remain separate throughout the year (Wakabayashi 

 footnote 6). These tagging data, together with dis- 

 tribution patterns and morphological differences 

 between populations of the two main wintering con- 

 centrations, suggested that these groups may con- 

 stitute independent northern and southern spawning 

 stocks. Other evidence supporting a two- stock con- 

 cept was 1) apparent differences in growth rate and 

 length-weight relationships between samples from 

 the two areas (Wakabayashi footnote 6), 2) differ- 

 ences in egg diameter between samples from north 

 and south of Nunivak Island, where independent 

 spawning areas for the two stocks might exist 

 (Kashikinal965),and3) distribution patterns shown 

 by research vessel surveys in spring and late summer 

 (Chikuni 1971; Yamaguchi 1972), which indicated 

 independent concentrations of fish in northern and 

 southern stock areas. 



However, results from other studies have not sup- 

 ported a two- stock concept. Fadeev (1970) found no 

 significant differences in growth rates, length- weight 

 relationships, body proportions, and meristic charac- 

 ters for samples from the two principal wintering 

 groups near Unimak Island and west of St. Paul 

 Island, and Wakabayashi (footnote 6) found no 

 significant differences in the relationship between 

 total body length and radius of the otolith for samples 

 from the two areas. Moreover, winter concentrations 

 of small yellowf in sole have only been found in Bristol 

 Bay (Fadeev 1970), and tag recoveries since 1974 

 have shown more intermixing offish between the pro- 

 posed northern and southern stock areas than had 

 earlier tagging data (Wakabayashi et al. footnote 

 4). 



In the Gulf of Alaska, yellowfin sole are much less 

 abundant and have not been targeted by directed 

 fisheries. As a result much less is known about the 

 geographic distributions of morphological and life 

 history traits in this area. 



In this paper, the geographic distributions of elec- 

 trophoretically detectable protein variants was used 

 to describe the genetic stock structure of yellowfin 

 sole in the eastern Bering Sea and the Gulf of Alaska. 

 Specifically, samples were collected in the inshore 

 spawning areas of the eastern Bering Sea to deter- 

 mine whether the northern and southern stocks are 

 genetically distinct. In addition to these data, the 

 data from one sample of Japanese yellowfin sole pro- 

 vide an estimate of the amount of ocean- wide genetic 

 differentiation among populations. Finally, the 

 genetic population structure of yellowfin sole is com- 

 pared with the genetic structures of four species of 

 flatfish using the gene- diversity analysis as a sum- 

 mary statistic. 



FISHERY BILLETIN: VOL. 81, NO. 4 



MATERIAL AND METHODS 

 Electrophoresis 



Tissue samples or whole fish were collected at 12 

 locations in the southeastern Bering Sea, at 3 

 locations in the Gulf of Alaska, and at 1 location off 

 Hokkaido, Japan, and shipped frozen to Seattle, 

 Wash. (Table 1, Fig. 1). Samples were held at -25°C 

 up to 5 mo until laboratory analysis. The tissues 

 assayed for specific proteins using horizontal starch- 

 gel electrophoresis were skeletal muscle, heart, 

 stomach muscle, vitreous fluid of eye, brain, liver, 

 spleen, kidney, gill, and gonad, but only skeletal mus- 

 cle, heart, vitreous fluid, and liver were examined in 

 all of the samples. Extraction procedures and elec- 

 trophoretic methods followed May et al. (1979). Gels 

 consisted of 13% hydrolyzed potato starch (Electro- 

 starch, Madison, Wis., lot 307; Sigma starch, lot 

 39c-0459) 7 . The locations of specific enzymes were 

 visualized in the gels using solutions described by 

 Harris and Hopkinson (1976). The peptidase stain- 

 ing method A of Harris and Hopkinson (1976) was 

 used to detect zones of activity of a number of pep- 

 tidases, except that 0-dianisidine diHCl was used as 

 a dye coupler. Three peptides, leu-ala, leu-gly-gly, 

 and phe-pro were used as peptidase substrates. 



Three buffer systems were used to achieve max- 

 imum resolution of the protein bands on the gels: (I) 

 gel, TRIS 0.03 M, citric acid 0.005 M (pH 8.5), tray, 

 lithium hydroxide 0.06 M, boric acid 0.3 M (pH 8.1) 

 (Ridgway et al. 1970); (II) gel, 1:20 dilution of tray 

 solution, tray, citric acid 0.04 M adjusted to pH 6.1 



'Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



Table 1.— Locations (see also Figure 1) and collection dates of 

 samples of yellowfin sole used for electrophoretic analysis. 



668 



