30 



THE AMERICAN NATURALIST [Vol. LII 



ferent from each other and from the Massachusetts race 

 (probably imported from France). All of them are dif- 

 ferent from the Japanese races and these again differ 

 in the different parts of Japan. The characters of dif- 

 ference are manifold; we shall confine ourselves here to 

 a single character, more interesting and more character- 

 istic than the others— the markings of the caterpillars. 

 Fig. 1 shows caterpillars of a few races after the first 

 moult. We see here some of the transitional stages from 

 a very light to an almost black caterpillar. The genetic 

 study of this character of marking shows that we are 

 dealing here with a primary type of marking which be- 

 longs to the entire group of moths in a similar form, that 

 is, the light pattern. All the darker forms have the same 

 genetic basis of marking, on which, however, dark pig- 

 ment encroaches increasingly until the markings prac- 

 tically disappear. "We may now divide this increasing 

 melanism into ten classes and place the lightest individ- 

 uals in Class X and the ones without marking in Class I. 

 It must be added that the dark series extends beyond Class 

 I, but the difficulty of classifying them is such that no 

 darker classes have been adopted. 



The young caterpillars of the different races show 

 markings which fluctuate around a mean at a certain point 

 of the series and this behavior is remarkably constant 

 for the different races. The following Table I gives a 

 few polygons for different European and Japanese races. 



TABLE I 



