6 May 1994 Aust. Syst. Bot., 7,89-124 Morphology and Taxonomy of Caloglossa (Delesseriaceae, Rhodophyta) R. J. in^^ and C. F. puttockAB *school of Biological Science, University of New South Wales, Sydney, NSW 2052, Australia. B~resentaddress: Australian National Herbarium, CSIRO Division of Plant Industry. GPO Box 1600, Canberra, ACT 2601, Australia. Contents Abstract ...................... ............................................................................................................. 89 Introduction ................ .......... ..........,,,.....................,,...............,,,...........................................90 90 Materials and Methods ........................................................................................................... Morphological Features ....................................... ............,....................................................9 1 Taxonomic Treatment ............................. . .............................................................................. 99 100 Caloglossa J. Agardh ................................................................................................. Key to Species.................................................................................................................. 101 1. Caloglossa bengalensis (G. Martens) R. J. King et Puttock ...................................102 2. Caloglossa adhaerens R. J. King et Puttock ............................................................ 103 3. Caloglossa triclada ( E . Post) R. J. King et Puttock ................................................106 4. Caloglossa ogasawaraensis Okamura .....................................,.............,.................107 5. Caloglossa beccarii (Zanardini) De Toni ................................................................ 109 6. Caloglossa stipitata E. Post ..................................................................................... 1 0 7. Caloglossa leprieurii (Montagne) J. Agardh ......................................................... 112 8. Caloglossa continua (Okamura) R. J. King et Puttock ............................................. 1 5 8a. Caloglossa continua subspecies continua ............................................... ... . . . .116 8b. Caloglossa continua subspecies axillaris R. J. King et Puttock ............................... 117 8c. Caloglossa continua subspecies saigonensis (Tanaka et Pham-Hohng HA) R. J. King et Puttock ............................................................................................ 1 7 118 8d. Caloglossa continua subspecies postiae R. J. King et Puttock ................................ Conclusion .........................................,. ... ..... .... ...... .................... .... ..,. .. . .1 18 119 Acknowledgments ................................................................................................................ Index to Genera, Species and Synonyms .............................................................................. 119 References ......................... ... ............,..................................,...............,,..............................120 Abstract The genus Caloglossa J. Agardh is revised and eight species are recognised. One new species is described (C. adhaerens R. J. King et Puttock, sp. nov.), one new combination is made (C. bengalensis (G. Martens) R. J. King et Puttock, comb. nov.), and two forms of C. leprieurii are raised to specific status (C. triclada (E. Post) R. J. King et Puttock, stat. nov. and C. continua (Okamura) R. J. King et Puttock, stat. nov.) In Caloglossa continua four subspecies are recognised: C. continua ssp. continua; C. continua ssp. avillaris R. J. King et Puttock, ssp. nov.; C. continua ssp. postiae R. J. King et Puttock, ssp. nov.; and C. continua ssp. saigonensis (Tanaka et Pham-Hokng HB)R. J. King et Puttock, stat. nov. Detailed descriptions, taxonomic notes and a key are provided for these taxa and the previously described species that we recognise in the genus (C. beccarii (Zanardini) De Toni, C. leprieurii (Montagne) J. Agardh, C. ogasawaraensis Okamura and C. stipitata E. Post). Caloglossa leprieurii var. angusta Jao is a synonym of C. continua ssp. saigonensis, and C, adnata f. divaricata E. Post is included within C. leprieurii. 1030-1887/94/020089$05.00 R.J. King and C.F. Puttock Introduction The genus Caloglossa J. Agardh is a distinctive member of the family Delesseriaceae. It differs from all other Delesseriaceae except Taenioma J. Agardh in that it exhibits exogenous primary branching. It differs from Taenioma on the basis of its lack of subapical pigmented monosiphonous filaments (Papenfuss 1944), monopodial rather than sympodial growth (see Papenfuss 1944, fig. 7), and the production of cortication from the tetrasporangial stalk cells on the dorsal and ventral surfaces. Caloglossa has been placed either with Taenioma in the Caloglossa Group, an informal taxonomic category roughly equivalent to a Tribe (Papenfuss 1961; see also Wynne 1983), or constitutes the sole member of the Caloglossa Group of Kylin (1956). The presence of exogenous branches precludes membership in the Claudea Group, where both Caloglossa and Taenioma were placed previously (Papenfuss 1937). In 1936, Post sumrnarised the status of the genus in her major revision of the systematics and biogeography of macroalgae associated with mangroves. This was followed by an extensive examination of the morphology and ecology of Caloglossa (Post 1943). Her later publications on the distribution and ecology of mangrove algae (for summaries see Post 1963, 1966b) added little taxonomic or systematic information, and even papers on particular species are essentially biogeographic notes (e.g. Post 1965, 1 9 6 6 ~ ~ Post ). (1936) recognised six species, and although she did not provide detailed descriptions of the taxa, considerable information can be extracted from her synoptic key (Table 1). In addition to the six species, she recognised two forms of C. adnata (forma typica and divaricata) and two varieties of C. leprieurii (variety hookevi and variety leprieurii-with five forms: typica, pygmaea, ceylonensis, continua and triclada). Since then, only three new taxa within Caloglossa have been described (C. leprieurii var. angusta Jao (1941), C. saigonensis Tanaka et Pham-Hohng H6 (1962), and C. ogasawaraensis var. latifolia Kumano (1978)), although these names have not been applied widely. On the basis of the characters used by Post (1936), the first two would be placed in C. leprieurii and the latter into C. beccarii. The present paper reassesses the species in the genus Caloglossa and recognises eight species. These species are described in detail, and a key based on vegetative morphology is provided. The taxa are therefore morphological or 'form' species, regardless of geographic or genetic isolation. The aim of the current paper is to establish clearly these morphological entities as the basis for further studies on genetic and geographic differentiation of populations. There is an extensive literature associated with the genus, and a concerted effort has been made to refer to all major papers that have contributed to its taxonomy. Species of Caloglossa are common and widely distributed in the eulittoral and uppermost sublittoral zones on sheltered and semisheltered coasts, from tropical to cool-temperate waters. They occur on a variety of substrata but are typically epiphytes on saltmarsh and mangrove plants, where, with Bostrychia, Catenella and Stictosiphonia they form the 'Bostrychietum' or 'Bostrychia-Caloglossa Assoziation' (Post 1936; King and Puttock 1989). Caloglossa also occurs epilithically on fully marine coasts but is more common in estuaries near the lower limits of salinity influence. It is also recorded in permanent, freshwater habitats (e.g. Entwisle and Kraft 1984; Seto and Jao 1984; Sheath et al. 1993). The capacity of Caloglossa to grow in a range of salinities, and to cope with fluctuating salinities, has been investigated (Yarish et al. 1979a,b, 1980; Mann and Steinke 1988) and attributed to both ecotypic variation and physiological tolerance. Fischer (1984) considered the swelling of the cell wall as the protoplast shrinks to be a protective measure against plasmolysis, and this aspect has been further investigated by Mostaert and King (1993). Karsten et al. (1992) reported that mannitol was a compatible solute in Caloglossa (see also West et al. 1992). Materials and Methods This revision is based on the examination of fresh, liquid preserved and herbarium material. Type material of all but Caloglossa bengalensis and C. stipitata has been examined. The types of these two Taxonomy of Caloglossa 91 species have not been located in any herbarium where they might have been expected to be found. Microscope preparations were made from field-collected material fixed in 4% glutaraldehyde-seawater, or from small fragments of dried herbarium material rehydrated in a detergent solution. Slides were stained using toluidine blue and mounted in corn syrup (Karo) solution. A selection of the herbarium specimens examined is listed in the species accounts. Specimens collected by, or for, us are lodged in the John T. Waterhouse Herbarium (UNSW); specimens in other herbaria are cited using registered acronyms following Holmgren et al. (1981). A representative collection of microscope slides has been deposited in UNSW. All drawings were made by CFP with the aid of a camera lucida. Light micrographs were made on a Leitz Orthoplan photomicroscope using Kodak technical pan 2415 rated at 50 ISO. Morphological Features The features used by Post (1936 et seq.) to characterise the species of Caloglossa and to distinguish between them were drawn entirely from vegetative material. Thirteen characters have been recognised from the key to the species in Post (1936) (Table 1) although some of these are interdependent. In addition, other authors have used the position of the rhizoids. Fertile material, which in field-collected specimens is uncommon, has not been used as a key character. In the present paper the following terminology has been adopted: the 'main axis' for the indeterminate primary row of cells; the 'blade' for the thallus between nodes; and 'wing' for the monostromatic lateral derivatives of the pericentral cells. Apical Growth Thallus growth is initiated by a single apical cell which cuts off a series of cells by transverse divisions. Post (1936) considered the ratio of the height of the cells to their width as being taxonomically useful (Table 1, Character I), but our observations contradict this. In all species, the width is greater than the height, although in type material of Caloglossa ogasawaraensis and C. bombayensis the ratio approaches unity. There is such considerable variation even within single collections that detailed analyses seem unwarranted. In general, the breadth of the apical cell is greater in plants of greater thallus width. The number of cells behind the apical cell before the formation of pericentral cells (Table 1, Character 2) is very variable within taxa, and provides no useful information. The pattern of pericentral cell formation follows that of the Delesseriaceae. Wing formation follows that of the Hypoglossum Group, being initiated from the lateral pericentral cells with production of secondary and tertiary cell rows (Fig. 1). Occasionally the marginal cells divide, and Papenfuss (1961) referred to such cells as fourth-order branches. Node/Internode Formation and Blade Morphology The regular constriction of the thallus in some species of Caloglossa, including the type of the genus, C. lepriezirii, gave rise to the notion of a nodehnternode system. Post (1936) and others have used this as a diagnostic character for species (Table 1, Character 3), but our examinations indicate that it is less useful than superficial observation suggests (Fig. 2). The axial cell from which an exogenous branch arises can be regarded as the nodal cell in all species of Caloglossa. The overall appearance of the thallus is the result of the degree of constriction at the nodes. Where there is essentially no constriction, as in C. adnata sensu Post, the main and side branches were described as congenital or fused (Table 1, Character 4) (Fig. 2d,e). This feature, in conjunction with the length-to-breadth ratio of the youngest developed internode (Table 1, Character 5), contributes to the blade shape (Table 1, Character 6). The usefulness of blade morphology is questionable given the range within species (e.g. see Fig. 2), although Post (1936) and other authors have recognised various forms of C. leprieurii on this basis alone. In some species, wing cells are not developed on the main axis in the nodal region, and the blade then appears to have a 'petiole' or short stalk (Table 1, Character 7). Post (1936) referred to this as a stipe, but the use of the term in this way is inappropriate given its more general use in the Ceramiales (e.g. in Bostryclzia and Stictosiphonia). Table 1. Characters and their states in the key to the genus Caloglossa (Post 1936) \O N Where observations in the present study differ from those of Post they are noted in square brackets C. adnata sensu Post C. lepn'eurii C. ogasawaraensis C. bornbayensis C. beccarii C. stipitata width > length width = length [width > length] width < length [width = length] width < length [width = length] width < length [width > length] width < length [width > length] from 4th [2-71 from 7th 11-71 from 9th [5-81 from 9th [5-81 from 9th 12-41 from 9th 12-41 not constricted constrictedA [or not] not constricted [constricted] not constricted [constricted] constricted constricted 4. Main and side branches congenital [= fused] independent [or fused] independent [or fused] independent [or fused] independent [or fused] independent [or fused] 5. Length:breadth of youngest internodesB equal 10.81 Character 1. Proportions of apical cell 2. Cells from the apex before blade differentiation 3. Nodes 6. Blade shape 7. 