( xvii ) 



but when this happened — and some kept it up for a fortnight 

 — I dug them up and started them in a fresh hole, and at 

 last they adapted themselves to circumstances. I was then 

 enabled to watch their habits underground. They deepened 

 the burrow by using their mandibles as picks and their 

 heads as shovels and elevators, for as they excavated heads 

 down, they drove their long-pointed claws into the sides above,, 

 and making a larger excavation on one side they passed the 

 sand above by simply turning up their heads, ramming the 

 sand past and above their tails ; until, having a sufficient 

 weight or load, the larva reversed its position, put its head 

 and first thoracic plate underneath, and simply pushed and 

 pushed until it raised the sand to the ground-level, where a 

 large heap accumulated, which from time to time was shot 

 away. 



" When the desired step was reached and the pitfall set at 

 the entrance, the larva quietly put its head and thoracic plate 

 to the top to bar the entrance, the feet were driven into 

 the sides below the mentum, and the fifth abdominal segment 

 drawn vip vmtil the hump was wedged close up and under the 

 thorax, bringing the two vertical doi'sal spines at right angles to 

 the burrow, into the sides of which the points were driven* the 

 remaining abdominal segments were curled round until the anal 

 spines touched the opposite side, into which they were driven, 

 thus securely holding the larva, in its peculiar zigzag form, 

 ready at a moment's notice to drop down to the bottom of 

 the burrow by releasing the anal spines, and straightening 

 the abdomen. Occasionally the larva in its deepening would 

 meet with a small quartz pebble, which it seized between its 

 mandibles and carried to the top, where, with a rapid back- 

 ward movement of its head, it jerked the pebble away a 

 distance of ten inches. 



" Watching the larva through the glass, I pushed a dead fly 

 near. When it was within five-eighths of an inch it seemed to 

 suddenly disappear, and appear at the bottom of the burrow 

 in company of the larva. The movement was almost too 

 rapid to follow and observe how it was made, but after end- 

 less watchings I was able to follow its movements. I caught 

 a Blow Fly, killed it, and tied a thread of floss silk to its leg, 



