Unidirectionality in the phylogeny of social organization, with special reference to birds

Ecological explanations for the diversity in parental care patterns and social organization in certain taxonomic groups of birds are not fully satisfactory. They need to be supplemented by phylogenetic explanations. In this article I discussed some aspects of the latter type of explanations, especially the difference between probabilities of certain evolutionary transitions occurring in the one and in the opposite direction.To explain the diversity in parental care and mating patterns in waders and related groups, I presented a model on the phylogenetic pathways in the evolution of parental care and social organization in birds. It departs from an ancestral state with pure male parental care and social organization in birds. It departs from an ancestral state with pure male parental care which may evolve via "uniparental care" (male cares, but female cares if male deserts) and "double clutching" towards biparental care with similar roles, polyandry and pure female parental care (polygyny, promiscuity). I have argued that certain transitions in this model (especially those from uniparental care and double clutching towards biparental care with similar roles and towards pure female parental care) may easily occur in the given direction, but not in the opposite one.The model predicts that pure male parental care and related patterns may be preserved in various lineages and may be associated with several other patterns in related species. It also predicts that pure female parental care is, in many instances, a final stage in the evolution, and hence quite often combined with pure female parental in related species. To investigate the value of the model I tested its predictions for the phylogenetic trees of (1) arctic sandpipers, (2) the complete order of Charadriiformes, and (3) birds in general. All predictions were met. To investigate the likelihood of the model I considered to what extent predictions by alternative models were met. These models either ignored the effect of phylogenetic factors, or departed from alternative ancestral stages. The fit of the data seemed to be bad with the predictions of all of these models. Thus, the original model presented in this article must be considered as a probable reflection of the phylogeny of parental care and social organization in birds.