Molecular Systematics of Red Algae. Jeffrey Craig Bailey Louisiana State University and Agricultural & Mechanical College

Home , Acrochaetiales, Ahnfeltia, Ahnfeltia plicata, Amphiroa, Bonnemaisonia, Bossiella

Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1996 Molecular Systematics of Red Algae. Jeffrey Craig Bailey Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Bailey, Jeffrey Craig, "Molecular Systematics of Red Algae." (1996). LSU Historical Dissertations and Theses. 6316. https://digitalcommons.lsu.edu/gradschool_disstheses/6316

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zed) Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MOLECULAR SYSTEMATICS OF RED ALGAE

A Dissertation Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

in The Department of Plant Biology

By Jeffrey Craig Bailey B.S., The University of North Carolina at Wilmington, 1990 M.A., Tho College of William and Mary, 1992 December 1996

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 9720330

This microform edition is protected against unauthorized copying under Title 17, United States Code.

UMI 300 North Zeeb Road Ann Arbor, MI 48103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS

My graduate advisor, Dr. Russell L. Chapman, is gratefully acknowledged for his guidance and support during the course of these studies. For critcal reviews of the manuscript I thank the members of my graduate committee: Dr. Meredith M. Blackwell, Dr. Mark S. Hafner, Dr. David J. Longstreth, Dr. Lowell E. Urbatsch, and Dr. Bradley C. Venuto. Dr. Blackwell is especially acknowledged for her encouragement and the interest taken in my studies. Special thanks are due Dr. William J. Woelkerling who freely shared his immense knowledge of the coralline algae, critically reviewed several manuscripts, and for his hospitality during my stay in Australia. Dr. Kevin G. Jones is thanked for his expert technical assistance. I am indebted to the following persons for helpful discussions of red algal biology and systematics: Dr. Sharon T. Broadwater, Dr. D. Wilson Freshwater, Dr. David J. Garbary, Dr. Max H. Hommersand, Dr. Curt M. Pueschel, Dr. Gary W. Saunders, and Dr. Joe L. Scott. Wilson Freshwater and Gary Saunders are especially thanked for their support, collaboration, and continued friendship. Research and travel was generously supported by the National Science Foundation, Louisiana Educational Quality Support Fund, Phycological Society of America, Sigma Xi Scientific Research Society, and the Department of Plant Biology at Louisiana State University.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS...... ii

LIST OF TABLES...... v LIST OF FIGURES...... vi

ABSTRACT...... viii CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW...... 1

SYSTEMATICS OF THE RHODOPHYTA...... 2

History...... 2 The early era ...... 2 The modern era ...... 7 Ultrastructure ...... 8 Molecular systematics ...... 13 Classification...... 20

BIOLOGY AND SYSTEMATICS OF THE CORALLINALES...... 23

Vegetative morphology ...... 23 Reproduction ...... 31 Ecology...... 40 Fossil corallines ...... 44 Taxonomic history and circumscription of the Corallinales ...... 45 Coralline systematics ...... 49

2 PHYLOGENY OF THE CORALLINALES BASED ON ANALYSIS OF SSU GENE SEQUENCES...... 55

Introduction...... 56 Materials and Methods ...... 60 Taxon sampling...... 60 DNA extraction ...... 60 PCR amplification and sequencing of SSU genes ...... 63 Sequence alignment and data analysis...... 64 Results ...... 66 Discussion...... 74

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sporolithaceae ...... 74 Melobesioicieae ...... 75 Corallinoideae ...... 79 Mastophoroideae, Metagonioiithoideae, Lithophylloideae, and Amphiroideae 81 Higher-level systematics of the Corallinales ...... 83 Summary...... 87 3 EVOLUTION OF PIT PLUG-ASSOCIATED FEATURES AND A RE-EVALUATION OF PHYLOGENETIC RELATIONSHIPS AMONG THE FLORIDEOPHY- CIDAEAN ORDERS OF RED ALGAE...... 89

Introduction ...... 90 Materials and Methods ...... 93 Results ...... 98 Discussion...... 102

4 GENERAL SUMMARY AND CONCLUSIONS...... 108 REFERENCES...... 114

APPENDIX A PERMISSION FOR REPRODUCTION OF DRAWINGS USED IN FIGURES 1.3 AND 1.4...... 140

B COLLECTION INFORMATION FOR CORALLINE SPECIES...... 143 C MORPHOLOGICAL CHARACTERS AND CHARACTER STATES FOR THE FAMILIES AND SUBFAMILIES OF CORALLINE ALGAE...... 145

VITA...... 147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

Table Page

1.1 Distribution of pit plug-associated features among the orders of red algae ...... 12

1.2 Classification of the red algae ...... 22 1.3 Suprageneric classification schemes devised for coralline algae since 1950 ...... 51

1.4 Intraordinal classification of the Corallinales according to Cabioch (1972,1988) ...... 52

1.5 intraordinal classification of the Corallinales according to Johansen (1981) and Harvey and Woelkerling (1995) ...... 54

2.1 List of 37 species of coralline red algae and number of SSU base positions determined for each ...... 61 2.2 Conceptacle positions for the genera of Corallinoideae included in this study...... 80

3.1 Sources of the SSU gene sequences for 70 red algal species included in the data alignment ...... 94

3.2 Coding of pit plug-associated features for the orders of red algae ...... 97

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

Figure Page 1.1 Cladogram depicting possible phylogenetic relation ships among 22 red algal orders ...... 14 1.2 Drawing of longitudinal section through the margin of the thallus of a coralline alga ...... 24

1.3 Non-geniculate coralline algae ...... 28

1.4 Geniculate coralline algae ...... 29 1.5 Schematic diagram illustrating the different types of genicula in the coralline algae ...... 30

1.6 Diagram illustrating female/carposporophytic development in the coralline red algae ...... 33

1.7 Line drawing of vertical median sections through mature (A) carposporangial and (B) spermatangial conceptacles of coralline algae ...... 34 1.8 Line drawing of vertical median section through the mature sporangial conceptacle of a coralline alga...... 38

2.1 Strict consensus of 72 equally most parsimonious trees 9CI=0.56; Rl=0.77) resulting from a cladistic analysis of SSU gene sequences for 37 species of coralline algae ...... 67

2.2 Single most parsimonious phylogram (Cl=0.70; Rl=0.60) depicting phylogenetic relationships inferred among 13 melobesioid species ...... 69 2.3 Strict consensus of 18 equally most parsimonious trees (Cl=0.84; Rl=0.71) inferred from SSU gene sequences for 13 corallinoidean species ...... 71

2.4 Simplified cladogram depicting relationships inferred among the families and subfamilies of coralline algae included in this study ...... 73

3.1 One of four equally most parsimonious trees (Cl=0.41 ; Rl=0.69) inferred from an analysis of the SSU gene sequences for 70 species of red algae ...... 99

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.2 Cladogram(s) depicting relationships among the orders of red algae inferred from a cladistic analysis of SSU gene gene sequence and pit plug data ...... 101

3.3 Character state optimization of pit plug data onto a tree constrained to the result of Ragan etal. (1994) in which the Ahnfeltiales is sister ...... 107

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

Phylogenetic analyses of nucleotide sequences for the nuclear-encoded small subunit ribosomal RNA (SSU) gene were used to test current hypotheses of relationships within the Corallinales and among the florideophycidaean orders of red algae. Analyses of SSU gene sequences indicate that the order Corallinales includes four major lineages. The Sporolithaceae is resolved as the earliest- diverging lineage within the order and forms the sister group to a monophyletic Corallinaceae. The molecular data also support the monophyly of the subfamilies Melobesioideae and Corallinoideae. The latter subfamily is inferred as sister to a fourth lineage including non-geniculate and geniculate species classified in the subfamilies Mastophoroideae, Metagoniolithoideae, Lithophylloideae, and Amphiroideae. Genicula are inferred to have arisen independently in three separate lineages and are not homologous structures. These results indicate that the common taxonomic practice of separating corallines on the basis of the presence or absence of genicula does not accurately reflect the evolutionary history of the group. Cladistic analysis of SSU gene sequences indicates that the Florideophycidae is a monophyletic taxon comprised, roughly, of two large sister clades. One clade includes all orders possessing pit plugs with two cap layers. Within this assemblage, the orders Corallinales and Rhodogorgonales are inferred to be closely related. The second florideophycidaean clade includes, with the exception of the Gelidiales, those orders in which pit plugs are characterized by the absence of cap layers and the presence of a cap membrane. These results support Pueschel's (1994) contention that: 1)

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. "naked" pit plugs are the plesiomorphic type in the Rhodophyta, 2) all orders possessing pit plugs with two cap layers are derived from a common ancestor, 3) outer cap layers are homologous structures, and 4) domed outer caps and cap membranes are derived features within the two cap layer lineage. Results also indicate that although the inner cap layers found in the Bangiales, Hildenbrandiales, and the two-cap-layer lineage may be homologous, inner caps in the Gelidiales are independently derived.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 SYSTEMATICS OF THE RHODOPHYTA

History. In terms of numbers of species, their range of forms, life histories, and reproductive development, the red algae are the most diverse group of marine seaweeds (Kraft 1981; Woelkerling 1990). The history of the systematic study of red algae can be divided roughly into two separate periods, the early and modern eras. In the following review, each period is treated separately with emphasis on those discoveries that have influenced most our current understanding of the biology and phytogeny of red algae.

The early era. Early developments in the systematic study of red algae have been reviewed previously by Papenfuss (1955). Much of the following synopsis of that era is based on Papenfuss's account and additional details provided by Garbary and Gabrielson (1990) and Dixon and Irvine (1977). The nomenclature of the majority of algae begins with Linnaeus' (1753) Species Plantarum. In the current sense, Linnaeus recognized four genera of algae ( Conferva, Ulva, Fucus, and Chara) which he classified with ferns, mosses, and fungi in the class Cryptogamia. Most algae described during the next half century were attributed to one or another of Linnaeus' original genera. Stackhouse was the first phycologist to break formally with the Linnaean tradition. Working for the most part on marine species classified within Fucus from the British Isles, Stackhouse (e.g., 1816) divided the genus into 67 new taxa based on comparative gross morphological features. Lamouroux (1813) established a number of new genera and was the first to propose an ordinal level of classification for the aigae based, in part, on color. Lamouroux removed certain red algae from other superficially similar morphological groups and placed them in a separate "ordre," the Floridees.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Nevertheless, the distinction between other red, green, and brown algae was not made clear in this scheme of classification (Lamouroux 1813; C. Agardh 1817). Harvey's (1836) treatment of the algae subsequently presented a more complete and natural separation of the algae on the basis of color and other characteristics. The class Rhodophyceae was erected by Ruprecht (1851, 1855). Previous and present day classifications of the red algae depend to great extent on observed variation among life histories, reproductive, morphological, and anatomical features (Kraft 1981; Hommersand & Fredericq 1990). Male reproductive structures (spermatangia) were described by Ellis (1767) and, in Polysiphonia, by C. Agardh (1828) who termed them "antheridia." Stackhouse (1801) was the first to emphasize that the life history of most species of red algae include two phases bearing distinct spore-bearing structures (the cystocarp and tetrasporangium). Before his conclusions were published, Stackhouse's contemporaries were content to ascribe each life history phase to separate species. Later, after Stackhouse's proposal was accepted generally, there ensued a lengthy debate over the roles these structures played in the life history of red algae. Decaisne (1842a) concluded that the tetrasporangium was the site of "typical" (i.e., sexual) reproduction in red algae and that the cystocarp represented a structure which functioned as a means of vegetative propagation. Harvey (1849), on the other hand, suggested that the development of the cystocarp was the result of sexual processes whereas the tetrasporangium produced vegetative "gemmules." Nageli (1847) regarded the tetrasporangium as representing the female and the "antheridium" (sensu C. Agardh 1828), the male sex organs of red algae. Concurring with Decaisne, Nageli concluded that the cystocarp was involved only in asexual reproduction. Ruprecht (1851)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. likened tetraspores and carpospores of red algae to the pollen and seeds of land plants, respectively. Sexual reproduction in a number of red algae (e.g., Ceramium, Dudresnaya, and Nemalion) was first demonstrated conclusively by Bornet and Thuret in 1866 and 1867. Bornet and Thuret discovered that the female egg (carpogonium) was borne terminally on a specialized filament, the carpogonial branch, and was not released from the thallus prior to fertilization. Instead, they observed spermatia attached to the trichogyne of the carpogonium, hypothesized that fertilization occurred in situ, and concluded that sexual reproduction in red algae was oogamous. The actual fusion of spermatial and carpogonial nuclei was first observed by Wille (1894). The complexities of post fertilization processes in red algae were first elucidated by Schmitz (1883). These processes result in the development of a diploid carposporophytic phase that is borne on, and is nutritionally dependent upon, the female plant. Schmitz discovered that the location of zygote formation and the site of its ultimate growth and differentiation are usually separate. Schmitz (1883) observed that in some red algae the zygote nucleus was transferred to a neighboring cell, which he termed the auxiliary cell, and that gonimoblast filaments originated from the auxiliary cell and not the carpogonium. Oltmanns (1898) later showed that the diploid zygote nucleus transferred from the carpogonium to the auxiliary cell did not fuse with the hapolid nucleus of the auxiliary cell as Schmitz had postulated. Oltmanns concluded correctly that the haploid auxiliary cell nucleus plays no role in the reproductive development of the cystocarp. Yamanouchi (1906 a,b) correllated the nuclear cycle with the isomorphic haploid gametophytic and dipolid tetrasporophytic life history phases of Poiysiphonia, thus finally establishing the tetrasporangium as the site of meiosis and elucidating completely the first life history of any red alga. Red algal life

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. histories departing from the triphasic Polysiphonia-Xype were subsequently documented by, among others, Svedelius (1915) and Kylin (1916). Beginning with C. Agardh (1820,1824,1828), the classification of red algae was increasingly influenced by finer scale and more detailed morphological and anatomical investigations. These studies focused not only on vegetative features, but also on differences among the reproductive structures of red algae (Greville 1830; J. Agardh 1842; Harvey 1849). Many taxonomic characters first introduced into the literature during this period remain the conerstones of modem classifications of the red algae at higher taxonomic levels. For example, J. Agardh (1842) first emphasized the manner of division of tetrasporangia (tetrahedral, cruciate or zonate) and Oltmanns (1904) noted that families within some orders could be separated based on the uniaxial or multiaxial nature of the thallus. Kylin (e.g., 1932) recognized as taxonomically informative the fact that cystocarps are embedded in the thallus of some species but are emergent in others. With Schmitz’s paper of 1883, differences in the development of the cystocarp, particularly the timing, place of formation, and function of the auxiliary cell, among different taxa of red algae were considered especially informative for the purpose of classification (Papenfuss 1955; Kraft 1981). Based on comparisons of the reproductive development of the cystocarp in a number of species, Schmitz (1892) divided the Florideophycidae into four orders, the Nemalionales (syn. Nemaliales), Gigartinales, Rhodymeniales, and Cryptonemiales. Schmitz' classification of the Florideophycidae was later emended by Oltmanns (1904) and Kylin (1923) who erected the orders Ceram iales and Gelidiales, respectively. The genera Bangia and Porphyra were originally placed in the green algae by Harvey (1836) but were transferred formally to the Rhodophyceae by Berthold (1882). Later, Schmitz (in Schmitz & Hauptfleish 1897) tentatively assigned a number of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. other morphologically simple organisms, including unicellular, palmelloid, and filamentous species, to the Rhodophyceae but noted that these species were very different from those classified in the Florideophycidae. Kylin (1937) placed all unicelluar species within this enigmatic group in the order Porphyridiales and all other multicellular members of the group in the Bangiales. The first comprehensive classification of this assemblage of species, the Bangiophycidae, was presented by Skuja (1939) who, in addition to recognizing Kylin's two orders, also established the orders Goniotrichales and Compsopogonales. Skuja's classification of the Bangiophycidae has been emended by Feldmann (1955,1967) and, most recently, by Garbary etal. (1980) who have subsumed the Goniotrichales in the Porphyridiales. Students of red algal systematics are especially indebted to the Swedish phycologist Harald Kylin. Kylin was a productive scientist and keen observer who made innumerable original discoveries. Perhaps more importantly, he also possessed the ability to logically synthesize and organize large amounts of systematic evidence. In a long series of papers, Kylin (e.g., 1928,1930,1932, 1956) continually refined the classification of the red algae into one that more accurately reflected natural relationships among taxa within the group. In addition to other morphological and anatomical features, Kylin considered the presence or absence of generative auxiliary cells and their mode, location, and time of formation (before or after fertilization) taxonomically informative at the ordinal level. In this scheme, families within each order were delimited by, among other important features, the manner of division of tetrasporangia and whether the thallus was uni- or multiaxial. Kylin's (1932,1956) system of classification remained the standard for nearly 50 years and his monograph of the red algae has yet to be updated in the modern era. In recent years the number of red algal taxa recognized at the levels of order and family have risen

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sharply (Kraft 1981; Woelkerling 1990; Ragan etal. 1994; Fredericq etal. 1996). Even allowing for these taxonomic changes, it remains clear that the present day system of classification for the red algae is based principally on Kylin's earlier outline. As stated by Kraft (1981, p. 6), for most red algal taxonomists, "Kylin's (1956) compendium of red algal genera, families, and orders remains the single most authoritative work on the group and the point of departure for any revisionist attempts to alter the classification of red algae at or above the genus level."

The modern era. Although certainly not mutually exclusive, systematics studies of red algae in the modem era can be divided into three broad catagories. The first catagory includes comparative studies of red algal anatomy, reproductive development, and life histories - studies which follow the tradition established by Schmitz, Oltmanns, Kylin and others. In contemporary systematics parlance such approaches are often referred to as "traditional" or "organismal." Recent reviews focusing on these type studies are numerous (for a review see, Cole & Sheath 1990) and, therefore, will not be discussed in detail here. Infomation on the vegetative growth and organization of the thalli of red algae can be found in Bold & Wynne (1978), Lobban & Wynne (1981), and Coomans and Hommersand (1990). Life history variation among red algae, and its taxonomic and evolutionary significance, have been reviewed by Searles (1980), West and Hommersand (1981), Guiry (1987), and Hommersand and Fredericq (1990). The intricacies of female and carposporophytic development in different red algal taxa are summarized by Hommersand and Fredericq (1990).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The second and third catagories of studies in the modem era, in contrast, are delimited by technological advances in the life sciences, particularly the advent of electron microscopic and molecular biological techniques. Recent advances in these areas of research are summarized below.

Ultrastructure. The cells of red algae are typically eucaryotic in most respects, but can be easily distinguished by plastid-associated features including unstacked thylakoid membranes and the presence of phycobilin pigments arranged in phycobilisomes. Other distinguishing features include the absence of centrioles and flagella and the storage of photosynthate (floridean starch) in the cytosol. One of the most interesting aspects of red algal cells are the consistent and unorthodox associations observed between the cis (forming) face of Golgi bodies and other organelles (Pueschel 1990). In most eucaryotes Golgi bodies are typically associated with the endoplasmic reticulum (ER). In contrast, the dictyosomes of red algae are more commonly associated with mitochondria. Although Golgi body-mitochondrial associations are most conspicuous and common, Golgi bodies have also been observed in association with other organelles including the ER, nuclei, and plastids (Pueschel 1990). The physiological basis for Gogi body-organelle associations in red algae is not known. Ultrastructural studies of the flagellar apparatuses of motile cell stages in other algae have contributed substantially to a better understanding of the taxonomy and phylogeny of these organisms (Moestrup 1982; Mattox & Stewart 1984; Green et al. 1989). Because both vegetative and reproductive cells of red algae lack flagella entirely, ultrastructural studies within the group have focused instead on other features that are of taxonomic importance at higher

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. levels of classification within the Rhodophyta (Duckett & Peel 1978). For example, electron microscopic investigations have allowed for a more detailed and accurrate understanding of the cellular interactions which take place during the development of the cystocarp (Duckett & Peel 1978; Wetherbee 1980; Broadwater & Scott 1982; Delivopoulos & Kugrens 1985). Features of particular taxonomic importance in red algae are those associated with cell division and pit plugs. Because it is difficult to fix and find dividing cells in red algae, ultrastructural studies of cell division are lacking for most taxa. During karyokinesis in red algae, the nuclear envelope remains intact except at the poles. Mitotic cell division in red algae follows one of two basic patterns, termed the "Polar Gap" (PG) and "Polar Fenestrations" (PF) types. The PG and PF type patterns are differentiated mainly by characteristics of the prophase - metaphase polar region (Broadwater etal. 1993). The PG type pattern is characterized by the complete break-down of the nuclear envelope at the spindle poles. In contrast, the nuclear envelope does not break-down entirely at the spindle poles in species characterized by the PF type pattern. In the latter species the nuclear envelope is fragmented or fenestrated in the polar regions. Detailed comparisons of the PG and PF type patterns of cell division are provided by Scott & Broadwater (1990), Broadwater etal. (1993,1995), and Klepacki et al. (1995). The PG type pattern has been found in a number of unicellular red algae classified in the order Porphyridiales as well as in Batrachospermum ectocarpum (Scott 1983). In contrast, the PF type pattern of cell division is characteristic of species attributed to the orders Ceramiales, Compsopogonales, Corallinales, and Gigartinales. In many other details, cell division in taxa attributable to either the PG or PF groups varies significantly (Scott & Broadwater 1990). Red algae lack centrioles. Non-centriolar nucleus-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. associated organelles (NAOs) are present that are similar to those found in non-flagellated fungi (Heath 1980 a,b, 1981). Although present in interphase cells (Scott etal. 1981), NAOs are most conspicuously associated with the spindle poles during karyokinesis. The NAOs of most red algae are composed of a pair of short, hollow, cylindrical rings (generally refered to as "polar rings" [McDonald 1972]) that differ in size. In relation to one another, the two rings are usually configured such that the smaller ring lies atop the larger ring. However, in Batrachospermum the smaller ring is tucked inside the larger ring (Scott 1983). A structurally distinct bipartite, non-hollow type of NAO has been observed in Porphyridium (Schornstein & Scott 1982). During prophase in ceramialean species, the distal (smaller) ring is attached to the proximal (larger) ring which is, in turn, attached to the nuclear envelope by fine struts. The size of NAOs also varies among taxa. Smaller NAOs are characteristic of the bangiophycidaean orders Bangiales, Compsopogonales, and Porphyridiales. The behavior of NAOs during karyokinesis is also variable. In most red algae the movement, timing, and association of the NAO with other organelles (particularly the nuclear envelope) follows the pattern found in the genera Polysiphonia, Dasya, and Lomentaria (Scott etal. 1980; Phillips & Scott 1981; Davis & Scott 1986). In contrast, the NAO cycle observed in Pophyridium is different from that found in other red algae and other eucaryotes (Schornstein & Scott 1982). Although NAOs are consistent elements of the spindle poles, it is not known if these structures play a role in nucleating the microtubles of the spindle apparatus. Spindle microtubules abut in either an osmiophilic region of "NAO- associated material" in which the NAO may be embedded or near the inner membrane of the nuclear envelope. Microtubules have not been observed

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 attached to NAOs. Ultrastructural observations suggest that NAOs may be replicated in a semiconservative fashion (Heath 1981). Differences in the patterns of cell division and the size, shape and behavior of NAOs suggest that these characters are potentially of taxonomic and phylogenetic value. Thus far, however, cell division has been investigated in species representing only six orders of red algae and the extent to which features of mitotic cell division may vary within orders is largely unknown. These data suggest that cell division characters may in the future prove to be valuable taxonomic markers for delineating red algal orders (Broadwater et al. 1993) and already indicate that there are significant differences between putatively "primitive" and "advanced" red algal taxa. With the exception of a few unicellular species and the blade-forming phases of some bangiophytes, the thalli of most other red algae are constructed of filaments. In most red algae the process of cytokinesis is completed only during the formation of spores. Division of all other cells is initially incomplete resulting in a septal pore with cytoplasmic continuity between daughter cells. Soon after the septal pore is formed it is filled by a glycoproteinaceous substance, forming a pit plug (Pueschel & Cole 1982). Brown and Weier (1970) noted that the deposition of pit plugs is associated with the activity of local ER and dictyosomes. The fine structure of red algal pit plugs is variable. Pit plugs can be characterized by the presence or absence of osmiophilic cap layers (0,1 or 2) and a cap membrane. In taxa in which they are present, outer cap layers can be divided into two types: those that are thin and plate-like, and those that are domed. Aspects of the fine structure of pit plugs in most red algal lineages have now been determined by various techniques (reviewed in Pueschel 1990; Table 1.1). Pueschel and Cole (1982) first demonstrated conclusively the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1.1. Distribution of pit plug-associated features among the orders of red algae. Data taken from Pueschel (1990,1994).

Order* Cap layers Cap membrane Outer cap morphology

Rhodochaetales 0 A Compsopogonales 0 A Bangiales 1 A Acrochaetiales 2 domed /thin P Nemaliales 2 thin P Palmariales 2 thin P Rhodogorgonales 2 domed A Corallinales 2 domed A Batrachospermales 2 domed A Hildenbrandiales 1 P Gelidiales 1 P Ahnfeltiales 0 A Gracilariales 0 P Bonnemaisoniales 0 P Gigartinales 0 P Halymeniales 0 P Sphaerococcales 0 P Dumontiales 0 P Plocamiales 0 P Rhodymeniales 0 P Ceram iales 0 P

P=present, A=absent * Pit plugs are lacking in the order Porphyridiales.

taxonomic utility of pit plug variation at higher levels of classification in red algae. Based on their observations and other data, Pueschel and Cole erected the orders Batrachospermales and Hildenbrandiales and suggested that the Bonnemaisoniales and Corallinales deserved recognition. The hypothesis (Pueschel 1994) that pit plug-associated features are phylogenetically, as well as taxonomically, informative is explored in Chapter 3 of this dissertation.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pueschel and Cole's (1982) ultrastructural survey represents one of the more important contributions to the study of red algal systematics in the last 50 years. This study, and those of Guiry (1974,1978), advanced and altered considerably for the first time the essentially Kylinian classification of the red algae. These studies ushered in a new era of understanding, beginning in the late 1970's, of red algal phytogeny and taxonomy unprecedented since the end of the 19th century (Garbary & Gabrielson 1990).

