10 U. S. BUREAU OF FISHERIES. 



The generalization that the high death rate in pike-perch eggs is 

 due to hick of impregnation tlius seems to be unwarranted. That the 

 present methods of preventing cohesion of the eggs are responsible 

 for a certain percentage of the mortality is probable, but they do not 

 account for all the loss. On the other hand, it has been shown that 

 about 25 to 30 per cent of representative samples of 29-hour eggs show 

 abnormalities that must lead to either malformation or death. If an 

 average loss is then considered as HO to GO per cent (and that is a fair 

 estimate), about half of this is due to the agency which manifet-ts 

 itself in abnormal development. This cause is in all probability to 

 be found in the i)ractice of retaining ca})tured fishes in pens for the 

 purpose of permitting eggs and sperms to mature. 



