Consequences of dispersal of subtropically spawned crevalle jacks, Caranx hippos, to temperate estuaries(*).

Abstract.--Caranx hippos spawn at subtropical and tropical latitudes, but some of their propagules are dispersed hundreds of kilometers north of Cape Hatteras into temperate waters of the western North Atlantic. The effect that this northward dispersal pattern has upon the population depends on whether these juveniles return south during autumn to overwinter or whether they become expatriated from the spawning population and die from hypothermal winter conditions at temperate latitudes. We evaluated whether repatriation was possible by comparing C. hippos seasonal abundance and size-structure from New York to Florida. Young-of-the-year C. hippos occurred annually during summer and autumn but were uncommon in relation to other species in subtropical and temperate estuaries. Sizes of C. hippos at temperate latitudes were as large as conspecifics at subtropical latitudes and juveniles of other species that are known to migrate during autumn from temperate nursery grounds to subtropical latitudes. As C. hippos disappeared from estuaries of the middle Atlantic states in autumn, similar-size fish appeared on the inner continental shelf. We postulate that at least some of the C. hippos observed migrating from temperate estuaries during the autumn eventually overwinter at subtropical latitudes, where they can return to the spawning population. This is unusual, because individuals of many other species whose larvae are transported north of Cape Hatteras do not appear to successfully migrate back to subtropical overwintering habitats. This life-history pattern, in which fish begin their first year in the Carolinian biogeographic province, are dispersed to the Virginian province, and return to the Carolinian province before their first winter, has been demonstrated for only one other western North Atlantic finfish species: bluefish (Pomatomus saltatrix). A few other species are likely to occupy and reproduce within such large-scale oceanographic systems because they have a combination of spawning, larval, and juvenile traits that is similar to that of P. saltatrix and C. hippos.

In the western North Atlantic, many species spawn at tropical and subtropical latitudes, and their larvae are dispersed to temperate and even boreal regions (e.g. Gill, 1904; Markle et al., 1979; Curran, 1989). This transport of larvae occurs in association with the Gulf Stream (Wroblewski and Cheney, 1984; Hare and Cowen, 1991, 1996) and disperses propagules from the Carolinian to the Virginian biogeographic province (Briggs, 1974. 1996). This process, which involves the entrainment of eggs and larvae into oceanographic currents and the dispersal of propagules to various water masses, is understood for only a few species and may vary considerably between species. Perhaps the most famous example is that of catadromous freshwater eels (Anguilla anguilla and A. rostrata), which spawn in the Sargasso Sea and whose larvae are transported to coastal nurseries in North America and Europe (McCleave et al., 1987). Larval A. rostrata drift for up to one year (Kleckner and McCleave, 1985) through gyres in the southwestern Sargasso Sea, the Antilles Current, and the Florida Current and its Gulf Stream extension before moving into coastal waters (McCleave, 1993). In the case of a wrasse, Xyrichtys novacula, northward transport across hundreds of kilometers occurs in as little as eight days, and cross-shelf transport in the northern portion of the Carolinian province is aided by interactions between western-edge, warm-core eddies and the Gulf Stream itself (Hare and Cowen, 1991). Similar northward transport in association with the Gulf Stream occurs for "spring-spawned" bluefish, Pomatomus saltatrix, except that for this species dispersal time is longer--about 2 months (McBride and Conover, 1991)--and its juveniles may actively swim out of the Gulf Stream system (Hare and Cowen, 1996). Although some species are transported annually in this manner (e.g. Pomatomus), others are transported less regularly (Xyrichtys, Bothus, Syacium), and still other species do not move between slope and shelf water masses (Cowen et al., 1993).

Those individuals dispersed across biogeographic boundaries become expatriated, in terms of passing on their genes if they do not return to spawning grounds to reproduce. Anguilla species remain in coastal habitats for several years before migrating back to the Sargasso Sea (Tesch, 1977), but many species that spawn at subtropical latitudes are physiologically incapable of overwintering at temperate latitudes. To our knowledge, only P. saltatrix spawns at subtropical latitudes and its propagules, once dispersed to temperate latitudes, are known to migrate south within the same year. This species is successful largely because it spawns early in the year in the subtropics and juveniles enter temperate estuaries at a time when their principal prey is becoming available (Juanes et al., 1994). Direct evidence of a successful autumn migration by young-of-the-year P. saltatrix is available from recaptures of tagged individuals and size-frequency analyses (Lund and Maltezos, 1970; McBride et al., 1993). This "spring-spawned" cohort of P. saltatrix was more abundant than other intraspecific cohorts in the western North Atlantic during the 1980s and early 1990s (Chiarella and Conover, 1990; McBride et al., 1993), and this species repeats this life-history pattern in several oceanographic systems worldwide (Juanes et al., 1996).

The effects of this transport process, from the Carolinian province to the Virginian province, is known for few other species and these examples suggest a different fate for these species than for P. saltatrix. Moss (1973) concluded that the "sluggish" swimming ability and the critical thermal minimum of 8.7 [degrees] C for planehead filefish, Monocanthus hispidus, prevented individuals of this species from surviving once they had been dispersed to temperate latitudes. Hare and Cowen (1991) observed a wrasse, Xyrichtys novacula, settling on the continental shelf at temperate latitudes, but successive cruises could not find survivors from the initially observed cohort. And McBride and Able (1998) reported on the annual appearance of butterfly-fishes (Chaetodon spp.) at temperate latitudes, but field collections and laboratory experiments led them to conclude that none of these individuals survived through their first winter. Sinclair (1988) would categorize these unfortunate individuals as "vagrants" from, as opposed to "members" of, a population.

