African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 7, Num. 1, 1999, pp. 1-7

K. P. BAIYERI¹, A. TENKOUANO, B.N. MBAH¹ and J.S.C. MBAGWU¹

Plantain and Banana Improvement Program, International Institute of Tropical Agriculture, Oyo Road, PMB 5320, Ibadan, Nigeria
¹Faculty of Agriculture, University of Nigeria, Nsukka, Nigeria

Post-harvest characteristics of 36 Musa genotypes were evaluated under two cropping systems. Genotypes included AAA x AA, AAB x AA and ABB x AA (or BB) and their landraces AAA, AAB and ABB grown under monocropping and in the alleys of natural multi-species hedgerows. Significant differences (P < 0.01) were found among genomic groups for bunch and fruit weights, pulp yield, dry matter content, pulp firmness, shelf-life, and market potential index. Cropping systems were also significant for all traits except for pulp firmness and shelf-life. Significant interactions between genotypes and cropping systems were found for all traits except dry matter content and pulp firmness. Tetraploid hybrids had lower pulp firmness but higher shelf-life and market potential index than the triploid genotypes. Among the triploid landraces, AAA genotypes had longer shelf-life but expressed the lowest market potential index due to their lower pulp yield and dry matter content. The AAB x AA hybrids consistently expressed higher pulp firmness, shelf-life and market potential index than other tetraploid hybrids. The alley-cropping system resulted in higher values for most post-harvest characteristics than the monocropping system.

Key Words: Bananas, biological value, market potential index, ploidy shift, posthavest traits

Nous avons évalué les caractéristiques post-récolte de 36 génotypes de Musa comprenant des écotypes AAA, AAB et ABB et des hybrides AAA x AA, AAB x AA et ABB x AA (ou BB) dans deux systèmes culturaux: monoculture et culture intercalaire. Des différences significatives (P < 0.01) ont été observées entre les génotypes pour le poids du régime, le poids des fruits, le rendement en pulpe, la teneur en matière sèche, la consistance de la pulpe, la durée de vie verte et l'indice de marché. On a également observé un effet siginificatif des systèmes culturaux pour tous les traits sauf la consistance de la pulpe et la durée de vie verte. De màme, l'interaction entre génotypes et systèmes culturaux était significative pour tous les caractères observés, hormis la teneur en matière sèche et la consistance de la pulpe. Par rapport aux génotypes triploïdes, les hybrides tétraploïdes ont exprimé une durée de vie verte plus longue et un indice de marché plus grand, mais leur pulpe était moins consistante que celle des triploïdes. Parmi les cultivars triploïdes, les types AAA ont exprimé la meilleure durée de vie verte mais aussi l'indice de marché le moins bon, à cause de leur faible rendement en pulpe et de leur faible teneur en matière sèche. Les hybrides AAB x AA ont invariablement présenté une meilleure consistance de la pulpe, une meilleure durée de vie verte et un meilleur indice de marché que les autres hybrides tétraploïdes. On a obtenu en général de meilleures caractéristiques post-récolte en culture intercalaire plutôt qu'en monoculture.

Mots Clés: Banane, valeur biologique, indice de marché, changement de niveau de ploïdie, caractéristiques post-récolte

Bananas and plantains (Musa spp.) are important cash and subsistence crops in the developing world (Dadzie and Orchard, 1996; Robinson, 1996). In recent years, increasing pest and disease pressures have impelled genetic improvement of the species. High-yielding, pest and disease resistant tetraploid hybrids were obtained from inter-specific crosses between triploid and diploid accessions (Rowe and Rosales 1993; Robinson, 1996; Vuylsteke et al., 1997). However, post-harvest characteristics, particularly dry matter content and storability, may have been altered as a consequence of introgression of genes from wild diploid germplasm into the triploid landraces and the resultant change in ploidy level (Vuylsteke et al., 1997).

The concept of marketable yield has been used by potato (Solanum tuberosum L.) breeders to designate the fraction of the production that is accounted for by tubers that are suitable for consumers based on a weight criterion (Ortiz et al., 1988). Market potential of plantains and bananas may be defined, following Bressani's (1976) approach, as the utilisable fraction of the yield which is a function of pulp weight of the harvested fruit, dry matter content of the pulp, and shelf-life of the fruit.

Dry matter content has been considered as an indicator of cooking quality of bananas and plantains (Dadzie and Orchard, 1996), and may be positively correlated with storability (Ferris et al., 1996). Dry matter content determines the fraction of a harvested crop that is available for processing into food products (Sanchez et al., 1968), i.e., the biological value of the crop. However, shelf-life may determine the market potential of the crop, since it dictates processing options and the array of food products that can be derived from the crop.

The extent to which ploidy shift and introgressed alien genes, and their interaction with the crop production environment, affect biological and marketable yield of plantain and banana have not been investigated. The objective of this study was to assess the effect of cropping systems, and inter- and intra-genomic variability on biological value and market potential of Musa hybrids and landraces.

