McDonough et al : Fecundity and spawning season of Mugil cepha/us 



823 



In the southeastern United States the spawning season 

 lasts from two to five months depending on the coastal 

 area involved (Jacot, 1920; Broadhead, 1956; Anderson, 

 1958; Arnold and Thompson, 1958; Stenger, 1959; Dindo 

 and MacGregor, 1981; Greeley et al., 1987; Render et al., 

 1995; Hettler et al., 1997). Striped mullet are considered 

 isochronal spawning fishes (Greeley et al., 1987; Render et 

 al., 1995), i.e. they have synchronous gamete development 

 and individuals spawn all their reproductive material at 

 once or in batches over a very short time period (days, as 

 opposed to weeks). There have been limited observations of 

 offshore spawning activity (Arnold and Thompson, 1958), 

 and few examples of eggs and larvae collected offshore (An- 

 derson, 1958; Finucane et al., 1978; Collins and Stender, 

 1989). Collins and Stender (1989) concluded that striped 

 mullet spawn in and around the edge of the continental 

 shelf off the coasts of North Carolina, South Carolina, 

 Georgia, and the east coast of Florida (an area often re- 

 ferred to as the South Atlantic Bight) and have a protracted 

 spawning season from October to April. This spawning 

 season contrasts with that estimated by most other stud- 

 ies (Jacot, 1920; Broadhead, 1956; Anderson, 1958; Arnold 

 and Thompson, 1958; Stenger, 1959; Dindo and MacGregor, 

 1981; Greeley et al., 1987; Render et al., 1995; Hettler et al, 

 1997 ). These studies based their estimates on the reproduc- 

 tive condition of migrating adults and the subsequent re- 

 cruitment of juvenile fish back into coastal estuaries — not 

 on actual data on the offshore presence of striped mullet 

 larvae. Female mullet have been shown to mature at two 

 to three years of age at a size range from 230 mm to 350 

 mm standard length (Thomson, 1951, 1963; Greeley et al., 

 1987). Determination of spawning activity in mullet has 

 been estimated by using gonadosomatic indices (Dindo and 

 MacGregor, 1981; Render et al., 1995), examination of oo- 

 cyte size and maturity stage (Stenger, 1959; Greeley et al., 

 1987 ), and by the presence of enlarged, developing ovaries 

 in migrating fish (Jacot, 1920; Anderson, 1958). 



The purpose of the present study was to develop size- or 

 age-related estimates of fecundity in striped mullet from 

 South Carolina estuaries. These fecundity estimates were 

 determined in order to develop models for estimating 

 potential fecundity from catch data, such as length and 

 weight data. In addition, other indicators of reproductive 

 activity such as gonadosomatic indices and oocyte size were 

 examined to provide information on the duration of the 

 spawning season. Potential fecundity estimates can give a 

 barometer of reproductive potential based on morphologi- 

 cal information from catch curve data and size-frequency 

 distributions. 



Materials and methods 



Striped mullet were collected monthly from January 1998 

 through December 2000 by using a randomly stratified 

 sampling regime within three different estuarine systems 

 along the South Carolina coast: (Ashepoo-Combahee- 

 Edisto (ACE) Basin, Charleston Harbor, and the Cape 

 Remain estuary, Fig. D.The female striped mullet used for 

 fecundity estimates were collected from October through 



February each year because these were the only months 

 when fecund females were present. Fecund female striped 

 mullet were defined as specimens in the tertiary stage of 

 vitellogenesis with oocyte diameters greater than 400 ^m 

 and that had gonadosomatic indices greater than 5.0. The 

 vitellogenic stage for each specimen used in the fecundity 

 estimates was confirmed histologically. Fish were captured 

 during daylight ebbing tides with water levels ranging 

 from 0.3 to 2.0 meters, and the majority (67.8%) were 

 caught during late ebb. The fish were primarily caught 

 with a 184-m trammel net with an outside stretch mesh of 

 350 mm and an inside stretch mesh of 63.5 mm, although 

 a few of the 1999 samples (5) and the 2000 samples (3) 

 were obtained by using a cast net 1.8 m in diameter and 

 equipped with 10-mm mesh. Eight fecund females were 

 also captured by using an electroshock boat in the fresh- 

 water and low-salinity areas of the Cooper River (one of the 

 three rivers that make up the Charleston Harbor Estuary) 

 in October 2000. The areas included in the present study 

 have been sampled on a monthly basis since 1991 with 

 trammel nets by the Inshore Fisheries Group of the South 

 Carolina Department of Natural Resources as part of a 

 gamefish monitoring program. During this period repro- 

 ductively developing striped mullet of both sexes were 

 generally observed from October through February and 

 were presumably heading offshore to spawn (Jacot, 1920; 

 Anderson, 1958; Arnold and Thompson, 1958). Male striped 

 mullet that were reproductively developed were easy to 

 discern because they were usually leaking milt and were 

 not analyzed further. All other fish were brought back to 

 the laboratory and eviscerated in order to determine sex 

 and to collect ovaries for reproductive analysis. All of the 

 samples were kept on ice and were generally examined 

 within twenty-four hours of capture. 



Standard morphometric measurements included total 

 length (TL) in mm, fork length in mm (FL), and standard 

 length in mm (SL), total weight (TW) in grams, ovary 

 weight (OW) in grams, sex, and maturity. Body weight 

 (BW) was calculated as total weight minus ovary weight 

 {BW=TW-OW). Saggital otoliths were removed for aging. 

 A small section of the posterior end of the ovaries, at the 

 junction of the two lobes, was also removed for histo- 

 logical examination. The whole ovaries were fixed in 10% 

 seawater-buffered formalin and the histological sample 

 was fixed in 10% neutral-buffered formalin. Histological 

 samples were processed by using standard procedures for 

 paraffin embedding and sectioning (Humason, 1967). The 

 sections were dried on slides and stained by using standard 

 haematoxylin and eosin-Y staining techniques (Humason, 

 1967). Examination of the histological sections for maturity 

 stage was done by using a compound microscope at lOOx 

 magnification. Each histological section was evaluated by 

 two separate readers to determine agreement on maturity 

 stage. If there was a discrepancy in maturity staging for 

 any specimen, the discrepancy was either resolved by the 

 two readers or that specimen was not used in the analysis. 

 There were no discrepancies between readers on any of 

 the reproductively developing females. Maturity was as- 

 sessed according to a modified version of the schedule used 

 by Wenner et al. (1986) adapted to work with isochronal 



