David M. Wyanski H. Scott Meister

Analytical Report on the Sex Ratio, Maturity, Reproductive Seasonality, and Annual Fecundity of Wreckfish, Polyprion americanus, off the Southeastern United States David M. Wyanski H. Scott Meister South Carolina Department of Natural Resources P. O. Box 12559 Charleston, SC 29422 Contact person: David M. Wyanski December 2002 This work represents partial fulfillment of the Marine Resources Monitoring, Assessment, and Prediction (MARMAP) Program contract (No. 50WCNF606013) sponsored by the National Marine Fisheries Service (Southeast Fisheries Center) and the South Carolina Department of Natural Resources.

The following individuals assisted with putting together this Data Report: Interpretation of Reproductive Tissue: David M. Wyanski, H. Scott Meister Data Analysis: David M. Wyanski, H. Scott Meister Writing: David M. Wyanski 2

ABSTRACT Otoliths and reproductive tissue were obtained during 1989-97 and 2000-02, primarily from specimens (n = 1617) landed in Charleston, South Carolina, by commercial fishermen that fished on a complex bottom feature of great topographic relief (100+ m) known as the Charleston Bump. Specimens from the Azores (n = 38), Madeira (n = 5), Bermuda (n = 5), and the Bahamas (n = 1) were also sampled. The life history of wreckfish in the western North Atlantic has not been thoroughly described because, except for the Bermuda fishery, wreckfish were rarely encountered in this region prior to 1987. All reproductive tissue samples were processed histologically and potential annual fecundity was estimated for 98 specimens. The overall sex ratio for wreckfish was significantly different from 1:1 in samples from both periods, 1989-93 and 2000-02. Males were smaller than females in three data sets. Females were predominant in fishery-dependent samples from 1989-93 (median TL = 1005 mm), whereas males were predominant in fishery-independent samples from 1989-93 (median TL = 983 mm) and fishery-dependent samples from 2000-02 (median TL = 985 mm). Median TL was smaller in samples where males predominated. Mature gonads were present in 60% of females at 751-800 mm, 57% at 801-850 mm, and 100% at larger sizes. The estimate of L 50 was 790 mm (gompit model; 95% CI = 733-820). Mature gonads were present in 40% of males at 651-800 mm and 100% at larger sizes. L 50 was not estimated because transition to maturity was abrupt. Based on the presence of hydrated oocytes and POFs, female wreckfish spawned from December through May, with a peak during February and March. The relationships between total fecundity and total length, fork length, whole body weight, and ovary-free body weight were highly significant. Wreckfish have determinate fecundity as early stages of yolked oocytes were absent at the beginning of the spawning season. Annual fecundity estimates, uncorrected for atresia, for 3

wreckfish 933-1280 mm ranged from 1,397,500 to 4,114,200 stage-3 yolked oocytes. 4

INTRODUCTION The wreckfish Polyprion americanus (Schneider 1801) has a global anti-tropical distribution, but it was rarely captured in the western North Atlantic until the late 1980s, when a bottom hook-and-line fishery that targets wreckfish developed on the Blake Plateau. Wreckfish occur on both sides of the Atlantic Ocean, on the Mid-Atlantic Ridge, on Atlantic islands and seamounts, and in the Mediterranean Sea, southern Indian Ocean, and southwestern Pacific Ocean (Heemstra 1986, Sedberry et al. 1994, Sedberry 1995). In the western Atlantic, they are distributed from Grand Banks (44 o 50' N) off Newfoundland (Scott and Scott 1988) to the Valdes Peninsula (43 o 30' S) in Argentina (Menni et al. 1981). Juvenile wreckfish (< 60 cm TL) are pelagic, and often associate with floating debris, which accounts for their common name. The absence of small pelagic and demersal wreckfish on the Blake Plateau has led to speculation that young wreckfish drift for an extended period, up to four years, in surface currents until reaching the eastern Atlantic, or perhaps that they make a complete circuit of the North Atlantic (Sedberry et al. 2001). Active adult migration is also possible as the frequent occurrence of European fish hooks in western North Atlantic wreckfish suggests migration across great distances. Adults are demersal and attain lengths of 200 cm (Heemstra 1986) and 100 kg (Roberts 1986). Wreckfish landed in the southeastern United States average 15 kg and 100 cm total length (Sedberry et al. 1994). Wreckfish have supported substantial fisheries in the eastern North Atlantic, Mediterranean, Bermuda, and the western South Atlantic, but concentrations of wreckfish adequate to support a fishery off the southeastern United States were not discovered until 1987. The fishery off the southeastern United States occurs over a complex bottom feature of great topographic relief (100+ m), known as the Charleston Bump, that is located 130-160 km southeast of Charleston, South Carolina, at 31 o 30 N and 79 o 00 W on the Blake Plateau (Sedberry et al. 2001). Fishing occurs at water depths of 450-600 m. Primary fishing grounds 5

