Pacific Science (198 1), vol. 35, no. 1 © 1981 by The University Press of Hawaii. All rights reserved

A cropora in Hawaii. Part 1. History of the Scientific Record, Systematics, and Ecology!

RICHARDW. G RIGG,2 JOHN W. WELLS,3 and CARDEN WALLACE4

ABSTRACT: Present occurrence of the coral genus Acropora in Hawaii has long been questioned. This paper reviews the scientific literature concerning this controversy and presents the results of a recent resource survey of the entire Hawaiian Archipelago tha t clearly estab lishes the presence of three species of Acropora in Hawaii. These species are Acropora cytherea, A . valida, and A. humilis. Taxonomic descriptions for each species are presented, along with notes on their worldwide geographic distribut ions. In Hawaii, the three species are found only on six islands in the middle of the chain . Extension of their ranges throughou t the archipelago may be limited by discontinuous and sporadic larval recruit ment.

THE HAWAIIAN ARCHIPELAGO consists of ap­ results of the United States Exploring Ex­ proximately 132 high volcanic islands, rocky pedition in 1840 and 1841, the "Albatross islets, atolls , reefs, banks, shoals , seamo unt s, Expedition" in 1902, and the Ta nager Expedi­ and guyots (Armstrong 1973). The island tion in 1923. In 1976, a five-year research chain stretches northwest from the Island of program on wildlife and fisheries manage­ Hawaii diagonally across the Pacific to Kure ment in the Northwestern Hawaiian Islands Atoll, a distance of 2450 km. Beyond Kure was planned by the National Ma rine Fisheries Atoll , a series ofdrowned atoll s (guyots) and Service, the U.S. Fish and Wildlife Service, seamounts extend the chain to the Emperor the Hawaii State Division of Fish and Game, Seamounts. The Emperor Seamounts con­ and the University of Hawaii Sea Grant tinue northward all the way to the juncture Program. between the Kuril and Aleutian trenches. In 1978 and 1979, in connection with the In spite of the vast geography of the five-year research plan , the species compo­ Hawaiian Archipelago, knowledge of the sition of coral reefs off the islands of Nihoa, marine biota of the Hawaiian Islands was, Necker, French Frigate Shoa ls, Gardner until abo ut 1976, based largely on research Pinnacles, Maro Reef, Laysan, Lisianski, done in the major high islands (Hawaii to Pearl and Hermes, Midway, and Kure was Niihau) . Prior to this time, knowledge of surveyed. This inventory confirmed and doc­ marine flora and fauna of the Northwestern umented the presence of the coral genus Hawaiian Islands consisted primarily of the A cropora in Hawaii. Three species were re­ corded. Unti l this discovery, a controversy over the occurrence of Acropora in Hawaii persisted for many years. In part, this con­ I Hawaii Institute of Marine Biology contribution troversy was due to the presence in Hawaii number 608. The Nation al Sea Grant Program and the of Acropora in the geological record from Hawaii State Office of the Marine Affairs Coordinator provided funding for this research. Manuscript accepted the Miocene followed by apparent disap­ 12 January 1981. pearance during the Pleistocene. Thus, the 2 University of Hawaii, Hawaii Institute of Marine purpose of Part I of this paper is to sum­ Biology, P.O. Box 1346, Kaneohe, Hawaii 96744. marize the scientific record of the genus in 3 Cornell University, Department of Geological Sciences, Ithaca, New York 14850. Hawaii, substantiate the taxonomic position 4 James Cook University of North Queensland , of the three species discovered, and describe Queensland, Australia 4811. their present patterns of distribution and 2 PACIFIC SCIENCE, Volume 35,January 1981 abundance. In Part 2 of this paper, zoogeo­ SCIENTIFIC RECORD graphic aspects of Acropora are considered. The protean reef coral genus Acropora, with some 325 nominal species, is one of the most widely distributed of living reef cora ls METHODS in the Indo-Pacific, but its current presence Stations were selected off all major islands in Ha waiian waters has long been doubted. or reefs in the Hawaiian Archip elago. At Its absence from early collections of recent each station, the species composition and Ha waiian corals was notable. J. D. Dana, in community structure ofmacrobenthic species the first (1872) and subsequent editions of ( » 2 em in the greatest dimension) were his Corals and Coral Islands, suggested that quantitatively analyzed by conducting 25-m the "Hawaiian Islands . . . are outside of the or 50-m line transects at depths between 5 Torrid Zone . .., and the corals are con­ and 15m (Grigg and DoIlar 1980). Qualitative sequently less luxuriant and much fewer in estimates of abundance of the macrobenthos species. There are no Madrepores [A cropora] covering much larger areas (about 5000 m-) and but few of the Austraea and Fungia were also coIlected. The number of stations tribes, while there is a profusion of coral of selected per island ranged between two and the hardier genera, Porites and Pocillopora." eight, depending proportionately on island Earlier, in 1846, in his great report on the size and diversity of habitats. Where present, zoophytes of the U. S. Exploring Expedi­ fragments from mature (largest) colonies of tion, Dan a pointed out that "at the Sand­ Acropora were coIlected and preserved in wich Island s, which are near the northern formalin. The reproductive condition of limits of the coral seas, Porites and these colonies was determined by directly Pocillopora prevail. . .. " Quelch, in his mon­ examining the mesentaries of decalcified ograph of the Challenger reef corals (1886), specimens (G rigg and Boucher, ms) and by observed that "it is a noteworthy peculiarity standard histological techniques. AIl sam­ of the coral faun a of the Sandwich Island s ples used to determine reproductive condi­ that no representative of the widely dis­ tion were collected in the months of Jun e, tributed genus M adrepora [A cropora] is September, and November. At one station at found on the reefs." But Brook , in his 1893 French Frigate Shoals, a core was obtained monograph of the genus, listed the Sand wich from a water depth of 10m to ascertain the Islands as one of the several localities of presence of Acropora in the recent past . Madrepora echinata Dana, although he listed Material from the core was dated using only three specimens of the species in the radiocarbon dating. British Mu seum collections, none of them The taxonomic position ofthe three species from Hawaii . of Acropora was determined from colonies For nearly a century there have been on coIlected in 1978 and 1979 at French Frigate exhibit in the vitrines of the Galerie de shoals. The Hawaiian specimens were com­ Zoologie, Museum Nationale d'Histoire pared with a large number ofAcropora corals NatureIIe, Paris, two specimens of Acropora collected from the Great Barrier Reef. (The labeIled "Madrepora longicyathus M. E. & latter were collected by John Veron and H. lies Sand wich. M. Ballieu 1874," on the Carden Wallace for a revision of the family shell of a large Pinctada and " M adrepora Acroporidae.) Type material of A . variabilis durvillei M. E. & H. lies Sandwich. M. Klun zinger was received on loan from the Ballieu 1875." These were "discovered" in Museum fiir Naturkunde, Humboldt Uni­ 1978 by Denni s M. Devaney, the curator of versity, Berlin. Radial corallites are illus­ invertebrates at the Bishop Museum in trated by scanning electron micrograph s Honolulu, Hawaii. To an inquiry concerning taken at James Cook University, Australia. their origin, J. P. Chevalier replied that he The terminology used in this paper is defined could find no record of their acquisition and by Wallace (1978). that M. BaIlieu was unknown to the Museum. Acropora in Hawaii. Part I -GRIGG, WELLS, AND WALLACE 3

