FISHERY BULLETIN: VOL. 83, NO. 1 



in Figure 2 differed significantly from zero (t-test, 

 P < 0.001). However, there was no significant rela- 

 tionship between otolith nucleus length and fe- 

 male fork length {t-test, P > 0.05, Fig. 3), or egg 

 dry weight {t-test, P > 0.10). We also investigated 

 the utility of otolith nucleus lengths as a racial 

 characteristic by calculating D^, a part of a dis- 

 criminant function analysis. In this instance, D^ 

 is a measure of the power of discrimination of 

 nucleus length in separating juvenile sea-run and 

 freshwater S. gairdneri. D^ was 0.063 and was 

 not significant {P > 0.1). 



A major source of the variability in the otolith 

 nucleus length-female parent length relationship 

 (Fig. 3) was apparently related to the ontogeny of 

 otolith nuclei in the salmonid embryos. Otolith 

 nuclei result from the fusion of primordia. Pri- 

 mordia, the first calcified structures to arise in S. 

 gairdneri during embryonic development, ap- 

 peared at 115-214 Centigrade degree-days. Indi- 

 vidual primordia increase in size by concentric 

 accretions, ultimately fusing with neighboring 



0.24 n 



0.22 ^ 



^ 0.20 - 



Z 

 UJ 



o 

 o 



0.18- 



O 



X 



I— 

 o 



z 



UJ 



_J 

 Ul 



o 



§ 



0.16- 



0.14- 



0.12 



0.10- 



0.08- 



0.06 



I I r -1 1 1 1 



300 400 500 600 700 800 900 



FEMALE PARENT FORK LENGTH (mm) 



Figure 3. — Scatter plot of Salmo gairdneri female parent size 

 on otolith nucleus length of progeny. The origin of the adults is 

 given in the caption of Figure 2. 



84 



primordia to form the nucleus of the otolith at 

 226-241 degree-days (Fig. 4). Hatching occurred at 

 about 320 degree-days. The pattern of nucleus de- 

 velopment was similar in both rainbow and 

 steelhead trout. Although we did not follow otolith 

 development in O. tshawytscha, examination of 

 their nuclei suggested that they also arose from 

 fusion of multiple primordia. Deposition of growth 

 increments commenced immediately after fusion. 



The number of primordia fusing to form the 

 otolith nucleus in the salmonid species we 

 examined was variable, even within the progeny 

 of a single female. In rainbow trout, there was an 

 average of 8. 2 ± 2.7 primordia (±1 standard devia- 

 tion indicated). In steelhead trout and O. 

 tshawytscha numbers of primordia averaged 10.7 

 ± 2.4 and 10.1 ± 2.7, respectively. There were no 

 significant differences in mean primordia counts 

 among the three stocks of rainbow trout or the two 

 stocks of steelhead trout examined (analysts of 

 variance, P > 0.05). Figure 5 shows the relation- 

 ship between the number of primordia deposited 

 and otolith nucleus length. 



The variable location of primordia within the 

 nucleus also affects nuclear dimensions and 

 further increases variability. In some instances 

 (<5%), primordia were formed at the periphery of 

 the nucleus, resulting in a local distortion of 

 otherwise regular growth increments (Fig. 6). 



Otolith nucleus length (mm) ±1 SE in S. 

 gairdneri from the Deadman River was also af- 

 fected by incubation temperature as shown below: 



Water temperature 



6.5°C 



9.5°C 



15.0°C 



Mean nucleus length (mm) 



Rainbow 



trout 0.142 ± 0.009 0.174 ± 0.009 0.172 ± 0.008 

 Steelhead 



trout 0.154 ± 0.004 0.197 ± 0.008 0.191 ± 0.005 



One-way analysis of variance and the Student- 

 Newman-Keuls test indicated that the mean 

 otolith nucleus length in rainbow or steelhead 

 trout reared at 6.5°C was significantly less (P < 

 0.01) than at 9.5° or 15.0°C, although no significant 

 differences in otolith nucleus length (P > 0.05) 

 existed in fish reared at the two higher tempera- 

 tures. The number of primordia formed in both 

 Deadman River steelhead and rainbow trout was 

 independent of the water temperature at which 

 the eggs and alevins were incubated (analysis of 

 variance, P > 0.05). 



