660 



Fishery Bulletin 88(4), 1990 



Details of the microprobe counting procedures, preci- 

 sion, and standards employed in this study are outlined 

 in Table 1 . Background was measured at offsets both 

 above and below peak position for 50% of the peak 

 counting time. The average counts from these two 

 background measurements were subtracted from the 

 peak. Background measurements were made with each 

 analysis. Corrected X-ray intensity ratios were calcu- 

 lated using the ZAF method (Reed 1975), and the final 

 elemental ratios are normalized atom ratios based on 

 concentrations derived from the standards. 



Elemental data for otolith life-history transects from 

 adult salmonids were collected with approximately 40 

 ^m separating each 10 x 10 ^im sample region (50 ^ni 

 separating the center point of each sample). This sam- 

 pling spatial frequency was judged to be adequate 

 because it was desired to collect data relating to major 

 life-history changes only, and not recurring or seasonal 

 events. 



For a comparison of otolith composition among the 

 progeny of sea-farmed fish and freshwater fish, micro- 

 probe data were collected from five individual primcn'- 

 dia within each sagitta or lapillus. In some lapilli it was 

 necessary to make multiple measurements on individ- 

 ual primordia because five primordia were not always 

 exposed or visible. In addition, five microprobe mea- 

 surements were made on each otolith at points near 

 the otolith edge that were indicative of the otolith 

 composition at a time after the completion of yolk 

 absorption. 



Results 



Otolith life-history transects representative of several 

 types of adult salmonids are presented in Figure 1 . 

 Differences in otolith transects among anadromous or 

 sea-farmed fish and non-anadromous individuals are 

 only evident in the Sr/Ca data and, therefore, only 

 Sr/Ca life-history transects are presented. Sea trout, 

 sea-farmed Atlantic salmon, and sea-farmed rainbow 

 trout exposed to the marine environment display a 

 clear increase in otolith Sr/Ca that is coincident with 

 entry into seawater. 



There is an initial peak, associated with the otolith 

 nucleus, in the Sr/Ca ratio of the sea trout and Atlan- 

 tic salmon, but this peak does not appear in the sea- 

 farmed or freshwater rainbow trout. In Tasmania, 

 freshwater broodstock are used for the production of 

 sea-farmed rainbow trout. The peak in otolith Sr/Ca 

 in the otolith nucleus of some individuals is presumably 

 due to the presence of strontium se(juestered in the egg 

 yolk proteins during the seawater phase of ovarian 

 development of the fishes' female parent. 



Figure 1 



Transects of otolith Sr/Ca ratios measured with a waveleiigth- 

 dispersive electron microprobe from the primordium (0 ^m) to the 

 otolith edge along transverse sections of sagittae from 4 female adult 

 salmonids with differing life histories. Each point represents a single 

 measurement made over an area of 100 (jnr. (A) Wild non-anad- 

 romous 0>ic<irbynrhus mukiitx from Great Lake, Tasmania; (B) sea- 

 farme<l Oiicnvhynfhux »iykif:s: (C) sea-farmed Siilmo sidar; and (D) 

 wild aii.idronious Siiltnn Irnltn from the Derwent River esUiary. 



The liypothesis that the seawater strontium contribu- 

 tion to the egg would result in increased otolith stron- 

 tium was confirmed by the results of the rearing ex- 

 periment. Results of this experiment are presented in 

 Table 2. Mean Sr/(^a ratios based on measurements in 

 five primordia from each of the 20 sea-farmed brood- 

 stock progeny was 0.008 1.'3 + 0.00068, significantly 

 greater than the mean value of 0.00114 + 0.00014 

 measured in the nrimocilia of the freshwater brood- 



