MACY Y WosnyA SNR ES Ayal ye! re eek | sn, Hat Wy egy Derinizye Pry nt the font ‘hdeentn ats SEN mY SX meni cane, MMI na tarp 4 Nee Haat mew cha Nitta cevaris ns) : ri ae ie Saree iia 2 : Sy i we ite wna y Ronis rear} La in Tete Arath ve ST TUN or wr nly SEM ete PTY ih Tio menting APACER RV ITO eeu fee wealthy ‘ : 2 Maker te eeay ay 1 . TERE Ses Bey .S ee ny tgee yg A ree SUA Lodcotin poy ge iat Ik ee fat ig y GLEE 5 OOO CSET asa SAS te wae ta eta i tite aw “yr aviet fe Ns ¥ ASE 7 bys " dig iat i aT pak i on Dynes Ao Oe eters. McA SES Maem Seite eta tore TaN Tas GhAdin wa IEAM cae ta at a vats ens. tut US EW a ta Teena Fee Wea WAN, Nee tay any P Baa : PE henge: fartyas ay cer Wen dey Par ti aq “pA A EE se) ec (SE) gs) A ‘ < fe} jo} oy a SO” S Sud) © Su) & ONY E uy) 5 Gs) 2 GS ao” ro} YNoseSY = S wasnt ro) ¥y S es) Sash’ ro} ao” ‘ a =a = za 4 =| z F a) a LLALILSNI” NVINOSHLIWS S31YvVuUdIT LIBRARI ES SMITHSONIAN INSTITUTION NE) Ser NVINOSHLINS $3 = z r = c , o S o 2 oO =) = ax. oe >" = > Fs > E > = a re = - = is D a ® Z ® = ite SAE 2 4 2 }/BRARIES, SMITHSONIAN _ NOILNLILSNI_ NVINOSHLINS S31YVYGI7_ LIBRARIES SMITHSONIAN INS ea n” z a) z on z oe w = = = = < ES ieee Sie Sj z = z a z NS 4 = ro) = ro) etek D9 QYR = n ” nw 2 2 ’ a : : g 2 = 2 E z = z FS 8 z i's a ca a \. 2 A [ ILALILSNI Saluvuas7_ LIBRARIES SMITHSONIAN _ INSTITUTION NOILNLILSNI_ NVINOSHLINS Sa1u¥ = o 2 M4 z He Ss « @ 7) = no = Fe) os a x wi = ox = x. = ox at SX oe 4 | = 3 2 e Ps = Ps = Ps 5 ea 3 2 3 a Zs re) = re) = Zz | 2 ma =z - = XS 'BRARI ES_ SMITHSONIAN INSTITUTION _ NOILNLILSNI_NVINOSHLIWS ~53 lyYvuag oe LIBRARI ES_ SMITHSONIAN _ S ff, = eI a 3 S = — 5 a 5 a aX: 5 a 5 g Pe) = ve = Sas SN = i EF = re, ss - = \S > = -, as ILALILSNI = NVINOSHLIWS LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SJIUt z 7) = 9 ” as on z 7) < = < = < = = = z 5 z og = 3 z a § z .§ z 5 NO i § Z iurtaks [e) 3 oO = > BS r Oo -r Oo E Zz, AN 4 = SNS z = z = : 3 z Ba ee Ria 3 S BRARIES SMITHSONIAN _INSTITUTION | NOILNLILSNI_NVINOSHLINS S31UYV4NG17 LIBRARIES. SMITHSONIAN INSTITU = w = wo _ -_ B i a - = 3 2 e a a S = oc yee < viel < = _< = = A — S c - ee S ma + =f fo) = ro) : = ro) a” 5 a a = z a = a] = ILALILSNI NWINOSHLINS S31iu¥vHgI7 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS $93] uvd = z Es z = z & z o ° o = “Oo S o ° : = 5 ze 5 a 5 re) = a = 2 = 3 2 F at = e 5 - s = : é 5 2 ae 5 Z Z r 3 SMITHSONIAN — NOILNLILSNI_ NVINOSHLINS S31YVYSI7_ LIBRARIES SMITHSONIAN INSTI g z o z o z jh o z= = < = < = ~< s = =< a z a z = ‘Zz \) = z = ro) x re) =i o MY’ 6 S 2 g Z gy 2 he 2 Z e 2 E 2 = 2 = > = a 2 a 2 ra wiz s ILNLILSNI_NVINOSHLIWS S31uvugI7 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31UW é i j 2 B : 2G 2 4 = 4 ~ = = 5 a = = =i . 2 S ac = = S c S — ro} = 3 = f 5 = 5 = 4 pets SPE oa = = = za ee BRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IuvY 811 LIBRARIES SMITHSONIAN INST = if c z ey z = z £ ae — lip o — o — = j ao E Ly, 2 5 2 = = 5 ~ i dy > = > Ke > = ‘ in paw i? 2 - a F 2 Ce = 2 o Zz a oa Cat re Z aL. ILNLILSNI >NVINOSHLINS Sa 1yvYyusii = LIBRARI ES SMITHSONIAN INSTITUTION NOILALILSNI_ NVINOSHLINS, S23 f B ’ a - d : < SONS = Le < = Laim <= * = : < aT ; >, 2 GR 5 hy, = ws 3 KAM = AQ = GL, = Lar SANS fo) SX es Xe Wass rae aS 3RARIES SMITHSONIAN re - fo) i= Eff = a La Ee ga” 5 ILMLILSNI_NVINOSHLINS Sa1uVUGIT_LIB z = z =| (e) ae o wo wn cE ) = 2 a = 3RARIES SMITHSONIAN _ wo 2 us : Pav) re Ln fe 28 4 <> oY, = 2 “gy = a t4 rs) a is Zz INLILSNI_ NVINOSHLINS i= S = ae iS) es) AS = 2 SX .e 2 MS i RARIES, SMITHSONIAN INSTITUTION | = z =1 za iin a AN = E = 5 LALILSNI_ NVINOSHLIWS ie oe a) ul : E (= 7 we sl 5 2 za —Jj RARIES SMITHSONIAN = Ge = ee 2H 2 Db . z o SaIuvVNSIT LIB Zz (2p) < = z =] (@) a5, w w ac Oo ¥ [= = = 5 (7) 7A RARIES SMITHSONIAN _ INSTITUTION uw Jets ca, Lh Guan 4 “ff ed a GG 3 a is 2 LM.LILSNI_NVINOSHLINS — — <2 [@) w » — Ps] S = 2 Ne: = NS = RARIES SMITHSONIAN _ < i A a KV = AX ~ INSTITUTION NOILOALILSNI INSTITUTION NOILNL SMITHSONIAN Was gi1_LIBRAF stir 4 S tNoswsY RARIES INSTITUTION NOILNLILSNI S3Iu¥VYdIT LIBRARIES INSTITUTION NOILOALILSNI SIIYVYUdIT LIBRARIES INSTITUTION NOILALILSNI LIBRARIES Ssasluvaugi LIBRARIES SMITHSONIAN NOILNLILSNI INSTITUTION saluvugi LIWS NVINOSHLINS NOILNLILSNI za oe. 2 & : = \. = S, w 2s NOILNLILSNI wo WwW oc << [ag joo} = SMITHSONIAN INSTITUTION NOILNLILSNI NIAN NVYINOSHLINS S3IYVvy INSTITUTION NOILALILSNI NVINOSHLIINS S3iyvudiT_LIBRARIES NY NN aes NE NOILALI NVINOSHLIWS i oa be INSTITUTION SMITHSONIAN NVINOSHLIWS LIBRARIES NG SMITHSONIAN saluvugi NVINOSHLINS S31uvuglt Yin NOILNLILSNI NVINOSHLINS INSTITUTION SMITHSONIAN SMITHSONIAN NVINOSHLIWS IT LIBRARIES ENG SMITHSONIAN ® saluvug SMITHSONIAN NVINOSHLIWNS (=! > *A\ = Ile % be aT) SS me f{2, eat) = Gus) SOY EL ARS a Ns Washi? ro) XS ad Ns | a z Sage = LIBRARIES SMITHSONIAN INST Salyuvug ars z c ES = wilh Zs pe > 2G yy > a = Gps 6 x es Ke ~ Yi, fost Se 2 5 INSTITUTION, NOILOLILSNI_ NVINOSHLINS S312 Ko |S ae = . ar Si Mw) Fe =] ye NEE gg z= Lip = 2 @ saiuvy g In LIBRARI ES SMITHSONIAN _ INST a Ww 5 zal as 4 = fos. — c < 3; a = aS = e 5 = dl N Peas Vie aa = w = 5 a 5 = > = = a = ~ LIBRARIES SMITHSONIAN INSTI ene aan re z fi Zz = = & Vy. 9 = Biles oe : 72) Oe J fi? = NWS? =e ne aie =i INSTITUTION pNOMOLILSNI_NVINOSHLINS Sa 1% us a W = o ss — ao < 2 _< oc =] oc 2 fo) & a Zz a Saiuvugiq LIBRARIES SMITHSONIAN _INSTI (im = i 5 og o = wee) x 5 2 fae Fe % co o z o INSTITUTION NOIIALILSNI oale Ks Z BEC = ‘ \. =] me 4 LAK 2 = - = WE ea ne = ah S ryvugin LIBRARIES SMITHSONIAN INSTI 3 bi i re b A ee ttf Jy a S ez % Lge % S 5 a a Za = 4) z INSTITUTION ~ NOILALILSNI _NVINOSHLIWS ~34 is} e) =f vg re) = i =) y =>) : = Se iB — WAX 5 z Gave ett Unies NVINOSHLINS S31YVvudl ot! B RARI BS eu SONA Dios = gam = b>, > WOW. 5 Ge™m = INS 29 NO Dy ky 1 Od Nos. 233-239 EE 65° * ATOLL RESEARCH BULLETIN 233. Check list of recent coral records from Aldabra (Indian Ocean ) by Brian Roy Rosen 234. Récifs coralliens, constructions alguaires et arrécifes a la Guade- loupe, Marie Galante et la Désirade par R. Battistini et M. Petit 235. Systematics and ecology of the land crabs (Decapoda: Coenobitidae, Grapsidae and Gecarcinidae) of the Tokelau Islands, Central Pacific by J. C. Yaldwyn and Kasimierz W odzicki 236. Some aspects of the ecology of reefs surrounding Anegada, British Virgin Islands: by R. P. Dunne and B. E. Brown 237. The intertidal algae of the main- land coast in the vicinity of Towns- ville, Queensland by Yinam Ngan and Ian R. Price 238. Blue-green algae (Cyanobacteria) of the oceanic coast of Aldabra by B. A. Whitton and M. Potts 239. Vegetation of Aldabra, a reassess- ment by R. J. Hnatiuk and L. F. H. Merton Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 ay 7] my. ; ai Vi epi Lh, jh blah me p ey uigrey ier jal!" ris Mere che cla di sao aaa : Fin Fat, i BA : fi nor) Oe ae ee Veattul, Maree 4 ; . + Atay er } foals sy. mabe Se ' | | + ree Riga) \ Gabe 7 . Wie Tebliy deus | é 7 De iis) ¢ is ae a ee FSP b+: Dev) -” ne Vs Veli a a 3 F F ‘lp | ne ee tog By) tre ite Pina ae | oa f bs oh, S ry / Re | | | , i 7 Y ri + Aas oF ATOLL RESEARCH BULLETIN 233. Check list of recent coral records from Aldabra (Indian Ocean ) by Brian Roy Rosen 234. Récifs coralliens, constructions alguaires et arrécifes a la Guade- loupe, Marie Galante et la Désirade par R. Battistini et M. Petit 235. Systematics and ecology of the land crabs (Decapoda: Coenobitidae, Grapsidae and Gecarcinidae) of the Tokelau Islands, Central Pacific by J. C. Yaldwyn and Kasimierz W odzicki 236. Some aspects of the ecology of reefs surrounding Anegada, British Virgin Islands by R. P. Dunne and B. E. Brown 237. The intertidal algae of the main- land coast in the vicinity of Towns- ville, Queensland by Yinam Ngan and Ian R. Price 238. Blue-green algae (Cyanobacteria) of the oceanic coast of Aldabra by B. A. Whitton and M. Potts 239. Vegetation of Aldabra, a reassess- ment by R. J. Hnatiuk and L. F. H. Merton Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 ACKNOWLEDGEMENT The Atoll Research Bulletin is issued by the Smithsonian Institution, as a part of its activity in tropical biology, to place on record information on the biota of tropical islands and reefs, and on the environment that supports the biota. The Bulletin is supported by the National Museum of Natural History and is produced and distributed by the Smithsonian Press, The editing is done by members of the Museum staff and by Dr. D. R. Stoddart. The Bulletin was founded and the first 117 numbers issued by the Pacific Science Board, National Academy of Sciences, with financial support from the Office of Naval Research. Its pages were largely devoted to reports resulting from the Pacific Science Board's Coral Atoll Program, The sole responsibility for all statements made by authors of papers in the Atoll Research Bulletin rests with them, and statements made in the Bulletin do not necessarily represent the views of the Smithsonian nor those of the editors of the Bulletin. Editors F. R. Fosberg Ian G. MacIntyre M.-H. Sachet Smithsonian Institution Washington, D.C. 20560 D. R. Stoddart Department of Geography University of Cambridge Downing Place Cambridge, England ATOLL RESEARCH BULLETIN No. 233 CHECK LIST OF RECENT CORAL RECORDS FROM ALDABRA (INDIAN OCEAN) by Brian Roy Rosen Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 a MAE 1S FS EN, riwepian Lia TALI ae way 3 Pee Ow AOL ITI. AsO ST AA) DA eekgeiden — 1 = —_ ian | Cred sodamevel iL - . ha ; ea a a + CHECK LIST OF RECENT CORAL RECORDS FROM ALDABRA (INDIAN OCEAN) by Brian Roy Rosen! INTRODUCTION The purpose of this check list is to bring together all records of corals at Aldabra since no proper listing and bibliography exist at present. I have also taken this opportunity to make several revised identifications and to add some depth data which were not available when I presented my earlier list (Rosen 197la). ‘It is in fact possible to extract partial depth information for the species in this earlier list from the accompanying account by Barnes et al. (1971, especially their Table 2). I have incorporated this information here, and therefore only cited Bellamy's MS figures where they supplement the published information. I should like to express my thanks to Dr. David Bellamy (University of Durham) for allowing me to use this information. I should also like to thank Professor John Wells (Cornell University) for permission to use and refer to a Manuscript currently in preparation. Dr. Michel Pichon (James Cook University) and Ms Zena Dinesen (James Cook University) kindly checked the taxonomic list, and Dr. C.G. Adams (British Museum (Natural History) ) offered helpful criticisms of the manuscript. COLLECTIONS AND LITERATURE Table 1 analyses known coral collections made at Aldabra and related publications. It therefore serves as an Aldabra coral bibliography, and extends the information given in the more general bibliographies by Stoddart (1971), and by Peters & Lionnet (1973). 1 Department of Palaeontology, British Museum (Natural History), Cromwell Road, London SW7 5BD. (Manuscript received January 1978 -- Eds.) It is undoubtedly true that published coral records from Aldabra are based on less than half of the total amount of coral material which has actually been collected there. The present check list therefore provides only a partial picture of Aldabra's coral fauna. Large coral collections were evidently made by Voeltzkow and by Fryer (Sladen Expedition), but only a fraction of their material has ever been studied. Stoddart (1971) has given an account of these earlier expeditions. He points out, however, that until the visits by the Calypso Expedition (Cousteau, Cherbonnier) and by J.L.B. Smith, both in 1954, no specialist marine collections had been made at Aldabra. Cherbonnier collected corals and other groups, and Smith collected fishes. The Cousteau-Cherbonnier coral collection has recently been identified by Dr. Michel Pichon (James Cook University of North Queensland) and myself, and an account is currently in preparation. The next most important period of coral collection is the most recent, This is represented by the series of phases of the Royal Society Expedition to Aldabra, and associated visits, reconnaissance, etc., which began in 1966. Only the corals from Phase VI have so far been identified. Drew (1977) recently published a few photographs of the ecological transects carried out by the Phase VI party, and I have included a few identifications that were possible from the photographs. There remain the Phase I, II and V collections. The corals from the first two of these Phases were collected by Dr. John Taylor (British Museum (Natural History)), and together with the Phase VI corals constitute the most important of the Aldabra coral collections. Unfortunately, only the sight records for corals from Phases I and II have so far been published. For Phase V, Brander et al. (1971) mentioned a few names, but presumably a more extensive collection awaits published identifications. A number of identifications of both Fryer's and Taylor's material were incorporated in an account of coral zoogeography in the Indian Ocean (Rosen 1971b, see tables), but in that paper, Aldabra was considered together with neighbouring islands ("Aldabra-Glorioso group") and the information in the tables cannot therefore be quoted here. All but two possible generic records in that list however are covered in the present check list. The opportunity is also taken here to give the register numbers of specimens of Aldabra material but only where authors have previously published them. Finally it may be noted that of all the coral material from Aldabra, only six genera (Ddoderlein, 1902, Matthai, 1914, 1928) have ever been fully described. REMARKS ON THE CORAL FAUNA Fifty five coral genera* have so far been described or listed. * subgenera are counted as separate genera in this discussion Of these, 4 are non-scleractinians, and 4 are ahermatypic scleractinians one of which is only a sight record (Paracyathus) . Of the remaining 47 (hermatypic) genera, one (Stylocoeniella) is only a sight record. In addition to the corals listed here, two further genera may be present at Aldabra: Merulina and Plerogyra, since they appear in the Aldabra-Glorioso group list (Rosen 1971b). On the other hand, two genera (Physophyllia and Gyrosmilia) in the present check list require verification by re-examination of the original specimens. The figure of 46 hermatypic coral genera seems therefore to be a reasonable interim working total. This figure differs from the 50 given by Rosen (1971b, table 1, line b) for the Aldabra-Glorioso group. Three genera in that tabulation required confirmation, two of these being the genera discussed above and also listed here. The third however, Anomastraea, has now been re-examined and is in fact a Coscinaraea (see check list for details). Plesioseris in the earlier lists is now no longer regarded as a distinct subgenus of Psammocora (see below), and finally there are the two genera (Merulina and Plerogyra) known from elsewhere in the Aldabra-Glorioso group but not as yet from Aldabra itself. Four genera may therefore be subtracted from the Aldabra-Glorioso. list in order to make it comparable with the present one, and this leaves 46 hermatypic genera, i.e., one fewer than here. This does not however signify the addition of a new generic record as the new name, Gardineroseris appears by reason of synonymy changes affecting Agariciella and Leptoseris. It is less easy to be precise about numbers of coral species at Aldabra, even if the perennial problem of what constitutes a coral species is disregarded. There are 95 named species here, taking all "cf." forms as distinct. In addition, there are 8 genera for which there are only unnamed species records ("sp.", "spp.", etc.),or no mention of species at all. I have arbitrarily counted all these records as one species for each genus concerned, so bringing the total to 103 species. (Note that in the list which follows, unnamed species records have been treated as genus records. See next section). Of the 103 species reckoned in this way, 6 are non-scleractinians and a further 4 are ahermatypes, leaving a total of 93 hermatypes. Two of these require verification. Sight records include 1 ahermatype, 5 hermatypes. By comparison with other Indian Ocean reef regions, the above figures suggest that few new scleractinian genera will be added by further work on Aldabra corals, but that many more species (or "species") may well be recorded. On the first point, Rosen (1971b) interpolated only a further 5 genera, 3 of them fungiids, for the Aldabra-Glorioso'group. Verrillofungia, Ctenactis, Halomitra, Alveopora and Oulophyllia. For non-scleractinians, there is a good possibility of more hydrocoral genera and of species, because in various works by Boschma there are species known from nearby islands not as yet recorded from Aldabra. Because of the Cousteau-Cherbonnier list currently in preparation, it is not useful to discuss zoogeographical aspects here, except perhaps to draw attention to "shrinkage" in the Indian Ocean distribution of Anomastraea, following its elimination from Aldabra (see Rosen 1971b, fig.8). NOTES ON THE PRESENTATION OF THE CORAL RECORDS The list gives all known scleractinian records, together with the usual coralline reef octocorals and hydrozoans. The corals are set out in Treatise order (Wells 1956) but with the octocorals and hydrocorals at the end. Square brackets have been used for both species and generic records to indicate names now placed in synonymies and records which are otherwise incorrect for Aldabra. Their current names, where applicable are also given in the appropriate places in the list, and pairs of names affected in this way are cross referenced. For the most part the synonymies of more recent works, where they affect Aldabra records, have been accepted uncritically because the present list is primarily intended as a compilation. Generic names which have been used in the past, but which are now widely regarded as synonyms, are given in square brackets immediately beneath the currently valid names (e.g. Platygyra and [Coeloria]). The status of each record can be assumed to be a formal taxonomic treatment (check list, description, etc.) unless additional information is given: SR - sight record TC - citation of taxon in a text account, e.g. in diagrams or non-taxonomic tables. Plate and figure numbers have been given for species only where the coral illustrated is indicated in an author's caption as being from Aldabra. Specimen numbers are quoted only if authors have published them. Abbreviation used: BM(NH) - British Museum (Natural History), Dept. of Zoology. Depth data for species follow each author's record and consist only of the actual depths stated by individual authors. No interpolated depth ranges have been given. For generic depth ranges, all the depth records for each genus and its species are brought together, and again, no interpolations are made except in the topmost 5m. Here "surface" records have been taken to be Om and extended down to the next recorded depth which falls within the range O-5m. For depths given as "Bellamy MS", see the Introduction. Genera and species. Records of unidentified species are listed directly under the generic name, even when an author has distinguished several such "sp." names by number, or given a "spp." indication. There is no way of knowing whether more than one "sp." record within a genus represents one or more species (see remarks on the species totals in the previous section). PUBLISHED CORAL RECORDS FOR ALDABRA 1. Stylocoeniella Yabe & Sugiyama, 1935 Recorded depth range: surface Price 1971, 150 SR TC surface 2. Psammocora Dana, 1846 Recorded depth range: 18m Psammocora haimeana Edwards & Haime, 1851 Voeltzkow 1902, 565 Psammocora nierstraszi van der Horst, 1921 Rosen 197la, 109 Bellamy MS: 18m 3. Stylophora Schweigger, 1819 Recorded depth range: O-5m Brander et al. 1971, 417, 423, 424 TC less than 5m Fryer 1911, 411 SR TC Price 1971, 166 SR TC surface S. mordax (Dana, 1846) Barnes et al. 1971, 101, 102 TC 1-2m Brander et al. 1971, 423, 425 TC less than 5m Rosen 197la, 110 S. pistillata (Esper, 1797) Barnes et al. 1971, 103 TC 2m Rosen 197la, 110 4, Seriatopora Lamarck, 1816 Recorded depth range: surface, 12m, 27m, 33m Barnes et al. 1971, 103 TC 12m Fryer 1911, 411 SR TC Rosen 197la, 110 Bellamy MS: surface [S. hystrix Dana, 1846] Barnes et al. 1971, 97, 103 TC 27, 33m (probably should be S. c£. hystrix as in Rosen 197la). S. cf. hystrix Dana, 1846 Rosen 197la, 110 Bellamy MS: surface (see also S. hystrix, Barnes et al.) 5. Pocillopora Lamarck, 1818 Recorded depth range: surface to) Sm 2 eee ol Brander et al. 1971, 417, 423, 424 TC less than 5m Drew 1977, pl.la (BRR) Fryer 1911, 411, 412 SRe re Price 971), 150 SR TC surface P. damicornis (Linnaeus, 1758) Barnes eb al. 1971,) 103) Teen Rosen 1971la, 110 P. danae Verrill, 1864 Barnes et al. 1971, 103 Torro “27,7 33m Rosen 197la, 110 Bellamy MS: 20-27m P. cf£. danae Verrill, 1864 Price 1971, 166 SR TC surface P. eydouxi Edwards & Haime, 1860 Barnes et al. 1971, 102 Te- 3,5, 8-10, 12,° 255m Brander et al. 1971, 423 TC less than 5m Rosen 197la, 110 Bellamy MS: 5-18m 6. Acropora Oken, 1815 Recorded depth range: surface to 33m [Madrepora auctt. (non Linnaeus, 1758) ] Barnes et al. 1971, 93, 94, 102, 103 (including Spp. nos’. 2, 3, 4) TC lees Dy fp le Bellamy et al. 1969, 103 SR TC dead Drew 1977; “pls la; aby aici sayy esb es eee) Fryer 1911, 410, 411 (as Madrepora) SR TC Rosen 197la, 110 (spp. "Nos. 1-4"), Bellamy MS: surface to 15m A. cf. cuneatus [sic] (Dana, 1846) Price 1971, 166 SR TC surface A. digitifera (Dana, 1846) Price 1971, 150 SR TC surface A. c£. digitifera (Dana, 1846) Price 1971, 166 SR TC surface A. cf£. diyvyersa (Brook, 1891) "No.1" Barnes et al. 1971, 103 TC 9m Rosen 1971a, 110 Bellamy MS: 2-8m A. cf. diversa (Brook, 1891) "No.2" Barnes et al. 1971, 103 TC 2m Rosen 197la, 110 A. glochiclados (Brook, 1893) Barnes et al. 1971, 101, 102, 107 TC 1-2m Rosen 197la, 110 A. irregularis (Brook, 1892) Barnes et al. 1971, 102 TC Im Price 1971, 166 SR TC surface Rosen 197la, 110 A. monticulosa (Brueggemann, 1879) Barnes et al. 1971, 103 TC 13m Rosen 197la, 110 Bellamy MS: surface A. palifera (Lamarck, 1816) Barnes et al. 1971, 102 TC 1-3, 5-7m ?Drew 1977, pl.ic, 3b (BRR) Brander et al. 1971, 417, 423, 424 TC less than 5m (sometimes as "Acropora A") Rosen 197la, 110 Bellamy MS: 1-8m Voeltzkow 1902, 565 as Madrepora palifera Lamarck A. spicifera (Dana, 1846) Barnes et al. 1971, 103 TC 1, 5-6, 20, 28m Rosen 197la, 110 Bellamy MS: 1-33m A. tubicinaria (Dana, 1846) Brander et al. 1971, 417, 423, 424, 425 TC less than 5m (sometimes as "Acropora B") A. variabilis (Klunzinger, 1879) Barnes et al. 1971, 103 TC 15, 28, 33m Rosen 197la, 110 Bellamy MS: 13m 7. Astreopora Blainville, 1830 Fryer 1911, 411 SR TC Recorded depth range: 3-7m Astreopora myriophthalma (Lamarck, 1816) Barnes et al. 1971, 101, 102 TC 2-3, 5-6m Rosen 197la, 110 Bellamy MS: 3-7m 8 8. Montipora Quoy & Gaimard in Blainville, 1830 Recorded depth range: surface to 39m Barnes e& al. W971, 97, 103) (as “wontiporarspeca) TE 28,. 85, Sopa on Payer JOM aa! Ke SR Rosen 197la, 110 (Spp "Nos. 1, 2"). Bellamy MS: 6-39m Taylor) LOW alo;, V92 TC SR surface 2Drew 1977, poll.2a,, 365 ) (BRR) Montipora elschneri Vaughan, 1918 Barnes Crate LOW le NOS aC elm Rosen 197la, 110 Bellamy MS: 8m [Montipora erythraea von Marenzeller, 1907] Barnes et al. 1971, 103" TC LO-12,, 20), 22a emu (probably should be M. cf. erythraea as in Rosen 1971la) Montipora cf. erythraea von Marenzeller, 1907 Rosen 197la, 110 Bellamy MS: 8-23m (see also M. erythraea, Barnes et al.) Montipora cf. punctata Bernard, 1897 Barnes Ct al. 1971, p TOs Zhe 12m Rosen 197la, 110 Bellamy MS: 10m [Montipora tuberculosa (Lamarck, 1816) ] Barnes et al. 1971, 102 TC 1-3, 5-8, 10, em (probably should be M. cf. tuberculosa as in Rosen 1971la) Montipora cf. tuberculosa (Lamarck, 1816) Rosen 197la, 110 Bellamy MS: 1-8m (See also M. tuberculosa, Barnes et al.) Montipora verrucosa Lamarck, 1816) Barnes et al. 1971, 102 TC 2-3, 5-6, 10, 17m Rosen 197la, 110 Bellamy MS: 3-15m 9. Pavona (Pavona) Lamarck, 13802 Recorded depth range: 4-27m Pavona (P.) clivosa Verrill, 1869 Rosen 197la, 110 Bellamy MS: 5m Pavona (P.) explanulata (Lamarck, 1816) Barnes et al. 103 Te 12m Rosen 197la, 110 Bellamy MS: 4m Pavona (P.) frondifera Lamarck, 1816 Rosen 197la, 110 Bellamy MS: 7m Pavona (P.) gardineri van der Horst, 1922 Rosen 197la, 110 Bellamy MS: 23m Pavona (P.) cf. minor Brueggemann, 1879 Rosen 197la, 110 Bellamy MS: 6m Pavona (P.) praetorta (Dana, 1846) Rosen 197la, 110 Bellamy MS: 4m Pavona (P.) varians Verrill, 1864 Barnes et al. 1971, 103 TC 10-12, 18, 27m Rosen 197la, 110 Bellamy MS: 8-27m 10. Pavona (Pseudocolumnastraea) Yabe & Sugiyama, 1933 Recorded depth range 18, 20m Barnes et al. 1971, 103 TC 20m Rosen 197la, 111 Bellamy MS: 18m 11. Leptoseris Edwards & Haime, 1849 Recorded depth range 22, 25m Leptoseris columna Yabe & Sugiyama, 1941 Rosen 197la, 111 Bellamy MS: 25m [Leptoseris? mycetoseroides Wells, 1954] Barnes et al. 1971, 97, 103 (as L? mycetoserioides [sic] TC .27,33m Rosen 197la, 111 (This species is a synonym of Agariciella minikoiensis Ma, 1937 according to Wells MS (1977) ), Leptoseris tubulifera Vaughan, 1907 Rosen 197la, 111 Bellamy MS: 22m 12. Agariciella Ma, 1937 Recorded depth range 27, 33m Agariciella minikoiensis Ma, 1937 Barnes et al. 1971, 97, 103 (as Leptoseris? mycetoserioides [sic] Wells, 1954). TC 27,33m Rosen 197la, 111 (as L? mycetoseroides) (L? mycetoseroides Wells 1954 is a synonym of present species according to Wells MS (1977)). 10 [Agariciella ponderosa (Gardiner 1905) ] Barnes et al. 1971, 103, 106 LOR I28;, 933m Rosen 197la, 111 (This species is a synonym of Gardineroseris planulata (Dana, 1846) according to Scheer & Pillai (1974) and Wells MS (1977)). 13. Gardineroseris Scheer & Pillai, 1974 Recorded depth range 28-33m Gardineroseris planulata (Dana, 1846) Barnes et al. 1971, 103, 106 (as Agariciella ponderosa Gardiner, 1905)) TC 28, 33m Bellamy MS: 28-33m Rosen 197la, 111 (as A. ponderosa (Gardiner) ) (Agariciella ponderosa (Gardiner, 1905) is a synonym of present species according to Scheer & Pillai (1974) and Wells MS (1977)). 14. Pachyseris Edwards & Haime, 1849. Recorded depth range: 15, 18, 20-22m Pachyseris laevicollis (Dana, 1846) Barnes et al. 1971, 97, 103, 106 TC 15, 18, 20=22n Rosen 197la, 111 Pachyseris rugosa (Lamarck, 1816) Rosen 197la, 111 [Anomastraea von-Marenzeller, 1901] [Anomastraea (A.) irregularis von Marenzeller, 1901] Barnes et al. 1971. 103 TC (with "?") 10-11m Rosen 197la, 111 (Subsequent examination by BRR and MP of the particular specimen on which these records are based shows that it is actually Coscinaraea columna (Dana, 1846)). 15. Coscinaraea Edwards & Haime, 1848 Recorded depth range: 9-28, 33m Barnes et al. 1971, 103 (as. C. sp. No.1) TC 2ieZan Bellamy MS: 11-27m Rosen 197la, 111 (as C spp. Nos. 1, 2). Bellamy MS: 18m 11 Coscinaraea columna (Dana, 1846) This MS - after re-determination of the specimen on which previous records of Anomastraea (A.) irregularis von Marenzeller, 1901 were based (see above). TC 10-llm Bellamy MS: 9m Coscinaraea monile (Forskal, 1775) Barnes et al. 1971, 103 TC 11, 28, 33m Rosen 1971la, 111 16. Cycloseris Edwards & Haime, 1849 Recorded depth range: 7, 10-llm Cycloseris cyclolites (Lamarck, 1816) Barnes et al. 1971, 103 TC 10-1lm Rosen 1971la, 111 Bellamy MS: 7m Cycloseris distorta (Michelin, 1843) Doderlein 1902, 44, 74, pl.3, figs. dd, ee, ff, pl.5, figs. 3, 3a. (as Fungia distortaf, Cycloseris), Voeltzkow 1902, 565 (as Fungia distorta) . Fungia Lamarck, 1801 Recorded depth range: 3-12, Drew 1977, 3 15, 17, 19-20m [Fungia distorta Michelin, 1843] Voeltzkow 1902, (see Cycloseris distorta, above) [Fungia distorta Michelin, 1843 £. Cycloseris] Doderlein 1902 (see Cycloseris distorta, above) [Fungia fungites (Linnaeus, 1758) var. confertifolia Dana, 1846 Doderlein 1902, 44, 155, pl.23, figs 2, 2a Voeltzkow 1902, 565 (see Fungia (F.) fungites, below) [Fungia scutaria Lamarck, 1801 var. placunaria Klunzinger, 1879] Doderlein 1902, 44, 93, 96 (see under Fungia (Pleuractis), below) [Fungia scutaria Lamarck, 1801 var tenuidens Quelch, 1886] Voeltzkow, 1902, 565 (see under Fungia (Pleuractis) below) 17. Fungia (Danafungia) Wells, 1966 Recorded depth range: 19-20m Rosen 197la, 111 Bellamy MS: 19-20m 12 18. Ie) 20. 21. 228 Fungia (Fungia) Lamarck, 1801 Recorded depth range: no information Doderlein 1902, 44, 155, pl.23, figs 2, 2a (as F. fungites var. confertifolia). Voeltzkow 1902, 565 (as F. fungites var. confertifolia) Fungia (Pleuractis) Verrill, 1864 Recorded depth range: 3-12, IS besa Fungia (Pleuractis) paumotensis Stutchbury, 1833 Barnes etal. 1971, 102 TC. 3795, “Opel2ewom Rosen 197la, 111 Bellamy MS: 3-12m Fungia (Pleuractis) scutaria Lamarck, 1801 Barnes eé al; JQ71, 102 °2C13, 5, Lop ym Déderlein 1902, 44, 93, 96 (as F. scutaria var. placunaria) Rosen 197la, 111 Bellamy MS: 6-10m Voeltzkow 1902, 565 (as F. scutaria var. tenuidens) Herpolitha Eschscholtz, 1826 Recorded depth range: 20-23, 27m Herpolitha limax (Esper, 1797) Barnesneét)al 991971; 103 ,clO06(.TeC 212, 27m Rosen 197la, 111 Bellamy MS: 20-23m Podabacia Edwards & Haime, 1849 Recorded depth range: 18, 22, 27m Barnes»etsal.. 19715597, <103,.106 TC 227) 27m Rosen 197la, 111 Bellamy MS: 18m Goniopora Blainville, 1830 Recorded depth range: lm Goniopora stokesi Edwards & Haime, 1860 Rosen & Taylor, 1969 TC 1m Porites Link, 1807 Recorded depth range: surface to 22m Brander et al. 1971, 417, 424 TC less than 5m Fryer 194), 41079411 9S TEC Ma 1958, 10 TC Mal 9597, 33 TC Price W97il 166 SR Te surface Taylor 1971b, 181, 186, 188, 190-2, SR TC surface 13 23. Porites (Porites) Link, 1807 Recorded depth range: surface to 22m Porites (P.) andrewsi Vaughan, 1918 Barnes et al. 1971, 101, 102 TC 2, 5-6, 8m Rosen 197la, 111 Bellamy MS: 3-8m Porites (P.) lichen Dana, 1846 Barnes et al. 1971, 97, 103, 106 TC 20, 22m Rosen 197la, 111 Bellamy MS: 18m Porites (P.) lutea Edwards & Haime, 1851 Porites lutea Edwards & Haime, 1851 Barnes et al. 1971, 102 TC 2-3, 5-7m Price 1971, 166, 169 SR TC surface Rosen 197la, 111 Bellamy MS: 2-7m Taylor 1971b, 183 SR TC surface Voeltzkow 1902, 565 Porites (P.) nigrescens Dana, 1846 Porites nigrescens Dana, 1846 Barnes et al. 1971, 102 TC 5, 8-9, 11-12, 15, 17-18m Brander et al. 1971, 423 TC less than 5m Rosen 197la, 111 Bellamy MS: 2-22m Taylor 1971b, 192 SR TC surface 24. Porites (Synaraea) Verrill, 1864 Recorded depth range: 1m Porites (Synaraea) iwayamaensis Eguchi, 1938 Barnes et al. 1971, 103 TC 1m Rosen 197la, 111 25. Plesiastrea Edwards & Haime, 1848 Recorded depth range: 27-28, 33m Barnes et al. 1971, 103, 106 TC 28, 33m Rosen 197la, 111 Bellamy MS: 27m Plesiastrea versipora (Lamarck, 1816) Rosen 197la, 112 Bellamy MS: 27m 26. Favia Oken, 1815 Recorded depth range: surface to 33m Barnes et al. 1971, 103 TC 7m Fryer 1911, 411 SR TC Price 1971, 150, 166 SR TC surface Rosen 197la, 112 Taylor 1971b, 183, 186, SR TC surface 14 [Favia acropora (Linnaeus, 1767) ] Matthai 1914, 102 (1 specimen) (Matthai's F. acropora is a synonym of F. stelligera (Dana, 1846); see Vaughan 1918, 101). [Favia bertholleti Valenciennes MS] Ma 1958, 12 TC Ma 1959, 42 BM(NH) Reg. Nos. 1927.5.12.6, and 1927.5.12.49 [Matthai's material] TC Matthai 1914, 95 pl.24, fig.1 (2 specimens, above) (Matthai's (and hence Ma's) F. bertholleti is a synonym F. favus (Forsk&al, 1775); see Rosen (1968, 343)). Favia favus (ForskAl, Abe / 7/53) Rosen 1968, 330 (F. bertholleti of Matthai (1914) BM(NH) Reg. Nos. 1927.5.12.6, and 1927.5.12.49) [Favia halicora (Ehrenberg, 1834) ] Matthai 1914, 106 (2 specimens) (Matthai's F. halicora is a synonym of Favites abdita (Ellis & Solander, 1786); see Wijsman-Best (1972, 33)). Favia pallida (Dana, 1846) Barnes et al. 1971, 102 TC 2,5-6,9,11=-12,15,27—ts, 20-22,28, 33m Rosen 197la, 112 Bellamy MS: 2-33m Favia stelligera (Dana, 1846) Barnes et al. 1971, 103 TC 9m Matthai 1914, 102 (as F. acropora) (1 specimen) Rosen 197la, 112 [Favia vasta (Klunzinger 1879) ] Matthai 1914, 108, pl.27 fig.6 (1 specimen) (Matthai's F. vasta is a synonym of Favites virens (Dana, 1846); see Vaughan (1918, 111); but see also Chevalier (1971, 229) who regards Matthai's F. vasta as Favites vasta). 27. Favites Link, 1807 Recorded depth range: surface to 21m [Prionastrea Edwards & Haime 1857] Barnes et al. 1971, 103 (as "F. sp." following a Favites species, but should probably be Favia sp. as in Rosen 197la, 112) TC 6m. Fryer 1911, 411 (as Prionastrea) SR i Favites abdita (Ellis & Solander, 1786) Barnes et al. 1971, 102 SC (2-3, 5-8, 10-1) im 15 Matthai 1914, 106 (as Favia halicora (Ehrenberg, 1834) (2 specimens) Rosen 197la, 112 Bellamy MS: 2-33m Favites cf. pentagona (Esper, 1794) Barnes et al. 1971, 103 TC 3, 7, 21m Rosen 197la, 112 Bellamy MS: 3-21m Favites virens (Dana, 1846) [Favites vasta (Klunzinger, 1879) ] Matthai 1914, 108, pl.27, fig.6 (as Favia vasta (Klunzinger, 1879. See under Favia vasta, above) (1 specimen). 28. Goniastrea Edwards & Haime, 1848 Recorded depth range: surface, 9-33m Taylor 1971b, 183, 188, 190 SR TC surface [Goniastrea incrustans Duncan, 1889] Barnes et al. 1971, 97, 103 TC 27, 33m Rosen 197la, 112 Bellamy MS: 23-33m (The specimen on which these records are based was mistakenly identified. It should be Goniastrea palauensis (Yabe & Sugiyama, 1936)). Goniastrea palauensis (Yabe & Sugiyama, 1936) Barnes et al. 1971, 97,103 (as G. incrustans) TC 27, 33m Rosen 197la, 112 (as G. incrustans) Bellamy MS: 23-33m Goniastrea pectinata (Ehrenberg, 1834) Barnes et al. 1971, 102 TC 9-11,15,18, 20-22, 27, 33m Price 1971, 166 SR TC surface Rosen 197la, 112 Bellamy MS: 9-33m Goniastrea retiformis (Lamarck, 1816) Barnes et al. 1971, 103 TC 11m Ma 1958, 14 Ma 1959, 48 B.M.(N.H.) Reg. No. 1927.5.4.186 [Matthai's material] TC Matthai 1914, 118 (2 specimens) Rosen 197la, 112 Bellamy MS: 11-19m 29. Platygyra Ehrenberg, 1834 Recorded depth range: surface to 33m [Coeloria, Edwards & Haime, 1848] Fryer 1911, 411 (as Coeloria). SR TC Price 1971, 166 SR TC surface Platygyra astreiformis (Edwards & Haime, 1849) Ma 1958, 15 TC Ma 1959, 53 BM(NH) Reg. No. 1928.4.18.599 "Eldabra" TC 16 Platygyra daedalea (Ellis & Solander, 1786) Ma 1958, 15 TE Matthai 1928, 24 (as Coeloria daedalea) (2 specimens) Platygyra lamellina Ehrenberg, 1834 Barnes et al. 1971, 102 TC 2-3,5-7,9-10,12,15,17-18, 20-21, 33m Rosen 197la, 112 Bellamy MS: 2-33m [Platygyra phrygia (Ellis & Solander, 1786) ] Matthai 1928, 112 2 specimens (Matthai's "Platygyra" is actually Leptoria; see Crossland (95277 ASO) 30. Leptoria Edwards & Haime, 1848 Recorded depth range: surface [Platygyra, Matthai 1928, only] to 12m Brander et al. 1971, 417, 423-5 TC less than 5m Price 1971 54166 SR TC surface Leptoria phrygia (Ellis & Solander, 1786) Bamnes \e& als 19717) 102) tC 25 eSa od Aa Matthai 1928, 112 (as Platygyra phrygia (Ellis & Solander, 1786)) (2 specimens) Rosen 197la, 112 Bellamy MS: 2-12m 31. Hydnophora Fischer, 1807 Recorded depth range: surface to 3m, 15-33m Fryer 1911, 411 SR. TC [Hydnophora contignatio (Forskal, 1775) ] Matthai 1928, 155, pl.46, fig.2 (1 specimen, BM (NH) Reg. No. 1928,3.1.56). (H. contignatio is a synonym of H. exesa (Pallas, 1766); see Wijsman-Best, 1972, 51) Hydnophora exesa (Pallas, 1766) Barnes et.al. 1971, lO3*TC. 2?2,,15,) 27-28, 35 thee species appears twice in the table on p.103, the second time at 2m. Is this second record meant to be H. microconos?) Matthai 1928, 140 (2 specimens) Matthai 1928, 155, pl.46, fig.2, (as H. contignatio (1 specimen, BM(NH) Reg. No. 1928.3.1.56) Rosen 197la, 112 Bellamy MS: 15-33m Hydnophora microconos (Lamarck, 1816) ?Barnes et al. 1971, 103 (see under H. exesa, above) Price 1971, 166 SR TC surface TC 2m Rosen 197la, 112 Bellamy MS: 2-3m IL7/ 32. Diploastrea Matthai, 1914 Recorded depth range: no information Diploastrea heliopora (Lamarck, 1816) Rosen 197la, 112 33. Leptastrea Edwards & Haime, 1848 Recorded depth range: 2-22m Leptastrea immersa Klunzinger, 1879 Barnes et al. 1971, 103 TC 2m Rosen 197la, 112 Bellamy MS: 2-1lm Leptastrea purpurea (Dana, 1846) Barnes et al. 1971, 102 TC 9, 11-12, 20, 22m Rosen 197la, 112 Bellamy MS: 9-22m 34. Cyphastrea Edwards & Haime, 1848 Recorded depth range: surface to 33m Taylor 1971b, 183, 186, 192 SR TC surface Cyphastrea chalcidicum (Forskal, 1775) Barnes et al. 1971, 103 TC 3, 7, 21, 28m Rosen 197la, 112 Bellamy MS: 3-33m 35. Echinopora Lamarck, 1816 Recorded depth range: 2-39m Echinopora gemmacea (Lamarck, 1801) Barnes et al. 1971, 102 TC 2, 5-6, 9, ll, 15, 18, DOA=22, By Bsa Rosen 197la, 112 Bellamy MS: 2-39m 36. Oulangia Edwards & Haime, 1848 Recorded depth range: 18m Rosen 197la, 112 Bellamy MS: 18m [Madrepora Linnaeus, 1758] Fryer 1911, 410, 411 (this record is almost certainly Acropora) . SR TC 37. Galaxea Oken, 1815 Recorded depth range: surface to 13m, 15m Fryer 1911, 411 SR TC Galaxea fascicularis (Linnaeus, 1767). Recorded depth range: surface to 13m, 15m Barnes et al. 1971, 102 TC 2-3,5,7,10,15m Matthai 1914, 59 (6 specimens) Rosen 197la, 112 Bellamy MS: 2-13m 18 38. Blastomussa Wells 1968 Recorded depth range: 24m Blastomussa merleti Wells, 1961 Rosen 197la, 112 Bellamy MS: 24m 39. Acanthastrea Edwards & Haime, 1848 Recorded depth range: surface, 20-23) 27), 3am Fryer 1911, 411 SR Abs Price 1971, 166 SR TC surface Acanthastrea echinata (Dana, 1846) Barnes et al. 1971, 103 Te “2227 2p Rosen 197la, 112 Bellamy MS: 20-23m 40. Lobophyllia Blainville, 1830 Recorded depth range: 5-23, 27-28, 35, 39,43m Lobophyllia corymbosa (Forskal, 1775) Barnes et als 97, OZ 51 Te) 15-6, 10,15 ,27m Rosen 197la, 113 Bellamy MS: 5-23m Lobophyllia hemprichii (Ehrenberg, 1834) Barnes et al. 1971, 103 TEeO2Qy, 27-28, 35, 59 ,4om Rosen 197la, 113 41. Symphyllia Edwards & Haime, 1848 Recorded depth range: 20-21, 27m Symphyllia nobilis (Dana, 1846) Rosen 197la, 113 Bellamy MS: 20m Symphyllia valenciennesii Edwards & Haime, 1849 Barnes et al. 1971, 103; 106 Te 21,27m Rosen 197la, 113 Bellamy MS: 20m 42. Echinophyllia Klunzinger, 1879 Recorded depth range: 18-33m Echinophyllia aspera (Ellis & Solander, 1786) Barnes et al. 19 7ie os, WOE fe 20 2m Rosen 197la, 113 Bellamy MS: 18-33m 43. Mycedium Oken, 1815 Recorded depth range: 18-33m Mycedium tenuicostatum Verrill, 1901 Barnes et -al..« 1974;).1039eTe 18,22,28m Rosen 197la, 113 Bellamy MS: 18-33m 19 Mycedium tubifex (Dana, 1846) Barnes et al. 1971, 103 TC 18,21,28,33m Rosen 197la, 113 Bellamy MS: 18-27m 44, ?Physophyllia Duncan, 1884 Recorded depth range: 26m ?Physophyllia ayleni Wells, 1934 Rosen 197la, 113 Bellamy MS: 26m 45. Pectinia Oken, 1815 Recorded depth range: 12,18, 20, 26m Barnes et al. 1971, 103 TC 27m Rosen 197la, 113 Pectinia lactuca (Pallas, 1766) Barnes et al. 1971, 103 TC 12, 18m Rosen 197la, 113 Bellamy MS: 20m 46. Paracyathus Edwards & Haime, 1848 Recorded depth range: surface Taylor 1971b, 192 SR TC surface 47. Physogyra Quelch, 1884 Recorded depth range: 10-28, 33m Barnes et al. 1971, 103 TC 10, 12, 18,20-22,28, 33m Rosen 197la, 113 Bellamy MS: 10-28m 48. ?Gyrosmilia Edwards & Haime, 1851 Recorded depth range: 15,17- 33m ?Gyrosmilia interrupta (Ehrenberg, 1834) Barnes et al. 1971, 103 TC 17-18, 27m Rosen 197la, 113 Bellamy MS: 15, 18-33m 49. Dendrophyllia Blainville 1830 Recorded depth range: surface to 5m Barnes et al. 1971, 93 106 TC Bellamy et al. 1969, 103 SR aXC Brander et al. 1971, 400 TC less than 5m Fryer 1911, 411 SR AS Taylor 1971b, 192, 204 SR TC surface (Some of these records are probably Tubastrea) Dendrophyllia cf. florulenta van der Horst 1922 Rosen 197la, 113 Bellamy MS: surface 20 [Dendrophyllia micrantha (Ehrenberg, 1834) ] Barnes et al. 1971, 94,97,103 TE 39, 41-44m Rosen 197la, 113 Bellamy MS: surface to 44m + (This species is now regarded as a Tubastraea) 50. Tubastraea Lesson, 1834 Recorded depth range: surface [Dendrophyllia, in part] to 44m + taylor W971i; eLIO,, 192 SR TC surface see also Dendrophyllia, above Tubastraea micrantha (Ehrenberg, 1834) Barnes et al. 1971, 94, 97, 103 (as Dendrophyllia micrantha) TC 39, 41-44m Rosen 197la, 113 (as D. micrantha), Bellamy MS: surface to 44m + 51. Turbinaria Oken, 1815 Recorded depth range: 20-33m Barnes et al. 1971, 103 Leh 2158 Zi OS Rosen 197la, 113 Bellamy MS: 20-33m 52. Tubipora Linnaeus, 1758 Recorded depth range: surface to lm Eryer” 190) 420) SR? re Tubipora musica Linnaeus, 1758 Barnes et al. 1971, 103 TC lm Rosen 197la, 113 Bellamy MS: surface 53. Heliopora Blainville, 1830 Recorded depth range: surface Fryer 1911, 410 SR TC Stoddart 1967, 17, 18 SR TC (as sediment component) Stoddart et al. 1971, 50 SR TC (as sediment component) Heliopora coerulea (Pallas, 1766) Rosen 197la, 113 Bellamy MS: surface Voeltzkow 1902, 565 54. Millepora Linnaeus, 1758 Recorded depth range: surface to 12m Brander et al. 1971, 417, 423, 424 TC less than 5m Drew 1977, pl.la, 3a (BRR) Fryer 1911, 410-412 SR TC Price 1971, 166, 169 SR TC surface Taylor 197ib, 192 SR TC surface 21 [Millepora cf. clavaria Ehrenberg, 1834] Voeltzkow 1902, 565 (M. clavaria is a synonym of M. exaesa Forskal, 1775; see Boschma (1948, 28)). Millepora exaesa Forskal, 1775 Barnes et al. 1971, 101, 102, 107 TC 3,5-7,9m Rosen 197la, 114 Bellamy MS: 2-9m ?Voeltzkow 1902, 565 (as M. cf£. clavaria). Millepora platyphylla Hemprich & Ehrenberg, 1834 Barnes et al. 1971, 102 TC 1-2,5-6,12m Brander et al. 1971, 423,425 TC less than 5m Rosen 197la, 114 Bellamy MS: 1-llm Taylor 1971b, 183 SR TC surface 55. Distichopora Lamarck, 1816 Recorded depth range: surface to 2m Distichopora fisheri Broch, 1942 Barnes et al. 1971, 103 AK Dyan Rosen 197la, 114 (forma alpha Wells, 1954) Bellamy MS: surface Distichopora violacea (Pallas, 1766) Rosen 197la, 114 Bellamy MS: surface BIBLIOGRAPHY Works preceded by "ac" contain references to Aldabra corals by genus or species name, or provide photographs from which such names may be inferred. These works form the basis for the present check list. All authors of a single work have been given entries for "ac" references. ac Barnes, J., Bellamy, D.J., Jones, D.J., Whitton, B.A., Drew, E.A., Kenyon, L., Lythgoe, J.N. & Rosen, B.R. 1971. Morphology and ecology of the reef front of Aldabra. In D.R. Stoddart & C.M. Yonge (Eds.) 1971. Regional variation in Indian Ocean coral reefs. Symp.zool.Soc.Lond. 28, 87-114. ac Bellamy, D.J. 1971 (see Barnes, J. et al. 1971). ac Bellamy, D., Drew, E., Jones, D. & Lythgoe, J. 1969. Aldabra: a preliminary report of Phase VI of the Royal Society Expedition. Rep. Underwater Ass. (1969), 100-104. Boschma, H. 1948. The species problem in Millepora. Zool.Verh., Leiden 1, 1-115, pls. 1-15. 22 ac ac ac ac ac ac ac ac ac ac ac Brander, K.M., McLeod, A.A.Q.R. & Humphries, W.F., 1971. Comparison of species diversity and ecology of reef-living invertebrates on Aldabra atoll and at Watamu, Kenya. In D.R. Stoddart & C.M. Yonge (Eds.) 1971. Regional variation in Indian Ocean coral reefs. Symp.zool.Soc.Lond. 28, 397-431. Chevalier, J.-P. 1971. Les scléractiniaires de la Mélanésie frangaise (Nouvelle Calédonie, Iles Chesterfield, Iles Loyauté, Nouvelles Hébrides): Premiere partie. Expédition frangaise sur les récifs coralliens de la Nouvelle-Calédonie organisée sous 1'égide de la Fondation Singer-Polignac 1960- PIGS Volume Cinquiéme. Editions de la Fondation Singer- Polignac: Paris. Chevalier, J.-P. 1975. Les scléractiniaires de la Mélanésie frangaise (Nouvelle Calédonie, Iles Chesterfield, Iles Loyauté, Nouvelles Hébrides): Deuxiéme partie. Expédition frangaise sur les récifs coralliens de la Nouvelle-Calédonie organisée sous l1'égide de la Fondation Singer-Polignac 1960-1963. Volume Septiéme. Editions de la Fondation Singer-Polignac: Paris. Crossland, C. 1952. Madreporaria, Hydrocorallinae, Heliopora and Tubipora. Scient.Rep.Gt.Barrier Reef Exped. 6, 85-257, pis. L=56. Déderlein, L. 1902. Die Korallen-Gattung Fungia. Abh.senckenb. naturforsch.Ges. 27(3), 1-162, pls. 1-25. Drew, E.A. 1969. (see Bellamy, D.,et al. 1969). Drew, E.A. 1971 (see Barnes, J., et al. 1971). Drew, E.A. 1977. A photographic survey down the seaward reef- front of Aldabra atoll, Atoll Res. Bull. 193, 7pp., 3 pls. Farrow, G.E. 1971. (see Stoddart, D.R., et al. 1971). Fosberg, F.R. 1971. (see Stoddart, D.R., et al. 1971). Fryer, J.C.F. 1911. The structure and formation of Aldabra and neighbouring islands - with notes on their flora and fauna. Trans.Linn.Soc.Lond.2nd.Ser.Zoology 14, 397-442, pls.22-29. Humphries, W.F. 1971. (see Brander, K.M.,et al. 1971). Jones, D. 1969. (see Bellamy, D.,et al. 1969). Jones, D.J. 1971. (see Barnes, J., et al. 1971)-. Kenyon, L. 1971. (see Barnes, J., et al. 1971). ac ac ac ac ac ac ac ac ac ac ac 23 Lythgoe, J. 1969. (see Bellamy, D., et al. 1969). Lythgoe, J. 1971. (see Barnes, J., et al. 1971). Ma, T.Y.H. 1958. The relation of growth rate of reef corals to surface temperature of sea water as basis for study of causes of diastrophisms instigating evolution of life. Res.past Clim. contin. Drift., Taipei 14, 1-60, pls. 1-24. Ma, T.Y.H. 1959. Effect of water temperature on growth rate of reef corals. Oceanographica Sinica Spec. Vol. 1, i-v, 1-116, pls. A, 1-320. Matthai, G. 1914. A revision of the Recent colonial Astraeidae possessing distinct corallites. (Based on material from the Indo-Pacific Ocean and the collections of Paris, Berlin, Vienna, Copenhagen, London and Glasgow). Trans.Linn.Soc. London 2nd. Ser.Zoology 17, 1-140, pls. 1-38. Matthai, G. 1928. Catalogue of the madreporarian corals in the British Museum (Natural History). Volume VII, a monograph of the Recent meandroid Astraeidae. Trustees of the British Museum (Natural History): London. McCleod, A.A.Q.R. 1971. (see Brander, K.M., et al. 1971). Peters, A.J. & Lionnet, J.F.G. 1973. Central western Indian Ocean bibliography. Atoll Res. Bull. 165, 322pp. Price, J.H. 1971. The shallow sublittoral marine ecology of Aldabra. Phil. Trans. R.Soc. Ser.B. 260, 123-171, pls. 11-14. Rosen, B.R. 1968. An account of a pathologic structure in the Faviidae (Anthozoa): a revision of Favia valenciennesii (Edwards & Haime) and its allies. Bull. Br.Mus.nat.Hist. Zoology 16(8), 323-352, pls. 1-8. Rosen, B.R. 197la. Provisional check list of corals collected during the Royal Society Expedition to Aldabra, phase 6. In D.R. Stoddart & C.M. Yonge (Eds.) 1971. Regional variation in Indian Ocean coral reefs. Symp.zool.Soc.Lond. 28, 87-114. (see also Barnes, J., et al., 1971). Rosen, B.R. 1971b. The distribution of reef coral genera in the Indian Ocean. In D.R. Stoddart & C.M. Yonge (Eds.) 1971. Regional variation in Indian Ocean coral reefs. Symp.zool. Soc.Lond. 28, 263-299. Rosen, B.R. & Taylor, J.D. 1969. Reef coral from Aldabra: new mode of reproduction. Science, N.Y. 166, 119-121. 24 ac ac ac ac ac ac Scheer, G. & Pillai, C.S.G. 1974. Report on the Scleractinia from the Nicobar Islands. Zoologica, Stuttg. 42(3)122, 1-75, pls. 1-33. Stoddart, D.R. 1971. Scientific studies at Aldabra and neighbouring islands. Phil.Trans.R.Soc. Ser.B. 260, 5-29, pie ely Stoddart, D.R., Taylor, J.D., Fosberg, F.R. & Farrow, G.E. 1971. Geomorphology Aldabra atoll. Phil.Trans.R.Soc. Ser.B. 260, 31-65, pls. 2-9. Stoddart, D.R. & Wright, C.A. 1967. Geography and ecology of Aldabra atoll. In D.R. Stoddart (Ed.) 1967. Ecology of Aldabra atoll, Indian Ocean. Atoll Res. Bull. 118, 11-52. Taylor, J.D. 1969. (see Rosen, B.R. & Taylor, J.D. 1969). Taylor, Ji-Dy L97La. (see Stoddart, D.R., et al. 1971). Taylor, J.D. 1971b. Intertidal zonation at Aldabra atoll. Phil.Trans.R.Soc. Ser.B. 260, 173-213, pl. 15. Vaughan, T.W. 1918. Some shoal-water corals from Murray Island (Australia), Cocos-Keeling Islands and Fanning Island. Pap.Dep.mar.Biol.Carnegie Instn Wash. 9, 51-234, pls. 20-93 (Publs Carnegie Instn 213). Voeltzkow, A. 1902. Die von Aldabra bis jetzt bekannte Flora und Fauna. Abh.senckenb.naturforsch.Ges. 26, 461-537. Wells, J.W. 1977. Unpublished MS on Leptoseris, Agariciella and Gardineroseris. Whitton, B.A. 1971. (see Barnes, J., et al. 1971). Wijsman-Best, M. 1972. Systematics and ecology of New Caledonian Faviinae (Coelenterata-Scleractinia). Bi jar. Dierk 2A2@), 1-90) \pisiea=14e2 Wright, C.A. 1967. (see Stoddart, D.R. & Wright, C.A. 1967.) Table 1 Collections and studies of corals of Aldabra Abbreviations: BM(NH)DZ - British Museum (Natural COLLECTION/EXPEDITION Voeltzkow History), Department of Zoology. —_ 7 7 ' SMEM Station Marine d'Endoume, DATE Marseille UDDB - University of Durham, Department of Botany SR - Sight records COLLECTION CURRENTLY DEPOSITED AT:- The number of identifications is based on those actually listed by individual works, no allowance being made for validity, synonyms, etc. COMPLETENESS OF STUDY OF MATERIAL Identifications presumed in- complete "‘Identifications' here includes all levels of taxonomic study. GENERA IDENTIFIED 5 era Ti SPECIES IDENTIFIED 7 AUTHORS STATUS OF IDENTIFICATIONS GENERR Sica if ’ IDENTIFIED | IDENTIFIED Barnes et al. 1971 Ecological account using names in Rosen 197la ie 43 78) al Bellamy et al. 1969 General account. Sight records in text 2 = Brander et al. 1971 Ecological account, records in text [ 7 7 Déderlein 1902 Monograph on the genus Fungia iL 3 xX Drew 1977 Discussion of transect method. Corals named from plates by BRR in present list 5 1 Fryer 1911 General account. Sight records in text 17 Ma 1958 Coral growth account based on many regions. Lists of 4 specimens for numerous = Ma 1959 species 4 4 Matthai 1914 Matthai 1928 Pichon & Rosen (in prep.) Monograph of faviid species Monograph of meandrine coral species Check list with remarks Price 1971 Rosen 1968 Ecological account. Sight records in text Taxonomic revision of some faviid genera Rosen 197la ——|-----_ Check list (Rosen 1971b, Table 1, line e) Generic list for Aldabra- Glorioso group Rosen & Taylor 1969 Reproduction in Goniopora, at Aldabra : F Stoddart & Wright 1967 General account. Sight record in text il = Stoddart et al. 1971 General account. Sight records in text 1 Taylor 1971 Ecological account. Sight records in text 9 3 Voeltzkow 1902 Check list, partly after Déderlein 1902 5 7 xX Cherbonnier ROYAL SOCIETY EXPEDITION TO ALDABRA Fryer-Sladen ercaee oan 1908-9 1954 1966 1967 1967 1968 1968-9 DZ Identifications | Identifications | Identifications Identifications not yet Identifications of only a small in hand not yet made on | made on collected material complete collected except Phase V in part material (x) ATOLL RESEARCH BULLETIN NO. 234 RECIFS CORALLIENS, CONSTRUCTIONS ALGUAIRES, ET ARRIECIFES A LA GUADELOUPE, MARIE GALANTE ET LA DESIRADE par R. Battistini et M. Petit Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 MIEROO AVELLIASOO Baa AU IOISA IO FIT Ca, De eG +1 54 SIMA AO Ae ri booted a COM SI) Va Oe eiite SET 03 3.0 .sobgalden Wo” —— CTCL eodueaeol Fig. Fig. Fig. Fig. Fig. Fig. LISTE DES FIGURES Croquis de situation des principales constructions littorales de Grande-Terre et Guadeloupe Nord. Température de l'eau de surface, houle et vent. La Désirade 1971-1974. Grés de plage formant des arrécifes a 1'Est de Moule (Grande-Terre) . Marie Galante et Désirade. "Récifs en crochet" de Petite Anse et "récifs en S" de Facheux a Marie Galante. Coupe de la Caye au Vent a la plage (Marie Salante). MBANOTS 290 STERY « belerot?il anoldovtdanns sefeqicaizq éab aclsayiia 66 FE Sto eqveisberoD 46 esseTt-obee | ofatictht af oarev 3 wt ,wastive sb une"! of stusetkas prer-feg eM af 7? “us5?) ' ineoro) epaliqg eh saue T16T -abaa xe) RECIFS CORALLIENS, CONSTRUCTIONS ALGUAIRES, ET ARRIECIFES A LA GUADELOUPE, MARIE GALANTE ET LA DESIRADE par R. Battistini et M. Petit Certaines régions des Caraibes possédent une vie corallienne florissante : ainsi sur le Banc d'Argent en face de Porto-Rico, dans les Iles Vierges, ou au large des cdétes du Yucatan. Si nous considérons les "petites Antilles", et plus spécialement les iles frangaises, on constate le peu d'intérét que ces formations ont suscité jusqu'a présent. Le but de cet article! est de faire un court inventaire des récifs de la Guadeloupe et de ses deux dépendances, Marie Galante et la Désirade, et surtout de dégager les caractéres trés originaux des formes que l'on y observe. La morphologie des "récifs" guadeloupéens résulte en effet de la combinaison, a des degrés variables, de trois types de formations tout a fait différentes : ie les coraux proprement dits 7Xe les constructions alguaires 3 les arrécifes. L'exposition joue un réle déterminant dans la répartition des récifs, et dans leur morphologie. Dans la région considérée, les vents dominants viennent des secteurs NE, E et SE (voir la figure n° 2), les cétes W et NW étant au contraire en position d'abri. Ainsi que cela est habituel, c'est dans les secteurs les plus exposés que les constructions alguaires prennent leur développement Maximum : ainsi sur la cdte orientale de la Grande Terre de la Guadeloupe, sur la cdte orientale et méridionale de Marie Galante, et sur la c6te sud-est de la Désirade. On s'attendrait a trouver dans ces secteurs exposés de larges platures coralliennes émergeant légérement a basses mers, comme c'est le cas par exemple dans 1'Océan 1 Mission effectnée dans le cadre de 1'E.R.A. S45uGdu, (CNRS. avec l'aide de 1'U.E.R. de Lettres et Sciences Humaines de l'Université Antilles-Guyane. (Manuscript received April 1976 -- Eds.) 2 Indien : or il n'en est rien. Seuls subaffleurent d'étroits segments de constructions alguaires, larges de quelques métres, sur lesquels brisent les grands déferlements. Paradoxalement, c'est en situation relativement abritée, dans le Petit et le Grand Culs de Sac Marin, que l'on trouve les éléments de platures coralliennes de beaucoup les plus étendus, en face de Goyave et de Petit Bourg, et dans la barriére qui ferme le Grand Cul de Sac. Sur la céte orientale de la Grande Terre de la Guadeloupe (secteur de Moule en particulier), coraux et algues calcaires ne font, sans doute, qu'habiller un systéme d'arrécifes, constitué par 4 45 alignements d'anciens grés de plages décollés du rivage, et donnant autant de lignes de déferlements. On trouve donc ici une morphologie comparable 4 celle décrite par Ottmann (1960 et 1963) et Laborel (1965) sur la cdte nord-est du Brésil. Les cétes occidentales et méridionales de Basse-Terre ne possédent pas de récifs coralliens, sans qu'on en sache la raison, a l'exception d'un récif frangeant embryonnaire a 1'Ilet Pigeon. LA GUADELOUPE La céte orientale de la Grande Terre est ourlée de bancs discontinus, étroits, donnant autant de brisants. A 1'Anse Maurice, au Moule, a 1'Anse 4 1"Eau ainsi qu'aé la Gourde, apparafit l'importance de telles constructions linéaires et étroites, de nature essentiellement alguaire. Les algues calcaires qui les construisent sont d'une étonnant vitalité et d'une grande diversité. Ces éléments massifs, longs de plusieurs métres, sont séparés entre eux par des sillons perpendiculaires a la céte : il ne s'agit donc en aucun cas de véritables barriéres continues. Ces constructions alguaires incluent de nombreux patés coralliens morts ou vivants. Les principales espéces répertoriées (Porites astreoides, Favia, Diploria, Montastrea et Siderastrea) sont essentiellement massives ; rares sont les formes tubulaires, branchues ou palmées du type Acropora!. Le long du rivage, il existe a4 peu prés partout une ou deux lignes de grés de plage affleurant a basse mer, présentant la disposition classique en micro-cuestas hautes de quelques dizaines de centimétres a front tourné vers la plage. Certains de ces bancs de grés décollent complétement du rivage, comme 4 1'Anse de la Gourde, ou 4 1'Est de Moule, et se perdent vers le large (fig. 3). 1 Nous remercions trés vivement Monsieur J.P. Chassaing, de 1'INRA de Guadeloupe, qui a bien voulu nous initier a la détermination des principales espéces coralliennes et qui nous a accompagné lors de certaines plongées. Nous pensons que plus au large, les lignes de brisants successives correspondent a de telles lignes de grés de plage encore davantage décollées, qui ont servi de support aux constructions coralliennes et alguaires. C'est dans le secteur compris entre Moule et la Pointe de la Couronne qu'apparait le mieux la réalité de ces "arrécifes" (photographies aériennes verticales, mission 006-100 n°S 293 A 296). L'aspect rigide, rectiligne, des différentes lignes de brisants décalées les unes par rapport aux autres, ne peut avoir d'autre origine. Le littoral méridional de la Grande Terre est plus simple, sans succession rythmique de bancs a l'entrée des anses. Si les constructions demeurent 6étroites et parfois plus localisées, leur nature change radicalement. Nous sommes ici en présence de récifs frangeants classiques quasiment morts. Le platier, de quelques dizaines de métres de largeur maximum, est constitué presque exclusivement de grandes feuilles ou de fragments d'Acropora palmata et de débris de Porites porites Pallas sur lesquels végétent quelques touffes de corail vivant parmi lesquels dominent toujours 1'Acropora paimata mais aussi des formes branchues, fragiles, telles que 1'Acropora prolifera, les Porites porites Pallas, et les Millepora. Comme sur la c6te Est, la faible profondeur exclut 1'existence de tombant et permet un ennoiement rapide de la construction sous ses propres débris sableux. Un chenal d'embarcations, d'une centaine de métres de largeur, peut passer a un véritable lagon, comme au droit de Saint Fran¢ois ou de Sainte Anne, couvert par une 6troite prairie a Cymodocées. Enfin la plage révéle de rares grésifications, généralement tendres et de faible développement comme aux Raisins Clairs, 4 l'entrée de Saint Frang¢ois, ou a l'anse du Mancenillier plus a 1'Est. La céte Ouest de laGrande Terre rappelle, par bien des aspects, le littoral méridional : les récifs frangeants s'étendent de 1'Anse Laborde au Nord jusqu'’a Port Louis au Sud. La proportion de corail mort est toujours trés importante et les espéces peu variées avec predominance d'Acropora palmata et Porites porites. Actuellement, la vie corallienne serait plus active que sur le littoral méridional ; 1l'Acropora palmata domine, ainsi que la forét naine de Millepora complanata et les énormes Diploria strigosa et D.clivosa. Notons la déformation des Acropora palmata, fortement dissymétriques, le plateau se développant essentiellement vers l'intérieur, perdant leur belle platitude pour prendre l'aspect de longs tuyaux d'orgues. Ces faits sont certainement 4 mettre au compte des conditions nautiques et de l'action des houles d'Ouest. Les grés de plage prennent une réelle importance de la pointe d'Antigues a 1'Anse Bertrand ot les lignes de brisants se succédent. L'Anse Bertrand, largement ouverte, abrite trois alignements successifs, phénoméne unique sur l'ensemble de ce littoral occidental et méridional. Ainsi la Grande Terre se caractérise-t-elle par une ceinture étroite, discontinue, de faible 6paisseur, et sans véritable tombant, de platures coralliennes frangeantes au Sud et a l'Ouest passant a4 des constructions alguaires sans doute ancrées sur des arrécifesa l'Est. Ces récifs plus ou moins vivants se localisent systematiquement au large d'anciennes constructions coralliennes soulevées a+4 out5 m , riches en Diploria, et probablement d'age pré-flandrien (datations en cours a Gif-sur-Yvette) : il existe donc une certaine constance dans la répartition géographique de la vie corallienne depuis un Quaternaire encore non précisé. Les "Culs de Sac Marins" De part et d'autre de l'isthme reliant Grande et Basse Terre s'étendent d'importantes platures dans une position abritée, mais en eau souvent trouble, surtout lors des fortes houles. Toutefois il est remarquable d'observer l'indigence de la vie corallienne sur la cdéte Ouest de Basse Terre, liée peut 6tre au trouble des eaux occasionné par l'apport alluvial des torrents qui descendent de la montagne proche. Cependant une exception importante doit é6étre signalée avec l'ilet a Pigeon, en face de Malendure, ot les constructions vivantes descendent a -48 nl. Depuis l'ilet Fajou, dans le Grand Cul de Sac Marin au Nord, jusqu'au droit de Sainte Marie, au Sud, au dela du Petit Cul de Sac Marin, les récifs s'étendent considérablement. Ils atteignent 200 4 300 métres de largeur sur une distance d'une quinzaine de kilométres d'Ouest en Est, a l'entrée du Grand Cul de Sac. Ils s'interrompent au niveau de deux profondes passes correspondant probablement 4 un ancien écoulement fluvial pré-flandrien, contemporain d'un bas niveau marin. Le platier massif porte des flots (cayes) sablonneux sur la bordure Nord, argileux et marécageux 4 l'arriére, colonisés par les palétuviers (ilets Fajou et Caret). Le vaste lagon est parsemé de patés et pinacles de coraux vivants, émergeant d'un fond sableux. Les principales espéces répertoriées sont Acropora palmata, Porites porites, Oculina, Mussa angulosa. A l'accore, il semble que la morphologie du récif prenne une structure en peigne avec é6éperons et sillons ; comme dans les récifs d 1'Océan Indien ou de la Mer Rouge, la créte alguaire reste de dimension restreinte”. En fait, les ilots sont accrochés a un ancien récif envasé, émergé d'une quarantaine de centimétres. On distingue une premiére auréole corallienne vivante accrochée au rivage ot dominent Porites porites, Porites divaricata et Millepora alcicornis, 1 Communication orale de J.P. Chassaing 2 Voir l'étude de A. Guilcher, sous-presse. puis une large plage sableuse, propre, ot proliférent les Acropora palmata, enfin le récif barriére proprement dit oti foisonnent les espéces. Notons qu'il s'agit toujours de tétes et de patés dispersés, jamais de constructions massives. Toutefois le secteur se distingue par l"abondance des formes vivantes. Le petit Cul de Sac Marin est également riche en constructions vivantes qui se disloquent en redans successifs que séparent de profonds chenaux, dans l'alignement strict du drainage continental actuel. Si cette série de platures s"apparente a celle du Nord, on ne distingue plus cependant la morphologie a cayes!, ni l'accore 4 6perons et sillons. Parmi les espéces les plus fréquentes nous relevons* Siderastrea, Madracis mirabilis, Stephanocoenia intercepta, Agaricia agaricites, Colpophyllia, Acropora cervicornis, Dendrogyra cylindrus, Meandrina meandrites. Il s'agit en fait d'une colonisation des anciens interfluves flandriens aujourd'hui immergés ; les passes profondes, dans le prolongement du drainage continental, méme indigent, pourraient étre interprétées comme d'anciens axes d'écoulement pré-flandrien. Ainsi nous relevons de profondes différences dans les diverses constructions récifales frangeant la Grande Terre ou barrant le Grand Cul de Sac ; ces disparités sont-elles a mettre en relation avec l'agitation de l'eau, les fortes houles 6tant plus favorables aux algues calcaires encroltantes, les littoraux protégés s'avérant plus aptes au développement de la vie corallienne ? La zone intermédiaire aurait-— elle connu dans un passé récent une vie corallienne plus florissante étouffée ensuite par la prolifération des algues ? Les quelques renseignements sur la direction, la force des vents au sol et de la houle ne sont pas décisifs (fig 2). Il serait d'un grand intérét d'étudier également les conséquences, au niveau des constructions vivantes, des vents destructeurs qui accompagnent les cyclones comme cela a déja été entrepris dans le golfe du Mexique. LA DESIRADE La petite ile de la Désirade posséde quelques constructions généralement immergées sur son littoral méridional (fig 4), dans 1'Anse de Baie Mahault, de 1'Anse Petite Riviére a la pointe du Désert, et dans l'inflexion, largement ouverte, de Grande Anse. Ce littoral rappelle celui de la céte orientale de la Grande Terre avec ses grés de plage qui se dégagent des sables calcaires blancs de la plage, un étroit herbier a Cymodocées qui n"existe pas partout, un chenal d'embarcations a patés de corail vivant, et enfin des constructions alguaires peu mMassives, discontinues, qui se dédoublent localement, fortement battues 1 Le terme local de caye est vidé de son sens morphologique. 2 aétermination et récolte de J.P. Chassaing. 6 par une houle a peu prés constante. Seule la plature barrant l'entrée de Grande Anse est largement percée par deux passes dans 1l'alignement d'un ancien drainage. Ce littoral méridional est bordé par un récif soulevé a 4-5 m ; les plus beaux témoins s'étendent a la pointe des Colibris, a l'Ouest, mais a 1'Est ils couvrent l'ensemble de la Pointe Gros Rempart sur une profondeur de 300 m. Comme 4 la Grande Terre, il y a donc ici aussi une certaine correspondance entre la localisation de la vie corallienne actuelle et celle des coraux anciens. En face du Souffleur, ot nous l'avons examinée, la créte alguaire se présente sous la forme de petits éléments a sommet relativement plat, de quelques métres 4 une vingtaine de métres d'allongement, et larges de 3 4 5 métres en moyenne, grossiérement alignés et séparés les uns des autres par de profondes coupures. Le sommet, qui émerge a mer basse, de 50 cm en moyenne, entre les déferlements des vagues, est couvert par des algues brunes, dont des Sargasses, et par des Ulves. Du cdté du large, ces éléments discontinus qui constituent la créte alguaire sont limités par un tombant brutal, parfois avec apparence d'encorbellement, entiérement tapissé par des algues calcaires encrotitantes de couleur rose. Vers le lagon, le tombant est moins marqué : le calcaire alguaire est de ce cdte troué de multiples loges d'oursins. Il y a aussi quelques tétes de corail vivant, mais il apparait nettement que ce corail ne joue pratiquement aucun ré6le dans la construction de la créte. Par contre dans le lagon, profond de 3 a 5 métres, il existe de nombreuses grosses patates isolées de corail sur fond de sable. MARIE GALANTE Cette ile, aux contours massifs, posséde aussi sur presque tout son pourtour un récif corallien ancien a 4 ou 5 métres d'altitude, qui localement (Folle Anse) pénétre 4 plus de 3 km a 1l'intérieur des terres. Ce vieux récif est absent du Nord de l'ile, qui correspond a4 un compartiment affaissé, dominé par un escarpement de faille morphologiquement bien net : c'est la un argument péremptoire pour un age pré-flandrien de ce vieux récif. Un récif frangeant actuel.n'existe que sur les facades W, S.W et Sud de l'ile, depuis 1'Anse Piton jusqu'a Grand-Bourg. Comme sur la céte orientale de la Grande-Terre et a la Désirade, les constructions alguaires jouent ici aussi un réle essentiel, et ce sont elles qui constituent tous les points hauts sur lesquels se produisent les déferlements. Comme 4 la Désirade, ces déferlements sont séparés de la ligne de rivage par un chenal d'embarcations mais qui ne dépasse pas ici 3 métres de profondeur. A la vision stéréoscopique sur photographies aériennes verticales, les éléments constituant la créte alguaire apparaissent souvent avec —" une forme en S ou en crochet (fig. 5) ; la créte alguaire elle-méme peut se dédoubler en deux ou méme trois alignements paralléles, comme en face de la Plaine des Galets au Nord de la Caye a Facheux (fig 5). Les plages de sable fin calcaire se terminent par un trottoir de grés a la fois 6pais, trés dur, et en forte pente ; mais cette grésification n"appartient qu'au littoral oriental. Les lagons sont occupés par un maigre herbier a Cymodocées auquel succéde rapidement un platier mort, blanchdatre, de Porites porites Pallas, ot vivent quelques souches aux moignons fragiles et quelques formations encrotitantes de Porites astreoides. Face au large se développent les Acropora palmata morts, puis le platier encrotité, propre, dénué de blocs basculés, qui se termine par une levée alguaire. Ce banc alguaire est Sillonné de chenaux de 2 a 3 m de profondeur, a fond sableux vers l"intérieur, s'enrichissant en galets vers le large. Ces sillons fréquemment réunis, soudés, déterminent des tunnels ou se réfugient poissons et langoustes. Les patés ont généralement un profil en champignon, avec plateau sommital horizontal, alvéolé, envahi par les algues courtes et dévoré par une infinité de petits oursins lithophages qui y ont creusé des loges ; les flancs, en encorbellement, sont tapissés de plaques d'algues brunes ou vertes, le pied est enfoui dans le sable calcaire. Au-dela succéde un paysage d"une grande désolation, champ de ruines figées d'Acropora morts, cassés, parmi lesquels s"accrochent quelques bouquets vivants qui se font de plus en plus nombreux vers le large (fig. 6). Cette description du littoral, valable de la caye a Facheux au Nord a la Pointe des Basses au Sud, est particuliérement caractéristique au niveau de la caye a Tonnerre. Ces constructions alguaires, communes aux littoraux orientaux étudiés, face a 1"Océan, prennent un beau développement et se présentent ici d'une maniére schématique. On observe nettement en plongée le passage progressif de la formation corallienne a Acropora palmata morts au platier construit par encrottement des algues qui prennent appui sur les ruines ; puis les Porites porites Pallas se multiplient 4 l'arriére. On observe également la coalescence des différents patés coralliens par les algues qui jettent des ponts, respectant les couloirs initiaux, lesquels deviennent alors des chenaux étroits ot croissent Porites asteoides, Agaricia agaricites, Diploria strigosa et D. clivosa, enfin Isophyllia. L'aspect général en S ou en crochet est illusoire : la vue d'avion révéle des constructions en patés successifs, individualisés : ilya une juxtaposition plutd6t qu'une véritable coalescence. L' aspect continu, que donne la carte au 1/20 000 e, est une interprétation abusive du dessinateur. Le dédoublement de la créte alguaire en deux ou méme trois alignements paralléles pose un probléme. L"hypothése la plus vraisemblable est qu'ici aussi, comme sur la c6te orientale de la Grande Terre, l'ossature est fournie par des lignes de grés de plage qui donnent le dessin d'ensemble des brisants. Ces grés de plage ne seraient plus apparents, car complétement habillés et recouverts par des constructions coralliennes et alguaires auxquelles ils ont servi de socle. “a4 2 MHC YRPy SYB221 a0 905 ead 2p 52VD Mig ee Tamalous reuuayyasr 5B) voor puiobje a2uauumpard © rasNJO/4 (Sptmumsopasd 5404) BHD) “SIY(AID SOUUA///OLOD SDINJO{f -******* * S29800/ YOO PUM) we ~— “al- =P *yogeT JIINVYD epeatsed eT g : OOS sale Jip wiv : 4219 29-14P PPNOP VOI72INC7 fal Pp sposv (ays MOG epezeu QPIWsD Z 7 2 DIY 4 tee We] © 2perey *2UeA ZO STNOY ‘se0eFANS sp nes,T op saznjeszsedusyz *PL6L-ITL6L *¢ “Sta gqQNO0O Cv Ff rPHnvyp ds Lt =. piw2Ko us yney 2: Ss NS we Bppwinoud yocy - meas SC o/euiusas @ 090 LO gypsy xDUd e 22 ob6f * (eIZTaL-espuerzy) STNOW SP 3SH,T & SeytToOsrzze sep jueuAOZ aHeTd ap saazy “eo °6tw auuanpye)f# DUIP OAD Speowy/iafop Wobd aun wr, aay sasodahs swsawaubyo ~ -—- -abojd op saub 2p sjuowaubyo oe ayog FILOY, 7, (ANNES Vi Ae aD Perwi —~ Le fen de age veo’ souleve a¢hm. ws dé ferlements Sor élément; 3 predominance algarve 10 tsobathe de moins 10 melves Gres de plage Fle Ses Cohbris Grosse /einle MARE GALANTE Echelle es ———— ort Fig. 4. Marie Galante et Désirade. SJUBAIA dv *(aqueTe5 3oT7eW) abetd eT e jue,q ne 2eheD eT ap adnoo sejjed sayisod S804 gd eyeujed eiodoiny gy Sjuod XNBUGYD ~ }UBWA)}eqsOIUa~ aja1> esie6)e jUuewajnosoue p au07 alles abejd ap syi9 Vincent A Geaye go vert Cand: (Y = ee ee fares a oe (VY Caye au ver. a Lo Kigole — a Rects en crochet” de ketile -hse tepom _——“Aragle oes vents 0 Caye o River ee, _10m | Ze 7 7 7 f Yipes’ WG \ U | ! ! l { \ "Récifs en crochet" de Petite Anse et "récifs en S" de = Fig. 5. Facheux a Marie Galante. CONCLUSION Comparée aux fonds si riches de Porto Rico, d'Haiti ou du golfe du Mexique, la vie corallienne parait peu active sur la 'bordure orientale de la mer Caraibe. Pourtant les conditions écologiques semblent favorables au premier abord (températures, clarté et agitation des eaux sont amplement satisfaisantes) comme 1l'indiquent les quelques relevés effectués par la météorologie nationale a la Désirade. Si la vie corallienne a été plus exubérante dans le passé (vieux récifs), pourquoi l'est-elle si peu actuellement ? Une modification du climat et des courants marins ne peut é6tre envisagée localement sans imaginer une répercussion autour des fles plus septentrionales qui sont encore plus exposées. La pollution ne peut étre non plus invoquée. Mais le fait essentiel demeure l'extréme vitalité des constructions alguaires dans les secteurs les plus exposés a la houle. La créte alguaire constitue le trait morphologique majeur de ces récifs. Elle détermine aussi largement leur évolution, en g@énant en particulier le transit des matériaux coralliens morts vers le rivage : peut étre est-ce la cause essentielle d'une différence majeure avec la morphologie des récifs de 1'Océan Indien, qui eux sont caractérisés par l'affleurement 4 basse mer de larges éléments de plature, alors qu'ici il y a généralement plusieurs métres d'eau entre la plage et les déferlements de la pente externe. Un autre caractére important est le rdéle que jouent les bancs de grés de plage. On peut supposer qu'en de nombreux endroits, ce sont eux qui constituent l'ossature de l'ensemble, et qui déterminent, sous leur habillage de corail et d'algues calcaires, le dessin général des brisants. BIBLIOGRAPHIE Ottmann, F. 1960. Une hypothése sur l'origine des ‘arrecifes' du Nordeste bresilien. C. r. somm. Soc. géol. Fr.7: 175-176. Laborel, J. 1965. Note préliminaire sur les récifs de grés et récifs de coraux dans le Nord-est brésilien. Rec. Trav. SES mace Endoume, Bull. 37, fase. 53: ° 341-344, ATOLL RESEARCH BULLETIN NO. 235 SYSTEMATICS AND ECOLOGY OF THE LAND CRABS (DECAPODA: COENOBITIDAE, GRAPSIDAE AND GECARCINIDAE) OF THE TOKELAU ISLANDS, CENTRAL PACIFIC by J.C. Yaldwyn and Kasimierz Wodzicki Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 Seren Bhat a deoteal v# Sivie”’) AAICOe4HMn? ANI #5, AL ootyniden 4 P01. soderrraY, CONTENTS Abstract page Introduction Environment and Natural History Situation and Climate People Soils and Vegetation Invertebrate Animals Vertebrate Animals Material and Methods Systematics of the Land Crabs Coenobitidae Coenobita Coenobita brevimana Coenobita perlata Coenobita rugosa Birgus Birgus latro Grapsidae Geograpsus Geograpsus crinipes Geograpsus grayi Metopograpsus Metopograpsus thukuhar Sesarma Sesarma (Labuanium) ?gardineri 20 20 ii Gecarcinidae page 23 Cardisoma 24 Cardisoma carnifex 25 Cardisoma rotundum 27 Tokelau Names for Land Crabs 30 Notes on the Ecology of the Land Crabs 37 Summary 43 Acknowledgements 44 Literature Cited 45 IV. Fibers LIST OF FIGURES (following page 53) Map of Atafu Atoll, based on N.Z. Lands and Survey Department Aerial Plan No. 1036/7A (1974) . Map of Nukunonu Atoll, based on N.Z. Lands and Survey Department Aerial Plan No. 1036/7B sheets 1 and 2 (1974). Map of Fakaofo Atoll, based on N.Z. Lands and Survey Department Aerial Plan No. 1036/7C (1974). Sesarma (Labuanium) ?gardineri. Dorsal view of male, carapace length 28 mm from Nautua, Atafu. (Photo T.R. Ulyatt, National Museum of N.Z.) Cardisoma carnifex. Dorsal view of female, carapace length 64 mm from Atafu. (Photo T.R. Ulyatt) Cardisoma rotundum. Dorsal view of male, carapace length 41.5 mm from Village Motu, Nukunonu. (Photo T.R. Ulyatt) LIST OF TABLES Surface temperature in the Tokelau Islands (°c) page 5 Mean rainfall in the Tokelau Islands (mm) 6 Comparative list of crab names from the Tokelau Islands, Samoa, Niue and the Cook Islands. 35 Numbers of land crabs in five quadrats on Atafu Atoll, January-February 1973. 39 ie trae Yeapn TAy 10) fai “bas Sib 2) Tas a (E2 eveq patiwal fo?) . 7 hsaad .LLocgA otAga b5e5) AT\SEOL .of get neo Deeed .LIGIA pohomeuy oe pt net 7 , ‘ a eee ny asigog { iocttses te ‘ : a= vw oei¢g Ieee c he _ = SYSTEMATICS AND ECOLOGY OF THE LAND CRABS (DECAPODA: COENOBITIDAE, GRAPSIDAE AND GECARCINIDAE) OF THE TOKELAU ISLANDS, CENTRAL PACIFIC by J.C. Yaldwyn! and Kazimierz Wodzicki” ABSTRACT The Tokelau Islands consist of three atolls (Atafu, Nukunonu and Fakaofo) approximately 500 km north of Western Samoa. Their numerous islets are formed mainly of coral sand and rubble with no standing freshwater. Sixty-one plant species have been recorded, 13 of these being introduced and 10 being adventives. There are three vegetation zones, the beach, the beach-crest, and the interior coconut/fern zone with the physiognomy of a humid tropical forest. Marine invertebrates have not been studied. One hundred and fifty insect species in 83 families have been recorded with most being widely distributed South Pacific species including several introduced agricultural pests, e.g. Rhinoceros Beetle. Some marine fishes have been listed and 7 species of lizards are known from the group. Twenty-six bird species (15 sea birds, 8 shore birds and 3 land birds) are known but none are endemic races. Domestic pigs, cats, man and the Polynesian Rat (Rattus exulans) are the only mammals. R. exulans is an economic pest as it Causes considerable damage to the coconut crop and assists in the spread of filariasis. Ten species of "land crabs" are identified from the Tokelaus, 4 being terrestrial hermit crabs of the anomuran family Coenobitidae (3 Coenobita spp. and Birgus latro) and 6 being terrestrial crabs of the brachyuran families Grapsidae (2 Geograpsus spp., a Metopograpsus and a Sesarma) and Gecarcinidae (2 Cardisoma spp.). Eight of these species have been recorded before from the Tokelaus under one name or another but two, Geograpsus grayi and Metopograpsus thukuhar, are new records for this group of atolls. The total of 10 land crab species 1 National Museum of New Zealand, Wellington, New Zealand. 