Abstract
DNA fingerprint similarity is now being used widely to make inferences about the genetic structure of natural and domesticated populations, often with little regard to the limitations of such data. This paper provides an overview of the statistical theory of DNA fingerprint analysis with special focus on applications to natural populations for which little if anything is known about the detailed genetics of the DNA profiles. Approaches to estimating individual and population homozygosity, effective population size, population subdivision, and relatedness are reviewed, and issues concerning the biases and sampling properties of the statistics are discussed.