Bulletin of the British Museum (Natural History) Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. IV. Rhodophyta (Florideae) 2. Genera G James H. Price, David M. John and George W. Lawson Botany series Vol 18 No 3 29 September 1988 The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in four scientific series. Botany, Entomology, Geology (incorporating Mineralogy) and Zoology, and an Historical series. Papers in the Bulletin are primarily the results of research carried out on the unique and ever-growing collections of the Museum, both by the scientific staff of the Museum and by specialists from elsewhere who make use of the Museum's resources. Many of the papers are works of reference that will remain indispensable for years to come. Parts are published at irregular intervals as they become ready, each is complete in itself, available separately, and individually priced. Volumes contain about 300 pages and several volumes may appear within a calendar year. Subscriptions may be placed for one or more of the series on either an Annual or Per Volume basis. Prices vary according to the contents of the individual parts. Orders and enquiries should be sent to: Publications Sales, British Museum (Natural History), Cromwell Road, London SW75BD, England. World List abbreviation: Bull. Br. Mus. nat. Hist. (Bot.) ©British Museum (Natural History), 1988 The Botany series is edited in the Museum's Department of Botany Keeper of Botany: Mr J. F. M. Cannon Editor of Bulletin: Mr J. R. Laundon Assistant Editors: Dr A. J. Harrington and Miss M. J. Short ISBN 0 565 08021 0 ISSN 0068-2292 Botany series VollSNo 3 pp 195-273 British Museum (Natural History) Cromwell Road London SW7 5BD Issued 29 September 1988 Seaweeds of the western coast of tropical Africa and y adjacent islands: a critical assessment. IV. Rhodophyta (Florideae) 2. Genera G James H. Price Department of Botany, British Museum (Natural History), Cromwell Road, London SW7 5BD David M. John Department of Botany, British Museum (Natural History), Cromwell Road, London SW7 5BD George W. Lawson 23 Sheffield Terrace, London W8 7NQ Contents Synopsis 195 Introduction 195 Species list •. •: 197 Acknowledgements 255 References 255 Synopsis This paper assembles and, so far as is possible without extended field and herbarium studies, examines critically the validity of records of marine and brackish-water Rhodophyta (Florideae) for the western coast of tropical Africa. The whole mainland coastline from the northern boundary of Western Sahara southwards to the southern boundary of Namibia, the oceanic islands from the Salvage Islands southwards to Ascension and St Helena, and all islands close to the African mainland are included in the area covered. Each species entry includes all traced records for the species, the names which have previously been applied to it for the area, and additional comments or evaluation, as necessary. Comments are also provided at generic or generic group levels in very complex cases. One new combination is effected: Polycavernosa domingensis (Kiitzing) J. Price & D. John (Sphaerococcus Domingensis Kiitzing). Introduction The area dealt with in this part of the work is identical with that covered in parts published previously (Lawson & Price, 1969; Price, John & Lawson, 1978, 1986; John, Price, Maggs & Lawson, 1979). Country names employed and their earlier equivalents, and the names of island groups included, are either listed in the legend or both listed and shown on the map in Fig. 1. Genera with the initial letter G and constituent species are listed in alphabetical order; space and organisational requirements within the Bulletin have required subdivision of a text originally intended to cover as one issue all the remaining genera for the Rhodophyta (Florideae). Each main entry consists of: (i) The major bold heading, representing the currently-accepted name and authorities, (ii) Subsidiary italicised headings at intervals within the entry. These are in square brackets and essentially subdivide the overall entry. They represent the different ways in which the species has been referred to throughout the past publication patterns for the area. Incorrect citations from past literature have been maintained in these subsidiary heads so that there shall be no doubt as to which record we attribute to which species or lower Bull. Br. Mus. nat. Hist. (Bot.) 18 (3): 195-273 Issued 29 September 1988 196 J. H. PRICE, D. M. JOHN & G. W. LAWSON Fig. 1 The coastline of tropical west Africa and the offshore islands. 1, Salvage Islands; 2, Canary Islands; 3, *Western Sahara [- former Spanish Sahara, Spanish West Africa] (includes the often quoted Rio de Oro, the southern region of the country, but excludes Ifni); 4, Mauritanie; 5, Senegal; 6, Gambia; 7, Guinea-Bissau [= Portuguese Guinea]; 8, Guinee; 9, Sierra Leone; 10, Liberia; 11, Cote d'lvoire; 12, Ghana; 13, Togo; 14, Benin [= Dahomey]; 15, Nigeria; 16, Cameroun; 17, t Bioko [= Macias Nguema Biyogo, Fernando Poo]; 18, Principe; 19, Sao Tome; 20, t Equatorial Guinea [= Spanish Guinea]; 21, Gabon; 22, $ Republic of the Congo; 23, Cabinda; 24, Zaire [= Congo Republic]; 25, Angola; 26, Namibia [= South West Africa]; 27, Ascension Island; 28, Saint Helena; 29, Pagalu [ = Annobon] . The Cape Verde Islands, which lie immediately to the west of Dakar (Senegal) , have been omitted from this map but are included in the species list that follows. * The former colony of Spanish Sahara no longer officially exists, the territory it once covered being divided, by agreement, between Morocco and Mauritanie. The effective date of the division, Spain concurring, was 28 February 1976, although guerrilla opposition delayed matters until a formal agreement on 14 April 1976. The attempt to maintain the territory as the Democratic Sahara [Saharan] Arab Republic has apparently entered the 'realm of myth' (Gretton, 1976). t Nos 17 (Bioko) and 20 (Spanish Guinea, = Rio Muni) on the original map (part I) are now jointly administered, with Annobon (29), as Equatorial Guinea. Bioko and Annobon are entered separately, where appropriate, in the species list. $ Loango, a name much used by earlier collectors such as Welwitsch, was formerly a coastal region of West Africa. Its application appears to have included much of the coastline of the Republic of the Congo (22), as well as of Cabinda (23) and Zaire (24). Because by far the longest and rockiest part of the Loango coast lies now within the Republic of the Congo we have attributed all marine algal records from Loango to the Congo. RHODOPHYTA (pLORIDEAE) OF TROPICAL AFRICA 197 taxon level; only when clarification was required for comprehension have changes been made in sub-head citation, in which case explanation is given in intermediary or terminal notes. (iii) The distributional data, with countries and island groups arranged in a single alphabetical order. More generalised but still relevant statements of distribution follow the specific country list. Complete distribution patterns require a scan of records established under all names by which a species is known for this or adjacent areas. Hence, generalised distribution statements are included verbatim since it is not always clear for precisely which countries within the area they establish records. In all these cases, countries/ islands/generalised statements, numbers within parentheses after the names refer to corresponding numbers in the references. In the present reference list, for ease of readjustment for subsequent parts, references have not been renumbered but simply omitted or added terminally and additionally numbered as appropriate for the present part. Again, lists of references are therefore only partially interchangeable between different parts of the overall list. As in 1986, we have considered the alphabetical sequence of authors' names to be of greater importance than the strict numerical sequence. References therefore appear in alphabetical order but only cross-referred as to numerical order, where there is an anomaly for various reasons of rearrangement or addition. 'References' in the listing cover also manuscript and expeditionary sources, as well as works currently in press. (iv) Additional qualifying notes, intermediary or terminal, have proved to be required in many cases. These notes appear below whole entries or individual parts of entries to which they specifically refer. Citation of references in the notes depends for its form on whether those references contain species records (when they consist of authors' names, followed by the reference number in the terminal list and, where appropriate, the relevant page numbers after a colon) or do not (when they consist of authors' names, date of publication, and sometimes page numbers after a colon). Species nomenclature has been revised as far as possible and the complete author citation is given for each currently-accepted combination. The subsidiary italicised headings (and any other discarded combinations that require reference in the context of records of accepted species for the list area) are included as cross-referencing entries to the currently-accepted names in the overall list. The necessarily preliminary nature of all parts of the treatments presented has been emphasised for each previously-published part of the list and applies no less here. As indicated in previous parts, critical updating of the overall text is kept firmly in mind for the whole work, although feasibility of that process will remain the final determinant at the appropriate time. We would appreciate notification of any detected errors and omissions from any of the parts. Species list Galaxaura The complexities of taxonomy/nomenclature in this genus have long been recognised, although resolution of the problems still leaves much to be desired. Complexities have been hitherto compounded by a general tendency to apply different epithets to plants later recognised as tetrasporic and gametophytic generations of the same life-history. The anatomy exhibited by these different life-history stages is occasionally sufficiently dimorphic for some species to be entirely misunderstood and the different stages thus to be placed in separate anatomically- distinguished sections of the genus (cf. Howe, 1917, 1918; Cordeiro-Marino, 108; Papenfuss et al., 436) -examples are G. obtusata (gametophyte hitherto placed in Spissae*, tetrasporophyte in Cameratae*) and G. marginata (gametophyte in Vepreculae*, tetrasporophyte in Brachy- cladia*). The major earlier work on this genus, that of Kjellman (313), has been subject to major * Sensu Chou (103), Spissae = Umbellatae; Cameratae = Robustae; Vepreculae is maintained; Brachycladia = Arboreae. 198 J. H. PRICE, D. M. JOHN & G. W. LAWSON taxonomic reorganisation of conclusions by Papenfuss & Chiang (435) and Papenfuss et al. (436) ; this has resulted in the reduction of large numbers of species to synonymy , details of which will be clear from the species entries. We have generally followed the decisions made by Papenfuss and colleagues (434; 435; 436), although even then the overall situation remains complex and confusing. That situation is best summarised by Papenfuss (434: 271) and Papenfuss et al. (436: 401-402): The genus Galaxaura has suffered a great deal of misunderstanding. . . . The various species have been assigned to nine genera if Galaxaura is included.' Steentoft (535: 121) summarised aspects of the local western Africa situation, outlining the putative nature of the species distributions deducible from available reports and stating: 're-examination of the S. Tome and Principe material shows that these islands probably have the richest Galaxaura flora in the Gulf of Guinea.' On the matter of wider consideration of interrelationships and status of the Galaxauraceae as a whole, see recent publications by Huisman (especially 1987). Galaxaura adriatica Zanardini See Galaxaura oblongata (Ellis & Solander) Lamouroux. Note. Considered a synonym of G. oblongata by Papenfuss & Chiang (435), Papenfuss et al. (436), and Magruder(383). Galaxaura annulata Lamouroux See Galaxaura rugosa (Ellis & Solander) Lamouroux. Note. Placed in synonymy of G. rugosa (q.v.) by Papenfuss et al. (436: 421). Galaxaura coarctata Kjellman See Galaxaura rugosa (Ellis & Solander) Lamouroux (for records) and Galaxaura decaisnei J. Agardh (for taxonomic comment in terminal notes). Note. Supporting information for this synonymy was published in Papenfuss & Chiang (435) and (especially) in Papenfuss et al. (436: 422), where the full picture is presented. Galaxaura comans Kjellman See Galaxaura lapidescens (Ellis & Solander) Lamouroux and Galaxaura oblongata (Ellis & Solander) Lamouroux. Note. Relationships between the taxa named above need clarification. According to Papenfuss et al. (436: 407-8, 427), the type (a tetrasporophyte) of G. comans, from Guadeloupe, West Indies, is actually a plant of G. lapidescens, not of G. oblongata. G. lapidescens was thought to be the tetrasporophyte of G. oblongata, but this has more recently been questioned and demonstrated to be untrue for at least one geographical area (Hawaii); see the notes to G. lapidescens for further detail. Despite the statements, referred to above, in Papenfuss et al., the same authors elsewhere (p. 410) reiterated Howe's (1918: 197) suspicion that G. oblongata (gametophyte) and G. comans (tetrasporophyte) represent the same species. Galaxaura cylindrica (Ellis & Solander) Lamouroux See Galaxaura oblongata (Ellis & Solander) Lamouroux and Galaxaura lapidescens (Ellis & Solander) Lamouroux. Note. According to opinions expressed by Papenfuss et al. (436: 415-418), G. cylindrica is a synonym of G. oblongata (q.v.); Chou (104) had earlier (1947) indicated that she felt differences between these taxa to be arbitrary rather than real. Following generally the approach of Taylor (540), who commented that G. cylindrica and G. oblongata are not too difficult to distinguish despite branch-diameter convergence, Norris & Bucher (416: 193) were less definite. They maintained the application of the name G. cylindrica for their Belize material, pending further study, but acknowledged morphological overlap and possible conspecificity. Galaxaura decaisnei J. Agardh See Galaxaura obtusata (Ellis & Solander) Lamouroux [for general concept] and Galaxaura rugosa (Ellis & Solander) Lamouroux [for records under G. decaisnei for the Canaries]. Note. It has been established that the material representing the type of this taxon, and therefore the basis for the general concept, is a gametophyte from Barbados, West Indies, and should actually be attributed to RHODOPHYTA (FLORIDEAE) OF TROPICAL AFRICA 199 G. obtusata (Ellis & Solander) Lamouroux (Papenfuss et al. 436: 418-421); Seagrief (570) followed Papenfuss et al. in accepting this equivalence. By contrast, Vickers's (547) material, from the Canaries and named as G. decaisnei, was re-examined by B0rgesen (68: 70) and found to be what he then called Galaxaura squalida Kjellman; this latter was considered to be the sexual generation of Galaxaura flagelliformis Kjellman, both these names subsequently being placed in the synonymy of G. elongata J. Agardh, which is now considered a synonym of Galaxaura rugosa (Ellis & Solander) Lamouroux. G. flagelliformis has also been variously implicated in Galaxaura lapidescens auct. and, through the latter, in G. cylindrica (q.v.) and G. oblongata (q.v.). Some Despreaux Canaries specimens, originally held in Paris under the name G. decaisnei}. Agardh (Herb. Thuret), were transferred to Uppsala, where they were then held as Galaxaura coarctata Kjellman, another name now considered to be a synonym of G. rugosa (q.v.). Galaxaura elongata J. Agardh See Galaxaura rugosa (Ellis & Solander) Lamouroux. Note. Papenfuss et al. (436: 421-424) and Cribb (113: 26-27) both reduced G. elongata J. Agardh to the synonymy of G. rugosa. Re-attributions of records for the list areas have required the assumption that the original determination represented a correct application of the previous name. Thus (largely following Papenfuss et al., 436) records previously attributed in our text to G. elongata have been reallocated according to the following equivalence: records as: G. decaisnei J. Agardh: see G. rugosa (Ellis & Solander) Lamouroux. G. elongata J. Agardh: see G. rugosa (Ellis & Solander) Lamouroux. G. flagelliformis Kjellman: see G. lapidescens (Ellis & Solander) Lamouroux. G. lapidescens auct.: see G. lapidescens (Ellis & Solander) Lamouroux. G. rugosa auct.: see G. rugosa (Ellis & Solander) Lamouroux. G. squalida Kjellman: see G. rugosa (Ellis & Solander) Lamouroux. Galaxaura filamentosa Chou in W. Taylor [As Galaxaura filamentosa Chou ex W. Taylor] 'tropical West Africa' (561). [As Galaxaura filamentosa Chou] Sao Tome (350; 535; 586). 