CN de - ye re Fee aa tig hn AI ke ie foe ed OO ee ere ea aed ef ke De PP AR PLN 2 ee ELA OES Feta ot Patel eel" ey Pk Sage oats ypenele ea on ee MUS. COMP. ZOOL LIBRARY JUN 1 4 106% HARVARD LINIIV EE RSITY Postilla PEABODY MUSEUM OF NATURAL HISTORY YALE UNIVERSITY NEW HAVEN, CONNECTICUT, U.S.A. Number 108 9 June 1967 ENVIRONMENTAL FACTORS INFLUENCING THE DISTRIBUTION OF BARBATIA DOMINGENSIS (MOLLUSCA, BIVALVIA) ON THE BERMUDA PLATFORM SARA S. BRETSKY DEPARTMENT OF GEOLOGY, YALE UNIVERSITY ABSTRACT Eight localities, representing four generalized environments, were sampled to determine the environmental relations of the small, byssally attached arcid bivalve Barbatia (Acar) domingen- sis on the Bermuda Platform. The species was most abundant in turbulent environments and least abundant in the protected bays and sounds. It usually attaches to the underside of corals; occa- sionally it is found attached to rocks. The principal limiting factor on the distribution of the species appears to be the availability of suitable coral substrata for attachment. i) Postilla Yale Peabody Museum No. 108 INTRODUCTION Except for a few commercially important groups such as the pearl oysters, scallops, shipworms, and edible oysters, almost nothing is known of the factors which control the distribution pat- terns of marine bivalve molluscs. This is particularly true for the abundant and diverse Superfamily Arcacea, a widely distributed group which is most common in tropical and warm temperate seas. As a contribution toward understanding the environmental relations of this important superfamily, the small arcid bivalve Barbatia (Acar) domingensis (Lamarck) — often erroneously called Barbatia reticulata (Gmelin) — was intensively studied on the Bermuda Platform for three weeks in August, 1965. This species is found in the West Indies, Bermuda, and along the east coast of the United States as far north as Cape Hatteras. It is one of seven species of arcids, representing the genera Arca, Anadara, Barbatia, and Arcopsis, known from Bermuda (Dall, 1889; Warmke and Abbott, 1961; Neumann, 1965). B. domin- gensis, like most other arcids, has been little studied except for reports of distribution based mostly on empty shells from dredge samples. Preliminary observations showed that B. domingensis nor- mally lives attached to the underside of corals and rocks in shallow water. To further analyze the occurrences, corals were collected by shallow diving at eight localities (Fig. 1) and examined in the laboratory, where the attached Barbatia were counted and their attachment sites noted. In collecting the corals, no effort was made to determine the presence or absence of Barbatia; instead, at each locality a sample of corals was obtained which represented the overall coral diversity and abundance at that place. Other hard bottom materials, such as rock fragments and bivalve shells, were also examined to discover whether they might serve as attachment sites for Barbatia. This sampling produced 221 Bar- batia specimens: 215 from 95 corals and the other 6 removed directly from rocks. Some individuals were kept alive in the lab- oratory for several days and their behavior noted. Sediment samples were taken at some Barbatia collecting localities in order to determine the relative abundance of Barbatia valves in the sediment. 1967 Distribution of Barbatia domingensis 3 a ] NORTH e ROCK wa Ke ABET Dayo Lyk Yi ag) ies y Lae EXPLANATION LN GIA TRAV LEX leey Ue \D <22 GENERALIZED ee & ponh? ; DISTRIBUTION OF SHOALS y EW ry oY THREN G5 fA LOCALITIES Nee ‘ 1 THREE HILL SHOALS C® 5 &y WH iy ‘ 2 NORTH ROCK & y eo 3 SOUTHWEST OF NORTH ROCK V7 ‘ ae 4 LEDGE FLATS; WEST REEF WHALEBONE , 5 SOUTH SHORE BOILERS Bc) DP 6 HARRINGTON SOUND an Ley's ie 7 BAILEY'S BAY . BAV@) ye 8 OUTSIDE WHALEBONE BAY — EPR = L Aboped ‘ a HARRINGTON’ {I ENVIRONMENTS ~ CSOUNDS 5 = ore, [7A NORTH SHOALS x SOUTH SHOALS FE SHALLOW SUBLITTORAL (SHELTERED | SHALLOW SUBLITTORAL (OPEN OCEAN) BERMUDA ISLANDS 3 4 miles N