Although the ecological information about these tropical forms as they occur in temperate waters is limited, the common paradigm is that "[the] Gulf Stream has a profound effect upon the distribution of shore animals in the western Atlantic ... many tropical forms are left stranded along the inhospitable shores of northeastern North America." (Briggs, 1996; p. 238). We believe that Briggs's conclusion can be applied to most of these species; nonetheless, juveniles of several "southern" species grow and survive well at temperate latitudes during the summer and, based on their size by autumn and their general mobility, may migrate successfully back to subtropical latitudes before winter (see also Hare and Cowen, 1993). We propose that at least one other species is capable of following a dispersal-migration pattern that links the temperate and subtropical latitudes during its first year. The crevalle jack, Caranx hippos (Linnaeus), is a migratory, coastal species that is distributed worldwide at subtropical and tropical latitudes (Briggs, 1960; Kwei, 1978; Grosslein and Azarovitz, 1982). In the western North Atlantic, it occurs primarily in the southeastern United States and the Gulf of Mexico (U.S. Department of Commerce, 1986), and spawning is known to occur only in the subtropical Straits of Florida (Berry, 1959; Fahay, 1975) and in the tropical Caribbean Sea (Montolio, 1978). Young-of-the-year (YOY) C. hippos occur in subtropical estuaries of the south Atlantic states, and they use these habitats as nurseries before presumably migrating offshore in autumn (Berry, 1959). However, YOY C. hippos are also reported from many temperate estuaries north of Cape Hatteras (McBride(1)). Despite the broad geographic distribution of YOY C. hippos, little is known about their early life history at any latitude. From archived museum collections, recent field collections, and the literature, we assembled data about C. hippos in both southern and northern estuaries and postulate on the significance of dispersal of YOY C. hippos across two biogeographic provinces.

Materials and methods

Archival collections were examined at the New York State Museum, the American Museum of Natural History, and the Academy of Natural Sciences of Philadelphia, and collection dates, locations, and sizes of C. hippos from coastal habitats of New York and New Jersey were recorded (material examined is listed in McBride(1)). Similar data for C. hippos in subtropical estuaries (from Cape Hatteras, North Carolina, to Cape Sable, Florida) were taken from Berry (1959). Densities and size data from samples collected in Great South Bay (New York) and southern New Jersey embayments during 1987 and 1988 were also examined (see McBride and Conover [1991] for seine-survey design; weir-sample data from Rountree et al.(2)). Fish densities in these other estuaries and from the published literature were calculated as

CPUE = (Catch per unit of effort [number of fish per seine haul])

for the period May-October unless stated otherwise. Field sampling (described below) of temperate coastal habitats was also completed, and abundance, habitats, seasonality, and sizes of C. hippos were examined. Unless noted otherwise, fish size is reported to the nearest cm fork length (FL). A Gompertz model was used to model growth:

FL = [L.sub.[infinity]] exp(-exp{-G[DOY - [X.sub.0]]}),

where FL = fork length in mm;

[L.sub.[infinity]] = asymptotic length;

G = instantaneous rate of growth at age [X.sub.0];

DOY = day of the year, and

[X.sub.0] = inflection point of the curve.

Growth modeling and other statistical analyses were performed by using SAS (SAS, 1990).

Field sampling--Hudson River

Recent data (1986-1993) were analyzed from an ongoing (since 1974) survey of the entire Hudson River (Fig. 1). Fish were collected during daylight over a 1-wk period every other week between mid-June and mid-November. During each sampling week, a 30.5-m x 2.4-m (max. mesh=10 mm) beach seine was set in an arc from a boat at approximately 100 stations. Station locations were selected in a stratified random design by using 12 strata between Manhattan and Albany, New York (from river kilometer [rkm] 19 to 246). The length of each stratum varied from 8 to 44 km, and typically 5-20 hauls were completed in each stratum during each biweekly period. Mean abundance, by month or stratum, was calculated from transformed values of the number of fish in each seine haul ([log.sub.e] [n+1]) and was expressed as a geometric mean (antilogged mean values of -1). Annual abundances were also calculated as geometric means but only for data from August to September (when C. hippos were consistently collected).

[Figure 1 ILLUSTRATION OMITTED]

Field sampling--Haverstraw and Jamaica bays

Further analyses of the ecology of Caranx hippos in the Hudson River were based on data from an independent survey for the same 8-yr period (1986-1993) as the survey of the entire river. This additional survey focused on two subregions of the estuary: Haverstraw Bay and Jamaica Bay. Haverstraw Bay corresponds to strata numbers 2-4 of the Hudson River survey (Fig. 1); Jamaica Bay is a satellite embayment close to the mouth of the Hudson River.

Seining in Haverstraw Bay occurred biweekly from July to November (see also McKown and Young, 1992). Typically, 25 stations were sampled during daylight over a 2-d period; at each station, a single haul was made with a 61-m x 3-m beach seine (max. mesh=6.4 mm) set in an arc from a boat. This seine was used similarly in Jamaica Bay from May to November, but sampling frequency was generally monthly and only 4-5 stations were sampled each time. Temperature and salinity were measured with a hand-held thermometer and refractometer from surface-water samples after the net was hauled. Abundance of C. hippos was calculated as above (i.e. by geometric means).

Field sampling--the continental shelf

To determine seasonal abundance of C. hippos above the U.S. east coast continental shelf, we examined data collected during 1972-1996 in the National Marine Fisheries Service's bottom-trawling program. This program followed a stratified random statistical design to survey fish stocks of the continental shelf (9-366 m) from Cape Fear, North Carolina, to Nova Scotia, Canada. Standard methods were to tow a 24.4- or 30.5-m otter trawl (13-mm codend liner) for 30 min at randomly assigned stations within fixed strata (strata were delimited largely by depth and latitude). All samples combined covered all seasons, but sampling intensity was greatest during spring (March-May) and autumn (September-November), when about 350 stations were sampled during each 6-8 wk period. Fish were measured to the nearest cm. General survey design and its specific applications for other species can be found in many other studies (e.g. Grosslein and Azarovitz, 1982; Despres-Patanjo et al., 1988; McBride et al., 1998).