comprise an area of approximately 175-260 km 2, characterized by a rocky ridge and trough feature with a slope greater than 15 o (Sedberry et al. 1994; 1999; 2001). Because this ridge system is at an angle transverse to the northerly-flowing Gulf Stream, it creates a semipermanent upwelling by deflecting the Gulf Stream offshore. The life history of wreckfish in the western North Atlantic has not been thoroughly described because, except for the Bermuda fishery (Sedberry et al. 1999), wreckfish were rarely encountered in this region prior to 1987. A preliminary study on sex ratio, sexual maturity, reproductive seasonality, and total fecundity was conducted by the Marine Resources Monitoring Assessment and Prediction (MARMAP) program in 1989-94 (Wyanski 1994). More recently, it has been determined that there is one genetic stock of wreckfish in the North Atlantic and Mediterranean (Sedberry et al. 1996). The objectives of Year 2 in the present study were: 1) to compare the population sex ratios in 1989-93 and 2000-02, 2) to estimate size at maturity, 3) to describe reproductive seasonality, and 4) to estimate potential annual fecundity for wreckfish on the Charleston Bump. The present study represents a second cooperative effort between MARMAP and the National Marine Fisheries Service (NMFS) laboratory in Beaufort, North Carolina, to investigate the life history of wreckfish. MARMAP biologists are conducting the study of reproductive biology and J. Potts of the NMFS is examining otolith sections to estimate the age of specimens. MATERIALS AND METHODS Sample acquisition and histology Otoliths and reproductive tissue were obtained during 1989-97 and 2000-02, primarily from specimens (n = 1617) landed in Charleston, South Carolina, by commercial fishermen that fished on the Charleston Bump. Specimens from the Azores (n = 38), Madeira (n = 5), Bermuda (n = 5), and the Bahamas (n = 1) were also sampled. Sedberry et al. (2001) describes the fishing gear and fishing methods for all these areas, except for the Bahamas. Specimens from the Charleston Bump usually represented the catch from one or two days of each fishing trip. 6

At the dock, each specimen was weighed (+ 0.01 kg) and total (TL) and fork lengths (+ 1 mm) were measured. The left sagitta was removed and stored dry in a coin envelope. The gonads were excised and packed in ice. In the lab, a posterior portion of gonad tissue was fixed in 10% seawater formalin buffered with marble chips for histological processing using the methods of White et al. (1998); tissue samples from the Azores were fixed in aqueous Bouin s solution, transferred to 70% ethyl alcohol, and then processed histologically using the methods of White et al. (1998). Whole ovaries for fecundity analysis were weighed (+ 1 g) and tissue samples from the anterior, middle, and posterior regions of one lobe were fixed in 10% seawater formalin. Histological sections were viewed under a compound microscope at 40-400 X to determine sex and reproductive state. Two readers independently assigned sex and reproductive state (Table 1), and if the assessments differed, the slide was viewed simultaneously by both readers. If the disagreement persisted the sample was excluded from data analyses. To assess size and age at maturity, specimens from both sides of the Atlantic were examined because very few immature individuals have been captured on the Charleston Bump (Sedberry et al. 2001). The recent finding that there is one genetic stock of wreckfish in the North Atlantic and Mediterranean (Sedberry et al. 1996) supported our decision to pool data. Specimens with developing, ripe, spent, or resting gonads were considered sexually mature. For females, this definition of sexual maturity included specimens with oocyte development at or beyond the cortical granule (alveoli) stage and specimens with beta, gamma, or delta stages of atresia. To ensure that females and males were correctly assigned to the immature and resting stages categories, the length-frequency histograms for immature, resting, and stages that represent certain maturity (i.e. developing, ripe, or spent) were compared. Size at 50% maturity (L 50 ) was determined with the PROBIT procedure (SAS Institute Inc. 1989). The LOGISTIC procedure was used to determine which model (gompit, logit, or probit) to use in the PROBIT procedure. 7