'in tropical parts of the Atlantic and Pacific Oceans' (350; 586). Tropical Africa (N. Gambia- Congo river)' (598). [As Galaxaura lapidescens Lamouroux] Sao Tome (251; 263; 264; 265). Note. There has been much confusion as to the authorities to be correctly cited for this taxon. The type derives from a collection by W. R. Taylor in the Galapagos Islands and the first description by R. C. -Y. Chou was published in Taylor's (1945) paper on the Allan Hancock Expedition collections from there. This Latin description is verbatim the same as that published later in the same year (September, as opposed to May) by Chou (103) in her own paper. Thus, the authorities are correctly to be stated as above. In their most recent revision of Galaxaura, Papenfuss et al. (436: 404, Table I) mentioned a report by Isaac (1971) of the species from Kenya, but seem not otherwise to have dealt with the matter. The material referred to above, and published by Hariot (251) and Henriques (263; 264; 265) as representing G. lapidescens Lamouroux (q.v.), is included here on the basis of the opinion expressed by Steentoft (535) that it was misdetermined. Steentoft nevertheless observed that Sao Tome plants were more fastigiately branched and possessed somewhat shorter assimilators. Note that the report by Hariot (251) is merely secondary, being directly based on the earlier papers of Henriques (263 et seq.). Galaxaura flagelliformis Kjellman See Galaxaura lapidescens (Ellis & Solander) Lamouroux [for records] and Galaxaura elongata J. Agardh [for taxonomic comment]. Galaxaura fragilis (Lamarck) Lamouroux ex Decaisne [and other authority combinations] See Galaxaura oblongata (Ellis & Solander) Lamouroux. Note. Authorities hitherto quoted for this combination were usually '(Lamouroux) Decaisne', 'Lamouroux' or 'Decaisne'; critical nomenclatural and taxonomic revision by Papenfuss et al. (436: 415-418) resulted in recognition of the above more accurate citation. Those authors presented also detailed synonymy, which we have accepted, consequent upon examining the type (a presumed gameto- phyte from the West Indies, filed under Dichotomaria fragilis, ex Herb. Lamarck, in Herb. Mus. Paris). 200 J. H. PRICE, D. M. JOHN & G. W. LAWSON Galaxaura frutescens Kjellman See Galaxaura marginata (Ellis & Solander) Lamouroux. Note. Papenfuss et al. (436: 413-4) indicated these two taxa (inter alia) to share many attributes, but they did not have available material in sufficient amounts or state adequate to arrive at a final decision. The type of G. frutescens Kjellman is ex Herb. Areschoug and derives from Bahia, Brasil; the material, like that of G. marginata, shows a marginal ridge. Later in the same publication (436: 427), Papenfuss et al. indicated their opinion that G. frutescens is likely to be shown by future research to be conspecific with G. marginata. Galaxaura fruticulosa Kjellman See G. subfruticulosa Chou in W. Taylor and the notes to G. fruticulosa (Ellis & Solander) Lamouroux. Galaxaura fruticulosa (Ellis & Solander) Lamouroux [As ' zoophyta fruticulosa (galaxaura)'] Canaries (219). Note. See also the entry for Galaxaura subfruticulosa Chou in W. Taylor in the context of the possible conspecificity of the present taxon with G. fruticulosa Kjellman and with G. subfruticulosa. The above name is used in referring to the organism that provided the substrate for Conferva villum Agardh (q.v.). The Galaxaura taxon involved is clearly the same as that recorded earlier in the same work (Gaudichaud, 219: 6-8, 1826) as 'galaxaura lapidescens'. See also the entry for the latter species. Galaxaura lapidescens (Ellis & Solander) Lamouroux Canaries (128A; 306B; 436; 598; 604; 605). Cape Verde Islands (598). Salvage Islands (598). [As Galaxaura lapidescens Lamouroux] Canaries (38; 126; 127; 142; 145; 219; 259; 263; 264; 318; 391; 535). Note. See also the record from Gaudichaud (219: 156), as Galaxaura fruticulosa (q.v.). Cape Verde Islands (38; 145; 259). St Helena (142; 259; 260; 391). Throughout all tropical seas' (410). [As Galaxaura lapidescens (Solander) Lamouroux] Canaries (27; 131). [As Galaxaura lapidescens (Solander & Ellis)] Salvage Islands (381; 439). Note. Piccone (439: 35; 55) noted this taxon as 'G. oblongata (Sol. et Ell.) cf. G. lapidescens (Sol. et Ell.)', based entirely on the record in Lowe (381). [As Galaxaura (Microthoe) lapidescens Lamouroux] Canaries (257). Tropical Ocean' (257). [As Galaxaura (Dichotomaria) lapidescens] Canaries (313). [As Galaxaura lapidescens (Solander) Lamouroux et var. tomentosa (Kiitzing)] Canaries (390; 436; 441; 444). [As Galaxaura lapidescens Lamouroux var. tomentosa Kiitzing] Canaries (439). [As Actinotrichia lapidescens Schmitz] Canaries (547). Note. See the note at the entry for Actinotrichia lapidescens Schmitz. [As Galaxaura flagelliformis Kjellman] Canaries (13; 38C; 68; 191; 226; 227; 379; 490). Salvage Islands (38C; 556A). [As Galaxaura flagelliformis (Kjellman) B0rgesen] Canaries (14; 16; 230; 247; 375; 489; 556). Cape Verde Islands (556). Salvage Islands (231; 375; 556). [As Galaxaura flagelliformis (Kjelin.) B0rgs. [sic!]] RHODOPHYTA (pLORIDEAE) OF TROPICAL AFRICA 201 Canaries (229; 610). [As Galaxauraflagelliformis (Kjellman) emend. B0rg.] Canaries (38B; 68; 71; 78; 214; 235). Salvage Islands (38B; 381). [As ? Galaxauraflagelliformis (Kjellman) emend. B0rgs. (G. lapidescens Pice.)] Salvage Islands (452). [As Galaxauraflagelliformis Foslie] Canaries (235). Note. Records published by Henriques (263; 264; 265) for Sao Tome [Ins. Rolas] under the name Galaxaura lapidescens Lamouroux, based on collections by Quintas, are actually representative of Galaxaura filamentosa (q.v.) (Steentoft; 535). The latter author (535), nevertheless, found the Sao Tome plants to be more fastigiately branched and to have somewhat shorter assimilators. Steentoft (535) also, following Howe (1918: 196), suggested that G. lapidescens represents the asexual (tetrasporophytic) phase of the sexual Galaxaura cylindrica (Ellis & Solander) Lamouroux, now considered (Papenfuss et al. 436: 415-418) as synonymous with Galaxaura oblongata (Ellis & Solander) Lamouroux (q.v.). The tetra- sporophyte/gametophyte relationship of G. lapidescens and G. oblongata has been questioned by some workers and has recently (Magruder; 383) been shown to be untrue for Hawaii, where the G. oblongata tetrasporophyte is small, inconspicuous, and filamentous. See the notes to G. oblongata and the discussion in Magruder (383) for additional comment. See also the notes to Galaxaura comans Kjellman, Galaxaura fruticulosa (Ellis & Solander) Lamouroux, and Galaxaura subfruticulosa Chou in W. Taylor. According to Audiffred (38C: 175), G. flagelliformis Kjellman is the tetrasporophytic phase of G. squalida Kjellman. Galaxaura marginata (Ellis & Solander) Lamouroux Cameroun (288; 350; 586). Cote d'lvoire (287; 288; 350; 586). Gabon (288; 294; 350; 586). Ghana (288; 299; 338; 350; 535; 586). Liberia (129; 287; 350; 586). Principe (288; 350; 535; 586). Sao Tome (251; 288; 350; 535; 539; 586). Senegambia(296;319). Warm Atlantic (535). 'in warm temperate and tropical parts of the Atlantic Ocean.' (350; 586). 'Tropical Africa (N. Gambia- Congo river)' (598). [As Galaxaura marginata (Solander) Lamouroux] Ghana (153). [As Galaxaura marginata (Solander in Ellis) Lamouroux emend. Kjellman] Cameroun (139). Note. The De Toni (139) record is based directly on data provided in Pilger (454). [As Galaxaura marginata Lamouroux] Sao Tome (251; 263; 264; 265). Note. Reports in 263 and 264 include records from the adjacent islet of Rolas. [As Galaxaura marginata Kiitzing f. marginata J. Agardh] Senegambia (390). [As Galaxaura (Microthoe) marginata Lamouroux] Tropical Atlantic . . . Ocean' (255). 'Tropical Oceans' (257). [As Galaxaura tenera Kjellman] Gabon (350; 435; 570). 'Tropical Africa (N. Gambia - Congo river)' (598). 'widespread in warm temperate and tropical seas' (350). Note. The Gabon records have often been presented in the form of the phrase 'Gabon River, West Africa' (435; 570). [As Galaxaura ventricosa Kjellman] Gabon (104; 139; 313; 435; 436; 535; 561; 570). Tropical eastern Atlantic' (277; 561). 202 J. H. PRICE, D. M. JOHN & G. W. LAWSON Tropical western Africa' (277). Note. The records from Gabon have often been presented (313; 435; 436; 570) in the form 'Gabon River, West Africa'. De Toni gave even further detail (139: 198) in his Latin statement: 'in Oceano Atlantico inferiore in ostio fluminis "Gabon" Africae occidentalis meridionalis (E. Jardin)'. He also repeated Kjellman's (313) observation that the taxon was most closely allied to G. veprecula. [As Brachydadia australis Sonder [= Zanardinia marginata (Solander) J. Agardh]] Gabon (250). [As Brachydadia marginata (Solander) Schmitz] Cameroun (454). Senegambia (131; 390). [As Brachydadia marginata f. marginata (Kiitzing) J. Agardh] Senegambia (131). [As Brachydadia marginata (Solander) Schmitz f. linearis (Kiitzing) J. Agardh] Cameroun (454). [As Brachytrichia marginata (Solander) Schmitz] Sao Tome (251). [As Zanardinia marginata J. Agardh] 'Warm Atlantic' (410). [As Zanardinia marginata J. Agardh f. marginata J. Agardh] Senegambia (27). Note. Cordeiro-Marino (108) repeated Howe's (1918) comments on dimorphism in Galaxaura; the overall conclusion then was that G. marginata, for which only the tetrasporophyte was known, could possibly have G. occidentalis B0rgesen (60: 109) as its sexual phase. A preliminary revision of Galaxaura by Papenfuss & Chiang (435) concluded that G. tenera Kjellman embraced the 'species' G. stupocaula Kjellman [stupicaula], G. clavigera Kjellman, G. veprecula Kjellman, and G. ventricosa Kjellman. These authors further commented that various characteristics of G. tenera agreed very well with those of G. marginata and that the two may have been conspecific. On the basis of variable, often overlapping, local Brazilian specimens, Cordeiro-Marino (108) agreed but hesitated to formalize the issue pending study of types and further field-populations. Subsequently, Papenfuss et al. (436: 411-415) reduced G. tenera Kjellman to synonymy with G. marginata (Ellis & Solander) Lamouroux, along with G. clavigera Kjellman, G. occidentalis B0rgesen, G. stupocaula Kjellman, G. ventricosa Kjellman, and G. veprecula Kjellman. In work on material from the list area, Steentoft (535: 122) noted three growth forms amongst specimens from Sao Tome and Principe; one, including sexual material, manifested characteristics intermediate between G. ventricosa Kjellman (from Gabon) and G. veprecula Kjellman (from the Indian and Pacific Oceans). Asexual material amongst the specimens had much in common with G. marginata, G. stupicaula, and G. frutescens (both segregates from G. marginata and created by Kjellman). Steentoft (535) therefore suggested 'the whole complex should be re-examined together.' Seagrief (570), dealing with material from nearby South Africa, concluded that Galaxaura marginata sensu Krauss (1846: 214), Zanardinia marginata sensu Barton (1893: 171, pro parte), Z. marginata b. diesingiana (Zanardini) J. Agardh, and Brachydadia marginata f. diesingiana (Zanardini) De Toni were all related to Galaxaura diesingiana Zanardini, not to the true Galaxaura marginata as recorded here. Galaxaura oblongata (Ellis & Solander) Lamouroux Canaries (2; 8; 13; 16; 68; 72; 80; 97; 128A; 191; 226; 227; 230; 233; 306B; 375; 435; 436; 489; 490; 535; 584; 598; 604; 610). Cape Verde Islands (535; 598). Cote d'lvoire (288; 350; 586). Gabon (288; 294; 350; 586). Ghana (153; 288; 299; 338; 350; 535; 586). Liberia (129; 350; 586). Principe (288; 350; 436; 535; 586). Sao Tome (288; 350; 436; 535). Salvage Islands (38B; 231; 375; 381; 439; 452; 598). 'a tropical Atlantic alga' (2). 'circumtropical' (280; 435). 'cotes occidentals d'Afrique et aux Canaries' (184). RHODOPHYTA (pLORIDEAE) OF TROPICAL AFRICA 203 'in warm temperate and tropical seas' (350; 586). 'Pantropical and subtropical' (535). Tropical Africa (N. Gambia- Congo river)' (598). 'world wide distribution in tropical and subtropical waters' (383). [As Galaxaura oblongata (Kjellman) Borgs.] Canaries (229) [As Galaxaura oblongata Lamouroux] Canaries (493). [As Galaxaura adriatica Zanardini] Canaries (177). 'cotes occidentales d'Afrique et aux Canaries' (184). [As Galaxaura cylindrica (Solander [or Ellis & Solander]) Lamouroux] Canaries (13; 68; 71; 191; 227; 277; 390; 436; 439; 535). Cape Verde Islands (535). Principe (251; 265; 436). Sao Tome (251; 263; 264; 265; 436; 535). 'pantropical' (535). Note. The Sao Tome records in 263 and 264 include presence on Ins. Rolas, the nearby islet. [As Galaxaura cylindrica Lamouroux] Canaries (38). Cape Verde Islands (38; 145; 260). [As Galaxaura cylindrica (Solander in Ellis & Solander) Kjellman] Canaries (108). Cape Verde Islands (100; 183). [As Galaxaura fragilis (Lamouroux) Decaisne] Cape Verde Islands (41). [As Galaxaura fragilis Lamouroux] Cape Verde Islands (38; 261; 262; 408; 528; 551). Gabon (250). 'Mers chaudes en general' (38). 