VARIATION AND HEREDITY IX TWINS 179 



figures conceals the fundamental facts. Studies in the 

 heredity of armor units reveal not infrequently a 

 situation in which, so far as individual bands are con- 

 cerned, there is a close approach to identity between 

 mother and offspring; in other bands, however, there 

 may be a pronounced lack of heredity. In some cases, 

 moreover, one offspring in a set is almost identical, 

 band for band, with the mother, while another offspring 

 of the same set shows marked differences from the 

 mother. Once more then we shall have to call upon 

 the mechanism of somatic segregation, which is respon- 

 sible for the segregation of biparental units, so that one 

 individual of a monozygotic set shows the maternal 

 scute count and another shows a very different scute 

 count, presumably like that of the father. 



Hence, although a coefiicient of correlation derived 

 by averaging the conditions in a large number of mono- 

 zygotic sets of offspring may be a valuable measure of 

 the average performance of the species, it has no value 

 when appHed to individual cases. One must conclude, 

 therefore, that no definite law is to be posited as to the 

 relative potency of ''nature and nurture" or of the pre- 

 determinative versus the epigenetic factors of develop- 

 ment. Every character evidently has a genetic basis in 

 the zygote, but the exact expression of the character 

 is dependent upon developmental or epigenetic factors 

 that vary in each individual case. There appears no 

 longer to be any point to an attempt to determine the 

 relative potency of predetemiinative and epigenetic 

 factors in development. 



