Hunter et al : Fecundity, spawning, and maturity of Microstomus pacificus 



125 



atretic oocytes. A few females suffered substantial 

 losses in total fecundity, but such fish were rare and 

 they had little effect on population means. Histological 

 and anatomical evidence indicated that females with 

 a-atretic advanced yolked oocytes were more common 

 in central California than in Oregon waters. Atresia 

 might be more common in central California Dover sole 

 because bottom sediments are contaminated. Alter- 

 natively, females with atretic ovaries may be more com- 

 mon in central California waters because they are living 

 near the southern end of their range where food supply 

 and other habitat conditions may be less than optimal. 

 Both explanations seem equally plausible at present. 



The third assumption, that females used to estimate 

 potential annual fecutidity have not spawned in the cur- 

 rent reproductive year, would be rejected for females 

 taken in January through May. The assumption prob- 

 ably held for the females used to estimate annual fecun- 

 dity in November- December because only 2.9% of the 

 females from California and only 1% of the females in 

 Oregon showed any histological signs of past or immi- 

 nent spawning. The few females that showed histo- 

 logical signs of spawning were not used, of course, to 

 estimate annual fecundity. Spawning may have gone 

 undetected in some of the females used to estimate 

 fecundity since postovulatory follicles are eventually 

 resorbed. This does not seem likely for the November- 

 December case because the spawning season had just 

 begun and resorption is probably slow at the low tem- 

 peratures of Dover sole spawning habitat. 



Our fourth assumption, that all the oocytes that con- 

 stitute the potential annual fecundity were included in 

 our oocyte counts, is supported by two lines of evi- 

 dence. The first is that no positive correlation existed 

 between the mean diameter of the advanced oocytes 

 and total fecundity. Such positive correlations were 

 eliminated by excluding all ovaries in which the mean 

 diameter of the advanced oocytes was less than 

 0.86 mm. A positive correlation between diameter and 

 fecundity existed when all ovaries were considered 

 (range in mean diameter of the advanced oocytes, 

 0.71-1.04 mm). This is evidence that recruitment of 

 oocytes into the advanced class continued until the ad- 

 vanced stock was well separated from early vitellogenic 

 oocytes (stages 1 and 2, Fig. 6). The second source of 

 evidence is the form of the oocyte size-frequency 

 distribution. A prominent gap between stage-2 and 

 stage-3 oocytes existed when the mean diameter of 

 stage-3 oocjd;es was between 0.84 and 0.96mm (Fig. 

 6). The absence of significant numbers of oocytes in the 

 intervening diameter classes (0.55-0.65 mm) indicates 

 maturation of oocytes across this range either had 

 ceased or was proceeding at a very slow pace. We con- 

 clude that recruitment of significant numbers of 

 oocytes into the advanced stock probably ceases in 



Dover sole when the mean diameter of the advanced 

 stock is between 0.86 and 0.96 mm. 



Some authors working with other species (Hislop and 

 Hall 1974 on Melangius merlangus (L.), Horwood and 

 Greer Walker 1990 on Solea solea) consider all yolking 

 oocytes to comprise the potential annual fecundity. In 

 Dover sole this would mean that in addition to stage 

 3, the most advanced yolked oocytes, stages 1 and 2 

 would also be used to estimate annual fecundity. Such 

 broad criteria are acceptable if all oocytes that began 

 vitellogenesis ultimately become a part of the mature 

 stock of oocytes that are spawned. This was not the 

 case in Dover sole because oocytes in the early stages 

 of vitellogenesis (stages 1 and 2) occurred in nearly all 

 mature ovaries, including those in which some of the 

 batches had already been spawned. The fate and 

 dynamics of these small partially-yolked oocytes in ad- 

 vanced ovaries is uncertain; their numbers might either 

 decrease due to resorption, increase and become part 

 of next year's production, or remain in stable numbers 

 until later in the year. It would seem impractical to ad- 

 just estimates of potential annual fecundity based on 

 all vitellogenic oocytes for the fraction of those oocytes 

 which do not continue vitellogenesis. Therefore, we 

 believe use of the more mature yolked oocytes for 

 estimating the potential annual fecundity is preferable. 



An important implication of our discussions of the 

 third and fourth assumptions is that timing the sam- 

 pling of females is a critical element in estimating 

 potential annual fecundity: Sample too early in the 

 reproductive cycle and the ovaries are not sufficiently 

 mature; sample too late and spawning is prevalent. The 

 optimal time to sample Dover sole ovaries is when the 

 average diameter of the advanced stock is between 0.86 

 and 1.1mm (Fig. 12). When the diameter is less than 

 0.86 mm, the numbers of advanced oocytes are still in- 

 creasing (indicated by the t value for the diameter coef- 

 ficient in the fecundity equation. Fig. 12). When the 

 diameter exceeds 1.1mm, 20% or more of the females 

 show histological signs of past or imminent spawning, 

 and the assumption of no spawning cannot be safely 

 made. 



Spawning rates and reproductive energetics 



The spawning season of Dover sole was protracted with 

 postovulatory follicles occurring as early as December 

 and hydrated oocytes as late as May, indicating a 

 season of six months. This is a long season for a fish 

 of determinate fecundity, since typically they are high- 

 latitude species with short, 1-2 month spawning 

 seasons. Batch fecundity was low, averaging about 10 

 oocytes per gram female weight, except for the first 

 and last batch which average about 5 oocytes per gram. 

 Dover sole spawn about nine times during their 