2 Zoology Department, Victoria University of Wellington, New Zealand. (Manuscript received December, 1977. --Eds.) from the Tokelaus can be compared numerically with 15 species on one atoll (Arno) in the Marshall Islands, 9 on one atoll (Kapingamarangi) in the Carolines and 7 from one atoll (Raroia) in the Tuamotus. All the land crab species from the Tokelaus (except for Sesarma ?gardineri known elsewhere only from New Guinea and southern Micronesia) are wide- ranging Indopacific forms known from at least the western Indian Ocean to the eastern central Pacific. Fourteen different crab names in the Tokelau language are recurded and identified. Vernacular names for common land crabs such as uga (Coenobita brevimana), ugauga (Birgus latro), paikea (Cardisoma rotundum) and kalamihi (Geograpsus crinipes) are used with minor modifications for similar species in Samoa, Niue and the Cook Islands. A traditional story about the kaviki (the shore crab, Ocypode ceratophthalma) recorded by Dr. Judith Huntsman of the University of Auckland, is given in translation. All land crab species on the Tokelaus are nocturnal scavengers. Population counts of Coenobita brevimana and C. perlata together, Birgus latro, Cardisoma rotundum, C. carnifex, and Geograpsus crinipes give a combined density of about 560 crabs per 5000 sq. m. This figure can be compared to approximately the same land crab density figure on Kapingamarangi Atoll in the Carolines (Niering, 1956) but is in dramatic contrast to counts of approximately 30,000 Cardisoma planatus per 5000 sq. m on Clipperton Atoll in the eastern tropical Pacific (Ehrhardt, 1968). The numerous land crabs on the Tokelaus have an indirect effect on rat control programs and a direct effect on R. exulans themselves. Crabs interfere with rat traps and poison baits, eating both warfarin and zinc phosphide. The effect of the anticoagulant warfarin on crabs is unknown, but quadrat counts show that crabs are severely affected by zinc phosphide. The slight risk to humans of secondary poisoning from eating affected crabs can be virtually eliminated by banning crab collecting in poisoned areas for periods after poison baiting, and by putting poison baits in aluminium tubes of a diameter small enough to exclude large and sought-after species such as Birgus latro and the two Cardisoma spp. Large land crabs presumably prey on young rats, thereby forcing rats to nest above ground level, and compete with rats for at least one of their important foods, the coconut meat available in man-opened and abandoned nuts. INTRODUCTION There is no comprehensive information on the systematic status and biology of “land crabs" (both terrestrial hermit crabs and terrestrial true crabs) on islands in the central and south Pacific. Holthuis (1953) in his enumeration of crustacean species from coral atolls in the Marianas, Marshalls, Gilberts and Tuamotu Islands lists, but does not comment on, seventeen species of land crab in the families Coenobitidae (land hermit crabs), Grapsidae and Gecarcinidae (true crabs) . Niering (1956: 16-18) recorded nine land crab species of these families from Kapingamarangi Atoll in the Carolines and gives a general, but vivid, description of their habits. An extensive summary of the published information on the general ecology of land crabs on Pacific atolls is given in the excellent review of atoll environment by H.J. Wiens (1962: 432-439). One of us (K.W.) visited the atolls of the Tokelau group on four separate occasions (November 1966 - February 1967 and April - June 1968, Nukunonu Atoll; July - September 1971, Fakaofo Atoll; December 1972 - February 1973, Atafu Atoll) working on the problem of rat damage to coconuts (Wodzicki, 1968a, 1968b, 1969a, 1970, 1972a, 1972b, 1973a, 1973b, 1973c; Mosby and Wodzicki, 1972; Mosby, Wodzicki and Shorland, 1974; Mosby, Wodzicki and Thompson, 1973). During his early visits he made incidental collections of land crabs and was surprised at both the systematic variety and the abundance of these animals. On the suggestion of J.C.Y., he made a special collection of land crabs during his last trip concentrating on obtaining a coverage of the different species present, particularly those with "specific" Tokelauan names. The 1972/73 visit to Atafu Atoll produced a list of seven different land crab species and these have already been recorded in a preliminary report on this visit (Wodzicki, 1973c) . K.W. had previously recorded that land crabs seriously interfere with snap and live-trapping of rats on the Tokelaus (Wodzicki, 1968a: 56) and then added details of interference with ground-placed poison baits (1973c: 21,29). As the coconut crab (Birgus latro) and to a lesser extent other land crabs are a part of the Tokelauan diet, he Was aware that the use of acute poisons in rat control, such as zinc phosphide, might create the danger of secondary poisoning of humans. During a concurrent survey (Wodzicki, 1969b) of rat ecology and damage on Niue Island, an isolated raised atoll south-east of Samoa, K.W. made similar collections of land crabs and these will be the subject of a future study. The purpose of the present paper is to provide a detailed account of the systematic status and local names of the ten land crabs on the Tokelau Islands, to comment on their ecology and significance in these atolls, and to discuss the local problem of rat-crab relationships. ENVIRONMENT AND NATURAL HISTORY Situation and Climate The Tokelau (or Union) Islands consist of three atolls, named from west to east Atafu, Nukunonu (sometimes misspelt "Nukunono", see Wodzicki and Laird, 1970: 247) and Fakaofo. Their eighteenth and nineteenth century European names, not now used, are Duke of York, Duke of Clarence and Bowditch Islands respectively. The Tokelaus are bounded by latitudes 8°S and 10°S and by longitudes 171°W and 173°w. The atolls are approximately 500 km to the north of Western Samoa; Nukunonu is about 90 km south and east of Atafu and Fakaofo is about 65 km east of Nukunonu. They are typical atolls, each being surrounded by a coral reef with an inner lagoon. The numerous islets (or motus) of each atoll vary in number and size but do not rise more than 3 to 5 metres above sea level. The total land area of the Tokelaus is about 11.5 sq.km and the areas of the three atolls are approximately as follows (revised 1975 figures, N.Z. Ministry of Foreign Affairs): Total land Area largest Area of area islet lagoon (sq. km) (sq. km) (sq. km) Atafu (fig. 1) 3 eS 19 Nukunonu (fig. 2) ae) eS 109 Fakaofo (fig. 3) 4 ik 59 The mean annual temperature is about 28°C, with July or August being the coolest months and April or May the warmest. The rainfall is heavy but inconsistent and a daily precipitation of 8 cm or more can be expected at any time of the year. The highest mean annual rainfall is 2829 mm at Fakaofo for the years 1958-74. The islands lie in the zone of the south-easterly trade winds but from November to February north-easterly and northerly winds predominate (Kennedy, 1966). The Tokelaus are within the hurricane belt and severe tropical storms occur irregularly. Further details of daily temperatures and annual rainfall provided by the New Zealand Meteorological Service (J.D. Coulter, in litt.) are given in tables I and II. People The Tokelau atolls are in a border area between Micronesia and Polynesia, and are inhabited by about 1600 Polynesians. Although the islanders retain linguistic and cultural ties with Samoa, Tokelauan culture is of composite origin including elements from both eastern and western Polynesia (Macgregor, 1937). It is distinctly moulded by the atoll environment. The population has shown a significant decline over recent years due to emigration to New Zealand (fore- shadowed by Doumenge, 1966, and documented by Hooper and Huntsman, UO 73))e The islands are within the political boundaries of New Zealand and the Tokelauans are New Zealand citizens (Annual Report, 1974; Kennedy, 1966). Nearby Swains Island (or Olosenga), regarded by some authors as geographically part of the Tokelau group (see Krauss, 1970) is about 200 km south of Fakaofo and about 300 km north of Samoa. It is United States territory and is excluded from the present study. Bibliographies of the Tokelau Islands and of Swains Island have been compiled by Krauss (1969, 1970). VL6T “ZL6T-6961 POtazed [enjoy © VLET ‘ZLOT-696T POTzed Tenjou VL6OT ‘ZLET-TLET ‘6961 POtzed TenqoVy I GLE whe LLG Ole Mike GLE Lhe SLE BLE BLE ELE ULE Q°Se ueow ATTeq WG Dive LW GWE GWE WG SVG CVG Weve Wie LEG Weve sg > whawaseham “Greely wish 8° OE EOE BCE GE GE OOF Glee Gi PMU yw LCE LCE B°S unutxeu ATTep uesw ¢ (VL6T-696T) ONONOXAN Weg ESC BC Wee EG Ole Yhe wee WR Le S°OC CBE BLE ueow ATTeq GSE SSG L°SG E°SE WSEG WSC Se SSG GSC BSC E°Se LSE O°S~ unututu ATTep ueeW L°O€ Lue ows O@E COE BO BOC ECE BCE SWE EWE Sls SCE unutxeu ATTep ueSy z(VL6T-696T) OJOWNWA 78 ECG SSC WEG YBG O'LG OBE Ge L8G OBE EOE SSE TL-Se ueeu ATTeq 9°SZ v SE 8S€ GSE YSG |e ESE SVE WME SSE LSE Sse W°SZ unututu ATTep uesyW ipeke Eve OWE Oise Owe SOE LOE LOE CWE EWE CUE LE Oe unutxeu ATTep uesy tT (VL6T-696T) NawLw G*82é O°8c O°'8€ O°8 S°*8~ O°8% O°8% S°8Z S*8Z S°8Z S°8Z O°8Z O°8Z ueseu ATTeq 0°92 OVW O8VE OMG OF OV OOS |F9E OFC O'F% OYE E°Gs O'S? umututTu ATTep uesW O°OE QO OG OC@E OOS OCE GGG OCF Gee SCE GOCE O'GE OOF unutxem ATTep uesW (TP6T-626T) NAWLW ueepl oeq AON 290 deg pnw Tne une Aen rdw ZTeW qe uer (0,) SPUCTSI NeTSYoL, sy} uT eznjezeduez sovzmms I eTger T6972 6782 cSSZ TT6Z TeOK 6LE SCE 8ST TO? LOG V8z 6T? L9Z 390 LUG 9072 €07 S8T das S6T GLE 802 GEG L8T Tv TnL EST €6T VL6T EEL GIT ates SEC 7S 4A TA-TL6T “Oo-seer ONONNYNN S6T SCT ©0¢G SOE. EBE VL-OL6T ‘O9-8S6T OdJOWIWA 6LT OPT 812 PST 86€ PLET‘TL6T‘6S-8S6T SGile Sle 69¢ ‘Sre ] 6c €S6T-626T NAVLW Kew ady Zew qeq wep pxooax Fo potred (um) SPUPTST NeTSeyYOL, syj UT [TTesuTeA ueaW TI STP, Soils and Vegetation The islets of the three atolls are, with a few local exceptions, composed entirely of decaying organic matter, coral sand and rubble. There is no natural standing freshwater and the islanders use sub- surface sources and roof catchment systems. The plant cover of the islets has not been greatly modified by man and despite centuries of human intrusion, and the ubiquitous introduction of the coconut palm, the vegetation of the Tokelaus still remains in a state of natural equilibrium. The relative paucity of plant species (61, of which 13 are introduced cultivated plants and 10 are adventives) is another important characteristic of the vegetation of the Tokelau atolls (Parham, 1971). It is appropriate to consider the vegetation of each atoll islet as a single unit, a comparatively simple ecosystem, composed of three main zones: (1) the lower beach or foreshore of sand with more stable areas of coral rubble or beachrock; (2) the beach- crest, and (3) the interior strip on relatively flat stable coral sand. The characteristics and the principal plants of these three zones have been described by Parham (1971: 587-593, figs 4-8) in some detail and can be summarised here. The narrow foreshore zone of either the ocean or lagoon side of the islet is "practically devoid of vegetation" probably due to the unstable nature of this unprotected beach area. A creeping sand-binder, a grass and sedge occasionally occur aS pioneers on recently-modified beaches. The beach-crest zone, as the name indicates, stands about 3 to 5m above the high tide level and may extend 10 to 20m inland. There are four distinct plant associations including the Messerschmidia argentea facies which is characteristically seen on exposed sea-side frontages. Other distinctive shrubs and trees found in this zone are the windbreak- forming Scaevola taccada, the tall straggling shrub Pemphis acidula, the screwpine Pandanus (probably P. tectorius) and the tree Guettarda speciosa. The interior strip or central zone of the islets is typically occupied by a dense coconut/fern (Cocos nucifera/Asplenium nidus) forest with a closed 20m or more high canopy formed by coconut palms with an understory of other trees including Cordia subcordata, Pisonia grandis, Guettarda and Pandanus spp. The luxuriant character is enhanced by low shrubs and a dominant crowded ground cover of ferns, including Nephrolepis hirsutula, Phymatodes scolopendria and Psilotum nudum in addition to Asplenium, as well as mosses, lichens and fungi. This plant association has the general physiognomy of a humid tropical forest. Human existence on the Tokelaus is still largely based on fishing and coconuts. All vegetated islets of any significant size carry coconuts planted by man. The eastern islets of each atoll carry more coconut palms to the ground area than the western and southern islets, and are reputed to provide larger returns of copra per comparative area. Other introduced cultivated plants are only found on the village motu of each atoll which in all three cases is a western islet. These cultivated introductions are mainly used for food or as ornamental plants. The interior forest zone of certain eastern islets and areas such as Olopuka, Natama, Te Ahaga and Laualalava on Atafu Atoll (see fig. 1), Long Motu on Nukunonu Atoll (fig. 2), and Fenualoa and Lefu on Fakaofo Atoll (fig. 3) appears to be denser and more luxuriant than the forest zone of western and southern islets. This luxuriance in plant growth may have some connection with the predominence of easterly trade winds. Observations by K.W. and reports from local inhabitants would indicate that coconut crabs (Birgus latro), and possibly other land crabs, are significantly more abundant on these uninhabited eastern islets. In addition to coconut palms, two other plants in the coconut/fern forest zone are of especial importance to the Tokelauans. The kanava (Cordia subcordata) is the only timber tree used for canoe hulls on these atolls and the lau mea fern (Asplenium nidus) is extensively used aS a green vegetable (K.W., personal observation; Parham, 1971: 603-604) . Invertebrate Animals There is a rich fauna of corals and other marine invertebrates around the Tokelau Islands, not yet studied. Reef and lagoon crabs, and other crustaceans are well represented. Small collections of marine crabs made by K.W. are in the National Museum of New Zealand and Tokelauan names of a few species are recorded below. Hinds (1971) in a report to the South Pacific Commission mentions the presence of the Crown-of-Thorns starfish at these islands and discusses van Pel's observations (1958) on the possibility of transplanting pearl oysters and trochus shells to the lagoons. Although a recent issue of Tokelau Islands stamps features selected corals (September 1973) there is no published record of the actual occurrence of these forms in the area. A stamp issue featuring named cowrie shells (November 1974) is, however, based on species recorded from one of the atolls (Ingam, 1940). The terrestrial invertebrates of the Tokelaus have been the subjects of several studies. Laird (1956) recorded some freshwater protozoans, Hoyt (in Wodzicki, 1968a) listed an unidentified earthworm, Laird (1956) listed several unidentified entomostracan crustaceans, while Dale (1959) identified three widely distributed isopods. Land crabs are listed by Laird (Sesarma sp. only), Hoyt, Hinckley (1969) and Wodzicki (see below). Insects have been treated in some detail by Dale (1959), Hoyt (in Wodzicki, 1968a) and Hinckley (1969). The latter records a total of 150 insects in 83 families most being widely distributed South Pacific species including several introduced agricultural pests. The most important of these introductions is obviously the Rhinoceros Beetle (Oryctes rhinoceros) accidentally brought in from Western Samoa and still well established on Nukunonu Atoll in spite of a planned programme of eradication sponsored by the United Nations Development Program/South Pacific Commission Rhinoceros Beetle Control Project. Arachnids have been recorded by Marples (1955), who found 13 spiders all known from Samoa, Hoyt (in Wodzicki, 1968a), who added two mites and a scorpion, and by Hinckley (1969) who listed further mites. Centipedes and millipedes have been listed by Hoyt and by Hinckley. Additional collections of terrestrial arthropods made on the Tokelau atolls by K.W. are in the Entomology Department of the National Museum of New Zealand, but have not as yet been studied. Three terrestrial molluscs, all known from other Pacific islands, are recorded by Dale (1959). Finally the Tokelau Islands were selected as the site of a WHO supported research project on the control of mosquitos as carriers of filariasis and the results of this project have been reviewed by Laird (1966). The geographic relationship of the Tokelau Island terrestrial invertebrates can be summarized in the words of Hoyt (in Wodzicki, 1968a) “It is doubtful if any of the species of land inhabiting invertebrates are endemic to the Tokelaus and probably a few of the species come and go from time to time. In general most of the insects (and spiders) appear to have come from Samoa, although many are widely distributed throughout the South Pacific." In contrast to the Central Pacific atolls of Canton (Phoenix Group), Palmyra (Line Islands) and Johnston, which have faunas mainly derived from Hawaii, "the Tokelaus have obtained most of their fauna from the south and much of it from Samoa." Vertebrate Animals There is a rich and varied lagoon and offshore fauna of marine fishes still undocumented and fishing remains the mainstay of the Tokelauan way of life. Van Pel (1958) and Hinds (1971) give lists of many edible fishes taken by local fishermen. A stamp issue featuring named reef fishes was issued in November, 1975. Seven species of Pacific Island lizards (4 skinks and 3 geckos) have been recorded from the Tokelaus (Wodzicki, 1968a: 69; Whitaker, 1970). The commonest species is the blue-tailed skink, Emoia cyanura, found practically everywhere within coconut plantations and near human habitation. Four marine turtles have been identified from local reports (Wodzicki, 1968a: 67) and it appears that the Green Turtle (Chelonia mydas) nests on some of the islets of the group but is subjected to considerable human predation. A recent survey (Wodzicki and Laird, 1970) revealed a relatively rich avifauna of at least 26 species (15 sea birds, 8 shore birds and 3 land birds). Of these, seven sea birds and one land bird (the Pacific Pigeon, Ducula pacifica) are known to breed on the Tokelaus and the populations of these eight species are at present on the decline as they are being taken for food by the islanders. All 26 bird species are known widely on other Pacific islands and no endemic 10 races have been described from the Tokelaus. Most have been recorded from nearby central Pacific island groups such as the Gilbert and Ellice Islands and the Samoan Archipelago. All but one of the shore birds and two of the land birds (a duck, ? Anas super-ciliosa, and the Long- tailed Cuckoo, Urodynamis taitensis) are migratory which demonstrates the importance of these atolls as mid-Pacific stepping stones. Apart from domestic pigs, which are retained in enclosures or on small islets, cats (almost entirely kept as pets, rarely feral), and man, the Polynesian Rat (Rattus exulans) is the only other mammal found in the Tokelau Islands (Kirkpatrick, 1966; Wodzicki, 1972b). Fortunately no other rodents are found on these atolls but the Polynesian Rat inflicts considerable damage to coconuts, thereby affecting both the local food supply and copra production (Wodzicki, 1968a, 1972a). Rat damage is restricted to immature green coconuts on the palm (Laird, 1963) and rats do not attack unopened nuts on the ground (Wodzicki, 1972a: 311). The Polynesian Rat gnaws a small pit or an opening in the fibrous exterior of the green nut near the peduncle. A few days later the damaged nut drops to the ground and lies unused until eventual disintegration. These damaged nuts provide ideal habitats for the larvae of the diurnal mosquito (Aedes polynesiensis), carrier of filariasis (caused by the nematode Wucheria bancrofti), which breed in either the water-filled, rat-gnawed pits or in the decaying interior of the nuts (Laird, 1963, 1966). MATERIAL AND METHODS All specimens in the "Material examined" lists given below were collected by K.W. unless otherwise stated. All examined material is in the collection of the National Museum of New Zealand. Measurements of both hermit crabs and true crabs are middorsal carapace lengths in millimetres. SYSTEMATICS OF THE LAND CRABS CRUSTACEA Order DECAPODA ANOMURA Family COENOBITIDAE Coenobita Latreille, 1825 Coenobita Latreille, 1825: 276. Alcock, 1905: 139-141 (key to Indian supp.), 192-193 (world checklist). Fize and Seréne, 1955: 2-7 (key to South East Asian spp.). Gordan, 1956: 311-312 (bibliography 1905-1954). Three species of terrestrial hermit crabs were found on the Tokelau Islands. These can be recognized as belonging to the characteristic terrestrial hermit crab genus Coenobita by their dual habitat, by their shell-bearing habit, and by the compressed eye stalks with the cornea both terminal and lateral. Their generic identity can be confirmed by the twisted, asymmetrical abdomen bearing uropods but lacking paired pleopods and by the peduncle of the antennule (antenna 1) being as long as, or longer than, the carapace and ending abruptly (i.e. lacking a flagellum). All three are typically nocturnal species and can be specifically identified using the key given by Alcock (1905; translated into French by Fize and Serene, 1955). Coenobita brevimana Dana, 1852 (Tokelau name - uga) Coenobita clypeata Latreille, 1826: 277. [Not C. clypeatus Herbst, 1791 = C. diogenes of various authors, from the West Indies. ] Coenobita clypeata; Dana, 1852: 473; 1855: pl.30 fig. 4a. Coenobita clypeata var. brevimana Dana, 1852: 473; 1855: pl.30 fig. 4b. Coenobita clypeatus; Alcock, 1905: 142, pl.15 figs. 1, la. Coenobita brevimanus; Rathbun, 1910: 314 (synonymy with C. clypeata Latreille). Coenobita hilgendorfi Terao, 1913: 388 (replacement name for C. cylpeata Latreille used by some authors). Coenobita brevimanus; Holthuis, 1953: 36. Coenobita clypeata; Fize and Serene, 1955: 7, figs. 1 A-C, pl.l fig. 1 Material examined: Atafu Atoll 7 Feb. 1973, Nauta, 1 male 27mm, 2 females 25mm (1 ovigerous) . 14 Feb. 1973, Te Hepu, 1 damaged male, 3 females 25-28mn. 19 Feb. 1973, Vao islet, 2 males 19-20mm, 25 females 20-29 mm. (collected with C. perlata). Fakaofo Atoll 1 Aug. - 19 Sept. 1970, 4 males 19-44mm. Remarks: This characteristic purplish species can be readily identified as "C. clypeata Latreille" from Alcock's 1905 key by the reduced antennal scale not being fused with the 2nd segment of the antennal peduncle, by the slightly (rather than "Strongly") compressed eyestalks, and by the presence of a “bunch of hairs" on the inner surface of the right palm only. The form of the large (left) hand in the larger specimens from Fakaofo and Nautua is as shown in Fize and Serene, 1955: fig. 1B. The lower border is long and relatively straight. The lower border of the large hand in the smaller specimens from Fakaofo and in the Atafu collections of 14 and 19 February differs from that shown by Fize and Seréne in being relatively short and more convex in outline (cf. the characters given for Dana's so-called variety brevimana by Fize and Seréne, 1955: 11). 12 We have not followed Fize and Seréne (1955) in their use of the name "C. clypeata Latreille" for this species. Rathbun (1910) accepted that the name "C. clypeata" was pre-occupied by C. clypeata Herbst, 1791, used for a West Indian species of this genus. She applied the name C. brevimana Dana, originally used for a variety of C. clypeata Latreille, to this well known Indopacific species and was followed in this usage by Edmondson (1923) and Holthuis (1953). The name C. hilgendorfi Terao, 1913, has been used by some authors for this species, e.g. Forest (1954). The mollusc shells used by this species and retained with the above collections belong to three genera. The commonest was Turbo (Marmarostoma) argyrostoma, and the other two were Mancinella armigera and Casmaria ponderosa. Colour notes: Carapace, hands and legs purplish in life and colour retained in alcohol. Distribution: Widely distributed in the Indopacific ranging from the East African coast and Madagascar through the Indian Ocean and the Indonesian Archipelago to the Marshall Islands (Holthuis 1953), Ellice Islands (Whitelegge, 1897), Line Islands (Edmondson, 1923), Tahiti and the Tuamotu Archipelago (Holthuis, 1953; Forest, 1954). This species was first recorded from the Tokelau Islands by Hinckley (1969), and later (under the name C. clypeata) by Wodzicki (1973c). Coenobita perlata H. Milne Edwards, 1837 (Tokelau name - uga kaifala) Coenobita perlata H. Milne Edwards, 1837: 242. Coenobita perlatus; Alcock, 1905: 145, pl.XIV figs 2, 2a. Coenobita perlatus; Holthuis, 1953: 37. Coenobita perlata; Fize and Seréne, 1955: 24, figs 3C, 4A-C; pl.l aig se Di Material examined: Atafu Atoll 31 Dec. 1972, Motu-ite-Fala, 1 female 3lmn. 16 Jan. 1973, Nautua, 1 female 29mm. 26 Jan. 1973, Olopuka, 1 male 28mn. 10 Feb. 1973, 1 male 26m. 19 Feb. 1973, Vao islet, 1 male 21mm (collected with C. brevimana). Fakaofo Atoll 1 Aug. - 19 Sept. 1970, 3 males 8-30mm. 13 Remarks: The striking red colouring allows immediate visual recognition of this species in its natural habitat. It can be readily identified as C. perlata from Alcock's 1905 key by having the reduced antennal scale fused with the 2nd segment of the antennal peduncle, by the strongly compressed eyestalks, by the presence of a “brush of hairs" on the inner surface of both palms, by the presence of an oblique stridulating row of laminar teeth on the upper part of the outer surface of the left palm, and by having the outer surface of the propodus of the 3rd left leg (2nd walking leg) convex and not sharply separated from the anterior surface. In mature males the coxa of the 5th right leg is produced into a long curved tube (the vas deferens), considerably longer than the short vas deferens tube of the left coxa. Specimens of this species were taken in shells of the mollusc Turbo (Marmarostoma) argyrostoma. Colour notes: As mentioned above, the characteristic colour of the carapace and legs of this species is red, varying in the specimens examined from a pale to a vivid red. The juvenile specimen with a carapace length of 8mm was creamy-white in general colour with red bands on the wrists of chelae and walking legs. Distribution: Widely distributed in the Indopacific ranging from the Red Sea and Madagascar through the Indian Ocean and the Indonesian Archipelago to the Marshall Islands (Holthuis, 1953), Samoa and the Tuamotu Archipelago (Holthuis, 1953; Forest, 1954). This species was first recorded from the Tokelau Islands by Hinckley (1969), and later by Wodzicki (1973c). Coenobita rugosa H. Milne Edwards, 1837 Coenobita rugosa H. Milne Edwards, 1837: 241. Coenobita rugosus; Alcock, 1905: 143, pl.XIV figs. 3, 3a. Coenobita rugosus; Barnard, 1950: 469, fig. 86. Coenobita rugosa; Holthuis, 1953: 40. Coenobita rugosa: Fize and Seréne, 1955: 12, figs. 2A-C, 3A; pl.l figs. 3, 5, 7-10. Material examined: Atafu Atoll 5 Dec. 1972-20 Feb. 1973, 2 males 9-12mm, 2 females 7.5-9.5mm. 14 Remarks: These small specimens were identified as this species from Alcock's 1905 key by having the reduced antennal scale fused with the 2nd segment of the antennal peduncle, by the strongly compressed eyestalks, by the presence of a "brush of hairs" on the inner surface of both palms, by the presence of an oblique stridulating row of lamina teeth on the upper part of the outer surface of the left palm, and by having the outer surface of the propodus of the 3rd left leg flat and separated from the anterior surface by a well defined crest. In mature males the coxa of the 5th right leg is produced into a curved vas deferens tube but this is only slightly longer than the vas deferens tube of the left coxa. Colour note: The colour of these small specimens was creamy-white with light, orange-brown marks on hands and walking legs. Distribution Widely distributed in the Indopacific ranging from Natal and East Africa (Barnard, 1950), through the Indian Ocean and the Indonesian Archipelago to the Marshall and Gilbert Islands (Holthuis, 1953), Ellis Islands (Whitelegge, 1897), Line Islands (Edmondson, 1923), Tahiti and the Tuamoto Archipelago (Holthuis, 1953; Forest, 1954); apparently recorded from the west coast of the Americas (Barnard, 1950). This species was first recorded from the Tokelau Islands by Wodzicki (1968a) . Birgus Leach, 1815 Birgus Leach, 1815: 337. Alcock, 1905: 148. The genus Birgus is monospecific. The general appearance and habits of the distinctive Robber or Coconut Crab, B. latro, are well known from numerous general accounts and illustrations that have been published in the semi-popular literature on Indopacific animals. The place of this large land "crab" in the general ecology of coral atolls has been summarized by Wiens (1962: 432-434). Birgus latro (Linnaeus, 1767) (Tokelau name - ugauga) Cancer latro Linnaeus, 1767 (not "1758" as given by Gordan, 1956): 1049. Birgus latro; Henderson, 1888: 50 (synonymy). Birgus latro; Allcock; 1905: 50; spie xVia eto. e. Birgus latro; Reyne, 1939: 283 (habits and distribution). Birgus latro; Gordan, 1956: 304 (bibliography 1905-1954). 15 Material examined: Atafu Atoll 8 Jan. 1973, Nautua, 2 females 62-74mm. Fakaofo Atoll 1 Aug. - 19 Sept. 1970, 1 male 33mm. Remarks: Although many specimens of this large and readily identified land crab were seen by K.W. during his visits to the Tokelau Islands only three small specimens were brought back for the systematic collections. Larger specimens were collected during rat poisoning trials on Atafu in 1973 and have been used for chemical analysis. Distribution: Widely distributed on islands in the Indian Ocean; not recorded from the Malaysian-western Indonesian area, but ranging from the eastern Indonesian Islands and the Philippines, through northern New Guinea, to the Marshall Islands (Holthuis, 1953), Ellis Islands (Whitelegge, 1897), Niue Island (Yaldwyn, 1970), Line Islands (Edmondson, 1923), Tahiti and the Tuamotu Archipelago (Holthuis, 1953; Forest, 1954). Recorded from Swains Island just to the south of the three atolls of the Tokelau group by Dana (1852: 474; 1875). Dana visited Swains Island during the United States Exploring Expedition 1838-42 and recorded that "Great numbers of Birgi (large Crustacea) were burrowing over the island, some of which were six inches in breadth" (Dana, 1875: 160). First recorded from the Tokelau Islands by Wodzicki (1968a) and Hoyt (in Wodzicki, 1968a), and later by Hinckley (1969). BRACHYURA Family GRAPSIDAE Subfamily GRAPSINAE Geograpsus Stimpson, 1858 Geograpsus Stimpson, 1858: 101. Rathbun, 1918: 231 (American and eastern Pacific sp.). Tesch, 1918: 74 (Indo-West Pacific spp.). Banerjee, 1960: 157 (key to Indo-West Pacific spp.). Two species of the land crab genus Geograpsus were collected on the Tokelau Islands. These can be readily recognised in the field as belonging to this distinctively terrestrial grapsid genus by the obvious toothed plate on the chelipeds. This feature is a prominent, plate-like expansion of the inner distal margin of the merus (arm). It is found in some other members of the subfamily Grapsinae, such as the intertidal genera Grapsus and Pachygrapsus, the semi-terrestrial and distinctively-shaped genus Metopograpsus (see below) and the 16 temperate genus Leptograpsus. The generic identity of the Tokelau material dealt with here can be confirmed by the following features: the carapace is quadrate and dorsally flattened with the front (the interorbital margin) wider than either orbit but less than half the carapace width; the pterygostomial region of the carapace (on either side of the mouth-field) is neither reticulated nor "hairy"; the fingers of the chelae are distally acute and not spoon-shaped, and there is an opening into the gill chamber fringed with setae between the bases of the 3rd and 4th legs (see Barnard, 1950: 9, 76, 111). Both species of Geograpsus recorded here can be specifically identified using the key given by Banerjee (1960). Geograpsus crinipes (Dana, 1851) (Tokelau name - kalamihi) Grapsus crinipes Dana, 1851: 249. Grapsus crinipes Dana, 1852: 341, pl.XXI fig. 6. Geograpsus crinipes; Alcock, 1900: 396 (synonymy) . Geograpsus crinipes; Edmondson, 1959: 162, fig. 4a. Geograpsus crinipes; Banerjee, 1960: 163, figs. le, 3q, 3r, 4a-c. Material examined: Atafu Atoll 8 Jan. 1973, Nautua, 1 male 48mm. 4 Feb. 1973, from house in Atafu village, 1 male 40.5mn. 14 Feb. 1973, Kokoloa, 1 female 30mn. May 1976, coll. P.C. Cotton, 2 females 23.5-3lmn. Nukunonu Atoll 26 Dec. 1966, Teahua Motu, 1 female 41mm. 8 Feb. 1967, Avelau, Long Motu, 1 female 38mm. 26 May, 1968, 1 male 45.5.mn. Fakaofo Atoll 1 Aug. - 19 Sept. 1970, Fenualoa, 1 female 38mm. 1 Aug. - 19 Sept. 1970, 3 females 26-45mm. Remarks: This was the commoner of the two species of Geograpsus found on the Tokelau Islands. It can be identified as the relatively light- coloured, large species G. crinipes, from Banerjee's key (1960: 158). It has the lateral margins of the carapace parallel and not converging posteriorly, keeled throughout their lengths; the cardiac and intestinal regions of the carapace have a series of transverse irregular striae; the suborbital border between the external orbital angle and notch is dentate; the upper border of the buccal cavern is straight; the sternite of the chelipeds is not pubescent, and the lower margins of the meri of the walking legs are only very weakly dentate distally. 17 Although most of the twelve specimens available showed no trace of pubescence on the sternite of the chelipeds, two large females had some faint low pubescence on the anterior part of this sternite. Even the smallest specimens, two females from Fakaofo and a female from Atafu, all with immature triangular abdomens (carapace lengths 26, 27 and 23.5mm respectively), show the characteristic features of this species as listed above, including the fully keeled lateral margins of the carapace (cf. the small females identified below as Geograpsus grayi) . Colour notes: Three more or less different colour patterns were recorded in the preserved material. These differences could not be related to sex or size. Pale coloured individuals were straw or creamy-orange above and paler below. Intermediate coloured individuals were pale to dark grey above, often with darker or more orange frontal and gastric regions, and paler below. A dark individual had a dark brown carapace with the chelipeds, walking legs and sternites paler but with brown patches. Distribution: Widely distributed in the Indopacific ranging from the Red Sea, through islands in the Indian Ocean and the Indonesian Archipelago to the Marshall Islands (Holthuis, 1953), Ellis Islands (Whitelegge, 1897), Line Islands (Edmondson, 1923), Hawaiian Islands (Edmondson, 1959), Tahiti and the Tuamotu Archipelago (Holthuis, 1953), extending to Easter Island in the eastern Pacific (Garth, 1973). The species was first recorded from the Tokelau Islands by Wodzicki (1973c), though the "Geograpsus grayi" records from the Tokelaus given by Hinkley (1969) probably refer to this common species of Geograpsus. Geograpsus grayi (H. Milne Edwards, 1853) Grapsus grayi H. Milne Edwards, 1853: 170. Geograpsus grayi; Alcock, 19200: 395 (synonymy) . Geograpsus grayi; Banerjee, 1960: 159, figs. 1f, 3n-p. Material examined: Nukunonu Atoll Nov. 1966 - Feb. 1967, 1 female 24mm. Fakaofo Atoll 1 Aug. - 19 Sept. 1970, 1 female 19mm. Remarks: Two small female specimens of Geograpsus from the Tokelaus with unkeeled posterolateral margins on the carapace are tentatively identified as G. grayi. Following Banerjee's key (1960) they have the lateral margins of the carapace more or less parallel (but not 18 noticeably converging posteriorly) and keeled anterolaterally but not keeled posterolaterally; the epistome is poorly developed; the suborbital border between the external orbital angle and notch is very slightly dentate, and the sternite of the chelipeds bears a distinct irregular patch of pubescence. Though the underlined features do not agree with Banerjee's key to G. grayi, the unkeeled posterolateral carapace margin and the pubescence on the sternite separate these small specimens from the other species of Geograpsus. The two females identified by us as G. grayi have rounded abdomens of the mature type at carapace lengths of 19 and 24mm. The Fakaofo specimen of G. grayi was collected in the same period and on the same atoll as the smallest available specimens of G. crinipes, two females of 26 and 27mm. These female G. crinipes at that larger size have abdomens in the immature (triangular) abdominal stage as described above. We consider this to be additional evidence that two species of Geograpsus are present on the Tokelau Islands. Mature male specimens of this second species would be needed to confirm this tentative specific identification. Colour notes: The Nukunonu female after preservation was pale creamy-orange above and below, while the Fakaofo female was dark brown on the anterior part of the carapace and pale brown on the posterior part of the carapace, the chelipeds, walking legs and sternites. The hands of the latter specimen were slightly iridescent on the outer surface. Distribution: Widely distributed in the Indopacific ranging from the Red Sea area, through islands in the Indian Ocean and the Indonesian Archipelago to the Marshall Islands (Holthuis, 1953), Wake Island and Fiji (Banerjee, 1960), Niue Island (Sendler, 1923; Yaldwyn, 1970), Cook Islands (Sendler, 1923), Tahiti and the Tuamotu Archipelago (Golthui's> 1953). This is probably a new record for the Tokelau Islands as the "Geograpsus grayi" records from these atolls given in Hinckley (1969) are considered by us to refer to the commoner species G. crinipes. Metopograpsus H. Milne Edwards, 1853 Metopograpsus H. Milne Edwards, 1853: 164. Tweedie, 1949: 466 (key to spp.). Banerjee, 1960: 172 (key to spp.). The species of the genus Metopograpsus are not generally regarded as typical "land crabs" as the few references to habitat published usually record them from mangrove swamps or intertidal sand flats (e.g. Macnae and Kalk, 1962: 27; Macnae, 1966: 80; McNeill, 1968: 80). One species of the genus was collected for K.W. on Atafu Atoll by Dr. Iuta Tinielu during work on rat control and forwarded to us with a 19 collection of land crabs. Metopograpsus as a genus is characterized as follows: the Carapace is quadrangular with the posterior margin narrower than the anterior margin; the front (i.e. the anterior margin between the orbits) is wider than half the carapace width; weak striations are present laterally, but absent medially, on the dorsal surface of the carapace; the anterolateral surfaces of the carapace on each side of the mouth field (the pterygostomial regions) are smooth rather than reticulate; and the third maxillipeds lack an oblique setose ridge on their outer surfaces (see Barnard, 1950). The species of Metopograpsus from Atafu was readily identified as M. thukuhar using the key given by Banerjee (1960). Metopograpsus thukuhar (Owen, 1839) (Tokelau name - lala) Grapsus thukuhar Owen, 1839: 80, pl.24 fig. 3. Metopograpsus thukuhar; Tesch, 1918: 80 (synonymy) . Metopograpsus thukuhar; Banerjee, 1960: 186, figs. 6f, 6g. Metopograpsus thukuar (sic) Forest and Guinot, 1961: 155, figs. 162, 167. Metopograpsus thukuhar; Crosnier, 1965: 25, figs. 20-22, 27. Material examined: Atafu Atoll Dec. 1973, coll. Iuta Tinielu, 2 males 15-18mm, 2 females 13-15mm. Remarks: Dr. Tinielu sent a small collection of crabs from Atafu to K.W. in December 1973 in answer to a request for "land crabs". This consisted of two species of Cardisoma, a species of Sesarma and the four specimens of Metopograpsus listed above. He commented (in litt. 14 Dec. 1973) on the latter as follows "The lala are darkish in colour small in size." From Banerjee's key the following features were found to be diagnostic for M. thukuhar: no tooth on lateral margin of carapace posterior to external orbital angle; free edges of postfrontal lobes rounded and blunt, postfrontal region with ridges and markings (cf. Crosnier, 1965: fig. 21); suborbital tooth blunt and not keeled from tip to base (cf. Crosnier, 1965: fig. 27); base of antenna thickly pubescent; no “pubescent areas" on anterior surface of propodi of lst to 3rd walking legs, no "linear fringe" of setae on upper margin of propodus of 4th walking leg; male abdomen with 6th segment (penultimate) slightly longer than 5th; male lst pleopod with terminal chisel-like chitinous projection (i.e. chitinous projection not T-shaped or apically concave, see Forest and Guinot, 1961: fig. 162; Crosnier, 1965: fig. 20), and female oviducal aperture partly obstructed with a blunt lobe rather than a chitinous projection (cf. Forest and Guinot, 1961: fig. 167). 20 Distribution; Widely distributed in the Indopacific ranging from the east coast of Africa and Madagascar, through islands in the Indian Ocean and the ) Indonesian Archipelago to Japan, Australia, Fiji and Samoa (Banerjee, 1960), Hawaii (Edmondson, 1959; Banerjee, 1960) and Tahiti (Forest and Guinot, L961). This is the first record of this species from the Tokelau Islands. Subfamily SESARMINAE Sesarma Say, 1817 Sesarma Say, 1817: 76. Tesch, 1917: 128 (synonymic list of spp.), 234 (key to Indopacific spp.). Crosnier, 1965: 46 (key to subgenera). Seréne and Soh, 1970: 387 (generic and subgeneric subdivision of Sesarma s.1l.). Members of the tropical and subtropical genus Sesarma s.1l. are commonly referred to as "marsh crabs" and most species are recorded as living in mangrove swamps, mud flats or saline marshes. Even in the absence of mangroves and true marshes (cf. Parham, 1971: 592), a species of Sesarma s.1. occurs on the Tokelau Islands. This is regarded by us as a "land crab" as all specimens taken were collected on kanava trees (Cordia subcordata) or coconut palms at a distance from the beach. This Sesarma has already been recorded from the Tokelau Islands by Laird (1955, 1956). His specimen was taken with mosquito larvae from a "rot-hole" in a puka tree (Hernandia peltata) on "Motusanga" (Motuhaga), Nukunonu in June 1953. There are over one hundred species of Sesarma s.l. in the Indopacific area (Seréne and Soh, 1970). This multiplicity of species has made classification difficult both at the specific and generic level. Recently the genus Sesarma in its wide sense has been broken up into a number of new genera and subgenera by Seréne and Soh (1970) as a step towards the ultimate revision of this whole group of allied species. The Tokelau Sesarma belongs to the new genus Labuanium as defined in this reclassification. In our essentially faunal study we will take a conservative systematic view and use Labuanium Serene and Soh, 1970, in a subgeneric sense. Sesarma (Labuanium) ? gardineri Borradaile, 1900 (Tokelau name - ataata o hiliao) Fig. 4 References to S. gardineri: Sesarma gardineri Borradaile, 1900: 593, pl. XLII fig. 8. | Sesarma gardineri; Nobili, 1905: 497. . Sesarma gardineri; Tesch, 1917: 194 (in synonymy of S. rotundatum Hess, 1865). 21 Material examined: Atafu Atoll 16 Jan. 1973, Nautua, 1 male 28mm. Dec. 1973, coll. Iuta Tinielu, 3 males 26-28mm. May 1976, coll. P.C. Cotton, 1 female 33.5mm. Nukunonu Atoll 18 Feb. 1967, 1 male 30mm. Remarks: This relatively uncommon, pinkish orange, tree-climbing crab is readily recognizable as a species of Sesarma s.1l. by its lateral carapace margins being nearly straight rather than strongly convex, by the distinctly reticulate nature of the pterygostomial region, by the front of the carapace being bent downwards almost at right angles to the dorsal surface of the carapace with strong postfrontal lobes at this angle, by the basal segment of the antenna being in communication with (i.e. not excluded from) the orbit, and by the absence of teeth on the posterior edge of the walking leg meri (cf. Barnard, 1950; Crosnier, 1965). In Tesch's (1917: 235) or Crosnier's (1965: 47) subdivision of the genus the Tokelau species would be placed in the subgenus Sesarma as it has two teeth on the lateral border of the carapace behind the external orbital angle, and the upper border of the hand has only one or two simple longitudinal ridges (and no pectinated crest). In Tesch's key (1917: 238) to the Indopacific species of Sesarma, subgenus Sesarma, the Tokelau material can not be identified as S. rotundatum Hess in couplet 46 as the sides of the carapace are posteriorly divergent and not regularly convex, nor as S. trapezoidea Guérin in couplet 49 as the upper border of the male movable finger is not "regularly and transversely milled". Using Seréne and Soh's (1970) key to genera and subgenera of the Sesarma complex the Tokelau species has the following diagnostic features (in addition to those listed above): basal segment of antennule somewhat swollen and about as broad as long, walking legs long with anterior and posterior borders of meri subparallel for most of segment, breadth of front just a little shorter (regarded as "Subequal" for the key) than breadth of posterior border of carapace, postfrontal lobes strongly ridged anteriorly, carapace lateral border slightly diverging posteriorly and almost concave, walking leg dactyls about half length of propodi, male abdomen relatively narrow and elongate. Thus it can be placed in Labuanium though the lateral borders of the carapace are not "slightly convex". Seréne and Soh list eight Indopacific species in their Labuanium group. Serene (in litt., 10 June 1975) considers that the species of this group are probably all palm-tree dwellers. 22 Following correspondence and discussions with Dr. Seréne in 1975- 76 on the identity of this Sesarma species, we now regard it as conspecific with the species recorded as the "red tree-crab" and described from Funafuti (in the Ellice Islands) and Rotuma (north-west of Fiji) as Sesarma gardineri by Borradaile in 1900. S. gardineri has been synonymized with S. rotundatum Hess by Rathbun (1907: 33) and Tesch (1917), consequently it was not listed as a separate species under Labuanium by Seréne and Soh in 1970. Using an unpublished key to the species of Labuanium provided for us by Seréne, the Tokelau specimens are identified as "L. gardineri" from the following characters: frontal margin of carapace with a weak median concavity, anterior margin of postfrontal lobes not strongly crested and armed with blunt (rather than sharp) tubercles, lateral border of carapace diverging somewhat from behind external orbital angle and nearly straight rather than medially convex, innder surface of male palm with scattered granules arranged neither in vertical nor in transverse rows, upper borders of male palm with a finely granulate longitudinal line, upper border of male free finger (cheliped dactyl) with irregular longitudinal row of 13-15 acute conical tubercles, lower border of fixed finger with acute tubercles, walking leg dactyls about half length of propodi and thickly tomentose on both borders. As the status of S. gardineri vis-a-vis S. rotundatum is not at all clear we have followed Sexéne's advice and used this specific name with a question mark. It would appear that the nearly straight (though diverging) lateral carapace margins and the walking leg dactyls being at least half the length of the propodus may serve to distinguish Borradaile's species from S. rotundatum with its medially convex carapace margins and walking leg dactyls less than half propodi. S. rotundatum was originally described from Sydney in eastern Australia but this temperate locality is probably incorrect as the species has not been recorded from Australia since. The correct nomenclature and relationship of these oceanic island, tree-climbing Sesarma must await a complete revision of the whole group. Note on arboreal habits: This Sesarma was seen only on overgrown eastern islets, such as Nautua and Te Ahaga, of Atafu Atoll (see fig. 1) by K.W. in 1973. It was observed on rainy days running on coconut trunks or on kanava branches and sometimes on the ground. Laird took his specimen (1955) on a puka tree at Nukunonu, and Holthuis (1953: 33) recorded what was probably the same Sesarma from a "hole in live coconut trunk" on Ujae Atoll in the Marshalls. Colour notes: As mentioned above this crab on the Tokelau was recorded as being “pinkish orange in life". Borradaile's specimens (1900) were described as "red", while Holthuis (1953) noted that his specimen from Ujae was “orange-gray". ——— 23 Distribution: Sesarma gardineri has been recorded in the Indopacific from New Guinea, the Ellice Islands, Rotuma and the Tokelaus. It was first recorded from the Tokelau-Islands by Laird (1955, 1956) and later by Wodzicki (1973c). S. rotundatum has been recorded from the Nicobar Islands in the Indian Ocean, Java, New Guinea, Duke of York Island near New Britain, the Caroline and Marshall Islands in Micronesia, Samoa and Hawaii (Tesch, 1917). The Duke of York Island between New Britain and New Ireland, recorded as a locality for S. rotundatum by Miers (see Tesch, 1917), is not the same as the older European name (Duke of York Island) for Atafu Atoll in the Tokelaus. Family GECARCINIDAE The typical land crab family 1905. Catalogue of the Indian decapod Crustacea in the collection of the Indian Museum. Part II. Anomura. Fasciculus I. Pagurides. Indian Museum, Calcutta. 197 pp., 16 pls. Annual Report, 1974. Reports on Niue and the Tokelau Islands [by the] Maori and Island Affairs Department for the year ended 31 March 1974. Government Printer, Wellington. 44 pp. (Appendix Jour. House of Representatives N.Z. 1974, E14). Banerjee, S.K. 1960. Biological results of the Snellius Expedition XVIII. The genera Grapsus, Geograpsus, and Metopograpsus (Crustacea Brachyura). Temminckia 10: 132-199, 6 figs. 46 Barnard, K.H. 1950. Descriptive catalogue of South African decapod Crustacea (crabs and shrimps). Ann. "SS. Afr. Muss So. 1-837, 154 figs. Boardman, D.W. 1969. Vocabulary Tokelau - English English - Tokelau. Islands Education Division, N.Z. Department of Education for Department of Maori and Island Affairs, Wellington. 66 pp. Boone, L. 1934. Scientific results of the world cruise of the yacht "Alva", 1931, William K. Vanderbilt, commanding. Crustacea: Stomatopoda and Brachyura. Bull. Vanderbilt Mar. Mus. 5: 1-210, 109 pls. Borradaile, L.A. 1900. On some crustaceans from the South Pacific. Part IV. The crabs. Proc. Zoo. Soc. Lond. 1900: 568-596, pls XL-XLII. Bright, D.B. and Hogue, C.L. 1972. A synopsis of the burrowing land crabs of the world and list of their arthropod symbionts and burrow associates. Contr. Sci. Nat. Hist. Mus. Los Angeles 220: —58r Churchward, C.M. 1959. Tongan Dictionary (Tongan-English and English-Tongan) . Oxford University Press, London. 836 pp. Crosnier, A. 1965. Crustacés Décapodes Grapsidae et Ocypodidae. Faune Madagascar 18: 1-143, 260 figs, 11 pls. Dalia bese L959. Notes on some insects and other invertebrates collected in the Tokelau Islands. N.Z. Entomologist 2(4): 1-8. Dana; wu. Ds) 185153 Conspectus crustaceorum quae in Orbis Terrarum circumnavigatione, Carolo Wilkes e Classe Reipublicae Foederatae duce, lexit et descripsit. Catometopa or Grapsoidea. Proc. Acad. Nat. Sci. Philad. 5: 247=254. ----------- 1852. United States Exploring Expedition during the years 1838, 1839, 1840, 1841, 1842 under the command of Charles Wilkes, U.S.N. Vol. 13, part 1, Crustacea. Text 1393 pp. Atlas (l855)" 27" pp’.; lo pus. === == === iL{3}7/S)— Corals and coral islands. 2nd ed. Sampson Low, Marston, Low and Searle, London. 348 pp., maps and figs. Doumenge, F. 1966. L'homme dans le Pacific Sud. Musée de 1'Homme, Paris. 635 pp., 83 photo pls. (Publs Soc. Océanistes 19). Edmondson, C.H. 1923. Crustacea from Palmyra and Fanning Islands. Bull. Bernice P. Bishop Mus. 5: 1-43, 2 pls. SSSSSSS Sasa 1959. Hawaiian Grapsidae. Occ. Pap. Bernice P. Bishop Mus. 22(10): 153-202, 27 figs. SOS SSSSSeeS 1962. Hawaiian Crustacea: Goneplacidae, Pinnotheridae, Cymopoliidae, Ocypodidae, and Gecarcinidae. Occ. Pap. Bernice P. Bishop Mus. 23(1): 1-27, 10 figs. Ehrhardt, J.P. 1968. Recensement en 1968 de la population de Gecarcinus planatus Stimpson sur 1'Ilot de Clipperton. Centre de Recherches du Service de Santé des Armées. (Division de Biologie Générale et Ecologie) 40/BIO-ECO: 1-9. 47 Fize, A. and Serene, Reo Se Les Pagures du Vietnam. Notes Inst. Oceanogr. Nhatrang 45: 1-228, 6 pls. Forest, J. 1954, Crustacés decapodes marcheurs des Iles de Tahiti et des Tuamotu. - I. Paguridea (suite). Bull. Mus. natn. Hist. nat., Paris (2)26(1): 71-79, figs 15-24. Forest, J. and Guinot, D. 1961. Crustacés Décapodes Brachyoures de Tahiti et des Tuamotu. Foundation Singer —- Polignac, Paris. (Exped. Franc. Recifs Coral. Nouvelle-Calédonie 1960-62, Vol. Prelim.) 195 pp., 178 figs, 18 pls. Garth, J.S. 1973. The brachyuran crabs of Easter Island. Proc. Calif. Acad. Sci. 39(17): 311-336, 6 figs. Gordan, J. 1956. A bibliography of pagurid crabs, exclusive of Alcock, 1905. Bull. Am. Mus. Nat. Hist. 108(3): 253-352. Guinot, D. 1967. Les crabes comestibles de 1'Indo-Pacifique. Foundation Singer-Polignac, Paris. 145 pp., 23 figs, 10 pls. Henderson, J.R. 1888. Report on the Anomura collected by H.M.S. Challenger during the years 1873-76. Rep. Scient. Res. Voy. Challenger (69) Zool. 27(1): 1-221, 21 pls. Herbst, J.F.W. 1791-1796. Versuch einer Naturgeschichte der Krabben und Krebse nebst einer systematischen Beschreibung ihrer verschiedenen Arten. Band 2. G.A. Lange, Berlin und Stralsund. 225 pp., pls. 22-46. Hinckley, A.D. 1969. Ecology of terrestrial arthropods on the Tokelau Atolls. Atoll Res. Bull. 124: 1-18. Hinds, V.T. 1971. A rapid fisheries reconnaissance in the Tokelau Islands August 18th-25th, 1971. South Pacific Commission, Noumea. Mimeographed report (1069/71). 14 pp. 48 Hodehuas, i. Bo O53) Enumeration of the decapod and stomatopod Crustacea from Pacific coral islands. Atoll Res. Bull. 24: 1-66, 2 maps. Hooper, A. and Huntsman, J. 1973. A demographic history of the Tokelau Islands. J. Polynesian Soc. 82(4): 366-411. Ingram, W.M. 1940. Cypraeidae from Atafu Island, Union Group. J. Conch., Lond. 21(7): 213-214. Kennedy, T.F. 1966. A descriptive atlas of the Pacific islands. A.H. & A.W. Reed, Wellington. 65 pp., 62 figs. Kalifi, H. and Webster, J.H. n.d. Tokelau-English Dictionary. Tokelau Islands. 127 pp. Kirkpatrick, R.D. 1966. Mammals of the Tokelau Islands. J. Mamm. 47: 701-704. Kramer, A. 1903. Die Samoa-Inseln. Entwurf einer Monographie mit besonderer Berucksichtigung Deutsch-Samoas. E. Schweizerbartsche Verlagsbuchhandlung, Stuttgart. Vol. II. 445 pp. 147 figs. Krauss, N.L.H. 1969. Bibliography of the Tokelau or Union Islands, Central Pacific. N.L.H. Krauss, Honolulu. 11 pp. ----------- LOPFO'S Bibliography of Swains Island, American Samoa. N.L.H. Krauss, Honolulu. 7 pp. Laird, Mew dS55s Notes on the mosquitoes of the Gilbert, Ellice and Tokelau Islands, and on filariasis in the latter group. Bull. Ent. Res. 46(2): 291-300, 1 fig. SSSSSSSS 55> 1956. Studies of mosquitoes and freshwater ecology in the South Pacific. Bull. .R. Soc. °N.Z. 62 1-213; 62) tigen ----------- 1963. Rats, coconuts, mosquitoes, and filariasis. In: Gressitt, J.L. (editor). Pacific Basin Biogeography (Tenth Pacific Science Congress, Honolulu, 1961). Bishop Museum Press, Honolulu. Pp. 535-542, 4 figs. ----------- 1966. Integrated control and Aedes polynesiensis: an outline of the Tokelau Islands project, and its results. WHO/ Vector/Control 66. 204. 10 pp. Latreille, P.A. 1825. Familles naturelles du Regne Animal, exposées succinctement et dans un ordre analytique avec l'indication de leurs genres, etc. Paris. 5/70 pp. 49 =---------- 1825-28. Entomologie, ou histoire naturelle des Crustacés, des Arachnides et des Insectes. Encyclopédie Methodique vol. X. Paris. 832 pp. [Sherborn and Woodward, Proc. Zool. Soc. Lond. 1899: 595 state that pp. 1-344 were published in 1825 and pp. 345-832 in 1828. ] Linnaeus, C. 1767. Systema Naturae per Regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. 12th ed., vol.2. Pp. 533-1327. McEwen, J.M. 1970. Niue Dictionary. Department of Maori and Island Affairs, Wellington. 386 pp. Macgregor, G. 1937. Ethnology of Tokelau Islands. Bull. Bernice P. Bishop Mus. 146: 1-183, 10 pls. Macnae, W. 1966. Mangroves in eastern and southern Australia. Aust. J. Bot. 14(1): 67-104, 9 figs, 3 pls. Macnae, W. and Kalk, M. 1962. The ecology of the mangrove swamps at Inhaca Island, Mogambique. J. Ecol. 50: 19-34, 2 figs. McNeill, F.A. 1968. Crustacea, Decapoda & Stomatopoda. Scient. Rep. Gt. Barrier Reef Exped. 7(1): 1-98, 2 pls. Marples, B.J. 1955. Spiders from some Pacific islands. Pacif. Sci. 9(1): 69-76, 2 figs. Milne Edwards, H. 1837. Histoire Naturelle des Crustacés, comprenant l’anatomie, la physiologie et la classification de ces animaux. Vol.II. Lib. Roret, Paris. 532 pp. SSSSSSS Saas 1853. Mémoire sur la famille des Ocypodiens. Ann. Sci. Nat. Zool. (3)20: 163-228, pls 6-11. Milner, G.B. 1966. Samoan Dictionary Samoan-English English-Samoan. Oxford University Press, London. 465 pp. Miyake, S. 1939. Notes on Crustacea Brachyura collected by Professor Teiso Esaki's Micronesia Expeditions 1937-1938 together with a check list of Micronesian Brachyura. Rec. Oceanogr. Work. Japan 10: 168-247, pls 12-17. Mosby, J.M. and Wodzicki, K. 1972. Some parasites of the Kimoa (Rattus exulans) on the Tokelau Islands. N.Z. Jl Sci. 15(4): 698-704. Mosby, J.M., Wodzicki, K. and Shorland, F.B. 1974. Fatty acid composition of the depot fats of the Polynesian rat, Rattus exulans, Tokelau Islands. WN.Z. Jl Zool. 1(1): 67-70. 50 Mosby, J.M., Wodzicki, K. and Thompson, H.R. 1973. Food of the Kimoa (Rattus exulans) in the Tokelau Islands and other habitats in the Pacific. N.Z. Jl Sci. 16(4): 799-810. Niering, W.A. 1956. Bioecology of Kapingamarangi Atoll, Caroline Islands: terrestrial aspects. Atoll Res. Bull. 49: 1=32,7 333higse Nobili juGeee L905. Decapodi e Isopodi della Nuova Guinea Tedesca raccolti dal Sign. L. Bir6é. Ann. Mus. Natn. Hungar. 3: 480-507, pls XII-XIII. Parhamije BekiaVicu 1 LOWS The vegetation of the Tokelau Islands with special reference to the plants of Nukunonu Atoll. NwZ. Jl Bot. 9(4): 576-609, 11 figs. Pell, Hs vani, L958: A survey of fisheries in the Tokelau Islands. South Pacific Commission, Noumea. Mimeographed report. 16 pp. Pukui, M.K. and Elbert, S.-H. 1957. Hawailan-English Dictionary. University of Hawaii Press, Honolulu. 362 pp. Quoy, J. and Gaimard, J.P. 1824-1826. Zoologie. In: Freycinet, L. Voyage autour du monde, entrepris par ordre du Roi, sous le ministere et conformement aux instructions de S. Exec. M. le Vicomte du Bouchage, Secrétaire d'Etat au Départment de la Marine, execute sur les corvettes de S.M. 1'. Uranie et la Physicienne, pendant les annees 1817, 1818, 1819 et 1820. Pillet aine, Paris. 712 pp. Atlas, 96 pls. [Sherborn and Woodward, Ann. Mag. Nat. Hist. 7, 7: 392 state that pp. 1-328 were published in 1824, pp. 329-616 in 1825, and 617-712 in 1826. } Rathbun, M.J. 1907. Reports on the scientific results of the Expedition to the tropical Pacific, in charge of Alexander Agassiz, by the U.S. Fish Commission Steamer Albatross, from August, 1899, to March, 1900, Commander Jefferson F. Moser, U.S.N., commanding. IX. Reports on the scientific results of the Expedition to the eastern tropical Pacific, in charge of Alexander Agassiz, by the U.S. Fish Commission Steamer Albatross, from October, 1904, to March, 1905, Lieut.—-Commander L.M. Garrett, U.S.N. commanding. X. The Brachyura Mem. Mus. Comp. Zool. Harv... (35(2):421—747 29 Epis: DSSS SS 5555= 1910. Decapod crustaceans collected in Dutch East India and elsewhere by Mr. Thomas Barbour in 1906-1907. Bull. Mus. Comp. zool. Harv. 5216): 305-317, 6) pis. ----------- 1918. The grapsoid crabs of America. Bull. U.S.) Nate Mus. 97: 1-461, 161 pls. Sul Reyne, A. 1939. On the food habits of the Coconut Crab (Birgus latro L.), with notes on its distribution. Archs Neerl. Zool. 3(2/3): 283-320, 1 fig. Savage, S. 1962. A Dictionary of the Maori language of Rarotonga. Department of Island Territories, Wellington. 460 pp. Say, T. 1817. An account of the Crustacea of the United States. J. Acad. Nat. Sci. Philad. 1: 57-80. Sendler, A. 1923. Die Decapoden und Stomatopoden der Hanseatischen Sudsee-Expedition. Abh. Senckenb. naturf. Ges. 38(1): 21-47, pls 20-21 (also numbered 5-6). Serene, R. and Soh, C.L. 1970. New Indo-Pacific genera allied to Sesarma Say 1817 (Brachyura, Decapoda, Crustacea) . Treubia 214): 387-416, Si pls. Sharples, P.R. 1970. An orthography of the language of the Tokelaus. University of Auckland Anthropology Depart. Working Papers in Maori Studies and Linguistics 13: 1-6. SSS SS = 1976. Tokelau syntax: studies in the sentence structure of a Polynesian language. Unpublished Ph.D. thesis, University of Auckland. Smith, Hod. 1969). Rodent research in the Gilbert and Ellice Islands Colony 1967-1969 with control recommendations. Mimeographed report issued by the Department of Agriculture, Gilbert and Ellice Islands Colony, Tarawa. 96 pp., 2 maps. Stimpson, W. 1858. Prodromus descriptionis animalium evertebratorum, quae in Expeditione ad Oceanum Pacificum Septentrionalem, a Republica Federata missa, Cadwaladaro Ringgold et Johanne Rodgers Ducibus, observavit et descripsit. Proc. Acad. Nat. Sci. Philad. 10: 93-110. Storer, T.I. (editor) 1962. Pacific island rat ecology report of a study made on Ponape and adjacent islands 1955-1958. Bull. Bernice P. Bishop Mus. 225: 1-274, 73 figs. Terao, A. 1913. A catalogue of hermit-crabs found in Japan (Paguridea excluding Lithodidae), with descriptions of four new species. Annot. Zool. Japonenses 8: 355-391, 4 figs. Tesch, J.J. 1917. Synopsis of the genera Sesarma, Metasesarma, Sarmatium and Clistocoeloma, with a key to the determination of the Indo-Pacific species. Zool. Meded., Leiden 3(2-3): 127-260, 8 figs, pls XV-XVII. ----------- ORB: The Decapoda Brachyura of the Siboga Expedition I. Hymenosomidae, Retroplumidae, Ocypodidae, Grapsidae and Gecarcinidae. Siboga Exped. Monogr. 39c: 149-295, 6 pls. Tinker, S.W. 1965. Pacific Crustacea: an illustrated handbook on the reef-dwelling Crustacea of Hawaii and the South Seas. Charles E. Tuttle, Rutland and Tokyo. 134 pp., 52 pls. Turkay, M. 1973a. Die Gecarcinidae Afrikas (Crustacea: Decapoda) . Senckenbergiana biol. 54(1-3): 81-103, 18 figs. —---=----- 1973b. Zur Synonymie von Epigrapsus notatus und Cardisoma carnifex (Crustacea: Decapoda). Senckenbergiana biol. 54(1-3): 105-110, 8 figs. Se 1973c. Bermerkungen zu einigen Landkrabben (Crustacea, Decapoda). Bull. Mus. natn. Hist. nat., Paris 3, 142 (Zool. 106): 967-978, 2 pis. ----------- 1974. Die Gecarcinidae Asiens und Ozeaniens (Crustacea: Decapoda) . Senckenbergiana biol. 55(4-6): 223-259, 19 figs. Tweedie, M.W.F. 1949. The species of Metopograpsus (Crustacea, Brachyura) . Bijdr. Dierk., Amsterdam 28: 466-471, 1 fig. Whitaker, A.H. 1970. A note on the lizards of the Tokelau Islands, Polynesia. Herpetologica 26(3): 355-358. Whitelegge, T. 1897. The Crustacea of Funafuti. Aust. Mus. Mem. 3(2): 127-151, pls VI-VEL. Wiens, H.J. 1962. Atoll environment and ecology. Yale University Press, New Haven. 532 pp., 88 pls. Wodzicki, K. 1968a. An ecological survey of rats and other vertebrates of the Tokelau Islands 19 November 1966-25 February 1967. Mimeographed report prepared for Tokelau Islands Administration and issued by Departments of Maori and Island Affairs and Scientific and Industrial Research, Wellington =...132 .ppsj, 710 (igs: ----—————- 196 8b. The Tokelau rat survey 2. Follow-up report, 18 April-15 June, 1968. Mimeographed report prepared for Tokelau Islands Administration and issued by Departments of Maori and Island Affairs and Scientific and Industrial Research, Wellington. 41 pp., 3 figs. --------—— 1969a. Preliminary report on damage to coconuts and on the ecology of the Polynesian rat (Rattus exulans) in the Tokelau islands. Proc. N.Z. Ecol. Soc. 16> 7—12, 1 fig: 53 ----------- 1969b. A preliminary survey of rats and other land vertebrates of Niue Island South Pacific 2 November - 4 December 1968. Mimeographed report prepared for Niue Island Administration and issued by Departments of Maori and Island Affairs and Scientific and Industrial Research, Wellington. 39 pp., 2 figs. ----------- 1970. Report on results of rat control trials in the Tokelau Islands from 30 July to 20 September 1970 and recommendations for a comprehensive scheme of rat control. Mimeographed report issued by Departments of Maori and Island Affairs and Scientific and Industrial Research, Wellington. 42 pp. == 1972a. Effect of rat damage on coconut production on Nukunonu Atoll, Tokelau Islands. Oléagineux 27(6): 309-314, 5 figs. ----------- 1972b. The Tokelau Islands (men and introduced animals in an atoll ecosystem). South Pacif. Bull. 22(1): 37-41, 7 figs. ----------- 1973a. The Tokelau Islands - environment, natural history and special conservation problems. In: S.P.C. Regional symposium on conservation of nature - reefs and lagoons. Noumea, New Caledonia, 5 to 14 August 1971. Proceedings and papers. South Pacific Commission, Noumea. Paper 10: 63-68. SSS SSS SSS5= 1973b. Some problems arising from the invasion and deliberate introduction of exotic plant and animal species in the South-west Pacific. In: S.P.C. Regional symposium on conservation of nature - reefs and lagoons. Noumea, New Caledonia, 5 to 14 August 1971. Proceedings and papers. South Pacific Commission, Noumea. Paper 31: 239-244. ----------- 1973c. A survey of rats on Atafu Atoll, Tokelau Islands and recommendations for a scheme of control 6th December 1972— 21st February 1973. Mimeographed report prepared for the Tokelau Islands Administration and issued by Department of Maori and Island Affairs, Wellington. 49 pp., 1 fig. Wodzicki, K. and Laird, M. 1970. Birds and bird lore in the Tokelau Islands. Notornis 17(4): 247-276, pls XXX—-XXXIV. [Yaldwyn, J.C.] 1970. Oceanic crabs for Niue Island's new issue. Phil. Bull., Wellington (N.Z. Post Office) 4: 2, 3 figs. h care th £. 3 | 3 (1 gurwe nO aut: hipip ektetemst pt IHV1dOW & | OW 48 $9 3 YALAWN i E ‘ VOVHY SL iy : 9 : é VAVIVIVhW) & Bees. Nd3H 3h vO1Ox0Ox TOKELAU TE FAKANAVA Zz TEPUKA TEKAMU NUKUNONU VILLAGE MOTUHAGA NAHUA 600 te) 1000 Fig. 2. Map of Nukunonu Atoll, based on N. Z. Lands and Survey Department Aerial Plan No. 1936/7B sheets 1 and 2 (1974). \ RM a a re i a roa x.) a a ne MATAGI & ss a ~ ~ a na = F 2 a > rR : @ NY am, Z S ri Lg L Mig i NS % Bp ¥ 895 FENUAFALA ty a # % Z “hig, * 3 4, 3 4 a 4,0 Ee FAKAOFO VILLAGE & 2 Re i 7 %, he = i, Dn, rs eet, is 4 om Ye O% % 2 7% %, 0%, TETIALAU N %, %y % ag a - a Ard 40 3° % 2? em SAUMAFATAGA % Fe ~ & z SAKEA FENUALOA 600 ct) 1000 2000 EY Fig. 3. Map of Fakaofo Atoll, based on N. Z. Lands and Survey Department Aerial Plan No. 1036/7C (1974). Cardisoma rotundum. Dorsal view of male, carapace length 41.5 mm from Village Motu, Nukunonu. (Photo T.R. Ulyatt) Fig. 4. Sesarma (Labuanium) ?gardinert length 28 mm from Nautua, Atafu. (Photo T.R. Ulyatt, National Museum of N.Z.) Dorsal view of male, carapace Fig. 5. Cardisoma carnifex. Dorsal view of female, carapace length 64 mm from Atafu. (Photo R.R. Ulyatt) ATOLL RESEARCH BULLETIN NO. 236 SOME ASPECTS OF THE ECOLOGY OF REEFS SURROUNDING ANEGADA, BRITISH VIRGIN ISLANDS by R.P. Dunne and B.E. Brown Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 YrImUa2s0 40 18944 @pNiBU07 ne . a2 Uo o-< yaeg ui yBiey poy poy bog f2UuopW eueS < S| sewer! r t yeoys ) p21 $2} 49 91225 rome PEA PS SGNV1SI NIDUIA 3HL ou ae 1 Dee) HayD peog | 22Buin * . tA & yoy Aas wie . abe wajesnier vayey B x Sy) snoqieyy O8LI abessey y> d7 °Y PL, “n 0, Pry : % IM © | 06eq5)~ Vg up = e Be) orzt vIOluoL ed Apurse ary hed v991 be) 001. SMUUNDUOW | oBeqoy VaYOO NIDYIA ie begin bog aw) = | ig soueleg 1's ®0, a > 1 ae ‘0, 2o T3) Sur ou ova 7A ayhq|uen sof 9 oF i unos u, P Si oUnbsoyw bu, | @199KN o | SSA : V13NZ3N3A ess0H) ouM ey! pepiuiy o6e40, epevaing ‘ qereved Qiva2u, 1S Qersny 4S Renbwuew Q e21uiwog 1 m. The corals, Millepora squarrosa, ('boxwork form' Stearn and Riding 1975) and Agaricia agaricites, var. crassa were particularly abundant at the edge of spurs where the water depth was approximately 1 m. These spurs measured up to 4-5 m in height and ranged from 4-10 m in width. Beyond this zone the bottom sloped gently away at 6-8 m with no marked drop off seaward. In contrast to the previous areas, the buttress zone of Jack Bay (transects 1 and 2) yielded a much greater proportion of live coral together with a greater variety of coral species (Table 1). The spur and groove effect, however, was not as evident at Jack Bay as at sites towards the west end of the island. At the seaward end of transect 1 the dominant corals were similar to those quoted in the previous paragraph. Particular note was made of the growth form of Acropora palmata which was encrusting as opposed to branching. Species such as Isophyllia multiflora and Isophyllastrea rigida were recorded in the buttress zone but were limited in their abundance. The buttress zone of transect 2 was particularly significant in terms of the presence of the plate-like form of Montastrea annularis and also the abundance of Agaricia agaricites, the more typical form with thick leaf-like outgrowings and corallites on all sides of the colony. Aliso on this transect, at a depth of 10 m, a record of Mussa angulosa was obtained - this was the only site at which the coral was recorded. Beyond the buttresses formed by Montastrea annularis, the bottom levelled off at approximately 12 m, with evidence of an ‘old' spur and groove formation at this depth. These spurs were aligned simil- arly to those recorded at West End, although their origins were not investigated by coring. The height of the spurs was 0.5 - 0.8 m above the sand filled grooves between them - the spurs themselves being covered by sand and algae, particularly Turbinaria spp. From the final marker on transect 2, snorkellers swam offshore, beyond the reef, for a distance of 1 km. The bottom was similar in type to that area just beyond the buttress zone, with intermittent large pillars of Montastrea annularis arising out of a sand substrate which was traversed by 'tongues' of rock outcrops covered by sand and algae. The depth gradually increased to 15 m about 2 km offshore at this point. 20 Leeward Shore The southern and western patch reefs border on the shallow waters (2-8 m) of Gorda Sound. Charts of the area show a considerable number of coral heads breaking surface, for a distance of up to 3 km from the southern shore of the island, and patch reef development is quite extensive (Plates 1 and 13). The size of the patches varies from between 3-4 m diameter to over 100 - 200 m diameter. Profiles of the patch reefs are shown in Fig. 13. The limited amount of time available for study resulted in the detailed description of only a small number of areas. Eight patch reefs were studied in all; their position relative to the island is shown in Fig. 4. Of those patch reefs investigated, the majority showed an essentially similar structure, independent of their size, apart from patch reefs III and IX which were situated in shallow waters (1-2 m) and were much closer inshore than the other sites visited. The basic structure of most of the patch reefs consisted of a central core of Acropora palmata surrounded on the outer edges by Montastrea annularis heads and large colonies of Diploria spp. The latter corals formed discrete heads which were based on sand in depths of water ranging from 5-6 m. A comparison of the profiles obtained for each patch reef illustrates the similarity in basic structure. The dominant coral on all patches was Acropora palmata and was frequently dead and encrusted by coralline algae. Generally both the abundance and diversity of coral species was greater on the leeward side of the island than on the windward shore (Table 1) with a maximum of 22 species recorded on patch reef II. The abundance of Acropora prolifera was particularly noticeable on the leeward shore especially on patch reefs II, VII and IX. In each case the colonies were all localised on the shoreward facing side of the patch. Acropora cervicornis was also relatively abundant - particularly on patch reefs VI, VII and VIII. A notable feature on patch reef VIII was an intermediate form of Acropora palmata and Acropora cervicornis, in which cervicornis-like protuberances projected from Acropora palmata branches. Similar growth forms have been noticed by Roos (1971) on Bonaire and St. Martin and obviously such observations cast doubts about the species distinction. A variety of growth forms of other corals were noted, e.g., on patch reef II and IV, two forms of Porites astreoides were observed - the typical encrusting form and also a plate-like variety (Plate 14). Roos (1967) has already shown that in localities where light is limiting the coral shows a conspicuous flattening in response to ambient light conditions. 21 Agaricia agaricites var. crassa and var. fragilis were also recorded on patch reef II. Generally Agaricia agaricites var. fragilis was restricted to overhangs and gullies and var. crassa to shallow water on top of the patch reef. Millepora also displayed several forms on this patch reef (Plate 15). In this example both the bladed Millepora complanata and the branching Millepora alcicornis are illustrated, in a water depth of 3 m. At only one site did Montastrea annularis exhibit a vertical plate form similar to that recorded in the buttress zone of Jack Bay - this growth form is characteristic of the species when growing on shadowed vertical surfaces (Roos 1971). Coral species which were either absent or rare at sites visited on the leeward side of the island included Mycetophyllia lamarkana, Isophyllastrea rigida, Tubastrea coccinea and Manicina areolata. The latter species, however, was rated as common on patch reef VI where detached specimens were found on the sand at 3-4 m water depth, between the pillars of Montastrea annularis surrounding the patch. In general Gorgonids were abundant at all sites - the dominant species being Gorgonia flabellum, Pseudopteryogorgia americana and Plexaura spp. Apart from patch reefs III and IX, relatively few algae species were recorded on the other patch reefs visited. Calcareous species noted in sand around the reefs included Halimeda opuntia, Penicillis capitatus and Udotea flabellun. The Queen conch, Strombus gigas, was recorded around patch reef IV (approximate density 8 per 200 sq. m.) and also around patch reef VI (5 per 200 sq. m.). Patch reefs III and IX were unlike other sites visited on the leeward shore of the island since they were in much shallower water (1-2 m) and as a result were colonised by many algae species. Patch reef III at Pomato Point was located only 50 m from the shore and consisted of mainly dead Acropora palmata which had been encrusted and consolidated by coralline algae and also extensively bored. Coral species, resistant to sedimentation such as Siderasterea radians and Diploria clivosa were common. Patch reef IX consisted of similar dominant coral species and was also covered by abundant algal growth with dense Thalassia between coral outcrops. A complete list of dominant algae species collected at these sites can be found in Table 2. 22 SUMMARY Altogether 31 species of coral were recorded on Anegada - that is including sites visited on both the windward and leeward sides of the island. This figure compares with 37 species from Cuba (Duarte Bello 1961); 34 species from Puerto Rico (Almy and Carrion Torres 1963); 34 species from Barbados (Lewis 1960); 29 species from Curacao (Roos 1964); 62 species from Jamaica (Goreau and Wells 1967) and 46 Scleractinian species from Bonaire (Scatterday 1974). All these collections involved the use of similar skindiving/snorkelling techniques. However, as Goreau and Wells state in their paper on the shallow water Scleractinia of Jamaica, it is important to define the depth range in which these studies were carried out, e.g., Lewis worked to 10 m; Roos and Scatterday to 30 m, Goreau to 96 m, while other workers restricted their collections to depths of less than 10m in areas that were mainly on the inner lagoon side of reefs. The maximum depth worked around Anegada by members of the Expedition was 10 nm. The apparent absence of Madracis decactis may be accounted for by the worked depth of only 10 m, since this species is usually restricted to deeper water, the shallowest record from material examined in the Netherland Antilles being 9.4 m (Roos 1971). Scatterday (1974) however reports abundant Madracis decactis colonies in well concealed cavities in shallow waters on the Kralendyk reefs of the Netherland Antilles. Of those species noted on Anegada, the largest number recorded on the windward shore occurred within the rear zone of Jack and Loblolly Bay. Up to 24 species were observed here and these figures compare well with the numbers of coral species recorded on individual patch reefs on the leeward shore (17 - 24). The number of corals recorded on the reef top and buttress zone however was relatively low at all sites on the windward shore. Goreau (1973) describes similar results, in part, for the Jamaican north coast reefs where he shows that the rear zone is an area containing a varied Scleractinian community (i.e., 9 families: 17 genera: 29 species) and the reef flat Scleratinian fauna is impoverished (7 families: 10 genera: 22 species). However, in addition, the buttress zone is described as the richest habitat of the reef with 41 species of 23 genera of herma- typic Scleractinia and 2 species of Hydrocorallina. Certainly this is not the case in Anegada where the maximum number of species recorded within the buttress zone of Jack Bay was 15 (6 genera) as compared with the more diverse rear reef. There may be several reasons for this marked difference in observations: firstly, the buttress zone on Anegada is found in relatively shallow water, certainly not in excess of 10 m whereas similar zones in Jamaica extend down to depths of 20 m. Secondly, the north coast of Jamaica may not be exposed to the same physical forces ensuing from wave action as the windward reef of Anegada where there is a considerable amount of dead and broken coral within the mixed coral/algal ridge system. The effect of wave action 23 is quite obvious, particularly at East End, where branches of Acropora palmata are found detached from their original bases. As reported by other workers (D'Arcy 1975) a 'ground-sea' during the winter months increases wave action considerably on the exposed northern shores of the island. Thirdly, hurricane damage is quite possible, as noted elsewhere in the Caribbean (Stoddart 1963, 1974). The most recent hurricane to affect Anegada has been Hurricane Donna in 1960 which removed many houses from their foundations in the 'Settlement'. As a point of interest in this context, a severe tropical storm (later to become Hurricane Elouise) hit the island during the late summer of 1975 after the completion of the current survey. The resemblance of the Anegada rear reef to sites affected by hurricane damage on the British Honduras reefs is quite marked (Stoddart pers. comm.). Although it is not perfectly clear why the buttress zone of the windward reef is so depauperate in terms of coral species, it should be mentioned that in a survey of Grand Cayman reefs, Roberts (1974) describes the upper fore-reef terrace of the northern fringing reef (water depths 5-10 m) as a ‘barren plain’. He believes that the intense energy of the waves is dissipated in this zone and as a result only encrusting and low relief growth forms of coral are found here. In terms of dominant coral species the most evident must surely be Acropora palmata which dominates both the rear reef and reef-flat on the north-eastern reef and also constitutes the central core of many of the leeward patch reefs. Such findings are in line with the observations of Milliman (1973) who states that although Acropora palmata may be the dominant coral in Northern Caribbean reefs (Newell et al. 1951, 1959; Ginsburg 1956; Shinn 1963; Logan 1969) and Montastrea annularis the dominant on south-western Caribbean atolls (Milliman 1969a) this generalisation does not always hold. For instance Acropora palmata has also been shown to be the dominant coral at St. Croix (V.I.) even though the climate resembles the southern rather than the northern Caribbean (Milliman 1973). With regard to the considerable amounts of dead Acropora palmata noticed on the windward reef of Anegada, it has been previously suggested by Shinn (1963) that dead Acropora on the Florida reefs is the result of ‘over-crowding': Roos (1971) however does not attribute this cause to the death of colonies in Bonaire. In a more recent review of reefs around Bonaire Scatterday (1974) notes that prolific growth of this species is often limited to areas with heavy wave action (Goreau 1959; Storr 1964; Hoffmeister and Multer 1968). In Bonaire as in Anegada healthy Acropora palmata may be found on both the seaward margins of the reef as well as along channels (presumably there is sufficient wave action along channels to promote growth). According to Scatterday a similar situation exists in the windward reefs of the Caribbean Gulf of Mexico where parts of the colonies in more shore- ward locations are killed as a result of being forced into a position where wave action is diminished to an intolerable level, in the lee of other colonies that are established seaward. 24 Further detailed observations obviously need to be obtained around Anegada, particularly on the leeward side of the island and along Horse Shoe Reef where patch reefs abound. In terms of coral growth, patch reefs in the lee of the island showed not only a relatively good diversity of coral species but also a relatively greater abundance of these species when compared with the northern reef - particularly large colonies of the following species being found: Acropora cervicornis, Acropora prolifera, Diploria labyrinthiformes and Diploria strigosa. Scatterday (1974) also reports vigorous reef growth in leeward areas of Bonaire. This is in contrast to many other Caribbean reef systems where reefs located on the leeward sides of islands are poorly developed when compared to those found on the exposed windward coasts (Milliman 1973). REF ERENCES Adey, W.H. and Burke, R., 1976. Holocene bioherms (algal ridges and bank barrier reefs) of the Eastern Caribbean. Bull. geol. Soc. Am." 67, 95-109, Almy, Jr., C.C. and Carrién-Torres, C., 1963. Shallow-water stony corals’ of Puerto Rico. “Caribb. J. Sci. 3, (2 6&3), ilgeiegs D'Arcy, W.G., 1975. Anegada Island: Vegetation and Flora. Atoll Res. Bull. 188, 1-40. Duarte-Bello, P.P., 1960. Corales de los arrecifes Cubanos. Acuario Nacional, (Educacional) 2, 85 pp. Marianao, Cuba. Ginsburg, R.N., 1956. Environmental relationships of grain size and constituent particles in some south Florida carbonate sediments. Bull. Amer. Assoc. Petrol. Geol. 40, 2384-2427. Glynn, P.W., 1973a. Aspects of the ecology of coral reefs in the Western Atlantic region. In: 'Biology and Geology of Coral Reefs' 2, 271-324. Ed. by 0.A. Jones and R. Endean. New York: Academic Press. 480 pp. Glynn, P.W., 1973b. Ecology of a Caribbean Coral Reef. The Porites Reef-Flat Biotope: Part 1. Meteorology and Hydrography. Mar. Biol. 20, 297-310. Goreau, T.F., 1959a. The ecology of Jamaican coral reefs. I. Species composition and zonation. Ecology, 40, 67-90. Goreau, T.F., 1959b. Further studies on the buttress zone of Jamaican reef corals. Int. Oceanogr. Congr. United Nations, New York. Goreau, T.F., 1966a. Coral reef studies in Discovery Bay - Runaway Bay area on the north coast of Jamaica. Final Progress Report to Biology Branch, Office of Naval Research under Contract Nonr 4811 (00) NR, 104-845. 25 Goreau, T.F., and Wells, J.W., 1967. The shallow-water Scleractinia of Jamaica: revised list of species and their vertical distri- bution range. Bull. mar. Sci. 17, 442-453. Goreau, T.F., and Goreau, N.I., 1973. The ecology of Jamaican coral reefs. II. Geomorphology, zonation, and sedimentary phases. Bull. mar. Sci. 23, 399-464. Hoffmeister, J.E. and Multer, H.G., 1968. Geology and origin of the Florida Keys. Bull. geol. Soc. Am. 79, 1487 - 1502. Howard, J., 1970. Reconnaissance geology of Anegada Island. Carib. Res. Inst. Spec. Geol. Publ. 1, 1-19. Lewis, J.B., 1960. The coral reefs and coral communities of Barbados W.i. Can. J. Zool. 38, 1113-1145. Logan, B.W., 1969. Coral reefs and banks. Yucatan Shelf, Mexico (Yucatan Reef Unit) In: 'Carbonate Sediments and Reefs, Yucatan Shelf, Mexico' (By B.W. Logan, J.L. Harding, W.M. Ahr, J.D. Williams, and R.G. Snead). Mem. Am. Ass. Petrol. Geol. 11, 129-198. Martin-Kaye, P.H.A., 1959. Reports on the Geology of the Leeward and British Virgin Islands; Voice Publishing Co., Ltd., St. Lucia, W.1. Milliman, J.D., 1969a. Four southwestern Caribbean atolls: Courtown Cays, Albuquerque Cays, Roncador Bank and Serrana Bank. Atoll Res. Bull. 129, 1-41. Milliman, J.D., 1973. Caribbean Coral Reefs. In 'Biology and Geology of Coral Reefs' 1, 1-51. Ed. by 0.A. Jones and R. Endean. New York: Academic Press. 410 pp. Newell, N.D., 1959. The coral reefs. Nat. Hist. N.Y. 68, 226-235. Newell, N.D., Rigby, J.K., and Whiteman, A.J. and Bradley, J.S. 1951. Shoal-water geology and environments, eastern Andros Island, Bahamas. Bull. Am. Mus. nat. Hist. 97, 1-30. Newell, N.D., Imbrie, J., Purdy, E.G. and Thurber, D.L., 1959. Organism communities and bottom facies, Great Bahama Bank. Bull. Am. Mus. nat. Hist. 117, 177-228. Roberts, H.H., 1974. Variability of reefs with regard to changes in wave power around an island. Proceedings of the Second International Coral Reef Symposium 2, 497-512. Great Barrier Reef Committee, Brisbane. Roos, P.J., 1964. The distribution of reef corals in Curacao. Stud. Fauna Curacao, 20, 1-51. 26 Roos, P.J., 1967. Growth and occurrence of the reef coral Porites astreoides Lamarck in relation to submarine radiance distribution. Doct. Diss., Univ. Amsterdam: 72 pp. Roos, P.J., 1971. The shallow-water stony corals of the Netherlands Antilles. Stud. Fauna Curacao, 37, 1-108. Scatterday, J.W., 1974. Reefs and associated coral assemblages off Bonaire, Netherlands Antilles, and their bearing on Pleistocene and recent reef models. Proceedings of the Second International Goral Reef Symposium 2, 85-106. Great Barrier Reef Committee, Brisbane. Shinn, E., 1963. Spur and groove formation on the Florida Reef Tract. J. sedim., Petrol. 33, 291-303. Stearn, C.W., and Riding, R., 1973. Forms of the hydrozoan Millepora on a Recent coral reef. Lethaia, 6, 187-200. Stoddart, D.R., 1963. Effects of Hurricane Hattie on the British Honduras reefs and cays, October 30-31, 1961. Atoll Res. Bull. 95, 1-142; Stoddart, D.R., 1974. Post-hurricane changes on the British Honduras reefs, re-survey of 1972. Proceedings of the Second International Coral Reef Symposium 2, 473-483. Great Barrier Reef Committee, Brisbane. Storr, J.R., 1964. Ecology and oceanography of the coral reef tract, Abaco Island, Bahamas. Spec. Pap. geol. Soc. Am. 79, 98 pp. Walton Smith, F.G., 1971. Atlantic Reef Corals. University of Miami Press. 