Molecular systematics.Molecular biological techniques have rejuvenated systematics studies by extending the range of questions being studied and providing detailed information that was previously unobtainable. In addition, molecular biology has opened many new fields of investigation in which certain algae (e.g., Chlamydomonas, Acetabularia, and Fucus) play important roles as model organisms. Molecular systematics studies of red algae have burgeoned in the last decade and have significantly contributed to a better understanding of the evolutionary history of the group. Most molecular systematics studies of red algae have utilized nucleotide sequence information from the nuclear encoded rRNA cistron, particularly the small-subunit rRNA (SSU) gene, or the plastid encoded gene for the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase (rbcL). Sequence data for a limited number of other genes of nuclear or plastidial origin have been used to provide phylogenetic estimates for the relationship of red algae to other eucaryotic lineages as well (e.g., Liaud etal. 1994; Bouget etal. 1995). The following sections summarize some aspects of published phytogenies for red algae, emphasizing, at the ordinal level, those based on SSU and rbcL gene sequences. A cladogram depicting possible phylogenetic lines of descent among 22 red algal orders is presented in Figure 1.1. This figure represents a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sphaerococcales Dumontiales Gigartinales Ceramiales Gelidiales Halymeniales

Rhodymeniales Bangiophycidae Florideophycidae Plocamiales Gracilariales Bonnemaisoniales Ahnfeltiales Hildenbrandiales Rhodogorgonales Corallinales Batrachospermales Nemaliales Palmariales Acrochaetiales Bangiales Rhodochaetales Comsopogonales Porphyridiales

Figure 1.1. Cladogram depicting possible phylogenetic relationships among 22 red algal orders.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. consensus of the results obtained from phylogenetic investigations of both SSU and rbcL sequence variation. Figure 1.1 highlights both the substantial progress that has been made in our understanding of red algal evolution as well as areas requiring further study. With the exception of a few genera (notably Bangia and Porphyra), our knowledge of the biology and taxonomy of bangiophycidaean species is, compared to that of other red algae, rudimentary. The Bangiales have long been considered a natural group (Skuja 1939), but the taxonomic affinities of certain species presently classified in other bangiophyte orders are not clear. The Porphyridiales, for example, has historically served as a catch-all for morphologically simple (pseudo-) filamentous, palmelloid, or unicellular forms whose relationships to one another and other red algae are uncertain (Kylin 1937; Skuja 1939; Garbary etal. 1980) Unfortunately, sequence data for a broad sampling of bangiophyte taxa are lacking and phylogenetic relationships among the bangiophycean orders Porphyridiales, Compsopogonales, and Bangiales are not clearly resolved based on molecular data gathered thus far. In both SSU and rbcL trees representative bangiophycean taxa terminate long branches and support for the relative branching order of these taxa is poor (Freshwater etal. 1994, Ragan etal. 1994). Inferences from SSU sequence data suggest that the Bangiophycidae is polyphyletic. The Porphyridiales is inferred to be a non-monophyletic taxon with the unicellular genera Dixoniella and Rhodella diverging after species having either filamentous or blade-like thalli (Ragan etal. 1994). This result implies that some unicellular red algae may be derived from ancestors of greater morphological complexity (Fritsch 1945; Lee 1974; Garbary etal. 1980) and argues against the notion that unicelluar forms are necessarily primitive members of the phylum (e.g., Dixon 1973). In general, molecular data support the notion that the red algae can be

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. divided into two more or less distinct groups; the early diverging bangiophyte and the relatively advanced florideophyte lineages. The ancient (Campbell 1980) bangiophycean genus Porphyra presents an interesting case-study in evolutionary biology. The taxonomy of the genus is based mainly on certain aspects of relatively simple, one or two-cell thick, gametophytic blades. The blades of different species of Porphyra exhibit little variation and are nearly indistinguishable. Nucleotide sequences and other types of molecular data (Lindstrom 1993; Lindstrom &Cole 1992a,b; Stiller & Waaland 1993,1996) have, however, demonstrated a large amount of genetic divergence among species within the genus. These observations seem to indicate that in Porphyra, processes regulating morphological and molecular evolution are largely uncoupled. The Florideophycidae is inferred to be a monophyletic assemblage of taxa that are, in relative terms, distantly related to the Bangiophycidae. The branching order of florideophyte lineages is not fully resolved (Figure 1.1). Ribosomal DNA sequence analyses (see Chapter 3) clearly support earlier indications (Ragan et al. 1994) that the basal lineages of the florideophyte clade include the orders Corallinales, Palmariales, Acrochaetiales, Nemaliales, Batrachospermales, and Rhodogorgonales. A relationship among these orders is supported by ultrastructural evidence as well; these taxa all possess pit plugs with two cap layers (Pueschel 1990,1994; Pueschel etal. 1992a). The large order Gigartinales (Kraft & Robins 1985) is resolved as a polyphyletic taxon by both SSU and rbcL sequence analyses. Based on these, and additional molecular and organismal data (Fredericq et ah 1996; Saunders & Kraft 1994, 1996), four formerly gigartinalean lineages have been separated and elevated to the rank of order as the Dumontiales, Plocamiales, Halymeniales, and Sphaerococcales. These revisions have resulted in a putatively more natural

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17 concept of the Gigartinales, but relationships among the remaining families within the order are not clear. Phylogenetic hypotheses robustly supported by molecular data challenge several fundamental tenets of red algal evolution that were widely accepted in the preceding half century. The hypotheses that the Ceramiales represents the most highly derived red algal taxon (Kylin 1932,1956; Bold & Wynne 1978; Silva & Johansen 1986; Gabrielson & Garbary 1987; Garbary & Gabrielson 1990) and that procarpy is a derived character in red algae are not consistent with molecular data. Molecular systematics studies also indicate that at higher (ordinal) taxonomic levels some morphological, life history, and reproductive features are not phylogenetically informative. For example, sequence data indicate that the Gelidiales and Nemaliales are not closely related (Dixon 1973; Abbott & Hollenberg 1976; Dixon & Irvine 1977), although life histories in most species in the two taxa are wholly comparable and both orders are uniaxial. Indeed, molecular data have provided convincing support for the conclusion, reached independently by some morphologists (e.g., Hommersand 1963; West & Hommersand 1981; Hommersand & Fredericq 1990; Masuda etal. 1996), that parallel and convergent evolution of life history strategies and female reproductive systems are not uncommon in the Florideophycidae. There seems, in red algae at least, little hope that life history variation can be used to infer phylogenetic relationships (West & Hommersand 1981). Sequence data do not clearly support the widely held notion, based on analyses of red algal structure, biochemistry, and reproductive features, that the Acrochaetiales is the most primitiveflorideophycidaean lineage (Garbary 1978; Gabrielson etal. 1985; Gabrielson & Garbary 1987; Magne 1989; Garbary & Gabrielson 1990), but do indicate that the group diverges early in the florideophycean radiation. Ultrastructural (Pueschel & Cole 1982) and molecular evidence (Freshwater et

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. al. 1994; Saunders etal. 1995; Chapter 3) suggest that the Acrochaetiales is paraphyletic. Other authorities have argued that Rhodochaete parvula, the only species classified in the order Rhodochatetales, may be more like the last common ancestor of the Rhodophyta than any other extant species. This hypothesis has yet to be tested with molecular data; nucleotide sequences for any gene of Rhodochaete parvula are presently unavailable. Sequencing studies have also been used to address systematic questions at lower taxonomic ranks. Detailed phylogenetic hypotheses of relationships among genera and/or species within the orders Corallinales (Bailey & Chapman 1996), Gigartinales (Hommersand etal. 1994; Saunders & Kraft 1994, 1996; Fredericq etal. 1996), Gracilariales (Rice & Bird 1990; Bird et al. 1990, 1992a,b) and Gelidiales (Freshwater & Rueness 1994; Freshwater et al. 1995) have been presented. Robust phylogenies at the specific level are important elements of studies focusing on adaptation, character evolution, coevolution, and phytogeography (e.g., Hommersand etal. 1994; van Oppen et al. 1995a,b; Bailey & Chapman 1996; Goff etal. 1996; Pakker etal. 1996). For example, the biology of, and systematic relationships between, red algal hosts and their parasites (which are also red algae) have been the subject of numerous studies (e.g., Fan 1961; Goff 1982; Goff & Hommersand 1982; Fredericq etal. 1989; Goff & Zuccarello 1994). Relationships among some red algal hosts and parasites have been investigated recently by comparison of ribosomal DNA sequences (Goff etal. 1996). These data confirm earlier studies, based on anatomical and reproductive features, suggesting that hosts and their parasites are closely related species. The observation that hosts and their parasites are sister taxa is intriguing, and supports the notion that these specialized, host-specific parasites (adelphoparasites) may have evolved directly from their hosts (Goff et al. 1996). The adelphoparasitc mode of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. speciation is not otherwise widely known and aptly demonstrates the complex nature of those processes that can result in the formation of new species. Other molecular techniques, including restriction fragment length polymorphism (RFLP) analyses of plastid and nuclear DNAs, randomly amplified polymorphic DNAs (RAPDs), and polymerase chain reaction (PCR)- mediated detection of variable length sequences have been used to assess polymorphism within the red algae (Goff & Coleman 1988; Scholfield etal. 1991; Steane etal. 1991; Maggs etal. 1992; Bhodayefa/. 1994; Dutcher& Kapraun 1994). These methods provide useful data that are appropriate for species and population level investigations and such techniques are becoming common tools used by researchers involved in alpha-taxonomic studies of the algae. Molecular data also provide the means to address other basic and applied biological questions in the red algae. For example, Parsons et al. (1990) used plastid DNA RFLP analysis to confirm the identity of uncoupled, heteromorphic life history phases for Gymnogongrus devoniensis. Patwary and co-workers (Patwary etal. 1993; Patwary & van der Meer 1994) have successfully used RAPD banding patterns to assess heterosis quantitatively in laboratory-induced hybrids of the commercially valuable red alga Gelidium vagum. Brief mention is warranted here concerning the evolutionary relationships of the red algal "host" cell to other eukaryotic lineages. Because red algae share chlorophyll a and phycobilin photosynthetic pigments with some eubacteria (cyanobacteria) and lack flagella and centrioles at any stage of their life histories, it has been proposed that red algae represent the most primitive extant eukaryotic lineage. This hypothesis is soundly rejected by molecular data. Red algae are instead inferred to be a member of the "eukaryotic crown" group (Knoll, 1992), including the chlorobionta, chromistans, fungi and animals

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Bhattacharya etal. 1990; Hendriks etal. 1991; Cavalier-Smith 1993; Bouget et al. 1995; Liaud etal. 1994; Ragan & Guttell 1995). In the cited studies, the divergence of the red algal lineage does not appear to significantly pre-date that of other crown group taxa. The evolutionary position of rhodophytes in eukaryotic SSU gene trees (see refs, above) leaves little doubt that ancestors of the group possessed flagella. Apparently, flagella were lost early in red algal evolution. Molecular data (e.g., Bhattacharya etal. 1990; Hendriks etal. 1991; Cavalier-Smith 1993) also do not support the hypothesis that the marine red algae are the progenitors of terrestrial ascomycetous fungi (the so-called “floridean hypothesis:" Denison & Carroll 1966; Kohlmeyer 1975; Demoulin 1975,1985). Evidence for a specific relationship between red algae and fungi is lacking, and recent molecular studies (e.g., Hasegawa etal. 1993; Wainright etal. 1993; Nikoh etal. 1994), which are supported by biochemical data as well, indicate that fungi are more closely related to animals than they are to plants or protists. Ragan and Gutell (1995) have shown that the hypothesis that red algae and chlorobionta (green algae + land plants) are sister groups cannot be rejected on the basis of SSU gene sequence comparisons. A growing body of evidence from additional sequence data, intron-location and characteristics of nuclear and mitochondrial protein-coding genes also supports a possible relationship between the red algae and chlorobionta (Ragan & Gutell 1995; Zhou & Ragan 1996). This hypothesis remains to be explored further.

Classification. Below, a brief and up-to-date account of the classification of the Rhodophyta is provided. Emphasized here, first, are those features that distinguish the red algae from other eukaryotic lineages. Second, the diversity and higher-level classification of the group are summarized.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The Rhodophyta can be distinguished from other eukaryotic phyla by a combination of at least five features, including: 1) the lack of flagellated cells during any life history phase, 2) products of photosynthesis stored as floridean starch in the cytoplasm, 3) presence of the accessory photosynthetic pigments phycoerythrin, phycocyanin, and allophycocyanin, 4) plastids with unstacked thylakoid membranes, and 5) the absence of an external membrane around the plastid that is continuous with the endoplasmic reticulum (cf. chromophyte algae; Woelkerling 1990). The Rhodophyta includes a single class, Rhodophyceae, and ca. 2500 - 6000 species (Woelkerling 1990) that are classified in approximately 690 genera and 75 families. The red algae have been divided historically into two subclasses, the Bangiophycidae and Florideophycidae (Kylin 1956; Cole & Conway 1975; Bold & Wynne 1978). As many as 23 orders are recognized on the basis of anatomical, reproductive, ultrastructural, and other features (Woelkerling 1990; Fredericq etal. 1996). The classification of the Rhodophyta is summarized in Table 1.2.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1.2. Classification of the red algae. Modified and updated from Woelkerling (1990).

Estimates of numbers of included taxa

Taxon Families Genera

Phylum Rhodophyta Class Rhodophyceae Subclass Bangiophycidae Order Bangiales 1 2 -3 Compsopogonales 3 10 Porphyridiales 2 18-24 Rhodochaetales 1 1 Subclass Florideophycidae Order Acrochaetiales 1 1-7 Ahnfeltiales 1 1 Batrachospermales 3 9 Bonnemaisoniales 2 9 Ceramiales 4 325 Corallinales 2 37-43 Dumontiales 4 43 Gelidiales 1 9 Gigartinales 29 8 Gracilariales 2 10 Halymeniales 2 26 Hildenbrandiales 1 2 Nemaliales 2 20 Palmariales 2 12 Plocamiales 2 3 Rhodogorgonales 1 2 Rhodoymeniales 3 40 Sphaerococcales 5 11

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23

BIOLOGY AND SYSTEMATICS OF THE CORALLINALES

Vegetative morphology. Like most other florideophycean red algae, the thallus of corallines is composed of aggregated filaments (Figure 1.2). Thalli are multiaxial and pseudoparenchymatous. With the exception of a few species classified in the Delesseriaceae, corallines are the only red algae that possess both apical and intercalary meristems. In a few species apical cells divide synchronously in and produce regular (coaxial) tiers of medullary cells. In most other species, however, apical cells do not divide synchronously and regular tiers of cells are absent in the medullary cell region. Cells of the medullary region contain few plastids or lack plastids entirely, and may serve as a sink for photosynthate (Cabioch 1970; Johansen 1976; Giraud & Cabioch1977; Millson & Moss 1985). Depending on the orientation of the alga to the substratum, apical cell divisions increase the length of the alga in either the horizontal or vertical planes. Filaments arching toward the thallus surface are differentiated into photosynthetically active cortical and epithallial cell regions. Intercalary meristematic subepithallial initial cells divide transversely to produce epithallial cells distally and new cortical cells proximally. The intercalary meristem increases the thickness or girth of the alga. Corallines are, compared to other seaweeds, notoriously slow-growing. On average, the rate of extenstion for coralline algae is between 1 and 4 mm per month (Johansen & Austin 1970; Colthart & Johansen 1973; Adey & Vassar 1975; Matsuda 1989; Sartoretto 1994). Because they are produced from intercalary meristematic cells not found in most other red algae and are physiologically distinct (see below), epithallial

* cells can be considered unique to the Corallinales. The epithallial cell layer

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1.2. Drawing of longitudinal section through the margin of the thallus of a coralline alga. A=apical initials, C=cortical cells, M=medullary cells, l=intercalary meristematic cells (subepithallial initials), F=cell fusions. Epithallial cells are shaded. (Re-drawn and modified from Kylin 1956).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. may be uni- or multistratose. The distal (outermost) cell wall of epithallial cells is uncalcified. In all corallines critically examined, the epithallial cell layer is periodically sloughed off, usually in sheets (e.g., Johansen 1969a; Keats etal. 1993). Previous authorities have suggested that the process of epithallial cell sloughing is an antifouling mechanism (Johnson & Mann 1986), but recent ecological and electron mircroscopic investigations do not support this hypothesis convincingly (Keats etal. 1993,1994; Pueschel & Miller 1996). With the exception of only a few specialized cell types, cells of red algae generally

remain in the G 2-phase of the cell cycle (Goff & Coleman 1990). Once formed, epithallial cells never enter the DNA replication (S-phase) of the cell cycle or divide (Pueschel & Miller 1996; Bailey & Scott, unpubl. observations). The ploidy of gametophytic and sporophytic epithallial cells never exceeds basic 1C or 2C values of DNA content, respectively. Ultrastructural observations have demonstrated that prior to sloughing or senescence cytoplasmic organelles and inclusions are mobilized and translocated to cells (epithallial or cortical) below the layer of sloughing (Giraud & Cabioch 1983; Millson & Moss 1985; Pueschel 1996; Pueschel & Miller 1996; Pueschel etal. 1996). Epithallial cell senescence and sloughing appear to be part of a specialized, determinate program of cell death which is probably genetically regulated. Cells of filaments are joined by primary pit connections. Depending on the species, cells of adjacent filaments may fuse with one another or are joined by secondary pit connections. In a very few other species both cell fusions and secondary pit connections are present (Townsend etal. 1994,1995). Cell fusion in coralline algae is often accompanied by fusion of nuclei as well (Johansen 1981; Bailey pers. observ.). Secondary pit connections in corallines are formed directly; conjunctor cells are absent in the group.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The thalli of coralline algae are heavily calcified throughout. The only cells, or portions of cells, that are never calcified are those of the reproductive systems and the outer walls of epithallial cells. The cell walls of corallines are comprised of three wall layers. The innermost cell wall layer is uncalcified, but the middle and outer wall layers are impregnated with calcium carbonate. Unlike most other calcifying seaweeds that precipitate or deposite aragonite, corallines deposit calcium carbonate as calcite in their cell walls. Little is known of the mechanisms responsible for the deposition of calcite in these algae. Johansen (1981) provides an overview of several proposed models of calcification for coralline algae. Experimental studies have linked the rate of deposition of calcite to photosynthetic activity and pH (e.g., Gao etal. 1993). Photosynthetic activities have been demonstrated to generate an alkaline microenvironment in intercellular spaces which favors the formation of calcium carbonate. It is clear, however, that photosynthetic activity alone does not drive the process of biomineralization in coralline algae since some calcified cells lack plastids entirely. Calcification in non-photosynthetic cells of corallines is believed to be supported by translocation (Pearse 1972; La Velle 1979). Additional evidence suggests that the organic cell wall matrix plays a crucial role in the process, perhaps serving as the site for the nucleation of calcite crystals (Borowitzka etal. 1974; Borowitzka 1982a,b; Millson & Moss 1985; Somers etal. 1990; Pueschel etal. 1992b). Depending on their fate, cells of corallines that were initially calcified may later be decalcified. Cell sloughing, as well as the differentiation or formation of conceptacle primordia, cell fusions, and genicula (see below) all involve the decalcification of localized areas of vegetative tissue. Processes accounting for decalcification in coralline algae are unknown.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Corallines can be divided into two distinct morphological types based on the absence or presence of genicula. Non-geniculate corallines lack genicula and most take the form of plaque-like crusts (Figure 1.3). However, depending on the species and environmental conditions, non-geniculate corallines can assume a bewildering variety of forms. Non-geniculate species may be smooth or beset with unbranched or branched protuberances, and a few species are arborescent (Woelkerling etal. 1993). In contrast, geniculate species (Figure 1.4) take the form of upright, branched fronds that are attached to the substrate via a crustose basal holdfast system. The thalli of geniculate corallines are composed of alternating uncalcified nodes (genicula) and calcified intemodes (intergenicula) (Figures 1.4 & 1.5). Most corallinologists follow Johansen (1969a, 1976,1981) in recognizing three subfamilies of geniculate coralline algae. These subfamilies are delimited from one another by developmental and anatomical features of the geniculum (Figure 1.5). In the Corallinoideae genicula are unizonate and are comprised of a single tier of specialized medullary cells that elongate greatly during development (Borowitzka & Vesk 1978; Figure 1.5a). Genicula of the Amphiroideae are multizonate. In the latter subfamily genicula are composed of many tiers of cells that, in most species, are of equal height. Unlike the unizonate genicula of the Corallinoideae, in the Amphiroideae genicula are sometimes secondarily corticated (Figure 1.5b). The genicula of Metagoniolithon, the only genus classified in the subfamily Metagoniolithoideae, are also multizonate and corticated but the cells of the geniculum are not arranged into regular tiers (Ducker 1979). In marked contrast to all other types of genicula, the genicula of Metagoniolithon are meristematic and give rise to false whorls of branches (verticillasters) (Figure 1.5c).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28

Figure 1.3. Non-geniculate coralline algae. (A) Lithothamnion, (B) Lithophyllum, and (C) Mesophyllum. Scale=1cm. (Reproduced with permission from Abbott & Hollenberg 1976; see Appendix A).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29

Figure 1.4. Geniculate coralline algae. (A) Serraticardia, (B) Lithothrix, and (C) Corallina. Scale=1cm. (Reproduced with permission from Abbott & Hollenberg 1976; see Appendix A).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A IGSI IG J Corallinoideae

Amphiroideae

Metagoniolithoideae

Figure 1.5. Schematic diagram illustrating the different types of genicula in the coralline algae. (A) unizonate, ecorticate geniculum of the Corallinoideae, (B) multizonate, corticate geniculum of the Amphiroideae, and (C) multizonate, corticate, and meristematic geniculum of Metagoniolithon. G=genicula, IG=intergenicula, co=corticated area, br=branch.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31 Five coralline genera ( Austrolithon, Choreonema, Ezo, Kvaleya, Lesueuria), each monotypic and assigned to a different subfamily, are known only as host-associates of other coralline algae. The thalli of Kvaleya and Ezo are much reduced in size and develop on the external surface of the host (Adey & Sperapani 1971; Adey etal. 1974). In these genera haustoria are produced which penetrate the host's surface layers and fuse with host cells below. The vegetative thalli of Austrolithon, Choreonema, and Lesueuria are even further reduced and develop entirely within the host; only their conceptacles emerge above the host's surface (Woelkerling 1987; Woelkerling & Ducker 1987; Harvey & Woelkerling 1995). The thalli of the latter species are formed just beneath the surface of the host and, in general, take the form of diffusely organized pseudoparenchymatous pads of cells which are not differentiated into distinct cellular regions. These pads of cells ultimately give rise to conceptacles externally and may produce few-celled filaments internally as well. Ultrastructural evidence clearly indicates that Choreonema is a true parasite that derives nutrition for growth and reproduction from the cells of its host (Broadwater & LaPoint, in press). However, the nature of the relationship between other associates and their hosts is not clear.

Reproduction. The life history of coralline red algae is of the uPolysiphonia-\ype,* consisting of isomorphic gametangial and sporangial phases (West & Hommersand 1981). Field observations indicate that sporangial plants are often more common than are gametangial plants and also suggest that, at least in some species, gamete production may be controlled by seasonal factors (Adey 1973; Chihara 1973; Johansen 1976; Athanasiadis 1995). Most coralline species are dioecious, but monecious plants and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hermaphrodite conceptacles have been reported for some species (e.g., May & Woelkerling 1988; Chamberlain 1991). Excepting species of Sporolithon and Heydrichia, all male, female and sporangial reproductive structures of corallines are housed in calcified chambers termed conceptacles (Figures 1.6 & 1.7). In the former genera sporangial conceptacles are lacking and spores are instead formed in "sporangial complexes" or sori (Verheij 1992; Keats & Chamberlain 1995; Townsend etal. 1994,1995; Alongi etal. 1996). Although anatomically and developmentally distinct, the sori of Sporolithon and Heydrichia are unquestionably homologous to sporangial conceptacles found in other corallines. Conceptacles are initiated by the differentiation of cells in either the intercalary meristematic (cortical) or medullary cell regions. Although mature conceptacles may appear identical, the ontogeny of these structures varies among species. The conceptacle chamber and roof may be formed by filaments overarching the fertile disc. In other species, the conceptacle is formed as cells above the fertile disc degenerate (May & Woelkerling 1988; Penrose 1992, Woelkerling & Campbell 1992; Woelkerling & Harvey 1992; Townsend et al. 1995). Other modifications of these two basic modes of conceptacle construction have been documented as well; for example, in the Sporolithaceae sporangial chambers are formed as the tetrasporocyte enlarges and pushes adjacent filaments aside. Depending on the species, the conceptacle roof may be perforated by one to many ostioles, or pores. The conceptacle pores of some taxa are occluded by a mucilaginous plug formed at the tip of the tetrasporangium which disintegrates prior to the release of spores. Although detailed comparative analyses of the female reproductive apparatus has historically played an important role in the classification of red

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. □ EDO D0DD

□□□□ □□□□

Figure 1.6. Diagram illustrating female/carposporophytic development in the coralline red algae. (A) fertilization, (B) transfer of the zygote nucleus to the supporting cell via fusion through the hypogynous cell, (c) alternative mode of transfer of zygote nucleus directly to the supporting cell by a delicate transfer tube, and (D) formation of diploid carposporophytic tissues including the central fusion cell, gonimoblast filaments, and carposporangia. s=spermatia, t=trichogyne, cp=carpogonium, hc=hypogynous cell, sc=supporting cell, bc=basal cell, tr=transfer tube, cr=carposporangial remnants, cs=carpospore, g=gonimoblast filament, fc=fusion cell.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34

Figure 1.7. Line drawing of vertical median sections through mature (A) carposporangial and (B) spermatangial conceptacles of coralline algae. C=carpospores, F=fusion cell, G=gonimoblast filament, P=pore, S=spermatia (in black). Scale=50pm. (Re-drawn and modified from Lebednik 1978 and Chamberlain 1994).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. algae, details of female (carposporophytic) development in corallines are poorly known. Many separate studies present conflicting results which are not easily reconciled with one another. It is not presently known if all developmental variation recorded in separate studies can be attributed to natural variation within the group or other factors. Based on the best available evidence, a brief discussion of the mode of female reproductive development in corallines is presented below. Carpogonial branches in corallines are two celled and include the carpogonium (with trichogyne) and an hypogynous cell which are derived from a subtending supporting cell (Lebednik 1977; Bold & Wynne 1978; Silva & Johansen 1986; Woelkerling 1988; Townsend etal. 1995; Figure 1.6a). Trichogynes of mature carpogonia protrude outward through the conceptacle pore(s). Following fertilization in most corallines, e.g., Metamastophora (Woelkerling 1980) and Calliarthron (Johansen 1969a), a channel forms between the fertilized carpogonium and the supporting cell as a result of direct fusion through the hypogynous cell (Figure 1.6b). Suneson (1937) and Lebednik (1977) report that in some taxa (e.g., Melobesia, Mesophyllum and Clathromorphum) a transfer tube arises from the base of the carpogonium and connects directly with the supporting cell (Figure 1.6c). Presumably the zygote nucleus is translocated to the supporting cell through the transfer tube in these species. The existence of transfer tubes in the group has, however, been questioned and should be confirmed (Woelkerling 1980; Johansen 1981; Silva & Johansen 1986). Once the zygote nucleus, or its derivatives, is (are) transferred to the supporting cell this cell enlarges, expands laterally, and fuses with the supporting cells of adjacent carpogonial filaments. Schmitz (1883) referred to the multiple fusion of cells in adjacent carpogonial branch systems as the production of a "syncarpium." Suneson (1937) and Lebednik (1977)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36 have also reported fusion between cells of the supporting and basal cell layers in some melobesioid genera. The resulting large, multinucleate, plate-like or lobed fusion cell is centrally located within the conceptacle (Lebednik 1977; Woelkerling & Foster 1989; Woelkerling & Harvey 1992; Figures 1.6d & 1.7a). Fusion cells are absent or are comparatively reduced in size in Sporolithon and Heydrichia (Verheij 1992; Townsend etal. 1994,1995). Reports of discontinuous fusion cells in some species have been made; however, these observations should be confirmed. Woelkerling and Foster (1989) and Woelkerling and Harvey (1992) noted that multiple fertilizations of carpogonia in Synarthrophyton and Mesophyllum result in the initiation of several, separate fusion cells that may later fuse with one another. Depending on the stage of the section examined, multiple initial fusion cells could present the appearance of a discontinuous fusion cell. Also, sections through lobed fusion cells could be misinterpreted as discontinuous. In most species, short, usually unbranched gonimoblast filaments arise from the periphery of the fusion cell and bear a terminal carposporangium which produces a single carpospore (Woelkerling 1988; Figures 1.6d & 1.7a). There are at least three different modes of female reproductive development in the Sporolithaceae (Townsend etal. 1995). In Sporolithon durum (Townsend etal. 1995), carposporangia develop directly from the fertilized carpogonium, but in S. episoredion (Verheij 1992) gonimoblast filaments arise from both the dorsal and peripheral sides of a fusion cell and carposporangia are produced across the conceptacle floor. In corallines, then, the diploid carposporophytic generation includes the central fusion cell, gonimoblast filaments, and carposporangia. Diploid carpospores are released, germinate and develop into independent, sporophytic plants. Critical studies of female-carposporophytic development are needed for a number of coralline species for comparative assessment of the variation

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. documented in both studies cited here and others. These data are certain to be of use for delimiting higher-level taxa within the group and for understanding the phylogenetic relationships among them. Male reproductive development (Figure 1.7b) in corallines does not differ considerably from the process observed in other florideophyte red algae (Hommersand & Fredericq 1990). In some, but not all, coralline species male and female conceptacles are dimorphic. Spermatangial initials may be simple or occur in branched, dendroid clusters (Lebednick 1978; Woelkerling 1988; Chamberlain & Keats 1994). In some species both types of spermatangial branch systems may be found in a single conceptacle. Spermatangial initials may be found only on the floor of male conceptacles or on the walls and roofs of conceptacles as well. Differences in the size and shape of male conceptacles (cf. carpo- and bi-/tetrasporangial conceptacles) and the types and distribution of spermatangial initials have been used to delimit genera of Corallinales within certain subfamilies. The initiation and development of tetrasporangial or bisporangial (see below) conceptacles is as described above (Figure 1.8). Tetrasporogenesis in corallines follows generally the pattern found in other florideophytes with certain exceptions. Culture studies (Chamberlain 1987) indicate that meiosis occurs in the tetrasporangium. The result is four haploid tetraspores half of which, after they are released, settle and germinate into male gametophytes while the other half produce female gametophytes. In most corallines, tetraspores are zonately cleaved from the terasporangium following meiosis. Simultaneously three cell wall infurrowings are initiated and progress centripetally inward, eventually separating the four meiotically-derived nuclei into separate spores. The simultaneous zonate cleavage pattern of the tetrasporangium observed in the Corallinales has not been found in any other red algal taxon. Cleavage of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 1.8. Line drawing of vertical median section through the mature sporangial conceptacle of a coralline alga. T=tetraspores, B=bispores. P=conceptacle pore. Scale=100(im. (Re-drawn and modified from Kapraun 1980).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tetraspores from the tetrasporangium in species of Sporolithon and Heydrichia follows a cruciate pattern not found in other coralline taxa. Comparative electron microscopic studies of tetrasporogenesis among several coralline species representing different subfamilies have been completed (Vesk & Borowitzka 1984; Agee & Broadwater 1995; Kamas et al. 1995; Griffon & Broadwater 1996; Mays etal. 1996). At the ultrastructural level, the primary differences in the patterns of tetrasporogenesis among different corallines include: 1) the presence or absence of electron dense material surrounding the nucleus, 2) the degree of development of perinuclear membranes, and 3) the associations of other organelles, e.g., mitochondria and plastids, with the dividing nuclei. Patterns of tetrasporogensis are conserved within a genus but differ among genera and among subfamilies indicating that features of tetrasporogenesis are taxonomically, and perhaps phylogenetically, informative. Ultrastructural studies have also demonstrated that a well- developed cell wall is associated with the spores of coralline algae (Vesk & Borowitzka 1984; J. Scott pers. comm.). Detectable cell walls are absent from spores of other red algae. Corallines also reproduce by asexual means. In many coralline species bispores are produced in sporangia! conceptacles instead of tetraspores (e.g., Duwel & Wegeberg 1996; Figure 1.8). Bispores and tetraspores may also both be present in the same conceptacle (e.g., Athanasiadis 1995). Bispores are cytokinetically produced by a single transverse division of the tetrasporocyte and in most species each spore receives a single nucleus. Because uninucleate bispores regenerate the sporophytic phase, it is widely accepted that the nuclei of these spores are haploid and are the result of mitotic divisions. Binucleate bispores have been reported in some species and Suneson (1950) demonstrated that these spores develop into male or female plants. In Lithothrix aspergillum bispores were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. found to be produced exclusively by populations on the extreme northern and southern ranges of the species (Tyrrell & Johansen 1995). Some corallines are able to propagate themselves through asexual vegetative means. Attached plants, or portions of attached plants, may break free and are capable of continued growth and reproduction (Rosenvinge 1917; Johansen 1969a; Gittins & Dixon 1976). These free-living algae are, in some cases, rolled across the sea bottom in surge areas and may take the form of rounded "nodules,” or rhodoliths. Often the rhodolith has at its center rock or other material on which the alga was originally attached. Rhodoliths are abundant off fringing reefs in both the northern and southern hemispheres. Common rhodolith-forming taxa include species of Lithothamnion and Sporolithon (Adey 1968; Adey & McKibbin 1970; Bosellini & Ginsberg 1971; Siesser 1972; Keats etal. 1996). Structures termed "propagules" have also been recorded for some species, but their role, if any, in apomictic reproduction is uncertain (Hollenberg 1970; Coppejans 1978, 1983; Woelkerling 1988).

