43.5 cm, 42.3 cm, and 38.3 cm in 1951, 1965, and 

 1982, respectively). Because fecundity is propor- 

 tional to size, this decline implies that the repro- 

 ductive capacity of the stock has decreased 

 (Bagenal 1973; Borisov 1979), provided that rela- 

 tive fecundity or size and age at first maturity 

 have not undergone compensatory changes. 

 Data on fecundity and state of maturity of Dover 

 sole off Oregon have not been collected since 

 1950 (Harry 1959). 



In this paper, we describe fecundity of Dover 

 sole from the Columbia area during the 1985-86 

 spawning season as a function of length, weight, 

 and age, and compare the relationship between 

 length and fecundity with that previously esti- 

 mated from fish collected between 1948 and 1950 

 in the same geographical area (HaiTy 1959). Size 

 and age at maturation are also assessed and com- 

 pared vdth the hmited information presented by 

 Harry (1959). 



Materials and Methods 



Samples for estimating fecundity were ob- 

 tained from commercial trawlers fishing off 

 northern Oregon (about lat. 46°N) during De- 

 cember 1985. Fish total length (TL, nearest mm) 

 was measured and both otoliths were removed 

 and stored in 50% ethanol for age determination. 

 Ovaries were preserved in a 10% phosphate-buf- 

 fered formaldehyde solution. Maturity stages 

 were assigned using macroscopic inspection of 

 ovaries and oocytes and applying criteria de- 

 scribed by Hagerman (1952). Comparable cri- 

 teria were used by Harry (1956). Because most 

 specimens were filleted prior to sampling, the 

 relationship between total length and ovary-free 

 body weight (nearest 0. 1 g) was established from 

 intact fish collected at the same time. Body 

 weight at length for filleted fish was estimated 

 from this relationship. Additional specimens 

 (collected from 44° to 45°N) were sampled in 

 December 1985 and January 1986 from proces- 

 sing plants in Newport, OR and used to describe 

 state of maturity. Total length was measured 

 and otoliths were removed. 



Oocyte counts for fecundity estimates were 

 made using the gravimetric subsampUng method 

 for MacGregor (1957), as described by Hunter et 

 al. (1985). Both ovaries preserved from each fish 

 were blotted dry and weighed to the nearest 0. 1 

 g. Three subsamples of oocytes were removed 

 from each fish (one each from the anterior, 

 middle, and posterior regions of one ovary, 

 either left or right). Subsamples were weighed 



to the nearest 0.1 mg, placed on microscope 

 shdes in 33% glycerin, and teased apart to form 

 one layer of oocytes. Minimum and maximum 

 diameters of mature, yolked oocytes were esti- 

 mated from each subsample. Subsamples 

 weighed between 30 and 80 mg, and contained 

 about 100-300 oocytes. 



Ovaries were advanced enough to allow dis- 

 crimination of large, mature oocytes from 

 smaller partially-yolked oocytes and non-yolked 

 oocytes by the unaided eye. Oocyte size-fre- 

 quency distributions were determined micro- 

 scopically from the ovaries of 10 fish (MacGregor 

 1957). Using a Zeiss Videoplan 11'^ image ana- 

 lyzer, and a dissecting microscope with camera 

 lucida attachment, approximately 250 oocytes 

 lying along transect lines etched in the micro- 

 scope slide were measured to the nearest 

 micron, and size-frequency distributions were 

 evaluated. The modal size group of advanced 

 oocytes was determined by visual inspection of 

 the size-frequency plots. An average minimum 

 size-threshold for oocytes associated with the 

 most advanced and distinct modes in these 10 

 plots was determined. Ovaries with advanced 

 oocytes that all exceeded this size-threshold 

 were used in estimating fecundity. Individual 

 fecundity was calculated by multiplying total 

 ovarian weight by mean number of advanced 

 oocytes per mg in the three subsamples. 



Age was determined for each fish used in esti- 

 mating fecundity and for a subsample of those 

 fish used for maturity assessment. The left oto- 

 lith was prepared and sectioned as described by 

 Boehlert and Yoklavich (1984). Annuli were 

 counted on a dissecting microscope at 80 x 

 magnification using reflected light and a black 

 background. Age was determined twice for each 

 fish, approximately one month apart, to estab- 

 lish precision of the age estimate. There was no 

 difference between the two age determinations 

 for each of 75 fish (paired t-test, P > 0.50). The 

 first estimate was used in further analyses. 

 There has been no age validation for Dover sole, 

 but similar methods have accurately aged other 

 long-lived species {Sebastes diploproa, Bennett 

 et al. 1982; A. fimbria, Beamish et al. 1983; S. 

 flavidus, Leaman and Nagtegaal 1987). 



Results 



To determine fecunditv, 97 ovaries were col- 



'Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



989 