He also pointed out in correspondence "a is the case with the several [actually oneI species of curious thing: besides this sample, there is Acropora recorded from the Hawaiian Island s by Brook. another example of Acropora longicyathus Their occurrence there certainly needs confirmation, for in the large authentic collections of corals I have studied which is fixed also on a shell of oyster from those islands no Acropora has occurred. (Pinctada) but it comes from Tahiti (coil. Clone, no. I, 1871)." Like Studer's M. Studer's specimen was figured again by echinata, also attached to a pearl oyster Vaughan in his monograph on the Ha waiian shell, it is probable that M. Ballieu's corals corals (1907) with no comment on its prove­ were either purchased in Hawaii or were nance, save that "this species has been re­ obtained in Tahiti but were mistakenly attri­ ported from the Hawaiian Islands by Brook buted to Hawaii. and Studer. I have seen no specimens of it In 1875, when the Challenger stopped in from there." Later (1910) he apparently ac­ Honolulu, Moseley noted (as Verrill did later) cepted it: "A cropora is possibly but not that corals not from Hawaiian waters were probably, except for A. echinata, entirely offered for sale: "There is a large shop of absent from Hawaii ." Notwithstanding the Chinese and Japanese curiosities, and two very dubious specimen of A . echinata, it has photographers' shops, where corals , imported sincebeen accepted that the genus isnot found mostly from the Marquesas, and spurious in the Hawaiian coral 'fauna, the nearest imitations of native implements manufac­ occurrence being at Johnston Island, 720 km tured for sale, are disposed of, at exorbitant to the southwest, where there are at least five prices to passengers from the mail steamer" species. Dana's 1872 statement has also been (Moseley 1879:456). accepted as sound, although as late as 1943 The first published evidence of the pres­ Vaughan and Wells pointed out that a ence of Acropora in Hawaii came in 1901 number of tropical Pacific coral genera were when Studer described some corals in the lacking in Hawaii, and that " Acropora is collections of the Berne Museum, including doubtfully represented by a single species." some collected in 1896 and 1897 by H. H. In 1936, a sizable colony of Acropora, Schauinsland at Laysan. Among the Ha­ probably A . cytherea (Dana), was collected waiian species were several not previously at French Frigate Shoals northwest of the recorded and still not known to occur in the main island s. About that time, in islands: Madrepora (Acropora) echinata (non Washington, D.C., Paul Galtsoff told John Pallas), Fungia (Fungia) fungites (Linn.), Wells ofthe find but said that its whereabouts Fungia (Ctenactis) echinata (Pallas), Ha­ were unknown. In 1946, C. H. Edmondson lomitra tiara (= H. pileus [Linn.]) and published a photograph of this specimen Merulina regalis? Dana. However, there is in his Reef and Shor e Fauna of Hawaii. no certainty that all of these actually came In reply to an inquiry in 1947 about the from Hawaii. Studer noted that at least two specimen, he wrote that it was not in the had been purchased in Honolulu and thought Bishop Museum, and that "the Hawaiian that they had probably come from elsewhere boys who collected it did not inform me of in the Pacific. His figured specimen of A. its disposition." Was it broken up and sold for echinata, a small colony on a large pearl souvenirs? A search at the National Museum shell, if it did come from Hawaii, is possibly of Natural History failed to find any record. this species but is different from all three Nevertheless, this lost coral was clear ev­ species discussed in the present paper. A idence of the presence of living Acropora in year later Verrill (1902) said that A . echinata the Hawaiian Islands, and indicated that was "probably imported," and that he had further search at French Frigate Shoals would provide a final answer. never seen an authentic species of Acropora from Hawaii. Nearl y 40 years later, an A cropora colony Local collectors assert that the genus does not occur there. But great quantities of corals, etc., are broug ht was found living at a depth of 40 ft off Poipu from the Polynesian Island s to Honolulu by missionary on the south coast of Kauai. R. A. Kinzie vessels and sold there as curiosities . .. and probably this III sent it to John Wells, who tentatively 4 PACIFIC SCIENCE, Volume 35, January 1981 identified it as A . paniculata Verrill, known Acropora cy therea (Dana, 1846) from one or two localities in the western M adrepora cytherea Dana , 1846 :441, Pacific. It was later figured (Ma ragos 1977) pI. 32, fig. 3a, 3b; Brook 1893: 99 with notes on its origin. (synonymy). Then, in 1977 came the major discovery of Acropora cytherea: Crossland 1952: 215; Hawaiian A cropora. John Naughton found a Wallace 1978: 289, pis. 63, 64A, 64D, tabulate colony about 2 ft acro ss at La 66C, 66D, 67. Perouse Pinnacles, French Frigate Shoals, M adrepora efflorescens Dana , 1846:441, some 500 miles northwest of Kauai. The pI. 33, fig. 6; Brook 1893:35 colony was subsequently determined to be (synonymy). A . cytherea. More colonies were located by ? Acropora efflorescens: Pillai and Scheer Leighton Taylor. In 1978 and 1979, R. W. 1976:26, pI. 3, fig. 3. Grigg found still more A. cytherea and two M adrepora armata Brook, 1892:452; other species, A . valida and A . humilis, the 1893: 100, pI. 10, figs. A, B (synonymy). first two in relatively large numbers at depths M adrepora reticulata Brook, 1892:461; ranging from 3 to 20 m at a variety of 1893:68, pI. 4, figs. A, B. localities at French Frigate Shoals. This dis­ Madrepora reticulata var. cuspidata Brook , pelled any idea that Acropora is a rare 1893:69. Hawaiian coral and ' initiated the present A cropora reticulata: Wells 1954:422, pI. paper. 11 0, figs. 4-6, pI. 114, figs. 1-6 (synonymy); Pillai and Scheer 1976:28, pI. 7, fig. 1. TAXO NOM Y M adrepora arcuata Brook, 1893: 102, pI. 12; Studer 1901 :395. The three species of Acropora in Hawaii Acropora cytherella Verrill 1902 :253, pI. identified from specimens collected at French 36, fig. 7, pI. 36a, fig. 7, pI. 36f, fig. 1 Frigate Shoals are A . cy therea (Dana), A . (synonymy). valida (Dana) (= A. variabiiisi, and A . A cropora corymbosa "cytherea Form ": humilis (Da na). All of these species have broad Indo-Pacific distributions, and all von Marenzeller 1907:32, pI. 1, figs. 1, occur in some locations of the type dubbed 2, pI. 2, fig. 3. "marginal belts" by Crossland (1949), char­ Acropora hyacinthus (part): Hoffmeister acterized by "absence or almost absence of 1925:64; Wells 1954:421. the great reef builder, the genus Acropora." MATERIAL EXAMINE D: 4 specimens col­ Unfo rtunately, with knowledge of the lected at French Frigate Shoals, July 1979. biology of Acropora in a poor state, it is not possible to discuss special life-history strat­ DESCRIPTION: This species occurs as stalked egies of these species which might explain tables, up to 480 em in diameter, with a their broad distribution. However, all can be more or less rounded outer edge, sometimes associated with either severe or unpredic­ with additional platelike layers. Color is pale table conditions. A. cy therea is known on the brown. Great Barrier Reef (mainly from unpub­ SKELETON: The specimens are all portions lished observation s) as one of the plate­ of flat plates (figure l a). The primary direc­ forming species which are early colonizers of tion of branching is horizontal and there is disturbed areas. A . valida characteristically considerable, but not complete, ana stomosis occurs as the dominant (and sometimes the of horizontal branches. From these branches, only) A cropora in the austere conditions of short vertical to oblique branchlet s mainly the algal ridge (e.g., in the Marshall Island s). 5-10 mm in length arise singly or in groups Similarly A . humilis is a dominant (or (figure Ib). The radial corallites are ap­ codominant) cora l on windward reef fronts pressed tubul ar with the outer wall extended receiving extreme wave action on the Great upward as a lip. Many radial corallites are in Barrier Reef. tran sition to axial corallite status, and these Acropora in Hawaii. Part i -GRIGG, WELLS, AND WALLACE 5

