dent rainbow trout spawn from May to mid-July 

 (Fessler 1972). Mean water temperature during 

 the period of steelhead egg incubation was 8.4°C 

 for 1967-69 and 7.6°C for 1982-83. Mean water 

 temperature during the period when resident 

 rainbow trout eggs were incubating in the main 

 stem of the river was 12.6°C in 1967-69 and 

 11.9°C in 1982-83. Incubation temperature for 

 steelhead at Round Butte Hatchery was 10°C and 

 did not vary. 



The dimensions of otolith nuclei from resident 

 rainbow trout and steelhead in our study were 

 indistinguishable from those in fish from British 

 Columbia. No significant difference (a = 0.05) in 

 mean length of otolith nuclei existed between the 

 British Columbia steelhead incubated at 9.5° or 

 15°C and suspected wild steelhead from Bakeoven 

 Creek or Round Butte Hatchery steelhead incu- 

 bated at 10°C. Among resident rainbow trout, the 

 mean length of otolith nuclei for fish from the 

 Deschutes River was also not significantly differ- 

 ent from that for fish from British Columbia incu- 

 bated at 9.5° or 15°C. Because Rybock et al. ( 1975) 

 did not provide variances, we were unable to test 

 the hypothesis that means from our study coin- 

 cided with theirs. However, mean length and 

 width of otolith nuclei in our study were 29 and 

 559c less, respectively, for resident rainbow trout 

 and 49-70% less, respectively, for steelhead than 

 those studied by Rybock et al. (1975). 



Discussion 



The similarity of our results to those of Neilson 

 et al. (1985), who used similar methods, might be 

 expected for different populations under similar 

 genetic and environmental control. The disparate 

 results of our study and that of Rybock et al. 

 (1975) for the same populations after little ge- 

 netic change (based on comparisons of unpub- 

 lished, biochemical genetic data for these popula- 

 tions from 1972 to 1974 and 1984 to 1986) and 

 little environmental change partly reflected the 

 use of different definitions for the nucleus. We 

 defined the nuclear boundary as the first growth 

 ring surrounding all the fused primordia, 

 whereas Rybock et al. (1975) defined the nucleus 

 as the hyaline area in the center of the otolith 

 that is bounded by a metamorphic check formed 

 at hatching; they resolved the check by rendering 

 the otolith with HCl. We also measured the 

 length of the check surrounding the nucleus, as- 

 defined by Rybock et al. (1975), which we found 

 either to correspond with the area enclosed by the 



first check or to increase in density of growth 

 increments surrounding both the central and ros- 

 tral primordia. The close similarity between our 

 estimate for Round Butte Hatchery steelhead 

 (0.349 mm) and the mean calculated by Rybock et 

 al. (1975) for steelhead (0.354 mm) suggested sim- 

 ilar checks. It is unclear, however, why values for 

 resident rainbow trout for this dimension and the 

 results of tests to discriminate races differed be- 

 tween the two studies. Rybock (1973) noted that 

 the nuclear check could not be distinguished in 

 29% of the otoliths and that the use of HCl may 

 have caused the frequent confusion between the 

 metamorphic check and other groups of daily 

 growth rings. The grinding and polishing of 

 otoliths greatly reduce this source of error. Neil- 

 son et al. (1985) also discouraged the use of meta- 

 morphic checks as boundaries because the causal 

 links between checks and developmental events, 

 such as hatching, have not yet been established. 



Neilson et al. (1985) demonstrated that nuclear 

 length increased significantly with increase in in- 

 cubation temperature from 6.5° to 9.5°C but not 

 from 9.5° to 15°C. Although average water tem- 

 peratures in the Deschutes River were 0.8°C 

 lower during 1982-83 than in 1967-69, it is un- 

 likely that such differences completely explain 

 the greater estimates of mean length and width of 

 otolith nuclei in the earlier study by Rybock et al. 

 (1975). Rybock et al. (1975) calculated mean nu- 

 clear lengths and widths of 0.354 and 0.230 mm 

 for steelhead and 0.243 and 0.154 mm for resident 

 rainbow trout in the Deschutes River. Our esti- 

 mates were 29-70% less than their estimates for 

 a 0.8°C difference; whereas under controlled con- 

 ditions in British Columbia, mean nuclear length 

 for resident rainbow trout at 6.5°C was 18% less 

 for resident rainbow trout and 21% less for steel- 

 head than the nuclear length for fish incubated at 

 9.5°C, a difference of 3°C (Neilson et al. 1985). 



Comparisons of otolith nuclear dimensions be- 

 tween resident rainbow trout and steelhead incu- 

 bated at similar temperatures would establish 

 whether significant differences exist for these 

 measurements between the two races from the 

 Deschutes River. The use of a common definition 

 of nuclear boundaries would allow better com- 

 parisons between studies. However, given the dis- 

 parate results of our study, which were similar to 

 the results of Neilson et al. (1985), and the origi- 

 nal study for steelhead and resident rainbow 

 trout in the Deschutes River, as well as our fail- 

 ure to discriminate between races using both nu- 

 clear definitions proposed by Neilson et al. (1985) 



162 



