Population Biology and Life History of the North American Menhadens, Brevoortia spp.

DEAN W. AHRENHOLZ

Introduction sumers, feeding on phytoplankton, or Geographic Ranges secondary consumers, feeding on zoo­ Menhaden are members ofthe world­ plankton, or both. Many clupeids are Reintjes (1969) summarized the geo­ wide family Clupeidae, one ofthe most in turn prey for various piscivorous graphic ranges for the four menhaden important families of fishes both eco­ predators through virtually their entire species. Atlantic menhaden are seasonal­ nomically (Hildebrand, 1963), and eco­ lives. Life history patterns for this fam­ ly found from Nova Scotia, Can., to logically. Clupeids are characteristically ily of fishes include species which can southeastern Florida, near West Palm very numerous and form large, dense complete their entire life cycle in either Beach. Gulf menhaden range from schools which enhance our ability to fresh or marine waters, or are anadro­ southwestern Florida, near Cape Sable, harvest them. Many of the species are mous species, or marine migratory (estu­ to Veracruz, Mex. Yellowfin menhaden filter feeders, being either primary con- arine dependent) species. overlap the ranges of all three other The large-scaled menhadens, the At­ menhaden species and are found from lantic menhaden, Brevoortia tyrannus, Cape Lookout, N.C., to the Mississippi and the Gulf menhaden, B. patronus, River Delta. Finescale menhaden over­ have received considerable attention in lap the ranges of both the Gulf and ABSTRACT-Four recognized species oj fishery science research due to theirlarge yellowfin menhaden, and are found from menhaden, Brevoortia spp. , occur in North population sizes and resulting economic just east of the Mississippi River Delta Americanmarine waters: Atlanticmenhaden, and ecological importance. The small­ (Turner, 1971) to Campeche, Mex. B. tyrannus; Gulf menhaden, B. patronus; scaled menhadens, the yellowfin men­ The numbers of Gulf menhaden rela­ yellowfin menhaden. B. smithi;andfinescale haden, B. smithi, and the finescale men­ tive to numbers of yellowfin menhaden menhaden, B. gunteri. ThreeoJthemenhaden species are known toJorm two hybrid types. haden, B. gunteri, are less numerous become reduced proceeding southward Members oj the genus range from coastal and have received far less consideration on the Gulf of Mexico coast of Florida. waters oj Veracruz, Mex.. to Nova Scotia, in the scientific literature. The contrast in There appears to be a similar distribu­ Can. Atlantic and Gulf menhaden are ex­ relative importance is quite marked. On tion pattern for relative numbers of At­ tremely abundant within their respective ranges and support extensive purse-seine one extreme, the purse-seine reduction lantic and yellowfin menhaden pro­ reduction (to fish meal and oil) fisheries. All fishery (to fish meal and oil) for Gulf ceeding southward along the Atlantic menhaden species are estuarine dependent menhaden was the largest U.S. fishery by coast of Florida. The coastal area be­ through late larval and juvenile stages. De­ weight from 1963 through 1988, and tween West Palm Beach and Miami, Fla., pending on species and location within the Atlantic menhaden purse-seine reduction where menhaden are relatively rare range, spawning may occur within bays and sounds to a substantial distance offshore. landings, currently one-third to two­ (Dahlberg, 1970), geographically sep­ Menhaden are consideredto befilter-Jeeding, thirds thosefor Gulfmenhaden, werethe arates the Atlantic menhaden from the planktivorous omnivores as juveniles and largest for the U. S. from 1947 to 1962. Gulfmenhaden, as well as apparenteast­ adults. Menhaden eggs, immature develop­ On the other extreme, finescale menha­ ernand western populations ofyellowfin mental stages, and adults are potentialprey Jor a large anddiverse number ojpredators. den are apparently not directly sought by menhaden. North American menhadens, including two any recognized fishery, and yellowfin A large amount ofhybrid introgression hybrids, are hosts Jor the parasitic isopod, menhaden (and their hybrids) are only occurs between Atlantic and yellowfin Olencira praegustator, and the parasitic cope­ harvested by specialized bait fisheries on menhaden on the Atlantic coast of pod, Lemaeenicus radiatus. Althoughthedata both coasts of Florida. The following is Florida, and Gulf and yellowfin men­ are quite variable, a dome-shaped Ricker function is frequently used to describe the a general description of the population haden on theGulfcoast ofFlorida. Areas spawner-recruitment relationshipJor Atlan­ biology and life history of these four with pure strains ofyellowfin menhaden ticand Gulfmenhaden. Each oJthese species North American menhaden species. are yetto be defined. As the relative den­ is treatedas a single stock with respect to ex­ sity of Gulf menhaden decreases pro­ ploitation bythepurse-seine reductionfishery. Dean W. Ahrenholz is with the Beaufort ceeding southward, the number of Gulf Estimates oj instantaneous natural (other) Laboratory. Southeast Fisheries Science Center. mortalityrates are O. 45JorAtlanticmenhaden National Marine Fisheries Service. NOAA, x yellowfin menhaden (B. patronus x and 1.1 Jor Gulfmenhaden. Beaufort, NC 28516-9722. B. smithi) hybrids increases along with

53(4). 1991 3 Table 1.-0istinguishing and comparative characteristics of North American coastal menhadens (modified from pure strains of yellowfin. For example, Dahlberg, 1970). Turner (1969) reported that collections ofmenhaden from PanamaCity to Cedar Large-scaled menhaden Small·scaled menhaden Keys, Fla., consisted of94 % Gulfmen­ Character B. tyrannus B. patronus B.smith; B. gunteri haden and 6% yellowfin menhaden, Frontal groove Complete Complete Absent Absent while samples from farther south, Tam­ Lateral spots Usually present Usually present Absent Absent above and below above and below paBaytoCapeSable, Fla., were7% Gulf the level of the level of menhaden, 56 % yellowfin menhaden, shoulder spot shoulder spot Ventral fin Middle rays and Inner rays equal Inner rays about Inner rays about and 37 %Gulf x yellowfin hybrids. Het­ sometimes inner to or longer than one-half to two­ one-half to two- tler (1968) reported on two collections rays equal in length outer rays thirds length of thirds length of made along the southern Gulf coast of to outer rays fin fin Scale Pointed, length Pointed Rounded tip, Rounded tip, Florida; one near Naples consisted of pectinations 1 medium or long shorter shorter 17% Gulf menhaden, 9% yellowfin Body mucus 2 Copius Copius Sparse Sparse menhaden, and 74% Gulf x yellowfin Flesh' Soft Soft Firm Firm Ovarian color Yellow Yellow White ? hybrids, and the other from near Sanibel Lateral scale rows 43-53 (40-58) 42-48 57-73 (54-80) 65-72 (60-76) Island consisted of5 % Gulf menhaden, Opercular Prominent Prominent Faint or absent Faint or absent 54% yellowfinmenhaden, and41 %Gulf striations (12-31) (13-25) (0-15) (0-18) yellowfin hybrids. A similar situation Predorsal scales 35-44 (33-46) 29-37 (28-39) 39-51 (37-56) 40-49 (39-52) x Vertebrae 46-48 (44-49) 44-46 (43-47) 44-45 (43-46) 42-43(41-43) apparently exists on the east coast of Ventral scutes 31-34 (29-34) 29-31 (28-32) 30-32 (29-34) 28-30(27-31) Florida with the distributions ofAtlantic 1 Older adults. and yellowfin menhaden and the Atlan­ 2 Fresh specimens. tic x yellowfin hybrids; for example, the menhaden gill-net fishery in Indian River, Fla., is dominated by yellowfin two large-scaled menhaden species can shoulder spot, and have larger and fewer menhaden and the Atlantic x yellowfin be separated from the two small-scaled scales. The small-scaled menhadens hybrids (Dahlberg, 1970). species by a variety of characteristics lack the former characteristics and have (Table 1). Fresh specimens can be sep­ smaller and more numerous scales. Gulf Species Characteristics arated simply by feel, as the large-scaled menhaden have a deeper (more convex) Menhaden are generically distin­ menhadens have large amounts of body body shape and fewer predorsal scales, guished from other clupeids by their mucus and relatively soft flesh, while the vertebrae, and ventral scutes than their relatively large heads, pectinated scales, small-scaled species have relatively small Atlantic congener (Table 1, Fig. 1,2). absence ofteeth (beyondjuvenile stages), quantities ofbody mucus and their flesh The yellowfin menhaden can be separ­ and by their dorsal fin being over the in­ is firm. Additionally, the large-scaled ated from the finescale menhaden by the terval between the pelvic and anal fins species possess a frontal groove, acces­ yellowfin'sgreaternumber ofvertebrae (Reintjes, 1969; Hildebrand, 1963). The sory lateral spots beyond the large and ventral scutes, and relatively smaller

Figure I.-Adult Atlantic menhaden, 250 mm FL (J. W. Reintjes photo).

4 Marine Fisheries Review Figure 2.-Adult Gulf menhaden, 167 mm FL (R. B. Chapoton photo).

Figure 3.-Adult yellowfin menhaden, 300 mm FL (J. W. Reintjes photo).

head (Table 1, Fig. 3,4). More detailed crossing with either parental population Hybrids of B. gunteri X B. patronus descriptions are available from Dahlberg will be predominantly by male hybrids, (finescale X Gulf menhaden) and B. (1970) and Hildebrand (1963). Dahlberg as they dominate the hybrid population. gunteri X B. smithi (finescale x yellow­ (1970) also provides divergent char­ A self-sustaining population ofhybrids is fin menhaden) have not been reported. acteristics between the Atlantic and Gulf unlikely due to the preponderance of Although the ranges ofthe three species populations of yellowfin menhaden. males. Hettler (1968) found no female overlap, yellowfin and finescale men­ The morphological and morpho­ hybrids (B. patronus X B. smithi), while haden are not abundant in the area of metrical appearances of the large­ Turner (1969) reported finding 4 females overlap (the Mississippi Delta region). scaled menhaden and yellowfin men­ out of390 hybrids examined. Dahlberg Except for the southeastern Texas coast, haden hybrids are intermediate to those (1970) discovered one female hybrid (B. fine scale menhaden are apparently not forthe parents (Dahlberg, 1970) (Fig. 5). tyrannus X B. smithi) from an unknown abundant in U. S. Gulf coastal waters The presence ofa gradient ofcharacter­ number of hybrids examined on the where Gulf menhaden predominate. istics between the parental types sug­ Atlantic coast, and found no females Definitive studies on finescale menhaden gests back-crossing also occurs. Back­ among Gulf hybrids. are lacking.

53(4), 1991 5 Figure 4.-Adult finescale menhaden, 320 mm FL (R. B. Chapoton photo).

