S"R t 1 . C3 5 1 1 975 The lnternational Exchange and Testing of CASSAVA GERM PLASM 1n A FRICA Proceedings of an interdisciplinary workshop held at liTA, lbadan, Nigeria 17-21 November 1975 Editors· Eugene Terry and Reginald Mac l ntyre 1 ¡\ 1 r-A ~; .JE Cosponsored by tbe lnternational Development Researc !~ Centre andthe lnternational lnstitute of Tropical Agriculture IDRC -063e ~!3 el J 1 ·C~ .7 51 19 1~ - ¡/ The International Exchange and Testing of Cassava Germ Plasm in Africa o / Proceedings pf an lnterdisciplinary Workshop held a t liTA, lbadan, Nigeria , 17-2 1 November 1975 o Editors: Eugene R. Terry and Reginald Maclntyre 42675 Cosponsored by the lnternational Development Research Centre and the lnternational lnstitute of Tropical Agriculture ( ¿ ! ,' ' .:? . t 1 t'? ( / l e--·~/- ·"i( ,f/tf/ 1 1 1 IDRC-063e t · 1' / - -) , ; ~· .. ; / , .. .! t Contents Foreword 5 Participants 9 Welcoming Address, S. K . Hahn 11 Theme Papen Possibilities fur economic research intu cas,ava productiun systcms in Africa. J . C. Flinn 15 lmprovcment of cassava at the lntcrnational lnstitute of Tropical Agriculture. S. K. Hahn 21 eassava bacteria! blight in Africa. E. R. Tcrry 23 Advance; in re~earc h un the economic significancc of thc grcen cassava mitc (!\1onom·chef/us lttllt(ioa) in Uganda. Z. M. Nyiira 27 CoulllrY Prese111lltio11s 31 Summary of General Di~cussion J5 Appendix 1 Agronomic aspccts of the lnternational cxchange and testing of cassava germ plasm Part A euoperat ive tcsting and 'elec tion 39 Part B Germinatiun and pollination 41 Part e A rapid mu lt iplication techn iquc. A. K. Huwland 42 Pan D Guidelines for the establ is hment uf a ca;,ava improvement project: thc Zaú·c modcl. H. C. Ezumah. S. Kabonyi. K. Beya 45 Appcndix 2 Phytosa nitary aspect~ uf the international exchange and testing of cassava gern1 p lasrn Part A Suggestcd guidel inc~ rclati ng to thc intcrnational movemcnt of cassava planting materiab 5 1 Pan B Description and cvaluation of ca,sava mosaic diseasc in Africa, E. R. Terry 53 Part e Major pests of cassava in A frica and preliminary g uidelines for screen ing of re~istance. K. Leuschner 55 Appendix 3 A note on the liTA training program 57 3 There is a wide range of variation in the cluJracteristics encountered with the major cassava germ plasm collections. The suggested ideal p/ant type in Africa is one tluJt carries its branches on abouttwo metres of stem. 4 Foreword This is the ninth IDRC rcport on cassava and the seventh recording the proceedings of a mullidi sc iplinary workshop. Previous reports are listcd al the end of this Foreword. A s imilar workshop held al C IAT in Palmira , Colombia. in February 1975. was designed to establish guidelines fo r the international exchange and testing of cassava gcrm plasm with particular refercnce lo Lalin America and Asia. Beca use o f \he severe phytosanitary constrai nts posed by thc presence of African cassava mosa ic disease in A frica the only representation from that continent atthe CIAT meeting was from liTA. This parlic ipation was planned with the idea of structuring a similar workshop for African cassava-producing countries and this report describes the proceedings of that workshop. The meeting ope ned with a presenta! ion from Ha/111 . the liT A Root Crop Program leader. who set the framework for the subsequent discussions and described how liT A perce ived its role in the develo pme nt o f an international exchangc and testing program for cassava germ plasm in Africa. During thc ncxt two days reprcscntatives from Sri Lanka (Rajaguru). the Camerouns (Eckbeil and Lyonga ). Ghana (Doku). Liberia (Henries). the Malagasy Republic (Rafiringa). Malawi (Edje and Chapo/a). Sierra Leone (Dahniya). Tanzania (Msabaha), Togo (Adam). Uganda (Nyiira). and za·ire (Kabeya) prescnted situation papers relating to cassava produc!ion in their countries. Thc author of each presentation was requcsted to cover thc following points: 1) cassava gcrm plasm available in hi s country; 2) state o f the cvaluation ofthis gc rm plasm. including the charactcri stics being noted: 3) the area devoted to cassava and whether thi s was principally sing le- or mull iple-cropped and an estímate of whether this arca was like ly to increase in thc future; 4) the status of research on thc crop with a brief sum mary on the work in progress and plans for thc future; 5) the human and financ ia) rcsources clevotcd to research and cxtens ion on cassava: 6 ) thc na mes and instances ofthe most imporlanl cliseases and pests of the crop; and 7) the quarantine requirements for imporling cassava germ plasm. The informat ion prov icled by thc partic ipants is presen ted in tabular form in this report. Following each presentat ion there was ample time for qu~st ions and comments. These di scussion sessions brought out man y useful points which are outlined in the Summary of General Discussions. Thc participants made a tour ofthe liTA cassava fie lds and laboratories and spen t one da y v i ~iting a major cassava-growing area near Warri . more lhan 200 miles from lbadan. At this location liT A has a large-scalc cooperative program with bo th the S tate Government and a Shei i/ BP community devdopment program. Country presentations wcre summarized by Terrv after which four liTA staff members presentcd guidelinc documents relating to the tcsting aml selection of cassava varieties. The in itial paper by Hahn discusscd thc breeding aspects. followed by Terry who covered pathology. Leuschner entomology. and E::.umah agronomy. The next item on the age nda was a panel discussion relating tu quarantine regulations for cassava. The panel of Addoh. Aluko, Lozano. and Terr_v discussed this subject within the 5 context of the suggested guide lines that had been drawn up at the earl ier C IAT workshop. Following the panel diseussions. a subcommittee was appointcd to draw up reports on the agronomic and phytosanitary aspects of the international exehange and tcsting of cassava gcrm plasm. The subcommittee's repon was presentcd to the full works hop for review and adopt ion and is presented in this report as Append iccs 1-A and 2-A. The final da y ofthe workshop wa.~ given over toa panel discussion fol lowed by a series of special papers. T he panel discussion covered liT A 's T raining Program and was moderated by Reeves with Arin::.é, Adam. Ezeilo. Persley and Henries participating. Particular stress was placed in the discussion on thc role that LIT A might play in training seientists in root crop researc h. Flinn then presented a paper on thc possib ili ties for cconomic rcscan:h into ca~~ava production systems in Africa in wh ich he stressed the need for greater agrocconomic knowledge of thc crop in order to formu late souncl researeh programs. He was fo llow ed by Hahn who provided a description ofthe brecding program being u sed forcassava at liTA . Terry then deseribed cassava bacteria! bl ight. a d iseasc which wa~ previously considered unimportant, but in the last few years has beco me a major source of losses to cassava producer~ in A frica. The final paper by Nyiira al so dealt with a recent introduction, cas,ava green mi te . from the neo-tropics to Uganda less than fi ve years ago and which has already causcd scrious losses in both Uganda and ne ighbouring countrics. The final item of thc program \\a' a de monstration of the rapid multiplication teehnique being uscd for cas,ava at liTA by How/and. This report presents fi rst thc spceial papers followcd by atable ~ummarizing thc country presentations and the "Summary of the General Discussion'' on them. The latter part of thc report comprises three appendices: Appendix 1 deals wi th agronomic aspcct~ of the international excha ngc and testing of cassava germ plasm . This includes not only the report of the subcommitlee on cooperative test ing and sclcct ion but also provicles notes on germination and poll ination anda description of thc rapid multiplication tec hniquc u sed titutc of Tropical Agriculturc. P.:vt .B . 5320. lbadan. Nigcria L. Kabeya Agriculturc Expert. Departrncnt of Agricul tu re. P.O . Box 312. Kinshasa. Za'irc D. N. Kala barc MA NR Midwest. Agbaro-Warri . Midwest State of Nigeria Samuel J . Kamara Rice Research S tation. Sio.:rra Leone (Rokupr) Kla us Le uschner lntcrnational lnstitute o f Tropical Agriculture. P.M.B. 5320. lbadan. Nigcria J osé Carlos Lozano Plant Patho logi:.t. Centro Internacional de Agricultura Tropical. Apartado Aéreo 67-13. Cal i. Colombia 9 Simon Ngale Lyonga Chief of Division of Root and Cereal Crops. National Office for Scientific and Technical Research (ONAREST) . P.O. Box 2123. Yaounde. Cameroon Reginald Maclntyre Publications Coordinator. lntcrnational Dcvelopment Research Centre. P.O. Box 8500, Ottawa. Canada Eric C. MacKenzie Communauté Baptiste du Zai·rc Oucst. B.P. 4728. Kin~hasa 2. Za"ire M. A. M. Msabaha Ministry of Agriculture, A.R.I.