resulting in with 2n = 20, 40, 44, 60 and 80 (Fedorov 1974; lesser inter-population differences. Grouping of material Srivastava and Purnima 1990). Although it is primarily based on statistical numerical procedures helps in apomictic in nature, it shows high genetic variability in understanding of variability, which further becomes the subhumid dry regions of south India, probably due to basis for identification of core germplasm, which can the presence of residual sexuality in a few accessions. be exploited in breeding programs. Development of a Genetic diversity in the species has been reported by core subset is effective when it is truly representative Carino and Daehler (1999) utilizing molecular studies of the variation present in the germplasm collection. within and among Hawaiian populations of the species. A comparison of the data shows that almost all the Roy (2004) and Bhat and Roy (2007; 2014) have also variability has been captured in the core subset. reported high genetic variability among germplasm The study established high genetic variability and based on morphological and isozyme studies. A correlation of some traits with biomass production. considerable amount of localized variation was reported Tiller number, high nodal number and longer leaf in H. in the early botanical literature because of occasional contortus were reported to be associated with forage sexual reproduction in H. contortus (Soromessa 2011). yield (Roy 2004). Morphological traits like plant height, The lack of uniformity among individuals in Hawaiian number of tillers and leaf length were also reported to be populations of this apomictic grass, based on RAPD associated with forage yield and considered as important studies, indicated frequent sexual reproduction (Carino traits in constructing selection criteria for forage yield and Daehler 1999). Diverse forms collected from in the perennial grass Sehima nervosum (Roy et al. the same location (Bhat and Roy 2007) also indicate 1999). Earlier studies involving tropical perennial presence of some recombinations taking place through grasses indicated a wide range of diversity for different residual sexuality among some plants. characters and the clustering pattern was observed to Other factors could also be responsible for such be independent of their geographical distribution in variation. Introduction of germplasm from other countries Dichanthium (Agarwal et al. 1999; Chauhan et al. 2007), over different time periods is one potential source of Sehima nervosum (Roy et al. 1999), H, contortus (Roy variation, but there are no documented reports of this 2004; Bhat and Roy 2007; 2014) and Guinea grass (Jain occurring. However, dispersal of seed through other biotic et al. 2003a, 2003b, 2006; Roy et al. 2020). The annual and abiotic sources cannot be ruled out. One plausible accessions, with soft awns, were quite low in biomass. explanation of seed dispersal may be its hard awns, which In fact, these accessions were short-lived perennials and get stuck in animal or human bodies or coats. could not survive in the harsh climate of Jhansi, India The collections represent different agro-ecological (max temp 45 °C during May and June). Such annual zones of the country and variation was found within types have been described from India earlier (Soromessa local populations as well as between populations, which 2011). In earlier studies, evaluation of accessions of indicates a distinct possibility of sexual reproduction H. contortus collected from different parts of India taking place, unless the species was introduced into revealed high values for heritability, genetic advance India from more than one source. Further domestication and genotypic and phenotypic coefficients of variation and recombination might have contributed more to for tiller number/tussock, green fodder yield and dry the variability. The presence of among‐population matter yield (Roy 2004), and were considered to be differentiation but lack of between‐island differentiation useful for selection of accessions. An isozymic study was considered to indicate that H. contortus was an early on H. contortus accessions indicated high genetic intra- Polynesian introduction to the Hawaiian Islands (Carino species diversity; however, clustering could not be and Daehler 1999). Hence, further molecular studies correlated with the geographical origin of the accessions involving Indian germplasm and germplasm from some (Bhat and Roy 2014). other countries, particularly neighboring ones, may give The present study identified the wide range of genetic some idea on possible movement of germplasm. The study variation among the germplasm set evaluated, which is in will help researchers to focus future studies on this core congruence with earlier reports of high genetic variation. subset in developing genetic improvement programmes. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Genetic diversity and core subset in Heteropogon contortus 369 Acknowledgments Multivariate Behavioral Research 1:245–276. doi: 10.1207/s15327906mbr0102_10 Authors are thankful to National Bio-resource Chauhan R; Tiwari R; Roy AK; Kaushal P; Malaviya DR; Development Board, Department of Biotechnology, Chandra A; Mahanta SK. 2007. Variation for morphological Government of India for financial support. Authors are traits in Dichanthium Bothriochloa complex. Range also thankful to Indian Council of Agricultural Research, Management and Agroforestry 28(2):293–294.Dabadghao PM; Shankarnarayan KA. 1973. The Grass Covers New Delhi, India. of India. Indian Council of Agricultural Research, New Delhi, India. Conflicts of interest Daehler CC; Goergen EM. 2005. Experimental restoration of an indigenous Hawaii grassland after invasion by Buffel The authors declare that they have no conflict of interest. grass (Cenchrus ciliaris). Restoration Ecology 13:380– 389. doi: 10.1111/j.1526-100X.2005.00047.x References Fang Y; Xiong L. 2015. General mechanisms of drought response and their application in drought resistance (Note of the editors: All hyperlinks were verified 30 July 2021). improvement in plants. Cellular and Molecular Life Sciences 72:673–689. doi: 10.1007/s00018-014-1767-0 Agarwal DK; Gupta S; Roy AK; Gupta SR. 1999. Study FAO. 2007. Plant genetic resources of grassland and forage on agromorphological variation vis-à-vis geographical species. Report of Commission on Genetic Resources for distribution in marvel grass [Dichanthium annulatum Food and Agriculture. Background Study Paper No. 40. L. (Stapf.)]. Plant Genetic Resources Newsletter Rome, Italy. bit.ly/2WEjNfD 118:27–29. hdl.handle.net/10568/105418 Fedorov A. 1974. Chromosome numbers of flowering plants. Ahmad L; Kanth RH; Parvaze S; Mahdi SS. 2017. Agro-climatic Otto Koeltz Science Publishers, D-624, Koenigstein, West and agro-ecological zones of India. In Ahmad L; Kanth RH; Germany. Parvaze S; Mahdi SS, eds. Experimental agrometeorology: Goergen E; Daehler CC. 2001. Reproductive ecology of a A practical manual. Springer, Cham, Switzerland. p. 99– native Hawaiian grass (Heteropogon contortus; Poaceae) 118. doi: 10.1007/978-3-319-69185-5_15 versus its invasive alien competitor (Pennisetum setaceum; AOAC. 1980. Official methods of analysis. 13th Edn. Association Poaceae). International Journal of Plant Sciences of Official Analytical Chemists, Washington, DC, USA. 162(2):317–326. doi: 10.1086/319587 Bhat BA; Roy AK. 2007. Genetic diversity in Heteropogon Goergen E; Daehler CC. 2002. Factors affecting seedling contortus. Range Management and Agroforestry 28:300– recruitment in an invasive grass (Pennisetum setaceum) 301. and a native grass (Heteropogon contortus) in the Bhat BA; Roy AK. 2014. Genetic diversity based on isozymic Hawaiian Islands. Plant Ecology 161:147–156. doi: banding pattern in Heteropogon contortus - A perennial 10.1023/A:1020368719136 tropical forage grass. Indian Journal of Agricultural Goering HK; Van Soest PJ. 1970. Forage fiber analyses: Sciences 84(12):1464–1469. bit.ly/3xewgD0 Apparatus, reagent, procedures and some Applications. Bor NL. 1960. The grasses of Burma, Ceylon, India and Agricultural Handbook No. 379, ARS, USDA, Washington, Pakistan. Pergamon Press, London, UK. doi: cabi.org/isc/ DC, USA. bit.ly/3ygoqdp abstract/19766710676 Hoffmann BD. 2008. Response of ant communities to Brown AHD. 1989a. Core collections: a practical approach to experimental fire regimes on rangelands in the Victoria genetic resources management. Genome 31:818–824. doi: River District of the Northern Territory. Austral Ecology 10.1139/g89-144 28(2):182–195. doi: 10.1046/j.1442-9993.2003.01267.x Brown AHD. 1989b. The case for core collections. In: Brown ICAR. 2018. Agricultural Research Data Book 2018. Indian AHD; Frankel OH; Marshall DR; Williams JR, eds. The Agricultural Statistics Research Institute (ICAR), New use of plant genetic resources. p. 136–156. Cambridge Delhi, India. bit.ly/38E4upI University Press, Cambridge, UK. bit.ly/3jGU2o0 IRRI. 2020. STAR (Statistical Tool for Agricultural Research) Carino DA; Daehler CC. 1999. Genetic variation in Version 2.0.1. International Rice Research Institute (IRRI). an apomictic grass, Heteropogon contortus, in the bbi.irri.org Hawaiian Islands. Molecular Ecology 8:2127–2132. doi: Jain A; Roy AK; Kaushal P; Malaviya DR; Zadoo SN. 2003a. 10.1046/j.1365-294x.1999.00786.x Principal component analysis in Guinea grass (Panicum Carino DAM. 1999. Competitive ability, susceptibility to maximum Jacq.) germplasm. Indian Journal of Plant invasion, and genetic variation of piligrass (Heteropogon Genetic Resources 16:96–99. bit.ly/2VompgE contortus). M.Sc. Thesis. University of Hawaii at Manoa, Jain A; Zadoo SN; Roy AK; Kaushal P; Malaviya DR. 2003b. Honolulu, HI, USA. bit.ly/3mY9nlZ Meiotic system and probable basic chromosome number Cattell RB. 1966. The Scree test for the number of factors. of Panicum maximum Jacq. accessions. Cytologia 68:7– Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 370 A.K. Roy, D.R. Malaviya, P. Kaushal, S.K. Mahanta, R. Tewari, R. Chauhan and A. Chandra 13. doi: 10.1508/cytologia.68.7 in sen grass. Range Management and Agroforestry Jain A; Roy AK; Kaushal P; Malaviya DR; Zadoo SN. 2006. 20(2):131–135. Isozyme banding pattern and estimation of genetic diversity Roy AK; Agrawal RK; Bhardwaj NR; Mishra AK; Mahanta among Guinea grass germplasm. Genetic Resources and Crop SK. 2019b. Revisiting national forage demand and Evolution 53:339–347. doi: 10.1007/s10722-004-0574-0 availability scenario. In: Roy AK; Agrawal RK; Bhardwaj Malaviya DR; Roy AK. 2021. Time to return green to NR, eds. Indian Fodder Scenario: Redefining State Wise grasslands. In: Narain S; Mahapatra R; Das S; Misra A; Status. ICAR- AICRP on Forage Crops and Utilization, Parrey AA, eds. State of India’s environment. p. 290–297. Jhansi, India. p. 1–21. Centre for Science and Environment, New Delhi, India. Roy AK; Malaviya DR; Kaushal P. 2019a. Genetic Malaviya DR; Roy AK; Kaushal P. 2018. Rangelands/ improvement of dominant tropical Indian range grasses. Grasslands of India: Current status and future prospects. Range Management and Agroforestry 40(1):1–25. bit.ly/ In: Squires VR; Dengler J; Feng H; Hua L, eds. Grasslands 3j1JwWQ of the world: Diversity, management and conservation. p. Singh AK. 2012. Probable agricultural biodiversity heritage 223–240. CRC Press. bit.ly/3rJdlPB sites in India: XI. The Upper Gangetic Plains Region. Moata MRS. 2009. Characterizing Heteropogon contortus Asian Agri-History 16(1):21–44. bit.ly/2WTUeaB (piligrass) as living mulch for tropical vegetable crops. Soromessa T. 2011. Heteropogon contortus (L.) P.Beauv. ex M.Sc. Thesis. College of Tropical Agriculture and Human Roem. & Schult. In Brink M; Achigan-Dako EG, eds. Resources, University of Hawaii, Manoa, USA. bit.ly/3yIL91j PROTA (Plant Resources of Tropical Africa), Wageningen, Rhind PM. 2010. Terrestrial biozones – a rough guide to the Netherlands. bit.ly/3BUU6Yc world’s major terrestrial ecosystems: plant formations in Sparks DN. 1973. Algorithm AS 58: Euclidean cluster analysis. the Upper Gangetic Plain bio-province. Upper Gangetic Journal of the Royal Statistical Society. Series C (Applied Plain Moist Deciduous Forest. bit.ly/37bO33a Statistics) 22:126–130. doi: 10.2307/2346321 Roy AK. 2004. Genetic variability, character association and Srivastva AK; Purnima. 1990. Agamospermy in some path analysis in black spear grass (Heteropogon contortus polyploidy grasses. Acta Botanica Indica 18:240–246. L.). Range Management and Agroforestry 25(1):11–16. cabi.org/isc/abstract/19921631394 Roy AK: Malaviya DR; Kaushal P; Mahanta SK; Tewari R; Tilley JMA; Terry RA. 1963. A two-stage technique for the in Chauhan R; Chandra A. 2020. Diversity study among Guinea vitro digestion of forage crops. Grass and Forage Science grass (Megathyrsus maximus Jacq.) (Poales: Poaceae) 18:104–111. doi: 10.1111/j.1365-2494.1963.tb00335.x genotypes and development of a core germplasm set. Wang X; Zhao L; Yan B; Shi L; Liu G; He Y. 2016. Morphological Plant Genetic Resources: Characterization and Utilization and physiological responses of Heteropogon contortus to 18(6):470–482. doi: 10.1017/S1479262120000507 drought stress in a dry-hot valley. Botanical Studies 57:17. Roy AK; Agarwal DK; Gupta S. 1999. Character association doi: 10.1186/s40529-016-0131-0 (Received for publication 4 September 2020; accepted 8 July 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):371–375 371 doi: 10.17138/TGFT(9)371-375 Short Communication What drives the adoption of fodder innovation(s) in a smallholder dairy production system? Evidence from a cross-sectional study of dairy farmers in India ¿Qué impulsa la adopción de nuevas opciones forrajeras en un sistema de producción de leche a pequeña escala? Evidencia de un estudio transversal entre productores de leche en India D. THIRUNAVUKKARASU1, N. NARMATHA2 AND S. ALAGUDURAI1 1Krishi Vigyan Kendra, Kallakurichi, Tamil Nadu Veterinary and Animal Sciences University, Tamil Nadu, India. tanuvas.ac.in 2Department of Veterinary and Animal Husbandry Extension Education, Veterinary College and Research Institute, Tamil Nadu Veterinary and Animal Sciences University, Namakkal DT, Tamil Nadu, India. tanuvas.ac.in Abstract The study in India involving 384 households found that 42.7% of dairy farmers adopted new forage varieties when varieties were released. The farmer’s resources, their caste, access to markets for milk and price received for milk had positive effects on the decision to adopt. Management of farms by women-headed households had negative effects on the adoption decision. Increased forage yield and ease of propagation and establishment were important reasons for adoption of varieties, e.g. the relative advantage of pearl millet × Napier grass (Cenchrus americanus × C. purpureus) vs. hedge lucerne (Desmanthus virgatus). Thus, researchers need to address these issues when developing new germplasm, if farmers are to readily adopt new varieties, especially in the case of resource-poor farmers. Keywords: Forage attributes; surveys; smallholder dairying; tropical forages; variety acceptance. Resumen El estudio en India que involucró a 384 hogares encontró que el 42.7% de los productores de leche adoptaron nuevas variedades de forrajes cuando se liberaron. Los recursos del agricultor, su casta, el acceso a los mercados de la leche y el precio recibido por la leche tuvieron efectos positivos en la decisión de adoptar. Las mujeres enfrentaron más dificultades para adoptar las nuevas opciones forrajeras en sus fincas. El aumento del rendimiento del forraje y la facilidad de propagación y establecimiento fueron razones importantes para la adopción de variedades, p.ej. la ventaja relativa del mijo perla × pasto elefante (Cenchrus americanus × C. purpureus) frente al frijolillo (Desmanthus virgatus). Por lo tanto, los investigadores deben abordar estos problemas al desarrollar nuevo germoplasma, para que los agricultores adopten fácilmente nuevas variedades, especialmente en el caso de agricultores de escasos recursos. Palabras clave: Aceptación de variedad; atributos del forraje; encuestas; forrajes tropicales; lechería en pequeña escala. Correspondence: D. Thirunavukkarasu, Tamil Nadu Veterinary and Animal Sciences University, Kallakuruchi DT, Tamil Nadu, India. Email: dthirunavukkarasu@gmail.com Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 372 D. Thirunavukkarasu, N. Narmatha, and S. Alagudurai Introduction fodder and 25% for dry fodder (residues of cereal and pulse crops). This scenario holds good for Tamil Nadu Attributes (characteristics) of an innovation are considered state, a tropical region and one of the leading milk- to be drivers of its adoption or rejection by end-users, producing states of India. For a substantial period, explaining 49‒87% of the variance in adoption (Rogers various stakeholders, including Tamil Nadu Agricultural 2003). These attributes include: relative advantage over University (TNAU), have been addressing the shortage the existing technology; compatibility with values, of feed resources and between 1976 and 2019 TNAU lifestyle and needs; ease of use; trialability on a small released 22 fodder varieties/hybrids (Table 1). scale; and results/benefits and risk/uncertainty easily Mean annual yields of pearl millet × Napier, multi- seen; these attributes influence the decision to adopt or cut sorghum and hedge lucerne are 80, 49 and 20 not (Rogers 2003; Trope and Liberman 2003; Castaño tonnes dry matter (DM)/hectare, respectively, while the et al. 2008). Socio-demographic factors also impact on nearest competitor, single-cut sorghum, yields about 10 decisions to adopt (Arts et al. 2011). These inferences tonnes DM/ha. From 2010 onwards, Animal Husbandry are based mainly on adoption of consumer goods, such Department of Tamil Nadu intensively promoted and as durable goods, fast-moving consumer goods, fashion propagated perennial forage germplasm, namely: pearl etc. and have been used to modify and design innovations millet × Napier, multi-cut forage sorghum and hedge and/or reposition them in the market. However, research lucerne as mixed fodder crops/individual crops through on factors driving adoption of improved fodder varieties various incentive programs across the state (Government by smallholder dairy-farmers, who account for 90% of of Tamil Nadu 2018). Continuous efforts of the various milk produced and are associated with 80 million rural stakeholders resulted in an increase in forage cropping. A households in India, is limited. micro-study by Thirunavukkarasu et al. (2014) reported Since 1970, various stakeholders have made attempts that the area under pearl millet × Napier had increased to enhance productivity of dairy animals through from 0.01 ha to 0.08 ha during the period 2001‒2011 at improvement in feed and fodder resources, inter alia. an individual household level. Thirunavukkarasu et al. These innovations encompassed: enrichment of crop (2011a; 2011b) reported wide variation in availability residues; promotion of concentrate feeding; and fodder of green fodder and deficits of dry fodder across the cultivation. However, livestock are still under-nourished state. To provide base data for planning future fodder and there is an estimated 35.6% deficit of green fodder development programs, a deeper understanding of the in India (Indian Grass and Fodder Research Institute socio-demographic factors which affect adoption of new 2013). Furthermore, for the year 2025, Singh et al. forage varieties was needed. We conducted a study of (2013) predicted the deficit to increase to 65% for green dairy farmers to clarify the situation. Table 1. Major fodder varieties developed in Tamil Nadu Agriculture University between 1976 and 2019. Fodder Year of release Purpose Hedge lucerne 1976; 2019 Introduced as a perennial multi-cut crop to meet the protein requirements and minimize (Desmanthus virgatus) costs of protein supplementation. In 2019 for the first time mutational breeding was carried out and an improved variety was released. Pearl millet × Napier grass 1982‒20121 To replace cereal-based green crop residue; to reduce grazing dependence; and as a (Cenchrus americanus × perennial fodder. Five varieties released. C. purpureus) Multi-cut fodder sorghum 2001‒20141 To replace single-cut sorghum as perennial green and dry fodder. Two varieties (Sorghum bicolor × released. S. sudanense) Guinea grass (Panicum 1993‒20091 Introduced as a perennial multi-cut crop; shade-tolerant. Three varieties released. maximum, now Megathyrsus maximus) Lucerne (Medicago sativa)1980‒20131 Early-maturing leguminous fodder crop. Two varieties released. Cowpea (Vigna 1986‒20041 Introduced as leguminous fodder crop; resistant to root-rot and cowpea yellow mosaic unguiculata) virus. Two varieties released. 1Periodic releases of improved varieties/hybrids with higher yields and better attributes. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Drivers of forage technology adoption by smallholder dairy farmers in India 373 Material and Methods prevailing in India). Fifty-three percent of respondents belonged to the rural middle economic class and owned A sample of 384 dairy-farming households, 1‒3 adult dairy animals, producing around 10 L milk/ proportionally representing different farming systems of day/household. Family women operated more than one- Tamil Nadu, were selected using stratified multi-stage third of the farms (35%) on their own with occasional random sampling (D. Thirunavukkarasu unpublished support of men, while remaining farms were operated data). The selected farmers were interviewed regarding by men or men plus women. Fifty-eight percent of their status in relation to adoption of new forage varieties respondents reared dairy animals primarily, to supply and socio-demographic factors were recorded. During their household needs, with any surplus sold as market the course of interviewing, fodder adoption status, milk. The majority (53.1%) of farmers were members of membership in farmers’ collectives, caste and gender farmers’ collectives (farmers’ producer organizations). role were captured at a categorical level and other socio- About 40% of farmers were able to market milk through demographic factors were measured at a continuous either co-operative dairies or privately-owned dairy level. For the purpose of triangulation of the collected processors; 19.3% of households had provisions to data, to improve understanding and obtain additional market milk through both co-operatives and private information, ‘focus group’ discussions were organized processors; 25.5% were able to market milk only through in villages. In this exercise, participants, including a milk vendor; and the remainder had no marketing farmers (both female and male) and the village-level opportunities. On average farmers produced 3,613 L animal health service providers, identified the reasons milk annually, for which they received 22 INR/L (1 for planting or not planting new varieties. USD = 75 INR). These farmers access mass media and To understand the socio-demographic differences mass contact programs (exhibitions, campaigns, etc.) to between adopters and non-adopters, the chi square obtain information. In addition, farmers interact with (for categorical variables) and Mann-Whitney U test the extension system (propagators of livestock-related (for continuous variables) were used taking account of innovations), including veterinarians, para-veterinarians nature of variables and non-normal distribution of data. and other associated stakeholders at village level. Binary logistic regression was used to understand the The data indicated that 42.7% of the dairy-farming causal factors that promote adoption of new varieties households have adopted at least 1 or more improved along with descriptive statistics. In performing binary fodder varieties promoted through the Animal logistic regressions, if a farmer adopted any of the Husbandry Department and others. Adopters and non- promoted fodders, the farm was coded as 1, with 0 for adopters differed significantly in terms of land holdings non-adopters. The binary logit model was as follows: and animal numbers, socio-economic class, reasons for dairying, gender role, caste, milk production, access to Fodder adoption status = β0+ β1x1+ β2x2+ β3x3+ [...] + βnxn markets, price received for milk, availability of mass where: media and extension agency contact but differences x1, x2, x3 [...] xn represent independent variables; were not necessarily influential in terms of adoption of β0 = constant; β , β are logistic variables; new varieties. On average households cultivated 0.1 ha 1 2 regression coefficients (estimates) (range 0‒1 ha) of fodder on their own or leased land. Adoption status = 0, (adoption ≤0) Among the adopting farmers (164 farmers), 87.8, 22.0 Adoption status = 1, (adoption >0). and 1.2 % of farmers planted pearl millet × Napier, multi- The above statistical analysis was carried out using Statistical cut fodder sorghum and hedge lucerne, respectively. Of Package for Social Sciences and spread sheets at stat-help. the total cultivated area devoted to fodder production, com/spreadsheets.html pearl millet × Napier accounted for 64.8% and fodder sorghum 35.2%, with very little under hedge lucerne. Results and Discussion Among the above discussed variables, those which differentiated between adopters and non-adopters were Mean age of respondents was 46 years, while land checked for bivariate relationships. Animal numbers holdings were marginal (20% of respondents were and socio-economic class, which had highly significant landless) for sustaining their family needs. Respondents bivariate correlation with land, caste and annual milk were mostly (58% of households) the third lowest caste production, were excluded for understanding the role in the Indian hierarchy (‘caste’ is a social stratification of predictor variables in logit regression. Among the Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 374 D. Thirunavukkarasu, N. Narmatha, and S. Alagudurai selected variables, land, milk sale price, annual milk are in agreement with observations by other researchers production, market opportunity, gender role and caste that incentives or extrinsic motivational factors drive explained about 47.3% (Pseudo R2 = 0.473) of the adoption of innovations (Donkor et al. 2018). variance in fodder adoption among the farmers (Table Women-operated farms were less likely to have 2). Even though many variables differed between cultivated fodder than those operated by men only or adopters and non-adopters, the only ones displaying men plus women. Mobility restriction of women, on positive relationships with fodder adoption were: account of cultural factors and limited access to transport land size, annual milk production and access to milk systems, limits their access to extension services, marketing opportunities (P<0.001), caste (P<0.05) and technological solutions and external inputs, which may price received for milk (P<0.10). For one unit changes limit their adoption of innovations (Theis et al. 2018). in acreage of land, liters of annual milk produced and Castes that are higher in social stratification tend to marketing opportunity score (ranged from 0 to 3), the adopt fodder innovations more than lower castes, which estimated odds of adoption (Odds ratio) are multiplied may be related to limited access to extension services by 1.3, 1 and 2.876, respectively. Members of the third and quality information among the lower castes (Krishna lowest caste were 2.2 times as likely to adopt as those in et al. 2019). Nguthi (2007) suggested that an indirect the second lowest caste, while gender had a significant inference for rejection by non-adopters may be a lack negative relationship (P<0.05) with adoption. Women- of fit with existing livelihood assets, available options operated farms were only 0.49 times as likely to adopt and activities for resource-poor farmers. Similarly, poor fodder innovations as male-operated farms, i.e. women- market access is a disincentive (due to market disparities) operated farms were less likely to adopt fodder varieties to adoption. than farms operated by men only or men and women. At the same time all new varieties of fodder are not Thus, farmers with better resources (relatively large uniformly adopted by farmers and group discussion land holdings) and access to commodity markets (markets with farmers and others revealed the following facts: plus good milk price), high milk production and higher While hedge lucerne was released earlier than pearl in caste hierarchy are more likely to adopt new forage millet × Napier and multi-cut forage sorghum, the latter varieties. Crossing of Indian-origin Zebu cattle breeds 2 are certainly preferred. Biomass yield of pearl millet with European origin dairy breeds for increasing milk × Napier is higher (a relative advantage) than that of productivity demanded improved feeding strategies, hedge lucerne, and planting materials (vegetative setts) such as feeding of cultivated fodder. Thus, in tandem, are readily available through exchange between farmers cattle breeding programs and adoption of fodder varieties (less complex). Collection of planting material (seeds) might have improved milk production. Large landholders of hedge lucerne is a tedious, laborious process and have the option to divert some land from cropping to seed is not readily available. In addition, hedge lucerne green forage production. Better marketing opportunities seeds need seed treatment prior to sowing to break seed in the form of better access to markets and higher dormancy. Therefore, relative ease of obtaining planting milk prices (incentives) act as extrinsic motivational materials, nature of planting material and forage yields factors for adoption. Even landless farmers had acted are obvious reasons for greater adoption of pearl millet to lease land for cultivation of fodder. These findings ×Napier and multi-cut sorghum than hedge lucerne. Table 2. Estimated coefficients of logistic regression for factors influencing adoption of new fodder varieties (n=384). Variable Estimated coefficient Standard error Odds ratio1 Land size 0.246 0.053 1.279*** Price received for milk 0.052 0.031 1.054# Annual milk production 0.000 0.000 1.000*** Access to milk marketing opportunities 1.056 0.162 2.876*** Gender role (women-operated farms coded as 1; otherwise as 0) -0.705 0.289 0.494* Caste (lowest ranked in caste hierarchy–Schedule caste coded as 0) Most backward caste (second lowest in caste hierarchy coded as 1) 0.408 0.388 1.053 Backward caste (third lowest in caste hierarchy coded as 2) 0.799 0.404 2.223* Constant -0.5480 0.910 0.004 Pseudo R2 = 0.473; Log likelihood = 357.4 ***P<0.001; **P<0.01; *P<0.05; #P<0.10 1Increased odds of adoption from a unit increase in the variable. