296 
BULLETIN OF THE BUREAU OF FISHERIES 
Table 31 . — Length frequencies of summer herring — Continued 
McClure Bay 
Dogfish Bay 
Shuyak Strait 
Kachemak Bay 
Russian 
Harbor 
Length in millimeters 
1927 
1925 
1925 
1926 
1926 
1927 
1925 
Ac- 
Per 
Ac- 
Per 
Ac- 
Per 
Ac- 
Per 
Ac- 
Per 
Ac- 
Per 
Ac- 
Per 
tual 
cent 
tual 
cent 
tual 
cent 
tual 
cent 
tual 
cent 
tual 
cent 
tual 
cent 
245-249 
26 
13.0 
4 
4.0 
6 
3.3 
3 
2.8 
17 
2.5 
15 
4.3 
1 
1.0 
250-254 
16 
8.0 
12 
12.0 
15 
8.3 
6 
5.7 
29 
4.3 
11 
3. 1 
3 
2.9 
255-259 
10 
5.0 
9 
9.0 
15 
8.3 
10 
9.3 
38 
5.6 
14 
4.0 
7 
6.6 
250-264 _ ___ 
3 
1.5 
28 
28.0 
18 
10.0 
9 
8.4 
46 
6.8 
19 
5.5 
14 
13.3 
265-269 
3 
1.5 
19 
19.0 
28 
15.6 
13 
12.1 
79 
11.6 
20 
5.7 
12 
11.4 
270-274 
2 
1.0 
18 
18.0 
29 
16.1 
19 
17.7 
104 
15.3 
25 
7. 1 
14 
13.3 
275-279 
1 
.5 
2 
2.0 
23 
12.8 
12 
11.2 
125 
18.4 
21 
6.0 
11 
10.5 
280-284 _ 
1 
1.0 
14 
7.8 
9 
8.4 
91 
13.4 
20 
5.5 
12 
11.4 
285-289 
1 
.5 
1 
1.0 
4 
2.2 
3 
2.8 
43 
6.3 
16 
4.1 
7 
6.6 
290-294 
8 
4.4 
12 
11.2 
22 
3.2 
5 
1.4 
16 
15.2 
295 299 
3 
1.7 
1 
.9 
10 
1.5 
2 
.6 
3 
2.9 
300-304 
1 
.9 
5 
.7 
2 
.6 
3 
2.9 
30b 309 

2 
.3 
2 
1.9 
310-314 
1 
2 
.3 
Total frequency... 
200 
100 
180 
107 
679 
350 
105 
6 
1 
4 
3 
6 
11 
1 
The lengths were first grouped in half- 
centimeter categories and then smoothed 
twice by three’s to remove minor modes 
due to chance sampling. (Fig. 35.) These 
smoothed length distributions reveal a shift- 
ing of sizes from year to year, the mode 
shifting from 20 centimeters in 1925 to 
just over 22 centimeters in 1926. This 
same mode appears to have shifted to 
23.5 centimeters in 1927. The bimodal 
distribution of 1924 suggests that the single 
mode of the later years is composed of 
at least two age groups. The mode at 19 
centimeters in 1927 appears at 17 centi- 
meters in 1926. This group is numerous 
enough in 1927 to cause the mode at 23.5 
centimeters to appear relatively less im- 
portant than in the previous year. Some 
of the modes are no doubt partly con- 
cealed, others distorted or exaggerated by 
various factors, but, nevertheless, these 
factors are not sufficient to seriously ob- 
scure the presence of a dominant size group 
and its progression through the commer- 
cial catch. 
Although the data in this form show 
Figure 35. — Length frequency distributions for the sum- Well the dominance and progression of Cer- 
mer herring of Elrington Passage and immediate vicinity tain size groups, yet they do not show the 
from 1924 to 1927 i , • i , j. , ■ 
relative lack of certain size groups, not 
only as compared to the other sizes for the same year but as compared to the 
average of the same sizes over the entire period of four years. The average curve 
for these four years was obtained by summing the weighted frequencies (percentage 
