BULLETIN OF THE UNITED STATES FISH COMMISSION. 157 



there was nothing to be done but to direct all my attention to the ap- 

 paratus, with a view of improving it if possible. 



I must state here that the Americans, in their attempt to develop cod 

 eggs in 1879, used many of the older forms of apparatus, but first ob- 

 tained satisfactory results after the construction of an entirely new ap- 

 pliance. This, which was called from its inventor the Chester rotating 

 apparatus, was, however, also encumbered with various drawbacks, and 

 I determined, therefore, to experiment not ouly with the last named but 

 also with one of the older forms, and I selected the so called Clark's, 

 which appeared to me to be the most successful. The experiments with 

 this were, to be sure, unsuccessful in America, but when I based my 

 calculation upon cleaner water 1 believed that it would prove effective 

 in our country. The result showed that so far 1 was right ; but now the 

 fatal circumstance interfered — the salinity of the water at certain times 

 was too low — and a new difficulty also intervened, which was not, and, 

 without somewhat extensive preliminary experiments, could not have 

 been foreseen. The difficulty was that the cod eggs, which are naturally 

 developed at the surface, remained lying on the bottom, and that in an 

 apparatus in which there is a descending current. 



By accident I made the discovery that the greater or lesser inclina- 

 tion of the apparatus had considerable influence on the direction of the 

 current, and after having made some experiments I had the Clark's ap- 

 paratus set up in such a way as to secure a fall of 8 inches in 8 feet, 

 instead of 1 inch in the same distance, as proposed by an American 

 fish-culturist, Mr. E. E. Earll. By way of explanation allow me to give 

 a short description of the structure of this apparatus. It consists simply 

 of a water tight box, 8 feet long, 2 feet wide, and 1 foot deep. It is 

 divided lengthwise into two equal parts, and each of these again into 

 five smaller compartments, by means of partitions. 



As a result the box contains ten compartments of equal size, besides a 

 smaller one in each end, which serve for the inflow and exit of the water. 

 In each partition there is a depression containing a tin spout, through 

 which the water flows from one compartment to another. Now, when 

 the water is brought into the uppermost compartment this will be filled 

 to within about one inch from the top, whereupon it will flow into the 

 next division, and so on until all the compartments are filled. From 

 the last an exit-pipe extends through the floor. In each of the ten com- 

 partments there is a wooden box, furnished with a wire-gauze bottom, 

 and somewhat smaller than the space wherein it is contained. This 

 wooden box has one of its edges pressed under the tin spout through 

 which the water comes, and then it is naturally elevated at the opposite 

 side in such a way that it rests somewhat oblicpiely. 



In these small boxes, then, the eggs are placed, and, since the fine 

 openiugs iu the bottom furnish the only escape for the water, it is clear 

 that the principal current must be descending. Now, by giving the box 

 a greater inclination, as directed above, the water obtains, in leaving 



