206 
BULLETIN OF THE BUREAU OF FISHERIES 
years was probably due to the effect of new regulations restricting operations rather 
than to a material decline in the abundance of salmon. These regulations prohibited 
all commercial fishing within 1,000 yards of the mouth of Eshamy Creek and required 
a distance interval of 200 yards between all set nets in the bay and lagoon. The 
regulations effective in 1925 and continued without modification in 1926 and 1927 
completely closed the lagoon to commercial fishing for salmon and prohibited fishing 
operations within 1,000 yards of the mouth of Rubberboot Creek. It seems reason- 
able, then, to attribute the reduced catch at Eshamy from 1924 to 1927 to the enforce- 
ment of these regulations, and consequently the catches in these four years are not 
comparable with those of preceding years. 
Eshamy Creek is the outlet of a small lake and gained prominence among the local 
fisheries in the early days of the packing industry by reason of the red salmon it 
produced. It was the stream most preferred by the fishermen from the canneries 
at Odiak and later by the cannery at Orca. A saltery was once operated within a few 
yards of its mouth near the head of the lagoon. Fishing was easy and often destruc- 
tive; competition was keen between fishing crews; and the law was frequently disre- 
garded, as court records at Valdez and Cordova show. Red salmon alone were wanted. 
No effort was made to take pinks until 1912, at least none was reported from Eshamy 
before that year. The runs of cohos, chums, and kings, as indicated by the catch, are 
unimportant. A few hundred cohos were taken in each year since 1912 except 
two, 1916 and 1926. Chums were caught in slightly larger numbers without inter- 
ruption from 1916 to 1927. A few kings were taken in most years, but the largest 
catch in any one season was only 64. No analysis of the data for these species of 
minor importance seems advisable at this time. 
A graphic picture of the catch of red salmon at Eshamy from 1904 to 1927 is 
shown in Figure 4, and the trend of the catch, calculated on a 5-year moving average, 
is also shown for the period up to 1923 inclusive. The trend was not figured beyond 
that year on account of the influence of the new laws and regulations, mentioned 
above, on the catch in 1924 and subsequent years. In general, it is seen that good 
catches were secured up to 1910. Then followed a period of six years in which the 
catches were fight and the trend correspondingly lower. Beginning in 1917 there 
was another period marked by large catches — nearly, but not quite, the equal of 
those immediately preceding 1910. This lasted until 1924, when the new regulations 
became effective and since which time the catch has been held at a very low level. 
For a small stream which has been intensively fished and which has supported a 
relatively small run, probably never over a few hundred thousand red salmon, the 
Eshamy run has held up well and apparently shows no serious depletion. 
The percentage deviation of the catch from the trend is given in Figure 5. (See 
Pt. I, pp. 61-63 for an explanation of the use here made of the deviation from the 
trend and the correlations in the deviations at intervals.) The fluctuations in these 
deviations are distinct and regular, indicating definite cycles in the catches (and 
presumably therefore in abundance) at regular intervals. These cycles appear 
consistently through both good and poor years and are as well marked during the 
relatively unproductive years from 1910 to 1916 as during the productive periods 
that preceded and followed. Coefficients of correlation at 4 and 5 year intervals are 
as follows: Four-year interval r— 0.69 ±0.102; 5-year interval r = 0.76 ±0.085. Both 
of these correlations are high and sufficiently greater than their respective probable 
errors so that their significance is undoubted. Coefficients were not calculated for 
