W&M ScholarWorks

Dissertations, Theses, and Masters Projects Theses, Dissertations, & Master Projects

1994

Quantification of settlement and ecruitmentr processes in bivalve mollusks

Patrick Kelly Baker College of William and Mary - Virginia Institute of Marine Science

Follow this and additional works at: https://scholarworks.wm.edu/etd

Part of the Ecology and Evolutionary Biology Commons, Marine Biology Commons, and the Zoology Commons

Recommended Citation Baker, Patrick Kelly, "Quantification of settlement and ecruitmentr processes in bivalve mollusks" (1994). Dissertations, Theses, and Masters Projects. Paper 1539616555. https://dx.doi.org/doi:10.25773/v5-psh5-eg48

This Dissertation is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Dissertations, Theses, and Masters Projects by an authorized administrator of W&M ScholarWorks. For more information, please contact [email protected]. INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313/761-4700 800/521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with with permission permission of the of copyrightthe copyright owner. owner. Further Further reproduction reproduction prohibited prohibitedwithout permission. without permission. Order Number 9414201

Quantification of settlement and recruitment processes in bivalve m ollusks

Baker, Patrick Kelly, Ph.D.

The College of William and Mary, 1994

UMI 300N.ZeebRd. Ann Arbor, M I 48106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with with permission permission of the of copyrightthe copyright owner. owner. Further Further reproduction reproduction prohibited prohibitedwithout permission. without permission. QUANTIFICATION OF SETTLEMENT AND RECRUITMENT PROCESSES IN BIVALVE MOLLUSKS

A Dissertation Presented to The Faculty of the School of Marine Science The College of William and Mary

In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

by Patrick Kelly Baker 1993

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPROVAL SHEET This dissertation is submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Patrick Kelly Baker

Approved January, 1994

Roger (Mann, Ph.D.

Joiin Brubaker, Ph.D.

Romuald Lipcius, Ph.D.

il vv r i i u u ) Anson H. Hines, Ph.D. Smithsonian Environmental Research Center Edgewater, Maryland

ii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

Page ACKNOWLEDGEMENTS...... iv LIST OF TABLES...... V LIST OF FIGURES...... viii

ABSTRACT...... X CHAPTER 1. Review of the Ecology and Study of Recruitment of Bivalve Mollusks...... 2 CHAPTER 2. Competency to settle in oyster larvae, Crassostrea virginica: wild versus hatchery-reared larvae...... 47 CHAPTER 3. Temporal and spatial variability in abundance of late stage bivalve larvae in Chesapeake Bay...... 71 CHAPTER 4. A field comparison of planktonic abundance of oyster larvae, Crassostrea virginica and Ostrea equestris, with subsequent settlement...... 12 3 CHAPTER 5. Proportional settlement and recruitment of Crassostrea virginica: a quantitative field test using larval enclosures...... 176 CHAPTER 6. Conclusions and Summary...... 240 APPENDIX A. Review of Cues for Marine Invertebrate Larval Settlement and Metamorphosis...... 256 APPENDIX B. Effect of neutral red stain on settlement ability of oyster pediveligers, Crassostrea Virginia...... 330 APPENDIX C. Occurrence of post-metamorphic bivalves the plankton in lower Chesapeake Bay...... 341 VITA...... 381

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS Portions of this research were made possible by Minor Research Grants from the Virginia Institute of Marine Science (1989, 1992, 1993), and by a Grant-In-Aid of Research from Sigma Xi (1993). Most other funding was provided by the Commonwealth of Virginia, through the Bivalve Ecology program at VIMS. The following VIMS facilities and services were instrumental in my research. The oyster hatchery provided oyster larvae and algae; thanks to Valerie Shaffer for doing a great job. VIMS East at Wachapreague provided bivalve larvae and laboratory facilities. Vessel Operations maintained vessels; special thanks to Shirley Crossley and Raymond Forrest. The photo lab (Bill Jenkins) and APRC (Kay Stubblefield and Harold Burrell) cheerfully, skillfully, and characterfully provided graphics for presentations. Thanks to Marilyn Lewis and Diane Walker of the library for fast and friendly reference assistance (It's a bird! It's a plane! No, it's Reference Librarian!) Danny Gouge of the dive locker provided and maintained scuba gear. Thanks to fiscal agents Regina Burrell and Carol Tomlinson for keeping my head screwed on straight. Appreciation is due to persons assisting in laboratory and field work, especially Shirley Baker, Reinaldo Morales- Alamo, and Kenneth Walker. Many thanks to scuba divers who assisted me, including Aaron Adams, Mark Brotman, Marty Cavalluzzi, Judy Haner, Kirt Moody, David Plotner, and Rochelle Seitz. Thanks also to the members of my dissertation committee for their reviews of this dissertation. The Aquaculture Division of the Harbor Branch Oceanographic Institute, and the Smithsonian Marine Station at Linkport, graciously permitted me the use of their facilities at Fort Pierce, Florida. Thanks to Sherry Reed of SMSL for her help. And thanks to the 'gators for not eating me during night field work. Thanks to Dan Hornbach and Macalester College in St. Paul, Minnesota, for the use of their facilities while preparing the final drafts of this dissertation, and to Tony Deneka for use of his personal computer during this time. Thanks to Roger Mann, my primary advisor. I realize that, in many ways, this was as much a learning experience for him as for me, but in the long analysis, he was usually right. I appreciate his patience, support, and his stubborn British syntax, throughout this process. Special thanks to my wife, Shirley, for her many hours of assistance, for her patience and understanding, and for iv

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

Tables numbered separately for each chapter.

Page Chapter 2. 1. Summary of proportional settlement of Crassostrea virginica larvae from the York River...... 69 2. Proportional settlement of hatchery-reared Crassostrea virginica larvae...... 70

Chapter 3. 1. Correlation coefficients r for planktonic larval abundance data, between pairs of species...... 106 2. Mean abundance of larvae for four sample times, 1990...... 107 3. Summary of analysis of variance for the effects of tidal phase and depth on larval abundance...... 108 4. Summary of analysis of variance for the effects of depth near the benthos on proportional abundance of larvae...... 109 5. Summary of analysis of variance for the effects of time of day and tidal phase on proportion of larvae in near-bottom water, for three species...... 110

Al. Correlation coefficients r between larval abundance, and water temperature and salinity, lagged for 17 days prior to commencement of sampling...... 118 Bl. Abundance of three species of pediveliger larvae at three depths in 1991, for four tidal cycles..... 119

Chapter 4. 1. Mean proportional abundance of three species of bivalve pediveliger larvae at two times and two depths, at the Indian River, Florida site...... 162 2. Summary of analysis of variance for the effects of time of day and depth on abundance of three species of bivalve larvae at the Indian River, Florida, site...... 163

v

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES (cont.) Tables numbered separately for each chapter.

Page Chapter 4. (cont.) 3. Values used to estimate mortality of newly settled Crassostrea juveniles, over time, at the York River Virginia, site...... 164 4. Observed Crassostrea recruitment, estimates of total weekly settlement, intermediate steps, and mean planktonic abundance of pediveliger larvae, in the York River, Virginia, 1990...... 165 5. Summary of analyses of regression of oyster settlement on planktonic abundance, for York River, Virginia, and Indian River, Florida...... 166

Chapter 5. 1. Results and analysis of variance of sedimentation within and adjacent to experimental enclosures..... 219 2. Summary of mean Crassostrea settlement-100 cm'2, across distance from center of enclosure...... 220 3. Summary of settlement and survival for reef and defaunated treatments...... 222 4. Summary of estimated proportional settlement and survival from Day 0 of Crassostrea, for primary experiment...... 224 5. Summary of regression analysis for Crassostrea survival, for the primary experiment...... 225 6. Proportional survival over time for Crassostrea juveniles held in a laboratory flume, July 1-28.... 226 7. Settlement of Crassostrea in back-up experimental run...... 227 8. Results and analysis of field settlement and survival of Crassostrea in the back-up experimental run...... 228

vi

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES (cont.) Tables numbered separately for each chapter.

Page Chapter 5. (cont.) 9. Results and analysis of Crassostrea juvenile survival in the laboratory flume and suspended from the pier, in the back-up experimental run...... 229 10. Results and analysis of pier survival inthe back-up experimental run...... 230 11. Proportions of Crassostrea juveniles recorded on the convex surface of shell substrates from enclosures in back-up experiment...... 231 12. Settlement of Crassostrea larvae onto upper and lower surfaces of adult shells in the laboratory...... 232 13. List of common fouling organisms found in reef enclosures, including higher level taxonomy and common name, following the species.. name...... 233 14. Summary of regression analyses of effect of various shell substrate characteristic on settlement of Crassostrea on Day 0...... 234 Al. Relationship of shell mass to shell outline area...238

Appendix B. 1. Settlement of Crassostrea, expressed as proportions of total larvae...... 339 2. Summary of two factor analysis of variance on effects of stain and substrate on Crassostrea settlement...... 340

Appendix C. 1. Occurrence of postlarval drift by byssal threads in Bivalvia...... 379

vii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

Figures numbered separately for each chapter. Page Chapter 3. Figure 1. Map of study area, lower York River, Chesapeake Bay, Virginia...... Ill Figure 2. Diagram of vertical sampling array .....112 Figure 3. Diagram of horizontal sampling array intake...... 113 Figure 4. Abundance of three species of pediveliger larvae in near-bottom water, 1990, and simultaneous temperature and salinity...... 114 Figure 5. Abundance of three species of pediveliger larvae in near-bottom water, 1991, for four tidal cycles...... 115 Figure 6. Proportional abundance of three species of pediveliger larvae at three depths, 1991...... 116 Figure 7. Proportional abundance of four species of pediveliger larvae at three depths, 1992...... 117

Chapter 4. Figure 1. Abundance of Crassostrea virginica pediveliger larvae, July 5 - August 9, 1990, at the York River, Virginia, site...... 168 Figure 2. Abundance of three species of pediveliger larvae, May 3-31, 1993, at the Harbor Branch, Florida, site...... 169 Figure 3. Survival of recently settled Crassostrea virginica at the York River, Virginia, site...... 170 Figure 4. Estimated settlement of Crassostrea virginica onto shell strings at the York River, Virginia, site, versus pediveliger larval abundance in the plankton...... 171

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES (cont.)

Page Chapter 4 . (cont.) Figure 5. Settlement of Crassostrea virginica onto shell strings at the Harbor Branch, Florida, site, versus pediveliger larval abundance in the plankton...... 172 Figure 6. Settlement of Ostrea equestris onto shell strings at the Harbor Branch, Florida, site, versus pediveliger larval abundance in the plankton...... 173 Figure 7. Residuals of the regression of Crassostrea virginica settlement on planktonic larval abundance (direct data set), plotted against planktonic larval abundance...... 174 Figure 8. Residuals of the regression of Ostrea equestris settlement on planktonic larval abundance (direct data set), plotted against planktonic larval abundance...... 175

Chapter 5. Figure 1. Diagram of experimental enclosure...... 235 Figure 2. Estimated proportional survival of Crassostrea juveniles in experimental enclosures and flume control...... 236

ix

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. QUANTIFICATION OF SETTLEMENT AND RECRUITMENT PROCESSES

IN BIVALVE MOLLUSKS

ABSTRACT Studies were carried out to quantify abundance, mortality, and variability in these parameters, during settlement and recruitment of bivalve mollusks, using the oyster, Crassostrea virginica, as a primary model species. Most work was undertaken in the York River, Chesapeake Bay, Virginia, with additional work in the Indian River, Florida. The period chosen, in the bivalve early life history, was from the late planktonic larva to the early benthic juvenile. Studies were designed to specifically examine a) abundance of late-stage larvae in the plankton, b) the relationship between larval abundance and settlement, and c) mortality immediately following settlement. Variability in abundance or mortality was also examined at each of these stages. It was found that, of larvae in the plankton with the morphological characteristics of competency-to- settle, about 80% would settle within 24 hours, under laboratory conditions. Temporal variation in abundance of planktonic larvae was high and apparently random, but separate species covaried in observed abundance. Time of day and tidal phase had little or no effect on larval distribution, but late stage larvae showed a consistent depth preference, which varied depending on species. Crassostrea late larvae tended to be most abundant near the benthos, both at the Virginia site and at the Florida site. The relationship between planktonic abundance and settlement density of Crassostrea was weak, within one order of magnitude of variation in larval abundance, although the same relationship for a sympatric species in Florida, Ostrea equestris, was significant. Settlement onto a natural oyster reef was more variable but not significantly lower than settlement onto adjacent defaunated shell substrate. The increased variability could not be accounted for by coverage by dominant fouling macroorganisms. Mortality of newly-settled juvenile Crassostrea was high, with near 100% mortality within 28 days on a natural oyster reef, with high but significantly lower (about 96%) mortality on previously defaunated shell substrate over the same time.

Patrick Kelly Baker SCHOOL OF MARINE SCIENCE THE COLLEGE OF WILLIAM AND MARY IN VIRGINIA

X

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. QUANTIFICATION OF SETTLEMENT AND RECRUITMENT PROCESSES IN BIVALVE MOLLUSKS

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1 REVIEW OF THE ECOLOGY AND STUDY OF RECRUITMENT OF BIVALVE MOLLUSKS

2

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3

Table of Contents Introduction...... 4 Life Cycles of Bivalve Mollusks...... 5 Egg...... 6 Planktonic Larva...... 9 Non-Planktonic Larva...... 11 Competent Larva...... 13 Planktonic Postlarva...... 17 Benthic Postlarva...... 19 Juvenile...... 21 Life History Research...... 23 Zygote Production...... 23 Planktonic Larval Abundance...... 24 Postlarval Abundance...... 28 Present Research...... 31 References...... 34

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4

Introduction Animal populations depend upon dispersal of individuals to maintain population levels in the presence of environmental variability, and to maintain genetic variability. This requirement is met at the juvenile and adult levels of many highly mobile marine and aquatic organisms, including most fish, decapod crustaceans, and cephalopod mollusks. Benthic invertebrates which are sedentary or completely sessile as juveniles and adults depend upon planktonic larvae or postlarvae for dispersal. The planktonic stage of sedentary invertebrates includes both dispersal and juvenile recruitment, and is thus critical to our understanding of the ecology of these organisms. Research in the early life history of sedentary invertebrates can be divided into two general categories. The first category is descriptions of factors and mechanisms of early life history processes, including development and developmental physiology, stage-specific mortality factors, and recruitment factors. Steady advances have been made in these areas, some of which will be reviewed later in this chapter. The second research category is quantitative population ecology, including fecundity, planktonic survival, and recruitment success. Advances have been made in some areas of this, especially fecundity, and recruitment

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 of juveniles of over 1 mm in size, but little has been achieved or attempted on some intermediate stages between fecundity and juveniles. Research problems in these areas will be discussed later in this and other chapters. Bivalve mollusks, because of their relatively large size and commercial importance, are the best-studied group of sedentary benthic invertebrates. Where abundant, bivalves play major roles in nutrient cycling, regulating planktonic biomass, and biodeposition, as a result of their rapid and efficient suspension feeding (Newell, 1988; Ulanowicz, 1988; Bunt et al ., 1991). The shells of some species, in aggregate or individually, are critical habitat for many other organisms (Wells, 1961; Driscoll, 1968; Seed, 1976). The rest of this chapter will concentrate on bivalve mollusks, although other taxa will be discussed, when appropriate.

Life Cycles of Bivalve Mollusks Based upon larval dispersal, sedentary invertebrates can be divided into three groups, which grade together. These three groups are: species with long-lived planktonic larvae, species with short-lived planktonic larvae, and species without planktonic larvae. "Long-lived" is here defined in terms of tidal rhythms (and hence dispersal distance), which are approximately 12.4 hours in duration; long-lived larvae are in the water column for more than one

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tidal cycle. For freshwater ecosystems, planktonic durations could also measured in terms of periods of stream velocity or wind-driven circulation (e.g. Martel, 1993; Martel et al., 1993), but these vary greatly between river or lake systems, so 12.4 hours is meaningful primarily for coastal marine ecosystems. The above three categories — long-lived planktonic larvae, short-lived planktonic larvae, and no planktonic larvae, are functionally similar to the divisions proposed for bivalve larvae by Ockelmann (1965). Mileikovsky (1971) has an alternate scheme for dividing invertebrate development which further divides planktonic development into pelagic versus demersal development. As will be discussed later, the duration of the planktonic larval period may not be as critical to dispersal as it has been historically assumed.

Egg Most bivalve mollusks produce small eggs, unprotected by a shell or case. The eggs are small (40-60 p in diameter is typical), and may number in the millions (e.g. Stanley, 1985; Stanley and Sellers, 1986; Borcherding, 1991). The geoduc, Panope generosa, the largest infaunal bivalve, releases up to 20 million eggs 80 /z in diameter, although the ovaries may hold many more eggs than this (Goodwin and Pease, 1989). The giant clam, Tridacna gigas, the largest bivalve mollusk, may release up to 500 million eggs 100 /z in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 diameter (Crawford et a l ., 1986). The size and nutritional content of each egg affects proportional larval survival (Kraeuter et al., 1982; Gallager and Mann, 1986), and may be a means whereby adults can trade larval abundance for proportional survival. A few bivalves produce a protective coating for the planktonic egg, such as Turtonia, in which

larval development occurs entirely within a drifting egg case (Ockelmann, 19 64); and Solemya, which has a planktonic larval stage very different from most other bivalves (Gustafson and Lutz, 1992). On both intraspecific and interspecific levels, egg strategy is one of the few known forms of parental investment that can vary, among bivalve mollusks. Within intertidal populations of the mussel Mytilus edulis, egg size and number, which are roughly inversely proportional, vary according to adult size and position (Bayne et a l ., 1983). Egg size affects total nutrient content, which has been shown to affect larval growth and survival (Gallager and Mann, 1986). Most bivalves, especially those with small eggs, gain nutrition for growth from planktonic feeding (planktotrophy), but increased egg size or nutrient content allows a larva to grow on reserves alone (lecithotrophy), insurance against potentially poor feeding conditions. Egg fertilization is problematical in sessile invertebrates. In species that release both eggs and sperm, fertilization failure can be viewed as planktonic mortality

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 of an early free-living life stage. Recent research by Pennington (1985), Babcock (1993), and Lasker (1993) has shown that, for a variety of invertebrates, fertilization success varies greatly with gamete age and density, both affected by adult proximity. Fertilization is also affected by microscale hydrodynamics (turbulence) in the water column (Mead and Denny, 1993). There are no known copulatory organs in bivalve mollusks, so the most common strategy to ensure high fertilization success is adult proximity. Usually this is achieved simply by high population density, but in some cases, small or dwarf males associate directly with females, either attached externally (e.g. Pascual et a l ., 1989), or brooded internally (e.g. Turner and Yakovley, 1983; O Foighil, 1985). In brooding oysters of the genus Ostrea, sperm are released in aggregates termed "sperm balls", or "spermatozeugmata", which may enhance fertilization success (Strathmann, 1987; O Foighil, 1989). Females of the freshwater clam Margaritifera margaritifera, at low population densities, can become hermaphrodites and self-fertilize (Bauer, 1987). In the majority of invertebrates, including bivalve mollusks, the egg phase lasts only a few hours (e.g. Loosanoff and Davis, 1963), and proceeds rapidly to a swimming larval stage. See Strathmann (1987) for reviews of development to the larval .stage. An important ecological significance of the egg, compared to the larva, is that the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9 egg is nonmotile, and cannot regulate its position in the water column. An egg cannot move to avoid stressful variations in temperature, salinity, oxygen, or toxicants. The role of the egg phase in dispersal is probably relatively minor for most bivalves, since its duration is brief compared to subsequent larval development. The egg in most species is very slightly negatively buoyant until a motile stage is attained (e.g. Carricker, 1951; Culliney, 1975; Fitt et al., 1984). Aside from fertilization failure (discussed above), agents of mortality are probably similar to those for planktonic larvae (discussed below).

Planktonic Larva

Following the egg phase, in bivalve mollusks, is the larva. In most invertebrates, the larval phase has been divided by researchers into a number of developmental stages, but the ecological significance of these stages is usually unclear. In bivalve mollusks, there is an initial gastrula stage, sometimes called a "conchostome" (Flyachinskaya and Kulakovinskiy, 1992), which is motile in some species (e.g. Culliney, 1974). This proceeds to a trochophore stage, during which the larval shell first

appears. The trochophore is followed in most marine groups by the veliger stage, usually within 24 hours of fertilization. The veliger stage is often further divided into the "D" stage, or "straight hinge" stage (so named for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the shape of the shell), followed by the "umbo" stage, and finally the "pediveliger" stage, which is equipped with a crawling foot. This sequence is based primarily upon development of the well-studied oysters (Ostreidae), but does not vary significantly for most other marine lamellibranch taxa (e.g. Loosanoff and Davis, 1963; Galtsoff, 1964; Chanley and Andrews, 1971), although in some taxa the development may occur within an "egg capsule" (not strictly equivalent to an egg capsule in gastropod mollusks; see Ockelmann, 1964) or within a brooding adult (see below). Protobranch bivalves have a different larval form termed the "pericalymma" (first described by Drew, 1896; reviewed by Gustafson and Lutz, 1992). All of these stages are ciliated and can swim, if in the plankton, at rates up to 10 mm • s1' (Mann and Wolf, 1983). See also Cragg (1980) for discussion of swimming behavior of veliger larvae. Size of veligers ranges from 50 /n at early stages of some species, to 350 n (Chanley and Andrews, 1971; Gustafson and Lutz, 1992). Larval stages are important in species dispersal of bivalve mollusks, but are also subjected to high mortality. Yoo and Ryu (1985) calculated the instantaneous mortality rate for early and late stages of the oyster Crassostrea gigas in the plankton, and found a consistently lower mortality rate for larger, older larvae. Jorgensen (1981) found a similar trend for Mytilus edulis. This indicates that not all larval stages are ecologically equal, but since

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 losses are due to a combination of many factors, including predators, flushing, disease, and water quality (see Berg, 1971; Hines, 1986; and Rumrill, 1990; for reviews of planktonic mortality), larval mortality in the plankton is a complex issue that is far from being understood. Water quality and food levels can affect larval survival directly, but can also affect larval growth rates or behavior, and therefore indirectly affect survival (e.g. Diaz, 1973; Dupuy, 1975; Widdows et al., 1989; Mann and Rainier, 1990). The planktonic larval stage is clearly a critical dispersal stage, but for the majority of bivalves, which can produce byssal-drifting larvae (see below), it may not be the only dispersal stage. Important taxa that would be exceptions to this include oysters (Ostreidae), which cement permanently upon settlement (Galtsoff, 1964), and shipworms (Teredinidae), which quickly bore into wood upon settlement (Culliney, 1975).

Non-Planktonic Larva The planktonic phase of bivalve larva may be shortened, or eliminated entirely, in some taxa. This is usually accomplished by internal fertilization of the female (taking up shed sperm) and subsequent brooding. Brooding, along with egg size (above) is one of relatively few forms of parental investment available to bivalves. Female bivalve mollusks that brood have modified gill chambers to hold

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. developing larvae, and in some freshwater clams of the family Unionidae, the sexual dimorphism extends to the shape of the shell (McMahon, 1991). Freshwater bivalves of the

order Unionoida have a non-planktonic larva, termed the "glochidium", that is parasitic on fish (see review by McMahon, 1991). Brooding times vary greatly between taxa. In the some oysters (Ostreidae) and shipworms (Teredinidae), brooding lasts part way through the veliger stage, and there is a subsequent planktotrophic stage of days to weeks (e.g. Hopkins, 1936; Calloway and Turner, 1986; Pascual et a l ., 1989) . The number of larvae produced numbers in tens to hundreds of thousands (e.g. Walne, 1963). Among some groups of sedentary invertebrates, a planktonic phase lasting only minutes or hours is typical. The best-known examples are among the ascidians (Urochordata), which produce a few, large larvae. See Barnes (1986) for larval descriptions. The oyster genus Tiostrea produces large eggs (3 00 ju, compared to 50 fi for

non-brooding Crassostrea), broods the larvae to the stage at which they are competent to settle and metamorphose, and has a planktonic phase lasting minutes to hours, although settlement can be delayed up to 24 hours (Walne, 1963; Galtsoff, 1964). Tiostrea females produce only about 10% of the larvae produced by similar-sized Ostrea edulis, a species which broods only part of the way through development, followed by a prolonged planktonic larval stage

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Walne, 1963). Mortality throughout the veliger phase for Crassostrea gigas has been estimated at about 90% (Yoo and Ryu, 1985), so the number of surviving final-stage larvae may be similar for oyster species with different larval strategies. Other examples of bivalves with brief planktonic larval periods include the freshwater clam Corbicula fluminea (King et al., 1986), and some shipworms, of the family Teredinidae (Calloway and Turner, 1986). The advantage of long-term brooding may lie in the ability of the female to time larval release with favorable environmental conditions, but this might be balanced by reduced dispersal.

Competent Larva The final larval phase in most marine invertebrates is the competent-to-settle larva, (usually referred to simply as "competent" larva) which in bivalve mollusks is represented by the pediveliger. The pediveliger has a crawling foot with byssal glands, a rudimentary gill, and in some taxa, a distinct photosensitive organ, or "eyespot". Coon et al. (1990) and Fitt et a l . (1990) have given

definitions of competency. See Galtsoff (1964) and Loosanoff et al. (1966) for detailed descriptions of Crassostrea virginica and Ensis directus pediveligers, respectively; Cranfield (1973a), Gruffydd et al. (1975), and Lane and Nott (1975) give detailed descriptions of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14 larval foot and byssal gland. Chia and Rice (1978) give various examples of settlement and metamorphosis among marine invertebrate larvae. Behavior is more important than morphology, however, in identifying this stage. The pediveliger can actively swim, but at some point it approaches the substrate, explores it by crawling and

swimming, and eventually ceases to swim and attaches, either by cementation in oysters (Galtsoff, 1964), or by byssal threads in most other taxa. This process is termed "settlement" (sometimes "setting"), and is behaviorally mediated by a variety of physical and chemical cues. As is also true for larval development, the best-known model for settlement is oysters, but examples exist for other bivalves also. For examples of settlement behavior in bivalve mollusks, see Carriker (1961), Cranfield (1973b), Keck et a l . (1974), and Cragg (1980). A complete discussion of various physical and chemical cues used for settlement of benthic invertebrates is given in Appendix A, towards the end of this volume. Metamorphosis in bivalve larvae involves the loss of the velum (by absorption or abscission), the growth of the gill, the loss of the eyespot, and other anatomical changes. Tissue and organ changes during molluscan metamorphosis has been described for Mytilus edulis by Bayne (1971), and reviewed by Bonar (1976) and Fioroni (1982). The metamorphosis of the oyster Crassostrea virginica has been

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15 described by Baker and Mann (in press), and comparative anatomy before and after metamorphosis is presented for the scallop Patinopecten yessoensis by Bower and Meyer (1990) . Metamorphosis is a continual process, starting before settlement and continuing after; see Pechenik and Heyman (1987) and Zimmerman and Pechenik (1991) for definitions of metamorphosis in molluscan larvae. In oysters, the gill is rudimentary prior to the loss of the velum (Baker, in press), but in clams of the genus Macoma, the gill is fairly large and ramified prior to the loss of the velum (see Appendix C). Competency to settle has definite physiological and morphological characteristics (for review, see Fitt et al., 1990; Coon et al., 1990). In an ecological sense, however, competency begins only when settlement behavior- approaching the substrate, searching, and crawling- begins, and ends when that behavior ends in settlement, or a permanent cessation of swimming. During this phase, larvae are out of the plankton, and subject to a variety of mortality agents, and benthic sensory stimuli. Predation by benthic planktivores becomes important at this stage. For bivalve larvae, known common predators include adult conspecifics (Andre et al., 1993), barnacles and small sea anemones (Steinberg and Kennedy, 1979), spionid polychaetes (Breese and Phibbs, 1972), and large tunicates (Osman et a l ., 1989). These are probably only a few of many benthic predators that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16 consume bivalve larvae. Some suspension feeders such as large bivalve mollusks and polychaete worms do not feed on particles as large as bivalve larvae, but may capture them and reject them fatally entangled in pseudofeces (Mackenzie, 1981). In addition to subjecting competent larvae to new forms of mortality, the benthos is teeming with signals that inhibit, enhance, or excite settlement behavior. Many of these are related to habitat suitability. Shipworras (Teredinidae) require wood for settlement (Turner, 1966); oysters (Ostreidae) require hard substrate, such as shell, rock, or wood, for settlement (Cake, 1983); and mussels (Mytilidae) and scallops (Pectinidae) require hard or filamentous substrates (Fay et a l., 1983; Mullen and Moring, 1986; Shaw et a l., 1988; Newell, 1989). Many bivalves merely require mud or sand, but even this can be limiting, such as by heavy production of macroalgae (Olafsson, 1988). These substrates often have tactile or chemical cues associated with them; response to tactile cues is termed "rugotaxis", and response to chemical cues is termed "chemotaxis". The adjective "positive" or "negative" is also used when describing a taxis. Other settlement cues include light intensity (resulting in phototactic behavior), gravity (geotactic) or pressure (barotactic), or allelochemicals released by other benthic invertebrates. Chevolot et a l . (1991) has reviewed chemical cues in bivalve

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17 molluscan settlement; other examples of all of various cues, and a complete review of settlement cues used by marine invertebrates, are given in Appendix A.

Planktonic Postlarva Some sedentary invertebrates, including some bivalve mollusks, lack a planktonic larval phases, and in the past these species were thought to lack any significant dispersal ability. This should lead to genetic drift, but while a test of this with an intertidal snail, Nucella lamellosa, which lacks planktonic larvae, showed some evidence of genetic drift, the drift was slight over a large range, and showed no consistent trends based on geographic location (Grant and Utter, 1988). This, plus this species widespread distribution on isolated rocky outcroppings, indicates some dispersal ability other than planktonic larvae. Evidence for an increasing number of marine species indicates post- larval planktonic dispersal mechanisms. Most bivalve species can produce long byssal threads as juveniles, which catch the current and can carry the juveniles for extended periods of times (e.g. Sigurdsson et al., 1976; Beukema and de Vlas, 1989). Most of these byssal-producing species also have planktonic larvae, but some bivalve species that lack planktonic larvae can also disperse by byssal drifting, as juveniles.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18 The term "postlarva" has been chosen for bivalve mollusks by this author on the basis of its ecological role, rather than upon a specific set of morphological characteristics. Metamorphosis is complete, yet the organisms has not yet taken up an adult lifestyle, and in the case of byssal drifting, the organisms still functions much like a larva in terms of dispersal. See Appendix C for discussion and references regarding the postlarval concept and terminology. Carriker (1961) used the term "plantigrade" to describe the post-metamorphic stage of the clam Mercenaria mercenaria which does not burrow (as do older juveniles and adults), but attaches to surface substrates with byssal threads. Benthic postlarval stages will be discussed separately, below. It is probable that most small bivalve species which lack planktonic larvae have a byssal drifting stage. Bivalve mollusks which lack a planktonic larval phase are usually small as adults (l cm or less), and brood their larvae past metamorphosis. Sellmer (1967) and McMahon

(1991) review brooding in a number of bivalve mollusks. Byssal drifting threads are invisible under light microscopy, and have been overlooked by most researchers, but most bivalve taxa produce byssal threads as juveniles (Sigurdsson et al., 1976). In the small bivalve Turtonia minuta, the females produce egg capsules attached to long byssal threads (Ockelmann, 1964), and it is possible that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19 both eggs and juveniles in this species can disperse by byssal drifting. Byssal-drifting in Transenella and Lasaea, two small brooding bivalves, has been described (Martel and Chia, 1991). Byssal drifting threads have not specifically been reported in the clam Gemma gemma, but juveniles of this species have been found in the water column, and this species apparently has a byssal gland (Sellmer, 1967). These bivalve species produce only a few hundred juveniles annually, which is in part due to their small size, but possibly in part to the space occupied by the large eggs (120-150 n in Turtonia), larvae, and unreleased juveniles (375 n in Gemma, 500 /z in Transenella) (Ockelmann, 1964; Sellmer, 1967; Asson-Batres, 1988). Further review of byssal drifting, and examples from Chesapeake Bay collected during the course of this dissertation research, are given in Appendix C, towards the end of this volume.

Benthic Postlarva The concept of a benthic postlarva is most useful in

deep-burrowing clams— that is, bivalves which actively burrow throughout most of their juvenile and adult life, but which do not relocate as adults unless uncovered. The newly settled juveniles (postlarvae) of most of these go through an epibenthic phase, characterized by crawling and byssal attachment, before digging their first semipermanent burrows. Examples are given by Carriker (1961), Hoese

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 (1973), Berg and Alatalo (1985), Yankson (1986), and Baker and Mann (1991). This stage has been called the "plantigrade" by some (Carriker, 1961). Although the shell, siphon, and mantle develop throughout this phase to a juvenile configuration, there is not a distinct period at which active epibenthic behavior ceases, but rather a gradual increase in the frequency of burrowing by individuals. By about 10 mm, active crawling has usually ceased (Carriker, 1961; Baker and Mann, 1991). The benthic postlarva of deep-burrowing clams probably serves as to allow the animal to "sample" the local environment before choosing a site for a semipermanent, and to readjust for a changing environment. This is not migration because crawling postlarvae probably react to negative stimuli (e.g. competitors, unfavorable water quality), and therefore move away from a location, rather than towards a specific goal (Yankson, 1986). It is not dispersal in the sense that planktonic larvae disperse, nor does it necessarily result in an even or random distribution of juveniles. Mortality factors include siltation, water quality, and predators (reviewed by Baker and Mann, 1991). In oysters (Ostreidae) and shipworms (Teredinidae), which immediately take up a sedentary lifestyle upon settlement (see Galtsoff, 1964; Culliney, 1975), the concept of a postlarva is not ecologically significant. This post- metamorphic stage remains significant to the researcher,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21 because it starts out too small to see with the naked eye, and therefore requires a special sampling scheme, but in terms of mortality factors and trophic function is probably not different from slightly larger juveniles which are visible to the naked eye. For species which are active long into their juvenile stage, such as Mytilus (Newell, 1989), it is likewise difficult to separate a benthic postlarval phase from the following, only slightly less active, juvenile phase, although as discussed above, there may be a significant planktonic postlarval phase.

Juvenile "Juvenile" is normally defined as an organism past metamorphosis from the larva, but not yet sexually mature. A common term used for O-year class bivalve mollusks is "spat", especially for organisms under 1 cm in shell length. There is no clear ecological division between juvenile and postlarvae, among bivalves, as described above, because postlarval behavior is apparently interchangeable with juvenile behavior, up to a point. It is also difficult to the researcher to determine the onset of sexual maturity of individual bivalves, which may occur at a small size, although Lucas (1975) has defined postlarval and juvenile sexuality in bivalve mollusks. Fisheries ecologists will usually simply describe animal size, in sampling programs.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 The early life history of bivalve mollusks, upon reviewing the strategies examined above, seems to be a compromise between parental care and larval production. Increased parental investment, such as egg size or brooding, decreases larval mortality, but also decreases initial larval production. Without further examination, this may appear to fit the "r/K" strategy continuum (see Horn and Rubenstein, 1984; for a review of r and K), but in fact other life history traits of bivalve mollusks do not match the r/K continuum. Species which produce hundreds of thousands or millions of small larvae with long planktonic periods tend to be large and long-lived as adults, with stable populations. Examples include Crassostrea (Stanley and Sellers, 1986), Mercenaria (Stanley, 1985), and Margaritifera (Bauer, 1987). The "r" model predicts relatively small, short-lived species, with highly variable population levels. Bivalve species which produce the fewest number of offsprings per individual are small and short­ lived, with rapidly fluctuating population levels. Examples include Gemma (Sellmer, 1967) and Transenella (Asson-Batres, 1988). The r and K life history models are, therefore, not easily applicable to most bivalve mollusks.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23 Life History Research When discussing research techniques, there are three separate inodes in the early life history of sedentary invertebrates: zygote production, planktonic phases, and benthic phases. "Zygote" is here defined as the production of either fertilized eggs, or the first larval or juvenile stage released from the female, in the case of brooding species. Planktonic phases can include eggs, larvae, and postlarvae, as discussed above, but benthic phases, at least among bivalve mollusks, are virtually always restricted to postlarvae or juveniles. For all three life history aspects, the small (microscopic) size of the eggs, larvae, or postlarvae is the primary research difficulty.

Zygote Production Estimation of zygote production in bivalve mollusks is usually approached by examining the reproductive status of the adults in a population. For taxa that release swimming larvae or metamorphosed juveniles (see previous discussion

of brooding in bivalves in this chapter) this is a rational method (e.g. Sellmer, 1967; Jansen and Hanson, 1991), but for species with planktonic fertilization, this techniques is uncertain. Examples of estimates of fecundity in species with planktonic fertilization include Galtsoff (1964), Calabrese (1969), Brousseau (1978), and Peterson (1986). Although examination of gonadal tissue prior to and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. following spawning can give a reasonable estimate of the number of eggs released, Pennington (1985) has shown that small differences in distance between spawning individuals (males and females) leads to large differences in fertilization success, and hence zygote production. For colonial bivalves, such as oysters in a reef, this factor would be less important than for solitary bivalves, such as deep-burrowing clams (e.g. Mercenaria, Mya). For these clams, estimates of zygote production would require knowledge of the effects of distance on fertilization success (via gamete age and dilution), and the spatial distribution of the spawning adults. Both of these questions can be done with techniques available in the literature, but since to this point "gamete ecology" has been studied for very few species, few generalizations can be made.

Planktonic Larval Abundance Aside from the small size of planktonic larvae, several serious difficulties have limited quantitative studies of larval abundance or survival in coastal waters. First, tidal and other currents and bottom topography move masses of water in unpredictable manners, and continually dilute the water in any particular system with waters from adjoining systems. It is therefore difficult to define a body of water which contains invertebrate larvae. Second,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 swimming larvae are able to control their position in a body of water to some extent, and are not randomly distributed (e.g. Perkins, 1931; Seliger et al., 1982; Mann, 1988; Tremblay and Sinclair, 1990; Newell et al., 1991). Furthermore, larvae of different ages may have different distributions in the plankton, on a scale of meters (discussed in Chapter 3 of this volume). Quantitative sampling schemes must therefore take into account planktonic distribution of the larvae. A third difficulty is that many populations spawn nearly continuously over a period equal or longer to the planktonic larval period (e.g. Stanley and Sellers, 1986; Shaw et al., 1988; Newell, 1989). Near- continuous spawning, combined with inherent variability in larval development period (Pechenik, 1990) can completely obscure cohorts of larvae, so that no estimates can be made of cohort abundance or mortality. These difficulties have apparently discouraged attempts to directly estimate larval mortality in the plankton, because very few plankton-based studies exist. Jorgensen (1981) and Yoo and Ryu (1985), however, have both used

direct examination of planktonic larvae to estimate planktonic mortality, and have arrived at similar values for bivalve larval mortalities, from the "D"-stage to the pediveliger. In both cases, the larvae were from populations that exhibited a single major spawning episode, and were within a partially enclosed embayment, so that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26 interchange with the ocean was limited. A distinct spawning peak is frequently available even in populations that can also spawn continuously (e.g. Newell, 1989), and in many bivalve species discrete spawning episodes are typical, cued by environmental conditions (e.g. LaSalle and de la Cruz, 1985). If the flushing rate and larval planktonic distribution are known, it is feasible to estimate both larval abundance and mortality of a cohort. Several authors have discussed the possibility of tracking larvae by the use of genetic markers, tracer, or

staining, in a mark-and-recapture scheme (reviewed by Levin, 1990). Common stains like neutral red last for a period of days and apparently do not harm bivalve larvae or significantly modify their behavior (Baker, 1991). As for previous studies, however, actual field applications of these techniques are rare, with a single study using a stain, and two studies using a genetic marker (Levin, 1990). The primary limitation is the huge number of marked larvae required for quantitative estimates upon recapture. The lower York River, VA, for example, contains over 2 00 million cubic meters of water. Larval identification to species is another problem that has discouraged some researchers. At the early veliger stages, there are very gross morphological differences between bivalve mollusk species, even for very distantly related taxa (e.g. Loosanoff et al., 1966; Chanley and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27 Andrews, 1971). Lutz et al. (1982) and Fuller and Lutz (1989) have shown that larval hinge characteristics can be used for identification at a very early age, but this requires electron microscopy, and is therefore not practical for rapid identification of many individuals. At the pediveliger stage, larval identification is much easier (Loosanoff et al., 1966; Chanley and Andrews, 1971), but still requires laborious hand-picking of samples under a microscope. In shallow coastal waters, towed plankton nets are rarely satisfactory for quantitatively sampling invertebrate larvae, because it is difficult to regulate their depth and they cannot be used close to the bottom. The preferred tool, therefore, is the plankton pump, which has changed little in overall design since Perkins (1931). Pumps are less efficient than nets for some large, fast swimming plankton such as chaetognaths (arrow worms), but equal to or

better than nets in capturing slow-swimming plankton, which include bivalve larvae. See Miller and Judkins (1981), Pillar (1984), Taggart and Leggett (1984), and Mehlenberg (1987) for comparisons of pumps and nets. Although many delicate plankton are damaged by pumps, bivalve larvae usually seem unaffected by passage through a high-speed

impeller (P. Baker, unpubl. data).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28 Postlarval Abundance Peterson (1978) outlined a perennial question in benthic ecology; are population patterns fixed at settlement or by subsequent mortality patterns? To approach this problem, the researcher must first understand the difference between "settlement" and "recruitment" of marine benthos. The term "recruitment" is used frequently in studies of populations of benthic invertebrates, and is sometimes confused with the term "settlement". Recruitment for an individual means the act of entering a life phase that has been defined by the researcher, while recruitment for a population means proportional survival up to the defined life stage. Common life stages defined as recruitment points include recruitment to the sexually mature population, recruitment to a fishery (usually a specific size), or recruitment to a new, well-defined stage. Recruitment to the benthos from the plankton is an example

of the latter, and in invertebrates, is synonymous with the term "settlement". A frequent error, however, is for researchers to claim to have measured "settlement", when in fact they have measured survival of settled individuals over the period between sampling events. These researchers have actually measured a type of recruitment. For example, if the sampling periods are one week apart, the researcher has measured (settlement on day 1 minus 6 days of mortality) + (settlement on day 2 minus 5 days of mortality)... etcetera.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29 If the recruits are measured daily, the error is probably small, but any period longer than that is suspect. Evidence of extremely high mortality- and mortality rate variability- in a period of days for bivalve mollusks is given by Powell et al. (1984, 1986), Harrold et al. (1991), and Roegner (1991). Examples of researchers who have implied or stated that their measurements of recruitment are equal to settlement levels include Dean and Hurd (1980), Gaines et a l . (1985), Peterson (1986), Shanks and Wright (1987), Thresher et al. (1989), and Buyanovski (1991). For invertebrates that settle on hard substrate, such as oysters or mussels, newly settled postlarvae, or "spat" are relatively easy to see and quantify, under magnification. For invertebrates that settle in sand or mud, however, the process is far more exhaustive, and in practical terms, quantitative sampling of newly settled postlarvae can normally be done only if the larvae settled in densities high enough that very small sample replicates (e.g. no more than 100 cm2) can be used to estimate population size. Newly settled postlarval clams in Chesapeake Bay are usually under 3 00 n in diameter, which is comparable to sand grain size (Chanley and Andrews, 1971). Furthermore, as discussed previously in this chapter, postlarval clams attach themselves by byssal threads to sand grains or larger particles, which makes sorting them mechanically difficult.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Once the size of a cohort of newly settled postlarvae has been estimated, the next problem for a researcher is to chose an appropriate time period for re-sampling, to determine subsequent mortality rates. The evidence is that mortality rates are initially extremely high; many bivalves do not successfully metamorphose upon settlement (Baker and Mann, in press), but for those that survive the first few days, subsequent mortality rates are lower (Roegner, 1991), and periods between sampling can increase with a low chance of missing critical information. In most natural populations, new settlement will occur continuously, however, and the day-cohort of interest must somehow be differentiated from subsequent day-cohorts. This is most commonly approached by size frequency analysis (e.g. Brousseau, 1978), but this requires large samples for any degree of accuracy, especially between cohorts only a day or two apart, and is still only approximate. In most cases, there is no simple answer to this problem.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31

Present Research The research outlined in the following chapters was

designed to a) determine abundance patterns and mortality rates at several early life history phases of a bivalve mollusk at, within a specific system (the lower York River,

Chesapeake Bay, Virginia), b) to determine general sources of high variation in abundance or survival within the early

life history, and c) design and test techniques to quantitatively estimate population levels at the life history phases of interest. The life history phases of interest were those of the competent larva and the benthic post-metamorphic phase, up to a about 1-2 mm, at which size they are reliably detected by existing benthic sampling schemes. The following general questions were used a research guidelines: 1) Of all late-stage, or pediveliger, larvae observed in the plankton, how many were actually available, or competent, to settle, within a given time period? This was a question which evolved during preliminary research on Question 2 (below), and is primarily descriptive research, although experimental work comparing wild larvae to hatchery-reared larvae is also described.

2) How many pediveliger larvae are present in the water column at a give site? The effects of several factors, and their influence on abundance variability, were examined.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32 The general null hypothesis will be of uniform or random distribution of late-stage larvae.

3) How closely does planktonic pediveliger larval abundance relate to subsequent settlement, and what is the residual variation? A null hypothesis of no relationship will be used as the basis of research.

4) From a known number of pediveliger larvae, how many settle and survive onto natural substrate, and how does a natural benthic community effect survival and variability? A general null hypothesis of no effects, relative to controls, will be examined. The American oyster, Crassostrea virginica (Gmelin, 1789) (Mollusca: Bivalvia: Ostreidae), was the primary model species for this research. Galtsoff (1964) provides a comprehensive review of the life history of this species. The lack of a postlarval stage (described above) within

Ostreidae simplifies the concept of settlement, in that possible crawling or byssal drifting movements do not need to subtracted from mortality estimates of benthic post- metamorphic cohorts. Studies on pediveliger larval abundance and distribution made use of some additional bivalve mollusk species (see Chapters 3, 4). The American oyster was once an ecologically important species in Chesapeake Bay, (for reviews of ecological role, see Bahr and Lanier, 1981; Newell, 1988; Dame and Libes, 1993) but in recent years its numbers have declined

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33 catastrophically. Reviewed by Hargis and Haven (1988) and Mann et al. (1991) the decline has continued since then (Barber, 1992). Crassostrea remains one of the most abundant large bivalve mollusks in many areas of Chesapeake Bay, however (this author, unpubl. data), and its pediveliger larvae are consistently among the most abundant species in summer months (see Chapter 3).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34 References Andre, C. , P.R. Jonsson, and M. Lindegarth. 1993. Predation on settling bivalve larvae by benthic suspension feeders: the role of hydrodynamics and larval behaviour. Mar. Ecol. Prog. Ser. 97:183-192. Asson-Batres, M.A. 1988. Reproduction and growth of the brooding bivalve Transenella tantilla. Veliger 30:257- 266. Babcock, R.C. 1993. The real thing: In situ measurements of fertilization rates from natural spawnings of marine invertebrates. (Abstract). Proceedings of the Larval Ecology Meetings, Port Jefferson, NY, p. 31. Bahr, L.K. and W.P. Lanier. 1981. The ecology of intertidal oyster reefs of the south Atlantic coast. U.S. Fish and Wildlife Service, Office of Biological Services. FWS/OBS-81/15. 105 pp. Baker, P. 1991. Effect of neutral red stain on settlement ability of oyster pediveligers, Crassostrea virginica. J. Shellfish Research 10:455-456. Baker, P. and R. Mann. 1991. Soft shell clam. 4.1-4.18. In Funderburk, S.L., J.A. Mihursky, S.L. Jordan, and D. Riley (eds.). Habitat Requirements for Chesapeake Bay Living Resources, 2nd Ed. Living Resources Subcommittee, Chesapeake Bay Program, Annapolis, MD. Baker, S.M. and R. Mann. In press. Description of metamorphic phases in the oyster, Crassostrea virginica, and effects of low oxygen on metamorphosis. Mar. Ecol. Prog. Ser. Barber, B.J. 1992. Oyster spatfall in Virginia waters: 1992 annual summary. Virginia Marine Resource Special Report, December 1992. 12 pp. Barnes, R.D. 1986. Invertebrate Zoology, 5th Ed. Saunders College Press, Philadelphia, PA. 893 pp. Bauer, G. 1987. Reproductive strategy of the freshwater pearl mussel Margaritifera margaritifera. J. Animal Ecol. 56:691-704. Bayne, B.L. 1964. Primary and secondary settlement in Mytilus edulis L. (Mollusca). J. Animal Ecology 33:513- 523 .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35 Bayne, B.L. 1971. Some morphological changes that occur at the metamorphosis of larvae of Mytilus edulis. 4th European Marine Biology Symposium: 259-280. Bayne, B.L. P.N. Salkeld, and C.M. Worrall. 1983. Reproductive effort and value in different populations of the marine mussel, Mytilus edulis L. Oecologia 59:18-26. Berg, C.J., Jr. 1971. A review of possible causes of mortality of oyster larvae of the genus Crassostrea in Tomales Bay, California. California Fish and Game 57:69-75. Berg, C.J., Jr. and P. Alatalo. 1985. Biology of the tropical bivalve Asaphis deflorata (Linne, 1758). Bull. Marine Science 37:827-838. Beukema, J.J. and J. de Vlas. 1989. Tidal-current transport of thread-drifting postlarval juveniles of the bivalve Macoma balthica from the Wadden Sea to the North Sea. Marine Ecology Progress Series 52:193-200. Bonar, D.B. 1976. Molluscan metamorphosis: a study in tissue transformation. American Zoologist 16:573-591. Borcherding, J. 1991. The annual reproductive cycle of the freshwater mussel Dreissena polymorpha Pallas in lakes. Oecologia 87:208-218. Bower, S.M. and G.R. Meyer. 1990. Atlas of anatomy and histology of larvae and early juvenile stages of the Japanese scallop (Patinopecten yessoensis). Canadian Special Publication of Fisheries and Aquatic Sciences 111. 51 pp. Breese, W.P. 1972. Ingestion of bivalve molluscan larvae by the polychaete annelid Polydora ligni. Veliger 14:274-275. Brousseau, D.J. 1978. Spawning cycle, fecundity, and recruitment in a population of soft-shell clam, Mya arenaria, from Cape Ann, Massachusetts. Fishery Bull. 76:155-166. Bunt, C., H.J. Maclsaac, and W.G. Sprules. 1991. Pumping rate capacities of juvenile Great Lakes Dreissena polymorpha (Pallas). (Abstract). Proceedings of the Second International Zebra Mussel Research Conference, Rochester, NY: p 16.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36 Buyanovski, A.I. 1991. Ecology of larvae and settlement of spat of Pacific mussels in Avacha Bay (Eastern Kamchatka). Soviet J. Marine Biology 16:189-195. Cake, E.W., Jr. 1983. Habitat suitability index models: Gulf of Mexico American oyster. U.S. Fish Wildl. Serv. FWS/OBS-82/10.57 Calabrese, A. 1969. Mulinia lateralis: molluscan fruit fly? Proc. Natl. Shellf. Assoc. 59:65-66. Calloway, C.B. and R.D. Turner. 1988. Brooding in the Teredinidae. 215-226. In Thompson, M.F., R. Sarojini, and R. Nagabushanam (eds.). Marine Biodeterioration: Advanced Techniques Applicable to the Indian Ocean. International Conference on Marine Biodeterioration, Goa, India, 1986. Carriker, M.R. 1951. Ecological observations on the distribution of oyster larvae in New Jersey estuaries. Ecological Monographs 21:19-38. Carriker, M.R. 1961. Interrelation of functional morphology, behavior, and autecology in early stages of the bivalve Mercenaria. J. Elisha Mitchell Scientific Soc. 77:168-241. Chanley, P. and J.D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia. Malacologia 11:45-119. Chevolot, L., J.-C. Cochard, and J.-C. Yvin. 1991. Chemical induction of larval metamorphosis of Pecten maximus with a note on the nature of naturally occurring triggering substances. Marine Ecology Progress Series 74:83-89. Chia, F.-S. and M.E. Rice (eds.). 1978. Settlement and Metamorphosis of Marine Invertebrate Larvae. Elsevier/North Holland Biomedical Press. New York, NY. Coon, S.L., W.K. Fitt, and D.B. Bonar. 1990. Competence and delay of metamorphosis in the Pacific oyster Crassostrea gigas. Marine Biology 106:379-387. Cragg, S.M. 1980. Swimming behavior of the larvae of Pecten maximus (L.) (Bivalvia). J. Mar. Biol. Assoc. U.K. 60:551-564.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37 Cranfield, H.J. 1973a. A study of the morphology, ultrastructure, and histochemistry of the foot of the pediveliger of Ostrea edulis. Marine Biology 22:187- 202. Cranfield, H.J. 1973b. Observations on the behaviour of the pediveliger of Ostrea edulis during attachment and cementing. Marine Biology 22:203-209. Crawford, C.M., W.J. Nash, and J.S. Lucas. 1986. Spawning induction, and larval and juvenile rearing of the giant clam, Tridacna gigas. Aquaculture 58:281-295. Culliney, J.L. 1974. Larval development of the giant scallop, Placopecten magellanicus (Gmelin). Biol. Bull. 147:321-332. Culliney, J.L. 1975. Comparative larval development of the shipworms Bankia gouldi and Teredo navalis. Marine Biology 29:245-251. Dame, R. and S. Libes. 1993. Oyster reefs and nutrient retention in tidal creeks. J. Exp. Mar. Biol. Ecol. 171:251-258. Dean, T.A. and L.E. Hurd. 1980. Development in an estuarine fouling community: the influence of early colonists on later arrivals. Oecologia 46:295-301. Diaz, R.J. 1973. Effects of brief temperature increases on larvae of the American oyster (Crassostrea virginica). J. Fisheries Research Board Canada 30:991-994.

Drew, G.A. 1896. The anatomy, habits, and embryology of Yoldia limulata. Ph.D. thesis, John Hopkins University, Baltimore, MD. 45 pp. Driscoll, E.G. 1968. Sublittoral attached epifauna development in Buzzards Bay, Massachusetts. Hydrobiologia 32:27-32. Dupuy, J.L. 1975. Some physical and nutritional factors which affect the growth and setting of the larvae of the oyster, Crassostrea virginica, in the laboratory. 319-331. In Vernberg, F.J. (ed.). Physiological Ecology of Estuarine Organisms. University of South Carolina Press, Columbia, SC.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38 Fay, C.W., R.J. Neves, and G.B. Pardue. 1983. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (Mid-Atlantic) — bay scallop. U.S. Fish Wildl. Serv. Biol. Rep. 82(11.12). 17 pp. Fioroni, P. 1982. Larval organs, larvae, metamorphosis, and types of development of mollusca — a comprehensive review. Zoologische Jahrbiicher. Abteilung Anatomie 108:375-420. Fitt, W.K., C.R. Fisher, and R.K. Trench. 1984. Larval biology of tridacnid clams. Aquaculture 39:181-195. Fitt, W.K., S.L. Coon., M. Walch, R.M. Weiner, R.R. Colwell, and D.B. Bonar. 1990. Settlement behavior and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supernatants. Marine Biology 106:389-394. Fitt, W.K., C.R. Fisher, and R.K. Trench. 1984. Larval biology of tridacnid clams. Aquaculture 39:181-195. Flyachinskaya, L.P., and E.Ye. Kulakovskiy. 1992. Larval development of the bivalve Mytilus edulis (Mytilida, Mytilidae). Hydrobiological Journal 28:59-66. Fuller, S.C. and R.A. Lutz. 1989. Shell morphology of larval and post-larval mytilids from the north-western Atlantic. J. Marine Biological Assoc. U.K. 69:181-218. Gaines, S.D., S. Brown, and J. Roughgarden. 1985. Spatial variation in larval concentrations as a cause of spatial variation in settlement for the barnacle Balanus glandula. Oecologia 67:267-272. Gallager, S.M. and R. Mann. 1986. Growth and survival of larvae of Mercenaria mercenaria (L.) and Crassostrea virginica (Gmelin) relative to broodstock conditioning and lipid content of eggs. Aquaculture 56:105-121. Galtsoff, P.S. 1964. The American oyster. Fishery Bulletin 64: 1-480. Goodwin, C.L. and B. Pease. 1989. Species profiles: Life histories and environmental requirements of coastal fish and invertebrates (Pacific Northwest)— Pacific geoduc clam. U.S. Fish Wildl. Serv. Biol. Rep. 82(11.120). 14 pp.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39 Grant, W.S. and Utter, F.M. 1988. Genetic heterogeneity on different geographic scales in Nucella lamellosa (Prosobranchia: Thaididae). Malacalogia 26:275-287. Gruffydd, Ll.D., D.J.W. Lane, and A.R. Beaumont. 1975. The glands of the larval foot in Pecten maximus L. and possible homologues in other bivalves. J. Mar. Biol. Assoc. U.K. 55:463-476.

Gustafson, R.G. and Lutz, R.A. 1992. Larval and early post-larval development of the protobranch bivalve Solemya velum (Mollusca: Bivalvia). J . Mar. Biol. Assoc. U.K. 72:383-402. Hargis, W.J., Jr., and D.S. Haven. 1988. The imperiled oyster industry of Virginia. Virginia Institute of Marine Science Spec. Rep. Applied Mar. Sci. Ocean Engin. 290. 130 pp.