Results

Latitudinal comparisons

Young-of-the-year C. hippos occurred in subtropical estuaries between North Carolina and Florida from June to November and in temperate estuaries of New York and New Jersey during July-November (Fig. 2). On average, we observed only about one fish collected in three or more seine hauls, and these arithmetic mean densities were the same for temperate and subtropical estuaries (Fig. 3).

[Figures 2-3 ILLUSTRATION OMITTED]

At subtropical latitudes, individuals [is less than] 4.0 cm FL were present in all months, June-November (Fig. 2). Small individuals were also present for three months, July-September, at temperate latitudes. In both biogeographic regions, some individuals had reached approximately 20 cm FL by October and November, suggesting that absolute growth could exceed 1 mm/d. The prolonged appearance of small fish in southern estuaries, in particular, and the lack of age-specific data precluded more detailed comparisons of growth rates.

Estuarine residency

In the Hudson River system, C. hippos occurred annually but its density varied by an order of magnitude among years (Table 1). This species occurred as far upstream as rkm 102 (i.e. lower part of stratum 7) but was most abundant in the upper portion of Haverstraw Bay (Fig. 4A). Seasonally, C. hippos resided in the Hudson River system from July to November and was most common from late July to early October (Figs. 4B, 5A). The median temperature for collections of C. hippos was 26 [degrees] C in Haverstraw Bay and 24 [degrees] C in Jamaica Bay (Fig. 6). Near Haverstraw Bay, where salinities were much lower than those of Jamaica Bay, C. hippos was capable of invading nearly freshwater regions of the Hudson River (Fig. 6). Abundance measured with a 61-m net, set in a comparable manner in both bays, showed that the abundance of C. hippos was much higher in Jamaica Bay than in Haverstraw Bay. Abundance in Great South Bay, New York, was also consistently higher than in Haverstraw Bay (Fig. 5A), even though a smaller, 30-m net was used.

[Figures 4-6 ILLUSTRATION OMITTED]

Table 1 Catch data for all Hudson River sampling sites; a 30.5-m seine or 61-m seine net was used. Geometric mean (GM) number of C. hippos per seine haul is calculated only from collections during peak abundance (August-September).

                                 No. of       Total
                      Total      hauls       no. of
                      no. of   containing   C. hippos
Year                  hauls    C. hippos    collected     GM

Entire Hudson River
(30.5-m seine)                   All sampling strata

1986                   1000         6           10      0.0058
1987                   1101         2            3      0.0038
1988                   1100        12           22      0.0235
1989                   1100        15           40      0.0198
1990                   1000        19           32      0.0284
1991                   1000        11           58      0.0134
1992                   1000         9           53      0.0084
1993                   1000        17           30      0.0171
Total                  8301        91          248          --

Haverstraw and
Jamaica bays
(61-m seine)                        Haverstraw Bay

1986                    222         6           21      0.0398
1987                    225         7            9      0.0406
1988                    220        21           45      0.1578
1989                    225        17           64      0.0730
1990                    217        16           39      0.1274
1991                    215         9           20      0.0649
1992                    221         5            5      0.0140
1993                    225        11           20      0.0437
Total                  1770        92          223          --

                                 No. of       Total
                      Total      hauls       no. of
                      no. of   containing   C. hippos
Year                  hauls    C. hippos    collected     GM

Entire Hudson River
(30.5-m seine)                Haverstraw Bay (strata 2-4)

1986                     429        4            8      0.0135
1987                     473        2            3      0.0089
1988                     407        6           12      0.0344
1989                     429       14           39      0.0468
1990                     364       13           23      0.0395
1991                     364       11           58      0.0341
1992                     364        8           50      0.0130
1993                     364       15           27      0.0403
Total                  3,194       73          220          --

Haverstraw and
Jamaica bays
(61-m seine)                         Jamaica Bay

1986                      38        2           10      0.732
1987                      56        4           14      0.468
1988                      42        4           54      1.151
1989                      49        2            2      0.000
1990                      48        7           11      0.475
1991                      58        4            6      0.000
1992                      44        2            3      0.292
1993                      43        3           11      0.000
Total                    378       28          111         --

Fork length offish collected during 1986-1993 in the Hudson River estuary ranged from 2.9 to 17.6 cm. Growth rate was modeled by using data for New York and New Jersey estuaries (Fig. 5B; n=439), and the resulting equation,

FL = 136.6exp(-exp{-0.0315[DOY - 211]}), ([r.sup.2]=0.75),

predicted a peak instantaneous growth rate for July 30, after which growth slowed and fish reached a mean asymptotic length of 13.7 cm. The distribution of residuals for this growth model suggested that the asymptotic length was biased (i.e. lowered) by the presence of small, presumably young, juveniles in August-September; therefore we consider these growth estimates to be preliminary.

Continental shelf distributions

A total of 657 C. hippos were collected at 134 stations between 33 [degrees] 47'N and 41 [degrees] 00'N and at depths from 6 to 38 m (Fig. 7A). Although present in all but four years during 1972-1996, C. hippos were collected only in the months of July (1 of 134 stations=0.8%), September (84.3%), October (14.2%), and November (0.8%) during this period. Fish ranged in size from 3 to 29 cm FL, and modal size (17 cm) was the same for fish collected both north and south of 36 [degrees] 00'N (Fig. 7B).

