PROPAGATION OF TROUT IN AMERICA. 191 
supply of water, was first selected, and operations commenced at once. 
Across the stream, which was scarcely thirty inches in width, a rude 
weir was thrown so as to check the progress of the water, and to form a 
dam which extended several yards toward the source of the rill, with 
the average depth of twelve inches. The necessary continuous supply 
of water was thus obtained, the remainder passing away in an artificial 
channel cut for the purpose. 
The boxes for containing the ova were twenty-four in number, each 
being six feet long, eighteen inches wide, nine inches deep, and open at 
the top. The whole were disposed in a double row, parallel with the 
original course of the rill. Each row consisted of twelve boxes, placed 
end to end, the beds of the foremost commencing shortly below the 
lower end of the dam. A piece of three inches in depth and nine in 
width was cut from each log in order to allow a free passage for the 
stream through the whole series. At the junction of each box was 
nailed a sheet of tin with turned-up sides to prevent the escape of the 
water. A couple of pipes" a yard in length and two inches in diameter 
conveyed the stream to the foremost box in each row, the end of the 
pipes inserted in the dam being covered with fine wire gauze to prevent 
the entrance of trout and insects. The whole were arranged on a gentle 
slope, so as to avoid stagnation, and ensure a tolerably rapid flow of water. 
The boxes being arranged, a strata on which to place the ova was 
then formed. It consisted of a mixture of sand and gravel, of the depth 
of several inches, upon which were deposited pebbles of the ordinary 
size of road metal. When properly prepared for the reception of the ova, 
the stream averaged two inches in depth above the average of the stones. 
At a short distance below the dam two ponds were constructed to 
contain the fry, the one receiving the stream from the double row of 
boxes, and the other from the bed of the rill. The superficial area of 
each was 240 yards, being much too small, as finally ascertained, for the 
hosts of fry with which they were ultimately tenanted. 
A spawning bed in the immediate vicinity afforded every facility. 
In order to obtain the spawn in a perfectly mature state, the fish were 
taken from the spawning bed in the very act of its deposition. They 
were caught with nets at night. When taken they were instantly, and 
without injury, put into an oval tub one-fourth full of water. So soon 
as a pair of suitable fish were captured, the ova from the female w r as 
immediately discharged into the tub by a gentle pressure of the hands 
from the thorax downward. The milt of the male was ejected in a 
similar manner, and the contents of the tub gently stirred with the 
hand. After the lapse of a minute the water was poured off, with the 
exception of sufficient to keep the ova submerged and fresh supplied in 
its place. This also was poured off and fresh substituted previously to 
removing the impregnated spawn to the boxes prepared for its reception. 
In discharging the ova from the abdomen of the female, all violence 
was carefully avoided. If on examination the ova were found to be 
