iiillSifl A : . ; , , . , , - A . ; : : ' : 'MWAmiA . ? ? ? :,.:-,.. ' ? . FRONTISPIECE.?Carrie Bow Cay reef complex, aerial photo mosaic, March 1976. S M I T H S O N I A N C O N T R I B U T I O N S TO THE MARINE SCIENCES - N U M B E R 12 The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay, Belize, I STRUCTURE AND COMMUNITIES Klaus Riitzler and Ian G. Macintyre EDITORS ISSUED JUN1019K MTHSONIAN PUBLICATIONS SMITHSONIAN INSTITUTION PRESS City of Washington 1982 A B S T R A C T Riitzler, Klaus, and Ian G. Macintyre, editors. The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay, Belize, I: Structure and Communities. Smith? sonian Contributions to the Marine Sciences, number 12, 539 pages, frontispiece, 232 figures, 5 plates, 47 tables, 1982.?The results of the first series of multidisciplinary investigations of the Caribbean barrier reef complex near Carrie Bow Cay, Belize, are reported in 34 papers in this volume, which begins with a summary of past work on the Belizean reefs and cays. The first section treats the structure of barrier reef habitats in the vicinity of Carrie Bow Cay, influential physical parameters such as tides and currents, geological and sedimentological history of lagoon, reef, and island substrates, and the island's environment, including its climate and the effects of hurricanes. Subsequent papers analyze the distribution of endolithic microorganisms in carbonate substrates, and the diversity, standing crop, and production in selected lagoon and back-reef habitats. Related contributions report on the benthos of an unusual submarine cave and on the surface zooplankton over reef and lagoon bottoms. One section is devoted to the systematics and local distribution of flora and fauna. Marine plants covered are plankton diatoms, benthic algae?including a detailed study of the red alga Polysiphonia?and sea grasses. Faunistic studies focus on hydroids, medusae, stony corals, octo- corals, sipunculans, anthurid isopods, pycnogonids, a marine chironomid, ophiuroids, and crinoids. In the papers on Polysiphonia, hydroids, stony corals, and anthurids, all species are illustrated for identification by nonspecialists; figures of important or unusual examples are shown in the other systematic contributions. New species are described among anthurids, pycnogonids, and ophiuroids. A section on ecological responses discusses the reaction of algae to grazing pressure, the life history of an ichthyo-parasitic hydroid, the growth response of the reef coral Montastrea annularis to a light gradient, and associa? tions between zoanthids and their sponge hosts. Included in this section are discussions of the ecology of the zoanthid Isaurus duchassaingi, settlement behavior and development of the bivalve Malleus candeanus, and behavioral ecology of two closely related reef fishes, genus Acanthemblemana. The volume concludes with two general surveys of the barrier reef and cays, which discuss the Carrie Bow reef section and cay in relation to the overall barrier reef complex. OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is recorded in the Institution's annual report, Smithsonian Year. Library of Congress Cataloging in Publication Data The Atlantic barrier reef ecosystem at Carrie Bow Cay, Belize. (Smithsonian contributions to the marine sciences ; no. 12) Bibliography: p. Contents: 1. Structure and communities. 1. Coral reef ecology?Belize?Carrie Bow Cay. 2. Carrie Bow Cay (Belize) I. Riitzler, Klaus. II. Macintyre, Ian G. III. Series. QH108.B43A87 574.5'26367'097282 81-607039 AACR2 Contents Page FOREWORD ix Porter M. Kier PREFACE x INTRODUCTION 1 Klaus Riitzler and Ian G. Macin tyre Topography, Bio-Geological Structure, and Physical Environment T H E HABITAT DISTRIBUTION AND COMMUNITY STRUCTURE OF THE BARRIER R E E F C O M P L E X AT CARRIE Bow C A Y , BELIZE 9 Klaus Riitzler and Ian G. Macin tyre T I D E S AT CARRIE Bow C A Y , BELIZE 47 Bjorn Kjerfve, Klaus Riitzler, and George H. Kierspe W A T E R CURRENTS ADJACENT TO CARRIE Bow C A Y , BELIZE 53 Jeffrey E. Greer and Bjorn Kjerfve W A T E R EXCHANGE ACROSS THE R E E F C R E S T AT CARRIE Bow C A Y , B E L I Z E 59 Bjorn Kjerfve G E O L O G Y AND SEDIMENT ACCUMULATION R A T E S AT CARRIE Bow C A Y , B E L I Z E 63 Eugene A. Shinn, J . Haro ld Hudson , Rober t B. Halley, Barbara Lidz, Daniel M. Robb in , and Ian G. Macin tyre TERRESTRIAL ENVIRONMENT AND CLIMATE, CARRIE Bow C A Y , BELIZE . . 77 Klaus Riitzler and J o a n D. Ferraris Benthic and Planktonic Communities DISTRIBUTION OF MICROBORERS WITHIN PLANTED SUBSTRATES ALONG A BARRIER R E E F TRANSECT, CARRIE Bow C A Y , BELIZE 93 Jeffrey A. May , Ian G. Macintyre , and Rona ld D. Perkins PRODUCTION OF SOME BENTHIC COMMUNITIES AT CARRIE Bow C A Y , B E L I Z E 109 Paul E. Hargraves MACROBENTHIC INVERTEBRATES IN BARE SAND AND SEAGRASS {Thalassia testudinum) AT CARRIE Bow C A Y , BELIZE 115 David K. Young and M a r t h a W. Young A SUBMARINE C A V E NEAR COLUMBUS C A Y , BELIZE: A BIZARRE CRYPTIC H A B I T A T 127 Ian G. Macin tyre , Klaus Riitzler, J a m e s N. Norris, and Krist ian Faucha ld v VI SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES SURFACE ZOOPLANKTON AT CARRIE B o w C A Y , B E L I Z E 143 J o a n D. Ferraris Diversity and Distribution of Flora and Fauna PLANKTON DIATOMS (BACILLARIOPHYCEAE) FROM CARRIE Bow C A Y , B E L I Z E 153 Paul E. Hargraves M A R I N E A L G A E AND SEAGRASSES FROM CARRIE Bow C A Y , B E L I Z E 167 J a m e s N. Norris and K a t i n a E. Bucher T H E R E D A L G A Polysiphonia GREVILLE (RHODOMELACEAE) FROM CARRIE B o w C A Y AND VICINITY, BELIZE 225 Donald F. K a p r a u n and J a m e s N. Norris HYDROIDEA (CNIDARIA: HYDROZOA) FROM CARRIE Bow C A Y , B E L I Z E . . . 239 Barry W. Spracklin M E D U S A E (CNIDARIA) FROM CARRIE Bow C A Y , BELIZE 253 Rona ld J . Larson STONY CORALS (CNIDARIA: HYDROZOA, SCLERACTINIA) OF CARRIE Bow C A Y , BELIZE 271 Stephen D. Cairns OCTOCORALLIA (CNIDARIA) FROM CARRIE Bow C A Y , BELIZE 303 Kather ine Muzik DISTRIBUTION OF SIPUNCULA IN THE C O R A L R E E F COMMUNITY, C A R R I E Bow C A Y , BELIZE 311 M a r y E. Rice and Ian G. Mac in ty re ANTHURIDEA (CRUSTACEA: ISOPODA) OF CARRIE Bow C A Y , B E L I Z E 321 Brian Kensley PYCNOGONIDA FROM CARRIE Bow C A Y , BELIZE 355 C. Allan Chi ld Pontomyia EDWARDS (DIPTERA: CHIRONOMIDAE) , A M E M B E R OF THE C O R A L R E E F COMMUNITY AT CARRIE Bow C A Y , BELIZE 381 G e m o t Bretschko OPHIUROIDEA (ECHINODERMATA) FROM CARRIE Bow C A Y , B E L I Z E 387 Frederick H .C . Hotchkiss S H A L L O W - W A T E R CRINOIDEA (ECHINODERMATA) FROM CARRIE Bow C A Y , B E L I Z E 413 D. Bradford M a c u r d a , J r . Behavior, Species Interaction, and Responses to the Environment C H E M I C A L D E F E N S E IN T R O P I C A L M A R I N E ALGAE 417 J a m e s N. Norris and Wil l iam Fenical L I F E HISTORY OF THE HYDROMEDUSA Stomotoca pterophylla H A E C K E L AND ITS ICHTHYOPARASITIC H Y D R O I D 433 Rona ld J . Larson NUMBER 12 vn VARIATION IN G R O W T H FORMS OF THE R E E F C O R A L Montastrea annularis ( E L L I S AND SOLANDER): A QUANTITATIVE EVALUATION OF G R O W T H RESPONSE TO L I G H T DISTRIBUTION USING COMPUTER SIMULATION . . . . 441 Richa rd R. Graus and Ian G. Macin tyre SPONGE-ZOANTHID ASSOCIATIONS: FUNCTIONAL INTERACTIONS 465 Sara M. Lewis O N THE ECOLOGY OF Isaurus duchassaingi (ANDRES) (CNIDARIA: ZOANTHI- DEA) FROM SOUTH W A T E R C A Y , B E L I Z E 475 Kath leen S. Larson and Rona ld J . Larson LARVAL SETTLEMENT BEHAVIOR AND S H E L L M O R P H O L O G Y OF Malleus candeanus (d'ORBiGNY)(MoLLUscA: BIVALVIA) 489 T h o m a s R. Waller and Ian G. Macin tyre H A B I T A T AND RESOURCE PARTITIONING BETWEEN T w o SPECIES OF Acan- themblemaria (PISCES: CHAENOPSIDAE) , WITH COMMENTS ON THE C H A O S HYPOTHESIS 499 David W. Greenfield and Theresa A. Greenfield Carrie Bow Cay as Part of the Belizean Barrier Reef Complex RECONNAISSANCE STUDY OF THE GEOMORPHOLOGY AND BENTHIC C O M ? MUNITIES OF THE O U T E R BARRIER R E E F PLATFORM, BELIZE 509 R a n d o l p h B. Burke SPECIES-AREA RELATIONSHIPS ON SMALL ISLANDS: FLORISTIC D A T A FROM BELIZEAN SAND CAYS 527 David R. Stoddar t and F. R a y m o n d Fosberg PLATES 259 Dedication This volume is dedicated to the 50th anniversary of the Great Barrier Reef Expedition 1928-1929, which pioneered the integration of many science disciplines for the better understanding of the coral reef system Foreword Museum scientists tend to be specialists in a particular discipline and to work alone. Those of us, however, who study living organisms learn sooner or later that we cannot hope to understand our animals and plants fully without some knowledge of the environment in which they exist. We also learn at some point to appreciate the benefits of collaborating with fellow scientists whose disciplinary focus may differ from ours, but whose interests are related to our own through the "environmental" link. That is to say, a great natural ecosystem such as the tropical coral reef draws together researchers of many diverse disciplines. The project "Investigations of Marine Shallow-Water Ecosystems" (IMSWE) off Carrie Bow Cay has done this very thing for many of us at the Smithsonian Institution and elsewhere, and thus we have learned far more about our organisms and their environment than we might have done otherwise. As a result, we are more than enthusiastic about IMSWE's progress. We started with only a few investigators, but as the years have passed we have grown into a multidisciplinary contingent. Needless to say, part of IMSWE's success stems from the efforts of the principal investigator, Klaus Riitzler. He is responsible for the organizing, scheduling, coordinating, and orchestrating. He has done all this with good humor and energy. Of course, such an effort would not have been possible without the support and encouragement of the Belizeans themselves. They love their barrier reef and have high regard for its economic, recreational, and aesthetic value. Recognizing the need for understanding and protecting this precious resource, Belizean officials have approved and assisted our various endeavors since IMSWE's modest beginnings. In particular, Winston Miller, Fisheries Officer, has helped us in every way, as have members of the staff at the Ministry of Trade and Industry (after 1979 the Ministry of Health, Housing, and Coop? eratives), and the Ministry of Finance. We all know that research cannot go forward without adequate funding. The Exxon Corporation has provided part of our support, and has done it in a most generous fashion, allowing the scientists to go about their studies independently. This volume thus represents the culmination of research effort and support from many quarters. It is the first of an open report series and serves as the basis for future contributions. It not only "sets the scene" by determining terminology and summarizing our knowledge to date, but it also points out the gaps yet to be closed. We look forward to the next decade. September 1980 Porter M. Kier, Director (1970-1979) National Museum of Natural History Smithsonian Institution ix Preface On a windy, overcast morning of February 1972, we discovered Carrie Bow Cay. Arnfried Antonius and I (K.R.) were on our way from Stann Creek (now Dangriga) to Glover's Reef, looking for the Tobacco Cay passage through the barrier reef. We could already hear the waves pounding the coral crest but neither of two islands in front of us matched our memory of Tobacco Cay, and the crew of the charter boat from Belize City, unfamiliar with southern waters, was uncertain too. The smaller but nearer one of the cays had buildings and a solid dock so we decided to have a closer look. Minutes later we tied up to the concrete pier and walked toward the stately main building. No one was around, except for a few mildly disturbed pelicans, but a weathered sign above the gate to the main porch said "Welcome to Carrie Bow Island." On that memorable day when we first walked around Carrie Bow?a speck of sand covered by about 80 coconut trees overlooking three cottages, two outhouses, and three water tanks?we could not have predicted the developments of the eight years to come. Despite its small size, this island has been the site of a simple but functional laboratory to which more than 70 scientists from 30 institutions have come to study the well-developed reef complex nearby. Records indicate that up to 1927 Carrie Bow Cay, then known as Ellen Caye or Bird Caye, was an uninhabited sanctuary for migrating birds and for sea turtles coming there to lay their eggs. The surrounding reef "abounded with conch, lobster, turtles, and parrot fish, all of which have been depleted by extensive trade in these commodities," according to the historical account of Henry T. A. Bowman, current owner of Carrie Bow Cay. Mr. Bowman bought the island in 1943 and decided to name it after his wife Carrie. "At that time the cay was about twice the size it is now, and was surrounded by mangrove, and had about 75 coconut trees that were planted in the early 1900's. The mangrove, breeding spot for mosquitos, was cleared off in 1944 when [he] decided to build a summer home. Of the original coconut trees only six remain today." Mr. Bowman's account also cites the stresses intro? duced by countless storms and hurricanes, which were particularly harsh during the past three decades but whose damage Carrie Bow Cay and the reef have managed to survive. Notwithstanding these changes, the area remains relatively undisturbed by scientific standards, and thus is an ideal location for an ecological study of coral reefs. More significant, it is a segment of an enormous reef complex that is reasonably accessible from major cities on the North American mainland. This volume presents the first extensive scientific account of the barrier reef around Carrie Bow Cay, Belize. It is an outgrowth of the Smithsonian Institution Program "Investigations of Marine Shallow-Water Ecosystems" (IMSWE), which began in 1970 under the guidance of W. H. Adey, A. L. NUMBER 12 Dahl, and the editors of this volume. It was our goal to introduce long-term cooperation and a truly interdisciplinary approach to the ecological study of coral reefs and related tropical environments. Our search for a study site that would be satisfactory to the special requirements of various disciplines led us to the reef off Belize. IMSWE evolved alongside a larger proposed project for the study of a coral-reef ecosystem that was to be a multi-institutional and multi-national program sponsored by the International Decade of Ocean Exploration. Glover's Reef atoll was contemplated as a site for that program's comparative studies and Carrie Bow Cay?although not yet seen by any of the planning committee?was discussed as an additional possibility because of its logistical advantages, particularly its location half way between the mainland of Belize and the atoll. Although the large program was not funded, Smithsonian Institution scientists carried forward the idea for an interdisciplinary study of a coral reef ecosystem. Since that time, many enthusiasts have joined in IMSWE's collaborative investigations. Because of the growing body of information resulting from our work, we are now able to present this volume, whose purpose is to give a general introduction to the topography, oceanography, geology, and biology of the Belizean barrier reef complex near Carrie Bow Cay. Some new terms have been introduced in this volume to describe the physiographic characteristics and zonation of the reef in this area. It is hoped X l l SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES that even if these terms are revised at a later date, they will serve as an important reference for future research reports. This contribution also opens a publication series that makes it possible to add information on this reef complex as it becomes available and to keep the results in focus for subsequent additions and for a later synthesizing monograph. The use of brand names in this publication is for descriptive purposes only and does not constitute endorsement by the Smithsonian Institution. ACKNOWLEDGMENTS.?Our program and the production of this volume has received, over the years, help and encouragement from so many supporters in direct and indirect ways that it is impossible to give a just account of them all. Those who have contributed specifically to papers published in the following pages are acknowledged in the article introductions. Here, we wish to highlight the accomplishments of individuals and groups whose efforts have made our program successful and therefore made possible the production of this publi? cation. The Exxon Corporation extended grant support to us over the past five years; we are grateful not only for the financial aid but also for the personal interest in our studies shown by R. E. Chandler and E. W. Markowski, Public Affairs Department. Other funds came from the Smithsonian Environmental Sciences Program, National Museum of Natural History, Research Award Program and Fluid Research Fund, for which we thank D. Challinor, P. M. Kier, J. F. Mello, and S. D. Ripley. Special donations from AMOCO, Cities Service, Getty, Gulf, and Union Oil companies, and from K. Sandved and L. Taylor made possible the printing of the color plates in this volume. The people of Belize greatly enriched our days in the field, and its NUMBER 12 Xll l government officials helped us in many ways. W. Miller, Fisheries Officer and our scientific sponsor, patiently helped with certifications and permits. The Ministry of Trade and Industry (after 1979 the Ministry of Health, Housing, and Cooperatives) sanctioned our work, the Ministry of Finance granted exemption of scientific equipment from duty. H.T.A. Bowman ("Sir Henry") and Mrs. Bowman were our generous hosts for the past eight years. At the same time, H.T.A. Bowman, Jr. ("Junior") worked hard as our agent and guide, assisted by his wife Alice, sister Norma, and by other members of his family. H.T.A. Bowman III ("Henry") has read the tide gauge installed by us at Pelican Beach Motel, Dangriga, for the past four years. For many years the single most important person for the program was M. R. Carpenter, who solved logistical problems at home and in the field. Mike was preceded, or at times relieved, as field manager by H. Adolfi, A. Antonius, R. J. Larson, A. B. Rath, R. H. Sims, and B. Spracklin. Others who contributed to our comfort on the cay include people known familiarly to us as Frank and Japs, staff members of Pelican Beach Motel, and Emily, Ernie, and Genevieve, our cooks. Invaluable logistical help was also provided by C. Moore of Sen & Co., Belize City. At the National Museum of Natural History our study benefited greatly from the administrative expertise of C. A. Ossola, M. Parrish, and M. R. Tanner and from the technical assistance of C. G. Ahearn, W. T. Boykins, A. C. Cohen, L. G. Cole, R. J. and K. S. Larson, and K. B. Sandved. J. Petroski, Travel Services Office, did miracles with airplane reservations. This volume is the product of our collaboration with a number of patient authors and several dedicated assistants and goodwilled advisors. Presentation was shaped at an early stage by the skillful editing of V. V. Macintyre and the proficient illustrating of I. F. Jewett. M. Parrish and A. Stonework typed or retyped most of the contributions. S. D. Cairns, M. R. Carpenter, M. Parrish, and D. S. Robertson helped as editorial assistants. C. Schbpfer-Sterrer created the accompanying ink sketches and G. J. Thomas took the aerial photographs for the frontispiece. The following colleagues helped us as reviewers or scientific advisors: W. H. Adey, F. M. Bayer, T. E. Bowman, M. x i v SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES A. Buzas, S. D. Cairns, D. R. Calder, F. A. Chace, B. B. Collette, M. E. Downey, G. L. Hendler, R. S. Houbrick, E. Kirsteuer, C. G. Messing, J. N. Norris, D. L. Pawson, H. Pulpan, J. Rosewater, and T. Waller. Finally, special thanks are due to A. L. Ruffin, managing editor, B. J. Spann, editor, and L. J. Long, production manager, of the Smithsonian Institution Press, for seeing this publication through the final stages of production. The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay, Belize, I Introduction Klaus Riitzler and Ian G. Macintyre The barrier reef off Belize, Central America, has received the concentrated attention of scien? tists only within the past two decades. Interest in this area dates back to the late nineteenth cen? tury, however, the first scientific reference to the Belizean reefs having been made by Charles Dar? win, himself, in an 1842 work, 77?^ Structure and Distribution of Coral Reefs (pages 201-202). Using information provided by Captain B. Allen, Dar? win discussed the bathymetry associated with Belizean reefs and included them in his classifi? cation of principal reef types. Early studies of the natural history of Belize, then known as British Honduras, focused not on the reefs but on incidental collections taken along the shore or by dredge hauls by residents of the colony and by visiting individuals and expedi? tions, most notably the "Pawnee" (1925) and "Rosaura" (1937-1938) expeditions. One of the oldest such records is the description of a new sponge, collected near Belize (now Belize City) by Priest (1881). Other systematic work on early collections dealt with algae (Taylor, 1935), sea- grass Thalassia (den Hartog, 1970), sponges (Bur? ton, 1954), cnidarians (Boone, 1928a), crusta- Klaus Riitzler, Department of Invertebrate Zoology, and Ian G. Macintyre, Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. ceans (Boone, 1927), mollusks (Boone, 1928b; Marcus and Marcus, 1962; Robertson, 1975), echinoderms (Boone, 1928c; John and Clark, 1954), and fishes (Breder, 1927; Tucker, 1954; Robins and Starck, 1965; Birdsong and Emery, 1968). In addition, studies relating to commercial fisheries were reported by Smith (1939, 1941) and Craig (1966). The first investigations focusing on the cays and reefs of Belize were undertaken by the "Cam? bridge Expedition to British Honduras" (1959- 1960) led by J. E. Thorpe (Thorpe and Stoddart, 1962). Some members of the expedition mapped 40 cays, including Carrie Bow Cay, and collected samples of their flora. They also discussed the origin, formation, and distribution of reef and mangrove cays, and showed the migration of some of these cays on the basis of exposed beach rock (Stoddart, 1960). Another expedition team mapped bottom characteristics and identified coral species from around Rendezvous Cay, and conducted experiments on coral behavior in re? sponse to external stimuli (Thorpe and Bregazzi, 1960). Subsequently Stoddart, one of this expe? dition's participants, surveyed in detail the biol? ogy and geology of the Belizean atolls, that is, Turneffe Islands, Lighthouse Reef, and Glover's Reef (Stoddart, 1962a). At this time he also ini? tiated an interesting series of studies concerning SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES the hurricane destruction and post-storm recov? ery of cays and coral reefs (Stoddart, 1962b, 1963, 1969, 1974). A second stage of intensive research activity was spearheaded by E. G. Purdy of Rice Univer? sity, under whom several major projects?for the most part doctoral disserations?treating various geological aspects of the barrier reef complex were completed between 1961 and 1967. Most of this work appeared in a volume edited by Wantland and Pusey (1975) that focused on sediments and processes of sedimentation and diagenesis (L. R. High, Jr., pages 53-96; M. R. Scott, pages 97- 130; W. C. Pusey III; pages 131-233; W. J. Ebanks, Jr., pages 234-296), paleoecology and diagenesis of Pleistocene limestone (G. E. Teb- butt, pages 297-331), and distribution of benthic Foraminifera (K. F. Wantland, pages 332-339) and Ostracoda (J. W. Teeter, pages 400-499). Another member of this group of doctoral scholars documented elsewhere the genesis of re? cent lime mud (Matthews, 1966). In addition, Purdy himself produced two thought provoking reports dealing with the origin of the topographic relief of the Belizean reef and its influence on sediment distribution (Purdy, 1974a, 1974b). Among other studies of the Belizean reefs some of the most notable have been conducted by the University of Miami under the direction of R. N. Ginsburg. This work, to date, has focused on shallow-water submarine cements (Ginsburg and James, 1976; James et al., 1976) and has included geological investigations (James and Ginsburg, 1978) and a study of deep-reef fishes (Colin, 1974) of the deep seaward margins of the barrier reef and Glover's Reef atoll. Several recent studies have concentrated on the barrier reef and the atolls. Devaney (1974) has reported on shallow-water echinoderms collected off Turneffe Island and Lighthouse and Glover's reefs. Wallace and Schafersman (1977) have ex? amined the ecology and sedimentology of patch reefs in the Glover's Reef lagoon. This atoll was also the point of origin for a series of ichthyolo- gical studies that were expanded to the barrier reef complex and are still under way (Greenfield, 1972, 1975a-c, 1979; Greenfield and Greenfield, 1973; Greenfield and Johnson, 1981). With the establishment of the Smithsonian Institution's field laboratory at Carrie Bow Cay in 1972, most of the marine research in Belize concentrated on this part of the barrier reef; only some early site surveys and comparative studies included other locations. Smithsonian-sponsored physiographic surveys (Dahl et al., 1974; Miller and Macintyre, 1977) were important early steps of our programs, while hurricane reports and damage and recovery surveys (Antonius, 1972; Stoddart, 1974; Highsmith et al., 1980) appear to have become a continuing necessity. Among the methods developed or modified during our reef research were mapping (Riitzler, 1978a), under? water coring (Macintyre, 1975), and ecological sampling techniques (Dahl, 1973; Kirsteuer, 1978; Riitzler, 1978b; Riitzler et al., 1980). Early in our program, attempts were made to monitor the physical environment of Carrie Bow Cay and the surrounding waters. Cooperation with colleagues at the University of South Caro? lina who contributed specialized equipment pro? duced information on the diurnal energy balance on the island (Kjerfve, 1978) and on tidal patterns relative to the Caribbean system (Kjerfve, 1981). Inventory of flora and fauna was a prerequisite for the new program and produced an increasing number of checklists, distributional tabulations, taxonomic revisions, and descriptions of new taxa: algae and seagrasses (Tsuda and Dawes, 1974); nemerteans (Kirsteuer, 1974, 1977); poly- chaetes (Fauchald, 1980); sipunculans (Rice, 1976); copepods (H. B. Cressey, 1981; R. Cressey, 1981); ostracods (Kornicker, 1978, 1981; Kor- nicker and Cohen, 1978); decapods (Kensley, 1981; Kensley and Gore, 1981), bivalves (Waller, 1978), holothuroids (Pawson, 1976); echinoids (Kier, 1975); and fishes (Greenfield, 1981; Green? field and Johnson, 1981; Johnson and Greenfield, 1982). A comparable inventory of novel chemical compounds from algae, sponges, and gorgonians is the subject of a series of publications of W. Fenical's group at Scripps Institution of Ocean? ography (McEnroe and Fenical, 1978; Mc- NUMBER 12 Connell and Fenical, 1978; Kokke et al., 1979; Kokkeet al., 1981). Participants of several programs have exam? ined organisms producing limestone and calcar? eous sand. Environmental influences on skeletal development of corals were studied by Barnard et al. (1974), Macintyre and Smith (1974), Graus and Macintyre (1976), and Highsmith (1979). Related work on mollusks has dealt with growth rates of gastropods {Cerithium: Houbrick, 1974) and shell calcification of bivalves (Arcoida: Waller, 1980). Calcium carbonate breakdown by biological processes, on the other hand, was the subject of studies by Riitzler (1975), Rice (1976), and Highsmith (1981), and the contribution of noncalcareous (siliceous) components to reef sands was treated by Riitzler and Macintyre (1978). Another topic of research at Carrie Bow Cay has been the availability and quality of food as a major factor determining growth and distribution of invertebrates and fishes. Primary production by benthic algae, including spatial and temporal variability, were discussed by Dahl (1974, 1976). Distribution of many algae is controlled by her? bivores (Hay, 1981) but some plants can defend themselves from grazers by producing toxic com? pounds that act as feeding deterrents (Gerwick et al., 1979; Sun and Fenical, 1979; Paul and Feni? cal, 1980; Gerwick and Fenical, 1981). On the other hand, strong symbiotic ties exist between certain algae and invertebrates that were studied by Kokke et al. (1980) and Riitzler (1981). An? other report on invertebrate feeding presents be? havioral observations on coronate medusae cap? tured in plankton tows near Carrie Bow Cay (Larson, 1979). The earlier work summarized here will give the reader an indication of research projects under way at the Carrie Bow Cay laboratory or spon? sored by the Smithsonian reef study program elsewhere in Belize. The following papers in this volume will add further data and important new perspectives to our knowledge. Many more con? tributions can be expected to follow. Literature Cited Antonius. 1972. Barnard, 1974. Birdsong, 1968. Boone, L 1927. Hurricane Laura, Witnessed in British Honduras. Atoll Research Bulletin, 162:11-12. L. A., I. G. Macintyre, and J. W. Pierce Possible Environmental Index in Tropical Reef Corals. Nature, 252:219-220. R. S., and A. R. Emery New Records of Fishes for the Western Caribbean. Quarterly Journal of the Florida Academy of Sciences, 30: 187-196. Scientific Results of the First Oceanographic Ex? pedition of the "Pawnee," 1925: Crustacea from Tropical East American Seas. Bulletin of the Bingham Oceanographic Collection, 1(2): 1-147. 1928a. Scientific Results of the First Oceanographic Ex? pedition of the "Pawnee," 1925: Molusca from Tropical East American Seas. Bulletin of the Bingham Oceanographic Collection, 1(3): 1-20. 1928b. Scientific Results of the First Oceanographic Ex? pedition of the "Pawnee," 1925: Echinodermata from Tropical East American Seas. Bulletin of the Bingham Oceanographic Collection, 1(4): 1-22. 1928c. Scientific Results of the First Oceanographic Ex? pedition of the "Pawnee," 1925: Coelenterata from Tropical East American Seas. Bulletin of the Bingham Oceanographic Collection, 1(5): 1-8. Breder, C. M., Jr. 1927. Scientific Results of the First Oceanographic Ex? pedition of the "Pawnee," 1925: Fishes from Trop? ical East American Seas. Bulletin of the Bingham Oceanographic Collection, 1(1): 1-90. Burton, M. 1954. The "Rosaura" Expedition, 1937-1938: Sponges. Bulletin of the British Museum (Natural History), Zool? ogy, 2:215-239. Colin, P. L. Observations and Collection of Deep-Reef Fishes off the Coasts of Jamaica and British Honduras (Belize). Marine Biology, 24:29-38. K. Geography of Fishing in British Honduras and Adjacent 1974. Craig, A 1966. SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Coastal Waters. 143 pages. Baton Rouge, Louisiana: Louisiana State University Press. Cressey, H. B. 1981. Ceratocolax mykternastes, New Species (Copepoda, Bomolochidae) Parasitic in the Nasal Sinus of Haemulon sciurus (Pisces, Pomadasyidae) from Be? lize. Proceedings of the Biological Society of Washington, 94(2) :514-524. Cressey, R. 1981. Parasitic Copepods from the Gulf of Mexico and Caribbean Sea, I: Holobomolochus and Neobomolo- chus. Smithsonian Contributions to Zoology, 339: 24 pages. Dahl, A. L. 1973. Surface Area in Ecological Analysis: Quantifica? tion of Benthic Coral-Reef Algae. Marine Biology, 23:239-249. 1974. The Structure and Dynamics of Benthic Algae in the Coral Reef Ecosystem. In A. M. Cameron, B. M. Campbell, A. B. Cribb, R. Endean, J. S. Jell, O. A.Jones, P. Mather, and F. H. Talbot, editors, Proceedings of the Second International Coral Reef Sym? posium, 1:21-25. Brisbane, Australia: The Great Barrier Reef Committee. 1976. Generation of Photosynthetic Surface Area by Coral Reef Algae. Micronesica, 12:43-47. Dahl, A. L., I. G. Macintyre, and A. Antonius 1974. A Comparative Survey of Coral Reef Research Sites. In M.-H. Sachet and A. L. Dahl, editors, Comparative Investigations of Tropical Reef Eco? systems: Background for an Integrated Coral Reef Program. Atoll Research Bulletin, 172:37-120. Darwin, C. 1842. The Structure and Distribution of Coral Reefs. 214 pages. London, den Hartog. See Hartog, C. den Devaney, D. M. 1974. Shallow-Water Echinoderms from British Hon? duras, with a Description of a New Species of Ophiocoma (Ophiuroidea). Bulletin of Marine Science, 24(1): 122-164. Fauchald, K. 1980. Onuphidae (Polychaeta) from Belize, Central America, with Notes on Related Taxa. Proceedings of the Biological Society of Washington, 93(3): 797-829. Gerwick, W. H., and W. Fenical 1981. Ichthyotoxic and Cytotoxic Metabolites of the Tropical Brown Alga, Slypopodium zonate. Journal of Organic Chemistry, 46:22-27. Gerwick, W. H., W. Fenical, N. Fritsch, and J. Clardy 1979. Stypotriol and Stypoldione: Ichthyotoxins of Mixed Biogenesis from the Mrine Alga Slypopodium zonale. Tetrahedron Letters, 1979(2): 145-148. Ginsburg, R. N., and N. P. James 1976. Submarine Botryoidal Aragonite in Holocene Reef Limestones, Belize. Geology, 4:431-436. Graus, R. R., and I. G. Macintyre 1976. Light Control of Growth Form in Colonial Reef Corals: Computer Simulation. Science, 193:895- 897. Greenfield, D. W. 1972. Notes on the Biology of the Arrow Blenny, Lucay- ablennius zingaro (Bdhlke) from British Honduras. Copeia, 1972:590-592. 1975a. Emblemariopsis pricei, a New Species of Chaenopsid Blenny from Belize. Copeia, 1975:713-715. 1975b. Cleplicus parrae, an Additional Sponge-Dwelling Fish. Copeia, 1975:381-382. 1975c. Centropomus poeyi from Belize, with a Key to the Western Atlantic Species of Centropomus. Copeia, 1975:582-583. 1979. A Review of the Western Atlantic Starksia ocellata- Complex (Pisces: Clinidae) with the Description of Two New Species and Proposal of Superspecies Status. Fieldiana: Zoology, 73:9-48. 1981. Vancus imswe, a New Gobiid Fish from Belize. Copeia, 1981:269-272. Greenfield, D. W., and T. Greenfield 1973. Triathalassothia gloverensis, a New Species of Toad- fish from Belize (= British Honduras), with Re? marks on the Genus. Copeia, 1973:560-565. Greenfield, D. W., and R. K. Johnson 1981. The Blennioid Fishes from Belize and Honduras, Central America, with Comments on their Sys- tematics, Ecology, and Distribution (Blenniidae, Chaenopsidae, Labrisomidae, Tripterygiidae). Fieldiana: Zoology, new series, 8:1-106. Hartog, C. den 1970. Sea-Grasses of the World. Verhandelingen der Konink- lijke Nederlandse Akademie van Wetenschappen, Afdeel- ing Natuurkunde, 59:1-275. Hay, M. E. 1981. Spatial Patterns of Grazing Intensity on a Carib? bean Barrier Reef: Herbivory and Algal Distri? bution. Aquatic Botany, 11:97-109. Highsmith, R. C. 1979. Coral Growth Rates and Environmental Control of Density Banding. Journal of Experimental Marine Biology and Ecology, 37:105-125. 1981. Coral Bioerosion: Damage Relative to Skeletal Density. American Naturalist, 117:193-198. Highsmith, R. C , A. C. Riggs, and C. M. D'Antonio 1980. Survival of Hurricane-Generated Coral Fragments and a Disturbance Model of Reef Calcification/ Growth Rate. Oecologia (Berlin), 46:322-329. Houbrick, \J] R - 1974. Growth Studies on the Genus Cerithium (Gastro? poda: Prosobranchia) with Notes on Ecology and Microhabitats. The Nautilus, 88:14-27. NUMBER 12 James, N. P., and R. N. Ginsburg 1978. The Deep Seaward Margin of Belize Barrier and Atoll Reefs. 201 pages. Miami Beach, Florida: University of Miami. James, N. P., R. N. Ginsburg, D. S. Marszalek, and P. W. Choquette 1976. Facies and Fabric Specificity of Early Subsea Ce? ments in Shallow Belize (British Honduras) Reefs. Journal of Sedimentary Petrology, 46(3):523-544. John, D. D., and A. M. Clark 1954. The Echinodermata. In The "Rosaura" Expedi? tion, 1937-1938. Bulletin of the British Museum [Nat? ural History), Zoology, 2:139-162. Johnson, R. K., and D. W. Greenfield In press. Cling Fishes from Belize and Honduras, Central America, with a Redescription of Gobiesox barbalu- lus Starck. Northeast Gulf Science. Kensley, B. F. 1981. Notes on Axiopsis (Axiopsis) serratifrons (A. Milne Edwards) (Crustacea: Decapoda: Thalassinidea). Proceedings of the Biological Society of Washington, 93(4): 1253-1263. Kensley, B. F., and R. H. Gore 1981. Coralaxius abelei, New Genus and New Species (Crustacea: Decapoda: Thalassinidea: Axiidae): A Coral-inhabiting Shrimp from the Florida Keys and the Western Caribbean Sea. Proceedings of the Biological Society of Washington, 93(4): 1277-1294. Kier, P. M. 1975. The Echinoids of Carrie Bow Cay, Belize. Smith? sonian Contributions to Zoology, 206: 45 pages. Kirsteuer, E. 1974. Description of Poseidonemertes caribensis sp. n., and Discussion of Other Taxa of Hoplonemertini Monostilifera with Divided Longitudinal Muscu? lature in the Body Wall. Zoologica Scripta, 3:153- 166. 1977. Remarks on Taxonomy and Geographical Distri? bution of the Genus Ototyphlonemertes Diesing (Nemertina, Monostilifera). Mikrofauna Meeres- boden, 61:167-181. 1978. Coral-Associated Nemertina. In D. R. Stoddart and R. E. Johannes, editors, Coral Reefs: Re? search Methods. Monographs on Oceanographic Meth? odology, 5:315-327. Paris: Unesco. Kjerfve, B. 1978. Diurnal Energy Balance of a Caribbean Barrier Reef Environment. Bulletin of Marine Science, 28: 137-145. 1981. Tides of the Caribbean Sea. Journal of Geophysical Research, 86:4243-4247. Kokke, W.C.M.C, L. Bohlin, W. Fenical, and C. Djerassi In press. Minor and Trace Sterols of Marine Inverte? brates, X X : New Methyl Sterols from the Carib? bean Gorgonians Briareum asbestinum and Muriceop- sis flavida. Journal of Steroids. Kokke, W.C.M.C, W.Fenical, L. Bohlin, and C. Djerassi 1980. Sterol Synthesis by Cultured Zooxanthellae: Im? plications Concerning Sterol Metabolism in Host- Symbiont Association in Caribbean Gorgonians. Comparative Biochemistry and Physiology, 68B:281- 287. Kokke, W.C.M.C, C. S. Pak, W. Fenical, and C Djerassi 1979. Minor and Trace Sterols in Marine Invertebrates, XII: Occurrence of 24(#+.S)-Isopropenyl- cholesterol, 24(/c-hS)-MethylchoIesta-5,25-dien- 3/?-ol, and 24(/c-hS)-Methylcholesta-7,25-dien- 3/?-ol in the Caribbean Sponge, Verongia caulifor- mis. Helvetica Chimica Ada, 62(4): 1310-1318. Kornicker, L. S. 1978. Harbansus, a New Genus of Marine Ostracoda, and a Revision of the Philomedidae (Myodocopina). Smithsonian Contributions to Zoology, 260: 75 pages. 1981. Revision, Distribution, Ecology and Ontogeny of the Ostracode Subfamily Cyclasteropinae (My? odocopina: Cylindroleberididae). Smithsonian Con? tributions to Zoology, 319: 548 pages. Kornicker, L. S., and A. C Cohen 1978. Dantyinae, a New Subfamily of Ostracoda (My? odocopina: Sarsiellidae). Proceedings of the Biological Society of Washington, 91(2):490-508. Larson, R. J. 1979. Feeding in Coronate Medusae (Class Scyphoza, Order Coronatae). Marine Behavior and Physiology, 6:123-129. Macintyre, I. G. 1975. A Diver-operated Hydraulic Drill for Coring Sub? merged Substrates. Atoll Research Bulletin, 165:21- 26. Macintyre, I. C , and S. V. Smith 1974. X-Radiographic Studies of Skeletal Development in Coral Colonies. In A. M. Cameron, B. M. Campbell, A. B. Cribb, J . Endean, J. S. Jell, O. A. Jones, P. Mather and F. H. Talbot, editors, Proceedings of the Second International Coral Reef Sym? posium, 2:277-287. Brisbane, Australia: The Great Barrier Reef Committee. Marcus, E., and E. Marcus 1962. Opisthobranchs from Florida and the Virgin Is? lands. Bulletin of Marine Science of the Gulf and Car? ibbean, 12:450-488. Matthews, R. K. 1966. Genesis of Recent Lime Mud in Southern British Honduras. Journal of Sedimentary Petrology, 36:428- 454. McConnell, O. J., and W. Fenical 1978. Ochtodene and Ochtodiol: Novel Polyhalogen- ated Cyclic Monoterpenes from the Red Seaweed Ochlodes secundiramea. Journal of Organic Chemistry, 43:4238-4241. 6 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES McEnroe, F. J., and W. Fenical 1978. Structures and Synthesis of Some New Antibac? terial Sesquiterpenoids from the Gorgonian Coral Pseudopterogorgia rigida. Tetrahedron Letters, 1978(34): 1661-1664. Miller, J. A., and I. G. Macintyre 1977. Third International Symposium on Coral Reefs, Field Guidebook to the Reefs of Belize. 36 pages. Miami Beach, Florida: The Atlantic Reef Committee. Paul, V., and W. Fenical 1980. Toxic Acetylene Containing Lipids from the Red Alga Liagura farinosa Lamaroux. Tetrahedon Letters, 1980(21):3327-3330. Pawson, D. L. 1976. Shallow-Water Sea Cucumbers (Echinodermata: Holothuroidea) from Carrie Bow Cay, Belize. Pro? ceedings of the Biological Society of Washington, 89(31): 369-382. Priest, B. W. 1881. On an Undescribed Species of Sponge of the Ge? nus Polymastia, from Honduras. Journal of the Quekett Microscopical Club, 6:302-304. Purdy, E. G. 1974a. Reef Configurations: Cause and Effect. In L. F. Laporte, editor, Reefs in Time and Space. Society of Economic Paleontolgists and Mineralogists, Special Publication, 18:9-76. Tulsa, Oklahoma. 1974b. Karst-Determined Facies Patterns in British Hon? duras: Holocene Carbonate Sedimentation Model. The American Association of Petroleum Geolo? gists Bulletin, 58:825-855. Rice, M. E. 1976. Sipunculans Associated with Coral Communities. Micronesica, 12:119-132. Robertson, R. R. 1975. Systematic List of Commonly Occurring Marine Mollusks of Belize. In K. F. Wantland and W. C. Pusey III, editors, Belize Shelf-Carbonate Sedi? ments, Clastic Sediments, and Ecology. The Amer? ican Association of Petroleum Geologists, Studies in Ge? ology, 2:40-52. Tulsa, Oklahoma. Robins, C R., and W. A. Starck II 1965. Opsanus astrifer, a New Toadfish from British Hon? duras. Proceedings of the Biological Society of Washing? ton, 78:247-250. Riitzler, K. 1975. The Role of Burrowing Sponges in Bioerosion. Oecologia (Berlin), 19:203-216. 1978a. Photogrammetry of Reef Environments by He? lium Balloon. In D. R. Stoddart and R. E. Johan? nes, editors, Coral Reefs: Research Methods. Mon? ographs on Oceanographic Metholodogy, 5:45-52. Paris: Unesco. 1978b. Sponges in Coral Reefs. In D. R. Stoddart and R. E.Johannes, editors, Coral Reefs: Research Meth? ods. Monographs on Oceanographic Methodology, 5:299- 313. Paris: Unesco. 1981. An Unusual Blue-Green Alga Symbiotic with Two New Species of Ulosa (Porifera: Hymeniacidoni- dae). Marine Ecology, 2:35-50. Riitzler, K , J . D. Ferraris, and R. J. Larson 1980. A New Plankton Sampler for Coral Reefs. Marine Ecology, 1:65-71. Riitzler, K., and Macintyre, I. G. 1978. Siliceous Sponge Spicules in Coral Reef Sedi? ments. Marine Biology, 49:147-159. Smith, F.G.W. 1939. Sponge Mortality at British Honduras. Nature, 144:785. 1941. Sponge Disease in British Honduras, and its Transmission by Water Currents. Ecology, 22:416- 421. Stoddart, D. R. 1960. The Reefs and Sand Cays of British Honduras. In Cambridge Expedition to British Honduras, 1959-1960, General Report, pages 16-22. Cambridge, England. 1962a. Three Caribbean Atolls: Turneffe Islands, Light? house Reef and Glover's Reef, British Honduras. Atoll Research Bulletin, 87:1-151. 1962b. Catastrophic Storm Events on the British Hon? duras Reefs and Cays. Nature, 196:512-515. 1963. Effects of Hurricane Hattie on the British Hon? duras Reefs and Cays, October 30-31, 1961. Atoll Research Bulletin, 95:1-142. 1969. Post-Hurricane Changes on the British Honduras Reefs and Cays: Re-survey, 1965. Atoll Research Bulletin, 131:1-31. 1974. Post-Hurricane Changes on the British Honduras Reefs: Re-survey, 1972. In A. M. Cameron, B. M. Campbell, A. B. Cribb, R. Endean, J . S. Jell, O. A. Jones, P. Mather and F. H. Talbot, editors, Proceedings of the Second International Coral Reef Sym? posium, 2:473-483. Brisbane, Australia: The Great Barrier Reef Committee. Sun, H. H., and W. Fenical 1979. Rhipocephalin and Rhipocephenal: Toxic Feed? ing Deterrents from the Tropical Marine Alga Rhipocephalus phoenix. Tetrahedron Letters, 5:685-688. Taylor, W. R. 1935. Botany of the Maya Area; Miscellaneous Papers, VII: Marine Algae from the Yucatan Peninsula. Carnegie Institution of Washington Publication, 461: 115-124. Thorpe, J . E., and P. K. Bregazzi 1960. Experiments and Observations on the Corals at Rendezvous Cay. In Cambridge Expedition to British Honduras, 1959-1960, General Report, pages 22-28. Cambridge, England. Thorpe, J . E., and D. R. Stoddart 1962. Cambridge Expedition to British Honduras. Ge? ography Journal, 128:158-171. NUMBER 12 Tsuda, R. T., and C. J . Dawes 1974. Preliminary Checklist of the Marine Benthic Plants from Glover's Reef, British Honduras. Atoll Research Bulletin, 173:1-13. Tucker, D. W. 1954. Fishes, Part I: Families Carcharhiniidae, Torpe- dinidae, Rosauridae (nov.), Salmonidae, Alepo- cephalidae, Searsidae, Clupeidae. In The "Ro- saura" Expedition, 1937-1938. Bulletin of the British Museum (Natural History), Zoology, 2:163-214. Wallace, R. J., and S. D. Schafersman 1977. Patch-Reef Ecology and Sedimentology of Glovers Reef Atoll, Belize. In S. H. Frost, M. P. Weiss, and J . B. Saunders, editors, Reefs and Related Car? bonates?Ecology and Sedimentology. The Ameri? can Association of Petroleum Geologists, Studies in Geol? ogy, 4:37-52. Tulsa, Oklahoma. Waller, T. R. 1978. Morphology, Morphoclines and a New Classifi? cation of the Pteriomorphia (Mollusca: Bivalvia). Philosophical Transactions of the Royal Society of London, B, 284:345-365. 1980. Scanning Electron Microscopy of Shell and Man? tle in the Order Arcoida (Mollusca: Bivalvia). Smithsonian Contributions to Zoology, 313: 58 pages. Wantland, K. K , and W. C Pusey III, editors 1975. Belize Shelf-Carbonate Sediments, Clastic Sedi? ments, and Ecology. The American Association of Petroleum Geologists, Studies in Geology, 2: 599 pages. Tulsa, Oklahoma. The Habitat Distribution and Community Structure of the Barrier Reef Complex at Carrie Bow Cay, Belize Klaus Riitzler and Ian G Macintyre ABSTRACT The reef complex near Carrie Bow Cay, which is representative of the entire Belizean barrier reef, is separated from the mainland by a deep and wide lagoon, which grades into shallow sea- grass bottoms, patch reefs, and mangrove cays on the outer barrier platform. A study transect west (lagoon) to east (open ocean) shows a distinct zonation of substrates and organisms that reflects primarily water depth and the prevailing wave and current regime. The shallow back reef shows massive coral growth, extensive pavement areas, and large rubble accumulations caused by hur? ricane surge; it is separated from the inner fore reef by a narrow intertidal reef crest pounded by waves. The inner fore reef (to 14 m depth) shows a characteristic spur and groove structure, with high buttresses in the shallow depth zone (to 10 m) and low-relief formations on the deeper ter? race. The outer fore reef includes a steep inner- reef slope, a sand trough and an outer coral ridge. The steep fore-reef slope drops off at the top of the outer ridge. Many topographic features are comparable to those present on north Jamaican reefs. Corals of the genus Acropora suffer heavy damage but also gain wide distribution from storm swells. Halimeda plates dominate the coarse fraction of the sand substrates across the entire transect even far below the occurrence of the alga. Submarine pavement lithification is most pro? nounced in areas of low sediment accumulation. The outer ridge, although now dominated by Klaus Riitzler, Department of Invertebrate Zoology, and Ian G. Macintyre, Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, D. C. 20560. Acropora cervicornis on the study transect, appears to be built up by more solid framework. The reef at Carrie Bow Cay has a community composition representative of the central barrier reef province to which it belongs, but structurally it is a com? posite, including features characteristic of the discontinuous northern and southern province reefs. Introduction The barrier reef complex?10-32 km wide (James et al., 1976) and approximately 250 km long?off Belize, Central America, is said to be the largest continuous reef in the Caribbean Sea (Smith, 1948; Adey, 1977). Beginning as a fring? ing reef off the Pleistocene peninsula of Ambergris Cay, it extends southward into the Gulf of Hon? duras. This reef complex consists of an almost unbroken barrier reef and numerous patch reefs and mangrove cays in its shoreward lagoon. The shelf lagoon is 20-25 km wide in the northern section of the complex, but where the lower third of the reef bends eastward towards Gladden Spit, the lagoon becomes more than 40 km wide before it opens into the Gulf of Honduras. The point at which the reef complex bends eastward is marked by two islets, one of which?Carrie Bow Cay?is the subject of this volume (Figure 1; Plate 1). Carrie Bow Cay (16?48'N, 88?05'W), known as Ellen Cay up to 1944, is situated on top of the barrier reef proper, 22 km southeast of Dangriga (Stann Creek) (Figure 2) and 18 km east of Sittee 10 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 1.?Tobacco Reef looking south toward South Water and Carrie Bow cays; note cuts isolating Carrie Bow Cay from barrier reef trend. NUMBER 12 11 88?W :':' Blue Ground Range 1 Tobacco Range ?'.'.'.?? Twin Cays Lagoon Pafch Reefs ?/'.?'-?'?? BiRS -.5 ?':'.'p South Water C a S Carrie M Ca) Carrie Bow Cut jM\Cur\e\ ^ ^ Bank Curlew Cut : ; : ; : : : ^ ^ u i ? - - - : v : ?;Wee Wee Cay ?:vU:.5fc": : # & * ' * * ? ^ 2 FIGURE 2.?Index map of barrier reef complex in the vicinity of Carrie Bow Cay; area of larger map located on inset by rectangle. 12 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Point, which is the nearest land. Shoreward from Carrie Bow Cay, the eastern portion of the lagoon (on the barrier platform) is less than 5 m deep whereas the western part is as much as 20 m deep. The reef at Carrie Bow Cay (Figure 3; Plate 1: center right) is separated from the main barrier trend by two channels, South Water Cut to the north (0.4 km wide and 4 m deep) and Carrie Bow Cut to the south (0.7 km wide and 5 m deep). These and other channels through the barrier reef allow oceanic waters from the Car? ibbean Sea to flow onto the shallow barrier plat? form and to transport platform sediments to deep off-reef areas. South Water Cut also separates Carrie Bow Cay (120 m long and 40 m wide) from the larger South Water Cay (440 m long and 100 m wide), a low island less than 1 km to the north that, like Carrie Bow Cay, is composed of reef rubble and sand. Twin Cays, locally known as Water Range, is a swampy mangrove island 2 km to the north? west of Carrie Bow and is about 1 km in diameter and is divided by a meandering canal (Figures 3, 28 a). Approximately 0.5-1.5 km west and south? west of Carrie Bow Cay and within the range of strong tidal currents passing through breaks in the barrier reef, numerous patch reefs having low relief occur among Thalassia seagrass in the shal? low water (3-6 m) of the lagoon. Similar coral build-ups known as "sand bores" that reach the water surface and that are topped by intertidal sand are clustered about 3 km to the southwest. Although many aspects of the Belize Shelf have already been studied and described?most nota? bly by Stoddart (1963), Wantland and Pusey (1971), Purdy (1974), Purdy et al. (1975), and James and Ginsburg (1978)?most earlier work has been geologically oriented. In contrast, our description of reef zonation is based on detailed information about composition of flora and fauna FIGURE 3.?Aerial view of Carrie Bow Cay looking northwest. South Water Cay (right) is a reef island on the barrier trend; Twin Cays (left) is a mangrove island in the shallow lagoon. NUMBER 12 13 as well as topographic and substrate characteris? tics. Our study focuses on a 650 m research transect (just north of Carrie Bow Cay) that extends west to east from the lagoon to the deep fore reef and is representative of the entire barrier reef (Burke, herein: 509). ACKNOWLEDGMENTS.?We are indebted to I. Jewett for patiently preparing the maps and graphs, some of which are based on low-altitude helicopter aerial photographs taken by Captain G. Thomas. H. Adolfi, A. Antonius, G. Bretschko, M. Carpenter, R. Larson, B. Spracklin, and P. E. Videtich assisted in establishing and surveying the main transect and in field mapping other reef and lagoon habitats. A. Antonius, W. T. Boykins, R. B. Burke, M. R. Carpenter, and K. Sandved assisted with photography. Methods Vertical overlapping aerial photographs of the Carrie Bow Cay environs were made by a heli? copter-mounted 70 mm camera with timed mo? torized film advance. Black and white or color prints (12 X 12 cm) were used to make photo composites (for instance, Frontispiece) from which topographic features could be traced on transparent overlays. The scales of the original maps, shown here in reduced figures, are 1: 15,000 (Figure 2), 1:1500 (Figure 4), and 1:800 (Figure 5). Low-altitude aerial photographs (like Figure 8) were taken by balloon-suspended camera (Riit? zler, 1978a). The bottom topography in depths greater than eight meters is not discernible on aerial photographs and therefore had to be ascer? tained from underwater wide-angle photographs, as well as compass, depth-gauge, and tape mea? surements. Sediment thickness in each zone was determined with the aid of a steel probe (1 cm diameter) that could be extended in 3 m sections. The study transect is approximately 200 m north of Carrie Bow Cay, midway across and perpendicular to the crescent-shaped reef crest that half encloses the island (Figure 4). Metal stakes driven into the coral and submerged and surface buoys anchored permanently mark the transect along its projected length (not bottom contour) of 650 m. The zero point was established arbitrarily in a seagrass community in the shallow lagoon. Zones were determined on the basis of bottom configuration, substrates, and relative abundance of predominant organisms. Surveys were carried out along both the transect line and a 50 m strip on either side of it. Relative abun? dance of organisms within zones was measured by pointcounting organisms inside a 0.25 m2 frame randomly positioned (Riitzler, 1978b: 310, fig. 3). Objects overlayed by a 16-intersection grid (line stretched across every 10 cm) were recorded. Vertical projections of the points were used where the three-dimensional configuration of the sub? strate did not permit direct contact. In areas of high bottom relief, this method was more reliable and efficient than surface coverage estimates em? ployed by Riitzler (1975:206) for it avoids distor? tions due to nonhorizontal substrate surfaces (Dahl, 1973). The Transect On the basis of dominant biological and geo? logical characteristics, the barrier reef along the transect off Carrie Bow Cay can be divided into five major units: lagoon, back reef, reef crest, inner fore reef, and outer fore reef (Figure 5; Table 1). Each unit except for the reef crest can be subdivided into distinct zones. The movement and depth of water apparently are the main factors controlling the biological and geological zonations of this area. The lagoon unit (1.5-2.0 m depth) is marked by weak currents and a significant accumulation of fine sand and silt; the back reef (0.1-1.0 m) is subjected to strong cur? rents and the lagoonward transportation of sedi? ments; the water over the intertidal reef crest is in an almost constant state of agitation; the inner fore reef (1-12 m deep) similarly is strongly af? fected by both storm waves and waves related to normal trade wind conditions; the outer fore reef (at least 12 m deep), on the other hand, is affected only by long-period storm waves. Following is a detailed description of each unit. LAGOON.?This environment consists of a sand and seagrass {Thalassia) zone (Figure 6) and a 14 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES 5and and Rubble Pavement Rock Scattered Coral Dense Coral Thalassia 0 FIGURE 4.?Map of sea-floor characteristics of Carrie Bow reef with location of permanent study transect. NUMBER 12 15 sand and rubble zone (Figure 7) that together extend only 50 m along the transect line. The first section, which extends for about 20 m from the zero point northwest of Carrie Bow Cay, covers a patch of seagrass {Thalassia testudinum Banks ex Kbnig, along with a few Syringodium filiforme Kiitzing) on a silty sand bottom (Figure 6a; Plate 2: top left). This sand is fair sorted, dominantly very fine to medium, and 1.0-1.2 m thick. It consists mainly of coral, mollusks, benthic Foraminifera (including Homotrema), and Halimeda plates. During one of our surveys (May 1975) the lower 10-15 cm of the plants were buried in sediments, presumably as a result of hurricane Fifi in 1974. The alga Dictyota sp., which was sparse in 1975, was very common during May and June of 1976, 1977, and 1978. Other algae regularly interspersed with Thalassia belong to the genera Halimeda, Udotea, and Peni- cillus. Empty conch shells and other rubble are covered by algal felts and patches of crustose coralline algae. Uncommon but conspicuous in? vertebrates include the anemone Bartholomea an? nulata (Lesueur) inside conch shells, the corals Acropora cervicornis (Lamarck) and Siderastrea radi? ans (Pallas), the gastropods Strombus gigas Lin? naeus and Turbinella angulata Lightfoot, the echi- noid Tripneustes ventricosus (Lamarck), and the hol- othuroid Holothuria mexicana Ludwig. Benthic ma- croinvertebrates associated with this Thalassia community were studied by Young and Young (herein: 115). Between 20 and 50 m along the transect, sand appears along with rubble (sand and rubble zone) largely covered by algal felts (13%) and by Dictyota spp. (6%) (Figure 7). The poorly sorted sediment in this zone ranges in size from silt to gravel and its composition is similar to that of sediments in the Thalassia zone. The metal probe recorded sediment thickness of 1.0-1.4 m over a hard bot? tom. Acropora cervicornis was absent from the sand and rubble zone in 1972, appeared in only two percent of samples by 1975 (but was very com? mon in patches nearby, particularly just north of the transect), and by 1977 had taken over exten? sive areas on both sides of the transect (Figure la). A subsequent survey in spring 1979 after hurricane Greta (September 1978) showed that thickets of A. cervicornis had been broken up and carried deep into the lagoon by storm surge. New growth, however, was evident on some coral branches that had not been completely buried in sediment. During this survey, thick growths of the red alga Champia parvula (C. Agardh) Harvey (Norris and Bucher, herein: 201) appeared in patches throughout both the Thalassia and the sand and rubble zones (Figures 6b, lb). BACK REEF.?This unit, in which the bottom rises steadily from an average of 1 m to the intertidal reef crest, consists of a patch-reef zone (Figures 8, 9) and a rubble and pavement zone (Figure 11) that together extend from the 50 m mark to the 245 m mark along the transect. The substrate in the patch-reef zone (50-150 m marks) consists of gravel scattered in a poorly sorted silt to very coarse sand. The coarse fraction is com? posed mainly of Halimeda, coral, Homotrema and other benthic foraminiferans, mollusks, crustose coralline algae, and echinoids. Maximum sedi? ment thickness recorded with the metal probe was 0.3 m. This zone is characterized by local build-ups of dead coral framework forming iso? lated patch reefs that consist primarily of Montas- trea annularis (Ellis and Solander) and some Diplo? ma labyrinthiformis (Linnaeus) and Agaricia agaricites (Linnaeus) (Figure 9). Approximately 50 percent of the surface area of the coral is dead and overgrown by crustose coralline algae and algal turfs (Figure 8). Characteristically, the vertical sides of many coral heads and boulders are alive but the top surfaces are dead, evidently because exposure at low tide has limited the upward growth of the corals. At the same time, consider? able damage to this coral population is being caused directly or indirectly by the blue-green alga Oscillatoria submembranacea Ardissone and Strafforello (Antonius, 1973, 1977). Overhangs and crevices are commonly populated by the sea urchin Diadema antillarum Philippi, which, along with several species of parrot fishes, is responsible for extensive bioerosion of dead coral surfaces typical of this zone. Clusters of the corals Acropora 16 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Transect Across Barrier Reef at Carrie Bow Cay, Belize Lagoon Sand a Sea Grass Zone Sand a Rubble Zone Back Reef Patch Reef Zone 100 Rubble a Pavement Zone I2OO Reef Crest Inner Fore Reef High Spur a Groove Zone I3OO L00 - 10 - 2 0 - 3 0 .'.' ? : Sand and rubble / / / / / / Coral rock * // /<*' Tf-???????///????- III" ?? ?? ??????///????? \t^^\\ Mounds or spurs 8 ? Corol heads ???????:? . ? . ??? . ? / / / " ? 5 " ' " ^ ""??&!- ? Plaly corol ^ M / T h " 1 0 " 1 0 testudinum ft P\ Ponfera \ P i \ M i l , e P ? r a complanata %i Millepora- Agaricia- Pontes ?