CH.Iv.] “LIGHT FLY FOR LIGHTNESS: QU.? 57 
converse to hold good during the day. There 
are, however, doubtless occasional exceptions, such 
for instance as the yellow cow-dung fly, which I 
have found a great killer on a stormy day, par- 
ticularly among meadows, where the natural fly 
would abound, and when it would be swept off 
upon the water. 
On “stickles” (the expressive Devonshire word 
for the broken water at the head of pools) I think 
I have generally found a small bright fly (yellow 
floss-silk body and starling wing, for instance) 
kill better than a larger sombre one, but the 
latter to answer better in the still dark pools. 
This, which I have noticed quite independently of 
the last theory, certainly tends to support it, as 
Trout, in common with most fresh-water fish, un- 
doubtedly lie upon the shallows when the weather 
is bright and hot, and retire to the deeper water 
as it becomes cold and gloomy. It is for the 
same reason, as I believe, that Trout do not, in hot 
weather, so much affect those parts of a river 
which are densely wooded, and consequently shad- 
ed, as those which lie open and exposed to the 
sun. A red fly I have always found kill better 
rather late in the evening than at any other time 
of the day. In the white-moth tribe, as night-flies, 
