164 COLORATION IN I^EIPTINOTARSA. 



having periods of interrupted growth, stages and characters found at the 

 resumption of growth which they beHeve to be the atavistic reappearances of 

 ancestral adult characters. More recently, Shull, investigating these stages 

 in Sium cicutcefolium, comes to the conclusion that these stages represent, not 

 ancestral adult characters, but simpler conditions of structure which are 

 explainable upon physiological instead of phylogenetic grounds. In insects 

 these stages have not, as far as I am aware, been investigated. 



In the ontogeny of both the larval and imaginal coloration in Leptinotarsa 

 there are clearly distinguishable the following stages : (a) The preparatory 

 stage, the time when the color is the uniform yellow of the pupa, and the 

 period of active zymogen formation ; (b) the initial stage, during which the 

 color appears over the centers of coloration and the color pattern is marked 

 out; and (c) color intensification stage, when the colors develop to their full 

 intensity. These stages are clearly defined and recognized in all insects 

 whose color development has been studied. Do these three stages, which are 

 common to all insects in their development, represent three great steps in the 

 evolution of the coloration of the class ? Are we to believe that the prepara- 

 tory stage is the atavistic repetition of the adult coloration of the hypothetical 

 ancestor of the tracheate phylum? that the initial or second stage represents 

 the first developments of coloration, and the third the condensed recapitula- 

 tion of more modern specialization and differentiation in insect colors ? Such 

 a view is perfectly in accord with those of Cope, Hyatt, and others. The 

 interpretation of these stages, widespread though they are in insects, does not 

 rest with studies in phylogeny; it is a question solely of the investigation of 

 adaptation to ontogenetic ends ; that is, they are physiological and develop- 

 mental, and not phylogenetic phenomena. 



Almost all tracheates are subject to periodic changes of the chitinous por- 

 tion of their body wall, and at these ecdyses the color pattern, as far as the 

 markings arising from the dark cuticula pigment are concerned, is removed 

 and must be replaced by a new development thereof. The removal of the 

 chitinous layer reduces the animal to a condition which, as far as the integu- 

 ment is concerned, is practically embryonic. This is a fact which any one 

 who wishes may verify. From this condition the animal must retrieve itself 

 as soon as possible by the development of a new cuticula carrying new color 

 markings. At each recurring ecdysis this process is repeated, and the three 

 main stages are passed through. We admit that ecdysis is necessary to 

 growth in these forms, and that growth is in a sense periodically interrupted 

 by ecdysis. It is probably true that in the evolution of insect color there were 

 three stages corresponding roughly with the stages in development already 

 described; but that these should be repeated so exactly at each ecdysis is a 

 phenomenon the reasons for which it is hard to imagine. On the other hand, 

 its interpretation on the basis of an ontogenetic adaptation alone would involve 



