Temporal and spatial patterns of shell growth in hard clams (Mercenaria spp.) from the Indian River Lagoon, Florida, USA

Citation
Temporal and spatial patterns of shell growth in hard clams (Mercenaria spp.) from the Indian River Lagoon, Florida, USA

Material Information

Title:
Temporal and spatial patterns of shell growth in hard clams (Mercenaria spp.) from the Indian River Lagoon, Florida, USA
Creator:
Arnold, William Smith
Place of Publication:
Tampa, Florida
Publisher:
University of South Florida
Publication Date:
Language:
English
Physical Description:
x, 140 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Mercenaria -- Growth -- Florida -- Indian River ( lcsh )
Dissertations, Academic -- Marine Science -- Doctoral -- USF ( FTS )

Notes

General Note:
Includes vita. Thesis (Ph. D.)--University of South Florida, 1996. Includes bibliographical references (leaves 129-140).

Record Information

Source Institution:
University of South Florida
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
023490831 ( ALEPH )
36979695 ( OCLC )
F51-00195 ( USFLDC DOI )
f51.195 ( USFLDC Handle )

Postcard Information

Format:
Book

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

TEMPORAL AND SPATIAL PATTERNS OF SHELL GROWTH IN HARD CLAMS (Mercenaria SPP.) FROM THE INDIAN RIVER LAGOON, FLORIDA, USA by WILLIAM SMITH ARNOLD A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Marine Science University of South Florida December 1996 Major Professor: Norman J. Blake, Ph.D.

PAGE 2

Graduate School University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Ph.D. Dissertation This is to certify that the Ph.D. Dissertation of William Smith Arnold with a major in Marine Science has been approved by the Examining Committee on November 8, 1996 as satisfactory for the dissertation requirement for the Doctor of Philosophy degree Examining Committee: Major Professor: Norman J. Blake, Ph.D. Member: Thomas G. Bailey, Ph.D. Member: John J. Walsh, Ph.D. Member: Joseph J. Torres, Ph. D. Member: Gabriel A. Vargo, Ph.D.

PAGE 3

Copyright by William Smith Arnold 1996 All rights reserved

PAGE 4

DEDICATION This manuscript, and all of the time and effort that I devoted to its completion, is dedicated to the memory of my beloved mother, Mrs. Norma D. Arnold, whom I miss dearly, and to my father, Dr. Carl J. Arnold, Jr., whose love and support provided me with the luxury to indulge in this extremely protracted project.

PAGE 5

ACKNOWLEDGEMENTS I would like to thank Dr. Norman J. Blake for his patience, guidance, and perserverance throughout my tenure in the graduate program at the University of South Florida. I also appreciate the support of my committee members, Dr. Thomas Bailey, Dr. Jose Torres, Dr. John Walsh, and Dr. Gabe Vargo, each of whom made unique and substantial contributions to the success of this endeavor. I would especially like to acknowledge the faithful and constant support of my friend and colleague, Dr. Theresa Bert of the Florida Marine Research Institute. Dr. Dan Marelli, Ms. Paige Gill, and Mr. Don Hesselman provided the field and laboratory assistance that made this research possible. I am extremely grateful for the logistic and financial support provided by the Florida Department of Environmental Protection Marine Research Institute, without which this project would not have been possible. Finally, I must acknowledge the love, faith, and support provided by my wife Vicki and my children Lyndsey, Michael, and Jack. If not for them, the entire endeavor would have been pointless.

PAGE 6

TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT CHAPTER 1. INTRODUCTION AND LITERATURE REVIEW CHAPTER 2. CONTEMPORANEOUS DEPOSITION OF ANNUAL GROWTH BANDS IN Mercenaria mercenaria, M. campechiensis, AND THEIR NATURAL HYBRID FORMS Introduction Materials and Methods Results Discussion CHAPTER 3. HABITAT-SPECIFIC GROWTH OF THE NORTHERN HARD CLAM, Mercenaria mercenaria (L.), FROM THE iii v viii 1 11 11 18 28 40 INDIAN RIVER LAGOON, FLORIDA 48 Introduction 48 Materials and Methods 52 Classification of Individuals 52 Field Sampling Locations and Techniques 53 Determination of Growth Periodicity 58 Microgeographic Variation in Growth 61 Results 67 Classification of Individuals 67 Growth Periodicity 67 Microgeographic Variation in Growth 69 Discussion 78 Growth Periodicity 78 Microgeographic Variation in Growth 82 i

PAGE 7

CHAPTER 4. GENOTYPE-SPECIFIC GROWTH OF HARD CLAMS (GENUS Mercenaria) IN A HYBRID ZONE: VARIATION AMONG HABITATS 90 Introduction Materials and Methods Sample Collection Genetic Analysis Habitat-specific Growth Analysis Results Discussion CHAPTER 5. SUMMARY LIST OF REFERENCES VITA 90 93 93 97 99 102 121 127 129 EndPage

PAGE 8

LIST OF TABLES Table 1. Numbers of hard clams (Mercenaria spp.) assigned to each genotype class by discriminant function analysis performed using genetic index scores derived from four sets of diagnostic loci 29 Table 2. Significance tests for growth-band formation patterns among seasons within each of three hard clam (Mercenaria spp.) genotype classes from the Indian River lagoon, Florida 33 Table 3. Significance tests for growth-band formation patterns among genotype classes, within each of four seasons, from the Indian River lagoon, Florida 38 Table 4. Replicated Goodness-of-Fit G-test for heterogeneity of growth increments in Mercenaria mercenaria from the Indian River lagoon, Florida 68 Table 5. Games and Howell analysis of differences in w values among shellfish harvesting areas within the Indian River lagoon, Florida 73 Table 6. Cluster analysis based upon habitat components in the Indian River lagoon, Florida 74 Table 7. Games and Howell analysis of differences in w values among clusters within the Indian River lagoon, Florida 75 Table 8. Games and Howell analysis of differences in w values among depth zones within the Indian River lagoon, Florida 76 iii

PAGE 9

Table 9. Games and Howell analysis of differences in w values among submerged aquatic vegetation (SAV} categories within the Indian River lagoon, Florida Table 10. Comparison of range of w values, compared among the Indian River lagoon, Florida, various other sites throughout Florida and south Georgia, Narragansett Bay, Rhode Island, and for Mya arenaria from various sites along the northwest 77 Atlantic coast of the USA 83 Table 11. Pattern of w values along a latitudinal transect from shellfish harvesting areas c to IR1 in the Indian River lagoon, Florida 86 Table 12. Results of nonparametric multiple range test comparing mean annual salinity among eight sampling stations in the Indian River lagoon, Florida 103 Table 13. Frequency (%) of occurrence of three genotype classes (Mercenaria mercenaria, M. campechiensis, hybrids) in three subdivisions (C, D/E, F) of the Indian River lagoon, Florida 107 Table 14. Characteristics of four microhabitat types defined using cluster analysis performed on quantitative estimates of water depth, % sand and % organics in the substrate, and presence (1} or absence (2} of submerged aquatic vegetation (SAV) 108 Table 15. Frequency (%) of occurrence of microhabitat types within subdivisions of the Indian River lagoon, Florida 109 iv

PAGE 10

LIST OF FIGURES Figure 1. Boundaries of shellfish harvesting areas c, 0 and E, and locations of monthly sampling stations for hard clam (genus Mercenaria) growth-band formation study in the Indian River lagoon, Florida 15 Figure 2. Growth band phases for each of the translucent (T) and opaque (0) growth stages in thick-sectioned shells of hard clams (Mercenaria spp.) from the Indian River lagoon, Florida 23 Figure 3. Mean bi-weekly bottom-water temperature ( C one standard deviation collected at each of six sampling stations in the Indian River lagoon, Florida, August 1987 through August, 1988 26 Figure 4. size distribution (percent frequency) of three hard clam (Mercenaria spp.) genotype classes collected from the Indian River lagoon, Florida 31 Figure 5. Seasonal growth-band formation patterns of hard clams (Mercenaria spp.) in the Indian River lagoon, Florida 35 Figure 6. Boundaries of shellfish harvesting areas (A-F, IR1) and locations of monthly sampling stations for growth-increment formation study in the Indian River lagoon, Florida 55 v

PAGE 11

Figure 7. A) Shell of Mercenaria mercenaria illustrating position of radial cut along maximum axis of dorso-ventral growth; B) Cross-section view of M. mercenaria shell, illustrating the pattern of opaque and translucent growth increments and their relationship to the annual growth increment 60 Figure 8. Monthly pattern of T-O growth-increment formation in Mercenaria mercenaria, pooled over shellfish harvesting areas C, D, E, and F 66 Figure 9. A} Monthly pattern of temperature variation among shellfish harvesting areas within the Indian River lagoon, Florida, during September, 1987, through August, 1988; B) Monthly pattern of salinity variation among shellfish harvesting areas within the Indian River lagoon, Florida, during September, 1987, through August, 1988 Figure 10. Boundaries of shellfish harvesting areas c, D, E, and F and locations of salinity monitoring stations for the study of habitat-specific growth rates of hard clams (Mercenaria spp.) in the 71 Indian River lagoon, Florida 95 Figure 11. Variation in salinity (ojoo) in each of four shellfish harvesting areas (C, D, E, F) of the Indian River lagoon, Florida 105 Figure 12. Mean annual growth rate, as defined by the w parameter, for three genotype classes of hard clams from the Indian River lagoon, Florida 113 Figure 13. Variation among cluster-analysis-defined microhabitats in mean annual growth rate, as defined by the w parameter, of three genotype classes of hard clams from three subdivisions of the Indian River lagoon, Florida 115 vi

PAGE 12

Figure 14. Variation among depth categories in mean annual growth rate, as defined by the w parameter, of three genotype classes of hard clams from three subdivisions of the Indian River lagoon, Florida 117 Figure 15. Variation among submerged aquatic vegetation (SAV) categories in mean annual growth rate, as defined by the w parameter, of three genotype classes of hard clams from three subdivisions of the Indian River lagoon, Florida 119 vii

PAGE 13

TEMPORAL AND SPATIAL PATTERNS OF SHELL GROWTH IN HARD CLAMS (Mercenaria SPP.) FROM THE INDIAN RIVER LAGOON, FLORIDA, USA by WILLIAM SMITH ARNOLD An Abstract Of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Marine Science University of South Florida December 1996 Major Professor: Norman J. Blake, Ph.D. viii

PAGE 14

Hard clams of the genus Mercenaria were sampled from the Indian River lagoon, Florida, during July and August, 1986, and periodically from September, 1987 through August, 1988. During the 1986 sampling episode, I collected clams from a variety of habitats throughout the lagoon to determine variations in shell growth patterns of genetically defined specimens. During biweekly sampling in 1987 and 1988, I collected clams from previously defined habitats to determine the annual pattern of internal growth band deposition and to relate that to hard clam genotype class and to environmental patterns. Results from the periodic study clearly indicate that the alternating pattern of translucent (dark band in thicksectioned shells) and opaque (white band in thick-sectioned shells) paired internal growth bands are annual in nature. No latitudinal difference in the timing of annual growthband formation was discerned, nor were there any major differences in the timing of annual band formation among the three genotype classes (Mercenaria mercenaria, M. campechiensis, hybrid forms) that I defined. Overall similarities among genotype classes in the annual pattern of growth-band formation suggest that differences in those patterns among locations throughout the geographic range of Mercenaria along the Atlantic and Gulf of Mexico coasts of the United States are primarily mediated by environmental rather than genotypic factors. ix

PAGE 15

When comparing shell growth of hard clam genotype classes among habitats within the Indian River lagoon, I discovered substantial differences among habitats both within Mercenaria mercenaria and among M. mercenaria, M. campechiensis, and their hybrid forms. Shell growth of M. mercenaria was significantly faster in vegetated than unvegetated habitats and in shallow-water than in deep-water habitats. Among genotype classes, M. campechiensis maintained a growth advantage in deep-water habitats but was rare in other habitats, suggesting that other physiological constraints may prevent that species from numerically dominating the Indian River lagoon hard clam population. Both M. mercenaria and hybrid forms maintained a growth advantage in certain habitats; M. mercenaria grew fastest in deep-water habitats and hybrid forms grew fastest in shallow-water habitats. My results provide one explanation for the maintenance of a hybrid zone in the Indian River despite apparently substantial gene flow among the three hard clam genotype classes that inhabit that lagoon. Abstract Approved: Major Professor: Norman J. Blake, Ph.D. Professor, Department of Marine Science Date Approved: //-i' f -Y X

PAGE 16

CHAPTER 1. INTRODUCTION AND LITERATURE REVIEW The analysis of internal shell growth-increment patterns is a reliable technique for determining the age and growth of bivalve molluscs (Jones, 1983). Shell growthincrements (i.e., growth layers, growth bands, growth lines) are the bands or rings expressed on the external surface or within the skeletal components of bivalves and a variety of other organisms (e.g., corals, fish, polychaetes [Rhoads and Lutz, 1980)). Sclerochronology, the study of periodic growth structures contained within the skeletal components of aquatic organisms such as bivalve molluscs (Hudson et al., 1976; Jones, 1983; Quitmyer et al., in press), dates at least to the time of Leonardo da Vinci (Jones, 1983). However, the functional relationship between skeletal growth patterns and time, as mediated by the environment, was not established until the mid-1950's when Chave {1954) and Lowenstam {1954) related internal shell growth-increment patterns to environmental temperature. In bivalves, Barker {1964) categorized and described five internal shell-growth layers, ranging from a subdaily 5th-order layer influenced by tidal patterns to an annual 1st-order layer influenced by seasonal temperature and salinity variation. The underlying 1

PAGE 17

periodicity of those shell growth-layers has been subsequently exploited in a variety of paleochronological (e.g., Rosenberg and Runcorn, 1975), archaeological (e.g., Quitmyer et al., 1985), and ecological (e.g., Greene, 1975) investigations. Although growth-increment patterns of bivalve molluscs are not always periodic (Gould, 1979; Jones, 1981) and may be driven by stochastic as well as cyclical events (Kennish, 1980), it is apparent that certain repeating structural features contained within the shells of bivalve molluscs exhibit a predictable temporal periodicity of formation (Barker, 1964; Kennish, 1980). Of the five growth layers described by Barker (1964) in the shells of bivalve molluscs, clearly the most reliable and interpretable is the 1st-order (annual) layer. This 1st-order growth layer (the "annual band") visible along the exposed surface of radially-sectioned shells, is composed of two growth bands. In thick-sectioned shells (i.e., viewed with the aid of reflected rather than transmitted light) a complete annual shell growth-increment is composed of a translucent (dark) band characteristic of periods of relatively slow shell growth and an opaque (light) band characteristic of periods of relatively rapid shell growth. However, these macro-scale growth bands owe their existence to fundamental metabolic processes occurring on a sub-daily to daily time scale. Each of the translucent and opaque components of the annual 2

PAGE 18

band are constructed from a repeating series of 5th-order layers that reflect environmental conditions at the time of formation. Lutz and Rhoads (1977) have provided a coherent theory to explain the relationship between the environment, metabolic processes within the organism, and the pattern of formation of growth layers. When the shell is open and the animal is actively respiring, calcium-carbonate rich shell material is deposited within the shell organic matrix and a "light" band (in reflected light) is formed. When the animal closes its shell, aerobic respiration ceases, anaerobic metabolism begins, and the acidic end products of anaerobic respiration (e.g., succinic acid, alanine) begin to build up. Those end products are neutralized by calcium carbonate dissolved from the shell; the remaining organic matrix is then left to comprise a "dark" band. The formation of the translucent and opaque components of the annual band simply reflects the relative ratio of aerobic to anaerobic respiration over the daily cycle. During growthfavorable periods (e.g., during summer in cold-temperate climates), aerobic respiration predominates and the opaque band forms, whereas during unfavorable growth conditions anaerobic respiration predominates and the translucent band forms. Shell growth rate is an important component of fitness in bivalve molluscs for a variety of reasons. The mechanical force required to crush a bivalve shell is 3

PAGE 19

positively correlated with shell size (Blundon and Kennedy, 1982}, so the shell becomes a more effective predatordefense mechanism as shell size increases (Whetstone and Eversole, 1981; Arnold, 1984; Juanes, 1992}. Other components of fitness, including sedimentation-induced mortality, maturation and fecundity, and competitive interactions, also may be directly or indirectly sizedependent (Bricelj and Malouf, 1980; Peterson, 1983; Rawson and Hilbish, 1991). As a result, the fitness of individuals having rapid growth rates may be enhanced compared to that of slower-growing individuals. Therefore, understanding the annual cycle of shell growth of bivalve molluscs has important ecological implications. Documenting the rate and patterns of annual growth in bivalves and correlating those patterns with internal and external processes can provide important insights into physiological and ecological factors that affect growth (Lutz and Rhodes, 1977; 1980; Kennish, 1980}. For example, shell growth consumes energy otherwise available for tissue growth and reproduction (Peterson and Fegley, 1986} but also indirectly provides a pathway to refuge from numerous predators (Juanes, 1992}. Knowledge of intra-annual shell growth patterns of bivalves also may have important applications in the determination of paleoseasonality (Coutts, 1970), paleobiology (Sato, 1995} systematics, (Cunliffe, 1974) and seasonal occupation patterns of prehistoric man (Quitmyer et al., in press). 4

PAGE 20

Furthermore, there are important economic implications to the rate and pattern of bivalve shell growth. Bivalve molluscs comprise an extremely important resource to coastal communities in the USA, averaging over $370 million in annual dockside landings value during 1990 through 1994 (Anonymous, 1991-95), and an understanding of shell growth patterns may facilitate economic utilization of bivalve species by influencing the timing of harvest (Holiman, 1995) Hard clams of the genus Mercenaria (Bivalvia: Veneridae) are particularly amenable to internal shell growth-increment analysis because they are relatively abundant, occur in easily accessible nearshore habitats, deposit distinct annual growth bands, and support substantial commercial and recreational fisheries. Hard clams are ubiquitous in shallow marine habitats along the Atlantic and Gulf of Mexico coasts of the United States (Stanley and DeWitt, 1983). Two species of hard clams have been described from within that range: Mercenaria mercenaria (Linnaeus) occurs from the Gulf of st. Lawrence to the Gulf of Mexico, and M. campechiensis (Gmelin) occurs from New Jersey to Texas (Abbott, 1974; Brown, 1989). A subspecies, M. mercenaria texana (Dall), has been described from the Texas Gulf coast (Abbott, 1974), although the phylogenetic classification of that subspecies has been reconsidered (Brown, 1989). The distributions of the two species along 5

PAGE 21

the Atlantic Ocean coast generally do not overlap (M. mercenaria occurs inshore and M. campechiensis occurs offshore}, but in the Indian River lagoon on the eastcentral coast of Florida (Figure 1} the two species cooccur and hybridize extensively (Dillon and Manzi, 1989; Bert et al., 1993; Bert and Arnold, 1995). Although information on the basic ecology of hard clams is available for a variety of locations (e.g., McHugh et al., 1982; McHugh and Sumner, 1988}, useful data on the ecology of hard clam populations from the east coast of Florida have become available only recently. Those studies have shown that Indian River hard clams grow more rapidly (Jones et al., 1990), spawn more frequently (Hesselman et al., 1989}, and die at a younger age (Jones et al. 1990} than do their northern counterparts. Geographic variation in the shell growth rate of hard clams was summarized by Ansell (1968), but data upon which this summary was based were inadequate in several respects. For example, several important localities, including the Atlantic coast of Florida, were not represented. Furthermore, most of the data were obtained from non-native clams or from clams cultured under artificial conditions. Shell growth rates of cultured clams vary with the source of the stock (Walker and Humphrey, 1984), stocking density (Walker, 1985), and frequency of sampling (Walker, 1985). Nevertheless, controlled field studies of hard clam growth 6

