FISHERY BULLETIN: VOL. 84, NO. 3 



Figure 12.— Age composition of Centropristis 

 striata at the two experimental mark-recapture 

 sites. 



50 



30 



x Age 3.4 



Site 1 

 1981 



n 



Site 1 

 1982 



50-, 



30 



10 



x Age3.3 



Site 2 

 1982 



a 



>i _ 



50 



30 



10 



x Age3.1 



Site 2 

 1983 



n 



3 



AGE 



DISCUSSION 



Age and Growth 



The cause of the variation in size and shape of the 

 sagitta's central field in C. striata is unknown, how- 

 ever, differing size of the nuclei of Atlantic herring, 

 Clupea harengus, can be related to spawning season 

 (Postuma 1974). Further studies are needed to 

 determine if these differences can be related to 

 spawning time of C. striata in the South Atlantic 

 Bight. 



Inadequate validation in many studies that esti- 

 mate age have been noted by Beamish and McFar- 

 lane (1983), and they have reemphasized the need 

 for verification of aging technique. Our attempts at 

 validation have shown that one annulus is formed 

 each year during April-May. Also, our counts of 

 presumed daily growth rings have provided circum- 

 stantial support for the formation of the first an- 

 nulus. A similar approach was used by Radtke et 

 al. (1985) in their study of the oyster toadfish, Op- 

 sanus tau. 



Our mean back-calculated lengths agree well with 

 Mercer's (1978) data for C. striata from the South 



Atlantic Bight up to age 5; however, ours are much 

 smaller than Cupka et al. (fn. 5). Our lengths at age 

 are consistently smaller than C. striata from the 

 Middle Atlantic Bight (Mercer 1978). Mercer (1978) 

 attributed size at age differences between the two 

 areas to gear selectivity, yet our results suggest that 

 C. striata from the South Atlantic Bight are smaller 

 than those of the Middle Atlantic Bight. The larger 

 size at age found by Cupka et al. (fn. 5) may reflect 

 the population of C. striata in the South Atlantic 

 Bight prior to heavy exploitation that began in 1969. 

 Since estimates of L m , K, and t are affected by 

 several nonbiological, methodical factors, direct 

 comparisons of these growth parameters between 

 different studies are of limited value. However, 

 when viewed in relative terms, they can indicate 

 general differences or similarities between studies, 

 species, or areas. Our estimate of L^ (341 mm SL) 

 was much closer to Mercer's (1978) value (L^ = 

 352 mm SL) than that of Cupka et al. (fn. 5) (°°625 

 mm SL). Our growth coefficient (K) was higher, in- 

 dicating that C. striata achieves maximum attain- 

 able size more rapidly than previously reported. 

 These differences could have been caused by sam- 

 pling methodologies and/or conditions of the popula- 



738 



