Proceedings of the XLV Italian Society of Agricultural Genetics - SIGA Annual Congress

Salsomaggiore Terme, Italy - 26/29 September, 2001

ISBN 88-900622-1-5

 

Poster Abstract

 

 

ASSESSEMENT OF VARIATION IN LATHYRUS SATIVUS L. AND SOME RELATED LATHYRUS SPECIES

 

POLIGNANO G. B.*°, UGGENTI P.*, ALBA E.**

 

* Istituto del Germoplasma, C. N. R., Via Amendola 165/A, 70126 Bari, Italy

° (Author for correspondence Fax 080 5587566 - germnp07@area.ba.cnr.it)

** Dipartimento di Biologia, Difesa e Biotecnologie Agro Forestali, Università della Basilicata, Potenza, Italy

 

 

grass pea, morpho-agronomic traits, multivariate analyses

 

Efforts to collect and conserve wild relatives, land races and cultivars of grass pea (Lathyrus spp.) have only recently begun in many countries all around the world. The main objective of the present work is to broaden our knowledge on the Bari  grass pea collection. The information gathered should enable us to create a data bank of great usefulness for future evaluation and utilisation programmes. Of the 332  entries of Lathyrus spp. collection only 280 have been included in this study. The entries were grown on red clay soil at Valenzano (Bari) from November 1999 to July 2000. 18 quantitative and qualitative descriptors were observed. For each trait descriptive statistics were determined. Quantitative data together with qualitative information, were subject to multivariate statistical procedures. Analysis of the principal components (PRINCOMP procedure SAS, 1987) on the average standardised values of quantitative traits was carried out to study the total variation of the analysed entries. To summarise between-species variation and to acquire information in accordance with classification of each entry a canonical discriminant analysis was applied (DISCRIM procedure SAS, 1987). A highly significant “among entries” variance was detected for all quantitative traits. Descriptive statistics for 8 quantitative traits relative to the species investigated are reported. Frequency distributions for 8 qualitative traits have also been considered. The species L. ochrus showed earliness, tallest plants, smallest seeds, lowest 100-seed weight and more seeds/pod; late flowering data, late maturity data, shortest plants and intermediate values for the other traits were observed for L. cicera. Biggest seeds and highest 100-seed weight have characterised the L. sativus entries.

 

Principal Component Analysis: the first three components accounted for 69% of the total variation. The first component, which explained 38% of the variance is positively associated with both seed and pod traits. A further 18% and 14% of the variation were explained by the second and third principal components. The highest weightings in the second component were for plant height and pod length, while in the third were for flowering time, days to maturity and pod width.

 

Canonical Discriminant Analysis: the generalised squared distances between Lathyrus species derived from the discriminating analyses and estimated on the first three principal component were all found to be significant. L. sativus and L. cicera were the less distant, and thus more similar, species; conversly, L. ochrus was the most distant ones; intermediate distances were observed for the unclassified entries group. However, this latter group did not diverge much from L. cicera and L. sativus. Also, the discriminant analysis was performed to calculate the percentage of entries misclassified present in the collection tested. In conclusion, this preliminary analysis showed wide differences among the Lathyrus spp. tested. The phenotypic similarity among entries of diverse Lathyrus spp. was well defined by both principal component and canonical discriminant analyses. The presence of high diversity within the L. sativus germplasm suggests use of the primary genepool as a valuable source of genetic material for grass pea improvement. Thus, the Bari collection should be characterised and evaluated further in order to get additional information useful in to identify a core set of L. sativus entries for future breeding work.