-Ukiriguru Box 1433, Mwanza, Tanzania N'Guetta Bosso Director of Research, B.P. 8035. Abidjan. lvory Coast Z . M. Nyiira Kawanda Research Station. P.O. Box 765. Kampala. Uganda G . O. Obigbesan Dcpartment of Agronomy. Univcrsity of lbadan. lbadan. Nigcria P. U. Ohunyon Senior Extension Adviser. Shcii/BP. CPDW Box. Warri. Midwest S tate of Nigeria J. E. Opeke Federal Agricultura( Rescarch and Training Station. Umudike. Umuahia. East Central State. Nigeria Gabrielle Persley Research Fellow. lnternat ional lnstitute ofTropical Agriculturc. P.:vt.B . 5320. lbadan, Nigeria M. Ratiringa lngénieur d' Agriculture. B.P. 1690. Tananarive. République malgache A. S. B. Rajaguru Department of Animal Husbandry. Univer~it) of Sri Lanka. Peradeniya. Sri Lanka W . H. Reeves lnternationallnstitute ofTropical Agriculture. P.M.B. 5320. lbadan. Nigeria S. Sadik lnternational lnstitute of Tropical Agriculture. P.M .B. 5320. lbadan. Nigeria F. M. Shao Ministrv of Agriculture, A.R.I.-Ukiriguru Box 1433. Mwanza, Tanzania S. A. Shoyinka I. A.R. & T. , University of lfe. P.M. B. 5029. lbadan. Nigeria R . Soenarjo c/o J l . Merdeka 99. Bogor. Indonesia Eugene R. Terry lnternational lnst itute of Tropical Agriculturc. P .M. B. 5320. lbadan. Nigcria Jili E. Wilson lnternationallnstituteofTropical Agriculture. P.M.B. 5320. lbadan. Nigeria 10 Welcoming Address S. K. Hahn Imanationa/ Imtitllle l introdm:cd into thc A fr ican con ti ncnt atthc cm! of thc 16th ccntury. it ha' hccn '' idel) acccpted. gnm n. and u-,cd a-.nnc nf the mo't preferrcd -.tapie food~ in man) pan,ofthe contincnt. \\ 'orld production now amount~ lo ahout 100 million ton, annuall) from ahout 11 millinn hc~: tarc,. of whi.:h Afril.."a pnxlu~:cs 42 million ton~ from about 6 million hcc tarc,. Thi~ compri-.c., about -lO pcrn~nt of thc production and more than 50 perccnt nf thc arca dcv(>ted to th i' crop throughoutthc world . In A frica thc tn:nd in produrtion >hO\\ S a ~tcady inrrca'c. Ca,sava b potcntially ablc lO produce more food caloric~ per unit arca than any othcr food crup . nw ing to ih high yic lding ahility. adaptation to divcr~e dimatic ami ~:ultural conuitions. anu ahi lity to wrvive long H-6month~) dr)' \Ca,on\ . lt general!) require' k" labour. le'' co~t of production. and k" \."are in managementthan ~:.: real crop~. Both root anu lcaf are valuablc a~ human fo,•d and li' c'toclo. fecd anu the roo t i' w 1dcl) u sed for indu<,trial production of ~lan:h and alcohol. Somc externa) f<~~:tors could ha ve >igniticant influenc.:' on the role of l·a~sava in tropil·al Afril·a. Fir, t of al l. r.:ccnt high pril.."e of' ccrcab in the intcrnational marlo.ct' has enc(•uraged farmcr~ lo gnm more ca>sava. A' an cxamplc. hoth Sierra Lco ne and Libcria. w hich ha ve produl..".:tl and d.:pcmlcd mo~tly on rice. are d1anging from m<~jor rice produ~:tion lo general foou product ion w ith \."nn~ idcrahlc .:mpha'i' on cassava. Thc arrcagc of \."assava in hoth coun tric' ha, bccn notin:ahl) inrrea,ing in the last few ) car\ . Se\."ontll). dried cas~ava i ~ be~:oming an importan! source of .:arboh)tlrat.:' in liwstoclo. feed' anu fccd indu,tric,. particularly in Nnrth.:rn Europc anu Japan . Thailand. andIna le,,cr extent Indonesia and thc South American \."ountric,. are cxporting elart an cxport program of dricd ca,,ava. Lowcr price> of ca~'ava rclalive lo fc.:cd grain' in the intcrnationalmarkct coulu lcad to 'ignifieant incrca'c in Jcmand fur the cassava as li ve,tod fecd . Sinec ca~~av¡¡ i~ vcry toleran! to drought. its acreage in thc drought-affecled arca~ in Northern A frica ha' al'u been increa~i ng remarkably. Many African countric~ which suffered from drought in thc pa\l are now empha>izing ca-.sava c ultivation . lt ~hould be pointed out that the ca,~ava-growing arca in India has had no famine problem even whcn it had a severe drought. Though ca~~ava ha~ playcd an importan! role in A frica. and it has such desirable fe¡¡tu rcs. improvement in th b crop has bccn given vcry littlc attenlinn comparl!d w ith other erops. The avaage yield of ca.,..ava in A frica i~ about7 ton\ pcr hectare which i ~ about half of that in South America . Cassava mo-.aic di ~ea~c i' pre~ent and 'cr ious only in Africa and India. Ca~sava bacteria) blight di.,ca'e ha' recen ti y becn identiticu as vcry seriou> in severa) African countric~. lt mu~t be empha .. it.cd that thc futurc Jevelnpm.:nt of the crop in A frica can make a u~cful contrihution lo thc cconomy and 10 the qualit) of lifc on the continent JI This mee ting on cassava is the first of its kind in thc hi~tory of ca~~ava production in Africa. lt thcrcforc affords a unique o pportunity to discuss the current pro blems of cassava production in Africa and India and to lay down the future s trategies to sol ve the probkms. The objectives of the workshop are: 1) to idcntify the con~traints in production. rcscarch and developmcnt of cassava in A frica and India; 2) to explore ways and mean~ of solv ing the constraints; 3) to improve ways of mutual collaboration among ca~~ava workcrs and institutions; and 4) to establish a ne t\.\ o rk for cassava improvement in A frica and India . 12 Promising new cultivars of cassava need 10 be tested under a wide range of ecologica/ conditions. Here ~>·e see clean and hea/thy lea ves which suggest a good yield in cassava roots. 14 Possibilities for Economic Research into Cassava Production Systems in Africa J. C . Flinn lmernationa/ lnstitute of Tropical Agriculrure 1 badan, Nigeria Ca~~ava is of as much imcrcst to the economi~t as it is to thc cas~ava brccdcr. physiologist. patholo- gist. entomologist. agronomist or food sc ient i't working in the tropics. Thc economic importam:c and u'c~ of thc c rop. the farming sy~tems or region' in which it is gnll\n. and its role in national market' or in intcrnational trade are discussed in April et a l. (1974). Butler et al. ( 1971). John,on (1963). Jone~ (1969. 1972). Nc,tcl and Maclntyre (1973). and Phillip> ( 1973). Thc objective here is to suggest some arcas of cconomic research in cassava produc- tion \\ hich are complcmentary to that of thc biologi~t. or which are uscful to pol icymakers and planners to asscss the i mpacts and trade-otTs of al ternat ivc strategics fnr inc reasing the product ion of the crop. Some fca tu res of cassava of particular i n tcre~t to thc economist includc: (a) it is !!rown in farm ing 'ystem~ ranging from subs isten::c agricultu rc t~ highly commcrciali zcd systcms: (b) it i ~ prodtKcd using hand labour alone or with some aspect~ of production being mechanized: (e} it can be storcd in the ground for long periods of time. but \\hen harve~ted dctcriorates rapidly unless processed: (el) depcnding on type. it may be consumed with vcry little or vcry ex ten~ive proccssing: (e) proce~sing may be labour-intensive or capital-intens ivc: (!) 111<1rketing channels may be traditional. or may be highly integrated. large-scale operations: (g) the potential marke ts for cassava products are vcry broad . and include human consumption. animal consumption . medicinal purposes. and industri al uses. Thu there are choices available relatcd to system~ of production. processing. and marketing of ca'- sava. Thcsc choiccs in vol ve technical. economic. and sodal decisions. When evolving improvcd ca~~ava production systems, one must be concemed not only with the economic gains from the techno- log~ but with thc di stribution ofthese gains as wcll. Thu~ ocial goals . politics. and government policies cannot be di vorced from thc tcchnical and economic c ritcria u sed. in pan . to cvaluate .. relevant' · ~y~tems of ca a va production. lndeed . government polic ies and infrastructural arrangements are j ust as vital as thc tet' hnology itsc lf when the objective is to incrcasc thc marketed surplus of a commodity. A s implified produclion systcm Tn help focus thc fo llowing discussion. a highly s implified ca~~ava prod uction system is presented in Fig. 1 together wi th a number of the more important factor~ (and thdr principal linkages) which wi ll influcncc the structurc. conduct. and performance of thc ')'stcm in rcali ty . Each of these dec ision or intcrvcnt ion points provide foci for collaborati ve re,earch bet\ICen the cconomist and other workers. There is no suggestion that the economist would be the principal rcsearchcr in each of thcse groups; obviom.ly in some cases. he would be supportive of thc biologist (e .g. on the impact of environmental factor. on ca~sava production) . the political sc ienti st (policies and institutions). or the sociologist (human factor~). depending on the nature of the study. The important point is that there is a need for coopera- tivc. integratcd research between sc ientists of various disc iplines if cassava systcms . in their broader scnse. are to be quant itativcly understood . Agroeconomic rcscarch Toa cena in extent. the research undertaken at one point in t ime becomes the technology of the future which. by implication. will influence the future // 15 { ( r: • r·· L.:- ,• ¡-. l ---- Govt Ob¡ect •v es 8 PoJ1c1 es " .... .... ' Env 1 ronment lnst•tu t •onol Human Foc tors ---; Cl1mo te ><" Arra ngements - Ob¡ect•ves So•ls, etc - Extens1on - Monogement 1 - Regulot•ons 7<,,~ r Re lo! 1ve Pr~ ces Product1on ~ Products/ lnputs ~ Reseorch 7( 11 Form Product1on / \ of ~/ / \ ~ Mon1oc / Resource Base ~\::::: _______ ----- __/ - Technology ~------ ------- - Lond, Lobour ~ Re lol• ve Y1elds -Ca p•tol ~ Relot1ve Resource Requ~rements I ~ Subs•stence ~ Process•ng Requ~rements 1 L ~ St oroge J Morketed Sur plus 1 ~ Morket 1 ng/ D1st n but10n ~ Tr on sport/ Access to Morket Consumer Tostes 8 Preferences lncome Levels Fig. 1. A simplified cassava producrio11 system. 