Cappo et al : A new approach to validation of periodicity and timing of opaque zone formation 



479 



using SAS date functions as an integer 

 between 1 and 365. The closing dates of 

 anomalous extra cycles for comparison 

 with environmental stressors were esti- 

 mated by using 



CD,^, = TETJN + 

 365((-l +IF}/V, 



(7) 



where TET,IN 



t + i 



the date of the first 

 OTC injection coded in 

 SAS internal format in 

 units of days: and 

 the /((^ opaque zone 

 completed after the 

 OTC mark. 



The mean of the CCD estimates could be 

 used as an estimate of the calendar date 

 (MCCD) on which the opaque zones were 

 most likely to have been completed. Days 

 or months of the year are circular scales of 

 measurement and best expressed in Polar 

 coordinates as angles and radius unity. 

 Means, angular deviations (analogous 

 to the linear standard deviation), and 

 confidence intervals were therefore cal- 

 culated for circular von Mises distri- 

 butions of CCD for each species by using 

 the methods of Zar ( 1996). The Hotelling 

 paired sample test was used to determine 

 the significance of differences in angular 

 CCD data between reading axes. 



L. sebae 



S1980 



S1983 



S1982 



24-A uq-93 



HI 



Date fisfi killed 



n 



1-IVIar-95 



U 



16-Feb-95 

 17-Feb-95 



n 



17-IVIar-94 



U 



1-Mar-95 



n 

 2-Mar-95 



] u 



1-Mar-95 



Figure 4 



Measurements of the position of opaque zones and the OTC marks (unbroken 

 or extended vertical bar) on the ventral LI a.xis from the primordium to the 

 otolith edge for L. malabaricus and L. sebae. identified by slide numbers. 

 Bars representing field-tagged recaptures are shaded darker than those for 

 captives. Dates of injection and sacrifice are shown. Otoliths marked on the 

 same date are arranged with OTC marks aligned vertically, with a subjective 

 horizontal scale for the spread of injection dates. 



Results 



Recoveries of marked fish 



A total of 82 individuals from 11 species were recovered 

 with OTC marks visible on sections of their otoliths (Table 

 1 ), equating to a recovery rate of less than 20'7( from the 

 field-tagged and captive groups. Times at liberty were less 

 than 22 months and the marked fish were all in younger 

 year classes, less than 8-1-, in relation to the longevities 

 estimated in some current studies (e.g. Williams et al.^). 

 The median ages estimated for the most numerous species 

 were only 3+ for L. erythropterus. L. malabaricus, and L. 

 sebae, and S-t- torL.joh n li . Apart from a few early recoveries 

 of fatalities in the cages, all captive fish were sacrificed in 

 the beginning of March 1995. Fish were recaptured in the 

 field mainly over the summer period between September 

 1994 and April 1995. 



' Williams, D. McB., S. J. Newman, M. Cappo, and P. J. Doherty. 

 1995. Recent advances in the ageing of coral reef fishes. South 

 Pacific Commission, Forum Fisheries Agency Workshop on the 

 Management of South Pacific Inshore Fisheries. South Pacific 

 Commission Integrated Coastal Fish Management Project Tech- 

 nical Document 12. Forum Fisheries Agency, P.O. Box 629, 

 Honiara, Guadalcanal, Solomon Islands. 



Recoveries of field-tagged L.johnii, L. argentimaculatus 

 and L. sebae were low (only 8-W7( of the initial number 

 released were recovered, and tag loss was evident (607f 

 of field recoveries had lost one tag and three others, not 

 included in our study, had an OTC mark but no tags). Of 

 the captive group, only 60 fish from 10 species, including 

 six dead fish, were recovered. None of the smallest species, 

 L. fulviflamma and L. kasmira, and only 5% of the 

 numerous L. vitta remained at the end of captivity. Only 

 eight captive fish had retained anchor tags, and individual 

 lengths at release for captive fish without tags were 

 generally not distinguishable. One field-tagged L. johnii 

 and four captive L. erythropterus had two OTC marks 

 visible on their otoliths. 



Otolith growth and interpretation near OTC marks 



The strength of the working assumptions used in the 

 models developed in our study depended on the sources of 

 variability apparent in rate of growth of the otoliths outside 

 the OTC mark amongst species, age classes, field-tagged 

 and captive groups, and individuals in Figures 4—7. This 

 region generally comprised only a small proportion of the 

 otolith width, especially in the oldest fish, L.johnii (Fig. 6), 



