208 KELLOGG AND BELL 



variations) are not directly heritable, the only v'ariations which 

 give Natural Selection its validity as a species-forming factor 

 are the congenital ones. If the relation of variation to species- 

 forming is to be studied, the congenital variations, clearly apart 

 from those adaptively acquired during the development of the 

 individual in response to the immediate influence of the condi- 

 tions of life, are the only ones that neo-Darwinians can consist- 

 ently take into account. The great majorit}'^ of published 

 variation studies have apparently, however, left this distinction 

 between blastogenic and acquired variations wholly out of ac- 

 count, with results that seem to us to invalidate seriously the 

 application of the generalizations arrived at to the problem of 

 species-forming, granting for the argument the truth of what is 

 believed to be true by apparently the majority of working natur- 

 alists, namely, the non-heritability of acquired characters. 



As this matter of the importance of distinguishing between 

 congenital and acquired variations is referred to again in some 

 detail at the beginning of the next section of this paper, its dis- 

 cussion may be left now for the presentation of a brief account 

 of those conditions in insect-development which we believe af- 

 ford a criterion for makincr the distinction between the two sorts 

 of characters in the case of the variations presented by the full}' 

 developed imagines (adults) of insects with complete metamor- 

 phosis. 



Character of Development of Specialized Insects Affording 

 Means of Distinguishing Between Blastogenic and Acquired 

 Variation. — "Without by any means exhausting the subject of 

 the post-embryonic development of insects, entomologists have 

 become sufFiciently well acquainted with the phenomena attend- 

 ing this development to be able to confirm absolutely (in essen- 

 tial character) Weismann's discoveries in the larva of the " im- 

 aginal discs" as the independent embr3'onic centers from which 

 develop the wings, legs, antenna) and some other parts of the 

 winged imagines (adults) of insects with complete metamor- 

 phosis. That is to say, in all the insects which hatch from the 

 egg in a larval condition markedly different from the definitive 

 condition of the species in its fully developed, mature stage, 

 many of the adult organs, as, conspicuously, the external parts 



