120 



Fishery Bulletin 97(1), 1999 



It was possible that Sr could affect fish growth. 

 Thus, the change in mean fish weight per tank between 

 the initial stocking of the tanks and the sampling of 

 the fish remaining after 36 days was compared between 

 treatments by using one-way analysis of variance. 



Experiment 2: concentration and exposure period 



The aim of this experiment was to investigate a range 

 of strontium treatments and exposure periods in or- 

 der to determine minimum concentrations and times 

 required for marking the dorsal spines of juvenile 

 snapper. It was also possible that different batches 

 could be identified by different levels of strontium 

 incorporation. 



The experiment was initiated on 3 March 1994 

 when the fish were 110 days old and 64 mm mean 

 fork length (SD ±4.1 mm). The 16 treatments com- 

 prised three strontium concentrations (Sr-* five 

 times, ten times, and forty times ambient) matched 

 with each of five exposure periods (6, 12, 24, 36, and 

 48 hours) and a control treatment. Replication com- 

 prised two 100 litre tanks per treatment, each con- 

 taining six fish. Snapper were sampled seven days 

 after the termination of the chemical treatment. Sta- 

 tistical analysis involved comparisons between treat- 

 ment concentrations, exposure periods, and tanks us- 

 ing an asymmetrical design analysis of variance and 

 the Student-Newman-Keuls (SNK) multiple range test. 



Strontium levels in wild fish 



Juvenile P. auratus were caught between November 

 1993 and April 1994 with Opera House fish traps 

 (100 X 80 X 60 cm, stretched mesh size 11 mm) from 

 each of four estuarine locations around the Sydney 

 region. From each estuary 20 fish of similar size to 

 those used in experiments one and two were selected: 

 those from Botany Bay had a mean fork length of 63 

 mm (SD ±5.9), from Port Jackson, 80 mm (SD ±9.9), 

 from Middle Harbour, 82 mm (SD ±4.4), and from 

 Port Hacking, 80 mm (SD ±5.6). 



The main aim of this experiment was to compare 

 naturally occurring levels of dorsal spine strontium 

 in wild-caught juvenile snapper from multiple loca- 

 tions, of potentially different water chemistries, with 

 1) snapper immersed in Sr 40x ambient for 48 hours, 

 and 2) snapper immersed in Sr lOx ambient for 5 

 days. Statistical analysis involved one-way analysis 

 of variance and the SNK multiple range test. 



Treatment of experimental Fish 



The experiments were undertaken in an enclosed 

 room containing thirty-two cylindrical 100-litre 



fibreglass tanks, each with a 500-|im mesh filter, an 

 aquarium air stone and a clear plastic cover. Twice 

 daily water exchange from Salamander Bay was pro- 

 vided by a 1000-litre supply tank for each row of eight 

 tanks. Photoperiod was ten hours of light per day, 

 and tanks were cleaned weekly. Approximately 0.125 

 g of feed per fish per day was distributed in three 

 equal portions throughout the day. The specially for- 

 mulated dry snapper diet was a 50% protein extruded 

 diet based on 649c fish meal, similar in composition 

 to that in Quartararo et al. (1992). To maintain a 

 constant stocking density, fish mortalities were com- 

 pensated for by replacement with similar-size un- 

 treated snapper, fin-clipped for identification. 



The water quality parameters measured every sec- 

 ond day were pH (which averaged approximately 8.1), 

 electrical conductivity (approx. 52 mS/cm), turbidity 

 (generally or 1 NTU), dissolved oxygen (approx. 

 8.6 mg/L), water temperature (approx. 19-23°C ), and 

 salinity (approx. 34.3 ppt). No significant differences 

 were observed between tanks for any of these pa- 

 rameters, which remained within a range generally 

 acceptable for the maintenance offish health ( Poxton 

 andAllouse, 1982). 



Treatments were randomly allocated to tanks. All 

 fish chosen appeared healthy and in good condition. 

 Wet weight and fork length were measured prior to 

 introduction of representative groups to the experi- 

 mental tanks. Sedation was necessary to allow safe 

 handling and was achieved with 0.5 mL/L of ben- 

 zocaine solution, which comprised 100 g/L ethyl p- 

 amino benzoate (CgH^jNO,) in 70% ethanol solvent 

 (Summerfelt and Smith, 1990). The fish were accli- 

 matized to the tanks for a period of 5 days for experi- 

 ment 1 and 8 days for experiment 2. Strontium treat- 

 ments were initiated by dissolving the appropriate 

 quantity of chemical salt in each tank. For the dura- 

 tion of the treatment there was no water exchange 

 or cleaning activities. There was also no feeding of 

 the fish during treatment to avoid uneven uptake of 

 the chemical marker among individuals as a result 

 of differential feeding (e.g. Gallahar and Kingsford, 

 1996). Treatment was terminated after the prescribed 

 exposure period by allowing the exchange of new 

 water. Fish were removed by dip net from the tanks 

 and placed into a lethal dose of 1.0 niL/L benzocaine 

 solution at the appropriate time of sampling, then 

 measured, and frozen for storage. 



Dorsal spine tissue samples were dissected from 

 the treated and control snapper Each sample con- 

 sisted of four sturdy spines from the midfront of the 

 dorsal row. The weaker posterior spines were avoided 

 where possible. These samples were then placed into 

 polypropylene eppendorf tubes and dried overnight 

 in a 75°C oven. The tubes were then coaled for half 