Harrold, C., S. Lisin, K.H. Light, and S. Tudor. 1991. Isolating settlement from recruitment in sea urchins. J . Exp. Mar. Biol. Ecol. 147:81-94. Hines, A.H. 1986. Larval problems and perspectives in life histories of marine invertebrates. Bull. Marine Science 39:506-525. Hoese, H.D. 1973. Abundance of the low salinity clam, Rangia cuneata in southwestern Louisiana. Proc. Natl. Shell fisheries Assoc. 63:99-106. Horn, H.S. and D.I. Rubenstein. 1984. Behavioral adaptations and life history. 279-303. In Krebs, J.R. and N.B. Davies (eds.). Behavioral Ecology: An Evolutionary Approach. Blackwell Scientific Publications, Boston, MA. Hopkins, A.E. 1936. Ecological observations on spawning and early larval development in the Olympia oyster (Ostrea lurida). Ecology 17: 551-566. Jansen, W.A. and J.M. Hanson. 1991. Estimates of the number of glochidia produced by clams (Anodonta grandis simpsoniana Lea), attaching to yellow perch (Perea flavescens), and surviving to various ages in Narrow Lake, Alberta. Canadian J. Zoology 69:973-977.

J0rgensen, C.B. 1981. Mortality, growth, and grazing impact of a cohort of bivalve larvae, Mytilus edulis L. Ophelia 20:185-192.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40 Keck, R., D. Maurer, and R. Malouf. 1974. Factors inducing the settling behavior of larval hard clams, Mercenaria mercenaria. Proc. Natl. Shell fisheries Assoc. 64:59-67. King, C.A., C.J. Langdon, and C.L. Counts, III. 1986. Spawning and early development of Corhicula fluminea (Bivalvia: Corbiculidae) in laboratory culture. Amer. Malacol. Bull. 4:81-88. Kraeuter, J.N., M. Castagna, and R. Van Dessel. 1982. Egg size and larval survival of Mercenaria mercenaria (L.) and Argopecten irradians (Lamark). J . Exp. Mar. Biol. Ecol. 56:3-8. Lane, D.J.W. and J.A. Nott. 1975. A study of the morphology, fine structure and histochemistry of the foot of the pediveliger of Mytilus edulis L. J . Marine Biological Assoc. U.K. 55:477-495. LaSalle, M.W. and A.A. de la Cruz. 1985. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (Gulf of Mexico) — common rangia. U.S. Fish Wildl. Serv. Biol. Rep. 82 (11.31). 16 pp. Lasker, H.R. 1993. Variation in fertilization success of a broadcast-spawning gorgonian. (Abstract). Proceedings of the Larval Ecology Meetings, Port Jefferson, NY, p. 26. Levin, L.A. 1981. Bottoms up: juvenile terebellid polychaetes feeding in the water column. Bull. Southern California Academy Sciences 80:131-133. Levin, L.A. 1990. A review of methods for labeling and tracking marine invertebrate larvae. Ophelia 32:115- 144.

Loosanoff, V.L. and H.C. Davis. 1963. Rearing of bivalve mollusks. Advances in Marine Biology 1:1-13 6. Loosanoff, V.L., H.C. Davis, and P.E. Chanley. 1966. Dimensions and shapes of larvae of some marine bivalve mollusks. Malacologia 4:351-435. Lucas, A. 1975. Sex differentiation and juvenile sexuality in bivalve mollusks. Pub. Staz. Zool. Napoli 39 Suppl.:532-541. VIII European Marine Biology Symposium.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41 Lutz, R., J. Goodsell, M. Castagna, S. Chapman, C. Newell, H. Hidu, R. Mann, D. Jablonski, V. Kennedy, S. Siddall, R. Goldberg, H. Beattie, C. Falmagne, A. Chestnut, and A. Partridge. 1982. Preliminary observations on the usefulness of the hinge structures for identification of bivalve larvae. J . Shellfish Research 2:65-70. MacKenzie, C.L., Jr. 1981. Biotic potential and environmental resistance in the American oyster (Crassostrea virginica) in Long Island Sound. Aquaculture 22:229-268. McMahon, R.F. 1991. Mollusca: Bivalvia. 315-399. In Thorpe, J.H. and A.P. Covich (eds.). Ecology and Classification of North American freshwater Invertebrates. Academic Press, San Diego, CA. Mann, R. 1988. Distribution of bivalve larvae at a frontal system in the James River, Virginia. Marine Ecology Progress Series 50:29-44. Mann, R., E.M. Burreson, and P. Baker. 1991. The decline of the Virginia oyster fishery in Chesapeake Bay: Considerations for introduction of a non-endemic species, Crassostrea gigas (Thunberg, 1793). J. Shellfish Res. 10:379-388. Mann, R. and J.S. Rainer. 1990. Effect of decreasing oxygen tension on swimming rate of Crassostrea virginica (Gmelin, 1791) larvae. J. Shellfish Research 9:323-327. Mann, R. and C.C. Wolf. 1983. Swimming behavior of larvae of the ocean quahog Arctica islandica in response to pressure and temperature. Marine Ecology Progress Series 13:211-218.

Martel, A. 1993. Wind-induced drifting and settlement of early post-metamorphic zebra mussels. (Abstract). Proceedings of the Larval Ecology Meetings, Port Jefferson, NY, p. 29. Martel, A. and F.-S. Chia. 1991. Drifting and dispersal of small bivalves and gastropods with direct development. J. Experimental Marine Biology Ecology 150:131-147. Martel, A., A. Mathieu, S. Findlay, S. Nepszy, and J. Leach. 1993. Daily settlement rates in zebra mussels correlate with abundance of veligers. (Abstract). Proceedings of the Larval Ecology Meetings, Port Jefferson, NY, p. 29.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42 Mead, K.S. and M. Denny. 1993. The effects of turbulence on the fertilization and early development of the purple sea urchin, Strongylocentrotus pupuratus. (Abstract). Proceedings of the Larval Ecology Meetings, Port Jefferson, NY, p. 31. Mileikovsky, S.A. 1971. Types of larval development in marine bottom invertebrates, their distribution and ecological significance: a re-evaluation. Marine Biology 10:193-213. Miller, C.B. and D.C. Judkins. 1981. Design of pumping systems for sampling zooplankton, with descriptions of two high-capacity samplers for coastal studies. Biological Oceanography 1:29-56. Mohlenberg, F. 1987. A submersible net-pump for quantitative zooplankton sampling; comparisons with conventional net sampling. Ophelia 27:101-110. Mullen, D.M. and J.R. Moring. 1986. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (North Atlantic) — sea scallop. U.S. Fish Wildl. Serv. Biol. Rep. 82(11.67). 13 pp. Newell, C.R., H. Hidu, B.J. McAlice, G. Podniesinski, F. Short, and L. Kindblom. 1991. Recruitment and commercial seed procurement of the blue mussel Mytilus edulis in Maine. J. World Aquaculture Society 22:134- 152. Newell, R.I.E. 1988. Ecological changes in Chesapeake Bay: Are they result of overharvesting the American oyster, Crassostrea virginica? 536-546. In Lynch, M.P. and J.A. Mihursky (eds.). Understanding the Estuary: Advances in Chesapeake Bay Research. Chesapeake Research Consortium Publication 129.

Newell, R.I.E.. 1989. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (North and Mid-Atlantic) — blue mussel. U.S. Fish Wildl. Serv. Biol. Rep. 82(11.102). 25 pp. Ockelmann, K.W. 1964. Turtonia minuta (Fabricus), a neotenous veneracean bivalve. Ophelia 1:121-14 6. Ockelmann, W.K. 1965. Developmental types in marine bivalves and their distribution along the Atlantic coast of Europe. Proc. 1st European Malacological Conference, 1962: 25-35.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43 O Foighil, D. 1985. Form, function, and origin of temporary dwarf males in Pseudopythina rugifera (Carpenter, 1864) (Bivalvia, Galeommatacea). Veliger 27:245-252. O Foighil, D. 1989. Role of spermatozeugmata in the spawning ecology of the brooding oyster Ostrea edulis. Gamete Res. 24:219-228. Olafsson, E.B. 1988. Inhibition of larval settlement to a soft bottom benthic community by drifting algal mats: an experimental test. Marine Biology 97:571-574. Osman, R.W., R.B. Whitlatch, and R.N. Zajac. 1989. Effects of resident species on recruitment into a community: larval settlement versus post-settlement mortality in the oyster Crassostrea virginica. Marine Ecology Progress Series 54:61-73. Pascual, M.A., 0.0. Iribarne, E.A. Zampatti, and A.H. Bocca. 1989. Female-male interactions in the breeding system of the puelche oyster Ostrea puelchana d'Orbigny. J. Experimental Marine Biology Ecology 132:209-219. Pechenik, J.A. 1990. Delayed metamorphosis by larvae of benthic marine invertebrates: Does it occur? Is there a price to pay? Ophelia 32:63-94. Pechenik, J.A., and W.D. Heyman. 1987. Using KC1 to determine size at competence for larvae of the marine gastropod Crepidula fornicata (L.). J. Exp. Mar. Biol. Ecol. 112:27-38. Pennington, J.T. 1985. The ecology of fertilization of echinoid eggs: the consequences of sperm dilution, adult aggregations, and synchronous spawning. Biological Bull. 169:417-430. Perkins, E.B. 1931. Progress of oyster investigations in Barnegat Bay during 1930. Report of the Department of Biology of the New Jersey State Agricultural Experimental Station, 1931: pp 116-122. Peterson, C.H. 1978. Enhancement of Mercenaria mercenaria densities in seagrass beds: is pattern fixed during settlement season or altered by subsequent survival? Limnology and Oceanography 31:200-205. Peterson, C.H. 1986. Quantitative allometry of gamete production by Mercenaria mercenaria into old age. Marine Ecology Progress Series 29:93-97.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44 Pillar, S.C. 1984. A comparison of the performance of four zooplankton samplers. South African J. Marine Science 2:1-18. Powell, E.N., H. Cummins, R.J. Stanton, Jr., and G. Staff. 1984. Estimation of the size of molluscan settlement using the death assemblage. Estuarine, Coastal, and Shelf Science 18:367-384. Powell, E.N., R.J. Stanton, Jr., D. Davies, and A. Logan. 1986. Effect of a large larval settlement and catastrophic mortality on the ecological record of the community in the death assemblage. Estuarine, Coastal, and Shelf Science 23:513-525. Roegner, G.C. 1991. Temporal analysis of the relationship between settlers and early recruits of the oyster Crassostrea virginica (Gmelin). J . Experimental Marine Biology Ecology 151:57-69. Rumrill, S.S. 1990. Natural mortality of marine invertebrate larvae. Ophelia 32:163-198. Seed, R. 1976. Ecology. 13-65. In Bayne, B.L. (ed.). Marine Mussels: Their Ecology and Physiology. Cambridge University Press, New York, NY. Seliger, H.H., J.A. Boggs, R.B. Rivkin, W.H. Biggley, and K.R.H. Aspden. 1982. The transport of oyster larvae in an estuary. Marine Biology 71:57-72. Sellmer, G.P. 1967. Functional morphology and ecological life history of the gem clam, Gemma gemma (Eulamellibranchia: Veneridae). Malacologia 5:137-233. Shanks, A.L. and W.G. Wright. 1987. Internal-wave-mediated shoreward transport of cyprids, megalopae, and gammarids, and correlated longshore differences in the settling rate of intertidal barnacles. J. Experimental Marine Biology Ecology 114:1-13. Shaw, W.N., T.J. Hassler, and D.P. Moran. 1988. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (Pacific Southwest) — California sea mussel and blue mussel. U.S. Fish Wildl. Serv. Biol. Rep. 82(11.84). 16 pp. Sigurdsson, J.B., C.W. Titman, and P.A. Davies. 1976. The dispersal of young post-larval bivalve mollusks by byssal threads. Nature 262: 386-387.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45 Stanley, J.G. 1985. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (mid-Atlantic) — hard clam. U.S. Fish. Wildl. Serv. Biol. Rep. 82(11.41). 24 pp. Stanley, J.G. and M.A. Sellers. 1986. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (Gulf of Mexico) — American oyster. U.S. Fish. Wildl. Serv. Biol. Rep. 82 (11.64). 25 pp. Steinberg, P.D. and V.S. Kennedy. 1979. Predation upon Crassostrea virginica (Gmelin) larvae by two invertebrate species common to Chesapeake Bay oyster bars. Veliger 22:78-84. Strathmann, M.F. 1987. Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. 670 pp. Taggart, C.T. and W. C. Leggett. 1984. Efficiency of large-volume plankton pumps, and evaluation of a design for deployment from small boats. Canadian J. Fisheries Aquatic Sciences 41:1428-1435. Thresher, R.E., G.P. Harris, J.S. Gunn, and L.A. Clementson. 1989. Phytoplankton production pulses and episodic settlement of a temperate marine fish. Nature 341:641- 643. Tremblay, M.J. and M. Sinclair. 1990. Diel vertical migration of sea scallop larvae Placopecten magellanicus in a shallow embayment. Marine Ecology Progress Series 67:19-25. Turner, R.D. 1966. A survey and illustrated catalogue of the Teredinidae. Museum of Comparative Anatomy, Harvard University, Cambridge, MA. 265 pp.

Turner, R.D. and Y. Yakovley. 1983. Dwarf males in the Teredinidae. Science 219:1077-1078. Ulanowicz, R. 1988. Trophic structure of the Chesapeake mesohaline ecosystem. In Lynch, M.P. and J.A. Mihursky (eds.). Understanding the Estuary: Advances in Chesapeake Bay Research. Chesapeake Research Consortium Publication 129. 629 pp. Walne, P.R. 1963. Breeding of the Chilean oyster (Ostrea chilensis Philippi) in the laboratory. Nature 197:676.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46 Wells, H.W. 1961. The fauna of oyster beds, with special reference to the salinity factor. Ecological Monographs 31:239-266. Widdows, J., R.I.E. Newell, and R. Mann. 1989. Effects of hypoxia and anoxia on survival, energy metabolism, and feeding of oyster larvae (Crassostrea virginica, Gmelin). Biological Bull. 177:154-166. Yankson, K. 1986. Observation on byssus systems in the spat of Cerastoderma glaucum and C. edule. J. Marine Biological Association U.K. 66:277-292. Yoo, S.K. and H.Y. Ryu. 1985. Occurrence and survival rate of the larvae of Pacific oyster Crassostrea gigas in Hansan Bay. Bull. Korean Fisheries Soc. 18:47-476. Zimmerman, K.M., and J.A. Pechenik. 1991. How do temperature and salinity affect relative rates of growth, morphological differentiation, and time to metamorphic competence in larvae of the marine gastropod Crepidula plana? Biol. Bull. 180:372-386.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2

COMPETENCY TO SETTLE IN OYSTER LARVAE, Crassostrea virginica: WILD VERSUS HATCHERY-REARED LARVAE12

1 Contribution No. 1805 of the Virginia Institute of Marine Science

2 A version of this chapter has been accepted for publication in Aquaculture.

47

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48

Table of Contents

Page Abstract...... 49 Introduction...... 50 Materials and Methods...... 53 Results...... 58 Discussion...... 60 References...... 65 Tables...... 69

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49 Abstract Competency to settle in larvae of the oyster

Crassostrea virginica is defined as proportional settlement of a given number of larvae within 24 hours. It is distinguished from apparent competency, which is based on observed morphological characteristics. Competency of pediveliger larvae in lower York River, Chesapeake Bay, Virginia (taken by plankton pump from near the bottom of the water column) was 81.1% in 1990, and 79.9% in 1992. Larvae taken from midwater samples and near-surface samples showed a competency of 70.0 and 58.9%, respectively, but these values were not significantly different from the 1992 near­ bottom samples. There were no handing effects, or larval density effects, on competency, based on assays using hatchery-reared larvae. Settlement over three trials of hatchery reared larvae ranged from about 7% to 32% in 24 hours. These percentages are much lower than those for York River larvae, but similar to daily settlement reported by commercial oyster hatcheries elsewhere.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50 Introduction Dispersion of most marine bivalve mollusks is accomplished via a planktonic larval stage, followed by settlement to the benthos. Shortly before settlement and metamorphosis, the larvae develop morphological and physiological characteristics which enable them to explore the substrate, and then attach to it. This final period of larval development is termed "competency to settle", or simply "competency", and the larva itself is termed a "pediveliger" (after Carriker, 1961). Characters commonly used to determine competency include shell size and shape, the presence of a foot, and for some species, the presence of a pigmented eyespot. For oyster larvae (Ostreidae), all of these characters are useful, but since they all correlate closely with shell size, oyster culturists usually use size (as determined by a fine-mesh screen) to determine competency, only periodically checking for the presence of eyespots (Dupuy et al., 1977; Gibbons et al., 1992). In commercial oyster hatcheries, measured settlement rates are low relative to apparent (based on morphology) competency. Proportional settlements per day for hatchery reared larvae, calculated from literature and hatchery reports, range from 4% to 13% (Canzonier, 1989a,b), 10% to 13% (Dupuy et al., 1977), or up to 38% (Baker and Mann, 1992) for Crassostrea virginica; about 8% (Walne and Helm, 1974), 9% (Holiday et a l ., 1991) or up to 30% (Lipovsky,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51 1984) for C. gigas', and about 10% for Saccostrea commercialis (Holiday et al., 1991). More than 90% of the larvae in each case had the morphological characteristics of competency, and the entire behavioral sequence of settlement can occur in minutes (Cranfield, 1973; Coon et al., 1990). Oyster hatcheries allow up to eight days for settlement (Dupuy et al., 1977). Reported mortality rates of juveniles over a period of days, in the absence of predators, is less than 10% (Dupuy et al., 1977; Baker and Mann, 1992), and individuals dying during this period usually leave a valve attached to the substrate, readily observed under low magnification (pers. obs.). Reported settlement rates, therefore, are probably very close to actual settlement. All of the above infers that morphological characteristics alone do not adequately predict the ability or tendency of oyster larvae to settle. Unlike oysters, which cement one valve to the substrate, most other bivalves do not have a clearly defined time or visible sign of settlement, making it difficult to determine settlement rates. Eyster and Pechenik (1987) reported more than 50% settlement of laboratory reared larvae of the mussel Mytilus edulis in 24 hour under certain conditions, but mussels, like many others bivalves, may reenter the plankton after "settlement" (e.g., Bayne, 1964; Williams and Porter, 1971; Sigurdsson et al., 1976). In contrast to oyster settlement rates, laboratory reared

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52 barnacle larvae, Balanus amphitrite, have been induced to settle onto polystyrene plates at proportions of up to 81% in 24 hour (calculated from Maki et al., 1990). Like oysters, barnacles have planktonic larval development, and cement themselves to the substrate upon settlement. The problem of low proportional settlement is not only of interest to bivalve culturists, but raises the question of proportional settlement in nature, and its impact on the abundance of larvae available to settle from the plankton. Little or no quantitative data exist on proportional competency of wild oyster larvae, and if settlement of wild larvae is as low as reported for hatchery-reared larvae, settlement is a significant recruitment bottleneck for oysters. This paper describes research to examine the proportional settlement, or competency, of both wild and hatchery-reared Crassostrea virginica oyster pediveliger larvae, under similar conditions. The term "competency" will be used here to describe proportional settlement of "apparently competent" oyster larvae, within a 24 hour period, given the best possible laboratory settlement conditions. "Apparently competent" larvae are those with the morphological characteristics (size, shape, presence of foot and eyespot) associated with competency.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53 Materials and Methods The research site was the Virginia Institute of Marine

Science, in Gloucester Point, Virginia, situated at a narrow point in the York River estuary. The York River is a large estuary in the Virginia portion of Chesapeake Bay, poorly stratified in the downstream portions. Water temperatures during sampling periods were 25-29° C, salinities were 20-

25%o,and the water column (3 m) was always well-mixed at the study site (this author, unpubl. data). Larvae were collected in 1990 and 1992, using plankton pumps, from the end of a pier in the York River. In 1990, a modified submersible sump pump was used, situated 20 cm above the bottom. In 1992, a vertical array of three intakes was used, with one intake 2 0 cm below the surface ("near-surface), one 2 0 cm above the bottom ("near-bottom"), and one in between ("midwater"). Each intake was attached to a pump on the pier, and all three operated simultaneously. The volume of water pumped was about 1-3 m3 of seawater per sample. The seawater was pumped first

through a 400 fi mesh Nitex screen, and the pediveliger

larvae were subsequently retained on a 15 0 ju. screen. A hand-sprayer was used to transfer samples from the screen to a beaker. In the laboratory pediveliger larvae were identified under a dissecting microscope, and separated by micropipette. The presence of a clearly-developed eyespot was used to discriminate apparently competent larvae from

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54 pre-competent larvae, which were similar in most other morphological aspects. Despite the fact that the water column was well-mixed, oyster larval abundance in 1990 was clearly stratified, with the highest proportion of the pediveliger larvae near the bottom of the water column (this author, manuscript in preparation). Prior research has shown that oyster larvae prefer to settle on living or recently-living shells of conspecifics (Crisp, 1967; Vietch and Hidu, 1971), with a 1-3 day growth of marine bacteria (Fitt et al., 1990), under low light conditions (Ritchie and Menzel, 1969). In the laboratory, 2.5 cm circles of sterile but recently (within a few months) living oyster shell were placed in 0.5 n filtered seawater for 1-3 days prior to use. When a sample of oyster larvae was obtained, a shell was transferred to a culture dish with about 15 ml of filtered seawater, and the larvae were placed in the culture dish. Cultured alga, the flagellate Tahitian

Isochrysis galbana, was added to bring the cell concentration to about 20,000 ml'1 in the culture dish. This alga, used for larval nutrition in oyster hatcheries, was added to ensure that larvae did not fail to settle simply because ambient food levels were too low. Neutral red, a vital stain, was also added, to a concentration of about 30 ppm in the culture dish, to aid in locating larvae and newly-settled juveniles later. This stain has no effect on larval settlement (Baker, 1991). The culture dish was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55 covered, first with a watch glass, and then with an opaque cover, leaving the larvae and substrate in near-total

darkness, and left for 24 hours. This experimental duration was selected because it gave the larvae sufficient time to explore the substrate, but not enough time for significant maturation or mortality. At the end of 2 4 hours, all free- swimming, dead larvae, and settled juveniles were counted. Settlement for each sample was calculated as a proportion (percent). In 1990, 20 samples were taken, with six or more pediveliger larvae. In 1992, when no more than eight samples with pediveliger from any depth (near-bottom, midwater, and near-surface) were taken, larval abundance overall was much lower than 1990. Results of samples of fewer than five larvae were combined with results from the sample nearest in time (usually the following day), to reduce the effect of a single settlement event on proportional data. Proportional settlement of larvae from each depth in 1992 was determined separately. An unbalanced, one factor analysis of variance (Zar, 1984), was used to test for differences in proportional settlement from each depth. The power of the test was determined a posteriori (Zar, 1984). Although larvae visibly damaged by pump impellers or screens were rarely observed, an assay was developed to test the effect of handling on larval settlement. Soft-bodied

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56 animals collected during the sampling period, such as fish larvae or polychaete setiger larvae, were usually damaged or destroyed, but samples with towed plankton nets at the same site collected these organisms intact, so it is conceivable that tissue damage occurred to oyster larvae during pumping. Although hatchery-reared larvae are screened every several days, they do not normally experience the high pressure differences of a pump impeller, as did the larvae collected from the field. This assay also permitted an estimate of competency of hatchery-reared larvae under the experimental conditions described above. Crassostrea virginica pediveliger larvae were reared at the Virginia Institute of Marine Science oyster hatchery. The parent stock of the hatchery-reared larvae were collected in 1990 from Mobjack Bay, a portion of Chesapeake Bay contiguous with the York River. Larvae were fed Isochrysis galbana until about one week old, and thereafter were fed a mixture of Isochrysis and the diatom Thalassiosira weissflogii. Pediveliger larvae were obtained from the hatchery in 1990, and divided into two groups. One group was placed in a large bucket of

seawater, then pumped and screened, as were the larvae collected from the York River. This treatment was termed "handling". The other group was handled as little as possible ("no-handling" treatment). Twelve shells and culture dishes were prepared for each treatment as described above, except that no neutral red stain was used in the no­

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57 handling treatment. Using a micropipette, 100 larvae were placed in each culture dish, and given 24 hours to settle, as described above for the wild larvae. At the end of 24 hours, settlement in each dish was determined, and recorded as a proportion. Each value was transformed by the arcsine- square root method, so that a t-test could be used to determine the effect of handling on settlement (Zar, 1984). The power of the test was determined a posteriori (Zar, 1984). This assay was repeated three times, with three different cultures of larvae. The proportion of eyed pediveligers from each culture was > 95%. An assay was also performed to test the effect of pediveliger larval density on proportional settlement. This was done because oyster larvae are scarce in the York River, and most settlement trials using wild larvae were performed with 20 or fewer larvae. On the other hand, because settlement of hatchery reared pediveliger larvae was usually low (this paper), assays using these larvae were performed with 100 larvae, as described above. This lead to criticism that settlement of hatchery-reared larvae might be inversely density dependent. To test this, an assay was performed with two density treatments, 10 and 100 larvae per culture dish, respectively. Substrates, settlement period, sample size per treatment, and statistical analysis were performed as described above for the assay for handling effects.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58

Results In 1990, twenty field samples were assayed. No dead larvae were found in any assay, and non-settled larvae remained active. A total of 333 larvae were collected, with a mean sample size of 16.7. In 1992, seven samples were assayed from the near-bottom intake, with 54 larvae for a mean sample size of 7.7. The only mortality was one larva accidentally crushed, not included in the calculation of proportional settlement. Four samples were assayed from the midwater intake, with 45 larvae for a mean sample size of 11.3, and no mortality. Four samples were assayed from the near-surface intake, with 31 larvae for a mean sample size of 7.8, and no mortality. Table 1 summarizes the results for wild larvae. Analysis of variance was unable to detect a significant difference between proportional settlement of larvae from the three water depths in 1992, at a significance level of a = 0.05, although power of the test was low (0.30) due to low sample size. Proportional settlement of hatchery-reared larvae in the handling effects assay are given in Table 2. In none of the three trials was a significant difference between handling and no-handling treatments detected at a significance level of a = 0.05. Power of the test was low (0.52, 0.30, 0.38, for Trials 1, 2, and 3, respectively), primarily because the differences between treatments were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59 very small. No mortality in assays was observed, and non­ settled larvae remained active. Proportional settlement of hatchery-reared larvae in the density effects assay did not differ significantly between the two density treatments (10 and 100 larvae per culture dish). Mean proportional settlement at low density was 43.8%, with a standard deviation of 27.0%. Mean proportional settlement at high density was 56.7%, with a standard deviation of 12.7%. Power of the test was low (1 - /3 = 0.30), primarily because of the high standard deviation of the low density treatment, but in any case, there was no trend towards lower proportional settlement at higher densities.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60 Discussion Competency to settle (within 24 hours) of Crassostrea virginica pediveliger larvae from the York River appears to be high, but not equal to 100% of apparently competent larvae, based on morphological characteristics. Furthermore, at least for larvae from near-bottom water, proportional competency (about 80%) appears to have a fairly low variability (about 10% in 1990 and 18% in 1992), and is similar between years. This settlement rate is much higher than reported for hatchery-reared larvae in 24 hours. It should be pointed out that the values for competency

reported here are not absolute competency rates, but only an index based upon 24 hours. Given more time, more settlement would probably occur, since non-settled larvae in the assays were still active. Oyster culturists typically allow three or more days for hatchery-reared larvae to settle ( e.g., Walne and Helm, 1974; Dupuy et al., 1977; Lipovsky, 1984; Holiday et al., 1991).

Competency of larvae higher in the water column was not significantly different, although the actual competency values found in this study (about 7 0% in midwater and 56% in near-surface water) are lower than for larvae from near­ bottom water. The continuing and catastrophic decline of the oyster in Chesapeake Bay (Mann et al., 1991) has severely limited the availability of larvae for this study,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61 and thus the ability of the test to detect differences in competency within the water column. Despite rigorous handling of the larvae (pumping and sieving), no effects from this were detected in three trials of the assay to determine handling effects. The power of the tests were low primarily because the differences between

treatments were so small, relative to the variance. Interestingly, although larvae from the York River settled at high (80%) levels, hatchery-reared larvae settled at much lower levels (7-32%), similar to those reported in the literature for hatchery-reared oyster larvae. Gregariousness, or the tendency of settling pediveligers to seek out other settling larvae or recently- settled juveniles, has been reported for Crassostrea virginica (Vietch and Hidu, 1971; Hidu et al., 1978). Despite this, concern was raised that the observed differences in competency between hatchery-reared and wild pediveligers was due to crowding avoidance (inverse density dependence), so that the larger sample sizes in the assays using hatchery-reared larvae resulted in lower proportional settlement. The density assay showed that this was clearly not the case; proportional settlement was slightly higher (nonsignificant) at higher densities. The settlement rate

of the hatchery-reared pediveliger larvae during this time was higher (56.7%) than in 1990 trials (although still lower than that of wild larvae in 1990), but because of slightly

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62 altered hatchery techniques since 1990, is not directly comparable to 1990 settlement rates. Genetic differences between wild and hatchery-reared larvae is an unlikely explanation for the observed

differences in settlement, between hatchery-reared and wild larvae, for two reasons. First, the parent oysters of the hatchery-reared larvae were taken in 1990 from Mobjack Bay, which is contiguous with the lower York River, and therefore are part of the same population. Second, competency to settle is a critical aspect of oyster life history, highly correlated with fitness, and thus unlikely to be subject to high genetic variability. The larval rearing environment is the probable source of the difference in competency between wild and hatchery- reared larvae. Two major differences between the hatchery and the natural environment are; a) the types of mortality factors, or larval selection within a cohort (termed "culling" in oyster hatcheries); and b) larval diet. Water quality in the York River was stable throughout the sample period, (this author, unpubl. data), and therefore unlikely to have been a source of mortality to wild larvae. The nature of larval selection, or agents of mortality, differs between the hatchery and the natural environment. Hatcheries usually select (cull) for rapid growth, within a larval cohort, while larvae in the natural environment are

subjected to other mortality factors, particularly predation

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63 (see Rumrill, 1990, for review). It has not been demonstrated, however, that predation mortality of larvae, within a cohort, is non-random. Even if selection were non- random, it is difficult to see how selection for larval predator-avoidance traits would relate to selection for larval competency to settle. On the other hand, it can be argued that, because of continual predation in the plankton, rapid growth to competency is a favorable trait in the natural environment, as well as in the hatchery. Larval selection differences between the hatchery and the natural environment, therefore, do not seem as likely as an explanation for the difference in competency rates as does diet differences. Larval diet is the most obvious and best defined difference between the hatchery and the natural environment. Numerous studies have been performed on growth and survival of oyster larvae fed various cultured algal diets (e.g., Dupuy, 1975; Helm, 1977; Nascimento, 1980; Laing and Millican, 1986; Tan Tiu et al., 1989). Helm (1977) and Tan Tiu et al. (1989) demonstrated that some cultured diets favored oyster larval settlement over others, while Gallager et al. (1986) and Utting (1986) reported that increased dietary lipid and protein, respectively, enhance oyster larval settlement rates. Oyster larvae in Chesapeake Bay feed selectively upon a wide range of taxa (Fritz et al., 1984; Baldwin and Newell, 1991), and may select food items

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with nutrition content that favors the development of competency. It is also possible that there are minor

nutrients (i.e., vitamins or trace minerals) essential to the development of competency, which are present in natural plankton communities, but at below optimal levels in many cultured algae. At present the biochemistry of specific organ systems of oyster larvae is poorly known, and more research is required before the nature of these hypothesized nutrients can be speculated upon. Nonetheless, existing evidence strongly implicates diet as an important factor in competency to settle, and the most likely of various explanations for the difference in proportional settlement, observed in this study, between wild and hatchery-reared larvae.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65

References Baker, P., 1991. The effects of neutral red stain on settlement of oyster pediveligers, Crassostrea virginica (Gmelin, 1791). J. Shellfish Res., 10:455- 456. Baker, S.M. and Mann, R., 1992. Effects of hypoxia and anoxia on larval settlement, juvenile growth, and juvenile survival of the oyster Crassostrea virginica. Biol. Bull., 182:265-269. Baldwin, B.S., and Newell, R.I.E., 1991. Omnivorous feeding by planktotrophic larvae of the eastern oyster Crassostrea virginica. Mar. Ecol. Prog. Ser., 78:285- 301.

Bayne, B.L., 1964. Primary and secondary settlement in Mytilus edulis L. (Mollusca). J. Animal Ecol., 33:513- 523. Canzonier, W.J., 1989a. Maurice River Oyster Culture Foundation Summary Progress Report Jan. 12, 1989. 1 p. Canzonier, W.J., 1989b. Maurice River Oyster Culture Foundation Summary Progress Report Sept. 27, 1989. 1 p. Carriker, M.R., 1961. Interrelation of functional morphology, behavior and autecology in early stages of the bivalve Mercenaria mercenaria. J. Elisha Mitchell Scientific Soc., 77:168-241. Coon, S.L., Fitt, W.K., and Bonar, D.B., 1990. Competence and delay of metamorphosis in the Pacific oyster Crassostrea gigas. Mar. Biol., 106:379-387. Cranfield, H.J. 1973. Observations on the behavior of the pediveliger of Ostrea edulis during attachment and settlement. Mar. Biol., 22:203-209. Crisp, D.J., 1967. Chemical factors inducing settlement in Crassostrea virginica (Gmelin). J. Animal Ecol., 36:329-335. Dupuy, J.L., 1975. Some physiological and nutritional factors which affect the growth and settling of the larvae of the oyster, Crassostrea virginica, in the laboratory. 319-331. In Vernberg, J.F. (ed.). Physiological Ecology of Estuarine Organisms. University of South Carolina Press, Columbia, S.C.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66 Dupuy, J.L., Windson, N.T., and Sutton, C.E., 1977. Manual for design and operation of an oyster seed hatchery for the American oyster Crassostrea virginica. Virginia Institute of Marine Science, Special Report in Applied Marine Science and Ocean Engineering, #142; 104 pp. Eyster, L.S. and Pechenik, J.A., 1987. Attachment of Mytilus edulis L. larvae on algal and byssal filaments is enhanced by water agitation. J . Exp. Mar. Biol. Ecol., 114:99-110. Fitt, W.K., Coon. S.L., Walch, M., Weiner, R.M., Colwell, R.R., and Bonar, D.B., 1990. Settlement behavior and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supernatants. Mar. Biol., 106:389-394. Fritz, L.W., Lutz, R.A., Foote, M.A., Van Dover, C.L., and Ewart, J.W., 1984. Selective feeding and grazing rates of oyster (Crassostrea virginica) larvae on natural plankton assemblages. Estuaries, 7:513-518. Gallager, S.M., Mann, R . , and Sasaki, G.C., 1986. Lipid as an index of growth and viability in three species of larvae. Aquaculture, 56:81-103. Gibbons, M., Kurkowski, K., and Castagna, M., 1992. VIMS hatchery operations manual. Virginia Institute of Marine Science, Special Report in Applied Marine Science and Ocean Engineering, #318; 44 pp. Helm, M.M., 1977. Mixed algal feeding of Ostrea edulis larvae with Isochrysis galbana and Tetraselmis suecica. J. Mar. Biol. Assoc. U.K., 57:1019-1029. Hidu, H., Valleau, W.G., and Vietch, F.P., 1978. Gregarious setting in European and American oysters— response to surface chemistry versus waterborne chemicals. Proc. Natl. Shell fisheries Assoc., 68:11-16. Holiday, J.E., Allan, G.L., and Frances, J., 1991. Cold storage effects on setting of larvae of the Sydney rock oyster, Saccostrea commercial is, and the Pacific oyster, Crassostrea gigas. Aquaculture, 92:179-185. Laing, I., and Millican, P.F., 1986. Relative growth and growth efficiency of Ostrea edulis L. spat fed various algal diets. Aquaculture, 54:245-262. Lipovsky, V.P., 1984. Oyster egg development as related to larval production in a commercial hatchery. Aquaculture, 39:229-235.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67 Maki, J.S., Rittschof, D. , Samuelsson, M.-O., Szewzyk, U., Yule, A.B., Kjelleberg, S., Costlow, J.D., and Mitchell, R., 1990. Effect of marine bacteria and their exopolymers on the attachment of barnacle cypris larvae. Bull. Mar. Sci., 46:499-511. Mann, R., Burreson, E.M., and Baker, P., 1991. The decline of the Virginia oyster fishery in Chesapeake Bay: Considerations for introduction of a non-endemic species, Crassostrea gigas (Thunberg, 1793). J. Shellfish Res., 10:379-388. Nascimento, I.A., 1980. Growth of the larvae of Crassostrea gigas Thunberg, fed with different algal species at high cell concentrations. J. Cons. Int. Explor. Mer., 39:134-139. Ritchie T.P. and Menzel, R.W. , 1969. Influence of light on larval settlement of American oysters. Proc. Natl. Shellf. Assoc., 59:116-120. Rumrill, S.S., 1990. Natural mortality of marine invertebrate larvae. Ophelia, 32:163-198. Sigurdsson, J.B., Titman, C.W., and Davies, P.A., 1976. The dispersal of young post-larval bivalve molluscs by byssal threads. Nature, 262:386-387. Tan Tiu, A., Vaughan, D., Chiles, T. , and Bird, K., 1989. Food value of eurytopic microalgae to bivalve larvae of Cyrtopleura costata (Linnaeus, 1758) , Crassostrea virginica (Gmelin, 1791), and Mercenaria mercenaria (Linnaeus, 1758). J. Shellfish Res., 8:399-405. Utting, S.D., 1986. A preliminary study on growth of Crassostrea gigas larvae and spat in relation to dietary protein. Aquaculture, 56:123-138. Vietch, F.P. and Hidu, H., 1971. Gregarious setting in the American oyster Crassostrea virginica Gmelin: I. Properties of a partially purified "setting factor". Chesapeake Science, 12:173-178. Walne, P.R. and Helm, M.M., 1974. The routine culture of the Pacific oyster (Crassostrea virginica) at Conwy during 1973. Great Britain Ministry of Agriculture, Fisheries, and Food. Shellfish Information Leaflet No. 32. 10 pp. Williams, A.B. and Porter, H.J., 1971. A ten-year study of meroplankton in North Carolina estuaries: occurrence of postmetamorphal bivalves. Chesapeake Science, 12:26-32.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Zar, J.H., 1984. Biostatistical Analysis. 2nd Ed. Prentice Hall, Englewood Cliffs, NJ. 718 pp.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69

Table 1 Summary of proportional settlement of Crassostrea virginica larvae from the York River. # of Total Proportional Standard Year______Depth_____Samples Larvae Settlement Deviation

1990 near-bottom 20 333 81.1% 9.7% 1992 near-bottom 7 54 79.9% 18.1% 1992 midwater 4 45 70.0% 17.9% 1992 near-surface 4 31 56.1% 24.2%

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70

Table 2 Proportional settlement of hatchery-reared Crassostrea virginica larvae. Sample size for each treatment (handled versus non-handled) is 12. Standard deviations for each result are shown in parentheses. Handled Non-handled Trial 1 13.6% (6.6%) 11.6% (6.9%) Trial 2 11.0% (4.4%) 6.6% (5.5%) Trial 3 32.4% (12.6%) 27.6% (6.6%)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3.

TEMPORAL AND SPATIAL VARIABILITY IN ABUNDANCE OF LATE STAGE BIVALVE LARVAE IN CHESAPEAKE BAY

71

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72

Table of Contents

Page Abstract...... 73 Introduction...... 74 Materials and Methods...... 76 Research Site...... 76 Sampling: Abundance over Time and Water Chemistry...... 76 Sampling: Abundance over Tide and Depth...... 78 Sampling: Abundance acrossTidal Currents...... 79 Analysis: Abundance over Time and Water Chemistry...... 80 Analysis: Abundance over Tide and Depth...... 82 Analysis: Abundance across Tidal Currents...... 85 Results...... 86 Species Collected...... 86 Abundance over Time andWater Chemistry...... 87 Abundance over Tide andDepth ...... 89 Abundance across Tidal Currents...... 90 Discussion...... 92 References...... 101 Tables...... 106 Figures...... Ill Appendix A ...... 118 Appendix B ...... 119

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73 Abstract Several stationary sampling designs were used to examine the variation in abundance of late stage bivalve larvae, on temporal and spatial coordinates, in the York River of Chesapeake Bay, Virginia. Primary species examined included the clam Cyrtopleura costata, the oyster Crassostrea virginica, the shipworm Bankia gouldi, and the mussel Geukensia demissa. Abundance over time, whether on a scale of days or hours, showed fluctuations of one to two orders of magnitude between samples. These fluctuations in abundance appeared random, and correlated weakly or not at all with water temperature, salinity, time of day, and tidal phase, but in most samples different species showed parallel patterns of abundance. Most of the species examined showed a strong vertical distribution pattern at all times, however, with fewest late stage larvae near the surface. Geukensia, however, showed the reverse, with more late-stage larvae near the surface than lower in the water column. On a scale of meters, horizontal variation in abundance on an axis perpendicular to the tidal current was low.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74 Introduction For researchers studying population dynamics of many marine organisms, including bivalve mollusks, one of the most difficult aspects to quantify is the planktonic larval stage. Part of the reason for this is the limited knowledge of advective loss of larvae from their site of origin; the few attempts to follow cohorts of larvae through time have been conducted in relatively closed systems (Jorgensen, 1981; Yoo and Ryu, 1985). Furthermore, larvae are not distributed evenly throughout a given body of water, but may vary with depth (e.g. Nelson, 1927, 1928; Carriker, 1951; Yoo et a l ., 1977, 1979; Seliger et al., 1982; Tremblay and Sinclair; 1990), hydrographic conditions (e.g. Perkins, 1931, 1932; Carriker, 1951; Mann, 1988; Grigor'eva and Regulev, 1992), time or tide (Tremblay and Sinclair, 1990; Sekiguchi et al ., 1991), or randomly within those variables (e.g. Newell et al ., 1991; Sephton and Booth, 1992). Reported patterns differ for species and the body of water involved.

The primary reason to quantify larvae is to relate larval abundance in some way to either parent stock or subsequent juvenile recruitment. If the emphasis is on juvenile recruitment, one can largely ignore the question of larval mortality rates in the plankton by focusing on abundance of the final stage larvae, at or near a recruitment site. What remains is the problem of late stage

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75 larval abundance in the water carried by currents over the recruitment site, during the recruitment period. The research described in this paper used the final stage, or pediveliger, larvae of several species of bivalve mollusk to examine the abundance patterns of larvae in a shallow estuary of Chesapeake Bay, Virginia. Specific aspects of abundance variability included time, on scales of days and hours, tides, depth, and small-scale spatial, on the order of meters.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76 Materials and Methods Research Site The study site was at Gloucester Point, on the seaward side of the U.S. Hwy. 17 bridge, in the York River, an estuary within Chesapeake Bay, Virginia (see Figure 1). Sampling was conducted in 3 m of water (mean low tide), at the end of a pier. Tidal currents flowed perpendicular to the long axis of the pier at a mean velocity of about 8 cm-s'1 near the bottom, and the water column was unstratified throughout the study (this author, unpubl. data).

Sampling: Abundance over Time and Water Chemistry In 1990 a single submersible pump, modified from a sump pump rated at 1000 gallons (3785 liters) per hour, was used. The pump was enclosed in a plastic-coated wire basket that held it 2 0 cm above the bottom, and lowered from the end of the pier. Preliminary research showed that 2 0 cm was approximately as close as the pump could be held above the substrate without entraining benthic particles under some conditions. Seawater was pumped through a 4 00 n mesh Nitex screen to filter out large particles, and the samples were retained on a 150 n mesh. Both meshes were held by a segment of 6 inch PVC (polyvinyl chloride) pipe and two sleeve fittings. At the end of the sample, which lasted from 30 to 90 minutes, the sample was rinsed into a beaker with a hand sprayer. The approximate volume sampled was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77 calculated from the duration of sampling, over the mean pumping rate (pumping rate calculated at start and finish of each sample by measuring time required to fill a known volume). Daily mean water salinities and temperatures are recorded at the sample site by the Virginia Institute of Marine Science, Department of Oceanography. Mean daily salinities were 18.1-20.4 %oin 1990 (July 5 - Sept. 9); 19.9- 21.4%oin 1991 (July 17 - Sept. 5); 19.1-23.6%oin 1992 (June 29 - Sept. 7); and 22-24%oin 1993 (August 28-30) during summer samples. Mean daily temperatures were 2 6.5-28.9°C in 1990; 26.5-29.3°C in 1991; 24.1-27.7°C in 1992; and 27.5-

28.5° in 1993, during summer samples, with highest temperatures in late July and early August. V7ater currents were recorded with an Inter-Oceanics Model S-4 electromagnetic current meter during a neap tide series in October 1993. Tidal currents and heights at that time were very close to predicted, due to stable meteorological conditions. Over one full tidal cycle, mean near-bottom current velocity was 7.9 cm-s'1. When a vertical

tow was made at a near-bottom velocity of 7.8 cm-s'1 (flood tide), surface velocity was 11.0 cm-s'1, velocity at 1 m depth was 17.4 cm-s'1, and velocity at 2 m was 10.7 cm-s'1.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78 Sampling: Abundance over Tide and Depth In 1991 and 1992, three pumps were used to take three samples simultaneously. For sampling down through the water column, a piling made from 4 inch PVC pipe was driven into the sediment three m away from the end of the pier, and plankton samples were taken at this piling. Three pumps, each rated at 1200 gallons (4542 1) per hour, were situated on the pier, with hoses running to intakes attached to the PVC piling. One intake was about 2 0 cm above the bottom in 1991 (near-bottom), but in 1992 storms scoured the bottom, increasing the distance to 30 cm. Another intake floated about 20 cm below the surface (near-surface), and a third could be adjusted over a range of depths between these two (midwater) (Figure 2). The end of each intake was protected by a plastic ball perforated with many 6 mm holes, to prevent entrainment of large debris. All three intakes could be lifted clear of the water in between uses, to avoid fouling. Screening, and volume estimations were made as described above for 1990. Tidal phase and time of day were recorded at each sample interval. A fourth intake was used to examine plankton at the very bottom of the water column in 1992 and 1993. This intake was fastened to an Plexiglas acrylic plate, 50 cm in diameter. A flat lead weight of approximately 4 kg was fastened to the bottom of the acrylic plate, and the intake was protected by a plastic hemisphere, with perforations at

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79 the junction of the hemisphere and the acrylic plate. A very similar device was first used by Carriker (1951), differing significantly in design only in that Carriker's sampler had a 1 cm-wide, upward-directed flange, at the rim of the benthic plate. For identification elsewhere in this document, this apparatus will be termed a "Carriker sampler". This apparatus rested on the bottom during use. During a sampling episode, the Carriker sampler was operated in conjunction with the near-bottom intake, which was raised to 0.5 m above the bottom, and the midwater intake, which was lowered to 1 m above the bottom.

Sampling: Abundance across Tidal Currents To sample the plankton on a horizontal scale, three pumps were again used, but the three intakes were spaced at 5 meter intervals, starting 5 meters away from the end of the pier and in a line perpendicular to tidal current vectors. Each intake was a segment of 2 inch PVC pipe, perforated with many 6 mm holes for a length of 15 cm. A weight at one end and a float at the other held each intake vertically, about 2 0 cm above the substrate (Figure 3). Mohlenberg (1987) found no avoidance of a plankton pump intake by bivalve mollusk larvae, which swim slowly compared to many zooplankton species. Plankton pumps have been used by a number of researchers to sample bivalve larvae quantitatively (e.g. Carriker, 1951, 1961; Tremblay and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80 Sinclair, 1990, 1992; Newell et al., 1991; Sekiguchi et al., 1991). Plankton samples were examined under a dissecting microscope with the aid of a Ward plankton counting wheel, and pediveliger larvae were counted by species. Identification to species or developmental stage was assisted by Jorgensen (1946), Sullivan (1948), Rees (1950), Loosanoff et al. (1966), Chanley and Andrews (1971), Culliney (1975), Fuller and Lutz (1989), and Kennedy et al. (1989). In some cases (Geukensia demissa, Anadara transversa, and Cyrtopleura costata) larvae were cultured in the laboratory, using techniques suggested by Loosanoff and Davis (1963), Gustafson et al. (1991), until they settled and grew to a size large enough to identify to species.

Analysis: Abundance over Time and Water Chemistry Samples were collected four times daily in 1990 using the single near-bottom pump, at approximately 00:00 (midnight), 06:00 (morning), 12:00 (noon), and 18:00 (evening), Eastern Daylight Time. Pediveliger larvae were identified and counted, and abundance per m3 was calculated. All samples taken in 1990 were used to examine the

relationship of temperature and salinity to the abundance of pediveligers of the three most abundant species (Cyrtopleura, Crassostrea, Bankia). Abundance data were transformed by logarithm of (X + 1) (Zar, 1984), and cross-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81 correlational analysis was used to examine the effect of temperature or salinity up to seventeen days prior to the commencement of the sample period (before that, there was a long gap in the temperature and salinity data). Cyrtopleura metamorphose in 16-21 days following fertilization under laboratory conditions at 30°C (Gustafson et al., 1991);

Crassostrea are competent to settle in the Virginia Institute of Marine Science oyster hatchery 13-15 days following fertilization, in water of 24-28°C. (V. Shaffer, Virginia Inst. Marine Science, unpubl. data); and Bankia are competent to settle 2 0 days after release from brood chambers under laboratory conditions at 20°C (Mann and

Gallager, 1985). Seventeen days prior to the commencement of sampling, therefore, encompassed most, if not all, of the planktonic phases, for the species analyzed. All samples taken in 1990 were combined to test for independence of abundance between the three most abundant

species (Cyrtopleura, Crassostrea, Bankia). Pediveliger abundance data were transformed by logarithm of (X + 1) (Zar, 1984), and partial multiple correlation analysis with t-tests (Steel and Torrie, 1980) was used to test null hypotheses of independence within pairs of the three most abundant species. For this, as for other statistical tests, the confidence level a was set at 0.05. Samples taken in 1991, at three depths, were analyzed in the same manner, for each depth, and for the same three species as above.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82 Abundance data from 199 0 were converted to proportions of daily totals (pediveliger abundance per m3 summed over

midnight, morning, noon, and evening). The proportions were transformed by arcsine of the square root (Zar, 1984), and one-factor analysis of variance (Zar, 1984) was used to test the null hypothesis of no difference in distribution over the four sample times (morning, noon, evening, night) for each species. Frequently a sample was missed due to weather or other factors; analysis was completed only for days in which all four samples were taken, and only in which larvae of any given species were present in two or more of the four samples.

Analysis: Abundance over Tide and Depth In 1991, sampling was continuous over a series of tidal cycles (high tide to following high tide), broken into eight equal time periods of about 1.5 hours, and simultaneously at three depth (near-surface, midwater, near-bottom). The tidal time periods were numbered, with #1 including high water, #5 including low water, and #8 ending just prior to the following high water. Simultaneous samples were taken at three depths; near-surface, midwater, and near-bottom. Abundance was recorded for each species as pediveligers per

m3, and the data were transformed by logarithm of (X + 0.1).

Three-factor analysis of variance with incomplete crossing was used to test the relationship between tidal period (1-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83 8), depth, and pediveliger abundance; the third factor, day of sampling (1-4), was included to account for between-day abundance differences, and was not included in the model for interactions.

In 1992 and 1993, sampling was also done over three depths, but not over tidal cycles. In 1992, two sets of depths were sampled; near-surface, midwater, and near­ bottom, as in 1991; and separately at 0 m (using the Carriker sampler), 0.5 m, and 1 m above the benthos. In 1993, only the three near-bottom set of depths (0, 0.5, 1.0 m) were sampled. About equal numbers of day and night samples were taken in both cases. Abundance at each depth was converted to proportions of all pediveligers (of each species) at each depth. The proportion of larvae at each depth was arcsine-square root transformed, and one-factor analysis of variance was used to examine the effect of depth on larval distribution, separately for whole water column and near-bottom sampling series. Data in 1991, taken at the near-bottom intake over tidal cycles, was used to examine the effect of tidal phase

and time of day on pediveliger distribution. Abundance at each sample period was expressed as a proportion of pediveliger larvae (of that species) over all three depths (near-surface, midwater, near-bottom), and for data analysis, was arcsine-square root transformed. Only near­ bottom samples were used, for two reasons; first, larvae

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 84 were most abundance in near-bottom samples, and second, depth distribution was consistent over time (this paper). Tidal phase was divided into four (instead of eight) periods, to increase cell replication. Each tidal period therefore had two sequential replicates, for each day sampled. The time of day was classified either as day (noon to 9:00 pm) or night/morning (9:00 pm to noon the following day). These times were chosen because data for three species from 1990 indicated that afternoon (6:00), following the period of greatest surface water light and warmth, had consistently lowest near-bottom abundances of pediveligers, although the trend was significant only for Cyrtopleura (this paper). To decrease cell inequality of the two-factor analysis of variance, samples taken for the three hours following noon (the warmest period of air temperatures for the day) were added to afternoon/evening samples. Unbalanced two-factor analysis of variance (Zar, 1984) was used to test the effect of time of day and tidal phase on the proportion of larvae in near-bottom waters.

Data from the Carriker sampler and two other near­ bottom intakes, collected in 1992 and 1993, were examined for the effect of distance from the benthos and tidal phase on larval distribution. A one factor analysis of variance was used to test the null hypothesis of no difference between depths on proportions of larval collected in each sample (Zar, 1984).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85 Analysis: Abundance across Tidal Currents Samples taken simultaneously at three near-bottom intakes placed on a horizontal line, normal to tidal currents, were used in 1992 to examine small-scale spatial variability in pediveliger larval abundance. Data was transformed by logarithm of (X + 0.1), and one-factor analysis of variance was used to test the null hypothesis of no difference in abundance (per species) between three intakes. The variability was standardized by dividing the standard deviation of total abundance per m3 sampled, by the mean of total abundance, for each sample period, since the mean abundance varied by a factor of ten or more between samples. This standardized variability was regressed against duration (minutes) of sample time, to test the effect of duration of sampling on horizontal spatial variability. Regression analysis (Zar, 1984) was used to test the null hypothesis of no trend.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86 Results Species Collected Eight species or genera of bivalve larvae were positively identified from samples: the ark clam Anadara transversa (Say, 1822) (Arcidae); the mussel Geukensia demissa (Dillwyn, 1817) (Mytilidae); the jingle shell Anomia simplex Orbigny, 1842 (Anomiidae); the American oyster

Crassostrea virginica (Gmelin, 1791) (Ostreidae); the tellin Tellina agilis Stimpson, 1857 and tellinid clams of the genus Macoma (Tellinidae); the angel wing clam Cyrtopleura costata (Linne, 1758) (Pholadidae); and the shipworm Bankia gouldi Bartsch, 1908 (Teredinidae). Of these, only Geukensia, Crassostrea, Cyrtopleura, and Bankia were used in statistical analysis. Anadara, Anomia and Tellina were not found in sufficient numbers for statistical analysis in any year. Geukensia were found in sufficient numbers for statistical analysis in 1992 only. Crassostrea, Cyrtopleura, and Bankia were present at sufficient levels for analysis in all three years. Macoma species were the most abundant bivalve in the plankton, but were not used in statistical analysis for two reasons. First, the Macoma were probably a mixture of M. balthica (Linne, 1758) and M. mitchelli Dali, 1895 (Baker, manuscript in prep.), but the ratios of the two species in the plankton could not be determined. Both were present as adults at the sampling site and abundant in adjacent tidal

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87 creeks; M. balthica adults were about twice as common as M. mitchelli in benthic surveys. Larvae and postlarvae of M. mitchelli (Kennedy et al., 1989) closely resemble those of M. balthica (Jorgensen, 1946; Sullivan, 1948), and Macoma individuals could not be reared to juveniles in the laboratory for further identification. Second, a large percentage of the Macoma present were planktonic postlarvae (Baker, manuscript in prep.), and were difficult to distinguish from pediveligers. This is a life history pattern also described for M. balthica in the Wadden Sea in Europe (Beukema and de Vlas, 1989).

Abundance over Time and Water Chemistry In 199 0, pediveligers of the oyster Crassostrea virginica, the clam Cyrtopleura costata, and the shipworm Bankia gouldi were collected in sufficient numbers for statistical analysis. Over a period of days or weeks (100 total samples) within the annual reproductive period, abundance of pediveliger larvae any single species in the lower portion of the water column appeared random (Figures 4a-c). Temperature and salinity varied only slightly, with a general rise in temperature from about 24°C at the

beginning of the sample period, to about 28°C at the end,

while salinity remained near 20 ppt (Figure 4d). Although cross-correlations back to 17 days prior to the commencement of sampling resulted in some significant correlations, there

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88 was no consistent pattern in correlations. Bankia correlated most strongly, because the preponderance of zeroes (no larvae in a sample) in its data set most closely matched the nearly flat temperature and salinity trends. See Appendix A for correlation coefficients between larval abundance and temperature and salinity, for 0-17 days of lag. Data from 1990, in near-bottom water, and from 1991, taken at three depths, was analyzed at each depth for correlation in abundance between pairs of species. For all three depths in both years, Cyrtopleura and Crassostrea were significantly correlated (p < 0.05). Bankia was significantly correlated with Crassostrea in 1990 midwater samples, but was not significantly correlated with either of the other two species in other depths. See also Table 1 for the results of partial correlational analysis. Abundance of three species, in near-bottom water, in 1991, are presented graphically together for each of four tidal cycles, in Figures 5a-d. For abundance at other depths, refer to Appendix B.

