Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mechanical and chemical convergence of joints in three lineages of articulated coralline algae Janot, Kyra G. 2018

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

Item Metadata

Download

Media
24-ubc_2018_september_janot_kyra.pdf.pdf [ 2.86MB ]
Metadata
JSON: 24-1.0370962.json
JSON-LD: 24-1.0370962-ld.json
RDF/XML (Pretty): 24-1.0370962-rdf.xml
RDF/JSON: 24-1.0370962-rdf.json
Turtle: 24-1.0370962-turtle.txt
N-Triples: 24-1.0370962-rdf-ntriples.txt
Original Record: 24-1.0370962-source.json
Full Text
24-1.0370962-fulltext.txt
Citation
24-1.0370962.ris

Full Text

    MECHANICAL AND CHEMICAL CONVERGENCE OF JOINTS IN THREE LINEAGES OF ARTICULATED CORALLINE ALGAE  by Kyra G. Janot  B.Sc., The University of British Columbia, 2011  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (BOTANY)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2018  © Kyra G. Janot, 2018         ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:  Mechanical and chemical convergence of joints in three lineages of articulated coralline algae   Submitted by  Kyra G. Janot     in partial fulfillment of the requirements for  the degree of  Doctor of Philosophy          in  Botany              Examining Committee: Patrick Martone, Botany           Supervisor  Geoff Wasteneys, Botany           Supervisory Committee Member  Shawn Mansfield, Wood Sciences          Supervisory Committee Member  Brian Leander, Botany and Zoology          Supervisory Committee Member iii  Abstract Macroalgae living in wave-swept environments experience a high degree of mechanical stress. Flexible algae can mitigate this stress by bending with the waves, taking on smaller, more streamlined shapes that reduce drag. Coralline algae are limited in their flexibility due to having cell walls enriched in calcium carbonate, which cause thalli to be mostly rigid. While crustose species may avoid drag by growing prostrate along the substrate, most upright coralline species have uncalcified articulations (genicula) that allow them to retain flexibility despite their calcified constraint. Articulated corallines have evolved from crustose ancestors at least three separate times, leading to articulated species within the Corallinoideae, Lithophylloideae, and Metagoniolithoideae. The repeated evolution of genicula, and the rarity of upright coralline species without them, suggests that they play a key role in the ecological success of erect corallines.  While previous studies have noted structural and developmental differences among genicula in the three evolutionary lineages, I address multiple levels of genicular organization to investigate the depth of convergent evolution of these structures. I found that genicular tissues are stronger and more extensible than other fleshy seaweed tissues, reflecting the fact that genicula must undergo a high degree of stress and strain to compensate for rigidity elsewhere in the algal thallus (Ch. 2). Differences exist between articulated clades; corallinoids are particularly strong, while lithophylloids are often highly extensible (Ch. 2). Articulated clades also differ in the way genicular morphology and tissue properties are adjusted to increase thallus flexibility; corallinoids possess a high number of genicula, metagoniolithoids possess long genicula, and lithophylloids possess particularly pliant genicular tissues (Ch. 3). Both the content and structure of polysaccharides in the genicular cell wall varies depending on subfamily, reflecting differences in genicular development and potentially causing differences in material properties (Ch. 4). Results from polarized microscopy suggest that iv  the arrangement of polysaccharides within the cell wall also plays a role in how genicular tissue responds to mechanical stress (Ch. 5).  In summary, while genicula may serve similar functions in corallinoids, lithophylloids, and metagoniolithoids, I show that there is more than one way to build an articulated coralline.    v  Lay Summary Corallines are a group of calcifying red algae that dominate many marine intertidal and subtidal habitats worldwide. Most corallines grow as crusts or as upright, branching forms, particularly in areas of high-wave impact, where they provide both food and habitat for a wide array of organisms. Upright corallines often possess uncalcified joints called genicula, which are regularly spaced along the whole calcified plant and allow these otherwise rigid organisms to retain flexibility when hit by waves. Flexibility is very important in wave-swept habitats, as it allows algae to collapse and become smaller and more streamlined to avoid being dislodged. Genicula have evolved separately in three different coralline subfamilies, underscoring the importance of this structure in the survival and ecological success of articulated corallines. I compare the morphology, mechanical properties, and biochemistry of genicula produced by species in different subfamilies to determine which factors are consistently important in the repeated evolution of articulated coralline algae.  vi  Preface Components of Chapter 1 have been published in the introduction of: Janot, K.G., and Martone, P.T. (2016) Convergence of joint mechanics in independently evolving, articulated coralline algae. J. Exp. Biol. 219: 383-391; Janot, K.G., and Martone, P.T. (2018) Bending strategies of convergently-evolved, articulated coralline algae. J. Phycol. (in press, doi: 10.1111/jpy.12639). I wrote the manuscript, with input from Patrick Martone. A version of Chapter 2 has been published as: Janot, K.G., and Martone, P.T. (2016) Convergence of joint mechanics in independently evolving, articulated coralline algae. J. Exp. Biol. 219: 383-391. Experiments were designed in part by both me and Patrick Martone. Specimen collection was the group effort of myself, Patrick Martone, Mark Denny, John Huisman, Gerry Kraft, Rebecca Martone, Kathy Ann Miller, and John Statton. Experiments were conducted in the Martone Lab at UBC, as well as in lab space provided by John Huisman (Murdoch University in Perth, Australia) and Gary Kendrick (University of Western Australia in Perth, Australia). Raw data of tensile tests on Lithothrix aspergillum and Cheilosporum sagittatum was collected by Patrick Martone. Specimens imaged under TEM were prepared by Derrick Horne of the UBC BioImaging Facility. I performed all other data collection and analysis. I wrote the manuscript with the assistance of Patrick Martone.  A version of Chapter 3 has been published as: Janot, K.G., and Martone, P.T. (2018) Bending strategies of convergently-evolved, articulated coralline algae. J. Phycol. 54.3: 305-316. I designed the experiments, with input and advice from Patrick Martone. The computational model used for comparing bending strategies was originally designed by Patrick Martone and Mark Denny, and I subsequently modified it for use on a wider array of coralline species. I carried out all the experiments, with some assistance from Genie Chung, Wade Britz, and Cassandra Jensen. vii  Experiments were performed in either the Martone Lab at UBC, or in lab space provided by John Huisman (Murdoch University in Perth, Australia).    A version of Chapter 4 will be submitted as a research article for peer-review. Experiments were designed by me, Patrick Martone, and Shawn Mansfield, with protocols provided by Shawn Mansfield and Faride Unda. I performed all chemical extraction experiments in the Mansfield Lab at UBC, with the exception of the linkage analysis, which was performed by staff at the Complex Carbohydrate Research Center (University of Georgia, Atlanta). I performed the sequencing with the assistance of Cassandra Jensen and Katy Hind, and I carried out all data analysis. The chapter was written by me, with input and advice from Patrick Martone and Faride Unda.  A version of Chapter 5 will be submitted as part of a research article for peer-review, following additional experiments on microfibril angle in collaboration with other researchers. Enzyme extraction experiments were designed by me and Shawn Mansfield and carried out by me in the Mansfield Lab at UBC. Polarized microscopy experiments were designed by me, Patrick Martone, and John Gosline, and carried out in the Shadwick Lab at UBC. The custom-built micro-tensometer used for tensile tests was conceived by Patrick Martone, John Gosline, and me, and built by John Gourlay. Wade Smith provided equipment for and assisted with sectioning of corallines for tensile tests. I performed all data analysis and wrote the chapter with input from Patrick Martone.         viii  Table of Contents  Abstract ......................................................................................................................................... iii Lay Summary .................................................................................................................................v Preface ........................................................................................................................................... vi Table of Contents ....................................................................................................................... viii List of Tables .............................................................................................................................. xiii List of Figures ............................................................................................................................. xiv List of Symbols and Abbreviations ......................................................................................... xvii Acknowledgements .................................................................................................................... xix Dedication ................................................................................................................................... xxi Chapter 1: Introduction ................................................................................................................1 1.1 A unique group of macroalgae ........................................................................................ 1 1.2 Flexibility is key ............................................................................................................. 2 1.3 Genicula evolved multiple times .................................................................................... 3 1.4 Are genicula convergent at all levels of organization? ................................................... 6 Chapter 2: Convergence of material properties in articulated coralline algae ......................10 2.1 Introduction ................................................................................................................... 10 2.2 Materials and methods .................................................................................................. 12 2.2.1 Specimen collection .............................................................................................. 12 2.2.2 Pull-to-break tests ................................................................................................. 13 2.2.3 Transmission electron microscopy ....................................................................... 16 2.2.4 Cell wall analysis .................................................................................................. 16 ix  2.2.5 Statistics ................................................................................................................ 18 2.3 Results ........................................................................................................................... 19 2.3.1 Tissue breaking stress ........................................................................................... 19 2.3.2 Breaking strain ...................................................................................................... 20 2.3.3 Young's modulus ................................................................................................... 22 2.3.4 Toughness ............................................................................................................. 22 2.3.5 Cell wall thickness and stress ............................................................................... 23 2.4 Discussion ..................................................................................................................... 27 Chapter 3: Morphology and bending strategies of articulated coralline algae ......................34 3.1 Introduction ................................................................................................................... 34 3.1.1 Flexibility has consequences ................................................................................. 34 3.1.2 Articulated corallines are not analagous at all levels of organization .................. 36 3.2 Materials and methods .................................................................................................. 37 3.2.1 Specimen collection .............................................................................................. 37 3.2.2 Flexibility .............................................................................................................. 38 3.2.3 Morphometrics ...................................................................................................... 40 3.2.4 Modelling .............................................................................................................. 42 3.2.5 Bending stress ....................................................................................................... 42 3.2.6 Mechanical consequences of morphology ............................................................ 43 3.3 Results ........................................................................................................................... 44 3.3.1 Flexibility .............................................................................................................. 44 3.3.2 Morphometrics ...................................................................................................... 46 3.3.3 Bending stress ....................................................................................................... 46 x  3.3.4 Mechanical consequences of morphology ............................................................ 47 3.4 Discussion ..................................................................................................................... 53 3.5 Conclusions ................................................................................................................... 56 Chapter 4: Chemical composition of coralline cell walls .........................................................58 4.1 Introduction ................................................................................................................... 58 4.2 Materials and methods .................................................................................................. 63 4.2.1 Specimen collection .............................................................................................. 63 4.2.2 Species determinations .......................................................................................... 65 4.2.3 Tissue preparation ................................................................................................. 66 4.2.4 Neutral monosaccharide analysis .......................................................................... 66 4.2.5 Uronic acid analysis .............................................................................................. 67 4.2.6 Linkage analysis.................................................................................................... 68 4.2.7 Statistics ................................................................................................................ 69 4.3 Results ........................................................................................................................... 70 4.3.1 Species determinations .......................................................................................... 70 4.3.2 Monosaccharide composition ............................................................................... 71 4.3.3 Principal components anlaysis .............................................................................. 75 4.3.4 Linkage analysis.................................................................................................... 78 4.4 Discussion ..................................................................................................................... 80 4.4.1 Monosaccharide composition and linkage analysis .............................................. 81 4.4.2 Comparison of calcified and uncalcified tissues ................................................... 85 4.5 Conclusions ................................................................................................................... 88 Chapter 5: Correlating cell wall composition and tissue mechanics in coralline genicula ...90 xi  5.1 Introduction ................................................................................................................... 90 5.2 Materials and methods .................................................................................................. 94 5.2.1 Correlating monosaccharides with mechanical properties ................................... 94 5.2.2 Specimen collection .............................................................................................. 94 5.2.3 Enzyme digestions and tensile tests ...................................................................... 95 5.2.4 Quatification of extracted cellulose ...................................................................... 97 5.2.5 Polarized microscopy on corallines in tension ...................................................... 98 5.2.6 Estimating microfibril angle ............................................................................... 102 5.3 Results ......................................................................................................................... 103 5.3.1 Correlating monosaccharides with mechanical properties ................................. 103 5.3.2 Enzyme digestions and tensile tests .................................................................... 104 5.3.3 Microfibril angle of corallines in tension ............................................................ 107 5.4 Discussion ................................................................................................................... 112 5.5 Conclusions ................................................................................................................. 118 Chapter 6: Conclusions .............................................................................................................119 6.1 Major findings of this work ........................................................................................ 119 6.2 Future directions ......................................................................................................... 123 6.2.1 Biomechanical testing ......................................................................................... 123 6.2.2 Chemical analysis ............................................................................................... 124 6.2.3 Connecting chemistry and material properties ................................................... 126 6.2.4 Cellulose microfibril angle ................................................................................. 126 References ...................................................................................................................................128 Appendices ..................................................................................................................................139 xii  Appendix A: Genicular dimensions and breaking force ......................................................... 139 Appendix B: Mathematical details of MatLab bending model ............................................... 140 B.1 Cable model ............................................................................................................ 142 B.2 Solid model ............................................................................................................. 144 B.3 Model accuracy ....................................................................................................... 145 Appendix C: Estimation of cellulose thickness in genicular tissue ........................................ 146    xiii  List of Tables  Table 4.1 Monosaccharide composition (% dry weight) of coralline tissues ............................... 72 Table 4.2 Monosaccharide composition (mol%) of coralline tissues ........................................... 74 Table 4.3 Uronic acid composition of genicular tissue ................................................................. 75 Table 4.4 Linkage analysis of genicular tissue ............................................................................. 78 Table A.1 Area, length, and breaking force of genicula ............................................................. 138   xiv  List of Figures  Figure 1.1 Evolutionary relationships between select coralline subfamilies .................................. 4 Figure 1.2 Long-sections of genicula under light microscopy ....................................................... 6 Figure 2.1 Cross-section of Amphiroa anceps geniculum under light microscopy ...................... 17 Figure 2.2 Material properties of genicular tissue in tension ........................................................ 20 Figure 2.3 Average percent of genicular area taken up by primary and secondary cell wall ....... 24 Figure 2.4 Cross-section of Cheilosporum sagittatum geniculum under light microscopy ......... 24 Figure 2.5 Cross-sections of genicula showing differences in cell wall structure ........................ 25 Figure 2.6 Cell wall material breaking stress ................................................................................ 26 Figure 2.7 The correlation beween whole tissue breaking stress and cell wall percent ............... 27 Figure 3.1 Illustration of the method used for measuring frond flexibility .................................. 39 Figure 3.2 Close-up of genicula and intergenicula of different articulated corallines species ..... 41 Figure 3.3 Flexibility ranges of articulated coralline species, and average morphometrics of                   corallines that bent 50-70% or 70-90% of their initial length ..................................... 45 Figure 3.4 The correlation between predicted bending stress and genicular tissue strength ........ 47 Figure 3.5 The mechanical effect of computationally varing intergenicular radius ..................... 49 Figure 3.6 The mechanical effect of computationally varing genicular radius ............................ 50 Figure 3.7 The mechanical effect of computationally varying intergenicular length ................... 51 Figure 3.8 The mechanical effect of computationally varying genicular length .......................... 52 Figure 4.1 Evolutionary relationships between select coralline species ....................................... 63 Figure 4.2 Principal components of neutral monosaccharides from genicular, intergencular, and                    crustose coralline tissues ............................................................................................. 76 xv  Figure 4.3 Principal components of neutral monosaccharides in calcified tissues of different                         articulated and crustose coralline species ................................................................... 77 Figure 4.4 Principal components of neutral monosaccharides in genicular tissue of different                   articulated coralline species ......................................................................................... 78 Figure 5.1 Diagram of method used for tensile tests with an inverted polarized microscope ...... 79 Figure 5.2 Correlations between monosaccharide content and material properties ................... 101 Figure 5.3 First round enzyme treatment effects on material properties of genicular tissue ...... 105 Figure 5.4 Second round enzyme treatment effects on material properties of genicular tissue . 106 Figure 5.5 Enzyme treatment effects on glucose content in intergenicular and genicular tissue 107 Figure 5.6 Stress, strain, and estimated microfibril angle of genicula in tension ....................... 108 Figure 5.7 Longitudinal section of a coralline geniculum under increasing strain, visualized with                   polarized microscopy ................................................................................................. 110 Figure 5.8 Longitudinal section of a coralline geniculum beginning to rip under increasing strain,                   visualized under polarized microscopy ...................................................................... 111 Figure 5.9 The relationship between estimated microfibril angle and strain with different cellulose                    content estimates ....................................................................................................... 112 Figure B.1 Cable and solid modelling of genicula bending before and at the moment of                    intergenicular contact ................................................................................................ 132 Figure C.1 Schematic of top-down and side views of a longitudinal genicular section under                   tension ....................................................................................................................... 149 xvi  List of Symbols and Abbreviations A  planform algal thallus area aarea  cross-sectional area of cell lumen B  birefringence barea  combined cross-sectional area of cell lumen and cell wall carea combined cross-sectional area of cell lumen, cell wall, and half the extracellular matrix Cd  drag coefficient C.I.  confidence interval CO1 gene for mitochondrial cytochrome c oxidase subunit CW%  percentage of cell cross-sectional area accounted for by cell wall Dcell.  density of cellulose Dfrond.  density of whole coralline frond Dgen.  density of genicular tissue Dint.  density of intergenicular tissue E  Young’s modulus Fdrag  drag force k  half the stretched length of a geniculum in bending Lexposed  length of coralline segment projected horizontally during bending Linitial  starting length of coralline segment before bending l  length lo  initial length m  incremental addition of length to a geniculum in bending xvii  Mcell.  cellulose mass Mgen.  genicular mass Mexternal external bending moment Minternal internal bending moment ne refractive index of extraordinary ray of polarized light propagating through birefringent material parallel to the retardation azimuth of that material n’e refractive index of extraordinary ray of polarized light propagating through birefringent material in any direction no refractive index of ordinary ray of polarized light propagating through birefringent material in any direction Pcell. proportion of cellulose in genicular tissue Pgen. proportion of genicular tissue in coralline frond psbA  gene for photosystem II protein D1 R  retardation of light passing through material R  radius of curvature r1  long genicular radius r2  short genicular radius rbcL  gene for ribulose-biphosphate carboxylase s.e.m.  standard error of the mean T  initial thickness of longitudinal genicular section in tension T’  thickness of the middle of a longitudinal genicular section in tension t  effective thickness of a birefringent material   U  velocity at which a fluid is travelling xviii  Vcell.  volume of cellulose Vgen.  volume of geniculum w  initial width of longitudinal genicular section in tension w’  width of the middle of a longitudinal genicular section in tension x  genicular tissue hidden by intergenicular lips y  intergenicular radius   z  distance of any elemental area of geniculum to neutral axis β  half the contact angle at which the intergenicula of a bending coralline meet δ  lever arm  Ɛ  tensile strain η  distance between neutral axis and midline of geniculum in bending θ  angle of elemental area relative to centre of geniculum θi  angle of incidence of light propagating through a material ρ  density of a surrounding fluid σ  tensile stress σCW  tensile stress of cell wall σtissue  tensile stress of tissue φ  contact angle at which the intergenicula of a bending coralline meet  xix  Acknowledgements  Many, many people were involved in getting this thesis completed. First and foremost, I want to thank my supervisor, Dr. Patrick Martone, without whom this work would never have been conceived let alone brought to fruition. It is perhaps obvious that he would have a large hand in the making of this thesis, but his support goes beyond the academic. Patrick, I would like to thank you for not only being a source of intellectual inspiration and support, but also for keeping me simultaneously grounded and optimistic when various experiments have fallen apart. I do not believe I would have been able to complete a thesis with anyone else, let alone pursue a PhD.  