FIGURE I. Acropora cytherea: a, general view, O.5x; b, S.E.M. of branchlet bundle , 8x ; c, S.E.M. of coenos teum between corallites, 180x ; d, S.E.M. ofradial corallites, 18x . appear tubular with dimidiate openings. The fura Sea, China Sea, Great Barrier Reef, axial corallites are usually 2-5 mm exert. Solitary Islands, Funafuti Atoll, Fiji Islands, Their outer diameter is 1.5-2.5 mm, inner Marshall Islands, Tahiti, Northwestern diameter 0.6-1.2 mm. The primary septa are Hawaiian Islands, Hawaii. all present, up to tR; the secondary cycle is absent, or occasionally 1-3 septa are just DISCUSSION: This is a common reef-front visible. The coenosteum is costate on radial species,sometimes referred to as A. reticulata, a synonym, and sometimes confused with A. corallites and reticulate with scattered, later­ hyacinthus. On many reefs A . hyacinthus and ally flattened spines elsewhere (Figure Ic, d) . A. cytherea have overlapping distributions, DEPTH RANGE AT FRENCH FRIGATE SHOALS : with the former shallower than the latter. A. 3-20m. hyacinthus does not have the long axial co­ rallites of A . cytherea; its radial corallites GEOGRAPHIC DISTRIBUTION: Madagascar have more rounded lips and are more evenly (Pichon, personal communication), Masca­ arranged on branchlets. There is a slight rene Archipelago (Faure 1977), Amirante possibility that some of the table colonies at Islands, Seychelles, Maldives, Minicoy, Ara- French Frigate Shoals are A . hyacinthus . 6 PACIFIC SCIENCE, Volume 35, January 1981