'Tropical seas' (410). [As Galaxaura fragilis Decaisne] Cape Verde Islands (41; 42). [As Galaxaura fragilis (Lamarck) Decaisne] Gabon (250). [As Galaxaura fragilis Decaisne var.] Canaries (439). Note. Piccone (439) gave no information regarding a varietal name. Levring (375) reported this species to be well represented in the Salvage Islands, but not to be present elsewhere in the Madeira group. Piccone (439) noted this taxon as 'G. oblongata (Sol. et Ell.) cf . G. lapidescens (Sol. et Ell.),' based directly on the record in Lowe (381). There has been long debate on the possible conspecificity of G. oblongata and G. cylindrica (q.v.). Steentoft (535: 121) appeared from her comments to be of the opinion that differences between these two taxa were that G. oblongata is coarser and has more divergent branching. She further indicated that 'G. cylindrica is perhaps the sexual phase corresponding to the asexual phase known as G. lapidescens.' We have not followed this - see the details below. Cordeiro-Marino (108: 32) indicated general acceptance of the close relationships between G. oblongata and G. cylindrica, not only because of the merely narrower and thinner segments of the latter, but also in a manifest lack of data on tetrasporophytic plants, only the sexual phases being known. Following data variously presented in Howe (1918) and B0rgesen (68; 80), Steentoft (535) considered the corresponding tetrasporophyte of the sexual phase G. oblongata to be the taxon known as G. comans Kjellman (q.v.), not known from the eastern Atlantic. Cordeiro-Marino (108) also repeated Howe's (1918) comments that G. lapidescens (Ellis & Solander) Lamouroux (q.v.) and G. comans were possibly the tetasporophytic phases, respectively of G. cylindrica and G. oblongata. Thus, despite accepting (see above) close relationship between the latter two taxa, Cordeiro-Marino still maintained them as separate, in common with many other phycologists, including Howe and B0rgesen, mentioned above. Howe and B0rgesen both considered the demarcation between the entity [G. comans 204 J. H. PRICE, D. M. JOHN & G. W. LAWSON and G. cylindrical and the entity [G. comans and G. lapidescens] to be arbitrary. None of this accords well with the more recent opinions of Papenfuss & Chiang (435) and Papenfuss et al. (436), who submerged G. cylindrica in the synonymy of G. oblongata and believed G. lapidescens to be a good species, with G. comans as its synonym. We have adopted the views of these more recent workers, especially as outlined in 436, the most recent paper. On other aspects of the differences/distinctness of G. cylindrica and G. oblongata, see the notes presented at G. cylindrica. It is worth mentioning that B0rgesen also variously stated (68) that Lamouroux's specimen of G. oblongata, which he (B0rgesen) examined, is a form of G. obtusata (Ellis & Solander) Lamouroux, and that (80) it is impossible to separate the Red Sea species G. schimperi Decaisne from G. oblongata. Chou (104) considered that G. oblongata perhaps represented the sexual phase of more than one species, whilst Itono (277: 14, 16) indicated G. fastigata, G. pilifera, G. cylindrica, and G. oblongata closely to resemble each other in their external features, leading most Japanese phycologists to regard the latter three species as G. fastigiata. Despite his arbitrariness in dividing the four species on the basis of ecologically- or physiologically-affected characteristics, however, Itono felt they should be maintained pending further work. Recent work by Magruder (383) has shown that Hawaiian Galaxaura oblongata has a triphasic life history, with conspicuous gametophytes and small inconspicuous filamentous tetrasporophytes. In this it accords with data established for two other genera in the Galaxauraceae [= Chaetangiaceae], Scinaia and Pseudogloiophloea, even having much the same tetrasporophyte form. Most other known species of Galaxaura and Actinotrichia have conspicuous macroscopic tetrasporophytes, with only a few Galaxaura spp. (where the tetrasporophyte is not with absolute certainty known) having possibly small filamentous tetrasporophytes. Magruder (383: 407) commented that this presence of inconspicuous tetrasporophytes was certainly responsible for difficulties in aligning putative tetrasporophyte 'species' with the correspond- ing gametophyte 'species' in Galaxaura. Details of both spermatangial conceptacle development and carpogonial branch development are very similar in G. oblongata, G. obtusata, G. diesingiana, and G. marginata, according to Magruder (op. cit.). Galaxaura obtusata (Ellis & Solander) Lamouroux Canaries (68; 71; 97; 191; 227; 247; 277; 372; 435; 436; 598). Cape Verde Islands (598). Gabon (250; 294; 350; 586). Salvage Islands (598). Sierra Leone (30; 350; 586). 'eastern Atlantic (Canary Islands) westward to East Africa' (435). 'The tropical Ocean, in all longitudes' (256). 'Tropical Africa (N. Gambia - Congo river)' (598). Tropical and subtropical oceans' (257). 'widespread in warm temperate and tropical seas' (350; 586). [As Galaxaura obtusata J. Agardh] Angola (41; 42). 'Warm Atlantic' (41; 42). [As Galaxaura umbellata Lamouroux] Canaries (37; 41; 42; 408). Cape Verde Islands (408; 528; 551). 'Warm Atlantic' (410). [As Galaxaura umbellata (Esper) Lamouroux] Canaries (141A). Salvage Islands (38B; 215). [As Galaxaura umbellata Montagne] Canaries (38). Cape Verde Islands (38; 551). Note. Authorities quoted as '(Solander) Lamouroux' (e.g. in Itono (277: 7)) have been treated throughout the species entry as though the form in the present entry heading had been employed. B0rgesen (68) examined Lamouroux's specimens labelled as G. umbellata and concluded that they represent G. obtusata. Papenfuss et al. (436: 418-421) and Seagrief (570: 30) concluded equally that this was correct, at least pro parte for G. umbellata. Askenasy's (37; 38) records as G. umbellata may really represent G. rugosa (Ellis & Solander) Lamouroux. In 1897, Askenasy (38: 165), although presenting his entry under the heading 'G. umbellata Mont, non Lamour.', added a terminal comment in French, which translated as: RHODOPHYTA (PLORIDEAE) OF TROPICAL AFRICA 205 This species, also found on the Canaries and named by Montagne G. umbellata, seems to me to be different from the species so named by Espen [sic!], Lamouroux, Kiitzing and J. Ag. (Epicrisis). It is nearer to G. rugosa (Soland.) Kiitz.' See the entry for the latter species, as well as the note presented at G. decaisnei J. Agardh, for other comment of relevance to the present taxon. Itono (278: 8) regarded G. obtusata and G. umbellata as very similar and not separable into two species. All these forms were suggested as related to the sexual phase of Galaxaura robusta, which is not reported for coasts of the list area. Galaxaura rugosa (Ellis & Solander) Lamouroux Ascension (37; 535). Cameroun (288; 350; 454; 535; 586). Canaries (37; 128A; 232B; 306B; 436; 598; 604; 605). Cape Verde Islands (37; 38B; 535; 598). Cote d'lvoire (287; 288; 350; 586). Gabon (586). Ghana (586; BM (unpublished)). Liberia (287). Principe (436; 586). Salvage Islands (38B; 556A; 598). Sao Tome (251, pro parte; 263; 265; 288; 350; 436; 535; 586). 'in oceano calidiore atlantico' (27). 'Pantropical' (535). Tropical Africa (N. Gambia -Congo river)' (598). 'warmeren Atlantischen Ocean (Kanaren, Antillen)' (37). 'widespread in warm temperate and tropical seas' (350; 586). [As Galaxaura rugosa Lamouroux] Cape Verde Islands (38; 145; 259; 260). Sao Tome [including Rolas] (263; 264). 'Ocean Atlantique tropical' (38). 'Warm Atlantic' (410). [As Galaxaura rugosa (Solander) Lamouroux] Sao Tome (265 pro parte). 'in warmeren atlantischen Ocean' (503). [As Galaxaura rugosa Solander] Canaries (439). Sao Tome (251 pro parte, 265 pro parte). [As Galaxaura annulata Lamouroux] Cape Verde Islands (38; 408; 528). [As Galaxaura coarctata Kjellman] Canaries (139; 313; 436; 538). [As Galaxaura Decaisnei 3. Agardh] Canaries (547). 'in oceano Atlantico calidiore' (27; 131). [As Galaxaura elongata J. Agardh] Canaries (435). Gabon (294; 350). Sao Tome (350). 'eastern Atlantic (Canary Islands) westward to East Africa' (435). 'widespread in warm temperate seas and pantropical' (350). [As Galaxaura squalida Kjellman] Canaries (13; 38B; 38C; 68; 97; 226; 227; 302; 304; 351; 379; 414; 535). Cape Verde Islands (38B; 38C; 100; 183). Salvage Islands (38B; 556 A). Sao Tome (535). 'pantropical' (535). 206 J. H. PRICE, D. M. JOHN & G. W. LAWSON Note. The unpublished record from Ghana is based on a specimen (in BM) collected by V. J. Foote in 1946. Records that appear above as pro parte from Sao Tome require reference to the other taxon involved; in all cases, this appears to be Galaxaura subfruticulosa Chou in W. Taylor (q.v.) - see additional notes below. The report from the Salvage Islands (38B; 556A) initially (Prud'homme van Reine, 15 Nov. 1982, in litt. to JHP; 556A) indicated the records as new for that island group. The inclusion under this species entry of the records established as Galaxaura annulata Lamouroux for the Cape Verde Islands is based on the synonymy deduced and presented by Papenfuss et al. (436); the nature and accuracy of determination of the actual material involved have not been checked . The correctness of placement here of the records established as G. Decaisnei J. Agardh may not be firmly authenticated, but has been effected on the basis of the area concerned (Canaries; warm Atlantic ocean); see the notes to the species entry for G. decaisnei. Although the type of G. decaisnei (Barbados) proved (cf. 436) to be Galaxaura obtusata, Vickers's (547) material from the Canaries emerged on examination by B0rgesen (68) as G. squalida [= G. elongata, now synonymised with G. rugosa]. For a contrasting opinion on G. squalida status, see Audiffred (38C: 175) and the note to G. squalida here. For a full recent statement of the synonymy, similarities, and previously employed application of names to all or parts of this taxon in different geographical areas, see Papenfuss et al. (436). See also Cribb (113: 26-27). Steentoft (535: 123) suggested that G. rugosa and G. subfruticulosa Chou have hitherto been confused amongst Sao Tome material, hence the 'pro parte' recording of certain previous records, mentioned above. The significance of such confusion depends on more firm understanding than currently available of morphological relationships and life-history connections between these taxa, and between both and G. subverticillata Kjellman. See the detailed notes to G. subfruticulosa Chou and G. obtusata (Ellis & Solander) Lamouroux; Papenfuss et al. (436: 425) suggested that the tetrasporophyte of G. rugosa is probably that referred to as G. subverticillata Kjellman, but further work is required to establish the connection. G. subverticillata is not recorded for the eastern Atlantic. It is possible (cf. Itono 277: 8-9) that the Pacific Galaxaura pacifica and the Atlantic G. rugosa are different forms of the same species; both have free assimilatory filaments and similar overall morphologies, although it seems that the free filaments may persist in older parts of G. rugosa, as opposed to being early deciduous in G. pacifica. Galaxaura squalida Kjellman See Galaxaura rugosa (Ellis & Solander) Lamouroux and G. elongata J. Agardh. Note. According to Chou (104), the type of G. squalida (from the Virgin Islands) is in Herb. Areschoug under the name G. rugosa. According to Audiffred (38C: 175), G. squalida is the sexual phase of G. flagelliformis Kjellman (see G. lapidescens (Ellis & Solander) Lamouroux). Galaxaura stupicaula ['stupocaula'] Kjellman See Galaxaura marginata (Ellis & Solander) Lamouroux. Note. According to Chou (103), the form of the epithet should be stupicaulis, to agree with the generic substantive form. The sexual phase involved could be G. veprecula Kjellman [= G. marginata, q.v.]. Papenfuss & Chiang (435: 304) rendered the epithet as [G.] stupocaula and considered it to be representa- tive of Galaxaura tenera Kjellman [? = G. marginata, q.v.]; these data are repeated and elaborated in Papenfuss et al. (436: 411-415). Galaxaura subfruticulosa Chou in W. Taylor Sao Tome (350; 535; 586). 'in tropical parts of the Atlantic and Pacific Oceans' (350; 586). Tropical Africa (N. Gambia- Congo river)' (598). [As Galaxaura rugosa (Ellis et Solander) Lamouroux, pro parte] Sao Tome (251; 263; 264; 265). Note. See the detailed note to Galaxaura rugosa (Ellis & Solander) Lamouroux. There has been much debate about the status and relationships of this present taxon. Steentoft (535: 124) observed that G. subfruticulosa Chou is closely related to G. subverticillata Kjellman, only the latter having been previously recognised in the Atlantic (and then only the western Atlantic). Itono (278: 5-6) repeated the comment that G. subfruticulosa is most closely related to G. subverticillata; he concluded, after examining the supposed differences between the taxa, that 'it is not inconceivable that G. subfruticulosa and G. subverticillata will be found to be conspecific.' According to Papenfuss et al. (436: 409), G. fruticulosa Chou in W. Taylor comes close to G. lapidescens (q.v.). G. fruticulosa was established on the basis of material from Revillagigedo Is., Pacific Mexico, and said by Chou to be closely similar to G. ramulosa and G. tomentosa. G. lapidescens and G. fruticulosa may well eventually be thought conspecific. The significance of this is that Chou renamed G. fruticulosa RHODOPHYTA (pLORIDEAE) OF TROPICAL AFRICA 207 Kjellman, calling it G. subfruticulosa and designating a new type (from Clarion Is., Revillagigedo). Papenfuss et al. (436: 409) went on 'Whether Chou's species and the one of Kjellman actually are representative of the same species remains to be determined.' Later in their 1982 publication (436), Papenfuss et al. (pp. 424-5) commented that Galaxaura subverticillata Kjellman, G. fruticulosa Kjellman, and G. subfruticulosa Chou in W. Taylor all have swollen basal cells to both long and short assimilatory filaments. The conspecificity of G. fruticulosa Kjellman, G. fruticulosa (Ellis & Solander) Lamouroux, 1816 (from Bahamas), and G. subfruticulosa Chou in W. Taylor (Revillagigedo) cannot really be established without detailed examination of the types. Ellis & Solander's Corallina fruticulosa (1786) illustration shows that G. subverticillata is not of the same taxon; the branches of the latter do not 'grow smaller towards the ends.' The verticillate arrangement of assimilatory filaments in East African G. subverticillata was not mentioned by their description of C. fruticulosa, nor is it clear in Kjellman's (313) treatment of his G. fruticulosa, although it is so in his description of G. subverticillata. Papenfuss et al. (436: 425) went on to state: 'Extended assimilatory filaments of Galaxaura subfruticulosa Chou are evenly distributed over the entire surface of the thallus. . . . and slightly verticillate only near the tips. This taxon is distinct from G. subverticillata Kjellman. On this basis Chou (1945, p. 42) placed G. subfruticulosa very close to G. ramulosa and G. tomentosa, taxa which we have reduced to synonymy of G. lapidescens. ' Confusion between G. subverticillata and G. lapidescens may have partially arisen because the verticillate arrangement in the former is seen clearly only in young branches, the hairs disappearing with ageing of the branch. G. subverticillata Kjellman is probably the sporophyte of G. rugosa (q.v.); 'Until this can be established with certainty, however, it seems prudent to retain G. subverticillata as an autonomous species' (Papenfuss et al. 436: 425). Galaxaura subverticillata Kjellman See the terminal notes to the entry for Galaxaura subfruticulosa Chou in W. Taylor. Galaxaura tenera Kjellman See the entries for Galaxaura marginata (Ellis & Solander) Lamouroux [records] and for both G. marginata and G. veprecula Kjellman [notes]. Galaxaura umbellata Lamouroux See Galaxaura obtusata (Ellis & Solander) Lamouroux [records; notes] and Galaxaura rugosa (Ellis & Solander) Lamouroux [notes]. Note. The rationale for first attribution of valid publication to Lamouroux (331: 262) is presented in Papenfuss et al. (436; 419, footnote 11). The combination has sometimes been attributed to (a) Montagne; (b) (Esper) Lamouroux; (c) (Esper) J. Agardh, as authorities. Galaxaura ventricosa Kjellman See Galaxaura marginata (Ellis & Solander) Lamouroux. Note. Papenfuss et al. (436: 411, 412) referred to the type of G. ventricosa Kjellman, from [near the mouth of] the Gabon River, West Africa; they placed the taxon in synonymy with G. marginata (Ellis & Solander) Lamouroux (q.v.). Chou (104) had earlier indicated it to be impossible to distinguish between G. ventricosa and G. veprecula on the basis of anatomy; both these taxa are in any case now accepted as lying within the synonymy of G. marginata (Ellis & Solander) Lamouroux. Galaxaura veprecula Kjellman See Galaxaura marginata (Ellis & Solander) Lamouroux. Note. Jaasund (19776: 417) stated that the conspecificity of Galaxaura veprecula (from Madagascar) with Galaxaura tenera Kjellman (type locality: Mombasa) is still an open question. More recent opinion (e.g. Papenfuss et al., 436: 411 et seq.) does not, however, confirm this, since many workers accept both these taxa as lying within the synonymy of G. marginata (Ellis & Solander) Lamouroux; we concur in the latter view. See also the notes to G. ventricosa Kjellman. Galaxaura spp. Canaries (3; 71; 89; 117; 118; 214; 229; 253A; 301; 436; 567). Cape Verde Islands (191; 500). Ghana (491). Senegal (59; 282; 611). 