[5] RETARDING DEVELOPMENT OF SHAD EGGS. 791 



ing the feasibility of using the same water over and over again when 

 its temperature is kept below the normal condition. 



Our apparatus, Fig. 3, as arranged for these experiments, consisted of 

 two reservoirs, a hatching-cone, and a steam-pumi), and their connec- 

 tions. The first or supply reservoir (Fig. 3, a), was a small hogshead 

 elevated about three feet above the floor or just above the level of the 

 hatching-cone 6, which was of the ordinary pattern, with a rim of wire 

 sieving around the inside at the top with a gateway in it for the purpose 

 of letting dead eggs, pass oft" into the gutter, which rau around the top 

 of the cone outside of the sieve rim and so into the escape-pipe. The 

 cone was swung on braces attached to the side of the wall, and was con- 

 nected with the supply tank by means of a rubber pipe, c, passing from 

 the bottom of the reservoir to the bottom of the cone. The second res- 

 ervoir, d, was smaller than the first, and was placed under the floor and 

 below the hatching-cone with which it communicated by means of an 

 escape-pipe, e, passing from one side of the gutter at the top of the cone 

 down and over the edge of the reservoir can. 



The apparatus was completed by placing the two reservoirs in communi- 

 cation by means of a long tube, /, passing throujih the steam-pump g and 

 entering the top of the first reservoir. \^^ater being placed in the large 

 reservoir a flow would take place into the cone, and the cone when full, 

 would overflow into the second reservoir, from whence the once- used water 

 -could be pumped back by the pump into the supply-tank, again to run its 

 circuit through the cone. In this manner we had a constant flow of water 

 in our cone, and as the end of the escape-pipe from the cone into the 

 lower reservoir and that of the supply-pipe from the pump into the 

 supply-tank were considerably above the level of the water in their 

 respective reservoirs, there was also a slight amount of aeration from 

 the falling water. 



The temperature of the water was regulated by keeping ice in greater 

 «or less quantity in the supply-reservoir. Our first trial of this appara- 

 tus was begun on the evening of June 8. The supply-reservoir and 

 €one were filled with water from the bay and 50,000 freshly-taken eggs 

 placed in the cone. The temperature of the water was reduced to 45^, 

 then allowed to rise to 52°, and kept at that temperature until the 

 morning of the 11th, when the eggs were found to be dead and were 

 thrown away, and the apparatus thoroughly cleaned. It is probable 

 that the low temperature of the water had its effect in destroying this 

 lot. At 10 a. m. of the 12th the reservoir and cone were refilled and a 

 second lot of eggs placed in the cone. The temperature was 67=^, and 

 was gradually reduced to 56°, and kept for the most part at that point, al- 

 though on the 15th it rose to 04°, for lack of ice, consequent upon our 

 change of locality, but was brought back to 56^ on the IGth. These 

 eggs had been taken on the night of the 11th and kept until 10 a. m. 

 of the 12th in fresh running water, at 77°, before being placed in the 

 cone. When placed in the cone the segmentation cap entirely covered 

 the yelk, and the young fish, as was the case with lot No. 3 which was 