Acropora valida (Dana, 1846) to less than i-R. The coenosteum consists of lines of laterally flattened spines (Figure 2b). M adrepora valida Dana, 1846:461, pI. 35, fig. 1; Brook 1893: 168. DEPTH RANG E AT FR ENCH FRIGATE SHOALS: Acropora valida: Hoffmeister 1925:60, pI. 3-20 m. 12, figs. 1, la-c; Wells 1954:429, pI. 130, GEOGRAPHIC DISTRIBUTIO N: Madagascar figs. 7-9; Nemenzo 1967:99, pI. 30; Pillai and Scheer 1976:29, pI. 7, figs. 3,4; Zou (Pichon , personal communication), Ma sca­ rene Archipelago (Faure, personal com­ Ren-lin 1978: 107. munication), Red Sea, Maldives, Andamans, Madrepora variabilis Klunzinger, 1879: 17, pI. 1, fig. 10, pI. 2, figs. 1, 5, pI. 5, figs. 1, Nicobar Island , Mergui Archipelago, Singa­ pore, Philippines, China Sea, Great Barrier 3, pI. 9, fig. 14; Brook 1893: 161. Reef, Solitary Islands (Veron et al. 1974), Acropora variabilis: von Marenzeller 1907:49, pI. 15, figs. 40-44; Vaughan Fiji Islands, Marshall Islands, New Cale­ 1918: 181, pI. 80, figs. 2, 3, 3a, 3b; donia, Tonga, Tuamotus, Northwestern Hawaiian Islands. Faustino 1927:276; Wells 1950:38; 1954:428, pI. 128, figs. 1,2, pI. 130, figs. DISCUSSION: The two species A . valida 1, 2; Rossi 1954: 52; Stephen son and (type locality, Fiji Islands) and A. variabilis Wells 1956: 19; Scheer and Pillai 1974: (type locality, Red Sea) are here combined 23, pI. 8, fig. 2; Pillai and Scheer 1976: for the first time, as the relatively little­ 31 ; Zou Ren-lin 1978: 107; Wallace treated species A. valida falls well within the 1978:299, pI. 80C, D. large range of variability recorded for A. Acropora variabilis var. pachyclados variabilis. The French Frigate Shoals speci­ Klunzinger: Crossland 1952:222, pl. 38, mens are similar in all characteristics, except figs. 1, 6. the inner axial corallite diameter, to some MATERIAL EXAMI NED: 5 specimens, collect­ specimens from the Palm Island group, ed at French Frigate Shoals, July 1979. Great Barrier Reef.