= -^ - Acropora palmato FIGURE 5.?Map and bottom profile of the permanent study transect with zonation terminology (? ? permanent markers). cervicornis and Pontes astreoides Lamarck, and the octocoral Plexaura spp. are dispersed throughout the sandy flats (Figure 8), with a variety of biota including Dictyota sp., Cliona caribbaea Carter, Thal? assia testudinum, Strombus gigas, Udotea flabellum (El? lis and Solander) Lamouroux, Amphiroa sp., and Penicillus capitatus Lamarck?in order of abun? dance?growing in between. In the spring of 1979, as a result of hurricane Greta, most of the A. cervicornis showed extensive damage, and drifts of A. cervicornis debris were piled up against the patch reefs (Figure 10; Plate 2: center left). The corals Acropora palmata (Lamarck), Pontes pontes (Pallas), Diploria strigosa (Dana), Siderastrea siderea (Ellis and Solander), and Agaricia fragilis Dana were observed close to this zone but were not included in the counts. The rubble and pavement zone appearing be? tween 150 and 245 m along the transect consists of gravel in a silt to very coarse sand matrix that grades into a relatively smooth and undulating rock pavement adjacent to the reef crest (Figure 11). As can be seen on the map (Figure 4) the pavement is generally 40 m wide and forms the shoreward border of the entire reef crest trend. This pavement (described in detail by James et al., 1976) consists of a conglomerate of coral {Millepora), mollusk, and crustose coralline algal fragments lithified by a magnesium calcite sub? marine cement. We noted a maximum pavement thickness of only 4 cm although James et al. (1976) reported lithification down to 0.5-1.0 m in other areas of the barrier reef off Belize. Rock surfaces in this zone are overgrown by isolated NUMBER 12 17 ;y. *,'"} Low Spur a Groove Zone Uoo Outer Fore Reef Inner Reef Slope Sand Trough 1600 Outer Ridge Fore Reef Slope meters _ V Acropora cervicornis Gorgonocea E coral heads {Siderastrea siderea and Pontes as- treoides), scattered Dictyota spp., coralline crusts, and the boring sponge Cliona caribbaea. Less abun? dant organisms are algae of the genera Halimeda, Caulerpa, and Penicillus and the corals Agaricia agaricites, Diploria clivosa (Ellis and Solander), and Acropora palmata. The depth of water in this zone averages 0.6 m. Acropora cervicornis was abundant in this zone in spring 1978 (Figure 11a) but had nearly disappeared by the 1979 survey (Figure 1 lb), apparently owing to storm surges associated with hurricane Greta, which transported almost all living A. cervicornis lagoonward. Due south from the 150-200 m marks, between the transect and the island, the same zone changes into a Thalassia-dom'mated rubble flat, with Pontes por- ites, Siderastrea radians, crustose corallines (on rub? ble), and Diadema (Plates 1: bottom right, 2: top right). REEF CREST.?A transition zone (0.2 m deep) occurs between the back reef and fore reef at the 245-265 m marks along the transect. This zone is distinguished by a framework of dead (60%) and living Acropora palmata along with Agaricia agaricites and Pontes astreoides (Figure 12; Plate 2: bottom left, bottom right). The bottom consists mainly of rubble and rock pavement covered by crustose coralline algae along with patches of fine to medium, well- sorted sand, the coarser fraction of which is mainly Halimeda, coral, Homotrema, and crustose coralline algae. This sand supports scattered growths of Dictyota sp. whereas the dead coral rock is commonly covered by algal turfs and dense stands of Caulerpa racemosa (Forsskal) J. Agardh. The section of the reef crest unit between 265 and 270 m along the transect (depth of 0.1 m) is dominated by the coral Acropora palmata, the hydrocoral Millepora complanata Lamarck, and the 18 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES TABLE 1.?Relative abundance of dominant organisms on the Carrie Bow Cay main transect (column heads designate region, structural and biological zone, transect position in meters from 0, depth in meters as average or range, and substrate in order of abundance; + + + = Biota Fleshy macro-algae Calcareous macro-algae Crustose Corallinacea Algal felts and turfs Thalassia testudinum Massive Demospongea Cliona spp. Millepora alcicornis M. complanata Palythoa caribaeorum Stephanocoenia michelinii Madracis mirabilis Acropora cervicornis A. palmata Agaricia agaricites A. fragilis A. lamarcki A. tenuifolia Leptoseris cucullata Siderastrea siderea Porites astreoides P. porites furcata Diploria clivosa D. labyrinthiformis D. strigosa Colpophyllia nalans Montastrea annularis M. cavernosa Meandrina meandrites Dichocoenia stokesi Dendrogyra cylindrus Mycetophyllia danaana Eusmilia fasligiata Erythropodium caribaeorum Erect Gorgonacea spp. Strombus gigas Diadema antillarum Didemnidae Lagoon Sand-mud Thalassia 0- 2 M 20 uddy sand ++ ++ + + + P Sand- rubble Algal felts 20-50 1.8 Sand rubble ++ +++ + Back reef Patch reef- sand Montastrea- Diploria 50-150 1 Sand rock rubble + + ++ + + + ++ P ++ P P ++ P +++ P +++ ++ + ++ Rubble- pave? ment Siderastrea- Porites 150-245 0.6 Rubble rock sand +++ ++ ++ + ++ P + ++ + + + +++ + + P P + + Coral rock- sand Acropora- Agaricia 245-265 0.2 Rock sand rubble ++ ++ ++ + + -r-r-r- -r-r-t- + ++ + Reef crest Coral rock- rubble Acropora- Millepora 265-270 0.1 Rock rubble + ++ + + + + + + + + + + + +++ ++ Rubble- coral rock Corallines- Millepora 270-275 0.1 Rubble rock + + + + + + + ++ ++ + + NUMBER 12 19 very common (>10% presence); + + = common (5%-10% presence); + = less common ( 1 % - 5% presence); P = present and obvious but not encountered in the statistical samples) Coral pin? nacles Millepora- Acropora 275-330 1.8 Rock sand rubble + + Inner fore reef Spur High relief Agaricia- Acropora 330-410 5 Sand rock rubble + + and groove Low relief Gorgonacea- Montastrea 410-550 10 Sand rubble rock + + Inner reef slope Acropora- Montrastrea 550-575 15-22 Rubble + + Sand trough Gorgonacea- Montastrea 575-615 23 Sand rubble Outer fore reef Outer ridge Inner slope Acropora- Diploria 615-625 12-22 Rubble + + + Top Acropora- Gorgonacea 625-645 14 Rubble sand + + Fore reef slope Montaslrea- Gorgonacea 645-655 15-30 Rock rubble sand + + + +++ +++ ++ + + + + + + + + + + + + + + p p + + + + + + + + + + + + + + + + + + + + + + + + p p + + + ++-f- + + 4- + + + + + + + +?+ + + + + + + + + + + + + + + + + + + +++ ++ p p ++ ++ ++ +-1- +++ p p + + + + + + + + + + + + + + + + + ++-1- + + + + + + + + + ++ ++ ++ +++ p p ++ ++ p +++ +++ ++ ? I ? I ? I ? ++ +++ 20 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 6.?Sand and seagrass zone before and after hurricane Greta: a, March 1978; b, April 1979. (Note extensive development of epiphytic red alga Champia parvula on Thalassia testudinum blades; scale = 40 cm.) FIGURE 7.?Sand and rubble zone before and after hurricane Greta: a, March 1978, stands of Acropora cervicornis on plain substrate; b, April 1979, heavy red algal growth (Champia parvula) on substrate devoid of living A. cervicornis. (Scale = 40 cm.) zoanthid Palythoa caribaeorum Duchassaing and Michelotti (Figure I3a,b; Plate 3: top left, top right). Coralline algal crusts, algal turfs, and Por? ites astreoides are attached to massive, extensively bored coral rock. The last windward section of the intertidal reef crest unit (270-275 m along the transect) is built up by rubble and coral rock (60%), covered by coralline crusts, Palythoa cari? baeorum, Millepora complanata, some Porites porites, Agaricia agaricites, algal felts, and extensive crusts of the alcyonacean Erythropodium caribaeorum (Du? chassaing and Michelotti). Just south of the tran? sect line and at four or more other locations the reef crest is interrupted by perpendicular chan? nels (Figure 13 r%. NUMBER 12 25 FIGURE 14.?Coral rubble storm ridge, south of the transect, photographed after hurricane Fifi (1974); rubble prograding over back-reef pavement, partly burying Montastrea annularis colony. Bow Cay are formed by build-ups of dead Acropora palmata, the rest by live Millepora complanata, A. palmata, and Agaricia tenuifolia Dana. The dead framework is extensively covered by encrusting Cliona caribbaea, coralline algae, and some Palythoa caribaeorum. Diploria strigosa, Porites astreoides, P. porites, Montastrea annularis, and M. cavernosa (Lin? naeus) are massive corals of secondary impor? tance. The spur and groove zone of high relief (330- 410 m along the transect) has an average depth FIGURE 13.?Seaward margin of reef crest: a, Millepora com? planata, with Porites porites (foreground) and Acropora palmata (background); b, A. palmata community; c, Reef channel through Millepora ridge looking toward shallow inner reef crest (note overturned live colony of A. palmata in fore? ground). (Scale frame = 50 X 50 cm.) of 5 m, but depth ranges from 3 to 10 m between the highest coral spurs and deepest sand grooves (Figure 16; Plate 3: center right). This well-defined zone has been discussed by Wantland and Pusey (1971) and has been described in detail by James et al. (1976). The high-energy oscillating move? ment of water in this zone has promoted coral growth on the spurs (Shinn et al., herein: 63) and has caused erosion in the grooves. The but? tresses are characterized by foliate Agaricia tenui? folia and Millepora complanata enclosing clusters of Porites porites (Figure 17; Plate 3: bottom left). The tops of many buttresses are dominated by stands of Acropora palmata and A. cervicornis (Figure 18) and their flanks and the sand grooves in between by Agaricia agaricites, Diploria strigosa, various spe? cies of gorgonians (Muzik, herein: 303) and fleshy 26 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 15.?Pinnacles of Millepora complanata and Agaricia tenuifolia in an area of transition between reef crest and high-relief spur and groove zone. green algae (Figure 16; Plate 3: center right, bottom right). Of lesser importance are Siderastrea siderea, Pontes astreoides, Dichocoenia stokesi Milne Edwards and Haime and Stephanocoenia michehnii Milne Edwards and Haime. A characteristic and abun? dant?but quantitatively unimportant?compo? nent of this fauna is the purple hydrozoan Stylaster roseus (Pallas) that occurs in niches and overhangs along the sides of the buttresses. Another hydro? zoan, Millepora alcicornis Linnaeus, commonly grows over the dead skeletons of gorgonians, mainly of Gorgonia ventalina Linnaeus. The me? dium sand to gravel-sized sediment in the grooves has a maximum thickness of 0.3 m. It has fair sorting and is composed mainly of coral, Halimeda, Homotrema, and mollusk and echinoid debris. The spur and groove system of low relief ex? tends between 410 and 550 m along the transect (average depth, 10 m). A diverse population of gorgonians dominates the rock and rubble sub? strates of the sand flats as well as the low (about 1 m relief) coral spurs, which are formed of Montastrea annularis, M. cavernosa, Acropora cervicor? nis, and Diploria strigosa (Figure 19, 20; Plate 4: top left, top right). A few island-like coral pinnacles attain 3 m in height and diameter. Massive De- mospongea become quantitatively important in this zone of reduced agitation of near-bottom water, for example, Neofibularia nolitangere (Du- chassaing and Michelotti), Callyspongia spp., Aply- sina spp., Geodia neptuni (Sollas), Ircinia spp., as well as the thickly encrusting Anthosigmella vahans (Duchassaing and Michelotti). Halimeda spp. and Dictyota spp. are common algae. Conspicuous cor? als having patchy distribution are Agaricia agari? cites, A. tenuifolia, Diploria labyrinthiformis, Dendro- NUMBER 12 27 FIGURE 16.?High-relief spur and groove zone, having 5 m relief. (Water depth at sand bottom: 6 m.) gyra cylindrus Ehrenberg, and Eusmilia fastigiata (Pallas). Probes of the sand-filled grooves indi? cated a maximum depth of 1.2 m at the shallow end of this zone. Seaward, these sand lenses thin out and eventually give way to sand pockets in a rock pavement. The sediment is similar to that found in the high-relief spur and groove zone. James et al. (1976:532) referred to this low-relief spur and groove zone as the "deep spur and groove," which was described as being separated from a "shallow spur and groove" system by a "rubble-covered terrace." There is no indication, however, that a terrace separates the high- and low-relief spur and groove zones off Carrie Bow Cay. In fact, the spurs having low relief are commonly a continuation of the shallower high- relief spurs, with one or more shallow grooves spilling into one of the deeper and wider grooves. Furthermore, the relief of the spurs is considerably less than that reported by James' group who described spurs having relief of 3-4 m that rise to within 2 m of the surface of the waters. OUTER FORE REEF.?This region begins with a 25-degree slope?the inner reef slope?where the transect drops from 15 m depth at 550 m along the transect to 22 m at a position 575 m along the transect. Most of the bottom is covered by a thicket of living and dead Acropora cervicornis that offers substrate to some massive sponges?Veron- gula gigantea (Hyatt), Callyspongia vaginalis (La? marck)?and various gorgonians. Columnar col? onies of Montastrea annularis at the top of the slope give way to large platy colonies towards the base that are accompanied by Porites astreoides, Sideras? trea siderea, and Agaricia tenuifolia (Figure 21; Plate 4: center left). Very poorly sorted sediment com- 28 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 17.?Dominant framework builders of buttresses in high-relief spur and groove zone: a, Agaricia tenuifolia capped by Acropora palmata; b, close-up view of characteristic coral association consisting of A. agaricites, Millepora complanata, and Porites porites. (Scale = 40 cm.) FICURE 18.?Acropora palmata development on top of buttress in high-relief spur and groove zone. (Depth: 4 m.) NUMBER 12 29 FIGURE 19.?Low-relief spur and groove zone with scattered coral heads (foreground: Montastrea annularis) among octo corals on the rock pavement of a low spur. (Depth: 10 m.) FIGURE 20.?Sand-covered hard ground dominated by oc- tocorals with islands of coral heads, near outer edge of low- relief spur and groove zone. (Depth: 13 m.) posed of silt to medium-sized sand with scattered coarser debris, occurs in patches on this slope. Halimeda plates, mollusks, and echinoid spines make up most of the readily recognized coarse fraction. This inner reef slope is comparable to "the steep coral-veneered rock slope" that James and Ginsburg (1978:33-35) called the "reef step." The next section along the transect is a sand trough 40 m wide (575-615 m marks) and an average of 23 m deep (Figure 22). The substrate is a poorly sorted, silt size to very coarse sand sediment plain. This sediment is mainly very fine to fine sand, but coarser material consisting of 30 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 21.?Platy coral development (Montastrea annularis) at base of inner reef slope; note brick with coral transplant (Graus and Macintyre, herein: 441). (Depth: 21 m; scale = 40 cm.) Halimeda plates, mollusks, benthic foraminiferans, and echinoids is scattered throughout. Probings indicated that sediment varies in thickness from about 1 m at the toe of the inner reef slope to more than 12 m in the axis of the trough. Pieces of rubble support gorgonians and sponges, as well as some corals that form several small, isolated coral patches, predominantly of Montastrea annu? laris and M. cavernosa (Figure 22a). This sand trough zone on the transect also encompasses one tall coral pinnacle in the western part of the trough that has a coral-gorgonian composition similar to that of the inner slope with which it is connected by a low Acropora cervicornis ridge (Fig? ure 4). This zone correlates with the seaward- dipping sediment terrace off Tobacco and But- tonwood cays, where the slope is not bordered by an outer ridge (James and Ginsburg, 1978). At Carrie Bow Cay an outer ridge runs parallel to the intertidal reef crest and delineates the continental shelf (Figure 4). On the transect it is formed mainly by a thicket of Acropora cervicornis (Figures 22b, 23a), but south of the Carrie Bow transect Montastrea annularis becomes the principal framework builder (Figure 23 b). The steep 45- degree landward slope of the outer ridge (Figure 22b) supports, among the branches of A. cervicor? nis, massive Diploria labyrinthiformis, Porites as- treoides, and Stephanocoenia michelinii, gorgonians, large sponges?Pseudoceratina crassa (Hyatt), Xes- tospongia sp.?and conspicuous algae?Halimeda spp., Slypopodium zonale (Lamouroux), Peyssonelia sp. The top of the outer ridge lies between 625 and 645 m along the transect and its depth ranges from 12-14 m (Figure 23a). Acropora cervicornis and gorgonians are the dominant organisms, with NUMBER 12 31 - 4* a "/??**? , ^ - * ?'>-:-? & FIGURE 22.?Sand trough: a, isolated coral patches at the base of inner reef slope, 23 m deep; b, view from the top of the outer ridge over Acropora cervicornis thicket, 12 m relief. 32 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES NUMBER 12 33 FIGURE 24.?Fore-reef slope: a, looking up sand shoot from 25 m; b, platy coral-octocoral community on transect at 30 m. local accumulations of Montastrea annularis, Agari? cia agaricites, A. tenuifolia, and Madracis mirabilis (Duchassaing and Michelotti). Fleshy green al? gae, and sponges of the genera Aplysina, Verongula, Callyspongia, Xestospongia, and Agelas are common associates (Plate 4: center right). Poorly sorted, very fine sand to gravel-sized sediments that are rich in Halimeda plates occur in small, 1 m deep de? pressions scattered along the crest of the ridge. An identical shelf-edge ridge occurs off South Water Cay (James and Ginsburg, 1978). Approximately 645 m along the transect, the fore-reef slope (Figure 24) drops at an angle of 50?-70? from depths of 14 m down to 30-60 m, FIGURE 23.?Outer ridge: a, coral community dominated by Acropora cervicornis on transect line, 12 m depth; b, platy Montastrea annularis and Porites astreoides framework 100 m south of transect, 14 m depth. (Scale = 40 cm.) 0 - 2 0 - 4 0 - 1 6 01 E 8 0 - 100- 120- / WJ 1 j j ; A \ , Shallow Reef / Ikm m FIGURE 25.?Profile and dominant sea-floor characteristics of fore reef off South Water Cay (adapted from James and Ginsburg, 1978). 34 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 26.?Close-up view of representative fore-reef slope community of sponges, corals, and octocorals, including (clockwise from center bottom) Cinachyra kuekenlhali, Amphimedon compressa, Agaricia lamarcki, Mycale sp., Iciligorgia schrammi, Porites porites, Pseudopterogorgia sp., and encrusting Millepora sp. (Depth: 30 m.) where it diminishes to 40?-50?. According to previous observations from the research submers? ible Nekton off nearby South Water Cay (James and Ginsburg, 1978), a vertical wall continues from a depth of 60 m to about 110 m where a talus of large reef limestone blocks, coral rubble, and Halimeda sand grades into a gently sloping lime-mud bottom at about 200 m (Figure 25). Our surveys, which were restricted to a maximum depth of 30 m, show an abundance of platy Montastrea cavernosa and M. annularis, Gorgonacea, Agaricia fragilis, Leptoseris cucullata (Ellis and Solan? der), and Demospongea (Figures 246, 26, 27; Plate 4: bottom left, bottom right). Commonly asso? ciated are A. agaricites, A. lamarcki Milne Edwards and Haime, Porites porites forma furcata Lamarck, Diploria labyrinthiformis, Colpophyllia natans (Hout- tuyn), and Mycetophyllia danaana Milne Edwards and Haime. Abundance of gorgonians decreases with depth whereas that of sponges increases. A conspicuous gorgonian under 20 m is Iciligorgia schrammi Duchassaing (Figure 27). Quantitatively important sponges are rope-shaped species of Aplysina, Niphates, and Haliclona, tubular Agelas sp. and Verongula sp., massive Ectyoplasia ferox (Du? chassaing and Michelotti) and Cinachyra kueken? lhali Uliczka, and coral-eroding Cliona delitrix Pang and Siphonodictyon coralliphagum Riitzler. Hal- NUMBER 12 35 FIGURE 27.?Vertical fore-reef slope with platy development of Montastrea cavernosa and abundant sponges and octocorals. (Depth: 40 m.) 36 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES <-? u to C c T3 ? 5 c ? g (fl u o to "5 u 5 c C 3 3 & ? ^-i o ~ e (fl =o 35 (fl + * + fir? " ^ -a c 2 a P c b ? 5 II u o ~ B S I j 2 .a ? (fl c to _C O >~ C t ? o ^ >< V ? ? Io s S..S K M> Ml a -a . c 3 bo D "n 6S A S * tfl a 8 S I ? "5 T3 tfl a. w> bo + ? - cf -if u o 3 (A (fl H E (fl u s I ^ u- o > 3 o u (fl ? S -l fd 2 O5 3^ s rie r l OQ ^ Si o -?2! ^j CO | s? 1 g>o 1 " ] 2 a &o ? o s ^ <>J ^ > ~ ^ ? 3 i ? O j ^ "*-> ? ^ ^ ^ * t - ) ? C M y* OS > Si co A ; CO r l " 05 K ? ? got- <3 ?i l < | to 13 R t s Is , ^ y , s CM OO r^ CO ?< ? * ?S C- 1 + + + + + + + + + 1 1 + + + + 1 1 + + + Co" CM- CM* cxi TH ? * 1 + + + + + + S " OC? CO- CM CO CO w ' ^ / W ' 1 + + + + + + + + + CM CO CO CM O r?." ^^ c ^ ^ 1 + + + + + + + *?. S- v ^?v Tf TT" r^ iii p~i ?>*< 1 + + + + + + + + + f ^ CC? i n t o m ~?' 1 + + 1 + + + CO n 1 + + + 1 1 1 1 1 1 0 o ?~ V > o h be C (fl 0. 5 0 0 - l < z 0 P u z 3 , ^ L O 00, + + + + + + + + 1 1 1 , s L O O, 1 + ST c\i CM + + + u CO (fl Iw be U c 'C (fl i + + + + co CO + CT) O^ 1 + r?> CO + -^^ CM CO + + (D CO + + + (fl bp (fl 6 u U (fl 6 >^ to (U fc 1 1 + 1 + + 1 + r~ I - I ?J CM + + 00 S ^ in + + + ^?N ID ' t o to (fl s ^ - N /??s ^?s y^^v t o CM r~~ r~ O r-' oS o 1 + + 1 + + + + + + 1 1 1 + + 1 1 1 + + + in o in oq CM o CO o ? ' CO ? s - x + + + 1 + + + + + q oo -< m_ i n i r i (?> o " - ' CO v ~ ' v?' + + + 1 + + + + _^^ _^^^ (?> ? r~~ - ; ?; o S2- ^ ^ ^ 1 + + 1 + + + , s f s f ^ f s p i n cq i n CO ? * CO o v JVJ - w + + + 1 + + + + Co" Co" o^ o^ 1 1 1 1 + + to OJ VI to c g- o ' C 2 s (fl c o c tfl p bo C 3 B s -s (3 ffi o o NUMBER 12 37 imeda-rich, very fine sand to gravel-sized sediment occurs in pockets or small ledges between the living cover of this slope. Other components of this sediment include mollusks, echinoids, corals and benthic foraminiferans. Lagoon Environment Reef-forming organisms and other character? istic and quantitatively prominent associates ob? served on the barrier reef transect off Carrie Bow Cay are all sessile and require stable substrates to keep them from being washed away. In the pro? tected lagoon fine sediments tend to bury such substrates and, by settling on organisms as well, constitute a selective stress factor for the living populations. Thus, we examined the lagoon within a radius of 2 km from Carrie Bow Cay in order to compare its characteristic benthic biota with those of the barrier reef (Table 2). A similar comparison of zooplankton is presented by Fer? raris (herein: 143) and complemented by Riitzler et al. (1980). Although seagrass flats and man? groves were surveyed only qualitatively, patch reefs were examined both quantitatively and qualitatively. The Carrie Bow Cay reef flat (be? tween the cay and the reef crest) is the subject of a separate detailed study (Riitzler, in prep.). SEAGRASS COMMUNITY.?An area of approxi? mately 6 km of lagoon bottom immediately due west of Carrie Bow Cay was found to be less than 6.5 m deep, commonly 4-6 m deep. More than 90 percent of this area is flat soft bottom covered by Thalassia testudinum; the rest consists of rubble, reef patches, and large sponges. Most of the rub? ble originates from the commercial conch Strombus gigas. The shells, abandoned by fishermen in large piles or fields, provide substrate for a variety of algae (for instance, Amphiora fragilissima (Lin? naeus) Lamouroux), sponges {Desmapsamma an- chorata (Carter)), hydrocorals {Millepora sp.), cor? als {Porites sp., Siderastrea siderea), gorgonians {Plexaura sp.), and many less conspicuous orga? nisms. The dark interior of these shells constitutes a well-vented miniature cave habitat because fishermen puncture the spire near the apex where they cut the retractor muscle to remove the soft parts of the snail. Extent of colonization depends on exposure time and position of aperture relative to the sediment substrate. Small reef fishes, en? crusting coralline algae, foraminiferans {Homo? trema rubrum (Lamarck)), sponges {Spirastrella sp., Clathrina sp.), bryozoans and ascidians, as well as anemones {Bartholomea annulata), crabs {Mithrax sp.), and ophiuroids {Ophiotrix spp.) are common inhabitants. Several species of Cliona excavate the walls of old shells, some of which are occupied by the hermit crab Petrochirus diogenes (Linnaeus). Distribution of living conchs is patchy and their depletion in large areas observed by us over a seven-year period indicates over fishing. On the other hand we noted concentrations near certain conch rubble patches comparable to those re? ported by Hesse (1979). Another large gastropod, Turbinella angulata, is a common associate. Sponges are quantitatively the most important organisms in the Thalassia meadows, even outside the rubble fields. In 1975 and 1978 many were found loose and only partly alive, together with gorgonians in similar condition (Muzik, herein: 303) and can be assumed to have been torn from the reef and washed into the lagoon by hurricane surge (Riit? zler and Ferraris, herein: 77). The presence of recently sunken coconut trees confirms this as? sumption. Other sponges, however, were healthy and attached to small pieces of rubble and sea? grass {Desmapsamma anchorata, Iotrochota birotula (Higgin), Aplysina fistularis (Pallas), Niphates erecta Duchassaing and Michelotti) or, like some very large forms (5-40 1 volume), were rooted in sand {Spheciospongia vesparium (Lamarck), Ircinia spp.). PATCH REEFS.?These lagoon reefs of low relief are clustered about 0.3-2.0 km to the west and southwest of Carrie Bow Cay (Figure 2). Favor? able substrate conditions, together with trade- wind and tide-induced currents passing through South Water and Carrie Bow cuts probably pro? moted the development of these structures that now exhibit a richness of reef fauna that is sur? passed only by the outer barrier fore reef. Diver? sity and biomass of sponges, in particular, are higher than in any other habitat of similar depth because most species favor areas having a high rate of water exchange but lack resistance against 38 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES . A _.:> * FIGURE 28.?Twin Cays: a, aerial view of mangrove island showing channel and intertidal mud flats behind red mangrove fringes; b, red mangrove (Rhizophora mangle) on mud bank bordering channel. detachment from substrates by undulating water movement. The patch reefs are circular or oval, some are arranged in string-of-pearl fashion, 5-60 m in diameter and raised 0.2-1.0 m above the surrounding flat sand and Thalassia bottom in depths of 3-6 m. Abundance data were derived from two perpendicular transects across a char? acteristic patch reef, named "Spaghetti Reef," after a new species of stringy sponge of the genus Ulosa (Riitzler, 1981). Sand and coral rock make NUMBER 12 39 FIGURE 29.?Seagrass community of the shallow lagoon at the entrance to Twin Cay channel: Thalassia testudinum, Halimeda incrassata, and Manicina areolata. (Depth: 1 m.) up 45% of the surface area. Gorgonians are the most abundant organisms (20%), followed by cor? als and milleporids (17%), sponges (9%), and Thalassia seagrass (on sand, 9%). Large algae are notably absent. Gorgonia ventalina and Pseudoptero- gorgia spp. are the most conspicuous octocorals on the patch reefs. Massive forms dominate among the corals {Siderastrea siderea, Diploria labyrinthifor? mis, D. strigosa, Montastrea annularis), the most com? mon hydrocoral is Millepora alcicornis, which en? crusts numerous gorgonian skeletons. Sponges, although they score comparatively low in the point counts, contribute most to the standing crop. The principal species are Ircinia spp., Xesto- spongia sp., Iotrochota birotula, Desmapsamma anchor- ata, Amphimedon compressa Duchassaing and Mich? elotti, and Callyspongia spp. (Plate 5: center left, center right). MANGROVE.?Toward Twin Cays from the southeast a belt of shallow (1 m) Thalassia flat grades into the larger entrance of the channel that divides this island (Figure 28a). The flat is a transition zone between the more agitated deeper lagoon and the protected mangrove where moderate tidal currents control water exchange. Most abundant among the Thalassia are some algae (for instance, Penicillus capitatus, Halimeda incrassata (Ellis) Lamouroux), sponges {Tedania ignis (Duchassaing and Michelotti), Oligoceras vio? lacea (Duchassaing and Michelotti), Haliclona vir- idis (Duchassaing and Michelotti)), and the coral Manicina areolata (Linnaeus) (Figure 29). Millepora sp. encrusts large surface areas of submerged wood. Some more sponges, Aplysina fulva (Pallas) and the large loggerhead Spheciospongia vesparium, and numerous starfish {Oreaster reticulatus (Lin- 40 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES naeus)) occur on the bottom of the channel en? trance. The meandering channel is 0.5-2.0 m deep and is lined by red mangroves, Rhizophora mangle Lin? naeus, which grow on intertidal mud banks and extend their arched prop roots into subtidal water (Figure 28.6). In places the mud banks are washed out to form vertical walls and even overhangs and caves. Mud caves, probably of similar origin, are found outside the present mangrove margin just north of Twin Cays. They extend horizon? tally into mud banks surrounding 4 m depressions in the seagrass floor. The caves are large enough to harbor 2 m long sharks. Mangrove roots offer the only substrates that are not subject to accumulation of fine sediments. In this habitat strong competiton for the limited space takes place among diverse flora and fauna (Figure 30). On the light-exposed roots and bank edges, clusters of the algae Caulerpa verticillata J. Agardh and Halimeda spp. compete with the sponges Tedania ignis, Ircinia felix (Duchassaing and Michelotti), and Lissodendoryx sp. (Figure 30a,b; Plate 5: bottom left, bottom right). The walls of the shaded overhangs are dominated by the sponge Ulosa ruetzleri Wiedenmayer and, at the north entrance of the channel, by Mycale sp. Also, locally important in biomass are the anemones Bartholomea annulata and Condylactis gigantea (Wein- land), the tunicates Ecteinascidia turbinata Herd- man and Ascidia nigra (Savigny), and sabellid polychaetes. The intertidal parts of the mangrove roots are dominated by the oyster Crassostrea sp., whereas the Thalassia mud bottom supports a dense population of Scyphomedusae {Cassiopea spp.), with specimens of Oreaster reticulatus inter? spersed (Figure 30 c). Mud caves and roots pro? vide hiding places for a diverse fish fauna (Figure 30-6). Summary and Conclusions This paper presents the first detailed descrip? tion of the biological-geological zonation of the barrier reef complex off Belize. Despite some variation along the barrier reef (see Burke, herein), the zonation of the Carrie Bow Cay segment is typical of the entire reef platform (Figures 4, 31). Except for its large lagoon and greater distance from land, the Belizean barrier reef is comparable to well-investigated fringing barrier reefs off the north coast of Jamaica (Go- reau, 1959; Goreau and Land, 1974). Seaward of the Thalassia-dormm.ted lagoon, the back reef occurs between an area of massive coral heads on rock pavement and the breakers of the reef crest (Figure 31). On our transect we did not consider this "reef flat" as part of the crest (Goreau, 1959:74) because, although this zone is intertidal, it is very narrow and almost always flooded by waves; instead, we defined "reef flats" herein as large intertidal areas be? tween the eastern shores of South Water and Carrie Bow cays and their nearby crests (Larson and Larson, herein; Riitzler, in prep.). Only the narrow intertidal breaker zone is included in the reef crest. The inner fore reef begins at this point with a spur and groove (= buttress) zone of high relief?which in Jamaica is considered a part of the crest (Goreau, 1959)?but it changes abruptly into a gently sloping terrace of spurs and grooves having low relief ("seaward slope" or "upper fore- reef terrace" in Jamaica). The outer fore reef on our transect has a steep inner reef slope, a per? pendicular sand trough parallel to the reef crest, and an outer coral ridge where the fore-reef slope begins to drop off to the deep (vertical) fore reef. The comparable feature off Jamaica is a lower fore reef having a lower fore-reef escarpment. Both the trough and the ridge are missing in Jamaica (Goreau and Land, 1974). Water movement (direction and force) appear to control the development of zonation patterns. Wave action determines not only reef types, by influencing coral zonation (Geister, 1977), but also influences the distribution of all other sed? entary organisms, some of which?for instance, octocorals, sponges, and algae?have consider? able ecological importance. The quantitative sig? nificance of many coral associates is undervalued by most field methods (point and chain-link counts, estimates of area coverage) because pres? ence or surface area is measured but not biomass (massive sponges) or space occupied (swaying NUMBER 12 41 FIGURE 30.?Underwater views of man? grove root system at edge of Twin Cay channel: a, mudbank stabilized by root? lets and overgrown by Halimeda mat; b, juvenile barracuda finding shelter and food among the Rhizophora roots; c, Oreas- ter reticulatus on mangrove roots penetrat? ing sediment bottom of the channel. FIGURE 31.?Carrie Bow reef zonation: above, oblique aerial view looking toward lagoon; below, block diagram of transect. frs frs = fore-reef slope hsg = high spur and groove irs = inner reef slope lsg = low spur and groove or = outer ridge pr = patch reefs re = reef crest rf = reef flat sg = seagrass st = sand trough NUMBER 12 43 gorgonians). Objective evaluation of nonframe- building organisms is essential when comparisons are made between reef biota and nearby lagoon habitats that are not dominated by corals, or only partly dominated by them. We documented an impressive proliferation of Acropora cervicornis in the back-reef and lagoon zones during the four years following the destruc? tion caused by hurricane Fifi in 1974. Our spring 1979 survey indicated that destruction associated with hurricane Greta (September 1978) was mainly the breaking up and lagoonward trans? portation of the shallow-water A. cervicornis. Few colonies of the fragile branching coral escaped this movement and subsequent burial in the sand and rubble zone as well as in the patch reef zone. Almost all of this coral has been washed out of the rubble and pavement zone (Figure 11). Al? though many of the living and partly buried fragments will, in time, develop into large colo? nies, the number of living A. cervicornis in shallow water has been drastically reduced. This constant cycle of vigorous development, destruction, and resurgence gives rise to the commonly observed high proportion of A. cervicornis rubble in compar? ison to living A. cervicornis in many shallow-reef areas. Although the branching corals Acropora palmata and A. cervicornis suffer extensive mechanical dam? age during hurricanes, the transportation and reestablishment of living fragments are significant factors in the distribution of these corals in the shallow-water environment (Plate 2: center left; Highsmith et al., 1980). Similar observations have been made in Florida (Shinn, 1972; Gilmore and Hall, 1976) and in Jamaica (Tunnicliffe, 1980) where the dispersal of A. cervicornis in these reefs was reported to be largely related to asexual reproduction by regeneration of broken and transported branches. Our observations and those of James and Gins? burg (1978) have documented Halimeda as a ma? jor contributor to sediment in the Belizean bar? rier-reef complex. This calcareous green alga forms a major fraction of these sediments, extend? ing from the shallow lagoon down to at least 200 m on the fore-reef slope, well below its living depth range (approximately 100 m). Commonly comprising the dominant component of the ex? tremely coarse fractions (2-4 mm), the readily identifiable calcareous plates of Halimeda are more characteristic of reef-derived sediments than are the fragments of any other organisms, including the corals. The narrow rock pavement that occurs directly shoreward of the reef crest off Carrie Bow Cay (Figure 4) is a characteristic substrate of shallow reef areas that are constantly having their sedi? ment cover swept away by turbulent waters. The dates of 480?90 years and 534?90 years obtained from coral fragments embedded in a pavement off South Water Cay (James et al., 1976) indicate a long period of formation. These dates also support Macintyre's (1977) observation that sub? marine lithification is most highly developed in reef areas of high agitation and/or slow accu? mulation, where the substrate is exposed to nor? mal marine conditions for long periods of time. James and Ginsburg (1978) speculated that a shelf-edge ridge off South Water Cay is a sub? merged reef similar to the relict shallow-water, Late Holocene reefs described by Macintyre (1967, 1972), Adey et al. (1977), and Lighty et al. (1978). In contrast, Burke (herein) proposed that these ridges are active accumulations of the rap? idly growing coral Acropora cervicornis. Burke points out that not only is A. cervicornis dominant on these ridges, but that the ridges along the barrier reef complex are restricted to areas protected from long-period storm waves by the outlying atolls. The difficulty with which we probed this ridge (an average penetration of 1 m, maximum of 1.5 m) in contrast to the ease of probing through Acropora cervicornis at Rhomboid Shoals near Victoria Channel (Macintyre et al., 1977) indicates that this ridge does not have a similar open-frame network. Our earlier observation that Montastrea annularis constructs most of the modern framework of the ridge south of our transect also indicates that this shelf-edge ridge is not merely an accumulation of A. cervicornis. Core samples of the internal structure are needed to establish the 44 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES relative importance of relict and modern frame? work in the construction of this ridge system. The reef off Carrie Bow Cay has a species composition and community zonation that is rep? resentative of the entire barrier reef system (Burke, herein). The structure of shallow zones of this central province reef, however, is somewhat more similar to that of the discontinuous reefs in the northern and southern provinces because cur? rent flow through South Water and Carrie Bow cuts influences the sediment and coral distribu? tion patterns in the back reef and lagoon. The Carrie Bow Cay fore-reef structures, on the other hand, have a degree of development and flourish? ing coral communities that are characteristic of the central province. Literature Cited Adey, W. H. 1977. Shallow Water Holocene Bioherms of the Carib? bean Sea and West Indies. In D. L. Taylor, editor, Proceedings, Third International Coral Reef Symposium, 2:xxi-xxiv. Miami, Florida: Rosenstiel School of Marine and Atmospheric Science. Adey, W. H., I. G. Macintyre, and R. Stuckenrath 1977. Relict Barrier Reef System off St. Croix: Its Im? plication with Respect to Late Cenozoic Coral Reef Development in the Western Atlantic. In D. L. Taylor, editor, Proceedings, Third International Coral Reef Symposium, 2:15-21. Miami, Florida: Rosenstiel School of Marine and Atmospheric Science. Antonius, A. 1973. New Observations on Coral Destruction in Reefs [Abstract]. In [Abstracts of] Tenth Meeting of the Island Marine Laboratories of the Caribbean, page 3. Mayaguez, Puerto Rico: University of Puerto Rico. 1977. Coral Mortality in Reefs: A Problem for Science and Management. In D. L. Taylor, editor, Proceed? ings, Third International Coral Reef Symposium, 2:617- 623. Miami, Florida: Rosenstiel School of Marine and Atmospheric Science. Burke, R. B. 1979. Morphology, Benthic Communities, and Structure of the Belize Barrier Reef. 78 pages. Master's thesis, University of South Florida, Tampa. Chevalier, J. P. 1973. Geomorphology and Geology of Coral Reefs in French Polynesia. In O. A. Jones and R. Endean, editors, Biology and Geology of Coral Reefs, 1:113-141. New York: Academic Press. Dahl, A. L. 1973. Surface Area in Ecological Analysis: Quantifica? tion of Benthic Coral-Reef Algae. Marine Biology, 23:239-249. Geister, J. 1977. The Influence of Wave Exposure on the Ecological Zonation of Caribbean Coral Reefs. In D. L. Taylor, editor, Proceedings, Third International Coral Reef Symposium, 1:23-29. Miami, Florida: Rosen? stiel School of Marine and Atmospheric Science. Gilmore, M. D., and B. R. Hall 1976. Life History, Growth Habits, and Constructional Roles of Acropora cervicornis in the Patch Reef En? vironment. Journal of Sedimentary Petrology, 46:519- 522. Goreau, T. F. 1959. The Ecology of Jamaican Coral Reefs, I: Species Composition and Zonation. Ecology, 40:67-90. Goreau, T. F., and L. S. Land 1974. Fore-Reef Morphology and Depositional Proc? esses, North Jamaica. In L. F. Laporte, editor, Reefs in Time and Space. Society of Economic Pa? leontologists and Mineralogists, Special Publication, 18: 79-89. Tulsa, Oklahoma. Hesse, K. O. 1979. Movement and Migration of the Queen Conch, Strombus gigas, in the Turks and Caicos Islands. Bulletin of Marine Science, 29:303-311. Highsmith, R. C , A. C. Riggs, and C. M. D'Antonio 1980. Survival of Hurricane-generated Coral Fragments and a Disturbance Model of Reef Calcification/ Growth Rate. Oecologia (Berlin), 46:322-329. James, N. P., and R. N. Ginsburg 1978. The Deep Seaward Margin of Belize Barrier and Atoll Reefs. 201 pages. Miami Beach, Florida: University of Miami. James, N. P., R. N. Ginsburg, D. S. Marszalek, and P. W. Choquette 1976. Facies and Fabric Specificity of Early Subsea Ce? ments in Shallow Belize (British Honduras) Reefs. Journal of Sedimentary Petrology, 46:523-544. NUMBER 12 45 Lighty, R. G., I. G. Macintyre, and R. Stuckenrath 1978. Submerged Early Holocene Barrier Reef South- East Florida Shelf. Nature, 275:59-60. Macintyre, I. G. 1967. Submerged Coral Reefs, West Coast of Barbados, West Indies. Canadian Journal of Earth Sciences, 4: 461-474. 1972. Submerged Reefs of Eastern Caribbean. American Association of Petroleum Geologists Bulletin, 56:720- 738. 1977. Distribution of Submarine Cements in a Modern Caribbean Fringing Reef, Galeta Point, Panama. Journal of Sedimentary Petrology, 47:503-516. Macintyre, I. G., R. B. Burke, and R. Stuckenrath 1977. Thickest Recorded Holocene Reef Section, Isla Perez Core Hole, Alacran Reef, Mexico. Geology, 5:749-754. Purdy, E. G. 1974. Reef Configurations: Cause and Effect. In L. F. Laporte, editor, Reefs in Time and Space. Society of Economic Paleontologists and Mineralogists, Special Publication, 18:9-76. Tulsa, Oklahoma. Purdy, E. G., W. C. Pusey III, and K. F. Wantland 1975. Continental Shelf of Belize?Regional Shelf Attri? butes. In K. F. Wantland and W. C. Pusey III, editors, Belize Shelf-Carbonate Sediments, Clastic Sediments, and Ecology. The American Association of Petroleum Geologists, Studies in Geology, 2:1-52. Tulsa, Oklahoma. Riitzler, K. 1975. The Role of Burrowing Sponges in Bioerosion. Oecologia (Berlin), 19:203-216. 1978a. Photogrammetry of Reef Environments by He? lium Balloon. In D. R. Stoddart and R. E. Johan? nes, editors, Coral Reefs: Research Methods. Mon? ographs on Oceanographic Methodology, 5:45-52. Paris: Unesco. 1978b. Sponges in Coral Reefs. In D. R. Stoddart and R. E. Johannes, editors, Coral Reefs: Research Meth? ods. Monographs on Oceanographic Methodology, 5:299- 313. Paris: Unesco. 1981. An Unusual Blue-Green Alga Symbiotic with Two New Species of Ulosa (Porifera: Hymeniacidoni- dae). Marine Ecology, 2:35-50. In prep. The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay, Belize: The Reef Flat Community. Riitzler, K., J. D. Ferraris, and R. J. Larson 1980. A New Plankton Sampler for Coral Reefs. Marine Ecology, 1:65-71. Shinn, E. A. 1972. Coral Reef Recovery in Florida and the Persian Gulf. 9 pages. Houston, Texas: Conservation Depart? ment, Shell Oil Company. Smith, F.G.W. 1948. A Handbook of the Common Atlantic Reef and Shallow- Water Corals of Bermuda, the Bahamas, Florida, the West Indies, and Brazil. 112 pages. Coral Gables, Florida: University of Miami Press. Stoddart, D. R. 1963. Effects of Hurricane Hattie on the British Hon? duras Reefs and Cays, October 30-31, 1961. Atoll Research Bulletin, 95:1-142. Tunnicliffe, V. Z. 1980. Biological and Physical Processes Affecting the Survival of a Stony Coral, Acropora cervicornis. Ph.D. dissertation, Yale University, New Haven, Connecticut. Wantland, K. F., and W. C. Pusey III 1971. A Guidebook for the Field Trip to the Southern Shelf of British Honduras. New Orleans, Louisiana: New Orleans Geological Society. Tides at Carrie Bow Cay, Belize Bjorn Kjerfve, Klaus Riitzler, and George H. Kierspe ABSTRACT The tide at Carrie Bow Cay, Belize, is micro- tidal (mean range of 15 cm) and is of the mixed semidiurnal type. Comparison with conditions at Key West, Florida, indicates that high and low waters off Carrie Bow occur earlier than at Key West by 45 and 2 minutes, respectively. Because of differences in tidal type and meteorological conditions, corrections for height and time differ? ence applied to the predicted tide at Key West yield only approximate tide predictions for Carrie Bow Cay. Introduction Because the tide in the Caribbean Sea is micro- tidal, it might be expected to have little influence on the water flow regime. Velocity measurements indicate, on the contrary, that tidal forcing is a major cause of currents in coastal regions of the Caribbean (Roberts et al., 1975). Study of a shallow reef flat and back reef shows that even small tidal fluctuations have strong influence on the distribution and succession of organisms (Glynn, 1973; Riitzler, in prep.). This note de? scribes, characterizes, and predicts the tide at Carrie Bow Cay (16?48'N, 88?05'W), Belize, as a necessary first step in the investigation of flow and water exchange characteristics as well as Bjorn Kjerfve, Marine Science Program, Department of Geology, and the Belle W. Baruch Institute for Marine Biology and Coastal Research, University of South Carolina, Columbia, S. C. 29208. Klaus Riitzler, Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C 20560. George H. Kierspe, Harbor Branch Foundation, Inc., Fort Pierce, Fla. 33450. intertidal and shallow subtidal communities at this barrier reef location. ACKNOWLEDGMENTS.?Many persons helped to collect the tide data used in this study. We are especially grateful to J. D. Ferraris, Mount Desert Island Biological Laboratory, R. Larson and M. Carpenter, Smithsonian Institution, and J. Greer, R. L. Crout, S. Ferguson, and G. Pickler, Univer? sity of South Carolina. J. E. Fancher and D. C. Simpson of the National Ocean Survey provided valuable advice on the analytical procedures. L. Kjerfve did the drafting. All computer work was performed on the University of South Carolina IBM 370/168 system. The Belle W. Baruch Insti? tute for Marine Biology and Coastal Research has listed this paper as its Contribution Number 269. Measurements, Analyses, and Results A Benthos 2820 submergible in situ tide re? corder, which senses the hydrostatic pressure, was installed below the pier on the west side of Carrie Bow Cay in the barrier reef lagoon. This location is 120 m west of the reef crest, 24 km southeast of Dangriga (Stann Creek). Two major navigation cuts through the barrier reef, less than 1 km away, connect the barrier reef lagoon to the Caribbean Sea. The gauge intake was, on the average, 33 cm off the bottom, 59 cm below water, 153 cm below the top of the cement dock (the local reference datum) and has been operated intermittently since early 1976. Several 29-day tide records were digitized to hourly intervals and subjected to harmonic anal- 47 48 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Carrie Bow Cay Tide (MSL) FIGURE 32.?Comparison of measured and computer-predicted tides at Carrie Bow Cay, Belize, 10 April-8 May 1976. ysis (Schureman, 1940; Dennis and Long, 1971) for the purpose of computing amplitudes and epochs for the 24 major tidal constituents, which correspond to more than 99% of the actual am? plitude. The nine constituents with the greatest amplitudes for a typical 29-day series are com? pared in Table 3 with the same constituents at Key West, Florida. The 24 amplitude and epoch values for Carrie Bow Cay were entered into the National Ocean Survey tide prediction computer program (Pore and Cummings, 1967), which al? lows monthly prediction of both hourly tidal height values and times and heights of high and low water. Results indicated reasonable agree? ment between measured and predicted tides for a 29-day test series (Figure 32). The prediction of time of high and low water is in general more successful than reproduction of water elevation. The 29-day test record, 10 April-8 May 1976, was also tabulated and subjected to high-and-low water analysis on the basis of instructions for Form 2211 (National Ocean Survey, 1974:44-59). TABLE 3.?Comparison of values for the nine major tidal constituents at Carrie Bow Cay, Belize, and Key West, Florida (symbol in parentheses follows each constituent; columns 1 and 2 computed for Carrie Bow Cay using Program "Harmonic" (Dennis and Long, 1971); columns 3 and 4 supplied by D. Simpson, Predictions Branch, National Ocean Survery; columns 5 and 6 from Defant, 1960: 267) Tidal constituents SEMIDIURNAL COMPONENTS Principal lunar Principal solar Larger lunar elliptic Luni-solar Larger lunar evectional DIURNAL COMPONENTS Luni-solar Principal solar Principal lunar Larger lunar elliptic (M2) (&) (N2) (K2) ("2) (* i ) (A) (Oi) (Oj) Carrie Bow Amplitude (cm) 5.7 3.5 2.5 1.0 0.5 7.9 2.6 2.5 0.5 Cay Epoch C) 251 228 243 228 244 185 185 244 273 Key Amplitude (cm) 17.8 5.4 3.5 1.5 0.7 8.9 2.9 9.2 2.2 West Epoch C) 285 302 269 302 271 282 287 284 275 Period (mean solar hour) 12.42 12.00 12.66 11.97 12.63 23.93 24.07 25.82 26.87 Theoretical coefficient ratio (M2 = 100) 100.0 46.6 19.2 12.7 3.6 58.4 19.4 41.5 7.9 NUMBER 12 49 The pertinent results of this analysis are summa? rized in Table 4. Discussion The tide at Carrie Bow Cay is microtidal and of the mixed semidiurnal type. Its mean range is 15 cm and its semidiurnal and diurnal amplitudes are of approximately equal importance (Table 3). The form number F, an amplitude ratio between harmonic constituents, may be used to quantify the tide type (Defant, 1960:306-308). It is defined by F = (Ki + Ox)/{M2 + S2), where Kx is the diurnal luni-solar, 0\ the diurnal principal lunar, Mi the semidiurnal principal lunar, and Si the semidiurnal principal solar component (Table 3). If F< 0.25 the tide is semidiurnal; if 0.25 3.0 the tide is diurnal. The Carrie Bow Cay form number is 1.13 in comparison with 0.75 at Key West, Florida. Although both locations have the same tide type, the diurnal influence is greater at Carrie Bow Cay. From the amplitude of the various harmonics (Table 3) it is possible to compute additional statistics (Marmer, 1954). With respect to the semidiurnal tidal constituents, the spring tide range is 2{M2 + S2) or 18.4 cm and the neap range approximately seven days later of 2(M2 ? S2) or 4.4 cm. The mean semidiurnal tide is 2.2 M2 or 12.5 cm. With respect to the diurnal con? stituents, the tropic tide range measures 2(K\ + Oi) or 20.8 cm, the equatorial range is 2(K\ ? Oi) or 5.4 cm, and the mean diurnal tide is \.b(K\ ? Oi) or 15.6 cm. The values above are only ap? proximate as the P\ and N2 constituents show values significantly larger than could have been expected from the theoretical coefficient ratio based on the magnitude of the constituents' tide- producing forces (Table 3) at this location. Of course, such discrepancies between the magnitude of the tide-producing force and actual response of the water mass is quite common and is due to basin resonance characteristics. The diurnal and semidiurnal partial tides are approximately equal; however, the spring-neap- spring cycle of the semidiurnal tide is 29.5 mean solar days (synodic month) and is related to the phase of the moon. The tropic-equatorial-tropic cycle of the diurnal tide is somewhat shorter, 27.3 mean solar days (sidereal month), and is related to the declination of the moon from the equator. Because of this time difference, several longer cycles are introduced. If the occasional times of high water in M2, S2, N2, K\, P\, and Oi occur simultaneously, the total tide range could be as great as 50 cm. Of course, wind tides are likely to cause extreme sea level changes more so than the astronomical forces. The epoch or phase relative to the Greenwich meridian yields the following information about the inequality of the timing of highs and lows in the component tides (Marmer, 1949). If epochs are expressed in degrees, the difference D = \M?2-{K0iA- 0?i)\ indicates whether tidal in? equality is entirely in the high waters {D ~ 0?), is equally great in the high and low waters {D ~ 90?), or is entirely in the low waters {D ~ 180?). If D is greater than 180?, its value is subtracted from 360?. For Carrie Bow Cay, the D value of 120? indicates inequality in both high and low waters, which is especially pronounced in the low water elevations (refer to Figure 32). This so- called diurnal inequality on the average measures 5.6 cm between low waters and 2.4 cm between high waters (Table 4). The epoch also yields information about the phase age and the diurnal age. The phase age TABLE 4.?Tide statistics for Carrie Bow Cay, corrected for the longitude of the moon's node and changes in the decli? nation of the sun, based on computations using National Ocean Survey Form 2211 for 10 April-8 May 1976 (refer to Figure 32) Tide datum (mean low water)1 Mean range (Mn) Mean tide level (MTL)2 Diurnal high water inequality (DHQ) Diurnal low water inequality (DLQ) Greenwich lunitidal high water interval Greenwich lunitidal low water interval 0 cm 15.0 cm 7.5 cm 2.4 cm 5.6 cm 2.14 h 8.44 h 1 101 cm below local reference datum. 2 Marked "0" on Figure 32, right hand scale. 50 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 33.?Co-tidal lines of the predominant tides for the Caribbean Sea (after Defant, 1960); tide regression indicated by relative arrival time of high or low water, expressed in lunar hours: a, semidiurnal tide, M2 constituent, progresses along the Belizean coast from north to south; b, diurnal tide, K\ constituent, progresses along the Belizean coast from south to north. (Carrie Bow Cay indicated by arrow). refers to the semidiurnal tide and is the lag of the spring tide relative to full or new moon. The phase age is computed as 0.98 {S2 ? M2) and equals ?22 hours for Carrie Bow Cay, which indicates that spring tide leads the new and full moon by 22 hours. The diurnal age, on the other hand, is a measure of the timing of tropic tide relative to maximum declination of the moon. It is computed as 0.91 (A"? ? 0?) and is 0 hours for Carrie Bow Cay, indicating a maximum diurnal tide range at the time of maximum lunar decli? nation. Figure 33, which shows the main Caribbean semidiurnal {M2) and diurnal {K\) amphidromic systems (Defant, 1960) indicates that the M2 tide progresses from north to south along the Belizean barrier reef, whereas the K\ tide progresses in the opposite direction, from south to north, along the reef crest. Because it is inconvenient, though possible, to publish tidal predictions for Carrie Bow Cay, we instead computed correction factors and applied them to predicted tides for Key West, Florida, a National Ocean Survey reference gauge for which daily predictions are published. The Greenwich lunitidal intervals at Key West are 2.89 hours for high water (HW) and 8.48 hours for low water (LW) (D. Simpson, pers. comm.). Comparison of these figures with the lunitidal intervals for Carrie Bow (Table 4) indicates that on the average HW at Carrie Bow Cay leads HW at Key West by 45 minutes, and LW at Carrie Bow Cay leads LW at Key West by 2 minutes. Using the predicted tides for Key West for April and May 1976, we also found that HW at Carrie Bow Cay is 23 cm below HW at Key West and that LW at Carrie Bow Cay is 4 cm above LW at Key West on the average. However, because Carrie Bow Cay and Key West can be expected to experience different meteorological and oceanographic conditions at given times, the resulting tide predictions for Carrie Bow Cay now and then will deviate sig? nificantly from actual tide variations. NUMBER 12 51 Literature Cited Defant, A. 1960. Physical Oceanography. Volume 2, 598 pages. Lon? don: Pergamon Press. Dennis, R. E., and E. E. Long 1971. A User's Guide to a Computer Program for Har? monic Analysis of Data at Tidal Frequency. Na? tional Oceanic and Atmospheric Administration, Technical Report, 4L: 31 pages. Rockville, Maryland. Glynn, P. W. 1973. Ecology of a Carribbean Coral Reef: The Porites Reef-Flat Biotope, Part I: Meteorology and Hy? drography. Marine Biology, 20:297-318. Marmer, H. A. 1954. Tides and Sea Level in the Gulf of Mexico. Fishery Bulletin, 55(89): 101-118. National Ocean Survey 1974. Manual of Tide Observations. National Oceanic and Atmospheric Administration Publication, 30-1: 72 pages. Washington, D.C. Pore, N. A., and R. A. Cummings 1967. A Fortran Program for the Calculation of Hourly Values of Astronomical Tide and Height of High and Low Water. United States Weather Bureau [now: National Weather Service] Technical Memorandum, TDL-6: 17 pages. Silver Spring, Md. Roberts, H. H., S. P. Murray, and J. N. Suhayda 1975. Physical Processes in a Fringing Reef System. Journal of Marine Research, 33 (2): 233-260. Schureman, P. 1940. Manual of Harmonic Analysis and Prediction of Tides. United Slates Coast and Geodetic Survey, Special Publication, 98: 31 pages. Washington, D.C. Riitzler, K. In prep. The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay, Belize: The Reef Flat Community. Water Currents Adjacent to Carrie Bow Cay, Belize Jeffrey E. Greer and Bjorn Kjerfve ABSTRACT Tide, wind, and waves significantly influence the water currents off Belize, as indicated by continuous current-meter measurements at four locations in or behind the barrier reef at Carrie Bow Cay. Tidal currents reaching speeds of up to 40 cm s~ are dominant in the major reef en? trances. In the lagoon behind the reef crest, the water is primarily wind driven, without an ob? vious tidal signature, and reaches maximum speeds of 33 cm s_1. On the leeward side of Carrie Bow Cay a slow wave drift accounts for the maximum 6 cm s_1 flow. During the late part of the study, a tropical depression forming just east of Carrie Bow Cay caused an abnormally high mean tide that increased the flow of water into the lagoon. Introduction Coral reef ecosystems are controlled by the complex interaction of biological, chemical, geo? logical, and physical parameters (Odum and Odum, 1955), of which the least understood are probably the physical forces?waves, winds, tides (see for example Macintyre et al., 1974). Early studies of the physical oceanography of reefs fo? cused primarily on Pacific atolls (Munk and Sar? gent, 1948; von Arx, 1948) and paid considerably less attention to coral reefs in the Caribbean. Jeffrey E. Greer, Marine Science Program, University of South Carolina, Columbia, S.C. 29208. Bjorn Kjerfve, Marine Science Program, Department of Geology, and the Belle W. Baruch Institute for Marine Biology and Coastal Research, University of South Caro? lina, Columbia, S.C. 29208. Although recent investigations have increasingly shifted to the Caribbean, mainly because of prox? imity for American workers, physical oceano? graphic investigations of the well-developed bar? rier reef of Belize are only beginning. Previous physical research in this area was concerned with waves and wave-related processes (Roberts, 1974; Roberts et al., 1975; Shinn, 1963; Storr, 1964; Wilson et al., 1973), so that the interaction of tides with other physical processes has received little attention. Recent work on Grand Cayman (Roberts et al., 1977) suggests strong tidal influ? ence on physical processes within the shallow fore-reef region. Because the tide in the Carib? bean is microtidal and of mixed semidiurnal type (Kjerfve et al., herein), it might be expected to have considerably less influence on the reef system than wind-driven currents and waves. Our study around Carrie Bow Cay indicates that just the opposite may be true?that is, tidal processes may play a major role in controlling the water move? ments within the reef system. ACKNOWLEDGMENTS.?We are grateful for the valuable field assistance given by L. Greczy, Louisiana State University; J. A. Proehl and F. M. Drescher, the University of South Carolina; and especially to L. Kjerfve. Vice consul J. Walsh of the Unites States Embassy in Mexico City and United States Senator S. Thurmond of South Carolina gave us invaluable logistic assistance during the 1978 Belize trip, which allowed us to complete the study. The Belle W. Baruch Insti? tute for Marine Biology and Coastal Research has listed this paper as its Contribution Number 270. 53 54 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Site, Measurements, and Analysis The barrier reef off Belize, which extends in a north-south direction for over 200 km, is sepa? rated from the mainland by a lagoon that is 10- 25 km wide and 5-20 m deep. The study area, Carrie Bow Cay (16?48'N, 88?05'W), a small, sandy cay approximately 120 m long and 30 m wide, lies 100 m west of a well-developed section of the barrier reef crest (Figures 34, 40). Two large cuts, 12 m deep and approximately 1 and 2 km wide, respectively, separate the Carrie Bow reef section from Tobacco Reef to the north and Gladden Cay Reef to the south (Figure 2). Carrie Bow receives steady northeasterly trade winds and has a mixed semidiurnal tide with a mean range of 15 cm. Current, tide, wind, and atmospheric pressure were measured in June 1978. A Bendix Q-15 recording current meter suspended from a ^ \*. ? LT L2, L4 ? South Water fm^\ J- jronsect Carrie Bow L 3 * 88' Cut 04.7 ' W Cut ? t N 0 100 meters 6 ? 4 8 . l ' N- dense coral Thalassia FIGURE 34.?Location of current meter stations around Car? rie Bow Cay. moored boat was monitored at five locations (Fig? ure 34) in the vicinity of Carrie Bow Cay for time periods ranging from 17 to 74 hours. At each location the Q-15 was positioned at mid-depth. At location LI it was placed in the lagoon 125 m WSW of the island, where the total water depth was 4 m (duration, 24 hours); at locations L2 and L4 it was placed in South Water Cut above a large, flat sand plain in depths of 11 m (duration, 43 and 44 hours, respectively); at location L4 it was in Carrie Bow Cut above a 10 m deep sand channel (duration, 74 hours); and at location L5 it was in the lagoon 150 m W of the main study transect (Riitzler and Macintyre, herein), where total water depth was 3 m (duration, 17 hours). Each record of current speed and direction was digitized at 10 min intervals, resolved into E-W and N-S components, and then averaged vecto- rially to obtain resultant hourly speeds and direc? tions. The time-series plots of the data obtained in the South Water Cut and Carrie Bow Cut are shown in Figures 35 and 36. Current records for stations LI and L5 are presented as progressive vector diagrams in Figure 37. A Benthos 2820 submergible recording tide gauge was installed below the dock on Carrie Bow Cay to obtain the complete tidal record during the study (Figure 38). Wind speed and direction were read on a cup anemometer with vane every hour on the hour; readings were av? eraged by eye for 30 s to smooth high-frequency noise. The wind data are plotted on a frequency isopleth in Figure 39. The atmospheric pressure was measured using a Weather-Measure B211 recording microbarograph. Results and Discussion Northeasterly winds with speeds of 4-5 m s~ were most frequent during the study period. These conditions seem to prevail during 70% of the year (Riitzler and Ferraris, herein). Sustained winds did not exceed 9 m s_1 and the wind direction varied little from northeast. During the latter part of the study (16-20 June), the average wind speed increased slightly, to 6.5 m s -1, and water levels became abnormally high. The in- NUMBER 12 55 South Water Cut (L2, L4) Current 20 4 20 Jun FIGURE 35.?Time-series of current speed and direction in South Water Cut at current meter locations L2 and L4, using vectorially averaged hourly values. 50- 4 0 ? 3 0 - 20 - 10- 0 - ^Direction A i i 1 1 Carrie Bow Cut(L3) Current Speed . _//, 360 288 ?216 ? 144 ?72 16 0 16 Jun 8 0 17 Jun 16 0 18 Jun FIGURE 36.?Time-series of current speed and direction in Carrie Bow Cut at current meter Location L3, using vectorially averaged hourly values. 