TR1CHOP TERA . 1 89 



it lives. Very little respect for the architectural skill of 

 these builders is commanded by their rude dwellings. But 

 if one looks a little farther, something will be found that is 

 sure to excite admiration. The dweller within this rude re- 

 treat is a fisherman ; and stretched between two stones near 

 by can be seen his net. This is made of silk. It is usually 

 funnel-shaped, opening up-stream ; and in the centre of it 

 there is a portion composed of threads of silk extending in 

 two directions at right angles to each other, so as to form 

 meshes of surprising regularity. It is as if a spider had 

 stretched a small web in the water where the current is the 

 swiftest. These nets occur in rapids between stones, but in 

 many places they are to be found in greater numbers along 

 the brinks of falls. Here they are built upon the surface of 

 the rock, in the form of semi-elliptical cups, which are kept 

 distended by the current. Much of the coating of dirt with 

 which these rocks are clothed in summer is due to its being 

 caught in these nets. We have not yet observed the owners 

 of the nets taking their prey from them ; but we cannot 

 doubt that they are made to trap small insects or other ani- 

 mals that are being carried down-stream ; for the larvae of 



^> 



the sub-family to which these net-builders belong, the Hy- 

 dropsy china? , are known to be carnivorous. It should be 

 noted here, however, that the greater number of Caddice- 

 worms are herbivorous. 



When a Caddice-worm gets ready to change to a pupa it 

 retires into its house and builds a door to keep intruders 

 out ; but the door always has an opening to allow the water 

 to flow in so that the pupa can breathe. Sometimes a simple 

 Sfratincr of silk is made over the entrance. 



o & 



On one occasion the writer had the good fortune to ob- 

 serve a Caddice-fly leave the water and take its first flight. 

 The specimen was one of the net-building species, Hydrop- 

 sycJie (Hy-drop-sy'che), which I was breeding in an aquarium 

 in my laboratory. It swam to the surface of the water 

 repeatedly, using its long middle legs. When swimming, 