General Life Cycle eXlstmg knowledge of migration and ward. By June, the population is redis­ distribution was further strengthened tributed from Florida to Maine. Even Menhaden are estuarine dependent, by an analysis of the age and length though some Atlantic menhaden migrate marine migratory species. Spawning distributions ofAtlantic menhaden in the north and south along the U.S. Atlantic generally occurs during the cooler landings (Nicholson, 1971), and finally coast, because the fish distribute them­ months in the marine environment, from results ofan internal, ferromagnetic selves on the basis of size and age, the and larvae undergo early growth and tagging program (Dryfoos et al., 1973; movement actually represents a seasonal development at sea. About 1-2 months Kroger and Guthrie, 1973; Nicholson, expansion and contraction of the Atlan­ later, those larvae that have been trans­ 1978). tic menhaden's range. ported shoreward enter estuarine bays, During summer, Atlantic menhaden Geotemporal aspects of spawning for sounds, and streams, and metamorphose are generally distributed from northern this species are closely associated with the into juveniles. Menhaden juveniles Florida to Maine. The adult population migratory behavior of the adults, and (young-of-the-year) normally reside in stratifies by age and size, with the older some degree of spawning activity is estuarineareas until the following fall or and larger individuals farther northward believed to occur during virtually every early winter when many migrate into and the younger and smaller fish in the month of the year. Some fish ripen and marine waters. Adults generally occur in southern half of the species' range. Al­ some spawning occurs in the more north­ nearshore oceanic waters and frequent­ though localized movements occur dur­ erly portions of the fishes range as the ly reside in large estuarine systems. ing summer, no major systematic move­ fish begin moving southward in Septem­ ment occurs until September, when the ber. Spawning continues with increasing Migratory Behavior more northerly portion ofthe population intensity as the fish move progressively and Spawning Season begins to migrate southward. By Decem­ farther southward in October and No­ ber, a significant portion of the adult vember. Spawning intensity is believed Atlantic Menhaden population that was north ofChesapeake to peak in waters offthe North Carolina Early hypotheses ofthe migratory be­ Bay during summer has moved south­ coast during winter. Spawning con­ haviorofAtlantic menhaden were based ward to waters off the North Carolina tinues, but with decreasing levels of in­ upon observations ofschools appearing coast. These fish are followed by large tensity as the fish move northward the and disappearing along the U.S. Atlan­ numbers ofjuvenile (young-of-the-year) following spring and early summer. Sup­ tic coast, and from the examination ofthe menhaden, which have recently emi­ porting evidence for these conclusions age and size composition of catches grated from nursery areas farther north. was obtained earlier by Higham and among fishing ports along the U. S. Atlan­ Usually by late January, menhaden Nicholson (1964), subsequently by Ken­ tic coast (June and Reintjes, 1959). An schools disappear and schools disperse dall and Reintjes (1975), and later by analysis ofthe frequency and distribution from nearshore surface waters of North Judy and Lewis (1983). Atlantic men­ ofpurse-seine sets contributed additional Carolina. During March or early April, haden are believed to spawn in oceanic information with respect to the timing schools ofadult menhaden reassemble in waters over much of the continental of migrations (Roithmayr, 1963). The coastal waters and move rapidly north­ shelf, and in bays and sounds in Long

6 Marine Fisheries Review Figure 5.-Adult yellowfin menhaden (upper), 280 nun FL, and Atlantic x yellowfin menhaden hybrid (lower), 285 mm FL (J. W. Reintjes photo).

Island waters and northward (Nelson et ofthe agelsize structure ofthe spawning u.S. Gulfcoast in nearshore waters. Be­ aI., 1977; Ferraro, 1980b). Evidence for population. For example, with the bulk ginning in October, they move offshore recent spawning activity was based on of the spawning stock in recent years into deeper waters for winter. Roithmayr the presence ofmenhaden larvaeand eggs consisting of late age-2 fish, relatively and Waller (1963) reported that during in plankton samples. Evidence ofimmi­ less spawning activity would have been summer Gulf menhaden occurred in nent spawning was also provided by expected in the New England and Mid­ depths of 1-8 fathoms, while during the presence of near-ripe specimens in dle Atlantic areas as compared to the winter months they were found in 4-18 fish samples obtained from commercial 1950's when a broader and stronger age fathoms east and west ofthe Mississippi purse-seine landings. However, spawn­ structure was more extant in the popula­ Delta, and at 20-48 fathoms in a smaller ing has not been directly observed in tion (Ahrenholz et al., 1987b). area east and northeast of the Delta. the marine environment, and running­ Results of tagging studies failed to ripe females are rarely captured. Gulf Menhaden identify any east-west component of The relative magnitude of temporal Gulfmenhaden do not exhibitan exten­ annual migration for Gulf menhaden spawning activity within and between sive migratory pattern. During late spring (Pristasetal., 1976; Kroger and Pristas, geographic regions is in part a function and summer they distribute along the 1975); however, multiple-year juvenile

53(4),1991 7 tag-recovery data indicated a tendency by February 8 about 25 %offemales were sexually mature during their second year for Gulfmenhaden from the eastern and ready to spawn. Hettler's (1970) speci­ of life, while from two-thirds to nearly western extremes oftheir range to move mens from the Atlantic coast were taken all are sexually mature by theend oftheir toward the center oftheir range with age from the Indian River, Fla., in Febru­ third year (age 2) (Higham and Nichol­ (Ahrenholz, 1981). ary. Dahlberg (1970) concluded that son, 1964; Lewis et al., 1987). Female The spawning season for Gulf men­ the spawning season for yellowfin men­ Gulf menhaden about 150 mm FL and haden was determined by observations of haden was February and March on both larger are generally sexually mature by larvae, gonadal development, and pres­ the Atlantic and Gulf coasts of Florida. the spawning season (Lewis and Roith­ ence ofeggs in plankton samples. Spawn­ His conclusion for the Gulfcoast was at mayr, 1981), while the smallest sex­ ing has not been directly observed. From least in part based on Hettler's (1968) ually mature female Atlantic menhaden observations of the occurrence of lar­ collection just north of Naples during are at least 180 mm FL (Lewis et al., vae in Lake Ponchartrain, La., Suttkus mid-March oftwo ripe female yellowfin 1987). (1956) concluded that spawning prob­ menhaden and Turner's (1969) collection Age and size at maturation data is ably began in October and ceased in of ripe females during February and limited for the small-scaled menhadens. February; he presumed that this period March off the southern Gulf coast of Gunter (1945) observed a ripe female could fluctuate among years. Combs Florida. Spawning may occur as early as finescale menhaden 150 mm TL, and a (1969) concluded from a histological November, as Houde and Swanson ripe male 125 mm TL. The smaller ofthe examination of ovaries that spawning (1975) collected yellowfin menhaden two ripe female yellowfin menhaden that ranged from October through February eggs during this month from Atlantic Hettler (1968) found was 186mmFL. No or early March. Christmas and Waller waters off the Florida coast. standing stock ova counts for either (1975), after a literature review and an species of small-scaled menhaden are examination of plankton samples col­ Finescale Menhaden available. lected from much ofthe GulfofMexico, There is no evidence from which to Atlantic and Gulf menhaden are con­ concluded that spawning " ... for the deduce any systematic seasonal migra­ sidered to be multiple (fractional or inter­ most part ... " occurred from October tion by the finescale menhaden other mittent) spawners (Higham and Nichol­ through March. Shawetal. (1985a)pre­ than the notation ofan apparent seasonal son, 1964; Combs, 1969). As noted by sented arguments and evidence for an shift of larger finescale menhaden be­ Combs (1969), the fishes' ovaries could even more protracted season. tween Texas bays (Gunter, 1945). Like not contain all the developing ova ifthey Spawning areas have been determined the yellowfin menhaden, the finescale matured at the same time. Thus, ova by noting the geographic collection sites menhaden appears to occur more in es­ mature and are spawned in batches over where Gulf menhaden eggs were taken. tuarine or nearshore areas. Gunter (1945) a protracted spawning season. Based on their own collections and the referred to it as a bracldsh-water form, as The potential number ofova produced work ofFore (1970) and Turner (1969), opposed to the more saline Gulf men­ by an individual female during a spawn­ Christmas and Waller (1975) concluded haden, although this species was not ing season has been determined (esti­ that Gulf menhaden spawn from near formally described until 3 years later. mated) by counting the standing stock of shore to 60 miles offshore along the en­ Gunter (1945) discovered a ripe male advanced oocytes in Atlantic menhaden tire U. S. Gulf coast. during February and a ripe female dur­ (Higham and Nicholson, 1964; Dietrich, ing the latter part of March, and noted 1979; Lewis etal., 1987) and Gulfmen­ Yellowfin Menhaden that the spawning season was probably haden (Suttkus and Sundararaj, 1961; Adult yellowfin do not appear to dis­ from midwinter to early spring. He also Lewis and Roithmayr, 1981). For this play any systematic, annual migratory observed post-larval finescale menhaden technique to provide a reasonable esti­ behavior. Dahlberg (1970) referred to from January to May. Simmons (1957) mate of true annual fecundity, the num­ them as " ... common near shore along reported that this species spawned in the ber ofova produced during a season must both Florida coasts throughout the year. " upper Laguna Madre of Texas during be annually determinate, like that of the He considered them an inshore or bay February. Given these observations, a multiple spawning Atlantic silverside, form (in contrast to the large-scaled spawning period ofNovember to March Menidia menidia (Conover, 1985), as menhadens). Some larger individuals are appears realistic. Both Simmons (1957) opposed to being annually indeterminate, occasionally found as far north as Cape and Gunter (1945) reported that spawn­ similar to the multiple spawning northern Lookout, N.C., during summer. ing occurs in inside (estuarine) Texas anchovy, Eng raulis mordax (Hunter and Spawning seasons and some spawning waters. Macewicz, 1985). While some workers areas have been identified by collecting (e. g. , Lewis et al., 1987) felt that deter­ specimens for artificial spawning and Maturation and Fecundity minate fecundity is likely for the Atlan­ rearing. For the Atlantic coast popula­ Gulf menhaden become sexually ma­ tic menhaden (and thus likely for the tion, Reintjes (1962) began sampling ture near the end of their second year other menhadens), this condition has near Sebastian, Fla., in November. He of life (age 1) (Lewis and Roithmayr, not been demonstrated, nor has batch noted ripening males in December, 1981). Bycomparison, only a small per­ fecundity been estimated for any species several ripening females in January, and centage of Atlantic menhaden become of menhaden.