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Drivers of forage technology adoption by smallholder dairy farmers in India 375 Conclusions Agriculture 8:121. doi: 10.3390/agriculture8080121 Government of Tamil Nadu. 2018. Animal Husbandry Policy Similar to the situation with consumer goods, socio- note 2018-19. bit.ly/3BIMB5V cultural factors and forage attributes influence the IGFRI (Indian Grassland and Fodder Research Institute). adoption/rejection of new forage varieties. These varieties 2013. Vision 2050. bit.ly/3mTGTK2 are more readily adopted by farmers with adequate Krishna VV; Aravalath LM; Vikraman S. 2019. Does caste determine farmer access to quality information? PLoS land, producing larger volumes of milk and with ready ONE 14(1):e0210721. doi: 10.1371/journal.pone.0210721 access to milk markets paying acceptable milk prices. Nguthi FN. 2007. Adoption of agricultural innovations by New forages with higher yields and ease of propagation smallholder farmers in the context of HIV/AIDS: The are also more readily adopted. When developing new case of tissue-cultured banana in Kenya. Ph.D. Thesis. germplasm and propagating new varieties, researchers Wageningen University, Netherlands. doi: 10.3920/978- need to consider these issues by focusing on material 90-8686-641-0 that does not require complex processes to establish Rogers EM. 2003. Diffusion of Innovations. 5th Edn. Free and manage. Extension professionals need to improve Press, New York, USA. access to the extension system and technological inputs Singh KM; Singh RKP; Jha AK; Kumar A. 2013. A micro by marginalized sections within the dairy-farming analysis of fodder production and marketing in India: The case of Bihar. MPRA Paper No. 53548. mpra.ub.uni- community, if all sectors of the community are to take muenchen.de/53548/ advantage of the new resources, especially resource- Theis S; Lefore N; Meinzen-Dick R; Bryan E. 2018. What poor farmers. happens after technology adoption? Gendered aspects of small-scale irrigation technologies in Ethiopia, Ghana, and References Tanzania. Agriculture and Human Values 35:671–684. doi: 10.1007/s10460-018-9862-8 (Note of the editors: All hyperlinks were verified 15 July 2021). Thirunavukkarasu M; Sankaran VM; Kathiravan G; Karunakaran R. 2011a. Estimating dry fodder availability Arts JWC; Frambach RT; Bijmolt THA. 2011. Generalizations and requirement for bovines. Indian Journal of Animal on consumer innovation adoption: A meta-analysis on Sciences 81:744–750. drivers of intention and behavior. International Journal Thirunavukkarasu M; Sankaran VM; Kathiravan G; of Research in Marketing 28:134-144. doi: 10.1016/j. Karunakaran R. 2011b. Green fodder availability in Tamil ijresmar.2010.11.002 Nadu – A district wise analysis. Indian Veterinary Journal Castaño R; Sujan M; Kacker M; Sujan H. 2008. Managing 89(10):18–20. krishikosh.egranth.ac.in/handle/1/66613 consumer uncertainty in the adoption of new products: Thirunavukkarasu D; Jothilakshmi M; Murugesan S; Temporal distance and mental simulation. Journal of Doraisamy KA. 2014. Transition of smallholder dairy Marketing Research 45:320−336. doi: 10.1509/jmkr.45.3. farming system - a micro study in Tamil Nadu, India. 320 Livestock Research for Rural Development 26, Article Donkor E; Owusu V; Sekyere EO; Ogundeji AA. 2018. The #88. lrrd.cipav.org.co/lrrd26/5/thir26088.htm adoption of farm innovations among rice producers in Trope Y; Liberman N. 2003. Temporal construal. Psychological Northern Ghana: Implications for sustainable rice supply. Review 110(3):403−421. doi: 10.1037/0033-295X.110.3.403 (Received for publication 2 May 2020; accepted 6 July 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):376–382 376 doi: 10.17138/TGFT(9)376-382 Nota Técnica Uso de sensores remotos en la determinación del forraje disponible de Urochloa humidicola cv. Llanero bajo pastoreo en la Altillanura colombiana Use of remote sensors to determine forage availability in grazed pastures of Urochloa humidicola cv. Llanero in the Colombian Altillanura RAÚL ALEJANDRO DÍAZ GIRALDO, MAURICIO ÁLVAREZ DE LEÓN Y OTONIEL PÉREZ LÓPEZ. Corporación Colombiana de Investigación Agropecuaria (Agrosavia). C.I. La Libertad. Villavicencio, Colombia. agrosavia.co Resumen La modernización de los sistemas pastoriles basados en pasturas del género Urochloa en los Llanos Orientales de Colombia requiere de técnicas que usan sensores remotos desde plataformas satelitales para estimar la oferta de forraje. En el C.I. Carimagua de la Corporación Colombiana de Investigación Agropecuaria (Agrosavia) se evaluó una pastura de Urochloa humidicola cv. Llanero con imágenes Landsat 8 y Sentinel 2A. Se utilizaron los índices de vegetación NDVI, SAVI, EVI y GNDVI, calculados a partir de las bandas azul, verde, rojo e infrarrojo cercano. Los resultados fueron analizados con el software de estadística R y se compararon con el aforo (= mediciones en campo) del forraje disponible bajo pastoreo en época seca. El aforo fluctuó entre 290 y 656 kg MS/ha y los índices de vegetación fueron, para los sensores Landsat 8 y Sentinel 2A, respectivamente: NDVI = 0.67 (±0.037) y 0.69 (±0.061); SAVI = 0.48 (±0.048) y 0.41 (±0.046); EVI = 0.70 (±0.052) y 0.41 (±0.047); y GNDVI = 0.60 (±0.028) y 0.70 (±0.034). La relación entre los índices de vegetación con la oferta de forraje fue lineal directa; para la valoración de los modelos predictivos se usaron los criterios coeficiente de determinación R2 (0.56‒0.72) y el error cuadrático medio (RMSE) (63.95‒80.16) de las ecuaciones de regresión. Se concluye que para las condiciones del estudio el EVI (para Landsat 8) y el NDVI (para Sentinel 2A) son índices apropiados para predecir la oferta forrajera del pasto Llanero. Palabras clave: EVI, imágenes satelitales, índices de vegetación, NDVI. Abstract Modernization of pastoral systems based on the use of Urochloa species in the Colombian Eastern Llanos need the use of remote sensing techniques from satellite platforms to estimate amount of offered forage. In the Carimagua Research Centre of the Colombian Corporation for Agricultural Research (Agrosavia), an Urochloa humidicola cv. Llanero pasture was evaluated using Landsat 8 and Sentinel 2A images. The NDVI, SAVI, EVI y GNDVI vegetation indices were determined by using the blue, green, red and near infrared bands and the results analyzed with the R free software, to relate those indices with forage availability field measures taken during the dry season. Forage availability ranged between 290 and 656 kg DM ha-1 and the vegetation indices for the Landsat 8 and Sentinel 2A sensors were: NDVI = 0.67 (±0.037) and 0.69 (±0.061); SAVI = 0.48 (±0.048) and 0.41 (±0.046); EVI = 0.70 (±0.052) and 0.41 (±0.047); y GNDVI = 0.60 (±0.028) and 0.70 (±0.034), respectively. The relationships between vegetation indices and forage availability were linear. The Coefficient of Determination (R2= 0.56‒0.72) and the Mean Square Error (MSR =63.95‒80.16) of the prediction equations were used. In conclusion, under the conditions of the study, the EVI for Landsat 8 and NDVI for Sentinel 2A were considered adequate for estimating forage availability of Urochloa humidicola cv. Llanero. Keywords: EVI, satellite images, NDVI, vegetation indexes. Correspondencia: Mauricio Álvarez de León, Agrosavia C.I. La Libertad. Km 17 vía a Puerto López. Villavicencio, Meta, Colombia Correo electrónico: malvarez@agrosavia.co Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Sensores remotos en la determinación del forraje de Urochloa humidicola en la Altillanura Colombiana 377 Introducción Por tanto, el objetivo de este trabajo fue estimar la oferta de forraje en una pastura de Urochloa humidicola Para el monitoreo de grandes extensiones de tierra donde cv. Llanero a partir de imágenes multiespectrales están incorporados los sistemas ganaderos, se propone provenientes de Landsat 8 y Sentinel 2A, en la subregión el uso de información proveniente de plataformas de la altillanura de los Llanos Orientales de Colombia. satelitales (Beaulieu et al. 2007). Con esta información, que es cada vez más detallada en términos de resolución Materiales y Métodos espectral, espacial y temporal, es posible caracterizar la cobertura herbácea y derivar prácticas de manejo de las La investigación se realizó en el Centro de Investigación áreas pastoriles (Beaulieu et al. 2006). Carimagua de la Corporación Colombiana de Investigación Los índices de vegetación (IV) derivados de Agropecuaria – Agrosavia, ubicado en el municipio de imágenes satelitales, como Landsat 8 y Sentinel 2A, son Puerto Gaitán, Meta, en la subregión de la altillanura combinaciones algebraicas de varias bandas espectrales, plana de los Llanos Orientales de Colombia (Figura 1). diseñadas para destacar el vigor de la vegetación y sus El promedio anual de las precipitaciones es de 2,000 propiedades (biomasa del dosel, radiación absorbida, mm; su distribución es monomodal, correspondiendo contenido de clorofila, entre otros) (Bannari et al. 1995). la época seca al período noviembre‒febrero. Los suelos Los IV más usados en la evaluación de pasturas y praderas en el área de estudio son Oxisoles y Ultisoles ácidos, son: NDVI (Normalized Difference Vegetation Index); isohipertérmicos (Álvarez y Rincón 2010). SAVI (Soil Adjusted Vegetation Index); EVI (Enhanced Vegetation Index); y GNDVI (Green Normalized Difference Vegetation Index) (Huete 1988; Candiago et al. 2015; Hoffmann 2018). Para áreas pastoriles se han aplicado estas técnicas de uso de sensores remotos en los Pirineos españoles (Barrachina et al. 2009, 2010); la Pampa argentina (Cristiano 2010; Irisarri et al. 2013); la región Este de Uruguay (Baeza et al. 2011); el departamento de Antioquia en Colombia (Ramírez 2014; Padilla 2017); y el estado de Sao Paulo en Brasil (Cisneros et al. 2020). Estudios realizados anteriormente en los Llanos del Orinoco en Colombia por Girard y Rippstein (1994), usando radiometría terrestre e imágenes SPOT, y por Serna-Isaza (2001), usando sensores AVHRR de NOAA, el sensor MSS/TM de Landsat 4 y 5 y los sensores Figura 1. Ubicación del área de pastoreo en el C.I. Carimagua: HRV a bordo de SPOT 3 y 4, han mostrado el valor de a) Departamento del Meta en la República de Colombia; b) los datos de la teledetección en plataformas satelitales Municipio de Puerto Gaitán en el Departamento del Meta; c) para la cartografía de pasturas mejoradas y sabanas Área de pastoreo y lotes evaluados. nativas, bajo diferentes manejos. En los llanos del Orinoco en Venezuela, Chacón (2004) realizó mapeos Los aforos de forraje y las imágenes corresponden de los ecosistemas de sabana utilizando imágenes al día 8 de diciembre de 2015 (época seca). Las áreas multitemporales del satélite NOAA, apoyándose en el bajo pastoreo por bovinos utilizadas son de pasturas conocimiento experto e índice NDVI. de Urochloa humidicola CIAT 6133 cv. Llanero (antes En la región de los llanos de Colombia es limitada considerado como Brachiaria dictyoneura; Rincón la información sobre el uso de sensores remotos para la et al. 2018) con manejo rotacional. El aforo se realizó estimación de la oferta forrajera, pese a que la región es en ocho potreros de 2.61 ha (±0.15) cada uno: En cada típicamente ganadera con una población bovina de 5.75 potrero (unidad experimental) se tomó en 20 puntos una millones de cabezas (ICA 2020) y un área estimada en muestra con un marco de 0.25 m2 cortando con hoz a 15 pasturas mejoradas, especialmente del género Urochloa, cm de altura, se pesó el forraje verde con una balanza de más de dos millones de hectáreas (Álvarez y Rincón electrónica y se secaron las muestras a una temperatura 2010). de 70 ºC durante tres días para determinar la materia seca. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 378 R.A. Díaz Giraldo, M. Álvarez de León y O. Pérez López Las imágenes Landsat 8 y Sentinel 2A fueron obtenidas En las Figuras 2, 3, 4 y 5 se presentan las rectas de del portal del Servicio Geológico de los Estados Unidos regresión de ajuste lineal entre los IV y la oferta de (earthexplorer.usgs.gov). Las correcciones atmosféricas y forraje (biomasa de las pasturas). el cálculo de los IV se realizaron con el módulo SCP (Semi- Con respecto a los índices obtenidos por Landsat 8, el Automatic Classification Plugin) y la calculadora ráster en EVI presentó el valor de R2 más alto (0.721) y el RMSE el Software libre Q-GIS. En el Cuadro 1 se presentan los más bajo (63.95) (Cuadro 3), por lo cual con base en los IV y las bandas espectrales utilizados en este trabajo. resultados experimentales es razonable inferir que este es el IV de elección para estimar la oferta forrajera. Con Cuadro 1. Fórmulas de cálculo de los índices de vegetación relación a los índices calculados a partir de Sentinel 2A, con base en las bandas espectrales utilizadas. el NDVI es el más apropiado por cuanto alcanza el valor Índice espectral de vegetación Referencia de R2 más alto (0.712) y RSME más bajo (64.89) en la NDVI=(IR – R) / (IR + R) Rouse et al. (1974) estimación de la oferta forrajera (Cuadro 3). SAVI= 1.5 × (IR – R) / (IR + R + 0.5) Huete (1988) En la Figura 6 se muestra la distribución espacial de EVI =2.5 × ((IR - R) / (IR + 6 × R – 7.5 Jiang et al. (2008) x A + 1)) la biomasa obtenida para cada sensor. GNDVI = (IR-V) / (IR+V) Wang et al. (2007) NDVI = Normalized Difference Vegetation Index; SAVI = 700 Soil Adjusted Vegetation Index; EVI = Enhanced Vegetation Index; GNDVI = Green Normalized Difference Vegetation 600 Index. A = banda espectral del azul; V = banda espectral del verde; R = banda espectral del rojo; IR = banda espectral del 500 infrarrojo cercano. 