Principal component analysis in Paspalum germplasm chacarterization.

In a survey, though principal component analyses, for representative descriptors to be used in Paspalum gerplasm characterization, 16 agronomic traits, 16 reprosuctive traits and 21 vegetative traits werw evaluates in 32 accessions comprising Paspalum plicatulum (19) and Paspulum compressifolim (13) from the institutional germoplasm collection. ANOVA for agronomic descriports showed significant differences among accessionsm except for percentage of dry matter in winter (DMW) which was not significant. Princiapal component anlyses allowed reduction of 69% and 62%, respectively, in the intial number of descriptors, and the election of the most important for agronomic traits: DMYS, DMYW, RGS, CPPW and NDFW (dry matter yield in summer, dry matter yield in winter, regrowth score in summer, crude protein percentage in winter and neutral detergent fiber in winter respective); reproductive descriptors: NSG, PE, PFA, PS and LE (number of seeds per gram, pubescense of the escape respective); and vegetative descriptors: PLDI, PLL, PINDE, LL, LWH, SWB, SWT (pubescence of lealf - distribuition, pubescence of lealf length, lealf winth - half, sheath width - base and sheath width - top,respective)