June-Sept., 1919.] MuRPHY : EGG-LAYING OF BRACHYCENTRUS. 157 



screen fastened over the rear end of the tube prevented the larvae 

 from escaping and did not hinder the attachment of rubber tubing 

 for an outlet. At the other end was fitted another piece of tubing 

 which connected with a small glass funnel. This funnel was fastened 

 just under the faucet. The desired amount of water could be easily 

 regulated at any time, and the funnel furnished a means of intro- 

 ducing food. 



Larvae introduced into the stream passing through this tube soon 

 attached their cases and assumed their characteristic attitude. Others 

 were artificially fastened to stones in various positions. So long as 

 the larvae were facing the current, the legs were in the normal posi- 

 tion. It did not seem to make much difference whether the larvae 

 were right or wrong side up in their cases. One specimen stayed 

 ventral side up for over a week before it took the trouble to turn 

 over. Those larvae that did not face the current, after vain attempts 

 to detach their cases, left them, and proceeded to build new ones. 

 These they attached in the characteristic manner. 



The larvae apparently fail to recognize any food that is not moved 

 into their grasp. Cyclops and water fleas, when held in the stream 

 just in front of the larvae, were absolutely ignored; but when allowed 

 to float down with the current, were quickly seized and devoured. A 

 waterflea, fully one-half the length of a larva was grasped and 

 hugged bear-fashion by the legs that are all armed with powerful 

 curved claws and strong spines (figs. 5, 7). Two very tough 

 Hydrachnids suffered a like experience with the same larva. They 

 were torn in pieces by the strong mandibles armed with sharp teeth 

 (fig. 9) with a speed and regularity that well may be compared with 

 a saw-mill. 



Brachycentrus larvae when placed in still water will assume their 

 characteristic attitude of out-stretched legs. With Cyclops and 

 water fleas within easy reach, passing to and fro, in and out of the 

 case, not the slightest attempt was made to obtain a morsel. The 

 current again started, however, the first unsuspecting intruder was 

 caught and greedily devoured. 



The spines on the femur of the meso- (fig. 7) and metathoracic 

 legs (very similar) may serve as plankton seives. Finely powdered 

 carmine introduced into the stream could be seen strained and caught 

 on these. Then the short spines and row of long straw-colored hairs 



