244 PROPAGATION OF FISH. 
orange, but wlien impregnated change to a pale, trans- 
parent yellow, witli a small black spot in the centre, and 
on losing their vitality turn white. When one dies in 
artificial breeding, it communicates disease to the rest 
unless it is removed, but in the natural method the cov- 
ering of sand isolates them and ap^^ears to prevent this. 
The water must flow steadily, and, if not sufficiently 
fresh and abundant, should be allowed to fall in little 
cascades, to give it life. As the operation proceeds, the 
change in the egg can be readily seen with the micro- 
scope, the cells of w^iich it is composed clustering to one 
side. Any one curious on this subject is referred to Dr. 
Garlick's interesting little work on fish culture. 
Every old fish having been carefully removed, the 
breeding-pond is refilled immediately, and the young, 
awkward, unwieldy little infants, soon make their appear- 
ance, and after their thirty days of self-sustenance are fed 
on worms cut in small pieces, on grated or pounded fish or 
meat cooked. Before Summer is over they are a few 
inches long, and are driven down into their new home 
by again drawing oS the water of the breeding-pond 
into the latter, and frightening them into it. 
The number raised in this manner is perfectly surpris- 
ing. From about one hundred breeders, male and female, 
we produced over seven thousand fish, the latter fact being 
determined not by guesswork but by actual count. An 
average of nearly one hundred and fifty young, in spite 
of frogs and accidents. One is apt to under-estimate in- 
stead of exaggerating the number by guessing ; they are 
so small and lie so quiet that they almost escape obser- 
vation. 
