PACIFIC HERRING 
281 
year groups do notenter the commercial catch in any numbers, while the 4-year group 
is well represented. Because of this fact, the 3-year group, being close to the 4-year 
group and overlapped by it, does not show clearly in the length-frequency curves. 
BODY L EN&TM IN MILLIMETERS * 
Figure 28.— Length frequency of 463 young herring from Cape Elrington, Prince William Sound, August 28, 1925 
Otoliths and scales were collected from 50 herring from Elrington Passage, Prince 
William Sound, on July 6, 1927. Both the scales and the otoliths were legible for 
45 of these 50 fish. The results of the readings are tabulated in Table 27. 
Table 27 . — Relationship of age readings by means of scales and otoliths 
The two readings agree except in five cases. In two cases the scale readings 
showed the fish to be in the sixth year, whereas the otolith readings showed the fish 
in the fifth year. In three cases fish seven years by the scales were six years by the 
otoliths. There was no disagreement among the younger age groups. In each case 
of disagreement it was due to the otolith lacking one year. The otoliths are very 
small, and the outer rings are so close together in the older fish as to make it difficult 
to distinguish them by either reflected or transmitted light. 
The positive correlation between the readings of the scales and of the otoliths 
helps in confirming the results of our scale readings and shows that the annual rings 
on the scales are due to some general physiological change, probably associated with 
growth or spawning, that affects other hard parts as well as the scales. 
During the middle of April, at Halibut Cove, the last ring had not yet formed 
on many of the scales but was barely discernible on the majority. This is in accord- 
ance with Lea’s results with the spring spawning herring of Norway (1911). On 
following the growth of the herring in one locality throughout the year he found that 
