PROPAGATION OF FISH. 241 



thousands by a yearling happening accidentally to get 

 among the fry just hatched. 



The ponds from which my experience is mostly drawn 

 I have no part nor parcel in. further than what may be 

 claimed from having furnished extensive advice ; but I 

 have taken a deep interest in their success, and if I speak 

 of them as my ponds and my experiments, the party that 

 really made them will understand and excuse me, and 

 the public will comprehend that no egotism is intended. 



They are five, and will soon be more, built of gravelly 

 sand along a bank that is filled with pure, clear springs, 

 and about five feet deep. In the first place, the musk- 

 rats are annoying, having in one instance tunnelled a 

 dam and let out over a hundred breeders, and when the 

 expense is not an object, it is well to put a fence of 

 boards or even stone-work in the middle of the dam. 

 These ponds all communicate with one another at the 

 surface of the water by a sluiceway that can be closed 

 at pleasure, and each can be drawn off dry separately. 

 In that which might be called the highest, and communi- 

 cates directly with the principal springs, are placed a 

 row of boxes about twelve feet square and one foot deep, 

 divided into compartments by divisions of wood, with 

 openings cut to let the water circulate. The boxes are 

 kept on a level by being supported on stout pegs driven 

 into the sand, and are filled about half full of pure white 

 gravel, leaving the water about four inches deep. There 

 is an open passageway to the outside and through the 

 partitions, whereby the fish can enter, and doors, com- 

 posed mainly of wire, keep off leaves and twigs that 

 would foul the water, but can be lifted at pleasure. 



Jl 



