MODES OF EMERGINGFROM PUP^. 317 



dergo their transformation, emerge from their sub- 

 aqueous galleries and come to the surface of the 

 water. As they must keep their wings dry, the pro- 

 cess would appear to be one of considerable diffi- 

 culty ; yet an observer may remark that they perform 

 it with the utmost ease. In the instance of the gnat, 

 this process of emerging from the v/ater is still more 

 conspicuous on account of the difference of form in 

 the pupa and the fly. 



About eight or ten days after the larva of a gnat is 

 transformed into a pupa, it prepares, generally to- 

 wards noon, for emerging into the air, raising itself 

 up to the surface so as to elevate its shoulders just 

 above the level of the water. It has scarcely got into 

 this position for an instant, when, by swelling the 

 part of its body above water, the skin cracks between 

 the two breathing tubes, and immediately the head of 

 the gnat makes its appearance through the rent. 

 The shoulders instantly follow, enlarging the breach 

 so as to render the extrication of the body compara- 

 tively easy. The most important and indeed indis- 

 pensable part of the mechanism is the maintaining 

 of its upright position so as not to get wetted, which 

 would spoil its wings and prevent it from flying. Its 

 chief support is the rugosity of the envelope, which it 

 is throwmg off, and which now serves it as a life-boat 

 till it gets its wings set at liberty and trimmed for 

 flight. The body of the insect serves this little boat 

 for a mast, which is raised in a manner similar to 

 moveable masts in lighters constructed for pass- 

 ing under a bridge, with this difference, that the 

 gnat raises its body in an upright direction from the 

 first. " When the naturalist," says Reaumur, " ob- 

 serves how deep the prow of the tiny boat dips into 

 the water, he becomes anxious for the fate of the 

 little mariner, particularly if a breeze ripple the 



T 3 



