OF THE NORTH SEA. 
No. 2] 
15 
place gradually and the intervening length-groups, between 12 and 
14 cm. might be supposed to inhabit the area of the Firth of Forth, 
having a depth between 25 and 50 m. A suggestion in this direction 
we find in sample 2, which has a length composition lying some¬ 
what intermediate between the two groups but tending very decidedly 
towards the latter. The cause of the segregation is very difficult to 
locate, it might be the hydrographical conditions, the search for sui¬ 
table food, the swimming capacity or, most probably, a combination 
of all these factors. The question might arise whether it is a segregation 
of the larger individuals. The same result might have been found if 
a lot of the individuals had by chance emigrated to the deeper water 
and under the different conditions grown more rapidly. The first-named 
explanation is, however, more in conformity with features displayed by 
segregation among the larger herrings, viz: the segregation of the larger 
individuals of the fat herring and their subsequent congregation with 
the spawning herring. 
The third part of the figure shows the curves representing the 
length frequency of the samples from Heligoland Bay to off Nissum 
Fiord (Samples 8—11). As seen they are fairly uniform and display 
marked modes by 13 cm. The average length composition of the 
four samples is shown by the continuously-drawn line. By com¬ 
paring these samples with those from the Firth of Forth we find that 
the length composition is intermediate between the two groups found 
at the latter place. As these samples have been taken at depths from 
18—30 m. they confirm the idea of a segregation of the larger indi¬ 
viduals of the group and a migration of these individuals to deeper 
water, in such a manner that the average length found from the 
samples becomes higher the farther out the samples are taken. 
All the above-mentioned samples are taken by trawl: the fourth 
part of the figure, however, represents two samples taken by nets. 
The number of individuals in these samples is very small, only 8 and 
6 respectively and they would have been of little interest had not, 
especially sample 12, displayed remarkably larger individuals of the 
I-group than any sample taken by trawl. As seen from table I the 
average length is found to be 16.7 cm., about 2 cm. higher than found 
in any trawl sample. The 6 individuals of the I-group are taken 
