1865.] 



203 



just as reasonably first lay it down as a law, that all imagos that have 

 abdomens of the same size, shape and color are identical, and then tor- 

 ture Nature to fit the Procrustean bed, which we have predetermined in 

 our own minds that she shall lie upon, whether or nay. 



There is a partial parallel to the case of these two Halesidnta in the 

 "alternate generations" of certain Radiata. "It is curious," we are 

 told, "that while very dissimilar Jelly-fishes may arise from almost 

 identical Hydroids, we have the reverse of the proposition, in the fact 

 that Hydroids of an entirely distinct character may produce similar 

 Jelly-fishes." {Seaside Studies by E. C. and A. Agassiz, p. 43 and 

 see p. 75.) Here two given Hnes either diverge after converging, or 

 converge after diverging. But in Halesidota the lines first converge, 

 then diverge, and then converge again. For it has been shown, that 

 the very young larvae of tessellaris and HarrisH are very nearly or quite 

 undistingui-shable, that the mature larvae differ as widely as any two 

 species of the same genus can well do, and that finally the imagos be- 

 come absolutely undistiuguishable. 



On the supposition that tessellaris and Harrisii sprang from the same 

 pre-existing species, and consequently that they were primordially un- 

 distiuguishable in the larva state, as they still are in the imago ; we 

 may account for their larval differences by assuming, that the colora- 

 tioaal peculiarities of the two larvae afford them a partial protection 

 against birds and against ichneumon-flies and other predaceous insects, 

 and were gradually assumed on the Darwinian theory of Natural Selec- 

 tion. We know how many lepidopterous larvae are partially protected 

 from birds kc. by simulating twigs or the bark of the tree on which 

 they live ; and it is not at all impossible that the orange pencils kc. of 

 Harrisii and the black pencils of tessellaris may be mistaken by birds 

 and insects for a process of the particular plants on which they feed. 

 If we reject this assumption, we can only attribute the differences of 

 the two larvcTe to what Darwin calls " Divergence of Character," super- 

 induced by breeding "in-and-in" for ages; in the same manner as ge- 

 ographical varieties come to differ in process of time from one another 

 and from the original type. 



Sphingicampa distigma Walsh and Duyocampa bicolor Harris. 



(Lepidoptera.) 



Having been fortunate enough to meet this year with three larvae 



exactly similar to that which I have described as D. bicolor, {Proc. &c. 



III. p. 425,) I had hoped to solve definitively the question of what imago 



they would produce. Being confined, however, in a cage with milliuet 



