BLASTOGENIC VARIATIONS. 121 



never throw intermediate characters, the result of 

 Heape's experiment is without significance, as far as it 

 bears on the inheritance of acquired characters. Heape 

 considers that Romanes is mistaken in this view, for he 

 has himself obtained experimental evidence to show 

 that some of the young obtained by crossing are of an 

 intermediate character. It is nevertheless true that in 

 the majority of cases the young are apparently pure 

 bred of one type or the other, and hence the value which 

 ought to attach to Heape's experiments, so far as they 

 relate to the production of somatic variations by change 

 of environmental conditions during embryonic develop- 

 ment, is probably not very great. 



The evidence so far available seems to render it 

 highly probable, therefore, that the major part of the 

 variation exhibited by organisms is of blastogenic 

 rather than somatic origin. It is due more to dif- 

 ferences in the germ cells than to external influences 

 acting during ontogeny. If it be found possible to col- 

 lect considerable series of anthropometric measure- 

 ments of identical twins, and to compare them, as re- 

 gards variability, with similar measurements on dissimi- 

 lar twins, then we may hope to obtain some adequate 

 conception as to what proportion of the variation ex- 

 hibited by adult individuals is due to external influences 

 acting during pre- and post-natal existence. Also by 

 comparing the variability of dissimilar twins with that 

 of members of families produced in the normal manner 

 of one at a birth, we may hope to determine what 

 changes, if any, are produced by the slight differences 

 in the maternal fluids which must doubtless exist dur- 

 ing the development of the different offspring. 