164 pp. JIPUS UeoTY Oeng ey} Jo UISIIQ | BIy sieoh UOT] [TUL 09 FIaUS ueoIy OjIENg jo uolyeUuIO; ut sj[nsel UISTUvoTNA SNOFOVLEUD | DIOZOSAIN quasaid moreq [eAg| Bag OOO9T OL OOO'FI Ov GI susd0e[eq yUSd.IEUIS auas0q OOO0T ik ssulooeq jleyus AUVILYEL 000 JIeys JO ButpooTy vag or ewso08i0| or ewso08i0| OIOZONAD 000 Sutstey AqyIqeig jor lor susoorng asueyo [eae] ag eee se. auss0[0H (L9G) a!zuaysey pue ajomyeay sayje) 4184S UedIy OVaNg ayy yo dew siydesBoaboajeg Zz fly Ajiiejo 10} palyrwo spuejsi JajjeWs 000'9 —000'8 Bets 000'8 —000'0l [ee] d@ 000'01 — 000'b! pabsewqns seare (>| auaao}siajq Gulsnp Guisamoy janay eas xew — autjyno ysabse] 0314 OjJang *soqoyd aySN 6S61 wory pesodwod :o9TesoW [etszoy °] aed S76 Saji ‘u ~ 7 { 0 ” 2h VaVDOUNV : (OL6L ‘PAeMOH_ 4a)4e) SUOIsIAIpgns a1ydesGolsdyg ¢ Gry e1odoiny @ euodig jissoy ¥ (3983) ayBiay yods ggg [a] anosbuey [nea] mee afpi 4901-pag (TU xajdwoo afpiy yoeaq pue aunp pasijiqeis ee! S}e|} 4901-pag = Plate 2 Potholes in Limestone: East End. mr 5 s f . sais Apmis — epebauy yp S14 Ai s}gasuel) Pe | sjaa1 yayeq [een] pug 3b] TT eased a wiog = 6— uesijag «= L 28 wo1jeys auouazol eaosGuey wed Guineas ¢€ 1w6ig Ss s & 2 SSE|PUINA auog heg 2 g ett, = SS yaaa MOD i NS & PS ae —_ a vee ‘2 COO CS oc gece Saas CO BEEN 9.907 ARNG aQGIOIOC Ong ace DAOC EE © @ @@eeee & eee ee e PSOUI@ARD PAaIZSeRUOW °@@O@-e7e e080 e@ @0C@0 0°@ © @e i & & @ 2 . a & e@ ee eee Cd sTwzoyry3uraAqerT erzrordra @e@ee@gee & e°@ a i ) e@ Ge e@ Bee @ ee e esofT13s erzrordra & @e ee @ @e ee ee @e@ee @ee0e 2. e @e e PsOATTO erzordta e e @ ee e & & e r ) unzbery ePraey avai TAwd e ® (eqeoiny *reA) saztrod sezrz0d ©@e@e@g@eesre @ °@ @ °@ e] oo e © a (BTIPAPTO *reA) seqTrod sarod ry) (ezeoTIeATp *reA) sazraod sezr10dg a 2 SaepToem3gse SsazTro0d avai LiIwod Pareprs earjseiaprs ee e SUPTPpPI PaIZSeIAPTS Val aeLSveadrs Oo (stTrbeay *reA) sajzrorzrebe erorazeby e@®ee8e°@ ee & GS @226 t © @ & e op ee ® (esseio *TeA) sazroriebe erorzeby AVGITOLYVOW ee) e @ e erazrjoad erodo1imy @0ee0@ ©O@e CB °@@*e *@e@ °@ © ++ eo en, Bee & e@ e .@ @ Ge & sTuqooTArIa9 erodoroy awdIwodowow WINILOWYSATOS WOZOHLNY H < < < < H H H Dp DP rH Ali Gs ee Ee a wo wwe Wye eke he! (es ey ee “ad * H H H < H H o 0 » 35 2 o c£ 2 Oo ® c 0 0 f 6c 2 O@ c 0 0 H H H o 9 Q o 2 Q +t © @ @Q rt o 9 Q ct Oo 2 Q ch De soleil H Pit inde eye m HK O (an dm Inf To) ym siaatat fe) rot Ans) st fo) ch En ie tO Oo Oo ° K ie) lad fe) K ° tal ° aN 5 R yw ON! 6S ® HN 3 ®o HB N 5 o WN 3 oO Ww NSN! 3 cee ose RG os Se FER FS 8 5 5 4 3 > 5 SaIodds AVE @ LOAUSNVEL T LOUSNWaL AWd Sdaqgu HOLWd ana wJLsSwa ATIOTAOT Awa wowle AMOGYM MOD GNa JSaM GaYOHS duyWwMdaT GaOHS GuvMaNIoM “SCGNWISI NIOSWIA HSILIYG ‘YQVDENW * GONVGNNaWY YIGHL GNW SAIOgadS TwuOO 4O LSIT | aTawh XI IIIA IIA TA Sdaau HOLWd qYOHS CawMaaT @e @ ® ee e e e eee ee ee DB Pp fH o o » 3 mhoR SO & anu 8 R (0) @) ) to 3 o aNa SWa @ wepunay @ wouwos @ 201e9S eee e @ ® ( ) C) C) @ C) @ e@ @ a ie ) e @ ® e©e@ee e @ ® @ @ e°e e ee © t ) @ e e e @ DW wo DDH oO wy eHe oD DW eH Oo o p c& 0 oO po cs 0 Oo fp ££ 0 OO fo o p @Q rt O@ 9 WQ Gre () fd (Ce) +t oO 9 WQ rh OR 8) fay Mh RR @ fay fh OR iB fay mh OR @ Nie © HoH N S mM F) Mm o BN 3S {o) fo) n (eo) fo) 10) {e) fo) 0) {e) fo) Lo} =) nh i=) nh 3% J n =] 0) (0) ) (0} AW ¢@ GLOUSNVWaL T DOSSNWaL AWd ATIOTEOT AWad Mowe MOHYM MOD HYyOHS CavyMaNnrImM e a1ey sseiz4qng doi Fe0"u Na : AA auozZ 7e8y uoobey LSaM eqeuetdues eriodeTrrw esorzienbs erfodeTTTw STUIOSTOTS eLlodeTrtw AVdIuOdaT TIN WNIYOdaTITIN WOZOUdCAH PaeuTII0D Parzseqny AWdITTITAHAONGNaG PRETHhTASeEZ ETfrusng AVGITIAHdOLaVO ePueyoreueT erpTrAydo,aoAwn PIOTFIATOU eTTTAYdosT eprbhrzi eezrzset TAydosrt esornbue essny erlaseT eruAtToos avdISssnw FISeyozs eTusoooyorq AVGILITINSOHOONL e2eTOserTe euTOTUPH suezeu etTfTAydodtTop snyjuezeue etTrTAydodtop STTeBTiestTAsig erpTpfhydodtop *43uoo AWdIIAWA SaIogds ———————=_ $334 HOLVd XI Bin Th IA A A nif I 0 cs ole mn o Siz ww” Oz “Buestes AVG @ LOISNVHYL LLOIAISNVUYL AV@ GNI LSVWI A11019801 AVG Naver AVS Nove WIJHYM MOI QN3I LSIM :VWIHV Gpvezt Gy & 2 G pee 2 t yf ie Sema | femme | amet ey Aaa | Gbpeéd bt :3n0z 0 eo) pe, o $3!34dS 40 ON oO N auozZ SSoetqqng ¢C GZ doy jooy ¥ ouozZ uocosey Z V jJooy rAeoy ¢ QUuOZ VIOYSUT | sAo0y -epesouy JO e10YyS paemMoody oy. (*q pue eLoys preMputM 2x47 (*Vy UO peptodea Sotoeds [etod Jo Aoquny ¢°3Td : Bec: en Plate 6 End of inshore transect: porites var. divaricata. re Plate 7 Sea-whips Plexaura spp- and Pseudopteragorgia anericana in Jack Bay. v Ne ae ¥ Plate 8 Millepora squarrosa on reef top at East End. Plate 9 Algal ridges on seaward side of reef at Jack Bay. fe Gi Bi 2 Si St vi El et Porolitheae. Specimen of algal ridge, Plate 10 palmata on reef top at Extensively bored Acropora Plate 11] Jack Bay. it 2 d3 ia WS te 7 18 Ne 2i0 ali 22 213 24 2i5 2 Plate 12 Specimen of bored Acropora palmata. Shee, 2 F SS > e Plate 14 Plate like growth of Porites astreoides. Plate 15 Bladed Millepora complanata and branched alcicornis on Patch Reef II. ies SU te ae ke et "heed + patchas of coral ie ; rf ‘ Ferauine algal ridges ea. " BUTTRESS > 30 tae aes poorly evelaped spur and grobve depths in feet 0 500 m Plate 16 Jack and Loblolly Bays; showing position of Transects 1, 2 and 3. deg gover : ojzord 1 yoosuwsL 9 3a aNnvs anvs w OcEe OOE aNvs sore Ope agIOHLOd # GauOe sse1}3ng (0) Ose 092 Ove Oce anvs anvs zy bL Uf Vi Z aga 1IOHLOd #7 GauOR AISAIGNALXA - TVUO9 ONIBONVUE AVaAd < yy eee 4 me Hrd bis . ei oe Ost ro. 09 ‘ Ovl Oel aNvs dNId wz UgLVMA MOT 4eq your : aijoad g qoosuvsy 2314 mg p / We aguOe ATIAVEH i y v “cs / } ay / tS Yay J) 1 Sg / : typhi LT A PLLA TOT HEL. ‘OS \ (ean? : / a51,15 BAY Vi ly / //] [| Ze LL) ; SSe.})nNg doy soy 0 seljow Ore O¢cEe OO€ oge 7 @Uu0Z Jeoy 09¢ Ope Oee 002 ES i oo eee = Sen $$ Cll = ees — 5 OF! Od ol & nN epls 18y3;@ seyozed [esr05 HOOUHOVaE keq Ajjorqoy : ettyord g yoosuvsy 8 31a eeZ[e oul[[sioo JZuyenioue + CHHOd do, joey sere Op YO}rd 08 [3 O9E y // } y / MLIAL! Lt Ly 1 hows L Lf /K Uy Mylyf LU LLL / OVE OdE OO€E Ose anvs W111 PIAA A hh hod, VU < ie ( y / YUM / / / nile np iNLI le ® : LZ PACS Ove Og? 002 ~ O08 ~ O S 7, @ Y ® S ep pte IMD, euoz 4e0y ergany r uoobey UaLVM MOT sand & algae, . Pt solidated rubble “a ue 2 ee eee, all aptonies Sh: REAR TONES SRG pari” dead, coral spurs living colonies 0 500m Plate 17 West Ends; showing position of transect 4. pug 789M : etypord p yoosuBIy aNnvs 9801.)ng Vv 6 31a O09” LL] ag Ayo, LAGOON sand + algae and coral Patches INSHORE fringing reef Plate 18 East End; showing position of Transects 5, 6, 7, 8 and 9. pug 38¥q ‘3Ujog UvoTIed : 8 pue LD so[yjord yoosuesy, Ol 3ta ‘ds vinexe[g * ‘de vascdelttAl TT pruosi0p se es ‘ds vrsojdiq c.. winjposursr4g Jo ByesBteyy THIN sl1B[nNuUe “Wl eh Ss saoztjoid -“ ¥ BSG sepjoeryse “J ay syUIOo;AIeD “VY = eye148qns peed 1/// sojysod ‘g meh a syeujud vy ay tc B: | ape pil PA OTE MMA IVADY VITITTLTI LIT TII LLL TLL. / Janes * W809 BVEE/////// ny Lid Yani OSS Se ie ee | . 10) (s03}10d ) (syaxo9;az00) @TaHNu algo a 1agonu TVuooO Tvuoo + GNVS wz A Let UaLYM MOT euaTe e13sstT + anvs bc ta: €: dat: | ugdAoo ur Ivuoo avad +aNvs Zz WL Wt WIN tt IvVD1V dsNaa aL aL L 8 LOASNVaL UaLVM MOT o serjemr Og OZ 09 os ov O€ er4 ol (0) TRANSECT 5 ———__] Thalassia =e ———_]Halimeda _—————— ———____] Penicillis —— | ———Coralline alga _ Wee] 7777 Porites ——__] 8. radians vie fe ie TRANSECT 9 = 10 20 30 7.¥ 40 50 OSs BON. oa punto bete “br ty Ug Pw Pe we Py Tf ® i th tlt | eT Im an [_ : = — Thalassia SS = aa | Halimeda ae = = | | Penicillis ewes = a : a Coralline alga ve —— Caan e er ee SSE P. porites SSS Syringodium NO A. palmata #8 P. astreoides —] KEY [Common | —— Abundant Rare [] sn (7A Dead coral substrate L = = J Fig. 11 Profiles of the inshore fringing reef at East End, Anegada, showing zonation and abundance of the dominant species SPERMATOPHYTA : Thalassia testudinum, Syringodium filiforme ALGAE : Penicillis capitatus, Halimeda incrassata SCLERACTINIA - Acropora palmata, Porites porites var divaricata, Porites astreoides, Siderastrea radians. Fig 12 PATCH REEF PROFILES The profiles are presented in an order corresponding to a geographical distribution running from West to East, see fig.4 Patch Reef I was only used for a fish rotenone station and is not therefore represented here. Transects were not possible on Patch Reef III due to poor visibility but a plan view is available. UA dgau HOLVd @ etjo1g Scrat CD Salad Lil hh eign 0z OL A ettsOtd V ANd XI daau HOLVd a eltserd Il AgaY HOLVd / Ved le Pep Ll fA ah = ne oz V %1sJ014 joar mol[eys yo Arupunog xoidde PUMAD AL MME Zi : as aS wit oe V eltyoid AI duau HOLVd qulog o;BUIOg ~~ ure'0 sro _Pug ™ ™ m™ m™ 7 am a UIntposuss 5 or UPOUNTCH we a eyorg are TOLL V eltyord A daau HOLVd usid Ill daau HOLVd @ eorg V OO IA daagu HOLVd da o1Njorg ~° | Vv 101d 55 NOTES ON THE ALGAE OF ANEGADA, BRITISH VIRGIN ISLANDS The marine algae of the Virgin Islands have been extensively studied by Borgesen (1913-20). The most useful recent reviews include the manuals of Taylor (1967) and Chapman (1961, 1963), and both these works have been used in the identification of algae species collected on Anegada. More recently Earle (1972) has listed 154 species of plants from Lameshur Bay, St. John, including 26 species that were new records for the Virgin Islands. The collection made is not intended as a comprehensive survey of the island's marine algae - only dominant species were collected along the transect lines and therefore the species list (Table 2) is certainly not exhaustive nor complete. A total of 46 species are recorded from 14 families. The data is presented in tabular form so that it can be readily used as both a total species list and for area reference. Algae were mounted and preserved for taxonomic reference as dried specimens. The collection is now held for reference by the Cambridge Anegada Expedition, c/o Dr. B.E. Brown, Dove Marine Laboratory, University of Newcastle upon Tyne, Cullercoats, Northumberland. ~—s ~ a a —— eohiktet axpart vartine AGATA "UO TALIA SEE : ALGAE and SPERMATOPHYTAE SPECIES LIST | oe ‘ a x woe SV5EF Tal ranTiV addy ‘Fo mug la ‘so.)tem " a . f Mee 3 orerl ele a , t's ; Dee iri 9g 708 «di Bi aii v val lo alhunee ong is ; hts - Ps SPECIES od at ag 10g One 3 P : ‘ Si 28.709 f ia g : 4 5 az & 5 hy ' te > CHLOROPHYCEAE i 30 Cladophorales Family : Cladophoraceae , : » a Chaetomorpha sp. = + - Cladophora crispula Vickers Siphonocladiales Family : Dasycladaceae 52 Acetabularia crenulata Lamouroux + Family : Valoniaceae Dictyosphaeria cavernosa (Forsskal) Bérgesen + + Siphonales Family : Caulerpaceae Caulerpa cupressoides var. mamillosa (Montagne) + Weber-van Bosse Caulerpa sertularioides var. brevipes (J. Agardh) + Svedelius Caulerpa racemosa var. Clavifera (Turner) Weber-van Bosse + Caulerpa racemosa var. occidentalis (J. Agardh) Bérgesen Family : Codiaceae Avrainvillea (longicaulis) (Kutzing) Murray and Boodle Udotea conglutinata (Ellis and Solander) Lamouroux Udotea flabellum (Ellis and Solander) Lamouroux Penicillis capitatus Lamarck Penicillis dumetosus (Lamouroux) Blainville Halimeda opuntia (Linnaeus) Lamouroux Halimeda tuna (Ellis and Solander) Lamouroux Halimeda incrassata (Ellis) Lamouroux Halimeda monile (Ellis and Solander) Lamouroux Codium (isthmocladium) Vickers PHAEOPHYCEAE Dictyotales Family : Dictyotaceae Dilophus guineensis (Kiitzing) J. Agardh Dictyota dichotoma (Hudson) Lamouroux Dictyota divaricata Lamouroux Dictyota indica Sonder in Kiitzing Dictyopteris justii Lamouroux Dictyopteris delicatula Lamouroux Stypopodium zonale (Lamouroux) Papenfuss Lobophora variegata (Lamouroux) Papenfuss Padina sanctae-crucis Bérgesen Sporochnales Family : Sporochnaceae Sporochnus pedunculatus (Hudson) C. Agardh ALGAE and SPERMATOPHYTAE SPECIES LIST SPECIES continued Fucales Family : Sargasseae Sargassum platycarpum Montagne Turbinaria tricostata Barton Turbinaria turbinata (Linnaeus) Kuntze RHODOPHYCEAE Cryptonemiales Family : Corallinaceae Sub Family : ilelobesieae Lithothamnion sp- Lithophyllum sp. Goniolithon sp. Porolithon sp. Amphiroa fragilissma (Linnaeus) Lamouroux Amphiroa rigida Lamouroux var. antillaria Bérgesen Gigartinales Family : Hypneaceae Hypnea sp- Family : Gracilariaceae Gracilaria sp- Rhodymeniales Family : Champiaceae Coelothrix irregularis (Harvey) Bérgesen Cermiales Family : Ceramiaceae Ceramium nitens (C. Agardh) J. Agardh Ceramium sp. Family : Rhodomelaceae Polysiphonia sp. Bryothamnion triquetrum (Gmelin) Howe Acanthophora spicifera (Vahl) Bérgesen Laurencia poitei (Lamouroux) Howe SPERMATOPHYTA Halophila baillonis Ascherson Syringodium filiforme Kutzing Thalassia testudinum Konig WEST END JACK & LOBLOLLY BAY EAST END PATCH REEF III PATCH REEF IX 58 Acknowledgements We are indebted to Nancy Ogden of the Fairleigh Dickinson West Indies Laboratory at St. Croix, for checking and correcting our identifications of the algae. REFERENCES Borgesen, F., 1913-1920. The marine algae of the Danish West Indies I Chlorophyceae. Dansk. Bot: Arkiv, 1: 1-158 +25eoiiee Id., Il Phaeophyceae, ibid., 2: 1-66-42, i914. ids, III Rhodophyceae, a, ibid., 3: 145-240, 1917. Ids, d,ibidagmer 241-304, 1918. Id., e, ibid., 3: 305-368, 1919); id.) fa jeaibaen S79" 369-504, «920% Chapman, V.J., 1961. The Marine Algae of Jamaica Part I. Myxophyceae and Chlorophyceae. Bull. Inst. Jamaica Sci. Ser. 12..No..1..5 159.ppr. Chapman, V.J., 1963. The Marine Algae of Jamaica Part 2. Phaeophyceae and Rhodophyceae. Bull. Inst. Jamaica Sci. Ser. 12 No. 2., 201 pp. Earle, S.A., 1972. The influence of herbivores on the marine plants of Great Lameshur Bay, with an annotated list of plants. In "Results of the Tektite Program: Ecology of Coral Reef Fishes' Ed. by B.B. Collette and S.A. Earle. Natural History Museum Los Angeles County Science Bulletin 14, 17-44. Taylor, W.R., 1960. Marine algae of the eastern tropical and subtropical coasts of the Americas. University of Michigan Studies, Scientific Series Volume XXl. University of Michigan Press, Ann. Arbor. 870 pp. 59 CORAL REEF FISH OF ANEGADA, BRITISH VIRGIN ISLANDS The first mention of the abundant fish life around Anegada is that of Schomburgk (1832), but it was not until 1973 that any attempt was made to assess the fisheries and mariculture potential of the island (Iversen et al). During the latter survey a brief dive was carried out on patch reefs off East Point, Anegada and 30 species of reef fish were recorded. Approximately 185 species (55 families) were recorded in the present survey at sites on both the windward and leeward sides of the island. METHODS Collection of information involved firstly, fishwatching to obtain estimates of easily visible fish on the reef; rotenone collections in selected areas to capture cryptic species and thirdly, photography of fish species so collected as a record and aid to later identification. Fishwatching was carried out at selected sites. These included Jack Bay, Loblolly Bay, West End and East End on the northern shore and nine patch reefs on the southern shore. Brief surveys of Cow Wreck Bay and Bone Bay are also included in the summary of results. At each site particular habitats were chosen that were representative of features in a zone, e.g., in the lagoon zone at East End such habitats included a sand substrate, a sand and algae region and an inshore fringing reef. Within these areas three pairs of divers would fishwatch for 60 minutes, noting not only the presence of fish species but also their abundance. The following scoring scheme was adopted for the survey: Number of individuals Species allocated 1 to one of 6 groups 2-5 according to numbers 6-10 seen. 11-30 31-100 100+ During the 60 minute fishwatch, the first 20 minutes were spent in a general swim of the area; the second and third 20 minute periods were spent examining a relatively small area (approximately 10 m in extent) paying particular attention to cryptic and retiring fish species. Thus a general overall impression of relative abundance in the area was obtained. Results obtained in the field from each pair of divers appeared to show surprisingly good agreement for each site visited. Note was also made during fishwatches of the presence/absence of juvenile fish and their abundance on the reef. 60 Several night dives were carried out both on the northern shore and on patch reefs on the leeward shore, in an attempt to describe any changes that might occur in the fish population on the reef after dark. By assessing abundances in this way, factors such as time of day, state of tide, meterological conditions, etc., are likely to play an important role in determining numbers of fish on the reef at any one time. In this preliminary survey it was impossible to standardise all these factors and this limitation must be borne in mind in interpretation of the final results. "Rotenone' or fish poison stations were carried out at a limited number of sites according to methods described by Randall (1963). Rotenone is an alkaloid with an empirical formula of C..H..0, and its effect upon fishes is to cause vasoconstriction of the capillaries of the gills (Hamilton 1941) and hence respiratory impairment. Powdered rotenone, however, has relatively little effect upon invertebrates except the groups Cephalopoda and Turbellaria. In the present study rotenone was used in the form of derris powder (3-6% rotenone). The powder was mixed with water immediately before use, in the following proportions, for dispersal in approximately 10° m~- seawater: 1.2 Kg derris powder/2.5 litres water A pair of divers first selected the site of the poison station and then estimated the strength and direction of currents using fluoroscein dye. Appropriate quantities of the poison were released by the divers, and on its dispersal two additional pairs of divers assisted in the collection of fish by netting both on the surface and on the bottom. The specimens were then returned to the boat, moored nearby, where the collection was sorted and fish species placed in appropriate tanks of seawater before immediate return to the field laboratory on completion of the station. Time spent on each site varied between 3-4 hours. At patch reef I, the first poison station attempted’, subsequent visits indicated that fish returned to the patch within 24 hours of application and dispersal of the poison. On return to the laboratory the specimens were prepared for photography, according to methods used by Randall (1961). Each specimen was pinned out in a wax-bottom dissecting dish and the fins fixed in an erect position before applying formalin with a fine paint brush to the fins and other parts of the body such as the gill covers @Ec. A few minutes later the specimen was removed and covered with a small amount of water in a perspex container - a photographic record was then made. All specimens collected at rotenone or poison stations were weighed (net wt.), measured (standard length) and preserved in 10% formalin. The fish collection is now housed at the British Museum (Natural History), Cromwell Road, South Kensington, London SW7. 61 Identification of specimens was made using Bohlke and Chaplin (1968); Randall (1968) and Chaplin and Scott (1972). OBSERVATIONS All results are contained in the species list (Table 3) for the island, where abundances are recorded for each habitat visited. Information is based on fishwatches during the day. The key at the end should provide a guide to information on the abundance of juvenile species; specimens collected in rotenone stations, etc. 20 pueaqepTTH eueorrowe STZPASE <. # ©, Joad bie visendd Biren t Bot “xaeys uoweT (Ae0d) spz9sorraerq uoTideben yaeys drayoeta (seuuetouaTen) sngequrr snuryreyore) — ‘) (aepeocryos pue motebra) raeburads snuryreyore) AVGINIHYWHOUNVO q34* wnjerzro ewo,sowATbury aVdIdOTOLOIYO " pues pues pues syita sitidot Snteu' sbao see a Ge eqeuted geoy burbutzg ejeuTed ebpta Tebtw eqeuted « sayojzed Tez0D seusjzed Tez0D sseipg + pues adr 3e235qNS «< oPblyw + pues 1?) < ) bes SYOHS cuvMagT o 4 Iqen } SGNWISI NISYIA HSILIY8 ‘WOWOENW SAIOgdS HSId dO LSIT lagtuny 4 Aezou AwbAg emezeuey ePTysoA setTyozreuy Aezour [tTejuepTo (dney) sTzerrrmm eueerny Xezou peqqods (aeTanD) ebuTzou xezoyzouwAy Aezow uwser5 Tuezuey stziqeunz xezoyzouwAD exzow 3nuyzsseyo ‘ds ezoohTeyoug aHVGINSvanW © © © 2G. © Tee jFeoy (ueuIg73S) seprozozidohy shyzyotdney @ Tee ssezbeas Q) uex4INT Tuosuens snuTYyIOTTYyD AVAL YSNODONAX Te2 T3Ieybeds (ueuT TO; @) pue uepzor) Tspzempe enburzroy AVGINSNIYOW ystyexeus @) (z89S204) sdohu snTeydesoutyoezy ZeATp pues O90 ® © O (2) ee e @ e (ztsseby) snrpewzezur snpoufs USTJPAeZTT ezAOYSUT © e (SneeuUuTT) suaze0z snpouAs USTJPAeZTI peu @ (sneeuuTT) snpoufAs snpoufs “HVaI.LNOGONAS fo ow c= w iS) Fe Le} yy n n se] n > he) n Q Q n n n ny K o iy o 9» a) fe) (o) p 9 ¢ [I oe 2) =) ES e) Q i= 3 K K 3 t=) lor E) =] ao 3 a o 3 o o 0 a a 7) os f =e e oe g § 5 s » 8 9 re 2 Salogds Q D> Be o o Q ct no g 8, a 8 i (0) oy) ) a a a 4 o © © © a) Ke hh i} 10} wo} oO Sdaqau HOLYd qa a v r) a W tai ro) ist W W qd laf ro) Vv GaOHS GuvMaaT aNd LSVa A'iTOTIOT AVad wove aNad JSaM ystyteaaztnbs jyooy oe em @) @ (Av0d) snosnzoo xfAzorpy ystyTeazatnbs meCbuoy ({2eqS0) . @ @ ¢ e@e . d eee sjTuoTsueose snz2UaD0TOH WaT ULNA SO'TOH @prsraaTTs peeypazey (Teyosozz, pue e e Z®TINN) Sedr3s snzowourzreyzy SVGINIYSHLY xeOqFTPH (Tuezuey) e e snjerosezyrun snydueyrodiy ooyATTea (sneeuut) e e sTsuetTrseaq snyduezrmey avd IHdWvulWaH ystsJeTpeceu OTAUPTIW (umeqtTem) nonuTz earnTAbuorsS UsTFPUNOoH (anensetT eo pue uozed) snfrposor2 sninsoyfAy YySTFJeTpesu uTFpey (Avsod) e3e}20U ernTAbu0o7r3S SVGINO Isa uor4yeys yee wI0m peTyoeds QC) uexanI snzejzound stydozhw Tee wz0m fay CQ) (uepazor) sTjuowbe erryy Tae pez zodspTop (dney) snzerno0 sAyjyorzAw AVGIHLHOIHdO euouUs{OI GBAOIZUVM yu PUNO} oOsTy ezeUTed pues eqeUTed pues ejeuTed pues pues Salodds Jeey butbutzg eebtw + pues ebptu Tebtw seyoqzed TerT0OD ssei9 + pues adAy e3erz4SqnS & sayoqed [e109 a Q Oo « ro) a Af a 2) a W a io) < Sdaga HOLvd GYOHS dayvMaa'T ana Jswa ATIOTHOT Awa wowle aNO@ MOD aNa LSaM Jeuuss uzrSBYyANOS e Keod ePTTrpnord eueezhyds epnoezzeq 7eeID e e e e ee @e e ee e e e d e e e (umeqTeM) Bpnoezrzeq eueezrhyds SVGINAWaAHdS ystgedtd utz3z0ys Xeo0q suaonte snyzeubuls aeejTezqeW Treyounp snyzeubuhs 1 ustyedtd ssoubng yustygedtd paqse15 (A804) UOl4ZS4S BUOTA4OI BAOISUGM 4B pus 4yuIog Suryy0eg FJO pszyO9T[OD snTeydeshyseizq shyzyotoyzAz0D SAVGIHLYWNONAS ystzzedumazy, SouusTouSTeA e e@e@ e@ eee e@ e ee e d e e @ sn3eTnoew snuozsoTay SVGIWOLSOTNW ysno Aysng (eMezeuey pue G) SpooMm) T3pruyos uorprydoreg avdIIqiIHdo @) “ds erqrr6o awd 1n.Lowd USTJASTPTOS AeqyoeTEA zetang snqooel st3zstTidtzAw @ e e e e e e e e e e @ e e a e ustsTezztnbs (umeqTeM) snjnz SnZIZUSIOTOH YSTIASTPTOS TeuTpzeD (ROusYyoTND) sTurdsoz,zez1 sdodfAz329eTq ustyTezatnbs Aysnqg ° © @ @ @ ee 6) e : © e. .@ ° (A90d) snTzeTTTxXea xAzorpy o oO J © @ & YW ® F vy oS Hn NH oo > oy fo) an 4 a eee ee ae ee ee ie ie] iS fey fa 5 8 9 #8 2 9 & @& a 2 a Ep e » ct ct Be + + a oP ai fies cy a oD C) oY) 9 Qa »p i ry » © c ‘4 a ct ct oH 0 a 8 3 5 Bg Deh a mie i SalogdS hh n n 0 Sdaqaud HOLWd a a wv ro) a W ta r) a W W a a r) W GYOHS GuWMdaaT GNa LsSwa ATIOTHOT Awa Mowe aNOa Moo ana Sam Aqskez (epedgoet) wnjejuenro ucdojawor3ag — Keauop (sneeuuTT) PATAZ sTroydoreyde avaINvaNas hens ; ysTZyouoy | (adop) snarTrays uoboder3sy ‘ ystsTeuppaeo Aysng_ ee@-: > ” > © @) (Avog) snrrejuewbrd uobody e YSTFTeUTpzAeS paTyxeId be : - ‘) (t938eATTS) TurfTyuoo uobody YsT3TeuTpzeo pozzeg : @) @ (Aa0d) snje30urq uobody — : ystjoweTa @) ® (@) (Av0d) sn3eTnoew uobody aVaINosodyv azepunoTy pedAg 0) (ztsseby) snjerTres0 snyz0g zepunoTy yooorsad ° @°* ee . . . . (sneruuTy) sngeunT snyjog ‘ avdIniod JeTINW TPeQUueZ = Aeod uopoyorz} Trb6nw Poles aval TrIonw SG) 1 Se NS 9 UH oH » 9 Q a g a » ry » o ° ° i oe oe PEE a - e e or sp 5 + MY » 8 9 ig ry a p ios » 2 9 aa a t a 0 . 3 3 E B 8 ® 0 D so FI Saloads » v wo t ~ Sdaqa HOLWd a a W re) a Ww a ro) a W W a d ro) W AWE OTEK AYOHS CuyMagT ana wsva ATIOTHOT AWG = wowre aNOa Moo ana Sam USTFSTTI pues @ e e e (yooOTad) TzeTunTd snyzueseTen AVdIOaLSOTHONWad Zedeams Asset Teosor, pue ZeTTaAwW rySznquioyos strzreyduag aval wHdAWad okebtg e e ZeTAND snzeueTe snyzUeDeT Ig zeddeus sAasseT) (apadgoe7) e e e ® snze7zuenzIoD snyzUPoeTIg SWCLHLNWOVIdd qeTsseq Aztegq C) e e e e @ @) Kaog o7ezoT eumezry SYGIWNWeD qeTsseq efSsmogq (emMezeuey) Gd) sTsuepnuizeq eumezbopnesd SVaILSTIWWWad qeTsseq pzenBsstms e (Ae0d) ezqnz umtzsTzsTz0y) zednozb yorta e e e (A80dg) Toeuog eozadozazoAW puty pew (sneeuuTT) sn2e23n5 snfeydeutdg © © pury 3xoou (392qS0) sTuoTsueospe snTeydeutdg eqeulted pues eqzeuTed pues eqeUulTed pues pues SaTOgdS Jooy butbutz4 oeblw + pues ebpta TeStw seyoqed [ezr0D seyo jeg Te1z0D ssez5 + pues ed&y, e3e74SGQnS iH 13) (=) « =} 13) (2) « q a a 1) a Sdaqad HOLWd tat a HYOHS GYvMaaT GNa JLSWa ATIOTOT Ava wowle aNd GSM sdgagd HOLWd GYOHS CYVvMagT pqeuTed jJeoy Hbutbutz4 n na a) 7) o o o a =] =} re 3 Q a 8 a o + ct o > = Qa o o Vv 2) a WV LSwa ATIOTAOT ebptu Tebtw oO AWE eqeuted xowle . . . d e eo e d . e © . . . . * . 6 d e e . e n Q Qo 4 o re ° ra oy 5 H nm 3 3 roy w o roy o = ron + S S oO ct ct K 0 i) o > ez n i) o n 7] nH v Vv a aq re) Vv AOFTaM “A 2 aNa LSaM adAy, a3exzysqns E wi zetand (sneeuu (yoo (sneeu (ame pue (aeTa e uepz —qutes qunab youezg (3se2euUSaq) INZeEeUTTOAPTJ uoTNueeH azequo, wn3jeeuTToine uofTnweey awd IASvdvwod zeddeus auey TI) srabeufhs snuel3n7 azeddeus [Te\MOTTEA Ta) snznsfzyo sninhoo aaddeus Aer uT I) snestz6 snuel(3jnT az04sewWTOOYDS QTeM) snpode snuel3n7 aeddeus bog (aepteuyos yoota) nool snuel3yn7T azeddeus Aueboyew np) ruoboyeu snueljn7 SWGINWCLN'T ejoWOTed uuewlreAg of Tapoob snjzouTyoriy yoerl eAa-asz0H ztsseby snjer xuezred aqauuna ent o2ATeTTH Aoazzya0h snsnz xuerep yort aeg (yYOoTd) zeqnz xuezeD AVGIONWaWD Saloads uotye4s auOuTI4,0OL DAOIZULU 4% poayxdaT[[ON ite} @ “I 0) uw C= w Ls) re Le} ny n n 9 HK pgp BoB 5 5 I) fen fa 2 «@ Be i ee Q >» he mw @Q o © oO rh Sdaqa HOLWd aq 4 v HYOHS CeavMaa'T QNa LSva ejeuTed pues abptu Tebtw eReUTed pues seyoqed Ter0D sayozed Ter0D sseizj + pues pues adAy, e7ez4SsqnS cH 1) a ezzefCom utszzods paeazt) pue piteg snezuebzre snuiozsouTong ezzefom zepueTs zeuyund5 Trseuof snuiozsoutong Auuel zeatts (pzeures pue Aond) ern6 snuiozsoutong ezzefom uTIMOTTOA (umeqTeM) snezeuTD sezza5 avdI awa Abzod peayuytor (z@pTeuyos pue yooTd) opeuofeq snueTeD awdrlawds qunz6 ystueds Zey_UNy umuioZsorToeu uoTNUAaeH eqebreu yoeTE (ys0Tdq) STSueuweutins snuiezz0sTuYy USTs 370d (SNeeUuUTT) SNoTUTSITA snuezzOsTuy qunzb pedtzj4senta@ (Meus) snznTos uoTnuaey qunib zeseep Keog untzeuoqzeo. uopTnuaey gunzb 23TUM (epedgoeq) TzerTwnTd uoTnueey qunz6 yjnomy tems Zeyquny unezAbreshzyo uoTnuaeey oqebren Zetanyd unqTe uotnuseey SsIOadS YSTITESMep TTEIMOTTSA © 6@ € 6 6 “e@e5see# e+ @ & & & d e eee (zeTanD) snanshazyo uopoyzedsoz tH SWdIwLINaOWWOd YysSTZATJAOAIANq pepueg eee @ e eee . @ e 7. @ @ e SNSeUUTT SNzPTIZS UOpOZaPYD ystgzATyazeqgnq eAerznog eee e e Qe . ¢@ e er e (sneeuuTT) sn3ez3stTdeo uopozeryD ystytTebue Aer e e (SNeeUuUuTYT) snjzenore snyjQuPoPUOg ystgztTebue youezg e e e . e e 2 (4Y2OTd) nzed snyjuPoPUl0g ystyTebue uaend ee ee C) e (sneeuUuTT) STzeTTTo snyzueoeTOH Aqneaq yooy e @- ee . . d e . ee (yooOTd) OTOooTIZQ snyjZueoeTOH aVaT LNOGOLaVHD ystyepeds ofquetaw e e (Jeuossnorzg) Taqez snzaqdrpojzaryD avalddIHda qnyo epnuzeg eee e ee e.° ee ° d & @e (sneeuUuT]) xrz3eR20as snsoydAy HVdTSOHdAN ysty 3eob pajjods e e e e e .- @ e e . e a6 8 (yYooTd) sngeTnoew snauednpnesg YysSTy 3eOb MOTTSA eoeee® e & e eee e . e d e e ‘O) . (aeTAND) snorurgzew sAyZYOTpTOoTToW avdt TIOW wunap pez jods (zepteuyos d pue yooTd) snzejzound snjenbgZ AVGINaAYTIOS a wn oa w N br se] "y n n se] na > se] n Q Q n n 2) o K o o ry rf b o ro ° ° ry ou c i laa =] =] = ro] a = =] tal tal =] 2 on i=] 5 a Q =] a oY) =] a o o Qa o7) 77) a Q o _ o e =) ct ct B + ct ct + K » 3 i) i) i) ‘vv a] o = 2 a Fs o R ° SaIOddsS yp Q aQ Te) fe) o a oO o 0 4 > a 8 o o ® i) un < mh 7) 7) | Sdaqu HOLWd a a Vv 9) a W a re) a WV Y qa dq e) W AWd AOFM GYOHS dCuywMaa'T ana JSwa ATIOTAOT Awd wowe ana LS 9SSSPIM UMOTD (TeyosozL pue @ e e e C) ee e @ ® @ C) ® @ e d e @ @® e Z2TINW) euuTdrpnoeu sazrsoyotrTeu peeyenta @ ele) unjZeTOISeFT PUOSSPT®2l ©0000 0@0O@eGe2> OO @e@e- 4 @ @@e (OPT Eo ie oa yustyboy utzjods e (A80d) snTTeyotnd snuerpog usty50y ystueds e @ e @ @ @ @ e @ t) d @ e @ e (SneeuUuTT) sngznz snuetpog awdlaavt usTs4Mey peqjodspey G) (Aezqmow) sourd snzryzzroAtTquy aVdILIHeerS sTwozyo oentTa@ @ ° @ C) d e r) (A80dg) veueho stuozyD sTwozyo umozg e e C) d @ (JousyoTNd) e,eeUuTTTATNU SsTWOIYD zofem quesbieas @eeee5sexeee e©e@ @ ce @ @e e d @ °° @ e (SneeuUuTT) STTT#exes Jnpzepnqu Azobezbnesg (Teyssozz, pue eT THW) @ e e e .}@ e e e@ @ e @ e @ e@©e @ C ) d x. @ ® @ snj}oT}sooneT snzzuedPuU0dng ustyTesuep Asuoy ° es ek eee G) ® 0 eo es eo @ e ‘ds snzzueoeucdng ystgyTesuep eoD05 (neutTeqsep) eoeee@eee# @ @) e.° ee ee @ @ d ° @ STTTQ®TIPA snzzUeDePUOdNg ustjyTesuep jodseezyr, ©*eeee:-e e @ ® ee e d e (Zetand) suozf#rTuetd snzzueoeucdng ystyTesuep Aysnq © @ee @ ee @) e @ ® e @ e @®e @ d e@ [S) (420d) sueoturdoszop snzzueDeuodng USTFTesmep TOTOOTY Ct ) e e 8 e e @ e@ e d e e@ (As0d) sn3zTQz2ed snz}zuad.eu0dng © 8 SI @ G S Ww fH fs) tal th) uy oN > Uy NH Q Q MW a CS ee 9 9 e 9 fo} © i & Sy bees hae l=! Boy @ |S R im fs} AS ely fay 3 A m js fay i) i) fe Fah a 9 o Be 9 B B ct ae ee ct ct + K es ae ; a <8 s do 5 mn a & 6 a oD e 0 © 2 ah en a Sa1OadS rh a) 7) J © Sdagu HOLWd a ad W 2) ad W a 2) ad W WV a da ro) W GYOHS duavMaaT GqNa JSva ATIOTAOT Ava wWowl aNa LSaM 90 -@-: eo -; e 8 e e e 6° fon) w cS w nN Lad sa] y o K be oa 8 3 oy) a cr - o 3 a a o 1) hh Sdgqu HOLVd dq a SHYOHS duyvMag'T ONa e ne ebtw pues pues pues Ter0D pues Tez0 ejeulTed T eqeuTed eebtw + pues abpty seyoqed seyojed < < ATTOTHOT ang LSamM ystyqozred [Tejpey (aapteuyos pue yoota) wnrzeydoshayo ewosTieds ystyjorzed TTeIMOT TOA (sauuetoueTea) @uurdraqnz ewostreds ystgqoazed 3yhtuptw So@uUueTOUeTRA SnUT_SeTEe0D snzP]lS ystyj0aaed moquTey A@TAND ereweoend snzreos ystyj0aaed anta (Y9OTa) SnaTnze0co snzP9sS ystyqoOzred sseoutad jsereuseq snzegdoruerz, snzP9s ystyjorzed uaaend aepreuyos pue YSOTa PTNQeA snzPoOS aVaIYvoOS ystjazozez uaaar5 (neuTejseo) suepueTds snjouoreqd Twey ystjazozea AyTaved (sneeuutT) eTnoeaou snjo0u0rTaRd WAH eazy tmMbutppnd aRPUUTT) SNZePTPePT SATAOYOTTeH @SSCAIM PROYUMOTTAA (SeuueToUueTeA) TIOUTeS SaTeOYOTTeH esseam pe JUTed (Av0d) snjotd sazraoyorT eH yotp Aaeddrts 9TH) SNQPRITALQ SaraoYyorTyTeH eSseamM rAeeYoRTA 2eputTeagsS) rheod saraoyoTTeH Sadlogds Auuetq pepueg () (ZeuyoepuTeysS) snzZerosez snuTjoereg Auuetq utyzizeg (Ae0d) ZOTODTSTaA snusezD0DePTeW Auuatq Aztex (pzeutey e @) pue Aond) stuurdtyonu snuiostzqet Auustq efe-aTbboy (seuusToueTeA) OTqob snuostaqeT AuuseTq yeeyosgnd (As0d) snzezroonq snuostiqet Auusetq yoo ueA-say, pue eqeeq STSUeTFTeY snuiostiqe-T SWGINITO © Oo Zezebieqjs aTppes ©) KaTbuoy snzouTo0zqnzZ sn{TTeTIrTy zezebreqjs akebtg syze}S snzossozra sndoosoTAzoeWd Zezebiejs pues C) @ e- @2) TI? sn3ze3TbTprazz sndoosoTAzoeG Zezebzejs dt Tyjoous S@xTUQA UODe snTTYyooZNST AVAIdOosoTALOWG ystyqozzed 3ybtTdojs @e@ee @ ® e ° d e@ @ e (ezzZeVZeUUOg) apTzrA eulosrzeds ystzjz01zed pedtzqs 20 utyeTI30N yooTd sTsueoToZTD snzeos @ @ ® © ° ) ystyjozzed pueqpey (SeuusToUSTeA) e® @ e e e @ e e e d e@ @ e@ unzeuaeTzorne euiostzreds ystyqorzed yQ00R yong (SeuueTOUSTeA) sUueTper eulosTrTeds >) (i) > Ww N (= is) Ry n n ue} n QP ue) WY Q Q n n 1) 9 K v » 9 » e o fo) fe} » 9 ‘3 ee PF 8 3 =) = 43) q Es K int =) 3 i=} =) a Qu i=] Qu iy 5B Qu o i) Qu Qu 7) » Q o = o (= = ct + fe + a + ste N i) 3 i) a i) ‘J ‘J o Q ta eo o o 9) ct t=! Q ct fag in] o BS a 2 ff 4 SaIOgdS co} 0 o o i) Ke mh 10) a he} (0) Sdaga HOLWd qd a W ro) a W a 2) ad W W a fag 2) W HYOHS CGuyvMaaT GNa LSva ATIOIAOT Ava Wowle GNa 2SamM Sdagau HOLWd GYOHS duWwMag'T ejeuted jeoy butbutzg e e e e e n n "0 n ee 8 &§ a oe 5B a a cr o od z a » o v re) q W isva ATIOTAOT ebptu Tebtw AWa eQemTed pues seyoqed [e109 Pa @ ©© ® e © © 9 7) R 5 wo oT) i) + 0 o 19) ad tal ie) o z a o 7) i) v GNa LSaM pues adAL e3ez4sqns Aqo6 Aqsny (zee T2z29n) r3rrodry snrrrnbsrnd Aqob utyeTx0e14 (SanqsutS) unzeuuwed5 ewosorqoo Aqob 5utueato suTqoy pue ayTUQga eruedS euosorqoy Aqob perpt3za TTT uwnueezyooneTh6 snzezdoydAzrop awaIrHoS 29TTIW ATTON sneeuutTy sngeqsTro snTuueT_ Auuetq drtpow (SauueToueTRA) snorjueTze snruueTqorydq SWaTINNa'Ta utgeTdtazq3 abpexoeta q9PTQUeSOY snz0z72e sazoeuPeUUg Auustq eAepey (teTMOg) STTeIOZ0ed sazOeuPauUg SVGIISANA LAT AuuetTqey7td zeozyzenta Aeaod e2ePTTa00 srsdouery) Auuetq Azejzar90g exXTUQA eTreW PTIewAaTqueyjueoy Auuetq Aysna (zeuyoeputays) TTTTh snue,o00RTeN AuusTq Tez0D (ysaey pue uueulzaAg) sn3zeTnhuro snurrorred AuuetTq Asoy (A90qg) sndozoru snue,209PT PW AuusTq OTWTW (ueuz0n) rAddnb snwostrzqe7 SaIoadds 1014848 euoTe,OI SAOIZUCU 4B P2zDeT[OQ 1014848 euoU9}{01 SAOIZSUBU 4B pee T[O| © OO YD HW B WwW SY oy mM HN Nn gy 2 es. » PoP nS = 3 ior) (ory B or » @Q o a og a a Q Pp 6 a» Q vy p o © Fh Sdaay HOLWd ta a W fo) a W YOHS CeawMaaT GNa JSwWa ATIOTAOT a > Le) n Qe - § Dp 8 B BP ct Cn) ee a ire} 0) 2) i= WV AWd Mowe seyoqzed [Tez0D rt Q n R 8 » a = + S © ct! Q » Zs 0) a a dq ro) Ww aNd LSaM pues adAy, a3e23SqnS UuSTJAS9HHTIA uesd9 (TTFUSI TH) uaueTgZjns stuzepryzue) UuSTJAeH6T7Q uesand SNSeCUUTT PTNZAA SeASTTeg AVGLLST Iva uoebans uesc0 neujeysep snuertyeq snznyzueoy ysTJ10390q (yooTa) snéznzTyo snznyzueoy Hue} ent Zepteuyos pue yooTd snayTnzeco snzny2ueoy avdTYNHLINVOWw ystyuotdzo0os woozysnw ZeTaAnD sTwreur eusedz00S aVadINavdydOos Agqob zeazeq (94eqTTS pue uepiof) euliosoaTog snTTeuorqoy &qob esouxzeus sutqou pue eyTyuod seuATaae eulosoTgoy &qob uojtop sutqoy pue eayxTUod snTOTp snzezdoyohzo09D &qob jodspto5 uepz0e tuosduoyz sTdeToyzeuD Aqob6 utyaeg sutTqoy pue eyxTUOM SeprorTrTe snzezdoydAzoD Aqob uts Tita (SeuusToueTeA) Zozezodos sntqobhyzeg SaTOads sseizjng ‘Gd doy jaay 5) pea2eTT0> suewyIeds yo zequinu euoZ 1e9y "¢ yar® uot jIeIs suouaj oT uoco8e 7 oy. @peWw plode1 aouepunge ou 3anq pajou sayzoeds SYaLLAT ANOZ ° ee e eee @ e e . e e e ° e e e e e. e e e CNC copa (GQ SINS) iret Li | Hn Nn ae] n o K sy o w o Pe i=) 3 - 35 B =] a a 5 a o Q o ct oa + ct o 2 a a > is yo) a o o o o mh Sdaqgu HOLWd dq a v ro) a W @YOHS duvMaaT ana JSva ATIOTHOT @ d abptu 1ebty @ oO AWE oo _— OO ' ' or a] T STENPTATpuy Jo raquny e Qa e 0 Nn (2) » ° re 5 n 5 Bo o o - cr o a) » ing o) ® a a W v a AWE AOaAdM wowle Q 177) ie) o K 2 ry a e + ve] o 9 ct x ie} » a un o i) a re) v aNa LSaM *Aay ystsbutTo pew (Av0dqd) snsourb6rqnz sooiy avaroosaigod ystyeutdnosz0d snoeeuutTy xrzqshy uopord aval LNodord aesgnd asoudaeys (yooTd) e3erQSOr TaQZsPbTYyQueD aezgnd TyeQpueg (yoota) rrerbueds saprozraoyds AVGILNOdOVeLaAL ysTyxuNIy YyQoOoUS (sneeuuTT) ze9;enbrzq sAaydojoe7 avdI lovaLso USTJETTF peTMezos (yoeqso) eydTzos ezragnNTY USTJeTTS payjzodsebuez9 (Tuezuey) snprnd seuryzeyqued uobanp oe Ta (epedgoeq) szebru shyQyoT ren pues —_—— Satoadds adAL a3e23SqQnNS 77 ADDENDUM TO FISH SPECIES LIST Species Site SERRANIDAE Hypoplectrus puella (Cuvier and Patch Reef 2 Valenciennes) Barred hamlet Hypoplectrus nigricans (Poey) Patch Reef 6 Black hamlet Mycteroperca tigris (Valenciennes) East End Tiger grouper Area D LUTJANIDAE Lutjanus analis (Cuvier) Patch Reef 9 Mutton snapper LABRIDAE Clepticus parrai (Bloch and Patch Reef 6 Schneider) Creole wrasse OST RACLIIDAE Lactophrys bicaudalis (Linnaeus) Patch Reef 2 Patch Reef 7 Spotted trunkfish Patch Reef 9 Bone Bay 78 Species changes at night Observations at night indicated that a marked difference in distribution of fish on the reef occurred after dusk. Although abundant during the day, members of the family Pomadysidae were strikingly absent from the reef at night as a 'fish watch! on patch reef V would indicate. Table 4 shows the relative abundance of reef species during the day (mid-day fishwatch) and at night (fishwatch during and immediately after the evening change-over) and the contrast in the number of species present is quite clear. Similar observations have already been recorded in some detail by members of the Tektite Program (Collette and Talbot 1972). During the evening change-over period the snappers (Lutjanidae) and the grunts (Pomadysidae) leave the reef, some in schools and some alone, for the sandflats and seagrass areas where they feed. It was also noticed that the grunts were accompanied in their excursions over the sand by the holocentrids, Holocentrus rufus and Holocentrus coruscus. Myripristis jacobus does not appear to move far away from the reef at night. Results depicted in the Table also indicate the replacement of diurnal species by nocturnal species such as Apogon maculatus, Apogon binotatus and Pempteris schomburgki and such findings are in general agreement with the work of the Tektite Program in Lameshur Bay, St. John, U.S. Virgin Islands. The mangrove area Other interesting observations were made on a mangrove area, also on the leeward side of the island, approximately 1.5 km east of Setting point. Here, on a discrete mangrove patch (measuring approximately 25 m by 15 m in extent and situated in a depth of water of 1m) a rotenone station yielded 22 species of fish (Fig. 13), the majority of which were juveniles. Paricularly abundant were the French grunts Haemulon flavolineatum, the Silver jenny Eucinstomus gula, the Ballyhoo Hemiramphus brasiliensis and the Yellowfin mojarra Gerres cinereus. As Fig. 14 shows, the mangrove area can be regarded as a ‘nursery’ with large numbers of juvenile fish recorded. Indeed, mangroves are known to be highly productive areas which also serve as shelters to young reef fish (Ogden, Yntema and Clavijo 1975). SUMMARY A total of 185 species (55 families) of reef fish were recorded on Anegada. A similar study on Tague Bay Reef, St. Croix, U.S. Virgin Islands, yielded 125 species (44 families) - Ogden et al (1972), although the total number of marine species for the entire island was estimated at 300 by Ogden, Yntema, and Clavijo (1975). Randall (1968), in ‘Caribbean Reef Fishes', cites 300 common reef fish which may be found on reefs and sandflat/seagrass communities in the Caribbean - this figure 79 does not include many of the cryptic species which would be collected in a rotenone station. Similar trends in the distribution of reef fish were observed on the northern shore of Anegada as those recorded in the Taque Bay study, St. Croix; with relatively greater numbers of species being found in the rear and buttress zones as compared to the shallow reef flat or reef top (Fig. 15). The increased number of species in these areas is probably due in part to the greater water depth and diversity of habitat provided by the regions. The highest number of species of reef fish recorded on the northern shore of Anegada was at West End in the buttress zone where 70 species were recognised. Particular note was made at this site of the spadefish Chaetodipterus faber and the tarpon, Megalops atlantica. As is to be expected, generally lower numbers of fish species were recorded over the sand and sand/grass areas. Numbers of species ranged from 18-28 in the sandy lagoons of sites on the northern shore - predominant species including the jacks, (Carangidae); the mojarras (Gerridae); and the sanddivers (Synodontidae). In an area extensively colonised by algae, in the lagoon at East End, up to 40 species were noted; also at East End 44 species were recorded on the inshore fringing reef. The french grunt Haemulon flavolineatum, the reef squirrelfish Holocentrus coruscus, the bluehead Thalassona bifasciatun, and the striped parrotfish Scarus croicensis were particularly abundant at the latter site and a small rotenone station yielded 21 species, including large numbers of the dusky squirrelfish Holocentrus vexillarius, redtail parrotfish Sparisoma chrysopterium, striped parrotfish Scarus croicensis, ocean surgeon Acanthurus bahianis, freckled cardinalfish Apogon conklini and the chestnut moray Enchelycore sp. On the leeward side of the island the number of species recorded varied between 54 - 78 on the nine patch reefs visited. The highest number of species was recorded on patch reef VI at the east end of the island. The most outstanding feature on the patch reefs was the general abundance of the Pomadsyidae - particularly Haemulon flavolineatum and Haemulon sciurus, both species being found shoaling in stands of Acropora prolifera and Acropora palmata. REFERENCES Bohlke, J.E. and Chaplin, C.C.G., 1968. Fishes of the Bahamas and adjacent tropical waters. Livingston Publ. Co., Wynnewood, Pa. 777 pp. Chaplin, C.C.G. and Scott, P., 1972. Fishwatchers guide to the West Atlantic coral reefs. Livingston Publ. Co., Wynnwood, Pa. 65 pp. 80 Collette, B.B. and Talbot, F.E., 1972. Activity patterns of coral reef fishes with emphasis on nocturnal-diurnal changeover. In "Results of the Tektite Program: Ecology of Coral Reef Fishes' Ed. by B.B. Collette and S.A. Earle. Natural History Museum Los Angeles County Science Bulletin 14, 98-124. Hamilton, H.L., 1941. The biological action of rotenone on fresh-water animals. lowa Acad. Sci. 48, 467-479. Iverson, E.S., Krantz, G.E., Rehrer, R., and Beardsley, G., 1973. Fisheries and Mariculture Potential of Anegada Island (B.V.1.) Report to Interbankhouse Inc. 10 pp. Ogden, J.C., Helm, D,, Peterson, J., Smith, A., and Weisman, S. (eds) 1972. Distribution of Fishes on Tague Bay Reef. p. 12-21. In 'An ecological study of Tague Bay Reef, St. Croix, U.S. Virgin Islands.' West Indies Lab., Special Publication 1, 90 pp. Ogden, J.C., Yntema, J.A., and Clavijo, I., 1975. An annotated list of the fishes of St. Croix U.S. Virgin Islands. West Indies Lab., Special Publication 3, 63 pp. Randall, J.E., 1963. Methods of Collecting Small Fishes, Underwat. Natur. 1,5; 6=11, & 32-36). Randall, J.E., 1961. A technique for fish photography. Copeia, 2, 241-242. Randall, J.E., 1968. Caribbean reef fishes. Tropical Fish Hobbyist Publications, Inc., Jersey City, N.J. 318 pp. SPECIES SYNODO ND IDAE Synodus intermedius (Inshore lizardfish) BELONIDAE Strongylura notata (Atlantic needlefish) Tylosurus crocodilus (Houndfish) HOLOCENTRIDAE Holocentrus Holocentrus Holocentrus Myripristis coruscus (Reef squirrelfish) rufus (Squirrelfish) jacobus (Blackbar soldierfish) AULOSTIMIDAE Aulostomus maculatus (Trumpet fish) SPHYRAENIDAE Sphyraena barracuda (Barracuda) APOGONIDAE Apogon maculatus (Flamefish) Apogon binotatus (Barred cardinal fish) SERRANIDAE Cephalophilis fulva (Coney) Epinephelus adscensionis (Rock hind) PRIACANTHIDAE Priacanthus cruentatus (Glasseye snapper) PEMPHERIDAE Pempheris schomburgki (Glassy sweeper) CARING IDAE Caranx ruber (Bar jack) LUTJANIDAE mahogoni (Mahogony snapper) jocu (Dog snapper) Lutjanus apodus (Schoolmaster) Lut janus griseus (Grey snapper) Ocyurus chrysurus (Yellowtail snapper) Lutjanus Lutjanus POMADASYDAE aurolineatum (Tomt ate) flavolineatum (French grunt) sciurus (Bluestriped grunt) Haemulon plumieri (White grunt) Haemulon macrostomum (Spanish grunt) Anisotremus virginicus (Porkfish) Haemmlon Haemulon Haemulon SPARIDAE Calamus bajonado (Jolthead porgy) GERRIDAE Gerres cinereus (Yellowfin mojarra) Eucinostomus gula (Silver jenny) MULLIDAE Mulloidichthys martinicus (Yellow goatfish) KYPHOSIDAE Kyphosus sectatrix (Bermuda chub) CHAETODONTIDAE Holacanthus ciliaris (Queen angelfish) Pomacanthus paru (French angel fish) Chaetodon capistratus (Foureye butterflyfish) Chaetodon striatus (Banded butterflyfish) POMACENT RIDAE ABUNDANCE DAY ascensionis (Longjaw squirrelfish) *** tokik Kk xi tek we kik Has Microspathodon chrysurus (Yellowtail damselfish) ** Eupomacent rusdorsopunicans (Dusky damselfish) Eupomacentrus variabilis (Cocoa damsel fish) Eupomacentrus sp- (Honey dameelfish) Eupomacentrus leucostictus (Beaugregory) Abudefduf saxatilis (Seargeant major) Table 4 by night. telek toiek ek Kk tek Relative abundance NIGHT kk (local) ** wk aekick kick tetoiek (off reef) (local) (near reef) * * SPECIES LABRIDAE Thalassoma bifasciatum (Bluehead wrasse) ou ny ( " juvenile) Halichoeres poeyi (Blackear wrasse) Halichoeres radiatus (Puddingwife) Halichoeres garnoti (Yellowhead wrasse) SCARIDAE Scarus vetula (Queen parrot fish) Scarus taeniopterus (Princess parrot fish) Scarus coelestinus (Midnight parrot fish) Sparisoma rubripinne (Yellowtail parrot fish) Sparisoma chrysopterum (Redtail parrot fish) Sparisoma aurofrenatum (Redband parrot fish) Scarus croicensis (Striped parrot fish) Sparisoma viride (Stoplight parrot fish) BLENNIDAE Ophioblennius atlanticus (Redlip blenny) ACANTHURIDAE Acanthurus coerulus (Blue tang) Acanthurus chirurgus (Doctorfish) Acanthurus bahiansis (Ocean surgeon) OSTRACIIDAE Lactophrys triqueter (Smooth trunkfish) TOTAL SPECIES kk 6-10 week 11-30 ABUNDANCE DAY dete ak 55 Cee? SiloilO@) of reef fish on patch reef V by day and NIGHT * 20 NO. OF INDIVIDUALS COLLECTED —_ a 1 5 0) 2 s o = s 2 r=] — = o punctatus Myrophis Synodus foetens Fig. 13 Strongylura timucu NO. OF FISH COLLECTED Coral reef fish species list and numbers of individuals collected at a rotenone station - a mangrove patch on the leeward shore of Anegada. WZ ge = 2 = o é & o E i= D4 3 3 g g = g Fa e& §& & o = Bees s = = 5 é 2 ‘- s e > = s . 3 8 @¢ & 3B s . £ oe SS & SoS oe sais a & $ s gg 3 8 Se EP aE z E 3 S = v Z * s = £ = = & = S 5 2 = z 3 2 g Ee = = 2 « s s =} 2 (= 2 - = = = s = = Ss — Ss £ 5 a s > 6 & S = = S Ss s = 6 a = & See eis Fig.14 Length - frequency distribution of the French Grunt Haemulon flavolineatum at a rotenone station - a mangrove patch on the leeward shore of Anegada. JUVENILES ADULTS ce) 20 40 60 80 100 120 140 160 180 200 LENGTH (mm) Eucinostomus argenteus Scarus sp. Coryphophopterus dicurus Number of coral reef fish species recorded on A.) the Fig-15 Windward shore and B.) the Lecward shore of Anegada. SSiu1iNng 2 3NOZ YVIY - yoay Gur Gursy < aebjy+ pues NoOoV1 pues az As. dOl 1344 S < INOZ YV34 = pues jNo009V1 —_ > SSaH1ing < dOl 4334 = 3NOZ YV3AY E pues NOOSV1 > = r-) w sayayeg j2109 NOOdVI ) r=) = o us aS ssau ing =a o o sSa¥ling S dO 4334 a sayajyeq e109 — bs sse19+PUES Lo ooy) 3 pues [e) [s) [o} [e) [o} [o) iS) Sy CON TS 2 ESO) AES aye SN S3N0Z 03040934 S$3193dS HSI4 40 ON $3193dS 40 ON & 60 40 20 PATCH REEF ae be teehee aeete Gs 149 em not heey ‘temo \s Ts i.e eo er olan Ce eee rte 4nvea3 vit) pe CIMA ‘aa sang aytew WOOD ‘ae , 7 vae | ATOLL RESEARCH BULLETIN NO. 237 THE INTERTIDAL ALGAE OF THE MAINLAND COAST IN THE VICINITY OF TOWNSVILLE, QUEENSLAND by Yinam Ngan and Ian R. Price Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 *Aeg ATTOQOTYOTL ‘uosn31eq edeg ‘IIIA uot zeIS - g fAeg eTIAN] ‘uosn31aeq adep SITIA uoTIeIS - / SpueTeAEeTD adeg JO WS SIA UOTIeIS — g SAeqeMYyPeIg UTe}ZSeg ‘A UOTIeIS - ¢ $qutog Butssty SAI uot eqs - 4 S‘epueret[eg edeg ‘III uotzeqIg - € SepuereTTeg adep JO MN SII woTIeIS - Z fyoeoeg S,tepunes ‘7 uoTqeIg - T ‘suoT}eIS BuTJDeT[OO Fo uoT}eIOT BuTMoys uoTZe1 aTTTASuUMO] Jo dew *T “3Ta uosn6ie4 ede “ FTIIASNMOL wiod Buissin %, epueseljeg ede! “tn, Puelere|D edes »% %, 4a, pier KOSS On THE INTERTIDAL ALGAE OF THE MAINLAND COAST IN THE VICINITY OF TOWNSVILLE, QUEENSLAND 1/ by Yinam Ngan and Ian R. Price- ABSTRACT One hundred and thirty-nine taxa of intertidal marine algae are recorded from eight stations in the vicinity of Townsville, on the north-east tropical coast of Australia. The flora comprises 25 species of Chlorophyta, 19 of Phaeophyta, 89 of Rhodophyta, and 6 of Cyanophyta, most of which are new records for the area. INTRODUCTION The benthic marine algal flora in the Townsville region has remained virtually unstudied to the present time. Only three algal species have been reported previously (Endean et al., 1956). However, some collections and studies have been made in Queensland waters to the north (see Lucas, 1931; Cribb, 1961; Price et al., 1976) and also further south (see Lucas, 1931; May, 1951; Cribb, 1954, 1956, 1958a, 1958b, 1960, 1965a, 1965b, 1966, 1969, 1971, 1972, 1973, 1975). Earlier references relating to the marine algae of north-east Queens- land are included in Cribb (1954). The following list of species includes references to published descriptions and illustrations, which should facilitate and stimulate further studies of the algal flora in tropical Australia. The list includes many widely distributed tropical Indo-West Pacific taxa, and will enable more detailed biogeographical comparisons with other regions. The check-list relates chiefly to rocky intertidal areas along the mainland coast. It is based on collections at eight different stations (Fig. 1) over a two-year period (February 1975 - November 1977), as part of a detailed ecological study of the intertidal algal communities. Stations IIL, IV, V and VII were sampled monthly throughout this period. 1/ Botany Department, James Cook University of North Queensland, Townsville, Qld 4811, Australia (Manuscript received May 1978--Eds.) A total of 139 taxa, comprising 25 Chlorophyta, 19 Phaeophyta, 89 Rhodophyta and 6 Cyanophyta, is reported. The arrangement of orders and families in the Divisions Chlorophyta, Phaeophyta, Rhodo- phyta, and Cyanophyta generally follows those of Womersley and Bailey (1970), Papenfuss (1951), Kylin (1956), and Desikachary (1973) respectively. In each family, genera and species are listed in alphabetical order. Voucher specimens of all species listed are deposited in the Herbarium of the Botany Department, James Cook University of North Queensland, Townsville, Queensland, Australia (JCT), and an incomplete set of duplicates is housed in the Herbarium of the Botany Department, University of Queensland, Brisbane, Queensland, Australia (BRIU). Voucher specimen numbers (YN...) are given in parentheses in the station listing for each taxon. LOCATION AND DESCRIPTION OF THE STATIONS The study area is located approximately midway along the eastern coast of the State of Queensland, between latitudes 19910" and 19917" S, and longitudes 146°37' and 147904' E (Fig. 1). This dry tropical zone fits essentially into the Aw category in the KSppen scheme (Dick, 1975), with a prolonged and intense drought in the low-sun period of the hemisphere, followed by substantial rainfall during the warmer part of the year (December to March). The shoreline consists generally of alternating granitic rocky headlands and sandy beaches, and includes the estuaries of rivers such as the Bohle, Ross and Haughton. Mangrove vegetation occurs at numerous places along the coast. The study area is sheltered from oceanic swell from the south-west Pacific Ocean by the Great Barrier Reef system. The tides are semi-diurnal with significant diurnal inequality (Easton, 1970). The mean tidal range is 2.5m at spring and 0.8m at neap periods (Queensland Department of Harbours and Marine, 1977). The south-east trade winds prevail for much of the year. The occasional passage of tropical cyclones in summer is generally accompanied by torrential rain, storm surge, and high wave energy. Station I (Saunder's Beach): Gently sloping sandy beach with occasional patches of pebbles and shell-fragments; the latter is the only type of substratum available to larger algae in this unstable habitat. Algae were also collected in the neighbouring estuary of Saunder's Creek. Station II (NW of Cape Pallarenda): Rocky point with frequent sand and silt covered boulders at higher intertidal levels. A very gently sloping sand flat occurs at lower intertidal levels. Station III (Cape Pallarenda): Similar to station II, but the shore densely covered with boulders except at the lowest intertidal levels. Station IV (Kissing Point): Rocky point consisting in the intertidal of a mass of sediment covered boulders and rocks, with a sandy floor at the lowest intertidal levels. The Point is bounded to the east by a sandy beach, and to the west by a small, shallow, muddy bay containing sparse mangrove vegetation near the shore- line and extensive lag gravels at lower levels. Station V (Eastern Breakwater): Artificial harbour wall consisting of relatively smooth granitic rocky surfaces at higher intertidal levels, and irregular boulders at lower levels. Conditions inside the breakwater are sheltered (and heavily silted) in comparison with the exposed outer face. Station VI (SE of Cape Cleveland): Steeply sloping rocky shore consisting of sediment free boulders above and pebbles below. Station VII (Cape Ferguson, Turtle Bay): Steeply sloping rocky shore consisting mostly of large boulders providing numerous shaded situations, with some sandy patches at lower levels. Stations VIII (Cape Ferguson, Ticklebelly Bay): Rocky shore varying from large boulders to a gently sloping muddy beach with occasional boulders, and adjacent to extensive mangrove vegetation. The stations listed above may be divided into two groups. The area between Capes Cleveland and Ferguson, which includes Stations VI and VII, is characterized by steeply sloping sandy beaches and rocky outcrops, higher wave energy, and generally less turbid water. In contrast, stations to the west of Cape Cleveland (I, II, III, IV and V) and to the south of Cape Ferguson (VIII) generally slope more gently, are more sheltered, and the water is usually more turbid. These differences in environmental conditions are reflected in the composition of the intertidal algal flora of these two groups of stations, as will be discussed in detail in a subsequent publication. DIVISION CHLOROPHYTA Order ULOTRICHALES Family Ulvaceae Enteromorpha clathrata (Roth) J. Agardh. Bliding 1963: 107, fig. 67. Station IV (YN 12). E. flexuosa (Wulfen ex Roth) J. Agardh. Bliding 1963: 73, fig. 38. Station Tl; *Lits. IV “GN 373), We Vi, VEL Cine Sap. E. ralfsit Harvey. Bliding 1963: 43, fig. 18. Station IV (YN 390). Occasional lateral proliferations are present as in E. chaetomorphotdes Boergesen (1913: 5, fig. 1), but the cells contain 2-4 pyrenoids. Ulva rigida C. Agardh. Bliding 1968: 546, figs. 6-10. Stations III, IV (YN 112, YN 367), V, VI, VII. Order CLADOPHORALES Family Cladophoraceae Chaetomorpha antennina (Bory) Kuetzing. Boergesen 1913: 16, figs. 4-5; 1940: 37. Cribb 1954: 17, pl. 1, fig. 6. Dawson 1954: 386, filo. 960. Stations LIL, IVS Ve Vien 55), VEL. C. spiralis Okamura 1912: 162, pl. 95. Segawa 1968: 11, fig. 49. Stations IV, VII (YN 389). Comparison with authentic Japanese material is required. Chaetomorpha sp. Stations III, IV (YN 393). Cladophora fascicularis (Mertens ex C. Agardh) Kuetzing 1853: 26, pl. 90, fig. 2. Boergesen 1940: 34, fig. 10. Pham-hoang 1969: ASS MElo. 14538' Stations III, IV, V, VI, VIL (YN 94). C. rugulosa von Martens. Okamura 1910: 103, pl. 80, figs. 1=7- Sakai 1964: 67, figs. 31-2, pl. 15, fig. 1. Pham-hoang 1969: 43 thle. “453566 Stations ile skVve Vv Vie WGN) 151) Vik. Rhizoclonium capillare Kuetzing. Cribb 1965b: 262, pl. 1, figs. 1-8. Stations: EEE IGN 272) nV. Cribb considers that the transfer of R. capillare to Chaetomorpha by Boergesen (1925: 45, fig. 13) on the basis of the number of nuclei is not justified. R. tmplexum (Dillwyn) Kuetzing. Cribb 1965b: 264. R. kochtanum Kuetzing. Boergesen 1913: 19, fig. 7. Stations III (YN 309), IV. R. rtpartum (Roth) Harvey. Taylor 1960: 76. Dawson 1964: 11, pl. 8, figs. B-E. Pham-hoang 1969: 413, fig. 4.18. Stations III (YN 296), IV, VIL, VIII. Family Anadyomenaceae Microdiectyon okamurat Setchell. Yamada 1934: 40, figs. 6-7. Taylor 1950: 46, pl. 27, fig. 1. Dawson 1956: 32, fig. lla. Pham-hoang 1969: 439, fig. 4.43. Station VII (YN 313). The cell wall thickness is about 2 wm and shows little variation. Order SIPHONOCLADALES Family Valoniaceae Valonta aegagropitla C. Agardh. Egerod 1952: 348, pl. 29, fig. b. Dawson 1954: 388, fig. 8j. Cribb 1960: 13. Station IV (YN 320). Family Siphonocladaceae Cladophoropsts herpestica (Montagne) Howe. «Dawson 1954: 390, fig. 8h. Cribb 1960: 10, pl. 4, figs. 5-6. Station IV, V, VI (YN 154), VII. Family Boodleaceae Boodlea compostta (Harvey) Brand. Boergesen 1940: 21, fig. 6; 1946: 15, fig. 5. Dawson 1954: 390, fig. 9c-d. Cribb 1960: 14 Station VII (YN 392). Struvea anastomosans (Harvey) Piccone & Grunow ex Piccone. Boergesen 1913: 54, fig. 39. Dawson 1954: 390, fig. 8g. Cribb 1960: 13. Stations III, IV, V, VI (YN 153), VII. Order CODIALES Family Bryopsidaceae Bryopsts tndtca Gepp & Gepp 1908: 169, pl. 22, figs. 10-11. Boergesen 1953: 6, fig. 1. Cribb 1954: 18. Dawson 1956: 34, fig. 14. Stations IV (YN 27), V. Order CAULERPALES Family Caulerpaceae Caulerpa lentillifera J. Agardh. Weber-van Bosse 1898: 380, pl. 34, figs. 1-2. Cribb1958b: 213, pl.,4,. figs. 1-4, pl. 5,. figs liye Station VII (YN 260). C. racemosa var. laetevirens (Montagne) Weber-van Bosse 1898: 366, pl. 33° figs. 8; 16-22. Cribblosebs 2125 pli35) tics ie Stations? Lil, IVs (YN 69)peV.. VEL. C. taxtfolta (Vahl) C. Agardh. Boergesen 1913: 131, figs. 104-105. Cribb 1958b: 2105 pl. 1, figs. 8=11, pill. 25,fdieseed—oe Stations) LLLP, Tv, Vo° Vil iGyNe95),. C. vertictllata J. Agardh. Weber-van Bosse 1898: 267, pl. 20, figs. 7-10. Boergesen 1913: 121, figs. 95-98. Dawson 1954: 392, fig. 10b. Station V (YN 409). Family Udoteaceae Chlorodesmis hildebrandtii Gepp & Gepp 1911: 16, 137, figs. 74-75. Ducker 1967: 164, pls. 6, 16. Station TLE (YN 391)’. Ducker (1967: 163) queried the occurrence in Queensland of this species, which occupies an intermediate position between C. major and C. fastigiata (given as C. comosa, but see Ducker 1969: 17). Udotea flabellun (Ellis & Solander) Howe. Gepp & Gepp 1911: 131, pl. 3, figs. 26-28. Lucas 1931: 49. Durairatnam 1961: 25, pls 205 ft'ow 2. Stations III (YN 399), IV. Order DASYCLADALES Family Dasycladaceae Acetabularia calyeculus Quoy & Gaimard. Valet 1969: 617, pl. 44, figs. 7-9. Stations TILT, IV (YN 72): DIVISION PHAEOPHYTA Order ECTOCARPALES Family Ectocarpaceae Bachelotia anttllarwn (Grunow) Gerloff 1959: 37. Cardinal 1964: 10. Pylatella antillarum (Grunow) De Toni. Blomquist 1958: 25, figs. 1-17. Station V (YN 407). Feldmannia trregularis (Kuetzing) Hamel 1939: XVII, fig. 61£. Cardinal 1964: 54, fig. 29. Eetocarpus trregularis Kuetzing. Boergesen 1941: 23, figs. 8-11. Dawson 1954: 398, fig. 14e-f. Station III (YN 297). Giffordia mttchellae (Harvey) Hamel 1939: x, fig. 6l1c. Eetocarpus mttchellae Harvey. Boergesen 1941: 7, figs. 1-5. Dawson 1954: 400, fig. 14c-d. Stations III, IV (YN 117), VII. Family Ralfsiaceae Ralfsta sp. Stations II, III, IV (YN 413), V, VI, VII. Order SPHACELARIALES Family Sphacelariaceae Sphacelarta furctgera Kuetzing 1855: 27, pl. 90, fig. 2. Dawson 195ke AOD, Hig, Me, Stations III (YN 292), IV, VII. S. trtbulotdes Meneghini. Boergesen 1941: 41, fig. 18. Dawson 1954: 400, fig. 14i-j. Station VII (YN 281) Order DICTYOTALES Family Dictyotaceae Dietyopterts delicatula Lamouroux. Boergesen 1914: 60, figs. 40-41. Yamada 1950: 187, figs. 5-6. Pham-hoang 1969: 333, fig. 3.32. Station VII (YN 256). D. woodwardit (Brown ex Turner) Schmitz. Durairatnam 1961: 35, pl. 23. Misra 1966: 151, fig. 79. Haltserts woodwardii (Brown) J. Agardh. Kuetzing 1859: 22, pl. 53, fig. 2. Station VII (YN 331). Dictyota bartayrestt Lamouroux. Cribb 1954: 20. pl. 3, fig. 6. Stations LILI, IV (YN 397), V, VIL. Some specimens are similar to D. bartayrestt Lamouroux sensu Vickers (see Jaasund 1970: 72, figs. 1D, 2C). D. ctltolata Kuetzing. Womersley 1958: 148, pl. 1. Taylor 1960: 223, pl. 32, fig. 3, pl. 59, fig. 1. Jaasund 1970: 76, fig. 2A. Stations III, IV, V (YN 249), VII. D. dtchotoma var. tntricata (C. Agardh) Greville. Cribb 1954: 20, pl. 3, fig. 4. Stations III (YN 395), IV. 8 Lobophora vartegata (Lamouroux) Womersley 1967: 221. Womersley & Bailey 1970: 292. Pococktella variegata (Lamouroux) Papenfuss 1943: 467, figs. 1-14. Dawson 1954: 400, fig. 14k. Stations IV, VII (YN 97). Padina tetrastromatica Hauck. Boergesen 1930a: 172, fig. 10, pl. 23 1935: 35. Gaillard 1967: 447, figs. 1-6. Stations III, IV (YN 113), V, VII. Spatoglossum asperum J. Agardh. Boergesen 1935: 35, pl. i5; °1937b2/1313; 1941: 48. Durairatnam 1961:34. Misra 1966: 160, fig. 85. Station VII (YN 335). Order DICTYOSIPHONALES Family Punctariaceae Colpomenta stnuosa (Roth) Derbés & Solier. Okamura 1907b: 86, pl. 19, figs. 11-12, pl. 20, figs. 10-12. Boergesen 1914: 20, fig. 12. Clayton"l975: 187, fies. 5-7, 12-13° Stations ©tL, IV GN 414), Ve Vern. Rosenvingea orientalis (J. Agardh) Boergesen 1914: 26; 1930a: 168; 19387as>25.- Cribb 1954:°24, pl. 2, fig. 3. “Misra 1966s io. Station IV (YN 396). Order FUCALES Family Cystoseiraceae Cystosetra trinodis (Forssk91) C. Agardh. Papenfuss & Jensen 1967: 21, figs. 1-2. Jaasund 1976: 53, fig. 108. Stations IV (YN 400), V, VII. Family Sargassaceae Sargassum olitgocystwn Montagne. Womersley & Bailey 1970: 299, fig. 8, plsa2one ti pe 6 S. bindert Sonder ex J. Agardh. Sonder 1871: 11. J. Agardh 1889: 875, pl. 26, fig. 2. Jaasund 1976: 5/5 fic. ei? peer Stations LV, VeeVie GN S72)is The specimens show many of the characters described for S. oltgocystwn by other authors, although the branches are only slightly compressed. Spines occur on the very short leaf petiole. Sargassum sp. Station VII (YN 401). The material shows a number of the features of S. polycystwun C. Agardh as described by Durairatnam (1961: 46, pl. 10, figs. 14-18) and Jaasund (1976: 57, fig. 115), for example muricate branches, but fertile specimens have not been found. DIVISION RHODOPHYTA Order NEMALIALES Family Helminthocladiaceae LItagora sp. Station VII (YN 354). Family Chaetangiaceae Galaxaura oblongata (Ellis & Solander) Lamouroux. Boergesen 1927: 71, figs. 39-41. Chou 1947: 7, pls. 2-3, 9. Papenfuss & Chiang 1969: 310, fig. 5. Stations IV (YN 394), VIL. Setnata moretonensts Levring 1953: 509, figs. 39-40. Womersley 1958: 153. Stations IV, VII (YN 356). Family Bonnemaisoniaceae Asparagopsts taxtformts (Delile) Trevisan. Boergesen 1919: 352, fig. 347-351. Dawson 1953: 57. Station IV, V (YN 415). Falkenbergta hillebrandit (Bornet) Falkenberg. Boergesen 1919: 331, figs. 332-333. Dawson 1954: 414, fig. 252. Stations IV, V (YN 46). Order GELIDIALES Family Gelidiaceae Geltdtum corneum (Hudson) Lamouroux. Boergesen 1920: 114, fig. 124. Sreenivasa Rao 1970: 71, fig. 5A-E, pl. 2, fig. h. Taylor 1928: 1425 pl. 28), fig. 2: Stations IV, V (YN 342), VIL. Further comparative studies are required to confirm this identification. G. ertnale (Turner) Lamouroux. Feldmann & Hamel 1936: 240, fig. 22. Station III (YN 277; YN 370). The vegetative and reproductive (cystocarpic and tetrasporic) structure fits that described by Feldmann and Hamel. G. ertnale var. perpustllum Piccone & Grunow. Weber-van Bosse 1921: 225. Dawson 1954: 421, fig. 3le-f. Stations II, III (YN 337), IV. G. heteroplatos Boergesen 1934: 3, fig. 3. Jaasund 1976: 73, fig. 145. Stations III, IV (YN 202), V, VII (YN 279). 10 Certain specimens referred to this species show some features of G. spathulatum (Kuetzing) Bornet (see Pham-hoang 1969: 123, fig. 2.53). G. pustllum (Stackhouse) Le Jolis. Boergesen 1924: 279, fig. 26, Dawson 1954: 420, fig. 3la-c. Taylor 1960: 354, pl. 45, fig. 4: Stations III, IV (YN 119), V, VII (YN 161). Geltdtum sp. 1 Station V (YN 411). The shape and arrangement of the cystocarps, and the anatomy of the thallus are similar to those described for G. heteroplatos. However the rather short and somewhat terete thalli and the form of the tetrasporangial stichidia seem distinct. Geltdium sp. 2. Station IV (YN 198, YN 326). The specimens superficially resemble those referred to Geltdium corneum , but they are more slender and less cartilaginous. Rhizines are confined to the inner cortex, but as female material has not been collected the generic relationships remain uncertain. Order CRYPTONEMIALES Family Rhizophyllidaceae Chondrococeus hornemannit (Lyngbye) Schmitz. Weber-van Bosse 1921: 255. Womersley & Bailey 1970: 305. Desmta hornemannit Lyngbye. Kylin 1956: 166, fig. 113. Station VII (YN 374). Family Hildenbrandtiaceae Htldenbrandtia prototypus Nardo. Dawson 1953: 95, pl. 7, fig. 43 1954: 424, fig. 36a-b. Stations II, III, IV, V. VI (YN 152), VII, VIII. Family Corallinaceae Amphtroa fragilissima (Linnaeus) Lamouroux. Weber-van Bosse 1904: 89, pl. 16, figs. 1-2, 5. Dawson 1954: 430, fig. 40g-h. Taylor 1960: 403, pl. 47, figs. 1-2. Stations DL, iVA@Nedds) > Voss Cheilosporun spectabile Harvey ex Grunow. Weber-van Bosse 1904: 106. Boergesen 1935: 51, fig. 23. Pham-hoang 1969: 146, fig. 2.75. Womersley & Bailey 1970: 314, fig. 22, pl. 26. Station VII (YN 378). Jania adhaerens Lamouroux. Boergesen 1917: 195, figs. 184-7. Pham-hoang 1969: 142, fig. 2.71. Jaasund 1976: 77, fig. 154. Stations \IVs \ V5. Vire@yN 416). Janta sp. Station VEIT (YN 255). 11 The conceptacle morphology is identical to that described for J. rubens (Linnaeus) Lamouroux by Kuetzing (1857: 40, pl. 84, figs. 2-3) and Pham-hoang (1969: 140, fig. 2.68). The branching pattern is very variable in our material and further confirmation is required. Family Grateloupiaceae Grateloupta divaritcata Okamura 1915: 55, pls. 116, 117, figs. 12-18; 1942: 99. Stations III (YN 307), IV (YN 196), V, VII (YN 162). The morphology of this species is highly variable. Comparison with authentic Japanese material is required. Order GIGARTINALES Family Gracilariaceae Ceratodietyon spongtosum Zanardini. Okamura 1909: 1, pls. 51-52. Dawson 1954: 438, fig. 48c. Stations III, IV (YN 408). Geltdiopsts scoparta (Montagne & Millardet) Schmitz. Boergesen 1952: 26, figs. 13-14; 1954: 22, fig. 7. Pham-hoang 1969: 178, fig. 2.108. Station VII (YN 376). G. vartabtlis (Greville ex J. Agardh) Schmitz. Weber-van Bosse 1928: 426. Pham-hoang 1969: 177, fig. 2.107. Jaasund 1976: 87, fig. 176. Station IV (YN 357), VII. Gractlarta crassa Harvey ex J. Agardh. Boergesen 1936: 86, fig. 8. Dawson 1954: 438, fig. 48b. Ohmi 1958: 25, fig. 11, pl. 5, figs. D-E. Jaasund 1976: 85, fig. 170. Stations III, IV (YN 115), V, VII. G. edults (Gmelin) Silva. Ohmi 1958: 16, fig. 6, pl. 3, fig. B. Durairatnam 1961: 62, pl. 14, figs. 4-5. Jaasund 1976: 85, Ie, IZ Stations IV (YN 199), V, VII (YN 310). G. purpurascens (Harvey) J. Agardh. Weber-van Bosse 1928: 437. Gjombl IUS3Z) S05 seslfys WAG jlo (5 seslersiq (C10), Stations IV (YN 365), VII (YN 375). G. rhodotricha (Dawson) Papenfuss 1966: 100. Gractlartopsts rhodotricha Dawson 1949: 47, pl. 19, figs. 3-7. Ohmi 1958: 47, fig. 23, pl. 10, figs. A-B. Stations III (YN 193, YN 371), IV. Papenfuss includes Gractlartopsts rhodotricha in the genus Gractlarta. The similarity of our material to the descriptions given by Dawson and Ohmi is striking. However, the plants from this area are smaller and more slender, and male reproductive organs have not been found. ly G. textortt (Suringar) De Toni. Weber-van Bosse 1928: 438. May 1948: Gi YOhmit 1958405 “£igs. 20-21: Stations III, IV (YN 350), VII (YN 96). G. verrucosa (Hudson) Papenfuss. Dawson 1954: 438, fig. 49. Ohmi 19593 16, tigs. 1-2. epi. 1, £igs.. A—D. Stations fil GYN) 192) LV. This species is often difficult to distinguish from G. rhodotricha in the field, but can be recognized by the three- layered cortex, the hemispherical cystocarps with numerous nutritive filaments, and the many-layered pericarp. Family Sphaerococcaceae Caulacanthus ustulatus (Mertens) Kuetzing 1868: 3, pl. 8. Boergesen 1933b: 115; 1950: 19, figs. 5-6. Feldman & Hamel 1936: 256, figs. 31-33. Stations ILE (YN°274), IV’ @N 203). V CN 351) vine A very variable species in this area, and may include C. indicus Weber-van Bosse (1921: 222, fig. 67). Family Solieriaceae Sareconema fitliforme (Sonder) Kylin 1932: 22. Papenfuss & Edelstein L974: SL, Eiess I=35 rs, 20-25. Station IV (YN 321, YN 369). Solterta mollis (Harvey) Kylin 1932: 20, pl. 6, fig. 12. Segawa 1968: 84, fig. 397. Pham-hoang 1969: 189, fig. 2.121. Rhabdonia mollis Harvey 1863, synop.: 41. Stations III (YN 269; YN 301), IV. The specimens agree well with the original description given by Harvey and appear to be distinct from S. robusta. S. robusta (Greville) Kylin 1932: 18. Boergesen 1950: 13. Jaasund 1976: 93, fig. 188. Stations III (YN 270), IV. Family Rhabdoniaceae Catenella ntpae Zanardini. Tseng 1942: 143, fig. 2. Dawson 1954: 443, fig. 52£. Stations if) FTL ‘(YN 276)’, IV QIN 157), Viet, vere Family Hypneaceae Hypnea boergesenit Tanaka 1941: 233, figs. 6-8, pl. 53, fig. 1. Dawson 1954: 436, fig. 46K. Stations III, IV (YN 346), VII. H. cenomyce J. Agardh. Tanaka 1941: 250, fig. 21. Pham-hoang 1969: Oy rig. 2.129 Stations V, VI (YN 165). 13 H. cervicornts J. Agardh. Tanaka 1941: 240, fig. 13. Pham-hoang 1969: IQA. ait, Zo IAS Stations III (YN 340), IV (YN 347), VII. H. cornuta (Kuetzing) J. Agardh. Tanaka 1941: 242, fig. 14. Dawson 1954: 436, fig. 46c. Jaasund 1976: 99, fig. 200. Station IV (YN 360). H. espert Bory. Kuetzing 1868: 9, pl. 26, figs. A-C. Tanaka 1941: 243, fig. 15. Dawson 1954: 436, fig. 46h-}j. Stations III, IV (YN 345), VII. H. pannosa J. Agardh. Kuetzing 1868: 9, pl. 27, figs. i-k. Weber-van Bosse 1928: 445, fig. 193. Tanaka 1941: 247, fig. 20. Jaasund 1976: 97, fig. 196. Stations V, VII (YN 384). H. valentitae (Turner) Montagne. Dawson 1954: 436, figs. 462, 47. H. charoitdes Lamouroux. Kuetzing 1868: 8, pl. 22, figs. a-b. Sikzizaloms Iny (aN SY) 5 WEIL. Family Dicranemaceae Dieranema rosaltae Setchell & Gardner 1924: 745, pl. 22, fig. 6. DawSong 95/2 lel oheet leche Zales ay. Station VIL (YN 304). Our specimens, like those of Setchell and Gardner and of Dawson, are sterile. Family Phyllophoraceae Gymnogongrus grtffithstae (Turner) Martens. Kuetzing 1869: 24, pl. 65, figs. e-g. Gayral 1958: 398, pl. 107. Pham-hoang 1969: AVS, iS, Do ilsye. Siegueiopre, IY (aN IS) 5 Wa WALES G. pygmaeus J. Agardh. Kuetzing 1869: 24, pl. 64, figs. c-d. Dawson IOSyAg “AO. ites Sile. Stations III, IV (YN 349), V (YN 250), VII. Family Gigartinaceae Gigartina sp. Statvonse LVAIAVNIGY NeSSS) en Wille Order RHODYMENIALES Family Rhodymeniaceae Coelothrix indica Boergesen 1944: 14, figs. 9-11; 1950: 40, figs. 20-21. 14 Stations) IV @Ni377) vik Dawson (1957: 115) suggests that C. tndica is conspecific with C. trregularis Boergesen (1920: 389, figs. 373-4), but our specimens agree better with Boergesen's description of the Mauritius species. Rhodymenta leptophylla J. Agardh. Weber-van Bosse 1928: 461. Kylin 1931: 20, pl. 6, fig. 16. Chapman & Dromgoole 1970: 134, plows Station Vil (YN 282): This species has been considered endemic to New Zealand (Chapman & Dromgoole, 1970). However, Dawson (1941: 144, pl. 20, fig. 18; pl. 27, fig. 39) described C. leptophyllotdes as a very closely related and possibly identical species from Hawaii. Family Lomentariaceae Champta parvula (C. Agardh) Harvey. Boergesen 1920: 407, figs. 392-393. Dawson 1954: 443, fig. 52C. Womersley & Bailey 1970: 321. Stations III, IV, VII (YN 353). Order CERAMIALES Family Ceramiaceae Centroceras clavulatum (C. Agardh) Montagne. Dawson 1954: 466, fig. 54h. Womersley & Bailey 1970: 323. Stations’ TEE, TV, Vs VEWQN 147), VEL; Ceramtum fasttgtatun (Roth) Harvey. Boergesen 1918: 241, fig. 231. Taylor 1928: 191, pl. 27. Nakamura 1965: 129, fig. 4. Pham- hoang, 1969 3237/5" fi. ee Look Station’ TIL (YN 363). C. gractllimum var. byssoidewn (Harvey) Mazoyer. Dawson 1954: 448, fig. 55e-f. Nakamura 1965: 136, fig. 6. Itono 1972: 76, figs. 2-3. Stations TLL, LV (YN 380), V; VIL. C. maryae Weber-van Bosse 1923: 324, figs. 117-118. Dawson 1954: 448, fig. S6e—i). Station IV (YN 364). C. mazatlanense Dawson 1950: 130, pl. 2, figs. 14-15; 1954: 448, fig. S5e=]. Ltono 1972:82,, fies 110 B-ce Staton) Viele nG@yNi O79) Ceramtum sp. Station VII (YN 352). The specimens have forcipate tips, cortical bands with a central whorl of larger periaxial cells, and whorled tetrasporangia. They approach C. marshallense Dawson (1957: 20, fig. 27a-b), but differ in their broader branches (200 um), unequal dichotomies, and wider cortical bands (130 um). 15 Griffithsta sp. Station VII (YN 382). Pleonosportum sp. Station IV (YN 319), V (YN 412). Rhizoids occur on the lower parts of the main axes, and polysporangia are present. Spyrtdta filamentosa (Wulfen) Harvey. Boergesen 1917: 233, figs. 222-6. Dawson 1954: 444, fig. 541. Jaasund 1976: 111, fig. 224. Station IV (YN 381). Family Delesseriaceae Caloglossa bombayensis Boergesen 1933b: 127, figs. 10-12. Stations II, III, IV (YN 280). The status of this species in relation to C. ogasawaraensts Okamura has been discussed by Tseng (1945: 163) and Post (1943: 139; 1955: 371). C. leprteurtt var. hookert (Harvey) Post 1943: 127, figs. 3-5, 24. Taylor 1960: 544. Stations I, II, III, IV (YN 158), V, VI, VIII. Family Dasyaceae Heterostphonta multtceps (Harvey) Falkenberg 1901: 654. Dasya multtceps Harvey. Kuetzing 1863: 27, pl. 77. Station IV (YN 368). H. wurdemannt var. laxa Boergesen 1919: 324, figs. 327-8. Dawson 1956: 57, fig. 60; 1963: 404, pl. 129, fig. 1. Station IV (YN 325). Family Rhodomelaceae Acanthophora muscotdes (Linnaeus) Bory. Okamura1907a: 38, pl. 8, fig. 8-10. Taylor 1960: 619, pl. 72, fig. 3. Jaasund 1976: 137, fig. 278. Stations III (YN 191), IV, VII. A. sptctfera (Vahl) Boergesen 1918: 259, figs. 235-8. Weber-van Bosse 1923: 347, figs. 131-2. Dawson 1954: 456, fig. 6la-b. Stations III (YN 190), IV (YN 110), V, VIL. Acrocystts nana Zanardini. Okamura1907a: 23, pls. 6-7. Dawson 1954: 461, fig. 63b-c. Jaasund 1976: 143, fig. 291. Station VII (YN 334). Bostrychta bindert Harvey 1847: 68, pl. 28. Tseng 1943a:177, pl. 1, figs. 7-8. Stations III (YN 317), IV (YN 159), VII, VIIL. 16 B. kelanensts Grunow ex Post 1936: 20; 1955: 356; 1957: 90. Tseng W943a: 1695 pil 2, fies. d—5. Station IV (YN 402). B. radicans (Montagne) Montagne. Post 1936: 13. Tseng 1943a: 168, pl. 1, figs. 1-3. ‘Dawson 1954: 452, fig. 59d-e. Jaasund 1976: 127, fig. 257. Stations I, II, III, IV (YN 156, YN 252), V, VII, VIII. B. tenella (Vahl) J. Agardh. Boergesen 1918: 300, figs. 299-303. Tseng 194342176, pl. 1, fig. 6. Jaasund 1W76s 127) sticcmeoor Stations ehms@yNn 275)), LV. Chondrta dasyphylla (Woodward) C. Agardh. Taylor 1928: 170, pl. 34, fig. de) Jaasund =e SS, Seta eZee lemons Statdons Li @N'S03)., LVe Gu ratnfordty ‘Lucas 1927, 560; pilss*45, 475 fig. 2. Stations III (YN 189), IV. Chondrta sp. 1. Station VII (YN 386). The specimens superficially resemble C. dasyphylla, although cystocarpic and tetrasporic plants are only 1-2 cm tall. Branch tips are truncate, but with a projecting apical region. The walls of internal cells are markedly thickened over their whole surface. Chondrta sp. 2. Station VII (YN 385). The specimens show good agreement with the descriptions of C. collinstana Howe given by Taylor (1960: 617) and Jaasund (1976: 135, fig. 275), except that the pericentral cell walls are strongly thickened over their whole surface. Herpostphonta tnstdiosa (Greville ex J. Agardh) Falkenberg. Okamura 1930: 25, pl. 264, figs. 10-16. Dawson 1954: 452, figs. 58h-i. Station VII (YN 362). H. tenella forma secunda (C. Agardh) Hollenberg 1968: 556. H. secunda (C. Agardh) Falkenberg. Boergesen 1918: 469, fig. 428. Stations Lit, LV (YN 404) V> Vibe Laurencia dotyt Saito 1969: 154, figs. 9 A-C, 10 A-B. Station VIL. (YN 257). L. gracilis Hooker & Harvey. Yamada 1931: 212, fig. M, pl. 12, Eatp! ible Stations IV (YN 358), VII. i) 17 majuscula (Harvey) Lucas 1935: 223, Saito & Womersley 1974: 819, figs. 1A, 6. Station VII (YN 92, YN 283, YN 329). nidtftea J. Agardh. Yamada 1931: 202. Boergesen 1945: 47, figs. 21-24. Cribb 1958a: 168, pl. 5, fig. 12, pl. 6, figs. 1-3. Saito 1969: 152, fig. 5. Stations III, IV (YN 253), VII. paptllosa (C. Agardh) Greville. Yamada 1931: 190, pl. 1, figs. a-b. Dawson 1954: 458, fig. 61i. Cribb 1958a: 169. pl. 7, figs. 6-8. Saito 1969: 158. Stations III (YN 146), IV, VII. perforata (Bory) Montagne. Kuetzing 1865: 18, pl. 49, figs. e-g. Boergesen 1930b: 69, fig. 26. Yamada 1931: 193, figs. A-B, pl. 3, fig. b. Cribb 1958a: 164, pl. 3, figs. 1-2. , Stations IV (YN 201), VII (YN 287). pygmaea Weber-van Bosse 1913: 122, pl. 12, fig. 6. Dawson 1954: 458, fig. 62k. Cribb 1958a: 166, pl. 4, figs. 1-6. Stations IV (YN 70), V, VII. . Sucetsa Cribb 1958a: 163, pl. 1, figs. 1-3. Saito 1969: 157. Stations VII (YN 288, YN 327, YN 359). tenera Tseng 1943b: 200, pl. 1, fig. 6, pl. 2, figs. 5-6. Dawson 1954: 458, fig. 62b-c. Cribb1958a 167, pl. 5, figs. 1-10. Saenger 1973: 25, figs. 12-13. Station VII (YN 289, YN 333). The specimens agree very well with the detailed descriptions of the sterile and tetrasporic materials given by Tseng. However, no peripheral haptera are present in our specimens. Laureneia sp. Station VII (YN 388). This species belongs to the subgenus Chondrophycus (see Saito 1969: 148), and approaches LZ. cartilaginea Yamada (1931: 230, fig. 0, pl. 19a) in anatomy (see Saito 1967: 53, fig. 43). However, our material appears to be more slender (to 1 mm broad at the base), more or less cylindrical throughout, and the ultimate branchlets are not densely arranged on the upper branches. Levetllea jungermanntotdes (Martens & Hering) Harvey. Falkenberg 1901: 392, pl. 6, figs. 1-13, pl. 14, figs. 18-27. Dawson 1954: 461, fig. 63a. Station VII (YN 355). Polystphonta coacta Tseng 1944: 71, pl. 2. Pham-hoang 1969: 258, tie, 2 Alte), Station IV (YN 361), VII (YN 343). 18 P. fragtlts Suringar. Okamura 1929: 7, pl. 255. Dawson 1954: 452, fig. 60a-b. Station III (YN 338). P. subttltsstma Montagne. Kuetzing 1863: 10, pl. 28, fig. la-e. Tseng 1944: 70, pl. 1. Pham-hoang 1969: 255, fig. 2.185. Stations LiL (YN 300), EV, V. VEL. Tolyptoecladta glomerulata (C. Agardh) Schmitz. Falkenberg 1901: 177, pl. 21, figs. 27-29. Dawson 1954: 452, fig. 59b REFERENCES Agardh, J.G. 1889. Species Sargassorum Australiae. K. svenska Vetensk-Akad. Handl. 23(3), 1-133, pls. 1-31. Bliding, GC. »1963- 9-4 eritical survey of European taxa in Ulvales. Part I. Capsostphon, Percursarta, Bltdingia, Enteromorpha. Op. bot. Soe. bot. Lund. 8(3), 1-160. Bliding, C. 1968. A critical survey of European taxa in Ulvales. Part II. Ulva, Ulvaria, Monostroma, Kornmannia. Bot. Nottser 121: 535-629. Blomquist, H.L. 1958. The taxonomy and chromatophores of Pylaiella antillarum. J. Elisha Mitchell Set. Soc. 74, 25-30. yr Boergesen, F. 1913. The marine algae of the Danish West Indies. Vol. 1. Chlorophyceae. Dansk bot. Ark. 1, 1-160. Boergesen, F. 1914. The marine algae of the Danish West Indies. Part 2. Phaeophyceae. Dansk bot. Ark. 2, 1-66. Boergesen, F. 1917. The marine algae of the Danish West Indies. Vol. 2. Rhodophyceae, Part 3. Dansk bot. Ark. 3, 145-240. Boergesen, F. 1918. The marine algae of the Danish West Indies. Vol. 2. Rhodophyceae, Part 4. Dansk bot. Ark. 3, 241-304. Boergesen, F. 1919. The marine algae of the Danish West Indies. Vol. 2. Rhodophyceae, Part 5. Dansk bot. Ark. 3, 305-368. Boergesen, F. 1920. The marine algae of the Danish West Indies. Vol. 2. Rhodophyceae, Part 6. Dansk bot. Ark. 3, 369-504. Boergesen, F. 1924. Marine Algae from Easter Island. In C.Skottsberg (ed.), The Natural History of Juan Fernandez and Laster Tsland, Vol. 2, pp. 247-309. Almqvist and Wiksells, Uppsala. Boergesen, F. 1925. Marine algae from the Canary Islands, especially from Teneriffe and Gran Canaria. I. Chlorophyceae. Biol. Meddr. 5(3), 1-123. Boergesen, F. 1927. Marine algae from the Canary Islands, especially from Teneriffe and Gran Canaria. III. Rhodophyceae. Part fe Bangiales and Nemalionales. Btol. Meddr. 6(6), 1-97. Boergesen, F. 1930a. Some Indian green and brown algae especially from the shores of the Presidency of Bombay. Jd. Indian bot. Soc. GO. 151174, pls. A=Z. Boergesen, F. 1930b. Marine algae from the Canary Islands, especial]; from Teneriffe and Gran Canaria. III. Rhodophyceae. Part 11 Ceramiales. Btol. Meddr. 9(1), 1-159. 21 Boergesen, F. 1933a. Some Indian green and brown algae especially from the shores of the Presidency of Bombay. III. J. Indian bot. Soe. 12, 1-16, pls. 1-5. Boergesen, F. 1933b. Some Indian Rhodophyceae especially from the shores of the Presidency of Bombay. III. Kew Bull., 113-142. Boergesen, F. 1934. Some Indian Rhodophyceae especially from the shores of the Presidency of Bombay. IV. Kew Bull., 1-30. Boergesen, F. 1935. A list of marine algae from Bombay. Btol. Meddr. 12(2), 1-64, pls. 1-10. Boergesen, F. 1936. Some marine algae from Ceylon. Ceylon J. Sct. A, 1D, S78 Boergesen, F. 1937a. Contributions to a South Indian marine algal Plea, Lo do Hpdicm Ob. SOG. NO, I-5O, joil, ihe Boergesen, F. 1937b. Contributions to a South Indian marine algal PL@iegi5 Miko do shacklon, [dice SOEs) WG) sis) Boergesen, F. 1940. Some marine algae from Mauritius. I. Chlorophyceae. Btol. Meddr. 15(4), 1-81, pls. 1-3. Boergesen, F. 1941. Some marine algae from Mauritius. II. Phaeophyceae. Btol. Meddr. 16(3), 1-81, pls. 1-8. Boergesen, F. 1944. Some marine algae from Mauritius. III. Rhodophyceae. Part 3. Rhodymeniales. Btol. Meddr. 19(6), 1-32. Boergesen, F. 1945. Some marine algae from Mauritius. III. Rhodophyceae. Part 4. Ceramiales. Btol. Meddr. 19(10), 1-68. Boergesen, F. 1950. Some marine algae from Mauritius. Additions to the parts previously published. IT. Btol. Meddr. 18(11), 1-46. Boergesen, F. 1952. Some marine algae from Mauritius. Additions to the parts previously published. IV. Btol. Meddr. 18(19), 1-72, joukiss5 dkea5 ye, Boergesen, F. 1953. Some marine algae from Mauritius. Additions to the parts previously published. V. Btol. Meddr. 21(9), 1-62, pls. 1-3. Boergesen, F. 1954. Some marine algae from Mauritius. Additions to the parts previously published. VI. Bztol. Meddr. 22(4), 1-51. Cardinal, A. 1964. tude sur les Ectocarpacées de la Manche. Beth. Wova Hedwtgta 15, 1-86, figs. 1-41. 22 Chapman, V.J. and F.I. Dromgoole. 1970. The marine algae of New Zealand. Part 3. Rhodophyceae. Issue 2. Florideophycidae: Rhodymeniales. Cramer, Lehre. Chou, R. C-Y. 1947. Pacific species of Galaxaura. II. Sexual types. Pap. Mich. Acad. Set. 31, 3-24, pls. 1-13. Clayton, M.N. 1975. A study of variation in Australian species of Colpomenia (Phaeophyta, Scytosiphonales). Phycologta 14, 187-195. Cribb, A.B. 1954. Records of marine algae from south-eastern Queensland. I. Pap. Dep. Bot. Univ. Qd. 3, 15-37. Cribb, A.B. 1956. Records of marine algae from south-eastern Queensland. II. Polysiphonta and Lophostphonia. Pap. Dep. Bot. Univ. Qd. 3, 131-147. Cribb, A.B. 1958a. Records of marine algae from south-eastern Queensland. III. Laurencta.Pap. Dep. Bot. Univ. Qd. 3, 159-191. Cribb, A.B. 1958b. Records of marine algae from south-eastern Queensland. IV. Caulerpa. Pap. Dep. Bot. Univ. Qd. 3, 209-220. Cribb, A.B. 1960. Records of marine algae from south-eastern Queensland. V. Pap. Dep. Bot. Univ. Qd. 4, 1-31. Cribb, A.B. 1961. Some marine algae from Thursday Island and surrounding areas. Pap. Dep. Bot. Univ. Qd. 4, 51-59. Cribb, A.B. 1965a. The marine and terrestrial vegetation of Wilson Island, Great Barrier Reef. Proc. Roy. Soc. Qd. 77, 53-62, pls. 2-3. Cribb, A.B. 1965b. An ecological and taxonomic account of the algae of a semi-marine cavern, Paradise Cave, Queensland. Pap. Dep. Bot. Untv. Qd. 4, 259-282. Cribb, A.B. 1966. The algae of Heron Island, Great Barrier Reef, Australia. Part I. A general account. Univ. Qd. Pap, Heron Is. Ress Stas. (its) d=23% Cribb, A.B. 1969. The vegetation of North West Island. Qd. Nat. 19, 85-93. Cribb, A.B. 1971. Queensland algae. Qd. Nat. 20, 14-20. Cribb, A.B. 1972. Vegetation of Hoskyn I. and Reef. Qd. Nat. 20, 92-100. 23 Cribb, A.B. 1973. The algae of the Great Barrier Reefs. In 0.A. Jones and R. Endean (eds.), Btology and Geology of Coral Reefs, Vol. IT: Biology 1, pp. 47-75. Academic Press, New York. Cribb, A.B. 1975. Algal vegetation of Masthead Island Reef. Qd. Nat. 21, 79-83. Dawson, E.Y. 1941. A review of the genus Rhodymenta with descriptions of new species. Allan Hancock Pactf. Exped. 