Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Gelidiales (rhodophyta, red algae) in British Columbia and Northern Washington : taxonomy, morphology,… Renfrew, Dawn Elizabeth 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1988_A1 R46.pdf [ 35.6MB ]
JSON: 831-1.0098264.json
JSON-LD: 831-1.0098264-ld.json
RDF/XML (Pretty): 831-1.0098264-rdf.xml
RDF/JSON: 831-1.0098264-rdf.json
Turtle: 831-1.0098264-turtle.txt
N-Triples: 831-1.0098264-rdf-ntriples.txt
Original Record: 831-1.0098264-source.json
Full Text

Full Text

GELIDIALES (RHODOPHYTA, RED ALGAE) IN BRITISH COLUMBIA AND NORTHERN WASHINGTON: TAXONOMY, MORPHOLOGY, DEVELOPMENT AND LIFE HISTORY by DAWN ELIZABETH RENFREW MSc, Acadia University, 1983 BSc, The University of Toronto, 1980 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES Botany We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1988 c Dawn Elizabeth Renfrew, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. 1 further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of P)( The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT A study of the red algal order Gelidiales in British Columbia and northern Washington was conducted. Gelidiales are represented in the study area by four species, Gelidium coulteri, G. purpurascens, G. vagum and Pterocladia caloglossoides, and a key to identify the taxa is provided. Earlier reports of G. crinale, G. pusillum, G. robustum and G. sinicola from British Columbia and northern Washington have been shown to be misidentifications of the other taxa and are excluded from the flora. The occurrence of G. vagum in the British Columbian flora is significant because it appears to have been introduced from the northwest Pacific and now has a disjunct distribution, occurring in Japan and China and on two islands in the Strait of Georgia, British Columbia. Gelidium vagum and the Gulf of California species, G. johnstonii, are considered to be sister species. They share the unique character of monoecy (i.e. unique in the genus) and a morphological resemblance, but differ in several other characters, including arrangement of cortical cells, shape and flattening of tetrasporangial stichidia, shape of stichidial apices and position of the apical initial with respect to the adjacent cortex, presence/absence of an apical furrow and sterile margin in tetrasporangial stichidia and presence/absence of intact third order filaments in mature cystocarps. The life history of Gelidium vagum was completed in culture, only the second Gelidium species for which this has been accomplished. Gelidium vagum has a triphasic Polysiphonia-type life history with isomorphic (monoecious) gametophytes and tetrasporophytes, and carposporophytes growing on gametophytes. The rare ii occurrence (G. purpurascens, G. coulteri) or absence (Pterocladia of gametophytes in the field and the lack of successful completion of a life history in culture suggests that, whereas these species may occasionally complete a Polysiphonia-type life history in situ, vegetative or apomictic mechanisms are more important in maintaining field populations. Chromosome counts of n= 14-15 were made on undivided tetrasporangia of G. vagum. The pattern of spore germination in Gelidium coulteri, G. purpurascens and G. vagum is similar for carpospores and tetraspores. A single germ tube grows from the spore, the entire cytoplasmic contents of the spore evacuate into the germ tube and a wall cuts the germ tube off from the empty spore. The germ tube divides unequally to form a concave and a fusiform cell. The primary attachment rhizoid forms from a derivative of the concave cell and the sporeling apical cell forms from a fusion cell derivative. As apical organization is established, the subapical cell produces two lateral periaxial cells. Carpogonium and carposporophj'te development was followed in Gelidium purpurascens and G. vagum. The functional carpogonium is intercalary. Non-functional carpogonia divide transversely with a concave wall and become sessile and degenerate. The fertilized carpogonium consistenty forms a fusion cell by fusing with adjacent cortical cells. Fusion cell lobes cut off gonimoblast filaments that send out processes that fuse with haploid gametophyte cells of nutritive chains. Carposporangia are produced terminally and laterally by gonimoblast cells and protrude into two locules that have formed as the cortex is pushed away from the plate of second order filaments. Carpospores are released iii through a single ostiole in each locule. Spermatogenesis, tetrasporogenesis and rhizine morphology were studied ultrastructurally for the first time in Gelidiales. In spermatangial development, apical chloroplasts in the spermatangial mother cell are cut off by a furrow, a mechanism previously not reported in red algae for exclusion of chloroplasts from spermatangia. As the spermatangium matures, spermatangial vacuoles form from the coalescence of vesiculated endoplasmic reticulum and dictyosome-derived vesicles. Spermatangial vacuoles are discharged from the spermatangium just prior to the release of the spermatium. Stages of tetrasporangium development are marked b}' wall development. Prior to karyokinesis, the young tetrasporangium is dominated by a large nucleus. Following karyokinesis, chloroplasts, starch grains and dictyosomes increase in abundance, a wall layer is deposited, and dictyosomes undergo a series of morphological changes from small and flat, to large and flat producing fibrillar vesicles, to hemispherical and producing cored vesicles. At cytokinesis an electron dense wall layer is deposited around the tetrasporangium and in the cleavage furrow, and cored vesicles secrete a mucilage-like material inside this wall layer. iv TABLE OF CONTENTS Abstract ii Table of Contents v List of Figures ~ x List of Tables V 1 1 1 Acknowledgements xii Chapter I. Introduction 1 A . Taxonomic History 1 B. Characterization of Gelidiales 5 C. Differentiating Gelidium and Pterocladia .... 7 D. Need for Additional Information 9 Chapter II. Materials 10 Chapter III. Taxonomy and Nomenclature 12 A . Introduction 12 B. Methods 15 C. Taxa Present in British Columbia and Northern Washington 17 1. Gelidium coulteri Harvey 18 a. Habitat and Habit 18 b. Vegetative Anatomy 19 c. Reproductive Anatomy 20 d. Discussion 22 e. Seasonality and Distribution 24 2. Gelidium purpurascens Gardner 25 a. Habitat and Habit 26 b. Vegetative Anatomy 27 c. Reproductive Anatomy 30 d. Discussion 34 e. Seasonality and Distribution 37 3. Gelidium vagum Okamura 38 a. Habitat and Habit 39 b. Vegetative Anatonvy 40 c. Reproductive Anatomy 42 d. Discussion 46 e. Seasonality and Distribution 52 4. Pterocladia caloglossoides (Howe) Dawson 53 a. Habitat and Habit 53 b. Vegetative Anatom}' 55 c. Reproductive Anatonvy 56 d. Discussion 57 e. Seasonality and Distribution 62 v D. Previously Reported Taxa Excluded from the Revised Flora 63 1. Gelidium robustum (Gardner) Hollenberg et Abbott . 64 2. Gelidium crinale (Turner) Lamouroux 67 3. Gelidium pusillum (Stackhouse) Le Jolis 69 4. Gelidium sinicola Gardner 70 Chapter IV. Life History of Gelidium in Culture 73 A. Introduction 73 B. Methods 75 1. Culturing 75 2. Chromosome Counts 78 C. Spore Germination and Early Development 78 D. Reproduction in Cultured and Field Plants 80 1. Culture Results 80 2. Field Observations of Reproduction 84 E. Discussion 85 1. Spore Germination 85 2. Life Histories 89 3. Chromosome Counts 91 Chapter V. Carpogonium and Carposporophyte Development 93 A. Introduction 93 B. Methods 96 C. Carpogonium Development in Gelidium 96 D. Carposporophyte and Cystocarp Development 98 E. Discussion 100 Chapter VI. Spermatiogenesis in Gelidium 105 A. Introduction 105 B. Methods 107 C. Spermatiogenesis 108 D. Comparison of Spermatiogenesis in Gelidium and Other Florideophycean Genera 112 Chapter VII. Ultrastructure of Tetrasporogenesis in Gelidium 122 A. Introduction 122 B. Methods 123 C. Ultrastructural Observations 123 1. Pre-meiotic Tetrasporangium 124 2. Post-meiotic Tetrasporangium 126 3. Tetrasporangium at Cytokinesis 127 D. Discussion 128 Chapter VIII. General Discussion and Conclusion 137 References 142 Figures 163 vi Tables 262 Appendix 1. Herbarium specimens examined 275 Appendix 2. Procedure for Embedding Material in JB4 Methacrylate 288 Appendix 3. Preparation of Tetrasporangial and Spermatangial Material for Transmission Electron Microscopy 290 vii List of Tables Table 1. Historical summary of genera included in Gelidiaceae 262 Table 2. Sites and dates of field collections of Gelidiales from the study area 264 Table 3. Herbaria from which loans were obtained and their abbreviations .... 267 Table 4. Species of Gelidiales recorded from the northeast, northwest and southeast Pacific Ocean areas 268 Table 5. Differences between Gelidium purpurascens and Gelidium robustum .... 273 Table 6. Anatomical differences between Gelidium johnstonii and Gelidium vagum 274 v i i i List of Figures Abbreviations used in figures 163 Fig. 1. Map of study area 165 Fig. 2. Barkley Sound collection sites 167 Fig. 3. Southern Vancouver Island, Gulf Islands and San Juan Islands collection sites 169 Fig. 4. Strait of Georgia collection sites 171 Fig. 5. Key to Gelidiales in British Columbia and northern Washington 173 Figs. 6-11. Gelidium coulteri vegetative morphology 175 Figs. 12-17. Gelidium coulteri tetrasporophytes 177 Figs. 18-20. Gelidium coulteri male gametophytes 179 Figs. 21-24. Gelidium coulteri habit 181 Figs. 25-29. Gelidium purpurascens habit and vegetative morphology 183 Figs. 30-34. Gelidium purpurascens vegetative morphology 185 Figs. 35-40. Ultrastructure of rhizines in Gelidium purpurascens and G. vagum 187 Figs. 41-44. Gelidium purpurascens tetrasporoph3'tes 189 Figs. 45-48. Gelidium purpurascens female gametophytes 191 Figs. 49-51. Gelidium purpurascens carposporophytes 193 Figs. 52-56. Gelidium purpurascens male gametophytes 195 Figs. 57-63. Gelidium purpurascens apical and lateral initials, historical and type collections 197 Figs. 64-68. Gelidium vagum habit and vegetative morphology 199 Figs. 69-76. Gelidium vagum vegetative and tetrasporophyte morphology 201 Figs. 77-83. Gelidium vagum reproductive morphology 203 Figs. 84-89. Gelidium vagum gametophytes 205 Figs. 90-95. Gelidium vagum reproductive morphology 207 ' i x Figs. 96-100. Gelidium vagum Japanese material 209 Figs. 101-106. Gelidium johnstonii habit, vegetative and reproductive morphology 211 Figs. 107-111 Gelidium johnstonii reproductive morphology and G. vagum habit 213 Figs. 112-118. Pterocladia caloglossoides habit and vegetative morphology 215 Figs. 119-121. Pterocladia caloglossoides vegetative morphology 217 Figs. 122-128. Pterocladia caloglossoides reproductive morphology 219 Figs. 129-133. Pterocladia caloglossoides reproductive morphology 221 Figs. 135-140. Historical collections and type specimens 223 Fig. 141. Polysiphonia-type life history 225 Figs. 142-153. Sporeling development of Gelidium vagum in culture 227 Figs. 154-162. Sporeling development of Gelidium vagum in culture 229 Figs. 163-168. Reproduction of Gelidium vagum in culture 231 Figs. 169-173. Reproduction of Gelidium purpurascens and G. coulteri in culture 233 Figs. 174-177. Chromosomes of Gelidium vagum 235 Figs. 178-183. Carpogonium and carposporophyte development in Gelidium purpurascens and G. vagum 237 Figs. 184-188. Carposporophyte development in Gelidium purpurascens and G. vagum 239 Figs. 189-192. Ultrastructure of spermatiogenesis in Gelidium vagum 241 Figs. 193-196. Ultrastructure of spermatiogenesis in Gelidium vagum 243 Figs. 197-200. Ultrastructure of spermatiogenesis in Gelidium purpurascens and G. vagum 245 Figs. 201-204. Ultrastructure of spermatiogenesis in Gelidium purpurascens and G. vagum 247 Figs. 205-208. Ultrastructure of spermatiogenesis in Gelidium purpurascens and G. vagum 249 Fig. 209. Summary of spermatiogenesis in Gelidium 251 Figs. 210-215. Ultrastructure of tetrasporogenesis in Gelidium purpurascens 253 Figs. 216-220. Ultrastructure of tetrasporogenesis in Gelidium purpurascens 255 Figs. 221-226. Ultrastructure of tetrasporogenesis in Gelidium purpurascens 257 Figs. 227-232. Ultrastructure of tetrasporogenesis in Gelidium coulteri and G. purpurascens 259 Acknowledgements I have received assistance and suppport from a number of people, to whom I am very grateful. Many thanks to my supervisors for their generous all-round support and concern, to Dr. D. J . Garbary who got me interested in this project, to Dr. R. F . Scagel who kindly took over when Dr. Garbary left for another position and to Dr. K . M . Cole who gave support from beginning to end. I am deeply indebted to Dr. P. W. Gabrielson who always showed interest and offered much of his time for discussion, advice and as my diving buddy. I am grateful to Ira Borgmann, Dr. W. Wheeler, Dr. B. R. Oates, Dr. M . W. Hawkes, Cindy Durance and Gary Kendrick who also were diving buddies, and to Dr. G. I. Hansen who collected material intertidally for me. These field collections were very important to my work. Dr. B. Hunt and Don DeReussi gave much appreciated hospitality and access to dive sites at Whalebone Bay and Georgina Point. I thank Dr. J . P. van der Meer, Dr. P. C. Silva, Dr. M . H . Hommersand and Dr. I. Akatsuka for valuable discussions. I appreciate the assistance of Dr. V . L. Anderson (at A H F H ) , Dr. S. Honey (at BM), Dr. P. B. Hamilton (at C A N A), Dr. G. J . Cacavio (at F); Dr. D. Duggins (at FHL) , Dr. M . J . Wynne (at MICH), Dr. J . Tanaka (at TNS), Julie Oliveira (at UBC), Dr. P. C. Silva (at UC), Dr. J . N . Norris (at US) and Dr. M . F. Denton (at WTU) for sending herbarium loans or looking for possible Pterocladia caloglossoides isot3rpes. Dr. J . W. Waaland and Dr. M . J . Wynne graciously loaned specimens from their private collections. xii Funding for this research was provided by the National Science and Engineering Research Council of Canada grants to Dr. Cole (0645), Dr. Scagel (A4471) and Dr. Garbary (U0014), an H. R. Macmillan Family Scholarship, a Kit Malkin Scholarship and a British Columbia Post Secondary Scholarship. To all those acknowledged here, to Frank and Dancer who helped me through the hardest part and to my friends who helped keep it all in perspective, thank you — you are greatly appreciated. xiii CHAPTER I. INTRODUCTION A. TAXONOMIC HISTORY Gelidiales Kyl in (1923: 132) currently encompass the single family Gelidiaceae Kutzing (1843: 390) which contains 11 genera: Acanthopeltis Okamura (in Yatabe, 1892: 157), Beckerella Kyl in (1956: 139), Gelidiella Feldmann et Hamel (1934: 529), Gelidium Lamouroux (1813: 128), Onikusa Akatsuka (1986a: 63), Porphyroglossum Kutzing (1847a: 775), Pterocladia J . G. Agardh (1852: 482), Pterocladiastrum Akatsuka (1986c: 55), Ptilophora Kutzing (1847b: 25), Suhria J . G. Agardh ex Endlicher (1843: 41) and Yatabella Okamura (1900: 1). Historically, a heterogeneous assemblage of genera (now included in four orders) was ascribed to the family (Kutzing, 1849; Schmitz, 1889). A historical summary of genera included in Gelidiaceae in seven major classifications is presented in Table 1. The family Gelidiaceae was erected by Kutzing (1843: 390, 405 as "Gelidieae"; vide Silva 1980: 83) to encompass cartilaginous, pinnately branched plants, with fibrous internal structure, exserted cystocarps, small globose spermatia and cruciate tetrasporangia on distinct tetrasporangial branches. The family originally comprised Acrocarpus Kutzing (1843: 405), Ctenodus Kutzing (1843: 407), Echinocaulon Kutzing (1843: 405) and Gelidium, with the later addition (Kutzing, 1849) of Polycladia Montagne (1847: 378), Thysanocladia (Endlicher) Lindley (1846: 25), Delisea Lamouroux (1819: 41), Chondrodon Kutzing (1847b: 5) and Phacelocarpus Endlicher et Diesing (1845: 289), and the substitution of Euctenodus Kutzing (1847b: 5) for Ctenodus. J . G. Agardh's (1851) concept of Gelidiaceae 1 Introduction / 2 (as Gelidieae) was narrower, including only the genera Gelidium [composed of the subgroups Acrocarpus ( — Gelidium in part), Echinocaulon ( = Gelidiella) and Gelidium], Pterocladia (newly erected), Suhria and Ptilophora, all genera currently belonging to the family. Wurdemannia Harvey (1853: 245) was added (with a query) because of a similarity in thallus construction to Gelidium (J. G. Agardh, 1876). Schmitz (1889) devised a classification scheme based on thallus construction and cystocarp construction and development and recognized in Gelidiaceae 14 genera in five tribes. Schmitz's family concept was broad and heterogeneous, and all genera in four of the tribes, i.e. Binderelleae [including Binderella Schmitz (in Engler and Prantl, 1897: 342) and Choreocolax Reinsch (1874-1875: 61)], Harveyelleae [including only Harveyella Schmitz et Reinke (1889: 28)], Wrangelieae [including Wrangelia C. A . Agardh (1828: 136), Atractophora H . M . Crouan et P. L . Crouan (1848: 371) and Naccaria Endlicher (1836: 6)] and Caulacantheae [including only Caulacanthus Kii tzing (1843: 395)] were transferred eventually to other families. Schmitz did recognize, however, the similarities of gelidiaceous plants, and he placed Gelidium, Pterocladia, Suhria, Porphyroglossum, Ptilophora, Acropeltis Montagne (1837: 355) and Schottmullera Grunow (in Schmitz 1889: 440; nomen nudum) in the tribe Gelidieae and excluded Wurdemannia from Gelidiaceae. Schottmullera was renamed Acanthopeltis by Okamura (in Yatabe, 1892) and, along with Spencerella Darbishire (1896: 199), it was added to Gelidiaceae (Schmitz and Hauptfieish, 1897). Oltmanns (1904) removed Harveyella, Wrangelia, Atractophora and Naccaria to Wrangeliaceae (Gigartinales). De Toni (1924) apparently disagreed with Oltmanns' removal of Wrangeliaceae, because he added Haliacantha J. G. Agardh (1899: Introduction / 3 109) and Gulsonia Harvey (1855: 334) to the tribe Wrangelieae in the Gelidiaceae. He also added Yatabella and Spencerella to the tribe Gelidieae. Choreocolax was removed to Cryptonemiales (Sturch, 1926), Harveyella and Caulacanthus were transferred to Gigartinales (Sturch, 1926 and Feldmann and Hamel, 1934, respectively) and the new genus Gelidiocolax Gardner (1927b: 340) was added to Gelidiaceae. Feldmann and Hamel (1934) renamed Echinocaulon Gelidiella, because the name Echinocaulon was a later homonym for a genus of Polygonaceae. Kylin (1956) revised the contents of Gelidiaceae as it had developed from Schmitz, transferring Binderella to Gigartinales, Haliacantha, Gulsonia and Spencerella to Ceramiales, and adding Beckerella. Fan and Papenfuss (1959) removed Gelidiocolax to Cryptonemiales. Fan (1961) placed Gelidiella in its own family, Gelidiellaceae, based on the absence of rhizines and apparent lack of a sexual generation. Recently, Santelices and Montalva (1983) merged the monotypic genus, Acropeltis, characterized by shield-shaped tetrasporangial stichidia, with Gelidium, stating that this character "has no taxonomic significance in presentty accepted schemes of classification of the Gelidiaceae." As a result of studies of vegetative characters that emphasized surface cell morphology, Akatsuka (1986a) erected Onikusa, based on Gelidium pristoides (Turner) Kiitzing (1849: 786) from South Africa. Also using morphological characters with an emphasis on cortical cell morphology Akatsuka (1986c) erected Pterocladiastrum, based on the "Robust" and "Poor Knights" forms of Pterocladia lucida from New Zealand. The position of Gelidiaceae has fluctuated between inclusion in Nemaliales Schmitz Introduction / 4 in Engler (1892) and recognition as an independent order, Gelidiales Kylin (1923). When Gelidiaceae was erected, Kiitzing (1843) included it in the order Periblasteae. Schmitz (1883) placed Gelidiaceae in its own order, but later (Schmitz, 1889) included it in Nemaliales (as "Nemalioninae"), where it remained until Kylin (1923) raised it to ordinal rank, coordinate with Nemaliales (as "Nemalionales"; see Nicolson and Norris, 1983 for correct spelling and use of Nemaliales), C^ptonemiales Schmitz in Engler (1892), Gigartinales Schmitz in Engler (1892), Rhodymeniales Schmitz in Engler (1892) and Ceramiales Oltmanns (1904). Kylin saw Nemaliales and Gelidiales as similar in that the fertilized carpogonium in both was the starting point for the gonimoblast. According to Kylin (1923), auxiliary cells were absent in Nemaliales but present in Gelidiales, functioning in a nutritive capacity. Kylin noted that Gelidiales were believed to be diplobiontic, in contrast to the supposed haplobiontic Nemaliales, although he specifically refrained from using life history characters to separate the orders. However, misinterpreting that Kylin (1923) had elevated Gelidiaceae to ordinal rank on the basis of life history differences (diplobiontic in Gelidiales versus haplobiontic in Nemaliales), Dixon (1961) returned Gelidiaceae to the Nemaliales (as "Nemalionales"). He cited the questionable existence of a regular Polysiphonia-type life histor}' (Dixon, 1961) in some Gelidium and Pterocladia species, based on the rare occurrence or absence of generations, usually gametophytes, and the lack of reports of sexual plants for Gelidiella species, as an arguement against regarding Gelidiaceae as strictly diplobiontic. Dixon also pointed out that some members of Nemaliales (as Nemalionales) (e.g. Bonnemaisonia) appeared to be diplobiontic. Hence life history type no longer Introduction / 5 could be used to separate Gelidiales from Nemaliales. Papenfuss (1966) argued, however, that ordinal status for Gelidiales was warranted as they were characterized uniquely by nutritive chains, the production of only two pericentral cells and Gelidium-type spore germination. Gelidiales contained the single family Gelidiaceae until Fan (1961) erected a second family, Gelidiellaceae, to contain species of Gelidiella (discussed above). The report of rhizines in Gelidiella calcicola (Maggs and Guiry, 1987) and cystocarps in a Gelidiella species from India (Sreenivasa Rao and Trivedi, 1986, cited in Maggs and Guiry, 1987) contradict Fan's (1961) characters for separation of the family and Maggs and Guiry (1987) merged Gelidiellaceae with Gelidiaceae. B. CHARACTERIZATION OF GELIDIALES At present, Gelidiales is recognized as an independent order (Santelices, 1974; Garbary et al., 1982; Gabrielson and Garbary, 1986, 1987; Hommersand and Fredericq, 1988). The order appears monophyletic (Gabrielson and Garbary, 1986, 1987), having several unique characters that set it apart from other orders. Gelidiales are uniaxial, pseudoparenchymatous. have pit plugs with a single plug cap (Pueschel and Cole, 1982), and show "Gelidium-type" spore germination (Chemin, 1937) with germling development proceeding directly to prostrate or upright axes without an intervening discoid stage (Boillot, 1963). Axial cells produce two periaxial cells at 180° (Fan, 1961). and inner cortical cells produce elongate, thick-walled, unicellular rhizines that grow basipetally. Gelidium-type spore germination and sporeling development (Papenfuss, 1966) and rhizines are Introduction / 6 features unique to Gelidiales (Fan, 1961; Santelices, 1974). Rhizines, also referred to as rhizoids and hyphae (Feldmann and Hamel, 1936; Dixon, 1958; Fan, 1961), are unique anatomical features of the vegetative thallus of Gelidiales. Present in all genera of Gelidiales, rhizines are found throughtout the thallus in all except Gelidiella where they are reported to occur only near attachment pads(Maggs and Guiry, 1987). Rhizines differ from multicellular rhizoid filaments that grow between medullary cells of some other genera of red algae, in that gelidioid rhizines are unicellular. This distinguishes rhizines sensu Feldmann and Hamel, 1936) from external attachment rhizoids. Spermatangia are produced by the transverse division of spermatangial mother cells (Tazawa, 1975), a character unique to Gelidiales at the ordinal rank (Gabrielson and Garbary, 1987). In a recent study of pre- and post-fertilization events in Gelidium pteridifolium Hommersand and Fredericq (1988) revised some previous interpretations. According to Hommersand and Fredericq the carpogonium is intercalary and has associated chains of nutritive cells. An auxiliary cell sensu Drew (1954) is absent, because as the fertilized carpogonium enlarges, it makes non-obligatory fusions with adjacent chains of nutritive cells to form a fusion cell. The nutritive cells fused with are not "specified" (sensu Drew), and thus cannot be interpreted as auxiliary cells. The fusion cell initiates a filamentous carposporangium-producing gonimoblast within a pericarp. Tetrasporangia are cruciate, irregular cruciate or tetrahedral (Guiry, 1978). The life history is presumed to be of the Polysiphonia-type (Santelices, 1974), but has been completed for only two species of Gelidium (Macler and West, 1987; van der Introduction / 7 Meer, pers. comm.). The life history of the type species of Gelidium still needs to be followed in culture. C. DIFFERENTIATING GELIDIUM AND PTEROCLADIA The most speciose and geographically widespread genera of Gelidiaceae, Gelidium and Pterocladia (Santelices, 1974), are both represented in the British Columbia flora. Gelidium and Pterocladia are similar morphologically, and their differentiation has long been problematic (see e.g. Santelices, 1974, unpubl. mscr.; Stewart, 1976; Akatsuka, 1986a; Rodriguez and Santelices, 1987, unpubl. mscr.). Rodrigeuz and Santelices (unpubl. mscr.) reviewed characters previously proposed for separating these genera. Currently, only the number of cjrstocarp locules (two in Gelidium and one in Pterocladia) can be used consistently to separate the genera. Okamura (1934) had suggested that rhizine position and medullary cell form are useful differentiating characters, generalizing that in Gelidium rhizines are distributed in the cortex and outer medulla and medullary cells are rounded and loosely packed, whereas in Pterocladia rhizines are distributed only in the inner medulla and medullary cells are angular and closely packed. Despite this distinction, Okamura noted some exceptions to these generic rhizine patterns. Others (i.e. Santelices, 1974; Stewart, 1976) have found these characters to be variable and inconsistent, and thus incapable of discriminating between Gelidium and Pterocladia (Rodriguez and Santelices, unpubl. mscr.). Stewart (1976) studying the California flora, proposed the presence or absence of basally incurved (geniculate) branches as a character to separate these two genera with geniculate branches present in Gelidium and absent in Pterocladia. However, this character Introduction / 8 was found to be inconsistent when tested on more species of both genera (Rodriguez and Santelices, unpubl. mscr.). Recently, Akatsuka (1981, 1986a, 1986c) proposed that surface cortical cell morphology, arrangement and orientation, in surface view and longitudinal section, are capable of differentiating Japanese species of Gelidium and Pterocladia. Akatsuka's criteria were found only to be consistent for the most basal parts of (upright) axes (Rodriguez and Santelices, unpubl. mscr.). Rodriguez and Santelices (1987, unpubl. mscr.) have proposed apical architecture characters that almost consistently separate Gelidium from Pterocladia (except P. bulbosa and possibly other Pterocladia species with acute apices). They suggested that apical and lateral initials are morphologically similar in Pterocladia but are different in Gelidium and that lateral initials m a 3 ? be in cortical indentations in Pterocladia but are never indented in Gelidium. These criteria were tested further on gelidioid species in British Columbia and northern Washington. Rodriguez and Santelices (1987, unpubl. mscr.) implied that other features of apical architecture (lateral initials close to or far from the apical initial, apical initial protruding beyond or indented between adjacent cortical lobes, and lateral initials protruding from or indented between cortical lobes) also might be used for generic segregation of Gelidium and Pterocladia. While discussing the states of these characters in small groups of species of Gelidium and Pterocladia Rodriguez and Santelices (unpubl. mscr.) do not demonstrate how the characters may be used to vegetatively discriminate Gelidium and Pterocladia. Introduction / 9 D. NEED FOR ADDITIONAL INFORMATION Given the economic importance of Gelidiales as a high quality source of agars (Santelices, 1974, 1986), surprisingly little is known about the biology of members of the order. Manj' taxa have been described, but the morphological plasticity of species has caused great taxonomic confusion (e.g. Dixon, 1958), and few careful, comparative, monographic studies have been made. Gelidiaceae occupied a pivotal position in the definition of an auxiliary cell, yet there are still many aspects of pre- and post-fertilization development to be resolved, with few species having been studied in detail. Developmental studies of spermatiogenesis and tetrasporogenesis, particularly ultrastructural, are needed to better understand developmental patterns that may help establish the position of Gelidiales with respect to other orders. The life history for members of the order has been presumed Polysiphonia-type and was recently determined from culture studies for only two species (Macler and West, 1987; van der Meer, pers. comm.); further confirmation is needed in other species, particular^' the generic types from their type locality. The seaweed flora of British Columbia and northern Washington is being revised (Garbary et al., 1980, 1982; Hawkes and Scagel, 1986a, 1986b; Lindstrom and Scagel, 1987), and this research on Gelidiales contributes to this goal. C H A P T E R II. M A T E R I A L S Collections of Gelidium coulteri, G. purpurascens, G. vagum and Pterocladia caloglossoides were made on the southern part of British Columbia and some of the San Juan Islands (Washington). In addition, herbarium specimens were examined from previous collections throughout the study area. Most of this study was carried out on freshly collected or 5% Formalin or acetic alcohol (3:1 ethanohglacial acetic acid) liquid preserved plants. Herbarium collections were relied on for type specimens, for representatives from a taxon's range, and for comparisons of local species with specimens which do not occur in the study area. A total of 56 field collections were made from 27 sites between Nov. 1983 and Aug. 1987 (Table 2). Collections were made intertidally and from the shallow subtidal using SCUBA. Some sites were selected based on previous reports of gelidiaceous algae (e.g. South, 1968; Scagel, 1973; Garbary et al., 1984), or from herbarium records. Other sites were selected on the basis of a suitable habitat and substratum (e.g. bedrock, boulders). All collection sites (with the exception of the Barkley Sound locations) were chosen because of their relative accessibility to shore diving; a boat usually was not available. Of the collection sites (Figs. 1-4), eleven sites were in Barkley Sound, west coast of Vancouver Island (Fig. 2), seven in the Strait of Juan de Fuca and Gulf of Georgia (Fig. 3), five on Hornby and Denman Islands, mid-way along the western side of the Strait of Georgia (Fig. 4), and four mid-waj' along the 10 Materials / 11 eastern side of the Strait of Georgia on the mainland (Fig. 4). Frequent collections were made at Orlebar Point, Gabriola Island (Table 2) in the Strait of Georgia, a site chosen for its accessibility (suitable for a day trip), good diving conditions (safe entry, some protection from winter storms and availability of accurate marine weather reports) and the abundance of three gelidiaceous species. Gelidium and Pterocladia populations were studied approximately ever3' 6-8 weeks throughout the year. Collections were qualitative with presence/absence, growing condition, and reproductive state of the plants noted. Herbarium specimens were obtained from AHFH, CANA, FHL, MICH, TCD, UC, US, USM and WTU (see Table 3 for abbreviations; Holmgren et al, 1978). After the AHFH loans used in this study were obtained, the AHFH collection was moved to LAM. A detailed list of all herbarium specimens studied is given in Appendix 1. CHAPTER III. TAXONOMY AND NOMENCLATURE A. INTRODUCTION Since the first published records of Gelidiaceae in the study area (Setchell and Gardner, 1903), a total of nine species of Gelidium Lamouroux (1813: 128) and Pterocladia J . G. Agardh (1852: 482) have been reported. Setchell and Gardner (1903) reported Gelidium amansii Lamouroux (1813: 41) from Vancouver Island as have subsequent reports (Collins, 1913: 114; Kyl in , 1925: 13 and Connell, 1928: 100). Scagel (1957: 142) interpreted all of these reports to be G. crinale (Turner) Lamouroux (1825: 191). [Kylin's (1925) report of G. amansii is a repeat of Setchell and Gardner's (1903) record, because he stated that he did not see G. amansii in his collections]. A t least some early "G. amansii" records can probably be attributed to G. purpurascens Gardner (1927a: 275), a taxon recorded [by Scagel (1957: 141, as G. amansii] from the southern part of Vancouver Island (CANA 3843, C A N A 4346, C A N A 4349). A second species, Gelidium cartilagineum (L.) Gaillon (1828: 15) was reported by Kyl in (1925) from San Juan Island, Washington. Gardner (1927a: 280) described G. cartilagineum var. robustum to encompass the Californian plants previously referred to G. cartilagineum. Previous British Columbian and northern Washington records of G. cartilagineum were considered to be G. cartilagineum var. robustum by Scagel (1957). These large, robust plants were later reported by Hollenberg and Abbott (1965: 1179) as a distinct species, G. robustum. 12 Taxonomy and Nomenclature / 13 Gelidium coulteri Harvey (1853: 117) was collected first from the southern part of Vancouver Island (as G. crinale in 1908, CANA 3474, Scagel, 1957). The identification of this species in the study area as G. coulteri first was made by Harlin (1969: 258) and Abbott (1970: 1) from Washington, and by Scagel (1973: 138) from British Columbia. Scagel (1957) first reported G. crinale in British Columbia and Washington (CAN 207, now CANA 3474, and V 1456 (V = herbarium of the British Columbia Provincial Museum). Gelidium sinicola Gardner (1927a: 278) was mentioned first in descriptive ecological works (Stephenson and Stephenson 1961a: 15 and 1961b: 229) and later as a taxonomic record (Norris and West, 1966: 176). Gelidium pulchrum Gardner (1927a: 279) has been recorded only once in an ecological study of zonation in the Strait of Georgia (Stephenson and Stephenson, 1961b: 229). Norris and Wynne (1968) reported G. pusillum Le Jolis (1863: 139) from Washington, but there are few (Widdowson, 1974; Scagel et al., 1986) there are few subsequent reports of this taxon in the study area. Similarly, there are few records of Pterocladia caloglossoides Dawson (1953: 76), since it was reported by Norris and Hollenberg (1969: 116). The most recent addition to the gelidiaceous flora of British Columbia was G. purpurascens Gardner (1927a: 279) (Scagel, 1973; Waaland, 1973). A larger number of gelidiaceous taxa occur in adjacent regions of the Pacific (Table 4). To the south, 11 species of Gelidium, six species of Gelidiella Feldmann et Hamel (1934: 529) and four species of Pterocladia are present in California, Pacific Mexico and the Gulf of California (Dawson, 1944, 1953; Stewart 1976; Stewart and Norris, 1981; Santelices and Stewart, 1985). From Pacific Panama, two species of Gelidium and one species of Gelidiella were Taxonomy and Nomenclature / 14 reported in an ecological study (Lubchenco et al., 1984). Nine species of Gelidium, one Gelidiella and four Pterocladia are recorded from Pacific Colombia, Peru and Chile (Santelices and Stewart, 1985; Santelices and Montalva, 1983; Schnetter and Bula Meyer, 1982; Dawson et al, 1964). From Hawaii, four species of Gelidium, four Gelidiella and four Pterocladia are currently recognized (Santelices, 1977; Santelices and Stewart, 1985). Numerous species of Gelidium and Pterocladia, plus species of Beckerella, Acanthopeltis, Gelidiella, Onikusa and Yatabella have been reported from the northwest Pacific. From Japan, 24 Gelidium species, two Gelidiella species, four Pterocladia species, two Beckerella species and a single species each of Onikusa, Acanthopeltis and Yatabella are reported (Akatsuka, 1986b). For China, eight Gelidium species, one Gelidiella species and two Pterocladia species are recorded (Bangmei et al., 1983; Santelices and Stewart, 1985) some in common with Japan. Hommersand (1972) noted some floristic elements in common between northeast and northwest Pacific areas, suggesting possible floristic affinities of some Gelidiaceae between British Columbia and the northwestern Pacific. At present, the distinction between Gelidium and Pterocladia can only be definitively determined by the number of locules in the cystocarp (Santelices, unpubl. mscr.). Several other vegetative features have been suggested for generic separation. Characters such as rhizine distribution (subcortical or outer medullary in Gelidium versus inner medullary in Pterocladia by Okamura, 1934), morphology of emergence of branches from the main axis (branches basally bent in Gelidium and unbent in Pterocladia, Stewart, 1968) and shape and orientation of cortical cells (Akatsuka, 1981, 1986a) have been found inconsistent and unreliable when Taxonomy and Nomenclature / 15 tested on larger numbers of taxa (Rodriguez and Santelices, unpubl. mscr.). Recently, Rodriguez and Santelices (1987, unpubl. mscr.) proposed shape and position of the apical and lateral initials to distinguish these two genera. They proposed that in Pterocladia apical and lateral initials are indistinguishable in shape and always found depressed below the level of outer cortical cells, whereas in Gelidium, the lateral initials are different in shape from apical initials and are never present in cortical depressions (or could be, in a few cases, in indistinct depressions). These characters were consistent for a large number of Pacific species, with the exception of Pterocladia bulbosa Loomis (1960: 7) (Rodriguez and Santelices, unpubl. mscr.). All of the features were tested in the species of Gelidium and Pterocladia in this study and are discussed under each taxon. B. METHODS Material was hand-sectioned, or sectioned to 10-20 um on a Bailey Instruments Tissue-Freez freezing microtome, thick enough to determine relationships between cells as indicated by pit-connections and to trace files of cells. Transverse and longitudinal sections were stained using 1% aqueous aniline blue for 15-45 s, fixed with 10% HC1, washed with distilled water and mounted in 40% Karo clear syrup to which a small amount of Formalin and aniline blue stain had been added. Nuclei were observed by staining sections with Wittmann's hematoxylin (Wittmann, 1965) for one hour, followed by fixing with 45% acetic acid, and washing and mounting in either distilled water or in 40% Karo (with a few drops of Formalin added). In some specimens, apices, tetrasporangial stichidia, male or female Taxonomy and Nomenclature / 16 gamete-producing branchlets and/or cystocarps were excised. Tissue was fixed in 2.5% glutaraldehyde in Sorensen's phosphate buffer (pH 7.2), washed in buffer, dehydrated through a graded methanol series, infiltrated through a graded series of JB4 catatysed solution A concentrations and embedded in activated JB4 methacrylate (Ruddell, 1967; Appendix 2). The JB4 blocks were sectioned on a Sorvall JB4 microtome. Other material was embedded in Spurr's low-viscositj' epoxy resin (Spurr, 1969) and fixed following the recipe described under transmission electron microscope (TEM) fixation (Appendix 3). Blocks were sectioned with a glass knife on a Reichert OM U3 ultramicrotome at approximately 2.5 M m . Sections of embedded material were then stained with toluidine blue (pH 4.4) for 20 s, rinsed in tap water and mounted in 40% Karo clear syrup with Formalin and aniline blue stain added. For each taxon, measurements giving maxima or minima (e.g. maximum height) are taken from all specimens studied. Axis width and branch angle measurements were made from 10 plants, selected haphazardly. For cortical cell, medullar}' cell and rhizine sizes, 20-50 measurements were made on 4-5 plants of each taxon from different localities for which fresh or Formalin-preserved specimens were available. Carpogonial size was taken from 4-5 measurements on 4-5 plants; cystocarp diameter was from 10 measurements on 1-2 plants; carposporangial size was from 20 measurements on 4-5 plants. Twent}7 measurements of spermatium diameter were made on 2 plants. Tetrasporangial stichidium length and diameter of tetraspore release holes was based on 16-20 measurements of 4 plants; tetrasporangial size was taken from 30-40 measurements of 3-4 plants, except for Pterocladia caloglossoides where sporangial size was from 15 measurements on 3 Taxonomy and Nomenclature / 17 plants. The "!" notation, given with the information on type specimen for the species in this study, indicates that I have seen the type(s), and n.v. (non vide) indicates that I have not seen the specimen(s). Rhizine ultrastructure was observed in sterile (from Orlebar Point, Gabriola Island) and spermatangial thalli (from male gametophytes cultured from tetraspores of whalebone Bay, Gabriola Island) of Gelidium purpurascens, and in the spermatangial part of G. vagum gametophytes from Tribune Bay, Hornbj' Island. Longitudinal sections of axes 1 mm long, from 3-5 mm proximal to the branch apex (for young rhizines) and