The Origins of Inversion Polymorphisms

Abstract
The expected distribution of inversion lengths in natural populations differs among different hypotheses as to the mode of establishment of inversion polymorphisms. A theory of the adaptive significance of inversion lengths is presented and applied to several families of Diptera. The observed distributions differ strikingly among families. The only hypothesis found generally applicable is that several genes are required on the average for an inversion (or appreciably heterotic loci are rare but required) and that double crossing over subsequently removes many of the longer inversions. Two additional predictions of this hypothesis cannot yet be clearly tested from available data. Position effect does not seem to be important in the establishment or maintenance of inversions.