Most benthic invertebrates and reef-associated fish undergo a dispersive, planktonic larval stage prior to settlement and metamorphosis into the juvenile and adult stages. In some species, settlement may be decoupled from adult abundance at local spatial scales if hydrodynamic conditions or larval behavior do not promote local retention. Similarly, spatial variability in postsettlement mortality or secondary dispersal by juveniles and adults may decouple spatial patterns of adult abundance from those of settlement. as a consequence, spatial patterns of settlement and adult abundance may be functionally related in a complex fashion. Whether biotic/environmental factors control spatial patterns of abundance may have profound implications for conservation of exploited benthic marine species, particularly when patterns of exploitation are themselves spatially structured as they are under management by marine reserves. As part of this dissertation, a spatially-explicit population dynamics model for the Caribbean spiny lobster in Exuma Sound, Bahamas was developed. The model is stage- and age-structured, and features dispersal of larvae from hatching sites via advection by hydrodynamic currents and diffusion, active migration of postlarvae into shallow nursery habitats, density-dependent survival and dispersal of benthic life-history stages (juveniles and adults), size-specific fecundity, and spatially-explicit exploitation rates. The population dynamics model was used heuristically to investigate the joint effects of reserve design (i.e., size, location, number), exploitation, population regulation and larval dispersal via hydrodynamic currents on population abundance and fishery yield. Principal findings were that fishery yield and larval production were idiosyncratic functions of reserve size, substantially influenced by interactions between current patterns and reserve location. Also, management strategies which implemented a single large reserve outperformed those using a network of small reserves, a reduction in total effort, or no action whatsoever. Results support the efficacy of marine reserves as a tool for rebuilding overexploited marine populations and creating sustainable fisheries. However, haphazard reserve creation may lead to a false sense of security, and poorly-designed reserves can perform worse than taking no action at all. Thus, designing successful marine reserves requires knowledge of local and regional patterns of hydrodynamic transport and larval dispersal, as well as other species' life-history characteristics.

Over the past 100 years, riparian vegetation communities throughout the Southwest United States have been extensively invaded by Tamara spp. (saltcedar). Saltcedar derives its common name from its physiological adaptation to excrete salts. The production of Tamarix detritus with associated secondary chemicals may affect the quality of aquatic invertebrate food and habitat resources. An alteration in food and habitat quality may affect the composition and structure of aquatic invertebrate assemblages. A series of experiments was conducted contrasting aquatic invertebrate assemblage densities, colonization rates, and growth rates associated with Tamarix versus native vegetation, Populus fremontii (cottonwood) and Salix exigua (willow), to determine if aquatic invertebrate assemblages have been altered by the invasion of Tamarix. Results of invertebrate growth rates over 13 weeks indicate that Tamarix is minimally different in food quality to cottonwood and willow. I failed to find differences in invertebrate colonization rates or invertebrate assemblage densities associated with Tamarix compared to cottonwood and willow over two 6-week time periods.

Carbon export from the land to the ocean are an important part of the global carbon cycle, linking terrestrial watersheds and the global carbon cycle. Burial of terrestrial organic carbon represents a long term sink for atmospheric CO2. Approximately 0.4 Pg Cy-1 is delivered to the global ocean from rivers, equally divided between POC and DOC. However, the amount of carbon entering the ocean is a small portion of the total amount entering rivers from the terrestrial environment, suggesting a large amount of processing in inland waters and estuaries. Most monitoring efforts have focused the processing of organic matter on baseflow conditions. However, recent studies have shown that POC and DOC exported during storm events, a small time period during a hydrologic year, can account for the majority of the annual carbon exported from small watersheds. This dissertation identifies the impact different magnitudes of storm events have on the source, composition and reactivity of organic carbon released to downstream waters from the terrestrial environment at Taskinas Creek, Virginia. The proximity of the Creek to the York River estuary, the changes in water table at the site, along with the small size of the watershed allowing opportunity to examine the connectivity between the watershed processes and delivery of organic matter made the site ideal for identifying how hydrology and environment alter POM and DOM export and reactivity. The sources, composition and flux of DOM and POM were measured during four storm events of different magnitudes to determine how events impacted the sources and fluxes of organic matter and the % reactive DOC exported. Events of different magnitudes with varying sources of DOC and POC had similar % reactive DOC that was not predicted using excitation emission spectroscopy. The events resulted in DOC fluxes 1.5-490 fold higher than baseflow. POC fluxes for storm were 6.7-55 times higher than DOC fluxes. Although the % reactive DOC did not increase during storm event conditions, coupled with the overall flux, storm events represent a considerable pulse of % reactive DOC to downstream waters, well above baseflow levels. When considered with increases in storm intensity due to climate change, storm event fluxes of reactive OM may have broad impacts on estuaries and the global carbon cycle through changes in carbon storage.

