at this site are, for the most part, continuous and 

 distinct. The annular markings on the scales are not 

 as distinct or as sharp in contrast as on the otoliths. 

 The scale annuli from the first three ages are well 

 defined, but in ages 6 and 7 the outer annuli start 

 to run together and the scale margin shows wear 

 from resorption and deterioration. In contrast, the 

 age-6 and -7 otoliths still reveal distinct annular 

 rings. 



PA, APE, and V values for all 10 treatments are 

 presented by treatment comparisons in Tkble 1. The 

 scale vs. otolith treatments were compared to deter- 

 mine how similar the two readers were in inter- 

 preting scale and otolith annuli. The PA values 

 (82.6%, 80.4%) are relatively close to one another 

 along with the values for APE and V. These values 

 did reveal that R^ had more agreement and less 

 variability between aging scales and otoliths than did 

 Rg. Tkble lb shows a higher percentage agreement 

 and less variability between Rj and R2 for otoliths 

 than for scales. 



Comparisons made between treatments of scales 

 to true age and treatments of otoliths to true age 

 are shown in Tkble Ic The agreement of otolith ages 

 to true ages was 10% greater and had half the varia- 

 tion of scale ages. Treatment comparison between 

 male and female scale values were about the same 

 (Ikble Id), whereas the otolith values between sexes 

 are quite different (Ikble le). The female otolith 

 treatment had a higher PA value and lower APE and 

 V values than did the male treatment. 



Discussion 



The scale vs. otolith treatment (Tkble la) reveal- 

 ed no real difference in the aging ability of each 

 reader. There were also higher PA and lower V 

 values for the treatment of otolith to true age (Tkble 

 Ic). Therefore, the outcome for other treatments 

 could be attributed to differences in the aging 

 methods rather than the aging ability of the readers. 



The results of the statistical tests (Tkble lb) sug- 

 gest that the method of estimating alewif e ages from 

 otoliths is superior to the method using scales. Ages 

 derived from otoliths revealed a higher reproduc- 

 ibility and less variability than those obtained from 

 scales. These differences between methods were in- 

 herent within the otoliths of female alewives, which 

 provided the greatest accuracy and lowest variabil- 

 ity (Tkble le) within all types of statistical 

 comparisons. 



Female alewives, at one time during their lives, 

 possibly achieve a faster rate of growth than males. 

 Cooper (1961), Havey (1961), and Libby (1982) show- 

 ed that females are larger than males at aga This 

 attribute of females would have the same effect on 

 the growth of the otolith resulting in wider opaque 

 zones with more distinct hyaline rings than male 

 otoliths. Female scales would also have greater 

 growth but possibly because of scale resorption and 

 deterioration they were less easy to read than 

 otoliths. 



Other investigators have mentioned the ease of in- 

 terpreting annuli of otoliths compared with scales, 

 but have remarked that otoliths are less convenient 

 to collect and store (Norden 1967; Kornegay 1978). 

 The method presented in this paper was found to 

 be more effective in obtaining both otoliths, un- 

 broken, than the commonly used transverse cut 

 made to the side of the head. The method of storage 

 eliminated transferring and handling individual 

 otoliths each time they were used. A disadvantage 

 in using otoliths was that the fish had to be sacri- 

 ficed. More care was required initially to obtain 

 otoliths than scales, but the technique was quickly 

 learned. The use of otoliths for aging alewives was 

 more accurate and less subjective than aging with 

 the use of scales. 



Acknowledgments 



I wish to thank Sherry Collins, Maine Department 



Table 1.— Percent agreement (PA), average percent error (APE), and mean coefficient of varia- 

 tion (V) values for five treatment comparisons: a) scales and otoliths for R, against Rji b) 

 scales of R^ and Rj against otoliths of R, and Rg; c) scales of R, and Rg and true age against 

 otoliths R, and Rj and true age; d) scales of R, and Rg for males against scales of R, and 

 Rj for females; and e) otoliths of R, and Rj for males against otoliths of R, and Rg for 



700 



