652 



Fishery Bulletin 98(3) 



10 20 30 



Age (days) 



40 



Figure 2 



The relation between increment counts 

 and known-age for spotted seatrout, 

 Cynoscion nebulosus. Solid line repre- 

 sents regression line; dotted, 45% line. 

 Values in parentheses indicate number 

 of specimens. 



Table 2 



Analysis of variance results for increment counts versus 

 age, and inverse regression for age versus increment counts 

 (inverse regression) for all larvae and juveniles ln=69) of 

 known-age spotted seatrout, Cynoscion nebulosus. SE = 

 standard error. 



Dependent 

 variable Slope 



SE Intercept 



SE 



Counts 1.0296 0.0172 -1.9867 0.3.574 0.0001 



Age 0.9538 0.0159 2.2273 0.3135 0.0001 



SL, were marked at 7-d intervals, and which were pre- 

 served 14 d after marking), we observed seven to eight 

 (.ic=7.0) increments between marks and 13-15 (.v=14.2) 

 increments between the last mark and the edge. 



Discussion 



Results 



The slope of the line describing the relation between incre- 

 ment counts and known age was not significantly different 

 from one (Table 2; Fig. 2). This result indicated that rings 

 were being deposited daily and that we were correctly 

 interpreting daily rings from subdaily rings or optical arti- 

 facts, a problem that was encountered with juveniles in our 

 initial readings (see "Materials and methods" section). 



A large percent of the laboratory-spawned and labora- 

 tory-reared fish that were marked with ALC failed to take 

 up the stain on their otoliths. Only 15'7i (^=61) of the 

 known-age fish exhibited a readable mark and only for 

 two age groups. On the other hand, the majority (75%) of 

 wild-caught juveniles exhibited a readable mark on their 

 otoliths. 



Increments were estimated to form at age 2-3 d. An 

 inverse regression (counts regressed on age) yielded an 

 intercept with a value of 2.3 d (Table 2). The intercept 

 was significantly different from zero. Analysis of increment 

 counts on ALC-marked otoliths from known-age larvae 

 also revealed the first increment formed on day 2 or 3. For 

 14-d-old larvae marked at age 9 d, seven increments were 

 counted: for 23-d-old larvae marked at age 9 d, six and 

 seven were counted. 



Increment counts on marked otoliths were fairly accu- 

 rate. On 14-d-old larvae (n=4) marked at day 9, we counted 

 seven increments from the core to the ALC mark, and four 

 to six (.v=5.0: five expected) from the mark to the otolith 

 edge. On 23-d larvae (/i=5) that were marked on day 9 

 and day 16, we counted six to seven (.v=6.6) increments 

 from the core to the first mark, 14 to 15 (.v=14.7) from 

 the core to the second mark, and four to nine (.v =7.6) from 

 the second mark ( 16-d-old) to the edge (23-d-old). For wild- 

 caught larvae (;; = 12) (which ranged from 34.4 to 70.2 mm 



Our overestimation of ages, in our initial analysis (see 

 "Materials and methods" section), from increment counts 

 of older juveniles is difficult to relate to environmental fac- 

 tors. Overestimation of age is not common, although Fives 

 et al. (1986) reported that larger bay anchovies, Anchoa 

 mitchelU, in any known age group generally had more 

 growth increments than smaller fish. Nielson and Geen 

 (1982, 1985) noted that for salmonids. feeding frequency, 

 exposure to warm or cool temperature cycles twice in a 

 24-h period, and an enforced increase in activity increased 

 the rate of increment formation. Fish in our study did not 

 undergo such cyclic events. 



Campana and Moksness ( 1991 ), from a detailed appraisal 

 of accuracy and precision of age estimates derived from oto- 

 liths, concluded that accuracy of age determination from 

 validation experiments is probably "optimistic" owing in 

 part to the constraints and limitations of validation stud- 

 ies done under laboratory conditions. The marked size dif- 

 ference of our known-age 32-d-old juveniles ( 16.2-34.6 mm 

 SL) was undoubtedly a result of laboratory rearing condi- 

 tions and indicated that even with appropriate photoperiod 

 and feeding conditions, caution should be used when inter- 

 preting rings from older laboratory-reared juveniles, espe- 

 cially when size ranges are highly variable as in our study. 

 For older known-age laboratory-reared material, multiple 

 markings should be employed to discern those areas where 

 interpretation is difficult and to discern subdaily rings. 



Previous validation studies of spotted seatrout have 

 indicated that rings are deposited daily, but central rings 

 are difficult to interpret (McMichael and Peters, 1989). 

 McMichael and Peters ( 1989) used otoliths mounted whole 

 from tetracycline-marked wild-caught larvae and juve- 

 niles (7-10 mm). Although they were able to observe the 

 formation of daily rings, they had difficulty in interpret- 

 ing central rings. They used the average measurement to 

 the tenth ring (50 pm, SD=2t taken from few exception- 