[Figure 7 ILLUSTRATION OMITTED]

Discussion

Densities of C. hippos in Haverstraw Bay, where our study was focused, were generally low compared with other New York estuaries. We did observe considerable variation in densities of C. hippos between months, years, and estuaries, which we believe is at least partially the result of this species' relatively low density and shoaling (e.g. aggregating) behavior (Kwei, 1978). Larval dispersal processes or habitat may also affect juvenile C. hippos densities between temperate estuaries. This variability appears to have led to contradictory reports of C. hippos abundance at temperate latitudes: some investigators concluded that this species is "rare" or "scarce" (e.g. Bean, 1900; Greeley, 1937), whereas others have considered it to be "common" or "abundant" (DeKay, 1842; Smith, 1985).

Our review of seining data across a latitudinal range from 27 [degrees] N to 41 [degrees] N demonstrated that densities of C. hippos propagules transported to subtropical estuaries were not higher, as would be expected for a "subtropical" species, than densities of those propagules transported to temperate estuaries. Not all CPUE values we examine in our study were directly comparable because of potential differences in deployment procedures, sampling habitats, and range of seine sizes. We hope that others will report more data for such comparative purposes in the future. We are also unaware of comparative data for tropical regions to determine if this species is more abundant in a more central portion of its latitudinal range. Young-of-the-year C. hippos abundance is generally low compared with other estuarine finfishes from Massachusetts to Florida (e.g. Ayvazian et al., 1992; Tremain and Adams, 1995; Able et al., 1996), and we conclude that this species is uncommon throughout this range.

Having found juvenile C. hippos in reasonable densities in temperate estuaries, we also observed them to occupy habitats similar to those of juvenile C. hippos in subtropical estuaries. In both temperate and subtropical regions, juveniles of this species appear to use estuaries as nurseries (see also Berry, 1959; Kwei, 1978). We observed individuals as far upstream as the freshwater interface (about 1 ppt), at about rkm 90-100 in the Hudson River during July-October (Cooper et al., 1988; Geoghegan et al., 1992). Several others have reported YOY C. hippos in oligohaline habitats (Christensen, 1965; Tagatz, 1968; Smith, 1985; Beebe and Savidge, 1988) but none in fresh water. The presence of C. hippos within the Hudson River was associated with warm temperatures; this species occurred in similarly warm temperatures in subtropical estuaries (18-33 [degrees] C: Tagatz and Dudley, 1961; Christensen, 1965; Tagatz, 1968). The concentration of fish in and near Haverstraw Bay is probably due to this region's broad width and shallow bathymetry, which slows waterflow rates and responds rapidly to solar radiation (Cooper et al., 1988). In addition, three power plants that release heated effluent are located near Haverstraw Bay (Cooper et al., 1988). Temperature eventually becomes a very important parameter affecting YOY C. hippos distribution, and cold temperatures have been implicated in several cases of overwintering mortality for C. hippos. We know of no experimental study to document the lower lethal temperature for this species, but Bean (1903) noted that C. hippos could overwinter in aquarium conditions above 10 [degrees] C, and Hoff (1971) observed dead C. hippos in waters colder than 9 [degrees] C, whereas conspecifics in a downstream portion of the same Massachusetts river survived in waters above 9 [degrees] C. Water temperatures fall below 10 [degrees] C for at least the 4-mo. period of December-April near Haverstraw Bay and above the continental shelf of the middle Atlantic states (Cooper et al., 1988; Mountain and Holzwarth, 1989); therefore there is no suitable overwintering habitat for C. hippos north of Cape Hatteras. Yet although the predictability of the seasonal cycle may be an important determinant of a species' latitudinal range, sudden and irregular freezes are more likely agents for actual hypothermal mortality, regardless of latitude (Storey, 1937). There are, for example, several instances of hypothermal mortality of C. hippos on Florida's Atlantic and Gulf coasts (Storey and Gudger, 1936; Miller, 1940; Galloway, 1941; Gilmore et al., 1978; Snelson and Bradley, 1978; Provancha et al., 1986). If individuals dispersed to temperate estuaries react in a similar manner to those dispersed to subtropical estuaries and emigrate from estuaries during autumn, then they may successfully migrate south to suitable overwintering habitats. We observed similar sizes of YOY C. hippos disappearing from estuaries and appearing on the continental shelf during September and October, which is strong circumstantial evidence of a migration pattern before hypothermal temperatures are reached.

During the summer, coastal water temperatures at temperate and subtropical latitudes are similar (e.g. Mountain and Holzwarth, 1989) and produce a favorable growth climate for C. hippos across a wide latitudinal range. Monthly sizes of YOY C. hippos were similar across latitudes, and the apparent growth rates were high in relation to other YOY estuarine fish species in temperate waters (Rountree and Able, 1992). Our comparative data suggested no size disadvantage for individuals dispersed to temperate estuaries. Overwintering mortality is size-dependent for some species that migrate offshore (Conover, 1990), so that if C. hippos in temperate waters were smaller by autumn than those in subtropical estuaries, then their overwinter survival could be lower. Nevertheless, individuals in temperate waters may be at a disadvantage to equal-size individuals in subtropical waters because the latter do not need to migrate as far to reach overwintering habitats. Our preliminary growth model was confounded by a prolonged (about 3 months) presence of small fish, and presumably there was considerable age variation within our late-summer samples. The continued presence of small fish into September confounded analysis of length frequencies and created the appearance of sudden shifts in average fish size in late summer. The discrepancy between predicted size in temperate estuaries during autumn and the modal size observed on the shelf (e.g. 14 vs. 17 cm FL) could be the result of smaller fish having a higher mortality rate during this habitat shift. Further work with growth modeling will improve with age-specific data. The growth and size data of our study, however, suggest that C. hippos emigrated from estuaries largely as YOY and at the same time and sizes across latitudes. These data suggest that the same behavioral cue to migrate operated on individuals dispersed to both biogeographic regions.

