FISHERY BULLETIN: VOL. 76, NO. 3 



streambed for as long as 18 mo ( McNeil et al. 1964). 

 Once hatching is completed, no new dead eggs can 

 be added to the streambed. Because hatching of 

 live eggs was well underway on 1 December ( about 

 35% completed), many of the dead eggs present in 

 March had already died by 1 December (Table 10). 



Most of the eggs and alevins that disappeared 

 over the winter (about 11 million; Tables 7, 8, 9) 

 were individuals that had been alive on 1 De- 

 cember because the number of dead eggs and ale- 

 vins was essentially unchanged from December 

 (4.6 million) to March (4.2 million) (Table 9). Mor- 

 tality (in the form of disappearance) of live eggs 

 and alevins in the streambed between 1 December 

 and 25 March was 74% in the upper area, 77% in 

 the middle area, and 85% in the lower area (Table 

 11). Of the 11.1 million pink salmon eggs and 

 alevins that disappeared within the streambed be- 

 tween 1 December and 25 March, 10.7 million 

 were alive on 1 December. 



As previously mentioned, the cause of the dis- 

 appearance of dead eggs and alevins in the 

 streambed may differ from the cause of their 

 deaths. This apparently occurred with the 1967 

 brood year pink salmon progeny in Sashin Creek, 

 and I offer the following theoretical sequence to 

 explain the major overwinter disappearance of 

 eggs and alevins. 



The greatest number of fry produced in Sashin 

 Creek since 1940 was 5.9 million (Table 1). On 1 

 December 1967, 13.7 million live pink salmon 

 eggs and alevins were in the Sashin Creek 

 streambed (Table 9), a number that appears to 



Table 10. — Estimated densities of dead pink salmon eggs and 

 alevins in three areas of Sashin Creek on 1 October 1967, 1 

 December 1967, and 25 March 1968. 



Table ll. — Estimated densities of live pink salmon eggs and 

 alevins in three areas of Sashin Creek on 1 December 1967 and 

 25 March 1968 and disappearance of live eggs or alevms between 

 the two dates. 



Area 



Upper 



Middle 

 Lovi/er 



Live eggs and 

 alevins per square meter 



1 Dec. 1967 25 Mar 1968 



Alevins Eggs Alevins 



Eggs 



899 



769 

 463 



Percentage of live 



eggs or alevins that 



disappeared between 



dates 



exceed the capacity of the streambed for pink 

 salmon fry production. I postulate that the high 

 initial density of eggs led to a severe mortality of 

 embryos in the early alevin stage, probably be- 

 cause of widespread oxygen privation or a combi- 

 nation of oxygen privation and a buildup of toxic 

 metabolites. The rate of oxygen consumption by 

 embryos increases steadily with development 

 (Wickett 1954, 1962) and coincides with the gen- 

 eral lowering of streamflows during the late fall, 

 followed by stabilization of streamflows at near 

 the normal winter levels.^ This combination of 

 conditions permitted the embryo population to 

 survive up to, but not much beyond, the hatching 

 period. These recently hatched dead alevins then 

 apparently disappeared rapidly within the 

 streambed through the combined action of 

 biochemical decomposition and intragravel inver- 

 tebrate scavenging. As I will show later, the rapid 

 disappearance of recently hatched dead alevins in 

 the streambed seems consistent with this 

 hypothesis. 



Although no intragravel water quality data are 

 available from Sashin Creek during or shortly 

 after hatching to support the above theory, a com- 

 parison of the rates of oxygen consumption by pink 

 salmon embryos of various ages indicates that 

 oxygen requirements do steadily increase during 

 the hatching period. The rates of oxygen consump- 

 tion reported for early stage eggs (7-26 days old) 

 have ranged from 0.0003 mg O2 /egg per h ( Wickett 

 1954) to 0.0005 mg 02/egg per h (Brickell 1971). 

 Brickell found that the rate of oxygen consump- 

 tion by 35-day-old pink salmon eggs was 0.0018 mg 

 Oj/egg per h, almost four times the rate he mea- 

 sured for 7-day-old eggs. Faintly eyed 38-day-old 

 eggs had an oxygen consumption rate of 0.002 mg 

 Og/egg per h (Wickett 1962) while 7-day-old ale- 

 vins had a consumption rate of 0.01 mg 02/alevin 

 per h (Wickett 1954). x 



575 



345 



249 



382 



260 

 108 



74 



77 

 85 



""Seasonally, stream discharge in Sashin Creek is usually 

 highest in fall and lowest in summer. Discharge in winter 

 months may also be low, but is normally above summer levels. 

 Because unseasonably low winter discharge could reduce oxygen 

 delivery to embryos below the normal seasonal pattern, I com- 

 pared the low monthly discharge during December, January, 

 February, and March for 1967-68 with low discharge patterns in 

 the same months for the period 1951-52 to 1966-67. The low 

 mean monthly discharge from Sashin Creek during December, 

 January, February, and March ranged from 18 to 62 ft^/s and 

 averaged 33 ft^/s for the 16-yr period. The mean minimum 

 monthly discharge during these same 4 mo in 1967-68 was 30 

 ft^/s (U.S. Geological Survey 1969), suggesting that low 

 streamflow levels during these months in 1967-68 were near 

 normal. 



576 



