KETTLEWELL1 gives the frequencies of the three forms carbonaria, insularia and typical of the moth Biston betularia, for many localities in the British Isles. In this paper he points out that the form insularia is extremely variable and has a distribution which would repay more intensive study. He also states2 that the two melanics carbonaria and insularia are controlled by genes at separate loci with no evidence of linkage. He bases his analysis1 of insularia frequency changes during the evolution of industrial melanism on this assumption. However, the extreme variability of the form indicates that it may be controlled in more than one way and some recent breeding results3 suggest that on occasion the form may be controlled by a third allelomorph at the carbonaria locus.