Divergência genética e avaliação de modelos de degradação em genótipos de amendoim forrageiro (Arachis pintol)

Abstract

The nutritional value of ten genotypes of forage peanut (Arachis pintoi): 13251, 15121, 15598, 30333, 31135, 31496, 31534, 31828, Itabela and RIO were evaluated. The statistical design used was a randomized block design with ten treatments (genotypes) and three replicates. At the CEPLAC Zootecnia station, in Itabela-BA, they were cultivated and harvested in two seasons (dry period, with lower rainfall, and rainy season, referring to the period with higher rainfall). The dry matter production (kg / ha), the production of green matter (kg / ha), crude protein, neutral detergent fiber, acid detergent fiber, insoluble protein of acid detergent from crude protein (% PB), insoluble protein of acid detergent from dry matter (% DM) and accumulated gas production adjusted to the Gompertz, Logistic, Brody, Von Bertalanffy and Bicompartmental Logistic models and evaluated by R², QMR, AIC and BIC were evaluated. The multivariable analysis used was the analysis of clusters with cophenetic correlation coefficient and the Rand index to verify the quality of adjustments and the number of groups. In the period of greatest rainfall, genotypes 15121, 15598, 30333 and 31496 stood out as the ones with best productivity and nutritional value. In the period of lower precipitation, genotypes 31828 and Itabela stood out as the best genotypes for feeding ruminants. The best model adjusted for both genotypes was the Logistic Bicomportamental model because it presented lower AIC and BIC. Therefore, it has superior adjustment and should be used to describe cumulative curves of gases in forage peanut (Arachis pintoi).