1100 09 Jun Meter Location LI 1200 08 Jun I 0 I East-West Distance (km) FIGURE 37.?Progressive vector diagram of currents at cur? rent meter locations LI and L5, using vectorially averaged hourly values. creased wind speed and water level were probably related to a tropical depression that was devel? oping east of Glover's Reef. The atmospheric pressure showed a net drop of 6 mb from 15 to 20 June. The predominant flow feature observed in the major reef cuts was a strong tidal periodicity with flood currents greatly exceeding the ebb currents. Maximum current velocities in South Water Cut were 26 cm s"1 and 43 cm s_1 during the first (L2) and second (L4) study periods, respectively. In 56 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Carrie Bow Cay Tide FIGURE 38.?Measured tide at Carrie Bow Cay dock, 8-22 June 1978; numbered segments indicate water elevations when Q-15 current meter was in operation at locations L1-L5. W? Density: 2.07 1.49 Frequency (%) FIGURE 39.?Frequency isopleth diagram (modified from Seppala, 1977) of 312 hourly recordings of wind data at Carrie Bow Cay, Belize, 8-22 June 1978, indicating density of wind observations (in observations per m s_1 and 15? segments) and percentage of frequency corresponding to a given direction. Carrie Bow Cut, observed hourly peak velocities measured 40 cm s - . Directional changes during each half tide (rising or falling) were negligible in both South Water and Carrie Bow cuts. In South Water Cut the average resultant direction for inflow was toward 260?, and for outflow toward 70?. In Carrie Bow Cut the flood flow was toward 275? and the ebb flow toward 105?. The predom? inance of the flooding tide over the ebb during the study period indicates a slow, continual in? filling of the lagoon, which is also indicated by data for the mean increase of the tide level (Figure 38). The tidal pattern was due to the stationary tropical depression to the east of Carrie Bow, which had strengthened the northeast trade winds. Current patterns observed at lagoon stations LI and L5 are shown in Figure 39 on a progressive vector diagram. Neither location exhibited the obvious tidal activity that had occurred at sta? tions L2, L3, and L4. Meter location LI showed weak water movement and little effects of tidal or wind forcing. Hourly averaged currents at LI did not exceed 6 cm s~ and were generally on the order of 2 cm s_1. This location, being on the lee side of the island, was shaded from direct wind influence. The net flow direction to the northeast at LI suggests that the current was due to wave drift. The wave crests that propagated into Carrie Bow Cut were greatly refracted by the reef crest and entered the Carrie Bow lagoon from the southwest, as indicated by waves break- NUMBER 12 57 FIGURE 40.?Vertical high altitude aerial photograph of Carrie Bow Cay (center) showing wave refraction patterns; island to the north (top) is South Water Cay, sand bank to the south (bottom) is Curlew Bank (picture area = 2.5 X 3.5 km). ing on the lee side of the island and refraction patterns on aerial photographs (Figure 40). Although higher water levels and increased wind speeds due to the developing tropical de? pression were atypical conditions for station L5, these conditions probably did not alter the typical flow patterns there, but merely modulated their intensity. Maximum hourly velocities at L5 were 33 cm s_1, with a net flow direction due southwest, varying almost 180? from that at LI. The current 58 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES at L5 did not have the strong tidal signature evident in the major reef cuts. Rather, water was being driven across the reef crest and into the lagoon by the northeasterly trade winds. While the higher than usual water levels and increased wind speeds tended to move more water across the crest than usual, the current at L5 appeared to be primarily wind-driven. Conclusions Current flow through the major reef cuts bor? dering Carrie Bow Cay appears to be tidally dominated despite a small tidal range. Measure? ments within the cuts showed a predominance of flooding over ebbing currents that was related to increased northeasterly winds associated with the development of a tropical depression to the east of the island. Depending on location, net currents inside the lagoon may be wave-driven, wind- driven, or both. Current records from this area lack the clear tidal signature that was observed in Carrie Bow and South Water cuts. Because our study was conducted in the early summer, however, caution must be exercised in extrapo? lating the results to other seasons. Literature Cited Arx, W. S. von 1948. Circulation Systems of Bikini and Rongelap La? goons. Transactions, American Geophysical Union, 29(6) :861-871. Macintyre, I. G., S. V. Smith, and J. C. Zieman, Jr. 1974. Carbon Flux through a Coral Reef Ecosystem: A Conceptual Model. Journal of Geology, 82:161-171. Munk, W. H., and M. C. Sargent 1948. Adjustment of Bikini Atoll to Ocean Waves. Trans? actions, American Geophysical Union, 29 (6):855-860. Odum, H. I., and E. P. Odum 1955. Trophic Structure and Productivity of a Wind? ward Coral Reef Community on Eniwetok Atoll. Ecological Monographs, 25:291-320. Roberts, H. H. 1974. Variability of Reefs with Regard to Changes in Wave Power around an Island. In A. M. Cameron, B. M. Campbell, A. B. Cribb, R. Endean, J. S. Jell, A. O. Jones, P. Mather, and F. H. Talbot, editors, Proceedings, Second International Symposium on Coral Reefs, 2:497-512. Brisbane, Australia: Great Barrier Reef Committee. Roberts, H. H., S. P. Murray, and J. H. Suhayda 1975. Physical Processes in a Fringing Reef System. Journal of Marine Research, 33(2): 233-260. 1977. Physical Processes in a Fore-Reef Shelf Environ? ment. In D. L. Taylor, editor, Proceedings, Third International Coral Reef Symposium, 2:507-516. Miami, Florida: Rosenstiel School of Marine and Atmospheric Science. Seppala, M. 1977. Frequency Isopleth Diagram to Illustrate Wind Observations. Weather, 32(5): 171-175. Shinn, E. A. 1963. Formation of Spurs and Grooves on the Florida ReefTract. Journal of Sedimentary Petrology, 33:291- 303. Storr ,J . R. 1964. Ecology and Oceanography of the Coral Reef Tract, Abaco Island, Bahamas. Geological Society of America Special Paper, 79: 98 pages von Arx. See Arx, W. S. von Wilson, W. S., P. G. Wilson, and J. A. Michael 1973. Analysis of Swell near the Island of Aruba. Journal of Geophysical Research, 78:7834-7844. Water Exchange across the Reef Crest at Carrie Bow Cay, Belize Bjorn Kjerfve ABSTRACT Water movement across the reef crest off Carrie Bow Cay, Belize, is affected by at least five mi- crogrooves, 2-4 m wide and 1-3 m deep, cutting across the reef crest and connecting the shallow reef lagoon and the open Caribbean Sea. The currents in these microgrooves flow into the la? goon during rising tides and out to sea during falling tide. Over one or more tidal cycles, how? ever, net flow inside the microgrooves is directed toward the ocean and along the remainder of the reef crest it is directed toward the lagoon owing to water being pumped over the crest by breaking waves. Introduction Few environments are as hazardous or as dif? ficult to sample as the reef crest on an exposed coast. Breaking waves that pound the outside edge of the reef dissipate vast quantities of energy, and cause water to be pumped across the crest. In addition, the water flow over the reef is typi? cally influenced by tidal currents, hydraulic head differences (Tait, 1972), and persistent 4-8 m s~ , trade winds. Few studies have examined this complex of water motions. Munk and Sargent (1948) noted that "surge channels," cutting across the reef crest and reef flat on the windward side of Pacific atolls, are tuned to the average wave characteristics and Bjorn Kjerfve, Marine Science Program, Department of Geology, and the Belle W. Baruch Institute for Marine Biology and Coastal Research, University of South Carolina, Columbia, S. C. 29208. play an important role in the dissipation of wave energy. Although Roberts et al. (1975) found no evidence of surge channels in exposed fringing reefs around Grand Cayman and Barbados, a section of the Belizean barrier reef at Carrie Bow Cay in the western Caribbean (16?48'N, 88?05'W) has at least five channels cutting across the reef crest. Because of their small size and other characteristics that differ from the atoll surge channels, the Carrie Bow channels are called "microgrooves" in the following descrip? tion of their role in the water exchange across the reef crest. ACKNOWLEDGMENTS.?For help with the collec? tion of field data in 1977 and 1978 I am especially indebted to J. B. Atkins, F. M. Drescher, J . E. Greer, and J. A. Proehl, University of South Carolina; L. Greczy, Louisiana State University; and to L. Kjerfve. J. Stallings, United States Geological Survey, and D. Middaugh, Environ? mental Protection Agency, donated Rhodamine dye and practice golf balls, respectively, for nu? merous qualitative flow experiments. The Belle W. Baruch Institute for Marine Biology and Coastal Research has listed this paper as its con? tribution Number 271. Results MICROGROOVES.?The Carrie Bow section of the barrier reef lies just off the south end of Tobacco Reef. It is separated from South Water Cay by a cut, 800 m wide and 12 m deep. The reef crest forms a half-ellipse 1000 m long (Figure 41) that consists of dense growths of Acropora 59 60 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Reef Crest Microgrooves -16 88 04.7 W FIGURE 41.?Schematic representation of the Carrie Bow Cay reef crest, indicating location of five microgrooves (A- E), refraction of typical wave crests before breaking, (curved lines) and direction of predominant trade winds and deep water waves (arrows). (Scale in m.) palmata, Millepora complanata, Agaricia agaricites, Porites astreoides, and other corals. The elevation at the crest coincides approximately with the mean low water level. The crest measures 20-35 m across from a 1 m deep sand-filled moat on its lagoon side to another 4 m deep sand moat on its seaward side. The lagoonward edge of the reef crest is irreg? ularly indented and gives the impression that numerous channels open into the ocean. Most of these cuts, however, extend only part of the way across the crest, except for five locations (Figure 41) where the channels traverse the crest and thus make it possible for a swimmer to reach the fore reef even during moderate wave conditions. These small passages are the microgrooves (2-4 m wide and 1-3 m deep) that wind their way across the crest; they typically exhibit a con? stricted area somewhere between the lagoon and ocean where the water becomes significantly shal? low compared with the rest of the microgroove. The bottom of the microgrooves consists of poorly sorted carbonate sand and large amounts of bro? ken coral branches and rubble. Large numbers of schooling fish appear to be associated with the microgrooves. Microgroove E (Figure 41) across the well-developed northern section of the reef at Carrie Bow Cay is by far the deepest and widest one. The microgrooves differ from atoll surge chan? nels (Munk and Sargent, 1948) in several ways: they are shallower and narrower; they extend across the reef crest with several bends and branchings; they usually do not align with a fore- reef groove; they appear less well tuned to the predominant waves. These differences, however, may all be due to the calmer wave climate in the Caribbean. Whereas the trade-wind-generated waves on Bikini Atoll are 2.0-2.5 m high with steady 7-9 s periods, the Carrie Bow reef usually experiences 1.0-1.5 m, 4-5 s waves with a great variability over time. WATER EXCHANGE.?As wave crests progress toward a reef they shoal and refract. Before waves reach the breaking point, refraction causes their crests to become almost parallel to the outer reef edge (Figure 41). When a wave breaks, water is pumped across the reef crest (Shinn, 1963) and some water surges back before the arrival of the next breaker. The available wave energy is trans? formed into kinetic energy and is dissipated, and thus helps to maintain a superelevated water level at the breaking point, in contrast to the still-water levels of the lagoon (Munk and Sargent, 1948). This reef set-up may be as much as 20% of the incident wave height (Tait, 1972). If the waves are high, the set-up can drive over the reef crest a steady net inflow, which becomes superimposed on the more or less symmetrical, oscillatory pumping and return-surge action due to breakers. Because of refraction, waves usually traverse the barrier perpendicular to the local crest axis and NUMBER 12 61 along the mean direction of the microgrooves. As in the case of atoll surge channels, this pattern suggests a close relationship between micro- grooves and wave action. Qualitative features of the reef-crest water ex? change were assessed by means of current-meter measurements and dye drops during March 1977 and June 1978. A Bendix B-10 impeller-type, ducted, bidirectional current meter was placed at approximately middepth in microgrooves A, B, and C, as well as in two locations on the reef crest between microgrooves B and C. At each location two 25 h time series were obtained via a 200 m cable connecting the sensor and a shore-based read-out unit and recorder. The signal was sub? jected to a 2.5 s RC low-pass filter, and was then read every 5 s for 3 min on the hour every hour. Statistically significant hourly mean values based on 36 readings were plotted versus time, and compared with the tide for the same period. The resulting plot for microgroove B is shown in Fig? ure 42, that for the reef crest in Figure 43. In " 15 10 o 9 " 6 " 3 - U -6 ' Flow 7* ^ * \ \ / In Microgroove B Tide Toward Ocean \ f\ ' ^ / ' * f l / V A / \ 1 \ 1 \ i \ J \ / \ / \ i \ 1 AV /? ' \ / Current \j Toward Lagoon i i i i i i o -10 12 00 15 June 16 June FIGURE 42.?A 25 hour time series of low-pass filtered current in microgroove B, 15-16 June 1978, with simultaneous tide record at the Carrie Bow dock (positive current = lagoon- Ward flow, negative current = oceanward flow). -10 " - 2 0 19 June FIGURE 43.?A 25 hour time series of the low-pass filtered current on top of the reef crest between microgrooves B and C, 19-20 June 1978, with simultaneous tide record at the Carrie Bow dock (positive current = lagoonward flow, neg? ative current = oceanward flow). addition, Rhodamine B dye was released at ap? proximately 50 locations along the Carrie Bow section of the barrier, in front of the reef, in the lagoon, on top of the crest, and in all microgrooves during both falling and rising tides. In addition, water movement was traced by means of 200 plastic practice golf balls released at several lo? cations oceanward of the reef and observed as they crossed the reef crest; these were subse? quently collected. Conclusions Although these experiments yielded primarily qualitative results, several generalizations can be made. Currents in microgrooves as well as on top of the reef crest varied on at least two time scales: 62 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES wave and tide frequencies. The wave-induced motions were largely oscillatory and perpendicu? lar (back and forth, up and down) to the crest with typical instantaneous microgroove speeds of 10-60 cm s_1. The hourly means, however, reflect the tidal forcing and were typically 1-20 cm s_1 m in the microgrooves, directed oceanward (neg? ative) during falling tide and lagoonward (posi? tive) during rising tide. The current on top of the reef crest behaved similarly but exhibited a much greater variation than it did in the microgrooves. Both microgroove and reef-top currents clearly display the tidal signature. When averaged over a 25-hour period (2 semidiurnal or 1 diurnal cycle), however, the microgrooves showed a net oceanward flow, whereas the water on top of the crest is directed toward the lagoon (Figure 43). Dye experiments during the falling tide indi? cated a jet-like flow from the microgrooves into the forereef region. Once outside the reef, some of this water, is transported southward with the longshore current. Most of the dye seemed to return via wave-pumping over the reef crest both up- and down-stream of the microgroove exits, thus setting up a series of circulation cells along the outer reef edge. The currents are much more vigorous at microgrooves C, D, and E compared with B and A. Instantaneous and time-averaged values were greater here because in this region the incident waves have been subjected to a minimum of refraction. The features described thus far reflect trade wind conditions estimated to occur 70 percent of the time (Riitzler and Ferraris, herein). During minor storm events and other times of high wave activity, however, both reef crest and microgroove net currents flow toward the lagoon and do not display any tidal influence. At these times, the wave set-up is probably sufficiently greater to dominate any tidal influence by exceeding any differences in tidal hydraulic head. Water then leaves the shallow Carrie Bow lagoon both at the north and south of the island. Although Hernandez-Avila et al. (1977) found that during storms, coral rubble was transported from the deep fore reef to form coral boulder ramparts along the south coast of Grand Cayman, elsewhere in the Caribbean sand-sized material appears to be slowly transported away from the reef crest. These observations are not contradic? tory. In view of the measured ocean-directed net flow at Carrie Bow Cay, it can be assumed that the microgrooves act as a passway for sediment from the lagoon and reef crest during typical conditions. When a storm strikes, however, the wave drift is directed up the fore reef and the forces are great enough to transport boulders toward and over the reef. The rubble rampart that makes up the SE reef crest of Carrie Bow Cay (Riitzler and Macintyre, herein:9) undoubt? edly owes its existence to wave transport during heavy storms. Literature Cited Hernandez-Avila, M. L., H. H. Roberts, and L. J. Rouse 1977. Hurricane-generated Waves and Coastal Boulder Rampart Formation. In D. L. Taylor, editor, Pro? ceedings, Third International Coral Reef Symposium, 2: 71-78. Miami, Florida: Rosenstiel School Of Ma? rine and Atmospheric Science. Munk, W. H., and M. C. Sargent 1948. Adjustment of Bikini Atoll to Ocean Waves. Trans? actions, American Geophysical Union, 29(6):855-860. Roberts, H. H., S. P. Murray, and J. N. Suhayda 1975. Physical Processes in a Fringing Reef System. Journal of Marine Research, 33(2):233-260. Shinn, E. 1963. Spur and Groove Formation on the Florida Reef Tract. Journal of Sedimentary Petrology, 33(2):291- 303. Tait, R . J . 1972. Wave Set-up on Coral Reefs. Journal of Geophysical Research, 77(12):2207-2211. Geology and Sediment Accumulation Rates at Carrie Bow Cay, Belize Eugene A. Shinn,J. Harold Hudson, Robert B. Halley, Barbara Lidz, Daniel M. Robbin, and Ian G Macintyre ABSTRACT A 24 km long transect of cores consisting of four rotary drill cores and seven vibrocores was drilled in a line extending from the seaward side of the reef crest at Carrie Bow Cay to the main? land at a point between Sittee Point and the town of Stann Creek. Two of the four rotary cores were drilled seaward of the reef crest, one through a spur to a depth of 7.6 m and the other into the adjacent groove to a depth of 18.3 m. The two cores show no evidence that the spur and groove system was erosional in origin; rather, they dem? onstrate that it was constructional. Submarine cementation, chiefly in the form of fine-grained high-magnesium calcite, was found mainly in cemented internal sediment in both cores drilled seaward of the reef crest. Two other rotary cores were drilled landward of the reef crest. One was drilled on the reef flat to a depth of 8.8 m, and the other was drilled on the southwest tip of Carrie Bow Cay to a depth of 17.7 m. Both cores encountered essentially uncemented carbonate reef sands with some coral rubble. Of the four rotary cores, only the Carrie Bow Cay core encountered Pleistocene bedrock. Radiocarbon dating of a large head of Siderastrea siderea, growing on bedrock from the Carrie Bow Cay core at a depth of 15.04 m below sea level, gave an age of 6960?110 years. The leached calcitic coralline bedrock, at a depth of 15.7 m below sea level in the Carrie Bow Cay core, contained root marks, and iron staining indicative of subaerial exposure. Eugene A. Shinn, J. Harold Hudson, Robert B. Halley, Barbara Lidz, Daniel M. Robbin, United States Geological Survey, Fisher Island Station, Miami Beach, Fla. 33139. Ian G. Macintyre, Division of Sedimentology, Department of Paleobiology, Smithsonian Institution, Washington, D.C. 20560. Five of the seven vibrocores, with sediment recoveries ranging from 1.6 to 5 m, contained peat. Generally, the peat was located near the bottom of the cores, and in the four most land? ward cores the peat overlay clay, silt, quartz sand, and in some a few quartzite pebbles. The peats are interpreted to record flooding of the coastal plain during the last transgression. Radiocarbon dates of peats overlying terrigenous sediments range from 6804?150 to 8808?600 years. Introduction This paper reports the results of rotary rock coring on and around Carrie Bow Cay and de? scribes a transect of seven sediment vibrocores drilled in a line from Carrie Bow Cay to the mainland, 24 km to the west. Carrie Bow Cay, literally within a stone's throw of the reef at 16?48'10"N, 88?04'45"W, is situated between two tidal passes that cut through an otherwise continuous reef flat (Figure 2). Tidal passes through the Belize barrier reef are rare, but wherever they occur, they are adjacent to islands situated near the edge of the reef tract. Carrie Bow, like other islands on the shelf south of the Belize River, is a Holocene sedimentary accumulation. Although there are extensive ex? posures of Pleistocene limestones at the northern limits of the Belizean barrier reef complex, nota? bly Ambergris Cay, nowhere within Belizean wa? ters south of the Belize River do Pleistocene or older limestones extend to or above sea level to form an island, as is common in other coral reef areas of the Caribbean. Sediments, vegetation, 63 64 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES 88"10' 16 50 Tobacco Cay Tobacco Reef CBf * V Twin Cays / \l3 7 $ \f * ? NJJB2 ^South Water Cay /jfcoV o 0 * > ^ C B 1 ) '^ ' ) . o ? y ^^")0Came Bow Cay Columbus Reef - i i i i I 5 km -16*50' 88*10' FIGURE 44.?Map of Carrie Bow Cay study area, Belize, showing transect of sediment vibrocores (CB1-CB7). FIGURE 45.?Map of immediate Carrie Bow Cay area show? ing rotary core locations (RC1-RC4) and the easternmost sediment vibrocore (CB1). A CB1 RC4^ JVansect ? / ) Carrie Bow C ay RC3 ^ ^ RC 1 % _ RC2 ^ Reef Crest and effects of storms on Belizean islands, includ? ing Carrie Bow Cay, have been described by Stoddart (1962, 1963). Vibrocore locations are shown in Figures 44 and 46 and rotary drill core locations are shown in Figures 45 and 46. Core RC2, immediately adjacent to RC1, was taken on a spur (Figure 47). This core was compared with core RC1, taken in a groove, to determine whether the spur and groove system was formed by erosion, as NUMBER 12 65 proposed by Cloud (1959) for Saipan reefs or by construction, as proposed for Jamaica reefs by Goreau (1959) and determined for a Florida reef by Shinn (1963). This work, conducted on and around Carrie Bow Cay, is part of a larger United States Geo? logical Survey Belize project, the objectives of which are to determine (1) porosity distribution as controlled by submarine cementation, (2) fac? tors that control reef distribution, (3) nature and origin of the numerous patch reefs lagoonward of the barrier reef, and (4) rate of sediment accu? mulation during the past 10,000 years. Only that work conducted on and around Carrie Bow Cay is reported here. ACKNOWLEDGMENTS.?We gratefully acknowl? edge the gracious cooperation of the Belize gov? ernment, in particular the fisheries department. This work would not have been possible without the aid of R. Gaensslen, owner and captain of the M / V Sea Angel. We also acknowledge the inval? uable aid of P. Shea, who acted as diver and technician. The adapted jackhammer used in sediment vibrocoring was designed and perfected by D. Lanesky of the Comparative Sedimentol? ogy Laboratory, University of Miami, Fisher Is? land Station, Miami Beach, Florida. The use of brand names in this report is for descriptive pur? poses only and does not constitute endorsement by the United States Geological Survey. RC4 RC3 FIGURE 46.?Cross sections of study area: a, entire area shown in Figure 44, with vibrocore (CB) and rotary core (RC) locations; b, detailed cross section of Carrie Bow Cay and associated reef area comparable to Figure 45. Note distribution of corals seaward of reef crest, which is composed of coral rubble encrusted with Millepora sp. The inferred transition zone from in situ coral accumulation to sand and rubble is indicated by the zig-zag line. Staghorn corals shown near cores RC3 and RC4 are inferred. Staghorn corals in RC 1 and RC2 were recovered because of cementation. R C l and RC2 are at the same distance along the E-W section, RC2 being immediately to the N of R C l . (For explanation of symbols, see Figure 47.) 66 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES 5 - 10- 15- 2 0 - 25 J R C 2 7175 ?100 S ^ ^ ^c- Coralline algae & Agaricia spp. >< Acropora cervicornis (2& Montastrea annularis ^ Porites astreoides ^ Diploria 10m FIGURE 47.?N-S cross section between cores RCl (drilled in a groove) and RC2 (drilled on adjacent spur) showing distribution of major coral components. Radiocarbon dates indicate that the section is less than 7175 years old and that growth rates are rapid. See Table 6 for accumulation rates. Also note apparent date reversal in corals between 14 and 17 m. Both ages are within the margin of error for the method; the samples should therefore be considered approximately.the same age. These two dates indicate an extremely rapid rate of accumulation for that interval. Methods ROTARY DRILL CORING.?Core drilling into the reef (Figure 48a) was accomplished with a version of the underwater hydraulic rotary drill described by Macintyre (1975). Modifications include (1) a portable, lightweight power source (manu? factured by Custom Hydraulics of Miami, Fla.) measuring 1 m in length, 60 cm in width, 80 cm in height, and weighing only 125 kg, which is easily operable from a small skiff; (2) a light? weight plastic hydraulic hose, Vi inch (12.7 mm) in diameter; and (3) a lightweight, collapsible aluminum drilling tripod (Figure 48a). Standard length (5 ft, 1.5 m) N drilling rods and both X and BX diamond-tipped core barrels were used. Although the spur and groove zone was drilled to a depth of 18 m (RCl), only nine meters of the reef flat could be penetrated because of collapsing sand. Drilling the sands of Carrie Bow Cay re? quired drilling mud, but commercial drilling mud FIGURE 48.?Underwater coring equipment employed: a, diver-operated hydraulic drilling rig at location of R C l ; b, hydraulic jackhammer in operation; c, equipment used for vibrocoring (A = hydraulic jackhammer; B = adapter that mates jackhammer to core tube; C = plastic cap for capping core before extraction; D = section of 3-inch (76 mm) diameter aluminum core tubing, with vents for letting out water as core is taken; E = aluminum clamps with handles used for extracting core from sediment). 68 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES was not available; a mixture of clay and silt dug from an area near Stann Creek on the mainland and diluted with sea water sufficed to accomplish the drilling. Five garbage cans containing sedi? ment were brought to the island and dumped into a pit lined with plastic. The diluted mixture, necessary to prevent collapse of the hole, was used to drill the site near the south tip of Carrie Bow Cay to a depth of 16.2 m, where Pleistocene limestone appeared. Total depth including Pleis? tocene bedrock was 17.7 m. VIBROCORING.?Sediment vibrocoring (Figure 48b) was accomplished by adapting a 36 kg hy? draulic jackhammer (manufactured by Fairmont Hydraulics of Fairmont, Minn.), similar to the familiar air-powered jackhammers used to break up concrete, so that it could use 6 and 9 m lengths of standard 3 in (7.6 cm) aluminum irrigation tubing. The advantages of this hydraulic equip? ment were reduced corrosion and the ability to utilize the same hydraulic power source as that used for rotary drilling. The jackhammer was joined to the thin-walled tubing by an adapter that was constructed by press fitting an internally fitted steel sleeve to a standard, hardened steel chisel made especially for the hammer. The sleeve is a friction-fitting device that slips approximately 10 cm into the tubing. Further insertion is pre? vented by a collar (Figure 48c). Water is able to escape from the core tubing as it penetrates sediment by means of relief ports added near the upper end of the tubing (Figure 48c). Slits were cut into the tube with a hacksaw and one side of each cut hammered inward, so that the resulting ports resemble gill slits. Ini? tially, ports were made with an electric drill, but the saw and hammer method was found to be more efficient. Coring was accomplished by attaching either the 6 or 9 m lengths of tube to the jackhammer and floating the entire device with an inverted air-filled garbage can. When the core tube was vertical over the selected site, one diver released air from the garbage can while another diver guided the unit to the bottom. When the tube contacted bottom, the diver depressed the trigger and the machine-gun action of the hammer drove the core tube into the sediment (Figure 48b). Generally, 6 to 9 m of penetration was achieved in less than 30 seconds with an average of 80 percent recovery. In very clear water the tube was first driven approximately xh m vertically into the bottom with a 1 kg sledge hammer. When the tube was correctly oriented and free standing, the jackhammer was floated into position and at? tached to the adapter, and coring began. Because the tube itself does not rotate during coring, orientation of the core can be determined by putting a scratch on one side of the soft aluminum tube before extracting the core from the bottom. Cores were removed from the bottom with the aid of adjustable clamps with handles that were attached to the tube at the sediment/water inter? face. Excess tubing was sawed off underwater by hacksaw. An externally fitting plastic cap (Caplug EC-48, manufactured by Caplugs Divi? sion of Protective Closures Company, Inc., Buf? falo, N.Y.) was affixed to the protruding tube end, thereby creating a vacuum and preventing loss of core material from the buried end during extraction. The cores were extracted and brought to the surface by attaching the garbage can and filling it with air from the SCUBA regulator. Although core' catchers were not used, sediment was seldom lost during extraction. All vibrocores were dried and impregnated with plastic according to the methods of Ginsburg et al. (1966). Peats were removed prior to impreg? nation. Carbon-14 dating was carried out at the University of Miami's Radiocarbon Dating Lab? oratory. Sediment types are described according to the classification of Dunham (1962). In keeping with his scheme, loose sediment is described as if it were the rock it will eventually become. Thus, "packstone" is used to describe sand-size carbon? ate grains whose interstices are filled with lime mud; "grainstone" is used if mud is absent. In addition, mud-supported sediments are termed "mudstones" if they contain less than 10 percent sand-size grains and "wackestones" if they have more than 10 percent. NUMBER 12 69 Results OUTER REEF.?Of the four cores drilled in the Carrie Bow Cay area, only RCl and RC2 con? tained in situ corals. Core RCl from a groove in 7 m of water was by far the most difficult to drill (see Figures 45, 46, 47). The core was drilled to 18.3 m before caving and other difficulties pre? vented further penetration. Pleistocene bedrock was not encountered. Table 5 shows that core recovery was about 36 percent. Core RC2, ap? proximately 10 m to the north on the adjacent spur, was drilled to a depth of 7.6 m, the first 3.7 m being in the spur. Because the spur is composed primarily of the lettuce coral, Agaricia sp., and little cementation or infill of internal sediment is present, there was practically no core recovery in this interval. Recovery of material began when the drill reached the level of the adjacent groove. Because the material from both groove and spur base was identical, we concluded that the spur is of constructional origin. We recognize, however, the possibility of a deep-seated erosional spur and groove in the Pleistocene strata below 18 m which might serve as a template for later growth. None? theless, the visible portion of the spurs in this area is clearly constructional, and the interval 3.7 to 7.6 m (below the spurs) is probably also construc? tional. The most noticeable feature in core RCl and the lower 3.7 to 7.6 m of RC2 is the occurrence of both cemented and uncemented internal sedi? ment. Fine-grained internal sediment was gener? ally light gray in color and contained sedimentary laminations, commonly inclined. The cement is Mg-calcite, similar to that reported elsewhere (Ginsburg et al., 1967; Macintyre et al., 1968; James et al., 1976; Land and Goreau, 1970; Mac? intyre, 1977; Shinn, 1969, 1971). In partially filled small cavities (less than 1 cm across), inter? nal sediment is commonly geopetal, that is, the upper surface of the fill is horizontal. In larger voids, however, such fillings are generally inclined (Figure 49). Very little uncemented sediment was recovered from large voids, but its presence was indicated by bursts of muddy drilling water as- E A/**''1 *? a FIGURE 49.?Two pieces of core from R C l : a, section from 10.75 m showing cemented internal sediment at steep angle (arrow); note unusual network of voids in cemented internal sediment; cause of the void network is not known, although it does not appear to be caused by leaching; b, sample from 7.4 m; gray internal sediment resting on Porites coral; note steeply inclined laminations within cemented internal sedi? ment. Cement is Mg-calcite; origin of voids is not known, but they are thought to be organic borings. sociated with sudden drops of the drill bit and by small amounts of sediment recovered in the cores. Fibrous aragonite, probably the most diagnos? tic form of submarine cement, was present only in small cavities, usually within shell and coral chambers, and was visible only in thin sections. This observation is identical to that reported by Macintyre (1977) but is in strong contrast to the fibrous aragonite crystals 1 to 2 cm long described by Ginsburg and James (1976) and James and Ginsburg (1978) from the deep fore reef at Belize below 100 m. REEF FLAT AND ISLAND.?Core RC3 was drilled on the reef flat approximately 15 m behind the reef crest in water about 25 cm deep. The seabot- tom at the RC3 site consisted of a hard pavement 70 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES composed of cemented sediment sparsely popu? lated by small head corals. Drilling showed the pavement to be only 1 cm thick and underlain by coral rubble imbedded in coarse carbonate reef sand. From approximately 2 m to total depth (8.8 m), only carbonate sand with sparsely scat? tered coral fragments was encountered. Except for the thin cemented pavement exposed at the surface, no evidence of submarine cementation was found, nor did any corals appear to be in growth position. Core RC4 was drilled on the southwest tip of Carrie Bow Cay (Figures 45). Drilling revealed a section consisting mainly of coarse-grained car? bonate sand. At a depth of 15.5 m, a single large head of Siderastrea siderea appeared in growth po? sition on Pleistocene bedrock. Bedrock was reached at 16.2 m and cored to 17.7 m. It is coralline limestone that has been leached and contains brown caliche staining. The coral over? lying the bedrock was dated by carbon-14 tech? nique and is discussed below. There was no evi? dence of cementation of the Holocene sands be? neath Carrie Bow Cay. TABLE 5.?Locations, water depths, core recovery, and general comments on rotary drill cores and sediment vibrocores taken in the Carrie Bow Cay area, Belize Core ROTARY DRILL CORES R C l RC2 RC3 RC4 VIBROCORES CB1 CB2 CB3 CB4 CB5 CB6 CB7 Lat, Long. 16?48'13"N, 88?04'41"W 16?48'13"N, 88?04'41"W 16?48'13"N, 88?04'43"W 16?48'10"N, 88?04'45"W 16?48'10"N, 88?05'10"W 16?47'50"N, 88?06'30"W 16?49'30"N! 88?07'45"W 16?50'20"N, 88?08'40"W 16?53'25"N, 88?14'00"W 16?54'15"N, 88?15'50"W 16?54'15"N, 88?15'88"W Location Description About 60 m seaward of reef crest 10 m N of core R C l On reef flat 15 m W of reef crest SW tip of Carrie Bow Cay About 150 m W of Carrie Bow Cay S of Twin Cays See Figures 44 and 45 Just W of dropoff into Victoria Channel See Figures 44 and 45 ~2 km from mainland ~300 m from shore Water depth (m) 6.7 3.7 0.25 +0.5 2.5 7.3 11.5 20.2 12.2 6.7 4.3 Core hole depth (m) 18.3 7.6 8.8 17.7 4.6 3.5 5.4 2.0 3.4 3.9 3.1 Recovery (m) 6.55 1.8 =0.5 =0.03 3.7 2.8 5.0 1.6 2.7 3.7 2.4 Comments Drilled in groove Drilled on spur; recov? ery only below base of spur Poor recovery due to uncemented sand No core recovery above 15m; good recovery from Pleistocene 16.2-17.7 m Carbonate sand bot? tom with scattered Thalassia species Peat extends from about 2.3 m to bot? tom of core recovery at 2.8 m Contains peat from 4.5 m to bottom of core recovery at 5.0 m Sticky red clay pre? vented further core penetration Peat in bottom 50 cm of core; water visibil? ity about 25 cm Peat in upper section of core; see Table 6 Bottom 3/5 of core is peat NUMBER 12 71 r% JA* 10cm FIGURE 50.?Selection of plastic-impregnated sediment cores from transect shown in Figure 44: a, core CB1, interval 40 to 70 cm; note burrowing, grass roots, and coral fragment at upper right; smaller white fragments are Halimeda; b, same interval from core CB2; note large burrow; c, core CB2 from interval 2.2 to 2.5 m showing transition from lime sediment to peat; note coral fragments at top, Oculina sp., and large unidentified oysters and other mollusks at transition; d, core CB4 showing transition from carbonate grainstone to orange terrigenous clay, contact is at base of large gastropod, cracks in lower part of core are due to dessication in clay, which would not impregnate because of its fine-grain size; e, top of core CB6 (interval 0 to 30 cm); upper portion consists of coarse, poorly sorted quartz sandstone merging with peat near bottom; large white fragments are quartz pebbles. VIBROCORE TRANSECT.?Location of all seven vibrocores is shown in Figures 44 and 46, and brief comments on depth, location, and penetra? tion appear in Table 5. Selected photographs of these plastic-impregnated cores are shown in Fig? ure 50. Core CBl consists entirely of burrowed, reef- derived grainstone but lime silt content increases slightly near the base, locally creating a packstone facies associated with burrows. Despite a few sticks oi Acropora cervicornis, the principal sand-size constituents are Halimeda plates and soritid fora? minifera. Homotrema sp., a foraminifer generally abundant in reef-derived sands, is rare. Core CB2 is predominantly a packstone mot? tled by abundant burrows. Mollusks are the most obvious large fossils and soritid foraminifera are abundant, approximately as abundant as Hali? meda plates. At 2 m, there is a transitional change to lime silt, grading into a mixture of peat and 72 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES lime silt. Large oyster shells are mixed into the peat down to 2.7 m. Broken oyster shells extend deeper. Coral fragments of Oculina sp. occur be? tween 2.1 and 2.2 m, just above the peat. Core CB3 consists of burrowed packstone and silty lime mudstone with pelecypods to 3.7 m. Sediment becomes darker at 3.7 m owing to abundance of peat flecks. Pelecypods occur be? tween 4 and 5 m. Sediment is lime silt between 4 and 4.5 m with an admixture of clay that gives it a gray-brown color. Peat begins at 4.5 m and extends to the bottom of the core at 5 m. Gastro? pods, such as Astrea sp. and various cerithids, occur just at the transition from lime sediment to peat. Core CB4 was taken in 20 m of water imme? diately landward of a continuous shallow rock ridge (see Figure 46). The upper 1 m of the core consists of burrowed mixed grainstone and pack? stone. The grains are Halimeda, mollusk frag? ments, and foraminifera. Soritid foraminifera are rare, and obvious darkened grains make up 1 to 3 percent of the section. At 1.24 m the lime sediment changes into brownish-orange clay con? taining red-brown concretions. The unexpected occurrence of such coarse-grained sediment at this depth is probably explained by the nearby shallow ridge. Sediment is probably periodically swept landward off this ridge. Core CB5 consists of brown mud and wacke- stone with terrigenous clay containing scattered mica flakes. The predominant fauna is small, thin-shelled pelecypods. At 2.3 m there are large oyster shells, and at 2.5 m the sediment grades to a darker color clay and organic-rich mudstone. Peat was collected from 1.78 and 2.67 m for dating. Core CB6 consists of 30 cm of very poorly sorted quartz sandstone with quartzite pebbles up to 1 cm in diameter. The only carbonate is pelecypod fragments. Peat begins at 30 cm and extends to 1.3 m. The remainder of the core down to 3.65 m consists of massive, dirty sandstone with root marks. Between 3.65 m and the bottom of the core at 3.7 m, the sediment is gray clay with bright-red clay mottles. Core CB7 consists of almost 1 m of burrowed, poorly sorted quartz sandstone with pelecypod fragments, scattered Halimeda plates, and few gas? tropods. Sediment contains peat fragments and abundant quartzite pebbles at 75 cm. The section becomes peat at 0.92 m, and peat extends to the bottom of the core at 2.4 m. The peat contains thin stringers of quartzite pebbles. Peat was exposed on the bottom in 1 m of water landward of site CB7, and a peat-forming swamp is present at sea level all along the shore, just landward of a thin quartz-sand beach ridge. The change from coarse-grained carbonates to terrigenous elastics and peats in a shoreward direction follows the basic pattern described by Purdy et al. (1975). Although red mangroves are abundant on the seaward edge of the swamp, numerous hard? woods, grasses, and small palms indicate the peat is not entirely of mangrove origin. Nevertheless, the peat is accumulating at sea level and can be visually correlated with the submarine pet just offshore and is thought to be continuous with the peat in cores CB6 and CB7. RADIOCARBON AGE DATING?Corals from var? ious depths within the rotary cores RC 1 and RC4 were sampled and dated by the radiocarbon method, as were peats from sediment vibrocores CB5, CB6, and CB7. Dates, calculated accumu? lation rates, and sediment and water depths are provided in Table 6. In this table there are two columns of figures under the heading "Accumu? lation rates (m/1000 y)". The first column (core top to sample) lists the accumulation rates in meters/1000 years, assuming the top of the reef is growing and has a 14C age of approximately zero. The second column (intervals between samples) lists the calculated accumulation rates between dated samples. This figure is more meaningful for obvious reasons. All peat dates were obtained by D. S. Introne in 1978 at the University of Miami's Radiocarbon Dating Laboratory. Coral material was dated at the same laboratory by J. J. Stipp. It can be seen from Table 6 that the reef off Carrie Bow Cay has been growing upward at a rapid rate (ranging from approximately 1 to 6 m/1000 y), considerably faster than accumula? tion of the lagoonal sediments overlying the peats NUMBER 12 73 TABLE 6.?Radiocarbon dates and accumulation rates (m/1000 y) of selected material in rotary drill cores and sediment vibrocores (further explanation in text; location coordinates listed after each core number; note unexplained age reversal in two samples from RCl) RCl RC4 CB5 CB6 CB7 Core 16?48'13"N, 88?04'41"W 16?48'10"N, 88?04'45"W 16?53'25"N, 88?14'00"W 16?54'15"N, 88?15'50"W 16?54'15"N, 88?15'88"W Lab. no. UM-1009 UM-1010 UM-1011 UM-1012 UM-1013 UM-1249 UM-1250 UM-1252 UM-1251 UM-1248 Material dated Montastrea annularis Porites sp. M. annularis Porites sp. Siderastrea siderea Peat Peat Peat Peat Peat Depth in core hole (m) 5.79 8.23 10.97 17.67 15.54 1.78 2.67 0.30 0.62 0.92 Inter? val be? tween samples (m) 2.44 2.74 6.70 0.89 0.32 Water depth (m) 6.7 6.7 6.7 6.7 +0.5 12.2 12.2 6.7 6.7 4.3 Depth from water surface to sample (m) 12.49 14.93 17.67 24.37 15.04 13.98 14.87 7.00 7.32 5.22 Age differ? ence between samples (y) 540 0 1035 618 2004 UC age (yBP) 5625?85 6165?90 6140?90 7175+100 6960+110 7619-320 8237+270 6804+150 8808+600 2861 + 190 Accumulation (m/WOOy) Core top Inter- to sam? ple 1.03 1.33 1.79 2.46 2.23 0.23 0.32 0.04 0.07 0.32 vats be? tween samples 4.52 0 6.47 1.44 0.16 (<1 to = 1.4 m/1000 y). This agrees with Purdy's (1974a) observation of an approximately 10 to 1 difference in accumulation rate between the bar? rier platform and the shelf lagoon. Our reef ac? cumulation rates exceed rates determined for the Florida reef tract. Shinn et al. (1977) determined accumulation rates in Florida to range from 0.38 m/1000 y to 4.85 m/1000 y. More rapid accu? mulation rates have been reported at Alacran Reef (north of the Yucatan Peninsula in the Gulf of Mexico) by Macintyre et al. (1977), where the average of four dated intervals is 5.6 m/1000 y and one interval indicates a rate of 12 m/1000 y. Further discussion of reef accumulation rates, both in the Atlantic and Pacific, has been pro? vided by Adey (1978). Discussion Figure 46 is an interpretation of facies changes across the Carrie Bow reef. Core RC3 on the reef flat just behind the reef crest clearly demonstrates that the reef flat is not composed of corals that have grown up to sea level, but instead is com? posed of reef sands and some rubble, which have probably been thrown up and over the reef crest during storms. In many areas along the lagoon- ward edge of the reef-flat sands, patch reefs are in the process of being buried as these sands accrete landward to extend the reef flat. Reefs off the Florida Keys are also growing landward (Shinn et al., 1977). From his work on Alacran Reef, Logan (1969:189) termed such accumula? tions off-reef clastic drape reefs (model 3). If underlying topography accounts for the lo? cation of Carrie Bow Cay, the proof cannot be found in our drilling. Previous work (Halley et al., 1977) proved conclusively, however, that Boo Bee Patch Reef, near Wee Wee Key about 8 km to the southwest, was initiated over a 6 m Pleis? tocene topographic high. Drilling on two man? grove islands situated on patch reefs also con- 74 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES firmed Pleistocene bedrock topography to be the controlling factor. In both cases Pleistocene cor? alline limestone reached to within 12 m of sea level under the island. Surrounding the island, the rock was more than 21 m below sea level. On the basis of his 1960 studies, Purdy (1974a,b) was the first to suggest that rock floor topography is the factor controlling island distribution on the Belize shelf. We suspect, therefore, that the reef off Carrie Bow Cay was initiated on a pre-existing rock high and that initial flooding, determined by 14C age of coral growing directly on bedrock (see Table 6), occurred before approximately 7000 y BP. Whether there has always been an island since early Holocene flooding or whether the island (actually little more than a sand spit) sprang into existence during the past few thou? sand years has not been determined. On the basis of our observations and drilling, the spurs and grooves appear to be relatively recent in age and clearly are not of erosional origin. Because the Holocene reef is greater than 18 m thick, however, it seems unlikely that the present day spur and groove systems have existed in the same place throughout the reefs history. This conclusion cannot be borne out by only two core holes. We suspect that such a system has existed in the past but that the position of the spurs and grooves has shifted laterally as the reef built up. Furthermore, it seems unlikely to us that the present system is simply patterned over a Pleistocene spur and groove system more than 18 m below the present reef surface. Determining the origin of the spur and groove system at Carrie Bow Cay reef was beyond the purpose of the expedition; however, such a deter? mination should be made. We encourage coral reef researchers to drill spur and groove systems at many sites in the Caribbean or Pacific to determine the significant factors that lead to spur and groove development. Conclusions The Holocene coral reef seaward of the reef crest is in excess of 18 m thick. The present spur and groove system is constructional. The reef flat is composed primarily of reef-derived carbonate sand with scattered coral rubble. Carrie Bow Cay is Holocene and composed primarily of carbonate sand over 15 m thick. Leached calcitic limestone, which was subaerially exposed during the last glacial period, underlies the island at a depth of 16 m. Flooding of the bedrock occurred prior to 7000 years ago. Submarine cementation is, for the most part, restricted to the Holocene section seaward of the reef crest. Except for a thin ex? posed pavement directly shoreward of the reef crest, all reef flat and lagoonal sediments over a distance of 24 km are uncemented. Peat, repre? senting former sea level, occurs in sediment cores in the lagoon. Peat in cores near the mainland overlie terrigenous clastic sediments; thus, the peats probably record the initial flooding of the coastal plain during the most recent transgression. Literature Cited Adey, W. H. 1978. Coral Reef Morphogenesis: A Multidimensional Model. Science, 202(4370):831-837. Cloud, P. E. 1959. Geology of Saipan, Mariana Islands, Part 4: Sub? marine Topography and Shoal Water Ecology. United States Geological Survey Professional Papers, 280-K:361-445. Dunham, R. J. 1962. Classification of Carbonate Rocks according to Depositional Texture. In W. E. Ham, editor, Clas? sification of Carbonate Rocks. The American Asso? ciation of Petroleum Geologists, Memoir, 1:108-121. Tulsa, Oklahoma. Ginsburg, R. N., H. A. Bernard, R. A. Moody, and E. E. Daigle 1966. The Shell Method of Impregnating Cores of Un- NUMBER 12 75 consolidated Sediments. Journal of Sedimentary Pe? trology, 46:1118-1125. Ginsburg, R. N., and N. P. James 1976. Submarine Botryoidal Aragonite in Holocene Reef Limestones, Belize. Geology, 4:431-436. Ginsburg, R. N., E. A. Shinn, and J. Schroeder 1967. Submarine Cementation and Internal Sedimen? tation within Bermuda Reefs [Abstract]. In Pro? gram, 1967 Annual Meetings, The Geological Society of America, New Orleans, Louisiana, pages 78-79. Goreau, T. F. 1959. The Ecology of Jamaican Coral Reefs, I: Species Composition and Zonation. Ecology, 40:67-90. Halley, R. B., E. A. Shinn, J. H. Hudson, and B. Lidz 1977. Recent and Relict Topography of Boo Bee Patch Reef, Belize. In D. L. Taylor, editor, Proceedings, Third International Coral Reef Symposium, 2:29-36. Miami, Florida: Rosenstiel School of Marine and Atmospheric Science. James, N. P., and R. N. Ginsburg 1978. The Deep Seaward Margin of Belize Barrier and Atoll Reefs. 201 pages. Miami Beach, Florida: University of Miami. James, N. P., R. N. Ginsburg, D. S. Marszalek, and P. W. Choquette 1976. Facies and Fabric Specificity of Early Subsea Ce? ments in Shallow Belize (British Honduras) Reefs. Journal of Sedimentary Petrology, 46(3):523-544. Land, L. S., and T. F. Goreau 1970. Submarine Lithification of Jamaican Reefs. Jour? nal of Sedimentary Petrology, 40(l):457-462. Logan, B. W. 1969. Coral Reefs and Banks, Yucatan Shelf, Mexico. In B. W. Logan, J. L. Hardin, W. H. Ahr, J. D. Williams, and R. G. Snead, editors, Carbonate Sediments and Reefs, Yucatan Shelf Mexico. The American Association of Petroleum Geologists, Memoir, 11:129-196. Macintyre, I. G. 1975. A Diver-operated Hydraulic Drill for Coring Sub? merged Substrates. Atoll Research Bulletin, 185:21- 26. 1977. Distribution of Submarine Cements in a Modern Caribbean Fringing Reef, Galeta Point, Panama. Journal of Sedimentary Petrology, 47(2):503-516. Macintyre, I. G., E. W. Mountjoy, and B. F. D'Anglejan 1968. An Occurrence of Submarine Cementation of Car? bonate Sediments off the West Coast of Barbados, W. I. Journal of Sedimentary Petrology, 38(2) :660- 664. Purdy, E. G. 1974a. Karst Determined Facies Patterns in British Hon? duras: Holocene Carbonate Sedimentation Model. The American Association of Petroleum Geolo? gists Bulletin, 58 (5):825-855. 1974b. Reef Configurations: Cause and Effect. In L. F. Laporte, editor, Reefs in Time and Space. Society of Economic Paleontologists and Mineralogists, Special Publication, 18:9-76. Purdy, E. G , W. C. Pusey, and K. F. Wantland 1975. Continental Shelf Attributes. In K. F. Wantland and W. C. Pusey, editors, Belize Shelf?Carbonate Sediments, Clastic Sediments and Ecology. The American Association of Petroleum Geologists, Studies in Geology, 2:1-40. Shinn, E. A. 1963. Spur and Groove Formation on the Florida Reef Tract. Journal of Sedimentary Petrology, 33(2):291- 303. 1969. Submarine Lithification of Holocene Carbonate Sediments in the Persian Gulf. Sedimentology, 12: 109-144. 1971. Aspects of Diagenesis of Algal Cup Reefs in Ber? muda. Transactions, Gulf Coast Association of Geologi? cal Societies, 21:387-394. Shinn, E. A., J. H. Hudson, R. B. Halley, and B. Lidz 1977. Topographic Control and Accumulation Rate of Some Holocene Coral Reefs: South Florida and Dry Tortugas. In D. L. Taylor, editor, Proceedings, Third International Coral Reef Symposium, 2:1-7. Miami, Florida: Rosenstiel School of Marine and Atmospheric Science. Stoddart, D. R. 1962. Castastrophic Storm Effects on British Honduras Reefs and Cays. Nature, 196(4854):512-515. 1963. Effects of Hurricane Hattie on the British Hon? duras Reefs and Cays: October 30-31, 1961. Atoll Research Bulletin, 95:1-142. Terrestrial Environment and Climate. Carrie Bow Cay, Belize Klaus Riitzler and Joan D. Ferraris ABSTRACT Severe hurricane activity during the past 20 years has reduced Carrie Bow Cay (16?48'N, 88?05'W) to half its pre-1960 size of 0.8 ha. At present, Carrie Bow Cay (0.4 ha area) is one of the smallest inhabited sand cays on the barrier reef of Belize. The island measures 120 X 36 m, rises 40 cm above mean tide level, and supports three wooden cottages with freshwater tanks. The only permanent terrestrial plants are about 60 coconut trees. Other vegetation appears periodi? cally and spreads until it is destroyed by inter? mittent storm tides. Conspicuous animals include a few birds, a lizard, and some supratidal crus? taceans. About one-third of the island's surface is intertidal and occupied mainly by algae, crusta? ceans, and mollusks that are adapted to this habitat. The climate is oceanic and is dominated largely by northeasterly trade winds. Introduction Previous terrestrial investigation of Carrie Bow Cay was based on brief topographic and floristic surveys (Stoddart, 1963; 1969; 1974) and a short- term meteorological study (Kjerfve, 1978). Our own first topographic survey was prompted by the severe impact of hurricane Fifi, in 197, on the shape and size of the island. From then on we monitored morphological changes of the cay, con- Klaus Riitzler, Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Joan D. Ferraris, Mount Desert Island Biological Laboratory, Salsbury Cove, Maine 04672. dition of the remaining coconut tree population, and recovery of the vegetation that had been entirely eliminated by salt water flooding. Our other observations on the terrestrial and intertidal flora and fauna of the islet are of casual qualita? tive nature and restricted to large and conspicu? ous organisms. Meteorological records were taken regularly during the months of our field work, mainly in spring and early summer, but are spo? radic during the remaining parts of the year. Carrie Bow Cay has been the base of the Smith? sonian Institution coral reef study since the initi? ation of the program in 1972. The small island provided the necessary support in close proximity of reef and lagoon habitats without having no? ticeable terrestrial effects on these environments. ACKNOWLEDGMENTS.?We thank H. Pulpan and M. Carpenter for their help in surveying Carrie Bow Cay. The following colleagues pro? vided or confirmed identification of organisms mentioned in this report: S. H. Brawley (blue- green algae); J. N. Norris (algae, except blue- greens); M. E. Hale (lichens); M.-H. Sachet (higher plants); J. C. den Hartog (cerianthids); W. A. Newman (barnacles); B. Kensley (crusta? ceans, except barnacles); R. S. Houbrick (mol? lusks); D. L. Pawson (echinoids); R. I. Crombie, F. J. Irish (reptiles); and W. Trivelpiece (birds). C. H. Sprinkle helped with information on me? teorological monitoring practices. The following persons assisted in maintaining and evaluating weather records: H.T.A. Bowman III, M. Car? penter, B. Kjerfve, R. J. Larson, K. Leslie, A. 77 78 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Rath, D. S. Robertson, R. Sims and B. Spracklin. I. Jewett drafted the graphs. Special thanks are due P. J. Herbert for information on hurricane events in Belize, particularly for case histories of Fifi and Greta, and D. R. Stoddart for use of his 1972 map of Carrie Bow Cay. Methods GEODETIC SURVEYS.?Tape measure, sighting compass, and sighting along permanent or tem? porary markers were used for determining shape and size of the island and position of trees and artificial structures. The concrete boat dock on the lagoon side and the wooden main house in the center of the cay provided the principal points of reference. Vertical photographs from a helicop? ter, in March 1976, helped to improve the docu? mentation of the island's physiography. PLANT ABUNDANCE.?The first quantitative survey of plants made in 1978 consisted of sub? jective visual estimates of five categories of rela? tive abundance. More objective measurements of area of plant coverage were made in 1979 by placing a 50 X 50 cm (0.25 m2) frame subdivided into 10 X 10 cm (100 cm2) fields over all surfaces of the island showing vegetation. The number of frames occupied by a given species was recorded to the nearest half frame (50 cm2). METEOROLOGY.?General ly , meteorological conditions were recorded three times a day (0600- 0800 h; 1200-1400 h; 1800-2000 h) whenever the laboratory on Carrie Bow Cay was in operation, most commonly during the periods January-June and October/November, 1976-1978. Tempera? ture was measured with ?0.5? C accuracy in shaded air, in sun-exposed sand (5 cm below surface), in water on the reef flat (0.2 m average bottom depth), and below low-tide level at the boat dock (lagoon, 0.8 m average bottom depth). Some continuous analog chart recordings of solar radiation were made by pyranograph (Weather- Measure B211). Wind speed and direction were read from a cup anemometer with air foil vane (WeatherMeasure W121). Precipitation was mea? sured with two rain gauges (10 cm diameter), one installed on the cay, and the other one on the mainland at Pelican Beach Motel, Dangriga. The rainy season (June-September) of 1979 was also monitored by an unattended tipping-bucket rain- gauge (WeatherMeasure P501-I) with solar-pow? ered event recorder on Carrie Bow Cay. Humidity was calculated from psychrometer (Psychro-Dyne PP100) readings. Physiography LOCATION.?Carr ie Bow Cay (16?48'N, 88?05'W) is a small sand island located on top of the barrier reef that lines the outer shelf edge of Belize (formerly British Honduras), Central America (Plate 1: center right). Its former name, Ellen Cay, is still recorded on many nautical charts. The nearest significant settlement is Dan? griga (Stann Creek), a town of 7000 inhabitants on the mainland, 24 km due 320? (NW). The cay belongs to H.T.A. Bowman of Dangriga and is used as a vacation place for his family. The nearest islands are South Water Cay, 1.5 km due 0? (N), a sand cay populated by a few fishermen and occasional vacationers, and Twin Cays (also known as South Water Range) 4 km due 323? (NW), an uninhabited mangrove development. Carrie Bow Cay is protected from open ocean waves by a crescent-shaped reef crest to the east and a 100 m wide reef flat that extends from the crest to the island's seaward shore (Plate 5: top left). SHAPE AND SIZE.?With a surface area of less than 0.4 ha, Carrie Bow Cay belongs among the smallest inhabited cays on the Belizean shelf (Stoddart and Fosberg, herein: 527) (Figure 51; Plate 1: bottom left). The island formerly was double its present size (H.T.A. Bowman, pers. comm.) and bordered by mangrove, but clearing of these trees in 1944 led to progressive erosion by storm tides. Stranded beachrock as far as 30 m east and south of the present seaward shore doc? uments both a shift in dimensions and slow mi? gration leewards. At present, the cay has an elipt- ical shape with approximate north-south exten- NUMBER 12 79 FIGURE 51.?Carrie Bow Cay, aerial view from the south, May 1973; note exposed beach rock on the reef flat east of the island. sion. The longer axis is directed due 30? (NNE) and measures 120 m between mean tide level (MTL) points; the greatest dimension perpendic? ular to this axis lies along a line transecting the center of the isle and measures 36 m. Surface area calculated from planimetry is 0.36 ha to MTL, 0.25 ha if only dry-land (supratidal) area is mea? sured. Highest elevation, which is approximately 40 cm above MTL, occurs at the central portion of the island. SUBSTRATES.?Reef-derived carbonate sand and rubble on a base of Pleistocene bedrock (Shinn et al., herein: 63) make up the entire natural substrate of the cay (Figure 52a). Accu? mulation of beach sand varies with the direction and force of wind and currents. Under the influ? ence of the predominant northeasterly trade winds, sand is deposited at the north point and northwest beach and around the south tip; at times separate intertidal sand spits are formed to the north. Concrete block seawalls and rubble and coral rock landfills built up over many years to delay erosion dominate the northwest (Figure 52b) and southeast shorelines, which also have a few small sandy beaches here and there. Conch shells abandoned by generations of local fisher? men are accumulated along the southwest coast. STRUCTURES.?Other than seawalls, artificial structures on Carrie Bow Cay include two docks and three buildings with water vats (Figure 51; Plates 1: bottom right, 5: top left). The main or boat dock to the west (lagoon side) of the cay is 26 m long and built of concrete. A smaller wooden dock over the reef flat (SE) serves the two out? houses. The buildings are wooden and include the main house, 14 X 12.5 m, "Junior's House," 13 X 3.5 m (now serving as our project's labora? tory), and a small cabin, 5 X 5 m (Figure 60). 