400 A partir de las imágenes de Landsat 8 y Sentinel 2A se calcularon los IV según las fórmulas en el Cuadro 1 y se 300 procedió a establecer las ecuaciones de regresión lineal que explican la relación entre el IV y la oferta forrajera 200 para cada una de las imágenes. Una vez obtenidos los 0.55 0.60 0.65 0.70 0.75 0.80 modelos que estiman la producción de forraje se determinó Índice NDVI cuál es el de mejor ajuste, para lo cual se utilizaron los NDVI-L8 NDVI-S2 criterios raíz del error cuadrático medio (root mean square Linear (NDVI-L8) Linear (NDVI-S2) error, RMSE) y el R2 de las ecuaciones. Adicionalmente Figura 2. Rectas de regresión de ajuste lineal para el índice se estableció, mediante la prueba t de Student, si había NDVI obtenido por los sensores de Landsat 8 (NDVI-L8) y diferencias entre los valores estimados por las ecuaciones Sentinel 2A (NDVI-S2) y la oferta de forraje. de regresión para cada uno de los sensores. Para todos los análisis se usó el software libre R (R Core Time 2016). 700 Resultados 600 La oferta de forraje determinada en campo varió entre 500 290 y 656 kg MS/ha para el día de la evaluación, con un promedio de 501 kg MS/ha. Los valores de los 400 índices que se obtuvieron de las imágenes de Landsat 8 y Sentinel 2A se muestran en el Cuadro 2. 300 Cuadro 2. Valores medios y dispersión de los índices de 200 vegetación obtenidos de los dos sensores. 0.80 0.25 0.45 0.65 0.85Índice EVI Índice Landsat 8 Sentinel 2A EVI-L8 EVI-S2 NDVI 0.67 ± 0.037 0.69 ± 0.061 Linear (EVI-L8) Linear (EVI-S2) SAVI 0.48 ± 0.048 0.41 ± 0.046 Figura 3. Rectas de regresión de ajuste lineal para el índice EVI 0.70 ± 0.052 0.41 ± 0.047 EVI obtenido por los sensores de Landsat 8 (EVI-L8) y GNDVI 0.60 ±0.028 0.70 ± 0.034 Sentinel 2A (EVI-S2) y la oferta de forraje. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Forraje (kg MS/ha) Forraje (kg MS/ha) Sensores remotos en la determinación del forraje de Urochloa humidicola en la Altillanura Colombiana 379 700 700 600 600 500 500 400 400 300 300 200 0.50 0.55 0.60 0.65 0.70 0.75 2000.75 0.30 0.40 0.50 0.60 Índice GNDVI Índice SAVI GNDVI-L8 GNDVI-S2 SAVI-L8 SAVI-S2 Linear (GNDVI-L8) Linear (GNDVI-S2) Linear (SAVI-L8) Linear (SAVI-S2) Figura 4. Rectas de regresión de ajuste lineal para el índice Figura 5. Rectas de regresión de ajuste lineal para el índice GNDVI obtenido por los sensores de Landsat 8 (GNDV-L8) y SAVI obtenido por los sensores de Landsat 8 (SAVI-L8) y Sentinel 2A (GNDVI-S2) y la oferta de forraje. Sentinel 2A (SAVI -S2) y la oferta de forraje. Cuadro 3. Ecuaciones de regresión lineal y los criterios de selección R2 y RMSE para los sensores e índices utilizados. Índice Sensor Ecuación R2 RMSE NDVI Landsat 8 Y= 2,376.5 X - 1,100.2 0.628 73.84 Sentinel 2A Y= 1,547.7 X - 572.71 0.712 64.89 EVI Landsat 8 Y = 1,912.6 X - 844.28 0.721 63.95 Sentinel 2A Y = 1,638 X - 187.98 0.655 71.08 GNDVI Landsat 8 Y = 3,131.5 X – 1,375.1 0.628 73.81 Sentinel 2A Y = 2,596.3 X – 1,307.6 0.645 72.15 SAVI Landsat 8 Y = 1,861 X - 400.93 0.648 71.83 Sentinel 2A Y = 1,788.2 X - 228.29 0.561 80.16 RMSE = error cuadrático medio. Figura 6. Representación gráfica de la biomasa estimada en el área de estudio. a) oferta de forraje estimada con la ecuación lineal a partir del EVI obtenido con Landsat 8; b) oferta de forraje estimada con la ecuación lineal a partir del NDVI obtenido con Sentinel 2A. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Forraje (kg MS/ha) Forraje (kg MS/ha) 380 R.A. Díaz Giraldo, M. Álvarez de León y O. Pérez López Discusión Conclusiones La oferta promedio de forraje del pasto Llanero obtenida Con la información obtenida a partir de los sensores de en esta investigación (501 kg MS/ha) fue inferior a los Landsat 8 y Sentinel 2A fue posible establecer, en forma valores reportados por Pérez et al. (2019) para el mismo preliminar, un modelo de relación lineal directa entre pasto, también en época seca y en el C.I. Carimagua los índices EVI y NDVI con el forraje disponible de (645.7 kg MS/ha con periodos de descanso de 35 a 40 U. humidicola cv. Llanero bajo pastoreo en la Altillanura días), Carulla et al. (1991) en el Piedemonte llanero del colombiana. El uso de Landsat 8 y Sentinel 2A tiene el Meta (669 kg MS/ha) y Pardo y Pérez (2010), también potencial para convertirse en una herramienta predictiva en el Piedemonte llanero (1,365 kg MS/ha para el mismo para el manejo de praderas en sistemas de producción pasto y 770 kg MS/ha para Brachiaria humidicola). pastoriles de esta región. Las diferencias en los valores de biomasa que se Estudios futuros deben considerar la validación de la observan en la Figura 6 pueden atribuirse a factores información de las plataformas satelitales para la gestión como las propiedades del suelo, la microtopografía, la de áreas pastoriles mayores, e incorporar en ellos tanto humedad y temperatura las cuales pueden ocurrir aún en series multitemporales de imágenes asociadas a los áreas relativamente pequeñas (Tamme et al. 2016). ciclos hidrometeorológicos como para otras especies de Respecto a los valores de oferta forrajera estimados pastoreo. con base en Landsat 8 y Sentinel 2A para cada uno de los índices, no se encontraron diferencias significativas Agradecimientos (P>0.05) según la prueba t de Student. Los modelos utilizados muestran su potencial para predecir el aforo Los autores agradecen al Ministerio de Agricultura y de pastizales, pero deben tomarse con precaución sobre Desarrollo Rural de Colombia por el apoyo financiero todo los obtenidos con Landsat 8, considerando los R2 y al personal administrativo y operativo de los Centros entre 0.56 y 0.72 y RMSE entre 63.95 y 80.16, para de Investigación Carimagua y La Libertad de Agrosavia, Landsat 8 y Sentinel 2A, respectivamente. quienes con su apoyo hicieron posible el desarrollo de Mientras que los cuatro índices mostraron relaciones las actividades del proyecto. positivas con el aforo de los pastos (Cuadro 3), es necesario reconocer que las observaciones realizadas a Referencias partir de Sentinel 2A tiene un mayor detalle respecto a la variabilidad espacial de la oferta forrajera, debido a las (Nota de los editores: Los enlaces se verificaron el 2 de agosto de 2021). características de la resolución espacial en comparación con Landsat 8. Esto permite una mejor estimación de la Álvarez M; Rincón A. 2010. Características agroecológicas oferta de forraje, para áreas de estudio ≥2 hectáreas, tal de la Orinoquia colombiana. En Rincón A; Jaramillo CA, como es evidente en la Figura 6. eds. Establecimiento, manejo y utilización de recursos forrajeros en sistemas ganaderos de suelos ácidos. La estimación de la oferta forrajera del pasto debe Corporación Colombiana de Investigación Agropecuaria ser contextualizada con la magnitud del área a evaluar: – Corpoica. Villavicencio, Colombia. p. 9–26. hdl.handle. Desde luego estudios a una escala mayor (1:2,500) net/20.500.12324/12703 (potrero o predio) exigen sensores cuya resolución Baeza S; Paruelo J; Ayala W. 2011. Eficiencia en el uso de la espacial corresponda a tamaños de pixeles relativamente radiación y productividad primaria en recursos forrajeros pequeños (10 m × 10 m) como es el caso de Sentinel del este de Uruguay. Agrociencia Uruguay 15(2):48–59. 2A, mientras que en el caso de trabajo a escala menor bit.ly/3jfqhZP (1:50,000; regional, nacional) pueden ser utilizados Bannari A; Morin D; Bonn F; Huete AR. 1995. A review of sensores como Landsat 8 (30 m × 30 m), con tamaños vegetation indices. Remote Sensing Reviews, 13(1-2), de pixeles mayores. Para una escala aún menor 95–120. doi: 10.1080/02757259509532298Barrachina M; Cristobal J; Tulla AF. 2010. Los recursos (1:100,000), Baeza et al. (2011) y Ramírez (2014) han ganaderos en los sistemas extensivos de la montaña usado exitosamente el sensor MODIS con pixeles de pirenaica catalana: aproximación al cálculo de la producción 250 m × 250 m, para estudios de sabanas tropicales y de biomasa herbácea mediante el uso de la teledetección. subtropicales. Serie Geográfica 16:35–49. hdl.handle.net/10017/14221 Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Sensores remotos en la determinación del forraje de Urochloa humidicola en la Altillanura Colombiana 381 Barrachina M; Cristobal J; Tulla A; Pons X. 2009. Análisis Instituto Colombiano Agropecuario ICA. 2020. Censo de la producción de biomasa de los prados y pastos de la pecuario año 2020. ICA, Bogotá, Colombia. bit.ly/3lonJv9 Vall Fosca (Pirineo Axial-Nogueres). Memorias del XIII Irisarri JGN; Gundel PE; Clavijo MP; Durante M; Sosa P. Congreso de la Asociación Española de Teledetección. 2013. Estimación de la PPNA y la capacidad de carga Calatayud, 23–26 de septiembre de 2009. p. 225–228. bit.ly/ por ambientes mediante información satelital en un 3A3CPKB establecimiento ganadero en la Pampa Deprimida. Revista Beaulieu N; Álvarez M; Dewispeleare G; Rosa R; Sano E; Argentina de Producción Animal 33(1):11–20. hdl.handle. Yamamoto Y. 2007 Potenciales y limitaciones de las net/11336/4255 imágenes multiespectrales de satélite para el monitoreo de Jiang Z; Huete AR; Kamel D; Miura T. 2008. Development of pasturas degradadas en las sabanas tropicales. Innovación a two-band enhanced vegetation index without a blue band. y Cambio Tecnológico 5(5):48–53. hdl.handle.net/20.500. Remote Sensing of Environment 112(10):3833–3845. doi: 12324/754 10.1016/j.rse.2008.06.006 Beaulieu N; Álvarez M; Rincón Á. 2006. Detección de la Padilla GA. 2017. Estimación de la productividad primaria degradación en pasturas usando series temporales de neta aérea (PPNA) del forraje de pastos para los sistemas imágenes multiespectrales en los Llanos Orientales de de producción ganadera mediante sensores remotos. B.Sc. Colombia. Boletín de investigaicón No. 8. Corporación Tesis. Universidad del Valle, Cali, Colombia. hdl.handle. Colombiana de Investigación Agropecuaria – Corpoica. net/10893/10315 Villavicencio, Colombia. hdl.handle.net/20.500.12324/ Pardo O; Pérez O. 2010. Alternativas forrajeras para los llanos 1218 orientales de Colombia. En Rincón A; Jaramillo CA, Candiago S; Remondino F; De Giglio M; Dubbini M; Gattelli eds. Establecimiento, manejo y utilización de recursos M. 2015. Evaluating multispectral images and vegetation forrajeros en sistemas ganaderos de suelos ácidos. indices for precision farming applications from UAV Corporación Colombiana de Investigación Agropecuaria – images. Remote Sensing 7(4):4026–4047. doi: 10.3390/ Corpoica. Villavicencio, Colombia. p. 27–74. hdl.handle. rs70404026 net/20.500.12324/12703 Carulla JE; Lascano CE; Ward JK. 1991. Selectivity of resident Pérez O; Bueno GA; Baquero JE; Onofre HG; Cassalett E; and oeophageal fistulated steers grazing Arachis pintoi Molina AM; Arango MA; Velásquez JG; Rodríguez NS; and Brachiaria dictyoneura in the Llanos of Colombia. Moreno LM; Amaya CV; Bernal JH; Correal WA; Pabón Tropical Grasslands 25:317–324. hdl.handle.net/10568/ DM; Villalobos MA; Velásquez JH; Fuentes JC. 2019. 44085 Modelo productivo tecnologías integrales para sistemas de Chacón-Moreno EJ. 2004. Mapping savanna ecosystems producción de bovinos de cría en la altillanura colombiana. of the Llanos del Orinoco using multitemporal NOAA Corporación Colombiana de Investigación Agropecuaria satellite imagery. International Journal of Applied Earth (Agrosavia), Mosquera, Colombia. hdl.handle.net/20.500. Observation and Geoinformation 5(1):41–53. doi: 12324/35060 10.1016/j.jag.2003.08.003 R Development Core Team. 2016. R: A Language and Cisneros A; Fiorio P; Menezes P; Pasqualotto N; Van Environment for Statistical Computing. Version 4.0.0 Wittenberghe S; Bayma G; Nogueira SF. 2020. Mapping 2020-04-24, Arbor Day. productivity and essential biophysical parameters of Ramírez A. 2014. Evaluación del crecimiento de pastos usando cultivated tropical grasslands from sentinel-2 imagery. índices de vegetación calculados a partir de información Agronomy 10(5):711. doi: 10.3390/agronomy10050711 satelital. M.Sc. Tesis. Universidad Nacional de Colombia, Cristiano PM. 2010. Estimación de la productividad primaria Medellín, Colombia. repositorio.unal.edu.co/handle/unal/ mediante el modelo de eficiencias y sensores remotos. 75325 Ph.D. Tesis. Universidad de Buenos Aires. Buenos Aires, Rincón A; Flórez H; Ballesteros H; León LM. 2018. Efectos Argentina. hdl.handle.net/20.500.12123/6206 de la fertilización en la productividad de una pastura de Girard C; Rippstein G. 1994. Utilisation de données SPOT Brachiaria humidicola cv. Llanero en el Piedemonte de HRV pour la cartographie de savanes et pâturages dans los Llanos Orientales de Colombia. Tropical Grasslands- les Llanos de Colombie. Bulletin - Société Française de Forrajes Tropicales 6:158–168. doi: 10.17138/tgft(6)158- Photogrammétrie et de Télédétection 133: 11–19. agritrop. 168 cirad.fr/387246/ Rouse JW; Haas RH; Schell JA; Deering DW. 1974. Hoffmann M; Butenko Y; Traore S. 2018. Evaluation of satellite Monitoring Vegetation Systems in the Great Plains with imagery to increase crop yield in irrigated agriculture. Agris ERTS. Third ERTS Symposium, NASA, Washington, DC, on-line Papers in Economics and Informatics 10(3):45–55. USA. p. 309–317. ntrs.nasa.gov/citations/19740022614 doi: 10.22004/ag.econ.281644 Serna-Isaza RA. 2001. Teledetección para la cartografía de la Huete AR. 1988. A soil-adjusted vegetation index (SAVI). vegetación de Sabana. En Rippstein G; Escobar G; Motta Remote Sensing of Environment 25(3):295–309. doi: F, eds. Agroecología y biodiversidad de las sabanas en los 10.1016/0034-4257(88)90106-X Llanos Orientales de Colombia. Publicación CIAT No. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 382 R.A. Díaz Giraldo, M. Álvarez de León y O. Pérez López 322. Centro Internacional de Agricultura Tropical (CIAT), 27(5):1012–1022. doi: 10.1111/jvs.12431 Cali, Colombia. p. 97–110. hdl.handle.net/10568/55153 Wang F; Huang J; Tang Y; Wang X. 2007. New vegetation index Tamme R; Gazol A; Price JN; Hiiesalu I; Pärtel M. 2016. and its application in estimating leaf area index of rice. Co-occurring grassland species vary in their responses to Rice Science 14(3):195–203.doi: 10.1016/S1672-6308(07) fine-scale soil heterogeneity. Journal of Vegetation Science 60027-4 (Recibido para publicación 28 septiembre 2020; aceptado 27 julio 2021; publicado 30 septiembre 2021) © 2021 Tropical Grasslands-Forrajes Tropicales es una revista científica de acceso abierto publicada por el Centro Internacional de Agricultura Tropical (CIAT), en asocio con la Chinese Academy of Tropical Agricultural Sciences (CATAS). Este trabajo se publica bajo la licencia Creative Commons Atribución 4.0 Internacional (CC BY 4.0). