Trout Breeding. 97 



do as well and would not buckle as my rod did if there 

 was any obstruction. A tumbler rested on a strip at the 

 head of each trough, and in the bottom of it was bored 

 a J-inch hole ; in this hole worked in and out a bit of 

 wood, about half the diameter of the hole, attached to 

 the rod by an arm. The tumblers were filled with food 

 wet to the proper consistency ; the wheel revolved ; the 

 long rod worked back and forth, forcing the wooden 

 pins in and out of the tumblers and dropping the food 

 automatically. I put one in the hatchery at Cold Spring 

 Harbor, N. Y., but found that it required constant at- 

 tention to see that the tumblers were not clogged and 

 that their contents were fluid enough but not too fluid. 

 In theory the thing was perfect. In practice the water 

 settled to the bottom and went out first, and when it 

 was gone the liver would not flow ; it remained in a 

 solid mass and formed an arch above the wooden pin, 

 which tried to do its duty but could not. 



Such experiments are of value, even if the results 

 are failures, for they show us what to avoid. Mr. Hox- 

 sie looked after his own trout and had comparatively 

 few. I had more, and, working for the State, it was 

 desirable to produce the best results without too great 

 regard for expense, and so I went back to hand-feeding 

 with brains behind it. I had a man, Foster Van Aus- 

 dall, who was the most .persistent trout-feeder I ever 

 knew, and — I am glad to say so out loud — he loved to 

 see trout feed, either old or young, and he was on his 

 feet all day feeding. When he came to the lower end 

 of the last trough he would begin at the head of the 

 first again, and so he went the rounds day by day while 

 the other men were on the road with trout for stocking 

 public waters. He loved his work, and the trout 

 showed it. 