8 Marine Fisheries Review 600 .------, Fecundity estimates currently used in spawner-recruitment analyses (when number of potential eggs produced is 500 used as a measure of spawning stock) are derived from the results of Lewis and Roithmayr (1981) for Gulf menha­ 400 Atlantic den (Fig. 6). Fecundity values for the Atlantic menhaden are results ofpooled o '0 300 data from Higham and Nicholson (1964), Vl "0 C Dietrich (1979), and Lewis et al. (1987)

Description and Development Ol--.L---'----'------'----+-----"_l--.L---'---+-----'------"------l_l--+---'---'------'------"-_I--.L--l.----.J of Immature Life Stages 140 190 240 290 340 Eggs Fork length (mm) Figure 6.-Potential number ofova (in thousands) as a function ofFL for Atlantic A description of the early life history menhaden (line) and Gulf menhaden (dashes). forms of Atlantic menhaden is given by Kuntz and Radcliffe (1917). These authors collected a developmental series ofripe adults through eggs, embryos, lar­ distinguishable to species with morpho­ ported hatching times of 46 hours from vae, and juveniles during summer from logical characteristics (Table 2). fertilization for yellowfin menhaden Woods Hole Harbor, Martha's Vine­ Egg hatching time varies as a function eggs held at temperatures of 18.5° to yard, and Nantucket Sound. They de­ of temperature and species. Kuntz and 19.0°C. Additional data on hatching time scribed the eggs as spherical in shape, Radcliffe (1917) reported incubation obtained from the study reported by highly transparent with a thin, horny time for Atlantic menhaden eggs as less Reintjes (1962) are given by Hettler egg membrane and a relatively wide than 48 hours. Ferraro (1980a) devel­ (1968) as 46 hours at 18°C, 34 hours perivitelline space. Each egg contained oped a temperature-dependent empirical at 21 DC, and 26 hours at 26°C. Hettler a single oil globule. Their recorded egg equation from which temporal estimates (1968) further reported that the yellow­ dimensions are summarized in Table 2. ofduration for any stage ofembryolog­ fin x Gulf menhaden eggs hatched in Descriptions of eggs from other spe­ ical development, including hatching, about 38-39 hours when held at 19.5 to cies of menhaden followed a number could be obtained. Reintjes (1962) re­ 21.5°C. Hettler (1984) reported Gulf ofyears later. Reintjes (1962) described yellowfin menhaden eggs obtained on the Atlantic coast of Florida from both planktonic sampling and artificial fer­ Table 2.-Comparative characteristics 01 North American coastal menhaden eggs by species and source.

tilizations. Hettler (1968) described Species and Egg Yolk Oil globule Source of yellowfin x Gulf menhaden eggs from source diameter (mm) diameter (mm) diameter (mm) eggs the Gulfcoast ofFlorida, obtained by arti­ B. tyrannus ficial cross fertilization. Houde and Fore Kuntz and Radcliffe (1917) 1.4-1.6 0.9 0.12-0.14 Planktonic Jones et al. (1978) 1.30-1.95 0.90-1.20 0.11-0.17 ? (1973) described Gulf menhaden eggs Hettler (1984) 1.54-1.64 0.82-0.95 0.20-0.23 Laboratory reared from planktonic collections. An addi­ B. patronus tional description of yellowfin menha­ Houde and Fore (1973) 1.04-1.30 0.08-0.20 Planktonic Hettler (1984) 1.18-1.34' 0.95 (0.05)' 0.16-0.22' Laboratory den eggs obtained from the Atlantic coast reared ofFlorida was given by Houdeand Swan­ B.smithi Reintjes (1962) 1.21-1.48 0.77-1.04 0.05-0.18 Planktonic son (1975). Hettler (1984) described eggs Reintjes (1962) 1.15-1.30 0.77-0.95 0.07-0.16 Artificially obtained for laboratory-spawned Atlan­ spawned Houde and Swanson (1975) 1.21-1.34 0.80-1.19 0.12-0.17 Planktonic tic and Gulf menhaden (Table 2). Powell B. gunteri No data and Phonlor (1986) indicate that Atlan­ B. smithi x B. patronus tic menhaden eggs tend to be larger than Hettler (1968) 1.05-1.18 0.98 Artificially those of Gulf menhaden for a particular spawned set ofconditions; however, due to dimen­ , Combined resulls from two spawning series. , Range for spawning with larger yolks not given, mean and one standard deviation shown; range for spawning with sional overlap, menhaden eggs are not smaller yolks 0.66-0.79 mm.

53(4),1991 9 menhaden eggs hatched in 40-42 hours at 19-20°C. Larvae Larval development through the pre­ juvenile stages are described by Kuntz and Radcliffe (1917) and Lewis et al. (1972) for Atlantic menhaden, by Hettler (1984) for Gulfmenhaden, and by Houde and Swanson (1975) for yellowfin men­ haden. Additional information on early larval yellowfin menhaden is given by Reintjes (1962) and Hettler (1970). The size ofmenhaden larvae at hatch­ ing is thought to be a function ofegg size (Powell and Phonlor, 1986). Observed sizes at hatching ranged from 2.6 mm to Figure 7.-111us­ about 3.7 mm SL (Houde and Swanson, trations of At­ 1975; Hettler, 1984; Powell and Phonlor, Iantic menhaden 1986). The smallest individuals were larva (27 mm Gulfmenhaden and the largest, Atlantic TL), prejuvenile menhaden, with yellowfin menhaden (32mmTL),and juvenile (64 mm about midrange. TL); upper, mid­ Thelarvae are relatively undeveloped dle, and lower, upon hatching. Themouth is not formed, respectively, the eyes are unpigmented and thus non­ from Lewis etal. (1972). functional, and the fin rays are undevel­ oped (Houde and Fore, 1973; Reintjes, 1962). Depending on temperature, lar­ val menhaden begin feeding within 2-6 average rate of0.30 mm day-lover 60 not be as critical a life-history event for days. Most of the yolk is absorbed dur­ days for ocean-sampled Gulfmenhaden these species as it is for the large-scaled ing the prefeeding developmental period, larvae. species. but some may remain after the onset of Atlantic menhaden larvae ranging feeding (Houde and Swanson, 1975). from 14 to 34 mm FL (Reintjes and Juveniles Once the yolk sac is absorbed, the lar­ Pacheco, 1966) enter estuarine nursery When menhaden larvae undergo vae are slender and rodlike. areas at about 45-60 days of age (Nel­ metamorphosis to the juvenile stage, The subsequent rate of growth for son et aI., 1977). Larvae are reported they haveall thecharacteristics ofan adult each species depends on temperature as entering estuaries in New England except for sexual maturity (Fig. 7). Dur­ and food availability. Hettler (1984) during May through October, in the ing transformation (prejuveni1e stage), reported that larval Gulf menhaden U.S. Middle Atlantic from October to they undergo a substantial increase in growth averaged 0.30 mm day-l for June, and along the U.S. South Atlantic relative body depth and weight, while the first 90 days of rearing. He also coast from November to May (Reintjes achieving only a slight increase in length. reported that growth of yellowfin men­ and Pacheco, 1966; Wilkins and Lewis, The pronounced difference in relative haden larvae averaged 0.36 mm day-l 1971). body proportions is shown graphically through 32 days (Hettler, 1970). Houde The length of the oceanic period for for Atlantic menhaden by Lewis et and Swanson (1975) observed larval larval Gulf menhaden is estimated at al. (1972), and for Gulf menhaden by yellowfin menhaden growth of0 045 mm 6-10 weeks (Deegan and Thompson, Deegan (1986). Length during the pre­ day-l for the first 20 days of rearing, 1987). Estuarine immigration was ob­ juvenile period varies between in­ albeit at a higher temperature than Het­ served from late October through April, dividuals and species, but is between 30 tler (1970). Additional comparative and larvae ranged in size from 10 to 32 and 40 mm TL for Atlantic menhaden growth values were obtained by adjusting mm TL (Fore, 1970; TagatzandWilkins, (Lewis et aI., 1972; June and Carlson, the exponential expressions given by 1973). 1971). Gulfmenhaden metamorphoseat Powell and Phon10r (1986). Growth for Systematic observations oflarval im­ a slightly smaller size, with complete their Atlantic menhaden was about 0 Al migration for small-scaled menhadens transformation by 28-30 mm SL (Sut­ mm day-l and for their Gulf menha­ are unavailable. Since some spawning tkus, 1956). An even smaller size was den, 0.33 mm darl for about 18 days presumably occurs near or in some estu­ reported for yellowfin menhaden (lab­ of rearing. Warlen (1988) reported an arineareas, oceanic larval transport may oratory-reared) by Houde and Swanson

10 Marine Fisheries Review Table 3.-Reported ranges in lenglh of Juvenile men­ (1975), with the transformation com­ haden near the end of estuarine nursery habitation. Trimodality in length frequency distribu­ plete between 20 and 23 mm SL. Gunter Numbers in parentheses are approximate conversions tions is sometimes observed and could from lolal lenglh 10 fork lenglh using paramelers from (1945) observed post-larval finescale Jorgenson and Miller (1968). result from a large portion of the mi­ menhaden as small as 21 mm TL. gratory spawners moving even farther Range of Month In addition to the more apparent ex­ Species FLin mm sampled Source south, or from a cessation in spawning by the overwintering fish, or from differ­ ternal changes in body shape, significant B. tyrannus 38-171 September Krogeretal. (1974) internal morphological changes occur B. patronus 38-110 October Unpublished 1 ential survival of larvae or juveniles in B. patronus (78-103) August Tagatz and during metamorphosis as well. Gill Wilkins (1973) winter. rakers increase in number, length, and a.smith; 63-88 August Unpublished A substantial variance in mean size of overall complexity (June and Carlson, B. gunter; (74-94) "1-yearo old" Gunter (1945) olderjuvenilesexists among years. This 1971). The functional morphology of 1 Range from tagging records over several years; NMFS, can be partly due to density-dependent SEFC Beaufort Laboratory, Beaufort, N.C. the resulting elaborate branchial bas­ 'Range tram tagging records, Turnbull Creek, Fla., 1971; growth, as size-at-age data was shown ket is described by Friedland (1985). NMFS, SEFC Beaufort Laboratory, Beaufort, N.C. to be inversely related to year class size, Intestine length increases dramatically, at least as early as the estuarine growth and a gizzard-like pyloric stomach and phase (Reish et al., 1985; Ahrenholz et