f | | = | Figure |. Map of Bermuda Islands showing sampling Iocalities and generalized environments. ANATOMY AND BEHAVIOR The anatomy of Barbatia domingensis has been studied in detail by Heath (1941). No additional anatomical studies were made, but observations of general behavior patterns were made to supplement Heath’s descriptions of preserved specimens. The average shell length is about 10 mm, with an observed range between 3 and 25 mm. The ligament is long and opistho- detic. The byssus is strong. If the clam is dislodged from its attach- ment site, the remnants of the old byssus are expelled from the foot and a new byssus is produced. The process requires about 30 minutes in the smaller, more active, and presumably younger 4 Postilla Yale Peabody Museum No. 108 animals. In the larger Barbatia, byssus re-formation may require 6 to 24 hours and sometimes does not occur at all. Lacking the accustomed coral substratum, the animals may attach to the sides of the aquarium or to one another’s shells. In nature, Barbatia domingensis is found attached by its byssus to the undersurfaces of corals and, less commonly, of rocks (Fig. 2). The species has no adaptive specialization for boring into hard materials, but shows a definite preference for nestling 1 crevices in corals and in the burrows formed by other organisms, such as the sponge Cliona and the mytilid bivalve Lithophaga. Many specimens, especially the larger ones, are excellently camou- flaged by encrustations of serpulid worm tubes and the pink foraminifer Homotrema rubrum (Lamarck). The inhalant current appears to enter ventrally and somewhat anteriorly, and the exhalant current is posterior and dorsal (Fig. 2). 0 5 1 1.5 SSS ee cm Al) y SS ( SS RSLS CORAL G\\\y, NY Ly WSK 2S Zod E SB GY %neE BAY EQ] S| S ap, 2 We 222s = = ms Wn N\//, ZI Se = MKE = Me WW ZN Ay = Zi, = Sel 1B Y Wee Se E SS (MF BZ W/Z S )) UTNE \' NV; €< Se AS wn? S ZF Wy, = Bw; Se Mie, PAD Lil} a & 8 Wipf Z : 4 G : = iy & Aye / o, WINS INHALANT : CURRENT = SS 1 -——— la | EXHALANT ee: sl IL CURRENT ‘CALCAREOUS j | ROCK — | 1 ae as FiGURE 2. Vertical section showing living position of Barbatia domingensis, mode of attachment, and direction of inhalant and exhalant currents. 1967 Distribution of Barbatia domingensis 3) ECOLOGY OF SAMPLING LOCALITIES The sampling localities (Fig. 1) represent four major environ- ments: the North Shoals (localities 1, 2, 3, 4); the South Shoals (locality 5); the sheltered shallow sublittoral (localities 6 and 7); and the open-ocean shallow sublittoral (locality 8). Table 1 sum- marizes the major features of these environments. SHOALS The coral communities of the North and South Shoals show probably the greatest faunal diversity of any environment on the Bermuda Platform. The epifauna of the shoals is dominated by scleractinian, alcyonarian, and hydrozoan corals (Table 1). A number of small epifaunal animals, including sponges, foramini- fers, tunicates, bryozoans, and bivalves, are attached to the bases of these corals. Barbatia domingensis was the most common bivalve found on coral bottoms; it was sometimes associated with the pectinid Lima lima (Linné). The North Shoals and South Shoals are faunally quite similar. The main difference between them is that, though both may be characterized by notable turbulence in appropriate weather, wave action is particularly spectacular at the South Shore Boilers (locality 5). Table 2 indicates the considerable variation in abundance of Barbatia among the four localities of the most extensively collected environment, the North Shoals. A total of 62 Barbatia were col- lected from 68 corals, the number of Barbatia on a single coral ranging from 0 to 7. Barbatia was actually living on only 34% of the corals examined. Barbatia was strikingly more common at locality 5 on the South Shoals than anywhere on the North Shoals, with 15 corals harboring an average of 2.4 Barbatia each and Barbatia living on 80% of the corals collected (Table 2). Up to 10 