'Stipe' (blade petiole) x3 [2.5] linear lanceolate-elliptical lanceolate linear lanceolate elliptical on adventitious rosettes only on adventitious rosettes onlyA [and on exogenous branches] not recorded [on any blade] not recorded [on any blade] at all internodes at all internodes monopodial [and pseudodichotomous] monopodial [and pseudo dichotomous^ Yes [various] yes [various] Yes [usually] Yes [usually] less less less less 8. Branching monopodial monopodial [and or pseudodichotomous pseudodichotomous] pseudodichotomous [monopodial] monopodial [and pseudodichotomous] monopodial [and pseudodichotomous] E 2 W 9. Thallus differentiated Yes into long and short shoots [various] 10.Relativedevelopment of not recorded side and main branches [sub-equal] Ba 0 7 equal [sub-equal] s w 8 c3 Table 1. Continued Character 8 C. adnata sensu Post C. leprieurii C. ogasawaraensis C. bornbayensis C. beccarii C. stipitata 3 i v 1 1.Number of first order lateral branches 1 12,Adventitiousrosettes wndogenous branching] 2 4 blades per rosette, or absent [absent] 13.Proventive branches [Adventitious branching] linear 2-loA or absent [and dorsal] irregular on the dorsal side [lateral] irregular on the dorsal side [lateral] irregular on the dorsal side [lateral] irregular on the dorsal side [lateral] lanceolate-elliptical lanceolate [linearj linear lanceolate elliptical A~nternal inconsistencies in the key. B ~ r o m type material. g 3 R.J. King and C.F. Puttock Fig. 1. Lateral blade structure in Caloglossa showing development of secondand third-order cell rows from the first-order cell row (main axis). Occasionally fourth-order cell rows (sensu Papenfuss 1961) result from division of marginal cells (not shown). lpc, lateral pericentral cell; lo, firstorder cell row; 2", second-order cell row; 3",third-order cell row. Exogenous Branching and Branching Patterns Exogenous branching develops from an oblique division of a sub-apical cell in the uniseriate apical region; as Nageli (1855) showed, the branching development is monopodial. These lateral indeterminate branches arise alternately. Post (1936), however, contrasted this general branching pattern to that in Caloglossa leprieurii, which she considered to be pseudodichotomous (Table 1, Character 8) (Fig. 2a). Her interpretation was that, as a result of confusing appearance and development, there is almost equal development of the indeterminate lateral branches producing a pseudodichotomous thallus. Post (1943, fig. 12) illustrated specimens of C. adnata sensu Post which likewise exhibit a pseudodichotomous thallus form (see also Fig. 2d,e). Post (1936) used the presence of a thallus differentiated into long and short shoots to distinguish Caloglossa adnata sensu Post from C. leprieurii (Table 1, Character 9). By inference she placed the remaining four species in the group with long and short shoots. There is, however, no underlying difference in the thallus plan derived from exogenous branching in these taxa. All species recognised by Post (1936), except C. leprieurii, have forms with short and long shoots. A related character is the development of the main axis and the side branches. It appears from Post (1936) that short and long shoots are recognised simply by the dominance of the main axis (Table 1, Character 10). In all specimens of all species examined, we observed only one sub-apical exogenous branch, although Post (1936) recorded one or two in some species (Table 1, Character 11). In these cases she appears to have confused later-formed adventitious branches with exogenous branches. Endogenous and Adventitious Branching In addition to exogenous branches, both endogenous and adventitious branches develop in Caloglossa (Fig. 3). Endogenous branches are derived from the indeterminate axis and arise from the first axial cell above the nodal cell. Of the species considered by Post (1936), only C, leprieurii exhibits this (Fig. 3b). Repeated endogenous branching, in which successive Taxonomy of Caloglossa 95 branches always arise from the first axial cell of the preceding endogenous branch, results in the formation of a cluster of branches. This was interpreted by Post (1936) as being an adventitious rosette (Table 1, Character 12). Post (1936, 1943) reported the same phenomenon in some specimens of C. adnata sensu Post, but those she illustrated showing this feature (Post 1943, fig. lq, r, t, u, x) are here considered to be C. leprieurii. Adventitious branches arise from cells other than those of the indeterminate axis and form in two ways: a) from a lateral pericentral cell of the first axial cell above the node on the main axis (Fig. 3c), with successive branches near the node developing from the lateral pericentral cell of the first to third axial cells of the branches; or b) from wing cells at the margin (Fig. 3 4 . It is the former to which Post (1936) referred in relation to the irregular adventitious (proventive) branches on the dorsal surface of Caloglossa ogasawaraensis, C. bornbayensis, C. beccarii and C. stipitata (Table 1, Character 13). Fig. 2. Variation in thallus form in Caloglossa. Scale: 1 mm. a-c: range of form in C. leprieurii: (a) a relatively unconstricted form (s.n.); (b) an intermediate form as shown by type material (Leprieur s.n.); (c) a form in which the nodelinternode system exhibits marked constrictions at the nodes (s.n.). d,e: variation in C. adnata: (d) a form with so-called short and long shoots (UNSW 18121);( e ) a pseudodichotomous form (UNSW 20983). Blade Characters Not Used by Post (1936) Considerable emphasis in the older literature is given to descriptions of thallus morphology. In an attempt to quantify blade morphology, a number of characters was measured. The number of cell rows on any given axial cell (i.e. second- and third-order cell rows) that reach the margin at the midpoint of the blade shows considerable overlap between R.J. King and C.F. Puttock Fig. 3. Modes of branching in Caloglossa. (a) C. continua ssp. axillaris (UNSW 18166), showing the lateral swelling (arrow) of the subapical cell in the early development of an exogenous branch. Scale 40 pm. (b) C. leprieurii (UNSW 22022) with an endogenous branch arising from the first axial cell above the node. Scale 100 pm. (c) C . ogasawaraensis (B@rgesen5208), showing adventitious branching developed from the lateral pericentral of the first axial cell above a node. Scale 100 pm. (d) C. adhaerens (Tanaka slide #193), showing the development of an adventitious branch from marginal wing cells. Scale 100 pm. species and is not diagnostic. The width of the internode, expressed as the number of cells in the second-order branches from the central axial filament to the blade margin is, as already noted, so highly variable that it cannot be used to distinguish between species. One of the generic characters emphasised by Pappenfuss (1944) is that all second- and third-order cell rows reach the margin; this is the case except at the point of exogenous branch divergence. At nodes, the transition from a single blade to two blades is accommodated by the reduction in the number of third-order cell rows produced by the nodal cells or the axial cells immediately above the node, or a reduction in the development of cells in these rows so that the second- and/or third-order cell rows do not reach the margin. The extent to which this occurs is particularly useful to distinguish species in that group of species with non-constricted nodes. In two species (Caloglossa leprieurii and C. triclada), the pericentral cell on the adaxial side is not developed at all. The terminology used in this paper is defined in Fig. 4. Taxonomy of Caloglossa Fig. 4. Branch terminology employed in the monograph. ab, abaxial; ad, adjacent; ax, adaxial; i, internode; la, lateral axis; ma, main axis; n, node; op, opposite. The angles of branching have been measured in an attempt to quantify a major determinant of thallus shape (Fig. 5). Consistent ranges of these angles provide support for the separation of Caloglossa bengalensis (C. adnata sensu Zanardini) and C. adhaerens (C. adnata sensu Post). Cortication The vegetative thallus of Caloglossa is essentially ecorticate, although there are two exceptions: the proliferation of cells associated with the nodal rhizoids in C. leprieurii and C. triclada where the pencentral cells of the node cell and the proximal pencentral cell of each branch divide irregularly to form a cushion of cells on the lower surface, and regions of the main axis adjacent to procarps and cystocarps in all species. Internal hyphae, analogous to those described in Bostrychia (King and Puttock 1989), are conspicuous in older parts of the thallus in Caloglossa continua ssp. axillaris. These are relatively long and narrow cells running within the main axis and arising from axial cells. Rhizoids Although most authors have noted multicellular rhizoids in Caloglossa, there has been no attempt to use them in distinguishing species or species groups. We recognise five arrangements of rhizoids that are distinct and consistent within taxa (Fig. 6). Except in the case of C. adhaerens, the rhizoids are clustered near the nodes, although there are differences in detail between species. In Caloglossa leprieurii and C. triclada, rhizoids develop from cortical cells formed from the nodal pencentral cell and the two pencentral cells immediately above the node. Papenfuss (1961) referred to these cortical cells as rhizoid initials, but this terminology is perhaps inappropriate since such cells sometimes divide further to produce the stipe ('St-chen' of Post 1936,1943). R.J. King and C.F. Puttock Fig. 5. Cell-row and branch-divergence angles measured in Caloglossa, here shown in C. adhaerens (UNSW 21329). Scale 100 pm. (a) the angle (a)formed by the firstformed third-order cell rows and the main axis, and the angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral. (b) the divergence angle (6) of the exogenous lateral branch from the prolongation of the main axis. la, lateral axis; ma, main axis. ( c ) the angle (0) occupied by cells of an exogenous lateral branch immediately above the node. In C. leprieurii, therefore, the rhizoids are very closely associated with the node and its corticating cells, where they generally coalesce to form the stipe. On a normally bifurcating thallus the stipe begins as a small swelling below the node, and develops few rhizoids. After attachment to the substratum, the stipe progressively corticates by irregular cell divisions. It may reach several rnillimetres in length and support an independent rosette of blades, the formative bifurcating thallus having long since degenerated. In Caloglossa continua, the rhizoids develop in a defined axillary cluster (Fig. 6b) directly from wing cells of the main axis and the lateral exogenous branches. There is no associated cortication and the rhizoids remain discrete. In Caloglossa stipitata, C. ogasawaraensis (including C. bornbayensis) and C. beccarii, the rhizoids develop from the pericentral cells of the node and the two pericentral cells directly above the node (Fig. 6c). In contrast to the case in C. leprieurii, however, these pericentral cells do not divide further to form a cortex, and the rhizoids do not coalesce. In Caloglossa bengalensis (= C. adnata sensu Post), the clusters of rhizoids arise directly from wing cells adjacent to the axis and generally near the nodes (Fig. 6 4 . Taxonomy of Caloglossa 99 Caloglossa adhaerens is the only species in which the rhizoids are not clustered and not clearly associated with the node or main axis. In this species the rhizoids arise from blade cells and are scattered on the ventral surface (Fig. 6e), although none occurs near the margins, except where the blade tissue has been damaged. Fig. 6. Distribution of rhizoids in Caloglossa. (a) C. leprieurii (Leprieur s.n.) with rhizoids clustered at the nodes, arising from the nodal pericentral cell and the two pericentral cells above the node and generally coalescing to form a stipe. Scale 1 mm. (b) C. continua ssp. axillaris (UNSW 18166), with rhizoids in an axilliary cluster developing from wing cells of the main axis and lateral exogenous branches. Scale 300 ym. (c) C. stipitata (UNSW 21334), with rhizoids as in C. leprieurii but the rhizoids not coalescing. Scale 1 mm. (4 C. bengalensis (UNSW 19550), with rhizoids arising from wing cells near the main axis and mostly near nodes. Scale 1 mm. (e) C. adhaerens (UNSW 21329), with scattered rhizoids arising from blade cells. Scale 1 mm. Taxonomic Treatment Caloglossa was used first by Hooker and Harvey (1845a,b)for C. hookeri, but as there was no description this is considered to be a nomen nudum. Hooker and Harvey (1845a,b) could not find enough character states to separate Caloglossa from Delesseria, but later Harvey (1853)proposed Caloglossa as a subgenus of Delesseria, following Agardh (1852) who had used Caloglossa as a subdivision of unspecified rank within the genus Delesseria. In 1849, Kiitzing transferred 100 R.J. King and C.F. Puttock D. leprieurii Montagne to Hypoglossum. Subsequently Martens (1869), in dealing with Hypoglossum, discussed Harvey's subgenus Caloglossa in a brief description of C. leprieurii P subtilissima. It is our opinion that Martens (1869) did not intend to raise Caloglossa formally to generic rank, and in later papers (Martens 1870, 1871a,b) he continued to use Kiitzing's genus Hypoglossum for these taxa, as did other authors (e.g. Zeller 1873; Reinsch 1875). Zanardini (1872b) was even more conservative, treating Caloglossa as synonymous with the genus Delesseria, transferring H. vieillardii to it as D. vieillardii, and describing two new species, D. adnata and D. beccarii. In 1876, Agardh formally proposed the genus Caloglossa, with the two species Caloglossa leprieurii (Montagne) J. Agardh and C. mnioides Harvey ex J. Agardh, but made no reference to Zanardini (1872b). All subsequent authors have attributed the genus to Agardh (1876) rather than Martens (1869), since Martens did not provide a generic description of Caloglossa, and furthermore continued to use Hypoglossum. Caloglossa J . Agardh Caloglossa J. Agardh 1876: 498. Lectotype: C. leprieurii (Montagne) J. Agardh 1876: 498. De Toni 1900: 728. Okamura 1908: 180; 1936: 793. Post 1936: 45. Fan 1952: 1. Kylin 1956: 398. Post 1957: 105. Taylor 1960: 544. Mori 1961: 19. Joly 1965: 211. Pham-Hohng HB 1969: 247. Lawson and John 1982: 302. Bourrelly 1985: 274. Delesseria pro parte J. Agardh 1852: 680,682. Delesseria subgenus Caloglossa Harvey 1853: 98. Thalli consisting of a polysiphonous, monopodial, ecorticate indeterminate axis, bearing alternate, exogenous, pseudodichotomous indeterminate lateral branches; four pericentral cells, produced in the delesseriaceous sequence, first the two lateral cells, then the ventral or rhizoid-bearing side, followed by the dorsal side; the lateral pair of pericentral cells functioning as the first cells of the second-order cell rows which produce strictly organised cortication forming monostromatic wings (flanges, alae or blades). Transverse divisions of the pericentral cells and intercalary cell divisions in the second-order cell rows are lacking. Only the inner cells of the second-order cell row produce third-order cell rows. These thirdorder cell rows are produced by oblique abaxial divisions of the second-order cells, i.e. on the posterior side of the parent second-order cell, followed by divisions more-or-less parallel to the main axis and synchronous with other rows of this order. The second- and third-order cells become multinucleate by formation of secondary pit-connections. In the internodal region, i.e. between exogenous branches, all second- and third-order cell rows reach the margin. Margins entire, flat, not revolute or convolute, constricted at the nodes in some species. The angle of divergence of the indeterminate lateral branch from the main axis is variable, the main axis also being deflected from the original direction. The transition of a single flat blade to two blades is accommodated by the obliteration of, and the reduction of, the