ARTIFICIAL PROPAGATION OF THE RAINBOW TROUT. 
249 
the food in such manner as to prevent each one from getting its proper share. If 
liver is used as food it is very difficult to distribute it evenly through the water 
so that all the fish will partake of it without the water in the trongii being made 
milky from its use. This is very objectionable and, in my opinion, very injurious to 
the young fish. 1 tliink that it produces iufiammation of the gills and a slimy, itching 
disease of the skin, often causing heavy mortality among the young fish. The fry 
will be ready to take food as soon as the yolk sac is absorbed; the time required for 
this will depend upon the growth of the fish, superinduced by the temperature of the 
water. At Wytheville, where the Avater temjierature is regular at 53°, they will take 
food in about 30 days after hatching. The time to commence feeding can be closely 
determined by watching the movements of the fish. Before the yolk sac is entirely 
absorbed they begin to break up the schools on the bottom of the trough and scatter 
through the water, rising higher and higher from the bottom of the trough each day 
until they can balance themseh^es gra<;efully in a horizontal position, all heading 
against the current and swimming Avell up in the Avater. They are then apt to be ready 
for food, but, to make sure of it, drop some small bits of cork or a nap from red flannel 
on the surface of the Avater, and if they stiike at it as the current cari’ies it doAvu,giA"e 
them food, but do not give it to them until they strike at the substance floating. 
Their food should be prepared very fine, and if it consists of liver it will be found 
necessary to put some water Avith it before feeditig, in order to make it distribute 
evenly. The liA^er can best be given to the fish Avith a feather, by dipping the feather 
into the liver and skimming it over the surface of the water in the trough. After the 
fish grow to be to l.f inches in length they Avill begin to take the food that settles 
on the bottom of the trough, and as it is then not so tedious and difficult to feed them 
the food can be given from the haaid and it will not be necessary to mix it Avith water. 
The young fry should be fed five or six times a day, giving them their food sloAvly and 
sparingly. After they learn to take their food well from the bottom of the trough 
three feeds a day will be suflicient, as they can then be given more at each feeding. 
The rearing ponds . — Ponds for rearing fish should be narrow, say from 8 to 12 feet 
Avide, and any desired length, u]> to 00 feet. For convenience in draAving otf the 
ponds, etc., I would not advise making them more than 00 feet long. The size, shape, 
and arrangement of the ponds must depend altogether upon the topography of the 
ground where they are to be constructed. If it can be so arranged, I prefer to have 
these ponds built on a hill-side, one pond above the other, Avith earth and piling 
embankments at the lower sides and at the ends. A pond of this kind is shoAvn 
at plate 91, and is the one which I shall endeavor to describe. In the construction of 
these ])onds various materials may be used for damming the Avater. The embank- 
ments may be made of earth altogether, or they may be lined on the inside next to the 
Avater with stone, brick, cement, or timber. In all cases they are constructed Avith 
the same end in view, and the general principle is very much the same. It is not 
always convenient to use certain materials, but where the gi'ound is of a porous or 
loose formation it will be necessary to use piling or cement for the inside of the embank- 
ments, and i)ossibly cement for the bottoms. 
I prefer in all cases to have earth bottoms, where the nature of the ground will 
lierinit. The water should enter the pond at one end and discharge from the corner 
having the deepest water in the opposite end. The bottom should be graded as 
shoAvn in the cross section 2, ]date 91, and Avith a slope toward the outlet, so that all the 
