184 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



• 



introducing" a gate into one side of the cone, through which the shells 

 and fish and dead eggs might go out into appropriate receptacles. This 

 device, so far as it served for the collection of the young fish, was quite 

 successful; but it was not found capable of doing the work for which it 

 was first planned by Mr. Clark, and was abandoned. Similar experi- 

 ments, looking to the same result, were made by him with the Chase 

 jar — the form of apparatus employed for the whitefish work at the 

 Northville station. The result of these experiments, however, led Mr. 

 Clark to the conclusion that an automatic or self-picking arrangement 

 for effecting the complete separation of the dead from the live eggs was 

 not practicable, and a paper to that effect was written and published 

 by him in Vol. I, Bulletin of the United States Fish Commission (1881, 

 p. (>2). The present method employed by him for the separation of the 

 dead whitefish eggs is to siphon off the dead eggs and such live eggs as 

 are necessarily drawn over witli them, and to transfer them to what he 

 terms "hospital jars," the live eggs thus drawn over being left to take 

 their chances with the dead ones. 



This mode of treatment undoubtedly has served to diminish materi- 

 ally the percentage of loss in the eggs, thus treates by him, as in this 

 way, by the sacrifice of a small proportion of the eggs, he secured the 

 complete separation of all elements of contamination and disease from 

 the great bulk of the eggs. 



In 1881, while I was in charge of a shad-hatching station on the 

 Potomac River, and in position to observe closely the performance of 

 the hatching apparatus in use, the question of the separation of the 

 dead from the living eggs was taken up systematically, with the view of 

 devising a form of apparatus which would accomplish the purpose and 

 which would be of such shape as to be of easy and convenient use in 

 practice. Knowing that there was an apparent difference in the specific 

 gravity of the living and the dead eggs I determined to see if I could 

 not avail myself of this difference to effect the separation. The first 

 form of apparatus employed is represented in Fig. 1.* 



*This consists essentially of an oblong trough with wooden ends and sloping glass 

 sides, glass being used in order to be able to observe the movement of the eggs under 

 the influence of the currents. This trough rests upon a rectangular box made of 

 boards, which serves at once as a firm base for the support of the trough, and as a 

 chamber for the equable distribution of the water pressure. The water, which 

 enters the rectangular box forming the base of the apparatus through the supply 

 pipe I, passes to the trough proper through a slot extending the whole length. The 

 influx of the water to the trough is regulated by the valve V V, which, by means of 

 the set rods S S, can be pushed down so as to cut off the flow of water entirely. By 

 setting so as to have the opening between the valves and the glass sides about one 

 thirty-Second of an inch, the water enters the hatching trough in thin sheets which are- 

 directed up the glass sides of the trough. The effect of this is to give the eggs a con- 

 tinuous movement in the direction shown by the arrows. Tlie water Hows over the 

 edges of the central trough, and escapes from the apparatus at O. The dead eggs in 

 their circuit float higher than the li vine,, and the force of the entering current may bo 

 bo regulated that the former will be swept out by the escaping water. 