By the time YOY C. hippos reach shelf habitats during autumn, we postulate that they are of adequate size to continue south to subtropical latitudes. For example, YOY C. hippos are as large as "summer-spawned" YOY P. saltatrix; McBride et al. (1993) concluded the latter migrate from temperate nursery grounds to subtropical overwintering habitats at sizes of 10-15 cm FL. Moreover, several fish species have been tagged in temperate waters and recaptured in subtropical waters at sizes similar to those of C. hippos in autumn, including Atlantic croaker, Micropogonias undulatus (Pearson, 1932; Haven, 1959); spot, Leiostomus xanthurus (Pearson, 1932); weakfish, Cynoscion regalis (Nesbit, 1954); Atlantic thread herring, Opisthonema oglinum (Pristas and Cheek, 1973); bluefish, P. saltatrix (Lund and Maltezos, 1970); and Atlantic menhaden, Brevoortia tyrannus (Kroger and Guthrie, 1973). We reject the hypothesis that YOY C. hippos migrate offshore to the shelf edge, because they are distributed close to shore ([is greater than] 38 m) during autumn and they are not collected in winter or spring sampling out to 366 m depths by the National Marine Fisheries Service's trawling program. We similarly reject the hypothesis that the YOY C. hippos observed in coastal habitats during autumn have moved inshore from "offshore" nursery grounds. This cohort was not evident in summer trawl samples and such a hypothesis would require an alternative life history tactic for using shelf habitats as nurseries--a tactic that up until now has not been reported for this species.

Direct evidence (e.g. mark-recapture) of successful migrations by marine fishes is rare, and the migration patterns of many species are inferred from seasonal changes in distribution (Leggett, 1977), as we have done in our study. In addition, we have examined growth rates and sizes to link the seasonal changes in geographic distribution to the YOY cohort. The above evidence indirectly links some YOY C. hippos that have dispersed to temperate estuaries back to suitable overwintering habitats and suggests that these individuals retain "membership" in the spawning population (sensu Sinclair, 1988). We do not, however, assume that all individuals leave temperate estuaries before hypothermal conditions can develop, nor that all individuals survive an autumn migration to subtropical latitudes. The relative importance of temperate estuaries over subtropical estuaries for YOY C. hippos depends on the relative contribution to future spawning by individuals dispersed to each biogeographic region, but this contribution can not be calculated without further study.

Cape Hatteras represents a major faunal (and floral) break along the U.S. east coast (Pielou, 1979; Briggs, 1996), but the larvae of many species that spawn in coastal habitats, particularly of those that broadcast their eggs into the water column and have moderate-to-long planktonic larval durations, are capable of being transported around this geographic barrier (e.g. Curran, 1989). Among these species, we postulate that there is a subset of species that have juvenile traits that allow them to exploit nursery habitats in both biogeographic provinces, and an additional subset of species for which some individuals can return from temperate estuaries to reproduce. Traits such as early spawning season (i.e. winter or spring), fast growth rate, large YOY body size, generalized habitat requirements, and fast swimming speeds are favorable traits for any species attempting to complete a life-history circuit within one year between subtropical and temperate waters of the western North Atlantic. These traits are found in several finfish species in the Carangidae, Scombridae, and Mugilidae. Grouping species by spawning, larval, and juvenile characteristics and comparing their fate in these large oceanographic systems will help define the appropriate spatial scale for sampling these populations in the western North Atlantic, as well as in other systems such as the Sea of Japan (Nishimura, 1965), the Agulhas Current (Beckley, 1985), and the East Australian Current (Miskiewicz, 1981) where similar dispersal patterns of subtropically spawned species to temperate waters have been described. An initial screening for such characters allows researchers to begin to evaluate whether a specific "subtropical" species can or cannot survive once dispersed to temperate habitats.

Some of these subtropically spawned species are dispersed to temperate waters during less favorable conditions (e.g. winter) and, presumably, others do not grow to a sufficient size or have the behavior characteristics to migrate in autumn. A nonadaptive outcome of dispersal for many species on an ecological timescale is not, however, necessarily surprising. An examination of the paleontological record, which is most complete for invertebrates, shows that biogeographic assemblages lag notably behind environmental changes at an evolutionary timescale (Briggs, 1974; Pielou, 1979). A modeling and theoretical method, such as that used by Hare and Cowen (1993), could help explain why the larvae of so many species are dispersed into large oceanographic systems even when expatriation is likely, but more natural history data are necessary before such a method can be widely applied. A more complete record of which species are and are not expatriated within large-scale systems will contribute to understanding the processes underlying these biogeographic patterns, provide opportunities to examine further the evolution of reproductive strategies, and help explain oceanographic effects on population dynamics.

Acknowledgments

R. Daniels, W. Saul, and C. L. Smith assisted with the collections at the New York State Museum, the Academy of Natural Sciences of Philadelphia, and the American Museum of Natural History, respectively. Data for the entire Hudson River were collected for Consolidated Edison and were made available by J. Young and G. Sentell of EA Engineering, Science, and Technology. B. Young, L. Pasciutti, M. Penski, and other biologists of the New York State Department of Environmental Protection helped collect data for Haverstraw Bay and Jamaica Bay. Trawling data were collected by the Northeast Fishery Science Center, National Marine Fisheries Service, and made available by T. Azarovitz and B. O'Gorman. Valuable discussions or comments were provided by K. Able, F. Berry, E. Blair, D. Conover, R. Crabtree, J. Hare, S. Kaiser, R. Loveland, W. Smith-Vaniz, and J. Waldman. Final data analyses and manuscript preparation were completed while R.S.M. was employed at he Florida Marine Research Institute. Editorial assistance was provided by L. French, J. Leiby, and J. Quinn. This research was supported in part by a Tibor T. Polgar Fellowship to R.S.M. from the Hudson River Foundation. We are grateful to all of the above.