bivalves were found on one coral. No preference for any coral species or kind of coral was apparent. The corals collected from the shoals ranged in diameter from 2 to 27 cm; the largest corals usually had the most bivalves, but many large corals had none. 6 Postilla Yale Peabody Museum No. 108 SHALLOW SUBLITTORAL The shallow sublittoral collecting localities (6, 7, and 8) have carbonate-sand bottoms littered with rock debris from the shoreline cliffs. Harrington Sound (locality 6) has only a narrow connection to the open ocean; hence wave activity is subdued. Corals are sparse and usually small, and some of the abundant shoal corals are lacking or rare (Table 1). Large bivalves (10- 15 cm long) are quite common and form a major element of the epifauna. My observations and the little comparative information available (Nielsen, n. d.) indicate that the diversity of both epifaunal and infaunal bivalves is much greater in Harrington Sound than on the shoals. The epifaunal bivalves listed in Table 1, except for Barbatia, were also identified by Neumann (1965), in his comprehensive study of carbonate sedimentation in Harring- ton Sound. Most of these bivalves are byssally attached to shells, corals, or rocks in shallow water, thus occupying about the same ecological niche as Barbatia. Six corals were collected from Harrington Sound, near Abbott’s Cliff on the northwest side of the Sound. Three Barbatia individ- uals were found on one of these. A number of rocks and bivalve shells were also examined, but Barbatia was not found on any. Bailey’s Bay (locality 7) is partly cut off from the open ocean by a chain of small islands, and thus is nearly as sheltered as Harrington Sound. These two localities are therefore grouped as the sheltered shallow sublittoral environment. Small corals occur sparingly here; those examined contained no Barbatia. Twenty- four arcids were collected from under rocks. Only six of these, however, were Barbatia domingensis; the other 18 were the closely related Arcopsis adamsi (E. A. Smith), a slightly smaller arcid with a triangular ligament between the umbones. The shallow sublittoral area just outside Whalebone Bay (locality 8) was grouped with the two previous localities in Table 1 because of the similarities in water depth and type of bottom, but in some respects it is more similar to the shoal local- ities. It is exposed to the open ocean and thus to intensive wave action. In size and diversity of corals (Table 1), it is intermediate between the shoals and the sheltered environment. The only bivalves found here were Barbatia and Spondylus. All the speci- 1967 Distribution of Barbatia domingensis a mens of B. domingensis collected here, a total of 112, were found on a single large (39 cm diameter) coral of the genus Diploria. Such a coral is unusually large for the shallow sublittoral environ- ment of Bermuda, though its size would not be unusual for a shoal coral. It was collected near the beginning of my study, and it was expected at first that similar densities of concentration of Barbatia would occur on the larger corals at other localities. Comparable aggregations were found, however, on only two corals, both from the South Shoals; these, with diameters of 14 and 15 cm (area therefore about 1/9 that of the Whalebone Bay coral) had, respectively, 9 and 10 Barbatia. All other corals in the study on which Barbatia was found were considerably less densely pop- ulated with Barbatia. Possible ecological explanations for the presence of dense aggregations of bivalves will be discussed in the section “Factors Affecting Distribution.” COMPARISON OF BARBATIA AND ARCOPSIS DISTRIBUTION The apparent contrast in distribution between Barbatia domin- gensis and another small arcid bivalve, Arcopsis adamsi, which was found with B. domingensis at locality 7 (Bailey’s Bay), strongly indicates that, while the two species are not mutually exclusive, B. domingensis is specialized for life on corals and A. adamsi for rocks. Arcopsis