558 REPORT OF THE COMMISSIONER OF FISHERIES. 
generally true with all storms, whatever their direction, it would 
surely be known to the fishermen, who utilize this movement of the 
fish in the river for their capture, as will be explained later. 
As mentioned above, the water level at the upper end of Lake Erie 
is very variable. The long axis of the lake lies nearly west-southwest 
and east-northeast, so that both westerly and easterly winds have a 
ereat*influence in piling the water at one end or the other. The pre- 
vailing winds of summer are southwesterly to westerly, so that the 
level is almost constantly changing. This gives a great resemblance 
to tides, except that the changes are, of course, much less regular, and 
generally of less amplitude. A strong southwest wind, however, 
blowing steadily for a day or two, will lower the general water level 
in Sandusky Bay, for instance, a foot or more, while a long-continued 
storm may result in an even greater change of the level. As soon as 
the wind ceases, or shifts around to the opposite direction, as is usually 
the case in our cyclonic storms, the reverse current sets in, affecting 
the water for miles up the Sandusky and Portage rivers. 
Just how far this variation of the water level and the consequent 
reversion of flow of the rivers influence the movements of the earp Iam 
unable to say. This much, however, is certain. A fall of a foot or even 
less in the general water level means the laying bare of great expanses 
of marsh land, and the carp which were feeding over this area have to 
seek deeper water as that on the flats gradually becomes shallower. 
In this way they work into the smaller streams, and so into the larger 
erecks, and from these into the river. It is at such times that they 
are taken in large numbers in a seine which has previously been 
stretched across the mouth of the creek, as will be described more 
fully in connection with the methods of fishing (p. 613). The fish 
appear to be quick to appreciate the lowering of the water, for they 
begin to run out very soon after it has begun to fall. Conversely, they 
run up again and spread out over the marshes as the water rises. 
This movement, which seems to depend upon the gradual lowering 
of the water in the shallow places, is distinetly different in nature from 
the ordinary reaction of most fishes to a current of water. As is 
well known, most fishes, when placed in running water, immediately 
react by turning head-up into the current.¢ That this is true of 
young carp, I have ascertained by experimentation. It may also be 
the explanation of the crowding of these fish around the inlet when 
fresh water is being pumped into a pond, a phenomenon which will be 
described more fully in the discussion of their reaction to fresh water 
(p. 560). It is equally true that most fish become uneasy as the water 
in a vessel or other container is gradually lowered without producing 
a definite strong current. It is probably this ‘‘ uneasiness” which 
causes the fish to leave the marshes as described above. 
aFor a discussion of the orientation of fish to running water see a recent paper by Lyon (1904). 
