Microevolution of the Chibcha-speaking peoples of lower Central America: rare genes in an Amerindian complex.

Models are developed for the survival, history, and spread of variant alleles, in order to consider what can, and what cannot, be inferred from this type of data. The high variances of the processes involved, and questions of sampling, place severe limitations on inferences. Nonetheless, by combining information on a number of rare variants observed in a group of interrelated populations, reliable qualitative inferences are possible. These ideas and models are developed in the context of data on five rare variants and six private polymorphisms observed in eight Chibcha-speaking tribes of Costa Rica and Panama. The decline and fragmentation of the Amerindian populations of Central America over the last 300 years create considerable difficulties in attempting inference of past genetic events. However, these tribes have been well studied genetically, anthropologically, and linguistically and thus provide an excellent framework for the study of rare-variant spread.