Phylogenetic and historical biogeographic hypotheses were explored for three groups of teleost fishes that are hypothesized to share a common biogeographic track. Distributed across the eastern Pacific/eastern Atlantic (EP/EA) biogeographic track of Rosen (1975) are the Scomberomorus regalis (Scombridae), Strongylura marina (Belonidae) and Hyporhamphus unifasciatus (Hemiramphidae) species groups comprised of one eastern Atlantic; three, three and four western Atlantic; and two, two and four eastern Pacific species, respectively, for each group. In addition, two species in the Strongylura marina and Hyporhamphus unifasciatus groups, have invaded freshwater drainages of Central and South America. Each of the three species groups were found to be monophyletic using on average 3000 base pairs (bps) of nucleotide sequence. Genes sequenced included: the entire ATP synthase 8 (ATPase 8; 169 bp) and 6 (ATPase 6; 684 bp), NADH ubiquione oxidoreductase subunit 2 (ND2; 1047 bp), and partial cytochrome b (cyt b; 800 bp) mtDNA protein coding genes; partial 12S (12S; 429 bp) and 16S (16S; 576 bp) ribosomal mtDNA genes; and portions of the nuclear encoded aldolase exon 5/intron 5 (ALD; &\sim&400bp) and creatine kinase b introns and exons 2 and 3 (CK; &\sim&1042 bp). The analysis of multiple genes permitted an evaluation of their phylogenetic signal and usefulness in biogeographic reconstructions. Mitochondrial protein-coding genes resulted in well-supported resolution of sister species relationships, while the nuclear-encoded genes helped resolve basal relationships. However, contrary to expectations, the mtDNA ribosomal genes did not perform as well. Additionally, in Scomberomorus a morphology-based phylogeny was included in the analysis. In general phylogenetic hypotheses inferred for each species group were consistent with expectations of the EP/EA generalized biogeographic track. The results for each species group indicated a basal position for the eastern Atlantic taxa S. tritor, St. senegalensis and Hy. picarti, with closest relationships being either transisthmian (geminate) or, within the eastern Pacific or western Atlantic, for the derived New World taxa. Vicariant and dispersal hypotheses, consistent with paleogeological and paleoceanographic evidence, and life history and dispersal abilities are offered to explain the biogeographic histories of.

Otoliths, scales, dorsal spines, and pectoral fin rays, of Atlantic croaker, Micropogonias undulatus, were compared for legibility of presumed annuli and precision in repeated readings, to determine the best structure for ageing. Marks on transverse otolith sections were easiest to read and showed the best agreement between readings. Fish collected in Chesapeake Bay and in Virginia and North Carolina coastal waters during 1988-1991 were then aged using otolith sections. Ages 1-8 were recorded, but eight-year-old fish were rare. Marginal increment analysis showed that for ages 1-7 annuli are formed once a year during the period April-May. Otolith age readings were very precise, with percent agreement within and between readers greater than 99%. Observed lengths-at-age were highly variable and showed a rapid decrease in growth after the first year. Observed lengths for ages 1-7 showed a very good fit to the von Bertalanffy growth model (r&\sp2& = 0.99; n = 753). No differences were found between sexes. Total annual instantaneous mortality (Z) ranged from 0.55 to 0.63. Atlantic croaker are multiple spawners with indeterminate fecundity. Mean length at first maturity for males and females was 182 and 173 mm TL, respectively. More than 85% of both sexes were mature by the end of their first year and all were mature by age 2. Spawning extends over a protracted period (July-December), but individual fish spawns for only 2-3 months. Spawning starts in Chesapeake Bay and continues offshore and south as Atlantic croaker migrate from the estuary. Seasonal fluctuations in sex ratios suggest that males start leaving the estuary earlier than females. A high incidence of atretic advanced yolked oocytes in spawning females suggests that a surplus production of yolked oocytes is part of Atlantic croaker reproductive strategy. Yield-per-recruit modeling results indicated that, over a likely range on natural mortality values, present levels of harvest in Chesapeake Bay are below the maximum potential yield-per-recruit. Results from this study do not indicate the existence of a group of larger, older fish in the Chesapeake Bay region and suggest that the hypothesis of a different population dynamics pattern for Atlantic croaker north and south of North Carolina, should be reevaluated.