I would also like to thank my ever-patient graduate committee. I would like to thank Dr. Shawn Mansfield for allowing me use of his lab space, his equipment, his materials, and most of all his time and intellectual energy to pursue avenues of questioning that would otherwise be out of reach. I would like to thank Dr. Geoff Wasteneys, for his constant willingness to meet and discuss cell walls and for his enthusiasm, which were a constant reminder that this work had worth. Finally, I would like to thank Dr. Brian Leander, who inspired me to pursue marine biology when I was but a flailing undergraduate student, and who has been a voice of practicality in matters of publication and work-life balance.   I am extremely thankful to have had amazing lab mates throughout the course of my graduate school career. Katy Hind, thank you for your patient instruction, your grounded feedback, and your impressive frisbee skills. Lauran Liggan, you have been not only a fantastic person to share lab space with, but an amazing friend who has taught me to allow my creative side out occasionally. Sam Starko, thank you for being an excellent source of input for all aspects of my thesis, and for your constant willingness to talk about science in pretty much any scenario. Liam Coleman, you are a xx  constant source of brightness when things get frustrating and that has been invaluable in the home stretch.  Laura Borden, thank you for your constant willingness to offer input on both manuscripts and conference presentations, even while maintaining such a busy work and school schedule. To Soren Huber, Rebecca Guenther, Laura Anderson, Jenn Clark, and Matt Whalen; I’m so happy I had the chance to work alongside you, and to share a rabid love of seaweed with you.  I have had the opportunity to work with and learn from many amazing individuals over the course of my graduate career. Dr. Mark Denny and Dr. John Gosline both offered a great deal of feedback and assistance in various aspects of this thesis, and I am immensely thankful for being able to pick their brains about various aspects of biomechanical theory.  Dr. Faride Unda has patiently walked me through many a chemistry protocol and put an inordinate amount of time into assisting me with analyzing hundreds of coralline tissue samples. Dr. Wade Smith offered me the breakthrough in coralline sectioning techniques that I needed to perform tensile tests under a microscope, and without him Ch. 5 would likely not exist.  I would be remiss if I didn’t acknowledge the initial sources of inspiration for my love of phycology. Samantha Iverson, thank you for speaking so excitedly about the seaweed class you were taking, all those years ago. I chose to take a phycology course as a direct result of your enthusiasm, and it has apparently gotten somewhat out of hand. To Dr. Rob DeWreede, thank you for teaching that course, and for making me realize how nuanced and fascinating phycology is. Thank you for also making sure I was “really sure” before giving me a letter of recommendation for graduate school.   Thank you to my family (especially my parents), my friends, and my amazing husband Beau Gravlin. Thank you for believing in me, even when I didn’t believe in myself. I couldn’t have gotten this far without you.       xxi  Dedication This thesis is dedicated to my parents, Mark and Jonnita Janot, and my little brother, Stefan Janot.  This is what happens when you teach someone to love the ocean and ALL the weird, wonderful things in it. 1  Chapter 1: Introduction  1.1 A unique group of macroalgae The red algal subclass Corallinophycideae is composed of four orders: Corallinales, Hapalidiales, Sporolithales, and Rhodogorgonales (Le Gall et al. 2010, Nelson et al. 2015). Most algae within this subclass are referred to as “corallines,” morphologically united by their coral-like hard, calcified thalli, which result from calcite crystals that are deposited within their cell walls (Johansen 1981). Algal calcification is not unique to the corallines - the orders Nemaliales (Rhodophyta), Caulerpales (Chlorophyta), and Dasycladales (Chlorophyta), as well as the family Peysonneliaceae (Rhodophyta) and the genus Padina (Ochrophyta), are all examples of other algal groups that contain calcareous species (Bilan and Usov, 2001). Corallines are notable, however, for making up a large clade of over 500 species that calcify without exception (Bailey and Chapman 1998, Bilan and Usov 2001, Brodie and Zuccarello 2007). The location of calcification is also unusual; while corallines calcify within their cell walls, most other groups calcify intercellularly or extracellularly (i.e. within the extracellular matrix) (Borowitzka and Larkum 1987, Bilan and Usov 2001). Corallines exhibit a variety of growth forms, ranging from smooth crusts to branching, upright articulated morphologies (Johansen 1981). Between these two extremes are crusts of varying degrees of bumpiness, free-living rhodoliths, and reduced articulated corallines composed of miniscule uprights with large crustose bases (Martone et al. 2012). Corallines occupy oceans worldwide in habitats ranging from the high intertidal to depths of over 200 m (Littler et al. 1986), and they are particularly successful on rocky, exposed shores (Steneck 1986).  2  1.2 Flexibility is key Wave-exposed marine environments are places of extreme hydrodynamic stress, due to waves that can reach velocities of up to 25 m s-1 when they crash upon the shore (Denny et al. 2003). Sessile organisms such as algae cannot relocate to avoid exposure and must contend with the drag forces imposed upon them. Drag forces are dependent on size and shape, and can be calculated from the drag equation: Fdrag = 0.5ρ𝑈𝑈2𝐴𝐴𝐶𝐶d Eq. 1.1  where, ρ is the density of the surrounding fluid (seawater is approximately 1025 kg m-3), U is the velocity at which the fluid is travelling, A is algal area projected into flow (also sometimes measured as planform area), and Cd is the drag coefficient, a dimensionless number that takes into account shape and reconfiguration (Denny 1995, Denny and Gaylord 2002, Harder et al. 2004). Therefore, algae can minimize the drag imposed on them by remaining small (to reduce A) or by exhibiting a streamlined shape (to reduce Cd) (Denny 1999, Harder et al. 2004, Boller and Carrington 2006). Many macroalgae are small (e.g. Fucus, Ulva, Mastocarpus), but there are exceptions, such as kelps that reach several metres in length. Furthermore, branched seaweeds are generally not considered streamlined (Denny and Gaylord 2002, Starko and Martone 2016). Large or branching seaweeds must nonetheless employ some method of reducing drag to survive. One strategy exhibited by macroalgae is flexibility, which allows them to reconfigure branches and blades into smaller, more streamlined shapes (Denny and Gaylord 2002, Harder et al. 2004, Boller and Carrington 2006); this means that flexible algae take on a drag-reducing form only when drag is imposed, via passive reorientation. 3  Most fleshy algae are flexible along their entire lengths. However, the calcification of corallines causes this group to be mostly rigid. How do these algae withstand wave-induced drag? Crustose species grow prostrate on the substrata and remain in slower moving water; when water velocity is higher, they use the rock for support. There are, however, many upright coralline species that also survive in habitats with high wave action. To achieve flexibility, most upright corallines have evolved uncalcified joints, called genicula, that separate calcified segments, called intergenicula; together, these components make up calcified-yet-flexible upright fronds. This key innovation of joints in an otherwise rigid thallus may contribute to the success of articulated corallines in wave-swept habitats worldwide.  1.3 Genicula evolved multiple times Various evidence suggests that upright articulated corallines evolved from prostrate, crustose coralline ancestors. This evolutionary trajectory is supported both by the fossil record (Aguirre et al. 2010, Kundal 2011) and by molecular phylogenetics (Bailey and Chapman 1998, Bittner et al. 2011, Kato et al. 2011). Moreover, these data suggest that articulated thalli are polyphyletic and evolved from crustose corallines multiple times, leading to articulated coralline algae in three different subfamilies: Corallinoideae, Lithophylloideae and Metagoniolithoideae (Fig. 1.1, and see Johansen 1981). Genicula in these lineages are nonhomologous, and therefore represent a clear example of convergent evolution.   4   Fig. 1.1. Evolutionary relationships between articulated coralline subfamilies and select crustose subfamilies, showing similarities in thallus morphology and differences in genicular morphology. (A) Phylogenetic summary of relationships between coralline subfamilies (in bold) (information from Kato et al., 2011). (B, D, F) Coralline fronds in situ. (C, E, G) Close-up of genicula (g) under a dissecting microscope. Scale bars=700 μm. (B, C) Calliarthron tuberculosum.  (D, E) Amphiroa anceps. (F, G) Metagoniolithon stelliferum #1.  Convergent evolution is generally defined as the independent evolution of phenotypic similarities (Stern 2013, Stayton 2015), and it is a fairly common phenomenon – the presence of wings in birds, bats, and insects, the similarly streamlined body plan of dolphins and sharks, and the complex camera eye of mammals and cephalopods are all well-known examples of convergent evolution (Losos 2011, Stayton 2015, Speed and Arbuckle 2017). Convergence at smaller scales is also well documented, such as tetrodotoxin resistance in the sodium channels of some snakes (Feldman et al. 2012) and increased hemoglobin-oxygen affinity in high-altitude birds (Natarajan et al. 2016). Despite the seeming ubiquitous nature of convergent evolution, the way in which it is interpreted is complicated – while some researchers see convergence as a testament to the capability of natural selection to produce “optimized” solutions to environmental pressures, others see it as an indication of constraint resulting from either genetic or 5  developmental biases (for a discussion of varying interpretations, see Losos 2011, Stayton 2015, Speed and Arbuckle 2017). One way to differentiate between these two general causes of convergence is to look for some adaptive significance behind the evolution of the convergent trait (Losos 2011). If it can be demonstrated that (1) the species or populations with a convergent trait have some similarity in their environmental setting, and (2) that the convergent trait provides a similar functional advantage within that setting, this suggests that the trait has arisen due to adaptation (although it does not rule out the effect of drift or other constraints, see Losos 2011).   When determining the presence and cause of convergent evolution, it is important to distinguish between form and function (Doolittle 1994, Losos 2011, Speed and Arbuckle 2017). Convergence may exist with regard to one or the other, or both. In a loose sense, convergence of function seems evident in genicula, as they allow for otherwise rigid coralline structures to retain flexibility to some degree. However, convergence of form is not as clear, and could include overall genicular morphology, cell wall morphology, and even chemical composition. Has evolution yielded functionally similar structures that have converged at all levels of organization, or do genicula perform similar functions despite diverging in composition or structure?  Developmental and structural divergence between genicula in Corallinoideae, Lithophylloideae and Metagoniolithoideae have already been described (Johansen 1969, 1981, Ducker 1979, Woelkerling 1988). One of the most striking differences is how genicula are formed: corallinoid and lithophylloid genicula form via decalcification of medullary cells and a cracking of the surrounding cortex (Johansen 1969, Johansen 1981), whereas metagoniolithoid genicula form when meristematic cells switch production from calcified to uncalcified cells (Ducker 1979). In addition, genicula in the Corallinoideae and Lithophylloideae can form at 6  intervals along a branch and at the branching nodes, but genicula in Metagoniolithoideae form only at branching nodes (Johansen 1981, Ducker 1979). Differences between the corallinoids and the lithophylloids also exist. Joints in the corallinoids are completely decalcified except for the ends inserted into adjoining intergenicula, while joints in some lithophylloid species retain calcification on the surface (Johansen 1969, 1981). Moreover, the genicula in Corallinoideae are composed of a single tier of cells, while those in the Lithophylloideae may be single- or multi-tiered (Johansen 1969, 1981; Fig. 1.2). Metagoniolithoids have joints that lack a clearly tiered structure (Ducker 1979).  Fig. 1.2. Long-sections of genicula under light microscopy, dyed with 1% Aniline Blue. Scale bars, 100 μm. Arrows and ‘g’ labels indicate location of genicular tissue – note that all tissue shown in C is genicular tissue. (A) Calliarthron tuberculosum (Corallinoideae). (B) Amphiroa gracilis (Lithophylloideae). (C) Metagoniolithon stelliferum #1 (Metagoniolithoideae).   1.4 Are genicula convergent at all levels of organization? The goal of my thesis is to compare the mechanical and chemical properties of these independently derived joints, to determine the degree of convergence that has resulted from overcoming the biomechanical constraint of calcification in wave-swept habitats. How mechanically effective are the genicula of these groups given their documented developmental and structural differences? Are all genicula built the same way in terms of chemical composition? I compared the genicula produced by species within the Corallinoideae, 7  Lithophylloideae, and Metagoniolithoideae at multiple levels of organization, and attempted to link these levels together to clarify the process of convergent evolution and to build a more complete picture of what it means to be an articulated coralline.  In Chapter 2, I compare the performance of genicular tissue from multiple species under tension. I demonstrate that coralline joints are often stronger and more extensible than other algal tissues, and that corallinoids are particularly strong overall. I use histological techniques to demonstrate that the high strength of corallinoid species is largely due to the presence of secondary cell walls, which strengthen the joint tissue without adding bulk to the joint itself. Cell wall thickness is shown to be a contributing factor to strength across all groups, except for the corallinoid Cheilosporum sagittatum, which may possess a distinct chemical composition within its walls.  In Chapter 3, I explore the interaction between flexibility and morphological variation in articulated corallines, and investigate whether representatives of convergently evolving clades follow similar strategies to generate mechanically successful articulated fronds. By using computational models to explore different bending strategies, I show that there are multiple ways to generate flexibility in upright corallines, although not all morphological strategies are mechanically equivalent: corallinoids have many genicula, metagoniolithoids have long genicula, and lithophylloids have pliant genicular tissues. While these strategies can lead to comparable thallus flexibility, they also lead to different levels of stress amplification in bending. Moreover, genicula at greatest risk of stress amplification are typically the strongest, mitigating the trade-off between flexibility and stress reduction.  In Chapter 4, I investigate whether the abundance and identity of polysaccharides in genicular and intergenicular tissue differs between articulated clades. I also analyze the chemical 8  composition of select crustose coralline species to see whether phylogenetic relationships can predict cell wall chemistry of calcified tissue across different coralline morphologies. I analyze the monosaccharide composition of multiple species from each articulated clade and interpret the results of linkage analysis on representative species from each clade. Data presented in this chapter indicate that developmental and mechanical differences in genicular tissues from different articulated clades are underscored by differences in monosaccharide composition; conversely, chemical similarity between calcified tissues from both articulated and crustose species suggests that little has changed chemically in these tissues over the course of evolution.   In Chapter 5, I explore the link between material properties of genicular tissue and chemical composition of genicular cell walls. A correlation between glucose content and strength in different articulated species suggests that cellulose may play a role in the strength of genicular tissues, and I use enzyme extractions and polarized microscopy to further investigate this correlation. By performing tensile tests under a polarized microscope, I visualize changes in the birefringence of genicular tissues from Calliarthron tuberculosum and estimate the change in cellulose microfibril angle with increasing strain. Data suggest that shifts in stiffness with increased strain are related to realignment of the cellulose network within the cell wall.  In sum, I utilize comparative, biomechanical, computational, histological, and chemical techniques to gain an integrated understanding of what is involved in “building” an articulated coralline. Studying species from multiple articulated clades allows me to determine which aspects of genicula are universal, so I can begin to elucidate which factors may have adaptive significance. Factors that are mechanically necessary would be expected to show up in all articulated clades, while factors that are clade-specific may be the result of divergent selection, stochastic mechanisms, or genetic/developmental constraints.  By investigating genicula at 9  multiple levels of organization, I gain a more complete picture of the relationship between these levels, and the role each plays in overall mechanical performance of articulated corallines in wave-exposed, rocky shores.   10  Chapter 2: Convergence of material properties in articulated coralline algae  2.1  Introduction Wave-swept, rocky shorelines are a place of extreme hydrodynamic stress. Organisms living in these habitats are subject to water velocities that regularly reach 2 m s−1 as waves break, with velocities as high as 25 m s−1 being recorded in intertidal surf (Denny 1988, Denny et al. 2003). These waves impose high drag forces upon sessile organisms such as algae, and the magnitude of that drag depends upon both the size and the shape of the organism. Flexible seaweeds optimize both size and shape by bending over to minimize projected area and reconfiguring branches or blades into more streamlined shapes (Denny and Gaylord 2002, Harder et al. 2004, Martone 2006, Martone et al. 2012). Most upright algae are generally flexible along their entire thallus, but coralline algae demonstrate a unique and interesting exception; many upright species have evolved uncalcified joints, called genicula, that allow them to retain flexibility in an otherwise rigid thallus.  Although an articulated morphology allows upright corallines to bend over and reconfigure in a manner similar to fleshy algae, it also presents unique biomechanical challenges. Bending occurs only at discrete joints along articulated thalli, and so joints must be composed of materials that are both extensible enough to retain flexibility, and strong enough to resist amplified bending stress (Martone and Denny 2008a). Furthermore, joints must also resist tensile forces associated with drag, after bending has occurred. Genicula in the corallinoid Calliarthron cheilosporioides are composed of tissues that are often more extensible than other red algal tissues (Hale 2001), as well as 35–400% stronger than other red algal tissue (Hale 2001, Kitzes 11  and Denny 2005, Martone 2006), likely indicating a necessity for high performing materials in these structures. The exceptional material properties of C. cheilosporioides likely contribute to its dominant abundance in wave-swept intertidal habitats where it is found, but do other articulated corallines display similar properties? Structural differences between genicula in the corallinoids, lithophylloids and metagoniolithoids previously described (Johansen 1969, 1981, Ducker 1979) could affect the mechanical performance of joints under bending stress. For example, corallinoid genicula are unique in being composed of a single tier of cells that are anchored to adjacent intergenicula, but only loosely connected to one another laterally (Johansen 1969, 1981, Martone and Denny 2008a, Denny et al. 2013). Lithophylloid genicula are often multi-tiered, whereas metagoniolithoid genicula lack a tiered structure altogether (Johansen 1969, 1981, Ducker 1979). Genicular cells in C. cheilosporioides (Corallinoideae) also possess secondary cell walls that likely play a role in strengthening genicular tissue (Martone 2007b, Martone et al. 2009), whereas no similar feature has been documented in either lithophylloids or metagoniolithoids.  This study aims to investigate the organizational level at which joints have convergently evolved in articulated coralline algae by comparing their material properties, which are integral to the function of joints under hydrodynamic stress. Given the unique mechanical challenges posed by possessing a jointed morphology, I hypothesize that genicular tissue in all three groups is both stronger and more extensible than other fleshy red algal tissues. The maximum material stress and strain required to break joint tissue was measured, as well as the stiffness of joint tissue during loading in tension. Tensile toughness (strain energy density, i.e. the energy absorbed before breaking) was calculated from the area under a stress–strain curve; an alga can achieve toughness by being very strong, very extensible, or both. Toughness has been widely 12  reported for marine plant tissues (Koehl and Wainwright 1977, Armstrong 1988, Patterson et al. 2001, Harder et al. 2006); however, the biological significance of this property is unclear (Denny and Gaylord 2002, Denny and Hale 2003). Finally, I explored whether any apparent differences in material properties among the three subfamilies could be attributed to differences in cellular structure or thickness of the cell wall.  2.2 Materials and methods  2.2.1 Specimen collection Cheilosporum sagittatum was collected from Gleneuse Reef, Point Lonsdale, Victoria, Australia (38°17′37″ S, 144°36′47″ E), in January 2009, from a depth of ∼3.0 m. Calliarthron tuberculosum, Corallina officinalis var. chilensis, and Johansenia macmillanii were collected subtidally at a depth of ∼3.0 m from Botanical Beach (48°31′48″ N, 124°27′18″W) on Vancouver Island, BC, Canada, in June/July 2012. Lithothrix aspergillum was collected from Potato Harbor (32°02′52″ N, 119°35′31″ W) on Santa Cruz Island, CA, USA, at depths of 4.6–5.2 m, in September 2006. Amphiroa anceps and Amphiroa gracilis were collected in Point Peron (32°16′01″ S, 115°41′14″ E), Perth, Western Australia, at depths of 3.0–4.6 m, in December 2012. Metagoniolithon stelliferum #1 and #2 indicate specimens that currently fall under the name Metagoniolithon stelliferum, but that appeared morphologically distinct in the field. Sequencing of psbA, CO1 and rbcL genes indicates that these two groups represent distinct species (K.G.J., unpublished data), and so they have been treated as such in this study. Both M. stelliferum “species” were collected in December 2012 from Point Peron (32°16′01″ S, 13  115°41′14″ E), Perth, Western Australia, at depths of 3.0–4.6 m, where they were growing epiphytically side by side on seagrass. Metagoniolithon chara was collected off Carnac Island (32°07′07″ S, 115°39′52″ E) near Perth, Western Australia, at depths of ∼4.6 m, in January 2014. All plants were collected in their entirety and kept in flowing seawater in the laboratory prior to mechanical testing. Mechanical tests were performed no later than 72 hrs after collection, and remaining specimens were air dried for later microscopic analysis. Representative vouchers for each species were deposited into the University of British Columbia Herbarium for future taxonomic reference: Cheilosporum sagittatum (A88599); Calliarthron tuberculosum (A91564); Corallina officinalis var. chilensis (as Corallina officinalis, A91563); Johansenia macmillanii (A91561); Lithothrix aspergillum (A88575); Amphiroa anceps (A91566); Amphiroa gracilis (A91572); Metagoniolithon stelliferum #1 (as Metagoniolithon stelliferum, A91576); Metagoniolithon stelliferum #2 (as Metagoniolithon sp., A91579); and Metagoniolithon chara (A91464).  2.2.2 Pull-to-break tests Calliarthron tuberculosum (n=15), Corallina officinalis var. chilensis (n=15), and Johansenia macmillanii (n=12) were tested using a standard tensile method on a computer-interface tensometer (model 5500R, Instron Corp., Canton, MA, USA). Basal 2–3 cm segments were held in pneumatic clamps lined with neoprene and sandpaper, which provided both cushioning and friction. Each segment included multiple genicula that floated between the clamps – the exact number varied depending on the species. Samples were wetted with seawater after being mounted in the clamps and before testing. Extension was continuously measured via 14  movement of the crosshead, and force was measured via a 500 N tension load cell. Specimens were directly observed during testing to monitor slippage in the clamps, and tests in which slippage occurred were not included in analysis. The crosshead was set to move at a rate of 10 mm min−1 until tissue failure, as measured by a sudden drop in force. All data were collected and initially processed using Instron Bluehill 3 software (Instron Corp.). Lithothrix aspergillum (n=9) was tested with the same custom built, portable tensometer described in Martone (2006). In short, fronds were held between two sets of aluminum clamps that moved along a tensometer track. Clamps were positioned on the intergenicula and lined with rubber pads to prevent the calcified tissue from being crushed. Force was quantified as the deflection of a stationary clamp mounted to two steel beams, measured by a linearly variable differential transformer (LVDT; model 100HR, Schaevitz Engineering, Pennsauken, NJ, USA). Strain was measured directly using a video camera (model TMC-S14, Pulnix Sensors, Sunnyvale, CA, USA) and video dimension analyzer (model V94, Living Systems Instrumentation, Burlington, VT, USA), which tracked the relative position of intergenicula flanking individual joints in each stretched specimen. Specimens were pulled at a rate of 60 mm min−1 until failure. Cheilosporum sagittatum (n=10), Amphiroa anceps (n=15), Amphiroa gracilis (n=12), Metagoniolithon stelliferum #1 (n=20), Metagoniolithon stelliferum #2 (n=15), and Metagoniolithon chara (n=14) were tested with a second custom-built, portable tensometer. Fronds were clamped in a manner similar to that described in Martone (2006), with the aid of a motor (model SM2315D, Moog Animatics, Milpitas, CA, USA) controlled via the SmartMotor Interface (Moog Animatics). Force was measured with a 5 kg beam transducer (model FORT5000, World Precision Instruments, Sarasota, FL, USA), which was amplified through a 15  transducer amplifier (model SYS-TBM4M, World Precision Instruments) and collected in real time using LabVIEW SignalExpress software (National Instruments Canada, Vaudreuil-Dorion, QC, Canada). Extension was measured as the displacement of the mobile clamp, calculated from the number of rotations of the motor. Specimens were pulled at a rate of 60 mm min−1 until failure. As this study includes data collected over a span of 5 years, using two different extension rates, I compared results for five specimens of Calliarthron tuberculosum that were also tested in the portable tensometer at a rate of 60 mm min−1, as described above. Breaking stress, breaking strain, Young’s modulus and breaking energy all fell within the ranges found for C. tuberculosum specimens tested in the Instron tensometer. After testing, samples were dissected under a dissecting microscope (model SZ61, Olympus Canada, ON, Canada) with an attached camera (model DP20, Olympus Canada) to measure cross-sectional area of the broken interface (estimated as elliptical) and cumulative genicular length (i.e. length of all genicula in the testing area added together). All specimens tested with the custom portable tensometer were first dried for transport, then rehydrated in saltwater for at least 10 min prior to morphometric measurements. Stress (σ, MPa) was obtained by dividing force measurements by cross-sectional area of the broken geniculum, and strain (ε) was calculated by dividing extension (l) by initial cumulative genicular length (lo). The resulting stress–strain curve was used to calculate Young’s modulus (E, MPa), a measurement of initial tissue stiffness, by taking the slope of the curve from 0 to 0.1 strain. Breaking strain energy density (MJ m−3), or toughness, was calculated from the total area under the stress–strain curve when specimens were pulled to break. 16  Data from fleshy red, green, and brown algal tissues were compiled from Hale (2001). Three species from each group were selected to represent a large range of values of breaking stress, breaking strain, Young’s modulus and breaking energy. These species were graphed alongside data from this study for comparative purposes.  