Pilling et a!.: Validation of annual growth increments in the otoliths of Lethnnus mahsena and Apnon virescens 



605 



L. mahsena Marginal increment analy- 

 sis indicated an overall increase in the 

 marginal increment between January and 

 May, falling in August (Fig. 7). This pat- 

 tern was clearest in the younger, faster- 

 growing growth group. A unimodal distri- 

 bution was found in the plot of the propor- 

 tion of stained edges found in each month, 

 as well as a high proportion of individu- 

 als showing opaque growth commencing 

 between May and August (Fig. 8). 



A. virescens With relatively few samples 

 at certain key times during the year, pat- 

 terns in the marginal increment could not 

 be identified clearly (Table 4). The width 

 of the outer zone increased until June 

 in younger, faster-growing individuals. In 

 subsequent months, a lack of samples pre- 

 vented the identification of an indicative 

 pattern (Fig. 9). No pattern was found in 

 the marginal increment of older individu- 

 als. Staining improved the clarity of the 

 type of growth present on the edge of the 

 otolith. Edge analysis of these specimens 

 showed a unimodal distribution of the per- 

 centage of stained edges by month. A high 

 proportion of individuals exhibited opaque 

 zone formation in the period between Octo- 

 ber and March (Fig. 10). 



P. filamentosus No overall pattern of 

 increase in the marginal increment was 

 found (Fig. 11). This result was partly due 

 to the lack of samples in the months of 

 May and June. The legibility of the incre- 

 ments in the otoliths of this species was 

 poor, and staining failed to improve the 

 clarity of the otolith zones. Edge analysis 

 was confused by indeterminate increments 

 formed in the translucent growth period. 

 Their presence on the edge of the stained 

 otolith at the time of capture led to the 

 misidentification of opaque zone formation 

 and thus confused the identification of an overall pattern. 

 As a result, validation could not be achieved with this 

 method. 



Discussion 



Back-calculation indicated that there was a direct linear 

 relationship between fork length and otolith radius. The 

 otoliths of each species showed an internal structure, with 

 increments widely spaced near the nucleus and narrower 

 and more evenly spaced toward the edge. Growth curves 

 derived from back-calculated length-at-age data indicated 

 an exponential decrease in increment width with assumed 

 increasing age. The otoliths of the three species therefore 

 fulfilled the first two criteria for their use in aging, at least 



over the age range seen in the samples, in that the otolith 

 grew throughout the life of the fish and showed an inter- 

 nal structure of increments. The pattern in P. filamentosus 

 was, however, confused by the presence of indeterminate 

 increments. 



Confirmation of the third criterion (that increments seen 

 in the otoliths corresponded to a regular time scale) was 

 not demonstrated so readily. Validation could not occur 

 through methods such as tetracycline injection owing to 

 practical limitations imposed by the fisheries and to the 

 depths inhabited by certain species. Pristipomoides fila- 

 mentosus, for example, is caught at over 75 m depth. The 

 majority of specimens would be dead or heavily stressed 

 through barotrauma on reaching the surface (DeMartini 

 et al., 1996) and useless for tagging programs. Marginal 

 increment and edge analysis was therefore used. 