PAGE 22

in the southeastern United States indicate that annual growth of Mercenaria campechiensis is faster than that of M. mercenaria and that the growth of their reciprocal hybrids is intermediate between that of the two parental species (aaven and Andrews, 1956; Chestnut et al., 1956; Menzel, 1961; 1962). Recent research on growth of natural hard clam populations from the Atlantic and Gulf of Mexico coasts of Florida suggest that comparisons of the relative shell growth of the two species are more complex than first perceived. In Florida, growth of M. mercenaria from Atlantic coastal waters was similar among locations but growth of M. campechiensis from Gulf of Mexico coastal waters differed substantially among locations, with some M. campechiensis growing faster and others more slowly than Atlantic populations of M. mercenaria. It is clear that a review of hard clam shell growth rates, as determined by analyzing the internal shell growth-increments of natural populations from throughout the geographic range, is needed to clarify local and latitudinal variations in shell growth rates of this animal. The timing of annual band formation also appears to differ, both within and among geographic locales and within 7

PAGE 23

1992). However, geographic variation in the timing of annual band formation within each of the two parental species may be substantial. For M. mercenaria, the translucent band forms during winter in populations north of New Jersey (Panella and MacClintock, 1968), during summer in populations from North Carolina southward (Peterson et al., 1983; Arnold et al., 1991), and during both summer and winter in populations from New Jersey (Grizzle and Lutz, 1988). In M. campechiensis from the Gulf of Mexico, translucent band formation predominates during July through March in populations from the waters off southwest Florida, but during July through February in populations from northwest Florida waters (Jones et al., 1990). Finally, in the southeastern United States, there are apparent differences in the temporal pattern of annual band formation between the two parental species. The period during which translucent bands are formed appears to be longer in M. campechiensis populations from Florida Gulf of Mexico waters than it is in populations of M. mercenaria from comparable latitudes along the Atlantic Ocean coast of Florida (Quitmyer et al., 1985; Arnold et al., 1991; Quitmyer and Jones, 1992). It is unclear whether differences in the pattern and timing of shell growth within and among Mercenaria genotype classes are due to ambient environmental conditions (including those that may be geographically correlated) or 8

PAGE 24

are genetically mediated. Growth in both species appears to be influenced by factors related to water temperature (Ansell, 1968; Clark, 1979; Arnold et al., 1991), but seasonal allocation of energy between somatic and reproductive growth also affects hard clam shell growth patterns (Peterson and Fegley, 1986) and may influence the timing of growth. Previous studies of hard clam shell growth have been conducted in geographically separated areas or in separate years. Furthermore, although shell growth in M. mercenaria x M. campechiensis F1 hybrid forms has been documented (Menzel, 1962; Dalton, 1977), the rate and timing of shell growth in naturally occurring hybrid forms of the two species has not been defined. To differentiate between environmental versus genotypic influeces on the patterns of shell growth in hard clams, it is necessary to conduct contemporaneous studies of shell growth among the two species and their hybrids within similar habitats. The Indian River lagoon provides an ideal location for such comparisons. Both species and their hybrids occur in the lagoon (Dillon and Manzi, 1989; Bert and Arnold, 1995). Furthermore, the benthic structure of the lagoon is composed of a variety of microhabitats (Arnold et al., 1991a), and various combinations of the two species and their hybrids occur in each of those microhabitats. This offers a unique opportunity to study contemporaneous growth among genotype classes within a diverse array of 9

PAGE 25

habitat types. I have conducted studies of spatial and temporal growth among three defined hard clam genotype classes (Mercenaria mercenaria, M. campechiensis, and their hybrid forms) from the Indian River lagoon. I first define temporal patterns of annual growth band deposition among genotype classes sampled from similar habitats during September, 1987, through August, 1988. After determining that 1st-order bands within the shells of Indian River hard clams are truly annual in nature, I then compare habitatspecific growth patterns among genotype classes collected from the lagoon. I first show that, within the M. mercenaria genotype class, considerable variation in shell growth does occur among habitats within the lagoon, but that variation is substantially less than that observed between widely-separated locations or between bivalve genera. I then compare habitat-specific growth among genotype classes within the lagoon, and conclude that growth variation among habitats is significant, that each genotype class has a growth advantage in certain habitat types, and that those species-specific growth advantages may provide a mechanism for stabilization of the hard clam hybrid zone in the Indian River lagoon. 10

PAGE 26

CHAPTER 2. CONTEMPORANEOUS DEPOSITION OF ANNUAL GROWTH BANDS IN Mercenaria mercenaria, M. campechiensis, AND THEIR NATURAL HYBRID FORMS Introduction Understanding the annual cycle of shell growth of bivalve molluscs is important from both an economic and an ecological perspective. Economically, bivalve molluscs comprise an extremely important resource to coastal communities in the USA, averaging over $370 million in annual dockside landings value during 1990 through 1994 (Anonymous, 1991-95). Knowledge of growth periodicity may facilitate economic utilization of bivalve species by influencing the timing of harvest (Holiman, 1995). Ecologically, documenting annual growth cycles in bivalves and correlating those cycles with internal and external processes can provide important insights into physiological and ecological factors that affec t growth (Lutz and Rhodes, 1977 ; 1980; Kennish, 1980). For example, shell growth consumes energy otherwise available for tissue growth and reproduction (Peterson and Fegley, 1986) but also indirectly provides a pathway to refuge from numerous predators 11

PAGE 27

(Juanes, 1992). Knowledge of intra-annual shell growth patterns of bivalves also may have important applications in the determination of paleoseasonality (Coutts, 1970), paleobiology (Sato, 1995) systematics, (Cunliffe, 1974) and seasonal occupation patterns of prehistoric man (Quitmyer et al., in press). Hard clams of the genus Mercenaria (Bivalvia: Veneridae) deposit repeating growth bands within their shells. A variety of bands, formed in response to numerous periodic and aperiodic internal and external cues, with periodicities ranging from hours to years, may be deposited within the shell of an individual hard clam (Barker, 1964; Kennish, 1980). Although many of those banding patterns may be difficult to interpret ( Kennish, 1980), properly verified (e.g., Peterson et al., 1983; Grizzle and Lutz, 1988; Arnold et al. 1991a) annual growth bands (sensu Jones et al., 1990) are relatively easy to identify and interpret in hard clams. Annual bands (composed of one translucent band and one opaque band in thick-sectioned shells) have been successfully analyzed in a variety of investigations (e.g., Green, 1975; Rosenberg and Runcorn, 1975; Quitmyer et al., 1985; Jones et al., 1989; Arnold et al., 1991a; 1996; Bert and Arnold, 1995). Hard clams are ubiquitous in shallow marine habitats along the Atlantic and Gulf of Mexico coasts of the United states (Stanley and DeWitt, 1983). Two species of hard 12

PAGE 28

clams have been described from within that range: Mercenaria mercenaria (Linnaeus) occurs from the Gulf of st. Lawrence to the Gulf of Mexico, and M. campechiensis {Gmelin) occurs from New Jersey to Texas (Abbott, 1974; Brown, 1989}. A subspecies, M. mercenaria texana (Dall), has been described from the Texas Gulf coast {Abbott, 1974), although the phylogenetic classification of that subspecies has been reconsidered (Brown, 1989}. The distributions of the two species along the Atlantic Ocean coast generally do not overlap (M. mercenaria occurs inshore and M. campechiensis occurs offshore), but in the Indian River lagoon on the east-central coast of Florida (Figure 1), the species co-occur and hybridize extensively (Dillon and Manzi, 1989; Bert et al., 1993; Bert and Arnold, 1995). In inshore areas, the timing of annual band formation is similar among habitats within both Mercenaria mercenaria (Peterson et al., 1985} and M. campechiensis (Quitmyer and Jones, 1992}. However, geographic variation in the timing of annual band formation within each of the two parental species may be substantial. For M. mercenaria, a translucent band (indicative of relatively slow shell growth) forms during winter in populations north of New Jersey (Panella and MacClintock, 1968), during summer in populations from North Carolina southward {Peterson et al., 1983; Arnold et al., 1991a}, and during both summer and 13

PAGE 29

Figure 1. Boundaries of shellfish harvesting areas c, D, and E (---) and locations of monthly sampling stations (e) for hard clam (genus Mercenaria) growth-band formation study in the Indian River lagoon, Florida.

PAGE 30

0 0 0 ..-co 0 ('t) 0 co N (.) 15 0 0 0 co N 0 CD \ ; '. 0 1.0 CD

PAGE 31

winter in populations from New Jersey (Grizzle and Lutz, 1988). In M. campechiensis from the Gulf of Mexico, translucent band formation predominates during July through March in populations from the waters off southwest Florida, but during July through February in populations from northwest Florida waters (Jones et al., 1990). Finally, in the southeastern United States, there are apparent differences in the temporal pattern of annual band formation between the two parental species. The period during which translucent bands are formed appears to be longer in M. campechiensis populations from Florida Gulf of Mexico waters than it is in populations of M mercenaria from comparable latitudes along the Atlantic Ocean coast of Florida (Quitmyer et al., 1985; Arnold et al., 1991a; Quitmyer and Jones, 1992). It is unclear whether differences in the timing of rapid and slow growth between Mercenaria mercenaria and M. campechiensis are due to ambient environmental conditions (including those that may be geographically correlated) or are genetically mediated. The timing of growth in both species appears to be influenced by factors related to water temperature (Ansell, 1968; Clark, 1979; Arnold et al., 1991a), but seasonal allocation of energy between somatic and reproductive growth also affects hard clam shell growth patterns (Peterson and Fegley, 1986) and may influence the timing of growth. Previous studies of seasonal shell growth 16

PAGE 32

in the two species have been conducted in different years or in geographically separated areas. Furthermore, although some work on the seasonal pattern of shell growth in M. mercenaria x M. campechiensis F1 hybrid forms has been done (Menzel, 1962; Dalton, 1977), the seasonal timing of shell growth in natural hybrid forms of the two species has not been defined. To compare the seasonal patterns of translucent and opaque growth-band formation within and among the two hard clam species and their hybrid forms, without the potential complications of geographic variation, I examined the seasonal growth patterns of Mercenaria mercenaria, M. campechiensis, and their naturally-occurring hybrid forms collected from the Indian River lagoon. Individually for each of those three genotype classes, I first tested the hypothesis that each of the translucent and opaque components of the annual band are equally represented within each season. Acceptance of the null hypothesis, that no difference exists in the proportions of the two components of the annual band among seasons within each genotype class, would suggest that growth-band formation is not seasonally dependent in Florida waters. I next tested the hypothesis that there is no difference in the proportional representation of each growth band among the three genotype classes within each season. Acceptance of that null hypothesis would support the contention that there is no 17

PAGE 33

inherent difference in the timing of annual band deposition among the three hard clam genotype classes that I tested. My results support a seasonal pattern of growth-band formation within each genotype class, and further indicate that differences among hard clam genotype classes in the seasonal timing of annual band deposition within similar habitats are minor. However, subtle differences in the timing of annual-band formation among the three genotype classes suggest that some genotypic divergence has occurred in the timing of annual band formation between M. mercenaria and M. campechiensis. Materials and Methods The Indian River lagoon extends approximately 240 km along the east central coast of Florida and is separated from the Atlantic Ocean by a narrow barrier island system (Figure 1). For management purposes, the lagoon has been latitudinally subdivided into a series of shellfish harvesting areas (e.g., Poole and Arnold, 1985); the present study was conducted in shellfish harvesting areas c, D, and E of the lagoon (Figure 1). Those areas have a tidal range of less than 5 em (Smith, 1993), similar water temperatures (Arnold et al., 1991a), and significant differences in salinity among areas (Arnold et al., 1991a; Arnold et al., 1996). The lagoon is shallow, with an average depth of 1.5 18

PAGE 34

m and a maximum depth of approximately 4 m (White, 1986), but is generally well mixed by wind-driven circulation (Costa et al., 1987). To study the seasonal pattern of growth-band formation among Mercenaria mercenaria, M. campechiensis, and their hybrid forms, I collected ten hard clams each month from September 1987 through August 1988 from each of six sampling stations (two stations per shellfish harvesting area) located along the central axis of the study area (Figure 1) Samples were collected from similar, unvegetated habitats in approximately 2 m water depth. The clams were returned alive to the laboratory, where they were immediately dissected. Mantle, gill, and adductor muscle samples were excised and stored at -80 C for allozyme electrophoresis. The shells were cleaned, numbered, and stored for subsequent analysis of growth bands. Shell height (SH = maximum distance from umbo to ventral margin) of each clam was measured to the nearest 0.1 rom using Vernier calipers. I used a YSI Model 33 Salinity-Conductivity-Temperature meter to collect water-temperature data from approximately 0.5 m above the sediment-water interface at each of the six sampling stations. Temperature data, collected biweekly from August 31, 1987, through August 30, 1988, were pooled among the sampling stations within each sampling date because standard deviation of temperature among stations was generally < 1 0 C. 19

PAGE 35

Each clam was assigned to a genotype class following previously established procedures (Bert et al., 1993; Bert and Arnold, 1995). Clams collected from the Indian River, and from 13 other sites throughout the reported ranges of the two species in the United States, were assayed for four diagnostic protein loci (phenylalanine-specific dipeptidase-2 [DPEP-2; E.C. 4.3.11.*], glucose phosphate isomerase [GPIP; E.C. 5.3.1.9], aspartate amino transferase [AAT; E.C. 2.6.1.1], and phosphoglucomutase [PGMP-2; E.C. 5.3.2.2]) using horizontal starch gel electrophoresis. All electrophoretic assays were conducted by Bioche.mical Systematics Laboratory personnel at the Florida Department of Environmental Protection Marine Research Laboratory. Using principal components analysis (PCA), I selected 9 of the 20 U.S. populations to serve as pure-species reference standards (Mercenaria mercenaria: Connecticut, North Carolina, Georgia, and three locations in northeast Florida; M. campechiensis: Texas and two locations in northwest Florida). Individuals who were obvious outliers from these populations were excluded from the pure-species standards. For each allele at each locus, I calculated an allele diagnostic value (ADV) as the difference between the mean frequency of the allele in the pure-species M. mercenaria populations and the mean frequency of the allele in the pure-species M. campechiensis populations. For each individual, a genetic index score for each locus was then 20

PAGE 36

calculated as the sum of the two ADVs specific to the two alleles that the individual possessed at that locus. Using the pure-species individuals as knowns and the Indian River individuals as unknowns, I applied discriminant function analysis (DFA) to the genetic index scores for the four loci to calculate the probability of an individual belonging to each of the three genotype classes. Each Indian River individual was then assigned to the genotype class to which it had the highest probability of belonging. Although DFA is a parametrically based statistical technique, it yielded more reasonable results than analogous nonparametric statistical methods (see Bert et al., 1993). Because all individuals did not resolve and therefore were not scored for all four loci, I repeated both the PCA and the DFA four times, using sequentially the four-locus combination, two three-locus combinations (DPEP-2, GPIP, AAT and DPEP-2, GPIP, PGMP-2), and the two-locus combination with the best discriminatory power (DPEP-2, GPIP) to maximize the number of individuals classified. All individuals possible were used in each analysis; I used the maximum number of loci possible to assign each individual to a genotype class. Most individuals were classified using all four loci (Table 1) I used the translucent/opaque staging technique (Jones et al., 1990) for analysis of annual growth bands on thicksectioned shells because it provides a reliable profile of 21

PAGE 37

Figure 2. Growth band phases for each of the translucent {T) and opaque {0) growth stages in thick-sectioned shells of hard clams (Mercenaria spp.) from the Indian River lagoon, Florida. Shells were sectioned from the umbo to the ventral margin along the axis of maximum growth.

PAGE 38

Growth Phase Tl Translucent increment forming on the marginal edges T2 Translucent increment one-half complete T3 Translucent increment complet e 01 Opaque increment forming on the marginal edges 02 Opaque increment one-half complete 03 Opaque increment complete Radial Cross-Section Hinge-Plate Margin Ventral Margin 23

PAGE 39

the seasonal timing of shell growth in hard clams (Claasen, 1990). The stage of terminal growth was identified as being within either the translucent (narrow dark) band or the opaque (wide white) band (Arnold et al., 1991a) on one radially sectioned valve from each individual hard clam collected. To increase resolution in the analysis of the timing of translucent and opaque growth band formation, I also determined the growth "phase" (sensu Quitmyer and Jones, 1992; Quitmyer et al., in press) within each stage based upon the relative thickness of the terminal growth band (Figure 2). If the band was just beginning to form, it was classified as being in either the T1 (translucent phase 1) or 01 (opaque phase 1) phase. A band that was approximately 1/2 as large as the previous band of the same stage was classified as being in the T2 or 02 phase. Finally, a band that was as large or larger than the previous band of the same stage was classified as being in the T3 or 03 phase. Monthly growth stage data for each species were pooled into seasonal groups. The winter season (December-February) was the three-month period during which water temperatures were lowest (Figure 3). The other three seasons were structured incrementally, in three-month groupings, around the winter season. Thus, spring was designated as MarchMay, summer as June-August, and fall as September-November. 24

PAGE 40

Figure 3. Mean bi-weekly bottom-water temperature ( C one standard deviation) collected at each of six sampling stations (see Figure 1) in the Indian River lagoon, Florida, August 1987 through August 1988. Temperature data were averaged among shellfish harvesting areas within each biweekly sampling episode.

PAGE 41

..-... 30 () 0 ()) 25 s..... ::J +-' C\l s..... ()) 20 c.. E ()) I15

PAGE 42

I conducted a variety of non-parametric statistical analyses in preparation for and in effecting tests of my hypotheses. I first applied the Kruskal-Wallis test to ranked shell height data to determine differences in mean shell height among genotype classes (Zar, 1984). I then used the RxC G-Test of Independence to test for differences among shellfish harvesting areas in the pattern of growthband formation within each of the Mercenaria mercenaria and hybrid genotype classes within each season (Sokal and Rohlf, 1995}. That test was inappropriate for the M. campechiensis genotype class because all but one member of that class was collected from within area c. Next, within each of the three genotype classes, I tested for differences in the frequenc y of occurrence of each of the two growth bands among seasons using a Replicated Goodness-of-Fit G-test (Sokal and Rohlf, 1995). Finally, within eac h of the four seasons, I again used the Replicated Goodness-of-Fit G-test to test for differences in the frequency of occurrence of each of the two growth bands among genotype classes. For both Goodness-of-Fit G-tests, the overall G value was decomposed into a pooled component (describing the proportional contribution of the translucent versus opaque growth bands pooled among genotype class or among season) and a heterogeneity component (describing the degree of heterogeneity in the proportional contribution of the two bands among genotype class or among season) Again for both 2 7

PAGE 43

Goodness-of-Fit G-tests, the Simultaneous Test Procedure for frequencies (STP, Q=0.05} was conducted to define heterogeneous groups of percentages (Sokal and Rohlf, 1995). When conducting multiple tests of the same general hypothesis within each of the four seasons or within each of the three genotype classes, I applied the sequential Bonferroni technique to compensate for multiple testing of the hypothesis (Rice, 1989}. Growth phase data generally were not amenable to statistical analysis because of low sample sizes in many season by genotype-class combinations. Results Of the 720 hard clams collected in the Indian River lagoon from September, 1987 through August, 1988, I successfully genotyped 252 Mercenaria mercenaria, 33 M. campechiensis, and 111 hybrid forms (Table 1}. The remaining 324 clams could not be resolved for one or both of the DPEP-2 or GPIP loci. The distribution of principal component 1 (PC1} scores from the Indian River samples spanned the range of PC1 scores recorded for pure-species populations from throughout the Atlantic and Gulf of Mexico coastal waters of the United States but was skewed toward M mercenaria. The genotype distribution of individuals used in this study is given in Bert et al. (1993). 28

PAGE 44

Table 1. Numbers (percentage in parentheses) of hard clams (Mercenaria spp.) assigned to each genotype class by discriminant function analysis performed using genetic index scores (defined in Materials and Methods) derived from four sets of diagnostic loci. Locus combination PGJP, DP2, PGIP DP2, PGIP, DP2 PGJP DP2 Genotype class AAT, PGMP-2 AAT PGMP-2 Mercenaria mercenaria 227 (90.1) 11 (4.4) 13 (5.2) 1 (0.4) Mercenaria campechiensis 27 (81.8) 1 (3.0) 5 (15.2) 0 (0.0) Hybrid forms 98 (88.3) 7 (6.3) 6 (5.4) 0 (0.0) Total 352 (88.9) 19 (4.8) 24 (6.1) 1 (0.2) 29

PAGE 45

Figure 4. Size distribution (percent frequency) of three hard clam (Mercenaria spp.) genotype classes collected from the Indian River lagoon, Florida. Size classes are in 5 rom increments of shell height (maximum distance from umbo to ventral margin); each x-axis label represents the maximum height within that size class (e.g., 40 rom includes all clams between 35.1 and 40.0 rom shell height).