2.2.3 Transmission electron microscopy One representative species was chosen to illustrate each subfamily – Calliarthron tuberculosum for Corallinoideae, Amphiroa anceps for Lithophylloideae, and Metagoniolithon stelliferum for Metagoniolithoideae. One specimen of each representative species was rehydrated for 1 hr in seawater and fixed overnight in 5% formalin seawater. Fixed specimens were decalcified overnight in HCl, and then dehydrated in increasing concentrations of ethanol (25%, 50%, 75% and 100%) for 1 h per treatment. Specimens were left in 100% ethanol overnight, then placed in medium-grade LR White embedding resin overnight. Specimens were placed in gel capsules, immersed in fresh LR White embedding resin, and baked at 62°C for 1.5 hr. Resin blocks were sectioned using a diamond knife mounted on an ultramicrotome (model Ultracut T, Leica Biosystems, Nussloch, Germany). Sections were mounted on formvar-coated 100 mesh copper grids and stained with uranyl acetate for 17 min and Reynold’s lead citrate for 6 min. Sections were visualized and photographed on a transmission electron microscope (model H7600, Hitachi High-Technologies Canada, Toronto, ON, Canada).  2.2.4 Cell wall analysis Five specimens of each species were rehydrated in saltwater for a minimum of 10 min, after which they were decalcified in 0.1 M HCl for between 2 and 24 hrs (the time required for 17  full decalcification varied widely between species). Decalcified samples were placed in ethanol for 10 min, embedded in TissueTek OCT compound (Sakura Finetek, Europe), then cross-sectioned within the basal genicular region using a freezing microtome (model CM1850, Leica Biosystems). Thickness of sections varied between 10 and 20 μm. Sections were dyed with 5% potassium permanganate for approximately 5 min, then washed with freshwater and viewed under a light microscope (model BX51wi, Olympus Canada). Photos were taken using a camera (DP21, Olympus Canada) attached to the microscope. Photos were analyzed using ImageJ (US National Institutes of Health, Bethesda, MD, USA). Cell wall proportion in a cross-section was estimated by drawing and measuring the area of polygons around the lumen (a), lumen+cell wall (b), and lumen+cell wall+half the extracellular matrix/middle lamella (c) (Fig. 2.1). Cell wall percentage was calculated as: CW% = �barea − aareacarea � ∗ 100 Eq. 2.1 Cell wall percentage was calculated for 20 randomly selected cells per cross-section, and then averaged to obtain one value per specimen. In the case of the metagoniolithoids, both cortex and medulla cells were visible and had slightly different morphologies; measured cells were split evenly between the two layers, and the final average weighted these measurements depending on the proportion of each tissue layer in the overall cross-section.  18   Fig. 2.1. Cross-section of Amphiroa anceps geniculum. Letters indicate polygons used to measure different cell layers: a, cell lumen; b, cell wall; and c, extracellular matrix (halved to account for portion associated with other cells). Scale bar=5 μm.  Given that coefficients of variation (CV) for CW% were generally low, under 0.1 for all species except Cheilosporum sagittatum (CV=0.17), an average cell wall percentage was calculated for each species. This average was used to correct breaking stress values obtained from pull-to-break tests, calculated as: σCW = σtissue ∗ 100CW% Eq. 2.2 One specimen of Cheilosporum sagittatum was embedded in LR White resin using the same protocol described for transmission electron microscopy. Sections of 10 μm were obtained with an ultramicrotome (Porter-Blum MT-2, Sorvall Products, New Castle, DE, USA), and stained and visualized with the same methods used for the cryosections.  2.2.5 Statistics As unequal variances between species could not be solved with either logarithmic or square root transformations, non-parametric Kruskal–Wallis tests and post hoc Dunn’s tests were 19  performed to compare breaking stress, breaking strain, Young’s modulus and breaking energy, as well as cell wall stress. Statistical comparisons were made at the species level only. This was done in R 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria) using the RStudio interface (version 0.98.1056, RStudio, Boston, MA, USA) and the dunn.test() function from the dunn.test package (dunn.test: Dunn’s test of multiple comparisons using rank sums, version 1.2.3, Alexis Dinno 2015). The relationship between tissue stress and cell wall percent in cross-section was tested in R 3.0.1 with a one-way ANOVA using the lm() and anova() functions from the base stats package. Means and standard errors reported for each subfamily were calculated by pooling all data from all species within each subfamily. Statistics were not performed at the subfamily level.  2.3 Results  2.3.1 Tissue breaking stress Average tissue breaking stress (mean ± s.e.m.) was 31.9 ± 2.0 MPa for the corallinoids, 10.7 ± 0.6 MPa for the lithophylloids, and 5.6 ± 0.5 MPa for the metagoniolithoids. Average stress varied significantly between species (Kruskal–Wallis test, p<0.001), and these differences were consistently segregated among subfamilies (Dunn’s test; Fig. 2.2A). With the exception of Johansenia macmillanii, all corallinoid species were significantly stronger than all lithophylloid and metagoniolithoid species tested. Cheilosporum sagittatum was the strongest of the corallinoids, with an average breaking stress of 56.3 ± 2.4 MPa, over 75% stronger than the next strongest species, Corallina officinalis var. chilensis, with an average breaking stress of 31.7 ± 2.0 MPa. 20  All species tested appeared stronger than typical green and brown algal tissues (Fig. 2.2A). While corallinoid and lithophylloid species were consistently stronger than other red algal tissues, metagoniolithoid species fell within the range for fleshy red algae reported by Hale (2001).  2.3.2 Breaking strain Average breaking strain (mean ± s.e.m.) was 0.77 ± 0.04 for the corallinoids, 0.84 ± 0.06 for the lithophylloids, and 0.56 ± 0.07 for the metagoniolithoids. Average breaking strain was significantly different among species (Kruskal–Wallis test, p<0.001), although this variation was unrelated to subfamily (Dunn’s test; Fig. 2.2B). Both lithophylloids and metagoniolithoids contained species with some of the highest average breaking strains (e.g. Amphiroa anceps: 1.02 ± 0.09, and Metagoniolithon chara: 1.07 ± 0.14) and some of the lowest (e.g. Lithothrix aspergillum: 0.35 ± 0.04, and Metagoniolithon stelliferum #1: 0.32 ± 0.02). Many of the species tested had breaking strains that were double or more than that of other red, green, and brown algal tissues (Fig. 2.2B). Even the lowest strains (i.e. that of Lithothrix aspergillum and Metagoniolithon stelliferum #1) were on the higher end of the range for other algal tissues. 21  Fig. 2.2. Material properties of genicular tissue in tension. (A) Breaking stress, (B) breaking strain, (C) Young’s modulus, and (D) breaking energy of each species. Corallinoid species are in green, lithophylloid species are in purple, and metagoniolithoid species are in orange. Significant differences between species were found for breaking stress, breaking strain, Young’s modulus, and breaking energy (Kruskal-Wallis tests, p<0.001 in all cases). Lowercase letters indicate results of a non-parametric post hoc Dunn’s test (p<0.05). Grey bars show comparative data for fleshy algae: red (=Rhodophyta), brown (=Ochrophyta), and green (=Chlorophyta) from Hale (2001). Error bars represent s.e.m. 22   2.3.3 Young’s modulus Average Young’s modulus (initial stiffness, mean ± s.e.m.) was 51.7 ± 4.6 MPa for the corallinoids, 19.5 ± 3.7 MPa for the lithophylloids, and 15.8 ± 2.2 MPa for the metagoniolithoids. Although average Young’s modulus was significantly different among species (Kruskal–Wallis test, p<0.001), it was highly variable for all species, and did not differ consistently among subfamilies (Dunn’s test; Fig. 2.2C). Cheilosporum sagittatum had the highest average stiffness, with a Young’s modulus of 92.1 ± 14.4 MPa, over 75% higher than the modulus of Corallina officinalis var. chilensis (51.8 ± 6.5 MPa). While the stiffest species came from the corallinoids, high modulus values were also found in the lithophylloids and metagoniolithoids – Lithothrix aspergillum had a modulus of 43.6 ± 11.6 MPa, and Metagoniolithon stelliferum #1 had a modulus of 30.3 ± 3.3 MPa. With the exception of Cheilosporum sagittatum, which was almost twice as stiff as the stiffest fleshy red species tested by Hale (2001), most coralline species tested fell within the range of stiffness reported for other red algal tissues (Fig. 2.2C). Consistent with fleshy red algae, all coralline species were stiffer than green algal tissues, though not notably different from brown algal tissues.  2.3.4 Toughness Average toughness (breaking strain energy density, mean ± s.e.m.) was 15.7 ± 1.5 MJ m−3 for the corallinoids, 5.0 ± 0.5 MJ m−3 for the lithophylloids, and 2.2 ± 0.6 MJ m−3 for the metagoniolithoids. Average toughness differed among species (Kruskal–Wallis test, p<0.001). All corallinoid species were tougher than all metagoniolithoid species, while lithophylloid 23  species were not significantly different from either corallinoids or metagoniolithoids (Dunn’s test; Fig. 2.2D). Cheilosporum sagittatum had the highest breaking energy at 31.6 ±  3.8 MJ m−3 – more than double that of the next toughest species, Corallina officinalis var. chilensis, which had a breaking energy of 12.1 ± 1.2 MJ m−3. Almost all species tested had a higher toughness than the fleshy red, green, and brown algal tissues tested by Hale (2001). Metagoniolithon stelliferum #2 fell within the range of other red algae.  2.3.5 Cell wall thickness and stress With the exception of Cheilosporum sagittatum, corallinoid species had genicula with proportionally thicker cell walls than genicula in lithophylloid and metagoniolithoid species (Fig. 2.3). This was largely due to the presence of a secondary cell wall, which roughly doubled the total cell wall area in cross-section. Secondary walls were not consistently visible in C. sagittatum, with the exception of the most basal tissue (Fig. 2.4). In addition, all lithophylloids tested, as well as one metagoniolithoid species (Metagoniolithon stelliferum #1), had large non-fibrillar extracellular spaces/middle lamellae that may have been either not present or not visible to the naked eye in other metagoniolithoids, or in any of the corallinoids (see Fig. 2.5).  24   Fig. 2.3. Average percent of genicular area taken up by primary and secondary cell wall. The remaining percentage (not shown) represents cell lumen and extracellular matrix/middle lamella. Corallinoid species are in green, lithophylloid species are in purple, and metagoniolithoid species are in orange.    Fig. 2.4. Resin-embedded cross-section of Cheilosporum sagittatum geniculum under light microscopy, dyed with 5% potassium permanganate. Arrows indicate location of unidentified layer peeling away from inside the primary cell wall, which may represent a secondary cell wall of distinct chemical composition. Scale bar=2 μm. 25   Fig. 2.5. Cross-sections of genicula showing differences in cell wall structure between the three articulated coralline clades. (A-C) Cross-sections of genicula under light microscopy, dyed with 5% potassium permanganate. Scale bars=10 μm. (D-F) Cross-sections of genicular cells under transmission electron microscopy (TEM). Scale bars=2 μm. (G-I) TEM images of cell wall layers and extracellular matrix between genicular cells in cross-section. Scale bars=500 nm. (A, D, G) Calliarthron tuberculosum. (B, E, H) Amphiroa anceps. (C, F, I) Metagoniolithon stelliferum #1.  Correcting cross-sectional area with approximate cell wall percentage yields breaking stresses that are much closer across species and articulated groups (Fig. 2.6), with the exception of Cheilosporum sagittatum. Cell wall breaking stress (mean ± s.e.m.) was 50.1 ± 4.3 MPa for the corallinoids (but 32.6 ± 1.6 MPa with the exclusion of C. sagittatum), 24.8 ± 1.4 MPa for the lithophylloids, and 15.2 ± 1.2 MPa for the metagoniolithoids. Although cross-sectional area did not account for all of the variation in breaking stress among species (Kruskal–Wallis test, p<0.001), many differences among individual species are lost after accounting for the cell wall (Dunn’s test; Fig. 2.6), so that the previously clear relationship between subfamily and strength is blurred. Cell walls in the corallinoid Johansenia macmillanii are statistically indistinguishable 26  from any of the lithophylloids, as well as Metagoniolithon chara. Cell walls in Lithothrix aspergillum are comparable in strength to all of the corallinoids except for Cheilosporum sagittatum. Some differences are maintained – Metagoniolithon stelliferum #2 is still weaker than all other species tested, with a cell wall stress of 9.1 ± 0.9 MPa. Cell walls in Cheilosporum sagittatum are still the strongest of all articulated species, with a cell wall stress of 102.4 ± 14.2 MPa - over twice that of Corallina officinalis var. chilensis at 40.7 ± 2.4 MPa.  Fig. 2.6. Cell wall material breaking stress of each species. Corallinoid species are in green, lithophylloid species are in purple, and metagoniolithoid species are in orange. Lowercase letters indicate results of a non-parametric post hoc Dunn’s test (p<0.05). Error bars represent s.e.m.  Cell wall proportion and tissue stress do not show any correlation when all species are included in the analysis (ANOVA, p=0.1231). However, after excluding Cheilosporum sagittatum as an outlier, tissue stress increased significantly with cell wall proportion across all other species (ANOVA, p<0.001, R2=0.84; Fig. 2.7). 27   Fig. 2.7. Whole tissue breaking stress in genicula increases with an increase in the percentage of cross-section taken up by the cell wall. Each point represents species averages. Corallinoid species (excluding Cheilosporum sagittatum) are in green, lithophylloid species are in purple, and metagoniolithoid species are in orange. Cheilosporum sagittatum is shown as an outlier in red. The trend line represents the line of best fit through all points excluding C. sagittatum (y=0.45-9.79, R2=0.84). Error bars represent s.e.m.   2.4 Discussion My results support the hypothesis that unique challenges faced by articulated corallines contribute to extraordinary mechanical properties of joint tissue. Genicula were generally tougher, and often stronger and more extensible, than fleshy algal tissues. This is particularly striking given the evolutionary and structural differences of the joints among the three subfamilies. Corallinoids were much stronger and tougher than both fleshy algae and other articulated corallines, as well as much more extensible than fleshy species. Lithophylloid species were stronger and tougher than fleshy algae, and either exceeded or fell at the high end of the range for extensibility in fleshy species. Metagoniolithon stelliferum #1 and Metagoniolithon chara were tougher than fleshy algae, while M. chara was also more extensible than fleshy 28  algae. In all other instances, metagoniolithoid species fell within the ranges of strength/extensibility/toughness for fleshy algae. Tensile toughness is measured as the area under the stress–strain curve, and high breaking stress or high breaking strain can both result in ‘tough’ biological materials. In the case of articulated corallines, both properties appear to play a role – this is most apparent when comparing tissues of lithophylloid and metagoniolithoid species with other red algal tissues. Metagoniolithon stelliferum #1, for example, is neither obviously stronger nor more extensible than fleshy red algal tissues, but moderate performance in both traits results in a comparably high toughness. Toughness in corallinoids is also related to both high stress and strain relative to fleshy algae; however, it is the high strength of this group that pushes its toughness past that of other coralline subfamilies. Although the almost universally high toughness of articulated corallines distinguishes their genicular material from other algal tissues, it is not clear whether this ability to absorb energy is beneficial to survival in wave-swept environments. The amount of energy actually absorbed by an alga in flow is likely negligible compared with the vast kinetic energy available in a given wave (Denny and Gaylord 2002). Furthermore, energy that is absorbed by seaweeds in this way could be released via propagation of cracks through the tissue being loaded, ultimately leading to catastrophic failure (Denny and Hale 2003). Thus, the biological significance of high toughness in algal tissues is unclear and deserves more study. That the metagoniolithoids were weaker than other articulated corallines is perhaps not surprising, given the difference in substrates and tissue composition. First, while the corallinoids and lithophylloids tested in this study were found growing predominantly on rock, all three metagoniolithoid species tested were growing as epiphytes on seagrass (mainly Amphibolus sp.). There are two potential biomechanical consequences of this epiphytic habit: (1) dislodgement is 29  partially dependent on how much force is resisted by the host seagrass, and (2) drag may be lessened by growing epiphytically, because of both the potential ‘drafting’ effect of the host as well as the host’s reconfiguration capabilities (see Anderson and Martone 2014). This means that an epiphytic metagoniolithoid might not require tissue strength as high as an epilithic corallinoid or lithophylloid – indeed, having the capability to withstand more force than that of the host seagrass would be superfluous. It should be noted that the only known epilithic metagoniolithoid species, Metagoniolithon radiatum, was not included in this study because of the failure to procure fresh samples. Data from this species would be valuable to start disentangling the effects of taxonomy and environment in this subfamily. Additionally, the unique biomechanical challenges faced by articulated corallines may not apply to metagoniolithoids. The segmented body plan of articulated corallines can result in amplification of bending stress at the joints, the degree of which is affected by a variety of morphological factors (Martone and Denny 2008a). Shorter joints, as well as joints that are flanked by long calcified ‘lips’ (see Fig. 1.2A), experience more tissue stress. All of the corallinoid species tested possess calcified lips. All corallinoid and lithophylloid species had much shorter joints than any of the metagoniolithoid species (see Table A.1). In contrast, joints up to 6–8 mm in length are seen in Metagoniolithon stelliferum #1. By having long joints that are unhindered by calcified lips, metagoniolithoids may experience drag in a way that is more similar to fleshy algae than it is to other articulated corallines, making such high strength requirements unnecessary. Although all of the corallinoid species tested possess large breaking stresses compared with other articulated groups, Cheilosporum sagittatum was particularly impressive. High material strength in this species may be necessary to offset its slender genicula; with an average cross-sectional area of 0.04 mm2, joints in C. sagittatum were anywhere from 3 30  to 15 times narrower than joints of other species (Appendix A, Table A.1). Across both red and brown algal species, there is a tendency for algae with more slender thalli to be composed of stronger tissues than algae with thicker thalli (Martone 2007a). By increasing the quality, rather than the quantity, of joint tissue, C. sagittatum may withstand forces similar to seaweeds of much larger sizes. This may mean that C. sagittatum is “over-designed” for the drag forces it encounters: frond area affects drag in flow, and it is a diminutive species relative to the others tested. Frond area was not measured in this study, but would be a key factor to consider in future mechanical comparisons. One factor that we were unable to control for in this study was the mechanical history of the specimens tested. Algae in wave-swept environments are subject to constant, repetitive stress that, over time, can lead to breakage at stresses far below the maximum strength of the tissue (Hale 2001, Mach 2009, Mach et al. 2011). This phenomenon, known as fatigue, is due to the accumulation of small imperfections in the tissue that can increase the likelihood of a crack propagating, ultimately leading to tissue failure (Vincent 1990). Although it is impossible to determine the degree to which fatigue played a role for each species in this study, it is likely that some species are more resistant to fatigue than others. For example, the genicula of Calliarthron cheilosporioides are known to be highly resistant to fatigue, because of the loose connection between genicular cells, which minimizes propagation of cracks (Denny et al. 2013). Although other corallinoids have a joint structure similar to that of C. cheilosporioides, genicular cells in lithophylloids and metagoniolithoids appear to be much more adherent to one another, potentially allowing for more energy transfer between adjacent cell walls. Lithophylloid and metagoniolithoid species may be more susceptible to fatigue, thereby breaking at lower stresses 31  that reflect imperfections accumulated during previous wave impacts in the field. Additionally, all non-corallinoid species except for Lithothrix aspergillum had multi-tiered joints – this could increase the number of weak points in the tissue, allowing cracks to propagate around the cells (through the middle lamella) rather than through the cell wall. The combination of differences in cell–cell adherence and tier structure could help explain the comparatively high strength of the corallinoids as a group, though these differences did not correlate with breaking strain. To explore the contributions of cell wall composition and thickness to tissue strength, I corrected breaking stress measurements by the amount of cell wall. As much of the tensile load is likely to be taken up by the cell wall, this essentially calculated the breaking stress of the wall itself. For most species, cell wall quantity appeared to account for much of the difference in strength between groups. That is, articulated corallines appear to strengthen primarily by increasing the amount of cell wall within their tissues (Fig. 2.7). This is a very different strategy from that documented in other algae; kelps increase breaking force by adding cells near the stipe surface to increase girth (Martone 2007b), whereas fleshy red algae add cells to medullary tissue to increase blade thickness (Demes et al. 2011). Corallinoids generally had more cell wall than lithophylloids and metagoniolithoids. However, Cheilosporum sagittatum was a notable exception (Fig. 2.2). Although all other corallinoid species tested had clear secondary cell walls that accounted for roughly half of the cell wall volume – consistent with previous findings in Calliarthron cheilosporioides (Martone 2007b, Martone et al. 2009) – none were immediately visible in the C. sagittatum sections investigated. However, closer inspection of resin-embedded specimens (as opposed to the cryosections used for cell wall measurements) revealed a layer within the primary cell wall that may represent a secondary cell wall (Fig. 2.4). If this is the case, 32  this layer is chemically and mechanically distinct from the secondary walls present in other corallinoids – not only did it not stain with potassium permanganate, indicating a chemical composition differing from that of the primary wall, it also appeared to pull away from the primary wall in some cells. Ultimately, cell wall strength of C. sagittatum was even greater than that of other articulated corallines, suggesting that the cell walls in C. sagittatum may be doing something unique at the chemical level. Remaining differences in strength, after accounting for the amount of cell wall, may be due to differences in the types and quantities of different polysaccharides within the wall. Cell walls in most red algae are characterized by skeletal polysaccharides such as cellulose, as well as an amorphous matrix composed mostly of sulfated galactans (Frei and Preston 1961, Usov 1992, Tsekos 1999, Vreeland and Kloareg 2000). In land plants, variation in the proportion of cellulose to matrix has been found to affect tensile strength (Girault et al. 1997, Genet et al. 2005). Furthermore, angle of the cellulose microfibrils may affect stiffness, as less steeply angled cellulose will take more time to reorient in the direction of the applied force (Koehl and Wainwright 1977, Kohler and Spatz 2002). Strength, extensibility, and stiffness may also depend on the type of sulfated galactans produced by different life stages of red algae (Carrington et al. 2001). The high material strength of corallinoids, in particular Cheilosporum sagittatum, could be due to either high levels of cellulose relative to other corallines, or a unique set of matrix polysaccharides linking the cellulose together. Articulated corallines represent an interesting example of convergent evolution, in which multiple calcified algal groups have come to the same general solution for mitigating drag: growing upright thalli that are flexible via segmentation. Given the mechanical challenges inherent in a jointed morphology, articulated corallines have converged on a similar set of 33  mechanical properties. Coralline joints are generally stronger and tougher than tissues of fleshy algae, while maintaining high strains comparable to fleshy algae. Tensile stiffness is highly variable among corallines. Differences in the cellular structure of joints, such as cell-to-cell adherence and the number of cell tiers, likely contribute to the slight remaining differences in mechanical behaviour between subfamilies. Data suggest that articulated corallines universally strengthen joints by augmenting the quantity of cell wall, with remaining differences in strength pointing to a potential contribution of cell wall composition. This is particularly evident in the unusual strength and toughness of the corallinoid Cheilosporum sagittatum, which warrants further investigation. 34  Chapter 3: Morphology and bending strategies of articulated coralline algae  3.1 Introduction Corallines occupy a diverse array of habitats, but they are particularly successful in the intertidal and shallow subtidal of wave-swept shores, where flexibility is considered a prerequisite for the survival of upright macroalgae (Harder et al. 2004). The importance of flexibility may have been a contributing factor to the evolution of joints in upright corallines, which has occurred at least three separate times as supported by both molecular (Bailey and Chapman 1998, Bittner et al. 2011, Kato et al. 2011) and fossil evidence (Aguirre et al. 2010, Kundal 2011). Representatives of all three articulated coralline clades - Corallinoideae, Metagoniolithoideae, and Lithophylloideae - can be found in abundance in wave-swept environments. The structure and development of joints in the three lineages is not homologous (see Johansen 1981); furthermore, while joints in articulated corallines universally aid in flexibility, we know little about how the degree of flexibility and the depth of convergence compare. While the basic jointed morphology is analogous, the structural and developmental details are not, so what about the function?  3.1.1 Flexibility has consequences For macroalgae living in wave-exposed environments, flexibility is an important drag-mitigating mechanism (Harder et al. 2004). Flexible macroalgae can bend towards the substrate, lowering the area projected into flow (Vogel 1984, Denny 1985, Martone and Denny 2008a). They may also fold up blades or branches, reconfiguring passively into more streamlined shapes (Denny and Gaylord 2002, Boller and Carrington 2006, Martone et al. 2012). These shifts in size 35  and shape reduce the amount of drag that is experienced, thereby reducing the chance of dislodgement.  While the benefits of flexibility for fleshy macroalgae have been well documented (Vogel 1984, Denny and Gaylord 2002, Harder et al. 2006, Starko et al. 2015), flexibility in articulated corallines comes with unique challenges that make the benefits of flexibility to survival less apparent. By limiting bending to discrete joints, stress in joint tissue is amplified, and this stress increases as bending increases (Martone and Denny 2008a).  Computational modelling of the articulated corallinoid species, Calliarthron cheilosporioides, revealed that morphological trade-offs exist between increasing flexibility and decreasing stress; for example, decreasing joint width might increase flexibility of the overall thallus, but it also decreases the tissue available to withstand stress (Martone and Denny 2008a). For this species, genicular morphology is somewhat balanced between stress mitigation and flexibility maximization (Martone and Denny 2008a). An improved Calliarthron geniculum would have longer genicula than those observed; this may be the result of some developmental constraint in corallinoids, as genicular cells lose cytoplasm and organelles as they lengthen and may have some upper limit to how far they can extend (Martone and Denny 2008a). Corallinoid joints are composed of only a single tier of cells, so limitation on individual cell length translates to a limitation on total genicular length (Johansen 1969, 1981, Martone and Denny 2008a).  Single-tiered cell structure is specific to corallinoid genicula – articulated corallines in other groups may be able to lengthen their joints further by adding additional cell layers. While the model does not answer the question of why an articulated coralline possesses a certain morphology, it can be used to investigate the consequences of that morphology. Modelling is thus an ideal tool for exploring the degree to which the genicula of different articulated coralline clades are functionally convergent. 