from 2-3 cm proximal to the branch apex (for older rhizines), were prepared for T E M study. C. TAXA PRESENT IN BRITISH COLUMBIA AND NORTHERN WASHINGTON Four gelidiaceous taxa in the study area are morphologically distinct and reproductive characters confirm their identities (Fig. 5). The Gelidiales in the local flora are Gelidium coulteri, G. purpurascens, G. vagum Okamura (1934: 58) and Pterocladia caloglossoides. Specimens upon which records of G. robustum, G. crinale, G. sinicola and G. pusillum were based were re-examined and found to be misidentifications of the four taxa reported herein. The report of G. pulchrum Gardner (1927a: 279) (Stephenson and Stephenson, 1961b) is unsubstantiated. Taxonomy and Nomenclature / 18 1. Gelidium coulteri Harvey 1853: 117 Synonym: G. undulatum Loomis, 1960: 4, pi. 6, fig. 1-4, n.v. Lectotype: TCD, collected by Dr. Coulter. Selection of the lectotype was by an unknown person, and it is not known if the designation was ever published (Parnell, pers. comm. 1988). Harvey (1853) described a tetrasporangial plant, so the lectotype specimen should be a tetrasporophyte, n.v. Type locality: California, U.S.A. Isotypes: TCD!, AHFH 53929! a. Habitat and Habit This is the only local Gelidium species that is restricted to the intertidal, occurring epilithically at relatively protected sites from the mid-littoral to the 0 m tidal level (Canadian chart datum). Plants are soft and non-cartilaginous, although short plants can be slightly stiff. They are brown to reddish-brown when fresh and dry to brown-black. The growth form is a turf, small thalli (to 2 cm tall) are dense, whereas taller thalli (to 5 cm tall) are less dense. Plants from protected sites tend to form taller, looser turfs. Upright axes arise from branched, creeping axes, and may divide to form 2-3 main axes, up to 5 cm tall, but more commonly 1-3 cm tall. Branching is sparse and usually only to two orders, though occasionally some Taxonomy and Nomenclature / 19 third order branches may be present (Fig. 6). Branch angles are wide, 45-60°, on widety spaced, irregularly to suboppositely positioned branches. Many . branches are short and determinate, giving axes a narrow outline. Some short plants appear densely branched but to a higher order, since branches are more closely spaced together with relatively fewer short proliferous-like branches. Axes are terete and usually narrow, to 0.5 mm wide although occasionally axes may be up to a maximum of 1 mm and compressed. This wider, compressed form was less common. Each branch order is narrower than the one it arises from. Branches taper abruptly at the apex, but branch bases are untapered and unconstricted. Creeping, terete axes of smaller diameter than uprights, that produce occasional, scattered, pads attach the thalli to the substratum. Attachment pads are colourless, formed of elongated cortical and medullary cell extensions amassed together, and somewhat spreading on contact with the substratum (Fig. 8). Attachment pads are absent in upright axes. b. Vegetative Anatomy Thallus construction is uniaxial, with a domed apical cell (Fig. 7) producing a subapical cell that cuts off two periaxial cells. In transverse section, cortical and medullary layers are apparent (Figs. 9, 10). The outer cortex is two to three cell layers and is composed of cells of approximate^ ' the same size. Outer cortical cells of uprights are isodiametric, 3-8 uffi long and 3-8 um wide, those of the creeping axes are larger and more oval, 6-16 um long and 4-9 um wide. Taxonomy and Nomenclature / 20 In surface view, cortical cells of both upright and creeping axes (viewed approximately halfway between plant apex and base) are randomly oriented and equidistant, showing no pattern or groups (e.g. pairs or tetrads) (Fig. 11). Secondary pit-connections are absent between the outermost cortical cells, but are present between second and third cortical cell layers of the cortex. The inner cortex, of 1-2 cells, grade to the medulla (Figs. 9, 10), and become progressively larger and more elongate inward. The medulla occupies approximately 75% of the axis radius (Fig. 9), with elongate, thick walled cells, 28-228 Mm long and 10-21 Mm wide, that are widely spaced, and have abundant secondary pit-connections to adjacent cells (Fig. 10). Intercellular spaces are packed with rhizines (Figs. 9, 10), which are also present in the inner cortex, being sparse in young axes but abundant in older ones. Unicellular and thick walled, rhizines are small diameter (3-4 Mm) and are very elongate (at least 500-700 Mm long) (Fig. 10), and also present in creeping axes. c. Reproductive Anatomy Tetrasporangia are restricted usually to determinate fertile branches (stichidia). Stichidia in Gelidium coulteri are elongate, (proportionately longer and narrower than in other local Gelidium species), 0.7-1.9 mm long and 0.2-0.4 mm wide, and general^ appear somewhat like proliferous branchlets (Fig. 12). The stichidia maj' themselves bear a few short branchlets (Fig. 13), arranged irregularly alternate to subopposite on higher branch orders or on distal parts of major axes (Fig. 12). Tetrasporangia are scattered irregularly on stichidia (Fig. 14), with a tendency for older tetrasporangia to be in the more basal parts of stichidia. Taxonomy and Nomenclature / 21 Older stichidia produce additional tetrasporangia following release of those formed first (Fig. 16). Developing tetrasporangia first are recognizable 90-300 nm behind the apex, are cut off from a cortical cell 3-4 cells below the surface. The pit-connection to the bearing cell is basal initially and becomes lateral as the sporangium increases its size with much inward expansion (Fig. 15). Tetrasporangia divide successively [i.e. cytokinesis of the first (transverse) division is completed before the second cytokinesis is initiated], to form four cruciately arranged tetraspores (Figs. 15, 17). At maturity, tetrasporangia are oval and deeply embedded in the cortex, but with their outer ends close to the thallus surface (Fig. 17). Mature tetrasporangia are 26-41 um long and 16-31 Mm wide, with adjacent cortical cells usually elongated and curved around them. Gametophj'tes of Gelidium coulteri have not been collected from the field in the study area, but male gametophytes have been grown in culture from released tetraspores. Female gametophytes were not detected in cultures, possibly because of low numbers of germinated sporelings. Gametophytes are assumed to be dioecious because male gametophytes only produced male gametes. Spermatangia are borne in superficial sori near the bases of indeterminate branches. Strongly fertile male gametophytes produce spermatangia along the branch to within a few millimeters of the apex. At low magnification, spermatangial areas appear pale pink (vs. brownish red where not fertile) because spermatangia are not pigmented. The fertile area appears slightly thicker because the wall is ruptured by spermatangial release, and an abundance of mucilage is present (Fig. 18). Spermatangial mother cells differentiate from cortical cells and Taxonomy and Nomenclature / 22 are half the diameter of vegetative cortical cells. Each spermatangial mother cell cuts off a single spermatangium by a transverse (periclinal) division (Figs. 19, 20). Spermatia released from spermatangia are small, 2.1-3.1 Mm at widest diameter, and hemispherical to spherical in shape. d. Discussion The intertidal habitat, few orders of branches, narrow branch outline, spindle-shaped tetrasporangial stichidia, turf-like growth habit and dark colour distinguish Gelidium coulteri from other gelidiaceous taxa in the study area. Particularly diagnostic are the abundant, short, simple branchlets. I found that British Columbia plants of G. coulteri correspond well with G. coulteri in California (Figs. 21-24). A comparison of type material (lectotype and isotype in TCD, Figs. 22, 24; isotype A H F H 53929, Fig. 23) from Monterey Bay, California with British Columbia plants indicates that plants from the two localities are conspecific. They have similar branching patterns, but Californian plants are more branched and appear bushier. The fertilitj' (tetrasporangial) of some of the Californian plants also adds another order of (stichidial) branches, contributing to their bushy appearance. Cystocarps of Gelidium coulteri are known from California, but evidence for the placement of this species in Gelidium is supported additional^' by features of apical and lateral branch initials. Apical initials are large and domed, protruding beyond adjacent cortical cells and form acute branch apices (Fig. 7). Lateral initials differ from apical initials, being inverted conical, but are similar in Taxonomy and Nomenclature / 23 position to apical cells as both initials are not in depressions of the cortex. These observations agree with Rodriguez and Santelices' (1987, unpubl. mscr,) proposed use of vegetative features to characterize Gelidium. disagreement with their Pterocladia characters. A number of reports pertaining to collections of Gelidium coulteri from British Columbia and several misidentified herbarium specimens of G. coulteri (in UBC), show that this species has been confused with other taxa. At times G. coulteri has been referred to erroneously as "G. crinale" (e.g. Scagel, 1957 - CANA 3474 (formerly as CAN 207); Scagel 1973 - UBC A40467, UBC A40468, UBC A40469, UBC A40470, UBC A31363, UBC A31426, UBC A37402, UBC A53975, UBC A60236, UBC A60254, UBC A60268, UBC A64934), as "G. pusillum" (e.g. UBC A24837, UBC A28351; FHL 3055), as "G. sinicola" (e.g. UBC A1449, UBC A14403, UBC A36545), as "G. robustum" (e.g. South, 1968 vouchers UBC A28645, UBC A29575, UBC A36198, UBC A36199), as "Caulacanthus ustulatus" (UBC A31363) and as "Endocladia muricata" (UBC A12298, UBC A13259, UBC A19739). In a few cases (UBC A30401, UBC A312511, UBC A37605), Pterocladia caloglossoides has been misidentified as "G. coulteri". These other taxa (G. crinale, G. pusillum, G. sinicola and P. caloglossoides) are generally poorly understood, and probably accounting for their confusion with G. coulteri. The report of Gelidium sinicola by Stephenson and Stephenson (1961b) from the area of Nanaimo, British Columbia may represent a mistaken identification of G. coulteri. Unfortunately, the authors gave no voucher numbers, description or Taxonomy and Nomenclature / 24 figures of the reported taxon, and therefore the plants' identity cannot be confirmed. Collections from the area (Orlebar Point, False Narrows, Lock Bay and Davisons Beach, all on Gabriola Island, near Nanaimo), however, show that G. coulteri is common. Stephenson and Stephenson were listing plants in the infralittoral zone, and they referred to G. sinicola as a short, moss-like alga. They also reported the presence of G. pulchrum, a larger plant with a wiry texture (probably G. purpurascens, see discussion later in this chapter). This allusion to a turf-like growth habit and short stature suggests that G. coulteri is probably the taxon that Stephenson and Stephenson (1961b) were calling G. sinicola. e. Seasonality and Distribution Gelidium coulteri is perennial, persisting as short uprights when not growing actively. Year round observations of G. coulteri were made at Orlebar Point, Gabriola Island (Fig. 3). Plants began showing growth from regenerated apices in early spring (March) and continued growing throughout the summer. They became fertile (tetrasporic) in August, remained fertile until November, but those few plants that were fertile did not produce tetrasporangia abundantly. Growth slowed in November and December, indicated by increasing overgrowth of apices by diatoms. Plants ceased growth in January and axes were eroded back to shorter lengths. Apices could not be detected and branch ends appeared truncated and broken. Gelidium coulteri is common throughout most of the study area and was recorded Taxonomy and Nomenclature / 25 from almost all of my collection sites that had Gelidium or Pterocladia. Its range extends north on both the open Pacific and Strait of Georgia shores of Vancouver Island, and on the British Columbia mainland to the Queen Charlotte Islands. The most northerly collection is from Pincer Island, British Columbia (52°11'N latitude). Gelidium coulteri ranges south, through Oregon (Doty, 1947) and California to Punta Pequena, (Pacific) Baja California del Norte, Mexico (Dawson, 1953). 2. Gelidium purpurascens Gardner 1927a: 275 Synonyms: Gelidium contortum Loomis, 1960: 4, n.v. Gelidium densum Gardner, 1927a: 278, n.v. Gelidium distichum Loomis, 1949: 2, n.v. Gelidium gardneri Loomis, 1960: 5, n.v. Gelidium papenfussii Loomis, 1949: 1, n.v. Gelidium polystichum Gardner, 1927a: 276, n.v. Gelidium pulchrum Gardner, 1927a: 279, n.v. Gelidium ramuliferum Gardner, 1927a: 279, n.v. Gelidium setchellii Gardner, 1927a: 275, n.v. Holotype: UC 93572 cystocarpic! Type locality: Moss Beach, San Mateo County, California, U.S.A. a. Habitat and Habit Taxonomy and Nomenclature / 26 Gelidium purpurascens is one of the more conspicuous gelidiaceous species in the study area. It occurs epilithically in protected sites in the shallow sublittoral from 0-12 m depths, but is most abundant and vigorous at 0-5 m depths. Gelidium purpurascens may be present at localities where there is much siltation, but it is absent from kelp and seagrass beds. This species is often one of the dominant seaweeds where it occurs. Plants are robust, cartilaginous, stiff (they are stiff enough to support their branches when out of the water), and dark red to maroon and dry to blackish-red. Plants grow as individuals or in small clumps and do not form a turf. Individuals may be much-branched and "brush-like" if older, or pyramidal and flat when younger (Fig. 25). Most of the biomass is in upright axes, which are attached to the substratum by a smaller system of creeping axes. Usually a single upright axis arises from creeping axes, but the upright may divide to produce 2-4 main axes that can grow to 15 cm tall. Plants are well-branched with up to four orders of branches (Figs. 25, 27). Branches are distichous to irregularly arranged, being subopposite to alternate in different parts of the plant or even along the same axis. Branches are indeterminate, such that older branches are longer and more branched than younger ones. This gives a pyramidal aspect to at least well-branched sectors of the plant. A divergent branching angle of 30-45° also contributes to the triangular outline of plants. Upper portions of plants appear best developed, Taxonomy and Nomenclature / 27 whereas lower portions often are less well-branched, possibly because of shading or damage from grazers or abrasion. Proliferous branches may develop, but these are not common or abundant. Axes are terete to compressed. Main axes may be 1 mm wide, and progressively higher orders of branches are successively narrower with the highest order branches to 250 (im wide. Branch apices are acute to rounded, with bases that are not tapered or constricted. Creeping axes are terete, narrower than major and second order upright axes, less deeply pigmented and sparsely and irregularly branched (Fig. 27). Attachment of creeping axes to the substratum is by occasional attachment pads (Fig. 33) that are colourless and formed from elongated cortical cells amassed together into a solid structure (Fig. 34). They are found only on creeping axes (there is no secondary attachment of upright axes). b. Vegetative Anatomy The thallus is uniaxial with a domed apical cell, not obvious at low magnification but easily visible in longitudinal or saggital sections (Fig. 28). Each subapical cell cuts off two, laterally opposite, periaxial cells. In transverse sections a pigmented cortex and unpigmented medulla can be distinguished with the inner cortex grading into the medulla without an abrupt transition (Figs. 26, 29). Surface cortical cells are smallest (4-9 Mm long and 2-7 Mm wide), and inner cortical cells are progressively larger (Figs. 29, 30). Outer cortical cells are isodiametric to oval with their longer axes, if detectable, randomly oriented (Fig. 31). Inner cortical cells are oval to short cylindrical, their length being approximately twice Taxonomy and Nomenclature / 28 their width (Fig. 30). Secondary pit-connections are abundant between inner cortical cells but are absent from the outer two cortical layers. Medullary cells are cylindrical, 18-107 jum long and 10-25 um wide with innermost cells longest (Fig. 30), and have abundant secondary pit-connections (Fig. 32). Rhizines are abundant in the inner cortex and outer medulla, but in decreased abundance in the central medulla (Fig. 26). They are present in creeping axes in the same position and abundance as in uprights. Rhizines are unicellular, narrow (3-4 um wide), very thick walled and unbranched (Figs 29, 30). They are cut off from the inner proximal corner of inner cortical cells 280-325 jum and develop proximal to the apex as a small protuberance, growing basipetally between inner cortical and medulla^ cells to at least 2-3 times the length of the longer medullary cells (i.e. at least 300 um long). The cell wall is the rhizine's most conspicuous feature. Its radius (1000-1500 nm) is greater than or equal to the diameter of the protoplast, and the wall is thicker in older cells (Fig. 36) than in younger ones (Fig. 35). Fibrils of the wall are interwoven, but generalty run parallel to the long axis of the cell (Fig. 39) and are loosest near the plasma membrane (Fig. 35). The ground matrix between fibrils is electron transparent. The metabolism of young rhizines appears to be directed toward wall deposition. The plasma membrane has a convoluted profile (Fig. 35). Mitochondria are prominent and closely associated with the forming face of dictyosomes (Fig. 39). Dictyosomes are abundant and large, 500-1000 nm wide and 600-800 nm tall (Fig. 39), their mature vesicles containing fibrillar material similar in diameter and electron densit}*- to the cell Taxonomy and Nomenclature / 29 wall fibrils (Fig. 39). However, there is no evidence of vesicles with fibrillar contents secreting to the wall, even in cells showing numerous secretory vesicles. Another form of vesicle (Figs. 37, 39), containing spherical and tubular bodies and resembling multivesicular bodies or lomasomes (Scott and Dixon, 1973b), is present. The origin of these vesicles is unclear, but they do not appear to be directly dictyosome-derived. There is evidence of many of these vesicles fusing with the plasma membrane and releasing their contents to the wall (Figs. 37, 39). Outside and immediately adjacent to the plasma membrane are many tubular body fragments that are not present in older walls, farther away from the cytoplasm. The tubular bodies must either degenerate or become dispersed, forming new wall. The abundance of both types of vesicles suggests the maturation of the fibrillar contents into tubular bodies and their subsequent release to the cell wall. In young rhizines, near their point of initiation from a subcortical cell, proplastids are present (Fig. 38, arrow) with onty an inner encircling thj'lakoid, a few plastoglobuli and areas of DNA. These proplastids, included in the rhizine cytoplasm at the time of division of the subcortical cell, do not replicate; they are seen only occasionally and are absent from older or more distant parts of the rhizine. There is much less secretion and wall deposition activity in old rhizines. The cytoplasm is vesiculate and vacuolate (Fig. 40), many of the vesicle membranes look degenerate, mitochoindria are few and the cytoplasm is thin and sparse. The plasma membrane profile is smooth (Fig. 36) and not convoluted as in more active cells. Taxonomy and Nomenclature / 30 c. Reproductive Anatomy Tetrasporangia are located in determinate, stichidial branchlets of upright axes (Fig. 41) that are short and clavate, compressed, and approximately twice as wide as vegetative branchlets of the same order or one order lower. Stichidia usually form on distal parts of plants or axes and impose an additional order of branching over the vegetative branching pattern. Thus tetrasporangial plants appear denser and more branched. Tetrasporangia are scattered over the stichidial surface, but are absent from the narrow sterile margin (Fig. 42). Sporangia are not oriented in rows, and different ages are interspersed due to the continuous production of tetrasporangia (Fig. 42) after mature ones are released. Similar ages of sporangia are approximately equidistant. Tetrasporangia are first distinguishable 30-60 um proximal to the apex (Fig. 43). The pit-connection to the subtending cortical cell is initially basal (arrowheads in Fig. 43), but later becomes lateral as the tetrasporangium expands deeper into the branch. Tetrasporangia are cruciately divided with the first division periclinal and the second (in each half) anticlinal (Fig. 44). At maturity tetrasporangia are 40-81 am long and 20-49 um wide. Tetraspores are pyramidal on release because of mutual compression in the sporangium, but soon round up. The cortex of the tetrasporangial part of the stichidium is thicker than the vegetative axes and occupies 2/3 of the axis radius. Inner cortical cells adjacent to the widest part of the tetrasporangium are elongated. Gelidium purpurascens is dioecious, and male and female gametophytes occur in Taxonomy and Nomenclature / 31 low numbers in field collections. Gametophytes are isomorphic with tetrasporophytes. Carpogonia and cystocarps occur terminally on highest order branches or on short, side branchlets of upright axes (Figs. 45, 48). These branchlets have the capacity to continue growing, but do not if reproduction is successful. Occasionally a second fertile area occurs when apical growth continues beyond the first set of carpogonia (Fig. 48). The apex of the carpogonial stichidium of G. purpurascens is acute, with the apical cell not recessed into an apical notch as reported by Hommersand and Fredericq (1988) for G. pteridifolium from South Africa. Numerous (20-32) carpogonia develop in the fertile area, and are initiated five to seven axial cells proximal to the apex (Fig. 47). The apex continues to grow, and later in development the cystocarp appears 1-3 mm proximal to the apex. Carpogonia are 13-29 um long and 3.5-5.5 um wide. Sizes are approximate as measurements did not account for the curvature of cells. Carpogonia differentiate in third order filaments above and below the second order plate (i.e. a plate of second order filaments connected by secondary pit-connections), and their trichogynes project to both surfaces of the branchlet (Fig. 47). Carpogonia are intercalary in vegetative filaments and have two pit-connections (Fig. 46). The carpogonium is the second cell in a third order filament cut off from a cell of the second order plate, close to the axial filament. The upper pit-connection, of the carpogonium is to a vegetative cell (Fig. 46) that produces higher order vegetative branches. In non-fertile axes, the cell in the position of the carpogonium has a second upper pit-connection (three pit-connections in total) to another branched vegetative chain. Thus, the carpogonium replaces a vegetative Taxonomy and Nomenclature / 32 chain whose outermost cells contribute to a small patch of thallus surface. The absence of surface cortical cells, which have been replaced by the carpogonium, leaves a small gap or "hole" in the thallus surface. The confluence of adjoining "holes", resulting from a double row of carpogonia, creates a medial furrow in the upper and lower surface of the female branchlet. Trichogynes protrude through this gap. Carpogonia farther back from the apex are sessile, cut off from the subtending cell by a distinctive concave wall. It was determined (see Chapter 5) that intercalary carpogonia are functional and may be fertilized. When they are past fertility, the concave wall cuts off the distal part of the carpogonium, which then degenerates while the "supporting" cell remains functional. Upon fertilization, nutritive filaments composed of short, curving chains of small, densely staining, globular cells, proliferate from basal cells of third order filaments and curve towards and around second order plate cells. The fertilized carpogonium forms a large, multilobed, multinucleate fusion cell by fusing with surrounding cortical cells. The fusion cell cuts off gonimoblast initials that form a branching, filamentous gonimoblast and that produce uninucleate carposporangia terminally and laterally (Fig. 51). A more detailed description of post-fertilization development is given later in Chapter 5. Mature cystocarps of Gelidium purpurascens, 420-560 um in diameter, are bilocular with one ostiole per locule (Fig. 50). They are spherical, protruding from both surfaces of the branch and are wider than the sterile portion of the branch Taxonomy and Nomenclature / 33 (Fig. 50). The carposporophyte is centered around the second order plate (Fig. 50) of gametophyte cells. The cystocarp cortex is the same thickness as the vegetative cortex, but it is composed of more cell layers. Cystocarp cortical cells are smaller and somewhat more tightly packed than in the vegetative cortex. Mature cystocarps of G. purpurascens have characteristic filaments between second order plate cells and the cortex (Figs. 49, 50). These filaments are third order files of cells that arise from cells of the second order plate, contribute to the cortex and extend to the surface. As the cj^ stocarp cavity expands and the cortex is pushed away from the second order plate, third order filaments stretch but do not break and are thus intact in the mature cystocarp. There is no evidence to suggest a nutritive function for these stretched filaments; they are simply part of the vegetative construction of the thallus over which cystocarp morphology is imposed. Filaments of a similar appearance have been reported as "nutritive filaments" in other genera, e.g. Gracilaria (Dawson, 1949), but they are not homologous to "nutritive filaments" of Gracilaria. Carposporangia are 20-39 um long and 12-22 um wide, ovate to obpyriform, uninucleate, pigmented, have a large, stellate chloroplast and are formed singly, i.e. not in chains (Fig. 51). Thej' have a thick coating of mucilage or polysaccharide that fills the cystocarp cavit3'. Carpospores are released through an ostiole in the pericarp wall. Spermatangia are produced in superficial sori on short stichidial branchlets of upright axes (Fig. 54), similar in position and arrangement (Fig. 52) to tetrasporangial stichidia. In living plants at low magnification the fertile area has Taxonomy and Nomenclature / 34 a colourless halo (Fig. 54) that corresponds to the spermatangial mother cell wall ruptured by gamete release, partially empty of spermatia and surrounded with large amounts of mucilage. Surface cortical cells 200-300 um proximal to the apex are transformed by one to two longitudinal divisions into two to four spermatangial mother cells replacing a single outer cortical cell. Each spermatangial mother cell cuts off a single spermatangium by a characteristic transverse (periclinal) division (Fig. 56). As the spermatangium develops, an enlarging spermatangial vacuole appears basally as a clear area at high magnification. A single, conical spermatium, 3.1-5.2 um diameter, is released through a narrow channel in the thickened wall (Figs. 55, 56), and rounds up into a spherical or short oval shape (Fig. 53). d. Discussion The bilocular cystocarp of this species dictates generic placement in Gelidium, as opposed to Pterocladia. Rodriguez and Santelices' (1987, unpubl. mscr.) proposed that Gelidium could be separated from Pterocladia using vegetative characters, of which two have been investigated here: 1) apical and lateral initials are morphologically different in Gelidium, whereas they are similar in Pterocladia, and 2) lateral initials never are found in indentations of the cortex in Gelidium whereas they are in Pterocladia. Apical architecture of G. purpurascens plants from the study area and from California agree with this suggested pattern: lateral initials are not in depressions and they are inverted conical in shape (Fig. 58), whereas the apical initial is domed and hemispherical (Fig. 57). Rodriguez and Santelices (unpubl. mscr.) also observed that the cortical cell pattern of the Taxonomy and Nomenclature / 35 basal 1-2 cm of upright axes could be used to discriminate between Gelidium and Pterocladia, but the cortical cell pattern was not consistent elsewhere on the axes as proposed by Akatsuka (1981, 1986a). In surface view, basal cortical cells of G. purpurascens are isodiametric, equidistant and are not aligned parallel to the axis (Fig. 60), in agreement with Rodriguez and Santelices' (unpubl. mscr.) characterization of Gelidium. Localized in small patches at the very base of uprights, elongate, elliptical, cortical cells have their long axes perpendicular to the branch axis (Fig. 61). This is unlike the pattern noted for both Gelidium (isodiametric and no alignment) and Pterocladia (elliptical and parallel to branch axis) (Rodriguez and Santelices, unpubl. mscr.), but this pattern is not universal or widespread on a branch. The British Columbian entity called Gelidium purpurascens corresponds well with the holotype of G. purpurascens (UC 93572) (Figs. 62, 63) and other Californian material belonging to this species. Diagnostic characters given by Gardner (1927a) are that G. purpurascens (Figs. 62, 63) is tall, but not as large and coarse as G. robustum (Hollenberg and Abbott, 1965); well branched, including lower parts of axes; has compressed, basally constricted and geniculate pinnate branches to four or five orders; produces initially single cystocarps on fertile branchlets and later exhibits proximal, lateral, cystocarpic branchlets. Stewart (1976) modified the description to include geniculate or non-geniculate plants, with or without abundant basal branching and did not specifiy basal branch constriction as a diagnostic character. Plants collected in the study area fit the redefined description, particularly as branches of G. purpurascens are not basally constricted when compared to the marked basal constriction of higher order branches of G. robustum. Taxonomy and Nomenclature / 36 Stewart (1976) generally described Gelidium purpurascens as large and similar in branching pattern to G. robustum, but not as coarse or tall as the latter species. A comparison of the holotype (UC 93572, Figs. 62, 63) and of specimens of G. purpurascens from the study site, with the holotype of G. robustum (UC 294572) (Fig. 138) showed that G. purpurascens and G. robustum are different morphologically. Although both species show the same number of branch orders, G. purpurascens is more proliferous and more densely branched. In G. purpurascens there are often several dominant axes that may divide into equivalent axes, whereas G. robustum has a single main axis obvious throughout the length of the plant. Transverse sections of vegetative branches of G. purpurascens show that the subcortex may be thick but that ordered rows of cells are not evident; similar sections of G. robustum branches also show a thick subcortex, but it is ordered into pallisade-like rows. Tetrasporangial stichidia also differ between these species; stichidia of G. purpurascens have a narrow, almost inconspicuous, sterile margin, whereas stichidia of G. robustum have a wide sterile margin. Thus, despite some similarity in branching pattern, G. purpurascens can be distinguished morphologically and anatomically from G. robustum (Table 5). Some of the earliest Gelidium collections from British Columbia are of G. purpurascens (CANA 3473, CANA 3740 (Fig. 59), CANA 3843, CANA 4346, CANA 4349, UBC A1952). These were called G. cartilagineum var. robustum (Scagel, 1957), and later G. robustum when Hollenberg and Abbott (1965) Taxonomy and Nomenclature / 37 elevated the variety to species status. Wherever possible, subsequent reports of G. purpurascens in the study area (Scagel, 1973; Foreman, 1977; Lindstrom and Foreman, 1978; Pueschel and Cole, 1982) have been confirmed. Publications that reported "G. robustum" in the study area (Scagel, 1967; South, 1968, record from Twin Beaches, Gabriola Island; Scagel, 1973), cited voucher herbarium specimens or illustrations, correspond not to G. robustum, but to G. purpurascens. e. Seasonality and Distribution Gelidium purpurascens is perennial, growing from April through October, but only persists as uprights with little growth (demonstrated by heavy diatom epiphytism) over winter months. Tetrasporophytes are fertile from February to December, most abundantly in May to October. Female gametophytes bear carpogonia in August and November collections, and male gametophytes are fertile in June, Jul}' and November. Gelidium purpursacens is widespread throughout the study area and found along the British Columbia mainland, Queen Charlotte Islands, Hornby Island and Denman Island, Strait of Georgia, Gulf Islands, Strait of Juan de Fuca and along the length of the west coast of Vancouver Island. It is abundant at sites where it occurs and is present at all of m}' collection sites where any other species of Gelidium and/or Pterocladia are found. Tetrasporophytes are fertile throughout the study area but fertile gametophytes are rare. Female gametophytes have been collected onty from Diana Island and Taxonomy and Nomenclature / 38 Geer. Islets, both in Barkley Sound (Fig. 2), and male plants have been identified only from Nootka Sound, Vancouver Island (Fig. 52), Esteban Point, Vancouver Island and Barkley Sound. Gelidium purpurascens is distributed from Baja California del Norte (Stewart, 1976), to southeast Alaska (UBC A69432, UBC A69433). Its morphology is plastic, although in British Columbia it is distinctly different from other local species. In some environments plants are much less robust, paler and branches become long and flexuous, especially near apices, appearing similar to G. vagum Okamura (1934). The two species are still distinguishable by the cartilaginous texture of G. purpurascens (vs. lax and soft in G. vagum), the terete to compressed axes of G. purpurascens (vs. compressed to flattened in G. vagum), the narrow sterile margin of tetrasporangial stichidia in G. purpurascens (a sterile margin is absent in G. vagum), dioecious gametophytes of G. purpurascens (vs. monoecious gametophytes in G. vagum), and the presence of stretched third order filaments in mature cystocarps of G. purpurascens (these filaments are broken in G. vagum cystocarps). 3. Gelidium vagum Okamura 1934: 58 Type: A holotype was not designated by Okamura (1934) in the protologue. Syntype specimens are in MAK (Akatsuka, pers. comm. 1987) but not available for loan. It is not known if Okamura designated a holotype, but no lectotype has been published. Taxonomy and Nomenclature / 39 Type locality: Not specified, although the distribution in Japan (around Honshu and southern .Hokkaido) was described (Okamura, 1934). a. Habitat and * Habit Gelidium vagum usually grows in the shallow subtidal from 0-13 m depths as well as in very low intertidal pools. Wherever plants or parts of plants have been exposed to air they are bleached and dead. This species is most abundant in the shallow subtidal, 0-4 m, where it may be a dominant, but it is not a dominant in populations of deeper water. Gelidium vagum grows epilithically on vertical walls and tops and sides of boulders, occuring in somewhat silted (e.g. Tribune Bay, Hornby Island) to protected localities (e.g. Denman Island and Ford Cove, Hornby Island) with a moderate amount of water motion. Plants are soft and lax and do not remain upright when exposed in air as does the more rigid G. purpurascens. They range from red to yellowish-red (deeper plants), and dry to brownish red, and grow as individuals, not as a turf, even though several upright axes may arise from the same creeping basal axis. Plants may be up to 10 cm tall, but are usually shorter (Figs. 64, 65), and can be fertile (tetrasporic or gametophytic) in the field at 1 cm tall. Gelidium vagum shows considerable morphological variation in growth form, from thin and sparsely branched, to wide, robust and well branched. Generally, deeper-growing plants are thinner, lighter in colour, smaller and less branched. Plants are branched to the fourth order, and occasionally well-developed plants have fifth order branches (Fig. 64). Branching is divergent with a branch angle Taxonomy and Nomenclature / 40 of 45-60° for all ranks of branching. One to three main axes usually are distinguishable, although in some plants the main axis divides subdichotomously distally, and neither branch can be recognized as derivative. Branching is distichous and irregularly alternate to subopposite, and successive branches are well separated (approximately 1-2 mm apart). Shallow-growing plants may be quite bushy, whereas deeper plants are sparsely branched. Axes vary from compressed to flattened, with the most robust plants being conspicuously flattened, especially at branch points, and smaller, less robust plants terete to compressed. Main axes and lower order branches vary in width from 0.2-2.5 mm. Branch bases often are narrowest (but never constricted), with the branch axes widening gradually to the widest point, about 1-2 cm behind the apex. Higher order branches are usually progressively narrower and constricted basally, but may taper abruptly to acute apices, the distal 1 mm of which appears whitish or unpigmented in fresh material. Occasionally, ultimate branches become long and whip-like (Fig. 66). Higher order axes branch in the same way as main axes. Creeping axes are pigmented, irregularly branched, terete, narrower than most upright branches and bear attachment pads. Attachment pads are unpigmented, scattered irregularly along prostrate axes formed from the confluence of elongated cortical cells, and spread out at the point of contact with the substratum. b. Vegetative Anatomy Thallus construction is uniaxial, with a conspicuous, domed apical cell domed (Fig. 68). Cells of the axial row each cut off laterally two periaxial cells (Fig. 68). In transverse section cortical and medullary layers are distinctive (Figs. 67, 71), Taxonomy and Nomenclature / 41 with a smooth and rapid transition between inner cortex and outer medulla. Outer cortical cells are isodiametric to slightly elongate, 4.0-11.3 /xm long and 3.0-6.2 jim wide, with their longest axes irregularly oriented in the periclinal plane and equidistant (i.e. not associated into clusters) (Figs. 69, 70). However, near the apex, where cell divisions are occurring, the longest axes of outer cortical cells are in the anticlinal plane (Fig. 68). Secondary pit- connections are absent between cells of the outer two cortical cell layers, but are abundant between inner cortical cells. Inner cortical cells are larger and more elongate (16-60 ixm long, and 12-29 Mm wide) than outer cortical, and short and cylindrical with their long axis at approximately 30-45° to the branch axis. Medullary filaments are oriented periclinally and cortical filaments are anticlinal, so that the transition part of filaments of the inner cortex and outer medulla are oriented at oblique angles as noted by Akatsuka for G. vagum in Japan (1981, 1986a). Medullary cells are larger (75-290 Mm long and 10-29 Mm wide) than inner cortical cells, with the longest cells deepest in the medulla (Fig. 72). The}' are cylindrical, unpigmented, thick-walled and have abundant secondary pit-connections (Fig. 68). Rhizines are intercalary between inner cortical and medullar}' cells (Figs. 67, 71) and increase in abundance farther behind the apex. They are cut off from the proximal end of inner cortical and outer medullary cells (Fig. 73) approximately nine axial cells behind the apex. Rhizines are abundant in the inner cortex and outer medulla (Fig. 71), and later grow between medullary cells (Fig. 67). They Taxonomy and Nomenclature / 42 are unicellular, unpigmented, refractive, elongate (measured to at least 970 um and most probably are longer), have a narrow diameter (3-4 um), very thick walls (1.1-1.5 Mm), and do not form secondary pit-connections with adjacent cells. c. Reproductive Anatomy Tetrasporangial stichidia are produced on upright axes along with the vegetative branches (Fig. 76). Unbranched or once branched, 1.5 mm long and 1 mm wide, stichidia are usually determinate and arranged distichously, subopposite to alternate, cylindrical to compressed, ovate to lanceolate with a blunt-rounded to blunt-tapered apex (Fig. 75). Occasionally the stichidial apex continues to grow bej'ond a fertile area and produces another tetrasporangial area a short distance beyond the original patch (Fig. 74), however, usually the apex does not continue vegetative growth but produces one to three short stichidial branchlets just proximal to the original fertile area. Tetrasporangial stichidia may be produced on narrow or wide plants over one cm tall and may be present on any branch order, but usually are not present on the main axis if there is more than one branch order. Tetrasporangia are scattered randomly, i.e. not arranged in rows or V pattern, over the stichidial surface. There are generally more mature tetrasporangia in the basal part of the stichidium as these differentiate first, but, after release of the first-formed sporangia, secondary production of tetrasporangia results in a mixture of sporangia of different ages (Figs. 79, 81). Tetrasporangia occur around the