SOME EXPERIMENTS IN THE PROPAGATION OF 
RAINBOW TROUT. 
BY MR. ALBERT ROSENBERG OF KALAMAZOO, MICHIGAN. 
In presenting this paper for your consideration, I feel some- 
what diffident, as I have no means of knowing that there is any- 
thing new or original in it. 
In view, however, of the discouraging work with these fish at 
other stations as well as ours, and the radical and gratifying 
résults obtained by these experiments, it will, no doubt, prove 
interesting to those fish culturists engaged in this work. 
If you will pardon the digression, I will begin by giving you 
a short resume of the season of 1904, so that a comparison may 
be made: 
Our take Of e858 Was. ',25e sale howler. 252,000 
TGS: Ol SOMOS: WAS Seeks Sites ca eck ote 95,000 
Heschel Auleynaise tet eat ctan clea eee 81,840 
LOSSY) the bg Ae a Po eo ee 11,262 
The eggs were taken at side of pond, regardless of air tem- 
perature, using double pan; lower one containing water. When 
separated, they were carried in pails to the hatchery, washed, 
measured, and placed on trays. 
When hatching was thoroughly started, alevins were taken 
up and placed in fish trays. Here they commenced to die in large 
numbers, when from one to four days old; and all fish culturists, 
who have had similar experiences can realize the awfulness of 
picking dead alevins, bursted sacs, ete. ‘ 
We tried leaving them in tank where they were hatched ; but 
the result was the same—the losses ran from 5 per cent to 100 
per cent in different lots of eggs. 
The next serious, and to me, new feature, occurred when the 
sac was nearly absorbed, and the fish were turned out in a tank 
preparatory to feeding. The little fellows scattered all over and 
lay on their sides gasping—only a few swimming naturally. 
For want of a better name, we christened these “side-wheelers.” 
The losses in batches affected in this way ran about 90 per cent., 
87 

