Morphological characterisation of East African AAB and AA dessert bananas (Musa spp.)
MetadataShow full item record
Onyango, M.; Karamura, D.; Keeley, S.; Manshardt, R.; Haymer, D. -2011-Morphological characterisation of East African AAB and AA dessert bananas (Musa spp.) -ISHS 897-p. 95-105
Permanent link to this item: http://hdl.handle.net/10568/42417
Internet URL: http://www.actahort.org/books/897/897_9.htm
Within the small-fruited Musa AAB dessert bananas in East Africa are cultivars that have potential for export, and the diploid AA bananas known in Kenya as "Muraru" are socially valued in the country. It is important to be able to distinguish these cultivars from other similar cultivars and identify the various subgroups within the AAB and AA genome groups, and possibly come up with some recommendations for future economical considerations. Objectives of this study were: 1) to identify morphological characters that distinguish a) the various subgroups of AAB dessert bananas found in East Africa, and b) the Muraru from other cultivated AA bananas; 2) to evaluate the relationship among the AAB and among the AA Muraru dessert banana groups of East Africa in relation to other bananas. Forty-three (43) cultivars of AAB, AA groups and outgroups from a large banana collection at the Kenya Agricultural Research Institute, Kisii were characterised in 2007 using morphological traits. Morphological data were collected using 84 characters derived from a modified version of the descriptors for bananas developed by Bioversity International in conjunction with CIRAD. Techniques of multivariate analysis were employed. Based on unweighted pair group using arithmetic mean (UPGMA), two major clusters of Musa acuminata-derived cultivars (AAs and AAAs) and hybrids of Musa balbisiana and M. acuminata (AAB) were produced. Within the major clusters were subclusters conforming to various subgroups. Within the AAB dessert cluster, four distinct subclusters were formed, i.e. Sukari Ndizi, Prata, Mysore and Silk. Muraru also formed a well-defined cluster. Thirty-three (33) characters contributed 71% of the total variation within the 43 accessions on the first and second principal components, allowing separation of clusters corresponding to genome groups and subgroups. The analysis further revealed that morphological traits, particularly of male bud, fruit and sucker, can be used to make distinctions within genome groups and subgroups, and to isolate various subgroups within genome groups. Morphological traits can be used confidently to describe various banana subgroups.