130 Dr. T. A. Chapman's notes on Pupa", 



as sliowing that the Limacodids must take even a lower 

 place than I had given them. 



The other error was as to the structure of the pupa of 

 the Nolidx, due to my having only dead or empty pupo3. 

 An examination of living pupa3 shows that this pupa 

 possesses only one free segment, viz., the fifth (abdominal), 

 and is, therefore, probably in no way directly related to the 

 Macro-heterocera. I hope to deal with this more fully in 

 some future communication. 



There is a pupal piece I have not hitherto referred to, 

 having no clear idea as to its relationships, and have 

 passed over as a portion of the anterior head cover (or 

 face-piece), which no doubt it is. Nor have I studied it 

 enough to know its range, except that it occurs chiefly in 

 low micro forms. It is often a separate piece between 

 the face-piece proper and the dorsal head-piece ; it does 

 not divide in the middle line on dehiscence, but remains 

 in one piece. It is probably the dorsal-plate belonging 

 to the antennal section of the head, as distinguished 

 from the face-piece proper, which is the anterior plate 

 or plates belonging to the mouth parts. Though it has 

 no bearing on the subject of this paper, I take this 

 opportunity of calling attention to it. 



In my paper, read Feb. 22nd, 1893, I made it suflS- 

 ciently clear that the section Incomplete included the 

 more generalised and ancient forms, and that the Ohtectx 

 were more specialised and more modern. By putting 

 this, however, into a tabular form, I believe I conveyed the 

 impression that there was some sort of gulf between 

 them, and what is especially erroneous, though much the 

 same thing in another form, that the Ohtedx. were more 

 closely related to each other than any of them were to 

 any of the Inconipletse^ and this misapprehension was 

 probably favoured by my taking the number of free 

 abdominal segments as the most typical indication of the 

 extent to which a pupa belonged to the more generalised 

 foi'ms. No doubt that is so, but the other directions in 

 which fusion of parts (the character of Ohtectie) has 

 failed to obtain, is also of importance. 



The feature of an Incomplete pupa that is usually most 

 persistent — that is, appears most frequently in the highest 

 forms — is the presence of the portion, that I at one 

 time supposed might be an anterior portion of the pro- 

 thorax, but which I think there can be no doubt is the 



