166 



Fishery Bulletin 91(1), 1993 



metrically (Bagenal & Braum 1978), using the hydrated 

 oocyte method (Hunter et al. 1985). A 0.2 g subsample 

 of fresh oocytes was taken from the middle of the right 

 ovary, and all hydrated oocytes in each subsample were 

 counted under a dissecting microscope at a magnifica- 

 tion of 24 x. 



Oocytes were separated from one another and the 

 ovarian membrane through a washing process. Each 

 ovary was slit longitudinally, turned inside out, and 

 held under vigorously flowing tapwater. This flushed 

 the oocytes out of the ovarian membrane and into a 

 0.01 mm mesh sieve, which was held beneath the ovary. 

 Oocytes collected in the sieve were again rinsed with 

 fully-flowing tapwater to help separate them from one 

 another. The whole procedure took 5-10 min per ovary 



After draining the water, oocytes were transferred 

 to containers where they were preserved in 2% neu- 

 trally-buffered formalin. This formalin concentration 

 was chosen because it was the lowest possible con- 

 centration that would ensure proper oocyte preserva- 

 tion while minimizing changes in oocyte size and 

 appearance. 



The equipment necessary for the washing process is 

 very basic. We used two standard faucets (2 cm diam- 

 eter), with flow rates of 133 and 286 mL/s, respectively. 

 Both faucets had sufficient hydraulic pressure to dis- 

 lodge oocytes of all stages from ovarian tissue. How- 

 ever, the faucet with the higher flow rate, and thus 

 greater water pressure, worked best. Any sieve with 

 mesh small enough to retain less-developed oocytes, 

 and deep enough to keep them from being flushed over 

 the edge during washing, can be used as a collecting 

 sieve. We used a sieve made from a piece of nylon 

 plankton net (0.01 mm mesh) inserted between two 

 sections of 10 cm diameter PVC pipe, with a depth 

 (from lip to the mesh layer) also of 10 cm (Fig. 1). 



Preserved oocytes were measured 3-4 mo after col- 

 lection and, again, 6-7 mo after collection. Samples 

 were stirred before oocytes were removed to reduce 

 bias due to settling differences caused by oocyte size 

 or density. Oocytes were then dipped out of the forma- 

 lin with a spoon and placed in a gridded petri dish. 

 The first 20 undamaged hydrated oocytes were mea- 

 sured along the median axis as described for fresh 

 oocytes. Oocyte damage, due to the washing process, 

 was evaluated by assessing the percentage of dam- 

 aged oocytes in subsamples of 50 hydrated oocytes from 

 each of 10 preserved samples. We considered as dam- 

 aged those oocytes which were partially collapsed and 

 thus not appropriate for diameter measurements. 



Batch fecundities were also estimated gravimetri- 

 cally from preserved samples (after 3-4 and 6-7 mo 

 preservation) using oocyte samples from the same 10 

 fish originally used to estimate batch fecundities from 

 fresh samples. Oocytes were stirred, decanted into a 



sieve, drained of formalin, and washed with tapwater. 

 Oocytes were removed from the sieve, spread on the 

 bottom of a petri dish, and blotted dry with tissue 

 paper. A 0.2 g subsample was then transferred to a 

 gridded petri dish. A small amount of tapwater was 

 added to keep the oocytes moist and to help distribute 

 them evenly over the bottom of the dish. 



One-way analysis of variance (ANOVA) was used to 

 evaluate differences between fresh and preserved 

 oocyte diameters and batch fecundity estimates. Indi- 

 vidual females were used as blocks to remove the ef- 

 fect of variation among females. To compare batch 

 fecundities based on fresh samples with those based 

 on preserved samples, it was important to evaluate 

 the within-ovary positional effect. This was necessary 

 because fresh oocyte samples were taken from the 

 middle of the ovary, whereas preserved samples came 

 from mixed areas (due to the washing process). Hy- 

 drated oocytes were counted in 0.2 g oocyte samples 

 taken from the anterior, middle, and posterior areas of 

 28 fresh ovaries. 



Mean oocyte shrinkage was calculated for each of 

 the 28 ovaries after 3-4 and 6-7 mo preservation. Mean 

 oocyte shrinkage was then plotted against mean fresh 

 hydrated oocyte diameter to evaluate whether oocyte 

 shrinkage was consistent over the size-range of hy- 

 drated oocytes. 



All data were analyzed using statistical methods 

 available through the Statistical Analysis System (SAS 

 1988). Model assumptions were evaluated by exami- 

 nation of residuals (Draper & Smith 1981). Batch fe- 

 cundity data was log ln -transformed to meet the as- 

 sumption of homogeneity of variances. 



10 cm 



PVC pipe 



0.01 mm mesh 



Figure 1 



Schematic representation of the sieve used to collect weakfish 

 Cynoscion regahs oocytes dislodged during the washing 

 process. 



