[3] PROPAGATION OF CODFISH. 781 



instead of copper wire to cover the jars and placed these in the box 

 with tidal motion as before. This lot of eggs was hatched out with a 

 small percentage«of loss. 



From my own experience I have come to the conclusion that cod eggs 

 will float for five or six days, but at the end of that time they begin 

 clotting together and sink. 



It would be impossible for me to give in detail all the different ap- 

 pliances and means that were resorted to, but they were all carried on 

 with the view that it is necessary for the cod eggs to be submerged for 

 awhile and then allowed to rise to the surface, and every motion that 

 could be conceived was tried to attain this end. I think it a great mis- 

 take to use any metal whatever in fitting up any kind of apparatus for 

 cod hatching. 



Although prepared for applying the tidal motion, I had never up to 

 this time fitted the tubs up, applying this motion. On the 23d of De- 

 cember I fitted up one glass aquarium and wash-tub with tidal motion, 

 using cheese-cloth screens made to fit tight on the inside, about 4 inches 

 from the bottom. 



On the 25th I found that most of the eggs had gone to the bottom in 

 both apparatus, owing to the density of the water having fallen from 

 .025 to .021 degrees, and upon examination found that the pumps were 

 drawing fresh as well as salt water, which, of course, put an end to this 

 experiment, as the eggs were all destroyed. 



On the 28th I put in another lot of eggs, which did very w r ell, but do 

 not think that I got more than 50 per cent, of young fish ; but even 

 this was an improvement on anything heretofore accomplished. 



On the 6th of January, 1880, 1 put a fresh lot of eggs in the aquarium 

 and one tub, which did well until the 9th ; but for some unaccountable 

 reason at least one-half the eggs in the aquarium had gone to the 

 bottom and were dead, while those in the tub were still doing splendidly. 

 These commenced hatching on the 19th, and by the 22d all had hatched 

 out, not more than 10 per cent, having been lost. 



After the many experiments tried both by Captain Chester aud my- 

 self I have no hesitation in saying that the best conditions for success 

 in cod- hatching are : 



(1) As little motion as possible, with just sufficient change of water 

 to keep it fresh. 



(2) To use entirely filtered water, which can be easily done by filling 

 a McDonald jar with cotton, aud fitted up as is done in shad work. 



(3) To avoid the use of anything like metal in fitting up an apparatus. 



The work now ceased for a time, as' the codfish in live cars had all 

 died on account of the extremely cold weather, and I was instructed to 

 proceed to Florida with half a million of the young fry. Up to the 

 time I left I estimate that we had taken about 15,000,000'eggs, all of 

 which were lost in experimenting, with the exception of about 2,000,000, 