PAGE 46

w ...... 35 >. 30 (.) c Q) ::::J cQ) L-LL. ....... 25 20 c Q) 15 (.) L-Q) a.. 10 5 M. mercenaria r,,,,,,,,,,, ,,,,, , ,,,,,,,,1 M campech1ens1s :::: I I Hybrid forms :: Q I , 35 40 45 50 55 60 65 70 75 80 85 90 95 100105 Size Class (mm)

PAGE 47

Mean shell height (Figure 4) differed slightly but significantly among the three Mercenaria genotype classes sampled during my study (x2=12.40, df=2). Mean shell height of M. mercenaria (61.0 mm, S.D.=10.3) was significantly smaller than that of either M. campechiensis (64.9 mm, S.D.=9.7) or hybrids (64.8 mm, S.D.=10.9) but there was no significant difference in shell height between M. campechiensis and hybrids. For both Mercenaria mercenaria and hybrid genotype classes, I detected no significant difference in the proportional occurrence of the translucent and opaque growth bands among shellfish harvesting areas within each season. Thus, there was no influence of location on the pattern of growth-band deposition within each genotype class by season combination. Within each genotype class, I therefore pooled samples among locations within each of the four seasons. For the spatially-pooled data, I next tested the frequencies of occurrence of the translucent and opaque growth bands for significant differences among seasons within each genotype class (Table 2). The overall proportion of individuals in the opaque vs. translucent growth stage (pooled G) differed significantly in Mercenaria mercenaria but not in M. campechiensis nor hybrid forms (Table 2). Within each genotype class, the heterogeneity G accounted for 91.2% to 99.2% of the total G value. Seasonal differences in the proportions of the two growth bands 32

PAGE 48

Table 2. Significance tests for growth-band formation patterns among seasons within each of three hard clam (Mercenaria spp.) genotype classes from the Indian River lagoon, Florida. Seasons are defined in Materials and methods. Significant differences in the proportion of translucent and opaque growth bands, overall (pooled), among seasons (heterogeneity), and within each season (fall, winter, spring, summer) were determined using the Goodnessof-Fit G-test (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). Significant differences among seasons were further defined using the Simultaneous Test Procedure (STP) with p < 0.05. For the STP, seasons for which the proportions of translucent and opaque growth bands were not significantly different within each genotype class are denoted by a common letter. Genotype class M mercenaria M campechiensis Hybrid Test df G STP df G STP df G STP Pooled 1 6.4 1 5 1 0 4 Heterogeneity -. -3 169 3 3 15 4 3 52.0 -5 1 Fall 1 41. 0 A 1 A8 1 21. 1 A -Winter 1 53 7 8 1 1 6 8 1 3.2 8 Spring 1 -58 6 8 1 0.5 8 1 17.4 -8 --Summer 1 22 3 A 1 9.7 8C 1 10 7 A Total 4 175 7 4 16.9 4 52 4 -33

PAGE 49

Figure 5. Seasonal growth-band formation patterns of hard clams (Mercenaria spp.) in the Indian River lagoon, Florida. Large bars depict percent frequency of occurrence of the opaque (0) or translucent (T) growth-band stages of (A) M. mercenaria, (B) M. campechiensis, and (C) hybrid forms. Numbers above each set of bars are the total number of hard clams in that genotype-class x season combination. Within each large bar, small bars represent, from left to right, the percent frequency of each of the three growth phases (01, 02, 03 or Tl, T2, T3) within each of the opaque and translucent growth stages. Within each growth phase, vertical bars represent the first month, hatched bars the second month, and horizontal bars the third month of collection within each season.

PAGE 50

100 75 50 25 A) 1 54 62 6 69 7 8 39 0 c B) 1 7 7 3 5 3 o 7 (J) 100 ::J o75 (J) L... LL 50 c 25 Q) 0 0 ___._____---=..___.. L... (J) a_ 100 75 50 25 C) 1 20 21 11 23 3 7 25 0 --'-----=L...-...1. 0 T 0 T 0 T 0 T Fall Winter Spring Summer Season 35

PAGE 51

accounted for most of the heterogeneity in the seasonality of growth. For M. mercenaria, growth was clearly seasonally partitioned; significantly different fractions of individuals were in the translucent or opaque growth stage each season (Table 2), and most individuals were experiencing rapid growth in winter and spring and slow growth in summer and fall (Figure 5A). The seasonal patterns of growth-band formation in M. campechiensis and hybrids were essentially the same as those in M. mercenaria, but growth stages were not as distinctly seasonally partitioned in those two genotype classes. In M. campechiensis (Figure 5B), significantly higher percentages of individuals were in the slow-growth stage during summer and fall, but the proportions of individuals in the slow versus rapid growth stages in winter and spring were not significantly different (Table 2). In hybrid forms (Figure 5C), as with the other genotype classes, significantly higher percentages of individuals were in the slow-growth stage during summer and fall (Table 2). During winter and spring, growth-band formation demonstrate d characteristics reminiscent of the deposition patterns of each of the parental species. As in M. campe chiensis, no significant difference in the proportion of individuals in the two growth stages occurred in winter; but as in M. mercenaria, nearly all individuals were in the opaque growth stage in spring. 36

PAGE 52

I next tested the frequencies of occurrence of the translucent and opaque growth bands for significant differences among genotype classes within each season (Table 3). Within each season the contribution of the pooled G predominated, comprising 82.4% (winter) to 98.6% (fall) of the total G value. This indicates a predominance of animals in one of the growth stages within each season, although the predominant growth stage varied among seasons. In each of the fall, spring, and summer seasons the relative frequencies of the two growth stages did not differ significantly among genotype classes within each season (heterogeneity G); only 1.4% to 7.3% of the total G value was attributable to differences among genotype classes in the frequencies of the two growth stages in those seasons (Table 3). In winter the heterogeneity G was significant and contributed 17.6% of the total G value. That difference was attributable to 91% of Mercenaria mercenaria being in the opaque growth stage during winter, whereas only 70% of M. campechiensis and 66% of hybrid forms were in the opaque stage during that season. However, the difference was significant only between M. mercenaria and hybrid forms (STP, Table 3 ) but not between the parental species. For all other pairwise comparisons among genotype classes within each season, the STP results were non-significant. Growth-phase analysis clarified the differences in growth-band formation described above (Figure 5). For all 37

PAGE 53

Table 3. Significance tests for growth-band formation patterns among genotype classes, within each of four seasons, from the Indian River lagoon, Florida. Seasons are defined in Materials and methods. Significant differences in the proportion of translucent and opaque growth bands, overall (pooled), among genotype classes (heterogeneity), and within each genotype class (Mercenaria mercenaria, M. campechiensis, and hybrid forms), were determined using the Goodness-of-Fit G-test (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). Significant differences among genotype classes were further defined using the Simultaneous Test Procedure (STP) with p < 0.05. For the STP, genotype classes for which the proportions of translucent and opaque growth bands were not significantly different within each season are denoted by a common letter. Sea s on Fall Winter Spring Summer Test df G STP df G STP df G STP df G STP ---Pooled 1 66. 2 1 48.2 1 72 6 1 39 6 -Heterogeneity 2 1 .0 2 10 3 2 4 0 2 3 1 ---M mercenaria 1 41. 1 A 1 53 7 A 1 58 6 A 1 22 3 A -M campechien s i s 1 5 1 A 1 1 6 AB 1 0 5 A 9 7 A Hybrid form s 1 21. 1 A 1 3 2 8 1 17 4 A 1 10 7 A -Total 3 67.2 3 58 5 3 76 6 3 42. 7 38

PAGE 54

three genotype classes, the distribution of growth phases was similar during fall and summer when most clams were in the translucent stage of growth. Most of those clam s that were in the opaque stage during fall were sampled during November and were in the 01 phase, whereas most clam s that were in the opaque stage during summer were sampled during June and all were in the 03 phase. Howe ver, during winter and spring growth-phase patterns differed substantially among genotype classes. During those seasons, most Mercenaria mercenaria were in some phase of opaque growth; most animals in the translucent growth-stage during winter were collected in December and most animals in the translucent growth-stage during spring were collected in April or May (Fig. 5A). In contrast, 30% of the winter M campechiensis collection remained in some phase of translucent growth and T-phase animals were collected as late as February (Fig. 5B). No clear pattern was evident in the spring M. campechiensis collection; a Tl-phase animal was collected in May and T3-phase animals were collected in March and April. For hybrid forms, T-phase animals were abundant during all winter months; only Tl phase animals were collected during December but all three phases were collected during January and February (Fig. 5C). During spring the growth-phase pattern of hybrid forms was similar to that observed for M. mercenaria, with few animals 39

PAGE 55

remaining in the translucent growth stage during that season. Discussion The formation of coupled translucent-opaque growth bands in the shells of hard clams from the Indian River lagoon follows an annual cycle and is similar for all three Mercenaria genotype classes. For each genotype class the opaque growth band, indicative of rapid shell growth, forms during winter and spring and the translucent growth band, indicative of slow shell growth, forms during summer and fall. Seasonal growth patterns of all hard clam genotype classes in the Indian River lagoon are therefore similar to seasonal growth-band formation patterns reported for natural assemblages of M. mercenaria from North Carolina (Peterson et al., 1983), Georgia (Clark, 1979; Quitmyer et al., 1985), and Florida (Arnold et al., 1991a) and of M. campechiensis from various sites in Florida (Jones et al., 1990; Quitmyer and Jones, 1992). Consistency in the pattern of annual growth-band formation in hard clam shells has been documented both among years within a single habitat (Quitmyer et al., in press) and among habitats within a single year (Peterson et al., 1985; Quitmyer and Jones, 1992). However, there exist substantial latitudinal differences in the timing of growth40

PAGE 56

band formation within Mercenaria mercenaria (Peterson et al., 1983; Grizzle and Lutz, 1988), within M. campechiensis (Jones et al., 1990), and between the two species (Quitmyer et al., 1985; Arnold et al., 1991a; Quitmyer and Jones, 1992). It has not been clear whether those differences were due to environmental factors associated with differences in date or site of collection, or to inherent genetic differences among populations. The contemporaneous patterns of growth-band formation that I found among hard clam genotype classes collected from similar habitats in the Indian River lagoon, Florida, suggest that previously reported differences in the timing of growth-band formation among hard clam populations in the southeastern United States are environmentally rather than genetically mediated. Temperature has direct or indirect effects on the shell growth of Mercenaria (Ansell, 1968; Clark, 1979) and a variety of other bivalve genera [e.g., Rangia (Aten, 1981), Mytilus (Almada-Villela et al., 1982}, Macoma (Beukema et al., 1985), Crassostrea (Brown, 1988), and Phacosoma (Tanabe and Oba, 1988)], and is the primary environmental factor influencing the formation of annual growth bands in shallowwater bivalves such as Mercenaria (Barker, 1964; Pannella and MacClintock, 1968; Kennish, 1980; Peterson et al., 1983; Jones et al., 1990; Arnold et al., 1991a). For example, M. mercenaria deposits a distinct translucent growth band within its shell during winter in northern latitudes 41

PAGE 57

{Pannella and MacClintock, 1968) and during summer in southern latitudes (Peterson et al., 1983). Similar temperature-related (Beukema et al., 1985) latitudinal variation has been reported for the bivalve Macoma balthica; spring growth typifies populations from southerly latitudes whereas summer growth was described for populations from more northerly latitudes (Bachelet, 1980). However, other environmental influences on the timing of bivalve growth also have been described, including salinity (e.g., Craig et al., 1988), reproductive activity (e.g., Jones et al., 1978; Peterson and Fegley, 1986; Sato, 1995), food availability (Beukema et al., 1985; Peterson and Beal, 1989; Page and Ricard, 1990), hydrodynamics (Kerswill, 1949; Peterson and Beal, 1989), light regime (Nielsen and Stromgren, 1985), sediment composition (Pratt, 1953), and source of stock (Mallet et al., 1987; Peterson and Beal, 1989). Of these, the first four probably are most significant to my considerations of seasonal growth-band formation patterns. Salinity may influence the annual cycle of growth-band formation in bivalve molluscs (Barker, 1964), and intraspecific differences in the shell structure of Mercenaria in response to salinity have been noted (Cunliffe, 1974), but the effect of salinity on seasonal growth of Indian River hard clams is probably limited (Arnold et al, 1991; 1996). Although growth cessation in hard clams from Texas bays was attributed to limiting 42

PAGE 58

salinity (Craig et al., 1988), Indian River lagoon waters do not express a strong seasonal pattern of salinity variation (Costa, 1986). Both species of hard clam grow well within the 15-35 ppt salinity range (Castagna and Chanley, 1973; Stanley and DeWitt, 1983) typical of the Indian River lagoon (McCall et al., 1970; Costa, 1986). Although sudden salinity changes are typical of certain areas of the lagoon (Arnold et al., 1991a), those events should be reflected as a unique and easily distinguishable disturbance break within the shell (Kennish, 1980). Reproductive development, food availability, and hydrodynamics are seasonally-scaled events that may interactively influence bivalve shell-growth cycles. For example, current patterns will affect food supply (Grizzle and Morin, 1989) which in turn affects reproductive development and growth (Peterson and Fegley, 1986). Spawning events are generally reflected in the shells of bivalve molluscs by discrete but relatively short-term changes in the pattern of daily growth-line deposition (Kennish, 1980; Sato, 1995). However, spawning in Indian River hard clams is bimodal and poorly synchronized among individuals within the population (Hesselman et al., 1989), and is probably reflected in their shells by multiple narrow disturbance bands. Hesselman et al. (1989) described a "resting/spent" phase in the reproductive cycle during August and September, possibly in response to water 43

PAGE 59

temperature > 30C, but a disturbance band associated with tnat event would be embedded within the translucent component of the annual growth band and therefore would be indistinguishable from that component. Nevertheless, developmental stages of the reproductive cycle prepatory to spawning may affect seasonal shell growth patterns of bivalve molluscs (Peterson and Fegley, 1986). Those effects may be interactive with food supply and hydrodynamics, but such influences on annual shell-growth cycles are poorly understood and in need of further study. Despite the overall similarities among hard clam genotype classes in the seasonal pattern of shell growth in the Indian River lagoon, some differences in the timing of shell growth among my defined genotype classes are detectable. Those differences cannot be attributed to environmental factors because all animals lived in a similar environment. Nor were those differences related to differences in shell size among the genotype classes because, although a significant difference in mean shell size was detected between Mercenaria mercenaria and the other two genotype classes, that difference was slight. overall, the size distribution among genotype classes was very similar and most clams were well within the size range typical of rapidly growing hard clams in the Indian River lagoon (Jones et al., 1990). 44

PAGE 60

The growth pattern of Mercenaria mercenari'a was seasonally distinct: more than 90% of the individuals I sampled were in the rapid-growth stage during winter and spring and more than 80% of sampled individuals were in the slow-growth stage during summer and fall. Those few animals in the less-represented stage during any season were either "late" in initiating the predominant growth stage (e.g., during summer) or were "early" in exiting that stage (e.g., during fall) The distinct seasonal pattern of growth-band formation that I describe for Mercenaria mercenaria was not as evident for M. campechiensis nor for hybrid forms. Only 65-70% of my M. campechiensis were in the rapid-growth stage during winter and spring, and those animals in the slow-growth stage showed no clear phase-related pattern. Although nearly all hybrid-form individuals were in the rapid-growth stage during spring, less than 66% of hybrids were in the rapid-growth stage during winter. Again, no phase-related pattern was evident during winter. Maximum shell growth rate of Mercenaria mercenaria is reported to occur at 20 c with a gradual decrease above and below that temperature (Ansell, 1968). Maximum calcium deposition rate within the shell is reported at approximately 13 C and again at 24 C (Storr et al., 1982). That information is consistent with my observations of rapid growth in shells of M. mercenaria during winter and spring, 45

PAGE 61

when water temperature fell below 20 C in the lagoon, and slow growth during summer and fall when water temperature generally exceeded 20 C. Although calcium deposition may occur at equal rates both above and below 20 c (Storr et al., 1982}, calcium dissolution in response to anaerobic metabolism (Lutz and Rhoades, 1977) may be more pronounced during conditions of relatively high ambient water temperature, low environmental oxygen availability, and high basal metabolic rate generally expected during summer and fall. However, I saw no indication of continued growth of M. campechiensis during the warm summerjfall months, suggesting that the trend of increasing growth of that species with increasing water temperature (Ansell, 1968) does not hold for natural assemblages of M. campechiensis in Florida. Unfortunately, information on temperature-specific calcium deposition rates in M. campechiensis is not available. The annual growth cycle of hybrid forms is intermediate between that of the two parental species, being similar to M. campechiensis in winter and to M. mercenaria during spring. Similarly, the annual rate of shell growth of cultured F1 hybrid hard clams tends to be intermediate between that of the parental species (Menzel, 1962; Dalton, 1977). Shell growth rate of Mercenaria is a heritable trait (Hadley et al., 1991} and clams have been selected for rapid growth (Crenshaw et al., 1996). I suggest that the timing of growth also may be heritable, in which case it may be 46

PAGE 62

possible to exploit the subtle differences in timing of growth that I describe to minimize the time to market size in this commercially valuable organism. Seasonal patterns of growth-band formation have been extensively applied in paleontological analyses, including determination of coastal occupation patterns of humans (Quitmyer et al., in press), reconstructing sexual development patterns of fossil bivalves (Sato, 1995), verifying changes in environmental rhythms over time (Pannella and MacClintock, 1968), assessing changes in the rate of rotation of the earth (Rosenberg and Runcorn, 1975), and numerous other paleoecological and ecological problems (Rhoads and Lutz, 1980). My results provide further evidence that growth bands in the shells of bivalve molluscs are deposited in response to environmental cues. Additionally, I show that a faithful seasonal signature can be derived from mixed-species assemblages even when those species are closely related but only rarely occur in sympatry. This provides assurance that, when drawing conclusions based upon samples collected from widely different sites inhabited by congeric specimens, observed differences will be environmentally rather than genetically based. 47

PAGE 63

CHAPTER 3. HABITAT-SPECIFIC GROWTH OF THE NORTHERN HARD CLAM, Mercenaria mercenaria (L.), FROM THE INDIAN RIVER LAGOON, FLORIDA Introduction The analysis of internal shell growth-increment patterns is a reliable technique for determining the growth rate of bivalve molluscs. Shell growth patterns were described by Barker (1964), and have been applied in a variety of paleochronological (Rosenberg and Runcorn, 1975), sclerochronological (Jones et al., 1989), archaeological (Quitmyer et al., 1985), and ecological (Greene, 1975) investigations. Although the formation of shell growth patterns is not always periodic (Gould, 1979; Jones, 1981) and may be driven by stochastic as well as cyclical events (Kennish, 1980), it is apparent that certain repeating structural features exhibit a periodicity of formation (Kennish, 1980). Hard clams of the commercially important g .enus Mercenaria are particularly amenable to internal shell growth-increment analysis. Barker (1964) described five distinct growth layers found in the shell of Mercenaria 48