pyloric caeca develop (June and Carl­ Table 4.-Maximum sizes reporte~ foradultmenhaden by al.,1989). son, 1971). These changes are diagnos­ species. Values in parentheses are approximateconver­ A relatively wide range injuvenile size sionsfrom values reported, forpurposesof comparison. tic and necessary for a lifetime trophic Conversions 10 fork lenglh (FL) used equalions from has also been observed for Gulf menha­ Jorgensonand Miller(1968), whilelolallenglh(TL)10 sian· den (Table 3). Some of the broad range habit of filter-feeding, microphagous dard lenglh(5L) and vice versa used combined ralios (Iwo planktivory. large-scale species' values combined, and two small· in sizes is expected as a result of their scale species' values combined) from ranges in sludy Ultimate juvenile size achieved dur­ specimens reported by Hildebrand (1963). protracted spawning season. There is ing estuarine residence is a function of additional evidence for bimodality in favorable growth conditions and absolute Species TLmm FLmm SLmm Source the juvenile population, and hence sug­ age ofindividual fish. The observed size B. tyrannus 500 (419) (409) Hildebrand (1963) gestions oftwo spawning peaks (Tagatz B. patronus1 265 (223) (214) Hildebrand (1963) range of juvenile Atlantic menhaden is B.smithi,,2 (341) (281) 257 Christmas and and Wilkins, 1973). The bulk of a year quite large (Table 3), and is attributed Gunter (1960) class is 10-14 months old on their first B. gunter; 3 (351) (289) 264 Christmas and to the broad geographic and temporal Gunter (1960) designated birthday (by convention spawning range. Most ofa new yearclass January I). Density-dependent growth 1 Some larger individuals of B. patronus and B. smithi were is 10-18 months old on their first reported by Dahlberg (1970). It appears, however, that some during the juvenile stage is not prom­ of his reported standard lengths may be fork lengths. designated birthday (by convention, 2 Fish on Figure 3 is larger than from this earlier report. inant, if present at all, and Nelson and March 1). Hence, some individuals in a 3 Fish on Figure 4 is larger than from this earlier report. Ahrenholz (1986) could find no evi­ year class may be 8 months older than dence ofdensity-dependent growth from other individuals. an examination offishery landings size­ The spawning which initiates a year at-age data. However, Guillory and class begins offthe New England coast in spawning during the southern migra­ Bejarano (1980) reported evidence for September, then proceeds southward to tion and one from spawning during the density-dependent growth among mean the U.S. Mid-Atlantic and Chesapeake northern migration (McHugh et a!., lengths ofjuveniles sampled from several Bay coasts in October and November and 1959). Also, preliminary length-fre­ estuarine areas and subsequent catch­ subsequently to the oceanic waters offthe quency analyses ofjuveniles sampled in per-unit-effort (CPUE) data for age-l Carolinas. Spawning resumes as fish South Carolina waters indicate a fair fish in the purse-seine reduction fishery. move north in early spring and continues degree of unimodality from that area, Much narrower ranges in size were into summer. The longest temporal presumably from a relatively unbroken noted for juvenile finescale and yellowfin period between spawning origins of a spawning period. menhaden (Table 3). This may be due developing year class occurs in the most Some juvenile length-frequency dis­ to sampling limitations, as well as from northern waters and the least in the more tributions from northern coastal estuaries a relatively less protracted spawning southern. Hypothetically, this type of ofNorthCarolinaappearto have two, and season. spawning pattern should result in bi­ in some cases three, modes. The first modal length-frequency distributions in of these modes may be attributable to Age and Growth of Adults geographic regions with a detectable spawning in October or early November Relative to absolute size, Atlantic hiatus in spawning due to migration, and by sexually maturing age-2 and some menhaden are the largest of the genus, a single mode in the more southerly age-3 fish which summered in North Gulf menhaden are the smallest, with reaches of the migratory route. Some Carolina or Virginia waters (Wilkins and both small-scaled species being of in­ supporting evidence for these hypotheses Lewis, 1971). The second (and major) termediatesize(Table4). Similarly, the exists. A seasonal bimodality in length mode probably represents progeny from Atlantic menhaden is probably the most frequencies has been observed for young­ subsequent winter spawning by migra­ long lived (10-12 years) and the Gulf of-the-year in the Chesapeake Bay area, tory adults which summered in the U. S. menhaden the shortest (5-6 years). The presumably one mode resulting from Mid-Atlantic and New England waters. longevity ofthe small-scaled menhadens

53(4),1991 Jl is unknown, but probably is equal to 350 or greater than that for Gulf menhaden, and certainly less than that for Atlantic 300 menhaden. Size-at-age data, and hence rates of growth in years are only available for 250 700 large-scaled menhadens. The technique 600 for ageing Atlantic menhaden with scales E .s 200 was developed by June and Roithmayr .r:; 0, 500 (1960). Additional validation ofthe tech­ c ~ :§ 150 nique was provided by Kroger et al. -'" 400 E a Ol (1974). Gulf menhaden are aged with LL 'iii 300 ~ scales by criteria developed by Nicholson 100 and Schaaf (1978). Since the density-dependent growth 200 effect persists for several years in the 50 fishery dependent size-at-agedata for the 100 Atlantic menhaden, descriptive growth o 0 equations were estimated for each year 0.0 ~5 1.0 15 2~ 25 30 3.5 ~o ~5 5.0 5.5 6.0 ~5 class (Ahrenholz et al., 1987b). Com­ Age (years) parative growth curves for length and Figure 8. -Comparativefitted von Bertalanffy curves oftwo dissimilar sized Atlantic weight are given for a relatively large menhaden year classes (numbers offish), 1970 (small year class, solid curves), and year class (1975) and a small year class 1975 (large year class, dashed curves). Upper curves are fork lengths in millimeters, (1970) in Figure 8. Only one equation lower curves are weights in grams, as a function of age in years. was used to describe adult growth in Gulf menhaden in a stock assessment and population simulation study by and phytoplankton, but interestingly, Juneand Carlson, 1971). Darnell (1958) Nelson and Ahrenholz (1986) (Fig. 9). someofthe phytoplankton they consume found relatively large quantities of Trophic Relationships are an order of magnitude smaller than phytoplankton along with detritus and the smallest phytoplankton consumed some zooplankton in the guts of small Because the general morphology of during the larval phase (Chipman, 1959; juvenile Gulf menhaden. menhaden is similar, a high degree of similarity among species with respect to their roles as both predators and prey is assumed. Observations and study results are given here by the particular 300 300 species upon which they were made, but, in general, parallel conclusions should be possible for the other men­ 250 250 haden species.

Menhaden As Consumers 200 200 E From the first-feeding larval stage .§. § into the prejuvenile stage, Atlantic men­ .r: 1: C, 150 150 Ol haden selectively sight-feed on individ­ c 'iii ~ 3: ual planktonic organisms (Chipman, :£ <5 1959; June and Carlson, 1971). Govoni LL 100 100 etal. (1983) noted that small Gulfmenha­ den larvae feed heavily on larger phyto­ plankton (predominant!y dinoflagellates) 50 50 and some zooplankton. As the larvae grow, phytoplankton become less impor­ tant in the diet, and (larger) zooplankton, 0 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 55 especially copepods (all life stages) be­ come more important. After metamor­ Age (years) phosis, filter feeding omnivory becomes Figure 9. - Fitted von Bertalanffy curves ofmean size-at-age for Gulfmenhaden, the rule. Juveniles consume zooplankton solid curve is fork length in millimeters, dashed curve is weight in grams.

12 Marine Fisheries Review Adult Atlantic menhaden stomach to bea function offish size, the abundant In estuarine and marine waters of the contents examined by Peck (1893) con­ 40-90 mm FL juveniles present in estu­ U. S. Atlantic and Gulfcoasts, menhaden sisted ofphytoplankton (especially dino­ arine systems during spring and summer, juveniles and adults are potential prey for flagellates and diatoms); zooplankton; probably take advantage ofthe predom­ a large number of species and sizes of greenish, brownish, or yellowish organic inating primary productivity occurring piscivorous fishes (Sykes and Manooch, "mud" or amorphous matter; anddetri­ within the nanoplankton segment of the 1979). The relative degree ofpredation tus. He also examined stomach contents plankton community. will again be a product ofthe coincidence from juveniles and found the same type On the other extreme, Friedland et al. in space and time of the potential pred­ oforganic material that was in adult fish. (1984) noted maximum filtration effi­ ators and prey, and their relative sizes. Because of the menhaden's elaborate, ciency occurred for objects about 100 For the most part, prey selection among highly specialized gill-rakers for filter­ J.lm in diameter for the 138 mm FLjuve­ menhaden predators appears to be pre­ feeding, Peck (1893) hypothesized that niles, but did not give an estimate for dominantly opportunistic. However, what was in the surface waters could also maximum acceptable prey size. Durbin since menhaden are so widespread and be found in menhaden stomachs. He and Durbin (1975) did not give a prey size abundant in estuarine and nearshore further demonstrated this conclusion for maximum filtration efficiency, but systems, they are frequently an important by pouring sampled seawater through a they did note that the maximum accept­ component of many fishes' diets during gauze and white sand filter. This paper able prey size was between 1,200 J.lm one or more time periods within the year. emphasized that similarities and differ­ and 10 mm, as prey (copepods) of the For example, Atlantic menhaden were ences in the composition of the fishes' smaller size were consumed while those reported as an important component of diet are due to local variations. of the larger (adult brine shrimp) were the diet ofstriped bass, Morone saxatilis, Peck's (1893) hypothesis has been only not. in Albemarle Sound, N.C. (Manooch, slightly modified by more recent studies. In addition to the living organisms 1973), but of variable importance to The relative composition of micro­ consumed, varying quantities ofdetritus weakfish, Cynoscion regalis (Merriner, organisms/materials within the fishes' and/or amorphous material is also in­ 1975). Peck (1893) noted that bluefish, stomachs, as compared to that in the jested while filter feeding. While the Pomatomus saltatrix, and bonito, Sarda surrounding water, is a function of the actual organic source of this material, sarda, are major predators of menha­ menhaden's filtering efficiencies for and the actual magnitude ofthe energetic den, and pointed out that the potential different sizes and types oforganisms. In contribution it makes, can vary by habi­ breadth ofthe role ofAtlantic menhaden addition, there is some minimum size tat type and is not well known, evidence as prey is well demonstrated by its pop­ threshold, below which the fish is in­ of the digestion and subsequent absorp­ ularity as bait. capable of capturing by filtration, as tion of this material by menhaden is Menhaden are thought to be an impor­ well as a maximum size, above which accumulating (Jeffries, 1975; Lewis tant forage for piscivorous birds, e.g. the fish will simply avoid (and/or the and Peters, 1984; Peters and Lewis, brown pelicans, Pelecanus occidentalis, organism avoids the fish). A knowledge 1984; Deegan et aI., 1990). and are known to be heavily preyed upon ofthese maximum and (especially) min­ by osprey, Pandion haliaetus (Spitzer, imum thresholds is critical for the deter­ Menhaden As Forage 1989) and common loons, Gavia immer mination ofthe ecological role ofAtlan­ All life history stages of menhaden (Spitzer'). Menhaden were also re­ tic menhaden (Durbin and Durbin, 1975; from egg through adults are potential ported as prey for marine mammals Friedlandetal., 1984). Ofmajorconcem prey for a large variety of predators. (Hildebrand, 1963). is the nanoplankton (2-20 J.lm by classi­ Moreover, the potential exists for men­ fication of Sieburth et aI., 1978), which haden to feed on their own eggs (Nel­ Parasites and Disease can be a dominant fraction ofthe phyto­ son et al., 1977), as well as the eggs and Two common parasites encountered plankton production in estuarine and larvae of other fishes and invertebrates on Atlantic menhaden are the parasitic nearshore systems, especially during (Peck, 1893; McHugh, 1967). Larvae isopod, Olencira praegustator, and the summer (see Durbin et al., 1975; McCar­ andjuveniles ofa number ofpiscivorous parasitic copepod, Lernaeenicus radia­ thyetal.,1974). species of fish potentially prey upon tus. The relatively common occurrence The minimum size threshold for adult menhaden larvae, depending on their of O. praegustator in the mouth and menhaden (about 260 mm FL) was deter­ coincidence in space and time, along with throat ofB. tyrannus is reflected by one mined from feeding experiments to be compatible body sizes. Many inverte­ of the Atlantic menhaden's early com­ 13-16 J.lm (Durbin and Durbin, 1975). brate predators, especially in oceanic mon names, i.e., "bug-fish" (Smith, Friedland et al. (1984) determined the waters, can be expected to prey upon 1907). Kroger and Guthrie (1972a) noted minimum size threshold for juveniles menhaden larvae; notable among this that the highest rate of infestation of (about 138 mm FL) to be 7-9 J.lm. They group are the abundant chaetognaths this isopod among juvenile Atlantic noted an increase in filtering efficiency (Clements, 1990). Other potential inver­ for some types oforganisms when detri­ tebrate predators include, but are not I P. R. Spitzer, Univ. Md. System, Cent. Environ. tus was present in the water column. limited to, squids (mollusks) and cteno­ Est. Stud., Horn Point Environ. Lab., P.O. Box Since the minimum threshold appears phores and jellyfishes (coelenterates). 775, Cambridge, MD21613. Personal commun.

