NATURAL SCIENCES OF PHILADELPHIA. 49 



as suitable protections against inclement weather during this 

 critical period of radical changes. 



If the views embodied in the last paragraph have any semblance 

 of reality, we should expect to meet with, in the habits of existing 

 species, confirmatory proofs. Among the Hydrocampidse, Para- 

 ponyx closely resembles in larval form and habits the Phryganese. 

 Its larva is possessed of large branchiae, besides spiracles, and its 

 pupa is found in a cocoon amongst leaves under water. There is 

 another species quite common in France, which subsists upon a 

 species of pond weed. In the caterpillar state it cuts two pieces 

 of leaf and fashions them so as to become nearly oval in shape 

 and equal in size. These are joined by their margins by means of 

 a little silk, the larva taking special care to leave an opening for 

 the head and the first segments of the body. It drags this house 

 under water, occasionally destroying it for a more substantial 

 domicile. When it is ready to assume the chrysalis state, it at- 

 taches its leafy house securely to plants or stones in the immediate 

 vicinity. The China Mark's caterpillar lives underneath the 

 leaves of a species of Lemma in the water, and protects itself in 

 a cylindrical case of silk covered with leaves, in which it subse- 

 quently becomes metamorphosed into a chrysalis. Here, it is evi- 

 dent, is the starting point for the development of the higher moths. 

 It is not to be presumed that the Hydrocampidae are the immedi- 

 ate descendants of the Phryganese. There may be other forms 

 somewhat lower, of whose existence we have no knowledge. In 

 the accompanying tree I have preferred to represent these forms 

 under the name of ideal Paraponyx, and to consider them as the 

 immediate progenitors of those whose life-history I have partially 

 described. 



The pupa of Eudryas unio, Riley, has been found in winter in 

 the stems of a species of Hibiscus, as though the larva had been 

 feeding in that location. Psychomorpha epimenis, Clem., fre- 

 quently perforates a piece of old wood and changes into a chrysalis 

 therein. Its habit of boring into some substances to prepare for 

 the change is inveterate, and it always neatly covers up the ori- 

 fice so that it is difficult to detect. Other instances might be cited. 

 The species just cited belong to the Zygsenidse from which it has 

 been assumed the JEgeridee were evolved. There is a manifest 

 resemblance to the latter, with the important difference that the 

 larvae of the JEgeridse pass through their early stages in the steins 