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):383–390 383 doi: 10.17138/TGFT(9)383-390 Short Communication The effects of increasing concentrations of Trichanthera gigantea leaves in pellets on the nutritive value and short-term intake of diets of grass plus pellets offered to lambs reared under tropical conditions in the Caribbean Efectos de las concentraciones crecientes de hojas de Trichanthera gigantea en pélets sobre el valor nutritivo y la ingesta a corto plazo de dietas de pasto más pélets ofrecidas a corderos criados en condiciones tropicales en el Caribe H.A. JACK1,2, L.M. CRANSTON1, J.L. BURKE1, M. KNIGHTS3 AND P.C.H. MOREL1 1School of Agriculture and Environment, Massey University, Palmerston North, New Zealand. massey.ac.nz 2Caribbean Agricultural Research and Development Institute, St Augustine, Trinidad and Tobago. cardi.org 3Biosciences, Agriculture & Food Technologies, University of Trinidad and Tobago, Centeno, Trinidad and Tobago. utt.edu.tt/baft Abstract There is currently limited information on the benefits of increasing the concentration of Trichanthera gigantea leaves in pelleted diets offered to lambs reared under tropical conditions in the Caribbean. Twelve crossbred Barbados Blackbelly rams aged 5 months were used to determine the effects of increasing the concentrations of T. gigantea in pelleted diets, on the nutritive value and intake of grass forage plus pellets offered to lambs. Animals were randomly assigned to a basic diet (4 kg) of chopped Cenchrus purpureus plus 1 of 6 pelleted diets (500 g) comprised of either 100% intact commercial pellets or a pelleted mixture of ground commercial pellets and ground (dry fallen) T. gigantea leaf in the following ratios (T. gigantea leaves:ground commercial pellets): 20:80 (T20); 40:60 (T40); 60:40 (T60); 80:20 (T80); and 100:0 (T100). Total intakes of forage and pellets (TPI) were measured at the end of each day during a period of 7 days, and the average daily nutrient intakes of the different treatment diets were calculated. Overall, there was no significant difference in the intakes of pellets containing 0 to 80% T. gigantea leaves (P>0.05) but intakes of pellets comprising 100% T. gigantea leaves were significantly lower (P<0.0001). Both CP and soluble protein intakes declined progressively as the percentage of T. gigantea leaves in the pellets increased. While level of T. gigantea leaves in pellets fed to lambs did not generally affect total intakes of pellets, grass, or grass+pellets, animal performance on these various rations cannot be assumed to be similar until longer-term feeding studies have been performed, as reduced protein and energy concentrations in the pellets could significantly lower weight gains as level of leaf in the pellets increased. Keywords: Barbados Blackbelly sheep, multi-purpose trees, pellet feeding. Resumen Actualmente, existe información limitada sobre los beneficios de aumentar la concentración de hojas de Trichanthera gigantea en las dietas peletizadas que se ofrecen a los corderos criados en condiciones tropicales en el Caribe. Se utilizaron doce carneros Barbados Blackbelly mestizos de 5 meses de edad para determinar los efectos del aumento de las concentraciones de T. gigantea en dietas peletizadas, sobre el valor nutritivo y la ingesta de forraje de pasto más pélets ofrecidos a los corderos. Los animales fueron asignados aleatoriamente a una dieta básica (4 kg) de Cenchrus purpureus picado más 1 de 6 dietas peletizadas (500 g) compuestas por pélets comerciales 100% intactos o una mezcla peletizada Correspondence: Heidi Jack, School of Agriculture and Environment, Massey University, Private bag 11-222, Palmerston North, New Zealand. E-mail: h.jack@massey.ac.nz Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 384 H.A. Jack, L.M. Cranston, J.L. Burke, M. Knights, and P.C.H. Morel de pélets comerciales molidos y hojas secas de T. gigantea en las siguientes proporciones (hojas de T. gigantea:pélets comerciales molidos): 20:80 (T20); 40:60 (T40); 60:40 (T60); 80:20 (T80); y 100:0 (T100). Se midieron las ingestas totales de forraje y pélets (TPI) al final de cada día durante un período de 7 días, y se calcularon las ingestas promedio diarias de nutrientes de las diferentes dietas de tratamiento. En general, no hubo diferencias significativas en la ingesta de pélets que contenían de 0 a 80% de hojas de T. gigantea (P> 0.05), pero la ingesta de pélets compuestos 100% de hojas de T. gigantea fue significativamente menor (P <0.0001). Tanto la ingesta de proteína cruda como la de proteína soluble disminuyeron progresivamente a medida que aumentaba el porcentaje de hojas de T. gigantea en los pélets. Si bien el nivel de hojas de T. gigantea en los pélets dados como alimento a los corderos generalmente no afectó la ingesta total de pélets, pasto o pasto+pélets, no se puede suponer que el rendimiento de los animales en estas diversas raciones sea similar hasta que se hayan realizado estudios de alimentación a más largo plazo, ya que las concentraciones reducidas de proteína y energía en los pélets podrían reducir significativamente las ganancias de peso a medida que aumenta el nivel de hojas en los pélets. Palabras clave: Alimentación con pellets, árboles polivalentes, oveja de Barbados Blackbelly. Introduction there is no known study on the use of fallen leaves as a prospective feed ingredient for lambs. Further, there Trichanthera gigantea is a common non-leguminous is currently no information on the nutritive value of dry multi-purpose tree species (MPT) used in small ruminant fallen leaves of T. gigantea and effects of feeding them production systems in the Caribbean (Heuzé et al. 2017). to lambs in the Caribbean. Therefore, the objective of The nutritive value of fresh intact T. gigantea leaves this study was to determine the effects on nutritive value is attributed to its high protein concentration, which and intake of pellets of increasing concentrations of dry ranges between 150 and 220 g/kg DM (Rosales 1997; fallen T. gigantea leaves in pellets offered to lambs with Rosales and Rios 1999). In addition, the presence of grass forage. hydrolyzable tannins in T. gigantea may increase rumen undegradable or bypass protein, which can be a direct Materials and Methods benefit to ruminants when consumed (Rosales 1997; Edwards et al. 2012). Compared with other MPTs at the The effects of including dry ground fallen T. gigantea same stage of maturity, T. gigantea is typically higher leaf in commercial pellets at 0 (T0), 20 (T20), 40 in non-structural and storage carbohydrates and lower (T40), 60 (T60), 80 (T80) and 100% (T100) fed with in structural carbohydrate, which results in its high a fresh grass forage basal diet on intake by lambs were rumen degradability (Rosales and Rios 1999). Further, examined over 2 periods: Period 1 (10‒15 May 2019) T. gigantea has cystoliths on leaf and stem surfaces, and Period 2 (22‒28 May 2019). Due to limitations which result in high ash concentration and a large with the facilities (spacing), all 6 treatments could not percentage of calcium, which is typically greater than be compared at the same time so intakes of treatments 20% DM (Benton and Benton 1963; Barahona 1999). T0, T20 and T40 were measured during Period 1 The higher ash concentration may be used to improve and intakes of Treatments T60, T80 and T100 were the mineral concentrations in the diets of livestock in the measured during Period 2. The study was conducted tropics, where mineral deficiencies in tropical pastures at the Eastern Caribbean Institute for Agriculture and often occur (McDowell and Arthington 2005). Forestry (ECIAF) – University of Trinidad and Tobago Apart from fresh leaves, leaf fall may be a potential (10.56° N, 61.32° W). dry season feed for animals, despite the possible lower nutritive value relative to intact T. gigantea Harvesting and pelleting material leaves as a result of senescence (Charlton et al. 2003). During periods of prolonged drought, there is often an Dry fallen T. gigantea leaves (mature flowering stage; abundance of biomass available as leaf fall (Wright and approximately 88% DM) were collected one week Cornejo 1990). This may be significant, particularly in prior to the study period from the plantation at the “Up the Caribbean, where prolonged severe dry periods are the Hill Farms”, which is located in Moruga, Trinidad frequent and are predicted to become more common (10.11° N, 61.29° W). (Lallo et al. 2017). Though there are several studies Dry leaves and a commercial ration were the primary focused on the use of fresh intact T. gigantea leaves, ingredients used to produce the pellets examined in this Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Pellets with Trichanthera leaves in the Caribbean 385 study. On the Control diet (T0, i.e. 100% commercial were randomly assigned on the basis of live weight to pellets) intact commercial pellets made of 80% (DM 3 groups of 4 animals, which were allocated to 1 of 3 basis) wheat middlings, 20% (DM basis) corn and diets (T0, T20 and T40) and intakes were recorded for 7 a vitamin and mineral mix were fed with fresh grass days. Lambs were then returned to the same diet as fed forage (see below). Pellets fed in the other treatments prior to Period 1 for 5 days. Period 2 then commenced, included mixtures of ground commercial pellets and where the concentrations of T. gigantea in pellets were ground dry fallen T. gigantea leaves in the following 60, 80 and 100%. Groups of lambs fed T0, T20 and T40 ratios (T. gigantea leaves:ground commercial pellets): diets in Period 1 were assigned to dietary treatments 20:80 (T20); 40:60 (T40); 60:40 (T60); 80:20 (T80); T60, T80 and T100, respectively. This was done to and 100:0 (T100). Firstly, the commercial pellets minimize between-lamb variation within treatment and dry T. gigantea leaves were ground separately to groups. During the experiment, all lambs were confined pass through a 0.635 cm screen (screen was initially to well-ventilated individual pens (1.22 × 1.22 m) and 2.54 cm and modified to 0.635 cm) of a Craftsman had unrestricted access to water and a mineral block shredder-hammer mill (Model 247.776380). The (Alphablock), which contained: 55,000 IU Vitamin A; ground materials were weighed according to the ratios 27,500 IU Vitamin D3; 300 IU Vitamin E; 30,000 mg for the different pellet treatments. After weighing the calcium; 5,000 mg magnesium; 1,800 mg iron; 2,500 respective ratios for the different treatment groups, mg manganese; 50 mg cobalt; 1,500 mg zinc; 10 mg the ground materials were mixed manually for 10‒15 selenium; and 35 mg iodine per kg DM. minutes and pelleted using a Changchai-ZS1115 Pellet Mill (22 Horse-Power Diesel Engine) with a die length Experimental procedure and design and diameter of 2.54 and 1.27 cm, respectively. Prior to Periods 1 and 2, a single batch of pellets for each Animals were fed twice daily at 09:00 h (4 kg forage) treatment group was produced and fed to the respective and 15:00 h (500 g pellets). Total forage and total pellets treatments throughout the respective periods. offered and refused for each animal were recorded daily In addition to the pellets, mature (6‒8 weeks to calculate intake of each component of the diet. At regrowth and 1.5 m high) Cenchrus purpureus (syn. 06:30 h daily, both total forage intake (TFI) and total Pennisetum purpureum) grass was manually harvested pellet intake (TPI) were recorded. Total dry matter intake with a machete each day from the Eastern Caribbean (TDMI) was calculated as the sum of TPI and TFI. Institute for Agriculture and Forestry Campus – University of Trinidad and Tobago (ECIAF-UTT) Sampling and analytical procedures according to Gemeda and Hassen (2014). C. purpureus was used as the basal diet for both Periods 1 and 2. Feed samples (forage and pellets) were collected at the Once harvested, the C. purpureus (including leaves end of each week for DM determination and chemical and stem) was manually chopped to lengths of about analysis. The pellet samples included 2 subsamples from 5‒10 cm according to Schnaider et al. (2014), for daily a total of 2 batches used for feeding. Forage on offer feeding. during the week was consistent and a representative sample was selected to determine nutritive value. The Animals and diets total nutrient concentration in the diets was calculated by determining the concentrations of each nutrient The same 12 crossbred (Barbados Blackbelly × West (on a DM basis) in both forage and pellets fed, from African) intact rams, aged 5 months, were used in both which total daily intakes on the various treatments were periods (Periods 1 and 2) to measure the intakes of the calculated. treatment diets. Mean live weight at the commencement of Period 1 was 22 ± 2.2 kg and for Period 2 was 27 Chemical analysis ± 2.4 kg. Before Period 1 commenced, the lambs were subjected to a 19-day adaptation period, where they Samples were dried at 60 °C for 72 h and ground to were examined, treated for internal parasites, fed a diet pass through a 2 mm sieve using a Thomas Scientific of 4 kg of chopped C. purpureus (including leaves and mill. These were then packaged (package included stem) plus 500 g of commercial pellets and allowed to Export permit no. 139517 for Research) and exported become familiar with their enclosures. The 12 lambs to Cumberland Valley Analytical Services (CVAS; Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 386 H.A. Jack, L.M. Cranston, J.L. Burke, M. Knights, and P.C.H. Morel Waynesboro, PA, USA) for further analysis. Dry matter pellets (181 g/kg DM) and C. purpureus (150 g/kg of C. purpureus (modified method) was determined DM) was higher than those reported for T. gigantea by drying samples at 105 °C for 3 h (National Forage pellets (98‒145 g/kg DM) and T. gigantea leaves (81 Testing Association 2002). Dry matter concentrations g/kg DM). Actual CP% in feed consumed could be for pellets and T. gigantea were determined by drying higher than these data suggest as only about 75% of the samples at 35 °C for 2 h (method no. 930.15, AOAC grass offered was eaten. Since sheep are very selective 2000). ADF was determined using a Whatman 934-AH and chopped material of whole plants was offered, glass micro-fiber filter with 1.5 μm particle retention one could assume that the lambs selected for leaf and in place of a fritted glass crucible (modification of rejected the stem, which would have much lower CP method no. 973.18, AOAC 2000). NDF was obtained concentration than the average figures quoted for the using Whatman 934-AH glass micro-fiber filters with grass. Unfortunately we did not analyze the forage 1.5 μm particle retention used in place of a fritted glass rejected by the lambs to clarify this point. Soluble crucible (a modification of Van Soest et al. 1991). Ash protein of C. purpureus was up to 