Females that possessed hydrated oocytes or postovulatory follicles (POFs) <48 h old were considered to be in spawning condition. POFs were assigned approximate ages according to criteria developed by Hunter and Goldberg (1980) for northern anchovy, Engraulis mordax. Most wreckfish on the Charleston Bump have been caught in areas with bottom temperatures of 9.0-16.3 o C (Sedberry et al. 1999), similar to the temperatures (13-19 o C) at which northern anchovy spawn (Hunter and Macewicz 1985). A gonadosomatic index (GSI) was calculated with the formula: GSI = (gonad weight/(fish weight gonad weight))*100. Sex ratios were examined by sample source, period and size class using a chi-square goodness of fit test to determine if the ratios differed from the expected 1:1. All statistical tests were conducted using SAS, and the results were considered significant if P was <0.05 (SAS Institute Inc. 1989). All measurements in the text refer to total length. Fecundity: Definitions of total fecundity (TF), potential annual fecundity, determinate fecundity, and indeterminate fecundity follow Hunter et al. (1992). Three stages of yolked (vitellogenic) oocytes, migratory nucleus oocytes, hydrated oocytes, and atretic oocytes, were identified in samples from formalin-preserved gonads (sensu Hunter et al. 1992). Densities of hydrated oocytes from five random locations (three samples per location) in the ovaries of four fish were compared to determine if stage-3 yolked oocytes were randomly distributed. Samples weighed approximately 150 mg and consisted of 200-400 oocytes. The effects of location and individual fish on density were assessed with a two-factor ANOVA. To estimate total fecundity, stage-3 yolked oocytes were counted in two 100-300 mg samples from random locations in the ovaries of specimens collected during 1992-96 and 2000-01. Total fecundity was calculated with the formula: TF=preserved ovary wt (g) * oocyte density (number of stage 3 oocytes/sample wt (g)). 8

The relationship between total fecundity and total length was described for four months (October through January) using least squares linear regression and an analysis of covariance (ANCOVA) was performed to examine the effect of time interval on total fecundity. RESULTS Histology The overall sex ratio for wreckfish was significantly different from 1:1 in samples from both periods, 1989-93 and 2000-02 (Tables 2, 3 and 4). Males were smaller than females in all three data sets. Females were predominant in fishery-dependent samples from 1989-93 (median TL = 1005 mm), whereas males were predominant in fishery-independent samples from 1989-93 (median TL = 983 mm) and fishery-dependent samples from 2000-02 (median TL = 985 mm). Median TL was smaller in samples where males predominated. Females were predominant in fishery-dependent samples from 1989-93 and 2000-02 during the five months (August through January) leading up to the spawning peak in February and March. Sex ratio was significant different from 1:1 during August, October, December, and January (Table 5). Males were more abundant at sizes <950 mm in 1989-93 fishery-dependent samples and sex ratio was significantly different from 1:1 in size classes between 801 and 900 mm (Table 2). At sizes >950 mm, females were more abundant and sex ratio was significantly different from 1:1 in size classes between 1001 and 1150 mm. Males were more abundant at sizes <1000 mm in 1989-93 fishery-independent samples and sex ratio was significantly different from 1:1 in size classes between 851 and 1000 mm (Table 3). At sizes >1050 mm, females were more abundant and sex ratio was significantly different from 1:1 in size classes between 1051 and 1100 mm. Males were more abundant at sizes <1000 mm in 2000-02 fishery-dependent samples and sex ratio was significantly different from 1:1 in size classes between 801 and 950 mm (Table 4). At sizes >1000 mm, females were more abundant and sex ratio was significantly different from 1:1 in size classes between 1101 and 1150 mm. 9

The near overlap in the length histograms for specimens of certain maturity and specimens that were resting and the minimal overlap in the histograms for immature and resting specimens indicated that reproductive tissue was correctly assigned to the immature and resting categories for males (Fig. 1a) and females (Fig. 1b). Most of the immature specimens (three of 18 males and seven of 21 females) were collected in the eastern North Atlantic. Mature gonads were present in 60% of females at 751-800 mm, 57% at 801-850 mm, and 100% at larger sizes (Table 6). The smallest mature female was 692 mm, and immature females were 576-831 mm. The estimate of L 50 was 790 mm (gompit model; 95% CI = 733-820). Mature gonads were present in 40% of males at 651-800 mm and 100% at larger sizes (Table 6). The smallest mature male was 661 mm, and immature males were 518-883 mm. L 50 was not estimated because transition to maturity was abrupt. Vitellogenesis in wreckfish oocytes is a prolonged process, four to five months in length, owing to their large size (1.4 mm dia. prior to hydration). Vitellogenic oocytes were prevalent by July, but the first signs of spawning did not occur until December (Fig. 2). Based on the presence of hydrated oocytes and POFs, female wreckfish spawned from December through May, with a peak during February and March. POFs were present in 38 specimens. In 36 of 38 specimens, the POFs were 24-48+ hr old. The sharp decline in the GSI between February and April confirms February and March as the spawning peak (Fig. 3). The highest percentages of ripe males occurred during December through May, which corresponded with the female spawning season (Fig. 4); however, males in spawning condition were collected throughout the year. The male spawning peak was also during February and March. Fecundity The abundances of earlier stages (1 and 2) of yolked oocytes relative to stage-3 oocytes were very small one to three months prior to the spawning season (Fig. 5A-D). By the beginning of the spawning season, earlier stages of yolked oocytes were absent (Fig. 5E,F), which was evidence of determinate fecundity. 10

