370 SALMON AND TROUT. 
mount to saying that the object of a scientific dry-fly fisher 
should be to so manceuvre his artificial fly as to make it as far 
as possible copy in its movements, as it should in its appearance, 
those of the natural insect. The natural insect emerges from 
the nymph-envelope on the surface of the stream, and as far as 
it drifts down on the water is carried along at the same speed 
and in the same direction as the run in which it happens to 
be when first clear of the shuck. Under no condition is it 
very likely for a shy fish like a trout to take a fly deviating from 
this natural course, and the more a river is fished the shyer the 
trout become, and the less likely they are to forgive a mistake 
in this respect. 
Wherever the run of the water has the effect of causing the 
artificial fly to drag, there the fisherman is likely to find himself 
foiled in all his efforts to rise the fish, and the place should, as 
a rule, be avoided. On the other hand, wherever the run of 
the water causes the artificial fly to follow exactly the course 
taken by the natural, there a rising fish is likely to be tempted 
by a good imitation delicately and accurately placed. As a 
general rule, wherever the action of the water on the line 
eauses the artificial fly to deviate in pace or direction from that 
which the natural insect would follow in a similar position, a 
wake is produced behind the fly, and this is technically termed 
dragging. 
There are three conditions under which dragging may take 
place. A fly may travel either faster or slower than the natural 
insect, or in a different direction from it. 
The fly travels faster than the natural insect in a place where 
the angler has to throw across the stream, and where the most 
rapid portion of the current is between him and the spot where 
the fish is feeding. The fly then drags because the action of 
the stream on the line causes the fly to travel at the pace of 
this the more rapid stream, instead of at the rate of the portion 
of the river where the fly is floating. It further has the ten- 
dency of dragging the artificial fly more or less across the 
normal direction of the stream. This form of dragging can be 
