320 KELLOGG AND BELL 



never exposed in definitive unchangeable condition to outside 

 influences) are blastogenic variations, as are also the structural 

 variations in the character of the venation and the number of 

 costal hooks in A^is and in the black ant that was studied but 

 the variations in the pattern of the pronotum of Tettigonia^ 

 CorisUy the capsid and flower bug, and the number of tibial 

 spines in Cicada and Mclanopltis^ may be in part acquired, for 

 in these latter cases the insects are exposed during their imma- 

 ture life (development), with these color and structural charac- 

 ters in formative condition, and to some extent in use, to the 

 continuous influence of their environment. 



Given a criterion (either in the character of the variation, or 

 the variable character itself, or in the life history of the animal 

 showing the variations) that will enable one to distinguish be- 

 tween strictly blastogenic variations and those which may be 

 wholly or in part acquired — and we believe we have such an 

 one and a particularly valid one, in the case of the insects with 

 complete metamorphosis — and this criterion may be applied 

 not only to an individual but to an hereditary series of individ- 

 uals and the larger question — that Sh'cilfrage of modern biol- 

 ogy — as to the transmission of acquired variations be approached 

 through breeding. By comparison of series of generations 

 whose immature life has been exposed to various (experimen- 

 tally controlled and quantitatively determined) conditions of 

 life, with series of generations whose immature life is exposed 

 to only a single rigidly controlled set of life conditions may be 

 determined not only the effect of varying conditions on the pro- 

 duction of variations, but the heritability of these variations. 

 The most satisfactory answer to the question of the hereditary 

 transmission of acquired characters will come as the result of 

 a quantitative (statistical) study of variations known to be blas- 

 togenic compared with a similar study of variations known to 

 be acquired, both studies to be made on complete series of in- 

 dividuals bred under quantitatively determined life conditions. 

 Such studies are certainly not impossible, with a criterion for 

 the distinction between blastogenic and acquired variation once 

 obtained. The character of this criterion, for insects, exists by 

 reason of the effective difference in the course of the life his- 



