DISCONTINUOUS VARIATION. 



39 



ing in close communion with each other under the same 

 stones. 



In this instance there could be no doubt as to the di- 

 morphism, just as in many of the instances previously 

 quoted there could be practically no doubt as to the 

 monomorphism, but obviously there must be inter- 

 mediate stages in which the fusion is much closer. In 

 these an irregular asymmetrical curve with only one 

 summit may be obtained, instead of a distinct double 



FIG. 8. I. High male, II. Low male, of Common Earwig. 



humped curve. Weldon * obtained such a curve for 

 the distribution of the frontal breadths of Naples speci- 

 mens of the crab, Carcinus mcenas. This is repro- 

 duced in Fig. 9, the horizontal scale representing thou- 

 sandths of the carapace length, and the vertical scale 

 numbers of individuals. It seemed to Professor Wel- 

 don very probable that this asymmetrical curve was 

 produced by the fusion of two races of individuals, clus- 

 tered symmetrically about separate mean magnitudes, 

 and Professor Pearson tested this supposition for him. 

 Pearson calculated that by mixing 41.45 per cent, of 

 individuals with a mean frontal breadth of 630.62 thou- 



*Proc. Roy. Soc., liv. p. 324, 1893. 