DESCRIPTION: This species is corymbose in growth form (Figure 2a). Colonies up to 30 Acropora humilis (Dana, 1846) em high and 40 em wide occur at French M adrepora humilis Dana, 1846:483, pI. Frigate Shoals. Color is tan. 31, fig. 4, pI. 41, fig. 4. SKELETON: From a central attachment Acrop ora humilis: Wells 1954:425, pl. 100, area, branches arise vertically to obliquely, fig. 1, pI. 126, figs. 1-6, pI. 127, figs. 3, sometimes branching again, and sometimes 4, pI. 128, figs. 3-5 (synonymy) . Rossi proliferating at the tip. Branches are 7-12 1954:50; Stephenson and Wells 1956: mm greatest diameter, up to 70 mm in length, 15; Scheer 1967:424, figs. 4, 5; Pillai and only slightly tapering. Radial corallites and Scheer 1976:32; Zou Ren-lin 1978: are of mixed length, or mainly all of similar 105; Wallace 1978:300, pis. 81, 82, 83. length, up to 5 mm long, partly to fully (For a complete synonymy refer to Wells appressed tubular (Figure 2c). The openings 1954.) are oval, directed at 45° to 90° to the branch, MAT ERIAL EXAMINED: 1 specimen collected the outer edge of the wall sometimes extends at French Frigate Shoals, 8 m depth, July upward and is slightly liplike. Primary septa 1979. are all present, usually up to tR, but some­ times to tR; the secondary cycle is partly to DESCRIPTION: The specimen examined fully present, up to i-R. Axial corallites are comprises small pieces from a corymbo se or around 1 mm exert, with outer diameter caespitose colony (Figure 3a). The longest 2.0-2.8 mm, inner diameter 1.0-1.3 mm. branch is 55 mm in length, and the broadest Primary septa are present, up to t R; 22 mm. Radial corallites are mixed sizes, secondary septa are absent , or 1 to 3 present , up to 3 mm diameter and 3 mm in length, Acropora in Hawaii. Part I-GRIGG, WELLS, ANDWALLACE 7

FIGURE 2. Acropora valida: a, general view, O.7 x ; b, S.E.M. of coenosteum between corall ites, 16x ; c, S.E.M . of radial corallites, 160x. projecting at 90° to branch, tubular with di­ on radial corallites, with lines of laterally midiate openings and thickened walls (Figure flattened spines between radial corallites. 3b). Axial corallites are 1.5 mm exert, 4.0­ 5.5 mm outer diameter, 1.0-1.5 mm inner GEOGRAPHIC DISTRIBUTION: Madagascar diameter. Both septal cycles are developed, (Pichon 1978), Mascarene Archipelago primaries to t R, secondaries to ±R. Co­ (Faure 1977), Red Sea, Arafura Sea, Mal­ enosteum is a dense arrangement of costae dives, Cocos-Keeling, Philippines, China 8 PACIF IC SCIENCE, Volume 35, Jan uary 1981

FIGURE 3. Acropora humilis: a, general view, I x; b, S.E.M. of radial corallites, 16x .

1000 FFS DISTRIBUTION AND ABUNDANCE 500 OF ACROPORA IN THE HAWAIIAN ARCHIPELAGO .. "'E 100 ABU,NDA NT MAR a Acropora cytnerea 0 CJ Acropora valida 0 50 0 COMMON ISSS'3 Acropora humilis 10 - ...... , (J) lJJ Z GAR DNE R 0 --l 10 -r- 0 UNCOMMON o , u, 5 0 -r- a: RARE LAYSAN lJJ CD , ::E KAUA I NI HOA ::::>z 05

H M 0 NE LI PH MK 0 18 19 20 21 22 23 24 25 26 27 28 29 30 L ATITUDE (ON)

FIGURE 4. Distrib ution and abundance of Acropora spp. in the Hawaiian Archipelago in the most favorable habitats. Note that abunda nce is expressed on a log scale. This is done to emphasize the importance of one colony, that is, the capability of colonization. Only one colony of A . cytherea was found on Kauai and Nihoa and two on Laysan, clearly setting the outer limit of the distribution. The center of the distribution is French Fr igate Shoa ls, where over 1000 colonies ofA. cytherea can be found in a 5000 m2 area. Acroporain Hawaii. Part I -GRIGG, WELLS , AND WALLACE 9