Table 2 summarizes the results for proportion of pediveligers collected each of four times of day (midnight, morning, noon, evening), in 1990. For all three species, the lowest mean abundances in near-bottom water were observed during evening samples, but significant differences (p < 0.05) were detected only for Cyrtopleura, for which the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89 most data was available. Table 2 presents the results of analyses of variance for each species.

Abundance over Tide and Depth In 1991, pediveliger larvae of the oyster Crassostrea virginica, the clam Cyrtopleura costata, and the shipworm Bankia gouldi, were collected in sufficient numbers for analysis of the relation of depth and tidal phase on pediveliger abundance. Variation in abundance for any one species sometimes exceeded two orders of magnitude, within one tidal cycle (12.4 hours), and tidal phase had no detectable effect on abundance, but abundance was strongly affected by depth. Table 3 presents the results of multi­ factor analyses of variance for tidal phase and depth. Values for all depths are given in Appendix B. Figure 6 graphically presents distribution for all depths in 1991, expressed as proportions, for three species. In 1992, pediveliger larvae of Crassostrea, Cyrtopleura, and Bankia, in addition to the mussel Geukensia demissa, were collected in sufficient levels for statistical analysis. Depth distributions for Cyrtopleura and Bankia were similar to those for 1991, but pediveligers of Crassostrea were most abundant in midwater samples, rather than in near-bottom samples as in 1991. Pediveligers of Geukensia were most abundant in near-surface water and least abundant in near-bottom water; the opposite of most other

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90 observed depth distributions. Figures 7 graphically represents results for depth distributions for 1992, expressed as proportions. Data from 1992 and 1993 from the three near-bottom intakes (0, 0.5, 1.0 m above the benthos) were combined, to give 10 replicate series in time for Cyrtopleura and 8 series for Crassostrea, the two species abundant enough for statistical analysis in these samples. For both species, analysis of variance showed significantly (p < 0.05) higher abundances of pediveliger larvae 0.5 m above the benthos, than on the benthos (0 m). Table 4 summarizes the results and analyses of variance.

In 1991, pediveliger larvae of Crassostrea, Cyrtopleura, and Bankia, were collected by near-bottom intakes in sufficient numbers for statistical analysis of the relationship between time of day, tidal phase, and depth distribution. Tidal phase had no detectable effect of depth distribution, nor did time of day, for any species. Table 5 summarizes the analyses of variance.

Abundance across Tidal Currents

In 1992, pediveliger larvae of the oyster Crassostrea virginica, the clam Cyrtopleura costata, and the shipworm Bankia gouldi, were collected in sufficient numbers for statistical analysis of variation in abundance between three near-bottom intakes distributed horizontally 5 m apart, on

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. an axis perpendicular to tidal current vectors. No significant differences in abundance between the three intakes, for any species, were found, and variation between the three intakes within a single sample rarely exceeded a factor of two. The most abundant species, Cyrtopleura, was analyzed for the effect of the length of any one sampling event abundance variability, as measured by the coefficient of variation. Variation in abundance (arcsine-square root transformed) decreased, but not significantly, with increasing sample length (r2 = 0.298, p = 0.082).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92 Discussion Variation in water temperature or salinity did not adequately explain pediveliger larval abundance for any species. Although cross-correlational analysis showed some significant correlations, these were either explained by a relative lack of variation (e.g. temperature and salinity versus abundance of Bankia), or were inconsistent (e.g. correlation at the 0 and 17 days of lag for Crassostrea). Diaz (1973) was unable to detect changes in the growth rate of Crassostrea larvae due to sudden temperature changes less than a2 0° C, so perhaps Chesapeake Bay species are little

affected by the temperature and salinity changes observed in this study. Zimmerman and Pechenik (1991), however, reported that growth rates and time to competency of larvae of the gastropod Crepidula plana, another estuarine species varied with temperature (20-29° C) and salinity (4-30 ppt) ,

so sudden, significant shifts in these parameters might temporarily affect the supply of competent larvae in a coastal system. The lack of variation in temperature or salinity was probably fortuitous for this research, by reducing the variables of concern. The consistent within-tidal cycle correlation between species, however, especially between Cyrtopleura and Crassostrea (Figure 5), suggested that larval abundance patchiness on a scale of hundreds of meters (the length of the body of water flowing past the sample intakes during any

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 93 one event) existed in the plankton. If present, it is probable that this patchiness is related to some, environmental variable, such as phytoplankton abundance or patches of planlton predators. (In the York River, Crassostrea and Cyrtopleura adults occupy different habitats, and it is unlikely that they would release larvae into the same localized parcels of water.) Sephton and Booth (1992) described a system in which oyster larval density cycled chaotically, with patches on the scale of hundreds to thousands of meters, around a semi-enclosed embayment, over a period of days, apparently in response both to tidal circulation and water column parameters. The implication of discrete larval patchiness is that a researcher using an Eulerian approach (fixed sampling location) must develop a sampling regime to detect patches as they move past, which depends on the size of the patches and current velocity. A researcher using a Lagrangian

approach (following water body movements), however, may be able to plot plankton patches and current vectors, and thus predict plankton abundance at a given site, for a given time. A test of the latter approach was made with some success by Shanks and Wright (1987) to predict barnacle larval abundance at shore sites in a coastal embayment, and correlate it with subsequent settlement. An alternate explanation for the apparent patchiness could be vertical larval migration with relationship to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94 tidal phase, during which in apparent period of low abundance, the larvae move to near the benthos, and thus avoid detection by the plankton samplers. Larva must approach the benthos to settle, and it is possible that they do so more often at a particular tidal phase. No clear tidal migration was detected here, however, so evidence for

tidal migration driving apparent larval abundance patterns is not strong. Apparent day-night patterns in larval distribution were discovered for Cyrtopleura in 1990, with significantly higher abundances in near-bottom waters at midnight than during the late afternoon. The most probable explanation is vertical migration, upwards during the day, because abundance over time was otherwise apparently completely random. Subsequent data, however, suggests that, at this site, vertical migration with time of day is not a strong or consistent phenomena, and is overriden by depth-specific distribution patterns. Patterns of Crassostrea abundance in near-bottom water were similar in 1990, but not significant, and patterns of Bankia abundance differed, although again, they were not significant. In 1991, data from all three depths failed to show a significant effect of time of day. It therefore appears that if diurnal migration exists for any of these species in the York River, the phenomena is relatively weak compared to that reported for scallop larvae, Placopecten magellanicus, in a weakly stratified

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95 system (Tremblay and Sinclair, 1990) . Part of the cause for the weak patterns observed, however, was the high variability observed across this factor; thus the role of time of day on larval depth distribution or abundance is a concept deserving closer examination with other sampling techniques.

Tremblay and Sinclair (1990), reported that scallop (Placopecten magellanicus) larvae in the Bay of Fundy were at shallower depths at night than during the day, and used this as clear evidence of diurnal depth migrations in this species. In that study, however, all developmental stages

of veliger larvae larger than 85 /urn shell length were considered together. In this present study, only pediveliger larvae, the final developmental stage before settlement and metamorphosis, were considered. In a separate study of Placopecten larvae, it was found that while as that many as 9.5% of the larvae were larger than 233 pm in shell length, which includes pediveligers, usually only about 0.3-1.7% of the larvae were in this largest size class (calculated from Tremblay and Sinclair, 1992). It is possible, therefore, that any depth or time related patterns (or absence of patterns) of abundance of Placopecten pediveliger larvae, reported in Tremblay and Sinclair (1990), were lost in overall larval distribution patterns. For four species in this study, including the oyster Crassostrea virginica, abundance within the water column was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96 related primarily to water depth, with peak abundance either near-bottom (Crassostrea, Cyrtopleura, and Bankia) or near­ surface (Geukensia). Nelson (1927, 1928) and Perkins (1931, 1932) also recorded the abundance of larvae of Crassostrea over depth, in a 3 m water column in Barnegat Bay, New Jersey, over a period of years. Their results varied, but generally showed a peak in abundance near midwater, associated with a halocline. No halocline was present at the site for the present study. Nelson (1927, 1928) and Perkins (1931, 1932) did not distinguish between various developmental stages of the larvae, which may also explain in part the differences between their findings, and the findings from the present study. Carriker (1951), who distinguished between phases of Crassostrea larvae, reported that older larvae tended to congregate near the benthos. Banse (1986) reported that all stages of several species of polychaete larvae and larvae of three echinoderm species were concentrated in the lower part of the water column. Yoo et al. (1977) also found that all stages of the ark clam Anadara broughtonii tended to be concentrated near the

benthos, but that this pattern strengthened in shallow regions (« 8.0 m) over regions with deeper water (« 16.0 m). By the definition of Yoo et al. (1977), the present study took place in shallow water. The mixing of the water column that inhibited both haloclines and thermoclines at the site of this study did

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97 not prevent vertical stratification of pediveliger larvae, since all species analyzed favored either near-surface water (Geukensia) or near-bottom water (Crassostrea, Cyrtopleura, and Bankia). Higher abundance of the latter three species in near-bottom water could be explained by passive sinking, and Banse (1986) hypothesized that late-stage larvae of some species are near the bottom (in high-density water) because they gradually lose swimming ability, but the higher near­ surface abundance of Geukensia reported in the present study can only be behavioral in origin. It is thus possible that the observed depth distributions in all four species were primarily due to swimming behavior. Adult Geukensia are primarily intertidal (Kraus and Crow, 1985), which may be in part a reflection of settlement choice. Maximum abundance of Geukensia pediveligers in this study was in near-surface waters. Cyrtopleura as adults are exclusively infaunal in benthic sediments (Turgeon, 1968), which could in turn correlate with the high near-bottom abundance of pediveliger larvae of this species. Although Crassostrea adults are found throughout the water column in Chesapeake Bay (Turgeon, 1968), Roegner and Mann (1990) found that settlement throughout the intertidal zone in the York River, Virginia, increased with depth. This trend is compatible with the abundance distribution for Crassostrea pediveligers found in this study, at least for 1991. Bankia settles throughout the water column in Chesapeake Bay

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98 (Scheltema and Truitt, 1954), but it is probable that the majority of available substrate (water-logged wood) is found on the benthos, near where the majority of pediveliger larvae were found in this study. Pediveliger depth distributions in the York River for these species are therefore consistent with settlement or adult habitat depths. Within the lowest 1 m of the water column, however, pediveliger larval abundance did not vary significantly with depth, at current velocities of 5-10 cm-s'1. This field

observation differs remarkably from the behavior of Cerastoderma edule pediveligers observed by Jonsson et al. (1991). In that study, at current velocities of 2 cm-s'1, pediveligers distributed themselves throughout the water column, but at velocities of 5 cm-s'1, they confined

themselves to the viscous sublayer, within 1 mm of the bottom. Pediveligers must approach the substrate to settle, which at vertical sinking or swimming velocities of up to 5 mm-s'1 (Hidu and Haskin, 1978; Mann and Rainer, 1990), could

be done from a meter away in minutes. It is not known how close pediveliger larvae must be to the substrate to detect chemical cues (e.g. Vietch and Hidu, 1971; Fitt et al., 1990; Chevolot et al., 1991), but staying very close to the substrate increases the encounter rate with benthic planktivores (e.g. Breese and Phibbs, 1972; Steinberg and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99 Kennedy, 1979; Cowden and Chia, 1984). For this reason, it may be to the advantage of pediveliger larvae to stay near but not necessarily at the bottom of the water column, until immediately prior to settlement. Jonsson et al. (1991) reported that Cerastoderma pediveligers in a flume at low current velocities (2 mm-s'1) spent short periods in the viscous sublayer, before returning to the water column. This suggests substrate "sampling" by the pediveligers, and could in part explain the near-bottom distribution of pediveligers observed in this study. Carriker (1951) and Kunkle (1958) both suggested that late-stage Crassostrea larvae remained near the bottom to take advantage of net upstream tidal transport. Although this may be true in some systems, the lack of water column stratification at the sampling location in this study suggests that the pediveligers have other reasons for being near the benthos. The role of water currents in larval settlement processes is poorly understood, however. There also remains uncertainty regarding the extent to which settling invertebrate larvae resemble inert, slightly negatively buoyant particles, as opposed to active swimmers. Analysis of the depth distribution patterns of abundance observed in this study indicate that such patterns are fairly strong and consistent. The Geukensia example demonstrates that not all species have similar depth distributions, so depth distribution patterns must be

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. determined separately for each species, and possibly for each sampling site. Once determined, however, depth distribution is not a large source of variability in abundance estimates, and plankton sampling at a single depth is probably sufficient for estimates of instantaneous larval abundance.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101 References Banse, K. 1986. Vertical distribution and horizontal transport of planktonic larvae of echinoderms and benthic polychaetes in an open coastal sea. Bull. Mar. Sci. 39:162-175. Beukema, J.J. and J. de Vlas. 1989. Tidal-current transport of thread-drifting postlarval juveniles of the bivalve Macoma balthica from the Wadden Sea to the North Sea. Marine Ecology Progress Series 52:193-200. Breese, W.P., & F.D. Phibbs. 1972. Ingestion of bivalve molluscan larvae by the polychaete annelid Polydora ligni. Veliger 14 (3): 274-275. Carriker, M.R. 1951. Ecological observations on the distribution of oyster larvae in New Jersey estuaries. Ecol. Monogr. 21:19-3 8. Carriker, M.R. 1961. Interrelation of functional morphology, behavior and autecology in early stages of the bivalve Mercenaria mercenaria. J. Elisha Mitchell Scientific Soc. 77:168-241. Chanley, P. and J.D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia. Malacologia 11:45-119. Chevolot, L., J.-C. Cochard, & J.-C. Yvin. 1991. Chemical induction of larval metamorphosis of Pecten maximus with a note on the nature of naturally occurring triggering substances. Mar. Ecol. Prog. Ser. 74:83-89. Cowden, C., C.M. Young, & F.-S. Chia. 1984. Differential predation on marine invertebrate larvae by two benthic predators. Mar. Ecol. Prog. Ser. 14 (2/3): 145-149. Culliney, J.L. 1975. Comparative larval development of the shipworms Bankia gouldi and Teredo navalis. Marine Biology 29:245-251. Diaz, R.J. 1973. Effects of brief temperature increases on larvae of the American oyster (Crassostrea virginica). J. Fish. Res. Board Can. 30:991-993. Fitt, W.K., S.L. Coon, M. Walch, R.M. Weiner, R.R. Colwell, & D.B. Bonar. 1990. Settlement behavior and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supernatants. Mar. Biol. 106:389- 394.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102 Fuller, S.C. and R.A. Lutz. 1989. Shell morphology of larval and post-larval mytilids from the north-western Atlantic. J. Marine Biological Assoc. U.K. 69:181-218. Grigorieva, N.I. and V.N. Regulev. 1992. Vertical distribution of larvae of the scallop Mitzuhopecten yessoensis and mussel Mytilus trossulus in Posyet Bay, Sea of Japan. Soviet J. Marine Biology 17:105-109. Gustafson, R.G., R.L. Creswell, T.R. Jacobsen, & D.E. Vaughan. 1991. Larval biology and mariculture of the angelwing clam, Cyrtopleura costata. Aquaculture 95:257-279. Hidu, H. and H.H. Haskin. 1978. Swimming speeds of oyster larvae Crassostrea virginica in different salinities and temperatures. Estuaries 1:252-255.

Kunkle, D.E. 1958. The vertical distribution of oyster larvae in Delaware Bay. Proc. Natl. Shellf. Assoc. 48:90-91. Jonsson, P.R., C. Andre, and M. Lindegarth. 1991. Swimming behavior of marine bivalve larvae in a flume boundary- layer flow: evidence for near-bottom confinement. Mar. Ecol. Prog. Ser. 79:67-7 6

J0rgensen, C.B. 1946. Lamellibranchia. 277-311. In Thorson, G. & C.B. Jorgensen. Reproduction and Larval Development of Danish Marine Bottom Invertebrates. Meddelser Fra Kommissionen for Danmarks Fiskeri-og Havundersogelser: Plankton Ser. 4: 523 pp. Jorgensen, C.B. 1981. Mortality, growth, and grazing impact of a cohort of bivalve larvae, Mytilus edulis L. Ophelia 20:185-192. Kennedy, V.S., R.A. Lutz, and S.C. Fuller. 1989. Larval and early postlarval development of Macoma mitchelli Dali (Bivalvia:Tellinidae). Veliger 32:29-38. Kraus, M.L. and J.H. Crow. 1985. Substrate characteristics associated with the distribution of the ribbed mussel, Geukensia demissa (Modiolus demissus), on a tidal creek bank in southern New Jersey. Estuaries 8:237-243. Loosanoff, V.L. and H.C. Davis. 1963. Rearing of bivalve mollusks. Advances in Marine Biology 1:1-13 6. Loosanoff, V.L., H.C. Davis, and P.E. Chanley. 1966. Dimensions and shapes of larvae of some marine bivalve mollusks. Malacologia 4:351-435.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103 Mann, R. 1988. Distribution of bivalve larvae at a frontal system in the James River, Virginia. Marine Ecology Progress Series 50:29-44. Mann, R. and S.M. Gallager. 1985. Physiological and biochemical energetics of larvae of Teredo navalis L. and Bankia gouldi (Bartsch) (Bivalvia: Teredinidae). J. Exp. Mar. Biol. Ecol. 85:211-228. Mann, R. and J.S. Rainer. 1990. Effect of decreasing oxygen tension on swimming rate of Crassostrea virginica (Gmelin, 1791) larvae. J . Shellfish Res. 9:323-327. Nelson, T.C. 1927. Report of the Department of Biology of the New Jersey State Agricultural Experiment Station, for the year ending June 30, 1926: 103-113. Nelson, T.C. 1928. Report of the Department of Biology of the New Jersey State Agricultural Experiment Station, for the year ending June 30, 1927: 77-83. Newell, C.R., H. Hidu, B.J. McAlice, G. Podniesinski, F. Short, and L. Kindblom. 1991. Recruitment and commercial seed procurement of the blue mussel Mytilus edulis in Maine. J. World Aquaculture Soc. 22:134-152. Perkins, E.B. 1931. A study of oyster problems in Barnegat Bay. Report of the Department of Biology of the New Jersey State Agricultural Experiment Station, for the year ending June 30, 1930: 25-47. Perkins, E.B. 1932. Progress of oyster investigation in Barnegat Bay during 193 0. Report of the Department of Biology of the New Jersey State Agricultural Experiment Station, for the year ending June 30, 1931: 116-122. Rees, C.B. 1950. The identification and classification of lamellibranch larvae. Hull Bulletins of Marine Ecology 3 (19):7 3-104. Roegner, G.C. and R. Mann. 1990. Settlement patterns of Crassostrea virginica (Gmelin, 1791) larvae in relation to tidal zonation. J. Shellfish Res. 9:341-346. Scheltema, R.S. and R.V Truitt. 1954. Ecological factors related to the distributions of Bankia gouldi Bartsch in Chesapeake Bay. Maryland Board of Natural Resources, Department of Research and Education. Publ. #100. 31 pp.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 104 Sekiguchi, H., H. Saito, and H. Nakao. 1991. Spatial and temporal distributions of planktonic and benthic phases of bivalves in a tidal estuary. Nippon Bentosu Gakkaishi (Bull. Japan Assoc. Benthology) 40:11-21. Seliger, H.H., J.A. Boggs, R.B. Rivkin, W.H. Biggley, and K.R.H. Aspden. 1982. The transport of oyster larvae in an estuary. Marine Biology 71:57-72. Sephton, T.W. and D.A. Booth. 1992. Physical oceanographic and biological data from the study of the flushing of oyster (Crassostrea virginica) larvae from Caraquet Bay, New Brunswick. Can. Manu. Rep. Fish. Aquatic Sci. 2162. 61 pp. Shanks, A.L. and W.G. Wright. 1987. Internal wave-mediated shoreward transport of cyprids, megalopae, and gammarids and correlated longshore differences in the settling rate of intertidal barnacles. J. Exp. Mar. Biol. Ecol. 114:1-13.

Steel, R.G.D. and J.H. Torrie. 1980. Principles and Procedures of Statistics: A Biometrical Approach. 2nd. Ed. McGraw-Hill Book Co., New York, NY. 633 pp. Steinberg, P.D. & V.S. Kennedy. 1979. Predation upon Crassostrea virginica (Gmelin) larvae by two invertebrate species common to Chesapeake Bay oyster bars. Veliger 22 (1): 78-84. Sullivan, C.M. 1948. Bivalve larvae of Malpeque Bay, P.E.I. Fisheries Research Board of Canada Bulletin 77. 59 pp. Tremblay, M.J. and M. Sinclair. 1990. Diel vertical migration of sea scallop larvae Placopecten magellanicus in a shallow embayment. Marine Ecology Progress Series 67:19-25. Tremblay, M.J. and M. Sinclair. 1992. Planktonic sea scallop larvae (Placopecten magellanicus) in the Georges Bank region: broadscale distribution in relation to physical oceanography. Can. J. Fish. Aquat. Sci. 49:1597-1615. Turgeon, D.D. 1968. Guide to estuarine and inshore bivalves of Virginia. M.A. Thesis, College of William and Mary, Williamsburg, VA. 126 pp. Vietch, F.P. & H. Hidu. 1971. Gregarious setting in the American oyster Crassostrea virginica Gmelin: I Properties of a partially purified "setting factor". Chesapeake Sci. 12:173-178.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 105 Yoo, S.K., Y.U. Kim, and K.Y. Park. 1979. Improvement of the method of seed scallop production. (In Korean with English summary). Bull. Natl. Univ. Busan 19:55-62. Yoo, S.K., K.Y. Park, and M.S. Yoo. 1977. Biological studies on arkshell culture 1. Distribution of drifting larvae of the arkshell, Anadara broughtonii Schrenk. (In Korean with English summary). J . Oceanological Soc. Korea 12:75-81. Yoo, S.K. and H.Y. Ryu. 1985. Occurrence and survival rate of the larvae of Pacific oyster Crassostrea gigas in Hansan Bay. Bull. Korean Fisheries Soc. 18:47-476. Zar, J.H. 1984. Biostatistical Analysis. 2nd Ed. Prentice- Hall, Englewood Cliffs, NJ. 718 pp. Zimmerman, K.M. and J.A. Pechenik. 1991. How to temperature and salinity affect relative rates of growth, morphological differentiation, and time to metamorphic competence in larvae of the marine gastropod Crepidula plana? Biol. Bull. 180:372-386.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106 Table 1. Correlation coefficients r for planktonic larval abundance data, between pairs of species. Values derived from partial correlational analysis of data from 1990 and 1991. Cyrtopleura Cyrtopleura Crassostrea & Crassostrea & Bankia & Bankia

1990 Near-Bottom 0.582* 0.306* 0.013 N = 100 1991 Near-Surface -0.512* 0.345 0.129 N = 32 1991 Midwater 0.690* 0.110 0.483* N = 32 1991 Near-Bottom 0.824* 0.074 0.155 N = 32 * Significantly non-independent at a = 0.05

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107

Table 2. Mean abundance of larvae for four sample times daily, 1990, expressed as proportions (percentages) of all larvae collected that day. Cyrtopleura Crassostrea Bankia

# Days 14 11 3

Midnight 36.6 31.9 13.3 Morning 29.5 30.7 45.4 Noon 26.8 26.2 41.3 Evening 12 .7 11. 2 <1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108

Table 3. Summary of analysis of variance for the effects of tidal phase (Tide) and depth (Depth) on larval abundance, with day of sample (Day) as a non-interacting factor, for three species. DF = degrees of freedom; Seq. SS = sums of squares, Adj. SS = adjusted sums of squares; Adj. MS = adjusted mean squares.

Table 3a. Analysis of Variance for Cyrtopleura. Source______DF______SS______MS______F_____P Tide 7 9.575 1.368 0.91 0.507 Depth 2 101.527 50.763 33.61 0.000 Tide*Depth 14 31.472 2.248 1.49 0.139 Day 3 55.259 18.420 12 .19 0. 000 Error 69 104.226 1.511 Total 95 302.059

Table 3b. Analysis of Variance for Crassostrea. Source DF SS MS F P Tide 7 6.244 0.892 0.78 0.605 Depth 2 97.867 48.933 42.88 0. 000 Tide*Depth 14 17 .951 1.282 1.12 0.354 Day 3 7.729 2.576 2.26 0. 089 Error 69 78.741 1.141 Total 95 208.532

Table 3c. Analysis of Variance for Bankia. Source DF SS MS F P Tide 7 13.261 1.894 1.36 0.238 Depth 2 120.363 60.182 43 .10 0.000 Tide*Depth 14 23.914 1.708 1.22 0.279 Day 3 3 . 657 1. 219 0.87 0.459 Error 69 96.341 1.396 Total 95 257.537

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109 Table 4. Summary of analysis of variance for the effects of depth near the benthos (0.0 m, 0.5 m, 1.0 m above the benthos) on the proportional abundance (Prop. Abundance) of larvae. N = sample size, s.d. = standard deviation, DF = degrees of freedom, SS = sums of squares, MS = means squares.

Table 4a. Results and analysis of variance for Cyrtopleura Prop. Abundance Depth N Mean s.d. 0.0 m 10 0.4339 0.2176 0.5 m 10 0.7687 0.2089 1.0 m 10 0.5698 0.2493 Analysis of Variance SourceDF SS MS Depth 2 0. 5673 0.2836 5. 56 0 . 010 Error 27 1.3785 0.0511 Total 29 1.9458

Table 4b. Results and analysis of variance for Crassostrea Prop. Abundance Depth_____ N______Mean_____ s.d. 0.0 m 8 0.3344 0.3542 0.5m 8 0.7539 0.1162 1.0 m 8 0.5965 0.3106 Analysis of Variance Source____ DF______SS______MS______F______p Depth 2 0.7187 0.3594 4.58 0.022 Error 21 1.6481 0.0785 Total 23 2.3668

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110 Table 5. Summary of analysis of variance for the effects of time of day (Time), and tidal phase (Tide), on proportion of larvae in near-bottom water, for three species. DF = degrees of freedom; Seq. SS = sums of squares, Adj. SS = adjusted sums of squares; Adj. MS = adjusted mean squares.

Table 5a. Analysis of Variance for Cyrtopleura Source DF Sea. SS Adi. SS Adi. MS F P Tide 3 0.4630 0.3830 0.1277 1.19 0.333 Time 1 0.0034 0.0022 0.0022 0.02 0.887 Tide*Time 3 0.3180 0.3180 0.1060 0. 99 0.413 Error 24 2.5651 2.5651 0.1069 Total 31 3.3495

Table 5b. Analysis of Variance for Crassostrea. Source DF Sea. SS Adi. SS Adi. MS FP Tide 3 0.40695 0.33057 0.11019 1.26 0.309 Time 1 0.00529 0.00625 0. 00625 0. 07 0.791 Tide*Time 3 0.12442 0.12442 0.04147 0.48 0.702 Error 24 2.09208 2.09208 0.08717 Total 31 2.62874

Table 5c. Analysis of Variance for Bankia. Source DF Sea. SS Adi. SS Adi . MS FP Tide 3 0.0979 0.0967 0.0322 0.20 0.895 Time 1 0.0034 0.0032 0.0032 0. 02 0.889 Tide*Time 3 0.5617 0.5617 0.1872 1.17 0.343 Error 24 3.8522 3.8522 0.1605 Total 31 4.5152

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ill

Figure 1. Map of study area, lower York River, Chesapeake Bay, Virginia.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Q.

Research » Site V

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 112

Figure 2. Diagram of vertical horizontal sampling array.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cable Pulley

Piling

To Pumps

Float

20 cm Upper (floating) Hoses Intake

Middle (adjustable) Intake

Bottom (fixed) Intake

25 cm 'i'

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3. Diagram of horizontal sampling array intake.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Float

r \ o0o0 o ° § o ? ) ° 0 ° n ° n O S Intake Area OO q O OOnn ° f f o ° ( 25 cm

To Pump

Weight

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 114

Figure 4. Abundance of three species of pediveliger larvae in near-bottom water, 1990, and simultaneous temperature and salinity. Larval abundance values are individuals • m'3, taken as a mean for each date; temperature is in degrees Celsius; salinity are in parts per thousand.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cyrtopleura Bankia Crassostrea Salinity Temperature — ------1 # A X @— ------F-+-l-H SAMPLE SAMPLE DATE (July 4-August 7, 1990) H-HH X'X'X-x-X -^''^'XX-X-X-X'^'X'X'X-X-Xi^-X-X-X-XkX-X->jX -x|X-X'X-X-X-X 70 60 50 40 E co Hi o z < Q z cc — m < UJ o _l UJ > Q UJ Q.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115

Figure 5. Abundance of three species of pediveliger larvae in near-bottom water, 1991, for four tidal cycles.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcd ih emsin fte oyih onr Frhrrpouto poiie ihu permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

1000 o O o U • 0VNe 1VAUV1 30NVQNneV gUJ • eBS 6oj) (e|BDS o S •S 1: a 8 o 1 O o o o o o 0 0 CM 00 in

TIDAL SAMPLING PERIOD TIDAL SAMPLING PERIOD Figure 6. Proportional abundance of three species of pediveliger larvae at three depths, 1991.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcd ih emsin fte oyih onr Frhrrpouto poiie ihu permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

1991 Nwmoo Nwmoo H31VM NWniOO NWniOO U3J.VM

% OF ALL LARVAE % OF ALL LARVAE Figure 7. Proportional abundance of four species of pediveliger larvae at three depths, 1992.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcd ih emsin fte oyih onr Frhrrpouto poiie ihu permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

1992 •32 I 5 Nwmoo U31VM o U. UJ < 0C § a g Co CO Co CO o wo uaxvM wwmoo

% OF ALL LARVAE % OF ALL LARVAE 118 xs G (0 ********* (0 hnHininn(NH(ONuiNHNHi"joj a) h monoincomw^NMnnmainoMH 3-4 onM^j,'^,^t,in'>^ri,in'5i,ntMf'jcNf')CNog JI3 CJ •*tii«it*i(i«iiiii P 3 oooooooooooooooooo 03 IX) I I I I I I I I I I I I I I I I I I QJ a g 03 a) (u P Sh P <#l'(JlinHNrHO(Mhy)n«)l^niDH03 3-4 to Hnc'JCNr-vO'to^cnvoconcNvot-Hir) Q) >iO OOOHOOOHOOHHHHHHHO 4J (/j •••••••••••••••••• (d -HI/) oooooooooooooooooo S C 03 I H 3-1 xi ho G • 03 03 &> W G 0) O0) H 3>4 * * * * G a 0) r''inoinr-'t''Covocou3 CTiincor'VDHH'3< 3 g M HVOn^HWincOrllMHtfOCOinHVDO 0303 CL, n'3,^('INHOOr)H(\l('l(NrlNt001>J H3 (/] Q •••••••••■•«•••••• 3 P OOOOOOOOOOOOOOOOOO P H S4 I I I I I I I 03 O o 3n H p 03 G > a) 3-4 g fd d) *K *K *K *K *K ho 03 ■:fUincocooiflr"d'H<*^uinMnr)('i G h r^^HnHoininoinirifocoLnnr^no GO Ai ^nn^'fnoinNHHHHNNNnoi Q) ^ c] •••••••••••••••••• 03 g 03 OOOOOOOOOOOOOOOOOO Is O PC) I t I I I I I I I 1 I I ! I I I I I Pa) o P o P 03 R 0) U Q) Jh 030 M P M'jinHcoHoino^NUimncoi'Hin PH 303 HHH^,nC^C^3nc^^^'lnC3^OC^^C0•1 a 3-4 H 03 6 0 H T3 0) 0) EH O C" 03 lO U H in o I 3 G O Q) O O h 'j,r'nv£>'i,cMt''COt-'Hcoincovoincoaico h >-1 ci, (N'OOMONOHoiifinin'oiDiniom'oco || PO O HOINHHNNHriNMNNrlOOHO fd *u • ••••••••••••••••• y H >H oooooooooooooooooo o) n p U 0) O 3 3-1 &> 0 cr> p u 3 G • rH 3 < ,— 0 • k» m H H > 1 >> 4H •rl rtj P 3 OHCNjrO^riDVDt'' H •H Q OHNn^flOIOf'COOIrlHHHrlHHrl C G Q) G -—■ &> a> H H •H a P H &> 03 a 3 3 3 < Eh U3 P

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 119

Appendix B.

Table Bl. Abundance of three species of pediveliger larvae at three depths (near-surface, midwater, near-bottom) in 1991, for four tidal cycles, sampling eight times per tidal cycle. Sample #1 included high tide, sample #5 included low tide.

Table Bla. Start July 18, 14:32; End July 19, 00:43 Cyrtopleura costata Abundance (larvae •m-3> Sample # 1 2 3 4 5 6 7 8 Near-Surface 0.3 1.2 1.4 1.8 2 . 3 1.2 2.4 3 . 3 Midwater 8.7 1.7 1.9 21 21 4.6 8 . 5 5.2 Near-Bottom 12 32 3 . 2 51 190 17 10 10

Crassostrea virginica ______Abundance (larvae-m'3) Sample # 1 2 3 4 5 6 7 8 Near-Surface 0 1.8 0 0.6 1.4 0.6 0.6 2.4 Midwater 15 2.2 1.9 6.5 4.7 2.3 3.7 1.4 Near Bottom 19 14 1.9 45 144 9.8 4 . 3 5.2

Bankia gouldi Abundance (larvae- m'3t Sample # 1 2 3 4 5 6 7 8 Near-Surface 0 0 0 0.6 0.5 0 0 1.9 Midwater 1.2 0.4 0 6.5 2 . 3 1.2 0 0 Near-Bottom 9.3 7.4 0.6 17 92 6.4 0.6 1.4

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120 Table Bib. Start July 21, 05:11; End July 21, 16:48

Cyrtopleura costata ______Abundance (larvae-m'3)______Samole # 1 2 3 4 5 6 7 8 Near-Surface 1.2 0 0.5 0.9 0 0.9 0.5 0 Midwater 4.5 2.4 1.5 1.4 0 0 4.1 2.0 Near-Bottom 8.9 6.9 2.9 186 121 43 37 4.5

Crassostrea virginica Abundance (larvae-m'3t Samole # 1 2 3 4 5 6 7 8 Near-Surface 0 0 0.5 0.9 0 1.8 0 1.0 Midwater 3.3 2.4 0.5 0.9 1.0 0.5 6.4 0.5 Near Bottom 11 6.9 1.5 40 23 27 12 4.0

Bankia gouldi Abundance (larvae- m'3} Samole # 1 2 3 4 5 6 7 8 Near-Surface 0 0.8 0 0 0 1.8 0.5 Midwater 1.2 2.4 0.5 0 1.0 0 3.7 Near-Bottom 4.9 1.6 0.5 41 17 12 23

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table Blc. Start July 29, 11:22; End July 29, 23:16

Cyrtopleura cos tata ______Abundance (larvae - m~3)______Samole # 1 2 3 4 5 6 7 8 Near-Surface 1.5 0 3.0 3.2 1.1 2 . 6 6.4 12 Midwater 2.6 9.6 2.2 5.1 5.6 6.3 9.4 20 Near-Bottom 15 16 17 0.9 5.3 9 . 0 11 19

Crassostrea virginica ______Abundance (larvae-m'3) Samole # 1 2 3 4 5 6 7 8 Near-Surface 1.1 0 1.7 0.5 0.7 3 . 2 3 . 0 2 . 6 Midwater 6.6 7 . 1 3.5 2.3 3.9 3 .2 3.0 11

Near Bottom 25 16 12 1.4 4.2 4.2 4.1 6.4

Bankia gouldi Abundance (larvae- m ’3t Samole # 1 2 3 4 5 6 7 8 Near-Surface 0 0 0 0 0.7 1.6 0.7 1.1 Midwater 0.4 3 . 0 0.9 2 . 3 1.4 1.6 2 . 5 3.4 Near-Bottom 4.4 7.1 5.2 0.9 2.5 2 .1 5.2 3.4

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 122 Table Bid. Start August 1, 00:29; End August 1, 12:21

Cyrtopleura costata ______Abundance (larvae-m'3)______Sample # 1 2 3 4 5 6 7 8 Near-Surface 8.9 3.7 4 . 3 0.9 6.7 39 0.6 2 .1 Midwater 10 29 41 34 48 0 127 13 Near-Bottom 47 21 48 155 85 150 179 15

Crassostrea virginica Abundance C larvae-m'3) Samole # 1 2 3 4 5 6 7 8

Near-Surface 2 . 5 1.2 0.5 0.5 3.7 4.2 0 0

Midwater 3 . 0 4 . 4 3 . 3 1.4 1.3 1.4 2 . 6 2.5 Near Bottom 5.0 11 4.8 5.6 3.4 6.0 1.9 1.2

Bankia gouldi Abundance (larvae- m'3t Samole # 1 2 3 4 5 6 7 8 Near-Surface 0.5 0.4 0 0.5 1.0 1.9 0 0.4 Midwater 1.0 6.2 7.6 1.9 1.3 0.5 0.6 0.8 CO o

Near-Bottom 6.4 12 21 7.9 4.4 2 . 3 5.2 •

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4

A FIELD COMPARISON OF PLANKTONIC ABUNDANCE OF OYSTER LARVAE, Crassostrea virginica AND Ostrea equestris, WITH SUBSEQUENT SETTLEMENT

123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 124

Table of Contents

Page Abstract...... 125 Introduction...... 126 Materials and Methods...... 132 Research Sites...... 132 Plankton Sampling...... 133 Estimation of Juvenile Mortality...... 134 Estimation of Settlement...... 138 Analysis of Settlement as a Function of Larval Abundance...... 141 Results...... 143 Planktonic Abundance...... 143 Juvenile Mortality...... 144 Settlement...... 144 Settlement as a Function of LarvalAbundance ...... 145 Discussion...... 147 Depth Distribution...... 147 Juvenile Mortality...... 150 Settlement as a Function of Larval Abundance...... 151 Implications for Shell String Survey...... 153 References...... 155 Tables...... 162 Figures...... 168

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125

Abstract Field studies were effected to compare planktonic abundance estimates of pediveliger larvae of oysters, Crassostrea virginica and Ostrea equestris, with subsequent settlement, using shell strings to monitor settlement. The shell string has been used for many years by the Virginia Institute of Marine Science to monitor Crassostrea settlement in Chesapeake Bay. At two locations, the lower York River in Chesapeake Bay, Virginia (Crassostrea only), and the Indian River lagoon near Fort Pierce, Florida (Crassostrea and Ostrea), small scale variations in larval abundance over time and with depth were clearly observed. Larvae tended to remain near the bottom of the water column, except for Ostrea during the night. Settlement onto shell strings was not significantly related to larval abundance for Crassostrea at either site, at planktonic variations of

about one order of magnitude. Settlement of Ostrea was significantly related to planktonic abundance in Florida.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 126 Introduction Most benthic marine invertebrates have, as part of their life cycle, a planktonic larval phase, which is followed by settlement to the benthos and metamorphosis to a juvenile phase. (See Mileikovsky, 1971, for a discussion of types of marine invertebrate larval life histories.) Research on the relationship between larval abundance and subsequent requires quantification of larvae just prior to settlement, followed by quantification of the settled postlarvae or juveniles. Quantitative research on the planktonic larval phase is complicated by the small size, wide dispersal, and non- homogeneous distribution of larvae (Perkins, 1931; Mann, 1988; Sekiguchi et a l ., 1991; Tremblay and Sinclair, 1992; see also Introduction of Chapter 3). Rumrill (1990) and Yoo and Ryu (1985) have discussed prior attempts to quantify larval processes in the plankton. Quantitative research on the early juveniles, or "settlers" is complicated by high mortality (Luckenbach, 1984; Powell et a l ., 1986; Roegner, 1991) and even in the case of hard-shells mollusks, rapid dissolution of the remains (Powell et al., 1984, 1986). Life history studies of bivalve mollusks often commence with the smallest stage that is detected in conventional sampling techniques (Brousseau, 1978, Peterson, 1986), or sample at intervals of many days (Shaw, 1965; Nosho and Chew, 1972; Dean and Hurd, 1980; Gaines et al., 1985; Shanks and Wright,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 127 1986; Thresher et a l ., 1989; Buyanovskii, 1991), during which time individuals may die and their shells dissolve (Powell et a l ., 1984, 1986). The annual oyster (Crassostrea virginica) population survey conducted by the Virginia Institute of Marine Science quantifies individuals detectable by the naked eye, or greater than about 2 mm in

diameter (B. Barber, pers. comm.). Some relatively recent studies have attempted to tightly relate larval abundance to subsequent settlement. This relationship has been examined for bivalve mollusks (Feller et a l ., 1992; Martel et al., 1993), barnacles (Shanks and Wright, 1987; Minchinton and Scheibling, 1991) and decapod crustaceans (Lipcius et al., 1990; Olmi et a l .,

1990). On one level of scale, there must be a relationship between larval abundance and subsequent settlement; no settlement can occur without larvae, and heavy settlement must be preceded by high larval density. Martel et al. (1993) found a strong relationship between late-stage zebra mussel (Dreissena spp.) larval abundance, and subsequent settlement, over a larval abundance (density) range of three orders of magnitude. Shanks and Wright (1987) describe a system in which larvae arrive in patches; there was approximately one order of magnitude of difference between within-patch and outside-patch larval abundance, and recruitment was significantly higher on substrates within a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 128 planktonic larval patch, for two barnacle species (Balanus glandula and Semibalanus cariosus), despite a recruitment sampling interval of days. With a similar difference in crab Callinectes sapidus megalopa larval abundance, Lipcius et al. (1990) found a significant relationship between larval abundance and subsequent settlement to artificial collectors. Not all studies have shown a relationship between larval abundance and settlement. Feller et a l . (1992) was unable to establish a relationship between larval abundance and recruitment of several juvenile invertebrate taxa. Olmi et a l . (1990) found a significant relationship between crab, Callinectes sapidus, larval abundance and settlement to a natural habitat, but not to adjacent artificial settlement substrates. Minchinton and Scheibling (1991) reported a significant relationship between barnacle (Semibalanus balanoides) larval abundance and settlement density, but in that study, larval abundance was corrected for tidal immersion time, and was not found to vary within the water column. There are several possible reasons why settlement might not be expected to correlate with larval abundance, one of which is interference of settlement by prior organisms (Scheltema, 1974; Woodin, 1976; Dean and Hurd, 1980; Watzin, 1983; Osman et al., 1989) or passive exclusion, such as by drifting macrophytes (Olaffson, 1988). To eliminate

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 129 interference by pre-existing organisms, researchers usually provide settling larvae with clean (defaunated) settlement substrates (Dean and Hurd, 1980; Butman, 1989; Duggins et al., 1990; Hurlbut, 1991; Menge, 1991). The Virginia Institute of Marine Science has for many years monitored Crassostrea virginica recruitment using a defaunated substrate known as the "shell string", described below in the Materials and Methods (Haven and Fritz; 1985, and Barber, 1992). If the above interference factors are eliminated by providing larvae with clean, defaunated settlement substrate, and temperature and salinity remain fairly constant (Zimmerman and Pechenik, 1991), water column processes may still make analysis of a larva-settler relationship difficult by providing high larval abundance variability. Larvae are usually not distributed evenly throughout a given body of water, and abundance may vary with depth (Seliger et al., 1982; Tremblay and Sinclair; 1990), hydrographic conditions (Mann, 1988; Grigor'eva and Regulev, 1992), time or tide (Tremblay and Sinclair, 1990; Sekiguchi et al., 1991), or apparently randomly (Newell et al., 1991; Sephton and Booth, 1992). Reported patterns differ for species and the body of water involved (see also Chapter 3). To get around this difficulty, a researcher must account, as much as is possible, for variation in larval abundance within the larval sampling design. Factors

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 130 of importance in a shallow coastal system (Chesapeake Bay) and sampling methods have been discussed in Chapter 3 of this dissertation. A final source of settlement variability is larval behavior itself, and its interaction with the microscale aquatic environment. Qualitative observations of variable behavior of settling bivalve larvae have been given by Prytherch (1934), Cole and Knight-Jones (1949), Carriker (1961), and Cranfield (1973). Apparently competent-to- settle larvae may delay metamorphosis (Coon et al., 1990; Pechenik, 1990). The hydrodynamics of chemical settlement cues are poorly understood; without a chemical cue, a larva may fail to detect suitable settlement substrate only centimeters away. Chemical cues for settlement are reviewed in Appendix A of this dissertation, and have also been reviewed by Rodriguez et al. (1993). Currents may also interact positively with settlement behavior, up to a point. Examples of interactions of larvae and water currents have been given by Eyster and Pechenik (1987), Pawlik et al. (1991), and Snelgrove et al. (1993). At this point, research on the quantitative roles of larval behavior, small scale hydrodynamics, and comparable agents of variability, on the relationship between larval supply and settlement remain laboratory exercises, and acceptable methods for studying them in the field have not been developed. If, however, large scale larval abundance

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 131 variation, and interference factors are accounted for or eliminated, as discussed above, and if one assumes that under hypothetical ideal conditions, there will be a direct relationship between larval abundance and subsequent settlement, then the observed variability that remains in a field study is a measure of the importance of larval behavior and small scale interactions with the aquatic environment. The research described here makes use of the shell string, for many years a part of the Virginia Institute of Marine Science's monitoring of local settlement of oyster, Crassostrea virginica, in many parts of Chesapeake Bay, in Virginia (Haven and Fritz; 1985, and Barber, 1992). Planktonic larval abundance estimates (using techniques developed previously) are compared to subsequent settlement of oyster larvae onto shell strings. From this, the strength of the relationship between larval abundance and settlement is estimated. Inferences will then be drawn, regarding the precision of the shell string method in predicting Chesapeake Bay oyster settlement.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 132 Materials and Methods Research Sites An initial study was carried out July 5 to August 8, 1990, in the lower York River, at the end of a pier adjacent to the Virginia Institute of Marine Science, in Gloucester Point, Virginia. The mean depth of the water column was 3 m. Water temperatures varied from 26-29° C, salinity ranged from 19-2 3 ppt, and the water column was well mixed (this author, unpubl. data). Water currents near the bottom were typically 5-10 cm • s'1. The tidal range is about

1.0 m. The largest populations of oysters in recent history in the York River have been on large, subtidal oyster reefs, in recent years these populations have become sparse or extinct (Virginia Institute of Marine Science, oyster monitoring programs, unpublished data). At present, about equal numbers of oysters are present near the research site on intertidal, artificial substrates (pilings, rocks), and subtidally on scattered mollusk shells and anthropogenic debris.

A second study was undertaken June 1-28, 1993, in the blind ship channel of Harbor Branch Oceanographic Institution, Inc., on the Indian River Lagoon near Fort Pierce, Florida. The mean depth of the water column at the sample site was 1.5 m. Water temperatures were 26-29° C,

salinities were 28-32 ppt, and the water column was well

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 133 mixed (this author, unpubl. data). Water currents were typically 1-3 cm • s'1, and were primarily tidal in origin. The tidal range was less than 0.5 m. In and around the Harbor Branch ship channel, in Florida, highest densities of Crassostrea, by at least an order of magnitude, were found intertidally, and abundances quickly declined subtidally. Ostrea, on the other hand, was never observed intertidally, and most individuals were observed subtidally, on the lower surfaces of limestone boulders, at depths of greater than 0.5 meters below mean low tide (this author, unpubl. data).

Plankton Sampling In Virginia, plankton was sampled with a modified sump pump, suspended 20 cm above the bottom to avoid entraining sediment. Plankton was sampled four times daily, every six hours, at midnight, 6:00 am, noon, and 6:00 pm. Samples were usually about one hour in duration, and about 1-2 m3 in

volume. Samples were first run through a 400 p Nitex mesh to remove large particles, and pediveliger larvae were retained on a 150 /x Nitex mesh. Pediveliger larvae were identified and quantified in the laboratory. Abundance was calculated as pediveligers per cubic meter. In Florida, plankton was sampled with two modified 12 V bilge pumps, rated at 500 gallons (1800 1) per hour. One pump was suspended 2 0 cm below the surface, and the other

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 134 was suspended 20 cm above the bottom. Plankton was sampled twice daily, at mid-morning and mid-evening (daylight and dark) , for about two hours each time (1-2 m3 in volume) .

Preliminary sampling showed generally low variation in larval abundance within a day, probably due to the low currents and thus long residence period of any particular larval patch (Sephton and Booth, 1992; see also Chapter 3 of this dissertation); the above sampling regime was thought to account for variations in larval abundance. Larvae were collected and processed as in Virginia (above). The relative proportions of pediveliger larvae (from the sum of larvae per cubic meter in both near-surface samples and near-bottom samples) in near-surface and near-bottom water were calculated. The mean of these proportions were calculated over the sampling period, and null hypotheses of no effect of time of day (morning versus evening), no effect of depth (near-surface versus deep waters), and no interaction between the above two factors, were tested with two-factor analysis of variance (Zar, 1984) .

Estimation of Juvenile Mortality Daily mortality of settled juveniles on shell strings in Virginia may have affected apparent settlement of samples collected at seven day (or longer; one sample was not retrieved for 14 days) intervals. Virginia shell strings (described below, under Estimates of Settlement) were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 135 collected at only approximately seven-day intervals, and recruitment on lower (concave) surfaces only was quantified. To estimate total settlement, daily mortality rates of settled juveniles were estimated using several existing data sets collected in the York River. These data sets are described below. Roegner (1989, 1991) allowed Crassostrea virginica settlement onto unglazed ceramic plates, which were then suspended in the water column in the York River. Survival was measured at seven-day intervals. Data points are taken from four extreme low intertidal (-0.75 m) studies, each with ten substrate replicates, described in Roegner (1989). These data points are referred to in the Results as "Roegner 1-4". In May 1992, hatchery-reared3 Crassostrea virginica larvae were permitted to settled onto 10 x 10 cm squares of scallop shell (Placopecten magellanicus). After counting the settled juveniles on one surface, substrates were placed adjacent to the plankton sampling site, suspended on lines through holes drilled in the center of each shell. Seven replicate shells were used, and to obtain an estimate of the variance, data were pooled. Survival was determined at 11 and 19 days following settlement. The data from this is referred to in the results as "Scallop Shell" data.

3 Virginia Institute of Marine Science oyster hatchery.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 136 In July 1993, hatchery-reared Crassostrea larvae were permitted to settle onto defaunated but conditioned (Fitt et al,, 1990) oyster shells, both in the laboratory and in enclosures in the field. Twelve shells were used in the laboratory, and after permitting the larvae 24 hours for settlement, the settled juveniles on one surface only (concave) were counted. These shells were placed together in a mesh bag and suspended in the water column. Data from these were pooled, and are designated "Lab Set". Four samples of three shells each, from the field enclosures (designated "Enclosure Set 1-4"), were similarly treated. Juvenile survival on both Lab Set and Enclosure Set substrates were quantified at three and six days following deployment. The results are also part of the data set presented in Chapter 5; for further description of methods, refer to that manuscript. In September 1993, hatchery reared Crassostrea larvae were permitted to settle onto ten shell strings, which were prepared as described for shell strings used at the Florida site (below). Larvae were left with the shell strings for 24 hours, with neutral red stain at a concentration of about 5 ppm. Neutral red stain does not affect larval settlement (Baker, 1991), but does make them readily visible. Settled juveniles on the lower (concave) surfaces of the middle ten shells only, on each shell string, were counted, and the shell strings reassembled in the same order

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 137 as during settlement. Shells with no initial settlement were not included in subsequent analysis. The shell strings were then suspended in the water column, at the same site as previous substrates (above). Nine days later, they were

retrieved, and surviving juveniles counted. Each shell string was treated as one replicate, with an estimate of variance across the ten shells of the shell string. These data are designated "Shell String" data, in the results. Using the above data sets, proportional survival was plotted and regression analysis was used to compare it to time since settlement. For each data sequence, 100% survival at day 0 was included, although the regression models were not forced through 100% at the y-intercept. Both logarithmic and linear (Type I and Type II) models of mortality (Smith, 198 0) were used to determine best fit to the observed mortality pattern. From this, mean daily mortality rates were estimated, and used to calculate total Crassostrea settlement for each of the four York River shell string samples collected in 1990. Untransformed, logI0- transformed, and logc-transformed survival data were all

used to determine the characteristic of the relationship; the final results of these are summarized in Results. The equation of the linear model used was: y = 86.3 - 6.50x (Equation 1) where y = percent survival at time x. Constants (the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 138 y-intercept and the slope) are expressed in percentages; the daily mortality rate, or slope of the equation, expressed as a proportion, is 0.065. The y-intercept was not 100 (percent survival at day 0), because the equation was not forced through that value.

Estimation of Settlement The Virginia Institute of Marine Science shell string substrate was used as settlement substrate in both Virginia and Florida. The shell string consisted of 12 single valves from Crassostrea virginica adults, each with a mean total surface area of about 1000 cm2, with a coefficient of variation of about 46% (calculated from Morales-Alamo, 1992). A hole was drilled through the center of each valve, and valves were threaded onto a galvanized fencing wire, with the smooth, concave surfaces facing down. In Virginia, shell strings were deployed singly, as part of the continuing spatfall (oyster juvenile recruitment) monitoring program, at predetermined stations, suspended just above the bottom, and left for seven days. Upon processing, settled spat, or newly settled juveniles, were counted on lower (concave) surfaces only, and only on the middle 10 shells. One shell string station was present adjacent to the plankton sampling site in the York River. These were also described by Haven and Fritz (1985), and Barber (1992) further described shell string deployment.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 139 Using the relationship developed for juvenile oyster mortality (Estimation of Juvenile Mortality, above), weekly recruitment R in the York River was predicted by the equation:

6 R = 7 S - [S • M • (Et)] (Equation 2) 1 = 0

where S is daily settlement, M is daily proportional mortality, or 0.065, and t is time in days. It was assumed that there was no mortality on the first day; thus, mortality was summed over only six days, even though there were seven days of settlement. Equation 2 simplifies to: R = S • (7 - 21 • 0.065) or: R = S • 5.635 (Equation 3) It was assumed, for the sake of this equation, that the settlement rate was constant throughout the period each shell string was in the water. Daily settlement S can be estimated by using known values for recruitment R by reordering the above equation:

S = R / 5.635 (Equation 4) Daily settlement multiplied by 7 is the estimate of total, or weekly, settlement. Between July 12 and July 26, 1990, the shell string was not removed as usual; the resulting recruitment R was divided by 2 prior to putting into Equation 4, to make that data point comparable to others.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 140 The shell string monitoring program quantified settlement onto lower (concave) surfaces of shell strings only. In Virginia in 1990, four shell strings retrieved from the Piankatank River were examined in the laboratory to determine relative settlement on upper (rugose) and lower (concave) surfaces of shells. The ratios from this were used to estimate total abundance on shell strings deployed in the York River in 1990 during the plankton sampling period, in concert with the methods above. A t-test (Zar, 1984) was used to test the difference between upper and lower surfaces. In Florida in 1993, the Virginia shell string method was modified. Three shell strings were deployed simultaneously, adjacent to the plankton sampling site, and retrieved daily. They were threaded on plastic-coated fencing wire, and two days prior to deployment, were soaked in seawater, to permit bacterial growth, which enhances oyster larval settlement (Fitt et al., 1990). After retrieval, the shell strings were soaked in seawater with 10 ppm of neutral red stain, a vital stain which brightly colors recently settled oysters, and increases the accuracy of counting them. Settled juveniles were identified to species, and quantified for each day. The assumption was made that no significant mortality of settled juveniles had occurred.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 141 Analysis of Settlement as a Function of Larval Abundance Regression analysis was used to determine the strength of the relationship between observed larval abundance and the abundance of newly settled juveniles (settlers) on shell strings, testing the null hypothesis of no relationship at a = 0.05 (Zar, 1984). In Virginia, weekly settler abundance, corrected by the methods outline above, was regressed against the mean Crassostrea pediveliger abundance corresponding to the week that the shell strings were in the water. In Florida, daily settler counts of each species (Crassostrea and Ostrea) was regressed against

a) pediveliger abundance for the 24 hour period that the shell string was in the water (designated "direct" data),

and b) pediveliger abundance for a 24-hour period ending 12 hours before the shell strings were removed, and thus over­ lapping the shell string deployment period by only 12 hours (designated "off-set" data). The off-set data set was used to examine the possibility that larvae have a protracted substrate exploration period prior to settlement, so that larvae observed in the plankton a few hours before the termination of settlement would be unlikely to settle. Settlement by 24 hours, however, is high (see Chapter 2 of this dissertation). The variation in the larva-settlement relationship accounted for by the regression equation was determined. It was assumed that planktonic abundance was accounted for by

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 142 the sampling regime (above), and that there was no inhibition of or significant variation in settlement caused by the shell strings themselves (as discussed in the Introduction). The residual variation was therefore inferred to be primarily due to individual larval behavior variability and interaction with small-scale hydrodynamic processes, as discussed in the Introduction. The residuals, using the data set (see above) which gave the strongest relationship, were plotted and regressed against larval abundance to look for patterns.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 143

Results Planktonic Abundance Pediveliger larvae of the oyster Crassostrea virginica were present in Virginia in 1990, at peak mean daily densities of 2 5.7 larvae • m'3. Abundance and variation over

time at this site (Figure 1) have previously been reported and analyzed in Chapter 3 of this dissertation. Weekly abundance values are discussed separately, below. Abundance and variation of other species collected are reported separately in Chapter 3. Pediveliger larvae of the oysters C. virginica and Ostrea equestris were present in Florida, in addition to pediveliger larvae of an unknown shipworm species (Teredinidae). Abundance of these species over time are shown in Figure 2. For Crassostrea and the teredinids, a significantly (p < 0.05) higher proportion of pediveliger larvae were in the lower part of the water column, while the effect of time of day was not significant. Ostrea pediveligers tended be uniformly dispersed in the water column at night, but largely demersal during the day; this was born out by the significant (p = 0.012) interaction factor between time of day and water depth. Table 1 summarizes the above results, and Table 2 summarizes the analyses of variance.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 144 Juvenile Mortality Regression analysis was used to compare proportional survival of newly settled oysters from the various data sets (Roegner, 1991; Scallop Shell, Lab Set, Field Set, and Shell Strings) against time. The data used are summarized in Table 3. Untransformed, log10-transformed, and logc-transformed

survival data all produced a significant relationship (p < 0.0005), but the untransformed data (linear model) accounted for the highest proportion of the variation (r2 = 0.645). The equation derived from this, y = 86.3 - 6.50x has been discussed above, as Equation 1 in Materials and Methods. The relationship between survival and time is shown graphically in Figure 3.