Ecology. Coralline algae are found only in marine environments and as a group include the overwhelming majority of the calcareous algae of the world. Corallines are among the most abundant and geographically widespread of all macrophytic seaweeds. Representatives of the order can be found in all marine biogeographic regions, from the cold waters of the arctic and antarctic to warmer, tropical equatorial waters. Most corallines occupy sublittoral habitats, although certain species flourish intertidally and in tide pools in the surge zone. The abundance of corallines generally decreases with depth; however, a few species are found in deep waters (Littler 1973; Meneses 1993). The deepest marine macroscopic alga known is a coralline alga found on an uncharted seamount at a depth of ca. 268 meters (Littler & Littler 1985).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. As with other seaweeds, the geographical distribution of species of coralline algae is widely believed to be determined by oceanic isotherms. Adey and Macintyre (1973) have noted that the species diversity of separate taxa can be correlated with seawater temperatures. The Melobesioideae appears to be a predominately arctic and subarctic assemblage, whereas species classified in the subfamilies Mastophoroideae and Lithophylloideae achieve their greatest species diversity in subtropical and tropical waters. The known geographical ranges of the genera of coralline algae were summarized by Johansen (1981) and Woelkerling (1988). The distribution of coralline species also varies with depth. The vertical distribution and abundance of species is dependent on light quality and quantity, substrate availability, wave action, dessication tolerance, interspecific competition, and grazing pressure (Adey & Macintyre 1973; Hackney etal. 1989; Meneses 1993; Leukart 1994). On some coasts dense populations of coralline algae represent the dominant members of macroalgal climax communities. Corallines are excellent competitors for primary space in the marine environment and are among the first marine organisms to colonize bare substrates (Adey & Vassar 1975; Paine 1984; Matsuda 1989; Benedetti-Cecchi & Cinelli 1994). Different species of corallines often compete with one another at both early and later stages of succession (Ady & Vassar 1975; Paine 1984; Steneck 1986; Steneck & Paine 1986; Keats etal. 1994). Excepting in areas of high wave energy, corallines are at later successional stages overgrown by filamentous algae, tunicates, invertebrates (e.g., anthozoans, bryozoans, barnacles, and sponges), and other macrophytic seaweeds (Matsuda 1989). Corallines provide grazing, habitat, and refugia for a number of invertebrate and finfish taxa. Geniculate corallines typically harbor large populations of various molluscs, crustaceans, and polychaete worms. A

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. number of boring marine worms and other invertebrates make their homes inside the thalli of non-geniculate species. Corallines are grazed actively by herbivorous fishes, molluscs (e.g., limpets), and various echinoderms (Paine 1984; Steneck 1986; Steneck & Paine 1986). In the Precambrian, reefs were constructed mainly by cyanobacteria, but hard corals and calcified red algae began to dominate and construct reef superstructures during the Ordivician period (Sorokin 1993). Today, coral reefs cover approximately two million square kilometers of the earth's surface and are home to one quarter of all marine species (Shreeve 1996). Coral reefs are restricted to tropical and subtropical waters where the annual mean minimum temperature is 18° C or higher (Nybakken 1988). In the absence of corals, bioherms of colder waters are constructed primarily by coralline algae. Coralline algae play fundamental roles in the ecology and geology of coral reef ecosystems. Corallines overshadow all other calcareous organisms in terms of standing stock in reef ecosystems, often forming continuous pavements many kilometers in length (Zaneveld 1958; Littler 1972,1973; Adey & Macintyre 1973). The productivty of coralline algae is well within the range reported for other reef primary producers and the rate of calcification in corallines is equivalent to that of hermatypic scleractinian corals (Littler & Doty 1974; Adey 1978a; Sorokin 1993). Geologically, corallines are involved in the processes of reef growth and consolidation. Depending on depth, corallines may cover 40- 100% of the available reef substrate (Marsh 1970; Littler 1973). Actively growing corallines significantly contribute to the accretion of calcium carbonate on reefs, particularly on the seaward edges of reefs that are exposed to high wave energy. Along the windward edges of reefs in the Pacific and Caribbean coralline algae form massive "algal ridges" (Adey 1978a,b). These algal ridges are dominated by a few coralline species that outcompete other plants and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. animals (Lee 1967; Littler & Doty 1974; Steneck & Adey 1976). Coring studies of reefs in such areas indicate that the rate of calcium carbonate accretion by coralline algae is between 3 and 6 meters per 1000 years. Ridge-forming hermatypic species are are especially adapted to cope with hydrodynamic and physiological stresses associated with relatively prolonged exposure to the atmosphere (Littler 1972; Littler & Doty 1974). When fragmented, or upon decomposition, corallines, as well as a few taxa of calcifying green algae (e.g., Halimeda, Udotea, and Peniclllus), release calcium carbonate which forms sediment. Approximately 20 - 40% of the total amount of soft carbonaceous sediments produced by reef biota can be attributed to species of algae (Sorokin 1993). A portion of this sediment is swept back into fissures and surge channels of the reef as filler. These sediments are-cemented in place by hermatypic organisms and, in time, are metamorphosed and incorporated into the reef structure (Ladd etal. 1950; Moberly 1970; Ginsberg & Schroeder 1973). The structural integrity and development of reefs depends crucially on the cementing and stabilization of framework elements by coralline algae (Ladd etal. 1950; Adey & Macintyre 1973; Ginsberg & Schroeder 1973; Adey 1978a). In recent decades, a fundamental shift in the community structure of coral reef ecosystems has been noted on a worldwide scale (Johnson et al. 1995). The species composition of reefs previously dominated by corals, coralline algae, and fleshy macrophytic seaweeds are being transformed by the invasion of turf-forming algae. Changes in the species composition of reefs has resulted in changed patterns of carbon flux through the ecosystem (Johnson et al. 1995). Many studies have noted that decreases in reef species diversity can be correlated to eutrophication of adjacent waters (e.g., Naim 1993; Peyrot- Clausade et al. 1995). Earlier studies suggested that phosphate-enriched

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. waters may enhance the growth of corallines. However, field and laboratory experiments conducted by Bjork etal. (1995) on a Zanzibar reef indicate that the abundance, growth, and calcification rates of corallines are adversely affected by high phosphate levels.

Fossil corallines. Unlike other macroalgal groups, the calcareous nature of corallines has allowed the preservation of an extensive fossil record (Johnson 1961; Wray 1977; Braga & Aguirre 1995). Wray (1977) divided fossil calcareous red algae into four families, the Solenoporaceae, Gymnocodiaceae, Squamariaceae, and Corallinaceae. The first two families listed are extinct, whereas the latter two families include both fossil and extant species. Well- developed coralline algae assigned to the Solenoporaceae were present by the early Devonian period (400 million years ago; Mya) and several lines of evidence suggest that the Corallinaceae may have evolved from ancestors within the latter family. Undoubted fossil representatives of the Corallinaceae, includingSporolithon (syn. Archaeolithothamnion), Lithophyllum , and Lithothamnion, first appear in the late Jurassic period (ca. 163 Mya). Sixty fossil coralline taxa have been described, 13 of which have been assigned to extant genera. There has been little cross-communication between palaeontologists and phycologists who study coralline algae. The taxonomy of extant taxa is based mainly on vegetative anatomical and reproductive features. Conceptacles, or their precursors (?), and their uncalcified contents are not preserved in fossil taxa recovered from deposits predating the early Paleocene epoch. For this reason, the taxonomic circumscription of fossils assigned to extant genera often differs considerably from the modern concepts for these genera. The geological ranges of fossil corallines attributed to extant genera are presented by Wray (1977). Adey and Macintyre (1973) have proposed a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. phylogenetic hypothesis for the evolution of the Corallinaceae which is based, in part, on fossil evidence.

Taxonomic history and circumscription of the Corallinales.In the early 18th century geniculate corallines were treated as plants (e.g., Tournefort 1719; Ray 1724), but non-geniculate corallines were considered animals. Later, Jussieu (1745), Ellis (1755), and Ellis and Solander (1786) concluded that all calcareous organisms were compound animals composed of polyps. These authors, and later Linnaeus (1758), classified both types of corallines as animals, apparently mistaking the conceptacles of coralline algae for the skeletal polyp-cups of scleractinian corals. The family Corallinaceae was erected in 1812 by Lamouroux who also referred the taxon to the animal kingdom. In addition to corallines, Lamouroux included in the Corallinaceae certain other calcifed red algae (e.g., Galaxaura) as well as some siphonous green algae. The plant-like nature of both geniculate and non-geniculate corallines, and the relationship of corallines to other red algae, was firmly established in studies by Schweigger (1819,1826) and Philippi (1837). Non-coralline calcified marine algae were removed from the Corallinaceae, and non-geniculate and geniculate corallines were brought together for the first time, in studies by Decasine (1842a,b) and Chauvin (1842). Schmitz (1892) and Schmitz and Hauptfleish (1897) classified the family Corallinaceae within the order Cryptonemiales and aligned this family with the Squamariaceae (=Peyssonneliaceae). The alliance of the Corallinaceae and Peyssonneliaceae within the Cryptonemiales was based on the key character that carpogonia and auxiliary cells were borne on separate branches in-special structures (conceptacles and nemathecia, respectively). Vegetative similiarites also seemed to ally the Corallinaceae and Peyssonneliaceae since most taxa in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. both families can be characterized as having a crustose habit. Kylin (1928, 1956), who was influenced by the work of his former student Suneson (1937, 1943,1945), maintained the same classification. Kylin (1956) emphasized the "accessory” nature of the carpogonial and auxiliary cell filaments and keyed the Corallinaceae with those taxa in which the carpogonium first fuses with a "nurse" cell in the carpogonial branch and, like Schmitz (1892), emphasized the nemathecial nature of the system. According to Kylin (1956) the conceptacle of the corallines represented a specialized nemathecium. The Corallinaceae was classified within the Cryptonemiales for nearly 60 years. The intrafamilial classification of the Corallinaceae begins with Lamouroux (1816) who divided the Corallinaceae into two categories, "polypiers articules" and "polypiers inarticulds," but assigned no rank to either group. Following Lamouroux, Areschoug (1852) established the tribes Melobesieae and Corallineae for non-geniculate and geniculate coralline species, respectively. The first validly published subfamilial name, Corailinoideae, was proposed by Gray (1821). Taxonomic studies of corallines in the period 1895- 1911 were dominated by two persons, Mikael Foslie and Franz Heydrich. During this period, Foslie and Heydrich described approximately 470 and 84 specific and intraspecific taxa, respectively. Foslie and Heydrich were bitter foes who early in their careers often exchanged unseemly, vitriolic remarks in print. The taxonomic viewpoints of these authors changed considerably (sometimes illogically) between successive papers and were fueled, no doubt, by their mutual animosity. Foslie's work, in particular, has been criticized because he frequently failed to provide descriptions for taxa in sufficient detail to be of use to subsequent workers and also failed to properly catalogue or annotate material used in his studies. Further complicating matters, most of Heydrich's personal collections of coralline algae were destroyed as a result of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the Allied bombing of Berlin (1943-44) during the Second World War. This situation has resulted in a great deal of confusion about the generic and subgeneric limits of the numerous coralline taxa validly erected by these men (Johansen 1981). Also, it is well worth noting that the taxa erected by these authorities were based almost exclusively on differences in the external appearance of the specimens. Because reproductive features were considered in only a superficial manner, the type collections of many species do not include representatives of each life history phase. This unfortunate circumstance has created innumerable practical and nomenclatural problems for present-day taxonomic studies of corallines, since the delineation of most genera in the modern era is based, at least in part, on the anatomy and development of the reproductive structures. The legacy left by Foslie and Heydrich continues to frustrate corallinologists today. Foslie's systematic treatment of the coralline algae, nevertheless, remains the cornerstone of recent schemes of classification. Foslie (1903, 1904) divided the coralline algae into eight separate sections, but failed to provide valid descriptions for any of these groups. Subsequently, later workers legitimately established several subfamilies based on Foslie's original sections. Yendo (1902) erected the subfamily Melobesioideae and Setchell (1943) established the subfamilies Mastophoroideae, Lithophylloideae, and Sporolithoideae. Additional historical and nomenclatural details on the classification of coralline algae are provided by Cabioch (1972), Lebednik (1977), Johansen (1981), and Woelkerling (1988). For many years (see above) the Corallinaceae was classified within the order Cryptonemiales. In 1985 Kraft and Robins argued that there was no logical basis for the separation of the orders Cryptonemiales and Gigartinales based on either vegetative or reproductive features and therefore subsumed the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cryptonemiales in the Gigartinales. Although it was suggested by some (e.g., Lebednik 1977) that the Corallinales should also be subsumed in the Gigartinales, ultrastructural data (Pueschel & Cole 1982) clearly indicated that the Corallinaceae was not closely related to any taxon included in Kraft and Robins' (1985) expanded concept of the Gigartinales. The implicit suggestion by Pueschel and Cole (1982) that the Corallinaceae should be elevated to the rank of order was formally affected by Silva and Johansen (1986), who erected the order Corallinales and delineated the group from other red algal orders based on certain reproductive, anatomical, and ultrastructural features. Many features considered characteristic of the Corallinales by Silva and Johansen (1986) have now been found in other red algal taxa. For example, calcite deposition in marine red algae was thought to be restricted to the Corallinales but has now been documented in the specialized "calciferous cells" of Rhodogorgon and Renouxia (Rhodogrogonales; Norris & Bucher 1989; Pueschel etal. 1992b; Fredericq & Norris 1995). Pit plugs of identical morphology to those found in corallines have now been documented for species classified within the orders Batrachospemales and Rhodogorgonales (Pueschel 1989,1994; Pueschel etal. 1992a; Table 1.1) And, the cruciate pattern of division of the tetrasporangium, once thought to be lacking in corallines, has now been found in two putatively primitive coralline genera, Sporolithon and Heydrichla. The circumscription of the order Corallinales, therefore, needs to be emended. Seven features, which are found in most coralline species, distinguish the Corallinales from other red algal orders: 1) vegetative thalli (excepting genicula) calcified throughout, 2) simultaneous zonate cleavage of tetraspores, 3) presence of well-developed spore cell walls, 4) presence of reproductive structures housed in calcified conceptacles or sori, 5) presence of both apical and intercalary meristems, 6) presence of physiologically distinct

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. epithallial cells that are derived from intercalary meristematic initial cells, and 7) post-fertilization events involving a cluster of procarpial filament systems and the production of a large, coenocytic fusion cell as part of the development of the carposporophyte. An up-to-date account of selected intraordinal systems of classification for coralline algae is presented in the following section.

Coralline systematics.The Corallinales is among the largest and most studied of all red algal taxa (Woelkerling etal. 1993). The group is also widely acknowledged to be among the most difficult groups of algae to investigate taxonomically due, at least in part, to their great diversity and morphological variability (Taylor 1960; Kraft 1981). Presently, ca. 27 genera of non-geniculate and 15 genera of geniculate corallines are recognized. The number of species is still largely guesswork, and is perhaps between 400 and 500 (Woelkerling. pers. comm.). The diversity and the relative antiquity (Johnson 1956,1961; Wray 1977) of the corallines suggest that the group represents one of the major evolutionary lineages of florideophycean red algae. Most coralline genera and species, particularly non-geniculate forms, are indistinguishable on the basis of their external appearance. The accurate identification of corallines requires the study of decalcified reproductive material, i.e., conceptacles and their contents. Species and generic concepts for coralline algae are poorly developed and remain in a state of flux. Phenetic approaches to the classification of these algae based upon analyses of anatomical and developmental characters have not resulted in a universally accepted scheme at any taxonomic level. Many genera have not been studied in a modem context and only recently have reproductive characters been incorporated among the criteria used for generic delineation within the group

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Townsend 1981; Turner & Woelkerling 1982; Woelkerling 1983, 1988; May & Woelkerling 1988; Penrose & Woelkerling 1988,1992; Penrose 1992; Penrose & Chamberlain 1993; Woelkerling & Harvey 1993; Townsend et al. 1994; Chamberlain & Keats 1994). The lack of data for some taxa and differences of opinion over the taxonomic importance of certain characters have resulted in controversy over the status of certain taxa within the group. Current generic circumscriptions are not universally accepted (e.g., Cabioch 1972; Woelkerling 1985; Penrose & Woelkerling 1988; Campbell & Woelkerling 1990; Chamberlain & Keats 1994; Wilks & Woelkerling 1994). For these reasons, generic boundaries in most non-geniculate subfamilies are considered “provisional"(Chamberlain & Keats 1994). There is also little consensus among higher-level taxonomic arrangements for the coralline algae presented by various authors. Since 1950, 15 different suprageneric classification schemes have been proposed for the group (Table 1.3). Two merit special attention here because they are most commonly encountered in the current literature and represent perhaps the most dramatic and divergent hypotheses of suprageneric and generic interrelationships within the Corallinales. The classification scheme proposed by Cabioch (1971,1972,1988; Table 1.4) is distinctive in that major emphasis is placed on characters relating to conceptacles and cell connections, rather than on the presence or absence of genicula. The latter character is used by Cabioch at the ranks of tribe and genus. Cabioch's scheme, which she suggests represents a natural classification of the coralline algae, requires genicula to have arisen independently at least three separate times, presumably from crustose antecedents. Cabioch's system of classification has not been widely accepted. Instead, most ecological, anatomical, and taxonomic studies of coralline algae

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1.3. Suprageneric classification schemes devised for coralline algae since 1950. # = taxon recognized by the author(s), ■■ = taxonomic status of these taxa not discussed in the cited work.

Mason 1953 — _ _ Hamel & Lemoine 1953 • — — — Kylin 1956 • • • • Maslov 1962 • — — — Zinova 1967 • • — — — Johansen 1969 • • • • • • Adey & Johansen 1972 • • • • • Cabioch 1972 • • • • • Johansen 1976 • • • • • Wray 1977 Johansen 1981 • • • • • • Woelkerling 1988 • • • • • • Verheij 1993 • • • • — Harvey & Woelkerling 1995 • • • • • • •

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1.4. Intraordinal classification of the Corallinales according to Cabioch (1972, 1988).

Taxon Included genera

Family Corallinaceae

Subfamily Sporolithoideae Sporolithon Subfamily Lithothamnioideae

Tribe Lithothamnieae Kvaleya, Melobesia, Mesophyllum, Lithothamnion Polyporolithon Subfamily Corallinoideae

Tribe Mastophoreae Lithoporella, Mastophora Metamastophora

Tribe Neogoniolitheae Fosliella, Metagoniolithon Neogoniolithon, Porolithon

Tribe Corallineae Alatocladia, Arthrocardia Bossiella, Calliarthron, Cheilosporum, Chiharaea, Choreonema, Corallina, Haliptilon, Jania, Marginisporum, Serraticardia, Yamadaea

Subfamily Lithophylloideae

Tribe Amphiroeae Amphiroa

Tribe Lithophylleae Lithophyllum, Pseudolithophyllum

Tribe Dermatolitheae Dermatolithon, Goniolithon, Lithothrix, Tenarea

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53 have been conducted within the suprageneric frameworks for the geniculate and non-geniculate genera erected by Johansen (1969a, 1981) and Woelkerling (1988; Harvey & Woelkerling 1995), respectively. In this system the Corallinales are divided into eight subfamilies; five subfamilies of non- geniculate genera and three subfamilies that include only geniculate genera. In this scheme (Table 1.5) the presence/absence of genicula is treated as the overriding phylogenetic criterion at higher levels of classification. The question of whether genicula or secondary pit connections are of greater phylogenetic significance remains unresolved, and the same can be said of characters presently used to delimit subfamilies, tribes, and genera (Johansen 1972,1976; Woelkerling 1988; Chamberlain & Keats 1994). No single scheme depicting relationships between genera or higher level taxonomic divisions is universally accepted at present. Thus, as noted by Woelkerling (1988), determining which classification scheme to follow is based primarily on subjective and pragmatic considerations. The relationship of the Corallinales to other red algal lineages also remains unclear. Published molecular data suggest that the Corallinales represents an early diverging lineage within the florideophycean radiation and that the Corallinales may be allied to the Hildenbrandiales (Freshwater et al. 1994, Ragan etal. 1994). This hypothesis is, however, weakly supported by the data and is inconsistent with PuescheFs (1994) assertion that the Corallinales is closely related to other red algae that possess pit plugs with two cap layers. The phylogenetic affinities of the Corallinales are re-evaluated in Chapter 2 of this dissertation.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 1.5. Intraordinal classification of the Corallinales according to Johansen (1981) and Harvey and Woelkerling (1995).

Taxon Included genera

Family Sporolithaceae Heydrichia, Sporolithon Family Corallinaceae

Subfamily Amphiroideae Amphiroa, Lithothrix

Subfamily Corallinoideae Arthrocardia, Alatocladia, Bossiella, Calliarthron, Cheilosporum, Chiharea, Corallina, Haliptilon, Jania, Marginisporum, Serraticardia, Yamadaea

Subfamily Metagoniolithoideae Metagoniolithon

Subfamily Austrolithoideae Austrolithon, Boreolithon

Subfamily Choreonematoideae Choreonema

Subfamily Lithophylloideae Ezo, Lithophyllum, Titanoderma, Tenarea Subfamily Mastophoroideae Fosliella, Lesueuria, Lithoporella, Mastophora, Metamastophora, Spongites, Neogoniolithon, Pneophyllum

Subfamily Melobesioideae Clathromorphum, Exilicrusta, Kvaleya, Lithothamnion, Leptophytum, Mastophoropsis, Melobesia, Mesophyllum, Phymatolithon, Synarthrophyton

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2 PHYLOGENY OF THE CORALLINALES BASED ON ANALYSIS OF SSU GENE SEQUENCES

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION

The marine red algal order Corallinales is comprised of a diverse assemblage of species that deposit calcite in their ceil walls. Coralline species are found in all the world's oceans and occupy habitats ranging from the intertidal down to the lower limits of the the photic zone (Johansen 1976; Littler & Littler 1985; Leukart 1994). Corallines are hermatypic organisms that play especially important ecological and geological roles in coral reef ecosystems (Adey & Macintyre 1973; Sorokin 1993). Coralline algae can be divided into two types; geniculate and non- geniculate. Geniculate corallines take the form of upright, branched fronds that are attached to the substrate by a crustose basal holdfast system. The thallus of geniculate corallines is divided into uncalcified nodes (genicula) and calcified intemodes (intergenicula). In contrast, non-geniculate, or crustose, corallines lack genicula. Non-geniculate corallines can assume a bewildering variety of different growth forms depending on the species and local environmental conditions (Woelkerling et al. 1993). Excepting only a very few species, phenotypically plastic non-geniculate corallines cannot be identified on the basis of external appearance alone. Accurate identification of these species depends fundamentally on detailed studies of their vegetative anatomy and reproductive structures (conceptacles and their contents). As presently circumscribed, the Corallinales contains two families, the Sporolithaceae and Corallinaceae, and approximately 42 genera. The taxonomic status of some coralline red algae is the subject of conjecture and controversy. Most species are poorly known and uniform, meaningful species concepts are lacking for nearly all non-geniculate taxa (Woelkerling 1988;

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Woelkerling & Foster 1989). The circumscription genera varies so much that some authorities recognize some taxa that are not accepted, or are only considered provisional, by others (e.g., cf. Cabioch & Giraud 1978; Woelkerling & Irvine 1986; Woelkerling 1988; Campbell & Woelkerling 1990; Chamberlain 1990, 1996; Wilks & Woelkerling 1994; Chamberlain & Keats 1994). The phylogenetic significance, if any, of many organismal features used to delimit coralline taxa remains unclear. In the recent literature, two intraordinal classification systems for the Corallinales have been employed which differ in both objective and scope. Cabioch (1971,1972) divided the Corallinaceae into five subfamilies including the Schmitzielloideae, Sporolithoideae, Lithothamnioideae, Lithophylloideae, and Corallinoideae. (Schmitziella endophloea, on which the Schmitzielloideae was based, has since been removed from the Corallinales). Cabioch postulated that, at the subfamilial level of classification, certain morphogenetic and vegetative (e.g., presence/absence of cell fusions and secondary pit connections) features were phylogenetically more informative than the presence or absence of genicula. Based on her observations (for review see Cabioch 1988), Cabioch concluded that geniculate forms had evolved independently from within three separate non-geniculate lineages. Thus, in Cabioch's scheme the subfamilies Corallinoideae and Lithophylloideae include both geniculate and non-geniculate taxa. Within these subfamilies the presence or absence of genicula is considered diagnostic of tribes and/or genera. Cabioch (1972,1988) contends that her arrangement of the coralline algae accurately reflects the natural history of the group. Cabioch's basic premise that some non-geniculate species may be more closely related to geniculate species than to other non-geniculate species has been accepted, albeit with critical reservation, by other authorities (Johansen

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1969a,b, 1976; Adey & Johansen 1972; Adey & Macintyre 1973; Woelkerling 1988; Athanasiadis 1989,1995). In practice, however, this hypothesis has not influenced the intraordinal classification of the Corallinales. Instead, most ecological, taxonomic, and systematic studies of corallines have been performed within the taxonomic frameworks erected by Johansen (1969a, 1976, 1981) and Woelkerling (1988; Woelkerling & Harvey 1995). In this system, eight subfamilies are recognized; three include geniculate species only and five include only non-geniculate species. The classification of non-geniculate species by Woelkerling (1988; Harvey & Woelkerling 1995) is an explicitly phenetically-based system that is designed to organize all available systematic information for the group efficiently. Because the classification scheme of Johansen (1976, 1981) and Woelkerling (1988; Harvey & Woelkerling 1995) is likely to be more familiar to those not specializing in the group, it has been selected to provide the basis for discussions of coralline systematics in this study. Attempts to devise a natural classification of the group based on organismal characters are plagued by a number of problems. The polarity of many organismal character states used in coralline systematics is uncertain. Hypothesized character-state transformation series lack objective support and instead rely on assumptions made based on the experience of the investigator. Although corallines are well-represented in the fossil record (Johnson 1961; Wray 1977) most systematically informative vegetative and reproductive features are rarely, or imperfectly, preserved. Thus, the fossil record is of limited value for helping determine plesiomorphic and derived character states within the group. In addition, recent studies have demonstrated that some organismal characters traditionally used to delimit non-geniculate coralline taxa aretaxonomically, and therefore phylogenetically, unreliable (Woelkerling 1985,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1987; Penrose & Woelkerling 1988; Campbell & Woelkerling 1990; Penrose 1991,1992; Woelkerling & Harvey 1992; Penrose & Chamberlain 1993). Perhaps most importantly, many species have simply not been thoroughly investigated in a modem context and remain poorly known to science. The lack of relevant information for some taxa contributes to questions regarding the distribution and taxonomic utility of various characters. For this reason, only recently have sufficient data accumulated to allow reproductive features to be included among those criteria used for generic delineation within the group (Keats etal. 1996). Persistent problems such as these must eventually be resolved before robust phylogenetic trees can be inferred from organismal data for coralline algae. In this investigation an independent molecular data set was used to assess the phylogenetic relationships among 37 species of coralline red algae. In a separate study Bailey & Chapman (1996) constructed a phylogenetic hypothesis for 23 species classified within the Corallinaceae based on an analysis of nuclear-encoded small-unit ribosomal RNA (SSU) gene sequences. The results of that investigation were inconsistent with present systems of classification for the group. In the present investigation molecular systematics studies of the Corallinales have been extended by determining SSU sequences for representatives of several phylogenetically important lineages not included in the previous study. The new data matrix was used to: 1) test conflicting hypotheses of relationships among coralline taxa at the suprageneric and intergeneric levels based on organismal features, 2) test the robustness of the earlier SSU-based hypothesis, and 3) examine, using character optimization techniques, the phylogenetic utility of organismal characters presently used to delimit the families and subfamilies of the Corallinales.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MATERIALS AND METHODS

Taxon sampling. Species included in this study and the number of bases determined for the SSU gene of each are listed in Table 2.1. The SSU gene sequences are deposited in GenBank under the following accession numbers: U60738-U60746, U60943-U60948, and U62113-U62120. Collection information for each species is presented in Appendix B. Liquid-preserved and dried voucher specimens will eventually be transferred to the Jepson Herbarium at the University of California at Berkeley (UC).

DNA extraction. Total cellular DNA was extracted from coralline red algae as follows: All centrifugations were performed at room temperature in either a Beckman J2-HC floor-model ultracentrifuge equipped with rotor type JS-13.1 (Beckman Instruments, Inc., Palo Alto, CA, USA) or Eppendorf model 5415 C table-top microcentrifuge with rotor F-45-18-11. Stock lysis buffer was prepared by combining 12.5 mL 2 M Tris-HCI, pH 7.5, 50 mL 0.5 M ethylene diamine tetracetic acid [EDTA], pH 8.0, and 15g of sodium dodecyl sulfate [SDS] (Lee & Taylor 1990). The solution was brought to a final volume of 500 mL with distilled water and sterilized prior to use. One - 10g of non-decalcified material was ground in liquid nitrogen. Heavily calcified non-geniculate species were crushed with a hammer before grinding. Ground tissue was then combined (in 2 mL or 15 mL plastic tubes depending on the amount of starting material) with pre-heated (60° C) stock lysis buffer to which had been added 5% [w:v] polyvinylpyrrolidone (MW 40 KD; PVP-40) and 10% [v:v] 2- mercaptoethanol in a ratio of 1g tissue: 3.5 volumes of lysis buffer. This algal/lysis buffer slurry was incubated at 60° C for 1 hr and mixed at 10 min

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.1. List of 37 species of coralline red algae and number of SSU base positions determined for each.