16 viability and social structure of the agricultura! sector. Thus. a case exists for cooperation between the agronomist and the eco nomist in the continuum from identifying research priorities through to the eventual o n-farm testing ofthe subseque nt agricultu- ra! technology. OECD (1964). Fishel ( 197 1) , and Andersen ( 1972) di scuss and provide exte nsive references related to the identi fication. exccu tio n. and interprctation of coopcrative agricultura! re- search p rograms. There are. o f course. obvious are as where the payoff from biological research is apparent and docs not call for the input of an economist (e .g. breeding for resistance to C BB. CMD). However. when considering other breeding objective (c.g. time to maturity of the crop. quality and storage attributes) there are economic trade-offs which will influence the relevance of the breeder's varieties to diffcrcnt target g roups of farrners. processors. and consumers. While the economist may interact with the plant breeder in defining breeding objectives ( Ryan et al. 1974). he has a more obvious role when working with the biologis t in de fining the more economicall y importan! management facto rs that become eompo- nents of the agricultura! technology adopted by the farmcr. For example . what are the economic combinations of inputs to use under various s itua- tions for "improved" cassava production? What are the appropriate methods of cultivation and land management'~ Which cropping systems are relevant to cassava production (e.g . sole versus intercrop- ping. rotatio n sequcnccs)'l How often should the crop be weedcd (hand weed ing or chemical weed- ing)? How and when should the crop be harvested (hand or mcchanically)? Yalid so lutions to thesc questions involve economic a well as technical and biological considerations. in addition toan apprec ia- tion by thc researchers of the resource base of the farmcr. hi s objectives. and management skills. Cassava production systems lf researchers are concerned with identifying and so lving problems o f cassava production. it is necessary that they ha ve a clear unders tanding o fthc role of cassava wi thin farming systems . Also required are specifics o f cassava production practi- ces and how factors both within and beyond the control of the farmer impinge o n his decisions to allocate his resources between cassava. other crops. and competing activities. The quantity o f cassava produced by a farmer. and the amount that he may be induced to produce in the future. wi ll be inlluenced by a complex of: environmental factors limiting yields ( inc idcnce o f di sease , soi l type. drought. e tc .); the prices o f eassava. o ther crops. and inputs; the relative yields of cassava and other crops; hi s resource base, including the level o f technology available; his management object ives (pro fit motives. altitudes to risk . concern for producing a supply of food over time . e tc.). The above class of info rmation is best collected through e mpírica! studies of farming systems- facts. no t imprcssions. must be assembled . These farm-levcl studies o f cassava production ystems should be undertaken with a view to: (a) de cribing aml quantifying present ca sava productio n systems in the more importan! cassava-producing regions of Africa: (b) determining the teehnical. biological . eeonomic. and manageme nt rclationships which exis t betwcen cassava and other crops grown by the farmers: and (e) deterrn ini ng the effects of e nviron- ment (disease. soil s. etc. ) on the o utput of cassava production systems. Such s tudies are multidiscipli- nary in na ture and. for max imum effectiveness. require the cooperation o f thc economist, biologist , and soil scientist. The major payoffs from such studics. in addition to helping the rescarcher understand the systcm in which changes will take place. include: (a) pecifica- tion o f thc problem arcas in cassava production which should ha ve prio rity in b iological rescarch: (b) specification of thc problcm arcas wh ich need to be tack led through changes in institutio nal and marke- ting syste ms: (e) identification of constraints by priority whic h. unless re moved. will impede the adoption o f improvcd cassava technology: (d) idcntificatio n o f technology with a high expected ratc of adoption : (e) estimation of the impact o f new technology in terms ofboth economic gains and the ir distribution betwee n farmcrs: (f) helping to identify the target g roup for the extens ion cffort in cassava productio n; (g) enabling planners to prcdict resour- ces and prices (credit. fertilizer. planting material) required for new technology to be adopted by farme rs: (h ) estimation of future demands for cassava products. Examples of farm-levcl studies of cassava produc- tion systems are: Diaz ( 1973). Ezeilo et al. ( 1975), Pinstrup-Andcrsen and Diaz ( 1975) . and Rankine ( 1972). Procedures and methodological issues invol- ved in conducting farm-level production studies are fo und . for example. in Collinson ( 1972). Norman ( 1 973). and Spencer ( 1 972). The present large di screpancy betwee n bio logical or potential yield and economic yie ld (i.e. that portio n o f the cassava crop harvested by the farmer) suggests that if the price of cassava wcre more attractive. rapid increa es in supply could be forthcoming. This. coupled with the advances being made by cassava breeders in identify ing resistance or 17 tolerance toa number of the more importan! ca~~ava disease~. suggests that future yie l d~ ma) not be the mo~t importan! limiting factor to increased cas~a\a production. Problcms are likel) to occur in thc ar.:a~ of processing. storagc. and marketing. Cassava processing systems Studie~ havc shown that. within tradi tional farming S) stcms, labour requiremcn ts forproce~~ing cassava (often the woman's job) may account for a third or more of thc total cnergy rcquircd to produc.: a consumable or sa lcabl c itcm (Cicave 1974). In con~cqucnce. if it "ere tcchnically possiblc w incrcase cassava production at the farm leve! dueto thc rcsearch endca\ our~ of biologi~ts . "mlld the traditional proces~ing ~)~ tcm (both~ un- or off-farm) havc the capac it) to procc~~ the incrca~ed outpur' Where \\Ould thc bottlened. lie'! Would it be in (a) tran~porting the root~ to the processing ~ itc·> (b) washing/peeling thc roots·> (e) grating or othcr handling of thc root~·> (d) fcrmcnting or rctting thc roots? (e) drying thc produl·t (energy supply'1)? ( n storing the processcd ca,,ava? lf thc proce~~ing sub~cctor of the ca:,,ava ')'tcm i!> likely to represen! thc effective constraint on an increa,ed ~uppl y of ca~sava products reaching thc markct. it must be quantitatively understood bcforc a solution is sought through rc~carch. Herc thc economist should coopcratc "ith the enginccr and food tcchnologist in ~pcc ifyi ng the capacit). capital. and uperating co,ts of cassava processing equipmcnt that i~ more attractivc to the processor than that presentl) available. The di~tribution effects of propo~ed tcchnology for cassava proce~~ing rcqu ire equally careful considcration (e.g. "mcchaniLed .. gari production). Industrial plants capabl c of pnx:essing over 40 ton~ of root~ a day are on thc market (e.g. thc Ne"ell Dunford plant: Anon. 1974). Othersofan "intermc- diate leve!" of technology are avaílable \\hich procc~~ up to 4 ton'> of roots per da) (e.g. thc "FABRICO" and "PRODA" plant~ in Nigeria: Ngodd) 1974). lt i~ likcly that while the lar!!c. capi tal-intens ivc planb are extreme! y efficicnt frZ)J11 an cconomic vie" point. thcy could al so e reate ma'>'> unemployment in the traditional , labour-intcnsive production sectors. From both economic and ~ocíal viewpoints. design critcria for ca~sava proces~ing plants should pcrmit more cfficient labour producti- vity than traditional sv~tem~ "ithout creating uncm- ployment. This rcpr.;scnt~ a challcnge for the food technologist and enginccr. ~ Cassava marketing systems lf ca~~ava production increases be) ond the point rcquircd to mcct