Sea, Great Barrier Reef, Solitary Island s 40 percent of the bottom cover (Figure 5). (Veron 1974), Solomon Island s, Samoa, Fiji, The largest colonies of A . cytherea measured Marshall Islands, Tuamotus, French Frigate at French Frigate Shoals are 480 em in Shoals, Hawaii . diameter. At French Frigate Shoals A. cytherea DIS CUSSION: This species, as revised by occurs between .depths of 3 and 20 m in Wells (1954), has a long list of synonyms. man y habitats : inside the lagoon on patch Studies underway on the Great Barrier Reef reefs, in areas partially protected from open suggest that this lumping may be too ex­ sea, and off all exposed sides of the island. treme, and that two, or even three, species The most well-developed colonies occur at may be involved. However, there is no doubt the southwestern end of the "atoll" near that the French Frigate Shoals material is Disappearing Island . A . valida occurs in the same species as A. humilis s. s. A. humilis more restricted habitats partially protected is characteristically a shallow-water Acropora from open sea but at similar depths. One occurring as a dominant species on outer reef colony of A . humilis was found at French flats and the upper few meters ofexposed reef Frigate Shoals inside the barrier reef at a fronts , but it does also occur commonly on depth of 10 m. Several other colonies of A. horizontal reef surfaces that are subtidal and humilis were found at Maro Reef also near not exposed to strong wave action. In the first 10m depth. Of the three species of A cropora situation, colonies tend to have a broadly in Hawaii, only A . cytherea has a variable encrusting base (of a similar diameter to the growth form dependent upon habitat. Inside corallum) from which branches arise verti­ the barrier reef, vasiform colonies of A. cally with little or no secondary branching (e.g., see Wells 1954, pI. 128, fig. 5; Morton cytherea grow higher off the bottom with 1974, fig. IOD; Wallace 1978, pI. 8IA). In the taller and narrower supporting basal stalks than colonies outside the reef. Colonies in second, the attachment diameter is smaller areas exposed to open sea are more heavily than that of the corallum, and secondary branching is more frequent (e.g., see Wallace branched and thicker than colonies in pro­ 1978·, pI. 8IB). The French Frigate Shoals tected habitats. material is in the second category. The distributional patterns of all three species are very patchy. For example , the sighting of one colon y during a survey would invariably increase the probability of finding DISTRIBUTIO N, ABUNDA NCE, AND another colony of the same species. Aggre­ REPRODUCTIV E BEHAVIOR gation could result from gregarious larval The results of coral reef surveys off all behavior or immediate settlement or could major islands in the Hawaii an Archipelago be due to physical fragm entation. At French indicate that species of Acropora occur on Frigate Shoals many settlement sites can be only six islands (Kauai, Necker, French distingui shed, hence larval settlement must Frigate Shoals, Gardner Pinnacles , Maro occur from time to time. Even so, larval Reef, and Laysan, Figure 4). Three species recruitment may not be a continuous pro­ of Acropora were found at French Frigate cess. The size frequency distribution of A. Shoals and Maro Reef, whereas only one cytherea is relatively continuous, suggesting was collected at Kauai, Necker, Gardner low year-class variability; however, this Pinnacle s, and Laysan islands. The distribu­ could also be a result of frequent frag­ tional center for all three species appears to mentation. In June 1978, 36 colonies of A. be at French Frigate Shoals, where they are cytherea at 10m adjacent to La Perouse most abundant (Figure 4). A . cytherea and Pinnacle were measured and tagged in order A . valida are about three times more abun­ to determine growth. In September 1979, dant at French Frigate Shoals than at Maro the same area was resurveyed, but only 16 Reef. In some localities at French Frigate tagged colonies could be found . It was evi­ Shoals, A . cytherea occupies an average of dent that man y colonies in this area had 10 PACIFIC SCIENCE, Volume 35, January 1981

FIGURE 5. Benthic photo-transect conducted at French Frigate Shoals at 10 m depth on the outer reef south of Disappearing Island. In this area Acropora cytherea accounts for 40% of the bottom cover. been fragmented or overturned. In January few months to detect whether or not an 1979, a severe "Kona" storm hit this portion annual cycle exists or whether or not gona­ of the archipelago and may have caused the dal material in colonies from these islands observed destruction. In any case, it would ever reaches reproductive maturity (Table 1). appear that the aggregated patterns of distri­ The largest oocyte or spermary recorded bution for both A. cytherea and A. valida are in the three species of Acropora on any due to both fragmentation and larval be­ Hawaiian island was 41011 in A . cytherea, havior, at least at French Frigate Shoals. 24011 in A . valida, and 25011 in A. humilis In order to determine the degree to which (fable I). Recent work in Australia by larval settlement within the archipelago is Carden Wallace and Jamie Oliver (personal due to local reproduction versus outside re­ communication) on reproductive cycles of cruitment, colonies of A . cytherea and Acropora indicates that mature or near­ A . valida were collected at French Frigate mature colonies of A . cytherea contain Shoals, Maro Reef, and Laysan Island oocytes up to 93411 in length and spermaries during as many months of the year as pos­ up to 132111 in length. Equivalent measures sible. Each collection consisted of fragments for A . valida are 128011 for oocytes and broken from the largest (most mature) 328011 for spermaries. These data leave little colonies present. The mesenterial septa of doubt that all the colonies examined in polyps from these colonies were examined Hawaii of at least these two species, and and all oocytes and spermaries were mea­ probably of A . humilis as well, were in an sured. Unfortunately, the results are incon­ immature state. Since the Hawaiian data clusive, primarily because the data cover too were collected over a six-month time period Acropora in Hawaii. Part I -GRIGG, WELLS, AND WALLACE 11

