FISHERY BULLETIN; VOL. 76, NO. 3 



year progeny (Figure 5). Mortality coefficients for 

 1963 and 1965, the only years besides 1967 when 

 estimates of live eggs and alevins were made after 

 spawning, at hatching, and before emergence, are 

 from McNeil ( 1968); the equation and notation for 

 computing the instantaneous mortality 

 coefficients for the 1967 brood year follow McNeil 

 (1966). Time intervals assigned the three periods 

 were 1.4 mo (from potential egg deposition 

 through spawning), 2.0 mo (from spawning to 

 hatching), and 3.7 mo (from hatching to 

 emergence). The survival percentages by time 

 periods are given in Table 6. 



The impact of the heavy mortality between 

 hatching and emergence in 1967 brood year prog- 

 eny is evident in Figure 5. Monthly mortality 

 coefficients during spawning were also high in 



Figure 5. — Number of live pink salmon in Sashin Creek at the 

 beginning and end of three periods in freshwater for 1963, 1965, 

 and 1967 brood years. Numbers in parentheses show instan- 

 taneous monthly mortality coefficients from the egg through 

 alevin stages. Mortality during fry migration (April and May) 

 for the 1963 and 1967 brood years was negligible when measured 

 as the difference between streambed and weir estimates of total 

 fry production. The dotted extension of the 1965 brood year 

 assumes no mortality during this period. 



1963 and 1967 (0.76 and 0.71); the lower mortality 

 during spawning in 1965 (0.31) probably reflects 

 efficient spawning during the low streamflow con- 

 dition prevailing that year (McNeil 1968). Mortal- 

 ity from spawning to hatching was similar, but 

 mortality from hatching to emergence was strik- 

 ingly different in each of the 3 yr (Figure 5). 

 McNeil (1968) suggested the increase in over- 

 winter mortality of 1965 brood year progeny (0.26) 

 over 1963 brood year progeny (0.07) might have 

 been related to a delayed mortality from low con- 

 centrations of dissolved oxygen in early embryo 

 development during drought conditions in late 

 summer and early fall in 1965. The number of 

 spawners in 1967 was more than double the 

 number in 1963 and 1965 (Table 1). The over- 

 winter mortality of 1967 brood year progeny (0.42) 

 was considerably higher than the high mortality 

 of the 1965 brood year (0.26). The heavy over- 

 winter mortality experienced by the 1967 brood 

 year progeny may also have been caused by low 

 dissolved oxygen concentrations. However, be- 

 cause no drought conditions existed while the 

 progeny were in the gravel, these poor oxygen 

 conditions probably resulted from the high density 

 of eggs and alevins in the streambed. 



SUMMARY 



1. In 1967, 38,067 pink salmon spawned in 

 Sashin Creek on Baranof Island, Alaska. Fifty- 

 two percent of the spawners ( 19,639) were females; 

 mean fecundity was 2,260 eggs/female and the 

 potential number of eggs available for deposition 

 totaled 44.4 million. 



2. Entry of spawners into the stream was the 

 second earliest on record; based on the previously 

 consistent relation between time of entry and 

 freshwater survival, the production of fry should 

 have been greater than any previously recorded, 

 but the 3 million fry produced were less than half 

 the predicted number. 



3. Mean female densities on the spawning 

 grounds were 1.74/m2 in the upper area, 1.62/m2 

 in the middle area, and 1.17/m2 in the lower area. 

 Densities were higher in the upper area at the 

 beginning of spawning before significant levels of 

 spawning occurred in the middle or lower areas. 

 The tendency for spawners in the odd-year line to 

 utilize the upper area of Sashin Creek may be due 

 to genetic factors, including timing of escape- 

 ments, and possibly differential marine survival 

 favoring fry produced in the upper area. 



580 