20 g/kg DM more was determined using 0.35 g sample, which was ashed than that of the commercial pellets and more than for 4 h at 535 °C (a modification to method no. 942.05, twice the average value of 21.4 g/kg DM, reported for AOAC 2000). the pellets containing T. gigantea leaf. ADF and NDF concentrations of the feed components ranged between Statistical analysis 109 and 425 g/kg DM and 302 and 660 g/kg DM, respectively, for all feeds. Cenchrus purpureus had the Statistical analysis was conducted using R environment highest concentrations of both ADF and NDF. for statistical computing and visualization (R Core The average feed and nutrient intakes for the Team 2013). Intake measurements obtained from different treatment groups are presented in Table 2. each lamb at different times were treated as repeated Total forage intake (TFI) and total dry matter intake measures. Package nlme (Pinheiro et al. 2018) was (TDMI) were comparable across all treatment groups used to apply a linear mixed effect model to the intake (P>0.05), ranging between 0.770 and 0.795 kg DM/ data. The model consisted of treatment, day and day × hd/d for TFI and 1.13 and 1.21 kg DM/hd/d for TDMI, treatment interaction as fixed effects and animal as the while total pellet intake (TPI) of the T100 group was random effect. An analysis of variance (Anova) from lower (P<0.0001) than that of all other treatments. Package car (Fox and Weisberg 2011) and Agricolae Treatment had a significant effect on intakes of (de Mendiburu 2019) was used to obtain the P-value nutrients (Table 2). CP intake declined progressively for the model differences. Means and superscripts were from 194 g CP/hd/d for T0 group to 148 g CP/hd/d for generated using the R package emmeans (Lenth et T100 group (P<0.0001). Similarly, intake of soluble al. 2019) and multcomp (Hothorn et al. 2016), which protein (SP) declined from 55 g SP/hd/d for T0 to 45 g helps in separating significantly different means using SP/hd/d for T100 (P<0.0006). The average ADF intake Tukey’s multiple comparison test. Differences were for Groups T0, T20 and T40 (380 g/hd/d) was less considered statistically significant if P<0.05. than that for Groups T60, T80 and T100 (448 g/hd/d) (P<0.0001). Results Total pellet intake/day (TPI) did not vary throughout the study for T0, T20, T40, T60 and T80, nor did it vary Chemical composition of the dry fallen T. gigantea between these treatments (P>0.05) but increased from leaves used, C. purpureus and all pelleted feeds offered Day 1 to Day 5 for T100, before declining again (Table to lambs in the current study is presented in Table 1. 3). TPI on T100 was lower (P=0.0001) than on other The crude protein (CP) concentration of commercial treatments on all days except Days 3 and 5. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Pellets with Trichanthera leaves in the Caribbean 387 Table 1. Chemical composition (g/kg DM) including crude protein, soluble protein, acid detergent fiber, neutral detergent fiber, ash and organic matter for Cenchrus purpureus, dry fallen Trichanthera gigantea leaves (TGL) and pellets fed in various treatments (T0, T20, T40, T60, T80 and T100). Parameter C. purpureus TGL1 T02 T20 T40 T60 T80 T100 Dry matter (g/kg) 270 878 870 876 871 854 843 833 Crude protein 150 81 181 145 143 103 109 98 Soluble protein 53 19 33 25 28 21 20 13 Acid detergent fiber 425 308 109 142 152 295 287 340 Neutral detergent fiber 660 430 351 302 316 365 432 459 Ash 132 194 64 85 88 151 159 159 Organic matter 138 684 806 791 783 703 684 674 1TGL: Trichanthera gigantea (ground dry fallen leaves of T. gigantea). 2Commercial pellets were offered intact with grass forage for the Control group (T0). Other treatments (T20, T40, T60 and T80) comprised ground commercial pellets mixed with increasing proportions of ground dry fallen T. gigantea leaf; T100 represents pellets with ground T. gigantea leaves as the sole ingredient. Table 2. Average daily feed and nutrient intakes for the different treatment groups (n=4 lambs per treatment). Parameter T01 T20 T40 T60 T80 T100 s.e. P-value Forage intake (kg DM/hd/d) 0.770 0.738 0.762 0.764 0.767 0.795 0.023 0.678 Pellet intake (kg DM/hd/d) 0.435a 0.438a 0.435a 0.423a 0.417a 0.330b 0.013 <0.0001 Total intake (kg DM/hd/d) 1.21 1.18 1.20 1.19 1.18 1.13 0.028 0.4049 Crude protein (kg/hd/d) 0.194a 0.179ab 0.172bc 0.162cd 0.161cd 0.148d 0.003 <0.0001 Soluble protein (kg/hd/d) 0.055a 0.052ab 0.051ab 0.051ab 0.049bc 0.045c 0.001 0.0006 Acid detergent fiber (kg/hd/d)) 0.374a 0.389a 0.378a 0.459b 0.447b 0.439b 0.010 <0.0001 Neutral detergent fiber (kg/hd/d) 0.660 0.640 0.621 0.673 0.689 0.659 0.015 0.0707 Ash (kg/hd/d) 0.129a 0.139a 0.135a 0.168b 0.168b 0.154b 0.003 <0.0001 1T: Trichanthera gigantea (ground dry fallen leaves); T0: Commercial pellets (Control group fed grass forage + 100% commercial pellets); T20 group fed forage + pellets comprised of 20% ground dry fallen T. gigantea leaf and 80% commercial; T40 group fed forage + pellets comprised of 40% dry fallen T. gigantea leaf and 60% commercial; T60 group fed forage + pellets comprised of 60% dry fallen T. gigantea leaf and 40% commercial; T80 group fed forage + pellets with 80% dry fallen T. gigantea leaf and 20% commercial; and T100 fed forage + pellets comprised of 100% dry fallen T. gigantea leaf. Means within rows with the same letters are not significantly different (P>0.05). Table 3. Total pellet intake (TPI) (kg DM/day) of lambs fed rations made up of Cenchrus purpureus forage plus commercial pellets or forage plus pellets made of a mixture of ground commercial pellets and ground dry fallen leaves of Trichanthera gigantea in varying proportions (n=4 lambs per treatment). Treatment P value Day T01 T20 T40 T60 T80 T100 s.e. Treatment Day Treatment × Day 1 0.435ax2 0.438ax 0.435ax 0.402ax 0.418ax 0.319bx 0.0179 0.0001 0.6092 0.8518 2 0.435ax 0.438ax 0.435ax 0.426ax 0.421ax 0.316bx 3 0.435ax 0.438ax 0.435ax 0.427abx 0.421abx 0.354bxy 4 0.435ax 0.438ax 0.435ax 0.427ax 0.419ax 0.322bxy 5 0.435ax 0.438ax 0.435ax 0.427ax 0.411ax 0.376ay 6 0.435ax 0.438ax 0.435ax 0.427ax 0.412ax 0.316bx 7 0.435ax 0.438ax 0.435ax 0.427ax 0.414ax 0.309bx 1T: Trichanthera gigantea (ground dry fallen leaves); T0: Grass forage + commercial pellets (Control group fed pellets comprised of 100% commercial ingredients); T20 group fed grass forage + pellets comprised of 20% ground dry fallen T. gigantea leaf + 80% commercial; T40 group fed grass forage + pellets comprised of 40% ground dry fallen T. gigantea leaf and 60% commercial; T60 group fed grass forage + pellets comprised of 60% ground dry fallen T. gigantea leaf and 40% commercial; T80 group fed grass forage + pellets with 80% ground dry fallen T. gigantea leaf and 20% commercial; and T100 fed grass forage + pellets comprised of 100% ground dry fallen T. gigantea leaf. 2Means followed by the same letters (a,b,c,d) within rows are not significantly different (P>0.05) and means followed by the same letters (x,y,z) within columns are not significantly different (P>0.05). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 388 H.A. Jack, L.M. Cranston, J.L. Burke, M. Knights, and P.C.H. Morel Discussion CP consumed and, based on CP intakes, the daily intakes of SP barely reached the minimum required amounts for The inclusion of MPTs above 50% in ruminant diets is growing lambs (0.05‒0.07 kg SP/h/d) in all treatment often associated with reduced intake as a result of anti- groups. An adequate supply of degradable and bypass nutritional factors inherent to these species (Reed 1995; protein from diets is associated with increased efficiency Min et al. 2003). However, in the current study, the total of microbial fermentation; improved digestion; increased intakes of pellets with up to 80% T. gigantea (T80) were throughflow from the rumen; and therefore increased comparable with that of the commercial pellets which intake and improved performance (Lazzarini et al. 2009; lambs were accustomed to being fed. Unlike many other Sampaio et al. 2009). multi-purpose tree species, there is no known report of While total intakes of dry matter were not affected anti-nutritional factors that limit the intake of T. gigantea by amount of T. gigantea leaves included in the pellets, (Barahona 1999; Wanapat 2009). This may explain the performance of animals on the different rations could vary comparable intakes of pellets comprising up to 80% substantially as CP concentrations in the different rations T. gigantea leaves. were quite different. Before any conclusions are drawn Trichanthera gigantea is typically reported as having a about appropriate levels of T. gigantea leaves to incorporate moderate to low palatability because of the hirsute nature in pellets, longer-term feeding studies with animals where of its leaves (Mejía and Vargas 1993); however the TPI liveweight gains are recorded need to be conducted. was generally high for all pellet treatments except T100. This may be as a result of the pelleting process, which is Acknowledgments often associated with higher levels of palatability and the presentation of a more favorable form of the feed (Wallace The authors are thankful to the Caribbean Agricultural et al. 1961; Dobie 1975). For instance, the smaller unit Research and Development Institute (CARDI) and the size of pellets makes it more prehensile and easier to School of Agriculture and Environment (SAE), Massey ingest compared with the bulkier form of unprocessed University for providing the funding support required forage. This smaller denser form of feed is also associated to undertake the research. The authors are also thankful with more rapid flow of feed through the gastro-intestinal to the University of Trinidad and Tobago (UTT) for tract resulting in characteristically higher intakes when donating animals, granting access to housing facilities, compared with bulkier unprocessed forage (Blaxter and forage banks and the provision of technical support and Graham 1956; Minson 1963). In addition, the pelleting access to laboratory equipment required to conduct the process involves the drying, grinding, mixing and studies. The donation of animals and forage from the compression of leaves with more favorable ingredients, Trinidad and Tobago Goat and Sheep Society is also which is often associated with reduced selection and gratefully acknowledged. increased intake (Wanapat et al. 2013). There are no current studies on the impact of pelleting on the intake References T. gigantea leaves in small ruminants; however according to Beardsley (1964), pelleting can increase intake of (Note of the editors: All hyperlinks were verified 26 August 2021). forage feeds by up to 25%. Therefore, pelleting may provide an opportunity for improving the intake of and AOAC. 2000. Official Methods of Analysis. 17th Edn. Association therefore performance on T. gigantea. of Official Analytical Chemists, Gaithersburg, MD.Barahona R. 1999. Condensed tannins in tropical forage Daily CP intake by the Control group barely satisfied legumes: their characterisation and study of their the CP requirement for finishing lambs (4‒7 months of nutritional impact from the standpoint of structure-activity age) weighing 30 kg and growing at a daily rate of 295 relationships. Ph.D. Thesis. University of Reading, UK. g/d (191 g CP/d) (NRC 1985), while those of groups fed bit.ly/3myY4k3 pellets containing T. gigantea leaf would not support Beardsley DW. 1964. Symposium on forage utilization: gains of this magnitude. According to Hoover and Miller Nutritive value of forage as affected by physical form. (1996) the amount of soluble protein (SP), that fraction Part II. Beef cattle and sheep studies. Journal of Animal of the rumen degradable protein that is immediately Science 23(1):239–245. doi: 10.2527/jas1964.231239x available for utilization by rumen microbes, should Benton W; Benton H. 1963. Encyclopaedia Britannica. represent about 35% of feed protein in order to optimize University of Chicago Press.Blaxter KL; Graham NM. 1956. The effect of the grinding rumen function. SP in this study was less than 35% of and cubing process on the utilization of the energy of dried Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Pellets with Trichanthera leaves in the Caribbean 389 grass. The Journal of Agricultural Science 47(2):207–217. Mejía CE; Vargas JE. 1993. Análisis de selectividad de ovejas doi: 10.1017/S0021859600040132 africanas con cuatro tipos de forrajes. Livestock Research for Charlton JFL; Douglas GB; Wills BJ; Prebble JE. 2003. Farmer Rural Development 5, Article #22. lrrd.org/lrrd5/3/mejia.htm experience with tree fodder. Using trees on farms. Grassland Min B; Barry T; Attwood G; McNabb W. 2003. The effect research and practice series 10:7–16. doi: 10.33584/rps.10. of condensed tannins on the nutrition and health of 2003.2989 ruminants fed fresh temperate forages: a review. Animal De Mendiburu F. 2019. Package ‘agricolae’. R package Feed Science and Technology 106(1–4):3–19. doi: 10.1016/ version, 1-2. cran.r-project.org/package=agricolae S0377-8401(03)00041-5 Dobie JB. 1975. Cubing tests with grass forages and similar Minson D. 1963. The effect of pelleting and wafering on the roughage sources. Transactions of the ASAE 18(5):864– feeding value of roughage—a review. Grass and Forage 866. doi: 10.13031/2013.36697 Science 18(1):39–44. doi: 10.1111/j.1365-2494.1963.tb00 Edwards A; Mlambo V; Lallo CHO; Garcia GW; Diptee MD. 324.x 2012. In vitro ruminal fermentation parameters of tanner National Forage Testing Association. 2002. Method 2.2. 2.5: grass (Brachiaria arrecta) supplemented with leaves Dry matter by oven drying for 3hr at 105° C. National from three forage trees. Livestock Research for Rural Forage Testing Association, Omaha, NE, USA. Development 24, Article #102. lrrd.cipav.org.co/lrrd24/6/ NRC (National Research Council). 1985. Nutrient edwa24102.htm requirements of dairy cattle 6. National Academy Press, Fox J; Weisberg S. 2011. An R companion to applied Washington, DC, USA. regression. 