80 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 52.?Intertidal substrates: a, sandand rubble (mainly conch shells) on western shore, looking north (coconut palms felled by hurricane Greta); b, northwest seawall, coated by algae Cladophoropsis and Oscillatoria, toppled by Greta. Flora and Fauna Carrie Bow Cay's small size, lack of a fresh? water lens, and exposed location near the open ocean are responsible for the absence of a complex permanent terrestrial environment. We distin? guish between an intertidal zone along the shore and a central island area above the high-tide beach undercut, which, in our experience, has been flooded by sea water only during hurricane tides (see below). SUPRATIDAL ORGANISMS.?The most conspicu? ous, and only permanent, plants, except perhaps for the lichens, are coconut palms {Cocos nucifera L.) most of which were planted during the past 35 years in at least five recognizable N-S rows. In NUMBER 12 II May 1979 this population consisted of 58 healthy trees, of which 38 were mature and showed either nuts or flowers, 8 were immature (one or more years established), and 12 were freshly planted after hurricane Greta (less than one year estab? lished). The remaining vegetation observed in May 1978 and May 1979 (before and after hur? ricane Greta; Table 7) recolonized Carrie Bow Cay after salt water flooding associated with hur? ricane Fifi (September 1974) had washed away TABLE 7.?Systematic list and relative abundance of Carrie Bow Cay plants, excluding coconut and others artificially introduced; figures for 1978 show recolonization of the island after hurricane Fifi (September 1974), when all vegetation was destroyed, and are visual estimates; data for 1979 reflect minor changes and losses (indicated by dash) in plant cover caused by hurricane Greta (September 1978) and are compiled from quadrat counts (see "Methods"); likely methods of dispersal are indicated for each plant (B = bird, D = drift, W = wind); approximation of size and frequency of plants is given for 1978 Species (Family) Paspalum distichum L. (Gramineae) Sesuvium portulacastrum (L.) L. (Aizoaceae) Philoxerus vermicularis (L.) Beauvois (Amaranthaceae) Suaeda linearis (Elliott) Moquin (Chenopodiaceae) Portulaca oleracea L. (Portulacaceae) Coccoloba uvifera L. (Polygonaceae) Cakile lanceolata (Willdenow) O. E. Schulz (Cruciferae) Rhizophora mangle L. (Rh izophoraceae) Euphorbia blodgettii Engelmann ex Hitchcock (Euphorbiaceae) Euphorbia mesembrianthemifolia Jacquin (Euphorbiaceae) Ipomoea pescaprae brasiliensis (L.) van Ooststroom (Convolvulaceae) Ipomoea stolonifera (Cyrillo) Gmelin (Convolvulaceae) Tournefortia gnaphalodes (L.) Kunth (Boraginaceae) Eclipta alba (L.) Hasskark (Compositae) Unidentified seedling 1 2 Total plant cover 1978 m 2 > 5 0.02-0.50* 0.02-0.50* <0.01 1-5** <0.01 0.5-1.0f - 1-5** 1-5** >5 <0.01 <0.01 <0.01 <0.01 <0.01 not determined m 2 0.010 2.585 - 0.015 4.825 - 3.275 0.010 0.185 1.820 0.190 - 0.035 - 0.015 - 12.965 1979 % total 0.08 19.94 - 0.12 37.22 - 25.26 0.08 1.43 14.04 1.47 - 0.27 - 0.12 - 100.03 rank 9a 3 8a 1 - 2 9b 6 4 5 - 7 8b - Method of dispersal D? D?, B? D, B? D, B? B? D, B? D D B? B? D ? D W, B? D D * few medium-sized plants, each 0.02-0.30 m ** numerous small plants, each 0.01-0.02 m2 f few large plants, each >0.3 m 82 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES or killed all plants except the majority of coconut trees (Figure 53). In addition to the species listed in Table 7, two were recently artificially intro? duced: Casuarina equisetifolia L. (Casuarinaceae) and Hymenocallis littoralis (Jacquin) Salisbury (Lil- iaceae, sensu lato). At least three species of lichens are common on the northeast surfaces of wind exposed palm trunks: Lecanora subfusca (L.) Achar- ius, Pyxine cocoes (Swartz) Nylander, and Chiodecton sp. Although we have noted a variety of insects and a few spiders on Carrie Bow Cay, we have not determined the species and do not know whether they are breeding resident populations. Some ants, cockroaches, and spiders are no doubt FIGURE 53.?Plant cover on north point, May 1978: Ipomoea pescaprae, Cakile lanceolata (foreground, with 0.25 m frame), and freshly planted Casuarina tree. introduced by supply boats carrying produce. Fleas and ticks have been left behind by dogs, the former at times plaguing sensitive investigators. Flying insects are commonly blown over from land or larger islands during westerly winds. Most of the island's invertebrate fauna, however, con? sists of three crustaceans: the hermit crab Coenobita clypeatus (Herbst), and the crabs Ocypode quadrata (Fabricius) and Gecarcinus lateralis (Freminville). Only the lizard Anolis sagrei Dumeril and Bi- bron, a species widespread in the West Indies and apparently expanding its range onto Caribbean Mexico and Middle America (R. Crombie, pers. comm.), occurs as resident population of verte? brates on Carrie Bow Cay. A sea turtle, Caretta caretta (L.), was last seen laying eggs on the island on 28 May 1972 (A. Antonius, pers. comm.). Birds that feed regularly around the cay are the Brown Pelican {Pelecanus occidentalis L.), Frigate- bird {Fregata magnificens Mathews), and Osprey {Pandion haliaetus (L.)). Other birds commonly seen include the Boat-tailed Grackle {Cassidix mexicanus (Gmelin)), Common Tern {Sterna hirundo L.), Brown Booby {Sula leucogaster (Boddaert)), Snowy Egret {Leucophoyx thula (Molina)), and Barn Swallow (Hirundo rustica L.). An assortment of involuntary visitors from the mainland, such as warblers and flycatchers, arrive exhausted on the island after periods of strong westerly winds. All birds, except the grackle and the swallow, roost eleswhere, most likely on South Water Cay. The grackle may even breed on Carrie Bow Cay because a female was observed gathering mate? rials for nest building. INTERTIDAL ORGANISMS.?The mean tidal range at Carrie Bow Cay is only 15 cm (Kjerfve et al., herein: 47, Table 4). The observed maximum range, however, partly because of wind forcing is more than 40 cm. With a shoreline slope of 90? to 4? the width of the intertidal zone on Carrie Bow Cay ranges between 40 cm on vertical cinder block walls and 6 m at the flat northern point, on the average between 0.5 and 2.0 m. Only during spring tides are wide areas on the reef flat exposed (Plate 1: bottom right). Sandy beaches have a diverse and rich interti? dal meiofauna (Kirsteuer, in prep.) but only one NUMBER 12 83 benthic macro-organism, the cerianthid Arachhan- thus nocturnus den Hartog, could be observed at low tide buried in exposed sand on the northeast shore. Ocypode quadrata crabs, however, temporar? ily establish burrows in sand areas exposed at low tide. Rocky substrates support a more varied inter? tidal flora and fauna but differences in abun? dance can be observed between the leeward (west) and windward (east) sides of the island. Coral rock, rubble, and concrete blocks of the leeward sea wall are thickly covered by algae (Figures 52b, 54a). Oscillatoria submembranacea Ardissone and Strafforella and Schizothrix mexicana Gormont (Cyanophyta), and Cladophoropsis membranacea (C. Agardh) B0rgesen (Chlorophyta) occupy the up? per zone, Padina Jamaicans (Collins) Papenfuss (Phaeophyta) and Neomeris annulata Dickie (Chlo- FIGURE 54.?Intertidal organisms; a, algae Cladophoropsis and Oscillatoria (top), Neomeris and Padina (bottom) on concrete block seawall; b, gastropod Liltorina ziczac on coral boulder. (Picture width, a = 70 cm; b = 5 cm.) rophyta) the zone below. On the windward side only the calcareous green alga Halimeda opuntia (L.) Lamouroux, red Laurencia papillosa (Forsskal) Greville, and some of the Oscillatoria were found exposed. A few specimens of the actinian Stoichactis ane? mone (Ellis) and barnacle Tetraclita stalactifera (La? marck) were also encountered on the windward side. The most abundant crustaceans on the la? goon shore are the hermit crab Clibanarius tricolor (Gibbes), which clusters in great numbers on intertidal rock and rubble, and the elusive isopod Ligia olfersii Brandt, which is particularly com? mon around empty conch shells near the concrete boat dock. Several crabs are common among rubble and concrete blocks all around the cay. Grapsus grapsus L. is the largest and most abun? dant; other crabs include Pachygrapsus transversus (Gibbes), Cyclograpsus integer Milne Edwards, and Petrolisthes quadratus Benedict. Among the mol? lusks only gastropods occur intertidally at Carrie Bow Cay. On the windward side Nerita peloronta L., N. versicolor (Gmelin), Littorina nebulosa La? marck, L. ziczac (Gmelin) (Figure 54b), and Tec- tarius muricatus (L.) are found on vertical coral rock and concrete block surfaces of the seawall. Several size classes of juvenile Cittarium pica (L.) cluster among rubble or on beach rock below. Nerita versicolor, L. ziczac, and T muricatus also occur on the leeward seawall but are less abun? dant there. A few specimens of a single species of echinoderm, the echinoid Echinometra viridis Agas? siz, are found here and there under tide-exposed rocks. Climatic Parameters The climate of Belize is subtropical to tropical, with temperatures ranging from 10? to 36? C (average range in Belize City, 23?-33? C), and rainfall averaging 125-450 cm a year. Tempera? tures are lowest in the highlands and during the cool period of the year (November to March). Average rainfall increases from north to south; the rainy season lasts from June to October. The overall climate of the country, particularly of the outer cays, is influenced by northeasterly trade 84 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES winds that prevail at velocities of 4-5 m/s during about 70% of the year. Our meteorological rec? ords from Carrie Bow Cay, although not contin? uous, indicate major patterns of temperatures, solar radiation and cloudiness, wind, and rainfall and allow some comparisons with conditions pre? vailing on the mainland. In addition, 12-day continuous measurements of radiation, evapora? tion, wind, and air-water-sand temperatures in June 1975 were reported by Kjerfve (1978). TEMPERATURE AND SOLAR RADIATION.?Figure 55 presents monthly temperature records, except for July and December. Data for January, Feb? ruary, and August to October are the result of a single year's readings; other data were derived from at least three consecutive years of observa? tion. Values are plotted against a background of ten-year average minimum and maximum tem? perature readings provided by the Melinda Forest Station near Dangriga, on the mainland of Belize. Temperature conditions on the cay follow closely those on land, where the highest averages occur during May and August (33? C) and the lowest during January and February (22? C, 21 C). Solar radiation measurements are only available for the cay and for the months of March through June, and November. The highest total radiation reaching the ground on a single day was recorded during April and May and amounted to 490 cal/ cm2. Monthly averages of daily radiation related to this value give an indication of cloudiness and haze (Figure 55). WIND.?Measurements of wind direction, speed, and frequency on Carrie Bow Cay are summarized in Figures 56 and 57. Values for March-June show the typical situation: northeas? terly trade winds predominate and compare well with published wind roses from the open ocean surface off Belize (United States Naval Oceano? graphic Office, 1963). Our observations on wind for the rest of the year are sparse and may not be representative of long-term averages. Speed val? ues for the infrequent winds from the northwest sector are somewhat low because of the shading effect of the big house and of coconut trees. RAINFALL AND HUMIDITY.?Long-term rainfall ?c 40 -, 35 - 30 - / 8 3 % \ / 8 6 % \ / 7 7 % \ f70%\ (?) {JJ {?) (j^) 25 20 - Id A l l n M ( ^ m i- 40 - 35 - 30 - 25 - 20 1 1 1 I I I I I I I 1 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC ' A i r ? Sand A Reef Flat Lagoon W!\ Mainland FIGURE 55.?Monthly temperatures (mean, range) 1976-1980 and solar radiation (percentage of maximum) 1978-1980 for Carrie Bow Cay; monthly temperature range (shaded area) at Melinda Forest Station on mainland, averaged over a ten-year period, 1965-1975. NUMBER 12 85 Jan /Feb (25) 40.6 13.8 ? 20 30.7 < = - ( 371 Jun (98) 12.2 5.| >vl4.3 43.9 28.8 Oct/Nov (66) 19.7 FIGURE 56 (above).?Wind roses for Carrie Bow Cay indi? cating direction, speed, and frequency (figures in parentheses are numbers of observations during 1976-1980). FIGURE 57 (right).?Monthly summaries of wind speed fre? quencies during 1976-1980, Carrie Bow Cay. data taken at the Melinda Forest Station indicate an average annual accumulation of 218 cm for the Dangriga district. The range is from 4.4 cm in March to 30.4 cm in September. Values for Carrie Bow Cay are presented in Figure 58 and compared with the mainland averages. The is- 36.6 140.8 Legend 1-5 6-8 9-14 > I4 m/s -c i r 20 ? i 1 1 40 60% Frequency Legend /ColrnX ( ~i 0 i 20 -5 Speed 6 -8 1 ? Frequency 1 1 ' 40 ' 60 9- 14 > I 4 IT 80 i /s 100% Jan/Feb Mar & Apr May Jun Oct/Nov ^ I I I 1 L_ I I I I I 20 40 60 80 100% 86 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES mm 300 250 - 2 0 0 - 150 1 0 0 - 50 - 0 ?*d 1. Wm T: W. w - 3 0 0 - 250 200 M M N - 150 i-100 - 50 0 0 Daily Maximum Mainland FIGURE 58.?Monthly average (total bar) and daily maximum rainfall on Carrie Bow Cay (1976-1980), compared with mainland monthly rainfall (Melinda Forest Station) averaged over a 71-year period (1906-1977). land receives, on the average, only 42 prcent of the mainland rainfall if one excludes February and December for which comparative data are lacking. The high value (279 mm) for November may be a peculiarity of the year (1979) in which the record was taken. On the other hand, a second but incomplete measurement of 213 mm (1978, November 14-26) indicates a similarly high or even higher rainfall during that month. Humidity measured between March and June averaged 78 percent, with a range of 58-96 per? cent. Recent Hurricane Effects on Carrie Bow Cay Computer files of the United States National Hurricane Center (P. J. Hebert, pers. comm.) indicate that at least 20 hurricanes and 45 tropi? cal cyclones have passed within 100 nautical miles (185 km) of Belize City (17?30'N5 88?18'W) dur? ing the last century (records date back to Novem? ber 1889). From these data it can be determined that nine hurricanes and seven tropical storms have passed Carrie Bow Cay within a 50 km radius. Storm activity in this area seems to have increased recently as six of the hurricanes and the most violent of tropical storms (Laura) have oc? curred within the last 20 years (Table 8). Hattie is the only storm for which the long- term effects on Belizean reefs and cays, including Carrie Bow Cay, have been monitored (Stoddart, TABLE 8.?Hurricanes passing within 50 km radius of Carrie Bow Cay, 1960-1980, including name, date, and maximum sustained wind speed while storm center was within 50 km of Carrie Bow Cay Name Abby Anna Hattie Francelia Laura* Fifi Greta Month/ Year Jul 1960 Jul 1961 Oct 1961 Aug 1969 Nov 1971 Sep 1974 Sep 1978 Wind speed (km/h) 128 148 259 182 111 176 176 Officially declared a tropical storm. NUMBER 12 87 1963, 1969, 1974). Other recent hurricane reports include a brief eyewitness account of tropical cyclone Laura passing over Glover's Reef and Stann Creek (Dangriga) (Antonius, 1972), and observations on the impact of hurricane Greta on the reef community near Carrie Bow Cay (High- smith et al., 1980). HURICANE FIFI (14-22 September 1974).?A tropical depression south of Puerto Rico and Hispaniola moving westward developed into hur? ricane Fifi on 17 September. Fifi, as reported by Hope (1975), acquired its maximum sustained winds of 95 kt (176 km/h) while it moved along the coast of Honduras, 18-19 September, where heavy rains caused a high number of deaths by inland flooding of rivers. The hurricane crossed the barrier reef approximately 20 km south of Carrie Bow Cay and reached the coast of southern Belize during the afternoon of 19 September. Our observations on the effects of Fifi on Carrie Bow Cay rely on a survey in December 1974 as no eyewitness reports are available. Storm surge flooded the entire island, and most of the uncon? solidated sand was either piled up high inside the buildings or carried away, leaving a surface of coral rubble and exposed palm tree roots. A comparison of photographs (Figure 59) and a map of Carrie Bow Cay prepared by D. R. Stod? dart in 1972 (Figure 60a) indicate that coastal erosion was strongest to the north, northeast, and %JE w mtt ?** FIGURE 59.?Carrie Bow Cay silhouettes looking east: a, February 1972; b, December 1974, three months after hurricane Fifi. Note reduction of island size, in number of trees, and density of leaves caused by the hurricane. SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES a Old New AAAA Mean Tide Level Undercut Shoreline Submerged Sand Spit Coconut Palms ? Live o Planted after Fif & Stump FIGURE 60.?Maps of Carrie Bow Cay showing hurricane effects and poststorm recovery: a, changes 1972-1974, caused principally by hurricane Fifi, September 1974; b, recovery, Decem? ber 1974 to May 1978. NUMBER 12 89 ?-*? FIGURE 61.?Effects of hurricane Fifi: a, eroded north point of Carrie Bow Cay with fallen coconut palms, looking northeast; b, submerged bank of sand originating from northern portion of the island, looking northwest. 90 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES south of the island. About 30% of the island's surface area was lost, but some of it was regained by subsequent deposition along the western shoreline. Most island sand, however, settled as a subtidal sand bank to the northwest of the cay (Figure 61 b). The eastern seawall and outhouse dock were destroyed and a huge tree stump, situated for years on the reef flat to the east, was floated to the new south tip of the cay. At least 14 coconut trees, predominantly from the north point and northeast shore were uprooted by ero? sion and either fell in place (Figure 61a) or were carried into the lagoon and sank. Others lost their tops in the storm or withered from overexposure to salt water. All other plants previously recorded (Stoddart, 1969), such as low Tournefortia bushes, Euphorbia, Ipomoea, and Sesuvium ground cover, and grasses, disappeared and did not recover to the approximate prehurricane condition until spring 1978. By that time, with the help of seawalls and rubble fills, much of the prehurricane island out? line was restored (Figure 60b). HURRICANE GRETA (13-23 September 1978).? The track of Greta was almost identical to that of Fifi, and both storms occurred at almost the same time in September. The meterorological history of Greta is described by Lawrence (1979). A depression formed northwest of Trinidad on 13 September. Hurricane force with sustained winds of 115 kt (213 km/h) developed at a position south of Jamaica on 16 September. Moving over the Bay Islands off the north coast of Honduras, Greta weakened and made its landfall with 80 kt (148 km/h) winds near Dangriga on the evening of 19 September. Greta was a much more severe hurricane than Fifi but despite locally heavy rain it did not cause devastating river floods (P. J. Hebert, pers. comm.). The eye of hurricane Greta passed Carrie Bow Cay about 6 km to the north and brought the island winds of approximately 95 kt (176 km/h). Although direct observations are lacking, a con? siderable storm tide (about two meters above normal, estimated from events at Dangriga) must have flooded the island because the smallest cot? tage disappeared and the ocean-side wall of the laboratory caved in. Despite damage to buildings, coastal erosion was considerably lower than dur? ing Fifi. About 20 coconut trees were lost, most of these from the leeward side of the island. Other plants were much less affected by Greta than by the 1974 hurricane as only one common and four minor species disappeared (Table 7). Summary and Conclusions From its position, structure, and flora, Carrie Bow Cay can be classified as a reef-derived sand cay. It is located at the seaward margin of the barrier reef, is composed of reef rubble and sand, and is held together primarily by coconut rootlets and, to a lesser degree, by ground cover and artificial structures. The island measured a little over two acres (0.8 ha) when it was bought by the present owners in 1943. Today it is less than half that size and exposed beachrock on the wind? ward side indicates westward (leeward) migra? tion, which confirms the view that sand cays of this nature are slowly migrating sand waves (Mil- liman, 1973). Because of its small size, low elevation, and porous substrate, Carrie Bow Cay lacks a fresh? water lens and it has not developed a complex terrestrial environment. Considering the occa? sional salt water flooding during storm tides, the island may be described as a supralittoral habitat. The climate, too, is dominated by the surround? ing ocean and by northeasterly trade winds; it is moderate in comparison with the nearest main? land. Clearing of vegetation during this century and increased hurricane activity in the area during the past two decades are mainly responsible for the rapid shrinking of Carrie Bow Cay. Captain Owen, who mapped the island as "Jack Ellin's Cay" in 1830, noted "tops of bushes 20 feet" (Stoddart, 1963), presumably seagrape, baycedar, and mangrove. Coconuts planted in the early 1900s, and repeatedly again since, may not be equally effective in holding the sand, also, they do not protrude into the intertidal to trap sedi? ments or break the power of waves or currents. NUMBER 12 91 Although physiographic change of the cay was minor during hurricane Hattie (Stoddart, 1963, 1969), later storms, Fifi in particular, took severe toll. Recovery of plant cover destoyed by Fifi took about four years. Colonization of sand cays is thought to be primarily by floating seeds or by seeds carried by birds (Stoddart, 1960), or by wind, but direct observations on these processes are sparse. Possible means of dispersal judged from seed type are listed in Table 7 (data pro? vided by M.-H. Sachet). Our own findings sug? gest that only four of 16 species of plants?coco? nut, red mangrove, and two unidentified seed? lings?arrived by sea and sprouted. None of them survived beyond two years because of the unsuit? able location of settlement. Experimental studies on natural means of island colonization should be the next step in elucidating the terrestrial devel? opment of Carrie Bow Cay. Literature Cited Antonius, A. 1972. Hurricane Laura, Witnessed in British Honduras. Atoll Research Bulletin, 162:11-12. Highsmith, R. C , A. C. Riggs, and C. M. D'Antonio 1980. Survival of Hurricane-generated Coral Fragments and a Disturbance Model of Reef Calcification/ Growth Rate. Oecologia (Berlin) 46:322-329. Hope, J. R. 1975. Atlantic Hurricane Season of 1974. Monthly Weather Review, 103:285-293. Kirsteuer, E. In prep. Horizontal and Vertical Migrations of Ololyphlo- nemertes (Nemertinea) in a Sandy Beach at Carrie Bow Cay, Belize. Kjerfve, B. 1978. Diurnal Energy Balance of a Caribbean Barrier Reef Environment. Bulletin of Marine Science, 28: 137-145. Lawrence, M. B. 1979. Atlantic Hurricane Season of 1978. Monthly Weather Review, 107:477-491. Milliman, J. D. 1973. Caribbean Coral Reefs. In O. A. Jones and R. Endean, editors, Biology and Geology of Coral Reefs, 1:1-50. New York and London: Academic Press. Stoddart, D. R. 1960. The Reefs and Sand Cays of British Honduras. In Cambridge Expedition to British Honduras, 1959- 1960, General Report, pages 16-22. Cambridge, England. 1963. Effects of Hurricane Hattie on the British Hon? duras Reefs and Cays. October 30-31, 1961. Atoll Research Bulletin, 95:1-142. 1969. Post-Hurricane Changes on the British Honduras Reefs and Cays: Re-survey, 1965. Atoll Research Bulletin, 131:1-31. 1974. Post-Hurricane Changes on the British Honduras Reefs: Re-survey, 1972. In A. M. Cameron et al., editors, Proceedings of the Second International Coral Reef Sympsoium, 2:473-483. Brisbane, Australia: The Great Barrier Reef Committee. United States Naval Oceanographic Office 1963. Oceanographic Atlas of the North Atlantic Ocean, Section IV: Sea and Swell. United States Naval Oceanographic Office, Publication, 700: 227 pages. Washington, D.C. Distribution of Microborers within Planted Substrates along a Barrier Reef Transect, Carrie Bow Gay, Belize Jeffrey A. May, Ian G Macintyre, and Ronald D. Perkins ABSTRACT A diverse assemblage of endolithic microorgan? isms was identified in a series of carbonate sub? strate stations planted along a transect across the barrier reef near Carrie Bow Cay, Belize. These tropical endoliths at the sediment-water interface include the cyanophytes, Hyella tenuior Bornet and Flahault, Mastigocoleus testarum Lagerheim, and Plectonema terebrans Bornet and Flahault, the chlo- rophytes, Ostreobium brabantium Weber Van-Bosse and Phaeophila engleri Reinke, the rhodophyte Por? phyra sp. (Conchocelis-phase), and various fungi. This assemblage was subdivided into an upper photic zone assemblage dominated by Mastigoco? leus, Hyella, Phaeophila, and Ostreobium species, and a lower photic zone assemblage dominated by Porphyra sp. Subsurface endolithic activity detected at the shallow lagoon station included filamentous irreg? ular polygonal networks, irregular flattened masses, and regular crenulate discoids, which dif? fered from and were less diverse than the assem? blage at the sediment-water interface. Affinities of these subsurface microborings are unknown but they resemble endolithic traces and organic scars variously attributed to fungi, bacteria, and Actinomycetes. The regular discoids and irregular masses occurred only in association with the fila? mentous form, and therefore may be related re? productive bodies. For reasons not fully under- Jeffrey A. May, Department of Geology, Rice University, Box 1892, Houston, Texas 77001. Ian G. Macintyre, Department of Paleo? biology, National Museum of Natural History, Smithsonian Institu? tion, Washington, D. C. 20560. Ronald D. Perkins, Department of Geology, Duke University, Box 6665, College Station, Durham, N. C 27708. stood, microborings were not present in the sec? ond subsurface station, in fore-reef sand at a depth of 24 m. Introduction Endoliths are microorganisms (generally less than 1 jtim to 100 jum in diameter) that penetrate calcareous substrates by chemical and/or me? chanical means and that leave post-mortem mi? croscopic networks. They are distinguished from "epiliths," which live only on a substrate's sur? face, and from "chasmoliths," which adhere to surfaces of fissures or cavities within the substrate (Golubic et al., 1975). Endoliths include cyano? phytes, chlorophytes, rhodophytes, fungi, and possibly bacteria and sponges. The most diverse and ecologically important microborers occur in the marine setting; their boring patterns?size, mode of branching, spatial arrangement, and growth directions?are taxo- nomically characteristic (Golubic, 1972). The ma? rine endoliths have been subdivided on the basis of bathymetric and regional assemblage distri? butions, which are controlled by geographic, cli? matic, photic, and environmental factors (Perkins and Halsey, 1971; Perkins, 1972; Rooney and Perkins, 1972; Golubic et al., 1975; Green, 1975). Various studies of ancient and modern forms have indicated that endoliths may be used to interpret paleoclimatic conditions and to recon- 93 94 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES struct depositional environments (for example, see Swinchatt, 1965; Gatrall and Golubic, 1970; Perkins and Tsentas, 1976). As well, microboring organisms have been used to establish the posi? tions of relict shorelines (Perkins and Halsey, 1971; Edwards and Perkins, 1974) and to detect sediment transport (Rooney and Perkins, 1972). Microborers modify both lithified and unlithi- fied carbonate coasts (Purdy and Kornicker, 1958; Hodgkin, 1970; Schneider, 1976), in the colonization of shifting upper sublittoral sub? strates?where they create finer carbonate sedi? ments, preferentially remove certain components, and initiate micrite rind formation (Bathurst, 1966; Golubic, 1969; Alexandersson, 1972; Roo? ney and Perkins, 1972; Perkins and Tsentas, 1976)?and in their alteration of the deep-sea sedimentary record (Zeff and Perkins, 1979). En? dolithic algae not only dissolve carbonates, but also induce precipitation of calcium carbonate within shallow marine corals (Schroeder and Ginsburg, 1971; Schroeder, 1972a, b; Scherer, 1974). Furthermore, their biologically related physico-chemical processes may influence low- temperature sedimentary mineralization within boring networks (Taylor, 1971; Kobluk and Risk, 1977). Recently, planted substrates have been used to identify modern microboring assemblages and to establish rates of infestation (Golubic, 1969; LeCampion-Alsumard, 1975; Perkins and Tsen? tas, 1976). The present investigation examined and identified endoliths within planted substrates in order to determine the distributional patterns of these organisms at the sediment-water interface along a reef transect. Two stations planted below the sediment surface allowed a cursory examina? tion of subsurface microboring activity. ACKNOWLEDGMENTS.?The authors thank M. R. Carpenter (Smithsonian Institution) and R. G. Zingmark (University of South Carolina) for field assistance, ar\d G. W. Lynts and R. B. Searles (Duke University) for helpful comments on the manuscript. Support for part of this re? search was provided by National Science Foun? dation Grant OCE74-12555 A01 to Duke Uni? versity. Methods This study examined two types of carbonate substrates planted at and below the sediment- water interface along a reef transect of Carrie Bow Cay (Figure 62): (1) crushed, fresh inner shell parts of the queen conch, Strombus gigas Linnaeus, and (2) cleaved calcite rhombohedra. Samples of each type were retained as controls. Fragments from 1 to 10 mm in size were attached to 15 cm2 plexiglass plates (one type per plate) and 40 cm long polyvinyl chloride (PVC) pipes (types mixed) by a thin film of epoxy resin. Fourteen substrate-covered plexiglass plates were mounted on short lengths of protruding PVC pipe to maintain the samples above shifting sub? strata (Figure 63). These plates were placed in pairs at seven locations extending from the la- goonal Thalassia zone (depth 1.2 m) to the fore- reef slope (depth 27.4 m). One subsurface pipe station (consisting of one pipe) was inserted into the sea floor in the Thalassia zone (Figure 64) and another into the Halimeda-rich sand of the fore- reef sand trough (depth 24 m). Exposure time of substrates ranged from 21 to 24 months. After being harvested, the planted substrates were preserved in 4 percent formaldehyde in 0.1 M phosphate neutral buffer. Fragments intended for light microscopic study were carefully scraped to remove epilithic organisms, then dissolved with 5 percent EDTA solution (van Reine and van den Hoek, 1966) at a pH of 6. Although the three- dimensional configuration of the microborers is lost because of their collapse, organic structure and color are not damaged by this slow-dissolving solution. Extracted tissues were then mounted on glass slides. Scanning electron microscopic (SEM) analysis was based on the casting-embedding technique of Golubic et al. (1970). An alcohol dehydration series was followed by an infiltration series using Durcupan ACM Araldite Base Embedding Agent. After polymerization within plastic hold? ers, the substrate fragments were exposed by means of a rotary grinding tool, then etched with 3 percent hydrochloric acid. This technique re? vealed plastic casts of the microboring networks, NUMBER 12 95 DEPTH AND tOCATION | 2 m OF SUBSTRATE ' STATIONS ft ft SPORANGIAL STRUCTURES ( Fungal filaments found in al l substrate stations FIGURE 62.?Idealized cross section of Carrie Bow Cay transect showing locations of substrate stations planted at sediment-water interface and the distribution of endoliths collected from these plates (hatched squares = great abundance; hatched rectangles = presence observed). which retained the original spatial relationships of the endoliths and their three-dimensional con? figurations. The mounted plastic blocks were scanned with an International Scientific Instru? ments Super II electron microscope after vacuum shadowing with gold-palladium alloy. Casts of boring networks were correlated with the endo? liths isolated by acid dissolution. Results ENDOLITHS AT THE SEDIMENT-WATER INTER? FACE.?Blue-green algae were ubiquitous and consisted of Hyella tenuior Bornet and Flahault, Mastigocoleus testarum Lagerheim, and Plectonema terebrans Bornet and Flahault. Green algal micro? borers, likewise abundant, included Ostreobium bra- bantium Weber Van-Bosse and Phaeophila engleri Reinke. Red algae were much less abundant and were represented only by the Conchocelis-stage of Porphyra sp. Fungal forms were found in almost all samples. Plectonema terebrans was the most common cy- anophyte at all stations. Diagnostic are its smooth, elongate, thread-like filaments 2 to 4 jum in diameter, which may run along the interior 96 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 63.?Patch reef zone substrate station, Carrie Bow Cay, showing location of plates above the sediment-water interface. FIGURE 64.? Thalassia-zone subsurface substrate station, Car? rie Bow Cay, showing a closeup of the buried substrate- covered 40 cm long PVC pipe. Note that only the protective collar and a small portion of the pipe protrude above the sediment-water interface. surface of the substrate or may form dense, inter? woven meshworks (Figure 65a,b). Also ubiquitous was Mastigocoleus testarum, which is composed of sharply curved to elongate filaments 5 to 8 /xm in diameter that have numerous short lateral branches and heterocysts (Figure 65c,d). The mor? phology of M. testarum penetrating inorganic sub? strates is much more affected by the rhombo- hedral microstructure than is the morphology of P. terebrans; this observation corresponds with findings of LeCampion-Alsumard (1975). Hyella tenuior, which is less common, appears as a cluster of slender, elongate, relatively straight to bent filaments 5 to 8 /xm in diameter (Figure 66a,b). These filaments grow subperpendicular to the surface of the substrate. FIGURE 65.?Scanning electron and transmitted light pho? tomicrographs of endolithic algae: a, Plectonema terebrans form? ing a typical dense network of filaments, acid-etched mollusk fragment, Thalassia zone (note the smooth, elongate, fine nature of the plastic casts); b, P. terebrans isolated by disso? lution of a mollusk fragment, reef-crest zone; c, characteristic heterocyst development of Mastigocoleus testarum shown on plastic casts, acid-etched mollusk fragment, Thalassia zone; d, M. testarum isolated by dissolution of a mollusk fragment, Thalassia zone (note the heterocyst development marked by arrows). (Scale = 50 /xm for a, d; 40 jLtm for b; 25 fim for c.) NUMBER 12 97 98 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES 1 ( j < ^ ^ 9 k i ' v ^*v i ; >' ** d NUMBER 12 99 Phaeophila engleri was the most abundant and widespread chlorophyte. This species is charac? terized by rectilinear branching and pronounced bulbous or irregularly ovoid, 15 to 20 jum swell? ings at points of branching or along the irregular 5 to 10 /im filaments (Figure 66c,d). The largest boring species detected, Ostreobium brabantium has digitate growths of single or bilobate branches (Figure 67a,b). Single plants up to 1 mm in length radiate into the substrate; individual branches of 40 to 60 jum may enlarge up to 120 /xm before bifurcation. The Conchocelis-stage of the rhodo- phyte Porphyra sp. is characterized by rectilinear branching of long and fine, 2 to 3 /xm filaments running along slightly beneath the substrate sur? face (Figure 67 c,d). Extremely fine filaments from less than 1 up to 4 /xm in diameter occur in a wide variety of forms, from non-branched to extensively branched and fused, sparse to massive networks (Figure 68a,b). These filaments probably represent fungal hy- phae. The hyphae appear to be directed towards algae and can be observed penetrating these or? ganisms, presumably in the act of feeding (Figure 68b). Three different types of structures observed with the scanning electron microscope were at? tributed to fungal spore cases. Form A has 5 to 15 /xm ovoid to pyriform reproductive bodies, from the bases of which radiate long and thread? like, i to 2 jum hyphae (Figure 69a). These hyphae are typically unbranched and connect spore cases. Form B has 8 to 20 /xm, globose to oblong fruiting bodies with connective hyphae attached at their FIGURE 66.?Scanning electron and transmitted light pho? tomicrographs of endolithic algae: a, elongate casts of the blue-green alga Hyella tenuior directed away from the sub? strate surface, some branching near their bases, plastic cast within an acid-etched calcite rhomb, Thalassia zone (very little control is exerted by the substrate microstructure); b, filaments of//, tenuior displaying elongate, somewhat rectan? gular cells, isolated by dissolution of the molluscan substrate, Thalassia zone; c, green alga Phaeophila engleri exhibiting characteristic ovoid swelling at points of branching and rectangular branching pattern, plastic cast of etched mollusk fragment, Thalassia zone; d, P. engleri demonstrating probable sporangia (arrows) and swellings at branching points, fila? ments isolated by dissolution of a molluscan substrate, Thal- assia zone. (Scale = 50 ju,m for a; 25 iim for b, c; 40 ttm for d.) sides, so that these spore cases appear positioned along filaments (Figure 696). The irregular hy? phae of less than 1 jum diameter become branched and fused, and form interconnected networks. Fungal form C has 5 to 15 /xm globose to discoid sporangial bodies with indented ends that appear "doughnut-shaped" (Figure 68c). These bodies occur alone or in clusters directly upon the surface of endolithic algae; connective hyphae are lack? ing. ENDOLITHS BELOW THE SEDIMENT-WATER INTER? FACE.?In contrast to the diverse assemblage of microorganisms boring into the substrates planted at the sediment-water interface, only a few endolithic forms were found below the sur? face. These boring organisms were present only in the molluscan fragments planted within the lagoonal Thalassia zone. Classification is problem? atic as these forms have not previously been described. No microboring activity was recorded at the subsurface station in the fore-reef sand- trough zone. An extremely irregular, polygonal network of variable and intermittent filaments 5 to 7 /xm in diameter (Figure 70a,b) commonly crosses the regular and parallel lines representing remnants of the organic matrix that separated the inorganic crystals of the gastropod shell. This network of variously spaced filaments bores just below the surface of the molluscan substrates. The filaments are curvilinear, have a twisted appearance, and branch sideways. This form is recurrent in ap? proximately 10 percent of the molluscan frag? ments throughout the pipe planted in the Thal- lasia zone, and was not found in the control samples or in samples from the sediment-water interface. Possibilities for taxonomic assignment include marine fungi (Phycomycetes, Ascomy- cetes, and Deuteromycetes) and filamentous bac? teria (Actinomycetes). Another group occurs as regular patches or regular crenulate discs associated only with the irregular polygonal networks described above (Figure 70c,d). These may be reproductive struc? tures or separate endolithic forms; they range from 125 to 200 /xm in diameter and average 2 /xm in thickness. Although they may be remnants 100 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Ms?!;** - * NUMBER 12 101 FIGURE 68.?Scanning electron photomicrographs of endolithic fungi: a, little-branched, thin plastic casts believed to be fungal hyphae, intertwined with the larger blue-green alga Plectonema terebrans, etched mollusk fragment, Thalassia zone; b, network of fine fungal borings covering and possibly feeding (arrows) upon the underlying alga, plastic casts within an etched mollusk substrate, Thalassia zone. (Scale = 5 /xm for a; 25 /xm for b.) of the organic matrix, no analogous structures were found in any control sample or in any sample infested at the sediment-water interface. FIGURE 67.?Scanning electron and transmitted light pho? tomicrographs of endolithic algae; a, large, radiating growth form of the green alga Ostreobium brabantium, plastic cast within an acid-etched mollusk fragment, Thalassia zone (note both the single and bilobate branches, background forms are algae and fungi); b, 0. brabantium filaments isolated by dis? solution of a molluscan substrate, patch-reef zone; c, char? acteristic rectilinear pattern of the Conchocelis-stage of the red alga Porphyra sp., plastic cast, enhanced on a microscale by boring within a calcite rhomb, fore-reef slope zone (larger forms are an unidentified alga with strong microstructural control upon its boring pattern); d, Conchocelis-stage of Por? phyra sp. displaying fine filaments in the typical rectilinear pattern, isolated by dissolution of a molluscan substrate, fore-reef slope zone. (Scale = 200 /xm for a; 100 jum for b; 25 /xm for c, d.) Discussion ENDOLITHS AT THE SEDIMENT-WATER INTER? FACE.?The bathymetric distribution of Belizean endoliths was compared with similar tropical mi? croboring assemblages recovered from artificial substrate stations planted in reefs off St. Croix and Jamaica (Green, 1975; Perkins and Tsentas, 1976). The distribution of autotrophic endolithic organisms is related to light penetration in the sea?both the intensity of illumination and spec? tral composition (Golubic et al., 1975). Although endolithic organisms cannot be assigned to abso? lute depths?owing to variations in water clarity, currents, and other environmental factors?Per? kins and Tsentas (1976) pointed out that "clear- water" assemblages might be used to estimate maximum depths for endolithic algae. Their di- 102 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 69.?Scanning electron photomi? crographs of endolithic fungi: a, form A, characterized by ovoid to pyriform repro? ductive bodies, distributive unbranched casts of hyphae radiate from the base of the fruiting bodies just below the surface of the substrate, plastic cast of an etched calcite rhomb, fore-reef slope; b, repro? ductive bodies of fungal form B laterally connected by hyphae, etched mollusk fragment, Thalassia zone (these bodies are typically globose to oblong, note twisted or irregular appearance of the hyphal casts); c, "doughnut-shaped" or indented sporangial bodies of fungal form C, plas? tic cast in an etched mollusk substrate, Thalassia zone (note that these occur in close proximity to the outer substrate sur? face or to algal borings). (Scale = 10 /xm for a, b; 25 /xm for c.) NUMBER 12 103 vision of such assemblages into an upper photic zone of Mastigocoleus, Hyella, Phaeophila, and Os? treobium species, and a lower zone dominated by Porphyra sp. in its Conchocelis-phase parallels, to some extent, the zonation found in the present study (Figure 62). Ostreobium brabantium was observed only in the shallowest (1.2 and 1.5 m) sites of the Thalassia and patch-reef zones off Belize. In St. Croix, although present to 30 m, 0. brabantium was pre? dominant in depths less than 15 m (Perkins and Tsentas, 1976). Also, only the three shallowest sites off Belize contained Hyella tenuior, which was found down to 45 m in Jamaica (Green, 1975). Four species of algae occurred at all depths along the Carrie Bow Cay transect. Phaeophila engleri was most abundant from 1.2 m to the 7.6 m spur and groove zone of the inner fore reef; similarly, this species is very common in the shallower zones of St. Croix and Jamaica. Masti? gocoleus testarum was most abundant in the upper 12 m of the Belize sites, as was the case in St. Croix. Off Belize, however, it decreased slightly in the reef-crest and Thalassia zones. Plectonema terebrans was ubiquitous off Belize as well as Ja? maica and St. Croix, except that in Belize it was less abundant in the Thalassia and reef-crest zones. Patchily distributed below 1.5 m, Porphyra sp. was most common in depths of 12 to 27 m, the deepest zone examined off Belize. This pat? tern has also been reported from the Australian Barrier Reef (Rooney and Perkins, 1972), the Puerto Rico shelf (Perkins, 1972), and Woods Hole, Massachusetts (Golubic et al., 1975), as well as Jamaica and St. Croix (Green, 1975; Perkins and Tsentas, 1976). Although thin filaments believed to be fungal hyphae were present at all depths, the distribu? tion of different sporangial structures was depth dependent. Fungal form A, with pyriform to ovoid bodies, from which basal hyphae radiate, was found only in the sample from 27 m on the fore-reef slope. Similarly, Perkins and Tsentas (1976) found a reticulate fungal form at 30 m in St. Croix. Forms B and C with laterally arranged globose spore cases and isolated "doughnut- shaped" bodies, respectively, were found only in the shallow reef-crest, patch-reef, and Thalassia zones off Belize. ENDOLITHS BELOW THE SEDIMENT-WATER INTER? FACE.?Fungi may be as important as bacteria in breaking down organic matter into a nutrient source for other organisms. Marine fungi may be as versatile and potent in their feeding activity as their terrestrial and fresh-water counterparts, and are probably able to attack the entire spectrum of plant and animal detritus (Johnson and Spar? row, 1961). Endolithic fungi in living and dead shells at the sediment-water interface produce intertwined, anastomosing, and branched net? works, and are able to use the energy contained in organic conchiolin and chitin matrix and to parasitize algal microborers (Kohlmeyer, 1969; Calvaliere and Alberte, 1970; Green, 1975; Zeff and Perkins, 1979). Although marine fungi are generally believed to be restricted to the aerobic surface layers of bottom sediments, the Thraustochytriacea and Chyritidiaceae commonly occur well below the sediment surface (Clokie, 1970; Bremer, 1976; Johnson, 1976). The similarity between irregular polygonal networks observed in this study and microborings attributed to fungi leaves little doubt that these polygonal networks are of fungal origin. The associated irregular masses and crenulate discoids may be the reproductive bodies of these boring fungi. Kohlmeyer (1969) described irreg? ular black peritheca and conidia 100 to 125 /xm in diameter as fruiting bodies for endolithic ma? rine Ascomycetes and Deuteromycetes. Alterna? tively, these masses may be separate endolithic fungal colonies, similar to irregular patches and to crenulate rosettes attributed to fungal colonies attacking spores, pollen, and other organic micro? fossils (Elsik, 1971). In addition, two other groups of organisms, the bacteria and Actinomycetes, could produce en? dolithic scars resembling these forms. Similar scars on modern and ancient spores and pollen have also been attributed to bacteria and Acti? nomycetes (Moore, 1963; Elsik, 1966, 1971; Hav- 104 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES NUMBER 12 105 inga, 1971). Both groups are abundant at all levels within bottom sediments (ZoBell and Fel- tham, 1942) and at all depths sampled in lakes and the ocean (ZoBell and Rittenberg, 1938; Weyland, 1969; Willoughby, 1976). Boring by bacteria and Actinomycetes into carbonate sub? strates has not yet been investigated. The heterotrophic or chemoautotrophic mode necessitated by endolithic life within buried sed? iments may explain the occurrence of these forms only in the molluscan shell fragments planted in the Thalassia zone. The lack of endoliths within inorganic calcite rhombs planted below the sedi? ment-water interface may indicate parasitic or saprophytic requirements of these organisms. Bor? ing might be an effort to obtain nutrients from the organic matrix of the molluscan substrates. No reasons are known for the lack of these endo? liths in the deeper fore-reef sand-trough zone of Carrie Bow Cay. Although both this environment and the Thalassia zone consist of a carbonate sand bottom, the latter may contain more interstitial detrital organic matter or may be a more reducing environment. The availability of nutrients may control the distribution of these endoliths and therefore may explain the paucity of these forms within coarse sediments lacking necessary nutri? ent sources. Eh conditions within sediments may also be a controlling factor. Endolithic boring in this case may not be an "active" search for or? ganic matrices in carbonate substrates, but a "passive" result of metabolic reaction or a form of protection from interstitial grazers. FIGURE 70.?Scanning electron and transmitted light pho? tomicrographs of unidentified microborings collected in sub? surface station in the Thalassia zone: a, irregular polygonal network of varying intermittent borings, plastic cast within an etched mollusk fragment, 20 cm below the sediment- water interface; b, irregular polygonal endolithic network within molluscan substrate, collected 35 cm below the sedi? ment-water interface; c, irregular, flattened aggregates asso? ciated with a polygonal network, these patches might be related reproductive bodies or separate colonial forms, plastic cast of acid-etched molluscan substrate; d, crenulate discoid associated with a polygonal network, this also might be a separate organism or related reproductive body, plastic cast of an etched mollusk fragment, 5 cm below the sediment- water interface. (Scale = 50 /xm for a, c, d; 25 /xm for b.) PALEOECOLOGIC AND GEOLOGIC SIGNIFICANCE.? Endoliths and their borings found in carbonate sediments that are exposed on the sea floor may provide valuable information for the study of paleoecological conditions of carbonate sedi? ments. The microborers and their distributional patterns at the sediment-water interface of Belize closely resemble those of assemblages previously examined in St. Croix, Jamaica, and Florida (Perkins and Tsentas, 1976). It thus becomes possible to identify endoliths typical of a tropical shallow marine setting and to establish their oc? currence in upper photic and lower photic zones. Microborings commonly are preserved within an? cient carbonates (Hessland, 1949; Gatrall and Golubic, 1970; Golubic et al, 1975), but they have not been examined in relation to recent zonations for the interpretation of paleoenvironmental con? ditions. Conclusions Carbonate substrates, both conch shell frag? ments and cleaved calcite, planted just at the sediment-water interface in various depths along the reef transect off Carrie Bow Cay, Belize, contained a diverse assemblage of microboring forms. The blue-green alga Hyella tenuior and the green alga Ostreobium brabantium were restricted to the shallowest sample sites of the upper photic zone, which is also characterized by the abundant blue-green alga Mastigocoleus testarum and the green alga Phaeophila engleri and very little of the Conchocelis-stage of the red alga Porphyra sp. The lower photic zone, below approximately 12 m, is characterized by abundant Porphyra sp. and con? siderably less M. testarum and P. engleri. The blue- green alga Plectonema terebrans was abundant at all depths examined. Hyphae of fungal endoliths were present at all sample sites, although different sporangial forms were bathymetrically restricted. This study of algal endoliths supports previous findings of a distinct tropical assemblage that may provide a basis for paleoecological studies of ancient assemblages. Below the sediment-water interface off Belize, endoliths infested only the molluscan (conch) 106 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES fragments in the Thalassia-zone station and were not present in material buried in a deeper fore- reef sand-trough station. Restriction of infestation to the molluscan shell fragments suggests that these subsurface endoliths require organic matri? ces as nutrient sources. Three types of endoliths occur below the sediment-water interface: (1) irregular filamentous networks, (2) irregular flat? tened amorphous masses, and (3) regular crenu? late discoids. These microborers have unknown taxonomic affinities but they closely resemble endolithic traces and scars attributed to fungi, bacteria, and Actinomycetes. Further studies are needed to explain the origin and geological im? portance of these possible environmental indica? tors and post-depositional carbonate degraders. Literature Cited Alexandersson, T. 1972. Micritization of Carbonate Particles: Processes of Precipitation and Dissolution in Modern Shallow- Marine Sediments. Uppsala University Geological In? stitutions, Bulletin, new series, 3:201-236. Bathurst, R.G.C. 1966. Boring Algae, Micrite Envelopes, and Lithifica? tion of Molluscan Biosparites. Geological Journal, 5: 15-32. Bremer, G. B. 1976. The Ecology of Marine Lower Fungi. In E.B.G. Jones, editor, Recent Advances in Aquatic Mycology, pages 313-333. New York: John Wiley and Sons. Cavaliere, A. F., and R. S. Alberte 1970. Fungi in Animal Shell Fragments. Journal of the Elisha Mitchell Scientific Society, 86:203-206. Clokie, J. 1970. Some Substrate Relationships of the Family Thraustochytriaceae. Verbffentlichungen des Instituts fur Meeresforschung in Bremerhaven, 12:329-351. Edwards, B. B., and R. D. Perkins 1974. Distribution of Microborings within Continental Margin Sediments of the Southeastern United States. Journal of Sedimentary Petrology, 44:1122- 1135. Elsik, W. C. 1966. Biological Degradation of Fossil Pollen Grains and Spores. Micropaleontology, 12:515-518. 1971. Microbiological Degradation of Sporopollenin. In J. Brooks, P. R. Grant, M. Muir, P. van Gijzel, and G. Shaw, editors, Sporopollenin, pages 480-511. New York: Academic Press. Gatrall, M., and S. Golubic 1970. Comparative Study on Some Jurassic and Recent Endolithic Fungi Using Scanning Electron Micro? scope. In T. P. Crimes and J. C. Harper, editors, Trace Fossils. Journal of Geology, Special Issue, 3:167- 178. Golubic, S 1969. Distribution, Taxonomy, and Boring Patterns of Marine Endolithic Algae. American Zoologist, 9: 747-751. 1972. Scanning Electron Microscopy of Recent Boring Cyanophyta and Its Possible Paleontological Ap? plication. In R. V. Desikachary, editor, Proceedings of the Symposium on Taxonomy and Biology of Blue-Green Algae, pages 167-170. India: Bangalore Press. Golubic, S., G. Brent, and T. LeCampion 1970. Scanning Electron Microscopy of Endolithic Al? gae and Fungi Using a Multipurpose Casting- Embedding Technique. Lethaia, 3:203-209. Golubic, S., R. D. Perkins, and K. J . Lukas 1975. Boring Microorganisms and Microborings in Car? bonate Substrates. In R. W. Frey, editor, The Study of Trace Fossils, pages 229-259. New York: Sprin? ger. Green, M. A. 1975. Survey of Endolithic Organisms from the North? east Bering Sea, Jamaica, and Florida Bay. 112 pages. Master's thesis, Department of Geology, Duke University, Durham, North Carolina. Havinga, A. J. 1971. An Experimental Investigation into the Decay of Pollen and Spores in Various Soil Types. In J. Brooks, P. R. Grant, M. Muir, P. van Gijzel, and G. Shaw, editors, Sporopollenin, pages 446-478. New York: Academic Press. Hessland, I. 1949. Investigation of the Lower Ordovician of the Sil- jan District, Sweden, II: Lower Ordovician Pene? trative and Enveloping Algae from the Siljan District. Uppsala University Geological Institutions, Bulletin, 33:409-428. Hodgkin, E. P. 1970. Geomorphology and Biological Erosion of Lime? stone Coasts in Malaysia. Geological Society of Ma? laysia, Bulletin, 3:27-51. NUMBER 12 107 Johnson, T. W. 1976. The Phycomycetes: Morphology and Taxonomy. In E.B.G. Jones, editor, Recent Advances in Aquatic Mycology, pages 193-211. New York: John Wiley and Sons. Johnson, T. W., and F. K. Sparrow 1961. Fungi in Oceans and Estuaries. 668 pages. Weinheim: J. Cramer Publishing. Kobluk, D. R., and M. G. Risk 1977. Algal Borings and Frambiodal Pyrite in Upper Ordovician Brachiopods. Lelhaia, 10:135-143. Kohlmeyer, J. 1969. The Role of Marine Fungi in the Penetration of Calcareous Substances. American Zoologist, 9:741- 746. LeCampion-Alsumard, T. 1975. Etude experimentale de la colonisation d'eclats de calcite par les Cyanophycees endolithes marines. Cahiers de Biologie Marine, 16:177-185. Moore, L. R. 1963. Microbiological Colonization and Attack on Some Carboniferious Miospores. Palaeontology, 6:349- 372. Perkins, R. D. 1972. Microboring Organisms and Environmental In? dicators and Sediment Tracers: S. W. Puerto Rico Shelf [Abstract]. Geological Society of America, Ab? stracts with Programs, 1972 Annual Meeting, page 625. Perkins, R. D., and S. D. Halsey 1971. Geologic Significance of Microboring Fungi and Algae in Carolina Shelf Sediment. Journal of Sedi? mentary Petrology, 41:843-853. Perkins, R. D., and C. I. Tsentas 1976. Microbial Infestation of Carbonate Substrates Planted on the St. Croix Shelf, West Indies. Geo? logical Society of America, Bulletin, 87:1615-1628. Purdy, E. G., and L. S. Kornicker 1958. Algal Disintegration of Bahamian Limestone Coasts. Journal of Geology, 66:97-99. Reine, W.F.P. van, and C. van den Hoek 1966. Isolation of Living Algae Growing in the Shells of Molluscs and Barnacles with EDTA (Ethylenedi- aminetetraacetic Acid). Blumea, 14:331-332. Rooney, W. S., Jr., and R. D. Perkins 1972. Distribution and Geologic Significance of Micro? boring Organisms within Sediments of the Arling? ton Reef Complex, Australia. Geological Society of America, Bulletin, 83:1139-1150. Scherer, M. 1974. The Influence of Two Endolithic Microorganisms on the Diagenesis of Recent Coral Skeletons. Neues Jahrbuch fiir Geologie und Palaeontologie, Monatshefte, 9:557-566. Schneider, J. 1976. Biological and Inorganic Factors in the Destruc? tion of Limestone Coasts. Contributions to Sedimen? tology, 6:1-112. Schroeder, J. H. 1972a. Calcified Filaments of an Endolithic Alga in Re? cent Bermuda Reefs. Neues Jahrbuch fiir Geologie und Palaeontologie, Monatshefte, 7:16-33. 