number of second- or third-order cell rows produced by the axial cells in the immediate vicinity of the node. Multicellular rhizoids are scattered on the blades or variously organised into clumps at or near the nodes. Endogenous and adventitious proliferations of three types are present: emergent at the nodes, 'intercalary' near the nodes, and marginal. These proliferations often bear reproductive sori. Tetrasporangial sori develop on the main thallus or on proliferating blades directly from second- or third-order cells. There may be a reduction in the number of third-order rows produced within sori. The cells of both second- and third-order rows become fertile, cutting off a tetrasporangial parent cell on the adaxial (anterior) side of the second- or third-order cell rows. The residual stalk cell then cuts off cover cells in the dorsal and ventral planes which do not produce further tetrasporangia. Sporangia mostly have tetrahedrally arranged tetrads (occasionally cruciate or decussate), and are rarely bisporic or unisporic. Spores are released through linear slits in the cuticle on the dorsal side of the blade. Spermatangial sori develop on both surfaces of the blade, covering the whole of the blade with the exception of the margin cells, some lateral cells and all transverse pericentral cells. Taxonomy of Caloglossa 101 One or occasionally two cortical cells are cut off from the blade cells (usually two for the larger inner cells and lateral pericentral cells). These cortical cells divide again, initially on the abaxial side by anticlinal walls, into a branched chain, forming a layer of cortical cells. Any superficial cell may be spermatogenous and produce several spermatangia. The side directed towards the substratum is less well developed than the upper side. One cortical cell is cut off by periclinal divisions on the dorsal and ventral side of the wing cells for the smaller wing cells and two for the larger wing cells close to the axis. This description differs from that of Pappenfuss (1961), who considered that C. leprieurii follows the pattern described by Womersley and Shepley (1959) for the Sarcomenioideae. The so-called patterns of development appear to be related to cell size rather than having any taxonomic significance. This is detailed in the account for Caloglossa continua (as C, leprieurii f. continua) given by Tanaka (1992) and in that for C, ogasawaraensis by Tanaka and Karniya (1993). Carpogonial branches are borne on the first-order (axial) cell row, usually in series although only one is fertilised. The fertile pericentral (supporting) cell divides by an anterior division to form a cover cell. The supporting cell then cuts off a carpogonial branch initial which undergoes further divisions to form a four-celled carpogonial branch that terminates with a more-or-less forwardly directed trichogyne. Following fertilisation, the supporting cell cuts off an auxiliary cell which fuses with the carpogonium via a connecting cell. These cells, together with the fertile axial cell, appear to form the fusion complex from which arise several, much branched, gonimoblast filaments bearing terminal, subspherical to ovoid carposporangia in short chains. At maturity, the cystocarps are subspherical, ostiolate, and subsessile on the parent blade. The pericarp is formed by the lateral pericentral cells which produce 8-10 filaments, each of which produces further cortication of several cell layers. Distribution Widespread in tropical to cool-temperate marine and fresh waters; epiphytic on trunks, prop roots and pneumatophores of mangroves; on stems of herbs and shrubs in saltmarshes; and epilithic on rocky shores, in freshwater streams, often in crevices and caves. Key to the Species of Caloglossa Branching of main axis exogenous, otherwise adventitious branching but only from cells of the wing margin ............................................................................................................................... 2 Branching of main axis exogenous, otherwise adventitious (new blades arising from pericentral cells in the same plane as the thallus) and/or endogenous (new blades arising from the first axial cell above the node and out of the plane of the thallus, usually on the upper surface) ....3 Rhizoids grouped together, arising from clusters of wing cells above the nodes and adjacent to the polysiphonous axis ........................................................................................ C . bengalensis Rhizoids scattered, arising from isolated cells of the wings ......................................C. adhaerens Branching both adventitious and endogenous, on a single thallus (known only from Mauritius) .. .................................................................................................................................... C. triclada Branching adventitious or endogenous, never both on the same thallus ....................................... 4 Branching adventitious (C. beccarii complex) ............................................................................5 Branching endogenous .................................................................................................................. 7 Thallus internodes mostly less than 0.33 mm broad (2-8 wing cells wide) .....C. ogasawaraensis Thallus internodes usually more than 0.7 mm broad 0 1 4 wing cells wide) ................................6 Thallus internodes narrowly elliptical, blades not markedly constricted at nodes; internodes usually 5.5-8.3 mm long, 5-7 lateral second- and third-order cell rows produced by each C. beccarii axial cell .................................................................................................................... Thallus internodes broadly elliptical, blades markedly constricted at nodes; internode length usually 2.4-3.5 mm long, 4 or 5 lateral cell rows produced by each axial cell ........C. stipitata R.J. King and C.F. Puttock First lateral adaxial pericentral cell absent; rhizoids arising from transverse and lateral pericentral cells of the nodal and first axial cells of the exogenous branches; rhizoids coalescent, developing in tandem with a corticated stipe .......................................C. leprieurii First lateral adaxial pericentral cell present, forming a short series of wing cells, the rhizoids arising from these cells; rhizoids discrete, divergent and not forming a stipe (C, continua complex) ...................................................................................................................................8 Thallus internodes usually more than 0.8 mm broad (>I4 wing cells wide)................................9 Thallus internodes usually less than 0.5 mm broad ( 4 4 wing cells wide) ................................10 Thallus internodes usually 0.8-1.6 mm broad (14-25 wing cells wide), primary cell row hyphae absent .....................................................................................C continua ssp. continua Thallus internodes usually 1.8-2.9 mm broad (30-36 wing cells wide), primary cell row hyphae present in mature thalli ...........................................................C continua ssp, axillaris Thallus internodes 4.2-5.8 mm long, usually 3 lateral rows produced by each axial cell, adaxial cell row slightly reduced ...............................................................C. continua ssp, saigonensis Thallus internodes 1.4-2.2 mm long, usually 4 or 5 lateral rows produced by each axial cell, adaxial cell row markedly reduced ...................................................C. continua ssp. postiae 1. Caloglossa bengalensis (G. Martens) R. J. King et Puttock, comb, nov. Type: Lower Bengal, Mutlah, Kurz 2024 (not located, presumably B destroyed). Neotype: INDIA: Pegu, Elephant Point, xii. 1868, Kurz 3266 (L! isoneotype BM!). Hypoglossum bengalense G. Martens Proc. As. Soc. 39, 258 (1870) basionym. Zeller 1873: 193. Srinivasan 1965: 247. Krishnamurthy and Joshi 1970: 24. Delesseria bengalensis (G. Martens) Prain 1905: 331 as D. bengalense. Delesseria adnata Zanardini 18726: 141, pl. 5B, figs 1-3, syn. nov. Type: MALAYSIA: Sarawak, without date, Beccari (holotype FI!). Caloglossa adnata (Zanardini) De Toni 1900: 730. Islam 1976: 61, fig. 339. Caloglossa adnata f. typica E. Post 1936: 46. Thalli prostrate, closely adpressed to the substratum and occasionally imbricate forming a mat-like weave; pink to pale brown, epiphytic. Attached by clusters of rhizoids that arise from second- and third-order cells in more-or-less linear patches of 6-10 cells which arise from the first three axial cells above the node on the main axis; rhizoids also scattered singly or in groups of several adjacent cells further from the axes; mature blades linear to narrowly elliptical, 0.9-1.4 mm wide with 0.8-1.2 mm between nodes; stipeless; adventitious branches formed from initials on the margin near the blade divergences; endogenous branches absent. Internode morphology: the inner (2-)3(-4) second-order cells produce third-order rows, thus on one side of the blade from any one axial cell the apical cells of (3-)4(-5) rows form the blade margin, with 14-34 cells per row to the margin decreasing to 17-35 cells per row at the nodes. Second- and third-order cell rows close to the margin rarely divide to produce single fourth-order cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 54-78", and the average angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral is 10" (see Fig. 5 for definitions of cell angles). The length of axial cells in the mature thallus is 140-190 pm. Node morphology: the angle of divergence of the lateral branch from the main axis is 60-90"; the lateral branch axis is deflected by 46-62" from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle o,Fig. 5c) is 37-4g0(-54"). At the node, on the main axis, all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin and the number of branches is the same as, or one fewer than, that in the internode region. On the lateral axis the first axial cell on the adaxial side produces only a second-order cell row, usually only 1-1 1 cells long and not extending to the margin; the second axial cell produces one or occasionally two rows and these sometimes do not extend to the margin. On the abaxial side, the nodal axial cell produces only one row, 7-10 cells long and not extending to the margin; the first three axial Taxonomy of Caloglossa 103 cells of the lateral branch produce one row each, the lower two usually not extending to the margin. By the fifth axial cell above the node the regular internodal condition is attained. Rhizoids arise from groups of blade cells on the main branch immediately above an exogenous branch (first to third axial cell) in the region of the second to fifth cells from the axis (inner blade), the 10th to 15th cells (mid blade) and on the lateral branch associated with the fifth and sixth axial cell and second to fifth cells from the axis, and few internodal mid blade cells; rhizoids absent from the outer blade. The rhizoids are 0.35-0.60 mm long, 3 0 4 5 pm in diameter, comprising 4-6 cells; the last may be branched. Adventitious blades are formed from the blade margin in the nodal region by 'marginogenous' proliferations arising from any cell terminating lateral exogenous branch second- or third-order rows. At later stages of development several neighbouring cells corticate and rhizoids are formed basally from several cells; these new plantlet-forming blades are not connected to the main axis by a false vein; adventitious blades often produce tetrasporangial son. Tetrasporangia, spermatangia and carpogonia unknown. Distribution Epiphytic on mangrove trunks, prop roots and pneumatophores, and Nipa palms. Restricted to south-east Asia. Cultured material of this species from Daintree River, north Queensland and from Singapore, is in the collection of M. Kamiya, Tsukuba University. Selected Specimens Examined (7 Specimens examined.) BRUNEI-DARASSALEM: Kota Batu, 28.iii.1989, Booth UNSW 19550. INDIA: Pegu, Elephant Point, xii.1868, Kurz 3266 (BM, L). MALAYSIA: Sarawak, [Kuching] without date, Beccari (F). Sungei, Tebrau, Johore, 13.vi.1963, Sinclair 10810 (L). Sementa, Selangor, 31.vii.1985, Sasekumar UNSW 18121. Taxonomic Notes Many algal collections made by Kurz are available in European herbaria. We have searched un~uccessfullyfor the holotype of Hypoglossum bengalense, which may have been destroyed at Berlin during World War 11. A second collection (Kurz 3266) was referred to by Zeller (1873) as H. bengalense. This was likely to have been determined by Martens, but he died before his checklist was published by Zeller (1873). Despite the absence of the holotype, we have decided to place Caloglossa adnata in its synonomy rather than regard H. bengalense as a nomen dubium. 2. Caloglossa adhaerens R.J. King et Puttock, sp. nov. Type: AUSTRALIA: Ballina, NSW, 22.viii.1983, King & Wheeler UNSW 15020. Alga membranacea prostrata substrato arcte adhaerens; ramificatio endogena et constrictiones ad nodos absentes; ramuli adventitii ex cellulis in margine laminarum orientes. Affinis C. bengalensi sed rhizoideis dispersis in alis non prope secusve axes principales aggregatis differt. Membranous prostrate alga adhering closely to the substratum, lacking endogenous branching and nodal constrictions; adventitious branchlets produced by cells at the margin of the blades. With affinity to C. bengalensis but differing in possessing scattered rhizoids on the wings, not in clusters near or along the main axes. Caloglossa adnata sensu Post 1936: 47, fig. 1. Borgesen 1937: 342; 1938: 267. Post 1943: 124, pl.1, 2, 19, 22, 23, 31, 35, 36, 38. Dawson 1954: 453, fig. 58b; 1956: 57. Post 1957: 105, fig. 3a-e, 4, 5; 1963: 99. Pham-Hoing H8 1969: 250, fig. 2.180. Saenger, Specht, Specht and Chapman 1977: 318. Fortes and Trono 1979: 61. King and Wheeler 1985: 102. Tanaka and Chihara 1985: 42, figs 1,2. Silva, Meiies and Moe 1987: 60. Tanaka and Chihara 1988a: 98, fig. 9; 1988b: 31, fig. 1-2-8. 