36  3.1.2 Articulated corallines are not analogous at all levels of organization The development of genicula progresses differently in all three articulated coralline clades; corallinoid and lithophylloid genicula form via decalcification of medullary cells and a cracking of the surrounding cortex (Johansen 1969, 1981), whereas genicula in Metagoniolithoideae form when meristematic cells switch production from calcified to uncalcified cells (Ducker 1979). Additionally, while corallinoid joints are always single-tiered and anchored in intergenicular tissue at either end (Johansen 1969, 1981, Martone and Denny 2008), lithophylloid joints are typically multi-tiered (Johansen 1969, Johansen 1981), and metagoniolithoid joints are composed of many layers of irregularly shaped cells that lack tiered organization altogether (Ducker 1979). The cells of corallinoid genicula are unique in being very loosely connected laterally, which causes whole joints to behave more like a rope than a solid when bending (Martone and Denny 2008a). Lack of attachment between neighboring cells also mitigates the propagation of cracks throughout the tissue, causing corallinoid genicula to be highly resistant to fatigue from repeated stresses (Denny et al. 2013). Lithophylloid and metagoniolithoid joint tissues, on the other hand, are much more cohesive in structure. Structural and developmental constraints may affect the degree to which each articulated clade can achieve a mechanically balanced morphology, or the morphological strategy used to do so.  There are two main morphological ways in which a frond can achieve flexibility without sacrificing the capability to withstand stress – a frond can increase the number of joints or increase the length of individual joints. Both options result in a similar level of uncalcified tissue per unit length. Having thin joints can increase flexibility, but this also comes with an increased likelihood of breakage. A fourth strategy to mitigate mechanical challenges is adjusting material properties of genicular tissue - corallinoids, lithophylloids, and metagoniolithoids all possess 37  joints with high material strength and extensibility relative to other red algal groups (Ch. 2), which may help them resist stress amplification in their genicula. However, because there are significant differences in material properties among the three articulated groups (Ch. 2), the role of material properties in stress mitigation is uncertain. In this study, I aimed to determine whether articulated corallines universally mitigate or resist bending stress in the same way, despite having convergently-evolved bending structures. Given differences in structure and development, I hypothesize that articulated lineages may employ different strategies to achieve similar levels of flexibility. Using a computational model developed by Martone and Denny (2008a), I explore the morphological trade-off between resisting stress and maintaining flexibility, as well as the consequences associated with different bending strategies.  3.2 Materials and methods  3.2.1 Specimen collection Corallinoid species Calliarthron tuberculosum, Corallina officinalis var. chilensis, and Johansenia macmillanii were collected subtidally at a depth of approximately 3 m from Botanical Beach (48°31′48″N, 124°27′18″W) on Vancouver Island, BC, Canada, in June/July 2012. Lithophylloid species Amphiroa anceps and Amphiroa gracilis were collected in Point Peron (32°16′01″ S, 115°41′14″ E), Perth, Western Australia at depths of 3.0–4.6 m in December 2012.  Both Botanical Beach and Point Peron are located on open coasts unsheltered by any surrounding land features, and are generally considered “wave-exposed” locations. Metagoniolithon stelliferum #1 and #2 are used here to denote two morphotypes that 38  currently fall under the name Metagoniolithon stelliferum. The two morphotypes are genetically distinct from one another at a species level, based on sequencing of psbA, CO1 and rbcL genes (K.G.J., unpublished data). Both metagoniolithoid species were found growing epiphytically on Amphibolis seagrass in the same bed, with no obvious habitat delineation between them. They were collected from Point Peron at 3.0-4.6 m in December 2012.  Specimens were kept in cooled seawater in the lab for no more than 48 hours prior to mechanical testing. Representative individuals of each species were deposited at the University of British Columbia Herbarium for future reference: Calliarthron tuberculosum (A91564), Corallina officinalis var. chilensis (A91563, as Corallina officinalis), Johansenia macmillanii (A91561), Amphiroa gracilis (A91572), Amphiroa anceps (A91566), Metagoniolithon stelliferum #1 (A91576, as Metagoniolithon stelliferum), and Metagoniolithon stelliferum #2 (A91579, as Metagoniolithon sp.).  3.2.2 Flexibility Bending tests followed the protocol outlined by Martone and Denny (2008a). Basal segments just over 2 cm in length were trimmed of all branches; for dichotomously branching species, one side of each dichotomy was removed to produce a straight, unbranched segment. Trimmed segments were mounted horizontally in a vice clamp (Fig. 3.1), with the first 1-3 joints held stationary within the grips. The insides of the grips were lined with neoprene and sandpaper to prevent crushing or slippage of the specimen. Dental floss was looped and tied around the segment at a point ~2 cm distal to the anchored end, and fixed in place with cyanoacrylate glue. Weights of 5, 20, and 100 g (corresponding to 0.05, 0.20 and 0.98 N of force) were hung from the end of the floss. Chilled seawater was pipetted onto bent segments between each load. 39  Fifteen specimens per species were tested.  Fig. 3.1. Illustration of the method used for measuring frond flexibility.  Photos were taken at each load from the side (see Fig. 3.1) using a digital camera, and flexibility was determined by analyzing photos in ImageJ (NIH Image, http://rsb.info.nih.gov/ij).  Flexibility was quantified by measuring the “percent bent” of the original segment length. As bending occurs, the height projected into the hypothetical flow decreases, so percent bent was measured as: percent bent = �1 − �LexposedLinitial �� ∗ 100 Eq. 3.1  where, Lexposed is the length projected horizontally during bending, and Linitial is the starting segment length (see Fig. 3.1).  After visualizing the data, I focused on the percent bent at 0.05 N only, as larger forces 40  only caused individuals to converge upon 100%. Due to unequal variances between species, statistical comparisons were done with a non-parametric Kruskal-Wallis test and post hoc Dunn’s test. Both tests were performed in R 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria) using the RStudio interface (version 0.98.1056, RStudio, Boston, MA, USA) and the dunn.test() function from the dunn.test package (Dunn’s test of multiple comparisons using rank sums, version 1.2.3, Alexis Dinno 2015).  3.2.3 Morphometrics The following genicular and intergenicular dimensions were measured for each bent specimen: short and long genicular radii, genicular length, short intergenicular radius, intergenicular length, and intergenicular lip length (corallinoids only).  Corallinoid species in this study possess intergenicular lips that overlap genicula on either end (see Fig. 3.2A inset); this meant that corallinoid genicula had to be long-sectioned to obtain an accurate measurement of genicular length, precluding those same genicula from being cross-sectioned for measurements of genicular radii. For these species, measurements of each geniculum switched between measuring genicular length or radii starting from the anchored end, and missing values were replaced with an average of the values from adjacent genicula. For species with no lips (lithophylloids and metagoniolithoids), every joint that underwent bending was measured for all parameters.  In cases where genicula or intergenicula appeared to individually grow wider from the proximal to the distal end (e.g. Amphiroa anceps, Fig. 3.2E), an average of the two ends was taken. Measurements were performed under a dissection microscope (model SZ61, Olympus Canada, ON, Canada) with an attached camera (model DP20, Olympus Canada, ON, Canada). Averages of all morphometrics were calculated for each specimen. 41   Fig. 3.2. Close-up of genicula and intergenicula for each species studied. (A, B, C) Corallinoid species, Calliarthron tuberculosum, Corallina officinalis var. chilensis, Johansenia macmillanii. (D, E) Lithophylloid species, Amphiroa gracilis, Amphiroa anceps. (F, G) Metagoniolithoid species, Metagoniolithon stelliferum #1, Metagoniolithon stelliferum #2. Scale bars=0.5 mm.  Metagoniolithon stelliferum #1 has a joint morphology that required extra consideration in measuring intergenicular parameters (see Fig. 3.2F). This species possesses unusually long meristematic genicula, with false whorls of branches that originate from the midpoint (Ducker 1979). Branching points are much thicker than the surrounding genicular tissue, and did not appear to bend during tests despite being non-calcified. I thus chose to group them with intergenicula when measuring morphometrics - “intergenicular” radius and length values presented in this study for Metagoniolithon stelliferum #1 represent averages of both intergenicula and branching points, and should not be considered representative of the true 42  morphology of this species, except when modelling bending behaviour.  To compare similarly flexible individuals morphometrically, two ranges of flexibility were chosen: 50-70% bending and 70-90% bending at 0.05 N. Individuals that fell within these ranges were averaged by species, and all morphometric parameters were compared. I also calculated and compared intergenicular length/genicular length to look at relative levels of calcification. Statistical analysis was not applied to these qualitative comparisons, as some species had too few individuals within a given range to make statistical analysis informative.  3.2.4 Modelling Bending of different species was modelled in Octave 4.2.1 (Eaton et al. 2016, https://www.gnu.org/software/octave/) using code adapted from Martone and Denny (2008a) to explore the consequences of using different genicular traits to achieve flexibility. Like Martone and Denny (2008a), I modelled corallinoid genicula as cables; however, lithophylloid and metagoniolithoid genicula were modelled as solid beams. For thorough derivation of mathematical details, see Appendix A of Martone and Denny (2008a). For a summary of the distinction between the cable and solid models, see Appendix B and Fig. B.1 of this thesis.   3.2.5 Bending stress The model calculated total maximum stress in the first geniculum, with both tensile and bending stress components. The mathematical details of this are outlined in Appendix B of Martone and Denny (2008a). The morphometric parameters of fronds used in real bending tests were input into the model to estimate stress and to explore other mechanical consequences of morphology beyond flexibility. To separate the effects of morphology from tissue properties, all 43  species were modelled using an “average coralline” tissue stiffness, which was calculated as the mean of species tissue stiffness values taken from Ch. 2. Corallinoid species were again modelled using a cable model, while lithophylloid and metagoniolithoid species were modelled using a solid model. Species averages of estimated stress in bending were then compared with species averages of breaking strength obtained from Ch. 2.  3.2.6 Mechanical consequences of morphology The model was used to explore the individual effects of different morphometric parameters on flexibility and stress. Mean values for genicular and intergenicular length and width of each species were calculated from measurements taken during bending tests. These values were used to virtually build an “average” 2 cm tall frond of each species – genicula in a single frond were assumed to have constant morphology from base to tip. Virtual frond data were input into the model and tested at a force of 0.05 N, and both percent bent (Eq. 3.1) and stress in the first geniculum (MPa) were quantified. Each morphometric parameter was then varied while holding all other values constant, with flexibility and stress being recorded for each trial. Genicular radii were varied together to explore the overall effect of joint width. The amount to which each parameter could be varied differed depending on other morphological constraints, which in turn depended on the species being modelled. Genicula were never modelled as being wider than surrounding intergenicula, except in the case of Metagoniolithon stelliferum #2, a species in which this occurs naturally (Fig. 3.2G). All fronds were held at a 2 cm length as joint morphology varied, so variation of intergenicular length was limited to different extents in different species, due to the need to maintain at least one geniculum within the modelled frond. As with the morphometric comparisons, non-bending branching points of 44  Metagoniolithon stelliferum #2 were grouped with intergenicula. The percent change of each genicular/intergenicular dimension was graphed against both total stress in the first geniculum (MPa) and percent bent – genicular dimensions were considered “balanced” when low stress and high flexibility coincided.   3.3 Results  3.3.1 Flexibility Average flexibility (as measured by percent bending at 0.05 N) depended significantly on clade (Kruskal-Wallis test followed by post hoc Dunn’s test, H(6)=75.55, p<0.001, Fig. 3.3A). Metagoniolithoids generally bent more than lithophylloids, which bent more than corallinoids for a given force. Metagoniolithon stelliferum #1 was more flexible than all other species, with an average percent bending (mean ± s.e.m.) of 94.4 ± 1.2%. Metagoniolithon stelliferum #2 overlapped with Metagoniolithon stelliferum #1 and both lithophylloid species, bending 85.8 ± 1.1%. Amphiroa anceps and Amphiroa gracilis bent 76.2 ± 4.1% and 73.0 ± 3.6%, respectively. Johansenia macmillanii bent 44.1 ± 3.1%, Corallina officinalis bent 53.3 ± 2.9%, and Calliarthron tuberculosum bent 40.1 ± 4.9%.  45   Fig. 3.3. (A) Flexibility ranges of articulated coralline species, measured as the percent bent with the application of 0.05 N of force. Each point represents one individual. Corallinoid species are represented by green triangles, lithophylloid species by purple triangles, and metagoniolithoid species by orange triangles. Bold lowercase letters indicate the results of a non-parametric Kruskal-Wallis test and post hoc Dunn’s test (p<0.001). (B, C) Average morphometrics of corallines that bent between 50-70% (B) or 70-90% (C) of their initial length with the application of 0.05 N of force. Numbers above each column in subpanels B.I. and C.I. indicate the number of individuals measured for each species within a given flexibility range. Error bars represent s.e.m. 46   3.3.2 Morphometrics Specimens that bent 50-70% exhibited different morphological characteristics, depending on subfamily. All three corallinoid species and both lithophylloid species had individuals that bent 50-70% with an application of 0.05 N. No clear pattern was observed in either intergenicular or genicular radii (Fig. 3.3B.I-II). Both intergenicular and genicular lengths were higher in lithophylloid species than in corallinoid species (Fig. 3.3B.III-IV). Intergenicular length/genicular length was also higher in lithophylloids than corallinoids (Fig. 3.3B.V), indicating more calcified tissue per unit length. Almost all species in this study had at least one individual that bent within the 70-90%, except Johansenia macmillanii. Again, strategies for achieving this level of bending appeared to differ depending on subfamily. No clear pattern was observed in intergenicular or genicular radii (Fig. 3.3C). Lithophylloids once again had the longest intergenicula (Fig. 3.3C.III), whereas both metagoniolithoid species had notably long genicula (Fig. 3.3C.IV). Both metagoniolithoids had a very low level of calcification per unit length (Fig. 3.3C.V).  3.3.3 Bending stress Modelled total stress in the first geniculum at 0.05 N was highest in the corallinoids (Fig. 3.4): it was predicted Calliarthron tuberculosum would experience 20.3 ± 4.5MPa of stress (mean ± s.e.m.), Corallina officinalis var. chilensis would experience 16.6 ± 3.8MPa, and Johansenia macmillanii would experience 11.0 ± 3.4MPa. Modelled total stress was higher in lithophylloids than metagoniolithoids, with Amphiroa gracilis predicted to experience 7.03 ± 0.5MPa and Amphiroa anceps predicted to experience 10.7 ± 2.4MPa. The model suggests 47  metagoniolithoids experience the least amount of stress, with Metagoniolithon stelliferum #1 predicted to experience 1.2 ± 0.1MPa and Metagoniolithon stelliferum #2 predicted to experience 3.5 ± 0.4MPa. Species average stress in bending was positively correlated with species average tissue strength (y=1.3x+2.2, R2=0.71, Fig. 3.4). In almost all cases, tissue strength was greater than modelled stress: Metagoniolithon stelliferum #2 was the one exception, with a tissue strength of 2.8 ± 0.3MPa, 0.7MPa lower than the stress it was expected to experience (Fig. 3.4).  Fig. 3.4. The relationship between the material stress experienced in bending and the material strength of genicular tissue in articulated coralline species. Points represent species averages (n=15), and error bars represent s.e.m. Corallinoid species are shown in shades of green, lithophylloid species are shown in shades of purple, and metagoniolithoids are shown in shades of orange. The black dotted line represents a line of best fit (y=1.3x+2.2, R2=0.71), and the red dashed line represents a 1:1 line.   3.3.4 Mechanical consequence of morphology Intergenicular radius appeared to be close to balanced for most species, in that a decrease or increase would either increase stress or decrease flexibility with minimal benefit in the other 48  parameter (Fig. 3.5). Calliarthron tuberculosum (Fig. 3.5A) was an exception, and could drastically decrease stress with a minimal decrease in flexibility by decreasing intergenicular radius even slightly. According to the model, Amphiroa anceps (Fig. 3.5E) could increase flexibility slightly by increasing intergenicular radius, but this would also come with an increase in stress.  The consequences of varying genicular radius were variable between species (Fig. 3.6). Corallina officinalis var. chilensis (Fig. 3.6B), Amphiroa anceps (Fig. 3.6E), and Metagoniolithon stelliferum #2 would increase in flexibility with almost no effect on stress by possessing a slightly smaller genicular radius. Calliarthron tuberculosum (Fig. 3.6A) and Johansenia macmillanii (Fig. 3.6C) could decrease stress by increasing genicular radii, although the associated decrease in flexibility would be equally significant. Amphiroa gracilis (Fig. 3.6D) and Metagoniolithon stelliferum #1 (Fig. 3.6F) were both largely unaffected by minor shifts in genicular radii, although a large decrease would result in a slight increase in stress.  Calliarthron tuberculosum would lower stress while maintaining high flexibility by possessing a slightly lower intergenicular length (Fig. 3.7A). All other species appeared to have balanced intergenicular lengths (Fig. 3.7, B-G): in the case of the lithophylloid and metagoniolithoid species, shifting length positively or negatively within the range tested would have minimal mechanical effect. Calliarthron tuberculosum (Fig. 3.8A), Johansenia macmillanii (Fig. 3.8C), Amphiroa anceps (Fig. 3.8E), and Metagoniolithon stelliferum #2 would all increase flexibility and decrease stress by increasing genicular length. Corallina officinalis var. chilensis (Fig. 3.8B), Amphiroa gracilis (Fig. 3.8D), and Metagoniolithon stelliferum #1 (Fig. 3.8F) had sufficiently long genicula to both maximize flexibility and minimize stress, with little to be gained from a further increase in genicular length. 49    Fig. 3.5. The effect of varying intergenicular radius on the maximum stress experienced in the first geniculum (black triangles) and percent bent (grey circles) with the application of 0.05 N. Lines of no change (dotted red lines) indicate stress and flexibility for an “average” individual of a given species. Typical joint construction of each articulated clade is illustrated in the bottom right – solid red bars indicate where intergenicular radius would be measured.  50   Fig. 3.6. The effect of varying genicular radius on the maximum stress experienced in the first geniculum (black triangles) and percent bent (grey circles) with the application of 0.05 N. Lines of no change (dotted red lines) indicate stress and flexibility for an “average” individual of a given species. Dotted line breaks in panels B and C indicate regions where a data point was skipped, due to a mathematical anomaly in the MatLab model. Typical joint construction of each articulated clade is illustrated in the bottom right – solid red bars indicate where genicular radius would be measured.       51   Fig. 3.7. The effect of varying intergenicular length on the maximum stress experienced in the first geniculum (black triangles) and percent bent (grey circles) with the application of 0.05 N. Lines of no change (dotted red lines) indicate stress and flexibility for an “average” individual of a given species. Typical joint construction of each articulated clade is illustrated in the bottom right – solid red bars indicate where intergenicular length would be measured. 52   Fig. 3.8. The effect of varying genicular length on the maximum stress experienced in the first geniculum (black triangles) and percent bent (grey circles) with the application of 0.05 N. Lines of no change (dotted red lines) indicate stress and flexibility for an “average” individual of a given species. Typical joint construction of each articulated clade is illustrated in the bottom right – solid red bars indicate where genicular length would be measured.        53  3.4 Discussion Articulated corallines may have convergently evolved to possess genicula, but not without variation on the overall theme. Though species differed in average levels of bending, there was sufficient overlap to allow for morphological comparisons of similarly flexible individuals. From these comparisons a pattern emerges, where each subfamily utilizes a different strategy to achieve flexibility. Both corallinoids and metagoniolithoids have joint morphologies that minimize the amount of calcification per unit length, but in different ways. Corallinoids have many genicula, as evidenced by the fact that a low calcification level is maintained despite genicula being relatively short. Metagoniolithoids, in contrast, have a smaller number of longer genicula. The lithophylloid strategy is not immediately apparent based on morphology alone – while lithophylloid genicula were longer than corallinoid genicula, so too were lithophylloid intergenicula, resulting in a high proportion of calcified tissue per unit length compared to both corallinoids and metagoniolithoids. It is here that material properties become relevant; average genicular tissue stiffness is 11 MPa in Amphiroa gracilis and 12 MPa in Amphiroa anceps, less than half the stiffness of any corallinoid species tested (29-52 MPa; Ch. 2). This allows lithophylloids to maintain a higher level of calcification than other groups without sacrificing the capability to bend in flow.  Differences in bending strategy may reflect different developmental and structural constraints. As previously noted, genicula in corallinoids likely have an upper limit to how long they can grow, due to being composed of a single tier of elongated cells with minimal cellular content. This means that increasing the number of genicula may be the only option corallinoids have available to increase the overall proportion of genicular tissue. The multizonal structure of lithophylloid and metagoniolithoid genicula make limitations on individual cell length irrelevant 54  to genicular length. In addition, metagoniolithoid genicula are meristematic and often continue to grow throughout the life of the frond (Ducker 1979), so it is perhaps not surprising that the metagoniolithoid species tested here achieved flexibility primarily by having long genicula. The reason behind the low stiffness of lithophylloid genicula is unclear; it may be that morphological constraints necessitate low stiffness to allow for bending, or it may be that low stiffness allows for a more calcified morphology. Morphological and mechanical tests on a greater number of lithophylloid species might help clarify this relationship. Comparisons of expected bending stress also revealed a relationship between morphology and material properties. Species that are expected to incur the most stress in bending due to morphology are also composed of stronger tissues that can withstand that stress. This is at least partially due to secondary cell wall growth in the corallinoids, which fortifies the primary cell wall and increases the quantity of material resisting stress (Martone 2007b, Ch. 2). The chemical composition of genicular cell walls in the three articulated clades is also likely to play a role, but only Calliarthron cheilosporioides joints have been explored for this thus far (see Martone et al. 2009, Martone et al. 2010). As with the properties of lithophylloid genicula, a “chicken or egg” question arises; are corallinoid genicula strong because they cannot grow longer to minimize bending stress, or does joint morphology represent a “bare minimum” of joint formation that is possible because of genicular strength? Shifting one genicular parameter at a time using modelling revealed that corallinoids are not necessarily well balanced between flexibility and stress. Calliarthron tuberculosum specifically could benefit from minor changes in morphology – indeed, the average modelled frond for this species appears to be right on the edge of drastically reduced stress for every morphometric investigated. Johansenia macmillanii is also on the detrimental side of a tipping 55  point for genicular radii, intergenicular length, and genicular length. In contrast, the average Corallina officinalis consistently minimizes bending stress, which is notable given that it is also composed of the strongest material. It is possible that Corallina officinalis is “overdesigned” and could survive with weaker tissue – however, no part of this analysis considered frond size, density, or habitat, which could impact the amount of drag experienced in the field. Although all species in this study were collected from wave-swept environments, if Corallina officinalis lives in a particularly wave-exposed microhabitat or assumes a more drag-prone morphology, then it may experience more drag than other species nearby and may need stronger tissues to survive. Modelled lithophylloid and metagoniolithoid species consistently maintained high flexibility and low stress, although in some cases small shifts in genicular morphology could improve flexibility even farther without a detrimental effect on stress. Perhaps maximal flexibility is not necessarily the most ideal situation for all upright corallines, as excess flexibility might cause coralline thalli to flop over when it is not mechanically necessary, i.e. when wave action is minimal. This would directly contradict many of the benefits associated with upright growth, such as increased area available for photosynthesis and nutrient uptake, or avoidance of herbivores (Lubchenco and Cubit 1980, Padilla 1984). Given that each articulated clade exhibited a different flexibility range in bending tests, it is unlikely that the optimal level of flexibility is the same for all species or environments.  Two of the three corallinoid species could theoretically reduce stress by limiting the proportion of calcified tissue present, either by shortening intergenicula (Fig. 3.7) or lengthening genicula (Fig. 3.8) – this suggests some non-mechanical benefit to heightened calcification levels. That is, if decreasing calcified tissue could generally increase flexibility and decrease stress, then why remain calcified? While calcification has typically been considered a deterrent 56  against herbivory (Littler and Littler 1980, Steneck and Watling 1982, Pennings and Paul 1992), this is largely dependent on the type of herbivore and its associated feeding apparatus (Padilla 1989, Maneveldt and Keats 2008). Indeed, some degree of herbivory may help corallines thrive, as it facilitates the removal of epiphytes and fouling organisms from the epithallial surface (Steneck 1982, Littler et al. 1995, Berthelsen and Taylor 2014). Although calcification may not completely prevent herbivory, it could provide just enough resistance to limit the depth of grazing that occurs or to facilitate rapid recovery (Steneck et al. 1991). Calcification also provides protection for reproductive propagules, which are housed in cavities called “conceptacles” that reside exclusively on the calcified intergenicula (Johansen 1981), so increased intergenicular tissue could also correspond with an increased surface area available for reproductive activity. Most of the studies done on the benefits of calcification in corallines have been performed on crustose forms, and its adaptive significance in articulated corallines deserves further study.   