I1O FISH-CULTURAL ASSOCIATION. 
have not answered it satisfactorily to myself yet. They could 
not have been all caught out as the stream is preserved. From 
observations the writer thinks that many have gone down, find- 
ing their way into the Genesee river and Lake Ontario, just as 
the California salmon did several years ago; they have gone as 
suddenly as the salmon. Stories are afloat of large ones being 
caught miles below. As the spawning season approaches they 
also run up stream just as far as they possibly can, and as the 
stream is generally at its best at this season they cannot get 
back unless they do so before the water subsides. I have often 
found them in water holes that had no connection with the 
stream except during high water and where they would die in a 
short time. I heard of one found in a man’s garden this spring 
that was nearly a mile away from the stream, the fish had gone 
up there in a little stream that was formed by melted snow-and 
rain, and which run dry in aweek. Brook trout generally find 
their way back and don’t get stranded. You would suppose 
that the natural increase would keep the stock up in a preserved 
stream, but it does not in this case, and here I would call your 
attention to the fact that at the best not more than 50 or 60 per 
cent. of the many rainbow trout eggs taken at the hatcheries 
at Caledonia can be impregnated. There is no such percentage 
of empty eggs of others of the trout family that are handled 
here: 
During the past winter I made an experiment with eggs taken 
from a fine healthy brook trout, impregnated by a number of 
good males of the same. First, 1 took 350 of-her eves; placed 
the milt with them and then washing it off as quickly as possible, 
and forty-five seconds after taking the eggs placed them on the 
screens in the hatching trough, Next; Il took |350° moreveoos 
from the same fish and let them stand three minutes before 
washing off the milt. Next, the remainder of the eggs the fish 
contained, 335 in number, I let remain in the spawning pan the 
usual length of time—about thirty minutes. The three lots I 
carefully placed on trays, picking out the bad ones every day, 
until they were old enough to plainly show the eye spots, when 
I counted what I had left of each of them: 