53(4),1991 13 Figure 10.-Ulcerative mycosis lesions on juvenile Atlantic menhaden from Hancock Creek, N. C. (fall of 1986).

14 Marine Fisheries Review menhaden was in estuaries along the haden from estuarine systems from istics (hence potential genetic separation) mid-portion of the fishes' range, i.e., Delaware Bay to northern Florida. They are insignificant between eastern and Virginia through New Jersey. suggested that the infected fish captured western populations (GSMFC, 1988). In addition to Atlantic menhaden, O. in South Carolina and Georgia were ac­ However, population dynamics analysts praegustator has been reported in spe­ tually migrants from an area of primary treat the Gulf menhaden population as a cimens from the Atlantic and Gulf pop­ infection farther north. This report also single biological and managerial stock ulations ofyellowfin menhaden, as well noted fish which had lesions that resem­ relative to the purse-seine reduction as Gulf menhaden, and both the hybrids bled UM from collections made in New fishery (Nelson and Ahrenholz, 1986; B. patronus X B. smithi and B. tyran­ York estuaries in 1982. Vaughan, 1987; GSMFC, 1988). nus X B. smithi(DaWberg, 1969; Turner Although UM has been detected in On the east coast of Florida, the yel­ and Roe, 1967). Similarly, the cope­ various families ofestuarine dependent lowfin menhaden, the extreme south­ pod L. radiatus has been found on the fishes (Sindermann, 1988), it has not ern portion of the Atlantic menhaden large-scaled species of menhaden and been reported from the other three spe­ population, and their hybrids appear to their hybrids with B. smithi, as well as cies of North American menhadens. comprise the "stock" for a menhaden B. smithi from the Gulf of Mexico and However, Nogaetal. (1988) pointed out bait fishery. Similarly, on the west coast Atlantic populations (Dahlberg, 1969). that what was reported by Kroger and ofFlorida, the yellowfin menhaden, the Lists of additional parasites of Atlantic Guthrie (1972b) as wounds attributed to southeastern most portion of the Gulf menhaden are contained in Westman and predators on some juvenile Gulf men­ menhaden population, and their hybrids Nigrelli (1955), Hildebrand (1963), and haden as caused by predators, appeared represent the "stock" for another Reintjes (1969). similar to UM lesions. menhaden bait fishery. Yellowfin men­ Two major diseases are commonly haden from each coast of Florida are associated with Atlantic menhaden. Population Processes probably genetically separate popula­ Westman and Nigrelli (1955) reported tions. Dahlberg (1970) gives some on annual die-offs in the New York Stock Structure meristic comparisons for these potential­ area. The dying fish, "spinners" (hence Considerable debate relative to the ly distinct populations. Very little is "spinning disease") were characterized stock structure of the Atlantic menha­ known or speculated relative to genetic as having lost coordinated movements, den population has been expended, and mixing within the population offinescale with one or both eyes protruded, and with as many as three different stocks have menhaden. hemorrhages in the gills, eyes, and optic been advanced, primarily on the basis lobes of the brain. Similar mortalities of meristic and morphometric analyses Mortality have been noted among Atlantic men­ (June, 1958, 1965; Sutherland, 1963; Traditionally, losses in numbers ofin­ haden in Chesapeake Bay (Reintjes, June and Nicholson, 1964; Nicholson, dividuals from fish populations (total 1969). The cause of these mortalities 1972, 1978; Dryfoos et al., 1973; Epper­ mortality) are ascribed to either fishing was undetermined. A virus has subse­ ly, 1989). Some evidence for the pres­ or natural mortality. Analytical pro­ quently been identified as the agent of ence of more than one stock exists; cedures used to estimate instantaneous this disease, at least in Chesapeake Bay however, the fish reared in different rates oftotal (Z) and fishing (F) mortal­ (Stephens et al., 1980). geographic areas and those from differ­ ity for each of the large-scaled menha­ The second disease, ulcerative mycosis ent temporal spawning cohorts appear den populations assume constant rates (UM), became prominent in recent to mix rapidly due to the nature of their of instantaneous natural mortality (M) years. Atlantic menhaden with deep, movement patterns. Since potentially among time intervals and estimated rates crater-like lesions of UM, were ob­ different spawning groups are currently of fishing for each interval (Nelson and served in collections from the Pamlico inseparable in the Atlantic purse-seine Ahrenholz, 1986; Ahrenholz et al., River, N.C., during spring 1984 by per­ reduction fishery, the Atlantic menhaden 1987b; Vaughan, 1987; Vaughan and sonnel from the North Carolina Division population is treated as a single exploited Smith, 1988). Thecatch-at-agedata does of Marine Fisheries. Although these stock with respect to that fishery. not contain enough information to esti­ lesions occurred on most areas of the In marked contrast to Atlantic mate both M and F simultaneously, as body, they were most common in the menhaden, Gulf menhaden lack any they are additive exponential rates (Z = anal area (Noga et aI., 1988)(Fig. 10). systematic seasonal movement through F + M). Thus, the computational pro­ Hargis (1985) provided an early de­ their range and tend to mix very slowly. cedures have ascribed to F all the vari­ scription of this disease. Pathological Tagging studies revealed that movement ances observed in Z among time inter­ investigations of infected fish revealed across the Mississippi Delta is infre­ vals, even though true M also probably the presence of aseptate fungal hyphae quent, either within or between seasons varied as well, albeit to a lesser degree. of the genera Aphanomyces and Sapro­ (KrogerandPristas, 1975; Pristasetal., The estimate of M recently used in legnia in the area ofthe lesions (Dykstra 1976). Hence, ithasbeensuggestedthat assessment analyses for Atlantic men­ et aI., 1986; Noga and Dykstra, 1986). the Gulf menhaden population could be haden is M = 0.45. This estimate is a During 1985, Ahrenholz et al. (1987a) treated as two management stocks, even mean of a range of available estimates: found infected juvenile Atlantic men­ though differences in meristic character­ Dryfoosetal. (1973)estimatedM =0.52