104 R.J. King and C.F. Puttock Thalli prostrate, closely adpressed to the substratum and occasionally imbricate forming a mat-like weave; pink to pale brown, epiphytic and epilithic, attached by solitary rhizoids on the ventral side which arise at random from second- and third-order cells; mature blades (0.9-)1.7-2.3(-3.1) mm wide, with 0.9-2.9 mm between nodes; stipeless; adventitious branches formed from initials on the thallus margin; endogenous branches absent. Internode morphology: the inner (1-)2-3(-4) second-order cells produce third-order rows, so that on one side of the blade from any one axial cell the apical cells of (2-)3-4(-5) rows form the margin. There are 1 6 4 8 cells in the rows, with little reduction in the rows at the nodes; with 17-40 cells per row. Second- and third-order cell rows close to the margin rarely divide to produce single fourth-order cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 47-70", and the angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral is -2 to 9" (see Fig. 5 for cell angle definitions). The length of axial cells in the mature thallus is 120-210(-260) pm; the pericentral cells are of the same length as the axial cells. Node morphology: the angle of divergence of the lateral branch from the main axis is 59-83"; the lateral branch axis is deflected by 34-51" from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle u, Fig. 5c) is 54-60'. At the node on the main axis, all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin and the number of branches is the same as, or one fewer than, that in the internode region. On the lateral axis the first axial cell on the adaxial side produces only a second-order cell row, usually only 3-6(-10) cells long and not extending to the margin; the second axial cell produces one or occasionally two rows, and these generally extend to the margin. On the abaxial side, the nodal axial cell produces one or two rows which extend to the margin; the first three axial cells of the lateral branch produce one or two rows each. By the fifth axial cell above the node, the regular internodal condition is attained. Rhizoids arise from isolated or occasionally adjacent second- and third-order cells, never from the pericentral cells nor from the first or second cells from the central axis. Most rhizoids are in the region of the third to sixth cells from the axis (inner blade), the 10th-18th cells (mid blade), and 22nd-25th cells from the axis (outer blade). The rhizoids are 0.25-1.0 rnm long, 30-50 ym in diameter, and comprise 3-6 cells, the last of which may be branched. Adventitious blades are formed in the internodal region of the blade by 'marginogenous' proliferations arising from any cell terminating a second- or third-order row (Fig. 3 4 . At later stages of development apparently several neighbouring cells corticate and rhizoids are formed basally from several cells; these new plantlet-forming blades are not connected to the main axis by a false vein. Often these adventitious blades produce reproductive sori. Tetrasporangial sori are 12-15 axial cells long, with little or no reduction in third-order rows (1-4 rows) from the vegetative state; they are 5-7 row cells wide, often include the lateral pericentral cells, and are frequently associated with exogenous branching. Tetrasporangia are 80-90 pm in diameter at maturity. Distribution Epiphytic on mangrove trunks, prop roots and pneumatophores, and Nipa palms in areas with strong water flow and low turbidity. Restricted to south-east Asia, the western Pacific and northern Australia to 28" S. Selected Specimens Examined (15 Specimens examined.) AUSTRALIA: Ballina, NSW, 22.viii.1983, King & Wheeler UNSW 15020. Weipa, 22.vii.1984, King & Puttock UNSW 17023. BRUNEI-DARASSALEM: Puiau Barambang, 25.i.1990, Booth UNSW 21329. PAPUA NEW GUINEA: Riwo Village, Madang, 3.v.1988, King UNSW 20983. VIETNAM: station 12, Dawson 11392 (US, BISH, AHFH, Nha Trang). Taxonomy of Caloglossa 105 Taxonomic Notes This widely distributed species has, until now, been referred to Caloglossa adnata, especially by Post (1936 et seq.). Examination of the type of C. adnata reveals that it has rhizoids which form clusters associated with the nodes, as Zanardini (1872b) noted. Plants of this morphology are now referred to C. bengalensis. Caloglossa adhaerans is superficially similar to C. bengalensis but can be distinguished on a number of characteristics (Table 2). Caloglossa bengalensis differs from C. adhaerens in that the distribution of rhizoids is in clusters, the cell line of the nodal axial cell on the abaxial side is always compacted and never reaches the margin, and the angle of deflection of the lateral branch at the node (angle 6, Fig. 5b) is generally less than 36". The divergence of the seventh cells from the axis is greater than in C. adhaerans, and lateral branches are slightly constricted. Table 2. Morphological variation in Caloglossa adhaerans and Caloglossa bengalensis - Character Caloglossa adhaerans (measurements based on 11 specimens) - Caloglossa bengalensis Delesseria adnata Malaysia, Beccari (holotype) Number of cell rows produced per axial cell Hypoglossurn bengalense Pegu, Kurz 3266 (neot~~e) UNSW 18121 Selangor UNSW 19550 Brunei (3-)4(-5) (3-)4(-5) 3-5 Number of cells in 2nd order rows at internodes 26-34 20-28 16-24 14-22 Angle of divergence of lateral branch from main axis 60-79" 80-90" 75-83" 62-7 1" Row divergence (angle a) Angular divergence (angle P) Lateral branch deflection (angle 6) 54-69" 59-78" 64-76" 63-75" 10-15" 4-17" 5-13" 10-18' 48-62" 58-62" 53-61" 46-55" Angle of blade defined by branching (angle o) 42-48' 43-47" 45-48(-54)' 37-43" Adventitious branching from internodal margins near nodal margins near nodal margins near nodal margins near nodal margins? Rhizoid distribution scattered on wing cells in small groups, scattered on wing cells near nodes in small groups, scattered on wing cells near nodes in small groups, scattered on wing cells near nodes in small grows, scattered on wing cells near nodes 106 R.J. King and C.F. Puttock This species is clearly identical to that described by Tanaka and Chihara (1985) as Caloglossa adnata, although we have found that the acceptable variation within the species is greater than they described. Generally the width of the thallus is 1.7-2.3 mrn in the typical form, but the range is from very much narrower forms 0.9-1.3 mm wide to much broader forms 2.7-3.0 mrn wide. Such a degree of variation is similar to, or less than, that exhibited in other species such as C. leprieurii and C, beccarii. There are also two growth forms, a so-called monochasial (zigzag) form where the lateral branches are slower growing than the main axis, and a dichasial form where the lateral and main axes develop more or less equally, although the pattern has little direct effect on the angles of divergence of the lateral-main branches. 3. Caloglossa triclada (E. Post) R.J. King et Puttock, sp. et stat. nov. Lectotype: MAURITIUS: ArsBnal, 1868, Pike s.n. (ex Dickie BM!, isolectotype L 935329.26!) Alga membranacea arcuata rhizoideis ex cellulis pericentralibus in nodis orientibus. Differt a C. leprieurii et C. beccarii ramificatione adventitia et endogena simul praesenti. Membranous arching alga with rhizoids arising from the pericentral cells at the nodes. Differing from C. leprieurii and C. beccarii in the presence of both adventitious and endogenous branching on individual thalli. Delesseria leprieurii pro parte Dickie 1875: 193. Caloglossa leprieurii var. leprieurii f. triclada E. Post 1936: 47 (in clavi), 53. Thalli arching, loosely attached to the substratum and imbricate, forming a mat-like weave, reddish brown to pale brown, epiphytic, attached by a cluster of rhizoids on the ventral side that arise from cortication of pericentral cells at the nodes; mature blades linear to narrowly elliptical, 0.6-1.2 mm wide with 1.8-3.7 mm between moderately to weakly constricted nodes; stipeless; adventitious branches formed from lateral pericentral initials; endogenous branches formed from axial cells above the node. Internode morphology: the inner 5 or 6 second-order cells produce third-order rows, thus on one side of any blade from any one axial cell the apical cells of 6 or 7 rows form the blade margin, with 12-18 cells per row decreasing to 4-10 cells per row at the node. Second- and third-order cell rows close to the margin do not divide transversely to produce single fourth-order cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 65-77", and the average angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral is 12" (see Fig. 5 for cell angle definitions). The length of axial cells in the mature thallus is 120-210 ym. Node morphology: the angle of divergence of the lateral branch from the main axis is 58-70"; the lateral branch axis is deflected by 30-39" from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle o, Fig. 5c) is 59-67'. At the node, on the main axis, all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin and the number of branches is 2 or 3 compared with 6 or 7 in the internode region. On the lateral axis the first axial cell on the adaxial side produces no second-order cell row; the second axial cell produces one or occasionally two rows and these sometimes do not extend to the margin. On the abaxial side, the nodal axial cell produces only one row, not extending to the margin; the first three axial cells of the lateral branch produce 2-4 rows each, all extending to the margin. By the fourth axial cell above the node the regular internodal condition is attained. Rhizoids arise from a cortical pad that develops by the division of the lateral and ventral pericentral (second-order) cells of the node and first anterior axial cells. The rhizoids are 0.5-0.8 mm long, 35-50 ym in diameter, comprising 6-10 cells; the last may be dendriform. Endogenous blades are formed in the nodal region of the blade by a single proliferation of the first axial cells of the lateral branch above the node (occasionally the first axial cells of the main branch). The new blade-forming cell lies between a dorsal and lateral pericentral cell and the new blade develops out of the of the thallus. This new blade produces further blades which arise from the basal cell of successive new blades in a sympodial fashion. Taxonomy of Caloglossa 107 Adventitious blades are formed in the nodal region of the blade by a single interstitial proliferation arising laterally from the lateral pericentral cells of the first or second axial cell above a node. The production of sori on these adventitious blades is not known. Tetrasporangia, spemzatangia and carpogonia unknown. Distribution Epiphytic on mangrove trunks, prop roots and pneumatophores in areas with moderate water flow and low turbidity. Known only from Mauritius. Selected Specimens Examined (2 Specimens examined.) MAURITIUS: Arsbnal, 1868, Pike s.n. (BM, L). River Profunde, foot of Cornisard Mt, viii.1869, Pike s.n. (BM). Taxonomic Notes There are at least two independent collections from Mauritius made by Pike between 1868 and 1870. The specimen at BM, originally from Dickie's herbarium, is chosen as the lectotype. No corresponding specimen has been seen from L (see Post 1963: 50). A second specimen (R. Profunde), cited by Post (1936: 56) as Caloglossa leprieurii var. hookeri, is also C. triclada, but in this material there are fewer adventitious branches. Caloglossa triclada has both endogenous branching, otherwise restricted to the C. leprieurii and C. continua complex, and adventitious branching arising from the pericentral cells, as found in the C. beccarii complex. The species was recognised as a form of C. leprieurii by Post (1936). It appears to have characteristics that are intermediate between C. leprieurii and C. beccarii. The nodal anatomy is similar to that in C. leprieurii and it also produces endogenous branches. Although the rhizoids coalesce, the formation of a stipe from coalescing rhizoids, as in C. leprieurii, has not been observed. Caloglossa beccarii and related species have not been recorded for Mauritius. 4. Caloglossa ogasawaraensis Okamura 1897: 13, figs A-D. Type: JAPAN: Ogasawara-jima (Bonin Islands), iii.1879, Yatabe (TI). De Toni 1900: 730. Okamura 1902: 51; 1903a: exsicc. 68; 1903b: 130; 1908: 183, 185, pl. 37, figs 1-11; 1916: 60. De Toni 1924: 358 (in synonymy of C. zanzibarensis K.I. Goebel). Okamura 1936: 794. Post 1936: 60. Skuja 1938: 631, pl. 35; 1944: 54, pl. 6, figs 11-16. Tseng 1945: 163, pl. 1, fig. 7. Fan 1952: 1, figs 4,21,22,24. Post 1957: 114. Segawa and Kamura 1960: 59. Mori 1961: 19, figs I, 3-6,II, 1. Mori and Ikeda 1961: 225. Umezaki 1962: 366. Post 1963: 104; 1966c: 317, fig. 3; 1968: 273. Pham-Hohg H8 1969: 251, fig. 2.178. Ugadim 1976: 139, figs 63, 64, 66, 68-73, 82. Saenger, Specht, Specht and Chapman 1977: 318. Fortes and Trono, 1979: 62. Lawson and John 1982: 304, pl. 48, fig. 3. Schnetter and Bula Meyer 1982: 166, pl. 30, fig. A, pl. 31, fig. B. Soto 1982: 251 (as C. ogassuarensis Skuja orthographic error). Yoshizaki, Iura, Miyaji and Kasaki 1983: 191, fig. 3. Bourrelly 1985: 274, pl. 77, figs 7, 8, pl. 78. Hadlich and Bouzon 1985: 94, figs 20-22. King and Wheeler 1985: 103. Seto 1985: 317, figs 1-3. Silva, Mefies and Moe 1987: 60. Tanaka and Chihara 1988a: 98, fig. 11; 198827: 32, fig. 1-2-11. King 1990: 60. West 1991: 460, figs 4-8. Tanaka and Karniya 1993: 113. Delesseria leprieurii var. polychotoma Okamura ms. (fide Seto 1985: 317) Delesseria leprieurii var. subfasciculata Okamura ms. Vide Seto 1985: 317) Delesseria zanzibariensis K.I. Goebel 1898: 65, figs 1-6. Type: TANZANIA: Moos aus Bach Kibaoni, nordlich von Kokotoni, [Zanzibar], Stuhlmann 102 (holotype LD 31878!-labelled 'original specimen'; isotypes BM!, C! , L!) Caloglossa zanzibariensis (K.I. Goebel) De Toni 1900: 731; 1924: 357. Caloglossa bombayensis Boergesen 1933: 127, figs 10-12. Type: INDIA: Malabar Hill, Bombay, Borgesen 5208 (BM!, C!). Post 1936: 61. Fan 1952: 1, figs 1-3, 7-9, 17-20, 23, 25-35. Post 1957: 115; 1963: 107. Saenger, Specht, Specht and Chapman 1977: 3 18. 