3, 123-153, pls. 18-30. Dawson, E.Y. 1949. Studies of northeast Pacific Gracilariaceae. Occ. Pap. Allan Haneock Found. 9, 1-54, pls. 1-25. Dawson, E.Y. 1950. A review of Ceramitwm along the Pacific coast of North America with special reference to its Mexican representatives. Farlowta 4, 113-138. Dawson, E.Y. 1953. Marine red algae of Pacific Mexico. Part I. Bangiales to Corallinaceae Subf. Corallinoideae. Allan Hancock Pactf. Exped. 17, 1-239, pls. 1-33. Dawson, E.Y. 1954. Marine plants in the vicinity of the Institut Océanographique de Nha Trang, Viet Nam. Pactf. Sct. 8, 371-481. Dawson, E.Y. 1956. Some marine algae of the Southern Marshall Islands. Pactf. Set. 10, 25-66. Dawson, E.Y. 1957. An annotated list of marine algae from Eniwetok Atoll, Marshall Islands. Pactf. Set. 11, 92-132. Dawson, E.Y. 1963. Marine red algae of Pacific Mexico. Part 8. Ceramiales: Dasyaceae, Rhodomelaceae. Wova Hedwigta 6, 401-481, pls. 126-171. Dawson, E.Y. 1964. The Seaweeds of Peru. Cramer, Weinheim. Desikachary, T.V. 1959. Cyanophyta. Indian Council Agricultural Research, New Delhi. Desikachary, T.V. 1973. Status of classical taxonomy. In N.G. Carr and B.A. Whitton (eds.), The Btology of Blue-Green Algae. Bot. Monogr. No. 9, pp. 473-481. Blackwell Scientific Publs., Oxford. Dick, R.S. 1975. A map of the climates of Australia: according to Koppen"s principles of definition. Queensland Geogr. J. 3, 33-69. Drouvet, F. 1968. Revtston of the classtficatton of the Oscillatortaceae. Acad. Nat. Sci. Philadelphia Monogr. 15, 1-370. Ducker, S.C. 1967. The genus Chlorodesmts (Chlorophyta) in the Indo- Pacific region. Nova Hedwigia 13, 145-182, pls. 25-43. 24 Ducker, S.C. 1969. Additions to the genus Chlorodesmis (Chlorophyta) . Phycologta 8, 17-19. Durairatnam, M. 1961. Contribution to the study of the marine algae of Ceylon. Bull. Ceylon Fish. 10, 1-181. Easton, A.K. 1970. The tides of the continent of Australia. Horace Lamb Centre for Oceanographical Research, Res. Pap. 37, 1-326. Egerod, L.E. 1952. An analysis of the siphonous Chlorophycophyta, with special reference to the Siphonocladales, Siphonales, and Dasycladales of Hawaii. Univ. Calif. Publs. Bot. 25, 325-454, pls. 29-42. Endean, R., R. Kenny and W. Stephenson. 1956. The ecology and distribution of intertidal organisms on the rocky shores of the Queensland mainland. Austr. J. Mar. Freshw. Res. 7, 88-146, pls. 1-7. Falkenberg, P. 1901. Die Rhodomelaceen des Golfes von Neapel. In Fauna und Flora des Golfes von Neapel. Monogr. 26. Friedlander, Berlin. Feldmann, J. and G. Hamel. 1936. Floridées de France. VII. Gelidiales. Rev. Algol. 9, 209-264, pls. 1-5. Caillard, J. 1967. ftude monographique de Padina tetrastromattca (Hauck.). Bull. I.F.A.N. 29, Ser. A2, 447-463. Gayral.)2s 1958. Algues de la Cote Atlantique Marocaine. Soc. Sci. Nat. Phys. Maroc, Rabat. Gepp, A. and E.S. Gepp. 1908. Marine Algae (Chlorophyceae and Phaeophyceae) and marine phanerogams of the 'Sealark' Expedition, collected by J. Stanley Gardiner, M.A., F.R.S., F.L.S. Trane. Lim. Soc. Lond. (Ser. 2. Bot.) 7, 163-188, pls. 22-24. Gepp, A. and E.S. Gepp. 1911. The Codiaceae of the Siboga Expedition. Siboga Exped. Monogr. 62, 1-150, pls. 1-22. Ger lots ie OD Bachelotia (Bornet) Kuckuck ex Hamel oder Bachelotia (Bornet) Fox? Nova Hedwigta 1, 37-39. Hamel, G. 1939. Phéophycées de France. Fasc. V, 337-432. Paris. Harvey, W.H. 1847. WNerets Australts. London. Harvey, W.H. 1863. Phycologia Australica. Vol. 5; synop., pp- 1-73. Reeve, London. Hollenberg, G.J. 1968. An account of the species of the red alga Herpostphonia occurring in the central and western tropical Pacific Ocean. Pactf. Set. 22, 536-559. 25 Itono, H. 1972. The genus Ceramtum (Ceramiaceae, Rhodophyta) in southern Japan. Bot. Mar. 15, 74-86, Jaasund, E. 1970. Marine algae in Tanzania IV. Bot. Mar. 13, 71-79. Jaasund, E. 1976. Interttdal Seaweeds tn Tanzania. University of Tromsé, Troms¢. Kuetzing, F.T. 1853. Tabulae Phycologtcae, vol. 3. Nordhausen. Kuetzing, F.T. 1855. Tabulae Phycologtcae, vol. 5. Nordhausen. Kuetzing, F.T. 1857. Tabulae Phycologtcae, vol. 8. Nordhausen. Kuetzing, F.T. 1859. Tabulae Phycologtcae, vol. 9. Nordhausen. Kuetzing, F.T. 1863. Tabulae Phycologtcae, vol. 14. Nordhausen. Kuetzing, F.T. 1865. Tabulae Phyecologtcae, vol. 15. Nordhausen. Kuetzing, F.T. 1868. Tabulae Phyeologtcae, vol. 18. Nordhausen. Kuetzing, F.T. 1869. Tabulae Phycologtcae, vol. 19. Nordhausen. Kylin, H. 1931. Die Florideenordnung Rhodymeniales. Acta Univ. Lund 27(11), 1-48, pls. 1-20. Kylin, H. 1932. Die Florideenordnung Gigartinales. Acta Univ. Lund 28(8), 1-88, pls. 1-28. Kylin, H. 1956. Dte Gattungen der Rhodophyceen. Gleerups, Lund. Levring, T. 1953. The marine algae of Australia. I. Rhodophyta: Goniotrichales, Bangiales and Nemalionales. Ark. Bot. Ser. 2, 2(6): 457-530. Lucas, A.H.S. 1927. Notes on Australian marine algae, V. Proc. Linn. Soc. N.S.W. 52, 555-562, pls. 41-48. Lucas, A.H.S. 1931. The marine algae hitherto recorded from north-east Australia. Rep. Gt Barrier Reef Comm. 3, 47-57. Lucas, A.H.S. 1935. The marine algae of Lord Howe Island. Proc. Linn. Soe. N.S.W. 60, 194-232, pls. 5-9. May, V. 1948. The algal genus Gracilaria in Australia. C.S.I.R. Bull. 235, 1-64, pls. 1-15. May, V. 1951. The marine algae of Brampton Island, Great Barrier Reef, off Mackay, Queensland. Proce. Linn. Soc. N.S.W. 76, 88-104. Misra, J.N. 1966. Phaeophyceae in India. Indian Council Agricultural Research, New Delhi. 26 Nakamura, Y. 1965. Species of the genera Ceraniwn and Campylaephora especially those of Northern Japan. Set. Pap. Inst. Algol. Res. Hokkatdo Univ. 5, 119-180, pls. 1-14. Ohmi, H. 1958. The species of Gracilaria and Gractlartopsts from Japan and adjacent waters. Mem. Fac. Fish. Hokkaido Univ. 6, 1-66, pls. 1-10. Okamura, K. 1907a. Icones of Japanese Algae 1(2), 23-49, pls. 6-10. Tokyo. Okamura, K. 1907b. Icones of Japanese Algae 1(4), 65-92, pls. 16-20. Tokyo. Okamura, K. 1909. Icones of Japanese Algae 2(1), 1-20, pls. 51-55. Tokyo. Okamura, K. 1910. Icones of Japanese Algae 2(6), 89-108, pls. 76-80. Tokyo. Okamura, K. 1912. Icones of Japanese Algae 2(9), 143-165, pls. 91-95. Tokyo. Okamura, K. 1915. Iecones of Japanese Algae 3(4), 55-77, pls. 116-120. Tokyo. Okamura, K. 1929. Icones of Japanese Algae 6(1), 1-8, pls. 251-255. Tokyo. Okamura, K. 1930. Icones of Japanese Algae 6(3), 19-27, pls. 261-265. Tokyo. Okamura, K. 1942. Icones of Japanese Algae 7(10), 81-116. Tokyo. Papenfuss, G.F. 1943. Notes on algal nomenclature. II. Gymnosorus J. Agardh. Amer. J. Bot. 30, 463-468. Papenfuss, G.F. 1951. Phaeophyta. In G.M. Smith (ed.), Manual of Phycology, pp- 119-158. Waltham, Mass. Papenfuss, G.F. 1966. Notes on algal nomenclature. V. Various Chlorophyceae and Rhodophyceae. Phykos 5, 95-105. Papenfuss, G.F. and Y-M, Chiang. 1969. Remarks on the taxonomy of Galaxaura (Nemaliales, Chaetangiaceae). Proc. Intl. Seaweed Symp. 6, 303-314. Papenfuss, G.F. and T. Edelstein. 1974. The morphology and taxonomy of the red alga Sarconema (Gigartinales: Solieriaceae). Phyeologia 13, 31-44. Papenfuss, G.F. and J.B. Jensen. 1967. The morphology, taxonomy, and nomenclature of Cystophyllwn trinode (Forssk$1) J. Agardh and Cystosetra myrica (S.G. Gmelin) C. Agardh (Fucales: Cystoseiraceae). Blumea 15, 17-24, figs. 1-4. Pham-hoang Ho. 1969. Marine algae of South Vietnam. Study Centre, Saigon. (In Vietnamese). Post, E. 1936. Systematische und pflanzengeographische Notizen zur Bostrychta-Caloglossa-Assoziation. Rev. Algol. 9, 1-84. 27 Post, E. 1943. Zur Morphologie und Okologie von Caloglossa. Arch. Protistenk. 96, 123-220. Post, E. 1955. Weitere Daten zur Verbreitung des Bostrychietum IV. Arch. Prottstenk. 100, 351-377. Post, E. 1957. Weitere Daten zur Verbreitung des Bostrychietum VI. Arch. Prottstenk. 102, 84-112. Price, Ian R., A.W.D. Larkum and A. Bailey. 1976. Check list of marine benthic plants collected in the Lizard Island area. Aust. J. Plant Phystol. 3, 3-8. Queensland Department of Harbours and Marine. 1977. Offictal Tide Tables for the Coast of Queensland and Notes on Boating. Queensland Government Printer, Brisbane. Saenger, P. 1973. Additions and comments on the Rhodomelaceae of Inhaca Island, Mocgambique. Wova Hedwtgta 24, 19-31, figs. 1-14. Saito, Y. 1967. Studies on Japanese species of Laurencta, with special reference to their comparative morphology. Mem. Fac. Fish. Hokkaido Untv. 15, 1-81, pls. 1-18. Saito, Y. 1969. The algal genus Laurencta from the Hawaiian Islands, the Philippine Islands and adjacent areas. Pacif. Sez. 23, 148-160. Saito, Y. and H.B.S. Womersley. 1974. The southern Australian species of Laurencta (Ceramiales: Rhodophyta). Aust. J. Bot. 22, 815-874, figs. 1-27. Sakai, Y. 1964. The species of Cladophora from Japan and its vicinity. Set. Pap. Inst. Algol. Res. Hokkaido Univ. 5, 1-104, pls. 1-17. Segawa, S. 1968. Coloured illustrations of the Seaweeds of Japan. Osaka. (In Japanese). Setchell, W.A. and N.L. Gardner. 1924. The marine algae. Expedition of the California Academy of Sciences to the Gulf of California in 1921. Proc. Caltf. Acad. Sct. (4th Ser.) 12, 695-949, pls. 12-88, map. Sonder, 0.G. 1871. Die Algen des tropischen Australiens. Abh. naturw. Ver. Hamburg 5, 33-74, pls. 1-6. Sreenivasa Rao, P. 1970. Systematics of Indian Gelidiales. Phykos 9, 63-78, pls. 1-2. Tanaka, T. 1941. The genus Hypnea from Japan. Set. Pap. Inst. Alg. Res. Hokkaido Univ. 2, 227-250, pls. 53-4. 28 Taylor, W.R. 1928. The marine algae of Florida, with special reference to the Dry Tortugas. Pap. Tortugas Lab. 25, 1-219, pls. 1-37. Taylor, W.R. 1950. Plants of Bikint and other Northern Marshall Islands. Univ. Michigan Press, Ann Arbor. Taylor, W.R. 1960. Marine Algae of the Eastern Troptecal and Sub- tropteal Coasts of the Americas. Univ. Michigan Press, Ann Arbor. Tseng, C.K. 1942. Marine algae of Hong Kong, II. The genus Catenella. J. Washington Acad. Set. 32, 142-146. Tseng, C.K. 1943a. Marine algae of Hong Kong, III. The genus Bostrychia. Pap. Mich. Acad. Set., Arts Lett. 28, 165-183, pills. L=3. Tseng, C.K. 1943b. Marine algae of Hong Kong, IV. The genus Laurencia. Pap. Mich. Acad. Set., Arts Lett. 28, 185-208, pls. 1-4. Tseng, C.K. 1944. Marine algae of Hong Kong, VI. The genus Polystphonia. Pap. Mich. Acad. Set., Arts Lett. 29, 67-82, pls. 1-4. Tseng, C.K. 1945. New and unrecorded marine algae of Hong Kong. Pap. Mich. Acad. Scei., Arts Lett. 30, 157-171, pls. 1-2. Umezaki, I. 1961. The marine blue-green algae of Japan. Mem. Coll. Agric. Kyoto Univ. 83, 1-149. Valet, G. 1969. Contributions 4 1'étude des Dasycladales, 2-3. Nova Hedwigta 17, 551-644, pls. 133-162. Weber-van Bosse, A. 1898. Monographie des Caulerpes. Ann. Jard. bot. Buttenzorg 15, 243-401, pls. 20-34. Weber-van Bosse, A. 1904. Corallineae verae of the Malay Archipelago. Stboga Exped. Monogr. 61, 78-110, pls. 14-16. Weber-van Bosse, A. 1913. Marine Algae, Rhodophyceae, of the "Sealark* Expedition, collected by Mr. J. Stanley Gardiner, M.A. Trans. Linn. Soc. Lond. (Ser. 2. Bot) 8, 105-142, pls. 12-14. Weber-van Bosse, A. 1921. Liste des algues du Siboga. II. Rhodophyceae. Pt. 1. Protoflorideae, Nemalionales, Cryptonemiales. Stboga Exped. Monogr. 59b, 187-310, pls. 6-8. Weber-van Bosse, A. 1923. Liste des algues du Siboga. III. Rhodophyceae. Pt. 2. Ceramiales. Stboga Exped. Monogr. 59c, 311-392, pls. 9=10. 29 Weber-van Bosse, A. 1928. Liste des algues du Siboga. IV. Rhodophyceae. Pt. 3. Gigartinales et Rhodymeniales. Stbhoga Exped. Monogr. 59d, 393-483. Womersley, H.B.S. 1958. Marine algae from Arnhem Land, North Australia. In R.L. Specht and C.P. Mountford (eds.), Rec. Amer.-Aust. Scetent. Exped. Arnhem Ld., Vol. 3, pp. 139-161. Melbourne Univ. Press, Melbourne. Womersley, H.B.S. 1967. A critical survey of the marine algae of southern Australia. II. Phaeophyta. Aust. J. Bot. 15, 189-270. Womersley, H.B.S. and A. Bailey. 1970. Marine algae of the Solomon Islands. Phil. Trans. Roy. Soc. London B259: 257-352, pls. 24-27. Yamada, Y. 1931. Notes on Laurencita, with special reference to the Japanese species. Univ. Calif. Publs. Bot. 16, 185-310, pls. 1-30. Yamada, Y. 1934. The marine Chlorophyceae from Ryukyu, especially from the vicinity of Nawa. J. Fac. Set. Hokkatdo Univ. (Ser. 5) 3, 33-88. Yamada, Y. 1950. A list of marine algae from Ryukyusho, Formosa. I. Chlorophyceae and Phaeophyceae. Sct. Pap. Inst. Alg. Res. Hokkatdo Untv. 3, 173-194. i ee ee Tavior. W.8, J0P8.. chomeseboneiaberse . ‘ _ 7 in , . ce fn pe FURS A er deh) 3° at ond Veg i yer, a: 5, 1-4 C£oe-FPE , oC? eQantet G J 1. d22 OG bak wei nes bod Seeee seer” SeaRert CA , (68s) Sees capor rs tb $4908? .3,0 nt aT Ie sreth speed .abIOe Jem be vice FON) Ke OPT) | Ola aires @** ¥re es ors wav mo A oe! 2.808 cbleviend | web«a 7. 2, : ae 428 } khndopiy ear Se he ca he em s3e, 3)1-392, gota Ona } ATOLL RESEARCH BULLETIN NO. 2338 BLUE-GREEN ALGAE (CYANOBACTERIA) OF THE OCEANIC COAST OF ALDABRA by B.A. Whitton and M. Potts Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 A (AAR O-DLIG ws, CI0O ART s “4 4 bomen “CN USIP AAO ie ANT A213 OG worgetdan CvCl 2 < 4 um, > 4 < 8 um wide etc. In the floristic list, records are given first for the actual categories used, and then the most appropriate binomial added. RESULTS Rocks in the supralittoral and upper littoral Blue-green algae are present on almost all rock surfaces in the supralittoral and the upper part of the littoral. The nearer a true terrestrial environment, the more frequent is Tolypothrix byssoidea, the thinner any continuous endolith layer and the more scattered any true chasmolithic growths. Below the zone of T. byssoidea, but above the barnacle zone, the surface of most rocks may appear steel-blue, grey or almost black. The paler colours are due to Hyella balani, which grows both at the surface and endolithically. A bright green layer of chasmoliths is sometimes present, composed of forms resembling Pleurocapsa. In the same zone (below Tolypothrix byssoidea and above barnacles) Scytonema sp. forms small olive-green tufts in depressions in the champignon; it was found at every location studied, usually with many epiphytes e.g. Xenococcus spp., Dermocarpa spp. Slightly lower down the shore Calothrix and Rivularia are widespread on the north coast, and Isactis and Rivularia on the south coast. Areas of beachrock on Ile Picard (near the settlement) were studied more intensively. The rock is smooth, rounded and slopes towards the sea at an angle of about 10-15°. It has a blue-grey colour due to Hyella balani, and when fractured, a chasmolithic layer about 2 mm thick is visible at a depth of about 3-5 mn. Cells resembling Pleurocapsa and fungal hyphae are also present. In small cracks and crevices slightly further down the shore, but still holding water after the tide has receded, accumulated sand supports thin films of Lyngbya martensiana and Schizothrix calcicola, with some Nodularia harveyana. On passing down the shore from the blue-grey rocks, the colour changes to brown at about the level that barnacles appear. The brown colour is due to the sheaths of Calothrix crustacea and C. contarenii, which form a thin epilithic layer over Hyella balani. After a period of particularly high tides and stormy seas, an extensive area of supralittoral beachrock became exposed due to the removal of sand deposits several metres thick which had been present for at least six months (and probably longer) . The newly exposed rock appeared pale yellow, lacking the characteristic steel-blue colour of adjacent rocks. Examination of the surface of the beachrock showed no Obvious blue-green algal communities. Patches of pale blue were 4 just discernable after 5 days, while the whole rock had light steel-blue colour within 2 weeks. Visually the rock was indistinguishable from other rocks after 5 weeks, but the green chasmolithic layer had not yet appeared. Several species were recorded only in particular parts of the atoll, although this may well be simply a consequence of insufficient sampling elsewhere. Scytonema endolithicum grows in the upper supralittoral of {le Picard. The closely appressed filaments bore as much as 1200 um into the rock; the cells are bright blue-green and the sheaths yellow- brown. When chips of rock with this alga were placed on agar (high CaCO3 — seawater medium), filaments also developed above the surface, growing vertically, in small tufts, and reaching heights of 1000 — 2000 um. At Dune d'Messe, inspection of black rocks showed that lichens are frequent higher up, while the endoliths Brachytrichia sp. and Solentia stratosa become more frequent lower down. A zone of lighter coloured rocks below these black rocks showed Mastigocoleus testarum to be the dominant endolith here. Rocks elsewhere on the south coast (e.g. Dune Jean Louis) showed a similar vertical colour zonation, with a blue-black zone above a paler one. Brachytrichia sp., Solentia stratosa and Mastigocoleus testarum, but apparently not Scytonema endolithicum, are all widespread on cliffs of the lagoon shore. Reef-flat and reef-ridge on Tle Picard In general it appears that the longer rocks are covered by water each tidal cycle, the less conspicuous are the blue-green algae; however small areas of Lyngbya or colonies of Calothrix crustacea are not uncommon both in the area of the reef-flat and on the reef-ridge. Occasional patches of blue-green algae occur on the sediments between the shore and reef-ridge on fle Picard. These consist usually either of Nodularia or mixed populations of Lyngbya and other Oscillatoriaceae. Blue-green algal patches are apparently most frequent on sediments near the research station, in the same region that pink colourations due to phototrophic bacteria occur at some spring tides (Potts & Whitton, 1979b; in press). The Oscillatoriaceae here include Spirulina. Sublittoral Visually obvious growths of blue-green algae are only occasional, or even rare, in the upper 10 m, but become slightly more frequent at greater depths (down to 45 m, the maximum depth surveyed). The great majority of samples consist largely or entirely of Lyngbya, apart from the endolith Plectonema terebrans and occasional epiphytes, but Calothrix films also accur in the upper 10 m. Lyngbya trichomes show a wide range of widths, from about 5 to 40 um, with no easy separation into distinct species, though many fall into the size range of L. sordida «(14 — 3) 1m). Conspicuous growths of the broadest forms (L. majuscula) were found only down to 20 nm. Trichomes range in colour from green through olive to pink. Down to about 20 m examples occur of all colours, but by 40 m all are pink. Some of the best developed growths of L. sordida at 40 m were found to cover the surface of a sponge, itself red in colour (when viewed in daylight). A particularly interesting population of L. sordida was found at this depth off Tle Malabar, not far from Passe Gionnet. The alga formed branched structures attached at various points to dead coral and a dead clam; in at least some cases the alga formed distinct tubes. This population was closely associated with several shrimps, which apparently live inside the alga structure. LIST OF SPECIES poe Equivalent binomial computer Category used for records Uhereuapiticabile number 010801 Brachytrichia sp. 010918 Calothrix contarenii Bornet et Flahault 010919 C. crustacea Thuret 011560 Chroococcus, > 8 < 16 um, sheath C. turicensis (Nageli) not striated Hansgirg 018201 Calmatella buaensis Ercegovi¢ 012005 Dermocarpa leibleiniae (Reinsch) Bornet et Thuret 012006 D. olivacea (Reinsch) Tilden 012008 D. sphaerica Setchell et Gardner 012009 D. minima Geitler 012050 Dermocarpa sp. 012201 Entophysalis granulosa Kutitz. 013403 Hormathonema violaceo-nigrum Ercegovié 013604 Hyella balani Lehmann 013605 H. tenuior Ercegovié 018750 Isactis sp. 014204 Lyngbya confervoides Ag. 014205 L. digueti Gomont 014206 L. epiphytica Hieronymus 014211 *L. martensiana Menegh. ex Gomont 014212 L. norgardii Wille 014219 L. majuscula Harvey (014238) L. sordida (Zanard.) Gom. 014501 Mastigocoleus testarum Lagerheim 014801 Microcoleus chthonoplastes Thuret ex Gomont 014903 Microcystis reinboldii (Richter) Forti 015101 Nodularia harveyana Thuret 015103 N. spurnigeria Mertens 015814 Plectonema terebrans Bornet et Flahault 015932 Pleurocapsa > 4 < 8 um P. fuliginosa Hauck 015933 Pleurocapsa > 8 < 16 um P. crepidinum Collins 015934 Pleurocapsa > 16 um O1L6101 Pseudanabaena catenata Lauterborn 6 Durham Equivalent binomial computer Category used for records where applieenie : number 016572 Rivularia sp. D 016602 Schizothrix arenaria (Berk.) Gomont 016604 S. calcicola (Ag.) Gomont 016732 Scytonema endolithicum Ercegovié 016754 Scytonema > 16 um Scytonema sp. 018602 Solentia stratosa Ercegovié 016901 Spirulina subsalsa Oersted 016951 S. subtilissima Kitz. 017602 Tolypothrix byssoidea (Berk.) Kirchner 018401 Trichodesmium erythraeum Ehrenberg ex Gomont 018402 T. thiebautii Gomont 018052 Xenococcus > 2 < 4 um X. laysanensis Lemmermann 018053 Xenococcus > 4 < 6 um X. kerneri Hansgirg 018054 Xenococcus > 6 < 8 um X. schousboei Thuret *Lyngbya martensiana may perhaps be a misidentification of L. semiplena Ag. The latter, in contrast to the former possesses a calyptra; L. semiplena has been reported elsewhere from marine environments much more often than L. martensiana. DISCUSSION Blue-green algae are equally abundant in the supralittoral of oceanic and lagoon shores of Aldabra, but there is a marked contrast in their behaviour further down the shore. On the oceanic side they become less frequent the longer surfaces are covered by water during each tidal cycle; this contrasts with the lagoon where large areas of the lower littoral are covered by mats of Microcoleus chthonoplastes (Potts & Whitton, in press). Of all regions on the atoll with a dense photosynthetic cover, the uppermost 15 m of the sublittoral probably has fewer obvious growths of blue-green algae than any other. As ropes and buoys often developed Lyngbya tufts rapidly, it seems likely that the intensive grazing by animals is the main reason for the poor development of blue-green algae in the upper part of the coral zone. The importance of grazing by fish in suppressing algal growth on a shallow reef has been documented in some detail for Curagao by Wanders (1977) and van den Hoek et al. (1978). Too few observations have been made of the sublittoral in the lagoon to compare it with the ocean. Deeper in the ocean attached blue-green algae become slightly more frequent, in some cases being closely associated with a red sponge or with shrimps. The latter is apparently a similar association to that recorded by Cowles (1913) who summarized earlier data, Taylor (1950) for the Marshall Islands and Newhouse (1954) for Raroia. The previous records are all from shallow waters, but otherwise the present association appears very similar. Cowles described the blue-green alga as Plectonema, but Newhouse identified his material as Lyngbya sordida, the same species as recognized for Aldabra. ACKNOWLEDGEMENTS The study was made possible by a grant from the Natural Environment Research Council, and by the support of the Aldabra Research Committee and Messrs D.J.H. Griffin and L.U. Mole of the Royal Society. The facilities of the Aldabra Research Station proved invaluable during the two more recent study periods on the atoll. SCUBA diving to depths greater than 15 m was carried out by Dr D.J. Jones. Prof S. Golubié provided useful discussion on identification of endoliths. REFERENCES Braithwaite, C.J.R. 1975. Petrology of palaeosols and other terrestrial sediments on Aldabra, Western Indian Ocean. Phil Trans. R. Soc. B- B/s39 dos. Braithwaite, C.J.R., Taylor, J.D. and Kennedy, W.J. 1973. The evolution of an atoll: the depositional and erosional history of Aldabra. Phil. Trans. R. Soc. B. 266: 307-340. Cowles, R.P. 1913. The habits of some tropical Crustacea. Philippine J. Sci. 8: 119-124. Donaldson, A. and Whitton, B.A. 1977. Algal flora of freshwater habitats on Aldabra. Atoll Research Bulletin 215: 1-26. Frémy, P. 1933. Cyanophycées des cétes d'Europe. Mém. Soc. nat. Sci. nat. math. Cherbourg 41, 235 pp. + 66 plates. Hughes, R.N. and Gamble, J.C. 1977. A quantitative survey of the biota of intertidal soft substrata on Aldabra Atoll, Indian Ocean. Phil. Trans. R. Soc. B 279: 327-355. le Campion-Alsumard, T. 1969. Contribution a 1'étude des cyanophycées lithophytes des 6tages supralittoral et mediolittoral (Région de Marseille). Théthys 1: 119-172. Lewis, J.R. 1964. The Ecology of Rocky Shores, 420 pp. English U.P., London. Moul, E.T. 1957. Preliminary report on the flora of Onotoa Atoll, Bilter Islands. Atoll Res. Bull. 57: 1-48. Newhouse, J. 1954. Notes on Myxophyta of Raroia. Atoll Res. Bull. 33: 45-58. Potts, M. and Whitton, B.A. 1979a. pH and Eh on Aldabra. dn Comparison Of marine and freshwater environments. Hydrobiologia Potts, M. and Whitton, B.A. 1979b. pH and Eh on Aldabra Atoll. 2 Intertidal photosynehetic microbial communities showing zonation. Hydrobiologia 8 Potts, M. and Whitton, B.A. in press. Vegetation of the intertidal zone of the lagoon of Aldabra, with particular reference to the photosynthetic prokaryotic communities. Price; Ji-HiemL Ove The shallow sublittoral marine ecology of Aldabra. Phil. Lransat(R. Soc. By eZeOcwe 2S —Iy//2F Stoddart, D.R. and Mole, L.U. 1977. Climate of Aldabra Atoll. Atoll Res. Bull. 202: 1-27. Taylor; 1d «DEeLOTAS. Intertidal zonation of Aldabra Atoll. Piaseles Transiehe oO aE = 92602 SLU Taylor, J.D. and Way, K. 1976. Erosive activities of chitons at Aldabra Atoll. J..sed. Petrol. 46: 974-977. Taylor, W.R. 1950. Plants of Bikini and other Northern Marshall Islands. University of Michigan Press, Ann Arbor, U.S.A. apaorlestilils Grete allisi7Aéye The marine erosion of limestones on Aldabra Atoll, Indian Ocean. Z. Geomorph. N.F., Suppl. 26: 164-200. van den Hoek, C., Breeman, A.M., Bak, R.P.M. and van Buurt, C. 1978. The distribution of algae, corals and gorgonians in relation to depth, light attenuation, water movement and grazing pressure in the fringing coral reef of Curagao, Netherlands Antilles. Aquatic Bot. 5: 1-6. Wanders, J.B.W. 1977. The role of benthic algae in the shallow reef of Curagao (Netherlands Antilles) III: the significance of grazing. Aquatic. Bot .13:°%13857-390. Westoll, T.S. and Stoddart, D.R. 1971. A discussion of the results of the Royal Society expedition to Aldabra 1967-68. Phil. Trans R. Soc. B 260: 1-654. Whitton, B.A. 1971. Terrestrial and Freshwater algae of Aldabra. Phil. Trans. R. Soc. B 260: 249-255. Womersley, H.B.S. and Edmonds, S.J. 1958. A general account of the intertidal ecology of South Australian coasts. Aust. J. mar. fresh. Res. 9:\ 217=260. ATOLL RESEARCH BULLETIN NO. 239 VEGETATION OF ALDABRA, A REASSESSMENT by R.J. Hnatiuk and L.F.H. Merton Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. November 1979 - eo . fokeu, M. and Winbton,; BAC in oeeee, Pup good Of the legace Of 2) dames @dith pPhotonynthetss © yoRnacyperenie remem | Lee | = >.> Petce, JA 73 re ev . tclere) secifie ree = i SiaaAlie Ato - =. , BN . ‘ CT} ere oe } : > bencal WiuTiev! yo poewrine ae 4.2.3 3Al notgeiden v°?) eodarseell VEGETATION OF ALDABRA, A REASSESSMENT by R. J. Hnatiuk! and L.F.H. Merton” INTRODUCTION Aldabra's vegetation has been described and communities or types of vegetation distinguished for a part or all of the atoll on several occasions (Fryer, 1911-12; Vesey-Fitzgerald, 1942; Stoddart & Wright, 1967; Stoddart 1968b; Fosberg, 1971; and Grubb, 1971). The classifications of Vesey-Fitzgerald and Stoddart & Wright are essentially modifications of that of Fryer (1911-12, p 414) who recognized four major units: Mangrove Swamp, Pemphis Bush, Varied or Open Bush, and Shore Zone (table 1). Vesey-Fitzgerald (1942, p 7) added a "Spray Zone Community" and a "Herb Mat Community" while Stoddart and Wright (1967, p 26) included "Man induced vegetation". Stoddart (1968b, Fig. 1) included "Casuarina" as a separate unit. He also made important distinctions within the shrub dominated communities by using what appears as a combination of floristic and geomorphologic criteria. However, in his descriptions of the "Platin with open woodland" he did not specifically note the occurrence of tortoise turf although reference to Dactyloctenium pilosum and Eragrostis sp. clearly indicates he was aware of its presence. Grubb (1971, p 351) produced a detailed map of the east end of Grande Terre showing 10 "vegetation types" as distinguished on aerial photos with some ground control, but since he includes units called "champignon" and "mud flats", he is concerned with "habitats" and not only "vegetation". However, he has come closer to understanding the complexity of the scrub types than many of the other workers have. Fosberg's (1971) classification is by far the most detailed to date, recognizing 63 vegetation types almost exclusively on the basis of vegetation and floristic criteria. Only in the distinction of "swamp" does he overtly use habitat data in his definition of a type (Fosberg Ithe Royal Society Aldabra Research Station, current address: Western Australian Herbarium, Dept. of Agriculture, George Street, South Perth 6151, Australia. 2Deceased; formerly of the Botany Department, The University, Sheffield, England. (Manuscript received January 1977-- Eds.) (SMopeseu suTIeu pebrewaqns) so‘9z spuejs wnyoTzsoz2y (fT) 60 Tebte (ST) €9'Th' LZ SJeTSI AqTunuuop uoobey JO SMOopeeW SNosoeqzeH (ZT) 3eW GI9eH suepzep §(€T) Sa9AoID YnuoDs0D (Z) uoT7e e658, seAaorIy PuTTeNseD (T) Or‘8z‘Pz peonput—uewn seyoreg jo uot je jebaa AVeuCTG (TT) qnaos ound JanL Te#SeoD (pueTPOOM set 3 tunuoS snoTuTbITA snToqozods (g) TeySeOD) euoz Aeads pueTssez) y~oossuL 79‘T9‘09'6S‘LS Sp2eMS suoTzetoossy suoTzeToossy uoTARZOepozeToOs (L/L) "eep’67'72Z‘PT‘OT snjToqoizods snoTtydoumes gq sno TtTydoumesq auoz ez0ys LS‘9S'0S Spies FANL SSTORTOL (6) ‘IS‘OS‘Sb‘ pp‘ ce abpas sse19 * SMOPPOW obnTTOw-edoseg (OT) qnaog saeproeutndod 8S‘EG UOoTQeIebeaA ersedsayy—e19z3 TuUuNT) ‘eGtipemedog "9 [ood seaormH euTTenseD (T) OS‘SP‘ED pueTPOoM icv 6€ *Peacseos TeqzseoD “Gc‘ezc‘ET’Z2T‘TL 4sexog qnzos ysng uedo Qna9g PpexTW (9) fGtaee oo ere Z qniosg pextw qniog pexTtW qnazosg pexTtW 3 poeTtazaeprA ysng qnz9S *‘d (€) 9n' TZ qnaz5sg *d FexOTUL “d FexOTUL “d sTyduad 8b‘ Lp‘ BE ‘9E'SE durems durems uot}eER7ebea “W (Pf) OT Bl Z Tt OLASt eTuUaoTAY soetytunumop -W SeTyTuNUIOD “W eaorbuen (zeded sty) (TL6T) (TL6T) (L96T) (Zv6T) NOLYSW 8 ANILYNH Sugdso4i q@aned QLHOIYM 8 LYWadoOLs ATWHADZLIA-AYSAA (TT6T) WaAN *ezqepTw JO uoTRZeJebean ayQ JO SUOTRJeOTJTSSeTO JO uosTzeduios y “T eTqaeL 1967, p 120). Four vegetation maps, Or maps which at least portray vegetation units, for the whole of the atoll are known to exist (Baker, 1963, Stoddart 1968a, b, and Directorate of Overseas Surveys D.O.S. 6001 Vegetation Overlay, 1969). All maps appear to be drawn from an interpretation of aerial photographs taken in 1960. Baker's map was primarily intended as a geological map with some vegetation notes appended and thus criticism of his vegetation notes need to be tempered with this in mind. Both the D.O.S. map and Baker's suffer from what appears to be lack or insufficient use of ground control. Because of certain errors in photo-interpretation found on these maps caution is advised in attempting to use them. Stoddart's 1968b map clearly shows the distribution of the major landforms and some vegetation units for Aldabra. The only major unit missing on the scale of his map is dense shrub on medium champignon. Fosberg has applied the classification scheme which he devised for the International Biological Program (Fosberg, 1967). It isa general purpose scheme that is able to be applied in many places of the world. It uses a heirarchical structure so that, for instance, his 63 ultimate units on Aldabra may be grouped into 24 units at the next highest and six units at the top of the classification. With so much work already done on the vegetation of the 155 square kilometres of dry land on Aldabra, why is yet another study necessary? The reason is, that despite the past work, there is still only a rudimentary understanding of the atoll's vegetation. The earliest works gave the basic outline of much of the vegetation but did not specifically consider several important units such as the "tortoise tur£" (Grubb 1971, p 359) or "mixed orthophyll ‘tortoise pastures'" (Fosberg 1971, p 223). Grubb's work was not intended to be comprehensive for the atoll, and certain specific problems, treated below, have limited the use of Fosberg's classification. It is the objective of the current work to present both a comprehensive classification and a synthesis of Aldabra's vegetation such as has not been possible before. Before discussing our classification it is instructive to make a closer examination of Fosberg's detailed classification (1971). His primary subdivision criterion of "spacing" of plants, and his ultimate criterion of "floristic" composition have made it difficult to use on Aldabra. Of lesser importance but still contributing to the overall difficulty of use are the criteria of leaf size, and evergreenness. These latter two criteria have been found to be dependant upon the amount of water available during a particular year or growing season. Thus the plants on one piece of ground may appear "evergreen" and "mesophyllous" one year but the same plants deciduous and "microphyllous" another year, because of the unpredictable and very variable climatic regime found on Aldabra (see Stoddart & Mole, 1977). Such a characteristic is undesirable in a vegetation classification and the solution(of Asprey & Loveless 1958) to a similar problem encountered in Jamaica, that of using the phenological state prevailing under "average" x lo 1 a aa © am mol ‘ b * wae * (peysttqndun) Set yTunumi0o asayq uo pexATOM sey weyazangq *Adeq Auejog ‘uospTeuod *w ostTe ‘pz7z d ‘TLET Hreqsog ees ynq peTpn4s 98734TT) nUOTIeIeHaa aefhTe Hbutzoq pue wTty uesofkydourdy,, TWOTW suepze5 = (€T) (peonput ueul) AATunumo0D snoaseqray 3 APOoM pexTwW SQ2TSI ucobey uo SMOpeeW SNosoeqzeH (ZT) soyoreg JO uoTRZeZeheQ ABVBUCTd (TT) smopeew ObnTTOW-edooeg (OT) (dnoz5 satoads) sqzeyq jootd JaANL estozIOL (6) Jueutuiog sdnorz5-satoeds gan, TezseoD snorurbira snToqorods (g) puetsseir5 yOossn], uOTAROepoIeTIS (/) sjueutTwog setoeds-aTbuts ASE sqieH 3OOOUCHW SnogowdaaH squetTd Adoueos uodn peseq erzqepTW AOF uoTReOTJTSSeTO uoTWIeIOH|A VW qniosg uoTAxozepts -*o qnzos yoreeg “q qniog eunq ‘e isqueTzeA qna9S pexTW (9) qnazog seproautndod ersedseyy—-erazztTuunT (G) uotqze3e6e, eAozbuen (fp) jueutuog sdnozrz5-setoeds qnzos sTydueq (€) saaory euTZenseD (Zz) saaorzy ynuoD0D S(T) queutuog satoeads-aThuts AGOOM “¢ eTaeL 5 conditions is not feasible because the "average" condition on Aldabra is not known (Hnatiuk & Merton, 1979). The use of spacing of plants as the primary basis of division in classification (Fosberg, 1967, p 69) may appear at first sight to be a straightforward description of a feature of the plants, independent of the surrounding physical environment. "Closed" vegetation poses little problem but the "open" and "sparse" spacing groups used by Fosberg (1967, 1971) include space which on Aldabra is not part of the individual plants making up the vegetation. The unoccupied space between plants is part of the physical environment that Fosberg wishes to exclude from the classification criteria. Such a distinction may appear as "hair splitting" but two fundamentally different situations are being grouped under Fosberg's application of his "spacing" criteria. To take the extreme case of sparse vegetation in which there is more than twice the individual plant diameter between plants (Fosberg 1967, p 79), on the one hand plant cover may be sparse but the actual place where each plant grows in the sparsely vegetated area is usually assumed to be independent of the substrate (i.e. the plants could grow at any site within the area). On the other hand plant cover may be sparse because the sites where the plants can potentially grow are themselves sparsely distributed in a matrix of uninhabitable space. If plant cover is "closed" in the habitable areas then the vegetation should be seen as closed even though the "habitat" is open. The situation is somewhat analagous to plants growing in soil in pots on a bench. The spacing of plants is then easily seen as a two-level characteristic dependent upon firstly how close together plants can grow in a pot and secondly upon how closely spaced are the pots. We believe that it is the first of these levels of spacing which should be considered on Aldabra where this two-level distinction is to be found, for example at the east end of Grande Terre and along much of the south coast of Grande Terre inland of the "8-metre ridge". Floristic data have often been effectively used to define vegetation units because the presence or abundance of particular species or species groups are known to be good indicators of environmental conditions. Studies of species distribution patterns on Aldabra (S. & R. Hnatiuk, unpublished) have suggested that colonization patterns may be a prominent feature of the atoll's plants, and these patterns are made complex through the interaction of developing interspecific and species-environment interactions leading to establishment of species niches. Fosberg (1971) has been generally cautious in his use of floristics as criteria for classification, but even so he has used such units as Acalypha scrub, Guettarda scrub, and Scaevola scrub that subsequent observations have found to be too heterogeneous to be called single types. Fosberg's classification may thus indicate the potential vegetation that could develop on Aldabra, but it does not adequately portray the present vegetation. Our classification has been based upon features of the canopy plants only, understorey plants are not considered in the classification but are noted in the descriptions. Plant nomenclature follows that of Fosberg and Renvoize (1979). (1) COCONUT GROVES Location: planted at many sandy places but particularly on the west coast of the atoll, on Ile Esprit and Ile Michel. Small groves can be found on other beaches. Vegetation: the coconut palm, Cocos nucifera is the dominant plant. It is evergreen and often reaches over 20 m, but also much less in poor sites. Casuarina is frequently found with and usually somewhat taller than the palms. The cover of the coconut palm canopy varies from greater than 75% in densely reproducing groves to less than 25% where planting was sparse. The largest plantation with over 1000 trees is found at Settlement while small plantings of less than a dozen trees can be found at Grande Cavalier. The distribution patterns of mature plants often reflects the ordered rows of the original planting. The understorey is variable depending somewhat upon the degree and recentness of maintenance of the groves. Thus the better maintained groves tend to have a grass understorey about 5-20 cm high. The grass species are usually either Eragrostis sub-aequiglumis, Paspalum distichum, or Sporobolus virginicus. The sedges Cyperus ligularis, Cyperus niveus, and Fimbristylis cymosa may also occur in locally dense patches. In some groves there may be much self seeding of coconuts resulting in very dense groves as on parts of Ile Michel. The Anse Mais grove is developing similarly (R.J.H. unpublished work). Species from the Mixed Scrub (6) are commonly found in the understorey of the neglected groves. Soil: The calcareous mineral arenaceous soil of Trudgill (1979), also called the Farquhar Series(Piggott 1968), is the most common soil found, but sand of the Shioya Series (Piggott 1968) may be found in parts of the Ile Esprit Grove. Phenology: Seasonal changes in the coconut trees are limited to a reduction in crown size and a slowed development of leaves, flowers and fruit during the dry season, at which time the understorey herbs also become brown and dry. Flowering of understorey plants is largely restricted to periods of sufficient rain. Notes: The coconuts are not thought to be native to Aldabra and primarily occur in areas where they were introduced by man, although now they have become naturalized in some areas. Whereas formerly, maintenance of the groves kept the understorey clear of scrub and coconut regeneration, now that maintenance has essentially lapsed, except for sections of the Settlement grove, scrub species can be found invading some parts of groves, while in other places the development of coconut thickets with litter accumulations exceeding 1.5 m deep is occurring. There are no indications as yet that the groves will spread beyond the areas of deep sandy soil, though they may be extending into a few formerly scrub-dominated areas with sandy soil on the west coast of Grande Terre. It appears from simple transplant studies with Carica papaya, that competition for water may be an important limiting factor for the vegetation of these plantations. (2) CASUARINA GROVES Location: Casuarina Groves are found primarily on the north and west sea coasts of Aldabra, and on some lagoon-facing sandy beaches. Vegetation: The canopy is evergreen and dominated by Casuarina equisetifolia that may reach more than 20 m above ground. The Casuarina may occur together with Cocos and the vegetation type may then be referred to as "Casuarina and Cocos Grove". The understorey of Casuarina Groves is various and can be Cocos, Mixed Scrub (tall or low), grass, or barren. Soil: Two soil types have been found under Casuarina: Farquhar Series (Piggot, 1968), and organic brown calcareous soils (Trudgill, 1979). Occasionally a shallow, rendzZina-like soil consisting largely of Organic matter with a small amount of sand over rock is all that is found. In most places, the soil under Casuarina has a thick (2-10 cm) layer of Casuarina 'needles'. Phenology: There is still some doubt about how Casuarina reached Aldabra. Fryer (1911, p 416) and Fosberg (1971, p 215) believe Casuarina to be definitely introduced by man, but Ridley (1930, p 316- 17) believes it to be widely distributed by the sea. Wickens (1979) gives further information. What ever may have been its method of introduction, it is now well established and apparently spreading (as are other species) as seen in the small, dense groves of young saplings and seedlings at the periphery of some groves (e.g. west end of Passe Houareau grove, and south of Anse Var). Young plants are to be seen growing up through Mixed Scrub on both sides of Passe Houareau. The effect of the overtopping of scrub by Casuarina is variable. In some instances it appears that the scrub may either die out (Fosberg 1971, p 216) or may continue to thrive apparently little affected by the Casuarina as for example just north of Settlement. In sOme areas Acalypha claoxyloides is found to form a dense understorey shrub stratum to Casuarina (e.g. between Anse Owen and Anse Grande Poche, north west end of Ile Polymnie, part of the grove on the west side of Passe Houareau) as also does Plumbago aphylla. We would repeat Fosberg's statement that this is a "good problem for an ecological investigation" (1971, p 216). Heavy infestations of Casuarina by the woolly coccid, Icerya seychellarum, are associated with reduction in photosynthetic canopy, reduced or halted growth, and, in small plants at least, death. (3) PEMPHTS SCRUB Location: Pemphis Scrub is one of the widespread and common vegetation types found on the atoll. Vegetation: Pemphis Scrub as defined here includes those areas where Pemphis acidula grows in pure or virtually pure stands, and excludes those areas where Pemphis is merely one of many scrub species — such latter areas being classed Mixed Scrub (6). Most of the Pemphis in 8 the Pemphis Scrub occurs as multi-stemmed plants although large, single- stemmed, tree-like Pemphis also occur. Other plant species are rare in this type but the most common include Vernonia grandis, Acalypha claoxyloides, and Scaevola taccada. Certain Mixed Scrub species are often found growing in a mosaic with Pemphis Scrub, and it appears from observations that two vegetation types are involved (see notes below and discussion on 'Mosaics'). The height of Pemphis Scrub ranges from about O.5 m to more than 6 m. Mature Pemphis growing in nearly pure stands less than 1 m high, is common along parts of the trade-wind-exposed, south coast of Grande Terre. There is some evidence (D. Lewis, pers. comm -)that such plants may be genotypically dwarfed and often prostrate. That they have only been found on the seaward edge of scrub may suffice to distinguish them as a sub-type, which is part of Fryer's "Shore Zone" (Table 1). The canopy of tall Pemphis is very deep, often reaching 4 to 5 m. The "surface" of the canopy is very irregular with narrow, conical branches that are leafy to near their bases. Foliage density does not appear to be exceptionally high and direct sunlight penetrates to the ground; however, there is virtually no understorey development. Soil: The soil appears very poorly developed. It may consist of "shallow organic soil" (Trudgill, 1979) over solid rock (and is thus a very shallow rendzina-like soil) or no soil may be visible at all. Surface feeding roots may be abundant in the small pockets of organic accumulation but all that is generally visible is the tops of roots that pass down into crevices in the otherwise barren rock. Phenology: Pemphis is evergreen, its stems form no resting buds, and it can grow throughout the year, producing leaves, flowers, and fruits. Growth may decline or cease in periods of severe drought, particularly in plants that do not appear to have direct contact with sea water. Leaf fall is stimulated by the onset of dry conditions and by heavy infestations of the woollycoccid, Icerya seychellarum. Notes: As noted by Fosberg (1971, p 221) past estimates of the abundance of Pemphis have been exaggerated. This is particularly noticeable in the maps of Baker (1963) and the Directorate of Overseas Surveys (Vegetation overlay 6001, 1:25,000, 1969) where Mixed Scrub and Pemphis Scrub have been misinterpreted from the aerial photographs (see notes under Mixed Scrub (6)). The correlation between the occurrence of Pemphis Scrub and very rough limestone (champignon) has been noted since the earliest reports, but caution is necessary in using this relationship. For example, Braithwaite et al., (1973, p 337) found a high correlation between 'Pemphis dominated scrub' on the D.O.S. vegetation map and the occurrence of ‘dissected areas of the Takamaka limestone’ and used this correlation in their extrapolation of geological boundaries from interpretation of vegetation as seen on aerial photographs. The geological map is probably a fair representation of the surface geology, but the designation "Pemphis dominated scrub" is very much in error in several large areas. In our experience it would appear that Pemphis is the dominant plant on sites where the surface is sufficiently dissected to allow salt water to penetrate into the rooting zone, and also to prevent any substantial amount of soil from accumulating at the surface. Areas such as this can extend over several square kilometres (e.g. parts of south west Grande Terre) but most of the Pemphis Scrub occurs in a mosaic with a subdivision of the Mixed Scrub. Pemphis Scrub occupies the low ground, while Mixed Scrub occupies the knolls that rise 0.5 to 1.0 m above the low ground. These knolls may be only 5 to 10 m across so that if a larger unit of area is taken as the basis of study, the distinction between these types disappears. However, it appears useful in understanding the vegetation to recognize the elements as separate vegetation types because there does not seem to be a dynamic, vegetational relationship between the two types — each occurs in its own distinct habitat. From circumstantial evidence, it seems that Pemphis may be able to utilize salt water as a water source: live roots are often seen extending into the tidally inundated portions of pot holes; the canopy is evergreen with somewhat succulent leaves; measurements show growth to be continuous throughout the dry months on sites apparently devoid of soil when other shrub species nearby have long lost their leaves. The dynamic status of Pemphis Scrub is not clear. Seedling and sapling size individuals are very rarely reported for Pemphis. As noted above, a few large, tree-like individuals with diameters of about O.3 m and extending above ground for 2 to 2.5 m before branching, can be found throughout the atoll, but most Pemphis occurs as a prolifically, low-branching shrub. Until more is known about this species, further interpretation of these observations would be premature. The autecology and population structure of Pemphis is another good problem for ecological research. (4) MANGROVE VEGETATION Location: around most of the lagoon coast including on many of the numerous islets in the lagoon, and occasionally at isolated, inland places. Vegetation: The four most common species are Avicennia marina, Bruguiera gymnorrhiza, Ceriops tagal, and Rhizophora mucronata. Three less abundant species of mangrove are Sonneratia alba, Xylocarpus granatum and X. moluccensis. The trees range in height from 1 m to greater than 10 m, and canopy cover ranges from less than 25% to virtually 100%. A zonation of mangrove species relative to distance inland or nearness to tidal streams is not obvious on Aldabra as it is in continental areas (Macnae, 1971). Without knowing more about the reason why the mangrove species are distributed as they are on Aldabra, division of this vegetation on floristic bases seems premature. Macnae recognized two variants on the basis of the height of the trees: "high forests' (greater than about 6 m) being found mostly on the north, west and east lagoon shores and only sporadically on the south, while 10 the second 'thickets and low forests' being 1 to 6 m high predominate on the south shore of the lagoon. Soil: ranges from light grey, silty marl with only stunted mangrove (e.g. Dune Jean Louis landing) to deep, highly organic muds with tall mangrove forests (e.g. Cinq Cases Creek) (Macnae, 1971). Phenology: The trees are evergreen and for Bruguiera, Ceriops and Rhizophora, flowers and fruits can be found at all times of year, but not necessarily on the same plant at all times. The cycle of events for an individual is not known. Avicennia appears to have a seasonal response and to be more synchronized for all populations than are the former three species. Generally it appears that Avicennia produces leaves during the wet season, and flowers and fruits during the dry season but more observations are needed. Notes: The mangroves, especially the tall ones, have been cut for timber at manyplaces and regeneration is abundant in some places and absent in others. Macnae (1971) comments on the relatively old trees on the north lagoon shore of Grande Terre where soils are poor and the trees, though stunted, have large root systems. The detailed dynamics of this vegetation are not known and would repay study. (5) LUMNITZERA-THESPESIA POPULNEOIDES SCRUB Location: appears restricted to the east end of the atoll, in and around summer-flooded basins. Vegetation: The major canopy species are Lumnitzera racemosa and Thespesia populneoides. Pandanus tectorius Occurs around some of the basins, forming large dense clumps. These three species commonly occur in pure stands but may be inter-mixed. Canopy cover ranges from 100% beneath closely growing shrubs to less than 25% where shrubs are few and branches not numerous. The canopy ranges in height from about 1.5 m to more than 5 m above ground. The understorey is generally barren of plants but tussocks of the sedge Fimbristylis ferruginea are common beneath breaks in the shrub canopy and are largely, but not exclusively restricted to these sites. Cyperus ligularis is also occasionally found here. Soil: The soil is classed as mineral sediments: SM in Trudgill's system (1979). The soil is light to dark grey, fine-textured, saline, and commonly contains mollusc shells. The surface may have a litter layer and organic matter appears well mixed into the upper layers. Water logging is common during the rainy season, whilst during the dry months, the water table may fall below soil surface allowing the surface soil to become quite dry and powdery. The extent of the drying is dependent upon the amount and distribution of rainfall in any particular year. deal Phenology: Lumnitzera and Pandanus are essentially evergreen but the quantity and size of leaves present appears to be much greater during the wet season than during the dry. Thespesia may lose most if not all leaves during prolonged dry season drought. Flowers can be found at most times of year on Lumnitzera but they become very scarce during drought periods and very abundant with the return of the rainy season. Flowers on Thespesia and Pandanus are largely confined to the late wet season although occasional exceptions do occur. Ripe fruits are common in late wet and early dry season, and become more scarce as the dry season advances. Notes: It appears that Lumnitzera occurs most abundantly and luxuriantly beside pools and in soils that are more saline than those favoured by Thespesia, while Pandanus is at its best in the least saline conditions. However, the species are often found intermixed at one and the same site. Although Pandanus often occurs near fresh water pools (cf. Stoddart & Wright 1967, p 27) it is found sufficientlyoften in equally large and luxuriant groves or as isolated individuals in locations far from such pools (e.g. Anse Var, coastal scrub east of Passe Gionnet, Passe Houareau camp, and Point Vacqua) that the general value of Pandanus as an indicator of fresh water pools or of a particular vegetation type is largely restricted to the east end of Grande Terre. The dynamic status of this vegetation type is not known. However, it appears to have experienced change as evidenced by the apparently uniform size of most individuals around a pool although different pools appear to have different populations. About 1.3 km south west of Takamaka Grove, conspicuously "two-aged" stands of Lumnitzera, as judged by shrub height and stem diameter, are to be found but the reasons for this structure are not known. (6) MIXED SCRUB Location: widespread throughout the atoll. Vegetation: No single taxon characterizes this community, but some of the most common ones are: Apodytes, Canthium, Erythroxylum, Euphorbia pyrifolia, Ficus spp., Maytenus, Mystroxylon, Ochna, Polysphaeria, Sideroxylon, and Terminalia boivinii. Less abundant or locally common taxa are: Acalypha claoxyloides, Allophyllus, Clerodendrum, Dracaena, Flacourtia, Guettarda, Jasminum, Operculicarya, Pandanus tectorius, Phyllanthus casticum, Margaritaria cheloniphorbe, Scaevola, Scutia, Secamone, Tarenna trichantha, Tarenna supra-axillaris, Tricalysia, and Triainolepis. Many other taxa also occur. The understorey is generally barren, but dense to open patches of Cyperus niveus or Lomatophyllum aldabrense locally occur. The Mixed Scrub varies greatly in height from site to site. It is at its greatest in Takamaka Grove, reaching about 12 metres and at its least (less than 1 m) in some of the areas of shallow soil. In some of the most extensive areas of Mixed Scrub (e.g. west of Bassin Frigate) the height is about 3-5 m. 12 Soil: The Organic Brown Calcareous Soil (CMD) (Trudgill, 1979) is the most common found under Mixed Scrub. It often occurs in pits 5-20 cm deep and in some areas is overlain by leaf litter or even several centimetres of arthropod frass (mostly from a millepede). The latter is called a Shallow Organic Soil, Pellet Type (OSP) by Trudgill. In some areas the soil is sandy and belongs to the Farquhar Series. | In a few places, sand has become mixed with the upper layers of the CMD. Shallow Organic Soil (OS) (Trudgill, 1979) may also be found. Phenology: With such a large number of species, the phenological status of this type is complex at any time of year. However, certain generalizations are possible. Firstly, there are evergreen and deciduous species and a few which are neither entirely one nor the other depending upon just how dry it becomes. Flowering and fruiting is in general most prolific during the rainy season. For any particular species the peak may be at the start, middle or end of the wet period. A very few species flower most profusely during the dry months (e.g. Capparis cartilaginea). Some taxa are very opportunistic, flushing new leaves and flowers after any moderate rainy period and just as rapidly losing them when it dries up (e.g. Allophyllus, Erythroxylum). Notes: This vegetation type is perhaps the most complex on Aldabra and the one about which the least is known. As a type it extends into a wide variety of habitats defined on species dominance, canopy height, plant spacing, and substrate type. For the time being it is being left as a single, heterogeneous unit although variants are recognised below as sub-types on the basis of certain conspicuous features. However, the status of these sub-types is very uncertain and requires detailed study. sub-type a. Leeward Scrub Recognised primarily by its location in the lee of the trade winds on or at the base of dunes. Location: mostly associated with the large dunes on the south coast of Grande-Terre. Notes: As a habitat it is extensively used by both tortoises and birds as a cool, shadey place to avoid the heat of midday. As a vegetation category, it is essentially an extension of the Mixed Shrub. sub-type b. Dune Scrub Recognised by its location on dunes and by floristic composition. Location: primarily on large dunes (e.g. Dune Jean Louis). sub-type sub-type 1,3} Vegetation: dominated by woody plants from 1 metre to about 7 m high in open or closed communities. Understorey plants usually absent. The common species are Tournefortia argentea, Scaevola taccada, and Thespesia populnea. Notes; The species are characteristically, widespread, coastal, tropical taxa. ch Beach Scrub Recognised by its location and floristic composition. Location: at the head of small beaches. Vegetation: as for Dune Scrub but the common species are usually Cordia subcordata, Casuarina equisetifolia, Hibiscus tiliaceus, Scaevola taccada, Suriana maritima, and Tournefortia argentea. Notes: Resembles Dune Scrub closely, but the nearness to the sea and less deep sandy soil may account for the somewhat different species composition. Otherwise the two sub-types have much in common. d. Sideroxylon Scrub Recognised primarily on the basis of species composition. Location: Generally on knolls of limestone rising 0.5 to 1.0 m above the surrounding rough champignon terrain. Vegetation: Sideroxylon inerme dominates these knolls as relatively large single-stemmed plants with broadly spreading crowns and fairly dense, evergreen canopy. Maytenus and Scutia may be common associates in the canopy while Cyperus niveus often occurs in the understorey. Notes: This sub-type is one of the important variants of the Mixed Scrub in that it is common and widespread in a mosaic with the extensive areas of Pemphis Scrub. Because Sideroxylon Scrub occurs consistently in habitats distinct from those of Pemphis Scrub, although in a fine mosaic with it, it seems valid and useful to classify the Sideroxylon Scrub separately from the Pemphis Scrub. The dynamic status of the sub-type is not clear. Most Sideroxylon on Aldabra are large and no seedling or sapling individuals have been reported despite prolific fruit production that is known to be fertile at least in part. How this sub-type relates to Mixed Scrub is not clear, as the two merge imperceptibly in some areas. 14 The distribution patterns of Mixed Scrub in some areas of Aldabra are very conspicuous, particularly when seen on aerial photographs. In the present classification, the various distinct patterns are not classified separately, but some comments seem pertinent because of the very different habitats they create. Firstly, there is a conspicuous arrangement of shrubs in rows oriented more or less north west to south east. Fosberg (1971, p 218, and Grubb 1971, p 357) have suggested that the prevailing south east trade wind may be involved in the origin of this pattern. The present day trade winds may be ruled out as the immediate cause by noting that several areas of pronounced row-pattern are sharply demarcated from surrounding scrub which shows no such pattern. Examination on the ground showed that shrubs in areas of row-pattern are almost always growing in pits, depressions or on broken ground separated by smooth to undulating regions with very shallow pits and virtually no soil. The smaller the depression in which the shrubs grow, the more dwarfed were the shrubs. Thus the immediate cause of the rows would appear to be patterning of the habitable substrate and not to the direct influence of the wind. It is noteworthy that very similar patterns can be seen on aerial photo 42 SY 15 no. 031 where they occur not only on dry land but also on the wave cut platform of the south east coast of Grande TenEre . The trade winds do result in regular die-back of the previous seasons growth on shrubs. This die-back is most pronounced on the windward sides of shrubs and eventually may result in very windswept crowns. Some plants with such crowns do lean down wind, but whether this tilt is directly caused by the force of the wind or the stress resulting from the very asymmetric crowns is not clear, but this situation of wind shaping of crowns is not directly related to that of arranging the positions of individuals into a pattern of parallel rows. The second conspicuous pattern is that of "clumps" of shrubs throughout large areas of platin and pavé at the east of Grande Terre. Field observations again have shown the shrubs to be associated with patches of rough ground in an otherwise little broken terrain of limestone. The rough ground may be (as noted by Grubb 1971, p 357) in the form of depressions or sump holes. Such places accumulate soil and thus hold moisture, providing adequate rooting conditions for shrubs, or they may be local patches of moderately deeply pitted champignon or fissured rock. In all cases it seems that where adequate rooting by shrubs is possible, they occur, and where sufficient rooting conditions are not met, there is only low vegetation or bare rock. Takamaka Grove is the superlative development of one such shrub clump on Aldabra. The distribution of space habitable by shrubs in these areas relates to the origins of the substrate pattern Hnatiuk & Merton CLOT ae (7) SCLERODACTYLON TUSSOCK GRASSLAND Location: primarily around the sea coasts of the whole atoll although its greatest extent is along the south coast of Grande Terre. Occasional patches can be found well inland (e.g. 1.3 km south west of Takamaka Grove). 15 Vegetation: One species, Sclerodactylon macrostachyum dominates this vegetation type. This grass grows in tussocks from 0.05 m in height where grazing is heavy to over 0.5 m in height where grazing is absent. The crowns of the tussocks usually interdigitate with each other and produce virtually 100% canopy cover, but where grazing occurs, cover may be considerably less. There is no understorey vegetation, but a variant with Cyperus ligularis as dominant may occur (see notes below) . Soil: The sandy Farquhar Series (CMA of Trudgill, 1979), is the most common soil found in this vegetation type, but stoney soils and occasionally the fine Organic Brown Calcareous soil (Trudgill, 1979) form the substrate of the tussocks. Phenology: Seasonal changes are slight. The tussocks are evergreen and individual leaves appear to remain green throughout their length and then die rapidly, thus the canopy never appears brown from dead leaf tips as occurs in some tussock grasses. The leaves do turn pale grey from salt and fine calcareous deposits during long dry spells with strong onshore winds and ocean swell. Flowering occurs almost exclusively during the rainy season and can be prolific. Notes: The dynamic status of the community is complex. It would seem that under heavy grazing pressure and shade seeking activity of tortoises, the tussocks are killed and replaced by Sporobolus virginicus (Hnatiuk et al., 1976). Where extensive stands of Sclerodactylon occur, the plants often appear to belong to distinct clones, 10 to 50 m across, being distinguished on the basis of leaf colour (blue, brown, light green). In some places that are little disturbed by grazing, the Sclerodactylon grow in a mosaic, closed-canopy community with low, coastal Mixed Scrub on rocky, shallow, soil. Cyperus ligularis also forms tussocks and dense, one-species stands in habitats identical to those of the Sclerodactylon. The Cyperus appears to be an invasive species of disturbed habitats. It is only grazed by tortoises when it is producing new growth. Its loose tussocks and flexible leaves are not as susceptible to mechanical damage from trampling as are tussocks of Sclerodactylon. Until more is known about the sedge, it seems best not to classify it as a separate vegetation type. (8) SPOROBOLUS VIRGINICUS COASTAL TURF Location: Widespread on sandy deposits around the sea coast of the atoll, but most abundant along the south and east coast of Grande Terre. Vegetation: One species, Sporobolus virginicus, dominates this vegetation type, but several other species also are to be found: Euphorbia Stoddartii, Fimbristylis cymosa, Launaea sarmentosa, Lepturus repens, Portulaca mauritiensis, and Sida parvifolia. All but Lepturus occur most frequently with the Sporobolus wheretortoise grazing is intense; Lepturus tends to be found on stonier soil than is Sporobolus. The short turf of Sporobolus (less than 1 to 2 cm in height) with cover of less than 50% in places, found under intense grazing conditions, 16 becomes a tall sward (10 - 20 cm in height) with 100% cover and with a dense accumulation of litter (Hnatiuk et al., 1976). Soil: The type appears largely restricted to the Farquhar Series sand (CMA of Trudgill, 1979) of the low perched beaches but in some places Shioya sand is the substrate. Phenology: Sporobolus virginicus is evergreen. It produces most new growth during the rainy season. It appears to flower most prolifically during the late wet and early dry season, but relatively heavy rains during the dry season can stimulate sporadic flowering. The leaves may become grey on the windward trade coast from fine, air-borne, calcareous deposits during the trade wind season. The vegetation takes on a brownish cast during the dry months as the leaves begin to slowly die back from the apex. Notes: The dynamics of this vegetation type appear to be closely linked with that of the Sclerodactylon Tussock Grassland. The Sporobolus virginicus Coastal Turf may at some time in the past have been confined to unstable sandy soil at beach crests but appears to have become widespread through the feeding and shade seeking activity of tortoises at the expense of both low Mixed Scrub and the Tussock Grassland. However, having once become established over large tracts, it seems to persist (Hnatiuk et al., 1976). Its long term status in these places is not known. (9) TORTOISE TURF Location: primarily on the platin and pavé terrain at the east end of Grande Terre although rather small, often poorly developed patches are found on Ile Picard (e.g. Back Path, near Bassin Cabri, and near Anse Var) and on Ile Polymnie. Vegetation: Tortoise Turf is an assemblage of many species that change in relative dominance from place to place. The following dwarf grasses and sedges are most common: Bulbostylis basalis, Dactlyoctenium pilosum, Eragrostis decumbens, Fimbristylis cymosa, Panicum aldabrense, Cyperus pumilus, and Sporobolus testudinum. The most common dicot herbs are Euphorbia stoddartii, Phyllanthus maderaspatensis, Sida parvifolia, and Tephrosia pumila. A thallose liverwort, Riccia, is common on heavily grazed ground. The intensely grazed turf is mostly less than 1 or 2 cm in height and the canopy cover less than 20%. In lightly grazed conditions, cover can approach 100% and height, depending upon which species is dominant, can reach 10 - 15 cm. There does not appear to be any development of strata within the community. Soil: The Organic Brown Calcareous Soil (Trudgill, 1979) is almost all that is found in this type. The soil is only a pale coloured representative of this soil type, perhaps because the dark horizon only forms under a shrub canopy or perhaps the former, darker horizon has been eroded in recent years (Merton et al., 1976). 17 Phenology: The summer rains bring a slow greening of the turf and the onset of dry conditions brings a gradual browning as many of the plants die. Flowering and fruiting is largely done during the late rainy period and early dry period. Most of the species appear to be short lived perennials or facultative annuals during years of severe drought. Seed production can be high judging from the abundance of seedlings that can be found once the wet season is well established. Notes: Tortoise Turf, first named by Grubb (1971, p 359), isa community that appears well developed to withstand moderate to intense grazing. In fact, its species richness may in part be possible because grazing keeps the more vigorously growing members from overtopping the less vigorous ones. Exclosure plots have only recently been set up but strong species interactions appear to be possible as the various species grow to much greater stature when not or little grazed than when heavily grazed. The Tortoise Turf has been seen to be invaded by the large sedge Cyperus ligularis where grazing has been restricted. It does not appear that the turf survives under a shrub canopy. Thus where shrubs are prevented from growing by continual browsing and abrasion as for example with Ficus spp., and Ochna, the Turf thrives where it would otherwise be shaded out. In the absence of at least moderate grazing, the Tortoise Turf could well become much less extensive than it is today. The status of broad leaf herbs in the Turf is not clear. They are all readily eaten by the tortoises and some of these herbs are restricted to pits and crevices where the tortoises cannot reach. Many of these herbs appear capable of growing more rapidly and much larger than most of the dwarf grasses and sedges in the Turf. Thus, whether the dicot herbs would completely dominate the Turf if grazing were reduced is a matter of conjecture. Two sub-types may be recognized: sub-type a. Herbaceous pasture It is characterized by a dominance of broad leaved herbs. It occurs in areas little grazed by tortoises (e.g. pits in champignon near Cing Cases, on the 8 m ridge south of Anse Var, along part of Back Path, and near Bassin Cabri). The major taxa vary with geographic location but include: Asystasia, Euphorbia stoddartii, Evolvulus, Hypoestes, Lagrezia, Nesogenes, Hedyotis, Portulaca mauritiensis, Ruellia, and Tephrosia. Although quite distinct communities of Herbaceous Pasture can be found, because it grades into Tortoise Turf in some places, it here will be only distinguished as a sub-type. sub-type b. Fimbristylis cymosa Turf It is dominated by a single species and occurs Over extensive areas of platin and pavé. It does not appear to be as readily grazed by tortoises as the other Turf 18 species although exclosure sites indicate that there must be some degree of grazing of new leaves and inflorescences during the rainy season. Fimbristylis cymosa appears somewhat more abundant in the northern part of the platin-pavé at the east end of Grande Terre while the other monocot Turf community is more common further south, but the two ranges overlap considerably. Until more is known about the interrelationships of the various Turf communities, it seems best to recognize only sub- types of the main Tortoise Turf vegetation. (10) BACOPA-MOLLUGO MEADOWS Location: It has been found only at the east end of Grande Terre on summer flooded basins. Vegetation: The dwarf, succulent herbs, Bacopa monnieri and Mollugo oppositifolius, are the most common species found. They generally grow as very prostrate plants, less than 1.5 cm high and rooting at the nodes. Where protected from grazing Mollugo has been found to grow very much larger, reaching 20 cm in height, but Bacopa does not respond so vigorously. Cover is variable from greater than 75%, to less than 10% where grazing is intense or soil sparse. Soil: A “mineral sediment (SM)" of Trudgill, (1979). It, is: al fine grey, generally saline silt, that may crack on drying. Algal sediments (SA of Trudgill, 1979) may also occur here. An algal surface layer is not uncommon. The soil is periodically submerged when the basins fill after rain, and then slowly dries out. Depending upon the extent of the droughts each dry season, the soil may or may not completely dry out. Phenology: The plants are short lived perennials that remain green for as long as there is sufficient moisture available. Several weeks innundation does not halt either leaf or flower production although competition from algal blooms may eventually retard their growth. When the water recedes, the herbs flourish. Flowering and fruiting can thus occur at any time of year when moisture conditions permit. Notes: Bacopa-Mollugo Meadows are closely associated with Lumnitzera- Thespesia Scrub. However, since the former does not grow under the latter nor are they always present at the same basin, it seems best to classify these two very different structural units as separate vegetation types. The particular conditions favoured by each of the two dominant herbs is not clear. Bacopa may be more common in shallow soil and rock crevices while Mollugo may be more common on the deeper soil of the basin, but more observations are needed. 19 sub-type a. Bryodes Meadow A dicot herb community dominated by one species, Bryodes micrantha. It appears to form in some places, a zone which is transitional between Bacopa-Mollugo Meadow and Tortoise Turf. The soil is often innundated by water during the wet season, but as the colour of the soil is brown, the duration of water logged conditions must be brief. The Bryodes community is ephemeral, appearing for a brief period only after prolonged rainy periods and dieing out as the soil dries out. Flowering and fruiting occur readily during this brief, active season. If artificially watered, the individual plants can live at least a year. Because only a very few patches of this community have been seen, its status is uncertain. The pools occupied by both Bacopa-Mollugo Meadow and Lumnitzera Thespesia populneoides Scrub are found in depressions with impeded drainage in the hardened platin and pavé. The basins may fill with 0.5 m to 1.5 m of water during the rainy season, and slowly dry during the trade wind season. A complex succession of algae parallels the seasonal changes in water levels (A. Donaldson pers. comm.), and fish are occasionally to be found in the flooded pools. When pools dry out they are left with a hard, cracked crust that is matted by algae, or may be colonized by Bacopa and Mollugo. (11) PIONEER VEGETATION OF BEACHES Location: at the crest of most beaches around the atoll. Vegetation: A mixture of herbs common on the strand line throughout much of the tropics forms this type. The most common species are Cyperus conglomeratus, Dactyloctenium ctenoides, Ipomoea pes-caprae, and Sporobolus virginicus. The community may consist of several or only one of these species. The canopy height is dependent on the species but is usually less than 0.2 m except for Cyperus which rises to 0.6 m. The canopy cover is generally less than 25% but locally denser patches do occur. Soil: The soil is undeveloped beach sand. Phenology: Vegetative growth, flowering, and fruiting are prolific during the wet season and much reduced or halted during the dry months. Notes: This vegetation type is a pioneer on disturbed beach crests. It can vanish overnight if the beach is eroded by a storm, Or may gradually disappear as tall woody plants of the Mixed Scrub sub-type Beach Scrub become established on places of extended stability. 20 (12) HERBACEOUS MEADOWS ON LAGOON ISLETS Location: on some of the islets in the lagoon, particularly off the north shore of Grande Terre (e.g. Champignon des Os). Vegetation: The dominant species are different on different islets and even on different parts of the same island. But taken as a whole they form an assemblage quite distinct from anything else on the atoll and therefore they are recognized as a separate vegetation type. The commonest species are Achyranthes aspera, Boerhavia sp., Dactyloctenium pilosum, Lagrezia oligomeroides, Lepturus repens, Portulaca oleracea, P. mauritiensis, Sida parvifolia, and more rarely Sesuvium portulacustrum. The shrubs Acalyphya claoxyloides and Pemphis acidula are not uncommonly present though in a stunted form. Cover is generally high (greater than 75% during the wet season). Average height to the top of the canopy is less than 0.75 m and often less than O.3 m. Soil: Generally very little and composed of coarse sand with a few silty patches presumably both blown or washed up from the lagoon floor by wind and wave action. Bird dung and remains are locally abundant and in such areas the soil may be of the Desnoeufs Series (the organic soil, guano variety, OSG of Trudgill, 1979). Phenology: Vegetative growth, flowering and fruiting are largely confined to the wet season and early dry season until fresh water supplies are exhausted and then most of the community goes dormant. These islands can thus be bright green in January to March, and dull brown in August to November. Notes: The dynamic status of this vegetation is uncertain but since some islets are currently used as roosts and nesting sites, while others appear to have been so used in the past, they may have been modified from shrub to herb cover by the activities of the birds. Some dead shrubs can be seen on some of the islets which are now entirely covered by Herbaceous Meadow. (13) GARDENS Location: near current or old sites of habitation, primarily at Settlement but also near Anse Var, Anse Mais, Anse Malabar, Ile Michel and Passe Houareau camp. Vegetation: Many species of both cultivated and "weed" status are used to define this type. The commonest cultivated species are Ipomoea batatas, Moringa oleifera, Capsicum frutescens, Phaeseolus sp., Datura metel, Tamarindus indicus, Cymbopogon, Solanum melongena, Carica Papaya, Panicum maximum, Cucurbita spp., Agave sisalana, and Pedilanthus tithymaloides. Common weeds and ornamentals are Stachytarpheta jamaicensis, Tridax procumbens, Synedrella nodiflora, Catharanthus roseus, and Sida acuta. 21 Canopy height and cover is very variable depending upon the degree of cultivation, disturbance, and species composition. Soil: Primarily of the Farquhar Series although some phosphate rich soil has been reported (Baker, 1963,p 107). Phenology: Variable, depending upon the species concerned. However, most Of the species are active during the wet season and early dry season, becoming dormant during the driest months. Some species such as Carica produce most leaves during the rains, and ripen their fruits only during the mid to late dry season. Notes: Although the vegetation can be recognized on the basis of its floristic composition, certain special features can be noted. The species are all probably either deliberately or accidentally introduced by man. Some of the species do not appear able to spread either because all viable seed seems to be eaten by animals (e.g. Moringa), or no flowers are produced and vegetative spread is extremely limited (e.g. Cymbopogon, Bambusa) . Other species have become naturalized and are spreading into the native vegetation (e.g. Stachytarpheta, Passiflora, Agave) . REFERENCES Baker, B.H. 1963 Geology and mineral resources of the Seychelles Archipelago. Geol. Surv. Kenya Mem. 3. Braithwaite, C.J.R., Taylor, J.D., & Kennedy, W.J. 1973 The evolution of an atoll. Phil. Trans. R. Soc. Lond. B. 266: 307-340. Fosberg, F.R. 1967 A classification of vegetation for general purposes. In G.F. Peterken, IBP Handbook No. 4 Guide to the Check Sheet for IBP areas. Blackwell Scientific Publ. Oxford 8 Edinburgh. Fosberg, F.R. 1971 Preliminary survey of Aldabra vegetation. Phil. Trans. R. Soc. Lond. B. 260: 215-227. Fosberg, F.R. & Renvoize, S. 1979 Flora of Aldabra. Royal Botanic Gardens, Kew. Fryer, J.C.F. 1910-12 The structure and formation of Aldabra and neighbouring islands with notes on their flora and fauna. Linn. Soc. (Lond.) 2nd series (Zoology), 14: 397-442. Grubb, P. 1971 The growth, ecology and population structure of giant tortoises on Aldabra. Phil. Trans. R. Soc. Lond. B. 260: 327-372. Hnatiuk, R.J. & Merton, L.F.H. 1979 A perspective of the vegetation of Aldabra. Phil. Trans. R. Soc. Lond. B. 286: 79-84. 22, Hnatiuk, R.J., Woodell, S.R.J., & Bourn, D.M. 1976 Giant tortoises and vegetation interactions on Aldabra atoll. ib. (Coastal. Biol. Conserv. I=) 305—-s16e Macnae, W. als y/al Mangroves on Aldabra. Phid Lanse lh SOC hme. 260: 237-248. Merton, L.F.H., Bourn, D.M., & Hnatiuk, R.J. 1976 Giant tortoises vegetation interactions on Aldabra atoll. De SinVvand* Biol. Conserv. 9: 293-304. PAgGOR aC he 1968 A soil survey of Seychelles. Tech. Bull. Nouwee Land Research Division, Directorate of Overseas Surveys, Tolworth. Ridley, H.N. 1930 The Dispersal of Plants throughout the World. L. Reeve and Co. Ashford. Stoddart, D.R. 1968a The Aldabra affair. Biol. Conserv. 1: 63-69. Stoddart, D.R. 1968b The conservation of Aldabra. Geogrl. J. 134: 471-486. Stoddart, D.R. & Wright, C.A. 1967 Geography and ecology of Aldabra atoll. Atoll, Res. Budi V1Ss, 1-52. Stoddart, D.R. & Mole, L.U. 1977 Climate of Aldabra Atoll. Atoll? Res. Bully 20223) 1-2 Trudgill,, ST. 1979 The soils of Aldabra. Phil. Trans. BR; Soc. Lond. B. 286: 67-77. Vesey-Fitzgerald, L.D.E.F. 1942 Further studies on the vegetation of islands in the Indian Ocean. J. ECOL, 302 “L=16. Wickens, G.E. oye Speculations on seed dispersal and the flora of the Aldabra Archipelago. Phil. Trans. R. Soc. Lond. B. 286: 85-977. iy ATOLL RESEARCH BULLETIN 240. Man and the Variable Vulnerability of Island Life. A Study of Recent Vegetation Change in the Bahamas by Roger Byrne Issued by | ah THE SMITHSONIAN INSTITUTION a fea Washington, D. C., U.S.A. January 1980 ATOLL RESEARCH BULLETIN 240. Man and the Variable Vulnerability of Island Life. A Study of Recent Vegetation Change in the Bahamas by Roger Byrne Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. January 1980 ACKNOWLEDGEMENT The Atoll Research Bulletin is issued by the Smithsonian Institution, as a part of its activity in tropical biology, to place on record information on the biota of tropical islands and reefs, and on the environment that supports the biota. The Bulletin is supported by the National Museum of Natural History and is produced and distributed by the Smithsonian Press. The editing is done by members of the Museum staff and by Dr. D. R. Stoddart. The Bulletin was founded and the first 117 numbers issued by the Pacific Science Board, National Academy of Sciences, with financial support from the Office of Naval Research. Its pages were largely devoted to reports resulting from the Pacific Science Board's Coral Atoll Program. The sole responsibility for all statements made by authors of papers in the Atoll Research Bulletin rests with them, and statements made in the Bulletin do not necessarily represent the views of the Smithsonian nor those of the editors of the Bulletin. Editors F, R. Fosberg Ian G. MacIntyre M.