Kalish Otolith microchemistry to distinguish salmonid progeny 



659 



at 90°C for 60 seconds; ramp ashing from 90 to 700°C 

 for 20 seconds; ashing at 2600°C for 1 second; and 

 atomization, with no gas flow, at 2600°C for 3 seconds. 



To confirm the hypothesis regarding otolith nucleus 

 composition, eggs were obtained from sea-farmed and 

 freshwater broodstock rainbow trout which originated 

 from a similar hatchery stock (Cressy, Tasmania). 

 These two groups of hatchery fish derive from similar 

 stocks of rainbow trout imported into Tasmania; be- 

 cause of minimal gene flow, inbreeding is relatively 

 great and the diversity of the gene pool is low. Both 

 freshwater and sea-farmed adults had been maintained 

 on jack mackerel Trachurus decUvif^ based pellets 

 throughout the period of egg development. Sea-farmed 

 broodstock were kept at the same hatchery as fresh- 

 water broodstock for 3 weeks before stripping, and all 

 eggs were fertilized with sperm from a male of fresh- 

 water stock. The developing embryos frotn sea-farmed 

 and freshwater broodstock wet'e maintained in sep- 

 arate baskets in the same channel of a recirculating 

 water system where temperature was maintained at 

 10°C. A natural photoperiod was maintained through- 

 out the e.xperiment. A random sample of 20 freshwater 

 progeny and 20 sea-farmed progeny was taken 20 days 

 after hatching, and these fish were frozen for later 

 removal of otoliths. Sagittae and lapilli were removed 

 from fry, the adhering otolith capsule was removed, 

 and otoliths were rinsed in deionized water. Otoliths 

 were oven-dried at 40°C. The length of sagittae and 

 lapilli was measured with an ocular micrometer. 



Otoliths from adults were embedded in Araldite D 

 epo.xy resin (Ciba-Geigy) and polymerized in an oven 

 at 40°C for 48 hours. Several transverse sections of 

 approximately 200 /.im thickness were obtained from 



otoliths, including one section containing the pri- 

 mordium, with a low-speed saw (Struers Accutom) 

 equipped with a diamond cut-off wheel. Sections were 

 affixed to glass slides with epoxy. After drying, sec- 

 tions were ground with a graded series of carborun- 

 dum paper (wet/dry paper 600-1200 grade) until the 

 primordia were reached. The much smaller otoliths 

 from fry were mounted, sulcus-side-up, in Crystalbond 

 509 (Aremco Products, Inc.), a heat-labile thermoplas- 

 tic polymer (Neilson and Geen 1981), and ground in the 

 sagittal plane to the level of the primordia with 1200- 

 grade wet-dry paper. All otoliths were then gi-ound to 

 the precise level of the [jrimordia on a lapping wheel 

 with 0.25 t^m aluminium paste (Linde A) and then 

 finished with 0.25 f^m diamond paste. Polished speci- 

 mens were ultrasonically cleaned in reagent-grade 

 mineral spirits followed by ultrasonic cleaning in de- 

 ionized water and oven drying at 60°C. Samples were 

 then coated in a high-vacuum evaporator with a 25-nm 

 carbon layer and stored in a dessicator. 



Otolith elemental analyses were carried out with a 

 Cameca SX-50 wavelength-dispersive electron micro- 

 probe. Analyses were made with a square raster of 

 10x10 (Liiii. Probe current and accelerating voltage 

 were 10 nA (measured on Cu) and 15 kV, respective- 

 ly. Elements analyzed included Ca, Sr, Na, K, and S. 

 Salmonid otoliths are very susceptible to damage from 

 the electron beam, and it was necessary to utilize a low- 

 beam current and reduced counting times for all anal- 

 yses. In conjunction with these conditions that reduce 

 the total number of X-ray counts from a specimen, it 

 is important to be aware of detection limits and count- 

 ing statistics. A treatment of these subjects can be 

 found in Goldstein et al. (1981) and references therein. 