There was no difference in the density of stage-3 yolked oocytes among five randomly selected locations (F=2.04, P=0.11, DF=4) in four specimens, which indicated that samples for estimating total fecundity can be taken from any location without bias. Total fecundity as a function of total length was similar in the four months (October through January) prior to the peak of the spawning season (Fig. 6), as indicated by the lack of differences in slopes (F=1.21, P=0.311, DF=3) and intercepts (F=0.51, P=0.677, DF=3) among months. Given the similarity in the monthly equations, data were pooled for the estimation of total fecundity. The relationships between total fecundity and total length, fork length, whole body weight, and ovary-free body weight were highly significant (Table 7). Total fecundity estimates, uncorrected for atresia, for wreckfish 933-1280 mm ranged from 1,397,500 to 4,114,200 stage-3 yolked oocytes. DISCUSSION Contrary to the findings of studies on wreckfish off New Zealand (Roberts 1989) and in the Mediterranean Sea (Commission of the European Communities 2001), males were notably smaller than females in western North Atlantic samples. The presence of sexual dimorphism in our samples may be the result of a larger sample size (>1300 vs. 200-325 specimens). Hermaphroditism does not explain the size difference between sexes because Polyprion species have gonad morphology that is typical of gonochorists (Roberts 1989). Other hypotheses such as differential growth and differential longevity should be investigated. We found that the largest specimens were usually females, as did Roberts (1989) and Commission of the European Communities (2001), whereas the smallest specimens in the present study were predominantly males, a trend not evident in the other two studies. Roberts (1989) reported a sex ratio of 1:0.86 (male:female, n = 339) in fishery-dependent catches off New Zealand, which is similar to the ratios in 1989-93 fishery-independent samples (1:0.61) and 2000-02 fishery-dependent samples (1:0.73) from the present study. The presence of sexual dimorphism requires random samples from the entire population when assessing sex ratio. We found that the length distribution of samples has an obvious impact on overall sex ratio. The sex 11

ratio in 1989-93 fishery-dependent samples favored females (1:1.77) because median TL was ca. 20 mm greater (Table 2) than that in two other data sets (Tables 3 and 4). It is possible that the change in sex ratio in fishery-dependent samples from 1:1.77 to 1:0.73 between 1989-93 and 2000-02 represents a reduction in the percentage of females in the population. The length frequency distribution in the 2000-02 fishery-dependent samples should be compared with that from landings data to determine if our samples are representative of the population. Results of maturity analyses in the present study supported an earlier conclusion based on submersible observations and landings data that nearly all fish on the Charleston Bump are large, mature fish (Sedberry et al. 1994). The estimate of L 50 (790 mm or 31 inches) for females is clearly in the left tail of length frequency distributions from 1992-98 fishery-dependent samples (Vaughan et al. 2001). Our estimate of L 50 for females was very similar (790 mm vs. 773 mm) to that reported by the Commission of the European Communities (2001) for the wreckfish population off Crete and Italy. Histological methods were used in both studies. In the present study, forty percent of males were mature at 651-700 mm (n = 5) and the smallest mature male was 661 mm, whereas off Crete and Italy the L 50 for males was 685 mm and the smallest mature male was 720 mm. That the estimate of L 50 for males was less than the size of the smallest mature male is likely due to a small sample size (n = 88). We chose to not estimate L 50 for males by applying a model because Trippel and Harvey (1991) recommends reporting the size interval with the first occurrence of >50% maturity if transition to maturity is abrupt. The spawning seasons of wreckfish on the Charleston Bump and in the Mediterranean Sea (Commission of the European Communities 2001) appear to be very similar. Vitellogenic oocytes are prevalent during summer and fall. Although collections were not made during the months of spawning in the Mediterranean, the presence of post-spawning females in March and April in the Mediterranean indicates that spawning occurs during winter in both regions. Wreckfish off New Zealand appear to spawn in the winter as well. Based primarily on macroscopic observations, Roberts (1986) found that spent and resting females were prevalent in August and September, which corresponds to February and March in the northern hemisphere. 12

Our total fecundity estimates appear to represent estimates of potential annual fecundity because: 1) the abundances of earlier stages (1 and 2) of yolked oocytes relative to stage-3 oocytes were very small one to three months prior to the spawning season (Fig. 5A-D), and 2) earlier stages of yolked oocytes were absent at the beginning of the spawning season (Fig. 5E,F), indicating that potential annual fecundity (equivalent to total fecundity prior to the onset of spawning) became fixed prior to the spawning season. Species in which annual fecundity becomes fixed prior to the spawning season are said to have determinate fecundity (Hunter et al. 1992). Wreckfish in the Mediterranean Sea also have determinate fecundity (Commission of the European Communities 2001), as frequency distributions of oocyte diameter exhibit a welldefined hiatus between yolked and unyolked oocytes. Estimates of potential annual fecundity in the two regions are very similar. Using the equation for the relationship between fecundity and length (y = 21208e 0.0442x ) for wreckfish off Crete, the fecundity of a 1050 mm TL specimen is 2,200,000 oocytes. The equation used in the present study produced an estimate of 2,300,000 oocytes for the same sized specimen on the Charleston Bump. Determining the spawning season of wreckfish on the Charleston Bump partly fulfilled one aspect of the Sustainable Fisheries Act, namely that fishery management councils identify Essential Fish Habitat (EFH). EFH includes those waters and substrate necessary to fish for spawning, feeding, or growth to maturity (Schmitten 1999). Ongoing research within MARMAP and at NMFS Beaufort is focusing on the age and growth of wreckfish so that these aspects of life history can be meshed the results of the present study. An isotope method is being used to validate the age of wreckfish, which will hopefully provide a better understanding of increment pattern in otoliths. 13