Number of Taxon Bases

Corallinaceae Corallinoideae Arthrocardia filicula (Lamarck) Johansen 1649 Bossiella califomica ssp. schmitii (Manza) Johansen 1652 Bossiella orbigniana ssp. dichotoma (Manza) Johansen 1688 Calliarthron cheilosporioides Manza 1620 Calliarthron tuberculosum (Postels et Ruprecht) Dawson 1682 Cheilosporum sagittatum (Lamouroux) J. Areschoug 1655 Corallina elongata Ellis et Solander 1704 Corallina officinalis Linnaeus 1786 ♦ * 'Corallina sp.?' 1724 Haliptilon roseum (Lamarck) Garbary et Johansen 1599 Jania crassa Lamouroux 1717 Jania rubens (Linnaeus) Lamouroux 1358 Serraticardia macmillanii (Yendo) Silva 1725 Amphiroideae Amphiroa sp. (AUS) 1717 Amphiroa sp. (SAF) 1581 Amphiroa fragilissima (Linnaeus) Lamouroux 1728 Lithothrix aspergillum J. E. G ray 1702 Metagoniolithoideae Metagoniolithon chara (Lamarck) Ducker 1699 Metagoniolithon radiatum (Lamarck) Ducker 1518 Metagoniolithon stelliferum (Lamarck) Weber-van Bosse 1573 Lithophylloideae Lithophyllum kotschyanum (Unger) Foslie 1714 Mastophoroideae Spongites yendoi (Foslie) Cham berlain 1712 Melobesioideae Clathromorphum com pactum (Kjellman) Foslie 1636 Clathromorphum parcum (Setchell et Foslie) Adey 1666 Lithothamnion glaciate Kjellman 1729 Lithothamnion tophiforme Unger 1683 Leptophytum acervatum (Foslie) Chamberlain et Keats 1723 Leptophytum ferox (Foslie) Chamberlain et Keats 1676 'Leptophytum foveatumT 1664** Mastophoropsis caniculata (Harvey et Hooker) Woeikerling 1423 Mesophyllum englehartii (Foslie) Adey 1647 Mesophyllum erubescens (Foslie) Lemoine 1624 (table con'd.)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62 Phymatolithon laevigatum (Foslie) Foslie 1685 Phymatolithon lenormandii (J. Areschoug) Adey 1678 Synarthrophyton patena (Hooker et Harvey in Harvey) Townsend 1623 Sporolithaceae Heydrichia woelkerlingii Townsend, Chamberlain et Keats 1758 Sporotithon durum (Foslie) Townsend et Woelkerling 1575

* Ragan etal. (1994), GenBank accession no. L26184 ** Sequences for these species not submitted to GenBank (see text).

intervals. Samples were then spun for 5 min at 2000 x g to remove larger pieces of material and cellular debris from the solution. After centrifugation, the supernatant was transferred to new tubes, extracted with an equal volume ofphenol: chloroform (49:1), and spun for 8 min at 3000 x g. The aqueous phase was collected, transferred to new tubes, extracted with an equal volume of phenol: chloroform: isoamyl alcohol (25:25:1), and spun for 8 min at 3000 x g. After spinning, the aqueous fraction was removed and re-extracted with phenol: chloroform as above. Following centrifugation, the aqueous phase was transferred to new tubes, 0.8 volumes of isopropyl alcohol was added, and the solution was mixed for 2 min. The sample was then centrifuged at 16000 x g for 3 min and stored at -20° C for 1 hr. After freezing, the sample was re centrifuged at 16000 x g for 10 min. Nucleic acid pellets were air dried and

100-200 jiL of sterile water was added. Due to the presence of contaminating polysaccharides which co-precipitated with the nucleic acid pellet at this step, pellets were either placed in an incubator at 37° C for 15 - 45 min or were stored overnight at 4° C to resuspend the pellets. To the solution, an equal volume of 4M LiCI was then added to precipitate RNA, mixed for 2 min, spun at 16000 x g for 3 min, and stored at 4° C for 1 hr. After refrigeration, the sample was spun at 16000 x g for 10 min and the DNA-containing supernatant was transferred to new tubes. An equal volume of ice-cold 100% ethanol and 0.1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. volumes of 4M NaCI was then added to the DNA solution(s), samples were spun for 3 min at 16000 x g, and stored at -20° C for 1 hr. Following freezing, DNA pellets were precipitated by spinning at 16000 x g for 10 min. DNA pellets were air-dried, 100-150 pL of sterile water were added, pellets were put into solution by incubation at 37° C or overnight refrigeration as above, and then combined into a single sample if necessary. Finally, to remove any remaining contaminants (polysaccharides) DNA samples were purified with a GeneClean II Kit (Bio 101, La Jolla, CA, USA) following the vendor's suggested specifications. DNA was resuspended in 100-150 pL of sterile water and stored until needed at -20° C.

PCR amplification and sequencing of SSU genes. Oligonucleotide primers used to amplify and sequence the nuclear-encoded 18S rRNA gene included both primers originally developed for higher land plants and green algae (Hamby etal. 1988) and primers designed specifically for rhodophycean taxa (Saunders & Kraft 1994). Template DNA was polymerase chain reaction (PCR) amplified in 0.5 mL micro-centrifuge tubes (USA/Scientific Plastics, Ocala, FL, USA) in a total reaction volume of 100 pL Each amplification reaction consisted of 5-20 pL of a 10-2 dilution of DNA template, 4 pL each of the two flanking primers (10 pM), 6 pL MgCfe (25 mM), 6 pL of 1 mM deoxynucleotide mix (containing dATP, dTTP, dCTP, and dGTP), 10 pL 10X PCR reaction buffer (Perkin-Elmer, Branchburg, NJ, USA), 0.25 pL Tfl DNA polymerase (Epicentre Technologies, Madison, Wl, USA), and an appropriate amount of sterile water. Negative controls, lacking target DNA template, were included for each set of reactions. Double-stranded (symmetric) amplifications were performed in a Perkin-Elmer Cetus DNA thermocycler with an initial denaturation step of 95° C for 3 min, followed by 27 cycles of 1 min at 95° C,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. primer annealing for 1 min at 50° C, and extension for 2 min at 72° C. PCR amplification products were checked on ethidium bromide-stained 0.8% agarose 1X TBE gels for correct length, (against a standard 1 KB DNA ladder; Life Technologies, Inc. Gaithersburg, MD, USA) yield, and purity. Unincorporated dNTPs, excess primers, Tfl polymerase and salts were removed from pooled DNA amplification products with Millipore UFC3 TTK 00 (30,000 NMWL) filter units (Millipore Corporation, Bedford, MA, USA) in a variable speed Eppendorf 5415 C table-top microcentrifuge and amplification products were brought to a concentration of ca. 5-40 ng DNA/pL. Purified amplification products were sequenced following the manufacturer's specifications for the AmpliCycle® sequencing protocol (Perkin-Elmer). Sequencing primers were end-labeled with 22P or 33p dATP (DuPont, Boston, MA, USA) to allow visualization of the resulting fragments by autoradiography. DNA fragments were separated on 6% Long Ranger (FMC, Rockland, ME, USA) gels according to standard methods (Sambrook etal. 1989).

Sequence alignment and data analysis. Sequences were first aligned automatically with the PILEUP program (UNIX v. 8.0, Genetics Computer Group 1994) and then edited interactively using the Sequence Editor program (Olsen et al. 1991). Regions of the data matrix that could not be unambiguously aligned were excluded from subsequent analyses. Including gaps, the resulting data matrix, including all species, contains 1739 positions. Maximum parsimony and maximum likelihood analyses (see below) of these data were rooted with the SSU sequence for Rhodogorgon carriebowensis, which was kindly provided by Dr. G. W. Saunders. Because much of the sequence data excluded from the alignment containing all species are conserved among species classified in the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subfamilies Melobesidoideae and Corallinoideae, separate alignments were constructed for these taxa. Phylogenetic analyses of the data matrices for the Melobesioideae (1736 positions) and Corallinoideae (1744 positions) were rooted with Corallina officinalis and Amphiroa fragilissma, respectively. Phylogenetic hypotheses of relationships among coralline species were inferred using both maximum parsimony and maximum likelihood optimality criteria. Heuristic parsimony analyses were performed in PAUP (v. 3.1.1., Swofford 1993) using the tree bisection-reconnection (TBR) algorithm with random sequence addition (500 replicates), COLLAPSE and MULPARS options in effect. All sites were weighted equally and alignment gaps were treated as missing data. Support for nodes was assessed by calculating bootstrap proportions based on 100 resamplings of the set (Felsenstein 1985). Maximum likelihood (ML) analyses were performed with the fastDNAml program (v. 1.0.8, Olsen etal. 1994). The optimal transition:transversion (tn:tv) ratio was determined separately for each data set by plotting a range of tn:tv values against their corresponding inferred log likelihoods (data not shown). The optimal tn:tv was determined to be 2.15 for the taxon-replete matrix, 2.40 and 2.00 for the separate melobesioidean and corallinoidean data matrices, respectively. These values were used in all subsequent replicate ML analyses of the data sets. ML analyses were replicated ten times for each data set using empirical base frequencies and random orders of sequence addition. Local rearrangements were permitted during the sequential addition phase of the algorithm and global rearrangements were performed on the completed tree(s). User-defined analyses and the log likelihood ratio test (Felsenstein 1988) were used to determine if ML trees differed significantly from the parsimony trees. To examine the phylogenetic utility of seven morphological characters used to delineate the families and subfamilies of the Corallinales, character

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. states for each of these features were coded in binary form for each species and appended to the SSU data matrix. The seven characters, their possible character states, and the morphological data matrix are presented in Appendix C. Cladistic analysis of the combined SSU and morphological data was performed as above. Morphological character states were traced on the tree(s) under both delayed (DELTRAN) and accelerated (ACCTRAN) transformation criteria as implemented in the MacClade software package (v. 3.0, Maddison & Maddison 1992).

RESULTS

For parsimony, the taxon-replete data set contained 269 phylogenetically informative positions. Cladistic analysis of these data yielded 72 equally most parsimonious trees of 853 steps. The strict consensus of these trees is presented in Figure 2.1. Replicate ML analyses of the taxon-replete data matrix resulted identical trees (Ln likelihood = -7718.11453) that did not differ significantly from the parsimony trees (p<0.05). The consensus tree shown in Figure 2.1, with the exception of only the inclusion of new species herein, is consistent with the phylogenetic hypothesis for the Corallinales presented by Bailey & Chapman (1996). Four major lineages are resolved within the order (Figure 2.1). The Sporolithaceae, including Sporolithon durum and Heydrichia woelkerlingii, is resolved as the earliest divergence within the order and forms the sister group to a monophyletic (BP=100%) Coraliinaceae. The bootstrap confidence value for the sister group relationship between these species in notably low (51%). This is a result of the fact that the SSU sequences of Sporolithon durum and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Heydrichia woelkerlingii Sporolithon durum Amphiroa s p . (AUS) Amphiroa sp. (SAF) Amphiroa fragilissima Lithothrix aspergillum Lithophyllum kotschyanum Metagonioiithon chara Metagoniolithon radiatum Metagonioiithon stelliferum Spongites yendoi Cheilosporum sagittatum Haliptilon roseum Jania crassa Jania rubens Corallina officinalis Corallina elongata ■'Corallina sp. ?' Bossiella californica ssp. schmitii Bossiella orbigniana ssp. dichotoma Calliarthron cheilosporioides Calliarthron tuberculosum Serraticardia macmillanii Arthrocardia filicula Mesophyllum englehartii 'Leptophytum foveatum?' Mesophyllum erubescens Leptophytum acervatum Leptophytum ferox Synarthrophyton patena v— Clathromorphum compactum t— Clathromorphum parcum Mastophoropsis caniculata Phymatolithon laevigatum Phymatolithon lenormandii Lithothamnion glaciale Lithothamnion tophiforme

Figure 2.1. Strict consensus of 72 equally most parsimonious trees (Cl=0.56; Rl=0.77) resulting from a cladistic analysis of SSU gene sequences for 37 species of coralline algae. Circled numbers indicate the four major lineages resolved within the Corallinales and are numbered according to the order in which they are discussed in the text. Bootstrap values (> 50%) for each node are indicated above branches.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Heydrichia woelkerlingii are almost equally as likely to be associated with the outgroup, Rhodogorgon carnebowensis, as they are with each other. These results suggest that the Rhodogorgonales and Corallinales may be much more closely related that previously thought (Norris & Bucher 1989; Ogden 1992; Pueschel e ta i 1992a; Fredericq & Norris 1995). In fact, if any other red algal SSU sequence is added to the alignment and is selected as the outgroup Rhodogorgon carnebowensis is resolved within the ingroup (i.e., the corallines) (data not shown). The relationship between the Rhodogorgonales and Corallinales is discussed further in Chapter 3. The non-geniculate subfamily Melobesioideae is resolved as a monophyletic group and represents the second major lineage within the Corallinales. Relationships among most genera within the subfamily are not clearly resolved in the taxon-replete tree (Figure 2.1). For parsimony, the separate alignment constructed for the Melobesioideae contained 109 informative positions. Cladistic analysis of the melobesioidean data matrix resulted in a single most parsimonious tree of 302 steps (Figure 2.2). All ten replicate ML analyses of this data set resulted in trees which were identical to one another (Ln likelihood = -4302.36483) and the parsimony tree (Figure 2.2). All genera within the subfamily represented by at least two species are resolved as monophyletic, excepting Leptophytum (see discussion). Bootstrap values indicate that most nodes of the tree are moderately to strongly supported; however, the relationships of Synarthrophyton patena and Mastophoropsis caniculata to other melobesioidean species remains unclear. The third major clade includes geniculate species classified within the subfamily Corallinoideae (Figure 2.1). The subfamily comprises two sister clades, but relationships among some genera within these subgroups are either unresolved or are poorly supported by bootstrap confidence values. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69

Mesophyllum englehartii

'Leptophytum foveatum?'

Mesophyllum erubescens 87 100 Leptophytum acervatum

Leptophytum ferox

Synarthrophyton patena

70 Clathromorphum compactum

Clathromorphum parcum

Mastophoropsis caniculata

100 Phymatolithon laevigatum CPhymatolithon lenormandii 100 r " Lithothamnion glaciate

Lithothamnion tophiforme

Figure 2.2. Single most parsimonious phylogram (Cl=0.70; Rl=0.60) depicting phylogenetic relationships inferred among 13 melobesioid species. Bootstrap values (> 50%) for each node are indicated above branches.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. separate corallinoidean alignment contained only 62 phylogenetically informative positions. Cladistic analysis of these data resulted in 18 equally most parsimonious trees of 272 steps. The strict consensus of these 18 trees is presented in Figure 2.3. All replicate ML analyses of the corallinoidean data matrix yielded identical trees (Ln likelihood = -3994.02150) that are consistent with the parsimony tree (Figure 2.3). The topologies of parsimony and ML trees found following analyses of the separate alignment differ only slightly from that of corallinoidean subtree shown in Figure 2.1. In these trees Jania crassa is resolved as sister to Haliptilon roseum (Figure 2.3) and not as sister to a clade including H. roseum and Cheilosporum sagittatum. Although the monophyly of the subclade is well supported (Figures 2.1 & 2.3), robust character support for hypotheses of relationships among the genera Jania, Cheilosporum, and Haliptilon is lacking. In none of the analyses performed here are the genera Jania and Bossiella resolved as monophyletic taxa. For Jania, this observation can almost certainly be attributed to the fact that the sequence of Jania rubens is less complete than that of Jania crassa (Table 2.1). In most optimal solutions, Bossiella californica ssp. schmittii is resolved as sister to an unresolved polychotomy including species of Calliarthon and Serraticardia, as well as Bossiella orbigniana ssp. dichotoma (e.g., Figure 2.3). Sequence divergence values among species within this group of genera are very low (0.8%) and emphasize the fact that phylogenetic signal in this area of the tree is weak. More variable data are needed to determine the relationships among the latter genera. For these reasons, the hypotheses that Jania and Bossiella are not natural taxa are not considered further in this study. Crustose field-collected material for the taxon labeled 'Corallina sp.?' was originally identified as a non-geniculate coralline. The SSU sequence of this entity, however, differs from that of Corallina officinalis and Corallina

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. M Cheilosporum sagittatum Haliptilon roseum

Jania crassa

Jania rubens

Corallina officinalis A 'Corallina sp.?1 -H 60 Corallina elongata 75 Bossiella californica ssp. schmitii L Bossiella orbigniana ssp. dichotoma 94 81 Calliarthron cheifosporioides

Calliarthron tuberculosum

Serraticardia macmillanii

Arthrocardia filicula

Figure 2.3. Strict consensus of 18 equally most parsimonious trees (Cl=0.84; Rl=0.71) inferred from SSU gene sequences for 13 corallinoidean species. Mapped character states indicate conceptacle positions for genera within the group: A=axial, M=marginal, and L=lateral. Bootstrap values (> 50%) for each node are indicated above branches.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. elongata by only 0.7% and is resolved as closely related to the latter species in all analyses (Figures 2.1 & 2.3). This result clearly suggests that the "crust" collected is actually the basal crust (holdfast) of a species of the geniculate genus Corallina. Excluding the sequence fo r 'Corallina sp.?’ from parsimony or ML analyses of the taxon-replete or corallinoidean data matrices has no effect on the topology of the resulting trees and was therefore included in the study. The SSU data strongly (BP=97%) support the monophyly of a fourth clade comprised of species classified in the non-geniculate subfamilies Mastophoroideae and Lithophylloideae and the geniculate subfamilies Metagoniolithoideae and Amphiroideae (Figure 2.1). The clade is subtended by Spongites yendoi (Mastophoroideae), which diverges prior to a clade including species of Metagonioiithon. These lineages are, in turn, allied to species representing the subfamilies Lithophylloideae and Amphiroideae, which are resolved as sister taxa. Within the latter assemblage of taxa, the subfamilies Metagoniolithoideae (BP=100%) and Amphiroideae (BP=71%) are resolved as monophyletic. Analysis of the combined SSU and morphological data set resulted in 72 equally most parsimonious trees of 863 steps. These trees are identical to those found following the analysis of SSU gene sequences only. A minimum of ten steps are required to fit the character states of the seven morphological characters onto the tree. The inferred path of evolution for the morphological characters is graphically depicted on the "reduced" (cf. Figure 2.1) cladogram shown in Figure 2.4. Character states for six of the seven characters are fitted onto the tree without conflict (Cl=1.00) and identically under both DELTRAN and ACCTRAN optimization criteria. The seventh character, presence or absence of genicula, is more homoplastic (Cl=0.33). DELTRAN optimization of the latter feature suggests that genicula arose independently in three separate

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73

G 2 0 p » 4 Amphiroideae

Lithophylloideae

+— Metagoniolithoideae P- ■■■■■ Mastophoroideae SC i-H Corallinoideae MC + ■■■■■■ Melobesioideae S Tc F/2C Sporolithon H— h “ t - Heydrichia

Figure 2.4. Simplified cladogram depicting relationships inferred among the families and subfamilies of coralline algae included in this study. Key to mapped organismal character states: P-=sporangial plugs absent, 2C=secondary pit connections present, F/2C=both cell fusions and secondary pit connections present, S=sorus, SC=sporangial conceptacles present, Tc=division of tetrasporangium cruciate, MC=sporangial multiporate, G=genicula present. See Appendix B and discussion for further details.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lineages of coralline algae (Figure 2.4). The ACCTRAN character state optimization criterion suggests that genicula were present in the common ancestor of the subfamilies Corallinoideae, Amphiroideae, Metagoniolithoideae, Mastophoroideae and Lithophylloideae but were independently lost in the ancestors of the latter two (non-gen icu late) groups. The latter scenario is inconsistent with paleontological evidence (Johnson 1956,1961; Wray 1977), which indicates beyond any reasonable doubt that non-geniculate algae of the melobesioid, lithophylloid, and mastophoroid lineages precede all geniculate corallines in the fossil record. Based on these independent data, the ACCTRAN result is rejected in favor of the hypothesis for character state assignments at internal nodes of the tree suggested by DELTRAN optimization (Figure 2.4).

DISCUSSION

Sporolithaceae. The family Sporolithaceae (Verheij 1993) includes two genera, Sporolithon and Heydrichia. The molecular data support the monophyly of the Sporolithaceae; Sporolithon durum and Heydrichia woelkerlingii are inferred to be sister taxa. In relative terms, these genera differ considerably from other coralline algae. Sporolithon and Heydrichia lack tetrasporangial conceptacles found in all other corallines; instead, tetrasporangia in these genera develop within "sporangial complexes," or sori (Verheij 1992,1993; Townsend etal. 1994,1995). Other autapomorphic features of the group include the presence of both cell fusions and secondary pit connections as well as the cruciate pattern of cleavage of the tetrasporangium. Unlike most other algal taxa, the calcareous corallines have an extensive fossil record (Johnson 1961,1963; Adey & Macintyre 1970; Wray

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1977; Poignant 1985). The extant genus Sporolithon (= Archaeolithothamnion) has a continuous fossil record extending at least into the late Jurassic period (ca. 165 million years ago; mya) (Wray 1977). Based on these anatomical and fossil data, most corallinologists concur that Sporolithon likely represents the oldest extant lineage within the Corallinales (Johnson 1956; Cabioch 1970; Townsend 1982). This hypothesis is supported by the molecular data which indicate that Sporolithon and Heydrichia are the earliest-diverging taxa within the Corallinales. Before the family Sporolithaceae was erected by Verheij (1993), Sporolithon was classifed either within the subfamily Melobesioideae (e.g., Adey etal. 1982; Johansen 1981; Woelkerling 1988) or in a separate subfamily (the Sporolithoideae sensu Cabioch 1971,1972,1988). The notion that Sporolithon may be closely related to the Melobesioideae is based on the observation that Sporolithon shares certain vegetative anatomical and reproductive features with melobesioid taxa, and, in particular, Lithothamnion (Adey 1970; Adey etal. 1982). Although resolved as sister taxa, the long branch (70 steps) uniting the Sporolithaceae with other corallines implies that the phylogenetic relationship between the Sporolithaceae and the subfamily Melobesioideae is not a close one.

Melobesioideae. The subfamily Melobesioideae includes ten putative genera that are delimited from one another by a combination of nine characters (Chamberlain & Keats 1994). Members of this diverse taxon are easily distinguished from other non-geniculate corallines by the presence of both multiporate bi-/tetrasporangial conceptacles and apical plugs that occlude the pore above each sporangium prior to spore release (Woelkerling 1988; May & Woelkerling 1988). The taxonomic status of most genera within the group is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. uncertain and generic boundaries are considered "provisional"(Chamberlain & Keats 1994). Phylogenetic trees inferred from SSU gene sequence data support the monophyly of the Melobesioideae (Figure 2.1). The molecular data also provide strong support for several independent lineages (genera) within the subfamily, but the interrelationships among these lineages are not resolved robustly in all cases. The results do, however, provide valuable new insights into several taxonomic problems within the group. The taxonomic disposition of the genus Phymatolithon is not universally agreed upon by corallinologists. Phymatolithon has generally been considered distinct from, but closely related to, Lithothamnion (e.g., Adey 1970; Woelkerling & Irvine 1986; Chamberlain & Keats 1994). According to the latter authorities, Phymatolithon can be delimited from Lithothamnion on the basis of three vegetative features, viz. the relative lengths of subepithallial initial cells, epithallial cell morphology, and the mode of cell elongation in the region of the subepithallial initials (see also Adey & Johansen 1972). In contrast, Giraud and Cabioch (1978) did not consider the above features to be of significance at the generic level and, instead, recognized Phymatolithon as one of three (with Clathromorphum and Leptophytum) subgenera of Lithothamnion. Wilks and Woelkerling (1994) have also argued that the characters presently used to distinguish Phymatolithon from Leptophytum (Chamberlain & Keats 1994) need to be re-evaluated critically and concluded the the latter genera cannot, as presently diagnosed, be easily delimited from one another. The SSU data support the notion that Phymatolithon and Lithothamnion are closely related taxa. Pairwise comparisons indicate that, on average, the SSU sequences for species of Phymatolithon and Lithothamnion differ from one another by only 3%. In the taxon-replete tree (Figure 2.1) species of Phymatolithon and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77 Lithothamnion are resolved as a monophyletic group within the subfamily. In the separate parsimony and ML analyses of SSU data for melobesioidean taxa (Figure 2.2), Lithothamnion diverges first followed by Phymatolithon. With the latter data set, however, a penalty of only four additional steps is incurred if Phymatolithon and Lithothamnion are constrained to monophyly as shown in Figure 2.1. These results confirm the notion that the latter two genera are closely related. Although there is some question about whether the characters presently used to delimit Phymatolithon from Leptophytum will ultimately prove to be taxonomically useful (Wilks & Woelkerling 1994), the SSU data indicate convincingly that Leptophytum represents a lineage that is distinct from Phymatolithon and Lithothamnion. In addition, the SSU trees (Figures 2.1 & 2.2) do not support the hypothesis that Clathromorphum and Leptophytum are closely related to Lithothamnion (sensu Giraud & Cabioch 1978). The taxonomic disposition of Leptophytum also has been the subject of considerable debate (Woelkerling & Irvine 1986, Chamberlain & Keats 1994; Wilks & Woelkerling 1994). The SSU data indicate that Leptophytum acervatum and Leptophytum feroxate very closely related and represent an independent lineage within the subfamily. The SSU sequence for 'Leptophytum foveatum', however, strongly (BP=100%) allies this taxon with Mesophyllum engelhartii within a clade that also includes Mesophyllum erubescens. This result is consistent with earlier observations as well (Bailey & Chapman 1996). Under parsimony, a minimum penalty of 18 additional steps is assessed if the genus Leptophytum is constrained to be monophyletic. Comparsions of the SSU sequences themselves, and the results of the phylogenetic analyses presented here, strongly suggest that the 'Leptophytum foveatum' isolate has been misidentified and that the species is probably attributable to Mesophyllum.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78 If so, the SSU data appear to support the recognition of Leptophytum as a distinct genus within the Melobesioideae. Adey and Johansen (1972) recognized two tribes within the Melobesioideae, the Phymatolitheae (comprising the genera Phymatolithon, Leptophytum, and Kvaleya) and Lithothamnieae (Areschoug 1852; Lebednik 1977) including all other melobesioid genera. These tribes were established on the basis of variation among features including meristematic cell length, site of cellular elongation, and the anatomical origin of conceptacle primordia. The SSU sequence data do not support the monophyly of either tribe and indicate that both are paraphyletic. These results suggest that the features considered diagnostic of the tribes listed above have evolved convergently and are not phylogenetically informative. The paleontological record for the Melobesioideae is not a diverse one and most fossil representatives of the group have been assigned to the genus Lithothamnion, which first appears in the late Jurassic (ca. 165 mya). Because of similarities in their vegetative anatomy and the imperfectly preserved condition of most specimens, it is likely that Lithothamnion also includes the fossil ancestors of other melobesioid genera. Putative fossil representatives of Lithothamnion precede other distinct melobesioid genera (e.g., Mesophyllum and Melobesia) in the fossil record by ca. 90- 100 million years (Wray 1977). The early appearance of Lithothamnion in the fossil record, and its supposed close relationship to Sporolithon (e.g., Adey etal. 1982), have been considered as evidence suggesting that the genus is the most primitive among melobesioid taxa. This hypothesis is consistent with the SSU gene trees (Figures 2.1 & 2.2) which indicate that Lithothamnion (or Lithothamnion/Phymatolithon) is the basal lineage within the Melobesioideae and that the genus diverged earlier than Mesophyllum (Wray 1977).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Corallinoideae. The primary taxonomic feature distinguishing the subfamily Corallinoideae from other subfamilies within the order is the possession of unizonate, ecorticate genicula. The genera of Corallinoideae are classifed in two tribes which are delimited from one another on the basis of a number of morphological and reproductive features (Johansen & Silva 1978; Garbary et al. 1981; Garbary & Johansen 1982). The tribe Janieae is comprised of the genera Jania, Haliptilon, and Cheilosporum, and the tribe Corallineae includes all other corallinoidean genera. The SSU data support the monophyly of the tribes Janieae and Corallineae and indicate that these lineages diverged from one another early in the evolutionary history of the group (Figures 2.1 & 2.3). Due to low SSU gene sequence variation and missing data for some included species (Table 2.1) relationships among genera within either tribe are not clearly resolved. Johansen (1969a) regarded the anatomical origin of conceptacles to be of prime significance for classifying and understanding the evolutionary history of corallinoidean genera. Based on this postulate, and other considerations, Johansen (1969a, p. 68) presented a branching diagram depicting possible phylogenetic relationships among 11 corallinoidean genera. To test Johansen's proposal that conceptacle positions are phylogenetically informative, conceptacle positions were recorded for each genus included in this study (Table 2.2) and were mapped onto the corallinoidean tree(s). Regardless of whether the topology of Figure 2.1 or Figure 2.3 is used, three steps are required to fit the character states onto the tree. The SSU gene tree supports Johansen's contention that axial conceptacles are plesiomorphic and that marginal and lateral conceptacles are derived character states within the group (Figure 2.3). Johansen (1969a, p. 68) also speculated that "Bossiella is a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2.2. Conceptacle positions for the genera of Corallinoideae included in this study. Data summarized in Johansen (1969a, 1981).