3.5 Conclusions Genicula in articulated corallines likely evolved convergently from similar environmental pressures, and this has led to a variety of morphological forms. Different species utilize different strategies in response to the mechanical challenges inherent in being articulated, even when comparing within a set range of flexibility. Determining whether these strategies are consistent within each articulated clade requires further testing on additional species, particularly within the lithophylloids and metagoniolithoids. Morphological types are not equal in balancing flexibility with bending stress, raising additional questions about the developmental constraints imposed on genicula, the need for flexibility, and the benefits of calcification. Biomechanical challenges 57  caused by morphology may be partially mitigated with material properties – the stressful morphology of corallinoids is offset by strong tissues, while the inflexible morphology of lithophylloids is offset by pliant tissues. Structural and computational biomechanics can help clarify at which level of organization convergent evolution has occurred in ostensibly similar structures. 58  Chapter 4: Chemical composition of coralline cell walls  4.1 Introduction Articulated corallines from the subfamilies Corallinoideae, Lithophylloideae, and Metagoniolithoideae all have genicula that perform the same function – that is, they allow otherwise rigid thalli to retain flexibility under hydrodynamic stress. These key structures do, however, differ in several notable ways. Development is different between the three groups; genicular cells in both corallinoids and lithophylloids start out calcified and decalcify secondarily (Johansen 1969, 1981), while metagoniolithoid genicula are the result of a shift in production at the meristem from calcified to uncalcified cells (Ducker 1979). They also differ in their material properties; while genicula from all three clades generally outperform other fleshy algal tissues under tensile stress and strain, corallinoid genicula are stronger and tougher than genicula in other groups, and some lithophylloid genicula are particularly extensible (Ch. 2). The identity and amount of polysaccharides within the cell wall have been linked to both the mechanical performance of algal tissues in general (Kloareg and Quatrano 1988, Stockton et al. 1980, Kraemer and Chapman 1991, Carrington et al. 2001, Starko et al. 2018) and the deposition of calcium carbonate in corallines specifically (Borowitzka and Vesk 1978, 1979, Okazaki et al. 1982, Bilan and Usov 2001, Martone et al. 2010, Carvalho et al. 2017). Given the link between calcification and chemistry, it is expected that cell walls of intergenicula and genicula are distinct from one another, and differences in histological staining (Yendo 1904, Johansen 1969) and some analytical chemistry (Martone et al. 2010) indicate that this is the case. However, we do not know if a similar chemical shift occurs during genicular development across all articulated coralline groups. Articulated corallines all have a similar evolutionary trajectory, 59  having evolved from prostrate crustose ancestors (Bailey and Chapman 1998, Aguirre et al. 2010, Bittner et al. 2011, Kato et al. 2011), and this could hypothetically constrain the chemical building blocks found in genicular tissues. Alternatively, differences in the development and material properties of genicular tissue suggest that there are underlying chemical differences. As in terrestrial plants, cell walls in red algae are composed of both a crystalline, skeletal component and a surrounding amorphous matrix (Frei and Preston 1961, Kloareg and Quatrano 1988, Carvalho et al. 2017). Unlike terrestrial plant cell walls, however, red algal cell walls are dominated by matrix polysaccharides with the skeletal component typically comprising <20% dry weight of the whole tissue (Frei and Preston 1961, Kloareg and Quatrano 1988). The main skeletal polysaccharide is typically cellulose, although xylans and mannans are found instead in some members of the Bangiales (Frei and Preston 1961, Kloareg and Quatrano 1988). While “true” cellulose (i.e. what is found in terrestrial plants) is composed of β-1,4 glucose units, celluloses found in red algae are likely to contain other sugars integrated into the polysaccharide backbone, such as β-1,4 xylose (Cronshaw et al. 1958, Turvey and Williams 1970, Kloareg and Quatrano 1988).  Red algal celluloses are assembled into microfibrils just as they are in terrestrial plants, although the cross-sectional shape and width of the microfibrils varies by species (Kloareg and Quatrano 1988, Tsekos et al. 1996, Tsekos 1999, Niklas 2004, Nishiyama 2009).  The matrix of red algae is composed mostly of sulfated galactans, a family of polymers with a backbone made up of alternating β-1,3 and α-1,4 galactose units (Frei and Preston 1961, Kloareg and Quatrano 1988, Usov 1992, Tsekos 1999, Vreeland and Kloareg 2000, Navarro and Stortz 2002, 2008). While the β galactose units are always in the D configuration, the α-galactose units may be in the D configuration (carrageenan) or the L configuration (agar); the α-galactose 60  may also occur as a 3,6 anhydro derivative (carrageenose or agarose), so that a total of four major galactan groups exists (Kloareg and Quatrano 1988, Bilan and Usov 2001, Lahaye 2001, Navarro and Stortz 2002, 2008).  There is a high degree of variation within these groups, as the regular backbone of these polysaccharides may have different combinations of O-linked groups such as sulfate esters, methyl ethers, or xylose and galactose side chains (Kloareg and Quatrano 1988, Usov 1992, Bilan and Usov 2001, Navarro and Stortz 2002, 2008). The various combinations of these side groups are what lead to a range of gelling properties in the galactans of different red algal species, which in turn may affect their mechanical and physiological properties (Usov 1992, Kloareg and Quatrano 1988). Multiple studies on different species of both articulated and crustose corallines have revealed the existence of a group of galactans that are unique to corallines, dubbed “corallinans” by Cases et al. (1992, 1994). This group was first discovered in Corallina officinalis by Turvey and Simpson (1966), and later found to be present in several other species (Usov et al. 1995, Takano et al. 1996, Usov and Bilan 1998, Navarro and Stortz 2002, Navarro et al. 2011).  Classified as agarans due to the L-configuration of the α-galactose units, corallinans have varying levels of O-methylation at the 2 and 3 positions of the α-L galactose units, as well as on position 6 of the β-D galactose units (Cases et al. 1992, 1994). Sulfate groups may be present on positions 2 and 3 of the α-L galactose units and position 6 of the β-D galactose units, and 4-O-methyl-D-galactosyl units may be present at position 6 of the β-D galactose units (Cases et al. 1994, Stortz et al. 1997).  Finally, corallinans can possess xylosyl sidechains at the C-6 position of the β-D galactose units, with a total Xyl:Gal ratio of approximately 1:3 (Cases et al. 1992) – for this reason, corallinans are also referred to as “xylogalactans”.   61  A major limitation of the work done on coralline cell wall chemistry has been the failure to analyze tissue types separately. While this is not an issue when studying crustose species (e.g. Lithothamnion heterocladum, Navarro et al. 2011), it is likely that the genicular and intergenicular tissue of articulated corallines differ in their chemical composition. For example, Martone et al. (2010) compared the structure of galactans found in genicula and intergenicula of Calliarthron cheilosporioides and found that the levels of O-methylation and xylosyl substitution differed between the two tissues. Galactans found in the intergenicula had a typical xylogalactan structure, with low levels of O-methylation and significant xylosyl sidechains on C-6 of the β-D galactose units. Galactans found in the genicula were the inverse, with high O-methylation and only minor amounts of xylosyl side units on C-6 of the β-D galactose units. Martone et al. (2010) suggested that the xylosyl sidechains present in the intergenicular galactans might facilitate nucleation of calcium carbonate, which may explain the abundance of xylogalactans in corallines in general. In addition to possessing galactans with a unique structure, coralline cell walls are set apart from those of other red algae by the presence of alginates (Okazaki et al. 1982, Okazaki and Tazawa 1989, Usov et al. 1995, Bilan and Usov 2001, Navarro and Stortz 2002, 2008, Navarro et al 2011). Alginates are a family of polysaccharides composed of 1,4 linked β-D-mannuronic acid (M) and α-L-guluronic acid (G), in varying M:G ratios (Kloareg and Quatrano 1988, Bilan and Usov 2001). They are more typically associated with brown algae and are not known to occur in red algae outside of the Corallinaceae. Alginic acid can bind to calcium ions, and so it has been suggested to play a role in the calcification process of corallines (Okazaki et al. 1982, Okazaki and Tazawa 1989). Alginic acid levels are typically very low in corallines, 62  however, and calcification in alginate-abundant brown algae is rare (Bilan and Usov 2001, Navarro and Stortz 2002, 2008, Navarro et al. 2011).   The majority of work on cell wall polysaccharides in corallines has been performed on species within Corallinoideae, i.e. several Corallina and Bossiella species (Turvey and Simpson 1966, Cases et al. 1992, 1994, Usov et al. 1995, Navarro and Stortz 2002), Alatocladia modesta and Haliptylon splendens (Usov et al. 1995), Joculator maximus (Takano et al. 1996), Johansenia macmillanii (as Serraticardia macmillanii, Okazaki et al. 1982), Jania rubens (Navarro and Stortz 2002, 2008), and Calliarthron cheilosporioides (Martone et al. 2010). A small number of crustose species from the subfamily Melobesioideae have also been studied, namely Lithothamnion phymatodeum, Lithothamnion heterocladum, and Clathromorphum nereostratum (Usov et al. 1995, Navarro et al. 2011). Several species within the Lithophylloideae have been studied - the articulated species Amphiroa zonata (Okazaki and Tazawa 1989) and Amphiroa fragilissima (Okazaki and Tazawa 1989, Bilan and Usov 2001), and the crustose species Tenarea tortuosa (as Lithophyllum tortuosum, Okazaki and Tazawa 1989) -  but the focus of these studies was limited to the presence of alginic acids. The only species that has had intergenicular and genicular tissue analyzed separately is Calliarthron cheilosporioides (Martone et al. 2010). Nothing is known about the differences in chemical composition between genicular tissue and intergenicular tissue in the articulated corallines of Lithophylloideae, and no chemical work has been done on Metagoniolithoideae at all. In this study, I analyzed the chemical composition of cell walls in articulated species from Corallinoideae, Lithophylloideae and Metagoniolithoideae. I also examined the cell walls of three crustose species chosen for their phylogenetic relatedness to articulated groups (Fig. 4.1, Bailey and Chapman 1998, Aguirre et al. 2010, Bittner et al. 2011, Kato et al. 2011). I 63  hypothesized that the monosaccharide composition of intergenicular tissue in any given articulated group would be similar to that of the calcified tissue of closely related crusts, indicating a similarity in the polysaccharides involved in the calcification process. I also hypothesized that genicular tissue would be chemically distinct from all calcified tissue. Finally, I tested two competing hypotheses about the chemical composition of genicular cell walls: would they be chemically similar given their comparable function and convergent evolutionary trajectories, or chemically different given their distinct developmental patterns and varying material properties?   Fig. 4.1. Evolutionary relationships between study species (based on Bittner et al. 2011, Kato et al. 2011). Crustose species are in red.    4.2 Materials and methods  4.2.1 Specimen collection Species from the subfamily Corallinoideae, Calliarthron tuberculosum, Corallina officinalis var. chilensis, and Johansenia macmillanii were collected subtidally at approximately 64  3 m depth from Botanical Beach (48˚31’48’’ N, 124˚27’18’’ W) on Vancouver Island, BC, Canada, in June/July of 2012. Articulated species from the subfamily Lithophylloideae, Amphiroa anceps and Amphiroa gracilis, were collected in Point Peron (32˚16’01’’ S, 115˚41’14’’ E), Perth, Western Australia at depths of 3-4.6 m, in December 2012. Metagoniolithon stelliferum #1 and #2 indicate specimens that currently fall under the name Metagoniolithon stelliferum (Metagoniolithoideae), but that appeared morphologically distinct in the field. Sequencing of psbA, CO1 and rbcL genes indicates that these two groups represent distinct species (K.G.J, unpublished data), and so they have been treated as such in this study. Both Metagoniolithon stelliferum “species” were collected in December 2012 from Point Peron at depths of 3-4.6 m, where they were growing epiphytically side-by-side on Amphibolus seagrass. Metagoniolithon chara was collected off Carnac Island (32˚07’07’’ S, 115˚39’52’’ E) near Perth, Western Australia at depths of about 4-5 m in January 2014. Representative vouchers for each species were deposited into the University of British Columbia Herbarium for future taxonomic reference: Calliarthron tuberculosum (A91564); Corallina officinalis var. chilensis (as Corallina officinalis, A91563); Johansenia macmillanii (A91561); Amphiroa anceps (A91566); Amphiroa gracilis (A91572); Metagoniolithon stelliferum #1 (as Metagoniolithon stelliferum, A91576); Metagoniolithon stelliferum #2 (as Metagoniolithon sp., A91579); and Metagoniolithon chara (A91464). Specimens of the crustose species Lithophyllum sp.1 were collected from the low intertidal of 5th Beach Exposed Point on Calvert Island, BC, Canada (51˚38’20’’ N, 128˚09’23’’ W) in June 2016. This species was selected because of its close phylogenetic relatedness to articulated corallines within Lithophylloideae (Fig. 3-1). The crustose species Spongites tumidum (Porolithoideae) and Lithothamnion glaciale (Melobesioideae) were collected intertidally from 65  West Beach Boulders on Calvert Island, BC, Canada (51˚39’05’’ N, 128˚08’37’’ W) in June 2016. Spongites was selected because of its close phylogenetic position relative to the completely articulated subfamily Metagoniolithoideae, and Lithothamnion was selected as a species sister to all the articulated coralline clades (Fig. 3-1). No completely crustose sister clade is currently known for the Corallinoideae.  4.2.2 Species determinations In order to confirm the species identities of corallines, DNA was extracted from several specimens of each suspected species. Extraction of DNA followed the protocol outlined in Saunders (2008). Genes for photosystem II protein D1 (psbA, 863 bp) and the mitochondrial cytochrome c oxidase subunit (CO1-5P, 664 bp) were amplified and sequenced.  The psbA gene was amplified using the primers psbAF1 and psbAR2 (Yoon et al. 2002), and the CO1 gene was amplified using the primers GWSFn (Le Gall and Saunders 2010) and GWSRx (Clarkston and Saunders 2012). The PCR amplification process for both genes was performed following Hind and Saunders (2013). Products of the PCR reactions were sent to the Genome Quebec Innovation Centre (McGill University, Montreal, Quebec, Canada), where they were sequenced using standard Sanger sequencing methods on a 3730xl DNA Analyzer (Applied Biosystems, Foster City, CA, USA). In order to confirm species identities, sequence data was edited and aligned with Geneious 11.0.5 (Kearse et al. 2012) and compared against sequences in a local database, as well as BOLD (The Barcode of Life Data System, Ratnasingham and Herbert 2007) using BLAST (Basic Local Alignment Search Tool, Altschul et al. 1997).  66  4.2.3 Tissue preparation Both articulated and crustose species were inspected carefully for epiphytes, which were removed manually with either forceps or a toothbrush. Only healthy vegetative tissues were used. Specimens were dried via plant press (for articulated species) or silica gel (for crustose species), and intergenicula and genicula of articulated species were carefully separated using a razor blade. Tissues were macerated with a mortar and pestle, after which calcified tissues were decalcified via dropwise addition of 1 M HCl and constant stirring until no more bubbling was detected. Decalcified tissues were centrifuged at 10,000 RPM for 5 min, and the supernatant was removed and discarded. Decalcified pellets of intergenicula and crustose tissues, as well as macerated genicular tissues, were dried for 48 hrs in an oven at 50˚C. Decalcified tissues were reground in a mortar and pestle following oven drying, then placed back in the oven for a minimum of 1 hr prior to weighing for acid hydrolysis.  4.2.4 Neutral monosaccharide analysis A scaled down modification of the acid hydrolysis component of a klason procedure was performed on all tissues (Huntley et al. 2003). In brief, samples of ~1 mg of genicular/intergenicular tissue or ~10 mg of crustose tissue were weighed out into 2 mL graduated, free-standing microcentrifuge tubes with screw-caps and o-rings (catalog #02-681-374, ThermoFisher Scientific, MA, USA). These tubes were used because of their slightly >2 mL capacity, which was necessary for the following procedure.  50 μL of 72% H2SO4 was added to ~1 mg of ground tissue (articulated species), or 100 μL was added to ~10mg of ground tissue (crustose species). Samples were vortexed briefly, then incubated in a heating block at 30˚C for 1hr while being shaken at 500 RPM. 1 mL (for 10 mg samples) or 2 mL (for 1 mg samples) of 67  deionized H2O was added, and samples were vortexed briefly to resuspend tissue. Samples were then autoclaved at 120˚C for 90 min, allowed to cool, and centrifuged at 13,000 RPM for 10 min. The supernatant was separated from the remaining pellet, then filtered through 0.45 μm filter tips. Filtered supernatants were analyzed for monosaccharide composition using a high performance liquid chromatograph (HPLC) (Dionex DX-600, ThermoFisher Scientific, MA, USA) with an ion exchange PA-1 column and a Dionex AS-50 autosampler (ThermoFisher Scientific, MA, USA). Samples were injected in 25 μL volumes onto the column, which was equilibrated with 250 mM NaOH. Samples were eluted with deionized H2O at a rate of 1 mL min-1, followed by post-column addition of 200 mM NaOH at 0.5 mL min-1. All samples were run with technical duplicates. Neutral sugar standards containing glucose, galactose, xylose, fucose, arabinose, rhamnose, and mannose were hydrolyzed, autoclaved and analyzed by HPLC alongside tissue samples to account for monosaccharide breakdown.  Each sample represented a unique pooled set of 3-5 individuals, due to the minimal weight of genicula and decalcified tissues. Five samples per tissue type, per species, were prepared for articulated species, while only one sample was prepared for each crustose species due to the low number of individuals collected.  4.2.5 Uronic acid analysis A separate acid hydrolysis using trifluoroacetic acid (TFA) was performed on genicular tissue from the following articulated coralline species: Calliarthron tuberculosum, Johansenia macmillanii, Amphiroa gracilis, Metagoniolithon stelliferum #2, and Metagoniolithon chara. These species were chosen based on remaining tissue availability after initial monosaccharide analysis. Only one sample was run per species.  68  1 mL of 2 M TFA and 5 μL of 0.04 M erythritol internal standard was added to samples of ~15 mg of genicular tissue. Samples were autoclaved at 120˚C for 2 hrs, after which the TFA was evaporated off under nitrogen at a temperature between 73-80˚C. Samples were resuspended in 500 mL of deionized H2O, and centrifuged at 13,000 RPM for 10 min. The supernatant was removed and filtered through 0.45 μm filter tips, then analyzed on the same system used for the neutral monosaccharide hydrolysates. Samples were injected in 25 μL volumes onto the column and eluted at a rate of 0.4 mL min-1 with 100 mM NaOH and a linear ramp of sodium acetate from 10 mM to 400 mM, followed by post-column addition of 200 mM NaOH at 0.5 mL min-1. Between samples, the column was cleaned with 300 mM NaOH and 100 mM sodium acetate, then equilibrated back to starting conditions of 100 mM NaOH and 10 mM sodium acetate. All samples were run with technical duplicates. Standards containing galacturonic acid and glucuronic acid were hydrolyzed and run on the HPLC alongside genicular tissue samples. Samples of pure alginate (Kelco, CA, USA), as well as samples of alginate spiked with galacturonic acid and glucuronic acid, were also run for comparison. Due to uncertainty in the ratio of guluronic:mannuronic acid in the pure alginate sample, these sugars could only be assessed as present or absent rather than quantitatively.   4.2.6 Linkage analysis Linkage analysis of carbohydrates was performed by the Complex Carbohydrate Research Center at the University of Georgia, using a protocol modified from Heiss et al. (2009). Samples of ~1 mg of ground genicular tissue from Calliarthron tuberculosum, Amphiroa gracilis, and Metagoniolithon stelliferum #2 first underwent a desulfation step in which sulfated polysaccharides were converted to pyridinium salts, followed by removal of sulfate groups in a 69  heated mixture of dimethyl sulfoxide (DMSO) and methanol. After removal of DMSO, samples underwent permethylation, reduction, and acetylation as described in Heiss et al. (2009), and the resulting partially methylated alditol acetates were analyzed via gas chromatography-mass spectrometry (Hewlett-Packard 5890 GC, Hewlett-Packard, Palo Alto, CA, USA). Potential structural units were assigned to linkages where possible, based on what has previously been found in corallines (Turvey and Simpson 1966, Okazaki et al. 1982, Cases et al 1992, Cases et al. 1994, Usov et al. 1995, Takano et al. 1996, Navarro and Stortz 2002, 2008, Martone et al. 2010, Navarro et al. 2011).   4.2.7 Statistics Principal components analysis (PCA) was performed on neutral monosaccharide data in R 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria) using the RStudio interface (version 0.98.1056, RStudio, Boston, MA, USA) and the PCA() function from the FactoMineR package (Multivariate and Exploratory Data Analysis and Data Mining, version 1.39, Francois Husson, Julie Josse, Sebastien Le, and Jeremy Mazet 2017). All data were log transformed and scaled to avoid disproportionate effects of more abundant monosaccharides. PCAs were performed first on all neutral monosaccharide data for all tissue types to determine whether calcified tissues from articulated and crustose species were more similar than either tissue type compared to genicular tissues. PCAs were then performed on calcified tissues alone (i.e. intergenicula and crusts) and genicular tissues alone to look for separation based on phylogeny. Using the Kaiser criterion, only principle components with eigenvalues >1 were considered further. For further qualitative analysis, the ordiellipse() function from the vegan package (Community Ecology Package, version 2.4-6, Jari Oksanen, F. Guillame Blanchet, Michael 70  Friendly, et al. 2018) was used to draw 95% confidence intervals around subfamilies and tissue types.   A permutational multivariate analysis of variance (PERMANOVA) was performed on a model that included tissue type (calcified vs noncalcified), clade (articulated subfamilies and their most closely related crust), and the interaction as factors. This was done in RStudio using the adonis() function from the vegan package. Homogeneity of multivariate dispersion was tested using the betadisper() function from the vegan package.  4.3 Results  4.3.1 Species determinations Many species used in this study had CO1 sequences that matched (>99%) their field identification when compared against the BOLD database; Calliarthron tuberculosum, Johansenia macmillanii, Amphiroa gracilis, Metagoniolithon stelliferum #1, Metagoniolithon chara, and the crustose species Spongites tumidum were all positively identified. The species referred to here as Amphiroa anceps matched a currently undescribed species, Amphiroa sp.1WA (99.6% CO1 match), but further information on the matching sequences was not yet available and therefore I retained the field identification for the purposes of this study. CO1 sequences of the remaining species were not closely matched in the BOLD database, and so psbA sequences were compared against a local database. One crustose species matched Lithothamnion glaciale (100% psbA match, PTM442). Metagoniolithon stelliferum #2 was most similar (99.5% psbA sequence similarity) to an unknown Metagoniolithon species collected from Western Australia (Metagoniolithon sp.2WA, GWS024725) with the next closest match being Metagoniolithon 71  stelliferum (98.4% psbA sequence similarity, GWS016595). Compared against the local database, one articulated species matched an undescribed species (Corallina sp.1 frondescens, 100% psbA match, PTM1244) currently grouped under the name Corallina officinalis var. chilensis, and one crustose species matched an undescribed species (Lithophyllum sp.1, 100% psbA match, PTM674).  4.3.2 Monosaccharide composition Both glucose and galactose were found in all tissue types and all species, both crustose and articulated (Table 4.1). Xylose was found in almost all samples, except for the crust Lithothamnion glaciale (Table 4.1). Low amounts (<1% dry weight) of fucose were also present in all species and tissues except the genicula of Calliarthron tuberculosum and the intergenicula of Metagoniolithon stelliferum #2 (Table 4.1). Low levels (<1% dry weight) of rhamnose were found in the genicula of all articulated species other than Metagoniolithon chara and Metagoniolithon stelliferum #1. Rhamnose was also detected in low levels in some calcified tissues but not others, with no clear phylogenetic pattern (Table 4.1). Mannose was detected in most species and tissue types but was notably absent from both the genicula and intergenicula of all metagoniolithoids. It was also not detected in Lithothamnion glaciale (Table 4.1). Arabinose was absent from almost all species, but was detected at a very low level in the single sample of Spongites tumidum. It was also detected in one sample of genicular tissue from Amphiroa gracilis, but only in one of the two technical duplicates – the remainder of the monosaccharide content from the anomalous duplicate was also inconsistent with all other Amphiroa gracilis samples, so this duplicate was removed from analysis.72   Table 4.1. Monosaccharide composition (% dry weight) of genicular, intergenicular, and crust tissues. Values for articulated species indicate average ± s.e.m, n=5. Crustose species values indicate the average of two technical duplicates. Subfamily Species Glu Gal Xyl Fuc Ara Rha Man TotalCorallinoideae Calliarthron tuberculosum 6.23 ± 0.34 5.98 ± 0.42 0.12 ± 0.01 - - 0.02 ± 0.02 0.27 ± 0.03 12.62 ± 0.78Corallinoideae Corallina officinalis var. chilensis 15.38 ± 1.11 2.83 ± 0.20 1.68 ± 0.10 0.10 ± 0.01 - 0.79 ± 0.05 0.37 ± 0.04 21.14 ± 1.49Corallinoideae Johansenia macmillanii 10.30 ± 0.44 5.30 ± 0.28 0.20 ± 0.02 0.16 ± 0.01 - 0.12 ± 0.01 0.37 ± 0.02 16.46 ± 0.75Lithophylloideae Amphiroa gracilis 6.17 ± 0.32 10.81 ± 0.42 11.76 ± 0.54 0.32 ± 0.01 - 0.52 ± 0.02 0.94 ± 0.05 30.53 ± 1.24Lithophylloideae Amphiroa anceps 5.49 ± 0.36 11.40 ± 0.61 3.29 ± 0.36 0.16 ± 0.01 - 0.04 ± 0.01 0.30 ± 0.05 20.68 ± 0.59Metagoniolithoideae Metagoniolithon chara 2.88 ± 0.13 8.06 ± 0.84 4.57 ± 0.44 0.10 ± 0.01 - - - 15.61 ± 1.38Metagoniolithoideae Metagoniolithon stelliferum #1 3.71 ± 0.31 6.85 ± 0.41 2.04 ± 0.13 0.07 ± 0.01 - - - 12.68 ± 0.83Metagoniolithoideae Metagoniolithon stelliferum #2 1.48 ± 0.08 3.79 ± 0.15 2.80 ± 0.13 0.06 ± 0.01 - 0.09 ± 0.01 - 8.22 ± 0.32Corallinoideae Calliarthron tuberculosum 19.97 ± 5.65 0.98 ± 0.32 0.05 ± 0.05 0.14 ± 0.02 - - 0.20 ± 0.06 21.34 ± 6.01Corallinoideae Corallina officinalis var. chilensis 10.67 ± 0.63 0.61 ± 0.03 0.06 ± 0.01 0.07 ± 0.01 - 0.02 ± 0.01 0.08 ± 0.02 11.51 ± 0.68Corallinoideae Johansenia macmillanii 11.72 ± 0.93 0.76 ± 0.07 0.02 ± 0.01 0.09 ± 0.01 - - 0.16 ± 0.01 12.75 ± 1.02Lithophylloideae Amphiroa gracilis 31.49 ± 3.43 2.22 ± 0.35 0.77 ± 0.26 0.17 ± 0.02 - 0.13 ± 0.06 0.38 ± 0.05 35.17 ± 4.10Lithophylloideae Amphiroa anceps 11.03 ± 1.43 0.88 ± 0.12 0.07 ± 0.01 0.04 ± 0.01 - - 0.20 ± 0.01 12.22 ± 1.55Metagoniolithoideae Metagoniolithon chara 15.49 ± 1.07 3.65 ± 0.38 0.55 ± 0.10 0.08 ± 0.01 - 0.18 ± 0.02 - 19.93 ± 1.03Metagoniolithoideae Metagoniolithon stelliferum #1 5.02 ± 0.57 3.82 ± 0.35 0.28 ± 0.04 0.04 ± 0.01 - 0.06 ± 0.02 - 9.23 ± 0.76Metagoniolithoideae Metagoniolithon stelliferum #2 5.60 ± 0.58 1.98 ± 0.34 0.23 ± 0.05 - - 0.31 ± 0.24 - 8.12 ± 0.51Porolithoideae Lithophyllum sp.1 6.85 ± 0.89 0.92 0.10 0.06 0.01 0.08 0.35 8.35Lithophylloideae Spongites tumidum 6.05 0.76 0.10 0.02 - 0.18 0.14 7.25Melobesioideae Lithothamnion glaciale 4.88 0.26 - 0.01 - - - 5.15Monosaccharide composition (% dry weight)GENICULAINTERGENICULACRUSTS73   Glucose was typically higher in intergenicular tissue than genicular tissue within a given species, except for in Corallina officinalis var. chilensis where this trend was reversed (Table 4.1). Glucose was the most dominant monosaccharide in all calcified tissue, with levels ranging between 1.3 (Metagoniolithon stelliferum #1) to 20.3 (Calliarthron tuberculosum) times that of the next most abundant sugar, galactose (Table 4.2). In corallinoid genicula, glucose was the most abundant monosaccharide present; however, galactose was more abundant than glucose in the genicula of all lithophylloid and metagoniolithoid species (Table 4.2).   