ACKNOWLEDGEMENTS We are indebted to Captains Paul Reese and Sam Ray for their willingness to allow us to sample their catches. Colleen Fennessey, Carlos Andrade, Brian Luckhurst, and Joel Carlin are thanked for making samples from the eastern Atlantic, Bermuda, and the Bahamas available to us. Kathy Grimball, Oleg Pashuk, and Donna Schlegel processed the histology samples. Todd Kellison, Andrea Johnson, and Jack McGovern assisted with the processing of fecundity samples. 14

LITERATURE CITED Commission of the European Communities, Directorate-General for Fisheries, XIV. 2001. State of the stocks of European wreckfish (Polyprion americanus). Final Report, 54 p. Heemstra, P.C. 1986. Family No. 165: Polyprionidae. Page 509 in M.M. Smith and P.C. Heemstra, editors. Smith s Sea Fishes, 6 th edition. Springer-Verlag, Berlin. Hunter, J.R. and S.R. Goldberg. 1980. Spawning incidence and batch fecundity in northern anchovy, Engraulis mordax. Fish. Bull. 77:641-652. Hunter, J.R. and B.J. Macewicz. 1985. Rates of atresia in the ovary of captive and wild northern anchovy, Engraulis mordax. Fish. Bull. 83:119-136. Hunter, J.R., B.J. Macewicz, N.C.H. Lo, and C.A. Kimbrell. 1992. Fecundity, spawning, and maturity of female Dover sole Microstomus pacificus, with an evaluation of assumptions and precision. Fish. Bull., U.S. 90:101-128. Menni, R.C., H.L. Lopez and M.L. Garcia. 1981. Campanas de investigacion qesquere realizadas en el mar Argentino por los B/I "Shinkai Maru" y "Walter Herwig" y el B/P "Marburg", anos 1978 y 1979. Seccion IV. Lista comentada de las especies de peces colectadas durante la campana V del B/I "Shinkai Maru" en el Mar Argentino (25/8-15/9/1978). Contribucion. Instituto Nacional de Investigacion y Desarrollo Pesquero (Argentina). Mar del Plata 383: 267-280. Roberts, C.D. 1986. Systematics of the percomorph fish genus Polyprion Oken, 1817. Ph.D. thesis, Victoria University of Wellington. Roberts, C.D. 1989. Reproductive mode in the percomorph fish genus Polyprion Oken. J. Fish Biol. 34:1-9. SAS Institute Inc. 1989. SAS/STAT User's Guide, Version 6, 4th edition, Vol. 2, Cary, NC: SAS Institute Inc., 846 p. Schmitten, R.A. 1999. Essential fish habitat: opportunities and challenges for the next millennium. Pages 3-10 in L.R. Benaka, editor. Fish habitat: essential fish habitat and rehabilitation. American Fisheries Society, Symposium 22, Bethesda, Maryland. Scott, W.B., and M.G. Scott. 1988. Atlantic fishes of Canada. Can. Bull. Fish. Aquat. Sci. 219:731 p. Sedberry, G.R. 1995. Aspects of the biology and management of wreckfish, Polyprion americanus, in the western North Atlantic. Pages 105-116 in A.R. Lima, editor. 13 a semana das pescas dos AHores, 15 a 21 MarHo, Horta-Faial, Relat\rio 1994. Portugal. 15

