iSo NATURAL HISTORY OF AQUATIC INSECTS CH. 



against, and I find that the larva is seldom or never 

 swept away, even when its haunts are invaded by a 

 groping naturalist. Its swimming power, I should 

 explain, is inconsiderable, altogether insufficient for 

 propulsion against a rapid stream. If seriously 

 alarmed, the larva lets go, and immediately dis- 

 appears from sight. But by watching the place 

 attentively, we shall before long see the larva work- 

 ing its way back, and in a 

 minute or two it will be found 

 attached to the very same leaf 

 from which it started, or to 

 some other leaf, equally con- 

 venient, which it happens to 

 fall in with. I found the diffi- 

 culties of observation in fast- 

 flowing water crowded with 

 leaves very great, until at last 



XfO 



it occurred to me to push a 



FIG. 62. Head of larva of Simu- 



Hum, showing on left side the w hite plate in amonsf the 



position and form of the new fc> 



larval antenna and fringed ap- Ipovpc Then trie rlarl- 



pendage, which will become ICdVCb. ddl K- 



external at the next moult ; on coloured larvp^ herame r>er 

 right side the new imaginal larVSE 



antenna, which will become fart]\7 evident on the \vhite 

 external after pupation. ICC11J 



ground, and I was able to see 



exactly how they managed. When disturbed by the 

 plate, some of them let go, and drift a few inches away. 

 They are not very easily frightened, and most of them 

 remain holding on by their sucker. Those which quit 

 the leaf remain stationary in the torrent or nearly 

 so, and on close observation a thread, or perhaps a 

 number of threads, become visible on the white 

 ground. These threads are in general stuck all over 



