476 



Fishery Bulletin 97(3), 1999 



with low precision owing to the small sample sizes 

 available. No significant differences existed between 

 the spacing means for circuli pairs 1-10 (Table 3). In 

 contrast, the data for 1983 showed that spacing pat- 

 terns and growth were different for early versus late 

 season collections. Fish collected in September had 

 wider circuli spacings for circuli pairs 4-10 than fish 

 collected in October (Fig. 4B). The September samples 

 had mean spacings generally above 0.0625 mm; 

 whereas, the October samples had mean spacings 

 generally below 0.06 mm. Most of the statistical com- 

 parisons between the poststratified groups were sig- 

 nificant at either P=0.05 or P=0.1 (Table 3). 



The seasonal pattern in circuli spacing for Gulf of 

 St. Lawrence postsmolts was similar to the patterns 

 observed for the some of the comparison stocks by 

 year, and dissimilar in other cases. For smolt year 

 1982, Gulf of St. Lawrence postsmolts and Penobscot 

 and Saint John returns showed similar seasonal pat- 

 terns of circuli spacing versus circuli pair (Fig. 5A). 



as 

 a. 

 CO 



07. 



0.06- 



0.05- 



0.04- 



0.07- 



0,06- 



0.05- 



0.04. 



-o— Aug 8-17 

 -A— Aug 18-Sept 5 

 -V— Sept 7-Oct 18 



-.— I 1 1 1 I I I — < I ' I < I ' I < I — ' — I — r— r 



B 



Sept 24-30 

 Oct 1-11 



T 1 1 1— I— 1 1 I < I r— I r— I 1— p-i 1— I— I r— f- 



2 4 6 8 10 12 14 16 18 20 22 



Circuli pair 



Figure 4 



Circuli spacing versus circuli pair for Gulf of St. Lawrence postsmolts 

 for two smolt year classes, 1982 (A) and 1983 (B I. Samples are poststratified 

 by date of capture. Error bars mark 95'7f confidence intervals. 



This comparison included both the ISW and 2SW 

 maturity groups within a stock. The pattern of mean 

 circuli spacing for Connecticut River 2SW returns 

 was different from that for the other stocks and 

 showed an ontogenetic trajectory of narrower spac- 

 ings with age. When compared statistically, mean 

 spacings for Gulf postsmolts were significantly higher 

 and lower than the mean for the three comparison 

 stocks (Table 4). The mean spacing for circuli pair 1 

 was significantly lower for Gulf postsmolts compared 

 with the other stocks, but the relationship between 

 stock circuli spacing changed for pairs 4-8 as the 

 rank order changed. The rank order changed again 

 by circuli pair 10. In 1983, the spacing pattern for 

 the postsmolts could not be distinguished graphically 

 from the combined signals for the other stocks (Fig. 

 5B). which is supported by statistical comparison as 

 well (Table 4). Again, the Connecticut River stock 

 had the narrowest spacings, whereas the widest spac- 

 ings were observed in Penobscot and Saint John fish. 

 In 1984, mean circuli spacing varied over 

 a wide range, suggesting the different 

 stocks experienced different growth re- 

 gimes that year (Fig. 5C). What is par- 

 ticularly striking is that the wild-origin 

 postsmolts from the Gulf had the narrow- 

 est mean spacings for most of the post- 

 smolt growth season. Mean spacings for 

 Gulf postsmolts were significantly lower 

 than those for the comparison stocks with 

 the exception of only one spacing pair 

 (Table 4). 



The three stocks provided for compari- 

 sons with Gulf of St. Lawrence postsmolts 

 displayed a range of return rates and 

 ISW fraction, reflecting differences in 

 survival and maturity rates. For the Con- 

 necticut stock, return rates of 2SW fish 

 and ISW fraction averaged 0.13% and 

 0.01, respectively (Table 5). Penobscot 

 River smolts returned at higher rates, 

 0.097^ and 0.477f for ISW and 2SW fish, 

 respectively. Typically 10*^ of the Penob- 

 scot cohort matured at ISW. The Saint 

 John stock had the highest return rates 

 and the largest proportion of the cohort 

 maturing after one-seawinter. The return 

 rate for ISW fish averaged lAT/r and the 

 ISW fraction averaged 0.59. The 2SW 

 return rate for the Saint John stock is 

 similar to that observed for the Penobscot 

 stock. Thus, the Connecticut stock had 

 the lowest survival rate and produced few 

 early maturing fish; the Saint John stock 

 had the highest survival rate and pro- 