entire diameter of the axis, and a sterile stichidial margin is lacking (Fig. 75). Taxonomy and Nomenclature / 43 Six to seven axial cells proximal to the apex (Fig. 77) tetrasporangia are initiated from inner cortical cells (Fig. 79). They are terminal on one of the two branches borne by a cortical cell. The sporangial pit-connection is initially basal (Fig. 79), but becomes lateral (Fig. 83) by the basal expansion of the sporangium below the pit-connection. The first division of the tetrasporangium is transverse (Fig. 81), and the second division produces four cruciately arranged tetraspores (Fig. 78), 46-89 um long and 25-46 /um wide. Spores are released through small holes, smaller than spore widths, (diameter 5.1-10.0 um) in the sporangial wall (Fig. 80). Either spores must be squeezed during release, or holes are stretched. The space remaining after spore release contains a substance that stains weakly with aniline blue and appears to have been partially drawn or squeezed out of the cavitj' (Fig. 83). Based on similar observations of other workers studjdng tetrasporogenesis of red algae (Peyriere, 1970; Pueschel, 1982), this substance is probably mucilage or polysaccharide from sporangium wall breakdown or a spore secretion. Gelidium vagum is monoecious, with male and female gametes consistently formed in adjacent sections of apical parts of gametophyte branches (Fig. 85); no fertile gametophytes have been seen that produce only one sex of gametes. Gametangia are borne on short, determinate stichidia of upright axes (Figs. 82, 84), but occasionally stichidia bear two more stichidial branchlets. Stichidial apices occasionally may continue growth after a cystocarp has been produced and give rise to another patch of spermatangia and carpogonia 1.5-3 mm distal to the first gametangia. Gametangial stichidia are distichous and irregularly alternate to subopposite on second or third order branches (except on very small, 1-2 cm tall Taxonomy and Nomenclature / 44 plants where they are on the main axis). Spermatangia are located just proximal to the area that produces carpogonia (9-10 axial cells or 70-250 um behind the apex). At low magnification fertile male areas can be recognized as areas with a somewhat raised and thickened cuticle that corresponds to the layer of unpigmented spermatangia (Fig. 85). Carpogonia are produced three to six axial cells (50-220 um) proximal to the apex, and the region of carpogonial production extends proximally along the length of 4-5 axial cells (100-300 um). Carpogonia are elongate (10-25 nm long, and 3-6 um wide at the widest, basal part), curve forward and outward and taper gradually to a trichogyne that extends directly to the thallus surface (Fig. 90). The carpogonium is intercalary, the second to basal cell of a third order filament, and has two pit-connections, one to the next distal cell and one to the next proximal cell of the filament (Fig. 90). All other cells of the filament appear unmodified. A double row of carpogonia (from both lateral sides of the axial row) are directed toward the upper branch surface and another double row are directed to the lower branch surface because third order filaments extend both above and below the second order plate (Fig. 89). In the vegetative branch, cells in the same position as the carpogonium have one proximal pit-connection (to the basal cell of the file) and two distal pit-connections (to a distal cell in the same file and to a higher order file), but in the gametangial branch the carpogonium replaces one file and its subsequent branches, creating a gap in the branch surface. The alignment of gaps from adjacent carpogonia forms an axial furrow through which trichogynes protrude. Taxonomy and Nomenclature / 45 Associated with carpogonia and young carposporophytes are short, curved chains of two to four isodiametric cells, referred to as "nutritive filaments" (Fig. 88) (Hommersand and Fredericq, 1988). These are cut off from basal cells of third order filaments and curve toward the second order plate. A lobed, multinucleate fusion cell develops after fertilization, from the fusion of the carpogonium and cortical and subcortical cells (Fig. 92). Fusion cell lobes cut off gonimoblast initials that develop into gonimoblast filaments composed of small, elongate cells that do not stain darkly with aniline blue. Gonimoblast cells fuse with apical cells of nutritive filaments. The gonimoblast branches extensively around and between cells of the second order plate, producing single, obpyriform carposporangia, 35-75 um long and 15-25 um wide in one of two locules formed between the second order plate and cortex (Fig. 93). A large quantity of polysaccharide or mucilage, staining faintly with aniline blue, surrounds the carpospores, much of it remaining in the locules after carpospore release. A more detailed description of carpogonium and carposporophyte development is given in Chapter 5. As the cystocarp swells, third order filaments between the cortex and second order plate (Fig. 91) stretch and are broken, in contrast to Gelidium purpurascens cystocarps where filaments stretch but remain intact. Mature cystocarps are wider, 300-550 um in diameter, than the vegetative part of the bearing branch, bilocular, and locules are separated by the plate of second order filaments (Fig. 93). The carposporoplryte is restricted to the area around the second order plate. The cystocarp is domed at the ostiole, but not apiculate, and has one ostiole per locule (Fig. 93). Taxonomy and Nomenclature / 46 Spermatangia are cut off from spermatangial mother cells over most of the stichidial surface for 150-500 um immediately proximal (9-10 axial cells or 70-250 /um proximal to the apex) (Fig. 85) to the carpogonial area. Stichidia may include some small patches of vegetative cortical cells. At low magnification, fertile male areas can be recognized by their raised and thickened wall, corresponding to the layer of unpigmented spermatangia and mucilage (Fig. 85). Spermatangial mother cells are narrow and elongate (Fig. 86), and each produces a single spermatangium by a periclinal (transverse) division. A clear, rounded body corresponding to a spermatangial vacuole can be seen basally in some spermatangia. Each spermatangium releases a single spermatium (Fig. 87) 1.5-2.5 um in diameter. Spermatia are hemispherical to blunt-conical at release, later becoming spherical. From the first-formed (most proximal) spermatangial mother cells to the youngest (most distal) along the length of male area of the stichidium can be seen a progression and intermixing of spermatangial developmental stages, suggesting percurrent production of spermatangia. d. Discussion The bilocular cystocarp of this species indicates placement in Gelidium. Supporting this generic placement, apical and lateral initials are morphologically different and neither is in a cortical indentation. The monoecious nature of gametoplrytes of this species does not correspond to any Gelidium species previously reported from British Columbia. Gelidium purpurascens is morphologically distinct and dioecious. I have seen separate male and female gametophytes of G. purpurascens (male: UC 305373; female: UC 93572, UC 276633, UC 296689, UC 305364, UC 305371), Taxonomy and Nomenclature / 47 and G. contortion Loomis, G. distichum Loomis and G. irregulare Loomis, all considered synonyms of G. purpurascens, are known to be dioecious (Silva, 1978). Gelidium coulteri is dioecious and morphologically distinct (Macler and West, 1987), and Pterocladia caloglossoides also is morphologically distinct. Monoecj' has not been reported previously in Gelidium, but has been observed in Acanthopeltis japonica (Gelidiaceae) (Kaneko, 1968). Monoecy is a consistent feature of G. vagum from British Columbia and gametes are abundantly functional (judging from the large number of fertile cystocarps), eliminating the possibility of the observed condition being a genetic abberation. There are, however, a large number of Gelidium species for which males are unknown. Two such species, G. vagum from the northwestwern Pacific (Japan and China) and G. johnstonii from the Gulf of California, show a striking morphological resemblance to the British Columbian monoecious Gelidium. Illustrations of Gelidium vagum from China (Bangmei et al., 1983; Santelices, unpubl. mscr.) and Japan (Okamura, 1934; Segawa, 1959) are morphologically similar to the monoecious Gelidium from British Columbia. In particular, the original description and figures of G. vagum (Okamura, 1934 pi. 25) described a morphology completely within the range of form displayed by the British Columbian monoecious species. British Columbian plants also agree with Santelices' (unpubl. mscr.) description of Chinese G. vagum. To date, Okamura has provided the only illustrations of a cystocarpic plant and a cystocarp, but the diagnosis lacks illustrations or mention of internal cystocarp anatomy. Taxonomy and Nomenclature / 48 Because the type of Gelidium vagum, in the Makino Herbarium of the Tokyo Metropolitan University (MAK; Akatsuka, pers. comm.), was unavailable for loan, other specimens of Japanese G. vagum, borrowed from the National Science Museum, Tokyo [TNS 25817, TNS 25823 (Fig. 94), TNS 25824, TNS 25825, TNS 25847] and a single sheet with two to four Japanese plants in UBC (A56807), were examined. Two plants are cystocarpic [TNS 25824 (Fig 95), TNS 25825], and when sectioned were found to be monoecious (Fig. 96). Spermatangia in these plants are like those in the British Columbian monoecious species, occurring just proximal to the female gametangial area, and later at, and extending below, the cystocarp base. Cystocarp anatomy also is similar in British Columbian and Japanese plants (Fig. 96). Occasional^, however, cystocarps of Japanese plants have an intact third order filament between the second order plate and cortex (arrowhead in Fig. 97), whereas equivalent filaments in British Columbian material are broken during cystocarp expansion. As in the British Columbian material, tetrasporangial stichidia of G. vagum lack a sterile margin, and have tetrasporangia scattered irregularly over the stichidial surface (Fig. 98). Tetrasporangial stichidial anatomy is similar in both entities, although in Japanese plants, tetrasporangia are embedded somewhat more deeply in the thallus with one or two cortical layers lying outside the outer end of sporangia (Figs. 99, 100). In light of the similarity of other characters, i.e. the morphology of thalli with regard to branching pattern, branch size, shape and degree of constriction, presence of male and female gametangia not only on the same thallus (monoecy) Taxonomy and Nomenclature / 49 but in the same arrangement, the absence of a sterile margin in tetrasporangial stichidia and the arrangement of tetrasporangia in stichidia, the difference observed in breakage of sterile filaments linking cells of the second order plate and cortex, and in the degree of tetrasporangia embedding are considered minor. Thus, this monoecious British Columbian Gelidium is considered to be conspecific with G. vagum. Spermatangia also were unknown for Gelidium johnstonii Setchell et Gardner (1924: 742),1 when it was described from the Gulf of California, and illustrations (Setchell and Gardner, 1924 pis. 46a, 72, 73) showed a strong morphological resemblance between this species and G. vagum. Akatsuka (1986b) suggested that G. vagum was related closeh' to G. johnstonii. Examination of the holotype (Fig. 101), isotypes and paratype, as well as specimens identified as G. johnstonii collected by Dawson (AHFH 2211, AHFH 4150, AHFH 4156, AHFH 4179, AHFH 4192, AHFH 4193, AHFH 4194, AHFH 50267, AHFH 50268, AHFH 50299, LAM 52684 in AHFH, LAM 52894 in AHFH), revealed that small, to 5 cm tall, plants resemble G. vagum morphologically, but that larger plants are more robust and coarse, lacking the delicate appearance of G. vagum. Gelidium johnstonii and G. vagum differ anatomically in several vegetative and reproductive features listed below and summarized in (Table 6). Contrary to Setchell and Gardner (1924) the medulla of the holotype of G. johnstonii is not sparse, however the outer cortex is three to four cells as they had reported (Figs. 102, 1 Holotype: (Johnston 27) CAS 1343 in UC! Isotypes: (Johnston 27) CAS 484385 in UC!, CAS 484386 in UC!, CAS 484388 in UC! Paratype: (Johnston 13) CAS 484390 in UC! Type locality: Bahia San Francisquito,Gulf of California, Baja California del Norte, Mexico Taxonomy and Nomenclature / 50 103). In contrast, the outer cortical cells of G. vagum are unaligned. Tetrasporangial stichidia of Gelidium johnstonii are elongate, to 3 mm, spatulate (Fig. 106), flattened, and have a blunt apex (Fig. 104) and a sterile margin of variable width (Fig. 104) unlike the shorter, ovate, terete to compressed stichidia with more pointed apices of G. vagum. The apical initial of G. johnstonii is domed and protrudes slightly beyond adjacent cortical cells (Fig. 105), but in tetrasporangial and gametangial stichidia the apical cell is level with the adjacent cortex or recessed slightly between adjacent cortical lobes (Fig. 109). In G. vagum, the apical initial of both vegetative branches and fertile stichidia protrudes beyond adjacent cortical cells (Figs. 68, 75). In G. johnstonii a furrow develops immediately behind the apical cell in gametangial and tetrasporangial stichidia (Fig. 105, 109) which is not filled in by cells of higher order filaments for 250-350 Mm behind the apex. An apical furrow is absent in G. vagum tetrasporangial stichidia (Fig. 75), but is present in gametangial stichidia, as a result of cortical cells being replaced by carpogonia (Fig. 89, development described earlier). The furrow in G. vagum gametangial stichidia is not analogous developmentally to the furrow in G. johnstonii tetrasporangial stichidia. The mature cystocarp of Gelidium johnstonii (Fig. 107) has stretched but intact third order filaments (Fig. 108), as in G. purpurascens, and differs from G. vagum where third order filaments are broken. Significantly though, G. johnstonii is monoecious, male and female gametes appearing in the same position as in G. vagum (Fig. 111). Taxonomy and Nomenclature / 51 Despite the morphological similarity of smaller thalli of Gelidium vagum and G. johnstonii, and that both species are unique in being moneocious, these taxa are regarded as separate species because of several consistent and significant character differences (Table 6). Since they are both moneocious, a condition that is otherwise unknown in Gelidiales, and because of the uniformity of spermatangial position with respect to carpogonia, it is reasonable to propose that G. johnstonii and G. vagum are closely related sister species. A similar pattern of warm-temperate to sub-tropical, East Pacific—West Pacific disjunct distributions has been observed for Pachydictyon coriaceum (Holmes) Okamura, Tinocladia crassa (Suringar) Kylin, Endarachne binghamiae J. Agardh, Prionitis cornea (Okamura) Dawson, Ishige sinicola (Setchell et Gardner) Chihara and Lomentaria catenata Harvey, and for the species pairs Eisenia arborea Areschoug and E. bicyclis (Kjellmann) Setchell, and Carpopeltis divaricata Okamura and Binghamia California J. G. Agardh (Hommersand, 1972). In addition, numerous other species pairs have been suggested on the basis of preliminary observations (Hommersand, 1972), and Gelidium johnstonii and G. tenue Okamura (1934: 56) are found as a pair of taxa. The description and illustrations of G. tenue (Okamura, 1934, pis. 23, 31 fig. 8-10) are not strikingly reminescent of G. vagum, and a similarity between G. tenue and G. vagum has not been noted in discussions of Japanese or Chinese species of Gelidium (Akatsuka, 1986b; Santelices, unpubl. mscr.). No evidence has been presented that supports a sister taxa relationship between G. tenue and G. vagum. The combination of the 1) narrow range of distribution in the Pacific Northwest, Taxonomy and Nomenclature / 52 restricted to British Columbia of Gelidium vagum (see below), and 2) its great abundance and vigour within its distribution, suggests that this species may have been a recent introduction rather than a relict of a population of a once wider distributional range in the northwest Pacific. Many oyster farms are present in the Hornby Island-Denman Island area. As such, the importation of Japanese oyster spat is a likely vehicle for the introduction of G. vagum into British Columbia. Scagel (pers. comm.) has suggested that another species, Lomentaria hakodatensis, of the British Columbian algal flora may have also been introduced accidental^, probably in oyster spat. In view of the rapid colonization of Hornby Island and Denman Island, future range expansion of G. vagum in British Columbia might be expected. To date no other west coast of North America populations are known, although a single sterile specimen of G. vagum (UBC A64965) (Fig. 110) appeared in the drift at Ladysmith, farther south in the Strait of Georgia. e. Seasonality and Distribution Gelidium vagum is perennial in British Columbia. Fertile tetrasporophytes and cystocarpic plants have been collected in the fall (August, September, October), whereas plants collected in April were infertile. Japanese specimens (TNS; UBC) were tetrasporangial in May and July and gametangial in Jul}'. In British Columbia, Gelidium vagum is restricted to Hornby Island and Denman Island (Strait of Georgia). Taxonomy and Nomenclature / 53 In addition to the Strait of Georgia in the area of Hornby Island and Denman Island in British Columbia, G. vagum is distributed in the Huanghai Sea (Bangmei et al, 1983), and the Yellow Sea of China (Akatsuka, 1987; Santelices, unpubl. mscr.), the Korea Strait and Sea of Japan, (Tokida, 1954; Akatsuka, 1987; Kajimura, 1987), and the Pacific coasts of Honshu and southern Hokkaido (Okamura, 1934; Akatsuka, 1987). 4. Pterocladia caloglossoides (Howe) Dawson 1953: 76 Basionym: Gelidium caloglossoides Howe 1914: 96, fig. 7, pis. 34, 35 Synonym: Pterocladia parva Dawson 1953: 77 fig. 2, pl. 6, n.v. Holotype: Coker 59 (NY); lost on loan (Thiers, pers. comm. 1987), if no other specimens are found, a neotype should be made. Type locality: Island of San Lorenzo, Peru a. Habitat and Habit Pterocladia caloglossoides grows in the shallow subtidal, 1-7 m depths, in moderately exposed to protected locations. It is epilithic on tops and sides of boulders, on platforms or on walls, epiphytic on crustose corallines and also occurs on calcareous shells. Plants are soft, but not lax, and are dark red, drying to blackish red. They form a dense, low turf of upright axes that arise from an extensive system of prostrate axes (Fig. 112). Many uprights develop along the length of a single prostrate axis. Although the most vigorous plants Taxonomy and Nomenclature / 54 may reach 2 cm tall, the turf usually is not more than 1 cm tall. Branching in Pterocladia caloglossoides ranges from very sparse to abundant; the number of branch orders is low; uprights may be unbranched or may bear a maximum of two branch orders, though second order branches were uncommon in the study area. Branches are distichous, irregularly subopposite, and the branch angle is almost 90° (Fig. 113). Axes are compressed to flattened at branch bases, especially where branches are regenerating from cut ends. Upright axes vary greatly in width, and a single branch may range from 240-690 Mm. Axes taper gradually at their bases and abruptlj' at their apices to an acute tip. Prostrate and upright axes are similar, except that prostrate axes have attachment pads at frequent intervals whereas pads are lacking on upright axes. Occasionally an upright axis bends down, developing attachment pads and becoming prostrate. Branching of prostrate axes is subopposite distichous, but when producing uprights, one to five axes may arise from a single node. Uprights are perpendicular to prostrate branches and occur opposite an attachment pad (Figs. 114, 116). Prostrate axes are less variable in width (230-260 Mm diameter) than upright axes and are approximately the same width as narrower uprights. Prostrate axes do not taper at branch points, but have acute apices. Attachment of prostrate axes to the substratum is by numerous attachment pads formed medially on the ventral surface of creeping axes opposite the upright branches (Fig. 116) and formed on the lower surface of an upright branch that Taxonomy and Nomenclature / 55 is becoming prostrate. Cortical cells elongate, or produce elongate, unpigmented, rhizoid-like lobes that adhere to form attachment pads (Figs. 120, 121). Pads are to 1.0 mm long and are 0.2-0.5 mm wide and flare out distally, as the distal end of cells incorporated in the pad are swollen and bulbous. b. Vegetative Anatomy Thallus construction is uniaxial with a conspicuous, domed apical cell and two periaxial cells derived from each subapical cell (Fig. 115). Second order filaments grow laterally outward at a 45-60° angle to the axial filament. These are evident just behind apices where cells of second order filaments are in diagonal rows from the axial filament (Fig. 115). Transverse sections of Pterocladia caloglossoides reveal a cortex of only 1-2 cell layers (Fig. 117) and a medulla. The outer wall of superficial cortical cells is as wide as the protoplast diameter. Cortical cells in upright axes are pigmented, oval or short cylindrical, 7-10 um long and 5-10 um wide, and in surface view are equidistant, but unordered and irregularty arranged (Fig. 119). Cortical cells of creeping axes are slightly larger, but are otherwise like cortical cells of upright axes. Outer cortical cells lack secondary pit-connections, whereas the inner cortical cells (in specimens where two rows are present) have secondary pit-connections. The medulla is 3-5 cells thick and in upright axes composed of unpigmented, cylindrical cells, 12-27 urn long and 7-13 nm wide, that have abundant Taxonomy and Nomenclature / 56 secondary pit-connections (Figs. 117, 118). Medullary cells of prostrate axes are slightly larger. Cells of the axial filament are conspicuous in longitudinal section (Fig. 118). Rhizines, present in the medulla (Fig. 117), are very abundant in upright axes but are more sparse in creeping axes. They are initiated from the inner proximal corner of medullary or inner cortical cells, at a position 6-7 axial cells behind the apex. c. Reproductive Anatomy Tetrasporophytes are the only reproductive thalli collected from the study area. Gametophytes are unknown in British Columbia, northern Washington and southeast Alaska. Tetrasporangia form only on main axes and short lateral branches of upright axes (Fig. 113), occuring in sori but are not in stichidia; the branch apex grows vegetativery beyond the sorus. Short, fertile, first order branches narrow rapidty at the base, and taper abruptly to an acute apex, similar in shape of similarly positioned vegetative branchlets. One of the most striking features of Pterocladia caloglossoides is the characteristic "V" arrangement of tetrasporangia in the sorus (Fig. 122). The more proximal Vs are oldest. Sporangia occur in pairs of lines radiating laterally at 45° to the axial filament. The margin of the flattened branch is sterile, with tetrasporangia borne on both faces of branches which are recognizable approximately five axial cells behind the apex. They are cut off from medullary cells adjacent to the second order plate (Fig. 123) and have a lateral pit connection (Fig. 124) to their bearing cell. Tetrasporangia are Taxonomy and Nomenclature / 57 cruciately divided (Fig. 125), sometimes irregularly, and 20-45 um long and 15-40 Aim wide. Only rarely do young sporangia form between cavities of released sporangia, indicating that percurrent production of tetrasporangia is uncommon in P. d. Discussion Pterocladia caloglossoides originally was described by Howe (1914) as Gelidium caloglossoides from the Island of San Lorenzo, Peru. It was found on shells dredged from 2.5 fathoms (5 m depth) by Dr. Robert E. Coker. The holotype (Coker 59 in NY) was lost while on loan (Thiers, pers. comm. 1987) and it is unknown if isotypes exist. I have written to AHFH, BM, FH, MICH, NY and US, but their curators have not located isotypes of G. caloglossoides. Coker 59 was the only material mentioned by Howe in the diagnosis. Questions may be raised of the connection between the taxon Howe described as G. caloglossoides from Peru and Dawson's identification of material from Pacific Mexico as P. caloglossoides that was used in comparison with British Columbian P. caloglossoides. In Howe's original diagnosis, G. caloglossoides was characterized by it's creeping habit, small size, radiate clusters of 2-5 upright branches opposite attachment pegs, flattened axes, surface cells in distinct oblique rows at branch apices, narrow medulla, and tetrasporangia in "distinct oblique lines". Gametophytic plants either were not found or described. Howe noted the similarity of G. caloglossoides to G. pusillum but believed that G. caloglossoides was sufficiently distinct on a number of criteria to warrant recognition as a separate species. Taxonomy and Nomenclature / 58 Hollenberg (1942) reported Gelidium caloglossoides from southern California and Monterey, central California., which was included by Smith (1944) in the marine flora of the Monterey Peninsula. Diagnostic features used to recognize this taxon were its prostrate habit, small size, flattening of the branches, branch pattern and arrangement of tetrasporangia in Vs. Smith (1944) also noted that in the Monterey specimens the upright axes have few orders of branching and that branches are almost perpendicular to main axes. Dawson (1953) transferred G. caloglossoides to Pterocladia based on unilocular cystocarps found in material from Pacific Baja California del Norte, observing that the tetrasporangial plants from Baja California were the same as the material described from Peru by Howe (1914). I have compared British Columbian Pterocladia caloglossoides with Dawson's material from Pacific Mexico (AHFH Dawson 8593, liquid-preserved and specimens on microscope slides 1176, 1177 from Guadalupe Island, off Baja California, and AHFH Dawson 8733 liquid-preserved 'and microscope slides 1325, 1326, 1327 from Barra de Navidad, Jalisco on the mainland of Mexico). British Columbian P. caloglossoides resembles to both these collections, although it is most similar to the Guadalupe Is. plants. Specimens from Guadalupe Is. were a maximum of 1 cm tall, with unbranched uprights, the tallest of which were undamaged by grazing or abrasion. Liquid-preserved Guadalupe Is. specimens, those on slide 1176 and three of four plants on slide 1177 are tetrasporophytes, with tetrasporangia clearly in Vs (Fig. 126). On the creeping axes are numerous attachment pads opposite from which two to four upright branches arise (Fig. 114). These Pacific Mexico plants fit easily within the description of P. Taxonomy and Nomenclature / 59 caloglossoides given and illustrated by Howe (1914). Baja California plants are less well developed than those from Peru, but British Columbian plants from Orlebar Point, Gabriola Island, where frequent collections were made throughout the year, showed much variation in the development of upright axes from their almost complete absence to unbranched axes, and to well branched axes. British Columbian P. caloglossoides also corresponds well with all other illustrations and descriptions given for central California, and Pacific Mexico and Gulf of California P. caloglossoides (Smith, 1944; Stewart, 1976; Stewart and Norris, 1981), and thus is determined to occur in the study area. One plant of Dawson's slide 1177 from Guadalupe Is. (Fig. 127) is cystocarpic. Its morphology matches that of tetrasporophytes, and it is reasonable to assume that the specimen is representative of Pterocladia caloglossoides. From what can be seen from a whole mount of a cj^ stocarpic plant (giving a "top" view of a cystocarp), an ostiole appeared on only one face of the cystocarp (Fig. 129), indicating a single locule. Sections of a cystocarp from liquid-preserved material (AHFH Dawson 3733) show a single locule (Fig. 128). Thus the species as known from Pacific Mexico does indeed belong in Pterocladia, and Dawson (1953) was justified in transferring Gelidium caloglossoides to Pterocladia, though he admits that absolute confirmation of placement in Pterocladia could only be obtained by observing cystocarps in type localitj' Peruvian material. It is interesting that Dawson later used the combination G. caloglossoides when describing specimens from the Peruvian flora (Dawson et al., 1964). Added evidence for the generic placement of this taxon in Pterocladia comes from Taxonomy and Nomenclature / 60 apical and lateral initial morphology observed in British Columbian material. Both types of initials are domed hemispherical and protruding, and neither occur in cortical depressions. This finding is in agreement with Rodriguez and Santelices' (unpubl. mscr.) proposal that apical and lateral initials are similar in shape in Pterocladia and dissimilar in Gelidium, but it contradicts their statement that either the apical or lateral initials are depressed in cortical indentations in Pterocladia. The "indentation of initials" character while apparently reliable for other taxa of Pterocladia was not corroborated in British Columbia P. caloglossoides. Pterocladia caloglossoides as identified by E. Y. Dawson from Barra de Navidad, Jalisco, Mexico (AHFH Dawson 8733) are heavily damaged from abrasion or grazing, with truncated uprights that only reach a maximum of 0.5 mm tall. Some of the truncated axes have regenerated apical cells, but little regrowth occurred before the plants were collected and preserved. Two plants (slide 1327, AHFH) have many attachment pads and several uprights opposite the pads. Sections of vegetative plants (Fig. 132; slide 1326, AHFH) show a cortex and medulla that is well developed compared to British Columbian P. caloglossoides. Dawson's material does correspond to the illustrations and description given by Howe for Gelidium caloglossoides, particularly in the distinctive features of branching of the prostrate axis, attachment pad position and arrangement of tetrasporangia in Vs. Thus I agree with Dawson (1953) that the Mexican material with a single cystocarp locule is the same as G. caloglossoides (Howe, 1914) and it belongs in Pterocladia. Taxonomy and Nomenclature / 61 Dawson (1953) noted that "antheridia" were unknown in Pterocladia caloglossoides, however, specimens of his (slide 1325, A H F H ) are a whole mount of one female and two male thalli. The fertile apices of these gametophtyes are blunt rounded (Fig. 129), whereas vegetative apices are more acute and tapering. Spermatangia are in clusters of two to four in sori behind apices (Fig. 130). Each cluster replaces a vegetative cortical cell (Fig. 131), and spermatangia are cut off transversely and singly from spermatangial mother cells (Fig. 133). When Dawson made the new combination Pterocladia caloglossoides, he also described a new species, P. parva Dawson (1953: 77) 1 based on only female gametophyte material. Dawson's illustration (1953, pi. 6 fig. 2) of P. parva is strongly resembles P. caloglossoides, an observation supported by Dawson's comment that the illustration of P. caloglossoides in Smith (1944, pi. 44 figs. 3, 4) is more similar to P. parva than to P. caloglossoides. Pterocladia parva later was condsidered a synonym of P. caloglossoides by Stewart and Norris (1981). The similarity of Gelidium pusillum to Pterocladia caloglossoides was noted by Stewart (1976) and Stewart and Norris (1981), because both are small and turf-like. In California, however, G. pusillum has bilocular cystocarps, characteristic of the genus Gelidium. The name G. pusillum was used with some reservation (Stewart, 1976) for material from California. Other species names, G. latifolium, G. sinicola, P. parva and P. caloglossoides have been applied to this California material, although more definite determinations necessarily await a thorough comparison with type material of G. pusillum. holotype: Dawson 425 on sheet 4181, plus slides 1178-1181 in A H F H Type locality: San Felipe, Baja Californiadel Norte, Gulf of California Taxonomy and Nomenclature / 62 e. Seasonality and Distribution Pterocladia caloglossoides is perennial, persisting for most of the year as prostrate axes with few uprights that become more abundant and form a turf in the summer months, reaching their maximum development in August and September. Tetrasporangial sori are most abundant at this time. A collection was made in February 1987 at Georgina Point, Mayne Island (Fig. 3), however, in which tetrasporangial sori were present but sparse. Pterocladia caloglossoides is relatively common and widespread in the study area, but it likely has been overlooked and not reported from many sites due to its small size and creeping habit. This species has been collected from Orcas and San Juan Islands in northern Puget Sound (Fig. 3), Mayne and Gabriola Islands in the Gulf of Georgia (Fig. 3) and Hornby and Denman Islands in the Strait of Georgia (Fig. 4), Sooke (Juan de Fuca Strait), Barkley Sound (Fig. 2) and Nootka Island, on the west coast of Vancouver Island, and the Queen Charlotte Islands (Fig. 1). This study extends the northern limit of distribution of Pterocladia caloglossoides from northern Washington (Norris and Hollenberg, 1969) to British Columbia (e.g. UBC A12295, UBC A19667, UBC A28937, UBC A37605, UBC A64648, for others see Appendix 1) and southeast Alaska (UBC A69423, UBC A69424, UBC A69425, UBC A69426, UBC A69427, UBC A69428, UBC A69529, UBC A69430, UBC A69431). Pterocladia caloglossoides also is reported from Hawaii (Santelices, 1977), the Great Barrier Reef, Australia (Cribb, 1983), the Indian Taxonomy and Nomenclature / 63 Ocean south of Sumatra (Weber van Bosse, 1921), and Natal, South Africa (Norris, 1987). D. PREVIOUSLY REPORTED TAXA EXCLUDED FROM THE REVISED FLORA Gelidium crinale, G. pusillum, G. robustum and G. sinicola have been reported in the local flora (Widdowson, 1974; Scagel et al., 1986), but these are based on taxonomic misidentifications as none of these taxa actually occur in the study area. With the exception of G. robustum, these taxa are generally poorly understood and their species limits poorly defined. In the first two decades of this century, Gelidium amansii (type locality "in mari Indico ad ins. Franciae, Madagascar, Indiae orientalis", Kiitzing, 1849) was reported from southern British Columbia and northern Washington (Setchell and Gardner, 1903; Collins, 1913; Kylin, 1925; Connell, 1928), and reported plants were likened to P. B.-A., No. 585 (distributed as G. amansii, Collins et al., 1903, but in fact are now recognized to be G. robustum collected in California) by Setchell and Gardner (1903). Collins (1913) noted that British Columbian plants based on collections by Macoun from Ucluelet, Departure Bay and Victoria and by Tilden from Port Renfrew (Minnesota Seaside Station) he referred to G. amansii were much smaller than Californian specimens, Without exact collection data confirmation of Collins' detemination cannot be made but UBC and CANA have Macoun and Tilden collections from these locations (Macoun: UBC A1952, CANA 3843, CANA 4346, CANA 4349 (Fig. 135); Tilden: UBC A5195). All of Taxonomy and Nomenclature / 64 these specimens were determined to be G. purpurascens. Kylin (1925) included G. amansii in the Friday Harbour flora but had not seen any specimens. These early records of G. amansii have been treated by later authors as misidentifications (Scagel, 1957). There is a single report of Gelidium pulchrum from the area of Nanaimo (Stephenson and Stephenson, 1961b). Vouchers of this record do not exist and the Nanaimo area has been recollected without finding any specimens that are this taxon. This record has been included as misidentified G. purpurascens by Scagel et al. (1986) in the most recent floristic treatment of British Columbia, northern Washington and southeast Alaska. In the following discussion, I treat those taxa more recently reported for the local flora but that are, in fact, absent, based mostly on taxonomic misidentifications. 1. Gelidium robustum (Gardner) Hollenberg et Abbott 1965: 1179 Basionym: Gelidium cartilagineum var. robustum Gardner 1927a: 280, pl. 54 Holotype: UC 294572! Type locality: near Ensenada, Baja California del Norte, Pacific Mexico Gelidium robustum is a well defined species ranging from central Baja California, Mexico (Stewart, 1976) to central California (Silva, pers. comm.). First was Taxonomy and Nomenclature / 65 described by Gardner (1927a) as a variety of G. cartilagineum (L.) Gaillon, it was later elevated to species rank by Hollenberg and Abbott (1965). Based on an examination of the holotype from near Ensenada, Baja California del Norte (UC 294572, Fig. 138) and additional representative material (UC 395419 from San Pedro, California collected by Gardner; UC 647822, from White Pt., California collected by Dawson; and UC 756464, UC 756469, UC 756470, UC 756503, and UC 940173 from various localities in Baja California, Mexico collected by Dawson; and UC 1451987 from Portuguese Bend, California collected by Loomis), it was concluded that none of the British Columbian plants (with the possibility of one exception) correspond to this taxon. There is an interesting specimen in UBC (UBC A7861) of a plant that is tall, robust, tetrasporangial and clearly is Gelidium robustum (Fig. 136). It poses somewhat of a mystery. The annotation label states the collection locality as "Shoal B. Victoria" from 1917 by an unknown collector. There is no "Shoal B." near Victoria, British Columbia (Canadian Coast Pilot) although it is unclear whether the reference is to Shoal "B", or whether B is an abbreviation for bay, bight, beach etc.. However, there are no such locales near Victoria, B.C., but there are some in the San Juan Islands a few km to the east of Victoria (Canadian Coast Pilot). A considerable amount of seaweed collecting was done in southern Vancouver Island as early as 1917, and it is unlikely that if G. robustum existed at Victoria in 1917, such a large, conspicuous plant would have been collected only once and never at an3' time between central California and British Columbia. Admittedly, a small population could have been in Victoria (perhaps introduced from California in the early 1900's), that has subsequently Taxonomy and Nomenclature / 66 disappeared. It is also possible that collection data were confused between samples from a variety of locations along the coast. As it stands, even if G. robustum had been present in British Columbia, it does not exist here now, and G. robustum must be excluded from the flora. In the study area, several specimens of Gelidium purpurascens were misidentified as G. robustum. Differences between these taxa already have been discussed (Table 4). Several of the literature records alluding to the presence of G. robustum in the study area lacked voucher specimens or illustrations, and those records could not be investigated. None of the reports for which there are vouchers identified as "G. robustum" were in fact G. robustum, but are misidentifications. Scagel (1957) cited five specimens from Departure Bay and Victoria on Vancouver Island (as G. cartilagineum var. robustum) (CANA 3473 = CAN 209, CANA 3740 = CAN 74, CANA 3843 = CAN 353, CANA 4346 = CAN 208, CANA 4349 = CAN 310), and all were re-examined and found to be G. purpurascens, and Scagel's (1967) illustration is of G. purpurascens. South's (1968) records of "G. robustum" from Gabriola Island were of a variety of taxa: specimens from Orlebar Point (UBC A28937, UBC A29575) are Pterocladia caloglossoides and G. coulteri respectively, specimens from Lock Bay (UBC A36.199) and False Narrows (UBC A36198) are G. coulteri and specimens from Twin Beaches (UBC A28645, UBC A28646) are G. coulteri and G. purpurascens respectively. All records of "G. robustum" from Barkley Sound (Scagel, 1973) that could be located were found to be G. purpurascens. 2. Gelidium crinale (Turner) Lamouroux 1825: 191 Taxonomy and Nomenclature / 67 Basionym: Fucus crinalis Turner 1819: 4 Synonyms: Acrocarpus crinalis (Turner) Kutzing 1868: 11, pl. 33 figs, a-c, n.v. Acrocarpus spinescens Kutzing 1868: 12, pl. 33 figs, d-e, n.v. Acrocarpus corymbosus Kutzing 1868: 13, pl. 36 figs, a-c, n.v. Lectotype: Specimen annotated by Setchell as "This is Turner's idea of G. crinale and may be taken as type. f. W.A.S." (from Turner Herbarium in BM) (Dixon and Irvine, 1977b) Isotype: AHFH 55234, n.v. Type locality: Kilmouth, England Descriptions of Gelidium crinale (Gardner, 1927a; Dawson, 1944, 1953; Taylor, 1957; Santelices, 1977) demonstrate the lack of well defined diagnostic characters for this species. Santelices (unpubl. mscr.) especially has noted the confusion about the species limits and identifications of this taxon, and the resulting misapplications of the name. Collins (1913) first recorded G. crinale in the study area from Victoria, B.C. and cited P. B.