Settlement Sixty percent (60.0) of recruitment onto shell strings from the Piankatank River in 1990 was onto lower (concave) surfaces (surfaces quantified in normal shell string monitoring). The difference between upper and lower surface recruitment was significant (p = 0.0018), so estimated total settlement onto lower surfaces of shell strings (below) for York River data was multiplied by a factor of 1.67 for total settlement.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 145 Shell strings were placed and retrieved in Virginia in 1990, on the following dates: July 5-12, July 12-26, July 26-August 2, August 2-9. Using observed number of settled oysters and the above equations, total weekly settlement was estimated. Observed numbers for the four periods were 7, 3, 9, and 3 juveniles, respectively; using the procedures described above, the estimates of total settlement per week for each of the four periods were 14.5, 3.1, 18.6, and 6.2, respectively (values were not rounded to whole numbers). Intermediate steps and total weekly settlement estimates are summarized in Table 4, along with mean planktonic abundance of pediveliger larvae for the respective time periods. Mean weekly plankton abundance for those periods were 2.00, 2.85, 6.39, and 0.89 larvae • m'3, respectively.

Settlement as a Function of Larval Abundance

The estimates of total settlement per week were plotted against mean observed planktonic abundance of pediveligers for the corresponding time period (Figure 4). Although the regression was positive (b = 1.98) and accounted for 43.1% of the variance (r2 = 0.431), the regression coefficient was

not significant (p = 0.344). The regression and analysis or variance are summarized in Table 5. Using the Florida data for Crassostrea and Ostrea, separately, settlement was regressed against planktonic abundance of pediveliger larvae (Figures 5 and 6). The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 146 relationships were stronger for the direct plankton data (the period directly corresponding to the period the shell strings were in the water), than for the off-set data (the period preceding the shell strings by 12 hours). For Crassostrea, the relationship using the direct data was positive (b = 0.407), but the regression accounted for only 11.0% of the variance (r2 = 0.110), and the regression

coefficient was not significant (p = 0.165). The regressions using both the direct plankton data and the off­ set plankton data, and the analyses of variance, are summarized in Table 5. For Ostrea, the relationship using the direct data was positive (b = 0.361), and the regression accounted for 41.7% of the variance (r2 = 0.417), which was significant (p =

0.003). Use of the off-set plankton data also gave a significant relationship (p = 0.013), but the regression

coefficient was lower (r2 = 0.312). Regression results and analyses of variance are summarized in Table 5. The residuals of the regression of settlement on the direct plankton abundance data were examined for Crassostrea and Ostrea in Florida. The residuals were nearly uniformly distributed; the regression of the residuals on planktonic larval abundance accounted for no detectable level of the variation for Crassostrea (r2 < 0.0005, p = 0.957) (Figure

7), and virtually none of the variation for Ostrea (r2 = 0.062, p = 0.305) (Figure 8).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 147 Discussion Depth Distribution Plankton samples at discrete depths in the Indian River channel, Florida, established two important points. First, it demonstrated that, at these levels of larval abundance, clear and consistent distribution patterns could be detected by the sampling methods used. This infers confidence in the larval abundance estimates used in the relationship between larval abundance and settlement. Second, the depth distribution patterns are consistent with previous findings by this author in the York River, Virginia (Chapter 3 of

this dissertation), and the generally overlooked work by Yoo et al. (1977, 1979) who found that late stage bivalve larval abundance was strongly affected by water depth, particularly in shallow waters. Distribution patterns of pediveliger larvae of the oyster Crassostrea virginica have previously been determined in the lower York River (see Chapter 3); although there is some inter-annual variation, the lowest densities are near the surface. This pattern was also true for this species in the Indian River in Florida, as for an unrelated shipworm (Teredinidae) species. Roegner and Mann (1990) showed that settlement of this species at this site was higher subtidally than intertidally, using hatchery-reared larvae. In the York River, highest juvenile and adult densities, if not the highest total population abundance, are intertidal.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 148 Michener and Kenny (1991) found a similar two-week recruitment pattern for this species in South Carolina, where adult oysters are almost exclusively intertidal (Bahr and Lanier, 1981) . The same trend was apparent for the Florida site, in this study. Again, pediveliger larvae of Crassostrea were found in lowest densities near the surface, yet by far the highest juvenile and adult densities were intertidal. For the other oyster in Florida, Ostrea, the patterns were not so clear. Again, during the day, larval densities were lowest in the surface. At night, however, they were highest in the surface (not significantly higher, but significantly different from day-time distribution). The juveniles and adults, unlike those of Crassostrea, are almost exclusively subtidal, with highest densities occurring in deeper water. Except for day-time distribution of Ostrea, distribution of juvenile and adult oysters appears inconsistent with distribution and settlement of late stage larvae. Differential mortality patterns following settlement probably account for observed adult distribution, but based on abundance alone, it seems that the trait for settlement low in the water column would have low fitness. In a similar system using an intertidal barnacle, Semibalanus balanoides, Le Tourneaux and Bourget (1988) found that the depth distribution of newly-settled barnacles

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 149 in the complete absence of adult conspecifics was the same as the distribution of adults at adjacent sites. If the oyster models are similar, then the adult distribution patterns are set primarily by settlement site choice, not subsequent mortality. This leaves the researcher the problem of reconciling larval and adult depth distributions. One possible reason for the difference between depth distribution of larval and adult oysters is that, just before settlement, oyster pediveligers redistribute themselves in the water column, moving from areas of favorable larval conditions to areas of favorable settlement conditions. In other words, they may stay in relatively deeper water until hours or minutes prior to settlement, and then move into the shallows to settle. Although Michener and Kenny (1991) saw higher settlement lower in the intertidal, they examined only a 15 cm vertical range of substrates (corrugated tubes) in the field. Within any given small, homogeneous environment, larvae may select the lowest area on which to settle, but in a large,

heterogeneous environment, settling larvae may first exhibit initial broad-scale depth selection (shallow areas), followed by fine-scale site selection (the lowest point on a given piece of substrate). There are at least two mechanisms whereby preferential settlement at depths different from pediveliger larval distribution would not be detected by the plankton sampling

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 150 design used in this manuscript. First the larvae could spend a relatively very short time adjusting from the larval depth to the settlement depth. Evidence presented in Baker (in press), however, in which near-surface Crassostrea pediveligers did not settle more readily than near-bottom pediveligers, is not consistent with a hypothesis of larvae moving to near the surface to settle. As an alternate mechanism, the pediveliger larvae could move to the settlement site along the benthos, away from the plankton samplers. To do this, however, the larvae would be continuously exposed to a variety of benthic predators (Breese and Phibbs, 1972; Steinberg and Kennedy, 1979; MacKenzie, 1981; Osman et al., 1989). In conclusion, therefore, larval movement just prior to settlement as an explanation for the apparent inconsistency between pediveliger and adult depth distribution does not seem more plausible than differential mortality following settlement, as discussed above.

Juvenile Mortality

Roegner (1991) found a Type I survival curve for post- metamorphic oysters, with high initial mortality, and decreasing proportional mortality with age. When the raw data of Roegner (1989) is combined with other data sets, however, extremely high overall variation in mortality rates was found, even though all data sets came from the same site

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 151 (lower York River, Virginia), on discrete substrates suspended in the water column, during the summer. This variation obscured any logarithmic relationship (Type I mortality curve); a linear relationship served as the best model for the relationship between proportional survival and time. This is not to say that for any given cohort at any give site, a Type I mortality curve does not exist, but rather that across several cohorts or several sites, it is not a useful model to describe mortality.

Settlement as a Function of Larval Abundance Even with estimates of total settlement derived from the linear model for post-settlement mortality (see above), the relationship between pediveliger larval abundance and settlement was obscured by high variability, in the York River, Virginia. This may have been a result of the limited number of settlement data points (4), but results were similar for Crassostrea in the Indian River, Florida, site, with many more data points. As seen for Ostrea in the Indian River, a relationship can be found, but even in that case, the regression accounted for less than half (42%) of the variation. It is possible that for any given day, larval abundance was not adequately estimated by the sampling regime, but indications overall were that the probability of significant error was low. In Florida, daily error should have been

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 152 accounted for by the relatively high number of data points. In Virginia, sample intervals were seven days, with four sample points daily; this should have provided a good estimate of mean larval abundance over the seven day period. In addition, as noted above, the plankton sampling regime was sensitive enough to detect depth-related larval abundance patterns. Thus, the weak relationship is probably not due to plankton sampling error. Martel et al. (1993) found a high relationship between pediveliger larval abundance of zebra mussels (Dreissena spp.) and subsequent settlement. That study, however, contained nearly four orders of magnitude variation in larval abundance. In Virginia and Florida, in this study, variations in larval abundance were primarily within one order of magnitude. The data was near the detection limit for the level of variation observed; an r2 of about 0.16 is

required for significance of a 1-tailed test (assuming that the relationship between larval abundance and settlement is positive) at a confidence level of 95% (a = 0.05) and a sample size of 19 (Zar, 1984). To assure detection of a relationship, at this magnitude of abundance variability and an r2 of 0.16, for a statistical power of 90% (0 = 0.10) and a confidence level of 95% (one tailed test, a = 0.05), the regression sample size may need to be as high as 51 (calculated from Zar, 1984, Equation 19.13). As Martel (1993) has observed, however, at higher variations in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 153 abundance, the regression is stronger and a smaller sample size would be reguired. The residuals that remain from the above regressions, or 1 - r2, are indirect measurements of the effect factors

aside from larval abundance on settlement. As discussed above, the sample regime appears powerful, and the use of the shell strings avoided interference factors caused by fouling organisms or predators (Dean and Hurd, 1980; Butman, 1989; Duggins et al., 1990; Hurlbut, 1991; Menge, 1991). What remains are variation in larval behavior and interactions with small-scale hydrodynamic factors. An examination of the residuals (Figures 7, 8) shows no patterns with respect to larval abundance. Until a better understanding of the quantitative roles of larval behavior and interactions with small-scale hydrodynamics in the field is gained, therefore, these factors must be regarded as variability inherent in the system.

Implications for the Shell String Survey The assumption of the spatfall monitoring program of the Virginia Institute of Marine Science is that there is a reasonable relationship between observed numbers of newly recruited oysters and subsequent recruitment to the oyster fishery. Inter-annual variations in settlement density, however, for any one site, are normally within one order of magnitude, and sample sizes over the entire sampling season

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 154 are only 19 (Barber, 1992). Since, in this study, there was high confidence in larval abundance estimates, the implication is that the variability occurred on the settlement substrates themselves. By inference, therefore, a single shell string, or even 19 shell strings across the sample season, is not a reliable predictor of settlement onto adjacent sites. In addition, the regression of survival across time (e.g., the week that the shell strings are deployed) failed to detect a significant relationship, despite the relatively large number of samples used, so there is poor confidence that the observed number of Crassostrea juveniles observed on a shell string is closely related to total settlement for that week. Unless inter­ annual variations in abundance are greater than one order of magnitude, therefore, inferences regarding relative reproductive output of oyster populations between should be cautious.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 155 References Bahr, L.K. and W.P. Lanier. 1981. The ecology of intertidal oyster reefs of the south Atlantic coast. U.S. Fish and Wildlife Service, Office of Biological Services. FWS/OBS-81/15. 105 pp. Baker, P., 1991. The effects of neutral red stain on settlement of oyster pediveligers, Crassostrea virginica (Grnelin, 1791). J. Shellfish Res. 10:455-456. Baker, P. In press. Competency to settle in oyster larvae, Crassostrea virginica: wild versus hatchery-reared larvae. Aquaculture. Barber, B.J. 1992. Oyster spatfall in Virginia waters: 1992 annual summary. Virginia Marine Resource Special Report, December 199 2. 12 pp. Breese, W.P. 1972. Ingestion of bivalve molluscan larvae by the polychaete annelid Polydora ligni. Veliger 14:274-275. Brousseau, D.J. 1978. Population dynamics of the soft- shell clam, Mya arenaria. Mar. Biol. 50:63-71.

Butman, C.A. 1989. Sediment-trap experiments on the importance of hydrodynamical processes in distributing settling invertebrate larvae in near-bottom waters. J.Exp. Mar. Biol. Ecol. 134:37-88. Buyanovski, A.I. 1991. Ecology of larvae and settlement of spat of Pacific mussels in Avacha Bay (Eastern Kamchatka).Soviet J. Marine Biology 16:189-195. Carriker, M.R. 1961. Interrelation of functional morphology, behavior and autecology in early stages of the bivalve Mercenaria mercenaria. J. Elisha Mitchell Scientific Soc. 77:168-241.

Cole, H.A. and E.W. Knight-Jones. 1949. The setting behaviour of larvae of the European flat ouster, Ostrea edulis L., and its influence on methods of cultivation and spat collection. U.K. Ministry of Agriculture and Fisheries, Fishery Investigations Ser. 2, Vol. 17, No. 3. 3 9 pp. Coon, S.L., W.K. Fitt, and D.B. Bonar. 1990. Competence and delay of metamorphosis in the Pacific oyster Crassostrea gigas. Mar. Biol. 106:379-387.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 156 Cranfield, H.J. 1973. Observations on the behaviour of the pediveliger of Ostrea edulis during attachment and cementing. Mar. Biol. 22:203-209. Dean, T.A. and L.E. Hurd. 1980. Development in an estuarine fouling community: the influence of early colonists on later arrivals. Oecologia 46:295-301. Duggins, D.O., J.E. Eckman, and A.T Sewell. 1990. Ecology of understory kelp environments. II. Effects of kelp on recruitment of benthic invertebrates. J. Exp. Mar. Biol. Ecol. 143:27-45. Eyster, L.S. and J.A. Pechenik. 1987. Attachment of Mytilus edulis L. larvae on algal and byssal filaments is enhanced by water agitation. J. Exp. Mar. Biol. Ecol. 114:99-110. Feller, R.J., S.E. Stancyk, B.C. Coull, and D.G. Edwards. 1992. Recruitment of polychaetes and bivalves: long­ term assessment of predictability in a soft-bottom habitat. Mar. Ecol. Prog. Ser. 87:227-2 38. Fitt, W.K., S.L. Coon, M. Walch, R.M. Weiner, R.R. Colwell, and D.B. Bonar. 199 0. Settlement behavior and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supernatants. Mar. Biol. 106:389- 394. Gaines, S.D., S. Brown, and J. Roughgarden. 1985. Spatial variation in larval concentrations as a cause of spatial variation in settlement for the barnacle Balanus glandula. Oecologia 67:267-272. Grigor'eva, N.I. and V.N. Regulev. 1992. Vertical distribution of larvae of the scallop Mitzuhopecten yessoensis and mussel Mytilus trossulus in Posyet Bay, Sea of Japan. Soviet J. Marine Biology 17:105-109.

Hargis, W.J., Jr., and D.S. Haven. 1988. The imperiled oyster industry of Virginia. Virginia Institute of Marine Science Spec. Rep. Applied Mar. Sci. Ocean Engin. 290. 130 pp. Haven D. and L.W. Fritz. 1985. Setting of the American oyster Crassostrea virginica in the James River, Virginia, USA: temporal and spatial distribution. Marine Biology 86:271-282.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 157 Haven, D.S., W.J. Hargis, Jr., and P.O. Kendall. 1978. The oyster industry of Virginia: its status, problems and promise. Virginia Institute of Marine Science Spec. Rep. Applied Mar. Sci. Ocean Engin. 168. 149 pp. Hurlbut, C.J. 1991. Community recruitment: settlement and juvenile survival of seven co-occurring species of sessile marine invertebrates. Mar. Biol. 109:507-515. Le Tourneaux, F. and E. Bourget. 1988. Importance of physical and biological settlement cues used at different spatial scales by the larvae of Semibalanus balanoides. Mar. Biol. 97:57-66. Lipcius, R.N., E.J. Olmi III, J. van Montfrans. 1990. Planktonic availability and settlement of blue crab postlarvae. Mar. Ecol. Prog. Ser. 58:235-242. Luckenbach, M.W. 1984. Settlement and early post­ settlement survival in the recruitment of Mulinia lateralis (Bivalvia). Mar. Ecol. Prog. Ser. 17:245-250. MacKenzie, C.L., Jr. 1981. Biotic potential and environmental resistance in the American oyster (Crassostrea virginica) in Long Island Sound. Aquaculture 22:229-268. Mann, R. 1988. Distribution of bivalve larvae at a frontal system in the James River, Virginia. Marine Ecology Progress Series 50:29-44.

Mann, R., E.M. Burreson, and P. Baker. 1991. The decline of the Virginia oyster fishery in Chesapeake Bay: Considerations for introduction of a non-endemic species, Crassostrea gigas (Thunberg, 17 93). J. Shellfish Res. 10:379-388. Martel, A., A. Mathieu, S. Findlay, S. Nepszy, and J. Leach. 1993. Daily settlement rates in zebra mussels correlate with abundance of veligers. (Abstract). Proc. Larval Ecology Meetings, Port Jefferson, NY: 29. Menge, B.A. 1991. Relative importance of recruitment and other causes of variation in rocky intertidal community structures. J. Exp. Mar. Biol. Ecol. 146:69-100. Michener, W.K. and P.D. Kenny. 1991. Spatial and temporal patterns of Crassostrea virginica (Gmelin) recruitment: relationship to scale and substratum. J. Exp. Mar. Biol. Ecol. 154:97-121.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 158 Mileikovsky, S.A. 1971. Types of larval development in marine bottom invertebrates, their distribution and ecological significance: a re-evaluation. Marine Biology 10:193-213. Minchinton, T.E. and R.E. Scheibling. 1991. The influence of larval supply and settlement on the population structure of barnacles. Ecology 72:1867-1879. Newell, C.R., H. Hidu, B.J. McAlice, G. Podniesinski, F. Short, and L. Kindblom. 1991. Recruitment and commercial seed procurement of the blue mussel Mytilus edulis in Maine. J. World Aquaculture Soc. 22:134-152. Nosho, T.Y., and K.K. Chew. 1972. The setting and growth of the Manila clam, Venerupis japonica (Deshayes), in Hood Canal, Washington. Proc. Natl. Shellf. Assoc. 62:50-58. Olafsson, E.B. 1988. Inhibition of larval settlement to a soft bottom benthic community by drifting algal mats: an experimental test. Marine Biology 97:571-574. Olmi, E.J., III, J. van Montfrans, R.N. Lipcius, R.J. Orth, and P. W. Sadler. 1990. Variation in planktonic availability and settlement of blue crab megalopae in the York River, Virginia. Bull. Mar. Sci. 46:230-243. Osman, R.W., R.B. Whitlatch, and R.N. Zajac. 1989. Effects of resident species on recruitment into a community: larval settlement versus post-settlement mortality in the oyster Crassostrea virginica. Marine Ecology Progress Series 54:61-73. Pawlik, J „R., C.A. Butman, and V.R. Starczak. 1991. Hydrodynamic facilitation of a reef-building tube worm. Science 251:421-424. Pechenik, J.A. 1990. Delayed metamorphosis by larvae of benthic marine invertebrates: Does it occur? Is there a price to pay? Ophelia 32:63-94. Perkins, E.B. 1931. A study of oyster problems in Barnegat Bay. Report of the Department of Biology of the New Jersey State Agricultural Experiment Station, for the year ending June 30, 1930: 25-47. Peterson, C.H. 198 6. Enhancement of Mercenaria mercenaria densities in seagrass beds: is pattern fixed during settlement or altered by subsequent differential survival. Limnol. Oceanogr. 31:200-205.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 159 Powell, E.N., H. Cummins, R.J. Stanton, Jr., and G. Staff. 1984. Estimation of the size of molluscan settlement using the death assemblage. Estuarine, Coastal, and Shelf Science 18:367-384. Powell, E.N., R.J. Stanton, Jr., D. Davies, and A. Logan. 1986. Effect of a large larval settlement and catastrophic mortality on the ecological record of the community in the death assemblage. Estuarine, Coastal, and Shelf Science 23:513-525. Prytherch, H.F. 1934. The role of copper in the setting, metamorphosis, and distribution of the American oyster, Ostrea virginica. Ecol. Monogr. 4:47-107. Rodriguez, S.R., F.P. Ojeda, and N.C. Inestrosa. 1993. Settlement of benthic marine invertebrates. Mar. Ecol. Prog. Ser. 97:193-207. Roegner, G.C. 1989. Recruitment and growth of juvenile Crassostrea virginica (Gmelin) in relation to tidal zonation. M.A. Thesis, College of William and Mary, Williamsburg, VA. 145 pp. Roegner, G.C. 1991. Temporal analysis of the relationship between settlers and early recruits of the oyster Crassostrea virginica (Gmelin). J. Exp. Mar. Biol. Ecol. 151:57-69. Roegner, G.C. and R. Mann. 1990. Settlement patterns of Crassostrea virginica (Gmelin, 1791) larvae in relation to tidal zonation. J. Shellfish Res. 9:341-346. Rumrill, S.S. 1990. Natural mortality of marine invertebrate larvae. Ophelia 32:163-198. Scheltema, R.S. 1974. Biological interactions determining larval settlement of marine invertebrates. Thalassia Jugoslavica 10:263-296. Sekiguchi, H., H. Saito, and H. Nakao. 1991. Spatial and temporal distributions of planktonic and benthic phases of bivalves in a tidal estuary. Nippon Bentosu Gakkaishi (Bull. Japan Assoc. Benthology) 40:11-21. Seliger, H.H., J.A. Boggs, R.B. Rivkin, W.H. Biggley, and K.R.H. Aspden. 1982. The transport of oyster larvae in an estuary. Marine Biology 71:57-72.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 160 Sephton, T.W. and D.A. Booth. 1992. Physical oceanographic and biological data from the study of the flushing of oyster (Crassostrea virginica) larvae from Caraquet Bay, New Brunswick. Can. Manu. Rep. Fish. Aquatic Sci. 2162. 61 pp. Shanks, A.L. and W.G. Wright. 1987. Internal wave-mediated shoreward transport of cyprids, megalopae, and gammarids and correlated longshore differences in the settling rate of intertidal barnacles. J . Exp. Mar. Biol. Ecol. 114:1-13. Shaw, W.N. 1965. Seasonal setting patterns of five species of bivalves in Tred Avon River, Maryland. Chesapeake Science 6:33-37. Smith, R.L. 1980. Ecology and Field Biology. 3rd ed. Harper and Row, Publ., New York, NY. 835 pp. Snelgrove, P.V.R., C.A. Butman, and J.P. Grassle. 1993. Hydrodynamic enhancement of larval settlement in the bivalve Mulinia lateralis (Say) and the polychaete Capitella sp. I. in microdepositional environments. J. Exp. Mar. Biol. Ecol. 168:71-109. Steinberg, P.D. and V.S. Kennedy. 1979. Predation upon Crassostrea virginica (Gmelin) larvae by two invertebrate species common to Chesapeake Bay oyster bars. Veliger 22:78-84. Thresher, R.E., G.P. Harris, J.S. Gunn, and L.A. Clementson. 1989. Phytoplankton production pulses and episodic settlement of a temperate marine fish. Nature 341:641- 643.

Tremblay, M.J. and M. Sinclair. 1992. Planktonic sea scallop larvae (Placopecten magellanicus) in the Georges Bank region: broadscale distribution in relation to physical oceanography. Can. J. Fish. Aquat. Sci. 49:1597-1615. Watzin, M.C. 1983. The effects of meiofauna on settling macrofauna: meiofauna may structure macrofaunal communities. Oecologia 59:163-166. Woodin, S.A. 1976. Adult-larval interactions in dense infaunal assemblages: patterns of abundance. J. Mar. Res. 34:25-41. Yoo, S.K., Y.U. Kim, and K.Y. Park. 1979. Improvement of the method of seed scallop production. (In Korean with English summary). Bull. Natl. Univ. Busan 19:55-62.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 161 Yoo, S.K., K.Y. Park, and M.S. Yoo. 1977. Biological studies on arkshell culture 1. Distribution of drifting larvae of the arkshell, Anadara broughtonii Schrenk. (In Korean with English summary). J . Oceanological Soc. Korea 12:75-81. Yoo, S.K. and H.Y. Ryu. 1985. Occurrence and survival rate of the larvae of Pacific oyster Crassostrea gigas in Hansan Bay. Bull. Korean Fisheries Soc. 18:47-476. Zar, J. H. 1984. Biostatistical Analysis. 2nd Ed. Prentice-Hall, Englewood Cliffs, NJ, 718 pp. Zimmerman, K.M. and J.A. Pechenik. 1991. How to temperature and salinity affect relative rates of growth, morphological differentiation, and time to metamorphic competence in larvae of the marine gastropod Crepidula plana? Biol. Bull. 180:372-386.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 162

Table 1. Mean proportional (percent) abundances of three species of bivalve pediveliger larvae at two times (morning and evening) and two depths (near-surface and near-bottom), at the Florida site. Standard deviations are given in parentheses. Sample sizes (n) given for each species, each time of day. Pediveliger Taxon_____Morning______Evening______All Times Crassostrea virginica near-surface 14.6 (18.8) 16.8 (30.3) 15.5 (24.5) near-bottom 85.4 (18.8) 83.2 (30.3) 84.5 (24.5) n = 19 n = 13 n = 32 Ostrea equestris near-surface 18.6 (29.4) 61.0 (41.9) 33.2 (38.6) near-bottom 81.4 (29.4) 39.0 (41.9) 66.8 (38.6) n = 19 n = 10 n = 29 Teredinidae sp. near-surface 23.2 (39.1) 26.8 (33.4) 24.2 (38.3) near-bottom 76.8 (39.1) 73.2 (33.4) 75.8 (38.3) n = 20 n = 8 n = 28

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 163

Table 2. Summary of analysis of variance for the effects of time of day (morning versus evening) and depth (near-surface versus near-bottom) on abundance of three species of bivalve pediveliger larvae, at the Florida site.

2a: Analysis of Variance for Crassostrea virginica Source DF Sea SS Adi SS Adi MS F P Time of Day 1 7.779 7.779 7.779 1.89 0.174 Depth* 1 46.685 36.225 36.225 8.81 0.004 Time x Depth 1 7.967 7.967 7.967 1.94 0.169 Error 58 238.417 238.417 4.111 Total 61 300.848

2b: Analysis of Variance for Ostrea equestris Source DF Sea SS Adi SS Adi MS F P Time of Day 1 10.701 10.701 10.701 1.39 0.244 Depth 1 41.905 15.269 15.269 1.98 0.165 Time x Depth* 1 52 .386 52.386 52.386 6.80 0.012 Error 54 416.069 416.069 7.705 Total 57 521.062

2c: Analysis of Variance for Teredinids Source DF Sea SS Adi SS Adi MS F P Time of Day 1 0.0161 0.0161 0.0161 0. 02 0.898 Depth* 1 16.6334 10.9201 10.9201 11.26 0.001 Time x Depth 1 0.7875 0.7875 0.7875 0.81 0. 372 Error 52 50.4180 50.4180 0. 9696 Total 55 67.8550 * Indicates significance at a = 0.05

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 164

Table 3. Values used to estimate mortality of Crassostrea juveniles, over time (days following settlement) at the York River, Virginia, site, listed by source (see Methods and Materials). Sources designated "Roegner" are taken from Roegner (1989). N = number of replicate substrates in each data source, S.D. = standard deviation. Source Time N Survival S.D. Roegner 1 8 10 14.6 10.1 Roegner 2 8 10 5.6 7.2 Roegner 3 8 10 6.6 9.2 Roegner 4 8 10 22.8 8.7 Roegner 1 15 10 8.4 5.6 Roegner 2 15 10 1.4 3 . 0 Roegner 3 15 10 3 . 8 5.5 Roegner 4 15 10 8.7 6.6 Scallop Shell 11 7 60. 0 14.9 Scallop Shell 18 7 6.8 8 . 3 Lab Set 3 12 74 .7 11.2 Lab Set 7 12 59.8 16.2 Enclosure Set 1 3 3 5.5 7.8 Enclosure Set 2 3 3 5.5 7.2 Enclosure Set 3 3 3 14 . 5 14.1 Enclosure Set 4 3 3 17 . 3 12 . 5 Enclosure Set 1 7 3 0.9 1.2 Enclosure Set 2 7 3 1.7 2 . 5 Enclosure Set 3 7 3 4.9 3.7 Enclosure Set 4 7 3 7 . 6 7 . 3 Shell String 1 9 8 45.8 40.6 Shell String 2 9 9 27 . 8 41.6 Shell String 3 9 10 48 . 3 33.5 Shell String 4 9 9 29.6 32.2 Shell String 5 9 7 29.2 34.5 Shell String 6 9 9 44.9 38.4 Shell String 7 9 8 16.7 33 . 3 Shell String 8 9 8 29 . 5 34 . 6 Shell String 9 9 7 7 . 1 17.5 Shell String 10 9 8 6.7 11. 6

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 166 Table 5. Summary of analyses of regressions of oyster settlement on plankton abundance, for York River, VA and Indian River, FL.

Table 5a. Crassostrea virginica, York River 1990 Regression Equation: 4.60 + 1.98 Regression Coefficient: r 0.431 Analysis of Variance SOURCE______DF______SS ______MS Regression 1 66.58 66.58 1.51 0.344 Error 2 88.00 44.00 Total 3 154.58

Table 5b. Crassostrea virginica, Indian River 1993 (Simultaneous Plankton Data) Regression Equation: y = 0.486 + 0.407-X Regression Coefficient: r2 = 0 . 110 Analysis of Variance SOURCE______DF______SS ______MS______Regression 1 2.697 2 . 697 2 .10 0.165 Error 17 21.830 1. 284 Total 18 24.526

Table 5c. Crassostrea virginica, Indian River 1993 (Off-Set Plankton Data) Regression Equation: y = 0. 689 + 0.182-X Regression Coefficient: r~2 _= 0.033 Analysis of Variance SOURCE______DF______SS MS______Regression 1 0.807 0.807 0. 58 0.457 Error 17 2 3.72 0 1. 395 Total 18 24.526

Table 5d. Ostrea equestris, Indian River 1993 (Simultaneous Plankton Data)

Regression Equation: y = 0.181 + 0.361-X Regression Coefficient: r2 = 0.417 Analysis of Variance SOURCE______DF______SS ______MS Regression 1 19.874 19.874 12 . 15 0.003 Error 17 27.811 1.636 Total 18 47.684

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 167 Table 5. (cont.) Table 5e. Ostrea equestris, Indian River 1993 (Off-Set Plankton Data) Regression Equation: y = 0.12 6 + 0.583-x Regression Coefficient: r2 = 0.312 Analysis of Variance SOURCE______DF______SS______MS______F E__ Regression 1 14.900 14.900 7.73 0. 013 Error 17 32.784 1.928 Total 18 47.684

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 168

Figure 1. Abundance of Crassostrea virginica pediveliger larvae, July 5 - August 9, 1990, at the York River, Virginia, site.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcdwt priso o h cprgt we. ute erdcin rhbtd ihu permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

PEDIVELIGER ABUNDANCE 1990 .O 5 s co co co 2 co CO O O O O O LO O LO O LO O 's f CO CO C\J C \l \l C C\J CO CO f 's 4 ---- i i i i euj/aoNvaNnav ------1 ------t — 1 ------t — r~ < < I CO LO

SAMPLE SEQUENCE (DAYS) 169

Figure 2. Abundance of three species of pediveliger larvae, May 3-31, 1993, at the Harbor Branch, Florida, site.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Teredinid INDIAN RIVER, MAY INDIAN 1993 RIVER, DAY OF SAMPLINGDAY (1/2 day intervals) —is— Crassostrea—is— *= Ostrea ® I ® PEDIVELIGER ABUNDANCE LARVAL E CO LU O z < Q Z Z> CD < < _l > DC i

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 170

Figure 3. Survival of recently settled Crassostrea virginica at the York River, Virginia, site. Taken from various sources (see Methods and Materials).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcdwt priso o h cprgt we. ute erdcin rhbtdwtot permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

100* / o o o o o o ocoNcoin^cooji- o o o o CM O C o "si* LO II IBAjAjns % IBAjAjns 44 i r i i i «< 4

4 4 4 <

444 444 4 41 4

O CO LO C 10 -CO .CO .00 -"sf C\J o * c ■*" I ! £ o Q co M C sz 0 0 if) 0 0 171

Figure 4. Estimated settlement of Crassostrea virginica onto shell strings at the York River, Virginia, site, versus pediveliger larval abundance in the plankton.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcd ih emsin fte oyih onr Frhrrpouto poiie ihu permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

CRASSOSTREA VIRGINICA C\J O LO NfOO VdS lV O flO IN O) O LO - -CD -CD O

LARVAL ABUNDANCE / m3 172

Figure 5. Settlement of Crassostrea virginica onto shell strings at the Harbor Branch, Florida, site, versus pediveliger larval abundance in the plankton.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. <

LO 00

CO CD CD l-co ® E i— 1— - t— c c d LJJ § z II CM ■3 8 CD < L. £ Q 2 z

5 m E ^ < co f LO — 1

8 oc s

LO o

LO 'si* co CM t - o

i n h o o ivds

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 173

Figure 6. Settlement of Ostrea equestris onto shell strings at the Harbor Branch, Florida, site, versus pediveliger larval abundance in the plankton.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcd ih emsin fte oyih onr Frhrrpouto poiie ihu permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

OSTREA EOUESTRIS O O CM CO ^ LO ivds d v i o o n N i

LARVAL ABUNDANCE / nri 174

Figure 7. Residuals of the regression of Crassostrea virginica settlement on planktonic larval abundance (direct data set), plotted against planktonic larval abundance.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. erdcd ih emsin fte oyih onr Frhrrpouto poiie ihu permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

CRASSOSTREA CO O O O y e s j p n B i s • O T““ CNJ CO CNJ O

Pediveliger Abundance 175

Figure 8. Residuals of the regression of Ostrea equestris settlement on planktonic larval abundance (direct data set), plotted against planktonic larval abundance.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CN

00

5 CN K CD CO O Pediveliger Pediveliger Abundance

CN

CO CN T—O T—

s | B n p j S 0 y

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5

PROPORTIONAL SETTLEMENT AND RECRUITMENT OF Crassostrea virginica: A QUANTITATIVE FIELD TEST USING LARVAL ENCLOSURES

176

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 177

Table of Contents Page Abstract...... 178 Introduction...... 179 Materials and Methods...... 183 Research Site...... 183 Enclosure Treatments...... 184 Enclosure Deployment...... 186 Enclosure Sampling...... 189 Treatment Effect Analysis...... 192 Defaunated Treatment Study II...... 194 Settlement Orientation...... 195 Analysis of Within-Reef Variables...... 196 Results...... 199 Sedimentation...... 199 Effect of Distance fromInput Port...... 199 Treatment Effects...... 2 00 Defaunated Treatment Study II...... 202 Settlement Orientation...... 203 Within-Reef Variables...... 204 Discussion...... 206 Effectiveness of Larval Enclosures...... 206 Treatment Effects...... 207 Settlement Orientation...... 208 Within-Reef Variables...... 209 References...... 214

Tables...... 219 Figures...... 235 Appendix...... 237

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 178 Abstract The effect of an oyster reef fouling community on settlement and survival of oyster, Crassostrea virginica, larvae and juveniles, was examined. Larval enclosures were used to retain known quantities of larvae over plots and controls in the field, during initial settlement, and settlement and survival were subsequently sampled in each treatment. The overall effect of the reef community was strong, with significantly lower settlement and survival to 28 days compared to the defaunated controls. In addition, variability of settlement and survival, for the first few days, was higher in the reef treatment. Within the reef treatment, however, the correlation of settlement with percent cover of total fouling community or the three major fouling organisms (the sponge Cliona celata, the barnacle Balanus eburneus, and an unidentified cheilostome bryozoan) was poor. One alternate explanation for lower survival on reefs, backed by observations, is that the presence of small predators such as turbellarian flatworms strongly effect survival.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 179 Introduction The quantitative role of benthic communities on settlement or subsequent recruitment is a topic of interest in marine ecology. The vast majority of studies, however, have examined either settlement from the plankton, or subsequent recruitment (survival), but not both.

There have been numerous studies on the role of marine benthic communities, or components of those communities, on recruitment of arriving invertebrate larvae (Dean and Hurd, 1980; Watzin, 1983; Luckenbach, 1984; Hunt et al ., 1987; Davis et a l ., 1989; Young, 1989; Duggins et a l ., 1990; Hurlbut, 1991). Some other studies have taken this to the next logical step, and examined whether recruitment patterns are set during larval settlement, or by subsequent differential survival (Peterson, 1986; Osman et al., 1989; Menge, 1991). In most cases, the focus has been on the role of a specific species (Luckenbach, 1984; Young, 1989; Davis et a l ., 1991) or subset of the benthic community (Watzin, 1983; Maki et al., 1989; Duggins et al., 1990).

Fewer quantitative studies have been conducted on the role of larval supply to settlement (Shanks and Wright, 1987; Minchinton and Scheibling, 1991). Chapter 4 of this dissertation reviews these latter works. Determining larval abundance in the field is a daunting proposition, as discussed below.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 180 Settlement of benthic organisms from the plankton, onto natural substrates, has proven difficult to measure in the field. Research challenges and sources of error include a) difficulty in distinguishing settled, metamorphosed juveniles from the exploratory stage of the final larval stage (especially for most bivalve mollusks: this author, unpubl. data), or b) from a postlarval stage that may or may not resuspend in the plankton (for examples within Mollusca, see Sigurdsson et al., 1976; Board, 1983; Martel and Chia, 1991; Armonies and Hellwig-Armonies, 1992), c) difficulty in distinguishing newly settled juveniles from older juveniles (partly from a lack of data on immediate post-metamorphic growth rates), and d) mortality and disappearance of juveniles between sampling periods (e.g. Luckenbach, 1984; Powell et a l ., 1984; Powell et a l ., 1986). Furthermore, no prior field settlement study has started with a knowledge of the number of larvae available to settle from the water column. Certain groups of organisms attach to the substrate immediately upon settlement, and do not have a postlarval stage, making settlement a precisely defined event. Well- known examples include barnacles (Visscher, 1928; Crisp, 1961), ascidians (van Duyl et a l ., 1981; Davis, 1987), bryozoans (Maki et al., 1989), spirorbid polychaetes (Wisely, 1960), articulate brachiopods (Freeman, 1993), and oysters of the family Ostreidae (Prytherch, 1934; Cranfield,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 181 1973; Bonar et al., 1990). Oyster settlement and metamorphosis in particular is well-described, and in

addition, oyster larvae are readily available from hatcheries. Use of an oyster species as a research model overcomes the first two of the research difficulties mentioned above. Marking larvae and releasing them over the settlement site permits the researcher to precisely identify the age of subsequent juveniles, and confusion with older or subsequent settlement is eliminated (the third research difficulty mentioned above). Several marking techniques for bivalve larvae and juveniles have been tested (for review, see Levin, 1990). An additional advantage of marking is that it makes very small juveniles much easier to find and quantify, and increases the accuracy of sampling. Releasing marked larvae into an open system immediately raises the problems of dilution, and the number of larvae available to settle at the study site becomes difficult to estimate. A solution to this is to place temporary enclosures over the settlement substrate, and to place a known quantity of marked larvae within the enclosure. This approach has previously been used on a small scale by Young and Chia (1982). Enclosures to contain larvae must remain virtually watertight under field conditions, because larvae are small (300 ju for Crassostrea virginica larvae) . Following a period for settlement, the enclosures can be

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 182 opened or removed, permitting a full return to natural processes that might affect survival of juveniles. The previously stated problems with measuring settlement and survival in the field, therefore, can be overcome by a) using a species such as Crassostrea virginica, which has a precisely defined settlement event;

b) marking the larvae with a stain such as neutral red, so that they can be readily quantified, and distinguished from older or younger animals; and c) using enclosures to temporarily hold larvae in the field while they settle onto the study site. The research described in this paper makes use of experimental field enclosures to contain pediveliger larvae of oysters, Crassostrea virginica, over natural (termed "reef" hereafter) and defaunated oyster shell substrates. This study is designed to quantify proportional settlement

and survival of Crassostrea with some precision, starting from a known number of larvae. In addition, the quantitative effects of specific substrate characteristics, including substrate size, fouling, and specific fouling by dominant benthic fouling organisms (sponges, bryozoans, and barnacles) on settlement and survival of oysters, will be examined.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 183

Materials and Methods Research Site The research area was in the lower York River, a sub­ estuary of Chesapeake Bay, in Virginia. Two sites were used; a primary site at 2 m mean depth, and a back-up site at 1.5 m mean depth. Both sites were near the north shore of the York River. Temperature and salinity measurements, and flume water intakes, were situated at the end of a pier between the two sites, at a mean depth of 2.2 m. Salinity at this site varied from 19-22 ppt throughout the study period (June 28 to August 12, 1993), and temperature varied from 27-31° C. The Virginia Institute of Marine Science

maintains a Crassostrea settlement monitoring station at this site; no natural settlement at this or any nearby sites was observed during the research period. Natural oyster reefs in the York River have died out and been largely buried by sediment, although scattered shells remain on the surface (pers. obs). An artificial oyster reef was prepared, therefore, at part of the 2 m research site. In 1991, oyster shells were placed over the site, to form a layer about 15 cm thick. These shells mostly sank into, or were buried by, sediment, but provided a base for subsequent shell. In 1992, oyster shells, heavily fouled by sessile invertebrates, were dredged from areas in the lower York River where shells were still

present on the surface, and placed on the 2 m site, over the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 184 sediment-covered shells. The 1992 shells remained on the surface, and fouling organisms quickly recovered, according to visual surveys by scuba diving.

Enclosure Treatments Settlement enclosures were developed to retain known quantities of competent-to-settle Crassostrea larvae over the substrate treatments, during the settlement period. Design constraints included enclosure strength and integrity under tidal currents, and retention of larvae of 3 00 ju in size, while permitting water exchange to prevent oxygen depletion or other water chemistry imbalance within the enclosure, for a period of 24 hours. Following the settlement period, the bases of the enclosures also served as a sampling reference. Settlement enclosures were made from Plexiglas clear acrylic, 6.5 mm thick. Each enclosure was square; internal dimensions were 1 m on a side, and 15 cm in depth. There was a lid for each enclosure which fitted snugly over the enclosure, and rested on a 6.5 mm wide flange, 2 cm down from the edge all around the outside of each enclosure. The edges of the lid were made of 6.5 mm clear acrylic, but the top was made from 2.5 mm clear acrylic. Plexiglas acrylic is nearly neutrally buoyant, and requires weight to hold it down; to reduce the weight required, 2.5 mm acrylic was used wherever strength was not required. Each lid had four

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 185 holes, about 40 mm in diameter, around the edges, covered with 150 n Nitex mesh, to permit water exchange, and reducing potential water chemistry or dissolved oxygen imbalances. Although it is possible that slight hypoxia might develop, less than 24 hours of exposure would have an undetectable effect on survival (Baker and Mann, 1992). A fifth hole was placed in the center, and fitted with a screw-on lid, to permit the injection of larvae (Figure 1). Lead sheeting, in 0.5 kg squares, were attached to the upper side of two corners of the lids, for weight. Additional weights were placed on the lids while deployed. The lead weights, screw-on lid, and Nitex mesh were attached with non-toxic clear silicone sealant. The enclosures described above are much larger than, but otherwise very similar to larval enclosures used in the field by Young and Chia (1982), in both construction and deployment.

Sedimentation occurred onto shell substrates within both enclosures during the experiment. On Day 28, relative sedimentation onto shells inside and outside of the enclosures was calculated. This was to examine the

possibility that the enclosures significantly affected

sedimentation onto the substrates. Next to the defaunated treatment enclosures (see below) were areas of unused bare shells, placed on June 29. Four 10 x 10 cm shell samples each from within, and outside of, the enclosures were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 186 carefully collected, taking care not to dislodge sediment, and placed into sealable plastic bags. Samples within reef treatment enclosures (see below) and over adjacent shell reef were similarly taken. In the laboratory, the sediment was rinsed through a 500 p Nitex mesh (the size of the largest particles collected in previous near-bottom plankton samples; this author, unpubl. data), and then collected on pre-weighed P4 filter paper. Samples were oven-dried at 80° C for 12 hours, and reweighed. Sedimentation was

calculated both as sediment per 10 x 10 cm area, and by shell mass. Two factor analysis of variance was used to compare sedimentation within and outside enclosures, and between defaunated and reef treatments.

Enclosure Deployment Twelve enclosures were made, for two treatments: reef (the artificial oyster reef), and defaunated shells. The enclosures for the reef treatment did not have bottoms, and were dug by hand 1-3 cm into the sediment beneath the oyster shells, so that there was no water flow under the lower edge. Subsequent observations showed that, with a single exception, these remained firmly imbedded in the sediment for the one-month period of the experimental run. Each of the reef enclosures was weighted with 2 kg of lead sheeting, attached to the outer surfaces. The enclosures for the defaunated shell substrate had bottoms made of 2.5 mm

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 187 acrylic. Defaunated shell substrate consisted of adult Crassostrea shell, air dried for several weeks. About 30 kg of dry Crassostrea shell was placed into each enclosure, forming a layer about 5-10 cm thick, visually approximating a shell density similar to that in the reef enclosures. Enclosures within a treatment (reef or defaunated) were placed about 0.5 to 1.0 m from each other, and the defaunated shell enclosures were about 30 m from the reef enclosures (see Figure 2). Primary tidal current vectors were parallel to the axis created by the two treatments: whatever flowed over one flowed over the other within minutes. Lids were placed on both enclosures, and enclosures were allowed to sit for one day to allow shells to shift, and to permit bacterial growth on the defaunated shells. Bacterial films have been shown to enhance settlement of oyster larvae (Fitt et al., 1990). Crassostrea virginica larvae were raised to the pediveliger stage in the Virginia Institute of Marine Science oyster hatchery. Several hours before use, neutral red stain was placed into the water in which they were contained, to a concentration of about 10 ppm. Several marking techniques for bivalve larvae and juveniles have been used, including alizarin red (Hidu and Hanks, 1968), tetracycline (Dey and Bolton, 1978), light-induced pigmentation banding (Trevelyan and Chang, 1987), and others (for review, see Levin, 1990); however, neutral red, a vital

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 188 stain, is simple to use, and has been shown to have little or no harmful effects to the larvae (Loosanoff and Davis, 1947; Manzi and Donnelly, 1971; Baker, 1991). This treatment gave the larvae a strong red color that made them easier to quantify, and which stained the juveniles until they had grown substantially. Approximately 100,000 larvae were placed into each enclosure on June 30. In the laboratory, larvae were placed into a large beaker, and vigorously agitated with a perforated acrylic plunger, so that they were evenly distributed in the water. Twelve equal aliquots of known volume were siphoned from the beaker, along with the larvae. From each aliquot, three samples of 0.25 ml were taken (during agitation) and preserved, to be counted later. From the mean of these samples, multiplied by the volume of the aliquots, the actual number of larvae placed into each enclosure was estimated. These values are also given in

Table 3, in Results. Larvae from each aliquot were then placed into lengths of tubing, stoppered at each end. Under water, at the research site, a diver introduced the larvae into the enclosures via the central input port (Figure 1) of each enclosure lids. The stoppers were loosened (but not removed), and one end of the tubing was placed through the input port into the enclosure. The lower stopper was then knocked off against the shell substrate, allowing the negatively buoyant larvae to fall from the hose into the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 189 enclosure. The upper stopper was removed, and the hose blown through gently to evacuate any remaining larvae. The hose was then removed, and the lid on the center hole replaced.

Enclosure Sampling

Twenty four hours after placing the larvae in the enclosures, the enclosure lids were removed. This was designated as Sampling Day 0, and three subsamples were taken from each enclosure. A i m grid, divided into 10x10 cm squares, was placed over an enclosure. Gird squares were chosen randomly by tossing weighted flags into the grid. Perimeter grid squares (>4 0 cm from the center) were not sampled, to avoid possible edge effects. Flags that landed in a square adjacent to another were re-thrown, and the grid squares along the enclosure edges were not sampled. All exposed shells from each grid were collected, and shells that overlapped with another grid were assigned to that grid which contained their largest portion. Samples were placed into plastic bags, and their position within the enclosure- near (within 2 grids of the center) or far (greater than 2 grids from the center)- was recorded. Subsequent samples were taken in a procedure similar to that described above, on Day 1 (July 1), Day 3, Day 7, Day 14, and Day 28. Previously sampled grid sites were not used again. On Day 28 (July 28), reef Enclosure 1 was not

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 190 sampled, because it had come loose from the sediment and moved from its original site. To regain a balanced statistical design, the missing value was estimated using the Shearer technique (Zar, 1984). On Day 43 (August 12), 0.25 m2 samples were taken from the reef enclosures 2-5, to estimate abundance of living Crassostrea juveniles, because they were too sparse to be detected in 0.01 m2 samples on

Day 28. Day 43 data were used in regression analysis only. The Plexiglas acrylic lids and enclosures were visually examined for settlement of Crassostrea. Shell height (an axis across the shell that includes the hinge) of each Crassostrea juvenile observed in the defaunated enclosure subsamples were recorded on Day 28, and in the reef enclosures on Day 43. Juvenile Crassostrea (spat) were counted on all shell surfaces under a dissecting microscope. Counting error was estimated by re-examining subsamples of previously quantified substrate under higher magnification. Counting error on Day 0 and Day 1 was very low (<1%), because spat were brightly stained. On Day 14 and Day 28, counting error was also very low, because the spat were several mm or greater in size, and clearly visible to the naked eye. On Day 3 and Day 7, there was some counting error because spat had lost their stain, and remained small, or because dead spat were counted. Mean error was -3.2% (under-count) on Day 3 (n = 6), with a standard deviation of 22%, and 6.6%

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 191 (over-count) on Day 7 (n = 8) with a standard deviation of 18.3%. In light of the relatively low mean error and high variation, it was determined not to adjust recorded counts for Days 3 and 7. On June 29, when the larvae were introduced into the enclosures, an unused portion of the same larvae were used in a laboratory assay to determine proportional settlement in the absence of field conditions. Twelve circles of oyster shell, each approximately 3 cm in diameter, were previously conditioned in seawater for 24 hours. Each was placed into a culture dish with 15 ml unfiltered seawater from the York River, and 100 Crassostrea larvae were placed into each culture dish. A Captrol micropipette was used to count the larvae. The culture dishes were covered to prevent evaporation, and to reduce light levels, and left for 24 hours, at 23° C. At the end of 24 hours, the

proportion of settled larvae was determined. This was intended to be taken as the optimal settlement rate of the larvae. This assay was modified from Baker (1991). From Day 0 samples collected from defaunated enclosures, 12 samples (2 from each enclosure) were reserved for subsequent survival and growth rate measurements in a sheltered environment. Shells were placed in Nitex 1 cm mesh bags, in a flowing seawater (raw York River water) flume in the laboratory. Survival was measured on Day 3,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 192 Day 7, Day 14, and Day 28, and shell heights were measured on Day 28. Regression analysis was used to determined the strength of the relationship between Crassostrea juvenile survival in the flume, and time.

Treatment Effect Analysis Distance of any subsample from the central input port affected Crassostrea juvenile counts (see Results). The effect of distance on proportional settlement and survival data in particular was strong, and analysis of proportional data required that this effect be first accounted for. Determination of initial proportional settlement in each enclosure required a) the relationship of settlement density to distance from input port, and b) the distance of each subsamples within an enclosure. Using the recorded locations of each subsample, the relationship of count data to distance from the input port was determined with regression analysis (Zar, 1984). From this (Table 2), the distance at which predicted density was \ of the density at distance 0 (the y-intercept) was calculated. Data for each

subsample were then "corrected" for distance, using the equation: Cl = Cj * Xj/x,A (Equation 1) where C = count data for subsample i, Cl = corrected count

data, X; = distance of subsample i from the central input

port, and x,A the distance from the central port at which the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 193 predicted count is equal to one half of the y-intercept. The mean of corrected subsamples within an enclosure on Day 0 was taken as settlement • 100 cm'2. Proportional

settlement was then calculated as this value, divided by 1% of the larvae introduced to that enclosure. Settlement and survival of Crassostrea juveniles in each enclosure, for each sample date, was calculated as mean individuals • 100 cm'2 grid area. Juvenile abundance was analyzed separately for each sample date, when testing the null hypothesis of no effect of treatment, using the distance-corrected data (above), and two sample t-tests (Zar, 1984) at a = 0.05. The total variance (across all subsamples) of the

distance-corrected data was statistically examined. Total variance (n = 18) was expressed for each treatment and sample date as a coefficient of variation, or the standard deviation divided by the mean (Zar, 1984). For each date, the null hypothesis of no difference between treatment coefficients of variation was tested with an F-test (Zar, 1984), at a = 0.05: this test uses log10-transformed data,

which accounts for reduction of variance by many zero data points, and can thus sometimes give a different interpretation than a simple coefficient of variation. Regression analysis (Zar, 1984) was used to examine the null hypothesis of no relationship between proportional survival (using corrected data as described above) and date

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 194 of sampling (Days 1-28, and Day 43 for the reef enclosure treatment), for the two treatments (reef and defaunated substrate enclosures). A t-test was used to test the null hypothesis of no difference between the slopes of proportional settlement of the two treatments (Zar, 1984).

Defaunated Treatment Study II On August 3, four enclosures were used in a test to further examine dispersal of larvae within the enclosures, and to compare laboratory settlement and survival to field settlement and survival. Results were also used in the settlement orientation study (see below). The back-up research site was used, in 1.5 m mean water depth. The procedure for placement of shell and treatment of larvae was the same as described above for the June 3 0 experiment. Total larvae in each enclosure are given in Table 2, in Results. Samples were taken on Day 0 and Day 1, but could not be taken on Day 3, due to inclement weather. The final sample was taken on Day 6. The positions of the spat on

each shell (total number on smooth, concave surface, versus rough, convex surface), were recorded. As for the primary experiment (June 30), subsamples from Day 0 of this study were saved, but half of these were placed in the laboratory flume, and half were suspended in mesh bags from a pier adjacent to the research site. Proportional survival of these juveniles was calculated on

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 195 Days 3, and 7, with the aid of a dissecting microscope. The purpose of this was to examine the effect of the pelagic environment alone on survival of Crassostrea juveniles. Regression analysis was used to test the strength of the relationship between time and survival in both treatment, and the null hypothesis of no difference between treatments was tested with a t-test (Zar, 1984). A laboratory assay to determine settlement rates under sheltered conditions was executed for the back-up study; the details of this assay are given below.

Settlement Orientation On Day 0, the percent of the defaunated Crassostrea substrate shells from the enclosures in Defaunated Treatment Study II (see above) that had the concave (inner) surface facing upwards was estimated, by recording position of those collected in subsamples. Shells lying on their side (rare) and shells of other species (mainly Mercenaria mercenaria) were not counted. Using the proportional orientation of shells, and the proportional settlement of spat on concave versus convex surfaces (above), larval settlement orientation preference in the field was estimated. The laboratory settlement assay for the Defaunated Study II was modified and expanded from the June 3 0 experimental run assay (above). Whole oyster shells, from the same source as shells for the enclosures, were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 196 conditioned for one day in seawater. They were placed into containers with 1 1 of raw York River water, and placed into a laboratory at ambient temperature (mean = 29° C) , with low light levels. The oyster shell positions were alternated, so that half had the concave surface facing up, half down. Approximately 500 stained larvae (taken by aliquot from a volume of water with total larval abundance estimated by sampling) were placed into each container. Total proportional settlement, including onto each container, was determined 24 hours later. Two-factor analysis of variance was used to test the null hypothesis of no larval orientation (choice of upper or lower surface of shell) or rugosity preference (choice of settlement on rough, convex surface versus smooth, concave surface of shell).