thc incn:asing demands of an cxpanding population. what may be the con,cqucn- ces from marketing and pricc viewpoints'! Sc\cral workcr~ ha ve suggcstcd that the dcmand for Ca\ a products for human con~umption is rather incnmc- inela~tic in ~omc countrics of Africa. and l.jllitc income-clastic in othcrs (e.g. Phillips 1973. p. 17). but that demand for ca~sava product~ is not 'ery pricc-ela~tic (~ce Araullo ct al. 197-1 and Ed\\ ard., 1974 for a revie" of indu,trial proce"ing). Thu,. m those countries \\ hcrc thc inn>me and pricc cla.,tici- tic~ of dcmand for cassava are inelastic. if suppl) incrca-;cs ata more rapid ratc than does populat ion it is po~sible that ca~~av¡¡ prices may fall. Thc produn~r i-, then nn bcttcr offthan he \\as bcfore he incrca .. cd hi' production . Undcr '>Uch ci rcum~tanccs. pro- gram~ aimed at incrcasing the national pnxlm:uon nf cass¡¡,·a above that requircd to meet the necd' nf an cxpanding populat ion \\ ould ha ve littlc challL'C of ~uccc~~ - unless there \\a' price intervcntion b) government or unlcss alternativc markcts for ca.,,a\a products wcrc found. Thu~ in ordcr for policy maker' to as~e's the pricc cffcct'> of i ncrea,cd ca.,.,ava production. and to e<.,tima tc future demand. dcmand studie-, for cassa\ a must be undcnaken b) markcung Cl'llllllllli~b. Other bottlen.:cks in thc ca~'ava '''>te m llla\ be found in transpon and mark.:ting. J~hlNHl ( 10DJ and Joncs ( 1969. 1972) describcd thc markctin!.! systcms for ca~-.ava in Africa in genera l tcrn1<.~ However. dcspitc thc imponancc of thc crop. thcre i' littlc quantitativc informal ion availablc on ca""' a marketing in thc traditional '>CCtor. In mam cnun- tries in Africa tran~port ¡, poml) de,elopcd and cnstl). Progralll'> aimcd at increa~ing product1on ot cas~a\ a m ay be to no a,·a il if transportation to markct' is not a\ailablc. We need w a-,,e.,, thc capacit) of the transpon') 'te 1m to handle inneascd volume~ of ""~ava and othcr proJucts . and tn determine what modificatillll'> are rcquircd tn cnablc prc>dUl't> to be 1110\ cd at l1m L·ost. * i\larketing margins in dc,cloping countn.:' (the diffcrcnce bet\\ecn thc pm:c rcccivcd b\ the farmer and the price paid by thc con\umerl a~c ~a1d to he cxcessive. and middlemen appar.:ntl) makc htr!.!e profits at the expense of thc produca and thc * lt mu'l be poinled out. h<>\\cvcr. lhal roatb and <>lh~r improvcmcnl' of infra>lru.:ture do not í!Uarantee !!reater con!.umptinn of ca,.,ava. lt " po'sibl~ thal imprmed lran,portalion could 'ubjecl hcrclofnre ¡,olaled produ.:er' 10 cmnpelition from imponed commoditie' or commodlth!' from other rcgion,. 18 consumcr. ls this a fact. or is the nature of the cassava marketing systcm ~mall-scale ano high-co't in nature with the ino iviouals prnvioing scrvicc' within thc marketing systcms. like thc farmec making modcst incomcs') Many rnarl..eting stuoics (Thodey 1 96H) suggc~t thi~ is the case. but policyrnakers often arguc otherwise. Obviou,ly. work is requireo to measure thc prices of cassava producb. the vol u me of tlm\ at various points in the marketing 'ystem. and thc cost of prov iding such service' beforc statcmcnb rélating to the efficiency of the marketing ~y,tems can rcasonably be evalua- teo. Such information will abu hc lp 10 evaluate the oistribution con,cquencc' of altcrnative forrn' of cassava marketing k.g. govcrnment marketing boards. produce marketing coopcratin:s. etc.). Policy a nd pla nnin¡.: While poli t ical and social dccisions" ill influence policie~ relatcd to ca\sava produc t ion anclmarkcting (e.g. objcctives of sclf-su!Ticiency. the role of the publk ,·crsll!, thc privatc sector) it i' important that the social costs and the distribution of thc bcncfits (and pcnaltics) from altcrnative policies are krm\\ n to thc policymaker. Thc micro- and macrocconom ic studics alrcady identified providc part of thc information rcquircd by governrnent to a!>ses' the rclative merits of alternative mcans of achieving thcir food production objcctiv.:~. Once objcctive~ have been defined it thcn becomes the re~pomibi lity of thc planner. econo- mist. and technologist to identify the nature of thc in,titutions. policies. and programs which willmost efficicntly contributc to thc attainment of the political ob¡ectivcs. lnformed programs. whether they are in agricultura! rescarch. extension. input infrastructurc. marketing. or pricing. cannot be formulated. or emharras~ing consequences avoicled. withnut a quantitativc underqanding of the various componen!~ and links \1 ithin the ca~sava-based systcm. Summar~· Whcn considering ca~sava improve ment in Africa. it is vital that the componen!\ of cassava production. processing. marketing. ano policy. be studied asan inkgrated system . Thc ramification~ of a possiblc ch¡¡nge at onc point in the >y>tem ma: be magnificd. dampeneo. or "ip..:d out by a reactÍlln o r bottleneck in anothcr. To understand thc true n1nsequenccs of an objcctiH· to increa'e ca..sava production at the farm leve! requircs an under~tan­ ding ofthe linkagcs betwccn thc various subsystcms. a task which requircs the cooperation of the biologist. social scientist. and planner. In general. cassava sy,tcms in man_:, parts of A frica are not well understoncl or docu rncntcd. Cooperat ive research projccb should be initiatccl to reduce this gap in knowledge. Refercnccs Annn . 1 tel. Barry. and Rcginald Maclntyrc (cu.) 1973. Chronic ca"ava toxicity: procccdings of an interdi,cipli- nar) workshop. London. 29-30 Janual) 1973. !DRC-OIOe. lnternational Developmcnt Research Cen- tre . Ottawa. Canada. 163 p. Ngodd) . P. O. 1974. Gari mechanization in N1geria: a ca'e 'tudy in thc choice ot tcchnology. Paper p~cpared for th<' Stud) Group on Lo" Cost Technology and Rural lndu, triali zation. OECO. Paris. 19 Nonnan. D. W . 1973. Methodology and problem; of f ann managcment invcstigations: expcriences from northcrn Nigeria. Africa Rural Employment Paper No 8. Dcpl. of Agricultura! Economic;, Michigan S tate Univer,it). East Lansing. OECD. 1964. Co-operation bct\\een re;earch in agncultu- ral natural ~ciences and agricuhural economic;. OECD Documcnrarion in Food and Agricuhurc No . 65. Pari ~. Pinstrup-Andersen. P .. and R. O. Diaz. 1975. A ~uggcsted method for improving the information ba;c for c'ta- blishing prioritie~ in cassava re;earch, p. 51-59. In N este l. B .. and R. Maclntyrc (cd. 1 The international exchangc and testing of c a;sava germ pla\m: procee- dings of an interdisciplinary workshop held at CIAT. Palmira. Colombia, 4-6 Fcbruary 1975. lnternational Developmcnt Research Centre IDRC-049e. Ottawa. 74 p. Pholllps. T . P. 1973. Cas;ava uulization and potcntial markets. lntcrnat ional D.::vclopment Rc":arch Centre IDRC-020e. 183 p. Rankinc. L. B . 1972. Comparative economic~ of root crop product ion in sclcctcd countrie; of the commonwcalth of thc Caribbean. Dcpl. of Agricultura! Economic, and Farm Managcmcnt. the Unover,ity of the Wcst lndie>. Trinidad. Ryan. J . G .. R . Sheldrakc. and S. P. Yadav 1974. Human nutritional need, anJ crop breeding obj<:Cll\ e; in thc ~cmi-arid 1ropic1. Economic~ Dcpanmcnt. ICRISAT. Hyderabad. India. Spencer. D. S. C. 1972. Mocm-le1el farm managem~no and production ewnomic' rc,carch among tradotoonal African farmcr;: Jessons from Sierra Lconc. Atncan Rural Employment Papcr No. 3. Dept. of Agri<:ultural Econom ic~. Michigan S une Univ.::r;it). East Laol\ing. ThodC). A R. 196X. \1arkcting of,taple food' in ,,c,tcrn Nigcna. Stanford Research ln'>lllutc. Stanford. Cahtnr- nia. 3 Volume' 20 Improvement of Cassava at the International Institute of Tropical Agriculture S. K. Hahn /nternationa/ Jnstitute of Tropical Agricufture l badan. Nigeria The rnajor biological constraint to cassava production in Africa is d isease. especially cassava rnosa ic diseasc (CM D) which ex ists o nl y in Afri- ca and India. cassava bacteria! blight d isease (CBB). and anthracnosediseasc. lnsects (e .g . green mite recently introduced into East Afr ica . and meal y bug. recently identified in z a·ire and the Congo) are pote ntiall y serious pests in A frica. Most of the local cassava cultivars are susceptib le to these diseases and yields are low. be ing about 5-1 O tons of fresh yield per hectare in 12 months compared with potential y ie lds of more than 20 tons. Although cassava is a vcry importa n! staple food crop in tropical Africa. and has vcry serious problems . improveme nt o f thc crop had been given very little attention. Realizing th is. thc lnternational lnstitu te o f Tropical Agriculture (liTA) establ ished the Root and Tuber lmprovement Program in 197 1. This program covers cassava. yams. swcct potato. and cocoyams. with cassava receiving the highest priority. The broad objectives are to devclop improved cultural practices and varieties with high stable yields. high quality. and plant characteris tics suitablc for efficient cropping