Both galactose and xylose content were consistently higher in genicular tissue than intergenicular tissue for all articulated species tested, indicating a higher level of galactans in general (Table 4.1). Crustose tissues were also low in galactose and xylose, and similar to intergenicular tissues. Galactose and xylose were particularly abundant in lithophylloid and metagoniolithoid genicula when compared to corallinoid genicula. The ratio of xylose/galactose varied for both genicula and calcified tissues, from 0.02-1.31 (Table 4.2), with no clear pattern based on either tissue type or phylogeny.  The total monosaccharide content for most species was <25% of the original tissue dry weight regardless of tissue type, with the exception of Amphiroa gracilis which had a monosaccharide content >30% for both genicular and intergenicular tissue. Total monosaccharide content for crustose species was particularly low, at <10% of the total dry weight (Table 4.1). 74   Table 4.2. Monosaccharide composition relative to total sugar content (mol%) of genicular, intergenicular, and crustose tissues. Values for articulated species indicate averages of n=5, values for crustose species are averages of two technical duplicates.Subfamily Species Glu Gal Xyl Fuc Ara Rha Man Glu/Gal Xyl/GalCorallinoideae Calliarthron tuberculosum 49.32 47.30 1.10 - - 0.19 2.10 1.04 0.02Corallinoideae Corallina officinalis var. chilensis 71.32 13.11 9.32 0.51 - 4.00 1.73 5.44 0.71Corallinoideae Johansenia macmillanii 62.34 32.07 1.47 1.09 - 0.80 2.23 1.94 0.05Lithophylloideae Amphiroa gracilis 18.71 32.80 42.82 1.07 - 1.74 2.87 0.57 1.31Lithophylloideae Amphiroa anceps 25.71 53.40 18.48 0.81 - 0.21 1.39 0.48 0.35Metagoniolithoideae Metagoniolithon chara 17.41 48.73 33.20 0.66 - - - 0.36 0.68Metagoniolithoideae Metagoniolithon stelliferum #1 28.40 52.29 18.72 0.59 - - - 0.54 0.36Metagoniolithoideae Metagoniolithon stelliferum #2 16.86 43.10 38.14 0.80 - 1.09 - 0.39 0.89Corallinoideae Calliarthron tuberculosum 93.47 4.59 0.29 0.72 - 0.00 0.93 20.36 0.06Corallinoideae Corallina officinalis var. chilensis 92.53 5.32 0.66 0.65 - 0.16 0.67 17.39 0.12Corallinoideae Johansenia macmillanii 91.78 5.98 0.23 0.76 - - 1.25 15.34 0.04Lithophylloideae Amphiroa gracilis 89.07 6.28 2.62 0.54 - 0.40 1.09 14.18 0.42Lithophylloideae Amphiroa anceps 90.13 7.18 0.68 0.40 - - 1.61 12.55 0.09Metagoniolithoideae Metagoniolithon chara 77.19 18.20 3.30 0.41 - 0.90 - 4.24 0.18Metagoniolithoideae Metagoniolithon stelliferum #1 54.06 41.08 3.59 0.52 - 0.75 - 1.32 0.09Metagoniolithoideae Metagoniolithon stelliferum #2 68.30 24.12 3.41 0.00 - 4.16 - 2.83 0.14Lithophylloideae Lithophyllum sp.1 94.71 4.99 - 0.31 - - - 18.98 -Porolithoideae Spongites tumidum 81.72 10.92 1.36 0.74 0.13 0.98 4.15 7.49 0.12Melobesioideae Lithothamnion glaciale 83.00 10.37 1.67 0.33 - 2.68 1.94 8.00 0.16CRUSTSMonosaccharide composition (mol%) RatiosGENICULAINTERGENICULA75   Galacturonic acid and guluronic acid peaks could not be separated, as indicated by the fact that samples of alginate spiked with galacturonic and glucuronic acid only separated into three peaks on HPLC. It is therefore uncertain whether the corresponding peaks in the genicular tissue samples represent galacturonic or guluronic acid. One or the other, or both, was present in all samples (Table 4.3). Mannuronic acid was also detected in all samples tested, as was glucuronic acid at levels <3% (Table 4.3).   Table 4.3. Uronic acid composition of genicular tissue. Values for Glu A (glucuronic acid) are averages of two technical duplicates, x for Gul A/Gal A (guluronic acid/galacturonic acid) and Man A (mannuronic acid) indicates presence.  4.3.3 Principal components analysis The first two principal components (PC1, PC2) of the analysis on all tissue monosaccharide data accounted for a total of 66% of the variation. A third principal component explained 15% of the variation, but it was heavily based on arabinose (95%), which was only detected in one species (Spongites tumidum). This component was not considered further. Contributions to PC1 were fairly evenly distributed across xylose, fucose, mannose, galactose, Subfamily Species Gal A/Gul A Man A Glu ACorallinoideae Calliarthron tuberculosum x x 0.15Corallinoideae Corallina officinalis var. chilensis x x 1.04Lithophylloideae Amphiroa gracilis x x 2.79Metagoniolithoideae Metagoniolithon chara x x 2.36Metagoniolithoideae Metagoniolithon stelliferum #2 x x 2.08Uronic acid content (% dry weight)76  and rhamnose. Glucose alone contributed 48% of the variation of PC2, with lesser contributions from galactose, mannose, and xylose.  Clustering in PC1 and PC2 was evident between calcified and noncalcified tissues, with crusts clustering with intergenicula from articulated species (Fig. 4.2). There was a significant effect of both tissue type (PERMANOVA, R2=0.22, p<0.001) and clade (R2=0.22, p<0.001), as well as a significant interaction (R2=0.10, p<0.001). However, the assumption of homogeneity of multivariate dispersion was violated for both tissue type and clade, so these results must be interpreted with caution.  Fig 4.2. Principal components of neutral monosaccharides from genicular (blue), intergenicular (yellow) and crustose (red) data. Each point represents a single sample. The first principal component (PC1) was composed primarily of fucose (25%), xylose (25%), mannose (20%), galactose (18%) and rhamnose (12%). The second principal component (PC2) was composed primarily of glucose (48%), mannose (16%), galactose (16%), mannose (15%), and xylose (11%). All other contributions to dimensions 1 and 2 were <10%. Ellipses indicate 95% C.I. around tissue type clusters.  77  The first two principal components of the analysis on genicular tissues accounted for 80% of the variation. PC1 was mostly based on mannose and fucose, with lesser contributions from rhamnose and xylose. PC2 was mostly based on galactose, glucose, and xylose, with a lesser contribution from rhamnose. Data for PC1 and PC2 clustered clearly based on subfamily (Fig. 4.3).  Fig. 4.3. Principal components of neutral monosaccharides from genicular tissue of Corallinoideae (green), and Lithophylloideae (purple) and Metagoniolithoideae (orange). Each point is a single sample. The first principal component (PC1) was composed primarily of mannose (28%), fucose (28%), rhamnose (19%) and xylose (11%). The second principal component (PC2) was composed primarily of galactose (30%), glucose (28%), xylose (23%) and rhamnose (13%). All other contributions to dimensions 1 and 2 were <10%. Ellipses indicate 95% C.I. around subfamily clusters.  The first two principal components of the analysis on calcified tissues accounted for 68% of the variation. A third component explained 16% of the variation, but it was heavily based on arabinose (83%), so it was not considered further. PC1 was based predominantly on glucose 78  (32%) and fucose (31%), with lesser contributions from mannose and xylose. Galactose contributed 39% of the variation of PC2, with lesser contributions from xylose and rhamnose. Metagoniolithoids appear to separate out from other groups slightly, but clustering in PC1 and PC2 based on relatedness is not clear (Fig. 4.4).   Fig. 4.4. Principal components of neutral monosaccharides from intergenicular tissues (circles) and crustose tissues (diamonds). Colours indicate closely related groups: green is Corallinoideae, purple is Lithophylloideae, orange is Metagoniolithoideae and Porolithoideae, and black is Melobesioideae.  Each point represents a single sample. The first principal component (PC1) was composed primarily of glucose (32%), fucose (31%), mannose (24%), and xylose (11%). The second principal component (PC2) was composed primarily of galactose (39%), xylose (26%), and rhamnose (26%). All other contributions to dimensions 1 and 2 were <10%. Ellipses indicate 95% C.I. around articulated subfamily clusters.  4.3.4 Linkage analysis All species possessed 1-4-glucose, likely indicating the presence of cellulose (Table 4.4). The mol% of glucose was much higher overall in Calliarthron tuberculosum than the other two 79  groups; one linkage, 3-Glu, was only found in Calliarthron. What this linkage corresponds to structurally is unknown.  Table 4.4.  Linkage analysis of genicular tissue from Calliarthron tuberculosum, Amphiroa gracilis, and Metagoniolithon stelliferum #2. Units are expressed as mol%. aGlu=glucose, Gal=galactose, Xyl=xylose. bG=β-D-galactose, LG=α-L-galactose, X=xylose, S=sulfate, U=unknown, numbers indicate position of substitution on galactose unit.    Although D and L conformation were not tested here, it was assumed that 3-Gal units corresponded to β-D-galactose, while 4-Gal corresponded to α-L-galactose. This assumption was based on the ratio of 3-Gal to 4-Gal, which approached parity for all species (1.11 for C. Deduced Linkagea Possible Structural UnitsbCalliarthron tuberculosumAmphiroagracilisMetagoniolithonstelliferum  #2Glucoset-Glu cellulose 3.8 1.0 1.34-Glu cellulose 13.8 1.4 1.7Galactose3-Gal G 37.0 30.1 31.63,6-Gal G6X, G6S 3.1 5.1 6.42,3,6-Gal G2U6X, G2U6S - 2.0 2.64-Gal LG 32.4 23.4 27.32,4-Gal LG2S 0.9 3.6 6.03,4-Gal LG3S, G4S 2.2 - -2,4,6-Gal LG2S6U - 1.6 1.8Xyloset-Xyl G6X 0.8 4.5 3.9Unassigned Linkages3-Glu Unknown 2.3 - -2-Gal Unknown 0.3 0.4 1.02,3-Gal Unknown 0.7 15.4 13.84,6-Gal Unknown 0.7 10.4 2.12-Xyl Unknown 0.2 0.4 0.14-Xyl Unknown 1.9 0.6 0.380  tuberculosum, 0.95 for A. gracilis, and 1.09 for M. stelliferum #2), indicating likely involvement of these linkages in the agar backbone structure. Possible sulfation positions were not directly tested, but they were assumed to be comparable to what has been previously found in Calliarthron cheilosporioides (Martone et al. 2010). There was a greater proportion of galactose overall in A. gracilis and M. stelliferum #2 compared to C. tuberculosum, with much of this difference being the result of galactose units with previously undocumented linkages (Table 4.4). 2,3-Gal was found in all three species, but it was much more abundant in A. gracilis and M. stelliferum #2 than in C. tuberculosum. 4,6-Gal was also found in all three species, but more abundantly in A. gracilis than in either of the other two species. Additionally, both A. gracilis and M. stelliferum #2 possessed the highly substituted units, 2,3,6-Gal and 2,4,6-Gal, neither of which was detected in C. tuberculosum. Finally, t-Xyl was roughly 4x as abundant in A. gracilis and M. stelliferum #2 than C. tuberculosum, indicating a higher presence of xylose side-stubs on the agar backbone of these groups.   4.4 Discussion This work is the first to compare the chemical composition of genicula and intergenicula of all three articulated clades, as well as the first to make direct comparisons between calcified tissues of crustose and articulated species. These chemical data offer a new perspective on the commonality of calcified coralline walls, the convergent evolution of coralline genicula, and the biochemical underpinnings of these important biomechanical structures. I found that the chemical composition of calcified tissues across both articulated and crustose clades may not have changed much over the course of evolution, perhaps because the structural role of these tissues has remained consistent. In contrast, genicular tissues of different articulated clades 81  exhibit several notable differences in both content and structure of cell wall polysaccharides; this may have resulted from differences in genicular development, and it is reflected in differences in material properties.  4.4.1 Monosaccharide composition and linkage analysis Results from the linkage analysis indicate that much of the glucose in genicula of all three articulated clades is 1-4-linked (Table 4.4); this is suggestive of cellulose. Hemicellulosic polysaccharides may also have a backbone of 1-4-Glu, but they are not typically abundant in red algae relative to other polysaccharides (Kloareg and Quatrano 1988, Masarin et al. 2016). While it is possible that some fraction of the 1-4 Glu observed is involved in non-cellulosic carbohydrates, I henceforth consider glucose content to be an indicator of relative cellulose content between tissues and species. Similarly, galactose content is a reasonable proxy for galactans, as the ratio between 1-3 and 1-4-linked Gal units was close to 1:1 in all three linkage analysis samples (Table 4.4), indicating a typical agaran backbone structure (Kloareg and Quatrano 1988, Martone et al. 2010). If glucose is treated as a proxy for cellulose and galactose as a proxy for galactans, results from the monosaccharide analysis show that cellulose content is usually higher in intergenicular tissue than genicular tissue, while galactan content is lower (Table 4.1). Xylose content was also lower in intergenicula than in genicula, for all species.  This last result is inconsistent with Martone et al. (2010), who found that xylose sidechains were more abundant in the galactans of intergenicula than genicula in Calliarthron cheilosporioides. It is expected that results for corallinoid species, at least, would follow a similar pattern. This may be partially due to an incomplete extraction of the calcified tissues in my study – Navarro and Stortz (2002) compared previously published methods of coralline 82  galactan extraction and found that methods using higher concentrations and longer treatments of HCl for initial decalcification led to a higher total % yield of carbohydrates, suggesting that the calcium carbonate left in more gentle treatments blocked full extraction. Conversely, more gentle treatments had a higher ratio of Xyl:Gal, suggesting that harsh treatments may have cleaved and extracted xylose sidechains pre-emptively. Finding the ideal level of decalcification, then, is a balance between exposing as much of the cell wall to extraction as possible without pre-extracting too much of the polysaccharides being exposed. The treatment used here was less harsh than even the gentlest acid treatment compared in Navarro and Stortz (2002) – the method of Cases et al. (1992) exposed tissues to 1M HCl for 24 hrs, while tissues in this study were only exposed to 1M HCl until visible bubbling ceased, typically <4 hrs.  Therefore, it may have failed to expose all the xylogalactans present – if xylose sidechains are indeed an integral component of the calcium carbonate network, they could be particularly difficult to remove without complete decalcification. It should be noted that all calcified tissues in this study received the same decalcification treatment; therefore, while comparisons between intergenicula and genicula should be interpreted with caution, comparisons of calcified tissues between species are still reliable. The other potential source of discrepancy between the results of this study and Martone et al. (2010) lie in the source of the monosaccharides analyzed. While Martone et al. (2010) performed analysis specifically on galactan extracts, here I analyzed whole hydrolyzed wall tissue. Linkage analysis reveals that not all the xylose in genicular tissue is t-Xyl, the linkage associated with xylose sidechains; 2-Xyl is also present, and 4-Xyl content in Calliarthron tuberculosum is more than twice as high as that of t-Xyl (Table 4.4). If 2-Xyl and 4-Xyl are associated with some polysaccharide other than galactans, this might explain why corallinoid 83  genicula in this study had a higher xylose content overall than what was found in Martone et al. (2010). Incomplete extraction could explain the low xylogalactan levels found in intergenicula, while contributions of non-galactan associated xylose could explain the high levels in corallinoid genicula. The presence of uronic acids in genicular tissues was also somewhat unexpected. Glucuronic acid and guluronic acid could not be differentiated in the current HPLC analysis, but the presence of both mannuronic acid and galacturonic acid were identified. Potential guluronic acid and definitive mannuronic acid are suggestive of the presence of alginates, which are commonly found in many corallines species (Okazaki et al. 1982, Okazaki and Tazawa 1989, Bilan et al. 1995, Bilan and Usov 2001, Navarro and Stortz 2002, 2008, Navarro et al 2011). Some authors have suggested a role for alginates in coralline calcification based on their ability to bind calcium ions, as well as the observation that alginates are not found in closely related noncalcified algae (Okazaki et al. 1982, Okazaki and Tazawa 1989). Total uronic acid content is typically low, however, between 2-6% of dry weight (Okazaki and Tazawa 1989, Navarro and Stortz 2002, 2008, Navarro et al. 2011). This number includes uronic acids not associated with alginates, such as glucuronic and galacturonic acid. If alginates are indeed playing a role in calcification, then a high abundance is not required, so the presence of alginate at even low levels in the uncalcified genicula is surprising. I did not test for the presence of uronic acids in calcified tissues, but future studies would benefit from a comparison of alginate content in intergenicula and genicula. Results from both the monosaccharide analysis (Table 4.1, Table 4.2) and the linkage analysis (Table 4.4) revealed a higher glucose content in the genicula of corallinoid species than in lithophylloids and metagoniolithoids, and a higher galactose content in the genicula of 84  lithophylloids than in corallinoids and metagoniolithoids. These differences could play a role in the differing mechanical properties of the tissue between articulated clades. If glucose and galactose are indicative of cellulose and galactans, then cellulose in genicular tissues could contribute to strength while galactans could contribute to extensibility. The genicula of corallinoid species are typically stronger than genicula in both lithophylloids and metagoniolithoids, while lithophylloids are often more extensible (Ch.1). The possible connections between chemical composition and mechanical properties are explored further in Ch. 5. Structural differences in the galactans of genicular tissues from different species were revealed in the linkage analysis (Table 4.4). All the genicular galactans described exhibited notable differences when compared to the typical “corallinan” structure described by others (Turvey and Simpson 1966, Cases et al. 1992, 1994, Usov et al. 1995, Takano et al. 1996, Navarro and Stortz 2002, Navarro et al. 2011). Although the typical agaran backbone structure is suggested by 1-3 and 1-4 linked Gal, the presence of xylose side-stubs was low. The proportion of t-Xyl was higher in Amphiroa gracilis and Metagoniolithon stelliferum #2 than in Calliarthron tuberculosum, but the ratio of Xyl:Gal was universally lower than the 1:3 ratio suggested by previous works (Cases et al. 1992, 1994); ~1:15 for A. gracilis, ~1:20 for M. stelliferum #2, and ~1:95 in C. tuberculosum. This does not preclude the hypothesis put forward by Martone et al. (2010), wherein xylose branches may be removed during decalcification to prevent calcification in the genicular tissue; if xylose branching does indeed provide a site of nucleation for calcium crystals in calcified tissues, disruption of that network by a factor of 5 (in the case of Amphiroa gracilis) could have a significant effect. 85  In addition to differences in the amount of possible xylose sidechains, an overall difference in the degree of branching of galactans was observed. Of note were two units present in Amphiroa gracilis and Metagoniolithon stelliferum #2, but absent in Calliarthron tuberculosum; 2,3,6-Gal and 2,4,6-Gal. If these units are part of the galactan structure, a higher degree of branching than what is typically found in corallinans would be expected. Two linkages that were present in all species (but especially in A. gracilis and M. stelliferum #2) were 2,3-Gal and 4,6-Gal, but it is not clear whether these are involved in galactans or part of some other polysaccharide. Several linkages could not be assigned to potential structural units, as they have not been previously documented in corallines. Along with 2,3-Gal, 4,6-Gal, 2,3,6-Gal and 2,4,6-Gal, I also found 3-Glu, 2-Gal, 2-Xyl, and 4-Xyl (Table 4.4). These will require future investigation before their polysaccharide assignments can be assigned. Determining the absolute configuration of the galactose units could indicate whether 2,3-Gal and 4,6-Gal are involved in the agar backbone, with substitutions of sulfate groups at positions 2 and 4 respectively. Alternatively, these units could be part of a side-chain connected to the agar backbone, or they could be part of another polysaccharide entirely. 4-Xyl may be incorporated into a hemicellulose type structure or involved in cellulose directly, as has been occasionally documented in red algae (Cronshaw et al. 1958, Turvey and Williams 1970, Kloareg and Quatrano 1988).  4.4.2 Comparison of calcified and uncalcified tissues Chemical differences between tissue types were revealed with both principal component analysis (PCA) and PERMANOVA (Fig. 4.2). The visual clustering (Fig. 4.2), as well as significant effect of tissue type on chemical composition in a PERMANOVA test, indicate that 86  crusts and intergenicula from different subfamilies are more similar to one another than intergenicula are to genicula in the same species. The fact that there was also a significant effect of clade and an interaction between clade and tissue type suggests that genicular and intergenicular tissues differ among species. It should be noted that the assumption of homogeneity of multivariate dispersion was violated for both tissue type and clade, so it cannot be ruled out that some of this difference between groups was the result of different levels of variation; however, the visual clustering of the PCA (Fig. 4.2) supports the conclusion that there is a difference in chemical composition between tissue types. In the case of galactose and minor sugars, differences in ease of extraction between calcified and uncalcified tissues cannot be ruled out as a factor; it may be that xylogalactan content in particular is higher in calcified tissues than what data in this study shows. Differences in glucose (the primary driver of variation in PC2), however, are unlikely to be explained by this; glucose was typically either comparable or higher in calcified tissues compared to genicula, so this difference would only increase if there is an issue of incomplete extraction.   When PCAs were performed on tissue types separately, genicular tissues showed clear clustering based on clade (Fig. 4.3) while calcified tissue clades overlapped significantly (Fig. 4.4). Most of the separation between clades in genicula appeared to be in PC2, which was driven predominantly by galactose, glucose, and xylose. Crustose tissues did not appear to group with the intergenicula of closely related articulated species, but instead clustered together within the region of overlap between intergenicular tissues (Fig. 4.4). This suggests that relatively little has changed in the chemical composition of crustose species over evolutionary time. For intergenicular tissues, the radiating spread of the lithophylloid and metagoniolithoid clusters away from the region of overlap suggests a higher variation of chemical composition within 87  these groups, with the corallinoids displaying a greater similarity in composition among species (Fig. 4.4). The corallinoid cluster is situated completely within the lithophylloid cluster, while the metagoniolithoid cluster is more isolated, suggesting something is chemically unique within this group.  The most consistent difference in the intergenicular tissue of metagoniolithoids compared to that of other groups is the lack of mannose; while it is not abundant in corallinoids and lithophylloids (<1% dry weight, Table 4.1), the complete absence in metagoniolithoids is interesting. Metagoniolithoid intergenicula also typically have a higher amount of galactose than intergenicula in other groups (Table 4.1). The reason behind this difference is not clear, but might be elucidated in future studies if the polysaccharide containing mannose residues in lithophylloids and corallinoids can be identified. The results of this study support the hypothesis that calcified coralline tissues are similar to one another regardless of whether they come from an articulated or a crustose species. This should be considered with caution, however, given discrepancy in ease of cell wall extraction between calcified and uncalcified tissue types. Future work to confirm this result might require harsher decalcification, and extractions targeting specific components of the cell wall such as galactans. PCA analysis of calcified tissues indicates that crustose tissues of different subfamilies may be more similar to one another than to they are to related articulated species, which suggests that crustose species may not have changed significantly in chemical composition over time. Intergenicula of some articulated species may have developed different chemical compositions over time (i.e. metagoniolithoids), while others appear to be relatively similar to crustose species and may not have changed much at all (i.e. corallinoids).  Confirmation of this result would require a higher sample size per crustose species and would ideally include a higher number of crustose species from both the clades of interest and 88  more distantly related groups. For example, species from Neogoniolithoideae (a potential sister subfamily to Corallinoideae, Bittner et al. 2011) could be investigated, and analyzing species from the secondarily-derived, crustose corallinoid genus Crusticorallina (Hind et al. 2016) would be particularly informative. Crusts used in this study were chosen for their supposed phylogenetic relationships with each articulated clade, based on Kato et al. (2011) and Bittner et al. (2011). More recent work, however, indicates that the phylogenetic position of the genus Spongites is not clear (Rosler 2016). Analyzing species that are more clearly closely related to metagoniolithoids, such as Porolithon spp., would also be re.    The results from the monosaccharide analysis, PCA, and linkage analysis all suggest major differences in the chemical composition of the genicula between articulated coralline clades. Along with potential differences in cellulose content, galactans varied in both content and structure. Genicula of articulated lithophylloids and metagoniolithoids were much more similar to one another, while corallinoid genicula were distinct. More detailed investigation into the structure of the galactans using a technique such as nuclear magnetic resonance (NMR) would confirm the assumptions made in this study regarding D and L conformation of the galactose units, which would in turn aid in the structural assignment of some of the unknown linkages found.  4.5 Conclusions Genicula of articulated corallines have all evolved to serve a similar function, but not without substantial variation on the overall theme. Genicular tissues display similar material properties to one another when compared to tissues of fleshy algal species, but with significant differences between articulated coralline clades (Ch. 1); similarly, the chemical composition of 89  genicula is universally similar when compared to that of crustose tissues, but there are key differences in both the content and structure of polysaccharides when comparing genicular tissues of different subfamilies. The data presented here suggest that corallinoid genicula contain more cellulose than other groups, while lithophylloid and metagoniolithoid genicula possess more highly branching galactans than what has been found in corallinoids. Furthermore, the genicula of Amphiroa gracilis and Metagoniolithon stelliferum #2 possess previously undocumented galactose linkages that are not present in Calliarthron tuberculosum, which could indicate further structural differences in xylogalactans between articulated clades or a completely different polysaccharide altogether. The results of this study further solidify the importance of looking at multiple levels of organization when investigating convergent evolution.   90  Chapter 5: Correlating cell wall composition and tissue mechanics in coralline genicula  5.1 Introduction Cell walls are a defining feature of land plants and macroalgae, and have evolved multiple times across a wide array of phylogenetically disparate organisms (Niklas 2004). Cell walls perform multiple structural functions, from regulating growth and cellular expansion, to maintaining cell shape and tissue cohesion (Popper 2008).  