1972b. Fabrics and Sequences of Submarine Carbonate Cements in Holocene Bermuda Cup Reefs. Geoli- gische Rundschau, 61:708-730. Schroeder, J. H., and R. N. Ginsburg 1971. Calcified Algal Filaments in Reefs: Criterion of Early Diagenesis [Abstract]. American Association of Petroleum Geologists, Bulletin, 55:364. Swinchatt, J. P. 1965. Significance of Constituent Composition, Texture and Skeletal Breakdown in Some Recent Carbon? ate Sediments. Journal of Sedimentary Petrology, 35: 71-90. Taylor, B. J. 1971. Thallophyte Borings in Phosphatic Fossils from the Lower Cretaceous of Southeast Alexander Is? land, Antarctica. Palaeontology, 14:294-302. van Reine. See Reine, W.F.P. van Weyland, H. 1969. Actinomycetes in North Sea and Atlantic Ocean Sediments. Nature, 223:858. Willoughby, L. G. 1976. Freshwater Actinomycetes. In E.B.G. Jones, editor, Recent Advances in Aquatic Mycology, pages 673-698. New York: John Wiley and Sons. Zeff, M. L., and R. D. Perkins 1979. Microbial Alternations of Bahamian Deep-Sea Carbonates. Sedimentology, 26:175-201. ZoBell, C. E., and C. B. Feltham 1942. The Bacterial Flora of a Marine Mud Flat as an Ecological Factor. Ecology, 23:69-78. ZoBell, C. E., and S. C. Rittenberg 1938. The Occurrence and Characteristics of Chitino- clastic Bacteria in the Sea. Journal of Bacteriology, 35:275-287. Production of Some Benthic Communities at Carrie Bow Cay, Belize Paul E. Hargraves ABSTRACT The primary production of Thalassia testudinum, mixed algae on coral rubble, and coral sand in the vicinity of Carrie Bow Cay, Belize, was esti? mated from in situ oxygen flux (by titrimetry) in enclosed chambers. Thalassia beds were auto? trophic, yielding a daily net production in the range of 2.22-42.3 mg O2 per square meter (ap? proximately equivalent to 0.67-12.7 g C per square meter). The other habitats were hetero? trophic despite considerable microalgal develop? ment, with a production/respiration ratio (P/R) less than unity. Contribution by the phytoplank- ton was negligible. Diurnal oxygen flux in open waters over Thalassia beds was typical for similar tropical areas. Introduction In tropical coastal waters, coral reef ecosystems are generally considered to be highly productive in contrast to plankton communities, whose pri? mary production is quite low or even negligible (Lewis, 1977; Milliman and Mahnken, 1969; Sournia, 1969). The metabolism of coral reefs as a unit has been the subject of numerous studies beginning with Sargent and Austin (1949) and Odum and Odum (1955). The remarkable fertil? ity of reefs, their dynamic balance, and suscepti? bility to environmental perturbation have often been noted (see reviews by Lewis, 1977, and Johannes in Wood and Johannes, 1975). The plants responsible for primary production in tropical coastal waters include mangroves, sea Paul E. Hargraves, Graduate School of Oceanography, University of Rhode Island, Kingston, R. I. 02881. grasses, macroscopic algae, boring and epipelic algae, zooxanthellae, and phytoplankton. The relative importance of each group varies from area to area, but except for special environments such as enriched lagoons and upwelling areas (Sournia, 1969; Steeman-Nielsen, 1975), the con? tribution of phytoplankton is low. The techniques of investigating primary productivity have varied among investigators according to whether infor? mation is required on the productivity of the total community or its components, and according to the facilities and methods available. In general, chemical or polarographic measurements of oxy? gen flux are sufficiently precise for nonplanktonic biotopes, and these techniques have provided considerable information on the metabolism of tropical benthic communities (for instance, Lewis, 1977; McRoy and McMillan, 1977). Reported here are results of production mea? surements in several habitats at and near Carrie Bow Cay, Belize. Topographic, oceanographic, and ecological characteristics of the area are pre? sented elsewhere in this volume. ACKNOWLEDGMENTS.?The technical assistance of D. Hargraves and K. Zimmerly is gratefully acknowledged. Partial support was provided by National Science Foundation Grant OCE-76- 82280. N. Marshall reviewed an early draft of the manuscript. Methods Oxygen flux was measured in three habitats: seagrass beds (Thalassia testudinum Banks ex Konig with small amounts of Syringodium filiforme Kiitz- ing) at depths of 1 to 3 m on the lagoon side of 109 110 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Carrie Bow Cay; mixed algal communities on coral rubble at depths of 0.5 to 1.0 m in the rubble and pavement zone behind the reef crest; and coral sand at a depth of 1 m in the back-reef zone. The coral rubble was composed of cobble- sized pieces of coral mixed with sand and with small (unidentified) filamentous, crustose, and thallose seaweeds. The coral sand first seemed devoid of organisms, but microscope examination revealed considerable numbers of blue-green al? gae, diatoms, Foraminifera with symbiotic algae, filamentous algae, and a multitude of other rep? resentatives of the micro- and meiobenthos. Experiments were conducted over periods of dawn/noon, noon/sunset, and 24 h cycles with determinations at 1 or 2 h intervals, all during April/May 1977 and January 1978. For compar? ative purposes, the production of phytoplankton was also assessed. For each experiment three transparent and two opaque polystyrene boxes of 500 cm area were inverted over the substrate. The edges were pushed into the sediment to prevent leakage around the margin of the container. Water was withdrawn through sampling ports with a 50 ml capacity syringe. The total volume of enclosed water was 4.2 liters. For diurnal fluctuations in oxygen and phytoplankton production measure? ments, 300 ml biological oxygen demand (BOD) bottles were used. Dissolved oxygen was measured titrimetrically by the Winkler method, using phenylarsine oxide (PAO) as titrant. PAO is superior to thiosulfate in its longer shelf life and resistance to bacterial decomposition, and produces results with com? parable precision. During the experiments water temperature varied from 26? to 29?C (by mercury immersion thermometer) and salinity varied from 34.4 to 35.2 %o (by Endeco refractometer type 102). As much as possible, all experiments were conducted on days that were cloud free, or nearly so. Results and Discussion Results of a preliminary experiment to deter? mine the diurnal variation in dissolved oxygen in the open water over a Thalassia bed are shown in Figure 71. During predawn hours, characteristi? cally, minimum oxygen concentrations are ex? hibited. The subsequent rapid increase of dis? solved oxygen reflects photosynthetic activity (as light intensity increases). The daily oxygen max? imum occurs at or somewhat after noon, at which time minute streams of oxygen bubbles frequently issue from some Thalassia blades. A gradual de? cline in oxygen concentration occurs during the rest of the day, becoming more pronounced in the late afternoon. Minimum concentration is again reached in the predawn hours. This pattern is typical and well documented, not only for tropical seagrass communities (for instance, Odum et al., 1959; Qasim and Bhattathiri, 1971) but for coral reefs" as well (for instance, Sournia, 1976a). For part of the day, the water is supersat? urated, often in excess of 150%, with respect to oxygen. Some consequences of these conditions are discussed below. Table 9 summarizes the production of the dif? ferent communities. Oxygen data were converted to carbon assuming a photosynthetic quotient (PQ) of 1.25 (McRoy and McMillan, 1977; West- lake, 1963). The standing stock of Thalassia blades varied from about 34 to 126 g/m2 dry weight, with a mean of about 47 g. Other organisms were not analyzed quantitatively. Compared to published data for seagrass beds, summarized in Lewis (1977) and McRoy and Time (h) FIGURE 71.?Changes over a 24-hour period in dissolved oxygen over a Thalassia bed near Carrie Bow Cay, 28-29 April 1977. NUMBER 12 111 TABLE 9.?Oxygen production of communities at Carrie Bow Cay, 26 April-5 May 1977, unless otherwise indicated (gross and net carbon production assumes P Q = 1.25 (Westlake, 1963); phytoplankton assumes 1 m water depth) Community Thalassia beds Thalassia beds (Jan 1978) Mixed algae/ coral rubble Coral sand Phytoplankton Trials (n) 5 3 3 2 2 Production (mg 02/m2/d) Gross Net 19.3-118.1 14.9-33.2 60.0-94.3 75.5, 107.9 0.48, 1.00 6.03-42.30 2.22-14.50 163.0-31.6 -11.3,-32.1 0.14,-0.07 Respiration (mg 02/m2/d) 13.27-75.80 12.68-18.70 223.0-125.9 86.8, 140.0 0.34, 1.07 P/R 1.45-1.56 1.18-1.78 0.27-0.75 0.87,0.77 1.41,0.93 Production Gross 5.79-35.43 4.47-9.96 18.00-28.29 22.65, 32.37 0.14,0.30 (g C/m2/d) Net 1.81-12.69 0.67-4.35 -48.90-9.48 -3.39, -9.63 0.04, -0.02 McMillan (1977), the data for Thalassia at Carrie Bow Cay fall within previously noted ranges for gross and net production. The standing stock is rather low compared to several other localities in the Caribbean and Gulf of Mexico (refer to table 1 in McRoy and McMillan, 1977). In contrast to some Thalassia habitats in Puerto Rico (Odum et al., 1959), in the present case all measurements indicate a consistently autotrophic community, that is, ratio of production to respiration (P/R) in excess of 1. The wide range in measurements, in part attributable to density variations in Thal? assia and seasonal differences, make extrapola? tions to an annual production uncertain. For comparative purposes the mean production for all Thalassia measurements may be estimated at 1800 g C/m 2 /y net production. This amount is higher than those reported from seagrass beds in Puerto Rico (Odum et al., 1959), but consider? ably less than those reported from Cuba and Florida (Buesa, 1972; Odum, 1957, 1963). Differences in production were noted between late spring and winter. The extent to which these differences are significant in unknown because of the small number of replicates analyzed. Such variables as photoperiod (Marmelstein et al., 1968) and water temperature (citations in McRoy and McMillan, 1977) affect Thalassia production and add to the uncertainty of annual rate com? parisons based on extrapolation. A considerable area of the back-reef zone at Carrie Bow Cay consists of coral rubble and pavement rock overgrown with a variety of small but conspicuous filamentous and thallose algae. Despite a gross production generally exceeding that of the Thalassia beds (Table 9), the commu? nity as a whole is heterotrophic, having a P /R less than 1. This habitat harbors a diverse and abundant fauna of burrowing invertebrates that contribute to the high community respiration. Grazing effects of fishes and large invertebrates, not considered here, are probably also significant in the consumption of primary producers (Marsh, 1976). The relative contribution of microalgal species in this habitat has generally been ne? glected, although Qasim et al. (1972) reported net production rates of 365-800 g C/m 2 /y for similar types of algae from the Laccadive Archi? pelago. Overall, the algal oxygen production in this habitat is less than oxygen consumption by herbivores and other animals. Although the extensive coral sand areas north and northwest of Carrie Bow Cay appear to have few plant producers, microscopic examination revealed large numbers of benthic pennate dia? toms and filamentous blue-green algae (or cyano- bacteria), smaller numbers of Foraminifera with apparent endosymbiotic algae, and large num? bers of micro- and meiobenthos metazoans. Com? munity respiration slightly exceeded gross pro? duction, with P /R less than 1 (Table 9), although this habitat was closer to an autotrophic condi? tion than the rubble-pavement habitat. The con? tribution of bacteria to community respiration 112 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES was not evaluated but it may be significant (see for instance, Edwards, 1978). The gross produc? tion approached that of the Thalassia beds, and therefore indicated the high photosynthetic activ? ity here. That such sand areas are not necessarily het? erotrophic has been convincingly demonstrated by Sournia (1976b). Extensive development of the blue-green alga Oscillatoria limosa in sand at Moorea Island lagoon was responsible for an average gross production equivalent of over 6000 mg 0 2 / m 2 / d with P / R = 1.5-3.0, a highly au? totrophic habitat. Although blue-green algae are not abundant at Carrie Bow Cay, the potential contribution of coral sand areas should not be ignored in calculations of production in reef areas. Under some circumstances Foraminifera with en- dosymbiotic algae may dominate coral sands. Sournia (1976a) described sands from Takapota Atoll with a net production of 115-354 mg 0 2 / m 2 / h (= 43-133 mg C/m2 /h) and with pop? ulations consisting primarily of Foraminifera with endosymbionts. Small numbers of these protozoa were also present in sands at Carrie Bow Cay, but pennate diatoms dominated as primary pro? ducers. These sands could potentially assume the au? totrophic role that Sournia (1976a) described. A minor perturbation in the environment of orga? nisms already under natural stress (by light and temperature thermal, among others) may shift community composition to one of low density consisting of eurytolerant species (Wood and Jo? hannes, 1975). Several different species of blue- green algae, which certainly qualify as eurytoler? ant, can be observed in the sands of Carrie Bow Cay. Coral sand habitats have been largely ne? glected by ecologists, so that it would be useful to stress them experimentally in various ways in order to determine whether and how their trophic status is modified. Perturbation in the form of nutrient enrichment was attempted in one micro- atoll, resulting in significantly increased produc? tion (Kinsey and Domm, 1974). As expected, the production of the phytoplank? ton was very low, and near the limits for accuracy of the technique used (Table 9), supporting the conclusion that phytoplankton in the vicinity of coral reefs contribute a negligible amount to net production (Milliman and Mahnken, 1969; Sour? nia and Ricard, 1976). Although such shallow areas generally have, in addition to the normal complement of typically planktonic forms, a higher number of benthic diatoms swept into the water column by turbulence, net production re? mains negligible. For offshore Caribbean waters net production can be higher or lower than in reef areas, depending on locale. Beers et al. (1968) found 0.03-0.28 g C/m 2 /d offshore from Ja? maica; Steeman-Nielsen and Jensen (1957) found 0.14 and 0.19 g C/m 2 /d in the south-central Caribbean Sea. In contrast, Ricard (1977) found higher production in the lagoon than in the open ocean at Tahiti, but the converse was true at Lakeba lagoon. Higher productivity is unques? tionably possible under localized enriched con? ditions (Gordon et al., 1971; Margalef, 1975). The limitations of technique and procedures used in this work deserve brief mention. Produc? tion in tropical benthic communities has been studied in several ways: by flow respirometry, which measures changes in the flux of carbon dioxide or oxygen (upstream-downstream); by in situ light/dark bottle methods using oxygen flux (titrimetric and polarographic) and various radio? active tracers; and by direct measurement of changes in standing stocks. All these methods are subject to errors and various assumptions (for discussion, see Lewis, 1977; Sournia, 1976c; Vol- lenweider, 1974) that will not be examined here, except to note that the technique used here ap? pears to be satisfactory for benthic habitats. Un? questionably, accuracy decreases in the case of water supersaturated with oxygen; similarly, res? piration of animals is affected by reduced oxygen levels. Recycling of oxygen in lacunae of Thalassia blades is another potential source of error (Hart- mann and Brown, 1967). The accuracy of conversion of oxygen flux to its equivalent carbon varies widely depending on the photosynthetic quotient assumed. This cre? ates a problem in comparing the production rates NUMBER 12 113 calculated by different authors. Photosynthetic quotients ranging from 0.86 to 3.00 are possible. This work follows Westlake (1963) in the assump? tion that P Q = 1.25 is valid for natural tropical communities in favorable conditions. Two important sources of tropical primary pro? duction have not been considered here, namely the coral zooxanthellae and the larger seaweeds. The complex role of the former in reef trophic structure has been discussed by Lewis (1977) and Marsh (1976) among others, and the contribution of zooanthellae at Carrie Bow Cay is certainly worthy of investigation. Under some conditions the larger seaweeds contribute significantly to primary production, even more so than Thalassia beds (Doty, 1971; Wanders, 1976), but their de? velopment around Carrie Bow Cay is not exten? sive. Conclusions Of the benthic habitats investigated at Carrie Bow Cay, Thalassia beds were most productive, yielding a maximum gross production of 35.4 g C/m / d and a maximum net production of 12.7 g C/m 2 /d . Coral sand habitats and mixed algae with coral rubble habitats were heterotrophic in nature, despite considerable algal development. Net production by phytoplankton was negligible. The diurnal oxygen content of water flowing over Thalassia beds fluctuates in a manner typical of similar habitats previously described. Given the problems inherent in comparing a wide variety of techniques and procedures, the pattern of primary production at Carrie Bow Cay is typical of similar tropical areas. Addendum Since the writing of this contribution three pertinent papers have appeared. Two center on reef community metabolism elsewhere in the Car? ibbean (Puerto Rico, U.S. Virgin Islands) and also employ the Winkler method for oxygen de? terminations (Rogers, 1979; Rogers and Salesky, 1981). The third deals with primary production by microalgae in North Sea sediments comparing various techniques, including platinum electrode measurements that allow recording of dissolved oxygen microprofiles (Revsbech et al., 1981). Literature Cited Beers, J. 1968. Buesa, R 1972. Doty, M. 1971. Edwards 1978. R., D. M. Steven, and J . B. Lewis Primary Productivity in the Caribbean Sea off Jamaica and the Tropical North Atlantic off Bar? bados. Bulletin of Marine Science, 18:87-104. ?J- Produccion primaria de las praderas de Thalassia testudinum de la platforma noroccidental de Cuba. Cuba Centro de Investigaciones Pesquera, Revista Traba- 7'oj,3:101-143. S. The Productivity of Benthic Frondose Algae at Waikiki Beach, 1967-1968. University of Hawaii, Botanical Science Paper, 22: 119 pp. R.R.C. Ecology of a Coastal Lagoon Complex in Mexico. Estuarine and Coastal Marine Science, 6:75-92. Gordon, D. C , R. O. Fournier, and G. J. Krasnic 1971. Note on the Planktonic Primary Production in Fanning Island Lagoon. Pacific Science, 25:228- 233. Hartman, R. T., and D. L. Brown 1967. Changes in Composition of the Internal Atmos? phere of Submerged Vascular Hydrophytes in Relation to Photosynthesis. Ecology, 48:252-258. Kinsey, D. W., and A. Domm 1974. Effects of Fertilization on a Coral Reef Environ? ment?Primary Production Studies. In A. M. Cameron, B. M. Campbell, A. B. Cribb, R. En- dean, J . S. Jell, O. A. Jones, P. Mather and F. H. Talbot, editors, Proceedings of the Second International Coral Reef Symposium, 1:49-66. Brisbane, Australia: The Great Barrier Reef Committee. 114 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Lewis, J. B. 1977. Processes of Organic Production on Coral Reefs. Biological Review, 52:305-347. Margalef, R. 1975. Fitoplancton invernal de la laguna costera de Alvarado (Mexico). Anales del Instituto Botanico An? tonio Jose Cavanilles, 32:381-387. Marmelstein, A. D., P. W. Morgan, and W. E. Pequegnat. 1968. Photoperidism and Related Ecology in Thalasia testudinum. Botanical Gazette, 129:63-67. Marsh, J. A. 1976. Energetic Role of Algae in Reef Ecosystem. Micro- nesica, 12:13-22. McRoy, C. P., and C. McMillan 1977. Production Ecology and Physiology of Seagrasses. In C. P. McRoy and C. Helfferich, editors, Seagrass Ecosystem, Scientific Perspective, pages 53-87. New York: Marcel Dekker, Inc. Milliman, J. D., and C. V. Mahnken 1969. Appendix: Reef Productivity Measurements. Atoll Research Bulletin, 129:23-26. Odum, H. T. 1957. Primary Production of Eleven Florida Springs and a Marine Turtle Grass Community. Limnology and Oceanography, 2:85-97. 1963. Productivity Measurements in Texas Turtle Grass and the Effects of Dredging an Intracoastal Chan? nel. Publications of the Institute of Marine Science, University of Texas, 9:45-58. Odum, H. T., P. R. Burkholder, and J . Rivero 1959. Measurements of Productivity of Turtle Grass Flats, Reefs, and the Bahia Fosforescenti of South? ern Puerto Rico. Publications of the Institute of Marine Science, University of Texas, 6:159-170. Odum, H. T. and E. P. Odum 1955. Trophic Structure and Productivity of a Wind? ward Coral Reef on Eniwetok Atoll. Ecological Monographs, 25:291-300. Qasim S. Z., and P.M.A. Bhattathiri 1971. Primary Production of a Seagrass Bed on Kavar- atti Atoll (Laccadives). Hydrobiologia, 38:29-38. Qasim, S. Z., P.M.A. Bhattathiri, and C.V.G Reddy 1972. Primary Production of an Atoll in the Laccadives. Internationale Revue des gesamten Hydrobiologie, 57: 207-225. Revsbech, N. P., B. B. J0rgensen, and O. Brix 1981. Primary Production of Microalgae in Sediments Measured by Oxygen Microprofile, H14C03-Fix- ation, and Oxygen Exchange Methods. Limnology and Oceanography, 26:717-730. Ricard, M. 1977. Phytoplankton Contribution of Primary Produc? tivity in Two Coral Reef Areas of Fiji Islands and French Polynesia. In D. L. Taylor, editor, Proceed? ings of the Third International Symposium on Coral Reefs, 1:343-348. Miami, Florida; Rosenstiel School of Marine and Atmospheric Science. Rogers, C. S. 1979. The Productivity of San Cristobal Reef, Puerto Rico. Limnology and Oceanography, 24:342-349. Rogers, C. S., and N. H. Salesky 1981. Productivity oi Acropora palmata (Lamarck), Mac? roscopic Algae, and Algal Turf from Tague Bay Reef, St. Croix, U.S. Virgin Islands. Journal of Experimental Marine Biology and Ecology, 49:179-187. Sargent, M. C , and T. S. Austin 1949. Organic Productivity of an Atoll. Transactions of the American Geophysical Union, 30:245-249. Sournia, A. 1969. Cycle annuel du phytoplancton et de la produc? tion primaire dans les mers tropicales. Marine Bi? ology, 3:387-303. 1976a. Primary Production of Sands in the Lagoon of an Atoll and the Role of Foraminiferan Symbionts. Marine Biology, 37:29-32. 1976b. Ecologie et productivity d'une Cyanophycee en milieu corallien: Oscillatoria limosa Agardh. Phycol? ogia, 15:363-366. 1967c. Oxygen Metabolism of a Fringing Reef in French Polynesia. Helgolander Wissenschaftliche Meeresunter- suchungen, 28:401-410. Sournia, A., and M. Ricard 1976. Phytoplankton and its Contribution to Primary Productivity in Two Coral Reef Areas of French Polynesia. Journal of Experimental Marine Biology and Ecology, 21:129-140. Steeman-Nielsen, E. 1975. Marine Photosynthesis. 141 pages. New York: Elsev? ier Scientific Publishing Company. Steeman-Nielsen, E., and A. Jensen 1957. Primary Oceanic Production: The Autotrophic Production of Organic Matter Production in the Ocean. Galathea Report, 1:49-136. Vollenweider, R. A., editor 1974. A Manual for Measuring Primary Production in Aquatic Environments. 213 pages. Oxford and Edinburgh: Blackwell Scientific Publications. Wanders, J.B.W. 1976. The Role of Benthic Algae in the Shallow Reef of Curacao (Netherlands Antilles), II: Primary Pro? ductivity of the Sargassum Beds on the Northeast Coast Submarine Plateau. Aquatic Botany, 2:327- 335. Westlake, D. F. 1963. Comparisons of Plant productivity. Biological Re? views, 38:385-425. Wood, E.J.F., and R. E. Johannes, editors 1975. Tropical Marine Pollution. 192 pages. Amsterdam, Oxford, New York: Elsevier Scientific Publishing Company. [Elsevier Oceanography Series, 12.] Macrobenthic Invertebrates in Bare Sand and Seagrass {Thalassia testudinum) at Carrie Bow Cay, Belize David K. Young and Martha W. Young ABSTRACT The generally accepted view that seagrasses support a more dense and diverse invertebrate fauna than sand areas devoid of such vegetation was tested in Thalassia testudinum beds of Carrie Bow Cay lagoon, Belize. Mechanisms regulating the distribution of invertebrates were also exam? ined. Core samples from five stations, ranging from bare sand to dense seagrass, indicated no clear- cut effect of seagrass versus bare sand on densities of polychaetes and mollusks. The numbers of species from all seagrass stations were not signifi? cantly higher than those from sand stations. Spe? cies diversities for all stations were influenced mainly by species richness, except for the station in dense seagrass where species evenness was high. The most abundant polychaete and mollusk spe? cies sampled were found at all stations. The proportion of fine-grained sediments in the substrate increased progressively from bare sand into the seagrass. Encrusting coralline algae provide a major contribution to the fine sedi? ments of the grass beds. Seagrass standing crop increased from the small seagrass patch to the dense Thalassia bed. Inorganic components made up 49% of live and 70% of dead seagrass. This study found no relationship between sea? grass standing crop and species densities of inver? tebrates. Intense predation pressure is proposed as the primary regulating mechanism of species David K. Young, Naval Ocean Research and Development Activity, Department of the Navy, National Space Technology Laboratory Station, Miss. 39529. Martha W. Young, U.S. Fish and Wildlife Service, National Coastal Ecosystem Team, Slidell, La. 70458. densities in the Carrie Bow lagoon. Experimental field studies are needed to clarify predator-prey interactions. Introduction Seagrasses are widely thought to support a more dense and diverse invertebrate fauna than bare sediment devoid of such vegetation. Only recently has this generalization been subjected to rigorous quantitative sampling and experimental testing (Heck and Wetstone, 1977; Orth, 1977; Reise, 1977; Young and Young, 1978). The mech? anisms of distribution examined by these workers primarily concern (1) food, (2) living space, (3) predation, and (4) sediment stability. FOOD.?Because seagrasses decompose rapidly relative to other vascular plants in the marine environment, they are considered to be major contributors to the detrital food web of seagrass ecosystems (Fenchel, 1977). Heck and Wetstone (1977) suggested from their work in Panama that food reserves are abundant in seagrass ecosystems and that food is not a limiting factor. Young and Young (1978), however, have experimentally demonstrated that numbers of certain deposit- feeding invertebrates are food-limited in a sea? grass detritus-rich habitat in Florida and that significant increases in overall macrobenthic den? sities result from addition of detrital food with high nitrogen content. LIVING SPACE.?Increased surface area afforded 115 116 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES by topographically complex living and nonliving substrata has been defined as an important eco? logical parameter on the study transect at Carrie Bow Cay (Dahl, 1973; Riitzler and Macintyre, herein: 9). Dahl (1973) has estimated that the functional surface area of bare sand is increased over four times by the presence of seagrass in the Carrie Bow Cay lagoon. Young and Young (1978) have experimentally shown no significant de? creases in overall species densities of macroben- thos following the clipping and removal of sea? grass blades from an area in a Florida seagrass bed, although the abundance of several seagrass- associated species diminished. PREDATION.?Heck and Wetstone (1977) sug? gested that plant blades provide protection to epifauna from visually searching fish predators. Orth (1977) and Reise (1977) provided experi? mental evidence that seagrass roots and rhizomes decrease the effectiveness of crab predators upon infauna. Reise (1977) has also shown that coarse? grained sediments afford protection to infauna from crab predation more than fine-grained sed? iments. Because finer sediments accumulate more in seagrass beds relative to adjacent areas bare of vegetation (Orth, 1977), a clear-cut assessment of the effect of sediment size versus that of seagrass upon intensity of predation cannot be discerned in most seagrass areas. SEDIMENT STABILITY.?Seagrass roots and rhi? zomes bind loose sediments together, and their blades baffle waves; both actions assist in pre? venting sediment resuspension and erosion (Gins? burg and Lowenstam, 1958). Orth (1977) has demonstrated that the overall sediment stabiliz? ing effect of seagrasses is a major factor in in? creased densities and diversity of infauna of sea? grass beds in Chesapeake Bay and Bermuda. Our primary intent in this study was to deter? mine whether densities of macrobenthic inverte? brates in the Carrie Bow Cay lagoon were related to the presence of vegetation and, within the time constraints, which mechanisms were regulating the distribution of species. ACKNOWLEDGMENTS.?We thank B. Spracklin for his invaluable assistance during our stay in Belize. Appreciation is due T. Wolcott for sedi? ment analyses. This study was funded by a grant from the Atlantic Foundation to the Smithsonian Institution and by the Harbor Branch Founda? tion, Inc. Laboratory space was provided by Mis? sissippi State University Research Center and administrative support by the Naval Ocean Re? search and Development Activity. Study Area and Methods The lagoon floor of Carrie Bow Cay is covered by calcareous sediment comprised largely of coral, encrusting coralline algae, Halimeda, molluscan, foraminiferal, and echinoid material. The sand surface in areas lacking vascular vegetation is here and there covered with a fine filamentous algal "felt" (Dahl, 1973). The seagrass Thalassia testudinum Banks ex Konig grows profusely in these sediments, and grades from small, thinly vege? tated patches (from the back-reef region) to the extensive, dense beds of the lagoon. The Thalassia is interspersed with sparse stands of the seagrass Syringodium filiforme Kiitzing. Thalassia blades are heavily encrusted with coralline algae and con? spicuously devoid of epizoans, with the exception of Foraminifera. We extended the established transect (Riitzler and Macintyre, herein) from its beginning point at the buoy (0 m) lagoonwards 5 m into a dense bed of Thalassia (Figure 72). Station 1 (+23 m) was an area of bare sand containing no vegeta? tion, station 2 (+18 m) was a small patch of Thalassia, station 3 (+12 m) was in bare sand, station 4 (+5 m) was in Thalassia at the edge of an area densely covered with seagrass, and station 5 (?5 m) was inside the Thalassia bed. Because of the depth of water at the sampling locations (approximately 2 m) and the small tidal variation (0.2 m tidal range, Kjerfve, 1978), seagrasses in this area of the lagoon are never exposed at low tide. Salinity was 34 %o throughout the sampling period and water temperature varied from 27? to 28? C. The grass area sampled at station 2 is approximately 80 m from the nearest patch reefs. The sand areas (stations 1 and 3) are at the same depth as adjacent seagrass stations. These bare NUMBER 12 117 Lagoon 5 (-5m) x? Buoy (Om) 4 (+5m) 3 (+12m) 2 (+18m) Reef 1 (+2 3 m) FIGURE 72.?Diagrammatic representation of the transect in Carrie Bow Cay lagoon showing sampling stations for this study; the buoy marks the beginning of the main transect (Riitzler and Macintyre, herein), which was extended, for purposes of this study, 5 m into dense Thalassia (distance from zero point in parentheses). sand stations show no traces of scour or ripple marks. Samples were taken from 23 to 31 March 1976 with a PVC corer (9.45 cm in diameter) to a depth of 20 cm (= 1405 cm3) at five stations along a 28 m long transect. Each sample consisted of four replicate cores taken perpendicular to the transect at each sampling station. The number of replicate cores per sample was determined by analyses of pilot samples taken in dense Thalassia from the lagoon. Four replicate cores contained 90% or more of the species of polychaetes and mollusks expected according to the test of Gaufin et al. (1956). Samples taken in Thalassia included blades, roots, and rhizomes. Samples were sieved through 1.0 mm mesh Nytex screen, narcotized with 0.15% propylene phenoxytol in sea water, stained with rose bengal and fixed in 5% to 10% formalin in sea water. Organisms were transferred to 70% ethyl alcohol after 24 hours for sorting, identification and stor? age. In this study, the term "macrobenthos" refers to those benthic invertebrates retained on a 1.0 mm mesh screen. Sampling efficiency of 1.0 mm mesh was determined by screening one of the replicates from each station through both 0.5 mm and 1.0 mm mesh screen. Only whole animals, anterior fragments of polychaetes and live shells of mollusks were identified as to species and counted. All other macrobenthos were identified only to higher taxonomic categories and counted. One sediment sample (comprising four repli? cate cores) was taken with the PVC corer at each station. Particle sieve size was determined using a Ro-Tap and analyzed by the method of Folk (1974). Samples from each particle size category were examined microscopically. All seagrass blades were clipped at the sediment surface and shoots were counted from one-meter- square areas adjacent to stations 2, 4, and 5. Erect live seagrass blades and prostrate dead blades were collected separately from each clipped plot. Live and dead seagrass dried at 100? C for 48 hours and weighed provided an estimate of sea? grass standing crop at each station. Organic and inorganic composition was determined from the ash of both live and dead fractions processed at 550? C for 2 hours. A random sample of Thalassia blades that showed evidence of grazing by herbivorous fishes was collected, and bite marks were counted and measured. Data were analyzed (log2) using Shannon's information measure of species diversity, H ' (Pie- lou, 1966); species richness, d, was determined by the method of Margalef (1958), and species even? ness by E' (Buzas and Gibson, 1969). A one-way analysis of variance (ANOVA) using non-trans? formed data compared numbers of species as well as species densities at the 5 stations. The a pos? teriori Student-Newmans-Keuls (SNK) multiple comparison test was run, since in both cases the ANOVA indicated significant differences among means (Sokal and Rohlf, 1969). Significance for the ANOVA and SNK tests was chosen at the 95% level of confidence. 118 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES . 3 E 5 ^ CM CM CM CT) O") CO C O C O r^t-~ to to ID in in i n - * CO CO CO O O ? ?' ?c CM CM CM CO CO CO Th LO in io CO CM CO ^f o O ?i co ?i O CO O O O CO o o - CM CM o CM o CO o o o o Th o CM CM o o o o - - - o CO o o o o - S S "2 ?&. o o ? g E S B v -s ~3 ?? -c re -o re a re - 2 c ? 11i||i S C 2d b ' ? l 3 .B ?? -2 u 8 ^ h 2 5 ? ? ? 5 3 >- -S ? g *8 J= -a > ?S ? .S I 3 <5/) s_ q H 1 J is ?5 co CD O I-H CM CO i n UD i-? co r ~ ? . - < in oi in CM CT) CM Th Th o CO CM CO CM O CO o CM CM O T h c D i n c o c o i n o ? ' ? * ? o o c o o r - c r > i n i n o ~ ) C M c o - ^ < X ) ? ' ^ - ^ ^ i n c M i n o c M m o f - ^ c o o c o ? ? o c o i n i n c O ' ? i c o o c o c o ^ m r ^ ? ' c M i n o c o o c M o o o T h v o ? io NUMBER 12 119 co m CO C O C O CO CM CM CM CM CO COCO CO CT) O) o o o o i n o c M o o ? ' O O C M O C M i n T h ? ? o ?? ? o o ( M O O ? ? o ? ' ?' o o ? O ?i ?' CM uo ? co o o o o o o o o O O O O ?? ?' o o ?< o o o o o o o O O ?< ?' O O ?' O O ? ? O O O ?? ?" o o o o O ? CM ?. o o ? ? o o ? ? o o o o o o ?N a J3 r^> flj >-> ?3s * w "S. -c ?, C U S u ifl s pu st u a ^ a *o ^ bl in g; X S -2 be ia ~ c bo a be ?? C 1 0 >*?? -M 2 a ?? 8 -s ? -s ?? l&i ia la j l i s I aid!<5 ?5 ' ? -^w r^* ^** "^ C fr Cs sii ? re 53 OT -3-9 03 5 ? Q fc 0 R.S r ^ r~~ t^- ? ^ r ~ CM CM CM CM CM CM CM ^O o o O ?i o o o o o ? o ? ? o o o o o o O ?i o o o o o o o o o o ?> ?i ?i ? o o o o o o CM CM CM CM O O O ?? O O ? O ?' ?i O O o o o o Th co o Th CO CM ? o > 3 ' ? i id a V J3 OT i i sic i n ic 1*3 u -O T o 51 G -a ? - V da r a, ril l !H > a 3 a rr at O 2 -S a 3 J J. ?2 C 2 w a 1 s s ?S -2 .a re 5 "S g. 3 > <3 o -o 2 ?? ? fi a h E ~ ? re .c 6 3 ?3 OT ^ Zti .3 3 J3 ^T 03 120 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Results Polychaetes and mollusks represented 51% of 1960 individuals collected in 20 core samples. Nematodes, oligochaetes, and arthropods com? prised 47% and miscellaneous other groups (pre? dominantly sipunculids, anthozoans, and echi? noderms) comprised 2% of the total macroben? thos. Numerical dominance and the fact that our species identifications were most complete for polychaetes and mollusks led us to assume that these taxonomic groups adequately represent the macrobenthos at the community level of organi? zation. A comparison of rankings of polychaetes and mollusks from all five stations (Table 10), using a grand mean of two specimens per species per station as a criterion, shows that nine species were common to both bare sand and areas con? taining Thalassia; only three species {Platynereis dumerilii (Audouin and Milne-Edwards), Capitella capitata (Fabricius), and Phyllaplysia engeli Marcus) were found in seagrass alone. Mean densities (extrapolated) of 8000/m2 (polychaetes and mol? lusks) were sampled in bare sand and 6476/m (polychaetes and mollusks) in Thalassia. The faunal list given in Table 10 is not in? tended to be all inclusive. Several species found among grass blades from the clipped meter square areas did not appear in the core samples, for example, the mollusks Astrea phoebia Roding, Tri- colia thalassicola Robertson, and Alaba incerta (Or- bigny). Figure 73 shows numbers of individuals and species of polychaetes and mollusks per sample at each of the five stations. The number of individ? uals at station 3 was significantly (SNK) different 70 o 60 o pe r In di vid ua ls "o | 3 0 z 20 i _ - i A _ r - - i I I I 5 (-5m) Buoy (0m) 4 3 f*5m) (-?12m) Stations 2 H8m) 1 (+23m) FIGURE 73.?Numbers of individuals and numbers of species of polychaetes and mollusks per core (x + Sx) at five stations in the Carrie Bow Cay lagoon. 5.0 4.8 4.6 4.4 4.2 4.0 1 - i A 1 I I I ? ^ ^ ^ \ K / \ """""^ N. / \ \ / \ \ ? i I I I 0.8 0.7 LU 0.6 0.5 0.4 10.0 9.0 ??8 .0 7.0 6.0 - - - i i i i i - o-_ 1 I 1 1 1 1 1 1 1 1 ..A.. A . . . - ? " " ' " " ? ? - , , - ? ? " . . ? ? " " -A-"' A" A 5 (-5m) Buoy (0m) 4 3 (+5m) (+12m) Stations 2 (+18m) 1 (+23m) FIGURE 74.?Species diversity (H' with solid lines and solid circles), species evenness (E' with dashed lines and open circles), and species richness (d with dotted lines and open triangles) of polychaetes and mollusks at five stations in the Carrie Bow Cay lagoon; each sample consists of four repli? cate cores. NUMBER 12 121 TABLE 11.?Standing crop of Thalassia per square meter at three stations in the Carrie Bow Cay lagoon; dry weights and percentages of organic and inorganic (after ashing) fractions compared for live and dead blades Stati No. of shoots Total dry weight (g) Live blades Dead blades WL (g) Org. (%) Inorg. (%) Wt. (g) Org. (%) Inorg. (%) 117 136 422 104.04 79.59 338.07 58.73 57.49 178.67 51 51 52 49 49 48 45.31 22.10 159.40 27 34 31 73 66 69 .? 0 01 5 -1.0 0.0 +4.0 40 30 20 10 n - Station 4 -. / S \ . S* > . " ^ ^ , \ " 40 30 20 10 n i i i Station 5 T I 1 i 1 1 " \ \ * +1.0 +2.0 +3.0 Sediment Size (0) FIGURE 75.?Sediment size analyses at five stations in Carrie Bow Cay lagoon. Weight percentages of sediment sizes (?1.0 to +4.0 d>) at each station. (95% level of confidence) from tha t at stat ion 5. All o ther stations, whether conta in ing Thalassia or no vegetation, were not significantly different from one another . W h e n numbers of species at all stations were compared , stations 2 and 4 were significantly (SNK) different (95% level of confi? dence) from stations 1 and 5. Stat ion 3 was not significantly (SNK) different from all o ther sta? tions. Figure 74 shows that species diversity and rich? ness indices for all stations except station 5 reflect the same t rend shown by numbers of species. T h e higher species diversity at station 5 (dense Thal? assia) compared to stations 1 and 3 (bare sand) was largely due to the effect of high species evenness at station 5. Species evenness was higher in all seagrass stations than in bare sand stations. Grass blades included in core samples con? ta ined eight polychaetes that had burrowed into the tissue of the grass blade, in some cases leaving a trail of fecal pellets behind them. These repre? sent a variety of families: three capitellids, three syllids, one maldan id , and one sabellid. This phe? nomenon is not explained. Al though 216 addi t ional specimens of poly? chaetes and mollusks were added by the 0.5 m m mesh from five cores (one from each station), only one species (the mollusk, Caecum ryssotitum Folin) was found tha t did not occur in 1.0 m m mesh samples. Gravimetr ic measurements (Table 11) of bo th live and dead blades of Thalassia cl ipped from one meter square areas adjacent to stations 2, 4, and 5 show that seagrass s tanding crop increased from the pa tch at station 2 to the dense growth at station 5. Overal l , 49% of living plant mater ia l and 70% of dead grass were inorganic in compo? sition, whereas 5 1 % and 30% respectively were organic. Sediment size analyses (Figure 75) indicated tha t the percentage of fine sand (2 and 3 ) increases from station 1 to station 5. An examination of 20 randomly selected Thal? assia blades with bite marks revealed 5.5?0.9 {x + Sx) bites per blade, with a diameter of 7.8?0.3 mm {x + Sx). Discussion Although extrapolated densities of macroben? thos per meter square in the Carrie Bow Cay habitat (total fauna: 12,167/m2 in Thalassia; 16,750/m2 in bare sand) fall within the ranges reported for subtropical and tropical locations (Table 12), this study does not show a direct relationship among seagrass standing crop, spe? cies densities, and numbers of species. Although some invertebrates may benefit from the protec? tion and living space afforded by blades, roots, and rhizomes, the overall composition of seagrass and bare sand assemblages of macrobenthos in? dicates that there is no clear-cut effect of seagrass versus bare sand on species densities. In fact, species densities are significantly (SNK) higher at one of the bare sand stations (station 3) than at the dense seagrass station (station 5). Numbers of species from seagrass stations differ significantly (SNK) from those collected in bare sand stations, with the exception of station 5. Numbers of spe? cies at station 5, in dense Thalassia, do not sig? nificantly differ from those in bare sand stations (stations 1 and 3). The abundant species of polychaetes and mol? lusks are found in both sand and seagrass. Among the polychaetes, seven of the ten top ranked species are common to all stations. Only three species, the polychaetes Platynereis dumerilii and Capitella capitata and the opisthobranch Phyllaply- sia engeli, have relatively high densities in seagrass and not in bare sand. In this habitat, all three species live closely associated with seagrass blades: P. dumerilii builds tubes on the blades, C. capitata tunnels within the tissue of the blades, and P. engeli feeds upon seagrass epiphytes. In other studies comparing animal assemblages of sand and seagrass, the water depths, tidal range, hydrography, and sediment characteristics have differed from those considered here. For example, along the sampling transect at Carrie Bow Cay the sediment interface between Thalas? sia beds and adjacent bare sand is imperceptible and not evidenced by erosional ridges (Jackson, 1972) or sand bars (Orth, 1977). Nor is there any indication in the study area of erosion around the rhizomes of seagrass plants or exposure of the root system, in a pattern of "raised terraces" (Kikuchi and Peres, 1977). In other words, the transition from sand to seagrass off Carrie Bow Cay is marked only by the advent of vegetation and by a gradual lagoonward decrease in sediment grain size. There is no evidence of sedimentary bedload movement in the form of ripple marks (Santos and Simon, 1974). Because the bare sand areas studied are physically stable, no effect of sediment instability on densities or species composition of macrobenthos is evident. The overall small size of specimens indicates that few macrobenthic invertebrates survive to maturity in this habitat. Only a few specimens fall within the mature size ranges reported in the literature; the lack of such larger sizes off Carrie Bow Cay implies high and persistent predation. The Thalassia and adjacent bare sand environ? ments here may be defined, according to the model of Connell (1975), as having benign phys? ical conditions in which predators usually kill all young colonists each year, and thereby reduce competition for resources (for instance, food and space) and help to perpetuate a state of nearly continual recolonization. Connell's model ac? counts for an occasional event in time and space wherein certain prey species can reach invulner? able sizes. While seasonality may be less influential in a tropical than a temperate habitat, seasonal dif? ferences in densities of macrobenthos clearly should be expected in the habitat studied here. Nevertheless, our expectation that macrobenthic densities would correlate positively at any given time with seagrass was not substantiated by the evidence from this study. In Carrie Bow lagoon, a majority of Thalassia NUMBER 12 123 TABLE 12.?Published data on densities of macrobenthic invertebrates per square meter in seagrass and bare sand from subtropical and tropical benthic studies Reference Bloom et al., 1972 Virnstein, 1972 Santos and Simon, Young et al., 1976 McNulty, 1970 Brook, 1975 Brook, 1978 Murina et al., 1973 Young and Young, 1974 this study Seagrass - 33,485? 3994-14,236 - Substrate 4000 (1972) 1895 (1973) 292-10,644 4508 583 12,167 6476" Bare sand 510 17,151 8795" - 291 (1956) 108 (1960) - - 354 16,750 8000* Sieve mesh size (mm) 1.0 0.5 0.5 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Location Tampa Bay, Florida Tampa Bay, Florida Tampa Bay, Florida Indian River, Florida Biscayne Bay, Florida Card Sound, Florida Biscayne Bay, Florida Florida Bay, Florida Cuba Carrie Bow Cay lagoon, Belize ? Polychaetes only. Polychaetes and mollusks only. blades were thickly encrusted with epiphytic cor? alline algae at the time of sampling. Humm (1964) notes that such encrusted blades sink to the bottom when they are broken off the parent plant and are not readily carried to other areas by tidal currents as are nonencrusted blades. In addition, observations of seagrass along the tran? sect of this study suggest that individual Thalassia blades are increasingly weighted down by heavy growths of encrusting coralline algae at their tips, so that dead blades are commonly retained in situ. Gravimetric measurements of live and dead seagrass clipped from meter-square plots at sta? tions 2, 4, and 5 show that 49% of the total biomass of live grass and 70% of dead sea? grass consisted of inorganic material. Microscopic examination of this inorganic residue revealed mainly fragments of coralline algae. It is likely that coralline algal epiphytes on Thalassia con? tribute substantial carbonate sediments to the lagoon at Carrie Bow in a manner similar to that reported in Florida (Humm, 1964) and in Ja? maica (Land, 1970). Patriquin (1972) estimated that the annual carbonate production by encrust? ing coralline algal epiphytes on Thalassia in Bar? bados is 2800 g/m2. Grazing fish help to distribute a portion of the epiphytic crust on seagrass blades?sometimes widely beyond the seagrass beds (Randall, 1967). Numerous reef fishes, in mixed schools (mainly scarids and acanthurids), were observed moving between the patch reefs of the back reef area and the seagrass of the lagoon. Although schools of fish were observed among the seagrasses, individ? ual fish appeared to feed on seagrass blades and the sediment between. Direct feeding on seagrass is evident from many bite marks on the edges of the blades. Thomas et al. (1961) reported similar crescent-shaped bites on Thalassia blades, and Zieman (1974) attributed these bite marks to the scarid, Sparisoma radians. Conspicuous bands of bare sand between reef areas and adjacent sea? grass beds in the Virgin Islands have been related by Randall (1965) to heavy grazing by parrot fish (scarids) and surgeon fish (acanthurids). Primary invertebrate grazers of seagrass, such as echinoids (Margalef and Rivero, 1958), were not observed in the Carrie Bow lagoon. Until night observations are made, we cannot determine whether invertebrate-feeding fishes such as pomadasyids (grunts) migrate nocturnally from the reefs to the seagrasses of the lagoon, as 124 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES described in the West Indies by Kikuchi and Peres (1977). Decapod crustacean predation on other members of the macrobenthos may be im? portant as a regulator of species densities, as suggested by Young et al. (1976) in Florida sea? grasses. Caging experiments in both seagrass and bare sand areas, as in the Indian River lagoon, Florida, by Virnstein (1978), would help to deter? mine the relative impact of fish and decapod predation in the Carrie Bow lagoon. This study provides only indirect evidence to suggest that "within-community" predation may be intense among the macrobenthic species of Carrie Bow. Large numbers of bored shells of bivalves and gastropods were found in our sam? ples; in all cases, numbers of bored shells greatly exceeded numbers of live shells. The only drill species was Polinices lacteus (Guilding), a naticid reportedly responsible for heavy predation on a wide variety of mollusks in Thalassia beds in Jamaica (Jackson, 1972). Polychaetes of the fam? ily Syllidae were extremely abundant in samples from both bare sand and Thalassia of Carrie Bow lagoon; five species of syllids are ranked among the 10 most abundant macrobenthic species. Mu? rina et al. (1973) also reported large numbers of syllids in Cuban seagrass beds. Syllids are be? lieved to be carnivorous (Pettibone, 1963; Day, 1967), preying on other invertebrates. As often is the case, ecological information is lacking for most of the species collected here, so that details of predator-prey interactions among the macro? benthos remain largely speculative. Conclusions The data from this study do not substantiate the generalization that seagrasses support a more diverse and dense invertebrate fauna than areas without seagrasses. There are no significant dif? ferences in species densities of macrobenthos be? tween seagrass and bare sand off Carrie Bow Cay. Significant differences do not exist among num? bers of species of polychaetes and mollusks from all seagrass stations and from bare sand stations. We suggest that the primary mechanism re? sponsible for the regulation of macrobenthic spe? cies densities and diversities in the Carrie Bow lagoon is high and persistent predation. Intense predation pressure probably limits the presumed advantages of the more variable habitats and increased living space provided by seagrasses. Predator-prey interactions could be clarified by the use of field-experimental techniques such as caging. Sediment instability appears to be unimpor? tant in the regulation of macrobenthic species densities and diversities in the lagoon. Few macrobenthic species are associated exclu? sively with seagrasses off Carrie Bow Cay. The majority of species sampled are common to both bare sand and seagrass stations. Sediment size analyses showed a progressive increase of fine-grained sediments from bare sand into seagrasses. Important contributors to these fine sediments are coralline algal epiphytes on Thalassia. Literature Cited Bloom, S. A., J. L. Simon, and V. D. Hunter 1972. Animal-Sediment Relations and Community Analysis of a Florida Estuary. Marine Biology, 13: 43-56. Brook, I. M. 1975. Some Aspects of the Trophic Relationships among the Higher Consumers in a Seagrass Community (Thalassia testudinum Konig) in Card Sound, Flor? ida. 133 pages. Ph.D. dissertation, University of Miami, Coral Gables, Florida. 1978. Comparative Macrofaunal Abundance in Turtle Grass (Thalassia testudinum) Communities in South Florida Characterized by High Blade Density. Bulletin of Marine Science, 28:212-217. Buzas, M. A., and T. G. Gibson 1969. Species Diversity: Benthic Formaminifera in West? ern North Atlantic. Science, 163:72-75. Connell, J . H. 1975. Some Mechanisms Producing Structure in Natu? ral Communities: A Model and Evidence from NUMBER 12 125 Field Experiments. In M. L. Cody and J. M. Diamond, editors, Ecology and Evolution of Commu? nities, pages 460-490. Cambridge, Massachusetts: Belknap Press. Dahl, A. L. 1973. Surface Area in Ecological Analysis: Quantifica? tion of Benthic Coral-Reef Algae. Marine Biology, 23:239-249. Day, J. H. 1967. A Monograph on the Polychaeta of Southern Africa, Part I: Errantia. British Museum (Natural History), Publication, 656: 458 pages. London. Fenchell, T. 1977. Aspects of the Decomposition of Seagrass. In C. P. McRoy and C. Helfferich, editors, Seagrass Ecosys? tems, 4:123-146. New York: Marcel Dekker. Folk, R. J. 1974. Petrology of Sedimentary Rocks. 182 pages. Austin, Texas: Hemphill Publishing Company. Gaufin, A. R., E. K. Harris, and H. J. Walter 1956. A Statistical Evaluation of Stream Bottom Sam? pling Data Obtained from Three Standard Sam? plers. Ecology. 37:643-648. Ginsburg, R. N., and H. A. Lowenstam 1958. The Influence of Marine Bottom Communities on the Depositional Environment of Sediments. Jour? nal of Geology, 66:310-318. Heck, K. L., Jr., and G. S. Wetstone 1977. Habitat Complexity and Invertebrate Species Richness and Abundance in Tropical Seagrass Meadows. Journal of Biogeography, 4:135-142. Humm, H. J . 1964. Epiphytes of the Seagrass, Thalassia testudinum, in Florida. Bulletin of Marine Science Gulf and Caribbean, 14:306-341. Jackson, J.B.C. 1972. The Ecology of Molluscs of Thalassia Communi? ties, Jamaica, West Indies, II: Molluscan Popula? tion Variability along an Environmental Stress Gradient. Marine Biology, 14:304-337. Kikuchi, T., and J . M. Peres 1977. Consumer Ecology of Seagrass Beds. In C. P. McRoy and C. Helfferich, editors, Seagrass Ecosys? tems, 4:147-194. New York: Marcel Dekker Kjerfve, B. 1978. Diurnal Energy Balance of a Caribbean Barrier Reef Environment. Bulletin of Marine Science, 28: 137-145. Land, L. S. 1970. Carbonate Mud: Production by Epibiont Growth on Thalassia testudinum. Journal of Sedimentary Petrol? ogy. 40:1361-1363. Margalef, R. 1958. Temporal Succession and Spatial Heterogeneity in Natural Phytoplankton. In A. A. Buzzati-Trav- erso, editor, Perspectives in Marine Biology, pages 323-349. Berkeley and Los Angeles: University of California Press. Margalef, R., and J. A. Rivero 1958. Succession and Composition of the Thalassia Com? munity. In Association of Island Marine Laboratories, Second Meeting, Bermuda, Sep 17-21, 1958, pages 19- 21. McNulty, J. K. 1970. Effects of Abatement of Domestic Sewage Pollu? tion on the Benthos, Volumes of Zooplankton, and the Fouling Organisms of Biscayne Bay, Flor? ida. Studies in Tropical Oceanography, 9: 107 pages. Miami, Florida: University of Miami. Murina, V. V., V. D. Chukhchin, O. Gomez, and G. Suarez 1973. Quantitative Distribution of Bottom Macrofauna in Upper Sublittoral Zone of Northwestern Part of Cuba. In Investigations of the Central American Seas, pages 242-259. New Delhi: Indian National Sci? entific Documentation Centre. Orth, R. J. 1977. The Importance of Sediment Stability in Seagrass Communities. In B. C. Coull, editor, Ecology of Marine Benthos, 6:281-300. Columbia, S.C: Belle Baruch Library in Marine Science. Patriquin, D. G. 1972. Carbonate Mud Production by Epibionts on Thal? assia: An Estimate Based on Leaf Growth Rate Data. Journal of Sedimentary Petrology, 42:687-689. Pettibone, M. H. 1963. Marine Polychaete Worms of the New England Region, I: Families Aphroditidae through Troch- ochaetidae. United States National Museum Bulletin, 227: 356 pages. Pielou, E. C. 1966. The Measurement of Diversity in Different Types of Biological Collections. Journal of Theoretical Bi? ology, 13:131-144. Randall, J. E. 1965. Grazing Effect on Seagrasses by Herbivorous Reef Fishes in the West Indies. Ecology, 46:255-260. 1967. Food Habits of Reef Fishes of the West Indies. Studies in Tropical Oceanography, 5:665-847. Miami, Florida: University of Miami. Reise, K. 1977. Predation Pressure and Community Structure of an Intertidal Soft-Bottom Fauna. In B. K. Keegan, P. O. Ceidigh and P.J.S. Boaden, editors, Biology of Benthic Organisms, pages 513-519. New York: Pergamon Press. Santos, S. L., and J. L. Simon 1974. Distribution and Abundance of the Polychaetous Annelids in a South Florida Estuary. Bulletin of Marine Science, 24:669-689. Sokal, R. R., and F. J. Rohlf 1969. Biometry: The Principles and Practice of Statistics in 126 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Biological Research. 776 pages. San Francisco: W.H. Freeman and Company. Thomas, L. P., D. R. Moore, and R. C. Work 1961. Effects of Hurricane Donna on the Turtle Grass Beds of Biscayne Bay. Florida, Bulletin of Marine Science Gulf and Caribbean, 11:191-197. Virnstein, R. W. 1972. Effects of Heated Effluent on Density and Diver? sity of Benthic Infauna at Big Bend, Tampa Bay, Florida. 59 pages. M.A. Thesis, University of South Florida, Tampa. 1978. Predator Caging Experiments in Soft-Sediments: Caution Advised. In M. L. Wiley, editor, Estuarine Interactions, pages 261-273. New York: Academic Press. Young, D. K., M. A. Buzas, and M. W. Young 1976. Species Densities of Macrobenthos Associated with Seagrass: A Field Experimental Study of Preda? tion. Journal of Marine Research, 34:577-592. Young, D. K., and M. W. Young 1978. Regulation of Species Densities of Seagrass-Asso- ciated Macrobenthos: Evidence from Field Exper? iments in the Indian River Estuary, Florida. Jour? nal of Marine Research, 36:569-593. Zieman, J . C. 1974. Quantitative and Dynamic Aspects of the Ecology of Turtle Grass, Thalassia testudinum. In Recent Ad? vances in Estuarine Research, Proceedings of the 2nd International Estuarine Research Conference, Myrtle Beach, South Carolina, October, 1973, pages 541-562. A Submarine Gave near Columbus Cay, Belize: A Bizarre Cryptic Habitat Ian G. Macintyre, Klaus Riitzler, James N. Norris, and Kristian Fauchald ABSTRACT An unusual cryptic habitat having an extensive covering of serpulid worms has been discovered in a submerged Pleistocene cave in the Belize barrier-reef platform near Columbus Cay. Aggre? gates of serpulids, which have been named "pseu- dostalactites," project from the ceiling of this cave in the direction of a narrow opening (10 m long and less than 3 m wide) that breaches the roof of the cave at a water depth of 17 m. Apparently this opening has restricted the movement of water within the cave so that serpulid worms have become more abundant on the ceiling than other cryptobiota, which include some sponges, fila? mentous algae, mollusks, and bryozoans. The latter group, with the exception of boring bi? valves, occurs only within a radius of 25 m from the entrance of the cave. The pseudostalactites are composed mainly of serpulids belonging to two species of the Vermiliopsis glandigera infundibu- lum group and extend at least 40 m from the cave opening, which is the limit of our observations. Varying amounts of submarine cement consisting of magnesium calcite form a coating on, or matrix in, the serpulid aggregations. A barren sediment cone (very fine sand to mud) occurs at a depth of 30 m below the cave opening. and 5 cm in diameter that he had collected from a submarine cave near Columbus Cay. This frag? ment had two surprisingly different sides: one composed predominantly of serpulid tubes and the other composed of dense microcrystalline Mg calcite forming a knobby surface. In our subse? quent search for the source of this sample, we discovered an unusual megacryptic environment. This paper presents the results of our preliminary investigation of the biological and geological characteristics of a highly unusual habitat. ACKNOWLEDGMENTS.?We are grateful to H. Bowman, Jr., and A. Usher of Dangriga for bring? ing this submarine cave to our attention. We thank K. E. Bucher, W. Gerwick, G. L. Hendler, P. M. Kier, W. M. Kier, R. J. Larson, K. Muzik, A. B. Rath, and P. E. Videtich for their field assistance. G. L. Hendler, W. M. Kier, and A. B. Rath also took underwater photographs. For help with identifications we thank A. H. Cheetham (bryozoans), I. M. Goodbody (tunicates), and R. S. Houbrick and T. R. Waller (mollusks). R. B. Burke provided the aerial photograph of the cave site and V. Krantz provided other photographic assistance. I. Jewett drafted the diagrams. Introduction In 1977 a Belizean SCUBA diver sent us a "stalactite" sample approximately 15 cm long Ian G Macintyre, Department of Paleobiology; Klaus Riitzler and Kristian Fauchald, Department of Invertebrate Zoology; and James N. Norris, Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Description of the Submarine Cave AEREAL SETTING.?The submarine cave about 1 km northwest of Columbus Cay (17?00'N, 88?02'W) occurs in a Thalassia-covered lagoon (5 m deep) at a distance of about 3 km inside the outer edge of the Belizean barrier reef platform (Figure 76). The only other natural depression in 127 128 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES 88?10'W .^Mosquito Cay j A Columbus Cay \ & Cross C<> p Cay A % ?' B ' HKBIKM" ^ - Jf" 1 ,?--'~ > p Hnd?H.\ ^ B^k ^ l ^BB rf^ BBBfctBfl ?iW ^ . ^ ^ k X " - v A.i?V H^^H ? i ^ | J I V ' - BK?fe*a%fK^BBl E -*y| PF- BBF / : / - ) - H P ^ m.4*#;'?? ' %r JfJBjh -^^ R B/ ' i jJJ- '^ ? J i-'mP'Mmto- . < 44W , iB** J F Vxl ? B F f r w > ??'*-1m ?' .BBr ' L J B B T ' #)' . TBBB'P!^ '^ -^B/ 'i*r %?-^ttiw ?/ wr 4BB ?k V **T ?% ? !^9- ? f'???rjm-:,*fr UBB> ^SwBte^' V/J^s, ? ?' m t ?' /JjB y ?^5BBBBMiLi ._ i^BF'S. - , . ???....jfflBM. T B J i f c * ^ ^ . . . * * . - - . 2 B ? B * M H ^ " S B B W : ^ ^ S > - z ^ * ^ * ' ' > j y FA ^SP ^ a -liBBBBl B ^ ' '"^ # " Bf ^ ^' ^ ^ ^ | B W ^ E T ? -a, -jf m ? ? ?^? ' BBJ V r'J?>. J JP ?? 0.5%) on either side of the barrier reef at Carrie Bow Cay (number of samples in parentheses) Copepoda Nauplii Gastropoda larvae Fish ova Larvacea Polychaeta larvae Shrimp* larvae Brachyura larvae Pteropoda Cirripeda larvae Chaetognatha Lagoon (40) 53.0 31.5 3.0 2.4 2.2 1.9 1.1 1.1 0.9 0.8 0.7 Ocean (32) 52.7 29.7 2.0 1.2 8.3 3.9 0.3 0.3 1.0 0.7 0.5 * Euphausiacea, Penaeidea, Caridea. TABLE 14.?Results of Wilcoxon signed-ranks test applied to abundance of organisms in different sample groups com? pared on basis of mean number of individuals/m3 (data from Table 15; + = significant at P< 0.05; 0 = no significant difference) Sample groups compared Ocean : Lagoon Night : Day 0-0.5m : 2.5-3.0 m depth Total plankton Copepods and nauplii Others NUMBER 12 145 B 3 3 J= S ?7 II CM" g _ ^ K B ?7 II +1 co en co ? i Ol LO +1 tO CO ^ 01 o +1 CM , i "*< ? ' ? i C M +1 +1 +1 CM t O I O ?> m +1 o IO +1 o CO +1 -f +1 CO CO +1 to L.O 8? 6 co 9+8 o 0? 9 ? * ? * r~? ~^ i n ?i ? o 0+8 CM 0+6 oS ^ +i co o o +1 o CO o +1 r-~ LT) -^? +1 E- Z 2 $ fi o -C ?7 II IT) ^ S ?n II g ^ ? 7 II irj ^ c\i g cs t ^ ?7 II c\i CPi to CM Th +1 r- a> tO to CO +1 o ^ H to in C M ? i +1 +1 r-. o - ; to 01 o CO r - +1 o 30 .0 0? 1 to o C O ? ^-> C O +1 +1 CM i - i ?i d +1 +1 CO LO O C M - H ? ' ? i C M O l - H ?, _ r-: +1 +1 +1 CM CO CO o LO ?I ?i CO ? Ol +1 co to +1 co LO + 1 o C M ? I as J2 -a 3 o ci -a s O CO d d +1 +1 co ? C M ? < ? I C O E '> o .a R a > aj ? C . 3 j - ?- o 2 2 O -C CL 3 -3 C n O U C fi -5- < 146 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES 6 ? o o tan 11 a at o (A s- Ul J CO < D. fc' (0 o 1) X fc 3 C TJ C as C a o; TI U 3 O JO >^ XI -D OJ ts c V TJ U as rt X> oS X o ? V X fc p ti_ as oS TJ Ol Th e-> LO +1 LO CM to to O +1 o OJ + + -* ? ? ' -* +1 CO m co r~ O +1 o ? Th co CO +1 LO CM OS co o +1 "-" CM Th Th 0+0 ?' CM 2? 0 - ? ** 0+9 + o r?-0+9 m + OS co 0? 1 CO os ?* CO ?* CM ? +1 +1 j- os co tO CM ? "* CO r-' TH ^ CM CO O +1 +1 +1 ?i OS to CO CM O OS -* + + + + ? * Th + 1 o LO LO CO +1 + . ?i CO O +1 +1 +1 m in to O CM o CM CM CO ?l + + + + -: + + + Os CM 0+6 o to 0+6 o en to 0+8 o + + - H as t o Th O co to ^ d +1 +1 +1 CM cq r^ - _|_ co CM ? ^ r~ m CM d d +1 +1 +1 r - -H TJH o +1 +1 +1 +1 co o ? to ^ t\i N O + + + m O 7? 1 Th CM 8? 1 ? CM to 4? 4 r-? o CO 5? 0 o CM - H CM ? Q Th CO CM - d co - +1 +1 +1 +1 Th o o ?' Th co ? os ? m co o o to CM CM d co +1 +1 +1 _|_ CO ?I Tf t o CM O CO CO ?i CO OS ? d o +1 +1 +1 ? t o OS iri CM CM + tD r^ O +1 CM ^ H (TM rH +1 OS to to _ +1 to ^ 2 ?i u O ei O -C , o 2 > ? b H > oS ^ u y y T5 oS as 3 oj JS XI " p 3 O o T3 o. a o 2 ^ 2- fi o o T3 8,3 y ._, T3 3 ' TJ O oj CX W 2 y * - > a s a b "J ?' T3 J2 a l l as S u a i i j S - ' o ' o e - ^ ^ ' C ^ - 2 / g o J . b o ^ B g_^ 3 T3 >- o -c c OS 1 5 u O y 2 ?i . ?7 II >A>S. c\i S II Cs ^ B ^ Cs ? d +1 to oi L O L O d +1 Th c\i O OS d +1 CM CM 0 LO to 0 +1 CO os CM CM +1 00 CO 0 CO 0 +1 CO X va V Si ph on op ho ra C te no ph or a Pl at yh el m in th la rv ae r~- CO Th +1 r~ os to Th CM +i Th r-* co O CM +1 co r^ O O X ? +1 r-? co LO cs r~ +1 01 r^ OJ co X OS +1 r? oi CO 00 CO CM +1 OS iri CM co LO d +i CM t~ CM v a e Po ly ch ae ta la r + + CO Th d +1 x d + + + + u lts Po ly ch ae ta a d + + co L O d +1 ? f? os X +1 O oi to Th d +1 OS d rv a e Si pu nc ul id a la + + 0 r~~ d +1 r- d + + r~- co d +1 p X B ry oz oa la rv ae LO Th oi +1 os iri co ?n ? < +1 co CO CO m CO +1 CM Th 0 r~ OS +1 CO co Ol CO 0 co +1 X d OS to +1 os Th LO LO oi +1 r^ OS Th CM oi +1 ? oi rv a e G as tr op od a la ; _ n d +1 p - 1 m d +1 p m 0 d +1 OJ ? + X X d +1 0 oi O L O d +1 OS t? os d +1 ? ? oi 0 CM d +1 in d ? v ae Pe le cy po da la i CO 0 oi +1 X CO m X oS +1 co os 0 X d +1 co oi L O ?; .?1 +1 in ri 0 Th r~ +1 O CM OS X CM +1 04 OS OS CO X +1 OJ oi X ?; r^ +1 CO co Pt er op od a m n 1?1 +1 to ? CO I?H fl to co CO d +1 m d co L O d +1 p ? L O L O d +1 Th OJ CO X +1 CO oi ? iri +1 OS ri CO if) 1^ +1 OJ co Os X X +1 r^ r^ Th d +1 CO d + + + r-? CO d +1 m d m x 12 .1 40 .7 . 62 CM ? O +1 +1 +1 CM r- os ?^ d ? Tf ? CO CM OS O CO ^ OS CM ? d "I ? Th CM +1 +1 +1 cq CM in d os iri CO ? CM ? CM CO CO r~ co r* OS Th r^ m +1 -H CO Tf< - ) - X OS co OJ ? CO ?n 0 m Th co" ri 2 os os P r? co co +1 +1 +1 O CO O r^ CM co O r? co in r-- in 0 p Th in C M iri iri Oa Th ? +1 +1 +1 CM CM m N ri us Th CO ? CM M in ? CO d P ? x co ^ ? Oa m +1 +1 +1 co co os oi ? d os ? CO CO CM Oa to ri ?? CM r-~ ? Th +1 +1 , m 0 + CO ?\ r-~ to CM ? ? CM X CM CO Th CO CO ^ Tf ? O +1 +1 +1 00 X 00 oi co d 0 ?1 CM ? V C la do ce ra O st ra co da C op ep od a N au pl ii C ir ri pe da la rv + in 0 d +1 r-s d + co Th d +1 m d la rv ae St om at op od a in m d +1 X d M ys id ac ea r^ r^ d +1 co d + + L O CO d +1 ??. + 0 + + + + C um ac ea A m ph ip od a _ ?1 X +1 CO X co Ol X +1 in oi 01 r- d +1 Th rt in Th X +1 CO CO Ol + co OJ d +1 X d + OJ OJ d +1 in d V Sh ri m p* la rv a 0 Ol d +1 co d X m d +1 X d + + pi A no m ur a la rv ,_, OJ d +1 Th X as OS d +1 OS Th X d +1 OJ ?' 0 00 Th +i OJ oi OJ r-? r-? d +1 p ? c CO CO d +1 X d OJ Oa d +1 CO X + oS s> B ra ch yu ra la r CO d +1 CO X CO CO d +1 CO X CO d +1 X d m 0 d +1 r^ d CO CM d +1 X d + CO CO d +1 CO d + a E ch in od er m at la rv ae OJ X d +1 -1- 00 0 + + CO X +1 ?' co Th 0 +1 "1 + 0 0 OJ +1 OJ + + V . la rv a a rv a e H em ic ho rd at a U ro ch or da ta 1 Oa Th 0 x CM x ? to x y-; CM m CM d d co ? ? d +1 +1 +1 +1 +1 +1 to i~~ ?>? x ? r^ d d d co co d m 00 00 ? in 0 in co QJ t>- CM 0 CM 0 r^ d +1 +1 +1 +1 +1 -j. in C M in 00 in d os CM os d CO Th ? in x CO ^ P -H (N - +1 +1 +1 -|- _|- Th r-; CM CO t-^ CO CM 0 0 in CM in Th co d co 0 (o ? d +1 +1 +1 +1 + + r-- r-. p ?; r^ os co ? ? OJ Th CO 00 CM ^ X X CM ?< ? +1 +1 +1 -(--(- ? CO CO r^ co CM Tf Tf ? CO ? O OS * CO 1 - O CO ? CO +1 +1 +1 +1 X -)- Tf CO X -}- d 06 Th oi Th CO x os Th ? < ^ ri <=? <=>. O CM ? ? +1 , +1 +1 +1 , x + CM in p + d x CM oi in X co Th 0 os "* Th ^ ^ O CM O ? +1 +1 +1 +1 m CM oa co -|- d d ?