-H. Sachet Smithsonian Institution Washington, D.C. 20560 D. R. Stoddart Department of Geography University of Cambridge Downing Place Cambridge, England ATOLL RESEARCH BULLETIN NO. 240 MAN AND THE VARIABLE VULNERABILITY OF ISLAND LIFE. A STUDY OF RECENT VEGETATION CHANGE IN THE BAHAMAS by Roger Byrne Issued by THE SMITHSONIAN INSTITUTION Washington, D.C., U.S.A. January 1980 TABLE OF CONTENTS TABLE OF CONTENTS LIST OF Jessi (O}F Section IL IIe 6 Ito We Viele Walle. FIGURES TABLES INTRODUCTION THE THEME: MAN AND THE VULNERABILITY OF ISLAND LIFE Early Interpretations Recent Views Unresolved Issues THE SETTING: A LOW LIMESTONE ISLAND Insularity Geology and Geomorphology Soils Climate Hydrology The Environment as a Whole THE PRESENT VEGETATION: A GENERAL VIEW Salt-Water Habitats Man's Impact Seasonally-Flooded Fresh-Water Habitats Man's Impact Well-Drained Fresh-Water Habitats CLEARING AND BURNING FOR AGRICULTURE The Island Arawak The English The Free Negro Contemporary Agriculture SELECTIVE CUTTING OF INDIVIDUAL SPECIES Dyewoods Barks Timber Trees Local Exploitation THE INTRODUCTION OF ALIEN PLANTS AND ANIMALS Introduced Plants Domesticated Animals alah Section Weleier. IX. GES TEAL CCIE Appendix Tes ikke ILLES TeVis METHODS OF ANALYSIS Field Methods Analysis of Field Data Limitations of Field Data THE IMPACT OF SHIFTING AGRICULTURE: AGE DIFFERENCES IN THE WOODLAND The Whiteland The Flatland The Blackland THE IMPACT OF SELECTIVE PRESSURES: AGE DIFFERENCES IN THE WOODLAND Goats and the Woodland Selective Cutting Combined Selective Pressures THE INVASION OF ALIEN PLANTS The Species Involved The Extent of the Invasion THE STABILITY OF THE WOODLAND Adaptations to Clearing and Burning Resistance to Grazing and Browsing Resistance to Invasion by Alien Plants The Significance of Diversity The Nature of Insularity The Variable Intensity of Man's Impact REVIEW AND CONCLUSIONS ACKNOWLEDGMENTS BIBLIOGRAPHY A Systematic list of the Plant Species Encountered on Cat Island Index to Common Names of Plants Mentioned in the Text Data Sheet Used During Analysis of Old Fields A List of the Minor Species Encountered in each Habitat-Type and Age-Class 10. The 16. Ie LIST OF FIGURES Figure Map of the Bahama Islands Location Map of Cat Island Geomorphological Cross-Section of Cat Island Map of Cat Island Showing Major Habitat-Types Photograph showing Coastal Vegetation on Holocene Beach Ridges Photograph showing Coastal Vegetation On Holocene Dunes Photograph showing Vegetation on a Leeward Limestone Coast Photograph showing Vegetation on a Exposed Limestone Coast Photograph showing a Rhizophora Forest Photograph showing Tidal Channel Vegetation Photograph showing Competition Between Casuarina and Gundlachia Photograph of "Hammock" Vegetation in the MecQueens Savanna Photograph of Pothole Vegetation Photograph of Recently Burned Palmettoes Photograph of "Footpath Distribution" of Palmettoes Aerial Photograph of the Woodland Photograph of the Woodland looking east towards Freetown iii Page 28 28 31 31 32 32 34 36 36 38 38 40 44 iv 183. 19. 20. alee (efee oor. 24. 25K. 26. ra 28. 29. 30. Sila 32% 33. Se Bie Bb. Figure Photograph of the Whiteland being Cleared Photograph of a Whiteland Field Ten Years after Cultivation Photograph of a Flatland Field Close to the Savanna Photograph of Second Growth Vegetation on the Flatland Photograph of a Severely Burned Blackland Field in 1967 Photograph of a Severely Burned Blackland Field in 1970 Diagram showing Aerial Photograph Reference Grid Whiteland Succession Diagram Aerial Photograph of Whiteland, 1943 Aerial Photograph of Whiteland, 1958 Flatland Succession Diagram Aerial Photograph of Flatland, 1943 Aerial Photograph of Flatland, 1958 Blackland Succession Diagram Aerial Photograph of Blackland, 1943 Aerial Photograph of Blackland, 1958 Graphs showing Diversity and Height against Age, for each Habitat-Type Map showing Distribution of Distance Classes Photograph (1) showing Impact of Grazing by Goats Page 73 ne 45 45 46 46 81 90,91 94,95 94,95 97,98 100,101 100,101 104,105 106,107 106,107 110 114 116 Silke 38. 39. HO. 1. 42, 43. 4d, 45. 46. Figure Photograph (2) showing Impact of Grazing by Goats Graphs showing Age/Distance Unpalatable Species Graphs showing Age/Distance Distributions of Selectively-Cut Species, and Metopium toxiferum Photograph showing a Corn Field Infested with Leucaena leucocephala being Outgrown by Native Species Photograph of a Recently Burned Blackland Field Photograph of a Two Year Old Blackland Field Photograph of the Savanna/Woodland Transition Photograph of the Mangrove/Woodland Transition Diagram showing the Floristic Composition across a Low Limestone Ridge Page 116 118 124 135 135 143 143 144 144 146,147 vi Table LIST OF TABLES Climatic Data 1952-1962 The Bight, Cat Island Vegetation and Habitat-Types Age and Tonal Density The Number of Fields Sampled in Each Category Species Included in the Analysis of Browsing by Goats Species Included in the Analysis of Selective Cutting Diversity, Distance and Age Percentage Cover Alien Species Area and Population of Small West Indian Islands, 1950-1958 Page 21 26 84 86 is 123 128 136 154 MAN AND THE VARIABLE VULNERABILITY OF ISLAND LIFE. A STUDY OF RECENT VEGETATION CHANGE IN THE BAHAMAS 1 by Roger Byrne I. INTRODUCTION A question of increasing concern to scientists and lay- men alike is to what extent plant and animal communities can withstand disturbance by man. In this context the histori- cal biogeography of smell islands is especially relevant. Island life has proved to be particularly vulnerable to human disturbance. In the brief period of human settlement faunal extinction rates have been proportionally much higher on islands than on the continents. For plants the situation is less clear. Few island studies have dealt in detail with the consequences of man's impact and as a result the vulner- ability of insular plant communities is poorly understood. It was with this general problem in mind that the present Study was undertaken. More specifically, an attempt was made to determine the extent to which man has modified the vegetation of Cat Island, a small island in the Bahamas. Originally, it was intended to deal with the vegetation of the island as a whole, but for several reasons the detailed analysis was limited to the mixed evergreen-deciduous woodland, or "cop- pice" as it is locally known. The woodland covers more than 90 percent of the island and has been intensively disturbed by man. On a theoretical level the question was considered as to whether or not the vegetation of offshore islands such as Cat is vulnerable in the same way as that of the Hawaiian Islands or the Galapagos. The choice of Cat Island was to a certain extent fortuitous. As a low-limestone island it provided a comparatively simple setting, it was reasonably T. Present address: Department of Geography, Universi- ty of California, Berkeley, California, 94720. (Manuscript received November 1978--Eds.) 2 accessible, and like the Bahamas as a whole its vegetation was virtually unexplored. Field work was carried out during three visits: July to October 1967, June to August, 1968 and June to September 1970. In addition short visits were made to three other Bahamian islands: Bimini, Mayaguana, and New Providence. Historical evidence was gathered in London during the spring of 1968. Particular attention was paid to the manuscript collections at the Public Records Office, the British Museum, the British Museum of Natural History, and the Royal Botanical Gardens at Kew. This text represents a revised version of a doctoral dissertation submitted to the Univer- sity of Wisconsin in 1972. (Byrne,.-1972). II. THE THEME: MAN AND THE VULNERABILITY OF ISLAND LIFE The idea that island life is inherently vulnerable to disturbance by man was first proposed by Charles Darwin as Supporting evidence for his theory of natural selection. Since Darwin's time, however, the whole question of insular vulnerability has been variously interpreted and as_ yet there is no consensus. Early Interpretations During his voyage on the Beagle (1831-1836), Darwin was particularly impressed by the extent to which man had dis- turbed the plant and animal life of so many remote islands. On St. Helena, for example, the woodland had been virtually removed by the combined effects of selective cutting and grazing, and the native plants had been apparently replaced by introduced species. The many imported species must have destroyed some of the native kinds; and it is only on the highest and steepest ridges that the indigenous flora is now predominant (Darwin, 1839: 485). In his account of the voyage, Darwin made no attempt to explain why the plants and animals of remote islands should have been so vulnerable to disturbance. His explanation was to come twenty years later in The Origin of Species. Here he suggested that continental species have a competitive advantage over insular species because the struggle for existence is more severe on continents than on islands. Islands, because of their small size and inaccessibility, have fewer species and consequently the competition among those present is less vigorous. To support this idea he pointed out how relict floras and faunas had survived on islands long after they had become extinct on the mainland. On a small island the race for life will have been less severe, and there will have been less modifi- cation and less extermination. Hence, we can understand how it is that the flora of Madeira...resembles to a certain extent the extinct tertiary flora of Europe (Darwin, 1859: 108). This view of islands as natural museums in which plants and animals are preserved by isolation was generally accepted by the great nineteenth century naturalists. Wallace, for example, agreed with Darwin that species from continental areas, particularly Europe, were more aggressive than insular’ types. Like Darwin, he noted how the native flora and fauna of St. Helena had been drastically changed during the few hundred years of European settlement. When first visited by civilized man it was in all probability lie nady; stocked with plants’ and animals, forming a kind of natural museum or vivarium in which ancient types, perhaps dating back to the Miocene period, or even earlier, have been saved from destruction which has overtaken their allies on the great continents (Wallace, 1902: 308, 309). Darwin's close friend Joseph Dalton Hooker was. also impressed by the vulnerability of island life. He wrote at length on the success of European weeds on oceanic islands, and like Darwin attributed it to their supposedly superior competitive ability (Hooker, 1860, 1865, 1867a, 1867b). In general terms, Darwin, Wallace, and Hooker were all agreed that island life was inherently vulnerable. What was not clear was the detailed nature of this vulnerability and to what extent it was due to man. The nineteenth-century naturalists were limited in their thinking by the typologi- cal species concept, and because of this they reduced the intricacies of competition to a battle in which the length of species lists assumed an inordinate importance. This is not to say that they were entirely unaware of the _ signifi- cance of man's role. Wallace (1902: 306), for example, emphasized that European weeds could not have’ successfully invaded New. Zealand had man not disturbed the native vegeta- talon Hirst. Before reviewing more recent views of the vulnerability of island life, mention should be made of the important role oceanic islands played in the development of the theory of natural selection. In The Origin of Species Darwin argued strongly against the then popular land-bridge theory that had been used to account for the origin of insular floras and faunas. Oceanic islands, he maintained, had never’ been connected to the continents and consequently were poor in species of both plants and animals; furthermore, many impor- tant taxa were not represented at all. This impoverishment, he argued, was further evidence against the doctrine of independent creation. T. Allan (1936), in his Critique of the. hypothesis 3am insular vulnerability, failed to give Wallace credit for this observation. He who admits the doctrine of the creation of each separate species, will have to admit that a suffi- cient number of the best adapted plants and animals were not created for oceanic islands; for man has unintentionally stocked them far more fully and perfectly than did nature (Darwin, 1859: SO Ye This argument reinforces his earlier conelusion that the inhabitants of oceanic islands are inherently vulnerable to disturbance. Both Hooker and Wallace emphasized the difference between continental and oceanic islands; the former having at some time been connected by land bridges to the continents, whereas the latter had always been isolated. Wallace elaborated on the distinction and defined oceanic islands as follows: Islands of volcanic or coralline formation, usu- ally far from continents and always separated from them by very deep sea, entirely without indigenous land mammalia or amphibia, but with a fair number of birds and insects, and usually with some _ rep- tiles (Wallace, 1902: 243). Recent Views In the present century the broad approach of the natural scientist has generally been abandoned in favor of narrow specialization. Botanists have been divided on the question as_ to whether or not insular plants are inherently vulnerable to disturbance. One school of thought has held that Darwin's interpretation was basically incorrect. Allan (1936), for example, argued that continental plants are not inherently more aggressive than insular types. According to Allan, in New Zealand the introduced species owe their success to prior disturbance of the natural vegetation by man and his domesticated animals. Much the same conclusion was reached by Egler in his review of the status of alien plants in Hawaii. In the absence of anthropic influences, the evi- dence strongly favors the view that most of the aliens will be destroyed by the indigenes, such aliens surviving only in greatly reduced numbers and as very subordinate members of the resulting ecosystem (Egler, 1942: 23). At the same time, Egler argued against the need for any general theory to account for the processes involved, and suggested that the history of each alien species should be looked at individually. A somewhat different view of insular vulnerability has been presented by Fosberg (1936, 1965, 1972). Fosberg has argued that the ecosystem concept provides an especially useful means of evaluating the significance of man's impact on island life. In essence, he has restated the Darwinian hypothesis in modern terms. In his introduction to the sym- posium Man's Place in the Island Ecosystem, he characterizes the island ecosystem as follows: Limitation in organic diversity; reduced inter- species competition; protection from outside com- petition and consequent preservation of archaic, bizarre, or possibly ill-adapted forms; tendency toward climatic equability; extreme vulnerability, or tendency toward great instability when isola- tion is broken down (Fosberg, 1965: 5). In another paper on this theme, Fosberg emphasized the contrast between old continental ecosystems and young island ecosystems. The former, he suggests, are floristically and faunistically diverse, well-balanced, rarely invaded by aliens, and quick to recover after disturbance. The latter are poor in- species, often imbalanced, often invaded, and Slow to recover after disturbance (Fosberg, 1963: 557-561). The ecosystem concept has been of considerable value in guarding against too narrow a view of man's impact on island ltaher Another botanist who has emphasized the idea that island life is inherently vulnerable is Carlquist (1965, 1970). Like Darwin and Wallace, Carlquist has suggested that islands have been refugia for species that have become extinct on the continents. In his recent book ongatme natural history of Hawaii, he devoted a chapter to a con- Sideration of the loss of competitiveness in native plants, (Carlquist, 1970: 173-179). He notes that the flora is espe- cially poor in poisonous, strongly-aromatic, or spiny plants and concludes that this reflects the lack of any grazing pressure from mammals. He also points out that few Hawaiian plants are weedy and that most species are less competitive than their continental counterparts. These factors. together with inbreeding, small population sizes, and highly Specialized habitat requirements, have made the Hawaiian Species especially vulnerable to disturbance. Unfor- tunately, Carlquist avoids the subject of man's impact. His main concern is the fate of the rare endemics, and introduced species are dismissed as uninteresting weeds, whose story could "only have been a catalogue of sorts" Cearekquisitre | iO-sivacd) es lhas attitude nas) been — charac= teristic of many island botanists. Rare endemics and remote virgin forests have attracted much more attention than cosmopolitan weeds and secondary woodland. Unlike botanists, zoologists have rarely questioned the idea that island life is inherently vulnerable to distur- bance. Simpson and Mayr, for example, have accepted the vul- nerability thesis and have tried to place it in the frame- work of modern evolutionary theory. Simpson (1953: 306) suggested that islands, particu- larly small, strongly isolated islands, are "evolutionary traps," in which the possibilities of further evolution are extremely restricted. The organisms reaching such islands become specifically adapted to a small number of niches, and thereafter a rather static, closed ecological situation per- Sists. Populations are likely to be small, with little genetic variability available for change. If invasion occurs, the native organisms are particularly subject to rapid extinction. On the other hand, he rejects the idea that islands are ephemeral features and therefore unlikely to be very old. Mayr (1963: 74-76) has likewise noted that insular fau- nas are particularly vulnerable to competition from intro- duced species. In general terms he accepts the idea that Species from large areas have a competitive advantage over Species from small areas.~ At the same time he cautions against the acceptance of any sweeping generalizations and points out that there are many exceptions to the rule. Another zoologist who has been concerned with the consequences of the invasion of islands by alien species is Wilson (1965). From a statistical analysis of Hawaiian bird faunas he concluded that there was no evidence to suggest that continental species were intrinsically superior to insular species. His approach, however, involving as it did only faunal lists, could hardly be expected to provide any econelusive answers. A broader approach to the whole problem of insular vul- nerability has been advocated by Elton. In The Ecology of Invasions by Animals and Plants he devoted a whole chapter to a consideration of "The Fate of Remote Islands" (1958: 2. Mayr expressed the same idea in his concluding re- marks to the symposium on The Genetics of Colonizing Species (Baker and Stebbins, T9605: 559). 77-93). Having documented the drastic changes on islands such as Juan Fernandez and Hawaii, he repeats the argument that insular plant and animal communities are vulnerable to invasion because of their comparative simplicity. Natural habitats on small islands seem to be more vulnerable to invasion than those on the con- tinents. This is especially so on oceanic islands which have rather few indigenous species (Elton, 1958: 147). The idea that species diversity can be directly corre- lated with stability has since become a canon of the conser- vation movement. In essence it is a restatement of the Darwinian hypothesis put forward nearly a hundred years ear- lier. Unfortunately, Elton does not explain on just why small islands are so vulnerable or, more particularly, to what extent man is responsible for this vulnerability. As Elton himself admits, this is a poorly-researched topic and ecologists have in general avoided such complicated ques- GLONS). Geographers who have explored this theme have rarely been intimidated by the complexities of island life. In most cases they have taken an holistic approach which has included plants, animals, and man. A pioneer study of this kind is Clark's The Invasion of New Zealand by People, Plants, and Animals (1949). Here, however, the main concern was with the invaders, particularly man, and the changes in the native flora and fauna are only briefly described. More recently, Harris (1965) has explored ae similar theme in his Plants, Animals, and Man in the Outer Leeward Islands. This study involved a detailed review of the his- torical evidence for man's impact on the plant and animal communities of Antigua, Barbuda, and Anguilla. Harriss conclusions were much the same as Elton's, namely, that small oceanic islands are particularly vulnerable to inva- Sion by alien plants and animals, especially man. The plant and animal communities of the Outer Leewards are highly vulnerable because they lack the "ecological resistance" of more complex communities, and although the number of species that have become extinct since man first settled the islands is not known, it is thought to be large (Harris, 1965: 141). In another publication, Harris (1962) reviewed Darwin's hypothesis in the light of his research in the Outer Lee- wards. Like Allen and Egler, he emphasized the point that without prior disturbance by man, alien plants would make up only a small proportion of the total plant cover. In other words, alien plants are not inherently more aggressive than the native species. A very similar conclusion was reached by Watts (1966, 1970) in his study of man and vegetation change in Barbados. Having shown that alien species had not been able to invade areas of comparatively undisturbed vege- tation, Watts concluded that alien species were dependent for their success on prior disturbance by man. In a Similar study Kimber (1969) analysed the history of recent vegetation change in Martinique, a small volcanic island in the Lesser Antilles. Although her approach was primarily historical, it was strongly influenced by the ecosystem concept. The question of insular vulnerability waS not considered as such, but the evidence presented clearly shows that the vegetation of the island has. been disturbed drastically by man. In her conclusion she states that the island ecosystem of Martinique is out of balance and that the plant cover has been degraded and simplified by man's interference (Kimber, 1969: 599). A somewhat different approach to the problem has been cakenmebDy dad. ssauer (1960), 1967). Unlike Clark, Harris), Watts, and Kimber, Sauer combined the historical approach with detailed analyses of the contemporary vegetation. His research on the coastal vegetation of Mauritius and the Sey- chelles is especially relevant to the present study insofar as it shows that the native species were inherently well- adapted to withstand the impact of disturbance by man. These findings do not necessarily mean that Darwin's hypothesis should be rejected. They do mean that coastal Species with their cosmopolitan distributions are not typi- cally insular. Significantly, the inland vegetation on both Mauritius and the Seychelles has been extensively modified by man. Unresolved Issues In general terms Darwin's hypothesis has withstood the test of empirical research rather well, although it must be admitted that few biologists or geographers have concerned themselves with a detailed analysis of insular vulnerabil- ity. In some cases vulnerability is easily understood. Flightless birds without fear of predators are obvious can- didated for extinction. For plant populations the situation is less clear. With a few notable exceptions, such as Egler, Fosberg, and J.D. Sauer, botanists have tended to ignore the subject, and have concerned themselves with systematic studies or with the analysis of undisturbed vegetation. Several biogeographers have explored the theme but for the most part 10 have restricted themselves to the historical approach. Unfortunately, the historical record alone Ls rarely detailed enough to allow for anything more than a very qual- itative reconstruction of vegetation change. For some islands it has been possible to establish the chronology of introductions and extinctions, but these in a sense represent only the first and last chapters of the story. The complex processes of vegetation change that have been started by man on small oceanic islands are still only poorly understood. On a different level, the question as to what extent islands vary in their vulnerability remains unanswered. Theoretically, one would expect that the vegetation of inac- cessible islands such as the Hawaiian Islands or the Galapagos would be more vulnerable to disturbance than that of off-shore islands such as the Bahamas. The former are inhabited by rare endemics, the latter by wide-ranging Species. Somewhat unexpected, therefore, are the conclu- sions reached by Harris (1965), Watts (1966), and Kimber (1969), which suggest that the plant and animal life of the Lesser Antilles is vulnerable in much the same way as that of the Hawaiian Islands. If this is indeed the case, the plant and animal life of the Bahamas might also be expected to be vulnerable. With these questions in mind, it was decided to study in some detail man's impact on the vegetation of one small island in the Bahamas, Cat Island. More specifically, the study attempts to determine to what extent cutting, burning, browsing, and the introduction of alien plants and animals have brought about changes in the native vegetation. Although Cat Island is an oceanic island in the sense that it has never been joined to the continent, it is clearly less isolated than remote islands such as Hawaii or the Galapagos. As an offshore island with a lesser degree of insularity, it provides an interesting test. for =etne hypothesis of insular vulnerability. 11 III. THE SETTING: A LOW LIMESTONE ISLAND In spite of their early discovery, the Bahamas are still in many respects unknown. Wallace (1902:5), in his introduction to Island Life, briefly commented on _ the remarkable contrasts in flora and fauna between peninsular Florida and the Bahamas, only 50 miles to the east (Figure We The differences, he argued, could not be explained by existing conditions, as the climate and soil of the two areas were the same. The Bahamas were different, he implied, because they were islands. Unfortunately, although he went on to discuss, with the aid of many examples, "the complex causes of insular floras and faunas," the Bahamas were not referred to again. Then, as now, very little was known about them. Unlike the formerly prosperous’ sugar islands to the south, the Balmeul lel environment has attracted little attention from scientists. This has been particularly true for remote "Out Islands"“ such as Cat Island (Figure 2). Apart from some botanical exploration at the beginning of the present cen- tury (Britton, 1907) and a study of land snails in the 1930s (Clench, 1938), virtually no scientific research of any kind had been carried out on Cat Island before the present study was started in the summer of 1966. Lind's study of coastal landforms began shortly afterwards and his findings have since been published (1969). Cat Island provides a comparatively simple setting for a study of man and vegetation change. It is only 250 square kilometers in area and has a subdued relief, the highest elevation being just over 60 meters. Because of its small Size and low elevation it is climatically more or less. uni- form. The differences that do exist are due to differences in exposure; the eastern and southern coasts face the trade Winds, while the western coast is comparatively sheltered. As far as bedrock geology is concerned, again there is lit- tle variation; the entire island is composed of virtually pure calcium carbonate. There is, however, some local vari- ation in lithology which in large part reflects the degree to which the limestone has become indurated by exposure to the atmosphere. Also, there is considerable local variation I. IhiS 1S not strictly true in certain areas of biolo- gy and geology, as a lengthy bibliography compiled by Gillis et al. (1976) indicates. 2. The term "Out Island" in the Bahamas refers to is- lands other than New Providence, the seat of the capi- tal, Nassau. 12 yueg eweyeg }ea1h @USPIAOId "NG 4 3NI1 WOHILV4 OL yvaiwold yueg eweye ayy ap of the Bahama Islands M ile Figure 13 in the degree to which the limestone surface has. been dissected by solution. In some areas there are numerous potholes; an others the surface is still more or less intact. These” local differences’ in lithology and relief have an important influence on the composition and structure of the vegetation. Insularity Cat Island is located in the east central part of the Bahamian archipelago, less than a hundred miles north of the Tropic of Cancer (Figure 1). Like most Bahamian islands it fringes the windward margin of a shallowly submerged bank. Together the island and the bank form an easterly arm of the Great Bahama Bank which projects out into the deep water of the Atlantic. Off the northern, eastern and southern coasts preeipitous slopes descend to depths of more than 2,500 fathoms. To the west ten miles of bank lie between the island and the comparatively shallow Exuma Sound, only 1000 fathoms deep. To the northwest a narrow submarine ridge connects Cat Island with Eleuthera and the Great Bahama Bank. There is, however, no evidence of any geologically recent land connection between the Great Bahama Bank and either Florida or Cuba, so in this sense, Cat Island is an oceanic island. On the other hand, the sea level fluctuations of the Pleistocene drastically changed the relative areas of land and sea. A fall in sea level of only 10 fathoms, such as occurred several times during the Pleistocene, would double the present land area of the island and make it a peninsula of the Great Bahama Bank (Figure 1). For the Bahamas as a whole it would increase the land area from 5,400 to 60,000 Square miles and would make the minimum salt water distance to Cuba only 10 miles and to Florida 50 miles. Clearly, Cat Island's insularity has varied significantly in the recent geological past; even today this is not an oceanic island to the same degree as remote islands such as the Hawaiian islands or the Galapagos. Cat Island's insularity is also qualified by the proximity of other islands in the archi- pelago. Geology and Geomorphology The shallow water of the Bahama banks provides an ideal environment for the accumulation of a great variety of car- bonate sediments, and borehole data indicate that similar sediments have been accumulating in the area since at least the Early Cretaceous and possibly the Palaeozoic (Lynts, 14 CIITA ADAG NS INDUSTRIOUS HILL STEPHENSON THE COVE TEA BAY KNOWLES SMITH BAY LOCATION MAP CAT ISLAND [__] BLE Hotes McQUEENS DEVIL'S POINT Figure 2. BAIN TOWN ORANGE CREEK ARTHURS TOWN DUMFRIES WILSON BAY BENNETS HARBOUR 0) Q FREETOWN THE BIGHT THE BLUFF DOUDS MOSSTOWN OLD BIGHT ZANICLES BAIN TOWN PORT HOWE Location Map of Cat Island 15 1970: 1227). This accumulation has been made possible by a long-continued subsidence, and indicates that the Bahamian environment has in the long term been a remarkably stable one (Newell and Rigby, 1957). Even so, the present Bahamian islands are in a geologi- eal sense very young. The surface rocks are aeolianites and shallow- water marine sediments of late Pleistocene or Holo- eene age. On Cat at least three age-surfaces can be identi- fied. They are easily recognized in the field by the degree to which the limestone has been indurated and potholed. Because of the homogeneous nature of the bedrock, differ- ences in surface characteristics are important in determin- ing the character of the vegetation. For this and other reasons that will be discussed later it was decided to use landform types as a framework within which to analyse vege- tation change. Although some of the details of the origins of Bahamian landforms are still unclear, the general pattern seems plain; furthermore, it is a pattern that repeats itself on all the larger islands.’ What follows here is a brief account of the sequence of landform types observed on Cat Island. Figure 3 represents a geomorphological cross- section of the island and Figure 4 shows the distribution of the more important habitat-types. Dune Ridges. On the windward side of the island three discontinuous dune systems run sub-parallel to the present coast. The oldest lies furthest inland and is usually higher than the other two, reaching just over 60 meters in places. The aeo- lianite is indurated to an unknown depth, and its surface is pockmarked with potholes. Although no dates are available for the age of these dunes, they may have been formed during the Yarmouth or early Sangamon interglacial. That they were partly submerged prior to the present sea level rise is shown by a wave-cut bench at about 1 meter above the present high tide mark. Fossil coral was occasionally observed at about the same elevation, although nowhere was it as common as on other Bahamian islands, such as Andros and Mayaguana. Isotope dates for presumably synchronous corals on the Florida Keys range between 80,000 and 150,000 years (Broecker and Thurber, 1965; Newell, 1965). A mile or so to the east of the older dunes iS a dune System of intermediate age. These dunes are generally 3. Doran (1955) has provided a useful account of the landforms of the southeastern Bahamas. . In some areas, as for example behind the northern coast of the island, the intermediate-age dunes have overridden the older dunes. 16 tet 0 1 2 3km. 3 4 5 6km. WINDWARD COAST =3=£———————> Figure 3 A Geomorphological cross-section of Cat Island AGE LANDFORM TYPE HABITAT TYPE Holocene a Marine Beds Offshore b Beach Ridges Whiteland ec Dune Ridges Whiteland Young a Marine Plains Flatland Pleistocene b Beach Ridges Blackland e Dune Ridges Blackland Old a Marine Plains Flatland Pleistocene b Beach Ridges Blackland ec Dune Ridges Blackland ee ee ee ee me ee ee ee ee EE ET AT ST HABITAT TYPES E.] WHITELAND FLATLAND Boas]. BLACKLAND SAVANNA HS MANGRoveE Figure 4. Map of Cat Island showing Major Habitat Types 17 18 between 15 and 30 meters high. Their surfaces are also indurated, but lack the numerous’ potholes of the older dunes. In localized areas the limestone is loosely consoli- dated and breaks down to produce a sandy soil. On the ground, the intermediate dune system has a comparatively fresh appearance, with steep lee slopes and gentle dip slopes easily distinguished. Also Pleistocene in age, they do not show any evidence of marine erosion apart from that which has accompanied the recent rise in sea level. Directly behind the windward coast is the youngest dune sys- tem. These dunes rarely reach more than 15 meters in height and are usually around 6 meters high. The lime sand has been only shallowly indurated and is easily kept loose by cultivation. Like the two older dune systems, the younger dunes are fossil landforms, and erosion rather than dune formation characterizes the present coast. These Holocene dunes were studied in detail by Lind, who obtained dates for them of between 4,000 and 500 years (Lind, 1969: 126). Casual inspection with a hand lens suggested, all three dune systems are composed of similar carbonate sediments. Oolites are particularly important, with fragments of shell, coral, and calcareous algae also being common. Fossils are not abundant, although in certain layers shells of the land Snails Cerion and Cepolis can be found. The differences in the surface characteristics of the three dune systems are important as far as the cultivation of crops is concerned and aS a consequence they are recognized locally as distinc-— tive habitat-types. The two Pleistocene dune systems are known collectively as "the blackland," while the Holocene dunes are called "the whiteland". Beach Ridges. On the leeward side of the island, water has been a more important depositional agent than wind, and here beach ridges are the most prominent land form. They curve round in multiple series independent of whatever was the prevail- ing wind direction at the time they were being formed. Their distribution has apparently been influenced by the movement of currents in the lee of the dunes. Again three age surfaces can be distinguished, each one corresponding to a dune system on the windward side of the island (Figure 3). Collectively they differ from the dunes both in their lower elevation and in the symmetry of their cross profiles. In terms of surface characteristics, ~eacu beach ridge system is much the same as its equivalent dune System, and to this extent offers similar opportunities for plant —lsher This is true of cultivated aswell as wala plants, and as a result the local people make the same 19 distinction between the younger, older, and intermediate- aged surfaces as they do on the dunes. The former is’ known as "the whiteland," the other two "the blackland". Lagoonal Plains. Between the dunes, and between the dunes and _ beach ridges, are two comparatively level surfaces. Their origin is indicated by the marine shells they contain, including the conch Strombus. Apparently they are of different ages. The higher surface at about 6 meters above sea level is rid- dled with potholes, whereas the lower one, only a few meters above sea level, is comparatively intact. The higher sur- face is probably the same in age as the oldest dune/beach- ridge system while the lower surface corresponds to the intermediate dune/beach-ridge system. A third surface, corresponding to the youngest dune/beach-ridge system, can be seen below present sea level at an average depth of about 3 fathoms. Unlike the cross-bedded aeolianite that makes up the dunes, these marine sediments have bedding planes that are elose to horizontal and as a result the land surface tends to be much smoother. Especially level areas are known locally as "platey land" and, as will be shown later, have a distinctive vegetation cover. The plains are not entirely flat, being characterized by low undulations with wave- lengths in the order of 1 x 10 meters. The proximity of the water table in low-lying areas means that slight changes in elevation can produce sharp changes in vegetation. soils The soils of the island are thin and discontinuous. For the most part they vary according to the age of the land Surface and the extent to which it has been disturbed by man. Mooney (1905), in the only comprehensive account of Bahamian soils yet available, lists four soil types as being present on Cat: (1) Coral sand, (2) Bahama Black Loam, and (3) Bahama Red Loam, (4) Bahama Marl. The coral sand type refers to the soil found on the Holocene sand dunes and ridges, or in local terms, the whiteland. The term "coral sand" is a misnomer since coral fragments form only a very small percentage of these sedi- ments. The whiteland soil type is immature in the _ sense that there is little profile development. The A1 horizon is very thin, rarely more than a few inches thick if it is present at all. In most parts of the island, cultivation has disturbed whatever profile had developed in the pre- 20 agricultural period.” According to Mooney (1905: 157), this soil is quite rich in potash, phosphates, and nitrates. How- ever, a more recent government report states that the white- land soil is poor in nutrients and too droughty for success- ful agriculture (Anonymous, 1960). The Bahama Black Loam is the most widespread soil type on the island. It covers all but a few areas of the indurated Pleistocene surfaces. Like the whiteland soil it has been severely disturbed by cultivation and is rarely more than a few inches thick. During cultivation it col- lects in pockets and crevices in the limstone surface, leav- ing large areas of bare rock exposed. According to Mooney (1905: 158), it is a residual soil derived from the underly= ing limestone by weathering. An alternative explanation would be that it consists of broken-down organic matter. Because of the capillary rise of water through the’ underly- ing limestone, this soil type is less droughty than the whiteland soils (Anonymous, 1960: 2). It is recognized as the most productive soil type on the island and is therefore frequently cultivated. The Bahama Red Loam has a_ restricted distribution on Cat Island. It is found on the older dune and beach ridge Systems and in the land-locked depressions between them. This red soil is almost certainly a fossil soil that formed under different weathering conditions during the Pleisto- cene. Similar fossil soils have been described in several areas of the West Indies (Ruhe et al., 1961; Kaye, 1959). It is a lateritie soil rich in iron and aluminum and has been derived from the underlying limestone by weathering. Although it appears to be a loam in the field, particle size analysis has shown it to be a clay (Ahmad and Jones, 1969). This is also indicated by the way in which it floods after rain and becomes hard and compacted when dry. According to Mooney (1905: 166), it is rich in nutrients, particularly phosphates, nitrates, and potash, It has a slightly lower pH than the Bahama Black Loam (7.0-7.5 cf7.5-8.0) and probably because of this has been especially favored for pineapple cultivation (Anonymous, 1960: 2). Mooney's fourth soil type, the Bahama Marl has a very restricted distribution on Cat Island, being found ontyeag small areas of low ground east of McQueens settlement (Moo- ney, 1905: plate XXXI). Basically it is a freshwater marl overlying decayed organic material. Its proximity to the water table makes cultivation precarious, and as a result it has only been used for grazing in recent years. 5. Lind (1969) failed to take this into account in his Study of the coastal landforms of the island. 21 The importance of soil differences as far as the compo- sition of the wild vegetation is concerned is not clear. The whiteland is a distinctive habitat-type, but not because of its soil characteristics. The boundary between the Black SOneagdeekedus sOlte Tis usualdy Vdastinet, = but. as Mooney, reported (1905: 165), it does not appear to have any signi- ficance as far as wild vegetation is concerned. The Bahama Marl is too localized to be of major importance. For the island as a whole it is probably a safe assumption that soil differences per se are not an important cause of variation in the structure and composition of wild vegetation. Climate As might be expected from its location, Cat Island has a seasonally wet and dry sub-tropical climate (Table 1). TABLE 1 CLIMATIC DATA 1952-1962 THE BIGHT, CAT ISLAND Bahamas Dept. of Agriculture, cited in Lind (1969: 10) Temperature Rainfall F C inches mm. Jan T2 22.2 1.2 30.5 Feb 74H 23.23 102 30.5 Mar 73 22.8 Io H0.6 Apr 15 23.9 4.5 114.3 May 78 25..6 3.9 99.1 Jun 81 27.2 1.9 48.3 Jul 82 2708 Bo 90.55 Aug 83 BE. 3 5.0 127.0 Sep 82 27 o 8 9.1 231.1 Oct 80 26.7 9.1 231.1 Nov 76 24 4 3.6 91.4 Dec UT 25.0 Qo | 5330 3 Total 39.2 995.7 Mean 7B 25) 56 The basie climatic control is the north Atlantic sub- tropical high pressure cell. In winter the cell moves south and intensifies, limiting convection and reducing rainfall. Occasional outbreaks of modified polar air bring cloudy D2, weather from the north and temperatures fall as low as_ the 60s, but for the most part the sky is clear and temperatures average in the 70s. In summer the high pressure cell moves north, convection is stronger, and trade wind cumulus clouds may tower as high as 50,000 feet. This is the rainy season, during which sporadic convectional showers provide most of the total precipitation. Afternoon temperature may reach the low 90s in July and August, although the comparatively cool trade winds keep the averages in the 80s. As far as plants are concerned, perhaps the most impor- tant aspeet of Cat Island's climate is its’ mildnesse Separated from the continent by the warm waters of the Gulf Stream and under the influence of the trades for most of the year, the island has a truly oceanic climate. Wallace (1902: 5) underestimated this oceanicity when he argued that differences in plant and animal life between Florida and the Bahamas could not be explained by climate. Frost is virtu- ally unknown in the islands, and because of this many plants are able to grow that could never survive for long on the mainland. To the northerner Bahamian vegetation is very definitely tropical. In a negative sense rainfall iS a more important climatic variable than temperature. Although the data presented in Table 1 do not show it, summer droughts are a common occurrence in the Bahamas. In some years no rain may fall for several months and at such a time plant life is dependent upon either the formation of dew or the upward movement of water through the limestone by capillary action. A visitor to Cat Island from the West Indies or Central America would see few plant species that were new to him, but he would probably be surprised by the generally stunted nature of the vegetation. Early students of the Bahamas flora were especially impressed by the many morphological adaptations to drought shown by Bahamian plants, for exam- ple, hairiness and small leathery leaves (Coker, 1905: 215). Deciduousness is also a response to the seasonally-dry cli- mate, and during the "winter" many trees and nearly all the Shallow-rooted bushes lose their leaves. A phenomenon closely related to the occurrence of drought is fire. Unfortunately, the significance of natural fires as far as Cat Island vegetation is concerned is diffi- cult to assess. According to local informants, no lightning fires were known to have occurred within living memory. Even so, the possibility remains that they could occur, especially in the seasonally-flooded savannah. The ever- green woodland could also burn naturally if there were enough dry fuel available on the ground. This is rarely the case today, although it may have been more common in the 23 past. The significance of natural fires in the Everglades area of South Florida has been recently emphasised by Craig- head (1971) and Robertson (1962). Another important aperiodic climatic variable is the frequency of hurricane strength winds. In 186 years, 12 hurricane eyes have passed over the island, an average of one every 16 years (Lind, 1969: 132). Although no studies have been done on the effects of hurricanes on Bahamian vegetation, the conclusions drawn from studies in similar areas elsewhere are probably applicable (Craighead, 1962, 1964; Stoddart, 1965; Sauer, 1962; Wadsworth and Englerth, 1959). Native species are usually quick to recover after the storm. Flexible trunks and branches, bushy habits, pho- tosynthetic bark, and deep root systems are characteristic of many Bahamian species. Hydrology The porosity of the limestone exaggerates the droughti- ness of the surface, but on the other hand it means that rainfall is not quickly lost as run-off. In the indurated Pleistocene dunes, the water table may be at a considerable depth below the surface. At the Devils Point settlement, for example, deep wells are required to reach fresh water. In the summer of 1970 the water table was 12 meters below the ground surface and only 3 meters above sea level. Plant life on these hills must therefore rely upon water from the unsaturated zone. In the loosely-consolidated younger dunes and beach ridges the water table is very close to sea level. _ Although there are no permanent fresh water streams on the island, several low-lying, land-locked depressions are flooded after heavy rains in the summer and autumn. This flooding makes tree growth impossible and means that slight changes in elevation can cause sharp changes in vegetation. In these areas the location of the water table has an impor- tant influence on the character of the vegetation. The Environment as a Whole A low, limestone island such as Cat appears to provide a rather difficult environment for plant life. Droughts, hurricanes, and an apparently infertile limestone surface all combine to place limitations on plant growth, and at first acquaintance it is somewhat surprising that anything ean grow here, wild or cultivated. Yet, paradoxically, the island is floristically rich. In an area of only 250 square kilometers, probably a thousand species of vascular plants 24 are present.© In large part this diversity must reflect the fact that the low, limestone island is an ancient environ- ment. In the area that is now the Bahamas, islands similar to Cat have probably been discontinuously present since at least the early Cretaceous. Plants and animals have there- fore had along time to adapt to this sort of Setting jue is interesting to note here that of 60 genera identified in the Wilcox Flora of the Southeastern United States, which dates to the early Eocene (Berry, 1930), roughly 40 percent are living in the Bahamas today. This is not to suggest as the nineteenth-century authorities might have done, that Bahamian vegetation has survived undisturbed since the early Tertiary. The Bahama islands as they are today are geologi- cally young. Furthermore, the sea level oscillations of the Pleistocene must have caused drastic changes in hydrological conditions and corresponding changes in vegetation indepen- dent of any changes in regional climate. On a shorter time scale, natural disturbance, in the form of dune formation and erosion, flooding by salt and fresh water, hurricanes, and lightning fires, has undoubtedly played an important role in the development of Bahamaian vegetation. How this vegetation has been able to withstand the new types of dis- turbance introduced by man is the main concern of the present study. 6. This eStimate is based on the Bahama Flora (Britton and Millspaugh, 1920:vii), which lists 995 speriataa phytes and 33 pteridophytes for the archipelago as whole. These totals, however, only include cn tivaeee species such as have Shown "a strong tendency to become spontaneous." 25 IV. THE PRESENT VEGETATION: A GENERAL VIEW Ideally, a study of man and vegetation change would begin with the vegetation as it was before man arrived and then trace the changes that have occurred as a result of his presence. In practice, even for recently settled islands, this is rarely possible. The problem, of course, is that there are so few accurate accounts of aboriginal vegetation. Certainly for Cat Island there is nothing. The Arawaks left no written record, and the early European accounts are disappointingly vague. An alternative approach was there- fore called for. The one that is followed here is similar to that used by Harris in his study of the Outer Leeward Islands (1965). The present vegetation is described first in general terms. This general account then provides’ the basis for the later, detailed analysis of man's impact. In most respects, the vegetation of Cat Island is simi- lar to that of the rest of the Bahamas and indeed to that of Similar environments in the New World subtropics as a whole.’ A low, largely evergreen woodland covers all but the lowest ground, where it gives way to either salt-tolerant vegetation or seasonally-flooded savanna. Because of the island's small size and low relief, climate is not an impor- tant cause of variation in the composition or structure of the vegetation. The differences that do exist can largely be attributed to disturbance by man or to differences in topography, lithology and drainage. For purposes of general description, therefore, the vegetation was classified indirectly on the basis of landform or habitat- types. The resulting classification is shown in Table 2. It has the advantage that it avoids the arbitrary problem of classify- ing vegetation itself and at the same time provides a coherent framework within which to study the processes of vegetation change. The Original intention, was to study man's impact on the vegetation of the entire island. However, it soon became clear that such a broad approach would have to be of limited depth and it was therefore decided to restrict the study to an analysis of man's modification of the evergreen woodland, or coppice" as it is locally known. Unlike the vegetation of salt water habitats and seasonally-flooded T. It differs from several Bahamian Islands (New Provi- dence, Abaco, Andros, Grand Bahama, Caicos) in that the Caribbean pine (Pinus caribaea) is not present. There is no obvious @€xplanation for the anomalous distribu- tion of pine in the Bahamas. 26 pueTpoom pueTye ta pue TAoe Td sosptu SouoysaUTT yoeresq auTueuw pue saunp dUud004ST2ETd Q@Uds004STATd GAaNIVYd-T14aM eAOUZUeW aAOUZUCW szeTd Ssuoogsey TePtL SOUOYSOAUWTT spues auTuew auTuew aus004STATd BUdOOTOH PUUBARS ysuew PUeTEITUM ePUUBARS ysueyW sosptu yoeaq SOUOFYSOWTT pue saunp auTuew sqTueuw QUSDOTOH QUd004STITd GaqdOOTA-ATIVNOSVAS SLVIIGVH YALVM HSAYd °d GNWINI qnuos puedu4s qnuosg pueuys sesptu sosptd yoeaq yoeaq pue saunp pue saunp BUds004STOTd aUds0OTOH TWISVOO SIVIIGVH YALVM LIVS °V SddAL LVLIGVH GNV NOILVIASSA e Adee aUSDOTOH NOILVLIA9SA SAWVN TV9O1 ddAL LVLIGVH SOVNIVUC NOILVLADAA SSWYN TV¥OO1 adAL LVIIGVH LOAdSV ail areas, the woodland occupies ground on which cultivated crops can be grown, and as a result it has been drastically disturbed by man. In the general account that follows, the comparatively undisturbed vegetation of the uncultivated areas is described first, since it provid¢és the background for the later discussion of the woodland.“ Figures 5 through 15 illustrate the general aspect of the vegetation of the uncultivated areas. Salt Water Habitats Salt water is never far from the surface in the Baha- mas. No part of Cat Island is more than 3 kilometers from the sea, and the fresh water lens at its deepest is probably not more than 30 meters thick. All the larger lakes are Saline and tidal. Around the margins of these lakes there is a transitional zone of brackish water, the width of which varies from a few meters to over a kilometer. Its extent largely a function of topography and bedrock characteris- tics. The vegetation of the salt-water habitats is distinc- tive, few of the salt-tolerant species are adapted to life in a fresh-water environment. None of the species present is peculiar to Cat Island and most of them are widely dis- tributed throughout the New World sub- tropics. The differ- ences in vegetation within the salt-water habitats are largely to differences in parent material and exposure. The latter is most important, and accounts for the contrast between the vegetation of the coast and that of the tidal flats and lagoon margins. Coastal Environments The coastal environment presents problems that few Species have been able to overcome. The plants that live here are adapted to periodic flooding by salt water, salt Spray, high evaporation rates, and geomorphic instability. Soils are thin or non-existent and the species present are capable of growing in a raw sand or bare rock substrate. The nature of the substrate is the main determinant of local 2. Most of the species Listed in this account were col- lected in the field. A voucher collection including some 600 specimens has been deposited in the University of Wisconsin herbarium, and an almost complete dupli- cate set in the herbarium of the Arnold Arboretum, Har- vard University. A systematic list of the species en- countered is given as Appendix I, and a list of local names and their scientific equivalents as Appendix II. 28 Figure 5. Holocene beach-ridges at North Bird Point. Mal- latonia gnaphaloides is in the central foreground, Suriana maritima in the lower left. Herbaceous cover consists of Uniola ~ paniculata and Iva imbricata. The coconuts in the background are planted. Figure 6. Holocene dune coast east of Stevenson. Large thickets are primarily Coccoloba uvifera. In the foreground is Hymenocallis declinata together with Uniola Paniculata and Ipomoea | pes caprae. 29 variation in floristic composition. Holocene Dunes and Beach Ridges. Most of the dunes and beach ridges that fringe the coast were formed during the postglacial rise in sea level. They are composed of a variety of carbonate sediments that are as yet poorly conso- lidated and are in fact being constantly eroded and redepo- Sited, especially along the exposed east coast. This is a particularly unstable habitat and as one would expect, herbs, vines, and shrubs are especially important. The zonation of species that characterizes so many tropical coasts is not always evident on Cat Island. Dis- turbance by waves, wind, and man usually prevents the estab- lishment of any stable pattern. In several exposed areas along the eastern and northern coasts the sand is virtually bare of shrubs. Here sea oats (Uniola paniculata) and the sea lily (Hymenocallis declinata) are common. The sand is usually crisscrossed by trailing vines, Ipomoea pes-caprae, Cassytha filiformis, Canavalia maritima, and the bay mari- gold (Ambrosia hispida). several grasses (Paspalum vagina- tum, Cenchrus tribuloides, Distichlis spicata) and succu- Tents (Sesuvium portulacastrum, Iva imbricata, Chamaesyce mesembrianthemifolia, Cakile lanceolata). Not all of these Species are likely to be found in any one locality, but all are common along the coast. In other areas a thicket of evergreen shrubs reaches down to the edge of the beach. Probably the two most common species here are the sea grape (Coccoloba uvifera) and bay cedar (Suriana maritima). Also characteristic of the coastal thickets are Sceaevola plumieri, Mallatonia gnaphaloides and Guilandina bonduc. In some areas along the comparatively quiet western coast the Holocene sands have become partially indurated. The harder surface offers different opportunities for plant life and is characterized by different species combinations. This transitional type of habitat can in fact be included in the next sub-heading. Pleistocene Dunes and Beach Ridges. The indurated Pleistocene surfaces offer a more stable habitat for plant life and, the vines and stoloniferous grasses are less com- mon here. Fleshy beach plants such as Sesuvium portulacas- trum are occasionally seen but for the most part low xero- phytic shrubs provide the only cover. Two very common Species are Strumpfia maritima and Rachicallis americana. Suriana maritima is sometimes present as also are Antirrhea myrtifolia, Erithalis fruticosa, Ernodea littoralis, Eugenia Tongpipes, Jaquinia keyensis and Bumelia retusa. The lime- Stone surface is very droughty and even on sheltered sites the shrubs are rarely more than 2 meters tall. 30 Tidal Flats and Lagoon Margins. Along the exposed eastern coast no plants can _ survive for long in the inter-tidal zone. However, this is not the case along the more sheltered leeward coast. Here is an extensive area of tidal flat most of which occupied by mangroves. All four New World mangrove species are native to Cat Island, the red mangrove (Rhizophora mangle), the black mangrove (Avicennia germinans), the white mangrove (Laguncularia racemosa), and buttonwood (Conocarpus erecta). Rhizophora is by far the most common of the four, ranging from the high tide mark to the edges of the tidal channels. On the flats it rarely reaches more than a meter in height, whereas along the tidal channels it may reach 2 to 3 meters. Avicennia, more tree-like in habit and lecking the _ prop roots, appears to prefer a more stable substrate than lime mud. Characteristically, it is seen fairly ‘close Sto Mine high tide mark or in an area where the limestone bedrock is close to the surface. Laguncularia is the rarest of =the four and appears similar in habitat preferences to Avicen- nia. Conocarpus is very definitely restricted to a narrow zone just above the high tide mark. Also, salt-tolerant herbs such as Batis maritima, Salicornia perennis and Borri- chia arborescens are frequently found just above the high tide mark. Along the central axis of the island and behind the coastal dune barriers are several salt water lagoons, some narrowly onnected to the sea, others completely landlocked.’ For reasons that are not immediately apparent the lagoons vary considerably in their physical and biologi- cal characteristics. Some are comparatively clear while others have the consistency of thick soup due to the accumu- lation of algae. The vegetation around their margins, how- ever,: is similar to that found along the coast or on the tidal flats. and floristic differences can be largely attributed to differences in substrate. Unconsolidated sediments, such aS cover a wide area around the Orange Creek Blue Hole (Figure 2), are usually characterized by Rhizophora and other typical tidal flat species. On the other hand, salt-tolerant shrubs such as Rachicallis, Strumpfia, and Conocarpus are more common where the shoreline is rocky. This generalization does not always hold true, as for exam- ple along the rocky shoreline of the Blue Hole east of Dum- fries, where there is a forest of pure Rhizophora, with 3. several Of the deeper Iagoons are known as "blue holes." Some are thought to be inhabited by sea mon- sters and are therefore avoided by the local people. 31 Figure 7. Leeward limestone coast at Wilson Bay. Species present are Conocarpus erecta, aevola plumierii, Coccoloba Figure 8. Exposed limestone coast south of north Bird Point. Suriana maritima showing the effects of exposure to the trade winds. 32 Figure 9. Rhizo- phora mangle at the Dumfries Blue Hole. Epiphytic bromeliads are the only other vascular plants present here. Figure 10. Margin of a tidal channel south of Dumfries. Avicennia nitida dominates the foreground, its pneumato- phores are growing in a meter of peat and marl. The most common species in the background is Conocarpus erecta. trees as high as 8 meters, growing on peat that was measured to be at least 6 meters thick in places. Samples were taken from this peat for pollen analysis, but unfortunately they proved to be barren. Man's Impact on the Vegetation of Salt-Water Habitats. For the most part the vegetation of salt-water habitats has not been disturbed by man. Only near the settlements on the sheltered leeward coast has really significant modifica- tion taken place. Here mangrove swamps in tidal inlets have been locally cleared or filled to reduce the mosquito prob- lem and generally improve sanitary conditions. Rhizophora is cut for charcoal, although not in sufficient quantities to affect its overall distribution. Likewise Conocarpus has not become noticeably rare around the settlements even though it is the most popular source of firewood. A few horses are grazed on the whiteland, although it is doubtful that they have had much of an impact on the coa- stal vegetation as most of the common species are not really palatable. Fires were occasionally observed to have spread down to the coast from inland fields; however, their effect has probably been small in the long term. As far as the invasion of alien plants is concerned, only the Australian Pine (Casuarina equisetifolia) has been really successful. It was introduced to the island probably less than a hundred and fifty years ago and has since spread along most of the sheltered west coast. It is planted in the settlements aS an ornamental and occasionally on the whiteland as a_ windbreak. From these plantings it has rapidly colonized a narrow zone just above the high tide mark. Characteristically, it carpets the ground with a eover of needles that few species can cope with and has replaced the native plants over a considerable section of the leeward coast. On the windward coast it has been less successful, possibly because erosion is more active there. Apart from Casuarina, the only other exotics that have become established are Spanish Bayonet (Yucca aloifolia) and the Mahoe (Thespesia populnea), neither of which have spread far outside of the settlements. The coconut (Cocos nuci- fera), although often seen in the beach drift, has for some reason failed to establish itself spontaneously on Cat Island. The same is true of the Indian Almond (Terminalia catappa). In general, the vegetation of salt water habitats has proved remarkably resistant to disturbance by man. Except 34 Figure 11. Competition between Casuarina equisetifolia and Gundlachia corymbosa. The accumulation of Casuarina needles has somehow caused the death of the Gundlachia bushes. Crab holes are evident in the foreground. 35 on the leeward coast where Casuarina has become so impor- tant, there is little evidence of man's intervention. Seasonally-Flooded Freshwater Habitats Over most of the island fresh water is quickly absorbed at the surface. However, in low-lying areas, particularly land-locked areas, the surface may be flooded to a depth of several feet after heavy summer rains. The water may stand on the ground for several days before being absorbed by the limestone. These floods, together with drought in the dry season, provide a combination that few perennials can with- stand. Woody species may survive long enough to become saplings, but in the long-term, tree growth is not possible. The vegetation of these areas is in many respects similar to that of the Florida Everglades and to seasonally- flooded Savannas in general. The largest area of savanna on Cat Island is just to the east of McQueens (Figure 2). To the north and east it is bounded by a Pleistocene dune ridge whose fairly steep slopes provide a sharp limit to the flooded area, whereas to the south and west a gradual rise through a series of low ridges produces a more broken transition. The ridges appear as islands, or "hammocks," of evergreen woodland surrounded by savanna. With increasing elevation the woodland increases in area at the expense of the savanna, until the distribution is reversed and small pockets of savanna sur- vive in depressions surrounded by woodland. These in turn disappear on the higher ground. In the lower parts of the larger savannas fresh water may be present throughout the year. These marshy areas are characterized by Eleocharis caribaea, sawgrass (Cladium jamaicense), and cattail (Typha domingensis). In seasonal ponds several aquatics were collected, including Nymphaea ampla var. pulchela, Potamogeton heterophyllus, and Pros- perinaca platycarpa. Around the margins of these ponds Echinodorus berteroi Jussiaea suffruticosa, Centella erecta, Lippia stoechadifolia are commonly encountered, together with the sedges Albilgaardia monostachya, Dichromena eolorata and Rhynchospora cyperoides. Around the edges of the smaller, potholes the pond apple (Annona glabra) is often found. In the drier areas several species of herbs are charac- teristically present; for example, Pluchea rosea, Buchnera elongata, Eustoma exaltatum, Sabbatia stellaris, Eupatorium villosum, Walthera indica, Cynoctonium mitreola, and Linum bahamense. The savannas are not grasslands in the true 36 Figure 12. A "hammock" in the McQueen's savanna. A poison- wood (Metopium toxiferum) has become established on a small ridge. On either Side are palmettoes (Sabal palmetto); in the foreground Pluchea rosea, Andropogon gracilis, and Aris- tida ternipes. Figure 13. A smadli seasonally-flooded depression just above the McQueen's’ savanna. The white bracts of Dicromena colorata are visible in the foreground, as are saplings Of Tabebuia bahamensis. In the background are Coccoloba uvi- —_— - 37 sense of the word, as grasses cover only a small percentage of the total area. Andropogon gracilis and Aristida ter- nipes were the only grasses seen, and both were rare. Around the edges of the savannas, pioneer shrubs’ and small trees often obtain temporary foothold; especially com- mon are horsebush (Gundlachia corymbosa), sweet gale (Myrica cerifera), coco plum (Chrysobalanus icaco), beefwood (Torru- bia Longifolia), and five finger (Tabebuia bahamensis). Palmettoes (Sabal palmetto) mark the upper margins of the flooded ground and in turn give way to evergreen woodland on the well-drained sites. Man's Impact on the Seasonally-Flooded Freshwater Habitats. As in the salt-water habitats, the growth of crops is not possible in the seasonally-flooded areas, and conse- quently the native vegetation has been relatively undis- turbed. Fire has probably been the main cause of change. 4 All but the youngest palmettoes show scorch marks on their trunks (Figure 14). According to local information, savanna fires served no useful purpose and were simply the work of children. There were no reports of any lightning fires, but this does not mean they never occur. In the Everglades, for example, they are not uncommon events (Robertson, 1962). Grazing by horses was observed in the more accessible areas, although what species were affected was not deter- mined. The savannas are generally regarded as poor pasture because they dry out during the winter months. No alien species appear to have established themselves in the seasonally-flooded areas. Casuarina, which has colonized analogous areas in the Everglades (Egler, 1952), is conspicuously absent. On the other hand, the McQueens Savanna does provide a good example of man having locally extended the range of a native species. As can be seen from Figure 15, the trail through the savanna passes through an avenue of palmettoes. These have grown from berries dropped during the harvesting of palmetto inflorescences for hog feed. For the most part, however, the seasonally-flooded Savannas have been of little value to man, and for this rea- son they have remained relatively undisturbed. YT. There is no evidence to suggest that the area of sa- vannas has been enlarged by repeated fires. In all eases the woodland comes down to the edge of the sea- sonally flooded areas. 38 Figure 14. Palmet- toes (Sabal pal- metto) recently / burned by children. The palmettoes were not killed by the fire and were actu- ally sprouting new leaves at the time the photograph was taken. Figure 15. "Footpath distribution" of palmettoes near McQueens. The individuals alongside the trail have grown from accidentally dropped fruits, that had been gathered for hog feed. 39 Well-Drained Freshwater Habitats. Most of the island is covered by a low, largely ever- green woodland, locally known as "the coppice". And as its local name implies, it has been drastically disturbed by man. Seen from above it has the appearance of a patchwork quilt, the abandoned fields of shifting agriculturalists Showing progressively darker shades of green according to their age (Figure 16). On the ground, sharp changes in height and floristic composition give further clues as to the history of disturbance. It was this obviously disturbed aspect of the woodland that made it particularly interesting in the context of the present study In mature woodland on the more mesic sites, the dom- inant trees reach a height of 10 meters or so, with the larger trees having diameters at breast height of from 20 to 30 centimeters. On dry sites, where the fresh water lens is thin or the limestone surface is steep, the woodland degen- erates into a cactus’ scrub. Here the shrubby trees are rarely more than 3 meters tall, and the dildo cactus (Cephalocereus millspaughii) adds to the xerophytic charac- ter of the vegetation. As a vegetation type the Cat Island woodland is prob- ably equivalent to Beard's (1955) "Evergreen Bushland" and dry "Evergreen Thicket". This kind of vegetation is fre- quently encountered in low limestone environments throughout the New World sub-tropics. To the northerner the woodland is definitely tropical. The great majority of species present range south to the Greater Antilles or beyond, and very few are found north of the Florida Keys. Furthermore, it is floristically rich. Coker was certainly correct when he reported that: As one passes through a typical Bahamian coppice, different plants are met with at every step. The variety seems interminable and on first acquain- tance one is appalled with the difficulty of becoming acquainted with them (Coker, 1905: 232). In the present study, 120 species of trees and _ shrubs were encountered during systematic sampling, and herbarium eollections were made of 30 more (Appendix I). This floris- tic diversity is to a large extent masked by a remarkably uniform physiognomy. Most of the woody species have small, entire-margined and leathery leaves. This uniformity pro- vides a classic example of convergent evolution, species having adapted in similar ways to the problems of life on a low limestone island. Figure 16. An aerial view of the woodland south of Old Bight. Scale = 1:1,000 41 *7J8T 2U4 OF STQTSTA st ‘puog pay se umouy ATTeOOT Jeyzemyztes Vy auy wWouj 4seo ‘uooset “UMOJ90Ug JO UYIsIOU ABpPTu-eunp auUs004STAEeTd BUTHOOT pueTpOOM ayy JO MeTA “LL aInBtTy 42 On the other hand, the woodland is not entirely uniform in structure or floristie composition. As was indicated above, both vary from place to place, largely as a_ function of the availability of moisture. This in turn as Vargery determined by the location of the water table and the nature of the substrate. Although the former was difficult to identify in the field the latter was not. Therefore it was decided to simplify the analysis by subdividing the woodland into three types on the basis of landform or "habitat" characteristics (Table 2). More specifically the three types are: (1) Holocene dunes and beach ridges; (2) Pleisto- cene marine plains; and (3) Pleistocene dunes and beach ridges--or, in local terminology, the whiteland, flatland, and blackland, respectively. As will become clear later, these habitat-types are not floristically distinct, as many Species are present in all three. Even so, by subdividing the woodland in this way it was possible to control some of the environmental variability that would otherwise have com- plicated any analysis of man's impact. Figures 1 itommeee give a visual impression of the three habitat types. The Whiteland The largest area of whiteland on the island forms a discontinuous strip along the east coast (Figure 4). On the west coast there are localized areas near Orange Creek, Bennet's Harbour, and McQueens. Not all of the whiteland is covered by evergreen woodland. On its seaward margins it is occupied by the _ salt-tolerant vegetation of the coast or tidal flats, and inland by the seasonally-flooded savanna, or other varieties of evergreen woodland, depending upon the location of the water table. The woodland is generally lower on the whiteland than elsewhere. The trees have a characteristically bushy appearance, and are rarely more than 6 meters tall. The number of species present is proportionately lower. There are probably three basic reasons for this impoverishment. First, the loose sand does not have a very high moisture- retention capacity and ground water is not able to rise very easily by capillary action (Anonymous, 1960). Second, the sand provides a rather insecure rooting-medium for trees. so often subject to hurricane-strength winds. Third, and perhaps most important, the whiteland has in most areas. an exposed location on the coast and evapotranspiration rates are therefore high. This is particularly significant on the southern and eastern coasts, which face the persistent trade winds. In spite of the impoverished aspect of the whiteland vegetation, a considerable diversity of species is likely to 43 Figure 18. A whiteland field, the light area in the middle distance, is in the process of being cleared. In this case the fallow period has not been long enough to allow shrubs to become established. Uniola paniculata, Chloris petraea, Cenchrus echinatus and Bidens pilosa are common in the fore- ground. Figure 19. Corchorus hirsutus dominates this whiteland field The bushes are probably about ten years old Coccoloba uvifera is present in the left foreground and ALSO” alia wae distance. 44 be encountered at any one locality. Especially common in areas of older woodland are cassina (Acacia choriophylla), ramshorn (Pithecellobium keyense), dollen plum (Reynosia septentrionalis, poisonwood (Metopium toxiferum), milkberry (Bumelia retusa), and beefwood (Torrubie longifolia). On less favorable sites sea grape (Coccoloba uvifera), coco plum (Chrysobalanus icaco), black torch (Erithalis fru- ticosa), and white torch (Amyris elemifera) may be present. On exposed sites, the woodland degenerates into a coa- stal thicket in which the area of bare sand exceeds the area covered by vegetation. Such eas are occasionally washed by salt water and are therefore not cultivated. Apart from a few remote areas, as for example west of McQueens (Figure 2), the whiteland has been repeatedly cleared and burned for agriculture during the past three hundred years. The wood- land is in fact secondary vegetation in various stages of recovery after clearing, burning, and grazing. The Flatland Between the whiteland and the Pleistocene dunes and beach ridges is the flatland. In total area it accounts for the largest part of the island. On its lower margins it grades into either the whiteland, the seasonally-flooded freshwater areas, or the saline lagoons; on higher ground it is bordered by the Pleistocene dunes and beach ridges. The woodland is generally higher on the flatland than on the whiteland. In relatively undisturbed areas, trees were seen 10 meters in height with diameters at breast height of 20 to 30 centimeters. In contrast to the almost impenetrable whiteland thickets, individual trees are spaced 2 to 3 meters apart . The more common species are pigeon plum (Coccoloba diversifolia), poisonwood (Metopium toxi- ferum), wild tamarind (Lysiloma bahamensis), kamalamey (Bur- sera simaruba), and mastic (Sideroxylon foetidissimum). The ground is covered with a thick accumulation of leaves, which in turn may be covered by a dense growth of terrestrial bromeliads. The trees themselves characteristically support a rich growth of epiphytic orchids and _ bromeliads. The woodland as a whole has a distinctly tropical appearance. Such undisturbed areas are comparatively rare, as most of the flatland has been repeatedly cleared and burned for agriculture. Second-growth woodland in various stages of recovery covers most of this habitat-type. Here a great variety of species are present including hor sebush (Gundlachia corymbosa), granny bush (Croton linearis), jum- bay (Leucaena leucoephala), and soap bush (Corchorus hir- sutus). The individual trees and shrubs are closely spaced, 45 Figure 20. A flatland field close to the edges of the seasonally-flooded savannah. Cultivation is precarious here because of the risk of flooding. Palmetto fronds have been laid out to dry before burning. Figure 21. Second growth on the flatland south-east of Bennet's Harbour. The bushes are three meters tall and probably about twenty years old. The palmetto 46 Figure 22. A severely burned blackland field. Note that many of the bushes in the background are still without leaves. The photograph was taken in June 1967 after an unusually dry dwinter. Figure 23. The same field in 1970, from a slightly age ferent angle. The field had been abandoned in 1969. Recovery is slow probably because of the deep. burn. The sprout in the foreground (Coccoloba diversifolia) is one of the few to be seen. Bae Sp et . - > * & ¢ ee Aa ane * 4 oe 63 - 7 : ¥ bd <* ; - 47 making the woodland almost impossible to penetrate without a machete. Iie RCAanoipya Tn uhannermand asta lGesultea lauxur dant growth of epiphytes is not possible. Vines, however, are common, particularly the troublesome Smilax havanensis, together with Jaquemontia cayensis, and several morning glories (Ipomoea microdactyla, I. acuminata). The Blackland The Pleistocene dunes and beach ridges in a sense form the backbone of the island, and are covered with the highest and floristically most diverse woodland. Except in exposed areas facing the trade winds or in areas where slopes were excessively steep, this is the optimum habitat for tree growth. As on the flatland, comparatively undisturbed woodland was hard to find because most of the blackland has been intensively used for agriculture. However, in a few areas trees were seen on the order of 12 meters tall with diame- ters at breast height of around 30 centimeters. This mature woodland has a distinctly tropical appearance not only because of the great number of species present, but also because of the many epiphytic orchids and bromeliads. The dominant tree species are pigeon plum (Coccoloba diversi- folia), poisonwood (Metopium toxiferum), hog cabbage palm (Pseudophoenix vinifera), mastic (Sideroxylon foetidis- simum), and madeira (Dipholis salicifolia). In older woodland the differences between the flatland and blackla.d are not too obvious. This is not the case in the younger woodland. On the blackland the limestone _ sur- face is riddled with potholes and small crevices. These microhabitats play an important role in slowing soil erosion and reducing evaporation. The broken surface also reduces the severity of the fires set by the shifting agricultural- ists. As will be shown later, the rate of succession is more rapid on the blackland than on any other habitat-type. 49 V. CLEARING AND BURNING FOR AGRICULTURE Cat Island has been discontinuously inhabited for at least a thousand years, during which time the woodland has been repeatedly cleared and burned for agriculture. The history of this disturbance is unfortunately obscure. The impact of agriculture on the woodland is only occasionally referred to in the historical record, and the record itself is understandably thin. What follows here is a brief sum- mary of the evidence that is available. Particular attention is given not so much to the history of agriculture itself as to the way in which agricultural practises have affected the woodland. The Island Arawak (1000 A.D.