systems. T he ult imate goals and interactions among the disciplines are presented in Fig . l . Program objcctives The specific objectives for cassa va improveme nt at liT A are: 1) h igh y icld in terms of dry matter product ion per unit o f land and time in both monoculture and mixed cropping systems: 2) resis- tance to. and cultural control of, economica!ly important diseases and insects: 3) improved quality in terms of consumer acceptance. nutr itional value. and processing characteri stics; 4) improved plant type: canopy and root characteristics: and 5) adaptat ion to a wide range of environme nts. At liT A. a large germ plasm has been assembled in seed form from Africa . Latín America (especially from Brazil and CIAT ). and Asia. This has been cvaluatcd for res istance 10 the major economic d iseascs and for agronomic traits. Sources of res istance to the d iseases inc lud ing CMD. C BB . and anthrac nose d isease ha ve been identi fied. Res istance to both C MD and C BB showcd a posi tivc !!enetic rclationsh ip . Extensive hybridizations among selected parents are made and about 100,000 seed li ngs are ra ised and screened for disease resista nce and ~oot characteris- tics every year. About 2000- 4000 seed lings are selected a nnually in fi ve different locations in Nigeria and 19 different locations in za·ire for cloning, a nd for further evaluation forre istance to diseas~s and lodging, for root and plant c haracteris- tics and fo r high yield po tential. Promising clones ha ve becn put forward fo r advanced y ield trials and the mo t elite clones from these tr ials have bee n tcsted in three different ecological arcas. A few clones whic h have shown consis te ntl y superior performance over the years in terms of disease reasista nce, yield , resistance to lodg ing, and plant c haracteristics have been mult iplied and planted in farmers' fie lds in Nigeria. We can then o bserve performance under local conditions and farm practi- ces. and test farmers· reactions to improved mate- rials there by making the best clones available to farmers. Some selected seed lings have also been eval uated on farms in za·ire. S ignifican! progress has been made in producing improved cassava clones with resistance to diseases (espec ial! y CMD and C BB). higher yie ld , improved root charac teri stics, and resista nce to lodg ing. Their gari quality has been tested and is acceptable. 21 Soorce lbterials ( rt'obleiiiS P•tbology •• • EAtomoloo < ¡: • Biothtmistry <- ¡! • Physioloo ldoatific.ation & ...... tito _______ _,_ Breeding .-e~- Agronomy • Breediftg • Agronomy Recombiution lmpro•td mittits Economits Farming system • Higher • production -:--= UtilizJtion hnpmtd • cullllral DuelopiiHI methocls Stor•c• Mechuiution Fig. l . A flow dia~ram shm•mf? the mteractiom hetween disupline.1 and the goa/1 of the pro~wm. Our disca\c-resistant (CMD. CBB) ca ... .,ava in Nigeria maintain ~ it., re,istancc when planted in Za"ire. Resistance of our material to CMD was also confirmcd in Sierra Leone, Libcria. and Togo. Our material ~howed a high le\cl of resi..,l<;nce to anthracnosc di,ease in Za"ire '' hcre the di-,ea-,c i~ a ~erious probh:m. Cassava from exotic sources has becn ... ucce~s­ fully improvcd for re..,i\tance to CMD. CB B. and lodging v. ithout sacrificing desirablc agro- nomic lrait\. 8) continuou~ sckction for thn:c g.:n.:rations. cassava has becn improved for l for good 'ccd germination are high soil moislllre and a soil tempcrature of between 30 and 35°C. Thi ' gcrmination mcthod ha' enablcd the program to deal with large plant population\. The brceding mcthod \\hkh wc ha'c .,uccc~~fully uscJ i~ a half-..,ib famil} 'clection method in combination "ith a poi yero'' mcthod. lt ha~ bccn po~~iblc to incorporate gene' a\sociatcJ with re~i~tance to di.,casc' and lodging and good root charactcri~tics. Thi' mcthod also mal-.e' pos\ible thc introgression of C\Otic source-, into our bn:eding populations. Method' of sc lcction ha,·c hccn de' elo- ped based on somc gcnctic information. The kc} factor' which have made ca">a,·a improvement worl.. succc.,..,ful are fan>Urablc cnvironmcntal conditions forcvaluation.large popu- lation si1cs, appropriate brecding mcthod~. and good teamwork among the relevan! di-,cipline,. A number of national ca"a' a "orl..cr-. from man) countrics in /\frica and Asia havt: ... pent time at liTA to bccomc bettcr acquaintcd with our cassava improvcrncnt technolog) . .\1(1\t have rcturncd homc with improved secds and ha\c e~tabli-,hcd national ca~sava improvcmcnt programs in their own coun- trie~. The Root and Tubcr lmpnl\cment Program no'' consists of a corc program at liT/\ and t\\ O coopcrativc cassava programs in Nigcria and Zaú·c . Therc are 13 profe.,...ional staff and st:vcral postdoctn- ral fcllo\\ s within thc program. Through staff 'i'!ts and thc cxchangc nf infonnatinn and mataial. thc prograrn has e~tablishcd coopcrati\ e linl-.s '' ith man) national programs and in..,titutit>Jh in Africa. Asia. and Latin Amcrica (partil·ularl) CIAT). Cassava Bacterial Blight in Africa E. R. Terry lntemational lnstitute of Tropical Agriculture PMB 5320. !hadan. Nigeria Cassava is groils. and the high yield potential of sume cultiva rs. Within the la;.t 5 year\, ;.erious o utbrcaks of cassava bacteria! blight cau~ed by the bactaium Xamhomonas mwtilwtis have been reponed in Za'irc (Hahn and Wi lliam;. 1973), Nigcria (Williams el aL !973), and Cameroon (Terry and Ezumah unpub- lishcd data)_ This disease is cnnsidcrcd the mo~t devastating of severa! bac teria! d iseases nf cassava in Afr ica ~ince it often rcs ults in total loss o f both y ield and plantí ng mate rial under conditions favour- able for its deve lopment (Lozano and Sequci ra 1974b), The potcntia l for further sprcad of the disease by movcment of infectcd p lant ing material b hig h , and the rcs ulting cassava crop losses could affcct thc nutrition and cconomy of millions o f people in Africa. Thi~ rcvicw of the d iscasc in Africa hiohli !!hb somc relevan! fact~ that may \Crvc as a ba~is for chccking its ~prcad and eventual controL Etiolog~· and symptomatology Thc cassava bacteria] blight pathogen Xantlumw· nas manihotis i~ a gram- ncgative , motile. slcnder rod. wi th a s ing le polar llagell um ( Lozano ami 23 Scqueira 197-+a). Thc bactcrium pcnctratcs v ia thc s tomata or through wounds in epidcrmal tis~ucs o f lea ves and young shoots. invades the vascular tissue. and cause~ cxtcnsive brcakdown of parenchímatous tiss ue _ Thc charactcri;,tic 'ymptoms includc (a) angular_ water-soakcd leaf spots 11 hich are small initially hut la ter enla rg:e. coalcsce, and eventual! y turn brown_ Thc aiTectcd !caves hccomc h lightcd and dry and cventually ab~cise: (b) degrees of leaf wilt ranging from une wi lted lamina l(>be to a large number of wholc Jeavc;,: (el )Cikm-orang:e gum exudation on thc lcaf pctiole ancl )Oung \hnnb: (el) >evere dcfoliation; and (e ) tip dieback re~ulting from vascular necrosis ami death of the growing points. Geographic distribution In Zaú·e, the d ist:a\C was fi rst notcd in 1969-70 in thc Gungu rcgion of Bandundu provincc ( Hahn and W illia1m 1 973l and latcr rccognizcd throughoutthat prov ince as 11 ell as Bas-Za'irc and Kasa i provi nces (Maraite and Meyer 1975L In Nigeria it was first obscrvcd in 1972 in man) locat ions inthe cen tral and southern reg ions (Wi ll iams et aL 1973). Thc h ighcst incidcncc was obse rvccl in the East-Central and Mici-Wcstern states where 100'7c crop losses wcre recorcled (T cn:y unpuhlishcd data). In thc Came- roon. incidcncc has so far bccn reponed in 1974. on ly in thc Northwcs tern. Wcstcrn. and L inoral pnwincc' {Terry ami Ezumah u npublished data)_ !t sh nu!d be cmphasÍLed that the disease íncidencc survcys made ha ve as yct bccn vcry fragmentary, and becausc of thc case nf sprcad nf the disease, more comprchcnsive surveys may likely identify o thc r local ions in A fr ica 11·herc the discasc has not yet b~en identified and repon ed, Cassava bacteria( blight (CBB) is a deadly disease of cassava in certain parts of A frica. and can cause total crop failure. The p lants shown in the background have been bred at 1fT A and are resistan/ to both CMD and CBB. 24 Epidemiology The incidence of the di sease is highcst during the rainy season, and rain-spl ash is thc most importa n! mcans of spread over short distanécs. Thus . although scattered leaf spotting and wilt have been observcd during the dry season in Nigcria. the incidence of these sympto ms increases s ig nificantl y during the rainy season. Disease incidence in Colombia is highly correlated with amount of rainfall . and thc locations in Afri ca where the most severe incidence has been observed are also high rainfall areas IKikwit. za·ire. 1600 mm: Warr i. Nigeria. 2600 mm: Bambui . Cameroon. 