53(4),1991 IS from an analysis of returns of adult­ where the number of recruits to the pop­ analysis with environmental variables tagged Atlantic menhaden; Reish et al. ulation are dependent on one or more which were considered important a (1985) estimated M = 0.50 for ages 2 environmental factors. Removal ofadult priori. Emphasis was placed on Ekman and 3 from analyses of tag returns of fish from the population can have a pro­ transport, which was thought to be a both adult- and juvenile-tagged fish; nounced effect on subsequent recruit­ substantial contributor to oceanic cur­ Schaaf and Huntsman (1972) estimated ment ifthe stock-recruitment relationship rents which would bring larval Atlan­ M = 0.37 from an analysis of catch is strong. Results from analyses con­ tic menhaden from offshore spawning statistics. ducted by Schaaf and Huntsman (1972) areas to the vicinity of inlets and estu­ An estimate ofM = 1.1 has been used revealed a very weak association be­ arine nursery areas. The resultant model in stock assessment analyses for Gulf tween the spawning stock size and subse­ described the recruitment data for menhaden (Nelson and Ahrenholz, 1986; quent recruitment for Atlantic menha­ 1955-70 fairly well. Though concep­ Vaughan, 1987; GSMFC, 1988); it rep­ den. Later analyses used potential egg tually sound, this model did not effec­ resents the mean of six estimates of M production in place of spawning stock, tively describe the recruitment esti­ ranging from 0.69 to 1.61 , obtained from but did not substantially improve the mates obtained during the 1970's and an analysis of mark-recovery data (Ah­ analytical relationship (Nelson et aI., early 1980's. Reish etal. (1985) feltthat renholz, 1981). 1977). The data used in both analyses, much ofthe strong statistical correlation Since estimates ofM for both Atlantic when plotted as scatter diagrams, did not obtained by Nelson et al. (1977) for and Gulf menhaden were obtained from display any pattern well enough to sug­ Ekman transport was due to one data purse-seine reduction landings or tag gest anappropriate functional model. Of point, i.e., the exceptionally large 1958 recoveries from reduction plants, they the two theoretical recruitment func­ year class. Additional mixed regulatory include "other" losses. In addition to tions commonly used, Ricker's (1954) factor models were developed by predation and disease, losses due to by­ equation, which results in a dome-shaped Yoshiyamaetal. (1981), who preferred catch in other fisheries, as well as land­ curve, was selected on biological grounds a Ricker function with an environmen­ ings for bait are included as losses in M. for both ofthe earlier reports (Schaafand tal parameter (Ekman transport at lat. No estimates for M are available for Huntsman, 1972; Nelsonetal., 1977). 35°N., long. 75°W.) in the stock-inde­ either of the two small-scaled menha­ For example, menhaden may consume pendent term of the equation. dens. Values of M are probably inter­ their own eggs under certain circum­ Checkley et al. (1988) described a mediate between those for Atlantic and stances, which can contribute to the process-oriented study relative to spawn­ Gulf menhaden. descending right hand limb ofthe Ricker ing, larval transport, and early survival curve. Further, four other ocean­ of Atlantic menhaden. They suggested Recruitment spawning clupeids are thought to be that spawning off the North Carolina Tempered by the number ofage classes represented by a dome-shaped curve coast occurred along the western wall represented in the fisheries, fluctuations (Cushing, 1971). Based on statistical oftheGulfStream, and that the ultimate in year-class size have naturally con­ grounds, Reish etal. (1985) preferred the survival of larvae was dependent on tributed to the variability in landings function developed by Bevertonand Holt storm-induced upwelling and buoyancy­ .... among years (Smith, 1991). Estimates (1957) for simulation purposes. driven transport (the result of water and of recruitment into the Gulf menhaden Nelson and Ahrenholz (1986) found air temperature differentials). They also stock at age 1 have varied more than that potential egg production-recruitment postulated that events on the order of fivefold, while estimates ofrecruitment scatter plots for Gulf menhaden were days rather than months, were critical into the Atlantic stock at a similar age dome-shaped. Coupled with biological to spawning and larval transport and have fluctuated almost thirteenfold arguments, they used the Ricker function development. (Vaughan and Merriner, 1991). for both description and subsequent Shaw et al. (1985b, 1988) examined The observed uncertainty for men­ population simulation studies. onshore transport and subsequent estu­ haden recruitment among years and its Since deterministic, density-depen­ arine immigration processes for Gulf ramification for landings have fostered dent spawner-recruitment functions menhaden larvae. They hypothesized a number of different investigations alone would be of little value in predic­ that larvae spawned in the waters west ranging from those designed to deter­ ting subsequent year-class sizes of At­ of the Mississippi Delta moved shore­ mine iffishing was impacting recruitment lantic menhaden, Nelson et al. (1977) ward in a west-northwesterly direction to those designed to predict recruitment developed a mixed regulatory factor and subsequently enter more westerly and/or landings. Additionally, studies (both density-dependent and density­ estuaries, rather than estuaries nearer to were designed to directly sample and independent) model. First, they fitted where spawning actually occurred. estimate prerecruitment abundance. a density-dependent Ricker function to These studies may ultimately permit a Factors affecting recruitment are tradi­ the potential egg production-recruit­ more refined examination of potential tionally categorized as density depen­ ment data, then used the deviations from environmental factor influences on lar­ dent, where the absolute spawning stock the fitted model to develop a survival val survival over a reduced temporal size and the size of the subsequent year index. This index was in turn the depen­ and geographic scale. class are related; or density independent, dent variable for a multiple regression Two studies which emphasized envi­

16 Marine Fisheries Review ronmental variables and are at least tan­ atic, exploratory regression analyses geographic and temporal spawning gentially related to fishery recruitment, have a high probability of providing origins of prerecruits. Research along were conducted on Gulf menhaden. In some spurious relationships. these lines is being conducted. the first study, Stone (1976) conducted Guilloryetal. (1983) treated each year an extensive, systematic, multiple-re­ class separately, thus, their models may Literature Cited gression search ofvarious environmen­ provide some insight into important en­ Ahrenholz, D. W. 1981. Recruitment and exploita­ tal data. Methodical temporal lags and vironmental factors. As with Stone's tion of Gulf menhaden, Brevoortia patronus. Fish. Bull. 79:325-335. commercial fishing effort were used as (1976) work, however, caution must be ----:-=-::-=__~, J. F. Guthrie, and R. M. Clayton. independent variables to detect any exercised in applying the results because 1987a. Observations of ulcerative mycosis in­ potential relationships between these there is a strong possibility of spurious fections on Atlantic menhaden (Brevoonia tyran­ nus). U.S. Dep. Commer., NOAA Tech. Memo. variables and landings ofGulfmenhaden. correlations resulting from such a large NMFS-SEFC-196, 28 p. Some of the analyses were conducted number ofexploratory regressions. This ____ , , and C. W. Krouse. 1989. Results ofabundance surveys ofjuvenile using monthly time periods. He thought is especially true when using the CPSE Atlantic and Gulf menhaden, Brevoortia tyran­ that environmental factors which influ­ values, as they are the result ofan earlier nus and B. patronus. U.S. Dep. Commer., ence important life history events (espe­ exploratory series ofanalyses (Guillory NOAA Tech. Rep. NMFS 84, 14 p .. ----:-=-::-=_~, W. R. Nelson, and S. P. Epperly. cially factors affecting recruitment) could and Bejarano, 1980). Additionally, the 1987b. Population and fishery characteristics be detected by the analyses when ap­ predictive accuracy ofrelationships will ofAtlantic menhaden, Brevoortia tyrannus. Fish. propriately temporally matched. depend on how strongly CPUE ofage-l Bull. 85:569-600. Beverton, R. J. H., and S. J. Holt. 1957. On the The second study was reported by menhaden reflects true year class size. dynamics of exploited fish populations. Fish. Guillory et al. (1983), who used first­ The NMFS Beaufort Laboratory at­ Invest. Ser. II, Mar. FishG.B. Minist. Agric., Fish. Food 19,533 p. order linear regressions and stepwise tempted to obtain prerecruitment esti­ Checkley, D. M.,Jr.,S. Raman, G. L. Maillet, and multiple regressions to determine the mates of year-class strength by directly K. M. Mason. 1988. Winterstormeffectsonthe relationship between a wide variety of sampling juvenile menhaden in estua­ spawning and larval drift ofa pelagic fish. Nature 335:346-348. environmental factors, both singly and rine areas a few months prior to their Chipman, W. A. 1959. Use of radioisotopes in in combination. They examined catch­ emigration and recruitment into the studies of the foods and feeding activities of per-sampling-effort (CPSE) of young­ coastal populations. Some forms of ju­ marine animals. Publ. Sta. Zool. Napoli, 31 (Supp!.): 154-175. of-the-year Gulf menhaden from otter venile relative abundance sampling Christmas, J. Y., and G. Gunter. 1960. Distribu­ trawIshrimp abundance surveys in Loui­ began as early as 1956 on the Atlantic tion ofmenhaden, genus Brevoortia, in the Gulf of Mexico. Trans. Am. Fish. Soc. 89:338-343. siana estuaries (Guillory and Bejarano, coast (Ahrenholz et aI., 1989). Two­ ____ and R. S. Waller. 1975. Location and 1980), as well as landings ofage-l Gulf boat, surface-trawl surveys were initi­ time ofmenhaden spawning in the GulfofMex­ menhaden per-vessel-ton-week ofeffort ated for juvenile Atlantic menhaden in ico. GulfCoastRes. Lab., Ocean Springs, Miss., 20p. (CPUE as reported earlier) by the com­ 1962 and for juvenile Gulf menhaden Clements, L. C. 1990. Chaetognaths versus larval mercial purse-seine fleet from Louisiana during 1964 (Turner, 1973; Turner et fish. Fourteenth Larval Fish Conf., Am. Fish. ports. Both the CPSE and CPUE values al., 1974). The studies culminated with Soc., May 6-9, 1990, Beaufort, N.C. (Abstr.) Combs, R. M. 1969. Embryogenesis, histology were used as surrogates for recruitment an extensive two-boat, surface-trawl and organology of the ovary of Brevoortia estimates (year-class strength). relative abundance survey on each coast patronus. Gulf Res. Rep. 2(4):333-434. In general terms, Guillory et al. (1983) from the early 1970's through 1978. Conover, D. O. 1985. Field and laboratory as­ sessment of pattern in fecundity of a mUltiple and Stone (1976) found some strong Although density estimates from these spawning fish: the Atlantic silverside Menidia correlations for the temporally lagged surveys appeared to reflect local and menidia. Fish. Bull. 83:331-341. Cushing, D. H. 1971. The dependence of recruit­ variables such as temperature (air or regional abundance of menhaden, they ment on parent stock in different groups offishes. water), and wind speed and direction, did not correlate well with subsequent J. Cons., Cons. Int. Explor. Mer 33:340-362. with the CPUE and CPSE values and virtual population analyses estimates of Dahlberg, M. D. 1969. Incidence of the isopod, Olencira praegustator, and copepod, Lernae­ landings, respectively. Stone's (1976) year-class size for either coast (Ahren­ enicus radiatus, in three species and hybrid analyses provide some insight into im­ holz et aI., 1989). Nevertheless, these menhaden (Brevoortia) from the Florida coasts, portant environmental variables, but with five new host records. Trans. Am. Fish. Soc. surveys were invaluable for determin­ 98: 111-115. only at a more general level. Recruitment ingjuvenile menhaden distribution pat­ ____ . 1970. Atlantic and Gulf of Mexico variability is not clearly expressed in terns in estuaries and aided many life menhadens, Genus Brevoortia (Pisces: Clupeidae). Bull. Fla. State Mus., BioI. Sci. purse-seine landings because they are history studies, even though they had 15:91-162. dominated by two age classes. But, limited predictive application. Darnell, R. M. 1958. Food habits of fishes and predictions of landings, not year-class The limited success of attempts to largerinvertebrates ofLake Pontchartrain, Loui­ siana, an estuarinecommunity . Pub!. Inst. Mar. sizes, were apparently the desired pro­ model or estimate recruitment at relative­ Sci. Tex. 5:353-416. ducts of Stone's (1976) analyses. In ad­ ly young ages may well be due to over­ Deegan, L. A. 1986. Changes in body composition dition, the author stated that potential simplification in the models' or sampling and morphology ofyoung-of-the-year Gulfmen­ haden, Brevoortia patronus Goode, in environmental effects could be masked programs' designs relative to the com­ Fourleague Bay, Louisiana. J. Fish BioI. 29: in the regression models by effort and plexity ofthe recruitment patterns. With 403-415. -:----:-,----,_ , B. 1. Peterson, and R. Portier. 1990. temporal effects. Finally, the results have respect to at least Atlantic menhaden, an Stable isotopes and cellulase activity as evidence to be viewed cautiously because system­ accounting must be made of both the for detritus as a food source for juvenile Gulf