PAGE 64

mercenaria; the frequency of deposition of these layers ranged from tidal to annual. Subsequently, researchers have analyzed shell growth-increments to determine the shell growth rates of hard clam populations from a variety of Atlantic and Gulf of Mexico populations (e.g., Saloman and Taylor, 1969; Greene, 1978; Kennish and Loveland, 1980; Peterson, 1983; Peterson et al., 1983; Grizzle and Lutz, 1988; Jones et al., 1989; 1990). Most research efforts have focused on annual shell growth-increment analysis, a technique that is relatively easy to apply and interpret and that has implications in a variety of contexts. The annual pattern of first-order shell growth increments (sensu Barker, 1964) is well established for hard clam populations from locations throughout the Atlantic and Gulf of Mexico coasts of the United States (e.g., Peterson et al., 1983; Grizzle and Lutz, 1988; Jones et al., 1989; 1990). Information on the basic ecology of hard clams is available for a variety of locations (e.g., McHugh et al., 1982; McHugh and Sumner, 1988), but useful data on the ecology of hard clam populations from the east coast of Florida have become available only recently. Studies have shown that hard clams on the Florida east coast grow more rapidly (Jones et al., 1990), spawn more frequently (Hesselman et al., 1989), and die at a younger age (Jones et al., 1990) than do their northern counterparts. Shell growth rate can affect the life history strategy of hard 49

PAGE 65

clams in a variety of ways (e.g., Peterson, 1983; Arnold, 1984) and is important both ecologically and economically. More specific information is needed, particularly concerning variations in shell growth rate among local populations, to better manage the commercial fishery and to understand factors that control hard clam distribution and abundance. Considerable research has focused on the fact that the shell growth rates of hard clams vary throughout the geographic range of the animal. This geographic variation in the growth rates of hard clam shells was summarized by Ansell (1968), but data upon which this summary was based were inadequate in several respects. For example, several important localities, including the Atlantic coast of Florida, were not represented. Furthermore, most of the data were obtained from non-native clams or from clams cultured under artificial conditions. Shell growth rates of cultured clams vary with the source of the stock (Walker and Humphrey, 1984), stocking density (Walker, 1985), and frequency of sampling (Walker, 1985). A review of the growth rates of hard clam shells, as determined by analyzing the internal shell growth-increments of natural populations from throughout the geographic range, is needed to clarify local and latitudinal variations in shell growth rates of this animal. To fully understand variation in shell growth rates o f hard clams among populations throughout the range of the 50

PAGE 66

genus, variation within local populations must first be defined. To accomplish this, a single quantitative growth parameter that is comparable between populations is required. The w parameter (Gallucci and Quinn, 1979), based upon the k and parameters of the von Bertalanffy growth equation (von Bertalanffy, 1938), is suitable for this purpose. The von Bertalanffy growth function has traditionally received wide application in the analysis of bivalve growth (e.g., Brousseau, 1979; Gallucci and Quinn, 1979; Appeldoorn, 1983; Schick et al., 1988; Tanabe, 1988; Jones et al., 1990). When compared against the Gompertz, power, logistic, and exponential models (Kaufmann, 1981}, the von Bertalanffy function has been shown to be the model that best fits (highest r2 ) the pattern of shell growth recorded for Indian River hard clam populations (Jones et al., 1990}. Thew parameter, which corresponds to growth rate near t0 (Gallucci and Quinn, 1979), has previously been applied to comparisons of both habitat-specific growth rates (Appeldoorn, 1983} and location-specific growth rates (Jones et al., 1990) of bivalve molluscs. In this study I took a similar quantitative approach to the analysis of shell growth rates of Mercenaria mercenaria from various habitats within a single geographic entity, the Indian River lagoon, Florida (Figure 6). Internal shell growth-increment analysis was used to determine growth relationships in naturally occurring populations. I first 51

PAGE 67

established the periodicity of shell growth-increment formation in M. mercenaria from the Indian River lagoon; I then used that information to determine significant differences in shell growth rate, as described by the w parameter, among M. mercenaria collected from quantitatively definable benthic habitats within the Indian River. My results will help clarify the significance of variation in shell growth rates previously described for hard clam populations on a larger geographic scale. Materials and Methods Classification of Individuals The Indian River hard clam population is composed of Mercenaria mercenaria, M. campechiensis, and hybrids of the two species (Dillon and Manzi, 1989). The relative proportions of the three taxonomic entities vary among areas within the Indian River (Bert et al., 1988; Bert et al., 1993). Because shell growth-increment formation and shell growth rates may differ among these taxonomic groups, I restricted my analysis to genetically defined M. mercenaria. This approach permitted me to compare growth patterns of M. mercenaria from the Indian River with patterns reported from other areas within the species range. 52

PAGE 68

All clams obtained from the Indian River were taxonomically categorized (Mercenaria mercenaria, M. campechiensis, or hybrid) by applying discriminant function analysis to individual genotype index scores. Index scores were formulated from composite genotypes generated using allele frequencies from four diagnostic protein loci (glucose phosphate isomerase, phenylalanine-specific dipeptidase-2, aspartate amino transferase, and phosphoglucomutase) (Bert et al., 1993). The loci were identified using starch-gel electrophoresis on combined gill and mantle tissue from over 1600 individuals. All electrophoretic assays were conducted by Biochemical systematics Laboratory personnel at the Florida Department of Environmental Protection Marine Research Laboratory. Electrophoretic procedures and techniques followed Bert (1985; 1986) and are described in detail in those publications. Field Sampling Locations and Techniques The Indian River is a semi-enclosed lagoon connected to the Atlantic Ocean by several man-made and natural channels through the barrier islands that extend along the east-central coast of Florida (Figure 6). The study area 53

PAGE 69

Figure 6. Boundaries of shellfish harvesting areas (A-F, IRl) and locations of monthly sampling stations (denoted by closed circles) for growth-increment formation study in the Indian River lagoon, Florida.

PAGE 70

28.30' 28. zs as 81. ao 55 ATLANTIC OCEAN 8o3o

PAGE 71

extended from Haulover Canal south to CR 510. Water depth within the study area averages 1.5 m; maximum depths of 4 m are found in the Intracoastal Waterway, which roughly bisects the Indian River throughout its length, and in areas around Melbourne and Eau Gallie (White, 1986}. The tidal range in the lagoon is negligible except near Sebastian Inlet (Smith, 1987). The Indian River Mercenaria mercenaria used in this study were collected using two different sampling strategies. To determine variation in shell growth rate among habitats, hard clams were collected during July and August 1986 in a short-term, intensive survey conducted throughout the study area. Divers equipped with hand rakes collected hard clams from each of 75 randomly allocated 1-m2 quadrats located in each of the seven shellfish harvesting areas (A-F, IR1} within the Indian River (Figure 6). A total of 922 clams was obtained from 525 sampling quadrats. To document the annual periodicity of shell growth-increment formation in M. mercenaria, samples were collected monthly from September 1987 through August 1988 from established stations located in shellfish harvesting areas c, D, E, and F (Figure 6}. All clams were returned to the laboratory, where tissue samples were taken for electrophoretic analysis and shells were cleaned and numbered, disarticulated, and stored for later growth-increment analysis. 56

PAGE 72

Throughout the collecting period for the growth-increment formation study, temperature ( C) and salinity (ppt) were measured twice monthly at each repetitive sampling station, at a point approximately 0.5 m above the sediment-water interface, using a YSI Model 33 s-e-T meter. Biweekly temperature and salinity values were pooled over areas and averaged within each month to develop annual temperature and salinity profiles for the study area. During the short-term intensive survey, triplicate, 2.5-cm-diameter x 5-cm-deep sediment cores were extracted from each of the 525 1-m2 sample quadrats before raking for clams. Sediment samples were pooled in the field, numbered, returned to the laboratory, and stored at room temperature for later analysis. Percent cover of submerged aquatic vegetation (SAV) was visually estimated for each quadrat, and water depth at each quadrat was determined by sounding line. The SAV was generally composed of various combinations of the seagrasses syringodium filiforme Kuetz. and Halodule wrightii Aschers. or of the marine alga caulerpa prolifera (Forsskal) Lamoureux (Rice et al., 1983). Sediment composition was determined in the laboratory according to the methods of Folk (1974), with the exception that sediments were wet-sieved. Pooled samples from each station were analyzed for percent composition of gravel (grain size> 2.00 mm), sand (grain size 0.063-2.00 mm), and 57

PAGE 73

silt-clay (grain size < 0.063 mm). Percent organic matter was determined by ignition at 550 c for 6 h. Determination of Growth Periodicity One valve from each genetically identified Mercenaria mercenaria collected during the monthly sampling program was radially sectioned from umbo to ventral margin, along the axis of maximum height (Figure 7A), using a Highland Park Model SSP20 lapidary saw equipped with a diamond blade. Monthly samples for the growth-increment formation analysis were not embedded prior to sectioning. The alternating pattern of translucent (dark = slow growth) and opaque (white = rapid growth) increments was then examined (Figure 7B), and the periodicity of growth-increment formation was analyzed using the T-O staging method (Jones et al., 1978; Quitmyer et al., 1985). The terminal growth increment was identified as being in the translucent (T) or opaque (0) stage and was then classified to growth phase as determined by size of the increment (increment beginning to form: T1 or 01; increment approximately one-half as large as the previous increment of the same stage: T 2 or 02; increment as large as, or larger than, the previous increment of the same stage: T3 or 03). These six subdivisions established a reliable seasonal profile of incremental shell growth based upon the frequency with which M mercenaria were found in 58

PAGE 74

Figure 7. A} Shell of Mercenaria mercenaria illustrating position of radial cut along maximum axis of dorso-ventral growth; B) Cross-section view of M. mercenaria shell, illustrating the pattern of opaque and translucent growth increments and their relationship to the annual growth increment (from Quitmyer et al., 1985).

PAGE 75

A 8 Opaque lncrem ant Translucent I ncrement Ventral Margin POSTERIOR Ventral margin ANTERIOR Middle S h 60 Outer she l Ia yer ell Layer Umbo

PAGE 76

the various phases of growth (Claassen, 1990). Heterogeneity in the proportion of translucent and opaque growth increments was tested for significance by the Replicated Goodness-of-Fit procedure (G-test), and heterogeneity among months was analyzed by the Simultaneous Test Procedure (Sakal and Rohlf, 1981). Monthly samples from areas c, D, E, and F were pooled for analysis of the periodicity of growth-increment formation. Preliminary analysis indicated that no difference in the monthly pattern of growth-increment formation, among areas within the lagoon, could be discerned based upon the available sample size of Mercenaria mercenaria during certain month-by-area combinations. Previous work on variation in the pattern of growth-increment formation of M. mercenaria across habitats within Cape Lookout, North Carolina (Peterson et al., 1985) and Charlotte Harbor, Florida (Jones et al., 1990), indicated that such variation was small or nonexistent. I assumed a similar consistency in the pattern of growth-increment formation across habitats within the Indian River. Microgeographic Variation in Growth For growth rate analysis, one valve from each specimen collected during the July-August 1986 intensive study was 61

PAGE 77

embedded in EPON 815 resin (Kennish et al., 1980) to prevent breakage and then was sectioned along the axis of maximum shell height. In most cases, no further preparation was required prior to growth-increment analysis. However, sectioned shells with closely spaced or indistinct growth increments were polished through 240-, 400-, and 600-grit emery paper, smoothed with 1.0 micron alumina microgrit, etched with 0.1N HCl, and impressed upon sheet acetate (Kennish et al., 1980}. Each clam was analyzed for growth rate according to standard techniques (Peterson et al., 1983; Jones et al., 1990}. Each pair of one opaque plus one translucent growth increment constituted one annual growth increment; the ventral margin of each translucent growth increment defined the end of an annual cycle (Figure 7B). A complete record of size at age for each clam collected during the survey was obtained by measuring incremental growth as the distance from the ventral margin of each translucent band (marked by pencil) to the umbo using Vernier calipers accurate to 0.1 mm. Mercenaria mercenaria collected from the 525 sampling sites were classified into habitat-specific subpopulations by grouping the sites according to habitat type using three different criteria. The primary selective procedure was a cluster analysis (CLUSTER Ward's Method, SAS Version 5 [1985]) using percent composition of sand, water depth in 62

PAGE 78

meters, vegetative cover, and percent organic matter. Gravel and silt-clay composition were correlated with sand composition (Pearson Correlation Coefficients= -0.75 and -0.85, respectively) and were not used in the cluster analysis. The influence of sediment composition (Pratt, 1953; Pratt and Campbell, 1956) and vegetative cover (Peterson et al., 1984) on hard clam shell growth rate are well established. Two subjective secondary selective procedures were employed, based upon the results of the cluster analysis. These a posteriori classifications defined subpopulations by depth ( < 1.0 m, 1 .0-2.0 m, > 2 0 m) and, separately, by percent of vegetative cover (0% 1-20%, 21-40%, 41-60%, 61-80% 81-100%). Growth rate also was compared among the seven shellfish harvesting areas to determine if a gradient in shell growth rate existed along the north-south extent of the lagoon. To compare growth rates among subpopulations grouped by habitat or area, the w parameter was calculated for each subpopulation. To determine w shell growth was first modelled by fitting a von Bertalanffy growth function to the shell height-at-age data. The function is described by the equation SHt = [ 1 e-k
PAGE 79

function by iterative non-linear least squares regression (NLIN Multivariate Secant Method, SAS Version 5 (1985]). This procedure yielded estimates of SH_ k, and t0 plus asymptotic standard errors and an asymptotic correlation matrix of the estimates. Values of r2 (Kvalseth, 1985) for all populations exceeded 0.95, further justifying the use of the von Bertalanffy equation to describe growth in hard clam populations from the Indian River. The growth parameter w (= k x SH_), together with its variance, was calculated from the von Bertalanffy parameter estimates for each subpopulation (Gallucci and Quinn, 1979). The w parameter integrates k and SH_ to provide an estimate of growth rate near t0 (Gallucci and Quinn, 1979), and has been successfully applied to growth comparisons for a variety of bivalve populations (Appeldoorn, 1983; Jones et al., 1989; 1990) to compare growth among habitats or regions. Use of w ignores the influence of t0 (Kappenman, 1981), but this is not a serious problem since t0 simply positions the curve along the x-axis (Appeldoorn, 1983). The calculated subpopulation w values within each of the three habitat classifications, and among shellfish harvesting areas, were compared statistically by the Games and Howell method (Games and Howell, 1976; Sokal and Rohlf, 1981) This technique is not affected by heterogeneous variance and requires knowledge only of the mean and variance of each subpopulation. 64

PAGE 80

Figure 8. Monthly pattern of T-O growth-increment formation in Mercenaria mercenaria, pooled over shellfish harvesting areas c, D, E, and F. Dashed line connects monthly temperature means ( C S.D.); dotted line connects monthly salinity means (ppt S.D.). Horizontal bars above the histogram connect months in which the proportion of translucent (T) or opaque (0) increments are homogeneous (Simultaneous Test Procedure (Sokal and Rohlf, 1981] p < 0.05). See text for explanation of T1, T2, T3, 01, 02, and 03.

PAGE 81

T1 T2 T3 100 90 80 >-70 35 c.,:) ::z: 1.1..1 60 30 :::: c::::1 1.1..1 25 a:: 50 1.1.... 40 20 30 15 20 10 10 5 0 0 Sep Nov Dec Jan Feb Mar Apr May Jun Jul Aug 1987 1988

PAGE 82

Results Classification of Individuals Of the 856 hard clams obtained during the monthly sampling program, I applied the T-O growth periodicity procedure to 356 genetically identified Mercenaria mercenaria. Of the 922 hard clams obtained during the short-term intensive survey, I applied annual growth-increment analysis to 465 genetically identified M. mercenaria. Growth Periodicity Macroscopic growth increments visible in radial sections of shells of Mercenaria mercenaria from the Indian River show an annual pattern of formation (Figure 8). From July through November, most M. mercenaria were in some phase of translucent growth-increment formation. From January through May, most M mercenaria were in some phase of opaque growth-increment formation. A transition period, during which the proportions of translucent and opaque growth-increments were similar, occurred during both December and June. Replicated Goodness-of-Fit test results supported these observations (Table 4). Sample heterogeneity was significant (p 0.05) for all months 67

PAGE 83

Table 4. Replicated Goodness-of-Fit G-test for heterogeneity of growth increments in Mercenari a mercenaria from the Indian River lagoon. A) Overall ratios of translucent (T) to opaque (0) growth increments. B) Contributions due to individual months ( p < 0.05). A) Test df G Pooled 1 4 97 Heterogeneity 11 246 .24 Total 12 251.21 B) Test df G September, 87 1 26.26 October 87 1 24. 95 November, 87 1 6 .28 December, 87 1 1 .30ns January 88 1 6 .28 February 88 1 50. 05 March 88 1 30.19 April 88 1 26.66 May, 88 1 18. 03 June,88 2 50n s July, 88 1 18.50 August 88 40.20 Total 12 251 .21 68

PAGE 84

except June and December. Results of the Simultaneous Test Procedure indicated that changes in the relative proportion of translucent and opaque terminal growth increments were gradual but significant over time. Mean monthly bottom water temperature recorded during this study ranged from 14.5 C to 30.8C, with a maximum standard deviation among stations within any month of less than 1.0C. Mean monthly salinity ranged from 20.1 ppt to 25.3 ppt, with a maximum standard deviation among stations within any month of 4.2 ppt. Mean monthly temperature and salinity curves, pooled over all sampling stations, are plotted in Figure 8 for comparison with the monthly pattern of growth-increment formation. Mean monthly temperature and salinity, averaged within each shellfish harvesting area, are plotted in Figures 9A and 9B, respectively. Figures 9A and 9B illustrate the similarity of temperature, and the latitudinal trend in salinity, over areas within the lagoon. Temperature and salinity trends recorded during my study are similar to long-term temperature and salinity trends recorded for the Indian River (McCall et al., 1970; Costa, 1986}. Microgeographi c Variatio n in Growth The w values determined from growth parameters of Mercenaria mercenaria were significantly different among 6 9

PAGE 85

Figure 9. A) Monthly pattern of temperature variation among shellfish harvesting areas within the Indian River lagoon, Florida, during September, 1987, through August, 1988; B) Monthly pattern of salinity variation among shellfish harvesting areas within the Indian River lagoon, Florida, during September, 1987, through August, 1988.

