682 



Fishery Bulletin 92(4), 1994 



1957; Warlen, 1982; Lewis and Judy, 1983; Setzler- 

 Hamilton, 1987). However, the fact that during both 

 years we found spawning-phase females (stages 3— 

 5) in Chesapeake Bay from July through October and 

 that regressing and resting females — which probably 

 had completed spawning for the season — were col- 

 lected in the estuary indicates that the role of estu- 

 aries as additional spawning areas for Atlantic 

 croaker may be more important than previously 

 thought. Other sciaenids that were believed to be strict 

 marine spawners have also been reported to spawn 

 occasionally in estuaries (Castello, 1985; Johnson and 

 Funicelli, 1991). However, whether significant spawn- 

 ing of Atlantic croaker occurs in Chesapeake Bay or 

 other estuaries requires further investigation. 



The fact that spawning-phase Atlantic croaker 

 have not previously been found in Chesapeake Bay 

 can be attributed, at least in part, to their pattern of 

 multiple spawning and indeterminate fecundity. 

 Because in multiple spawning fishes the processes 

 of hydration, ovulation, and spawning usually occur 

 within a matter of hours (Hunter and Macewicz, 

 1985a; Brown-Peterson et al., 1988), the probability 

 of collecting gravid or running-ripe females is much 

 lower compared with other maturity stages. Addi- 

 tionally, contrary to what happens with total spawn- 

 ers, partially spent ovaries contain oocytes ranging 

 from primary growth to advanced yolked stage, mak- 

 ing the macroscopic identification of post-spawning 

 fish very difficult (Hunter and Macewicz, 1985a). In 

 most cases, we were not able to distinguish macro- 

 scopically between fully developed and partially 

 spent ovaries, and this also may have been a prob- 

 lem with previous studies (e.g. Wallace, 1940). 



Diel periodicity of spawning could also influence 

 the occurrence of hydrated females in samples from 

 different gears. The thousands of adult Atlantic 

 croaker examined by Haven (1957) and Wallace 

 ( 1940) were collected primarily from Chesapeake Bay 

 commercial pound nets and haul seines, which are 

 usually fished in the predawn or early morning hours 

 (Reid, 1955; Chittenden, 1991). During the rest of 

 the day and through most of the night, fish remain 

 alive in the pound-head or in the seine-bag until the 

 nets can be fished (emptied), usually during slack 

 water, and between 4:00 and 9:00 am. We hypoth- 

 esize that during this period Atlantic croaker spawn 

 within the nets at their usual spawning time of dusk 

 (Holt et al., 1985). Females collected from these nets 

 the following morning would probably show little or 

 no signs of spawning and be identified as "develop- 

 ing" (Wallace, 1940) or fully developed (this study). 

 However, contrary to what happens with pound nets 

 and haul seines, gill nets usually kill the fish within 

 a short time after capture. Females undergoing hy- 



dration or ovulation, especially those caught a few hours 

 before dusk, would die before they finished spawning, 

 and the presence of hydrated oocytes in the ovaries 

 could be recorded. This may explain why we observed 

 hydrated or recently spent females only in gillnet col- 

 lections. A similar pattern has also been observed for 

 weakfish, Cynoscion regalis, which, like Atlantic 

 croaker, spawn primarily between 6:00 and 9:00 pm. 1 



Size and age at maturity 



Our estimates of size and age at maturity are gener- 

 ally below values previously reported for Atlantic 

 croaker in the Chesapeake Bay and mid-Atlantic 

 regions. Disagreement with previous reports can be 

 attributed to three main factors: 1) failure of at least 

 some studies (Wallace, 1940; Morse, 1980) to sample 

 small, young fish from fishery-independent sampling 

 programs; 2) the inclusion of samples collected from 

 a period when resting (reproductively inactive) fish 

 were present in the estimation of the proportion of 

 mature fish by size or age; and 3) disagreement with 

 previous estimates of age at maturity probably re- 

 flects problems with age-determination methods pre- 

 viously used for Atlantic croaker. Because of the dif- 

 ficulty in distinguishing resting and immature go- 

 nads, estimates based on samples pooled over the 

 entire spawning season or during a period when rest- 

 ing fish were present (e.g. Wallace, 1940; Morse, 

 1980) are probably biased towards larger sizes or 

 older ages. Hunter et al. ( 1992 ) found that estimates 

 of L 50 for Dover sole were higher when females were 

 taken during the spawning season than when they 

 were sampled before spawning began. They sug- 

 gested that estimates of length or age at first matu- 

 rity should always be based on samples collected prior 

 to the onset of spawning, when postspawning females 

 with highly regressed ovaries are rare. However, for 

 species like Atlantic croaker, which show individu- 

 ally asynchronous gonadal maturation, sampling 

 before the onset of spawning will not prevent the 

 occurrence of prespawning, resting fish. To avoid this 

 problem we used only fish collected in September, 

 when no resting or developing stages occurred, to 

 estimate size and age at first maturity. Finally, dis- 

 agreement with previous estimates of age at matu- 

 rity probably reflects problems with age-determina- 

 tion methods previously used for Atlantic croaker. 

 The use of length frequencies (Welsh and Breder, 



1 Lowerre-Barbieri, S. K., M. E. Chittenden Jr., and L. R. Barbieri. 

 1993. The multiple spawning pattern of weakfish, Cynoscion 

 regalis, in the Chesapeake Bay and mid-Atlantic Bight, with a 

 discussion of annual fluctuations in reproductive output. Vir- 

 ginia Institute of Marine Science, Gloucester Point, VA 23062. 

 Unpubl. manuscr., 61 p. 