-1500 A.D.) The first known inhabitants of the Bahamas were the Island Arawak. According to the available archaeological evidence, these people probably reached the islands about 1000 A.D., having left the South American mainland at about the time of Christ (Rouse, 1964; Hoffman, 1967; MacLaury, 1970). At the time of European contact, Arawakan-speaking peoples occupied a wide area, including Amazonia, Central America, and the Caribbean, and although detailed population figures are not known, it appears that a very large number of them, several million according to the Spanish accounts, were in the Greater Antilles, Turks and Caicos, and _ the Bahamas (Rouse, 1964). The Arawak population of the Bahamas, or the Lucayas as the islands were called, was reported by Peter Martyr to be OnmuMewonder of ton ty thousand (Craton. 1968s 939)).)) ih. “that's estimate is correct, the number of people living on Cat Island must have been several thousand. Unfortunately, the Spanish accounts are not reliable on this point. What is eertain is that less than thirty years after Columbus's landing the whole of the Arawak population had been tran- sported to Hispaniola by Spanish slavers (Sauer, C.0., 1966; Craton, 1968). The Island Arawak were skilled agriculturalists, with a long inventory of cultivated plants, including cassava, Sweet potatoes, corn, beans, squashes, and tobacco (Stur- tevant, 1961). They were also expert fishermen and obtained much of their food from the sea. They lived in settled vil- lages ruled by hierarchies of chiefs, made good pottery, and had a relatively elaborate religion centering around the worship of deities known as Zemis (Rouse, 1964: 502). 50 Evidence of the former presence of the Arawaks is widespread on Cat Island. Pottery fragments and shell mid- dens are commonly encountered, and in virtually every cave human bones’ have been found. Some reconnaissance archaeo- logical work has been carried out (MacLaury, 1970), but the data uncovered tell little about Arawak subsistence or to what extent the woodland may have been cleared. The histor- ical record is a more fruitful source of information here. Although Cat Island is rarely referred to specifically, it is possible to draw some general conclusions on the basis of accounts of Arawak populations in the Greater Antilles. The Arawak way of life was remarkably similar throughout the Bahamas and the Greater Antilles (Sauer, C.0., 1966). It seems likely that the Arawaks practiced some form of shifting agriculture. The actual clearing of the woodland was probably done by fire as the thick accumulation of organic matter on the ground could have been easily burned during the dry season. Clearing of mature hardwood trees by felling or even. girdling would have been difficult if not impossible for people equipped with only shell tools. Arawak cultivation in the Greater Antilles involved planting in "conucos," mounds of earth surrounded by stones. After several years of cultivation fields were abandoned because of declining yields and competition from weeds. Just how much of the woodland was cleared in this way is not known. It would seem likely, however, that in five hundred years even a small population of shifting agriculturalists could have cleared a large part if not all of the woodland on a small island such as Cat. That this was in fact the case is suggested in a letter written by a loyalist who set- tled on Cat Island in 1784. He wrote that he had seen rocks piled up in little heaps by the Indians...and it plainly appears by this and other relics daily met with that the inhabi- tants have been very numerous, as there is none or but very little ground but what has been cleared and cultivated. Great quanti- ties of their bones are to this day found in different cavities of the rocks (Eve, 1784). There is an indirect suggestion in the historical record that the Arawak population of the islands at the time of contact was dangerously large. The Spanish accounts describe the Arawak as being close to starvation. Ferdinand Columbus, for example, noted that the amount a Spaniard would consume in a day would last the average Arawak a whole week (Craton, 1968: 24). In view of the crops the Arawaks had at their disposal there can be only two explanations for 51 this. Either the population was too large for the amount of cultivable land available, or there had been a recent crop failure due to a drought or hurricane. Unfortunately Columbus's journal contains little infor- mation about Arawak agriculture or the vegetation of the islands. Apart from the Arawaks themselves, the Bahamas contained nothing of value to the Spanish and therefore received little attention. After the Arawaks had been taken to the mines of Hispaniola, the Bahamas were to remain uninhabited for nearly two hundred years. During this time the woodland presumably recovered to something like its natural state. The English (1703-1834) Although the English settled Eleuthera and New Provi- dence in 1648 and 1666 respectively, Cat Island was at this time too vulnerable to attack from the French and Spanish to make permanent settlement worthwhile. In 1703, however, New Providence was attacked and 120 of the 150 inhabitants fled to Cat Island (Craton, 1968: 93). The exile was only a tem- porary one. In 1723 the Governor reported that "the people of Cat Island have lately quitted that remote place having been so often plundered and disturbed" (Phenney, 1723). It was during this brief period of settlement that Cat Island's reputation as the best agricultural island in the Bahamas was established. In 1730 Governor Rogers reported that settlers without land on St. Christophers were keen to develop sugar plantations on Cat Island, which “all people in general agree is much the best of the Bahama Islands hav- ing large valleys of fine land and plenty of water" (Rogers, 310)! Seven years later settlers from Barbados, the Lee- wards, and the Virgin Islands were also expressing interest in obtaining land on Cat, whose reputation had increased further: Cat Island contains at least as much land fit to cultivate sugar cane upon as Barbados, besides a large quantity of ground fit to produces corn “cotton trees), indigo, ec inigerr; and savannahs or low ground fit to raise and fatten cattle upon. The soil they say is much the same with that of Hispaniola or Cuba Gralte7wellisam, wliesh(e) In 1734 Thomas Coram petitioned the Commissioners for Trade and Plantations, advocating a settlement scheme involving Cat Island and Nova Scotia. He envisioned a trade By triangle in which the Bahamas, specifically Exuma, would provide salt for the cod fisheries in Nova Scotia. The set- tling of Cat Island would provide protection for the salt rakers from the Spaniards at Baracoa and "would otherwise be vastly advantageous to the crown" (Compston, 1918: 73). The Commissioners gave Coram every encouragement, but the settl- ment never took place, apparently because of the unsettled political situation in the Bahamas. For the next sixty years the island was again virtually uninhabited. However, in 1783 settlement began in earnest. In that year the loyalists began to arrive to claim the land grants they had been promised in return for their support of the Crown during the American Revolutionary War. In) ‘total! about two thousand loyalists came to the Bahamas from the former colonies, bringing with them nearly six thousand Slaves (Dunmore, 1789). Of these about 60 loyalists and 500 slaves appear to have settled on Cat (McKinnen, 1804: 198). At this time Cat still had the reputation of being the best agricultural island in the Bahamas (Johnson, 1783), and most of the loyalists who settled on the island were experienced planters from the former colonies. Their main hope was to establish successful plantations based on cotton. With few exceptions, the plantations were established on high ground overlooking the southern and eastern coasts. The blackland and flatland were regarded as the most produc- tive for cotton while the whiteland was used for food crops (Eve, 1784). Very quickly the woodland was cleared and large fields were planted in cotton. By 1788, 2000 acres had been planted on Cat, roughly a quarter of the total Bahamian acreage in cotton (Wylly, 1789). Initial yields were promising, but in 1788 and again in 1791 a large part of the crop was lost because of insect pests, specifically the chenille and red bug (Wylly, 1800). Clearing and plant- ing continued, but the planters were never able to recover their losses. Even before the Emancipation Act of 1834 the loyalists had left the island, leaving their slaves behind (Anonymous, 1840). The impact of loyalist agriculture on the woodland can only be assessed in general terms. It is interesting to note, however, that the planters themselves attributed a large part of their failure to indiscriminate cYearange According to one contemporary account, the loyalists “went to work in the true American way, cleared immense fields, and laid their lands open to every wind" (Wylly, 1880). This "first and most fatal error" accelerated soil erosion and encouraged the spread of insect pests (Kelsall, 1800). Having learned from their mistakes, the planters later cleared smaller fields, but the damage had already been 55) done. As one Cat Island planter pointed out, by 1800 there was little new land left to cultivate, and that which had been "too much exhausted or burned" was slow to recover (Eve, 1800). It seems clear that within the short period of twenty years virtually the whole of the woodland had been eleared. After the plantations were abandoned the woodland must again have reasserted itself, but this time its recovery was hindered by the activities of a now permanent population of shifting agriculturalists, the abandoned slaves. The Free Negro (1834-Present) The loyalists left behind 694 slaves and 55 free negroes (Cameron, 1805), most of whom were given land or allowed to farm the plantation lands on a _ share-cropping basis. After the passing of the Emancipation Act in 1834, some freed slaves bought land from the Crown in either 20- or 40-acre lots, while others continued to rent or simply occupied the land illegally as squatters. In the 1830s and 1840s settlements were established along the east coast of the island. Some, such as Orange Creek and Bennets Harbour, were located where small boats might anchor. For the rest the availability of Crown Land seems to have been the most important factor in determining their location. Emancipation did not mean a sudden change in the way of life of the Bahamian negroes, since most of them had been largely independent for twenty or thirty years. According to the report of a magistrate who visited Cat in the year 1836, all the freed slaves were quite satisfied with their lot. Their only complaint was they had no means of sending their surplus produce to Nassau (Stiles, 1836). Later reports from the island seem to indicate that the free negro population had become well-established. The old plantation grounds were being kept as grazing land, and a small supply of stock and corn was being shipped to Nassau (Anonymous, 1840). By this time (1840), the population was still only 750. Four years later the inhabitants of Cat Island, Rum Cay, and Watlings were close to starvation. An unusually long drought in that year led to the failure of crops and had it not been for emergency supplies from Nassau the peo- ple would have starved. According to Governor Mathew, the famine was partly caused by the very unusual drought, partly by the improvidence of the people, and partly above all by the exhaustion of the scanty soil (Mathew, 1844). 54 The last point is significant insofar’ asi lt samples that a large part of the land had already been cleared. That this was the case is also suggested in the report of an Anglican missionary who visited the island in 1855: I found much difficulty in visiting the people. They are very much scattered with poor roads leading to their dwellings. they, gor our aato their fields pretty early in the morning, which being far from their homes they do not return till sunset (Higgs, 1855). This account suggests that all the accessible land had been already cultivated and that people were being forced to farm in remote areas of the island. A somewhat more optimistic view of life on Cat Island is given in the report of a government surveyor, Thomas Harvey. In his opinion Cat Island was by far the best agricultural island in the Baha- mas, "the soil of the island is excellent and produces fine pineapples" (Harvey, 1858: 23). The negro practice of clearing the woodland and then burning it was almost universally criticized by English and American visitors. A good example of this attitude is con- tained in Johnston's report on the agricultural capabilities of the islands. He visited Cat Island in September 1867 and reported: I was much pleased with my examination of the lands of the Poitier's estate; a portion of this “tract is decidely the best land I sawein the out-islands. How this tract has escaped destruction so long I cannot conceive. But, alas! the work has begun in earnest. I saw very many patches recently cleared, burnt, and others burning, as I passed; cruel! cruel! (Johnston, 1867) The gradual clearing of the woodland in part reflected the necessities of shifting agriculture but was also a response to the needs of a growing population. In 1861 the total number of people on Cat Island was 2,378, while only thirty years later the population had doubled to its highest total ever of just over 5,000 (Sharer, 1955: 92). Unfor- tunately, the historical record provides little insight into the state of the island at this time. According to several of the older inhabitants inter- viewed on the island in 1970, the late nineteenth century was a time of considerable hardship. Not only had the attempt at sisal cultivation failed, but the population of 55 the island was too large to be supported by traditional agricultural methods. All the cultivable land had been cleared at least once and in many areas the fallow period was not long enough to allow an adequate recovery of fertil- ity. The only alternative was emigration, and many of the islanders left Cat for Nassau or the United States. In cer- tain parts of the island, particularly on level ground, the effects of this over-cropping can still be seen. In areas that have not been cultivated for half a century the wood- land still has a degraded appearance. During the present century the population has continued to decline. However,the practice of shifting agriculture has eontinued largely unchanged. An interesting commentary on eonditions in the 1930s was given by the Harvard concholo- gist Clench (1938: 501-502). He reported that the unfor- tunate practice of burning the vegetation was resulting in the rapid disappearance of the soil, and predicted that within a few more generations there would be little or no agriculture possible. According to Clench, most of the woodland being cleared at that time was between 10 and 15 years old, and although the resulting fields were productive for only a year, they were usually farmed for two to five years after each burning. In spite of emigration the pres- Sure of population on the land still appears to have been severe. That this was in fact the case was supported by the accounts of several older inhabitants interviewed on the island in 1970. It has only been within the past twenty years or so that the intensity of shifting cultivation has declined. The development of the tourist industry in Nassau has led to further emigration from Cat and has meant a return flow of money and food to the island. As a result the acreage in eultivation has declined. Even so, shifting agriculture is still practiced in much the same way it has been for the last century and a half. As one farmer expressed it, "I work after the old peoples' dispensations, and I find myself walking in the right track." Contemporary Agriculture Agriculture today is probably less intensive than it has been at any time during the past hundred years. The total population in 1971 was about 3,000, most of whom were either older people or young children. Accurate statistics for acreage cultivated are not available, but a figure of four acres for a family of eight is probably a fair esti- mate. 56 The four acres would consist of five or jsa@xieivelids scattered across the island. One reason for this is that the chances of drought are lessened if fields are widely spaced; on a small island such as Cat, summer thundershowers can be very localized. Another important reason is that the different habitat- types are adapted to different crops. For example, the blackland, which is regarded as the best farming land is especially well-suited to Sweet potatoes, cassava, peanuts, corn, benny seed, and beans. The flat- land, at least in lower areas where moisture is available, is recognized as being especially good for tomatoes, melons, onions, and vegetables in general. Cultivation is, however, a precarious proposition here because of the risk of flood- ing. On the higher flatland areas the problem is drought, and corn and pigeon peas are the most common crops. The whiteland soils are easily worked but also tend to be droughty. Corn and sorghum were the main crops in 1970, although the latter is much less important that it formerly was. The localized areas of red soil are usually planted in pineapples, although cassava, pigeon peas, and peanuts may also be grown. The clay soil is usually too compact for Sweet potatoes. The shallow potholes that are so common in certain flatland and blackland areas are commonly planted with bananas, pawpaws, yams, Sugar cane, and eddoes. The variety of habitat-types is important insofar as it enables the farmer to diversify his crop combinations. In any one year most farmers on the island cultivate two crops, a winter crop which is planted in May and June and harvested in December, and a summer crop which is planted in December and harvested in July and August. The winter crop is the most important in terms of yield, as the summer crop often fails because of droughts. A particular field is usually cultivated for two or three years and may therefore produce four or five crops before being abandoned. The decision as to where to cultivate in any one habitat- type is usually made on the basis of what trees are present in the area. Poisonwood (Metopium toxiferum), pigeon plum (Coccoloba diversifolia), mastic (Mastichoden- dron foetidissimum), and horseflesh (Lysiloma leucoephala) are all regarded as an indication of good soil. On the other hand, wild tamarind (Lysiloma bahamensis), horsebush (Gundlachia corymobosa), and granny bush (Croton linearis) are all poor soil indicators. Also important is the age of the woodland. As a general rule, the older the trees, the more productive will be the soil after clearing and burning. Clearing is done with a machete, usually during the cooler months from January to April. Apart from the ques- tion of fertility, people preferred to clear older trees 57 because they were more easily cut than younger bushy growth. Most of the single women and older men hired professional bush cutters to clear their fields. Fields cleared in this way are distinctive in their regular, usually square, shape. Fields are initially about a quarter of an acre in size, and are expanded later if yields are good. lv @ ele as a large one, several shade trees will be spared. Character- istically, these are mastic (Mastichodendron foetidissimum) or kamalamay (Bursera simaruba). Ii Mearily all” RIGS several trunks are left standing to provide support for the vines. After clearing, the lighter brush is laid out on the ground to dry, while the heavier trunks are pulled to the side orm the field > ethe reason for this being that the’ larger pieces of wood burn at a high temperature and reduce the fertility of the soil Many local farmers take pride in their burning tech- nique. The old tradition has long been criticized by Euro- peans, but is apparently the best way to make nutrients available to the crops. It removes debris, exposes the soil, and has the advantage of killing crop-eating insect larvae. Burning is usually done one or two days after rain in May or June. If the ground is too dry, the burn will be too hot and the productivity of the soil will be reduced. One experienced farmer claimed that the severity of the burn should be determined by the crop that is to be planted. Sweet potatoes, for example, would benefit from a hot burn, as they need more ash. Corn, on the other hand, needs a lighter burn. A very hot burn was also thought to encourage more weeds. On Cat Island, fires very rarely escaped from a field into the uncleared woodland, because the living trees are not dry enough to burn. The woodland is in fact a green firebreak. The only situation in which a wild fire might Start would be where there was a great deal of dry litter on the ground, as for example in an area of older woodland at the end of a severe drought. Frequent clearing and burning prevents the accumulation of litter and therefore reduces the chances of wildfires. A few days after a field has been burned, fertilizer is added to the ashes. A few days after that, ideally just after a rain, seeds and cuttings are planted. Planting is done with the aid of a simple dibbling stick. As an insurance against crop-eating insects and birds, especially the ground dove, several seeds are dropped in each hole. Nearly all farmers were aware of the advantages of seed selection. but few thought it was worth the effort. For most crops the time of planting is determined by the phases of the moon. During the next few months weeding takes a large part of the farmer's time. In the first year of cul- tivation sprout-weeding is the main activity, while in the 58 second and third years the control of herbaceous weeds is more important. If there is no drought the winter crop is usually harvested in November or December. Yields are dif- ficult to estimate because of the irregular methods of har- vesting and widely-scattered fields. Most farmers, however, produce more than enough for their own needs and ship the Surplus to Nassau. At present a field is rarely cultivated for more than two or three years. The pressure on the land is so low that it is easier to clear a new field and start again than to farm intensively. According to local opinion, the minimum required fallow period for sustained yields is on the order of ten years. However, in recent years younger woodland has rarely been cleared, and the fallow period is normally, at least fifteen years. Conclusion In conelusion, the three cultural groups, the Arawak, the loyalists, and the negro peasant farmers, have each had an important impact on the evergreen woodland. The conse- quences of Arawak settlement are unfortunately not well known, but historical evidence suggests they may have been Significant. The loyalists in a period of less than twenty years cleared a large part, if not all, of the woodland. The negro population in a hundred and fifty years of shift- ing agriculture repeatedly cleared and burned the whole of the woodland. At present it is safe to say that all of the woodland has been cleared and burned at least once, and in the accessible areas it has been cleared and burned at least a dozen times. In this sense it is all secondary vegeta- Gaon. What is not certain is just how the woodland has changed as a result of this clearing and burning, which species have become rare or extinct, and which have become more important. ———E——EEEE te —s 59 VI. SELECTIVE CUTTING OF INDIVIDUAL SPECIES Man's exploitation of individual species has had an important influence on the character of the woodland and indeed on the character of Bahamian vegetation as a whole. From the early part of the seventeenth century until the latter part of the nineteenth century the selective cutting of dyewoods, barks, and timber trees was an important activity in the Bahamas particularly during periods of econmic depression. So intensive was this exploitation that several sensitive species became rare. During the present eentury there has been a general decline in the demand for dyewoods, barks and native timber and many of the exploited Species appear to have recovered. Fortunately, the history of selective cutting is more amenable to analysis than the history of clearing and burning. Usually the species involved can be definitely identified and in some cases there is reliable evidence as to their former distribution and abundance. Unfortunately, there are few specific refer- ences to Cat Island, and what follows therefore deals largely with the Bahamas as a whole. Dyewoods The most important of the native dywoods was brasiletto, a shrubby legume valued for the red dye obtained from its heartwood. Apparently three species were exported from the Bahamas under this trade name: Caesalpinia vesi- caria, C. bahamensis and C. reticulata. All three are small trees generally found on rather dry sites. Brasiletto was probably first cut by the Spanish in the sixteenth century (Craton, 1968: 58), although on what scale is not known. In the early seventeenth century English woodcutters from Bermuda were cutting Brasiletto in the Bahamas. The sailing orders for a trading voyage from Lon- don to Barbados in 1650 illustrate well the scope of these early activities. The captain was instructed to search the Bahamas for basiletto, seal oil, ambergris, and wreck goods and take what he found to Barbados or any of the Leewards from where it could be shipped to England. From Barbados he was to return to the Bahamas two, three, or even four’ times if necessary to make the enterprise profitable (Lefroy, SiO), Lil: O08) s T. The three names Listed here are taken from the Baha- ma Flora (Britton and Millspaugh, 1920: 173). 60 In 1670 Simon Robinson, a Bermudan ship captain, reported to the Lords Proprietor.© that New Providence’ had only small quantities of brasiletto whereas "Egsuma had much brasiletto wood...and another island discovered last year also full of brasiletto wood" ) (Robinsons) 1670s Teor According to another Bermudan (Carrell 1670: 475), Jamaica was at that time the chief port for the proceeds of "shal- loping brasiletto, amber (Ambergris) and turtle shell". The effects of this early exploitation were quickly felt and by the 1670's the dyewood was in short’ supply. The Lords Proprietor were concerned about unlicensed cutting and in 1676 instructed Governor Chillingforth to prevent iit (Albemarle, 1676). Catesby, who visited the islands in 1725, reported that the value of the wood had made it scarce, the biggest trees remaining not being more than 8 to Gafeet stad GUS abla Si By the time the loyalists arrived it was still being exported in considerable quantity GSchoepf, -1778234):. However, after the development of syn- thetic dyes in the 1870's and 1880's the demand for bra- ziletto declined, and by the end of the nineteenth century its export had virtually ceased (Coker, 1905: 201). OneG@an Island in 1970 it was very seldom seen in the woodland. Another valuable dyewood exported from the Bahamas was Logwood (Haematoxylum campechianum), although this species is not native to the islands. Bahamian woodcutters, cut it in Honduras in the late seventeenth and early eighteenth centuries and brought seeds back with them to plant at home. According to Catesby (1731, II: 65), it was introduced from the Bay of Honduras by a Mr. Spatches in 1722. Apparently it became quickly established locally as it was included by Bruce (1782: 422) in his list of valuable dyewoods growing in the Bahamas in the 1740's. However, when Schoepf (1788:35) visited the islands in 1784 it was not ‘yeu important export. In the nineteenth century it was cut in large quantities on Exuma and to a lesser extent on Cat, New Providence, and Long Island. By 1880 it had become scarce and according to the Blue Book of that year the supply was virtually exhausted (Taylor, 1881: 55). At the end of the nineteenth century it was still being exported in consider- able quantities to New York, the most important amounts com- ing from Andros, Exuma and Cat Island (Coker, 1905: 9920208 In 1970 there was still a demand for chipped logwood in Lon- don but the lack of any chipping machinery on the islands prevented its export.’ On Cat Island in 1970 it was quite common in certain areas at the southern end of the island. 2. Ihe first English settlement in the Bahamas was ad- ministered by a proprietary form of government. 3. Personal communication from Mr. Leonard A Roaker, a Bahamian agent for barks and dyewoods. 61 Another dyewood mentioned in the early accounts is yel- low fustic, also known as yellow wood or satinwood. These names refer to Fagara flava a member of the citrus family. Yellow wood was initially cut as a dyewood but later became more important as a timber tree. Its fine grain made it the most valuable of all the woods exported from the island in the late nineteenth century. Again its value led to its being over-exploited and by the 1880 the supply had been largely exhausted (Taylor, 1881: 55). Yellow wood is cer- tainly a rare species on Cat Island at present. Of all the valuable woods this is probably the most sensitive to cut- ting. In 1723 the Governor reported that "Brown ebony of a strong rhodium scent" was being exported from the colony (Phenney, 1723: 54). The species referred to here is_ prob- ably Dalbergia ecastophyllum, which today does not grow in the Bahamas. Theoretically this is a species that could have become extinct because of over-cutting. However, it is rarely mentioned in the accounts of the islands and on the basis of the available evidence it seems unlikely that ever grew on Cat Island. There is a puzzling note in Governor Montfort Browne's report on the state of the islands in 1775. According to Browne (1775: 1), green ebony and bark were being exported to Britain. The trade name green ebony usually refers to the leguminous tree Brya ebenus, which is native to Jamaica and Cuba but not the Bahamas (Britton and Millspaugh, 1920: 196). Again the possibility exists that it was exploited to the point of extinction, although it seems more likely the Governor was referring to the re-export of wood imported from Jamaica. Barks Sweetwood bark (Croton eluteria) has long been an esta- blished Bahamian export. Its uses have been varied, although it seems to have been most important as a basic ingredient in tonic waters. Exploitation probably began in the seventeenth century. According to Stisser it was exported to England as a smoking mixture in 1686 (cited by Bacot 1869: 3). Catesby reported that it was common on most of the islands, although cutting had reduced the size of the trees. He described it as a fine aromatic bitter to be infused with wine or water (1731,II: 46). Throughout the eighteenth and nineteenth centuries sweetwood bark was exported on a fairly regular basis, although what statistics there are suggest cutting was especially important during times of economic difficulty. The preparation of the bark 62 is a rather time consuming operation and is only worthwhile when money is in short supply. By the end of the nineteenth century supplies were "steadly diminishing" (Morris, 1896). And in the Bahama Flora it is described as "Becoming scarce" (Britton Sana Millspaugh, 1920: 223). In recent years the demand for the bark has increased and in 1970 it was being actively cut on Cat Island. A total of about 20 tons were exported from the Bahamas in 1970, of which about a third came from Cat and the rest from Acklins. Sweetwood bark reproduces vigorously from sprouts and although it is not a common tree in the woodland it seems unlikely that its importance has been sig- nificantly reduced by cutting. The early accounts of it having become rare may simpyly refer to a declime in the size and yield of individual trees rather than an actual reduction in range. On Cat Island there has been some small scale cultivation of sweetwood bark and many families have a "Bark field" close to their homes. Another tree that has been cut for its bark is wild cinnamon (Canella alba). Like sweetwood bark it was one of the earliest Bahamaian exports and may in fact have been exploited by the Spanish. However, unlike the sweetwood bark, the demand for wild cinnamon was never very high and it was exported only on a small scale (Coker, 1905: 206). At present wild cinnamon is a very rare species on Cat Island. Apparently it has been less able than the sweetwood bark to recover from the effects of cutting. At one time it was thought that Cinchona was native to the islands. Gover- nor Phenny (1724: 55) reported that "the Spanish have told several people that the Jesuit's bark abounds. But it has not been found for want of a curious’ enquirer." Quite likely the species referred to here was princewood (Exostema caribaeum), a close reative to Cinchona and a common small tree in the woodland. Princewood apparently was never cut for its bark on a commercial scale, although it is used in local medicine as febrifuge. Timber Trees Generally speaking, Bahamian timber has been protected by its naturally small size and comparatively few species have been cut for export. The most important of those that have is mahogany (Swietenia mahagoni) Although widely intro- duced elsewhere in the West Indies, mahagony is native to the Bahamas. There is no record of its introduction and it appears quite at home in the woodland. According to Catesby it was the most valuable timber tree in the islands, being better than oak for shipbuilding because it resisted shot 63 Without splintering (1731, II: 81). The historical record indicates that it was extensively cut in the early eighteenth century. Governor Rogers (1730) reported that "one of the best employments the inhabitants have had of late is sawing mahogany and Madera plank to ship to Europe." At the end of the eighteenth century it was still the most important timber tree although it had become rare on New Providence and neighboring islands due to over-cutting. The demand for Bahamian mahogany declined,in the nineteenth cen- tury, presumably because of the small size of individual trees. During the course of the present study, mahogany was occasionally seen on Cat Island, particularly in the more remote areas. It was also seen in the settlements where it has been planted on a small scale. Its bark is valued in local medicine and most trees are characteristically scarred as a result. Horseflesh (Lysiloma leucocephala), a tall leguminous tree, has also been exported under the trade name mahogany. According to Catesby (1731, II: 42), it was the second most valuable tree in the Bahamas after Swietenia. Like the true mahogany, its export declined in the nineteenth century although it was still being exported in "considerable quan- tities at the beginning of the present century" (Coker, 1905: 202). Cutting has made it a rare tree on Cat Island, as it was very infrequently seen in 1970. Lignum vitae (Guiacum sanctum, G. officinale), whose resinous gum was thought to provide a cure for syphilis, was probably first cut in the sixteenth century by the Spanish. The early English accounts boast of its presence and its exploitation appears to have continued throughout the seven- teenth and eighteenth centuries. The hard wearing proper- ties of the wood made it one of the most valuable of Bahamian timber trees. In the nineteenth century the demand for it also declined because of its small size (Coker, 1905: 203). As in the case of horseflesh, cutting has made lignum vitae a rare tree on Cat Island. It was seen growing wild only in areas remote from the settlements. On the other hand, it has been planted on a limited scale as an ornamen- tal. A number of other native species have been cut for timber, although not on the same scale as those mentioned above. These include: ironwood (Krugiodendrom ferreum), boxwood (Buxus bahamensis), mastic (Sideroxylon foetidi is- simum), and princewood (Exostema caribaeum). With the exception of princewood, all were comparatively rare on Cat Island in 1970. MThe posssibility exists that they were 64 always rare, but the fact that they are more commonly encountered in the more remote areas of the woodland sug- gests that selective cutting has reduced their importance. Local Exploitation The exploitation of native plants for local usage has long been an important part of Bahamian life. Furthermore, virtually every species in the woodland iS recognized as being useful for some purpose. In the discussion that fol- lows only those species whose importance in the woodland has been changed by exploitation are considered. Several native fruits are gathered locally, most not- ably the sea grape (Coccoloba uvifera) and cocoa plum (Chrysobalanus icaco). Both are important on the whiteland and may have had their distribution patterns modified by accidental dropping of the fruits. Certainly the sea grape is commonly seen growing along the footpaths across the island although rarely in the woodland itselii. The same"footpath" distribution is characteristic of the pondtop (Sabal palmetto) whose fruits are collected for hog feed. As was mentioned earlier fruits accidentally dropped in this way have produced an avenue of palmettoes through the MecQueens savannah (Figure 14). The hog cabbage palmetto (Pseudophoenix vinifera) has been even more drastically affected by man's activities. Both its fruits and terminal buds have been collected for hog feed, and as a result it is now seen only in remote parts of the island. On the posi- tive side, it has been occasionally planted as an ornamen- tal. In recent years, houses have been built with imported pine, although in the past they were entirely built with local wood. For support posts and beams hardwoods such as dollen plum (Reynosia septentrionalis), mastic (Sideroxylon ee maderia (Swietenia mahogani), horseflesh (Lysiloma Teucocephala) , or cassada wood (D ipholis salici- folia) were used. For the more flexible cross-beams wattle (Eugenia spp.) or red mangrove (Rhizophora mangle) were pre- ferred. Thatching would be done with the leaves of the Buf- falo top Thrinax microcarpa). The traditional methods of thatching are African in origin. Only a few men in each settlement know the techniques’ and very likely they will soon be lost. Small fishing boats are still built Wocaiigg usually with horseflesh or mastic or if these are not avail- able with wild locust (Lysiloma bahamensis). The cutting of wood for fuel has probably reduced the local importance of several species. Buttonwood (Conocarpus 65 erecta) is generally regarded as the best firewood, while black torch (Erithalis fruticosa), dollen plum (Reynosia septentrionalis), and white torch (Amyris elemifera) are also used. : A great many native species are used in bush medicine, although it seems unlikely that many have become rare because of it. Two that might have been over-exploited are the boarhog bush (Callicarpa hitchcockii) and manroot (Val- lesia antillana). Both are ingredients in popular aphrodi- Siacs and are rarely seen in the woodland around the settle- ments. In summary, it seems clear that a great number of wood- land species have been cut for either export or local use. This exploitation began in the sixteenth century and contin- ued with varying degrees of intensity until the present. Although some cutting has always been done locally much of the activity appears to have been based in Nassau. As Schoepf pointed out in 1784, the white inhabitants of New Providence employed their slaves cutting wood wherever it eould be found: Wood-cutting is gradually becoming more diffi- cult and less lucarative. On the islands lying next to Providence the best wood is always cut off, and thus there must be recourse to islands lying farther away, or the woods must be more deeply gone into (Schoepf, 1778: 34). On Cat in 1836 it was thought necessary to have a man on guard in the north eastern part of the island to prevent Eleuthera men from cutting wood (Stiles, 1836). Even so by the middle of the nineteenth century the more accessible timber appears to have been taken. The surveyor Harvey, who visited the island in 1855, reported that: The timber on St. Salvador (Cat) is fine and large and might be made a profitable branch of commerce; maderia, mahogany, cassada, prince- wood and braziletta, yellow wood and lignum vitae are found in every part but in greatest abundance on the east side (Harvey, 1858: 78). In other words, the valuable timber was already depleted on the west side of the island that is in the areas close to the settlements. The exploitation of dyewoods and valuable timbers appears to have declined in the second half of the nineteenth century. Exploitation for local use has continued, but even here the pressure has eased because of the import of cheap pine from Nassau. 66 As might have been expected, the species that have been selectively cut have varied in their capacity to recover. Some like lignum vitae (Guaiacum sanctum), wild cinnamon (Canella alba), and yellow wood (Fagara flava) are still rare even though they have not been cut on any scale since the nineteenth century. Other species, such as sweetwood bark (Croton eluteria) and logwood (Haematoxylum campechia- num), have recovered comparatively quickly. The differences here are probably due to inherent differences in reproduc- tive capacities and habitat tolerances. What is surprising is that for the Bahamas as a whole not one economically valuable species is known to have become extinct. Even though the pre-settlement composition of the woodland is not known there is no evidence to suggest that any species has been exploited to the point of extinction. The species involved have in fact proved remarkably resilient. 67 VII. THE INTRODUCTION OF ALIEN PLANTS AND ANIMALS The Bahamas have not been isolated from the large-scale interchange of plants and animals that has characterized the tropical world during the last five hundred years. In spite of their late settlement by Europeans and the persistent failure of commercial agriculture, a great number of plants and animals have been introduced to the islands. Unfor- tunately, most of these introductions are undocumented, par- ticularly for remote out-islands such as Cat. What follows therefore is a general review which deals for the most part with the Bahamas as a whole. The discussion of plants is limited to those woody species capable of establishing them- selves spontaneously within the woodland. Except where Stated, the alien status of the species discussed is well established either by documentary evidence or by the artifi- cial nature of their distribution. The only animals con- Sidered are those that have had some impact on the composi- tion of the woodland. Introduced Plants It seems likely that the Arawaks brought several species of fruit trees to the Bahamas. The guava (Psidium guajava), sugar apple Annona squamosa), custard apple (A. reticulata), dilly (Manilkara zapota), and hog plum (Spon- dias mombin) were all cultivated in pre-Columbian times in the West Indies (Roumain, 1942). Each species is capable of spreading from cultivation, although to what extent any of them actually did is not known. Perhaps significantly the early English accounts make no mention of any "wild" fruit trees. The question is complicated by the fact that in the seventeenth and eighteenth centuries all the Arawak fruit trees were reintroduced as were several other New World species such as the avocado (Persea americana), genip (Mel- icoccus bijugatus), pawpaw (Carica papaya), and cashew nut (Anacardium occidentale). At the same time, the tamarind (Tamarindus indica), Indian Almond (Terminalia catappa), pomegranate (Punica granatum), and mango (Mangifera indica) were introduced from the Old World. Commercially the most important Old World introductions were species of citrus. The most valuable Bahamian exports in the early eighteenth century were limes (Citrus aurantium), lemons (C. limon), oranges (Cc. sinensis), and sours (Ge aurantium). They were exported to North America together with dyewoods, timber, and medicinal barks (Catesby, 1731, I: xxxviii). 68 To what extent any of the citrus were able to become naturalized is not -indicated in the “historical (econun although by analogy with what had happened in other parts of the West Indies it would seem likely that some escaped from cultivation. Harris (1965: 93) has emphasized the rapidity with which the lime and bitter orange spread spontaneously in the West Indies. According to Howard (1950: 345), Citrus aurantium has become established spontaneously on Bimini, in the north western Bahamas. On Cat Island, citrus were exten- sively planted in the 1850s (Harvey, 1858: 76); however, no feral trees were seen in the woodland in 1970. In 1783 a plan was formulated to establish a botanical garden in the Bahamas, with one of its proposed purposes being to test plants from the South Seas (Pownall, 1783). A shipment of live plants and seeds was sent to the Bahamas in 1799 from the botanical garden at St. Vincent (Anderson, 1802: 2-45 )k. Four important fruit trees included were the Indian Almond (Terminalia catappa), the mango (Mangifera indica), the otaheite gooseberry (Phyllanthus distichus). and the bread-fruit (Arctocarpus communis). All four’ were successfully established in private gardens and may, with the exception of the breadfruit, have escaped locally. In the nineteenth and present centuries, the development of private gardens, particularly in Nassau, led to the intro- duction of literally hundreds of new species and varieties of fruit trees. However, there is little indication that any of them have been able to become established as part of the wild vegetation. In contrast to the fruit trees, few ornamentals appear to have been introduced before the eighteenth century. Catesby, who visited several islands during his stay in the Bahamas (1725-1726), included only two introduced ornamen- tals in his Natural History, the red frangipani (Plumiera rubra) and the coral tree (Erythrina corallodendrum). Nei- ther species is capable of spreading spontaneously. The naturalist Schoepf, who visited the Bahamas in 1784, described a large silk cotton tree (Ceiba pentandra) in Nassau which presumably had been planted there early in the eighteenth century (Schoepf, 1788: 37). According to Gardner and Brace (1889: 369), this tree had originally been introduced from South Carolina and was the source of all the other silk cotton trees on New Providence. The species is included in the Bahamas Flora where it is reported as being "Spontaneous after cultivation" (Britton and Millspaugh, 1920: 275). There is a large silk cotton tree at the Bight settlement on Cat Island. However, it shows no evidence of successful regeneration. 69 Other ornamentals mentioned by Schoepf which may have become locally naturalized are the sand box tree (Hura crep- itans) and "Barbados Pride" (Poiniana pulcherrima). Accord-— ing to Gardner and Brace (1889: 376), the latter was intro- duced to New Providence in 1886 by a Mr. Sanders. This must have been a late reintroduction. Neither species was seen on Cat Island during the present study. During the nineteenth century a great many ornamentals were introduced to private gardens, but comparatively few appear to have spread spontaneously. Exceptions have been the poinciana (Delonix regia), the Australian pine (Casuar- ina equisetifolia), Jerusalem thorn (Parkinsonia aculeata) , Spanish bayonet (Yucca aloifolia), and the cactus-like Euphorbia lactea. All of these are included in the Bahama Flora and all were observed to have spread locally on Cat Island. In the present century the development of landscape gardening has meant a further increase in the introduction of ornamentals. However, few if any are adapted to life in the woodland and consequently their story is not relevant to the present study. Cotton is one of the few aliens known with certainty to have been introduced by the Arawaks. According to the early Spanish accounts, it was cultivated in the Bahamas on a con- Siderable scale (Craton, 1968: 25). Which species was involved is not certain, as both Gossypium barbadense and G. hirsutum were cultivated in the West Indies (Sauer, C.O., 1950: 535). Cotton presumably persisted after the islands were depopulated but for how long is not known. There is no mention of wild cotton in the early English accounts. As indicated earlier, cotton was cultivated on a_ small scale in the early eighteenth century, extensively in the late eighteenth century, and again on a small scale in the nineteenth century. On Cat Island it has probably not been cultivated since the American Civil War. In 1970 individual bushes were occasionally seen in the woodland. Sea island cotton (Gossypium barbadense) is weedy and has been able to persist on a small scale in disturbed sites. Sisal (Agave sisalina) was introduced to the Bahamas from Yucatan im (RAS (Morirass 183908 2c Aeeeie 2 Peeler slow start as a commercial crop, it was widely planted throughout the islands in the years 1887-1896. However, the productive life of the plant proved to be shorter’ than expected, and cultivation ceased. Since then it has been planted locally on a small scale and has proved remarkably persistent in the wild. 70 Bowstring hemp (Sansevieria thyrisiflora) was intro- duced to the Bahamas in the nineteenth century (Dyer, 1887). It was planted commercially on Cat Island in the 1940s, although like sisal it was not a commercial success. In 1970 it formed a thick carpet under second- growth woodland in the Orange Creek area, where it had formerly been cul- tivated. Presumably it will eventually be shaded out, but is currently slowing down the regeneration of the native Species. The "indigo weed" (Indigofera suffruticosa) was being cultivated on New Providence as early as 1698 (Craton, 1968: 89), and may even have been introduced in Arawak times. It is now widely distributed throughout the archipelago and was seen occasionally in weedy sites on Cat Island in 1970. The old-world indigo (Indigo tinctoria) was introduced in the late eighteenth century but did not grow well in the Bahamas (Brown, 1802: 27). As was indicated earlier, logwood (Haematoxylum cam- pechianum) was introduced to the Bahamas in 1722 from Hon- duras. Since then it has spread widely around the islands and has locally become important in the woodland. Another leguminous tree that was probably introduced in the eighteenth century is the sweet acacia (Acacia farnesiana). Catesby (1731, II: 45),includes a plate of what appears to be this species in his Natural History. Its original home is not known with certainty, although it has been accepted as native in Cuba (Little and Wadsworth, 1964:144). Its Spines make it a valuable hedge plant and probably because of this it has been widely distributed throughout the Baha- mas. On Cat Island it was occasionally observed in dis- turbed habitats such as roadsides. The divi-divi tree (Caesalpinia coriaria) was intro- duced into the Bahamas in the early part of the nineteenth century in the hope of exporting its pods for tanning (Ham- ton. © oso). Apparently the demand was not strong enough to justify large- scale planting, and the tree was never taken toe sthe™ Our) sisiltandis: According to Britton gang Millspaugh (1920: 174), it has spread locally from cultiva- tion on New Providence. Several species were introduced in the eighteenth cen- tury as fodder crops. The most important of these was Jum- bay (Leucaena leucocephala a leguminous shrub from Central America. Leucaena has been widely planted throughout the archipelago, and is probably the most invasive of all the introduced trees or shrubs. Also valued as a source of fodder in the eighteenth century was the horseradish tree (Moringa oleifera). It had been introduced to Jamaica from 71 the coast of Guinea in the seventeenth century (Edwards, 1819, 1:481), and may have been brought from there to the Bahamas. The loyalists used it as fodder for sheep (Brown, 1SO2 See in)! < At present it is occasionally seen as an orna- mental on Cat Island and is locally spontaneous. Also used by the loyalists for fodder were Sesbania grandiflora and the "Pride of India" (Melia azaderach). Both of these old-world species have spread spontaneously. The latter has been planted as an ornamental and is occa- Sionally seen in the settlements on Cat Island. Although the historical record is far from complete, it does indicate that very few aliens become firmly established as part of the wild vegetation. In spite of the fact that a considerable number of plants have been introduced to the Bahamas during the past three hundred years, comparatively few have been able to spread very far without man's help. Daniel McKinnen aptly described the situation in 1803: The exotics which are introduced seem feebly and unsuccessfully to struggle with cold winds; the droughts, and unfriendly seasons; while a crop of hereditary and worthless weeds take possession of the soil prepared for cultivation, and extract all its nourishment to administer fertility, as they decay, to the native and unprofitable forest trees succeeding them, the elemi, silver-leaved palmet- tos, and hungry aborigines of the rocks (McKinnen, 13068 BSiNe Just why this should be so, in contrast to the West Indies, where aliens covered extensive areas at very early dates (Harris, 1965: 113), is not immediately apparent. The question will be raised again later when the quantitative importance of aliens in Cat Island woodland is discussed. Domesticated Animals The introduction of alien animals has often brought about far-reaching changes in the vegetation of oceanic islands and this has certainly been the case in the Bahamas. Goats, horses, sheep, cattle, and hogs have all had an impact on the wild vegetation of the islands, especially in the areas close to the settlements. Whether or not the Spanish stocked the island with livestock is not known. When the loyalists arrived on Cat Island in the 1780s they found plenty of wild hogs, although no mention is made of any other animals (Eve, O., 1784). 72 Presumably the hogs brought about changes in the woodland, although just what these were is difficult to assess. Very likely the wild population was eliminated by hunting in the nineteenth century. Domesticated hogs have had an indirect impact on the woodland through the gathering of hog feed. The native hog cabbage palm (Pseudophoenix vinifera) has become rare because of this pressure. In 1970 most families had at least one hog in their yards. They are, however, penned and fed largely on household scraps. During the loyalist period Cat Island had the reputa- tion of being one of the better islands in the Bahamas for the raising of livestock. According to one source (Powles, 1888:234), thousands of head of cattle were raised on the island at this time. After Emancipation cattle were still raised, although not in the same numbers as_ before GUnderhill,: 1862: 480). In recent years the Bahamian government has made a concerted effort to develop livestock farming. Pastures were established in several settlements (Arthur stown , The Bluff, The Bight) and planted with African grasses.’ Unfortunately the scheme has had little success. The woodland was cleared with a bulldozer and the limestone surface hardened on exposure to the atmosphere and as a result even the drought-resistant African grasses have scarcely been able to survive. In spite of reports of large cattle herds during loyal- ist times, it seems unlikely that livestock farming was ever much of a success. During the dry season the native grasses dry out and there is a general scarcity of forage. Further- more, the broken, potholed nature of the limestone is dangerous for livestock. In 1970 only one herd of cattle was seen on the island, and this was being pastured on the whiteland near The Bluff settlement. Horses were apparently numerous on the island during loyalist times. One Cat Island planter even imported thoroughbred stallions from England (Stark, 1891: 149). The horse population declined in the late nineteenth century, although horses continued to provide the main means of tran- sportation on the island until automobiles arrived compara- tively recently. It is doubtful whether any of the loyalist thoroughbred stock made any contribution to the present population of horses on the island. Cat island horses in T. Three species were tested for use in the pastures: Rhodes grass (Chloris gayana), pangola grass (Digitaria decumbens), and star grass (Leptochloa plechtostachya). Three Others were planted as possible fodder crops: elephant grass (Pennisetum purpureum), para grass (Pan- icum muticum), and Johnson grass (Sorghum halepense). WS 1970 were generally small and lean and had the distinct appearance of being both underfed and overworked. Both horses and cattle are mainly pastured on the whiteland where they feed on grasses and herbaceous weeds. Their main influence at present is to slow down the recovery of shrubs and trees. Sheep were also introduced by the loyalists and were at one time quite numerous, especially at the southern end of the island. Grazing by sheep is of more interest to the present study in that it directly affects the woodland. Sheep, more so than cattle or horses, are content to browse. In the dry season when no grass is available they will eat the leaves of shrubs such a jumbay (Leucaena leucocephala ramshorn (Pithecellobium keyense) and pigeon berry (Rhacoma crossopetalum). This browsing has had a selective effect on the composition of the woodland. The pasturing of sheep also affects the woodland in other ways. Some farmers will scatter grass seeds and jum- bay before abandoning a field. When the last crop has been harvested, the field will be turned over to sheep. Grass seeds are cycled through the sheep's gut, and a thick growth of grass results. This means that soil development is accelerated but the recovery of the woody vegetation is retarded. In 1970 such effects were very localized on Cat Island, as only a few small herds of sheep were being raised. They are a white short-haired breed without horns. Only the goat has had a really significant impact on the woodland. Unlike the other domesticates, the goat is eontent to browse at all times of the year. In 1970 the goat was the most numerous domesticate on the island, the total population amounting to several thousand. Cat Island goats are mostly coarse-haired, black or brown animals with short horns that curve over backwards. In recent years attempts have been made to upgrade the quality of the local breed with American stock. This review of the history of plant and animal intro- ductions does not entirely support the idea that alien species have a competitive advantage over insular _ species. Certainly as far as plants are concerned, few aliens have become well-established. On the other hand, domesticated animals have at various times been numerous on the island, and may in some cases have had a significant impact on the woodland. The nature of this impact is considered more closely in the section dealing with areal variation in the composition of the woodland. - Dg Somech> »ortso ber: « Mie fel ae ae PO 4 behtowi see Dim) > Pe Tey oar p sa os So's “geles viv At oo bautesoe Pee Van ou hn ade : mene Lata y S| On ¢ce*7 49 ra] } ; im Og tiie ees eR 75 VIII. METHODS OF ANALYSIS Unfortunately, the historical record provides only a very qualitative indication of the extent to which the vege- tation of Cat Island has been modified by man. It shows that for a period of at least a thousand years the woodland has been discontinuously cleared, burned, browsed, and selectively cut, but it does not show what the detailed consequences of these activities were. With this deficiency in mind it was decided to analyse the composition of the present woodland in such a way as to show in quantitative terms the extent to which it has been changed by man. More Specifically the following questions were considered: 1. What are the detailed consequences of clearing and burning? How long does it take the woodland to recover? What species are involved and when do they become esta- blished? 2. To what extent has the woodland been changed by selective pressures such as grazing, cutting, and the use of fire? Which species have become less important as a result of these pressures? Which have become more important? 3. To what extent have alien plants been able to invade the woodland? Are they only a temporary feature of the vegetation or have they become permanently established? Field Methods Fortunately, it was possible to spend three field ses- Sions on Cat Island: July to October 1967, June to August 1968, and June to August 1970. In addition short visits were made to three other Bahamian islands--Bimini, May- aguana, and New Providence. Basically the field work con- Sisted of plant collection and identification, interviews, and vegetation analysis. Plant Collection and Identification The Bahamian woodland is typically tropical in that two individuals of the same species are rarely seen together. This, together with the physiognomic similarity of so many Species, made the taxonomic problem a difficult one in the early stages of the study. A large part of the first field season was therefore devoted to the collection of plant specimens. In total some 600 collections were made, not counting duplicates (Appendix I). As far as possible all 76 the species encountered were collected. In the case of dif- ficult genera, such as Pithecellobium, Eugenia, Cassia, Coc- coloba, and Lantana, several collections of each species were made. Provisional identifications were made in the field with the use of the keys in Britton and Millspaugh (1920) and Little and Wadsworth (1964). Local names provided useful if not always reliable clues as to the identity of many species. A nearly complete set of the plants collected was later sent to Dr. R.A. Howard of the Arnold Arboretum, and his determinations are the ones used in the present text. Local Interviews Because of the length of time spent in the field it was possible to develop close personal contacts with many of the local inhabitants, particularly in the settlements at the north end of the island. This in turn made it possible to gather a great deal of information relating to shifting agriculture, grazing, and the selective exploitation of woodland species. Interviews were usually carried out on an _ informal basis without the use of any set questionnaire. A compli- cating factor here was the custom whereby the person inter- viewed usually provided the answers he or she thought were expected. This problem was avoided by not asking leading questions and by checking the reliability of the informant with questions the answers to which were already known. The Older people were generally more knowledgeable since they had spent all their lives farming on the island. As far as the recent history of shifting agriculture was concerned, they were the only source of information. In general, Cat Islanders were both willing and knowledgeable informants. Furthermore, the information they provided was indispensable to the study as a whole. Vegetation Analysis Ultimately the success of this type of study depends on the ability of the investigator to accurately identify plants in the field. As was indicated above, the taxonomic problem was initially a difficult one; apart from a few weeds and aquatics, the Bahamian flora is very different from that of mid-latitude areas. Eventually, however, after walking through literally hundreds of miles of woodland and collecting every unknown, it became possible to identify on sight the vast majority of the species encountered. Once the flora was well-known it was possible to set up a sampling procedure that could be applied to the woodland as a whole. a In 1970 a sampling method was devised that would not only give answers to the questions posed above but would also provide a base-line against which further change could be assessed. Three hundred sampling sites were selected in different parts of the woodland, and since the entire wood- land had been cultivated at one time or another, each sample Site was in effect an abandoned field. The two criteria for selection were that the site could be agcurately located on the available aerial photograph coverage’ and that the vege- tation appeared to be, or was known to be, of a uniform age. The actual data collected consisted of a visual esti- mate of the cover within 25 x 1 meter quadrats in each field. Only woody species or succulents with a minimum cover of at least one quarter of a square meter in each qua- drat were included. Herbs were omitted because of the marked seasonality of their occurrence and because a new sampling method would have been needed to deal with them adequately. In younger fields the quadrats were laid out in the form of a cross centered in the middle of the field. In the older fields, or in fields difficult to penetrate, a 25 meter belt transect was run as far into the field as possi- ble. Measurements were made with a meter stick and a meas- ured length of rope. The cover data were dictated into a portable tape recorder and later transcribed onto data sheets (Appendix III). For convenience each species encoun- tered was given a code number. 120 species were encountered in the 300 fields sampled (Appendix I). Once a field had been sampled it was given a number and classified on the basis of the following characteristics: (1) age since abandonment, (2) habitat-type, (3) height of vegetation, (4) soil-type, (5) moisture characteristics, and (6) distance from the nearest settlement. Finally, its location on the 1958 aerial photographs was fixed by means of a six-figure cross reference. Not all of the distinc- tions made above proved to be relevant to the present study, but the way in which they were established is as follows: Age since abandonment. As can be seen from Figure 16, the woodland is in fact a patchwork of fields in different stages of recovery following clearing and burning. It was soon recognized that any analysis, if it were to be meaning- ful, would have to allow for these age differences. The age of the woodland was determined in two ways: local informa- tion on the history of land use, and tonal appearance on T. Good quality stereoscopic coverage was available for the island at a scale of 1:12,500. The photographs were taken in 1958 by Spartan Aviation Services on con- tract to the Bahamian Government. 78 aerial photographs. Local informants could usually provide a reasonably accurate age for fields abandoned less than 12 years ago, although beyond that memories were vague. For older aban- doned fields the 1958 aerial photographs provided further clues as to age. Generally speaking, the older the woodland the darker its appearance on the aerial photographs. The reasons for this are basically twofold: (1) reflection from the limestone surface is reduced as succession takes place, and (2) pioneer species have generally lighter-colored leaves than the shade-tolerant species that replace them. This meant then that tonal differences on the aerial photo- graphs provided a good indication the relative age of the woodland in the areas sampled, and by determining whether the area was white, light grey, dark grey or black, an approximate age-sequence was established. Habitat type. As was indicated earlier, there’) 3seae variability in the woodland which reflects underlying differences in surface characteristics. To minimize the effects of this variability it was decided to classify the woodland on the basis of land surface or habitat charac- teristics. Initially all five basic landform types were used: 1. Holocene Dunes and Beach Ridges. 2. Young Pleistocene Marine Plains. 3. Young Pleistocene Dunes and Beach Ridges. 4, Old Pleistocene Marine Plains. 5. Old Pleistocene Dunes and Beach Ridges. After analysis of the data collected, it became clear that habitat-types 2 and 4 could be combined, as also could 3 and 5. As far as the floristic composition was concerned the differences in each case did not appear to be signifi- cant. Furthermore, a low sample coverage in habitat type 4 limited its usefulness. The final breakdown therefore involved just three habitat-types: 1. Holocene Dunes and Beach Ridges. 2. Pleistocene Marine Plains. 3. Pleistocene Dunes and Beach Ridges. In terms of local names these habitat-types are equivalent to the whiteland, flatland, and blackland respec- tively. In classifying the woodland indirectly by means of surface characteristics, the arbitrary problem classifying vegetation itself was avoided. Furthermore, a coherent framework was provided within which the processes of change could be analysed. 79 Height of vegetation. For each abandoned field sampled the average height of the vegetation was measured and recorded to the nearest meter. In most cases this was a meaningful figure, as the vegetation sampled was of a more or less uniform height. In the younger fields, sprouts which would characteristically project above the seedling- bushes by a meter or so were not included in the average. Soil-type. Each sample site was classified as _ to whether the soil was white, black, or red. The main reason for making this distinction, in addition to that of habitat-type, was that the red soils are present in more than one habitat-type. Although they are mainly found in the flatland areas, they are also characteristic of the older Pleistocene dune ridges. However, the red soil areas are localized on the island and because of this only 34 red soil fields were sampled. Because of the small size of the total sample and because there was no indication that differences in soil type had any affect on the composition of the vegetation it was decided not to use the red soil fields as an independent unit. The white soils are simply equivalent to the whiteland habitat and the black soils to those Pleistocene surfaces without red soils. Moisture class. Within the woodland there are differ- ences that are independent of either soil or surface charac- teristics and are primarily due to differences in drainage eonditions or exposure. In some areas, as for example on the leeward side of the Pleistocene dune ridges, the land surface is steep and any rain that does fall is quickly lost as run-off. Such areas are naturally droughty. Similarly, low-lying areas where the fresh-water lens is thin or non- existent are also droughty. The same is true of areas exposed to the full force of the trade winds; evapotran- Spiration rates are high and the trees are consequently stunted and slow-growing. On the other hand, in some low- lying areas, close to the seasonally-flooded savannas, the woodland is occasionally flooded. In order to accommodate these differences, a three-fold elassification was set up: xeric for droughty fields, hydric for seasonally-flooded fields, and mesic for intermediate fields. As it happened, only 15 xeric and 15 hydric fields were sampled, and visual inspection of their cover data sug- gested their floristic composition was not significantly different from the rest. Consequently these distinctions were ignored in the later analysis. Distance index. Until recently few people on Cat Island had motor cars and even now most farmers walk to and from their fields along rocky footpaths. There is only one 80 road on the island and this joins the settlements along the western and southern coasts. For most of its length it is not surfaced, and travelling from one end of the island to the other is a major undertaking, so much so that people from the northern end rarely if ever visit the southern end and vice versa. All of this emphasizes the fact that the frictional effect of distance is strong on Cat Island. The importance of this to the present study is that man's modif- ications of the woodland through clearing, burning, selec- tive cutting, and grazing decreases in intensity with dis- tance from the settlements. In order to determine the sig- nificance of this, each sample site was classified in terms of its distance in kilometers from the nearest settlement. Aerial photograph reference. AS was indicated above, the choice of sampling locations was in large part deter- mined by whether or not the field in question could be accu- rately located on the 1958 aerial photographs. This was thought to be especially important insofar as it would make possible future sampling in the same location. The location of each field sampled was fixed by a cross-reference on a 1958 aerial photograph. A 12-figure reference was used, the first three numbers of which referred to the flight number of the photograph, the second three to the frame number. The last six figures’ provided the cross-reference on the photograph itself, the first three representing the distance in millimeters from the left-hand margin and the last three the distance from the bottom margin, (Figure 24). At a scale of 1/12,500, one millimeter on the photograph represents 12.5 meters on the ground, so the sample site is fixed quite accurately. Analysis of Field Data After returning from the field, the data gathered were organized for analysis. The analysis involved plant iden- tification, aerial photograph analysis, and computer analysis of the plant cover data. Plant Identification As was mentioned earlier, a full set of the plants col- lected was sent to the Arnold Arboretum and the names of all the species included were determined by Dr. R.A. Howard. In addition to this, several days were spent at various her- baria comparing duplicate specimens with earlier Bahamian collections. The Field Museum of Natural History collection of Bahamian plants was especially useful here, as it included a duplicate set of all plants collected during the 81 Figure 24. The aerial photograph reference grid. Each Sample site was fixed by a 12-figure reference number; for example: 012 : 180 :: O45 : 069 012 = The flight number. 180 = The frame number. O45 = Millimeters from left edge of photograph. 069 = Millimeters from bottom edge of photograph. 82 compilation of the Bahama Flora (Britton and Millspaugh, 1920). The herbarium work had two major objectives: first, to obtain some idea of the variability present in difficult taxa such as Pithecellobium, Eugenia, and Coceoloba; Jama second, to check the dates on the collections of jalaen species. In the last context the early collections by Catesby and others at the British Museum of Natural History were especially useful. Aerial Photograph Analysis The 1958 aerial photograph coverage of the island was an indispensable aid to field work. The photographs made it possible to pinpoint the location of the area sampled and also provided a good indication of the age of the woodland. Having returned from the field in 1970, further aerial pho- tograph analysis was made possible by the acquisifion of aerial photographs of the island taken in 1943. This further analysis consisted of two parts: confirmation of the habitat-type classification, and the establishment of time- control for the age of the woodland. Habitat-type classification. As every field sampled had been accurately located on the 1958 aerial photographs, it was quite a simple procedure to check whether or not the habitat-type assigned to the ground was confirmed by the aerial photographs. Stereoscopic coverage was available for 1943. and 1958, and on both series the land surface charac-— teristics were easily identifiable. Age determination. The 1958 aerial photographs pro- vided an approximate guide to the age of the woodland, as abandoned fields appear progressively darker with age. In spite of the time lapse between photography and sampling, this relationship appeared to be quite real on the ground, except of course for those fields cultivated and abandoned after 1958. The tonal differences, however, give only a relative time scale. It was safe to assume that a dark area of woodland was older than a light area, but the actual age difference involved could not be determined from the 1958 photographs alone. This problem was overcome by using the 1943 photographs in conjunction with those taken in 1958. More specifically, 2. The T943 photographs were taken by the U.S. Navy as part of a wartime defense operation. Also of quite good quality, they are available in stereoscopic cover- age ata sculie of es So 000 83 by comparing the differences in tonal density between photo- graphs of the same area it was possible to fix with a cer- tain degree of accuracy the age of any part of the woodland that had not been cleared since 1958 (Table 3). Computer Analysis Once all the field data had been checked they were transferred from the data sheets (Appendix III) to IBM cards for computer analysis. Each of the categories in the age, habitat, height, soil, moisture, and distance classes was given a code number, and each of the species encountered was given the same code number that had been used to refer to it in the field (Appendix I). Once the cards were in order programs were written to compute basically two kinds of information. First, how did the floristic composition of the woodland change through time? In other words, how did the woodland recover after clearing and burning? And second, how did the woodland vary with distance from the settlements? Limitations of the Data The methods employed in gathering the data were designed with two specific purposes in mind, namely, to evaluate the extent to which man had changed the floristic composition of the woodland and to provide a base-line against which future changes could be assessed. However, before proceeding with any discussion of the data, certain limitations should be made clear. Taxonomy As has been mentioned above, several genera contain Species that are not easily distinguished in the field. For example, in the genus Pithecellobium all the individuals encountered were recorded as P. keyense. However, the leaf characteristics by which this species is differentiated from P. unguis-cati are somewhat variable and both species may therefore be present. Similarly, in the genus Eugenia two Species, E. buxifolia and E. longipes, were quite distinct, but three others, E. lucaya, E. myrtoides, and E. monticola, were not, and some misidentification may have been made here. In the Euphorbiaceae two species may have been con- fused, Drypetes diversifolia and Savia bahamensis. Both are Similar in terms of leaf characteristics and general appear- ance and may have been misidentified when flowers or fruit were not available. In the genus Coccoloba, C. uvifera and ic” krugii are distinctive, but theljwildespread C-) diversi- folia was quite variable and may include more than one Species. 84 yoe td yoe tg do Kauy y4eqg OZ6L au0jeg (*S4k 0OS<) A Aaay yaed Aauy wntpeyw 6£61-O026L (“S44 OG=0€8 2 AT quesaug JON Asuy wnt pew do fauy 4UBTT GS6L-On6L (“sah 62=SG1)) IID quesaud 4ON Jo Kauy 4YU3TT quesadd ON S961L-9S6L Gigst Tl=G ie ie queseud JON quesaud ON 0L61-9961 (*s4h G>)-T SHdVYDOLOHd SHdVHYDOLOHd 8S6L NO JONVYVaddY €h6L NO JONVYWVAddY aqauvalo aLvad SSV190 J9V ALISNAG TVNOL GNV 4OV € ATaVL 85 In the final analysis the taxonomic problem was not a serious one. Because of the long period of time spent in Eicmielderat was pPOssabilles to Pidientiny, on saehts the vast majority of the species encountered. Furthermore, most of the 120 species included in the systematic sampling are quite distinctive. Coverage Roughly two-thirds of the sample locations are in the northern half of the island, the main reason being that the field research was based in Arthurstown. Although the road along the island is only about sixty miles long, for the HOStmepagelhtarSesimplys a rough track and ‘travelling is therefore very difficult and expensive. This uneven coverage is probably not a significant problem as there are few obvious differences between the northern and southern parts of the island. On the other hand, it does mean that the importance of alien species in the sample is less than it might have been. Logwood (Haema- toxylum campechianum) was seen to be especially common around the two southern settlements of Port Howe and Old Bight. As can be seen from Table 4, the coverage of the dif- ferent age-classes iS somewhat uneven. This reflects the variable character of the woodland itself. Large areas of woodland fall in the first three age-classes, whereas older woodland is comparatively hard to find. The same is true of the other variables indicated. Sample sites were chosen to give a good coverage of the woodland as a whole rather than to provide equal coverage within the different classes. This later proved to be a problem insofar as it necessitated the use of percentages in comparisons between classes with different sample coverage. Rare Species Even though the total sampled was large, rare species because of their very nature are not well represented. This is particularly unfortunate because it is most likely that it is the rare species have been most significantly affected by disturbance. In spite of their limitations, the data gathered pro- vide a comprehensive picture of oO (oe) cl Of 0S GS EL LL SPTeTd CUR G = VU S=t "wy h-€ “UH e=¢ ries ( aa “wy L g9oueystd Sl Ol2 Gl SPTeTd oTupAy oTSoW OTUdX aunqystow ft He MAG SPT9eTty oyTUM poy Hoel” TtOS 6S1L 66 en SPpTeTty pueTHoOe TT, pue Tye Td PpueT9yTyuM ZeqTqeH Ge AS €8 08 09 SPT9TY "suk OS< *suk OG-OF “suk 62-SlL “SUA FL-G SIO (Ga Ss aBy AYOOALVO HOWE NI Ga IdWVS (out X G2) SQTHI4 dO YSEWNN AHL fh ATAVL 87 the variable character of the woodland as it was in 1970.3 AS a cross-sectional sample of an uneven-aged population, they provide a basis for estimating what changes have occurred in the past and what changes are likely to occur in the future. 