Analysis of Within-Reef Variables Shell substrate mass was used within subsamples to approximate relative substrate area. Appendix 1 discusses tests of this approximation. On shells from the reef enclosures, the proportion of each shell substrate buried in the sediments was visually estimated, to the nearest 5%. Burial in sediments left a black stain on buried portions of substrate that did not completely fade for several days. The proportion thus stained on each piece of shell substrate was removed from

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 197 total shell used in calculation of settlement and survival per unit shell substrate mass. Proportional coverage by common fouling invertebrates was visually estimated to the nearest 5% (unless under 10%; then estimated to the nearest 1%), and non-clonal individuals (except when very small and abundant) were counted. To verify the accuracy of these estimates, color photographs were taken of several subsamples, and the mass of the area within the photograph was compared to visual estimates (for one surface of each shell only). The difference between these two methods was usually less than 10% of the estimate. The abundance and percent cover of infaunal species, particularly the polychaete Polydora websteri , which bores into shell (Blake and Evans, 1972), could not be estimated. To determine proportional cover by an organism x across all pieces of substrate within a subsample, the mass of a shell substrate i (corrected for the proportion buried in sediments) was divided by the mean mass of all pieces of substrate for that subsample, and then this product was multiplied by the percent cover of organism x on shell substrate i, to obtain a weighted proportional cover. Weighted proportional coverage was used in regression analyses (below). The effects of individual substrate mass, proportional coverage by all fouling species (calculated as the inverse of cover, or proportion of space unoccupied by

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 198 macroorganisms), and proportional coverage by the visually dominant fouling organisms (the sponge Cliona celata, an unidentified cheilostome bryozoan, and the barnacle Balanus eburneus) was examined by both mulitple regression and by single factor regression (Zar, 1984). Cliona is a boring sponge, with most of its tissue mass within the shell substrate that it occupies. Its presence can be detected on the surface, however, by many close-spaced "papillae", 2-3 mm distant from each other, and about 1 mm in diameter. When undisturbed, these papillae unfold as "fans" about 5 mm high, occluding the entire area. For this reason, an area of shell substrate thus occupied was considered to be 100% covered by Cliona, for the purposes of analysis. Each shell substrate piece was treated as a replicate datum point for this analysis. The null hypothesis of no relationship between the five variables (shell mass and proportional coverage by three species and by total fouling community) and Crassostrea settlement on Day 0 was tested with multiple regression analysis (Zar, 1984). Single factor regression was used separately to examine the precent (coefficient of

determination) of the relationship accounted for by each variable, but no hypotheses were tested with the single­ factor analyses of variance.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 199 Results Sedimentation Sedimentation onto shell substrates was not significantly affected by presence of enclosures, or by whether the substrate was defaunated or natural reef shells, and there was no significant interaction effect. The ratio of accumulated sediment to substrate was 8.5%, 8.7%, 14.8%, and 7.7%, respectively, for reef plots outside and inside enclosures, and defaunated plots outside and inside enclosures. Data and results of the analysis of variance are summarized in Table 1.

Effect of Distance from Input Port Distance from the larval input port within the enclosures affected subsequent settlement density, in both the primary experiment and the secondary defaunated-only experiment. The relationship appeared to be negatively exponential; log10-transformation of the data gave the

strongest regressions. Table 2 summarizes the data and the regression analyses (using log10-transformed data) for all

treatments. In the reef enclosure treatment, the predicted distance from the central input port at which the mean number of juveniles • 100 cm'2 fell to zero was 39 cm, giving

an effective settlement area of 0.4778 m2. In the

defaunated treatment and the defaunated treatement study II, predicted settlement did not decrease to zero within the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 200 area of the enclosure (>59 cm from the center)• Examination of the enclosure edges and lids, however, revealed no settlement onto these surfaces.

Treatment Effects Mean distance-corrected Crassostrea juvenile abundance data for each treatment and sample date are summarized in Table 3. Settlement, or juvenile abundance on Day 0, did not significantly differ between the reef and defaunated substrate on Day 0, but on each sample date thereafter, juvenile abundance was significantly higher in the defaunated substrate treatment than in the reef treatment. (On Day 28, no t-test was performed, because observed juvenile abundance in the reef treatment was 0, across all enclosures). Significant differences between treatments are indicated in Table 3. The coefficients of variation in Crassostrea juvenile abundance, for total variation within each treatment and sample date, are given in Table 3. Coefficients of variation were significantly higher for the reef treatment than for the defaunated substrate treatment, for sample dates Day 0 (p < 0.001) and Day 1 (p < 0.001). On Day 3, the coefficient of variation using untransformed data was higher for the defaunated substrate treatment, because of the high number of 0's in the reef data, but the log transformation of the data used in the F-test (Zar, 1984)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 201 indicated that the variance was higher for the reef treatment (p = 0.047). After Day 3, however, most subsample values within the reef treatment were 0, effectively eliminating most variation. Table 4 presents settlement (Day 0) as a proportion of initial larvae available, and subsequent survival (Days 1- 28) as proportions of estimated settlement density (see Materials and Methods: Enclosure Sampling). Table 5 presents the within-treatment regression of proportional survival, including the slope and coefficient of determination, using log10-transformed data, for both treatments. Both regressions were significant at a = 0.05. Based on a t-test, the slopes (regression coefficients) of the two relationships differed significantly (p < 0.0005). The June 30 assay designed to examine settlement of the Crassostrea larvae in the laboratory was lost: no results are available. Survival of juveniles settled in the defaunated enclosures on June 3 0 and collected on sampling date Day 0 (July 1) are presented in Table 6. The regression analysis of the relationship between juvenile survival and time is also summarized in Table 6. The mean shell height of 232 Crassostrea juveniles from the defaunated enclosures on Day 28 was 11.2 mm, with a range of 2-20 mm and a standard deviation of 8.4 mm. From the reef enclosures on Day 43, the mean shell height of 6 Crassostrea juveniles was 23.0 mm, with a range of 13-35 mm

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 202 and a standard deviation of 8.2 mm. In the flumes, the mean shell height of 437 juveniles was 6.3 mm, with a range of 0.3 mm (no growth since settlement) to 14 mm, and a standard deviation of 5.8 mm.

Defaunated Treatment Study II The settlement and survival data for the back-up experimental run using four defaunated shell substrate enclosures are presented in Table 7. Table 8 presents the proportional survival data (based on count data corrected for subsample location, as explained in Materials and Methods), and the summary of the regression analysis examining the relationship between Crassostrea juvenile survival and sample date. The effect of distance from the central input port on juvenile density is summarized in Table 2. Proportional survival data of Crassostrea juvenile collected on Day 0 of the back-up experiment (August 4), and held in laboratory flumes, or suspended from the end of the pier, are summarized in Tables 9 and 10, respectively. Regression analyses on the relationship between Crassostrea survival and time are also summarized in Tables 9 and 10. The slopes of the regressions (regression coefficients) did not differ significantly at a = 0.05 (p = 0.092). Mortality of Crassostrea on shells suspended from the pier showed very high variability, which seemed to correlate

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 203 to the presence of newly settled, unidentified turbellarian flatworms, normally no more than 1000 /x in length. These organisms had previously been collected in low numbers as planktonic larvae (this author, unpubl. data) which readily metamorphosed on almost any surface, and had also been observed occasionally in defaunated shell substrate enclosure treatments. In four of eight replicate mesh bags suspended from the pier during the back-up trial, however, high numbers (>10 individuals per shell substrate) were observed. It was not possible to quantify them, because many had been dislodged by handling of the replicates, but these replicates were characterized by high recent mortality of Crassostrea, detectable as vacant shells. Most of

turbellarians had fed on Crassostrea, and had acquired the neutral red stain, and many were still within the juvenile shells.

Settlement Orientation In four samples collected by hand from the enclosures on August 5, 63.9% (standard deviation =5.0%) of the Crassostrea shell substrates were oriented so that the concave surface of the shell clearly faced upward. (In these four samples, no shells were on their edges, but in the enclosures, approximately 1% of all shells were oriented on their edges, so that neither surface faced up or down.) Table 11 summarizes the proportions of Crassostrea juveniles

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 204 recorded on the convex surface from the enclosures on Days 0, 1, and 6. The results of the one-sample t-tests are also summarized in Table 11 as p-values. Settling Crassostrea larvae in the laboratory strongly favored the lower surfaces of shells in the laboratory, regardless of shell orientation, in the August 4 laboratory assay. Mean proportional settlement was 64.6% (standard deviation =16.7%) in 24 hours. Laboratory settlement data and the two-factor analysis of variance on larval settlement site choice are summarized in Table 12.

Nithin-Reef Variables On Day 0, within the reef treatment, 255 individual shells or shell fragments were examined for the presence of newly settled Crassostrea virginica and common or dominant fouling organisms. Major fouling organisms observed within the reef treatment, the frequency of occurrence, and for the most abundance species, their mean areal coverage, are summarized in Table 13. The results of the multiple regression analysis on the relationship of various shell substrate variables (shell mass, total fouling communtiy coverage, and coverage by three major species) and Crassostrea juvenile abundance are summarized in Table 14. Coefficients of determination for single factor regression analyses are also given in Table 14. In the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 205 multiple regression, only shell mass was significantly related to Crassostrea abundance (p < 0.0005). Crassostrea juveniles were never observed settling directly on exposed tissue of the boring sponge, Cliona, but some individuals were observed within 1 mm of exposed Cliona tissue. Crassostrea settlement directly onto living

bryozoan colonies were rare (although they did occur), but settlement onto dead bryozoan colonies was common. Settlement also occurred occasionally onto the bases of living barnacle, Balanus, tests, and commonly onto the shells of the jingle shell, Anomia, or the calcium carbonate tubes of living tube worms, Hydroides. The most abundant errant benthic macroorganism collected with the reef treatments subsamples was the polychaete Neries succinea (Nereidae), although these and other errants were not collected quantitatively by the sampling methods used. Other macro-organisms commonly observed were gastropods, including Lunatia mitrella (Columbellidae), Ilyanassa obsoleta (Nassariidae), Eupleura caudata, and Urosalpinx cinerea (both Muricidae);

unidentified gammarid amphipods; and decapod crustaceans, including snapping shrimp Alpheus sp. (Alpheidae) and several mud crab species (Xanthidae).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 206 Discussion Effectiveness of Larval Enclosures To this author's knowledge, this study is the first to use enclosures in the field to quantify settlement or recruitment of invertebrate larvae, on this scale. In concept, however, the enclosures are not substantially different from "traps" developed by Porter and Porter (1977) and reviewed by Cahoon and Tronzo (1989), used to quantify demersal plankton emerging and migrating from the benthos. One minor difference is that there is a stronger requirement for a tight fit between the substrate and the enclosure than in emergence traps, because settling Crassostrea larvae will locate narrow gaps, or be transported by currents flowing through gaps. While designing these enclosures and experimenting with alternatives, the author found that a

fairly robust enclosure was required for this application, in this environment. Young and Chia (1982) developed a very similar enclosure, but on a much smaller scale, to examine larval settlement in the field. There was concern that the enclosures would artificially enhance sedimentation by altering currents, but from the sedimentation study, this does not seem to have been that case. In fact, the one enclosure that came loose in the reef treatment did so because of scour (sediment removal) both within and outside the enclosures, and it was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 207 necessary to bank shells around the outer perimeter of the reef enclosure (no bottoms) to prevent scour. The combination of larval enclosures and staining the larvae before injecting them into the enclosures worked well. Although there was high variability in Crassostrea juvenile counts, this was not due to counting error, but to the heterogeneity of the habitat, and possibly to gregarious settlement by larvae (Vietch and Hidu, 1971). One problem that had not been prepared for was the very high gradient in settlement across the enclosures. In future designs, this could easily be accounted for either by creating currents to disperse larvae after injection into the enclosures, to by injecting larvae at multiple locations within each enclosure. Because of the sampling design used, however, in this case the settlement gradient did not effect the between-treatment comparisons.

Treatment Effects

Settlement and survival of Crassostrea was highly variable in the reef treatment enclosures, which partially obscured differences between the reef treatment and defaunated shell treatment, in which mean settlement and survival was higher. Differences were nonetheless significant, particularly in survival. By Day 7, survival in the reef treatment was barely detectable, and by Day 28 it was not detectable, using three 100 cm2 subsamples per

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 208 enclosure, although it had not dropped to zero. By contrast, in only one of the defaunated shell enclosures (on Day 14) was no survival of Crassostrea detected.

The initial assumption when designing this experiment was that survival in the protected environment of the laboratory flume would be higher than in the field, including in the defaunated enclosures, where Crassostrea juveniles would be subject to immigrating predators and siltation. Sampling error within the field enclosures was apparently high, however (for example, apparent proportional survival on Day 7 was higher than on Day 3, within the defaunated substrate treatment), and on Day 7, apparent proportional survival was higher in the field, in the defaunated substrate enclosures, than in the laboratory flumes. The assumption that the laboratory flume was a more benign environment was further challenged by the relative growth rates, which were higher in the field than in the flume. For these reasons, the results of the laboratory flume survival study should probably not be compared directly with the field data.

Settlement Orientation Crassostrea settled preferentially onto lower surfaces of shells in both the laboratory and in the field, in the defaunated treatment study II, regardless of how the shell was positioned. In the field, 64% of the shells were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. oriented with the concave (outer surface in living animals) surface down, probably due to the hydrodynamics of oyster shells (when a single, concave valve of an oyster is dropped into water, it will almost always fall with the convex surface down; this author, pers. obs.). About 62% of Crassostrea juveniles collected from Day 0 in the defaunated treatment study II were on the concave surface; thus, by inference, the lower surfaces were favored as settlement sites by greater than 60%. This result could be due to either a negative phototactic response or a negative geotactic response, or a combination of the two, on the part of the Crassostrea larvae, but it agrees with prior findings (Ritchie and Menzel, 1969; see also Chapter 4 of this dissertation) in both the lab and the field. In an estuary with a high sediment load, such as Chesapeake Bay, one advantage of settlement onto lower surface is clear; the avoidance of siltation which could cover and smother small individuals.

Within-Reef Variables It was difficult to correlate the difference in

Crassostrea survival between the reef and defaunated shell treatments with any particular substrate characteristic apart from shell mass, a close correlate to shell surface area. Multiple regression analysis of the relationship between the various substrate variables (mass, and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 210 proportional cover by total flouing comminity ot three dominant species) and Crassostrea abundance was significant, but this was due almost entirely to the effect of shell substrate mass. Sponges have been implicated in deterrence of settlement by other sessile invertebrates by allelopathic chemicals (Thompson, 1985; Davis et a l ., 1991). Allelopathic inhibition of recruitment has been reviewed by Davis et a l . (1989). In this study, while the direction of effect by Cliona was negative, settlement of Crassostrea was not significantly related to the presence or absence of the sponge Cliona. Cliona is largely endolithic; although colonies have many tubular papillae (with ostia or oscula), the majority of the organism is hidden within, and covered by, the shell substrate. If they lack allelopaths to inhibit settlement by Crassostrea, then larval Crassostrea may find adequate settlement space between the external papillae of the sponge. It is important to bear in mind the scale; a 2-mm space between sponge papillae is equal to about six body lengths of a Crassostrea pediveliger larva. Settlement of Crassostrea was not significantly related the proportional cover by encrusting cheilostome bryozoans (for which the apparent direction of effect was negative), nor to proportional cover by barnacles, Balanus eburneus (for which the apparent direction of effect was positive). Balanus improvisus, a similar species in Chesapeake Bay, has

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 211 been implicated as a predator of Crassostrea virginica pediveliger larvae (Steinberg and Kennedy, 1979), so one might predict the opposite trend, if any. Substrate and habitat requirements, however, may be similar for Crassostrea and Balanus, which both cement permanently to the substrate when settling, so at abundance levels too low to effect Crassostrea larval abundance, Balanus may merely be an independent correlate to Crassostrea recruitment. Young (1989) found that although two species of solitary ascidians preyed heavily on various larval taxa in the laboratory, their presence did not effect settlement patterns in the field. Some other fouling species, such as the jingle shell, Anomia, and the tubeworm, Hydroides, clearly did not reduce available settlement space by the area that they covered, as demonstrated by numerous Crassostrea individuals settling onto the shells or tubes of these organisms. The feeding zone of these organisms occupied only a small portion of the area occupied by total body area. For this reason, substrate space occupied by these organisms was not analogous to substrate space occupied by Cliona, bryozoans, or Balanus. The heavy recruitment event of turbellarian flatworms onto shells suspended from the pier on August 4 provides insight into causes of differential Crassostrea survival between the reef treatment and the defaunated shell

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 212 substrate treatment. The flatworm Stylochus ellipticus has been implicated as a predator of juvenile Crassostrea to 61 mm in shell length (Webster and Medford, 1959; Landers and Rhodes, 1970), while Provenzano (1959) observed settling turbellarians, thought to be S. ellipticus, associated with heavy Crassostrea juvenile mortality, as in the present study. This or similar turbellarian species may prey heavily on very small oysters. The reef treatment would not be dependent upon migration or recruitment of predators, since these organisms would already be present, but their small size, cryptic nature, and the ease with which they become dislodged from substrate made them difficult to detect in this study, unless they had recently preyed on Crassostrea which still retained the neutral red stain. Other predators might also be present; the oyster drills, Eupleura caudata and Urosalpinx cinerea may have preyed on larger juveniles. Ilyanassa obsoleta is an opportunistic feeder, and presence of this species has been shown to negatively correlate with newly recruited infaunal invertebrate survival, possibly by grazing them along with other benthic material (Hunt et al., 1987). Mitrella lunatia, a smaller but more abundant opportunistic gastropod, may have a similar effect.

Predation at the scale of newly settled Crassostrea has been poorly studied, and it is probable that there is much "crypto-predation"- that is, predation for which the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 213 predator species is unknown, and possibly by taxa that researchers are not familiar with as predators, at least upon juvenile bivalve mollusks. This problem deserves further work if the role of a natural community on invertebrate recruitment is to be quantified.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 214

References Armonies, W. and M. Hellwig-Armonies. 1992. Passive settlement of Macoma balthica spat on tidal flats of the Wadden Sea and subsequent migration of juveniles. Neth. J. Sea. Res. 29:371-378. Baker, P. 1991. Effect of neutral red stain on settlement ability of oyster pediveligers, Crassostrea virginica. Baker, S.M. and R. Mann. 1992. Effects of hypoxia and anoxia on larval settlement, juvenile growth, and juvenile survival of the oyster, Crassostrea virginica. Biol. Bull. 182:265-269. Blake, J.A., and J.W. Evans. 1972. Polydora and related genera as borers in mollusk shells and other calcareous substrates. Veliger 15:235-249. Board, P. 1983. The settlement of post larval Mytilus edulis. Journal of Molluscan Studies 49(l):53-60. Bonar, D.B., S.L. Coon, M. Walch, R.M. Weiner, and W. Fitt. 1990. Control of oyster settlement and metamorphosis by endogenous and exogenous chemical cues. Bull. Mar. Sci. 46:484-498. Cahoon, L.B. and C.P. Tronzo. 1989. A comparison of demersal zooplankton collected at Alligator Reef, Florida, using emergence and reentry traps. Fish. Bull. 86:838-845. Cranfield, H.J. 1973. Observations on the behaviour of the pediveliger of Ostrea edulis during attachment and cementing. Mar. Biol. 22:203-209.

Crisp, D.J. 1961. Territorial behavior in barnacle settlement. J . Exp. Biol. 38:429-446. Davis, A.R. 1987. Variation in recruitment of the subtidal colonial ascidian Podoclavella cylidrica (Quoy & Gaimard): the role of substratum choice and early survival. J. Exp. Mar. Biol. Ecol. 106-57-71. Davis, A.R., A.J. Butler, and I. van Altena. 1991. Settlement behavior of ascidian larvae: preliminary evidence for inhibition by sponge allelochemicals. Mar. Ecol. Prog. Ser. 72:117-123.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 215 Davis, A.R., N.M. Targett, O.J. McConnell, and C.M. Young. 1989. Epibiosis of marine algae and benthic invertebrates: natural products chemistry and other mechanisms inhibiting settlement and overgrowth. Bioorganic Marine Chemistry 3:85-114. Dean, T.A. and L.E. Hurd. 1980. Development in an estuarine fouling community: the influence of early colonists on later arrivals. Oecologia 46:295-301. Dey, N.D. and E.T. Bolton. 1978. Tetracycline as a bivalve shell marker. (Abstract). Proc. Natl. Shellf. Assoc. 68:77. Duggins, D.O., J.E. Eckman, and A.T Sewell. 1990. Ecology of understory kelp environments. II. Effects of kelp on recruitment of benthic invertebrates. J. Exp. Mar. Biol. Ecol. 143:27-45. Fitt, W.K., S.L. Coon, M. Walch, R.M. Weiner, R.R. Colwell, and D.B. Bonar. 1990. Settlement behavior and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supernatants. Mar. Biol. 106:389- 394. Freeman, G. 1993. Metamorphosis of the brachiopod Terebratalia: evidence for a role of calcium channel function and the dissociation of shell formation from settlement. Biol. Bull. 184:15-24. Hidu, H. and J.E. Hanks. 1968. Vital staining of bivalve mollusk shells with alizarin sodium monosulfonate. Proc. Natl. Shellf. Assoc. 58:37-41. Hunt, J.H., W.G. Ambrose, Jr., and C.H. Peterson. 1987. Effect of the gastropod, Ilyanassa obsoleta (Say), and the bivalve, Mercenaria mercenaria (L.) on larval settlement and juvenile recruitment of infauna. Hurlbut, C.J. 1991. Community recruitment: settlement and juvenile survival of seven co-occurring species of sessile marine invertebrates. Mar. Biol. 109:507-515. Levin, L.A. 1990. A review of methods for labeling and tracking marine invertebrate larvae. Ophelia 32:115- 144. Loosanoff, V.L. and H.C. Davis. 1947. Staining of oyster larvae as a method for studies of their movement and distribution. Science 106(2763):597-598.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 216 Luckenbach, M.W. 1984. Settlement and early post­ settlement survival in the recruitment of Mulinia lateralis (Bivalvia). Mar. Ecol. Prog. Ser. 17:245-250. McNulty, J.K. 1953. Seasonal and vertical patterns of oyster setting off Wadmalaw Island, S.C. South Carolina Contr. Bear Bluff Lab. 15, 17 pp. Maki, J.S., D. Rittschof, A.R. Schmidt, A.G. Snyder, and R. Mitchell. 1989. Factors controlling attachment of bryozoan larvae: a comparison of bacterial films and unfilmed surfaces. Biol. Bull. 177:295-302. Manzi, J.J. and K.A. Donnelly. 1971. Staining large populations of bivalve larvae. Trans. Amer. Fish. Soc. 1(3):58-90. Martel, A. and F.-S. Chia. 1991. Drifting and dispersal of small bivalves and gastropods with direct development. Journal of Experimental Marine Biology and Ecology 150(1):131-147. Menge, B.A. 1991. Relative importance of recruitment and other causes of variation in rocky intertidal community structures. J . Exp. Mar. Biol. Ecol. 14 6:69-100. Minchinton, T.E. and R.E. Scheibling. 1991. The influence of larval supply and settlement on the population structure of barnacles. Ecology 72:1867-1879.

Morales-Alamo, R. 1993. Estimation of oyster shell surface area using regression equations derived from aluminum foil molds. J . Shellfish Res. 12:15-19. Osman, R.W., R.B. Whitlatch, and R.N. Zajac. 1989. Effects of resident species on recruitment into a community: larval settlement versus post-settlement mortality in the oyster Crassostrea virginica. Marine Ecology Progress Series 54:61-73.

Peterson, C.H. 1986. Enhancement of Mercenaria mercenaria densities in seagrass beds: is pattern fixed during settlement or altered by subsequent differential survival. Limnol. Oceanogr. 31:200-205. Porter, J.W. and K.G. Porter. 1977. Quantitative sampling of demersal plankton migrating from different coral substrates. Limnol. Oceanogr. 22:553-556.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 217 Powell, E.N., H. Cummins, R.J. Stanton, Jr., and G. Staff. 1984. Estimation of the size of molluscan settlement using the death assemblage. Estuarine, Coastal, and Shelf Science 18:367-384. Powell, E.N., R.J. Stanton, Jr., D. Davies, and A. Logan. 1986. Effect of a large larval settlement and catastrophic mortality on the ecological record of the community in the death assemblage. Estuarine, Coastal, and Shelf Science 23:513-525. Provenzano, A.J., Jr. 1959. Effects of the flatworm Stylochus ellipticus (Girard) on oyster spat in two salt water ponds in Massachusetts. Proc. Natl. Shellf. Assoc. 50:83-88. Prytherch, H.F. 19 34. The role of copper in the setting, metamorphosis, and distribution of the American oyster, Ostrea virginica. Ecol. Monogr. 4:47-107. Ritchie, T.P, and R.W. Menzel. 1969. Influence of light on larval settlement of American oysters. Proc. Natl. Shellf. Assoc. 59:116-120. Shanks, A.L. and W.G. Wright. 1987. Internal wave-mediated shoreward transport of cyprids, megalopae, and gammarids and correlated longshore differences in the settling rate of intertidal barnacles. J. Exp. Mar. Biol. Ecol. 114:1-13. Sigurdsson, J.B., C.W. Titman, and P.A. Davies. 1976. The dispersal of young post-larval bivalve mollusks by byssal threads. Nature 262(5567):386-387. Thompson, J.E. 198 5. Exudation of biologically-active metabolites in the sponge Aplysina fistularis. I. Biological evidence. Mar. Biol. 88:23-26. Trevelyan, G.A. and E.S. Chang. 1987. Light-induced pigmentation in post-larval Mytilus edulis and its use as a biological tag. Mar. Ecol. Prog. Ser. 39:137-144. van Duyl, F.C., R.P.M. Bak, and J. Sybesma. 1981. The ecology of the tropical compound ascidian Trididemnum solidum. I. Reproductive strategy and larval behavior. Vietch, F.P. and H. Hidu. 1971. Gregarious setting in the American oyster Crassostrea virginica Gmelin: I Properties of a partially purified "setting factor". Chesapeake Sci. 12:173-178.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 218 Visscher, J.P. 1928. Reactions of the cyprid larvae of barnacles at the time of attachment. Biol. Bull. 54:327-335. Watzin, M.C. 1983. The effects of meiofauna on settling macrofauna: meiofauna may structure macrofaunal communities. Oecologia 59:163-166. Webster, J.R. and R.Z. Medford. 1959. Flatworm distribution and associated oyster mortality on Chesapeake Bay. Proc. Natl. Shellf. Assoc. 50:89-95. Wisely, B. 1960. Observations on the settling behavior of the tubeworm Spirorbis borealis Daudin (Polychaeta). Aust. J. Mar. Freshwater Res. 11:55-73. Young, C.M. 1989. Larval depletion by ascidians has little effect on settlement of epifauna. Mar. Biol. 102:481- 489. Young, C.M. and F.-S. Chia. 1982. Factors controlling spatial distribution of the sea cucumber Psolus chitonoides: settling and post-settling behavior. Mar. Biol. 69:195-205. Zar, J.H., 1984. Biostatistical Analysis. 2nd Ed. Prentice- Hall, Englewood Cliffs, NJ. 718 pp.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 219 Table 1. Results and analysis of variance of sedimentation within (inside) and adjacent to (outside) experimental enclosures, with either natural reef substrate or defaunated substrate.

Table la. Summary of sedimentation data. Mass units are grams. Standard deviations shown in parentheses. ______Shell Mass Sediment Mass Shell/Sed. Out. Reef 176 (79.7) 13.3 (3.29) 0.085 (0.037) In. Reef 280 (62.2) 17.8 (2.74) 0.087 (0.034) Out. Defaun. 219 (44.0) 32.6 (7.38) 0.147 (0.016) In. Defaun. 384 (37.4) 27.6 (6.24) 0.077 (0.025)

Table lb. Summary of two-factor analysis of variance for total sediment over substrate (reef versus defaunated) and position (within or adjacent to enclosures).

Source DF SS MS F P Substrate 1 848.4 848.4 8.19 0.014 Position 1 0.1 0.1 0. 00 0.973 Interaction 1 90.9 90.9 0.88 0. 367 Error 12 1243 . 0 103 . 6 Total 15 2182.4

Table lc. Summary of two-factor analysis of variance for proportion of shell-to-sediment ratio over substrate (reef versus defaunated) and position (within or adjacent to) enclosures.

Source DF SS MS F P Substrate 1 0.002734 0.002734 1.14 0. 306 Position 1 0.004633 0.004633 1.94 0.189 Interaction 1 0.004976 0.004976 2.08 0.175 Error 12 0.028671 0.002389 Total 15 0.041014

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 220 Table 2. Summary of mean Crassostrea settlement • 100 cm'2, across distance from center, or input point, of enclosure. Location = mean distance from center, N = number of samples. Values in parentheses are standard deviations. Also shown are regression statistics for the relationship of settlement density and distance, including the regression equation and the coefficient of determination, r2.

Table 2a. Primary experiment, June 30. Reef Enclosures Defaunated Enclosures Location______N Spat’ 100 cm-2______N Spat- 100 cm'2 14 cm 6 585 (570) 7 724 (394) 36 cm 12 7.8 (8.0) 11 87 (72)

Reef Enclosures Regression The regression equation is: D = 935 - 2 6 . 2 ' X Where D = density of settled juveniles • 100 cm'2 x = mean sample distance from center in cm r2 = 0.406

Analysis of Variance SOURCE______DF______SS______MS______F______p Regression 1 1334025 1334025 10.93 0.004 Error 16 1952167 122010 Total 17 3 286192

Defaunated Enclosures Regression The regression equation is: D = 1130 - 29.0-x Where D = density of settled juveniles • 100 cm'2 x = mean sample distance from center in cm r2 = 0.603 Analysis of Variance SOURCE______DF______SS______MS______F______p Regression 1 1735365 1735365 24.31 <0.0005 Error 16 1142280 71392 Total 17 2877645

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 221 Table 2. (cont.)

Table 2b. Backup Experiment, August 4. Defaunated Enclosures Location______N Spat-100 cm'2 6 cm 4 4915 (1992) 24 cm 4 812 (384) 42 cm 4 252 (384) Back-up Defaunated Enclosures Regression The regression equation is: D = 4719 - 143'x Where D = density of settled juveniles • 100 cm'2 x = mean sample distance from center in cm r2 = 0.519 Analysis of Variance SOURCE______DF______SS______MS______F______E__ Regression 1 35743268 35743268 10.79 ( .008 Error 10 33130162 3313016 Total 11 68873432

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to to to 28 (0) (0) (0) (0) (0) 0* — »— » 0 0 0 Dav 0.5* cm'2 2.95 Dav 14 1.2* 0 0 (0) 0 (0) 4.3* 0.669* 4.12 0.3 0.3 (0.4) 0 (0) 1.0 (1.5) 0 9.61.53* 7.6 2.7 0.7 0 2.14* Dav 0 Dav 1 Dav 3 Dav 7 11 11 (16) 25 (18) 4.5 (5.3) 0 (0) 0 (0) 27 27 (9.1) 2.5 (2.5) 0.5 (0.7) 0 (0) 1.8 (2.5) 167 167 (227) 5.3 (4.3) 0 (0) 0 (0) 0 (0) 0, ____ 19.1 145 94.5 154 9.9* Larval Means of Juvenile Counts - 100 96 96 (11.3) 94 94 (23.2) Count- 103 Count- 104 104 (1.1) 1 107 (10.8) 62 (83) 0.5 (0.7) 21 (29) 2 3 54 (25.1) 427 (573) 22 (21) 45 6 112 (11.2) 232 (311) 4.5 (5.3) 0 (0) 7.3 (10) 0 (0) X s.d. c.v. tlosure larvae introduced to each enclosure, and values are in thousands. Values for each Table 3. Summarycorrected of settlementfor distance and survivalfrom the forinput reef port and defaunated (see text). treatments,enclosures are means Larval ofcounts each aresubsample, of all and standard deviations within enclosures are given parentheses.in Means and (x) standard deviations (s.d.) for each day are given at the bottom of each column.each date are indicated by Significantan asterisk. differences between treatmentsstandard The coefficient deviation, for of variation and significant (c.v.) differences for between treatments are indicated with total variation within each treatment text) (see are given an belowasterisk. the mean and Table 3a. Reef Enclosures

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in cn CO 223 • •• • CN • o CO CM in CN H O * co vo '■w' ■■— 'w' *— CO • p> H P» • > O in in o cn co o (0 • CN H • H • p' CO __ in r— . rH •• ^-N P- • o H cr» coo H OV * cn CO e 'w' V— * _ _ - - - - - H • • o rH COH CO co o o CN o CM rH • i-l • o in P' rH

r— s. s .-- .^ r— ■» 0> H CN p" P' W VOH CN VO 00 * VO -P s— " s— "■■— '— ' ------P' o CN C P- vo • 3 CM CM in o o co H 0 CO o CN m vo U CN

c CO H CTl CO P» CN a) CO CO CN CN CO H * P- > P- VO cn 3 "d* co ho CO H O CO CN *3*

w a) r— S r H * p © CM CM O CM •sr 3 0^ r H rH CO O CO vo m ' w ’Sw» w w • K''s. r ^ l£» H CO O CO 0 VO CM CO in n- CO c rH C M H r H in w

' d in 0> H •p r H O ••••• CO 3 fd H o in rH H • C > CM H CM rH CM vo 3 -P 'w' *w> 'w» , t I 3 3 0 • H rH w 0 A A 0 3 3 G EH EH W

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to to 2.48 2.22 4.16 Dav 28 28 - 0 0 3.04 0.57 0 0 0 0 Dav 14 Dav 18.6 3.99 8.31 63.8 0.96 4.93 20. 6 20. 5.47 6.43 18.5 2.53 2.67 Dav 7 14 1Dav 0 0 1.02 0 1.48 0.50 0 8.98 29.1 7.43 % % Juvenile Survival 47.5 7.08 6.13 0 14.7 25.4 3.48 % % Juvenile Survival Dav 3 0.13 0 0 6.84 0.66 0.35 22.02.33 0 3.96 1.10 0 0 2.91 0 Dav 3 Dav 7 186 12.8 80.5 55.3 9.00 240 3.14 14.2 75.0 14.9 Dav 1 16701 8. Dav 1 for primary experiment, calculated from values 226 10.2 0.75 323 49.2 E . S. D. S. E . E . S . D. . S E . Crassostrea, 1.20 210 7.14 12.9 5.92 235 34.5 Dav 0 81.5 118 475 21.9 207 108 Dav 0 26.7 298 111 % % Settlement % % Settlement 1 3 6 18.3 206 107 5 3.02 2 4 21.4 245 139 1 15.4 X 3 36.8 345 2 24.7 301 50.2 456 20.9 15.8 46.7 212 311 296 X s.d. 27.7 41.9 s.d. 11.4 41.7 0 0 (Days 1-28) of corrected from Table 3 estimated (see text). settlement density Survival from (E.S.D.); Day apparent are0 proportionssurvival for of individualthe enclosures Table 4. Summary of estimated proportional settlement 0) (Day and survival from Day and standard deviations (s.d.) are given at the bottom of each column. Enclosure may exceed 100% (see Table 4 for actual values; see text for explanation). Table Means 4a. (x) Reef Enclosures Enclosure Table 4b. Defaunated Enclosures

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 225 Table 5. Summary of regression analysis for Crassostrea survival • 100 cm'2, for the primary experiment, with Day 0 values set at 1.0, and using proportions shown in Table 5 as percentages for Days 1-28. See also text. Data is log10 transformed.

Table 5a. Survival versus time, reef enclosures. Includes 4 data points for Day 43 (see text). The regression equation is: S = -1.46 - 0.101-t where S = Individuals • 100 cm'2 (log10 transformed) t = time (days) since settlement r2 = 0.419 Analysis of Variance SOURCE______DF______SS______MS______F______p Regression 1 75.373 75.373 26.69 <0.0005 Error 37 104.501 2.824 Total 38 179.874

Table 5b. Survival versus time, defaunated enclosures. The regression equation is: S = -0.374 - 0.055-t where S = Individuals • 100 cm"2 (log10 transformed) t = time (days) since settlement r2 = 0.680 Analysis of Variance SOURCE______DF______SS______MS______F______p Regression 1 10.285 10.285 22.24 <0.0005 Error 34 15.726 0.463 Total 35 26.011

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 226 Table 6. Proportional survival over time for Crassostrea juveniles held in a laboratory flume, July 1-28.

Table 6a. Proportional survival data (percentages) of Crassostrea juveniles held in a laboratory flume. Means (X) and standard deviations (s.d.) are given at the bottom of each column.

Initial # Proportional Survival Juveniles Dav 3 Dav 7 Dav 14 Dav 28 198 52.3 11.7 16.7 9.60 282 57.8 23.0 18.1 15. 3 146 77.4 38.0 37.0 31.8 46 58.7 24.4 17.4 17.4 35 71.4 40.0 25.7 25.7 757 55. 5 40.3 22.6 16.4 414 16. 9 2.90 2.70 2 .20 8 2.50 0 0 0 60 80. 0 38.3 28.3 28.3 112 67.0 40.2 34.0 28.6 1185 23.4 6.80 3 .90 3.00 16 68.8 25. 0 18.8 18.8

X 181 55. 2 24 .7 17 .9 16.4 S.d. 309 20. 7 14.5 11.5 10. 5

Table 6b. Regression analysis of survival versus time, of Crassostrea juvenile held in a laboratory flume. The regression equation is: S = -0.289 - 0.033-t where S = Individuals • 100 cm'2 (log10 transformed) t = time (days) since settlement r2 = 0.231 Analysis of Variance SOURCE______DF______SS______MS______F______P Regression 1 6.5684 6.5684 17.44 <0.0005 Error 34 21.8410 0.3766 Total 35 28.4094

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 227 Table 7. Settlement of Crassostrea in back-up experimental run (August 4-9). Within-enclosure standard deviations are given in parentheses, and means (x) and standard deviations (s.d.) for each day are given at the bottom of the columns.

Initial # Settlement Juvenile Counts tclosure Larvae Dav 0 Dav 1 Dav 6 1 83,660 2160 (1464) 1239 (1617) 324 (320) 2 115,824 5786 (992) 1187 (951) 232 (108) 3 129,684 902 (889) 1771 (1795) 349 (202) 4 114,700 2647 (3455) 1491 (1628) 1779 (2492)

X 110,967 2873 1422 671 s.d. 16,834 1780 232 641

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 228 Table 8. Results and analysis of field settlement and survival of Crassostrea in the back-up experimental run (August 4-9).

Table 8a. Proportional (percent) settlement and survival in back-up experimental run. Calculations are based on corrected count data (not shown) for all dates, as performed for the primary experimental data (see text). Estimated settlement density (E.S.D.) is set equal to initial larval density (see text). Means (x) and standard deviations (s.d.) for each day are given at the bottom of the columns. % Juvenile Survival Enclosure % Settlement E.S.D. Dav 1 Dav 6 1 192 837 71.0 19.8 2 89.9 1158 23.7 36.7 3 35.8 1297 99.9 19.7 4 94 . 2 1147 66. 5 3 .14

X 103 1110 65.3 19.9 s.d. 56.2 168 27.2 11.9

Table 8b. Summary of regression analysis for proportional survival, using log10-transformations of proportions given in Table 9a for Day 1 and Day 6 as percentages. Values for Day 0 are set at 1.0. The regression equation is: S = -0.048 - 0.133-t where S = Individuals • 100 cm'2 t = time (days) since settlement r2 = 0.625 Analysis of Variance SOURCE______DF______SS______MS Regression 1 1.4649 1.4649 16.63 0.002 Error 10 0.8807 0.0881 Total 11 2.3456

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 229 Table 9. Results and analysis of Crassostrea juvenile survival in the laboratory flume and suspended from the pier, in the back-up experimental run (August 4-9).

Table 9a. Proportional (percent) survival over time for Crassostrea juveniles. Means (x) and standard deviations (s.d.) are given at the bottom of each column. Initial # % Survival Juveniles Dav 3 Dav 7 76 53 .9 32.9 260 91.2 79.6 172 75. 6 69.8 26 80. 8 50.0 50 76.0 74.6 206 57 . 8 81.6 174 75.3 71.3 64 75. 0 42.2 17 94.1 58.0 207 64 . 7 51.2 137 76.6 70.1 30 75. 0 36.7

X 232 74.7 59.8 s.d. 181 11. 2 16. 2

Table 9b. Summary of regression analysis for proportional survival, using proportions given in Table 11a for Day 3 and Day 7 as percentages. Values for Day 0 are set at 1.0. Data are not transformed. The regression equation is: S - 0.969 - 0.056-t where S = Individuals • 100 cm'2 t = time (days) since settlement r2 = 0.645 Analysis of Variance SOURCE______DF_____ SS MS Regression 1 0.94017 0.94017 61.72 0.000 Error 34 0.51792 0.01523 Total 35 1.45809

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 230 Table 10. Results and analysis of pier survival in the back-up experimental run (August 4-9).

Table 10a. Proportional (percent) survival over time for Crassostrea juveniles suspended off the pier. Means (x) and standard deviations (s.d.) are given at the bottom of each column. Initial # % Survival ______Juveniles Dav 3_____Dav 7 311 5.5 0.4 341 3.5 1.8 661 14.5 4.9 630 17.4 7.6 266 80. 7 75. 6 473 74.8 72.0 247 78. 8 61.4 289 79. 2 62.2 402 44.3 35.8 155 36.7 34.7

Table 10b. Summary of regression analysis for proportional survival, using proportions given in Table 12a for Day 3 and Day 7 as percentages. Values for Day 0 are set at 1.0. Data are not transformed. The regression equation is: S = 0.893 - 0.088-t where S = Individuals • 100 cm'2 t = time (days) since settlement r2 = 0.409 Analysis of Variance SOURCE______DF______SS______MS______F______p Regression 1 1.5262 1.5262 15.21 0.001 Error 22 2.2075 0.1003 Total 23 3.7338

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 231 Table 11. Proportions of Crassostrea juveniles recorded on the convex surface of shell substrates from enclosures in back-up experiment. Means (x) and standard deviations are given at the bottom, and below that are p-values from one- sample t-tests of independence.

Dav 0 Dav 1 Dav 6 66.8 53.8 56.5 62.0 35.2 56.2 50.9 72.3 54.2 68.2 57.5 67.2 62 . 0 54.7 58.5 7.9 15. 3 5.9 0.0006* 0.0056 0.0003 *Significantly independent from 50% at a = 0.05

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 232

Table 12. Settlement of Crassostrea larvae onto upper and lower surfaces of adult shells in the laboratory.

Table 12a. Summary of laboratory settlement (Total Set.) for back-up experimental run, August 4, with proportions (percentages) of settling juveniles onto lower substrate surfaces (% Lower Surface). Initial larval density was approximately 500 in each settlement chamber; proportional total settlement (% Total Set) is based on that. Orientation (Shell Orient.) of the concave (inner) surface of the adult shell substrate is indicated by UP (concave surface oriented upwards) or DOWN. Means (x) and standard deviations (s.d.) for each day are given at the bottom of the columns. Shell Total % Total % Lower Orient. Set.______Set.___ Surface UP 396 79.2 73.0 DOWN 366 73.2 90.6 DOWN 280 56. 0 78.1 UP 291 58. 2 88. 3 UP 343 68. 6 81.9 DOWN 497 99.4 84.4 DOWN 219 43.8 89.6 UP 339 67.8 67.9 UP 189 90. 6 37.8 DOWN 254 93.3 50.8 DOWN 292 58.4 71.1 UP 411 82.2 90.9 X 323 64.6 s.d. 83.2 16.7 8.42

Table 12b. Summary of two-factor analysis of variance of the effects of shell substrate Orientation (settlement onto upper or lower shell substrate surface) and Concavity (settlement onto the smooth and concave, or rough and convex, surface of each shell substrate) on proportional settlement. Proportions were transformed by the arcsine of the square root. Analysis of Variance Source DF SS MS F b Orientation 1 1.69279 1.69279 113.65 0.000 Concavity 1 0.00310 0.00310 0.21 0.660 Interaction 1 0.01636 0.01636 1. 10 0. 325 Error 8 0.11916 0.01490 Total 11 1.83141

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 233 Table 13. List of common fouling organisms found in reef enclosures, including higher level taxonomy and common name, following the species name (when available). Percent frequency of occurrence on shell substrate (for each shell is also given as well as frequency of occurrence score 0 or 1), and for abundant species, mean proportional (percent) cover, with standard deviations in parentheses. N = 255 shell substrate units.

Species % Occurrence____% Cover Cliona celata 31.0 15.8 (29.6) Porifera: Hadromerida boring sponge Ectoprocta: Cheilostomata 37.6 8.82 (18.5) encrusting bryozoan Balanus eburneus 58.0 4.88 (8.30) Crustacea: Cirripeda barnacle Anomia simplex 10.2 2.63 (9.90) Mollusca: Bivalvia jingle shell Hydroides dianthus 16.5 Sabellaria vulgaris 5.5 Polychaeta: Sabellida tubeworms Crepidula fornicata 18.0 Mollusca: Gastropoda slipper shells Cnidaria: Hydroidia 2.5 arborescent hydroid Porifera 4.3 encrusting sponges Ascidea 4.3 tunicates Alcyonidium sp. 3.1 Ectoprocta: Ctenostomata Unoccupied zones 92.9 53.5 (76.6)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 234 Table 14. Summary of multiple regression analysis on effect of various shell substrate characteristic on settlement of Crassostrea on Day 0. Juvenile abundance data are log10- transformed. s.d. = standard deviation. Multiple Regression Equation: J = - 5.0 + 0.715M + 69.7Bal - 2.52bry - 17.6C + 10.4A Where J = Crassostrea juvenile abundance M = shell substrate mass Bal = proportional cover by Balanus bry = proportional cover by bryozoan C = proportional cover by Cliona A = proportion of substrate unoccupied Coefficient of determination r,2 _ 0.095 Table 14a. Multiple regression coefficient summary. Predictor Coefficient s.d. t-ratio P Constant -4.97 12 .97 -0.38 0.702 Mass 0.7154 0.1623 4.41 <0.0005 Balanus 69.68 47.21 1.48 0.141 bryozoan -2.516 2.043 -1.23 0.219 Cliona -17.65 17.13 -1. 03 0. 304 bare 10.37 14.02 0.74 0.460 Table 14c. Analysis of variance summary. SOURCE DF SS MS F Regression 5 92930 18586 5.24 <0.0005 Error 249 883947 3550 Total 254 976877 SOURCE DF SEO SS Mass 1 64965 Balanus 1 7655 bryozoan 1 3891 Cliona 1 14479 bare 1 1940 Table 14d. Single factor regression summaries, Coefficients of determination are given as r2. Factor______Regression Equation______r2 Mass J = 2.27 + 0.660M 0.067 Balanus J = -0.57 + 0.83Bal 0.042 bryozoan J = -0.51 + 0.0033bry <0.0005 Cliona J = -0.49 - 0.063C ~ 0.001 bare J = -0.68 + 0.33A 0.072 Where J = Crassostrea juvenile abundance M = shell substrate mass Bal = proportional cover by Balanus bry = proportional cover by bryozoan C = proportional cover by Cliona A = proportion of substrate unoccupied

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 235

Figure 1. Diagram of experimental enclosure.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cn O)

•o

E o CM O

o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 236

Figure 2. Estimated proportional (percent) survival of Crassostrea juveniles, on a log10 scale, in experimental enclosures and flume control. See Tables 5, 7, for standard deviations.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o CO

LO CVJ

I— O Z CM LU

LU I 1= LU m co LU o z CO

CO - Flume Control Defaunated Treament —as— h

< - Q

-in ReefTreament

(%)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 237 Appendix 1: Relationship of shell mass to shell outline area. For subsamples taken from within enclosures, it was assumed that total shell substrate mass corresponded closely to total shell area. The great majority of shells were of a single species, Crassostrea virginica. To tests the above assumption, individual shell mass was regressed against individual shell outline area (one of several possible measures: McNulty, 1953; Morales-Alamo, 1993). Outline area was estimated by weighing cut-outs of photocopies of the shells placed concave surface down on a photocopier. The mass of the paper outlines for each shell mass were converted directly to area, by determining the ratio of mass to area for the paper used. The coefficient of variation in photocopy paper mass per unit area was 1.7% (for n = 10). Regression analysis (Zar, 1984) was used to test the assumption of no relationship between shell mass and shell outline area. In addition to Crassostrea shells, the above procedure was repeated for the next three most common shell species in the reef enclosures: Anomia simplex (jingle shell, Anomiidae), Mercenaria mercenaria (hard clam, Veneridae), and Tagelus plebius (stout razor clam, Solecurtidae). Crassostrea is heavy and irregular in shape, while Anomia is thin and irregular in shape. Mercenaria is heavy and very regular in shape, while Tagelus is thin, and very regular in shape. In addition to whole shells, however, fragments of shells were also used in this analysis. Shells were rinsed and air dried before use, but fouling organisms were left on. The results are shown in Table Al. Crassostrea, with the most irregular shell, had the least strong relationship (r2 = 0.703), even though the test used a higher number of replicates (n = 56) than the other species, but even at that, over 70% of the variance in shell mass was accounted for by shell outline. When summed across a subsample of many shells, therefore, the confidence that shell mass is related to shell area (available to settling larvae) is very strong.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 238

Table Al. Relationship of shell mass to shell outline area.

Table Ala. Crassostrea virginica N = 56 Shell Mass: x = 37.8 g s.d. = 30.7 g Outline Area: x = 31.6 cm2 s.d. = 14.9 cm2 Regression equation: M = -16.9 + 1.7 3 • A r2 = 0.703 Analysis of Variance SOURCE______DF______SS______MS______F______P Regression 1 36447 36447 127.90 <0.0005 Error 54 15388 285 Total 55 51835

Table Alb. Anomia simplex N = 19 Shell Mass: x = 1.27 g s.d. = 0.99 g Outline Area: x = 5.34 cm2 s.d. = 2.24 cm2 Regression equation: M = -0.83 + 0.39 • A Where M = shell mass A = shell outline area. r2 = 0.778 Analysis of Variance SOURCE______DF______SS______MS______F______P Regression 1 13.743 13.743 59.73 <0.0005 Error 17 3.911 0.230 Total 18 17.655

Table Ale. Mercenaria mercenaria N = 24 Shell Mass: x = 41.4 g s.d. = 27.2 g Outline Area: x = 29.2 cm2 s.d. = 15.3 cm2 Regression equation: M = -10.5 + 1.77 Area Where M = shell mass A = shell outline area. r2 = 0.948 Analysis of Variance SOURCE DF SS MS F Regression 1 16708 16708 402.01 <0 Error 22 914 42 Total 23 17622

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 239 Table Al. (cont.) Table Aid. Tagelus plebius N = 21 Shell Mass: x = 3.57 g s.d. = 1.81 g Outline Area: x = 13.66 cm2 s.d. = 4.3 5 cm2

Regression equation: M = -1.60 + 0.378 • A Where M = shell mass A = shell outline area. r2 = 0.829 Analysis of Variance SOURCE______DF______SS______MS______F £ ___ Regression 1 54.210 54.210 91.99 <0. 0005 Error 19 11.196 0.589 Total 20 65.406

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 6. SUMMARY AND CONCLUSIONS

240

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 241

Table of Contents Page Introduction...... 242 Competency to Settle...... 244 Planktonic Abundance ofLate Stage Larvae...... 245 Settlement vs Larval Abundance...... 248 Settlement and Survival on the Benthos...... 251 References...... 254

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 242

Introduction The research outlined in the preceding chapters was

designed to a) determine abundance patterns and mortality rates at several early life history phases of a bivalve mollusk at, within a specific system (the lower York River,

Chesapeake Bay, Virginia), b) to determine general sources of high variation in abundance or survival within the early

life history, and c) test techniques to quantitatively estimate population levels at the life history phases of interest. The life history phases studied were those of the competent larva and the benthic post-metamorphic phase, up to a about 1-2 mm, at which size they can readily be observed by existing benthic sampling schemes. The following general questions were used as research guidelines: 1) Of all late-stage, or pediveliger, larvae observed in the plankton, how many were actually available, or competent, to settle, within a given time period? This was

a question which evolved during preliminary research on Question 2 (below). 2) How many pediveliger larvae are present in the water column at a give site? The effects of several factors, and their influence on abundance variability, were examined.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 243 3) How closely does planktonic pediveliger larval abundance relate to subsequent settlement, and what is the residual variation?

4) From a known number of pediveliger larvae, how many settle and survive onto natural substrate, and how does a natural benthic community effect survival and variability? The choice of Crassostrea virginica as a primary research model was made because of a) availability of mass- produced larvae (from the Virginia Institute of Marine Science oyster hatchery), and b) the precision of its settlement event. Oysters of the family Ostreidae, as discussed in Chapter 1, cement themselves permanently upon selection of a settlement site, so there is no ambiguity regarding their settlement, as there is for many other marine invertebrates (see Galtsoff, 1964, for a review of Crassostrea virginica life history). The choice of Crassostrea, however, also had a serious drawback; the abundance of oysters have declined precipitously in recent years in Chesapeake Bay (Hargis and Haven, 1988), and the decline continued during this author's research period. By 1992, larvae at the research site no longer attained levels sufficient to undertake some field research, and an alternate site was required in Florida. One consequence of this was that most of the data collected fell into the low range of normal larval and juvenile abundance levels for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 244 this species. This factor will be discussed where it affected interpretation of the results.

Competency to Settle Competency to settle in larvae of the oyster Crassostrea virginica was defined as proportional settlement of a given number of larvae within 24 hours, in a laboratory assay. Coon et al. (1990) has defined and reviewed competency to settle in Crassostrea. Preliminary work with hatchery-reared larvae demonstrated that settlement of larvae which had all of the morphological characteristics associated with settlement, or "apparent" competency, often did not do so, even after 24 hours of continuous exposure to suitable substrate. As reviewed in Chapter 2, this finding was consistent with settlement rates reported for commercial hatchery systems. Concern that this might also be true of pediveliger larvae observed in the plankton lead to assays to quantify competency to settle in larvae collected from the plankton. Competency of pediveliger larvae in lower York River, Chesapeake Bay, Virginia was very close to 80% in both 1990 and 1992, and the variance of these values were very low. It is possible that the laboratory assay in some manner provided less than optimal settlement substrate, and that the actual competency was higher, although the consistency of the results argues against that. In any case, the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 245 correction value would be a minor constant. Morphological characteristics define competent larvae, at least for Crassostrea; what a researcher sees in a plankton sample is what is available to settle, at least within 24 hours. In this case, hatchery data on oyster competency was misleading. This incidentally leads to the question of why competency is so much lower among hatchery-reared larvae, a topic discussed in Chapter 2. For the overall purpose of this dissertation, however, apparent competency to settle, as estimated by morphological larval characteristics, is not a large source of variation in estimates of larval abundance and availability.

Planktonic Abundance of Late Stage Larvae. Numerous studies (reviewed in Chapter 3) have demonstrated that invertebrate larvae are not distributed evenly throughout the water column. None of those studies,

however, specifically examined the distribution of late- stage larvae, which is represented among bivalve mollusks by the pediveliger. Thus, while the prior studies gave examples of possible larval distributions, the findings could not be used for direct inference of pediveliger distribution. Primary species examined in this study included the clam Cyrtopleura costata, the oyster Crassostrea virginica, the shipworm Bankia gouldi, and the mussel Geukensia

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 246 demissa, in the lower York river, Virginia. This research is reported in Chapter 3 of this dissertation. In addition, as part of research reported in Chapter 4, C. virginica, the oyster Ostrea eguestris, and an unidentified shipworm (Teredinidae) larva, were studied in a channel attached to the Indian River lagoon, in Florida. At the York River site, pediveliger larval abundance over time, whether on a scale of days or hours, showed fluctuations of one to two orders of magnitude between samples. These fluctuations in abundance appeared random, uncorrelated with tidal phase, and for Cyrtopleura only, weakly correlated with time of day. (Temperature and salinity, while possibly significant factors in some cases, did not vary significantly during the research.) Three species, however, Cyrtopleura, Crassostrea, and Bankia, covaried on a scale of hours, which infers that there are discrete patches of larvae. This gives a researcher the option of mapping and tracking patches of high larval

abundance, and predicting larval supply to specific locations at specific times. At least two studies, in fact, have attempted a version of this technique (Shanks and Wright, 1987; Sephton and Booth, 1992). A researcher using an Eulerian (fixed sampling location) approach, however, must develop a sampling regime which accounts for patchiness, which, in turn, is at least partially a function of current velocity. At the Florida site, where current

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 247 velocities were much lower than in Virginia, the rate of variation with time was much lower. Covariance of larval species with different adult habitats infers that larval behavior strongly affects their distribution. This was further supported by the depth distribution patterns observed for all species, in a well- mixed water column. Passive sinking could account for the near-bottom larval distribution reported for most species, but not for Geukensia, which had highest larval abundance in near-surface waters. The strong depth patterns, in combination with the near-random patchiness observed over time, accounts for the lack of an affect of time of day or tidal phase; for either of these factors to take effect, the larvae would have to move out of range of the plankton samplers (up or down in the water column), which they were unlikely to do. Only Ostrea, in Florida, showed a tendency to change depth with time of day, but during the day they remained clustered near the benthos, as did other species, and a researcher which knew this could account for that redistribution in a sampling regime. Most of the variability observed in pediveliger larval abundance was fairly straightforward. The consistent effect of depth generally overrode the effects of time of day in most cases, and there was no discernable effect of tidal phase. This left variation in time, which can also be described as patchiness in a moving fluid in an estuarine

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 248 system, as the primary source of variation for a researcher to account for.