2540 mm). Dissemination fro m one area to another occurs largely throug h infected planting material (Lozano and Sequeira J974b). The bacte ria may also be disseminated by movement uf soil during fa rm operations and by the use of contaminated tools. It is al so possible that the practicc of manually removing young cassava lea ves for cooking may play a ro le in disease spread . lnsect~ have bcen suggested as possiblc agents ot disseminatio n. but no suppo rting ev idence has bee n presentcd. However. at the onsct o f the rainy season in Nigeria when there is a partic ul arly high incidente of blig hting and wi lting. there are also large populations o f thc variegated grasshopper Zonoce- rus vuricgatus colonizing infected plants. 1t is known that grasshoppers · mouthparts and appcnda- ges become contaminated with moist bacter ia! exudate (Terry J974a), and it seems likc ly that they may aid in spreadi ng the di seasc. In za·ire. it was reponed that the disease was more severe on poor sandy soils (Hahn and Williams 1973). and observations in N igeria suppo rt thi s observation. In Cameroon. however. severe inci- dence was observed on volcanic soils (Terry and Ezumah unpublished data). Control The most promis ing method of controlling cas- sava bacteria! blight is to utilize di sease-res istant cultivars. Most ofthe local and improved cultivars in Nigeria are highly susceptible, but screening for resistance has been carried o ut at this lnstitutc. Observations for 3 years have revealed a large number of clones which show high tolerance in hig h- rainfall arcas favourab le fo r ctiscasc dcvclop- me nt: 202 local cultivars tested at the same location showed no tolerancc . Slower sprcad of bacter ia! blight by pruning most of the abovc-ground portio ns of in fected p lan ts has becn reponed (Lozano and Scqucira 1974b). T he disease can also be controllcd by careful sekction of di scase- free planting material and a successfu l me thod for producing bacteria-free plants has bccn developed (Lozano and Wholey 1974). Crop rotatio n has bccn suggcsted as a means o f control. but the survi val potential o f the pathogcn in A frica in not known. A possible mode of survival of Xalllhrmwnas manihotis is in thc form o f dry. bacteria! pelle ts. cach of which may comain 100.000-1 ,000.000 viable ce lis (Tcrry 197-lb ). References Hahn. S . K., and R. J. Williams. 1973. lnvestigationson cassava in the Republic of Za'ire. Repon to the Commissaire d'état á 1' Agriculture. Republic of za·ire 1973. Lozano. J. C.. and Luis Sequeira. 1974 a. Bacteria! blight of cassava in Colombia: etiology. Phytopathology 64: 74-82. 1974b . Bacteria! blight of cassava in Colombia: epidemiology and control. Phytopathology 64: 83- 88 . Lozano. J . C .. and D . W. Wholcy. 1974. Thc production of a hactcrial-frcc planting stock of cassava. World Crops 26 (3): 114- 11 7. Maraite. H .. and J . A. Mcycr. 1975 . Xamhomonas manihotis (Anhaud- Bcrthct) Starr. causal agcnt of bacteria! wilt. blight and lcaf spot> of cassava in Zalre. PANS 2 1: 27- 37 . Terry. E. R. 1974a . So me cpidcmio logical f¡tetors affec- ting thc survival and disscmination of Xmuhomonas manihotis. Proc . First Workshop on Cassava Bacteria! Blight in Nigeria. liTA. lbadan. May 1974. 1974b. A mode of survival of Xomhomonas manilwtis. The cassava bacteria! bl ight pathogen. Nigeria Soc ie ty of Plant Protection. Occasional Publ ica- tion 1: 19 Abstr . Will iams . R. J .. S. D. Agboola. and R. W. Schneider. 197 3. Bacteria! wilt of cassava in Nigeria . Plant Dis. Rept. 57: 824- 827. 25 Sorne cultivars are highly susceptible 10 pests and diseases. Grasshoppers are a major tlrreatto cassava /ea ves. sometimes causing complete defo/iation as shown here (/oregromrd) . 26 Advances in Research on the Economic Significance of the Green Cassava Mite (Mononychellus tanajoa ) in Uganda Z. M. Nyiira Kmranda Rl'.H!arch Sration P.O. Box 765. Kampala . Uganda The grecn cas~ava mi tc Mononrchellto lllnajoa ''a~ prt;bably accidentally introdu~ed into Uganda during recen! years. The fir~t documented outbrcak was around Kampala rNyiira 1972). Sincc 1972 thc pcst ha~ covcrcd wide arcas including Kenya, Congo Brauaville. Burundi. and pos.,ibly R\\anda. Prior to its disctncr~ in Uganda thcrc was link concern 01er mitcs "' pe~h of cassava. although C\lslcnn: of thc pc'>t \las recogni7cd. Nyiira ( 1975a) ,hm' s that '>pider mi tes infest cas~ava in South and Central Amer ica. thc Carib- bcan Zone. A~ia. and Africa. The specics include: Tetranrchu.\ urtic11e ( = T. telaritl.\ ). T. !Jimaculatus: T. cimwharinus. T. timidius. Mononycltellus umajoa. M. carriheanae. M. plan- l..i. M. chemo.1etosus. M. lwndari. M. planb.l'. and Olil(oltYclws s:os.n·pii. Togcthcr. they form "hat is no" knm'n a~ thc cas,ava mi te comple\. Althou!!h all ha1 e b.::c n rccorded on ca~sava. some are of min~Jr economic consc4uence . Moll(l/l."cltellus wnajoa is considercd a potential threat to cassava <.:ultivation throughout thc African tropi<.:s. Other Mononycltellus species and T. cinnabarinus could atta in economic pest ~tatus as cas~ava becomes more '' idel; cultivated . Econom ic importance M01tonrchel/us tanajua fced' on thc undersurface of the cassava Iea1 es. The ínfested leaves develop mottling. and a bron1.ed, mosaic appearan<.:c. On closc cxamination. lhc brontc mo~aic appearance Í\ dueto the yellow ~pots rcsembling p ín-point pricks intlictcd b) thc mite·\ fccding. Thc damage ¡, chíetly to top leave.,. although under severe infe~ta­ tion. stems gel scarified and die back. Scverel) ínfested leavcs drop off whilc thc you ng ca~~a1a ~tems damaged by the mi te turn rusty and rough. The damage 10 lea ve, and s te!n~ interfere~ \\Ílh phl)lü'>yn- thesí~ activity of thc plants. and that. combíned with the dropping off of lea ve'>. m ay lcad to rcduccd tubcr yield. Host range Host plant~ of thc green cassava mitc in Uganda índudc thc follm1 ing: .\1anilwt species: escule111a. utili.\sima. gla~im·ii. dichmoma . heptaphylla. piau- hren\is. ami cartagennis. In Brazil. the green cassava mi te al,o brccJ, and multiplies on Manilwt apii. The mire ¡, apparcntly restríctcd to Manihot spcc ie~. lt is po~siblc. h(mever. that its hosts may include other Euphorbiaccae. Natural cnemies The most common natural enemics of the grecn cassa\'a mi tes in Uganda includeO/i!(Oill (Staph) lin i- dae) and Stetlwrtt.\ (Coccincllidae ) ;,pecícs and the Ph) toseíidae mire complcx . Other predators include Syrplws (Díptera : Syrphidae). Chrysopa (Neurop- tera: Anthocoridac). Thrips (Thysanoptcra: Thripi- dae). Geocuris ( Hcmiptera: Chrysopidac) . Orius insiduous ( Hcmiptcra: L) gaeidae). and spiders (Aranca). A brief 1 ook at the poten tia! of predators a-, control agenb of the grccn cas'>ava mite sho\\ed themto be effectí' e ami fa, ,)u rabi e. In Trinidad. thc Common\\calth lnstítutc of Biological Conlrol carricd out a similar survcy. They identified Trph/odro111a/us /i111onictts ami T. rap(Lt as abundan! and importan! predators of general Mmwnw·ltel/us. They al~o encountercd thrips and cecidomriids (probabl) Feltiella sp.) assocíated with MmunnThellu~. These initial tindings suggestthat natural ene mies of M. twu~¡oa are nf s uffic icnt imponancc to mcrit 27 further investigation fo r application of control o f thc countries free of thc pcst might be developed. Such a green eassava mite. system would grcatl y a~~i~t planning of con trol acti vities. Life history The green cassava mitc i ~ very small (average length 350 JL ). lts egg is about half the ~ize of thc adult. Adult femalcs la y thc cggs o n thc undcrsurfacc of the lea ve~. Initially. the egg~ are depo~ited along the midrib and alo ng the veins. As the femalc population incrcascs. thc eggs are dcpositcd at random. The eggs when first dcpositcd are transparcnt and spherical. As incubalion pmg rcsses the eggs look glassy and eventually the colour of paJe stra w just before hatc hing. The number of egg~ laid per female varies, the average at an uncontro lled mean temperature of 23°C being 3.4 per femalc per da y. The incubation period is 4-5 days at 1 6-32°C. The hatchability percentage ranges from 25 to 100%. Thc nymph period at Kawanda Research Station (at 1 6-20°C) i~ 8- 13 days. The egg-lay ing lifc span averagcs 19 days in the laboratory at uncontrolled temperaturcs. and thc mean longevity of fcmalc~ i~ JO days. Aerial dispersa! A reasonably consi~tent activity pattern by g rccn cassava mi tes was recordcd between 24 and 29°C in slight air currents. Strong winds do not intlucnce aerial migration. altho ugh water. othcr insect~. and various a nimals contributc greatly to thcir movc- ment. Nyiira ( 1975b) dbcussed the importancc of meteorological factors in the develo pmen t. migra- tion , and dispersa! of thc g rccn cassava mitc. With sufficient knowledgc of the aerobio logical activit) of the green cassava mite. a warn ing system for Scasonalabundance Mite population dcnsity is highe~t during tht: driest periods, and high-humidi ty condition~ tcnd to suppre~s major outbreak~ and damagc. In rain) conditions thc population of mites dccrea~e~. Tht: reduction in the population in wet wcather ma) be d uc to a physiologica l phenomcnon linkcd with thc develo pment habits of thc mite. o r a rc~ult of mechanical de~truction of the mi te stages. Yariation in thc population dcnsity. foiiO\\Cd nvcr 12 months on thrcc varietie~ of cassava. ~hO\\ ed that mite popul ation~ remained lo\\ on all the varieue~ aftcr 1 O rnonth~ irrcspcct i\c of changes in "eatha conditions. Con trol mca~ure~ ma} be unnc..:c~sar) o n infe~ted cassava aftcr 1 O rnonth~. H1mcvcr. the\c rc~ults wcrc not followcd up w find out whether thC) had any bearing on planting time of thc ca~~ava. EfTect on yield Different leveb of infc~tatinn can rc~ult in mcr 40'7< lo~~ in frcsh tubcr \\eight. a~ a re~ult of gre.:n ca~~ava mite attark (Table 1 and 2). Thc magnitudc of the ) ield l o~~ " ill dcpcnd on thc age at '~ hkh the crop i ~ attackcd and thc nurnber of mite~ invohed. This aspect is under active \t udy at thi~ laborahX). Control measures lnit ial work (:-.lyi ira 1972) on the control of ca~sava mite~ u~ing chcmical~ indicatcd that thc three acaricidc~ tc~tcd ( Kelthanc ( Dicofol). Chlom- bcnzi late (Akar). ami Rogor (Dimethoatc) ) ''ere effective again~t grccn ca~\ava rnites. Kelthane and Rogor \\ere more pcr~btent than chlorobcn7ilatc. They provided ~ u ffic icnt protcct ion foro ver 30 da) s. Table l. Yield and quality components of cassava (variety Bukala~a 11 ) f1om green cassava mite attack compared lo the unprotecled cassava of the same va1iety. Mean fresh yield per plant (kg)b Estimated fresh tuber yield per hectare (kg l ha) Mean diameter of tube r (at 12 months) (cm) Mean length of tuber (at 12 months) ( cm ) Mean no. of tubers per pla n! Quality (laste scorec) Texture • Sprayed with Rogor (Dimethoate). " Based on 3-score sca le: Good, Fair, Bad. e Mean of tuber weight from 40 plant s/ treatmc nt. 28 Protected•1 (sprayed) 2.46 16,520 14.04 24.83 8 Good Good Unprotected ( unsprayed) 1.67 11 , 190 13.5 1 24.66 6 Good Good Table 2. Tuber weight (k!:) per plant recorded at differcnt infestation levels in six replicates o n cassava variety Bukalasa 1 l . Jnfestation levels (based on plots of 20 plants each ) 0% 2or¡- 40% 60% 80% 100% 1 2.22" 2.25 1.95 2.3-t 2.02 1.95 2 2.72 3.10 2.27 1.99 1.48 2.56 3 2.63 2.93 2.22 2.29 1.61 1.98 4 2.5\J 2.26 1.97 1.9'J 1.78 1.03 5 2. 18 1.71 2. 15 1.43 1.56 1.54 6 2.41 2.05 3. 16 2. 17 1.75 1.69 Total : 14.75 14.30 13.72 12.2 1 10.20 10.75 Mean: 2.50 2.40 2.30 2.10 1.70 1.80 • Weight of tuber (kg) per plant from 20 plants per treatment per rcplicate. Difference between yie ld of different infestat ion levels significan! at 1% leve! of probability. Thc mi te popul ation rcduc tion aftcr 2 ''ce k~ ''a' ovcr 70'ff in cassava trcatcd '' ith a ll thrce acaric idc~. The use of expcn~ive acaricide~ again~ t cas~ava mitcs b not fcas iblc by pca~ant farmers in Uganda. Morcover. more scvcrc mitc i n fc~tat ion occu rrcd aftc r cvcr} application of chcmicals. On furthcr invc~tigations. we fnund tha t thc chcmicab had an advcr~c cffect on thc common. abundan! prcdator' of thc gree n cas~ava mi te-. (0/is,:Ofa. S1c1homs. and Thrphlodromus). Currcnt rcsulb ~hm' that bi{>lngical control agcn t\ can ~uppre~s the mitc populat ion~ . particular! y in thc arca-. uf bimodal rainfall (th..: .. o uthern zonc nf Ugancla). Ccrtain varic tics of cw,sava can'' ithstand infc,tation '' ithout much darna!!c to lcavc~. lf a \aricty is fast-maturing (thcrcfo;c ~ubject tu fC\\Cr 'críe~ o f infestation~). ancl fairl ~ rc-.istant or tolcrant tn thc mi te damage. couplcd '' ith biological control agcnt-.. the effect o n that varict~ b~ thc grecn ca-.-.a\a mitc would be rcduced. 29 Effective mca-.u rc-. again~t thc grccn cassava mi te -.hould. thcrcforc. aim a t an integrated contro l utiliz ing qu ick-maturing . rc-.istant/tolerant varicties backed by a program of biological control. References yi ira. A. M. 19n Repon of inves1igation on cassava mil e. \fononychellu.' UIIWJOII ( Bondar). Depl. of Agríe. Ka\\anda Rc,carch S1a1ion. Kampala. Uganda (mi- meo.) 1975a. Biology. uis1ribu1ion and ccology in Uganda of lhe green ca;;ava milc. Mononychellus rmwjoa (Bondar) (Acarina: Tctranychidac). Ph. D. The~is. 3 14 p. 1975b. Ca"ava mil e' and meleorology. Paper prc:.cnlcd 10 lhc 2nd 'ympo-;ium of 1he Uganda Sociely of Agronomy. Kampala. Uganda (lo appear as Proc. U. Soc. Agron.) . Summary of information on cassava production in Africa and Sri Lanka Yield Genn plasm Germ plasm (tons/ Are a Country collected evaluated ha) (ha) Typc of culture Diseascs Pcsts Quarantine Personnel Finance Projects --- ---- --- - CAMEROON Local S 5 200,000 1 Monoculture 1 Mosaic 1 Mealy No cuttings Severa! Under 1 Establishment SEnégal 2 Mixed 2 Bac teria! bugs from expansion National Root Mada- few cropping blight 2 White fly surrounding Crops l mprovement gasear 3 Gra>s- countries Program hoppcr Exrand production in north and east Mixed cropping with cocoa 4 Provision of improved seed to fanners GHANA Local 90 12.5-37.5 270,000 1 Mixcd With 1 Mosaic 1 Grass- 1 Cuttings l{csearch: Limitcd 1 Demonstration Forei~n ~ maiLe 2 Anthracnose hopper through Severa! farms Monoculture 3 Bacteria! 2 Mealy quarant inc Extension: 2 Establishment south-easl blight bu~s 2 No cuttings 1 per 100 processong plants (new from Brazil. farmers 3 Effects of mosaic record) Nigcria on yie ld za·ire 4 Affect~ of age on tuber yield LIBERIA Local so - J4.000 1 Mixcd (72%) 1 Mosaic 1 Red mitc l No cuttings Re;earch: Re~carch: 1 Cassava evaluation Bra1il 1()(} 2 Monoculture 2 Brown 2 n cer from 27 $270.000 2 Fea~ibility pellet Malaysia 12 (2K<'(. } spot 3 Rodents Central & Extension: Extcnsion: plant Ni~eria 100 3 Whitc S. America, 220 $282,000 spot Gabon, lvory Coasr. Ni~ería, Tanzania za·ire MALAGASY Local and 6 220.000 1 Monoculture 1 Mosaic - Very ~tricl - Govt. of 1 Productivity REPUBLIC introduced 2 Mixcd culture 2 Root rot Malagasy 2 Resistancc to hybrid in the south Republic mosaic spccies 3 Resistance to root rot ---- ---·-- --- MALAWI 146 146 - 300,000 1 Monoculture 1 Mosaic 1 Mcaly 1 No cuttings 2 1 Gcrm plasm 2 Mixed with 2 Witches bug 2 Treatcd collection and bean' broom 2 Whire O y seeds evaluation J Gra\S- throu~h 2 Agronomic triab hoppcr quaran unc J Screcnin~ for mosaic resistance 4 Screcning for HCN 5 Fertilizcr trials 6 Multiplication di ... ca'\c-frcc mat~rial SIERRA Numerous 7 S 17,000 1 Monocul!ure 1 Mosaic 1 Termites Non~ 4 Liule 1 Effects delayed LEONE (m06t (88%) 2 White 2 Animals weed control on popular) 2 Mixed (12%) thread 3 Grass- yield hoppcr 2 Effccts plant ing density on yield Effects fertilize" on yield 4 Effects mulchin~ on yield ---~ ----- - SRI LANKA Mauritius M u 22 most 10 90,000 1 Monocul!ure Nonc scriou~ Nonc senou~ No cuuings Severa! 1 Min. of 1 Production manioc 76 popular 2 Mixed with Central me al for animal coconuts in Agric. leed so me are as Res. Stn. 2 Chip storage 2 Univ. or 3 Methods removal Sri HCN from tubers Lanka TANZANJA Local 73 Mosaic- 6-34 3,550,000 1 Mixed (coasl) 1 Mosaic 1 Green 1 Cullings 2 Limitcd 1 Gcrm plasm Brazil 1 resis tan! 11 2 Mo noculturc 2 Bro wn mite through collection and High- (West) streak quarantine cvaluation yielding 18 2 Location tests Mite- 3 Evaluation resistant 13 polycross pro¡¡eny 4 Varietal improvement and multiplication 1 ntcrcropping --- -- --- -- - ·-- TOGO Local 10 800,000 Mixed with 1 Mosaic lmport Togolesc 1 Creation o f Foreign 60 legumcs, maize, 2 CertoJ/Jora forbidden Govt. instuute for root rice. sorghum spp. IRAT crops 3 Anthracnosc (France) 2 Study o f dose~s 4 Root rot GTZ J l mproved culture 5 Bacteria! (Germany) pracuces blight (new record) -------- UGANDA Local over 40 All 5-15 300,000 1 Mo noculture 1 C•rcoJpora 1 Mites Phyto- 2 Limited 1 Selection for and 2 Mixed 2 Cassava 2 Scales sanitary resist ancc to introduced cropping mosaic 3 White fly cerufocate ¡\f tanatoa viius 4 Grass- o r 2 Crop lo. Class .J - Severe mosaic patlem. severe distortion of two-th irds of leaflets. and general reduction of leaf size. Class 5 - Severe mosaic. severe distortion of four-tifth s or more of leaflets. twisted and misshapen leaves. and severc reduction of leaf size. 54 Part C Major Pests of Cassava in Africa and Preliminary Guidelines for Screening of Resistance K. Leuschner lnrernationa/lnstitute ofTropical Agriculture /hadan , Nigeria Cassava has. compared wi th o ther crops. relative ly few insect proble ms. Traditional methods of cultivation in smaller plots and mixcd cropping are less favourable to severe insect attack. Ho" ever. there are sorne insect and mi te specie~ which are. o r rnay become. major pests in Africa. The main pests o f cassava in Nigeria are thc whitc fly (Bemisiarabaci), the grasshopper Zonocerus mriegarus. and the