(1) McBride, R.S. 1995. Perennial occurrence and fast growth rates by crevalle jacks (Carangidae: Caranx hippos) in the Hudson River estuary. In Final reports of the Tibor T. Polgar Fellowship Program, 1994 (E. A. Blair and J. R. Waldman, eds.), p. VI1-VI29. Hudson River Foundation, New York, NY.

(2) Rountree, R. A., K. J. Smith, and K. W. Able. 1992. Length frequency data for fishes and turtles from polyhaline subtidal and intertidal marsh creeks in southern New Jersey. Institute of Marine and Coastal Science (IMCS) report 92-34, Rutgers, the State Univ. of New Jersey, New Brunswick, NJ, 08903, 165 p.

Literature cited

Able, K. W., D. A. Witting, R. S. McBride, R. A. Rountree, and K. J. Smith. 1996. Fishes of polyhaline estuarine shores in Great Bay-Little Egg Harbor, New Jersey: a case study of seasonal and habitat influences. In Estuarine shores: environments and human alterations (K. F. Nordstrom and C. T. Roman, eds.), p. 335-355. John Wiley and Sons, Ltd. New York, NY.

Anderson, W. D., Jr., J. K. Dias, R. K. Dias, D. M. Cupka, and N. A. Chamberlain. 1977. The macrofauna of the surf zone off Folly Beach, South Carolina. U.S. Dep. Commer., NOAA Tech. Rep. NMFS-SSRF 704, 23 p.

Ayvazian, S. G., L. A. Deegan, and J. T. Finn. 1992. Comparison of habitat use by estuarine fish assemblages in the Acadian and Virginian zoogeographic provinces. Estuaries 15:368-383.

Bean, T. H. 1900. Report on the fishes of Long Island collected in the summer of 1898. N.Y. State Mus. Rep. 52. vol. 1, p. r103.

1903. Catalogue of the fishes of New York. Bull. N.Y. State Mus. 60, Zoology 9:428-429.

Beckley, L. E. 1985. The fish community of East Cape tidal pools and an assessment of the nursery function of the habitat. South African J. Zool. 20:21-27.

Beebe, C. A., and I. R. Savidge. 1988. Historical perspective on fish species composition and distribution in the Hudson River estuary. Am. Fish. Soc. Monogr. 4:25-36.

Berry, F. H. 1959. Young jack crevalles (Caranx species) off the southeastern Atlantic coast of the United States. Fish. Bull. 59:417-535.

Bigelow, H. B., and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. U.S. Fish and Wildl. Serv. Bull.74, vol. 53, 577 p.

Briggs, J. C. 1960. Fishes of worldwide (circumtropical) distribution. Copeia 1960:171-180.

1974. Marine zoogeography. McGraw-Hill Book Co, New York, NY, 475 p.

1996. Global biogeography. Elsevier, Amsterdam, The Netherlands, 452 p.

Briggs, P. T. 1975. Shore-zone fishes of the vicinity of Fire Island Inlet, Great South Bay, New York, N.Y. Fish Game J. 22:1-12.

Briggs, P. T., and J. S. O'Connor. 1971. Comparison of shore-zone fishes over naturally vegeetated and sand-filled bottoms in Great South Bay. N.Y. Fish Game J. 18:16-41.

Chiarella, L. A., and D. O. Conover. 1990. Spawning season and first-year growth of adult bluefish from the New York Bight. Trans. Am. Fish. Soc. 119:455-462.

Christensen, R. F. 1965. An ichthyological survey of Jupiter Inlet and Loxahatchee River, Florida. M.S. thesis. Fla. State Univ., Tallahassee, FL, 318 p.

Conover, D. O. 1990. The relation between capacity for growth and length of growing season: evidence for the implications of counter-gradient variation. Trans. Am. Fish. Soc. 119:416-430.

Cooper, J. C., F. R. Cantelmo, and C. E. Newton. 1988. Overview of the Hudson River estuary. Am. Fish. Soc. Monogr. 4:11-24.

Cowen, R. K., J. A. Hare, and M. P. Fahay. 1993. Beyond hydrography: can physical processes explain larval fish assemblages within the Middle Atlantic Bight? Bull. Mar. Sci. 53:567-587.

Curran, M. C. 1989. Occurrence of tropical fishes in New England waters. In Proceedings of the American Academy of Underwater Science; Ninth Ann. Sci. Diving Symp. Sept. 28-Oct. 1, 1989 (M. A. Lang and W. C. Jaap, eds.), p. 71-82. Am. Academy of Underwater Sciences, Costa Mesa, CA.

DeKay, J. E. 1842. Zoology of New York or the New York Fauna. Part III: reptiles and amphibia, p: 120-121. W. & A. White & J. Visscher, Albany, NY.

Despres-Patanjo, L. I., T. R. Azarovitz, and C. J. Byrne. 1988. Twenty-five years of fish surveys in the northwest Atlantic: the NMFS Northeast Fisheries Center's Bottom Trawl Survey Program. Mar. Fish. Rev. 50:69-71.

Fahay, M. P. 1975. An annotated list of larval and juvenile fishes captured with surface-towed meter net in the South Atlantic Bight during four RV Dolphin cruises between May 1967 and February 1968. U.S. Dep. Commer., NOAA Tech. Rep. NMFS-SSRF-685, 39 p.

Futch, C. R., and S. E. Dwinell. 1977. Nearshore marine ecology at Hutchinson Island, Florida: 1971-1974. IV: Lancelets and fishes. Fla. Mar. Res. Publ. 24, 23 p.

Galloway, J. C. 1941. Lethal effect of the cold winter of 1939-40 on marine fishes at Key West, Florida. Copeia 1941:118-119.