Thirty six genotype representatives of Musa spp. Subgroups (Table 1) were evaluated under sole cropping and alley cropping, with multiple species hedgerows, at the Onne Research Station (4° 43'N, 7° 01'E, 10 m a.s.l.) Of the International Institute of Tropical Agriculture (IITA), in southeastern Nigeria. The station is located in a degraded rain forest swamp, characterised by an Ultisol derived from coastal sediments and 2400 mm unimodal annual rainfall (Ortiz et al., 1997). Daily temperature and solar radiation average 27° C and 14 MJ m^-2, respectively, resulting in high relative air humidity (Ortiz et al., 1997).

TABLE 1. List of genotypes used in the study

FHIA and SH series are hybrids from Fundaçion Hondureña de Investigacion Agricóla, Honduras; PITA and BITA series are hybrids from the International Institute of Tropical Agriculture, Nigeria; EMB and EMC hybrids are from EMBRAPA and EMCAPA in Brazil respectively, others are landraces

Planting was done on 19 and 20 June 1995 utilising a 6 x 6 simple lattice design. Each genotype was grown in a single-row-plot of five plants per replication. Cultural practices used were those described by Swennen (1990). At maturity, banana bunches were harvested and weighed. Total fruit weight per plant was also measured after removal of the peduncle; fruits were separated into two samples. Fruit firmness was measured from one sample using a digital force gauge (Lloyd Instruments, model DFS 250) and fresh pulp weight measured after removal of the fruit peel. The pulp was oven-dried (50° C, 72 hours) and dry matter content was determined as the ratio of dry to fresh weight.

Fruit ripening was monitored using the second sample under laboratory conditions. In this study, three ripening stages were defined: Stage 1 was the phase from complete greenness (unripe fruits) to yellowing of fruit tip, Stage 2 was the early ripening phase with up to 60% yellowing of the fruit, and Stage 3 was reached when fruits were fully ripe with 100% yellowing of the peel. In some West African countries, the market price of ripe fruits is higher than that of the green (Ahenkora et al., 1996). However, there are more food processing options with green fruits as opposed to ripe ones. For example, Stage 1 plantain fruits may be processed into flour, chips and fried products. However, processing into flour is no longer feasible with fruits that have reached the second stage, while fried and/or boiled products are the only options for fruits in the third stage. Therefore, the duration of each stage was recorded for all genotypes, the sum of which was considered as the shelf-life. Market potential index was then estimated from biological yield components and the number of days in each ripening stage using the following formula:

U_{i( j )} = P_{iå j} (S_ij.V_j) [Eqn. 1]

Where U_{i( j )} is the estimated market potential index of the i^th genotype, P_i is the dry pulp yield of the i^th genotype, S_ij is the number of days spent in the j^th ripening stage by the i^th genotype, and V_j is a subjective index value attributed to the j^th stage. In this study, an index of 1 was given to Stage 1, 0.75 to Stage 2, and 0.5 to Stage 3. Clearly, the value attributed to each stage will vary with cultural and socio-economic contexts. Consideration of such aspects could allow construction of an aggregate market potential index for Musa genotypes. For the present study, a single socio-economic context was assumed.

Statistical analysis was carried out on plot means basis due to unequal number of observations per plot (Piepho, 1997). Due to missing values the statistical model used for the analysis was that of randomized complete block design instead of lattice design. Data were subjected to analysis of variance and separation of means using the GLM procedure in SAS (SAS Institute, 1992).

Ploidy and genomic group effect. Significant differences were found among genotypes, both within and across genomic groups, for all traits (Table 2). Intra-genomic differences accounted for a larger proportion of variation in all traits compared to inter-genomic differences.

Cooking bananas (ABB) produced the highest bunch and fruit weight compared to dessert bananas (AAA) and plantains (AAB). Similarly, cooking banana hybrids expressed higher bunch and fruit weight than other hybrids (Table 3). Higher yield was associated with a larger number of hands and fruits per bunch in both triploid and tetraploid backgrounds. The triploid genotypes had higher dry matter content (DMC) than the tetraploid hybrids. More specifically, DMC was highest in plantains (34.5%) and lowest in AAA x AA hybrids (23.7%). Similarly, pulp firmness was greater in triploid landraces compared with tetraploid hybrids. However, bunch and fruit weight of tetraploid hybrids were higher than those of the triploid genotypes.

The shelf-life and market potential index of tetraploid hybrids were greater than those of the triploid landraces. However, significant intra-ploidy variation was also observed for these traits. Duration of Stages 1 and 2 was longer for plantain and dessert banana hybrids than for the cooking banana hybrids and triploid genotypes (Fig. 1). On the other hand, duration of Stage 3 was similar irrespective of ploidy level (Fig. 1). Plantain hybrids (AAB x AA) had a longer shelf-life at each ripening stage, portraying a longer time for processing options at each of those stages than the other tetraploid and triploid genotypes. Shelf-life of triploid genotypes varied from 12 days for the plantains to 15 days for the dessert bananas (Table 3). However, dessert bananas had the lowest market potential index, due to their low pulp yield, pulp firmness and dry matter content. Among the tetraploid genotypes, the AAB x AA hybrids consistently expressed higher pulp firmness, shelf-life and market potential index (Table 3) which contrasted with the low market potential index of the AAB progenitors.

TABLE 2. Mean squares from analysis of variance of post-harvest traits of 36 Musa genotypes grown under two cropping systems in Nigeria