TABLE 1 REPRODUCTIVE CONDITION OF A cropora spp. IN THE NORTHWESTERN HAWAIIAN ISLANDS

SPECIES NUMBER NUMBER PERCENT AND MONTH COLONIES POLYPS POLYPS OOCYTE SPERMARY LOCATION COLLECTED COLLECTED EXAMINED WITH GONADS LENGTH (urn) LENGTH (um)

Acropora cytherea F FS* Sept. 1980 20 64 44% n = 45 n =40 x = 106 x = 126 SD = 44.5 SD = 36.9 Range = 35-300 Range = 80-220 FFS Nov. 1978 4 21 100% n = 82 n = ­ x = 238 x= - SD = 83.4 SD= ­ Range = 110-440 Range = ­ Maro June 1979 7 32 0% n= ­ n= ­ x= - x= - SD= ­ SD= ­ Range = - Range = - Laysan June 1979 10 0% n= ­ n= ­ x= - x= - SD= ­ SD= ­ Range = ­ Range = ­ Laysan Sept. 1980 10 10% n= ­ n=2 x= - x = 100 SD= ­ SD=O Range = - Range = - A cropora valida FFS June 1979 2 15 93% n = 85 n=2 x = 72 x = 110 SD = 46.5 SD = 70.7 Range = 15-240 Range = 60-160 Maro Sept. 1980 20 65 7% n = 18 n= ­ x = 58 x= - SD = 26.0 SD= ­ Range = 30-110 Range = - Acropora humilis Maro Sept. 1980 6 24 62% n =40 n=3 x = 128 x = 153 SD = 56.0 SD = 50.3 Range = 50-250 Range = 60-200

* French Frigate Shoals

(June, September, November), it would be Observations at Gardner Pinnacles (mid­ surprising but not impossible to have missed way between French Frigate Shoals and the season in which reproductive maturity Maro Reef) may shed further light on the occurs in all three species. On the basis of question of the source and frequency of existing data it therefore appears unlikely larval recruitment in the Hawaiian Archipel­ that species of Acropora in Hawaii success­ ago. At Gardner Pinnacles in 1978, most of fully produce larvae. Recruitment may de­ the reef at 10m depth was dominated by pend entirely on recolonization from outside very large (1-10 m diameter) but dead col­ the archipelago . If this conclusion is correct, onies of A . cytherea. Only three living col­ the patterns of distribution and abundance onies were sighted in 1978, all less than 4 em of Acropora in Hawaii should reflect the In diameter. On the surface of the dead present route of colonization (see Acropora colonies of A. cytherea, the largest colonies in Hawaii. Part 2. Zoogeography). of P. lobata that were present were collected 12 PACIFIC SCIENCE, Volume 35, January 1981

(n = 10) and later sectioned in the labor­ Marine Affairs Coordinator is gratefully atory so that growth bands could be counted acknowledged. Steven J. Dollar, Michael (Buddemeier, Maragos, and Knutson 1974; Palmgren, Lisa Boucher, Catherine Agegian , Grigg and Dollar 1980). The oldest colonies and Tony Chess assisted in various aspects were 16 years, suggesting that some catastro­ of the fieldwork . James Maragos made avail­ phic event prior to 1962 killed all of the A. able to us the only known sample of cyth erea at Gardner. Since that time, the Acropora from the Island of Kauai. Lisa only successful recruitment appears to have Boucher analyzed the reproductive condition been in the last year or two. Either survival of all of the Acropora samples from Hawaii. of recruits at Gardner Pinnacles is very low We would also like to thank J. Darley, E. M. or recruitment is a very sporadic process (on Centre , James Cook University, for advice the order of once in 15 years). This may and use of equipment; the Australian explain why Acropora spp. have not dis­ Institute of Marine Science for photographic persed to more islands in the chain. If the assistance; and Dr. D. Kuhlmann, Museum source of recruitment is within the chain (if fiir Naturkunde, Berlin, and Dr. Yossi the center of distribution is French Frigate Loya, Tel-Aviv University, for the loan of Shoals), the Gardner data suggest that repro­ specimens. duction is sporadic, possibly only in very warm years. Alternatively, recruitment may depend entirely on sporadic colonization LITERATURE CITED from outside the archipelago. While recruitment of Acropora in Hawaii ARMSTRONG, R. W., ed. 1973. Atlas of may be discontinuous and severe mortality Hawaii. University of Hawaii Press, may be intermittent, it appears as though Honolulu. 222 pp. A cropora has been in Hawaii for some time. BROOK, G. 1893.The genus Madrepora. Cat. In 1979, two 50-cm cores, 5 em in diameter, Madrep. Corals, Br. Mus . (Nat. Hist.), were extracted from the reef at a depth of 10 vol. 1. m at French Frigate Shoals . At the bottom BUDDEMEI ER, R. W., J. E. MARAGOS, and of one of the cores was a piece of fossil coral J. W. KNUTSON. 1974.Radiographic studies identified as Acropora (Michel Pichon , per­ of reef coral exoskeleton. 1. Rates and sonal communication). Unfortunately this patterns of coral growth. J. Exp. Mar. piece of coral was unsuitable for radiocar­ BioI. and Ecol. 14: 179-200. bon dating due to a high calcite content. CROSSLAND, C. 1949. Reef corals of the However, directly above the Acropora there South African coast. Ann. Natal Mus. was an intact fragment of Porites suitable 11(2): 169-203, 14 pis. for radiocarbon dating. The radiocarbon age DANA, J. D. 1846. Zoophytes. U.S. Expl. of this fragment is 414 ± 68 years. This is Exped., vol. 7. clear evidence that Acropora has been in DANA, J. D. 1872. Corals and coral islands. Hawaii since before the time of Western New York. contact (1778). The zoogeographic implica­ EDMONDSON, C. H. 1946. Reef and shore tions of this finding coupled with the data fauna of Hawaii. Bernice P. Bishop Mus. on reproduction, ecology, and possible re­ Spec. Publ. 22. cruitment from outside the archipelago are FAURE, G. 1977. Annotated check list of sufficientlycomplex to warrant separate treat­ corals in the Mascarene Archipelago, ment and therefore are considered in Part 2 Indian Ocean. Atoll Res. Bull. 203 : 1-26. of this paper. GRIGG, R. W., and S. J. DOLLAR. 1980. The status of reef studies in the Hawaiian Archipelago. In R. W. Grigg and R. ACKNOWLEDGEM ENTS Pfund , eds. Proceedings of the Sympo sium on Status of Resource Investigations The assistance of the National Sea Grant in the Northwestern Hawaiian Islands. Program and the Hawaii State Office of the UNIHI-SEAGRANT-MR-80-04. Acroporain Hawaii. Part I-GRIGG, WELLS, AND WALLACE 13

