Banana Cultivars - Classification of Cultivars

Classification of Cultivars

Banana plants were originally classified by Linnaeus into two species, which he called Musa paradisiaca – those used as cooking bananas (plantains), and M. sapientum – those used as dessert bananas. The primary center of diversity of cultivated bananas is Southeast Asia. Botanical exploration of this area led to many more species being named, along with subspecies and varieties. However, this approach proved inadequate to deal with the large number of cultivated varieties (cultivars) which were discovered, and many of the names later proved to be synonyms. Furthermore, it was discovered that most cultivated bananas are actually hybrids between two wild species, M. acuminata and M. balbisiana, and that Linnaeus' two "species" were both this hybrid, whichis now called M. × paradisiaca. Unlike the wild species, which have seeds, cultivated bananas are almost always seedless (parthenocarpic) and hence sterile, so they have to be propagated vegetatively.

In 1955, researchers Norman Simmonds and Ken Shepherd proposed abandoning traditional Latin-based botanical names for cultivated bananas. This approach foreshadowed the International Code of Nomenclature for Cultivated Plants which, in addition to using Latin names based on the International Code of Nomenclature for algae, fungi, and plants, gives cultivars names in a currently spoken language, enclosed in single quotes, and organizes them into "cultivar groups", also not given Latin names.

Banana cultivars derived from M. acuminata and M. balbisiana can be classified into cultivar groups using two criteria. The first is the number of chromosomes: whether the plant is diploid, triploid or tetraploid. The second is relationship to the two ancestral species, which may be determined by genetic analysis or by a scoring system devised by Simmonds and Shepherd. A cultivar is scored on 15 characters, chosen because they differ between the two species. Each character is given a score between one and five according to whether it is typical of M. acuminata or of M. babisiana or is inbetween. Thus the total score for a cultivar will range from 15 if all characters agree with M. acuminata to 75 if all characters agree with M. balbisiana. Intermediate scores suggest mixed ancestry: for example, 45 would be expected for diploids with equal genetic contributions from both species.

Groups are then named using a combination of the letters "A" and "B". The number of letters shows the ploidy; the proportion of As and Bs the contributions of the ancestral species. The AAB Group, for example, comprises triploid cultivars with more genetic inheritance from M. acuminata than M. balbisiana. A character score of around 35 is expected for members of this group. Within groups, cultivars may be divided into subgroups and then given a cultivar name, e.g. Musa AAA Group (Cavendish Subgroup) 'Robusta'.

In practice, the scoring system and the associated grouping is not as straightforward as the Simmonds and Shepherd naming system implies. For example, a member of the AAB Group should have a score about one third of the way between M. acuminata and M. balbisiana (i.e. about 35) if one third of its chromosomes come from M. balbisiana. However, the cultivars 'Silk' and 'Pome', both classified in the AAB Group, scored 26 and 46 repectively. The cultivar 'Pelipita' is placed in the ABB group, so should have 11 of its 33 chromosomes derived from M. acuminata. However, a technique called "genomic in situ hybridization" (GISH) showed that actually only 8 chromosomes were of this origin. Other lines of evidence suggest a more complex genome structure is present in other banana cultivars, so the group names should not be taken at face value.