females, and the smaller males spawning with 

 the smaller females. 



5. Jack salmon constantly attempted to 

 infiltrate an established redd but were driven 

 away by both males and females, usually 

 by the males. As a wave of spawning waned, 

 jack salmon finally gained brief access to 

 redd territories. 



6. The male often took an active part in 

 digging the redd but dug more sporadically 

 than the female. 



males. The instant a fresh newcomer started 

 a digging movement, all spawned-out females 

 in the vicinity converged in a concerted attack. 

 Perhaps this behavior is a protective instinct 

 to prevent disturbing existing redds. 



9. At the peak of spawning and for a day 

 or two after, dead and dying spent males were 

 observed floating through the spawning area. 

 The floating carcasses were almost never 

 females. Spawning fish paid no attention to 

 these carcasses even when they entered a 

 defended redd area. 



7. Males chased mostly males and fe- 

 males mostly females. This phenomenon would 

 tend to make full use of a spawning stock 

 with an unbalanced sex ratio. For instance, if 

 there were a surplus of females, a male 

 would be able to spawn with many females 

 without interference, as demonstrated in con- 

 trolled experiments at Wood River Lakes by 

 Mathisen (1955). Of 134 females chased from 

 an established redd with a 1;1 sex ratio, only 

 16 were chased by males. 



8. As a spawning wave waned, the number 

 of lone females increased rapidly (0:1 sex 

 ratio), and males, which died first, disappeared 

 (fig. 9). Lone females on completed redds 

 were very intolerant of digging by new fe- 



An attempt was made to follow the behavior 

 of individual fish that had distinctive markings 

 or tags. Only four such fish (all females) 

 could be positively identified through a com- 

 plete spawning sequence. The chronology of 

 their actions during spawning is shown in 

 table 5. For these four fish, the spawning 

 period averaged slightly longer than the post- 

 spawning period, and the whole spawning proc- 

 ess, from redd selection to death, averaged 

 about a week (7.75 days). 



If this life span of about a week holds true 

 for the entire spawning population, each of 

 the three spawning waves may be the result 

 of progressive buildups of numbers of spawn- 

 ers. Recruitment over a period of several 



TABLE 5. — Observations of four spawning female sockeye salmon. Brooks River, 



September 11-25, 1957 



Identifying 

 mark 



Date of 

 first obser- 

 vation 



Time with 

 male 



Time 

 alone 



Date of 

 last obser- 

 vation 



Tag, green- 

 green 



Net scar across 

 top of head 



Tag, red- 

 white 



Tag, red- 

 yellow 



Sept. 11 



Sept. 12 



Sept. 22 



Sept. 25 



Days 



6 



Ar 



3 



5 



Days 



3 

 2 

 2 



6 



Sept. 19 

 Sept. 17 

 Sept. 26 

 Oct. 5 



Average 



4.5 



3.25 



16 



