Recent tests of theory for sex ratio evolution in the social Hymnoptera have focused on the common occurrence of bimodally distributed (split) colony sex ratios.  A number of these studies have tested whether workers bias the sex ratio according to variable relatedness asymmetry among colonies (Boomsma J.J. and Grafen A. 1991. J. Evol. Biol. 4: 383-407), and a number of studies support this hypothesis.  However, split sex ratios also occur in populations where the relatedness asymmetry hypothesis does not apply.  New theory shows that when queens bias the primary (egg) sex ratio and workers bias the secondary (adult) sex ratio, that split sex ratios can often evolve in the absence of variable relatedness asymmetry because of how queen and worker biasing mechanisms change the colony sex ratio (Helms K.R. et al. 2005. Evolution 59: 2626-2638).  It also shows that split sex ratios can occur because of the evolution of either two divergent queen or two divergent worker biasing strategies.

A foundation of this theory is that the degree that queens and workers can adaptively bias the sex ratio is determined by egg availability.  In Helms et al. (2005), workers exhibited pure biasing strategies, and when queens produced few (if any) female eggs, workers programmed to bias strongly against males could not facultatively reduce bias to avoid eliminating males that could have been reared with available resources.  In the present study, we analyze the case where workers can assess resource availability and facultatively decrease male elimination when it is adaptive to do so.  The results show that split sex ratios still evolve frequently; however, they most frequently result from the evolution of two divergent queen strategies, which results in near-equal investment in the sexes at the population level.  These results are consistent with those found in nature in some ants.