Pollard et al.: Chemical marking of Pagrus auratus 



127 



poses. Removal of spines or scales causes minimal 

 physical defacement of the product, an aspect of 

 particular relevance to the Japanese market where 

 aesthetic appeal is highly valued. 



The immersion technique for incorporation of 

 strontium enables greater control over the degree of 

 exposure offish to strontium than does introduction 

 through the diet. The latter method may introduce 

 inconsistencies between experimental fish as a re- 

 sult of differential feeding, which often results from 

 size hierarchies within the tanks (Umino et al., 1993). 

 Intraperitoneal or intramuscular injection allows 

 similar control of exposure to that of immersion; how- 

 ever each fish must be dealt with separately. Scott 

 (1961) was unable to mark fish under 8 cm long by 

 using injection techniques. Nevertheless, with large 

 pelagic fish injection provides many advantages, in- 

 cluding the ability to tag directly from the capture 

 vessel. 



The strontium tag showed no signs of decay due to 

 leaching from the spine tissue, and tag retention 

 times may prove to compare favorably with conven- 

 tional methods. External physical tags are shed at 

 different rates for different species, fish sizes, tag 

 types, or attachment sites (e.g. Ingram, 1993), and 

 may facilitate disease. Tattoos may be rendered un- 

 readable by fish growth, and vital dyes may be 

 leached from the animal (Laird and Stott, 1978). 

 Clipped fins may regenerate, or there may be confu- 

 sion between clipped fins and fins that have been 

 naturally excised by predators. Population studies 

 also assume that the proportion of tagged to untagged 

 fish captured is representative of that existing in the 

 population. These assumptions hold for strontium 

 marking, whereas individuals marked with tags may 

 be less fit for survival, more conspicuous to predators, 

 or more likely to become entangled in capture nets. 



Radioactive isotopes are effective markers offish, 

 as evidenced by the permanent marking of many fish 

 by ^^C after atomic bomb testing in the 1950s (Kalish, 

 1995). A number of different radioactive isotopes have 

 been investigated for fish tagging, for example ^^Ca 

 (Bogoiavlenskaia, 1959;Anwand, 1966), -^^PiKarzin- 

 kin et al., 1959), ^^Fe (Scott, 1961), ^^"Cs (Scott, 1962), 

 I'^^Ce (Hoss, 1967), ^^il (Fitzgerald and Keenleyside, 

 1978), is^Eu and ^^^Eu (Hansen and Fattah, 1986), 

 ^^Sr (Carlson and Shealy, 1972; Lehtonen et al., 

 1992), s^Sr (Farrell and Campana, 1996), and ^^Sr 

 (Zhao et al., 1992). Marking with stable strontium 

 salts however has many advantages over radioactive 

 tagging, which is comparatively expensive, environ- 

 mentally unattractive, and a potential threat to hu- 

 man health. The latter is an important criticism 

 when radioisotope techniques are used to mark com- 

 mercially important species, such as radioactive iri- 



dium used to tag and release snapper (Kato, 1990; 

 Kato et al., 1991). Conversely, strontium immersion 

 has in fact been suggested as a low-level marking 

 technique for farm-reared brook trout destined for 

 human consumption (Guillou and de la Noue, 1987). 



A major disadvantage of the technique is that the 

 existence of the strontium mark cannot be detected 

 in recaptured fish prior to analysis. This means that 

 large batches of snapper of the appropriate size may 

 have to be analyzed when only a small proportion 

 may be marked individuals. Most conventional tag- 

 ging methods, including most radioactive isotope 

 markers, allow in vivo detection of tagged fish. The 

 number of fish requiring analysis should be mini- 

 mized by the preselection of appropriate-size fish 

 based on growth rate estimates for the appropriate 

 geographic location (e.g. Francis and Winstanley, 

 1989; Tsukamoto et al., 1989; Paul, 1992). Moreover, 

 other external features may also aid in the preselec- 

 tion of possibly marked fish; for example, a large 

 proportion of hatchery-reared Japanese snapper lack 

 an internostril epidermis (Sobajima et al., 1986). 

 Salmon stocks of different origin have been identi- 

 fied by using naturally occurring differences in scale 

 patterns (Bilton, 1972), otolith patterns (Hindar and 

 Abee-Lund, 1992), vertebral elemental composition 

 (Mulligan et al., 1983), or parasitic fauna (Margolis, 

 1963). 



Stock enhancement or reseeding programmes may 

 be of particular importance when there has been re- 

 cruitment failure due to natural or anthropogenic 

 environmental change, or as a result of overfishing. 

 Pagrus auratus have been ranched and released for 

 many years in Japan. Concerns have been raised 

 about alterations to the gene pool from the introduc- 

 tion of genetically inferior cultured fish (Hindar et 

 al, 1991; Harada, 1992). Smith and Hataya (1982) 

 have however estimated that the release of one mil- 

 lion snapper per year would increase annual net re- 

 turns to the fishery by about 35 metric tons, and the 

 Kagoshima Bay ranching operation appears to have 

 become highly profitable ( Matsuda, 1992 ). There has 

 also been consideration of the establishment of snap- 

 per reseeding operations in the southern hemisphere 

 (Smith and Francis^). Strontium tagging can assess 

 the economic or conservation value of such stock en- 

 hancement operations and lead to refinement of the 

 reseeding techniques, for example selection of the 

 optimal locations and fish sizes for release. Consid- 

 eration of migratory patterns (e.g. Crossland, 1976; 

 Edmonds et al., 1989) is required to achieve spatial 

 separation of studies, and temporal division of stud- 

 ies may also be achieved by the use of different co- 

 horts separated by sufficient time to allow size sepa- 

 ration. The success of the strontium immersion tech- 



