10 



Cincinnati Society of Natural History. 



ns they would to a Lithocolletis in escaping from its mine through the 

 delicate cuticle of a leaf, or to Bedell ia which does not have to escape 

 from anything except its pupa skin. In fact, while some pupne, as, 

 e. </., that of Prodoxus, have the armature of the head, but not 4 

 the lateral armature of the anal segment, and yet are hot known 

 to have tropin of the first form at any time, yet the fact that 

 these structures of the pupa are better developed in those genera 

 which retain for the longest time the tropin of the first form, suggests 

 that there mav be some connection or correllation of growth between 

 the two kinds of structure. That the course of evolution is influenced 

 by variations which have taken place in earlier stages, is stated by Mr. 

 Darwin in the "Origin of Species,'' chap. 5: "The whole organization 

 is so tied together during its growth and development, that when 

 slight variations in an}' one part occur, and are accumulated through 

 natural selection, other parts become modified. This is a very impor- 

 tant subject most imperfectly understood. The most obvious case is 

 that modifications accumulated solely for the good of the young or 

 larvae will it may safely be concluded affect the structure of the adult/' 

 But, however, if at all, the first form of tropin is related to the partic- 

 ular structures of the pupa which I have mentioned, there can be no 

 doubt that the genera above named have the first form of tropin in the 

 earlier stages of their existence, and afterward change it for the sec- 

 ond or ordinary form, without any change whatever in their external 

 surroundings, and it is difficult to see how the change can have been 

 produced by the effect of an}- external influence such as some natur- 

 alists of the mechanical school suppose to be sufficient to account for 

 the initiation of the variations which have resulted in the present 

 condition of the organic world. Many of these larvae (as all Lithocol- 

 letis larvae of the cylindrical group, and some species of Gracillaria. and 

 Ornix) pass the last four stages of their larval existence and the pupa 

 state under precisely the same conditions which characterized the pre- 

 vious stages — living in the same mine, eating the same food, subject to 

 precisely the same influences, and the only change is in the way in 

 which they bite the food; a consequence, not a cause of the changed 

 structure. It is one of the numerous cases of structure in advance of 

 function. Others, as Lithocolletis ornatella, Leucanthiza amphicarpeoz- 

 foliella, cease to feed after the change in the form of the trophi, and 

 pass the remaining two stages of their larval existence still in the mine 

 without eating, although their trophi are more perfect and better 

 developed than they were in their feeding stages; others again, as most 



