32 FISH CDLTUEE. 



or straining medium again before distribution. These 

 are matters whicb in their adaptation are so likely 

 to be governed by circumstances, that much must 

 be left to the ingenuity of the pisciculturist. As 

 regards the apparatus io be used for the simple 

 purposes of incubation, I„ prefer the French plan 

 as employed at Huningue, which is, I believe, the 

 invention of M. Coste. » 



This plan consists, as I have already partly pointed 

 out, in a series of trays, which may be made either 

 of glazed earthenware or slate. I prefer the slate for 

 reasons that will appear anon, and chiefly because 

 it is capable of being bored or sawed without diffi- 

 culty, and is less liable to break. "Within each of 

 these trays is deposited a grating or grille, resting 

 in about mid-water, and made of glass rods enclosed 

 in a wooden frame, placed sufficiently near to each 

 other to prevent the ova from slipping down between 

 the bars, while there is a sufficient opening to allow 

 any deposit which may be in the water to pass 

 through the bars, and settle to the bottom of the tray. 

 This apparatus is already shown at page 10. Level 

 with the bottom at each end is a small, round hole, 

 shown at (a), which can be stopped with a cork, but 

 which, when dirt accumulates at the bottom of the 

 tray, can be removed ; the dirt may then be swept 



