550 



Fishery Bulletin 103(3) 



1 



£ 



(•) 



(..) (....) 



B 



I 



1 



(-) (*•) (—) (• 



TSS-1 TSS-2 WS-1 WL-1 TSS-1 TSS-2 WS-1 WL-1 



Method (source of increment counts) 



Figure 5 



Precision evaluation for different aging methods employed for larval 

 (A) and early-juvenile (B) laboratory-reared spot: transverse section of 

 sagitta (polished on one side) (TSS-1), transverse section of the sagitta 

 (polished on two sides) (TSS-2), whole sagitta (polished on one side) (WS-1), 

 and a whole lapillus (polished on one side) (WL-1). The coefficients of 

 variation (CV) values were calculated for three independent increment 

 counts per otolith. Additionally, the confidence of the otolith reader in 

 increment recognition has been indicated by numbers of stars, i.e., poor 

 (*), relatively good (**), good (***), and very good (****). 



Accuracy and precision of age estimates among 

 different types of otoliths and preparation techniques 



Lack of distinct patterns in daily growth increments 

 in otoliths of laboratory-reared fish (e.g., David et al., 

 1994) could make it difficult to conduct laboratory- 

 based ecological experiments with larval fish. Hettler 

 (1984) attempted to validate increment formation rate 

 in the sagittal otoliths of laboratory-reared spot (13- 

 16 mm SL). Within eight days after tetracycline mark- 

 ing, otolith radii increased approximately 18%, but no 

 increments were observed. Siegfried and Weinstein 

 (1989) confirmed daily increment formation in the sag- 

 ittae of field-reared spot larvae, but those reared in 

 the laboratory produced 17 increments instead of the 

 expected 30. Our results, on the other hand, provided 

 direct validation of daily increment formation in the 

 sagittae and lapilli of laboratory-reared spot (Table 2). 

 Even though increment formation was found to occur 

 daily, there were inaccuracies in the estimate of age 

 from otolith increment counts. Twenty-four increments 

 were counted on the sagittae of 34-day-old larvae; if 

 five increments were added for time between first-incre- 

 ment formation and formation of the second check (the 

 starting point of counts used in the present study), age 

 was still underestimated by 4-5 days. Similarly, 24 

 increments were counted on the sagittae of 34-day-old 

 larvae; if 6-7 days were added to account for the tim- 

 ing of increment formation in the lapillus, age was un- 

 derestimated by 3-4 days. Similar inaccuracies in age 

 estimates were derived for 53-day-old juveniles. Peters 

 et al. 1 also found age inaccuracies of five days from 



sagittal increments and concluded that first-increment 

 formation occurred five days after hatching. Given our 

 results and those of Hettler (1984) and Siegfried and 

 Weinstein (1989), we conclude that the likely explana- 

 tion for age inaccuracies is that the increments near 

 the core of the otolith become harder to read as more 

 otolith material is laid down and this process results in 

 the appearance of fewer increments. These inaccuracies 

 would contribute to a 10-15% underestimation of age 

 from sagittae and a 3-11% underestimation of age from 

 lapilli. To account for these inaccuracies, five increments 

 should be added to increment counts to estimate age. 



Lapilli, compared with sagittae, exhibited very clear 

 patterns with increments (Fig. 2) and provided more 

 precise results for the ages of larval and juvenile spot. 

 Although there is no study presenting age data obtained 

 from lapilli for larval or juvenile spot, lapilli have been 

 used successfully for aging many other fish species. 

 Ichimaru and Katsunori (1995) preferred the lapillus 

 as a source of age data for two species of flyingfishes 

 larvae (Cypselurus heterurus doederleini and Cypselurus 

 hiraii) because increments were as clear as those in the 

 sagittae, yet the lapilli did not require any preparation. 

 Bestgen and Bundy (1998) reported increments depos- 

 ited on sagittae of Colorado squawfish (Ptychocheilus 

 lucius) were difficult to distinguish after fish were 30 

 days old and thus lapilli were used to age older fish. 

 Lapilli were the preferred otoliths for age determination 

 of young Lost River sucker (Deltistes luxatus) and short- 

 nose sucker (Chasmistes brevirostris) because of their 

 readability and conservative growth pattern (Hoff et 

 al., 1997). Escot and Grando-Lorencio (1998) concluded 