Catch, effort, fleet size and boat owner expenditure data were collected on Virginia's recreational marlin/tuna fishery for the 1983-1985 seasons. Logbooks, dockside interviews and a telephone survey were evaluated to determine which method was the most efficient and effective for collecting and estimating catch and effort for Virginia's pelagic recreational fishery. In 1984, logbooks were used to collect catch and effort data and fishing effort was estimated using Bochenek's method. Very few fishermen returned their logbooks and as a result this data is probably less reliable than the data collected in other years. Due to the poor return of logbooks, this method should not be used to assess Virginia's marlin/tuna fishery. For the 1985 season, Figley's telephone survey (1984) was compared to the NMFS dockside interview technique for large pelagics. Both the telephone survey using Figley's technique (1984) and dockside interviews using Bochenek's method for calculating effort appear to provide similar estimates of projected total catch. However, the dockside method is very labor intensive, costly and fraught with problems in estimating fishing effort. Therefore, the telephone survey technique using Figley's method for estimating effort appears to be a better method for analyzing this fishery. If telephone interviewing will not work in an area and dockside sampling methods must be relied upon to study the pelagic fishery, Bochenek's method appears to produce a better estimate of fishing effort. Using Figley's (1984) mark-recapture technique, Virginia's pelagic recreational fleet was estimated at 455 and 774 vessels in 1983 and 1985, respectively. Boat owner expenditures for this fleet were estimated at &3,863,045 in 1983, \&4,057,020 in 1984 and &5,538,191 in 1985. Bluefin tuna were caught at SST ranging from 58-83 F but seem to prefer SST of 70 to 75 F. Yellowfin tuna were caught at SST ranging from 68-86 F with the majority landed at SST of 76-80 F. White marlin appear to prefer SST of 74 to 81 F.

Perkinsus marinus is the causative agent of the oyster disease Dermo in the eastern oyster, Crassostrea virginica. In vitro propagation of the parasite has led to the establishment of multiple isolates by several investigators. Little work, however, has been done to characterize different isolates. In this study multiple isolates were examined for genetic and biochemical diversity. at two loci, the ITS region and the ATAN region, there was as much intra-isolate genetic variation among DNA sequences of some isolates as there was inter-isolate variation. Variation was also observed at a third loci, a subtilisin-like serine protease gene. This is the first report of a serine protease gene in P. marinus. In addition, a second, very similar subtilisin-like gene, may have also been isolated which was 95% similar in DNA sequence to the first and encoded amino acid changes in conserved regions. Examination of the extracellular proteins produced by eight P. marinus isolates from the Chesapeake Bay revealed differences in protein profiles, protease activity, protease profiles, growth rates, cell size and viability. The degree to which the differences observed were due to strain variation or to complex biochemical interactions of the parasite is unknown. However, isolates of similar viability, growth rates and cell size still had differences in protein and protease band profiles as well as in proteolytic activity, suggesting that there may be some genetic basis for the observed differences. The variation detected at both the genetic and molecular level suggests that multiple isolates should be examined when conducting biochemical and physiological studies on the parasite and for designing molecular diagnostic probes and PCR primers.

The stress protein response has been proposed as a general indicator of exposure to stress as their expression might integrate overall biological impact and interactions among multiple stressors. as a marker of contaminant effects, a major advantage is the premise of higher sensitivity over other indices such as condition index, scope for growth and survival. Laboratory exposure to suspended field contaminated sediments (SFCS) elicited a stress protein response (HSP70) in the eastern oyster, ( Crassostrea virginica). The stress response probably resulted from the combined effect of various contaminants including PAHs, metals and PCBs. Exposure to 1--2g suspended clay particles spiked with PAHs (Fluoranthene, Pyrene, Benzo(e)Pyrene and Benzo(a)Pyrene) caused an increase in the levels of HSP70. No changes in the condition index or mortality were observed. B(e)P accumulated at a higher level compared to other PAHs. This study appears to indicate that the HSP70 provides a sensitive indicator of exposure to PAHs. The effect of PCBs (Aroclor 1242, 1254 and 1260) on the HSP70 was investigated in sexually mature oysters fed algal paste containing 0, 0.10 or 1.00 mug PCBs daily for 15 and 30 days and immature oysters fed algal paste containing 0, 0.35 or 3.50 mug PCBs daily for 56 days. Exposure to PCBs did not produce significant changes in the HSP70 levels. Exposure to 15--30ppb Cd sorbed to algal food and 1--2g suspended clay particles for 40 days generally did not cause changes in HSP70 levels in oysters. Oysters seem to have a high tolerance for Cd toxicity without eliciting a stress protein response. Oysters collected form Point of Shoals in the James River, Virginia, exhibited variation in the HSP70 levels probably related to water temperature. There seems to be an inverse relation between ambient water temperature and levels of stress protein (i.e. low temperature related to high levels of stress protein). The stress protein response has potential as a biomarker of exposure to contaminants. However, the target of the contaminant, mechanisms of coping with exposure, synergistic interactions of contaminants, endogenous factors, and adaptations and life history traits of organisms may influence the stress protein response.

