7 6 Mosquitoes 



active so long as any of the water remained liquid. 

 Some larvae, kept with the water partially solidified 

 about them, were alive at the end of the fourth day, 

 but were killed that night when all of the water 

 congealed. On February 9th, Dr. Wright removed 

 the snow and ice which for two weeks had covered 

 a pool to a depth of from eighteen inches to two 

 feet ; in the water he found active Cnlcx,Anop1ieles, 

 Dixa, dragon flies, and other insects, but no culicid 

 pupae. Larvae which pupated in November, De- 

 cember, and January died before completing their 

 changes, being unable to resist the low temperature 

 as did the other, younger larvae. It is probable that 

 mosquitoes do not winter in the pupa stage. 



Sight of Larvae. — I have no reason to think that 

 the sight of larvae is well developed. It is true that 

 they distinguish between light and darkness, a sud- 

 den shadow often sending them wriggling wildly 

 away, but from the character of their movements 

 when any small water dweller, such as a water-flea, 

 approaches, the probability is that they do not see it, 

 but sense it with their feathered hairs when it is close 

 enough. It seldom comes near enough to touch the 

 hairs, however. They can evidently smell in some 

 manner, as they will gather around some large piece 

 of food, a dead larva for instance. 



Moulting. — The larva sheds his skin three times 

 before pupation, the exuviae cast when the pupa 

 emerges being sometimes referred to as the fourth 

 moult. The first stages often differ much from the 

 last. Before the primary moult the head is larger in 