GRIGG, R. W., and L. BOUCH ER. A new of the Hawaiian Islands and Laysan. Proc. method to assess reproductive condition 7th Int. Zool. Congr. Boston. 12 pp . in corals. Manuscript. VAUGHAN, T. W. 1910. Review of: J. S. MARAGOS, J. E. 1977. Order Scleractinia. In Gardiner, Flora and fauna of the Maldive Reef and shore fauna of Hawaii. 1. and Laccadi ve Islands, vol. 2, suppl. I, Protozoa through ctenophora. Bernice P. Madreporaria, parts 3 and 4. Science Bishop Mus. Spec. Pub!. 64(1):158-241. 21 :984-985. MARSHALL, S. M., and T. A. STEPHENSON. VAUGHAN, T. W., and J. W. WELLS. 1943. 1933. The breeding of reef animals. Part 1. Revision of the suborders, families and The Corals. Gt. Barrier Reef Exped . 1928­ genera of the Scleractinia. Geol. Soc. 29. Sci. Rep. 3(8) :219 -245. America, Spec. Pap. 44. MORTON, J. 1974. The coral reefs of the VERON, J. E. N . 1974. Southern geographic British Solomon Islands: a comparative limits to the distribution of Great Barrier study of their composition and ecology. Reef hermatypic corals. Proc. 2d Int. Proc. 2d Int. Symp . On Coral Reefs 2 : Symp . on Coral Reefs I :465-473. 31-53. VERON, J. E. N ., R. A. How, T . J. DONE, MOSELEY, H. N. 1879. Notes by a naturalist L. D. ZELL, M. J. DODKEN, and A. F. on the Challenger. . .. Macmillan, London. O'FARREL L. 1974. Corals of the Solitary PICHON, M . 1978. Recherche sur les peuple­ Islands, Central New South Wal es. Au st. ments adominance d'althozoaires dans les Jour. Mar. Freshwater Res. 25: 194- 208. recifs coralliens de Tulear (Madagascar). VERRI LL, A. E. 1902. Notes on corals of the Atoll Res. Bull. 222 : 1-446. genu s A cropora (Madrepora Lam.), with QUELCH, J. J. 1886. Report on the reefcorals. new descriptions and figures of types, and Challenger Reports, Zoology, vol. 16. of several new species. Conn. Acad. Arts RINKEVICH, B., and Y. LOYA. 1979. The Sci., Trans. 11:207- 266. reproduction of the Red Sea coral Stylo­ WALLACE, C. 1978. The coral genu s A cro­ phora pistil/ata. 1. Gonads and planulae. pora (Astrocoeniinae: Acroporidae) in the Mar. Ecol. Prog. Ser. 1: 133-144. central and southern Great Barrier Reef STUD ER, T. 1901. Madreporarier von pro vince. Mem. Queensland Mus. 18(2) : Samoa, den Sandwich-inseln und Laysan . 273-319, pis. 43-103. Ergebnisse einer Reise nach dem Pacific WELLS, J. W. 1954. Recent corals of the (Schauinsland 1896-1897). Zoo1. Jahrb. Marshall Islands. U. S. Geol. Survey, Syst. 14:388-428. Prof. Paper, 260-1: 385-486. VA UGHAN, T. W. 1907. Recent Madreporaria