VARIABILITY IN DIMENSIONS OF SALMONID OTOLITH NUCLEI: 



IMPLICATIONS FOR STOCK IDENTIFICATION AND 



MICROSTRUCTURE INTERPRETATION 



John D. Neilson.i Glen H. Geen ^ and Brian Chan^ 



ABSTRACT 



Sagittal otoliths in rainbow trout, Salmo gairdneri, and chinook salmon, Oncorhynchus tshawytscha, 

 arise by fusion of otolith precursors (primordia) before hatching. Size of the otolith nucleus exhibited 

 considerable variability even in the progeny of a single female. Otolith nucleus length was directly 

 related to the number and position of the primordia and water temperature at which the eggs were 

 incubated. This variability limits the utility of nucleus dimensions as criteria for separating sympatric 

 populations of juvenile steelhead and rainbow trout. Variability in otolith nucleus dimensions also 

 accounted for a significant error in otolith size-fish size relationships in recently hatched alevins. 



The early development of otoliths is poorly un- 

 derstood considering their potential use in stock 

 identification (Postuma 1974; Rybock et al. 1975) 

 and in the provision of data on fish age and growth 

 to the daily level of precision (Pannella 1971; 

 Wilson and Larkin 1982). Variability of otolith 

 nucleus size and shape is of particular concern in 

 stock identification studies since nucleus dimen- 

 sions may be racial characteristics. Rybock et al. 

 (1975) have suggested a positive correlation of the 

 rainbow trout, Salmo gairdneri, otolith nucleus 

 size and the mean egg size of the female which, in 

 turn, is positively correlated to the size of the 

 female. Their data on Deschutes River steelhead 

 trout (the sea-run form of S. gairdneri) females, 

 which were larger, on average, than females of the 

 sympatric population of freshwater resident rain- 

 bow trout, led to the suggestion that otolith nu- 

 cleus dimensions would differ significantly and 

 provide a basis for racial identification of juveniles. 

 This hypothesis was of particular significance 

 since no other meristic or morphometric trait is 

 known which permits identification of juvenile 

 sea-run and freshwater resident S. gairdneri. 



Nucleus dimensions might affect the widths of 

 concentrically formed daily growth increments 

 deposited around the otolith nucleus. Bipartite 

 daily growth increments consist of alternating 



^Department of Biological Sciences, Simon Fraser University, 

 Bumaby, B.C. V5A 1S6; present address: Marine Fish Division, 

 Canada Department of Fisheries and Oceans, Biological Station, 

 St. Andrews, N.B., Canada EOG 2X0. 



^Department of Biological Sciences, Simon Fraser University, 

 Bumaby, B.C., Canada V5A 1S6. 



^British Columbia Fish and Wildlife Branch, Ministry of Envi- 

 ronment, Kamloops, B.C., Canada V2C 5Z5. 



protein and calcium rich zones (Brothers 1981), 

 and their widths are proportional to fish growth 

 during the period of increment formation (Wilson 

 and Larkin 1982). If increment width and number 

 vary as a function of nucleus size and shape, then a 

 source of the 15% error described by Wilson and 

 Larkin in the estimation of fish growth from 

 otolith growth could be identified. 



In this paper, we describe development of sagit- 

 tal otoliths of S. gairdneri (sea-run and fresh- 

 water resident) and chinook salmon, Oncorhyn- 

 chus tshawytscha, and examine the effect of water 

 temperature on otolith nucleus dimensions. These 

 data permit a reexamination of the hypothesis of 

 Rybock et al. (1975). Finally, the implications 

 of variability in otolith nucleus size on otolith 

 microstructure and its interpretation are con- 

 sidered. 



METHODS 



To study otolith nucleus development in S. 

 gairdneri, we obtained eggs from steelhead trout 

 in the Deadman River, British Columbia (B.C.), in 



1981 and from the Nicola and Deadman Rivers in 



1982 (Thompson River tributaries). Rainbow trout 

 eggs were taken from the Deadman River in 1981, 

 and from stocks in Mission Creek and Pennask 

 Lake in south-central B.C. in 1982. Prior to fertili- 

 zation, samples of eggs (n = 20) were taken for dry 

 weight determination (17 of 18 fish collected in 

 1982). In all cases, eggs were fertilized with pooled 

 sperm from 2 to 3 males of similar size and origin 

 as the female. In total, eggs from 10 steelhead and 

 11 rainbow trout were used in this study. 



Manuscript accepted April 1984. 



FISHERY BULLETIN: VOL. 83, NO. 1, 1985. 



81 