< Tf m Sa lp a D ol io lid a L ar va ce a C ha et og na th a Fi sh o v a Fi sh la rv ae 148 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES mean abundance of organisms in different habi? tats and light periods (Table 14). In order to eliminate the influence of the dominant copepods and nauplii, these groups were tested as a separate category. Abundance of organisms in ocean sam? ples is greater than in lagoon samples with respect to all three counts. Night tows (0400 h and 2000 h) yield a larger combined number of individuals than daytime samples (0800 h and 1600 h), as would be expected from the bimodal pattern of plankton emergence (near dawn and dusk) indi? cated by Glynn (1973). This difference does not hold, however, for organisms other than copepods and nauplii which show no significant quantita? tive difference between night and day. On the other hand, surface tows (0-0.5 m) differ signifi? cantly from 2.5-3.0 m tows only with respect to plankton other than copepods and nauplii. These differences are not entirely reflected by settled volume or ash-free dry weights (Table 15). When mean values for settled volume generated by paired groups of samples are compared by the Wilcoxon signed-ranks test, significantly more plankton volume/m3 is found in ocean samples than lagoon samples and, contrary to individual numbers, in day tows than in night tows {P < 0.05). The latter result is probably due to the presence of significantly more siphonophores in day samples (see below). Depth of tow did not have bearing on volume of plankton. Ash-free dry weight, which is a more reliable estimate of bio? mass than settled volume, does not differ signifi? cantly when locations or times of day are com? pared (Table 17) using means generated by paired groups of samples (Table 15). Significantly more plankton biomass (mg/m3), however, is found at 0-0.5 m than in 2.5-3.0 m, regardless of location or time of day. The lack of agreement between these measurements of numbers of indi? viduals and biomass probably reflects differences in the composition of the samples (Tables 13, 16) as well as patchiness or unevenness in zooplank? ton distribution. Patchiness is indicated by the large standard errors of means calculated for all counts and biomass measurements in the present study (Tables 15, 16). It should be noted, how? ever, that mean ash-free dry weight (mg/m ) of ocean samples is greater than that of lagoon samples, that of night samples exceeds that of day samples, and that of 0-0.5 m samples is greater than that of 2.5-3.0 m samples (Table 17). The dominant organisms other than copepods and nauplii represented at both locations indicate a primarily oceanic plankton: larval polychaetes, gastropods, cirripedes, and shrimps (Euphausi- acea, Penaeidea, Caridea) as well as pteropods, larvaceans, chaetognaths, and fish ova (Tables 13, 16). Of these groups, polychaete larvae, gas? tropod larvae, larvaceans, and chaetognaths are significantly more numerous in ocean than in lagoon samples {P < 0.05). Similarly, larval cni- darians, platyhelminthes, bryozoans, pelecypods, and echinoderms, as well as siphonophores, ostra- cods, copepods, nauplii, salps, and doliolids are more numerous in ocean samples. Also, although the blue-green alga Oscillatoria {= Trichodesmium) sp. was not counted, it was observed to be a major component of the ocean but not the lagoon plank? ton. These results may indicate a net import of organisms to the reef because oceanic plankton is removed from the water column by the reef com? munity. Similar findings are reported by Tranter and George (1972) at Kavaratti and Kalpeni atolls in the Laccadives, by Glynn (1973) at Puerto Rico, and by Johannes and Gerber (1974) at Eniwetok Atoll. In contrast, Motoda (1940) and Johnson (1949, 1954) found zooplankton to be less abundant in ocean waters than in lagoon waters. However, as pointed out by several au- TABLE 17.?Carrie Bow Cay plankton biomass, expressed as ash-free dry weights (mg/m3 ? standard error) for the six major sample groups, with results of Wilcoxon signed-ranks test for paired groups (+ = significant at P < 0.05; 0 = no significant difference) Sample group Ocean Lagoon Night Day 0-0.5 m 2.5-3.0 m mg/m3 2.35?0.41 1.85?0.30 2.48?0.41 1.71?0.25 2.44?0.41 1.75?0.28 Wilcoxon test 0 0 + NUMBER 12 149 thors (Alldredge and King, 1977; Porter and Porter, 1977; Porter et al., 1978; Riitzler et al., 1980), plankton samples from the surface waters associated with coral reefs do not necessarily re? flect the amount of plankton actually available to reef organisms. In the present study, certain groups of organisms (mysids, cumaceans, brach- yuran larvae) are present in significantly {P < 0.05) greater numbers per m3 in the lagoon. Also, the percentage of the plankton represented by these groups, together with adult polychaetes and shrimp larvae, is greater in the lagoon than out? side the reef (based on data in Table 13, with the addition of lagoon versus ocean sample percent? ages for adult polychaetes, 0.2:0.04; mysids, 0.08: 0.004; and cumaceans, 0.4:0.005). Similarly, the foraminiferan Rosalina (= Tretomphalus) sp. was recorded as abundant in lagoon but not in ocean samples. These results may indicate a significant difference in the composition of the plankton between the two localities. Hence, this difference may also reflect the presence of a resident reef fauna in the lagoon, represented by organisms that are at times considered demersal, such as adult polychaetes, mysids, cumaceans, shrimp larvae, and brachyuran larvae (Tranter and George, 1972; Sale et al., 1976; Alldredge and King, 1977; Porter and Porter, 1977). Such a resident reef fauna would be sampled only par? tially by the methods used in the present study and more likely would appear in samples from the lagoon, where water depth is only 5-8 m and small patch reefs are common. Similarly, differ? ences in composition but not necessarily in bio? mass of lagoon and offshore plankton were found by Johnson (1954), Odum and Odum (1955), Bakus (1964), Tranter and George (1972), and Sale et al. (1976). This difference would indicate removal of oceanic plankton by reef organisms and replacement by a resident lagoon plankton (Emery, 1968). Tows taken at 0-0.5 m and 2.5-3.0 m appear to have similar composition. Mysids, cumaceans, and fish ova were more abundant {P < 0.05) at 0-0.5 m whereas pelecypod and anomuran larvae were more common in 2.5-3.0 m tows. Comparing day and night tows, significantly more cnidarian larvae, siphonophores, platyhel? minth larvae, sipunculid larvae and cladocerans appear in daytime (0800 h and 1600 h) tows. Crustacean groups (ostracods, mysids, cuma? ceans, amphipods, shrimp larvae, and anomuran larvae) as well as doliolids, however, are more abundant at night (0400 h and 2000 h). Results of this study agree, in general, with Emery (1968), Johannes et al. (1970), Glynn (1973), Porter (1974), and Renon (1977), who have noted both quantitative and qualitative differences between day and night samples of zooplankton over a variety of coral reefs. More specifically, Alldredge and King (1977) also report more ostracods, my? sids, cumaceans, amphipods, shrimps, and deca? pod larvae in night tows than in daylight tows at Lizard Island on the Great Barrier Reef. Conclusions Copepods and nauplii (mostly copepod) com? bined account for 84.5% and 82.4% of the zoo? plankton in lagoon and ocean samples, respec? tively, from Carrie Bow Cay. Other major groups of organisms represented at both locations (larval polychaetes, gastropods, cirripedes, and shrimp, as well as pteropods, larvaceans, chaetognaths, and fish ova) indicate a primarily oceanic plank? ton. The relative abundance of mysids, cumaceans, adult polychaetes, and brachyuran and shrimp larvae in the lagoon may indicate partial sam? pling of a resident reef fauna in this habitat. Samples from the open ocean contain a signifi? cantly greater number of organisms (copepods and nauplii, as well as all others) per m water than those from the lagoon, and therefore suggest a net import of zooplankton to the reef. Plankton biomass, expressed by ash-free dry weight was greater in ocean, night, and 0-0.5 m samples than in lagoon, day, and 2.5-3.0 m tows. Resident reef fauna will have to be sampled directly before final conclusions can be drawn about the impact of zooplankton on the energy budget of the coral reef community. 150 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Literature Cited Alldredge, A. L., and J. M. King 1977. Distribution, Abundance, Substrate Preferences of Demersal Reef Zooplankton at Lizard Island La? goon, Great Barrier Reef. Marine Biology, 41:317- 333. Bakus, G. J. 1964. The Effects of Fish Grazing on Invertebrate Evo? lution in Shallow Tropical Waters. Allan Hancock Foundation Publications Occasional Paper, 27:1-29. Davis, W. P., and R. S. Birdsong 1973. Coral Reef Fishes Which Forage in the Water Column. Helgb'lander Wissenschaftliche Meeresunter- suchungen, 24:292-306. Emery, A. R. 1968. Preliminary Observations on Coral Reef Plankton. Limnology and Oceanography, 13:293-303. Gerber, R., and N. Marshall 1974. Reef Pseudoplankton in Lagoon Trophic Systems. In A. M. Cameron, B. M. Campbell, A. B. Cribb, R. Endean, J. S. Jell, O. A. Jones, P. Mather, and F. H. Talbot, editors, Proceedings of the Second Inter? national Coral Reef Symposium, 1:105-107. Brisbane, Australia: The Great Barrier Reef Committee. Glynn, P. W. 1973. Ecology of a Caribbean Coral Reef. The Porites Reef-Flat Biotope, Part II: Plankton Community with Evidence for Depletion. Marine Biology, 22 :1- 21. Goreau, T. F., N. I. Goreau, and C. M. Yonge 1971. Reef Corals: Autotrophs or Heterotrophs? Biologi? cal Bulletin, 141:247-260. Hobson, E. S. 1974. Feeding Relationships of Teleostean Fishes on Coral Reefs in Kona, Hawaii. Fishery Bulletin, 72: 915-1031. Johannes, R. E. 1974. Sources of Nutritional Energy for Corals. In A. M. Cameron, B. M. Campbell, A. B. Cribb, R. En- dean, J. S. Jell, O. A. Jones, P. Mather, and F. H. Talbot, editors, Proceedings of the Second International Coral Reef Symposium, 1:133-137. Brisbane, Aus? tralia: The Great Barrier Reef Committee. Johannes, R. E., S. L. Coles, and N. T. Kuenzel 1970. The Role of Zooplankton in the Nutrition of Some Scleractinian Corals. Limnology and Oceanography, 15:579-586. Johannes, R. E., and R. Gerber 1974. Import and Export of Net Plankton by an Eni- wetok Coral Reef Community. In A. M. Cameron, B. M. Campbell, A. B. Cribb, R. Endean, J. S. Jell, O. A. Jones, P. Mather, and F. H. Talbot, editors, Proceedings of the Second International Coral Reef Symposium, 1:97-104. Brisbane, Australia: The Great Barrier Reef Committee. Johnson, M. W. 1949. Zooplankton as an Index of Water Exchange be? tween Bikini Lagoon and the Open Sea. Transac? tions of the American Geophysical Union, 30:238-244. 1954. Plankton of Northern Marshall Islands, Bikini and Nearby Atolls. United States Geological Survey Profes? sional Papers, 260-F:301-304. Lewis, J. B. 1977. Processes of Organic Production on Coral Reefs. Biological Reviews, 52:305-347. Moore, E., and F. Sander 1976. Quantitative and Qualitative Aspects of the Zoo? plankton and Breeding Patterns of Copepods at Two Caribbean Coral Reef Stations. Estuarine and Coastal Marine Science, 4:589-607. 1977. A Study of the Offshore Zooplankton of the Trop? ical Western Atlantic Near Barbados. Ophelia, 16: 77-96. Motoda, S. 1940. Comparison of the Conditions of Water in Bay, Lagoon and Open Sea in Palao. Palao Tropical Biological Station Studies, 2:41-48. Odum, H. T., and E. P. Odum 1955. Trophic Structure and Productivity of a Wind? ward Coral Reef Community on Eniwetok Atoll. Ecological Monographs, 25:291-320. Porter, J. W. 1974. Zooplankton Feeding by the Caribbean Reef- building Coral Montastrea cavernosa. In A. M. Cam? eron, B. M. Campbell, A. B. Cribb, R. Endean, J. S. Jell, O A. Jones, P. Mather and F. H. Talbot, editors, Proceedings of the Second International Coral Reef Symposium, 1:111-126. Brisbane, Australia: The Great Barrier Reef Committee. Porter, J . W., and K. G. Porter 1977. Quantitative Sampling of Demersal Plankton Mi? grating from Different Coral Reef Substrates. Lim? nology and Oceanography, 22:553-556. Porter, J. W., K. G. Porter, and Z. Batac-Catalan 1977. Quantitative Sampling of Indo-Pacific Demersal Reef Plankton. In D. L. Taylor, editor, Proceedings of the Third International Coral Reef Symposium, 1:105- 112. Miami, Florida: Rosenstiel School of Marine and Atmospheric Science. Porter, K. G , J. W. Porter, and S. L. Ohlhorst 1978. Resident Reef Plankton. In D. R. Stoddart and R. E. Johannes, editors, Coral Reefs: Research Methods, Monographs on Oceanographic Methodology, 5:499-514. Paris: Unesco. NUMBER 12 151 Qasim, S. Z., and V. N. Sankaranarayanan 1970. Production of Particulate Organic Matter by the Reef on Kavaratti Atoll (Laccadives). Limnology and Oceanography, 15:574-578. Renon, J . P. 1977. Zooplankton du lagon de I'atoll de Takapoto (Po- lynesie Francaise). Annates de I'Instilul de la Oceano- graphie (Paris), 53:217-236. Riitzler, K., J . D. Ferraris, and R. J. Larson 1980. A New Plankton Sampler for Coral Reefs. Marine Ecology, 1:65-71. Sale, P. F., P. S. McWilliam, and D. T. Anderson 1976. Composition of the Near-Reef Zooplankton at Heron Reef, Great Barrier Reef. Marine Biology, 34:59-66. Sargent, M. C , and T. S. Austin 1954. Biologic Economy of Coral Reefs. United States Geological Survey Professional Papers, 260-E:293-300. Sokal, R. R., and F. J. Rohlf 1969. Biometry. 776 pages. San Francisco: W. H. Freeman and Co. Stevenson, R. A., Jr. 1972. Regulation of Feeding Behavior of the Bicolor Damsel-Fish (Eupomacentrus parlitus Poey) by En? vironmental Factors. In H. E. Winn and B. L. Olla, editors, Behavior of Marine Animals, 2:278-302. New York: Plenum Press. Tranter, D. J., and J. George 1972. Zooplankton Abundance at Kavaratti and Kal- peni Atolls in the Laccadives. In C. Mukundan and C. S. Gopinadha Pillai, editors, Proceedings of the Symposium on Corals and Coral Reefs, pages 239- 256. Cochin: Marine Biological Association of India. Yonge, C. M. 1930. Studies of the Physiology of Corals, I: Feeding Mechanisms and Food. Scientific Reports of the Great Barrier Reef Expedition, 1:13-57. Plankton Diatoms (Bacillariophyceae) from Carrie Bow Cay, Belize Paul E. Hargraves ABSTRACT Seventy-three diatom taxa from the coastal waters around Carrie Bow Cay, Belize (Central America) were found in samples collected April/ May 1977 and January 1978. Chaetoceros, Rhizoso- lenia, and Coscinodiscus were the most common genera. Six species are new records for the Car? ibbean region. Distinct differences exist between spring and winter diatom flora. Although numer? ous spore-forming species were found, resting spores were absent. Cultivation of more tropical diatoms is seen as an aid in solving problems in the systematics of tropical phytoplankton. Introduction The species composition of plankton diatoms along the Carribean coast, unlike that of coastal areas in temperate regions, is poorly known. For oceanic areas of the Caribbean more information is available because a considerable number of oceanographic cruises have included sampling for phytoplankton (for instance, Hargraves et al., 1970; Hulbert, 1968; Marshall, 1973; Takano, 1960). Coastal floristic works are few and widely scattered geographically (Buchanan, 1971; Mar? galef, 1957, 1968; Sander, 1976). Apparently no publications deal with coastal plankton diatoms of Belize, although planktonic coccolithophorids have been examined (Kling, 1975). In terms of turnover rate and primary produc? tion in tropical waters, the nanoplankton (larger Paul E. Hargraves, Graduate School of Oceanography, University of Rhode Island, Kingston, R. I. 02881. than 1-10 jiim but smaller than 50-70 jum, de? pending on authors' definition) is often more important than the net plankton (Malone, 1971; Texeira, 1963); nanoplankton may occasionally dominate the biomass as well (Garrison, 1975). Owing to my limited sampling, the annotated list below does not completely represent plankton diatoms of Carrie Bow Cay. Also, primarily benthic species are excluded from this report, except for the few whose consistent presence in? dicates a pelagic as well as benthic existence. Generally, diatom species have been defined on the basis of morphological characteristics without a proven genetic basis. Variability within a spe? cies is often considerable, not only morphologi? cally (for example, Van Landingham, 1967, lists over 100 synonyms for Actinocyclus ehrenbergii that were formerly considered species, differentiated on morphological bases), but physiologically as well (for example, Carpenter and Guillard, 1971; Hargraves and Guillard, 1974). Although im? practical for routine purposes, genetic analysis by means of enzyme electrophoresis (Gallagher, 1979; Murphy and Guillard, 1976) shows consid? erable promise for separating diatom species. Nevertheless, morphological features are the main criteria for current classification schemes. The grouping of species into genera, of genera into families, and of families into orders is in a state of flux, primarily because of recent discov? eries about fine structure of the siliceous diatom frustule (see reviews in Werner, 1977). Organi? zation of these hierarchies is subjective, depend? ing on individual assessment of the relative im? portance of given structures, and diatomists have 153 154 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES probably not yet attained a wholly natural clas? sification system. For this reason, the list below is arranged in alphabetical order rather than in a particular phylogenetic scheme. Some recent at? tempts at an incisive phylogenetic arrangement have been reviewed by Hendey (1974) and Si- monsen (1974). ACKNOWLEDGMENTS.?The assistance of P. Boyd, D. Hargraves, and K. Zimmerly is grate? fully acknowledged. Partial support came from the National Science Foundation, Grant No. OCE 76-82280. Methods Samples were collected with 30 cm diameter plankton nets, having mesh sizes of 65 jum or 10 ju,m and were preserved in hexamine-buffered formalin or Lugol's iodine, without acetic acid. Aliquots were washed with distilled water to re? move salts. To remove organic matter, cells were oxidized by either boiling for one hour in 30% hydrogen peroxide, or by ashing samples for three hours on cover slips on a hot plate. Permanent slides of the oxidized material were prepared using Hyrax (Custom Research & Development, Inc., Auburn, California) as a mounting medium. These permanent mounts and the liquid-pre? served material were examined in a Zeiss Photo- microscope II using brightfield, phase contrast, and Nomarski interference contrast optics. When necessary, identifications were confirmed using a Cambridge Stereoscan electron microscope. Slides are maintained in the author's collection. Collection sites were in the vicinity of Carrie Bow Cay, in the central portion of the Belize barrier reef. Surface tows were made over Thal? assia, near the mangrove forests of Twin Cays, in the channel between Carrie Bow Cay and South Water Cay, in deeper water (over 200 m) about 0.5 km east of Carrie Bow Cay, and in the "South? ern Shelf of the lagoon, at approximately 16?55'N, 88?12'W in over 20 m depth. The sam? ples were collected in April and May 1977 and January 1978 at the same localities, and thus they cover the late spring and mid-winter periods. Annotated List of Diatoms (Species new to the Caribbean flora are designated by *) Actinocyclus ehrenbergii Ralfs in Pritchard (Figure 87a) [= A. octonarius Ehrenberg] Found only in lagoon samples in spring; never common. Previously reported from the Car? ibbean by Hagelstein (1938) and Margalef (1968). Characteristic of both benthic and pelagic environments. Reference: Hendey (1964). Actinocyclus subtilis (Gregory) Ralfs in Pritchard Found only in lagoon samples in spring; never common. Previously reported from the Car? ibbean by Hagelstein (1938) and Margalef (1968). Characteristic of both benthic and pelagic environments. Reference: Hendey (1964). Actinoptychus senarius (Ehrenberg) Ehrenberg [= A. undulatus (Bailey) Ralfs in Pritchard] Found only rarely in samples from the lagoon side of Carrie Bow Cay in winter. A cosmo? politan species, it is widespread in temperate coastal waters in all seasons, but apparently rarer in the tropics. Reported from the Car? ibbean by Hagelstein (1938) and Margalef (1968) among others. Reference: Hendey (1964). Asterionella notata (Grunow) Grunow in Van Heurck (Figure 8bd) Uncommon to common in all samples; more abundant on the ocean side of Carrie Bow Cay. A tropical coastal species straying rarely into temperate waters. Widespread in the Car? ibbean, but apparently misidentified occa? sionally as A. kariana, an Arctic species. Ref? erence: Hustedt (1931-1959). * Asteromphalus hookeri Ehrenberg (Figure 85c) Present rarely and only in spring from the ocean side. Normally oceanic in cooler waters. Not previously reported from the Caribbean. Reference: Lebour (1930). Bacteriastrum comosum Pavillard Occasionally found in winter only in lagoon NUMBER 12 155 and ocean-side samples. A predominantly tropical species, previously reported from the Caribbean by Margalef (1968). Reference: Hustedt (1927-1930). Bacteriastrum delicatulum Cleve Found in lagoon and ocean-side samples in spring and winter. An oceanic species, reach? ing its maximum abundance in temperate waters. Widespread and occasionally abun? dant throughout the Caribbean. Reference: Hendey (1964). Bacteriastrum elongatum Cleve (Figure 86b) Found rarely in ocean-side tows and only in spring. Commonly found in tropical waters, and widely reported from the Caribbean. Ref? erence: Hendey (1964). Bacteriastrum cf. furcatum Shadbolt (Figure 86a) This cell was seen in one sample from the ocean side in spring. It resembles B. furcatum Shadbolt, sensu Boalch (1975), but insuffi? cient material was found to make an identifi? cation. Bacteriastrum hyalinum Lauder (Figure 85a) Occasionally found in winter samples, from both lagoon and ocean side. Common in tem? perate and tropical coastal waters. Previously reported from the Caribbean by Hargraves et al. (1970) and Margalef (1968). Reference: Hendey (1964). Bacteriastrum varians Lauder Found once in a spring sample from the la? goon side of Carrie Bow Cay. Generally con? sidered a tropical oceanic species, it was re? ported in the Caribbean by Takano (1960). Boalch (1975) considers B. varians a synonym of B. furcatum Shadbolt. Reference: Boalch (1975). Cerataulina pelagica (Cleve) Hendey [= Cerataulina bergonii (Peragallo) Schutt] Present in small numbers in winter samples. Widely distributed in coastal temperate and tropical regions. Found throughout the Car? ibbean region. Reference: Hendey (1964). Chaetoceros affine Lauder Occasionally present in ocean-side samples in winter and spring. Widely distributed in tem? perate and tropical coastal waters. Commonly reported from the Caribbean. The morphol? ogy of this species is highly variable. Some chains appear transitional to C. diversum and C. laciniosum. Single cells are also seen rarely. Reference: Hustedt (1927-1930). Chaetoceros atlanticum Cleve Very rare in ocean samples in both winter and spring. A widespread oceanic species in tem? perate and tropical waters. Reported from the Caribbean by Margalef (1968), Roukiyainen et al. (1973), and Sander (1976). The variety neapolitana seems more common to tropical waters. The morphology of some chains re? sembles C. laciniosum, from which it can be distinguished by the multiple chloroplasts and central tubule of the valve. Reference: Hus? tedt (1927-1930). * Chaetoceros boreale Bailey Rarely found in a spring sample on the ocean side of Carrie Bow Cay. This rather distinctive species is normally found in temperate or cold waters. It has not been reported previously from the Caribbean. Reference: Hustedt (1927-1930). Chaetoceros coarctatum Lauder A rarely found species from ocean-side sam? ples in winter and spring. Generally found in the tropics, occasionally in temperate waters. Previously reported from the Caribbean re? gion by Hargraves et al. (1970), Margalef (1968), and Takano (1960). Rarely vorticellid protozoans are attached to several cells in a chain. Reference: Hustedt (1927-1930). Chaetoceros compressum Lauder Only seen once in a winter sample. This spe? cies is normally considered a coastal species from temperate or boreal waters. Nevertheless, it is widely reported from the Caribbean, sometimes as a dominant species (Sander, 1976). Reference: Hustedt (1927-1930). Chaetoceros curvisetum Cleve Occasionally seen in spring, but only in sam? ples from the ocean side of Carrie Bow Cay. Widely distributed in coastal waters in tem? perate and tropical regions. Reported by most 156 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 85.?Light photomicrographs of representative diatoms: a, Bacteriastrum hyalinum, valve view of cell with 17 setae (Hyrax mount); b, Chaetoceros peruvianum, girdle view of entire cell (only bases of setae are shown; Hyrax mount); c, Asteromphalus hookeri, valve view of cell (Hyrax mount); d, Astenonella notata, isolated cell in valve view (Hyrax mount); e, Coscinodiscusjonesianus, valve view (Hyrax mount) ; / , Thalassionema nitzschioides, eight-celled colony in girdle view (living material). (Scale = 10 /xm.) NUMBER 12 157 FIGURE 86.?Scanning electron photomicrographs of representative diatoms: a, Bacteriastrum sp., probably B. furcatum (sensu Boalch, 1975), valve view of terminal cell; b, B. elongatum, terminal and interior cells of a chain; c, Chaetoceros lorenzianum, girdle view; d, Hemialus hauckii, girdle view of an isolated valve. (Scale = 10 jum.). 158 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES NUMBER 12 159 authors throughout the Caribbean. Refer? ence: Hustedt (1927-1930). Chaetoceros danicum Cleve Occasionally found in lagoon samples in win? ter. Normally characterized as a temperate species favoring somewhat reduced salinity. Previously reported from the Caribbean by Margalef (1968) and Sander (1976). Refer? ence: Hustedt (1927-1930). Chaetoceros decipiens Cleve Rarely observed in winter samples from the lagoon and ocean side of Carrie Bow Cay. Nominally an oceanic species from temperate and boreal waters, it has been previously re? ported from the Caribbean by most authors. The morphology of the species is variable; the distinctive features distinguishing it from C. lorenzianum are the nature of fusion of the setae and the presence of resting spores in C. loren? zianum. Fusion of the setae is a somewhat unreliable character in nature, and spores are rarely seen; it is possible that misidentifica- tions have occurred, particularly in the case of narrow cells and short chains. Minor differ? ences between the two are apparent at the ultrastructural level (Evenson and Hasle, 1976). Reference: Cupp (1943). Chaetoceros didymum Ehrenbert Fairly common in ocean-side samples in both spring and winter. Widely distributed in tem? perate and warm waters; also widely distrib? uted in the Caribbean. Reference: Hustedt (1927-1930). Chaetoceros diversum Cleve Found in winter and spring in most samples. Considered to be a tropical coastal species, it is occasionally reported also in temperate wa? ters. Previously reported from the Caribbean by Sander (1976) and Takano (1960). This species is easily confused with C. laeve Leu- FIGURE 87.?Scanning electron photomicrographs of repre? sentative diatoms: a, Actinocyclus ehrenbergii, interior side of a small specimen; b, Trigonium formosum, external side of a five- sided form; C, Rhizosolenia calcaravis, side view of the valve; d, Triceratium pentacrinus, external side; e, Isthmia enervis, girdle view of the entire cell. (Scale = 10 ftm.) duger-Fort morel and perhaps the two should be synonymous. Sander and Takano include both these species in their lists. Short chains may resemble C. affine. Reference: Cupp (1943). Chaetoceros laciniosum Schutt [= C. distans Cleve] Occasionally present in ocean-side samples in spring. This species is common and widely distributed in temperate waters but also oc? curs in the tropics. Previously reported from the Caribbean by Hulburt (1968) and Sander (1976), the latter as "dominant." Chaetoceros laciniosum is another highly variable species, and in its growth forms can resemble C. pelag- icum Cleve and C. breve Schutt. Superficially some chains also resemble C affine and narrow forms of C. atlanticum. Reference: Hustedt (1927-1930). Chaetoceros lorenzianum Grunow (Figure 86c) Found in most samples in winter and spring. A common coastal form from temperate wa? ters, it also is widely distributed in the tropics. It is widely reported, sometimes in abundance, from the Caribbean. Narrow chains particu? larly can be confused with C. decipiens (see comment for that species). Reference: Hustedt (1927-1930). Chaetoceros pendulum Karsten Seen once in a spring tow from the ocean side. A fairly rare oceanic species characteristic of tropical waters. Reports from the Caribbean by Sander (1976) and Takano (1960). Refer? ence: Cupp (1943). Chaetoceros peruvianum Brightwell (Figure 85b) Found in a spring tow from the ocean side. A very common and widely distributed species in temperate and tropical waters, although rarely abundant. Present in both coastal and oceanic areas. Widely reported from the Car? ibbean (for instance, Sanders, 1976; Takano, 1960), as short chains and single cells. Refer? ence: Cupp (1943). Chaetoceros pseudocurvisetum Mangin Occasionally in winter samples. A coastal spe? cies widely distributed in temperate and trop- 160 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES ical areas, previously reported from the Car? ibbean by Hagelstein (1938), Margalef (1968), and T a k a n o (1960). Reference: C u p p (1943). Chaetoceros tortissimum Gran Seen only once in a winter tow on the ocean side of Carr ie Bow Cay. This is a characteris? tically t empera te coastal species only occa? sionally seen and rarely of any abundance . It has been reported from the Car ibbean by Margalef (1968) and T a k a n o (1960). Refer? ence: Hendey (1964). Climacodium biconcavum Cleve Rare ; observed in only one tow from the ocean side in spring. A widely dis tr ibuted tropical oceanic species. Previously reported from the Ca r ibbean by Sander (1976) a n d T a k a n o (1960). Reference: Hus ted t (1927-1930). Climacodium frauenfeldianum Grunow Fairly common in the spring in bo th lagoon and ocean-side samples. Like C biconcavum, this species is found only rarely outside the tropics, bu t it is apparen t ly more common that C. biconcavum. T h e shape of the intercel? lular aper tu re is apparent ly the only distin? guishing characterist ic between the two spe? cies: elliptical or quad ra t e in C frauenfeldianum, lanceolate or d iamond-shaped in C biconcavum. Sander (1976) and T a k a n o (1960) have re? por ted it in Car ibbean waters. Reference: C u p p (1943). Coscinodiscus asteromphalus Ehrenberg Occasionally seen in ocean tows in the spring and lagoon samples in the winter. This is one of the most common Coscinodiscus species in t empera te coastal waters. It is morphologi? cally variable and may be confused with other species, and has rarely been reported from the tropics. T h e only previous Car ibbean record is Hargraves et al. (1970). Reference: Hus ted t (1927-1930). Coscinodiscus centralis Ehrenberg Present in one sample in the spring from the ocean side. More common in oceanic waters, this species is also found in coastal t empera te areas. Repor ted from the Ca r ibbean by Bu- canan (1971), Hargraves et al. (1970), Mar? shall (1973), and T a k a n o (1960). Reference: Hus ted t (1927-1930). Coscinodiscus concinniformis Simonsen Occasionally seen in ocean-side samples in bo th spring and winter. This species was es? tablished by Simonsen (1974) and includes in par t C. concinnus in the sense of Hus ted t (1927- 1930). T h e relat ionship of C. concinniformis to C concinnus, C. concinnoides, C granii, C. nobilis and C. wailesii is discussed by Simonsen (1974: 14-16). At least some of the tropical records of C. concinnus are probably C. concinniformis, since the latter appears to be a tropical species, whereas the former is found in "higher lati? tudes ." Assuming this to be t rue, C. concinni? formis is widespread in the Car ibbean . Refer? ence: C u p p (1943); Simonsen (1974). Coscinodiscus jonesianus (Greville) Ostenfeld (Fig? ure 85 ; ^ v , ^ * 4 5Tv'' i^i' / J S o t s O D o " - ' - o o ? 0 ? & O j . O O <>Q - Q _ " ? "/JO 50jum FIGURE 112.?Polysiphonia sphaerocarpa: a, tetrasporangia in spiral series (DK-s.n., 21 Mar 1978); b, spermatangial branches (DK-s.n., 21 Mar 1978); c, mature pericarp (JN-6911); d, rhizoids cut off from pericentral cells (DK-s.n., 21 Mar 1978). 236 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Conclusions Our collections from Carrie Bow Cay add seven taxa of Polysiphonia to the previously reported marine flora of Belize. Five species belong to the subgenus Oligosiphonia?P. atlantica, P. ferulacea, P. flaccidissima, P. scopulorum var. villum, and P. sphaer? ocarpa?and two to the subgenus Polysiphonia?P. denudata and P. exilis. Polysiphonia exillis and P. flaccidissima are found in the Caribbean Sea for the first time. Prior to this study only P. havanensis and P. scopulorum were known from Belize, neither of which appeared in our collections. Although we cannot exclude the possibility that P. scopulo? rum reported by Tsuda and Dawes (1974) from Glover's reef, only 25 km east of Carrie Bow Cay, may belong to var. villum. Polysiphonia atlantica is a new name proposed for P. macrocarpa Harvey, a later homonym for P. macrocarpa (C. Agardh) Sprengel. The identity of C. Agardh's Hutchinsia macrocarpa, described from the Antilles but never reported since, remains unknown. Although members of the genus Polysiphonia occur predominantly in tropical waters, P. ferula? cea ranges into the subtropical Atlantic and P. atlantica and P. denudata extend even into the temperate Atlantic. The latter two species, at least, are remarkable for their physiological tol? erance of such a large range of environmental conditions. All our collections were made close to Carrie Bow Cay, during March and April. We expect that material sampled during other seasons and at different localities along the barrier reef, on the atolls, and along the mainland coast will contain interesting new finds and will help solve taxo? nomic problems. Literature Cited Abbott, I. A., and G. J. Hollenberg 1976. Marine Algae of California, xii + (2) + 827 pages. Stanford, California: Stanford University Press. Agardh, C. A. 1824. Systema algarum. xxxvii + 312 pages. Lund: Ber- ling. Agardh, J . G. 1863. Species, genera el ordines algarum, seu descriptiones suc- cinctae specierum generum et ordinum, quibus algarum regnum constituitur. Volume 2, part 3, pages [2] + 787-1291. Lund G.W.K. Gleerup. Ardre, F. 1970. Contribution a l'etude des algues marines du Por? tugal, I: La Flore. Portugaliae Acta Biologica (B), 10(1/4):5-423, 56 plates. B0rgesen, F. 1918. The Marine Algae of the Danish West Indies, Part IV [sic]: Rhodophyceae (4). Danske Botanisk Arkiv, 3(ld):241-304. Brauner, J. 1975. Seasonality of Epiphytic Algae on Zostera marina at Beaufort, North Carolina. Nova Hedwigia, 26:125- 133. Collins, F. S., and A. B. Hervey 1917. The Algae of Bermuda. Proceedings of the American Academy of Arts and Sciences, 53(1): 1-195. Dillwyn, L. W. [1809.] British Confervae 87 pages. London: W. Phil? lips. Dixon, P. S. 1960. Taxonomic and Nomenclatural Notes on the Flor- ideae, II. Botamska Notiser, 113(3):295-319. Greville, R. K. 1824. Flora Edinensis. lxxxi -I- 478 pages. Edinburgh: W. Blackwood. Harvey, W. H. 1853. Nereis Boreali-Americana . . . Part II: Rhodospermeae. [First issue, ii +] 258 pages, plates 13-36. Wash? ington, D . C : Smithsonian Institution [London: John Van Voorst]. [Third issue, 1853 [1858], Smithsonian Contributions to Knowledge, 5(5): ii + 258 pages, plates 13-36.] Hollenberg, G. J . 1942. An Account of the Species oi Polysiphonia on the Pacific Coast of North America, I: Oligosiphonia. American Journal of Botany, 29:772-785. 1944. An Account of the Species of Polysiphonia on the Pacific Coast of North America, II: Polysiphonia. American Journal of Botany, 31:474-483. 1961. Marine Red Algae of Pacific Mexico, Part 5: The NUMBER 12 237 Genus Polysiphonia. Pacific Naturalist, 2(6):345-375. 1968a. An Account of the Species of Polysiphonia of the Central and Western Tropical Pacific Ocean, I: Oligosiphonia. Pacific Science, 22(l):56-98. 1968b. An Account of the Species of the Red Alga Polysi? phonia of the Central and Western Tropical Pacific Ocean, II: Polysiphonia. Pacific Science, 22(2): 198- 207. Hollenberg, G. J., and J. N. Norris 1977. The Red Alga Polysiphonia (Rhodomelaceae) in the Northern Gulf of California. Smithsonian Contribu? tions to the Marine Sciences, 1: iii + 21 pages. Hooker, W . J . 1833. Mosses, Heptiae, Lichens, Characeae and Algae. In J . E. Smith, The English Flora, edition 2, 5(1): x + 4-1- 432 pages. London: Longman, Rees, Orme, Brown, Green & Longman. [Also issued in volume 2 of W. J. Hooker, The British Flora.] Howe, M. A. 1920. Class 2, Algae In N. L. Britton and C. F. Mills- paugh, The Bahama Flora, pages 553-618. New York: Published by the authors. Joly, A. B. 1965. Flora marinha do litoral norte do Estado de Sao Paulo e regioes circunvizinhas. Boletim Faculdade de Filosofia, Ciencias e Letras da Universidade de Sao Paulo, Boletim (Botanica), 21:1-393, plates 1-59. Kapraun, D. 1977. The Genus Polysiphonia in North Carolina, U.S.A. Botanica Marina, 20:313-331. 1979. The Genus Polysiphonia (Rhodophyta, Ceramiales) in the Vicinity of Port Aransas, Texas. Contributions in Marine Science, 22:105-120. Kiitzing, F. T. 1849. Species algarum. vi + 922 pages. Lipsiae: F. A. Brockhaus. 1863. Tabulae phycologicae. Volume 13, [i] + 31 pages. Nordhausen. Lauret, M. 1967. Morphologie, phenologie, repartition des Polysi? phonia marins du littoral languedocien, I: Section Oligosiphonia. Naturalia Monspeliensia (Botany), 18: 347-373. 1970. Morphologie, phenologie, repartition des Polysi? phonia marins du littoral languedocien. Naturalia Monospeliensia (Botany), 21:121-163. Mackay, J. T. 1836. Flora Hibernica. Volume 2, xxxiv + [3] + 354 + 279 pages. Dublin: Wm. Curry Jun. & Co. Meiiez, E. G. 1964. The Taxonomy of Polysiphonia in Hawaii. Pacific Science, 18(2):207-222. Oliveira Filho, E. C. de 1969. Algas marinhas do sul do Estado do Espirito Santo (Brasil), I: Ceramiales. Universidade de Sao Paulo, Faculdade de Filosofia, Ciencias e Letras, Boletim, 343 (Serie Botanica, 26): 277 pages, plates A-D + 1- 30, 1 map. 1977. Algas marinhas bentonicas do Brasil. [iii] + 407 pages. Sao Paulo, Brasil: Universidade de Sao Paulo (Instituto de Biociencias). Parke, M., and P. S. Dixon 1976. Check-List of British Marine Algae?Third Revi? sion. Journal of the Marine Biological Association, U.K., 56:527-594. Richardson, W. D. 1975. The Marine Algae of Trinidad, West Indies. Bul? letin of the British Museum (Natural History), Botany, 5(3):73-143, plates 16-27. Schnetter, R., and G. Bula Meyer 1977. Rodoficeas nuevas para la Costa Atlantica de Colombia. Anales del Instituto de Investigaciones Mar? inas Punta Betin, Santa Maria, Colombia, 9:81-90. Segi, T. 1951. Systematic Study of the Genus Polysiphonia from Japan and its Vicinity. Journal of the Faculty of Fisheries, Prefectural University of Mie, 1(2): 169-272. Setchell, W. A., and N. L. Gardner 1903. Algae of Northwestern America. University of Cali? fornia Publications, Botany, 1:165-418. Sprengel, K .PJ . 1827. Classis 24: Cryptogamia. In Caroli Linnaei . . . Sys- tema vegetabihum, editio decima sexta, 4:592 pages. Gottingen: Sumtibus Librariae Dieterichianae. Taylor, W. R. 1928. The Marine Algae of Florida, with Special Ref? erence to the Dry Tortugas. Papers from the Tor? tugas Laboratory, 25. Carnegie Institution of Wash? ington Publication, 379: v + 219, plates 1-37. 1935. Botany of the Maya Area, Miscellaneous Papers, VII: Marine Algae from the Yucatan Peninsula. Carnegie Institution of Washington Publication, 461: 115-124. 1960. Marine Algae of the Eastern Tropical and Subtropical Coasts of the Americas, ix + [iii] + 870 pages. Ann Arbor: University of Michigan Press. 1962. Marine Algae of the Northeastern Coast of North America. Second edition with corrections, viii + 509 pages. Ann Arbor: University of Michigan Press. 1969. Notes on the Distribution of West Indian Marine Algae Particularly in the Lesser Antilles. Contribu? tions from the University of Michigan Herbarium, 9(2): 125-203. Taylor, W. R., and A. J. Bernatowicz 1969. Distribution of Marine Algae about Bermuda. Bermuda Biological Station for Research, Special Publi? cation, 1: 42 pages. Bermuda: St. George's West. Taylor, W. R., and C. F. Rhyne 1970. Marine Algae of Dominica. Smithsonian Contributions to Botany, 3: i + 16. 238 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Tsuda, R. T., and C. J. Dawes 1974. Preliminary Checklist of the Marine Benthic Plants from Glover's Reef, British Honduras. Atoll Research Bulletin, 173:1-13. Womersley, H.B.S. 1979. Southern Australian Species of Polysiphonia Gre? ville (Rhodophyta). Australian Journal of Botany, 27(4):459-528. Yoneshigue-Braga, Y. 1972. Flora marinha bentonica da Baia de Guanabara e Cercanias, III: Rhodophyta; 3 Ceramiales. Pub- licacao do Instituto de Pesquisas da Marinha (Rw de Janeiro), 065:1-49, plates 1-9, map. Hydroidea (Cnidaria: Hydrozoa) from Carrie Bow Cay, Belize Barry W. Spracklin ABSTRACT Forty-five species of hydroids were found living on the reef and in the lagoon at Carrie Bow Cay. Collections were made each spring (March-June) from 1974 to 1978. Information was gathered on distribution, abundance, substrates, depth ranges, and reproductive maturity. Halcordyle disticha (Goldfuss), Halecium bermudense Congdon, and Hal? optens carinata Allman are the dominant hydroids in the outer reef areas and on patch reefs. Dyna- mena cornicina McCrady was common in all areas of the Thalassia (turtle grass) beds. Eleutheria di? chotoma Quatrefages, Halecium speciosum Nutting, and Egmundella grandis Fraser are recorded from the Caribbean Sea for the first time. Plumularia species is probably a growth form of P. floridana Nutting, the gonosome of which is undescribed. Halecium species is undescribed; corynid species, corymorphid species, and sertularid species are undescribed and appear to belong to new genera. Introduction This report on the hyrdroids of the Belizean barrier reef is based on collections made near Carrie Bow Cay during the spring months of each year from 1974 to 1978. Collections and obser? vations were made at nearly all areas during each of the five years. Many of the hydroids have been photographed in situ and almost all have been observed alive, prior to fixation. No major collection of hydroids has previously been made from the coast of Belize and little Barry W. Spracklin, Zoology Department, University of New Hamp? shire, Durham, NH 03824. material from this area of the Caribbean Sea was included by Nutting (1900, 1904, 1915), Fraser (1944, 1947), Gemerden-Hoogeveen (1965), or Vervoort (1968). In addition, most of the previous collections were made over a short period of time during a single year and were examined only after preservation. The major exception was Wed- ler's (1975) ecological study of the hydroids of Santa Marta (Colombia), which covered a 16- month period. ACKNOWLEDGMENTS.?I thank all of my numer? ous diving partners. S. Earle, P. M. Kier, M. E. Rice, and K. Riitzler collected some of the hy? droids; sponges were identified by K. Riitzler. C. W. Walker read the manuscript and made nu? merous useful suggestions. Methods During the spring months (March-June) of a 5-year period (1974-1978) 103 collections of hy? droids were made by skin diving and SCUBA diving. Conventional transecting techniques could not be applied readily because of the patchy distribution of hydroids in many areas, the cryptic occurrence of many hydroids, and the problem of identifying species with certainty underwater. Since, in most cases, a portion of each colony was collected for identification, the number of obser? vations of each species has been used for abun? dance estimates. The large number of hydroids (33 species) discovered on dead gorgonians and scleractinean corals does not necessarily indicate a preference for these substrates. Many small hydroids are commonly found on algae, but these are readily overlooked. 239 240 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES M a n y of the hydroids were pho tographed un? derwater or on the island before fixation, a n d most of the a theca te hydroids were d rawn from photographs . T h e thecate hydroids were d rawn from photographs and by camera lucida from fixed specimens. A reference collection is being prepared . Annotated Species List Suborder ATHECATA Family CORYMORPHIDAE Corymorph id species. Figure 113a. This hydroid is very similar to Corymorpha sym? metrica Harg i t t from the Phillipines; however, solitary polyps grow out of the ends of branches of the gorgonians Muriceopsis flavida (Lamarck) , Pseudopterogorgia acerosa (Pallas), Gorgonia flabel- lum Linnaeus , and G. ventalina Linnaeus ra ther t h a n on coralline algae. A m a t u r e hyd ran th has 16 to 22 proximal tentacles (usually coiled at the t ip) , 25 to 30 oral tentacles and seven to nine nematocyst clusters, each bear ing one to three lateral gonophores with no evidence of tentacles. Al though corymorphid sp. and C. symmetrica clearly belong in the family Cory? morph idae , these two species could easily be placed in a separate genus since they bo th have nematocyst clusters, a t tach to solid substrates, and lack tentacles on the gonophores or medusa buds . Family TUBULARIIDAE Ectopleura grandis Fraser. Figure 113c. T h e identification was m a d e from a colony that had only i m m a t u r e medusa buds. Family HALOCORDYLIDAE Halocordyle disticha (Goldfuss). Figure 113c. Family CORYNIDAE Corynid species. Figure H3d,e. This hydroid is common in the spur and groove zones on the encrust ing sponge Monanchora bar- badensis Hechtel and in the reef crest it was found on the bor ing sponge Cliona caribbea Carter . In the outer reef areas it was found on an unidentified sponge, and once on an uni ? dentified alga growing on a dead gorgonian. It has an oral whorl of four tentacles and a second whorl of eight to ten clusters of three tentacles each. T h e gonophores, which develop between the two whorls, were released in finger bowls immediate ly after collection with four apical nematocyst clusters, bu t no tentacles. Family ELEUTHERIIDAE Eleutheria dichotoma Quatrefages. Figure 113/. First record from the Car ibbean Sea. T h e me? dusa has been recorded from the Medi ter ra? nean Sea, British Isles, Sweden, and France (Brinkmann-Voss, 1970). On ly a single polyp with seven capi ta te tentacles was found on a b lade of turt le grass. Family ZANCLEIDAE Zanclea costata Gegenbaur . Figure 113g. T h e perisarc at the base of the h y d r a n t h was annu la t ed in the mater ia l on Sargassum and smooth in the mater ia l from the reef crest. Family CLAVIDAE Corydendrium parasiticum (Linnaeus). Figure 113/?. Not listed by Vervoort (1968) from the Carib? bean, this hydroid was recorded by Wedler (1975) from Colombia. Turritopsis nutricula (McCrady) . Figure 113?. Family BOUGAINVILLIIDAE Garveia humilis (Allman). Figure 113;'. Family EUDENDRIIDAE Eudendrium attenuatum Al lman. Figure \l3k,J. Eudendrium eximium Al lman. Figure 114a. Myrionema hargitti (Congdon) . Figure 1146. Suborder T H E C A T A Family CALICELLIDAE Egmundella grandis Fraser. Figure 114c. First Car ibbean Sea record. Family H A L E C I I D A E Halecium bermudense Congdon . Figure \\4d. Halecium namum Alder. Figure 114c. NUMBER 12 241 FIGURE 113.?Athecate hydroids, Corymorphidae-Eudendriidae: a, corymorphid species, some proximal tentacles removed to show the nematocyst clusters, X 10; b, Eclopleura grandis, some proximal tentacles removed to show the medusa buds, X 40; c, Halocordyle disticha, X \2; d, corynid species, X 45; e, corynid species, oral view, X 60; / , Eleuthena dichotoma, contracted after fixation, X 60; g, Zanclea costata, X 30; h, Corydendnum parasiticum, X 10; i, Turntopsis nutncula, X 10; j , Garveia humilis, X 50: k, Eudendrium attenuatum, main stem and side branch, X 8; /, Eudendrium attenuatum, male, X 32. 242 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 114.?Athecate hydroids, Eudendriidae (a,b), and thecate hydroids, Calicellidae-Hale- ciidae: (c-j) a, Eudendrium eximium, X 10; b, Myrionema hargitti, main stem and branch of female colony, X 13; c, Egmundella grandis, X 130; d, Halecium bermudense, X 30; e, Halecium namum, X 40; / , Halecium speciosum, X 50; g, Halecium speciosum, female gonophore, X 50; h, Halecium species, X 30; i, Halecium species, female gonophore, X 100; 7, Ophiodissa mirabilis, X 60. NUMBER 12 243 FIGURE 115.?Thecate hydroids, Campanulariidae-Sertulariidae: a, Clytia hemisphaerica, X 30; b, Clytia laxa, X 25; c, Clytia noliformis, X 25; d, Obelia dichotoma, X 40; e, Hebella calcarala, X 80; / , Hebella venusta, X 50; g, Scandia mutabilis, X 80; h, Cmdoscyphus marginalus, X 15; i, Diphasia tropica, X 3 0 ; / Dynamena cornicina, X 40. 244 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 116.?Thecate hydroids, Sertulariidae-Plumulariidae: a, Dynamena crisiodes, X 20; b, Sertularella parvula, X 50; c, Sertularella speciosa, X 15; d, Serlulana stookeyi, X 25; e, Sertularia turbinata, X 25; / sertularid species, X 40; g, sertularid species, showing tentacle-like nemato- phore, X 130; h, Antenella gracilis, X 70; i, Antenella quadriaurita, X 70; / Halopteris carinata, X 36; k, Halopteris cannata, male gonophore, X 36. NUMBER 12 245 FIGURE 117.?Thecate hydroids, Plumulariidae: a, Halopteris diaphana, X 35; b, Monostaechas quadridens, X 15; c, Plumularia halecioides, X 35; d, Plumulana halecioides, gonophore from base of stem or stolon, X 35; e, Plumularia margaretta, X 70; / Plumularia setacea, X 60; g, Plumularia species, X 50; h, Aglaophenia latecarinata, X 130; i, Aglaophenia pluma pluma, X 130. 246 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Halecium speciosum Nutting. Figure 114/g. Described from the western coast of North America, this is the first record from the Atlan? tic Ocean. Halecium species. Figure 1 \4h,i. The colonial growth form and strongly flared hydrophoral margin match the material that Vervoort (1968) referred to Halecium reflexum Stechow. Cornelius (1975a) reduced//, reflexum to a junior synonym of H. labrosum Alder. Ver- voort's material was not reproductive and the single gonophore present on my material is similar to H. nanum (Figure 114c) and not at all like the gonophore of H. labrosum. My material of H. nanum from the Carrie Bow Cay area has the normal growth form and hydrophoral mar? gin, and thus H. species probably represents an undescribed species. Ophwdissa mirabilis (Hincks). Figure 114;'. Cornelius (1975a) reduced Ophiodissa cacinifor- mis (Ritchie) to a junior synonym of 0. mirabilis. Family CAMPANULARIIDAE Clytia hemisphaerica (Linnaeus). Figure 115a. Listed as Campanularia {Clytia) johnstoni Alder by Vervoort (1968); see Millard (1966) for synonymy. Clytia laxa Fraser. Figure 1156. Listed as Laomedea (Phialidium) laxa by Ver? voort (1968). The genus Laomedea was reduced by Cornelius (1975b) to a junior synonym of Obelia. Clytia noliformis (McCrady). Figure 115c. Obelia dichotoma (Linnaeus). Figure ll5d. Listed by Vervoort (1968) as Laomedea (Obelia) congdoniHarght. Cornelius (1975b) referred nu? merous described species in the genus Obelia to three valid species. Family LAFOEIDAE Hebella calcarata (A. Agassiz). Figure 115c. Hebella venusta (Allman). Figure 115/. Scandia mutabilis (Ritchie). Figure 115g. Family SERTURLARIIDAE Cnidoscyphus marginatus (Allman). Figure 115a. Diphasia tropica Nutting. Figure 115z. Dynamena cornicina McCrady. Figure 115/ Dynamena crisioides (Lamouroux). Figure 116a. Sertularella parvula (Allman). Figure 1166. Sertularella speciosa Congdon. Figure 116c. The distinct hydrothecal base present in all of the shallow water specimens (31 m or less) was totally absent in the colony from 67 m. Sertularia stookeyi Nutting. Figure 116a". Sertularia turbinata Lamouroux. Figure 116c. Sertularid species. Figure 116/g. Superficially, this hydroid appears to be a nor? mal sertularid except that, in addition to the hydranth, each hydrotheca contains a large tentacle-like nematophore. In living material, either the nematophore or the hydranth, or both at once, may extend out of the hydro? theca. The operculum is composed of a single plate. This hydroid was collected in 1974 and not seen again until 1978 when it was found to be abundant on the underside of coral rubble in a different area of the reef crest. No repro? ductive material was evident. Family PLUMULARIIDAE Antenella gracilis Allman. Figure 116a. Antenella quadriaurita Ritchie. Figure 116z. Halopteris carinata Allman. Figure 116//:. Halopteris diaphana (Heller). Figure 117a. Monostaechas quadridens (McCrady). Figure 1176. Plumularia halecwdes (Alder). Figure 117c/ Plumularia margaretta (Nutting). Figure 117c. Plumularia setacea (Linnaeus). Figure 117/ Plumularia species. Figure 11Ig. This hydroid is probably conspecific with P. floridana Nutting; however, the gonophores were not described and the annulations were somewhat different than in my material. A more reliable identification will have to await an examination of the type material. Aglaophenia latecarinata Allman. Figure 117a. Aglaophenia pluma pluma (Linnaeus). Figure 117z. Distribution In order to facilitate collection and recording of data, the lagoon and reef around Carrie Bow Cay were subdivided into 10 areas. The lagoon includes patch reefs, mangrove islands, and the NUMBER 12 247 tur t le grass (Thalassia testudinium Banks ex Konig) beds. T h e reef was divided into the back reef, reef crest, high- a n d low-relief spur and groove, sand t rough, outer ridge, and fore reef slope zones (compare Riitzler and Macin tyre , herein: 9). T h e da ta in T a b l e 18 are the combined results of 103 collections m a d e over a five-year period (1974- 1978) and show the dis tr ibut ion, a b u n d a n c e , sub? strates, reproduct ive matur i ty , a n d dep th ranges of the hydroids found in these 10 areas. PATCH R E E F S . ? T h e pa tch reefs have the most diverse hydroid fauna (22 species) in the Carr ie Bow Cay area a n d even adjacent reefs vary con? siderably in composit ion and a b u n d a n c e of spe? cies. Density of hydroid colonies on the pa tch reefs is generally greater t han tha t in comparab le habi tats in the spur and groove, sand t rough, and outer ridge zones of the barr ier reef. Halocordyle disticha, Halecium bermudense, Halopteris carinata, An? tenella gracilis, and Plumularia setacea are found commonly on dead gorgonians and corals on patch reefs. Cnidoscyphus marginatus is unusual ly a b u n d a n t at one of the sites, overgrowing large areas of coral rock. A single hydroclad oi Antenella quadriaurita, the only record of this species in the present collection, occurred on a piece of coral rubble covered with numerous hydrocladia of A. gracilis. MANGROVE ISLANDS.?Dynamena cnsioides is found a b u n d a n t l y and exclusively on the man? grove roots at T w i n Cays and Wee Wee Cay. Myrionema hargitti and Halecium bermudense are con? centrated on the mangrove roots and banks at the nor th end of the channel dividing Twin Cays. Halecium bermudense, Clytia hemisphaerica, Obelia di? chotoma, Dynamena cornicina, and Plumularia hale? cioides are common on mangrove roots and algae at other locations a round the mangrove islands. T U R T L E GRASS B E D S . ? I n view of the large number of tur t le grass blades examined and the enormous surface area avai lable for set t lement it is surprising tha t only 12 species of hydroids are found here. Halecium bermudense and Dynamena cor? nicina are the only species c o m m o n th roughou t and bo th appea r to be most a b u n d a n t at the eastern sides of T w i n Cays a n d the pa tch reefs. Plumularia haleciodes, and to a lesser extent P. setacea, are also c o m m o n on the east side of Twin Cays. T h e only specimens of Eleutheria dichotoma (a single polyp) and Sertularia stookeyi (one colony) were collected in this habi ta t . In m a n y areas of the lagoon, algae replace hydroids as the major epihytic growth on turt le grass. BACK REEF.?Dynamena cornicina and Plumularia halecioides are common on the turt le grass in the back reef area between Carr ie Bow Cay and the reef crest (0-0.5 m). In 1974 Myrionema hargitti was found in a single pa tch of sand and rubble near the reef crest. Since then, this hydroid has steadily increased in a b u n d a n c e and in 1978 occurred in almost all the sand patches in this area, growing a t t ached to small pieces of rubble in the sand. At the nor th end of the island, Halecium speciosum N u t t i n g was collected on the calcareous green alga Halimeda species at a depth of approximately 0.5 m. Halecium species and H. nanum are also present here on Halimeda species; Halecium bermu? dense was collected from the brown alga Turbinaria species. R E E F C R E S T . ? M o s t of the nine hydroids from the reef crest were discovered on the underside of coral rubble slightly deeper than the normal low- tide level and were represented, for the most par t , by small stolonal colonies. T h e only common hydroid in this area is "sertularid species." A few hydran ths of "corynid species" were collected from one of the reef-crest channels on the boring sponge, Cliona caribbaea. Halocordyle disticha, Plu? mularia inermis, and P. setacea were collected in 2 m of water on dead gorgonians off the end of the reef crest south of Carr ie Bow Cay. H I G H - R E L I E F SPUR AND G R O O V E Z O N E . ? O n l y four hydroids were found a m o n g the high-relief coral spurs (2-6 m). T h e most common one in this zone is "corynid species," which occurs under overhanging coral on the sponge Monanchora bar- badensis. Halocordyle disticha and "corymorphid spe? cies" were collected on gorgonians at the seaward edge of the high spurs, whereas Halecium bermu? dense was found in crevices a m o n g the corals. L O W - R E L I E F SPUR AND G R O O V E Z O N E . ? N o n e of the 14 species collected here between depths of 7 a n d 12 m is a b u n d a n t , a l though Halocordyle disticha, Halecium bermudense, Halopteris carinata, and 248 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES V > I S B o cj5 5 T> TJ u bi n S o G O T J a se -O w TJ '3 y hy dr ? U B ow .ii C ar r o 0 3 ; f l 'C - D is 1. CO ? rA B LE OJ ra re II s e n t -O rt II 1 a te s B '? V u c C3 "0 c 3 o f a b c o e n ta ti -r. 1- (re p 19 78 r-. O) CO a so n - be c ;o ll ec ti o c II # - J G tfl :im e th J= u O Z ?? "5 fl'S ?TJ 3 h P ? CJ U |H OH CL E spiojpAj-i SUBIUOSJOS pUB SJBJOD pB3Q utnssvBivs ?UI}BO|j[ SJOOJ 3AOJ?UBJAJ s3pB|q mssvjvqj^ 3 B S J Y sdojs JSSJ-SJO^ s S p u J 3 j n o q S n o j j puBg 3UOZ 3AOOJ? puB j n d s Moq 3UOZ 3AOOJ? puB j n d s q S | H JS3J0 J33"?J J33J >|3Bg spsq vissvjm/j^ SpUB[Sl 3AOJ?UBJ/\J SJSSJ q^JBj CL t/3 r^ r-^ co ? r-~ ~ co ?? i? - H ^-i - C M - H M i r ! 0 i - . c o - ' i o c n ( N c o c n c o l f ! a ) S n i : r ) - H I I I I O O I I I I I I I I I O I ^ I I I * CM CN io l O C M c n - ^ r ^ c M C M C M c o uo in co * > - > s & 2 >, i C3 S Ju n CM 1 ri * G 3 ?* >-a! * G C 3 3 O ?? ? O >^ >^ CM ? CM ? ? w l I l l l l I l l I l ^ K ^ k y * I S> K> S> S> S/1 S^ ^ A n n I rS /S rS rS rS rS i*S 2 b C . p i eg rt a < CM CM I I I I X I I I I I I I I I I I I X I X I X I I I X I I l l l I I I I I I I X I X X I I I I X l I X X I I I X I l I I X I X I X I I I I X X X X X X X I X X X I I X X I X X X I X I X I I I CM I I ?i CO CM I CO I CO N I I | C M C M | j I CO CM CO CM I | C M C M | I I I CO I CO CO I I I I CM I ?' | CM I CM * I I I I I I I I I I I * l I CM i i i i i i I I I I I I I I I I I - I I I I - I I I I I I I l l l l l l | C M | C M | C O CO | CO | I I O I N I CM I | CO CM | | ? I I I co | | - N I I I I I I I CM CM ? CM CO CM | | M I I I CO CM | | CM | CM CM | CM I I I I I I I I I I I I CM | | CO | | CM | | t/3 Cl .ii "S y -e CL "5 ? 2 K ^3 J" ?f" !S -E o i i ^ O fe3 tt! s J 1 'E s ?c -2 .? S .2 "fe ? R ?J3 S s E .a .a ? ? i 1 ^ =1 ?? s .uis-s s ? NUMBER 12 249 ra te t o XI 3 C/3 a < _ C 4J ?Si bo--?s CL c G Q >- 0 *3 C* TJ a ' C 2 T J 5 ?fc 2 2 a, a E u spiojpAj-j SUBIUOSJOS p U B SJBJOD p B 3 Q utnssvSxvg ? U U B O [ j SJOOJ SAOJSUBJAJ sspBjq VISSV]VI{JL S B S J V 3 d o j S J33J -3 JO j 3 ? p u j s m o q S n o j j puBg 3UOZ 3AOOJ? puB j n d s Moq 3UOZ 3AOOJ? puB j n d s q S i H JS3JD J33>J J 3 3 J T\OVQ spaq mssvjvqj^ SpUBJSI 3AOJ?UBJAJ SJ33J q o j B j s CL C/2 CO 1 CM O l_ a < CM 1 S ? CM S CM m d 1 " O 1 O 1 1 o m o o rt G o **< ?< CM 1 1 * * * * * Q . CL < < ?n cn 1 X 1 1 1 1 1 X X 1 X X X X 1 X X X 1 CM CM - 1 1 1 1 O J CO CO s 5 "5 '-5 .? o 1 1 1 1 1 1 1 I 1 1 1 1 X X 1 1 1 1 1 X 1 1 1 1 1 1 1 1 CM | | CO | CM | CM | | CM | | | CM - H | CM | CM | | | | CM | CM j 1 1 1 1 1 1 1 CM | - | - | | 1 1 1 1 1 M 1 1 1 1 1 1 * 1 | | | | | CO ? * 1 1 1 ?* 1 ^ 1 3 a S r , 2 H eb ell a ca lca ra ta H . ve n u st a Sc an di a m u ta bi lis Cn id os cy ph us m a r D ip ha sia tr op ica D yn am en a co rn ici , D . cr isi oi de s CO 1 CM d # 1 1 1 1 X X CM 1 1 | 1 C J 1 O J 1 CM J3 3 5 5! a "?: t>3 - t3 o 1 1 i n ^ d co i i i - ? O CO > N cd * * cn 1 1 1 ? CO C N CO ? CM 1 1 1 i n - ^ C M CM 2 O J 1 Q . rt * CM C M 1 1 1 . Bl 1 1 X 1 1 1 1 1 1 1 1 1 1 1 1 1 <3 a ,g S -s: co i * 1 1 1 1 1 X 1 1 O J O J a 5 s "5, ? s o * O J * Cn * ??i r - O l r -? CO OJ co O J ? * ? * ? * a> o CO o CM ? o cn CM ? CM CM ? c o tfl ' ? CJ y CJ 3 2 CL O 5i u 1 M ?fe ,- *, . ".'T r ? FIGURE 121.?a, Madracis pharensis i. pharensis, fore-reef slope, 24-27 m, on underside oi Agaricia fragilis, specimen coated with NH4CI; b, c, Acropora palmata, off Cuba; d, A. cervicornis, exact locality at Carrie Bow Cay unknown; e, A. prolifera, off Belize, exact locality unknown. (Scale bars: a = 0.2 cm; b, d = 5 cm; c, e = 0.5 cm.) 278 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 122.?a, Acropora prolifera, off Belize, exact locality unknown; b, Agaricia agaricites f. carinata, fore-reef slope, 24-27 m; c, d, A. agaricites purpurea, sand trough, 21 m; e, A. agaricites agaricites, patch reef area, 2-3 m; / , A. tenuifolia, exact locality at Carrie Bow Cay unknown. (Scale bars: a, b,f= 5 cm; c = 0.5 cm; d, e ? 2 cm.) NUMBER 12 279 forms of the same species, whereas others (Smith, 1971; Roos, 1971; Wells, 1973) have retained them as separate species. Discussions of the prob? lem can be found in Vaughan (1901:312-314; 1919:482) and in Roos (1971:54). Family AGARICIIDAE Agaricia agaricites (Linnaeus, 1758) FIGURE I22b-e Agaricia agaricites.?Verrill, 1901a: 146-149 [in part, not va? riety e], figs. 5-7, pl. 26: figs. 2-3, pl. 27: figs 1-3, 5-7.? Squires, 1958:247-248, pl. 32: fig. 3, pl. 33: figs. 1-3.? Roos, 1971:56-58, pl. 14: figs, a-b.?Smith, 1971:74-75, pis. 5-6.?York, 1971:11, 13 [in part, pl. 4: figs. 3-8, pl. 5: figs. 1-6].?Wells, 1973:25, fig. 6a [not A. a. fragilis].?Tresslar, 1974:120, pl. 4.?Colin, 1978:220, 221 [color fig.], 226, 231. Agaricia crassa Verrill, 1901a: 145, pl. 30: fig. 6, pl. 34: fig. 2. Agaricia purpurea.?Verrill, 1901a: 149-150, pl. 27: figs. 4a-b. Colonies massive and encrusting; thick and flat, attached by pedicel; or flat with thick im? bricated, vertical, bifacial lobes. Up to 1 m in diameter. Cerioid to meandroid. Calices 1.5-3.0 mm in diameter, arranged in reticulate pattern or in rows of up to 20; 3-7 calices/cm. Columella small. Chocolate to purplish brown. HABITAT.?Agaricia agaricites forma agaricites is most common in back-reef and patch reef areas (0.5-2.0 m); A. a. forma purpurea (Lesueur, 1821) is common in the spur and groove zone and outer fore reef (7-30 m); and A. a. forma carinata Wells, 1973, was collected only from the fore-reef slope (18-30 m). Bathymetric range: 1-70 m. DISCUSSION.?At least five growth forms have been described and named for this variable spe? cies and several more forms could be described on the basis of similar criteria. The most common form at Carrie Bow Cay is purpurea, a large (up to 1 m), thick, flat colony with 5-6 calices/cm and long valleys having up to 20 calices. Each valley is bordered by prominent collines. Forma agaricites is smaller, hemispherical or encrusting, with larger calices arranged in a discontinuous or re? ticulate pattern and only 3-5 calices/cm. Forma carinata is like purpurea but has scattered, low (2-3 cm), thick, bifacial crests. The remaining two forms have not been collected at Carrie Bow Cay: forma humilis Verrill has small (1.5 mm in diam? eter), pinched calices in a crowded, reticulate arrangement; and forma danai Milne Edwards and Haime is massive with tall, erect, imbricated lobes. Agaricia fragilis forma fragilis Dana, 1848 FIGURE 123a-o Agaricia fragilis.?Verrill, 1901a: 142-145, pl. 26: fig. la-d.? Smith, 1971:75, pl. 8.?Wells, 1973:24 [not fig. 6a].? Colin, 1978:224 [color fig.], 234. Pedicelled or encrusting, very thin, unifacial fronds, either flat or vase-shaped. Usually less than 15 cm in diameter. Calices about 1.0 mm in diameter, 5-7/cm. Very low or no collines. Choc? olate or purplish brown. HABITAT.?Common on fore-reef slope (18-27 m). Bathymetric range: 3-40 m. DISCUSSION.?Agaricia fragilis is similar to A. agaricites forma purpurea, but is distinguished by its thinner, more delicate corallum and its lack of or very small collines. Roos (1971) considered A. fragilis a variety of A. agaricites. Agaricia tenuifolia Dana, 1848 FIGURE 122/; PLATE 3: center left, bottom left Agaricia agaricites.?York, 1971:9, pl. 3: figs. 5-6. Agaricia tenuifolia.?Wells, 1973:25 [key], fig. 13.?Erhardt and Meinel, 1975, fig. 5.?Werding and Erhardt, 1976, pl. 1: fig. 1.?Colin, 1978:224 [color fig.], 227 [fig.], 231. Large, bushy colonies up to several meters in size, composed of thin, vertical, dissected fronds. Corallum bifacial. Valleys short, sometimes re? ticulate. Collines prominent. Columella small. Brownish. HABITAT.?At Carrie Bow Cay this species is ubiquitous: it is common in the back-reef area (2-4 m) and is the major component of the buttresses and spurs in the spur and groove zone of the inner and fore reef (5-8 m). It is also found in the sand trough and fore-reef slope (20-27 m). Bathymetric range: 2-27 m. 280 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 123.?a, b, Agaricia fragilis f. fragilis, fore-reef slope, 24-27 m; c, d, Leptoseris cucullata, inner reef slope, 15-18 m; e,fi A. larmarcki, fore-reef slope, 24-27 m. (Scale bars: a, c ? 