BY FREDERICK A. A. SKUSE. 51 



which some authors have made out to be the stigmata of the 

 pupa. Winnertz suggests that certain tooth-like projections on 

 the pupa may serve for breaking through the cocoon ; in many- 

 cases after the escape of the imago the pupa skin is to be found 

 hanging outside the gall. The number of the abdominal segments 

 is nine, counting the anal joint, or that containing the genital 

 organs. These insects ordinarily continue a very brief time in 

 this state. The pupte of the celebrated Hessian fly [Cecidomyia 

 (Biplosis) destructor'] are known as the ''flax-seed" stage of 

 that fly. 



IV. Habits and hahitats of the perfect msects. 



In the perfect state I believe the majority of Cecidomyidse 

 live but a very short time, though individuals of some fe^v species 

 I know are obtainable for two or three months. In Australia these 

 flies may be found throughout the year, but they abound mostly 

 during the early sj)ring — August and September. They do not 

 vary very considerably in their modes of flight, and the majority 

 seem to resort to some shady retreat during the heat of the day, 

 such as the interiors of caves, the crevices in the rough 

 bark of trees, &c. ; towards evening they frequently fly into 

 houses, and may be seen vivaciously fluttering up and down 

 the window panes not directly e.xposed to the sunlight, 

 apparently all the while just touching the glass with their fore- 

 legs and butting with their head as they proceed, and by that means 

 possibly accelerating their pace ; aftei- dark, like many other 

 insects, they are attracted by artiflcial light. I have seen 

 a few species leisurely hovering over the tops of low 8hnal)s 

 in the fullest heat of the day, evidently preferring .sun-shine to 

 shade. Some only travel a very liinited distance ; one species 

 which I have noticed in very short grass (juite open to the sun in 

 the midst of the summer, appeared never to tly for more tlian a few 

 seconds at a stretch, and then only to reach a spot not more than 

 twelve inclies from its previous position. The flight never apj)oars 

 to be along in a direct Hue, Ijut is an irregular dancing move- 

 ment, totally precluding any sort of conjecture an to the direction 



