FISHERY BULLETIN: VOL. 75, NO. 4 



TABLE 11. — Total annual production (P), mean biomass (B), 

 and PIB ratios for all age 1 and older brook trout. 



of individual age-groups during winter 1969-70 

 were well below average and age 1 trout lost 

 weight. This was the only period in which an 

 age-group in Clubhouse Springs had a negative 

 growth rate, and it was probably due to immigra- 

 tion of yearling trout smaller than pond residents. 

 Overwinter production in 1969-70 was 2 kg/ha; 

 production during other winter periods ranged 

 fromlO to 21 kg/ha. 



PIB ratios for age 1 and older trout in Hoglot 

 and Maxwell springs tended to decline with 

 increasing biomass (Table 1 1 ), i.e., mean weighted 

 growth rates were inversely related to density. 

 As I have noted, age-specific, instantaneous 

 growth rates (G) were the only growth parameters 

 poorly correlated with density. Biased estimates 

 of G for individual year classes could have 

 obscured relationships with population density, 

 but did not markedly affect mean weighted 

 growth rates when all adult trout were combined. 



DISCUSSION 



Estimation of trout production in this study 

 required several assumptions and the data should 

 be interpreted accordingly. Major assumptions 

 were: 1) numbers of emergent fry were 80% of 

 total egg production, 2) growth and mortality 

 rates of age trout were constant from emergence 

 to fall, and 3) production could be estimated from 

 the product of G and B when immigration 

 occurred. 



Chapman (1967) suggested that production of 

 brown trout fry in Horokiwi Stream (Allen 1951) 

 could have been overestimated by fourfold due to 

 errors in estimating egg deposition and emer- 

 gence. 1 used fecundity data from two populations 

 of wild brook trout that were collected from ponds 

 in the same watershed as the study ponds. 



Fecundity differences among populations were 

 probably not large since growth rates of the trout 

 were similar. I assumed that all eggs were 

 spawned because egg retention was insignificant 

 in other stream populations of wild brook trout 

 (Wydoski and Cooper 1966). In addition, I as- 

 sumed emergent fry represented 80% of total egg 

 production. Percentage of live embryos in indi- 

 vidual redds exceeded 80% in my study. Brasch 

 ( 1949) studied brook trout reproduction in several 

 ponds; he found survival from egg to emergence 

 was 79%. In laboratory experiments, emergence 

 of brook trout fry exceeded 80% when the 

 substrate was composed of 5% or less sand and 

 concentrations of dissolved oxygen exceeded 

 7 ppm (Hausle 1973). Therefore, I do not believe 

 estimates of egg production or emergent fry 

 seriously biased production estimates. 



The assumption of constant mortality rates 

 from emergence to fall represents potentially 

 large errors in production estimates for age 

 trout. Hunt (1966) found that instantaneous 

 mortality rates from emergence in February to 

 June were about 10 times greater than mortality 

 from June to September; he based mortality rates 

 on 90% emergence of fry. To assess the influence 

 of variable mortality rates, I calculated produc- 

 tion for the 1970 year class at Maxwell Springs 

 from emergence to October with different mortal- 

 ity schedules. If mortality were five times greater 

 during the first half of the interval than during 

 the second, production would have been 63 kg/ha, 

 and if mortality rates varied by tenfold, produc- 

 tion would have been 60 kg/ha. With a constant 

 mortality rate from emergence to October, esti- 

 mated production was 109 kg/ha. Thus, if there 

 was an initial high mortality of fry, production 

 of age trout could have been overestimated by 

 50 to 60%, and annual production by all age- 

 groups would have been overestimated by 19%. 



Assumptions that instantaneous growth rates 

 were constant from emergence to fall certainly 

 oversimplify growth history of fingerlings, but 

 overall effects of this assumption on production 

 estimates did not appear significant. Hunt (1966) 

 found large variations in monthly growth rates 

 of brook trout from emergence to October; growth 

 rates increased to a maximum in May and then 

 declined the rest of the year. Average monthly 

 growth rates from February through April were 

 not different than those from May to October 

 U-test P>0.05). These periods correspond to 

 periods for which I calculated production by age 



762 



