476 



Fishery Bulletin 98(3) 



again about 3 months later. Most individuals were identified 

 before release into the cages by tagging with anchor tags 

 or by clipping of fin spines or rays. Some rarer species were 

 not identified this way to reduce handling stress, and other 

 individuals shed tags before recapture. The histories of these 

 captive fish were recognizable from unique combinations of 

 cage and species or, for some L. erythropterus. from twin 

 OTC marks visible on otolith sections (Table 1). 



Monthly measurements of surface seawater tempera- 

 ture and salinity recorded next to the cages and con- 

 tinuous records from the Australian Institute of Marine 

 Science weather stations nearby on the GBR were used to 

 derive information on environmental stressors and wind 

 speed and direction. Caged fish were subject to extremes 

 of wave energy, water temperature, turbidity, and salin- 

 ity that are characteristic of the inshore habitat in the 

 cyclone belt of the dry tropics. Water temperatures rose 

 by 11°C from the austral winter to summer, and fresh- 

 water plumes from cyclonic rains caused fluctuations in 

 salinity and turbidity. These influences extended out to 

 Pandora Reef, but not to Rib and Myrmidon Reefs where 

 there were clear waters and only a 6°C seasonal change 

 in water temperature. 



Laboratory techniques 



For each fish after death, the FL was measured where 

 possible and both sagittae (hereafter referred to as the 

 otoliths I were removed, weighed, and measured. One otolith 



from each fish was randomly selected and embedded in 

 soft epoxy resin. Three transverse sections were cut with a 

 low-speed saw and diamond wafering blade in the vicinity 

 of the primordium. The sections were 0.25-0.50 mm thick, 

 depending on width of the otolith, and were lightly polished 

 on wet ebony paper (1000 grade) and lapping film (9 and 

 3 pm). The sections were mounted on microscope slides 

 under cover slips in a colorless epoxy casting resin and 

 stored in darkness between short periods of examination. 



The counts and measurements of otolith features made 

 at the Australian Central Ageing Facility (CAP) were used 

 in the analyses in our study to avoid possible bias from 

 familiarity with the treatment of the individual fish. These 

 data were compared with counts made by the senior author 

 to assess precision and bias in otolith interpretation. 

 Opaque zones were counted under magnifications from 

 7.5x to 25. 2x with a stereo-dissecting microscope with 

 transmitted white light, or under reflected light against a 

 dark background (Newman et al.. 1996). The position of the 

 OTC mark was then measured by using epifluorescence 

 illumination with a compound microscope at 20x to 40x 

 magnification. 



Video images were obtained, enhanced, and digitized 

 with Bioscan's Optimas r^' and Optimate'i^^' software by 

 using customized macros (Morison et al., 1998) along two 

 planes (Fig. 2), from the primordium to the ventral apex 

 of the otolith (ventral LI axis), and from the primordium 

 to the proximal edge parallel to the sulcus on the christa 

 inferior (sulcal L3 axis). To define the position of the opaque 