3. In order that the old field analysis may be repeated in the future, all the data gathered have been deposit- ed in the Social Science Data Program Library at the University of Wisconsin. The full set of the 1958 aerial photographs may be obtained from either’ the University of Wisconsin Map Library; the Crown Lands Office, Nassau; or the Department of Overseas Surveys, Surbiton, England. E otcr ,norve oven Na ao if. 28 One lbdow ee Ce mer Yo sfqaga = ricwe 1! OCH) 2 Boe d) 89 IX. THE IMPACT OF SHIFTING AGRICULTURE: AGE DIFFERENCES IN THE WOODLAND seen from the air, the Cat Island woodland is a_ patch- work quilt of abandoned fields in various stages of recovery (Figure 16). It was soon recognized that any analysis of man's impact would have to include some consideration of these age differences. Not only were they significant in themselves, but they also represented an inherent variabil- ity that complicated the analysis of the whole. For- tunately, the availability of aerial photograph coverage for 1943 and 1958 made it possible to fix within certain limits the age of any part of the woodland. By sampling fields of different ages, it was possible to determine indirectly the nature of recovery following abandonment. An analysis of succession is in many ways relevant to the hypothesis of insular vulnerability. If the hypothesis is valid, several conclusions might be expected. First, the native species would be ill-adapted to this artificial form of disturbance and would be slow to recover after clearing and burning. Second, the native species would to a large extent be replaced by aliens, especially in the earlier stages of succession. Nal, wlaiiecl, whe lomg lalsvoiry Oi repeated clearing and burning would have brought about a marked reduction in the number of native species present. Because of the underlying differences in land surface characteristics, the woodland was dealt with in the context of the three habitat- types: the whiteland, flatland and blackland. Figures 25, 27, and 29 show in some detail the ehanging floristic composition in each habitat-type at dif- ferent stages of recovery. The distinction between "impor- tant" and "minor" species is arbitrarily based on whether or not the species in question accounts for more or less than 5 percent of the area sampled in any one age-class. The minor species are shown collectively according to the age-class in which they reach their maximum value. As the diagrams are largely self-explanatory, what follows here is simply an elaboration of their more important characteristics. The account of each habitat-type concludes with a discussion of a representative aerial photograph sequence. The Whiteland The whiteland is a distinctive habitat-type and it is not surprising, therefore, that succession here is different 90 (sefosds JueytoduyT) wetsetq uoTssao0ng pueTEyTUM ‘“VWG2e amsry NOISSIIINS GNVISLIHM B1BJINOAU! BUeJUeT sSqniys 10 Saat ON sminssty sns0y2s09 5 esogwAsoo eiyze/pung I eyeyda20Ine evaerney | asuahay wnigos/azay iid ! eop buoy eignssoy esayan @90/09909 H eyjAydorsoys e1seay sijeuoisuajdas eisousay ! I I esonjns, sieyusz I I ! erjojaeisnja eiseseg ; S$ | St | sz 1 Gt { $2 I ol st oz sz 0z ol of Siete) ler) ob a we SO ee pipe Ween ieee Sl geese free ta beet orm peep seen (OY fem eee Nell pea | tee peat Le Glz - Su - ti-S GLE - 62~41 ogi - 0S-0€ GL 0s< aajidWvs SHVIA ‘WOS NI 39V $3193dS LNVLYOdWI 91 (sefToeds JouTW) wetsetq uoTsseoons puel[s TUM °dGe ems 4 NOISSAIINS GNVISLIHM Wl xipuaddy aas , « SIBYIO E + siueaul] u0j019 1 x SWayyO 7 + i} Sapoigosad emumjes 1 ~~ S130 Z] + I I eyndiasid erplasid 1 ojjauyed jeges ! x 40YI0 | + i] I I 1 sisuahay eurnber I l ! l s | s | o | a | sol ry >) (=) a | i} Giz pe SUL tl-S GLE 62—SL OSl 0S—0E GL - 05¢ GI1d NWS SUVIA WOS NI d9V S3193dS YONI 92 from that on the flatland or blackland. The main difference is that, in all age classes, a comparatively few specwes aceount. for a large part of the tovall Veover-. Gundlachia corymbosa, Corchorus hirsutus, Leucaena Jleucocephala and Lantana involucrata dominate the first three age-classes (Figure 25A). All four are pioneer species, well-adapted to quickly colonizing open, droughty surfaces. Any particular field is usually invaded by only one or two of themiiioug species, and these will then dominate the early stages of succession. Floristic differences between recently aban- doned fields are probably due to chance factors, such as the timing of abandonment, rather than to basic environmental eonitrollse Gundlachia corymbosa is a shrubby composite rarely more than 2 meters tall. It grows naturally in the low, droughty areas around the margins of brackish ponds and salt-water lagoons, and from here has Spread rapidly into artificially-disturbed habitats such as roadsides and aban- doned fields. Its success can also be attributed to its prolific seed production and efficient dispersal capacities. Like many members of the Compositae, its seeds are dispersed by the wind. Corchorus hirsutus appears to be very similar to Gundlachia as far as habitat tolerances are concerned. It grows naturally in unstable dune environments inland from the coast. The fruit is a dehiscent capsule containing many small wind-dispersed seeds. Leucaena leucocephala unlike the other whiteland pioneers, | is an introduced species. Significantly, it is only found in artifically- disturbed habitats such as roadsides and abandoned fields. Like Gundlachia and Corchorus, its seeds are largely wind- dispersed. However, its success can also be attributed to its ability to reproduce from sprouts. The only other pioneer species to reach more than 5 percent in any particular age-class is Lantana involucrata. This aromatic shrub is somewhat smaller than the three species discussed above, rarely reaching 2 meters in height. It was seen growing in naturally-disturbed habitats, such as dunes and beach ridges, and also in droughty areas surround- ing the salt-water lagoons. Unlike Gundlachia and Cor- chorus, Lantana depends upon birds for seed dispersal. This species has small, fleshy fruits that are conspicuously blue Wi Woe S The pioneer shrubs are replaced after about 30 years by taller bushes, particularly Pithecellobium keyense, Torrubia longifolia, Coccoloba uvifera, and Acacia choriophylla. All of these species, with the exception of Torrubia, grow in dense thickets which gradually increase in area and eventu- ally merge. The decline of the pioneers is probably in 93 large part due to competition from the taller, more deeply- rooting species, although it is also possible that they are naturally designed for only a_ short lifespan. Older Gundlachia bushes, for example, were often seen to be dam- ‘aged by root rot and other pathogens. Pithecellobium keyense is a leguminous shrub or small tree that rarely reaches a height of more than 4 meters It grows naturally in droughty areas, such as dunes_ and beach ridges, or the scrub-lands just above the salt-water MAP OOnS ue SmSCedsmalgencoveredawith bright Ted aril, an obvious adaptation to encourage dispersal by birds. As Fig- ure 25 indicates, Pithecellobium is capable of colonising recently-abandoned fields but tends to become more important in the older age-classes. A very similar successional pattern is shown for Torru- bia longifolia. This small tree grows naturally in open, droughty habitats such as sand dunes and beach ridges. and the occassionally flooded areas around the edges of the savannas. Like Pithecellobium, its fruits are well-adapted to dispersal by birds. A Torrubia tree in fruit is virtu- ally covered with bright red berries. In the penultimate age-class (30-50 years), Coccoloba uvifera accounts for nearly 25 percent of the total cover. According to Britton and Millspaugh (1920:116), this bushy tree reaches 15 meters in height in certain areas of the Bahamas. However, on Cat Island it was rarely seen to be more than 6 meters tall. It covers extensive areas of the whiteland, both along the coast and back into the dunes and beach ridges. Characteristically it forms dense thickets which gradually expand into formerly-cleared areas. As might be expected from its wide distribution along the eoastlines of the New World tropics, its fruits are dispersed by ocean currents. Locally, however, birds, erabs, and man are important dispersal agents. As Figure 25A indicates, it was not encountered in the recently- abandoned fields. Its absence suggests that its seedlings need a certain amount of shade in order to become esta- blished. A similar successional pattern is shown by Acacia choriophylla. Like Coccoloba, this small leguminous tree was not encountered in the recently-abandoned fields. It is not a prolific seed-producer and probably depends on birds, erabs, and lizards as means of dispersal. Proportionally it is more common on the whiteland than elsewhere in the wood- land, which suggests that this may be its natural habitat. Its thick leaves are probably an adaptation to the dessicat- ing winds that characterize the areas close to the coast. 94 Figures 26 and 27 show a time-lapse sequence for a represen- tative whiteland area, at south Bird Point. The most sbvi- ous difference between the two photographs is the decrease in the area cultivated. The generally darker tone of the whiteland on the more recent photograph does in fact represent the recovery of the woodland. Three general stages of recovery can be identified. The flat grey tone, so widespread on the earlier photograph (for example at A), represents recently abandoned fields in which the cover is largely grasses and herbaceous weeds (see Figure 18). In 1958 many of these areas were characterised by a darker, fine stipple pattern, as for example tavens This represents the low shrub cover characteristic of fields between 5 and 15 years old (see Figure 19). The expansion of the taller, more deeply rooted, bushes can be identified at several locations in the older photograph (for example C). The beginnings of this later stage are also shown in Figure 41. In general, the south Bird Point sequence supports the idea of a comparatively slow recovery on the whiteland. Several fields in cultivation in 1943 can still be identifed on the 1958 photographs. This is rarely the. case sfomeeue other habitat types. On the other hand recovery is obvi- ously taking place and given no further disturbance it is not too difficult t® visualise the whiteland being eventually covered with tne woodland. Figure 26. Figure 27. Whiteland, Whiteland, SOuUela WaiLiec| Peo alinw SOvigia wBiliecel POLiMie 95 96 Older woodland was hard to find on the whiteland because it has been so intensively used for agriculture. Age-class 5 (more than 50 years) is therefore based on a comparatively small sample. The data do suggest, however, that the bushy character of the woodland eventually changes as trees such as Reynosia septentrionalis and Erithalis fru- ticosa become more important. Even so, several smaller bushes, for example, Pithecellobium, Torrubia, and Casasia, still account for a significant —percentage sof themucomal cover. The most important species in the oldest age-class is Reynosia septentrionalis. It is also common on droughty Sites on the blackland and flatland. Like moSt of "the species discussed above its natural habitat appears to be exposed limestone ridges just inland from the coasu- aces seeds are dispersed by birds and small animals such as crabs and lizards. The last two important species on the whiteland, Erithalis fruticosa and Casasia clusiaefolia,~ are (bora members of the madder family. Erithalis is a bushy tree which rarely reaches more than 4 meters in height. It is found in the woodland throughout the island but is _ espe- cially common on the whiteland. Its small purple fruits are well-adapted to dispersal by birds, although it does not become established on the recently abandoned fields (Figure 25Ay) Casasia clusiaefolia, has fruits that are adapted to dispersal by ocean currents, and probably because of this it is largely restricted to the whiteland. It is a shrub with a capacity for rapid growth and is characterasticalia present as a minor member of the whiteland thickets. es seeds are locally dispersed by birds, crabs, and lizards. An account of all 28 minor species encountered on the whiteland is beyond the scope of the present discussion. As can be seen from Figure 25B, they account for only a small percentage of the total cover in every age—-classe suum floristic simplicity contrasts sharply with the restvon me woodland. Figures 26 and 27 show a representative time-lapse sequence for the whiteland. The Flatland The flatland is also a distinctive habitat-type. The land surface consists of horizontally-bedded marine lime- Stones of Pleistocene age. The limestone is indurated-and the surface therefore tends to be droughty. This @iiecoums 97 (Sefosds queqzoduy) We4B8eTq uctssaoong pueTyetq ‘yee eun8ty NOISSIIINS GNVILVI4 siseaul) uojo1g Sqmys 10 saad ON esoguiAso3 B1y2e;puny asuahay Winigosjazay iid eyoy Guo e1gnssoy sisuaueyeg euojisAT winueysadwes wnjAxojewaey ! eqnseuis esasing ! ejAydolsoya e19e9y U ! e1/Opissanip &G0j09909 ! winsajixo) wnidojayy 1 wnuissipiaoy uojAxosapig | | I Ss | sz | gt eal g s a el Hal OL g s Se Lo Hie fee a) fenton ty paar | Fale fitael! SSNS Sl “Solve 'ssltatanleetcan 00s g> 000 62-SL OS 0S5-0€ 0Sz - 05< Q31d INS SHV3A Tana 39V $3194dS LNVLYOdII 98 (sefoeds ZOUTW) weuSeTq ucTsseoonsg pueTyety “ase aun3t4y NOISSIIINS GNVILV14 « SOMO ZZ + Al xipuaddy aas . esgers epsenang , Mayo gt + H (66niy eg0j03909 ! < Uayyo (| + | l seuolsjuajdas eisouday I | » MAO pL + H H ! e1eA0 e1sassnog l 1 1 » MAYO Z] + | ! | | ejeauna ewiuossAg 0z i Sz 1 02 { Sz \ 0z | ae ee a | Les) Ja Dr seas Lo al . | Gmc Woy 005 -S> S18 7 a 00v 62-SI 0Sb 05 0€ 0sz - 05¢ Baas \ } SUVIA NI 49V S$il94dS YONIW 99 means that recovery after clearing and burning is slow. As Calc CN ite UPC COA a jUStlOVeR Siipercent of the ground in fields abandoned less than 5 years ago was bare of trees or shrubs, a slightly higher percentage than on the whiteland. Even more important is the greater floristic diversity of the flatland. Although the number of important species is the same as on the whiteland, together they account for a much smaller percentage of the total cover. The balance is made up by 88 minor species (Figure 28B). Meadia, te is obviously impractical to deal individually with the minor species. The discussion that follows therefore deals largely with the dominants.. Croton linearis and Gundlachia corymbosa are the only two important species to reach their maximum cover values in Lhe ies trom second) Vagie—cllaisiseis- GroOwom IiMeceiets as e) small, narrow-leaved member of the spurge family. It is not a prolific seed producer and relies primarily on birds as a means of dispersal. It grows naturally in exposed areas, for example, behind the coast, above the salt-water lagoons, and around the edges of the seasonally-flooded savanna. From these naturally droughty habitats it has invaded the recently-abandoned flatland fields. It also grows on the whiteland and blackland, but reaches its maximum cover value on the flatland. As Figure 28A indicates, Gundlachia is clearly less important on the flatland than on the whiteland. Increased competition may be the reason for this, although environmen- tal differences between the two habitats could also be important. Even so, Gundlachia's age distribution is much the same on the flatland as on the whiteland. Like Croton linearis, it is a typical pioneer species. In the third age-class (15-29 years), four important Species reach their maximum cover values: Pithecellobium keyense, Torrubia longifolia, Lysiloma bahamensis, and the introduced dyewood Haematoxylum compechianum. Together they account for 40 percent of the total area sampled. Pithecel- lobium and Torrubia have very similar cover values in each age-class and presumably have similar ecological tolerances. Lysiloma bahamensis is a common species on the flatland. A fast-growing leguminous tree, it is well-adapted to rather droughty conditions. It has extensive, horizontal roots that run along the level limestone surface. According to Britton and Millspaugh (1920: 158), in some areas of the Bahamas it may reach 16 meters in height. On Cat Island, however, it was rarely seen to be more than 10 meters tall. It reproduces vigorously by Sprouts, and is also a _ prolific 100 Figures 29 and 30 show a time-lapse sequence for a represen-— tative flatland area just south of Flamingo Point. This ais a particularly interesting area in that it has been rela- tively undisturbed. It has been protected by inaccessibil-= ity, the nearest settlement being 5 kilomenters away. The low elevation of the flatland here is indicated by the salt-water marsh at A. In spite of the obvious differences in the quality of the photographs, their tonal variation gives a good indica- tion of the rate of recovery. The dark area on Figure 29 represents woodland that was probably over a hundred years old. The light grey tones around its edges indicate recent clearing. Some of these fields were abandoned shortly after 1943 because their outlines are still visible on the 1958 photograph (8B). However, their tonal appearance is dif- ferent; on the 1943 photograph they are light grey, on the 1958 photograph medium grey. This tonal change suggests that medium grey areas on the 1943 photograph were probably cleared around 1930. Another point to note is that small fields tengigeo recover more rapidly than large fields. This can be seen by comparing B and C on both photographs. The reasons for this are undoubtedly complex but probably involve seedling supply and microclimatic differences. In general the flatland fields appear t2 recover more quickly than the whiteland fields, recently cleared fields on the 1943 photograph being rarely visible in 1958. 101 Figure 29. Flatland, Flamingo Point 1943. Figuee 30, Pilenwilamcl, Wiles) Weeslialie 1')5yshe 102 seed-producer. Its pods are dehiscent, the small dark brown seeds being widely dispersed by the wind. In the intermediate-age fields it effectively replaces the pioneer bushes, such as Gundlachia and Croton linearis. After thirty years, however, it in turn is replaced by other species (Figure 28A). The introduced dyewood, Haematoxylum campechianum, accounts for just over 5 percent of the cover in the intermediate age-class. Surprisingly, it is the only alien among the 11 important species in this habitat-type. In the penultimate age-class (30 to 50 years), two important species reach their maximum cover values: Bursera Simaruba and Acacia choriophylla. Bursera is aerapidly- growing tree, easily distinguished by its birch-like bark. Like many of the woodland species, it fruits ‘proli@figaie and reproduces vigorously from sprouts. These characteris-= tics have made it an important species in the woodland. As Figure 28A shows, both Bursera simaruba and Acacia choriophylla have similar age distributions. Had the area sampled on the whiteland been greater, the same would prob- ably be true in Figure 25A. Both species have similar eco- logical tolerances. In the oldest age-class (greater than 50 years), Coc- coloba diversifolia and Metopium toxiferum emerge as dom- inants, accounting together for 45 percent of the total area sampled. Sideroxylon foetidissimum also reaches its highest value, accounting for just over 5 percent of the total. Coccoloba diversifolia is an erect fast-growing tree. On Cat Island it was rarely seen to be more than 10 meters tall. As a prolific seed-producer and efficient sprouter it is well-adapted to colonising abandoned fields. It) iwals encountered in all age-classes on the flatland (Figure 28A). Its importance in the older woodland can probably be attri- buted to its ability to outgrow the more widely-branching species such as Lysiloma bahamensis, Bursera simaruba, and Acacia choriophylla. The same is probably true of Metopium toxiferum. Like the pigeon plum, Metopium is an erect fast-growing tree. It also fruits prolifically, with birds being the main means of dispersal. Unlike the pigeon plum, it was not encountered in the recently-abandoned fields. Its absence suggests that it needs a certain amount of shade in order to become established. It is also a less efficient sprouter than Coccoloba. Sideroxylon foetidissimum is clearly restricted to the older woodland. According to Britton and Millspaugh (1920: 321), it grows to 25 meters in some parts of the Bahamas. On Cat Island, however, it was never seen to be more than 12 meters tall. This valuable timber tree has been selectively cut and is probably less important now than it was in the pre-European period. As far as the minor species are concerned, it can only be emphasized that they account for a large percentage of the total cover, over 40 percent in most age-classes. Their collective age distributions in Figure 28B are interesting in that they suggest that many of the species are present in several age-classes. Only the species reaching their max- imum cover values in the older woodland have a limited age distribution. As might be expected, these are sensitive species, intolerant of the open, droughty conditions that characterize the recently-abandoned fields. Figures 29 and 30 show representative time-lapse sequences for the flat- land. The Blackland The blackland differs from the flatland in that the limestone surface is broken and irregular. Soil erosion is therefore reduced and colonization by plants is facilitated. This is shown in the "no trees and shrubs" curve in Figure 31A. In every age-class, the area of bare ground is propor- tionally lower for the blackland than for the flatland. In fact, the rapidity with which succession takes place is an important reason for the early abandonment of fields on the blackland. The blackland is also floristically diverse. A total of 105 different species was encountered in the sam- pling, only 10 of which reached 5 percent in any one age- elass. The youngest age-class (less than 5 years) is particu- larly diverse. Although no important species reaches its maximum cover value here, 30 minor species do. iia wlag second age-class (5 to 14 years), Gundlachia corymbosa, Bourreria ovata, and Pithecellobium keyense reach their max- imum cover values. Of these, only Gundlachia is restricted to the younger age-classes. As on the whiteland and flat- land, it was not encountered in woodland over 30 years old. BowIeeCete OWE AS El Swell wreee Oe Seewloy leeleCuliy juiores than 5 meters tall. It grows naturally in droughty areas close to the coast and near the salt-water lagoons; from here it has spread into the abandoned fields. Its bright red fruits are primarily dispersed by birds. It also Sprouts vigorously, and because of this is commonly encoun- tered in the recently-abandoned fields. Pithecellobium keyense has almost the same age-distribution as Bourreria. Unlike the situation on the whiteland, where it reaches its maximum cover value in the oldest age-class (more than 50 years), here it has its maximum in age-class 2 (5 to 14 years). This illustrates well the basic difference between the xeric whiteland habitat and the more mesic blackland. 104 (sefosds queqioduy) wetsetq soTsseo ong pueTypoetd ‘“VIt aIMST A NOISSIJINS GNVINIVIE esoqwAso2 eiy2ejpung Sqnuys 10 Saad} ON eJeA0 elsasinog r . asuahay WNigo//azayiid 1 eo buoj eignssoy i sijeuolsjuajdas eisouday eqnsewis esasing eyjAydoroys eie7y | | | | 1 | ' sisuaumeyeg ewojisaAy ' ! ! ! ! ! | 1 ! | ! | | | I | | | | ! | \ B//OfISIAAIP 90/0990] H : wnsajixo) wnidojay I | I 1 G2 1 52 1 oO | ol | eee yal)! Oo ol ol ol oc 77 pan eee eer are dee Ue Raa) See el Wea Sa BS at ee > 5 056 e-5 o0et - 62-Gl 009 — 05-0€ oot 0s< anawys| J saya, NI 39V S$3193dS LNVLYOdWI 105 (Ssefosds JOUTW) weddeTq UOTSSeDoNs pUueTYOeTE “dIe sIMstTy NOISSIIINS GNVWINIV1E AL xtpuaddy aas x Slayyo GZ + eyeydasz0anay euaeona ~ * SUBIN0 GL + ejesmnjonul euejueyz % SUBWO ff + « S120 ZI + epijjed eingagey eyopayes syoydig + Saipo Qt + I esapiuaya SHAW Y I 4 ! Sl I SL ! o£ oa —T41 Fo I) 1 (antenatal aeta 5) 056 - t1-S OOE! » 62—-SL 009 0S—O€ 00€ - osc Bree J su NI 49V $3193dS YONI 106 Figures 32 and 33 show a time lapse Sequence for a represen- tative blackland area near Dumfries. Unlike the whiteland and flatland sequences discussed earlier, there was no decrease in cultivation here between the years 1943 and 1958. The reason for this was probably accessibility. The settlement of Dumfries is only half a kilometer away to the WE'SIG. Again the tonal variation between the two photgraphs Bives a good idea of the age of the woodland. Only a small area appears dark on both photographs, as for example at A. The dark area in the lower left part of the photographs (B) is an area of red mangrove around the Dumfries Blue Hole. Dark areas such as A were probably at least 50 years old in 1958. Again the change from light grey to medium grey, as for example at C, indicates the rate of recovery. By analogy it Seems likely that the medium grey areas on the earlier pho- tograph were cleared around 15 years before to 1943. For the area as a whole it is interesting to note that Such a large proportion (more than 75 percent) of the wood- land shows evidence of having been cleared within the time Span covered by the two photographs. On the other hand the rate of recovery is obviously rapid. Few recently cleared fields on the 1943 photograph are visible on the 1958 photo- graph. Figure 32. Figure 33. Aerial Photograph of Blackland, 1943 Aerial Photograph of Blackland, 1958 107 The same is true of Torrubia longifolia. On the whiteland Torrubia reaches its maximum cover value in the oldest age- Class (more than 50 years), while on the blackland it has its maximum in the intermediate class (15 to 29 years). In the penultimate age-class (30 to 50 years) , Reyno- Sia septentrionalis, Bursera simaruba, and Acacia choriophylla reach their maximum cover values. All three have been discussed above and therefore need little addi- tional comment here. It is interesting to note, however, that Bursera and Acacia have very similar age distributions. This reinforces the conclusion reached earlier that both species have similar ecological tolerances. In the oldest age-class (more than 50 years), Lysiloma bahamensis, Coccoloba diversifolia, and Metopium toxiferum all reach their maximum cover values, and together account for 55 percent of the total cover. Lysiloma appears tomie more persistent on the blackland than on the flatland, although why this should be so is not immediately apparent. Metopium and Coccoloba have very similar age-distributions in both habitat-types. As far as the minor species are concerned, the obvious eonelusion is that the woodland is floristically diverse in all age-classes (Figure 31B). The implications of this are discussed below. Figures 32 and 33 show representative time-lapse sequences for the blackland. The Woodland as a Whole The old field data show quite clearly that there are important differences in the nature of succession in the three habitat types. Two of the variables measured, height and floristic diversity, are shown in Figure 34. The white- land is clearly distinctive. Compared with with the flat- land and blackland it is floristically impoverished. An average of less than 7 species were encountered in each field sampled, regardless of age. Furthermore, only 12 species accounted for more than 75 percent of the total cover (Figure 25A). The whiteland is also different in that its rate of recovery after clearing and burning is rela- Gaviely’ ~ show: This can be seen from the aerial photographs and also from the fact that pioneer species, ‘SucHheae Gundlachia corymbosa, Leucaena leucocephala and Corchorus hirsutus are more important and more persistent 9on the whiteland than elsewhere. Similarly, those species that are characteristic of the older whiteland fields, such as Torrus bia longifolia, Reynosia septentrionalis, Erithalis fru- 109 eover values in intermediate-aged fields in the rest of the woodland. There is no equivalent on the whiteland to the "highwood" that characterizes the older woodland elsewhere on the island. This pioneer aspect of the whiteland prob- ably wemlects! the anherent instability of the habitat. The whiteland vegetation is also distinctive in its generally shrubby nature. Even the older trees were on the average less than 4 meters tall (Figure 34). The stunted aspect of the vegetation is probably in large part due to its exposed location near the coast. Other physiognomic variables were not recorded, but it is probably correct to say that there is a higher percentage of evergreens on the whiteland then elsewhere. Floristically there are no major differences between the whiteland and the rest of the woodland. A few species, such as Casasia clusiaefolia, Scaevola plumierii, and Coc- cothrinax argentea, are restricted to the whiteland; how- ever, most species, are found in all three habitat-types. Differences between the the flatland and blackland are less clear. In terms of floristic diversity, the number of species encountered was, on the average, slightly higher on the blackland than on the flatland, especially in the younger age-classes (Figure 34). The rate of recovery, at least in the earlier stages of Succession, appears to be more rapid on the blackland than on the flatland. This can be seen when one compares’ the aerial photograph sequences (Figures 29, 30; Figures 32, 33) and when one compares the "no trees and shrubs" curves on Figures 28 and 31. Vegetation recovers more quickly on the blackland than on the flatland, for several reasons. The surface of the limestone is more broken, which means less soil erosion and a generally more mesic edaphic situation. The broken surface also reduces the severity of fire, and therefore increases the frequency of sprouting. All of this is probably responsible for the greater floristic diversity of the blackland and the greater height of the vegetation in any particular age class. As far as the individual species are concerned, there are few major differences between the two habitat-types. Perhaps the most significant is the greater importance of Lysiloma bahamensisand Haematoxylum campechianum on _ the flatland. This may be a result of more droughty conditions on the flatland. The fact that the curves for individual species are generally smoother for the blackland than for the other two habitat-types is probably a reflection of differences in the area sampled in each case. The smooth- ness of the blackland curves is reassuring, in that it 110 AVERAGE NO OF HEIGHT SPECIES IN METERS <5 5-14 15-29 30-50 »>50 15-29 30-50 >50 AGE CLASS AGE CLASS ———— BLACKLAND = JFLATLAND rst. aa WHITELAND Figure 34. Graphs showing Diversity and Height against Age, for each Habitat Type 111 suggests that real processes are represented. Taking the woodland as a whole it is probably fair to say that the whiteland is quite distinctive in all the age- classes, whereas the differences between the flatland and blackland are only significant in the younger and intermedi- ate age-classes. The Nature of Recovery In view of the supposed vulnerability of insular vege- tation the old field data are in many respects surprising. Perhaps the most striking characteristic of the wood- land is the speed with which it recovers. As can be seen from the aerial photograph sequences, fields recently eleared in 1943 were barely visible in 1958. Although there are some exceptions (on the whiteland and in exposed areas), elsewhere woody plants very quickly colonize the limestone surface. The woodland is in fact inherently weedy. This is also shown, at least for the flatland and blackland, by the way in which most of the important species are encountered in every age-class. In other words, the recovery of the woodland does not involve the classical succession of Species, each replacing the other, but rather a change in the area covered by species already present. Even the older woodland consists, .for the most part, of species that are capable of colonizing recently-abandoned fields. It is interesting to speculate whether this weediness is in itself a result of repeated clearing and burning by man with species pre-adapted to this type of disturbance having increased at the expense of the more sensitive types. Whether or not this has actually happened is impossible to determine without evidence of the composition of the pre- settlement vegetation. Fortunately, the problem can be approached indirectly. In the following section, areal variation in floristic composition is used as a clue to determine to what extent man has modified the woodland. Another surprising aspect of the old field data is the great number of native species present in the recently aban- doned fields. According to the hypothesis of insular vul- nerability, very few native species would be expected to have adapted to the artificial habitats created by shifting agriculture. Yet on Cat Island, particularly on the flat- land and blackland, literally dozens of native species are involved in the early stages of succession. Furthermore, the number of species encountered actually declines with age (Figure 34). This decrease in species diversity is the 112 Opposite of what usually occurs in the early stages of suc- eession (Loucks, 1970: 17; Odum, 1971 ° 256). “Inepaqceme must be admitted, it may be an artifact of the constant area sampled. Fewer species were encountered in the older wood- land because the larger trees took up a greater proportion of the area sampled. Even so, it is generally true that the older woodland contained fewer species than the younger woodland. As can be seen from Figures 28 and 31, two species, Metopium toxiferum and Coccoloba diversifolia, dom- inate the older woodland. Again, it is interesting to speculate whether this floristic impoverishment is in itself a result of repeated clearing and burning and that slow-growing, shade-tolerant species were formerly more numerous in the woodland than they are today. This possibility is also considered in the following section. Another significant aspect of the old field data is the limited importance of alien species. Only seven aliens were encountered, and together they accounted for less than 6 percent of the area sampled. This low total was not entirely unexpected in view of the historical evidence discussed ear- lier, but even so it does not conform with the usual role of alien plants on small oceanic islands. According to several recent interpretations of insular vulnerability (Elton, 1958; Harris, 1965), alien plants have a competitive advan- tage over native species in vegetation that has been dis- turbed by man. The whole of the Cat Island woodland has been drastically disturbed during the past thousand years, and yet few aliens have been able to get established. In summary, the analysis of age-differences in the woodland has_' produced several results. First, the rate of recovery after clearing and burning has been shown to be Surprisingly rapid. Second, the data indicate that a great many native species are pre-adapted to withstand the effects of clearing and burning. And third, the invasion by alien species has had only limited success. In view of the sup- posed vulnerability of island life all three findings were unexpected. X. THE IMPACT OF SELECTIVE PRESSURES: AREAL VARIATION IN THE WOODLAND On many small islands, woody vegetation has been virtu- ally removed by grazing, selective cutting, and burning. On Cat Island this has not been the case. The woodland has Survived. Even so, the question remains as to what extent its floristic composition has been changed. As was emphasized, the data gathered in 1970 show the nature of vegetation change in an indirect way. Comparative analysis of abandoned fields of different ages indicates the probable nature of succession, but it does not show the actual changes that have occurred in a truly historical sense. These can only be determined from historical evidence, and HOGmaCatmetsieand ssvhe historical Mevadence is frustratingly thin. In spite of this deficiency, the question as to what extent man has changed the floristic composition of the woodland can be approached in other ways. One approach would be to monitor future changes. alii the frequency of shifting agriculture continues to decrease, it would seem likely that those species sensitive to clear- ing and burning will increase in importance at the expense of the weedy types. Just how long it would take the wood- land to recover to its pre-settlement condition is difficult to assess. Recovery at present may be rapid, but it is by no means complete. A second approach would be to analyse the areal varia- tion within the contemporary woodland. More specifically, by comparing disturbed areas close to the settlements with remote, comparatively undisturbed areas, it should be possi- ble to determine something of the selective nature of man's impact. This was the approach taken in the present study. As was indicated earlier, the intensity of many selec- tive pressures decreases with distance from the settlements. Because of this, each of the 300 sample sites was classified in terms of its distance from the nearest settlement, so that the nature of floristic variation with distance could be computed. In order to simplify the analysis, the 42 whiteland sites were not included. Their floristic composi- tion is so different from the rest of the woodland that to include them would have introduced too many uncontrolled Variables. On the other hand, as the blackland and flatland had been shown to be floristically similar in the age- vari- ation analysis, they were combined in this part of the study. The resulting 258 sample sites were classified, both as to age and to distance from the nearest settlement. 114 DISTANCE FROM SETTLEMENTS [ia <1KM fF] 1-2 km : 2-3 KM >3 KM Figure 35. Map showing Distribution of Distance Classes Cem inhi helchial ysis iat Decamenehear tbhaty the fifth age= class Cmore whan 50 Wears), Alacl” waa ied elec) eyes distance-classes (4 to 5 kilometers; 5 to 6 kilometers) did not have adequate sample coverage to make reliable interpre- tation possible. They were therefore combined with the fourth age-class and fourth distance-class, respectively. The areal distribution of the four distance-classes is shown in Figure 35. Goats and the Woodland Grazing and browsing by domesticated animals have drastically changed the vegetation of many parts of the world. Especially significant have been the changes brought about on small oceanic islands where native plants have evolved without defensive mechanisms against such pressures. On Cat Island, horses, hogs, sheep, cattle, and goats have all had an impact on the woodland. For several reasons, however, it was decided to limit the analysis to a study of the effects of grazing and browsing by goats. Initial observations in the field suggested that of all the domesticates on the island, goats had probably had the most important influence on the floristic composition of the woodland. They were the most numerous domesticate, and unlike sheep, cattle, and horses, which only ~browse under duress, they are quite happy to eat the leaves of bushes and trees. They prefer young leaves, shoots, or seedlings, and in this way influence regeneration rates. A recently-grazed area has a characteristically "clean" appearance, the sur- face having been stripped of all herbaceous growth and see- dlings (Figures 36, 37). In order to determine the extent to which goats had modified the woodland, a working hypothesis was proferred. It was assumed that palatable species would be rarer close to the settlements than in remote areas, while the opposite would be true of unpalatable species. The first problem, then was to determine which species were palatable and which were unpalatable. Palatable and unpalatable species. Contrary to popular opinion, goats do have discriminat- ing palates; some species they prefer and others they avoid. In order to determine these preferences, two approaches were used. First, local inhabitants, experienced in agricultural matters, were interviewed on the subject of goat-feeding habits. In this way, a basic check-list of preferred and 116 Figure 36. An area near Bennet's Harbour intensively grazed by goats. Note the Leucaena bushes in the center of the photograph have been browsed to a height 5f about one and a half meters. The other bushes in the foreground,Cassia bahamensis, are unpalatable to goats. Figure 37. Another heavily grazed area near Bennet's Harbour. Cassia bahamensis is very common on the level ground where virtually no herbaceous growth has survived. avoided species was set up. Second, the reliability of this information was checked by offering the species in question to hungry goats. Perhaps significantly, only five common woodland species were found that the goats really like to eat (Table 5). In contrast, 18 species were found that even hungry goats refused to eat (Table 5). Most of the unpalat- able species have strongly aromatic or poisonous leaves, the Spurge, legume, and myrtle families being well represented. Floristic composition and distance. Because of the cost of wire, there are few fenced pas- tures on the island, and as a result the goats have to be tethered. This is necessary because of the threat they pose to crops, and in turn means they have to be tended at least once every other day. The goats quickly exhaust the accessi- ble food supply and have to be moved to new areas. Also, they have to be watered because of the lack of surface water. Because they have to be tended so frequently, the local people are reluctant to tether them too far from home, and consequently the intensity of grazing pressure decreases as a function of distance from the settlements. In order to determine to what extent grazing and brows- ing had changed the woodland, the percentage cover values for the palatable and unpalatable species were calculated at the different distances from the settlements. The resulting graphs are shown as Figure 38. Originally it was intended to calculate the percentage cover values for each individual Species, but for the sake of simplicity they are presented here as two groups. Reassuringly, the cover values show that, with a few exceptions, there is a positive relationship between palata- bility and distance, and a negative relationship between unpalatability and distance. Furthermore, the species included account for about 60 percent of the total area sam- pled, so the curves can be taken as representative of the woodland as a whole. In the youngest fields it is interesting to note that the palatable species are less important than they are in the older age-classes. This is probably due to the custom of tethering goats in recently-abandoned fields. Just why the unpalatable species are also less important is not easy to explain. It may simply be due to the fact that such a great number of species, not included in either group, are present in the recently-abandoned fields. In spite of their low values, the two curves show very nicely the expected relationship between grazing pressure and distance. 118 550 400 425 lia Ages DISTANCE FROM SETTLEMENTS IN Km. UNPALATABLE — SPECIES _— x PALATABLE SPECIES % COVER AREA SAMPLED 0 1700 1700 1250 1800 KEY Graphs showing Age/Distance Unpalatable Species Figure 38. 119 TABLE 5 SPECIES INCLUDED IN THE ANALYSIS OF BROWSING BY GOATS Palatable Species Torrubia longifolia Acacia choriophylla Leucaena leucocephala Bursera Simaruba Pithecellobium keyense OIEwhy = Unpalatable Species Corchorus hirsutus Lantana bahamensis Croton linearis Gundlachia corymbosa Croton bahamensis Coccoloba diversifolia Metopium toxiferum Exostema caribaeum Lantana involucrata 10. Cassia bahamensis 11. Eugenia buxifolia 12. Eugenia monticola 13. Bourreria ovata 14. Croton eluteria 15. Malphigia polytricha 16. Lysiloma bahamensis 17. Piscida piscipula 18. Croton lucidus OMOA DU FW OY = In the second age-class, the sample-sites less than 1 kilometer from the settlements show very definitely the expected relationship, with unpalatable Species being roughly four times more important than palatable species. 120 However, at distances of more than 2 kilometers, the unpalatable species show an unexpected decline, and after 3 kilometers increase again. No obvious explanation can be offered for these changes, other than that some other vari- able besides grazing pressure is involved. Similarly, in the third age-class the percentage value for the palatable species is anomalously low in the sites closest to the settlements. This may reflect a somewhat lower sample coverage in this category, although just how is not clear. For the more distant sample sites, the curves behave as expected, and suggest that the impact of grazing is restricted to within 3 kilometers of the settlements. For the older sample sites the curves show very well the expected relationship in the first three distance categories. Only in the more remote fields does the curve for the palatable species drop unexpectedly. Again, this must be attributed to some variable other than grazing pres- sure, possibly increased competition from species not included in the two groups, conceivably from species that for other reasons are more important in remote areas, such as economically-valuable species or species sensitive to clearing and burning. When all the different age-classes are combined, it is interesting to note that the irregularities average out. The resultant curves show a nice symmetrical arrangement, which suggests that the impact of grazing is restricted to within a zone less than three kilometers from the _ settle- ments. At distances of more than three kilometers the aver- age cover for the two groups is more or less constant. The implications of the data. The first point to be made is that the working hypothesis was proved to be valid. Palatable species are rarer in heavily grazed areas than in lightly-grazed areas, while the reverse is true of unpalatable species. This apparently obvious’ relationship was not immediatedly apparent in the field, and only became clear after the analysis of the data. Also significant is the extent to which grazing and browsing have modified the compostion of the woodland. The woodland close to the settlements might be aptly described as "goat-proof." Also, the fact that so many unpalatable species could be identified in contrast to so few palatable Species was probably not fortuitous. Palatable species have undoubtedly become rarer in the woodland around the settlements. Whether or not this has actually happened can- not be proved without historical evidence, but in view of the data presented above, it seems more than likely that this has been the case. Certainly if the goat population declines it will be interesting to see how the woodland recovers. It is also interesting that grazing pressure is res- tricted to a zone within 3 kilometers of the settlements. This confirms the qualitative impression obtained in the field, few goats being seen beyond this distance. On the same point, it should be noted that just over two-thirds of the sites sampled in the old field study fall within the 3- kilometer limit. It follows, therefore, that the composi- tion of the woodland as described in Chapter 8 has to a large extent been determined by the goat. The differences in the cover values between the dif- ferent age-classes strongly suggest that the intensity of grazing had varied in the past. For example, the palatable Species are generally more important in the 15- to 30-year age-class than in any other age-class, even in the sample sites more than 3 kilometers from the settlements. The pos- “sibility exists that the goat population was low during this time period, although no local information was obtained which suggested this had been the case. The complicated nature of the data makes it difficult to evaluate as far as individual species are concerned. It is interesting to note, however, that the two dominants in the older woodland, Coccoloba diversifolia and Metopium tox- iferum, have very different cover values in the oldest age- elass. Even though both species are unpalatable, the former decreases in importance away from the settlements while the latter increases. Obviously some variable other than graz- ing determines the importance of Metopium. For Coccoloba it seems safe to assume that unpalatability is in large part responsible for its variable importance in the woodland. Other Domesticates The goat is probably not entirely to blame for the changes described above. Sheep, horses, hogs, and cattle have all exerted selective pressures on the woodland, although in each case the pressure has been different. Sheep, for example, prefer herbs, and will only browse reluctantly. Similarly, cattle and horses usually graze on the whiteland, or seasonally-flooded savanna, and only browse in the woodland during the dry season. Furthermore, the number of woody species that horses find palatable is 122 limited. The leaves and pods of the leguminous shrub Leu- caena leucocephala, that provide preferred feed for goats cause horses to lose their hair and even hooves (Little and Wadsworth, 1964). Hogs have had an indirect impact on the woodland insofar as the fruits and young leaves of several palms are gathered for hog feed. Pseudophoenix is known locally as the "Hog-cabbage palm," and is now rare in the woodland close to the settlements. There were no feral hogs on Cat Island in 1970, although they were reported to have been common in the past. Selective Cutting Of all the pressures that man has exerted on the wood- land, selective cutting was the easiest to define. Most of the species involved were identified in the historical record, and those that were not were determined on the basis of local information. Out of the 120 species encountered in the sampling, 15 were known to have been selectively cut on such a scale that their frequency in the woodland was. prob- ably affected. These consist of dyewoods, timber trees, and species valued for fuel or fodder (Table 6). The details of their use were discussed earlier and need not be repeated here. Suffice it to say that their value to man has made them vulnerable. The selective cutting of economically valuable or use- ful species is basically a similar process to the grazing and browsing of domesticated animals. The species that are cut decline in importance, while the species that have no value increase to take their place. However, in the discus- Sion that follows, only the former are included, as_ the increase in species that are not cut was less amenable to analysis than was the increase in unpalatable species. More specifically, it was impossible to determine which species increased at the expense of those that were selectively cut. Originally it was intended to deal with each of the economi- cally valuable species individually; however, because of the limitations of time and space, they are here considered col- lectively. Selective Cutting and Distance There are few roads on the island, and any heavy loads have to be carried by horses. Lighter loads are carried in traditional African style, on the head. Either way, the frictional effect of distance is strong, and as a result remote areas of the woodland have been comparatively undis- turbed. During field work this was recognized in a TABLE 6 SPECIES INCLUDED IN THE ANALYSIS OF SELECTIVE CUTTING Caesalpinia bahamensis Haematoxylum campechianum Croton eluteria Swietenia mahagoni Lysiloma latisiliqua Guiacum sanctum Krugiodendrom ferreum Mastichodendrom foetidissimum Erithalis fruticosa Pseudophoenix vinifera Dipholis salicifolia Amyris elemifera Thrinax microcarpa Callicarpa hitchcockii Vallesia antillana — =) =) —3 3 —) UEW DM = OWONI ADU EWpP = qualitative way. The question remained as to whether or not the same effect would be shown in the old field data. As with the analysis of grazing and browsing, a working hypothesis was established which assumed that valuable Species would be more important in remote areas of the wood- land than in areas close to the settlements. In the two youngest age-classes, the expected distribu- tion is generally followed. The 15 species increase in importance away from the settlements. Even so, it is prob- ably unwise to place too much weight on these curves, as the total values are so low (Figure 39). As was the case with the palatable and unpalatable species, the third age-class (15 to 30 years) includes some anomalies. When compared with the second age-class (5 to 14 years), the first two distance-classes have increased cover values while the last two actually decline, just the oppo- Site of what was expected. The species largely responsible for the high value in distance-class 2 is Erithalis fru- ticosa. Just why it should have been so important here is 325 550 400 425 400 425 225 650 DISTANCE FROM 1 SETTLEMENTS IN Km. All Ages a be > SELECTIVELY o CUT SPECIES = f METOPIUM ey TOXIFERUM AREA SAMPLED IN Sq. m 0 1700 1700 1250 1800 KEY Figure 39. Graphs showing Age/Distance Distributions of Selectively-Cut Species, and Metopium Toxiferum not immediately apparent. The oldest age-class very definitely shows the expected pattern. The cover-values for the sites less than 1 kilome- ter from the settlements are significantly lower than for the last three distance- classes. Furthermore, the cover- values for the last three classes are between 6 and 9 _ per- cent higher than in the previous age-class. The composite curve showing all ages together shows approximately the expected distribution. However, its use- fulness is questionable, because it masks what are very real differences between the different age-classes. Implications of the Data Again the working hypothesis was proved largely eorrect. The assumption that the selectively-cut species would be less common close to settlements than in remote areas is supported by the data. On the other hand, for the first three age-classes the total cover values for the 15 Species are very low. This means that a relationship between distance and selective cutting cannot be clearly defined for the younger woodland. It is interesting to speculate whether these low totals are themselves a result of selective cutting, although again without historical evidence this cannot be proved. It appears that the species involved do not have the ability to recover quickly after cutting. Certainly very few of the 15 Species are common in recently-abandoned fields. In other words, they are more vulnerable to disturbance than are the common woodland species. Somewhat surprising is the narrowness of the zone affected by selective cutting. In the oldest age-class, the impact is restricted to within a radius of 1 kilometer from the settlements. This probably reflects the general decline in demand for the 15 species during the present century. Had the areas sampled been greater the whole question as to how significant have been the effects of selective cutting would be more easily answered. Unfortunately, many of the species that have been selectively cut are still rare, and because of this they were not encountered in the sampling. Fagara flava the dyewood, Buxus bahamensis the timber tree, and Canella alba the medicinal bark were each seen only once in the woodland. The statistical signifi- cance of rare species cannot be adequately measured in a broad survey of the kind undertaken here. In spite of their 126 limitations, however, the old field data do show something of the consequences of selective cutting, and in view of the fact that the demand for many of the species has declined in recent years it should be interesting to monitor future changes. Combined Selective Pressures Any analysis of the effect of selective pressures is complicated by the fact that usually more than one pressure has been involved. For example, clearing and burning, which are non-selective in the sense that fields are established without too much concern as to what species are present, are selective insofar as they give an advantage to species that can sprout or that have the ability to colonize recently- abandoned fields. Furthermore, the frequency of clearing and burning decreases with distance from the settlements in generally the same way as the intensity of selective-cutting and browsing. The net result is that several selective pressures may be exerted on certain species at the same time. Originally it was hoped to analyse the impact of clear- ing and burning in the same way as grazing, browsing, and selective cutting, but unfortunately it was not possible to isolate a group of species that would unambiguously show this effect. It has only been possible to show the impact of grazing, browsing, and selective cutting because the Species involved were known with certainty. This. . Was) Soom the case with clearing and burning. Rather than avoid the problem completely, the following discussion considers the case of the poisonwood tree (Metopium toxiferum); as the second most important species in the older woodland, its distribution is worth considering. Metopium toxiferum Metopium, a close relative of poison ivy and _ poison sumach, is characterized by a caustic sap poisonous to the touch. It is a common tree in the older woodland, and grows in a wide range of edaphic conditions, from the driest sites to the wettest. As can be seen from Figure 39, it has an anomalous distance distribution in the woodland. In the first age-class (less than 5 years), it is rare, and is only present in the areas closest to and furthest away from the settlements. In the second age-class (5 to 14 years), its importance increases slightly while the same distance pattern is maintained. In the third age-class (15 to 29 years), there is a significant increase in the cover values while again the species is more important in the first and last distance classes than in the intermediate classes. In the older age-classes (more than 29 years), there is a continued increase in cover values, although this time there is a progressive increase in importance away from the settlements. The composite age-values show the charac- teristic U-shaped distribution. The variable importance of Metopium with distance from the settlements suggests several interesting possibilities. Its greater importance in the more remote areas suggests that it is sensitive to some pressure positively associated with the settlements. Metopium is cut on a small scale for its timber in certain parts of the West Indies (Little and Wadsworth, 1964: 290), although there is no evidence to sug- gest it has ever been exploited in the Bahamas (Coker, 1905: 205), and according to local reports it has never been cut on Cat Island. Selective cutting, therefore, does not account for the distribution. A more likely explanation is that Metopium is sensitive to clearing and burning. Unlike many of the woodland Species, it is not a prolific sprouter, and its seedlings were rarely seen in the recently-abandoned fields. Figures 28 and 31 suggest that it requires a certain amount of shade for successful germination. Assuming this to be correct, the question remains as to why the cover-values are higher in the sites closest to the settlements than they are in the intermediate classes. A plausible explanation here is that the species is unpalatable to the browsing animals. For obvious reasons it is avoided by goats and other domesti- cates. The variable importance of Metopium in the woodland is therefore probably due to the interaction of at least two selective pressures. Unpalatability gives it an advantage near the settlements, while sensitivity to clearing and burning puts it at a disadvantage; in effect, each pressure is working in the opposite direction. The U-shaped distri- bution suggests that the unpalatability-advantage overrides sensitivity to clearing and burning near the settlements, but that the opposite is true at intermediate distances. Just why this should be so is not immediately apparent, but could easily be determined with more field work. Obviously the combined effect of the selective pres- sures is going to be different for each individual species. A palatable species which happens to be economically valu- able and sensitive to fire is going to be rare even in the remote areas of the island. An unpalatable species of no 128 use to man and with a vigorous sprouting ability is going to be common. Ideally, it would have been preferable to study the effects of selective pressures on three floristically similar islands: one that had only been grazed and browsed; one that had only been selectively cut; and one that had been cleared and burned for agriculture. Unfortunately, three such islands do not exist. Distance and Diversity According to the hypothesis of insular vulnerability, the floristic composition of the woodland should have been significantly impoverished by grazing, browsing, selective cutting, clearing, and burning. The data discussed above hardly suggest that this has been the case. In orden euro consider the question of impoverishment more closely, it was decided to compute the number of species present per unit area sampled in the different age and distance-classes. The resultant totals are shown in Table 7. TABLE 7 DIVERSITY, DISTANCE AND AGE The average number of woody species encountered per 25 x 1m sample. KM. FROM SETTLEMENTS Less than 5 yrs. 15e 70 Saree 15.00 15.20 5-14 yrs. 14.11 14.87 14.95 Le a | 15-29 yrs. 11% 62 10.68 11256 10.41 30 yrs. or older 10.78 BATT 9.00 9.00 All ages 12.40 T2.28 rAey; iTrtz2 Significantly, there are no major differences in Species diversity between the different distance classes. In fact, apart from the second age-class, the sample sites less than 2 kilometers from the settlements have slightly higher averages than those more than 2 kilometers away from the settlements. The important conclusion to be drawn from this is that, although the species sensitive to selective pressures have become rare close to the settlements, they have been 129 replaced by a similar number of insensitive species. The discovery that so many native species are pre-adapted to withstand the impact of grazing, browsing, selective cut- ting, clearing, and burning was unexpected. The question as to why this is the case will be considered later. ae » 4 cranagnt 1a as8% aw ,.-Betoogs. oy -=tasae Se Bitp XI. THE INVASION OF ALIEN PLANTS Unlike the situation prevailing on remote islands such as the Hawaiian islands, comparatively few species of alien plants have become well-established in the Bahamas. Of the 120 woody species encountered during the course of the present study, only six were clearly identified as alien, and together they accounted for only 5 percent of the total cover. The Species Involved Leucaena leucocephala Leucaena is a leguminous shrub with a pantropical dis- tribution. Its original home is usually cited as Central America, although the details of its history are not clear. This is certainly so for the Bahamas. Its alien status is not confirmed by the historical record, but is strongly sug- gested by its localized distribution in artificially- disturbed habitats; characteristically it is a pioneer in abandoned fields, pastures, and roadsides. It was probably introduced to the Bahamas in the early eighteenth century as a fodder crop. One of Catesby's plates (1731, II: 42) looks very much like Leucaena and is identifed as such by Britton and Millspaugh (1920: 162). However, the text clearly refers to the native Lysiloma leu- eocephala, so the determination remains doubtful. The Naturalist Schoepf, who visited the islands in 1784, includes a Mimosa glauca in a check list of the more common plants, but gives no information as to its use or origin. Wery Likely wis Soeetes Me ieevciess wo LS IGCUCEISINE) SvCO@= eephala. On the other hand, there is surprisingly no men- tion of Leucaena in Brown's account of fodder plants in use on his plantation around the turn of the eighteenth century Gs8022 11): A plantation journal from Watlings Island for the year 1831 includes references to the planting of "cow bush" and "cow peas," but the identification of the species involved is not certain (Deans Peggs, 1957: v). What is certain is that by the end of the nineteenth eentury Leucaena was widely distributed around the archi- pelago (Hitchcock, 1893:166). On Cat Island in 1970 it was locally common in recently-abandoned fields, pastures,and along roadsides. 132 Haematoxylum campechianium Logwood is a thorny, leguminous tree with a rather shrubby habit. The heart-wood, which is deep reddish purple, is the source of the dye Haematoxylon. A native of Central America, logwood was widely introduced around the Caribbean during the late seventeenth and early eighteenth centuries (Wilson, 1936). Fortunately, there is a definite date for its introduction to the Bahamas. As was indicated earlier, Catesby reports that it was brought from the Bay of Honduras in 1722. As far as Cat Island is * concerned, it) seems liked that it was introduced by the loyalists at the end of the eighteenth century. Its present distribution is very local- ized in old plantation areas near Port Howe and Old Bight. Manilkara zapota The sapodilly is one of the most common fruit trees in the Bahamas. Supposedly Central American in origin, it was probably first introduced to the islands by the Arawaks. It is easily planted from seed in the Bahamas, and its fruits are characterized by a wide variability in size and _ shape (Morton and Morton, 1946: 88). It was probably re-introduced to Cat Island in the late eighteenth century by the loyalist planters. In 1970 few families did not have at least one tree in their yard. In the woodland it is seen characteristically by the side of footpaths and only occasionally in the middle of ae field. Presumably a large percentage of the woodland trees have grown from seeds, either intentionally planted or casually discarded. Both the fruit and seed are large and are prob- ably not dispersed far by natural mechanisms. Indigofera suffruticosa Whether or not the New World indigo is native to the Bahamas is uncertain. It was cultivated in South America in pre-European times and may have been brought to the islands by the Arawaks (Harris, 1965). There are references to the cultivation of indigo around the turn of the seventeenth century, although what species was involved is not clear (Craton 1968: 89). The Asiatic -indigo (indigofera’ )tinee toria) had probably been introduced by this time. Schoepf (1788) includes an Indigofera argentea in his list of common plants, and according to Britton and Millspaugh this is a synonym for the New World species (1920: 180). In nineteenth century botanical literature, Indigofera suffruticosa (syn. I. anil) is recorded for several of the Bahama Islands, although not for Cat (Hitchcock, 1893: 166). It is certainly rare on the Island at present, and its alien status is suggested by its localized distribution in artificially-disturbed habitats. Gossypium barbadense, G. hirsutum Cotton is one of the few cultivated plants known to have been introduced to the island by the Arawaks. However, that it could have persisted the wild after the Arawaks left seems doubtful. More than likely, the cotton seen in the woodland today can be attributed to eighteenth-century or even later introductions. At least four varieties were cul- tivated in the nineteenth century: Anguilla, Flyaway, Bour- bon, and Georgia (Johnston, 1867). The first three appear to have been varieties of sea island cotton (G. bar- badense), while the latter may have been a long-staple Variety of upland cotton (G. hirsutum). Both species are listed in the Bahama Flora, although upland cotton is only recorded for one island, Rum Cay (Britton and Millspaugh, 1920: 274). The species encountered on Cat Island in 1970 was sea island cotton. A weedy shrub, it was characteristi- cally seen in disturbed habitats, such as roadsides or gar- dens, and only occasionally in the woodland. Heliotropium angiospermum Unlike the five species discussed above, scorpion tail has never been planted. A woody herb rather than a bush, it is a widely-distributed tropical weed. Like many weeds its origins are uncertain. However, its irregular distribution in artificially-disturbed habitats strongly suggests that it is not native to the Bahamas. It has been present on Cat Since at least the late nineteenth century (Hitchcock, 1893: 168), but was rarely seen in the woodland in 1970. The Extent of the Invasion One of the more important conclusions reached by Elton in his The Ecology of Invasions (1958: 142) was that simple communities are more prone to invasion than communities which are rich in species. Harris reached much the same con- clusion in his study of the Outer Leewards. He suggested that the extent to which alien plants were able to become established varied from place to place according to the his- tory of disturbance and the floristic diversity of "ecologi- cal resistance" of the native vegetation. Whether this was also true of the Cat Island woodland was a basic question 134 considered in the analysis of the old field data. In™6the discussion that follows, the three habitat-types types and five age-classes are used as an organizing framework. The Whiteland As can be seen from Table 8, only 2 aliens were encoun- tered in the whiteland samples, Leucaena and Gossypium, with Leucaena being by far the most important. AS was shown ear- lier, Leucaena was in fact the third most important pioneer on the whiteland only exceeded in the first two age-classes by Gundlachia and Corchorus. In the third age-class it replaces Corchorus to a certain extent, but it is still not as important as Gundlachia. In the older woodland it declines quickly (Figure 25). The importance of Leucaena on the whiteland must in part be attributed to planting. In the pre-automobile era a great number of horses were grazed here, and it seems more than likely that some of the Leucaena thickets owe their origin to planting. Their persistence must in large part be attributed to the species's ability to sprout. This method of reproduction is especially effective in the loose _ sand substrate. Clearing and burning have little impact on the shrub, making it a troublesome weed (Figure 40). In intermediate-aged fields, Leucaena uSually forms dense thickets 2 to 3 meters in height (Figure 41). The close cover naturally prevents regeneration by other species. However, as Table 8 shows, in fields over 30 years old it accounts for only 2 percent of the average cover. What appears to happen is that the taller, more deeply-rooted shrubs, such as Coccoloba uvifera and Acacia buxifolia, gra- dually extend their cover at the expense of Leucaena. dif clearing and burning were to cease and natural regeneration allowed to take place, it seems unlikely that Leucaena would be able tp persist. Cotton (Gossypium barbadense) was the only other alien encountered on the whiteland. It was seen only once and in an area very close to the edge of the whiteland. Together with Leucaena and Guilandina, it formed a dense thicket 3 to 4 meters high. According to the early historical accounts, cotton never grew well on the whiteland, and there is no evidence to suggest it was ever planted there in any quan- cacy Although the data do not show it, cotton was more frequently seen on the blackland and flatland than on the whiteland. In either case, it accounts for a very small part of the total cover. Several other aliens were seen on the whiteland,; 135 Bugumges40.. A whiteland conn) field infested with leucaena leucocephala. This LSI 9 elose © une GCdEe of aA seasonally- flooded palmetto thicket, had been abandoned because of the difficulty of eradicating Leucaena sprouts. Bagurest il) Heme wan older Deucaena thicket, probably i5=20 eae, Seine accep Maced by Matave (species. saCasias ila clusiafolia is in the left foreground and Coccoloba uvifera €o the right and in the background. — 136 AGE IN YEARS WHITELAND Leucaendasc. «. Gossypium.... Total FLATLAND Leuicaenarcicre: ei. Haematoxylum. Total BLACKLAND Leucaena..... Haematoxylum. Manidlikarae << Heliotropium. Indigofera... moval ALL HABITATS Leucaena..... Haematoxylum. Gossiy plume. Manilkara.... Heliotropium. Indigofera... Towa! PERCENTAGE COVER ALIEN SPECIES | ul Oo Crorow (Sle) (oS iej(esiS)) <5 ~13 .00 TABLE 8 5-14 aime, SO.roroin) Centetey CU) 15=29 Nied 0.80 18.07 OO 0 OW je) oO LW de a 30-50 oO —_ (SUS MS) (SiS) SVQ ©‘ >50 137 although they were not encountered in the sytematic sam- pling. The most important by far was Casuarina equiseti- folia, the Australian Pine. Although usually restricted to a narrow zone just above the high tide mark, Casuarina has in certain areas escaped inland. It is occasionally seen along roadsides or in similar open habitats. Surprisingly, it has not been able to invade the recently-abandoned fields, although just what prevents it is not immediately obvious. Also occasionally seen on the whiteland was sisal (Agave sisalana). Extensively cultivated in the latter part ofthe nineteenth century, it is still planted on a very small scale. Most of the sisal on the whiteland had the appearance of being planted, although some natural increase may have occurred. More conspicuous was its wild relative Agave americana, locally known as "the Bamboo". Its flower- ing stalk reaches a height of 10 meters or so, while its basal leaves are 2 meters tall. Coconuts (Cocos nucifera) are often planted on _ the whiteland, usually as a means of satisfying government regu- lations, which require "improvement" of land grants. However,the coconut is definitely a cultigen here and no natural regeneration occurs. Overall, very few woody aliens have been able to suc- eessfully establish themselves on the whiteland. Leucaena, which was formerly planted as a fodder crop, is the only alien to have become an important part of the wild vegeta- tion. The Flatland Again, on the flatland only two aliens were encountered in the systematic sampling, Leucaena and Logwood. Taking the five age-classes as a whole, Leucaena was slightly more important than Logwood in both absolute and percentage terms (Table 8). On the flatland the indurated rock surface makes the ability to reproduce from sprouts less advantageous than on the whiteland, and perhaps because of this, Leucaena is less important. Competition from the greater number of species adapted to this habitat has probably also been a factor. Whatever the reason, Leucaena covers a much smaller area of the woodland than it did on the whiteland. On the other hand, in terms of importance in different age-classes the pattern is much the same. It also seems likely that if clearing and burning were to cease, Leucaena would not be 138 able to persist for long. The cover values for logwood cannot be properly inter- preted without some further information. Logwood has a very localized distributiion on Cat Island. It was seen to be especially common in the areas east of Old Bight and north- east of Port Howe. It was never encountered in the northern half of the island, where most of the sampling was carried out. This localized distribution, undoubtedly the result of its comparatively recent introduction, has meant that the average values in Table 8 are somewhat misleading. In the areas where it was encountered it was very common, but the averages are low because over much of the island it has not yet become established. Whether it will or not in the future is another question. The difference in the cover values for logwood in the different age-classes is probably a reflection of the his- tory of land use in the areas sampled. For example, in a large flatland area east of Old Bight, logwood trees 3 to 4 meters tall were surrounded by a great variety of native shrubs and small trees. Logwood regeneration was very poor, although twenty to thirty years ago it had apparently been good. This change in the woodland probably reflects a decrease in the intensity of land use in the area. AS over the rest of the island, in the fifteen years between 1943 and 1956 the area under cultivation decreased significantly, and since 1956 it has declined even further. In open condi- tions thirty years ago, logwood was able to reproduce well, but since then it has been less successful and the native Species have had a chance to recover. Furthermore, in the continued absence of disturbance, it seems likely that log- wood will not only have difficulty in reproducing but will be outshaded by the taller native species such as poisonwood (Metopium toxiferum) and pigeon plum (Coccoloba diversi- folia)i. Both were actually seen to be growing up through the laterally-branching arms of the logwoods. It is prob- ably significant that logwood was not encountered in the older woodland. Although logwood was planted in the past, it seems likely that most of the present growth is spontaneous. The demand for the dyewood declined in the late nineteenth cen- tury, and it probably has not been planted since. Even so, it is still recognized locally as a valuable tree and is spared because of this during clearing and burning. Again, as on the whiteland, several aliens were seen that were not encountered in the systematic sampling. Par- ticularly common are the two fiber plants, sisal (Agave sisalina) and bowstring hemp (Sansevieria sp.). Fruit trees 139 such as the sapodilly (Manilkara zapota), mango (Mangifera indica), tamarind (Tamarindus indica), and genip (Melicoccus bijugatus) were occasionally seen, although most of them had the appearance of being planted. The pawpaw (Carica papaya) was seen, both planted and wild, in cultivated or recently- abandoned fields. Overall, however, aliens have been less successful on the flatland than on the whiteland. The Blackland Five aliens were encountered on the blackland (Table Se The higher number can probably be attributed to the greater area sampled here. Leucaena has similar cover values to those recorded on the flatland and was generally observed to have the same patchy distribution. Logwood was much less important. In part, this may be due to thin sam- pling in the area where logwood is ill-adapted to the more mesic blackland habitat. In its natural habitat in British Honduras, logwood grows on the low, seasonally-flooded areas of the coastal plain (Wilson, A.M., 1936). The sapodilly was the only alien encountered whose importance increased in the older age-classes. Unlike the others it is capable of reaching a size equal to most of the native species. According to Britton and Millspaugh it is "Spontaneous after cultivation" in the Bahamas (1920: 324). However, aS was indicated above, its ability to reproduce independent of man was not clearly established on Cat Island, and, as the old field data indicate, it has made little progress in the woodland. The two doubtful aliens, Indigofera and Heliotropium, were each encountered only twice and therefore account for a very small fraction of the total cover. Furthermore, they were very rarely seen elsewhere in the woodland. Both are restricted to recently-abandoned fields and will be replaced when natural regeneration takes place. All the aliens listed above as seen but not sampled on the whiteland and flatland, were also present on the black- land, as also were Acacia farnesiana and Euphorbia lactea. None, however, were of anything more than local importance or showed any indication of becoming permanently esta- blished. In total, aliens were even less important on the blackland than on the flatland. ee ee In general terms the conclusion reached by Elton and Harris, that success of aliens is determined by the 140 diversity of the native vegetation, is supported by the data presented above. The progressive increase in the importance of aliens from the blackland, to the flatland, to the white- land is inversely proportional to the floristic diversity of the native vegetation. Just why this should be so is not immediately apparent. In part it appears to be determined by the behavior of only two of the aliens, Leucaena and log- wood; the former appears to be well-adapted to the white- land, while the latter is well-adapted to the flatland. Competition from native species is probably not as important as the diversity hypothesis suggests. Leucaena, for exam- ple, is more important on the blackland than the flatland in three of the five age-classes. The significant conclusion is that alien plants have had limited success in the woodland. In spite of repeated clearing, burning, and grazing, very few species have been able to get established. Furthermore, the area they cover is small and their presence promises to be ephemeral. This Situation contrasts sharply with that in the Outer Leewards, where, according to Harris (1965: 59), aliens “exceed natives in total mass of vegetation over extensive areas," or in Barbados, where Watts (1970: 100) has reported that aliens are dominant in certain grassland areas, secondary woodland, and thickets. 141 XII. THE STABILITY OF THE WOODLAND In view of the supposed vulnerability of island life, the most unexpected characteristic of the woodland to emerge from this study is its capacity to recover from disturbance by man. In spite of almost a thousand years of clearing, burning, selective cutting, and, more recently, grazing and browsing, the woodland has survived. This is not to suggest that it has survived unchanged. Sensitive species have become rare, while species pre-adapted to disturbance have increased in importance. In fact, the present stability of the woodland may in large part be the result of man having given a selective advantage to weedy types. Even so, the Surprising fact remains that so many of the native species are inherently weedy. Furthermore, there is no concrete evi- dence that any one species has become extinct. This does not, of course, preclude the possibility of extinctions in the prehistoric period, but it does provide a marked con- trast with the Hawaiian Islands, where literallly dozens of Species are known to have either become extinct or be seri- ously threatened with extinction (Fosberg, 1971: 6). The possible reasons for the unexpected resilience of the Cat Island woodland are the subject of the present chapter. Adaptations to Clearing and Burning In spite of a seasonally-dry climate and thin or non- existent soils, the woodland recovers very quickly after clearing and burning. The woodland is in fact inherently weedy. This weediness can be considered in three different eontexts: seed dispersal, seedling establishment, and regen- eration from sprouts. Seed Dispersal The discontinuous existence of an offshore archipelago such as the Bahamas, or indeed the West Indies as a whole, must have had an important influence on the evolution of plant-dispersal mechanisms. The Bahamas are not’ remote islands such as the Hawaiian Islands or the Galapagos; they anenc Losemmcomeach) other sande tomrLoradavand) the Greater Antilles. Ocean currents reach them after having passed through the whole of the Antilles. They are frequently erossed by hurricanes and other tropical storms. Fur ther- more, many migrating birds spend their winters in the islands, or at least visit them on their way north or south. With all these means of dispersal regularly available, the 142 arrival of a viable seed in the Bahamas as not ~ the rare, chance event it is in Hawaii. A great many of the woodland species are prolific berry-producers, an obvious adaptation to dispersal by birds. The vast majority of woodland species are adapted to dispersal by birds, a small number are wind dispersed, and an even smaller number are adapted to dispersal by ocean currents. Unlike many plant species, which on reaching remote islands have had the efficiency of their dispersal mechanisms reduced (Carlquist, 1965: 241), the species in the Cat Island woodland have retained efficient dispersal mechanisms. The relevance of all this to the present study is that a recently-cleared field is in fact an "island" in the ever- green woodland. The existence nearby of So many species pre-adapted to reach islands accelerates the rate of succes- sion. If the intensity of the "seed rain" on abandoned fields on Cat Island could be measured, it would undoubtedly be high. Seedling Establishment Weediness involves more than prolific seed-production and efficient dispersal mechanisms. The ability to grow in droughty, disturbed habitats is also important. Recently- cleared fields in the woodland are certainly droughty and disturbed habitats. 