red spider mi te Tetranychus te/arius. Outside Nigeria. the green ~pider mite (Mononychellus tanajoa) which occurred recentl y in Uganda and Ta nzania and a not-identified rnealy bug in za·ire have becomc serious pe~ts of cassava. Scirtothrips manihotis reponed from East Africa and termites can occas ionally be dangerous. Major pests of cassava Bemisia rabaci Distribution: Damage: Yicld losscs: Biology: Ho~t range: Control: Zonocerus l'ariegarus Distributio n: Damage: Yic ld losses: All over Africa Vector of cassava mosaic v irus; no damage by the insect itsel f Yicld Josses bctwcen 20 and 90o/r fo r CM B bctween 1930 and 1940: toda y cassava seems to live with the virus and is s till able to produce Bemisia lays its cggs on new developing leaves: larvae and pupae can be easi ly detected o n o lde r. fully grown leaves. One generation lasts about 4-5 weeks depending on climatic condítions (about 1 O generatio ns per year). The population level depends mo re on the physiological conditions o f thc plant than on climatic conditions Tomato. eggplant. tobacco. cas~ava. ~weet potato. cowpea, and many o thers Because of the wide host range and the fact that a s ingle insect can transmit thc rnosaic di se ase. o nly total immunity o f cassava would be of value. Becausc thi s rarcly can be obtained , either with insectic ides or in resistan! va rie ties. we should look o nly for varieties whic h do no t favour white tly dcvelopment Widespread in A frica bctwccn 10° north and south o f the equator Defoliation of plants bctwee n Novcmber and March: heavy outbreak can result in s tripping of the bark Yield can be affected whcn pla nts are younger than 1 month; 60% loss has been reported 55 Biology: Host range: Chemical control: Screening for resistance: Zonocerus is a dry-season pest. Eggs are laid in April and hatching occurs at the end of October and November. Two generation' are possible of which only the dry season one is of importance. Young hatchings do not touch cassava until they reach the 3rd instar. By thi s time cassava is often the only su itable food-plant becausc it can withstand drought. With incrcased cultivation of cassava Zono- cerus seems to gain in importance Cotton. citrus. swcct potato, cassava. tobacco. etc. Zonocems can be controlled by insect icides likc Gamalin 20 and Tenitroth ion Screening for rcs istance against Zonocems might be possiblc. Pre liminary tests showcd a prcfcrence for certa in varieties: 60444. lsunikakiyan. 581 O l. Screening i' not very easy beca use of high mobility and large numbers. but possible during carly moming. Po,- sible rating systcm: O - none: 1 - few resting only: 2- sl ight feedi ng; 3 - obvious fccding: 4 - numerous skeletonized leaves. no cxtcnsive bar k damage: 5- near dcfoliation with extensive bark feeding Tetranychus telarius (red spidcr mitc) Distribut ion: Worldwide Damage: Yield losses: Bio logy: Host rangc: Control: Resistance: Minor pests With increased mite population. ye llowish-white dots occur and later thc !caves turn yellowish-brown and cvcntually drop; heavy outbrcaks can cause complete defoliation; tips show light-coloured mottling, leavcs rcmain stunted. drop, and the growing point die' Unknown. but might be heavy if cassava is attacked by young a~ in thc case of Zonocerus Life cyc le varies frorn 7 to 12 days: 2-6 eggs are laid per da y with a total of 70 cggs per female. They hatch in a fcw days into six-legged larvac which rnolt into two recessive nymphal instars and an eight-legged adult. Thc mite is a dry-season pcst. It prcfcrs high temperature and low relat ive humidity. Mostly older and wdl- maintaincd plants are attacked. The mi te is spread by w ind. animals, and man Groundnuts . cassava. cotton. cucurbits. etc. Hcavy outbreaks might require the use of accricidcs likc ALadrin Dicrotophos. Dimethate fu nda!. etc. There may be differences in the susceptibility of cassava varieties to spider mites. but this has not yet been established. Possible rating system: O- no evidence of mi tes or damage: 1 -nil damage. only occasional speckling; 2- speckling localized. less than 25% of the leaf surface speckled; 3- spcckling exte nds length of thc mid-side and covcrs 30-50% of leaf surface: 4 - most leaves including young ones showing over 50% speckl ing; 5 - whole plant looks whitish wi th large amount of webbing and defoliation. Actual counting of mites is also possible but rather time-consuming. Minor pests should be carefully watched espec ially in connection wi th ncwly rcleased. high-yiclding and resistan! varie ties. 56 Appendix 3 A Note on the liT A Training Program Wade H. Reeves lnternational 1 nstitute of Tropical Agriculture /hadan, Nigeria To facilitatc communication and undcr~tanding. and for planning purposes, the Tra in ing Office of liTA has ddlned its area~ of re~pon~ibility and activity in the following terms. Seminars Weekly 'cminar!> are hcld at wh ich traincc and liTA staffprcscnt re\ultsoftheir research. thus providing a mean!> for trainee~ and other~ to beco me familiar" ith the "ork of the lnstitute a~ a whole and tu lcarn of the activities of scicntiMs working in other arca!> ufthe hum id tropic~. Conferenccs and workshops are al so hcld at thc institute and trainccs are cncouragcd to attcnd those of direct intcre!>t and conccrn to thcm. Postdoctoral fciiowship Fellowship\ are offcrcd cach ycar to ~clccted young scienti~b who ha ve rcccntly qualitled for advanccd dcgree~ tdoctorate level). The\t: fe llo\\ ships are offered for three reasons: ( 1) to affirm the re~earch !>kilb and oriental ion of young scicntísts by affording them the opportunity to \\Ofk with experienced research scicnt i ~h on relevan! probk:ms o f crop production in the lowland hum id tropics: ( 2) to assist thc in~titute to achieve it~ re~carch goa ls in priority arcas where insufficicnt regular staff ís availablc: and (3) to assist in thc idcnt ification and trai ning of staff for the in~titutc or for one of its cooperative program contracts. Research training Under the supervi~ion of membcr~ of liTA ·s scicntific staiT. trainees \\Ork on research problems that permit thc acquisition o r perfcction of needed re~earch ~"-ilb. All programs or projects are tailorcd to meet indiv idual nced' and all are planned in ~uch a way asto con tr ibute to liTA 's rescarch effort. Training in rcsearch is arrangcd both for postgraduatc dcgrce candidatcs and for those seeki ng dcgree>. Coopcrativc arrangcmcnt' have so far been madc wi th 18 uni versitics o f A frica. Australia. Europc. and North America for postgraduate degrcc candidates lo conduct the rcsearch portion of their degree rc4uirements at liTA. The insititute has also provided non-dcgree re>earch training for employec\ of 19 research institutes. universities. and departments of agriculture in 12 countrie~ of A frica and Asia. 57 Research scho/ars from many countries. including rhose shown from Zaire and Korea , are srudying cassam diseases and pesrs ar 1fT A. 58 Crop production technology and extens ion training This training leads toa thorough famil iarity with one c rop as well as with the methods of conducting an accelerated productio n campaign for that crop. The activities are designed to e reate production spec ialists who ha ve a high leve! of competency in ti ve arcas: ( 1) Technica/ competency Knowledge and understanding of the production practices involved and the physical e nvironment. This must includc the abi lity to diagnose correc tly and treat. if possible, typical problems and abnormalities of plants growing in the field . (2) Economic competency Ability to wcigh altemativc ofproduction input and marketing strategies. and to make recommendations in the light o f economic re lationships within the reality of the physical and soc ioeconomic systems in which thc farmer operates. (3) Scientijic competency Basic understanding of the philosophy of science and the ability to conduct s imple replicated ficld experiments wh ich objcctively test whcthcr an innovation is worthy of adoption under farm conditions. (4) Farming competency Willi ngness and skills to perform the range of physical tasks in volved in producing crops. Once the production specialist has acquired these four ski lis . a fifth becomes vital in his role of change agent. (5) Communication competency Ability to plan. prepare. and present adequatc mcssages for the farmer, thc crcdit agency. the input retailer. the market. or even the consumer. Thc broad objcctive of crop productio n tcchnology and extension courses ís to prepare extension leaders to promote. organize. and conduct crop productíon traini ng courses fo r extension agcnts and othcrs. thus mu ltiplying thc effects of their training at liTA. Training i~ done on a group basis. with emphasis on in-the-field experience . Adequate time is allowed in the classroom for lccture and discussion. 59 •. .. -