of the areas. The upper area had the highest num- 

 ber per square met-er (228), the middle area the 

 next highest (197), and the lower area the lowest 



(104). 



RELATION OF SURVIVAL OF EGGS AND 

 ALEVINS TO TIME OF SPAWNING OF 

 ADULTS 



The observed relation of fresh- water sui-vival to 

 the date by which 50 percent of the spawners en- 

 tered Sashin Creek is shown in figure 2. The regres- 



20 

 AUGUST 



^ET 



SEPTEMBER 



FiGiRE 2. — Relation of fresh-water .-iurvival of pink sal- 

 mon in Sashin Creek to (late by which 00 percent of 

 spawners entered the stream. The curve Y=26.66-0. 722 X 

 was fitted by least squares, X=0 corresponds to August 

 10. Relation of the 1965 brood year to the fitted regres- 

 sion is shown. 



sion line in the figure is fitted to points for 20 brood 

 years in the period 1940-64. Four years ( 1952, 1954, 

 1960, and 1962) were excluded from the regression 

 because potential egg deposition was very meager 

 (12,000 eggs or less) and one year (1964) because 

 the adults were transplanted to the creek. 



Fresh-water survi\'al has exceeded that of the 

 1965 brood year in only three other brood years — 

 1957, 1961, and 1963. In each year 50 percent of the 

 spawners had entered the stream on or before Au- 

 gust 26, the "index date" of stream entry for 1965. 

 The date was August 22 in 1957 (22-perc«nt sur- 

 vival) ; August 13 in 1961 (20-percent sur^-ival) ; 

 and August 26 in 1963 (20-percent survival). 



The predicted fresh-water' survival of tlie 1065 

 brcKxl year based on the regression line in figure 2 

 was 15 percent. The estimated survival based on the 

 niunber of alevins collected with a hydraulic sam- 

 pler (McNeil, 1964a) before the fry emerged from 

 the streambed was 18 percent. An estimated 2,235,- 

 ono fry were protluced. Although I could not com- 

 pare this figure with a count of fry at the weir in 

 the spring of 1966, sur\ival estimated with the 

 hydraulic sampler has agreed closely with survival 

 estimated at the weir (table 9). 



It is not entirely clear why fresh-water survival 

 of pink salmon should be high in Sashin Creek 

 when spawners enter early. Merrell (1962 ) hypotli- 

 esized that embryos from eggs deposited late in 

 the spawning season fail to develop sufficiently be- 

 fore the onset of cold weather and that embiyos of 

 retarded development are more sensitive to adverse 

 effects of cold water than are more developed em- 

 bryos. High mortality of pink salmon in streams 

 tributary to the White and Barents Seas in I960 

 was attributed to late spawning (Azbelev, Surkov, 

 and Yakovenko, 1962). 



Laboratory' experiments indicate that exposure 

 of salmon eggs to cold water soon after fertilization 

 is detrimental, and this etlect may partly explain 

 wliy late spawning is less successful than early 

 spa.wning. Eggs of sockeye salmon {0. nerka) hehl 

 at an initial incubation temperature of 7° C. expe- 

 rienced higher mortality than eggs held at 10°, 13°, 

 and 16° C. (Andrew and Geen, 1960), and early 

 exposure of salmon eggs to temperatures of 1° and 

 2° C. can be lethal (Seymour, 1956; Combs and 

 Burrows, 1957; Efimov, 1962). If incubation tem- 

 peratures are high enough initially, subsequent re- 

 duction to near freezing is not always harmful, Init 

 the duration of initial exposure to warm water is 

 important. Seymour (1956) incubated eggs of 

 chinook salmon (O. tshiiwytucha) 2 and 31/2 weeks 

 in warm water before exposing them to cold water. 

 He found that eggs reared 2 weeks in warm water 

 had a high mortality wlien exposed to cold water, 

 whereas eggs reared 3i/^ weeks in warm water had 

 a low mortality. Additional studies would be re- 

 quired to determine the effect of water temperature 

 on the survival of pink salmon eggs in Sashin 

 Creek and whetlier warm water during spawning 

 is generally necessary for high survival. 



PINK SALMON SPAWNERS IN SASHIN CREEK 



581 



