November— December 1S90.] 



PSYCHE. 



415 



spots are arranged in the same manner ; 

 In fact, there is no absolutely constant 

 ditlerencc, eitlier in structure, markings 

 or coloration, between the two so far as 

 I ha\e been able to observe, and I have 

 no doubt that specimens may occasional- 

 ly occur which it vrould not be possible 

 to refer positively to one species or the 

 other bv an examination of the imago 

 alone. It would at first seem that these 

 species must be more closely related than 

 others in the genus, perhaps hardly 

 yet distinct ; but when we consider the 

 larval stages, it is at once apparent that 

 they are widely separated, more widely, 

 in fact, than many other species of Data- 

 na which are readilv separable in the 

 imago state. This difference would seem 

 to indicate rather a long inherited simi- 

 larity between the moths than verv reoent 

 derivation from a common ancestor. 

 While the moths, it may be, have re- 

 mained constant to their former type, 

 or undergone a similar development, the 

 larvae have diverged greatly, which 

 would seem to correspond to the gen- 

 eral rule in lepidoptera, that the larvae 

 are often the first to vary. The eggs oi D. 

 major and D. drexelii differ markedly : 

 the larval differences appear in the first 

 stage and are kept up and intensified 

 throughout the larval period, and only 

 when the pupa stage is reached, does 

 the close resemblance observed in the 

 moths begin. The cause of the diver- 

 gence in the lan'ae,especially at maturity, 

 is not very obvious, particularly in the 

 peculiar ornamentation of D. major^ 

 which differs from that of all of the 

 other Datana larvae so far known. 

 D. drexelii is abnormal only in the 



predominance of vellow about the pro- 

 thoracic and the Sth, 9th and loth ab- 

 dominal segments,* and this might serve . 

 as an excellent protection, in that the 

 bunches of yellow striped larvae with 

 their extremities elevated in their cus- 

 tomary position, resemble the little 

 chisters of flowers of the Witch-Hazel 

 with their linear yellow petals, which 

 begin to blossom in September, just as 

 the larvae become conspicuous. What 

 protection, if any, their coloration affords 

 on the Deerberry, their other food-plant, 

 I am unable to say, as they occur almost 

 exclusively on the Witch-Hazel in this 

 vicinity. (Rhinebeck, N. Y.) 



But as to D. fjtajor its curious spots 

 do not resemble anything so far as I can 

 see. It is to be noted, however, that 

 the habits of the two species differ, for 

 while D. drexelii are always gregarious, 

 D. major, when they do not separate 

 entirely, feed farther apart, and tend to 

 scatter more in the last stage than al- 

 most any other species, and the change 

 in ornamentation from lines to spots, 

 which they undergo in the last molt, 

 may be of use to them with their 

 changed habits. 



Datana major,. G. S Ji. 

 Egg. Sub-pyriform, of less diameter 

 and cylindrical for a short distance at the 

 top, recalling the structure of Z^./a/w//; 

 flattened at base or a little hoMowed ; 

 evenly flat on top with a sharp angle 

 between top and sides. Color uni- 

 form sublustrous white, a rather 



* In this article, I have followed the nomenclature 

 used by Dr. Packard for the larval segments, as it is 

 not possible to describe intelligently the markings nf 

 Datana by the usual method. 



