ASSOCIATION between characters can be ephemeral, as that caused by linkage; or more persistent, as that caused by the pleiotropic effects of one or more genes. In the event of an association due to pleiotropy, selection for one character should produce a correlated response in the other of greater regularity and continuity than if the association was the result of linkage. Mice of both the brown and dilute phenotypes have been observed to be larger in body size than non-mutants (FELDMAN 1935; GREEN 1935; CASTLE et al. 1936; CASTLE 1941). Several studies have attempted to determine whether these observations are the result of genes influencing body size linked to the coat color loci or the pleiotropic effect of the mutants on body size. GREEN ( 1935) compared the body size of brown (bb) and heterozygous black (Bb) mice in various back-crosses involving different homozygous parent lines. He found that the body size difference between color genotypes varied from one backcross to another. In interpreting these observations, he hypothesized that this was caused by linkage between the brown locus and genes affecting body size. CASTLE et al. (1936) made a number of body size measurements on both brown and dilute (dd) mice and their black and non-dilute (D-) sibs in backcross and F, matings. The larger size which they found associated with mice having the recessive phenotypes they attributed to the physiological effect of the brown and dilute alleles. MACARTHUR (1949) reported in a two-way selection experiment for body weight that his high line was mostly bb dd and his low line mostly BB DD. He felt this corroborated CASTLE'S data and demonstrated that the brown and dilute alleles have a pleiotropic effect for larger body size. BUTLER (1954) crossed bb dd mice from MACARTHUR'S high line with two different inbred lines homozygous for the non-dilute (D) allele and compared body size in F, and backcross mice classified according to dilute or non-dilute coat color. He found that non-dilute mice were slightly larger in both the F, and backcross tests. BUTLER also examined an inbred line which was segregating for the dilute locus. After maintaining this line for seventeen generations and selecting only for general vigor, he again found that non-dilute mice were slightly larger than dilute mice. BUTLER noted that his results, while statistically nonsignificant were similar in magnitude to those reported by CASTLE but opposite ' SLientifiL Iournal .biicle No 0318. blmnesota 4gnculture Expninient StatLon lhe woik W~S suppurled bg the ~dtl~Jrld~ Science Foundation, Giants G 8878 and GI3 12%