Genus Conceptacle position

Arthrocardia A Bossiella L Calliarthron L Cheilosporum M Corallina A Haliptilon A Jania A Serraticardia L

A=apical, L=lateral, M=marginal

genus of relatively recent origin." The data also are consistent with this proposal and indicate that Bossiella is closely related to Calliarthron and Serraticardia. Bossiella, Calliarthron, and Serraticardia all bear lateral conceptacles, although axial conceptacles have been reported in the latter genus as well (Johansen 1969a). The earliest fossil Corallinoideae are referred to the genera Arthrocardia and Jania, which appear first in deposits of the late Cretaceous period (ca. 125 -100 mya) (Wray 1977). In the SSU trees (Figures 2.1 & 2.3) these genera diverge prior to others for which fossil representatives in the Mesozoic are unknown. Paleontological evidence indicates that most corallinoidean taxa evolved in the Cenozoic era (Johanson 1961; Wray 1977). Corallina appears in the Eocene (40 - 50 mya) and is followed by Calliarthron which has been found in deposits of the Miocene epoch (15-20 mya). These fossil data are consistent with, and provide independent support for, the relative order of branching of genera within the Corallinoideae inferred from the molecular data (Figure 2.3). The lack of SSU gene sequence variability among species of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Corallinoideae and the fossil data further suggest that this group is a recent divergence within the order.

Mastophoroideae, Metagoniolithoideae, Lithophylloideae, and Amphiroideae. The fourth major lineage of coralline algae resolved in this study includes species referred to the non-geniculate subfamilies Mastophoroideae and Lithophylloideae as well as the geniculate subfamilies Metagoniolithoideae and Amphiroideae. The observation that these taxa form a monophyletic group is not consistent with any previous taxonomic scheme proposed for the coralline algae (e.g., Cabioch 1971; Johansen 1976,1981; Woelkerling 1988). The geniculate subfamily Metagoniolithoideae includes a single genus, Metagonioiithon, and three species that are endemic to the continent of Australia. Cabioch (1972,1988) postulated that Metagonioiithon and certain mastophoroid genera (e.g., Neogoniolithon) are members of a distinct evolutionary lineage closely related to other mastophoroid genera and corallines possessing unizontate genicula (i.e., the Corallinoideae, sensu Johansen 1981). Comparing Metagonioiithon to other geniculate taxa, Ducker (1979) noted that although Metagonioiithon lacks secondary pit connections, in terms of reproductive features and their type of spore germination Metagonioiithon is most similar to the Amphiroideae. The SSU gene tree supports the notion that there is a phylogenetic link between Metagonioiithon and the non-geniculate subfamily Mastophoroideae. Metagonioiithon and the mastophoroid species Spongites yendoi are inferred to be sister taxa. The data do not, on the other hand, support Cabioch's contention that the latter assemblage is closely allied to the Corallinoideae. Instead, Metagonioiithon

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and Spongites are resolved within a separate clade that also includes species classified in the subfamilies Lithophylloideae and Amphiroideae. In species classified within the non-geniculate subfamily Lithophylloideae and the geniculate subfamily Amphiroideae, cell fusions are absent. In these subfamilies cells of adjacent filaments are, instead, joined by secondary pit connections. Johansen (1981) also commented that the smallest spores of coralline algae are those produced by species possessing secondary pit connections. De-emphasizing the taxonomic significance of the presence or absence of genicula, Cabioch classified all coralline taxa lacking cell fusions in a single subfamily (the Lithophylloideae, sensu Cabioch 1972). Although Cabioch's taxonomic concept of the Lithophylloideae has not been adopted by subsequent authorities, the idea that this arrangement may more accurately reflect phylogenetic relationships among these taxa has been widely acknowledged (e.g., Adey & Johansen 1972; Johansen 1976; Woelkerling 1988; Athanasiadis 1989,1995). The SSU sequence data indicate that the Lithophylloideae, represented by Lithophyllum kotschyanum, and the Amphiroideae are sister taxa. The bootstrap confidence value for the node conjoining Lithophyllum kotschyanum and the Amphiroideae is relatively high (BP=82%), indicating substantial character support for the relationship. The subfamiliy Amphiroideae, including the genera Lithothrix and Amphiroa, is resolved as a monopyletic group. Comparative studies of genicular anatomy and development made by Garbary and Johansen (1987) suggested to these authors that Amphiroa was derived relative to Lithothrix. This conclusion is consistent with the SSU gene tree in which Lithothrix aspergillum diverges prior to a strongly supported (BP=100%) subclade including three species of Amphiroa.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The fossil record for most genera included in this lineage is sparse and, for some, is limited to specimens found in Recent strata. Fossilized material for the non-geniculate members of this clade, the Lithophylloideae and Mastophoroideae, are not easily separated from fossil Melobesioideae unless secondary pit connections and/or sporangial conceptacles have been preserved. This is not often the case and it is possible that some fossils in both subfamilies have been assigned to the Melobesioideae as Lithothamnion. Of the genera within the group, two, Uthophyllum and Amphiroa, have extensive fossil records (Johnson 1961; Wray 1977). Lithophyllum has a continuous fossil record dating back to the late Jurassic or early Cretaceous periods (ca. 155 - 130 mya). The genus Amphiroa also appears in the early Cretaceous (ca. 120 mya), but in younger strata than does Lithophyllum. The SSU data indicate that the non-geniculate subfamily Lithophylloideae and geniculate subfamily Amphiroideae are sister taxa (Figure 2.1). The paleontological and phylogenetic evidence suggest that genicula may have evolved in the latter group within a frame of time of 10 - 35 million years.

Higher-level systematics of the Corallinales.The SSU tree is not consistent with any previously proposed system of classification for the Corallinales at the suprageneric level. A simplified cladogram depicting relationships among coralline families and subfamilies inferred from the molecular and morphological data is presented in Figure 2.4. Optimized onto this tree are those organismal features that are considered diagnostic of families or subfamilies within the order. The Sporolithaceae is delimited from the Corallinaceae by three autapomorphic features: 1) the production of tetrasporangia in sporangial complexes or sori [S], 2) the cruciate pattern of cleavage of tetrasporangia [Tc], and 3) the presence of both cell fusions and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. secondary pit connections [F/2C]. Because the Sporolithaceae is resolved as the basal sister group to all other coralline lineages, the above character states are interpreted to be plesiomorphic for the Corallinales. The monophyly of the Corallinaceae is supported by two synapomorphies, viz. the presence of sporangial conceptacles [SC] and the simultaneous zonate pattern of clevage of tetrasporangia. Presence of sporangial plugs is a plesiomorphic feature for coralline algae and one that unites the Sporolithaceae with the Melobesioideae. The latter two taxa are resolved as sister taxa diverging before all other coralline lineages which are characterized by the absence of sporangial plugs [P"J. Multiporate sporangial conceptacles [MC] are an autapomorphic feature of the Melobesioideae that is clearly diagnostic of this natural assemblage of species (Figure 2.1). These data suggest that multiporate-type sporangial conceptacles found in the Melobesioideae were derived from sori very much like, or identical to, those found in Sporolithon and Heydrichia and that, despite their anatomical and developmental differences, sori and conceptacles are homologous structures (Townsend etal. 1994). Uniporate sporangial conceptacles unite the subfamilies Corallinoideae, Mastophoroideae, Metagoniolithoideae, Lithophylloideae, and Amphiroideae to the exclusion of other coralline taxa. Secondary pit connections are inferred to have been lost in, or before the appearance of, the last common ancestor of the subfamilies Melobesioideae, Corallinoideae, Mastophoroideae, and Metagoniolithoideae; the latter subfamilies possess cell fusions. Secondary pit connections [2C] are independently derived along the common branch uniting the subfamilies Lithophylloideae and Amphiroideae and are synapomorphic for the latter clade. The hypothesis for a single evolutionary origin of the geniculum [G] is rejected. The SSU gene tree strongly indicates that the genicula of the Amphiroideae, Corallinoideae, and Metagoniolithoideae have evolved convergently and are not

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. homologous (Ridley 1993). As a corollary, it follows that any further comparisons of anatomical differences between genicula are comparisons of analogous features. Thus, although of undoubted taxonomic significance, features including whether or not genicula are uni- or multizonate, corticate or ecorticate, meristematic or non-meristematic, etc. are of no phylogenetic value when considered in the context of geniculate species alone. Thus, the SSU data support the hypothesis that most features used to delimit the families and subfamilies of the Corallinales are phylogenetically, as well as taxonomically, informative. The presence or absence of genicula is, however, not a phylogenetically informative feature for the coralline algae. Ribosomal gene sequences for species classified in the non-geniculate subfamilies Choreonematoideae and Austrolithoideae have yet to be determined. Nevertheless, the results of this study and a recent electron microscopic investigation (Broadwater & LaPoint, in press) provide important clues for resolving the natural affinities of these taxa to other coralline algae. Woelkerling (1987) established the Choreonematoideae for a single host-associate species, Choreonema thuretii, and delimited the subfamily from others based on a distinctive combination of vegetative and reproductive features. Based on light and scanning electron microscopic observations, Woelkerling (1987) concluded that in Choreonema: 1) cell fusions, epithallial cells, and haustoria are absent, 2) sporangial plugs are present, and 3) sporangial conceptacles are of the uniporate type. Woelkerling's assertion that Choreonema lacks epithallial cells is inconsistent with previous studies of the species made by Suneson (1937) and Kylin (1956). Woelkerling (1987,1988) also found no clear evidence suggesting that Choreonema is nutritionally dependent on its host(s). However, transmission electron microscopic studies clearly show that Choreonema is a true parasite and, contrary to Woelkerling

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1987), that the species possesses both cell fusions and epithallial cells (Broadwater & LaPoint, in press). In addition, and more importantly, it has also been found that the roofs of sporangial conceptacles of Choreonema are in fact multipored and not unipored (Broadwater, pers. comm.). These new findings indicate that there are no longer any features that can be used to distinguish the subfamilies Choreonematoideae and Melobesioideae (see Woelkerling 1988; Chamberlain & Keats 1994) from one another. Multipored sporangial conceptacles, sporangial plugs, and the absence of haustoria are also characteristic of the genera Austrolithon and Boreolithon, which are classified in the subfamily Austrolithoideae (Harvey & Woelkerling 1995). Unlike Choreonema, however, cell fusions are unknown in Austrolithon and Boreolithon. Thus, there is at present but a single useful (?) feature (presence or absence of cell fusions) for distinguishing the Austrolithoideae from the subfamilies Choreonematoideae and Melobesioideae. In this study SSU gene sequence trees support the hypotheses that the presence of multiporate sporangial conceptacles, sporangial plugs, and cell fusions (only) are phylogenetically informative characters for coralline algae. Together, the ultrastructural and molecular data clearly indicate that the taxonomic status of the Choreonematoideae and Austrolithoideae should be re-evaluated. The balance of evidence strongly suggests that the genera Choreonema, Austrolithon, and Boreolithon should be transferred from their respective subfamilies and reclassified within the Melobesioideae. Anatomical studies are needed to confirm whether or not cell fusions are indeed lacking in the Austrolithoideae. The hypothesis presented above predicts that cell fusions will eventually be found in the latter taxon.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SUMMARY

Phylogenetic relationships inferred from SSU sequences for 37 species of coralline algae are inconsistent with any previously proposed system of classification for the coralline algae. The molecular data indicate that the Corallinales comprises four major lineages (Figure 2.1). The Sporolithaceae is resolved as the earliest divergence within the order and forms the sister group to a monophyletic Corallinaceae. The data support the monophyly of the non- geniculate subfamily Melobesioideae, which is easily delimited from other coralline taxa by the presence of multipored sporangial conceptacles. The subfamily Corallinoideae is resolved as the sister to a fourth clade comprised of species referred to the subfamilies Mastophoroideae, Metagoniolithoideae, Lithophylloideae, and Amphiroideae. Within the latter assemblage of taxa, the non-geniculate subfamily Lithophylloideae is resolved as the sister to the geniculate subfamily Amphiroideae. The inferred order of branching among these lineages, and among some genera within each of the four clades, is supported by independent paleontological evidence. Except for the presence or absence of genicula, other characters used to delimit the families and subfamilies of Corallinales were inferred to be phylogenetically, as well as taxonomically, informative. Genicula, which are characteristic of the subfamilies Amphiroideae, Corallinoideae, and Metagoniolithoideae, have arisen independently in each of the latter groups and are not homologous structures. These results indicate that the common practice of separating coralline red algae on the basis of presence or absence of genicula is an artificial operation that does not accurately reflect the evolutionary history of the group. Ultrastructural observations, as well as phylogenetic evidence presented here, show that there is no longer any basis for maintaining the subfamily

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Choreonematoideae as distinct from the Melobesioideae. Although additional studies are required, evidence also suggests that the Austrolithoideae should be subsumed in the Melobesioideae as well. These results call for a reclassification of the coralline algae.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3 EVOLUTION OF PIT PLUG-ASSOCIATED FEATURES AND A RE- EVALUATION OF PHYLOGENETIC RELATIONSHIPS AMONG THE FLORIDEOPHYCIDAEAN ORDERS OF RED ALGAE

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION

Excepting the formation of spermatia and spores, cytokinesis in species of the Florideophycidae, as well as some Bangiophycidae, is initially incomplete resulting in a septal pore with cytoplasmic continuity between daughter cells. The septal pore is subsequently filled by the deposition of a glycoproteinaceous pit plug (Pueschel & Cole 1982; Pueschel 1990). The morphology of pit plugs is variable. Depending on the species, pit plugs can be characterized by the presence or absence of osmiophilic cap layers (0,1, or 2) and a cap membrane. In those taxa possessing two cap layers the outer cap may be thin and plate like or conspicuously domed. Cytochemical data suggest that outer cap layers in the Florideophycidae, whether thin or domed, are homologous structures (Trick & Pueschel 1991). In a series of electron microscopic studies Pueschel and co-workers have elucidated the fine structure of pit plugs for representatives of all currently recognized red algal orders (Pueschel 1990; Trick & Pueschel 1991; Pueschel etal. 1992a; Pueschel 1994). With one, or possibly two, exceptions pit plug features have been found to be stable among species classified within each order (Pueschel & Cole 1982; Pueschel 1989; Schnepf 1992). Subsequent anatomical, life history, and molecular data have corroborated Pueschel and Cole's (1982) contention that at the ordinal level pit plug-associated features are taxonomically informative characters in the Rhodophyta (Coomans & Hommersand 1990; Garbary & Gabrielson 1990; Hommersand & Fredericq 1990; Ragan etal. 1994). The fact that most ultrastructural features are evolutionary conserved led to speculation that pit plug-associated features are phylogenetically, as well as taxonomically, informative. Observations of "naked" (i.e., laking both cap

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. layers and cap membranes) pit plugs in the early-diverging (Freshwater et al. 1994; Ragan etal. 1994) bangiophyte genera Compsopogon(Scott etal. 1988) and Rhodochaete (Pueschel & Magne 1987) suggested that the naked pit plug is ancestral and that cap layers and membranes are derived features within the Rhodophyta (Pueschel 1987). These data challenged several widely-held tenets of red algal evolution including the assertion that the Acrochaetiales, which possess both inner and outer cap layers and a cap membrane (presumably derived features), are the earliest-diverging lineage within the Florideophycidae. These data also suggested that the "highly specialized" Corallinales (Silva & Johansen 1986), which lack a cap membrane, may have diverged prior to the purportedly primitive orders Acrochaetiales, Nemaliales, and Palmariales. Later, the fine structure of pit plugs in species classified in the Rhodogorgonales (Pueschel etal. 1992a) and Batrachospermales (Pueschel 1994) were determined and found to be identical to those of the Corallinales. Based on the accumulated evidence, Pueschel (1994) argued that naked pit plugs are ancestral, those orders with two cap layers form a monophyletic group, and that domed outer caps and the absence of a cap membrane are plesiomorphic character-states for this clade. Cladistic analyses of non-molecular (Gabrielson & Garbary 1987; Garbary & Gabrielson 1990) and molecular (Freshwater etal. 1994; Ragan et al. 1994) data sets have thus far failed to give robust support for PueschePs hypothesis of red algal evolution based on pit plug-associated features. Both non-molecular and molecular data support the contention that naked plugs are ancestral within the group, but a clade comprised of orders characterized by pit plugs with two cap layers has not been resolved. However, the lack of independent support for PueschePs hypothesis from the molecular studies is certainly attributable, at least in part, to inadequate taxon sampling.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Of particular interest in the present study are the phylogenetic affinities of the orders Corallinales, Rhodogorgonales, and Batrachospermales. These orders play a pivotal role in hypotheses of pit plug evolution, yet the relationships of these orders to other red algal lineages have not been made clear. Previously published studies of red algal phylogeny, based on sequence analyses of the genes for the nuclear-encoded small subunit rRNA (SSU) and the plastid-encoded large subunit of Rubisco {rbcL), weakly ally the Corallinales with the Hildenbrandiales (Ragan etal. 1994; Freshwater etal. 1994). Both of these orders are represented by a single species in each of the cited studies and terminate long branches near the base of the florideophycean clade. The alliance of the Hildenbrandiales (single cap layer) and Corallinales (two cap layers) is not consistent with Pueschel's (1994) hypothesis that red algal orders possessing two cap layers are derived from a common ancestor. In their SSU- based study, Ragan etal. (1994) did, however, note that constraining the Corallinales as sister to other orders with two cap layers resulted in only a one step penalty under parsimony. Before they were classified in a separate order, the Rhodogorgonales, by Fredericq and Norris (1995), the enigmatic genera Rhodogoron and Renouxia had been linked to either the Nemaliales (Ogden 1992) or Batrachospermales (Pueschel etal. 1992a). The idea that these genera might be referable to the Batrachospermales was derived primarily from the observation that the pit plugs of Rhodogorgon carriebowensis were morphologically identical to those found in the Batrachospermales. Although similar pit plugs are also found in the Corallinales, and in spite of the fact that the Rhodogorgonales and Corallinales are the only known marine red algae that deposit calcite in their cell walls, Pueschel etal. (1992a,b) and Fredericq and Norris (1995) did not consider a specific relationship between these two orders likely. The rbcL gene tree for the red algae presented by Freshwater et

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. al. (1994, p. 7282) united Rhodogoron and Renouxia with the Gracilariales, but the authors noted that the topological position of these genera was unstable and "may be incorrect." In the rbcL tree other orders characterized by pit plugs with two cap layers (viz. Acrochaetiales, Palmariales, Nemaliales, and Batrachospermales) were resolved as a grade at the base of the Florideophycidae (Freshwater etal. 1994). Sequences for species representing the orders Batrachospermales and Rhodogorgonales were not included in the SSU-based study of Ragan etal. (1994). In the present study the phylogenetic relationships among the Corallinales, Rhodogorgonales, and Batrachospermales and other red algal orders are re-examined. To evalute these relationships, and to test Pueschel's (1994) hypothesis for the evolution of pit plug-associated features, SSU data are included for 19 species representing all six orders possessing pit plugs with an outer cap layer. The alignment also includes SSU gene sequences for 51 other red algal species representing each of the remaining 13 currently recognized orders of red algae.

MATERIALS AND METHODS

A data set was assembled consisting of nuclear-encoded small subunit ribosomal RNA (SSU) gene sequences for 70 taxonomically diverse red algal species (Table 3.1). All sequences included in the alignment are available from GenBank. The alignment includes at least one representative from each presently recognized red algal order as well as 19 species classified in orders characterized as having pit plugs with two cap layers. The SSU sequences for Galaxaura marginata (Ellis et Solander) Lamouroux, Rhodogorgon carriebowensis J. Norris et Bucher, Nothocladus nodosus Skuja, and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.1. Sources of the SSU gene sequences for 70 red algal species included in the data alignment.

Taxon Reference

Bangiales Porphyra umbilicalis (Linnaeus) J. Agardh 4 Compsopogonales Erythrocladia sp. 4 Porphyridiales Dixoniella grisea (Geitler) Scott et al. 4 Sylonema alsidii (Zanardini) Drew 4 Acrochaetiales Audouinella dasyae (Collins) Woelkerling 4 Rhodochorton purpureum (Lightfoot) Rosenvinge 7 Ahnfeltiales Ahnfeltia plicata (Hudson) Fries 2 Batrachospermales Nothocladus nodosus Skuja 11 Psilosiphon scoparium Entwisle 11 Bonnemaisoniales Bonnemaisonia hamifera Hariot 4 Ceramiales Anotrichium furcellatum (J. Agardh) Baldock 10 Ceramium macilentum J. Agardh 10 Crouania attenuata (C. Agardh) J. Agardh 10 Griffithsia monilis Harvey 10 Dasya baillouviana (S.G. Gmelin) Montagne 4 Phycodrys rubens (Linnaeus) Batters 4 Rhodomela confervoides (Hudson) P. Silva 4 Corallinales Amphiroa fragilissima (Linnaeus) Lamouroux 12 Bossiella orbigniana ssp. dichotoma (Manza) Johansen 12 Cheilosporum sagittatum (Lamouroux) J. Areschoug 12 Clathromorphum compactum (Kjellman) Foslie 12 Corallina officinalis Linnaeus 4 Lithothamnion glaciale Kjellman 12 Metagoniolithon chara (Lamarck) Ducker 12 Phymatolithon laevigatum (Foslie) Foslie 12 Gelidiales Gelidiella acerosa (Forsskal) Feldmann et Hamel 13 Gelidium pusillum (Stackhouse) Le Jolis 10 Gelidium vagum Okamura 4 Pterociadia lucida (Brown exTumer) J. Agardh 13 Ptilophora pinnatifida (J. Agardh) Norris 13 (table con'd.)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gigartinales Dasyphloea insignis Montagne Farlowia mollis (Harvey et Bailey) Farlow et Setchell Endocladia muricata (Postels et Ruprecht) J. Agardh Gloiopeltis furcata (Postels et Ruprecht) J. Agardh Furcellaria lumbricalis (Hudson) Lamouroux Chondrus crispus Stackhouse Sarcothalia crassifolia (C. Agardh) Edyvane et Womersley Callophyllis rangiferina (Turner) Womersley Mychodea camosa Hooker et Harvey Nizymenia australis Sonder Mastocarpus stellatus (Stackhouse in Withering) Guiry Phacelocarpus peperocarpos (Poiret) Wynne, Ardre et Silva Schottera nicaeensis (Lamouroux ex Duby) J. Agardh Areschougia congesta (Turner) J. Agardh Eucheuma denticulatum (N. Burman) Collins et Hervey Kappaphycus alvarezii (Doty) Doty Sphaerococcus coronopifolius Stackhouse Gracilariales doococod-Phdcocodivsiocococod

Curdiea flabellata Chapman co

Gracilaria tikvahiae McLachlan —*■

Gracilariopsis lemaneiformis (Bory) Dawson, Acleto et Foldvik co Halymeniales

Cryptonemia undulata Sonder Grateloupia filicina (Lamouroux) C. Agardh Sebdenia flabellata (J. Agardh) Parkinson Hildenbrandiales Hildenbrandia rubra (Sommerfelt) Meneghini Nemaliales Galaxaura marginata (Ellis et Solander) Lamouroux Nemalion helminthoides (Velley in Withering) Batters Palmariales Palmaria palmata (Linnaeus) Kuntze Meiodiscus spetsbergensis (Kjellman) Saunders et McLachlan "j r\D —»• ^ -£■ cococo Rhodophysema elegans (P. et H. Crouan) Dixon Rhodothamniella floridula (Dillwyn) J. Feldmann in Christensen 's i Plocamiales Plocamiocolax pulvinata Setchell Plocamium angustum (J. Agardh) J.D. Hooker et Harvey Rhodogorgonales Rhodogorgon carriebowensis J. Norris et Bucher Rhodymeniales Champia affinis (Hooker et Harvey) J. Agardh Lomentaria australis (Kutzing) Levring Lomentaria baileyana (Harvey) Farlow Erythrocolon podagricum (Harvey) J. Agardh Gloioderma fruticulosa (Harvey) De Toni

Rhodymenia linearis J. Agardh cn co oo ^ co oo —*. oiui (table con'd.)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96

1. Bird etal. (1990) 2. Bird etal. (1992a) 3. Bird etal. (1992b) 4. Ragan etal. (1994) 5. Saunders & Kraft (1994) 6. Lluisma & Ragan (1995) 7. Saunders etal. (1995) 8. Millar etal. (1995) 9. Saunders & Kraft (1996) 10. Saunders etal. (1996) 11. Saunders (pers. comm.) 12. Bailey & Chapman (1996) 13. Present study

Psilosiphon scoparium Enwistle were kindly provided by Dr. G. W. Saunders. Sequences were aligned by eye using the SeqPup program (Gilbert 1995). Regions of the data matrix which could not be unambiguously aligned were excluded from subsequent analyses. Including gaps, the SSU alignment contains 1731 positions. Phylogenetic hypotheses were inferred from the data with the optimality criterion of maximum parsimony as implemented in PAUP (v. 3.1.1., Swofford 1993). All characters were treated as unordered and weighted equally. Gaps were coded as a fifth character. Heuristic searches were performed using the tree-bisection-reconnection algorithm with random sequence addition (500 replicates), COLLAPSE and MULPARS options in effect. Support for nodes of the parsimony tree was assessed by calculating bootstrap porportions (BP) based on 100 resamplings of the data set (Felsenstein 1985). The SSU sequence for the bangiophyte Erythrocladia sp. was selected to root the trees. Pit-plug associated features characteristic of each red algal order were coded as unordered, binary-state characters as shown in Table 3.2. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3.2. Coding of pit plug-associated features for the orders of red algae (Data from Pueschel 1990,1994).

Cap Inner Outer Outer cap Order membrane cap cap morphology

Bangiales 0 1 0 ? Compsopogonales 0 0 0 ? Porphyridiales ? ? ? ? Acrochaetiales i i 1 ? Ahnfeltiales 0 0 0 ? Batrachospermales 0 1 1 2 Bonnemaisoniales 1 0 0 ? Ceramiales 1 0 0 ? Corallinales 1 1 2 Gelidiales 1 1 0 9 Gigartinales 1 0 0 ? Gracilariales 1 0 0 ? Halymeniales 1 0 0 ? Hildenbrandiales 1 1 0 ? Nemaliales 1 1 1 3 Palmariales 1 1 1 3 Plocamiales 1 0 0 ? Rhodogorgonales 0 1 1 2 Rhodymeniales 1 0 0 ?

0=absent, 1=present, 2=outer cap domed, 3=outer cap plate-like, ?=missing or not applicable.

coding scheme assumes that character states for pit-plug associated features are consistent for all species within an order, and that the pit plugs ofErythrocladia (Compsopogonales) are identical to those found in Compsopogon. Because the Acrochaetiales violate the first assumption for a single character (outer cap morphology) this feature was coded as missing (?) for the group (Table 3.2). The evolution of pit plug-associated features was determined using two different approaches. The pit-plug data were appended to the SSU sequence alignment and the combined matrix subjected to parsimony analysis, and the ultrastructural data matrix was subjected to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. parsimony analysis using the parsimony tree as a constraint (Swofford 1993). Character states at internal nodes of the parsimony tree were optimally reconstructed using the criteria of both accelerated (ACCTRAN) and delayed (DELTRAN) transformation as implemented in MacClade (v.3.0, Maddison & Maddison 1992).

RESULTS

Cladistic analysis of the SSU gene sequence alignment resulted in four equally most parsimonious trees of 3847 steps. The four topologies differed only in the relationships inferred among species within the orders Gelidiales and Gigartinales. One of the four parsimony trees was selected for presentation and is shown in Figure 3.1. With the exception of the Porphyridiales, Acrochaetiales, and Gigartinales, all other orders represented by two or more species are resolved as monophyletic. The Porphyridiales and Acrochaetiales are inferred to be paraphyletic taxa. The SSU data indicate that Schizymenia dubyi (Schizymeniaceae) is not closely related to other species classified in the order Gigartinales, rendering that order polyphyletic. These data weakly (BP<50%) ally Schizymenia dubyi as sister to the Gracilariales. The monophyly of the subclass Florideophycidae is strongly supported (BP=100%) and bootstrap confidence values indicate that most nodes along the backbone of the tree are strongly to moderately supported. The SSU sequence of Hildebrandia rubra (Hildenbrandiales) is inferred to represent the earliest divergence within the Florideophycidae. The remaining florideophycidaean orders are divided between two large sister clades. The first includes Ahnfeltia plicata (Ahnfeltiales) which is resolved as sister to a clade comprised of all

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Coralllna officinalis Bossielfa orbigniana ssp. dichotoma Cheilpsporum ?agittatum Amphiroa fragilissuna Metagoniolithon chara Clathromorphum compactum Phymafofttffon laevjgatum Litnothamnion glacials Rhodogorgon cam'ebowensis Psilosipnon scoparium Nothocladus nodosus Audoufnella dasyae Rhodochorton purpureum Nemalion hefminthofdes Gaiaxaura marginata Rhpdothamnlellafloridula Meiodiscus spitsbergansis Rhodophysema elegans Palmaria palmata Afinfeltia pllcata Ceramium macitentum Crouania attenuafa Anotrichum furcallatum Griffithsia monilis Oasya baillouviana Phycodrys rubens Rhodomeia confervoides Gracilaria tikvahiaa Gracilariopsis lemaneiformts Curdiea flabellata Schizymenia dubyi Plocamiocolax pulvinata Plocamium angustum Rhodymenia linearis Champia affinis Gloioderma fruticulosa Lomentaria australis Lomentana baileyana Erythrocolon podagricum Sebdenia flabellata Cryptonemia undulata Grateloupia filfcina Mastocarpus stellatus Chondrus crispus Sarcothaiia crassifolia Schottera nicaeensis Furcellaria lumbricalis Areschougia congesta Mychodea camosa Kappaphycus alvarezii Eucneuma denticulatum Callophyiris rangiferina Fariowia mollis Dasyphloea insignis Phacelocarpus peperocarpos Nizymenia australis Sphaerococcus coronopifolius Gloiopeltis furcata Endocladia muricala Bonnemaisonia hamifera Gelidium pusitlum Gelidium vagum Ptiiophora pinnatifida Gelidiella acerosa Pterocladia lucida Hildenbrandia rubra Oixoniella grisea Porphyra umbilicalis Styionema alsidii Erythrocladia sp.