Geoghegan, P., M. T. Mattson, J. J. Reichle, and R. G. Keppel. 1992. Influence of salt front position on the occurrence of uncommon marine fishes in the Hudson River Estuary. Estuaries 15:251-254.

Gill, T. 1904. State ichthyology of Massachusetts. Science (new series) 20:321-338.

Gilmore, R. G., Jr. 1988. Subtropical seagrass fish communities: population dynamics, species guilds and microhabitat associations in the Indian River Lagoon, Florida. Ph.D. diss., Fla. Inst. Tech. Melbourne, FL, 199 p.

Gilmore, R. G., Jr., L. H. Bullock, and F. H. Berry. 1978. Hypothermal mortality in marine fishes, south-central Florida, January, 1977. Northeast Gulf Sci. 2:77-97.

Greeley, J. R. 1937. Fishes of the area with annotated list. In Survey of the lower Hudson watershed, p. 45-103. State of New York Conservation Department, Albany, NY.

Grosslein, M. D., and T. R. Azarovitz. 1982. Fish distribution. MESA New York Bight Atlas Monograph 15. New York Sea Grant Institute, Albany, NY, 182 p.

Gunter, G., and G. E. Hall. 1963. Biological investigations of the St. Lucie estuary (Florida) in connection with Lake Okeechobee discharges through the St. Lucie canal. Gulf Res. Rep. 1:189-307.

Hare, J. A., and R. K. Cowen. 1991. Expatriation of Xyrichtys novacula (Pisces: Labridae) larvae: evidence of rapid cross-slope exchange. J. Mar. Res. 49:801-823.

1993. Ecological and evolutionary implications of the larval transport and reproductive strategy of bluefish, Pomatomus saltatrix. Mar. Ecol. Prog. Ser. 98:1-16.

1996. Transport mechanisms of larval and pelagic juvenile bluefish (Pomatomus saltatrix) from South Atlantic Bight spawning grounds to Middle Atlantic Bight nursery habitats. Limnol. Oceanogr. 41:1264-1280.

Haven, D. S. 1959. Migration of the croaker, Micropogon undulatus. Copeia 1959:25-30.

Hoff, J. G. 1971. Mass mortality of the crevalle jack, Caranx hippos (Linnaeus) on the Atlantic coast of Massachusetts. Chesaapeake Sci. 12:49.

Hoff, J. G., and R. M. Ibara. 1977. Factors affecting the seasonal abundance, composition and diversity of fishes in a southeastern New England estuary. Estuarine Coastal Marine Sci. 5:665-678.

Juanes, F., J. A. Buckel, and D. O. Conover. 1994. Accelerating the onset of piscivory: intersection of predator and prey phenologies. J. Fish Biol. 45 (suppl. A):41-54.

Juanes, F., J. A. Hare, and A. G. Miskiewicz. 1996. Comparing early life history strategies of Pomatomus saltatrix: a global approach. Mar. Freshwater Res. 47:365-79.

Kleckner, R. C., and J. D. McCleave. 1985. Spatial and temporal distribution of American eel larvae in relation to North Atlantic ocean current systems. Dana 4:67-92.

Kroger, R. L., and J. F. Guthrie. 1973. Migrations of tagged juvenile Atlantic menhaden. Trans. Am. Fish. Soc. 102:417-422.

Kwei, E. A. 1978. Food and spawning activity of Caranx hippos (L.) off the coast of Ghana. J. Nat. Hist. 12:195-215.

Leggett, W. C. 1977. The ecology of fish migrations. Ann. Rev. Ecol. Syst. 8:285-308.

Lund, W. A., and G. C. Maltezos. 1970. Movements and migrations of the bluefish, Pomatomus saltatrix, tagged in waters of New York and Southern New England. Trans. Am. Fish. Soc. 99:719-725.

Marcellus, K. L. 1972. Fishes of Barnegat Bay, New Jersey, with particular reference to seasonal influence and the possible effect of thermal discharge. Ph.D. diss., Rutgers Univ., New Brunswick, NJ, 172 p.

Markle, D. F., W. B. Scott, and A. C. Kohler. 1979. New and rare records of Canadian fishes and the influence of hydrography on resident and nonresident Scotian shelf ichthyofauna. Can. J. Fish. Aquat. Sci. 37:49-65.

McBride, R. S., and K. W. Able. 1998. Ecology and fate of butterflyfishes Chaetodon spp. in the temperate, western North Atlantic. Bull. Mar Sci. 63(2):401-416.

McBride, R. S., and D. O. Conover. 1991. Recruitment of young-of-the-year bluefish (Pomatomus saltatrix) to the New York Bight: Variation in abundance and growth of spring- and summer-spawned cohorts. Mar. Ecol. Prog. Ser. 78:205-216.

McBride, R. S., J. B. O'Gorman, and K. W. Able. 1998. Interspecific comparisons of searobins (Prionotus spp.) movements, size structure, and abundance in the temperate western North Atlantic. Fish. Bull. 96:303-314.

McBride, R. S., J. L. Ross, and D. O. Conover. 1993. Recruitment of bluefish (Pomatomus saltatrix) to estuaries of the South Atlantic Bight. Fish. Bull. 91:389-395.

McCleave, J. D. 1993. Physical and behavioural controls on the oceanic distribution and migration of leptocephali. J. Fish Biol. 43 (suppl. A):243-273.

McCleave, J. D., R. C. Kleckner, and M. Castonguay. 1987. Reproductive sympatry of American and European eels and implications for migration and taxonomy. Am. Fish. Soc. Symp. 1:286-297.

McKown, K. A., and B. H. Young. 1992. Effects of year class strength on size of young-of-the-year striped bass. In Estuarine research in the 1980's (C. L. Smith, ed.), p. 265-276. State Univ. New York Press. Albany, NY.

