122 



Fishery Bulletin 97(1), 1999 



ambient (SNK: ControkSr 5x <Sr lOx). Slight fluc- 

 tuations, particularly in the control and Sr lOx am- 

 bient treatments (Fig.l), caused significant differ- 

 ences to be found between some sampling times 

 (Table 1), however these were not picked up by the 

 less sensitive SNK test (0 days=36 days=12 days=24 

 days). The ranks of different treatments remained 

 similar irrespective of sample time (i.e. no sample 

 time X treatment interaction) and there was no indi- 

 cation of a significant decay of the strontium signal 

 over the 36 days of the experiment. There were no 



LU 



g 



O 



c 

 o 

 O 



1,000 



900 



800 



700 



600 



500 



400 



0) 



o 

 O 



m i 



Treatment 



25 

 c/3 CO 



Figure 1 



Persistence over 36 days of "^Sr in the spines of juvenile 

 snapper immersed in zero, 5x ambient and lOx ambient 

 strontium for 5 days. 



Table 2 



Analysis of variance for experiment 2: asymmetrical design analysis of 

 variance to test for differences in spine ''^Sr incorporation among treat- 

 ment concentrations (cone.) (Sr^* 5x, lOx and 40x ambient; fixed factor), 

 exposure periods (exp.i i6-48 hours; random factor), and tanks irandom 

 factor). Nonsignificance at P=0.05 is indicated by n.s. Cochran's test: de- 

 grees of freedom idf) = 5; no. of variances ik) = 32; C = 0.1329; variances 

 homogeneous; no transformation. 



Source 



df Mean square F-ratio Probability 



Concentration 2 100,5648 18.167 



Exposure period 4 204047.2 163.683 



Cone. X exp. 8 ' .5.5355.9 44.405 



Tanks (cone, xexp.) 15 1246.6 0.575 



Control vs. rest 1 237036.4 190.146 



Residual 150 2166.3 



significant differences between tanks within each 

 treatment. 



There were no significant differences observed 

 among treatments for the growth of snapper as mea- 

 sured by the change in mean wet weight per tank 

 between the initial stocking and sampling at 36 days 

 (AN OVA: df=2, 9; F=3.27; P=0.086, n.s.). Mortality 

 rates were also very low for all treatments. These 

 results suggested that effects on health or growth from 

 the immersion offish in strontium were minimal. 



Strontium concentration and exposure period 



'^•'Sr was incorporated rapidly into the dorsal spines 

 of treated snapper and on average the concentrations 

 were significantly higher in treated fish than in con- 

 trols (Table 2; Fig. 2). Significant differences were 

 found among groups of snapper treated with differ- 

 ent SrCl.2 concentrations, and the magnitude of these 

 differences varied with exposure period, resulting in 

 a significant concentration x exposure period inter- 

 action (Table 2). Strontium incorporation increased 

 as the exposure period increased, and the magnitude 

 of this change was also not equal between treatments. 

 The signals from the Sr 40x treatment were gener- 

 ally much greater than either of the Sr 5x or Sr lOx 

 treatments, which were similar to one another, and 

 the Sr 40x signal also increased much more rapidly 

 between exposure periods (Fig. 2). 



Strontium in wild fish 



The experimental snapper immersed in Sr lOx for 5 

 days and Sr 40x for 48 hours had spine "^^Sr concen- 

 trations higher than those of any of the control groups 

 (ANOVA: df=6, 105; F=70.9; P<0.001; Fig. 3), with 

 respective concentrations 1.37 (mean 833 |ig/g) and 

 1.55 (942 |ig/g) times greater than the 

 maximum mean concentrations of any 

 wild fish (Port Hacking; 607 ng/g). 



There were also significant differ- 

 ences between estuaries of origin of the 

 wild-caught fish (ANOVA: df=: 3, 74; F= 

 5.8; P < 0.01; Fig. 3). Fish from Botany 

 Bay had lower levels of strontium than 

 those from the other estuaries (SNK: 

 Botany Bay<Port Jackson = Middle 

 Harbour=Port Hacking). There was a 102 

 |ig/g difference in mean spine strontium 

 between the highest and the lowest es- 

 tuary of origin. Fish from Port Jackson 

 and Middle Harbour had intermediate 

 values that were similar to one another. 

 The fish used as controls during the ex- 

 perimental period at Port Stephens also 



<0.01 

 <0.01 

 <0.01 

 n.s. 

 <0.01 



