MACKEREL INVESTIGATIONS IN 1897. 19 



Durino- the fourth and fifth days of development, most of the eggs of 

 sublot J)A hiy ou the cheesecloth at the bottom of the box and the 

 embryos gradually grew weaker until they finally succumbed. A very 

 few, about several hundred, hatched, but all died within 24 hours. The 

 eggs of DC were almost all destroyed by being injured or stranded on 

 the side of the floating box. Those of DB continued to develop beau- 

 tifully until the close of the fourth and beginning of the fifth day, up 

 to which time they had continued to float with only a very small per- 

 centage of loss. At this time they were nearly ready to hatch, having 

 developed at a higher temperature and more rapidly than DA, but 

 now they began to die rapidly and within a few hours the entire lot 

 succumbed. A heavy rain-storm occurred about this time, and the den- 

 sity of the supply of Salter water was found to have dropped to 1.024. 

 But I do not attribute the mortality to this, because many of the 

 embryos were observed to die while floating at the surface, to contract, 

 turn opaque, and sink just as they do when afl'ected by some deleteri- 

 ous substance. 



Eesults practically similar to these were obtained in all experiments 

 with higher densities, the encouraging feature being that, by this 

 means, a much larger proportion usually remained alive until late 

 stages of. development than when low-density waters were employed. 

 In no case was a greater number hatched, and in no case did the fry 

 thus reared appear to be more vigorous. 



Another rather interesting result was obtained by taking eggs which 

 had sunken to the bottom of the usual apparatus and placing them in 

 a circulating current in the McDonald shad jar, the outflow being pro- 

 tected by cheesecloth cage. Under such circumstances the numbers 

 hatched were always larger than in samples of the same eggs which 

 were left undisturbed, and fry already hatched could be kept alive and 

 vigorous for a much longer time. Some were thus kept for 5 days, 

 when the yolk-sac was absorbed, and they were ready to begin feeding. 

 Half of one lot (lot G) of over 400,000 eggs, all of which sunk at once 

 after fertilization, were thus treated. The remaining half being placed 

 in a McDonald cod-box, every one of the eggs which had lain at the 

 bottom of the box was dead before 36 hours had passed. The eggs 

 placed in the closed current jar continued to develop up to the last day 

 and some of them hatched. Probably the proportion hatched would 

 have been much greater could the eggs have been prevented from 

 piling up on the outlet screen and injuring one another by the pressure. 

 Numerous other experiments were carried out, but none were more 

 conclusive or satisfactory than those just indicated. It was found 

 impossible, with the limited facilities, to be sure that only one condition 

 had been varied, and that some other factor had not arisen to vitiate 

 the results. In view of the fact that in no case was a greater per cent 

 than 50 hatched, and that only once attained, it seems useless to 

 describe the results with further detail. It seems proper to add that 

 Mr. Corliss, superintendent of Gloucester station, advises me that he 