5 cm; b, f= 0.5 cm; d = 1 cm; e = 2 cm.) NUMBER 12 281 Agaricia lamarcki Milne Edwards and Haime, 1851 FIGURE I23e-f Agaricia lamarcki.?Wells, 1973:26-28, figs. 8-10.?Colin, 1978:234. Thin, unifacial, centrally attached fronds up to 2 m in diameter. Calices about 2 mm in diameter, 3-5/cm along valley. Valleys long, 4-6 mm wide. Septa thinner than interspaces, alternating in height and thickness. Collines prominent and thick. Columella large. HABITAT.?Collected only on fore-reef slope (25-27 m). Bathymetric range: 4-46 m. DISCUSSION.?Distinguished from Agaricia gra- hamae Wells, 1973, by its thinner septa, which alternate in size, and its wider valleys. Leptoseris cucullata (Ellis and Solander, 1786) FIGURE I23c-d Agaricia nobilis Verrill, 1901a:150-151, pl. 28: figs. 1-2.? Smith, 1971:75-76, pl. 7. Agaricia agaricites.?Roos, 1971:56-58 pl. 15. Helioseriscucullata.?Wells, 1973:25 [key], figs. 14,33.?Colin, 1978:229, 232 [color figs.], 234-235. Helioceris cucullata.?Tresslar, 1974:121, pl. 6. Leptoseris cucullata.?Dinesen, 1980:187-188, pl. 1: figs. 1-3. Very thin, unifacial, pedicelled fronds up to 40 cm in diameter. Underside costate. Calices about 3 mm in diameter, each of which is bordered by a prominent, crescent-shaped hood, which is strongly inclined toward the edge of the frond. Adjacent "hoods" eventually unite, forming a long colline. No columella. Green polyps on brown background. HABITAT.?Very common on fore-reef slope (25-28 m), particularly in slightly recessed cavi? ties. Also present in sand trough (22 m), inner- reef slope (14-16 m), and high spur and groove zone (6-9 m). Bathymetric range: 3-90 m. Family SIDERASTREIDAE Siderastrea siderea (Ellis and Solander, 1786) FIGURE 124 3 < 4 = 5 = 6 > 7 . Pereonites 4, 5, and 6 each with middorsal shallow pit. Pleonites 1-5 fused, bu t indicated ventrolaterally, wi th shallow dorsal grooves indicat ing lines of fusion; pleonite 6 free. Telson thin, hardly indura te , with hyaline border, widest at midlength , distodorsally slightly concave, with few distal setae; 2 large basal statocysts present. An tennu la r peduncle 3-segmented, basal seg? ment equal in length to 3 distal segments; flagel- lum of 3 articles. Antenna l peduncle 5-segmented, second segment grooved to accommoda te anten- nule; flagellum of 2 articles. M a n d i b u l a r pa lp 3- segmented, terminal segment short, bear ing 2 setae, middle segment longest, wi th elongate dis? tal seta; incisor of 2 b lun t cusps; l amina den t a t a wi th 4 margina l serrations; molar distally b lunt ly bilobed. Maxi l la slender with one strong spine 324 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES TABLE 24.?Distribution and morphological comparison of four species of Apanthura Distribution Pigment Dorsal pits Antennular flagellum basal article $ Antennal flagellum articles Maxillipedal endite Uropodal endite Size? A. magnifica Georgia to Florida absent slight depression on pereonites 4-6 elongate 4 present notched 8.5 mm A. signata Puerto Rico; Belize present slight depression on pereonites 4-6 short 3 present notched 4.5 mm A. significa Venezuela absent slight depression on pereonites 4-6 elongate 3 absent unnotched 5.0 mm A. geminsula Belize absent marked pits on pereonites 4-6 short 2 present notched 8.1 mm and 5 smaller distally curved spines. Maxilliped 5-segmented, terminal segment situated on outer distal angle of fourth segment, with 5 setae; third segment narrowed, less than half length of fourth segment; slender endite on inner surface with single terminal seta. Pereopod 1 unguis almost half length of dactylus; propodal palm with few simple spines, rounded hyaline lobe at about midpoint: carpus triangular, distally narrowed. Pereopods 2 and 3 considerably smaller than pereopod 1. Posterior pereopods with propodus having strong posterodistal spine and several small fringed scales on posterior margin; carpus small, triangular, underriding propodus, with strong posterodistal spine. Pleopod 1 exopod op- erculiform, distal margin bearing several plumose setae; endopod slightly shorter and about half width of exopod, with few distal plumose setae. Uropodal exopod distally notched, extending to distal end of basis; endopod extending beyond telsonic apex, distally rounded, bearing numerous simple setae. DESCRIPTION OF MALE.?Eyes considerably larger than those in female. Antennular flagellum of 8 or 9 articles, each bearing whorl of filiform aesthetascs. Mandibular palp as in female, but incisor, lamina dentata, and molar reduced to blunt nonsclerotised lobes. Pereopod 1 with hya? line process at about midpoint of propodal palm narrower than in female; numerous simple spines on medial margin; carpus with narrowly rounded hyaline posterodistal lobe. Pleopod 2 endopod with cylindrical, apically narrowly rounded cop- ulatory stylet extending beyond apices of rami. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble, coarse sediments, 0.2-1.5 m; Twin Cays, under mangroves, 0.2 m. Holotype: 2 (TL 8.1 mm), Twin Cays, USNM 171166. Allotype: 6 (TL 4.8 mm), Twin Cays, USNM 171167. Paratypes: 38, 3$, Twin Cays, USNM 171168. Additional Material: 120$, 3 sub<5, 18, 50 juve? niles. REMARKS.?Superficially the species of the ge? nus Apanthura are all remarkably similar, but subtle differences are to be found in the mandib? ular palp spination-setation, telsonic shape, max? illipedal segment proportions, and in the struc? ture of the first maxillipeds. Apanthura magnifica Menzies and Frankenberg (1966), A. signata, and A. significa Paul and Menzies (1971) have been recorded either from the Caribbean or from ad? joining areas, and are all figured as having pleon- ites 1-3 free, and pleonites 4 and 5 dorsally fused. Clearing specimens of A. magnifica and A. geminsula in lactic acid and Chlorozol Black shows that pleonites 1-5 are dorsally fused, but that pleonites 1-3 have a groove or fold over the dorsum which appears as an articulation. Careful examination of the type material of A. significa and A. signata shows a similar fusion. The main distinguishing features of the four species of Apanthura discussed here are summarized in Table 24. NUMBER 12 325 ETYMOLOGY.?The specific name geminsula is a Latinized form of Twin Cays, the locality where the species was abundant. Apanthura signata Menzies and Glynn FIGURE 143 Apanthura signata Menzies and Glynn, 1968:28, fig. 10.?Paul and Menzies, 1971:42. DESCRIPTION OF FEMALE.?Body hardly indur? ate. Body proportions: C < 1 = 2 > 3 < 4 < 5 > 6 > 7. Pereonites 4, 5, and 6 with shallow middorsal depression. Pleonites 1-5 dorsally fused, laterally distinct; pleonite 6 free. Telson not indurate, gently convex dorsally with rela? tively broad hyaline border, widest at midpoint; with 2 proximal statocysts. Antennular peduncle 3-segmented, basal seg? ment slightly shorter than 2 distal segments to? gether; flagellum of 3 articles, second article rel? atively elongate. Antennal peduncle 5-segmented, flagellum of 3 articles. Mandibular palp 3-seg? mented, terminal segment with 3 terminal setae, first and second segments each with single elon? gate seta; incisor of 3 blunt cusps, lamina dentata with 5 marginal serrations, molar blunt. Maxil- liped 5-segmented, with thin-walled endite on inner surface. Pereopod 1 unguis almost half the length of dactylus; propodal palm with rounded distal lobe and convex hyaline flange, few simple setae; carpus triangular, posterodistal angle pro? duced, narrowly rounded. Posterior pereopods with triangular carpus bearing strong posterod? istal spine. Pleopod 1 exopod operculiform, en? dopod more than half as wide and almost as long as exopod. Uropodal exopod notched, extending beyond basis. DESCRIPTION OF SUBMALE.?Eyes larger than in female, but not as large as in mature male. An? tennular flagellum lacking filiform aesthetascs. DESCRIPTION OF MALE.?Cephalon with eyes much larger than in female. Antennular flagel? lum of 9 articles bearing filiform aesthetascs. Mandible with palp as in female, but incisor, lamina dentata, and molar somewhat reduced. Pereopod 1 with propodal palm having low rounded process at about midpoint, hyaline con? vex flange on inner surface shorter than in female, but with more setae than in female. Pleopod 2 endopod with cylindrical, distally rounded cop- ulatory stylet extending well beyond rami. COLOR NOTES.?Pigment pattern fairly regu? lar, especially the dark band between eyes and narrow bands on anterior and posterior dorsal parts of pereonites, and on pleonite 6 in female. Pigmentation in male lacking regularity found in female; chromatophores scattered over dorsal and ventral surfaces. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble and coarse sediments, intertidal to 24 m: ~ 1209, 2 subo\ 16, 25 juveniles. PREVIOUS RECORDS.?Puerto Rico. Apanthuroides Menzies and Glynn Apanthuroides millae Menzies and Glynn FIGURES 144, 145 Apanthuroides millae Menzies and Glynn, 1968:30, fig. 11. DESCRIPTION OF FEMALE.?Integument hardly indurate, with numerous pits on cephalon, per- eon, pleon, and telson, and fine scales (seen with difficulty). Body proportions: C < 1 > 2 = 3 = 4 = 5 > 6 > 7 < P . Cephalon with short tri? angular rostrum not extending as far as antero? lateral corners. Eyes dorsolateral. Pereonite 7 very short, one-third length of 6. Pleonites 1-5 fused, with shallow grooves indicating lines of fusion; pleonite 6 fused with telson. Latter proximally broad, distal two-thirds narrowed, apically rounded, with strong middorsal longitudinal ridge. Antennular peduncle 3-segmented, basal seg? ment broad, equal in length to 2 distal segments plus 2 basal flagellar article; flagellum of 4 arti? cles, two distal articles each with single aesthetasc. Antennal peduncle 5-segmented, second segment grooved to accommodate antennule; flagellum of 7 articles. Mandibular palp 3-segmented, middle segment broadest and longest, terminal segment with 3 distal finely fringed spines; incisor of 2 blunt cusps, narrow lamina dentata with 6 or 7 326 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 143.?Apanthura signata Menzies and Glynn: a, complete specimen, 9; b, cephalon, 8; c, antennule 9; d, mandible; e, maxilliped;/, pereopod 1, 9; g, pereopod 1, 8; h, pereopod 2, 8; i, telson. NUMBER 12 327 FIGURE 144.?Apanthuroides millae Menzies and Glynn, holotype, 2: a, complete specimen; b, antennule; c, antenna; d, left mandible; e, right mandible; / , maxilla; g, pereopod I; h, pleopod 1; i, maxilliped; ^, uropodal exopod; k, pereopod 6. 328 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 145.?Apanthuroides millae Menzies and Glynn, 8: a, pleopod 2; b, cephalon; c, pereopod 1. serrations; molar of right mandible elongate, slen? der, finely ridged, absent on left mandible. Max? illa elongate, slender, apically with 3 or 4 broad? ened cusps, no spines distinguishable. Maxilliped 5-segmented, terminal segment semicircular, se? tose, second segment elongate, with well-devel? oped thin-walled endite on inner face. Pereopod 1 not markedly larger than following pereopods; unguis one-third length of dactylus, with short basal spine; propodus not very broad, palm straight, with 3 spines. Posterior pereopods, pro? podus with 2 distal finely-fringed spines; carpus slightly underriding propodus, anterior margin somewhat shorter than posterior margin; latter with strong posterodistal spine. Pereopod 7 ab? sent. Pleopod 1 exopod twice width and subequal in length to endopod, rami overlapping slightly, together forming opercular surface closing off branchial chamber. Uropodal exopod extending well beyond basis, twice as long as wide, distally rounded, margin entire, sparsely setose; endopod almost circular, sparsely setose. DESCRIPTION OF MALE.?Integument some? what less pitted than in female. Eyes enlarged. Antennular flagellum with 6 or 7 articles bearing filiform aesthetascs. Propodus of pereopod 1 with thin convex flange along palm, latter with 5 finely fringed spines in distal half. Pleopod 2 endopod with cylindrical copulatory stylet, apically rounded. MATERIAL EXAMINED.?Carrie Bow Cay, coarse sediments, 6-24 m: 3? (TL 2.3-2.8 mm), \6 (TL 2.8 mm), USNM 171154. REMARKS.?The type material of this species, consisting of two males from Puerto Rico, agrees well with the male of the present material. Men? zies and Glynn (1968), in describing A. millae, neither figured nor mentioned the characteristi? cally pitted integument, neither did they figure the unusual mandible. Several features of this species agree with Na- talanthura foveolata Kensley, 1978a, described from the southwest Indian Ocean. These similarities include the pitted integument, the 5-segmented maxilliped with endite, the elongate mandible with a slender molar present only on one side, pelopod 1 with the exopod and endopod together forming an operculum over the branchial cham? ber, and pleonites 1-5 fused, pleonite 6 fused with the telson. (With regard to the latter feature, Kensley (1978a) described the pleon of N. foveolata as having pleonites 1-3 free, 4 and 5 fused, 6 fused with the telson. Clearing of a specimen with lactic acid and Chlorozol Black has shown that pleonites 1-5 are completely fused, as in the present species). The aforementioned features, especially the unique mandibular structure, leave no doubt that Natalanthura should be regarded as a junior synonym oi Apanthuroides. Belizanthura, new genus DIAGNOSIS.?Eyes present. Antennal flagellum of 7 articles. Mandibular palp 3-segmented; in? cisor, lamina dentata, and molar reduced in 8, normal in 9. Maxilliped 7-segmented with endite in 9; 5-segmented and lacking endite in 8. Per? eopods 1-3 similar, subchelate. Pereopods 4-7 with triangular carpus underriding propodus. Pleopod 1 not operculiform. Pleonites 1-6 free. Telson lacking statocysts. NUMBER 12 329 TABLE 25.?Comparison of anthurid genera possessing a seven-segmented maxilliped Character Neohyssura Ocsanthura Minyanthura Belizanthura Eyes 8 mouthparts Carpus of pereopods 4-7 Pleopod 1 Pleonites Telson Statocysts absent p triangular non-operculiform 1-5 free, 6 fused with telson spiniform 7> absent similar to $ rectangular non-operculiform 1-6 free flattened 2 present similar to 2 rectangular operculiform 1-5 fused, 6 fused with telson flattened 2 present very reduced triangular non-operculiform 1-6 free flattened absent TYPE-SPECIES.?Belizanthura imswe, new species. ETYMOLOGY.?The generic name is derived from the country of Belize, plus anthura, the suffix often used for anthurid genera. REMARKS.?A 7-segmented maxilliped as in the female of Belizanthura also occurs in Neohyssura Amar (1952), Ocsanthura Kensley (1978b), and Minyanthura, new genus. The features separating these genera are summarized in Table 25. Belizanthura imswe, new species FIGURES 146, 147 DESCRIPTION OF FEMALE.?Body very slender, semi-transparent when alive, integument thin, not indurate. Body proportions: C > 1 < 2 = 3 < 4 = 5 > 6 > 7 . Cephalon with low triangular rostrum, tiny dorsal eyes. Pleonites free, 1-5 sub- equal, 6 with notch in posterodorsal margin. Tel? son widest at midlength, tapering to rounded setose apex, lateral margins distally with 7 or 8 shallow serrations; statocysts absent. Antennular peduncle 3-segmented, flagellum of 4 articles, terminal article tiny. Antennal pe? duncle 5-segmented; flagellum of 7 articles. Man? dibular palp 3-segmented, middle segment twice length of first, third segment bearing 3 distal fringed setae; incisor of 3 cusps; lamina dentata with 4 serrations and several tiny spinules; molar bluntly rounded. Maxilla with one large and 4 small distal spines. Maxilliped 7-segmented, ter? minal segment small, with 3 setae; third segment short; endite reaching to distal margin of fourth segment, thin-walled, with single terminal seta. Pereopods 1-3 similar, unguis half length of dac- tylus; propodal palm very slightly sinuous, with few scattered setae; carpus triangular, with pos? terodistal point. Posterior pereopods with unguis about one-quarter length of dactylus, propodus with 3 posterodistal sensory spines; carpus almost triangular, with short free anterior margin. Pleo? pod 1 not operculiform, subequal to following pleopods. Uropodal exopod pyriform, basally broad, outer margin with 4 or 5 serrations, distally rounded, setose; endopod longer than basis, ex? tending slightly beyond telson, distally rounded, setose. DESCRIPTION OF MALE.?Antennular flagellum of 8 articles, each with whorl of filiform aesthe? tascs. Eyes enormously enlarged, almost meeting middorsally and midventrally, with just sufficient space between to accommodate mandibular palps; remainder of mandible reduced to short nonmasticatory segment. Maxillae absent. Max? illiped 5-segmented, only terminal segment bear? ing 3 setae; endite absent. Pereopod 1 unguis about one-third length of dactylus; propodal palm straight, armed with about 8 short, fringed spines; carpus triangular, with apical fringed spine and short, acute process. Pereopod 2 pro? podus slightly shorter than pereopod 1, palm armed with 3 strong ventrodistal sensory spines. Posterior pereopods as in female. Pleopod 1 basis with 2 slender retinaculae, exopod and endopod subequal in length, broadly rounded distally both bearing elongate plumose setae. Pleopod 2 basis 330 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 146.?Belizanthura imswe, new species, holotype, 9: a, complete specimen; b, antenna; c, antennule; d, mandible; e, maxil la; / , maxilliped; g, pereopod 1; h, pleopod 1; i, uropod; ;', telson. NUMBER 12 331 FIGURE 147.?Belizanthura imswe, new species, 8: a, cephalon; b, mandible; c, maxilliped; d, pereopod 1; e, pereopod 2 ; / pereopod 7; g, pleopod 2. with 3 retinaculae; endopod bearing stout, api- rubble and shallow coarse sediments, 0.1-0.3 m; cally rounded copulatory stylet extending well Twin Cays, in algal mat under mangroves, 0 .1- beyond ramus; exopod with distinct transverse 0.3 m. suture at midlength. Holotype: 9 (TL 3.4 mm), Twin Cays, USNM MATERIAL EXAMINED.?Carrie Bow Cay, coral 171172. 332 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES Allotype: 6 (TL 2.5 mm), USNM 171173. Paratypes: 49 (TL 3.0-3.7 mm), 46 (TL 2.3 mm-2.7 mm), Twin Cays, USNM 171174. Additional Material: 279, 28. ETYMOLOGY.?The specific name is the acro? nym for "Investigations of Marine Shallow Water Ecosystems." FIGURE 148.?Mesanthura fasciata new species, holotype, 9: a, complete specimen; b, antennule; c, antenna; d, mandible; e, maxilla;/, maxilliped; g, pleon in lateral view. NUMBER 12 333 FIGURE 149.?Mesanthura fasciata new species, 6": a, pereopod 1; 6, pereopod 7; c, pleopod 1. Mesanthura, Barnard Mesanthura fasciata, new species FIGURES 148, 149 DESCRIPTION OF FEMALE.?Integument not in? durate. Body proportions: C < 1 < 2 = 3 < 4 = 5 > 6 > 7. Cephalon with low rostrum, not extending beyond anterolateral corners; eyes dor? solateral. Pereonites 2 and 3 each with anterodor- sal hollowed area. Pleonites 1-5 fused, indicated laterally, and with dorsal grooves marking fusion lines; pleonite 6 free, with middorsal notch in posterior margin. Telson dorsally slightly convex, distal margin rounded. Antennular peduncle 3-segmented, flagellum 3-segmented, distal segment one-quarter length of second segment, with 3 aesthetascs. Antennal peduncle 5-segmented, second segment grooved to accommodate antennule; flagellum 3-articu- late. Mandibular palp 3-segmented, third seg? ment one-third length of second, with 4 finely fringed spines; incisor of 3 cusps; lamina dentata with 4 serrations; molar bluntly rounded. Maxilla with 1 strong spine and 5 smaller distally curved spines. Maxilliped 5-segmented, terminal seg? ment small, set obliquely at outer distal angle of fourth segment; tiny thin-walled endite present at base of third segment. Pereopod 1 dactylus with small spine at base of unguis; propodal palm with rounded lobe at about midlength; 5 simple setae on inner face of propodus; carpus triangular, distally somewhat produced and rounded. Per? eopods 2-7 similar; posterior pereopods with tri? angular carpus bearing short posterodistal spine, underriding propodus; latter with 2 serrate pos? terodistal spines; posterior margin bearing tiny fringed spines. Pleopod 1 exopod operculiform, extending somewhat beyond endopod; both rami with distal plumose setae; basis with 3 retinac- ulae. Uropodal exopod with strong distal notch, fringed with plumose setae, extending beyond base of endopod; latter oval, extending slightly beyond telsonic apex. COLOR NOTES.?Red-brown pigment pattern constant, characterized by irregular patch on cephalon, delicate reticulation on pereonites 1-3; thin anterior submedian lines and posterior solid transverse bars on pereonites 4-7; pleon with 5 short transverse bars on fused pleonites 1-5, 334 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES pleonite 6 with fine middorsal tracery; patches of pigment on telson and uropodal bases and endo- pods. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble and coarse sediments, intertidal to 24 m. Holotype: 9 (TL 4.5 mm), Carrie Bow Cay, USNM 171162. Paratypes: 69 (TL 3.7-4.5 mm), USNM 171163. Additional Material: 289, 25 juveniles. REMARKS.?On the basis of the color pattern, the present species bears some resemblance to M. occidentalis Menzies and Barnard, 1959, from southern California, especially in the delicate tra? cery on pereonites 1-3. The Californian species, however, lacks the strong posterior pigment bars on pereonites 4-7 and the 5 shorter bars on the pleon. The 2 species can further be separated on the form of the maxilliped. ETYMOLOGY.?The specific name, derived from the Latin fasciata (striped), refers to the transverse bands of pigment on the posterior pereonites and on the pleon. Mesanthura paucidens Menzies and Glynn FIGURES 150, 151 Mesanthura paucidens Menzies and Glynn, 1968:27, fig. 9A-G. DESCRIPTION OF FEMALE.?Anterolateral lobes of cephalon rounded, not extending beyond ros? trum. Cephalon with marked dorsolateral ridge; midventral margin posterior to maxilliped evenly convex. Pereon and pleon not indurate. Pleonites 1-5 fused, sutures barely discernible along ven? trolateral margin; pleonite 6 free, with narrow middorsal notch in posterior margin. Telson dor? sally gently convex, distal margin evenly rounded. Antennular peduncle 3-segmented, distal seg? ment very short, flagellum of 3 articles, terminal article bearing several setae and 2 aesthetascs. Antennal peduncle 5-segmented, second segment grooved to accommodate antennule; flagellum of 3 (?4) short setose articles. Mandibular palp 3- segmented, terminal segment shortest and nar? rowest, with 6 stout spines; incisor of 3 blunt cusps, linked to somewhat reduced and rounded molar by 6-serrate lamina dentata. Maxilla slen? der, with 6 terminal spines. Maxilliped 5-seg? mented, second segment with reduced endite on inner face. Pereopod 1 unguis half length of dactylus; propodal palm finely crenulate, with small convex transparent process at midpoint bearing simple setae. Pereopod 2 unguis one-third length of dactylus; propodus cylindrical, with strong serrate distoventral spine; carpus triangu? lar. Pereopods 5-7 with carpus and propodus each with distoventral spine. Pleopod 1 exopod operculiform, about 3 times width of endopod; basis with 4 retinaculae. Uropodal exopod (Fig? ure 150^) distally sinuous rather than notched. DESCRIPTION OF MALE.?Eyes larger than in female. Antennular flagellum of 7 articles, each with whorl of filiform aesthetascs. Pereopod 1 with lobe at about midpoint of propodal palm, inner surface of propodus with many finely ser? rate spines. Pleopod 2 copulatory stylet cylindri? cal, extending beyond rami, distally spinulose, apically rounded and slightly sclerotised. COLOR NOTES.?Female with roughly rectan? gular dorsal patches of chromatophores on ce? phalon and pereonites. Pleon with 5 laterally linked transverse bars. Telson and uropodal en? dopod and basis with proximal chromatophores. Male with pigmentation of pereon heavier and less defined than in female, but with 5 pleonal bars as in female. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble and shallow sediments: 259 (TL 6.6 mm), 16* (TL 6.4 mm), 9 juveniles. Twin Cays, under mangroves: 19. PREVIOUS RECORDS.?Puerto Rico. REMARKS.?The palm of pereopod 1 has a definite rounded process at midlength, but as this is transparent, it may have been overlooked and therefore not figured by Menzies and Glynn (1968). The presence of a small maxillipedal en? dite is unusual in Mesanthura, but may be related to the relatively small body size, and the almost interstitial habit. The fact that the holotype is only about one-third the length of the present mature male and females may account for the omissions in the original description. NUMBER 12 335 w ?????: y?' m?&s lex FIGURE 150.?Mesanthura paucidens Menzies and Glynn, 9: a, complete specimen; b, antenna; c, antennule; d, cephalon in lateral view; e, pleon in lateral view; / mandible; g, maxilla; h, maxilliped; z, pereopod 1;/ , pereopod 2; k, pereopod 7; /, pleopod 1. 336 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 151.?Mesanthura paucidens Menzies and Glynn, 8: a, pereopod 1; b, antennule; c, pleopod 2. Mesanthura pulchra Barnard FIGURES 152, 153 Mesanthura pulchra Barnard, 1925:145, fig. 9e.?Schultz, 1969:109, fig. 151. Mesanthura decorata Menzies and Glynn, 1968:26, fig. 8A- I . DESCRIPTION OF FEMALE.?Cephalon with ros? trum reaching as far forward as anterolateral angles. Body proportions: C < 1 = 2 > 3 < 4 = 5 = 6 > 7. Anterior 5 fused pleonites equal in length to pereonite 7; individual pleonites indi? cated by very short ventrolateral incisions; pleon? ite 6 free, with middorsal slit in posterior margin. Telson broadly rounded distally. Antennular peduncle 3-segmented, basal seg? ment longest and broadest, flagellum of 3 articles. Antennal peduncle 5-segmented, second segment broadest, hollowed dorsally to accommodate an? tennule; flagellum of 3 short articles. Mandibular palp 3-segmented, middle segment longest, distal segment shortest, with row of 10 spines; incisor of 4 rounded cusps, linked to blunt rounded molar by lamina dentata having 5 serrations. Maxilla slender, with 5 or 6 distal spines. Lower lip bi- lobed, ending in narrowly rounded process, with fine lateral setae. Maxilliped 5-segmented, second segment longest; third segment slightly narrower than fourth; latter with 4 short setae on median margin; distal segment triangular, with 2 stout, fringed setae and few simple setae. Pereopod 1 unguis one-third length of dactylus; propodal palm with rounded process at about midpoint bearing row of 6 setae, remainder of propodus and carpus with few scattered setae. Pereopod 2 unguis one-quarter length of dactylus; propodus with distoventral spine and few setae; carpus very short, triangular. Pereopods 5-7, propodus with distoventral simple spine plus fringed spine and several short spinules; carpus triangular, under? riding propodus, with distoventral simple spine. NUMBER 12 337 FIGURE 152.?Mesanthura pulchra Barnard, 9: a, complete specimen; b, antenna; c, antennule; d, pereopod 1; e, m a n d i b l e ; / maxilla; g, maxilliped; h, lower lip; i, pleopod \;j, pereopod 2; k, pereopod 7; /, pleon in lateral view. 338 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 153.?Mesanthura pulchra Barnard, 8: a, antennule; b, pereopod 1; c, pleopod 2. Pleopod 1 exopod operculiform, 3 times width of endopod, with numerous plumose setae on distal margin; endopod with 9 plumose setae on distal margin; basis with 6 retinaculae. Uropods typical of genus. DESCRIPTION OF MALE.?Antennule with 3-seg? mented peduncle, flagellum of 7 articles, each bearing whorl of filiform aesthetascs. Pereopod 1 similar to female, but with unguis half length of dactylus, propodus with dense band of spines on inner face of paim. Pleopod 2 copulatory stylet of endopod simple, cylindrical, extending well be? yond rami. COLOR PATTERN.?Broad band of chromato? phores on cephalon posterior to eyes; hollow, roughly rectangular dorsal patch on pereonites 1-6, pereonite 7 and pleon with pigment band laterally broad, narrowed middorsally. Scattering of chromatophores on telson and uropods. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble and shallow sediments: 129 (TL 9.3 mm), 4 juveniles. PREVIOUS RECORDS.?St. Thomas and St. John, U.S. Virgin Islands; Puerto Rico; Dry Tortugas, Florida. TYPE MATERIAL.?Barnard's type series from the Copenhagen Museum consists of 28 and 1 ovigerous 9 syntypes, St. Thomas and St. John, 10-18 fathoms (18-33 m); ovigerous 9 lectotype (TL 6.5 mm); 2d (TL 5.4 mm, 4.7 mm) paralec- totypes. REMARKS.?Although Barnard described this species in 1925, examination of the 3 type speci? mens still reveals the pigment pattern quite clearly. Menzies and Glynn (1968) based their species M. decorata on differences in the pigment patterns but made no comparison of appendages. Comparison of the Carrie Bow Cay material and that from the Dry Tortugas, Barnard's types, and Menzies and Glynn's types shows that despite slight differences (for example, in the presence or absence of a clear area on the cephalon), the basic pigment pattern is the same. Comparison of the appendages, especially the first pereopods in both the male and female shows no differences in the 4 groups of specimens available. Schultz's figure NUMBER 12 339 (1969, fig. 151) misrepresents the pigment pat? tern. Mesanthura punctillata, new species FIGURES 154, 155 DESCRIPTION OF FEMALE.?Integument moder? ately indurate. Cephalon with large dorsolateral eyes, low triangular rostrum extending slightly beyond anterolateral corners. Body proportions: C < 1 = 2 > 3 < 4 = 5 > 6 > 7 . Pleonites 1-5 fused, only pleonite 1 indicated laterally; pleonite 6 free, with middorsal slit in posterior margin. Telson broadly rounded distally. Antennular pe? duncle 3-segmented, flagellum of 3 articles. An? tennal flagellum 3-articulate. Mandibular palp 3-segmented, middle segment twice length of dis? tal segment, latter armed with 7 finely fringed spines; incisor with 3 cusps; lamina dentata with 5 blunt marginal serrations; molar acute, slightly sclerotised. Maxilla with 1 strong and 4 smaller spines. Maxilliped 5-segmented, distal segment semicircular, with 4 setae; penultimate segment with 2 medial distal setae. Pereopod 1 unguis half length of dactylus, with tiny spine at base; pro? podus proximally broad, palm with hyaline ser? rate lobe at midlength; carpus narrowly triangu? lar, distal rounded part bearing 5 serrations. Pos? terior pereopods with short triangular carpus un? derriding propodus; posterior margin of propodus bearing short, fringed scales; posterodistal corner with 1 simple and one serrate spine. Pleopod 1 exopod operculiform, about 3 times wider than endopod; both rami with distal plumose setae; basis with 4 retinaculae. Uropodal exopod with outer sinuous margin, but not notched; endopod almost circular. DESCRIPTION OF SUBMALE.?Antennular flagel? lum elongate and swollen, but lacking aesthetascs. Pereopod 1 as in female. Pleopod 2 not yet differ? entiated. DESCRIPTION OF MALE.?Eyes larger than in female, extending dorsally and ventrally. Anten? nule with flagellum of 10 articles bearing whorls of filiform aesthetascs. Mandible with palp as in female; incisor and lamina dentata reduced and not sclerotised; molar absent. Pereopod 1 as in female but with dense band of simple spines on inner face of propodus. Pleopod 2 endopod with copulatory stylet extending well beyond rami, apically rounded. COLOR NOTES.?Female with pigment pattern consisting of almost solid red-brown bar between eyes and extending in lobes posteriorly; pereon, pleon, telson, and uropods bearing scattered and separate pigment spots. (When chromatophores are expanded, pigment pattern is still scattered and does not become reticulate.) Male with chro? matophores scattered dorsally and ventrally over entire body, denser than in female and with no discernible pattern. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble, coarse sediments, intertidal to 12 m. Holotype: Ovigerous 9 (TL 6.4 mm), Carrie Bow Cay, USNM 171157. Allotype: 8 (TL 4.5 mm), Carrie Bow Cay, USNM 171158. Paratypes: 8 (TL 4.5 mm), Carrie Bow Cay, USNM 171159; 39 (TL 5.7 mm, 5.2 mm, 5.2 mm), Carrie Bow Cay, USNM 171160. Additional Specimens: 59, 2 sub<5, 8 juveniles. REMARKS.?The distinctive pigment pattern distinguishes this species from its congeners from the same area. Other differences may be seen in the number of mandibular palp spines, shape of the uropodal exopod, the degree of lateral indi? cation of the fusion of pleonites 1-5, and the armature of the first pereopod. ETYMOLOGY.?The specific name punctillata, de? rived from the Latin word for small spots, refers to the overall scattered spots of pigment. Mesanthura reticulata, new species FIGURE 156 DESCRIPTION OF FEMALE.?Integument hardly indurate. Body proportions: C = 1 < 2 = 3 = 4 < 5 > 6 > 7. Cephalon with dorsolateral eyes; tiny triangular rostrum. Pleonites 1-5 fused; 6 free, with middorsal notch in posterior margin. 340 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 154.?Mesanthura punctillata new species, holotype, 9: a, complete specimen; b, antenna; c, antennule; d, mandible; e, maxil l iped; / pereopod 1; g, pereopod 7; h, maxilla; i, pleopod 1. NUMBER 12 341 FIGURE 155.?Mesanthura punctillata new species, 8: a, pereo? pod 1; b, mandible; c, pleon in lateral view; d, pleopod 2. Telson dorsally flattened, distal margin evenly rounded. Basal antennular peduncle segment broader than, but equal in length to, 2 distal segments; flagellum of 3 articles. Antennal flagellum of 3 articles. Mandibular palp, distal segment with 6 finely serrate spines; incisor with 3 acute cusps; lamina dentata with 5 marginal serrations; molar thumb-like, blunt. Maxilla with 1 strong and 4 slender distal spines. Maxilliped 5-segmented, lacking endite. Pereopod 1 unguis about one- third length of dactylus, propodus broad, palm with hyaline toothed lobe at midlength; carpus triangular, distal rounded part with about 6 ser? rations. Posterior pereopods with propodus bear? ing 3 strong posterodistal sensory spines; several spinules on posterior margin; carpus with short anterior margin, underriding propodus, with sin? gle sensory spine at anterodistal corner. Pleopod 1 exopod operculiform, 3 times width and sub- equal in length to endopod, both rami with distal plumose setae; basis with 5 retinaculae. Uropodal exopod ovate, outer margin sinuous, extending slightly beyond distal end of basis, fringed with plumose setae. COLOR NOTES.?Red-brown chromatophores form reticulate pattern on dorsal surface of ce? phalon, pereon, and pleon. Scattered chromato? phores on telson and uropods. MATERIAL EXAMINED.?Carrie Bow Cay, coarse sediment, 24 m. Holotype: 9 (TL 6.1 mm), Carrie Bow Cay, USNM 171161. REMARKS.?Mesanthura reticulata somewhat re? sembles M. punctillata in having scattered chro? matophores with a concentration of pigment on the cephalon between the eyes, as well as in the shape of the maxilliped and first pereopod of the female. Mesanthura punctillata, however, does not have a notched uropodal exopod, the third man? dibular palp segment has 7 more slender spines, rather than the 6 found in M. reticulata, and the body proportions of the 2 species also differ. Although only 1 specimen of this species was collected, the pigment pattern is distinctive enough to warrant the formation of a new species. ETYMOLOGY.?The specific name derives from the reticulate dorsal pigment pattern. Minyanthura, new genus DIAGNOSIS.?Antennular flagellum of 1 article; antennal flagellum of 4 articles. Mandible lacking palp and molar process. Maxilliped 7-segmented, bearing endite. Pleonites 1-5 fused; pleonite 6 fused with telson. Telson with 2 basal statocysts. Pleopod 1, exopod and endopod together forming operculum over branchial chamber. Carpus of pereopods 4-7, rectangular, not underriding pro? podus. TYPE-SPECIES.?Minyanthura corallicola, new spe? cies. 342 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES FIGURE 156.?Mesanthura reticulata new species, holotype, 9: a, complete specimen; b, antenna; c, antennule; d, mandible; e, maxi l l iped; / pereopod 1; g, maxilla; h, pereopod 7; i, uropodal exopod; j , pleopod 1. NUMBER 12 343 ETYMOLOGY.?The generic name is derived from the Greek minus (tiny), and anthura, the suffix used for many anthurid genera. REMARKS.?Three anthurid genera, Belizan? thura, Ocsanthura, and Neohyssura, possess a 7-seg? mented maxilliped with an endite. All 3 genera, however, possess nonoperculiform first pleopods and pleonites 1-5 free, and Ocsanthura and Neo? hyssura have a short triangular carpus on the posterior 3 pairs of pereopods, which underrides the propodus. The present specimens (assigned herein to Minyanthura corallicola), with pleonites 1- 5 fused, and pleonite 6 fused with the telson, and operculiform pleopod 1, and a rectangular carpus on the posterior pereopods, obviously cannot be members of any of the above-mentioned genera. Table 25 summarizes these differences. Minyanthura corallicola, new species FIGURES 157, 158 DESCRIPTION OF FEMALE.?Integument not in? durate. Body proportions: C > 1 < 2 < 3 < 4 = 5 > 6 > 7. Cephalon with broadly rounded rostrum extending beyond anterolateral corners; eyes dorsolateral. Pereonite 7 very short. Pleon only slightly longer than pereonite 7; pleonites 1- 5 fused, only indicated ventrolaterally; posterior margin of pleonite 5 with row of plumose setae; pleonite 6 fused with telson. Telson broad, distal margin crenulate, broadly rounded or truncate, with few setae; gently convex, longitudinal mid? dorsal ridge present; broad hyaline margin; 2 large basal statocysts. Antennular peduncle 3-segmented, basal seg? ment as long as rest of appendage, segments 3 and 4 equal in length, and half width of second segment; flagellum reduced to single very short setose article. Antennal peduncle 5-segmented, second segment curved ventrally, subequal in length to segments 3 and 4; flagellum of 4 articles, basal article longer than 3 distal articles together. Mandible with palp represented by single simple seta; incisor of 3 cusps, lamina dentata narrow, with 6 marginal serrations; molar absent. Maxilla with single stout spine and 5 shorter hooked spines. Maxilliped 7-segmented, 5 distal segments together shorter than second segment; thin- walled endite tipped with single seta, which reaches base of terminal palp segment. Pereopod 1 unguis one-third length of dactylus; propodus with straight unarmed palm, with single stout serrate spine and irregular row of fine combs of setules on inner face. Pereopod 2 similar to per? eopod 1. Posterior pereopods propodus with 2 distal serrate spines; ventral margin with row of setule-combs; carpus rectangular, not underrid? ing propodus. Pleopod 1 exopod and endopod lying side by side, subequal in length, together forming operculum over branchial chamber; ex? opod broadening distally, almost 3 times wider than endopod; both rami with distal plumose setae. Uropodal basis with row of plumose setae on outer margin: exopod widening distally, outer distal angle produced into acutely triangular lobe; inner distal angle rounded and dentate; endopod oval, distal margin serrate, endopod and exopod with broad hyaline border. DESCRIPTION OF MALE.?Body proportions as in female. Eyes larger than in female. Antennular peduncle 3-segmented, flagellum of 2 articles with single terminal aesthetasc. Pereopod 1 as in fe? male. Pleopod 2 endopod with copulatory stylet formed by club-shaped extension of distal end of ramus. COLOR NOTES.?Female, cephalon with broad dark-red-brown band between and posterior to eyes, with lateral unpigmented spots, centrally continuous with pattern on pereonite 1; latter with anterodorsal branching patch; pereonite 2 with dark medio-dorsal ramifying pattern; per? eonite 3 with 2 slender lateral bars; pereonite 4 with broad posterodorsal rectangle; pereonite 5 with slender posterodorsal rectangle; tiny wedge between pereonites 6 and 7; pleon with solid middorsal patch with lateral extensions. Male, pigment pattern less defined than in female, with scattered ventral patches. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble, 6-24 m. Holotype: larvigerous 9 (TL 1.7 mm), Carrie Bow Cay, USNM 171169 (2 larvae in brood pouch, with pigment pattern developed). Allotype: 6 (TL 1.3 mm), Carrie Bow Cay, USNM 171170. 344 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 157.?Minyanthura corallicola new species, holotype, 9: a, complete specimen; b, antenna; c, antennule; d, lateral margin of anterior pleon; e, mandible; / maxilla; g, maxilliped; h, pereopod 1; i, pereopod 6. NUMBER 12 345 FIGURE 158.?Minyanthura corallicola new species, 8: a, pleopod 2; b, pleopod 1; c, antennule; d, telson and uropod. Paratypes: 49 (TL 1.5-1.8 mm), Carrie Bow Cay, USNM 171171. ETYMOLOGY.?The specific name corallicola which means coral-dwelling, is used because all the specimens were removed from coral rubble. Pendanthura, Menzies and Glynn Pendanthura tanaiformis Menzies and Glynn FIGURES 159, 160 Pendanthura tanaiformis Menzies and Glynn, 1968:32, fig. 12A-I . DESCRIPTION OF FEMALE.?Integument moder? ately indurate. Cephalon half length of pereonite 1, with rostrum extending beyond anterolateral corners; eyes present in bases of anterolateral lobes. Body proportions: C < 1 > 2 = 3 < 4 = 5 > 6 > 7 > P. Brood pouch formed by 3 pairs oostegites on pereonites 3-5. Pleon very reduced, one-third length of pereonite 7; pleonites indi? cated only on ventrolateral margins by very short sutures. Telson dorsally flattened, distal margin broadly rounded, pair of statocysts situated at about midlength. Antennular peduncle 3-segmented, basal seg? ment longest and broadest, distal segment with reduced flagellum of 2 short articles, plus seta- bearing papilla; peduncle armed with large pin? nate setae, each articulating on distinct slightly 346 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 159.?Pendanthura tanaiformis Menzies and Glynn, 9: a, complete specimen; b, antenna; c, antennule; d, maxilla; e, mand ib l e ; / maxilliped; g, pleopod 1; h, pereopod 1; i, pereopod 7. NUMBER 12 347 FIGURE 160.?Pendanthura tanaiformis Menzies and Glynn, 8: a, antennule; b, pereopod 1; c, pleopod 2 endopod. broader base. Antennal peduncle 5-segmented, segments with minute setules; flagellum reduced to single short article. Mandibular palp reduced to small papilla bearing single seta; incisor of 3 rounded, slightly sclerotised cusps; molar re? duced; lamina dentata plate with 7 marginal serrations. Maxilla slender, with 8 distal curved spines, terminal spine strongest. Maxilliped 3-seg? mented, terminal segment with several distal sim? ple setae; thin-walled endite present on inner face, with 2 fine distal setae. Pereopod 1 unguis almost half length of dactylus, ventral margin of latter with fine spinules, strong supplementary spine at base of unguis; propodus with 6 serrate spines on inner face, rounded, very thin transpar? ent lobe on palm with few simple setae. Pereopods 2-7 similar, propodus with posterior margin with crenulations bearing clusters of short spinules and with strong serrate posterodistal spine; carpus short, triangular, underriding propodus; propo? dus, carpus, merus, and ischium bearing numer? ous short fine setules. Pleopod 1 exopod opercu? liform, with numerous laterodistal plumose setae; endopod half width of exopod, with 6 distal plumose setae; basis with 2 retinaculae. Uropodal exopod distally narrowly rounded, outer (dorsal) margin crenate, with numerous plumose setae, reaching slightly beyond basis; endopod ovoid, bearing plumose and elongate simple setae. DESCRIPTION OF MALE.?Antennular peduncle 3-segmented, flagellum of 4 articles, each with cluster of aesthetascs. Pereopod 1 propodus with dense cluster of curved serrate spines on inner surface, palm with rounded transparent lobe at about midlength. Copulatory stylet of pleopod 2 endopod attached at about proximal third, cylin? drical, longer than ramus, apically narrowly rounded, with minute scattered spinules. COLOR NOTES.?Strong reticulate red-brown pigmentation dorsally on cephalon, pereonites, antennae and first pereopods, becoming diffuse on pleon, telson and uropods. The young are released from the brood pouch fully pigmented. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble: 37 ovigerous 9, 829, 456*, 4 juveniles. PREVIOUS RECORDS.?Puerto Rico. 348 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES REMARKS.?This is the second record of this unusual genus and species, and the first record of the male. It was thought useful to supplement Menzies and Glynn's description and to figure most of the appendages. A few features require further comment. Menzies and Glynn (1968) de? scribed the mandibular palp as being reduced to 2 setae. In fact, the palp may be regarded as 1- segmented, this seta-bearing segment being re? duced to a small papilla. Menzies and Glynn mentioned the possible existence of a second fla? gellar ramus of a single article, on the antennule. More likely, this apparent segment is the slightly enlarged articulated base of one of the large specialized pinnate setae. Family PARANTHURIDAE Accalathura, Barnard Accalathura crenulata (Richardson) Calathura crenulata Richardson, 1901:509, figs. 1-4; 1905:74, figs. 58-61. Accalathura crenulata.?Barnard, 1925:147, pl. 4: fig. 18.? Nierstrasz, 1941:242.?Menzies and Glynn, 1968:33, fig. 13A-H.?Schultz, 1969:96, fig. 128. MATERIAL EXAMINED.?Carrie Bow Cay, inter? tidal to 12 m, sediments, 99, 38, 9 juveniles; Twin Cays, under mangroves, 19. PREVIOUS RECORDS.?Bahamas; Puerto Rico; Yucatan, 40 m; Brazil; Cape Verde Is. Paranthura, Bate and Westwood Paranthura caribbiensis, new species FIGURES 161, 162 DESCRIPTION OF FEMALE.?Integument not in? durate, with sparse scattered chromatophores. Cephalon with patch of chromatophores between dorsolateral eyes; tiny, triangular rostrum, not extending beyond anterolateral corners. Body proportions: C < 1 < 2 > 3 < 4 > 5 > 6 > 7 . Pleonites 1-5 free, subequal; pleonite 6 with bi- lobed posterodorsal margin. Telson dorsally flat, basally constricted, distal margin evenly rounded, with few setae. Antennular peduncle 3-segmented, basal seg? ment wider than and subequal in length to 2 distal segments; flagellum of 4 articles bearing aesthetascs. Antennal flagellum of single flattened triangular article. Mandibular palp 3-segmented, second segment two and one-half times length of basal segment, third segment with 4 stout fringed spines. Maxilla slender, with 8 distal serrations. Maxilliped 3-segmented, with short endite at base of third segment. Pereopod 1 unguis about one- third length of dactylus; propodus proximally broad, with convex flange and row of 12-14 spines on inner face. Pereopods 2 and 3 similar, subchelate, smaller than pereopod 1; propodal palm armed with 5 sensory spines. Pereopods 4- 7 ambulatory, propodus with 2 posterior sensory spines; carpus rectangular, with 2 posterior sen? sory spines. Pleopod 1 exopod operculiform, four times wider and slightly longer than endopod, both rami with distal plumose setae; basis with 4 retinaculae. Uropodal endopod subcircular; exo? pod oval, outer margin slightly sinuous, extending beyond endopod base. MATERIAL EXAMINED.?Carrie Bow Cay, coral rubble, shallow sediments, and under mangroves. Holotype: Larvigerous 9 (TL 4.5 mm), Carrie Bow Cay, USNM 171164 (8 larvae in brood pouch). Paratype: Ovigerous 9 (TL 4.5 mm), Carrie Bow Cay, USNM 171165. Additional Material: 109, 4 juveniles. REMARKS.?This small species of Paranthura is easily distinguished from the larger P. infundibu- lata, which also occurs at Carrie Bow Cay, by the serrate and curved telson in the latter species. Paranthura barnardi Paul and Menzies, 1971, re? corded from Venezuela, is more similar to P. caribbiensis in both size and structure. These 2 species can be separated on the uropodal struc? ture (broadly ovate and crenulate in P. barnardi, narrowly ovate/sinuous, and entire in P. caribbien? sis), the telson, which is distally more rounded in P. caribbiensis, the presence of a small maxillipedal endite in this species (absent in P. barnardi), the NUMBER 12 349 I /. FIGURE 161.?Paranthura caribbiensis new species, holotype, 9: a, complete specimen; b, antennule; c, antenna; d, mandible; 20 specimens; FC = fairly common, 5-19 specimens; P = present, < 5 specimens) Species Accalathura crenulata Apanthura geminsula A. signata Apanthuroides millae Belizanthura imswe Mesanthura fasciata M. paucidens M. pulchra M. punctillata M. reticulata Minyanthura corallicola Paranthura caribbiensis P. infundibulata Pendanthura tanaiformis Coral rubble 0-1.5 m A P P A FC FC FC FC FC A Coarse sediments 0-1.5 m FC A A P A A P P P Algal mat under 0 mangroves -1.5 P A - - A - P - - - - P - - m Coarse sediments 6-12 m P P P - P - P - - - Coarse sediments 24 m _ - P P P P P - - - mals, including isopods, amphipods, small mol? lusks, polychaetes, sipunculans, and pycnogonids. Ten species of anthurideans were collected here, several of which also occurred in coarse sediments of weed beds and at the bases of patch corals. Pendanthura tanaiformis and Paranthura infundibulata were found only in the coral rubble, the former in great numbers. The dark wine-red pigmenta? tion of both species, which live in the tiny holes in the Corallinacea-encrusted coral fragments, probably has a protective function. Probably also connected with this overriding red coloration of the coral rubble is the fact that 8 of the 10 species occurring here have some degree of integumental pigmentation. 2. Coarse white sandy sediments in shallow water. These sediments were taken from Thalassia weed beds, Syringodium weed beds, or from the sandy patches between patch reefs at a depth of 1.5 m or less. These sites are in protected areas with little wave action, either in the lagoonal area west of the island, or in the shallows between the main reef and the island. Eight species of anthur? ideans were common to these habitats and to the coral rubble area atop the reef, with Mesanthura fasciata, M. paucidens, and Apanthura geminsula being fairly common to abundant in both habitats. 3. Shallow water sediments and algae at bases of mangroves. This habitat is characterized by calm water in the shade of the mangrove trees, with high accumulations of organic debris, espe? cially mangrove leaves and rootlets, and Thalassia leaves. In the several large samples taken under the mangroves, three of the five species of an? thurideans collected occurred only rarely, whereas Belizanthura imswe and Apanthura geminsula were abundant, especially in the dense carpet of Caulerpa verticillata J. Agardh growing in less than 15 cm of water between the mangrove trunks and roots. Although Belizanthura was taken rarely from other living plants, it is abundant only in the Caulerpa verticillata. This algal mat, with its high organic debris content, and with a complex web of fine algal species (including Cladophora sp., Centroceras sp., Ceramium sp., blue-green algae and diatoms), which floats above the Caulerpa at high- tide and sinks onto the Caulerpa at low-tide, har? bors a rich fauna, including many asellote and gnathiid isopods, amphipods, a high diversity of pycnogonids (C. A. Child, pers. comm.), poly- 352 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES chaetes, nematodes, cumaceans, leptostracans, and small mollusks, especially the bivalved gas? tropod Berthelinia. 4. Coarse sediments and rubble from the chan? nel between Carrie Bow Cay and South Water Cay. This area is somewhat protected from wave action, but subject to the scouring action of water moving through the channel at a depth of 6 to 12 m. Five species of anthurideans were collected here, none of which appeared to be abundant. 5. Sand trough at base of coral and gorgonian- covered slope. Only two species, Apanthura signata and Mesanthura fasciata were collected from this area (~24 m depth) and from the intertidal zone. Both were abundant only in the very shallow sediments. Interestingly, Minyanthura corallicola re? sembles Pendanthura tanaiformis both in overall size and in having a very short anterior pleonal area. Both species seem to live in crevices in coral rubble, for which a reduction of the body length may be an advantage. GENERAL OBSERVATIONS.?Only male anthur? ids were taken in the horizontal plankton samples, in shallow water over the reefs suggesting that the males leave the substrate in search of females. The development of enlarged eyes, filiform aesth? etascs of the antennular flagella, and pigmenta? tion over the entire body (rather than confined to the dorsum in female Mesanthura spp., for exam? ple) are probably adaptations for this reproduc? tion-centered activity. It is unlikely that feeding is a reason for this increased activity, considering the reduction of mouthparts seen in the males of Belizanthura imswe, Apanthura geminsula, and Mes? anthura punctillata. The genus Mesanthura, represented here by five species (plus single specimens of two additional species not dealt with in this report) shows very successful radiation into several of the microha? bitats mentioned above. Four of these five species co-occur in the coral rubble environments, as well as in the coarse sediments from shallow water. Closer investigation will probably reveal distinc? tive feeding preferences, and even behavioral dif? ferences (considering the distinctive and constant dorsal pigmentation of each species), which would account for this apparent overlap of several species of the same genus. Literature Cited Amar, R. 1952. Isopodes marins du littoral corse. Bulletin de Societe Zoologique de France, 77:349-355. Barnard, K. H. 1925. A Revision of the Family Anthuridae (Crustacea Isopoda), with Remarks on Certain Morphologi? cal Peculiarities. Journal of the Linnaean Society of London, Zoology, 36:109-160. Camp, D. K., N. H. Whiting, and R. E. Martin 1977. Nearshore Marine Ecology at Hutchinson Island, Florida, 1971-1974, V: Arthropods. Florida Marine Research Publications, 25:1-63. Kensley, B. K. 1978a. The South African Museum's Meinng Naude Cruises: The Isopoda Anthuridea from the 1975, 1976, and 1977 Cruises. Annals of the South African Museum, 77(1): 1-25. 1978b. A New Genus and Species of Anthurid Isopod from Deep Water off the East Coast of the United States. Proceedings of the Biological Society of Washing? ton, 91:558-562. Menzies, R. J., and J. L. Barnard 1959. Marine Isopoda of Coastal Shelf Bottoms of Southern California: Systematics and Ecology. Pacific Naturalist, 1 (11 -12): 1 -35. Menzies, R. J., and D. Frankenberg 1966. Handbook of the Common Marine Isopod Crustacea of Georgia. 93 pages. Athens, Georgia: University of Georgia Press. Menzies, R. J., and P. W. Glynn 1968. The Common Marine Isopod Crustacea of Puerto Rico. Studies on the Fauna of Curacao and Other Car? ibbean Islands, 27:1-133. Nierstrasz, H. F. 1941. Gnathiidea, Anthuridea, Valvifera, Asellota, Phreatocoidea. In Die Isopoden der Siboga-Expe- dition, IV: Isopoda Genuina. Siboga-Expeditie, Vit- komslen op Zoologisch, Botanisch, Oceanographisch en NUMBER 12 353 Geologisch Gebied . . . uitgegeven door . . . M. Weber, Monographic, 32(d): 1-308. Paul, A. Z., and R. J. Menzies 1971. Sub-Tidal Isopods of the Fosa de Cariaco, Vene? zuela, with Descriptions of Two New Genera and Twelve New Species. Boletin de Instituto Universidade Oriente, 10:29-48. Richardson, H. 1901. Key to the Isopods of the Atlantic Coast of North America with Descriptions of New and Little Known Species. Proceedings of the United Stales Na? tional Museum, 23:493-579. 1902. The Marine and Terrestrial Isopods of the Ber? mudas, with Descriptions of New Genera and Species. Transactions of the Connecticut Academy of Sciences, 11:277-310. 1905. A Monograph of the Isopods of North America. Bulletin of the United States National Museum, 54:1? 