-A. No. 195 (Collins, et al., 1896, Fasc. IV) from Port Jefferson, Long Island, New York, G. crinale as a reference specimen, but this specimen is different from any species present in British Columbia or northern Washington. Collins' record for G. crinale is based on a Macoun collection from Victoria (Collins, 1913), which is probably CANA 3474 (formerly CAN 207) (Figs. 134, 137) from Beacon Hill, Victoria, collected June 1908 and labelled as G. crinale. Re-examination of the plants on this sheet Taxonomy and Nomenclature / 68 revealed them to be G. coulteri. The inclusion of Gelidium crinale in the British Columbia flora has continued with the repeated citation of CANA 3474 (as CAN 207); for the record from Departure Bay, B.C. (as V 1456) (Scagel, 1957), the specimen could not be located. A specimen (UBC A66778) collected by Scagel from Departure Bay as "G. crinale" probablj' corresponds to the missing V 1456 and is stamped "on loan from the B.C. Provincial Museum". This specimen is now recognized to be G. purpurascens. In UBC there are numerous specimens misidentified as "G. crinale" that were determined to be G. coulteri (UBC A31363, UBC A31426, UBC A37402, UBC A53975, UBC A60254, UBC A60268, UBC A60460, UBC A64934), and one that is Pterocladia caloglossoides (UBC A19667). Many other specimens originally labelled as "G. crinale" have been annotated as other species, e.g. G. coulteri, G. purpurascens, P. caloglossoides, reflecting the taxonomic confusion of G. crinale, and the assumption of its presence in the local flora, time of the collection. Gelidium crinale first was recorded from California by Collins (1903 in Collins et al., P. B.-A., Fasc. XXIII, No. 1138), and it was the basis for a new variety, G. crinale var. luxurians Collins (1906: 111). Later authors continued to record the presence of G. crinale var. luxurians from California (Gardner, 1927a) and Baja California del Norte, Mexico (Dawson, 1953). Dawson (1944, 1953, 1966) listed both G. crinale (var. crinale) and G. crinale var. luxurians from Pacific Mexico and the Gulf of California. Taxonomy and Nomenclature / 69 Stewart (1974) considered the Californian Gelidium crinale var. luxurians (i.e. P. B.-A., Fasc. XXIII, No. 1138; and the material mentioned in Gardner, 1927a) to be Pterocladia media Dawson (1958: 68). The remaining specimens of "G. crinale var. luxurians" Dawson and G. crinale var. crinale sensu Dawson (1944, 1953) were considered G. pusillum by Stewart and Norris (1981). Thus G. crinale var. crinale and G. crinale var. luxurians do not occur in the Gulf of California or in the northeast Pacific. 3. Gelidium pusillum (Stackhouse) Le Jolis 1863: 139 Basionym: Fucus pusillus Stackhouse 1801: 6 Synonyms: Fucus caespitosus Stackhouse 1801, pi. 12 n.v. ACrocarpus pusillus (Stackhouse) Kutzing 1849: 762 n.v. Lectotype: in BM (Dixon and Irvine, 1977b) Type locality: Sidmouth, England There is no evidence to support the presence of Gelidium pusillum in the study area. Of the reports of "G. pusillum" from British Columbia and northern Washington (given in Scagel et al., 1986), only Norris and Wynne (1968) cited voucher specimens and Garbary et al. (1984) did not specify which UBC herbarium specimens they used. Norris and Wynne (1968) first reported G. pusillum in the study area from San Juan Island, Washington, but the specimen they reported (WTU, Norris 5723) could not be located. Collections other than those of Norris and Wynne (1968) of gelidioid plants from San Juan Is. belong Taxonomy and Nomenclature / 70 to G. purpurascens (UBC A4402, UBC A4403; FHL 2849; private herbarium of W.R. Waaland #1583) and to Pterocladia caloglossoides (WTU 248018). The Scagel et al. (1986) report of "G. pusillum" from Barkley Sound, based on a specimen (UBC A61245), is now identified as P. caloglossoides. Other herbarium specimens from the study area labelled as "G. pusillum" are mis-identifications, and actually G. coulteri (UBC A24837, UBC A28351, FHL 3055) and P. caloglossoides (UBC A33645, UBC A64648). Gelidium pusillum has a worldwide distribution and, like G. crinale, is poorly circumscribed (Santelices, unpubl. mscr.). Different workers have used varying species concepts that have incorporated numerous other species (Dixon and Irvine, 1977a; Stewart, 1976; Stewart and Norris, 1981), or they have recognized many varieties (Dawson, 1944, 1953; Santelices, 1977; Schnetter and Bula Meyer, 1982). The name seems to have been applied commonly to any small, turf-forming, compressed to flattened gelidiaceous plants. This taxon clearly needs revision on a global scale, beginning with a thorough study of the type material. 4. Gelidium sinicola Gardner 1927a: 278 Holotype: (Gardner 2615) as UC 276620! Type locality: Point Cavallo, San Francisco Bay, California Records of "Gelidium sinicola" in the study area result from confusion as to the definition of the taxon. Gelidium sinicola Gardner (1927a) was described as Taxonomy and Nomenclature / 71 "sparse" (i.e. not abundant) and "of limited distribution", and based only on the type locality collection from Point Cavallo, San Francisco Bay, California (Gardner 2615 as UC 276620) (Fig. 139). Even Gardner was apparently confused as to the identity of this species. His later collections from the same tidepool as the type collection (Gardner 7179, UC 494898) (Fig. 21), were noted in his field book "G. sinicola ? at least from the same pool as type came" (Silva, pers. comm. 1985), and on the herbarium sheet he identified the plant only as "Gelidium". Added later, but not in Gardner's handwriting (possibly by E. Y. Dawson, fide Silva, pers. comm. 1985), is "coulteri Harv. f.". Both the type specimen and Gardner 7179 have a morphology that could fit within the limits of Gelidium coulteri. Gelidium sinicola differs from G. coulteri in that it is cylindrical rather than flattened (Fig. 140), and it is narrower (Gardner, 1927a). Stewart (1976) noted that G. coulteri varies widely in compression of branches, and I also have seen much variation of axis width and compression in G. coulteri from the study area. Silva (pers. comm., 1985) has visited the type locality of G. sinicola several times and has made extensive collections around San Francisco Bay, and writes of seeing plants he considers to be morphological^ ' variable G. coulteri, but never any cylindrical plants that could correspond to G. sinicola. Because the type material is infertile, there are few characters for species comparison. Based on a comparison of height, apical and basal axis diameter, branching pattern and rhizine abundance and position, in Pterocladia media, G. sinicola (holotype and Gardner's diagnosis), and Californian G. crinale var. luxurians, Stewart (1974) suggested that G. sinicola might be a synonym of P. media. Generic placement cannot be confirmed for G. sinicola Taxonomy and Nomenclature / 72 because no cystocarpic plants are known to exist. Despite the problems of the actual identity of Gelidium sinicola, records of it in the study area are based on misidentifications of other gelidioids. Specimens in UBC labelled originally as "G. sinicola" belong to G. coulteri (UBC A1449, UBC A14403) or Pterocladia caloglossoides (UBC A29576, UBC A39127). Of the reports of "G. sinicola" in the study area (given in Scagel et al., 1986), only Norris and West (1966) cite vouchers. The voucher cited by Norris and West (Norris 5087) could not be located. Thus G. sinicola does not appear to occur in the study area. CHAPTER IV. LIFE HISTORY OF GELIDIUM IN CULTURE A. INTRODUCTION Until recently, the life history had not been completed successfully in culture for anj' species of Gelidiales. This is somewhat surprising considering the economic potential and interest in the group. Most commercial uses of Gelidiaceae have been restricted to wild harvest. It was assumed from field collections that Gelidiaceae have a Polysiphonia-type life history (Fig. 141) (Dixon, 1961, 196.3) with a sequence of (haploid) gametophyte, (diploid) carposporophyte and (diploid) tetrasporophyte phases, where gametophyte and tetrasporophyte phases are free-living and isomorphic, and the carposporophyte develops on the female gametophyte. In order to confirm this, I attempted to culture the local Gelidium species. According to recent studies on the life histories of G. coulteri (Macler and West, 1987) and G. vagum (van der Meer, pers. comm.), both species require a short time to complete the life history (two months for G. coulteri, Macler and West, 1987). In the current study it was noted that both gametophytic and tetrasporophytic phases of Gelidium vagum are abundant enough within the species' restricted distribution (see Chapter 3) to suggest a Polysiphonia-type life history. Gametophytes of the other local species, however, are rare or absent in the study area, suggesting that a Polysiphonia-type life history may not commonty occur in the studj' area, even though the species may demonstrate it under different environmental conditions or geographical areas. Instead, populations may 73 Life History of Gelidium in Culture / 74 be propagating or persisting vegetatively, undergoing an asexual life history or repeating tetrasporophytes. In the study area, gametophytes of G. purpurascens rarely are found and have been collected from only a few restricted sites. Male gametophytes are difficult to differentiate in the field, although the stichidial nature of spermatangial branchlets imposes an additional order of branching making male plants detectable with some practice. It would be expected that if cystocarpic plants are collected, male gametophytes also might be present. Despite the paucity of G. purpurascens gametophytes, the species is abundant and widespread in the study area, and in late summer and early fall tetrasporangial plants are common. This raises the question of the kind of life history that predominates in most G. purpurascens populations in British Columbia and northern Washington. Are tetraspores viable, and do they produce gametophytes and ultimate^ cycle sexually, or do populations cycle asexually or persist vegetatively? Gametophytes of G. coulteri never have been collected in the study area, although they are known from farther south (e.g. California, Dawson, 1953). Tetrasporophytes are not often fertile, observations suggest a relatively restricted time period (August to November), and fertile plants are probably overlooked. Detection is more difficult when tetrasporophytes are only weakly fertile because tetrasporophytic branchlets of G. coulteri are not as distinctive as in G. purpurascens and G. vagum. Detection of tetrasporangia on Pterocladia caloglossoides plants is difficult because the small size of these plants (never taller than 2 cm) makes low magnification necessary for detection of fertility. When P. caloglossoides tetrasporophytes are fertile, tetrasporangia are abundant, Life History of Gelidium in Culture / 75 and most upright axes bear fertile sori. B. METHODS 1. Culturing Cystocarpic and tetrasporic Gelidium vagum was collected from Denman Island and Galleon Point, Hornby Island (Fig. 4) in late September, 1986. Tetrasporangial G. purpurascens was collected from Geer Islets and Dixon Island, both on the west coast of Vancouver Island (Fig. 2) in August 1984, Georgina Point, Mayne Island (Fig. 3) in November, 1985 and February, 1986, Orlebar Point, Gabriola Island (Fig. 3) in October and December, 1985, August and October, 1986, and from Whalebone Bay, Gabriola Island (Fig. 3) in October and November, 1986. Gelidium coulteri tetrasporophytes were collected from Denman Island in September, 1986 and from Orlebar Point, Gabriola Island in October, 1986. Tetrasporangial Pterocladia caloglossoides was collected from Georgina Point, Mayne Island, Orlebar Point, Gabriola Island and Galleon Point, Hornby Island in September, 1986. Field collected plants were brushed to remove detritus and as many epiphytes as possible before culturing. Fertile tetrasporangial or cystocarpic stichidia were excised and placed into 2X6 cm plastic petri plates containing culture medium. In the case of infertile plants, apices were excised and placed into petri dishes. For each isolate and set of growth conditions at least two dishes were prepared. The culture medium used was half strength, modified Provasoli's enriched Life History of Gelidium in Culture / 76 seawater medium (PES) (McLachlan, 1973). Initially, full strength PES was used, but it later was found that plants grew as well and with fewer contamination problems in 1/2 strength PES. This was a particular benefit in the first stages of isolation when unialgal cultures were being started. McLachlan's enriched seawater medium (SWM, McLachlan, 1973), without soil and liver extracts, also was tried. Growth was similar to that of 1/2 PES, but algal contaminants also seemed to thrive. Enrichment nutrients were filter sterilized with a 0.45 um millipore filter prior to mixing. Two walk-in environmental growth chambers, one at 20°C, 16:8 (light:dark photoperiod) and 20.15 uEm~ 2s~ 1 and one at 10°C, 8:16 photoperiod and 16.42 jiEm" 2s~ 1 were used, along with a smaller Percival 1-35-L incubator set at 24°C, 12:12 photoperiod and 7.91 uEm"2s"1. Each isolate was cultured under all three sets of conditions. Growth conditions were maintained as constant as possible but were subject to chamber breakdowns and disturbances (e.g. opening doors during dark periods). Several growth conditions were utilized to improve the chance of completion of a gelidioid life history. Conditions were not intended to test responses of plant growth and reproduction to differing environmental factors, as temperature, daylength and irradiance varied simultaneously between growth conditions. As the chambers were a group facility, they were set at temperatures and daylengths that were most amenable to all users, conditions that would not necessarily have been chosen for optimal Gelidium growth or life history studies. Aeration did not promote tetraspore or gamete production and made it harder to control contamination. Life History of Gelidium in Culture / 77 After incubation of spore producing tissue for 2-5 days (depending on chamber conditions), released spores and germinated sporelings were removed with a finely drawn out pipette operated by light suction and placed into petri plates of fresh medium. The medium was changed every one to two weeks depending on contamination and, after the first one to two months, once every two months. For isolates used in life history studies the medium was changed monthly. Tetraspores or carpospores released from isolates were removed to new medium, and the above process was repeated. When monoecious isolates produced gametes, the dishes were stirred and swirled each day to encourage fertilization. In dioecious cultures, fertile male and female gametophytes were placed together into a dish and stirred and swirled daify. Tetraspores of Gelidium purpurascens and G. vagum, and carpospores of G. vagum were released readily from plants under the culture conditions. Germination was most successful from the tetraspore and carpospore-bearing plants that were collected in September. At other times, spores released readily but most failed to germinate. Gelidium coulteri tetraspores released and germinated less readily so that few sporelings were obtained. Several techniques with fertile tetrasporophytes of P. caloglossoides such as drying the plants prior to immersion for release, cutting the spore-bearing branches and putting cheese-cloth covers over dishes to reduce light intensity, were unsuccessful at releasing tetraspores. For all species apical fragments also were cultured. Life History of Gelidium in Culture / 78 2. Chromosome Counts Chromosome counts were attempted on nuclei of undivided tetrasporangia of Gelidium vagum collected from Hornby Island. Squashes were made by first softening excised tetrasporangial stichidia in 4% KOH for 10 minutes. Tissue was stained with Wittmann's hematoxylin (Wittmann, 1965) for one hour, the cover slip added and tissue partly squashed. New stain was added at intervals to prevent drying out. Material was destained by drawing 45% acetic acid under the coverslip for 30-60 seconds, then rinsed with distilled water and permanently mounted in 40% Karo, and tissue was firmly squashed to spread cells thoroughly. C. SPORE GERMINATION AND EARLY DEVELOPMENT The pattern of spore germination, and early development of the sporeling to the stage of apical cell organization, are similar in Gelidium coulteri, G. purpurascens and G. vagum for both tetraspores and carpospores. The following account is of spore germination events in G. vagum. The released spore is spherical with a large, stellate chloroplast and prominent nucleus (Fig. 142). It germinates by producing a small hyaline protuberance (Fig. 143), which expands (Fig. 144) into a germ tube, slightly wider and longer than the spore. The spore contents evacuate into this tube, and a wall forms cutting off the cytoplasm from the spore (Fig. 145). In most cases the evacuated spore appears completely empty, but occasionally granular debris (Figs. 146, 147) or Life History of Gelidium in Culture / 79 parts of the chloroplast remain. This pattern of spore germination, referred to as "Gelidium-type" germination (Chemin, 1937), is characteristic of the order Gelidiales (Papenfuss, 1966; Santelices, 1974). In some spores a germ tube did not form, and the cytoplasm divided several times within the spore wall. Later germination stages were not seen, and it is unlikely that the sporeling survived. The sporeling divides longitudinally and unequally with a curved wall (Fig. 147), to form "concave" and "fusiform" daughter cells (after Boillot, 1963). The larger, concave daughter cell, divides several times transversely and longitudinally (to the axis established by the germ tube and first division), forming 10-14 isodiametric cells (Figs. 148, 149, 150). The most distal cell of this concave cell-derived group, opposite the spore, elongates, forming a primary attachment rhizoid (Figs. 151, 152). The rhizoid is unpigmented, thick-walled and most of the cytoplasm is contained in the tip. Adjacent distal cells may subsequently similarly elongate, forming additional rhizoids (Fig. 153) that attach the sporeling to the substratum. Similar to other workers (e.g. Macler and West, 1987) I have found that if attached sporelings are removed from the substratum they do not reattach later. Up to four primary rhizoids may form on a single sporeling. The smaller, fusiform daughter cell of the young sporeling divides several times transversely and longitudinally to form small, isodiametric cells (Figs. 154, 155). Two longitudinally oriented clusters of cells comprise the sporeling, each cluster originating from the concave or fusiform daughter cell and separated by the first division plane of the germ tube (Figs. 155, 156). When the entire sporeling is approximately 20-40 cells large, the most proximal cell of the fusiform Life History of Gelidium in Culture / 80 cell-derived group cuts off a shallow, hemispherical apical cell (Figs. 157, 158), adjacent to the original spore wall. Up to this point in development, the sporeling increases very little in size beyond that of the germ tube. The uniaxial construction typical for the genus develops following sporeling apical cell formation. The sporeling increases in size, growing in the opposite direction to the primary rhizoid and past the original spore wall (Fig. 160). Organization is uniaxial and shows a characteristic pattern. The domed, hemispherical, apical cell divides transversely (Figs. 159, 162). The subapical cell divides obliquely twice to form two lateral, opposite periaxial cells. Each periaxial cell cuts off a lateral second order filament that grows outwards contributing to thallus width (Figs. 68, 159). Periaxial cells also cut off third order filaments above and below the lateral plane that determine the thickness of the thallus. As the sporeling increases in length, branch initials are formed from modified cortical cells (Fig. 159), and attachment pads develop from the outward elongation and adhesion of cortical cells (Fig. 161). D. REPRODUCTION IN CULTURED AND FIELD PLANTS 1. Culture Results Reproductive structures appeared on cultured plants of G. vagum, when they were one to two cm tall, approximately 5-6 weeks after spore germination in isolates grown at 20°C. Cultured tetrasporophytes, with the same morphology as tetrasporophytes from the field, produced tetrasporangia on stichidia (Figs. 163, Life History of Gelidium in Culture / 81 166). Plants grew faster at 20 °C and high light (long day and high irradiance) and were narrower and less flattened than field collected plants. At 24°C and low irradiance, plants were narrow and weak and only sparsely fertile. Tetrasporophytes grown at 10 °C grew more slowly and took much longer (several months to one year longer) to become fertile, if they did so at all. Plants grown at 10°C more often appear similar to field material, with wider, flatter branches than plants grown at 20 °C. Tetrasporangial stichidia were determinate, cylindrical, with rounded or acute apices and produced tetrasporangia around the entire stichidium in an unordered manner. Tetraspores were released and germinated into gametophytes (Fig. 167). Gelidium vagum gametophytes produced carpogonia and spermatangia on short branchlets (Fig. 164), as in field collected material, with an axis producing a succession of fertile branchlets so that progressively older carpogonia and post-fertilization stages were observed in more proximal parts of the plant (Fig. 165). Developmental sequences of male and female gametes and carposporophytes were similar to those observed in field collected gametophytes. Fertilization occurred in culture, although not as extensively as in field material, evidenced by the lower frequency of cystocarps per cultured plant. Greater water motion in the field than in the occasional^ ' swirled, unaerated cultures may bring non-motile spermatia in contact with trichogynes more effectively. After three to five weeks, carpospores were released from bilocular cystocarps. Carpospores were obpyriform when in the cystocarp, but rounded-up on release. They contained a large, central, stellate chloroplast and germinated into tetrasporophytes, as previously described. Life History of Gelidium in Culture / 82 Gelidium vagum isolates originally initiated from tetraspores and carpospores, produced fertile gametophytes and tetrasporophytes (respectively), that in turn released ' carpospores, and tetraspores, and germinated into tetrasporophytes and gametophytes. These tetrasporophytes (Fig. 168) and gametophytes became fertile completing the life history (simultaneously from two different starting points). Life histories could not be completed in culture for the other local gelidiaceous species, but some insights into the life histories of Gelidium purpurascens and G. coulteri in the study area were provided by culturing. Tetrasporangial stichidia of Pterocladia caloglossoides were placed into culture several times, but tetraspores were not released under anj' of the experimental culture conditions, and the vegetative fragments grew well at 10°C and 20°C but did not become fertile. Cultures of G. purpurascens were started from tetraspores released from field-collected tetrasporophytes (Fig. 169). The resulting gametophyte sporelings were dioecious; both carpogonial and spermatangial plants were readily identifiable in the cultures grown at 20 °C. Fertile male gametophytes developed in four isolates (Geer Islets, Dixon Island, Majme Island and Whalebone Bay on Gabriola Island, Fig. 170). Female gametophj'tes were identified only in Whalebone Bay cultures (Fig. 171). Carpogonia and spermatangia in cultured plants appeared mature but evidence of fertilization was not observed, possibly due to inadequate water motion in culture dishes, so that non-motile spermatia could not contact trichogynes. Alternatively, fertilization might have occurred but gamete incompatability or inviability, or early carposporophyte abortion could make fertilization undetectable. Life History of Gelidium in Culture / 83 Tetraspores did not release as readily from stichidia of Gelidium coulteri, and germination was less successful than for G. purpurascens and G. vagum, but a few isolates of gametophytes were maintained. Fertile gametophytes (found only in 20°C cultures) were male (Figs. 19, 172, 173); female gametophytes were undetected (under all growth conditions), and it is unknown if they existed in the cultures, or if all non-fertile plants from tetraspores were male gametophytes or simply non-reproductive. Undivided tetrasporangia were used for chromosome counts rather than other cell types (e.g. apical cells, spermatangia or spermatia) because they had the largest nuclei and most visible chromosomes. Even so, nuclei were still small (3-7ym diameter), and it was difficult to obtain the resolution necessary for reliable chromosome counts. Counts were made only on nuclei squashed flat enough so that chromosomes were in a single focal plane. Counts of 14 bivalents and a single, additional small body were made on the best view of a dividing nucleus (Figs. 174, 175). The bivalents consisted of closely paired chromosomes at the most contracted stage of prophase I. It was unclear whether the small body was a separate chromosome, small bivalent or an extraneous body. If it was a small chromosome, then it should be noted that there are size differences between some chromosomes. Counts of 14 and at least 12 and 13 (Fig. 176) bivalents were obtained from other dividing tetrasporangial nuclei. Nuclei in which counts of 12 and 13 bivalents were obtained contained a large deeply staining body probablj' comprising two or more bivalents. Thus the chromosome number obtained from Life History of Gelidium in Culture / 84 all of these counts is n= 14-15. In some nuclei, five large bodies were observed (Fig. 177). 2. Field Observations of Reproduction Tetrasporophytes and gametophytes of Gelidium vagum are equally abundant in the field when reproductive plants were found (August to October). Gametophytes of G. purpurascens were rarely collected in the study area. Of 18 sites in the study area from which I collected G. purpurascens, female gametophytes were found at only two sites (Kirby Point, Diana Island and Geer Islets, both in Barkley Sound). Six specimens of G. purpurascens female gametophytes were found from a total of several hundred records of this species (UBC A41842, Fleming Island, Barkley Sound; UBC A39126, Lawn Point, Vancouver Island; UBC A41586, Wizard Island, Barkley Sound; UBC A36876, Amos Island, west coast of Vancouver Island; UBC A41946, Clarke Island, Barklej' Sound; and UBC A46026, Bamfield, Barkley Sound; UBC A67446, Kirby Point, Diana Island). Male gametophytes of G. purpurascens rarely are collected from the study area, and the only specimens known are three records in UBC (UBC A53657, Nootka Sound, Vancouver Island; UBC A10813, Esteban Point, Vancouver Island; and UBC A41415, Tzartus Island, Barkley Sound). Tetrasporophytes of Gelidium purpurascens are much more abundant than gametophytes in my own collections and in herbarium specimens. Whereas G. purpurascens gametophytes were found only in the southern half of the study area, while tetrasporophytes were found fertile throughout British Columbia and Life History of Gelidium in Culture / 85 to its northern range limit in southeast Alaska (UBC A69432). Where female gametophytes occur, they bear many fully developed cystocarps, suggesting that successful fertilization occurs in the field. Tetrasporangia are abundant and appear deeply pigmented. Fertile gametophytes of Gelidium coulteri are unknown in the study area, but occur farther south in California. Although tetrasporangial specimens have been collected in British Columbia, they are not often abundantly fertile. Gametophytes also are unknown for Pterocladia caloglossoides from the study area, but are reported for this species from California (Dawson, 1953). Pterocladia caloglossoides is tetrasporophytic throughout British Columbia, but infertile in southeast Alaska collections (UBC A69423, UBC A69424, UBC A69425, UBC A69426, UBC A69427, UBC A69428, UBC A69429, UBC A69430, UBC A69431). E. DISCUSSION 1. Spore Germination Spore germination of Gelidium. vagum, G. purpurascens and G. coulteri is uniform for all tj'pes of spores and the species. In this study, it was impossible to differentiate species or spore types using any aspect of spore morphology or germination pattern. The similarity between tetraspore and carpospore germination also was noted by Yamasaki (1960). Furthermore, spore germination in Gelidium (Chemin, 1937; Katada, 1949, 1955; Yamasaki, 1960; Boillot, 1963; Kaneko, 1966; Guzman-del Proo et al., 1972; this study), Pterocladia, Acanthopeltis Life History of Gelidium in Culture / 86 (Katada, 1955) and Gelidiella (Chihara and Kamura, 1963; Sreenivasa Rao, 1971) showed the same general pattern. Differences in reports of spore germination are minor and, in at least some instances, appear to result from the interpretation of observations. For several hours after release, and before attachment, carpospores of some Gelidium species reportedly showed amoeboid motion (Boillot, 1963; Guzman-del Proo et al., 1972). Chemin (1937) in particular was fascinated by carpospore mobility, reporting paths and rates for spores of several red algal species. Germination observations of the local species of Gelidium were started after spore attachment, thus amoeboid motion, if present, was not observed in the spores. Several authors observed one to three (mitotic) nuclear divisions before spore evacuation in G. vagum (Kaneko, 1966), Gelidiella acerosa (Sreenivasa Rao, 1971), Gelidium latifolium, G. pulchellum and G. pusillum (Boillot, 1963), but onlj' a single nucleus entered the germ tube. Empty spore germination (i.e. evacuation of the entire spore cytoplasm into the germ tube) occurs in other higher rhodophyte groups, but only in Gelidiales is it followed by direct development of the upright plants, without an intervening discoid phase. Chemin (1937) designated all species with emptj' spore germination, as having "Gelidium-type" germination regardless of subsequent developments, and Boillot (1963) later demonstrated the unique nature of this pattern in Gelidium. The first division of the germ tube after being cut off from the spore, is highly significant, as the plane of this division and the distinctive concave and fusiform Life History of Gelidium in Culture / 87 daughter cells determine the polarity of the sporeling. Observations of the sporeling from different orientations can be misleading. For example, Sreenivasa Rao (1971) noted that the first division of the germ tube in Gelidiella acerosa is sometimes transverse. This interpretation probably resulted from a face view of the fusiform or concave cell (analogous to a valve view of a diatom), where the curved wall separating them is not seen, and the first divisions of fusiform and concave cells is transverse. There appears to be general agreement that the concave cell is the larger of the two daughter cells and gives rise to the primary rhizoid (Boillot, 1963; Chihara and Kamura, 1963; this work). However, there is confusion in the literature as to the developmental role played by the fusiform and concave cells. In Gelidiella acerosa, Sreenivasa Rao (1971) reported that the rhizoid originated from the fusiform cell group, whereas Chihara and Kamura (1963) observed the opposite, a concave cell group origin of the rhizoid. It is most likely that the concave cell produces the primary rhizoid in Gelidiella as in the other members of the order. In a series of manipulations of the environment of germinating spores, Katada (1949, 1955) determined that the specific gravity of the water affected the length and number of rhizoids produced, but the pattern of development was invariant. Interpretation of the origin of the apical cell also has caused confusion. Observations made here indicate that the fusiform cell group is the site of apical cell production. Specifically, the cell of the fusiform cell group closest to the spore wall, differentiates as the apical cell. This agrees with the pattern illustrated or stated by some workers in Gelidium amansii, G. divaricatum, G. japonicum, G. pacificum, G. pusillum, Pterocladia. tenuis and Acanthopeltis japonica Life History of Gelidium in Culture / 88 (Katada, 1955), G. amansii (Yamasaki, 1960), and Gelidiella acerosa (Chihara and Kamura, 1963) but is contrary to observations of Boillot (1963). Boillot (1963) reported the formation of two apical initials in Gelidium latifolium, G. pulchellum' and G. pusillum, one contributed by the fusiform cell group and another from the concave cell group. It was unclear which initial takes over, but Boillot believed that it is the apical cell from the concave cell group. Yamasaki (1960) illustrated a developmental sequence like that observed in the local Gelidium species, but he referred indistinctly to "upper" and "lower" daughter cells. He stated that the upper daughter cell gave rise to the creeping part of the plant, and the lower daughter cell produced the primary rhizoid and upright or creeping branches. This is unlike any other gelidioid developmental pattern; it maj' correspond to the dual apical cells mentioned by Boillot. According to Yamasaki's (1960) illustrations, however, spore development patterns appear similar to those of other gelidioids. Some authors do not indicate the origin of the apical cell, and their illustrations are sometimes too confusing to determine where this occurs (e.g. Kaneko, 1966; Sreenivasa Rao, 1971). In conclusion, possibly the most notable feature of spore germination and early sporeling development in Gelidiales is the uniformity of the developmental pattern in all species, genera and spore types studied to date, and the distinctiveness this pattern with respect to other red algal groups. 2. Life Histories Life History of Gelidium in Culture / 89 The completion of the Gelidium vagum life history in culture shows that the British Columbian plants have a Polysiphonia-type life history and, based on the common occurrence of tetrasporophytes and gametophytes in the field, this seems to be an accurate description of the life history in the study area. The scarcity with which Gelidium purpurascens gametophytes are collected in the study area suggests that gametophytes either fail to survive in the field (for example, if environmental conditions are unsuitable for spore or sporeling survival or full development) or gametophytes are commonfy produced but rarely become fertile (and hence are not recognized as gametophytes). Low abundance of fertile male and female gametophytes in the field would decrease the chance of fertilization, accounting for the rarity of cystocarpic G. purpurascens in the study area. In British Columbia, G. purpurascens is probably capable of completing the Polysiphonia-type life history, but factors such as low tetraspore viability and low abundance of gametophytes or unfavourable environmental conditions could prevent the common occurrence of this sexual life history in situ. Occasional periods of favourable environmental conditions could permit the completion of a sexual Polysiphonia-type life history and range expansion, whereas more commonly, populations persist vegetatively or reproduce asexually. The British Columbian populations of Gelidium purpurascens persist by perennating as upright fronds or creeping axes and may recruit new individuals when environmental conditions are suitable for gametophytic sporeling survival or Life History of Gelidium in Culture / 90 gametophyte fertility. Drift plants of G. purpurascens appear to be whole plants torn from the substratum rather than fragments of plants. Drift plants and attached plants in the field do not exhibit the abundant production of attachment pads that can be seen in cultured plant fragments. These observations suggest that propagation by vegetative fragmentation does not appear be a mechanism operating here, but further field experimentation would be neccessary for confirmation. Although there was no evidence in cultures, tetraspores also could be recycling apomictically as alternate dispersal agents. However, further work would be required to determine if such a process is operating in British Columbia G. purpurascens. Information on whether populations are expanding or simply maintaining their current extent would also be needed to assess the existence and importance of the above mechanisms. The occurrence of fertile male Gelidium coulteri plants in 20°C culture was significant because, although none were recorded from the study area (from herbarium records or my own collections), their presence indicates that the capacity for the production of male gametophytes exists. Male plants could be overlooked in the field or they could be absent if environmental conditions are not suitable for their development. A seawater temperature of 20°C is uncommon in British Columbia but is occasionally reached, particularly in sheltered locations. Gelidium coulteri probably propagates vegetative^ by creeping axes and possibly apomictically by tetraspores in British Columbia, at the northern limit of its distribution. In culture, Pterocladia caloglossoides failed to release (or germinate in situ) Life History of Gelidium in Culture / 91 tetraspores, although they were abundant on stichidia. It is unknown if tetraspores function at all as propagules (meiotically or apomictically). Clearly vegetative propagation and perennation by creeping axes must be important in the maintenance or expansion of this species' local populations. Gelidium purpurascens, G. coulteri and Pterocladia caloglossoides display the pattern of latitudinal variation in life history expression proposed by Dixon, (1965), where progressively more northern populations of a species show reduced fertility of gametophytes and, even more northerly populations show reduced tetrasporophyte fertility. This trend ultimately results in only vegetative propagation at species' northern range limits. Gelidium purpurascens variesin part from this pattern because some of the most northerly collections of this species, from southeast Alaska, are tetrasporangial (UBC A69432). 