358 



Fishery Bulletin 104(3) 



1977). Correspondingly, I found the ML^g in the GOA 

 (35 cm) was 46% greater than that off the Oregon coast. 

 Although there is a possibility that rex sole MLr,,, in 

 the GOA was overestimated because sampling did not 

 occur prior to the onset of spawning, the bias cited by 

 Hunter et al. (1992) was relatively small and can not 

 fully account for the 11 cm difference in ML^^ between 

 the GOA and Oregon. However, some caution must be 

 exercised when maturity is compared between the GOA 

 and Oregon (Hosie and Horton, 1977) because about 

 25 years elapsed between studies and different criteria 

 were used to classify fish as mature. 



Differences in the histological criteria used to define 

 mature individuals and reliance solely on gross anatom- 

 ical methods can lead to a misclassification of maturity 

 and produce differences in estimates of MLjq (Hunter 

 et al., 1992; Zimmermann, 1997). In a comparison of 

 histological and gross anatomical methods. Hunter et 

 al. (1992) found 1% of active female Dover sole were 

 misclassified as inactive and 12% were visually clas- 

 sified as having advanced yolked oocytes and believed 

 capable of spawning, whereas histological analysis in- 

 dicated that they were inactive and future spawning 

 was unlikely. Likewise, macroscopic examination of 

 arrowtooth flounder misclassified 18.4% of maturing 

 or spent females as immature and 4.4%i of immature 

 females as spent (Zimmermann, 1997). Unfortunately, 

 in the only previous investigation on rex sole maturity, 

 Hosie and Horton (1977) limited their analysis of ova- 

 ries to macroscopic staging. However, the enormous dif- 

 ferences in MLgQ between the GOA stock and the West 

 Coast stock off Oregon likely stem from biological dif- 

 ferences between stocks rather than from inconsistent 

 methods used in the two studies. This interpretation is 

 consistent with the dramatic difference in length at age 

 between the two areas (Fig. 8) that was demonstrated 

 in this study. 



Given that Hosie and Horton (1977) used the otolith- 

 surface aging technique to analyze rex sole otoliths, 

 they may have underestimated ages and thereby biased 

 their estimates of both age at maturity and growth 

 rates. However, the tendency to underestimate age 

 with the otolith-surface aging technique is greater for 

 older fish where the outer edges of the otolith are worn 

 (Anderl"). Surface analysis of several GOA rex sole 

 otoliths revealed that, although break and burn is the 

 preferred technique for aging rex sole otoliths, it is pos- 

 sible, depending on the clarity of the annular pattern, 

 to age females less than 10 years old accurately by 

 using otolith surfaces (AnderF). Because the ages at 

 50% and 100% maturity off the Oregon coast were 5 

 and 9 years old, respectively (Hosie and Horton. 1977). 

 there is a lower potential for bias caused by the sur- 

 face aging technique because fish were <10 years old. 

 Although these different aging methods may limit 

 comparisons of age at maturity, the similarity in age 



60 



50 



£■ 40 - 



30 



20 



10 



•^ o o 



. .O . . o . - - 



° e 



ogo 



— GOA growth curve 

 o GOA data 



Oregon growth curve 



(Hosie and Horton, 

 1977) 



10 15 



Age (yr) 



20 



25 



30 



Figure 8 



The von Bertalanffy growth curve with 9.5% confidence 

 limits for female rex sole (Glyptocephalus zachirus) in 

 the Gulf of Alaska. Parameters of the Gulf of Alaska 

 growth model were L^ = 41.82, k = 0.388, and ?„ = -0.022. 

 For comparison, the growth curve for age 1-15 female 

 re.x sole off the Oregon coast is graphed with data from 

 Hosie and Horton (1977). 



Anderl, D. 2004. Personal commun. Alaska Fisheries 

 Science Center, National Marine Fisheries Service, 7600 

 Sand Point Way NE. Seattle, WA 98115. 



at maturity between the GOA and Oregon is strongly 

 indicated. Additionally, because the length-at-age data 

 presented in growth curves from Oregon were only for 

 rex sole <15 years old (Hosie and Horton, 1977), and a 

 difference in growth is evident at young ages (Fig. 8), it 

 seems unlikely that potential bias from otolith-surface 

 aging methods — rather than from biological differences 

 between stocks — could account for the markedly lower 

 growth rate of Oregon rex sole. To be sure, it would be 

 useful for future research to compare the reproductive 

 biology and growth data of female rex sole in the GOA 

 with a current study off the Oregon coast that employed 

 similar aging techniques and histological criteria for 

 maturity to determine if the distinct differences found 

 in the GOA stock (i.e., larger size at maturity, similar 

 age at maturity, faster growth rate) persist. 



Acknowledgments 



Sampling was a cooperative effort and fish were collected 

 with the assistance of National Marine Fisheries Ser- 

 vice, University of Alaska Fairbanks School of Fisher- 

 ies and Ocean Sciences, and Kodiak Island processing 

 plants (Trident Seafoods, Alaska Pacific Seafood, and 

 Cook Inlet Processing). I especially thank A. Barns, 

 D. Benjamin, E. Brown, H. Emberton, R. Foy, B. Hol- 

 laday, and N. Raring. I am grateful for the histology 

 advice from D. Nichol and B. Macewicz, and statistical 

 advice from M. Litzow. I thank E. Acuna for preparing 

 slides, and I thank D. Anderl, J. Lyons, and R. Katona 



