20 U. § BUREAU OF FISHERIES. 
in September, while the late or winter run spawns from early 
November to well into January. <A peculiarity of this species is 
that the fish in most instances are in spawning condition when they 
leave salt water to enter the streams, and most of the eggs collected 
in the Washington field are obtained in tidewaters. The increased 
collections of chum-salmon eggs within the past two years may be 
attributed to the lessened activity of commercial fishermen, the low 
prices prevailing offering but little inducement to fish. 
There was a satisfactory run of silver salmon at the Hood Canal 
substations, though egg collections fell below those of the preceding 
year because the run occurred at a time when floods prevented the 
successful operation of traps or seines, a condition that continued 
to prevail throughout the spring spawning season of the steelhead 
salmon. 
QUINAULT (WASH.) SUBSTATION. 
[PuHiILto B. Haw.ery, Foreman in Charge. ] 
The fish-cultural work at this substation during the year was less 
successful than usual, the aggregate output amounting to only 
5,820,800 as compared with 15,908,600 in 1920. Eggs at this point 
are obtained from fish taken in traps or by other means ithe tribu- 
taries of Quinault Lake, which, like most streams in a mountainous 
country, react quickly to climatic conditions, a heavy rain producing 
flood conditions almost immediately. The waters rapidly subside 
on the cessation of rain, while even a short period of dry weather 
causes low water stages at all points. Fishing operations are con- 
ducted in the streams at some distance from the hatchery, the eggs 
being transported by canoe. Every season there are times when the 
water stages in the upper Quinault and Big Creek, the principal 
spawning streams, are so low that it is difficult to operate a canoe, 
while at other times the water is so high that all traps are submerged 
to a depth of 1 to 5 feet. 
As it is characteristic of salmon to ascend the streams in the 
greatest numbers during high-water stages, it is not surprising that 
the ege collections should be subject to material fluctuation from 
year to year. Thus, the number of eggs of any species taken is not 
necessarily a just criterion of the number of salmon available at 
spawning time. The remedy for this situation, from a fish-cultural 
standpoint, is the construction of a trap at the lake outlet similar to 
the one used at the Baker Lake station, whereby all the fish are taken 
as they enter and are held in an inclosure until they are ready to 
spawn. 
The first sockeye-salmon eggs were taken November 3, and _ be- 
tween that date and November 20 the entire egg collections for the 
season, amounting to 1,750,000, were obtained. From November 20 
to the time that efforts at spawn taking were abandoned, on Decem- 
ber 15, high-water stages precluded all possibility of seining or fish- 
ing by other methods. Local collections of this species were 
augmented by the receipt of 2,000,000 eggs from the Afognak sta- 
tion. The run of silver salmon appeared to be larger than usual, 
and during the spawning season of that species, which is coincident 
with that of the sockeye, 1,565,000 eggs were secured. A small num- 
ber of chinook-salmon eggs was also taken. The run of chinooks, 
which was apparently not large, was earlier than usual. 
