BULLETIN OF THE UNITED STATES FISH COMMISSION. 183 



15.— HISTORY OF THE EXPERIMENTS LEADING TO THE DEVELOP- 

 MEIVT OF THE AUTOMATIC! FISH-HATCHINC} JAR. 



By MARSHALL McDONALD. 



The work of practical pisciculture was, until a comparatively recent 

 period, confined for the most part to the hatching of the different spe- 

 cies of the salmonidse. The incubation of the eggs was at first effected 

 in troughs having the bottoms covered with a layer of gravel, upon 

 which the eggs were placed and over which a current of fresh water was 

 allowed to flow. 



In succession followed the "grill system" of M. Coste and the differ- 

 ent devices of movable trays now in common use for handling this class 

 of eggs. In all these various methods the separation of the dead eggs 

 from the live ones was effected by means of hand-picking. The neces- 

 sity for the separation, although not so urgent in the case of the eggs of 

 the salmonidse as in that of those eggs which develop in warmer waters 

 and in much shorter periods of time, still entails a vast amount of labor 

 in connection with the hatching operations. 



Although the ingenuity of our fish-culturists has greatly improved the 

 forms of hatching apparatus for these heavy eggs, yet up to a compara- 

 tively recent period no other effectual means of separation than that 

 above indicated has been found practicable. The United States Fish 

 Commission, in the development of its work, had presented to it the 

 necessity of dealing with the eggz of the whitefish and the shad upon 

 a scale Unprecedented in the history of fish-culture. Millions of eggs 

 were to be hatched where fish-culturists formerly handled only thou- 

 sands, and the old methods of hand-picking were soon found to be im- 

 practicable. 



In all of the forms of apparatus for bulk hatching no adequate means 

 is employed for the separation of the dead eggs from the living. All, 

 as they come from the fish, the uninipregnated as well as impregnated, 

 are placed in the apparatus and remain together. 



In the case of the whitefish, and more especially in the case of the 

 shad eggs (which run through their period of incubation in a much 

 shorter time), fungus rapidly develops among the dead eggs, communi- 

 cates itself to the living, and large numbers of them, which would 

 otherwise reach the period of hatching, are destroyed. The percentage 

 of loss produced in this way is always considerable, and in many cases 

 none of the eggs undergoing incubation are saved. The attention of 

 fish-culturists was early directed to the serious losses thus arising, and 

 various experiments have been made with a view of effecting the sepa- 

 ration of the dead from the living eggs. 



In 1878 Mr. F. 2S". Clark, the superintendent of the United States 

 Hatchery at North ville, Mich., attempted to effect the separation by 



