II. 23. Numerical taxonomy as an aid to barley germplasm collection.
J. L. Molina-Cano, La Cruz del Campo, S. A., Malting and Brewing Co., Dept. of Barley Breeding, Apartado 53, Sevilla, Spain, "R"
The need for collecting, evaluating, and maintaining germplasm of crop plants has been repeatedly emphasized during the past few years. However, many problems need to be solved before this task can be performed with the highest effectiveness. In this connection there are certainly plenty of economic reasons for limiting the number of accessions in working collections and also for devising methods to detect duplicate phenotypes (Moseman and Craddock, 1976).
Numerical taxonomy may help to solve the problems outlined above.
To study the pattern of phenetic variation in the Spanish indigenous material of two-rowed barley, six presumed populations were collected in the summer of 1973 in the province of Soria (NE of Spain). With our previous knowledge of the distribution of the indigenous two rowed barley germplasm in this province (Molina-Cano, 1976b), the six locations were at first viewed as a satisfactory sample. Several spikes were taken at random from each sample and 66 of them were grown on a spike-line basis and studied together with 4 European cultivars as controls (Beka, Ingrid, Rika and Zemyr), by observation of 42 morphological and biochemical characters. These characters were chosen taking into account its nearly null response to environmental variations, at least from a comparative standpoint. Each character was coded into several states.
As a complementary tool in germplasm preservation, samples of the six populations were stored by using a long-term encapsulation method devised by Gomez-Campo (1972).
The Original data matrix obtained (Sneath and Sokal, 1973), by the observation of the character states possessed by the studied phenotypes, was subsequently subjected to several numerical taxonomic calculations, e.g. Product-moment correlation and Distance coefficients of similarity; UPGMA, Single Linkage and Complete linkage clustering strategies; and Principal Component Analysis (Sneath and Sokal, 1973). Some of these methods had been previously tested on a set of 38-European barley cultivars (Molina-Cano, 1976a), where most of the clusters formed proved to be correct in relation to the known pedigrees of the involved genotypes.
To exemplify the results obtained with the Spanish land material, one of the phenograms (Fig. 1), and one of the principal component diagrams (Fig. 2), are shown. These show no evidence of correlation between the former six land populations and the clusters formed at an intermediate level of similarity. There is, however, a clear separation between the Spanish genotypes and the four European controls. Though several groups could be detected within the Spanish material, no clear relationship was found to exist between them and the former populations. The phenogram shows, nevertheless, two pairs of identical phenotypes whose members were collected at different sites, Almazin-Judes and Almazin-Retortillo, respectively.
As general conclusions, we point out the following:
a) To collect the two-rowed barley germplasm in the studied zone, several random samples collected over a number of locations, also chosen at random, would prove enough.
b) The pattern of variation expressed by our results, suggests that the number of locations to be sampled could be smaller than previously thought.
c) Numerical taxonomic techniques proved to be a reliable tool to evaluate barley germplasm through the study of phenetic variability between populations. By using these techniques, it was even possible to detect duplicate phenotypes shared by different populations.
References:
Gomez-Campo, C. 1972. Preservation of West Mediterranean members of the Cruciferous Tribe Brassiceae. Biol. Conserv., 4(5):355-360.
Molina-Cano, J. L. 1976a. A numerical classification of some European barley cultivars (Hordeum vulgare L.s.l.). Z. Pflanzenzuchtg., 76: 320-333.
Molina-Cano, J. L. 1976b. Estudios sobre el germplasm indigena de cebada d'stica en la provincia de soria. Ph.D. Thesis. Universidad Politicnica de Madrid. Escuela T.S. de Ingenieros Agrdnomos. 208 pp.
Moseman, J. G. and J. C. Craddock. 1976. Genetic basis for collecting, evaluating and maintaining barley germplasm. In Barley Genetics III. Ed. by H. Gaul. Verlag Karl Thiemig. Munich. pp. 51-57.
Sneath, P. H. A. and R. R. Sokal. 1973. Numerical Taxonomy. Freeman. San Francisco and London. pp. 114-253.