2nd Edn. SAGE Publications, Thousand Oaks, Pinheiro J; Bates D; DebRoy S; Sarkar D. 2018. nlme: linear CA, USA. bit.ly/2WrdyeO and nonlinear mixed effects models. (Version 3.1-137). Gemeda BS; Hassen A. 2014. In vitro fermentation, cran.r-project.org/package=nlme digestibility and methane production of tropical perennial Reed JD. 1995. Nutritional toxicology of tannins and related grass species. Crop and Pasture Science 65(5):479–488. polyphenols in forage legumes. Journal of Animal Science doi: 10.1071/CP13450 73(5):1516–1528. doi: 10.2527/1995.7351516x Heuzé V; Tran G; Boudon A; Bastianelli D. 2017. Nacedero Rosales M. 1997. Trichanthera gigantea (Humboldt & (Trichanthera gigantea). Feedipedia, a programme by Bonpland.) Nees: A review. Livestock Research for Rural INRA, CIRAD, AFZ and FAO. Last updated on June 26, Development 9, Article #37. lrrd.org/lrrd9/4/mauro942.htm 2017. feedipedia.org/node/7270 Rosales M; Rios CI. 1999. Research into variation in nutritive Hoover WH; Miller TK. 1996. Feeding for maximum rumen value of provenances of Trichanthera gigantea. Proceedings function. Mid-South Ruminant Nutrition Conference of the FAO Electronic Conference on Agroforestry for Proceedings, Texas Animal Nutrition Council. p. 33–46. animal production in Latin America. (In Spanish). fao.org/ bit.ly/3topAlI livestock/agap/frg/agrofor1/rosale17.htm Hothorn T; Bretz F; Westfall P; Heiberger RM; Schuetzenmeister Sampaio CB; Detmann E; Lazzarini I; Souza MA de; Paulino A; Scheibe S. 2016. Package ‘multcomp’. Simultaneous MF; Valadares Filho SC. 2009. Rumen dynamics of neutral inference in general parametric models (Version 1.4-6). detergent fiber in cattle fed low-quality tropical forage cran.r-project.org/package=multcomp and supplemented with nitrogenous compounds. Revista Lallo CHO; Smalling S; Facey A; Hughes M. 2017. The Brasileira de Zootecnia 38(3):560–569. doi: 10.1590/ impact of climate change on small ruminant performance S1516-35982009000300023 in Caribbean communities. In: Ganpat W; Isaac W, Schnaider MA; Ribeiro-Filho HMN; Kozloski GV; Reiter T; eds. Environmental Sustainability and Climate Change Orsoletta ACD; Dallabrida AL. 2014. Intake and digestion Adaptation Strategies p. 296–321. doi: 10.4018/978-1- of wethers fed with dwarf elephant grass hay with or 5225-1607-1.ch011 without the inclusion of peanut hay. Tropical Animal Health Lazzarini I; Detmann E; Sampaio CB; Paulino MF; Valadares and Production 46(6):975–980. doi: 10.1007/s11250-014 Filho SC; Souza MA de; Oliveira FA. 2009. Intake and -0594-5 digestibility in cattle fed low-quality tropical forage and R Core Team (2020). R: A language and environment supplemented with nitrogenous compounds. Revista for statistical computing. R Foundation for Statistical Brasileira de Zootecnia 38(10):2021–2030. doi: 10.1590/ Computing, Vienna, Austria. URL. r-project.org. S1516-35982009001000024 Van Soest PV; Robertson JB; Lewis BA. 1991. Methods Lenth RV; Buerkner P; Herve M; Love J; Riebl H; Singmann for dietary fiber, neutral detergent fiber, and nonstarch H. 2019. emmeans: Estimated Marginal Means, aka polysaccharides in relation to animal nutrition. Journal of Least-Squares Means (Version 1.3.4). cran.r-project.org/ Dairy Science 74(10):3583–3597. doi: 10.3168/jds.S0022- package=emmeans 0302(91)78551-2 McDowell LR; Arthington JD. 2005. Minerals for grazing Wallace JD; Raleigh RJ; Sawyer WA. 1961. Utilization of ruminants in tropical regions. 4th Edn. University of Florida, chopped, wafered, and pelleted native meadow hay by Gainesville, FL, USA. weaned Hereford calves. Journal of Animal Science Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 390 H.A. Jack, L.M. Cranston, J.L. Burke, M. Knights, and P.C.H. Morel 20(4):778–781. doi: 10.2527/jas1961.204778x enhance rumen fermentation and ruminant production in Wanapat M. 2009. Potential uses of local feed resources the tropics. Journal of Animal Science and Biotechnology for ruminants. Tropical Animal Health and Production 4:32. doi: 10.1186/2049-1891-4-32 41:1035. doi: 10.1007/s11250-008-9270-y Wright SJ; Cornejo FH. 1990. Seasonal drought and leaf Wanapat M; Kang S; Polyorach S. 2013. Development of fall in a tropical forest. Ecology 71(3):1165–1175. doi: feeding systems and strategies of supplementation to 10.2307/1937384 (Received for publication 3 November 2020; accepted 13 August 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Tropical Grasslands-Forrajes Tropicales (2021) Vol. 9(3):391–396 391 doi: 10.17138/TGFT(9)391-396 Short Communication Effects of plant spacing and fertilizer level on forage yield and chemical composition of hybrid Urochloa cv. Mulato II grass during the first 150 days of growth under irrigation supplementation, in Chagni Ranch, Awi Zone, Ethiopia Efectos del espaciamiento de las plantas y el nivel de fertilizante sobre el rendimiento del forraje y la composición química de la gramínea híbrida Urochloa cv. Mulato II durante los primeros 150 días de crecimiento bajo riego suplementario, en Chagni Ranch, Awi Zone, Etiopía WONDIMAGEGN TADESSE1, BERHANU ALEMU2 AND MESGANAW ADDIS2 1Department of Animal Production and Technology, College of Dry Land Agriculture, Kebri Dehar University, Kebri Dehar, Ethiopia. kdu.edu.et 2Department of Animal Science, College of Agriculture and Natural Resources, Debre Markos University, Debre Markos, Ethiopia. dmu.edu.et Abstract A study was conducted to evaluate the effects of plant spacing and N fertilizer application on dry matter yield and chemical composition of Urochloa hybrid cv. Mulato II grass for the first 150 days after planting. A factorial experiment with 3 urea fertilizer levels (0, 50 and 100 kg/ha) and 4 spacings between plants and rows (20 × 20, 30 × 40, 40 × 60 and 50 × 80 cm) with 3 replications was used. Data collected were dry matter yield (DMY), leaf:stem ratio and chemical analyses, i.e. crude protein (CP), ash, neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL) concentrations. Results indicated that DMY, leaf:stem ratio, CP%, NDF% and ADF% were significantly (P<0.05) affected by interactions between plant spacing and fertilizer level. However, ash and ADL were significantly (P<0.05) affected only by main effects. The highest DMYs (9.18 t/ha and 8.93 t/ha) were recorded for narrowest plant spacing (20 × 20 cm) with higher urea fertilizer level (100 kg/ha) and narrowest plant spacing (20 × 20 cm) with medium urea fertilizer level (50 kg/ha), respectively. CP% ranged from 14.6 to 20% and leaf:stem ratio from 1.12 to 1.82:1. Similar studies need to be conducted over longer periods to determine to what extent these findings relate to performance over the life of a permanent pasture. Keywords: Dry matter yield, N fertilizer, nutrient composition, spacing, urea. Resumen Se realizó un estudio para evaluar los efectos del espaciamiento de las plantas y la aplicación de fertilizantes nitrogenados sobre el rendimiento de materia seca y la composición química del híbrido Urochloa cv. Pasto Mulato II durante los primeros 150 días después de la siembra. Se utilizó un experimento factorial con 3 niveles de fertilizante de urea (0, 50 y 100 kg / ha) y 4 espaciamientos entre plantas y surcos (20 × 20, 30 × 40, 40 × 60 y 50 × 80 cm) con 3 repeticiones. Los datos recopilados fueron el rendimiento de materia seca (DMY), la relación hoja: tallo y los análisis químicos, como las concentraciones de proteína cruda (CP), ceniza, fibra detergente neutra (NDF), fibra detergente ácida (ADF) Correspondence: W. Tadesse, Department of Animal Production and Technology, College of Dry Land Agriculture, Kebri Dehar University, Kebri Dehar, Ethiopia. Email: wondimagegntadesse2011@gmail.com Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 392 W. Tadesse, B. Alemu, and M. Addis y lignina detergente ácida (ADL). Los resultados indicaron que el DMY, la relación hoja: tallo, CP%, NDF% y ADF% se vieron afectados significativamente (P <0.05) por las interacciones entre el espaciamiento de las plantas y el nivel de fertilizante. Sin embargo, las cenizas y las ADL se vieron afectadas significativamente (P <0.05) solo por los efectos principales. Los DMY más altos (9.18 t/ha y 8.93 t/ha) se registraron para el espaciamiento de plantas más estrecho (20 × 20 cm) con un nivel de fertilizante de urea más alto (100 kg/ha) y el espaciamiento de plantas más estrecho (20 × 20 cm) con medio nivel de fertilizante de urea (50 kg/ha), respectivamente. El porcentaje de CP varió de 14.6 a 20% y la relación hoja: tallo de 1.12 a 1.82:1. Es necesario realizar estudios similares durante períodos más prolongados para determinar en qué medida estos hallazgos se relacionan con el rendimiento durante la vida de una pastura permanente. Palabras clave: Composición de nutrientes, espaciado, fertilizante N, rendimiento de materia seca, urea. Introduction chemical composition of grass pasture including crude protein (CP) concentration and digestibility, increases Livestock are an important component of nearly all in which improve livestock production (Marques et al. farming systems in Ethiopia, providing milk, meat, 2017). Nevertheless, information regarding the effects draught power, transport, manure, hides and skins and of fertilizer levels and plant spacing on biomass yield serve as a source of cash income (Funk et al. 2012). The and chemical composition of Mulato II grass is scarce subsector contributes about 16.5% of the national Gross in our country and specifically in the study area. We Domestic Product (GDP) and 35.6% of the Agricultural hypothesized that planting the grass more densely and GDP. It also contributes 15% of export earnings and 30% fertilizing with N would produce more dry matter (DM) of agricultural employment. The livestock subsector more rapidly than when planted at wide spacing and not currently supports and sustains livelihoods for 80% of the fertilizing. We conducted the present study in order to total rural population (Leta and Mesele 2014). Despite generate information on yield and chemical composition the importance of livestock in the country, productivity of Mulato II grass during the first 150 days at different is low (Gebremariam et al. 2010). One of the major plant spacings with different rates of nitrogen fertilizer. constraints leading to such low productivity is shortage of feed in terms of both quantity and quality, especially Materials and Methods during the dry season (Hassen et al. 2010), combined with high feed prices (Gebremariam et al. 2010). Description of the study area In order to solve the shortage of feed and increase livestock production, introduction and cultivation of The experiment was conducted in Chagni Ranch, high-quality forages with high yielding ability and Guangua Woreda, Awi Zone, Amhara National Regional adaptation to the biotic and abiotic environmental State, Ethiopia (10°57 ′N, 36°30′ E; 1,583 masl). The stresses may be an option (Kahindi et al. 2007). Improved area has average annual rainfall of 1,689 mm and mean grasses, many of African origin, have greater palatability minimum and maximum annual temperatures of 23 °C and productivity than other indigenous species and are and 30 °C, respectively. therefore desirable additions to pastures and common grazing areas (Mengistu 2002). Among the improved Experimental layout, Design and Treatments forage crops introduced into Ethiopia, Mulato II grass, which is the result of crosses of Urochloa ruziziensis, The study was conducted using a 3 × 4 factorial U. brizantha and U. decumbens, is claimed to have the arrangement in a randomized complete block design capacity to provide a significant amount of quality forage (RCBD) with 3 replications. The factors were 3 levels of (CIAT 2006). urea fertilizer (0, 50 and 100 kg/ha) and 4 spacings (20 The optimization of production and nutritive value × 20, 30 × 40, 40 × 60 and 50 × 80 cm; S1, S2, S3 and of grass can be achieved by planting on fertile soils and S4, respectively) between plants and rows, respectively, utilizing forage management tools such as plant spacing giving 12 treatment combinations and 36 experimental and utilizing when at high nutritive value (Yiberkew plots. Control treatment was regarded as the unfertilized et al. 2020). Nitrogen (N) fertilizer application is a treatment at each plant spacing. common practice, since this nutrient is found to be Each plot was 3 × 3.2 m and the total experimental one of the most limiting factors influencing yield and area was 12.6 × 41.5 m (522.9 m2). The spacings between Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Establishing Mulato II in Amhara, Ethiopia 393 plots and replications were 0.5 and 1.5 m, respectively. preserved in plastic bags pending analysis at Debre Treatments were randomly assigned to plots within each Birhan Agricultural Research Center Animal Nutrition replication. Laboratory. Ash and N were determined according to Soil samples were taken by auger from the center and the procedures described by AOAC (1990) and neutral corners of the experimental site prior to planting to a depth detergent fiber (NDF), acid detergent fiber (ADF) and of 15 cm and analyses revealed the following: organic acid detergent lignin (ADL) according to the procedures matter (OM) – 5.88%; organic carbon (OC) – 3.41%; total described by Van Soest (1985). N – 0.30%; available P – 4 ppm; and pH – 5.6. Statistical analysis Land preparation and Experimental management Data were subjected to analysis of variance (ANOVA) Land was oxen-ploughed and harrowing and bed using the General Linear Model (GLM) procedure of preparation were carried out before planting manually. the Statistical Analysis System (SAS 2007). Differences Root splits of Mulato II grass were collected from Finota among treatment means were determined using Duncan’s Selam grass nursery site at an age of 7 months regrowth Multiple Range Test (DMRT) at P<0.05. The statistical and planted at the experimental site on 6 September 2017. model used was: Urea was applied by split application with half applied Yijk = μ + Bi + Fj + Sk + (FS)jk + eijk, where: at planting and the remainder at 30 days after planting Yijk = the response variable; with different levels based on treatment. Weeding was μ = overall mean; done manually during the experimental period. The Bi = i th block effect; experiment was irrigated once a week when rain was F = jthj main factor effect (fertilizer level); limited, with precautions taken to avoid contamination S = kthk main factor effect (spacing); of treatments by cross flooding. (FS)jk = jk th interaction effect (fertilizer level × spacing); and Sample collection and Dry matter yield determination eijk = random error. Data on dry matter yield (DMY) and chemical Results composition of Mulato II grass were recorded at harvesting time, 150 days after planting. On 6 February Overall, there were significant interactions between the 2018, leaf:stem ratio was determined from 10 randomly effects of the main treatment variables (plant spacing selected plants in each plot by separating leaf and stem and urea level) on DMY, leaf:stem ratio and chemical portions, air-drying the leaves and stems and weighing composition. separately. DMY per plot was determined by hand-harvesting Dry matter yield and leaf:stem ratio plants in inner rows, i.e. excluding border rows, with sampling areas of 7.28 m2 for S1, 5.76 m2 for S2, 4.4 m2 DMY per hectare was significantly (P<0.01) affected for S3 and 2.4 m2 for S4 at a height of 5 cm from ground by both plant spacing and urea fertilizer level (Table 1). level. Fresh weight of forage was measured immediately Increasing plant spacing reduced DMY of forage at all after harvesting, before the forage was thoroughly mixed fertilizer levels (P<0.05) and urea application increased and a 0.5 kg fresh subsample was taken from each sample DMY at all plant spacings but differences were significant for DMY determination. The samples were oven-dried (P<0.05) at only the narrowest plant spacing. Highest and DMY/ha was calculated. yields were obtained at the narrowest plant spacing with urea applied (P<0.05). Chemical analysis of forage Leaf:stem ratio was increased by plant spacing at all fertilizer levels but differences were significant for only Following mixing of the forage a second 0.5 kg fresh the unfertilized Control and 100 kg urea/ha treatments subsample was taken from each plot for chemical (P<0.05; Table 1). Similarly, urea application increased analysis and dried in a forced-draft oven at a temperature leaf:stem ratio at all plant spacings but differences were of 105 ºC for 24 hours. The dried material was ground significant (P<0.05) for only the wider 2 spacings (S3 to pass through a 1 mm sieve for chemical analysis and and S4). