Lowerre-Barbieri et al .: A comparison of otolith and scale ageing methods for Cynoscion regalis 



565 



common otolith pattern found in many temperate fish 

 of thin translucent zones, which are considered an- 



nuli, interspersed with wide opaque zones (Hyndes 

 etal., 1992). 



Sectioned otoliths were consistently clear and easy 

 to read, as shown by the high precision of repeated 

 age readings. Although it was possible only to vali- 

 date ages 1-5 by separate marginal increment plots, 

 otolith annuli in all ages examined (1-10) were laid 

 down once a year during a discrete time period 

 (April-May). The constancy of annulus deposition at 

 older ages, the lack of severely crowded annuli in 

 older fish, and the similarity between weakfish 

 otoliths and other sciaenid otoliths that have been 

 validated at older ages (Beckman et al., 1990, 

 Murphy and Taylor, 1991; Barbieri et al., 1994) sug- 

 gest that otoliths are a reliable ageing technique for 

 weakfish, although older ages must still be validated. 

 In contrast, we found the scale method of ageing 

 weakfish to be imprecise and apparently inaccurate 

 at older ages. We found that scales form annuli over 

 an extended period, April-August, similar to the re- 

 sults of past studies (Perlmutter et al., 1956; 

 Massmann, 1963b). This protracted period of annu- 

 lus formation made it difficult to assign ages to fish 

 taken in midsummer with moderate growth on the 

 scale margin, as noted by Massmann (1963b). For 

 example, a fish taken in July with a medium mar- 

 ginal increment on its scale could have formed its 

 annuli in early April and have grown since then, or 

 it could have increased its growth increment before 

 forming an annulus in August. Thus, assigning an 

 age to these fish is purely subjective and can lead to 

 ageing errors ± one year, which may explain most of 

 the discrepancies between otolith and scale ages. 



The long period of annulus formation on scales and 

 the severe crowding of annuli at older ages make it 

 difficult to validate scales by the marginal increment 

 method — as Perlmutter et al. (1956) and Shepherd 

 and Grimes (1983) attempted for pooled age data. 

 Because scale annuli form over a protracted pe- 

 riod, the trough in the marginal increment plot is 

 shallow and the range of marginal growth during 

 other months is large. Additionally, validation by 

 the marginal increment method is not appropri- 

 ate if the hardpart shows severe crowding of an- 

 nuli at older ages, as we found with scales, and 

 has been previously reported (Shepherd, 1988). 

 Shepherd (1988) described annuli in fish older 

 than age 6 or 7 as being crowded and very diffi- 

 cult to detect, which could lead to marginal incre- 

 ments being measured from the last distinguish- 

 able annulus to the edge, rather than from the 

 last real annulus to the edge. This error would 

 inflate marginal increment estimates and there 

 would be no way to detect underaged, older fish in 

 marginal increment plots. 