108 R.J. King and C.F. Puttock Thalli arching, loosely attached to the substratum and imbricate forming a mat-like weave, reddish brown to pale brown, epiphytic, attached by several rhizoids on the ventral side that arise from pericentral cells at the nodes; mature blades linear, 0.12-0.33(-0.55) mm wide with 1.3-3.8 mm between strongly constricted nodes; stipeless; adventitious branches formed from lateral pericentral initials; endogenous branches absent. Internode morphology: the inner 1-3(-4) second-order cells produce third-order rows, thus on one side of any blade from any one axial cell the apical cells of 2-4(-5) rows form the blade margin, with 2-8 cells per row decreasing to 1-5 cells per row at the nodes. Second- and third-order cell rows close to the margin may divide transversely to produce single fourth-order cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 73-83"; the second cell rows from the axis are parallel to it. The length of axial cells in the mature thallus is 60-140(-160) pm. Node morphology: the angle of divergence of the lateral branch from the main axis is 67-75"; the lateral branch axis is deflected by 47-78' from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle o,Fig. 5c) is 25-33". The blade is reduced to one to several cells wide in the vicinity of the node. At the node, on the main axis, all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin and the number of branches is the same as, or one or two fewer than, that in the internode region. On the lateral axis the f i s t to third axial cells on the adaxial side produce only a second-order cell row, usually only 1 or 2 cells long and extending to the margin. On the abaxial side, the nodal axial cell produces two rows, 2 or 3 cells long, extending to the margin; the first three axial cells of the lateral branch produce one row each, extending to the margin. The regular internodal condition is not attained for several more cells above this. Rhizoids arise from the ventral and lateral pericentral (second-order) cells of the node and ventral pericentral cells of the first and second anterior axial cells. The rhizoids are 0.7-0.8 mm long, 25-45 pm in diameter, comprising 3-10 cells; the last may be dendriform. Adventitious blades are formed in the nodal region of the blade by interstitial proliferations arising laterally from the lateral pericentral cells of the first to third axial cell above a node. At later stages of development the parental pericentral cell produces several rhizoids. Often these adventitious blades produce sori. Tetrasporangial sori are 6-18 axial cells long arising from the second-order cell row only, 1 or 2 cells wide, and include the lateral pericentral cells and the submarginal cells. Dorsal and ventral cover cells are not formed. Tetraspores are 35-48 pm in diameter at maturity. Spermatangia and carpogonia are recorded by West (1991) and described in detail by Tanaka and Kamiya (1993). Distribution Epiphytic on mangrove trunks, prop roots and pneumatophores in areas with slow water flow and low turbidity. Occurs throughout south-east Asia, West Africa, the western Pacific and eastern Australia to 34' S. Selected Specimens Examined (15 Specimens examined.) AUSTRALIA: 'North Australia' Damael (L 941311.121). Careel Bay, 3.ix.1983, Wheeler UNSW 15052. Cape Melville, 1l.vii.1984, King & Puttock UNSW 16849. CHWA: Shantin, tidal cove, April [?1940], Tseng 2727. Shantin, tidal cove, July [?1940], Taam A145. INDIA: Bombay, Bqrgesen 5208 (BM, C). JAPAN: rivermouth of Kojahagi-gawa, Mikawa, vii.1902, Okamura 68 (BM, C , L). TANZANIA: Zanzibar, Stuhlmann s.n. (BM, C, L, LD). Taxonomic Notes Okamura (1903b) placed Caloglossa zanzibariensis in synonomy with C. ogasawaraensis. Tseng (1945) amalgamated C. ogasawaraensis (including C. zanzibariensis) with C. bornbayensis, noting that supposed differences between them could be the result of differing Taxonomy of Caloglossa 109 environmental conditions. Borgesen in a personal communication with Tseng, agreed as do we, that morphologically they are essentially identical. Goebel (1898), in his description of Delesseria zanzibariensis (= Caloglossa zanzibariensis), noted that it may be a form of C. amboinensis (= C. beccarii), but we distinguish it from C. beccarii on the basis of blade width. In this revision we recognise three closely related species, C. ogasawaraensis, C. beccarii and C. stipitata, essentially on the basis of blade morphology (Table 3). Whether this is valid, especially in view of the documented variability in other species, requires detailed investigation. Seto (1985) considered Okamura's manuscript names Delesseria leprieurii var. polychotoma and D. leprieurii var. subfasciculata to apply to specimens of C. ogasawaraensis. Table 3. Morphological variation in Caloglossa beccarii complex Character C. ogasawaraensis C. beccarii C. stipitata Internode width (mm) 0.12-0.33(-0.55) 0.7-1.3(-1.9) 1.1-2.1(-2.4) Internode length (mm) 1.3-3.8 (2.2-)5.5-8.3 2.4-3.5(-4.5) Number of cell rows produced per axial cell Number of cells in 2nd order rows at internodes Length of axial cell (pm) 60-140(-160) 130-230(-260) 120-150(-160) Angle of divergence of lateral branch from main axis 67-75" 74-8 1" 48-66" Row divergence (angle a) 73-83" 76-83" 73-82" 0" 16-25" 12" 25-33" 26-30" 52-60" Angular divergence (angle P) Lateral branch deflection (angle 6) Angle of blade defined by branching (angle o) 5. Caloglossa beccarii (Zanardini) De Toni 1900: 730. Type: MALAYSIA: super lapides torrentis Sodomae ad pedes montis Gunong Poe, Sarawak, Beccari s.n. (lectotype Fi!, isolectotype Fi!, L!). De Toni 1924: 358. Post 1936: 64; 1957: 117; 1963: 113; 1965: 184. Dixit 1968: 22. Womersley and Bailey 1970: 327. Schnetter and Bula Meyer 1982: 166, pl. 30, fig. C. Delesseria beccarii nom. nud. Zanardini 1872a: 387. Delesseria beccavii Zanardini 187217: 140, pl. 5A. Delesseria amboinensis G. Karsten 1891: 265, pl. 5. Lectotype (here chosen): INDONESIA: Amboina I. [Ambon], ix-x.1889, Karsten (L 935274.13!). Caloglossa amboinensis (G. Karsten) De Toni 1900: 731. Weber-van Bosse 1923: 390. Jadin 1934: 167. Caloglossa ogasawaraensis var. latifolia Kumano 1978: 103, fig. 5. syn. nov. Type: MALAYSIA: Sungei Cherok, Perak, 4.v.1971, Kumano s.n. (Kobe U ! ) . Thalli arching, loosely attached to the substratum and imbricate forming a mat-like weave, reddish brown to pale khaki green, epiphytic, attached by clusters of rhizoids on the ventral side that arise from pericentral cells at the nodes; mature blades linear to narrowly elliptical, 0.7-1.3(-1.9) mm wide with (2.2-)5.5-8.3 mm between strongly to moderately constricted nodes; stipeless; adventitious branches formed from lateral pericentral initials; endogenous branches absent. 110 R.J. King and C.F. Puttock Internode molphology: the inner (4-)5,6 second-order cells produce third-order rows, thus on one side of any blade from any one axial cell the apical cells of (4-)5-7 rows form the blade margin, with 14-18(-21) cells per row decreasing to 2-6 cells per row at the nodes. Secondand third-order cell rows close to the margin do not divide transversely to produce single fourth-order cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 76-83", and the angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral is 16-25" (see Fig. 5 for cell angle definitions). The length of axial cells in the mature thallus is 130-230(-260) ym. Node morphology: the angle of divergence of the lateral branch from the main axis is 74-81"; the lateral branch axis is deflected by 4 3 4 8 " from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle o, Fig. 5c) is 26-30". At the node, on the main axis, all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin and the number of branches is the same as, or one or two fewer than, that in the internode region. On the lateral axis, the first to third axial cells on the adaxial side produce only a second-order cell row, usually only 1 or 2 cells long and extending to the margin. On the abaxial side, the nodal axial cell produces two rows, 2 or 3 cells long, extending to the margin; the first three axial cells of the lateral branch produce one row each, extending to the margin. The regular internodal condition is not attained for several more cells above this. Rhizoids arise from the ventral pericentral (second-order) cells of the node and first anterior axial cells. The rhizoids are 0.6-3.5 mrn long, 3 0 4 5 ym in diameter, comprising 8-12 cells; the last may be dendriform. Adventitious blades are formed in the nodal region of the blade by interstitial proliferations arising laterally from the lateral pericentral cells of the first to third axial cell above a node. At later stages of development the parental pericentral cell produces several rhizoids. Often these adventitious blades produce sori. Tetrasporangia, spermatangia and carpogonia unknown. Distribution Epilithic on stones in fresh water coastal streams and epiphytic on mangrove trunks, prop roots and pneumatophores in areas with moderate water flow and high turbidity. Restricted to south-east Asia (India, Malaysia, Indonesia), the western Pacific and northern Australia to 28" S. Selected Specimens Examined (8 Specimens examined.) AUSTRALIA: Forest Creek, Tributary of the Daintree River, Sept.1991, King & Puttock UNSW 19624. BRUNEI-DARASSALEM: Tutong R., 25.i.1990, Booth UNSW 21388. INDIA: Calcutta Botanic Gardens, R. Hooghly, 23.vii.1936, Biswas s.n. (L). INDONESIA: Ambon, vii-xi.1913, Robinson 2396 (L). Ambon, vii-xi.1913, Robinson 2405 (L). Arnbon, 1922, Jensen s.n. (C). Irian Jaya: Noord R., Sabang-Geihen, 19.vii.1907, Lorenz 5 (L). Taxonomic Notes Post (1936) placed Caloglossa amboinensis into C. beccarii. We have also placed C. ogasawaraensis var, latifolia into synonomy, as it is also inseparable on morphological grounds. 6. Caloglossa stipitata E. Post 1936: 62, fig.4. Syntypes: ad rupes littoreas vel radices arborum ad Nova Guinea occidentalis [Naumann], Sumatra serpentrionalis [Troll (Belawan)], Amboina [Troll from Negerilama, Roemahtiga, Roehoe, Lateri and Seri] et insulas Am, [Troll from Wakoea]. Post 1957: 116; 1963: 112; 1966a: 109, fig. 1. Pham-Hoing H6 1969: 250, fig. 2.181. Schnetter and Bula Meyer 1982: Taxonomy of Caloglossa 111 166, pl. 30, fig. D, E, pl. 31, fig. C, D. Tanaka and Chihara 1988a: 98, fig. 12; 1988b: 32, fig. 1-2-10. West 1991: 463, figs 9, 10. Thalli arching, loosely attached to the substratum and imbricate fonning a mat-like weave, reddish brown to pale brown, epiphytic, attached by a cluster of rhizoids on the ventral side that arise from pericentral cells at the nodes; mature blades elliptical to suborbicular, 1.1-2.1(-2.4) mm wide with 2.4-3.5(-4.5) mm between strongly constricted nodes; stipeless; adventitious branches formed from lateral pericentral initials; endogenous branches absent. Internode morphology: the inner 3 or 4 second-order cells produce third-order rows, thus on one side of any blade from any one axial cell the apical cells of 4 or 5 rows form the blade margin, with 18-22 cells per row decreasing to 1-2(-3) cells per row at the node. Second- and third-order cell rows close to the margin do not divide transversely to produce single fourthorder cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 73-82", and the average angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral is 12" (see Fig. 5 for cell angle definitions). The length of axial cells in the mature thallus is 120-150(-160) pm. Node morphology: the angle of divergence of the lateral branch from the main axis is 48-66"; the lateral branch axis is deflected by 32-54" from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle o,Fig. 5c) is 52-60". The blade is reduced to one to several cells wide in the vicinity of the node. At the node, on the main axis, all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin and the number of branches is the same as, or one or twifewer than, that in the internode region. On the lateral axis the first to third axial cells on the adaxial side produce only a second-order cell row, usually only 1 or 2 cells long and extending to the margin. On the abaxial side the nodal axial cell produces two rows, 2 or 3 cells long, and extending to the margin; the first three axial cells of the lateral branch produce one row each, extending to the margin. The regular internodal condition is not attained for several more cells above this. Rhizoids arise from the ventral pericentral (second-order) cells of the node and first anterior axial cells. The rhizoids are 0.8-1.5 mm long, 25-35 pm in diameter, comprising of 6-12 cells; the last may be dendriform. Adventitious blades are formed in the nodal region of the blade by interstitial proliferations arising laterally from the lateral pericentral cells of the first to third axial cell above a node. At later stages of development the parental pericentral cell produces several rhizoids. Often these adventitious blades produce sori. Tetrasporangial sori are 8-12 axial ;ells long of 2 or 3 rows, 7-9 row-cells wide and usually include the lateral pericentral cells and the submarginal cells. Cover cells are not formed. Tetraspores are 45-60 pm in diameter at maturity. Spermatangia and carpogonia unknown. Distribution Epiphytic on mangrove trunks, prop roots and pneumatophores in areas with moderate water flow and low turbidity. Restricted to south-east Asia, the western Pacific and northern Australia to 28" S. Selected Specimens Examined (5 Specimens examined.) AUSTRALIA: Beach at Cape York, 18.vii.1984,King & Puttock UNSW 16895. BRUNEI-DARASSALEM: Pulau Barambang, 25.i.1990, Booth UNSW 21334. SINGAPORE: Jurong R., 1896, Ridley 6926a (BM); Tanjorg Penura, 30.iii.1936, Hassan s.n. (BM). Taxonomic Notes The type specimen has not been located in this study, but we have seen specimens provided by Troll to other herbaria that are annotated by E. Post as C. stipitata. 112 R.J. King and C.F. Puttock 7. Caloglossa leprieurii (Montagne) J. Agardh 1876: 499. Type: FRENCH GUIANA: nr Cayenne, without date, Leprieur 356 & 362 (?PC, isolectotypes (see note below) L 939285.13!, LD 31909-30912!). Hauck 1888: 461. Cramer 1891: 1, pl. 1-3. De Toni 1900: 729. Howe 1902: 149. Borgesen 1911: 46, fig. 4. Okamura 1916: 60. Collins and Hervey 1917: 116. Borgesen 1919: 341, figs 338, 339. Howe 1920: 564. De Toni 1924: 357. Weber-van Bosse 1923: 390. Borgesen 1933: 127. Post 1936: 49. Borgesen 1937: 342. Taylor 1937: 345, pl. 53, figs 2, 3. Borgesen 1938: 267. Jao 1941: 274. Taylor 1950: 140 (= Hypoglossum caloglossoides Wynne et Kraft). Fan 1952: 1, figs 5, 6, 10-16. Dawson 1956: fig. 59 (= Hypoglossum caloglossoides). Kylin 1956: 398, fig. 312, AE. Post 1957: 105. Segawa and Kamura 1960: 59. Taylor 1960: 544, pl. 68, fig. 1. Pappenfuss 1961: 8, figs 1-30. Post 1963: 99. Joly 1965: 212, pl. 44, figs 559-562. May 1965: 395. Post 1968: 274. Pham-Hohng Hi3 1969: 249, fig. 2.179. Pocock 1969: 27. Womersley and Bailey 1970: 327. Islam 1976: 61, fig. 339. Ugadim 1976: 138, figs 80, 81. Saenger, Specht, Specht and Chapman 1977: 318. Fortes and Trono 1979: 61, fig. 7. Davey and Woelkerling 1980: 59. King 1981: fig 12.7 C, D. Beanland and Woelkerling 1982: 98. Lawson and John 1982: 304, pl. 48, figs 1, 2. Schnetter and Bula Meyer 1982: 167, pl. 30, fig. B, pl. 31, fig. A. Yoshizaki, Iura, Miyaji and Kasaki 1983: 191, fig. 2. Bourrelly 1985: 274, pl. 76, 77, figs 1-6. Hadlich and Bouzon 1985: 92, figs 4-19. King and Wheeler 1985: 102. Silva, Mefies and Moe 1987: 60. Tanaka and Chihara 1988a: 98, fig. 10; 1988b: 31, fig. 1-2-9. King 1990: 60. Delesseria leprieurii Montagne 1840: 196, pl. 5, fig. 1. Hooker et Harvey 1845b: 541. Montagne in dlOrbigny 1846: 658. Montagne 1850: 285. J. Agardh 1852: 680, 682. Harvey 1853: 98, pl. 22, fig. C. Dickie 1875: 193. Goebel 1897: 436. Hypoglossum leprieurii (Montagne) Kiitzing 1849: 875. Nageli 1855: 69, pl. 8. Kiitzing 1866: 4, pl. 10, figs d-h. Martens 1870: 259; 1871b: 469. Zeller 1873: 193. Reinsch 1875: 55. Caloglossa leprieurii var. leprieurii f. typica E. Post 1936: 47, 51. Caloglossa hookeri nom. nud. J.D. Hooker and Harvey 1 8 4 5 ~270. : Caloglossa leprieurii var. hookeri E. Post 1936: 53. Type: NEW ZEALAND: Bay of Islands, without date, Hooker (lectotype (here chosen) BM!). Borgesen 1937: 342; 1938: 267, fig. 2; 1945: 26. Taylor 1960: 544, pl. 68, fig. 1. Hadlich and Bouzon 1985: 92, figs 11-13. Tanaka and Chihara 1985: 42, figs 3,4. Silva, Mefies and Moe 1987: 60. Tanaka and Chihara 1988a: 98; 1988b: 32. Delesseria mnioides Harvey ms Harvey exsiccate 33 (1855). Delesseria mnioides (Harvey ex J. Agardh) Prain 1905. Caloglossa mnioides Harvey ex J. Agardh 1876: 500. Type: In Oceano pacific0 [Tonga], without date, Harvey 33 (lectotype (here chosen) BM!, isolectotypes L!, LD!, MICH!, NSW!). De Toni 1900: 729. Delesseria opuntioides J. Agardh 1854: 109. syn. nov. Type: SENEGAL: Dakar, Guinea, without date, Le Normand (LD 31920!). Hypoglossum vieillardii Kiitzing 1863: 95. Lectotype (Wynne and Kraft 1985): NEW CALEDONIA: [Wagap, 18631, Vieillard 2026 (L!). Kiitzing 1866: 4, pl. 10, figs a-c. Martens 1869: 237; 1871b: 469. Reinsch 1875: 