3. Chromosome Counts There have been few chromosome counts made on any Gelidiaceae. Kaneko (1966) observed n=7-10 in tetraspore germlings of G. vagum, Dixon (1954) obtained n = 4 and 5 for cells of the cortex and nutritive tissue of female gametophytes of G. corneum, and Kaneko (1968) counted n=15 for tetrasporangia of Acanthopeltis japonica. A demonstration of the uncertainty often associated with such counts is offered in the observations made on G. latifolium. Dixon (1954) obtained "4, 5, 9 and 10 chromosomes" for carposporophytes, but Boillot (1963) obtained n=18 for young gametophyte sporelings, and Magne (1964) counted n or 2n = 25-30 for vegetative cells of G. latifolium var. luxurians. It is possible that Life History of Gelidium in Culture / 92 some varieties could represent polyploid groups (e.g. is G. latifolium var. luxurians a polyploid of G. latifolium var. latifolium!) or that polyploids could be included with diploids in the same, or other, species. Counts of n= 14-15 obtained from undivided tetrasporangia of G. vagum in this study, are closest to counts made by Boillot (1963) for G. latifolium and by Kaneko (1968) for Acanthopeltis japonica. The G. vagum counts of n= 14-15 made here are not similar to and do not appear to be multiples of the only other counts made for G. vagum (n = 7-10) by Kaneko (1966). The importance of good preparations and accurate information about cell type and generation counted must be stressed. Too few reliable studies have been made and corroborated to confidently state chromosome numbers of species, genera or base numbers for the family. C H A P T E R V. CARPOGONIUM AND CARPOSPOROPHYTE DEVELOPMENT A. INTRODUCTION As with spermatiogenesis and tetrasporogenesis, there have been few detailed studies on pre- and post- fertilization development in Gelidium. The most recent investigations were those of Dixon (1959) on G. latifolium and G. pulchellum, Fan (1961) on G. robustum (as G. cartilagineum var. robustum) and Hommersand and Fredericq (1988) on G. pteridifolium. Additional selected species of genera of Gelidiales, Gelidium, Pterocladia, Suhria, Beckerella, and Acanthopeltis, have been observed for comparative purposes (Fan, 1961; Hommersand and Fredericq, 1988) or briefly touched on in other studies (Kraft, 1976). The interpretations made by Hommersand and Fredericq (1988) were significantly different from previous authors, most notably as to the nature of the carpogonium, and their findings were corroborated in the current investigation. In view of the paucity of studies on these developmental processes, additional descriptive observations based on local Gelidium species were warranted. The lack of studies of pre- and post- fertilization development in members of Gelidiales is surprising given the healthy debate that has taken place historically over the ordinal recognition of the group. Gelidiales now are recognized almost universally at the ordinal level by a suite of characters other than female reproductive and carposporophyte characters, such as a single pit plug cap, Gelidium-type spore germination, the presence of rhizines, the production of two distichous periaxial cells, and the transverse division of the spermatangial mother 93 Carpogonium and Carposporophyte Development / 94 cell in forming a spermatangium (see Chapter 1, herein; Santelices, 1974; Gabrielson and Garbary, 1986, 1987; Hommersand and Fredericq, 1988). Kylin (1923) elevated the Gelidiaceae, a family in the Nemaliales (as "Nemalionales"), to ordinal status based on the unchanged, fertilized carpogonium that was the starting point for the gonimoblast, and auxiliary cells that are present but function as nurse cells in nutrition of the carposporophyte. Kylin's (1923) system of ordinal classification, was based entirely on features of the female gametophytic and carposporophytic reproductive apparatus, namely characters of the carpogonial branch, auxiliary cells, connecting filaments and the gonimoblast. It must be noted that contrary to Dixon's (1961) belief that life history differences (Kylin, 1923 believed Nemaliales were haplobiontic and Gelidiales diplobiontic) were also a criterion in the elevation of Gelidiaceae to an order, Kylin (1923) discussed these differences but stated that the life history character should not be used systematically at that time. Both Nemaliales and Gelidiales lacked a "typical" auxiliarj' cell, i.e. any cell that the carpogonium fused with and from which the gonimoblast developed (Kylin, 1928, 1956), and thus were set apart from the other orders (Cryptonemiales, Gigartinales, Rhodymeniales and Ceramilales) which possessed a "typical" auxiliary cell. Kylin (1928) studied pre- and post-fertilization events in detail in Gelidiaceae, and the family was important in his formulation of the distinction between "generative" ("typical") and "nutritive" auxiliary cells. Even after his addition of the criterion that a typical auxiliary cell also not be a carpogonial branch cell (Kylin, 1935, 1937), he still maintained that a "typical" auxiliary cell was absent in Gelidiales. Kylin (1928) believed that the gonimoblast in Gelidiales developed directly from Carpogonium and Carposporophyte Development / 95 the unchanged, fertilized carpogonium. When this was found to be erroneous and that the carpogonium underwent non-obligate fusions with adjacent cells (Dixon, 1959), the presence of generative auxiliary cells became open to interpretation and dependent on the definition of the auxiliary cell. Drew (1954) gave the most useful definition, as a cell of specified position in the thallus with which the carpogonium fuses prior to gonimoblast formation. Accordingly, Gelidiales lack an auxiliary cell. Details of the auxiliary cell debate are given by Santelices (1974) and Hommersand and Fredericq (1988). The ordinal position of Gelidiales has been questioned extensively (Dixon, 1959, 1961; Papenfuss, 1966), being largely dependent on the perceived presence or absence of an auxiliary cell and, in Dixon's (1961) view, the lack of life history differences, i.e. haplobiontic Nemaliales (as Nemalionales) and diplobiontic Gelidiales. Dixon (1961) argued for returning Gelidiales to Nemaliales as Kylin's (1923, 1928, 1956) characters separating the orders were not valid. On the other hand, Papenfuss (1966) argued that Gelidiales was supported at ordinal rank by the presence of unique chains or nutritive auxiliary cells, by the presence of only two periaxial cells, and by the unique pattern of spore germination. It is important that carpogonium and carposporophyte development be studied in more gelidioids to assess variation at the genus and species levels and to determine features of the developmental processes that are common to all members of the order. The present study contributes to a more complete understanding of carpogonium and carposporophyte development in the order. Carpogonium and Carposporophyte Development / 96 B. METHODS Of the local gelidiaceous species, fertile (female) gametophytes of only Gelidium vagum and G. purpurascens occurred in the study area. Carpogonium and carposporophyte development followed the same pattern in both species. The process also was observed, in less detail, in sections from herbarium specimens of G. robustum (UC 395419, 647822, 756470, 940173; UBC A62199). Observations were made on dried and Formalin preserved material following the light microscope methods described in Chapter 3. Hematoxylin stained material also was used, prepared as in Chapter 4. The only modification for the purpose at hand (as opposed to chromosome counts) was to make light squashes of apices and cystocarps, enough to spread out clusters of filaments, but not enough to break the continuity of filaments. Hand made razor blade sections were stained with hematoxylin for 1 h, destained with 45% acetic acid for 15-30 s, and rinsed with water. C. CARPOGONIUM DEVELOPMENT IN GELIDIUM The carpogonium of Gelidium differentiates from the second to basal cell of a third order filament that is a normal part of the vegetative thallus. In the vegetative condition, the second to basal cell of a third order filament is intercalary with three pit-connections: one to the basal cell of the filament, one to the third cell of the filament, and one to another distal cell basal in a fourth order filament. When differentiated as a carpogonium, the second to basal cell of Carpogonium and Carposporophyte Development / 97 a third order filament bears only two pit-connections (Fig. 182), one to the basal cell and one to the third cell of the third order filament. The fourth order vegetative filament and any of its higher order branches, is replaced by the gradually tapering, arched trichogyne of the carpogonium. Carpogonia are produced sequentially along the axis in the fertile area (Fig. 47). Two rows of carpogonia are produced lateral to the axial row, on third order filaments, extending to both surfaces of the compressed to flattened axis. Thus four rows of carpogonia are formed, with two rows visible in either a longitudinal or saggital section. Replacement of a cluster of vegetative cortical cells by the trichogyne of each carpogonium results in the production of a medial gap or furrow on both thallus surfaces (Fig. 89). Trichogjmes protrude to the thallus surface through this gap. The carpogonium is initially intercalary. Farther back in the fertile zone are sessile carpogonia, cut off from the supporting cell by a distinctive concave wall (Fig. 179). These carpogonia have a single pit-connection to an intercalary supporting cell, and a second to basal cell in a third order filament. The question arises whether intercalary carpogonia are immature, and require a final division to become mature and sessile, or whether intercalary carpogonia are mature and functional, and undergo a division that discards the trichogyne after being past receptivity to spermatia. Many sessile carpogonia were observed in a degenerating condition, whereas the supporting cell appeared healthy (i.e. cytoplasm dense as in adjacent cells) (Fig. 180). An intercalary carpogonium, just fertilized, was observed at an early stage of fusion cell initiation, fusing to cortical cells through two expanded pit-connections (Fig. 181). Thus functional carpogonia are intercalary in Gelidium. Carpogonium and Carposporophyte Development / 98 D. CARPOSPOROPHYTE AND CYSTOCARP DEVELOPMENT Following fertilization, the carpogonium expands and fuses with several adjacent cells through widened pit-connections, forming a large, irregularly lobed, multinucleate fusion cell (Figs. 182, 183). Occasionally two fusion cells can be seen in a single apex, suggesting that two carpogonia may be fertilized and develop a genetically heterogenous carposporophyte in a single cystocarp. It is difficult to be certain of this, however, as fusion cells may have very long, narrow lobes. When the fusion cells are widely spaced it seems likely that the carposporophytes are discrete. Concomitant with fertilization, short chains of small, isodiametric cells, referred to as "nutritive filaments" (Hommersand and Fredericq, 1988), are cut off from the bases of cells of third order filaments (Figs. 183, 184). In Gelidium vagum, fusion cells always are present in apices where nutritive filaments are seen, suggesting that nutritive filaments are initiated at fertilization. The timing of nutritive filament formation may be slightly earlier in G. purpurascens, as nutritive filaments are present in apices that appeared to lack a fusion cell. Nutritive filaments curve in, towards and around the plate of second order cells (Fig. 183). They reach a maximum length of six cells, the apical cell of which is slightty larger and round, other cells being short and cylindrical (Fig. 184). All nutritive filament cells have prominent nuclei. Fusion cell lobes cut off weakly staining, uninucleate gonimoblast cells, which form branching chains winding around and between cells of the second order Carpogonium and Carposporophyte Development / 99 plate (Fig. 185). Gonimoblast cells cut off elongated processes that contact and fuse with apical cells of nutritive filaments (Fig. 186). This fusion is between the diploid gonimoblast (carposporophyte) cells and haploid nutritive filament (female gametophyte) cells. The gonimoblast produces single carposporangia laterally and terminally (Fig. 188), which project into one of the two locules created between the second order plate and cortex at cystocarp expansion (Fig. 187). Uninucleate carposporangia expand to become ovate to obpyriform with a darkly staining cytoplasm and stellate chloroplast. Carposporangia are produced continually by the gonimoblast, and a variety of ages is seen in a cystocarp. The growing carposporophyte pushes the cortex away from the second order plate, creating locules and causing third order filament cells to stretch. In Gelidium vagum these filaments break as the cystocarp matures (Fig. 187), whereas they remain intact in G. purpurascens (Fig. 50). The cystocarp cortex does not increase in thickness over the vegetative cortex, although some cells stretch laterally. At maturity the cystocarp has expanded beyond the width of the vegetative branch. Locules are filled with carposporangia and faintly staining mucilage (Fig. 187). Nutritive filaments no longer are distinguishable, and the entire carposporophyte (except the sporangia) appears vacuolate, staining poorly (with aniline blue). One ostiole per locule forms in the cystocarp cortex, due to the failure of part of the cortex to fill in. It is a simple, round opening and is not beaked or protruding. There is no evidence that the ostiole forms from tearing of the cortex. Carpospores are released, leaving behind some mucilage in the Carpogonium and Carposporophyte Development / 100 cystocarp. There is no evidence of percurrent production by carposporangia. E. DISCUSSION The most significant observation made concerns the intercalary carpogonium in Gelidium. Dixon (1959) and Fan (1961) both noticed intercalary carpogonia but their interpretations differed. Intercalary carpogonia were interpreted as early developmental stages by Fan and as aberrant gametes by Dixon. They both agreed that intercalary carpogonia are non-functional. They saw sessile carpogonia as mature, but no basis for this assumption was provided, on the other hand, Hommersand and Fredericq (1988) regarded intercalary carpogonia as functional and sessile carpogonia as non-functional; while the}' examined the material no illustrations were provided. In this study on Gelidium, an intercalary carpogonium was observed fusing with cortical cells through two widened pit-connections at a verj' ear^' stage of fusion cell formation. Sessile carpogonia never formed fusion cells and often were seen in stages of degeneration, suggesting a non-functional condition. Reports of sessile carpogonia fusing with the supporting cell following fertilization (Dixon, 1959; Fan, 1961), stem from observations of fusion cells where the outline of the carpogonium is visible, and where the fusion cell narrows somewhat before expanding at what is believed to be the supporting cell. Earl}' stages of this supposed fusion are not reported or illustrated. In G. vagum and G. purpurascens, however, intercalary carpogonia are expanded at the base, particular!}' near pit-connections. Hommersand and Fredericq (1988) reported the Carpogonium and Carposporophyte Development / 101 retention of part of the carpogonium outline in the fusion cell, and this could explain the earlier reports of Fan (1961) and Dixon (1959), if the basal carpogonium outline is retained as well as the trichogyne lobe. It should be noted that fusion cells in G. vagum and G. purpurascens showed no carpogonial outline, but this observation probably depends on fusion cell age. Thus the functional carpogonia of Gelidium are intercalary, as noted by Hommersand and Fredericq (1988). If unfertilized, the carpogonium can cut off the trichogyne lobe, and the basal portion can then revert to a vegetative cortical cell. An apical notch is reported in female branchlets of Gelidium robustum (Fan, 1961) and G. pteridifolium (Hommersand and Fredericq, 1988), but is absent in British Columbia G. purpurascens and G. vagum. Hommersand and Fredericq (1988) speculated that retarded growth of the apical cell and axial row, evidenced by overgrowth of the apical cell by adjacent cortical lobes, was responsible for development of the axial furrow. Growth of third and higher order filaments near the axial row was outpaced by equivalent filaments lateral and distal to the axial row. Since there does not appear to be retarded axial row development in species lacking an apical notch, the axial furrow was interpreted as originating from the absence of cortical cells that belonged to filaments replaced by the carpogonium. Timing of the formation of nutritive filaments varied between studies and may be a taxonomically useful character. In Gelidium pteridifolium nutritive filaments are initiated as the carpogonium differentiates, but before fertilization (Hommersand and Fredericq, 1988). In G. latifolium, and G. pulchellum (Dixon, Carpogonium and Carposporophyte Development / 102 1959), G. robustum (Fan, 1961) and G. vagum (herein), fertilization is probably the stimulus for nutritive filament production. It is difficult to be certain from reports by Dixon and Fan, as both believed the functional carpogonium was intercalary. Thus when Dixon (1959) observed that nutritive filaments formed after the carpogonium was mature, the observation corresponds to filaments forming after the carpogonium became sessile and non-functional, while other functional carpogonia probably were being fertilized. Fan (1961) noted that nutritive filaments form during carpogonium development, but the exact timing is not known with respect to functional carpogonia or fusion cell formation. While there is general agreement that, upon fertilization, the carpogonium enlarges and becomes lobed, there is no agreement on the nature of fusions with surrounding cortical cells. Fan (1961) reported no fusions between the fusion cell and cortical cells, whereas Dixon (1959) accepted that fusions may occur but are not obligate. In this study and in all fusion cells seen by Hommersand and Fredericq (1988) there were fusions with cortical cells. My findings agree with Hommersand and Fredericq (1988) that fusions of the carpogonium with cortical cells are a regular feature of earty carposporophyte development in Gelidium, but that the fusions may not be with cells in a specified position with respect to the carpogonium. Fusions regularly occur between cells of nutritive filaments and gonimoblast cells. Nutritive filament cells clearly function as nutritive auxiliary cells, as their initially dense cytoplasm becomes sparse and vacuolate after fusions and gonimoblast growth. I suspect that the reported lack of fusions between nutritive Carpogonium and Carposporophyte Development / 103 filaments and gonimoblast (Dixon, 1959) resulted simply from failure to observe them and not their absence. Confirmation should be made, as substantiation of Dixon's (1959) observation could be important systematically. Gelidium latifolium and G. pulchellum differ from G. purpurascens and G. vagum in carposporangial initiation. In the former two species, clusters of uninucleate initials are cut off from gonimoblast cells that develop into carposporangia (Dixon, 1959). In the British Columbian Gelidium species, carposporangia are produced singly, terminal^ and laterally on gonimoblast cells, rather than in clusters. In G. latifolium and G. pulchellum, the cystocarp cortex is thicker than the vegetative cortex and ostioles do not form; carpospores are released by degeneration of the pericarp (Dixon, 1959). This is different from G. purpurascens and G. vagum where the cystocarp cortex is not thickened over the vegetative cortex and one ostiole is present per locule. In G. pteridifolium the "potential ostiolar regions" that develop have plugs that later break down allowing carpospore release (Hommersand and Fredericq, 1988). Returning briefly to the question of the presence of an (generative) auxiliary cell in Gelidiales, discussion of this problem is given by Hommersand and Fredericq (1988). It is clear that the gonimoblast develops from lobes of the fusion cell and not from an unchanged carpogonium as Kylin (1928) proposed. Having shown in this work that the functional carpogonium is intercalary and not sessile, observations interpreted as carpogonia fusing with their supporting cells are erroneous and are actually of the unequal enlargement of some parts of the carpogonium (e.g. base and pit-connection region) relative to others. Using Drew's Carpogonium and Carposporophyte Development / 104 (1954) definition of an auxiliary cell, the possibility that the supporting cell functions as an auxiliary cell can be eliminated, as a supporting cell does not exist in Gelidium. Given that fusions with cortical cells occur, the question of the nature of the auxiliary cell is reduced to whether any particular cortical cell is always the cell with which the carpogonium fuses and whether it functions in initiation of the gonimoblast or is strictly nutritive. I concur with Hommersand and Fredericq (1988) that there is, as yet, too little information available on cortical cell specificity or function. Recent observations do not suggest that a specified cortical cell is involved, and gonimoblast was not seen developing from the part of the fusion cell that fused with cortical cells. In light of Hommersand and Fredericq's (1988) findings, and because some previously reported developmental patterns have been generalized as "typical for the Gelidiales" (e.g. Kraft, 1976), re-investigations of previously studied species would be in order to clarify debated aspects. CHAPTER VI. SPERMATOGENESIS IN GELIDIUM A. INTRODUCTION In 1925, Grubb wrote the following concerning the state of knowledge of male organs in the Florideae (Grubb, 1925). Although nearly 150 years have passed since the first record of spermatia in the red algae occurred in print, our knowledge of these minute bodies which play so important a part in the reproductive processes of Rhodophyceae is surprisingly inadequate and limited. Records of the observation of antheridia in more than 120 European species of the Florideae are to be found scattered in algal literature, but of these the vast majority are simply notes to the effect that male plants have been seen and recognized. A certain number give a short description detailing the position of the antheridia, and some include a slight account of their structure, but owing to the nature of the material, modern cytological methods of investigation have only been brought to bear on a few forms. A similar situation still exists in 1988, even though spermatangia are known for many more species, and their development at the light microscope level has been studied to some extent. Few observations of the development of red algal male gametes have been made using electron microscope techniques. The male gamete and associated structures generally do not play an important role in red algal taxonomy. The presence of spermatangia and their position on the thallus are noted, but spermatangial morphology usually is not used as a taxonomic character. Part of the reason for this is the often cryptic nature of male gametangia and the resulting rare collections of male gametophytes or their recognition as such in collections. The importance of male characters in taxonomy 105 Spermatiogenesis in Gelidium I 106 also depends on the taxonomic rank at which the characters can be applied. Gabrielson and Garbary (1987) recently have used spermatangial characters at the ordinal rank in the construction of phylogenetic trees. For example, the transverse (as opposed to oblique) division of the spermatangial mother cell to form a spermatangium is characteristic of Gelidiales (Gabrielson and Garbary, 1987). At the generic rank, spermatangial position has been used to separate Gracilaria into subgenera (Yamamoto, 1975) and to differentiate the genera Gracilaria and Polycavernosa (Bangmei and Abbott, 1985). As has been discussed earlier in this work, the character of spermatangia occurring on the same plants as carpogonia (i.e. monoecy) instead of on separate male gametophytes (i.e. dioecy) can be used at the specific rank to recognize Gelidium vagum and G. johnstonii. Spermatangia are known for appproximately 30-40% of gelidiaceous species, but their development seldom has been followed (Dixon, 1959; Akatsuka, 1970, 1973; Tazawa, 1975). This is understandable given the small size of spermatangial mother cells, spermatangia and spermatia. Among the nine electron microscope studies of male gamete development in higher rhodophytes, none has been carried out on a member of the Gelidiales. Electron microscopj' of male gametes was performed primarily on ceramialean genera (Peyriere, 1971, Griffahsia flosculosa; Kugrens and West, 1972b, Levringiella gardneri and Erythrocystis saccata; Scott and Dixon, 1973a, Ptilota densa; Kugrens, 1974, Janczewskia gardneri; Scott et al., 1980, Polysiphonia denudata and P. harveyi; Kugrens, • 1980, Polysiphonia hendryi), and Bonnemaisonia Spermatiogenesis in Gelidium I 107 hamifera (Bonnemaisoniales) also was studied (Simon-Bichard-Breaud, 1971, 1972a, 1972b). In addition, there are some fragmentary reports on other higher rhodophytes (Peyriere, 1974, Polysiphonia, Rhodomela, Laurencia, Polyneura, Nitophyllum, and Furcellaria). In this study spermatiogenesis in two of the four gelidioids was followed using light and electron microscopy, as well as in a third species using only light microscopy. B. METHODS The species used in this study were Gelidium purpurascens, G. vagum and G. coulteri (light microscopy only). Light microscopy methods are included in Chapter 3. Both field and cultured material of G. vagum was fixed for electron microscopy. Gelidium purpurascens male gametophytes were not seen in the field, consequently material for this study on spermatiogenesis was obtained from male gametophytes grown to maturity in culture from tetraspores. Freshly collected field material was acclimated to a constant photoperiodic regime (12:8 LD in 20° C) in order to sjmchronize cell divisions with light and dark periods. Material was fixed 1 to 2 h after chamber lights came on to maximize the chance of seeing cell division (van der Meer, pers. comm.). Fixation was carried out in the refrigerator for 7 h in 2.5% glutaraldehyde, a 1:1 mixture of 5% glutaraldehyde and Sorensen's phosphate buffer (pH 7.2). For rapid fixation and to prevent blockage of intercellular spaces by air bubbles, tissue was cut from plants in a drop of fixative. Fixed tissue was postfixed in 1% osmium tetroxide in the refrigerator for 16 h and then dehydrated. Dehydration was Spermatiogenesis in Gelidium I 108 carried out in a graded methanol series. Material was infiltrated in propylene oxide, followed by embedding in a graded series of (10-100%) Spurr's epoxy resin. In the case of overnight steps, tissue was placed on the rotator at room temperature for the first and last hours, and in the refrigerator (not rotating) for the time in between. A sample protocol of the transmission electron microscope fixation procedure for spermatangia is provided in Appendix 3. Gold-purple to silver sections (170-75 um thick) of embedded material were cut on a Reichert OM U3 ultramicrotome using a Dupont diamond knife. Grids were stained in saturated uranyl acetate in 70% methanol for 45 min, and then with Reynolds' lead citrate (Reynolds, 1963) for 5.5 min. Sections were viewed and photographed in Zeiss EM10A and EM10C transmission electron microscopes. C. SPERMATIOGENESIS Spermatangial development is the same in all Gelidium species in this study. Spermatangia are borne in a superficial layer (Fig. 189) on short, ultimate branchlets of gametophytes. In monoecious G. vagum, they occur proximal to the female area of branchlets, a short distance behind the apex. Spermatangia are initiated and mature at the same time as the more distal carpogonia. In dioecious G. coulteri and G. purpurascens, spermatangia extend practically to the apex of the branchlet, and the size of the fertile area expands as the branchlet lengthens. Spermatangia are conspicuous by being colourless and associated with a "halo" of mucilage. Spermatiogenesis in Gelidium I 109 Vegetative cortical cells are pigmented; their dominant cytological components are a few large chloroplasts located peripherally (Fig. 191). The nucleus is usually basal, and the central part of the cell is occupied by a vacuole (Fig. 191). In surface view, cells in male areas are paler and smaller in diameter than vegetative cortical cells, and grouped in two's or four's (Fig. 190). These smaller diameter cells are spermatangial mother cells produced by the anticlinal halving of a surface cortical cell and the subsequent division of these cells in half (Fig. 192). Spermatangial mother cells thus have pit-connections to a subtending cortical cell and/or to another spermatangial mother cell (Fig. 192). Elongated spermatangial mother cells contain apical plastids ranging from proplastids to fully developed chloroplasts, typical of red algae, with an inner encircling thylakoid and aligned single thylakoids (Figs. 193, 203) bearing phycobilisomes. Starch grains accumulate in the basal half of spermatangial mother cells (Figs. 192, 193), and the nucleus also is basal, elongating into a central position (Fig. 193) prior to nuclear division (Fig. 194). Although a large number of cells were observed, and plants were fixed at times when, according to previous reports, the possibility of seeing dividing cells was highest (Scott et al., 1980; Scott, pers. comm.; Davis and Scott, 1986; van der Meer, pers. comm.), mitotic stages were not seen. After karyokinesis, the spermatangial mother cell nucleus returns to a basal position (Fig. 194). Cytokinesis occurs in a transverse plane characteristic of the Gelidiales, dividing the spermatangial mother cell approximately in half (Fig. 204) and producing a distal spermatangium. In longitudinal sections of fertile male branches, the transverse divisions of spermatangial mother cells help distinguish spermatangial from Spermatiogenesis in Gelidium I 110 vegetative areas where cortical cells divide somewhat obliquely. The spermatangium contains a prominent nucleus with condensed chromatin (Figs. 197, 201), endoplasmic reticulum and numerous mitochondria sometimes associated with dictyosomes (Figs. 195), but lacks chloroplasts and proplastids. If all chloroplasts are not successfully excluded prior to cytokinesis of the spermatangial mother cell, they must degenerate very soon after the spermatangium is cut off as extrusion of membranes is sometimes observed in the young spermatangium (Fig. 198). In some views, spermatangial mother cells appear to be expelling an entire chloroplast by either a furrow or by reforming the plasma membrane proximal to the chloroplast (Fig. 193). Soon after the spermatangium is cut off, many cored vesicles appear in the cytoplasm. Their origin is unclear, although similar cored vesicles were dictyosome-derived in developing tetrasporangia (see Chapter 7). Cored vesicles fuse with the plasmalemma (Fig. 199), contributing to a finely fibrillar, mucilage-like layer surrounding the spermatangium. The spermatangium is surrounded by the spermatangial mother cell wall (Fig. 205). One or two large spermatangial vacuoles form in the basal part of the spermatangium (Fig. 199), apparently from the coalescence of uncored, dictyosome-derived vesicles (Fig. 195), vesiculated ER (Fig. 196), or both. Spermatangial vacuoles can be discerned in the light microscope as unstained areas at the base of spermatangia (Fig. 189). Spermatangial vcauoles are filled with whorls of fibrillar material (Figs. 197, 200) and enlarge to a volume approximately equal to that of the remainder of the spermatangial cytoplasm. Release of the spermatangial vacuole along with adjacent cytoplasm (Fig. 203) is Spermatiogenesis in Gelidium I 111 followed by release of a single spermatium from the spermatangium. In early stages of spermatium release, the discarded pit plug (originally between the spermatangial mother cell and spermatangium) is seen between the spermatangial vacuole and spermatangial mother cell. The new spermatangial mother cell wall is thicker and interspersed with vesicles near the pit plug (Fig. 205). The spermatangial mother cell adjacent to this area of thickened wall contains additional vesicles, some of which are released to the wall. The space between the spermatangial vacuole and spermatangial mother cell often contains many vesicle remnants. (Figs. 204, 205) that could function in gamete release. The spermatium being released is squeezed through a weakened and ruptured area of the thallus wall (Fig. 202). It is conical in shape (Fig. 206), rounding up when free of the gametophyte. The spermatium is wall-less and contains a prominent nucleus with much condensed chromatin (Fig. 206), several large, oval mitochondria and abundant cored vesicles that continue to be released to the cell exterior (Fig. 206). Chloroplasts or proplastids are absent. Percurrent production of spermatangia occurs when the spermatangial mother cell expands into the space vacated by the previous spermatangium and cuts off another spermatangium (Figs. 205, 207). Occasionally, two spermatia are seen near the spermatangial mother cell that produced them (Fig. 205), suggesting rapid spermatium production. Loose pit plugs and successive old spermatangial mother cell wall layers indicate that at least four to five spermatangia can be produced percurrently from a single spermatangial mother cell (Figs. 207, 208). Spermatiogenesis in Gelidium I 112 D. COMPARISON OF SPERMATIOGENESIS IN GELIDIUM AND OTHER FLORIDEOPHYCEAN GENERA Detailed ultrastructural studies of florideophyte spermatiogenesis have been made on Griffithsia flosculosa (Peyriere, 1971), Levringiella gardneri, Erythrocystis saccata (Kugrens and West, 1972b), Ptilota densa (Scott and Dixon, 1973a), Janczewskia gardneri (Kugrens, 1974), Polysiphonia harveyi, P. denudata (Scott et al., 1980), P. hendryi (Kugrens, 1980) and Bonnemaisonia hamifera (Simon-Bichard-Breaud, 1971, 1972a, 1972b) and the general pattern of male gamete development is similar in all, although there are some differences in certain details. For clarification, it should be noted that Dixon (1959) used a terminology that is contrary to other authors. His "spermatangial mother cell", which expands giving rise to "spermatangia" by oblique divisions, is equivalent to a cortical cell producing two to four spermatangial mother cells (Dixon, 1959). Likewise, his "spermatangium", which divides transversely, is equal to the the spermatangial mother cell dividing to produce spermatangia (Dixon, 1959). Kugrens' (1980) interpretation of spermatangium and spermatium in Polysiphonia hendryi generated terminology that also is at variance with that of other authors. He interpreted the spermatangial mother cell as producing spermatia, the wall matrix surrounding the spermatia representing the spermatangium. It is unclear whether the spermatangium also surrounds the "spermatial" mother cell, but if so, this spermatangium cannot be homologous to the spermatangium of other genera which excludes the spermatangial mother cell. Clearly the outer wall matrix, surrounding the spermatangial mother cell and developing male gametes, Spermatiogenesis in Gelidium I 113 was produced by the periaxial cell before production of spermatangial mother cells and is equivalent to the wall produced by other vegetative periaxial cells. Thus I have retained the usage of other authors for (e.g. Scott and Dixon, 1973a) for "spermatangial mother cell" which produces a "spermatangium" and releases a "spermatium". I do agree with Kugrens' (1980) designation spermatangial "vacuole", as opposed to "vesicle" and have followed his interpretation. Spermatangial mother cells differ from vegetative cortical cells in that their chloroplasts are less well developed (Kugrens, 1974; Peyriere, 1974), or, if well developed (i.e. with many single parallel thylakoids), are fewer and less dominant components of the cell (Kugrens and West, 1972b), resulting in the reduced pigmentation seen at the light level. The latter is true for the Gelidium species studied here. Spermatangial mother cells also are less vacuolate than cortical cells (Peyriere, 1971; Scott and Dixon, 1973a; Kugrens, 1974; this study), with smaller and more basal vacuoles, except in Polysiphonia hendryi where spermatangial mother cells have differentiated directly from pericentral cells and are thus vacuolate (Kugrens, 1980). The transverse division of the spermatangial mother cell to produce a spermatangium is characteristic of Gelidiales and of the genera Gracilaria and Hypnea (Tazawa, 1975). Spermatangia are cut off by oblique divisions in other genera (Tazawa, 1975). In Gelidium each spermatangial mother cell produces a single spermatangium at any one time, and over time sequential production of spermatangia occurs. In some species, however, (e.g. Levringiella gardneri and Erythrocystis saccata) several spermatangia are produced simultaneously by a single spermatangial mother cell Spermatiogenesis in Gelidium I 114 (Kugrens and West, 1972b), and percurrent production is not mentioned or visible in illustrations. In other species (e.g. Ptilota densa, Janczewskia gardneri and Polysiphonia hendryi) multiple, percurrent spermatangia are cut off from a spermatangial mother cell (Scott and Dixon, 1973a; Kugrens, 1974, 1980 respectively). Chloroplasts in varying forms may be present within spermatangia or are absent (e.g. Gelidium). If present, they most often are rare (Ptilota densa: Scott and Dixon, 1973a), degenerating (Levringiella gardneri: Kugrens and West, 1972b), proplastids (Erythrocystis saccata: Kugrens and West, 1972b; Polysiphonia hendryi: Kugrens, 1980), or have fewer single thylakoids than vegetative cells (Griffithsia flosculosa: Peyriere, 1971). Organelles such as vacuoles, starch grains and the spermatangial mother cell nucleus in Griffithsia flosculosa (Peyriere, 1971) are positioned basally in the dividing spermatangial mother cell and are readity excluded from the spermatangium. But in Gelidium, well-developed chloroplasts occur apically in spermatangial mother cells. These chloroplasts appear to be released prior to spermatangial formation and then degenerate outside the cell. Evidence of expelled membranous material was seen in Gelidium and probably also occurs in Levringiella gardneri, Erythrocystis saccata (Kugrens and West, 1972b) and Ptilota densa (Scott and Dixon, 1973a). Spermatangia show abundant signs of metabolic activity. In Gelidium (this study), Griffithsia (Peyriere, 1971, 1974) and Bonnemaisonia (Simon-Bichard-Breaud, 1972a, 1972b), mitochondria are closely associated with dictyosomes. Cored vesicles were more abundant in Gelidium compared to Ptilota (Scott and Dixon, Spermatiogenesis in Gelidium I 115 1973a) and Janczewskia (Kugrens, 1974) and are probably dictyosome-derived. There is evidence that their contents are secreted extracellularly, contributing to wall development. Although in Griffithsia (Peyriere, 1971, 1974), Bonnemaisonia (Simon-Bichard-Breaud, 1972a), and Janczewskia (Kugrens, 1974), small amounts of cytoplasmic starch are present early in spermatangial development, none was seen in spermatangia of Gelidium. Because starch is always basal in spermatangial mother cells, it is unlikely to be included in newly formed spermatangia. Since plastids are absent, starch could not be formed de novo in spermatangia. The layers surrounding spermatangia in Gelidium differ from the three layered wall described in Janczewskia^ (Kugrens, 1974), Erythrocystis and Levringiella (Kugrens and West, 1972b). The outermost and thickest layer in Gelidium consists of very loose, fibrillar, mucilage-like material (similar in appearance to the contents of the spermatangial vacuole). Late in spermatangial development a thin, granular, inner wall layer appears. From the description given by Kugrens (1974), the second layer in Janczewskia spermatangia may correspond to the thick mucilage-like layer of Gelidium. The spermatangial wall of Polysiphonia hendryi (Kugrens, 1980) does not appear to have any layers analogous to those in Gelidium. In P. hendryi, a dark line of compressed fibrils forms a "separating layer" at the beginning of spermatium enlargement (Kugrens, 1980, Figs. 7, 12). In Kugrens' terminology, a spermatial wall (equivalent to spermatangial wall) with a distinctive reticulate fibrillar construction forms around the spermatangium and is gelatinous in nature (Kugrens, 1980, Fig. 12). Both a separating layer and a reticulate spermatangial wall are absent in Gelidium, but the reticulate Spermatiogenesis in Gelidium I 116 wall occupies the same position as the mucilage layer in Gelidium. A histochemical comparison of the reticulate wall and mucilage layer would be valuable. The spermatangial nucleus in Gelidium contains large amounts of condensed chromatin, but the components of the mitotic apparatus were not seen. It was proposed for Griffithsia flosculosa and Ptilota densa in which condensed chromatin also occurs that the spermatangial nucleus is arrested at prophase (Peyriere, 1974; Scott and Dixon, 1973a), hence the condensed chromatin. Contrarily, Kugrens (1974, 1980) noted that in released spermatia of Janczewskia and Polysiphonia hendryi the chromatin was dispersed. Other electron microscope studies of spermatiogenesis in florideophytes have reported a predominance of dictyosomes or ER, or a temporally changing dominance of both in spermatangia, the amount and kind of activity varying with species (Pe3rriere, 1974). Dictyosomes and ER vesicles are the source of secreted mucilages, wall material and the spermatangial vacuole. In Gelidium, spermatangial vacuoles were large, indicating much secretion. The source of this large volume of material in Gelidium is not obvious. Dictyosomes occasionally are seen but are not abundant, although cored vesicles believed to be derived from dictyosomes are abundant, suggesting rapid production of vesicles from few dictyosomes. Occasionally in the early stages of spermatangial vacuole formation, vesiculated ER is abundant basalry, seemingly involved in formation of the spermatangial vacuole. However, ER is not apparent in spermatangia alreadj' possessing small spermatangial vacuoles. Spermatiogenesis in Gelidium I 117 Spermatangial vacuoles are the most prominent developmental feature of spermatangia. They are usually one to three in number and at maturity occupy up to one half the spermatangial volume. Their origin is debated; an endoplasmic reticulum (ER) origin is suggested in Levringiella (Kugrens and West, 1972b), Janczewskia (Kugrens, 1974) and Polysiphonia hendryi (Kugrens, 1980), whereas in Erythrocystis (Kugrens and West, 1972b) and Bonnemaisonia (Simon-Bichard-Breaud, 1972a) the evidence indicates a dictyosome origin. In Gelidium, as in Ptilota (Scott and Dixon, 1973a), both dictyosome-derived vesicles and ER contribute to spermatangial vacuole formation. As determined by histochemical tests, the spermatangial vacuoles in Bonnemaisonia (Simon-Bichard-Breaud, 1972a) and Polysiphonia hendryi (Kugrens, 1980) contain polysaccharides. In Gelidium the internal structure consists of whorls of fibrils, but in Ptilota it also appears as bands of granular material (Scott and Dixon, 1973a). The spermatangial vacuolar contents of Polysiphonia hendryi initially are granular and later develop into fibrillar material (Kugrens, 1980), and a change in chemical reactivity or change in internal appearance occurs at maturity in Bonnemaisonia (Simon-Bichard-Breaud, 1972a). The position of spermatangial vacuoles also varies between taxa. In Bonnemaisonia two spermatangial vacuoles are produced, one of which is basal and the other apical (Simon-Bichard-Breaud, 1971, 1972a, 1972b). They mature and are released in these positions without migration. To date, Bonnemaisonia is the only species studied where all spermatangial vacuoles are not basal. The role of the spermatangial vacuole in the spermatangium has been debated. In Gelidium the spermatangial vacuole appears to function in release of the Spermatiogenesis in Gelidium I 118 spermatium. The vacuole always is released from the spermatangium before the gamete is released, and it remains approximately the same size and configuration after release. Thus it does not seem to contribute to the mucilage or adhesive coating of the spermatium, as has been suggested for Griffithsia (Peyriere, 1971), Levringiella and Erythrocystis (Kugrens and West, 1972b) and Ptilota (Scott and Dixon, 1973a). The spermatium mucilage probably is produced primarily from cored vesicles. The fact that the spermatangial vacuole never is retained within the Gelidium spermatium also suggests that it functions in