Figure 3.1. One of four equally most parsimonious trees (C!=0.41; Rl=0.69) inferred from an analysis of the SSU gene sequences for 70 species of red algae. Bootstrap confidence values (> 50%) for each node are provided at the branches.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 orders possessing pit plugs with two cap layers. Within the two cap layer assemblage a sister group relationship between the orders Rhodogorgonales and Corallinales is strongly (BP=100%) supported. The second florideophycidaean clade includes all orders lacking cap layers but possessing a cap membrane as well as the order Gelidiales (inner cap and cap membrane present). Nine steps were required to map the pit plug character data onto the parsimony tree whether the pit plug data were analyzed in combination with the SSU data or constrained to the result of the parsimony analysis. The following conclusions are supported by both ACCTRAN and DELTRAN character optimization (Figures 3.2a,b): 1) the naked type pit plug is the plesiomorphic condition, 2) outer caps evolved only once - in the common ancestor of the orders Corallinales, Rhodogorgonales, Batrachospermales, Acrochaetiales, Nemaliales, and Palmariales, 3) an inner cap layer had also evolved prior to the divergence of the latter orders, and 4) thin outer caps are derived relative to domed outer caps in the two-cap-layer group. The transition between domed and thin outer caps must have occurred prior to the divergence of the Nemaliales and Palmariales along the evolutionary path leading from the Batrachospermales. Data presented here, as well as ultrastructural evidence, suggest that thin outer caps probably evolved within a paraphyletic Acrochaetiales. Within the two cap layer assemblage, cap membranes evolved at roughly the same time as did thin outer cap layers. In other parts of the red algal tree hypotheses for the evolution of inner cap layers and cap membranes differ depending on the character state optimization method. ACCTRAN optimization (Figure 3.2a) suggests that the inner cap evolved prior to the divergence of the Bangiales and was lost prior to the divergence of the Hildenbrandiales. Inner caps are inferred to be derived

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bangiales Ahnfeltiales Rhodymeniales Plocamiales Porphyridiales Porphyridiales Hildenbrandiales Bonnemalsonlales Halymeniales Gracllariales Compsopogonales Corallinales Rhodogorgonales Batrachospermales Gelidiales Gigartinales Acrochaetiales Nemaliales Palmariales Ceramiales IC DELTRAN B Plocamiales Bangiales Ahnfeltiales Rhodymeniales Gracllariales Porphyridiales Hildenbrandiales Bonnemalsonlales Halymeniales Porphyridiales Compsopogonales Rhodogorgonales Batrachospermales Acrochaetiales Gigartinales Corallinales Nemaliales Palmariales Gelidiales Ceramiales IC 1C K> M ACCTRAN A cap, M=cap membrane. cap, membrane. M=cap indicates Star possible alternative origin for the thin cap outer (TC). Figure 3.2. Cladogram(s) depicting therelationships orders of among inferred red algae cladisticfrom ofanalysis SSU forSenarios the evolution of pit plug-associatedof methods features and are (B) DELTRAN presented for ACCTRAN (A) character • state optimization. =character gain, O outer =character TC=thin loss, outer cap, IC=inner cap, DC=domed . gene gene sequence and data. pit plug . Character forstate pit changes plug features are indicated along the branches.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. independently in the common ancestors of the two cap layer clade and the order Gelidiales. Cap membranes are inferred to have been present in the last common ancestor of the Florideophycidae, but were lost in the ancestor of the Ahnfeltiales and the two cap layer clade. Cap membranes are inferred to have arisen independently in the common ancestor of the the orders Acrochaetiales, Nemaliales, and Palmariales. DELTRAN optimization (Figure 3.2b) suggests that inner cap layers are independently derived in the orders Bangiales, Hildenbrandiales, and Gelidiales and in the last common ancestor of the two- cap-layer clade. Cap membranes evolved three times; in the ancestors of the Hildenbrandiales and Ceramiales, and within the two cap layer lineage.

DISCUSSION

As in previous studies of ordinal-level relationships among red algae (Freshwater etal. 1994; Ragan etal. 1994), results presented here indicate that the Florideophycidae is monophyletic. Within the subclass, two large sister clades were found, one of which includes the Ahnfeltiales and all orders characterized as having pit plugs with both inner and outer cap layers. Orders comprising the latter group are divided between two lineages, one including representatives of the orders Rhodogorgonales and Corallinales and the other including species classified in the Batrachospermales, Acrochaetiales, Nemaliales, and Palmariales. Relationships inferred among the orders resolved in the two cap layer assemblage are not consistent with traditional concepts of red algal phylogeny (Kylin 1956; Dixon 1973; Magne 1989; Garbary & Gabrielson 1990), but do provide support for Pueschel's (1994) contention that pit plug-associated features are phylogenetically informative for the group. The second large florideophycidaean clade is composed of those red algal orders,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with the exception of the Gelidiales, that lack cap layers but possess a cap membrane. Although the Rhodogorgonales have been previously considered closely related to either the Batrachospermales or Nemaliales (Norris & Bucher 1989; Ogden 1992; Pueschel etal. 1992a; Fredericq & Norris 1995), the SSU data strongly (BP=100%) support a close phylogenetic link between the Rhodogorgonales and Corallinales. In terms of their growth, vegetative organization, and reproductive development the Corallinales and Rhodogorgonales are markedly different (Fredericq & Norris 1995). Nevertheless, the SSU gene sequence data indicate clearly that the two lineages are closely related. The only anatomical evidence suggesting that the two orders share a common evolutionary history includes: 1) the presence of pit plugs characterized by two cap layers, the outer domed, and the absence of a cap membrane, and 2) the fact that the Rhodogorgonales and Corallinales are the only marine red algae known to deposit calcite in the their cell walls. Other calcified marine red algae (e.g., some Nemaliales) precipitate or deposit aragonite in their cell walls or in intercelluar spaces. Although phylogenetically informative, pit plugs identical to those of the Corallinales and Rhodogorgonales are also found in the Batrachospermales (Pueschel 1994); this feature is therefore not synapomorphic for the clade. Unlike corallines, which are heavily calcified throughout, the rhodogorgonalean genera Rhodogorgon and Renouxia deposit calcite only in a single, specialized cell type termed "calciferous cells" (Norris & Bucher 1989; Pueschel etal. 1992b). Calcite deposition in both groups involves a thick carbohydrate cell wall layer (Pearse 1972; Giraud & Cabioch 1978; Pueschel et al. 1992b). Members of the exclusively freshwater order Batrachospermales also deposit calcium carbonate as calcite but the process in this group does not depend on the presence of a specially modified

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. organic cell wall matrix (Probeguin 1954). Although calcite deposition in the Corallinales and Rhodogorgonales differs slightly in other ways (Pueschel et al. 1992b), the phylogenetic evidence suggests that the process may be homologous in the latter two orders but different from that occurring in the Batrachospermales. The hypothesis that calcification, depending on the presence of a specialized organic cell wall matrix, is synapomorphic for a clade including the Rhodogorgonales and Corallinales cannot be discounted, although the proposal should be tested with other sources of data. In any case, comparative studies of corallines, and Rhodogorgon and Renouxia may ultimately help elucidate the nature of calcification processes in red algae. A number of the hypotheses erected by Pueschel (1994) are supported in the current study. The ancestral type pit plug for the Rhodophyta is the naked type found in the orders Compsopogonales and Rhodochaetales (Pueschel & Magne 1987; Scott etal. 1988; Maggs & Pueschel 1989). Naked pit plugs are also characteristic of the Ahnfeltiales, although results presented here suggest that both absence of cap layers and membranes may be secondarily derived features in this order (but see below). Orders whose species are characterized by two pit plug caps are derived from a common ancestor. The molecular data also support the hypothesis, based on cytochemical evidence, that outer cap layers are homologous structures (Trick & Pueschel 1991). Within the two cap layer assemblage plate-like outer caps are derived relative to domed outer caps, which represent the plesiomorphic character state for the clade. The most highly derived pit plug type is characteristic of the orders Acrochaetiales (in part), Nemaliales and Palmariales. The pivotal position of the Acrochaetiales in the region of the tree where the transition from the domed to plate-like morphology occurs is consistent with the presence of both types of outer caps in Acrochaete species,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. . 105 and the observation that the Acrochaetiales is a paraphyletic taxon (Garbary & Gabrielson 1990, Freshwater etal. 1994; Saunders etal. 1995; Figure 3.1). Deciding between ACCTRAN or DELTRAN optimization for the best scenario of inner cap and cap membrane evolution is less clear. Unfortunately, convincing ultrastructural or cytochemical evidence for the compositional homogeneity of inner cap layers of pit plugs in the red algae is lacking (Trick & Pueschel 1991). Perhaps the fact that none of the many cytochemical tests employed by Trick and Pueschel (1991) differentiated between the inner cap layers of red algae can be considered as evidence that they are indeed identical, at least in composition? Such speculation does not seem exceedingly implausible. Even if all inner caps are compositionally identical, the phylogenetic evidence indicates that inner cap layers are not evolutionarily homologous structures throughout the red algae. Although the inner caps of the Bangiales, Hildenbrandiales, and the two cap layer clade may be homologous (Figure 3.2a), the inner caps of the Gelidiales, under any scenario, are inferred to be independently derived. The placement of Ahnfeltia plicata (Ahnfeltiales) as sister to the two cap layer lineage is not consistent with the topological postion of this species in other published gene-sequence trees for red algae (Freshwater etal. 1994, Ragan et al. 1994). This result may be artifactual and attributed to the sampling of species included in the alignment analyzed here. In the SSU-based hypothesis for the red algae presented by Ragan etal. (1994), the Ahnfelitiales is weakly resolved as sister to the Bonnemaisoniales, Ceramiales, Gelidiales, Gigartinales, Gracilariales, Halymeniales, Plocamiales and Rhodymeniales assemblage rather than the two cap layer lineage. Parsimony analysis of the molecular data constraining the Ahnfeltiales to the result of Ragan et al. (1994) yields six equally parsimonious trees that are eight steps longer than the most

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. parsimonious solutions found in the present investigation (data not shown). These six trees are not significantly worse than the optimal parsimony tree(s) found in the current study [Templeton Test as implemented in PHYLIP algorithm DNAPARS, Felsenstein (1995)]. A strict consensus of these six trees was constructed and used as a constraint for parsimony analysis of the pit plug data. Eight steps were required to fit the pit plug features onto this tree and character state transformations were similarly resolved under both ACCTRAN and DELTRAN optimization procedures. On this tree (Figure 3.3), inner caps are homologous in the Bangiales, Hildenbrandiales, and the two-cap-layer assemblage, and are inferred to be analogous to the inner caps of the Gelidiales. The most variable feature of the pit-plug-associated characters is the presence or absence of a cap membrane, which appears to have evolved independently in three different evolutionary lineages. Additional research, to clarify relationships at the weakly resolved nodes in the molecular tree as well as to test the extent of homology of cap membranes and inner caps throughout the red algae, is necessary if the evolution of these two features is to be understood clearly.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107

Porphyridiales Compsopogonales Bangiales Porphyridiales M Hildenbrandiales Corallinales Rhodogorgonales Batrachospermales Acrochaetiales Nemaliales Palmariales Ahnfeltiales Ceramiales Bonnemaisoniales Gelidiales Gigartinales Rhodymeniales Halymeniales Gracilariales Plocamiales

Figure 3.3. Character state optimization of pit plug data onto a tree constrained to the result of Ragan et al. (1994) in which the Ahnfeltiales is sister to those orders, excepting the Gelidiales, characterized by the absence of cap layers and the presence of a cap membrane. • =character gain, O =character loss, IC=inner cap, DC=domed outer cap, TC=thin outer cap, M=cap membrane. Star indicates possible alternative origin for the thin outer cap (TC).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4 GENERAL SUMMARY AND CONCLUSIONS

108

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109 In this dissertation, phylogenetic analyses of nuclear-encoded small subunit ribosomal RNA (SSU) sequences were used to test current hypotheses of relationships within the order Corallinales and among the florideophycidaean orders of red algae. Hypotheses inferred from the molecular data were then used to explore the possible pathways of evolution for selected organismal characters in each of the studies. In Chapter 2 conflicting hypotheses for the relationships among the subfamilies and genera of the red algal order Corallinales were tested against the results of phylogenetic analyses of a molecular data set comprised of SSU gene sequences. The SSU gene trees are consistent with some, but certainly not all, aspects of previously proposed classifications for the group. The order includes four major lineages that are well-supported by the molecular data and that are clearly delimited from one another by both reproductive and anatomical features. The earliest-diverging lineage within the order includes the genera Sporolithon and Heydrichia that are classified in the family Sporolithaceae. Paleontological, morphological, and molecular data indicate clearly that these genera are, in relative terms, distantly related to other coralline algae. The Sporolithaceae is resolved as sister to a monophyletic Corallinaceae. The Melobesioideae is resolved as monophyletic and can be delimited from all other coralline taxa by the presence of multipored sporangial conceptacles. The Corallinoideae, comprising the tribes Janieae and Corallineae, is also resolved as monophyletic and probably represents the most recent divergence within the group. Strong support was found for a clade including representatives of the subfamilies Mastophoroideae, Metagoniolithoideae, Lithophylloideae, and Amphiroideae. Within the latter assemblage, the non-geniculate subfamily Lithophylloideae and the geniculate subfamily Amphiroideae are resolved as sister taxa. The hypothesis that these subfamilies share a common

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110 evolutionary history has not previously been proposed in the literature, but is consistent with the notion that some non-geniculate species are more closely related to geniculate species than they are to other non-geniculate species. Character state optimization techniques were used to explore the evolutionary pattern of distribution for seven morphological characters considered diagnostic of families and subfamilies within the order. Results indicate that all features, with the exception of the presence or absence of genicula, are phylogenetically informative for coralline algae. Genicula are inferred to be convergent (nonhomologous) structures that were independently derived in three separate coralline lineages (i.e., the Amphiroideae, Corallinoideae, and Metagoniolithoideae). This observation supports Cabioch's (1971,1972,1988) contention that within the Corallinales parallel evolutionary lines, that include geniculate and non-geniculate taxa, appear as continuums marked by the progressive evolution of an erect habit. The presence or absence of genicula as the sole basis for separating coralline algae at the subfamilial level is no longer tenable and must be abandoned if the classification of the Corallinales is to accurately reflect the evolutionary history of the group. Morphological data (Broadwater & LaPoint, in press), as well as phylogenetic evidence presented here, suggest that the natural affinities of the genera Choreonema, Austrolithon, and Boreolithon are with the subfamily Melobesioideae. An informal proposal is made to subsume the Choreonematoideae and Austrolithoideae in the latter subfamily. If this proposal is made formally and accepted, then there remains the question of transforming the present (artificial) system of classification for the Corallinales into one that is more natural. The only difficulty here, and an admittedly

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 111 subjective one, is deciding on the taxonomic ranks afforded to each of the four major clades. In Chapter 3, cladistic analysis of SSU gene sequences for 70 species of red algae formed the basis for a re-evalutation of phylogenetic relationships among the orders of the subclass Florideophycidae. These data also were used to test hypotheses for the evolution of pit plug-associated features in the Rhodophyta outlined by Pueschel (1994). With the exception of the orders Porphyridiales, Acrochaetiales, and Gigartinales, all other orders represented by two or more species are inferred to be monophyletic. The subclass Florideophycidae is resolved as a monophyletic taxon comprised of two large sister clades. One clade includes a monophyletic assemblage of orders that all possess pit plugs bearing two cap layers which is resolved as the sister to the Ahnfeltiales. The second clade includes all orders, excepting the Gelidiales, possessing pit plugs characterized by the absence of cap layers and the presence of a cap membrane. Results of analyses of the independent molecular data support Pueschel's (1994) contention that: 1) naked pit plugs (i.e., lacking both cap layers and cap membranes) are the plesiomorphic type in the Rhodophyta, 2) outer cap layers are homologous structures, 3) orders possessing pit plugs with two cap layers are derived from a common ancestor, and 4) plate-like outer cap layers and cap membranes are derived features within the two cap layer clade. Orders possessing pit plugs with two cap layers include the Acrochaetiales, Batrachospermales, Corallinales, Nemaliales, Palmariales, and Rhodogorgonales. A more complete understanding of the evolution of pit plug- associated features will not be possible until independent evidence arguing for or against the homology of inner cap layers is available. Inner cap layers in the two cap layer lineage are certainly homologous and may be homologous to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. those present in the Bangiales and Hildenbrandiales. However, the inner cap layer present in the Gelidiales is inferred to be independently derived. Thus, these data suggest that inner cap layers are not homologous structures in all red algal taxa. The second key to elucidating the pathway of evolution of these features is to determine robustly the relationships of the orders Hildenbrandiales and Ahnfeltiales to other red algal lineages. In the present study both orders are represented by only a single species. The Hildenbrandiales is inferred to represent the earliest divergence within the Florideophycidae and the Ahnfeltiales is weakly resolved as sister to a clade including all orders possessing pit plugs with two cap layers. Often in molecular systematics studies further sampling within a taxon will help resolve its relationships to others. However, this is not an option for the Hildenbrandiales and Ahnfeltiales wherein more extensive sampling is not possible at the generic level. The Ahfeltiales includes only the genus Ahnfeltia ; the Hildebrandiales is comprised of only the genera Hildenbrandia and Apophlaea. Data from other sources, and in particular nucleotide sequences for other genes, are needed to develop a highly corroborated hypothesis for the relationships of these two orders to others in the Florideophycidae. Within the two-cap-layer assemblage, strong support was found for a sister group relationship between the Corallinales and Rhodogorgonales. The observation that these orders are closely related is unexpected, considering that they share very few morphological, anatomical, or reproductive features in common (Fredericq & Norris 1995). As far as is now known, there is but a single, tenuous organismal feature that indicates that the two orders may be phylogenetically linked. Members of both orders possess the ability to deposit calcite in their cell walls and in both the process of biomineralization involves a thick, specialized organic cell wall matrix. The thalli of coralline algae are

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heavily calcified throughout; however, calcite is deposited in only a single type of cell (calciferous cells) in rhodogorgonalean species. The phylogenetic observations raise an interesting, and highly speculative, question: Could the last common ancestor of the Corallinales have resembled the rhodogorgonalean species Rhodogorgon carriebowensis or Renouxia antillana ?

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES

Abbott, I.A. and G.J. Hollenberg. 1976. Marine Algae of California. Stanford University Press, Stanford.

Adey, W.H. 1968. The distribution of crustose corallines on the Icelandic coast. Scientia Islandica Anniv. Vol. 1968:16-25. Adey, W.H. 1978a. Coral reef morphogenesis: a multidimensional model. Science 202:831-837.

Adey, W.H. 1978b. Algal ridges of the Caribbean and West Indies. Phycologia 17: 361-367.

Adey, W.H. 1973. Temperature control of reproduction and productivity in a subarctic coralline alga. Phycologia 12:111-118.

Adey, W.H. and D.L. McKibbin. 1970. Studies on the maerl species Phymatolithon calcareum (Pallas) comb. nov. and Lithothamnium corallioides Crouan in the Ria de Vigo. Bot. Mar. 8:100-106.

Adey, W. H. and C.P. Sperapani. 1971. The biology of Kvaleya epilaeve, a new parasitic genus and species of Corallinaceae. Phycologia 10:29-42. Adey, W.H. and H.W. Johansen. 1972. Morphology and taxonomy of Corallinaceae with special reference to Clathromorphum, Mesophyllum, and Neopoiyporolithon gen. nov. (Rhodophyceae, Cryptonemiales). Phycologia 11: 159-180.

Adey, W.H. and I.G. Macintyre. 1973. Crustose coralline algae: a re-evaluation in the geological sciences. Geol. Soc. Am. Bull. 84:883-904.

Adey, W.H., T. Masaki, and H. Akioka. 1974. Ezo epiyessoense, a new parasitic genus and species of Corallinaceae (Rhodophyta, Cryptonemiales). Phycologia 13: 329-344.

Adey, W.H. and J.M. Vassar. 1975. Colonization, succession and growth rates of tropical crustose coralline algae (Rhodophyta, Cryptonemiales). Phycologia 14: 55-69.

Adey, W.H., R.A. Townsend and W.T. Boykins. 1982. The crustose coralline algae (Rhodophyta: Corallinaceae) of the Hawaiian Islands. Smithson. Contrib. mar. Sci. 15: iv+74 pp.

Agardh, C.A. 1817. Synopsis Algarum Scandinaviae. Lund, Berling.

Agardh, C.A. 1820. Species Algarum. Vol. 1, part 1. Lund, Berling.

114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115

Agardh, C.A. 1824. Species Algarum. Vol. 1, part 2. Lund, Berling.

Agardh, C.A. 1828. Species Algarum. Vol. 2. Greifswald, Moritz. Agardh, J.G. 1842. Algae maris mediterranei et adriatici. Masson, Paris.

Agee, M. A. and S.T. Broadwater. 1995. Light and electron microscopic study of tetrasporogenesis in Melobesia mediocris (Corallinales, Rhodophyta). J. Phycol. 31 (suppl.): 5. Alongi, G., M. Cormaci and G. Fumari. 1996. On the occurrence of Sporolithon ptychoides Heydrich (Corallinlaes, Sporolithaceae, Rhodophyta) in the Mediterranean Sea. Cryptogam. Algol. 17:131-137.

Areschoug, J.R. 1852. Ordo XII. Corallinaceae. In Species, genera, etordines algarum (ed J. Agardh), Vol 2, Part 2, C.W. K. Gleerup, Lund. pp. 506-576.

Athanasiadis, A. 1989. North Aegean Marine Algae. III. Structure and development of the encrusting coralline Titanoderma cystoseirae (Rhodophyta, Lithophylloideae). Nord. J. Bot. 9: 435-441. Athanasiadis, A. 1995. Morphology, anatomy and reproduction of the eastern Mediterranean coralline Tenarea tortuosa and its relationship to members of the Lithophylloideae and Mastophoroideae (Rhodophyta, Corallinales). Nord. J. Bot. 15: 655-663.

Bailey, J.C. and R.L. Chapman. 1996. Evolutionary relationships among coralline red algae (Corallinales, Rhodophyta) determined by 18S rRNA gene sequence analysis, In Cytology, Genetics and Molecular Biology of Algae (eds B.R. Chaudhary and S.B. Agrawal), SPB Academic Publishing, Amsterdam, pp. 363-376. Benedetti-Cecchi, L and F. Cinelli. 1994. Recovery of patches in an assemblage of geniculate coralline algae: variability at different successional stages. Mar. Ecol. Prog. Ser. 110:9-18.

Berthold, G. 1882. Die Bangiaceen des Golfes von Neapel und der angrenzenden Meeresabschnitte. Fauna Flora Golf. Neapel 8:1-28.

Bhattacharya, D., H.J. Elwood, L.J. Goff and M.L. Sogin. 1990. Phylogenyof Gracilaria lemaneiformis (Rhodophyta) based on sequence analysis of its small subunit ribosomal RNA coding region. J. Phycol. 26:181-186. Bhoday, R., S.J. Russell, J.H. Price, C.l. Scholfield and P. Gacesa. 1994. A molecular phytogeny of the British species of Callithamnion based on restriction site data. Biochem. Soc. Trans. 22:360S.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116 Bird, C.J., E.L. Rice, C.A. Murphy, Q.Y. Liu and M.A. Ragan. 1990. Nucleotide sequences of the 18S ribosomal RNA genes from the red algae Gracilaria tikvahiae McLachlan, Gracilaria verrucosa (Hudson) Papenfuss and Gracilariopsis sp. Nucl. Acids Res. 18: 4023-4024.

Bird, C.J., C.A. Murphy, E.L. Rice and M.A. Ragan. 1992a. The 18S rRNA gene sequences of four commercially important seaweeds. J. Appl. Phycol. 4:379- 384. Bird, C.J., E.L. Rice, C.A. Murphy, and M.A. Ragan. 1992b. Phylogenetic relationships in the Gracilariales (Rhodophyta) as determined by 18S rDNA sequences. Phycologia 31: 510-522.

Bjork, M., S.M. Mohammed, M. Bjorklund and A. Semesi. 1995. Coralline algae, important coral-reef builders threatened by pollution. AmbiolA: 502-505.

Bold, H.C. and M.J. Wynne. 1978. Introduction to the Algae. Prentice-Hall, Englewood Cliffs. Bornet, E. and G. Thuret. 1866. Sur la fecondation des Floridees. Compt. Rend. Acad. Sci. [Paris] 63: 444-447.

Bornet, E. and G. Thuret. 1867. Recherches sur la fecondation des Floridees. Ann. Sci. Nat. Bot. 7:137-166.

Borowitzka, M.A. 1982a. Mechanisms in algal calcification. Prog. Phycol. Res. 1:137-177.

Borowitzka, M.A. 1982b. Morphological and cytological aspects of algal calcification. Int. Rev. Cytol. 74:127-162. Borowitzka, M.A., A.W.D. Larkum and C.E. Nockolds. 1974. A scanning electron microscope study of the structure and organization of the calcium carbonate deposits of algae. Phycologia 13:195-203.

Borowitzka, M.A. and M. Vesk. 1978. Ultrastructure of the Corallinaceae. I. The vegetative cells of Corallina officinalis and C. cuvierii. Mar. BiolAQ: 295-304.

Bossellini, A. and R.N. Ginsberg. 1971. Form and internal structure of recent algal nodules (rhodolites) from Bermuda. J. Geol. 79:669-682. Bouget, F.-Y., C. Kerbourc'h, M.-F. Liaud, S. Loiseaux de Goer, R.S. Quatrano, R. Cerff, and B. Kloareg. Structural features and phytogeny of the actin gene of Chondrus crispus (Gigartinales, Rhodophyta). Curr. Genet. 28:164-172.

Braga, J.C. and J. Aguirre. 1995. Taxonomy of fossil coralline algal species: Neogene Lithophylloideae (Rhodophyta, Corallinaceae) from Southern Spain. Rev. Palaeobot. Palynol. 86:265-285.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Broadwater, S.T. and J. Scott. 1982. Ultrastructure of early development in the female reproductive system of Polysiphonia harveyi Bailey (Cermiales, Rhodophta). J. Phycol. 18:427-441. Broadwater, S.T. and E.A. LaPoint. Parasitic interactions and vegetative ultrastructure of Choreonema thuretii (Corallinales, Rhodophyta). J. Phycol. in press.

Broadwater, S., J. Scott, D. Field, B. Saunders and J. Thomas. 1993. An ultrastructural study of cell division in the coralline red alga Bossiella orbigniana. Can. J. Bot. 71: 434-446.

Broadwater, S.T., J.L. Scott, S.P.A. Goss and B.D. Saunders. 1995. Ultrastructure of vegetative organization and cell division in Glaucosphaera vacuolata Korshikov (Porphyridiales, Rhodophyta). Phycologia 34: 351-361. Brown, D.L and T.E. Weier. 1970. Ultrastructure of the freshwater alga Batrachospermum. I. thin section and freeze-etch analysis of juvenile and photosynthetic filament vegetative cells. Phycologia 9:217-235.

Cabioch, J. 1970. Sur I'importance de ph§nomenes cytologiques pour la systematique et la phylogenie des Corallinacees (Rhodophycees, Cryptonemiales). C.R. Acad. Sci. (Paris) ser. 271D: 296-299.

Cabioch, J. 1971. Essai d'une nouvelle classification des Corallinacees actuelles. C.R. hebd. Seanc. Acad Sci. Paris 272D: 1616-1619. Cabioch, J. 1972. Etude sur les Corallinacees. II. La morophogenese; consequences systematiques et phylogenetiques. Cah. Biol. mar. 13: 137-288.

Cabioch, J. 1988. Morphogenesis and generic concepts in coralline algae - a reappraisal. Helgolander Meeresunters. 42:493-509.

Campbell, S.E. 1980. Palaeoconchecelis starmachii, acorbonate boring microfossil from the Upper Silurian of Poland (425 million years old): implications for the evolution of the Bangiaceae (Rhodophyta). Phycologia 19: 25-36. Campbell, S.J. and Wm. J. Woelkerling. 1990. Are Titanoderma and Lithophyllum (Corallinales, Rhodophyta) distinct genera? Phycologia 29:114- 125.

Cavalier-Smith, T. 1993. Kingdom Protozoa and its 18 phyla. Microbiol. Rev. 57: 953-994.

Chamberlain, Y.M. 1987. Conceptacle production and life history in four species of Pneophyllum (Rhodophyta, Corallinaeceae) from the British Isles. Br. phycol. J. 22: 43-48.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chamberlain, Y.M. 1991. Observations on Phymatolithon lamii (Lemoine) Y. Chamberlain comb. nov. (Rhodophyta, Corallinales) in the British Isles with aa assessment of its relationship to P. rugulosum, Lithophyllum lamii and L melobesioides. Br. phycol. J. 26: 219-233.

Chamberlain, Y.M. and D.W. Keats. 1994. Three melobesioid crustose coralline red algae from South Africa: Leptophytum acervatum (Foslie) comb, nov., L foveatum sp. nov. and L. ferox (Foslie) comb. nov. Phycologia 33:111- 133.

Chauvin, J.F. 1842. Recherches sur I'Organisation, la Fructification et la Classification de Plusilurs Genres d'Algues. Caen. Chihara, M. 1973. The significance of reproductive and spore germination characteristics in the systematics of the Corallinaceae: articulated coralline algae. Jap. J. Bot. 20: 369-379.

Cole, K.M. and E. Conway. 1975. Phenetic implications of structural features of the perennating phase in the life hisory of Porphyra and Bangia (Bangiophyceae, Rhodophyta). Phycologia 14:239-245.

Cole, K.M. and R.G. Sheath. 1990. Biology of the Red Algae. Cambridge University Press, Cambridge.

Colthart, B.J. and H.W. Johansen. 1973. Growth rates of Corallina officinalis (Rhodophyta) at different temperatures. Mar. Biol. 18:46-49. Coomans, R.J. and M.H. Hommersand. 1990. Vegetative growth and organization, In Biology of the Red Algae (eds K.M. Cole and R.G. Sheath), Cambridge University Press, New York. pp. 275-304.

