NOTE Fey et al.: Effect of type of otolith and preparation technique on age estimation of Leiostomus xonthurus 547 



"/ 



* 



Figure 2 



Otolith microstructure of an early-juvenile 

 laboratory-reared spot (Leiostomus xan- 

 thurus): (A) transverse section of the sagitta 

 (polished on two sides); (B) whole sagit- 

 ta (polished on one side); and (C) whole lapil- 

 lus (polished on one side). Scale bar = 30 um. 



Increment formation occurred daily in both sagittae 

 and lapilli after the early larval period. The difference 

 in number of increments counted from sagittae and 

 lapilli from fish sampled 34 and 53 days after hatching 

 reflected the time elapsed between these two samplings 

 (Table 2) and indicated daily increment formation be- 

 tween the larval and early juvenile stage. The same 

 daily increment formation was also observed for larvae 

 sampled 12 and 27 days after hatching during the ex- 

 periment on first-increment formation (Table 2). 



The accuracy of larval age estimates were similar for 

 all the sagittae and lapilli preparation methods (ANO- 



Figure 3 



Central otolith area of early-juvenile labora- 

 tory-reared spot (Leiostomus xanthurus): (A) 

 transverse section of sagitta (polished on two 

 sides); five increments are visible between 

 hatching check (H) and, presumably, first feed- 

 ing check (FF); (B) whole lapillus (polished on 

 one side) with daily increments deposited after 

 the check was formed six days after hatching. 

 Scale bar = 10 i<m. 



VA, P>0.05; Fig. 4A). For juveniles, however, there was 

 a significant difference in the number of counted incre- 

 ments among sagitta preparation methods (ANOVA, 

 P<0.001) (Fig. 4B). A lower number of increments were 

 enumerated from transverse sections of sagittae (with 

 one side polished) (post hoc: Tukey HSD for unequal n, 

 P<0.001). Moreover, -25% of otoliths within this group 

 were not readable. 



All the otolith preparation techniques, except the PIS 

 transverse sections of sagitta from juveniles, underesti- 

 mated the age from hatching by 9-10 days. A 6-7 day 

 difference was expected between known age and lapilli 

 increment counts, owing to the time of first-increment 

 formation. Thus, actual fish age was underestimated by 

 approximately 2-4 days with lapilli increment counts. A 

 5-day difference was expected between known age and 



