Hamilton’s idea that haplodiploidy favors the evolution of altruism – the haplodiploidy hypothesis -- relies on the relatedness asymmetry between the sexes, caused by the sex-specific ploidies. Theoretical work on the consequences of relatedness asymmetries has significantly improved our understanding of sex-allocation and intra-colony conflicts, but the importance of haplodiploidy for the evolution of altruism came to be seen as minor. However, recently it was shown that haplodiploidy can strongly favor the evolution of eusociality, provided additional “preadaptations” are also present, such as the production of multiple broods per season and maternal ability to bias offspring sex ratios. These results were obtained assuming no influence of workers on the sex ratio, even though worker control of the sex ratio is known to occur. Here we model the evolution of sex-specific fratricide as a mechanism of worker control over the sex ratio. We show that fratricide can facilitate the initial evolution of helping. However, fratricide can also hamper the evolution of unconditional help. Instead, social polymorphism evolves, a mixture of helping and dispersing offspring. Finally, we show that the co-evolution of sex-allocation strategies of workers (fratricide) and queens leads to a split production of the sexes, with some colonies specializing in males and others in females. Thus, the model predicts that fratricide spawns a diversity of co-existing life cycles that strongly vary in degree of sociality and sex ratios.

All the data for the paper was generated by running individual based simulations written in c++ language. The code both for generating the data and for figure of the published article can be found in here: DOI: 10.5281/zenodo.3361665

Files names starting with the identifier "evol" correspond to the evolutionary trajectories, while files named starting with "dist" correspond to phenotypic distributions at the end of the evolutionary process. After the identifier, file names contain 3 numbers, each one followed by the information of the simulation they communicate. The first number gives the seed used in the random number generator for that particular simulation, it is followed by the word "seed". The second number provides the value used in parameter phi multiplied by 10, for convenience, and it is followed by the word "phi". Finally, the last number gives the value used in parameter b multiplied by 10, and it is followed by the letter b. 

The files of evolutionary trajectories contain data on the central tendency and spread of the 5 phenotypic trait along the evolutionary process. The 16 columns in the files correspond to: generation, mean for z1, mean for z3, mean for z5, mean for h, mean for omega, standard deviation for z1, standard deviation for z3, standard deviation for z5, standard deviation for h, standard deviation for omega, first quartile for z1, first quartile for z3, first quartile for z5,  first quartile for h,  first quartile for omega,  third quartile for z1,  third quartile for z3,  third quartile for z5,  third quartile for h,  third quartile for omega. 

The files with the phenotypic distribution at the end of the evolutionary process contain values for the traits of different individuals in the population. The five columns correspond to: z1, z3, z5, h, omega.