spermatium release. In Levringiella (Kugrens and West, 1972b) and Polysiphonia hendryi (Kugrens, 1980), spermatia commonly retain their spermatangial vacuoles. No other workers have reported retention of spermatangial vacuoles in spermatia. It could reflect a real generic difference between taxa, or it could be due to accidental release of immature spermatia during electron microscopy preparation. Kugrens and West (1972b) proposed that osmotic pressure generated by the polysaccharides contained in the spermatangial vacuole forced the pit plug between the spermatangial mother cell and spermatangium to break, enabling spermatial release. However, in Polysiphonia hendryi (Kugrens, 1980) the pit plug ruptures by spermatangial wall deposition and spermatangium enlargement and the separating layer breaks, facilitating release of the spermatium. In several cases the pit plug in Gelidium was. unattached on both faces, and the spermatangial mother cell showed a plasma membrane profile suggesting secretion of wall material under the plug that effectively breaks any solid connection between the spermatangium and the spermatangial mother cell. Although spermatium release could not occur with an intact pit-connection, the breaking of the pit-connection may be necessary, but not sufficient, for spermatium release. It may be difficult to separate the Spermatiogenesis in Gelidium I 119 consequences of pit-connection breakage and spermatangial vacuole release, as expulsion of the spermatangial vacuole and adjacent cytoplasm effectively breaks the pit-connection. A detailed ultrastructural study of mature but unreleased spermatia [sometimes seen as a second spermatium in series over a spermatangial mother cell (Fig. 205)] may resolve this question. Dixon (1959) suggested that the lower cell of his "spermatangium" in Gelidium, i.e. the spermatangial mother cell, also released a spermatangium. Electron microscopy clearly shows that this does not occur, at least in the Gelidium species studied here. In the Gelidiales there are morphological variations in male gametophytes or the manner in which the spermatangia are borne on gametophytes. With the exception of Acanthopeltis japonica (Kaneko, 1968), Gelidium johnstonii and G. vagum, all gelidioids where males are known are dioecious. Spermatangia in these plants form close behind the apex of fertile branchlets, and as the apex grows the male area lengthens. In monoecious plants, however, spermatangia are prevented from forming immediately behind the fertile branchlet apex, as this area differentiates carpogonia and nutritive filaments. In local G. purpurascens, male gametophytes are isomorphic with female gametophytes and tetrasporophytes and are not smaller than female gametophytes as reported for G. latifolium (Dixon, 1959). Male plants or male segments of monoecious gametophytes are paler in colour than vegetative regions (Fan, 1961; Akatsuka, 1970, 1973, 1979). Electron Spermatic-genesis in Gelidium I 120 microscopy has shown that spermatangia lack chloroplasts and thus should be colourless. Spermatangial mother cells contain fewer and less prominent chloroplasts than vegetative cortical cells and might be expected to be somewhat pigmented. Male areas also may be recognized at low magnification by the thickened mucilage layer over spermatangia (Dixon, 1959; Akatsuka, 1970), or by the dilated and flatter apices of fertile branches (Tazawa, 1975). The latter character applies only to certain species (Gelidium amansii, G. pacificum, Pterocladia nana, Tazawa, 1975; P. caloglossoides, this work) and was not evident in G. coulteri, G. johnstonii, G. purpurascens or G. vagum. In transverse or longitudinal sections, spermatangia are distinguished easily from vegetative cortical cells by the transverse walls between spermatangial mother cells and spermatangia, and the narrower width of spermatangial mother cells. In contrast, vegetative cortical cells have oblique division planes. Despite the differences in how and where spermatangia are borne on gametophytes, the pattern of their development in florideophytes including Gelidium is remarkably uniform. In addition, two species of Gelidium studied here could not be differentiated at the ultrastructural level. Too few ultrastructural studies of spermatiogenesis have been carried out to know yet if there are variations on the basic pattern that may be of taxonomic significance. The mechanism of exclusion or elimination of chloroplasts from spermatangia, the mode of formation of spermatangial vacuoles, the number and position of spermatangial vacuoles and the presence or absence of percurrent production of spermatangia vary in studies to date and may have some potential Spermatiogenesis in Gelidium I 121 for use in taxonomic studies. The mechanism of spermatium release also needs clarification. CHAPTER VII. ULTRASTRUCTURE OF TETRASPOROGENESIS IN GELIDIUM A. INTRODUCTION Tetrasporogenesis in red algae has been studied more extensively than spermatiogenesis at both light microscopy and ultrastructural levels. The origin of tetrasporangia and production of tetraspores have been noted in many gelidiaceous species. Fan (1961) remarked on the uniformity of tetrasporangial development in Gelidiales. Tetrasporangial characters (excluding arrangement on fertile branches) have not been used as taxonomic characters in Gelidiales, with the exception of the production of bispores in Suhria vittata and Onikusa pristoides (as Gelidium pristoides) (Fan, 1961). Tetrasporangial initials are borne laterally on subcortical cells, or "terminal in position on a lateral filament" as often stated (e.g. Fan, 1961). The young sporangium has a single pit-connection that is initially basal and becomes lower lateral as the sporangium expands. The tetrasporangium is deeply pigmented and divides into four cruciate or irregular cruciate, occasionally tetrahedral (Dixon, 1959) tetraspores. Release occurs through the degraded apical sporangial wall (Dixon, 1959). Additional sporangia are produced by the expansion of other initials after release of earlier formed sporangia. Ultrastructurally, tetrasporogenesis has been studied in the following Ceramiales, Griffithsia flosculosa (Peyriere, 1969, 1970), Levringiella gardneri (Kugrens and West, 1972c), Ceramium sp. (Chamberlain and Evans, 1973), Ptilota hypnoides (Scott and Dixon, 1973b), Callithamnion roseum (Konrad Hawkins, 1974b), 122 Ultrastructure of Tetrasporogenesis in Gelidium I 123 Polysiphonia denudata (Scott and Thomas, 1975; Alley and Scott, 1977), Erythrocystis montagnei (Santisi and De Masi, 1981), two members of Corallinales, Corallina officinalis (Peel et al., 1973), Haliptilon cuvieri (Vesk and Borowitzka, 1984), two species of Hildenbrandiales, Hildenbrandia rubra and H. occidentalis (Pueschel, 1982), one species of Rhodymeniales, Gastroclonium clavatum (Gori, 1982), and one of the Palmariales, Palmaria palmata (Pueschel, 1979). This ultrastructural studs' of tetrasporangial development in Gelidium, was undertaken in order to determine their patterns and compare them with previous studies of other red algae. B. METHODS Fertile tetrasporophytes of Gelidium coulteri, G. purpurascens and G. vagum were collected from the field or obtained from culture isolates originating from the study area. Tissue was prepared for electron microscopy as outlined in Chapter 6. However, fixation and osmication times were longer or at a warmer temperature to allow penetration of glutaraldehyde and osmium through the thick tetrasporangial walls. Material was fixed in glutaraldehyde for 4 h. at room temperature and postfixed in osmium for 18-19 h. at room temperature. The tissue preparation protocol that was used is given in Appendix 3. C. ULTRASTRUCTURAL OBSERVATIONS The process of tetraspore production and maturation in Gelidium coulteri, G. purpurascens and G. vagum is similar at both light and electron microscope Ultrastructure of Tetrasporogenesis in Gelidium I 124 levels, although there are differences between these species regarding the position of sporangia on stichidia. The account given here is applicable to all three local Gelidium species. Structures that exhibited the greatest amount of change from their appearance in vegetative cells were the nucleus, dictyosomes and wall. 1. Pre-meiotie Tetrasporangium The presence or absence of three distinctive wall layers serves as a reliable marker for events during development of mature tetraspores from the tetrasporangial initial cell. Vegetative cortical cells (similar to Fig. 191) from which tetrasporangia differentiate, have a central vacuole, basal nucleus, starch grains, and chloroplasts that are few in number, large and peripheral. Young sporangial initials contain the remnants of a vacuole (Fig. 210), and the vacuolar contents appear to be incorporated into vesicles and secreted extracellularly (Fig. 212). Occasionally, whorls of membranes are extruded from the cytoplasm (Fig. 211). In tetrasporangial initials, chloroplasts divide producing man3r smaller, ovate chloroplasts. Chloroplasts are unaligned and generally peripheral, but also may be scattered throughout the cytoplasm (Fig. 214). In very young tetrasporangia, dictyosomes are absent, or are so few that they were not seen in sections of cells, and starch grains are absent. The plasma membrane of young tetrasporangia is convoluted (Figs. 210, 216). Vegetative cells have a loose, fibrillar, electron transparent wall (layer 1) (Fig. 212) that is thickened to 100-250 nm wide during differentiation and early development of the sporangium (Fig. 215). Dictyosomes and vesicles that might be contributing wall material (Fig. 216) are not abundant, thus the source of wall components is unclear. Ultrastructure of Tetrasporogenesis in Gelidium I 125 In young tetrasporangia the interphase nucleus is enlarged and central (Figs. 214, 215) and contains a large nucleolus with an uneven granular composition (Figs. 213, 219). Nucleolar vacuoles are evident as areas of the nucleolus with the same electron density as the nucleoplasm (large arrows, Fig. 213), and have been associated with RNA synthesis (Peel et al., 1973). Pores are abundant in the nuclear envelope (arrowheads, Fig. 213) and perinuclear endoplasmic reticulum (PER) is present (small arrows, Fig. 213, Fig. 218, arrows, Fig. 219). Mitochondria are concentrated in the immediate vicinity of the nucleus. Dictyosomes appear just prior to karyokinesis and are small, flattened and up to 12 cisternae thick, with ends of distal cisternae inflated and saccate (Fig. 221). Larger cisternae contain an elongate core of moderately electron dense, fibrillar material (Fig. 221). The cisternal peripherj', as well as the entire contents of young cisternae, are electron transparent and sparsely fibrillar. There is a close association between mitochondria and the forming face of dictyosomes (Fig. 221). The narrow space between these organelles is unlike the surrounding cytoplasm in that it is free of ribosomes. Small, narrow, lenticular starch grains also appear prior to karyokinesis (Figs. 214, 215, 218, 219), first located centrally in the vicinity of the nucleus and between chloroplasts, and later scattered throughout the tetrasporangium. Actual meiotic division stages or synaptonemal complexes were not observed in these preparations. Several elongated and spindle shaped nuclei with depressions at their ends (Figs. 218, 219) and one with a narrow waist and chromatin (Fig. 217) probably were preparing for division. Somewhat older tetrasporangia with two nuclei but had not yet undergone cytokinesis (Fig. 220). Ultrastructure of Tetrasporogenesis in Gelidium I 126 2. Post-meiotic Tetrasporangium Following karyokinesis, as the tetrasporangium enlarges, starch grains, chloroplasts and dictyosomes become more abundant (Fig 220). Starch grains change from lenticular to ovate in shape (Figs. 220, 224). Chloroplasts divide by pinching into two and become somewhat more elongate. Thej' are well developed, with an encircling thylakoid interior to the double chloroplast membrane, many parallel, single thylakoids, plastoglobuli and areas of DNA (Figs. 221, 225). Dictyosomes become more prominent, particularly around the periphery of the sporangium. At low magnification they are noticable as clusters of sacs (Fig. 223) that still are flat but composed of more cisternae with larger inflated sacs. They are abundant in the tetrasporangium at all stages of development. Mitochondria are small relative to chloroplasts and are closely associated with the forming face of all dictyosomes (Fig. 221). Endoplasmic reticulum (ER) also is present in tetrasporangia, although often difficult to distinguish in the dense cytoplasm (Figs. 221, 222). A second tetrasporangial wall layer (layer 2) (Fig. 222) is deposited inside the first layer. This occurs when starch grains and chloroplasts have become abundant in the cytoplasm (Figs. 220, 224) during karyokinesis but before cytokinesis. Layer 2 (arrow, Fig. 227) is narrower (20-25 nm wide) and more densely fibrillar than layer 1 and of medium electron density, appearing as a grey band between white (layer 1) and black (layer 3) bands (Figs. 227, 232). It is deposited by abundant dictyosome- (Fig. 225) and ER-derived vesicles (Figs. 222, 225). Ultrastructure of Tetrasporogenesis in Gelidium I 127 3. Tetrasporangium at Cytokinesis The tetrasporangium undergoes cytokinesis at the two-nucleate stage, observations supported by light microscope studies on hematoxylin-stained material. The second meiotic division then occurs in both cells followed by another cytokinesis in perpendicular planes, producing a cruciately divided tetrasporangium with four uninucleate tetraspores (Fig. 78). Each tetraspore nucleus is small and inconspicuous, obscured by the great abundance of other organelles and inclusions. Chloroplasts and starch grains are very abundant (Fig. 226). In some mature tetraspores, chloroplasts are oriented radially around the nucleus (Fig. 229), but this arrangement does not seem to be universal. Starch grains have reached their maximum size in the tetrasporangium after cytokinesis (tetraspores) (Fig. 226). They are electron transparent, non-membrane bound and appear to lack substructure although dark, shadow-like marks make them appear three dimensional and angular (Fig. 228). Endoplasmic reticulum occurs in the peripheral cytoplasm of older tetrasporangia contributing vesicles to wall formation (Fig. 231). In tetrasporangia that are undergoing the second cytokinesis, and shortly thereafter, dictyosomes are abundant throughout the cytoplasm (Fig. 226), but not as prominent as in earlier stages. They are small and obscured by an abundance of chloroplasts and starch grains. Dictyosomes also have undergone a striking morphological change. The3' are distinctly concave; all of the approximately 9-12 cisternae per stack are strongly curved (Figs. 228, 230). There is a line of electron dense material inside the closely associated cisternae of the younger Ultrastructure of Tetrasporogenesis in Gelidium. I 128 (forming face) half of the dictyosome (arrowhead, Fig. 230) that is progressively more restricted in extent in the central region of more mature cisternae (Fig. 230). The dark material occurs in closely appressed, narrowest parts of cisternae and is absent from the saccate ends. Cisternae that are innermost in the cup that is formed are oldest. Their inflated sacs separate and are released as spherical vesicles (cored vesicles) which have an electron dense core and a medium density fibrous periphery (Fig. 230). Cored vesicles release their contents to the cell wall, contributing to its formation (Fig. 231). Mature tetraspores have few cored vesicles and visible dictyosomes. The third and final wall layer of the tetrasporangium (layer 3) is produced at cytokinesis (Figs. 226, 227, 232). It is 25-55 nm wide, stains strongly with osmium and is densely fibrillar, with parallel fibrils much more clearly aligned than in other layers (Fig. 227). Soon after layer 3 is completed, a layer of whorled fibrillar material, similar to the mucilage produced by spermatangia and contained in spermatangial vacuoles, appears internal to layer 3 and surrounds the tetraspores (Fig. 232). As in spermatangia, this material represents the contents of dictyosome-derived cored vesicles (Figs. 231, 232). D. DISCUSSION General trends of developmental changes during tetrasporogenesis in Gelidium agree with patterns observed in other florideophytes. An extensive comparison and discussion of tetrasporogenesis was presented by Vesk and Borowitzka (1984). Some differences in details do occur between the species, and these may be Ultrastructure of Tetrasporogenesis in Gelidium I 129 taxonomically important. The tetrasporangial wall differs from the wall of the vegetative cell in most species (e.g. Ptilota hypnoides, Scott and Dixon, 1973b; Levringiella gardneri, Kugrens and West, 1972c; Haliptilon cuvieri, Vesk and Borokitzka, 1984), and a proteinaceous (Pueschel, 1979) cuticle forms [e.g. in Hildenbrandia rubra, H. occidentalis (Pueschel, 1982), Palmaria palmata (Pueschel, 1979), Callithamnion roseum (Konrad Hawkins, 1974b) and Ptilota hypnoides (Scott and Dixon, 1973b)]. It has been proposed that the cuticle and tetrasporangial wall function to isolate the tetrasporangial cytoplasm from the influence of vegetative cells and sporangia at other developmental stages (Vesk and Borowitzka, 1984). In Gelidium, a cuticle was not apparent and neither was a unique wall formed early in tetrasporangial development; the initial wall thickens and a second wall layer is not deposited until karyokinesis. Division figures and synaptonemal complexes were not observed in Gelidium. Synaptonemal complexes are definitive proof of meiosis, but have been seen in few studies (Kugrens and West, 1972a, Janczewskia gardneri, Levringiella gardneri, Gonimophyllum skottsbergii and Polycoryne gardneri; Pueschel, 1979, Palmaria palmata; Broadwater, Scott and Pobiner, 1986a, Dasya baillouviana). They occur in few sections because of their small size and thus are less likely to be seen than larger, more conspicuous indicators of nuclear division, such as spindle shaped nuclei and division furrows. Increase in chloroplast number throughout tetrasporangial development occurs by pinching of single chloroplasts into two. This has been observed in Gelidium and reported in Griffithsia flosculosa (Peyriere, 1969), Ptilota hypnoides (Scott and Ultrastructure of Tetrasporogenesis in Gelidium 1 130 Dixon, 1973b), Palmaria palmata (Pueschel, 1979), Gastroclonium clavatum (Gori, 1982) and Haliptilon cuvieri (Vesk and Borowitzka, 1984). Early in the development of P. palmata tetrasporangia, rapidly replicating chloroplasts have the appearance of proplastids with few or no single parallel thylakoids (Pueschel, 1979), while later in development plastids are differentiated with several parallel thylakoids. Chloroplasts in early stages of tetrasporogenesis in H. cuvieri had few thylakoids, but later became fully developed (Vesk and Borowitzka, 1984). Chloroplasts in Gelidium and Erythrocystis montagnei (Santisi and De Masi, 1981) are well developed at all stages of tetrasporogenesis; proplastids were not observed in this study. The importance of dictyosomes and ER in secretory activity is recognized in all ultrastructural studies of tetrasporogenesis (e.g. Peyriere, 1970; Chamberlain and Evans, 1973; Konrad Hawkins, 1974b; Alley and Scott, 1977). During periods of wall deposition, dictyosomes and dictyosome-derived vesicles become more abundant, particularly in the peripheral cytoplasm, and cisternae increase in number, becoming more dilated and hypertrophied. After wall and mucilage secretion is reduced (in mature tetraspores), dictyosomes decrease in dilation, size and number. Many observers of tetrasporogenesis report or illustrate close associations between mitochondria and the forming face of dictyosomes (Peyriere, 1969, 1970; Kugrens and West, 1972c; Chamberlain and Evans, 1973; Peel et al, 1973; Scott and Dixon, 1973b; Konrad Hawkins, 1974b; Scott and Thomas, 1975; Alley and Scott, 1977; Pueschel, 1979, 1982; Santisi and DeMasi, 1981; Gori, 1982; Vesk Ultrastructure of Tetrasporogenesis in Gelidium I 131 and Borowitzka, 1984; this study). Dark layers occur between adjacent, closely-appressed cisternae in the younger half of the hemispherical dictyosome. These have been likened to cementing layers (Konrad Hawkins, 1974b), but their nature is not known, and they have not been reported in other plants and animals (Alley and Scott, 1977). Products of dictyosomes also change through time, evidenced by the variety of vesicles reported to be dictyosome-derived. Without histological information it is difficult to determine the homology of vesicles reported as fibrillar (Kugrens and West, 1972c, Levringiella gardneri), granular (Kugrens and West, 1972c, Levringiella gardneri; Chamberlain and Evans, 1973, Ceramium sp.), dense (Peyriere, 1969, Griffithsia flosculosa), with a fibrillar core (Chamberlain and Evans, 1973, Ceramium sp.; Santisi and De Masi, 1981, Erythrocystis montagnei), and granular with a fibrillar outer region (Chamberlain and Evans, 1973, Ceramium sp.). Late in cytokinesis, dictyosomes of Gelidium (this study), Levringiella gardneri (Kugrens and West, 1972c), Ceramium sp. (Chamberlain and Evans, 1973), Callithamnion roseum (Konrad Hawkins, 1974b), Erythrocystis montagnei (Santisi and De Masi, 1981), Palmaria palmata (Pueschel, 1979), Gastroclonium clavatum (Gori, 1982) and Callithamnion roseum carpospores (Konrad Hawkins, 1974a) have a distinctive hemispherical morpholog3' and produce cored vesicles. Peyriere (1970) reported positive tests for acid mucopolj'saccharides within cored vesicles formed at two different developmental stages in Griffithsia flosculosa, and Tripodi and De Masi (1975) found that cored vesicles of carpospores in Pterocladia capillacea and Polysiphonia sertularioides (similar in appearance to Gelidium cored vesicles in tetrasporangia) gave a positive reaction to Thiery's test for polysaccharides. These cored vesicles are sometimes close to the plasma membrane, suggesting their contents are being Ultrastructure of Tetrasporogenesis in Gelidium I 132 released to the wall (Tripodi and De Masi, 1975). The contents of cored vesicles in Gelidium (this study) and Erythrocystis montagnei (Santisi and De Masi, 1981) are secreted to the inner tetrasporangium wall or mucilage layer and are absent from released tetraspores (Santisi and De Masi, 1981). Similar appearing vesicles are believed to function in spore adhesion in other species (Vesk and Borowitzka, 1984, Haliptilon cuvieri). Dictyosome cisternae undergo a maturing process as they move through the stack (by addition and attrition of other cisternae) from the proximal (forming) face to the distal face. Alley and Scott (1977) proposed that they arise at the forming face from the fusion of small, ER-derived vesicles, and vesicles in transition between ER and the forming face of dictyosomes were seen in Erythrocystis montagnei (Santisi and De Masi, 1981). In Gelidium, ER occurs near the forming face, but vesicles never were seen in the space between mitochondria and the youngest cisternae. It is possible that small contributing vesicles were not sectioned or that thej' are added laterally and not centrally. Konrad Hawkins (1974b) stated that ER vesicles from chloroplast blebs fuse to form cisternae, and also that at least some cisternae contain chloroplast enzjanes capable of starch degradation, as cisternae were seen near eroding starch grains. In Gelidium, chloroplasts in the general vicinity of dictyosomes showed no evidence of special positioning near dictyosomes, or of blebbing, and starch grains were not in an eroding or eroded state. Vesicle production from dictyosomes is by release of entire cisternae (Peyriere, 1969, 1970; Alley and Scott, 1977) or release of dilated ends of cisternae as vesicles (Konrad Hawkins, 1974b). In Gelidium, dictyosome-derived vesicles are smaller than a whole cisternum, indicating that Ultrastructure of Tetrasporogenesis in Gelidium I 133 vesicles probably are derived from cisternal ends. Endoplasmic reticulum also is known to produce striated vesicles (Kugrens and West, 1972c) and mucilage vesicles (Pueschel, 1979). In Gelidium, vesicles containing membranes or smaller vesicles, similar to multivesicular bodies or lomasomes (Scott and Dixon, 1973b; Chamberlain and Evans, 1973), release their contents to the wall. Endoplasmic reticulum is associated with the nucleus at meiosis as PER. PER was not abundant in Gelidium, but is extensive, oriented parallel to the nuclear envelope or radially at different times in Corallina officinalis (Peep et al., 1973) and Haliptilon cuvieri (Vesk and Borowitzka, 1984), and believed to be related to high metabolic and synthetic activity of the nucleus prior to, and throughout, meiosis. Extensive, radially-oriented PER may be characteristic of Corallinales; it has only been reported from this order, although it is absent in some corallines (e.g. Jania rubens (Linnaeus) Lamouroux, Peel et al., 1973; Lithothrix aspergillum Gray and Fosliella sp., Vesk and Borowitzka, 1984). Endoplasmic reticulum also is important in inner tetrasporangium wall formation in Haliptilon cuvieri (Vesk and Borowitzka, 1984) and is present but not conspicuous at the cytoplasm periphery in Gelidium. Vesicles in Gelidium tetrasporangia release contents to the wall by fusing with the plasma membrane. No vacuoles, similar to those reported in Corallina officinalis (Chamberlain and Evans, 1973), Ptilota hypnoides (Scott and Dixon, 1973b), Polysiphonia denudata (Alley and Scott, 1977), Erythrocystis montagnei Ultrastructure of Tetrasporogenesis in Gelidium I 134 (Santisi and De Masi, 1981) and Griffithsia flosculosa (Peyriere, 1970) were seen. Mucilage sacs, present in Palmaria palmata (Pueschel, 1979), also were not observed in Gelidium. Deposition of a series of cell wall layers results from the cytoplasmic secretory activity described above. Unlike all other species where tetrasporogenesis has been studied (Vesk and Borowitzka, 1984), Gelidium does not deposit a tetrasporangial initial wall early in differentiation that is different from the vegetative wall. A histochemical study of tetrasporangium wall layers in Gelidium would enable comparisons with vegetative walls, as the vegetative wall of G. pacificum Okamura has an outer sulfated polysaccharide layer and an inner cellulosic layer (Akatsuka and Iwamoto, 1979). Based on position, appearance and time of formation, the outer, thick, loosely fibrillar wall (layer 1) of the Gelidium tetrasporangium is probably equivalent to the tetrasporangial mother cell wall in Levringiella gardneri (Kugrens and West, 1972c) and Ptilota hypnoides (Scott and Dixon, 1973b), layer three in Callithamnion roseum (Konrad Hawkins, 1974b), and the inner tetrasporangial wall in Palmaria palmata (Pueschel, 1979), Hildenbrandia rubra and H. occidentalis (Pueschel, 1982). The medium density, grey wall (layer 2) in Gelidium is different from layer 1 and is similar in position and appearance to the tetrasporangial initial wall in Ptilota hypnoides (Scott and Dixon, 1973b). The last wall layer laid down in Gelidium (layer 3) is continuous with the dark layer separating tetraspores at cytokinesis, and a similar layer is recognizable in other genera. A mucilage layer also cleaves the tetrasporangium in other florideophyte genera (e.g. Hildenbrandia, Pueschel, 1982; Haliptilon, Vesk and Borowitzka, 1984). This layer is recognizable by its position with respect to Ultrastructure of Tetrasporogenesis in Gelidium I 135 the dark, inner wall and its swirled fibrillar appearance. Peyriere (1970) found in Griffithsia flosculosa that the last type of tetrasporangial secretion, which appears similar to the mucilage-like material in Gelidium, reacted positively for mucopolysaccharides in histochemical tests. In all ceramialean and corallinalean species studied ultrastructurally to date, tetraspores are tetrahedrally or zonately (respectively) arranged in the sporangium. Both meiotic divisions of the nucleus are completed before cytokinesis, and cleavage is simultaneous (Kugrens and West, 1972c, Levringiella gardneri; Scott and Dixon, 1973b, Ptilota hypnoides; Chamberlain and Evans, 1973, Ceramium sp.; Alley and Scott, 1977, Polysiphonia denudata; Vesk and Borowitzka, 1984, Haliptilon cuvieri; Broadwater, Scott and Pobiner, 1986b, Dasya baillouviana) except in Erythrocystis montagnei where the cleavage furrow is half completed before karyokinesis is over (Santisi and De Masi, 1981). Cytokinesis is initially sequential, and later simultaneous in Palmaria palmata with cruciately divided tetrasporangia. (Pueschel, 1979) and in the zonately to cruciate^ divided tetrasporangia of Hildenbrandia rubra and H. occidentalis (Pueschel, 1982). Pueschel (1979, 1982) observed the first furrow forming after meiosis and not being completed until after the second perpendicular furrows had been initiated. Gelidium also has cruciately divided tetrasporangia, but cytokinesis was not observed ultrastructurally. Light microscope studies with hematoxylin-stained tetrasporangia suggested that division is completely sequential and that the second meiotic division occurs after the first cytokinesis. In summary, whereas tetrasporogenesis follows a similar pattern in the Ultrastructure of Tetrasporogenesis in Gelidium I 136 florideophytes studied to date, some differences between genera have been recorded such as extent and orientation of PER, types, functions and sources of vesicles, and timing of cleavage furrow development. As in spermatiogenesis, there are not yet enough studies on several members of a taxon (i.e. genus, family, order) to determine whether these differences are phylogenetically significant or whether differences merely represent variations among individuals. CHAPTER VIII. GENERAL DISCUSSION AND CONCLUSION In spite of the great confusion confounding the taxonomy of Gelidium and Pterocladia species (e.g. Dixon, 1958), it is possible to determine species, if the range of morphological and anatomical variability within a species is studied and determined for each taxon with in a restricted geographic area. In this study, conducted in British Columbia and northern Washington, species were assigned to genera on the basis of cystocarp morphology. Where cystocarps were not known for local species (i.e. G. coulteri, P. caloglossoides), ecological, vegetative and tetrasporangial characters were used to assign names to taxa. Although six taxa were recorded originallj' from the area, it was found in this study that only four species of Gelidiales are present: Gelidium coulteri, G. purpurascens, G. vagum and P. caloglossoides. A key was constructed (Fig. 5) to enable identification of these species. An emphasis on biogeographic and developmental patterns of reproductive structures as well as vegetative features, and on ecological characters may prove useful in further taxonomic determinations of gelidioid algae. To date, the only character that can be used reliably to distinguish between Gelidium and Pterocladia is the number of cystocarp locules. However, some of the apical architecture characters noted by Rodriguez and Santelices (1987, unpubl. mscr.) show potential for separating some of the species of these genera vegetatively. Two of these characters, similarit.y/dissimilarity of apical and lateral initials and the indentation of lateral initials, were examined in the three Gelidium, G. coulteri, G. purpurascens and G. vagum, and the one Pterocladia species, P. caloglossoides, from the study area. Apical and lateral initials of the 137 General Discussion and Conclusion / 138 local gelidiaceous species were found to conform to the pattern: both types of initials are similar and domed in Pterocladia, but apical initials are domed and lateral initials are dissimilar, inverted conical in Gelidium. The second character given by Rodriguez and Santelices (1987, unpubl. mscr.) was not consistent for P. caloglossoides although the}' suggested that this character might have exceptions in some Pterocladia species, including P. caloglossoides. Thus this second character of indentation of initials is incapable of differentiating between Pterocladia from Gelidium in the study area. In the present study, neither apical nor lateral initials are indented between cortical lobes in P. caloglossoides. Thus the similarity in shape of apical and lateral initials in Pterocladia species, and the dissimilarity of these initials in Gelidium species, in addition to number of cystocarp locules, are characters that may be used to separate Gelidium and Pterocladia. Gelidium vagum undergoes a regular Polysiphonia-type life history in culture, and the common occurrence of gametophytes and tetrasporophytes suggests the same is operating in the field. According to culture evidence and field observations G. purpurascens in the study area is capable of, but probably rarely actually completes, a Polysiphonia-type life history in situ. Fertile male gametophytes grew from tetraspores of G. coulteri in culture, but there are no field collections of fertile gametophytes. The complete Polysiphonia-type life history probably occurs only occasionally, if at all, in G. coulteri from the study area and vegetative or apomictic mechanisms probably sustain the population most of the time. Pterocladia caloglossoides populations also persist predominantly vegetatively or apomictically. Chromosome counts of n= 14-15 were obtained from undivided tetrasporangia of G. vagum. General Discussion and Conclusion / 139 The pattern of spore germination was found to be similar for both carpospores and tetraspores. A single germ tube grows from the spore, the entire cytoplasmic contents of the spore evacuate into the germ tube and a wall cuts off the germ tube from the empty spore. The germ tube divides unequally to form a concave and a fusiform cell. The primary attachment rhizoid forms from a derivative of the concave cell, and the sporeling apical cell forms from a fusiform cell derivative. As apical organization is established the subapical cell produces two lateral periaxial cells. Studies of developmental patterns in local gelidiaceous species revealed some previously unknown or unsubstantiated aspects. Hommersand and Fredericq (1988) recently proposed that carpogonia in Gelidiales are intercalary and that sessile carpogonia reported by previous workers (Dixon, 1959; Fan, 1961) are non-functional, but provided no data. The present study showed that the fusion cell is initiated from an intercalary carpogonium. In light of Hommersand and Fredericq's (1988) observations and the corroborative findings of this stud}', female gamete and carposporoprryte development needs to be followed in detail for a wide range of other gelidiaceous species. Such an expanded study is predicted to reveal developmental variations that may be important as ordinal characters. Ultrastructure of spermatiogenesis in Gelidiales was documented for the first time in this study. Exclusion of apical chloroplasts from the spermatangial mother cell is accomplished by the previously unreported mechanism of an ingrowing furrow General Discussion and Conclusion / 140 that cuts off the cytoplasm containing the chloroplast. The spermatangium cut off from the spermatangial mother cell is achloroplastic. A spermatangial vacuole develops in the spermatangium from both dictyosome-derived vesicles and vesiculated endoplasmic reticulum. The spermatangial vacuole is released from the spermatangium prior to spermatium discharge. The percurrent production of at least four spermatangia from a single spermatangial mother cell occurs in Gelidium. Ultrastructure of tetrasporogenesis in a member of the Gelidiales was documented here for the first time. The stage of wall development can serve as a marker for developmental events in the tetrasporangium. Prior to meiosis, the nucleus is large, dominating the cytoplasm, and the wall is thickened. Following karyokinesis, a second distinctive wall layer is deposited and chloroplasts and starch grains increase in number. Dictyosomes become more abundant and the number of cisternae increase and cisternae become more inflated. At the second cytokinesis a dark wall layer is deposited around the tetrasporangium and in the cleavage furrow. Starch grains and chloroplasts dominate the cytoplasm. Dictyosomes are strongly curved and young cisternae contain a dark-staining material where thej' are closely appressed. Cored vesicles are produced by the dictyosomes and subsequently release their contents forming a whorled fibrillar layer inside the dark tetrasporangium wall at the second cytokinesis. This study is presented as a contribution to the taxonomy and biology of Gelidiales with a focus on representatives of the order in British Columbia and northern Washington. It is hoped that other workers will find these observations General Discussion and Conclusion / 141 useful in their studies of gelidioid algae on both a local and wider geographic scale. REFERENCES Abbott, LA. 1970. On some new records of marine algae from Washington state. Syesis 3:1-4. Agardh, C.A. 1828. Species Algarum. Vol. 2 Section 1. Greifswald, 189 pp. Agardh, J.G. 1851. Species, Genera et Ordines Algarum, seu Descriptiones Succinctae Specierum, Generum et Ordinum, Quibus Algarum Regnum Constituitur. Vol. 2 Part 1 pp. 1-351. Gleerup, Lund. Agardh, J.G. 1852. Species, Genera et Ordines Algarum, seu Descriptiones Succinctae Specierum, Generum et Ordinum, Quibus Algarum Regnum Constituitur. Vol 2 Part 2 pp.337-351, 352-720 Gleerup, Lund. Agardh, J.G. 1876. Species, Genera et Ordines Algarum, seu Descriptiones Succinctae Specierum, Generum et Ordinum, Quibus Algarum Regnum Constituitur. Vol. 3 Part 1. De Florideis Curae Posteriores. Epicrisis Systematis Floridearum. Leipzig, 724 pp. Agardh, J.G. 1899. Analecta Algologica, Continuatio V. Lunds Universitets Arsskrift, Andra Afdelningen, Kongliga Fysiografiska Saalskapets Handlingar, 35:1-160. Akatsuka, I. 1970. Male organs of Gelidium amansii Lmx., Gelidium pacificum Okam., and Gelidium pusillum (Stackh.) Le Jolis. Bull. Jap. Soc. Phycol. 18:112-115. Akatsuka, I. 1973. Male organs of Pterocladia capillacea (Gelidiaceae. Rhodophyta). J. Jap. Bot. 48:52-54. 142 / 143 Akatsuka, I. 1979. A note on the male reproductive organs of Gelidium japonicum Okam., and Beckerella subcostata (Okam.) Kylin (Gelidiaceae, Rhodophyta). Rev. Algol., N.S. 14:17-20. Akatsuka, I. 1981. Comparative morphology of the outermost cortical cells in the Gelidiaceae (Rhodophyta) of Japan. Nova Hedwigia 35:453-463. Akatsuka, I. 1986a. Surface cell morphology and its relationship to other generic characters in non-parasitic Gelidiaceae (Rhodophyta). Bot. Mar. 29:59-68. Akatsuka, I. 1986b. Japanese Gelidiales (Rhodophyta), especially Gelidium. Oceanogr. Mar. Biol. Ann. Rev. 24:171-263. Akatsuka, I. 1986c. Pterocladiastrum, a new genus segregated from Pterocladia (Gelidiales, Rhodophyta). Bot. Mar. 29:51-58. Akatsuka, I. 1987. Taxonomy and distribution of Beckerella (Gelidiales, Rhodophyta). Aquatic Botany 28:257-274. Akatsuka, I. & Iwamoto, K. 1979. Histochemical localization of agar and cellulose in the tissue of Gelidium pacificum (Gelidiaceae, Rhodophyta). Bot. Mar. 22:367-370. Alley, CD. & Scott, J.L. 1977. Unusual dictyosome morphology and vesicle formation in tetrasporangia of the marine red alga Polysiphonia denudata. J. Ultrastr. Res. 58:289-298. Boillot, A. 1963. Recherches sur le mode de developpement des spores du genre Gelidium (Rhodophycees, Gelidiales). Rev. Gen. Bot. 70:130-137. Broadwater, S., Scott, J. & Pobiner, B. 1986a. Ultrastructure of meiosis in dasya baillouviana (Rhodophyta) II. Prophase. J. Phycol. 22:490-500. / 144 Broadwater, S., Scott, J. & Pobiner, B. 1986b. Ultrastructure of meiosis in Dasya baillouviana (Rhodophyta) II. Prometaphase I - telophase II and post-division nuclear behaviour. J. Phycol. 22:501-512. Chamberlain, A.H.L. & Evans, L.V. 1973. Aspects of spore production in the red alga Ceramium. Protoplasma 76:139-159. Chemin, M.E. 1937. Le developpement des spores chez les Rhodophycees. Rev. Gen. Bot. 49:205-234, 300-327, 353-374, 424-448, 478-536. Chihara, M. & Kamura, S. 1963. On the germination of tetraspores of Gelidiella acerosa. Phycologia 3:69-74. Collins, F.S. 1906. New species, etc., issued in the Phycotheca Boreali-Americana. Rhodora 8:104-113. Collins, F.S. 1913. The marine algae of Vancouver Island. Canada Geological Survey, Victoria Memorial Museum Bulletin 1:99-137. Collins, F.S., Holden, I. & Setchell, W.A. 1903. Phycotheca Boreali-Americana. Facs. 21-23, numbers 1001-1150. Maiden, Mass. Connell, R. 1928. Notes on marine algae collected at Departure Bay, B.C.. Can. Field-Naturalist 42:99-100. Cribb, A.B. 1983. Marine Algae of the Southern Great Barrier Reef—Rhodophyta. Australian Coral Reef Society Handbook No. 2. 173 pp., 71 pis. Crouan, H.M. & Crouan, P.L. 1848. Sur l'organisation la fructification et la classification du Fucus Whiggii de Turner et de Smith, et de VAtractophora hypnoides. Ann. Sc. Nat. Bot. Ser. 3, 10:361-376, pis. 11, 12. / 145 Darbishire, O.V. 1896. Spencerella australis, eine neue Florideen Gattung und Art. Ber. Deutsch. Bot. Ges. 14. Davis, E. & Scott, J. 1986. Ultrastructure of cell division in the marine red alga Lomentaria baileyana. Protoplasma 131:1-10. Dawson, E.Y. 1944. The marine algae of the Gulf of California. Allan Hancock Pacific Exped. 