Settlement versus Larval Abundance Settlement rate to a benthic community is dependent upon larval abundance, as has been demonstrated for several systems (see Chapter 4 for review), but the relationship can be partially obscured by other factors. With the notable exception of one system (Martel et a l ., 1993), researchers have all reported high residual variability in the relationship between larval abundance and subsequent settlement. Preliminary research using existing York River data on Crassostrea revealed high variability in this system as well. Some sources of variability can be accounted for; error in larval abundance estimates by a rigorous sampling regime, the effect of fouling communities by provision of defaunated settlement substrates. What residual variation remains, therefore, is a measure of the impact of larval behavior

variability, and its interaction with small scale hydrodynamics, neither of which has been studied quantitatively under field conditions. A channel attached to the Indian River lagoon in Florida was the site of field test on the relationship between larval abundance and subsequent settlement. In this system, larval abundances were on the low edge of the normal

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 249 scale, and variability was mostly within one order of magnitude, but this did not prevent the sampling system from detecting depth distribution patterns for three species, including two oysters, Crassostrea and Ostrea. The substrates used to examine settlement were shell strings, used for many years by the Virginia Institute of Marine Science to monitor settlement of Crassostrea (Haven and Fritz, 1985). The use of the shell strings eliminated the natural fouling community as a possible source of settlement variation, and also provided a test of the reliability of the shell strings as measurements of recruitment in Virginia. For both species, the regression of daily settlement on larval abundance was positive, but it was significant only for Ostrea, and even in that case, the regression accounted for less than half of the variation. The residuals were randomly distributed, indicating that there was not a single major factor that accounted for variability. More likely, it was a combination of larval behavior variability (described qualitatively by Prytherch, 1934), and small scale hydrodynamics (reviewed by Snelgrove et al., 1993). Assuming for the moment that the above sources of variability are a constant, there are two remaining factors which affect the observed strength of the relationship. One of these is sample size; in the system studied, to guarantee detection of a significant relationship, the sample size may

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 250 have to be as high as 50 (although for Ostrea, the relationship was detected with fewer samples). The approach in this study was that of predicting settlement from larval abundance. The variation observed, however, also has implications for the reliability of observed settlement data. Planktonic abundance data were considered reliable in this study, so the variability occurred on the settlement substrates, and suggests that replicate substrates would give different values, within the existing variability. For most Chesapeake Bay estuaries, the number of shell strings deployed by the Virginia Institute of Marine Science across the entire sampling season ranges from 209 for the James River, to 19 for some smaller estuaries, but no more than 19 for any one sampling location (Barber, 1992). The variability in Indian River data suggests that, at best, these data can be used to make predictions of recruitment across the entire estuary. An additional source of variability in the Virginia shell string monitoring program was mortality of early settlers during the seven days that each shell string is deployed; use of many similar data sets for a single site in the York River failed to produce a significant relationship between survival, and the time that the substrates were in the water column.

The other factor that affects the strength of the relationship between larval abundance and settlement is the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 251 order of magnitude in variation observed. In a system with nearly four orders of magnitude in larval abundance, settlement was very closely related to larval abundance (Martel et al., 1993), as one might expect. If there are no larvae, there can be no settlement, and if there is heavy settlement, there had to have been many larvae. The implication for the Virginia shell string monitoring program, and similar systems, is that inter-annual predictions have some merit when the difference between years is large; if there were 200 times as many recruits in 1993, for example, as 1992, it is fairly safe to say that recruitment was much higher in 1993 than in 1992, but if there were only 20 times more recruits in 1992 than in 1991, then confidence that recruitment was higher in 1992 (with the existing sampling system) is low.

Settlement and Survival on the Benthos The qualitative impact of specific benthic organisms on settling larva has been the source of many studies (reviewed by Scheltema, 1974), although quantitative studies have not universally shown an impact (Young, 1989). Other studies have examined the role of benthic communities on survival following settlement (see Chapter 5 for references), but to date, not the effect on both settlement and recruitment. The technical difficulties of a study to quantitatively examine the effects of the benthos on both settlement and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 252 survival (discussed in Chapter 5) were overcome by the use of larval settlement enclosures. Two treatments were used: a natural Crassostrea shell reef community, and a similar, but defaunated substrate consisting only of shells. Interestingly, despite many reports in the literature of the impacts of barnacles and other fouling organisms on settlement, settlement onto the reef treatment was not significantly lower than onto the defaunated treatment. A further examination of the roles of the most abundant fouling organisms (a barnacle, a bryozoan, and a sponge) showed no relationships, aside from a very weak positive relationship with barnacle coverage (the two species may share similar microhabitat requirements). Likewise, total coverage by all species was unrelated to Crassostrea settlement, onto individual shell substrates, despite of a very large sample size (n = 2 55). Variation in settlement, however, was significantly higher within the natural reef community, than within the defaunated substrates. Some factor or set of factors, apparently, was at least affecting settlement site choice, if not total settlement. Following settlement, mortality in the reef treatments was very high, so that by seven days following settlement, survival was barely detectable. Variability in abundance remained higher than in the defaunated treatments through Day 3, after which variability

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 253 in reef treatments was effectively eliminated by the lack of survivors. Based on qualitative observations, the mostly likely culprit for differential survival are small errant predators, especially (in this system) turbellarian flatworms. In some cases, nearly 100% mortality was attributed to these very small, very abundant predators. Previous researchers may have overlooked these predators in some cases; had not the flatworms taken up the stain that the Crassostrea were marked with, they might not have been noted by this researcher. The implication of these predators for researchers is that the sample interval for settlement needs to be brief, and there may be error even at 24 hours (predation rate is rapid). This is the first study known to this author which has deployed larval enclosures of this size; Young and Chia (1982) developed a similar but much smaller larval settlement enclosure. The limitations of this system are readily correctable, and it provides clear data.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 254 References Barber, B.J. 1992. Oyster spatfall in Virginia waters: 1992 annual summary. Virginia Marine Resource Special Report, December 1992. 12 pp. Coon, S.L., W.K. Fitt, and D.B. Bonar. 1990. Competence and delay of metamorphosis in the Pacific oyster Crassostrea gigas. Mar. Biol. 106:37 9-387. Galtsoff, P.S. 1964. The American oyster. Fishery Bulletin 64: 1-480. Hargis, W.J., Jr., and D.S. Haven. 1988. The imperiled oyster industry of Virginia. Virginia Institute of Marine Science Spec. Rep. Applied Mar. Sci. Ocean Engin. 290. 13 0 pp. Haven D. and L.W. Fritz. 1985. Setting of the American oyster Crassostrea virginica in the James River, Virginia, USA: temporal and spatial distribution. Marine Biology 86:271-282. Martel, A., A. Mathieu, S. Findlay, S. Nepszy, and J. Leach. 1993. Daily settlement rates in zebra mussels correlate with abundance of veligers. (Abstract). Proc. Larval Ecology Meetings, Port Jefferson, NY: 29. Prytherch, H.F. 1934. The role of copper in the setting, metamorphosis, and distribution of the American oyster, Ostrea virginica. Ecol. Monogr. 4:47-107. Scheltema, R.S. 1974. Biological interactions determining larval settlement of marine invertebrates. Thalassia Jugoslavia 10:263-296. Sephton, T.W. and D.A. Booth. 1992. Physical oceanographic and biological data from the study of the flushing of oyster (Crassostrea virginica) larvae from Caraquet Bay, New Brunswick. Can. Manu. Rep. Fish. Aquatic Sci. 2162. 61 pp. Shanks, A.L. and W.G. Wright. 1987. Internal wave-mediated shoreward transport of cyprids, megalopae, and gammarids and correlated longshore differences in the settling rate of intertidal barnacles. J. Exp. Mar. Biol. Ecol. 114:1-13.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 255 Snelgrove, P.V.R., C.A. Butman, and J.P. Grassle. 1993. Hydrodynamic enhancement of larval settlement in the bivalve Mulinia lateralis (Say) and the polychaete Capitella sp. I. in microdepositional environments. J. Exp. Mar. Biol. Ecol. 186:71-109. Young, C.M. 1989. Larval depletion by ascidians has little effect on settlement of epifauna. Mar. Biol. 102:481- 489. Young, C.M. and F.-S. Chia. 1982. Factors controlling spatial distribution of the sea cucumber Psolus chitonoides: settling and post-settling behavior. Mar. Biol. 69:195-205.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A.

REVIEW OF ENVIRONMENTAL CUES FOR MARINE INVERTEBRATE LARVAL SETTLEMENT AND METAMORPHOSIS

256

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 257

Table of Contents Page No. Introduction...... 258 Taxonomic Review...... 260 Porifera...... 260 Cnidaria...... 262 Ectoprocta...... 266 Brachiopoda: Articulata...... 268 Echinodermata: Echinoida...... 2 69 Echinodermata: Holothuroidea...... 271 Crustacea: Decapoda...... 272 Crustacea: Cirripeda...... 273 Ur ochor data...... 277 Annelida: Polychaeta...... 279 Echiura...... 281 Mollusca: Polyplacophora...... 281 Mollusca: Gastropoda: Opisthobranchia...... 282 Mollusca: Gastropoda: Prosobranchia...... 285 Mollusca: Bivalvia...... 287 Analysis of Cues...... 294 Abiotic Cues I: Gravity, Pressure, and Light...... 294 Abiotic Cues II: Rugosity, Hydrogosity, and Currents...... 298 The Problem of Inorganic ions...... 3 01 Chemical Cues I: Gregariousness and Conspecific Avoidance...... 303 Chemical Cues II: Prey, Habitat, and Bacterial Cues...... 306 Chemical Cues III: Allelopathy and Avoidance...... 310 The Roles of Larval Age and Behavior Variability..... 311 References...... 313

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 258

Introduction The majority of benthic marine invertebrates have a planktonic larval or postlarval stage with some swimming ability, which metamorphoses or develops into a benthic juvenile. Larval availability to the benthos can sometimes be accounted for by large scale (Shanks and Wright, 1987) or small scale (Butman, 1989) hydrodynamic processes, or by direct or indirect interactions with biotic communities (Gaines and Roughgarden, 1987; Duggins et a l ., 1990). Many benthic planktivores consume approaching larvae (e.g. Breese and Phibbs, 1972; Steinberg and Kennedy, 1979; Young and Cameron, 1989; Buyanovski and Solohina, 1991), and may thus affect settlement patterns (Young and Cameron, 1989) . In some cases, benthic settlement can be limited simply by the availability of suitable substrate (e.g. Olafsson, 1988). An increasing body of literature is accumulating on the role of larval behavior in regulating settlement. Studies on oyster larvae (e.g. Prytherch, 1934; Cranfield, 1974), barnacle larvae (e.g. Crisp, 1961), and polychaete annelid larvae (e.g. Wisely, 1960) demonstrated that larvae can have relatively complex and stereotypical behavior while searching for settlement sites. Larval settlement may be

divided into a) a swimming exploratory stage; b) a crawling

exploratory stage; c) a settlement stage; and d) metamorphosis, which is not behavioral but in some cases

seems to be directly linked to settlement behavior. (For

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 259 some active benthic invertebrates, such as decapod crustaceans, the crawling exploratory stage may be absent.) The swimming exploratory stage seems to be related strictly to developmental stage in the few species for which this has been examined, but the other steps in the settlement process may themselves be triggered by environmental cues. Examples of these will be discussed later. Larvae which have reached the developmental level which permits settlement and metamorphosis are termed "competent". Environmental cues can enhance or inhibit some of the above larval settlement stages, and can thus be divided into positive and negative settlement cues. Most examples of environmental cues are chemicals of biotic origin, but some well-defined abiotic cues also exist, including light and rugosity. For some environmental factors, however, it is difficult to determine whether the factor is a behavioral cue or effects settlement by some other means. Examples of these factors include water currents and water temperature, and will be discussed later in this review. The following review is divided into two major sections. The first of these sections examines known cues by major taxonomic groups, and within that, specific types of cues. The second section uses the same material, to identify major types of cues and behavioral mechanisms used by invertebrate larvae.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 260 The topic of settlement and settlement cues among marine invertebrates has recently been reviewed separately, with a somewhat different organization and approach, by Rodriguez et al. (1993). Pawlik (1990) has previously reviewed chemical settlement inducers.

Taxonomic Review Porifera Most marine sponges have a lecithotrophic parenchymella (also spelled parenchymula) larva, which is ovoid in shape and covered with flagellated cells, and has a brief (up to 56 hours) planktonic period (Fry, 1971; Barnes, 1980; Kaye and Reiswig, 1991). Some marine sponges produce chemical compounds which are though to inhibit settlement of larvae of space- competitors or predators (e.g. Bingham and Young, 1991; Thompson, 1985). This is known as allelopathy, and is an example of a negative settlement cue. Recruitment (two weeks of settlement and subsequent mortality) of an unidentified sponge was lower near two species of Halichondria (an encrusting sponge), relative to controls and other sponge species, and was interpreted to be the result of allelopathic effects on larval settlement (Bingham and Young, 1991). Since this study did not examine daily settlement, however, differential post-settlement mortality of sponges cannot be ruled out.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 261 Bergquist and Sinclair (1968) reported that Ophlitaspongia would not settle on glass until the glasshad been fouled by microorganisms, but this behavior was not observed for four other genera, or in two genera examined by Kaye and Reiswig (1991). Neither set of researchers identified the fouling microorganisms. Kaye and Reiswig (1991) examined some aspects of sponge parenchymella larval behavior in three species of Spongia and one species of Hippospongia. They observed "crawling" and apparent exploration interspersed with swimming prior to settlement in all species. Although larvae were negatively phototactic throughout most of their swimming phase, this behavior disappeared several hours prior to settlement. This phenomenon was also observed by Fry (1971) for Ophlitaspongia. Haliclona larvae, however, are positively phototactic throughout their mobile period, and in the laboratory settled towards the light source. No reaction to substrate orientation (angle) has been identified (Fry, 1971; Kaye and Reiswig, 1991).

Cnidaria Many marine cnidarians produce a planula larva, which is ovoid and ciliated. The larva may be either lecithotrophic or planktotrophic, and the planktonic period may last only hours or many days (Barnes, 1980).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 262 Brewer (1976) examined the general behavior of and a number of possible physical cues for larvae of the scyphozoan Cyanea capillata. He concluded that settling larvae are positively geotactic, because they dropped to near the bottom of the experimental dishes shortly before settling, but in a separate set of experiments he found that settlement occurred more heavily on the underside of coverslips, a behavior which requires at least some negative geotaxis. It is possible that the larvae decreased their swimming rate and merely sank passively prior to settling. Light had little or no effect on settlement in the observations of Brewer (1976) of the scyphozoan Cyanea (1976), but etched coverslips had greater settlement that smooth coverslips, indicating rugophilic behavior. Chemical alteration of the glass surface during etching might have affected larval response, but silanization alone, by a variety of chemicals, did not affect proportional settlement of two hydroids (Obelia sp. and Tubularia sp.) (Roberts et a l ., 1991). Lewis (1974) found planulae of the coral Favia fragum to be negatively phototactic during settlement, but detected no rugotactic preference. One of two coral (Agaricia) species examined by Morse et al. (1988) was positively phototactic when settling. Planulae of the sea pen Ptilosarcus gurneyi, a large sand-dwelling anthozoan, delay metamorphosis when deprived of sand grains, and by use

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 263 of antibiotics the authors were able to rule out a microbial role (Chia and Crawford, 1973). Virtually all other work on cnidarian larval settlement has examined chemical settlement cues. Copper ions and quinone can induce settlement in Tubularia larynx under some conditions (Pyefield and Downing, 1949). Williams (1976) examined the cause of aggregative recruitment on several species of hydrozoans, and found that the patterns were formed during settlement, or immediately following settlement, but prior to metamorphosis. Two species of Nemertesia sometimes settled in an apparently random spatial distribution, but the planulae then crawled into dense aggregations. A similar behavior was reported for the coral Pocillopora damicornis (Harrigan, 1972). Kirchenpaueria, a hydroid, settled much more onto isolated mucous of adult conspecifics than onto a bacterial film (Williams, 1976). The hydroid Bougainvillia principis is also reported to settle gregariously, although no description of the process or quantitative details are given (Mills and Strathmann, 1987) . An unidentified water-soluble compound present in seawater from adult habitat areas was required for settlement and metamorphosis of a benthic scyphozoan, Cassiopea andromeda (Fitt et al., 1987), although various non-natural cyclic adenosine triphosphate (cAMP) enhancers also triggered settlement. These behaviors fall under the general heading of "gregariousness".

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 264 Microbial films on firm substrates can affect larval settlement. Harrigan (1972) reported that Pocillopora preferred a mixed film of green algae, diatoms, and bacteria, and in lesser numbers on diatoms or bacteria alone. Clean surfaces were avoided. A Tubularia species (a

hydroid) settled more onto unidentified microbial films than onto clean glass (Pyefinch and Downing, 1949). Neumann (1979) demonstrated that water-soluble products produced by growing cultures of the ubiquitous marine bacterium Vibrio induced settlement and metamorphosis of the scyphozoan Cassiopea andromeda, and furthermore, that antibiotics in the larval culture inhibited settlement and metamorphosis indefinitely. In this case, it is inferred that bacterial supernatants are critical for recruitment. At the opposite extreme, Lewis (1974) reported that larvae of the coral Favia preferred unfilmed surfaces to bacterial-filmed surfaces when settling. Algae are preferred settlement substrates for a variety of cnidarians. Sertularella miurensis and Coryne uchidai, two hydroids, settle preferentially on two species of the brown alga Sargassum (Nishihira, 1967, 1968). For Coryne, it was further shown that water-soluble extracts of Sargassum induced settlement and metamorphosis (Nishihira, 1968). Extracts of a related alga, however, induced settlement but not metamorphosis. This was one of the first

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 265 pieces of evidence that settlement and metamorphosis are regulated separately in invertebrate larvae. Coralline red algae induce settlement in a number of corals (Harrigan, 1972; Sebens, 1983; Morse et al., 1988). Sebens (1983) found that settlement of the soft coral Alcyonium siderium is induced by three species of encrusting coralline algae, and Morse et al. (1988) reported the same for three species of the coral Agaricia; in this case only certain species of coralline algae induced settlement. In both cases, the chemical inducer involved seemed to be non­ water-soluble, and required the larvae to be touching or nearly touching the algae to induce settlement. For Agaricia, the chemical involved is thought to be a sulphated polysaccharide found in the algal cell wall (Morse, 1991). Gamma-aminobutyric acid (GABA), which has been shown to induce settlement in other invertebrates (discussed elsewhere in this review) did not effect settlement in either of the studies. The hydroid Proboscidactyla flavicirrata is symbiotic on the tube margins of large sabellid polychaete. The

planulae settle first onto the cirri of the polychaetes, and then transfer to the tube margins. Donaldson (1974) concluded that there were chemical cues involved in both steps. Interestingly, cirri of a sympatric, non-host sabellid induced settlement onto the cirri, but not transfer to the tube margin. Planulae of the anthozoan Peachia

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 266 quinquecapitata are parasitic through metamorphosis in several species of hydromedusae, and do not metamorphose until ingested by their host (Spaulding, 1972). The presence of the colonial ascidian Aplidium inhibited settlement of the soft coral Alcyonium, even when suitable substrate was present (Sebens, 1983). Continued presence of the ascidian eventually caused death of the soft coral larvae, however, so it is not certain whether the inhibition is behavioral modification or simply a toxic effect.

Ectoprocta Marine bryozoans produce two well-known larval types. The cyphonautes larva of Membraniporidae and Electriidae is broadly conical in form, planktotrophic, and long-lived. Most other marine bryozoans produce a coronate larva, which is cylindrical, lecithotrophic, and usually lasts only a few hours (Reed, 1987b). Virtually all bryozoans settle onto firm substrates and grow clonally after settlement. Frequently researchers have assumed that geotactic behavior accounts for a specific larval or postlarval distribution, but have not disproved the potential role of light or water pressure. Pires and Woolacott (198 3) controlled for all three factors for coronate larvae of two species of Bugula and observed negative geotaxis (swimming upwards) in the absence of light, despite the fact that the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 267 larvae themselves are negatively buoyant. In addition, settlement and metamorphosis was highest near the top of the experimental chambers. The authors were able to rule out the role of barotaxis (response to water pressure). Roberts et al. (1991) reported enhanced settlement by Bugula neritina by silanization of glass, without affecting surface texture, by various chemicals. Lynch (1961) reviews various ions that induce settlement of Bugula species and some other bryozoans. Copper, calcium, potassium, sodium, quinone, peroxide, iodine, and urea (ammonia) all induce settlement of Bugula under some conditions. Microbial films (presumably bacterial) enhances settlement of some bryozoans but not others (Crisp and Ryland, 1960; Ryland, 1976; Brancato and Woollacott, 1982). Maki et a l . (1989) examined this in more detail for Bugula neritina, and concluded that enhancement, when it occurred, was not due to chemicals produced by the bacteria, but by the presence of the bacteria increasing the "wetability" (hydrophilic effect) of the substrate surface. Cyphonautes larvae of Membranipora membranacea can be induced to settle by the placing pieces of an alga (species not given) with them, but it is not clear from the information available what the induction mechanisms was (Strickler, 1988). High levels of potassium ions in the water also induced settlement and metamorphosis in this

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 268 species. Coronate larvae of Alcyonidium polyoum settled preferentially on two algal species (Fucus and Chondrus) in the laboratory, and in this case, settled highest in pits on the algal thalli. This suggests rugophilic behavior, but does not rule out chemotaxis. One species of marine bacteria, Deleya marina, inhibited settlement of Bugula neritina onto otherwise acceptable substrates in the laboratory (Maki et al., 1988). B. neritina settlement was also inhibited by insoluble compounds isolated from an ascidian (Eudistom olivaceum), two of which were toxic as well (Davis and Wright, 1990). Thompson (1985) found evidence that allelopathic chemicals of a sponge (Aplysina fistularis) reduced settlement of the bryozoan Philodophora pacifica, but because prolonged exposure cause death of the larvae, it is not known whether the reduced settlement was due to behavioral modification or simply toxicity.

Brachiopoda: Articulata

The larvae of articulate brachipods are not widely studied and have no common name. Those species studied reveal a larval period of only a few days (Barnes, 1980; Reed, 1987a). Wisely (1969) studied recruitment of Waltonia inconspicua onto shells of grooved and smooth shells, and found highest recruitment in the field within grooves. From this he concluded that the settling larvae of this species

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 269 was rugophilic. Unreplicated observations by Percival (1956) also showed apparent rugophilism for larvae of this species in the laboratory. Phototactic behavior in Waltonia was not addressed by Wisely (1969), however, contrary to the statement by Reed (1987a).

Echinodermata: Echinoida Sea urchins and sand dollars have a bilateral, eight- lobed planktotrophic larva known as an echinopluteus, which remains in the plankton for weeks to months, and settles without attachment (Barnes, 1980; Strathmann, 1987c). Burke (1983) has described some components of neurological control of settlement and metamorphosis in the sand dollar Dendraster excentricus, and also gives detailed descriptions of the echinopluteus. Sand dollars as adults live in large aggregations, and there is strong evidence that larvae settle into beds of adults (gregariousness). Highsmith (1982) showed that Dendraster excentricus larvae are induced to settle in sand that has been exposed to adult. Burke (1984) later

identified the chemical responsible as a small (980 Daltons) peptide. Pearce and Scheibling (1990) showed the same settlement induction for Echinarachnius parma, and furthermore demonstrated that the chemical came from the adult, not adult-associated bacteria. They found the chemical responsible to be heat labile (e.g. peptide) and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 270 <1000 Daltons in size. In contrast, settling larvae of the urchin Strongylocentrotus droebachiensis do not settle proportionately more in the presence of adult or juvenile- associated chemicals (Pearce and Scheibling, 1991). Cameron and Hinegardner (1974) found that two urchins, Lytechinus pictus and Arbacia punctata, settled and metamorphosed onto surfaces that were filmed with bacteria, or simply in water that had been held in a container with bacteria-filmed surfaces, indicating that a chemical cue was involved. Pearce and Scheibling (1988), in contrast, were able to induce settlement of the urchin Strongylocentrotus droebachiensis by exposure to any of three species of coralline algae, and by aqueous extracts of the coralline alga Lithothamnion, and by the use of antibiotics were able to demonstrate that bacteria were not involved. Later, however, Pearce and Scheibling (1991), demonstrated that microbial films grown in then light, and therefore including microalgae, did induce settlement of S. droebachiensis. Pearce and Scheibling (1988) were also able to induce settlement of S. droebachiensis with gamma-aminobutyric acid (GABA), a neurotransmitter similar in activity to some molecules which have been isolated from coralline algae. Kitamura et al. (1992) were able to induce settlement of larvae of the sea urchins Pseudocentrotus depressus and Anthocidaris crassispina by simple lipids extracted from a coralline alga.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 271 Echinodermata: Holothuroidea Most sea cucumber larvae have planktotrophic auricularia larvae, followed by planktotrophic doliolaria larvae; however, some have a lecithotrophic vitellaria larva, followed by a competent-to-settle pentacula larva (Barnes, 1980). The cryptic sedentary Psolus chitinoides is an example of the latter. Young and Chia (1982) found that both vitellaria and pentacula larvae of these species were gregarious, and settlement was gregarious both with respect to other pentacula, and to adult conspecifics. Pentacula larvae also were negatively phototactic when settling.

Echinodermata: Asteroidea Sea stars with planktonic development have a bipinnaria larva, followed by a brachiolaria larva, which then settles and metamorphoses to a juvenile (Barnes, 1980). Johnson et a l . (1991b) studied settlement of the crown-of-thorns sea star, Acanthaster planci, onto the preferred juvenile habitat, coralline algae. High variability in settlement induction seemed to be related to the presence of epifaunal bacteria on the coralline algae; when antibiotics were used, settlement onto the algae was reduced. The compound responsible appeared to be water soluble and greater than 10,000 Daltons in molecular size.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 272 Crustacea: Decapoda Marine benthic decapod larvae are brooded to a free- swimming stage, and then develop planktotrophically to a postlarval mysis (shrimp and lobsters) or megalopa (crabs and anomurans), which is non-feeding and eventually settles to a more-or-less benthic stage (Barnes, 1980). Settlement has not been studied behaviorally in most decapods, possibly because it is seen as having little meaning in an animal that remains mobile after settlement. An increase in swimming activity in response to increased barometric pressure has been reported for megalopae of the crabs Cancer magister (Jacoby, 1982) and Callinectes sapidus (van Heukelem and Sulkin, 1983). This is probably a negative barotropic response. Callinectes megalopae are also negatively geotactic (van Heukelem and Sulkin, 1983), while Cancer megalopae are negatively geotactic at night and positively phototactic during the day (Jacoby, 1982). Mair (1980) reported that postlarvae of decapod shrimp species of Penaeus, which develop in the Gulf of Mexico but live as adults in estuaries, were attracted to relatively low-salinity water. Flume experiments demonstrated that postlarvae of two of the same species of Penaeus swam upstream in estuarine water compared to artificial seawater of the same salinity and temperature (Benfield and Aldrich, 1992). Benfield and Aldrich (1992) were unable to determine

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 273 the biological or geochemical source of the attraction, however. Blue crab (Callinectes sapidus) megalopae settle into sea grass beds. Experiments indicate a possible chemical

attraction in megalopae to sea grass, and more clearly showed avoidance behavior towards odors of adult Callinectes (Layman, 1992).

Crustacea: Cirripeda Barnacle have a relatively long, planktotrophic stage, and develop into an non-feeding, elongate, bivalved cypris larva (Barnes, 1980). The crawling behavior and metamorphosis of many species has been well-described (see Strathmann, 1987b, for review). Most behavioral work has been done with balanomorph barnacles (acorn barnacles), which cement themselves permanently to firm substrate upon settlement.

Taki et al. (1980) gave cyprids of Balanus amphitrite different-colored substrates of four different levels of reflectance ("luminosity"). The cyprids preferred substrates with the lowest reflectance, indicating a level of negative phototaxis. They also reported heavier settlement on red than green, indicating to them that the cyprids were most sensitive to green. Visscher (1928) had previously concluded that settling cyprids of Balanus

amphitrite, B. improvisus, and Chthamalus fragilis were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 274 negatively phototactic, and Crisp and Ritz (1973) had shown the same for Semibalanus (Balanus) balanoides. Crisp and Barnes (1954) demonstrated that cyprids of Semibalanus balanoides, B. crenatus, and Elminius modestus settle preferentially in pits and shallow grooves on substrates (rugophilic behavior). Wethey (1986) showed the same behavior for S. balanoides and Chthamalus fragilis, and used flume studies to show that the larvae selected microhabitats with lower shear. Crisp and Barnes (1954) also produced evidence that these cyprids are positively phototactic when settling, in contrast to the conclusions of Visscher (1928), later work by Crisp and Ritz (1973), and Taki et al. (1980) (above). Forbes et al. (1971) concluded that S. balanoides orient themselves with respect to light angle when settling, but also that the response to substrate rugosity was stronger than the response to light. Miller and Carefoot (1989) found that Balanus cyprids favor small pits for settlement sites, but are even more strongly attracted to the base of other barnacles, which provides even greater protection from disturbance. Small scale rugophilia may also be an adaptation to increase cement adhesion, as suggested and reviewed by Le Tourneaux and Bourget (1988). S. balanoides cyprids chose bare patches for settlement (and were observed to deliberately brush away detritus during exploration), and were positively rugotactic down to a scale of 3 00 ju (Le Tourneaux and Bourget, 1988) .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 275 Chemical alteration of glass plates by silanization, without affecting rugosity, had a negative impact of settlement by an unidentified barnacle (Roberts et al ., 1991). Knight-Jones and Morgan (1966) found that cyprids of Semibalanus balanoides reduced swimming when the barometric pressure dropped several millibars, and increased swimming with increasing pressure; in other words, a negative barotaxis. Knight-Jones and Stevenson (1950), Crisp (1961), Barnett and Crisp (1979), among others, have examined gregariousness and territoriality in Elminius modestus. Knight-Jones and Stevenson (1950) provided evidence for gregarious settlement caused by waterborne chemical cues, but also observed that the exploring cyprids tended to space themselves with regard to previous recruit when exploring the substrate, just prior to cementation. Crisp (1961) elaborated on this, and suggested that the post-settlement spacing for this species and for two other barnacles (Balanus and Semibalanus) was most likely due tactile cues, rather than chemical cues, although he did not provide a model for how the cyprids distinguished new recruits from surface irregularities by touch alone. Barnett and Crisp (1979) showed that gregariousness was biased towards conspecifics. Larman and Gabbott (1975) showed that boiled extracts of two barnacles and two bivalve mollusks induced settlement in cyprids of the two barnacles (S. balanoides

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 276 and E. modestus), suggesting that the compound involved was not a heat-labile peptide. Further work has partially isolated a peptide termed arthropodin, which seems to vary slightly between species (see Lewis, 1978, for review). Le Tourneaux and Bourget (1988) studied settlement patterns in the intertidal of S. balanoides in the absence

of adult chemical cues, and found that distribution patterns of newly settled juveniles were the same as those of adults in adjacent sites. Raimondi (1988) showed the same for Chthamalus anisopoma, although extract from the crushed adults also induced settlement in the field. Raimondi (1988) also demonstrated that extracts from a supra-littoral cyanobacteria (Callothrix crustacea) inhibited settlement, the recent presence of two sympatric herbivorous gastropods (Nerita funiculata and Tegula marinaria) also induced settlement, but not the recent presence of a similar allopatric gastropod (Tegula rugosa). Two species of intertidal barnacle (Tessopora rosae and Tetraclita pupurascens) that as adults live just above a zone inhabited by a polychaete tubeworm, Galeolaria caespitosa, settled onto plates more heavily in areas cleared of Galeolaria, than in areas with the polychaete present (Denley and Underwood, 1979). The results of these studies suggest that gregariousness or other responses to conspecific chemical cues in barnacles is critical primarily at a small scale (the scale of the barnacles themselves), and that other

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 277 factors, such as the presence of sympatric species, control broader scale patterns of distribution. Maki et al . (1990) found that three species of bacteria, Deleya marina, Vibrio vulnificus, and an unidentified species inhibited settlement of Balanus amphitrite onto polystyrene dishes, and that the inhibition increases with age of the culture. A mix of estuarine bacteria, however, enhances settlement. The authors concluded that water-soluble bacterial exopolymers were responsible in all cases. Rittschof et al . (1986) showed that settlement of Balanus amphitrite was inhibited by excess levels of several common seawater ions, including potassium, calcium, sodium, and magnesium.

Urochordata: Ascidacea Both solitary and colonial ascidians produce large tadpole larvae, which are short-lived and non-feeding (Barnes, 1980). Development and metamorphosis is described in detail by Cloney (1987). Tadpoles of the colonial ascidian Trididemnum solidum are negatively phototactic when settling (van Duyl et a l ., 1981). Crisp and Ghobashy (1971) reported the same for the colonial ascidian Diplosoma listerianum, and also concluded that the larvae were positively geotactic, but were unable to detect a response to changes in barometric pressure.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 278 Lynch (1961) reviews work on chemical inducers of tadpole settlement. A variety of ions, including copper, iron, zinc, calcium, sodium, iodine, and urea (ammonia), are reported by various authors to induce settlement in various ascidians. Models or mechanisms of induction are not given. Forms of gregariousness have been reported several times. Grave and Nicoll (1939) found that settlement in the solitary ascidians Ascidia nigra and Polyandrocarpa species was enhanced by crowding the larvae. Young and Braithwaite (1980), Van Duyl et al. (1981), and Grosberg and Quinn (1986) reported gregariousness in settlement with respect to recently-settled juveniles in the colonial ascidians Chelysoma productum, Trididemnum solidum, and Botryllus schlosseri, respectively. Davis (1987) and Davis et al. (1991) have investigated allelopathy by several species of sponges towards two colonial ascidians of the Podoclavella genus. Tadpole

larvae usually rejected sponge surfaces, unless they were heavily fouled, when settling. Organic-soluble compounds that inhibited settlement were isolated from two sponges, and shown to be non-toxic, indicating that the decreased settlement was due to behavioral modification, not toxicity. The solitary ascidian Ascidia nigra, however, recruited in greater numbers onto a sponge of the Terpios genus than onto

artificial sponge controls, indicating some sort of chemical attraction (Bingham and Young, 1991).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 279 Annelida: Polychaeta Most marine polychaetes larvae develop into a several- segment metatrochophore or nectochaeta shortly before settlement (Barnes, 1980; Strathmann, 1987a). Virtually all larval settlement research has been done on tube-building forms. Smith and Chia (1985) reported that larvae of Sabellaria cementarium are positively phototactic and positively geotactic when competent to settle, two normally opposing behaviors, and they did not report which was a stronger cue. They also concluded that sand is essential for metamorphosis, although they did not use antibiotics to eliminate bacteria. Wisely (1960) reported that larvae of the sessile polychaete Spirorbis borealis settle preferentially in grooves, and confirmed this in laboratory observations, but Crisp and Ryland (1960) found that this species avoided rugosity on a smaller scale. High levels of potassium induce settlement and metamorphosis of in Phragmatopoma californica (Yool et al., 1986). These authors propose that potassium ions, which also induce settlement in many other species but which is not common in seawater, artificially depolarize excitatory membranes of external chemoreceptors in larvae and initiates the settlement and metamorphosis process. The authors did

not propose a natural role in settlement for this ion under

normal conditions. Sulphide in seawater or sediments also

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 280 induces settlement and metamorphosis in a cultivar of the infaunal polychaete Capitella (Cuomo, 1985), but sulphide occurs naturally in sediments into which this species settled. It may be difficult, however, to distinguish toxic effects or narcotization by sulphide from settlement behavior, for some species. Many species settle gregariously. Wilson (1968) and Blake (1969) observed that larvae of the tube-building Sabellaria alveolata the small spionid Polydora ligni, respectively, settled near conspecifics. In S. alveolata, the response was strongest for living animals. Smith and Chia (1985) observed that larvae of the tube-building polychaete Sabellaria cementarium tended to settle in pairs, but are not induced to settle by sand from tubes of conspecifics. Hydroides dianthus (Scheltema et a l ., 1981), Phragmatopoma californica (Jensen and Morse, 1990; Pawlik, 1990; Pawlik et al., 1991), and Spirobranchus polycerus (Marsden, 1991), however, are all induced to settle in response to chemicals from the tubes of conspecifics. In Phragmatopoma, the inducer is part of the cement from the tube, and is a protein with many quinone cross-links and dihydroxyphenylalanine (DOPA), but DOPA alone will not induce settlement (Morse, 1991). Settlement of Phragmatopoma has been reviewed by Morse (1990) and Morse (1991).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 281 Kirchman et a l . (1982) studied the settlement induction effect of marine bacteria on the spirorbid Janua (Dexospira) brasilensis. A mixed-species culture of bacteria and a monoculture of the bacteria Deleya marina were both effective in enhancing settlement. Spirorbis borealis settlement was enhanced by an unknown microbial film (Crisp and Ryland, i960) . Woodin (1985) presents field evidence that infers that settlement of the spionid Pseudopolydora kempi is reduced in the presence of active Abarenicola pacifica (a large deposit-feeding polychaete), apparently due to behavioral avoidance by the larvae.

Echiura Echiurans have a planktotrophic trochophore larva, for which there is no specific term. Adults are sedentary, infaunal deposit feeders (Barnes, 1980). Pilger (1978) presents evidence for preferential settlement by Urechis caupo in organic-rich mud.

Mollusca: Polyplacophora Chiton larvae are planktotrophic, and develop into a trochophore larva, for which there is no unique term (Barnes, 1980), but the term protoconch, used for gastropods, is probably applicable here. All chitons are slow-moving as adults and require hard substrate.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 282 Late larvae of two species of Mopalia were apparently unable to metamorphose without a suitable substrate (Watanabe and Cox, 1976). The substrate used was algal- filmed Mytilus shell, but it was not determined whether the inducer was the shell or alga. Barnes and Gonor (1973) found that Tonicella lineata would settle only onto coralline red algae or tile soaked in coralline algal extract, out of all the substrates they tested. Compounds similar to the common neurotransmitter gamma-aminobutyric acid (GABA) have been isolated from red algae, and GABA will induce settlement in Mopalia mucosa and Katharina tunicata, but metamorphosis will not follow unless the larvae are in contact with coralline red algae (Lithothamnion) (Morse et a l ., 1979; Rumrill and Cameron, 1983).

Mollusca: Gastropoda: Opisthcbranchia Opisthobranchs usually have a planktotrophic, shelled larval stage termed a veliger. Upon settlement the larva of most species loose their shell. Most opisthobranchs are grazers on algae or colonial animals (Barnes, 1980). Havenhand (1991) reviews general aspects of opisthobranch larval behavior, and Hadfield (1978) reviews models of opisthobranch settlement induction. Little is known about opisthobranch larval response to light, gravity, or rugosity while settling. Miller and Hadfield (1986) reported that early veliger of Phestilla sibogae are positively

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 283 phototactic, but lose part or all of this response towards the end of the larval stage. Veligers of Haminoe solitaria apparently reguire chemicals from adults to settle and metamorphosis (Harrigan and Alkon, 1978). Potassium ions and choline will induce settlement and metamorphosis of Aldaria proxima (Todd et a l ., 1991), but this is though to occur by depolarization of neuroreceptors by these ions, rather than as a natural settlement cue. Aldaria in nature settles preferentially on living colonies of the bryozoan Electra pilosa, although Electra is not its prey, and Todd et a l . (1991) has demonstrated that a chemical from Electra is the cue. Almost all other known settlement inducers in the opisthobranchs are prey species. Species of sea hare (Aplysia) set preferentially on one or two species of fleshy algae, upon which they graze as adults (Kreigsten et al., 1974; Switzer-Dunlap and Hadfield, 1977). Doridella obscura and Rostangia pulchra, two dorid nudibranchs, settle preferentially onto the bryozoan Electra crustulenta (Perron and Turner, 1977) and the sponge Ophlitospongia pennata (Chia and Koss, 1978), respectively. The dorid Onchidoris bilamellata preys on barnacles; supernatants from barnacles in the seawater is required for the larvae to settle, but metamorphosis cannot occur until the larvae have touched living or dead barnacle tests or tissues (Chia and Koss,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 284 1988). Arkett et a l . (1989) have subsequently identified the specific neuroreceptor cells involved. Tritonia hombergi will settle only when its prey, the bryozoan Alcyonium, is provided (Thompson, 1962). A congener, T. diomedea, is predatory on the hydroid Vlrgularia, and the presence of the hydroid enhances settlement of the nudibranch, but the larvae to not settle directly onto the hydroid (Kempf and Willows, 1977). This infers that the settlement cue is a water-soluble chemical produced by the hydroid. Eubranchus doriae settlement is induced by water soluble chemicals from the prey hydroid, Kirchenpaueria pinnata. This chemical contains galactosidic residues, and settlement can also be induced by hexoses or galactosamide in which the hydroxyl groups on carbons 3 and 4 are in the Cis position (Bahamondes-Rojas and Dherbomez, 1990). Phestilla sibogae settlement and metamorphosis is induced by a small (<500 Daltons), polar chemical produced by their prey, corals of the genus Porites (Hadfield and Pennington, 1990). Non-polar organic solvents, however, also induce metamorphosis in this species, as well as polar organic solvents (Pennington and Hadfield, 1989).

Mollusca: Gastropoda: Prosobranchia Many benthic marine prosobranch snails produce a planktotrophic, shelled, veliger larva, which has a well- developed foot and eyespots (Barnes, 1980). As for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 285 opisthobranchs, however, virtually nothing is known about the role of light or gravity. Crepidula species are some of the relatively few examples of sedentary, suspension-feeding gastropods. McGee and Targett (1989) found that C. fornicata and C. plana settlement and metamorphosis was induced by water-soluble chemicals from conspecifics and congeners. In addition, the chemicals from the hermit crab Pagurus pollicaris, which occupies empty shells of the sort used by Crepidula, also induces metamorphosis. Pechenik and Heyman (1987), however, have also been able to induce settlement in C. fornicata by increased levels of potassium ions, which is not thought to have a natural role in settlement, so the mere ability of a chemical to induce settlement does not infer that it is normally a settlement cue. Gamma-aminobutyric acid (GABA), a common neurotransmitter, has been found to induce settlement an metamorphosis in many species of the abalone, Haliotis, although H. rufescens has been the primary model (Morse et al., 1979; Baloun and Morse, 1984; Barlow, 1990, Searcy- Bernal et al., 1992). GABA-like compounds have been isolated from coralline red-algae, but are apparently not released into the water by the algae, because the larvae must touch the algae to initiate crawling behavior and metamorphosis. GABA has been shown to initiate the formation of cyclic adenomonophosphate (cAMP), an

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 286 intracellular messenger molecule, which them apparently starts the behavioral/metamorphic process. It has further been shown that lysine and some other diamines greatly increase the larval sensitivity to GABA, through separate cellular binding sites. Potassium ions also induce settlement in H. rufescens, and affect the activity of GABA on settlement induction (Baloun and Morse, 1984). Johnson et al. (1991a), however, were unable to isolate GABA from bacteria associated with coralline algae, inferring that GABA is not the specific compound responsible for settlement induction. The pedal mucous of conspecifics, especially in association with a diatom film, has also been shown to induce settlement in Haliotis (Seki and Kan-no, 1981; Slattery, 1992), at higher levels than induction by GABA (Slattery, 1992). Chemical induction of settlement of H. rufescens, a commercial species, has been developed to the point of commercial applications (Searcy-Bernal et a l ., 1992). The Haliotis model is reviewed in detail by Morse (1990) and Morse (1991) .

Coralline red algae, GABA, and an undescribed algal film also induce settlement in the large top snail, Trochus niloticus (Heslinga and Hillman, 1981), although the details are not clearly described. Extracts of the red alga Laurencia poitei induce settlement and metamorphosis of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 287 larvae of the queen conch, Strombus gigas, as does elevated levels of potassium (Davis et al., 1990). Veligers of the mud snail Ilyanassa obsoleta settle onto benthic muddy sand. Sandy mud induces metamorphosis more than mud, but ashed sediments (organic-free) did not induce metamorphosis more than filtered seawater, while seawater conditioned with the substrate could induce settlement (Scheltema, 1961) . This is the only evidence available for a rugosity effect on gastropod veliger settlement, although there is in addition a chemical inducer in this particular case. The chemical inducer may be from adult conspecifics; Levantine and Bonar (198 6) isolated a small carbohydrate (> 1000 Daltons) from adult-inhabited habitat (mud flats) which induced settlement of T. obsoleta. Thompson (1985) found that allelochemicals produced by the sponge Aplysina fistularis inhibited settlement of Haliotis veligers, but since continued exposure to the sponge resulted in the death of the larvae, it cannot be determined whether the effect is due to behavioral modification or simply toxicity.

Mollusca: Bivalvia Most marine bivalves produce a planktotrophic, bivalved, veliger larva, with a planktonic period of days to weeks. The competent stage develops a foot and is known as a pediveliger. Some have apparently photoreceptive

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 288 "eyespots"; most do not (Barnes, 1980). Descriptions and figures of pediveliger larvae of a variety of taxa are given by Chanley and Andrews (1971). The issue of light and gravity roles in pediveliger settlement has been confused by a long history of poor experimentation and hasty interpretation. Hopkins (193 5) and Bonnot (1937) reported apparently contradictory results for the oyster Ostrea lurida; Bonnot reported higher settlement on the upper surfaces, while Hopkins reported higher settlement on lower surfaces of horizonal substrates. In both cases, however, they in fact examined recruitment after several months of settlement and subsequent mortality, and no conclusions should be drawn from their work regarding settlement. Cole and Knight-Jones (1949) review early research on this topic, and found that the literature consistently reported higher recruitment on lower surfaces, in contrast to their own studies, in which they found higher settlement for Ostrea edulis on upper surfaces when the substrate was shaded from above (although their results are not significant, due to high variance; this author, reanalysis). Shaw (1967) placed series two asbestos plates horizontally, one above the other, in the field for a week at a time, and observed settlement/recruitment on the underside of the top plate and the upper side of the lower

plate. When the plates were four inches (10 cm) apart, settlement of the oyster Crassostrea virginica was higher

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 289 down-facing surfaces, but when the plates were only one inch (2.5 cm) apart (and consequently darker between them), settlement was higher on the surface facing upwards. The pattern was reversed for mussel species. Ritchie and Menzel (1969) finally demonstrated negative phototaxis in settling Crassostrea virginica. This author (in review) has confirmed this, and further shown that the pediveligers of this species are about equally responsive to most visible wavelengths. Furthermore, gravity appears not to be a significant factor (this author, in review), at least for this species. Bayne (1964), found that the pediveligers of the mussel Mytilus edulis are also negatively phototactic, but in contrast to C. virginica, they were positively geotactic, with a much stronger response to gravity than to light (however, geotaxis was not separated from barotaxis in this study, a possible alternate explanation). Isham et a l . (1951) found a strong negative phototaxis in settling shipworm pediveligers Teredo (Lyrodus) pedicellata (which do not have visible "eyespots"), with little or no geotactic response. He also remarked on high larval behavior variability, however. Dons (1944) reported that shipworm pediveligers Teredo (Psiloteredo) megotara are rugophobic- that is settlement was higher on wood cut with the grain than across the grain, and highest on planed surfaces. Culliney (1974) reports a

form or rugophilic ("thigmophilic") behavior in settling sea

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 290 scallop pediveligers Placopecten magellanicus- that is, the larvae could be induced to settle simply by adding bits of shell or gravel. He did not use antibiotics to rule out the role of microbes, however. Keck et a l . (1974) reported that pediveligers of the hard clam Mercenaria mercenaria significantly preferred sand to mud in laboratory tests, and since both substrates had a high organic content, although organic-free sand induced slightly less settlement. Pediveligers of the mussel Mytilus edulis settle preferentially onto a variety of filamentous substrates, and this response is enhanced by water currents, but water currents alone are insufficient to induce settlement (Eyster and Pechenik, 1987). It has long been known that may species of oysters settle gregariously (see Cole and Knight-Jones, 1949; and Crisp, 1967, for reviews of early observations). Crisp (1967) found that Crassostrea virginica settlement was enhanced by water-soluble chemicals in both the shell and in pallial fluid from adults, and Vietch and Hidu (1971) isolated a large (>100,000 Daltons) protein from the shell of C. virginica which induces settlement; this protein contained, among other things, tyrosine compounds. Hidu et a l . (1978) showed that the respective inducers from O. edulis and C. virginica would both induce settlement and metamorphosis in larvae of either species. Coon et al . (1985) found that dihydroxyphenylalanine (DOPA), an

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 291 oxidative derivative of tyrosine, induces settlement and metamorphosis in Crassostrea gigas. DOPA is converted to dopamine, a neurotransmitter, within the larva. Epinephrine, norepinephrine, and other catecholamines also tyrosine derivatives, induce metamorphosis without settlement. The response to these compounds has since been shown for C. virginica (Bonar et al., 1990) and O. edulis (Shpigel et al., 1989). Gregarious pediveliger settlement has been shown for the mussel Mytilus edulis (Bayne, 1969), and a protein produced by the adult tissue was found to induce settlement. Petersen (1984) presents evidence for species specificity during settlement, between M. trossulus (formerly edulis)

and M. californianus, although experimental replication was low in that study. Keck et al. (1974) found that waterborne chemicals from adult hard clam Mercenaria mercenaria induced settlement in pediveligers of the same species, and that this cue was stronger than rugosity cues. Shipworm pediveligers (Teredo species), however, appear to avoid

settling on wood containing congeners (Hoagland, 1986), and pediveligers of the angel wing clam Cyrtopleura costata were not induced to settle by the odor of adult conspecifics, DOPA or GABA, but epinephrine, and to a lesser extent, norepinephrine (neurotransmitters which may or may not exist in the natural larval settlement environment), did induce settlement and metamorphosis.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 292 The filamentous red algae Platythamnion induces settlement and metamorphosis in Mytilus trossulus, and extracts from this alga also induce settlement, although to a lesser extent. Lanasol, a tyrosine derivative, is found in the cell walls of many filamentous red algae, and is similar to DOPA, which also induces settlement and metamorphosis in M. trossulus, while GABA does not induce settlement and metamorphosis (Cooper, 1982). Jacaranone, a tyrosine-derivative from the filamentous red alga Delesseria, induces settlement and metamorphosis in the scallop Pecten maximus, as will DOPA and epinephrine, but not norepinephrine (Chevolot et a l . 1991). Chevolot et al. also have pointed out that all known natural chemical inducers for bivalve settlement are or contain tyrosine derivatives which are quinones or readily oxidize to quinones. In contrast to that, obtusaquinone, and its derivative, obstusastyrene, which occur naturally in the tropical rosewood Dalbergia, inhibit settlement and metamorphosis of a tropical shipworm larva, Lyrodus pedicellatus (Turner, 1976). The endolithic mussel Lithophaga lessepsania settles almost primarily on the coral Stylophora pistilla, and tissue extracts of this coral could induce settlement and metamorphosis in this mussel (Mokady et a l ., 1991). Shipworms settle almost exclusively in wood, with relatively fresh wood a stronger inducer. The chemical inducer is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 293 water-soluble, but the nature of the chemical is unknown (Culliney, 1975). Bacterial supernatants from Alteromonas and Vibrio induce settlement of the oyster Crassostrea gigas (Fitt et al., 1990). These supernatants contain both DOPA- like compounds and ammonia, both of which have been demonstrated to induce settlement and metamorphosis in this species (Bonar et a l ., 1990). Higher concentrations have a lower inducing function, however (Fitt et al., 1990).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 294 ANALYSIS OF SETTLEMENT CUES Marine invertebrate larval settlement cues can broadly be divided into abiotic and biotic (chemical), although as will be discussed, there are a few cues for which the distinction is unclear. To review the introductory comments

in this review, larval settlement may be divided into a) a

swimming exploratory stage; b) a crawling exploratory stage;

c) a settlement stage; and d) metamorphosis, which is not behavioral but in some cases seems to be directly linked to settlement behavior. For some active benthic invertebrates, such as decapod crustaceans, the crawling exploratory stage may be absent.

Abiotic cues I. Pressure, gravity, and light. Most invertebrate larvae are negatively buoyant- that is, they sink when they cease to swim. It has also been shown for some species of bivalve mollusks that the larvae of various stages are not distributed randomly in the water column (e.g. Nelson, 1930; Wood and Hargis, 1971; Tremblay and Sinclair, 1990; Sekiguchi et al. 1991). Non-random distribution has also been reported for postlarvae of shrimp, Penaeus (Forbes and Benfield, 198 6). Taken together- negative buoyancy and non-random distribution- some sort of depth regulation behavior is required by larvae. The problem is that under different conditions, water pressure, gravity, or light could be the cue that the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 295 larvae respond to. In some cases, it has even been shown that hydrodynamic processes alone can account for larval distribution in the water column (Shanks and Wright, 1987) and/or larval settlement patterns (Shanks and Wright, 1987; Butman, 1989). This last result demonstrates the need to combine field studies with laboratory observations when studying larval settlement behavior. Larvae of the oyster Crassostrea virginica have a statocyst-like organ, which has been interpreted to be a statocyst (Galtsoff, 1964), although there have been no neurophysiological studies to demonstrate this. Pre- competent veligers of the clam Arctica islandica have been shown to be able to sense pressure independently of gravity and respond with negative barotaxis (Mann and Wolf, 1983), and cyprids of the barnacle Balanus balanoides are also negatively barotactic (Knight-Jones and Morgan, 1966). Barotaxis has been shown clearly for zoea larvae of numerous crab species (Foreward and Buswell, 1989), and there is evidence for it in megalopae as well (Jacoby, 1982; Sulkin, 1984) . In all species for which a response to barometric pressure has been studied, the response has been to increase swimming rate, which is to swim upwards. With the exception of barnacles, these species settle "down" prior to metamorphosis, so it appears that a negative barotactic response, at least by itself, is unlikely to be involved in settlement behavior. Although some barotactic responses are

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 296 remarkably sensitive, barotaxis is unlikely to be useful in final site selection in settlement, when larvae are exploring substrates, because (at least in shallow water) surface waves would provide confusing signals. Since pressure and gravity are usually intimately correlated, a barotactic response is also a geotactic response. For a few species, however, geotaxis has been shown to be independent of barotaxis. The bryozoan Bugula settles high in the water column, and has been shown to be negatively geotactic when settling (Pires and Woollacott, 1983) . As discussed above, some crab megalopae are negatively geotactic, but in a manner that is inconsistent with their settlement. Interestingly, however, Cancer appears to respond to light during the day and gravity at night, suggesting that different cues can become important under different conditions (Jacoby, 1982). Crisp and Gobashy (1971) found no barotactic response for larvae of the ascidian Diplosoma, but were able to show that they were positively geotactic when settling. Light, like pressure, is closely correlated to pressure. Light direction can be experimentally manipulated more readily then pressure, however, and its angle, magnitude, or presence is not constant like pressure or gravity. Phototaxis in some larvae is present throughout most of their planktonic phase, only to disappear shortly before competency. Examples include the sponge Hippospongia

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 297 (Kaye and Reiswig, 1991); the nudibranch Phestilla (Miller and Hadfield, 1986); and the clam Arctica (Mann and Wolf, 1983). In many others, however, negative or positive phototaxis is an important settlement cue, including positive phototaxis in the sponge Haliclona (Kaye and Reiswig, 1991), and in the coral Agaricia (Morse et a l ., 1988). Crisp and Ryland (1960) reported both positive phototaxis and positive geotaxis in the polychaete Spirorbis (or negative barotaxis), two apparently contradictory responses. This is not necessarily a paradox, if one cue is stronger than the other, so that the lesser cue acts only when the stronger cue is absent (e.g. rugotaxis and phototaxis in a barnacle Balanus; Forbes et a l., 1971). Negative phototaxis is more common; examples include a coral Flavia (Lewis, 1974), a sea cucumber Psolus (Young and Chia, 1982), a barnacle Balanus (Forbes et a l. 1971), tunicates Tridldemnum (van Duyl et al., 1981) and Diplosoma (Crisp and Ghobashy, 1971), oysters Crassostrea (Menzel and Ritchie, 1969), and shipworms Teredo (Isham et a l ., 1951). In settling, barotaxis, geotaxis, and phototaxis probably act primarily on the swimming exploratory phase, as general cues merely to get the larva in the right part of the water column for settlement. If a larva possess responses to more than one of these cues, interactions may exist, although this has not been shown for settling larvae. A hypothetical model for a larva that cements permanently

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 298 upon settlement (e.g. barnacles, spirorbid polychaetes, oysters) and thus cannot move after settling, in a sediment- rich environment, is as follows: positive barotaxis or negative phototaxis is used to place the larva at a desired depth in the water column, where a heterogeneous substrate is encountered. Negative phototaxis is used to find the darkest point available (e.g. the underside of a boulder). When light levels become very low, the negative phototaxis is overpowered by negative geotaxis, and the larva swims upward to find the underside of the substrate, which will be sediment-free. If an under-side is not encountered, the larva will continue upwards until barotactic or phototactic behavior is again triggered, to restart the process. This model could explain repeated observations by researchers for oyster larval settlement on the underside of objects. Unfortunately, a geotactic response for oysters during settling has never been confirmed in the laboratory.

