Mating system in a base population of Jatropha curcas L

Abstract

The aim of this study was to investigate the mating system of 30 progeny of a base population of Jatropha curcas L, using microsatellite markers. The base population was established as a progeny test in Selviria-MS, using a randomized block design with 30 progenies, three replications of five plants per plot, spaced 3 x 2 m. All trees of the test were sampled for microsatellite analysis. The estimate of multilocus outcrossing rate was high, but significantly lower than unity (t(m)= 0.888). The outcrossing rate varied among trees matrices from 0.645 to 0.986, which demonstrates that the selfing rate was higher in some trees. The estimated rate of mating among relatives was substantial and significantly higher than zero (t(m) - t(x) = 0.426), suggesting that the populations of origin of progenies may present spatial genetic structure. The paternity correlation was significantly higher than zero (r(p(m)) = 0.646), indicating a high proportion of full-sib within progeny. The coefficient of coancestry within progeny was higher than expected in half-sib progeny (Theta= 0.237) and the effective size was lower than expected in progenies originated from panmitic populations (N-i = 2.03). Therefore, to collect seeds for forest improvement, ex situ genetic conservation and environmental reforestation, seeds must be collected from at least 74 seed trees.