656 



Fishery Bulletin 94(4), 1996 



R 2 =N *(l-V)*exp fM2 + F ' 



(2) 



Smolt size and age 



Both equations contain N and thus can be simpli- 

 fied to the expression 



vp = 



* 



(0+1) 



where 



9= R l exp' 



■R. 2 exp 



(3) 



(4) 



This formulation assumes that all individuals ma- 

 ture as either 1SW or 2SW fish. This is reasonable 

 because returns of 3SW and older fish have been in- 

 significant during the study period (typically less 

 than 1% of the total runs). 



The 1SW:2SW ratio has frequently been used as 

 an index of 1SW maturation rate. However, sea-age 

 classes are often affected by different patterns of 

 marine mortality, particularly fishing mortality. 

 Thus, the 1SW:2SW ratio is a satisfactory measure 

 of maturation rate in only those instances where fish- 

 ing mortality is constant over time and in fact is lin- 

 early related to 1SW maturation rate under these 

 conditions. We know of considerable contrast in the 

 fishing mortality time series for North American 2SW 

 stocks, thus fmust be calculated. 



The size of smolts as a factor affecting return rate 

 and sea-age of maturation could not be directly evalu- 

 ated in this study because smolt sizes were not avail- 

 able. However, comparability of the smolt releases 

 in the two river systems and the transition from smolt 

 releases of predominantly two-year-old fish to re- 

 leases of yearling smolts raised concerns about the 

 role smolt size may have had on the observed pat- 

 tern of return rate and maturation. To address this 

 concern, for each scale the length (mm) of the fresh- 

 water zone (taken as the distance from the focus to 

 the end of the freshwater zone along the 360° axis of 

 the scale) was measured (Fig. 2). Back-calculated 

 lengths were not computed because of incomplete 

 data on length at recapture; therefore, freshwater 

 zone lengths were interpreted as an indication of 

 smolt size. The analysis of freshwater zone lengths 

 and other scale measurements were restricted to 

 2SW salmon owing to the extremely low returns of 

 1SW salmon to the Connecticut River (Table 1). 

 Cleaned and mounted scales were measured with an 

 Optimas image processing system. Sample sizes are 

 provided in Table 2. 



Statistical analysis of freshwater zone lengths was 

 complicated by the lack of a consistent pattern for 

 which stock had the greater zone-length mean. Fresh- 

 water zone lengths were compared among years and 

 between stocks with a two-way analysis of variance 