53(4), 1991 17 menhaden. Estuaries 13: 14-19. among yellowfin and Gulfmenhaden (Brevoor­ middle Atlantic coast from RV Dolphin cruises, ____ andB. A. Thompson. 1987. Growth tia) and their hybrid. Trans. Am. Fish. Soc. 1965-66. Fish. Bull. 73:317-335. rate and life history events of young-of-the­ 97:119-123. Kroger, R. L., andJ. F. Guthrie. 1972a. Incidence year gulf menhaden as determined from oto­ ____ . 1970. Rearing larvae of yellowfin of the parasitic isopod, Olencira praegustator, liths. Trans. Am. Fish. Soc. 116:663-667. menhaden, Brevoortia smithi. Copeia 1970: in juvenile Atlantic menhaden. Copeia 1972: Dietrich, C. S., Jr. 1979. Fecundity of the Atlan­ 775-776. 370-374. tic menhaden, Brevoortia tyrannus. Fish. Bull. _-----,-----,_ . 1984. Description ofeggs, larvae, and and . 1972b. Effect of 77:308-311. early juveniles of Gulf menhaden, Brevoortia predators on juvenile menhaden in clear and Dryfoos, R. L., R. P. Cheek, and R. L. Kroger. patronus, and comparisons with Atlantic men­ turbid estuaries. Mar. Fish. Rev. 34(11-12): 1973. Preliminary analysis of Atlantic menha­ haden, B. tyrannus, and yellowfin menhaden, 78-80. den, Brevoortia tyrannus, migrations, population B. smithi. Fish. Bull. 82:85-95. and . 1973. Migrations structure, survival and exploitation rates, and Higham, J. R., and W. R. Nicholson. 1964. Sex­ of tagged juvenile Atlantic menhaden. Trans. availability as indicated from tag returns. Fish. ual maturation and spawning of Atlantic men­ Am. Fish. Soc. 102:417-422. Bull. 71:719-734. haden. Fish. Bull. 63:255-27!. ____ , ,andM.H.Judy.1974. Durbin, A. G., and E. G. Durbin. 1975. Grazing Hildebrand, S. F. 1948. A review ofthe American Growth and first annulus formation of tagged rates ofthe Atlantic menhaden Brevoortia tyran­ menhaden, genus Brevoortia, with a description and untagged Atlantic menhaden. Trans. Am. nus as a function of particle size and concentra­ ofa new species. Smithson. Misc. Collect. 107 Fish. Soc. 103:292-296. tion. Mar. BioI. 33:265-277. (18): 1-39. ____ and P. J. Pristas. 1975. Movements Durbin, E. G., R. W. Krawiec,andT.J. Smayda. __-,--,- . 1963. Family Clupeidae. Mem. Sears oftaggedjuvenile menhaden (Brevoortiapatro­ 1975. Seasonal studies on the relative impor­ Found. Mar. Res., Yale Univ. 1(3):257-451. nus) in the Gulf of Mexico. Tex. J. Sci. 26: tance ofdifferent size fractions ofphytoplankton Houde, E. D., and P. J. Fore. 1973. Guide to 473-477. in Narragansett Bay (USA). Mar. BioI. 32: the identity of eggs and larvae of some Gulf of Kuntz, A., and L. Radcliffe. 1917. Notes on the 271-287. Mexico clupeid fishes. Fla. Dep. Nat. Resour., embryology and larval development of twelve Dykstra, M. J., E. J. Noga, 1. F. Levine, D. W. Mar. Res. Lab., Leafl. Ser. 4, pp. 1(23), 14 p. teleostean fishes. Bull. U.S. Bur. Fish. 35: Moye,andJ. H. Hawkins. 1986. Characteriza­ _-:-_:-- andL. J. Swanson, Jr. 1975. Descrip­ 87-134. tion of the Aphanomyces species involved with tion ofeggs and larvae of yellowfin menhaden, Lewis, R. M., D. W. Ahrenholz, andS. P. Epper­ ulcerative mycosis (UM) in menhaden. Myco­ Brevoortia smithi. Fish. Bull. 73:660-673. Iy. 1987. Fecundity ofAtlantic menhaden, Bre­ logia 78:664-672. Hunter, J. R., andB. J. Macewicz. 1985. Measure­ voortia tyrannus. Estuaries 10:347-350. Epperly, S. P. 1989. Ameristic, morphometric and ment ofspawning frequency in multiple spawn­ ____ and C. M. Roithmayr. 198!. Spawn­ biochemical investigation ofAtlantic menhaden, ing fishes. U.S. Dep. Commer., NOAA Tech. ing and sexual maturity ofGulfmenhaden, Bre­ Brevoortia tyrannus (Latrobe). J. Fish BioI. Rep. NMFS 36:79-94. voortiapatronus. Fish. Bull. 78:947-951. 35: 139-152. Jeffries, H. P. 1975. Diets of juvenile Atlantic ____ , E. P. Wilkins, and H. R. Gordy. Ferraro, S. P. 1980a. Embryonic development of menhaden (Brevoortia tyrannus) in three estu­ 1972. A description of young Atlantic menha­ Atlantic menhaden, Brevoortia tyrannus, and a arine habitats as determined from fatty acid com­ den, Brevoortia tyrannus, in the White Oak River fish embryo age estimation method. Fish. Bull. position ofgut contents. J. Fish. Res. Board Can. estuary, North Carolina. Fish. Bull. 70: 115-118. 77:943-949. 32:587-592. Lewis, V. P., and D. S. Peters. 1984. Menhaden -----,::-:------,-----, . 1980b. Daily time ofspawning of 12 Jones, P. W., F. D. Martin, and J. D. Hardy, Jr. - a single step from vascular plant to fishery fishes in the Peconic Bays, New York. Fish. Bull. 1978. Development offishes ofthe Mid-Atlantic harvest. J. Exper. Mar. BioI. Ecol. 84:95-100. 78:455-464. Bight. Vol. I, Acipenseridae through Ictaluridae. Manooch, C. S., III. 1973. Food habits ofyearling Fore, P. L. 1970. Oceanic distribution ofeggs and U.S. Fish Wildl. Serv., BioI. Servo Program and adult striped bass, Morone saxatilis (Wal­ larvae of the Gulfmenhaden. U.S. Dep. Inter., FWS/OBS-78/12, 314 p. burn), from Albemarle Sound, North Carolina. Fish Wildl. Serv., Circ. 341: 11-13. Jorgenson, S. C., and G. L. Miller. 1968. Length Chesapeake Sci. 14:73-86. Friedland, K. D. 1985. Functional morphology of relations of some marine fishes from coastal McCarthy, J. J., W. R. Taylor, and M. E. Loftus. the branchial basket structures associated with Georgia. U.S. Dep. Inter., Fish. Wild!. Serv., 1974. Significance of nanoplankton in the feeding in the Atlantic menhaden, Brevoortia Spec. Sci. Rep. Fish. 575, 16 p. Chesapeake Bay estuary and problems associ­ tyrannus (Pisces: Clupeidae). Copeia 1985: Judy, M. H.,andR. M. Lewis. 1983. Distribution ated with the measurement of nanoplankton 1018-1027. of eggs and larvae of Atlantic menhaden, Bre­ productivity. Mar. BioI. 24:7-16. _-----,__ , L. W. Haas, and J. V. Merriner. voortia tyrannus, along the Atlantic coast ofthe McHugh, J. L. 1967. Estuarine nekton. In G.H. 1984. Filtering rates of the juvenile Atlantic United States. U.S. Dep. Commer., NOAA Lauff(Editor) Estuaries, p. 581-620. Am. Assoc. menhaden Brevoortia tyrannus (Pisces: Clu­ Tech. Rep. NMFS SSRF 774,23 p. Adv. Sci. No. 83. peidae), with consideration of the effects of June, F. C. 1958. Variation in meristic characters ____ , R. T. Oglesby, and A. L. Pacheco. detritus and swimming speed. Mar. BioI. 84: ofyoung Atlantic menhaden, Brevoortia tyran­ 1959. Length, weight, and age composition ofthe 109-117. nus. Rapp. P.-v. Reun. Cons. Int. Explor. Mer menhaden catch in Virginia waters. Lirnnol. Govoni,J. J., D. E. Hoss,andA. J. Chester. 1983. 143:26-35. Oceanogr. 4: 145-162. Comparative feeding of three species of larval _--=-:-:-_ . 1965. Comparison ofvertebral counts Merriner, J. V. 1975. Food habits ofthe weakfish, fishes in the northern GulfofMexico: Brevoor­ of Atlantic menhaden. U.S. Dep. Inter., Fish Cynoscion regalis, in North Carolina. Chesa­ tiapatronus, Leiostomusxamhurus, and Micro­ Wild!. Serv., Spec. Sci. Rep. Fish. 513,12 p. peake Sci. 16:74-76. pogonius undulatus. Mar. Ecol. Prog. Ser. 13: and F. T. Carlson. 1971. Food of Nelson, W. R., and D. W. Ahrenholz. 1986. 189-199. young Atlantic menhaden, Brevoortia tyrannus, Population and fishery characteristics of Gulf Guillory, V., and R. Bejarano. 1980. Evaluation in relation to metamorphosis. Fish. Bull. 68: menhaden, Brevoortia patronus. Fish. Bull. of juvenile menhaden abundance data for pre­ 493-512. 84:311-325. diction ofcommercial harvest. Proc. Annu. Conf. ____ andW. R. Nicholson. 1964. Age and ____ , M. C. Ingham, and W. E. Schaaf. Southeast Assoc. Fish Wildl. Agencies 34: size composition of the menhaden catch along 1977. Larval transport and year-class strength 193-203. the Atlantic coast of the United States, 1958. ofAtlantic menhaden, Brevoortia tyrannus. Fish. ____ ,J. Geaghan, and J. Roussel. 1983. In­ U.S. Dep. Inter., Fish. Wildl. Serv., Spec. Sci. Bull. 75:23-4!. fluence of environmental factors on Gulf men­ Rep. Fish. 446, 40 p. Nicholson, W. R. 1971. Coastal movements of haden recruitment. La. Dep. Wildl. Fish., Tech. ____ andJ. W. Reintjes. 1959. Age and size Atlantic menhaden as inferred from changes in Bull. 37, 32 p. composition of the menhaden catch along the age and length distributions. Trans. Am. Fish. GSMFC. 1988. The menhaden fishery ofthe Gulf Atlantic coast ofthe United States, 1952-55; with Soc. 100:708-716. ofMexico United States: A regional management a brief review ofthe commercial fishery. U. S. ___-:- . 1972. Population structure and move­ plan. Gulf States Mar. Fish. Comm., Ocean Dep. Inter., Fish Wildl. Serv., Spec. Sci. Rep. ments of Atlantic menhaden, Brevoortia tyran­ Springs, Miss., 138 p. Fish. 317, 65 p. nus, as inferred from back-calculated length Gunter, G. 1945. Studies on marine fishes ofTexas. ____ and C. M. Roithmayr. 1960. Deter­ frequencies. Chesapeake Sci. 13: 161-174. Publ. Inst. Mar. Sci. Tex. 1(1):1-190. mining age of Atlantic menhaden from their ____ . 1978. Movements and population Hargis, W. J. 1985. Quantitative effects ofmarine scales. Fish. Bull. 60:323-342. structure of Atlantic menhaden indicated by tag diseases on fish and shellfish populations. Trans. Kendall, A. W., Jr., and J. W. Reintjes. 1975. returns. Estuaries 1:141-150. N. Am. Wild!. Nat. Resour. Conf. 50:608-640. Geographic and hydrographic distribution of and W. E. Schaaf. 1978. Aging of Hellier, W. F., Jr. 1968. Artificial fertilization Atlantic menhaden eggs and larvae along the Gulf menhaden, Brevoortia patronus. Fish.