PAGE 86

A B 32 3 0 2 8 2 6 2 4 "" = 2 2 "" "'-20 18 16 14 32 3 0 2 8 2 6 >24 -:z: --' ""' V') 22 2 0 18 ' ' ' O .. .. \\ i \:\o if \ I v 'I 'I ----....... 1--l o-o N 1987 M A M 11 II I I I '. I \ I \ I I \ 1988 DATE --0 0 c 0 E F ,' ............ ........... 0 II I I I I I I 0 / \ I \ I I I I \I \ \1 \ \ , \ A..ill c D E F A I I I / .... -/---6\ \ \ 0 r I 0 i \ X /o X \ 0 \ ,.. .. .... 0 1/ \ / r:...-x, \ I / 'x\ I \ X \ / \'.. X X-'" X \' N 1987 I I I M A M 1988 DATE 71 \ X \ \ X A

PAGE 87

(('IDalb>J.ee .:5, ) . 1-f.. .irn c:amaa 3:, -wuthw.arO. to ca Jniill1ia.'rfn.illn i.i.IJl IE;,, "l:tlllle:n to ca>::tfea. JiiRll.. lAl.l.J.l.. "C thr-ough rm1 { (!p m . !I!51 ) :!5;;2=;i :;:>jimpling -sites fu:Mtitta":t. ;p) :s-Je:s '9r-ouped 2 t:tm:G:se ,w..i::tttm.u:t tAV.. :I-n -< ::...1 .-m Wh.!=-re SA V is ?:t. a t-.i:.Pns wit-h mean SAV i._,n 'd!he :fo
PAGE 88

Table 5. Games and Howell (1986) analysis of differences in w values among shellfish harvesting areas within the Indian River lagoon, Florida. Numbers above diagonal are minimum significant differences {MSD; p = 0.05). Numbers below diagonal are differences between means for each comparison . ( p < 0.05). Also 1ncluded are results of the non-l1near curve-fitting analysis, including w value and standard deviation for each curve. See text for explanation of w k, SH .. and t0 Area comparison AB c D E F IR1 N w SOw k SH. AB 16. 9 16 5 16 5 16 5 16 6 5 29 1 5 9 0 .33 87 9 0 06 c 12.6 2 3 2 3 2 3 2.4 48 41.6 3 9 0 .61 68. 1 0 10 D 2 5 10.0* 0 .5 0 7 1 1 110 31. 6 1 2 0 .3 8 82.5 0 12 E 4 8 17.3* 7.3* 0 6 1 0 174 24 3 0.6 0.26 92.4 0 07 F 1 4 11.2* 1 .2* 6 .1* 1 1 85 30. 4 1.2 0.34 90.4 -0 .01 IR1 3 6 8.9* 1 .1* 8.4* 2 3 43 32.7 1.6 0 .41 79 4 -0 .06 73

PAGE 89

Table 6. Cluster analysis based upon habitat components in the Indian River lagoon, Florida. Values for mean submerged aquatic vegetation (SAV} have been calculated from the following relationships: 0 = 0%, 1 = 1-20% 2 = 21-40%, 3 = 41-60% 4 = 61-80%, and 5 = 81-100% vegetative cover. Cluster 1 2 3 4 5 6 X so RANGE -X so RANGE -X so RANGE -X so RANGE X so RANGE -X so RANGE Depth (m) 2 5 1 2 0.9-4.3 2.3 0.5 1 7-3 7 1.2 0 3 0.5-1 5 1 6 0 6 0 3-3 2 0 9 0 3 0 5-2 0 0 8 0 3 0 .3-1. 5 Sand % 65.6 25 4 2.5-93 5 81. 8 16 7 9 2-97 7 91. 6 9 6 46 5-98 4 89 2 10 8 50.4 98 2 92 7 6 4 71. 8-98 0 92 3 5 5 68 1-97 8 74 Organics% 4.0 6.0 0 8 22.0 2.2 2.4 0-17 6 1.2 1.4 0-9.3 1.4 1 1 0 2-5 6 1 0 0 7 0-3.3 1 2 0 7 0 1-3 2 SAV 2 1 2 1 0-5.0 0 0 0 0 0 0 1 1 0.2 1-2 2 6 0 5 2 3 4 7 0 5 4-5

PAGE 90

Table 7. Games and Howell (1976) analysis of differences in w values among clusters within the Indian River lagoon, Florida. Numbers above diagonal are minimum significant differences (MSD; p = 0.05). Numbers below dia9onal are differences between means for each comparison ( p < 0.05). Also included are results of the non-linear curve-fitting analysis, including w value and standard deviation for each curve. See text for explanation of w k, and t0 Cluster comparison 1 2 3 4 5 6 N w SOw k SH. to 1 1 5 1 5 1.8 3 9 1.9 39 34 5 2 3 0 .41 83 1 0 17 2 10 5 0.4 1 1 3 9 1 2 189 24 0 0 8 0.27 87 5 -0 12 3 4 3 6.2* 1 1 3 9 1 2 151 30 2 0 8 0 .35 85 8 0 .04 4 1 8 8 .7* 2 .5* 3 9 1 6 49 32 7 1 9 0.40 81. 4 0 .07 5 2 5 8.0* 1.8 0 .7 3.9 8 32 0 2 2 0 39 81. 7 0 .11 6 3 8 6 .7* 0.5 2 .1* 1 4 29 30 6 1 6 0.36 85.2 0 .01 75

PAGE 91

Table 8. Games and Howell (1976) analysis of differences in w values among depth zones within the Indian River lagoon, Florida. Numbers above diagonal are minimum significant differences (MSD; p = 0.05). Numbers below dia9onal are differences between means for each comparison ( p < 0.05). Also included are results of the non-linear curve-fitting analysis, including w value and standard deviation for each curve. See text for explanation of w k, and t0 Depth comparison <1m 1 -2m >2m N w Sd w K SH <1. 0m 0 4 0.4 76 33 7 1.2 0 42 80 6 0 00 1 0 2 0 0.2 211 28 3 0 7 0 .31 89 9 -0 .05 >2 0m 3 .7* 178 24.6 0.9 0 28 87 4 -0.06 76

PAGE 92

Table 9. Games and Howell (1976) analysis of differences in w values among submerged aquatic vegetation (SAV} categories within the Indian River lagoon, Florida. Numbers above diagonal are minimum significant differences (MSD; p = 0.05). Numbers below diagonal are differences between means ... for each compar1son ( p < 0.05). Also 1ncluded are results of the non-linear curve-fitting analysis, including w value and standard deviation for each curve. See text for explanation of w k, SH .. and t0 Vegetation comparison 0 1-20 21-40 41-60 6180 81100 >0 N w Sdw k 0 1 4 8.2 3.9 2 5 1.1 0.3 349 26 .5 0 5 0 30 87. 1 -0 .08 1 -20 6 .2* 7.9 3 9 2 8 1 8 44 32. 7 2 4 0 .41 79.7 0.05 2140 13.2* 7 .0 8.0 7 9 8 3 6 39.8 3.5 0 .49 81. 1 0 .31 41-60 5 .4* 0 8 7 8 4 2 3 8 8 32.0 2 2 0 39 81. 7 0 .11 61-80 1 0 .1* 4 .0* 3.1 4.7* 2 3 20 36 7 2 6 0.44 83.4 0 .21 81-100 7 .0* 0 8 6 2 1 6 3 .1* 38 33 6 1 6 0 .39 85 1 0 .10 >0 7 .0* 116 33 6 1 0 0 40 83. 2 0 .11 77

PAGE 93

cluster without SAV, and the shallow-water cluster with 60-100% SAV. Depth alone was strongly associated with the w value; w decreased significantly and monotonically with increasing depth (Table 8). The effect of SAV alone on thew value was determined by the presence or absence of vegetative cover. Mercenaria mercenaria in habitats without SAV had an w value that was significantly lower than the values observed in habitats characterized by any level of vegetative cover (Table 9). Discussion Growth Periodicity The seasonal pattern of growth-increment formation recorded for Mercenaria mercenaria from the Indian River is similar to that described for M. mercenaria populations from other sites along the Atlantic coast of the southeastern United States. Data forM. mercenaria from Kings Bay, Georgia, indicate that rapid growth occurs during December through March; that a transition in the phase of growth-increment formation occurs during April and May; and that slow growth occurs from June into November (Jones et al., 1990). Clark (1979) reported slow growth in M. mercenaria from Georgia during summer and early fall and 78

PAGE 94

fast growth during winter and spring. The pattern of growth described for North Carolina M. mercenaria also is similar to that described here, with slow growth recorded from June through October (Peterson et al., 1983). The only apparent difference in the pattern of annual growth-increment formation recorded in my study and the patterns of formation noted by Clark (1979), Peterson et al. (1983), and Jones et al. (1990), is that the intra-annual period of opaque (rapid) growth-increment formation extends for a longer period of time in M. mercenaria from the Indian River than in other M. mercenaria population from the southeastern United States. Opaque growth-increment formation begins in late fall in the Indian River, as in M. mercenaria from North Carolina and Georgia, and continues until early summer, as in M. campechiensis from Cedar Key, Florida (Jones et al., 1990). The observed difference in the pattern of growth-increment formation in hard clams from the Indian River compared with the pattern observed for other southeastern locations must be interpreted with caution. Inter-annual variation in the pattern of growth-increment formation may account for this difference (Claassen, 1990) For example, the late spring and summer of 1986 were warmer than normal in the Charlotte Harbor area (NOAA, 1986), and this may have contributed to a relatively early cessation of hard clam shell growth in this region during 1986 (Jones et 79

PAGE 95

al., 1990). In contrast, the Indian River region experienced a relatively cooler than normal spring and early summer during 1988 (NOAA, 1988), and this may have contributed to a relative lengthening of the period of hard clam shell growth in the Indian River during 1988. Contemporaneous multi-year monitoring of large sample numbers collected from a variety of sites, on a monthly or more frequent basis, is needed to separate the effects of inter-annual variability from subtle but real differences in the pattern and timing of seasonal shell growth of hard clams from the southeastern United States. The relationship between temperature and growth-increment formation observed in the present study also has been suggested for other southeastern Atlantic coast populations of Mercenaria mercenaria. Translucent growth-increment formation in M. mercenaria from North carolina corresponds with the seasonal water-temperature maximum (Peterson et al., 1983). Jones et al. (1990) and Clark (1979) also suggest that temperature is related to annual growth-increment formation in M. mercenaria from Georgia. I here substantiate the relationship between temperature and annual growth-increment formation in M. mercenaria from the southeastern Atlantic coast of the United states. Opaque growth-increment formation in M. mercenaria from this region occurs when the water temperature is low and begins to increase, and translucent 80

PAGE 96

growth-increment formation occurs during summer when water temperature is warmest. In contrast, translucent growth-increment formation occurs during the period of minimum water temperature in hard clam populations from Massachusetts, Connecticut, and Rhode Island (Rhoads and Panella, 1970; Jones et al., 1989). A transition zone, from winter slow growth to summer slow growth, may occur in the region from New York to Virginia, where translucent growth-increment formation has been reported during both winter and summer within the same year (Greene, 1978; Fritz and Haven, 1983; Grizzle and Lutz, 1988). It is noteworthy that shell growth in M. mercenaria from the Indian River essentially ceases at a water temperature of approximately 30C, which agrees well with the upper temperature limit of 31C for hard clam shell growth proposed by Ansell (1968). Mercenaria mercenaria from the Indian River lagoon exhibit no clear relationship between the pattern of shell growth-increment formation and ambient salinity. Hard clam shell growth continues at salinities as low as 17.5 ppt (Castagna and Chanley, 1973). The only instance of salinity less than 17.5 ppt recorded during my study was in area E during August 1988. Because this represents the period of minimum shell growth of hard clams in the Indian River, any effect that may be related to reduced salinity would be obscured by the overriding effect of temperature. Given the generally high salinity regime of the Indian River (McCall 81

PAGE 97

et al., 1970), it is unlikely that salinity exerts a major influence on the pattern of growth-increment formation of hard clams in the Indian River. Instead, the spatially restricted, short-term decreases in salinity characteristic of the Indian River (Barile and Doehring, 1986) would be manifested in hard clam shells as short-term growth "breaks" (sensu Kennish, 1980). These growth breaks are easily distinguished from annual shell growth increments that form with a seasonally predictable pattern in response to the annual temperature cycle. Microgeographic Variation in Growth Shell growth rates of Mercenaria mercenaria in the Indian River exhibit large variations among habitats, but in every case w values from the lagoon exceed those reported for Narragansett Bay, Rhode Island (Table 10). This result supports the contention of Jones et al. (1990) that annual growth of hard clams in Florida is more rapid than that observed for northern populations of M. mercenaria. However, the wide range of w values among habitats within the lagoon indicates that comparisons of shell growth rate for this bivalve mollusc, and perhaps other bivalves with a broad geographic distribution, must include a consideration of variation in shell growth rate within localized areas. 82

PAGE 98

Table 10. Comparison of range of w values, as computed by the methods of Gallucci and Quinn (1979), compared among the Indian River lagoon, Florida (this study), various sites (excluding the Indian River lagoon) throughout Florida and south Georgia (Jones et al., 1990), Narragansett Bay, Rhode Island (Jones et al., 1989), and for Mya arenaria from various sites along the northwest Atlantic coast (Appeldoorn, 1983). Percent difference represents degree of increase from smallest to highest w value within that population. Site Range of w values % Difference Indian River, Florida 24 0-41 6 71 Florida and south Georgia 14. 8-44 4 201 Narragansett Bay, Rhode Island 11. 3-21 9 93 Northwest Atlantic (Mya arenaria) 7 7 29.4 283 83

PAGE 99

Variation in shell growth rate of hard clams over habitat in the Indian River is proportionally similar to that observed for hard clams in Narragansett Bay. The maximum difference in w values observed between any habitat combinations in the Indian River is 71%, which is quite similar to the 93% difference between minimum and maximum w values observed in Narragansett Bay (Table 10). This range of variation is much less than the 201% difference observed for hard clams sampled from a variety of estuarine and coastal systems throughout Florida (Jones et al., 1990), or the 283% difference observed for Mya arenaria (Linnaeus) in the northwest Atlantic (Appeldoorn, 1983). A larger baseline data set, including w values within and among systems as well as within and among genera, will be required before a general trend in shell growth rate variation of bivalve molluscs can be established. The observed latitudinal trend in hard clam shell growth rates within the Indian River apparently reflects a fundamental feature of the lagoon rather than a difference in the proportional occurrence of various habitats within each shellfish harvesting area. If the observed depression in shell growth rate toward the center of the study area was simply a response to fewer high growth habitats (e.g., less vegetation) in the center of the study area, then this trend should be removed if w values are standardized within a particular habitat type and then compared across shellfish 84

PAGE 100

harvesting areas. To test this, I compared shell growth rates of hard clams occurring in non-vegetated habitats within the 1 -2 m depth stratum, across shellfish harvesting areas, and found the same trend of depressed w values toward the center of the study area (Table 11) One feature of the lagoon which may affect the observed latitudinal growth pattern is salinity. Although salinity within the study area is rarely low enough to halt hard clam shell growth, the general pattern of salinity recorded in this (Figure 7B) and other studies (McCall et al., 1970) is consistent with the observed latitudinal growth trend. Salinity in areas c, F, and IR1 is relatively high and rarely declines to 20 ppt except in spatially restricted areas around flood control canals (Costa, 1986). In contrast, salinity in areas D and E is usually less than 25 ppt, and salinity in the 17.5 20 ppt range occurs over large temporal and spatial scales within these areas. Although hard clam shell growth continues at salinities as low as 17.5 ppt, shell growth rate decreases as salinity decreases towards 17.5 ppt (Castagna and Chanley, 1973). I conclude that although salinity plays a minor role in controlling the seasonal pattern of shell growth-increment formation in hard clams from the Indian River, it may have an important influence on the observed latitudinal pattern of hard clam shell growth rate in this lagoon. 85

PAGE 101

Table 11. Pattern of w values along a latitudinal transect from shellfish harvesting area C to IR1 in the Indian River lagoon, Florida. Within each area, only Mercenaria mercenaria occurring in unvegetated habitats in water depths of 1 -2 m were considered. Area c w 39 6 D 27 7 86 E F IR1 25.4 29 6 32 7

PAGE 102

In 1-2 m and > 2 m depth zones, Mercenaria mercenaria growth was more rapid in areas characterized by submerged aquatic vegetation than in unvegetated habitats. In the 0-1 m depth zone, where submerged aquatic vegetation was recorded from all stations, the density of submerged aquatic vegetation had little effect on growth rate. My results agree with similar observations by Peterson et al. (1984), who observed an increase in the rate of hard clam shell growth at a seagrass density of 40 gjm2 but no additional increase in growth rate at seagrass densities abov e this level. I observed few significant differences in growth rate among vegetated habitats; even some comparisons between habitats with vastly different vegetative abundances (e.g., 1-20% vs. 81-100% SAV) exhibited no significant difference in shell growth rate. Peterson et al. (1984) hypothesized that submerged aquatic vegetation may act as a baffle to reduce flow and increase the availability of particulate food near the bottom, where M. mercenaria feeds. This hypothesis is supported by the decrease in current shear velocity observed over an artificial seagrass bed in the laboratory (Fonseca et al. 1982). Apparently, the structural properties of submerged aquatic vegetation which support enhanced hard clam shell growth are realized at an early stage of vegetative development and do not substantially increase with an additional increase in vegetative abundance. An alternative explanation is that 87

PAGE 103

hard clam predators are more abundant within vegetated habitats (Heck and Whetstone, 1977), with the result that only the fastest growing members of a cohort survive to achieve a refuge from predation (Arnold, 1984). Selective survival of faster-growing individuals would result in an apparently more rapid shell growth rate of hard clams in vegetated habitats, when in reality, the increased shell growth rate is attributable to extrinsic predator-prey interactions rather than to intrinsic metabolic considerations. Field studies on the growth and survival of juvenile hard clams in vegetated habitats will be required to differentiate between these alternative hypotheses. The observed decrease in shell growth rate with depth in all habitats may be due to the increased concentration of silt-clay with depth. Shell growth rate of hard clams decreases with an increase in silt-clay content (Pratt, 1953; Pratt and Campbell, 1956; Greene, 1975). Personal observations during 1986-90 reveal the Indian River to be an extremely turbid lagoonal environment, particularly in deeper water during the fall, winter and spring. Laboratory studies have shown that high levels of suspended sediment load reduce feeding in juvenile and adult Mercenaria mercenaria (Bricelj and Malouf, 1984; Bricelj et al., 1984). These observations suggest that the decrease in shell growth rate with increasing depth in the Indian River may be caused 88

PAGE 104

by reduced feeding due to interference by resuspended bottom sediments. Microgeographic variation in shell growth rate of Mercenaria mercenaria in the Indian River also has important economic implications. An aquaculture industry based upon culture of M. mercenaria is being developed in the Indian River (Vaughan, 1988), and the success of this industry is largely dependent upon the rate at which successive crops can be reared and sold. Differences in growth rates among habitats within the Indian River may translate into an additional six or more months in the time required to grow a seed clam to a legally marketable size of approximately 40 rom shell height. The results of the present study should be used to help determine sites for M. mercenaria aquaculture operations in the Indian River lagoon. 89

PAGE 105

CHAPTER 4. GENOTYPE-SPECIFIC GROWTH OF HARD CLAMS (GENUS Mercenaria) IN A HYBRID ZONE: VARIATION AMONG HABITATS Introduction Shell growth rate is an important component of fitness in bivalve molluscs because the bivalve shell provides a defense against many potential predators. The mechanical force required to crush a bivalve shell is positively correlated with shell size (Blundon and Kennedy, 1982), so the shell becomes a more effective predator-defense mechanism as shell size increases (Whetstone and Eversole, 1981; Arnold, 1984; Juanes, 1992). Other components of fitness, including sedimentation-induced mortality, maturation and fecundity, and competitive interactions, may also be directly or indirectly size-dependent (Bricelj and Malouf, 1980; Peterson, 1983; Rawson and Hilbish, 1991). As a result, the fitness of individuals having rapid growth rates may be enhanced compared to that of slower-growing individuals. Hard clams of the genus Mercenaria are a commercially important and ubiquitous component of shallow-water 90

PAGE 106

estuarine and marine benthic communities along the Atlantic and Gulf of Mexico coasts of the United States. Two species, Mercenaria mercenaria and M. campechiensis, occur within that range {Abbott, 1974). Along the Atlantic coast, the two species are generally parapatrically distributed (M. mercenaria occurs inshore and M. campechiensis occurs offshore), but in the Indian River lagoon on the eastcentral coast of Florida, the two species occur sympatrically and hybridize extensively (Dillon and Manzi, 1989; Bert et al., 1993; Bert and Arnold, 1995). Previous controlled field studies of hard clam growth in the southeastern United States indicated that the mean annual shell growth rate (hereafter termed "growth") of Mercenaria campechiensis was faster than that of M. mercenaria and that the growth of their reciprocal hybrids was generally intermediate between that of the parental species (Haven and Andrews, 1956; Chestnut et al., 1956; Menzel, 1961, 1962). However, recent research on growth of natural populations of hard clams from nearshore waters of the Atlantic and Gulf of Mexico coasts of Florida suggests that comparisons of the relative growth of the two species are more complex than first perceived. In Florida, growth of M. mercenaria along the Atlanti c coast was similar among locations, but growth of M. campechiensis from the Gulf of Mexico was highly variable among locations (Jones et al., 1990). Growth of some Gulf of Mexico M. campechiensis 91

PAGE 107

populations was faster than that of Atlantic M. mercenaria populations, but growth of other Gulf of Mexico M. campechiensis populations was slow compared to that of Atlantic M. mercenaria. Whether these observed differences in growth among natural hard clam populations are genotype-specific or reflect differences in relative growth among habitats is not clear. To differentiate between those alternatives, it is necessary to compare growth among the two species and their hybrids within similar habitats. The Indian River lagoon provides an ideal location for such a comparison. Both species and their hybrids occur in the lagoon (Dillon and Manzi, 1989; Bert and Arnold, 1995). Furthermore, the benthic structure of the lagoon is composed of a variety of microhabitats (Arnold et al., 1991a), and various combinations of the two species and their hybrids occur in each of those microhabitats. This offers a unique opportunity to compare growth among genotypes within a diverse array of habitat types. In this study, I compare the growth of three genetically identified genotype classes (Mercenaria mercenaria, M. campechiensis, and hybrids of the two species) collected from a variety of locations in the Indian River lagoon. I have previously reported the habitatspecific nature of the growth of M mercenaria in the lagoon (Arnold et al., 1991a). I now extend that analysis to 92

PAGE 108

include M. campechiensis and hybrids. I compare growth between pairwise combinations of the genotype classes within and between representative habitats within the lagoon. The complex habitat-specific and genotype-specific growth relationships I observed provide a potential mechanism for selection to contribute toward the maintenance of the Mercenaria hybrid zone in the Indian River. Materials and Methods Sample Collection The study area, comprising shellfish harvesting areas c, D, E, and F of the Indian River lagoon, is bordered on the north by State Road 405 and on the south by Sebastian Inlet (Figure 10) Water depth within the study area averages 1.5 m; maximum depths of 4 m are found in the Intracoastal Waterway, which roughly bisects the Indian River throughout its length, and in areas around Melbourne and Eau Gallie (White, 1986). The Indian River lagoon is ''microtidal" (Smith, 1987); within the study area, tidal range is 5 em or less (Smith, 1993). During July and August 1986, I collected all hard clams L 40 mm shell length from each of 75 randomly allocated, 1-m2 quadrats located in each of the four shellfish harvesting areas. Prior to collecting the clams, I recorded 93

PAGE 109

Figure 10. Boundaries of shellfish harvesting areas C, D, E, and F (dashed lines) and locations of salinity monitoring stations (e) for the study of habitat-specific growth rates of hard clams (Mercenaria spp.) in the Indian River lagoon, Florida. Eau Gallie and Melbourne represent major municipalities within the Indian River region.