3:189-454. Dawson, E.Y. 1949. Studies on the Northeast Pacific Gracilariaceae. Allan Hancock Found. Pubis., Occ. Papers 7:1-54. Dawson, E.Y. 1953. Marine red algae of Pacific Mexico. Allan Hancock Pacific Exped. 17:1-239. Dawson, E.Y. 1958. Notes on Pacific coast marine algae. VII. Bull. So. Calif. Acad. Sci. 57:65-80. Dawson, E.Y. 1966. New records of marine algae from the Gulf of California. J. Arizona Acad. Sci. 4:55-66. Dawson, E.Y., Acleto, C. & Foldvik, N. 1964. The seaweeds of Peru. Nova Hedwigia Beihefte 13:1-111. De Toni, G.B. 1924. Sylloge Algarum Omnium Hucusque Cognitarum. vol. 6 Florideae, sec. 5. Padua, 767 pp. Dixon, P.S. 1954. Nuclear observations of two British species of Gelidium. Phycol. Bull. 2:4. Dixon, P.S. 1958. The structure and development of the thallus in the British species of Gelidium and Pterocladia. Ann. Bot. 22:353-368. / 146 Dixon, P.S. 1959. The structure and development of the reproductive organs and carposporophyte in two British species of Gelidium. Ann. Bot. 23:397-407. Dixon, P.S. 1961. On the classification of the Florideae with particular reference to the position of the Gelidiaceae. Bot. Mar. 8:1-16. Dixon, P.S. 1963. Terminology and algal life histories, with particular reference to the Rhodophyta. Ann. Bot. N.S. 27:353-355. Dixon, P.S. 1965. Perennation, vegetative propagation and algal life histories, with special reference to Asparagopsis and other Rhodophyta. Botanica Gothoburensia 3:67-74. Dixon, P.S. & Irvine, L.M. 1977a. Seaweeds of the British Isles. Vol. 1. Rhodophyta Part 1. Introduction, Nemalionales, Gigartinales. British Museum (Nat. Hist.), London, 252 pp. Dixon, P.S. & Irvine, L.M. 1977b. Miscellaneous notes on algal taxonomy and nomenclature IV. Bot. Not. 130:137-141. Doty, M.S. 1947. The marine algae of Oregon Part II. Rhodophyta. Farlowia 3:159-215. Drew, K.M. 1954. The organization and inter-relationships of the carposporophytes of living Florideae. Phytomorphology 4:55-69. Endlicher, S.L. 1836. Genera Plantarum Secundum Ordines Naturales Disposita. Vindobonae. Endlicher, S.L. 1843. Genera Plantarum, Suppl. 3. Vindobonae, 56 pp. / 147 Endlicher, S.L. & Diesing, CM. 1845. Algarum natalensium diagnoses. Bot. Zeit. 3:288-290. Engler, A. 1892. Syllabus der Vorlesungen uber specielle und medicinisch-pharmaceutische Botanik. Gebruder Borntraeger, Berlin, 184 pp. Engler, A. & Prantl, K. 1897. Die ivai fir lichen Pflanzenfamilien Part 1(2). Engelmann, Leipzig, 580 pp. Fan, K-C. 1961. Morphological studies of the Gelidiales. Univ. Calif. Publ. Bot. 32:315-368. Fan, K-C. & Papenfuss, G.F. 1959. Red algal parasites occurring on members of the Gelidiales. Madrono 15:33-38. Feldmann, J. & Hamel, G. 1934. Observations sur quelques Gelidiacees. Rev. Gen. Bot. 46:528-549. Feldmann, J. & Hamel, G. 1936. Floridees de France VII. Gelidiales. Revue Algol. 7:209-264. Foreman, R.E. 1977. Benthic community modification and recovery following intensive grazing by Strongylocentrotus droebachiensis. Helgolander Wiss. Meersunters. 30:468-484. Gabrielson, P.W. & Garbar3', D. 1986. Systematics of red algae (Rhodophyta). CRC Critical Reviews in Plant Sciences 3:325-366. Gabrielson, P.W. & Garbary, D.J. 1987. A cladistic analysis of Rhodophyta: Florideophycidean orders. Br. Phycol. J. 22:125-138. / 148 Gaillon, B. 1828. Resume methodique des classifications des Thalassiophytes. Diet. Nat. Sci. 4. Garbary, D.J., Hansen, G.I. & Scagel, R.F. 1980. The marine algae of British Columbia and northern Washington: division Rhodophyta (red algae), class Bangiophyceae. Syesis 13:137-195. Garbary, D.J., Hansen, G.I. & Scagel, R.F. 1982. The marine algae of British Columbia and northern Washington: division Rhodophyta (red algae), class Florideophyceae, orders Acrochaetiales and Nemaliales. Syesis 15(Suppl. 1): 1-102. Garbary, D.J., Hansen, G.I. & Scagel, R.F. 1984. Additions to the marine algae of Barkley Sound, Vancouver Island, British Columbia. Syesis 17:43-45. Gardner, N.L. 1927a. New species of Gelidium of the Pacific coast of North America. Univ. Calif. Publ. Bot. 13:273-318. Gardner, N.L. 1927b. New Rhodophyceae from the Pacific coast of North America. Univ. Calif. Publ. Bot. 13:333-368. Gori, P. 1982. An investigation of tetrasporogenesis in the marine red alga Gastroclonium clavatum (Roth) Ardissone (Rhodymeniales). Ann. Bot. 50:451-457. Grubb, V. M. 1925. The male organs of the Florideae. Linn. J. Bot. 47:177-255. Guiry, M.D. 1978. The importance of sporangia in the classification of the Florideophyceae. In: Modern Approaches to the Taxonomy of Red and Brown Algae. (Ed. by D.E.G. Irvine and J.H. Price) pp. 111-144. Academic Press Inc., London. / 149 Guzman-del Proo, S.A., de la Campa-de Guzman, S. & Pineda-Barrera, J. 1972. Shedding rhythm and germination of spores in Gelidium robustum. Proc. Intl. Seaweed Symp. 7:221-228. Harlin, M.M. 1969. A phycological survey of Steamboat Island, Thurston Countj', Washington State. Syesis 2:257-261. Harvey, W.H. 1833. Algae. In: The Botany of Captain Beechey's voyage-Comprising an Account of the Plants Collected by Messrs Lay and Collie, and Other Officers of the Expedition, During the Voyage to the Pacific and Bering's Strait, performed in His Majesty's Ship Blossom, Under the Command of Captain F.W. Beechey, R.N., F.R., & A.S. in the Years 1825, 26, 27, and 28. Part 4. (W.J. Hooker and G.A.W. Arnott) pp. 163-165. Bohn, London. Harvey, W.H. 1853. Nereis Boreali-Americana. Part II Rhodospermae. Smithsonian Contrib. Knowl. 5(5): 1-258, pis. 13-36. Harve}', W.H. 1855. Short characters of some new genera and species discovered on the coast of Victoria, Australia. Ann. Mag. Nat. Hist. Ser. 2. 15:332-336. Hawkes, M.W. & Scagel, R.F. 1986a. The marine algae of British Columbia and northern Washington: division Rhodophyta (red algae), class Rhodophyceae, order Palmariales. Can. J. Bot. 64:1148-1173. Hawkes, M.W. & Scagel, R.F. 1986b. The marine algae of British Columbia and northern Washington: division Rhodophyta (red algae), class Rhodophyceae, order Rhodymeniales. Can. J. Bot. 64:1549-1580. Hollenberg, G.J. 1942. Phycological notes 1. Bull. Torrey Bot. Club 69:528-538. / 150 Hollenberg, G.J. & Abbott, LA. 1965. New species and new combinations of marine algae from the region of Monterey, California. Can. J. Bot. 43:1177-1188. Holmgren, P.K., Keuken, W. & Schofield, E.K. (Eds.) 1981. Index herbariorum. 1. The herbaria of the world. 7th ed. Regnum Veg. 106:1-452. Hommersand, M.H. 1972. Taxonomic and phytogeographic relationships of warm temperate marine algae occurring in Pacific North America and Japan. Proc. Intl. Seaweed Symp. 7:66-71. Hommersand, M.H. & Fredericq, S. 1988. An investigation of cystocarp development in Gelidium pteridifolium with a revised description of the Gelidiales (Rhodophyta). Phycologia 27:254-272. Howe, M.A. 1914. The marine algae of Peru. Mem. Torrey Bot. Club 15:1-185. Kajimura, M. 1987. Deep-water flora of benthic algae in the Oki Islands, Sea of Japan. Bot. Mar. 30:373-385. Kaneko, T. 1966. Morphological and developmental studies of Gelidiales. 'I. Behaviour of the nucleus in earty stages of tetraspore germination in Gelidium vagum Okamura. Bull. Jap. Soc. Phycol. 14:62-70. Kaneko, T. 1968. Morphological and developmental studies of Gelidiales II. On Acanthopeltis japonica Okamura. Bull. Fac. Fish. Hokkaido University 19:165-172. Katada, M. 1949. Studies on .the propagation of Gelidium (IV). Experimental studies of the germination of Gelidium amansii Lmx. (I). Bull. Jap. Soc. Sci. Fish. 15:359-362. / 151 Katada, M. 1955. Fundamental studies on the propagation of gelidiaceous algae with special reference to shedding and adhesion of the spores, germination, growth and vegetative reproduction. J. Shimonoseki College of Fisheries 5:1-87. Konrad Hawkins, E. 1974a. Golgi vesicles of uncommon morphology and wall formation in the red alga, Polysiphonia. Protoplasma 80:1-14. Konrad Hawkins, E. 1974b. Growth and differentiation of the golgi apparatus in the red alga Callithamnion roseum. J. Cell Sci. 14:633-655. Kraft, G.T. 1976. The morphology of Beckerella scalaramosa, a new species of Gelidiales (Rhodophyta) from the Philippines. Phycologia 15:85-91. Kugrens, P. 1974. Light and electron microscopic studies on the development and liberation of Janczewskia gardneri Setch. spermatia (Rhodophyta). Phycologia 13:295-306. Kugrens, P. 1980. Electron microscopic observations on the differentiation and release of spermatia in the marine red alga Polysiphonia hendryi (Ceramiales, Rhodomelaceae). Amer. J. Bot. 67:519-528. Kugrens, P. & West, J.A. 1972a. Synaptonemal complexes in red algae. J. Phycol. 8:187-191. Kugrens, P. & West, J.A. 1972b. Ultrastructure of spermatial development in the parasitic red algae Levringiella gardneri and Erythrocystis saccata. J. Phycol. 8:331-343. Kugrens, P. & West, J.A. 1972c. Ultrastructure of tetrasporogenesis in the parasitic red alga Levringiella gardneri (Setchell) Kylin. J. Phycol. 8:370-383. Kutzing, / 152 F.T. 1843. Phycologia Generalis Oder Anatomie, Physiologie und Systemkunde der Tange. Brockhaus, Leipzig, 458 pp., 80 pis. Kutzing, F.T. 1847a. Diagnosen einiger neuen auslandischen Algenspecies. Flora 30:773-776. Kutzing, F.T. 1847b. Diagnosen und Bemerkungen zu neuen oder kritischen Algen. Bot. Zeit. 5:1-5, 22-25, 33-38, 52-55, 164-167, 177-180, 193-198, 219-223. Kutzing, F.T. 1849. Species Algarum. Brockhaus, Leipzig, 922 pp. Kutzing, F.T. 1868. Tabulae Phycologicae. Vol. 18. Nordhausen, 35 pp., 100 pis. Kylin, H. 1923. Studien iiber die Entwicklungsgeschichte der Florideen. Svenska Vet. Akad. Handl. 63:1-139. Kylin, H. 1925. The marine red algae in the vicinity of the biological station at Friday Harbor, Wash. Lunds Universitets Arsskrift, N.F. Avd. 2, 24:1-127. Kylin, H. 1928. Entwicklungsgeschichtliche Florideenstudien. Lunds Universitets Arsskrift, N.F. Avd. 2, 24:1-127, 64 figs. Kylin, H. 1935. Remarks on the life-historj' of the Rhodophyceae. Bot. Rev. 1:138-148. Kylin, H. 1937. Anatomie der Rhodophyceen. In: Handbuch der Pflanzenanatomie Vol. 2. (Linsbauer). 347 pp. Kylin, H. 1956. Die Gattungen der Rhodophyceen. Gleerups, Lund, 673 pp. / 153 Lamouroux, J.V. 1813. Essai sur les genres de la famille des thalassiophytes non articulees. Ann. Mus. Nat. Hist. Nat. (Paris) 20:21-47, 115-139, 267-293, pis. 7-13. Lamouroux, J.V. 1819. Delisea In: Diet. Sci. Nat. Vol. 13 pp. 41-42. Levrault, Strasbourg. Lamouroux, J.V. 1825. Gelidium. Dictionnaire Classique d'Histoire Naturelle 7:190-191. Le Jolis, A. 1863. Liste des Algues Marines de Cherbourg. Mem. Soc. Imp. Sci. Nat. Cherbourg 10:5-168, 6 pis. Lindstrom, S.C. & Foreman, R.E. 1978. Seaweed associations of the Flat Top Islands, British Columbia : a comparison of community methods. Syesis 11:171-185. Lindstrom, S.C. & Scagel, R.F. 1987. The marine algae of British Columbia, northern Washington, and southeast Alaska: Division Rhodophyta (red algae), class Rhodophyceae, order Gigartinales, family Dumontiaceae, with an introduction to the order Gigartinales. Can. J. Bot. 65:2202-2232. Loomis, N.H. 1949. New species of Gelidium and Pterocladia with notes on the structures of the thalli in these genera. Allan Hancock Foundation Occasional Papers 6:1-8. Loomis, N.H. 1960. New species of Gelidium and Pterocladia from the Pacific coast of the United States and the Hawaiian Islands. Allan Hancock Foundation Occasional Papers 24:1-10. / 154 Lubchenco, J., Menge, B.A., Garrity, S.D., Lubchenco, P.J., Ashkenas, L.R., Gaines, S.D., Emlet, R., Lucas, J. & Strauss, S. 1984. Structure, persistence, and the role of consumers in a tropical rocky intertidal community (Taboguilla Island, Bay of Panama). J. Exp. Mar. Biol. Ecol. 78:23-73. Macler, B.A. & West, J.A. 1987. Life history and physiology of the red alga, Gelidium coulteri,, in unialgal culture. Aquaculture 61:281-293. Maggs, C.A. & Guiry, M.D. 1987. Gelidiella calcicola sp. nov. (Rhodophyta) from the British Isles and northern France. Br. Phycol J. 22:417-434. Magne, F. 1964. Recherches caryologiques chez les Floridees (Rhodophycees). Cah. Biol. Mar. 5:467-671. McLachlan, J. 1973. Growth media—marine. In: Handbook of Phycological Methods. Culture Methods and Growth Measurements. (Ed. by J.R. Stein) pp. 25-51. Cambridge Univ. Press, Cambridge. Montagne, J.F. 1837. Centurie de plants cellulaires exotiques nouvelles. Ann. Sci. Nat. Bot. Ser. 2 8:345-370. Montagne, J.F. 1847. Phycologie. In: Diet. Universel Hist. Nat. (Orbigny), 10:14-57. Nicolson, D.H. & Norris, J.N. 1983. Ordinal orthography: Nemalionales vs. Nemaliales (Rhodophyta). Taxon 32:288-292. Norris, R.E. 1987. Pterocladia (Gelidiaceae, Rhodophyceae), a genus previously unknown in South Africa, as it occurs in Natal. S. Af. J. Bot. 53:39-43. / 155 Norris, R.E. & Hollenberg, G.J. 1969. Notes on marine algae of Washington and southern British Columbia, IV. Syesis 2:115-119. Norris, R.E. & West, J. 1966. Notes on marine algae of Washington and southern British Columbia. Madrono 18:176-178. Norris, R.E. & Wynne, M.J. 1968. Notes on marine algae of Washington and southern British Columbia, III. Syesis 1:133-146. Okamura, K. 1900. Illustrations of the Marine Algae of Japan. Vol. 1. Tokyo, pp. 1-14, pis. 1-5. Okamura, K. 1934. Gelidium and Pterocladia of Japan. J. Imp. Fish. Inst. 29:47-67, pis. 13-33. Oltmanns, F. 1904. Morphologie und Biologie der Algen. Vol. 1 Gustav Fischer, Jena, 733 pp. Papenfuss, G.F. 1966. A review of the present system of classification of the Florideophycidae. Phycologia 5:247-255. Peel, M.C, Lucas, I.A.N., Duckett, J.G. & Greenwood, A.D. 1973. Studies of sporogenesis in the Rhodophyta I. An association of the nuclei with endoplasmic reticulum in post-meiotic tetraspore mother cells of Corallina officinalis L. Z. Zellforsch. 147:59-74. Peyriere, M. 1969. Infrastructure cytoplasmique de tetrasporocyste de Griffithsia flosculosa (Rhodophycee, Ceramiacee) pendant la prophase meiotique. C. R. Acad. Sci. Paris, Series D 269: 2332-2334. Peyriere, M. 1970. Evolution de l'appariel de de golgi au cours de la tetrasporogenese de Griffithsia flosculosa (Rhodophycee). C. R. Acad. Sci. Paris, Series D 270:2071-2074. / 156 Peyriere, M. 1971. Etude infrastructurale des spermatocystes du Griffithsia flosculosa (Rhodophycee). C. R. Acad. Sci. Paris, Series D 273:2071-2074. Peyriere, M. 1974. Etude infrastructurale des spermatocystes et spermaties de differentes Rhodophycees floridees. C. R. Acad. Sci. Paris, Series D 278:1019-1022. Pueschel, CM. 1979. Ultrastructure of tetrasporogenesis in Palmaria palmata (Rhodophyta). J. Phycol. 15:409-424. Pueschel, CM. 1982. Ultrastructural observations of tetrasporangia and conceptacles in Hildenbrandia (Rhodophyta: Hildenbrandiales). Br. Phycol. J. 17:333-341. Pueschel, CM. & Cole, K.M. 1982. Rhodophycean pit plugs: an ultrastructural survey with taxonomic implications. Amer. J. Bot. 69:703-720. Reinsch, P.F. 1874-1875. Contributiones al Algologiam et Fungologiam. Weigel, Leipzig, 103 pp., 131 pis. Reynolds, E.S. 1963. The use of lead citrate at high pH as an electron opaque stain in electron microscopy. J. Cell Biol. 17:208. Rodriguez, D. & Santelices, B. 1987. Patterns of apical structure in the genera Gelidium and Pterocladia (Gelidiaceae, Rhodophyta). Hydrobiologia 151/152:199-203. Rodriguez, D. & Santelices, B. unpubl. mscr. Separation of Gelidium and Pterocladia on vegetative characters. Ruddell, CL. 1967. Embedding media for 1-2 micron sectioning. 2 hydroxjrmethyl methacrylate combined with 2-butoxyethanol. Stain Techn. 42:253-255. / 157 Santelices, B. 1974. Gelidioid algae, a brief resume of the pertinent literature. Marine Agronomy U.S. Sea Grant Program, Hawaii. Technical Report No. 1. St. John Plant Science Laboratory, University of Hawaii, Honolulu, Hawaii, 111 pp. Santelices, B. 1977. A taxonomic review of Hawaiian Gelidiales (Rhodophyta). Pacific Science 31:61-84. Santelices, B. 1986. The wild harvest and culture of the economically important species of Gelidium in Chile. FAO Fisheries Technical Paper N. 281. pp. 165-192. Santelices, B. unpubl. mscr. Taxonomic studies on Chinese Gelidiales (Rhodophyta). Santelices, B. & Montalva, S. 1983. Taxonomic studies on Gelidiaceae (Rhodophyta) from central Chile. Phycologia 22:185-196. Santelices, B. & Stewart, J.G. 1985. Pacific species of Gelidium Lamouroux and other Gelidiales (Rhodophyta), with keys and descriptions to the common or economically important species. In: Taxonomy of Economic Seaweeds, with reference to some Pacific and Caribbean species. (Ed. by LA. Abbott and J.N. Norris) pp. 17-31. California Sea Grant College Program, University of California, La Jolla. Santisi, S. & De Masi, F. 1981. An electron microscopic study on tetrasporogenesis of the parasitic red alga Erythrocystis montagnei (Derb. and Sol.) Silva. Cytobios 31:163-178. Scagel, R.F. 1957. An Annotated List of the Marine Algae of British Columbia and Northern Washington. Bull. Nat. Mus. Canada 150:1-289. Scagel, R.F. 1967. Guide to the Common Seaweeds of British Columbia. British Columbia Provincial Museum Handbook No. 27, 330 pp. / 158 Scagel, R.F. 1973. Marine benthic plants in the vicinity of Bamfield, Barkley Sound, British Columbia. Syesis 6:127-145. Scagel, R.F., Garbary, D.J., Golden, L. & Hawkes, M.W. 1986. A Synopsis of the Benthic Marine Algae of British Columbia, Northern Washington and Southeast Alaska. Phycological Contribution No. 1, Botany Dept. of the University of British Columbia, Vancouver, 444 pp. Schmitz, F. 1883. Untersuchungen iiber die Befruchtung der Florideen. K. Preuss. Akad. Wiss. Berlin 10:215-258. Schmitz, F. 1889. Systematische iibersicht der bisher bekannten gattungen der florideen. Flora 72:435-456. Schmitz, F. & Reinke, J. 1889. Algenflora der westlichen Ostsee deutschen Antheils. Eine systematisch-pflanzengeographische studie. Ber. Kommiss. Wiss. Untersuch. Deutsch. Meere Kiel 6:1-101. Schmitz, F. & Hauptfleisch, P. 1897. Rhodophyceae. In Engler, A. & Prantl, K. Die Naturlichen Pflanzenfamilien. . Vol. 1, Engelmann, Leipzig, p. 298-544. Schnetter, R. & Bula Meyer, G. 1982. Marine Algen der Pazifikk'uste von Kolumbien: Chlorophyceae, Phaeophyceae, Rhodophyceae; Algas marinas del literal pacifico de Colombia: Chlorophyceae, Phaeophyceae, Rhodophyceae. Bibliotheca Phycologica 60:1-287. Scott, J., Bosco, C, Schornstein, K. & Thomas, J. 1980. Ultrastructure of cell division and reproductive differentiation of male plants in the Florideophyceae (Rhodophyta): cell division in Polysiphonia. J. Phycol. 16:507-524. Scott, J.L. & Dixon, P.S. 1973a. Ultrastructure of spermatium liberation in the marine red alga Ptilota densa. J. Phycol. 9:85-91. / 159 Scott, J.L. & Dixon, P.S. 1973b. Ultrastructure of tetrasporogenesis in the marine red algae Ptilota hypnoides. J. Phycol. 9:29-46. Scott, J.L. & Thomas, J.P. 1975. Electron microscope observations of telophase II in the Florideophyceae. J. Phycol. 11:474-476. Segawa, S. 1959. Coloured Illustrations of the Seaweeds of Japan. Hoikusha, Osaka, 175 pp. Setchell, W.A. & Gardner, N.L. 1903. Algae of northwestern America. Univ. Cal. Publ. Bot. 1:165-418. Setchell, W.A. & Gardner, N.L. 1924. Expedition of the California Academy of Sciences to the Gulf of California in 1921. The marine algae. Proc. Calif. Acad. Sci., Fourth Series 12:695-949. Silva, P.C. 1978. Type specimens of Gelidiaceae (rhodophyceae) described by Nina H. Loomis. Phycologia 17:257-261. Silva, P.C. 1980. Names of classes and families of living algae. Reg. Vegetabile #103. Bohn, Scheltema & Holkema, Utrecht. 156 pp. Simon-Bichard-Breaud, J. 1971. Un appareil cinetique dans les gametocystes males d'une Rhodophycee : Bonnemaisonia hamifera Hariot. C. R. Acad. Sci. Paris, Series D 273:1272-1275. Simon-Bichard-Breaud, J. 1972a. Origine et devenir des vacuoles a polysaccharides des gametocystes males de Bonnemaisonia hamifera Hariot (Rhodophycee). C. R. Acad. Sci. Paris, Series D 274:1485-1488. Simon-Bichard-Breaud, J. 1972b. Formation de a crypte flagellaire et evolution de son contenu au cours de la gametogenese male chez Bonnemaisonia hamifera Hariot (Rhodophycee). C. R. Acad. Sci. Paris, Series D 274:1796-1799. / 160 Smith, G.M. 1944. Marine Algae of the Monterey Peninsula, California. Stanford University Press, Stanford, California, 622 pp. South, G.R. 1968. Intertidal marine algae from Gabriola Island, British Columbia. Syesis 1:177-186. Spurr, A.R. 1969. A low-viscosity epoxy-resin embedding medium for electron microscopy. J. Ultrastr. Res. 26:31-43. Sreenivasa Rao. P. 1971. Studies on Gelidiella acerosa (Forsskal) Feldmann et Hamel. V. germination of tetraspores and nuclear changes of the germinating spores. Bull. Jap. Soc. Phycol. 19:65-72. Sreenivasa Rao, P. & Trivedi, M.K. 1986. Reproduction in Gelidiella. In: Taxonomy of Algae. Papers Presented at the International Symposium on Taxonomy of Algae held at the Centre of Advanced Botany, University of Madras, 9-16 December, 1974. (Ed. by T.K. Desikachary and V.N. Raja Rao) pp. 255-260. University Madras, Madras. Stackhouse, J., 1801. Nereis Britannica. Fascicle 3. S. Hazard, Bath, pp. 71-112. Stephenson, T.A. & Stephenson, A. 1961a. Life between tide-marks in North America IVa. Vancouver Island, I. J. Ecol. 49:1-29. Stephenson, T.A. & Stephenson, A. 1961b. Life between tide-marks in North America IVb. Vancouver Island, II. J. Ecol. 49:227-243. Stewart, J.G. 1968. Morphological variation in Pterocladia pyramidale. J. Phycol: 4:76-84. Stewart, J.G. 1974. Systematics of Pterocladia media from California. Bull. S. Calif. Acad. Sci. 73:105-108. / 161 Stewart, J.G. 1976. Gelidiaceae. In: Marine Algae of California. (LA. Abbott and G.J. Hollenberg) pp. 340-352. Stanford Univ. Press, Stanford. Stewart, J.G. & Norris, J.N. 1981. Gelidiaceae (Rhodophyta) from the northern Gulf of California, Mexico. Phycologia 20:273-284. Sturch, H.H. 1926. Choreocolax polysiphoniae Reinsch. Ann. Bot. 15:585-605. Taylor, W.R. 1945. Pacific marine algae of the Allan Hancock Expeditions to the Galapagos Islands. Allan Hancock Pacific Expeditions 12:1-528. Tazawa, N. 1975. A study of the male reproductive organ of the Florideae from Japan and its vicinity. Sci. Pap. Inst. Algol. Res., Fac. Sci. Hokkaido Univ. 6:95-179. Tokida, J. 1954. The marine algae of southern Saghalien. Mem. Fac. Fish. Hokkaido Univ. 2:1-264. Tripodi, G. & De Masi F. 1975. Cytological localization of polysaccharidic molecules in some red algae. J. Submicr. Cytol. 7:197-209. Turner, D. 1815-1819. Fuci. Vol. 4. London, 157 pp., pis. 197-258. Vesk, M. & Borowitzka, M.A. 1984. Ultrastructure of tetrasporogenesis in the coralline alga Haliptilon cuvieri (Rhodophyta). J. Phycol. 20:501-515. Waaland, J.R. 1973. New records of red algae from Washington and British Columbia. Syesis 6:269. Weber van Bosse, A. 1921. Liste des algues du Siboga. II. Rhodophyceae. Part 1. Protoflorideae, Nemalionales, Cryptonemiales. Siboga-Exped. Monographic 59B: 187-310. / 162 Widdowson, T.B. 1974. The marine algae of British Columbia and northern Washington: revised list and keys. Part II. Rhodophyceae (red algae). Syesis 7:143-186. Wittmann, W. 1965. Aceto-iron-haematoxylin-chloral hydrate for chromosome staining. Stain Tech. 40:161-164. Xia, B.M. & Abbott, LA. 1985. The genus Polycavernosa Chang et Xia (Gracilariaceae, Rhodophyta): a comparison with Gracilaria Grew, and a key to the species. In: Taxonomy of Economic Seaweeds with Reference to Some Pacific and Caribbean Species, (Ed. by LA. Abbott and J.N. Norris), pp. 157-162. California Sea Grant Program, University of California, La Jolla. Xia, B.M., Engzhan, X. & Junfu, Z. 1983. Gelidiaceae. In: Common Seaweeds of China. (Ed. by C.K. Tseng) pp. 66-68. Science Press, Beijing. Yamamoto, H. 1975. The relationship between Gracilariopsis and Gracilaria from Japan. Bull. Fac. Fish. Hokkaido Univ. 26:217-222. Yamasaki, H. 1960. Studies on the propagation of gelidiaceous algae VI. On the early development and morphogeny in Gelidium amansii Lmx. Bull. Jap. Soc. Sci. Fish. 26:116-122. Yatabe, B. 1892. Iconographia florae japonicae. Vol. 1, Tokyo. FIGURES Abbreviations Used in Figures. AP apical cell C carposporophyte CL chloroplast COR cortex CPG carpogonium CS carposporangium D dictyosome ER endoplasmic reticulum FC fusion cell G gonimoblast LO locule N nucleus NC nutritive chain NO nucleolus OS ostiole P attachment pad PA periaxial cell PC pit connection PL second order plate PR prostrate axis R rhizine SMC spermatangial mother cell 163 / 164 ST starch grain SV spermatangial vacuole UP upright axis W wall 1° cell of first order filament (axial file) 2° cell of second order filament / 165 Fig. 1. Map of study area and area of enlargement maps in Figs. 2, 3, 4. 166 / 167 Fig. 2. Barkley Sound collection sites, numbered as follows: 1. Brady's Beach, 2. Diplock Is., 3. Dixon Is., 4. Dodger Channel, 5. Geer Islets, 6. Haines Is., 7. Kirby Pt., 8. Meade Islets, 9. Roquefeuil Bay, 10. Ross Islets. 168 / 169 Fig. 3. Southern Vancouver Island, Gulf Islands and San Juan Islands collection sites, numbered as follows: 11. Madrona Pt., 12. Georgina Pt., 13. False Narrows, 14. Orlebar Pt., 15. Whalebone Bay. 170 / 171 Fig. 4. Strait of Georgia collection sites, numbered as follows: 16. Earls Cove, 17. Finn Cove, 18. Frolander Bay, 19. Halfmoon Bay, 20. Denman Is., 21. Ford Cove, 22. Galleon Pt., 23. Helliwell Park, 24. Tribune Bay. 1.72 / 173 Fig. 5. Key to Gelidiales in British Columbia and northern Washington. la. Plant dark red; mostly of branched, flattened, creeping axes, frequently forming attachment pads at branch nodes; uprights present or absent, never more than 2 cm tall, sparse branches approximately 90° to axis; tetrasporangia in V's... Pterocladia caloglossoides lb. Plant red to yellowish-red, dark red to maroon or brown to brownish-red; creeping axes branched, terete, with attachment pads usually not formed at branch nodes; uprights dominant, usuallj' more than 2 cm tall, well branched, branches 30-60° to axis; tetrasporangia not in V's 2 2a. Plant intertidal, brown to brownish-red; forms a turf; usually two orders of branches and many short second order branches; gametophytes not known from British Columbia or northern Washington Gelidium coulteri / 2b. Plant shallow subtidal (0-13 m); red to yellowish-red or dark red to maroon; does not form a turf; branching to three to four orders with abundant second, third and fourth order branches; gametophytes rare or common in British Columbia and northern Washington 3 3a. Plant robust, cartilaginous; portions of plant with pyramidal aspect; axes terete to compressed; tetrasporangial stichidia with narrow sterile margin; dioecious; third order filaments of mature cystocarp (between second order plate and pericarp) intact Gelidium purpurascens 3b. Plant soft and lax; not triangular in aspect; axes compressed to flattened; tetrasporangial stichidia lack sterile margin; monoecious; lacks intact third order filaments in mature cystocarp Gelidium vagum / 175 Figs. 6-11. Gelidium coulteri vegetative morphology. Fig. 6. habit of infertile plant, scale bar = 5 mm; Fig. 7. Apex, apical cell domed and protruding, whole mount, scale bar = 20 um; Fig. 8. Attachment pads from prostrate axis, whole mount, scale bar = 50 um; Fig. 9. Vegetative upright axis showing cortex, medulla and rhizines (arrowhead), transverse section, scale bar = 50 nm; Fig. 10. Vegetative, upright axis, secondary pit-connections (arrows) between inner cortical and medullary cells, rhizines, longitudinal section, scale bar = 20 um; Fig. 11. Outer cortical cells unoriented, isodiametric, whole mount, scale bar = 20 Mm. Figs. 6, 11 Kirby Pt., Diana Is.; Fig. 7 east side Denman Is.; Figs. 8-10 Haines Is. \ 1 76 / 177 Figs. 12-17. Gelidium coulteri tetrasporophytes. Fig. 12. Habit with tetrasporangial stichidia (arrows), scale bar = 5 mm; Fig. 13. Branched tetrasporangial stichidium, whole mount, scale bar = 300 M m ; Fig. 14. Tetrasporangial stichidium with embedded tetrasporangia (arrows), whole mount, scale bar = 50 M m ; Fig. 15. Undivided tetrasporangium with lateral pit connection (arrowhead), longitudinal section, scale bar = 20 M m ; Fig. 16. Various ages of tetrasporangia, longitudinal section, scale bar = 50 M m ; Fig. 17. Mature, cruciately divided tetrasporangium, transverse section, scale bar = 20 M m ; Fig. 12 Kirby Pt., Diana Is.; Fig. 13 Meade Islets; Figs. 14-17 Galleon Pt., Hornby Is. / 179 Figs. 18-20. Gelidium coulteri male gametophytes. Fig. 18. Spermatangial (arrow) and vegetative parts of branch, whole mount, scale bar = 20 urn; Fig. 19. Spermatangial mother cells (large arrows), spermatangia (small arrows) and spermatia (arrowheads), longitudinal section, scale bar=10 um; Fig. 20. Spermatangial mother cells (arrow) with periclinal wall cutting off spermatangium (arrowhead), transverse section, scale bar =10 um. Figs. 18-20 Plants cultured from tetraspores, source: east side Denman Is. 180 / 181 Figs. 21-24. Gelidium coulteri habit. Fig. 21. Herbarium specimens UC 494898, tetrasporophyte, Gardner's collection (#7179) from same tidepool as G. sinicola holotype, Point Cavalo, Marin Co., California, scale bar = 20 um; Fig. 22. Isotype in TCD, tetrasporophyte, scale bar = 20 mm; Fig. 23. Isotype, AHFH 53929, scale bar=10 mm; Fig. 24. Reproduction of photograph of lectotype in TCD, scale bar=10 mm. 182 / 183 Figs. 25-29. Gelidium purpurascens habit and vegetative morphology. Fig. 25. Habit of part of tetrasporophyte, scale bar=10 mm; Fig. 26. Vegetative branch showing cortex, medulla and rhizines (arrows), transverse section, scale bar = 50 Mm; Fig. 27. Habit of basal part of tetrasporophyte showing prostrate axes and tetrasporangial stichidia (arrows), scale bar =10 mm; Fig. 28. Apex of female gametophyte branch, apical cell domed, longitudinal section, scale bar =10 Mm; Fig. 29. Vegetative branch, rhizines (arrows) in inner cortex/outer medulla transverse section, scale bar = 20 Mm. Figs. 25, 27, 29 Meade Islets, Fig. 26 Orlebar Pt., Gabriola Is., Fig. 28 Kirby Pt., Diana Is. 184 / 185 Figs. 30-34. Gelidium purpurascens vegetative morphology. Fig. 30. Vegetative branch with cortex, medulla and rhizines (arrows), longitudinal section, scale bar = 20 um; Fig. 31. Surface cortical cells isodiametric, unoriented, equidistant, whole mount, scale bar = 20 um; Fig. 32. Medullary cells of second order plate interconnected by secondary pit connections (arrows), squashed longitudinal section, scale bar = 30 um; Fig. 33. Attachment pad on prostrate axis, whole mount, scale bar =100 um; Fig. 34. Attachment pad formed from elongate cortical cells, longitudinal section of prostrate axis, scale bar = 50 um. Fig. 30 Orlebar Pt., Gabriola Is., Figs. 31, 32 east side Denman Is., Figs. 33, 34 Whalebone Bay, Gabriola Is. 1 86 / 187 Figs. 35-40. Ultrastructure of rhizines in Gelidium purpurascens and G. vagum. Figs. 35, 36, 39, 40. Gelidium purpurascens. Fig. 35. Young rhizines with convoluted plasma membrane, large vesicles with fibrillar contents (arrows), thick wall, transverse section, scale bar = 500 nm; Fig. 36. Old rhizine with smooth plasma membrane profile, cytoplasm vesiculate, membranes thickened, very thick wall, transverse section, scale bar = 500 nm; Figs. 37, 38. Gelidium vagum. Fig. 37. Young rhizine with vesicles containing vesicular and tubular contents (arrows), some tubular bodies in wall outside indented plasma membrane (arrowheads), longitudinal section, scale bar = 200 nm; Fig. 38. Part of young rhizine with nucleus and chloroplast (arrow) with inner encircling thylakoid but lacking inner parallel thylakoids, longitudinal section, scale bar = 500 nm; Fig. 39. Young rhizine with dictyosome (arrow) producing vesicles with fibrillar contents and releasing contents to wall (at arrowheads), elongate mitochondrion (M), longitudinal section, scale bar = 400 nm; Fig. 40. Old rhizine, cytoplasm vesiculate, membranes thickened, plasma membrane profile smooth, longitudinal section, scale bar = 400 nm; Figs. 35, 39 Orlebar Pt., Gabriola Is., Figs. 36, 40 culture source: Whalebone Bay, Gabriola Is., Figs. 37, 38 culture source: Tribune Bay, Hornby Is. 188 / 189 Figs. 41-44. Gelidium purpurascens tetrasporophytes. Fig. 41. Tetrasporangial stichidia, habit, scale bar = 500 M m ; Fig. 42. Tetrasporangial stichidium with narrow sterile margin (arrow), whole mount, scale bar = 50 M m ; Fig. 43. Stichidial branch apex, young tetrasporangia with basal pit connections (arrowheads) and maturing tetrasporangia with one periclinal division (arrows), longitudinal section, scale bar = 50 M m ; Fig. 44. Cruciately divided tetrasporangium, transverse section, scale bar = 20 M m . Figs. 41, 44 Orlebar Pt., Gabriola Is., Fig. 42 Meade Islets, Fig. 43 Cape Suspiro, Alaska. 1 90 / 191 Figs. 45-48. Gelidium purpurascens female gametophytes. Fig. 45. Carpogonial stichidium, habit, scale bar =100 um; Fig. 46. Intercalary carpogonium, with two pit connections (arrowheads), squashed longitudinal section, scale bar = 5 um; Fig. 47. Apex with intercalary carpogonia (all arrows), most proximal carpogonia sessile (large arrows), longitudinal section, scale bar = 400 Mm; Fig. 48. Cystocarps single and in series on branch, habit, scale bar = 2 mm. Fig. 45 Whalebone Bay, Gabriola Is., Fig. 46 Geer Islets, Figs. 47, 48 Kirby Pt., Diana Is. / 193 Figs. 49-51. Gelidium purpurascens carposporophytes. Fig. 49. Stretched third order filament between second order plate and cystocarp cortex, longitudinal section, scale bar =10 um; Fig. 50. Mature cystocarp with two locules, carposporophyte concentrated around second order plate and producing carposporangia, longitudinal section through ostiole of upper locule, scale bar = 50 um; Fig. 51. Detail of carposporophyte, young carposporangia with mucilage coating (arrows), longitudinal section, scale bar = 25 um- Figs. 49, 51 Geer Islets, Fig. 50 Kirby Pt., Diana Is. 1 94 / 195 Figs. 52-56. Gelidium purpurascens male gametophytes. Fig. 52. Habit, whole mount, scale bar = 5 mm; Fig. 53. Spermatia, whole mount, scale bar = 20 jum; Fig. 54. Spermatangial stichidium with thickened and ruptured thallus wall over spermatangia, whole mount, scale bar=100 um; Fig. 55. Holes in wall (arrowheads) through which spermatia have been released, whole mount, scale bar =10 iim; Fig. 56. Spermatangial mother cells (large arrows) and spermatangia (small arrows) cut off by periclinal wall, longitudinal section, scale bar =10 um. Fig. 52 Nootka Sound UBC A53657, Figs. 53, 54, 56 cultured from tetraspores, source: Whalebone Bay, Gabriola Is., Fig. 55 Esteban Pt., Vancouver Is. UBC A10813. 196 / 197 Figs. 57-63. Gelidium purpurascens apical and lateral initials, historical and type collections. Fig. 57. Apical cell domed, protruding, whole mount, scale bar =15 um; Fig. 58. Lateral initial conical (arrow), whole mount, scale bar =15 um; Fig. 59. Macoun collection from Departure Bay, B.C., CANA 3740 (was CAN 74), scale bar=15 um; Fig. 60. Isodiametric, equidistant, unoriented, surface cortical cells near base of upright axis, cells branch axis runs left/right in figure, whole mount, scale bar = 50 um; Fig. 61. Patch of surface cortical cells near base of upright axis where cells are elliptical and oriented perpendicular to branch axis, branch axis runs left/right in figure, whole mount, scale bar = 40 um; Fig. 62. Holotype with herbarium labels, UC 93572, cystocarpic, from Moss Beach, San Mateo Co., California, scale bar = 40 um; Fig. 63. One holotype plant, UC 93572, cystocarpic, from Moss Beach, San Mateo Co., California, scale bar = 20 Um. Figs. 57, 58, 60, 61 Orlebar Pt., Gabriola Is. 198 / 199 Figs. 64-68. Gelidium vagum habit and vegetative morphology. Fig. 64. Habit of gametophyte, scale bar=10 mm; Fig. 65. Habit of tetrasporophyte, scale bar=10 mm; Fig. 66. Habit of gametophyte with whip-like ultimate branches, scale bar =10 mm; Fig. 67. Vegetative branch with cortex, medulla and rhizines (arrows), transverse section, scale bar = 25 um; Fig. 68. Apical organization in vegetative branch, apical cell cuts off cells of axial file/first order filament, periaxial cells cut off second order filament, rhizines (arrow) cut off inner cortical cells, longitudinal section in plane of second order plate, scale bar = 50 um. Figs. 64-68 Tribune Bay, Hornby Is. 200 / 201 Figs. 69-76. Gelidium vagum vegetative and tetrasporophyte morphology. Fig. 69. Surface cortical cells isodiametric, equidistant, unoriented, from base of upright axis, whole mount, scale bar = 50 um; Fig. 70. Surface cortical cells isodiametric, equidistant, unoriented, from upright axis midway between base and apex, whole mount, scale bar = 50 um; Fig. 71. Vegetative branch with cortex, medulla and rhizines (arrows) transverse section, scale bar = 30 ym; Fig. 72. Vegetative branch showing cortex and medulla (M), longitudinal section, scale bar = 30 urn; Fig. 73. Rhizine (arrow) cut off inner cortical cell, longitudinal section, scale bar =10 um; Fig. 74. Tetrasporangial stichidium with two fertile areas, whole mount, scale bar =100 um; Fig. 75. Tetrasporangial stichidium, lacks sterile margin, whole mount, scale bar = 30 um; Fig 76. Habit of tetrasporophyte with stichidia (arrow), scale bar = 5 mm. Figs. 69, 72, 73 Helliwell Park, Hornby Is., Fig. 70 Ford Cove, Hornby Is., Figs. 71, 75, 76 Tribune Bay, Hornby Is., Fig. 74 Galleon Pt., Hornby Is. 202 / 203 Figs. 77-83. Gelidium vagum reproductive morphology. Fig. 77. Stichidium apex with young tetrasporangia (arrows), longitudinal section, scale bar = 30 um; Fig. 78. Mature tetrasporangium, cruciately divided, tetraspore nuclei visible, squash preparation, scale bar = 20 um; Fig. 79. Young tetrasporangium with basal pit connection (arrowhead), longitudinal section, scale bar = 20 um; Fig. 80. Holes in thallus wall (arrows) where tetraspores have been released, whole mount, scale bar = 20 um; Fig. 81. Undivided tetrasporangium with lateral pit connection (arrowhead) and mature divided tetrasporangia, transverse section, scale bar = 30 um; Fig. 82. Gametangial stichidia, whole mount, scale bar=150 Mm; Fig. 83. Spaces vacated (arrows) by released tetrasporangia, filled with weakly staining substance, discarded pit plug in lateral position (arrowhead), longitudinal section, scale bar = 20 Mm. Figs. 77, 81, 82 Galleon Pt., Hornby Is., Figs. 78, 79, 80 Tribune Bay, Hornby Is., Fig. 83 Ford Cove, Hornby Is. 2 0 4 / 205 Figs. 84-89. Gelidium vagum gametophytes. Fig. 84. Habit of cystocarpic plant, scale bar = 4 mm; Fig. 85. Gametangial area with male gametes (arrowheads) proximal and young carposporophyte derived from fertilized female gametes distal (arrow), longitudinal section, scale bar = 30 um; Fig. 86. Spermatangium cut off spermatangial mother cell by periclinal wall (arrow), transverse section, scale bar =10 um; Fig. 87. Release of spermatium (arrow) from spermatangium, transverse section, scale bar =10 um; Fig. 88. Part of young carposporophyte with nutritive chains (arrows), longitudinal section, scale bar =15 um; Fig. 89. Young carposporophyte with furrows created by replacement of cortical cells by carpogonia (arrowheads), through which carpogonial trichogynes protrude, transverse section, scale bar = 40 um. Figs. 84-89 Tribune Bay, Hornby Is. 206 / 207 Figs. 90-95. Gelidium vagum reproductive morphology. Fig. 90. Intercalary carpogonium, lower pit connection out of focal plane, longitudinal section, scale bar=10 Mm; Fig. 91. Third order filaments (arrowheads) stretched between second order plate and pericarp in young, expanding cystocarp, longitudinal section, scale bar=15 Mm; Fig. 92. Fusion cell with two narrow lobes (arrows), longitudinal section, scale bar = 75 Mm; Fig. 93. Mature cystocarp, stretched third order filaments absent, longitudinal section, scale bar =100 Mm; Fig. 94. Habit, tetrasporophyte from Muroran, Hokkaido, Japan, UBC A56807, scale bar=10 mm; Fig. 95. Habit, tetrasporophyte from Hideshima, Iwate-ken, Japan, TNS 25823, scale bar=10 mm. Figs. 90-92 Tribune Bay, Hornby Is., Fig. 87 east side Denman Is. / 209 Figs. 96-100. Gelidium vagum Japanese material. Fig. 96. Mature cystocarp and adjacent male gametangial area (arrows, to right), longitudinal section, scale bar = 50 um; Fig. 97. Mature cystocarp, stretched third order filaments absent from central part of cystocarp but occasionally present (arrow) at less expanded ends, longitudinal section, scale bar = 50 um; Fig. 98. Tetrasporangial stichidium, sterile margin lacking, whole mount, scale bar = 50 p.m; Fig. 99. Immature (arrowheads) and mature (arrow) tetrasporangia, longitudinal section, scale bar = 20 um; Fig. 100. Tetrasporangial stichidium, note embedding of sporangia under two rows of cortical cells, transverse section, scale bar = 50 um. Figs. 96-98 Ta-no-hama, Iwate-ken, Japan, TNS 25824, Figs. 99, 100 Muroran, Hokkaido, Japan, UBC A56807. 210 / 211 Figs. 101-106. Gelidium johnstonii habit, vegetative and reproductive morphology. Fig. 101. Habit of holotype, CAS 1343 in UC, tetrasporophytes, Bahia San Francisquito, Gulf of California, Baja California del Norte, Mexico, scale bar = 40 mm; Fig. 102. Vegetative branch with cortical cells in anticlinal rows, medulla and rhizines (arrowheads), transverse section, scale bar = 50 nm; Fig. 103. Cortical cells in anticlinal rows, rhizines between inner cortical and medullary cells, transverse section, scale bar =100 jim; Fig. 104. Tetrasporangial stichidium with sterile margin, whole mount, scale bar = 20 nm; Fig. 105. Apical cell and furrow (arrowheads) of tetrasporophyte branch, whole mount, scale bar = 30 um; Fig. 106. Habit of tetrasporophyte with stichidia, scale bar=10 mm. Figs. 104, 105 Puerto Escondido, Gulf of California, Baja California del Sur, Mexico, Fig. 106 Punta Perico, Gulf of California, Mexico. 