Abiotic cues II. Rugosity, hydrophilic surfaces, & currents. Rugotaxis or some form of rugotaxis (e.g. thigmotaxis, rugophilia) has been observed several times. It has been widely observed that mussel Mytilus pediveligers settle preferentially on various fibrous substrates, including silk fibers, filamentous red algae, and adult byssal threads (e.g. Eyster and Pechenik, 1987). A polychaete Sabellaria

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 299 apparently requires sand for metamorphosis (Smith and Chia, 1985), while settlement in a sea pen Ptilosarcus (Chia and Crawford, 1973), a mud snail Ilyanassa (Scheltema, 1961), and a clam Mercenaria (Keck et a l ., 1974) is enhanced by sand as opposed to mud. Some species prefer small-scale (smaller than the larvae) roughness, such as the scyphozoan Cyanaea (Brewer, 1976), while others prefer smooth surfaces at this scale, such as a polychaete Spirorbis (Crisp and Ryland, 1960) and a shipworm Teredo (Dons, 1944). Rugotaxis on the scale of the larva has been reported for various larvae, including a brachipod Waltonia (Percival, 1956; Wisely, 1969), barnacles Balanus, Chthamalus, Elminius, and Semibalanus (Crisp and Barnes, 1954, Wethey, 1986), and a polychaete Spirorbis (Crisp and Ryland, 19 60). In all of the above cases, the larvae settle preferentially within grooves or pits, and Miller and Carefoot (1989) have shown that for barnacles, at least, this decreases mortality due to grazing or disturbance. Small scale rugosity could be detected by the larval appendages, but larger scale rugosity could only be detected by reference to a fixed external cue, such as light angle or gravity, which changes relative to the larva as the larva moves across the uneven surface. Unlike these external cues, however, rugosity cannot be detected until the larva has landed on the substrate, and thus begun the second phase of settlement.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The role of microbial films is still unclear, in part because most researchers have not identified the microbes involved (although when they have been involved, they are usually one of a handful of common bacteria). There are at least two viable models for the induction of settlement by bacteria based on experimental evidence, one of which, the chemical model, will be discussed later. The other model involves the hydrophilic nature of bacterial colonies. When comparing hydrophilic and hydrophobic artificial surfaces, Maki et al. (1989) found that settling bryozoan Bugula coronate larvae preferred the hydrophilic surface. The hydrophobic surface, however, could be made attractive to Bugula by permitting a growth of bacteria, which made the surface hydrophilic. Bugula, like some other marine invertebrates, cement themselves permanently to the substrate when settling, and most known cements require a hydrophilic surface. Many bivalve larvae attach themselves to the surface with a byssal thread, which also involves a

cement, although the attachment is not permanent. Although the hydrophilic nature of substances is not necessarily biotic in origin, because it is enhanced by bacteria it usually in fact is a sort of biotic cue. Researchers must take care, however, to separate bacteria from other microbes; microalgae apparently enhances settlement of the

urchin Strongylocentrotus droebachiensis (Pearce and Scheibling, 1991).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 301 Rugosity and the degree to which a surface is hydrophilic act upon the crawling exploratory phase of settlement, because they both require contact with the surface. The hydrophilic nature of a surface is also important in settlement itself, however, for species that attach themselves to the surface with cements or byssal threads. Water currents have been shown to enhance settlement of mussels Mytilus settling on fibrous substrates (Eyster and Pechenik, 1987), and the polychaete Phragmatopoma settling near conspecifics (Pawlik et al., 1991). If the direction or velocity of a current relative to the substrate surface were a cue, it would be termed "rheotaxis". It is not generally thought that currents are a settlement cue, however, but rather that currents tend to increase encounter rate of the larvae with substrates by passing them over more options. It can also be argued, however, that currents could decrease larval selectivity by decreasing time to "analyze" a potential site.

The problem of inorganic ions. A number of invertebrates can be induced to settle and metamorphosis by the addition of a variety of inorganic ions to the seawater in unnatural concentrations; most notably, potassium, but many other ions as well (e.g. Lynch, 1961; Baloun and Morse, 1984; Yool et al., 1986; Pechenik and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 302 Heyman, 1987; and Todd et al., 1991). These are recognized to be artificial inducers for the most part, which mimic or interact with intercellular neurotransmitters, depolarizing the sensitive membranes of the chemoreceptor cells and initiating the settlement and metamorphosis sequence. Bonar et al. (1990) and Coon et al. (1990) showed that ammonia, which is found naturally in low levels in the water column, could also induce settlement in oysters, Crassostrea. These authors proposed a model in which ammonia is produced either by adult oysters or by bacterial colonies as a byproduct and induces settlement of passing pediveligers. One possible problem with this interpretation is that ammonia is produced by a wide variety of invertebrates, since it is less energetically expensive to produce than urea and is readily dissolved and carried away in water. Ammonia is thus nearly ubiquitous in the marine environment, including over soft sediments that are unsuitable for settlement. It is also readily absorbed by most phytoplankton, however, for which nitrogen is usually limiting in coastal environments, and in shallow areas rarely becomes an abundant ion. It thus seems unlikely that larvae would detect a strong ammonia source, or that if they did, it would be a selectively advantageous cue. Ammonia should probably be added to the list of ions that merely depolarize sensitive chemoreceptor cell membranes.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 303 Sulphide, which is found to induce settlement in a polychaete Capitella (Cuomo, 1985), cannot be as readily dismissed as has been done for ammonia (above). Sulfides are released from organic-rich, anaerobic sediments, as a byproduct of bacterial degradation of proteins, and both the sediments and free sulfides induce settlement in the polychaete. The organic-rich mud is the natural habitat of this species, which is a deposit feeder, and thus sulfides are a good indicator of a food-rich environment. Although sulphide is an inorganic chemical, it is predominantly released by biotic processes, and in this sense is this a biotic, chemical cue. A potential problem for the researcher, however, is to distinguish narcotization effects from settlement behavior. In direct contrast to the findings reviewed above, Rittschof et al. (1986) showed that increased levels of most of the above ions (potassium, magnesium, calcium, sodium) inhibited settlement of a barnacle, Balanus amphitrite, probably through neuroreceptor interference.

Chemical cues I. Gregariousness and Conspecific Avoidance The advantages of gregariousness to the settling larva are obvious, especially for some sessile species. If there are adults or successfully metamorphosed juveniles present, the site must therefore be adequate habitat. For already- settled juveniles or adults, it is advantageous to attract

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 304 more potential mating partners, but beyond a certain point new settlers become significant competitors for food or space. The ideal strategy for adults is to release attractant pheromone until a certain density is reached, and then decrease or cease pheromone production. This model has never been tested; however, one of the major attractants for oyster Crassostrea pediveligers is from the shell of the adults (Vietch and Hidu, 1971), and it is unlikely that the shell is a mechanism of pheromonal control. For oysters, at least, it would appear that the behavioral evolution is at the larval level, responding to incidental adult odors. Aggregative or gregarious settlement in response to water soluble chemical cues from conspecifics has been shown for hydroids Nemertesia and Kirchenpaueria (Williams, 1976); a coral Pocillopora (Harrigan, 1972); sand dollars Dendraster (Highsmith, 1982; Burke, 1984) and Echinarachnius (Pearce and Scheibling, 1990); barnacles Chthamalus, Balanus and Elminius (Larman and Gabbott, 1975; Raimondi, 1988), solitary ascidians Ascidia and Polyandrocarpa (Grave and Nicoll, 1939) ; colonial ascidians Chelyosoma (Young and Braithwaite, 1980), Trididemnem (van Duyl et al., 1981), and Botryllus (Grosberg and Quinn, 1986) ; polychaetes Sabellaria (Wilson, 1968), Polydora (Blake, 1969), Hydroides (Scheltema et al., 1981), Phragmatopoma (Jensen and Morse, 1990; Pawlik et al., 1991), and Spirobranchus (Marsden, 1981); an opisthobranch Haminoe (Harrigan and Alkon, 1978);

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 305 prosobranchs Crepidula (McGee and Targett, 1989), Ilyanassa (Levantine and Bonar, 1986), and Haliotis (Seki and Kan-no, 1981; Slattery, 1992); oysters Crassostrea (Vietch and Hidu, 1971), mussels Mytilus (Bayne, 1969), and clams Mercenaria (Keck et al., 1974). Of these, compounds that induces settlement have been isolated (but never completely identified) for Dendraster, Balanus, Elminius, Phragmatopoma, Ilyanassa, and Crassostrea. Burke (1986) reviews evidence for some of these. In all of the above cases, the compound is thought to be a polypeptide or protein, and in Phragmatopoma and Crassostrea the compounds contain tyrosine derivatives. Chevolot et al. (1991) has pointed out that all identified natural settlement and metamorphosis inducers in bivalve mollusks are tyrosine derivatives that are either quinones or which readily oxidize to quinones.

Gregariousness may occur prior to settlement. Young and Chia (1982) found that pre-competent vitellaria larvae of the sea cucumber Psolus tended to aggregate, as did the competent pentacula larvae. Settlement in this species was gregarious, both with respect to each other or to adult conspecifics, although a specific chemical cue was not isolated. There are two main kinds of conspecific avoidance in settling larvae, differing primarily in the degrees of avoidance. Territoriality has been observed in barnacles

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 306 Elminius (Crisp, 1961), in which the larvae are first attracted to settle near conspecifics, and then space themselves uniformly relative to juveniles. Crisp (1961) observed crawling behavior prior to settlement and concluded that the larvae use tactile cues to space themselves, but if the larvae are able to distinguish between other barnacles and non-living irregularities, some chemical cue must be involved. The other kind of conspecific avoidance is found in blue crab Callinectes megalopae (B. Layman, VIMS, unpubl. data) and shipworm Teredo pediveligers (Hoagland, 1986), in which larvae apparently actively avoid settling near adults. Callinectes are cannibalistic (e.g. Mansour and Lipcius, 1991), and the advantages for avoiding adults are obvious. Teredo activity can destroy untreated wood in less than a year (pers. obs.), so space/food competition in shipworms is especially intense.

Chemical Cues II. Prey, Habitat, and Bacteria Most opisthobranch gastropods are predatory or grazers on sessile invertebrates or algae, and most species in which settlement have been studied settle preferentially onto their prey, in response to water-soluble chemicals produced by their prey, although the dorid Onchidoris cannot metamorphose until it makes physical contact with its barnacle prey. For Eubranchus, the cue has been shown to contain galactosidic residues from the hydroid prey

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 307 (Bahamondes-Rojas and Dherbomez, 1990). The queen conch Strombus feeds on algae, and extracts from the red alga Laurentia induces settlement of veligers of this species (Davis et al., 1990). The advantage of this behavior to the settling larva is obvious, while there is just as clearly no advantage to they prey species involved. This, therefore, is further indirect evidence for selection at the larval level for gregariousness (above), rather than at the adult level. A similar and sometimes indistinguishable settlement behavior is induction by specific, beneficial habitats. Gregariousness and attraction to prey, in fact, are merely two specialized forms of habitat selection. An extreme form of habitat selection is commensalism or parasitism. In these cases selectivity is usually assumed, but it has rarely been demonstrated experimentally. One for which selectivity has been demonstrated is the hydroid Aplidium, commensal on several sabellid polychaetes. Interestingly, the hydroid can also be induced to settle on sympatric related sabellids which do not support the commensalism (Donaldson, 1974). The opisthobranch Aldaria settles preferentially onto the bryozoan Electra, which is not its adult prey, but which, apparently, is a benign habitat for postlarval Aldaria. Pediveligers of the endolithic mussel Lithophaga are attracted to its primary host, the coral Stylophora (Mokady et a l ., 1991). An ascidian Ascidia

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 308 settles preferentially onto a sponge Terpios, relative to bare patches or artificial sponges (Bingham and Young, 1991). Very specific habitat selections by settling larvae are probably very widespread. Less specific habitat selection has been observed for the detritivorous mud snail Ilyanassa (Scheltema, 1961), settling onto organic-rich mud, hydroids Sertularella and Coryne settling onto algae

(Nishihira, 1968, 1969), and shrimp Penaeus to estuarine habitats (Benfield and Aldrich, 1992). Filamentous red algae, both encrusting calcareous (coralline) and non-calcareous forms, contain compounds that induce settlement in a variety of invertebrate larvae. Filamentous non-calcareous red algae can produce water- soluble compounds that induce settlement in mussels Mytilus (Cooper, 1982) and scallops Pecten (Chevolot et al., 1991). Coralline red algae induce settlement in a coral Agaricia (Morse, 1991); sea urchins Strongylocentrotus (Scheibling, 1988), Pseudocentrotus, and Anthocidaris (Kitamura et al., 1992); chitons Tonicella (Barnes and Gonor, 1973) , Mopalia (Morse et al., 1979), and Katharina (Rumrill and Cameron), a top snail Trochus (Heslinga and Hillman, 1981), and many abalone Haliotis species (Morse, 1991). In the case of the coralline algae, it has never been adequately determined whether the alga is prey or merely a food-rich, predator- free environment. Three classes of compounds that induce settlement have been isolated from red algae. DOPA-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 309 mimicking compounds, (but not DOPA specifically; Johnson et al, 1991a), which induce settlement in bivalve mollusks, are apparently water-soluble, and contain tyrosine-derived compounds (of which DOPA is an example), as discussed by Chevolot et al. (1991). GABA-mimicking compounds, which induce settlement in gastropods and chitons, are not released into the water by the algae, although they can be made water soluble. A sulphated polysaccharide from the cell wall induces settlement in the coral Agaricia, and is also not released into the water column. To confuse the issue, in at least one species, the sea star Acanthaster, settles onto coralline algae, but the use of antibiotics reduces this response (Johnson et al., 1991b), inferring that the chemical cue is produced by bacteria associated with the algae.

Non-aqueous chemical cues present a theoretical problem. If the larva cannot detect the desired substrate until it contacts it, much would appear to be left to chance in terms of settlement site selection. In laboratory observations, however, larvae do not show a response until they happen to physically encounter coralline algae. Coralline algae are very common in their respective habitats, but by no means ubiquitous, and compete for space with environments inimical to invertebrate juveniles such as colonial invertebrates and sessile benthic planktivores. It seems that some cue or set of cues, either positive towards

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 310 coralline algae or negative towards inimical substrates, is required. Bacterial films have often been shown to induce settlement, but in many cases it is not known whether the cues is chemical or physical. Chemical cues by bacteria have been demonstrated for a scyphozoan Cassiopea by Vibrio (Neumann, 1979), a barnacle Balanus by unidentified bacteria (Maki et al., 1990), a sea star Acanthaster by unidentified bacteria (Johnson et al., 1991b), and an oyster Crassostrea by Alteromonas or Vibrio (Fitt et al., 1990). Exopolymers have been implicated, as in the case of Crassostrea, DOPA- like compounds have been isolated, as predicted by the model by Chevolot et al. (1991) for settlement induction in bivalve mollusks.

Chemical Cues III. Allelopathy and Avoidance. Allelopathy with respect to invertebrate settlement is the production of noxious chemicals by existing epifauna that inhibit settlement, and is thus controlled by the

existing epifaunal organism. Avoidance is a behavioral response to an inimical environment detected chemically, and is controlled by the larva. The two can be confused without careful experimentation, and in some case both allelopathy and avoidance may exist. Territoriality and avoidance of conspecifics (discussed previously) are specialized kinds of avoidance behavior.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 311 Several sponges and colonial ascidians have been suggested to produce allelopathic compounds, and death of larvae by continual exposure has been shown several times (e.g. Sebens, 1983; Thompson, 1985; Davis et al., 1990). Avoidance has been shown for a bryozoan Bugula towards an ascidian Eudistoma, a polychaete Pseudopolydora towards another polychaete Abarenicola (Woodin, 1985), and an ascidian Podoclavella towards sponges (Davis et al., 1991). Several species of bacteria apparently inhibit settlement by some invertebrate larvae. Examples include Deleya marina, which inhibits settlement of a bryozoan Bugula (Maki et al., 1988) and a barnacle Balanus (Maki et al., 1990), and Vibrio vulnificus, which inhibits Balanus (Maki et al., 1990). The mechanism of inhibition is not known, however.

The role of larval age and behavioral variability. Some invertebrate larvae apparently cannot settle without a specific cue, such as an opisthobranch Haminoe (Harrigan and Alkon, 1978). This seems like an evolutionary dead-end, however, and in fact, in most cases inducers merely enhance settlement, and many studies have shown that older larvae are less particular, or will settle even on unsuitable habitats. Pechenik and Cerulli (1991) have shown that the cost of delayed settlement in a lecithotrophic polychaete Capitella is a lowered chance of post-settlement

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 312 survival, but of the survivors, there was no reduced growth or fitness. This simplifies the risk calculations that must be "considered" by a competent larvae. Planktotrophic larvae probably have fewer constraints on settlement delay. Several authors have noted high settlement behavioral variability in invertebrate larvae, so that the apparently optimal habitat is selected only part of the time. This is probably a parental strategy: to produce larvae with a range of settlement behaviors, and the variability should increase as the heterogeneity or instability of the environment increases.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 313 References Arkett, S.A., F.-S. Chia, J.I. Goldberg, and R. Koss. 1989. Identified settlement receptor cells in a nudibranch veliger respond to specific cue. Biol. Bull. 176:155- 160. Bahamondes-Rojas, I. and M. Dherbomez. 1990. Purification partielle de substances glycoconjuguees capables d'induire la metamorphose de larves competentes d'Eubranchus doriae (Trinchese, 1879), mollusque nudibranche. J. Exp. Mar. Biol. Ecol. 144:17-27. Baloun, A.J. and D.E. Morse. 1984. Ionic control of settlement and metamorphosis in larval Haliotis rufescens (Gastropoda). Biol. Bull. 167:124-138. Barlow, L.A. 1990. Electrophysical and behavioral responses of larvae of the red abalone (Haliotis rufescens) to settlement-inducing substances. Bull. Mar. Sci. 46:537-554. Barnes, J.R. and J.J. Gonor. 1973. The larval settling response of the lined chiton Tonicella lineata. Mar. Biol. 20:259-264. Barnes, R.D. 1980. Invertebrate Zoology. 4th Ed. Saunders College, Philadelphia, PA. 1089 pp. Barnett, B.E. and D.J. Crisp. 1979. Laboratory studies of gregarious settlement in Balanus balanoides and Elminius modestus in relation to competition between these species. J. Mar. Biol. Assoc. U.K. 59:581-590. Bayne, B.L. 1964. The responses of the larvae of Mytilus edulis L. to light and to gravity. Oikos 15:162-174. Bayne, B.L. 1969. The gregarious behavior of the larvae of Ostrea edulis L. at settlement. J. Mar. Biol. Assoc. 49:327-356.

Benfield, M.C. and D.V. Aldrich. 1992. Attraction of postlarval Penaeus aztecus Ives and P. setiferus (L.) (Crustacea: Decapoda: Penaeidae) to estuarine water in a laminar-flow choice chamber. J. Exp. Mar. Biol. Ecol. 156:39-52. Bingham, B.L. and C.M. Young. 1991. Influence of sponges on invertebrate recruitment: a field test of allelopathy. Mar. Biol. 109:19-26.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 314 Blake, J.A. 1969. Reproduction and larval development of Polydora from northern New England (Polychaeta: Spionidae). Ophelia 7:1-63. Bonar, D.B., S.L. Coon, M. Walch, R.M. Weiner, and W. Fitt. 1990. Control of oyster settlement and metamorphosis by endogenous and exogenous chemical cues. Bull. Mar. Sci. 46:484-498. Bonnot, P. 1937. Settling and survival of spat of the Olympia oyster, Ostrea lurida, on upper and lower horizontal surfaces. Calif. Fish Game 23:224-228. Brancato, M.S. and R.M. Woollacott. 1982. Effect of microbial films on settlement of bryozoan larvae (Bugula simplex, B. stolonifera, and B. turrita). Mar. Biol. 71:51-56. Breese, W.P. and F.D. Phibbs. 1972. Ingestion of bivalve molluscan larvae by the polychaete annelid Polydora ligni. Veliger 14:274 + 1 plate. Brewer, R.H. 1976. Larval settling behavior in Cyanea capillata (Cnidaria: Scyphozoa). Biol. Bull. 150:183- 199. Burke, R.D. 1983. Neural control of metamorphosis in Dendraster excentricus. Biol. Bull. 164:176-188. Burke, R.D. 1984. Pheromonal control of metamorphosis in the Pacific sand dollar Dendraster excentricus. Science 225:442-443. Burke, R.D. 1986. Pheromones and the gregarious settlement of marine invertebrate larvae. Bull. Mar. Sci. 39:323- 331. Butman, C.A. 1989. Sediment-trap experiments on the importance of hydrodynamical processes in disturbing settling invertebrate larvae in near-bottom waters. J. Exp. Mar. Biol. Ecol. 134:37-88. Buyanovski, A.I. and E.V. Solhina. 1991. Molluscan larvae as prey of Polydora limicola (Annekova). J. Moll. Stud. 58:79-81. Cameron, R.A. and R.T. Hinegardner. 1974. Initiation of metamorphosis in laboratory cultured sea urchins. Biol. Bull. 146:335-342.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 315 Chanley, P.E. and J.D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia. Malacologia 11 (1):45-119. Chevolot, L. , J.-C. Cochard, and J.-C. Yvin. 1991. Chemical induction of larval metamorphosis of Pecten maximus with a note on the nature of naturally occurring triggering substances. Mar. Ecol. Prog. Ser. 74:83-89. Chia, F.-S. and B.J. Crawford. 1973. Some observations on gametogenesis, larval development, and substratum selection of the sea pen, Ptilosarcus gurneyi. Mar. Biol. 23:73-82. Chia, F.-S. and R. Koss. 1978. Development and metamorphosis of the planktotrophic larvae of Rostangia pulchra (Mollusca: Nudibranchia). Mar. Biol. 46:109- 119. Chia, F.-S. and R. Koss. 1988. Induction of settlement and metamorphosis of the veliger larvae of the nudibranch, Onchidoris bilamellata. Int. J. Invert. Reprod. Develop. 14:53-70. Cloney, R.A. 1987. Phylum Chordata, Class Ascidiacea. 607- 639. In Strathmann, M.F. (ed.). Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. Cole, H .A. and E.W. Knight-Jones. 1949. The setting behavior of larvae of the European flat oyster Ostrea edulis L. U.K. Ministry Agriculture & Fisheries Fish. Invest. Ser. 2, Vol. 17, No 3. 39 pp. Coon, S.L., D.B. Bonar, and R.M. Weiner. 1985. Induction of settlement and metamorphosis of the Pacific oyster Crassostrea gigas (Thunberg) by l-DOPA and catecholamines. J . Exp. Mar. Biol. Ecol. 94:211-221. Coon, S.L., M. Walch, W.K. Fitt, R.M. Weiner, and D.B. Bonar. 1990. Ammonia induces settlement behavior in oyster larvae. Biol. Bull. 179:297-303. Cooper, K. 1982. A model to explain the induction of settlement and metamorphosis of planktonic eyed- pediveligers of the blue mussel Mytilus edulis L. by chemical and tactile cues. J. Shellfish Res. 2:117 (Abstract).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 316 Cranfield, H.J. 1973. Observations on the behaviour of the pediveliger of Ostrea edulis during attachment and cementing. Mar. Biol. 22:203-209. Crisp, D.J. 1961. Territorial behavior in barnacle settlement. J. Exp. Biol. 38:429-446. Crisp, D.J. 1967. Chemical factors inducing settlement in Crassostrea virginica (Gmelin). J . Animal Ecol. 36:329- 335. Crisp, D.J. and H. Barnes. 1954. The orientation and distribution of barnacles at settlement with particular reference to surface contour. J. Animal Ecol. 23:142- 162. Crisp, D.J. and A.F.A.A. Ghobashy. 1971. Responses of the larvae of Diplosoma listerianum to light and gravity. 4th Europ. Mar. Biol. Symp.:433-465. Crisp, D.J. and D.A. Ritz. 1973. Responses of cirripede larvae to light. I. Experiments with white light. Mar. Biol. 23:327-335. Crisp, D.J. and J.S. Ryland. 1960. Influence of filming and of surface texture on the settlement of marine organisms. Nature 185:119. Culliney, J.L. 1974. Larval development of the giant scallop Placopecten magellanicus (Gmelin). Biol. Bull. 147:321-332. Culliney, J.L. 197 5. Comparative larval development of the shipworms Bankia gouldi and Teredo navalis. Mar. Biol. 29:245-251. Cuomo, M.C. 1985. Sulphide as a larval settlement cue for Capitella sp. I. Biogeochemistry 1:169-181.

Davis, A.R. 1987. Variation in recruitment of the subtidal colonial ascidian Podoclavella cylindrica (Quoy & Gaimard): the role of substratum choice and early survival. J. Exp. Mar. Biol. Ecol. 106:57-71. Davis, A.R., A.J. Butler, and I. van Altena. 1991. Settlement behavior of ascidian larvae: preliminary evidence for inhibition by sponge allelochemicals. Mar. Ecol. Prog. Ser. 72:117-123.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 317 Davis, A.R. and A.E. Wright. 1990. Inhibition of larval settlement by natural compounds from the ascidian Eudistoma olivaceum (Van Name). J . Chem. Ecol. 16:1349- 1357. Davis, M . , W.D. Heyman, W. Harvey, and C.A. Withstandley. 1990. A comparison of two inducers, KC1 and Laurencia extracts, and techniques for the commercial scale induction of metamorphosis in queen conch Strombus gigas Linnaeus, 1758 larvae. J. Shellfish Res. 9:67-73. Denley, E.J. and A.J. Underwood. 1979. Experiments on factors influencing settlement, survival, and growth of two species of barnacles in New South Wales. J . Exp. Mar. Biol. Ecol. 36:269-293. Donaldson, S. 1974. Larval settlement of a symbiotic hydroid: specificity and nematocyst response in planulae of Proboscidactyla flavicirrata. Biol. Bull. 147:573-585. Dons, C. 1944. Marine boreorganismer. VIII. Teredo- angrepenes avhengighet av treverkets struktur og stilling. Norske Vidensk. Selsk. Forh. 17(10):40-42. (English abstract.) Duggins, D.O., J.E. Eckman, and A.T. Sewell. 1990. Ecology of understory kelp environments. II. Effects of kelps on recruitment of benthic invertebrates. J. Exp. Mar. Biol. Ecol. 143:27-45. Eyster, L.S. and J.A. Pechenik. 1987. Attachment of Mytilus edulis L. larvae on algal and byssal filaments is enhanced by water agitation. J . Exp. Mar. Biol. Ecol. 1987 114:99-110. Fitt, W.K., S.L. Coon, M. Walch, R.M. Weiner, R.R. Colwell, and D.B. Bonar. 1990. Settlement behavior and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supernatants. Mar. Biol. 106:389- 394. Fitt, W.K., D.K. Hofmann, M. Wolk, and M. Rahat. 1987. Recruitments of exogenous inducers for metamorphosis of axenic larvae and buds of Cassiopea andromeda (Cnidaria: Scyphozoa). Mar. Biol. 94:415-422. Forbes, A.T. and M.C. Benfield. 1986. Tidal behaviour of post-larval penaeid prawns (Crustacea: Decapoda: Penaeidae) in a southeast African estuary. J . Exp. Mar. Biol. Ecol. 102:23-34.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 318 Forbes, L., M.J.B. Seward, and D.J. Crisp. 1971. Orientation to light and shading response in barnacles. 4th Europ. Mar. Biol. Symp.:53 9-558. Foreward, R.B., Jr. and C.U. Buswell. 1989. A comparative study of behavioral responses of larval decapod crustaceans to light and pressure. Mar. Behav. Physiol. 16:43-56. Fry, W.G. 1971. Biology of larvae of Ophlitaspongia seriata from two North Wales populations. 4th Europ. Mar. Biol. Symp.:155-178 Gaines, S.D. and J. Roughgarden. 1987. Fish in offshore kelp forests affect recruitment to intertidal barnacle populations. Science 235:479-481. Galtsoff, P.S. 1964. The American oyster Crassostrea virginica Gmelin. Fish. Bull. 64:1-480. Grave, C. and P.A. Nicoll. 1939. Studies of larval life and metamorphosis in Ascidia nigra and species of Polyandrocarpa. Papers Tortugas Lab. 32:3-46. Grosberg, R.K. and J.F. Quinn. 1986. The genetic control and consequences of kin recognition by the larvae of a colonial marine invertebrate. Nature 322:456-459. Gustafson, R.G., R.L. Creswell, T.R. Jacobsen, and D.E. Vaughan. 1991. Larval biology and mariculture of the angelwing clam, Cyrtopleura costata. Aquaculture 95:257-219. Hadfield M.G. 1978. Metamorphosis and settlement in marine molluscan larvae. 165-175. In Chia, F.-S., and M.E. Rice (eds.). Settlement and Metamorphosis of Marine Invertebrate Larvae. Elsevier/North Holland Biomedical Press, New York, NY.

Hadfield, M.G. and J.T. Pennington. 1990. Nature of the metamorphic signal and its internal transduction in larvae of the nudibranch Phestilla sibogae. Bull. Mar. Sci. 46:455-464. Harrigan, J.F. 1972. Behavior of the planula larva of the scleractinian coral Pocillopora damicornis. (L.). Amer. Zool. 12:723. (Abstract). Harrigan, J.F. and D.L. Alkon. 1978. Laboratory cultivation of Haminoe solitaria (Say, 1822) and Elysia chlorotica (Gould, 1870). Veliger 21:299-305.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 319 Havenhand, J.N. 1991. On the Behavior of opisthobranch larvae. J . Moll. Stud. 57:119-131. Heslinga, G.A. and A. Hillman. 1981. Hatchery culture of the commercial top snail Trochus niloticus in Palau, Caroline Islands. Aquaculture 22:35-43. Hidu, H., W.G. Valleau, and F.P. Vietch. 1978. Gregarious setting in European and American oysters— response to surface chemistry and waterborne pheromone. Proc. Natl. Shellf. Assoc. 68:11-16. Highsmith, R.C. 1982. Induced settlement and metamorphosis of sand dollar (Dendraster excentricus) larvae in predator-free sites: adult sand dollar beds. Ecology 63:329-337. Hoagland, K.E. 1986. Effects of temperature, salinity, and substratum on larvae of the shipworms Teredo bartschi Clapp and T . navalis (Bivalvia: Teredinidae). Amer. Malacol. Bull. 4:89-99. Hopkins, A.E. 1935. Attachment of larvae of the Olympia oyster, Ostrea lurida, to plane surface. Ecology 16:82- 87. Hurlbut, C.J. 1991. Larval substratum selection and postsettlement mortality as determinants of the distribution of two bryozoans. J . Exp. Mar. Biol. Ecol. 147:103-119. Isham, L.B., F.G.W. Smith, and V. Springer. 1951. Marine borer attack in relation to conditions of illumination. Bull. Mar. Sci. Gulf Caribb. 1:46-63. Jacoby, C.A. 1982. Behavioral responses of the larvae of Cancer magister Dana (18 52) to light, pressure, and gravity. Mar. Behav. Physiol. 8:267-283. Jensen, R.A. and D.E. Morse. 1990. Chemically induced metamorphosis of polychaete larvae in both the laboratory and ocean environments. J. Chem. Ecol. 16:911-930. Johnson, C.R., D.G. Muir, and A.L. Reysenbach. 1991a. Characteristic bacteria associated with surfaces of coralline algae: a hypothesis for bacterial induction of marine invertebrate larvae. Mar. Ecol. Prog. Ser. 74:281-294.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 320 Johnson, C.R., D.C. Sutton, R.R. Olson, and R. Giddins. 1991b. Settlement of crown-of-thorns starfish: role of bacteria on surfaces of coralline algae and a hypothesis for deepwater recruitment. Mar. Ecol. Prog. Ser. 71:143-162. Kaye, H.R. and M.M. Reiswig. 1991. Sexual reproduction in four Caribbean commercial sponges. III. Larval behavior, settlement and metamorphosis. Invert. Repro. Develop. 19:25-35. Keck, R . , D. Maurer, and R. Malouf. 1974. Factors influencing the setting behavior of larval hard clams, Mercenarla mercenaria. Proc. Natl. Shellf. Assoc. 64:59-67. Kempf, S.C. and A.O.D. Williams. 1977. Laboratory culture of the nudibranch Tritonia diomedea Bergh (Tritoniidae: Opisthobranchia) and some aspects of its behavioral development. J. Exp. Mar. Biol. Ecol. 30:261-276. Kirchman, D., S. Graham, D. Reish, and R. Mitchell. 1982. Bacteria induce settlement and metamorphosis of Janua (Dexospira) brasilensis Grube. J. Exp. Mar. Biol. Ecol. 56:153-163. Kitamura, H., S. Kitahara, and K. Hirayama. 1992. Lipophilic inducers extracted from Corallina pilulifera for larval settlement and metamorphosis of two sea urchins Pseudocentrotus depressus and Anthocidaris crassispina. Nippon Suisan Gakkaishi 58:75-78. Knight-Jones, E.W. and E. Morgan. 1966. Responses of marine animals to changes in hydrostatic pressure. Oceanogr. Mar. Biol. Ann. Rev. 4:267-299. Knight-Jones, E.W. and J.P. Stevenson. 1950. Gregariousness during settlement in the barnacle Elminius modestus. J. Mar. Biol. Assoc. U.K. 29:281- 297. Kriegstein, A.R. , V. Castellucci, and E.R. Kandel. 1974. Metamorphosis of Aplysia californica in laboratory culture. Proc. Natl. Acad. Sci. 71:3654-3 658. Larman, V.N. and P.A. Gabbott. Settlement of cyprid larvae of Balanus balanoides and Elminius modestus induced by extracts of adult barnacles and other marine animals. J. Mar. Biol. Assoc. U.K. 55:183-190.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 321 Layman, B.C. 1992. Role of habitat features and chemical cues in substrate selection by blue crab megalopae: evidence from laboratory cues. M.A. Thesis, College of William and Mary, Williamsburg, VA. 68 pp. Le Tourneaux, F. and E. Bourget. 1988. Importance of physical and biological settlement cues used at different spatial scales by the larvae of Semibalanus balanoides. Mar. Biol. 97:57-66. Levantine, P.L. and D.A. Morse. 1986. Metamorphosis of Ilyanassa obsoleta: natural and artificial inducers. (Abstract). Amer. Zool. 26(5):14A Lewis, C.A. 1978. A review of substratum selection in free-living and symbiotic cirripeds. 207-218. In Chia, F.-S., and M.E. Rice (eds.). Settlement and Metamorphosis of Marine Invertebrate Larvae. Elsevier/North Holland Biomedical Press, New York, NY. Lewis, J.B. 1974. The settlement behavior of planulae larvae of the hermatypic coral Favia fragum (Esper). J. Exp. Mar. Biol. Ecol. 15:165-172. Lynch, W.F. 1961. Extrinsic factors influencing metamorphosis in bryozoan and ascidian larvae. Amer. Zool. 1:59-66. McGee, B.L. and N.M. Targett. 1989. Larval habitat selection in Crepidula (L.) and its effect on adult distribution patterns. J . Exp. Mar. Biol. Ecol. 131:195-214. Mair, J.McD. 1980. Salinity and water-type preference of four species of postlarval shrimp (Penaeus) from west Mexico. J. Exp. Mar. Biol. Ecol. 45:69-82. Maki, J.S., D. Rittschof, M.-O. Samuelsson, U. Szewzyk, A.B. Yule, S. Kjelleberg, J.D. Costlow, and R. Mitchell. 1990. Effect of marine bacteria and their exopolymers on the attachment of barnacle cypris larvae. Bull. Mar. Sci. 46:499-511. Maki, J.S., D. Rittschof, A.R. Schmidt, A.G. Snyder, and R. Mitchell. 1989. Factors controlling attachment of bryozoan larvae: a comparison of filmed and unfilmed surfaces. Biol. Bull. 177:295-302. Mann, R. and C.C. Wolf. 1983. Swimming behavior of larvae of the ocean quahog Arctica islandica in response to pressure and temperature. Mar. Ecol. Prog. Ser. 13:211- 218.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 322 Mansour, R.A. and R.N. Lipcius. 1991. Density-dependent foraging and mutual interference in blue crabs preying upon infaunal clams. Mar. Ecol. Prog. Ser. 72:239-246. Marsden, J.R. 1991. Response of planktonic larvae of the serpulid polychaete Spirobranchus polycerus var. augeneri to an alga, adult tubes and conspecific larvae. Mar. Ecol. Prog. Ser. 71:245-251. Miller, K.M. and T.H. Carefoot. 1989. The role of spatial and size refuges in the interaction between juvenile barnacles and grazing limpets. J. Exp. Mar. Biol. Ecol. 134:157-174. Miller, S.E. and M.G. Hadfield. 1986. Ontogeny of phototaxis and metamorphic competence in larvae of the nudibranch Phestilla sibogae Bergh (Gastropoda: Opisthobranchia). J . Exp. Mar. Biol. Ecol. 97:95-112. Mokady, 0., D.B. Bonar, G. Arazi, and Y. Loya. 1991. Coral host specificity in settlement and metamorphosis of the date mussel Lithophaga lessepsania (Vaillant, 1865). J. Exp. Mar. Biol. Ecol. 146:205-216. Morse, A.N.C. 1991. How do planktonic larvae know where to settle? Amer. Sci. 79:154-167. Morse, D.E. 1990. Recent progress in larval settlement and metamorphosis: closing the gap between molecular biology and ecology. Bull. Mar. Sci. 46:465-483. Morse, D.E., N. Hooker, A.N.C. Morse, and R.A. Jensen. 1979. y-aminobutyric acid, a neurotransmitter, induces planktonic abalone larvae to settle and begin metamorphosis. Science 204:407-410. Morse, D.E., N. Hooker, A.N.C. Morse, and R.A. Jensen. 1988. Control of larval metamorphosis and recruitment in sympatric agariciid corals. J. Exp. Mar. Biol. Ecol. 116:193-217. Neumann, R. 1979. Bacterial induction of settlement and metamorphosis in the planula larvae of Cassiopea andromeda (Cnidaria: Scyphozoa: Rhizostomeae). Mar. Ecol. Prog. Ser. 1:21-28. Nishihira, M. 1967. Observations on the selection of algal substrata by hydrozoan larvae, Sertularella miurensis in nature. Bull. Mar. Biol. Sta. Asamushi 13:35-48.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 323 Nishihira, M. 1968. Brief experiments on the effect of algal extracts in promoting the settlement of the larvae of Coryne uchidai (Hydrozoa). Bull. Mar. Biol. Sta. Asamushi 13:91-101. Olafsson, E.B. 1988. Inhibition of larval settlement to a soft bottom benthic community by drifting algal mats: an experimental test. Mar. Biol. 97:571-574. Pawlik, J.R. 1990. Natural and artificial induction of metamorphosis of Phragmatopoma lapidosa californica (Polychaeta: Sabellariidae), with a critical look at the effect of bioactive compounds on marine invertebrate larvae. Bull. Mar. Sci. 46:512-536. Pawlik, J.R., C.A. Butman, V.R. Starczak. 1991. Hydrodynamic facilitation of a tube-building reef worm. Science 251:421-424. Pearce, C.M. and R.E. Scheibling. 1988. Larval settlement in the green sea urchin Strongylocentrotus droebachiensis. Amer. Zool. 28:71A. (Abstract). Pearce, C.M. and R.E. Scheibling. 1990. Induction of settlement and metamorphosis in the sand dollar Echinarachnius parma: evidence for an adult-associated factor. Mar. Biol. 107:363-369. Pearce, C.M. and R.E. Scheibling. 1991. Effect of macroalgae, microbial films, and conspecifics on the induction of metamorphosis of the green sea urchin Strongylocentrotus droebachiensis (Muller). J. Exp. Mar. Biol. Ecol. 147:147-162. Pechenik, J.A. and T.R. Cerulli. 1991. Influence of delayed metamorphosis on survival, growth, and reproduction of the marine polychaete Capitella sp. I. J. Exp. Mar. Biol. Ecol. 151:17-27.

Pechenik, J.A. and W.D. Heyman. 1987. Using KC1 to determine size at competence for larvae of the marine gastropod Crepidula fornicata (L.). J. Exp. Mar. Biol. Ecol. 112:27-38. Pennington, J.T. and M.G. Hadfield. 1989. Larvae of a nudibranch mollusc (Phestilla sibogae) metamorphose when exposed to common organic solvents. Biol. Bull. 177:350-355. Percival, E. 1956. The settlement and metamorphosis of marine animals. Proc. New Zealand Ecol. Soc. 3:32-33.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 324 Perkins, E.B. 1931. A study of oyster problems in Barnegat Bay. Rep. Dept. Biol. NJ Agric. Exp. Sta. for 1930: 25- 47. Perron, F.E. and R.D. Turner. 1977. Development, metamorphosis, and natural history of the nudibranch Doridella obscura Verrill (Corambidae, Opisthobranchia). J . Exp. Mar. Biol. Ecol. 27:171-185. Petersen, J.H. 1984. Larval settlement behavior in competing species: Mytilus californianus Conrad and M. edulis L. J. Exp. Mar. Biol. Ecol. 82:147-159. Pilger, J. 1978. Settlement and metamorphosis in the Echiura: a review. 103-112. In Chia, F.-S., and M.E. Rice (eds.). Settlement and Metamorphosis of Marine Invertebrate Larvae. Elsevier/North Holland Biomedical Press, New York, NY. Pires, A. and R.M. Woollacott. 1983. A direct and active influence of gravity on the behavior of a marine invertebrate larva. Science 22 0:731-732. Prytherch, H.F. 19 34. The role of copper in the setting, metamorphosis, and distribution of the American oyster, Ostrea virginica. Ecol. Monogr. 4:47-107. Pyefinch, K.A. and F.S. Downing. 1949. Notes on the general biology of Tubularia larynx Ellis and Solander. J. Mar. Biol. Assoc. U.K. 28:21-43. Raimondi, P.T. 1988. Settlement cues and determination of the vertical limit of an intertidal barnacle. Ecology 69:400-407. Reed, C.G. 1987a. Phylum Brachiopoda. 486-493. In Strathmann, M.F. (ed.). Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. Reed, C.G. 1987b. Phylum Bryozoa. 494-510. In Strathmann, M.F. (ed.). Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. Ritchie, T.P. and R.W. Menzel. 1969. Influence of light on larval settlement of American oysters. Proc. Natl. Shellf. Assoc. 59:116-120. Rittschof, D., J. Maki, R. Mitchell, and J.D. Costlow. 1986. Ion and neuropharmacological studies of barnacle settlement. Neth. J. Sea Res. 20:269-275.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 325 Roberts, D., D. Rittschof, E. Holm, and A.R. Schmidt. 1991. Factors influencing initial larval settlement: temporal, spatial and surface molecular components. J. Exp. Mar. Biol. Ecol. 150:203-211. Rumrill, S.S. and R.A. Cameron. 1983. Effects of gamma- aminobutyric acid on the settlement of larvae of the black chiton Katharina tunicata. Mar. Biol. 72:243-247. Ryland, J.S. 197 6. Physiology and ecology of marine bryozoans. Adv. Mar. Biol. 14:285-443. Scheltema, R.S. 1961. Metamorphosis of the veliger larvae of Nassarius obsoletus (Gastropoda) in response to bottom sediment. Biol. Bull. 120:92-109.

Scheltema, R.S., I.P. Williams, M.A. Shaw, and C. Loudon. 1981. Gregarious settlement by the larvae of Hydroides dianthus (Polychaeta: Serpulidae). Mar. Ecol. Prog. Ser. 5:69-74. Searcy-Bernal, R., A.E. Salas-Garza, R.A. Flores-Aguilar, and P.R. Hinojosa-Rivera. 1992. Simultaneous comparison of methods for settlement and metamorphosis induction in the red abalone (Haliotis rufescens). Aquaculture 105:241-250. Sebens, K.P. 1983. Settlement and metamorphosis of a temperate soft-coral larva (Alcyonium siderium Verrill): induction by crustose algae. Biol. Bull. 165:286-304. Seki, T. and H. Kan-no. 1981. Induced settlement of the Japanese abalone, Haliotis discus hannai, veliger by the mucus of the juvenile and adult abalones. Bull. Tohoku Regional Fish. Res. Lab. 43:29-3 6. Sekiguchi, H., H. Saito, and H. Nakao. 1991. Spatial and temporal distributions of planktonic and benthic phases of bivalves in a tidal estuary. Nippon Bentosu Gakkaishi 40:11-21. Shanks, A.L. and W.G. Wright. 1987. Internal-wave mediated shoreward transport of cyprids, megalopae, and gammarids and correlated longshore differences in the settling rate of intertidal barnacles. J . Exp. Mar. Biol. Ecol. 114:1-13. Shaw, W.N. 1967. Seasonal fouling and oyster setting on asbestos plates in Broad Creek, Talbot County, Maryland, 1963-65. Chesapeake Sci. 8:228-236.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 326 Slattery, M. 1992. Larval settlement and juvenile survival in the red abalone (Haliotis rufescens): an examination of inductive cues and substrate selection. Aquaculture 102:143-153.

Smith, P.R. and F.-S. Chia. 1985. Larval development and metamorphosis of Sabellaria cementarium Moore, 1906. Can. J. Zool. 63:1037-1049. Spaulding, J.G.III. 1972. The life cycle of Peachia quinquecapitata, an anemone parasitic on medusae during its larval development. Biol. Bull. 143:440-453. Shpigel, M . , S.L. Coon, and P. Kleinot. 1989. Growth and survival of cultchless spat of Ostrea edulis Linnaeus, 1750 produced using epinephrine and shell chip. J. Shellf. Res. 8:355-357. Steinberg, P.D. and V.S. Kennedy. 1979. Predation upon Crassostrea virginica (Gmelin) larvae by two invertebrate species common to Chesapeake Bay oyster bars. Veliger 22:78-84. Strathmann, M.F. 1987a. Phylum Annelida, Class Polychaeta. 138-195. In Strathmann, M.F. (ed.). Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. Strathmann, M.F. 1987b. Phylum Crustacea, Class Maxillopoda, Subclass Cirripeda. 370-392. In Strathmann, M.F. (ed.). Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. Strathmann, M.F. 1987c. Phylum Echinodermata, Class Echinoida. 511-534. In Strathmann, M.F. (ed.). Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. Strathmann, M.F. and C.E. Mills. 1987. Phylum Cnidaria, Class Hydrozoa. 44-71. In Strathmann, M.F. (ed.). Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast. University of Washington Press, Seattle, WA. Strieker, S.A. 1988. Metamorphosis of the marine bryozoan Membranipora membranacea. Amer. Zool. 28:71A. (Abstract).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 327 Switzer-Dunlap, M. 1978. Larval biology and metamorphosis of aplysiid gastropods. 197-218. In Chia, F.-S., and M.E. Rice (eds.). Settlement and Metamorphosis of Marine Invertebrate Larvae. Elsevier/North Holland Biomedical Press, New York, NY. Switzer-Dunlap, M. and M.G. Hadfield. 1977. Observations on development, larval growth and metamorphosis of four species of Aplysiidae (Gastropoda: Opisthobranchia) in laboratory culture. J. Exp. Mar. Biol. Ecol. 29:245- 261. Taki, Y., Y. Ogasawara, Y. Ido, and N. Yokoyama. 1980. Color factors influencing larval settlement of barnacles, Balanus amphitrite subspp. Bull. Japan. Soc. Sci. Fish. 46:133-138. Thompson, J.E. 1985. Exudation of biologically-active metabolites in the sponge Aplysina fistularis. I. Biological evidence. Mar. Biol. 88:23-26. Thompson, T.E. 1962. Studies on the ontogeny of Tritonia hombergi Cuvier (Gastropoda: Opisthobranchia). Phil. Trans. Royal Soc. London 245:171-218. Todd, C.D., M.G. Bentley, and J.N. Havenhand. 1991. Larval metamorphosis of the opisthobranch mollusc Aldaria proxima (Gastropoda: Nudibranchia): the effects of choline and elevated potassium ion concentration. J. Mar. Biol. Assoc. U.K. 71:53-72.

Tremblay, M.J. and M. Sinclair. 1990. Diel migrations of sea scallop larvae Placopecten magellanicus in a shallow embayment. Mar. Ecol. Prog. Ser. 67:19-25. Turner, R.D. 197 6. Some factors involved in the settlement and metamorphosis of marine bivalve larvae. Sharpley, J.M. and A.M. Kaplan (eds.). Proc. 3rd Internat. Biodegradation Symp. pp. 409-416. van Duyl, F.C., R.P.M. Bak, and J. Sybesema. 1981. The ecology of the tropical compound ascidian Trididemnum solidum. I. Reproductive strategy and larval behavior. Mar. Ecol. Prog. Ser. 6:35-42. van Heukelem, W.F. and S.D. Sulkin. 1983. The behavioral basis of larval dispersal and recruitment of the blue crab Callinectes sapidus Rathbun. J. Shellfish Res. 3:103. (Abstract).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 328 Vietch, F.P. and H. Hidu. 1971. Gregarious setting in the American oyster Crassostrea virginica Gmelin: I Properties of a partially purified "setting factor". Chesapeake Sci. 12:173-178. Visscher, J.P. 1928. Reactions of the cyprid larvae of barnacles at the time of attachment. Biol. Bull. 54:327-335.

Watanabe, J.M. and L.R. Cox. 1976. Spawning behavior and larval development in Mopalia lignosa and Mopalia mucosa (Mollusca: Polyplacophora) in central California. Veliger 18 (Supplement):18-27. Williams, G.B. 1976. Aggregation during settlement as a factor in the establishment of coelenterate colonies. Ophelia 15:57-64. Wilson, D.P. 19 68. The settlement behavior of the larvae of Sabellaria alveolata (L.). J . Mar. Biol. Assoc. U.K. 48:387-435. Wisely, B. 1959. Observation on the setting behavior of larvae of the tubeworm Spirorbis borealis Daudin (Polychaeta). Aust. J. Mar. Freshwater Res. 11:55-72 + 1 plate. Wisely, B. 1969. Preferential settlement in concavities (rugophilic behavior) by larvae of the brachiopod Waltonia inconspicua (Sowerby, 184 6). New Zealand J. Mar. Freshwater Res. 3:273-80. Wood, L. and W.J. Hargis, Jr. 1971. Transport of bivalve larvae in a tidal estuary. 4th Europ. Mar. Biol. Symp.: 29-44. Woodin, S.A. 1985. Effects of defecation by arenicolid polychaete adults on spionid polychaete juveniles in field experiments: selective settlement or differential mortality. J . Exp. Mar. Biol. Ecol. 87:119-132. Yool, A.J., S.M. Grau, M.G. Hadfield, R.A. Jensen, D.A. Markell, and D.E. Morse. 1986. Excess potassium induces larval metamorphosis in four marine invertebrate species. Biol. Bull. 170:255-266. Young, C.M. and L.F. Braithwaite. 1980. Larval behavior and post-settling morphology in the ascidian, Chelyosoma productum Stimpson. J. Exp. Mar. Biol. Ecol. 42:157-169.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 329 Young, C.M. and J.L. Cameron. 1989. Differential predation by barnacles upon larvae of two bryozoans: spatial effects at small scales. J. Exp. Mar. Biol. Ecol. 128:283-294. Young, C.M. and F.-S. Chia. 1982. Factors controlling spatial distribution of the sea cucumber Psolus chitonoides: settling and post-settling behavior. Mar. Biol. 69:195-205.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX B

EFFECT OF NEUTRAL RED STAIN ON SETTLEMENT ABILITY OF OYSTER PEDIVELIGERS, Crassostrea virginica4

4 Contribution No. 1672 from the Virginia Institute of Marine Science. This appendix was published in a similar form in Journal of Shellfish Research 10:455-446. 330

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 331

Table of Contents

Page Abstract...... 332 Introduction...... 333 Materials and Methods...... 334 Results...... 336 Discussion...... 336 References...... 338 Tables...... 339

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 332

Abstract

The effect of neutral red stain on the settlement of oyster Crassostrea virginica (Gmelin) pediveligers was examined. Larvae were offered two types of substrate: oyster shell and acetate sheets. Settlement was measured as the proportion of pediveligers settled after 24 hours and analyzed with two-factor ANOVA. Staining did not significantly affect settlement, although settlement onto acetate was much lower than onto oyster shell.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 333 Introduction Studies of the early life history of bivalve mollusks often require examination and counting of larvae and newly-settled spat. Under some circumstances, this examination can be greatly facilitated by the use of a vital stain with the nearly transparent larvae or spat. It has been shown that a low concentration of neutral red stain colored oyster (Crassostrea virginica) larvae a bright red, with no apparent mortality or side effects (Loosanoff and Davis, 1947; Manzi and Donnelly, 1971).

In some cases, however, the swimming behavior of C. virginica pediveliger larvae in the laboratory was perceptibly altered for several hours after being stained with neutral red (unpublished data). Swimming rate decreased, compared to that of larvae without the stain, and there was a tendency for the stained larvae to clump. Because oyster settlement has a large behavioral component (Cranfield, 1973) , there is a concern that the use of neutral red to stain pediveliger larvae might affect their ability to settle. This article examines the effect of neutral red stain on the settlement of C. virginica eyed pediveliger larvae onto two types of substrates.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 334 Materials and Methods Crassostrea virginica pediveliger larvae were cultured in filtered York River water at 15%osalinity. Visual examination of the larvae confirmed that more than 95% had both a foot and a large, distinct eyespot, which are considered morphological characteristics of larval competency to settle (Coon et al. 1985). Manzi and Donnelly (1971) recommend culture concentrations of neutral red stain of 2.5-5.0 ppm for 24 hours in both distilled water and filtered seawater. In seawater however, neutral red stain forms a precipitate within several days, while solutions in distilled water can be kept at room temperature for months while no perceptible precipitation. Distilled water solutions were used in this experiment. The larvae were held in 15 mi culture dishes (watch glasses), filled with 10 ml of 0.2 /zm-filtered York River water at 15%osalinity. Salinities used by Loosanoff and Davis (1947) or Manzi and Donnelly (1971) were not given. Two settlement substrates were used: small, nearly flat

Crassostrea virginica valves, and circles of Mylar frosted acetate, both about 2.5 cm in diameter. The shells were scrubbed clean, the ligament was removed, and the concave surface was placed downward in the culture dish. The acetate circles were creased so that the frosted surface was concave, and the concave surface placed downwards in the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 335 culture dish. Mylar acetate has been used previously as a commercial settlement substrate (Dupuy et al., 1977). Three days prior to the experiments, the shells and acetate circles were placed in a flowing seawater trough with adult C. virginica. Dissolved substances from both oyster shells (Vietch and Hidu, 1971) and the common marine bacterium Alteromonas (Fitt et al., 1990) enhance settlement of oyster larvae. Exposure to seawater provided substrate accumulation of chemical substances which induce larval settlement. Three to four days of exposure permits optimal bacterial growth for settlement inducement (Fitt et al., 1990). The experimental design involved a two-factor analysis of variance (Zar, 1984). Factors were substrate treatment (shell versus acetate) and stain treatment (stain versus no stain), and each treatment combination had five replicates. Approximately 100 pediveliger larvae were added to each culture dish with a dropper pipette. Then three drops of the 0. I ppt neutral red stain solution were added to the stain-treatment dishes, yielding a culture stain

concentration of approximately 3 ppm. All of the culture dishes were covered with glass petri dishes, and covered with a dark cloth for 24 hours. The temperature was constant at 20°C.

At the end of 24 hours, the number of settled spat and free-swimming larvae were counted in each settlement

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 336 chamber. There were no more than one or two dead larvae per culture dish; most of these had been accidentally crushed by handling. One culture dish was lost; the missing value was estimated using the Shearer technique (Zar, 1984). Settlement was expressed as a proportion in each chamber. Prior to statistical analysis, each value was transformed by the arcsine square root method, to bring the data distribution closer to a normal distribution and satisfy the assumptions of analysis of variance (Zar, 1984).

Results Substrate type strongly affected settlement (Tables 1, 2). Mean proportional settlement on oyster shell was 64.2%, compared to only 1.4% on acetate. Staining did not significantly affect settlement (Tables 1, 2); the difference in mean proportional settlement between treatments was only 2.3%. The interaction effect was not significant (Table 2).

