Exploitation of heterosis for simultaneous improvement in both grain and stover yields of arid zone pearl millet (Pennisetum glaucum (L.) R. Br.)

Abstract

The requirement for simultaneous increases in stover as well as grain yields in pearl millet in arid zone environmentsmeans that conventional selection for grain yield improvement through increased harvest index (HI) is not applicable to such environments. In addition, there is a need to retain the adaptive traits present in local landrace germplasm, so that new cultivars for the arid zone do not trade increased productivity for reduced yield stability and increased risk of crop failure. This research was designed to test the hypothesis that it will be possible to meet these requirements by exploiting heterosis for overall biomass production in topcross hybrids (TCH) made with adapted, landrace-derived pollinators and dual purpose male-sterile seed parents, which partition the extra biomass of their hybrids equally to grain and stover. General combining ability (GCA) estimates for seven landrace-derived populations/varieties, derived from multi-environment tests in arid zone environments, indicated that selection history played a large role in determiningGCA for both biomass and HI,with prior selection (for grain yield) favoringGCA for HI at the expense of GCA for biomass. A similar analysis of a set of male-sterile seed parents indicated a wide range of GCA for both grain and stover yields, with a similar tradeoff of GCA for one trait against GCA for the other. It was, however, possible to identify several parental lines with a positive GCA for biomass, achieved by a positive GCA for growth rate, and neutral GCA for HI, resulting in positive/ neutral GCA for both stover and grain yields. A limited test of the ability of parental GCA to predict heterosis in TCH indicated that heterosis for stover yield was closely related to pollinator GCA for stover yield, and heterosis for grain yield was related to both pollinator and A-line GCA for HI. The same test confirmed the original hypothesis that crosses of parents with positive GCA for biomass/growth rate and neutral GCA for HI could produce TCH with positive heterosis for grain yield without an off-setting negative heterosis for stover yield. The frequency of such parental lines was limited, however.