a model as failing when a strong majority of studies reject its predictions. Conversely, models that are consistently supported by data are valued for accurately tracking evolutionary outcomes. By such standards, and contrary to its critics, kin selection theory has had major successes. However, contrary to its apologists, kin selection theory also has had some major failures. I will briefly review within the social Hymenoptera two successes, (i) sex investment ratios and (ii) conflict over reproduction by workers (i.e., worker policing), and two failures, (i) caste-biasing nepotism and (ii) reproductive skew theory. I will thereafter consider the mechanism for how kin are recognized as the driving force for the observed pattern of success and failure.
These four topics are chosen because I believe the underlying theory for the kin selective predictions is sound and that the possibility for kin nepotism to evolve is at least potentially present. This differs from two other cases, where kin selection predictions are suggested to have failed: the haplodiplody and monogamy hypotheses (Nowak et al., 2010). First, cooperative breeding has repeatedly evolved in the haplodiploid Hyme-noptera. Haplodiplody creates a genetic asymmetry, such that a female is more related to her full sister (r = 0.75) than she is to her own offspring (r = 0.5). Therefore, if a singly mated mother produces a female-biased offspring sex ratio, it is genetically more advantageous for a daughter to help her mother raise more sisters. However, the balance of evidence from existing species where cooperative breeding is facultative finds that such species are not monogamous, do not predictably bias sex ratios toward females, or both (Bourke and Franks, 1995). Hence, the haplodiploidy hypothesis is not a robust test of kin selection because the required patterns of genetic relatedness likely did not exist in the putative ancestors of eusocial species (Nonacs, 2010). The second example is the “monogamy hypothesis,” where cooperative breeding is predicted to be more likely to evolve in species where family groups are full siblings because of monogamy (Boomsma, 2009). However, a gene-based model for the evolution of cooperation found that helping actually often tended to spread more rapidly through populations with polygamy (Nonacs, 2011). This may be an instance where the underlying kin selection model actually produces erroneous predictions [as postulated by Nowak et al. (2010)].
SUCCESSES: BIASED SEX INVESTMENT RATIOS AND
PARENTAGE OF MALES THROUGH WORKER POLICING
Another genetic consequence of the asymmetrical relatedness patterns attributable to haplodiploidy is that with one singly mated queen (i.e., monogynous with monandry), workers are more closely related to full sisters than to their brothers. This preference should drive female investment