1In fact, on fiirst sight, dt,isye hamdeeeee imagine that any plant, wild or cultivated, could survive in such a difficult environment (Figure 42, 43). The existence of so many native species that can raises the question as to where they developed this ability. On Cat Island there are at least three possibilities: the savanna/evergreen woodland transition, the mangrove/evergreen woodland transition, and active sand dune situations. The savanna/evergreen woodland transition. Around the upper margins of the seasonally- flooded savanna is a zone that is only flooded after exceptionally heavy rains, perhaps once every four or five years. Here woody growth is possible on a short-term basis only, and the only species able to survive are those capable of quick dispersal and establishment on what is virtually a bare limestone surface (Figure 44). Gundlachia corymbosa and Tabebuia bahamensis, both wind-dispersed species, are common in these areas, as also is Torrubia longifolia, a prolific berry-producer. All three are also important in the early stages of succession on the abandoned fields. Randia mitis, Byrsonima cuneata, and Evolvulus squamosus are also often encountered in this 143 Figure 42. A recently burned blackland field. The area where the man is standing is the lowest part of the field and therefore has the thickest accumulation of ash, probably 10 to 15 centimeters. On higher ground the surface is more than 75 percent bare limestone. Figure 43. A two year old blackland field. The rocky nature of the blackland surface is well shown here. Most of the regeneration is a result of sprouting. As the photo- graph was taken in early June crops are not yet visible. The banana in the background is the only exception. 144 Figure 44. Savanna/woodland transition. This indicates something of the naturally open aspect of the savanna/woodland transition. Herbs on the lower ground give way to palmettoes (Sabal palmetto) and hardwoods on the well-drained sites. The two saplings that have become esta- blished in the foreground are: Torrubia longifolia in the center ad Tabebuia bahamensis to the right. Figure 45. Mangrove/woodland transition. Along the center of this photograph is the normal high tide mark. On the lower ground to the left is Rhizophora mangle. To the right are plants adapted to brackish conditions, Gundlachia corym- bosa in the foreground and Coccoloba uvifera in the back- 145 environment, although of these only Randia was commonly seen in the old fields. The seasonally-flooded savannas cover only a small area of Cat Island at present (Figure 22), but they were probably more extensive in the past, particularly when the Bahama Banks were only slightly above sea level. Looking at the North American subtropics as a whole, it seems likely that the savanna/woodland transitional zone has been an important breeding ground for weedy types. The salt water/evergreen woodland transition. Inland from the salt-tolerant vegetation of the coast and tidal flats, there is a transitional zone in which only the most drought-resistant of the woodland species can grow (Figure 45). Not only are these areas periodically washed with salt Spray, but the fresh water lens is either thin or non- existent. During droughts, the only source of fresh water is condensation in the form of dew. The floristic composition of this transitional zone appears to be largely determined by the nature of the sub- strate. On the loosely-consolidated dunes and beach ridges, several pioneer species are found, such as Lantana involu- erata, Corchorus hirsutus, Croton linearis, Gundlachia eorymbosa,and Salmea petrobioides. On the harder Pleisto- cene surfaces, Small trees are more common. Figure 46 shows the species present on a low Pleistocene beach ridge. In both environments, the salt water/evergreen woodland transi- tion is an important breeding ground for old field pioneers. The naturally droughty nature of the surface is basically Similar to the surface of a recently abandoned-field. Active sand dunes. There are no active sand dunes on Cat Island at present. A few small blow outs were observed on the Holocene dune ridges, but, in each case, in an area disturbed by man. The three dune ridges that comprise such a large part of the island are all fossil features. In _ spite of this, active sand dunes have probably played an important role in the development of weediness in Bahamian vegetation. Dune formation and erosion was a characteristic feature of the Bahamian environment even before the sea-level oscilla- tions of the Pleistocene. Furthermore, in several not very remote areas of the West Indies, there are dune systems still active today; for example, along the northern coast of Puerto Rico, where many of the Cat Island old field pioneers, or their congeners, are present in the sand dunes (Cook and Gleason, 1928). Not all the old field pioneers have been accounted for in this brief survey, but the implication is clear. 146 OBDANN SEWN KEY TO FIGURE 46 Rhizophora mangle LCaguncularia racemosa Conocarpus erecta Erithalis fruticosa Gundlachia corymbosa Rachicallis americana Reynosia septentrionalis Coccoloba diversifolia Metopium toxiferum Casasia clusiafolia Pithecellobium keyense Torrubia longifolia Jaquinia keyensis Coccothrinax argentea Manilkara jaimiqui 147 1S aSpty eucysowty mc] e& ssouoe UCTJTSOdWOD OTIST4UCTY 944 BuUTMOYS WeuseTq “Qh 24NBTY XZ =3 A ‘SH3LIW NI S$31V9S 148 Naturally open, droughty, or unstable habitats have always existed on Cat.. Island, and in similar environments throughout the North American subtropics. Certain plant species have adapted to these habitats, and when man created Similar habitats they simply expanded their range. As has been shown elsewhere, the expansion of naturally weedy types into artificially-disturbed habitats is a widespread phenomenon (Sauer, J.D. 1950; Anderson, 1952). Regeneration from Sprouts Clearing is rarely complete. A great many native wood- land species have the ability to sprout vigorously after cutting and burning. If a field is severely burned, all the pre-existing species may be killed. But this is rarely the case; most farmers are reluctant to burn too deeply because it Gesullts ini} Vower | yields. Consequently, sprout weeding occupies a large part of the farmer's time during the first year of cultivation. After a field is abandoned, any sur- viving sprouts have an obvious advantage over seedlings. Usually they can be identified by the way they project over the general level of the vegetation. The rapid recovery of the woodland after cultivation can in large part be attri- buted to the fact that so many specieS can reproduce from sprouts. Sprouting ability is probably, im part at leasteman adaptation that has developed as a response to breakage by hurricane-strength winds. Craighead (1962:17) has described how many of the tropical hardwoods in southern Florida recovered quickly after defoliation and breakage during hur- ricane Donna. Among the trees that responded most rapidly with new leaves were Lysiloma bahamensis, Bursera simaruba, Coccoloba diversifolia, Swietenia mahagoni,and Zanthoxylum fagara ( syn. Fagara pterota). Significantly, all of these Species are common in the old fields on Cat Island. Lightning fires have also played a role in developing the ability to sprout: In southern Florida, it sinew recognized that lightning fires have had an important influ- ence on the development of the natural vegetation (Robert- son, 2 1962). Furthermore, many of the tropical hardwoods have the ability to reproduce from sprouts. As Craighead has indicated: When fires destroy these hammocks..some roots usually survive. The regeneration from root suckers produces a new stand that is practi- cally identical to “the olldr Repeated fires, however, destroy all of the hardwoods in time (Craighead A974 1570 149 Even though lightning fires have not been common on Cat Island in the recent past, it seems unlikely that this has always been the case. Repeated clearing, burning, and graz- ing have prevented the accumulation of litter, and therefore reduced the chances of fire. In pre-human times, windfall and a deep litter accumulation would have provided ample RUC Om ethtnings fires. “Siteniificantily, fires’ are espe— cially severe in the southern Florida hardwood hammocks dur- ing the dry season after a hurricane. At such a time the accumulation of dead wood on the ground provides an ideal eondition for fire. Lightning fires and hurricanes have therefore both played a role in the development of sprouting ability. This background of natural disturbance has, in effect, preadapted the woodland to withstand many of the types of disturbance introduced by man. Because so many species have the ability to regenerate from sprouts, the impact of shifting agricul- ture has not been as drastic as it might have been other- wise. Resistance to Grazing and Browsing In view of the fact that no grazing or browsing mammals are native to the Bahamas, it was Surprising to find that so many species were unpalatable to goats. The situation on Cat Island is very different to that on Three Kings Island, New Zealand, where goats have reduced a floristically diverse woodland to a woodland composed of only one species (Turbott, 1963) or on St. Helena, where goats brought about ee complete removal of woody species (Darwin, 1839:486- 487). Just why so many species should be resistant to grazing and browsing pressures on Cat Island was not immediately obvious. The unpalatable species are unpalatable because of several reasons: some are poisonous, some are strongly aromatic, and some are armed with sharp spines or stinging hairs. More than likely, these characteristics developed in response to many pressures, of which grazing by mammals is only one. Seasonal drought may have played a role in the develop- ment of essential oils, which assist in water conservation by reducing evaporation rates. Similarly, spiny leaves are also often regarded as an adaptation to droughty conditions. Insect larvae consume a considerable volume of plant material every year, and must have played a significant role in the evolution of the woodland species. Mammals may also have been important. Many of the trees and shrubs now found 150 in the evergreen woodland were formerly present in what is now the south central United States. Presumably they could have developed defensive mechanisms against grazing there. Even today, deer browse in the floristically similar wood- land on the Florida Keys (Craighead, 1971: 97). The whole question is a complicated one and rather than speculate further the point might simply be made that many woodland species are pre-adapted to withstand grazing and browsing by domesticated animals. Resistance to Invasion by Alien Plants In spite of repeated disturbance, very few alien plants have been able to establish themselves in the woodland. There are probably three basic reasons for this: the nature of the woodland as an environment for plant life, the nature of the aliens that have been introduced, and the lack of efficient dispersal mechanisms. Life in the evergreen woodland poses many problems that each alien species must be able to cope with. As has been repeatedly stressed, this low limestone island is in many ways a difficult, environment for plant life. For the native species it is not difficult because they have adapted to it and similar environments over a geologically long period of time. This is not the case with most aliens: Drought, hurricanes, fire, an almost pure limestone surface, and a diverse array of plant pathogens are all natural hazards that any alien species must cope with if it is to become permanently established. Superimposed upon all this is the problem of competition with the native plant species, with their prolific seed production and rapid growth rates. The difficulties facing alien plants in the Bahamas are well illustrated by the repeated failure of attempts to establish commercial crops. Cotton, citrus, tobacco, and sisal have all failed commercially, either because of an insect pest or some other environmental problem (Mooney, 1905). Even with help from man, life in the Bahamas is too difficult for a great many invaders. Perhaps significantly, the only two successful aliens, Leucaena and Haematoxylum, are native to environments not too different from the Baha- mas. On the other hand, the general failure of aliens to invade the woodland must in large part be attributed to the nature of the aliens themselves. Most of them have been cultigens, incapable of spreading far without help from man. The herbaceous weeds that have been introduced are largely ee 151 restricted to open habitats, such as roadsides and gardens. Yet, paradoxically, even though the woodland has’ been repeatedly disturbed by man, very few aliens have been able to get established. This raises the question of dispersal mechanisms. Cattle have never been numerous on Cat Island. This has meant that an important plant-dispersal mechanism has been missing. Cattle have not played the role they have in parts of the West Indies, that of reducing the importance of native species and increasing the cover of aliens. All of this does not necessarily mean that there are no species in existence which, given a chance could become per- manently established in the woodland. For example, there are woodland species that grow on other Bahamian islands but not on Cat which presumably could become established if they were introduced. Looking further afield, it seems more than likely that there is a long list of potentially successful immigrants. Aliens can succeed without help from man, as Casuarina has shown along the coast. The Significance of Diversity Since Darwin's time, the assumption that diversity is Synonymous with stability has been generally accepted. Darwin originally emphasized the importance of diversity because he was impressed by the success of continental Species on oceanic islands. However, as has been shown more recently, the success of continental species on oceanic islands can largely be attributed to prior disturbance by man (Allen, 1936; Egler, 1942; Harris, 1962). In other words, most aliens do not have a competitive advantage because they evolved in floristically and faunistically more diverse environments; their advantage is due to the fact that they have evolved to withstand the types of disturbance associated with man. In many parts of the world man has brought about a marked reduction in diversity. Natural vegetation that was floristically diverse has been replaced by secondary vegeta- tion consisting of relatively few species. Surprisingly, on Cat Island this does not appear to have been the case. There are no significant differences in diversity between the remote, relatively undisturbed areas of the woodland and the areas close to the settlements. In fact, if the woodland is considered as a whole, man has actually increased its floristic diversity by the introduction of alien species. Although the number of aliens that have become established is not large, it probably exceeds the number that have SZ become extinct. The question whether the woodland is more stable now than it was in presettlement times is difficult to answer. Presumably it is more resistant to the types of disturbance introduced by man, but on the other hand it may be less resistant to natural forms of disturbance, such as hurricanes or disease outbreaks. It is probably true to say that the importance of the relationship between diversity and stability has been over- emphasized in recent island studies. Diversity offers a collective advantage in the face of selective pressures such as insect pests, but means little where large-scale pres- sures such as clearing and burning are concerned. The resi- lience of Cat Island woodland is not due to floristic diver- sity, but to the fact that so many species are preadapted to withstand the types of disturbance introduced by man. The Nature of Insularity The anomalous resilience of the woodland can also be attributed to the imprecise meaning of the word island. Some islands are more insular than others. Before the arrival of man, remote islands such as the Galapagos or the Hawaiian islands were colonized by rare, chance migrations. The successful colonists evolved in isolation, and often diverged to form new species or even genera. These endemics are by their very nature vulnerable, and because of their limited numbers are likely candidates for extinction, regardless of their reproductive capacities or ecological tolerances. On the other hand, islands such as Cat or indeed any island in the Bahamas qr West Indies, are not characterized by endemic populations. As far as plants are concerned, a constant interchange of seeds and pollen retards the evolu- tion of new types.~ Most of the species encountered in the Cat Island woodland are widely distributed around the North American subtropics. Each species consists of a large number of individuals, and the chances of extinction are therefore reduced. T. On this point it is interesting to note, that \Tayiese (1921) attributed the low rate of endemism in the Bahamian flora, less than 15 percent, to the geological youthfulness of the islands. As has been emphasized above, this has probably not been an important factor. 2. On several occasions, while at the northernmost point of Cat Island, a constant stream of butterflies, notably cloudless sulphurs (Phoebis spp.) and the Gulf fritillary (Agraulis vanillae), was seen being blown towards Eleuthera Dy the Trade Winds. 153 Insularity and, therefore, vulnerability depend not’ so much on distance, or the past presence or absence of land bridges, but on accessibility in terms of plant and animal dispersal capacities. Cat Island, like most islands, is MuEi LOSS EBCCSSSLIOLS WO ANNUMES Wein Sie Ah tee) jollelatens Its fauna is therefore impoverished, and the extinction of only a few species has a greater overall significance. The imnpieicatlony SOL all this) is) that, the vulnerability of island Like Varies, Mow Omily sero asilemcl to asilemel, join seein Species to species. Vulnerability is a more variable factor than has generally been recognized. Three recent studies in the Lesser Antilles have all emphasized the importance of limited accessibility as a basic reason for vulnerability (Harris, 1965; Watts, 1966; Majer, 1958) 6 While this may be true as far as animals are concerned, it seems unlikely that it is also true for plants. The Lesser Antilles are close to each other and to the continent of South America. Like the Bahamas, they are not inaccessible in the same way as Hawaiian Islands or the Galapagos. On the contrary, their accessibility has prob- ably played an important role in decreasing their vulnera- bility to disturbance by man. The Variable Intensity of Man's Impact As a final point in this general discussion of the sta- bility of the woodland, it should be emphasized that the human population density on Cat Island has always been low. As can be seen from Table 9, Cat is virtually uninhabited when compared to the islands of the Lesser Antilles. Barba- dos, for example, has 529 inhabitants per square kilometer, in contrast to only about 13 per square kilometer on Cat. The significance of the low population pressure is that man's impact on wild vegetation has been less intense on Cat Island than it has been elsewhere. On Barbados, the wood- land was almost completely cleared before the end of the seventeenth century (Watts, 1966: 45), and today some 50 percent of the cultivable land is permanently in sugar cane (Gourou, 1961: 211). The high population densities that are characteristic of so many islands must be considered in any discussion of insular yullaeralotinty, Ms Werrenilil (195s), Ilsererds (905) 5 Watts (1966), and Kimber (1968) have indicated, the native plant and animal life of the Lesser Antilles has been drast- ically disturbed by man. The question arises, then, to what extent the degree of disturbance is due to an inherent vul- nerability on the part of the plants and animals, and to 154 TABLE 9 AREA AND POPULATION OF SMALL WEST INDIAN ISLANDS, 1950-1958 (Gourou, 1961:208-209; *Sharer, 1955:92) Area in Inhabitants sq. km. Population per sq. km. Barbados 431 235,000 529 Grenada 340 91,000 267 Martinique 988 239,000 240 Saint Vincent 340 81,000 238 Saint Thomas 68 14,000 205 Montserrat 80 14,000 Lae Saint Christophe ila he 30,000 169 Anguilla 85 14,000 164 Guadeloupe i StS) 229,000 150 Sainte Lucie 620 92,000 148 Tortula 30 7,000 140 Antigua 440 57,000 130 Saint Barthelemy 21 2 S4 120 Saint Martin 52 SEBS vf hi f 103 Marie Galante 150 15, toe 101 Nevis 140 11,300 88 Dominica 790 65,000 80 Sainte Croix 205 14,000 68 Desirade (ead | 1,654 46 CAT ISLAND* 240 3,000 13 155 what extent to the density of human population. In a sense it is misleading to compare the extent of change on a small, densely-populated island with that on comparatively uninha- bited continental areas. Islands are vulnerable in the sense that they may become densely populated. This does not necessarily mean that island life is inherently more vulner- able to disturbance than that on the continents. The vari- able intensity of man's impact complicates any comparative analysis of insular vulnerability. In conelusion, the stability of the woodland is the result of many factors. Basically, it reflects a long period of adaptation to natural disturbances: hurricanes, fires, sea-level change, erosion, and deposition. Because the woodland species are adapted to these various types of disturbance, man's impact has been less important than it would have been otherwise. ‘clucaes teaud I6 .sttnen ond > rr ietve off sxsqios oF Brae sjmos © Sony fate Oriel wee ron 321 £493 VE ‘ yet ronhdébiamessd vem Vane jan rz tatel at neon ibn Lad! os PSEA NOseEeR XIII. REVIEW AND CONCLUSIONS The main purpose of this study was to determine the extent to which man has modified the vegetationof a small island in the Bahamas, Cat Island. In a broader sense, the question was considered whether or not the vegetation of Cat Island was vulnerable to culturally-induced disturbance in the same way as the vegetation of remote islands, such as the Hawaian Islands or the Galapagos. The nature of man's impact on vegetation of Cat Island was determined by the use of both historical and ecological evidence. Particular attention was given to the mixed evergreen-deciduous woodland that covers most of the island. The historical record for the Bahamas in general, and LOCA lS hande sine par ticular, ls snousitimatingiay thine. Even so, the evidence recovered clearly shows that the Bahamas have not escaped the processes of change that have affected nearly all tropical islands during the period of human _ set- tlement. In the comparatively short period of a thousand years, the Cat Island woodland has been drastically dis- turbed by clearing, burning, selective cutting, grazing, and browsing. Unfortunately, the historical record is painfully qualitative and gives no clear indication as to what the eonsequences of this disturbance have been. In order to remedy this deficiency, a detailed analysis was made of the present woodland. As no previous work of this kind had been attempted for this type of vegetation, it was necessary to develop what was basically a new methodology. With the aid of aerial photograph coverage for the years 1943 and 1958, 300 sam- pling sites were established, and at each site the percen- tage cover in 25 x 1 square meter quadrats was determined. These data were then analysed to determine the extent to which the floristic compostion of the woodland varied with respect to age and intensity of disturbance. The most important conelusion to be drawn from _ the analysis was that the Cat Island woodland is remarkably well-adapted to withstand the types of disturbance intro- duced by man. This resilience is shown in three basic ways. First, a great number of native trees and shrubs are capable of quickly colonizing disturbed habitats as abandoned fields. Second, there is no evidence to suggest that any particular species has become extinct. And third, with few exceptions, alien plants have not been able to establish themselves in the woodland. 158 On the other hand, the woodland has not persisted unchanged since pre-settlement times. As a result of clear- ing, burning, selective cutting, grazing, and browsing, sen- Sitive species have become rare, and have survived as impor- tant members of the woodland only in remote, relatively undisturbed areas. Conversely, weedy types have increased in importance; formerly restricted to naturally-disturbed or droughty habitats, they have expanded into the areas dis- turbed by man. This sequence of events is not a new one. The same changes have occurred wherever vegetation has been disturbed by man. What is unusual about the evergreen wood- land is that such a large proportion of the species present are inherently weedy. All of this does not mean that the hypothesis of insu- lar vulnerability should be rejected. The high rate of extinctions on the Hawaiian islands and the Galapagos is indisputable evidence of vulnerability. The important point is that islands vary in their vulnerability as also do dif- ferent species on any particular island. The differences between the conclusions reached in this study and those reached in similar studies in the Lesser Antilles can probably be attributed to several factors. First, Cat Island has differed from many tropical islands in its persistently low population density. This has meant that man's impact on the vegetation has not been as severe as it has been elsewhere. Second, the Bahamas, because of their off-shore location, seasonal drought, and background of natural disturbance, are characterized by vegetation which is pre-adapted to withstand the kinds of disturbance introduced by man. And third, in view of the inadequacy of the historical record and the absence of any previous analysis of this type, it was necessary to develop a basi- cally new methodology. This empirical approach prevented the easy acceptance of any theoretical interpretations. Although small and insignificant in some respects, Cat Island has provided a useful setting for a study of man and vegetation change. In spite of a thousand years of discon- tinuous settlement, man's impact on the woodland has been Surprisingly slight. What changes will occur in the future will depend largely on social and economic conditions in the colony. A decrease in the intensity of shifting agriculture will allow the woodland to continue its overall recovery. An increase will cause a reversion to conditions similar to those that existed in the late nineteenth century. Either way, the data gathered here will provide a useful base-line against which future change can be assessed. 159 ACKNOWLEDGMENTS This study owes much to the help and encouragement of others. I am particularly indebted to J. Gordon Nelson, who first introduced me to the "Man and Vegetation Change" theme in the Rocky Mountains of western Canada. At the University of Wisconsin I was given the chance to explore the theme further, “and™= Geeeived much help and ‘sound advice from Jonathan Sauer, William Denevan, the late Andrew Clark, and Hugh Tltis. F.R. Fosberg of the Smithsonian Institution also gave valuable advice in the early stages of the project. On Cat Island, the inhabitants of Wilson Bay and Arthurstown contributed a great deal to the success of the field work. Particular thanks are due to Japhet King, Con- Stable Samuel Seymour, Ned Isaacs, and my constant companion in the bush, Isaac Dean Junior. Thanks are also due to the late Biddle Page and his wife Valerie for their generous hospitality. Bryce Scott also kindly provided accommodation on the island. lid WaSSAU, OPIS RUSSELL, wae WhiintsBeie Or Agriculture, was helpful in many ways. Thanks are also due to Gerald Isaacs, the Chief of Customs, for his help in obtaining authorization to import fieldwork equipment. It is a pleasure to acknowledge that the cost of field work was partly alleviated by a Research Assistantship and a Knapp Travelling Fellowship from the University of Wisconsin. I completed the dissertation (Byrne, 1972) while at the University of Toronto, where I was a Lecturer in Geography at Erindale College. There I received valuable assistance in computer programming from Seigfried Schulte. Thanks are also due to Norman White, who drew all the maps and diagrams, and to Steve Jaunzems, who helped with many aspects of the pho- tography. The present text is a revised version of the Ph.D. dissertation, and for comments on style and content I would like to thank my father Henry Byrne, Jonathan Sauer, and F.R. Fosberg. The camera-ready copy of the manuscript was typed on the Unix system at the University of California, Berkeley Computer Center by Lucille Oberlander and Dianne Byrne _ and for their efforts I am especially grateful. 160 BIBLIOGRAPHY Ahmad, N., and R.L. Jones 1969. "Occurrence of Aluminous Lateritic Soil (Bauxites) in the Bahamas and Cayman Island." Economic Geol- ogy 64: 804-808. Albemarle, Earl of, et “al. 1676. Instruction to Mr. Charles Chillingworth, Governor of New Providence and the Rest of the Bahama Island, July 13, 1676. Egerton Ms. 2395, “fom 558-9, British Museum, London. Alexander, T.R. 1958. "High peu os Vegetation of the Southern Florida Mainland. The Quarterly Journal of the Florida Academy ee Sciences 21: 293-298. Alirany, Hi. Hi: 1936. “"“Indigene verses Alien in the New Zealand Plant World." Ecology 17: 187-193. Anderson, A. 1802. "An Account of the Exotic Plants sent from the Botanical Garden in St. Vincent to the Bahama Islands." In Communications on different Subjects Addressed to the Bahama Agricultural Society, pp. 45-59, Joseph Eve, Nassau, New Providence. Anderson, E. 1952. Plants, Man and Life. Little, Brown and Co., Bos- ton. Anonymous 1840. Statistical Account of the Bahama Islands, Ms., C/WI Box 3, United Society for the Propagation of the Gospel, London. 161 Anonymous 1960. "General Information of Farming Conditions in the Bahamas." Mimeographed. Department of Agricul- ture, Nassau, New Providence. Bacot, J.T.W. 1869. The Bahamas: A Sketch. Longmans Green, Belfast. Baker, H.G. and G. Ledyard Stebbins the First 2 ee Union of Biological Sich ences Symposia on General Biology. Academic Press, New York. (Goren GlenletneS mon Colonmzane Speeches: srroceedimngs On Beard, J.S. 1955. "The Classification of Tropical American Vegetation-types." Ecology 36:89-100. Berry, R.W. 1930. Revision of the Lower Eocene Wilcox Flora of the United States. U.S. Geological Survey Profes- Sional Paper 156:1-144. Britton, N.L. 1907. WRepors Of whe CoMmlmiectom Ot was isoweldteeil Exploration of the Bahama Islands." Journal of the New York Botanical Garden 8: 71-81. BrPleBeOm, Welbo, Bac Cole o IMs Shoe wyesle 1920. The Bahama Flora. Reprint (1962), Hafner, New York. Broecker, W.S., and D.L. Thurber 1965. “Uranium = Series Dating of Coralls and Oolites from Bahaman and Florida Key Limestones." Science 149:58-60. Brown, T. 1802, Lewiber t© Me. bell of Caicos, Gauec Sul Ov Wwele@ld, 1801. In Communications on Different Subjects 3-27, Joseph Eve, Nassau, New Providence. 162 Browne, M. 1775. Letter to the Earl of Dartmouth dated May 6, 1775. Colonial Office Ms. _ (© .0% 23723)72) Pubueie sRecormds Office, London. Bebe, IP olels 1782. Memoirs of Peter Henry Bruce Esq. A Military Off-= icer in the Services of Prussia, Russia, and Great Britain. London. Byrne, A.R. 1972 Man and the Variable Vulnerabiliy of Island Life: A study of Recent Vegetation Change in the Baha- mas. Unpublished Ph.D. dissertation; University of Wisconsin, Madison, Wisconsin. Byrne, A.R. and J.C. Munday, Jr. 1972 "Photodensity and the Impact of Shifting Agricul- ture on Sub-tropical Vegetation; A Case Study in the Bahamas." pp. 1311-1326. In Proceedings (on the 8th International Symposium on Remote Sensi of the Environment, October 2-6, 1972, be Sensing of Michigan, Ann Arbor. Cameron, C. 1805. List Containing the Number of Negro Slaves, Free Negroes and Colored People in the Bahamas. Colo- nial Office Ms., C.0. 23/48; 1, Public. Recamae Office, London. CanlqueaLrsitenssr. 1965. Island Life. Natural History Press, Garden City, New York. | 1970. Hawaii: A Natural History. Natural History Press, Garden City, New York. Catesby, M. 1731. Natural History of Carolina, Florida and the Baha- mas “Islands. 2 vols. London. Clark, A.H. 1949. The Invasion of New Zealand by People, Plants, and Animals: The South Tsland. Rutgers University Press, New Brunswick, New Jersey. Clench, W.J. 1938. "Origin of the Land and Freshwater Mollusk Fauna of the Bahamas, with a List of the Species Occur- ring on Cat and Little San Salvador Island." Bul- tetin of the Harvard Museum of Comparative Zoology 80: 481-547. Coker, W.L. 1905. "Vegetation of the Bahama Islands." In The Bahama Islands, G.B. Shattuck (ed.),pp.275-289, Geograph- ical Society of Baltimore, Macmillan, New York. Compston, H.F.B. 1918. Thomas Coram, Churchman, Empire Builder and Phi- lanthropist. Society for the Promotion of Chris- fian Knowledge, London. Cook, M.T. and H.A. Gleason 19285 VECOIOBICA SURVEY Oi wie lilore Oi IRoieiBe Trat@es¥ Porto Rico Department of Agriculture Journal 12:1-139. Craighead, F.L. 1962. "The Effects of Hurricane Donna on the Vegetation of Southern Florida." The Quarterly Journal of the Florida Academy of Science 25: 1-28. 1964. “Land, Mangroves, and Hurricanes." Fairchild Trop- 1eal Carden SwUllewiia 195 S320 1971 The Trees of South Florida, vol. 1. The Natural Environments “and ~ their Succession. University of Miami Press, Coral WGables. Craton, M. 1968. A History of the Bahamas. 2nd ed., Collins, Lon- Ona] Saae Sey Tee wee 164 Dairieciil, dio 1670. Letter to Lord Ashley dated March 15, 1670. Edg- erton Ms. 2395, fols., 474-478, British Museum, London. Darwin, ir 1839. The Voyage of the Beagle. Natural History Library Garden City, New York, 1962. 1859 The Origin of Species. Mentor edition, New Ameri- can Library, New York, 1958. Deans Peggs, A. 1957. "A Relic of Slavery." Mimeographed. Nassau, New Providence. DYOezlls Wop dheor 1955. Landforms of the Southeast Bahamas. University of Texas Publication Number 5509, Austin. Dunmore, Lord 1789. Report on the State of the Bahama Islands. Colo- nial Office Ms., C.0. 23/29; 25, Pubbic Regormans Office, London. Dyer, Ie 1887. Letter dated 15th May 1887. In Bahama Fibres 1854-1900. Miscellaneous Reports vol. 232, The Library, Royal Botanical Gardens, Kew. Edwards, B. 1819. The History, Civil and Commercial of the British West Indies, 5th ed., 5 vols., London. Heeieriry ak cE. 1942. “"Indigene versus Alien in the Development of Arid Hawaiian Vegetation." Ecology 23: 14-23. 1952. "Southeast Saline Everglades Vegetation, Florida, and its Management." WVegetatio Acta Botanica 3: 213-265. 165 Elton, CoS. 1958. The Ecology of Invasions by Animals and Plants. Methuen, London. Fives. nc sain ee oe ic lie Eve, Jd. 1800. Reply to a Questionnaire on the State of Planta- tion Agriculture in the Bahamas. Colonial Office MSo Gos 23/293 i, IPblolte Kacoeds Oprah iterate Eve, O. 1784. Letter dated 29th May, 1784. Colonial Office Ms. C.O. 23/26, Public Records Office, London. Fitzwilliam, R. 1737. Report on the State of the Bahama Islands. Colo- Mie Orestee MScy GoOs 23/3 BO, PUlolie iRnecords Office, London. Fosberg, F.R. 1963. "Disturbance in Island Ecosystems." Ii) PAC a Basin Gogeography: A Symposium, J. Linsley Gres- Siwe (eds) 5 SS. S572501. Bishop Museum Press, Hawaii. 1965 "The Island Ecosystem." In Man's Place in the Island Ecosystem: A Symposium, F.R. Fosberg (ed.), pp. 1-6. Bishop Museum Press, Hawaii. 1971. “Endangered Island Plants." The a ae of the Pacific Tropical Botanical Garden (Bye ots 1972. "Man's Effects on Island Ecosystems." In The Careless Technology: Ecology and International Development ots Baryvar Amel Joe. Wilton (cSs)5 pp. 869- B80. Doubleday, The Natural History Press, Garden City, New York. Gardner, J., and L.J.K. Brace 1889. "Provisional List of the Plants of the Bahama Islands. Arranged with Notes and Additions by Charles S. Dolley." Proceedings of the Academy of Naibualasclence, ehiskadelphaag | 3349 —a 0176 166 Gidiis. We 1975 Gournoun sr. 1965. Hamilton, 1836. HaeiceaS'.. Dy. 1962. 1965. Harvey, T. 1858. Higgs, J. 1855. Hitchcock, 1893. Hoffman, 1967. Toy ee yal. Bibliography of the Natural History of the Bahama Islands. Atoll Research Bulletin No. 191 "Pressure on Island Environment." In Man's Place in the Island Ecosystem: A Symposium, F.R. Fosberg Ced.), pp. 207-225. Bishop Museum Press, Hawaii. W. "On the Cultivation of the Divi-Divi Tree." Jour- nal of the Bahama Society for the Diffusion of Knowledge 6: 53-59. R. "The Invasion of Oceanic Islands by Alien Plants." Transactions and Papers, Institute of British Geo- graphers 31: 67-82. — Plants, Animals and Man in the Outer Leeward Islands. University of California Publications in Geography, vol. 18, Berkeley. C. Official Reports of the Out Islands of the Baha- mas. “TT. Darling, Nassau. auale Missionary Report for the Year Ending March 31st, 1855. Ms E/PRE, vol. E, fols. 161-168,.Umives Society for the Propagation of the Gospel, London. An Sie "List of Plants Collected in the Bahamas, Jamaica and Grand Cayman." Annual Report of the Missouri Botanical Garden 4: 47-179. 2 /ANs "Bahama Prehistory: Cultural Adaptation to an Island Environment." Unpublished Ph.D. disserta- tion, University of Arizona. Holdgate, 1961. Hooker, J. 1860. Hooker, J. 1865. 1867a. 1867b. Howard, R. 1950. Johnson, 1783. Johnston, 1867. Kelsall, 1800. 167 M.W. and N.M. Wace "The Influence of Man on the Floras and Faunas of Southern Islands." Polar Record 10: 473-493. D. The Botany of the Antarctic Voyage. eves — wibae Flora Tasmaniae, vol. 7 Dicotyledons. Lovell Reeve, London. Reece D. "Note on the Replacement of Species in the Colonies and Elsewhere." The Natural History Review N.S. 5:123-127. a eae? oe "On the Struggle for Existence Amongst Plants." The Popular Science Review 6: 131-139. "Insular Floras." Journal of Botany 5: 23-31. 1% "Vegetation of the Bimini MIsland Group B.W.1I." Ecological Monographs 20: 317-349. Lewis Lecwer Glatcee July, ~ 1703 8)a Colonial Office Ms., C.0. 23/26, Pulolie Records Orrlee, iLomclom. F.W. Report on the Agricultural Capabilities of the Bahamas. Colomial Orriee Mss, CGoOs 23/1898 171, Public Records Office, London. Shoreline Features and Quaternary Shoreline Changes Puerto Rico. “U.S. Geological Survey Pro- fessional Paper 317B: 49-140. J. Reply to a Questionnaire on the State of Planta- tion Agriculture in the Bahamas. Colonial Office Ms., €.0. 23/39; 17, Public Records Office, Lon- don. 168 Kimber, 1969. Lefroy, C. J. ine "Recent Historical Plant Geography of Martinique." Unpublished Ph.D. dissertation, University of Wisconsin. H. 1877-79.Memorials of Bermuda (1515-1685). 2 vols., “Lone Lind, A. 1969. Liteie: 1964. Loucks, 1970. Or ES QO. Loveless, UBT « Eymnitsi,, G-. 1970. MacArthu 1967. Las) don. Coastal Landforms of Cat Island Bahamas: A Study of Holocene Accretionary Topography and Sea Level Change. University of Chicago, Department of Geography Research Paper, No. 109. Ls, and HoH. Wadsworth Common Trees of Puerto Rico and the Virgin Islands. Agricultural Handbook No. 249, United States Department of Agriculture Forest Service, Washington,D.C. Iss "Evolution of Diversity, Efficiency, and Community Stability." American Zoologist 10: 17-25. ACR. and) Gob. Asiprey "The Dry Evergreen Formations of Jamaica, I. The Limestone Hills of the South Coast." Journal of Ecology 45: 799-822. W. "Conceptual Model of the Bahamian Platform for the Last 135 Million Years." Nature 225: 1226-1228. Roni. sand E.On Wasson The Theory of Island Biogeography. Monographs in Population Biology No. 1. Princeton University Press, Princeton, New Jersey. MacLaury, 1970. Mathew, G. 1844. Mayr, E. 1963. 1965. MecKinnen, 1804. Merrill, 1958. Mooney, C. 1905. Morris, D. 1896a. 1896b. 169 doGo Archaeological Investigations on Cat Island Baha- mas. State Museum Contributions in the Social Sciences Number 16, University of Florida, Gaines- ville. B. Letter dated August 7, 1844. Colonial Office Ms.C.0. 237118; 8, Public Records Office, London. Animal Species and Evolution. Belknap Press of Harvard University Press. “Cambridge. "Summary." In The Genetics of Colonizing Species, H.G. Baker and G. Ledyard Stebbins (eds.), pp. 553-562. Academic Press, New York. ID) A Tour Through the West Indies in the Years 1802 and 1803, Gi Giving Particular Account of the Bahamas Islands. London. a ae GriGr. "The Historical Record of Man as an Ecological Dominant in the Lesser Antilles." The Canadian Geographer 2:17-22. "Soils of the Bahama Islands." In The Bahama Islands, G.B. Shattuck (ed.), pp. 2147-181, Geo- graphical Society of Baltimore, Macmillan, New York. "Memo on the Cascarilla Bark." In Bahama oiere 1862-1900, Miscellaneous’ Reports, VOl>s 1085 WO-51. The Library, Royal Botanical Garden, ve, Bahamas - The Sisal Industry. Colonial Office Miscellaneous Report No. 5, H.M.S.O., London. 170 Morton, D., and J. Morton 1946. Fifty Tropical Fruits of Nassau. Text. House; Coral Gables. Newell, N.D. 1965. "Warm Interstadial Interval in the Wisconsin Stage of the Pleistocene." Science 148: 1488. Newell, N.D., and J.K. Rigby 1957. "Geological Studies on the Great Bahama Bank." In Regional Aspects of Carbonate Deposition, Society of Economic Paleontology and Minerology Special Publication No. 5. "pp. lib =O Odum, E.P. 1971. Fundamentals of Ecology, 3rd ed., W.B. Saunders, Philadelphia, Pa. Paxton, J. 1972-3.The Statesman's Yearbook. Macmillan, London. Phenney, G. 1723. Report on the State of the Bahama Islands. Colo- nial Office Ms. C€.0.. 23/1; 49. Public Records Office, London. 1724. Report on the State of the Bahama Islands. Colo- nial Office Ms. C.0. 23/71; 55. ~Pubilie) Recomum office, London. Powles,>L. Ds 1888. The Land of the Pink Pearl, or Recollections of Life in the Bahamas. Sampson Low, London. Pownall, J. 1783. A Plan for the Establishment of Botanical Garden in the Bahamas. Colonial Office Ms. C.0. 2373258. Public Records Office, London. 171 Rawson, R.W. 1866. oe ee Report for the Bahamas (1864). Parliamen- ary Papers, vol. 44 (Accounts and Papers vol. 8) Go SID, D0 IHSGly a Pipes Robertson, W.B. 1962. "Fire and Vegetation in the Everglades." In Proceedings, First Annual Tall Timbers Fire Ecol- ogy Conference, pp. 671-80. Tall Timbers Research Station, Tallahassee. Robinson, S. 1670. An Account of the Bahama Islands. Egerton Ms. 2395), %OlS> HA, So Bieaiasla MUSSUIN, iLoraelorals Rogers, W. 1730. Report on the State of the Bahama Islands. Colo- nial Offace Ms.) C0. 2372. Public Records Office, London. Roumain, J. iJ2ee ContErbuiron | ja) lEtude | de |) luEthnobotanique Precolomienne des Grandes Antilles. Bulletin du Bureau d'Ethology de la Republique d'Haiti, No.1. Rouse, I. 1964> “Prehistory of the West Indies." Science 144: 499-513. Ruhe, R.V.; J.G. Cady; and R.Z. Gomez 1961. "Paleosols of Bermuda." Bulletin of the Geological Society of America 72: 1T27eT TE. Sauer, C.O. 1950. "Cultivated Plants of South and Central America." In Handbook of South American Indians, Bureau of American Ethnology, Bulletin 143, vol. 6: 487-543. 1966. The Early Spanish Main. University of California Press, Berkeley and Los Angeles. 172 SEV, olDc 1950. "A Geography of Pokeweed." Annals of the Missouri Botanical Garden 39: 113-125. 1960. Coastal Plant Geography of Mauritius. Technical Report No. 15, Part A. Coastal Studies Series, Louisiana State University, Baton Rouge. 1962. "Effects of Recent Tropical Cyclones on the Coa- Stal Vegetation of Mauritius." Journal of Ecology 50: 275-290. 1967. Plants and Man on the Seychelles Coast: A Study in Historical Biogeography. University of Wisconsin Press, Madison. Schoe pity adie)! 1788. Travels in the Confederation. Reprinted in 1903 as Bulletin of the Lloyd Library, N. 6 (Reproduc- tion Series, no. 3), Philadelphia. shaiger .Ci. 1955. "Population Growth in the Bahamas." Unpublished Ph.D. dissertation. University of Michigan. Simpson, G 1953. The Major Features of Evolution. Columbia Univer- sity Press, New York. Stark, J. He 1891. History and Guide to the Bahama Islads. Boston. Stiles, Ji 1836. Report on Conditions to St. Salvador, Watlings. Rum Key Exuma, Long Island, Crooked Island. Colo- nials Offaice Ms.., _ C. O29 523)/9ie es Elblatc Records Office, London. Stoddart, D. 1965. "“Re-survey of Hurricane Effects on the British Honduras Reefs and Cays." Nature 207: 589-592. WAS) Sturtevant, W.C. 1961. “Taino Agriculture." In The Evolution of Horti- cultural Systems pe Native South America, Causes and Co pees Ueieee! FES A) So sso 0s ns 8 OY 5 on a Xe IPD iP pp. 69-82 Sociedad de Ciencias Naturales, La Salle, Caracas. TAILOR 5 lo lol\s 1881. Annual Report for the Bahamas (1880). Parliamen- tary Papers, Vol. 64 (Accounts and Papers vol. 8), cemscOo4eupp. J9=625 Taylor, N. 1921. “"Endemism in the Bahama Flora." Annals of Botany 358 D235 32 Sesses 25 Sees Turbott, E.G. 1963. "Three Kings Islands, New Zealand: A Study in Modification and Regeneration." In Pacific Basin Biogeography: A Symposium, J. Linsley Gressitt Ted.), pp. 485-498. Bishop Museum Press Hawaii. Under hawllite EB) 1862. The West Indies, their Social and Religious Condi- tion. Jackson, Walford and Hoddes, London. Wadsworth, F.H. and G.H. Englerth 1959. WEPFECCS OF wae 1956 Hurricane on Forests in Puerto Rico." Caribbean Forester 20: 38-51. Wallace, A.R. 1902. Island biG Brel PEW ed., Macmillan, London. Watts, D 1966. Man's Influence on the Vegetation of Barbados 1627 to 1800. —~Gecasional Papers in Geography No. 4, University of Hull. 1970. “Persistence and change in the Vegetation of Oce- anie Islands: An Example from Barbados, West Indies." Canadian Geographer 14: 91-109. 174 Wilson, 1936. Wilson, 1965.. Wilson, 1783. A. E. J. Wylly, W. 1789. 1800. M. "The Logwood Trade in the Seventeenth and Eighteenth Centuries." In Essays in the History of Modern Europe, D.C. McKay (Ced.), pp. 1-15. New York. OF "The Challenge from Related Species." In The Genetics of Colonizing Species, H.G. Baker and G. Ledyard Stebbins (eds.), pp. 7-24. Academic Press, New York. Report on the State of the Bahama Islands. Ms. Copy in the Hunt Collection, Boston Public JPalloeeve(, ff Jeyebc A short Account of the Bahama Islands. Additional Ms. 6058, fol. 51 British Museum, Lovdon. Letter dated 27th July, 1800. Additional Ms. 22901, fols. 84-89 British Museum, London. 175 APPENDIX I. A systematic list of the plant species encountered on Cat Island. The numbers without parentheses following the name are the collection numbers of voucher specimens deposited in the herbaria of the Arnold Arboretum, Harvard University, and the University of Wisconsin. Species without numbers are sight records. An asterisk indicates that the species was encountered in the old field study. The number in parentheses following these species represents the code number used to identify them during computer analyses. ACANTHACEAE % Anthacanthus spinosus (Jacq.) ace B34, CVS) o Dicliptera a assurgens (L.) Juss., 381. RieisetameUbenOsamlen 238), 553. AIZOACEAE Sesuvium portulacastrum L., 199. ALISMATACEAE Echinodorus berteroi (Spreng.) Fawe., 568. AMARANTHACEAE Achyranthes aspera (L.) Mill., 173. Alternanthera paronychioides St. Hil., 328. Amaranthus spinosus L., 547. Centrostachys indica (ie: ) Standley, 173. Gomphrena a globosa_ (eras Tresine diffusa H. & “Be ex Willd., 580. Philoxerus vermicularis (L.) P. Beauv., 456. AMARYLLIDACEAE Agave americana L., "Bamboo". % A. sisalana Perrine, "Sisal", (104). Hymenocallis declinata (Jacq.) M.J. Roem., 92. ANACARDIACEAE Mangifera indica L., "Mango". % Metopium toxiferum (L.) Krug. & Urban., "Poisonwood," 38, 47 (a9). Spondias mombin L., "Hog plum". 176 ANNONACEAE Annona glabra L., "Pond apple." Ay mumuecatask., NSoursop." A. squamosa Eo. 5 = tous alg sapp lent aie APOCYNACEAE snogadenis berterii: (A. ,,DC) sMierstyn SEiees BOO 2a RTOs Serie umbellata Jacq., 14. Catharanthus rosea (L.) G. Don., "Periwinkle." Nerium oleander L. * Plumiera obtusa L., "Milk bush," 249 (115). * Vallesia antillana Woods., "Man root", 126 (39). ASCLEPIADACEAE Asclepias curassavica L., 398. Calotropis procera UALt.) Ait. fo. 506 Cryptostegia grandiflora Re. Br. 469s Cynanchum eggersii (Schlt.) Alain, 435. C. northropiae (Schlecht.) Alain, 383. ie. spe. 590). BATIDACEAE Batis maritima L. BIGNONIACEAE Crescentia cujete L., 350. * Jacaranda coerulea (L.), Griseb., "Clock bush" 252 C22) ze * Neobracea bahamensis Britton, 422, 520 (110). * Tabebuia bahamensis (Northrop) Britton, "Five finger," 302, 479, 485 (21). BOMBACACEAE Ceiba pentandra (L.) Gaertn. BORAGINACEAE * Bourreria ovata Miers., "Strong bark," 36, 54, 80, sama (CT 6ey * Cordia bahamensis Urban, "Black granny bush", 213, 415, NG C11). c. a (Mill sp...). Macbria, 340. lucayana (Millsp.) Macbr., 455. C. sebestena L. 177 ‘2 Heliotropium angiospermum Murr. "Scorpion tail", 1, io LoCo Dae Hes curassavicum bog Bes. Siste H. inaguense Britton, Bre Mallatonia gnaphalodes L., "Bay lavender", 20. Tournefortia volubilis Lake "Soldier vine", BROMELIACEAE Ananas comosus, "Pineapple". BURSERACEAE * Bursera inaguensis Britton, "Beau Kamalamay", 251 % B. simaruba Sarg., "Kamalamay", 52, 298 (36). BUXACEAE Buxus bahamensis Baker, "Boxwood," 445, 531. CACTACEAE * Opuntia dillenia, (Ker-Gawl), Haw.5 “Prickly (94). CANELLACEAE Canella winterana (L.) Gaertn., 161. CARICACEAE Carica papaya L. CASUARINACEAE % Casuarina equisetifolia Forst., 91, (83). CELASTRACEAE % Crossopetalum rhacoma Crantz., 212, 439, (23). * MEECWUS AURA OItA TNs MiGs), Celse>o5 (Oy 22s CHENOPODIACEAE Atriplex pentandra (Jacq.) Steud., 110. Chenopodium murale L., 510, 545. Salicornia perennis Mill., 108. COMBRETACEAE Crore pear" (41). 178 * Conocarpus erecta L., "Buttonwood", 22)..299.5 400.50 90e Laguncularia racemosa GL.) Gaertn., 89. Terminalia catappa L., "Indian almond." COMMELINACEAE Rhoeo spathacea (Desv.) Stearn COMPOSITAE Ageratum conyzoides L. spp. latifolia (Cav.) Johnson, Ose7, oo A. muticum Girseb., 112. Ambrosia hispida Pursh., 104. A. paniculata Michx., 496. Aster bahamensi Britton., 117, 577. Bidens pilosa L., 204. Borrichia arborescens Chi.) BDGeeee ice Chaptalia dentata (L.) Cass. 574. bclaptarprostravan( Lam list sviey. Ennbvaysonchitolia-(i.)- DG... 509. Erigeron canadensis L., 197, 563. Eupatorium bahamense Northrop, 407. E. capillifolium (Lam.) Small, 589. lucayanum Britton, 70. villosum ow.; "Bitter sage," 335, C2). Flaveria Ti inearis Jacq., 500. F. trinervia (Spreng.) fons 440, Gaillardia pulchella Foug., 315. Gochnatia bahamensis (Urb.) Howard & Dunbar, 256, 276. Coal catolwvas esis ey ssii2e * Gundlachia corymbosa (Urb.) Brite, "Horsebush," SCteG>) a aes Iva imbricata Walt., 77. Lactuca intybacea Jacq., 484. Melanthera deltoidea Michx., 116. Parthenium hysterophorus L., 172. Feetis finifoiiarun, 450. Pluchea purpurascens (Sw.) DC., 4, 333, 337. P. rosea Godfrey, 526. Porophyllum ruderale (Jacq.) Cass., 557. * Salmea petrobioides | eraeee ne HOG. oops Sonchus oleraceus L., 266, 540, 560. Tridax procumbens L., 239. Verbesina encelicides (Cav.) Benth. & Hook., 316. Vernonia - bahamensis Griseb., 419. Vercinerea (L.) Lesses 5892 ¥ Wedelia bahamensis (Britton) 0O.E. Schulz, 102, 310, (106). W. Trilobata (L.) Hitch., 508. E E. 179 CONVULVULACEAE Evolvulus alSinoides L., 405. Bo SERUCGUS Sits § U2Z05- E. SCUAMOSUS eiweon, Weoom fOMeiay, 117, 290, (My. Ipomoea acuminata (Vahl) Ro@Ssg 30%. i, bacawas (55) Lewe, BOM, Moo, Mee. I. microdactyla Griseb. N65. it te J BieiLOoa bo, Ol, IAG. macrantha R. & So 352, acquemontia cayensis Brit., 436. jamaicensis (Jacq.) MALL Bop 2ilo 377 Merremia dissecta (Jacq.) Hall., 549. CRASSULACEAE Bryophyllum pinnatum (Lam.) Kurz., 185. CRUCIFERAE Cakile lanceolata (Willd.) D.E. Sehulz, 105. Lepidum virginicum L., 158. CYPERACEAE Abilgaardia monostachya (L.) Vahl, 181. Cyperus elegans L., 366. C.filiformis Sw., 511. Go fulgineus Chap. 347. Co Wipuilaris tbo ¢ 363. C. odoratus L., 565. planifolius’ lboGo Rieti. , S51. HACHAOTSTE Golorata CL.) A GChe - 12. Eleocharis cellulosa Torr., 119. "576. E. geniculata (L.) R.&S. 389, 512. Fimbristylis cymosa R. Br., 460. F. dichotoma (L.) Vahl, 516, 523. F. ferruginea (L.) Vahl, 63. Mariscus jamaicensis (Crantz) Britton, 90. Bhynehosper a cyperoides (Sw.) Mart., 578. 5 Steililewe OleiMs) Geiscios, I2ic a lithosperma (L.) Sw. 152. NAC DIOSCOREACEAE Dioscorea SIP: EBENACEAE % Diospyros caribaea (A.DC) Standley, 27S, MNG625 (290 180 ERYTHROXYLACEAE * Erythroxylon areolatum L., 487, (18). x E. rotundifolium Lunan, 255, 444, (17). EUPHORBIACEAE Acalypha alopecuroides oe 593% Argythamnia geedieans Sw. 458 A. lucayana Millsp., 515. * Bernardia dichotoma’ Muell’.. Arg... 0360; Cran * Bonania cubana A. Rich., 486, (80). Breynia nivosus (W. Smith) Smeaa, 498. Chemaesyce (blodgetvs (Engelm) Small, 69. GC; hirta Ciev=Milisp. 167, 246. (ee AyEeEeiralis CE.9) Milispe ae 215, 240. C. Techioides (Millsp.) Millsp. 4331:,, <53@:,! (C1 C. mesembrianthemifolia (Jacq.) “Dugand, iI@é * Croton bahamensis serge "Pepper bush", 45 1,7 ,@480R * C. eluteria (L.) Sw., "Sweetwood bark", 35 214, 270, ese (oop linearis Jacq., "Granny bush", 16; 432; 47lype@eous G % coe TMcrduswis cm (oamc ou tons G rosmarinoides Millsp., 254. * Drypetes diversifolia Krug & Urb., 382, 489, CiSoe Euphorbia heterophylla L. 3, 64. E. lactea Haw. * Grimmeodendrom eglandulosum (Rich.) Urb., 112, 533, Ci aoe ep ee % Gymnanthes lucida Sw., "Crabwood", 424, 480, (12). Hippomane | mancinella l., "Manchineel." Manihot esculenta Crant?. Pedilanthus tithymaloides Poit., 497. Phyllanthus acidus (L.) Skeels., "Otaheite gooseberry." - carolinensis Walt. ssp. saxicola (Small) Webster, Oli c= sie, * ee epiphylianthys L., "Rockbush", 134, (53). Pc na rurin ei edi(Or Ricinus communis L., 160. ® Savia bahamensis Brae. 365, 4465-454, Chae) FLACOURTIACEAE * Banara reticulata Griseb., 955, (116)- Casearia bahamensis Urb., 275. Xylosma ilicifolia Northrop, 591. GENTIANACEAE Eustoma exaltatum (L.) Griseb., 87: SAloogete) SwQiilaieidis Puiesing, B25) GOODENACEAE es SOASCVOLla jolwMiMteren Obo) Welailc, 195 (1O08)¢ GRAMINEAE Andropogon glomertus (Walt) B.S.P. "Bed grass", 297. Ko GracilLs Soreass7 CoO, 522s Ao PSRCUSUS Cho) WHINE. , Bide Reupeida tenipes Cav., 359. No Witla OA Set. 592. 2S Subquadriparia (Uriide) Eatelaes,;, 3535 538. Cenchrus echinatus L., 26, 231. Go CevsMlLotces Ile 5 208. Cc. viridis Soren. 13. Chloris barbata (1. ) Stis9 368. C. gayana Kunth, "Rhodes grass", 342. C. petraea (Sw.) Desv., 27. C. sagraeana A. Rich., 536. Cymbopogon citratus (D. Co) Swajoies Wo, Vienon Brass, Sie Cynodon dactylon (L.) Pers., "Bermuda grass", 535. Dactyloctenium aegyptium (L.) Wild., 459. Diglicearre Cnilerrs CRS%4>) K@Gls V/s eis. D. decumbens Steut., "Pangola grass", 343. D. diversiflora Swallen, 53) { o Eleusine indica (L.) Gaertn., 66s BPAGPOSELS Ciltaets Cho) RoBes, Sey Vb. Bo Guliaels (Clg) URie Bes Wales Meme Mevoog stiGe Bo Cemella (iL.) Beauv, ox Ro & S55 S8R6 Lasiacis divaricarta (L.) Hitch. 133, 221 o Leptochloa domingensis (Jacq.) Team. 115 L. plechtostachya K. Schum, 344. Panicum dichotomiflorum Michx., 386. pa maximum Jacq., "Guinea grass", 385. muticum Forsk. "Para er eeeet 345. spalum blodgettii Cage, 25, 835 S53. P. Paspa P. fimbriatum Kunth, 369. P: P. laxum Lam., 267. P. vaginatum Sw. 5 30, 2905 Pennisetum p purpureum Schumach, "Elephant grass", 341. Saccharum officinarum L. TS yee cane", Setaria geniculata (L.) Beauv., 85, 111. 283, 330. So Sevosa (Stn) Beauve, S86. Sorghum saccharatum (L.) Moench, "Guinea corn". Sporobolusppoiretit (R. & S.) Hitch., Boies So Viretmous Cho) Kimleicg 11, O55 BOS. Stenatophorum secundatum (Walt. ) Kuntze, 561. 182 Trichachne insularis (L.) Nees, 505. Uniola paniculata se WSea jOabs!Veeuk. U. virgata Griseb., 2. Zea mays L. GUTTIFERAE Mammea americana L., 362. HALORAGACEAE Proseripinaca platycarpa Small, 581. LABIATAE Leonotis nepetaefolia (L.) R.Br., 495. Salvia serotina L., "Catnip", 127, 145, 26 Is Scutellaria ha Vanensis Jacq., 505 - Teucrium cubense Jacq., 541. LAURACEAE % Nectandra coriacea Griseb., 514, 518, (42). Persea am americana a Mill., "Avocado", 48. LEGUMINOSEAE Abrus precatorius L., "Bead vine". Acacia acuifera Benth. 7» 356% * A. choriophylla Benth., "Cassina", 430, 478, (9). * A. farnesiana (L.) Willa. "Sail needle", 475, (97). * Ateleia gummifera (DC.) Dietr. » 3575 eT fax * eseselpinig pehemenss. La., "Braziletto"™, HOP. (aoe * GC ovalifoliarurp: Nickers", 268, (107). Cagenus cajan (L.) "Millsp. "Pigeon pea", 148. * alliandra formosa (Dth.) Benth., 353), (alae Canavalia rosea (Sw.) DC., "Sea Bean", 229; 338. * Cassia bahamensis Mill. "Stinking pea", T2425 U'C64Dr ® Giebad ora. sis ail Gps diffusa DC., 392. lineata, 159,.288, (8). obutsifolia L., 391. occidentalis L., 124, 463. Woe hag S22 Tentrosema virginianum (L.) Benth. 146, 269, 334, Crotalaria retusa L., 509: Delonix regia (Bojer) Raf., "Poinciana". Desmanthus virgatus (L.) Willd. » 306,2 9482 Desmodium canum (Gmel.) Schinz & Thellung, 329. Do giabrum (Milas )\ DELEs Syn LAAALALAL AI Da plotaxcis muraliss CE.) DC. 3935 Galactia rudolphioides (Gmel.) B. & H., 96, G. spiciformis Wore Gon 336. Gasitiatam@acq.) Urb...) 195. Guilandina bonduc L., "Nickers", 268. 305. * Haematoxylum campechianum Tes "Logwood", (92). * Indigofera s suffruticosa M Mill. aay (103) 6 * Leucaena leucocephala de Wit. L. latisiliqua Gillis., "Jumbay" 93, eae * Lysiloma bahamensis Benth. "Wild tamarind", Cie * L. latisiliqua (L.) Benth., "Horseflesh", (33) Macroptilium lathyroides (L.) Urb., 564. Parkinsonia aculeata L., "Jerusalem thorn". % Piscidia piscipula aL. ) Sargent, "Dogwood", * piehece ob tun keyense Britton "Ramshorn", BOW, SOS Ss Rhynchosia Sp. 195. Sophora tomentosa L., 287. Stylosanthes hamata (L.) Taub. 82, 554. Tamar indus indica L Ban Wtsels 233). Tephrosia cinerea Who) Perso, 32Vs LILLIACEAE Aloe barbadensis Mill Sansevieria sp., "Bowstring hemp". Smilax havanensis Jacq., 492. Yucca aloifolia L., "Spanish bayonet". LINACEAE Linum bahamense North., 529. LOGANIACEAE Cynoctonium mitreola (L.) Brit., 528. Spigelia anthelmia L., 293. LYTHRACEAE Ammania latifolia L., 562. MALPIGHIACEAE * Byrsonima lucida (Sw.) DC., 482, * Malpighia polytricha A. Juss., "Touch me S87, Coy % Bunchosia glandulosa (Cav.) DC., 513, (82). 82, 191, 183 (ibs) 284, 2513); Jay (C9))o 49, 81, AOwW 5 103, 140, 184 Iiniopteris jamaicensas Lb. “Coughtiveinel: MALVACEAE Abelmoschus esculenthus (L.) Moench "Okra". Abutilon permolle (Willd.) Sweet, 144, 248, 494. * Gossypium barbadense L., 427, (120). Herissantia crispa (L.) "Briz., 282. Malvastrum corchorifolium (Desr. ) Brit.,- 464, 50a Melochia tomentosa L. yug, Phymosia abutiloides on ) Desv., 590. Sida acuta Burm., 182, 194, 390. Se eiliiards Ikse 101, 573. Se ispinosack. "396. 5. URenswe. 559. Thespesia populnea (Ee) esolande MELIACEAE Melia azedarach L., "Pride of India". * Swietenia mahagoni (L.) Jacq., 178, (38). MORACEAE Artocarpus altilis (Park.) Fosb., 570. * Ficus jacquinifolia A. Rich, 468, (111). MORINGACEAE Moringa oleifera Lam., 506. MUSACEAE Musa paradisiaca L. MYRICACEAE Myrica cerifera L. MYRSINACEAE * Rapanea guyanensis Aubl., 452, (122). MYRTACEAE Eucalyptus tereticornis Sm., 517. Eugenia ax axillaris (Sw.) Wildida 423. E. buxifolia (Sw.) Willd. 262, 289, (65). * E. longipes Berg., "Sweet "Margaret". 222, -¢859%5 E. lucayana (Brit. ) Alain, 250: 185 - monticola DC., ek (66). E - E- myntoides Poir., 295. IP sidium guajava L., 123, 169. NYCTAGINACEAE Boerhaavea coccinea Mill., "Hogweed", 165, 243. Commicarpus scandens (L.) "Standley "Rat ears", 216. % oe eee eae: Torrubia Longifolia Britton, "Beefwood", 83, (1). * T. Oowuseca (JaGG.) BWeit., SIO, (121). Pisonia rotundata Griseb., 571. NYMPHAEACEAE Nymphaea ampla (Salisb.) DC. var. pulchella (DC.) CASIN6, 301s OLACACEAE Schoepfia Se ee (Rich.) Planch, 395. Ximenia americana L. So. OLEACEAE Jasminum fluminense Vell., 546 Sambac Ait., 566. Forestiera segregata (Jacq.) Krug & Urb., 384,443. ONAGRACEAE Jussiaea suffruticosa L., 567. ORCHIDACEAE % Epicdemeicevim Slo, 295 2e% PALMAE * Coccothrinax argentea Sarg., "Silver top", (OO) 6 Cocos nucifera L. * Pseudophoenix vinifera (Mart.) Becc., "Hog cabbage palm”, (Sys. ~ 50” % THPiAax MLCPrOCArOa SAPHo, VSUCFal© woo”, (5d % Sabal palmetto (Walt.) Lodd.. "Pond CO", (50). PAPAVERACEAE Argemone mexicana L., 33. 186 PASSIFLORACEAE Rassiplora ‘cuprae Ls, 320. Pe “pectinata Griseb., 18, 272. Pe suberosa Lk... 132, sete. Sie PHYTOLACCACEAE Rivina humilis L., 235, 241. PLANTAGINACEAE Plantago major L., 550. PLUMBAGINACEAE Plumbago scandens L., 399. POLYGALACEAE * Polygala obovata Blake, "Strip me naked", 317, (100). POLYGONACEAE Antigonon leptopus H. & A., 372. * Coccoloba diversifolia Jacq., "Pigeon plum", 46, 448, Wot, (CH6). ao Co krigit binds, 351, «479% * C. uvifera Jacq. "Sea grape", (59). PORTULACEAE Portulaca oleracea L., 164. PB. rubricaulis Hep eK., 345.. POTAMOGETONACEAE Potomogeton heterophyllus Schreb., 582. PUNICACEAE Punica granatum L.,."Pomegranate". RHAMNACEAE Auerodendron northropianum (Urb., 240. Colubrina elliptica (Sw.) Briz.,, & Stern, 373. * Cy ferruginosa (Mill.) Sarg-, "260,-Ci2- C. glandulosa Perkins, 418. * Krugiodendron ferreum (Vahl) Urban, "Ironwood", (34). * Reynosia Sepuene ona Urb., "Dollen plum", 47, (14). Zizyphus taylori (eto) JoOMAsvoin, 201 o RHIZOPHORACEAE Rhizophora mangle L., 332. ROSACEAE Chrysobalanus icaco L., "Coco plum", 466. RUBIACEAE * Avmeadielaee iiyiew vole) Weldon yp AG2ly (IAs )s Borreria laevis (Lam.) Griseb., 380. * Casasia clusiaefolia (Jacq.) Urban, "Sea bob", 6, (58). Catesbea parviflora Sw., ee 292. * Chiococca alba CHD Hivehcmee PHS! the! bed") 2108) = 31/3, Wiis ie C. pinetorum Brit., 524. * Erithalis fruticosa nog UBlack tore’, 15, 220, %21, ene % Braol@e Iiwworalsls Sis, 35 S9%5 BSD S505 CO) s 3 Exostema caribaeum (Jacq.) Roem & Schult., "Prince- wood", 2235 (So Guettarda elliptica Slog 232, “OO, SO, (27). G, Seaora (Gs) tame (26). % % % Hamelia patens TOC 5 N67 (28))o x Phialanthus myrtilloides Griseb., "Candlewood", 441, 5325 (ayo % Psychotria ligustrifolia (Northrop) Millsp., "Wild cof- LeQU BOS, Ci2e Rachicallis ager spaus (JaC@o) Oo IKUMEZEc, TOG. w Raachie mitts Bo, Veevyer lousiay, 122, 139, 278, 493, (20)/ Spermacoce confusa Rendle, 331. Seavenuion Lapeis4. Strumpfia maritima Jacq., 1, 219. RUTACEAE % Amyris elemifera L., "White torch", (57). Citrus aurantifolia (Christm.) "Swingle lime". Co AUiraincitim bo, VEbwear orale, 1tso Co Limon Cis) Burm. f. "Lemon". C. sinensis (L.) Osb."Sweet orange". Fagara flava (Vahl.) Krug & Urban. "Satinwood". % Ro puaeows G.5 Mie isimel, M9, (ss * Spathelia bahamensis Marie Wietoria, 9, 225, 30835 C99). 188 * Zanthoxylum coriaceum Rich. "Hercules Club", 420, (105) SAPINDACEAE * Exothea paniculata (Juss.) Radlk., 583, (19). * Hypelate trifoliata Sw., "Soapbush", 230, (96). Melicoccus bijugatus L. "Genip". Serjania diversifolia eon ), Radi ksged 305, cAS7e eee * Thouinia discolor or Griseb., "Three finger", 218, C69e SAPOTACEAE * Bumelia retusa Sw., 303, 338, 340, (6). * Chrysophylium oliviforme., "Wild saffron", 361, (102). * Dipholis salicifolia (L.) A.DC., "Cassadawood", 166m Lich en Gisy ye * Manilkara jaimiqui (Wright) Dubard, "Wild Dilly" 355, Kalono * Manilkara zapota (L.) V. Royen, "Sapodilly" 42, (11 * — Sideroxylon foetidissimum Jacq., "Mastic" 339, (40). SCROPHULARIACEAE Bacopa monnieri (L.) Pennell, 579. Buchnera elongata Sw., 521. Capraria bi ifiora L.,. "Goat weed" 5174. 542. SIMAROUBACEAE * Alvaradoa amorphioides ee 37-6; C76): * Picramnia pentandra Sw., "Bitter bush", 584, (45). Picrodendrom | macrocarpum (A.Rich) Britton, "Olive", Bea eso. SOLANACEAE Capsicum annum Ibe Datura inoxia Mill, 594. D. stramonium L., 585. Nicotiana tabacum L., "Tobacco". * Solanum bahamense L., "Canker berry", 28, 236, (61). * Ss erianthum D. Don..’,) Bil, 259,cato2s Se Higrum be, 150. STERCULIACEAE * Helicteres jamaicensis Jacq., 258, (30). H. semitriloba Bert. Dene (31). * Melochia tomentosa is. TA. 2855 (980. * 189 Waltheria americana L., 68. W. bahamensis Britton, 118, 374. io MAGMA a5 SS, S27. SURIANACEAE Suriana maritima L., 11. THEOPHRASTACEAE * daquinia keyensis Mex.; "Joewood", 5, 274, (109). TILIACEAE * COeCAOLUS INAPSUGUS Ibog WSOzApswsa | CB, AAS5 (10). GE siliquosus L. 403. Triumfetta Semitriloba bog WSO TURNERACEAE * Turnera diffusa Willd., 237, (63). * io Ulatwolie Is, 8, G5 “156, CTI) s TYPHACEAE Typha domingensis Pers. ULMACEAE Trema lamarckiana (R. & S.) Blume, 286 (71). UMBELLIFERAE Anethum graveolens L., 449. Centella erectarck.) Fern., 548. VERBENACEAE Avicennia germinans (L.), "Black Mangrove", 179. * Callicarpa hitchcockii Millsp. "Boarhog bush", 125, 279, (79). * Citharexylum PrUCLOOSUM Ibs, B33, (EM). % Dueanea weoems bog S65 “11, C16)- * Lantana bahamensis Brit., "Goldenrod", 75, 263, 309, Lae oT a % ho tavoltcrawe Ikon YSieee Sagel, 12, Aol, “On (sts Lippia nodiflora (L.) Michx. ~ tie S72 io SeOeGmacirolila Cibo) lloB.Ko, 26". % Petetia domingensis Jacq., "Banana wood", 125 (44). Priva Lapowilaeea Ol.) RPePSap 280 Stachytarpheta fruticosa B. be Robinson, 314. “> omy sealed “0 Peeve ee oT jamaicensis (L.)" VahiR3” 190:eneok ms . ic ry Ot! , edt RaG oe os : VDA CE ARS: Cissus intermedia A. Rich., 98, 324. Camtrit oltata- les 325. ZYGOPHYLLACEAE SARQATSR Guaiacum sanctum L., "Lignum vitae", 370, (91). iy 191 INPENIDIEN IIE Index to common names of plants mentioned in the text. Australian pine Casuarina equisetifolia Avocado Persea americana Bamboo Agave americana Barbados pride Poinciana pulcherrima Bay cedar Sligilananimaresthlman lla Bay geranium Ambrosia hispida Bay lavender Mallatonia gnaphaloides Bay marigold Borrichia arborescens Beans, bonavist Dolkichosmlablabens aan Beans, colored Phaseolus vulgaris Beans, lima Phaseolus lunatus Beefwood Torrubia longifolia Black mangrove Avicennia germinans Black torch Erithalis fruticosa Boarhog bush Callicarpa hitchcockii Bowstring hemp Sansevieria sp. Boxwood Buxus bahamensis Brasiletto Caesalpinia veSicaria Breadfruit Arctocarpus communis Broom bush Evolvulus squamosus — Brown ebony Dalbergia ecastophyllum Buffalo top Thrinax microcarpa Burr Grass CenelnrUs Soa.) Buttonwood Conocarpus erecta Caribbean pine Pinus caribea Cashew nut Anacardium occidentale Cassada Dipholis salicifolia © Cassava Manihot esculenta Cassina Acacia choriophylla Cattail Typha domingensis Coconut Cocos nucifera Coco plum Chrysobalanus icaco Coral tree Erythrina corallodendrum Corn Zea mays Cotton Gossypium barbadense Cow peas Vigna unguiculata Custard apple Annona reticulata Dildo Cephalocereus millspaughii Dilly Manilkara zapota = Divi-divi Caesalpinia coriaria Dollen plum Reynosia septentrionalis Elephant grass Pennisetum purpureum Fever bush Randia mitis Five finger Tabebuia bahamensis Frangipani Plumiera rubra 192 Genip Ginger Granny bush Green ebony Grease bush Guava Hog cabbage Hog plum Horsebush Horseflesh Horseradish tree Indian almond Indigo Ironwood Jerusalem thorn Joewood Johnson grass Jumbay Kamalamay Lemons Lignum vitae Lime Logwood Love vine Madeira Mahoe Mahogany Mango Manroot Mastic Melon, musk Melon, water Milkberry Milkweed Morning glory Nickers Onion Otaheite gooseberry Pangola grass Para grass Paw paw Pigeon berry Pigeon peas Pigeon plum Pineapple Poinciana Poisonwood Pomegranate Pond apple Pond top Pride of India Melicoccus bijugatus Zingiber officinale Brya ~ebenus Corchorus hirsutus Psdium guajava Pseudophoenx vinifera Gundlachia corymbosa Lysiloma Sabres leique Moringa oleifera Terminalia catappa Krugiodendron ferreum Parkinsonia aculeata Jaquinia keyensis Sorghum halepense Leucaena ~Teucocephala Bursera simaruba Citrus limon Guiacum sanctum, G. officinale Citrus aurantifolia Haematoxylum — codescute tal Thespesia populanea Swietenia mahogani Mangifera indica Vallesia antillana Mastichodendron foetidissimum Cucumis melo Colocynthis citrullus Bumelia retusa Chamaesyce mesembrianthemifolia Ipomoea pes-caprae Guilandina bonduc Allium cepa Phyllanthus distichus Digitaria decumbeus Panicum muticum Carica papaya Rhacoma crossapetalum Cajanus cajan Coccoloba diversifolia Delonix regia Metopium toxiferum Punica granatum Annona glabra Sabal palmetto Melia azaderach Princewood Ramshorn Red mangrove Rhodes grass Sail needle Salve bush Sandbox tree Sawgrass sea bean Sea bob Sea grape Sea lily Sea oats Sea pork Silver top Simon finger Sisal Soap bush Sorghum sours Spanish bayonet Squash Star grass Strong back Sugar apple Sweet gale Sweet margaret Sweet orange Sweet potato Sweet sage Sweetwood bark Tamarind Tobacco Tomatoes Wattle White mangrove White torch Wild cinnamon Wild lime Wild locust Wild tamarind Wild tobacco Yellow wood 193 Exostema caribaeum Pithecellobum keyense Rhizophora mangle Chloris gayana Acacia farnesiana Pluchea ros rosea Hura crepitans Mariscus jamaicensis aoe SS JUSS Canavalia maritima Casasia c clusiaefolia Hymenocallis declinata Uniola. paniculata _ Sesuvium portulacastrum Coccothrinax argentea Tabebuia bahamensis Nave Sisalime 9 Corchorus hirsutus Sorghum vulgare Citrus aurantium Yucca aloifolia Cucurbita Sp. Leptochloa plechtostachya Bourreria ovata | Annona squamosa Myrica cerifera Eugenia Tongipes Citrus sinensis Ipomoea batatas Lantana involucrata Croton eluteria Tamarindus indica Nicotiana tabacum Lycopersicon esculentum Eugenia spp. Laguncularia racemosa Amyris elemifera Canella alba Fagara pterota Lysiloma bahamensis Lysiloma bahamensis PlIuchea rosea Fagara flava APPENDIX III Data Sheet used during Analysis of Oldfields CAT OLD FIELD DATA Field Number Photo Reference Height Class _ Soid=Typevigsie’ tees eee Moisture Class Grazing Class __ Species Area Species Area Species Area Species Area Species Area No \ i NGOs , Nos , No i No. 001 ! 11 026 fie at) Lassa lia 0027. | Rr Ree eee Ps ees PV ee Ct 102 eee 003 "028 rt 053 a OT rt 403 4 004 | a0 eD rt 054 OTS ~ SO 005 1 0s0 : 11 055 a es esi 006 : reROsA i056 1) 081 : 11 106 1 007 11 032 Le Rit DT hcsci...° 008 T1 033 1 050 r+ 083 1 100 20s 009 rr 034 11 659 rt 084 109 laid Pa TE i Lie Lassa 0 OP ' i 036 1 it 061 1 an] 056 i it i =<... O12 (| 037 ! | 00 C aaa Laanae SaITURIODT liga Me Eviers ee pies) =a! rt 05 a ik | 074 1a) 039 1, O64 lads Hi Mt anni 015 ! 11 O40 065 i: 090 ' |, 415 - qos ci i ON 5 ota RGR rr 091 11116; | oe an aoe i oue ee 0GT = 7 ge Vit. 018 H rT O83 a ObUmen 0955 H imaitioss oa ee oe a 063 11008 020 11 O45 1! O70 11 095 owe Aaa airs ike 2 ain: 022 Ody, 1 Oe : EO : a. 023 11 O48 1 (Rh ae A list of the 195 APPENDIX IV. minor species encountered in each habitat- type and each age-class. Species are listed in order of impor- tance. Maximum cover Maximum cover OAD LEWh — Maximum cover Maximum cover ils Maximum cover Te Zo WHITELAND value in age-class 1 (less than 5 years). Croton linearis Turnera ulm ulmifolia Wedelia trilobata Croton eluteria value in age-class 2 (5-14 years). Salmia petrobiodes Casisilal lainearaninul Cordia bahamensis Chamaesyce lechiodes Ernodia littoralis Melochia tomentosa Amyris -elemifera Phyllanthus e epiphyllanthus value in age-class 3 (15-29 years). Piscidia piscipula Caesalpinia ovalifolia Calliandra formosa Bursera Simaruba Gossypium barbadense Bursera inaguensis _ Bunchosia glandulosa Solanum bahamense Agave americana Helicteres semitriloba Sceaevola plumierii Cassia bahamensis 0 Fagar ara pterota value in age-class 4 (30-50 years). Sabal palmetto value in age-class 5 (more than 50 years). Jaquinia keyensis Antirrhoea myrtifolia 196 FLATLAND Maximum cover value in age-class 1 (less than 5 years). Guettarda scabra Lantana ~involucrata Corchorus hirsutus Cassia lineata Wedelia trilobata Cordia bahamensis Guettarda elliptica Trema lamarcKiana Psychotria ligustrifolia 10. Solanum eriathum 11. Croton eluteria 12. Thrinax microcarpa 13. Chrysophyllum oliviforme 14. Helicteres semitriloa — 15. Duranta re repens 16. Anthacanthus Spinosus icteric domingensis 18. Solanum bahamense 19. Hamelia patens 20. Caesalpinia ovalifolia 21. Colubrina ferruginosa 22. Banara reticulata 23... Hacus jacquinifolia OOANNDN FSW = Maximum cover value in age-class 2 (5-14 years). 1. Coccoloba krugii 2. Tabebuia bahamensis 3. Leucaena leucocephala 4. Eugenia monticola Randia mitis 5 6. Erithalis fruticosa {fc Eugenia b buxifolia 8 9 Malphigia polytricha 5 Eupatorium | villosum 10. Phyllanthus epiphyllanthus 11. Turnera uUlmifoltansnanetes 12. Calliandra formosa 13. Torrubia obtusata 14. Croton lucidus 15. Croton bahamensis 16. Neobracea bahamensis 17. Picramnia pentandra 18. Callicarpa hitchcockii 19)s Opuntia dillenii Maximum cover value in age-class 3 (15-29 years). Reynosia septentrionalis MayGenus buxifolia Diospyros caribaea Exostema caribaeum Ernodea littoralis Gymnanthes 1 lucida Lantana bahamensis » Ca Cassia Sia bahamensis Turnera diffusa Crossopetalum rhacoma Melochia tomentosa _ Casasia clusiaefolia Bernardia d dichotoma Evolvulus squamosus ° Helicteres ja Jjamaicensis Maximum cover — —- OUO ODN LSWh = —. SS 1 Maximum cover =_— — — EW Mh e Des MH] OO dWONAY LWP = value in age-class 4 (30-50 years). Bourreria ovata Dipholis salicifolia Thouinea discolor Piscidia piscipula Conocarpus erecta Cassia lpdirt lore Coccoloba uvifera Erythroxylon rotundifolium Grimmeodendron n egland ulosum Coccothrinax argentea SADal, palmetto Manilkara emarginata Spathelia bahamensis Jaquinia keyensis Erythroxylon areolatum value in age-class 5 (more than 50 years). Byrsonima lucida Swietenia mahagoni Amyris elemifera Chiococca alba Savian bahamensis Plumiera obtusa Pseudophoenix | vinifera Drypetes diversifolia Jacaranda coerulea — Exothea paniculata Krugiodendron ferreum Brumelia retusa 197 198 13. Ateleia gummifera BLACKLAND Maximum cover value in age-class 1 (less than 5 years). Leucaena leucocephala Fagara pterota Corchorus hirstus 1 2 3 irstus 4, Eugenia monticola 5. oe ae Gr 7. Randia mi 8 9 Cordia bahamensis Wedelia trilobata 12. Psychotria ligustrifolia 13. Cassia biflora 14. Anthacanthus spinosus 15. Exothea paniculata 16. Trema lamarckiana 17. Malphighia polytricha er Jacaranda coerulea 21. Solanum bahamense~ Per Turnera ulmifolia 23. Croton eluteria 24. Evolvulus squamosus NS) « Croton bahamensis 26. Vallesia antillana 28. Banara reticulata 29. Coccothrinax argentea Maximum cover value in age-class 2 (5-14 years). 1. Lantana involucrata Croton linearis saum Maytenus buxifolia Exostema caribeum Eupatorium villosum Chiococca alba Croton lucidus Guettarda elliptica Agave americana Sabal palmetto . Atelia gummifera Nectandra coriacea me - OO ONAN WW LW lh pa 199 13. Cassia lineata 14. Heliotropium parviflorum V5 0 Bunchosia glandulosa Maximum cover value in age-class 3 (15-29 years). 1. Tabebuia bahamensis 2. Piscida piscipula Erithalis fruticosa Diospyros caribaea | Thrinax microcarpa . Crossopetalum rhacoma . Petitia domingensis Citharexylum fruticosum uPMera CASO SE TT 10. Eugenia longipes I. Caesalpinia " bahamensis 12. Bumelia retusa 130 Helicteres jamaicensis 14. Heliecteres semitriloba 155 Spal” bahamensis Maximum cover value in age-class 4 (30-50 years). Dipholis salicifolia Thouinea discolor Gymnanthes lucida Eugenia — buxifolia Bursera inaguensis Savia bahamensis Solanum erianthum Gasasia elusiaefolia Conocarpus erecta Erythroxylon areolatum 12. Polygala Obovata 13. Plumiera obtusa = OW OND LSWHY = [4 << n fh ! ce) 3 jab) ! it) ct a a ed | he BI i= iu) — — Maximum cover value in age-class 5 (more than 50 years). 1. Amyris elemifera Coccoloba krugeii Torrubia obtusata Guettarda scabra 5 Gr to Pseudophoenix vinifera 8 9 - Coccoloba uvifera 10. Erythroxylon n rotundifolium Wo Chrysophyllum oliviforme — a rr re 200 15. Zanthoxylum coriaceum 16. Guaiacum sanctum GOVERNMENT PRINTING OFFICE : 1980 0 - 309-619 > z Lal ee z =~" n RI toh en eee m Yosh > wha 7) = ” : SNI NVINOSHLINS S3I uVY gin ate e RARIES SMITHSONIAN INSTITUTION NOILNLILSNI _ NVINOSHLINS S314 Fe n z a pe o z i < = < = < = < = nese =] 2 =| Zz = = =I 5 zr ; = Ze 2 : g : & ‘2 “4 = $ E = E 2 ’ = = = = = BN. > = >" os Py a 2 2) 2 7) z 7) z= ae ES SMITHSONIAN INSTITUTION NOLLALILSNI_ NVINOSHLINS) S31 N¥Y17" LIBRARIES. SMITHSONIAN INSTITU = on = ie = WW a w = w = a Ho me a x a4 = + cc — oc 5 = ES a= “ 5 c i AN’: S a = = 5 = Se 5 a. 5 Zi Zz aT 2 4 — eT) a SNI NVINOSHLINS S3IYVYSIT LIBRARI ES _ SMITHSONIAN,_INSTITUTION NOILALILSNI _NVINOSHLIWS sa1uvy Lae = 7 = = Cs 5 cS = wo ° a oe 2 NY = 2 = = 5 an 5 & D AQ Ee Ps) — 2 = pe) E (a = ie = no m no im 2) m ~ > a = n = o z ES SMITHSONIAN INSTITUTION NOILALILSNI_ NVINOSHUNS SAluVe al LIBRARIES SMiiraaa INSTITL 2 2 g & = = eae z= = z Ks at ew a = = z WAS 5 z x Oo NAS 2 tg Sif oO aa O . a4 Ba oO g 2NR 32 GH ? g i 2M 2 2 3 = aN SY, = z S \. 2 a > = > j = > = > = eg ”) = w a wo eee ” SNI NVINOSHLIWS saiuvuali LIBRARIES SMITHSONIAN _ INSTITUTION’ NOLLN.LILSNI_NVINOSHLINS S31uVvYy 2 # z vi = uw 2 us (2) a 189) a fer th 4 = = wn = zd ox = &. % bp - ox = « [s =f a z zr = = ES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS (S31aYWy a1 LIB RAR ES SMITHSONIAN INSTITU! z iz z =e ip i a a o @ E @ = @ ie ly < \5 2G: - : : : 5 2 6 -, gre os = as - anh 4 os Zia Z 7 = f Z y Z an SNI NWINOSHLINS S31uVygIq CLIBRARITES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS saa z Z z g z : z 2 < < Ss =; Gh = =| > & = z = y | Wy.z Ww: = Xe : 7 = fe a) a Yy SQ a 7) s Ay a An 7) ae ro) Ko. = AA oO = ~ = = Bas >" = > ” 2 ” ” z ” 4 , ES SMITHSONIAN. INSTITUTION NOILNLILSNI_NVINOSHLINS S31¥Veaiy LIBRARIES SMITHSONIAN INSTITUT! no = (2p) — > ” = ars Ye 4 pe = aw a , ae . —_ — is n) Uy yrs = < c eS = 3 = a) 2 ag z sNI NVINOSHLINS S3IYVYg ee! BRARI ES _ SMITHSONIAN _INSTITUTION ZNOILNLILSNI NVINOSHLINS S3IYVY pe z = S = o = 6 ‘ae Pe) = x = 2 = bE > = > E 3 a = - = F : F = 5 @ o z Oo z a z , 2 ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS. S31uv¥aI7 LIBRARIES SMITHSONIAN _ INSTITU NVINOSHLIWS Ly Gs NVINOSHLINS S3IYVYUSIT SMITHSONIAN wwNosHuins SMITHSONIAN NVINOSHLIN SMITHSONIAN SMITHSONIAN _ sNI_NVINOSHLIWS S31MVYGIT~ LIBRARIES. SMITHSONIAN INSTITUTION NOILMLILSNI NVINOSHLIWS S31UW2 eX eS > INLILSNI SRARIES LNLILSNI JRARIES LALILSNI Lobe Pere PES ~ LALILSNI m PAS m 2 y “WN su on — wn — wn = on = IES SMITHSONIAN INSTITUTION NOILLILSNI_NVINOSHLIWS, Satuvugia LIBRARIES | SMITHSONIAN _| ES z g wy, z = y eels = Z = z= =| Y 5 ze Si = N z S z Ys. S z oO at o) \. ia 2) e a Uys oe) (de a . Sg 22) Ee : SOE : : : : 2 i ae = 2 Ao = (22) ?p) LSNI_ NVINOSHLINS S3IYVYEIT LIBRARIES _INSTITUTION NOILALILSNI_ NVINOSHLIWS = iw z ww 2 vf a uw 7) B = 7) n Oe i = = = © a = = AN ., = a 2 a ee = A f = S W YP oc S oe S o = RAS SN a = SS o = o- — (aa) oO S a (e) \S = ram (} SN al z aS a S a S =) 2 i. = 1ES_ SMITHSONIAN INSTITUTION NOILOLILSNI_NVINOSHLIWS Saluvudi1 LIBRARIES SMITHSONIAN _ = ~ ; = = as = Led ie, = a S c — ics} = w — OO — a@ = 4 f aa = 20 rE a iS 4.6f, gy x 2 GY 2 : S = 5 4 = = OG z = Za = 2 - Gap = 2° * 2 : Z 3 a a — = w » = = ISNI_NVINOSHLINS S3IYVUG!IT_ LIBRARIES SMITHSONIAN INSTITUTION NOILMLILSN! NVINOSHLINS z a7) Z pes n z= Re 2) = =< = ae = < x. = = P34 Fe =i “Z a iz SS = z 7 (oe) oe (S) : = ae eS = > GZ = n 2 7) OF a eta 7) Bo ae 7) TES SMITHSONIAN INSTITUTION NOILOLILSNI_NVINOSHLIWS S31YVYSIT_LIBRARIES SMITHSONIAN _ uw Fa wu > ae Ww 72 ul = a. i ff “s z 4 XE = = Boge oe = a S = = Uy, 7 Ss a Na 5 > = > = “> = = SS = 2 - 2 i 2 - NS" 5 = n = n° = on m 2 ee 2 n igs = 7) = w z ra) on = 1ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31iyvygi7_LIBRARIES SMITHSONIAN g CAR a) z oe) : Ze g z = < = vy < = x = < 3 2\o i G3 = 5 s z \ 8 2 IN 8 JOG 2 2 2 : SE E Nv 8G fi = 2 = = E 2 Ss Z S ze ie & LSNI_NVINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILMIILSNI_NVINOSHLINS | & ul Z us 5 ul & 2 = a = = 5a e = = c WINS 2 a < oa < 2 << =] AN « 5 « S o Sj co oO > jus} = oo. a oO = co i ‘a a 2 = S a 2 Fee a 1ES_SMITHSONIAN_INSTITUTION NOILMLILSNI NVINOSHLINS S3IYVYSIT_LIBRARIES_ SMITHSONIAN z z Serine E 5 = = ‘oO = RK = ow — ee) 5 a = s SK re a = Gy “2 2 z = UES 2 z —fY * a a4 = z = Gs - Gry ca ASNI_NVINOSHIIWS Sa1uVvud (Wu BRARI ES SMITHSONIAN INSTITUTION | NOILMLILSNI_ NVINOSHLINS = = a Cone Sse. = cs fy Z =] az NS = > = > tj (2) ae oO aS = ae (o) s 9 alt (e) y eB 2 Re 8 B zB 2 y = =. = » 2 = Z = = > = : > = . > g = (7) z, (/2) 0 Zz ” ~ = 7) CTS ION NOILNLILSNI_NVINOSHLINS S31YVYAIT_ LIBRARIES SMITHSONIAN _ ‘st > ens n Ss Fa a aa a a NE uw ie w 7) = Uae eS —_ wn XE = n —_ EPA We at =. ih jp z GX = = «gy = “f/f 2 < 2S = : a“iy 2 « YY cc S BOO 5 a YD F NVINOSHLIWS NVINOSHLIWS NSTITUTION | INSTITUTION | saiuvudit| INSTITUTION Saluvaugiy [odie eS eo ed ee INSTITUTION RILSER ECP 8 8 eA gi ape eed ae INSTITUTION saiuvaal INSTITUTION SL ew . * . . sa Git thie fice eh ne} eget meh SMITHSONIAN INSTITUTION LIBRARIES ee ate i ‘ 4 . ia yh pe haa I) UMN YEYA | ae ehacsues eet nel fate F f 2 Y | i 4 aires Wi i] | | | F ro , ne : | AAU AT WL thats tsar a iy eats 3 9088 01375 3785 A Wa: f y % ‘ , ‘ é ot es Fi Pee ay rer erry ‘ \) “cab ynint eis ives Y bode re ‘ * : . TA . tay re! antes phos 6 He : . if ‘ 3 natant ts ; RT: F a rprerb ey i ‘ y ¢ PAY | ~ 4 , Pe ; watt ‘ : : ad PCM ih tk a f t } Ae “ ren : 7 a eld Dab DPE Me aN bre ‘ , yh hve H taf Carona 2a ph uh . A 7 i . M Brie race , eitatts) i k brah Tey f if sy py TPacaish 52 titi) CALLER ea i ; tsi"; , ; wee ME ae q t ; Gath i hata + ‘ Hy, ‘ , ef z ‘ ‘ i. 44 ' J f ‘ f i Drercns A el Pte sh ‘ f \ - ia MARAE br dei fats its F ayypowta’ 7 r ‘ \ . Dae Sao Pe oe : sain wh & ‘ Lente : wi, : LAPT Ny . ,) "4 ae | DAS oh CX ttl bibs ‘ ‘ ‘ . 4 * - Rehan ; aay r : SHED SA ey, ee s) : slopes ‘a f . ~ mien { . rel be ST BLY NNsnta le Hpac tS pS cay VV Keay f } Cha A ann) ¥ « Aas ". mt, Se aE : as ‘ : Gury . ¢ “A> MRSA e e ‘ . ‘ \ ‘ Weal aS { Ma AN i ‘ CYS A SPs . a . * . MEO hal’ : ; manatee Pvisalan tier, Ota 5 WES ipootaxs : vi oat AE : Fisted Dyas SUSAN . P , ; UAT Skeet a Vue Sek t A’wiviine ash) i 7 7! , Ste anbysearntins Starrs ‘ bz ot ) STN eal . : wi gvarwivts apy SNe uranyl H : TMNT ese ewiere ely 4 ‘ a ; ; ' 3 Ne rurt Bae Pot < ; perils Se EDS : ras ES . h . KAP om SN : . 3 A Pah “ « oak 42 ARTS Sirs PVN L Rtas