55. Delesseria vieillardii (Kiitzing) Zanardini 1872b: 141. Caloglossa vieillardii (Kiitzing) Setchell 1924: 161. Caloglossa leprieurii var. P subtilissima G. Martens 1869: 234,237. Syntypes: INDIA: saltlakes, Calcutta, [ l l.xi.1868,I Kurz 1677,1680 (not located). Hypoglossumpygmaeum G. Martens 187la: 172. Lectotype (Wynne and Kraft 1985): INDIA: saltlakes, Calcutta, Kurz 3039 (L!, isolectotype BM). De Toni 1900: 695. Delesseria pygmaea (G. Martens) Prain 1905: 33 1. Caloglossa pygmaea Weber-van Bosse ms. Caloglossa leprieurii var. leprieurii f. pygmaea (G. Martens) E. Post 1936: 49. Dawson 1956: fig. 59 ( = Hypoglossum caloglossoides). Caloglossa adnata f. divaricata E. Post 1936: 46 (in clavi). syn. nov. Taxonomy of Caloglossa 113 Lectotype (here chosen): BURMA (MYANMAR): Kaladan R., Arracan [Arakan], x.1870, Kurz 1963 (L!). Post 1968: 273. Caloglossa leprieurii var. leprieurii f. ceylonensis J . Agardh ex Post 1936: 50. Type: SRI LANKA: Colombo, without date, Fergeson 13 (isolectotype BM!, L!, lectotype LD!). Delesseria leprieurii f. capillaris Kiitzing ms. Post 1936: 49. (based on material from New York, without date, Bailey 27 (BM!, L!). Thalli arching, loosely attached to the substratum at intervals and imbricate forming a matlike weave, reddish brown to pale brown, epiphytic and epilithic, attached by a coalescent cluster of rhizoids on the ventral side that arise from a cortical pad produced by the pericentral cells at the nodes; mature blades linear to suborbicular, (0.35-)0.6-2.1(-3.5) mm wide with (0.8-)3.1-6.2(-7.5) mm between weakly to strongly constricted nodes, sometimes arising as a rosette from a stipe, up to 4.5 mm high, 0.4-0.8 mm in diameter; endogenous branches produced by an axial cell above the nodes; adventitious branches absent. Internode morphology: the inner 5 or 6 second-order cells produce third-order rows, thus on one side of any blade from any one axial cell the apical cells of 6 or 7 rows form the blade margin, with 8-38 cells per row decreasing to 3-17 cells per row at the node. Second- and third-order cell rows close to the margin rarely divide transversely to produce single fourthorder cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 60-82", and the angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral is 15-25(-31)" (see Fig. 5 for cell angle definitions). The length of axial cells in the mature thallus is 120-180 pm. Node morphology: the angle of divergence of the lateral branch from the main axis is 48-75"; the lateral branch axis is deflected by 2 7 4 " from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle o,Fig. 5c) is 21-29". At the node, on the main axis, almost all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin and the number of branches is 2 compared with 6 or 7 in the internode region. On the lateral axis the fxst axial cell on the adaxial side produces no second-order cell row; the second axial cell produces one or two rows and these oft& do not extend to the margin. On the abaxial side the nodal axial cell produces one to three rows, extending to the margin; the fxst two axial cells of the lateral branch produce 2-4 rows each, all extending to the margin. By the fifth axial cell above the node the regular internodal condition is attained. Rhizoids arise from a cortical pad that develops by the division of the lateral and ventral pericentral (second-order) cells of the node and first anterior axial cells. The rhizoids are 0.5-0.8 mrn long, 40-60 pm in diameter, comprising 3-10 cells. Endogenous blades are formed in the nodal region of the blade by a single proliferation of the first axial cells of the lateral branch above the node (occasionally the first axial cells of the main branch). The new blade-forming cell lies between the dorsal and lateral pericentral cells and the new blade develops out of the plane of the thallus. The new blade produces further blades from the basal cefi of the new blades in a sympodial fashion. Tetrasporangial sori are 8-18 axial cells long of 1 or 2(3) rows, 7-9 row-cells wide and usually include the lateral pericentral cells and the submarginal cells. Cover cells are formed. Tetraspores are 50-75 pm in diameter at maturity. Spermatangia and carpogonia: see Pappenfuss (1961). Distribution Widespread species found commonly in mangroves and saltmarshes: also on rocky coasts and occasionally above the tidal limit of coastal rivers. Selected Specimens Examined (Over 200 Specimens examined.) AUSTRALIA: Port Fairy, Victoria, Harvey 2830 (BM). Burnett R. without date, Keys 43 (LD). Georgetown, Tasmania, s.d. Harvey 238 ( C ) . 114 R.J. King and C.F. Puttock BERMUDA: Ely Harbor, 21.iv.1914, Hewey 2038 (BM, C, L). Bermuda, 1881, Farlow 205 (L). BRAZIL: Barra do Ris CearB, Fortaleza Est CearA, 25.v.1967, Ferreira s.n. (C). BURMA (MYANMAR): Kaladan R., Arracan [Arakan], x.1870, Kurz 1963 (L). FRENCH GUIANA: nr Cayenne, without date, Leprieur s.n. (L 939285.13, LD 31909-30912). GHANA: Axim-Ancobra estuary, 25.iii. 1959, Lawson A1425 (L). GUADELOUPE: Pointe i Pitre, 12.vii [1870s], Maze 169 (BM). HONG KONG: Aberdeen, vii.1937, Tseng 2575 (L). INDONESIA: Lamakera, s.d. Weber-vanBosse 13 (L). INDIA: Mahim causeway, Bandra, Bombay, 26.xii.1927, B@rgesen5050 (C). MAURITIUS: Ile Maurice, vi.1890, Jadin 512 (C). NEW ZEALAND: Bay of Islands, without date, Hooker (BM). PUERTO RICO: Luquillo, 24.vii.1885, Sintenis A84 (BM, L). c.400-500m, Luquillo Mts, 20.vii. 1902, Wilson 339 (C, L). REPUBLIC OF SOUTH AFRICA: Durban, Natal, 1894, Weber-vanBosse s.n. (L).Kei mouth, s.d., Flanagan 216 (BM). UNITED STATES OF AMERICA: New York, without date, Bailey 27 (BM, L), specimens a and b used by Cramer 1891 (BM). Florida, 7.i.1852, Hooker (LD 31916). Port Leon, St Marks R., Florida, 26.i.1949, Drouet & Atwood 11466 (BM, L). VIRGIN ISLANDS: Christiansted Lagoon, St Croix, 17.i.1896, B@rgesen184 (BM, C, L). Taxonomic Notes Caloglossa leprieurii as recognised here is morphologically very variable and may encompass several subspecific taxa. None of the subspecific taxa recognised thus far appears to be more than a range on a continuum of variation. We agree with Post's view that Hypoglossum pygmaeum G. Martens is only a very small form of Caloglossa leprieurii. The specimens of this taxon which we have examined include female, male, and tetrasporic plants, and the taxon should be more closely examined in a detailed revision of the species. The available material of Caloglossa leprieurii var. leprieurii f. ceylonensis Agardh ex Post may be a single collection made at Colombo by Fergeson. Agardh described this material in his manuscript notes (LD 31879-31880), and the associated specimens on the same sheet 31981-31882 are annotated '13 Ceylon, Fergeson'. All other sheets available (31885-31897, L 94198.217, L 935329.23, 24 and BM) are also annotated with the above, in part or in full. It appears that Post (1936) divided part of this collection, determining it as C. leprieurii var. continua. The manuscript name Caloglossa leprieurii f. capillare was used by Kiitzing on specimens collected by Bailey from New York. They do not differ from the typical type material from Cayenne. Post (1936) described Caloglossa adnata f. divaricata (in clavi). It differs from f. typica (i.e. f. adnata) by having 'main axes with long internodes' compared with 'main axes genicu1ate'-presumably referring to the more-or-less straight main axis (referred to as monopodial by Tanaka and Chihara 1985) and the knee joint or 'zig-zag' (pseudodichotomous) main axis of the respective forms. In Post (1936), all illustrated internodes are about the same length, so that the designations of long and short shoots are inappropriate. As already noted, all Caloglossa species have psuedodichotomous branching with a monopodial main axis. In the original description of this form Post cited two specimens (=? syntypes) from Arakan and Belawan-Sumatra. The Arakan specimen was collected by S. Kurz [determined by Martens (1871b) as Hypoglossum vieillardii Kiitzing (1863), with a note that it may be juvenile H. leprieurii]. This specimen has been located [Arracan, S. Kurz 1963 (L)]and, on the criteria used in this paper (i.e. rhizoids arising from pericentral cells at nodes, and endogenous blades arising out of the blade at the node), is placed in C. leprieurii. There appears to be no doubt that this specimen is that referred to by Post, although none of it is reproductive (cf. Post 1943). The second specimen from Belawan-Sumatra was one of several collections made by Troll. In 1943, Post cited the specimen as Troll 30, although there is no indication which specimens are illustrated. Taxonomy of Caloglossa 115 Troll's collections were housed at Berlin and presumably no longer exist. We therefore consider the specimens Kurz 1963 to be the lectotype of C. adnata f. divaricata and place it in the synonomy of C. leprieurii. Caloglossa leprieurii and C. continua are distinguished from all other species except C. triclada by the production of endogenous branches at the nodes. Caloglossa leprieurii variety leprieurii was recognised by Post (1936) on the basis of having few endogenous blades associated with the nodes, and no stipe formation. Variety hookeri was recognised by having many endogenous blades associated with the nodes, and the formation of a stipe. The two morphologies can sometimes be seen on the same plant. In these, a rosette of blades is built up from repeated endogenous branching from each successive new blade. West et al. (in press) have suggested the recognition of Caloglossa apomeiotica sp, nov. inedit on the basis of the loss of sexual reproduction and the presence of viable bisporangia and abortive tetrasporangia. The taxon is restricted to Pacific Mexico, but on the basis of vegetative morphology it is indistinguishable from C, leprieurii. 8. Caloglossa continua (Okamura) R.J. King et Puttock, stat. nov. Type: JAPAN: river mouth of KO-yahagi-gawa,Mikawa, vii.1902, Okamura 67 (lectotype (here chosen) TI!, isolectotype C!, L!, TNS!). Caloglossa leprieurii var. continua Okamura 1903a: exsicc. 67; 1903b: 129. De Toni 1924: 357. Caloglossa leprieurii var. leprieurii f. continua (Okamura) Post 1936: 50. Tanaka 1992: 139. Caloglossa leprieurii var. alternatifolia Okamura 1902: 51 (nom. nud.) Caloglossa leprieurii auct. Okamura 1908: 179, 181, pl. 36, 37, figs. 12, 13; 1936: 794, pl. 383. Mori 1961: 19, figs I, 1,2,II, 2. Thalli arching, loosely attached to the substratum at intervals and imbricate forming a mat-like weave, pale brown to olive green, epiphytic, attached by a divergent cluster of rhizoids on the ventral side that arise directly from the pericentral cells at the nodes; mature blades linear to narrowly elliptical, 0.15-2.9 mm wide with 1.3-5.8 mm between weakly constricted nodes; stipeless; endogenous branches produced by an axial cell above the nodes; adventitious branches absent. Internode morphology: the inner 2-5 second-order cells produce third-order rows, thus on one side of any blade from any one axial cell the apical cells of 3-6 rows form the blade margin, with 6-36 cells per row and 3-34 cells per row at the node. Occasionally the ultimate cells of second- and third-order cell rows divide transversely to produce single fourth-order cells. The angle (a)formed by the first-formed third-order cell rows and the main axis is 66-86", and the angle (P) of divergence of the seventh cells of second- and third-order cell rows derived from a single lateral pericentral is 0-20" (see Fig. 5 for cell angle definitions). The length of axial cells in the mature thallus is 120-340 pm. Node morphology: the angle of divergence of the lateral branch from the main axis is 56-82"; the lateral branch axis is deflected by 27-53" from the prolongation of the main axis (angle 6, Fig. 5b). The angle of blade defined by the branching (angle o,Fig. 5c) is 57-85". At the node, on the main axis, all second- and third-order cell rows on the adjacent and opposite sides reach the blade margin; the number of branches on the opposite side is 1-5 compared with 3-6 in the internode region, and on the adjacent side is usually 1-4. On the lateral axis the first to third axial cells on the adaxial side produce one to occasionally three rows but those of the first and second axial cells do not extend to the margin. On the abaxial side the nodal axial cell produces 1-3 rows, most of which do not extend to the margin; the first two axial cells of the lateral branch produce 1--5 rows each, almost invariably extending to the margin. By the fifth axial cell above the node the regular internodal condition is attained. Rhizoids arise from the first and second adjacent and adaxial lateral pericentral (secondorder) cells above the node. The rhizoids are 0.6-1.5 mm long, 25-45 ym in diameter, comprising 6-10 cells. Endogenous blades are formed in the nodal region of the blade by a single proliferation of the first axial cells of the lateral branch above the node (occasionally the first axial cells of R.J. King and C.F. Puttock 116 the main branch). The new blade-forming cell lies between the dorsal and lateral pericentral cells and the new blade develops out of the plane of the thallus. This new blade produces further blades from the basal cell of the new blades in a sympodial fashion. Tetrasporangial sori are 8-21 axial cells long of 1 or 2(-3) rows, 2-16 row-cells wide and usually include the lateral pericentral cells and the submarginal cells. Cover cells are formed. Tetraspores are 50-85 pm in diameter at maturity. Spermatangia and carpogonia have been described for subspecies continua only (Tanaka 1992). Taxonomic Notes This species has been included until now in Caloglossa leprieurii, but is distinguished by the clustering of rhizoids within the axil, a character overlooked until this study, and the presence of the first lateral adaxial pericentral cell, this being absent in C. leprieurii. Within Caloglossa continua several subspecies are recognised. These are distinguished from one another on the basis of gross morphology (Table 4). These require further investigation to determine their relationships. Table 4. Morphological variation in Caloglossa continua Character ssp. continua ssp. axillaris ssp. saigonensis ssp.postiae Internode width (rnm) 0.15-0.45 0.3-0.45(-0.5) Internode length (rnm) 4.2-5.8 (1.4-)1.6-2.2 Number of cell rows produced per axial cell 3(4) 4-5 Number of cells in 2nd order rows at internodes 12-14 6-10 Number of cells in 2nd order rows at nodes 9-14 3-9 Length of axial cell (pm) 120-240(-300) 130-270 220-340 120-150 Hyphae associated with main axis absent present absent absent Angle of divergence of lateral branch from main axis 78-82" (56-)64-70" Row divergence (angle a) 66-69" 78-86" Angular divergence (angle P) 3-5" (7-) 12-20" 1-4" 0-6" Lateral branch deflection (angle 6) 38-46' 35-42' 42-53' 27-40' Angle of blade defined by branching (angle o) (58-)66-70" 57-67" 75-80" 75-85" 8a. Caloglossa continua subspecies continua Distribution Recorded from China and Japan, especially from brackish localities. Recent collections from north western Australia and cultures of these collections by M. Kamiya (Tsukuba University) are referrable to this taxon and point to the need for detailed study. Taxonomy of Caloglossa 117 Selected Specimens Examined (4 Specimens examined.) CHINA: Kutwang, South Quimoy, 17.v.1931, Tseng 133 (BM). Tien-bee, Kulangsu, 4.iii.1932, Tseng 133 (BM,C). Taxonomic Notes JAPAN: river mouth of KO-yahagi-gawa,Mikawa, vii.1902, Okamura 67 (C, L, TI, TNS). The specimen of Farlow et al. (Alg. Am. Bor. Exsicc. Fasc. I1 No. 66), which Okamura (1903b) placed with Caloglossa leprieurii var. continua, is not this entity, but C. leprieurii. 