208 J. H. PRICE, D. M. JOHN & G. W. LAWSON 'Gulf of Guinea' (611). 'West Africa' (290). Note. Bornet (89) stated he had '3 especes' of Galaxaura. Ganonema farinosa (Lamouroux) Fan & Wang See Liagora farinosa Lamouroux, and Abbott (1). Ganonema pinnatiramosa (Yamada) Fan & Wang in Fan et al. See Liagora farinosa Lamouroux, and Abbott (1). Gastroclonium clavatum (Roth) Ardissone Canaries (232B; 253; 583; 598; 604; 605). [As Gastroclonium ovatum (Hudson) Papenfuss] Canaries (226: 227). Note. Haroun Tabraue et al. (253: 273-4) recorded the taxon as 'Nueva cita para Canarias.' The first citation was of epiphytes on Corallina elongata, later (232B) extended to those on Codium taylorii. Haroun Tabraue et al. were also (loc. cit.) responsible for transferring the previous records as Gastroclonium ovatum (Hudson) Papenfuss (q.v.) in 226 and 227 to the present taxon; they stated 'La cita de Gastroclonium ovatum. . . . para Pta. Pechigueras, Lanzarote. . . . Gil-Rodriguez & Afonso-Carrillo (1980), tras la revision del pliego, debe ser eliminada del "Catalogo de las Algas marinas bentonicas para el Archipielago Canario" y ser incluido como Gastroclonium clavatum (Roth) Ardisson.' G. clavatum overall distribution was cited as "East Atlantic (Cadiz & Canaries) and Mediterranean.' See also the entry for Gastroclonium ovatum (Hudson) Papenfuss - when originally recording their material as that taxon from there (Canarias), Gil-Rodriguez & Afonso-Carrillo (226: 64) stated: 'Esta especie, cuya area de distribu- tion se extiende desde Inglaterra a Mauritania, no habia sido senalada precedentemente para Canarias.' Gastroclonium kaliforme (Goodenough & Woodward) Ardissone See Chylocladia verticillata (Lightfoot) Eliding. Gastroclonium ovatum (Hudson) Papenfuss Canaries (128A; 226; 598). Mauritanie (33; 222; 226; 273; 349; 546). 'Atlantique (de 1'Angleterre a la Mauritanie)' (33). 'Atlantique, depuis les cotes anglaises jusqu'en Mauritanie' (222). 'British Isles to Mauritania; Canary Isles' (273). 'desde Inglaterra a Mauritania' (226). 'desde las costas de Inglaterra hasta Mauritania' (546). [As Chylocladia ovalis (Hudson) Hooker] 'D'Angleterre aux Canaries' (89). [As Lomentaria ovalis (Hudson) J. Agardh] 'In Oceano atlantico calidiore' (27). Note. See also the entry for Gastroclonium clavatum (Roth) Ardissone. The Mauritanie record in Lawson & John (349) is based only on data in 33 and 222. It is probable that the older Canaries references (in 89) and the more recent secondary statement in Irvine (273) should both be referred (as with the bases in 226 and 227 for the more recent statement) to Gastroclonium clavatum (Roth) Ardissone. Haroun Tabraue et al. (253) established that point. Gastroclonium reflexum (Chauvin) Kiitzing Canaries (33; 227; 273). 'Atlantique (de 1'Angleterre aux Canaries)' (33). [As Lomentaria reflexa Chauvin] Canaries (70; 191). 'English coast southwards to the Canary Islands' (70). [As Lomentaria reflexa (Chauvin) J. Agardh] Canaries (441; 444). [As Lomentaria pygmaea (Lamouroux) Gaillon] Canaries (401). [As Lomentaria pygmaea Gaillon] Canaries (44). RHODOPHYTA (FLORIDEAE) OF TROPICAL AFRICA 209 Note. Ardre (33: 141) commented on distinguishing characters for Gastrodonium, indicating that her material closely approached the Coeloseira of Hollenberg, based on the constant occurrence of polysporangia in certain species; since this latter distinction lacked supporting characters, she preferred to retain the material in Gastrodonium. There is slender basis for many of the above records - Piccone (441), who recorded only a single young plant, formed the direct and only basis for his own repeated records in (444), as well as for those published in B0rgesen (70) and in Gil-Rodriguez & Afonso-Carrillo (227). This taxon was frequently referred to as Chylodadia reflexa [auctorum] and the situation as regards the synonymy is complex (cf. Irvine, 273: 84-85). The record in Montagne (401: 156-7), and therefore also that in Benitez (44), is attributed here on general nomenclatural grounds. Montagne referred in his synonymy only to Gigartina pygmaea Lamouroux and Lomentarla pygmaea of Gaillon and of Duby. His additional comments included: 'Among other algae , to which it adheres, this species, if red, is doubtfully of the animal kingdom' and he indicated that the species had a great affinity with Lomentaria pusilla De Notaris [in litt.] and with Chondria nana C. Agardh, all three being very probably only forms of one and the same species. De Toni (132: 566) recognized Gigartina pygmaea Lamouroux (with doubt) and Lomentaria pusilla Kiitzing in the synonymy of G. reflexum (Chauvin) Kiitzing; he nowhere in his works referred to Lomentaria pygmaea Gaillon, to Lomentaria pusilla De Notaris, or to Chondria nana C. Agardh, although he did place in that same synonymy Lomentaria exigua De Notaris. L. Irvine (273: 85) indicated G. reflexum to be 'closely related to G. davatum (Roth) Ardiss. . . . but without the reflexed habit.' (•d id id la acerosa (Forsskal) J. Feldmann & Hamel Gabon (294; 350; 586). Sao Tome (350; 535; 586). Sierra Leone (30; 350; 586). 'in warm temperate and tropical seas' (350; 586). 'most warm seas' (81). 'parait repandue dans toutes les mers tropicales (Atlantique . . .)' (194). Tropical Africa (N. Gambia - Congo river)' (598). 'tropical and subtropical seas' (614). [As Gelidium claviferum Kiitzing or its 'forma'] Sao Tome (251; 263; 264; 265). [As Gelidiopsis rigida (Vahl) Weber-van Bosse] 'Seems to occur in all warm seas' (64). [As Gelidium rigidum (Vahl) Greville] 'in oceano calidiori atlantico' (27). 'Throughout warm seas' (410). Note. Problems occasionally attach to accurate determination of material of this taxon from the area (see Steentoft, 535). B0rgesen (73: 5) earlier (but not acceptedly) recombined this taxon in the genus Echinocaulon. Of material from the Admiralty Islands (not directly relevant here), Hemsley (260) commented that This is probably a variety of the almost cosmopolitan Gelidium corneum, Lamour.' [He was writing under the heading of Gelidium rigidum Greville]. For a general statement of synonymy in Gelidiella acerosa see Seagrief (570: 30). Gelidiella pannosa (J. Feldmann) J. Feldmann & Hamel 'sporadically from the French Cote Basque south to Morocco . . . and Senegal' (614). [As Gelidiella tenuissima J. Feldmann & Hamel] Canaries (38B;556A). Cape Verde Islands (38B). Mauritanie (38B; 122; 556). Salvage Islands (38B; 556; 556A). Senegal (38B; 122; 556). 'From Biarritz and Portugal southward to Cape Verde.' (375). 'Portugal hasta Cabo Verde (Ardre 1970)' (518). [As Gelidiella tenuissima (Thuret) Feldmann & Hamel] Canaries (38C). Cape Verde Islands (38C). Mauritanie (38C; 121). Salvage Islands (38C). 210 J. H. PRICE, D. M. JOHN & G. W. LAWSON Senegal (38C). [As Gelidiella pannosa (Bornet) J. Feldmann & Hamel] Canaries (598). Salvage Islands (598). 'Atlantique (Biarritz, Portugal, jusqu'au Cap Vert)' (33). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). Note. Fan (1961: 340, note 4) has summarised the nomenclatural and taxonomic history of this taxon in demonstrating that authority and epithet citation should be as stated in the present entry heading. Nevertheless, we have considerable sympathy with the change of epithet to tenuissima effected by J. Feldmann & Hamel (195: 102 et seq. [Revue algol.] or 226 et seq. [collated reprint]) for this taxon in Gelidiella because of the potential confusion arising from the complexities of both taxonomy and nomenclature in the difficult Gelidiales; these complexities are well expressed in the recent work by Maggs & Guiry (614). It is not now possible to reject specific epithets on the grounds of confusion and it is clear that J. Feldmann's combination in Echinocaulon (?) of pannosum can be treated directly as the basionym for Gelidiella pannosa, given the homonymy involved in the earlier treatments of Gelidium pannosum Grunow (now Gelidiopsis pannosa (Grunow) Schmitz) and Gelidium pannosum sensu Bornet, the basis for much of the J. Feldmann & Hamel (194: 195) reasoning. Even in the case of the perhaps more desirable Gelidiella tenuissima, there has been frequent mis-citation of authorities; Thuret's herbarium name was Gelidium tenuissimum and, since Feldmann & Hamel were not taking up the whole combination in this MS name, there is no requirement to cite Thuret as authority. Gelidiella ramellosa (Kiitzing) J. Feldmann & Hamel Salvage Islands (38B; 231; 375; 598). Note. Levring (375) considered this species well-separated from Gelidiella tenuissima J. Feldmann & Hamel (= Gelidiella pannosa (J. Feldmann) J. Feldmann & Hamel, q.v.). G. ramellosa showed good representation in the Salvage Islands, although Levring (loc. cit.) presented its distribution only as the Salvage Islands and Tunis. See also details presented under Gelidiella tinerfensis J. Seoane-Camba. Gelidiella tenuissima J. Feldmann & Hamel See Gelidiella pannosa (J. Feldmann) J. Feldmann & Hamel. Gelidiella tinerfensis J. Seoane-Camba Canaries (227; 518; 574; 598; 614). Note. Seoane-Camba (518; 574) presented, especially in (518), comments on the distinctions between G. tinerfensis and its nearest relatives in Gelidiella - G. pannosa, G. ramellosa (q.v.), G. lubrica, and G. trinitatensis . Gelidiella sp. Canaries (38C). Salvage Islands (38B). Note. Reported as only sterile material. Audiffred (38C: 175) presented some description of the material from the Canaries. Gelidiocolax hemisphaerica Gardner See Gelidiocolax microsphaerica Gardner. Gelidiocolax microsphaerica Gardner Senegal (1A; 575). [As Gelidiocolax hemisphaerica Gardner] Senegal (122). 'Subtropical Africa [Senegal (N. Gambia), Mauritania, Former W. Sahara]' (598). Note. Fan & Papenfuss (575: 33) commented that: 'Dangeard in 1952 reported Gelidiocolax micro- sphaerica (in error as G. hemisphaerica) from Dakar.' Dangeard (122) did not explain his mis-use of the specific epithet as 'hemisphaerica'' , but it may possibly relate to his believing it to be a more accurate reflection of thallus form. Even Abbott & Hollenberg (1A: 342) opened their published description of Gardner's taxon with the words: Thallus a smooth, more or less hemispherical mound'. Fan & Papenfuss (575) considered various aspects of the relationships between species currently placed in this genus. In all probability, the Gelidiocolax sp. reported from Senegal and Mauritanie by Dangeard (121) in 1951 is also representative of G. microsphaerica; see the following entry. RHODOPHYTA (PLORIDEAE) OF TROPICAL AFRICA 211 Gelidiocolax sp. Mauritania (121). Senegal (121). Note. Found growing on Gelidiumfoliosum P. Dangeard, particularly on the 'stolons'; the Gelidiocolax sp. concerned is almost certainly that subsequently (122) referred to by Dangeard as G. hemisphaerica Gardner [= G. microsphaerica Gardner] (q.v.). Gelidiopsis gracilis (Kiitzing) Vickers Senegal (50; 55; 59). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). Note. Bodard (50) commented that this species was for long confused with Gelidium crinale (q.v.) in their collections from Senegal, since the two often occur intermixed. He continued: 'c'est a notre connaissance la premiere fois que ce Gelidiopsis est signale en Afrique; il etait uniquement connu aux Antilles, avec Gelidiopsis variabilis [q.v.] dont nous allons parler.' Later, Bodard & Mollion (59: 198) stated of this species: 'la position systematique est douteuse, il peut etre considere comme une algue tropicale a faible dispersion geographique.' The present species differs from Gelidiopsis planicaulis (q.v.) in that the latter lacks the cylindrical axis which is present in at least the distal parts of the former. Gelidiopsis intricata (C. Agardh) Vickers Ascension (474; 475). Canaries (38D; 598). Cape Verde Islands (38D; 191, 598). Salvage Islands (38D; 598). [As Gelidiopsis intricata (Kiitzing) Vickers] Canaries (38B; 70; 482; 605). Cape Verde Islands (38B; 100; 183). Salvage Islands (38B). 'does not extend (southwards) into the Gulf of Guinea' (487). [As Gelidiopsis intricata (Agardh) Schmitz] Senegal (542). [As Gelidium intricatum Kiitzing] Canaries (68; 118; 486; 489; 495). Note. It has been questioned (Steentoft, 535: 125-126) that this and G. variabilis (J. Agardh) Schmitz are really separate species. See the note to the latter for fuller explanation. Material recorded from Sao Tome by Tandy in Exell (539: 386) and from the Cape Verde Islands by J. Feldmann (183) and Chevalier (100), as G. intricata, was sterile but, according to Steentoft (loc. cit.), was actually referable to G. variabilis (q.v.). It is possible that other records here (as, e.g., 38B (Salvage Islands); 70; 100; 183; 482) should (although sterile/small specimens were again involved) be referred to the Gelidiopsis variabilis form, whether or not the latter and G. intricata are finally considered to be conspecific. B0rgesen (70) merely repeated the record from the Canaries given in Reinbold (482: 22), without additional information. Reinbold's material was also sterile. Audiffred & Weisscher (38B: 29) could not be sure of the determination of their Salvage Islands material, since their specimen was very small. Trochain's (542: 109) material from Senegal was merely drift - it, too, was referred by Steentoft (535: 126) to Gelidiopsis variabilis (q.v.). Sterile specimens may also have caused confusion in routine determination for areas of Atlantic Africa - G. intricata may have been misidentified with Wurdemannia miniata or Caulacanthus ustulatus; see the entries for these latter species. Rodriguez (486) misrendered the specific epithet as intrincatum in recording Gelidium intricatum from the Canaries. Schiffner (495) merely referred to B0rgesen's earlier (68) statements for Tenerife. Gelidiopsis planicaulis (W. Taylor) W. Taylor ?Bioko(346;350;586). Cameroun (346; 350; 586). Canaries (556A). Cape Verde Islands (598). Gabon (294; 350; 586). Liberia (129; 350; 586). Sierra Leone (295; 350; 586). 