PAGE 110

28.30' 30" 28"00' 2s 81"00' STATE ROAD 405 ,._,. ...-eo 95 ATLANTIC OCEAN 9o3o'

PAGE 111

water depth at each station, extracted triplicate 2.5-cmdiameter x 5-cm-deep sediment cores from within each quadrat, and visually determined the presence or absence of submerged aquatic vegetation (SAV) within the quadrat. The SAV was generally composed of various proportions of the seagrasses Syringodium filiforme and Halodule wrightii or of the marine alga Caulerpa prolifera (Rice et al., 1983). In the laboratory, I determined sediment composition (percentage of gravel, sand, silt-clay, and organic matter) following the methods of Folk (1974), except that sediments were wet-sieved. I analyzed pooled sediment samples from each station for percent composition of gravel (grain size >2.00 mm), sand (grain size 0.06-2.00 mm), and silt-clay (grain size <0.06 mm). Percent organic matter was determined by ignition at 550 C for 6 h. The clams were returned alive to the laboratory, where samples of gill and mantle tissue were dissected, quick-frozen in liquid nitrogen, and stored at -80 C for protein electrophoresis. The shell valves were cleaned and stored for subsequent age determination. From October 20, 1987 through August 30, 1988, I monitored salinity (ppt) biweekly at each of eight sampling stations (two stations per shellfish harvesting area) within the study area (Figure 10). Salinity was measured 0.5 m above the sediment-water interface using a YSI Model 33 Salinity-Conductivity-Temperature meter. Salinity was 96

PAGE 112

compared among stations using the nonparametric KruskalWallis test (Sakal and Rohlf, 1981). Significant differences in mean salinity among stations were determined using the simultaneous test procedure for unplanned multiple comparisons with equal sample sizes (Sokal and Rohlf, 1981}. Genetic Analysis Each clam was assigned to a genotype class following the procedures of Bert et al. (1993). All electrophoretic assays were conducted by Biochemical Systematics Laboratory personnel at the Florida Department of Environmental Protection Marine Research Laboratory. Starch gel electrophoresis was used to assay individuals for 4 semidiagnostic protein loci (phenylalaninespecific dipeptidase-2 [DPEP-2; E.C. 3.4.11.*], glucose phosphate isomerase [GPIP; E.C. 5.3.1.9], aspartate amino transferase [AAT; E.C. 2.6.1.1], and phosphoglucomutase [PGMP-2; E.C. 5.3.2.2)}. For each locus, each allele was assigned a numerical allele diagnostic value (ADV), calculated as f m f c where f m is the mean frequency of the allele in Mercenaria mercenaria and f c is the mean frequency of the allele in M. campechiensis. The M. mercenaria and M. campechiensis allele frequencies were derived from putative pure-species samples collected from within the respective reported ranges of the two species. 97

PAGE 113

For each individual, a genetic index score for each locus was calculated as the sum of the two ADVs for the locus. The four genetic index scores for each individual from the Indian River and for each individual from the purespecies samples were then entered as response variables into a standard discriminant function analysis (DFA) in which pure-species individuals were used to define the range of genetic index scores for each species, the hybrid range was defined as the range of scores between the two species' score ranges, and the Indian River individuals were entered as unknowns. For each Indian River individual, the DFA calculated the probability of the individual's belonging to each of the three genotype classes. Each individual was assigned to the genotype class to which it had the highest probability of belonging. Because not all loci could be resolved for all individuals, and to maximize the number of individuals classified, the DFA was repeated using two three-locus combinations (DPEP-2, GPIP, AAT; DPEP-2, GPIP, PGMP-2) and the two-locus combination in which the loci had the greatest differences in allele frequencies between the two species (DPEP-2, GPIP). My nine pure-species samples were chosen by performing principal components analysis (PCA) using the correlation matrix of the diagnostic locus genetic index scores of 20 samples collected from throughout the combined ranges of the two species in the United States. I selected the samples 98

PAGE 114

that were widely separated and tightly clustered on the PC axis that best separated the two species. Thus, as recommended by Neff and Smith (1978}, I used PCA to identify the groups of populations suitable for use as defined species samples. Because it is useful for allocating new observations to defined groups in a purely descriptive sense (James and McCulloch, 1990), the DFA was an appropriate method for assigning individuals from the Indian River lagoon to genotype classes, despite its limitations (e.g., intended mainly for continuous data) Habitat-specific Growth Analysis I have verified that shell growth-increments in hard clams from the Indian River lagoon are formed annually (Arnold et al., 1991a). Each annual growth-increment is composed of one opaque band (indicative of a rapid growth period) and one translucent band (indicative of a slow growth period) (Barker, 1964; Jones et al., 1990). Because the first identifiable band proximal to the umbo was always opaque, I defined the ventral margin of each translucent band as the end of an annual growth cycle {Arnold et al., 1991a). Thus, an annual growth increment consisted of an opaque band followed by a translucent band. By measuring the linear distance from the ventral margin of each translucent band to the umbo on one radially sectioned valve 99

PAGE 115

from each specimen, I was able to collect a complete record of size-at-age for each clam. To compare growth among genotype classes and habitats, I employed the w parameter (Gallucci and Quinn, 1979}, calculated for each genotype class independently of habitat and for each genotype class within defined habitat types. The w values were determined by first fitting the pooled shell-height-at-age data for each genotype class in each habitat to a von Bertalanffy growth function (von Bertalanffy, 1938} using iterative, nonlinear, least squares regression (NLIN, SAS Version 5, 1985)]. I had previously determined that the von Bertalanffy model best represents the growth patterns of Indian River hard clams (Arnold et al., 1991a}. Thew parameter is the product of the k and components of the von Bertalannfy equation (Gallucci and Quinn, 1979} and represents growth near t0 (Jones et al., 1990} Significant differences in w values between pairwise comparisons of genotype-habitat combinations were determined by the Games and Howell method (Games and Howell, 1976; Sakal and Rohlf, 1981}. The Games and Howell method (actually, it is the Behrens-Fisher modification of the Wholly Significant Difference method (Games and Howell, 1976] but is referred to as "the Games and Howell method'' in Sakal and Rohlf (1981]} controls the Type I error rate for both individual contrasts and for the familywise comparisons at a (I used a=0.05). The Games and Howell method is valid 100

PAGE 116

at sample sizes as low as n = 6 and is the most effective method available when sample sizes are unequal and variances are heterogeneous (Garnes and Howell, 1976). I first compared growth among the three genotype classes exclusive of location or habitat. Then, to analyze habitat-specific growth, I followed my sampling strategy of latitudinally subdividing the lagoon according to the shellfish harvesting area designations. I compared growth among genotype classes within four benthic microhabitat types and compared growth among the microhabitats within genotype classes. The microhabitats were determined based upon the results of a cluster analysis (CLUSTER Ward's method, SAS version 5 [1985]) in which water depth (rn), SAV, sand ( %of sediment, arcsin square-root transformed), and organic matter ( % of sediment, arcsin square-root transformed) were used as variables. Silt-clay and gravel compositions were correlated with sand composition (Pearson's correlation coefficients= 0.85 and -0.75, respectively) and were not used in the cluster analysis. For the microhabitat classification, all clams collected from a sampling station were classified as occupying the microhabitat into which the sampling station had been grouped. To further analyze genotype-and habitat-specific growth within each shellfish harvesting area, I performed separate sets of analyses using only depth 1.5rn, > 1.5rn) and only vegetation (SAV present, SAV absent) as habitat 101

PAGE 117

categories, following the procedure used in the analyses involving the cluster-defined microhabitats. Because the Garnes and Howell procedure is not valid for sample sizes < 6, I performed pairwise comparisons of samples only when the sample sizes of both members of the pair were 6. Results Salinity regimes at the monitoring stations differed considerably. In general, salinity was relatively high and stable in shellfish harvesting area c, low and stable in areas D and E, and highly variable in area F (Figure 11). Mean annual salinity differed significantly among stations (p < 0.001). Both stations in area C grouped together, but they were separate from the stations in areas D and E; all stations in areas D and E grouped together {Table 12). The two stations in area F did not cluster together; the northern station grouped with areas D and E, but the southern station grouped with area c. In addition, area F was characterized by substantial periodic fluctuations in salinity {Figure 11), principally due to aperiodic intrusions of fresh water from local flood-control canals {Costa, 1986). Because growth in at least one genotype class (Mercenaria mercenaria) is related to salinity {Arnold et al., 1991a) and because mean salinity in area C differs significantly from that in areas D and E, I retained area c 102

PAGE 118

Table 12. Results of nonparametric multiple range test (Sokal and Rohlf, 1981) comparing mean annual salinity among eight sampling stations in the Indian River lagoon, Florida. First letter of two-letter station abbreviation denotes shellfish harvesting area (C, D, E, or F); second letter denotes station location within each area (north (N] or south (S]) (see Figure 10). Underlining connects stations for which mean annual salinity did not differ significantly. Station CN FS cs DN OS FN ES EN Mean salinity (o/oo) 28 27 26 23 22 21 20 20 103

PAGE 119

Figure 11. Variation in salinity (ofoo) in each of four shellfish harvesting areas (C, D, E, F) of the Indian River lagoon, Florida. Locations of each of the areas and sampling stations are shown in Figure 10.

PAGE 120

1987 1988 Date 105 -----C North C South ---+D North D South ---+E North E South ---+F North F South .......

PAGE 121

(hereafter referred to as Subdivision C) as a separate entity for the analysis of genotype-specific growth among habitats but combined areas D and E (Subdivision D/E) to facilitate data interpretation and to increase sample sizes. I also retained area F (Subdivision F) as a separate entity of the study area because of its unique salinity properties. The patterns of salinity variation I observed in the Indian River are consistent with reported long-term trends (McCall et al., 1970; Costa, 1986}. Genotype classes were not distributed equally among the three subdivisions (Table 13). In Subdivision c, although Mercenaria mercenaria predominated, both hybrids and M. campechiensis were also relatively common. In Subdivisions D/E and F, M mercenaria comprised approximately three quarters, and hybrids comprised approximately one quarter, of the hard clams I collected; M. campechiensis were extremely rare. The paucity of M. campechiensis in Subdivisions D/E and F made it necessary to limit my comparisons of genotype-specific growth to the M. mercenaria and hybrid genotype classes in most analyses involving those subdivisions. The clustering method used to group benthic habitats based on depth, sediment composition, and vegetation cover requires that selection of the number of clusters produced be done a priori. I elected to group benthic habitats into four clusters (microhabitats) for two reasons: (1) problems 106

PAGE 122

Table 13. Frequency (%) of occurrence of three genotype classes (Mercenaria mercenaria, M. campechiensis, hybrids) in three subdivisions (C, D/E, F) of the Indian River lagoon, Florida. N = total number of hard clams collected in each subdivision. Subdivision N Genotype class M. mercenaria M campechiensis Hybrid c 108 44. 4 18.5 37. 0 DIE 395 71. 9 1 7 26.3 F 111 76. 6 0 0 23. 4 All combined 614 67. 9 4 4 27. 7 107

PAGE 123

Table 14. Characteristics of four microhabitat types defined using cluster analysis performed on quantitative estimates of water depth, % sand and % organics in the substrate, and presence (1) or absence (0) of submerged aquatic vegetation (SAV). N = number of 1 m2 quadrats sampled; SD = standard deviation; na = not applicable) Microhabitat type (N) 1 (56) 2 (82) 3 (115) 4 (47) Mean so Range Mean so Range Mean so Range Mean so Range Depth (m) 0 8 0 2 0 3 1 2 1 2 0 3 0 .5-1. 5 2.4 0 6 1 7-4 3 1 9 0 5 1. 4-3 7 % Sand 93.2 6.4 55 9-98 2 91. 6 9 6 46 5-98 4 81. 1 17 .7 2 5-97 7 85 3 11. 1 50 4-97 3 108 % Organics 1 0 0 6 0 0-3 2 1 2 1.4 0 0-9 3 2 4 2 8 0 0-22. 0 1 7 1 1 0 5 5.6 SAV 1 na na 0 na na 0 na na 1 na na

PAGE 124

Table 15. Frequency (%) of occurrence of microhabitat types (defined in Table 14) within subdivisions of the Indian River lagoon, Florida. N = the total number of quadrats in each subdivision. Subdivision N Microhabitat 1 2 3 4 c 75 40 0 13 3 20. 0 26 7 DIE 150 10.0 24 0 48 0 18 .0 F 75 14 7 48.0 37 3 0 .0 All 300 18 7 27 3 38 3 15 7 109

PAGE 125

with small sample sizes of some genotype classes within each cluster were minimized, and (2) distinct habitat differences among clusters were maximized. The microhabitats I defined can be characterized as follows: Microhabitat 1-shallow water, high percentage of sand and low percentage of organics in the sediment, vegetated; Microhabitat 2-shallow water, high percentage of sand, low percentage of organics, unvegetated; Microhabitat 3-deep water, low percentage of sand, high percentage of organics, unvegetated; and Microhabitat 4-deep water, low percentage of sand, high percentage of organics, vegetated (Table 14). The microhabitats were proportioned differentially among subdivisions (Table 15). Based upon an expected ratio (for the null hypothesis of equal distribution among subdivisions) of 1:1:1:1 for the four microhabitat types, Microhabitats 1 and 4 were more abundant than e xpected and Microhabitats 2 and 3 were less abundant than expecte d in Subdivision C (goodness-of-fit G-test; G = 11.5, df = 3, p < 0.01). In contrast, in Subdivisions D/E and F, Microhabitats 1 and 4 were less abundant than expected and Microhabitats 2 and 3 were more abundant than expected (Subdivision DjE: G = 45.8, df = 3, p < 0.001; Subdivision F: G = 14.5, df = 2, p < 0.01). Microhabitat 4 was completely absent from Subdivision F, but all other microhabitats occurred in all subdivisions. 110

PAGE 126

Overall, the growth of each hard clam genotype class differed significantly from that of the other two genotype classes (Figure 12). The growth of Mercenaria campechiensis was notably faster than that of either M. mercenaria or hybrids, whereas growth of M. mercenaria was only slightly faster than that of hybrids. Among subdivisions, both pure species grew significantly faster in Subdivision c than in Subdivisions D/E, and M. mercenaria grew significantly faster in Subdivision C than in Subdivision F; M. mercenaria also grew significantly faster in Subdivision F than in Subdivision D/E. There was no significant difference in the growth of hybrids between Subdivisions c and D/E, but hybrids in those subdivisions grew significantly slower than hybrids in Subdivision F. Within subdivisions, the pure species grew at similar rates in Subdivision c; both pure species grew significantly faster than hybrids in that subdivision. No difference in growth was detected between M. campechiensis and either M. mercenaria or hybrids in Subdivision D/E, but M. mercenaria grew significantly faster than hybrids in that subdivision. Finally, there was no significant difference in growth between M. mercenaria and hybrids in Subdivision F. For the cluster-defined microhabitats, I detected significant differences in hard clam growth within genotype classes among microhabitats and between genotype classes within microhabitats only in Subdivisions D/E and F (Figure 111

PAGE 127

Figure 12. Mean annual growth rate, as defined by the w parameter, for three genotype classes of hard clams from the Indian River lagoon, Florida. Values of w are plotted, within genotype classes, for the entire study area (ALL) and for each of the three subdivisions (C, D/E, F) of the study area. Vertical lines show one standard deviation from the mean. Numbers of individuals are in parentheses.

PAGE 128

90 --roo----------------, Q) ::J 80 70 -ro 6o > rn 50 C) Q) E 40 0 30 Q) 20 10 V//ZI M mercenaria I I M campechiensis t> Hybrids -0 N --(!) N .._... ALL c D/E F Subdivision 113

PAGE 129

Figure 13. Variation among cluster-analysis-defined microhabitats (see Table 14) in mean annual growth rate, as defined by the w parameter, of three genotype classes of hard clams from three subdivisions of the Indian River lagoon, Florida (defined in Results). Vertical lines show one standard deviation from the mean. Numbers of individuals are in parentheses. Groupings below histograms joined by common underlining show sets of microhabitats (upper table) or genotype classes (lower table) in which hard clam mean annual growth rates did not differ significantly for the listed genotype class or microhabitat, respectively. Genotype classes and microhabitats are arrayed, left to right, from those in which mean annual growth rate in the designated hard clam genotype class was fastest to those in which it was slowest.

PAGE 130

90 90 90 80 Q) 70 :::J M I c::::::J M campechlens i a 1!188111 Hybrids l D 80 Q) 70 :::J I c::::J M eam pechlensls 11!!!11!1 Hybrid s 80 Q) 70 :::J l r2lZ! M c::::J M llilillll Hybrids -;;; 60 > !tl 50 Cl Q) E 40 0 c: 30 !tl Q) :::iE 20 c; !::! :: N :!. --;;; 60 > !tl 50 Cl Q) E 40 0 c: 30 !tl Q) :::iE 20 --ill 60 !tl 50 Cl Q) E 40 0 c: 30 !tl Q) :::iE 20 c: e T e 10 10 10 0 0 0 2 3 2 3 2 3 M i crohabitat Type Microhabitat Type Microhab itat Type SUBDI V ISION C SUBDIVI S ION D / E SUBDIVISION F G enotype c l ass Mic r o habitat Gen o t y pe c l ass Mi cro habitat Genotype c l ass Mic rohabita t 4 M crcenaria mrwwia 4 2 2 3 Hy b rid 4 Hybrid 2 4 H y brid 2 Merc enaria campcc hjc n si s 4 Mcrcenaria c ampe c b jcns j s M.c.rc.c.naria c ampc c bjcruj s Mi c rohabitat Ge notype c l ass Mi c rohabitat Gen o type c lass Mi c rohabitat Ge not ype class Mm H y 2 2 H y Mm 2 Mm Hy 3 M e Mm H y 3 Mm H y 3 Mm Hy 4 M e H y Mm 4 Mm H y 4

PAGE 131

Figure 14. Variation among depth categories in mean annual growth rate, as defined by the w parameter, of three genotype classes of hard clams from three subdivisions of the Indian River lagoon, Florida (defined in Results). Vertical lines show one standard deviation from the mean. Numbers of individuals are in parentheses. Groupings below histograms joined by common underlining show sets of depth categories (upper table) or genotype classes (lower table) in which hard clam mean annual growth rates did not differ significantly for the listed genotype class or depth category, r espectively. Genotype classes and depth categories are arrayed, left to right, from those in which mean annual growth rate in the designated hard clam genotype class was fastest to those in which it was slowest.