In red algal cell walls, cellulose typically forms a structural skeleton, while other polysaccharides such as sulfated galactans crosslink and surround cellulose in an amorphous matrix (Frei and Preston 1961, Usov 1992, Tsekos 1999, Kloareg and Quatrano 1988, Vreeland and Kloareg 2000, Newman and Davidson 2004). These components together act as a fibre composite material, with cellulose contributing primarily to stiffness and strength, and the matrix contributing to extensibility and toughness (Cave 1968, Kloareg and Quatrano 1988, Kohler and Spatz 2002, Geitmann 2010, Cosgrove and Jarvis 2012, Denny and King 2016). The relationship between mechanics and chemical composition of the cell wall has been studied in both terrestrial plants and algae, either by examining different population/life stages (Rees and Conway 1962, Stockton et al. 1980, Girault et al. 1997, Carrington et al. 2001, Genet et al. 2005, Starko et al. 2018) or by direct manipulation via enzyme extraction, genetic mutation, or growth experiments (Kraemer and Chapman 1991, Turner and Somerville 1997, Toole et al. 2002, Ryden et al. 2003, Mine and Okuda 2003, Mine and Okuda 2007). Cellulose content has been linked to tensile strength in terrestrial plants ranging from the diminutive Arabidopsis thaliana (Turner and Somerville 1997, 91  Ryden et al. 2003) to the woody tissue of trees, including Pinus pinaster and Castanea sativa (Genet et al. 2005). Pectic polysaccharides have also been shown to affect strength and stiffness in the cell walls of Arabidopsis and the freshwater green alga Chara corallina (Toole et al. 2002, Ryden et al. 2003). Biomechanical studies of marine algae have focused primarily on matrix polysaccharides, although Starko et al. (2018) suggested links between strength and cellulose content in the kelp Laminaria setchellii. Carrington et al. (2001) demonstrated a correlation between mechanical properties (strength, stiffness, and extensibility) and the type of carrageenan produced by different life stages of Chondrus crispus, while others have noted differences in the gelling strength of matrix polysaccharides found in wave-exposed and sheltered populations of both red algae (Pyropia perforata, Rees and Conway 1962) and brown algae (Alaria esculenta, Stockton et al. 1980).   While the amount of cellulose present in the cell wall has been shown to play a role in tissue strength, the orientation of cellulose within the cell wall is equally as important. Cellulose microfibrils in terrestrial plants are often arranged in a preferred orientation relative to the cell axis, and this orientation can depend on both cell type and cell wall layer (Sugimoto et al. 2000). Both the initial deposition of cellulose and subsequent passive reorientation caused by cellular growth will affect the ultimate orientation (Baskin et al. 1999, Geitmann 2010). Moreover, both tensile strength and stiffness of the cell wall are expected to be higher in the direction parallel to the net orientation of microfibrils, due to the relative stiffness of cellulose compared to the surrounding matrix (Cave 1968, Koehl and Wainwright 1977, Cave and Walker 1994, Baskin et al. 1999, Geitmann 2010).  At the cellular level this can lead to anisotropic growth, as the direction of maximal cell expansion occurs in the direction perpendicular to microfibril alignment (Green 1980, Taiz 1984, Baskin et al. 1999, Geitmann 2010). At the tissue level, 92  microfibrillar angle can affect overall strength and stiffness, as has been demonstrated in both terrestrial plants (Cave 1968, Cave and Walker 1994, Altaner and Jarvis 2008) and algae (Koehl and Wainwright 1977).  Microfibrillar angle can also affect material properties of tissue in a dynamic manner; biological tissues loaded in tension do not exhibit constant stiffness, and in plants and algae this may be explained by the reorientation of cellulose under tension (Kohler and Spatz 2002, Toole et al. 2004, Cosgrove and Jarvis 2012).  More specifically, an increase in stiffness with increasing strain could be attributed to the reorientation of cellulose microfibrils to an angle more parallel with the direction of the applied load, while decreases in stiffness could be attributed to sliding of microfibril aggregates past one another due to shear in the surrounding matrix (Kohler and Spatz 2002, Cosgrove and Jarvis 2012). This paradigm is supported by work done on Arabidopsis thaliana by Kohler and Spatz (2002), who demonstrated that microfibrillar angle decreases with increased strain until a transition to a lower stiffness occurs, after which microfibrillar angle remains constant. Toole et al. (2004) also demonstrated reorientation of cellulose microfibrils with increased strain in Chara corallina, although they did not measure mechanical properties. Articulated corallines possess genicula with a high degree of both strength and extensibility relative to other algal tissues (Martone and Denny 2008b, see Ch. 2).  Species within the subfamily Corallinoideae are particularly strong, at least twice as strong as other articulated corallines and 3-5x stronger than other red algae (see Ch. 2). Some of this strength can be attributed to the development of secondary cell walls, which acts as a mechanism for adding material to resist force without increasing overall size of the thallus (Martone 2007a). Accounting for this extra cell wall material explains roughly 84% of the variation in tissue 93  strength across articulated corallines, indicating that strength is largely affected by the quantity of the cell wall (see Fig. 2.8). The remaining differences between species are likely due to the material properties of polymers synthesized within the genicular tissue. Given the documented relationship between cellulose content and strength in land plants (Ryden et al. 2003, Genet et al. 2005), as well as the potential relationship between galactans and material properties in general (Rees and Conway 1962, Stockton et al. 1980, Carrington et al. 2001), cellulose and galactans were targeted for further investigation in this chapter. We hypothesized that cellulose content would play a role in the strength of genicular tissue, while galactan content would play a role in extensibility, stiffness, and toughness. As is the case with many red algal tissues, coralline genicula exhibit an increase in stiffness with increased strain (Hale 2001, Martone 2007a, Denny and King 2016). Genicular tissues undergo a second transition if strain continues, this time to a lower stiffness effectively generating an extended yield (Martone 2007a, personal observation). Scanning emission electron microscopy (SEM) images of the inner layer of the genicular cell wall in Calliarthron tuberculosum reveal a seemingly random orientation of a fibrillar network resembling cellulose (Martone et al., in prep); reorientation of this network could explain the initial increase in stiffness of coralline genicula with increased strain, while the subsequent yield in stiffness could be caused by breakage of either bonds or matrix polysaccharide links between microfibrils (Denny and King 2016). Along with investigating the role of galactan and cellulose content in determining mechanical properties, I examined the role of cellulose angle in the dynamic properties of genicular tissue with increasing strain. I hypothesized that net microfibril angle in the cell wall would become more parallel to the direction of applied tensile force as strain increased, until some minimum angle was reached. This minimum angle might be set by the 94  length of matrix polysaccharides or proteins linking the fibrils together, and so hitting this minimum angle should correspond with a decrease in stiffness.  5.2 Materials and methods  5.2.1 Correlating monosaccharides with material properties In order to gain a preliminary understanding of the relationship between chemistry and mechanical properties in genicular tissue, I compared species averages of genicular mechanical properties from Ch. 2 – strength (breaking stress, σ, MPa), extensibility (breaking strain, Ɛ, mm mm-1), initial stiffness (Young’s modulus, E, MPa), and toughness (breaking strain energy density, MJ mm-3) - with species averages of the most abundant monosaccharides (glucose, galactose, and xylose) from Ch. 4. Model selection was performed in R 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria) using the RStudio interface (version 0.98.1056, RStudio, Boston, MA, USA) and the glmulti() function from the glmulti package (Model selection and multimodel inference made easy,  version 1.0.7, Vincent Calcagno 2013). The glmulti() function tests all possible combinations for the parameters given, and selects the best model based on AICc scores. Only main effects were explored. In cases where the AICc scores were within 1 point of each other, the model with the lowest number of explanatory variables was selected. The resulting linear regressions were tested for significance using the anova() function from the base stats package in R. In some cases, mechanical data had to be log transformed for models to meet normality and homoscedasticity assumptions.   95  5.2.2 Specimen collection Calliarthron tuberculosum used for enzyme digestions were collected intertidally from Hopkins Marine Station (36˚37’13’’ N, 121˚54’15’’ W) in Monterey Bay, CA, USA, in December 2017. Fronds were shipped overnight to the University of British Columbia Vancouver campus, packed in paper towel dampened with seawater.  Calliarthron tuberculosum used for birefringence analysis and a second round of enzyme digestions were collected subtidally from a depth of ~10 ft. at Ogden Point (48˚25’03’’ N, 123˚23’10’’ W) in Victoria on Vancouver Island, BC, Canada, in April 2017 (for birefringence) and January 2018 (for enzyme extractions). These fronds were brought back to the University of British Columbia Vancouver campus the same day in a cooler, packed in dampened paper towel.  5.2.3 Enzyme digestions and tensile tests Segments of Calliarthron tuberculosum weighing 1-1.2 g were individually placed in 15 mL falcon tubes containing 10 mL of either seawater or 50 mmol sodium acetate/acetic acid buffer (pH = 5.0). 40 units of cellulase (Trichoderma reesei ATCC 26921, Sigma Aldrich, St. Louis, MO, USA) were added to half of the buffer treatments. Half of the tubes with seawater were kept chilled in a water table at ~12˚C, while half of the tubes with seawater and all of the tubes with buffer (with and without cellulase) were placed in a hot water bath kept at 50˚C for 24 hrs, for a total of 4 treatments: cold seawater, heated seawater, buffer, and buffer with cellulase. Temperature, time, and buffer pH for maximal enzyme activity were selected based on Harun and Danquah (2011). A second round of enzyme digestions was performed on Calliarthron tuberculosum collected from Vancouver Island. Along with a set of treatments kept at 50˚C for 96  24 hrs, a second set of treatments was kept at 50˚C for 48 hrs. Cold seawater controls were kept chilled in the water table at ~12˚C for 24 hrs and 48 hrs. After treatment, all tubes were placed in an ice bath and tested for mechanical properties of the genicular tissue using a standard tensile method in a computer-interface tensometer (model 5500R, Instron Corp., Canton, MA, USA). Segments were held in clamps lined with neoprene for cushioning and sandpaper to prevent slippage. The bottom clamp was immobile, while the upper clamp was attached to a crosshead that moved upwards at a rate of 10 mm min-1 until tissue failure occurred. Extension was measured continuously via movement of the crosshead, while force was measured continuously via a 500 N tension load cell. All data were collected and initially processed using Instron Bluehill 3 software (Instron Corp., Canton, MA, USA). Only segments that broke cleanly and away from the clamps were included in statistical analysis. Cross-sectional area of the broken genicular interface and cumulative length of all genicula that underwent extension were measured by dissecting tested samples under a dissecting microscope (model SZ61, Olympus Canada, ON, Canada) with an attached camera (model DP20, Olympus Canada). Stress (σ, MPa) was calculated by dividing force measurements by cross-sectional area, and strain (Ɛ, mm mm-1) was calculated by dividing extension by the initial cumulative genicular length. These calculations were done within the Bluehill software, and yielded stress-strain curves that could then be used to calculate total modulus (stiffness, measured as the average slope of the curve, E, MPa) and breaking toughness (area under the curve, MJ mm-3). Stress, strain, total modulus and toughness were compared between treatments in R 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria) using the RStudio interface (version 0.981056, RStudio, Boston, MA, USA). A one-way ANOVA was run on results from the first round of extractions, while a two-way ANOVA was run on results from the second round, incorporating 97  both treatment and time as fixed effects. Both ANOVA tests were run using the base stats package in R. 5.2.4 Quantification of extracted cellulose Two methods were used to try and quantify the amount of cellulose extracted from coralline tissue in the second round of enzyme digestions. To investigate how much cellulose was removed from each segment, I filtered samples of the treatment liquid from the 24 hr and 48 hr buffer and cellulase treatments through 0.4 μm filter tips, and examined them on a high performance liquid chromatograph (HPLC, Dionex DX-600, ThermoFisher Scientific, MA, USA) to measure the concentration of glucose. The HPLC was equipped with an ion exchange PA-1 column and a Dionex AS-50 autosampler (ThermoFisher Scientific, MA, USA). Samples were injected in 25 μL volumes onto the column, which was equilibrated with 250 mM NaOH. Samples were eluted with deionized H2O at a rate of 1 mL min-1, followed by post-column addition of 200 mM NaOH at 0.5 mL min-1. All samples were run in duplicate against standards containing known amounts of glucose, galactose, and xylose. An additional control of only buffer and cellulase (i.e. not having contained coralline tissue) was also run. To narrow down which tissue cellulose was being extracted from, segments from all 8 treatments were dissected into their genicular and intergenicular components. Samples from each treatment were pooled, to ensure sufficient tissue weights for the following analysis. Tissues were prepared and hydrolyzed using the same methodology used in Ch. 4, using a scaled down version of the klason procedure described in Huntley et al. (2003). In brief, both genicula and intergenicula were ground using a mortar and pestle, after which intergenicular samples were decalcified via dropwise addition of 1 M HCl and constant stirring until no more bubbling was detected. Decalcified tissues were centrifuged at 10,000 RPM for 5 min, and the supernatant was 98  removed and discarded. All tissues were dried in an oven at 50˚C for 48 hrs; intergenicular samples were reground after drying, then placed back in the oven for a minimum of one hour prior to acid hydrolysis. Samples of ~1 mg of genicular tissue or ~10 mg of intergenicular tissue were weighed out into 2 mL graduated, free-standing microcentrifuge tubes, and either 50 μL (genicula) or 100 μL (intergenicula) of 72% H2SO4 was added. Samples were vortexed briefly, then incubated in a heating block at 30˚C for 1hr while being shaken at 500 RPM. 1 mL (for 10 mg samples) or 2 mL (for 1 mg samples) of deionized H2O was added, and samples were vortexed briefly to resuspend tissue. Samples were then autoclaved at 120˚C for 90 min, allowed to cool, and centrifuged at 13,000 RPM for 10 min. The supernatant was separated from the remaining pellet, then filtered through 0.45 μm filter tips. Supernatants were run on the same HPLC and using the same protocol as the treatment liquid samples. All samples were run in duplicate, against standards containing known amounts of glucose, galactose, and xylose.  5.2.5 Polarized microscopy on corallines in tension Crystalline cellulose is well known to possess strong birefringence, meaning that light passing through cellulose microfibrils is refracted differently dependent on both the polarization and direction of propagation of the light (Preston 1933, Iyer et al. 1968). This is the result of a combination of the biaxial, anisotropic structure of cellulose monocrystals, as well as the organization of those crystals within the microfibril (Iyer et al. 1968). The optical properties of birefringent materials can be described by two refractive indices, no and ne. Light that is linearly polarized in a direction parallel with the optical axis of a birefringent material experiences a single refractive index, no. Conversely, light that is polarized in a direction perpendicular optical axis breaks into an “ordinary ray” and an “extraordinary ray,” which oscillate perpendicular and 99  parallel to the optical axis, respectively. The ordinary ray experiences refractive index no, while the extraordinary ray experiences refractive index ne. Light that is polarized in any other direction will be broken into an ordinary and extraordinary ray, with the ordinary ray always experiencing no and the extraordinary ray experiencing n’e, which will have a value somewhere between no and ne depending on the polarization direction. In cellulose, the organization of the monocrystals within the microfibril is such that the optical axis is parallel to the long axis of the fibril (Preston 1933, Iyer et al. 1968, Abraham and Elbaum 2013). This conveniently allows one to utilize polarized microscopy to determine the direction of the optical axis, which in turn corresponds to the angle of the cellulose microfibrils within the tissue.  For this study, I measured birefringence of Calliarthron tuberculosum genicula using the LC-PolScope image processing system (CRi Inc., Woburn, MA, USA) mounted on an inverted microscope (Nikon Eclipse TE2000-U, Tokyo, Japan). For a detailed breakdown of the LC-PolScope system, see Oldenbourg (2007). In short, the system includes a computerized universal compensator which can rotate to a series of preset positions, polarizing light in different directions relative to the specimen. The system calculates retardance of light travelling through the specimen at each position, as well as the angle of maximum retardation (the “retardation azimuth”, Oldenbourg 2007, Eder et al. 2010, Abraham and Elbaum 2013, Hu et al. 2017). These data can then be used to estimate the angle of the optical axis, and therefore the microfibril angle. Segments of Calliarthron tuberculosum were longitudinally sectioned using an Isomet Low Speed Precision Cutter (Buehler, Illinois Tool Works Inc., Lake Bluff, IL, USA) equipped with two diamond wafering blades spaced 0.3 mm apart. Segments were mounted on glass slides using Crystalbond 509 mounting adhesive (Structure Probe Inc., West Chester, PA, USA), and the slides were then mounted in a precision vise perpendicular to the blades. The vise-mounted 100  slide was then slowly lowered onto the rotating wafering blades, which cut sections 0.3 mm thick from the centre of the longitudinally oriented coralline segment. Intergenicula on either side of longitudinal sections were attached to separate glass coverslips using Krazy Glue (Elmers Products, High Point, NC, USA) and an adhesive accelerator (Loctite 712 Tak Pak Accelerator, Henkel Adhesives North America, Westlake, OH, USA), so that segments were floating in between two coverslips (Fig. 5.1). Dental floss was glued to the opposite side of each coverslip and tied at one end to a 2.5 N beam force transducer (model FORT250, World Precision Instruments, Sarasota, FL, USA) and at the other end to a micrometer (non-rotating spindle head, WPI Model #502102). A dial on the micrometer was manually turned in small increments so that extension occurred on one side, while the force was read on the other via the force transducer. Mounted segments were stretched while placed on a large coverslip taped above the microscope stage opening with the objective underneath. Force measured by the transducer was amplified through a transducer amplifier (model SYS-TBM4M, World Precision Instruments, Sarasota, FL, USA) and collected in real time using LabVIEW Signal Express software (National Instruments Canada, Vaudreuil-Dorion, QC, Canada). Images were taken with a CCD camera (Nikon Digital Sight DS-Fi2, Tokyo, Japan) using Abrio software tools (CRi Inc., Woburn, MA, USA) after each increase in extension, and this process continued until the genicular tissue failed. The genicula being imaged were hydrated periodically with seawater using a pipette. 101   Fig. 5.1. Diagram of the method used for tensile tests with an inverted polarized microscope. Pink represents intergenicula and yellow represents genicula of a longitudinal section of Calliarthron tuberculosum. Blue indicates placement of glue to attach ends of coralline segments or dental floss to coverslips. This arrangement was placed on top of a large coverslip taped over the stage opening and microscope objective. Force was read on one end with a force transducer and applied on the other with a micrometer.    Force measurements after each manual extension were converted into stress (MPa) by dividing force by the original cross-sectional area of the geniculum. This was calculated by measuring the width of the geniculum in the first (pre-stretch) photo taken with the CCD camera using ImageJ (NIH Image, http://rsb.info.nih.gov/ij), then multiplying by the section thickness (0.3 mm). Strain was calculated as the change in length divided by the initial length; length measurements at each extension were also measured from the CCD camera photos using ImageJ. These data were used to generate stress-strain curves from which modulus (MPa) could be measured, using the slope between two points. Retardation values for light polarized in a direction parallel to the cell axis were extracted from photos using the Abrio software. Measurements were taken along the entire length of genicula, and values from the edge (i.e. near adjacent intergenicula) and the middle were recorded for further analysis.    102  5.2.6 Estimating microfibril angle Microfibril angle relative to the cell axis can be estimated from retardance measurements. Retardation (R) is equal to the birefringence relative to the thickness of the birefringent material: R = B ∗ t Eq. 5.1 Where B is the birefringence and t equals the thickness of the cellulose within the tissue rather than the total thickness of the genicular tissue (Abraham and Elbaum 2013, Hu et al. 2017). Several assumptions had to be made in the calculation of cellulose thickness (see Appendix C). Birefringence refers to the difference between the speed of the ordinary and extraordinary rays of polarized light as it propagates through birefringent material: B = no − n′e Eq. 5.2 And so, one can use the calculated birefringence extracted from retardance values, as well as published values of no (due to lack of values published for algae, I use the value for ramie fibres, 1.599, Iyer et al. 1968) to calculate n’e. The value of n’e is dependent on the angle of incidence, that is, the angle between the extraordinary ray and the retardation azimuth (Abraham and Elbaum 2013). This means n’e can be used to calculate angle of incidence (θi) by using the formula for a horizontal ellipse (Preston 1933, Abraham and Elbaum 2013), where no is the major axis, ne is the minor axis, and n’e corresponds to a radius at some angle between the two: n′e = none�no2cos2θi + ne2sin2θi Eq. 5.3 This equation can be rearranged to calculate θi: cos2θi =  no2ne2(no2 − ne2)n′e2 − no2no2 − ne2 Eq. 5.4 103  The microfibril angle corresponds to the angle between the extraordinary ray and the optical axis. As the optical axis and retardation azimuth are perpendicular to one another, microfibril angle is equal to 90˚-θi (Preston 1933, Abraham and Elbaum 2013). Calculations of effective thickness rely on an estimate of cellulose content (see Appendix C). In this study, I estimated cellulose content based on the average glucose content found in Calliarthron tuberculosum genicula in Ch. 4. This estimate (6.23% dry weight) was low compared to previous work, so I also calculated microfibril angle based on a 15% dry weight glucose content estimate from Martone (2007b). Microfibril angles were calculated from pre-rip retardation values of trials 1-5 using both cellulose content estimates and graphed against strain for comparison.  5.3 Results  5.3.1 Correlating monosaccharides with material properties Glucose content was positively correlated with the log of breaking stress (Fig. 5.2A, ANOVA, p<0.01, R2=0.74), the log of Young’s modulus (ANOVA, p=0.04, R2=0.45), and the log of toughness (ANOVA, p=0.01, R2=0.60). Corallinoid species exhibited both the highest strength and the highest glucose content, while metagoniolithoids had the lowest strength and glucose content (Fig. 5.2A). A positive correlation between galactose content and breaking strain approached significance (Fig. 5.2B, ANOVA, p=0.08, R2=0.33). Lithophylloid species had the highest galactose content, though not necessarily the highest breaking strain (Metagoniolithon chara was the most extensible, with a strain of 1.07) (Fig. 5.2B). 104   Fig. 5.2. The correlation between (A) glucose content and tissue breaking stress, and (B) galactose content and tissue breaking strain in genicular tissue of articulated corallines. Points represent species averages and error bars represent s.e.m. Material properties and monosaccharide content were measured on different sample sets. Corallinoid species are show in shades of green, lithophylloid species are shown in shades of purple, and metagoniolithoids are shown in shades of orange. Black dotted lines represent lines of best fit (A: y=2.05x+0.58, B: y=0.06x+0.33).  5.3.2 Enzyme digestions and tensile tests Treatment with enzymes to digest cellulose in the cell wall did not have a significant effect on breaking strain (Fig. 5.3B, ANOVA, p=0.49), total modulus (Fig. 5.3C, ANOVA, p=0.31), or toughness (Fig. 5.3D, ANOVA, p=0.21) in the first round of trials. The effect of treatment on breaking stress approached significance (Fig. 5.3A, ANOVA, p=0.06), which gave incentive for a second round of digestions with a higher sample size.  105   Fig. 5.3. Effect of 24 hrs of soaking in different solutions on (A) breaking stress (MPa), (B) breaking strain, (C) total modulus (MPa), and (D) breaking toughness (MPa) of genicula in Calliarthron tuberculosum. Numbers within each column in panel A indicate sample sizes for each treatment. Error bars represent s.e.m.   The second round of enzymatic digestions also failed to yield any significant effect of treatment on mechanical properties (Fig. 5.4). Only a significant effect of time was found, on both breaking strain (two-way ANOVA, p<0.01) and total modulus (two-way ANOVA, p<0.01); strain generally increased with an increase in time (Fig. 5.4B), while total modulus decreased (Fig. 5.4C). No significant interaction between treatment and time was found.  106   Fig. 5.4. Effect of 24 hrs (dark grey) and 48 hrs (light grey) of soaking in different solutions on (A) breaking stress (MPa), (B) breaking strain, (C) total modulus (MPa), and (D) breaking toughness (MPa) of genicula in Calliarthron tuberculosum. Numbers within each column in panel A indicate sample sizes for each treatment. Error bars represent s.e.m.   HPLC analysis of treatment liquid indicated glucose peaks in the cellulase treatments; however, the cellulase and buffer control also contained a glucose peak of a comparable size, indicating the presence of glucose within the enzyme mixture itself. For this reason, I was unable to estimate the amount of glucose released from tissues into the treatment liquid; it is likely that any glucose signal that resulted from cellulose degradation was masked out by the glucose already present from the enzyme. This was further complicated by the fact that any glucose coming from the tissue would have been heavily diluted in the 10 mL of buffer.  Similarly, 107  HPLC analysis of pooled samples of intergenicular and genicular tissue did not indicate lower glucose levels in cellulase treatments when compared to buffer treatments (Fig. 5.5).  Fig. 5.5. Glucose content (% dry weight) from (A) intergenicular tissue and (B) genicular tissue soaked in different solutions for 24 hrs (dark grey) and 48 hrs (light grey). Results are from pooled samples of individuals from each treatment.   5.3.3 Microfibril angle of corallines in tension Reliable mechanical and optical data were obtained in 5 of the 15 attempted trials. For these five, breaking stress was highly variable, between 17 and 50 MPa. Breaking strain was low, between 0.3 and 0.6. Modulus generally increased with an increase in strain (Fig. 5.6A-E), but no subsequent yield was observed.    108   Fig. 5.6. Stress, strain, and estimated microfibril angle of genicula in tension; vertically stacked plots indicate the same trial. (A-E) Stress-strain curves of tensile tests under polarized light microscopy. (F-J) The change in estimate microfibril angle with increased strain: black squares indicate measurements taken from the middle of the geniculum, while grey squares indicate measurements taken from the edge. (K-O) The relationship between microfibril angle and modulus of the stress-stress strain curve.   Estimated microfibril angle was between 70-85˚ relative to the longitudinal axis of the genicular cells at zero strain and decreased initially with an increase in strain (Fig. 5.6F-J). In four of the five successful trials, fraying of the genicular cells was observed early on; the comparison can be seen in Fig. 5.7, which shows a geniculum where all tissue is in tension, and Fig. 5.8, which shows the increased tearing of a geniculum that began to fray early in the tension 109  test. In some cases, estimated microfibril angle increased in the regions affected by the tear (trials 3-5, Fig. 5.6H-J). In cases where a tear continued to develop for an extended time before total tissue failure, estimated microfibril angle continued to increase. This was most evident in trial 5 (Fig. 5.6J). For four of the five trials, lower microfibril angle in the centre of the geniculum appeared to correlate with a higher modulus (Fig. 5.6K-N). This was not the case for trial 5 (Fig. 5.5O). Modulus was highly variable overall, with maximum moduli ranging between 67 and 528 MPa. No pattern between specific microfibril angle and modulus values was observed (Fig. 5.5K-O). The cellulose content estimate used for calculating effective thickness had a minimal effect on calculations of microfibril angle (Fig. 5.9). More than doubling the estimate from 6.23% dry weight to 15% dry weight only changed calculated microfibril angle 2-3 degrees. If the strain birefringence in genicula is the result of cellulose realignment alone, then calculations of microfibril angle used here are a reasonable estimate.       