Coppejans, E. 1978. Sur les propagules de Fosliella farinosa (Lamouroux) Howe var. farinosa (Rhodophyceae, Cryptonemiales). Bull. Soc. R. Bot. Belg. 111:55-61. Coppejans, E. 1983. Iconographie d'Algues Mediterraneenes. J. Cramer, Vaduz.

Davis, E. and J. Scott. 1986. Ultrastructure of cell division in the marine red alga Lomentaria baileyana. Protoplasma 131:1-10.

Decaisne, J. 1842a. Essais sur une classification des Algues et des Polypiers calciferes de Lamouroux. Ann. Sci. nat., Bot., ser. 217: , 297-380.

Decaisne, J. 1842b. Memoire sur les Corallines ou Polypiers calciferes. Ann. Sci. nat., Bot., ser. 2 ,18: 96-128.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Delivopoulos, S.G. and P. Kugrens. 1985. Ultrastructure of carposporophyte development in the red alga Gloiosiphonia verticillaris (Cryptonemiales, Gloiosiphoniaceae) Am. J. Bot. 72:1926-1938.

Demoulin, V. 1975. The origin of ascomycetes and basidiomycetes - the case for a red algal ancestry. Bot Rev. 40: 315-345.

Demoulin, V. 1985. The red algal-higher fungi phylogenetic link: the last ten years. BioSystems 18:347-356.

Denison, W.C. and G.C. Carroll. 1966. The primitive ascomycete: a new look at an old problem. Mycologia 58: 249-269.

Dixon. P.S. 1973. Biology of the Rhodophyta. Hafner Pregs, Macmillan, New York.

Dixon, P.S. and L.M. Irvine. 1977. Seaweeds of the British Isles, Vol. 1 Rhodophyta, Part 11ntroduction, Nemaliales, Gigartinales. British Museum (Natural History), London.

Ducker, S. 1979. The genus Metagoniolithon Weber-van Bosse (Corallinaceae, Rhodophyta). Aust. J. Bot. 27: 67-101.

Duckett, J.G. and M.C. Peel. 1978. The role of tranmission electron microscopy in elucidating the taxonomy and phytogeny of the Rhodophyta, In Modem Approaches to the Taxonomy of Red and Brown Algae, (eds. D.E.G. Irvine and J.H. Price), Academic Press, London, pp. 157-204.

Dutcher, J.A. and D.F. Kapraun. 1994. Random amplified polymorphic DNA (RAPD) identification of generic variation in three species of Porphyra (Bangiales, Rhodophyta). J. Appl. Phycol. 6:267-273.

Duwel, L. and S. Wegeberg. 1996. The typification and status of Leptophytum (Corallinales, Rhodophyta). Phycologia 35: 470-483. Ellis, J. 1755. An Essay Towards a Natural History of the Corallines. Privately published, London.

Ellis, J. 1767. Extract of a letter from John Ellis, Esquire, F.R.S. to Dr. Linnaeus of Upsal, F.R.S. on the animal nature of the genus of zoophytes, called Corallina. Phil Trans. R. Soc. Lond. 57: 404-427.

Ellis, J. and D. Solander. 1786. The Natural History of Many Curious and Uncommon Zoophytes. B. White and Son, London. Fan, K. -C. 1961. Studies on Hypneocolax, with discussion on the origin of parasitic red algae. Nova Hedwigia 3:119-128.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120 Farris, J.S. 1989. The retention index and rescaled consistency index. Cladistics 5: 417-419.

Feldmann, J. 1955. Un nouveau genere de Protofloridee: Colacodictyon, nov. gen. Bull Soc. BotF t. 102: 23-28.

Feldmann, J. 1967. Sur une Bangiophycees endozoique ( Neeva repens Batters) et ses affinites (Rhodophyta). Blumea 15:25-29. Felsenstein, J. 1985. Confidence limits on phytogenies: an approach using the bootstrap. Evolution 39: 783-791.

Felsenstein, J. 1988. Phytogenies from molecular sequences: inference and reliability. Ann. Rev. Genet. 22:521-565. Felsenstein, J. 1995. PHYLIP - Phytogeny Inference Package, v. 3.57c. Distributed by the author, Department of Genetics, University of Washington, Seattle.

Foslie, M. 1903. Den botaniske samling. Kongel. Norske Vidensk. Selsk. Aarsb. for 1902, pp. 23-25.

Foslie, M. 1904. Lithothamnionaeae, Melobesiaeae, Mastophoreae. In The Corallinaceae of the Siboga-Expedition, Siboga Expeditie (eds. A. Weber van- Bosse and M. Foslie), E.J. Brill, Leyden, pp. 10-77.

Fredericq, S., M.H. Hommersand and J.N. Norris. 1989. Morphological observations of the adelphoparasite Gracilariophila oryzoides (Gracilariales, Rhodophyta). Jap. J. Phycol. 37:167-179.

Fredericq, S. and J.N. Norris. 1995. A new order (Rhodogorgonales) and family (Rhodogorgonaceae) of red algae composed of two tropical calciferous genera, Renouxia gen nov. and Rhodogorgon. Crypt. Bot. 5:316-331. Fredericq, S., J.N. Norris, E.A. Zimmer, D.W. Freshwater, and M.H. Hommersand. 1996. Proposal of the Dumontiales ord. nov. and reinstatement of the Sphaerococcales Sjoestedt emend, based on family complexes previously placed in the marine red algal order Gigatinales. J. Phycol. 32(suppl.): 16.

Freshwater, D.W. and J. Rueness. 1994. Phylogenetic relationships of some European Gelidium (Gelidiales, Rhodophta) species, based on rbdL nucleotide sequence analysis. Phycologia 33:187-194.

Freshwater, D.W., S. Fredericq, B.S. Butler, M.H. Hommersand and M.W. Chase. 1994. A gene phytogeny of the red algae (Rhodophyta) based on plastid rbcL. Proc. Nat. Acad. Sci. USA 91: 7281-7285.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 121 Freshwater, D.W., S. Fredericq and M.H. Hommersand. 1995. A molecular phytogeny of the Gelidiales (Rhodophyta) based on analysis of plastid rbcL nucleotide sequences. J. Phycol. 31:616-632.

Fritsch, F.E. 1945. The Structure and Reproduction o f the Algae, Vol. 2. Cambridge University Press, Cambridge. Gabrielson, P.W., D.J. Garbary and R.F. Scagel. 1985. The nature of the ancestral red alga: inferences from a cladistic analysis. BioSystems 18:335- 346.

Gabrielson, P.W. and D.J. Garbary. 1987. A cladistic analysis of Rhodophyta: Florideophycidean orders. Br. phycol. J. 22:125-138.

Gao, K., Y. Aruga, K. Asada, T. Ishihara, T. Akano and M. Kiyohara. 1993. Calcification in the articulated coralline alga Corallina pilulifera, with special reference to the effect of elevated CO 2 concentration. Mar. Biol. 117:129-132.

Garbary, D. 1978. On the phylogenetic relationships of the Acrochaetiaceae (Rhodophyta). Br. phycol. J. 13: 247-254.

Garbary, D.J., G.l. Hansen and R.F. Scagel. 1980. A revised classification of the Bangiophyceae (Rhodophyta). Nova Hedwigia 33:145-166.

Garbary, D.J. and H.W. Johansen. 1982. Scanning electron microscopy of Corallina and Haliptilon (Corallinaceae, Rhodophyta): surface features and their taxonomic implications. J. Phycol. 18:211-219.

Garbary, D. J. and H.W. Johansen. 1987. Morphogenesis and evolution in the Amphiroideae (Rhodophyta, Corallinaceae). Br. phycol. J. 22:1-10.

Garbary, D.J. and P.W. Gabrielson. 1990. Taxonomy and evolution, In Biology of the Red Algae (eds K.M. Cole and R.G. Sheath), Cambridge University Press, New York. pp. 477-498.

Genetics Computer Group. 1994. Program Manual for the GCG Package, UNIX Version 8.0. 575 Science Drive, Madison, Wisconsin, 53711. Gilbert, D.G. 1995. SeqPup, a biological sequence editor and analysis program for Macintosh computers. Published electronically on the internet, available via anonymous ftp to ftp.bio.indiana.edu. Ginsburg, R.N. and J.H. Schroeder. 1973. Growth and submarine fossilization of algal cup reefs, Bermuda. Sedimentology 20:575-614.

Giraud, G. and J. Cabioch. 1977. Characters generaux de Pultrastructure des corallinees. Rev. Algol. 12:45-60.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Giraud, G. and J. Cabioch. 1978. Ultrastructure and elaboration of calcified cell- walls in the coralline algae (Rhodophyta, Cryptonemiales). Biol. Cellulaire 36:81-85. Giraud, G. and J. Cabioch. 1983. Inclusions cytoplasmiques remarquables chez les Corallinacees. Ann. Sci. nat., Bot. 5:29-43.

Gittins, B.T. and P.S. Dixon. 1976. Biology of Lithothrix aspergillum. Br. phycol. J. 11:194-199.

Goff, L.J. 1982. The biology of the parasitic algae, In Progress in Phycological Research, Vol. I, (eds F. Round and D. Chapman), Elsevier Biomedical Press, Amsterdam, pp. 289-370.

Goff, L.J. and M.H. Hommersand. 1982. Reproductive development of the parasitic red alga Gardneriella tuberifera (Solieriaceae, Gigartinales). J. Phycol. 18: 295-304.

Goff, L.J. and A.W. Coleman. 1988. The use of plastid DNA restriction endonuclease patterns in delineating red algal species and populations. J. Phycol. 24: 357-368.

Goff, L.J. and A.W. Coleman. DNA: Microspectrofluorometirc studies, In Biology of the Red Algae, (eds K.M. Cole and R.G. Sheath), Cambridge University Press, Cambridge, pp. 43-72. Goff, L.J. and G. Zuccarello. 1994. The evolution of parasitism in red algae: cellular interactions of adelphoparasites and their hosts. J. Phycol. 30: 695-720.

Goff, L.J., D.A. Moon, P. Nyvall, B. Stache, K. Mangin and G. Zuccarello. 1996. The evolution of parasitism in the red algae: molecular comparisons of adelphoparasites and their hosts. J. Phycol. 32:297-312.

Gray, S.F. 1821. A Natural Arrangement of British Plants. Vol. I. Baldwin, Cradock, and Joy, London.

Green, J.C., B.S.C. Leadbeater, and W.L. Diver, eds. 1989. The Chromophyte Algae: Problems and Perspectives. Systematics Association Special Volume Number 38, Clarendon Press, Oxford.

Greville, R.K. 1830. Algae britannicae. Maclachlan and Stewart, Edinburgh.

Griffon, B. and S.T. Broadwater. 1996. Ultrastructural study of sporogenesis in several genera within the Melobesioideae (Corallinales, Rhodophyta). J. Phycol. 32(suppl.): 19.

Guiry, M.D. 1974. A preliminary consideration of the taxonomic position of Palmaria palmata (Linnaeus) Stackhous e=Rhodymenia palmata (Linnaeus) Greville. J. Mar. Biol. Ass. U.K. 54: 509-528.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123

Guiry, M.D. 1978. The importance of sporangia in the classification of the Florideophycidae, In Modem Approaches to the Taxonomy of Red and Brown Algae, (eds. D.E.G. Irvine and J.H. Price), Academic, London, pp. 111-144.

Guiry, M.D. 1987. The evolution of life history types in the Rhodophyta: an appraisal. Cryptogam. Algol. 8:1-12. Hackney, J.M., R.C. Carpenter and W.H. Adey. 1989. Characteristic adaptations to grazing among algal turfs on a Caribbean coral reef. Phycologia 28:109-119.

Hamby, R.K., L. Sims, L. Issel and E.A. Zimmer. 1988. Direct robosomal RNA sequencing: optimization of extraction and sequencing methods for work with higher plants. Plant Mol. Biol. Rep. 6:175-192.

Hamel, G. and M. Lemoine. 1953. Corallinacees de France et d'Afrique du Nord. Archs Mus. Nat. hist. nat. Paris. Ser. 7 . 1:15-136.

Harvey, A.S. and Wm. J. Woelkerling. 1995. An account of Austrolithon intumescens gen. et sp. nov. and Boreolithon van-heurckii (Heydrich) gen. et comb. nov. (Austrolithoideae subfam. nov., Corallinaceae, Rhodophyta). Phycologia 34:362-382.

Harvey, W.H. 1836. Algae. In Flora Hibemica, Pt. 3, (ed. J.T. MacKay), William Curry and Co., Dublin, pp. 157-254.

Harvey, W.H. 1849. A Manual of the British Algae, 2nd ed. J. van Voorst, London.

Hasegawa, M, T. Hashimoto, J. Adachi, N. Iwabe and T. Miyata. 1993. early branching in the evolution of eucaryotes: ancient divergence of Entamoeba that lacks mitochondria revealed by protein sequence data. J. Mol. Evol. 36: 380- 388.

Heath, I.B. 1980a. Fungal mitoses, the significance of variations on a theme. Mycologia 72:229-250.

Heath, I.B. 1980b. Variant mitoses in lower eucaryotes: indicators of the evolution of mitosis? Int. Rev. Cytol. 64:1-80.

Heath, I.B. 1981. Nucleus-associated organelles in fungi. Int. Rev. Cytol. 69: 191-221.

Hendriks, L., R. De Baere, Y. Van de Peer, J. Neefs, A. Goris, and R. De Wachter. 1991. The evolutionary position of the rhodophyte Porphyra umbilicalis and the basidiomycete Leucosporidium scotii among other eucaryotes as deduced from complete sequences of small ribosomal subunit RNA. J. Mol. Evol. 32: 167-177.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 124

Hollenberg, GJ. 1970. Phycological notes IV, Including new marine algae and new records for California. Phycologia 9:61-72.

Hommersand, M.H. 1963. The morphology and classification of some Ceramiaceae and Rhodomelaceae. Univ. Calif. Publ. Bot. 35:165-366.

Hommersand, M.H. and S. Fredericq. 1990. Sexual reproduction and cystocarp development, In Biology o f the Red Algae, (eds K.M. Cole and R.G. Sheath), Cambridge University Press, Cambridge, pp. 305-345. Hommersand, M.H., S. Fredericq, and D.W. Freshwater. 1994. Phylogenetic systematics and biogeography of the Gigartinaceae (Gigartinales, Rhodophyta) based on seqeunce analysis of rbcL. Bot. Mar. 37:193-203.

Johansen, H.W. 1969a. Morphology and systematics of coralline algae with special reference to Calliarthron. Univ. Calif. Pubis. Bot. 49:1-98.

Johansen, H.W. 1969b. Patterns of genicular development in Amphiroa (Corallinaceae). J. Phycol. 5:118-123.

Johansen, H.W. 1976. Current status of generic concepts in coralline algae (Rhodophyta). Phycologia 15:221-244.

Johansen, H.W. 1977. The articulated Corallinaceae (Rhodophyta) of South Africa: 1. Cheilosporum (Decaisne) Zanardini. Jl. S. Afr. Bot. 43:163-185. Johansen, H.W. 1981. Coralline Algae, A First Synthesis. CRC Press, Boca Raton, FL.

Johansen, H.W. and L.F. Austin. 1970. Growth rates in the articulated coralline Calliarthron (Rhodophyta). Can. J. Bot. 48:125-132. Johansen, H.W. and P.C. Silva. 1978. Janieae and Lithotricheae: two new tribes of articulated Corallinaceae (Rhodophyta). Phycologia 17:413-417.

Johansen, H.W. and H.B.S. Womersley. 1994. Jania (Corallinales, Rhodophyta) in southern Australia. Aust. Syst. Bot. 7: 605-625.

Johnson, C., D. Klumpp, J. Field and R. Bradbury. 1995. Carbon flux on coral reefs: effects of large shifts in community structure. Mar. Ecol. Prog. Ser. 126: 123-143.

Johnson, J.H. 1956. Ancestry of the coralline algae. J. Paleont. 30: 563-567. Johnson, J.H. 1961. Limestone-building Algae and Algal Limestones. Colorado School of Mines, Golden, Colorado.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jussieu, B. 1745. Examen de quelques productions marines qui ontete mises au nombre des Plantes, et qui sont Pouvrage de'une sorte d'lnsects de mer. Mem. Acad. Roy. Sci. 1742:290-302.

Kapraun, D.F. 1980. An Illustrated Guide to the Benthic Marine ALgae of North Carolina I. Rhodophyta. University of North Carolina Press, Chapel Hill.

Karnas, K.J., J. Scott and S.T. Broadwater. 1995. Ultrastructural survey of sporogenesis in the genus Bossiella (Corallinales, Rhodophyta). J. Phycol. 31 (suppl.): 7.

Keats, D.W., A. Groener and Y.M. Chamberlain. 1993. Cell sloughing in the littoral zone coralline alga, Spongites yendoi (Foslie) Chamberlain (Corallinales, Rhodophyta). Phycologia 32:143-150. Keats, D.W., P. Wilton and G. Maneveldt. 1994. Ecological significance of deep-layer sloughing in the eulittoral zone coralline alga, Spongites yendoi (Foslie) Chamberlain (Corallinaceae, Rhodophyta) in South Africa. J. Exp. Mar. Biol. Ecol. 175: 145-154. Keats, D.W. and Y.M. Chamberlain. 1995. Heydrichia groeneri sp. nov.: an new species of crustose coralline alga (Rhodophyta, Sporolithaceae) from South Africa and Namibia. Phycologia 34:51-57.

Keats, D.W., R.S. Steneck, R.A. Townsend and M.A. Borowitzka. 1996. Lithothamnion prolifer Foslie: a common non-geniculate coralline alga (Rhodophyta: Corallinaceae) from the tropical and subtropical Indo-Pacific. Bot. Mar. 39: 187-200.

Klepacki, K.J., J.L. Scott, and S.T. Broadwater. 1995. Ultrastructure of cell division in Agardhiella subulata (Gigartinales, Rhodophyta). Eur. J. Phycol. 30: 159-167.

Kluge, A.G. and J.S. Farris. 1969. Quantitative phyletics and the evolution of anurans. Syst. Zool. 18:1-32.

Knoll, A.H. 1992. The early evolution of eucaryotes: a geological perspective. Science 256: 622-627.

Kohlmeyer, J. 1975. New clues to the possible origin of ascomycetes. BioScience 25: 86-93. Kraft, G.T. 1981. Rhodophyta: morphology and classification, In The Biology of Seaweeds, (eds. C.S. Lobban and M.J. Wynne), Blackwell Scientific Publications, Oxford, pp. 6-51.

Kraft, G.T. and Wm. J. Woelkerling. 1981. Rhodophyta: systematics and biology. In Marine Botany, an Australasian Perspective, (eds M. Clayton and R. King), Longman Cheshire, Melbourne.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 126

Kraft, G.T. and J. Robins. 1985. Is the order Cryptonemiales defensible? Phycologia 24: 67-77.

Kwok, S., T.J. White and J.W. Taylor. 1986. Evolutionary relationships between fungi, red algae, and other simple eucrayotes inferred from total DNA hybridizations to a cloned basidiomycete ribosomal DNA. Exp. Mycol. 10:196- 204. Kylin, H. 1916. Clber die Befruchtung und Redukionsteilung bei Nemalion multifidum. Ber. Deutsch Bot. Ges. 34:257-271.

Kylin, H. 1923. Studien uber die Entwicklungsgeschichte der Florideen. K. Svensk. Vet.-Akad. Hand!. 63:1-139.

Kylin, H. 1928. Entwicklungsgeschichtliche Florideenstudien. Lunds Univ. Arsskr.n.f. Avd. 24:1-127.

Kylin, H. 1930. Uber die Entwicklungsgeschichte der Florideen. Lunds Univ. Arsskr.n.f. Avd. 26:1-104.

Kylin, H. 1932. Die florideenordnung Gigartinales. Lunds Univ.Arsskr. n.f. Avd. 28: 1-88.

Kylin, H. 1937. Ubereine marine Porphyridium. Art. K. Fysiogr. Sallskap. Lund. Forhandl. 7:1-5.

Kyln, H. 1956. Die Gattungen der Rhodophyceen. C.W.K. Gleerups, Lund, xv + 673 pp.

Ladd, H.S., J.l. Tracey, J.W. Wells and K.O. Emery. 1950. Organic growth and sedimentation on an atoll. J. Geol. 58:410-425. Lamouroux, J.V.F. 1813. Essai sur les generes de la famille des thalassiophytes non articulees. Ann. Mus. Nat. Hist. [Paris], 20, 21-47,115-139, 267-293.

Lamouroux, J.V.F. 1812. Extrait d'un memoire sur la classification des polypiers coralligenes, non entierement pierreux. Nouv. Bull. Sci. Soc. Philom. Paris 3: 181-188. Lamouroux, J.V.F. 1816. Historie naturelle des animaux sans vertebres. Deterville, Paris.

La Velle, J.M. 1979. Translocation in Calliarthron tuberculosum and its role in light-enhancement of calcification. Mar. Biol. 55: 37-44.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lebednik, P.A. 1977. Postfertilization development in Clathromorphum, Melobesia and Mesophyllum with comments on the evolution of the Corallinaceae and the Cryptonemiales (Rhodophyta). Phycologia 16:379-406.

Lebednik, P.A. 1978. Development of male conceptacles in Mesophyllum Lemoine and other genera of the Corallinaceae (Rhodophyta). Phycologia 17: 388-395.

Lee, R.E. 1974. Chloroplast structure and starch grain production as phylogenetic indicators in the lower Rhodophyceae. Br. phycol. J. 9:291-295.

Lee, R.K.S. 1967. Taxonomy and distribution of the melobesioid algae on Rongelap Atoll, Marshall Islands. Can. J. Bot. 45:985-1001.

Lee, S.B. and J.W. Taylor. 1990. Isolation of DNA from fungal mycelia and single spores, In PCR Protocols, A Guide to Methods and Applications (eds. M.A. Innis, D.H. Gelfland, J.J. Sninsky and T.J. White). Academic Press, Inc. San Diego, pp. 282-287. Liaud, M.-F., C. Valentin, W. Martin, F.-Y. Bouget, B. Kloareg and R. Cerff. 1994. The evolutionary origin of red algae as deduced from the nuclear genes encoding cytosolic and chloroplast glyceraldehyde-3-phosphate dehydrogenases from Chondrus crispus. J. Mol. Evol. 38: 319-327.

Lindstrom, S.C. 1993. Inter- and intapopulation genetic variation in species of Porphyra (Rhodophyta: Bangiales) from British Columbia and adjacent waters. J. Appl. Phycol. 5: 53-62. Lindstrom, S.C. and K.M. Cole. 1992a. Relationships between some North Atlantic and North Pacific species of Porphyra (Bangiales, Rhodophyta): evidence from isozymes, morphology, and chromosomes. Can. J. Bot. 70: 1355-1363.

Lindstrom, S.C. and K.M. Cole. 1992b. The Porphyra lanceolata-P. pseudolanceolata (Bangiales, Rhodophyta) complex unmasked: recognition of new species based on isozymes, morphology, chromosomes and distributions. Phycologia 31: 431-448. Linnaeus, C. 1753. Species Plantarum. Vol 2. pp. 561-1200. Stockholm, Salvius.

Linnaeus, C. 1758. Systems Naturae, 10th edn. L. Salvii, Holmaiae. Lluisma, A.O. and M.A. Ragan. 1995. Relationships among Eucheuma denticulatum, Eucheuma isiforme and Kappaphycus alvarezii (Gigartinales, Rhodophyta) based on nuclear SSU-rRNA gene sequences. J. Appl. Phycol. 7: 471-477.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Littler, M.M. 1972. The crustose Corallinaceae. Oceanog. Mar. Biol. 10: 311- 325. Littler, M.M. 1973. The distribution, abundance, and communities of deepwater Hawaiian crustose Corallinaceae (Rhodophyta, Cryptonemiales). Pacific Sci. 27: 281-289.

Littler, M.M. and M.S. Doty. 1974. Ecological components structuring the seaward edges of tropical Pacific reefs: the distribution, communities and productivity of Porolithon. J. Ecol. 117-129.

Littler, M.M. and D.S. Littler. 1985. Deepest known plant life discovered on an uncharted seamount. Science 227: 57-59. Lobban, C.S. and M.J. Wynne. 1981. The Biology of Seaweeds. Blackwell Scientific Publications, Oxford.

Lockhart, P.J., M.A. Steel, M.D. Hendy and D. Penny. 1994. Recovering evolutionary trees under a more realistic model of sequence evolution. Mol. Biol. Evol. 1: 605-612.

Luekart, P. 1994. Field and laboratory studies on depth dependence, seasonality and light requirement of growth in three species of crustose coralline algae (Corallinales, Rhodophyta). Phycologia 33: 281-290. Luttke, A., T.L. Maier and E.A. Schenk. 1994. Codon usage adaptation in the ferredoxin-NADP+ oxidoreductase of Cyanophora paradoxa upon translocation from cyanoplast to nucleus. Gene 146:123-127.

Maddison, W.P. and D.R. Maddison. 1992. MacClade, version 3.0. Sinauer, Sunderland, Massachusetts.

Maggs, C.A., S.E. Douglas, J. Fenety and C.J. Bird. 1992. A molecular and morphological analysis of the Gymnogongrus devoniensis (Rhodophyta) complex in the North Atlantic. J. Phycol. 28:214-232.

Magne, F. 1989. Classification et phylogenie des Rhodophycees. Cryptogam. Algol. 10: 101-115.

Marsh, J. 1970. Primary productivity of reef-building calcareous red algae. Ecology 51: 255-263. Maslov, V.P. 1962. Fossil red algae of the USSR and their relation to facies. TrudyGeol. Inst, Lenningr. 53:1-222.

Mason, L.R. 1953. The crustaceous coralline algae of the Pacific Coast of the United States, Canada and Alaska. Univ. Calif. Pubs. Bot. 26: 313-390.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Masuda, M., K. Kogame and M.D. Guiry. 1996. Life history of Gymnogongrus griffithsiae (Phyllophoraceae, Gigartinales) from Ireland: implications for life history interpretation in the Rhodophyta. Phycologia 35:421-434.

Matsuda, S. 1989. Succession and growth rates of encrusting crustose coralline algae (Rhodophyta, Cryptonemiales) in the upper fore-reef environment off Ishigaki Island, Ryukyu Islands. Coral Reefs 7:185-195.

Mattox, K.R. and K.D. Stewart. 1984. Classification of the green algae: a concept based on comparative cytology, In Systematics of the Green Algae, (eds D.E.G. Irvine and D.M. John), Academic Press, London, pp. 29-72. May, D. I. and Wm. J. Woelkerling. 1988. Studies on the genus Synarthrophyton (Corallinaceae, Rhodophyta) and its type species S. patena (J.D. Hooker et W.H. Harvey) Townsend. Phycologia 27: 50-71.

Mays, K., J. Scott and C. Pueschel. 1996. Ultrastructural patterns of sporogenesis in geniculate coralline red algae. J. Phycol. 32(suppl.): 31.

McDonald, K. 1972. The ultrastructure of mitosis in the marine red alga Membranoptera platyphylla. J. Phycol. 8:156-166. Meneses, I.C. 1993. Vertical distribution of coralline algae in the rocky intertidal of northern Chile. Hydrobiologia 260/261:121-129. Millar, A.J.K., G.W. Saunders, I.M. Strachan and G.T. Kraft. 1996. The morphology, reproduction and small-subunit rRNA gene sequence of Cephalocystis (Rhodymeniaceae, Rhodophyta), a new genus based on Corylecladia furcellata J. Agardh from Australia. Phycologia 35: 48-60.

Millson, C. and B.L. Moss. 1985. Ultrastructure of the vegetative thallus of Phymatolithon lenormandii (Aresch., in J. Ag.) Adey. Bot. Mar. 28:123-132.

Moberly, R. 1970. Microprobe study of diagenesis in calcareous algae. Sedimentology 14:113-123.

Moestrup, 0.1982. Flagellar structure in algae: a review with new observations particularly on the Chrysophyceae, Phaeophyceae (Fucophyceae), Euglenophyceae, and Reckertia. Phycologia 21: 427-528. Nageli, C. 1847. Die neuem Algensysteme und Versuch zur Begrundung eines eigenen Systems der Algen und Florideen. Neue Denkschr. Allgemein. Schweiz. Ges. Gesammt. Naturwiss. 9:1-275. Naim, 0 . 1993. Seasonal responses of a fringing reef community to eutrophication (Reunion Island, Western Indian Ocean). Mar. Ecol. Prog. Ser. 99: 137-151.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Nikoh, N., N. Hayase, N. Iwabe, K. Kuma and T. Miyata. 1994. Phylogenetic relationship of the kingdoms, Animalia, Plantae, and Fungi inferred from 23 different protein sequences. Mol. Biol. Evol. 11: 762-768.

Norris, J.N. and K.E. Bucher. 1989. Rhodogorgon, an anomalous new red algal genus from the Caribbean Sea. Proc. Biol. Soc. Wash. 102:1050-1066.

Nybakken, J.W. 1988. Marine Biology, An Ecological Approach. 2nd edn. Harper and Row Publishers, New York. Ogden, N.B. 1992. Morphology, reproduction, and range extension of Rhodogorgon (?Nemaliales, Rhodophyta). Phycologia 31:470-477.

Olsen, G.J., N. Larsen and C.R. Woese. 1991. The ribosomal database project. Nucl. Acids. Res. 19: 2017-2021.

Olsen, G.J., H. Matsuda, R. Hagstrom and R. Overbeek. 1994. fastDNAml: a tool for construction of phylogenetic trees of DNA sequences using maximum likelihood. Comput. Biol. Sci. 10:41-48.

Oltmanns, F. 1898. Zur Entwicklungsgeschichte der Florideen. Bot. Zeit. 56: 99- 140.

Oltmanns, F. 1904. Morphologie und Biologie derAlgen. Band 1, G. Fischer, Jena.

Paine, R.T. 1984. Ecological determinism in the competition for space. Ecology 65: 1339-1348.

