BULLETIN OF THE UNITED STATES FISH COMMISSION. 419 
I have no doubt that four times this number were taken by the shad 
fishermen, who, as before stated, are very cautious about giving informa- 
tion. 
PROSPECTS FOR SPAWN GATHERING. 
There are but two places where the fish have been taken in sufficient 
numbers to enable us to get some spawning fish. These are Troy and 
Gravesend Bay. At Troy a number could be caught or bought from 
the shad fishermen. At Gravesend Bay they might be kept alive by the 
fishermen until the season was over, and then be purchased. The fish- 
ermen at this place did not know of any law on the subject, and openly 
sold their fish in the New York markets, receiving from 25 cents to $1 
per pound, the last figure being for the first two or three fish taken. 
I do not think the water in the Hudson below Troy is cool enough to 
pen salmon. Between Troy and Mechanics ville there are deep, cool 
spots which I think would answer. Mr. Burden thinks that the trout 
ponds at Cold Spring Harbor would be the best place to keep those capt- 
ured in the lower river until the spawning season, and in this I agree 
with him. At Mechanicsville the fish would require watching night and 
day, but this might be done by watchmen at the mills, or an inclosure 
might be made in the tail-races of the mills where it would be difficult 
of access. I believe that some eggs might be taken next season, but 
experiment only would determine the number and the cost of obtaining 
them. If the operations were confined to the upper river a few might 
be obtained at Hudson, through Matthew Kennedy, and taken up in a 
fish-car to add to the Troy catch. If the lower river was to be worked 
for fish to store at Cold Spring Harbor, a small sail-boat with a well in 
it would be needed. 
POLLUTIONS OF THE RIVER. 
In my opinion ordinary house or water-closet sewage does no harm 
to fish in a big river. In proof of this, I would call attention to the 
fact that shad have increased in the Hudson, through artificial culture, 
in spite of the growth of cities along its banks. Chemical works pol- 
lute the water to some extent, but the injury depends entirely on the 
relative amounts of chemicals and water. A poisonous stream enter- 
ing one side of a river does not mix at once with the whole stream, but 
continues down one shore, and is finally precipitated and becomes 
harmless. The poisoned water would kill a fish entering it, but the 
instinct of the fish teaches it to avoid it. At times the muddy water 
of the Missouri River can be seen for many miles below its junction with 
the Mississippi, and this will serve to illustrate this point. 
The paper-mills formerly poured great quantities of chloride of lime 
in the river, and this substance was claimed, rightfully or otherwise, to 
be the cause of the decrease of shad in the Connecticut River some 
years ago, because of the paper-mills at South Hadley Falls, Mass. 