was never found living on corals. Though both were only minor contributors to the sediment samples examined, Barbatia always made up 1-2% of the coarse fraction (>4 mm), whereas only three or four disarticulated Arcopsis valves were found in shoal sediments. Possibly the preference of Barbatia for nestling in burrows of other organisms precludes its living on talus fragments, which may be less likely than corals to be burrowed. Another suggestive observation is that the byssus of Arcopsis is smaller and less strong than that of Barbatia. Pos- sibly both nestling and the stronger byssus are adaptations of Barbatia to a more turbulent enironment; Arcopsis may require sheltered areas like Bailey's Bay. Aquarium observations, how- ever, indicated that Arcopsis is more mobile than Barbatia, and the byssus of Arcopsis is usually re-formed in an hour or less— faster than in all but the smallest Barbatia. Perhaps Barbatia requires an anchored attachment site, whereas Arcopsis may be able to live on a less stable sandy substratum because it can move No. 108 Postilla Yale Peabody Museum (aie) vidojdiq DIAJSDJUO JW DIDIADS Y DIAJSDJUO JW DIAJSDAIPIG SPO DIAISDAIPIG SaJOod DIAJSDAIPIS Salo pisojdiq pyjAydos] (jueulwop) vi40jdiq SURIUTJIBIIIIS SUPIUTJIBII[IS SURIUTIIBIIIIS s[eloy suonip SuON}Ip -UOd JOyJeOM -uod 194)RoM uo Juapusadap uo JUuapuddap ‘ysry Arey MOT MOT AIDA, ysiy Apuajysisuoyg $= ‘ysIYy Apuey souginqin ], SMO s1ydeisodo} ul Ssuneynuns -OB JUSIWIPIS a}BVUOgIvO w0}}0q SJUSWIS BAT S[BI09 AQ pdalIdUSA sajzlUP jo adh} Sn[B} YIM WOO puRsS d}ReUO0GIR yD -1]09 9U9D0}SI9[q Pas1UIGnsS *‘JUSWIPas Ul Z O} YIeW spy MOT Uva UW / 0} YIPUW Opt) MOT URITy yidap 19]e8 (8 “90T) (L °907T) (9 ‘90T) (¢ *90T) (y-T *90T) Avg Aeg punos auOgaTePU AA, s Agpieg UO}SULLIeH sjReoyg “OS sfeoys “ON NVd9O NadO dauaL THs IVAOLLITHNS MOTIVHS STVOHS qHLYNVaVd TVOIDOTO04 SLNHNNOUIANA AO AAUVNNOAS elds i\yll Distribution of Barbatia domingensis 1967 WS (p-¢ > snjkpuodsy DIIDGAV piodalpiy uvozoipAH 8 90°] wo ¢[-¢ :suRLIeUOADTY wuld (Q)[-$ > wid ¢7-¢ > Wd LT-T :SUBIUTIBII[IS DNUOUP ppvjould DIAISO snjkpuody TL) Speed SPHDAW sisdooip DIA DULIT DIIDGADE DIIDGAD DIIDGID panvxa)dopnasd DANDXA] d DIUOSAOL) suvIIvUOAD[Y pioda] iW pioda] pi uvozoIpAH, uvozo.IpAH EOOm tual Sab hoa | (aseq jo ‘WIeIp “xeuw) s[P1O09 payday] [Oo JO ddURI 9ZIS SOATRAIG jeuneyidy (ponunuos) | ATaVL No. 108 Postilla Yale Peabody Museum 10 SYIOI WOIF TV xx ISUDpD SISdOIAP QT Sd x (O01) 0 LI 08 6 LS LY OG at pipgivg YM s[e10o Jo % (ZIT) oe im) VT 90 0'7 it 1 Oo ria (Bae) [e109 /DIDGung ele a) € 9€ 07 ia IZ bas puvgivg “ON I ¢ 9 cI ZE L 61 (Oana ohne s[BI0D “ON 8 907 L‘90T 9 “90T ¢ 00T b 90T € ‘00T ZT ‘90T [ ‘90T NVdIOQ NadQ = davai tani —— —— LAO LAdOS MOVs SalVOHS HMOs STVOHS HLYON SLNANNOWIANT VGNWNYA NI PILV Edd JO AONVAUNNAV Cc ATV 1967 Distribution of Barbatia domingensis 11 about and reattach itself if necessary. The adaptive significance of the different types of ligaments—long and straight in Barbatia, short and triangular in Arcopsis—is not obvious, but a larger and perhaps stronger ligament may also be of more value to a bivalve living in turbulent waters. The most important control on the distribution of the two species, however, appears to be the distri- bution of the preferred attachment sites. FACTORS AFFECTING DISTRIBUTION Some of the important factors to be considered in accounting for the distributional pattern of any marine bivalve are salinity, temperature, depth, nutrients, competitors, larval ecology, and substratum. All the localities in this study were in areas of normal marine salinity, about 36% (Neumann, 1965). Since the study encom- passed only the last three weeks in August, and was limited to a small geographic area and a narrow depth range, temperature effects could not be assessed. Within the accessible range, no depth control on the distribution of Barbatia was evident. No precise information on nutrient distribution was available. Wave turbulence and current activity may be