Failure to detect pleiotropy of maternal traits in a rare honey bee mutant is not a test of ground plan hypothesis that explains origins of social behavior

Oldroyd and Beekman attempt to test the forager reproductive ground plan hypothesis (forager RGPH) using a selected strain of anarchistic worker honey bees. The authors conclude that their experiment does not support the forager RGPH, and they strongly criticize studies in support of the hypothesis that have used strains selected for high and low levels of pollen hoarding. We show here that the criticism is poorly founded. The results of Oldroyd and Beekman are flawed by contradicted assumptions about the reproductive behavior of anarchistic bees, by lack of variation in ovary phenotypes, by inference from insufficient sample sizes, and by the use of incomplete datasets from foraging specialists in their statistical analyses. Oldroyd and Beekman (2008) compared the ovary size, ovarian activation level, and foraging behavior of worker bees from commercial Australian stocks and a strain selected for abnormal reproductive behavior. In their study, they claim to test the RGPH of honey bee forager behavior proposed by Amdam et al. (2004, 2006). Normal worker honey bees are essentially sterile females. Yet the forager RGPH posits that variation among foragers in their age at foraging onset and in their bias toward collecting pollen (a protein source) or nectar (a carbohydrate source) is explained in part by variation in a gene network that can synchronize female foraging choice and maternal provisioning behavior with ovarian physiology. This pleiotropic regulatory network and its link to foraging behavior in workers derive from the reproductive biology of ancestral solitary insects (Amdam et al. 2004; 2006). Support for the forager RGPH comes from a battery of association studies that used unselected " wild type " sources of bees