'in tropical parts of the Atlantic Ocean' (350; 586). 212 J. H. PRICE, D. M. JOHN & G. W. LAWSON Tropical Africa (N. Gambia -Congo river)' (598). Note. A variable species with regard to flattening of branches; W. Taylor suggested (pers. comm.) that West African material was very similar to that from the western Atlantic, although flattening in the former was more localised (294; 350; 586). Gelidiopsis rigida auctorum [usually (Vahl) Greville or (Vahl) Weber-van Bosse] See Gelidiella acerosa (Forsskal) J. Feldmann & Hamel. Gelidiopsis variabilis (J. Agardh) Schmitz Angola (352). Bioko (346; 350; 586). Cameroun (122; 288; 350; 586). Cape Verde Islands (598). Cote d'lvoire (288; 350; 586). Gambia (296; 350; 586). Ghana (288; 290; 299; 300; 338; 350; 376; 491; 537; 586). Guinee/Guinea-Bissau (350; ? 529; 586). Liberia (129; 350; 586). Mauritanie (349). Sao Tome (288; 350; 586). Sierre Leone (295; 350; 586). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). Tropical Africa (N. Gambia -Congo river)' (598). 'widespread in warm temperate and tropical seas' (350; 586). [As Gelidiopsis variabilis (Greville) Schmitz] Annobon(456;457;535). Cameroun (500; 535). Canaries (7535). Ghana (535). Guinee (535). Mauritanie (192). Sao Tome (93; 535). Senegal (50; 121; 122; 529). [As Gelidiopsis variabilis (Greville ex J. Agardh) Schmitz] Ghana (297). [As Gelidiopsis variabilis Schmitz] Cameroun (454). [As Gelidiopsis intricata (C. Agardh) Vickers] Sao Tome (539). Sierra Leone (295). [As Gelidiopsis intricata (Agardh) Schmitz] Senegal (542). [As Gelidiopsis intricata (Kiitzing) Vickers] Cape Verde Islands (100; 183). Note. There is doubt as to the distinctions, or indeed the reality of separateness, between G. variabilis and Gelidiopsis intricata (q.v.). Steentoft (535: 125-126) firmly expressed this in the following terms: 'Larger tuft-forming plants seem to be placed in the former species [G. variabilis], low growing turf- forming ones in the latter [G. intricata].' In expanding this further, she commented on the Tandy in Exell (Sao Tome) material (539) and the Cape Verde Islands records (J. Feldmann, 183; Chevalier, 100). Other records that could possibly concern G. variabilis are discussed in the note to Gelidiopsis intricata (q.v.). The overall picture seems to be one of difficulty in distinguishing validly between ends of morphological/ anatomical clines. Steentoft's opinions were embodied in the following extract from 535: 125-126: 'Both G. variabilis and G. intricata. . . . are probably in need of critical re-examination. The latter name is probably invalid, since it is based on Sphaerococcus intricatus C. Ag., the "cotype" of which [J.] Feldmann has examined and reported as a Gelidium species, since intercellular rhizines are present. ... it seems to the present writer that there is some doubt as RHODOPHYTA (pLORIDEAE) OF TROPICAL AFRICA 213 to the distinctness of [these species]. . . . However a much greater acquaintance with the American forms of the so-called G. intricata is required before any conclusions can be reached.' Sterile material of G. variabilis may, for Atlantic Africa, have at times also been confused with Wurdemannia miniata (q.v.) or Caulacanthus ustulatus (q.v.). The reports from Senegal in Bodard (50) and Sourie (529) have similarly an appreciable degree of uncertainty about them. Sourie (529: 306) himself expressed doubt about his records of the species from Guinee/Guinea-Bissau, and his citation from Senegal stated: 'Gelidiopsis sp. (G. variabilis (Greville) Schmitz?).' Bodard (50: 83-84) indicated that G. variabilis from Senegal was much more robust (but much rarer) then Gelidlopsis gracilis (Kutzing) Vickers; his single certain specimen of Gelidiopsis variabilis from Pointe de Sarene, Senegal (February), proved identical with a plant collected by Sourie (529) at Cap Rosso. At first sight, G. variabilis could easily be confused with Gracilaria augustissima (sic! - for angustissima) (Bodard 50: 83-84), but the anatomical structure proved to be of Gelidiopsis, rather close to G. gracilis (Kutzing) Vickers. Gelidiopsis sp. Cameroun (337; 344; 393; 394). Ghana (42 A; 92; 220; 338; 344; 393; 394). Guinea-Bissau/Guinee (529). Guinee (269; 344; 384; 393; 394; 529). Nigeria (344; 393; 394; 415). Senegal (393; 394; 529). Note. For a possible attribution of the Senegal record in Sourie (529), see the entry for Gelidiopsis variabilis (J. Agardh) Schmitz. Gelidium Bornet's earlier categorization of Gelidium as 'genre diabolique' is hardly a matter for dispute; this applies whether the concern is nomenclatural, taxonomic, morphological, or ecological (Dixon & De Valera, 1961; Dixon & Irvine, 172). This complicated form- variation, often seasonally different in individual plants and populations, is therefore also reflected in the excessive names scattered throughout the literature, and probably lies behind the large number of species reported from South Africa sensu stricto and not from apparently similar areas to the north and within the check-list area. Cases in point are Gelidium amansii (Lamouroux) Lamouroux, G. arenarium Kylin, G. caespitosum Kylin, and G. pristoides (Turner) Kutzing. On the other hand, there are numbers of species names applied within the check-list area that have never been employed outside it, to north or south. Overall rationalization of the taxonomy and nomenclature on a pan- Atlantic basis is required, but the difficulties and time-consumption make the prospect daunting. Furthermore, approaches to species recognition via morphological characteristics alone are clearly insufficient if a meaningful result is to be obtained; biochemical and genetic levels will require exploration on a geographically widespread basis to rationalize the different patterns of understanding of the taxa, and to eliminate contrasting and purely local traditions of naming. No such vast and daunting project has been undertaken here and consequently no general reallocation of records depending on a totally reorganised system of name-allocation is presented. Transfers of individual records or groups of records have been made on restricted bases, where there has been good reason to believe that action necessary. In the absence of any such individual information, the publishing authors' names have been maintained, if for no other reason than to draw attention to current anomaly requiring further work. Thus, in the absence of real understanding of limits of taxa considered as species level across virtually the whole genus, the recording by name (and therefore the reality) within the present list area of the following 'species' must be considered at best speculative to doubtful, at worst spurious: Gelidium arbuscula asperum attenuatum canariensis [cartilagineum var. canariensis] 214 J. H. PRICE, D. M. JOHN & G. W. LAWSON cartilagineum clavatum corneum crinale elminense flaccidum foliosum latifolium melanoideum microdon micropterum pristoides pulvinatum pusillum reptans spathulatum spinulosum versicolor The numerous varieties described in many of the above, where not directly reflected in use by some authors at species level of names used by other authors at infraspecific levels, also require rationalisation. Amongst the above, the names canariensis (q.v.) and versicolor [previously cartilagineum] (q.v.) are perhaps the least likely to reveal anomalous application, whilst Gelidium sesquipedale (q.v.) has been omitted entirely from the list. G. sesquipedale is perhaps the least variable species in the genus and therefore the least likely name to be misapplied. Such confusion as exists about it has largely resulted from the involvement of Fucus corneus Hudson [Gelidium corneum auct.] in the situation, since F. corneus of Hudson has been shown (Dixon, 1967) to be based on Buddie (BM) specimens that proved to be G. sesquipedale. Hence, as a corollary, derives further confusion about the significance and application of the epithet corneum (q.v.). Gelidium apiculatum Kiitzing Note. See Gelidium spinulosum (C. Agardh) J. Agardh, for which this is a possible synonym, and the note to Gelidium microdon Kiitzing. Gelidium arbuscula Bory ex B0rgesen Canaries (306B; 582; 598). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). 'Tropical Africa (N. Gambia- Congo river)' (598). [As Gelidium arbuscula (Montagne) B0rgesen] Canaries (2; 229; 269; 270; 351; 374; 393; 394; 486; 490; 529; 567). Cote d'lvoire (287; 586). Guinee(350;586). Mauritanie (59). Senegal (59; 123; 483; 586). 'remonte du Senegal jusqu'a la Mauritanie.' (59). 'warm temperate and tropical parts of the eastern Atlantic Ocean.' (350; 586). [As Gelidium arbuscula Bory] Canaries (4; 5; 68; 191; 227; 379; 489; 493; 547; 581). Guinea-Bissau (529). Guinee (529). Senegal (122; 7529; 7530; 7531). [As Gelidium arbuscula (Bory) Montagne] Canaries (253; 583). Senegal (253). RHODOPHYTA (pLORIDEAE) OF TROPICAL AFRICA 215 [As Gelidium corneum (Hudson) Lamouroux var. nereideuml Lightfoot] Canaries (401). Note. See B0rgesen (68: 85-89) for details of the taxon and its background to 1927. Many of the above records involve considerable doubt as to their accuracy. All Sourie's (529; 530; 531) publications mentioned that doubt; Feldmann, who reviewed Sourie's material from Senegal, was said by Sourie (529: 116) to have considered specimens of this species 'assez differents de la plante des Canaries'. Essentially the same message lies behind the Bodard & Mollion statement (59: 198) that their Senegal plants represented 'une endemique possible que Dangeard avait appelee Gelidium arbuscula et qui semble plutot une espece nouvelle que J. Feldmann nous a fait remarquer.' Feldmann had earlier (191: 414) expressed the opinion that (for the Canary Islands) 'Le Gelidium arbuscula, endemique bien caracterise, peut neanmoins etre rapproche du Gelidium sesquipedale, espece lusitano-africaine.' Acuna Gonzalez (2: 3), also concerned with Canary Islands plants, commented that there 'tambien existen algunas [especies] que son endemicas, como. . . . Gelidium arbuscula' . It therefore seems a reasonable possibility that at least two separate entities are concealed within the application of this name in the list area; it may be possible partly to utilise ecological characteristics in distinguishing the entities - Santos, Acuna & Wildpret (490) indicated that what they called Gelidium arbuscula had an 'especial predilection por los acantilados verticales [vertical or steep cliffs]'. The record (401) in Montagne is included here on the basis of B0rgesen's comment (68: 86) that G. arbuscula Bory in herb, 'seems to be a rather common species at the [Canary] Islands. . . . further, according to a specimen in Herb. Montagne, the Gelidium corneum var. nereideum, too, is this plant'. See also the note to Gelidium as genus. Haroun Tabraue et al. (583: 111) commented that [on Gomera]: 'Gelidium arbuscula is easily recognisable by its purple coloured thalli, which clearly contrast with the brown-yellow of Cystoseria abies-marina [amongst which it grows] and the blackish colour of Gelidium versicolor '.' (Translation from the Spanish.) On p. 112, they added yet another difference - the greater rigidity of G. versicolor (q.v.). Gelidium asperum Montagne Note. See Gelidium spinulosum (C. Agardh) J. Agardh. The present name is probably a manuscript name appearing only on a specimen in the Herb. Montagne (PC); in any case, it appears much like Gelidium microdon Kiitzing, itself probably a synonym of G. spinulosum. Data also appear in B0rgesen (68: 90). Gelidium attenuatum (Turner) C. Agardh See Gelidium latifolium (Greville) Bornet in Bornet & Thuret. Gelidium caespitosum Kylin Note. This name was originally applied by Fox to BM specimens from Ghana (April 1952) and from Nigeria (March and November 1954); the records were subsequently published by her (1957) under the name Gelidium pusillum (Stackhouse) Le Jolis. Specimens were afterwards reassigned by Lawson & John (350) to Gelidium corneum sensu B0rgesen. Gelidium canariensis (Grunow) Seoane-Camba Canaries (518; 598). [As Gelidium versicolor (S. Gmelin) Lamouroux] Canaries (172; 180; 227; 253; 352). [As Fucus versicolor Gmelin] Canaries (90). Note. The taxon was recorded as both Fucus cartilagineus L. and Fucus versicolor Gmelin by Bory (90), who also (90: 304) stated the former to be a synonym of the latter. The alga was said to be: Tres comun sur toutes les pierres et les roches de la rade de Sainte-Croix.'. De Toni (131: 152 et seq.) also placed G. cartilagineum in the synonymy of G. versicolor Lamouroux. [As Fucus cartilagineus L.] Canaries (90). Note. See the note above, to Fucus versicolor Gmelin. [As Gelidium cartilagineum (L.) Gaillon] Canaries (25; 68; 81; 89; 131; 170; 229; 254; 269; 305; 318; 351; 386; 393; 394; 401; 482; 486; 489; 490; 567). Note. The records in Montagne (401: 158), J. Agardh (25), and De Toni (131) are based on Bory's (90) data. 216 J. H. PRICE, D. M. JOHN & G. W. LAWSON [As Gelidium cartilagineum var. canariense Grunow in Piccone or var. canariensis Grunow in Piccone] Canaries (68; 131; 191; 439; 482; 489; 567; 581). [As Sphaerococcus cartilagineus (L.) C. Agardh] Canaries (19). Note. Recorded by C. Agardh (19: 286-288) as 'Ad insulas Canarias; sec. Bory Voy.' [As Gelidium cartilagineum Gaillon] Canaries (44; 385; 493). Note. The record in Benftez (44) is directly based on Montagne (401). That in von Martens (385) is based (385: 38) on a record of Mertens, from Tenerife. [As Gelidium cartilagineum (L.) Greville] Canaries (37; 242). [As Corallina rubens Tournefort] Canaries (548). Note. For explanation of attribution of this record, see the notes to Jania rubens (L.) Lamouroux. [As Gelidium cartilagineum Greville] Canaries (493). Note. See the entry for Gelidium versicolor (S. Gmelin) Lamouroux for records from other geographical areas. The rationale for recognition of Gelidium canariensis is documented in the notes to G. cartilagineum (L.) Gaillon. Gelidium capillaceum Kiitzing [or (S. Gmelin) Kiitzing] See Pterocladia capillacea (S. Gmelin) Bornet & Thuret. Gelidium cartilagineum (L.) Gaillon See Gelidium canariensis (Grunow) Seoane-Camba and Gelidium versicolor (S. Gmelin) Lamouroux. Note. The name Gelidium cartilagineum (L.) Gaillon has to be rejected for that genus since it was based on material that has proved to be representative of the genus Plocamium (Dixon, 1967). Thus, the southern African species of Gelidium to which the epithet had largely by custom been applied required another name, the earliest available being shown by Dixon (1967: 58) to be Gelidium versicolor (S. Gmelin) Lamouroux, based on Fucus versicolor of Gmelin (1968). The most northerly attached records of the taxon to which the name G. versicolor (S. Gmelin) Lamouroux is thus to be applied are from the Canary Islands (Dixon, 170: 47, and subsequent publications, especially 172: 134). For that same area, Seoane-Camba (518) recently concluded that materials from 'mainland' Spain and from the Canary Islands were sufficiently different to represent distinct taxa, despite hitherto having been collectively called Gelidium cartilagineum (i.e. correctly G. versicolor). He raised the var. canariensis Grunow of G. cartilagineum to specific status as Gelidium canariensis (Grunow) Seoane- Camba; all previous Canaries records are here transferred to that latter species entry, although material behind all the earlier records has not been checked to confirm their validity. Previously, J. Feldmann (191: 414) had considered (without effecting it) that the Canaries material of then so-called Gelidium cartila- gineum was 'represente par une variete endemique (var. canariensis} assez differente du type pour meriter peut-etre d'etre consideree comme une espece distincte. Le G. cartilagineum type est une espece indo-pacifique, tres abondante en particulier en Afrique du Sud et en Californie.' See also the general note to Gelidium. Gelidium claviferum Kiitzing See Gelidiella acerosa (Forsskal) J. Feldmann & Hamel. Gelidium corneum sensu B0rgesen Canaries (486). Gabon (350; 586). Gambia (350; 586). Ghana (350; 377; 586). Liberia (350; 586). Nigeria (350; 586). Togo (350; 586). 