Pakker, H., H. Klerk, J.H. van Campen, J.L. Olsen and A.M. Breeman. 1996. Evolutionary and ecological differentiation in the pantropical to warm temperate seaweed Digenea simplex ( Rhodophyta). J. Phycol. 32:250-257. Pappenfuss, G.F. 1955. Classification of the Algae, In A Century of Progress in the Natural Sciences, 1853-1953 (ed E.L. Kessel), California Academy of Sciences, San Francisco, pp. 115-224. Parsons, T.J., C.A. Maggs and S.E. Douglas. 1990. Plastid DNA restriction analysis links the heteromorphic phases of an apomictic red algal life history. J. Phycol. 26: 495-500.

Patwary, M.U., R.M. MacKay, and J.P. van der Meer. 1993. Revealing genetic markers in Gelidium vagum (Rhodophyta) through the random amplified polymorphic DNA (RAPD) technique. J. Phycol. 29: 216-222. Patway, M.U. and J.P. van der Meer. 1994. Application of RAPD markers in an examination of heterosis in Gelidium vagum (Rhodophyta). J. Phycol. 30:91- 97.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pearse, V.B. 1972. Radioisotopic study of calcification in the articulated coralline alga Bossiella orbigniana. J. Phycol. 8: 88-97.

Penrose, D. 1991. Spongites fruticulosus (Corallinaceae, Rhodophyta), the type species of Spongites, in southern Australia. Phycologia 30:438-448.

Penrose, D. 1992. Neogoniolithon fosliei (Corallinaceae, Rhodophyta), the type species of Neogoniolithon, in southern Australia. Phycologia 31: 338-350. Penrose, D. and Wm. J. Woelkerling. 1988. A taxonomic reassessment of Hydrolithon Foslie, Porolithon Foslie and Pseudolithophyllum Lemoine emend. Adey (Corallinaceae, Rhodophyta) and their relationships top Spongites Kutzing. Phycologia 27:159-176. Penrose, D. and Wm. J. Woelkerling. 1992. A reappraisal of Hydrolithon and its relationship to Spongites (Corallinaceae, Rhodophyta). Phycologia 31:81-88.

Penrose, D. and Y.M. Chamberlain. 1993. Hydrolithon farinosum (Lamouroux) comb, nov.: implications for generic concepts in the Mastophoroideae (Corallinaceae, Rhodophyta). Phycologia 32:295-303.

Peyrot-Clausade, M. T. le Campion-Alsumard, P. Hutchings, J. le Campion, C. Payri, and M.-F. Fontaine. 1995. Initial bioerosion and bioaccretion on experimental substrates in high island and atoll lagoons (French Polynesia). Oceanologica Acta 18:531-341.

Philippi, R.A. 1837. Beweis, dass die Nulliporen Pflanzen sind. Arch. Naturgesch., Jahrg. 3, 1: 387-393.

Phillips, D. and J. Scott. 1981. Ultrastructure of cell division and reproductive differentiation of male plants in the Florideophyceae (Rhodophyta). Mitosis in Dasya baillouviana. Protoplasma 106:329-341. Pobeguin, T. 1954. Contribution a I'etude des carbonates de calcium precipitation du calcaire par les vegetaux comparaison avec le monde animal. Ann. Sci. Nat. Bot. 11:29-109.

Pueschel, C.M. 1987. Absence of cap membranes as a charcteristic of pit plugs of some red algal orders. J. Phycol. 23:150-156. Pueschel, C.M. 1989. An expanded survey of the ultrastructure of red algal pit plugs. J. Phycol. 25: 625-636. Pueschel, C.M. 1990. Cell Structure, In Biology of the Red Algae (eds K.M. Cole & R.G. Sheath), Cambridge University Press, New York. pp. 7-41.

Pueschel, C.M. 1994. Systematic significance of the absence of pit-plug cap membranes in the Batrachospermales (Rhodophyta). J. Phycol. 30: 310-315.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 132 Pueschel, C.M. 1996. Formation and mobilization of protein bodies in ephemeral epithallial cells of Clathromorphum loculosum (Corallinales, Rhodophyta). Eur. J. Phycol. 31: 249-254.

Pueschel, C.M. and K.M. Cole. 1982. Rhodophycean pit plugs: an ultrastructural survey with taxonomic implications. Am. J. Bot. 69: 703-720. Pueschel, C.M. and F. Magne. 1987. Pit plugs and other ultrastructural features of systematic value in Rhodochaete parvula (Rhodophyta, Rhodochaetales). Crypt. Algol. 8: 201-209.

Pueschel, C.M., H.N. Trick, and J.N. Norris. 1992a. Fine structure of the phylogenetically important alga Rhodogorgon carriebowensis (Rhodophyta, Batrachospermales?). Protoplasma 166:78-88. Pueschel, C.M., H.H. Eichelberger, and N.H. Trick. 1992b. Specialized calciferous cells in the marine alga Rhodogorgon carriebowensis and their implications for models of red algal calcification. Protoplasma 166:89-98,

Pueschel, C.M. and T.J. Miller. 1996. Reconsidering prey specializations in an algal-limpet grazing mutualism: epithallial c ell development in Clathromorphum circumscriptum (Rhodophyta, Corallinales). J. Phycol. 32:28-36.

Pueschel, C.M., T.J. Miller and B.B. McCausland. 1996. Development of epithallial cells in Corallina officinalis and Lithophyllum impressum (Corallinales, Rhodophyta). Phycologia 35:161-169. Ragan, M.A., C.J. Bird, E.L Rice, R.R. Gutell, C.A. Murphy, and R.K. Singh. 1994. A molecular phytogeny of the marine red algae (Rhodophyta) based on the nuclear small-subunit rRNA gene. Proc. Natl. Acad. Sci. USA 91: 7276- 7280.

Ragan, M.A. and R.R. Gutell. 1995. Are red algae plants? Bot. J. Linn. Soc. 118:81-105.

Ray, 1724. Synopsis methodica Stiripium Britannicarum. 3rd ed., The Ray Society, London. Rice, E.L and C.J. Bird. 1990. Relationships among geographically distnat populations of Gracilaria verrucosa (Gracilariales, Rhodophyta) and related species. Phycologia 29: 501-510.

Ridley, M. 1993. Evolution. Blackwell Scientific Publications, Oxford.

Rosenvinge, L.K. 1917. The marine algae of Denmark. Part II. Rhodophyceae II. (Cryptonemiales). Kongel. Danske Vidensk. Selsk. Skr., 7Raekke. Naturvidensk. ogMathem. Afd. 7:155-284.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ruprecht, F.J. 1851. Tange des Ochoskischen Meeres. In Sibirische Reise , (ed A.T. von Middendorff), K. Akad. Wiss., Lenningrad. pp. 191-435. Ruprecht, F.J. 1855. Ober das System der Rhodophyceae. Mem. Acad. Imp. Sci. Saint-Pdtersbourg, ser. 6, Sci. Nat Bot. 7:25-54.

Sambrook, J., E.F. Fritsch and T. Maniatis. 1989. Molecular Cloning, A Laboratory Manual. Vols. 1-3,2nd edn. Cold Spring Harbor Laboratory Press, New York.

Sartoretto, S. 1994. Structure et dynamique d'un nouveau type de bioconstruction a Mesophyllum lichenoides (Ellis) Lemoine (Corallinales, Rhodophyta). C.R. Acad. Sci. Paris 317:156-160. Saunders, G.W. and G.T. Kraft. 1994. Small-subunit sequences from representatives of selected families of the Gigartinales and Rhodymeniales (Rhodophyta): 1. Evidence for the Plocamiales ord. nov. Can. J. Bot. 72:1250- 1263.

Saunders, G.W. and G.T. Kraft. 1996. Small-subunit rRNA gene sequences from representatives of selected families of the Gigartinales and Rhodymeniales (Rhodophyta). 2. Recognition of the Halymeniales ord. nov. Can. J. Bot. 74: 694-707.

Saunders, G.W., C.J. Bird, M.A. Ragan and E.L Rice. 1995. Phylogenetic relationships of species of uncertain taxonomic position within the Acrochaetiales-Palmariales complex (Rhodophyta): inferences from phenotypic and 18S rDNA sequence data. J. Phycol. 31:601-611. Schmitz, F. 1883. Untersuchungen uber die Befruchtung der Florideen. Sitzungsber. K. Pruess. Akad. Wiss. Berlin 1883:215-258.

Schmitz, F. 1889. Systematische Ubersicht der bisher bekannten Gattungen der Florideen. Flora 72: 435-456.

Schmitz, F. 1892. [6. Klasse Rhodophyceae] 2. Unterklasse Florideae. In Syllabus...Grosse Ausgabe, (ed A. Engler), G. Borntraeger, Berlin, pp. 16-23. Schmitz, F. and P. Hauptfleisch. 1897. Corallinaceae. In Die naturlichen Pflanzenfamilien, (eds A. Engler and K. Prantl), 1. Teil, 2. Abt., W. Engelmann, Leipzig, pp. 537-544.

Schnepf, E. 1992. Electron microscopical studies of Thorea ramosissima (Thoreaceae, Rhodophyta): taxonomic implications of Thorea pit plug ultrastructure. PI. Syst. Evol. 181:233-244. Scholfield, C.I., P. Gacesa, J.H. Price, S.J. Russell and R. Bhoday. 1991. Restriction fragment length polymorphisms of enzymatically-amplified small-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 134 subunit rRNA-coding regions from Gracilaria and Gracilariopsis (Rhodophyta)-a rapid method for assessing 'species' limits. J. Appl. Phycol. 3: 329-334. Schornstein, K.L. and J. Scott. 1982. Ultrastructure of cell division in the unicellular red alga Porphyridium purpureum. Can. J. Bot. 60: 85-97.

Schweigger, A.F. 1819. Beobachtungen auf naturhistorischen Reisen, Anatomisch-Physiologische Untersuchungen uber Corallen. G. Reimer, Berlin.

Schweigger, A.F. 1826. Observations on the anatomy of Corallina opuntia, and some other species of corallines. Edinburgh New Philos. J. 1:220-224. Scott, J. 1983. Mitosis in the freshwater red algal Batrachospermum ectocarpum. Protoplasma 118:56-70. Scott, J., C. Bosco, K. Schornstein and J. Thomas. 1980. Ultrastructure of cell division and reproductive differentiation of male plants in the Florideophycidae (Rhodophyta): Cell division in Polysiphonia. J. Phycol. 16: 507-524.

Scott, J.L., D. Phillips and J. Thomas. 1981. Polar rings are persistent organelles in interphase vegetative cells of Polysiphonia harveyi Bailey (Rhodophyta, Ceram iales). Phycologia 20: 333-337. Scott, J., J. Thomas and B. Saunders. 1988. Primary pit connections in Compsopogon coeruleus (Balbis) Montagne (Compsopogonales, Rhodophyta). Phycologia 27: 327-333.

Scott, J.L. and S.T. Broadwater. 1990. Cell division, In Biology of the Red Algae (eds K.M. Cole & R.G. Sheath), Cambridge University Press, New York. pp. 123-145.

Searles, R.B. 1980. The strategy of the red algal life history. Am. Nat. 115:113- 120 . Setchell, W.A. 1943. Mastophora and Mastophoreae: genus and subfamily of Corallinaceae. Proc. Natl. Acad. Sci. USA 29:127-135.

Shreeve, J. 1996. Are algae - not coral - reefs' master builders? Science 271: 597-598.

Siesser, W.G. 1972. Relict algal nodules (rhodolites) from the South African continental shelf. J. Geol. 80: 611-616.

Silva, P.C. and H.W. Johansen. 1986. A reappraisal of the order Corallinales (Rhodophyceae). Br. phycol. J. 21: 245-254. Skuja, H. 1939. Versuch einer systematischen Einteilung der Bangioideen oder Protoflorideen. Acta Horti Bot. Univ. Latv. 11/12:23-40.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Somers, J.A., M.l. Tait, W.F. Long and F.B. Williamson. 1990. Activities of Corallina (Corallinales) and other Rhodophyta polymers in the modulation of calcification. Hydrobiologia 204/205: 491-497.

Sorokin, Y.l. 1993. Coral Reef Ecology. Ecological Studies Vol. 102. Springer- Verlag, Berlin.

Stackhouse, J. 1801. Nereis Britannica. Bathoniae and Londini. Stackhouse, J. 1816. Nereis Britannica, 2nd ed. Oxford, Collingwood. Steane, D.A., B.A. McClure, A.E. Clarke and G.T. Kraft. 1991. Amplification of the polymorphic 5.8S rRNA gene from selected Australian gigartinalean species (Rhodophyta) by polymerase chain reaction. J. Phycol. 27:758-762.

Steneck, R.S. 1986. The ecology of coralline algal crusts: convergent patterns and adaptive strategies. Ann. Rev. Ecol. Syst. 17:273-303.

Steneck, R.S. and W.H. Adey. 1976. The role of environment in control of morphology in Lithophyllum congestum, a Caribbean algal ridge builder. Bot. Mar. 19:197-215. Steneck, R.S. and R.T. Paine. 1986. Ecological and taxonomic studies of shallow-water encrusting Corallinaceae (Rhodophyta) of the boreal northeastern Pacific. Phycologia 25: 221-240. Stiller, J.W. and J.R. Waaland. 1993. Molecular analysis reveals cryptic diversity in Porphyra (Rhodophyta). J. Phycol. 29: 506-517.

Stiller, J.W. and J.R. Waaland. 1996. Porphyra rediva sp. nov. (Rhodophyta): a new species from northeast Pacific salt marshes. J. Phycol. 32: 323-332. Stoddart, D.R. 1969. Ecology and morphology of recent coral reefs. Biol. Rev. 44: 433-498.

Suneson, S. 1937. Studien uber die Entwicklungsgeschichte der Corallinaceen. Lunds. Univ.Arsskr. n.f. Avd. 2:33(2). 101 pp.

Suneson, S. 1943. The structure, life history and taxonomy of the Swedish Corallinaceae. Lunds. Univ.Arsskr. n.f. Avd. 2: 39(9). 65 pp.

Suneson, S. 1945. On the anatomy, cytology and reproduction of Mastophora. with a remark on the nuclear conditions in the spermatangia of the Corallinaceae. K. Sallsk. Lund. Forh. 15: 251-264.

Suneson, S. 1950. The cytology of bispore formation in two species of Lithophyllum and the significance of the bispores in Corallinaceae. Botaniska NotiserA: 429-450.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Svedelius, N. 1915. Zytologisch-entwicklungsgeschichtliche studien uber Scinaia furcellata ein beitrag zur frage der reduktionsteilung der nicht tetrasporenbildenden Florideen. Nova Act Reg. Soc. Sci. Upsal., ser 4, 4:1-55.

Swofford, D.L. 1993. PAUP: Phylogenetic Analysis Using Parsimony, Version 3.1.1. Computer Program distributed by the Illinois Natural History Survey, Champaign, Illinois. Taylor, W.R. 1960. Marine Algae of the Eastern Tropical and Subtropical Coasts o f the Americas. University of Michigan Press, Ann Arbor.

Tournefort, J.P. Institutiones reiherbariae, Vol. 1-3,3rd edn. Typographia Regia, Paris.

Townsend, R.A. 1981. Tetrasporangial conceptacle development as a taxonomic character in the Mastophoroideae and Lithophylloideae (Rhodophyta). Phycologia 20: 407-414. Townsend, R. 1982. Sporolithon, an important genus of Corallinaceae (Rhodophyta). Scientific Programme and Abstracts, First International Phycological Congress, St. John's Newfoundland, Aug. 8-14, p. a50.

Townsend, R.A., Y.M. Chamberlain and D.W. Keats. 1994. Heydrichia woelkerlingii gen. et sp. nov., a newly discovered non-geniculate red alga (Corallinales, Rhodophyta) from Cape Province, South Africa. Phycologia 33: 177-186.

Townsend, R.A., Wm. J. Woelkerling, A.S. Harvey, and M. Borowitzka. 1995. An account of the red algal genus Sporolithon (Sporolithaceae, Corallinales) in southern Australia. Aust. Syst. Bot. 8: 85-121. Trick, H.N. and C.M. Pueschel. 1991. Cytochemical evidence for homology of the outer cap layer of red algal pit plugs. Phycologia 30:196-204. Turner, J.A. and Wm. J. Woelkerling. 1982. Studies on the Mastophora - Lithoporella complex (Corallinaceae, Rhodophyta). II. Reproduction and generic concepts. Phycologia 21:218-235.

Tyrrell, B. and H.W. Johansen. 1995. Reproductive and regenerative strategies of Lithothrix aspergillum (Corallinales, Rhodophyta) in southern California. Phycologia 34: 39-44.

van Oppen, M.J.H., J.L. Olsen and W.T. Stam. 1995a. Genetic variation withi and among North Atlantic and Baltic populations of the benthic alga Phycodrys rubens (Rhodophyta). Eur. J. Phycol. 30:251-260. van Oppen, M.J.H., S.G.A. Draisma, J.L Olsen and W.T. Stam. 1995b. Multiple trans-Arctic passages in the red alga Phycodrys rubens: evidence from nuclear rDNA ITS sequences. Mar. Biol. 123:179-188.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 137

Verheij, E. 1992. Structure and reproduction of Sporolithon episoredion (Adey, Townsend et Boykins) comb. nov. (Corallinales, Rhodophyta) from the Spermonde Archipelago, Indonesia. Phycologia 31:500-509. Verheij, E. 1993. The genus Sporolithon (Sporolithaceae fam. nov., Corallinales, Rhodophyta) from the Spermonde Archipelago, Indonesia. Phycologia 32:184-196.

Vesk, M. and M.A. Borowitzka. 1984. Ultrastructure of tetrasporogenesis in the coralline alga Haliptilon cuvieri (Rhodophyta). J. Phycol. 20:501-515.

Wainright, P.O., G. Hinkle, M.L. Sogin and S.K. Stickel. 1993. Monophyletic origins of the Metazoa: an evolutionary link with fungi. Science 260:340-342. West, J.A. and M.H. Hommersand. 1981. Rhodophyta: Life Histories, In The Biology of Seaweeds (eds. C.S. Lobban and M.J. Wynne). Blackwell Scientific Publications, Oxford, pp. 133-193.

Wetherbee, R. 1980. Postfertilization development in the red alga Polysiphonia. I. proliferation of the carposporophyte. J. Ultrastr. Res. 70:259-274.

Wilks, K.M. and Wm. J. Woelkerling. 1994. An account of southern Australian species of Phymatolithon (Corallinaceae, Rhodophyta) with comments on Leptophytum. Aust. Syst. Bot. 7:183-223. Wilks, K.M. and Wm. J. Woelkerling. 1995. An accound of southern Australian species of Lithothamnion (Corallinaceae, Rhodophyta). Aust. Syst. Bot 8: 549- 583.

Wille, N. 1894. Uber die Befruchtung bei Nemalion multifidum (Web. et Mohr) J. Ag. Ber. Deutsch Bot. Ges. 12 [Suppl.]: 57-60. Woelkerling, Wm. J. 1980. Studies on Metamastophora (Corallinaceae, Rhodophyta). II. Systematics and distribution. Br. phycol. J. 15:227-245.

Woelkerling, Wm. J. 1983. A taxonomic reassessment of Lithothamnium (Corallinaceae, Rhodophyta) based on studies of R.A. Philippi's original collections. Br. phycol. J. 18:165-197. Woelkerling, Wm. J. 1983. A taxonomic reassessment of Lithophyllum (Corallinaceae, Rhodophyta) based on studies of R. A. Philippi’s original collections. Br. phycol. J. 18: 299-328.

Woelkerling, Wm. J. 1985. A taxonomic reassassment of Spongites (Corallinacae, Rhodophyta) based on studies of Kutzing's original collections. Br. phycol. J. 20:123-153.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 138 Woelkerling, Wm. J. 1986. The genus Litholepis (Corallinaceae, Rhodophyta): taxonomic status and disposition. Phycologia 25:253-261. Woelkerling, Wm. J. 1990. An introduction, In Biology of the Red Algae (eds K.M. Cole & R.G. Sheath), Cambridge University Press, New York. pp. 1-6.

Woelkerling, Wm.J. and L.M. Irvine. 1986. The typification and status of Phymatolithon (Corallinaceae, Rhodophyta). Br. phycol. J. 21: 55-80.

Woelkerling, Wm. J. 1987. The genus Choreonema in southern Australia and its subfamilial classification within the Corallinaceae (Rhodophyta). Phycologia 26:111-127. Woelkerling, Wm. J. 1988. The Coralline Red Algae, An Analysis of Genera and Subfamilies of Nongeniculate Corallinaceae. Oxford University Press, Oxford and British Museum (Natural History), London.

Woelkerling, Wm. J. and S.C. Ducker. 1987. Lesueuria minderiana gen. et. sp. nov. (Corallinaceae, Rhodophyta) from southern and western Australia. 26: 192-204.

Woelkerling, Wm. J. and M.S. Foster. 1989. A systematic and ecographic account of Synarthrophyton schielianum sp. nov. (Corallinaceae, Rhodophyta) from the Chatham Islands. Phycologia 28:39-60. Woelkerling, Wm. J. 1991. The status and disposition of Perispermon (Corallinaceae, Rhodophyta). Phycologia 30:135-144. Woelkerling, Wm. J. and S.J. Cambell. 1992. An account of southern Australian species of Lithophyllum (Corallinaceae, Rhodophyta). Bull. Br. Mus. nat. Hist. (Bot.) 22:1-107.

Woelkerling, Wm. J. and A. Harvey. 1992. Mesophyllum incisum (Corallinaceae, Rhodophyta) in southern Australia: implications for generic and specific delimitation in the Melobesioideae. Br. phycol. J. 27: 381-399. Woelkerling, Wm. J. and A. Harvey. 1993. An account of southern Australian species of Mesophyllum (Corallinaceae, Rhodophyta). Aust. Syst. Bot. 6571- : 637.

Woelkerling, Wm. J., L.M. Irvine, and A.S. Harvey. 1993. Growth-froms in non geniculate coralline red algae (Corallinales, Rhodophyta). Aust. Syst. Bot. 6: 277-293.

Wray, J.L 1977. Calcareous Algae. Elsevier, Amsterdam.

Yamanouchi, S. 1906a. The life history of Polysiphonia violacea. Bot. Gaz. 41: 425-433.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 139

Yamnouchi, S. 1906b. The life history of Polysiphonia violacea. Bot. Gaz. 42: 401-449. Yendo, K. 1902. Enumeration of corallinaceous algae hitherto known from Japan. Bot Mag. (Tokyo) 16:185-196.

Zaneveld, J.S. 1958. A Lithothamnion bank at Bonaire (Netherlands Antilles). Blumea 4: 206-219.

Zhou, Y.-H. and M.A. Ragan. 1996. Nuclear-encoded protein-coding genes of the agarophyte Gracilaria verrucosa (Gracilariales, Rhodophyta). Hydrobiologia 326/327: 429-436.

Zinova, A.D. 1967. The Red, Brown, and Green Algal Flora of the Southern Seas of the USSR. Academy of Sciences USSR, Moscow.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A: PERMISSION FOR REPRODUCTION OF DRAWINGS USED IN FIGURES 1.3 AND 1.4.

140

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 141 Ms. Nettie Debill 25 October 1996 Stanford University Press

Dear Ms. Debill,

I am writing for permission from the Stanford University Press to reproduce and include in my doctoral dissertation, "Molecular Systematics of Red Algae," several drawings from the following book:

Abbott, I. A. and GJ. Hollenberg, 1976. Marine Algae of California. Stanford University Press, Stanford. [ISBN 0-8047-0867-3] The drawings are as follows:

Figure Drawing Pag? 324 Lithothamnium crassiusculum 385 334 Mesophyllum lamellatum 390 337 Lithophyllum grumosum 394 350 Lithothrix aspergillum 402 357 Corallina vancouveriensis 406 360 Serraticardia macmillanii 408

These drawings are among the best for these types of organisms and their reproduction and presentation in my dissertation are important for the sake of comparison and enhance the overall quality of the manuscript considerably. The letter of permission will be included in the dissertation.

Thank you for your time and assistance. Your atttention to this matter is greatly appreciated.

Sincerely,

J. Craig Bailey

Department of Plant Biology Louisiana State University 502 Life Sciences Building Baton Rouge, LA 70803-1705

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 142

11/4/96.

DEAR PROFESSOR BAILEY:

YOU.HAVE OUR GRATIS! EEBMISSION TO USE'THE SIX DRAWINGS YOU REQUESTED. PERMISSION COVERS USE IN. YOUR DOCTORAL DISSERTATION ONLY;.-IF YOUR WORX IS EVER PUBLISHED IN ITS PRESENT'FORM OR IN A REVISED FORM; ’YOU MUST AGAIN REQUEST PERMISSION TO USE THE DRAWINGS.

YOURS SINCERELY,

NETTIE*DEBILL PERMISSIONS MANAGER STANFORD UNIVERSITY PRESS'

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX B: COLLECTION INFORMATION FOR CORALLINE SPECIES.

143

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Species Collection site

Arthrocardia filicula South Africa Bossiella califomica ssp. schmitii Stillwater Cove, Carmel, Monterey Co., CA, USA Bossiella califomica ssp. dichtoma Laguna Beach, Orange Co., CA, USA Calliarthron cheilosporioides Laguna Beach, Orange Co., CA, USA Calliarthron tuberculosum Vancouver Is., British Columbia, Canada Cheilosporum sagittatum Pt. Lonsdale, Victoria, Australia Corallina elongata Swakopmund, Namibia Corallina officinalis Peggy's Cove, Nova Scotia, Canada 'Corallina sp.?' Botany Beach, British Columbia, Canada Haliptilon roseum Pt. Lonsdale, Victoria, Australia Jania crassa Palm Beach, Natal, South Africa Jania rubens Galway, Ireland Serraticardia macmillanii MacKerracher St. Park, Mendocino Co., CA, USA Amphiroa sp. (AUS) Pt. Lonsdale, Victoria, Australia Amphiroa sp. (SAF) Pt. Alfred, South Africa Amphiroa fragilissima Key Largo, FL, USA Lithothrix aspergillum Laguna Beach, Orange Co., CA, USA Metagoniolithon chara Pt. Lonsdale, Victoria, Australia Metagoniolithon radiatum Beachport, South Australia Metagoniolithon stelliferum Portsea, Victoria, Australia Lithophyllum kotschyanum Fiji Spongites yendoi Holbaaipunt, South Africa Clathromorphum compactum Newfoundland, Canada Clathromorphum parcum Horseshoe Cove, Sonoma Co., CA, USA Lithothamnion glaciate Labrador, Canada Lithothamnion tophiforme Labrador, Canada Leptophyum acervatum South Africa Leptophytum ferox South Africa 'Leptophytum foveatum?' South Africa Mastophoropsis caniculata Beachport, South Australia Mesophyllum englehartii South Africa Mesophyllum erubescens Sao Sebastiao, Brazil Phymatolithon laevigatum Kimmeridge, Dorset, UK Phymatolithon lenormandii Kimmeridge, Dorset, UK Synarthrophyton patena Beachport, South Australia Heydrichia woelkerlingii Betty's Bay, South Africa Sporolithon durum Port MacDonnell, South Australia

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX C: MORPHOLOGICAL CHARACTERS AND CHARACTER STATES FOR THE FAMILIES AND SUBFAMILIES OF CORALLINE ALGAE.

145

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 146

1. Tetra-/bisporangial plugs: [1] present or[0] absent. 2. Sporangial conceptacles: [1] present or [0] absent. 3. Sporangial complexes (sori): [1] present or [0] absent. 4. Tetrasporangial cleavage pattern: [1] crucitate or [0] zonate. 5. Sporangial conceptacles: [0] uniporate, [1] multiporate, or [?] missing. 6. Cells of adjacent filaments connected by: [0] cell fusions, [1] secondary pit connections, or [2] both cell fusions and secondary pit connections. 7. Genicula: [1] present or [0] absent.

Taxon Character 1 2 3 4 5 6 7

Sporolithaceae 1 0 1 1 ? 2 0 Melobesioideae 1 1 0 0 1 0 0 Mastophoroideae 0 1 0 0 0 0 0 Lithophylloideae 0 1 0 0 0 1 0 Metagoniolithoideae 0 1 0 0 0 0 1 Corallinoideae 0 1 0 0 0 0 1 Amphiroideae 0 1 0 0 0 1 1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA

Jeffrey Craig Bailey was bom in Bluefieid, West Virginia on May 5,1968 the son of Phyllis K. and Henry H. Bailey, Jr. Craig Bailey completed a Bachelor of Science degree in Marine Biology, cum laude, with Honors in Biology, at the University of North Carolina at Wilmington in 1990. He was involved in several separate research projects as an undergraduate at UNCW, including studies focusing on cultivation, ecotypic variation, and the cytogenetics of selected marine green and red algal species. Craig Bailey earned a Master of Arts degree in Biology at the College of William and Mary in 1992 with a thesis describing nuclear DNA content variation among seventeen species of coralline red algae. Bailey began work on the Ph.D. at Louisiana State University in 1992. His research interests currently include the phytogeny, molecular, and cell biology of the algae. He is also interested in the biodiversity and ecology of seaweeds in coral reef ecosystems. Bailey's studies at LSU have earned him several honors and awards. They include an NSF-EPSCoR Graduate Fellowship in Molecular Evolution (1993-94), an NSF Dissertation Improvement Grant (1995-97), National Sigma Xi and Phycological Society of America Grants-in-Aid of Research Awards (1995), LSU and PSA Hoshaw Travel Awards (1993-96), and the William J. Luke LSU Botany Teaching Assistant Award (1995). His studies of algal biology and systematics have been published in internationally recognized journals and books.

147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidate: Jeffrey Craig Bailey

Major Field: B o tan y

Title of Dissertation: Molecular Systematics of Red Algae

Approved:

EXAMINING COMMITTEE:

Date of Kxamination:

10 / 21/96

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.