ARTIFICIAL HATCHING OF SPAWN. 71 



a large portion of them to some other tray that they 

 may not be overcrowded. Little by little, the bladder 

 is drawn up and absorbed into the system ; as it 

 disappears the fish grows more and more active, 

 darting and skipping about the troughs. Soon no 

 trace of it is to be seen, and yet the water, if not 

 filtered (which it needs not to be after the fry are 

 hatched), will afford them for a day or two sub- 

 sistence ; and now they begin darting to and fro 

 searching for more substantial fare, and they require 

 to be fed. And now it will be found that they require 

 still more room ; and if the pisciculturist does not care 

 as yet to turn them into the river, he should have 

 either a small shallow pond constructed, or a large 

 and long box or two of considerable extra capacity. 

 These boxes should be gravelled as the trays are, but 

 should be somewhat higher out of the water, as the 

 small fish are very active and apt to jump out. The 

 tops should be covered only with fine sparrow- 

 netting, to keep mischievous birds, as kingfishers, &c. 

 out, so as to admit plenty of light and air, and to 

 allow small flies to find their way to the water, 

 which at this period form the most .natural, and by 

 no means a scanty, supply of food. About two inches 

 depth of water will be sufficient, and a fair but gentle 

 stream should be turned on. In places the gravel 