'widespread in boreal-antiboreal to tropical parts of the Atlantic Ocean.' (350; 586). RHODOPHYTA (FLORIDEAE) OF TROPICAL AFRICA 217 [As Gelidium corneum (Hudson) Lamouroux] Angola (352). Canaries (66; 401). Gabon (294). Gambia (296). Ghana (288; 491). Liberia (129; 288). Nigeria (288). Togo (288). 'in [warmeren] atlantischen [Ocean]' (504). 'warmer parts of the Atlantic Ocean' (60). [As Gelidium corneum (Linn.) Lamour.] Canaries (331; 332). [As Gelidium corneum (Turner) Lamouroux] Canaries (68; 71; 499). Cape Verde Islands (499). 'Nordwestafrika' (499). 'Westafrika'(499). '[Madeira, but] . . . probably widely distributed.' (375). 'Seems to be widespread.' (68). [As Gelidium corneum (Lamouroux)] Canaries (196). [As Gelidium corneum Lamouroux] Angola (41; 42). Canaries (44; 547). Cape Verde Islands (38; 223; 578; 596). 'Throughout all oceans.' (410). [As Gelidium corneum Hudson] Cape Verde Islands (145). [As Gelidium corneum [sensu - as vgl.] Feldmann et Hamel] Angola (500). [As Gelidium arbuscula Bory] Liberia (286). Togo (293). [As Gelidium pusillum} Nigeria (213). [As Sphaerococcus corneus (Hudson) C. Agardh] Canaries (19). Note. C. Agardh (19: 279-285) recorded the 'parent' species as 'Ad Teneriffam; Bory'. He also recognised 17 varieties/subspecies, mostly with names still employed somewhere in Gelidium. Amongst those is ft. sesquipedalis, recorded from the Mediterranean and Cadiz. The Bory record and the significance of this S. corneus entry will need careful rationalization. [As Fucus corneus L.] Canaries (90). Notes. See the note at Sphaerococcus corneus (Hudson) C. Agardh subhead, above. General notes. For records under names often attributed as varieties of Gelidium corneum (Hudson) Lamouroux, refer as follows in this work: var. capillaceum (S. Gmelin) C. Agardh Pterocladia capillacea (S. Gmelin) Bornet et Thuret var. clavatum (Lamouroux) C. Agardh Gelidium pusillum (Stackhouse) Le Jolis var. crinale (Turner) C. Agardh Gelidium crinale (Turner) Desmazieres var. nereideuml Lightfoot Gelidium arbuscula Bory ex B0rgesen 218 J. H. PRICE, D. M. JOHN & G. W. LAWSON var. plnnatum (Hudson) Montagne Pterodadia capillacea (S. Gmelin) Bornet & Thuret var. sesquipedale Clemente Gelidium sesquipedale (Clemente) Thuret in Bornet & Thuret var. spinulosum C. Agardh Gelidium spinulosum (C. Agardh) J. Agardh. Amongst the attributive 'species' so involved, some (e.g. G. pusillum, G. crinale) are often considered as conspecific (e.g. by Wynne, 1986). Records established under the name Gelidium corneum in most cases present considerable interpretive problems, whatever authorities have been quoted for the combination. G. corneum represents perhaps the classic extreme of general confusion. Overlying all that is the fundamental difficulty of typification of the basionym Fucus corneus of Hudson. For general background to nomenclatural and morphological problems with this taxon, see Dixon (1967), together with the notes to the entries here for the genus Gelidium and for Gelidium sesquipedale (Clemente) Thuret in Bornet & Thuret. The name as utilized at present for the list area is only in the sense applied by B0rgesen (68: 85), and even that brings with it the need for a complete revision of whatever backing material is available for all the records involved. Therefore, what we offer here is merely a statement of rationalization through what others have said, not of what we conceive as taxonomic/nomenclatural truth, even assuming the latter to be approachable. If B0rgesen's taxon does finally turn out to be a good one, it will require whatever is the earliest available name, or a new one, as the case may be. At presents corneum as an epithet is completely confused in all respects. It is clear that even the reduction of the general chaos of names/taxa in Gelidium to the two aggregates 'latifolium' and ' pusillum' (leaving aside for now the more extreme morphologies of G. versicolor, q.v., and G. sesquipedale, q.v.), as has been suggested by Dixon & Irvine (172: 126 et seq.) leaves a plethora of problems in that these two aggregates are themselves not satisfactorily morphologically distinguishable apart at all times and from all locations; the artificial key presented in 172: 127 admits as much from the statement that it 'should be used with caution', and from the nature of the distinguishing characteristics provided. Since the epithet corneum has variously been applied, and conceptually recognized as applying, for our area to forms that could be placed in either or both of the aggregates, the point made by Lawson & John (350: 177) regarding Fox's (213) Nigerian plants reported as G. pusillum being more closely related to G. corneum sensu B0rgesen may have relatively little overall significance, although morphologically accurate for their material at that time. As observed by Seagrief (570), some applications of the name corneum (those sensu Harvey) for material from Sea Point, South Africa, do not represent Gelidium corneum (Hudson) Lamouroux (nor, apparently, Gelidium corneum sensu B0rgesen), but are attributed to Gelidium micropterum Kutzing (q.v.). On the basis of BM holdings of material collected during the her Benguellense of F. Welwitsch [those specimens circulated under the numbers 73 (Benguela); 74 (Mo9amedes); 75 (drift, from near Benguela)], it seems highly likely by comparison with the illustration presented by Kutzing (326: 21, Tab. 59 (4137), 1868) that this attribution to Gelidium micropterum Kutzing is also true for at least some material from Angola. It probably equally applies for the coast of Namibia, whence Gelidium micropterum Kutzing has been firmly reported. There has, however, been doubt about the way in which the name micropterum Kutzing has been applied in Namibia; it has been suggested that Gelidium pristoides (Turner) Kutzing was perhaps involved. For some of the supposed distinctions between G. corneum and G. micropterum, see P. S. Rao (1970: 71-73). Material recorded by Lawson, John & Price (352: 311), from Angola, as patches of short red algal 'turf of Gelidium corneum (Hudson) Lamouroux, included a mixture of forms that, in isolation, would have been considered as representing variously G. corneum sensu B0rgesen, G. pusillum (Stackhouse) Le Jolis, and Gelidium micropterum Kutzing. All the above complications are, of course, compounded by varying local traditions of epithet application in Gelidium such that, other than generic validity, the use of an epithet in one country or region may present no biogeographic information in any way of use for another area. This is so for most of the epithets of other than very local application and markedly so for 'corneum'. Gelidium coronopifolium (Goodenough & Woodward) Lamouroux See Sphaerococcus coronopifolius (Goodenough et Woodward) Stackhouse. Gelidium crinale (Turner) Desmazieres Gabon (350; 586). Gambia (296; 350; 586). RHODOPHYTA (FLORIDEAfi) OF TROPICAL AFRICA 219 Ghana (350; 586). Liberia (350; 586). Mauritania (349). Sao Tome (350; 586). Sierre Leone (350; 586). Western Sahara (349; 476). [As Gelidium crinale (Turner) Lamouroux] Angola (352). Canaries (2; 68; 108; 118; 191; 214; 252; 486; 517; 546; 547; 584). Gabon (294). Ghana (152; 338; 491; 535). Guinee-Frangaise (529). Liberia (129). Mauritanie(252;500;535). Salvage Islands (38B; 231; 375). Sao Tome (251; 265; 535). Senegal (33; 121; 122; 529). Sierra Leone (30; 295). 'Atlantic coast of Europe, N. Africa and N. America' (375). 'Atlantico de Inglaterra a Canarias' (517). 'Atlantique (de 1'Angleterre aux Canaries, Cap Vert . . .)' (33). 'English coast to the Canary Is.' (68). 'Gulf of Guinea' (269). 'in [warmeren] atlantischen Ocean' (504). 'repandu dans toutes les mers chaudes' (188). [As Gelidium crinale Lamouroux] Canaries (547). [As Gelidium crinale Lamouroux in Bory] Canaries (89). 'D'Angleterre aux Canaries' (89). [As Gelidium corneum (Huds.) Lamour. var. crinale (Turner) C. Agardh] Canaries (401). [As Gelidium crinale (Thuret) Lamouroux] Senegal (530). Note. The epithet rendering in Prime (476: 23) is as 'orinale\ a purely typographic error that has been ignored. It is now generally accepted that Lamouroux, despite employing the combination G. crinale (in Bory, Diet, class. d'Hist. nat. 7: 191 (1825)), presented neither basionym data and source, nor descriptions of the taxon, thus not effecting a proper transfer; the first to achieve correct recombination of Turner's epithet crinale in Gelidium was Desmazieres. Opinions on this taxon and the often-confused Gelidium pusillum (Stackhouse) Le Jolis have varied widely, from absolute synonymy to at least usage in very different and identifiable senses. Except where both the concepts used and the mode of application of the names have both been entirely clear and consistent, the possible equivalence of G. crinale with G. pusillum agg. has not been here taken as accepted. Names are therefore otherwise maintained as used by the publishing author, but it should be borne in mind that, given reasonable consistency of view and application, it is strongly likely that many records under the two names represent the same taxon. See also the entry for G. pusillum (Stackhouse) Le Jolis. Gelidium elminense Dickinson Cote dTvoire (288; 350; 586). Ghana (152; 288; 350; 586). 'in tropical parts of the eastern Atlantic Ocean.' (350; 586). 'Tropical Africa (N. Gambia-Congo river)' (598). Note. This species was originally described on the basis of plants from Elmina and Iture, both Ghana, collected in April 1945 (25th and 26th, respectively). According to Lawson & John (350; 586), the species falls within the Gelidium crinale complex sensu Dixon. From her own comments in the original description 220 J. H. PRICE, D. M. JOHN & G. W. LAWSON (152: 43), Dickinson was markedly reluctant to indulge in describing yet another new species in an already ill-understood and cluttered genus, but could see no other way out of such a tediously complex situation. Gelidium flaccid urn P. Dangeard Guinea-Bissau (529). Mauritanie (121). Senegal (121; 122; 529). 'Subtropical Africa [ Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). Note. In publishing the original description of G. flaccidum, Dangeard (121: 23) noted that 'La coupe transversale de la fronde montre la grande rarete des rhizines. Par ce caractere ce Gelidium se rapproche du G. melanoideum.' Gelidium foliosum P. Dangeard Mauritanie (121). Senegal (50; 121; 122; 123; 529). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). Note. In the process of describing new taxa from Senegal and Mauritanie, Dangeard (121 : 22) stated that 'Deux especes enfin de Gelidium [G. flaccidum; G. foliosum] de ses regions sont nouvelles et bien caracterisees'. Despite that, he went on to state [in both cases of G. foliosum]: 'rappelle un peux le Gelidium pusillum var. pulvinatum' and, later: 'ressemble un peu par son port au G. galapagense decrit par R. TAYLOR [sic!] (1945)'. He added that G. foliosum 'il en differe par ses ramules a tetrasporanges retrecis a la base et comme pedicelles.' The resemblance to Gelidium pusillum, especially var. pulvinatum, seems to have been too close for many workers to take the present taxon seriously. Soon after G. foliosum was published, Sourie (529: 289 and footnote 1) commented: 'Selon J. Feldmann (opinion inedite), ce Gelidium [foliosum] pourrait n'etre qu'une forme de G. pusillum.'' Even Bodard (50) doubted the validity of it, stating: 'Je pense que le Gelidium foliosum ne peut etre considere que comme une variete de Gelidium pusillum'. He did not give reasons for this opinion. More recently, one of us (DMJ) annotated a collection (John & Seku; no. 6949) from Harper, Liberia, 2 Jan. 1972, determined as Gelidium pusillum var. pulvinatum, 'Similar to description of G. foliosum by Dangeard (121: 22) from Senegal'. Gelidium galapagense W. Taylor See the entry to Gelidium foliosum P. Dangeard. Gelidium intricatum Kiitzing See Gelidiopsis intricata (C. Agardh) Vickers. Gelidium latifolium (Greville) Bornet & Thuret Canaries (253; 306B; 598; 604). Cape Verde Islands (598). Senegal (59). Western Sahara (253; 349; 393; 394). 'Atlantico (Noruega - Rio de Oro)' (604). 'Atlantico Oriental (Noruega - Rio de Oro)' (253). 'Atlantique: depuis les cotes anglaises jusqu'au Rio de Oro' (222). 'Nordwestafrika' (499). 'Southern Norway to Rio de Oro' (172). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). 'Tropical Africa (N. Gambia- Congo river)' (598). [As Gelidium latifolium (Greville) Thuret & Bornet] Senegal (59; 7529). 'Atlantic Ocean (European and African coasts, Canary Islands)' (177). 