PAGE 132

..... ..... ....:! 90 80 EZ;2l M mercenaria c:::::J M campechiensis 70 18!!!!!!1 Hybrids Q) :::J "' iii > 60 CIJ 50 Cl Q) E 40 0 c 30 CIJ Q) ::!! 20 10 0 LE 1.5 m GT 1 5 m Depth (m) SUBDIVISION C Genotype class Depth (m) Mcrcenarja mercenarja > 1.5 Hybrid > 1.5 M e rcenaria campecbiensis Depth (m) Genotype class 1 5 Mm Hy > 1.5 Me Mm Hy 90 90 r22Zll M mercenoria I 80] I EZ;2l M mercenoria c:::::J M campechiensis c:::::J M campechiensis I5Z5Z!Z!il Hybrids 70 IZ5ZIZ!ZI Hybrids Q) 80 70 Q) :::J :::J iii 60 > Iii 60 > CIJ 50 Cl Q) 401 -E 0 1111 c _I_ CIJ 30 Ill 50 Cl Q) E 40 0 c 30 CIJ Q) Q) ::!! 20 ::!! 20 10 10 0 0 LE 1 5 m GT 1 5 m LE 1 5 m GT 1 5 m Depth (m) Depth (m) SUBDIVISION D/E SUBDIVISION F Genotype class Depth (m) Genotype class Depth (m) Merceoada mercenarja > 1.5 Mercenarja merceoarja > 1.5 Hybrid > 1.5 Hybrid >1.5 Mercenarja campechjensjs Merceoaria campecbjensjs Depth (m) Genotype class Depth (m) Genotype class Hy Mm Hy Mm > 1.5 Mm Me Hy > 1.5 Mm Hy

PAGE 133

Figure 15. Variation among submerged aquatic vegetation (SAV) categories in mean annual growth rate, as defined by the w parameter, of three genotype classes of hard clams from three subdivisions of the Indian River lagoon, Florida (defined in Results). Vertical lines show one standard deviation from the mean. Numbers of individuals are in parentheses. Groupings below histograms joined by common underlining show sets of SAV categories (upper table) or genotype classes (lower table) in which hard clam mean annual growth rates did not differ significantly for the listed genotype class or SAV category, respectively. Genotype classes and SAV categories are arrayed, left to right, from those in which mean annual growth rate in the designated hard clam genotype class was fastest to those in which it was slowest.

PAGE 134

90 90 90 80 I'22Zl M mercenaria 80 M mercenana 80 i 111Z'21 M mercenaria c:::::J M campechlensis > N > 1 M Ill 50 Ill 50 Ill 50 01 01 01 Q) Q) Q) E 40 E 40 E 40 0 0 0 c 30 c 30 c 30 Ill Ill Ill Q) Q) Q) :::i! 20 :::i! 20 :::i! 20 10 10 10 0 0 0 Absent Present Absent Present Absent Present SAV SAV SAV ...... SUBDIVISION C SUBDIVISION D/E SUBDIVISION F ...... \D Genotype class Vegetation Genotype class Vegetation Genotype class Vegetation Mercenaria mercenaria Present Absent Merceoaria mercenarja Present Absent Mercenaria mercenaria Present Absent Hybrid Absent Present Hybrid Present Absent Hybrid Mercenaria camoechiensis Present Absent Mercenaria camoechiensis Mercenaria camoechiensis Vegetation Genotype class Vegetation Genotype class Vegetation Genotype class Absent Me Mm Hy Absent Mm Hy Absent Mm Hy Present Me Mm Hy Present Mm Hy Present

PAGE 135

13). In those subdivisions, growth of Mercenaria mercenaria and of hybrids was generally fastest in Microhabitat 1 and, when it occurred, Microhabitat 4. Growth of both genotype classes was slowest in Microhabitat 3. Within microhabitats, M. campechiensis could be compared with M. mercenaria and hybrids only in Subdivision c. There, growth of M. campechiensis was not significantly faster than that of either M. mercenaria or hybrids, although a trend of faster growth by M. campechiensis was observed. In Subdivisions D/E and F, neither M. mercenaria nor hybrids had a clear growth advantage. Growth of M. mercenaria significantly exceeded that of hybrids in Microhabitat 3, the microhabitat in which overall hard clam growth was slowest, in both subdivisions. However, in one instance (Microhabitat 2 in Subdivision D/E) the reverse situation occurred; growth of hybrids was faster than that of M. mercenaria. No significant differences in growth occurred between M. mercenaria and hybrids in the other three microhabitat comparisons made collectively in Subdivisions D/E and F. Variation in growth both within genotype classes between water-depth categories and among genotype classes within water-depth categories was also more pronounced in Subdivisions D/E and F than in Subdivision C (Figure 14). In Subdivision c, the only significant difference in growth was in deep water, where growth of Mercenaria campechiensis 120

PAGE 136

was significantly faster than that of the other genotype classes. In other subdivisions, growth of M. campechiensis could only be compared to the other genotype classes in deep-water habitats in Subdivision D/E; there, growth of M campechiensis was intermediate between that of M mercenaria and hybrids. In both Subdivision D/E and Subdivision F, growth of both M. mercenaria and hybrids was faster in shallow water than in deep water; the growth of hybrids significantly exceeded that of M. mercenaria in shallow water but was significantly less than that of M mercenaria in deep water. Growth of all genotype classes was as fast or faster in vegetated habitats than in unvegetated habitats (Figure 15) and growth of hybrids was usually significantly slower than that of both Mercenaria mercenaria and M. campechiensis in both vegetated and unvegetated habitats. Only in Subdivision c were numbers of clams in the two pure-species genotype classes sufficient for comparing growth. There, growth of M. campechiensis was not significantly faster than that of M. mercenaria in either vegetated or unvegetated habitats. Discussion The classical paradigm that describes genotype-specific growth of hard clams as being fastest in Mercenaria 121

PAGE 137

campechiensis, intermediate in hybrids, and slowest in M. mercenaria (Haven and Andrews, 1956; Chestnut et al., 1956; 1961, 1962, 1989) does not apply in the Indian River lagoon. Instead, relative growth of the two Mercenaria species and their hybrids is habitat-dependent. For example, the growth of M. campechiensis exceeds that of M. mercenaria and of hybrids in deep-water habitats in Subdivision c, but aside from that notable exception, little habitat-specific differentiation in growth occurs among genotype classes in that subdivision. In Subdivisions D/E and F, growth of hybrids is faster than that of M. mercenaria in shallow-water habitats but slower than that of M. mercenaria in deep-water habitats. An overall habitat-independent growth advantage for Mercenaria campechiensis was demonstrated in the present study, but this occurred principally because of the overwhelming number of M. campechiensis individuals that were collected from the subdivision and habitat in which that species grew fastest. The hierarchy of overall growth of the two species and their hybrids may have been quite different if a high percentage of M. campechiensis had been collected from Subdivisions D/E or F. That contention is supported by the lack of significant difference in habitatindependent growth between M. campechiensis and M. mercenaria in Subdivision c. Similarly, the overall growth advantage that M. mercenaria has over hybrids may be due to 122

PAGE 138

the habitat-specific proportions of individuals collected in those genotype classes. However, my detailed analyses are consistent with the idea that, overall, growth in M. mercenaria is more rapid than in hybrids. Nevertheless, in certain locations (Subdivisions D/E and F), hybrids had a growth advantage over M. mercenaria in shallow water. Two environmental factors that probably contributed to the location-specific and habitat-specific variation in growth among hard clam genotype classes in the Indian River are suspended sediment load (which is related to water depth and the presence or absence of submerged aquatic vegetation) and salinity. Both Mercenaria mercenaria and hybrids grew faster in shallow water than in deep water and grew faster in vegetated habitats than in unvegetated habitats. Silt-clay and combustible organics, the most easily resuspended components of the sediment (Levinton, 1982), comprise a greater proportion of the sediment in deep-water habitats than in shallow-water habitats in the lagoon. Additionally, growth of M. mercenaria is slower in unvegetated habitats than in vegetated habitats in North Carolina (Peterson et al., 1984); that growth difference has been attributed at least partially to greater sediment stability in vegetated habitats (Irlandi and Peterson, 1991). In M. mercenaria, the growth rate of soft tissues decreases with increasing concentration of suspended sediments, presumably because 123

PAGE 139

algal ingestion rate is reduced when suspended sediment load is high (Bricelj and Malouf, 1984; Bricelj et al., 1984). The effects of sediment load on the growth of M. campechiensis and hybrids have not been identified, but in the Indian River, the faster growth of M. campechiensis in deep-water habitats, which have relatively high organic and silt-clay content, suggests that this species is more tolerant of suspended sediment than is M. mercenaria. Hybrids may be most sensitive to the effects of sediment load or the absence of vegetation; the growth advantage in this genotype class was associated with both shallow water and the presence of submerged aquatic vegetation. Growth of hybrids was very slow in deep-water habitats, particularly when vegetation was absent. Growth of hard clams is also influenced by salinity. In field studies, variation in growth of Mercenaria mercenaria has been attributed to differences in the ambient salinity regime (Haven and Loesch, 1973; Arnold et al., 1991a); slow growth was related to low or fluctuating salinities. In a laboratory study, M. campechiensis grew very rapidly in high-salinity water (Figure 1 in Sunderlin et al., 1975), whereas growth of M. mercenaria was comparatively slow and mortality was high. High salinity may be a particularly important factor influencing growth in M. campechiensis; this species tends to occupy high-salinity coastal and open-ocean habitats rather than the euryhaline 124

PAGE 140

habitats occupied by M. mercenaria (Menzel, 1989). Mean salinity and salinity variation change substantially over the latitudinal extent of my study area, and these regional variations apparently influence mesoscale growth patterns in the Indian River hard clams. In my study, pure-species growth rates were higher in the relatively higher and more stable salinities of Subdivision C than in the subdivisions with lower {Subdivision D/E) or fluctuating (Subdivision F) salinities. The faster growth of hybrids in Subdivision F than in Subdivisions c and D/E may be a reflection of their mixed genome; hybrids may not be fully adapted to either relatively high or relatively low salinities. However, among genotype classes, habitat-specific differences in growth were clearly more pronounced in the subdivisions with relatively low or fluctuating salinities. It may be that when salinities are not optimal for growth in hard clams, other habitat components, including those I assessed, become important genotype-specific determinants of growth. Habitat-specific differences in growth among genotypes may contribute to the maintenance of the hard clam hybrid zone in the Indian River lagoon. In the Indian River hybrid zone, selection is complex. Selection acts differentially on specific genotypes (Bert and Arnold, 1995), and selection against hybrids does occur (Bert et al., 1993; Bert and Arnold, 1995). Lagoonal populations of pure Mercenaria mercenaria flank the hybrid zone (Bert and Arnold, in 125

PAGE 141

prep.), and influx of M. campechiensis larvae from nearby Atlantic continental shelf populations into the Indian River lagoon seems to be infrequent and aperiodic (Bert and Arnold, 1995). Therefore, the presence of considerable numbers of M. campechiensis in the deep waters of Subdivision C may be attributable to a selective advantage imparted to that species in that habitat; there, growth of M. campechiensis is comparatively very fast. Similarly, the presence of a significant and constant fraction of hybrid forms throughout Subdivisions D/E and F (Bert and Arnold, 1995) may in part be due to the more rapid growth of hybrids compared to the growth of the more predominant M. mercenaria in the shallow-water habitats of those subdivisions. The variation in growth among hard clam genotype classes in different habitats in the Indian River lagoon has important economic implications. A hard clam aquaculture industry has recently developed in the Indian River lagoon (Vaughan, 1988). Menzel (1989) suggested that h ybrids o f Mercenaria campechiensis and M. mercenaria would be suitable for aquaculture operations because the h ybrids hav e the fast-growth characteristic of M. campechiensi s and the longer shelf-life characteristic of M. mercenaria. The results of my study indicate that both selecting a suitable site for hard clam aquaculture in the lagoon (Arnold et al., 1991b) and carefully selecting a suitably adapted broodstock may be critical to the success of those ventures. 126

PAGE 142

CHAPTER 5. SUMMARY 1) Hard clams of the genus Mercenaria are an important constituent of the infuana in the Indian River lagoon, Florida. 2) The periodicity of annual growth-band formation is consistent among the three genotype classes that occur in the lagoon, with translucent band formation occuring during summer and fall and opaque band formation occuring during winter and spring. 3) Within the species Mercenaria mercenaria, rate of shell growth varies significantly among habitats, with faster growth in shallow-water than deep-water habitats and in vegetated than in unvegetated habitats. 4) The annual rate of shell growth differs significantly among hard clam genotype classes. Mercenaria campechiensis is restricted to deep-water habitats in the northern portion of the lagoon, but outgrows its conspecifics in that habitat. Mercenaria mercenaria and hybrid forms co-occur, but each maintains a growth advantage in certain habitats. 5) The genotype-and habitat-specific patterns of growth that I describe may explain why the hard clam hybrid 127

PAGE 143

zone in the Indian River lagoon is maintained over time in spite of apparently substantial gene flow among each of the three genotype classes. 6) My results can be used by the Indian River hard clam aquaculture industry to exploit favorable habitats for maximum crop yield. 128

PAGE 144

LIST OF REFERENCES Abbott, R.T., 1974. American Seashells. Van Nostrand Reinhold, New York, 2nd edition, 663 pp. Almada-Villela, P.C., J. Davenport & L.D. Gruffydd, 1982. The effects of temperature on the shell growth of young Mytilus edulis L. J. Exp. Mar. Biol. Ecol. 59: 275-288. Anonymous, 1991-1995. Fisheries of the u.s., 1990-1994. USDOC, NOAA, NMFS, Current Fisheries Statistics No. 9000-9400. Ansell, A.D., 1968. The rate of growth of the hard clam Mercenaria mercenaria (L) throughout the geographical range. J. Cons. perm. int. Explor. Mer 31: 364-409. Appeldoorn, R.S., 1983. Variation in the growth rate of Mya arenaria and its relationship to the environment as analyzed through principal components analysis and the w parameter of the von Bertalanffy equation. Fish. Bull. 81: 75-84. Arnold, w.s., 1984. The effects of prey size, predator size, and sediment composition on the rate of predation of the blue crab, Callinectes sapidus Rathbun, on the hard clam, Mercenaria mercenaria (Linne). J. Exp. Mar. Bioi. Ecol. 80: 207-219. Arnold, w.s., T.M. Bert, D.C. Marelli, H. Cruz-Lopez & P.A. Gill, 1996. Genotype-specific growth of hard clams (genus Mercenaria) in a hybrid zone: variation among habitats. Mar. Bioi. 125: 129-139. Arnold, W.S., D.C. Marelli, T.M. Bert, D.S. Jones & I.R. Quitmyer, 1991a. Habitat-specific growth of hard clams Mercenaria mercenaria (L.) from the Indian River, Florida. J. Exp. Mar. Biol. Ecol. 147: 245-265. 129

PAGE 145

Arnold, w.s., D.C. Marelli, c. Lund, 1991b. Suitability of the southern Indian River lagoon for hard clam (Mercenaria spp.) culture. Florida Department of Agriculture and Consumer Services, Aquaculture Market Development Aid Program 1989-1990: 11-73 Aten, L.E. 1981. Determining seasonality of Rangia cuneata from Gulf Coast shell middens. Bull. Texas Archaeol. Soc. 52: 179-200. Bachelet, G., 1980. Growth and recruitment of the tellinid bivalve Macoma balthica at the southern limit of its geographical distribution, the Gironde Estuary (SW France). Mar. Biol. 59: 105-117. Barile, D. & F. Doehring, 1986. Rainfall. In, Report o n rainfall event of September and October 1985 and the impact of storm discharge on salinity and the clam population (Mercenaria mercenaria) of the Indian River lagoon, edited by D.D. Barile and W. Rathjen, Marine Resources Council, Melbourne, Florida, pp. 8-24. Barker, R.M., 1964. Microtextural variation in pelecypod shells. Malacologia 2: 69-86. Bert, T.M., 1985. Geographic variation, population biology, and hybridization in Menippe mercenaria and evolution in the genus Menippe in the southwestern North Atlantic Ocean. Ph.D. dissertation, Yale University, New Haven, Connecticut, 306 pp. Bert, T.M. 1986. Speciation in western Atlantic stone crabs (genus Menippe) : the role of geological processes and climatic events in the formation and distribution of species. Mar. Biol. 93: 157-170. Bert, T.M. & W.S. Arnold, 1995. An empirical test of predictions of two competing models for the maintenance and fate of hybrid zones: both models are supported in a hard clam hybrid zone. Evolution 49: 276-289. Bert, T.M., w.s. Arnold & H. Cruz-Lopez, 1988. Species distributions in hard clams (genus Mercenaria) : complex patterns reflect natural processes and the influence of man. J. Shellfish Res. 7 : 547. Bert, T.M., D.M. Hesselman, W.S. Arnold, W.S. Moore, H. cruz-Lopez & D.C. Marelli, 1993. High frequency of gonadal neoplasia in a hard clam (Mercenaria) hybrid zone. Mar. Biol. 117: 97-104. 130

PAGE 146

Beukema, J.J., E. Knol & G.C. Cadee, 1985. Effects of temperature on the length of the annual growing season in the tellinid bivalve Macoma balthica (L.) living on tidal flats in the Dutch Wadden Sea. J. Exp. Mar. Biol. Ecol. 90: 129-144. Blundon, J.A. & V.S. Kennedy, 1982. Mechanical and behavioral aspects of blue crab, Callinectes sapidus (Rathbun), predation on Chesapeake Bay bivalves. J. Exp. Mar. Biol. Ecol. 65: 47-65. Bricelj, V.M. & R.E. Malouf, 1980. Aspects of reproduction of hard clams (Mercenaria mercenaria) in Great South Bay, New York. Proc. Natn. Shellfish. Assoc. 70: 216-229. Bricelj, V.M. & R.E. Malouf, 1984. Influence of algal and suspended sediment concentrations on the feeding physiology of the hard clam Mercenaria mercenaria. Mar. Biol. 84: 155-165. Bricelj, V.M., R.E. Malouf & C. deQuillfeldt, 1984. Growth of juvenile Mercenaria mercenaria and the effect of resuspended bottom sediments. Mar. Biol. 84: 167-173. Brousseau, D.J. 1979. Analysis of growth rate in Mya arenaria using the von Bertalanffy equation. Mar. Biol. 51: 221-227. Brown, B.L., 1989. Population variation in the mitochondrial DNA of two marine organisms: the hard shell clam Mercenaria spp. and the killifish, Fundulus heteroclitus. Ph.D. Dissertation, Old Dominion University, Norfolk, Virginia, 133 pp. Brown, J.R., 1988. Multivariate analyses of the role of environmental factors in seasonal and site-related growth variation in the Pacific oyster Crassostrea gigas. Mar. Ecol. Prog. Ser. 45: 225-236. Castagna, M. & P. Chanley, 1973. Salinity tolerance of some marine bivalves from inshore and estuarine environments in Virginia waters on the western mid-Atlantic coast. Malacologia 12: 47-96. Chave, K.E., 1954. Aspects of the biogeochemistry of magnesium. J. Geol. 62: 266-283. Chestnut, A.F., W.E. Fahy & H.J. Porter, 1956. Growth of young Venus mercenaria, Venus campechiensis, and their hybrids. Proc. natn. Shellfish. Assoc. 47: 50-56. 131