18 Marine Fisheries Review Bull. 76:315-322. Bull. Mar. Sci. 36:96-103. in Fish Communities and Their Role in Fish. Noga, E. J., and M. J. Dykstra. 1986. Oomycete ____ , W. J. Wiseman, Jr., R. E. Turner, Manage., Atlanta, 1978, p. 93-101. Sport Fish. fungi associated with ulcerative mycosis in L. J. Rouse, Jr., R. E. Condrey, and F. J. Kelly, Inst., Wash., D.C. menhaden, Brevoortia tyrannus (Latrobe). J. Jr. 1985b. Transport of larval Gulf menhaden Tagatz, M. E., and E. P. H. Wilkins. 1973. Sea­ Fish. Dis. 9:47-53. Brevoortia patrollUS in continental shelf waters sonal occurrence of young Gulf menhaden and ____ ,J. F. Levine, M.J. Dykstra,andJ. H. ofwestern Louisiana: A hypothesis. Trans. Am. other fishes in a northwestern Florida estuary. Hawkins. 1988. Pathology ofulcerative mycosis Fish. Soc. 114:452-460. U.S. Dep. Commer., NOAA Tech. Rep. NMFS in Atlantic menhaden Brevoortia tyrannus. Dis. ____ , B. D. Rogers,J. H. Cowan,Jr., and SSRF-672, 14 p. Aquat. Org. 4: 189-197. W. H. Herke. 1988. Ocean-estuary coupling of Turner, W. R. 1969. Life history ofmenhadens in Peck,J. 1.1893. On thefoodofthe menhaden. Bull. ichthyoplankton and nekton in the northern Gulf the eastern GulfofMexico. Trans. Am. Fish. Soc. U.S. Fish. Comm. 13:113-126. of Mexico. Am. Fish. Soc. Symp. 3:77-89. 98:216-224. Peters, D. S., and V. P. Lewis. 1984. Estuarine Sieburth, J. M., V. Smetacek, and J. Lenz. 1978. ____ . 1971. Occurrence of Brevoortia productivity: Relating trophic ecology to fish­ Pelagic ecosystem structure: Heterotrophic gunteri in Mississippi Sound. Q. J. Fla. Acad. Sci. eries.ln B. J. Copeland, K. Hart, N. Davis, and compartments ofthe plankton and their relation­ 33(4):273-274. S. Friday (Editors), Research for managing the ship to plankton size fractions. Limnol. Ocean­ _.,--__ . 1973. Estimating year-class strength nation's estuaries: Proceedings ofa conference ogr. 23:1256-1263. ofjuvenile menhaden. III A. L. Pacheco (Editor), in Raleigh, North Carolina, March 13-15, 1984, Simmons, E. G. 1957. An ecological survey ofthe Proc. Workshop on Egg, Larval, and Juvenile p. 255-264. UNC Sea Grant Coli. Publ. upper Laguna Madre ofTexas. Publ. Ins!. Mar. Stages ofFish in Atl. Coast Est. , p. 37-47. Tech. UNC-SG-84-08. Sci. Univ. Tex. 4(2):156-200. Publ. I, Mid-Atl. Coast. Fish. Cent., Natl. Mar. Powell, A. B., and G. Phonlor. 1986. Early life Sindermann, C. J. 1988. Epizootic ulcerative syn­ Fish., NOAA, Highlands, N.J. history of Atlantic menhaden, Brevoortia tyran­ dromes in coastal/estuarine fish. U.S. Dep. Com­ -----,~-----,-=, G. N. Johnson, and H. R. Gordy. nus, and Gulfmenhaden, B. patrollus. Fish. Bull. mer., NOAA Tech. Memo. NMFS-F/NEC-54, 1974. Compendium of juvenile abundance 84:991-995. 37 p. surveys in coastal streams of the northern Gulf Pristas, P. J., E. J. Levi, andR. L. Dryfoos. 1976. Smith, H. M. 1907. The fishes of North Carolina. ofMexico. U.S. Dep. Commer., NOAA, Natl. Analysis of returns of tagged Gulf menhaden. N.C. Geol. Econ. Surv., Vol. 2. E. M. Uzzell Mar. Fish. Serv., Data Rep. 89, 189p. Fish. Bull. 74:112-117. and Co., Raleigh, N.C., 453 p. ____ and R. B. Roe. 1967. Occurrence of Reintjes, J. W. 1962. Development of eggs and Smith,J. W.1991. The Atlantic and Gulfmenhaden the parasitic isopod Olellcirapraegustator in the yolk-sac larvae of yellowfin menhaden. U.S. purse seine fisheries: Origins, harvesting yellowfin menhaden, Brevoortia smithi. Trans. Dep. Inter., Fish Wildl. Serv., Fish. Bull. 62: technologies, biostatistical monitoring, recent Am. Fish. Soc. 96:357-359. 93-102. trends in fisheries statistics, and forecasting. Mar. Vaughan, D. S. 1987. Stock assessment ofthe Gulf _----,-_.,-- . 1969. Synopsis ofbiological data on Fish. Rev. 53(4):28-39. menhaden, Brevoortiapatrollus, fishery. U.S. Atlantic menhaden, Brevoortia tyrallllus. U.S. Spitzer, P. R. 1989. Osprey. III Proc. Northeast Dep. Commer., NOAA Tech. Rep. NMFS 58, Dep. Inter., Fish Wildl. Serv., Circ. 320, 30 p. Raptor Manage. Symp. Workshop, p. 22-29. 18 p. ____ andA. L. Pacheco. 1966. The relation Natl. Wildl. Fed., Wash., D.C. _.,--.,--_ and J. V. Merriner. 1991. Assessment ofmenhaden to estuaries. Am. Fish. Soc. Spec. Stephens, E. B., M. W. Newman,A. L. Zachary, and management ofAtlantic and Gulfmenhaden Publ. 3:50-58. and F. M. Hetrick. 1980. A viral etiology for stocks. Mar. Fish. Rev. 53(4):47-55. Reish, R. L., R. B. Deriso, D. Ruppert, and R. J. the annual spring epizootics of Atlantic men­ ____ andJ. W. Smith. 1988. Stock assess­ Carroll. 1985. An investigation ofthe population haden Brevoortia tyranllus (Latrobe) in Chesa­ ment ofthe Atlantic menhaden, Brevoortia tyran­ dynamics of Atlantic menhaden (Brevoortia peake Bay. J. Fish Dis. 3:387-398. IlUS, fishery. U.S. Dep. Commer., NOAA Tech. tyrannus). Can. J. Fish. Aquat. Sci. 42: 147-157. Stone, J. H. 1976. Environmental factors relating Rep. NMFS 63, 18 p. Ricker, W. E. 1954. Stockandrecruitment.J. Fish. to Louisiana menhaden harvest. La. State Univ., Warlen, S. M. 1988. Age and growth of larval Res. Board Can. 11:559-623. Sea Grant Publ. LSU-T-76-004, 69 p. Gulf menhaden, Brevoortia patronus, in the Roithmayr, C. M. 1963. Distribution offishing by Sutherland, D. F. 1963. Variation in vertebral northern GulfofMexico. Fish. Bull. 86:77-90. purse seine vessels for Atlantic menhaden, numbers of juvenile Atlantic menhaden. U.S. Westman, J. R., and R. F. Nigrelli. 1955. Prelim­ 1955-59. U.S. Dep. Inter., Fish Wildl. Serv., Dep. Inter., Fish Wildl. Serv., Spec. Sci. Rep. inary studies of menhaden and their mass mor­ Spec. Sci. Rep. Fish. 434, 22 p. Fish. 435, 21 p. talities in Long Island and New Jersey waters. ____ and R. A. Waller. 1963. Seasonal oc­ Suttkus, R. D. 1956. Early life history of the Gulf N.Y. Fish Game J. 2: 142-153. currence ofBrevoortia patronus in the northern menhaden, Brevoortiapatronus, in Louisiana. Wilkins, E. P. H., and R. M. Lewis. 1971. Abun­ Gulf of Mexico. Trans. Am. Fish. Soc. 92: Trans. N. Am. Wildl. Conf. 21:390-406. dance and distribution of young Atlantic men­ 301-302. ____ and B. I. Sundararaj. 1961. Fecundity haden, Brevoortia tyrannus, in the White Oak Schaaf, W. E., andG. R. Huntsman. 1972. Effects and reproduction in the largescale menhaden, River estuary, North Carolina. Fish. Bull. 69: of fishing on the Atlantic menhaden stock: Brevoortia patrollus Goode. Tulane Stud. Zool. 783-789. 1955-1969. Trans. Am. Fish. Soc. 101 :290-297. 8: 177-182. Yoshiyama, R. M., W. Van Winkle, B. L. Kirk, Shaw, R. F., J. H. Cowan, Jr., and T. L. Tillman. Sykes, J. E., and C. S. Manooch, III. 1979. andD. E. Stevens. 1981. Regression analyses of 1985a. Distribution and density of Brevoortia Estuarine predator-prey relations.lnH. Clepper stock-recruitment relationships in three fish patrollus (Gulfmenhaden) eggs and larvae in the (Editor), Predator-prey systems in fisheries populations. Oak Ridge Natl. Lab. , Oak Ridge, continental shelf waters of western Louisiana. management. Int. Symp. Predator-Prey Systems Tenn. Environ. Sci. Div. Publ. 1645,54 p.

53(4), /991 19