'Atlantique (de 1'Angleterre au Rio de Oro)' (33). [As Gelidium latifolium Bornet] 'Atlantic Ocean: . . . African coast.' (566). [As Gelidium attenuatum (Turner) C. Agardh] Western Sahara (393; 394; 476). Note. Despite the convenient recognition, in some parts of the world, of the contrasting morphologies of the 'species aggregates' known as G. latifolium and G. pusillum, it is by no means certain that the two RHODOPHYTA (FLORIDEAE) OF TROPICAL AFRICA 221 groups do not represent merely extremes of a morphological cline representing environmental or racial characteristics within the range of expression of a single genome. Haroun Tabraue et al. (253: 259) referred to their Canary Islands report as: 'nueva para el Archipielago Canario.' Dangeard (117; 118: 184-5) had previously stated that Gelidium attenuation was missing from the Canaries. It remains to be shown whether this and the other records above are significant or not. The expression of difficulty in determination and doubt occurs throughout many of the comments associated with records: Gayral (222: 373) indicated the wide range in morphology of the 'species', from 'var. luxurians" (wide branches; distichous) to 'var. hystrix' (cylindrical; minor laterals spiralled on axes), with all kinds of intermediates. Sourie (529) recorded his determination as doubtful, though the reason for doubt is not clear. Perhaps the extreme in confusion over this taxon was shown clearly by Dixon (170: 47), who revealed previous misdetermination as Gelidium sesquipedale (Clemente) Thuret in Bornet & Thuret (q.v.) of 'the entity known as G. attenuatuni '; G. sesquipedale is one of the few species in the genus with a very characteristic appearance. Two of us (Lawson & John, 349) were unable to find material in Western Sahara; our record is based on previous comments, of which the only original one is probably that in Primo (476). Gelidium melanoideum Schousboe ex Bornet Senegal (52; 529). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania Former W. Sahara]' (598). [As Gelidium melanoideum var. pseudopulvinatum J. Feldmann] Senegal (529). Note. For an analysis of characteristics that distinguish between G. melanoideum and the Gelidium crinalelG. pusillum complex, see Dixon & de Valera (1961). The species, originally recognised from the Tangier region, is principally of Mediterranean origin. See also the note to Gelidium flaccidum P. Dangeard. Gelidium microdon Kiitzing Western Sahara (349; 393; 394). 'Atlantique (. . . Canaries . . .)'. (33). [As Gelidium spinulosum (C. Agardh) J. Agardh] Western Sahara (476). Note. It is possible that the taxa G. microdon and G. spinulosum may be wholly conspecific, in which case the above records should all appear under the latter, being the older valid name. Ardre (33: 68), Bornet (89: 112 [272]) and B0rgesen (68: 90) all comment on conspecificity of these taxa; Bornet is the most unequivocal in his opinion, whilst Ardre suggests conspecificity and B0rgesen merely quotes Bornet's comments, under his heading of G. spinulosum (C. Agardh) J. Agardh. See also Gelidium asperum Montagne. Bornet's comment was: 'Les Gelidium apiculatum Kiitz. . . . et microdon Kiitz. . . . qui sont figures d'apres des echantillons de Tanger et de Cadix, ne sont que des formes de cette espece.' 'Cette espece' refers to the entry under which the statement appears, Gelidium spinulosum J. Agardh. The record above in 349 is based only on Michanek (393; 394) and Primo (476), Lawson & John not themselves having detected specimens. Gelidium micropterum Kiitzing Namibia (348; 522; 522 A). Senegal (121; 122). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). [As Gelidium corneum (Hudson) Lamouroux] Angola (352). Note. The Lawson & Isaac (348) record from Namibia was based directly on the data from 522 (Simons), since they did not locate material. Applications of the epithet micropterum, the concept behind the taxon, and any synonymy involved are all highly complex and opinions on the matter have varied greatly. For a detailed consideration of some aspects of this, see the notes to Gelidium corneum sensu B0rgesen as regards South Africa, Angola, and Namibia. Gelidium pristoides (Turner) Kiitzing (q.v.) is often considered to be involved, whilst De Toni (131: 152 et seq.) placed G. micropterum (albeit with '?') in the synonymy of Gelidium cartilagineum (L.) Gaillon [now Gelidium versicolor (S. Gmelin) Lamouroux or G. canariensis (Grunow) Seoane-Camba]. Gelidium pannosum Bornet See Gelidiella tenuissima J. Feldmann & Hamel. 222 J. H. PRICE, D. M. JOHN & G. W. LAWSON Gelidium pannosum Grunow See Gelidiella tenuissima J. Feldmann & Hamel. Gelidium pectinatum Montagne Canaries (68: 191; 392; 547; 584). 'Ad . . . litora Africae borealis' (318). [As Gelidium pectinatum (Schousboe) Montagne] Canaries (177; 227). Note. B0rgesen (68) knew of a single specimen found in the Canaries; he had not seen it. Citations of authorities that include the name of Schousboe are in error (as, e.g., in Boudouresque et al., 1984). The form used by the latter, 'Schousboe ex Montagne', is inaccurate since Montagne (403: 108) did not use the whole of the name by which Schousboe had referred to the taxon (Teloedema pectinatum Schousboe) on a specimen in Webb's herbarium. Gelidium pristoides (Turner) Kiitzing Namibia (348). Note. See the notes to Gelidium micropterum Kiitzing and to Gelidium corneum sensu B0rgesen. There has been past suggestion that Gelidium pristoides and Gelidium corneum auct. are directly related taxa; Montagne (404: 77), for example, referred to pristoides as 'variete pristoides du G. corneum', having more or less contemporaneously (403: 108) included 'Sphaerococcus corneus var. pristoides C. Ag.' in the synonymy of his Gelidium pectinatum. Gelidium pristoides is widely and abundantly reported for many localities in South Africa (Simons, 523: 259; Carter & Anderson, 579: 117; and others), so that whatever the real status of the morphology to which the name is applied, it occurs often in the general area here considered. A recent treatment of all non-parasitic genera of Gelidiaceae founded primarily on surface cell morphology (Akatsuka, 1986) resulted in the establishment of a new genus, Onikusa, to include both Gelidium pristoides (Turner) Kiitzing (the type species of Onikusa) and Gelidium japonicum (Harvey) Okamura. Onikusa is derived from the Japanese name for G. japonicum. Since the study affects also the entries for other genera of Gelidiaceae (Pterocladia; Suhria) found in the list area, it will be dealt with fully later. Gelidium pulchellum (Turner) Kiitzing Gambia (296). Senegal (121; 122). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). [As Gelidium pulchellum var. claviferum] Canaries (38C). Salvage Islands (38C). Note. It is commonly the case with most authors that Gelidium pulchellum is taken to be a synonym of Gelidium pusillum (Stackhouse) Le Jolis (q.v.), or at least to fall morphologically within the range taken to represent the 'Gelidium pusillum-G. crinale-G. pulchellum series' (see John & Lawson, 296: 289-290). Boudouresque et al. (1984: 44) and Wynne (1986) both believed at least pulchellum and pusillum to be synonymous, Boudouresque et al. further not distinguishing for the latter the var. pulvinatum (C. Agardh) J. Feldmann. Audiffred (38C: 173) cited the records from Canaries and Salvage Islands only in the context of the entry for Ceramium echionotum J. Agardh, the 'host'. The species list provided in the same publication included only Gelidium pusillum Stackhouse) Le Jolis, perhaps tending to imply that a nomenclatural and taxonomic change had been overlooked. Gelidium pulvinatum (Kiitzing) Thuret ex Bornet Canaries (68; 490). [As Gelidium pulvinatum Thuret] Canaries (493). Note. B0rgesen (68) found no specimens of this form amongst Canaries material, but it was listed by Sauvageau (493). It seems probable (B0rgesen, 68) that G. pulvinatum is closely related to G. pusillum (q.v.) and G. spathulatum (q.v.), all being forms of a single species. Schmidt (497: 110-111) relegated Gelidium pulvinatum Kiitzing to Gelidium pusillum f. pulvinata (Kiitz.), a process initially carried out by J. Feldmann (in Hamel, 1927) on the basis of C. Agardh's (19: 284) Sphaerococcus corneus o. pulvinatus. Gelidium pusillum (Stackhouse) Le Jolis Angola (352). Ascension (7474; 475; 567). RHODOPHYTA (pLORIDEAE) OF TROPICAL AFRICA 223 Benin (288; 293; 350; 586). Bioko (346; 350; 586). Cameroun (269; 288; 337; 350; 374; 454; 460; 484; 500; 537; 586). Canaries (13; 16; 38B; 38C; 38D; 68; 70; 71; 108; 128A; 191; 226; 227; 229; 236; 237; 253; 306B; 336; 351; 375; 379; 392; 486; 497; 499; 517; 556; 584; 585; 598; 604; 605; 608). Cape Verde Islands (213). Cote d'lvoire (287). Gabon (213; 350; 586). Gambia (350; 586). Ghana (213; 288; 297; 350; 377; 491; 586). Liberia (129; 287; 288; 350; 586). Mauritanie (38B; 38C; 38D; 349; 7529; 556). Salvage Islands (38B; 38C; 38D; 231; 375; 556; 556A; 598). Sao Tome (288; 350; 586). Senegal (38B; 38C; 38D; 59; 121; 213; 253; 336; 529; 556). Sierra Leone (30; 213; 295; 336; 339; 350; 374; 378; 586). Togo (288; 293; 350; 586). Western Sahara (38B; 38C; 38D). 'Atlantico (Noruega - Cabo Verde. . . .)' (253; 604). 'Atlantique (de 1'Angleterre aux Canaries)' (33). 'Atlantischer Ozean, von den englischen Kiisten an siidwarts bis zur afrikanischen Kiiste und den Kanaren' (499). 'Central Norway to Cape Verde' (172). 'extensive in warm & tropical seas' (81). 'Nordwestafrika' (499). 'Semble etre repandu dans toutes les mers chaudes' (188). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). 'Tropical Africa (N. Gambia- Congo river)' (598). Tropical West Africa' (611). 'Von den englischen Kiisten an sudwarts bis zur afrikanischen Kiiste' (497). 'Westafrika' (499). [As Gelidium pusillum var. pulvinatum (C. Agardh) J. Feldmann] Angola (352). Canaries (518; 519; 584). Gabon (294). Gambia (296). Liberia (129; 287; 288). Mauritanie (349). Senegal (121; 122; 529). Sierra Leone (295). [As Gelidium pusillum (Stackhouse)] Canaries (5). [As Gelidium pusillum Le Jolis] Canaries (547). [As Gelidium pusillum (Stark) le jolis var. pulvinatum (Ag.) Feldm.] Senegal (50). [As Gelidium pusillum var. conchicola Pice. & Grun.] Senegal (121; 122; 529). [As Gelidium pusillum var. minusculum Weber van Bosse] Senegal (121; 122; 529). [As Gelidium crinale (Turner) Lamouroux] Mauritanie (252; 500; 535). Western Sahara (476). 224 J. H. PRICE, D. M. JOHN & G. W. LAWSON [As Gelidium reptans (Suhr) Kylin] Mauritania (121; 122). Namibia (348). Sao Tome (93; 535). Senegal (121; 122; 296; 529). [As Gelidium corneum (Hudson) Lamouroux] Angola (352). [As Gelidium corneum (Hudson) Lamouroux var. davatum (Lamouroux) C. Agardh] Canaries (401). Note. According to P. S. Dixon (pers. comm., 1974), the taxon Gelidium pusillum (Stackhouse) Le Jolis var. pulvinatum (C. Agardh) J. Feldmann should correctly be known as Gelidium crinale (Turner) Lamouroux. Although Gelidium reptans (Suhr) Kylin has been recorded from Mauritanie, Namibia, Sao Tome, and Senegal, and tentatively from Gambia, we here follow B0rgesen (81) and others in considering it as no more than a variety of Gelidium pusillum. Sourie (529: 116), in commenting on material collected and recorded as G. reptans, stated 'les exemplaries recoltes ne seraient peut-etre qu'une forme de G. pusillum'. There is a similarly close relationship between G. foliosum of P. Dangeard and G. pusillum var. pulvinatum, as indicated by John (288); see also the notes in the entry for G. foliosum. Material recorded by Lawson, John & Price (352) from Angola as patches of short red algal 'turf of G. corneum (Hudson) Lamouroux included a mixture of forms that, in isolation, would have been considered as representing G. corneum sensu B0rgesen, G. pusillum (Stackhouse) Le Jolis and G. micropterum Kiitzing (q.v.). Gelidium reptans (Suhr) Kylin Note. See Gelidium pusillum Stackhouse) Le Jolis, especially the note to that species entry. Gelidium rigidum (Vahl) Greville [or simply Greville] See Gelidiella acerosa (Forsskal) J. Feldmann & Hamel. Gelidium senegalensis Bodard (nomen) [or J. Feldmann mscr.] Senegal (59). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). [As Gelidium senegalensis J. Feldm. mscr.] Senegal (59). Note. Included for completeness only, since no description or further data have been traced, and the relationships of material to which the name has been applied are not clear. Bodard & Mollion (59) included the form of the name used as heading here in their Table I (Baie de Goree), without description or further mention in the text. Later, in their Table IIIB, they referred to what must be taken as the same taxon as 'G. senegalensis J. Feld. mscr.' The name was presumably annotated on to a specimen or specimens by Feldmann and taken up temporarily by Bodard. Gelidium serrulatum J. Agardh [As Gelidium serrulatum Ag. var.] Angola (261; 262). Note. Originally described from Venezuela and known from Trinidad. Accepted by Taylor (540: 357) without comment, but not frequently employed as a name in the literature. Relationships of the material here involved are not clear. Gelidium sesquipedale (Clemente) Thuret in Bornet & Thuret Canaries (68; 117; 139; 191; 227; 598). Cape Verde Islands (122; 252; 598). Guinea-Bissau (529). Mauritanie (121; 122; 500; 529). Senegal (122; 252). Western Sahara (393; 394; 476; 529). 'de Inglaterra a Canarias' (517). 'does not extend [southwards] into Gulf of Guinea' (487). 'English coast southwards to Canary Islands' (68). 'French Channel coast to Mauretania' (172). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). 'Siidengland bis Mauritanien.' (567). RHODOPHYTA (FLORIDEAE) OF TROPICAL AFRICA 225 [As Gelidium sesquipedale (Clemente) Bornet & Thuret] Mauritanie (349). Western Sahara (349). [As Gelidium sesquipedale (Turner) Thuret in Bornet & Thuret] 'D'Angleterre aux Canaries' (89). [As Gelidium sesquipedale (Turner) Thuret] Cape Verde Islands (191; 423). Mauritanie (252). 'Atlantique: . . . sur les cotes anglaises et descend jusqu'a la pointe du Cap Blanc, en Mauritanie.' (222). 'Atlantique (de I'Angleterre a la Mauritanie' (33). 'du sud de I'Angleterre a la Mauritanie' (195). [As Gelidium sesquipedale Thuret in Bornet & Thuret] 'dall' Inghilterra alle Canarie' (390). [As Gelidium sesquipedale Thuret] Canaries (117; 118). Cape Verde Islands (38). 'De I'Angleterre aux Canaries.' (38). [As Gelidium resquipedalex] Canaries (237). [As Gelidium corneum Lamouroux var. sesquipedale J. Agardh] Cape Verde Islands (38). 'De I'Angleterre aux Canaries' (38). [As Gelidium corneum (Hudson) Lamour. var. sesquipedale Clemente] Canaries (401). [As Gelidium corneum (Hudson) Lamouroux var. sesquipedale Agardh] Cape Verde Islands (408). [As Gelidium corneum (Hudson) Lamouroux var. sesquipedale Greville] Canaries (141A). Note. B0rgesen did not find this species on the Canaries; he had only two previous records, from Montagne (401: 158) and from the herbarium of H. C. Lyngbye. He therefore referred to the species there as 'probably rare'. Although the Western Sahara record in Lawson & John (349) is supported by an original discovery from the Cap Blanc area, the statement for Mauritanie is secondary, being based on various of the data included here. See also the notes to Gelidium arbmcula Bory ex B0rgesen. Gelidium spathulatum (Kiitzing) Bornet Canaries (16; 68; 191; 375; 517; 584). 'Atlantique (de I'Angleterre aux Canaries . . .)' (188). 'Atlantique (du golfe de Gascogne aux Canaries)' (33). [As Gelidium spathulatum Kiitzing] Canaries (493). Note. Bornet (89: 108 [268]) suggested that this taxon covered a form of G. crinale (q.v.), not a good and separate species. B0rgesen (68) expressed general agreement with that view. For extended comments on the status of G. spathulatum, see Feldmann & Hamel (195: 115) and Ardre (33: 71-72 [203-204]). Gelidium spinulosum (C. Agardh) J. Agardh Canaries (2; 68; 191; 221; 227; 401; 431; 499). Western Sahara (393; 394; 476). 