212 / 213 Figs. 107-109, 111. Gelidium johnstonii reproductive morphology. Fig. 110. G. vagum habit. Fig. 107. Mature cystocarp, longitudinal section, scale bar = 50 um; Fig. 108. Stretched third order filament in cystocarp between second order plate and pericarp, longitudinal section, scale bar = 40 iim; Fig. 109. Apex of gametangial stichidium, note apical furrow and apical cell slightly recessed between cortical lobes, whole mount, scale bar = 50 um; Fig. 110. Gelidium vagum habit of drift plant from Ladysmith, B.C., UBC A64965, scale bar = 40 um; Fig. 111. Mature cystocarp (top) with ostiole (arrow) and adjacent proximal (bottom) spermatangial area (arrowheads), whole mount, scale bar=100 um. Fig. 107 Puerto Escondido, Gulf of California Baja California del Sur, Mexico, AHFH 50299, Fig. 108 Punta Escondido, Gulf of California, Baja California del Sur, Mexico, Dawson #7170, Fig. 109 Ensenada Bocochibampo, Sinaloa, Mexico, AHFH 4179, Fig. I l l Bahia Aqua Verde, Gulf of California, Baja California del Sur, Mexico, AHFH 502677. 214 / 215 Figs. 112-118. Pterocladia caloglossoides habit and vegetative morphology. Fig. 112. Habit, scale bar=10 mm; Fig. 113. Habit of upright axes, branches perpendicular to main axis, tetrasporangial, scale bar =150 um; Fig. 114. Habit of prostrate axes, several upright branches and attachment pad at a node, scale bar = 700 um; Fig. 115. Apex of vegetative branch, apical initial and lateral initial (upper left corner) domed and protruding, files of cells in V's, whole mount, scale bar=100 um; Fig. 116. Bottom view of flared attachment pad on prostrate axis and four radiating upright branches, whole mount, scale bar =100 um; Fig. 117. Upright axis with cortex, medulla and rhizines (arrows), transverse section, scale bar = 30 um; Fig. 118. Vegetative axis with narrow cortex, medulla and rhizines (arrows), transverse section, scale bar = 30 um. Fig. 112 Sea Otter Sound, Alaska, Fig. 113 Orlebar Pt., Gabriola Is., Fig. 114 Barra de Navidad, Jalisco, Mexico, Dawson #3733 slide #1327 (in AHFH) Figs. 115-118 Georgina Pt., Mayne Is. 21 6 / 217 Figs. 119-121. Pterocladia caloglossoides vegetative morphology. Fig. 119. Surface cortical cells oval, equidistant, unoriented, whole mount, scale bar = 50 um; Fig. 120. Attachment pad on prostrate axis, composed of elongated cortical cells, whole mount, scale bar = 30 Mm; Fig. 121. Attachment pad on prostrate axis composed of elongated cortical cells, longitudinal section, scale bar = 40 Mm. Figs. 119-121 Georgina Pt., Mayne Is. 21 8 / 219 Figs. 122-128. Pterocladia caloglossoides reproductive morphology. Fig. 122. Tetrasporangial stichidium, tetrasporangia in Vs, whole mount, scale bar =100 M m ; Fig. 123. Tetrasporangial stichidium, transverse section, scale bar = 40 M m ; Fig. 124. Tetrasporangium with lateral pit connection, squash preparation, scale bar = 20 am; Fig. 125. Mature, cruciately divided tetrasporangium, squash preparation, scale bar = 20 M m ; Fig. 126. Tetrasporangia in Vs in Dawson's material (Dawson #8593 slide 1177) from Guadalupe Is., Baja California del Norte, Mexico, whole mount, scale bar=100 M m ; Fig. 127. Cystocarp in Dawson's material (Dawson #8593 slide 1177) from Guadalupe Is., Baja California del Norte, Mexico, whole mount, scale bar = 40 M m ; Fig. 128. Cystocarp with single locule, Dawson's material from Barra de Navidad, Jalisco, Mexico, Dawson #3733 (wet preserved in AHFH), longitudinal section, scale bar =30 M m . Figs. 122-125 Georgina Pt., Mayne Is. 220 / 221 Figs. 129-133. Pterocladia caloglossoides reproductive morphology. Fig. 134. Gelidium coulteri. Fig. 129. Cystocarp with single ostiole (arrowhead), Dawson #8593 slide 1177, whole mount, scale bar=100 urn; Fig. 130. Spermatangial area, Dawson #3733 slide 1325, whole mount, scale bar = 20 ym; Fig. 131. Spermatangia, Dawson #3733 slide 1325, whole mount, scale bar = 40 um; Fig. 132. Vegetative branch with narrow medulla, Dawson #3733 slide 1326, transverse section, scale bar = 20 ym; Fig. 133. Spermatangia (arrow), Dawson #3733 slide 1325, optical section of whole mount, scale bar = 30 ym; Fig. 134. Habit of Macoun's collection of "Gelidium crinale" from Beacon Hill, Vicotria CANA 3474 (was CAN 207), scale bar=10 mm. Fig. 129 Guadalupe Is., Baja California del Norte, Figs. 130-133 Barra de Navidad, Jalisco, Mexico. 2 2 2 / 223 Figs. 135-140. Historical collections and type specimens. Fig. 135. Gelidium purpurascens, Macoun's collection as "G. amansii" from Departure Bay, CANA 4349 (was CAN 310), scale bar = 20 mm; Fig. 136. G. robustum, collection from "Shoal B. Victoria", UBC A7861, scale bar = 50 mm; Fig. 137. G. coulteri, Macoun's collection as "G. crinale" from Beacon Hill, Victoria, CANA 3474 (was CAN 207), scale bar = 50 mm; Fig. 138. G. robustum holotype, UC 294572, near Ensenada, Baja California del Norte, Mexico, scale bar = 50 mm; Fig. 139. G. sinicola holotype, UC 276620, Point Cavalo, Marin Co., California, scale bar=50 mm; Fig. 140. G. sinicola holotype (UC 276620), vegetative branch, terete, transverse section, scale bar = 30 mm. 224 / 225 Fig. 141. Polysiphonia-type life history. Triphasic (with tetrasporophyte, gametophyte and carposporophyte phases) life history, isomorphic tetrasporophyte and gametophyte, carposporophyte developing in female gametophyte axis, magnified cystocarp shown in longitudinal section; solid (black) structures diploid, hollow (white) structures haploid; not drawn to scale. 226 m a l e g a m e t o p h y t e N / 227 Figs. 142-153. Gelidium vagum sporeling development in culture. Fig. 142. Ungerminated tetraspore; Fig. 143. Germ tube (arrow) initiation, gametophyte; Fig. 144. Germ tube enlargement, gametophyte; Figs. 145, 146. Evacuation of cytoplasm from spore into germ tube and wall cutting off germ tube, tetrasporophytes; Fig. 147. First unequal division of germ tube into concave (larger) and fusiform (smaller) cells, tetrasporophytes; Fig. 148. Division of germ tube into concave (right) and fusiform (left) cells, gametophyte; Fig. 149. Transversely divided concave cell, most distal cell of group initiating primary rhizoid, tetrasporophyte; Fig. 150. Variety of developmental stages, original spores all from same cystocarp, tetrasporophytes; Fig. 151. Elongation of primary attachment rhizoid, originated from cell in concave group, tetrasporophyte; Fig. 152. Elongated rhizoid, gametophyte; Fig. 153. Second attachment rhizoid produced (before apical cell initiation), gametophyte. Figs. 142-153 scale bar = 30 jum. Figs. 142, 143 culture source: Tribune Bay, Hornby Is., Figs. 144, 148, 149, 153 culture source: Tribune Bay, Hornby Is., Figs. 145-147, 150, 151 culture source: east side Denman Is., Fig. 152 culture source: Ford Cove, Hornby Is. / 229 Figs. 154-162. Gelidium vagum sporeling development in culture. Fig. 154. First division of fusiform cell (arrow), tetrasporophyte; Fig. 155. Fusiform and concave cell groups distinct, tetrasporophyte; Fig. 156. Fusiform (arrowhead) and concave (arrow) cell groups distinct, tetrasporophytes; Fig. 157. Apical cell (arrowhead) formation opposite rhizoid, tetrasporophyte; Fig. 158 Apical cell (arrowhead) formation from fusiform group cell, tetrasporophyte; Fig. 159. Sporeling increased in size, with apical organization, tetrasporophyte; Fig. 160. Gametophyte sporelings with apical organization; Fig. 161. Attachment pad formed from adhesion of elongated cortical cells, tetrasporophyte; Fig. 162. Branch initiation (at arrow), gametophyte. Figs. 154, 155, 157-161 scale bar = 30 um, Figs. 156, 162 scale bar = 40 um. Figs. 154, 155 culture source: east side Denman Is., Figs. 156-159 culture source: Galleon Pt., Hornby Is., Figs. 160-162 culture source: Tribune Bay, Hornbj' Is. / 231 Figs. 163-168. Gelidium vagum reproduction in culture. Fig. 163. Tetrasporangial stichidia on tetrasporophyte from carpospores, scale bar = 400 um; Fig. 164. Gametangial stichidium on gametophyte grown from a tetraspore, scale bar =100 um; Fig. 165. Gametangial stichidia on gametophyte from a tetraspore, scale bar = 700 um; Fig. 166. Tetraspores being released from tetrasporophyte grown from a carpospore, scale bar = 500 um; Fig. 167. Germination of tetraspores released from tetrasporophyte grown from carpospore, scale bar = 300 um; Fig. 168. Fertile stichidia on tetrasporophyte grown from carpospore, originally grown from tetraspore, scale bar = 500 um. Figs. 163, 166, 167 culture source: east side Denman Is., Fig. 164, 165 culture source: Galleon Pt., Hornby Is., Fig. 168 culture source: Tribune Bay, Hornby Is. 232 / 233 Figs. 169-173. Reproduction of Gelidium purpurascens and G. coulteri in culture. Fig. 169. Gelidium purpurascens, tetrasporangial stichidia from field, excised for tetraspore release, scale bar = 200 um; Fig. 170. G. purpurascens, habit of fertile male gametophyte grown from tetraspore, scale bar = 500 um; Fig. 171. G. purpurascens, habit of fertile female gametophyte grown from tetraspore, scale bar = 500 um; Fig. 172. G. coulteri, spermatangial area of male gametophyte grown from tetraspore, scale bar = 200 um; Fig. 173. G. coulteri, habit of fertile male gametophyte grown from tetraspore, scale bar = 500 um. Fig. 174 Orlebar Pt., Gabriola Is., Figs. 170, 171 culture source: Whalebone Bay, Gabriola Is., Figs. 172, 173 culture source: east side Denman Is. / 235 Figs. 174-177. Chromosomes of Gelidium vagum. Figs. 174, 176, 177. Prophase I of undivided tetrasporangia, hematoxylin stained, Tribune Bay, Hornby Is.; Figs. 174-176 n= 14-15, Fig. 175. drawing of nucleus in Fig. 174; Fig. 177. five large chromosomal bodies; scale bar = 50 um for all figures. / 237 Figs. 178-183. Carpogonium and carposporophyte development in Gelidium purpurascens and G. vagum. Figs. 178. Gelidium vagum, intercalary carpogonium, longitudinal section, scale bar=10 um; Figs. 179-183 G. purpurascens. Fig. 179. Sessile carpogonium (arrow) with curved basal wall, longitudinal section, scale bar=10 um; Fig. 180. Sessile carpogonium (arrow) vacuolate and degenerate, longitudinal section, scale bar =10 um; Fig. 181. Fertilized carpogonium (arrow) fusing with cortical cells through two pit connections (arrowheads), showing functional carpogonium is intercalary, squash of longitudinal section, scale bar =10 um; Fig. 182. Multilobed, multinucleate fusion cells, squash of longitudinal section, scale bar=10 um; Fig. 183. Young carposporophyte, clusters of nutritive chains around second order plate cells and fusion cell (arrow), longitudinal section, scale bar = 300 um. Fig. 178 Ta-no-hama, Iwate-ken, Japan, TNS 25824, Figs. 179, 180, 183 Kirby Pt., Diana Is., Figs. 181, 182 Geer Islets. 238 / 239 Figs. 184-188. Carposporophyte development in Gelidium purpurascens and G. vagum. Figs. 184, 185. Gelidium purpurascens. Fig. 184. Nutritive chains, longitudinal section, scale bar=10 um; Fig. 185. Fusion cell and gonimoblast filaments (arrows), longitudinal section, scale bar = 20 ym; Fig. 186-188. G. vagum. Fig. 186. Lobe of gonimoblast cell (arrow) fused with apical cell (arrowhead) of (upper) nutritive chain, longitudinal section, scale bar =10 um; Fig. 187. One locule of mature cystocarp showing carposporophyte producing carposporangia, longitudinal sections through ostiole, scale bar = 50 um; Fig. 188. Gonimoblast cells producing carposporangia terminally and laterally, longitudinal section, scale bar=10 um. Figs. 184, 185 Geer Islets, Figs. 186, 188 Tribune Bay, Hornby Is., Fig. 187 Ta-no-hama, Iwate-ken, Japan, TNS 25824. 240 / 241 Figs. 189-192. Ultrastructure of spermatiogenesis in Gelidium vagum. Fig. 189. Spermatangial mother cells (arrow) and spermatangia (arrowhead), longitudinal section, scale bar=10 um; Fig. 190. Spermatangia (small cells) and vegetative cortical cells (larger), whole mount, scale bar = 30 um; Fig. 191. Ultrastructure of vegetative cortical cell with large plastids, basal nucleus (arrowhead) and central vacuole, scale bar=1000 nm; Fig. 192. Ultrastructure of two spermatangial mother cells each with a pit connection to base of a third spermatangial mother cell, which has a pit connection to a cortical cell, scale bar = 1000 nm. Figs. 189, 192 Tribune Bay, Hornby Is., Fig. 190 Galleon Pt., Fig. 191 east side Denman Is. 242 / 243 Figs. 193-196. Ultrastructure of spermatiogenesis in Gelidium vagum. Fig. 193. Spermatangial mother cell with elongate, central nucleus and furrow (arrowheads) possibly excluding apical chloroplast; Fig. 194. Spermatangial mother cell with two nuclei; Fig. 195. Spermatangium with two prominent dictyosomes (arrows) contributing vesicles to young spermatangial vacuole, mitochondrion (M); Fig. 196. Spermatangium with basal pit connection and young spermatangial vacuole, cored vesicles (arrowheads), large nucleus; scale bar = 500 nm for all figures. Figs. 193-196 Tribune Bay, Hornby Is. 244 / 245 Figs. 197-200. Ultrastructure of spermatiogenesis in Gelidium purpurascens and G. vagum. Figs. 197, 198. Gelidium vagum. Fig. 197. Spermatangium with 2-3 spermatangial vacuoles and large nucleus with condensed chromatin (dark areas), scale bar = 300 nm; Fig. 198. Spermatangial vacuole release, nucleus with condensed chromatin (dark areas), scale bar = 300 nm; Figs. 199, 200 G. purpurascens. Fig. 199. Two spermatangial vacuoles released from spermatangium, spermatangium with many cored vesicles (arrows), scale bar = 800 nm; Fig. 200. Spermatangial vacuole, contents whorled fibrillar, scale bar = 300 nm. Figs. 197, 198 Tribune Bay, Hornbyls., Figs. 199, 200 cultured Whalebone Bay, Gabriola Is. material. 2 46 / 247 Figs. 201-204. Ultrastructure of spermatiogenesis in Gelidium purpurascens and G. vagum. Figs. 201, 204. Gelidium purpurascens. Fig. 201. Spermatangium with many cored vesicles (arrows), mitochondria (M) and nucleus with condensed chromatin (dark areas), scale bar = 40 nm; Figs. 202, 203 G. vagum. Fig. 202. Spermatangium starting to release spermatium, spermatangial vacuoles already released, many cored vesicles (arrows) and thick mucilage coating at apex of spermatium, scale bar = 400 nm; Fig. 203. Spermatangium with cored vesicles (arrows), spermatangial vacuoles already released, scale bar = 500 nm; Fig. 204. Spermatium being released and wall material (arrowhead) secreted by spermatangial mother cell, spermatangial mother cell on right at cytokinesis, scale bar = 700 nm. Figs. 201, 204 cultured Whalebone Bay, Gabriola Is. material, Figs. 202, 203 east side Denman Is. 248 / 249 Figs. 205-208. Ultrastructure of spermatiogenesis in Gelidium purpurascens and G. vagum. Figs. 205-207.Gelidium purpurascens. Fig. 205. Release of percurrent spermatangia, spermatangial mother cell secreting new wall (arrowheads), note previous wall layer (arrows) surrounding spermatangial mother cell and spermatangium, scale bar = 800 nm; Fig. 206. Release of conical spermatium, large nucleus and many cored vesicles (arrows), scale bar = 500 nm; Fig. 207. Spermatangial mother cells and old wall layers (arrowheads) of percurrent spermatangia, scale bar = 2000 nm; Fig. 208. G. vagum spermatangial mother cell with old wall layers and discarded pit plugs (arrowheads) of percurrent spermatangia, scale bar=1000 nm. Figs. 205-207 cultured Whalebone Bay, Gabriola Is. material, Fig. 208 Tribune Ba3', Hornby Is. 250 / 251 Fig. 209. Summary of spermatiogenesis in Gelidium. a-c. Longitudinal division of an outer cortical cell to form 4 spermatangial mother cells; d-j. Summary of ultrastructural development of the spermatangium, d. Spermatangial mother cell (smc) with basal nucleus (n), starch grains (st) and chloroplasts (cl) throughout cytoplasm; e. Apical chloroplast (cl) of spermatangial mother cell (smc) being cut off from rest of cell by furrows (arrows); f. Spermatangial mother cell transversely divided to form a spermatangium (sp), spermatangium with a dictyosome (arrowhead) and vesiculated endoplasmic reticulum (arrow); g. Spermatangium (sp) with 2 spermatangial vacuoles (sv) developed from dictyosome-derived vesicles and endoplasmic reticulum, nucleus (n) with condensed chromatin; h. Spermatangial vacuoles (sv) released from the spermatangium, spermatangium with cored vesicles (cv) and an external layer of mucilage-like material (m); i. Spermatium (s) released from spermatangium, spermatium with cored vesicles, a layer of mucilage-like material and nucleus with condensed chromatin; j. Spermatangial mother cell expanded into space vacated by spermatium to form another spermatangium; drawings not to scale. 252 / 253 Figs. 210-215. Ultrastructure of tetrasporogenesis in Gelidium purpurascens. Fig. 210. Remains of vacuole (V) in very young tetrasporangium (before expansion of nucleus and thickening of vegetative wall), scale bar = 2000 nm; Fig. 211. Membranous material (arrows) extruded from pre-meiotic tetrasporangium, mitochondrion (M), scale bar = 400 nm; Fig. 212. Different section of very young tetrasporangium of Fig. 210, material from vacuole (V) incorporated in vesicles (arrows) and released to wall, scale bar = 400 nm; Fig. 213. Nucleus of pre-meiotic tetrasporangium with perinuclear endoplasmic reticulum (small arrows), many nuclear pores (arrowheads), large nucleolus, large nucleolar vacuoles (large arrows), scale bar=500 nm; Fig. 214. Pre-meiotic tetrasporangium with central nucleus, large nucleolus, lenticular starch grains (arrowheads) central around the nucleus, scale bar=2000 nm; Fig. 215. Pre-meiotic tetrasporangium, nucleus and nucleolus large, central, chloroplasts mostly peripheral, lenticular starch grains (arrows) around nucleus, scale bar=1000 nm; Figs. 210, 212, 213 Georgina Pt., Mayne Is., Figs. 211, 214, 215 Tribune Bay, Hornby Is. culture. 254 / 255 Figs. 216-220. Ultrastructure of tetrasporogenesis in Gelidium purpurascens. Fig. 216. Convoluted plasma membrane (arrowhead) of pre-meiotic tetrasporangium, small transparent vesicles (arrows) of unknown origin release their contents to wall, several mitochondria (M), scale bar = 200 nm; Fig. 217. Transverse section of young tetrasporangium, central nucleus (arrow) dumbell shaped, possibly undergoing meiosis, many chloroplasts, lenticular starch grains (arrowheads), scale bar = 2000 nm; Fig. 218. Spindle-shaped, single nucleus of pre-meiotic tetrasporangium with large nucleolus, small amount of perinuclear endoplasmic reticulum (arrowheads), lenticular starch grains, mitochondrion (M), scale bar = 400 nm; Fig. 219. Elongate nucleus of pre-meiotic tetrasporangium with large nucleolus, perinuclear endoplasmic reticulum (arrows), lenticular starch grains (arrowheads), scale bar = 500 nm; Fig. 220. Two nuclei in tetrasporangium after first meiotic division and before cytokinesis, many ovate chloroplasts, many lenticular to ovate starch grains, grey wall layer (layer 2) present (arrow) inside layer 1 wall (W), scale bar = 3000 nm; Fig. 216 Georgina Pt., Mayne Is., Figs. 217-220 Tribune Bay, Hornby Is., culture. 256 / 257 Figs. 221-226 Ultrastructure of tetrasporogenesis in Gelidium purpurascens. Fig. 221. Flat dictyosome associated with mitochondrion (M) in tetrasporangium about the time of meiosis, dictyosome cisternae contains fibrillar material, scale bar = 200 nm; Fig. 222. Vesiculate endoplasmic reticulum at periphery of post-meiotic tetrasporangium, vesicles contain fibrillar material, wall layer 2 present (W2), scale bar = 400 nm; Fig. 223. Part of a post-meiotic tetrasporangium with many chloroplasts and starch grains, peripheral dictyosomes flat with many cisternae per stack, mitochondrion (M), scale bar = 800 nm; Fig. 224. Tetrasporangium at about the time of karyokinesis, many chloroplasts, many lenticular to ovate starch grains, wall layer 2 (W2) present, scale bar = 3000 nm; Fig. 225. Peripheral cytoplasm of post-meiotic tetrasporangium, many dictyosomes and associated mitochondria (M), vesiculate endoplasmic reticulum (arrows), wall layer 2, scale bar = 400 nm; Fig. 226. Tetrasporangium at the time of cytokinesis, wall layer 3 (arrow) present interior to layer 2 (arrowhead), many ovate starch grains, many chloroplasts, scale bar =4000 nm; Figs. 221-226 Tribune Bay, Hornby Is., culture. 258 / 259 Figs. 227-232. Ultrastructure of tetrasporogenesis in Gelidium coulteri and G. purpurascens. Figs. 227, 228, 230-232 Gelidium purpurascens. Fig. 227. Tetrasporangial wall layers at cytokinesis, layer 1 (Wl), layer 2'(arrow), layer 3 (arrowhead), scale bar = 200 nm; Fig. 228. Cytoplasm of tetrasporangium after cytokinesis, starch grains ovate, large dictyosome curved and associated with mitochondrion (M), scale bar = 400 nm; Fig. 229. Gelidium coulteri, chloroplasts radiating around nucleus, starch grains between chloroplasts, scale bar =1000 nm; Fig. 230. Strongly curved dictyosome associated with mitochondrion (M) in post-cytokinesis tetrasporangium, dark material (arrowhead) between appressed younger cisternae, cored vesicles (arrows) produced at dictyosome maturing face, scale bar = 200 nm; Fig. 231. Cored vesicles (arrows) releasing contents to cleavage furrow (F) of tetrasporangium, scale bar = 400 nm; Fig. 232. Wall of mature tetrasporangium (after cytokinesis), layer 1 (Wl), layer 2 (black arrow), layer 3 (arrowhead), mucilage-like layer (white arrow), scale bar = 60 nm; Figs. 227, 228, 230-232 Tribune Bay, Hornby Is., culture, Fig. 229 Brady's Beach, Bamfield. 260 / 261 TABLES Table 1. Historical summary of genera included in Gelidiaceae (horizontal continuation of table on next page). Kiitzing, 1843 Kiitzing, 1849 Acrocarpus Ctenodus Echinocaulon Gelidium Acrocarpus Chondrodon Delisea Echinocaulon Euctenodus Gelidium Phacelocarpus Polycladia Thysanocladia J. G. Agardh, 1876 Gelidium Pterocladia Ptilophora Suhria Wurdemannia Schmitz, 1889 Acropeltis Atractophora Binderella Caulacanthus Choreocolax Gelidium • Harveyella Naccaria Porphyroglossum Pterocladia Ptilophora Schottmullera Suhria Wrangelia 262 Kylin, 1956 Fan, 1961 current, 1988 Acanthopeltis Acropeltis Beckerella Gelidiella Gelidiocolax Gelidium Porphyroglossum Pterocladia Ptilophora Suhria Yatabella Acanthopeltis Acropeltis Beckerella Gelidium Porphyroglossum Pterocladia Ptilophora Suhria Yatabella Acanthopeltis Beckerella Gelidiella Gelidium Onikusa Porphyrogloss um Pterocladia Pterocladias trum Ptilophora Suhria Yatabella / 264 Table 2. Sites and dates of field collections of Gelidiales from the study area. Site Latitude Longitude Dates collected Brady's Beach, 48°50'N Barkley Sound Diplock Island (NE 48°56'N corner), Barkley Sound Dixon Island (SE 48°51'N corner), Barkley Sound Dodger Channel, 48°51'N Barkley Sound Geer Islets, Barkley 48°56'N Sound Haines Island, 48°50'N Barkley Sound Kirby Point, Diana 48°51'N Island, Barkley Sound Meade Islets, Barkley 48°56'N Sound Roquefeuil Bay, 48°52'N Barklej' Sound 125°09'W 24/vi/86 125° 07'W 28/viii/84 125°07'W 27/viii/84 125°12'W 29/viii/84 125°07'W 28/viii/84 125° 12'W 29/viii/84 125°13'W 29/viii/84, 27/viii/85, 23/vi/86 125°07'W 28/viii/84 125°07'W 27/viii/84 / Ross Islets, Barkley 48°52'N 125°10'W Sound Madrona Point, Orcas 48°41.5'N 122°54'W Island, Gulf of Georgia Georgina Point, 48°52'N 123°17'W Mayne Island, Gulf of Georgia False Narrows, 49°08'N 123°49'W Gabriola Island, Strait of Georgia Orlebar Point, 49°12'N 123"49'W Gabriola Island, Strait of Georgia Whalebone Bay, 49°11'N 123°48'W Gabriola Island, Strait of Georgia Earls Cove, Strait of 49°45'N 124°00'W Georgia Finn Cove, Strait of 49°59'N 124°46'W Georgia 28/viii/86 24M/85 16/xi/85, 14/ii/86, 21/ix/86, 13/viii/87 8/x/85 8/x/85, ll/xii/85, 28/i/86, 3/iii/86, 7/v/86, 4/vi/86, 21M/86, 29/viii/86, 28/x/86, 17/U/87, 5/iii/87, 5/viii/87 28/x/86, 27M/86, 5/viii/87 6/iii/85 6/xi/83 / 266 Frolander Bay, Strait of Georgia Halfmoon Bay, Strait of Georgia Denman Island (E side), Strait of Georgia Ford Cove, Hornby Island, Strait of Georgia Galleon Point, Hornby Island, Strait of Georgia Helliwell Park, Hornby Island, Strait of Georgia Tribune Bay, Hornby Island, Strait of Georgia 49°45'N 124°17'W 5/xi/83 49°30'N 123°56'W 6/iii/85 49°30.5'N 124°44'W 29/ix/86, 26/viii/87 49°30'N 124°41'W l/x/86 49°33'N 124°40'W 30/ix/86 49°31'N 124°35'W 23/iv/87 49°32'N 124°38'W 15/x/84, 30/ix/86 / 267 Table 3. Herbaria from which loans were obtained and their abbreviations (Holmgren et al, 1981). Herbarium Abbreviation Allan Hancock Foundation, collection housed in LAM AHFH National Museum of Canada, Ottawa CANA California Academy of Sciences, collection housed in UC CAS Friday Harbor Laboratories, Friday Harbor FHL Los Angeles Countj' Museum, Los Angeles LAM School of Botany, Trinity College, Dublin TCD National Science Museum, Tokyo TNS University of British Columbia, Vancouver UBC University of California, Berkeley UC University of Washington, Seattle WTU / 268 Table 4. Species of Gelidiales recorded from the northeast, northwest and southeast Pacific Ocean areas. Numbered superscripts refer to references as follows: 1. Akatsuka, 1986b 2. Dawson, 1944 3. Dawson, 1953 4. Dawson et al., 1964 5. Santelices, 1977 6. Santelices and Montalva, 1983 7. Santelices and Stewart, 1985 8. Schnetter and Bula Meyer, 1982 9. Stewart, 1976 10. Stewart and Norris, 1981 11. Bangmei et al., 1983 Taxon Japan China Hawaii Mexico Central & Calif- and ornia South America Acanthopeltis japonica Okamura • Beckerella irregularis Akatsuka et Masaki Beckerella subcostata (Okamura) Kylin Gelidiella acerosa (Forsskal) Feldmann et Hamel Gelidiella adnata Dawson Gelidiella hancockii Dawson Gelidiella ligulata Dawson Gelidiella machristiana Dawson Gelidiella mexicana Dawson Gelidiella myrioclada Borgesen Gelidiella ramellosa (Kiitzing) Feldmann et Hamel Gelidiella refugiensis Dawson Gelidiella stichidiospora Dawson Gelidium amamiense Tanaka Gelidium amansii (Lamouroux) Lamouroux Gelidium arborescens Gardner Gelidium bulae Schnetter Gelidium cartilagineum (L.) Greville Gelidium chilense (Montalva) Santelices et Montalva Gelidium corneum (Hudson) Lamouroux Gelidium coronadense Dawson Gelidium coulteri Harvey Gelidium crinale (Turner) Lamouroux Gelidium deciduum Dawson Gelidium decumbensum Okamura Gelidium divaricatum vonMartens Gelidium isabelae Taylor Gelidium japonicum (Harvey) Okamura Gelidium johnstonii Setchell et Gardner Gelidium kintaroi (Okamura) Yamada Gelidium latifolium (Greville) Bornet Gelidium lingulatum Kutzing Gelidium linoides Kutzing Gelidium nanum Inagaki Gelidium nudifrons Gardner Gelidium pacificum Okamura Gelidium planisculum Okamura Gelidium pluma Loomis Gelidium polycladum Kutzing Gelidium polystichum Gardner Gelidium pristoides Turner Gelidium pseudointricatum Skottsberg et Levring Gelidium pulchrum Gardner Gelidium purpurascens Gardner Gelidium pusillum (Stackhouse) LeJolis Gelidium reediae Loomis Gelidium rex Santelices et Abbott Gelidium rigens Greville Gelidium robustum (Gardner) Hollenberg et Abbott Gelidium sclerophyllum Taylor Gelidium subfastigiatum Okamura Gelidium sinicola Gardner Gelidium. vagum Okamura Gelidium yamadae (Okamura) Fan Onikusa japonica (Harvey) Akatsuka Pterocladia bulbosa Loomis Pterocladia caerulescens (Kutzing) Santelices Pterocladia caloglossoides (Howe) Dawson Pterocladia capillacea (Gmelin) Bornet et Thuret Pterocladia densa Okamura Pterocladia mcnabbiana Dawson Pterocladia media Dawson Pterocladia nana Okamura Pterocladia tenuis Okamura Yatabella hirsuta Okamura / Table 5. Differences between Gelidium purpurascens and Gelidium robustum Character Gelidium purpurascens Gelidium robustum height main axis branching higher order branch bases inner cortex usually not > 15 cm 1 - several dense, branches close together, separated by 1-2 mm not constricted cells unordered commonly >15 cm 1 coarse, loose, branches separated by at least mm constricted cells ordered in pallisade-like rows tetrasporangial stichidium narrow sterile margin wide sterile margin / 274 Table 6. Anatomical differences between Gelidium johnstonii and Gelidium vagum Characters Gelidium johnstonii Gelidium vagum vegetative: outer cortical cells tetrasporangial stichidia: shape apex sterile margin apical cell apical furrow flattened in anticlinal files not in anticlinal files spatulate blunt present even or recessed present yes ovate, tapered apically pointed absent protruding absent no - at most compressed cystocarp: third order filaments intact, stretched not intact, broken APPENDIX 1. HERBARIUM SPECIMENS EXAMINED F = female gametophyte, G = gametophyte, M = male gametophyte, T = tetrasporophyte, specimens without a letter designation are vegetative Gelidium coulteri AHFH 53929 CANA 3474 FHL 3055 TCD California Beacon Hill, Victoria Cape Alava, Washington California ISOTYPE TCD California UBC A906 Monterey, California UBC A907 Monterey, California UBC A1449 Moss Beach, California UBC A2793 Amos Is., Kyuquot, Vancouver Is. UBC A4890 Mukkaw Bay, Washington UBC A10812 Mills Peninsula, Vancouver Is. UBC A10814 Esteban Point, Vancouver Is. UBC A11940 McLean Is.. Vancouver Is. UBC A11947 Spring Is., Kyuquot, Vancouver Is. UBC A12073 Miracle Beach, Vancouver Is. UBC A12294 Spring Is., Kyuquot, Vancouver Is. photograph of LECTOTYPE ISOTYPE T 275 UBC A12298 UBC A12299 UBC A12300 UBC A13259 UBC A13478 UBC A13658 UBC A14403 UBC A15173 UBC A19739 UBC A24527 UBC A24837 UBC A28351 UBC A31363 UBC A31426 UBC A36197 UBC A36198 UBC A36199 UBC A36372 UBC A36373 UBC A36374 UBC A36545 UBC A36711 UBC A36712 UBC A36713 McLean Is., Vancouver Is. McLean Is., Vancouver Is. McLean Is., Vancouver Is. Perez Rocks, Vancouver Is. Walters Cove, Vancouver Is. La Jolla, California Mukkaw Bay, Washington Malaspina Narrows, Queen Charlotte Strait Brooks Peninsula, Vancouver Is. Pescadero Point, San Mateo Co., California Mukkaw Bay, Washington Orlebar Point, Gabriola Is., B.C. Decanso Bay, Gabriola Is., B.C. Davison's Beach, Gabriola Is., B.C. Brooks Peninsula, Vancouver Is., B.C. False Narrows, Gabriola Is., B.C. Lock Bay, Gabriola Is., B.C. Jackobson Point, Brooks Peninsula, Vancouver Is., B.C. Bunsby Is., Vancouver Is., B.C. Bunsby Is., Vancouver Is., B.C. Amos Is., Vancouver Is., B.C. Bunsby Is., Vancouver Is., B.C. Bunsby Island, Vancouver Is., B.C. Bunsby Is., Vancouver Is., B.C. / 277 UBC A37077 Fossil Beach, Grassy Is., B.C. UBC A37078 Fossil Beach, Grassy Is., B.C. UBC A37079 Grassy Is., B.C. UBC A37219 Yellow Bluff, Vancouver Is., B.C. UBC A37220 Yellow Bluff, Vancouver Is., B.C. UBC A37221 Fossil Beach, Grassy Is., B.C. T UBC A37223 Yellow Bluff, Vancouver Is., B.C. UBC A37402 Pincer Is., Vancouver Is., B.C. T UBC A37403 Pincer Is., Vancouver Is., B.C. T UBC A37404 Nootka Is., Vancouver Is., B.C. T UBC A38472 Lawn Point, Vancouver Is., B.C. UBC A38473 Lawn Point, Vancouver Is., B.C. UBC A38474 Lawn Point, Vancouver Is., B.C. UBC A39128 Arab Cove, Vancouver Is., B.C. UBC A40404 Mukkaw Bay, Washington UBC A40405 Mukkaw Bay, Washington UBC A40415 Crescent Beach, Clallam Co., Washington T UBC A40467 Diana Is., Barkley Sound, B.C. UBC A40468 Diana Is., Barkley Sound, B.C. UBC A40469 Diana Is., Barklej' Sound, B.C. UBC A40470 Diana Is., Barkley Sound, B.C. UBC A42814 Lawn Point, Vancouver Is., B.C. UBC A43519 Fleming Is., Barkley Sound, B.C. UBC A43520 Reeks Is., Barkley Sound, B.C. UBC A43521 Aguilar House Beach, Bamfield, B.C. / 278 UBC A43522 Aguilar House Beach, Bamfield, B.C. UBC A43523 Effingham Is., Barkley Sound, B.C. UBC A47159 S. California (L.A. Co.) UBC A50505 Mussel Point, Monterey, California UBC A53814 Kirby Point, Diana Is., Barkley Sound, B.C UBC A53975 Brooks Peninsula, Vancouver Is., B.C. UBC A60236 Aguilar House Beach, Bamfield, B.C. UBC A60254 Diana Is., Barkley Sound, B.C. UBC A60268 Execution Rock, Barkley Sound, B.C. UBC A60460 Aguilar House Beach, Bamfield, B.C. UBC A61984 Pigeon Point, California UBC A62016 Montana de Oro, California UBC A64934 Ladysmith, B.C. UC 305332 Moss Beach, Pacific Grove, California WTU 246345 Steamboat Is., Thurston Co., Washington WTU 246351 Mukkaw Bay, Washington Gelidium johnstonii AHFH 69 San Francisquito Bay, Baja California, Mexico T, ISOTYPE AHFH 2211 Pond Is., off Angel de la Guardia Is., G Mexico AHFH 4150 Ensenada Bocochibampo, near Guayamas, T Mexico / 279 AHFH 4156 AHFH 4179 AHFH 4192 AHFH 4193 AHFH 4194 AHFH 50267 AHFH 50268 AHFH 50299 CAS 1343 in UC CAS 484385 in UC CAS 484386 in UC CAS 464388 in UC CAS 484390 in UC LAM 52684 LAM 52894 Bahia Carrizal, near Cabo Circo, Sonora, Mexico Ensenada Bocochibampo, Sonora, Mexico Isla Jorge, Gulf of California, Mexico Isla Patos, near Isla Tiburon, Gulf of California, Mexico Isla Patos, near Isla Tiburon, Gulf of California, Mexico Bahia Aqua Verde, Baja California, Mexico Punta Perico, Salinas Bay, Carmen Is., Gulf of California, Mexico Puerto Escondido, Baja California, Mexico San Francisquito Bay, Baja California, Mexico San Francisquito Bay, Baja California, Mexico San Francisquito Bay, Baja California, Mexico San Marcos Is., Gulf of California, Mexico Puerto Escondido, Gulf of California, Mexico SW end of Isla Partida, Gluf of California, Mexico G G,T G,T G,T G, T HOLO-TYPE T, ISOTYPE San Francisquito Bay, Baja California, Mexico T, ISOTYPE G, T ISO-TYPE T, PARA-TYPE T T / 280 Gelidium purpurascens CANA 3473 Beacon Hill, Victoria, B.C. T CANA 3740 Beacon Hill, Victoria, B.C. CANA 3843 Departure Bay, Vancouver Is., B.C. T CANA 4346 Departure Bay, Vancouver Is., B.C. CANA 4349 Departure Bay, Vancouver Is., B.C. FHL 1205 Santa Cruz, California T FHL 2849 Mitchell Point, San Juan Is., Washington UBC A903 Ensenada, California T UBC A1799 East Sound, Orcas Is., Washington UBC A1952 Ucluelet, Vancouver Is., B.C. T UBC A4263 Garden Is., Kyuquot, B.C. T UBC A4264 Garden Is., Kyuquot, B.C. T UBC A4265 Kains Is., B.C. UBC A4402 American Camp Beach, San Juan Is., Washington T UBC A4403 American Camp Beach, San Juan Is., Washington T UBC A5195 Minnesota Seaside Station, Vancouver Is., B.C. UBC A6744 Sunset Bay, Oregon UBC A6755 Cape Arago, Oregon T UBC A10310 Cortes Is., B.C. UBC A10514 Catala, Vancouver Is., B.C. T / 281 UBC A10813 Esteban Point, Vancouver Is., B.C. M UBC A11271 Ogden Point Breakwater, Victoria, B.C. UBC A11278 Rosebush Is., Vancouver Is., B.C. T UBC A11329 Dorcas Point, Strait of Georgia, B.C. T UBC A11929 McLean Is., Vancouver Is., B.C. T UBC A11937 McLean Is., Vancouver Is., B.C. T UBC A11988 Spring Is., Vancouver Is., B.C. T UBC A12296 McLean Is., Vancouver Is., B.C. T UBC A13479 Walters Cove, Vancouver Is., B.C. T UBC A17771 Acous Peninsula, B.C. T UBC A19032 Hisnit Is., B.C. UBC A28349 Orlebar Point, Gabriola Is., B.C. F UBC A28646 Twin Beaches, Gabriola Is., B.C. UBC A31414 Lock Bay, Gabriola Is., B.C. T UBC A35158 Arab Cove, Vancouver Is., B.C. T UBC A35175 Arab Cove, Vancouver Is., B.C. UBC A36371 Brooks Peninsula, Vancouver Is., B.C. T UBC A36710 Bunsby Is., Vancouver Is., B.C. T UBC A36875 Union Is., Vancouver Is., B.C. T UBC A36876 Amos Is., Vancouver Is., B.C. F UBC A37603 Blight Is., Vancouver Is., B.C. T UBC A38471 Lippy Point, Vancouver Is., B.C. T UBC A38475 Arab Cove, Vancouver Is., B.C. T UBC A38734 Arab Cove, Vancouver Is., B.C. T UBC A38908 Arab Cove, Vancouver Is., B.C. T / 282 UBC A39126 Lawn Point, Vancouver Is., B.C. F UBC A39622 Arab Cove, Vancouver Is., B.C. T UBC A40192 Amphitrite Point, B.C. T UBC A40743 Haines Is., Barkley Sound, B.C. T UBC A40793 Helby Is., Barkley Sound, B.C. T UBC A41214 Meade Islets, Barkley Sound, B.C. T UBC A41384 Helby Is., Barkley Sound, B.C. T UBC A41415 Fleming Is., Barkley Sound, B.C. M UBC A41491 Grappler Inlet, Bamfield, B.C. T UBC A41586 Wizard Islet., Barkley Sound, B.C> F UBC A41606 Effingham Is., Barkley Sound, B.C. T UBC A41607 Effingham Is., Barkle}' Sound, B.C. T UBC A41696 Roquefeuil Baj', Barkley Sound, B.C. T UBC A41842 Fleming Is., Barkley Sound, B.C. F,T UBC A41843 Fleming Is., Barkley Sound, B.C. UBC A41844 Fleming Is., Barkley Sound, B.C. T UBC A41946 Clarke Is., Barkley Sound, B.C. F,T UBC A42162 Reeks Is., Barkley Sound, B.C. T UBC A42436 Fleming Is., Barkley Sound, B.C. T UBC A42437 Fleming Is., Barkley Sound, B.C. T UBC A42438 Fleming Is., Barkley Sound, B.C. T UBC A42439 Fleming Is., Barkley Sound, B.C. T UBC A42733 Arab Cove, Vancouver Is., B.C. UBC A43525 Diana Is., Barkley Sound, B.C. T UBC A43526 Bamfield, B.C. T / 283 UBC A45812 UBC A46026 UBC A46960 UBC A53291 UBC A53297 UBC A53396 UBC A53447 UBC A53657 UBC A54172 UBC A55706 UBC A57371 UBC A57429 UBC A57442 UBC A57598 UBC A58562 UBC A58672 UBC A60374 UBC A60485 UBC A64402 UBC A64426 UBC A66660 Bamfield, B.C. T Bamfield, B.C. F,T Sear Is., B.C. Kyuquot Channel, B.C. T Kyuquot Bay, Kyuquot Sound, B.C. T Saanedra Is., Nootka Sound, B.C. Henderson Point, Saanich Inlet, Vancouver Is., B.C. Friendly Cove, Nootka Sound, B.C. M Hotham Inlet, Harmony Is., Jervis Inlet, B.C. Lasqueti Is., B.C. Helby Is., Barkley Sound, B.C. T Moorsam Bluff, Jervis Inlet, B.C. east of Brittain Rock, Jervis Inlet, B.C. mouth of Glacial Creek, Jervis Inlet, B.C. Station Is., B.C. island W of Fox Is., B.C. Diana Is., Barkely Sound, B.C. T Brady's Beach, Bamfield, B.C. T head of Pennell Sound, Queen Charlotte Is., T B.C. heaqd of Pennell Sound, Queen Charlotte Is., B.C. Tasu Sound, Queen Charlotte Is., B.C. / 284 UBC A66682 UBC A66778 UBC A67220 UBC A67249 UBC A67446 UBC A69432 UBC A69433 UBC A69434 UC 93572 UC 276633 UC 296689 UC 305332 UC 305364 UC 305371 UC 305373 UC 1452112 private collection of J.R. Waaland: 1583 Horn Rocks, Tasu Sound, Queen Charlotte Is., B.C. Departure Bay, Vancouver Is., B.C. Cunningham Is., B.C. Bachelor Bay, B.C. Kirby Point, Diana Is., Barkley Sound, B.C. F, T Cape Suspiro, Alaska T Cape Suspiro, Alaska Kanaka Bay, San Juan Is., Washington T Moss Beach, San Mateo Co., California F HOLO-TYPE Moss Beach, San Mateo Co., California F Moss Beach, San Mateo Co., California F, T Moss Beach, San Mateo Co., California F Pebble Beach, Monterey, California F Pescadero, California F Pescadero, California M Moss Beach, California F Kanaka Point, San Juan Is., Washington Gelidium robustum / 285 UBC A1448 UBC A7861 UBC A37790 UBC A47163 UBC A47164 UBC A50638 UBC A58591 UBC A60676 UBC A62199 UBC A63553 UC 294572 UC 395419 UC 647822 UC 756464 UC 756469 UC 756470 UC 940173 UC 1451987 Moss Beach, California T Shoal B, Victoria T Asilomar Point, California Palos Verdes, California Santa Catalina Is., California Point Pinos, Monterey Peninsula, California Santa Cruz Is., California T Pedros Blancas, California Carpinteria Reef, California F Monterey, California near Ensenada, California San Pedro, California White's Point, San Pedro, California Punta Santa Rosalia, Baja California, Mexico Punta Santa Rosalia, Baja California, Mexico Rosario, Baja California, Mexico Natividad Is., Baja California, Mexico F Portuguese Bend, San Pedro, California F T HOLO-TYPE F F Gelidium sinicola UC 276620 Point Cavalo, Marin Co., California HOLOTYPE / 286 Gelidium vagum TNS 25817 Harutachi, Hidaka, Japan T TNS 25823 Hideshima, Iwate-ken, Japan T TNS 25824 Ta-no-hama, Iwate-ken, Japan F TNS 25825 Ta-no-hama, Iwate-ken, Japan F TNS 25847 Same, Aomori-ken, Japan T UBC A56807 Muroran, Hokkaido, Japan T UBC A64965 Ladysmith, Vancouver Is., B.C. T Pterocladia caloglossoides UBC A12295 UBC A19667 UBC A28937 UBC A29576 UBC A30401 UBC A31251 UBC A33545 UBC A37605 UBC A39126 UBC A39127 UBC A61245 McLean Is., Vancouver Is., B.C. Bunsby Is., Vancouver Is., B.C. Orlebar Point, Gabriola Is., B.C. Orlebar Point, Gabriola Is., B.C. Whiffen Spit, Sooke, Vancouver Is., B.C. Whiffen Spit, Sooke, Vancouver Is., B.C. Arab Bay, Vancouver Is., B.C. Inner Bajo Reef, Nootka Is., Vancouver Is., B.C. Lawn Point, Vancouver Is., B.C. Lawn Point, Vancouver Is., B.C. Diana Is., Barkley Sound, B.C. T T UBC A64648 UBC A69423 UBC A69424 UBC A69425 UBC A69426 UBC A69427 UBC A69428 UBC A69429 UBC A69430 UBC A69431 WTU 248018 head of Rennell Sound, Queen Charlotte Is., B.C. Cape Suspiro, Alaska Cape Suspiro, Alaska Cape Suspiro, Alaska San Clemente Is., Alaska Sea Otter Sound, Alaska Sea Otter Sound, Alaska Sea Otter Sound, Alaska Sea Otter Sound, Alaska SeaOtter Sound, Alaska Mitchell Bay, San Jaun Is., Washington APPENDIX 2. PROCEDURE FOR EMBEDDING MATERIAL IN JB4 METHACRYLATE Fixation 2.5% glutaraldehyde in Sorensen's phosphate buffer (pH 7.2) (1:1) 4 h, room temperature 4-3 10 min washes in Sorensen's buffer 4-Dehydration 10%, 25%, 40%, 50%, 70%, 80%, 90%, 100%, 100%, 100% MeOH each step 10 min 10% - 80% in refrigerator, 90%, 100% at room temperature 4-Infiltration 1 part catalyzed solution A : 3 parts 100% MeOH 1 h 4-1 part catalyzed solution A : 1 part 100% MeOH 4 h 4-3 parts catalyzed solution A : 1 part 100% MeOH 8 h or overnight, refrigerator 4-100% catalyzed solution A overnight, refrigerator 288 / 289 4-Embedding 1 part solution B (activator) : 25 parts catalyzed solution A poured into mold , tissue arranged in mold, capped with a stub, hardened in refrigerator, overnight APPENDIX 3. PREPARATION OF TETRASPORANGIAL AND SPERMATANGIAL MATERIAL FOR TRANSMISSION E L E C T R O N MICROSCOPY Fixation 2.5% glutaraldehyde in Sorensen's phosphate buffer (pH 7.2) (1:1) t e t r a s p o r a n g i a l : 4 h, room temperature/spermatangial: 7 h, refrigerator 4-3 10 min washes in Sorensen's buffer 4-Osmication 1% OsO„ in buffer (1:1) t e t r a s p o r a n g i a l : 18-19 h, refrigerator/spermatangial: 16 h, refrigerator 4-3 10 min washes in buffer 4-Dehydration 10%, 25%, 40%, 50%, 70%, 80%, 90%, 100%, 100%, 100% MeOH each step 10 min 10% - 80% in refrigerator, 90%, 100% at room temperature 4-Infiltration 25%, 50%, 75%, 100% propylene oxide in MeOH 20 min, room temperature 4-290 / 291 Embedding 10% Spurr's epoxy resin in propylene oxide overnight, refrigerator, on rotator 4-25%, 40%, 55%, 70%, 80%, 90%, 100%, 100%, 100% Spurr's in propylene oxide each step \ day at room temperature on rotator, or overnight in refrigerator 4-Polymerization in Spurr's epoxy resin under 18 psi vaccuum \ day, then 600 min, 70°C, 18 psi 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items