748 REPORT OF COMMISIIONER OF FISH AND FISHERIES. [8] 



If the eggs are in a thick layer on the bottom, only a few, compara- 

 tively, develop, and this fact consequently necessitates the use of shal- 

 low jars, or dishes, having a large bottom area, during this part of the 

 process, or else some other form of ai)paratus which will produce the same 

 desired result, viz, freedom from pressure. During the past season we 

 accomplished this end, approximately, by placing the eggs when once 

 free of debris and unripe eggs or excessive spermatozoa, in soup plates 

 or large platters ; at Fair Haven I used the largest dishes I could pur- 

 chase and with good results. The jnincipal point is not to overcrowd 

 the eggs ; it is better to err on the side of too much caution, sacrificing 

 some of the eggs if necessary, rather than in the attempt to save all, 

 run the risk of very many failing to develop. I also devised another 

 method which gave fair results, and which appears susceptible of 

 improvement. It consisted of a small glass funnel holding about a 

 pint, and having a jet of air through a narrow orifice in the bottom. 

 The water containing the eggs was placed in the funnel and a small 

 stream of air, as little as possible, was forced up through the apparatus. 

 The eggs by this means were kept constantly in motion, rising in the 

 middle and sinking along the sides of the funnel. We got very good 

 results from this appliance but I am not prepared to say that it suc- 

 ceeded very much better than the platters. A slight imi)rovement in 

 the form of the funnel would add to its efficiency ; the nearer perfectly 

 conical its interior form the better. 



We did not find it necessary to add any water during the process of 

 segmentation, but having once placed the eggs in the plates or f.innel, 

 we allowed them to remain undisturbed until the first swimming stage 

 was reached. It is very important that w^en using the plates all the 

 preparations for operations subsequent to the time of arrival at the 

 swimming stage should be completed several hours before that event 

 occurs. The plates must not be moved nor the eggs disturbed, else it 

 will be impossible to avoid carrying into the second receptacle many 

 undeveloped eggs. In using the cone, a rest due to cessation of the air 

 jet for one hour or more was allowed, before any attemi)t to remove 

 the swimming embryos was made. 



Attention is called to Brooks' description of the development of the 

 egg, in order that the significance of this latter point may be appreciated. 

 The embryo swims at the surface but a comparatively short time; very 

 large numbers of eggs never reach this stage and the swimming embryos 

 must be separated from them before the disintegration of the former 

 renders the water so impure as to insure the destruction of the latter. 

 This removal is not always easy of accomplishment, but we succeeded, 

 partially, by using two plates. In one was placed the eggs during 

 segmentation, with water enough to nearly fill it. The second plate 

 was arranged under the first, so that any overflow from the latter would 

 fall into it. After the embryos began to swim the first plate containing 

 them was overflowed, drop by drop, by means of a minute stream of water 