727. Schultz, G A. 1969. How to Know the Marine Isopod Crustaceans, i-vii + 359 pages. Dubuque, Iowa: W. C. Brown. Pycnogonida from Carrie Bow Cay, Belize C. Allan Child ABSTRACT This is a systematic account of pycnogonids found in the area of Carrie Bow Cay, Belize. Thirty one identified species in 14 genera are discussed of which four are described as new species: Hedgpethius mamillatus, Callipallene belizae, Anoplodactylus imswe, and Rhynchothorax crenatus. Most species discussed were found within their previously known range of geographical distri? bution. Three species are reported for only the second time: Hedgpethius tridentatus, previously known from Florida, Parapallene bermudensis, from Bermuda, and Anoplodactylus bahamensis, from the Bahamas. At Carrie Bow Cay most pycnogonids occur within the littoral depth zone, some extend their range into the sublittoral to a depth of 33 m. Habitats include mangrove roots, algae, and seagrasses, as well as coral and coral rubble. Introduction This is the first report of pycnogonids from the Belizean barrier reef. Only a small collection from the northern portion of this barrier reef in the Territorio de Quintana Roo, Mexico, has been treated in a previous publication (Child, 1979). Previous to that report, the nearest occurrence of pycnogonids reported in the literature were deep water captures of Anoplodactylus lentus Wilson and Ascorhynchus serratus Hedgpeth off Yucatan (Hedg- peth, 1948:226, 259). Anoplodactylus lentus is also listed from deep water in the Yucatan Channel (Stock, 1975:1055). Other nearby captures are recorded from Cuba to the northeast and from the Caribbean coast of Panama to the south. C. Allan Child, Department of Invertebrate Zoology, National Mu? seum of Natural History, Smithsonian Institution, Washington, D.C. 20560. Most published work on pycnogonids centers on descriptive systematics. The small size, cryptic coloration, and very slow movement of these predators make it difficult to record their habits and habitats in the field. Therefore, our knowl? edge of the ecology of this group is rather poor. Although no attempt was made in this study to observe pycnogonids alive, ecologically relevant information on substrates and associations was obtained and is presented. All specimens are deposited in the pycnogonid collections of the National Museum of Natural History, Smithsonian Institution, and bear the catalog numbers of the United States National Museum collection (USNM). ACKNOWLEDGMENTS.?My appreciation is ex? tended to the various collectors listed under "ma? terial examined" and to P. M. Kier for supporting my 1976 field work in Belize. I am grateful to T. E. Bowman for his critical reading of the manu? script. Methods Pygnogonids were found in a wide variety of microhabitats such as bryozoans, hydroids, and encrusting sponges, algae, or seagrass, and rock rubble covered with one or more of these. Most samples were collected in a depth of less than one meter, some to a maximum depth of 33 m. The gross samples were either examined live under a low-power microscope or preserved before study. Plants and broken-up rock and rubble were agi? tated in a bucket with dilute formalin in seawater to separate and preserve the microfauna. Floating organisms were then skimmed off and the re? mainder of the liquid poured through a net to 355 356 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES concentrate the sample. Other suitable substrates such as dock pilings, coral heads, mangrove roots, and rock faces were scraped or chipped into plastic bags for preservation and later examina? tion. All specimens illustrated in this report were drawn by the author with the aid of a compound microscope and a camera lucida. Species List Family AMMOTHEIDAE Achelia Hodge Achelia sawayai Marcus Ammothella Verrill Ammothella appendiculata (Dohrn) A. exornata Stock A. marcusi Hedgpeth A. rugulosa (Verrill) Ascorhynchus Sars Ascorhynchus latipes (Cole) Ascorhynchus sp. cf. serratus Hedgpeth Eurycyde Schiodte Eurycyde raphiaster Loman Hedgpethius Child Hedgpethius mamillatus, new species H. Tridentatus Child Nymphopsis Haswell Nymphopsis duodorsospinosa Hilton Tanyslylum Miers Tanystylum birkelandi Child 7A. tubirostrum Stock Family CALLIPALLENIDAE Callipaltene Flynn Callipallene belizae, new species C. emaciata (Dohrn) Parapallene Carpenter Parapallene bermudensis Lebour Pigrogromilus Caiman Pigrogromitus timsanus Caiman Family PHOXICHILIDIIDAE Anoplodactylus Wilson Anoplodactylus bahamensis Child A. balangensis (Heifer) A. evelinae Marcus A. imswe, new species A. jonesi Child A. mantimus Hodgson A. monolrema Stock A. mulliclavus Child A. pectinus Hedgpeth A. partus Caiman Anoplodactylus sp. Family ENDEIDAE Endeis Philippi Endeis spinosa (Montagu) Family NYMPHONIDAE Nymphon Fabricius Nymphon floridanum Hedgpeth Family RHYNCHOTHORACIDAE Rhynchothorax Costa Rhynchothorax architectus Child R. crenatus, new species F a m i l y AMMOTHEIDAE Achelia Hodge, 1864 Achelia sawayai Marcus, 1940 Achelia sawayai.?Krapp and Kraeuter, 1976:342-343 [liter? ature].?Child, 1979:7-8. MATERIAL EXAMINED.?Carrie Bow Cay: On corals, 9.1 to 30.5 m, by hand with SCUBA, 7 Apr 1973; 16 with eggs. Tidal flats in 0.5 m, coll. K. Riitzler, 2 May 1974; 19. Tidal flats in 0.5 m, coll. K. Riitzler, 4 May 1974; 16. Tidal flats in 0-1 m, coll. K. Riitzler, 7 May 1974; 16. Wind? ward side, reef crest and flat in 0.5 m, 12 Jan 1976; 1 juvenile. Broken rock and rubble at tide line, coll. J. Clark, 17 Jan 1976; 66, 19, 2 juveniles. Tidal flats, coll. M. Jones, 8 Apr 1976; 16. Coral rubble on inner side of reef crest in 0.2-0.4 m, coll. A. Cohen, 15 Jan 1978; 19. Rubble at tide line, coll. C. A. Child, 27 Jan 1978; 3 specimens. Rubble at tide line, ocean side, coll. C. A. Child, 28 Jan 1978; 7 specimens. Rocks with algae on reef crest, coll. C. A. Child, 29 Jan 1978; 8 specimens. Algae of the genera Halimeda and Cau? lerpa from outer reef crest, coll. C. A. Child, 30 Jan 1978; 9 specimens. From the seagrass Syrin? godium on lagoon flats in 1.3 m, coll. B. Kensley, I Feb 1978; 19. Ocean side flats, plankton net on bottom in 0.4 m, coll. R. Larson, 30 Jan 1978; 1 juvenile. Syringodium bed at S end of lagoon in 1.3 m, coll. B. Kensley, 3 Feb 1978; 66, 1 juvenile. Rubble from tide line in 0.3 m, coll. C. A. Child, 4 Feb 1978; 3 specimens. Rubble in front of reef crest in 0.6 m, coll. B. Kensley, 5 Feb 1978; 26, 19. Carpet of coralline and fine red algae at shore, coll. B. Kensley, 5 Feb 1978; 39. From Halimeda NUMBER 12 357 in same place; 7 specimens. Halimeda and sparse rubble from top of reef ridge, in 18 m, coll. C. A. Child, 7 Feb 1978; 19. From other Halimeda and rubble nearby, coll. B. Kensley, 7 Feb 1978; 2 specimens. Tobacco Reef: About 500 m N of South Water Cay, debris at low tide, coll. M. Carpenter and R. Larson, 23 Mar 1977; 26. Twin Cays: N end of dividing channel, on Halimeda in 1 m, coll. C. A. Child, 31 Jan 1978; 19. NW coast, mat oi Caulerpa verticillata]. Agardh and mangrove rootlets beneath mangroves, inter? tidal, coll. C. A. Child, 31 Jan 1978; 3 specimens. NW coast, another Caulerpa mat under man? groves, coll. C. A. Child, 2 Feb 1978; 4 specimens. NW coast, mat of Halimeda under mangroves, intertidal, coll. C. A. Child, 2 Feb 1978; 4 speci? mens. REMARKS.?This species is by far the most com? mon pycnogonid in the Carrie Bow area and occurs from the tide line to a depth of 18 meters. It is also common elsewhere in the tropical North and South Atlantic to depths of 65 meters. It has also been found in Madagascar. ECOLOGY.?Several general collections of algae made in the shallow waters of various cays indi? cate that Achelia sawayai lives on associated rubble and other substrates rather than on the algae itself. This species and others were generally not found on clean algae at Carrie Bow Cay, but were common where the algae and rubble were associated with adherent detritus and sessile fauna. Therefore, it is concluded that algae do not provide food for A. sawayai, but only a sub? strate for food attachment. Ammothella Verrill, 1900 Ammothella appendiculata (Dohrn, 1881) Ammothella appendiculata.?Stock, 1955:250-252, fig. 18 [lit? erature]; 1975:973-975.?Child, 1974:497; 1979:9. MATERIAL EXAMINED.?Carrie Bow Cay: Sand trough behind outer reef ridge in 27 m, in coral sand and rubble, coll. C. A. Child, 7 Feb 1978; 2 juveniles. Rubble with Halimeda from top of outer reef ridge in 18 m, coll. B. Kensley, 7 Feb 1978; 16 juvenile. Blue Ground Range (Cays): At S end of north? ernmost cay, on mangrove roots in 0-1 m, coll. C. A. Child, 30 Apr 1976; 1 juvenile. Twin Cays: NW coast, from Rhizophora roots with algae, sponges, ascidians, and hydroids in 0.5 m, coll. C. A. Child, 31 Jan 1978; 36 with eggs, 19, 3 juveniles. From Halimeda mats along N channel wall in 1 m, coll. C. A. Child, 2 Feb 1978; 36 with eggs, 16, 79, 15 juveniles. From red sponge along wall of N channel in 1.0-1.5 m, coll. M. Carpenter, 2 Feb 1978; 26 with eggs. REMARKS.?There is some controversy over the status of this species and Ammothella rugulosa (page 358). The wide morphological variation displayed by A. appendiculata almost bridges the gap between the two species. Stock (1955:250-252, fig. 18) described two forms of A. appendiculata: one with the relatively short appendages of European spec? imens, and another form with very long append? ages from the Caribbean. Stock considered that his two forms reflected differences between the ultimate and penultimate molt in adults, the ultimate molt producing the "long form." The above material is in agreement with the "long form" but includes several chelate juveniles, thus invalidating the use of adult morphology as an explanation for the range of variation. Raising even more questions than these littoral specimens are the juveniles from depths of 18 and 27 meters with both abdomen and ocular tubercle more than twice as long as the "short form." As Stock pointed out, apparently no single visible character can be used to separate Ammoth? ella appendiculata from A. rugulosa taxonomically, if indeed they are separate species. It is somewhat easier to separate them on the basis of chelifore scape segment lengths (subequal for A. rugulosa) and ocular tubercle and abdomen length (meas? urably shorter for A. rugulosa). Stock (1975:973) noted that these differences are probably insuffi? cient to separate the two species. I have no stronger criteria for keeping the species separate, but with the inadequate material currently avail? able, I will regard them as separate until a larger 358 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES collection can be measured and diagnosed with more certainty. ECOLOGY.?This species has been found in a wide variety of littoral habitats throughout the Caribbean and on the Pacific side of Panama besides Europe and the Middle East. It has been collected in association with Rhizophora, several species of algae, Thalassia, sand, rubble, corals, sponges, and ascidians. This distribution indicates a wide food preference, probably including hy? droids and other coelenterates, sponges, and pos? sibly ascidians. Ammothella exornata Stock, 1975 Ammothella exornata Stock, 1975:975-978, figs. 7c-d, 8.? Child, 1979:9, fig. 3a-c. MATERIAL EXAMINED.?Twin Cays: NW coast, mat of Caulerpa verticillata and mangrove rootlets under Rhizophora in intertidal, coll. C. A. Child, 2 Feb 1978; 39, 2 juveniles. REMARKS.?The male characters were de? scribed and figured by Child (1979:9, fig. 3a-c). This is the third record of this species. Ammothella exornata is so distinctive that it is easily recognized, even as a juvenile. It is the only known species from the Caribbean littoral having median trunk tubercles. ECOLOGY.?In Belize, this species is known only in association with Rhizophora mangle Linnaeus and Caulerpa sp. I collected several juveniles as? sociated with algae (not Caulerpa) on the prop roots of R. mangle in the U.S. Virgin Islands. The type-specimens, from St. Martin, were also col? lected on and near this mangrove. The only other capture record is Stock's (1975) two specimens collected on algae at Bonaire. The species, how? ever, is not common in this mangrove habitat. At least six algae samples were taken from mangrove roots at about the time of the above capture but no other specimens of Ammothella exornata were found. The preferred sessile fauna used for food probably was not with the algae on and around the roots where the six samples were taken. No bryozoans, sponges, or hydroids were found in the Caulerpa sample. Therefore, the primary food of this species, as with most pycnogonid species, remains unknown. Ammothella marcusi Hedgpeth, 1948 Ammothella marcusi Hedgpeth, 1948:247-249, fig. 39b-g; 1954:427.?Stock, 1975:975, fig. 7a-b.?Child, 1979:9, 11. MATERIAL EXAMINED.?Carrie Bow Cay: tide flats in 0.5 m, coll. K. Riitzler, 4 May 1974; 16. Twin Cays: NW Coast, mat of Caulerpa verticil? lata and mangrove rootlets under Rhizophora in intertidal, coll. C. A. Child, 2 Feb 1978; 16. REMARKS.?These specimens agree well with Hedgpeth's figures of the type, also a male, and Stock's clarifying figures of the palp and terminal leg segments. Superficially, Ammothella marcusi is much more setose than the other species of Am? mothella in this report. There are many more clubbed, plain, and feathered setae on these spec? imens than have been figured before. The species is much less common than its nearest Caribbean relation, Ammothella appendicu? lata, which is known to have an extremely wide range of habitats in its amphi-Atlantic distribu? tion. The above records extend the known range of A. marcusi to Belize from Florida, Panama, the eastern Caribbean islands, and the Mexican Pa? cific coast. ECOLOGY.?This species has been found in shal? low water on debris, a sandy reef, Lithothamnion flats, rubble, algae, and on tide flats. This distri? bution shows no pattern except that the principal food of the species is not restricted to particular substrates. Ammothella rugulosa (Verrill, 1900) Ammothella rugulosa.?Stock, 1975:972 [literature].?Child, 1979:11. MATERIAL EXAMINED.?Carrie Bow Cay: Rub? ble along tide line on ocean side, coll. J. Clark, 17 Jan 1976; 16. Large Halimeda clump from SE shore, coll. B. Kensley, 5 Feb 1978; 96, 29. South Water Cay: SW end, scrapings from dock pilings in 0.5 m, coll. C. A. Child, 30 Jan 1978; 1 juvenile. NUMBER 12 359 Twin Cays: N channel, sponge with ectoprocts in 0.5 m, coll. R. Larson, 31 Jan 1978; 2 juveniles. NW coast, mat of Caulerpa verticillata and man? grove rootlets under Rhizophora in intertidal, coll. C. A. Child, 31 Jan 1978; 76 with eggs, 19, 3 juveniles. Another mat from nearby, coll. C. A. Child, 2 Feb 1978; 56 with eggs, 26, 39, 2 juve? niles. Halimeda mat from nearby, coll. B. Kensley, 2 Feb 1978; 19, 1 juvenile. N channel, red sponge from wall of channel in 1.0-1.5 m, coll. M. Car? penter, 2 Feb 1978; 16 with eggs. REMARKS.?All specimens of Ammothella with short ocular tubercle, short abdomen, and first chelifore segment approximately equal to the sec? ond are tentatively placed under A. rugulosa. Those without these qualifications are placed with A. appendiculata (see remarks under that spe? cies). The female from the Halimeda mat of 2 Feb 1978, has an extra eye situated below the normal left anterior eye. It is round, half the size of the one above, and is darkly pigmented, matching the four eyes above. Scattered records indicate that Ammothella ru? gulosa is distributed from Bermuda to Florida and through the Caribbean to Brazil. ECOLOGY.?This species lives among fouling organisms on ships and piers and has been taken in association with Sargassum, in addition to the habitats listed above. Ascorhynchus Sars, 1877 Ascorhynchus latipes (Cole, 1906) Ascorhynchus latipes.?Hedgpeth, 1948:256 [literature]; 1954: 427.?Fage, 1952:530.?Stock, 1953:304 [key]; 1954:116; 1975:969.?Child, 1979:15-16. MATERIAL EXAMINED.?Carrie Bow Cay: Tide line on ocean side among rocks and rubble, coll. J. Clark, 17 Jan 1976; 19, 1 juvenile. Tidal flats on ocean side, from plankton net resting on bot? tom in 0.5 m at night, coll. R. Larson, 30 Jan 1978; 19. REMARKS.?This is another species for which the range is extended to the western Caribbean. Its occurrence from Florida and the Bahamas to Bonaire, and also at Dakar, Senegal, gives it an amphi-Atlantic distribution. ECOLOGY.?This littoral species appears to have some preference for sand and rock habitats, but the collecting records are too scarce to be certain. Ascorhynchus sp. cf. serratum Hedgpeth, 1948 Ascorhynchus serratum Hedgpeth, 1948:259-260, fig. 44a-f. Ascorhynchus serralus.?Stock, 1975:969. MATERIAL EXAMINED.?Carrie Bow Cay: In coral and algae on outer vertical reef face in 27- 30 m, coll. C. A. Child, 29 Apr 1976; 1 juvenile. REMARKS.?This specimen probably represents a new species, but is sufficiently immature that a description should be postponed until an adult can be collected. It shows several similarities to Ascorhynchus serratum, in having 3-segmented che- lifores, tall ocular and median trunk tubercles, and a long curved oviger claw. The differences exhibited by this juvenile are that it (1) is without lateral process tubercles; (2) has Ammothella-\ike terminal leg segments, but without auxiliary claws; (3) has very short legs and oviger segments in comparison with A. serratum; (4) has very short first tibiae; and (5) has very long dorsodistal leg setae. Some or most of these characters may be the result of growth stage, but tubercles are usu? ally well developed by this stage. The absence of lateral process tubercles possibly places this spec? imen in a separate species. The depths at which Ascorhynchus serratum has been captured, roughly from 400 to 700 meters off Florida and in the Yucatan Channel, would remove this specimen from consideration as the same species except that we know nothing of vertical migration during growth in pycnogonids. We do know that some pycnogonids have a nar? row range of temperature tolerances, and perhaps vertical migration is limited by this factor. Cer? tainly, the temperatures from 30 meters to 700 meters cover a wide thermal spectrum. 360 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Eurycyde Schiodte, 1857 Eurycyde raphiaster Loman, 1912 Eurycyde raphiaster.?Stock, 1975:979 [literature].?Child, 1979:21, fig. 5i-j. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats in 0.5 m, coll. K. Riitzler, 4 May 1974; 16. Rock and algae from both sides of reef crest at low tide, coll. M. Carpenter, 20 Mar 1977; 19. Rocks and algae from just in front of reef crest in 0.4 m, coll. B. Kensley, 29 Jan 1978; 19. From sandy rubble with Halimeda behind outer reef ridge in 27 m, coll. C. A. Child, 7 Feb 1978; 1 juvenile. From Halimeda clump behind outer reef ridge in 27 m, coll. B. Kensley, 7 Feb 1978; 1 juvenile. REMARKS.?Females of this species usually have lateral process and first coxa tubercles that are much smaller than thos? of the male, and the lateral processes are sometimes placed closer to? gether, imparting a dimorphic appearance to the species. Otherwise, Eurycyde raphiaster is easily dis? tinguished from others of the genus in the western Atlantic. The juveniles collected from 27 meters appear to mark the deepest limit at which this species has been taken in the western Atlantic. Loman (1912:13) lists the capture depth as 91 meters for his type from the Cape Verde Islands. ECOLOGY.?Three of the above records show this species to be associated with Halimeda. It has many other habitats throughout its amphi-Atlan? tic range. It was one of the few pycnogonids found in the high energy wave action area of the reef crest, but it was also taken in the calm water of the sand trough behind the outer reef ridge. Hedgpethius Child, 1974 EMENDED DIAGNOSIS.?Ammotheidae. Ascorhyn- chus-Mke with minutely papillose body surface, without trunk or lateral process tubercles. Ante? rior trunk segment longer than combined length of posterior 3 segments; first lateral processes at extreme posterior of first segment, imparting "long necked" appearance. Proboscis with 3 an? terior-pointing tubercles arranged laterally and ventrally around its largest circumference. Scape 2-segmented, very short, chela vestigial. Palp 7- or 8-segmented, third segment with swelling. Fe? male oviger rudimentary, 3-segmented. Male ovi? ger 9-segmented, Ammothella-\ike, without dentic? ulate spines. Propodus with much reduced or missing main claw, auxiliaries large, very curved. Hedgpethius mamillatus, new species FIGURE 163a-/ MATERIAL EXAMINED.?Carrie Bow Cay: Bro? ken rock and rubble at tide line, coll. J. Clark, 17 Jan 1976; 1 ovigerous female, holotype (USNM 170997). DESCRIPTION.?First body segment 0.5 longer than combined length of posterior 3 segments. Lateral processes short, less than half trunk di? ameter, separated by greater than their own di? ameter, without setae or tubercles. Ocular tuber? cle a rounded cone as tall as neck diameter, situated at midlength of first trunk segment. Eyes lightly pigmented. Abdomen cylindrical, carried half erect" not extending beyond posterior lateral processes, armed with 2 distal setae. Proboscis large, ovoid, without marked con? strictions, with 2 dorsolateral and 1 ventral an? terior-pointing tubercles at widest diameter of proboscis and at one-third its length. Anterior to each tubercle is a slight bulge in same longitudi? nal axis. Mouth with 3 lateral and ventral slits with flat distinct lips. Scape 2-segmented, short, carried in cowling around anterior of first trunk segment. First scape segment longest, with single dorsodistal seta. Sec? ond segment minute. Chela tiny, with anterior crease and no fingers. Palp 7-segmented, thin. First 2 segments small, no longer than wide. Third segment about 5 times as long as wide, with a distinct posterior bulge. Fourth and sixth segments tiny, only slightly longer than wide. Fifth segment longest, ex? tremely slender, with 3-4 endal setae. Terminal segment thin, 4 times as long as wide, armed with NUMBER 12 361 FIGURE 163.?Hedgpethius mamillatus, new species (holotype): a, trunk; b, trunk, lateral; c, oviger; d, third leg; e, terminal segments of third leg; / palp. Hedgpethius tridentatus Child, male: g, anterior trunk segments, lateral; h, anterior of trunk, dorsal; i, femur with enlargement of femoral cement g l a n d ; / oviger; k, oviger terminal segments. 362 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES endal and distal setae longer than segment di? ameter. Oviger (female) rudimentary, tiny, of 3 seg? ments, arising ventrally just anterior to first lat? eral processes. Legs thin, femorae with eggs. Second tibia slightly longer than first, femur slightly shorter than both. Major segments with single long dor- sodistal seta, longer than segment diameter, and several short setae. Tarsus and propodus short, without spines but having a row of sole setae as long as segment diameter. Main claw lacking, auxiliaries large, strongly curved. MEASUREMENTS (mm).?Ocular segment length, 0.44; total trunk length (anterior tip to tip of fourth lateral processes), 0.74; trunk width (across second lateral processes), 0.23; proboscis length, 0.37; abdomen length, 0.1; third leg, coxa 1, 0.09, coxa 2, 0.15, coxa 3, 0.09, femur, 0.34, tibia 1, 0.37, tibia 2, 0.38, tarsus, 0.04, propodus, 0.13, auxiliary claw, 0.05. DISTRIBUTION.?Known only from the type-lo? cality, Carrie Bow Cay, Belize, littoral. ETYMOLOGY.?Named for the two breast-like lateral tubercles on the proboscis. REMARKS.?The new species differs from Hedg? pethius tridentatus in having one less palp segment, a longer abdomen, the ocular tubercle placed on the middle of the neck, and by having large auxiliary claws without a main claw, however small, between them. The male oviger of this new species remains unknown, but will probably re? semble that of H. tridentatus, with the same ex? treme sexual dimorphism displayed for that spe? cies. Hedgpethius tridentatus Child, 1974 FIGURE \63g-k. Hedgpethius tridentatus Child, 1974:494-497, fig. 1. MATERIAL EXAMINED.?Twin Cays: NW coast, mat of Caulerpa verticillata and mangrove rootlets beneath Rhizophora in intertidal, coll. C. A. Child, 31 Jan 1978; 16, 19. Another mat of the same substrate from just S of first collection, coll. C. A. Child, 2 Feb 1978; 13 specimens. DESCRIPTION OF MALE.?Proboscis tubercles slightly larger than those of female. Femoral ce? ment gland a tall tube placed at midlength of dorsal surface, perpendicular to femur. Oviger with 9 segments: second the longest; sixth ovoid; terminal 3 segments (strigilis) placed anaxially on middle of sixth segment; strigilis segments short, curved, with simple spines; all but first oviger segment armed with several setae; without ter? minal claw. EMENDED DESCRIPTION OF FEMALE.?Palp of 8 segments: first two very short; third and fifth longest; sixth and seventh very short; terminal segment thin, with ventral setae. Chelifores of 3 segments, partly hidden by anterior cowling of first trunk segment. Scape with 2 segments, first a short cylinder with dorsodistal seta, second a wrinkled bud, no longer than wide. Chela vesti? gial, with distal crease but no fingers. Juvenile chela with strongly curved fingers overlapping at tips. REMARKS.?The smallest palp and chelifore segments are extremely difficult to see and were missed in making the first description. The males of this species have the same number of segments in palps and chelifores as listed above in the emended female description. The extreme sexual dimorphism found in the ovigers of this species is very rare in pycnogonids. In most pycnogonid genera, either ovigers are lacking in the female or the ovigers are smaller slightly modified versions of the male ovigers. Enough females of this genus have now been collected to show that the very reduced ovigers reported in the original description are not mis? information based on damaged specimens, but are the natural state for females of Hedgpethius, in which they appear almost embryonic. This genus was named and published before I received the paper by Turpaeva (1973), in which she assigns a number of species of Rhopalorhynchus (Colossendeidae) to a new genus, Hedgpethia. Ac? cording to the International Code of Zoological NUMBER 12 363 Nomenclature, Article 56a, both names must be retained. The two genera are sufficiently different in most characters that they will never become synonymous. The known distribution for Hedgpethius triden? tatus is extended from Florida to the Belizean barrier reef islands; all records are littoral. Nymphopsis Has well, 1885 Nymphopsis duodorsospinosum Hilton, 1942a Nymphopsis duodorsospinosum Hilton, 1942a:303-305, pl. 45.? Hedgepeth, 1948:250-252, fig. 40; 1954:427.?Child and Hedgpeth, 1971:609 [list].?Kraeuter, 1973:496.?Stock, 1975:978.?Krapp and Kraeuter, 1976:342.?Child, 1979:21. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats in 0.5 m, coll. K. Riitzler, 7 May 1974; 19. REMARKS.?This species is easily recognized by its large size in comparison with tiny littoral species, its legs and trunk, which are crowded with dorsal tubercles, and particularly the two tall trunk tubercles, which separate it from all other pycnogonids known from the Belizean coast. It has been found from the tide line to about 60 meters of depth on various substrates, but its habits and feeding preferences remain unknown. Nymphopsis duodorsospinosa has a wide tropical and temperate distribution, from Georgia and Florida through the Caribbean to Panama and from the Gulf of California to the Pacific coast of Panama. It is also found in the Galapagos Islands. Tanystylum Miers, 1879 Tanystylum birkelandi Child, 1979 Tanystylum birkelandi Child, 1979:23, fig. 7. MATERIAL EXAMINED.?Carrie Bow Cay: Flats on ocean side, rubble and calcareous algae from tide line, coll. C. A. Child, 27 Jan 1978; 29. REMARKS.?These two females agree exactly with the female paratype from Galeta Island, on the Caribbean side of Panama. Since this is only the second capture record for the species, little can be said concerning its habitats, except that it has been taken with coralline and calcareous algae in the littoral. Its distribution is here ex? tended north from Panama to the Belizean coast. Tanystylum tubirostrum Stock, 1954 Tanystylum tubuostre Stock, 1954:117-120, figs. 24-25.?Bour- dillon, 1955:600, pl. 3: figs. 2-4.?Child, 1979:34-35. Tanystylum tubirostrum.?Stock, 1975:984. MATERIAL EXAMINED.?Twin Cays: NW coast, on roots of Rhizophora mangle with adherent red and green sponges, algae, ascidians and hydroids in 0.5 m, coll. C. A. Child, 31 Jan 1978; 26, 19. REMARKS.?These specimens agree in all re? spects with Stock's (1954) description and figures. The present capture extends the known distribu? tion of this species from Bermuda, Puerto Rico, Curasao and Bonaire, the western Caribbean, and the Pacific shores of Mexico and Panama. ECOLOGY.?There are few capture records for this species, but these are fairly well documented. It lives in littoral habitats and has been found associated with Sargassum, hydroids, and, in this case, with a wealth of potential food. Pycnogonids are known to feed on hydroids and ascidians and because they have been taken in association with sponges, it is assumed that they also feed on the soft parts of sponges. F a m i l y CALLIPALLENIDAE Callipallene Flynn, 1929 Callipallene belizae, new species FIGURE 164. MATERIAL EXAMINED.?Carrie Bow Cay: Hali? meda and rubble from outer reef ridge in 18 m, coll. B. Kensley, 7 Feb 1978; 16 with eggs, holo? type (USNM 171035), 2 juvenile paratypes (USNM 171036). Large clump oi Halimeda from same area, coll. C. A. Child, 7 Feb 1978; 19 364 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 164.?Callipallene belizae, new species (holotype): a, trunk; b, trunk, lateral; c, proboscis, ventral; d, chela with enlargement of tip of movable finger; e, third l e g ; / terminal segments of third leg; g, oviger with single egg; h, terminal segments of oviger with enlargement of two spines. paratype (USNM 171037). DESCRIPTION.?First 2 trunk segment lines pre? sent, third lacking. Neck short, unadorned. Lat? eral processes short, only as long as their diame? ters, separated by less than their diameters, gla? brous. Ocular tubercle a broad truncated cone capped with a thin tubercle as tall as ocular cone. Eyes large, lightly pigmented. Abdomen an in- NUMBER 12 365 flated cylinder slightly longer than twice its max? imum diameter, armed with 2 distal setae. Proboscis short, inflated distally and with rounded ventrodistal bulges. Chelifore scape equal in length to proboscis, armed with lateral and dorsodistal setae. Chela with slightly inflated palm, armed with 7-8 dorsal and 1 ventral setae. Fingers overlap at tips, with? out setae. Movable finger with single tiny distal tooth, broader than long. First 3 segments of oviger moderately short, fourth longer, fifth more than twice length of fourth, armed with several setae longer than fifth segment diameter. Distal apophysis of male fifth segment armed with 2 setae. Terminal 4 segments (strigilis) armed with denticulate spines in the formula 6 :5 :5 :5 . Denticulate spines dimorphic: proximal spines oval with fine denticulations; distal spine larger, with fan-like projection of denticulations pointing distally on segment. Legs long, slender, very setose distally. First and third coxae short, little longer than their diameters. Second coxa long, curved, over 5 times longer than its maximum diameter. Femur slightly longer than tibia 1, tibia 2 slightly longer than femur. Cement glands not found. Tarsus short, armed with 2 smooth ventral spines, 1 ventral and 1 dorsal setae. Propodus short, only slightly curved, armed distally and laterally with many long setae. Sole with 4 smooth heel spines and 7 distal spines. Main claw half as long as propodus, auxiliaries about 0.9 times length of main claw, without endal setae or teeth. MEASUREMENTS (mm).?Trunk length (cheli? fore insertion to tip of fourth lateral processes), 0.89; trunk width (across second lateral proc? esses), 0.37; abdomen length, 0.12; proboscis length, 0.28; third leg, coxa 1, 0.12, coxa 2, 0.65, coxa 3, 0.19, femur, 0.96, tibia 1, 0.77, tibia 2, 1.1, tarsus, 0.06, propodus, 0.28, claw, 0.14. DISTRIBUTION.?Known only from the type-lo? cality, Carrie Bow Cay, Belize, in depths of 18 meters. ETYMOLOGY.?Named for the country where it was discovered, Belize. REMARKS.?This new species resembles Amer? ican specimens of Callipallene brevirostris (John? ston), but has a shorter neck; it is also similar to Caribbean specimens of C. emaciata. It differs from both of these species in the following respects: the new species has a tall thin tubercle on top of its truncated ocular cone; it has smooth tarsus and propodus spines whereas the other two have var? iously crenulated spines; its chela fingers are with? out teeth except for a single tiny distal tooth on the movable finger; the auxiliary claws are longer than for most American specimens, although those of C. brevirostris are sometimes as long; and the posterior trunk segmentation line is lacking, although this is not a reliable taxonomic charac? ter with this genus. Taxonomic distinction among Callipallene spe? cies is often very difficult to make, not only because several species are very similar, but also because apparently they have more than one adult molt stage with resulting changes in setae, teeth, denticulate spine number and shape, and mensural characters and their ratios. These molt changes add vastly to the difficulty in deciding where variation ends and species begin, a dividing line that will undoubtedly remain uncertain for most Callipallene species until large numbers can be compared interspecifically and intraspecifi- cally. The propodus of Callipallene belizae is shorter, less curved, and has many more setae than either C brevirostris or C. emaciata. Neither of these species has the very marked dimorphism shown in the denticulate oviger spines of C. belizae. The termi? nal spine on each of the four distal segments is splayed out and canted forward so as to resemble a fan. ECOLOGY.?Since this species has been found only in rubble and Halimeda at depths of 18 meters, and the bottom samples from which the species was taken were not preserved for possible food preference organisms, its habits are un? known. Callipallene emaciata (Dohrn, 1881) Callipallene emaciata emaciata.?Stock, 1952a:8 [literature]. Callipallene emaciata.?Stock, 1975:1011.?Child, 1979:41-42. 366 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES MATERIAL EXAMINED.?Carrie Bow Cay: From Thalassia bed in 0.5 m, coll. R. Larson, 31 Mar 1977; 16 with eggs, 16, 29, 2 juveniles. From Dictyota on reef flat in 0.5 m, coll. R. Larson, 16 Mar 1977; 1 juvenile. Rubble and calcareous algae at wall on ocean side at tide line, coll. C. A. Child, 28 Jan 1978; 19. Syringodium and sediment from lagoon in 1.5 m, coll. B. Kensley, 1 Feb 1978; 19. Plankton sampler on bottom on ocean side flats in 0.5 m, coll. R. Larson, 30 Jan 1978; 1 larva (probably this species). South Water Cay: Piling scrapings from dock at S end on lagoon side, coll. C. A. Child, 30 Jan 1978; 1 juvenile. REMARKS.?These are the first records of Cal? lipallene emaciata from the western Caribbean. This species has been found in the Mediterranean, Portugal, the Azores, and from Florida and the Caribbean archipelago to the Guianas, primarily in littoral and shallow depths. ECOLOGY.?This species and two others from the Caribbean coasts, Callipallene brevirostris and C. phantoma (Dohrn) are generally found associ? ated with fouling organisms on bridge and dock pilings. Parapallene Carpenter, 1892 Parapallene bermudensis Lebour, 1949 FIGURE 165. Parapallene bermudensis Lebour, 1949:930-932, figs. 2-3. MATERIAL EXAMINED.?Carrie Bow Cay: Coral, sand and rubble from sand trough behind outer reef ridge at 27 m, coll. C. A. Child, 6 Feb 1978; 19 subadult. Sand and rubble from slightly S of above sample in 27 m, coll. C. A. Child, 7 Feb 1978; 16 juvenile. REMARKS.?Although neither of these speci? mens is fully mature, they agree with Lebour's (1949) description and represent only the second record of this species; the type-locality is Ber? muda. An unreported female from about the same depth in the Bahama Islands is in the collections of the National Museum of Natural History. Since Lebour's (1949) figures are somewhat stylized and several details are omitted from her drawings and description, I have prepared a set of figures of the male for clarification. Lebour stated that the chela fingers lack denticulations, but both the Bahama specimen and the above two have very small denticulations or teeth on one or both fingers of the chela. The species has raised lips and an oral fringe of tiny setae. The propodus has two major heel spines unlike the figure and description of Le? bour's (1949:931-932, figs. 3-7) specimen. The ventrodistal second tibia spine, major tarsal spine, and the sole spines all show a slight serration on their inner surfaces. The auxiliary claws are slightly shorter than those figured by Lebour. The ovigers of the Carrie Bow female are shorter than those of the type-specimen, but the female was about to molt and the next (adult) stage oviger can be seen clearly within the outer chitinous layer. The oviger of the Carrie Bow male is still the unsegmented curled appendage of a juvenile. A distinctive character of this species is its spination. There are short broad spines on the first coxae and a series of these spines around the insertion of the chelifores and proboscis. The spines have an annulated hollow interior without the annulations being carried through to the outer surface (Figure 165/). Elsewhere on the animal, where other typical spines and setae are present, these annulated spines are absent. The Carrie Bow specimens represent a range extension to the western Caribbean for this spe? cies. ECOLOGY.?The four available records for this species indicate that it has a very restricted depth preference. All four specimens reportedly have been taken in depths of 27 to 33 meters (90 to 100 ft). Lebour's (1949) type was taken "with fragments of hydroids and weeds." The Carrie Bow specimens were taken from rubble with some algae and probably a number of hydroids, al? though none of the latter were saved. No habitat data are available on the Bahamas specimen. The Carrie Bow specimens were light brownish green. Pigmentation and feeding preference have NUMBER 12 367 FIGURE 165.?Parapallene bermudensis Lebour, juvenile male: a, trunk; b, anterior trunk segments, lateral; c, third leg; d, terminal segments of third leg with enlargement of spine; e, chela with enlargement of movable finger; / annulated spine. not been correlated, but most literature references concerning live pycnogonids state that generally they are colored the same as the substrate making them difficult to see. Their color probably repre? sents some of the algal substrate on which many pycnogonids find their food source. Pigrogromitus Caiman, 1927 Pigrogromitus timsanus Caiman, 1927 Pigrogromitus timsanus Caiman, 1927:408-410, fig. 104a-f.? Hedgpeth, 1947:7 [text]; 1948:214-216, fig. 23.?Stock, 1968a:46; 1975:1015-1016.?Lipkin and Safriel, 1971: 9.?Arnaud, 1972:159-160.?Child, 1979:46,47. Clotenopsa prima Hilton, 1942b:52-53, fig. 8. MATERIAL EXAMINED.?Carrie Bow Cay: Among rocks and rubble on ocean side at tide line, coll. J. Clark, 17 Jan 1976; 2 juveniles. Twin Cays: Among Rhizophora roots along N edge of dividing channel at S end of Cays in 0.6 m, coll. C. A. Child, 25 Apr 1976; 16 with eggs. REMARKS.?This species has a pantropical dis? tribution. This is the second record of capture in the western and southwestern Caribbean, the first being from the Caribbean coast of Panama. 368 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES ECOLOGY.?The many and varied habitats of this species include corals, algae, zoanthids, man? groves, rubble, and molluscan cavities. It was also found among unspecified fouling organisms on pilings, almost always at intertidal and subtidal depths. Family PHOXICHILIDIIDAE Anoplodactylus Wilson, 1878 Anoplodactylus bahamensis Child, 1977 Anoplodactylus bahamensis Child, 1977:587-589, fig. 2. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats in 0.5 m, coll. K. Riitzler, 2 May 1974; 16. REMARKS.?This species was originally de? scribed (Child, 1977:587) as having teeth only on the movable fingers of the chelae. The male collected here has 3 tiny low teeth on the immov? able finger also, but it is otherwise indistinguish? able from the type, also a male. This Carrie Bow male extends the known dis? tribution to the western Caribbean. The habitats or associations of this species re? main unknown. It has been found at depths of 0.5 and 12 meters. Anoplodactylus batangensis (Heifer, 1938) Pycnosoma batangense Heifer, 1938:174-176, fig. 6a-c. Anoplodactylus batangensis.?Stock, 1968a:54 [literature]; 1975: 1082-1083, fig. 43c-d.?Arnaud, 1973:957, figs. 3-4.? Child, 1979:50. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats at tide line, coll. K. Riitzler, 23 Apr 1974; 16 with eggs, 19. Tidal flats at 0.5 m, coll. K. Riitzler, 2 May 1974; 16. From Dictyota on reef flat in 0.5 m, coll. R. Larson, 16 Mar 1977; 16 juvenile. From Thalassia bed in 0.5 m, coll. R. Larson, 31 Mar 1977; 1 juvenile. Rubble and Halimeda from outer reef crest in 0.5 m, coll. B. Kensley, 30 Jan 1978; 16 with eggs. Mixed carpet of red algae and compact corallines at SE end shore, coll. B. Kensley, 5 Apr 1978; 16. South Water Cay: Thalassia and red sponge beyond dock at S end in 1 m, coll. B. Kensley, 30 Jan 1978; 39. Twin Cays: NW coast, mat of Caulerpa verticil? lata and mangrove rootlets from under Rhizophora, intertidal, coll. C. A. Child, 31 Jan 1978; 16, several juveniles. Another C. verticillata and rootlet mat from nearby, coll. C. A. Child, 2 Feb 1978; 36. A mat of Halimeda from nearby, coll. B. Kensley, 2 Feb 1978; 19. REMARKS.?This species is easily recognized by its anteriorly curved and tapered proboscis, unique among the many species of this genus. Several live and even freshly killed specimens, particularly the 36 found on Caulerpa, 2 Feb 1978, had a broad chalk-white stripe running from the posterior of the ocular tubercle to the base of the abdomen. The remainder of these animals ranged from cream to slightly straw colored, except for a chalk-white band around each distal leg segment suture. This white color was not associated with the intestinal diverticula, which could be seen below and separate from the color line. ECOLOGY.?Stock (1975:1083) mentioned that males seem to be rare, but both in this collection and in another from Panama, males appear in equal or greater numbers than females. Of the 14 adults in this collection, there are 9 males and 5 females; 2 of the males bear eggs. This imbalance may be related to collecting methods. Pycnogon? ids are generally found incidentally by sorters who are looking for other organisms and therefore many are missed. The habitats of this species are quite varied, and include algae, Thalassia, sponges, rubble, and Rhizophora. All specimens were taken in littoral depths and most records in the literature are also from shallow water. Anoplodactylus evelinae Marcus, 1940 Anoplodactylus evelinae Marcus, 1940:55-58, pl. 4.?Hedgpeth, 1948:232, fig. 31.?Child, 1979: 53. Anoplodactylus (Labidodactylus) evelinae.?Stock, 1954:128; 1975:1083. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats at tide line, coll. K. Riitzler, 23 Apr 1974; NUMBER 12 369 16. Thalassia bed in 0.5 m, coll. R. Larson, 31 Mar 1977; 16 with eggs. Rubble and calcareous algae on ocean side in 0.3 m, coll. C. A. Child, 27 Jan 1978; 19. Halimeda, Caulerpa, and rubble from outer reef crest in 0.5 m, coll. B. Kensley, 30 Jan 1978; 19. Clump of Halimeda at shore, SE end, coll. B. Kensley, 5 Feb 1978; 19. Twin Cays: NW coast, Rhizophora roots with algae, hydroids, ascidians, and bryozoans in 0.5 m, coll. C. A. Child, 31 Jan 1978; 2 juveniles. NW coast, mat of Caulerpa verticillata and man? grove rootlets under Rhizophora, intertidal, coll. C. A. Child, 2 Feb 1978; 46, 29. Mat of Halimeda from same area, coll. B. Kensley, 2 Feb 1978; 19. REMARKS.?This is another easily recognized species. It is very "stumpy" in appearance, with a short and broad ocular tubercle, abdomen, and proboscis. The legs are robust and the very pointed heel with its short spine makes a reliable recognition character. It is common in the western Caribbean. The above collections produced 14 specimens from the littoral. ECOLOGY.?The habits of Anoplodactylus evelinae are unknown, but it is found together with other species of pycnogonids in algae, Thalassia, Rhizo? phora, rubble, and among sessile animals. The Belizean collections do not support Stock's (1975: 1083) suggestion that this species is a sand bur- rower. Anoplodactylus imswe, new species FIGURE 166 MATERIAL EXAMINED.?Tobacco Reef: Reef top about 500 m N of South Water Cay, interti? dal, coll. M. Carpenter and R. Larson, 23 Mar 1977; 16 holotype (USNM 171122). Carrie Bow Cay: Tidal flats in 0.5 m, coll. K. Riitzler, 4 May 1974; 19 paratype (USNM 171123). Tidal flats in 0.5 m, coll. K. Riitzler, 7 May 1974; 19 paratype (USNM 171124). DESCRIPTION.?Trunk with first two interseg? mental lines well marked, third incomplete, marked only by slight depression dorsally. Lateral processes separated by slightly less than their diameters, each as long as trunk diameter, with? out tubercles, armed with 1-3 dorsodistal setae. Ocular tubercle a moderately tall cylinder twice as tall as its diameter, capped by triangular cone. Eyes distally on cylinder, darkly pigmented. Neck armed with single seta lateral to, and in front of, ocular tubercle. Abdomen almost 3 times its di? ameter, bent erect, armed with several distal se? tae. Chelifores thin, scape armed with several distal setae. Chela ovoid, fingers shorter than palm, well curved at tips, without teeth. Palm with several distal setae; movable finger with 3 ectal setae. Oviger with robust curved basal segment. Sec? ond segment only slightly longer, cylindrical, armed with ectal row of 6-7 short setae. Third segment longest, slightly less than half again as long as second segment, armed with several ectal and endal setae. Terminal 3 segments each shorter than last, moderately setose distally with distal setae longer than segment diameter. Ter? minal segment thin, pointed at tip. Leg with first coxae armed with 2-4 dorsodistal setae, without tubercles. Second coxae of third and fourth legs with ventrodistal genital spur almost as long as segment diameter, carrying genital orifice at tip. Coxae of first and second legs without spurs or orifices. Third coxae armed with several distal setae. Femur the longest leg segment with first tibia longer than second, each armed with several short setae and a single long dorsodistal seta, as long or longer than segment diameter. Seta of femur mounted on short tuber? cle. Femoral cement gland a long, thin, flask- shaped tube, canted distally, situated at less than one-third length of femur. Tarsus roughly trian? gular, with several ventral setae. Propodus stout; heel perpendicular, armed with single large spine and 5 smaller spines. Sole armed with 6-7 curved spines and several lateral setae; distal lamina one- fifth sole length. Claw robust, strongly curved. Auxiliary claws lacking. MEASUREMENTS (mm).?Trunk length (cheli- fore insertion to tip of fourth lateral processes), 0.93; trunk width (across first lateral processes), 0.63; proboscis length, 0.52; abdomen length, 370 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES FIGURE 166?Anoplodactylus imswe, new species (holotype): a, trunk; b, trunk, lateral; c, proboscis, ventral; d, chela; e, third leg with enlargement of femoral cement g l a n d ; / terminal segments of third leg; g, oviger terminal segments. NUMBER 12 371 0.24; third leg, coxa 1, 0.21, coxa 2, 0.42, coxa 3, 0.2, femur, 0.55, tibia 1, 0.52, tibia 2, 0.47, tarsus, 0.09, propodus, 0.34, claw, 0.23. DISTRIBUTION.?Known only from the type-lo? cality, the Belizean barrier reef at Tobacco Reef and Carrie Bow Cay at a depth of 0.5 meters. ETYMOLOGY.?This species is named for the Smithsonian Institution Investigations of Marine Shallow-Water Ecosystems (IMSWE) Project. REMARKS.?The femoral cement glands of this species are not unique among species of Anoplo? dactylus. They are almost exactly like the glands of A. batangensis, but they are placed slightly farther back toward the proximal end of the segment in the new species. Other differences in A. imswe remove it from consideration as A. batan? gensis. Most notably, the proboscis in the new species is cylindrical whereas in A. batangensis it is constricted distally to a thin tube. The new spe? cies also has longer first oviger segments, longer and less setose leg segments, longer ocular tuber? cle and abdomen, and lacks low lateral process tubercles. The propodus, however, with its heel and heel spine, sole spines, and lack of auxiliaries, is strikingly similar in both species. This species is similar to Anoplodactylus erectus Cole in the shape of the trunk, ocular tubercle, oviger, and abdomen, but the propodus charac? ters and the cement glands are very different. It is also similar to A. maritimus. In comparison with Giltay's male type specimen of A. parvus {= A. maritimus, fide Stock, 1975:1069-1074), the ovi? gers, chelifores, trunk, and propodus shape of A. imswe are very close to those of the type of A. parvus. The differences are (1) the propodus of A. parvus has two heel spines and auxiliary claws, (2) the abdomen and ocular tubercle are shorter, (3) there are no discernible trunk segmentation lines, and (4) the cement glands are tiny short tubes where they emerge from the femorae. ECOLOGY.?The three specimens from Belize, although collected in separate locations, all share a distinct color pattern. When freshly killed, the Tobacco Reef specimen had a bright purple gut. After storage in alcohol for a year or more, all three specimens have deep blue intestinal tracts. This blue pigment appears to be a reliable rec? ognition character for this species, at least on the Belizean barrier reef. The purple or blue gut color is shared by Anoplodactylus lentus, a much larger pycnogonid which has not yet been found in Belize. The coloration of A. imswe indicates that it might feed on organisms containing blue-green algae, but since collecting data for this species do not contain information on habitat associates, its feeding preference remains unknown. Anoplodactylus jonesi Child, 1974 Anoplodactylus jonesi Child, 1974:497-500, fig 2; 1979:56, fig. 19a-b. Anoplodactylus (?) antillianus Stock, 1975:1081-1082, fig. 57. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats in 0.5 m, coll. K. Riitzler, 2 May 1974; 26. REMARKS.?Both of these specimens are normal males with full ovigers and a tall tubular cement gland on each femur. Several questionable fe? males have been found in the past that bear male ovigers, ova in the legs, and that lack cement glands. This seemingly abnormal situation has led to speculation on the status of this species in a genus in which the females are entirely without ovigers. Further discussion and figures of this species appear in Child (1979:56, fig. 19a-b). ECOLOGY.?Habits and habitats are lacking for the Belizean specimen, but in Panama the species has been found most often among algae. Anoplodactylus maritimus Hodgson, 1914 Anoplodactylus maritimus Hodgson, 1914:164; 1915:148; 1927: 357.?Marcus, 1940:60.?Stock, 1975:1069-1074, fig. 54. Not Anoplodactylus maritimus.?Hedgpeth, 1948:230, fig. 29d-e [= A. mleus Stock, 1975:1069]. Anoplodactylus parvus Giltay, 1934:1-3, figs. 1-5.?Hedgpeth, 1948:223-224, fig. 27e-f.?Stock, 1951:13, figs. 14-16; 1954:127; 1955:235; 1957:85; 1975:1069-1074, fig. 54.? Bourdillon, 1955:590-591, fig. 1, pl. 1: fig. 2.?Fage and Stock, 1966:326.?Kraeuter, 1973:494-495. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats, coll. M. L. Jones, 6 Apr 1976; 19. South Water Cay: Rubble from edge of chan- 372 SMITHSONIAN CONTRIBUTIONS TO T H E MARINE SCIENCES nel in 6 m, coll. C. A. Child, 1 Feb 1978; 26 with eggs, 2 juveniles. Twin Cays: NW coast, Halimeda mat from under mangroves in 0.3 m, coll. B. Kensley, 2 Feb 1978; 19. REMARKS.?In agreement with Stock (1975: 1072), I have combined the two species under Hodgson's (1914) earlier designation. There is ample cause for confusion here as Hodgson's type was never figured. Giltay's (1934) Anoplodactylus parvus has been figured by several authors, includ? ing Giltay, but was never compared with Hodg? son's type (which appears to have been lost). As presently designated, the species can also be easily confused with A. petiolatus (Kroyer), which itself is sufficiently variable to suggest the possibility of two species (Stock, 1975:1075). Stock (1975, figs. 53, 54) attempted to sort out and figure the differences between the present species and A. petiolatus. Association with floating substrates has given this species an extremely wide distribution in the Atlantic. It has been found from Chesapeake Bay in Virginia, south to Bermuda and through the West Indies to Brazil. Hodgson (1914) originally described it from south of the Azores and it is known also from the Cape Verde Islands. ECOLOGY.?This species is common over vast areas of the mid-Atlantic as one of the many inhabitants of the seaweed Sargassum (Bourdillon, 1955, fig 1). The type oi Anoplodactylus maritimus was collected in Sargassum, and many other rec? ords confirm this association. I have also seen specimens feeding on hydroids attached to Sar? gassum dipped from Florida waters. The present collection, however, shows that this species occu? pies other habitats as well. Anoplodactylus monotrema Stock, 1979 Anoplodactylus robustus.?Child and Hedgpeth, 1971:612-613 [literature].?Stock, 1975:1080. Anoplodactylus monotrema Stock, 1979:15-18, figs. 4-5.?Child, 1979:56, 58, fig. 19c. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats in 0-1 m, coll. 7 Apr 1973; 16 with eggs. REMARKS.?This species has a "fat" appear? ance with a short blunt proboscis, abdomen, and ocular tubercle. All specimens previously assigned to Anoplodac? tylus robustus, whether from European collections or from western hemisphere collections, were thought to agree. It recently became evident that the European specimens have more than one (usually 3) cement gland per femur whereas the American specimens have only one gland per femur. Other slight differences, when taken in combination, caused American specimens to be recognized as a separate species, A. monotrema. Therefore, the known distribution of the Ameri? can species is confined to the east and west coasts of southern North America and northern South America, including the Galapagos Islands. ECOLOGY.?The color of this littoral and sub? littoral species is usually straw or lighter. Its feeding habits are not yet identified. It has been taken on a wide variety of substrates. Anoplodactylus multiclavus Child, 1977 Anoplodactylus multiclavus Child, 1977:593-596, fig. 4; 1979: 58, fig. 19d. MATERIAL EXAMINED.?South Water Cay: SE side, in shallow grass bed with sand patches near Rhizophora stumps in 0.5 m, coll. J. Clark, 15 Jan 1976; 16 with eggs. REMARKS.?This specimen agrees with the type, also a male, in all respects including the multiple cement glands, except that it appears slightly more robust. ECOLOGY.?Both the type and this specimen have been found in association with or near man? groves. Anoplodactylus pectinus Hedgpeth, 1948 Anoplodactyluspectinus Hedgpeth, 1948:234-236, fig. 34; 1954: 427.?Stock, 1955:235; 1974:17.?Arnaud, 1973:955- 957.?Child, 1974:500; 1979;58. Anoplodactylus pectims [sic].?Stock, 1975:1050-1052, fig 41a. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats at tide line, coll. K Riitzler, 23 Apr 1974; 16. From Dictyota on reef flat in 0.5 m, coll. R. Larson, 16 Mar 1977; 16 with eggs, 16. Rubble NUMBER 12 373 from sand trough behind outer reef ridge in 27 m, coll. C. A. Child, 6 Feb 1978; 36, 29. Rubble from just S of 6 Feb collection, in 27 m, coll. C. A. Child, 7 Feb 1978; 36. Twin Cays: NW coast, mat of Caulerpa verticil? lata and mangrove rootlets beneath Rhizophora, intertidal, coll. C. A. Child, 2 Feb 1978; 19. Mat oi Halimeda from nearby, intertidal, coll. B. Ken? sley, 2 Feb 1978; 19. REMARKS.?This species, including its females, is easily recognized by the pectinate major heel spine that can be seen under high magnification. This marks the second time Anoplodactylus pec? tinus has been collected in the western Caribbean. Previously, it has been collected on the Caribbean coast of Panama, in Florida, the Caribbean Lee? ward Islands, and in Madagascar. ECOLOGY.?Freshly killed specimens had light green intestinal diverticula. This color may be a character useable to distinguish fresh specimens of this species, at least on the Belizean barrier reef. Anoplodactylus portus shares this green color, but retains coloration even after long storage, unlike A. pectinus. The color is undoubtedly a function of diet, although the particular green food matter is unknown. Anoplodactylus portus Caiman, 1927 Anoplodactylus portus.?Stock, 1975:1052-1053, fig. 41b-e [lit? erature].?Child, 1978:133-144, figs. 1-4; 1979:58-59. MATERIAL EXAMINED.?Carrie Bow Cay: Rub? ble and Halimeda from reef crest in 0.5 m, coll. C. A. Child, 29 Jan 1978; 16 with eggs. Lagoon flats, with Syringodium and sediment in 1.2 m, coll. B. Kensley, 1 Feb 1978; 16 juvenile, 1 larva. Mixed algae and compact corallines at shore, coll. B. Kensley, 5 Feb 1978; 19 juvenile. Halimeda and rubble from outer reef ridge in 18 m, coll. B. Kensley, 7 Feb 1978; 19 juvenile. South Water Cay: Rubble from edge of chan? nel in 6 m, coll. C. A. Child, 1 Feb 1978; 16, 29. Twin Cays: NW coast, mat of Caulerpa verticil? lata and mangrove rootlets under Rhizophora, intertidal, coll. C. A. Child, 2 Feb 1978; 19. REMARKS.?This is a pantropical species with a robust appearance which aids in distinguishing it from the many other tropical species of this genus. As with all species of Anoplodactylus, the configuration of the male femoral cement gland(s) is a key character. In conjunction with oviger, chela, and propodus characters, it facili? tates recognition among these otherwise difficult species. This observation does not hold for fe? males, which remain difficult if not impossible to identify if they are not taken in the same sample or area with recognizable males. ECOLOGY.?This species is often collected with gut diverticula in the legs showing green chloro- plasta or at least chlorophyllous coloration in? gested from the algal habitat in which it is often found. Whether or not it picks this up directly as food or as a biproduct of eating algal-grazing fauna has never been demonstrated. Anoplodactylus sp. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats, coll. K. Riitzler, 4 May 1974; 19 juvenile. Lagoon surface among Thalassia and Sargassum, coll. R. Larson with plankton net, 25 Jan 1978; 1 larva. Tidal flat, plankton net resting on bottom in 0.5 m, coll. R. Larson, 30 Jan 1978; 1 larva. With Syringodium in 1.0-1.5 m, coll. B. Kensley, 3 Feb 1978; 29 juveniles. REMARKS.?None of these specimens is mature enough to identify. Family ENDEIDAE Endeis Philippi, 1843 Endeis spinosa (Montagu, 1808) Endeis spinosa.?Hedgpeth, 1948:238-240 [early litera? ture].?Stock, 1952b: 185-186; 1954:128; 1957:85; 1962: 218; 1968a:59 [key]; 1968b:32, fig. 25.?Soyer, 1966:3.? deHaro , 1966:9; 1967:109, 112-113, fig. 5.?Krapp, 1973: 72.?Child, 1979:66. MATERIAL EXAMINED.?Carrie Bow Cay: Thal? assia beds in lagoon, 1.5 m, coll. J. D. Ferraris, 11 May 1975; 16 with eggs. Stann Creek: Surface plankton tow off Pelican 374 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES Beach, 0.5 mi out, coll. R. Larson, 13 Apr 1978; 19. REMARKS.?This species has been found from Norway to Argentina, including the Mediterra? nean Sea, at mainly littoral and sublittoral depths. ECOLOGY.?Endeis spinosa is a frequent inhab? itant of floating Sargassum in the Caribbean and elsewhere. It is also common in Thalassia grass that is supporting colonies of hydroids, bryozoans, and ascidians. Family NYMPHONIDAE Nymphon Fabricius, 1794 Nymphon floridanum Hedgpeth, 1948 Nymphon flondanum Hedgpeth, 1948:196-199, fig. 17 [long- necked form only].?Stock, 1955:215, fig. la [long-necked form only]; 1975:994-998, figs. 14-15.?Kraeuter, 1973: 494.?Krapp and Kraeuter, 1976:336-337.?Child, 1979: 37. MATERIAL EXAMINED.?Carrie Bow Cay: Tidal flats at low water, coll. K. Riitzler, 23 Apr 1974; 16. Tidal flats, coll. M. L. Jones, 5 Apr 1976; 19. Halimeda and rubble from outer reef ridge in 18 m, coll. B. Kensley, 7 Feb 1978; 1 juvenile. REMARKS.?These specimens are all the long- necked form of N. floridanum (sensu strictu) as defined by Stock (1975:994-998). In this exten? sive genus, this species (along with its look-alike, N. aemulum) is one of the few tropical members that is both littoral and sublittoral. It is known from Georgia through the Caribbean to as far south as French Guiana. It has been found in association with coral, algae, and sandy habitats. Family RHYNCHOTHORACIDAE Rhynchothorax Costa, 1861 Rhynchothorax architectus Child, 1979 Rhynchothorax architectus Child, 1979:68-72, figs. 23, 24a-g, 25a-e. MATERIAL EXAMINED.?Carrie Bow Cay: Bro? ken rock and rubble at tide line on ocean side, coll. J. Clark, 17 Jan 1976; 16. REMARKS.?This is the third capture locality for this variable species, the first two being the Caribbean and Pacific sides of the Isthmus of Panama. This specimen agrees well with male paratype specimens from Panama. It has low middorsal tubercles instead of the taller ones of the holotype. The species is known only from the intertidal. ECOLOGY.?This species probably came from the coral sand around or under rock rubble. The most common habitats of this genus, as reported in the literature, are interstitial, with the animals living between sand grains around or under rocks. The above capture probably substantiates this mode of living and suggests that coralline and coral sand are the primary substrates. Rhynchothorax crenatus, new species FIGURE 167 MATERIAL EXAMINED.?Carrie Bow Cay: Coral sand and rubble in sand trough behind outer reef ridge in 27 m, coll. C. A. Child, 6 Feb 1978; 1 subadult specimen, holotype (USNM 170996). DESCRIPTION.?Trunk compact, first 2 segment lines complete, third lacking. First 3 trunk seg? ments with tall conical median dorsal tubercles almost as tall as ocular tubercle. Surface of entire animal except proboscis with minute scattered papillae. Dorsal and ventral trunk surfaces with pattern of lightly pigmented molt sutures similar to reticulations. Ocular tubercle tall, pointing about 30 degrees anteriorly, with 2 tiny lateral tubercles flanking a conical cap, a posterior tri? angular median tubercle and 2 thin short tuber? cles lateral and posterior to triangular tubercle. Conical cap with single posterior seta. Eyes large with medium dark pigment. Abdomen extending to tip of second coxae of fourth legs, cylindrical, and with proximal and distal constrictions, armed with 3 dorsodistal setae. Proboscis cylindrical-conical, with 2 dorsolat? eral bulges flanking a tall spike-like tubercle in median line halfway along proboscis. Mouth with 2 laterally flattened antimeres confining it to a vertical slit. Palps 4-segmented, arising from rather long NUMBER 12 375 FIGURE 167?Rhynchothorax crenatus, new species (holotype): a, trunk; b, trunk, lateral; c, palp; d, terminal segments of palp, dorsally, with enlargement of endal spine; e, third leg. trunk tubercle bases lateral to and closely set against proboscis. First segment very short, half as long as its diameter. Second segment longest, with small triangular dorsodistal tubercle armed with 2-3 setae and tiny ventrodistal tubercle armed with single seta. Third segment slightly less than half as long as second, with narrow median dorsal tubercle armed with seta, and small ventrodistal bulge armed with 2-3 setae. Terminal segment round in lateral aspect, armed with many setae dorsally, distally, and ventrally. Third palp segment armed on endal surface with single large spine bearing denticulations on pos? terior surface only. Spine slightly longer than diameter of segment. Chelifores entirely lacking. Oviger incomplete (?) or vestigial (?) in holo? type, only small unsegmented bud present. Legs moderately thin, armed with a few short setae and a very long single seta on first and second tibiae. Femorae lacking long setae. Femur the longest segment; first and second tibiae each shorter than preceeding segment. Major leg seg? ments of anterior 4 legs slightly longer than pos? terior four. Tarsus with single ventral spine and seta. Propodus without heel, moderately curved, armed with 5-6 sole spines. Terminal segments of anterior 4 legs armed with 4-6 tarsus setae and up to 12 sole spines. Claw robust, less than half propodus length. Auxiliary claws lacking. DISTRIBUTION.?Known only from the type-lo? cality, Carrie Bow Cay, Belize, in a depth of 27 meters. ETYMOLOGY.?From Latin, meaning notched, 376 SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES in reference to the appearance created by the median trunk, ocular, and proboscis tubercles in lateral view. MEASUREMENTS (mm).?Trunk length (tip of ocular tubercle to tip of fourth lateral processes), 0.65; trunk width (across first lateral processes), 0.42; abdomen length, 0.14; proboscis length, 0.35; third leg, coxa 1, 0.1, coxa 2, 0.08, coxa 3, 0.07, femur, 0.23, tibia 1, 0.2, tibia 2, 0.17, tarsus, 0.05, propodus, 0.21, claw, 0.09. REMARKS.?This subadult specimen (sex pores not evident) has a unique set of characters al? though it superficially resembles Rhynchothorax un? icornis Fage and Stock from the Cape Verde Is? lands. The trunk of R. crenatus is thinner, with more widely spaced lateral processes. None of the appendage tubercles are nearly as long or thin, the ocular tubercle is very different with its cone and posterior tubercles, and the median trunk tubercles are simple cones instead of complex tuberculate and papillose tubercles. On the other hand, there are several striking similarities between the two species. Both share a single mid-dorsal proboscis tubercle, both have similar propodus configuration except for the strong distal sole spine of Rhynchothorax unicornis, and both have similar palps although the tall tubercle of the second segment and the tiny ter? minal segment are not present on R. crenatus. Both species lack auxiliary claws and have lateral pro? cess and coxa 1 tubercles in the same places although their sizes are very different. Finally, both species have small lateral tubercles placed behind the ocular tubercle and rudimentary ovi? gers (signifying that both are females?) of a but? ton-like appearance. The two species are similar enough perhaps to form a geminate pair, separated by the Atlantic Ocean and Caribbean Sea. Probably, each is a valid species, but the discovery of a specimen intermediate between the two, if such exists, would invalidate any suggestion that these ex? amples represent speciation by geographic isola? tion due to plate tectonics. Although the male ovigers in Rhynchothorax species are much alike, it might be worthwhile to study the character of male ovigers in these two species when more specimens are collected. Thompson (1909:535) first proposed a separate family for this genus and I concur that its com? bination of characters fit none of the currently accepted genera with which it has been placed. Conclusions The pycnogonid fauna of Carrie Bow Cay and its vicinity on the Belizean barrier reef is ex? tremely rich. Extensive collecting during this study has produced approximately 300 specimens from mainly littoral habitats. Thirty-one identi? fied species in 14 genera are represented, includ? ing four new species. The habitat diversity on the reefs makes it probable that collecting in nearby unsampled areas will further increase this num? ber. For instance, two samples taken from a single mangrove habitat on different days contained no less than 10 species of pycnogonids, and five small samples of rubble and algae from depths of 18 and 27 meters on the outer reef yielded nine species. Algae appear to be a preferred habitat but seldom is an alga associated with just one species of pycnogonid. Sargassum is regularly colonized by Anoplodactylus maritimus, Endeis spinosa, Tanystylum tubirostrum, Numphopsis duodorsospinosum, and occa? sionally other species. All these species, however, are also found in other habitats, such as rubble, other algae, sea grass, coral, and mangrove roots. Without an analysis of gut content there is no proof that pycnogonids ingest algae. More likely, they eat the soft parts of animals living on the algae or on rubble having associated algal growth. Specimens of Achelia sawayai, for example, were captured on rubble with algae supporting sessile organisms and were not found on "clean" algae collected at the same time. Pycnogonids with green intestinal diverticula {Anoplodactylus pectinus and A. portus) possibly form the second step in the food chain by ingesting the tissues of animals that feed primarily on green algae. Pycnogonids move very slowly so that it is almost impossible for them to feed on mobile fauna. A stand of foliose NUMBER 12 377 TABLE 27.?Geographical distribution of Belizean pycnogonid species Species Achelia sawayai Ammothella appendiculata A. exornata A. marcusi A. rugulosa Ascorhynchus latipes A. serratus Eurycyde raphiaster Hedgpethius mamillatus H. tridentatus Nymphopsis duodorsospinosa Tanystylum birkelandi T. tubirostrum Callipallene belizae C emaciata Parapallene bermudensis Pigrogromitus timsanus Anoplodactylus bahamensis A. batangensis A. evelinae A. imswe A. jonesi A. maritimus A. monotrema A. multiclavus A. pectinus A. portus Endeis spinosa Nymphon floridanum Rhynchothorax architectus R. crenatus P an tr o p ic al X X X u i s 1 CU rri M id dl e A m er ia X X X X X X -o G al ap ag os Is la n X X c ed