[3] SALMON HATCHING AT ROSLYN. 875 



had a large ball hanging by a string, the fish made a sudden turn to 

 escape the feather, which was under it, and I saw the cord break and 

 that portion of the sac contained in the ball fall to the bottom. That 

 particular fish was soon lost in the mass and could not be identified. I 

 preserved several specimens which had lost the pendant ball and were 

 about ready to take food. Of the original fifteen thousand in the in- 

 fected trough, about three thousand died with blue swelling, and two 

 thousand more from other causes, leaving ten thousand fry now taking 

 food, of which a greater portion have lost some part of their sac. I 

 firmly believe that had I not applied a remedy promjitly the whole lot 

 would have been past saving if left in that trough twenty -four hours 

 more. 



To those to whom it seems incredible that part of the sac of a trout 

 or a salmon should be thrown off by a mighty effort of nature when 

 found to poisoned, I would suggest following my experiment, if a blun- 

 der can be so called, and when the liver of the fry turns white, remove 

 the fish into a clean, healthy trough, and note the result. 



In this connection it has occurred to me that the reason that trout do 

 not flourish below saw-mills is on account of the water being impreg- 

 nated with either pine or oak. In 1875 I lost a lot of California salmon 

 at Blacksburgh, Va., in an oaken trough which one of the then fish com- 

 missioners of Virginia, in whose employ I was, insisted upon my using. 

 The impregnation of tannin was perceptible to the taste, and the fry 

 died as fast as hatched. The theory of the fishermen near saw-mills is 

 that the sawdust gets into the gills of trout and kills them. This may 

 be true to some extent, but I doubt it, for the reason that sand or other 

 material does not appear to injure the gills, and I have taken adult 

 trout below saw-mills. I incline to think that the mills are destructive 

 merely to the young, by covering tho spawning beds to some extent 

 with sawdust, but more by the absorption of turpentine from the pine 

 or tannin from the oak, the evil effects of which we know too well. 



From this insufficient tarring I probably lost 30,000 fry more than the 

 regular percentage to be expected, and a lot of 8,000 weak ones, which 

 were crowded against the lower end of a trough, were turned out into 

 Mr. Clapham's stream. By March 6 the white liver had largely disap- 

 peared and the dropsical fish had died and no new cases appeared. All 

 went well from this time. The charred troughs were kept for a lot of 

 fish, 57,000, which arrived March 17, making 344,500 eggs in all. 



Eggs received. 



January 28 120, 000 



February 4 80, 000 



February 16 37, 500 



February 17 50, 000 



March 17 57, 000 



Total ,.., 344,500 



