92 Puget Sound Marine Sta. Pub. Vol. 1, No. 11 



the per cent of those hatched in the light was 48.41 and of those 

 hatched in the dark only 33.95. By November 23, 82.62 per cent of those 

 in the light were hatched, while 69.23 per cent of those in the dark were 

 hatched. It was also observed that those in the light quickly wriggled out 

 of the shells and passed between the wires of the basket to the bottom of 

 the trough, while those in the dark were slow to separate themselves from 

 the shells. During the period of more active hatching many in the dark 

 were seen lying partly in and partly out of the shell, while in the light 

 very few were seen in this condition. Light hastened the hatching of the 

 embryos. 



Observations During Yolk-Sac Stage 



The fish that had been hatched in the light were removed December 

 15 to section 3 of trough B (Plate 17) which was painted white, and the 

 60-watt light placed directly over them within one foot of the surface of 

 the water. Shortly after this the fish in the light became markedly active 

 and swam to the upper part of the trough where the large majority of 

 them remained for the next 10 days, after which they distributed them- 

 selves fairly uniformly throughout the water of their compartment. This 

 free-swimming condition they kept from this time onward. 



Those in the dark remained quiescent for two weeks longer than those 

 in the light and were massed together at the lower end of the compart- 

 ment in which they were kept, being carried there by the current of the 

 water in the trough. At the end of the two weeks they became more 

 active and sought the upper end of their compartment. Within four days 

 from the time of the beginning of this activity they had distributed them- 

 selves quite uniformly throughout the water, but they did not exhibit the 

 same degree of activity as those that were kept in the light. 



About January 8 it appeared that those in the dark trough were a 

 little larger than those in the light trough. In order to verify this a num- 

 ber of weighings were made. In weighing the fish a beaker containing 

 about 50 cc. of water was accurately balanced on the scales, then the fish 

 were scooped up at random with a large sized coffee-strainer, in num- 

 bers varying between 45 and 66 at each sweep. The bottom of the strainer 

 was wiped comparatively dry with a cotton cloth, the flopping of the fish 

 causing most of the water to pass through the strainer. The fish were then 

 put into the beaker, which was again accurately balanced. As this 

 method would give rise to slight errors, a large number of fish were weighed 

 in several lots in order to reduce this inaccuracy to a minimum. Table 2 

 records the weights of five such lots of fish from each trough. The 

 difference in the average weight was 40 mg. in favor of those developed 

 in the dark. Thus up to this time those developed in the dark weighed 



