13 FISHES IN RELATION TO MOSQUITO CONTROL. 
This work around the ponds caused the top minnows to become 
tame, and that made it possible to perform certain feeding ex- 
periments which otherwise could not have been made. One of 
these feeding observations is described in the writer’s field notes as 
follows: ‘‘I took several large Anopheles larve from dense vegetation 
and placed them in open water among top minnows. With one 
larva was a small piece of bark. The larva hovered over this piece 
of bark and the fish did not detect it. When it was placed in open 
water, without the least protection, the fish swam around it, even 
‘nosed’ it, while the larva lay perfectly motionless. At last a rather 
small minnow seized and swallowed it. Placed another larva in 
open water among fish. This one too lay perfectly still, drifting like 
a small stick, while fish swam all about, nosing it a time or two, but 
apparently not detecting that it was alive and something to eat. 
Finally it drifted near a tuft of grass and with a surprisingly quick 
movement it swam into the vegetation. It was removed and placed 
in open water. There it lay motionless for about five minutes, when 
at last it was snapped up by an undersized minnow. A third was 
placed in open water; it too drifted along perfectly motionless for 
about five minutes before it was finally detected by an undersized 
minnow. Once this larva drifted very close to the grass from which 
it was originally removed, but it made no effort to get back into it. 
This may have been due to the presence of fish between it and the 
grass.” In some of the feeding experiments the larve were much 
more quickly detected by the fish than in the one just described. 
The rapidity with which they are found and eaten probably depends 
to a certain extent, at least, upon the eagerness with which food is 
being sought by the fish. 
These feeding experiments, which were repeated many times, 
demonstrated that the protective instinct in mosquito larve is highly 
developed. It was shown many times that the only protection an 
Anopheles larva has from fish in open water is inactivity. When the 
larva thus drifts along fish evidently mistake it for an inanimate 
object, for, as already shown, they may swim all around it for several 
minutes, even touch the larva with the snout and yet not discover 
that it is food. The slightest movement, however, on the part of 
the wriggler apparently never goes unseen and it is instantly seized 
and devoured by the fish. It often happens that a mosquito larva 
placed in open water drifts toward places of protection before it is 
discovered by the minnows and, if no fish are very near, or are present 
between the larva and the place of protection, it moves toward it 
with a remarkable rate of speed and quickly places itself over the 
object near the surface of the water where it can not be seen by 
fish. It, however, remains motionless if fish are near. 
It is not to be assumed from what has been said in the foregoin 
paragraphs that mosquito larve are as abundant in vegetation an 
débris when Gambusia is present as when absent. An Anopheles 
larva may find temporary protection over a blade of grass, but it is 
scarcely probable that this larva will spend its entire existence over 
a single blade of grass, and, if it moves, it is in great danger of losing 
its life. Then when it reaches the pupal stage the blade of grass is 
obviously not as well suited as previously to furnish protection. In 
this stage of life the mosquito appears to be much more active than 
in the larval stage. This would endanger its life still further, for it 