important both in the distribution of nutrients and in larval ecology. In this con- nection it may be noted that the area of greatest turbulence, the South Shoals locality, apparently exhibits the most consistent con- centration of Barbatia; and the occurrence of 112 bivalves on one coral at Whalebone Bay was also in an area of considerable wave activity. To determine whether size of coral (maximum diameter of base as an indication of surface area available for attachment) influences the number of Barbatia living there, data from the North Shoals, South Shoals, and Harrington Sound were combined. The corals collected at these localities ranged in size from 2 to 27 cm. Barbatia was never found on corals smaller than 9 cm. The Pearson product-moment correlation coefficient between size of coral and number of Barbatia per coral was 0.307 (N=89), which is a significant correlation at the 0.1% level but far from a perfect linear relationship. In the shoal environment, both inter- and intraspecific com- petition must be important in the densely populated “coral- 12 Postilla Yale Peabody Museum No. 108 bottom community.” The intensity of competition for living space and food may to some extent explain why the larger corals tend to support more bivalves. But in many cases Barbatia is absent from corals of large diameter. Here predation may have been particularly intense. In some large corals, burrows and crevices were lacking, so that surface area for attachment may actually have been relatively small. In Harrington Sound, competition from other epifaunal bivalves may have combined with the reduc- tion in number of corals to diminish the numerical importance of Barbatia. Direct competition with Barbatia appears more likely in the case of Arcopsis than in the larger bivalves, and the dif- ferent substratum preferences of the two small arcids may have evolved as a means of reducing competition between them. The patchy distribution may also be a function of the larval ecology of Barbatia. Unfortunately, there is no specific informa- tion on the larval development of Barbatia, and little is known about that of other West Indian arcids. Nothing is known, for example, about the length of the pelagic life of Barbatia. Marine pelagic larvae may spend from a few hours to several months in the water mass before settling (Johnson, 1964). Thorson (1957) noted that 85-90% of all tropical marine species have a long (3 weeks or more) pelagic larval life. On the other hand, Purchon (n. d.) estimated the pelagic lifetime of the Malayan arcid Anadara granosda as One to two weeks. Loosanoff and Davis (1963) found that spat of Anadara transversa, a species from the Atlantic coast of the United States, set in the laboratory 27 to 37 days after fertilization. Arctic species of arcids, however, may have a very brief or nonexistent pelagic stage (Ockelmann, 1958). A short pelagic larval life might result in a patchy distribution with high concentrations near the areas where spawning took place; a long larval life would permit more thorough coverage of the accessible environments but might be cancelled by increased exposure to predation, distribution patterns of currents, and other factors. It has been established (Thorson, 1957) that larvae of many marine invertebrates are able to select suitable substrata and to postpone metamorphosis until they find a substratum fitted to the requirements of the adults. Korringa (1940) described the pref- erence of oyster larvae for the lower surfaces of objects; such selectivity may account for the occurrence of Barbatia on the bottom of corals and rocks. 