The present work is an attempt to provide a systematic treatment of genetic linkage in diploid heredity. Part A presents a general account of statistical methods which can be brought to bear on the problem. The primary emphasis is on the practical aspects of estimation. A large proportion, if not the majority, of mutant genes fail to match up to 'textbook' genes-with faultless segregation ratios and expression-yet, these are the materials with which the practical researcher has to cope. For this reason, it is important to know how to deal with the assortment of genes which may display significant deviations from expectation. Part B examines the accumulated data on linkage for most of the laboratory mammals and provides a comprehensive and up-to-date survey. The need for a critical review has often been expressed and it is hoped that the present analysis will fill the gap. The volume of material is probably the most important in the animal kingdom other than that for Drosophila species.