Numerical analysis of variation patterns in the genus Stylosanthes as an aid to plant introduction and assessment

Abstract

An experiment is described in which a large collection of Stylosanthes introductions, comprising 14 named species, was grown as spaced plants and simple morphological and agronomic data recorded. After applying various methods of numerical analysis, and integrating the results, it was found that species and forms could be delineated. The importance of non-stable agronomic characteristics in the classification of the introductions is explored and the taxonomic status of the resultant groups briefly examined; the use of these systems for plant assessment programmes is discussed.

Geographical patterns of genetic variation in two species ofStylosanthesSw. using amplified fragment length polymorphism

“…S. viscosa is a perennial diploid species (2 n = 20) with a wide distribution in tropical America. It occurs in a wide range of environments, from the humid tropics to the semiarid tropics (Burt et al . 1971; Williams et al .…”

“…Some of these have included the use of rhizobial affinities (Date et al . 1979), M–A characters (Burt et al . 1971), seed protein analysis (Robinson & Megarrity 1975; Robinson et al .…”

A method for establishing character inter-relations in plant collections, and its possible application to plant improvement programmes

“…We suggest that the method described in this paper can offer at least a partial solution to this problem, providing that the data-set available contains not only morphological information, but also information of agronomic or economic interest. The importance of recording adequate morphological information has been stressed in BURT et al (1971); and whereas the experiments needed to establish estimates of yield or disease resistance are often expensive and time-consuming, morphological characters can be recorded easily and quickly, often from observations in a quarantine glasshouse (BURT &WILLIAMS, 1979a), or even from herbarium specimens. A method which can relate the two has obvious advantages.…”

The mixture method of clustering applied to three-way data

“…They can also be characterized by the performance values for a single attribute measured in a number of environments (G × E) matrix. Many methods of clustering have been applied to such two-way arrays to provide an appropriate grouping of the genotypes (Burt, Edye, Williams, Grof and Nicholson 1971;Mungomery, Shorter and Byth 1974;and Byth, Eisemann and De Lacy 1976). Such analyses have been very useful to plant breeders, but the restriction of being able to handle only two-way arrays has been a limitation.…”