PAGE 147

Claasen, c., 1990. Investigation of monthly growth in shellfish for application to archaeology. Final Report, National Science Foundation, Grant No. BNS8507714, 129 pp. Clark, G.R., II, 1979. Seasonal growth variations in the shells of recent and prehistoric specimens of Mercenaria mercenaria from St. Catherines Island, Georgia. Anthropol. Pap. Am. Mus. Nat. Hist. 56: 161-172. Costa, S.L., 1986. Salinity. In, Report on rainfall event of September and October 1985 and the impact of storm discharge on salinity and the clam population (Mercenaria mercenaria) of the Indian River lagoon, edited by D.O. Barile and W. Rathjen, Marine Resources Council, Melbourne, Florida, pp. 47-123. Costa, S.L., 1987. flushing study. R/IRL-3, 177 pp. Indian River lagoon circulation and Report to Florida Sea Grant College, Coutts, P.J.F., 1970. Bivalve-growth patterning as a method for seasonal dating in archaeology. Nature 226: 874. Craig, M .A., T.J. Bright & S.R. Gittings, 1988. Growth of Mercenaria mercenaria and Mercenaria mercenaria texana seed clams planted in two Texas bays. Aquaculture 71: 193-207. Crenshaw, J.W., Jr., P.B. Heffernan & R.L. Walker, 1996. Effect of growout density on heritability of growth rate in the northern quahog, Mercenaria mercenaria (Linnaeus, 1758). J. Shellfish Res. 15: 341-344. Cunliffe, J.E., 1974. Description, interpretation, and preservation of growth increment patterns in shells of cenozoic bivalves. Ph.D. Dissertation, Rutgers University, New Jersey, 169 pp. Dalton, R.C., 1977. The reproductive cycles of the northern and southern quahogs, Mercenaria mercenaria (L.) and M. campechiensis (Gmelin), and their hybrids, with a note on their growth. M.S. Thesis, Florida State University, Tallahassee, Florida, 89 pp. Dillon, R.T., Jr. & J.J. Manzi, 1989. Genetics and shell morphology in a hybrid zone between the hard clams Mercenaria mercenaria and M. campechiensis. Mar. Biol. 100: 217-222. 132

PAGE 148

Folk, R.L., 1974. Petrology of Sedimentary Rocks. Hemphill Publishing Company, Austin, Texas, 184 pp. Fonseca, M.S., J.S. Fisher, J.C. Zieman & G.W. Thayer, 1982. Influence of the seagrass, Zostera marina L., on current flow. Estuar. Coastal & Shelf Sci. 15: 351-364. Fritz, L.W. & D.S. Haven, 1983. Hard clam, Mercenaria mercenaria: shell growth patterns in Chesapeake Bay. Fish. Bull. 81: 697-779. Gallucci, V.F. & T.J. Quinn II, 1979. Reparameterizing, fitting, and testing a simple growth model. Trans. Am. Fish. Soc. 108: 14-25. Games, P.A. & J.F. Howell, 1976. Pairwise multiple comparison procedures with unequal N's andjor variances: a Monte Carlo study. J. Educational Stat. 1: 113-125. Gould, S.J., 1979. Time's vastness. Nat. History 88: 8-27. Greene, G.T., 1975. Incremental shell growth patterns as affected by environment in Mercenaria mercenaria. B.Sc. Thesis, Princeton University, Princeton, New Jersey, 77 pp. Greene, G.T., 1978. Population structure, growth and mortality of hard clams at selected locations in Great South Bay, New York. M.S. Thesis, SUNY Stony Brook, Stony Brook, New York, 199 pp. Grizzle, R.E. & R.A. Lutz, 1988. Descriptions of macroscopic banding patterns in sectioned polished shells of Mercenaria mercenaria from southern New Jersey. J. Shellfish Res. 7: 367-370. Grizzle, R.E. & P.J. Morin, 1989. Effect of tidal currents, seston, and bottom sediments on growth of Mercenaria mercenaria from southern New Jersey. J. Shellfish Res. 7: 367-370. Hadley, N.H., R.T. Dillon, Jr. & J.J. Manzi, 1991. Realized heritability of growth rate in the hard clam Mercenaria mercenaria. Aquaculture 93: 109-119. Haven, D. & J.D. Andrews, 1956. Survival and growth of Venus mercenaria, Venus campechiensis, and their hybrids in suspended trays and on natural bottoms. Proc. natn. Shellfish. Assoc. 47: 43-49. 133

PAGE 149

Haven, D.S. & J.G. Loesch, 1973. An investigation into commercial aspects of the hard clam fishery and development of commercial gear for the harvest of molluscs. Ann. Contract Rep., Contract No. 3-124 R, Virginia Institute of Marine Science, Gloucester Point, Virginia, 91 pp. Heck, K.L., Jr. & G.S. Whetstone, 1977. Habitat complexity and invertebrate species richness and abundance in tropical seagrass meadows. J. Biogeogr. 4: 135-142. Hesselman, D.M., B.J. Barber, & N.J. Blake, 1989. The reproductive cycle of adult hard clams, Mercenaria spp., in the Indian River lagoon, Florida. J. Shellfish Res. 8: 43-49. Holiman, S.G., 1995. Hard clam (Mercenaria mercenaria) aquaculture under production uncertainty. Paper presented at 15th Milford Aquaculture Seminar, Milford, Connecticut, 20 pp. Hudson, J.H., E.Shinn, R. Halley & B. Lidz, 1976. Sclerochronology: a new tool for interpreting past environments. Geology 4: 361-364. Irlandi, E.A. & C.H. Peterson, 1991. Modification of animal habitat by large plants: mechanisms by which seagrasses influence clam growth. Oecologia 87: 307-318. James, F.C. & C.E. McCulloch, 1990. Multivariate analysis in ecology and systematics: panacea or Pandora's box? Ann. Rev. Ecol. Syst. 30: 291-308. Jones, D.S., 1981. Repeating layers in the molluscan shell are not always periodic. J. Paleontol. 55: 1076-1082. Jones, D.S., 1983. Sclerochronology: reading the record of the molluscan shell. Am. Sci. 71: 384-391. Jones, D.S., M.A. Arthur & D.J. Allard, 1989. Sclerochronological records of temperature and growth from shells of Mercenaria mercenaria from Narragansett Bay, Rhode Island. Mar. Biol. 102: 225-234. Jones, D.S., I.R. Quitmyer, w.s. Arnold & D.C. Marelli, 1990. Annual shell banding, age, and growth rate of hard clams (Mercenaria spp.) from Florida. J. Shellfish Res. 9: 215-225. 134

PAGE 150

Jones, D.S., I. Thompson & W. Ambrose, 1978. Age and growth rate determinations for the Atlantic surf clam Spisula solidissima (Bivalvia: Mactracea), based on internal growth lines in shell cross-sections. Mar. Biol. 47: 63-70. Juanes, F., 1992. Why do decapod crustaceans prefer smallsized molluscan prey? Mar. Ecol. Prog. Ser. 87: 239-249. Kappenman, R.F., 1981. A method for growth curve comparisons. Fish. Bull. 79: 95-101. Kaufmann, K.W., 1981. Fitting and using growth curves. Oecologia 49: 293-299. Kennish, M.J., 1980. Shell microgrowth analysis: Mercenaria mercenaria as a type example for research in populations dynamics. In, Skeletal growth of aquatic organisms: biological records of environmental change, edited by D.C. Rhoads & R.A. Lutz, Plenum Press, New York, pp. 255-294. Kennish, M.J. & R.E. Loveland, 1980. Growth models of the northern quahog, Mercenaria mercenaria (Linne). Proc. natn. Shellfish. Assoc. 70: 230-239. Kerswill, C.J., 1949. Effects of water circulation on the growth of quahaugs and oysters. J. Fish. Res. Bd. Can. 7: 545-551. Kvalseth, T.O., 1985. Cautionary note about R2 Am. Statistn. 39: 279-285. Levinton, J.S., 1982. Marine Ecology. Prentice-Hall, Inc., Englewood Cliffs, New Jersey Lowenstam, H.A., 1954. Factors affecting the calcitearagonite ratios in carbonate secreting marine organisms. J. Geol. 62: 284-322. Lutz, R.A. & D.C. Rhoads, 1977. Anaerobiosis and a theory of growth line formation. Science 198: 1222-1227. Lutz, R.A. & D.C. Rhoads, 1980. Growth patterns within the molluscan shell: an overview. In, Skeletal growth o f aquatic organisms: biological records of environmental change, edited by D .C. Rhoads & R.A. Lutz, Plenum Press, New York, pp. 203-254. 135

PAGE 151

Mallet, A.L., C.E.A. carver, s.s. Coffen & K.R. Freeman, 1987. Winter growth of the blue mussel Mytilus edulis L.: importance of stock and site. J. Exp. Mar. Biol. Ecol. 108: 217-228. McCall, D., J.G. Cook, J.A. Lasater & T.A. Nevin, 1970. A survey of salinity levels in the Indian River-Banana River complex. Bull. Environ. Contam. T oxicol. 5: 414-421. McHugh, J.L. & M.W. Sumner, 1988. Annotated bibliography II of the hard clam Mercenaria mercenaria. United States Dept. Commerce, NOAA Tech. Rep. NMFS 69, 59 pp. McHugh, J.L., M.W. Sumner, P.J. Flagg, D.W. Lipton, & W.J. Behrens, 1982. Annotated bibliography of the hard clam (Mercenaria mercenaria). United States Dept. Commerce, NOAA Tech. Rep. NMFS SSRF-756, 845 pp. Menzel, R.W., 1961. Seasonal growth of the northern quahog, Mercenaria mercenaria, and the southern quahog, M. campechiensis, in Alligator Harbor, Florida. Proc natn. Shellfish. Assoc. 52: 37-46 Menzel, R.W., 1962. Seasonal growth of northern and southern quahogs, Mercenaria mercenaria and M. campechiensis, and their hybrids in Florida. Proc. natn. Shellfish. Assoc. 53: 111-119. Menzel, R.W., 1989. The biology, fishery and culture of quahog clams, Mercenaria. In, Clam mariculture in North America, edited by J.J. Manzi & M. Castagna, Developments in Aquaculture and Fisheries Science, Vol. 19, Elsevier Science Publishing Company Inc, New York, pp. 201-242. Neff, N A. & G.R. Smith, 1978. Multivariate analysis of hybrid fishes. Systematic Zoology 28: 176-196. Nielsen, M.V. & T. Stromgren, 1985. The effect of light on the shell length growth and defaecation rate of Mytilus edulis (L.). Aquaculture 47: 205-211. NOAA, 1986. Local Climatological Data, Ft. Myers, Florida. NOAA, 1988. Local Climatological Data, Daytona Beach, Florida. Page, H.M. & Y.O. Ricard, 1990. Food availability as a limiting factor to mussel Mytilus edulis growth in California coastal waters. Fish. Bull. 88: 677-686. 136

PAGE 152

Panella, G. & c. MacClintock, 1968. Biological and environmental rhythms reflected in molluscan shell growth. J. Paleontol. 42: 64-80. Peterson, C.H., 1983. A concept of quantitative reproductive senility: application to the hard clam, Mercenaria mercenaria (L.)? Oecologia 58: 164-168. Peterson, C.H. & B.F. Beal, 1989. Bivalve growth and higher order interactions: importance of density, site, and time. Ecology 70: 1390-1404. Peterson, C.H. & S.R. Fegley, 1986. Seasonal allocation of resources to growth of shell, soma, and gonads in Mercenaria mercenaria. Biol. Bull. 171: 597-610. Peterson, C.H., P.B. Duncan, H.C. Summerson & B.F. Beal, 1985. Annual band deposition within shells of the hard clam, Mercenaria mercenaria: consistency across habitat near Cape Lookout, North Carolina. Fish. Bull. 83: 671-677. Peterson, C.H., P.B. Duncan, H.C. Summerson & G.W. Safrit, Jr., 1983. A mark-recapture test of annual periodicity of internal growth band deposition in shells of hard clams, Mercenaria mercenaria, from a population along the southeastern United States. Fish. Bull. 81: 765-779 Peterson, C.H., H.C. Summerson, & P.B. Duncan, 1984. The influence of seagrass cover on population structure and individual growth rate of a suspension-feeding bivalve, Mercenaria mercenaria. J Mar. Res. 42 : 123-138. Poole, B.D. & D.W. Arnold, 1985. Shellfish survey of restricted and prohibited area classifications i n Brevard County, Florida. Florida Department of Natural Resources, Tallahassee, 64 pp. Pratt, D.M., 1953. Abundance and growth of Venus mercenaria and Callocardia morrhuana in relation to the character of bottom sediments. J. Mar. Res. 1 2 : 60-74. Pratt, D.M. & D.A. Campbell, 1956. Environmental factors affecting growth in Venus mercenaria. Limnol. Oceanogr. 1: 2-17. 137

PAGE 153

Quitmyer, I.R. & D.S. Jones, 1992. Calendars of the coast: seasonal growth increment patterns in shells of modern and archaeological southern quahogs, Mercenaria campechiensis, from Charlotte Harbor, Florida. In, Culture and environment in the domain of the Calusa, edited by W.H. Marquardt, Monograph 1, Institute of Archaeology and Paleoenvironmental Studies, University of Florida, Gainesville, pp. 247-264. Quitmyer, I.R., H.S. Hale & D.S. Jones, 1985. Paleoseasonality determination based on incremental shell growth in the hard clam, Mercenaria mercenaria, and its implications for the analysis of three southeast Georgia coastal shell middens. Southeast. Archaeol. 4: 27-40. Quitmyer, I.R., D.S. Jones & w.s. Arnold, in press. The sclerochronology of hard clams, Mercenaria spp., from the southeastern U.S.: a method of elucidating the zooarchaeological records of seasonal resource procurement and seasonality in prehistoric shell middens. J. Archaeol. Sci. Rawson, P.D. & T.J. Hilbish, 1991. Genotype-environment interaction for juvenile growth in the hard clam Mercenaria mercenaria (L.). Evolution 45: 1924-1935. Rhoads, D.C. & R.A. Lutz, environmental change. organisms: biological edited by D.C. Rhoads York, pp. 1-19. 1980. Skeletal records of In, Skeletal growth of aquatic records of environmental change, & R.A. Lutz, Plenum Press, New Rice, J.D. R.P. Trocine & G.N Wells, 1983. Factors influencing seagrass ecology in the Indian River lagoon. Florida Sci. 46: 276-286. Rice, W.R., 1989. Analyzing tables of statistical tests. Evolution 43: 223-225. Rosenberg, G.D. & S.K. Runcorn, 1975. Growth rhythms and the history of the earth's rotation. John Wiley, London, 538 pp. Saloman, C.H. & J.L. Taylor, 1969. Age and growth of large southern quahogs from a Florida estuary. Proc. natn. Shellfish. Assoc. 59: 46-51. SAS Institute, Inc., 1985. SAS User's Guide: Statistics, Version 5 Edition. SAS Institute, Inc., Cary, North Carolina, 956 pp. 138

PAGE 154

Sato, S., 1995. Spawning periodicity and shell microgrowth patterns of the venerid bivalve Phacosoma japonicum (Reeve, 1850). Veliger 38: 61-72. Schick, D.F., S.E. Shumway & M.A. Hunter, 1988. A comparison of growth rate between shallow water and deep water populations of scallops, Placopecten magellanicus (Gmelin, 1791), in the Gulf of Maine. Am. Malacolog. Bull. 6: 1-8. Smith, N.P., 1987. An introduction to the tides of Florida's Indian River lagoon. I. Water levels. Florida Sci. 50: 49-61. smith, N.P., 1993. Tidal and wind-driven transport between Indian River and Mosquito Lagoon, Florida. Florida Sci. 56: 35-246. Sokal, R.R. & F.J. Rohlf, 1981. Biometry: the principles and practice of statistics in biological research. W.H. Freeman and Company, San Fransisco, California, 2nd edition, 859 pp. Sokal, R.R. & F.J. Rohlf, 1995. Biometry: the principles and practice of statistics in biological research. W H. Freeman and Company, New York, 3rd edition, 887 pp. Stanley, J.G. & R. DeWitt, 1983. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (North Atlantic) Hard Clam. United States Dept. of Interior, FWS/OBS-82/11.18, 19 pp. Storr, J.F., A.L. Costa & D.A. Prawel, 1982. Effects of temperature on calcium deposition in the hard-shell clam, Mercenaria mercenaria. J. Therm. Biol. 7: 57-61. Sunderlin, J.B., M. Brenner, M. Castagna, J. Hirota, R.W. Menzel, and O.A. Roels, 1975. Comparative growth of hard shell clams (Mercenaria mercenaria Linne and Mercenaria campechiensis Gmelin) and their F1 cross in temperate, subtropical and tropical natural waters and in a tropical artificial upwelling mariculture system. Proc. 6th Annual Meeting, World Mariculture Soc., pp. 171-183. Tanabe, K., 1988. Age and growth rate determinations of an intertidal bivalve, Phacosoma japonicum, using internal shell increments. Lethaia 21: 231-241. 139

PAGE 155

Tanabe, K. & T. Oba, 1988. Latitudinal variation in shell growth patterns of Phacosoma japonicum (Bivalvia: Veneridae) from the Japanese coast. Mar. Ecol. Frog. Ser. 47: 75-82. Vaughan, D .E., 1988. Florida, U.S.A. Clam culture: state of the art in J. Shellfish Res. 7: 546. von Bertalanffy, L., 1938. A quantitative theory of organic growth. Human Biology 10: 181-213. Walker, R.L., 1985. Growth and optimum seeding time for the hard clam, Mercenaria mercenaria (L.), in coastal Georgia. The Nautilus 99: 127-133. Walker, R.L. & C.M. Humphrey, 1984. Growth and survival of the northern hard clam Mercenaria mercenaria (Linne) from Georgia. J. Shellfish Res. 4: 125-129. Whetstone, J.M. & A.G Eversole, 1981. Effects of size and temperature on mud crab, Panopeus herbstii, predation on hard clams, Mercenaria mercenaria. Estuaries 4: 153-156. White, c., 1986. Biological and environmental factors affecting the clamming industry. In, An o v erview of the Indian River clamming industry and the Indian Rive r Lagoon, edited by D. Busby, Florida Sea Grant Extension Program, Tech. Paper No. 44, University of Florida, Gainesville, Florida, pp. 9-13. Zar, J.H., 1984. Biostatistical analysis, 2nd edition. Prentice Hall, Englewood Cliffs, New Jersey, 718 pp. 140

PAGE 156

VITA William s. Arnold was born in Cocoa, Florida, on March 26, 1955, and graduated from Cocoa High School in June, 1973. He attended the University of Florida, where he obtained a Bachelor of Science degree in Zoology in June, 1978. Mr. Arnold was employed as an Environmental Educator from 1978 through 1979 and as a Quality Control Chemist during 1979 through 1980. Subsequent to his acceptance in the graduate program in Zoology at the University of Georgia, Mr. Arnold studied and published on the interaction between blue crab (Callinectes sapidus) predators and their hard clam (Mercenaria mercenaria) prey. He received the Master of Science degree in Zoology from that institution in 1983, after which he was recruited by the Harbor Branch Oceanographic Institution to assist in research on mid-water gelatinous organisms. Mr. Arnold accepted a job with the Florida Marine Research Institute in 1985, where he initiated and continues research on the ecology of hard clams and other commercially important fisheries throughout the southeastern United states.


printinsert_linkshareget_appmore_horiz

Download Options

close
No images are available for this item.
Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.