140 BRITISH LEPIDOPTERA. 



We do not know that anyone has attempted to really challenge 

 the view propounded by Weismann, but Chapman points out 

 (in lift.) that the larval evolution as here set forth exhibits some 

 hiatus and possibly some mistakes. After noticing that Weis- 

 mann's idea (that the primitive Eumorphid larva had a pale 

 subdorsal line, much as have the larvae of Hemaris and Sesia, and 

 which is still essentially present in Darapsd) may be accepted, he 

 goes on to urge that it was somewhere below Darapsa that the 

 Eumorphid stirps divided into two sections : 



Sect. i. Philampelina.- — This branch seems not to have developed eye-spots 

 on the abdominal segments in the subdorsal line of larva, but to have retained the 

 line, and often to have developed oblique stripes in the reverse direction to those 

 in Sphinginae — Philampelus, Dupo, Pachylia, Dilophonotus. Acosmeryx, &c. 



Sect. 2. Eumorphina. — This branch did develop larval eye-spots in the 

 subdorsal line on the abdominal segments of the larva. Of this group Darapsa 

 [myron) may be taken as a basal form — Eumorpha, Theretra, Hippotion, &c. 



Sect, i includes all the non-Eumorphine Eumorphids, and no 

 doubt contains several tribes of equal value with Eumorphina, yet 

 this broad subdivision thoroughly agrees with the pupal indications. 

 Darapsa is not only about as low pupally as Hemaris or Pterogon, 

 but is almost certainly lower than the best known Eumorphids and 

 Philampelids. Starting from this point (Darapsa), which is, from the 

 standpoint of larval morphology, fairly common to the lower mem- 

 bers of all the Sphingid subfamilies,* the Eumorphids began to 

 develop eye-spots along the subdorsal pale line. Weismann's 

 account of how the simple pale line gradually develops into an eye- 

 spot leaves nothing to be desired. From this point there are, as 

 he points out, several attractive hypotheses possible, though he does 

 not refer to the alternative ones till he has well enlisted our 

 sympathies in favour of the one he adopts, viz., that two branches 

 occurred here, in one of which the eye-spots originated on the anterior 

 segments, and were here carefully elaborated for terrifying purposes 

 on the ist and 2nd abdominal segments (elpenor, porcellus), whilst, 

 in the other, the eye-spot originated on the 8th abdominal segment 

 (for what object he leaves unelucidated), but, in each instance, as a 

 final stage, passing to the other abdominal segments, yet arising 

 in totally different ways. A difficulty is here obvious, although 

 it has to be conceded that the two divisions he makes correspond 

 with the two natural tribes — Eumorphidi (with retractile front seg- 

 ments, and cryptic coloration, terrifying markings and attitudes) 

 and Phryxidi (with much less retractile front segments, a conspicuous 

 warning coloration, and no terrifying attitudes). The difficulty is 

 in explaining why such remarkably similar markings, in such closely- 

 allied tribes, should separately originate for such very different 

 objects. Such markings, originating in a subdorsal line, are else- 

 where certainly very rare, if not unknown, and one is almost com- 

 pelled to believe that, with such different objects in view, the 

 necessary markings would have resulted in totally different forms, 

 because they would have originated in totally different ways. The 

 difficulty can only be got over by supposing that the tendency 

 to develop eye-spots was well expressed before the two branches 



* The still lower stage of "no markings" suggested by Weismann is quite 

 outside (i.e., below) the Sphingides (Chapman). 