Miller, E. M. 1940. Mortality of fishes due to cold on the southeast Florida coast. Ecology 21:420-421.

Miskiewicz, A. G. 1981. The distribution of tropical/coral reef fish larvae in temperate waters along the New South Wales coast of Australia. Rapp. P.-V. Cons. Int. Explor. Mer 191:452.

Montolio, M. 1978. Algunos aspectos sobre el desove y las concentraciones larvarias de las especies de la familia Carangidae en el Mar Caribe. Rev. Cubana Invest. Pesqueras 3:29-49.

Moss, S. A. 1973. The responses of planehead flefish, Monocanthus hispidis (Linnaeus), to low temperature. Chesapeake. Sci. 14:300-303.

Mountain, D. G., and T. J. Holzwarth. 1989. Surface and bottom temperature distribution for the northeast continental shelf. U.S. Dep. Commer., NOAA Tech. Memo. NMFS-F/NEC-73, 32 p.

Nesbit, R. A. 1954. Weakfish migration in relation to its conservation. U.S. Fish Wildl. Spec. Sci. Rep.-Fish. 115, 81 p.

Nishimura, S. 1965. The zoogeographical aspects of the Japan Sea, part 1. Publ. Seto Mar. Biol. Lab. 13:35-79.

Ogburn, M. V., D. M. Allen, and W. K. Michener. 1988. Fishes, shrimps, and crabs of the North Inlet Estuary, SC: a four-year seine and trawl survey. Baruch Institute Tech. Rep. 88-1, Univ. South Carolina. Columbia, SC, 299 p.

Pearson, J. C. 1932. Winter trawl fishery off the Virginia and North Carolina coasts. U.S. Bur. Fish. Invest. Rep. 10, 31 p.

Peters, D. J. 1984. Seasonality, residency, and spatial distribution of juvenile surf zone fishes of the Florida east coast. M.S. thesis, Fla. Inst. Tech., Melbourne, FL, 66 p.

Pielou, E. C. 1979. Biogeography. John Wiley & Sons. New York, NY, 351 p.

Pristas, P. J., and R. P. Cheek. 1973. Atlantic thread herring (Opisthonema oglinum) movements and population size inferred from tag returns. Fish. Bull. 71:297-301.

Provancha, M. J., P. A. Schmalzer, and C. R. Hall. 1986. Effects of the December 1983 and January 1985 freezing air temperatures on select aquatic poikilotherms and plant species of Merritt Island, Florida. Fla. Sci. 49:199-212.

Richards, C. E., and M. Castagna. 1970. Marine fishes of Virginia's eastern shore (inlet and marsh, seaside waters). Chesapeake Sci. 11:235-248.

Rountree, R. A., and K. W. Able. 1992. Fauna of polyhaline subtidal marsh creeks in southern New Jersey: composition, abundance, and biomass. Estuaries 15:178-220.

SAS Institute, Inc. 1990. SAS/STAT user's guide, vols. 1 and 2. SAS Institute, Inc., Cary, NC, 1686 p.

Schooley, J. K. 1977. Factors affecting the distribution of the nearshore fishes in the lagoonal waters of the Indian River, Florida. M.S. thesis, Univ. Florida. Gainesville, FL, 165 p.

Sinclair, M. 1988. Marine populations: an essay on population regulation and speciation. Washington Sea Grant Program. Seattle, WA, 252 p.

Smith, C. L. 1985. The inland fishes of New York State. The New York State Department of Environmental Conservation. Albany, NY, 522 p.

Snelson, F. F., and W. K. Bradley Jr. 1978. Mortality of fishes due to cold on the east coast of Florida, January, 1977. Fla. Sci. 41:1-12.

Storey, M. 1937. The relation between normal range and mortality of fishes due to cold at Sanibel Island, Florida. Ecology 18:10-26.

Storey, M., and E. W. Gudger. 1936. Mortality of fishes due to cold at Sanibel Island, Florida, 1886-1936. Ecology 17:640-648.

Tagatz, M. E. 1968. Fishes of the St. Johns River, Florida. Quart. J. Fla. Acad. Sci. 30:25-50.

Tagatz, M. E., and D. L. Dudley. 1961. Seasonal occurrence of marine fishes in four shore habitats near Beaufort, N.C., 1957-60. U.S. Fish Wildl. Spec. Sci. Rep. Fish 390, 19 p.

Tesch, F.-W. 1977. The eel: biology and management of Anguillid eels. Chapman and Hall. London, United Kingdom, 434 p.

Tremain, D. M., and D. H. Adams. 1995. Seasonal variations in species diversity, abundance, and composition of fish communities in the northern Indian River Lagoon, Florida. Bull. Mar. Sci. 57:171-192.

U. S. Department of Commerce. 1986. Marine recreation fishery statistics survey, Atlantic and Gulf coasts, 1985. Current Fishery Statistics 8327. Natl. Mar. Fish. Serv. Washington, D.C., 21 p.

Wroblewski, J. S., and J. Cheney. 1984. Ichthyoplankton associated with a warm core ring off the Scotian Shelf. Can. J. Fish. Aquat. Sci. 41:294-303.

(*) Contribution 2000-06 of the Institute of Marine and Coastal Sciences, Rutgers Univerisity, Tuckerton, New Jersey 08087.

Richard S. McBride
Rutgers University Marine Field Station
Institute of Marine and Coastal Sciences
800 Great Bay Boulevard
Tuckerton, New Jersey 08087
Present address: Florida Marine Research Institute
100 Eighth Avenue Southeast
St. Petersburg, Florida 33701-5095
E-mail address: [e-mail]

Kim A. McKown
New York State Department of Conservation
205 Belle Meade Road
East Setauket, New York 11733

Manuscript accepted 15 December 1999.