8b. Caloglossa continua subspecies axillaris R.J. King et Puttock, ssp, nov. Type: AUSTRALIA: Arnharn Highway bridge, South Alligator River, Northern Temtory, 12" 30' S 132" 30' E, 26.ii.1987, Gartrell & Harris UNSW18166. Ab subspecie typica laminis multo latioribus (1.3-)1.8-2.9 mrn latis et hyphis internalibus inter cellulas pericentrales axialesque orientibus differt. Differing from the typical subspecies in its much wider blades, (1.3-)1.8-2.9 rnm broad, and the production of internal hyphae between the pericentral and axial cells. Distribution Epiphytic on mangrove roots, pneumatophores and stems, often covered in mud. Known only from South Alligator river, northern Australia. Selected Specimens Examined (3 Specimens examined.) AUSTRALIA: Arnhem Highway bridge, South Alligator River, Northern Territory, 12" 30' S 132" 30' E, 265.1987, Gartrell & Harris UNSW 18166. South Alligator River, 6.5 km west of 'Old Munmarlary homestead', Kakadu National Park, 12" 28' S 132" 26' E, 19.vii.1987, King UNSW20819. Taxonomic Notes The axial cells at the node often produce hyphae, 4-8 cells long, directed towards the apex, and lying between the axis and the pericentral cells. In older thalli they may extend along the axes into the internodal regions. 8c. Caloglossa continua subspecies saigonensis (Tanaka et Pham-Hohng H6) R. J. King et Puttock, stat. nov. Type: VIETNAM: Cau Chu Y (Cholon), Saigon, 24.iv.1961, Tanaka & Pham-Hohng H6 2210 (holotype KAG!, isotype TNS!). Pham-Hoing H6 1969: 251, fig. 2.182. Caloglossa saigonensis Tanaka et Pham-Hoing H6 1962: 29, figs 7,8. Caloglossa leprieurii var. angusta Jao 1941: 274, pl. I, fig.1. syn. nov. Type: CHINA: Chialing R. nr Pehpei, Szechwan, 10.iv.1940, Jao SC 1105B (HBI). Seto and Jao 1984: 216, figs 1-4. Distribution Epiphytic on mangroves in rivers entering the South China Sea in Vietnam and China. Selected Specimens Examined (Only the two type specimens known.) CHINA: Chialing R. nr Pehpei, Szechwan, 10.iv.1940, Jao SC 1105B (HBI). VIETNAM: Cau Chu Y (Cholon), Saigon, 24.iv.1961, Tanaka & Pham-HoZmg H6 2210 (KAG,TNS). Taxonomic Notes This taxon was described by Tanaka et Pham-Hoing H6 from Vietnam. The variety angusta of Caloglossa leprieurii from China is here placed in synonomy with C. continua 118 R.J. King and C.F. Puttock ssp. saigonensis, the thallus morphologies being virtually identical. Reproductive material is unknown. The paper by Seto and Jao (1984) illustrates some of the regenerative morphologies typical of Caloglossa, which we regard as abnormal growth forms. Seto and Jao (1984) described the blades as being about 0.5-0.8 mm wide, but our measurements on the type material were in the range 0.15-0.45mm. 8d. Caloglossa continua subspeciespostiae R.J. King et Puttock, ssp. nov. Type: AUSTRALIA: Georges River at Georges Hall, Sydney, New South Wales, 21 .ix.1983, Wheeler & King UNSW 15205. Ab subspecie typica laminis multo angustioribus 0.3-0.45(-0.5) mm latis differt et ab subspecie saigonensi internodiis brevibus (1.4-)1.6-2.2 mm longis, seriebus lateralibus cellullarum 4-5 et seriebus adaxialibus valde reductis differt. Differing from the typical subspecies by its much narrower blades 0.3-0.45(-0.5) mm wide, and from the subspecies saigonensis by its short internodes (1.4-)1.6-2.2 mm long, four or five lateral rows and adaxial cell rows markedly reduced. Distribution Only from the Georges River, eastern Australia. Selected Specimen Examined AUSTRALIA: Georges River at Georges Hall, Sydney, New South Wales, 21.ix.1983, Wheeler and King UNSW 15205. Taxonomic Notes This species has until now been confused with Caloglossa ogasawaraensis, from which it is clearly distinguished by the presence of endogenous branches. It can also be confused with narrow forms of C. leprieurii, but the rhizoids originate in a different fashion. Conclusion This account of the genus Caloglossa is the first monographic treatment of the genus since Post (1936) and as part of this research we have made observations on all available type material. The taxonomic conclusions that we have reached are based on the presumption that there are consistent differences in vegetative morphology between the species and subspecies. Our conclusions will form the basis against which the outcomes of future studies (incorporating considerations of such aspects as biogeography, molecular genetics and interbreeding capacities) can be measured. In addition we have brought together the vast taxonomic literature on the genus. There is, as well, a wide range of ecological literature which refers to mangrove algae and specifically Caloglossa, and we have not attempted here to summarise the plethora of these ecological and biogeographical accounts. Within the scope of this project we have not been able to document fully the morphological variation within species, nor its underlying cause. This has contributed to some extent to the apparent uneven application of the species concept across the genus adopted in this study. The widespread Caloglossa leprieurii exhibits an extremely broad range of morphological variation within populations and no subspecies or even forms are recognisable. A similar diversity of morphology is observed within the C. beccarii complex but here the morphology within populations appears to be discrete. As a consequence, these taxa are maintained at species rank. The newly recognised C. continua complex likewise exhibits a broad diversity of population variation. Three of the four taxa recognised are currently very poorly known and as a consequence are given subspecific status until their relationships can be reassessed. Detailed studies on crossing experiments of the type currently being undertaken by J. West at the University of Melbourne, Australia and M. Kamiya at Tsukuba University, Japan will give new perspectives on these issues. In Taxonomy of Caloglossa 119 addition, molecular studies should clarify relationships. The conclusions reached here are based primarily on field collected material. It is essential that future studies include observations on material from a wide geographic range (including lesser collected areas such as South America), cultured under a variety of environmental conditions and taking into account age and reproductive status. Acknowledgments We express our sincere thanks to the many colleagues who assisted in the project and especially acknowledge the useful discussions and correspondence with Professor Y. Hara and M. Kamiya (Tsukuba University), Professor J. West (University of California, Berkeley), Dr J. Tanaka (National Museum, Tokyo) and Professor T. Yoshida (Hokkaido University). Bibliographic information was made freely available by Dr Paul Silva. Dr Peter Wilson kindly provided the Latin diagnoses, and Yvette Gray provided translation of Italian papers. We are grateful to many colleagues for answering specific requests for liquid preserved material. The following deserve special mention: P. Adam (UNSW), W. Kilkeary (Univ. Northern Territory), D. Walker (Univ. Western Australia), W.J. Woelkerling (La Trobe University) and W. Booth (Univ, of Brunei). We are also indebted to the directors of the following herbaria for the loan of specimens: ADU, AK, AKU, BISH, BRIU, CHR, HO, JCT, LTB, MEL, MELU and NSW. We particularly thank all of our colleagues who have reviewed the text and provided helpful criticism, especially Mitsunobu Karniya, Gerry Kraft and John West. This project was funded by the University of NSW Special Research Grants and Special Studies Program. In addition, we gratefully acknowledge the financial assistance of the Australia-Japan Foundation, the Joyce Vickery Foundation of the Linnean Society of NSW which supported field work, the Christensen Foundation for sponsorship of field work in Papua New Guinea, and Monbusho (Japan) for sponsorship of field trips to northern Australia, Baja California (Mexico) and Japan. Index to Genera, Species and Synonyms Taxa recognised in this study are in bold type. Caloglossa adhaerens R. J. King et Puttock, sp. nov. Caloglossa adnata (Zanardini) De Toni 1900 = C. bengalensis Caloglossa adnata f. typica E. Post 1936 = C. adhaerens Caloglossa adnata f. divaricata E. Post 1936 = C. beccarii Caloglossa amboinensis (G. Karsten) De Toni 1900 = C. beccarii Caloglossa beccarii (Zanardini) De Toni 1900 Caloglossa bengalensis (G. Martens) R. J. King et Puttock, comb. nov. Caloglossa bombayensis Boergesen 1933 = C. ogasawaraensis Caloglossa continua (Okamura) R. J. King et Puttock, stat. nov. Caloglossa continua ssp. axillaris R. J. King et Puttock, ssp. nov. Caloglossa continua ssp.postiae R. J. King et Puttock, ssp. nov. Caloglossa continua ssp. saigonensis (Tanaka et Pham-Hohng H8) R. J. King et Puttock, ssp. nov. Caloglossa hookeri nom, nud. J.D. Hooker et Harvey 1845a = C. leprieurii Caloglossa hookeri Harvey ms. = C, leprieurii Caloglossa leprieurii (Montagne) J. Agardh 1876 Caloglossa leprieurii var. alternatifolia Okamura 1902 = C. continua Caloglossa leprieurii var. angusta Jao 1941 = C. saigonensis Caloglossa leprieurii var, continua Okamura 1903a,b = C. continua Caloglossa leprieurii var. P subtilissima G. Martens 1869 = C. leprieurii Caloglossa leprieurii var. hookeri E. Post 1936 = C. leprieurii Caloglossa leprieurii var. leprieurii f. ceylonensis J. Agardh ex E. Post 1936 = C. leprieurii 120 R.J. King and C.F. Puttock Caloglossa leprieurii var. leprieurii f. continua (Okamura) E. Post 1936 = C. continua Caloglossa leprieurii var. leprieurii f. pygmaea (G. Martens) E. Post 1936 = C. leprieurii Caloglossa leprieurii var. leprieurii f. triclada E. Post 1936 = C, triclada Caloglossa leprieurii var. leprieurii f. typica E. Post 1936 nom. illeg. = C, leprieurii Caloglossa mnioides Harvey ex J. Agardh 1876 = C. leprieurii Caloglossa ogasawaraensis Okamura 1897 Caloglossa ogasawaraensis var. latifolia Kumano 1978 = C. beccarii Caloglossapygmaea Weber-van Bosse ms. = C. leprieurii Caloglossa saigonensis Tanaka et Pham-Hohng Hi3 1962 = C. continua ssp. saigonensis Caloglossa stipitata E. Post 1936 Caloglossa triclada (E. Post) R. J. King et Puttock, stat. nov. Caloglossa vieillardii (Kutzing) Setchell 1924 = C , leprieurii Caloglossa zanzibariensis (K.I. Goebel) De Toni 1900 = C. ogasawaraensis Delesseria adnata Zanardini 18722, = Caloglossa bengalensis Delesseria amboinensis G. Karsten 1891 = C. beccarii Delesseria beccarii nom. nud. Zanardini 1872a = C. beccarii Delesseria beccarii Zanardini l872b = C, beccarii Delesseria bengalensis (G. Martens) Prain 1905 = C. bengalensis Delesseria leprieurii Montagne 1840 = C. leprieurii Delesseria leprieurii f. capillaris Kiitzing ms. = C. leprieurii Delesseria leprieurii var. polychotoma Okamura ms. = C. ogasawaraensis Delesseria leprieurii var. subfasciculata Okamura ms. = C. ogasawaraensis Delesseria mnioides (Harvey ex J . Agardh) Prain 1905 = C. leprieurii Delesseria mnioides Harvey ms. = C. leprieurii Delesseria opuntioides J . Agardh 1854 = C. leprieurii Delesseria pygmaea (G. Martens) Prain 1905 = C. leprieurii Delesseria vieillardi (Kutzing) Zanardini 1872b = C. leprieurii Delesseria zanzibariensis K.I. Goebel 1898 = C. ogasawaraensis Hypoglossum bengalense G. Martens 1870 = C. bengalensis Hypoglossum leprieurii (Montagne) Kutzing 1849 = C. leprieurii Hypoglossum pygmaeum G. Martens 187 l a = C. leprieurii Hypoglossum vieillardii Kutzing 1863 = C. leprieurii References Agardh, J.G. (1852). 'Species genera et ordines algarum. Volumen secundum: algas florideas complectens.' Part 2 pp. 337-720, Part 3 pp. 701-786. (Gleerup: Lund.) Agardh, J.G. (1854). Nya Algformer. @versigt af Kongl. (Svenska) Vetenskaps-Akademiens Forhandlingar 11, 107-111. Agardh, J.G. (1876). 'Species genera et ordines algarum. Volumen tertium: de florideis curae posteriores. Epicrisis systematis Floridearum.' (Weigel: Leipzig.) 724 pp. Agardh, J.G. (1898). 'Species genera et ordines algarum.' Vol 3. Part 3, 1-239. (Gleerup: Lund.) Beanland, W.R., and Woelkerling W.J. (1982). Studies on Australian mangrove algae: 11. Composition and geographic distribution of communities in Spencer Gulf, South Australia. Proceedings of the Royal Society of Victoria 94, 89-106. Borgesen, F.C.E. (1911). 'The algal vegetation of the lagoons in the Danish West Indies.' pp. 1-55. Biologiske Arbejder tilegnede. (E. Warming.) Bargesen, F.C.E. (1919). The marine algae of the Danish West Indies, part 111: Rhodophyceae (5). Dansk Botanisk Arkiv 3, 305-368. Borgesen, F.C.E. (1933). Some Indian Rhodophyceae especially from the shores of the presidency of Bombay. Bulletin of Miscellaneous Information 1933(3), 113-142. Borgesen, F.C.E. (1937). Contributions to a South Indian marine algal flora-U. Journal of the Indian Botanical Society 16, 31 1-357. Borgesen, F.C.E. (1938). Catenella nipae used as food in Burma. Journal of Botany 76,265-271. Borgesen, F.C.E. (1945). Some marine algae from Mauritius. I11 Rhodophyceae. Part 4 Cerarniales. Biologiske Meddelelser 19(10), 1-68. Taxonomy of Caloglossa 121 Bourrelly, P. (1985). 'Les Algues D'eau Douce. Initiation B la SystCmatique. Tome 111: Les,Algues Bleues et Rouges Les EuglCniens, Peridiniens et Cryptomonadines.' (SociCtB Nouvelle Des Editions BoubCe: Paris.) Collins, F.S., and Hervey A.B. (1917). The algae of Bermuda. Contributions from the Bermuda Biological Station for Research 69, 1-195. leprieurii.' (Albert Muller's Verlag: Zurich.) 21 pp. Cramer, C. (1891). ' ~ b e Caloglossa r Davey, A., and Woelkerling W.J. (1980). Studies on Australian mangrove algae I. Victorian communities: composition and geographic distribution. Proceedings of the Royal Society of Victoria 91,53-66. Dawson, E.Y. (1954). Marine plants in the vicinity of the Institut OcCanographique de Nha Trang, Vi&t Nam. PaciJic Science 8,373-469. Dawson, E.Y. (1956). Some marine algae of the Southern Marshall Islands. Pacific Science 10,25-66. De Toni, G.B. (1900). 'Sylloge algarum.' Vol. 4. Florideae. Section I1 pp. 387-776. (Patavium: Padua.) De Toni, G.B. (1924). 'Sylloge algarum.' Vol6. Florideae. Section V. (Patavium: Padua.) 767 pp. Dickie, G. (1875). On the algae of Mauritius. Journal of the Linnean Society 14, 190-202. Dixit, S.C. (1968). Species list of marine algae-11. Journal of the University of Bombay 36,9-24. Entwisle, T.J., and Kraft G.J. (1984). Survey of freshwater red algae (Rhodophyta) of south-eastern Australia. Australian Journal of Marine and Freshwater Research 35,213-259. Fan, K.C. (1952). The structure, methods of branching and tetrasporangia formation of Caloglossa. Taiwan Fisheries Research Institute Laboratory Hydrobiology Report no. 4. 