114 



TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



not take any liberties with the figures that cannot be justified 

 by assuming the usual variability that occurs when one deals 

 with organisms of any species. The percentage of oil rises 

 from about 2% to a maximum of about 30% at the end of 

 August and then begins to decrease. At the same time the 

 percentage of water varies in a complementary fashion, falling 

 from nearly 80% to a minimum which appears to occur just 

 before the maximum of oil-contents occurs. Simultaneously 

 with these changes the percentage of " proteid " falls, decreasing 

 nearly steadily until the spawning occurs. 



5* 





1 











X 



■30 

 ■2S 



60 +■ 







^ 





■20 



20 

 15 



70 *■ 









-^V^£5__^ 



15 



10 

 5 



■60 





/& 





^UVotel^" I 







April 



May 



June 



July 



August September 



Fig. 1. 



No samples were obtained after 20th September, the 

 fishery being at an end by about that time and the herrings 

 having left the locality of Port Erin. There is no doubt that 

 had the shoals been sampled on their migration away from 

 Isle of Man it would have been observed that the percentage of 

 oil would have decreased to about 2%, that of water would have 

 increased to 80%, and that of " proteid " would also have 

 increased to its original percentage of about 19. That is, one 

 would be justified in completing the curves so as to represent 

 all the variabilities as cyclically repeated ones. 



The variation of " proteid " however is not so simple a 

 matter as the graph indicates. It is really nitrogen that is 