Sedberry, G.R., and eight authors. 1999. Wreckfish Polyprion americanus in the North Atlantic: fisheries, biology, and management of a widely distributed and long-lived fish. Pages 27-50 in J.A. Musick, editor. Life in the slow lane: ecology, and conservation of long-lived marine animals. American Fisheries Society, Symposium 23, Bethesda, Maryland. Sedberry, G.R., J.L. Carlin, R.W. Chapman, and B. Eleby. 1996. Population structure in the pan-oceanic wreckfish, Polyprion americanus (Teleostei: Polyprionidae), as indicated by mtdna variation. J. Fish Biol. 49 (Supplement A):318-329. Sedberry, G.R., J.C. McGovern, and O. Pashuk. 2001. The Charleston Bump: an island of essential fish habitat in the Gulf Stream. Pages 3-24 in G.R. Sedberry, editor. American Fisheries Society, Symposium 25, Bethesda, Maryland. Sedberry, G.R., G.F. Ulrich, and A.J. Applegate. 1994. Development and status of the fishery for wreckfish (Polyprion americanus) in the southeastern United States. Proc. Gulf Carib. Fish. Inst. 43:168-192. Trippel, E.A., and H.H. Harvey. 1991. Comparison of methods used to estimate age and length of fishes at sexual maturity using populations of white sucker (Catostomus commersoni). Can. J. Fish. Aquat. Sci. 48:1446-1459. Vaughan, D.S., C.S. Manooch, III, and J.C. Potts. 2001. Assessment of the wreckfish fishery on the Blake Plateau. Pages 105-120 in G.R. Sedberry, editor. Islands in the stream: oceanography and fisheries of the Charleston Bump. American Fisheries Society, Symposium 25, Bethesda, Maryland. Wallace, R.A. and K. Selman. 1981. Cellular and dynamic aspects of oocyte growth in teleosts. Amer. Zool. 21:325-343. Wenner, C.A., W.A. Roumillat and C.W. Waltz. 1986. Contributions to the life history of black sea bass, Centropristis striata, off the southeastern United States. Fish. Bull., U.S. 84:723-741. West, G. 1990. Methods of assessing ovarian development in fishes: a review. Aust. J. Mar. Freshwater Res. 41:199-222. White, D.B., D.M. Wyanski, and G.R. Sedberry. 1998. Age, growth, and reproductive biology of the blackbelly rosefish from the Carolinas, U.S.A. J. Fish Biol. 53:1274-1291. Wyanski, D.M. 1994. The reproductive biology of wreckfish, Polyprion americanus, off the southeastern United States. MARMAP Analytical Report, 7 p. 16

FIGURE LEGENDS Fig. 1 Comparison of length frequencies of male (A) and female (B) wreckfish Polyprion americanus that were categorized as immature, certain maturity, or resting/ca oocytes. Specimens of certain maturity were developing, ripe, or spent. All specimens were examined histologically. CA = cortical alveoli. Fig. 2. Percentage composition of reproductive states, based on histological criteria, by month for 809 female wreckfish Polyprion americanus from the western North Atlantic. Number of specimens examined is above each bar. POF = postovulatory follicle. Fig. 3. Gonadosomatic index for female wreckfish Polyprion americanus from the western North Atlantic. GSI = (gonad weight/(fish weight gonad weight))*100. Number of specimens examined is adjacent to each point. Fig. 4. Percentage composition of reproductive states, based on histological criteria, by month for 757 male wreckfish Polyprion americanus from the western North Atlantic. Number of specimens examined is above each bar. Fig. 5. Frequency distribution of oocyte diameter for four oocyte stages (see Hunter et al. 1992) in six wreckfish Polyprion americanus collected during October, December, and January on the Charleston Bump. Stage 1, Stage 2, and Stage 3 refer to yolked oocytes. MNO = migratory nucleus oocyte. Fig. 6. Estimates of total fecundity in wreckfish Polyprion americanus relative to total length during four time intervals. Stage-3 yolked oocytes were counted in 98 fishery-dependent specimens captured on the Charleston Bump in 1992-94 and 2000-01. 17

200 180 160 140 a) Immature (n = 18) Certain maturity (n = 689) Resting (n = 26) Male Number 120 100 80 60 40 20 0 40 60 80 100 120 140 Total length (cm) 200 180 160 140 b) Immature (n = 21) Certain maturity (n = 623) Resting or CA oocytes (n = 140) Female Number 120 100 80 60 40 20 0 40 60 80 100 120 140 160 Total length (cm) 18

Female P. americanus Reproductive Seasonality 1989-2002 100 124 9 13 83 73 29 42 122 53 49 96 116 80 Percentage 60 40 Developing Ripe Spent Resting Cortical-alveoli POF 20 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month 19

Female P. americanus 16 14 Mean gonadosomatic index 12 10 8 6 4 2 108 9 11 75 42 29 43 113 53 45 95 126 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month 20

Male P. americanus Reproductive Seasonality 1989-2002 100 49 14 24 121 98 29 61 84 48 35 91 103 80 Percentage 60 40 Developing Ripe Spent Resting 20 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month 21

40 30 A) B) Stage 3 Stage 2 Stage 1 40 30 Stage 3 Stage 2 Number 20 October 1041 mm TL 20 October 1028 mm TL 10 10 0 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Oocyte diameter (mm) 0 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Oocyte diameter (mm) 40 30 C) D) December 1000 mm TL 40 30 December 969 mm TL Number 20 20 10 10 0 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 0 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 40 30 E) F) Stage 3 MNO 40 30 Stage 3 MNO Number 20 January 1330 mm TL 20 January 1280 mm TL 10 10 0 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Oocyte diameter (mm) 0 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Oocyte diameter (mm) 22

