34 FISH CULTUKE. 



quently there is always at the bottom of the tray 

 a mass of water that does not change often, and 

 may almost be held to be stagnant. Now, as long 

 as the ova are supported to within one and a 

 half inch of the surface, this does all well enough. 

 The stream goes along the surface over the ova ; 

 impurity is buried inches below in the water ; and 

 when we consider that at Huningue they do not 

 undertake to hatch fish so much as to distribute 

 the ova in an advanced state of incubation, we must 

 own that no better plan for their purpose could 

 possibly be devised, but for the rearing of fry a 

 change is certainly advantageous. As, however, we 

 are not- come to the rearing of fry yet, we will 

 consider that point in its proper place. The. best 

 size for the trays is, I think, upon consideration, that 

 employed by the French, viz., twenty inches long, 

 five inches wide, and three and a half inches deep, 

 measured on the inside. The spout whence the water 

 is ejected should not be an open spout, as is shown 

 in the French trough, but should be a round hole of 

 at least three-quarters of an inch in diameter i in the 

 clear, and instead of an open spout, it should be 

 a bent metal pipe, curved slightly downwards Like 



1 It must be at least this size, or the tray will be useless for 

 rearing when a larger stream of water is needed. 



