Hannah el al Length and age at maturity for Eopsctta pidani 



713 



bias in some of these data, arising from spawning aggrega- 

 tioii, argues that a difPerent sampling approach niiglit he 

 usetiil I'or trying to generate unbiased maturity data from 

 samples of the commercial catch. A third objective of our 

 study was to evaluate problems with available maturity 

 data from fishery samples in light of histological evalua- 

 tions of maturity stages and to reconmiend an alternative 

 approach for sampling the commercial catch for age and 

 length at maturity. 



Materials and methods 



Petrale sole were collected by trawl during August and 

 September 2000 by using a chartered commercial fishing 

 vessel from Newport, Oregon. Sampling was conducted 

 over a very limited geographic range to minimize any 

 spatial variation in the maturity ogive. Individual sam- 

 pling sites were selected to attempt to sample across the 

 available size range of petrale sole, based on the skipper's 

 experience and expertise. Fish were sorted at-sea accord- 

 ing to total length (±0.5 cm), and fifteen fish from each 

 1-cm size interval were retained for maturity sampling. 

 Upon returning to port, the fish were measured again and 

 sex was determined by dissection. An ovary was removed 

 and assigned a macroscopic maturity stage following the 

 criteria of Hagerman (1952). Only experienced samplers 

 were used to assign macroscopic maturity stages to mini- 

 mize any errors in visual staging. Otoliths were removed 

 from all female fish for subsequent age determination, and 

 one ovary was preserved for histological examination to 

 determine maturity. Ages were determined by using the 

 break-and-burn technique for sagittal otoliths (Chilton 

 and Beamish, 1982). Consistency of age determination 

 was evaluated by using a blind double reading. 



Ovaries were preserved in WA buffered formalin and 

 later transferred to 70% ethanol for storage. Tissue sam- 

 ples from the midsection of the ovary were then embedded 

 in paraffin, sectioned at 5 pm, and stained with Harris's 

 hematoxylin and eosin V. The samples were examined and 

 classified as mature or immature based on the presence or 

 absence of dark-staining yolk globules (vitellogenin). 



The diameter (pm) of the largest spherical nonatretic 

 oocyte in the most advanced oocyte stage from each of five 

 microscope fields was also measured and used to calcu- 

 late a mean maximum oocyte diameter (MMOD) for each 

 sectioned ovary (West, 1990; Nichol and Pikitch, 1994). 

 MMOD was compared for mature and immature fish. 

 A divergence in MMOD between mature and immature 

 fish was taken as evidence that the presence or absence 

 of dark-staining yolk globules was an accurate indicator 

 of maturity status. Samples from August and September 

 were compared by using this approach. The accuracy of 

 visual examination of ovaries was then evaluated by com- 

 paring the maturity status from visual examination with 

 status determined after sectioning, staining, and micro- 

 scopic examination. 



Logistic regression was used to fit sigmoid length-ma- 

 turity and age-maturity curves. The model fitted had the 

 general form 



P = gl'iu + ''l»l V( 1 -I- {'"'() + ''l«ll) 



•>i ' 



where p = the probability that a fish is mature in a given 

 length (cm) or age category x-^; and 



6,1 and 6, are parameters that define the shape and loca- 

 tion of the fitted curve. 



The predicted length or age at 50% maturity was calcu- 

 lated as 



L (or A)r,|, = -hjhy 



Maturity data collected by the Oregon Department of Fish 

 and Wildlife (ODFW) from the 1986-2000 commercial 

 trawl fisheries were also analyzed and compared with the 

 data collected in our study. The most recent petrale sole 

 stock assessment (Sampson and LeeM relied on maturity 

 data used for the 1993 assessment (Turnock et al.'*), which 

 in turn were derived from combined ODFW port sampling 

 data for the years 1986-91. We fitted a maturity-length 

 relationship to these data for comparison with the matu- 

 rity data from our study. 



We also summarized the ODFW data from 1986-91 and 

 1992-2000, by month and maturity status, to compare with 

 our data and assess their adequacy for determining length 

 at maturity for female petrale sole. All of these samples 

 were classified by a simple visual evaluation of maturity 

 stage. To examine the effect that a lack of immature fish 

 may have had on estimated maturity curves from these two 

 time periods, we split the data into groups, according to sea- 

 son. Samples from the months of November through Febru- 

 ary were classified as "fall-winter" samples. These samples 

 represented the seasonal time period when the maturity 

 status of petrale sole was most readily determined visually, 

 but also the time period when samples were most likely 

 to he from spawning aggregations. Samples from April 

 through September were classified as "spring-summer" 

 and represented the seasons in which fish were not aggre- 

 gated for spawning, but the time when maturity status was 

 most difficult to determine by inspection of the ovaries. 



Results 



Comparison of visual maturity determinations with 

 results obtained from microscopic evaluation showed that 

 maturity of petrale sole cannot be reliably determined by 

 simple visual inspection in August or September (Tables 

 1 and 2). For ovaries classified as immature by inspection 

 in August, 16.7% proved to be mature upon microscopic 

 evaluation. Of those classified initially as mature, 6.5% 

 showed no microscopic evidence of vitellogenesis. By 

 September, all of the ovaries classified by inspection as 

 mature, did prove to be vitellogenic. However, 27.9% of 

 those classified as immature, proved to be vitellogenic 

 after viewing the stained sections with a microscope. 



MMOD increased markedly in vitellogenic ovaries 

 between August and September, but in nonvitellogenic 

 ovaries the increase was minimal (Fig. 2). The resulting 

 divergence in MMOD and the absence of fish with MMOD 