6e+6 Number of stage-3 yolked oocytes 5e+6 4e+6 3e+6 2e+6 1e+6 0 900 950 1000 1050 1100 1150 1200 1250 1300 Total length (mm) October (n=11) November (n=12) December (n=40) January (n=35) 95% CI 23

Table 1. Histological criteria used to determine reproductive state in wreckfish, Polyprion americanus. (see Hunter and Goldberg 1980; Wallace and Selman 1981; Hunter and Macewicz 1985; Wenner et al. 1986; West 1990). Reproductive state Male Female 0 Uncertain maturity 1 Immature (virgin) 2 Developing 3 Running ripe 4 Spent 5 Resting 7 Developing, recent spawn Inactive testes; unable to assess maturity; stage=1 or 5. Small transverse section compared to resting male; spermatogonia and little or no spermatocyte development. Development of cysts containing primary and secondary spermatocytes through some accumulation of spermatozoa in lobular lumina and ducts. Predominance of spermatozoa in lobules and ducts; little or no occurrence of spermatogenesis. No spermatogenesis; some residual spermatozoa in shrunken lobules and ducts. Larger transverse section compared to immature male; little or no spermatocyte development; empty lobules and ducts; some recrudescence (spermatogonia through primary spermatocytes) possible at end of stage. Inactive ovaries; previtellogenic oocytes only; unable to assess maturity; stage=1 or 5. Previtellogenic oocytes only, no evidence of atresia. In comparison to resting female, most previtellogenic oocytes <80 um (size is species dependent), area of transverse section of ovary is smaller, lamellae lack muscle and connective tissue bundles and are not as elongate, oogonia abundant along margin of lamellae, ovarian wall is thinner. See next page. Completion of yolk coalescence and hydration in most advanced oocytes; zona radiata becomes thinner. More than 50% of vitellogenic oocytes in alpha or beta stage of atresia. Previtellogenic oocytes only with traces of atresia possible. In comparison to immature female, most previtellogenic oocytes >80 um (size is species dependent), area of transverse section of ovary is larger, lamellae have muscle and connective tissue bundles, lamellae are more elongate and convoluted, oogonia less abundant along margin of lamellae, ovarian wall is thicker and exhibits varying degrees of expansion due to previous spawning. 8 Mature specimen, stage unknown 9 Unknown Mature, but inadequate quantity of tissue or postmortem histolysis prevent further assessment of reproductive stage. Postmortem histolysis or inadequate quantity of tissue prevent assessment of reproductive stage. See next page. Mature, but inadequate quantity of tissue or postmortem histolysis prevent further assessment of reproductive stage. Postmortem histolysis or inadequate quantity of tissue prevent assessment of reproductive stage. 24

Table 1. Continued. B Developing, recent spawn C Developing, recent spawn D Developing, recent spawn E Early developing, cortical alveoli F Developing, vitellogenesis G Final oocyte maturation Vitellogenic oocytes and postovulatory follicles <24 h old (sensu Hunter and Goldberg 1980). Formerly stage=7. Vitellogenic oocytes and postovulatory follicles 24-48 h old (sensu Hunter and Goldberg 1980). Formerly stage=7. Vitellogenic oocytes and postovulatory follicles >48 h old (sensu Hunter and Goldberg 1980). Formerly stage=7. Most advanced oocytes in cortical-alveoli stage. Formerly stage=2. Most advanced oocytes in yolk-granule or yolk-globule stage. Formerly stage=2. Most advanced oocytes in migratory-nucleus stage; partial coalescence of yolk globules possible. Formerly stage=2. 25

Table 2. Chi-square analyses of sex ratio by length interval for wreckfish, Polyprion americanus, in western Atlantic fishery-dependent samples from 1989-93. H O : male to female ratio is 1:1. TL = total length. Total length (mm) Males Females Male:Female X 2 Prob. H o 601-650 1 0 651-750 - - 751-800 1 0 801-850 10 1 1:0.10 7.36 0.007 Reject 851-900 28 3 1:0.11 20.16 0.001 Reject 901-950 48 31 1:0.65 3.66 0.056 Accept 951-1000 74 98 1:1.32 3.35 0.067 Accept 1001-1050 43 127 1:2.95 41.51 0.001 Reject 1051-1100 15 89 1:5.93 52.65 0.001 Reject 1101-1150 2 33 1:16.5 27.46 0.001 Reject 1151-1200 0 10 1201-1350 - - 1351-1400 0 1 No length 9 17 Total 231 410 1:1.77 49.99 0.001 Reject Median TL 964 1027 Overall median TL = 1005 mm Table 3. Chi-square analyses of sex ratio by length interval for wreckfish, Polyprion americanus, in western Atlantic fishery-independent samples from 1989-93. H O : male to female ratio is 1:1. TL = total length. Total length (mm) Males Females Male:Female X 2 Prob. H o 801-850 4 0 851-900 14 1 1:0.07 11.27 0.001 Reject 901-950 41 6 1:0.15 26.06 0.001 Reject 951-1000 43 15 1:0.35 13.52 0.001 Reject 1001-1050 23 23 1:1.00 0.00 1.000 Accept 1051-1100 8 17 1:2.12 3.24 0.072 Reject 1101-1150 0 8 1151-1200 0 1 1201-1250 - - 1251-1300 0 1 1301-1450 - - 1451-1500 0 1 No length 2 10 Total 135 83 1:0.61 12.40 0.001 Reject Median TL 962 1035 Overall median TL = 983 mm 26