1967 Distribution of Barbatia domingensis t3 In the latter stages of this study, I observed a young form of either Barbatia or Arcopsis from sediment from Bailey’s Bay which displayed the prominent, extensible foot and a structure resembling the valvular membrane described by Carriker (1961) for the plantigrade (latest larval) stage of Mercenaria mercenaria. This individual, about a millimeter long, was extremely active for several hours, eventually attaching itself to the side of the aquar- ium near the air-water interface. It appears likely, therefore, that the young of these arcids are able to choose where they will settle, and may be mobile after settling. In some bivalves, particularly sessile epifaunal forms, the larvae are apparently attracted to concentrations of adults of their own species (Thorson, 1957; Verwey, 1952). Possibly gre- gariousness, whether or not it operates in conjunction with brief larval life, may partly account for localized concentrations of Barbatia such as that on the Whalebone Bay Diploria and to a lesser extent on some of the shoal corals. The rarity of Barbatia in Harrington Sound is likely to be explained partly by the lack of many large corals for attachment, and partly by the sheltered and nearly enclosed situation of the Sound. Either Barbatia larvae are only rarely brought into Harring- ton Sound, or wave and current activity there is so limited that the young Barbatia are not widely distributed throughout the Sound and are therefore unable to chance upon the attachment sites that do exist. ACKNOWLEDGMENTS This study was carried out as part of a seminar on “Problems of organism-sediment interrelationships’, sponsored by National Science Foundation G. B. 3066 and given at the Bermuda Bio- logical Station during the summer of 1965. I thank Dr. W. H. Sutcliffe, Director of the Bermuda Biological Station, for provid- ing its facilities, and Dr. R. F. Schmalz of Pennsylvania State University, who supervised the seminar. Dr. A. Lee McAlester of Yale University very kindly read the manuscript and offered many helpful suggestions. This paper is Contribution Number 410 from the Bermuda Biological Station. Specimens of Barbatia domingensis and Arcop- sis adamsi collected in the study are permanently deposited in the collections of the Peabody Museum, Yale University. 14 Postilla Yale Peabody Museum No. 108 REFERENCES CITED Carriker, M. R., 1961. Interrelation of functional morphology, behavior, and autecology in early stages of the bivalve Mercenaria mercenaria: J. Elisha Mitchell Sci. Soc., v. 77, p. 168-241. Dall, W. H., 1889. A preliminary catalogue of the shell-bearing marine mollusks and brachiopods of the southeastern coast of the United States, with illustrations of many of the species: Bull. U. S. Nat. Mus., Ven Sia De l-22 Te Heath, H., 1941. The anatomy of the pelecypod family Arcidae: Trans. Am. Phil. Soc., v. 31, p. 287-319. Johnson, R. G., 1964. The community approach to paleoecology, in Imbrie, J. and Newell, N. D. (eds.), Approaches to paleoecology: New York, Interscience, p. 107-134. Korringa, P., 1940, Experiments and observations on swarming, pelagic life and setting in the European flat oyster, Ostrea edulis L.: Arch. INEcr ZOols Va). pe) 1-249. Loosanoff, V. L., and Davis, H. C., 1963, Rearing of bivalve mollusks: Adv. Mar. Biol., v. 1, p. 1-136. Neumann, A. C., 1965. Processes of Recent carbonate sedimentation in Harrington Sound, Bermuda: Bull. Mar. Sci., v. 15, p. 987-1035. Nielsen, T. R. A. [no date] Bermuda sea shells: Hamilton, Bermuda; Bermuda Mid-Ocean News, Ltd., 25 p., 47 figs. Ockelmann, W. K., 1958. The zoology of East Greenland — Marine Lamellibranchiata: Meddr. Grgnland, y. 122, pt. 4, p. 1-256. Purchon, R. D. [no date; seen only as undated reprint] The biology of “Krang”, the Malayan edible cockle: Proc. Sci. Soc. Malaya, v. 2. p. 61-68. Thorson, G., 1957. Bottom communities: Mem. Geol. Soc. Am., v. 67, pt. 1, p. 467-534. Verwey, J., 1952. On the ecology of distribution of cockle and mussel in the Dutch Waddensea: Arch. Neérl. Zool., v. 10, p. 171-239. Warmke, G. L., and Abbott, R. T., 1961. Caribbean seashells: Narberth, Penna.; Livingston, 348 p., 44 pls. a heh ON eaera ee eS ae 2 Pm Pete Sea te: = pe Ee Rete mE stn 9 - THEA Serer oY Sade cerepey renee prep enee S apne site n pans ova neteasighe ” = ~ 2 aX - ~~ . , = -. - : . . 5 = 2 > = = as