DUTIL AND COUTU: EARLY LIFE OF ATLANTIC SALMON 



not significant indicating that, in females as in 

 males, changes in size of the gonads were not 

 allometric (body length: P = 0.06; body weight: 

 P = 0.73; time: P - 0.68). 



Scale Marks 



Scales used for age determination were also ex- 

 amined for any mark that could be of use in stock 

 identification. Many individuals exhibited a tran- 

 sition zone on their scales. Circuli in this zone 

 were more wide-spaced than circuli laid down in 

 earlier summers, but they were more narrow- 

 spaced than circuli formed in summer in a marine 

 environment. This mark was present on 73% of 

 the scales in both males and females (72.6 and 

 73.87f) but tended to decrease as smolt age in- 

 creased: 84% (age 2 years), 75% (3 years), 68% (4 

 years), and 62% (5 years). However, the trend was 

 not significant (x^ = 5.74, P > 0.10). 



Scales also exhibited summer checks in 1982, 

 1983, and 1984. The number of circuli between 

 the smolt mark and the summer check was 

 10.9 ± 2.6 (mean ± SD) (range 7-15) in 1982 and 

 9.7 ± 2.2 (range 5-15) in 1983. The ratio between 

 the radius to summer check and the radius to 

 smolt mark averaged 2.00 ± 0.30 (range 1.49- 

 2.45) in 1982, and 1.67 ± 0.19 (range 1.26-2.32) 

 in 1983. The overall percentage of occurrence was 

 low in 1982 (3.6%) and involved only specimens 

 collected in the area Bale Trinite-Pointe aux 

 Anglais where 11 out of 13 postsmolts examined 

 showed a summer check. The incidence of sum- 

 mer checks on scales increased markedly in 1983 

 (55%) and 1984 (75%), and summer checks were 

 no longer restricted in distribution. Examination 

 of data concerning postsmolts possessing a sum- 



mer check showed no relationship between the 

 check and measured biological variables except 

 perhaps in 1982. The gonads of males having a 

 summer check on their scales in = 5) were heav- 

 ier than in males having no summer check (120 

 mg and 48 mg respectively). Their gonadosomatic 

 index was also higher (GSI = 0.042% and 0.025% 

 respectively). This was not so in females. Infor- 

 mation on gonads is not available for 1983, 1984, 

 and 1985. 



Back-calculated lengths at summer check for- 

 mation were larger in 1982: 22 ± 2.5 cm in 1982 

 and 20 ± 2 cm in 1983. Postsmolts measured 265 

 mm on 1 September. With a mean rate of increase 

 in length of 1.65 mm/day, the check must have 

 been formed some 26 (1982) to 40 (1983) days 

 earlier, i.e., in late-July. This is confirmed by ex- 

 trapolation of the length-date plot (Fig. 2). No 

 salmon were noted possessing 2 summer checks 

 on their scales. 



Food Items 



Drastic changes took place in prey selection be- 

 tween postsmolts collected in summer and in au- 

 tumn. The stomach contents of 40 salmon ranging 

 up to 188 mm in length (70 g) collected in the 

 second half of July 1984 in the estuary of the 

 Grande Trinite River, near Bale Trinite, revealed 

 a low diversity in prey items, indicative of a tran- 

 sition period between the riverine and marine en- 

 vironment. Small fish remains were observed in 

 only 5 out of 39 stomachs containing food re- 

 mains. Invertebrates were observed in 38 stom- 

 achs, classes Insecta and Crustacea occurring in 

 87 and 92% of the stomachs respectively. Food 

 items consisted mainly of 2 families: Chironomi- 



Table 4. - Functional regressions of gonad weight on date, body length, and 

 body weight of Atlantic salmon female postsmolts (n =213). Lengths and 

 weights are log-transformed. 



Significance 



Functional regression 



Date 



Length 

 (mm) 



Weight (g) 



2: P = 0.0001 



3 : P = 0.0001 



4 : P = 0.0058 



5 : P = 0.26 n.s. 



covariance : P 



2 

 3 

 4 

 5 



2 

 3 

 4 



5 



P < 0.0001 

 P < 0.0001 

 P < 0.0001 

 P = 0.34 n.s. 

 covariance 



P < 0.0001 

 P < 0.0001 

 P < 0.0001 

 P = 0.28 n.s. 

 covariance 



Y = 0.0075 X 



Y = 0.0044 X 



Y = 0.0052 X 



2.7127 

 1.9052 

 2.0229 



0.04 



y = 3.8015 X- 10.1089 

 y = 2.9105 X - 7.8659 

 / = 3.1893 X -8.4701 



P = 0.02 



Y= 1.31 10 X- 3.9428 

 Y = 1 .0385 X - 3.2227 

 Y= 1.1763 X- 3.4659 



P = 0.04 



203 



