Differences in recombination between sexes in the mouse occur frequently (see, for example, Castle, 1925;Parsons, 1958;Bodmer, 1961). Significant effects of maternal age on recombination are, however, less common. It is the purpose of this paper to present evidence for variability of recombination with age which differs between sexes.In the fifth linkage group, the order of the genes being agouti-undulatedpallid-fidget, Fisher (1949) found that the recombination fraction between agouti and undulated decreased with age in both sexes, with no significant difference between sexes. Wallace (1957), however, found a slight increase in recombination with age of heterozygous males in the segment agoutifidget which includes the agouti-undulated segment studied by Fisher, with no significant effect in females. Bodmer (1961) found that the recombination fraction between pallid and fidget decreased with age of female heterozygotes ; the data from male heterozygotes being perhaps inadequate to show any trend.With these observations in mind, some data for linkage group XIII previously analysed by Parsons (1958) were re-examined for age effects in the segment leaden (ln)-fuzzy (fz). The other mutants used by Parsons, polydactyly and Splotch, were omitted from this analysis as they are both liable to be misclassified. In particular the manifestation of polydactyly varies with the age of the maternal parent (Holt, 7948).The data for ln-fz are presented in table i. The recombination fraction appears to fall with age of female heterozygous parents, but with male parents, if anything, it increases. To test these trends further it is necessary to compute weighted regressions of the recombination fractions on time. The recombination fractions must be transformed using the angular transformation a = sin\/p where p is the percentage of recombinants, and a the transformed value of p (see table r), then a will have a variance of I i8o 2 -(_-) where n is the number of offspring counted to give p. Thevariance of a is therefore independent of p. This procedure was used by Fisher (7949) in his analysis of recombination between agouti and undulated.From the female parents there is a decrease in the regression of a on t of o57° per two months which is not significant, but from the males there is an increase of o87° per two months sub-significant at the io per cent. level. The difference between the two regressions was significant at the 24 per cent, level and almost at the i per cent, level. The mean recombination values also differed as expected from Parsons' (7958) results such that 107