'Nordwestafrika' (499). 'outre le Maroc aux Canaries' (221). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). [As Gelidium spinulosum J. Agardh] Canaries (242). 'De Cadix aux Canaries' (89). 226 J. H. PRICE, D. M. JOHN & G. W. LAWSON [As Gelidium spinulosum J. Agardh ex J. Agardh] 'In Oceano atlantico calidiore' (27). [As Gelidium corneum (Hudson) Lamouroux var. spinulosum C. Agardh] Canaries (318; 401). 'Ad oras Africae' (318). Note. See the notes to Gelidium microdon Kiitzing for comments on conspecificity of these two and other species. Although Acuna Gonzalez (2) stated 'perteneciente al continente Africano', Papenfuss (431: 174) excluded the species from the South African marine flora. Gelidium versicolor (S. Gmelin) Lamouroux Angola (298; 352; 611). Canaries (253; 306B; 582; 583). 'West African coast to the north of the Gulf of Guinea' (352) . [As Gelidium cartilagineum (L.) Gaillon] Angola (298; 393; 394; 424; 425). Note. For records of what was previously deemed the same taxon from the Canary Islands, see Gelidium canariensis (Grunow) Seoane-Camba. Although Simons in Day (523: 528) recorded the South African distribution as only 'Port Nolloth to Port Elizabeth', as usual in conditions of heavy surf on rock, the presence of strong populations in Angola and of the acknowledged records from Port Nolloth, near the Namibia/South Africa border, would certainly suggest that any rocky areas with surf along the Namibian shore-line could also bear populations. See also the general note to Gelidium and that to Gelidium arbuscula Bory ex B0rgesen. Gelidium spp. Angola (298; 352). Ascension (474). Cameroun (337; 344; 393; 394; 533). Canaries (2; 5; 8; 66; 71; 214; 237; 351; 393; 394; 486; 493; 583). Cape Verde Islands (423). Cote d'lvoire (287; 290; 394). Ghana (42A; 287; 290; 297; 299; 338; 340; 344; 376; 393; 394; 487; 567). Guinee (344; 384; 393; 394; 529). Liberia (287). Mauritanie(349;529). Namibia (348; 522). Nigeria (344; 393; 394). Senegal (121; 123; 344; 393; 394; 411; 529). Sierra Leone (339; 344; 393; 394). Western Sahara (349; 393; 394; 395; 476). 'Tropical west Africa' (611). West Africa (290; 344). Gigartina acicularis (Roth) Lamouroux Bioko (346; 350; 586). Cameroun (172; 243; 269; 337; 344; 350; 447; 537; 586). Canaries (2; 38B; 38C; 38D; 128A; 172; 227; 253; 306B; 392; 7490; 584; 598; 604). Gambia (350; 586). Ghana (42A; 153; 269; 291; 335; 338; 340; 344; 350; 374; 537; 586). Guinee (344; 350; 537; 586). Liberia (350; 586). Mauritanie (38B; 38C; 38D; 349; 500). Salvage Islands (38B; 38C; 38D; 598). Sao Tome (93; 350; 586). Senegal (38B; 38C; 38D; 53; 55; 59; 344; 537). Sierra Leone (350; 586). Western Sahara (36B; 38C; 38D). '[Associations on] , , , African west coast' (374). RHODOPHYTA (FLORIDEAE) OF TROPICAL AFRICA 227 ' Atlantico (Inglaterra - Sudafrica . . .)'(253). 'Atlantik Kiiste von Afrika (. . . Golf von Guinea)' (270). 'Atlantique (de 1'Angleterre a la Mauritanie)' (33). 'British Isles to the Cameroons' (172). 'British Isles south to Cameroun' (243). 'from the British Isles . . . south to Cameroun' (447). 'Gran Bretana a Camerun' (604). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara].' (598). Tropical Africa (N. Gambia- Congo river)' (598). Tropical West Africa' (611). 'widespread in warm temperate and tropical seas' (350; 586). [As Gigartina acicularis (Wulfen) Lamouroux] Angola (352). Cameroun (173). Canaries (13; 16; 70; 108; 189; 191; 375; 379; 439; 441; 444; 489; 499; 517). Gambia (296). Ghana (173). Guinea-Bissau (7529). Guinee(?529). Liberia (129). Mauritanie (349; 529). Salvage Islands (215). Sao Tome (252). Senegal (529). Sierra Leone (30; 295). Western Sahara (349). 'Atlantico (de Inglaterra a Canarias)' (517). 'Atlantique: depuis les cotes anglaises jusqu'en Mauritanie' (222). 'Atlantique (de 1'Angleterre aux Canaries . . . Sans doute repandu dans toutes les mers chaudes.'(189). 'Cosmopolite/subcosmopolite' (529) . 'D'Angleterre aux Canaries' (89). 'de 1'Irlande et de la Grande-Bretagne au Cameroun' (178). 'From the English coast southwards to the Canary Islands' (70). 'Golf von Guinea' (270). 'Nordwestafrika' (499). 'Seems to occur in all warmer seas.' (63). [As Fucus acicularis Turner] Ghana (271). [As Sphaerococcus acicularis (Wulfen) C. Agardh] 'In mari Atlantico, ab oris Galliae septentrionalis ad Africam.' (588). Note. B0rgesen's (70) specimens were all small and sterile. The Hornemann (271) record from 'Danish Guinea' [present day Ghana] is attributed here on nomenclatural equivalence only, but there is no reason to doubt its presence in Ghana, since modern records are plentiful. The P. E. Isert collection examined by Hornemann needs examination for full confirmation, but the relevant material may have been destroyed at C [Copenhagen] during the 1807 fire that resulted in the loss of some Isert specimens. There may be a very early (1803) Canaries record concealed amongst the heterogeneity of material described under Fucus perforatus Bory (90: 305, pi. V, figs 1 A, B, C); see the entries for Carpococcus perforatus (Bory) J. Agardh and Laurencia perforata Montagne. Gigartina bracteata (S. Gmelin) Setchell & Gardner Namibia (36B;522A). Note. See 36B: 314 for discussion by Wynne of various morphological, taxonomic, and nomenclatural aspects. 228 J. H. PRICE, D. M. JOHN & G. W. LAWSON Gigartina burmannii (C. Agardh) J. Agardh See Gigartina stiriata (Turner) J. Agardh. Gigartina confervoides (L.) Lamouroux [or simply Lamouroux] See Gradlaria verrucosa (Hudson) Papenfuss. Gigartina cylindrica Despreaux Note. See Gradlaria dura (C. Agardh) J. Agardh. The entry in Montagne (401: 160) for Gigartina dura Desmazieres states under 'Hab.': 'Gigartina cylindrica Despr. in schedula.' Gigartina dura Desmazieres See Gradlaria dura (C. Agardh) J. Agardh. Gigartina elegans Greville in St-Hilaire See Gigartina teedii (Roth) Lamouroux. Gigartina fastigiata J. Agardh See Gigartina scutellata (Hering) Simons in Seagrief . Gigartina flagelliformis (Sonder) Sonder 'Ad oras Africae occidentales: Lenormand!' (326). Note. Neither the alga being referred to nor the location from which it came are clear, 'West Africa' being referred to. Sonder (1845: 55) originally described his taxon in Polyides and certainly the plant illustrated in Kiitzing's (326: pi. 5, no. 4030, figs c, d) 1868 presentation has much in common with Poly ides/ Furcellaria, rather than with Gigartina, where Sonder (1846) recombined it. The 1868 reference to the species is the first time that West Africa is referred to, all previous details concerning material being from 'Novam-Hollandiam'. Presumably Kiitzing received or saw Lenormand Herbarium material citing the African data. If Kiitzing's drawing is an accurate representation of the Lenormand material, then it has to be taken that the entry is generically mis-placed in Gigartina. In 1869, when Kiitzing again (pi. 18, figs c, d, no. 4248) employed the name Gigartina flagelliformis of Sonder, the plant illustrated appears closer to Gradlaria then to either Gigartina or Poly ides/ Furcellaria. Kutzing then stated (p. 7) lnec Kg. Tab. ph. XVIII. tab. 5. ' and cited only 'Novam Hollandiam', remarking that the structure was still nearer to Polyides or Furcellaria. J. Agardh (24: 283, no. 38) maintained the taxon in Gigartina, but commented that the internal structure was like Poly ides/ Furcellaria and cited only New Holland as location. De Toni (131: 227) commented 'Videtur forsan Rhabdoniae species'. Gigartina griffithsiae (Turner) Lamouroux [or simply Lamouroux] See Gymnogongrus griffithsiae (Turner) Martius. Gigartina perforata (Bory) De Toni [or (Bory) J. Agardh] See Carpococcus perforatus (Bory) J. Agardh. Gigartina pistillata (S. Gmelin) Stackhouse Canaries (70; 191; 227; 229; 306B; 375; 489; 490; 499; 517; 598). Mauritanie (245B; 252; 349; 500; 529). Senegal (59; 394). Western Sahara (349). 'Atlantico (de Inglaterra a Canarias)' (517). 'Atlantique (de 1'Angleterre au Rio de Oro)' (33). 'Atlantique: depuis les cotes britanniques jusqu'au Rio de Oro' (222). 'Atlantischer Ozean, von der englischen Kiiste an siidwarts bis zur nordwestafrikanischen.' (497). 'British Isles to South Africa' (172). 'depuis le sud de la Grande-Bretagne jusqu-au Senegal (Bodard, com. verb.)' (173). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). 'From the English coast southwards to the Canary Islands.' (70). [As Gigartina pistillata Stackhouse] 'de 1'Angleterre aux Canaries' (89). Note. The record by two of us (Lawson & John, 349) of this taxon from Western Sahara is based on Ardre RHODOPHYTA (PLORIDEAE) OF TROPICAL AFRICA 229 (33); we did not collect the alga there. B0rgesen (70) based his comments solely on the report by Bornet (89: 273); there were no specimens in Herb. Thuret, nor did he find it. Gigartina pygmaea Lamouroux See Chylocladia reflexa Lenormand in Desmazieres [Gastrodonium reflexum (Chauvin) Kiitz- Gigartina radula (Esper) J. Agardh Namibia (161; 348; 437; 500; 522; 522A; 523). Note. Unusually, with the exception of the purely secondary reports in Schmidt & Gerloff (500), all the above statements of record are at least partly based on original observation and collection. This taxon is also well known from South Africa (570) and from many parts of the Southern Ocean system. Gigartina scabiosa (Kiitzing) Papenfuss See Gigartina scutellata (Hering) Simons in Seagrief . Gigartina scutellata (Hering) Simons in Seagrief Namibia (522 A). [As Gigartina scabiosa (Kiitzing) Papenfuss] Namibia (348; 522). Note. Both the above recordings are based, at least in part, on original observations and collections. The species is also widely known in South Africa, whence it was first described; there are nomenclatural problems, the nature of which is clear from the entry on G. scutellata in Seagrief (570: 31). Clearly, Simons had decided that, despite contrasting opinions held elsewhere in regard to synonymy in at least parts of the application (and perhaps the concept) of the name Chondrus scutellatus Hering, it is in fact the same plant that is being referred to. If this is so, then the plant hitherto mostly known as Gigartina scabiosa (Kiitzing) Papenfuss has an earlier valid name that must be invoked. Hence, the recombination proposed by Simons of Sphaerococcus (Chondrus) scutellatus Hering (1841: 91) in Gigartina, resulting in the binomial used as heading to the present entry. The inclusion of these data in Seagrief s list (570: 31) was clearly based on information provided by Simons in a letter to Seagrief, doubtless with Simons's permission. He (the latter) presumably expected his paper on Trematocarpus (Sarcodiaceae) in southern Africa and the exclusion of Chondrus scutellatus Hering to have appeared before that of Seagrief. Despite the delay in publication of the Seagrief check-list (the correspondence with Simons was in 1982; Seagriefs paper did not appear until 1984) no further publication details of the Simons work were given, implying that it had not appeared by then. In view of both previous correspondence and proximity, it is likely that Seagrief would have had his attention drawn by Simons to its appearance. We have searched carefully for any sign of the publication but have been unable to trace it. Hence, the recombination is cited as above, details provided by Seagrief having been adequate to validate Simons's proposed changes. Seagriefs suggested synonymy of G. scutellata included Gigartina fastigiata J. Agardh [non Postels et Ruprecht]. Gigartina stellata (Stackhouse in Withering) Batters See Mastocarpus stellatus (Stackhouse in Withering) Guiry in Guiry, West, Kim & Masuda. Gigartina stiriata (Turner) J. Agardh Namibia (348; 437; 522; 522A; 523). Note. A further, previously separate, taxon in Gigartina. G. burmannii (C. Agardh) J. Agardh, has been shown to be the tetrasporophyte of G. stiriata. Gigartina teedii (Roth) Lamouroux Angola (500). Cape Verde Islands (589; 598). Guinea-Bissau (529). Mauritanie (7344; 349; 7529). Namibia (36B; 161 ;522A). Senegal (47; 50; 52; 55; 59; 399; 529; 531). Western Sahara (349). 'British Isles to Angola . . . Cape Verde Islands' (172). 'Nordwestafrika' (499). 'Subtropical Africa [Senegal (N. of Gambia), Mauritania, Former W. Sahara]' (598). 'Westafrika' (499). 230 J. H. PRICE, D. M. JOHN & G. W. LAWSON [As Gigartina teedii Lamouroux] Cape Verde Islands (38). Senegal (38; 122; 408). [As Gigartina teedii Roth] Cape Verde Islands (259). [As Gigartina teedii J. Agardh] Angola (41; 42). Note. There seems no real evidence for the presence of this species in the Canary Islands; Dangeard (117; 118) specifically stated that it was missing from there. A certain number of expressed uncertainties have been ignored here - Bodard (47), for instance, recorded what may well have been the taxon as 'Gigartina aff. teedii', and Dangeard (122), more verbosely, commented for Senegal 'un Gigartina a ramification assez regulierement pennee, a branches et rameaux aplatis, qui se presente, soit a 1'etat sterile, soit porteur de nombreux cystocarpes . . . rapelle beaucoup le G. teedii . Askenasy's record was rationalised from probabilities, not fully recorded; he (38) stated, 'n'ont pas ete rapportees des iles du Cap Vert; il est tres probable qu'elles y croissent aussi et qu'on les trouvera plus tard'. Ardre (33: 130-131) stated the southern limit of the G. teedii distribution range to be Morocco. On the subject of the relationships between G. teedii and Gigartina elegans Greville in St.-Hilaire, see Cordeiro-Marino (108: 77-78). For general background, see Guiry (589). Gigartina spp. Cameroun(393;394). Ghana (344; 393; 394). Guinee(393;394). Mauritanie(393;394). Namibia (348; 438; 522; 611). Nigeria (393; 394). Senegal (393; 394). Sierra Leone (393; 394). Note. The records from Namibia could well include any (or a mixture) of Gigartina radula, G. scutellata, or G. stiriata (q.v.) in addition to other unidentified forms. Ginnania furcellata Montagne Note. See Scinaia forcellata Bivona-Bernardi. The genus Ginnania is based on material from the Canaries, published by Montagne (401: 162), repeated elsewhere by the same author (402) and subse- quently by others (e.g. Kiitzing, 318). Montagne later (408) cited the species with the addition '(Turn, sub Viva).' According to Papenfuss (434: 281), the report of Ginnania furcellata from the Cape of Good Hope (Harvey, 354: pi. 69) related to Pseudogloiophloea (now Scinaia} capensis. This agrees with the absence of records more southerly than the Congo for any species of Scinaia in this area, 5. capensis never having been reported. It is possible that the latter occurs in Namibia, and perhaps Angola. Gloiophloea verae Dickinson See Scinaia verae (Dickinson) Huisman. Goniolithon Afonso-Carrillo et al. (582: 25) accepted the characterization of this genus in the original (Foslie, 1898a) sense, as proposed by Cabioch (1972). All statements emanating from Afonso- Carrillo and/or colleagues must therefore be viewed in that light. The status of the genus Goniolithon Foslie has been explained by Johansen (1981: 218); Foslie's original concept of 1898 was later rejected by its author (1900