16 pp. Fischer, H. (1984). Turgor regulation in Caloglossa leprieurii (Montagne) J. Agardh (Delesseriaceae, Rhodophyta). Journal of Experimental Marine Biology and Ecology 81,235-239. Fortes, M.D., and Trono G.C. (1979). Marine algal microphytes new to the Philippines. Kalikasan, Philippine Journal of Biology 8, 51-68. Goebel, K. (1897). Morphologische und biologische Bemerkungen. 6. Ueber einige Susswasserflorideen aus Britisch-Guyana. Flora 83,436-444. Goebel, K. (1898). Morphologische und biologische Bemerkungen. 8. Eine Susswasserfloridee uas Ostafrika. Flora 85,65-68. Hadlich, R.M., and Bouzon Z.L. (1985). Contribution to the taxonomic survey of the benthonic marine algae from the Itacorubi mangrove-Florian6polis-Santa Catarina Island-Brazil-I1 Rhodophyta. fnsula 15, 89-116. Harvey, W.H. (1853). 'Nereis boreali-americana.' Part 11. Rhodospermeae. (Smithsonian Institution: Washington.) 258 pp. Hauck, F. (1888). Meeresalgen von Puerto-Rico. Botanische Jahrbucher fur Systematik, Planzengeschichte und Pflanzengeographie 9,457-470. Holmgren, P.K., Keuken W., and Schofield E.K. (1981). 'Index Herbariomm. Part 1. The Herbaria of the World.' 7th Edn. (Bohn, Schellema and Holkema: Utrecht.) Hooker, J.D., and Harvey W.H. (1845a). Algae antarcticae. London Journal of Botany 4,249-276. Hooker, J.D., and Harvey W.H. (1845b). Algae Novae Zelandiae. London Journal of Botany 4, 521,538-539,540-543. Howe, M.A. (1902). Caloglossa leprieurii in mountain streams. Torreya 2, 149-152. Howe, M.A. (1920). Algae. In 'The Baharna Flora'. (Eds N.L. Britton and C.F. Millspaugh.) pp. 553631. (Hafner Publishing Co.: New York.) Islam, A.K.M.N. (1976). Contribution to the study of the marine algae of Bangladesh. Bibliotheca Phycologia 19,l-253. Jadin, F. (1934) Algues des Iles de la RCunion et de Maurice. Annales de CryptogamieExotique 8,147-172. Jao, C.C. (1941). Studies on the freshwater algae of China. VIII. A preliminary account of the Chinese freshwater Rhodophyceae. Sinensia 12,274-277. Joly, A.B. (1965). Flora marinha do litoral norte do estado SSlo Paulo e regiBes circunvizinhas. Faculdade de Filosofia, Ci2cias e Letras, Universidade de SBo Paulo, Boletim 294, Serie Botrinica 21, 1-393. Karsten, G. (1891). Delesseria (Caloglossa Harv.) amboinensis. Eine neue Susswasser-Floridee. Botanische Zeitung 49,265-271. Karsten, U., West J.A., Mostaert A.S., King R.J., Barrow K.D., and Kirst G.O. (1992). Mannitol in the red algal genus Caloglossa (Harvey) J. Agardh. Journal of Plant Physiology 140,292-297. King, R.J. (1981). Mangroves and saltmarsh plants. In 'Marine Botany An Australian Perspective'. (Eds M.N. Clayton and R.J. King.) pp. 308-328. (Longman Cheshire: Melbourne.) King, R.J. (1990). Macroalgae associated with the mangrove vegetation of Papua New Guinea. Botanica Marina 33,55-62. 122 R.J. King and C.F. Puttock King, R.J., and Puttock C.F. (1989). Morphology and taxonomy of Bostrychia and Stictosiphonia (RhodomelaceaeRhodophyta). Australian Systematic Botany 2, 1-73. King, R.J., and Wheeler M.D. (1985). Composition and geographic distribution of mangrove macroalgal communities in New South Wales. Proceedings of the Linnean Society of New South Wales (1984) 108,97-117. Krishnamurthy, V., and Joshi H.V. (1970). 'A Check-List of Indian Marine Algae.' (Central Salt and Marine Chemicals Research Institute: Bhavnagar.) 36 pp. Kumano, S. (1978). Notes on freshwater red algae from western Malaysia. Botanical Magazine (Tokyo) 91,97-107. Kutzing, F.T. (1849). 'Species algarum.' (F.A. Brockhaus: Lipsiae.) 922 pp. Kutzing, F.T. (1863). Diagnosen und Bemerkungen zu drei und siebenzig neuen Algenspecies. Hedwigia 2,86-95. Kutzing, F.T. (1866). 'Tabulaephycologicae.' Vol 16. (Wilhelm Kohne: Nordhausen.) 35 pp. Kylin, H. (1956). 'Die Gattungen der Rhodophyceen.' (C.W.K. Gleerups Forlag: Lund.) 669 pp. Lawson, G.W., and John P.N. (1982). The marine algae and coastal environment of tropical West Africa. Nova Hedwigia 70,317-322. Mann, F.D., and Steinke T.D. (1988). Photosynthetic and respiratory responses of the mangroveassociated red algae, Bostrychia radicans and Caloglossa leprieurii. South African Journal of Botany 54,203-207. von Martens, G. (1869). Beitrag zur Algen-Flora Indiens von Dr. G. v. Martens in Stuttgart. Flora 52, 233-238. von Martens, G. (1870). A fourth list of Bengal algae, determined by Dr. G. v. Martens, communicated by S. Kurz, Esq. Proceedings of the Asiatic Society 39,257-260. von Martens G. (1871~).A fifth list of Bengal algae, determined by Dr. G. v. Martens, communicated by Mr. S. Kurz. Proceedings of the Asiatic Society 40, 170-173. von Martens, G. (1871b). List of algae collected by Mr. S. Kurz in Burma and adjacent islands, - by Dr. G. v. Martens, in Stuttgardt. Communicated by Mr. S. Kurz. Proceedings of the Asiatic Society 40,461-469. May, V. (1965). A census and key to the species of Rhodophyceae (red algae) hitherto recorded from Australia. Contributions from the New South Wales National Herbarium 3, 349-429. Montagne, J.P.F.C. (1840). Plantes cellulaires. Annales des Sciences naturelles, Botanique series 2 13, 193-207. Montagne, J.P.F.C. (1846). DelessCrie. In 'Dictionnaire Universel dlHistoire Naturelle'. (C.D. d'orbigny.) (d'orbigny: Paris.) Montagne, J.P.F.C. (1850). Cryptogamia Guyanensis. Annales des Sciences naturelles, Botanique series 3 14,283-309. Mori, M. (1961). Studies on the difference between the vegetative and tetrasporic thallus of Caloglossa in Japan. Japanese Journal of Phycology 9, 19-24 (57-62). Mori, M., and Ideka H. (1961). Studies on the freshwater Caloglossa of Japan. Japanese Journal of Limnology 22,225-229. Mostaert, AS., and King R.J. (1993). The cell wall of the halotolerant red alga Caloglossa leprieurii (Montagne) J. Agardh (Ceramiales, Rhodophyta) from freshwater and marine habitats: effect of changing salinity. Cryptogamic Botany (in press). Nageli, C. (1855). Wachsthumsgeschichte von Hypoglossum leprieurii (Mont.) Kg. In 'Pflanzenphysiologische Untersuchungen, Heft 1.' (Eds C. Nageli and C. Cramer.) pp. 69-75, t 8. Okamura, K. (1897). On the algae from Ogasawara-jima (Bonin Islands). Botanical Magazine (Tokyo) 11, 11-17. Okamura, K. (1902). 'Nihon Sorui Meii (Names of Japanese Algae).' 1st edn. (Tokyo.) Okamura, K. (1903a). 'Algae Japonicae Exsiccatae Fasciculus 11.' Okamura, K. (1903b). Contents of the 'Algae Japonicae Exsiccatae' Fasciculus 11. Botanical Magazine (Tokyo) 17, 129-132. Okamura, K. (1908). 'Icones of Japanese Algae.' Vol 1. (Tokyo.) Okamura, K. (1916). 'Nihon Sorui Meii (Names of Japanese Algae).' 2nd edn. (Seibido: Tokyo.) Okamura, K. (1936). 'Nippon Kaiso Shi (Japanese Seaweeds).' (Kazarnashobo: Tokyo.) d'orbigny, C.V.D. (1846). 'Dictionnaire Universel d'Histoire Naturelle.' (Paris.) Papenfuss, G.F. (1937). The structure and reproduction of Claudea multifida, Vanvoorstia spectabilis and Vanvoorstia coccinea. Symbolae Botanicae Upsalienses 2, 1-66. Papenfuss, G.F. (1944). Structure and taxonomy of Taenioma, including a discussion on the phylogeny of the Ceramiales. Madroiio 7, 193-214. Papenfuss, G.F. (1961). The structure and reproduction of Caloglossa leprieurii. Phycologia 1, 8-31. Taxonomy of Caloglossa 123 Pham-Hoing HO (1969). 'Rong Bi&nVi&tnam(Marine Algae of South Vietnam).' (Trung-Tbm HocLi&uXubt-Ban: South Vietnam.) 558 pp. Pocock, M.A. (1969). Preliminary list of marine algae collected at Inhaca and on the neighbouring mainland. In 'A Natural History of Inhaca Island, Mo~ambique'.(Eds W. Macnae and M. Kalk.) pp. 23-28. (Witwatersrand University Press: Johannesburg.) Post, E. (1936). Systematische und pflanzengeographische Notizen zur Bostrychia-CaloglossaAssoziation. Revue algologique 9, 1-84. Post, E. (1943). Zur morphologie und Okologie von Caloglossa. Ergebnisse der Sunda-Expedition der Notgemeinschaft der deutschen Wissenschaft 1929130. Archivfur Prostistenkunde 96, 123-220. Post, E. (1957). Fruktifikationen und Keimlinge bei Caloglossa. Hydrobiologia 9, 105-125. Post, E. (1963). Zur Verbreitung und Okologie der Bostrychia-Caloglossa-Assoziation. International Revue der Gesamten Hydrobiologie 48,47-152. Post, E. (1965). Caloglossa beccarii im Golf von Mexico. Hydrobiologia 26, 184-188. Post, E. (1966~).Caloglossa stipitata in Florida. Hydrobiologia 27, 109-112. Post, E. (19666). Neues zur Verbreitungsoekologie neuseelandischer und mittelamerikanischer Bostrychia-Caloglossa-Assoziation. Revue algologique 8, 127-150. Post, E. (1966~).Caloglossa ogasawaraensis in Westafrika. Hydrobiologia 27, 317-322. Post, E. (1968). Zur Verbreitungs-Okologie des Bostrychietum. Hydrobiologia 31,241-316. Prain, D. (1905). The vegetation of the districts of Hughli-Howrah and the 24-Pergunnahs. Records of the Botanical Survey ofIndia 3, 143-339. Reinsch, P.F. (1875). 'Contributiones ad algologiam et fungologiam.' Vol. 1 Rhodophyceae. (Weigel: Leipzig.) 103 pp. Saenger, P., Specht M.M., Specht R.L., and Chapman V.J. (1977). Mangal and coastal salt-marsh communities in Australia. In 'Ecosystems of the World. I. Wet Coastal Ecosystems.' (Ed. V.J. Chapman.) pp. 293-345. (Elsevier Scientific Publishing Company: Amsterdam.) Schnetter, R., and Bula Meyer G. (1982). Marine algen der Pazifikkuste von Kolombien. Chlorophyceae, Phaeophyceae, Rhodophyceae. Bibliotheca Phycologica 60,l-287. Segawa, S., and Kamura S. (1960). 'Marine Flora of Ryukyu Islands.' (Biological Institute of Ryukyus.) 72 pp. Setchell, W.A. (1924). 'Vegetation of Tutuila Island, American Samoa.' (Carnigie Institution: Washington.) 275 pp. Seto, R. (1985). Typification of Caloglossa ogasawaraensis Okamura (Ceramiales, Rhodophyta). Japanese Journal of Phycology 33, 317-319. Seto, R., and Jao C.C. (1984). Morphological study of the freshwater red alga, Caloglossa leprieurii (Montagne) J. Agardh var. angusta Jao (Rhodophyta, Ceramiales) from China. Japanese Journal of Phycology 32,216-220. Sheath, R.G., Vis M.L., and Cole K.M. (1993). Distribution and systematics of freshwater Ceramiales (Rhodophyta) in North America. Journal of Phycology 29,108-1 17. Silva, P.C., Mefies E.G., and Moe R.L. (1987). 'Catalog of the Benthic Marine Algae of the Philippines.' Smithsonian Contributions to the Marine Sciences, No. 27. (Smithsonian Institution Press: Washington D.C.) 179 pp. Skuja, H. (1938). Die Susswasserrhodophyceen der Deutschen Limnologischen Sunda-Expedition. Archivfur Hydrobiologie 15,603-637. Skuja, H. (1944). Untersuchungen uber die Rhodophyceae des Susswasser 12 tetrasporangien von Caloglossa ogasawaraensis Okam. Acta Horti Botanici Universitatis Lamiensis 14, 54-56. Soto, R. (1982). Caloglossa ogassuarensis Skuja (Rhodophyta, Ceramiales, Delesseriaceae) en Costa Rica. Brenesia 19, 251-253. Srinivasan, K. (1965). Algarum species ex India Oriundae. Bulletin of the Botanical Survey of India 67, 188-266. Tanaka, J. (1992). Reproductive structure of Caloglossa leprieurii f. continua (Ceramiales, Rhodophyceae) in Japan. Bulletin of the National Science Museum, Tokyo, Series B 18, 139-147. Tanaka, J., and Chihara M. (1985). Taxonomic studies of Japanese mangrove macroalgae 11. Two taxa of Caloglossa (Ceramiales, Rhodophyceae). Bulletin of the National Science Museum, Tokyo, Series B 11,41-50. Tanaka, J., and Chihara M. (1988~). Macroalgae in Indonesian mangrove forests. Bulletin of the National Science Museum, Tokyo, Series B 14,93-106. Tanaka, J.,and Chihara M. (1988b). Macroalgal flora in mangrove brackish areas of East Indonesia. In 'Biological system of Mangroves. A Report of East Indonesian Mangrove Expedition'. (Eds K. Ogino and M. Chihara.) pp. 21-34. (Ehime University: Ehime.) 124 R.J. King and C.F. Puttock Tanaka, J., and Kamiya M. (1993). Reproductive structure of Caloglossa ogasawaraensis Okamura (Ceramiales, Rhodophyceae) in nature and culture. The Japanese Journal of Phycology 41, 113-121. Tanaka, T., and Pham-Hohng H6 (1962). Notes on some marine algae from Viet-Nam I. Memoirs of the Faculty of Fisheries, Kagoshima University 11,24-40. Taylor. W.R. (1937). 'Marine Algae of the Northeastern Coast of North America.' (University of Michigan Press: Ann Arbor.) 427 pp. Taylor, W.R. (1950). 'Plants of Bikini and Other Northern Marshall Islands.' (University of Michigan Press: Ann Arbor.) 227 pp. Taylor, W.R. (1960). 'Marine Algae of the Eastern Tropical and Subtropical Coasts of the Americas.' (University of Michigan Press: Ann Arbor.) 870 pp. Tseng, C.K. (1945). New and unrecorded marine algae of Hong Kong. Papers of the Michigan Academy of Science, Arts and Letters 30, 157-171. Ugadim, Y. (1976). Ceramiales (Rhodophyta) do litoral sul do estado SBo Paulo e do litoral do estado do Parana (Brasil). Boletim Botanica, Universidade S6o Paulo 4, 133-172. Umezaki, I. (1962). Caloglossa ogasawaraensis Okarn. found in Wakasa Bay. Journal of Japanese Botany 37,366. Weber-Van Bosse, A. (1923). Liste des algues du Siboga, 111: Rhodophyceae. Seconde partie. Cerarniales. Siboga-Expeditie Monographie 59c, 3 11-392. West, J.A. (1991). New records of marine algae from Peru. Botanica Marina 34,459-464. West, J.A., Zuccarello G.C., Pedroche F.F., and Karsten U. (1992). Marine red algae of the mangroves in Pacific Mexico and their polyol content. Botanica Marina 35, 567-572. Womersley, H.B.S., and Bailey A. (1970). Marine algae of the Solomon Islands. Philosophical Transactions of the Royal Society of London, Bulletin of Biological Sciences 259,257-352. Womersley, H.B.S., and Shepley E.A. (1959). Studies on the Sarcomenia group of the Rhodophyta. Australian Journal of Botany 7, 168-223. Wynne, M.J. (1983). The current-status of genera in the Delesseriaceae (Rhodophyta). Botanica Marina 26,437-450. Wynne, M.J., and Kraft G.T. (1985). Hypoglossum caloglossoides sp. nov. (Delesseriaceae, Rhodophyta) from Lord Howe Island, South Pacific. British Phycological Bulletin 20, 9-19. Yarish, C., Edwards P., and Casey S. (1979~).Acclimation responses to salinity of three estuarine red algae from New Jersey. Marine Biology 51, 289-294. Yarish, C., Edwards P., and Casey S. (1979b). A culture study of salinity responses in ecotypes of two estuarine red algae. Journal of Phycology 15,341-346. Yarish, C., Edwards P., and Casey S. (1980). The effects of salinity, and calcium and potassium variations on the growth of two estuarine red algae. Journal of Experimental Marine Biology and Ecology 47,235-249. Yoshizaki, M., Iura K., Miyaji K., and Kasaki H. (1983). Two species of Caloglossa from Kujyukurihama, Chiba Prefecture. Nanki Seibutsu 25, 191-192. Zanardini, G.A.M. (1872~).Nota intorno ad un viaggio a Borneo recentemente intrapreso dal botanic0 fiorentino 0. Beccari. Atti Reale Istituto Veneto di Scienze, Lettere ed Arti, Series 4 1, 379-388. Zanardini, G.A.M. (1872b). Phycearum indicarum pugillus. Memorie del Reale Istituto Veneto di Scierzze, Lettere ed Arti 17, 129-170. Zeller, G. (1873). Algae collected by Mr. S. Kurz in Arracan and British Burma, determined and systematically arranged by Dr. G. Zeller. Journal of the Asiatic Society of Bengal42(2), 175-193. Manuscript received 24 August 1993, accepted 9 February 1994