During spring 1990 and 1991, ichthyoplankton surveys were conducted in lower Chesapeake Bay to estimate seasonal egg production, population biomass and the impact of predation by gelatinous zooplankton on early life stages of black drum, Pogonias cromis. Rearing experiments indicated that at least three species of sciaenid (silver perch, Bairdiella chrysoura; weakfish, Cynoscion regalis and P. cromis) were spawning in the survey area during both years. Specific identification of eggs based on previously published ranges of outside egg diameter (OED) were not reliable due to considerable overlap in diameter distributions. Analysis of weekly OED frequency revealed the presence of three modes which differed in temporal occurrence, suggesting the products of three species. Genetic typing of eggs using RFLP analysis of mtDNA confirmed the presence of three species, but demonstrated that eggs of a certain size class represented two species. Results illustrate that reliance on previously published ranges of egg diameter for specific identification of spring spawning sciaenids may overestimate the spawning biomass of black drum by as much as 50% owing to the misidentification of weakfish eggs as those of black drum. Black drum enter Chesapeake Bay in early spring and spawn throughout the day over a discrete time period from late April through May in a spatially limited area off the city of Cape Charles, Virginia. Seasonal egg production is low compared to other sciaenid stocks and estimates of spawner biomass indicate that the population is comprised of relatively few individuals. During 1990 and 1991, peak egg production of black drum occurred in the lower portion of Chesapeake Bay where potential encounter with gelatinous zooplankton predators was least likely. These observations were consistent between years with vastly different gelatinous zooplankton communities. Analysis of small-scale patterns of spatial coincidence between eggs and predators reveal that chances of physical contact are least in areas of peak production.

This work reports the development of an analytical methodology for the analysis of PCB congeners based on integrating relative retention data provided by other researchers. The retention data were transposed into a multiple retention marker system which provided good precision in the calculation of relative retention indices for PCB congener analysis. Analytical run times for the developed methodology were approximately one hour using a commercially available GC capillary column. A Tracor Model 700A Hall Electrolytic Conductivity Detector (HECD) was employed in the GC detection of Aroclor standards and environmental samples. Responses by the HECD provided good sensitivity and were reasonably predictable. Ten response factors were calculated based on the molar chlorine content of each homolog group. Homolog distributions were determined for Aroclors 1016, 1221, 1232, 1242, 1248, 1254, 1260, 1262 along with binary and ternary mixtures of the same. These distributions were compared with distributions reported by other researchers using electron capture detection as well as chemical ionization mass spectrometric methodologies. Homolog distributions acquired by the HECD methodology showed good correlation with the previously mentioned methodologies. The developed analytical methodology was used in the analysis of bluefish (Pomatomas saltatrix) and weakfish (Cynoscion regalis) collected from the York River, lower James River and lower Chesapeake Bay in Virginia. Total PCB concentrations were calculated and homolog distributions were constructed from the acquired data. Increases in total PCB concentrations were found in the analyzed fish samples during the fall of 1985 collected from the lower James River and lower Chesapeake Bay. Comparisons between the homolog distribution patterns in the fish samples with the previously mentioned Aroclor distribution patterns suggests a different source of PCBs for different areas. Sediments, oysters (Crassostrea virginica) and brackish water clams (Rangia cuneata) collected from the tidal James River in 1986 were also analyzed. Total PCB concentrations and homolog distributions were calculated for all samples. Sediment total PCB concentrations were relatively constant over the sampling range except in the region of the turbidity maximum which were significantly higher. Total PCB concentrations in the Rangia from the region of the turbidity maximum were the highest of all the biota samples. Rangia homolog distribution patterns from this area were distinctly different from the sediment distribution patterns or the other Rangia distribution patterns in this segment of the river. Alteration of the endemic distribution pattern may be due to physical-chemical processes occurring within the turbidity maximum.