6 NEUROPTERA. 



have never tried to rear these ; probably it would be rather 

 difficult, since their food in sufficient quantities ought to swim 

 past them, whilst the free-moving larvae seek their own food. 



In collections Ephemeridce can only be imperfectly pre- 

 served ; when pinned they lose almost entirely their colour 

 and markings, and their delicate abdominal skins change 

 their form very essentially. Neither do they keep well in 

 spirits of wine, their colours then going entirely. I have, 

 however, observed, that on taking great numbers, w^iich in 

 their swarms is often possible, individual specimens will 

 always partially preserve their forms, so that, in this way, 

 one may obtain by degrees approximately good specimens. 

 The alteration in the JEphemeridce after death is naturally 

 one of the main causes of the low state of our knowledge of 

 species. I am of opinion that this evil can only be remedied 

 in one way, namely, by the careful description of living 

 specimens, and by noticing at the same time how the same 

 specimens change after death. But since the males and 

 females are generally very different, and also the imago and 

 sub-imago differ essentially, this renders necessary for each 

 species four, and if we add thereto the dead insects, eight 

 separate descriptions ; a work the completion of which will 

 vet require considerable time, since at present we hardly 

 know ail the four-winged states of a dozen species, and only 

 possess descriptions of the living insects of single species. 

 Hence here there is rich unexplored field open to inves- 

 tigation ! 



One will seldom be in doubt as to the genus to which any 

 species should be referred, but to distinguish the species is, 

 on the other hand, a work of considerable difficulty, and the 

 differences which I have mentioned are those which are 

 apparent on the examination of dried cabinet specimens. In 

 the first place, the wings must be examined, the form and 