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 394 W. Tadesse, B. Alemu, and M. Addis Table 1. Total dry matter yield and leaf:stem ratio of Both NDF and ADF concentrations declined as Urochloa Mulato II at 150 days after planting as influenced by fertilizer level increased but differences were significant combinations of urea fertilizer level and plant spacing. (P<0.05) for only the narrower 2 plant spacings (Table Fertilizer Plant spacing 2). Similarly, NDF% declined as plant spacing increased level S1 S2 S3 S4 but differences were significant (P<0.05) in the Total dry matter yield (t/ha) unfertilized Control treatment only. ADF% also declined F1 6.5b 5.17bcd 4.22de 2.91e F2 8.93a 5.68bc 4.51cd 4.01de as plant spacing increased but there were no consistent F3 9.18a 6.46b 4.61cd 3.80e significant differences. Overall trends were for highest Leaf:stem ratio values for both NDF and ADF concentrations to occur in F1 1.12e 1.37bcde 1.16de 1.42bcd the Control (unfertilized) at the narrowest plant spacing F2 1.32bcde 1.27cde 1.38bcde 1.48bc and the lowest values with the higher urea level at the F3 1.38bcde 1.44bcd 1.59ab 1.82a widest plant spacing. S1 = 20 × 20 cm; S2 = 30 × 40 cm; S3 = 40 × 60 cm; and S4 = 50 × 80 cm spacing between plants and rows; F1= 0 kg urea/ Discussion ha; F2 = 50 kg urea/ha; and F3 = 100 kg urea/ha. Means for different treatments with different letters are The higher DMY at narrower spacing with application of significantly different (P<0.05). N fertilizer were to be expected as plant population was greater, plants were taller and soil fertility was improved Chemical composition with application of urea fertilizer. A combination of increased tiller numbers and number of leaves per plant The significant effects of N fertilizer level, plant spacing could have contributed to increased photosynthetic and their interactions on crude protein percentage (CP) are activity and hence higher dry matter production (Damry indicated in Table 2. CP concentration increased (P<0.05) et al. 2009). Those authors reported that increasing with increase in row spacing at all fertilizer levels and level of urea fertilizer application increased Mulato urea application increased CP% at all plant spacings but tiller numbers and DM production plus CP and NDF differences were significant (P<0.05) at only the narrowest concentrations. Responses to urea in this study also and widest spacings. Highest CP% (20.0%) was recorded confirm those of CIAT (2006) that Urochloa Mulato II where 100 kg urea/ha was applied at the widest plant is highly responsive to N fertilizer. Similarly, Bouathong spacing and the lowest (14.6%) for the unfertilized Control et al. (2011) reported a trend for hybrid Urochloa grass treatment at the narrowest plant spacing. yield components to increment as the level of N fertilizer Table 2. CP, NDF and ADF concentrations of Urochloa Mulato application increased with no significant benefit of II at 150 days after planting as affected by combinations of adding N at levels above 40 kg/ha. Similarly, Marques urea fertilizer level and plant spacing. et al. (2017) reported DMY of Mulato II significantly Spacing Fertilizer level increasing with increasing rates of N fertilizer. Yiberkew S1 S2 S3 S4 et al. (2020) reported that DMY of Mulato II hybrid was Crude protein (%) significantly affected by plant spacing where yields from F1 14.6h 16.6fg 17.7cdefg 18.5bcd spacings of 15 × 50 and 30 × 50 exceeded those at 45 × 50 F2 16.2g 16.6fg 18.3bcde 19.5ab cm at 3 months after planting. Buamool and Phakamas F3 17.2defg 18.1bcdef 19.1abc 20.0a (2018) also reported that DMY of tropical grasses like Neutral detergent fiber (%) Mulato II, Ruzi grass (Urochloa ruziziensis), Purple F1 52.8a 51.0ab 48.3abc 45.5bcde guinea (Megathyrsus maximus TD 58) and Mombasa F2 48.5abc 47.8bcd 43.6cde 46.2bcde F3 45.1cde 42.1e 46.6bcde 42.3de guinea (Megathyrsus maximus cv. Mombasa) were Acid detergent fiber (%) higher following urea application than with ammonium F1 39.3a 38.2ab 36.3bcd 34.4bcde sulphate or non-fertilized grasses. F2 36.9abc 36.3abcd 30.8de 33.3bcde Leaves are a good nutritional quality parameter for F3 33.7bcde 32.1cde 31.0de 30.1e forage grass species. The application of N fertilizer S1 = 20 × 20 cm; S2 = 30 × 40 cm; S3 = 40 × 60 cm; and S4 = increases soil fertility sufficiently to produce more leaves 50 × 80 cm spacing between plants and rows; F1= 0 kg urea/ and make the plant grow vigorously. In addition, at the ha; F2 = 50 kg urea/ha; and F3 = 100 kg urea/ha. wider spacings, the plants receive more light, which could Means for different treatments with different letters are be used for leaf formation but in grass grown at narrower significantly different (P<0.05). Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) Establishing Mulato II in Amhara, Ethiopia 395 spacings there could be shading effects, resulting in the feed intake, increased rumination time and decreased formation of fewer lateral shoots. Wider spacing reduces conversion efficiency of metabolizable energy (Reed and competition for light, nutrients and moisture so plants Goe1989). All forage produced in this study had NDF can grow more vigorously, which is stimulated by N concentrations below the critical value of 60%. application. Relatively lower leaf:stem ratios were The current results revealed that DMY of Mulato II recorded for narrower spacings due to competition during the first 150 days of growth can be improved by among plants, which resulted in increased stem growth urea fertilizer application. More specifically application (early maturity) rather than leaf development. These of 50 kg urea/ha produced good responses in both DMY results support the findings of Yiberkew et al. (2020) and CP concentration. While higher urea levels produced that leaf:stem ratio was higher at wider spacing (45 × 50 further increases in both parameters, the financial return cm) than at intermediate (30 × 50 cm) and narrow plant was unlikely to be positive given the lower response with spacing (15 × 50 cm) (1.39, 1.1 and 0.97, respectively). the extra amount of fertilizer applied. Similarly closer While Mulato II showed good response to N spacing of plants, i.e. closer planting, resulted in higher fertilizer, depending on the level of soil fertility, one yields of forage at reduced CP%. or more maintenance applications may be required to It must be remembered that this study covered only maintain high yields of good quality forage. Nutritive the first 150 days of growth, which is a very short value of forage, i.e. concentrations of CP, ADF, NDF time in the life of a perennial pasture. While narrow and digestibility, depends on soil fertility and stage of spacing resulted in additional forage growth in this early maturity. Planting Mulato II at wider plant spacings with stage, it would be expected that, as the stand matured, higher N fertilizer level produced excellent nutritional differences in yield between different spacings would value, particularly CP%, which is often a limiting nutrient disappear. Future studies should be continued for at least in forages. Forage produced at all plant spacings with urea 2 years to test this, while a range of harvest frequencies fertilizer application had CP concentration well above and maintenance fertilizer levels should be examined. the level required for effective rumen function (7.5%) Repeating these studies on a range of soil types and (Van Soest 1982) and for lactation (15%) (Norton 1982). under differing environmental conditions will determine This is a clear indication of the value of this particular how applicable the results are over a range of conditions, grass as a forage for livestock. Factors contributing to the while the true benefits of growing this grass will only higher CP percentage at higher fertilizer level and wider be known when performance of animals consuming the plant spacing would be higher N uptake by individual forage is assessed. plants, plus enhancement of leafiness and leaf:stem ratio of grass. This would agree with findings of Marques et References al. (2017), who reported that application of N produced a significant increase in forage production and a linear (Note of the editors: All hyperlinks were verified 31 August 2021). increase in CP% of Mulato II. NDF and ADF concentrations all declined as N AOAC. 1990. Official methods of analysis of Association fertilizer level and spacing increased. This would be of Official Analytical Chemists. 15th Edn. AOAC, a function of reduced stem percentage and reduced Washington, DC.Bouathong C; Hare M; Losirikul M; Wongpichet K. 2011. lignification at the wider spacing and with greater N Effect of nitrogen rates on plant growth, seed yield and availability in fertilized treatments. At lower fertilizer seed quality of three lines of Brachiaria hybrid grass. levels and narrower plant spacing, competition Khon Kaen Agricultural Journal 39:295-306. between plants for resources forces plants to prioritize Buamool P; Phakamas N. 2018. Effects of different forms development of structural components to cope with the of nitrogen fertilizer on growth and yield of four tropical environmental conditions. The current result is in line pasture grasses. International Journal of Agricultural with Marques et al. (2017), who reported that higher Technology 14(7):1065-1076. bit.ly/38SnxwK fertilizer doses led to higher protein and lower NDF and CIAT (Centro Internacional de Agricultura Tropical). ADF concentrations in Mulato II, which should increase 2006. CIAT Annual Report 2005. Project IP-5. Tropical nutritive value of grass and its intake. Increase in cell Grasses and Legumes: optimizing genetic diversity for multipurpose use. CIAT, Cali, Colombia. bit.ly/3tuFGKG wall contents (NDF) is a very important limiting factor in Damry; Marsetyo; Syukur SHj. 2009. Growth, production and terms of nutritive value of feeds (Van Soest 1985). NDF nutritive value of Brachiaria mulato as affected by levels values greater than 60% result in decreased voluntary of urea fertilization. Proceedings of the 1st international Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) 396 W. Tadesse, B. Alemu, and M. Addis seminar on animal industry, Faculty of Animal Science, and chemical composition of hybrid Brachiaria cv. Bogor Agricultural University. repository.ipb.ac.id/ Mulato II under system of cuts and nitrogen fertilization. handle/123456789/33826 Biological Science Journal 33:685-696. doi: 10.14393/BJ- Funk C; Rowland J; Eilerts G; Kebebe E; Biru N; White L; v33n3-32956 Galu G. 2012. A climate trend analysis of Ethiopia. Fact Mengistu A. 2002. Forage Production in Ethiopia: a case study Sheet 2012–3053. U.S. Geological Survey, Reston, VA, with implications for livestock production. Ethiopian USA. doi: 10.3133/fs20123053 Society of Animal Production, Addis Ababa, Ethiopia. hdl. Gebremariam S; Amare S; Baker D; Solomon A. 2010. handle.net/123456789/3432 Diagnostic study of live cattle and beef production and Norton BW. 1982. Differences between species in forage marketing. Constraints and opportunities for enhancing the quality. In: Hacker JB, ed. Nutritional limits to animal system. Bill & Melinda Gates Foundation and Ethiopian production from pastures. Commonwealth Agricultural Agricultural Transformation Agency, Seattle, WA, USA Bureaux, Farnham Royal, UK. p. 89–110. and Addis Ababa, Ethiopia. hdl.handle.net/10568/24698 Reed JD; Goe MR. 1989. Estimating the nutritive value of Hassen A; Ebro A; Kurtu M; Treydte AC. 2010. Livestock cereal crop residues: Implications for developing feeding feed resources utilization and management as influenced standards for draught animals. ILCA Bulletin No. 4. ILCA, by altitude in the Central Highlands of Ethiopia. Livestock Addis Ababa, Ethiopia. hdl.handle.net/10568/4610 Research for Rural Development 22, Article#229 lrrd. SAS (Statistical Analysis System). 2007. Statistical Analysis cipav.org.co/lrrd22/12/hass22229.htm System software. Version 9.0. SAS Institute Inc., Cary, Kahindi RK; Abdulrazak SA; Muinga RW. 2007. Effect of NC, USA. supplementing Napier grass (Pennisetum purpureum) Van Soest PJ. 1982. Nutritional ecology of the ruminant. with Madras thorn (Pithecellobium dulce) on intake, Cornell University Press, Ithaca, NY, USA. jstor.org/stable/ digestibility and live weight gains of growing goats. 10.7591/j.ctv5rf668 Small Ruminant Research 69:83–87. doi: 10.1016/j. Van Soest PJ. 1985. Analysis of forages and fibrous foods. smallrumres.2005.12.008 Cornell University Press, Ithaca, NY, USA. Leta S; Mesele F. 2014. Spatial analysis of cattle and shoat Yiberkew N; Mekuriaw Y; Asmare B. 2020. Effects of fertilizer population in Ethiopia: growth trend, distribution and types and plant spacings on plant morphology, biomass market access. SpringerPlus 3:310. doi: 10.1186/2193- yield and chemical composition of Brachiaria hybrid 1801-3-310 Mulato II grass grown in lowlands of Northwest Ethiopia. Marques DL; França AFS; Oliveira LG; Arnhold E; Ferreira Scientific Papers: Animal Science and Biotechnologies RN; Correa DS; Bastos DC; Brunes LC. 2017. Production 53(1):20-35. bit.ly/2YzdD0Z (Received for publication 23 December 2019; accepted 19 August 2021; published 30 September 2021) © 2021 Tropical Grasslands-Forrajes Tropicales is an open-access journal published by International Center for Tropical Agriculture (CIAT), in association with Chinese Academy of Tropical Agricultural Sciences (CATAS). This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775) TGFT Editorial Team A.A. 6713, Km 17 Recta Cali-Palmira, Cali, Valle del Cauca, Colombia. Phone: +57 2 4450100 Ext. 3084 Email: CIAT-TGFT-Journal@cgiar.org