Table 4. Chi-square analyses of sex ratio by length interval for wreckfish, Polyprion americanus, in western Atlantic fishery-dependent samples from 2001-02. H O : male to female ratio is 1:1. TL = total length. Total length (mm) Males Females Male:Female X 2 Prob. H o 651-700 0 1 701-750 1 0 751-800 5 0 801-850 14 1 1:0.07 11.27 0.001 Reject 851-900 48 4 1:0.08 37.23 0.001 Reject 901-950 70 20 1:0.29 27.78 0.001 Reject 951-1000 81 58 1:0.72 3.81 0.051 Accept 1001-1050 39 58 1:1.49 3.72 0.054 Reject 1051-1100 29 39 1:1.34 1.47 0.225 Reject 1101-1150 4 18 1:4.50 8.91 0.003 Reject 1151-1200 0 11 1201-1250 0 2 1251-1300 0 1 1301-1350 - - 1351-1400 0 1 No length 4 2 Total 295 216 1:0.73 12.21 0.001 Reject Median TL 952 1021 Overall median TL = 985 mm Table 5. Chi-square analyses of sex ratio by month for wreckfish, Polyprion americanus, in western Atlantic fishery-dependent samples from 1989-93 and 2001-02. H O : male to female ratio is 1:1. TL = total length. Month Males Male:Female X 2 Prob. H o Median TL Females (mm) January 23 51 1:2.22 10.60 0.001 Reject 1028 February 13 9 1:0.69 0.73 0.394 Accept 982 March 8 11 1:1.38 0.47 0.491 Accept 1011 April 40 38 1:0.95 0.05 0.821 Accept 980 May 81 58 1:0.72 3.81 0.051 Accept 1004 June 26 25 1:0.96 0.02 0.889 Accept 1000 July 46 33 1:0.72 2.1 0.144 Accept 962 August 69 104 1:1.51 7.08 0.008 Reject 996 September 48 49 1:1.02 0.01 0.919 Accept 1005 October 17 34 1:2.00 5.67 0.017 Reject 1015 November 77 86 1:1.12 0.5 0.481 Accept 995 December 64 109 1:1.70 11.70 0.001 Reject 996 Total 512 607 27

Table 6. Percentage of mature specimens by length interval for female and male wreckfish, Polyprion americanus, in western North Atlantic and eastern North Atlantic samples from 1989-2002. Western Atlantic specimens were from fishery-dependent and fisheryindependent sources. Eastern Atlantic specimens (n = 43) were from fishery-dependent sources. Specimens in the developing, spawning, spent, or resting stages were considered mature. All specimens were examined histologically. n = number of specimens. Total length Female Male (mm) n = 810 n = 741 % ( n ) % ( n ) 501-550 --- --- 0 ( 1) 551-600 0 ( 7) 0 ( 6) 601-650 0 ( 3) 0 ( 7) 651-700 20 ( 5) 40 ( 5) 701-750 0 ( 2) 100 ( 8) 751-800 60 ( 5) 100 ( 22) 801-850 57 ( 7) 100 ( 54) 851-900 100 ( 10) 99 (120) 901-950 100 ( 60) 100 (163) 951-1000 100 (187) 100 (197) >1000 100 (524) 100 (158) 28

Table 7. Linear regression coefficients for the relationship between total fecundity (TF; number of stage-3 yolked oocytes) and total length (mm), fork length (mm), and whole and ovary-free weight (g) in wreckfish, Polyprion americanus. Specimens were collected during October, November, December, and January. *P<0.0001. Linear equation TF = a + bx X a SE b SE Adjusted r 2 F N Range Total length -5906833 1069366 7828.9 1013.6 0.377 *59.66 98 933-1,280 mm Fork length -5858823 1014282 8121.6 1003.1 0.400 *65.56 98 889-1,235 mm Whole weight -325921 219759 136.1 10.9 0.613 *156.28 99 11,000-40,596 g Ovary-free weight -301811 243130 145.6 13.0 0.558 *124.86 99 9,858-36,314 g 29