Discussion The hypothesis that staining with neutral red affects settlement of C. virginica larvae was not substantiated, with an extremely small (2.3%) difference between staining treatments. The difference between proportional settlement on the substrate treatments (oyster shell versus acetate) shows the degree to which settlement of oyster larvae can be affected.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 337 Although settlement was not affected by neutral red stain, there may be effects on metamorphosis (a physiological event separate from settlement) or spat survival or growth. Metamorphic success and post metamorphic survival is affected strongly by physiological stress (Baker and Mann, 1992). Manzi and Donnelly (1971) examined the effects of neutral red and other stains on larvae of C. virginica and the venerid clam Mercenaria mercenaria, and reported no difference in survival and growth rates between the two species. Mercenaria or other bivalve pediveligers were not used in this study, but based upon Manzi and Donnelly's work, it is reasonable to infer that neutral red will have no significant effect on proportional settlement on other bivalve mollusk taxa.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 338

References Baker, S.M. and R. Mann. 1992. Effects of hypoxia and anoxia on larval settlement, juvenile growth, and juvenile survival of the oyster, Crassostrea virginica. Biological Bulletin 182:265-269. Coon, S. L., D. B. Bonar and R. M. Weiner. 1985. Induction of settlement and metamorphosis of the Pacific oyster, Crassostrea gigas (Thunberg), by L-DOPA and catecholamines. J. Exp. Mar. Biol. Ecol. 94:211-221. Cranfield, H.J. 1973. Observations on the behavior of the pediveliger of Ostrea edulis during attachment and cementing. Marine Biology 22(3):203-209. Dupuy, J.L., N.T. Windsor, and C.E. Sutton. 1977. Handbook for design and operation of an oyster seed hatchery. Virginia Institute of Marine Science Special Report in Applied Marine Science and Engineering 142, 111 pp. Fitt, W.K., S.L. Coon, M. Walch, R.M. Weiner, R.R. Colwell, and D.B. Bonar. 1990. Settlement behavior and metamorphosis of oyster larvae (Crassostrea gigas) in response to bacterial supernatants. Marine Biology 106(3):389-394. Loosanoff, V.L. and H.C. Davis. 1947. Staining of oyster larvae as a method for studies of their movement and distribution. Science 106(2763):597-598. Manzi, J.J. and K. A. Donnelly. 1971. Staining large populations of bivalve larvae. Trans. Amer. Fish. Soc. 1(3):58-90. Vietch, F.P. and H. Hidu. 1971. Gregarious setting in the American oyster Crassostrea virginica Gmelin: I. Properties of a partially purified "setting factor". Chesapeake Science 12(3):173-178. Zar, J. H. 1984. Biostatistical Analysis. 2nd Ed. Prentice-Hall, Englewood Cliffs, NJ, 718 pp.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1. Settlement of Crassostrea, expressed as proportions of total larvae. Oyster Shell______Mylar Acetate_____ Stain No Stain Stain No Stain 0 . 66 0.63 0.03 0.00 0.73 0.66 0 . 00 0. 06 0.71 0.72 0 . 00 0 .01* 0.36 0.82 0.00 0.03 0.82 0.31 0 . 01 0 . 00 ★Denotes estimated va see text.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 340

Table 2. Summary of two factor analysis of variance on effects of stain and substrate on Crassostrea settlement.

Source DF SS MS F P Stain 1 0.0005 0.0005 0.022 0.887 Substrate 1 3.6607 3.660715 7.8 <0.0001 Interaction 1 0.0079 0.0079 0.341 0.586 Error 16 0.3713 0.0232 Total 19 4.0403

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX C.

REVIEW OF PLANKTONIC BIVALVE MOLLUSK POSTLARVAE AND OCCURRENCE OF POST-METAMORPHIC BIVALVES IN THE PLANKTON IN LOWER CHESAPEAKE BAY

341

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 342 Table of Contents Page Abstract...... 343 Introduction...... 344 Materials and Methods...... 345 Results...... 347 Discussion...... 352 Definition of Postlarva...... 355 Bivalve Mollusk Postlarval Modes...... 362 Ecology of PostlarvalDrift ...... 369 References...... 371 Table...... 379

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 343

Abstract Plankton samples from the York River, Chesapeake Bay, Virginia, in 1990-1993, contained post-metamorphic bivalve mollusks. Four bivalves, Anadara transversa, Geukensia demissa, Tellina agilis, and Macoma spp. (M. balthica, M. mitchelli, or a combination of both), were present as post- metamorphic juveniles, apparently drifting passively by means of byssal threads. Macoma was highly abundant and present most of the year; other species were occasionally abundant and present primarily in summer. Seven other bivalve species were identified from plankton samples, but were present only as veliger larvae. The concept of a "postlarval" phase in bivalve mollusks is defined and discussed, in relation to some other marine invertebrate groups.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 344 Introduction Most marine bivalve mollusks disperse by means of a planktonic, swimming larval stage, or veliger. The paradigm of bivalve life history has been that larvae settle, metamorphose, and directly assume a benthic mode of existence. Increasing accounts of post-metamorphic juveniles, or "postlarvae", in the plankton, however, have challenged this model for many taxa (Bayne, 1964; Williams and Porter, 1971; Beukema and de Vlas, 1989; Martel and Chia, 1991; Martel, 1993). Sigurdsson et al. (1976) showed that many taxa can use long byssal threads to passively drift in currents ("byssal-drifting") after metamorphosis and loss of the swimming organ, the velum. The new life history model for many species includes a post-metamorphic, or postlarval drifting phase, followed by secondary settlement (Bayne, 1964). Following the discussion of the results in this manuscript, the concept of a postlarva in select marine invertebrate groups will be discussed.

As a consequence of extensive plankton sampling in 1990, 1991, and 1992 in Chesapeake Bay, Virginia, numerous post-metamorphic bivalves, drifting by means of byssal threads, were observed for several species. For some of these species, byssal drifting had not previously been reported. Species collected, and observations on them, are described here. .The term "postlarva" will be used hereafter

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 345 to refer to post-metamorphic bivalves in the plankton; refer to the discussion following Results for a full definition.

Materials and Methods The sampling site was in 3 m of water, off the end of a pier in the lower York River, at Gloucester Point, Virginia. In 1990, a single pump was used near the bottom of the water column, sampling four times daily from July 5 to August 9. In 1991 and 1992, a series of three pumps were used, near the bottom, near the surface, and midwater. Near-bottom samples were about 20 cm above the substrate, and did not entrain bottom sediments. Sampling in 1991 was primarily from July 17 to August 5, sampling for 12 hours at several- day intervals; and in 1992 from June 29 to September 7, sampling once or twice daily. Occasional other samples were collected between April 1990 and October 1993. Volumes sampled were typically 1-3 m3.

Larvae and postlarvae were retained on a 150 /zm mesh, and identified under a dissecting microscope in the

laboratory. In 1990 and 1991, a 400 lira mesh "pre-screen" was used to keep out large debris, but in 1992 the pre­ screen was not used. When the pre-screen was used, it was examined by eye for the presence of large bivalve postlarvae retained on the pre-screen. The following references were used in aid of identification of larvae and postlarvae: Jorgensen (1946); Sullivan (1948); Rees (1950); Loosanoff et

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 346 al. (1966); Chanley and Andrews (1971); Culliney (1975); Webb (1986); Fuller and Lutz (1989); and Kennedy et al. (1989). In some cases, larvae and postlarvae were cultured to juveniles, using techniques modified after Loosanoff and Davis (1963), Kennedy et al. (1989), and Gustafson et al. (1991), allowing more positive identification. The presence of shell growth rings past the prodissoconch (larval shell), and shell outline were normally sufficient to distinguish postlarvae from late-stage veliger (pediveliger) larvae. Byssal threads could not readily be seen under light microscopy, but their presence could be detected by dragging a fine probe gently through the water in a circle around a suspected postlarva, without touching the shell. If a byssus was present, the probe would catch it and the attached bivalve. Sigurdsson et al. (197 6) used Alcian blue (selective for acid mucopolysaccharides) to stain byssal threads, but this effect could not be duplicated by this researcher.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 347

Results The following species of larvae and postlarvae were identified: the ark clam Anadara transversa (Say, 1822) (Arcidae); the mussels Geukensia demissa (Dillwyn, 1817) and Mytilus edulis Linne, 1758 (Mytilidae); the jingle shell Anomia simplex Orbigny, 1842 (Anomiidae); the American oyster Crassostrea virginica (Gmelin, 1791) (Ostreidae); the jackknife clam Ensis directus Conrad, 1843 (Solenidae); the tellin Tellina agilis Stimpson, 1857 and tellinid clams of the genus Macoma (Tellinidae); the coot clam Mulinia lateralis (Say, 1822) (Mactridae); the angel wing clam Cyrtopleura costata (Linne, 17 58) (Pholadidae); and the shipworm Bankia gouldi Bartsch, 1908 (Teredinidae). The Macoma were probably a mixture of M . balthica (Linne, 1758) and M . mitchelli Dali, 1895. Both were present as adults at the sampling site and abundant in adjacent tidal creeks; M. balthica adults were about twice as common as M . mitchelli in benthic surveys. Both species recruited in the winters into seawater flumes connected by flowing, unfiltered seawater, to the York River. Illustrations of larvae and postlarvae of M . mitchelli (Kennedy et al., 1989) closely resemble those of M. balthica (Chanley and Andrews, 1971), and hinge characteristics of M. balthica have not been published. Macoma individuals could not be reared to juveniles in the laboratory for further identification. Anadara, Geukensia, Mulinia, and Cyrtopleura were positively

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 348 identified by rearing specimens in the laboratory to juveniles, at which point they could be distinguished from closely related, sympatric taxa. Other bivalve species were noted in the plankton, but could not be positively identified. Mean daily salinities were 18.1-20.4%oin 1990, 19.9- 21.4%oin 1991, and 19.1-2 3.6%oin 1992, during summer samples. Mean daily temperatures were 26.5-28.9°C in 1990,

26.5-29.3°C in 1991, and 24.1-27.7°C in 1992, during summer

samples, with highest temperatures in late July and early August. Postlarvae were clearly identified, by shell growth and morphology, to the following taxa: the ark clam Anadara transversa; the mussel Geukensia demissa; and the tellinid clams Tellina agilis and Macoma spp. All other bivalves observed in the plankton were veliger larvae. Anadara was rare to absent except in early July, 1992, where it attained a level of over 4 individuals per m3. In

July 1992 most individuals were veliger larvae, but in August 1992 and July of 1990 and 1991, most individuals were postlarvae. Larvae attained a shell length of about 300 jzm, and postlarvae attained a shell length of about 450 jim. Pediveliger larvae had an eyespot and a foot; postlarvae retained the foot but lost the eyespot. By late September 1992, when large numbers of newly recruited benthic juveniles were found attached to oyster shells collected by

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 349 a diver, Anadara was absent from the plankton as either larvae or postlarvae. Juveniles in the field were observed attached directly to shells, primarily of the oyster Crassostrea virginica. Individuals in the laboratory initially attached to filamentous substrates such as dead hydroid tubes (Halocordyle distichus), or byssus threads of the mussel Geukensia demissa. Geukensia was present throughout the summers of 1990 to 1992, but rarely exceeded 2 individuals per m3. About 90% of specimens observed were veliger larvae, with a shell

length of about 250 nm. The remainder were postlarvae, with

a shell length rarely exceeding 350 /urn. Both pediveliger larvae and postlarvae had a well-developed foot and eyespot (although eyespot was relatively smaller than that of Crassostrea pediveligers). No larvae or postlarvae remained in the plankton by late August 1990 or 1991, or late September 1992.

Tellina was rare to absent except in early July, 1990, when abundances reached more than 3 0 individuals per m3. The prodissoconch of this genus contains visible growth rings, which complicates differentiation of larvae from postlarvae. At least half of the specimens observed appeared to lack a velum, and had a well-developed, broadly triangular foot, and a byssus. These were about 300-350 /xm in shell length. Of the remainder, some as small as 250 jum, several were observed to have a velum, but the status of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 350 rest could not be determined satisfactorily. No Tellina of any size had a pigmented spot resembling an eyespot. Macoma was the most abundant bivalve larva or postlarva during summer and autumn, and in October 1990 abundances exceeded 1000 per m3, although in summer abundances were

typically closer to 100 postlarvae per m3. In summer,

Macoma larvae and postlarvae were observed together, but in October 1990 and other winter samples, only postlarval Macoma were observed. Postlarval Macoma were present but rare (1-2 per m3) in February and March 1992. Veliger

larvae as large as 400 jam in shell length were observed, but most Macoma in the plankton larger than 330 jam in shell length were postlarvae. The largest postlarvae observed were about 450 jam in shell length; most were closer to 400 jam. This maximum size was not affected by the presence or absence of the 400 /im mesh used to keep large debris off the 150 /am mesh. It was often difficult to distinguish large pediveligers from postlarvae, but postlarvae had a large, broadly triangular foot and a byssus. Neither pediveligers nor postlarvae had pigmented spots resembling eyespots. In this study, attempts to induce juvenile growth of Macoma in the laboratory were unsuccessful, using methods described by Kennedy et al. (1989). Postlarvae survived up to several weeks in conditions that were successful in inducing metamorphosis and growth in other infaunal species (Mulinia, Cyrtopleura), and remained active, but did not

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 351 grow. However, flow-through, ambient York River water flumes, with 1-2 cm of sediment, experienced recruitment of both Macoma balthica and M. mitchelli between November and February, in all years of this study. Of the remaining bivalves, Mytilus and Ensis were observed only in March 1992, and all specimens were veliger larvae. Water temperatures at this time were about 10° C, and salinity was about 18%o. Mulinia was observed only in April 1990, and all specimens were veligers. Water temperatures at this time were about 15° C, and salinity was

about 18% o . The remaining species were observed only between May and September. Cyrtopleura was the most abundant bivalve after Macoma, attaining a peak abundance of about 3800 late-stage larvae per m3 in August 1992, although

typical summer abundances were 10-100 per m3. Evidence of postlarval growth or byssal threads were not seen in any Cyrtopleura. Crassostrea and Bankia were both present throughout summers in all years, and reached peaks of about

140 and 90 per m3, respectively, although normal levels were

under 10 per m3. All specimens were veliger larvae.

Anomia was observed as rare veliger larvae in July of 1990 and 1992.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 352 Discussion This is the first record for planktonic postlarvae for Anadara or Geukensia. Planktonic postlarvae of Tellina have not been reported as such, but illustrations of Tellina species by Sullivan (1948) and Rees (1950) resemble postlarvae observed in this study, and Webb (1986) illustrated non-planktonic postlarval Tellina. Geukensia postlarvae have been illustrated by Fuller and Lutz (1989), but these were reared in the laboratory, rather than collected in the plankton. The species identification of Macoma postlarvae was uncertain. Two species are sympatric at the sampling site; Macoma balthica, and M. mitchelli. Blundon and Kennedy (1982) observed juveniles ("spat") of M . balthica primarily in the spring and juveniles of M . mitchelli throughout the year, in the Choptank River in northern Chesapeake Bay; however, Shaw (1965) reported two annual recruitment periods, in spring and fall, for M. balthica in the same region. At the York River site of this study, both species were present in late winter samples taken from ambient seawater flumes.

The biological costs and benefits of postlarval drifting remain unknown. In Chesapeake Bay, postlarval drifting in Macoma seems to form an integral part of a prolonged juvenile stage, in a manner similar to that described by Beukema and de Vlas (1989) for M . balthica in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 353 Europe, although unlike their European conspecifics (Beukema and de Vlas, 1989), no specimens larger than about 450 fim in

total shell length were observed. This was not an artifact of the sampling design, since the maximum size observed was not affected by the presence or absence of a 400 jum "pre- screen". Beukema (pers. comm.) has suggested that larger sizes of Macoma were not observed because of their relative scarcity. Many thousands of postlarvae of this species were collected, however, so if larger sizes are present but scarce, their ecological significance in Chesapeake Bay is probably slight. Successive dominance of first larvae and then postlarvae in the plankton by Anadara transversa in 1992 suggests that for this species, also, postlarval drifting is an integral part of the life cycle, although the postlarval period is weeks rather than months. This may be similar to the life cycle observed for Mytilus edulis in Maine by Newell et a l . (1991), in which larvae recruit to seagrass, and then as postlarvae transfer to the adult habitat. It is not known, however, whether Anadara settles before metamorphosing. Pediveliger larvae produce byssal threads (e.g. Lane and Nott, 1975), so it is possible that some species begin byssal drifting prior to metamorphosis, and metamorphose in the water column. Nelson (1925, p. 286) reported that larvae of Mytilus edulis could maintain

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 354 themselves in the water column while they metamorphosed, but the evidence for this was not made clear.

Planktonic postlarvae were not observed for some species for which they were expected. Planktonic postlarvae have been reported extensively elsewhere for Mytilus edulis (e.g. Bayne, 1964; de Blok and Tan-Maas, 1977; Board, 1983; Lane et al., 1985), but were not observed in the York River. Byssal drifting in postlarvae has been reported in the genus Ensis (Sigurdsson et al., 1976) and Williams and Porter (1971) found juveniles of this genus commonly in plankton samples, although they attributed this to swimming, not drifting. The relatively few Ensis seen in this study in the York River were all veliger larvae, however. Juvenile Anomia simplex possess a byssus (Fuller et al., 1989), and byssal drifting in postlarvae was reported in the related genus Heteranomia (Sigurdsson et al., 1976), but the very few specimens observed in the York River were all veliger larvae. It is possible that postlarval drifting of Mytilus, Ensis, and Anomia, occurs in Chesapeake Bay, despite not having been observed by this researcher. The oyster Crassostrea virginica lacks a planktonic postlarval stage, and permanently cements itself to the substrate immediately upon settlement (see Galtsoff, 1964, for a description of the life history). No planktonic postlarvae or byssal threads have been reported for any members of the families Pholadidae, of which Cyrtopleura

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 355 costata is a member, or Teredinidae, of which Bankia gouldi is a member. No evidence for planktonic postlarvae in either Cyrtopleura or Bankia was noted in this study, despite large numbers of animals observed. It is possible that this phenomena does not exist in these taxa.

Definition of Postlarva The term "postlarva" has come into increasing use in the literature to refer to a phase in the life cycle of an organism that is morphologically, behaviorally, or ecologically distinct from both the preceding larval phase and the subsequent juvenile phase. It has been used by researchers on a variety of taxa, and in several ways; use of the term in reference to three select invertebrate groups (decapod crustaceans, echinoderms, and bivalve mollusks) will be discussed here. The adjective "postlarval" is often applied to characteristics of invertebrates following the larval phase, with no intention of defining a specific life history stage; this latter use of the term will not be discussed. In crustacean zoology, the term "postlarva" is often used to refer to the morphologically distinct intermediate phase between the final instar zoea larva and the first instar juvenile. This may also be called the glaucothoe, mastigopus, megalopa, parva, or puerulus, depending on the taxon (see Gurney, 1942; Williamson, 1982; Felder et al.,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 356 1985; and Barnes, 1986, for general descriptions and terminology). The term "postlarva" is used widely by the aquaculture industry to refer to the parva (Caridea) or mastigopus (Penaeidea) stage of hatchery-reared shrimp, when they are sold to grow-out operations (e.g. West et al., 1983; Chiang and Liao, 1986). Among decapod crustacean ecologists, "postlarva" is used to distinguish the pre-juvenile instar (see above), which exhibits settlement behavior, from the preceding zoeal larval stages, which are strictly planktonic. Robertson (1968), Lyons (1970, 1980), Phillips and Sastry (1980), Phillips (1981), Lipcius et al. (1990), and Benfield and Aldrich (1992), offer examples of postlarval behavior and ecology among decapod crustaceans. In some species, the postlarva appears to be predominantly planktonic (e.g. Robertson, 1968), while in others it is predominantly nectonic (e.g. Phillips, 1981), but in all of the above taxa, the so-called "postlarva" a) disperses in the water column at least part of the time, and b) is followed by metamorphosis into a distinct juvenile instar. Lyons (1970) attempted to clearly define the ecological concept of the decapod crustacean postlarva; subsequent use of the term has been reviewed by Felder et al. (1985). Echinoderms, like crustaceans, have well-defined and taxonomically diverse larval stages. Barnes (1986) gives general descriptions and terminology. The "postlarva" of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 357 echinoderms is not a morphologically distinct stage, but resembles the later juvenile. The use of the term "postlarva" is not consistent within the study of echinoderm ecology. Harvey and Gage (1984) describe the "postlarva" of a burrowing sea urchin, Portalesia; the postlarva resembles the juvenile, but does not yet burrow, as do later juveniles. In this case, therefore, the postlarva would seem to be an behaviorally and ecologically defined phase. In Gage and Tyler (1985), however, the term "postlarva" is used to refer to small post-metamorphic urchins, Echinus, that were not shown to differ significantly from larger juveniles in behavior, habitat, or morphology. Hindler (1975) used the term "postlarva" to refer to both planktonic and benthic post-metamorphic phases of different species of ophiuroids (brittle stars). The post-metamorphic stages described often had fewer arms than the juveniles and adults, but the transition from benthic "postlarva" to juvenile was apparently continuous. Ebert (1975) used the term "post-larval" to refer to all sea urchins stages past metamorphosis from the final larval stage, or echinopluteus. There may be stages intermediate between larvae and juveniles among some echinoderms for which the term "postlarva" is appropriate, but there is no apparent agreement within the echinoderm literature on this. It is also apparent that if the term "postlarva" is applied to some echinoderms, the precise definition, in terms of a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 358 morphological and behavioral stage, will differ somewhat from the definition for decapod crustaceans. In terms of the concept of a postlarva, the study of bivalve mollusks is intermediate between that of decapod crustaceans and echinoderms. The existence of a behaviorally intermediate stage between the late larvae stage (pediveliger) and early juvenile was clearly shown for the hard clam, Mercenaria mercenaria, by Carriker (1961). This stage, called the "plantigrade", was epifaunal and highly motile, as opposed to the infaunal and sedentary juvenile. Prior to that, Nelson (1928) had observed post- metamorphic mussels, Mytilus edulis, in the plankton, but had simply used the malacological term "dissoconch" (referring to shell growth beyond the larval shell) to describe them. Bayne (1964) was the first to clearly show that post-metamorphic M . edulis re-entered the plankton on a population-wide scale, using long byssal threads. Since then, byssal drifting has been demonstrated for many bivalve taxa (see following discussion). The term "postlarva" has been used to refer to this phase by de Blok and Tan-Maas

(1977), Board (1983), Lane et al. (1985), Beukema and de Vlas (1989), and Armonies and Hellwig-Armonies (1992). Lane et al. (1985) also equated the postlarva with the plantigrade. In all of the above cases, the term

"plantigrade or "postlarvae" refers to a phase a) defined by unique behavior, but b) morphologically similar to the later

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 359 juvenile. In addition, at least in the case of byssal- drifting, the postlarval phase is not obligatory. The use of the term "postlarva" is not consistent within the bivalve mollusk literature, however. Campos and Ramorino (1981), Ramorino and Campos (1983), Lucas et al. (1986), Fuller and Lutz (1989), Thouzeau (1991), and Goodsell et al. (1992), are examples of references that have used the term "postlarva" to refer to post-metamorphic bivalves, but did not clearly define a difference between these and juveniles. Armonies and Hellwig-Armonies (1992), and Beaumont and Barnes (1992), in describing post- metamorphic drifting in two scallops species, have used the much older, more general term "spat", which lacks any specific morphological or behavioral connotations. Thus, while the concept of a behavioral or ecological phase is well developed in the bivalve mollusk literature, the use of the term "postlarva" remains inconsistent.

The protobranch bivalve Solemya velum, which has lethicotrophic development through a pericalymma larva, metamorphoses within an egg capsule, and hatches as a benthic juvenile. In describing this development, Gustafson and Lutz (1992) used the term "postlarva" to refer to the post-metamorphic individuals which had not yet hatched. Although they were morphologically similar to the juvenile, until they hatched, the "postlarvae" were functionally similar to the non-motile egg. Williams and Porter (1971),

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 360 however, reported high numbers of post-hatching S. velum in the plankton, so there may be a planktonic postlarval phase as well. If the planktonic post-hatching Solemya is considered a postlarva, (e.g. de Blok and Tan-Maas, 1977, Board 1983, Lane et al., 1985; Beukema and de Vlas, 1989, and Armonies and Hellwig-Armonies; 1992), then the term should not also be applied to the pre-hatching period, without some explanation or modifier. Several other possible bases for the definition of a postlarva have been presented in the bivalve mollusk literature, such as conchological features. The jingle shell, Anomia simplex, has a distinctive byssal foramen, or byssal gape, in the right valve as a juvenile and adult, but not as a larva. Fuller et al. (1989) have used the term "postlarva" to refer to the morphologically intermediate phase following development. Prezant (1990) also found shell differences between the "postlarva" and juvenile of the brooding bivalve Lissarca notorcadensis (Philobryidae). Le Pennec and Jiingbluth (1983) have found that there is an intermediate, "primitive" ligament which arises following metamorphosis from the larva, but is lost by the time the shell is 2 mm in length. Several researchers have shown a gradual, if continuous evolution of the hinge teeth in post- metamorphic bivalves (e.g. Lutz et al.,1982; Le Pennec and Yankson, 1985; Fuller and Lutz, 1989; Kennedy et al., 1989). More research will no doubt reveal conchological features

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 361 unique to the early post-metamorphic period in bivalve mollusks. It should be noted that a conchological definition of a postlarval phase does not exclude a behavioral or ecological definition. Anatomy or histology of the post-metamorphic period may also differ significantly from the later juvenile. Lucas (1975) defined the postlarval phase of various bivalve taxa on the basis of gonadal development. In the postlarva, the undifferentiated gonia, or primary germinal cells, are located in the pericardial region, but in the juvenile, they have begun to differentiate and migrate to the pedal region. As Lucas (1975) points out, however, this process is difficult to detect. Baker and Mann (in press) found that the pigmented eyespot of metamorphosed oysters, Crassostrea virginica, lingered following metamorphosis, but disappeared in the juvenile, and other anatomical features following metamorphosis (complete loss of the larval velum and assumption of gill activity) only gradually took on the juvenile form. As for conchological features, anatomical configurations unique to the post-metamorphic phase may only require further examination. Again, a sexual or anatomical definition of a postlarval phase does not exclude a behavioral or ecological definition. The following definition is hereby proposed for bivalve molluscan ecology. The "postlarva" of a bivalve mollusk is the life history phase immediately following metamorphosis

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 362 from the pediveliger larva, in which the behavior or habitat preference differs significantly from that of subsequent juvenile and adults phases. This difference may be absolute (e.g. planktonic drifting versus planktonic swimming or a benthic existence), or it may be relative (e.g. a largely epifaunal, very active mode versus a largely infaunal, largely sedentary mode). In the latter case, the definition given above must be qualified by that statement that, among mollusks, most aspects of the biology are on a continuum, and the division between postlarva and juvenile, especially, may be blurred in some cases. The modal state of a postlarva, however, should be readily discernable from the modal state of a juvenile. Although characteristics like distinctive morphology or gonadal development may also be present, the above definition of a postlarva is not dependent upon them. By the above definition of a postlarva, some bivalve mollusks lack a postlarval phase. Examples include oysters (Ostreidae), which permanently cement to the substrate following settlement (Galtsoff, 1964), and shipworms (Teredinidae), which bore immediately into wood following settlement and metamorphosis (Culliney, 1975).

Bivalve Mollusk Postlarval Modes Using the definition for a postlarva outlined above, there are two primary postlarval modes among bivalve

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 363 mollusks: benthic postlarvae, or plantigrades, and planktonic, byssal-drifting postlarvae. Although it is not known whether these modes are functionally separate (i.e., can an individual alternate modes at will?), reports in the literature are primarily of one mode or the other, and they will be discussed separately below. One of the first clear discussions of a bivalve mollusk life history phase intermediate between the larva and the juvenile was provided by Carriker (1961) , when he described the "plantigrade", or "plantigrade juvenile" postlarval phase of the clam Mercenaria mercenaria. The plantigrade in this species crawls actively, is epifaunal, and briefly attaches to substrates with a byssus. At a shell size of several millimeters, the juvenile becomes infaunal and sedentary, and the byssus gland eventually atrophies. Adults are not completely sessile, but are very sedentary (Stanley, 1985). A similar life history has been shown for the clam Mya arenaria (reviewed by Baker and Mann, 1991). The plantigrade postlarva is also a common feature among epifaunal bivalve mollusks that attach with a byssus as juveniles or adults, including ark clams, Arcidae (this manuscript), mussels, Mytilidae (e.g. Lane et al., 1982; Trevelyan and Chang, 198 3), pearl oysters, Pteriidae (e.g. Alagarswami et al., 1983), and scallops, Pectinidae (e.g Fay et al., 1983; Eckman, 1987). The plantigrade phase in the jingle shell, Anomia simplex (Anomiidae), is apparently non-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 364 obligatory (Loosanoff, 1961). Anomiids and scallops represent extremes in adult mobility; most adult anomiids are permanently attached by calcified byssal complexes, while most adult scallops are free-swimming, but both have plantigrade postlarval stages. A plantigrade phase is absent in oysters, which cement the shell to the substrate immediately upon settlement (Prytherch, 1934; Cranfield, 1973b), but is present in jewel box clams (Chamidae), which attach first with a byssus, then cement permanently to the substrate (LaBarbera and Chanley, 1971). Planktonic postlarvae have received increasing attention among bivalve molluscan ecologists, as it has become apparent that, for many species, it is important on a population scale. Bayne (19 64) was among the first to show this, for the mussel Mytilus edulis, although Nelson (1928) had previously observed M. edulis postlarvae in the plankton. Sigurdsson et al. (1976) demonstrated postlarval drifting ability in a wide variety of bivalve taxa. That paper also proposed the drifting mechanism currently favored by most researchers; that postlarvae produce elongated byssal threads which provide passive lift in water currents. This mechanism is functionally similar to "ballooning" by juvenile spiders (Araenea) in air currents (see Foelix, 1982; Decae, 1987; for reviews).

The majority of bivalve mollusks have byssal glands, at least in the pediveliger larval phase. Waite (1983) has

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. provided a general review of byssal threads and attachment in bivalve mollusks, and the larval byssal glands have been described for a number of species (Cranfield, 1973a; Gruffydd et al., 1975, Lane and Nott, 1975; Lane et al., 1982; Yankson, 1986). Lane et al. (1982) present evidence that in Mytilus edulis, there are separate glands for the production of byssal threads during "bysso-pelagic migration". Lane et al. (1985) described the byssal drifting threads M. edulis, which appear to be morphologically distinct from the attachment byssus. The chemical composition of this has not been determined; the adult byssus is composed primarily of polyphenolic proteins (Waite, 1983), but Sigurdsson et al. (1976) reported staining postlarval threads with Alcian blue, selective for acid mucopolysaccharides. Small gastropod mollusks use mucous strands to drift in the plankton (see reviews by Martel and Chia, 1991a, 1991b), and Beukema and de Vlas (1989) reported that the clam Macoma balthica (Tellinidae) used hyaline mucous threads for byssal drifting. Prezant and Chalermwat (1984) demonstrated that the drifting threads of juveniles of the freshwater clam Corbicula fluminea (Corbiculidae) were mucous threads produced by the gills. It is thus possible that the drifting threads of at least some bivalve postlarvae are muciferous rather than proteinaceous. Researchers have continued to refer to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 366 drifting threads as byssal in nature, however, and this convention will be used here. One difficulty for researchers studying postlarvae of bivalve mollusks is that the byssus or mucous threads, whether for benthic attachment or planktonic drifting, are not normally visible under light microscopy, even under high magnification (e.g. Cranfield, 1979b; Lane et al., 1985; Beukema and de Vlas, 1989; Beaumont and Barnes, 1992). Postlarvae are also able to release drifting threads at will (Sigurdsson et al., 1976; Lane et al., 1985). It has thus been difficult for researchers to verify that drifting postlarvae are indeed attached to byssal threads. Alcian blue stain is reported by Sigurdsson et al. (1976) to stain byssal drifting threads, although apparently no researcher since has duplicated that technique. Prezant and Chalermwat (1984) used carbon black to reveal the mucous threads of Corbicula. Lane et al. (1985) used transmission electron microscopy to view Mytilus byssal threads, but other researchers have relied upon the ability of postlarvae to

cling to vertical surfaces even when the water is drained, to detect byssus production (e.g. Castagna and Kraeuter, 1981; Yankson, 1986), or have dragged the postlarva via the unseen thread, using a fine probe (e.g. Beaumont and Barnes, 1992) . As a consequence of the difficulty in observing postlarval byssal threads, some researchers have not

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 367 reported byssal threads as a dispersal mechanism where they probably did occur. Nelson (1928) and Bayne (1964) both reported postlarval Mytilus edulis in the plankton, but did not recognize trailing byssal threads, which has subsequently been shown to the major postlarval dispersal mechanism for this species (Lane et al., 1982, 1985). Baggerman (1953), Williams and Porter (1971), and Grigor'eva and Regulev (1992) found planktonic postlarvae of a variety of species, but proposed other dispersal mechanisms. Subsequent evidence has shown that at least some of the species or closely related taxa they observed do enter the plankton by means of byssal threads (e.g. Frenkiel and Moueza, 1979; Yankson, 1986; Beukema and de Vlas, 1989; Beaumont and Barnes, 1992). Some juvenile and adult bivalve mollusks are also apparently able to drift passively in the plankton. Very small taxa, such as Musculus (Mytilidae), Lasaea (Lasaeidae), and Transenella tantilla (Veneridae), which as adults do not exceed several millimeters in shell length, are able to enter the plankton via byssal threads (Martel and Chia, 1991a). All of these taxa lack a free-living larval phase, so planktonic dispersal can only occur by this means. It can be argued that these taxa are neotenous, because the adults mimic the behavior of postlarvae of other taxa. The concept of neoteny has been discussed for another small brooding bivalve, Turtonia minuta (Turtoniidae), by

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 368 Ockelmann (1964) , which has subsequently been shown to be able to drift via byssal threads (Sigurdsson et al., 1976). Sellmer (1967) observed byssal production in Gemma gemma (Veneridae), another small brooding bivalve, and collected juveniles from the water column. Buoyancy by trapped gas has also been reported as a passive dispersal means in some bivalves. Nelson (1928) observed this in postlarval Mytilus edulis, and was able to demonstrate that it was not an artifact of sudden decrease in pressure caused by collecting methods. This or similar behavior has also been observed in Macoma balthica up to 14 mm in shell length (Sorlin, 1988). In the Macoma, however, floating by the means of trapped gas occurs in a distinctly juvenile or even small adult phase, rather than in a postlarval phase, while in Mytilus, subsequent researchers have found byssal drifting to be the primary means of postlarval drift (Lane et al., 1985). Some normally sedentary bivalve mollusks, aside from the well-known scallops (Pectinidae) and file shells (Limidae) (e.g Morton, 1964), are able to swim for short distances. The best known examples are jackknife clams (Solenidae). Williams and Porter (1971) reviewed swimming in non-pectinid bivalve mollusks, and found juvenile Ensis directus and Solen viridis in the water column, as well as several other taxa. Swimming, however, even in a relatively poor swimmer like Ensis, is functionally different from

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 369 passive drifting, and it clearly occurs during the juvenile or adult phases, not during the postlarval phase. Table 1 summarizes the bivalve mollusk taxa in which postlarval drifting by byssal threads has been observed.

Ecology of Postlarval Drift The byssal gland of most bivalve larvae is well- developed prior to metamorphosis (Cranfield, 1973a; Gruffydd et al., 1975, Lane and Nott, 1975; Lane et al., 1982; Yankson, 1986) , so it is possible that metamorphosis can occur in the plankton, for some species. Byssal drifting may act as merely a mechanism to move juveniles from a primary, "nursery" site to a final settlement site (Newell et al., 1991), or it can last for a period of months (Board, 1983; Beukema and de Vlas, 1989). Following this period, the postlarva may reattach briefly before drifting again (de Blok and Tan-Maas, 1977), but eventually select a final site for what has been termed "secondary settlement" (Bayne, 1964). The function of a prolonged byssal drifting phase for

bivalve mollusks that already have a planktonic larval phase is unclear. Beukema and de Vlas (1989) and Newell et al. (1991) have suggested that primary settlement requirements are different than requirements for growing juveniles, and that therefore the drifting postlarval phase is needed to transport the successfully metamorphosed postlarvae. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 370 role of the environmental factors that would cause this are only hypothetical, and have not been tested. If this model is correct, however, then the role of the drifting postlarva is not so much dispersal (in the sense of colonizing new habitats and gene flow) as it is "migration" to a preferred locale (e.g. Armonies and Hellwig-Armonies, 1992). For those bivalves which lack any free swimming larval stage, however, postlarval drifting is the primary form of dispersal (Martel and Chia, 1991a). Mortality factors for postlarval drifting bivalves have not been studied. Macoma can attain 2 mm while drifting (Beukema and de Vlas, 1989), which may put it in a different class of planktonic predators than veliger larvae (maximum of about 3 50 n). Long-term byssal drifting in Macoma occurs during the winter months, while larval development occurs in the summer and early fall (Beukema and de Vlas, 1989). Environmental pressures and the abundance and taxonomic composition of planktonic predators would be different between these two periods. During secondary settlement, however, the juveniles are still be vulnerable to benthic planktivores (e.g. Elmgren et al., 1986; Ejdung and Bonsdorff, 1992).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 371 References Alagarswami, K., 3. Dharmaraj, T.S. Velyudhan, A. Chellam, A.C.C. Victor, and A.D. Gandhi. 1983. Larval rearing and production of spat of pearl oyster Pinctada fucata (Gould). Aquaculture 34:287-301. Armonies, W. 1992. Migratory rhythms of drifting juvenile mollusks in the tidal waters of the Wadden Sea. Mar. Ecol. Prog. Ser. 83:197-206. Armonies, W. and M. Hellwig-Armonies. 1992. Passive settlement of Macoma balthica spat on tidal flats of the Wadden Sea and subsequent migration of juveniles. Neth. J. Sea. Res. 29:371-378. Baggerman, B. 1953. Spatfall and transport of Cardium edule L. Arch. Neer. Zool. 10:315-342. Baker, P. and R. Mann. 1991. Soft shell clam, Mya arenaria. 4.1-4.18. In Habitat Requirements for Chesapeake Bay Living Resources. Living Resources Subcommittee, Chesapeake Bay Program. Baker, S.M. and R. Mann. In press. Description of metamorphic phases in the oyster, Crassostrea virginica, and effects of low oxygen on metamorphosis. Mar. Ecol. Prog. Ser. Barnes, R.D. 1986. Invertebrate Zoology. 5th ed. W.B. Saunders Co., Philadelphia, PA. 893 pp. Bayne, B.L. 1964. Primary and secondary settlement in Mytilus edulis L. (Mollusca). J. Animal Ecology 33(3):513-523.

Beaumont, A.R. and D.A. Barnes. 1992. Aspects of veliger larval growth and byssus drifting of the spat of Pecten maximus and Aequipecten (Chlamys) opercularis. ICES J. Mar. Sci. 49:417-423. Benfield, M.C. and D.V. Aldrich. 1992. Attraction of postlarval Penaeus aztecus. Ives and P. setiferus (L.) (Crustacea: Decapoda: Penaeidae) to estuarine water in a laminar-flow choice chamber. J. Exp. Mar. Biol. Ecol. 156:39-52. Blundon, J.A. and V.S. Kennedy. 1982. Refuges for infaunal bivalves from blue crab, Callinectes sapidus (Rathbun) predation in Chesapeake Bay. J. Exp. Mar. Biol. Ecol. 65:67-82.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 372 Board, P. 1983. The settlement of post larval Mytilus edulis. J. Moil. Stud. 49(l):53-60. Campos, B. and L. Ramorino. 1981. Huevo, larvas y postlarva de Entodesma cuneata (Gray, 1928) (Bivalvia: Pandoracea: Lyonsiidae). Rev. Biol. Mar. Inst. Oceanol. Univ. Valparaiso 17:229-251. Castagna, M. and J.N. Kraeuter. 1981. Manual for growing the hard clam, Mercenaria mercenaria. Virginia Institute of Marine Science Spec. Rep. Applied Mar. Sci. Ocean Engineering 249. 110 pp. Chanley, P.E. and J.D. Andrews. 1971. Aids for identification of bivalve larvae of Virginia. Malacologia 11 (1):45-119. Chiang, P. and I.e. Liao. 198 6. The practice of grass prawn (Penaeus monodon) culture in Taiwan from 1968 to 1984. J. World Aquaculture Soc. 16:297-315. Cranfield, H.J. 1973a. A study of the morphology, ultrastructure, and histochemistry of the foot of the pediveliger of Ostrea edulis. Marine Biology 22:187- 202 . Cranfield, H.J. 1973b. Observations on the behaviour of the pediveliger of Ostrea edulis during attachment and settlement. Mar. Biol. 22:203-209. Culliney, J.L. 1975. Comparative larval development of the shipworms Bankia gouldi and Teredo navalis. Mar. Biol. 29(3):245-251. de Blok, J.W. and M. Tan-Maas. 1977. Function of byssal threads in young postlarval Mytilus. Nature 267(5611):558. Decae, A.E. 1987. Dispersal: ballooning and other mechanisms. 348-356. In Nentwig, W. (ed.). Ecophysiology of Spiders. Springer-Verlag, Berlin, Germany. Ebert, T.A. 1975. Growth and mortality of post-larval echinoids. Am. Zool. 15:755-775. Ejdung, G. and E. Bonsdorff. 1992. Predation on the bivalve Macoma balthica by the isopod Saduria entomon: Laboratory and field experiments. Mar. Ecol. Prog. Ser. 88:207-214.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 373 Elmgren R., S. Ankar, B. Marteleur, and G. Ejdung. 1986. Adult interference with postlarvae in soft sediments: the Pontoporella-Macoma example. Ecology 67:827-836. Fay, C.W., R.J. Neves, and G.B. Pardue. 1983. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (Mid-Atlantic) — bay scallop. U.S. Fish Wildl. Serv. Biol. Rep. FWS/OBS- 82/11.12. 17 pp. Felder, D.L., J.W. Martin, and J.W. Goy. 1985. Patterns in early postlarval development of decapods. 163-225. In Wenner, A.M. (ed.). Crustacean Issues: Vol. 2. Larval Growth. A.A. Balkema, Rotterdam, Netherlands. Foelix, R.F. 1982. Biology of Spiders. Harvard University Press, Cambridge, MA. 3 06 pp. Frenkiel, L. and M. Moueza. 1979. Developpement larvaire de deux Tellinacea, Scrobicularia plana (Semelidae) et Donax vittatus (Donacidae). Mar. Biol. 55:187-195. Fuller, S.C. and R.A Lutz. 1989. Shell morphology of larval and post-larval mytilids from the north-western Atlantic. Journal of the Marine Biological Association, U.K. 69(1):181-218. Fuller, S.C., R.A. Lutz, and Y.-P. Hu. 1989. Bilateral asymmetry in the shell morphology and microstructure of early ontogenetic stages of Anomia simplex. Biol. Bull. 177(1):83-95. Gage, J.D. and P.A. Tyler. 1985. Growth and recruitment of the deep-sea urchin Echinus affinus. Mar. Biol. 90:41- 53. Galtsoff, P.S. 1964. The American oyster. Fish. Bull. 64:1- 480. Goodsell, J.G., S.C. Fuller, A.G. Eversole, M. Castagna, and R.A. Lutz. 1992. Larval and early postlarval shell morphology of several venerid clams. J. Mar. Biol. Assoc. U.K. 72:231-255. Griffiths, R.W. 1991. Spatial distribution and dispersal mechanisms of zebra mussels in the Great Lakes basin. (Abstract). J. Shellfish Res. 10:248. Grigor'eva, N.I. and V.N. Regulev. 1992. Vertical distribution of larvae of the scallop Mitzuhopecten yessoensis and mussel Mytilus edulis in Posyet Bay, Sea of Japan. Soviet J. Mar. Biol. 17:105-109.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 374 Gruffydd, Ll.D., D.J.W. Lane, and A.R. Beaumont. 1975. The glands in the larval foot in Pecten maximus L. and possible homologues in other bivalves. J. Mar. Biol. Assoc. U.K. 55:463-476. Gurney, R. 1942. Larvae of Decapod Crustacea. Ray Society, London. 306 pp. Gustafson, R.G., R.L. Creswell, T.R. Jacobsen, & D.E. Vaughan. 1991. Larval biology and mariculture of the angelwing clam, Cyrtopleura costata. Aquaculture 95(3/4):257-279. Gustafson, R.G. and R.A. Lutz. 1992. Larval and early post-larval development of the protobranch bivalve Solemya velum (Mollusca; Bivalvia). J . Mar. Biol. Assoc. U.K. 72:383-402. Harvey, R. and J.D. Gage. 1984. Observations on the reproduction and postlarval morphology of pourtalesiid sea urchins in the Rockall Trough area (N.E. Atlantic Ocean). Mar. Biol. 82:181-190. Hindler, G. 1975. Adaptational significance of the patterns of development of ophiuroid development. Am. Zool. 15:691-715. Jorgensen, C.B. 1946. Lamellibranchia. 277-311. In Thorson, G. & C.B. Jorgensen. Reproduction and Larval Development of Danish Marine Bottom Invertebrates. Meddelser Fra Kommissionen for Danmarks Fiskeri-og Havundersogelser: Plankton Ser. 4: 523 pp. Kennedy, V.S., R.A. Lutz, and S.C. Fuller. 1989. Larval and early postlarval development of Macoma mitchelli Dali (Bivalvia:Tellinidae). Veliger 32:29-38. LaBarbera, M. and P. Chanley. 197 5. Larval and postlarval development of the corrugated jewel box clam Chama congregata Conrad (Bivalvia: Chamidae). Bull. Mar. Sci. 21:733-744. Lane, D.J.W., and J.A. Nott. 1975. A study of the morphology, fine structure and histochemistry of the foot of the pediveliger of Mytilus edulis L. Journal of the Marine Biological Association, U.K. 55(2):477-495. Lane, D,J.W., J.A. Nott, and D.J. Crisp. 1982. Enlarged stem glands in the foot of the post-larval mussel, Mytilus edulis: adaptations for bysso-pelagic migration. J. Mar. Biol. Assoc. U.K. 62:809-818.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 375 Lane, D.J.W., A.R. Beaumont, and J.R. Hunter. 1985. Byssus drifting and the drifting threads of the young post­ larval mussel Mytilus edulis. Mar. Biol. 84:301-308. Le Pennec, M. and J.H. Jiingbluth. 1983. The ligamental formation of Margaritifera margaritifera (L.) (Bivalvia: Margaritiferidae) and Mytilus edulis (L.) (Bivalvia: Mytilidae) during larval and postlarval ontogenesis. J . Mar. Biol. Assoc. U.K. 63:289-294. Le Pennec, M. and K. Yankson. 1985. Morphogenesis of the hinge in the larva and postlarva of the brackish-water cockle Cerastoderma glaucum. J. Mar. Biol. Assoc. U.K. 65:881-888. Lipcius, R.N., E.J. Olmi, III., and J. van Montfrans. 1990. Planktonic availability, molt stage, and settlement of blue crab postlarvae. Mar. Ecol. Prog. Ser. 58:235-242. Loosar.o.ff, V.L. 1961. Partial metamorphosis in Anomia simplex. Science 133:2070-2071. Loosanoff, V.L. and H.C. Davis. 1963. Rearing of bivalve mollusks. Advances in Marine Biology 1:1-13 6. Loosanoff, V.L., H.C. Davis, and P.E. Chanley. 1966. Dimensions and shapes of larvae of some marine bivalve mollusks. Malacologia 4(2):351-435. Lucas, A. 1975. Sex differentiation and juvenile sexuality in bivalve mollusks. Pub. Staz. Zool. Napoli 39 Suppl.:532-541. VIII European Marine Biology Symposium. Lutz, R., J. Goodsell, M. Castagna, S. Chapman, C. Newell, H. Hidu, R. Mann, D. Jablonski, V. Kennedy, S. Siddall, R. Goldberg, H. Beattie, C. Falmagne, A. Chestnut, and A. Partridge. 1982. Preliminary observations on the usefulness of the hinge structures for identification of bivalve larvae. J. Shellfish Research 2:65-70. Lyons, W.G. 1970. Scyllarid lobsters (Crustacea, Decapoda). Memoirs of the Hourglass Cruise 1(4):1-74. Florida Dept. Nat. Resources Mar. Res. Lab. Publ. 144. Lyons, W.G. 1980. The postlarval stage of scyllaridean lobsters. Fisheries 5(4):47-49. Martel, A. 1992. Occurrence of post-metamorphic drifting in zebra mussels: implications on dispersal and recruitment. (Abstract). J. Shellfish Res. 11(1):231- 232 .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 376 Martel, A. 1993. Wind-induced drifting and settlement of early post-metamorphic zebra mussels. (Abstract). Proceedings of the Larval Ecology Meetings, Port Jefferson, NY, p. 29. Martel, A. and F.-S. Chia. 1991a. Drifting and dispersal of small bivalves and gastropods with direct development. J . Exp. Mar. Biol. Ecol. 150(1):131-147. Martel, A. and F.-S. Chia. 1991b. Foot-raising behaviour and active participation during the initial phase of post-metamorphic drifting in the gastropod Lacuna sp. Mar. Ecol. Prog. Ser. 72:247-254. Morton, J.E. 1964. Locomotion. 384-423. In Wilbur, K.M. and C.M. Yonge (eds.). Physiology of the Mollusca, Vol. I. Academic Press, New York, NY. Nelson, T.C. 1925. Report of the Department of Biology of the New Jersey State Agricultural Experimental Station, 1925: 281-288. Nelson, T.C. 1928. Pelagic dissoconchs of the common mussel Mytilus edulis, with observations on the behavior of the larvae of allied genera. Biol. Bull. 155:180-192. Newell, C.R., H. Hidu, B.J. McAlice, G. Podniesinski, F. Short, and L. Kindblom. 1991. Recruitment and commercial seed procurement of the blue mussel Mytilus edulis in Maine. J . World Aquaculture Soc. 22(2):134- 152. Phillips, B.F. and A.N. Sastry. 1980. Larval ecology. 11- 57. In Cobb, J.S. and B.F. Phillips (eds.). The Biology and Management of Lobsters. Vol. II. Ecology and Management. Academic Press, New York, NY. Phillips, B.F. 1981. The circulation of the southeastern Indian Ocean and the planktonic life of the western rock lobster. Oceanogr. Mar. Biol. Ann. Rev. 19:11-39. Prezant, R.S. 1990. Shell and hinge morphology of the Antarctic bivalve Lissarca notorcadensis. J. Mar. Biol. Assoc. U.K. 70:841-855.

Prezant, R.S. and K. Chalermwat. 1984. Floatation in the bivalve Corbicula fluminea as a means of dispersal. Science 225:1491-1493.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 377 Prytherch, H.F. 1934. The role of copper in the setting, metamorphosis, and distribution of the American oyster, Ostrea virginica. Ecol. Monogr. 4:47-107. Ramorino, L. and B. Campos. 1983. Larvas y postlarvas de Mytilidae (Mollusca: Bivalvia). Rev. Biol. Mar. Inst. Oceanol. Univ. Valparaiso 19:143-192. Rees, C.B. 1950. The identification and classification of lamellibranch larvae. Hull Bulletins of Marine Ecology 3(19):73-104. Robertson, P.B. 19 68. The complete larval development of the sand lobster, Scyllarus americanus (Smith) (Decapoda, Scyllaridae) in the laboratory, with notes on larvae from the plankton. Bull. Mar. Sci. 18:294- 342 . Sellmer, G.P. 1967. Functional morphology and ecological life history of the gem clam, Gemma gemma (Eulamellibranchia: Veneridae). Malacologia 5:137-223. Shaw, W.N. 1965. Seasonal setting patterns of five species of bivalves in the Tred Avon River, Maryland. Chesapeake Science 6:33-37. Sorlin, T. 1988. Floating behavior in the tellinid bivalve Macoma balthica (L.). Oecologia 77:273-277. Sullivan, C.M. 1948. Bivalve larvae of Malpeque Bay, P.E.I. Fisheries Research Board of Canada Bulletin 11. 59 pp. Sigurdsson, J.B., C.W. Titman, and P.A. Davies. 1976. The dispersal of young post-larval bivalve mollusks by byssal threads. Nature 262 (5567):386-387. Stanley, J.G. 1985 Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (Mid-Atlantic) — hard clam. U.S. Fish Wildl. Serv. Biol. Rep. FWS/OBS-82/11.41. 24 pp. Trevelyan, G.A. and E.S. Chang. 198 3. Experiments on larval rearing of the California mussel (Mytilus californianus). J. World Mariculture Soc. 14:137-148. Waite, J.H. 1983. Adhesion in byssally attached bivalves. Biol. Rev. 58:209-231. Webb, C.M. 1986. Post-larval development of the tellinacean bivalves Abra alba, Tellina fabula, and Donax vittatus (Mollusca: Bivalvia) with reference to the late larva. J. Mar. Biol. Assoc. U.K. 66:749-762.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 378 West, D.C., R.K. Thompson, and J.M. Huron. 1983. Opto- electric counters for postlarvae and juveniles of Macrobrachium rosenbergii (De Man); design and tests of Type A counter. Aquacult. Eng. 2:119-133. Williams, A.B. and H.J. Porter. 1971. A ten-year study of meroplankton in North Carolina estuaries: occurrence of postmetamorphal bivalves. Chesapeake Science 12(1):26- 32. Williamson, D.I. 1982. Larval morphology and diversity. 43-110. In Abele, L.G. (ed.). The Biology of Crustacea. Vol. 2. Embryology, Morphology, and Genetics. Academic Press, Inc., New York, NY. Yankson, K. 1986. Observations on byssus systems in the spat of Cerastoderma glaucum and C. edule. J. Mar. Biol. Assoc. U.K. 66:277-292.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 379

Tan-Maas, 1977; & 1982, 1985; only the most important references are are references most important the only al.,

1992 1976 1976 1976 1977; de Blok 1976 al.,

al., al., Regulev, 1992 et al., et et al., & Macoma balthica, Macoma et et al., Board, Board, 1980; Lane et Newell et and and Porter and Williams, 1971 this manuscript References Sigurdsson this manuscript Sigurdsson et Sigurdsson Sigurdsson et Sigurdsson et al., 1976 Martel & Chia, 1991a Grigor'eva Sellmer, Sellmer, 1967 Beaumont & Barnes, 1992 Beaumont & Barnes, 1992 Grigor'eva Regulev, & 1992 Baggerman, 1953; Williams and Porter, 1971 Sigurdsson et al., 1976 Sigurdsson Martel and Chia, 1991a Griffiths, 1991; Martel, 1993

t

Mytilus Mytilus edulis

*t (Modiolarca tumida) Solemya velum*t Geukensia demissa Nucula tenuis Anadara transversa Pecten maximus Modiolus sp. Musculusmarmortatus Gemma gemma Transenella tantilla* Musculus sp.* Mytilus edulis Heteranomia squamata Patinopectenyessoensis Mytilus trossulust Venerupispullastra Turtonia minuta* Petricolapholadiformisi Aequipecten opercularis Dreissenapolymorpha listed; references. further text for see listed; Table 1. Occurrence of postlarval drift by byssal threads in byssal threads by drift postlarval of Occurrence 1. Table Bivalvia. For Solemyacea Taxa Arcacea Nuculacea Mytilacea Pectinacea Anomiacea Veneracea Cyamiaceae Dreissenacea

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 380 1976 al., 1976 1976 1976 1976 1976 1976 1976 1976 1976; Frenkiel Moueza, & 1979 1976 1976 1976 1976 al., al., al., al., al., al., al., al., al., et al., et al., et al., et al., Armonies andHellwig-Armonies, 1992 References Sigurdsson et Sigurdsson Sigurdsson Sigurdsson Baggerman, 1953; Yankson, 1986; Armonies,Yankson, 1992 1986 Sigurdsson et Sigurdsson et Sigurdsson et Sigurdsson et Porter Williams,& 1971 Sigurdsson Williams and Porter, 1971; Armonies, 1992 Sigurdsson et Frenkiel Moueza, & 1979 Porter andWilliams, 1971 Beukema de & Vlas, 1989; Armonies, 1992; this manuscript Sigurdsson et Sigurdsson et al., 1976 Sigurdsson et Baggerman, 1953; Sigurdsson et Sigurdsson et sp. sp.* Martel and Chia, 1991a sp. ______Lasaea Cardium echinatum Cerastoderma glaucum Cerastoderma edule Spisula Mysella ferruginosa Montacuta substriata Cultelluspellucidus Ensis directus Ensis Mactra corallina Tagelusplebiust Scrobiculariaplana Tellina agilis Tellina tenuis Gari fervensis Donax variabilist Donax vittatus Corbula gibba Abra alba Macoma balthica Hiatella arctica Hiatella gallicana Mya arenaria Table 1. 1. Table (cont.) Taxa Cardiacea Galeommatacea Solenacea Mactridae Tellinacea Myacea * Lacks* a free-swimming larval phase. Drifting t mechanism not shown, but byssus probable. Hiatellacea

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA Born April 17, 1964, in Berlin, Federal Republic of Germany, to Marjorie and David Baker. Graduated from Newberg High School, Newberg, Oregon, 1982. Obtained Bachelor of Science degree in Biology, Seattle Pacific University, Seattle, Washington, March 1986. Married to Shirley Marie Coleman in April, 1986. Obtained Masters of Science degree in Biology at the Oregon Institute of Marine Biology, University of Oregon, Charleston, Oregon, August 1988. Entered the School of Marine Science of the College of William and Mary, at the Virginia Institute of Marine Science, Gloucester Point, Virginia, in August, 1988.

381

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.