FISHERY BULLETIN: VOL. 75, NO. 4 



an impoundment that extended to within 5 m of 

 the pond. The impounded area was heavily silted 

 by fall 1968 and I counted only four redds there. 

 The owner had reported that large numbers of 

 brook trout spawned in this area prior to weir 

 installation. In fall 1974, 1 yr after the weir had 

 been removed, I counted 34 redds and about half 

 the streambed was covered with silt. Since effects 

 of impoundment were still evident, this portion 

 of the outlet may have provided much more 

 spawning area than was evident in 1974. Possibly, 

 immigration was an important source of recruit- 

 ment prior to this study. 



Population declines at Maxwell Springs were 

 accompanied by changes in age structure. In April 

 1969, density of age 3 and older trout was nearly 

 1,000/ha and they totaled 233 kg/ha, or 63% of 

 population biomass (Table 3). By September 1972, 

 density of age 3 and older trout was 22/ha and 

 biomass was about 6 kg/ha, the lowest in the 4-yr 

 period. 



Mortality 



Numbers of fall fingerlings in Hoglot and Max- 

 well springs represented from 0.2 to 1% of the 

 estimated number of eggs deposited the previous 

 fall. I sampled 52 redds in five different ponds 

 to assess preemergence mortality. Numbers of 

 eggs per redd ranged from about 30 to 220. 

 Percentage of live embryos in individual redds 

 ranged from 76 to 99 (mean = 89%); stage of 

 development of these embryos varied from eyed 

 egg to alevin. Due to additional mortality to 

 emergence, I used 80% of potential egg deposition 

 to estimate numbers of emerging fry. Although 

 highest mortality rates in both ponds occurred 

 during years of highest egg production, egg 

 production and fingerling mortality were not 

 significantly correlated (Table 4). 



To estimate age-specific total mortality rates 

 of trout in Maxwell Springs, I assumed that 

 immigration was negligible. At Clubhouse and 

 Hoglot springs, where immigration was sub- 

 stantial, unmarked residents and immigrants 

 could not be separated; therefore, mortality rates 

 were calculated using only permanently marked 

 trout. Numbers of age 2 and older trout were 

 usually too small to allow estimation of mortality 

 rates. 



Mean rates of oversummer mortality in Max- 

 well and Hoglot springs increased with age (Table 

 5). Overwinter mortality rates at Maxwell 



TABLE 4. — Estimated egg production of brook trout populations 

 and densities of fall fingerlings. Egg deposition was estimated 

 from number of mature female trout in fall and the relationship 

 of fecundity (Y) and trout length in millimeters (X); Y = -588 + 

 6.14X. Instantaneous mortality rates (Z) were based on 80% of 

 egg production and were corrected for 182-day intervals. 



Springs also increased with age, except that age 

 trout had higher mean mortality rates than did 

 age 1 trout. However, within years there was 

 considerable variability between age of fish and 

 mortality rates. In all ponds mean mortality rates 

 oversummer exceeded overwinter rates. 



Immigration 



Estimation of immigration rates at Clubhouse 

 and Hoglot springs were based on mortality rates 

 calculated from relatively small numbers of 

 permanently marked trout and from mean, age- 

 specific mortality rates of trout from Maxwell 

 Springs (Table 5). Although accuracy of these 

 estimates is suspect, they should be useful in 

 illustrating seasonal differences in immigration 

 and in assessing the effect of immigration on 

 recruitment. 



At Clubhouse Springs most immigration oc- 

 curred overwinter and age trout made up 55% 

 of all migrants (Table 6). Largest migrations into 

 Hoglot Springs occurred between April and 

 September when age 1 trout accounted for 73% 

 of all migrants. In both populations periods of 

 peak immigration coincided with highest popula- 

 tion densities. Immigration was the only source 

 of recruitment at Clubhouse Springs; at any one 

 time more than half the population consisted of 

 fish that had immigrated within the previous 

 6 mo. At Hoglot Springs percentages of recent 

 immigrants ranged from 8.2 to 54.9 (mean = 

 34%). 



If estimates of trout migrating into Hoglot 

 Springs are reasonable, immigration accounted 

 for a major portion of total recruitment. The four 

 year classes produced in the pond from 1968 to 

 1971 amounted to 7,700 fall fingerlings/ha. About 

 3,800 of these fish survived to the following spring. 



756 



