FISHERY BULLETIN: VOL. 86, NO. 3 



grain-size distribution comprised of fine-grained 

 material (i.e., <0.063 mm in size). This variable was 

 used as an index of sediment character because 

 many physical, chemical, and biological character- 

 istics of sedimentary environments covary with the 

 grain-size distribution (Buchanan 1984). 



During 1981-82, English sole were collected from 

 15 May to 6 July 1981 at 12 transects in Commence- 

 ment Bay, Elliott Bay, and Sinclair Inlet (Fig. 1). 

 Fishes also were collected during three additional 

 time periods (23-24 November 1981; 8-9 March 

 1982; 30-31 August 1982) to evaluate seasonal 

 patterns at three stations in Commencement Bay 

 (CB-1, CB-4, and CB-5; Fig. 1). These stations were 

 selected to represent a gradient of sediment grain 

 size. Although the seasonal sampling occurred over 

 a 2-yr period, it was assumed that interannual varia- 

 tion was not sufficiently large to obscure major 

 seasonal patterns. 



During the May- July sampling, English sole were 

 collected at each transect during morning (07:00- 

 09:30), midday (10:00-13:30), evening (20:30-23:30), 

 and midnight (00:30-04:00). During the remaining 

 sampling periods, fish were collected only during 

 midday and evening. For each transect, results ob- 

 tained for different periods of the diel cycle were 

 pooled before male proportions were calculated. Sex 

 determinations and evaluations of spawning condi- 

 tion were made for all English sole larger than 160 

 mm total length (TL). Age was not determined. 



During 1984, English sole were collected from 4 

 to 9 June 1984 at 15 transects in Commencement 

 Bay and at 2 transects in Carr Inlet (Fig. 1). Sam- 

 pling was conducted during daylight hours (06:30- 

 17:00) at all transects. Sex determinations and 

 evaluations of spawning condition were made for all 

 individuals larger than 225 mm TL. Age was deter- 

 mined by otolith analysis for all fish. 



During both surveys, English sole were collected 

 using a 7.6 m (headrope) Marinovich otter trawl 

 having a body mesh of 3.2 cm (stretched) and a cod 

 end liner mesh size of 0.8 cm (stretched). Trawling 

 was conducted along isobaths at a constant vessel 

 speed of approximately 2.5 knots. 



During both surveys, bottom sediments were sam- 

 pled using a modified van Veen bottom grab and the 

 grain-size distribution of the top 2 cm of sediment 

 was determined using standard sieve and pipette 

 techniques (Folk 1968). During 1981-82, sediments 

 were collected at five sampling points located at 

 approximately equal distances along each transect. 

 During 1984, sediments were collected at a variable 

 number of stations located within 300 m from each 

 transect. 



The association between percent fine-grained 

 sediment and male proportion of English sole was 

 tested using Spearman's rank correlation coefficient 

 {Vg). The association between station depth and 

 male proportion was tested in the same manner. 

 Differences in male proportion among age-classes 

 of fish and among the seasonal sampling transects 

 were tested using the G-test (Sokal and Rohlf 

 1981). 



RESULTS 



Station depth, percent fine-grained sediment, 

 sample size, and male proportion at each transect 

 from the 1981-82 and 1984 surveys are presented 

 in Tables 1 and 2, respectively. During 1981-82, 

 4,430 English sole were sampled from stations 

 ranging in depth from 8 to 36 m. Percent fine- 

 grained sediment ranged from 5.8 to 92.1% and 

 male proportion ranged from 0.20 to 0.91. During 

 1984, 1,007 English sole were collected from 

 stations ranging in depth from 5 to 20 m. Per- 

 cent fine-grained sediment ranged from 8.6 to 

 87.5% and male proportion ranged from 0.13 to 

 0.98. 



The correlation between percent fine-grained sedi- 

 ment and male proportion of English sole was 

 significant (P < 0.01; r, = 0.73) for the May- July 

 sampling of the 1981-82 survey (Fig. 2). This cor- 

 relation was also significant (P < 0.001; r^ = 0.80) 

 for the 1984 survey (Fig. 2). Correlations between 

 depth and male proportion were not significant {P 

 > 0.05) for either the 1981-82 (r, = -0.10) or 1984 

 (Vg = -0.04) survey. 



For the stations sampled seasonally in 1981-82 

 (CB-1, CB-4, and CB-5), heterogeneity of male pro- 

 portion among stations was significant {P < 0.05) 

 during all four sampling periods (Fig. 3). In all cases, 

 male proportion was lowest at CB-5, highest at 

 CB-4, and intermediate in magnitude at CB-1. This 

 gradient of increasing male proportion was consis- 

 tent with the gradient of increasing percent fine- 

 grained sediment at the three stations (Table 1). The 

 mean abundance of English sole at the three sta- 

 tions rose from a minimum level in March to a peak 

 in May-July, and then declined through August to 

 a level in November that was similar to that in 

 March (Fig. 3). 



For both surveys combined, 64 (1.2%) English sole 

 were found to be in spawning condition. Twenty- 

 four of those individuals were collected during 

 March, to represent a spawning component of 14.8% 

 during that month. The remaining 42 individuals 

 were sampled between May and July to represent 



518 