110   Fig. 5.7. Longitudinal section of a geniculum under increasing strain (left to right), under different imaging modes of the Abrio software. (A-D) Grayscale image processing of the retardance range 0-273 nm. Lighter gray indicates higher retardance parallel to the longitudinal axis of the geniculum. (E-F) Colour-encoded image processing of the retardance range 0-273 nm. Blue indicates lower retardance in the longitudinal axis of the geniculum, red indicates higher retardance. (I-J) Colour-encoded orientation of the retardation azimuth, which is perpendicular to the optical axis of the microfibrils. Red indicates the retardation azimuth is parallel to the longitudinal axis of the geniculum, while green indicates it is perpendicular.  111   Fig. 5.8. Longitudinal section of a geniculum under increasing strain (left to right), under three different imaging modes of the Abrio software. (A-D) Grayscale image processing of the retardance range 0-273 nm. Lighter gray indicates higher retardance parallel to the longitudinal axis of the geniculum. (E-F) Colour-encoded image processing of the retardance range 0-273 nm. Blue indicates lower retardance in the longitudinal axis of the geniculum, red indicates higher retardance. (I-J) Colour-encoded orientation of the retardation azimuth, which is perpendicular to the optical axis of the microfibrils. Red indicates the retardation azimuth is parallel to the longitudinal axis of the geniculum, while green indicates it is perpendicular. The beginning of a tear is visible in B, F, and J, and it propagates further with increased strain. 112   Fig. 5.9. Relationship between estimated microfibril angle (˚) and strain (Ɛ, mm/mm) when cellulose content is estimated at 6% dry weight (black circles) and 15% dry weight (grey circles). Data points are calculated from all pre-rip retardance values from trials 1-5.   5.4 Discussion The results presented here indicate that the relationship between chemical composition and mechanical properties is not straightforward. When comparing articulated species to one another, there is a strong correlation between glucose content and strength (Fig. 5.2A), stiffness, and toughness of genicular tissue. This suggests that cellulose has a structural role in resisting tensile forces. Unfortunately, enzyme extractions on Calliarthron tuberculosum were unable to verify this effect. While an initial round of testing with a low sample size indicated a possible relationship between cellulose and strength (Fig 5.3A), more rigorous tests could not confirm this (Fig. 5.4). HPLC results on the treatment liquid from the second round of testing were not useful in determining successful extraction of cellulose with the cellulase treatment, due to the presence of glucose within the enzyme formula. HPLC results on pooled tissue samples did not 113  show lower glucose levels for cellulase treated fronds, in either intergenicular or genicular tissues (Fig. 5.5). There are several possible explanations for these results. Based on HPLC data, it is possible that the cellulase did not successfully access cellulose, at least in the second round of trials. It was initially expected that cellulase would, at a minimum, degrade cellulose in the easily accessed intergenicular tissues; it is possible, however, that calcium carbonate blocks enzyme access to the cellulose network. Genicular tissue may also be difficult to access, although for a different reason. Most of the coralline surface area is composed of intergenicula, which take up most of the volume of the whole coralline frond (~93%, see Appendix C). Genicular surface area is more limited and is shielded by calcified overhangs from adjacent intergenicula. Additionally, glucose content in the cell wall is not overly high (6-15% dry weight, Ch. 4 and Martone 2007b), so any cellulose present is likely heavily surrounded by matrix polysaccharides and proteins, which could further inhibit enzyme access. The final possibility is that cellulose does not play a direct role in the strength of genicular tissue, despite the observed correlation. If cellulose acts as a scaffold for other components of the cell wall, a greater proportion of cellulose at the species level could correspond to a more organized matrix network, which could in turn affect mechanical properties. Disrupting cellulose after the rest of the network has been laid down may not be that disruptive to the tissue overall.  Further enzymatic investigation into the role of cellulose in the cell wall would involve several adjustments to the methods used here. Genicula could be sectioned prior to treatment to remove intergenicular overhangs, thereby giving enzymes more access to the target substrate. Higher enzyme concentrations could also be used. A major missing aspect of this study was enzymatic degradation of matrix polysaccharides – an enzyme that targets either β-1,3- or α-1,4-114  galactose linkages could be used to degrade agarans, the most abundant matrix polysaccharide found in corallines (Turvey and Simpson 1966, Cases et al. 1992, 1994, Usov et al. 1995, Takano et al. 1996, Navarro and Stortz 2002, Martone et al. 2010, Navarro et al. 2011).  A proteinase that functions in acidic conditions (pH of ~5.0) would also be an informative addition to this study. Neither of these enzymes were able to be sourced at the time of writing, but they may be available for future work. Degrading matrix polysaccharides and proteins would not only give insight into the role of these cell wall components on mechanical properties, it could also aid in making the cellulose network more vulnerable to enzymatic degradation. While I was unable to properly elucidate the role of cellulose content on tissue properties, I was able to gain some insight into the way the cellulose network behaves under tensile forces. Using retardance under polarized microscopy as a proxy for microfibril angle, I found that microfibril angle initially decreases when force is applied (Fig. 5.6K-O). In cases where a subsequent increase was observed, the start of that increase corresponded to the appearance of tearing in the tissue (for an example, see Fig. 5.8). The increase typically manifested first at the edge of the tissue, and in some cases appeared slightly afterwards in tissue at the middle. Cells appeared to fail in clusters and recoil slightly upon failure, so that measurements at the middle of the tissue continued to only include cells in tension while measurements at the edge included both cells in tension and cells in recoil. Though it could not be observed directly, cases where microfibril angle also began to rise in middle tissue could be the result of tearing and delamination on the opposite side of the longitudinal section. The rapid increase in estimated microfibril angle observed during tearing suggests a high degree of elasticity in the cellulose network, which may also be reflected in the high degree of fatigue resistance (Denny et al. 2013, Denny and King 2016). 115  In general, lower microfibril angle in the middle of the geniculum corresponded to increased tissue stiffness. Trial 5 was an exception, with no clear relationship observed. This could be an artefact of the way in which retardance data were collected; most rips that occurred started from the centre of the geniculum, so that only edge measurements included broken cells. The rip in trial 5 occurred near one edge of the geniculum, with cells appearing to pop out of the calcified intergenicular tissue completely. This meant that broken cells masked the birefringence signal of cells in tension for most of the length of the geniculum, including the middle. In other words, the tissue affecting material properties (only intact cells) was different than the tissue for which microfibril angle was being calculated (both intact and broken cells).  There was no clear relationship between calculated microfibril angle and a specific value of tissue stiffness; while all trials had a similar microfibril angle range, the range of stiffness varied greatly between trials. This may be the result of variability in width measurements of thin sections skewing stress and stiffness calculations. It was assumed that all sections were 0.3 mm in width at the start of the test, as this was the distance between blades in the diamond saw that was used for sectioning, but this may not have been a reasonable assumption. While the diamond saw was a useful method for sectioning corallines without prior decalcification, it was not particularly precise – cells near the outside of the section could get caught in the blade and fray, causing sections that were functionally <0.3 mm in width. Additionally, sections sometimes appeared to be wider on one side than the other. Variation in section width would cause a variation in cross-sectional area that was not accounted for, which in turn would lead to error in stress and stiffness calculations. An assumption was also made regarding cellulose content, namely that it was similar for all trials, and variation in cellulose content could cause minor variation in microfibril angle. However, the value for estimated cellulose content does not cause 116  microfibril angle to change more than 2-3 degrees (Fig. 5.9).  Thus, the cause of the disconnect between microfibril angle and stiffness is likely due to error in morphological measurements, and minimizing this error should be a focus of future work.  I did not observe a yield in stiffness in the stress-strain curve, and so I was unable to test the hypothesis that microfibril angle would hit some minimum and then remain constant as the surrounding matrix was sheared. Instead, stiffness continued to climb with increased strain until the geniculum failed, at a lower extensibility than what has been observed for whole genicula (Ch. 2). This may be indicative of the genicular tissue being insufficiently hydrated, causing it to become more brittle. The thin sections utilized in this experiment likely dried out faster than the whole genicula used in previous experiments (i.e. Ch. 2), and they are therefore not shielded by intergenicular overhangs that might otherwise trap water. Furthermore, tensile tests performed under the microscope took much longer (up to 20 min per trial) than those performed in Ch. 2, due to the need to periodically pause the test and take pictures. Kamiyama et al. (2005) observed a similar trend in comparisons between hydrated and dry wood; where hydrated wood experienced a sharp decrease in microfibril angle and a yield in stiffness, dried wood experienced a lesser decrease in microfibril angle and no yield prior to complete tissue failure.  The discrepancy between the behaviour of whole genicula and the longitudinal sections tested here could also be partially caused by the fraying behaviour observed under the microscope. While I initially hypothesized that shifts in stiffness in the stress-strain curve were the result of micromechanical factors within the cell wall, it was assumed that the whole tissue experiences and reacts to tensile forces equally. Tears on the surface of the geniculum occurred much sooner than expected, demonstrating that the whole tissue may fail in stages as cells fray from the outside of the geniculum first (see also Martone and Denny 2008b). Fraying likely plays 117  a role in the high strength and fatigue-resistance of corallinoid genicula; weak lateral connections between cells prevents the propagation of cracks throughout the whole tissue, as energy is dissipated in the break of individual cells rather than being transferred to the adjacent ones (Denny et al. 2013). This decrease can still come at a price, however, as the cells that remain intact must now resist a higher proportion of the tensile load. In the case of thin sections, the effect of one cell fraying is proportionally higher than it is for a whole geniculum. Losing larger amounts of area at a time could cause sections to fail more abruptly (i.e. without an extended yield) than what is observed in whole genicula. To tease apart the effects of hydration and fraying, two experiments could be performed. The first would involve tensile tests on whole genicula at varying levels of hydration, while the second would involve tensile tests on longitudinal sections of varying thicknesses. Looking for changes in the presence/absence or degree of yield in the stress-strain curve would determine whether either factor was at play for this study. Either result would contribute insight into the mechanical behaviour of genicular tissue. Effects of hydration on stiffness could indicate a substantial role of the matrix, as the gel formed by polysaccharides such as galactans would be the first to become hydrated. Effects of fraying would suggest that the composition of the middle lamellae governing cell-to-cell cohesion is equally as important as the composition of the cell wall - this aspect of tissue properties is often overlooked. To properly isolate the effect of the cellulose network on the stress-strain curve would be more difficult, as it would necessitate performing tensile tests or other mechanical manipulations on an individual cell.  118  5.5 Conclusions In this study, I found initial support for the hypothesis that cellulose plays a role in strength of genicular tissues. By comparing the chemistry and mechanical properties of multiple articulated coralline species, I found a strong correlation between glucose and strength, as well as stiffness and toughness. I was not able to confirm this relationship with enzyme digestion tests, and further investigation is needed to determine the efficacy of our methods. I found partial support for the hypothesis that microfibril angle is connected to stiffness, as increased stiffness often corresponded with a decrease in microfibril angle. I could not test the second part of this hypothesis, that yield in the stress-strain curve would correspond to microfibril angle ceasing to change as the surrounding matrix underwent shear. To do so, it will be necessary to devise a way of performing tensile tests on single cells under polarized microscopy.    119  Chapter 6: Conclusions  6.1 Major findings of this work With this thesis, I aimed to gain a more integrative understanding of the convergent evolution of articulated corallines by using a variety of techniques targeting genicula at multiple levels of organization. By comparing both form and function of genicula across three articulated coralline clades, I demonstrate that the similarity between these groups is actually fairly shallow. Genicula convey the advantage of flexibility to upright coralline forms in Corallinoideae, Lithophylloideae, and Metagoniolithoideae, and they are composed of tissues that are strong and extensible when compared to other fleshy algae (Ch. 2). However, genicula in the three subfamilies are not mechanically equivalent, due to significantly different morphologies that lead to differing levels of bending stress (Ch. 3). Furthermore, material properties between articulated clades are significantly different, albeit in ways that offset the mechanical consequences of differing morphologies (Ch. 2). Differences in material properties are caused by both quantity and quality of cell walls, where quantity is affected by the presence or absence of secondary cell walls and quality is affected by chemical composition (Ch. 2 and Ch. 4). Corallinoids possess a high number of relatively short genicula, which leads to an overall decrease in thallus calcification but also makes this group susceptible to high bending stress (Ch. 3). Perhaps consequently, corallinoids also possess genicular tissues with significantly higher material strength compared to other articulated corallines, which allow them to cope with high bending stress (Ch. 2 and Ch. 3). Part of this tissue strength is caused by the presence of secondary cell walls, which do not appear to be present in either lithophylloid or metagoniolithoid species (Ch. 2). High cellulose content of the cell wall may also play a role in 120  the high strength of corallinoid species (Ch. 4). These findings are similar to what has been found in Calliarthron cheilosporioides, a corallinoid species that exhibits high strength and extensibility (Martone and Denny 2008b), thick secondary cell walls (Martone 2007b), and high cellulose content relative to other red algae (Martone 2007a).  Lithophylloids possess long genicula, but they also possess particularly long intergenicula, so that the calcification level of the total thallus is much higher in this group compared to either corallinoids or metagoniolithoids (Ch. 3). To compensate for what could be considered a relatively inflexible morphology, lithophylloid genicula have a relatively low material stiffness, which allows lithophylloid species to attain high degrees of flexibility that are comparable to other groups (Ch. 2 and Ch. 3). Two of the three lithophylloid species tested (Amphiroa anceps and Amphiroa gracilis) were also highly extensible compared to many other articulated coralline species, which could be due, at least partially, to high galactan content (Ch. 2 and Ch. 4). Galactans in lithophylloids may also be highly branched in comparison to those found in corallinoids, though the mechanical significance of this finding is unknown (Ch. 4).  This work is the first to use biomechanical techniques on articulated lithophylloid species and the first to investigate chemical composition of genicular tissue specifically, but additional work on other lithophylloid species is required to determine if results presented here are generalizable. Metagoniolithoid genicula were the least impressive at the tissue level, though they still had strength and extensibility comparable to some of the strongest and most extensible fleshy red algal species investigated previously (Ch. 2, and see Hale 2001). Genicula in this group were universally longer than those of either corallinoid or lithophylloid species, and intergenicula were so short as to be functionally nonexistent in some cases (Metagoniolithon stelliferum #1, see Ch. 3). The chemical composition of metagoniolithoid genicula was similar to that of 121  lithophylloid genicula in both monosaccharide content and polysaccharide structure, although metagoniolithoids were the only subfamily to completely lack mannose (Ch. 4). Histological analysis, however, revealed a dominant extracellular matrix that was not present in either corallinoids or lithophylloids (Ch. 2). The role of this matrix in total tissue properties is unknown.  The fact that metagoniolithoids are minimally calcified compared to other articulated corallines may explain why the tissue properties were more in line with fleshy algal species. If most of the thallus has some level of flexibility, there is no need for the tissue to be unusually strong or extensible to compensate for lack of flexibility elsewhere. This clade may have also evolved in response to different evolutionary pressures compared to the corallinoids and lithophylloids: almost all species (except for Metagoniolithon radiatum, not tested in this study) grow epiphytically on seagrass (Ducker 1979), so the type of hydrodynamic stress they experience is likely to be fundamentally different from what is experienced by epilithic corallinoids and lithophylloids. Drag may be dampened and reduced by living epiphytically (Anderson and Martone 2014), lessening the necessity for particularly strong or extensible materials. Additionally, high strength is not advantageous for epiphytes if the host is likely to break first (Anderson and Martone 2014). As with articulated lithophylloids, metagoniolithoids have not been previously studied in terms of biomechanics and chemical composition. To determine which factors are the result of ancestry versus environmental selection would require including the epilithic Metagoniolithon radiatum.  When all organizational levels are considered, each articulated clade seems to have a different way of attaining flexibility. Corallinoids possess many genicula made of particularly strong tissue, which is bolstered by the deposition of a secondary cell wall and relatively high 122  cellulose content. Lithophylloids possess a less flexible morphology, but they have highly pliable genicular tissues with high galactan content. Metagoniolithoids possess long genicula with middling material properties, but their microhabitat is such that they may avoid high hydrodynamic stress altogether. This suggests that genicula in the three subfamilies are convergent in function but not exactly in form. Furthermore, the reasons behind the evolution of genicula in each clade may differ. As Losos (2011) notes, convergence is not necessarily indicative of adaptation; similarly, any adaptive significance that does exist does not have to be the same across each evolutionary iteration. Genicula could convey flexibility to corallinoids and lithophylloids, while having less overall calcification could decrease total thallus weight in metagoniolithoids, for example.  Along with subfamily-level patterns in the mechanical behaviour and chemical composition of genicular tissues, some interesting chemical patterns were found in calcified tissues (i.e. intergenicula and crusts). Despite clear differences in genicular chemistry between articulated clades, calcified tissues were much more similar. Both intergenicular tissue from articulated species and calcified tissue from crustose coralline species were found to be chemically similar to one another, which could suggest that little has changed in calcifying tissues over evolutionary time. It also suggests that the chemical building blocks available for the evolution of genicula in different articulated coralline clades were likely similar. My study indicates that a similar starting point - crustose, prostrate morphologies with similar chemical composition – has led to three articulated coralline clades that are functionally convergent but phenotypically divergent. That is, evolution has given rise to flexible joints in three distinct ways. Given that prior work has documented both structural and developmental differences in genicula (Johansen 1969, 1981, Ducker 1979, Woelkerling 1988), the differences 123  in morphology and chemical composition are perhaps not surprising. What is notable is the way in which morphology and material properties vary synergistically to produce similar mechanical outcomes overall, especially when comparing corallinoids and articulated lithophylloids. This synergy suggests that the hydrodynamically stressful conditions of wave-swept environments have played a significant role in the evolution of genicula in most articulated coralline algae.  6.2 Future directions  6.2.1 Biomechanical testing This study identified several clade-specific patterns in genicular morphology and material properties, but studies on additional species would help bolster these conclusions. Species with atypical habitats should be targeted, such as the epilithic Metagoniolithon radiatum, or epiphytic species within Corallinoideae (e.g. Jania spp., particularly Jania rosea which grows on the seagrass Amphibolus spp., Joll and Phillips 1984) and Lithophylloideae (e.g. Amphiroa rigida, which can grow on the seagrass Posidonia oceanica, Brahim et al. 2014). This would help differentiate between the effects of ancestry and hydrodynamic environment on genicular properties. For example, finding that M. radiatum has strong tissues like epilithic corallinoids (and unlike epiphytic metagoniolithoids) would demonstrate that hydrodynamic environment likely plays a larger role in genicular properties than phylogenetic history.  However, if epiphytic corallinoids have strong tissues like epilithic corallinoids, it could be concluded that phylogenetic history plays a larger role given the distinctive hydrodynamic environment.   The strongest species tested in Ch. 2, Cheilosporum sagittatum, was not available for further work, but flexibility and morphological analysis of this species would also be 124  informative. Given the high material strength of C. sagittatum, it would be interesting to see if the morphology of this species incurs a particularly high bending stress, requiring a particularly high strength to resist it. Lithothrix aspergillum was another species that was only tested for material properties in Ch. 2 - this species represents a fourth type of genicular structure development that is quite distinct from that of other lithophylloids (see details in Johansen 1981). Genicula of L. aspergillum superficially appear more similar to corallinoid genicula, so perhaps the bending strategy of this species would be similar to corallinoids as well.  6.2.2 Chemical analysis This study was the first to investigate the chemical composition of genicular tissue and intergenicular tissues separately in lithophylloids, and the first to investigate the chemical composition of metagoniolithoids at all. While it provides an initial comparison of monosaccharide composition, linkage analysis was only performed on a select number of species, and then only on genicular tissues. Further work should include linkage analysis on intergenicular tissues from each articulated clade as well as crusts, to determine whether calcified tissues really are as similar as the monosaccharide data suggests. Previous work has documented the structure of agarans (a sulfated galactan) in a variety of coralline species (Turvey and Simpson 1966, Cases et al. 1992, 1994, Usov et al. 1995, Takano et al. 1996, Usov and Bilan 1998, Navarro and Stortz 2002, 2008, Navarro et al. 2011), and it is likely that the results of these studies are mostly representative of calcified tissues due to the high proportion of intergenicular to genicular tissue in articulated species. The species included in these studies were limited to those from Corallinoideae and Melobesioideae, however, so no in-depth analysis of the agarans in intergenicular tissue of lithophylloids and metagoniolithoids has been done. 125  More crustose species from the Lithophylloideae and Porolithoideae should be included for a more robust comparison of intergenicular and crustose tissues. Including species from other morphologies, such as rhodoliths (Neogoniolithoideae), or oddities such as the secondarily-derived crustose corallinoid genus Crusticorallina (Hind et al. 2016) would help to further discern whether morphology has any role in calcified tissue chemistry. Notably absent from chemical analysis was any data from Cheilosporum sagittatum. The reason for this was twofold: (1) tissue availability of this diminutive Australian species was limited, and (2) separating genicular and intergenicular tissue of this species was prohibitively difficult (genicula were approximately 220 μm in diameter and 200 μm in length). Regardless, finding a method of tissue separation to analyze the impressively strong genicular tissue of this species would be hugely informative, particularly regarding the potential role of cellulose in material strength. The possible lack of secondary walls suggests that this species may gain strength from its unique chemical composition. The developmentally atypical Lithothrix aspergillum would also be a good candidate for further chemical analysis of genicular tissue. Several assumptions were made in Ch. 4 regarding the structure of galactans. Many of these assumptions could be tested with the use of nuclear magnetic resonance (NMR), such as D and L conformation of linkages and sulfation and methylation levels of galactans. Some previously undocumented galactose and xylose linkages were found in both Amphiroa gracilis and Metagoniolithon stelliferum #2, and therefore NMR would be useful in more definitively placing these linkages in overall polysaccharide structures.  126  6.2.3 Connecting chemistry and material properties Enzyme digestion failed to confirm a relationship between cellulose and material strength in Calliarthron tuberculosum. This experiment could be repeated with methodological adjustments that would make genicular tissues more vulnerable to enzymatic degradation. One possibility would be to perform enzymatic treatments and subsequent tensile tests on thin sections such as those used for the polarized microscopy experiment. This method would both remove the intergenicular overhangs that typically block genicular tissue from the outside medium and encourage full penetration of the enzyme throughout the entire width of the tissue of interest. At the time of writing a suitable enzyme for degradation of agarans has not been sourced, but an enzyme that targets β-1,3- or α-1,4-galactose linkages would be an invaluable addition to future work. Similarly, extractions using a proteinase could begin to elucidate the mechanical role of non-carbohydrate cell wall components. Enzymatic treatments on lithophylloid and metagoniolithoid species may indeed be more effective, given that genicular tissues are typically more exposed.  6.2.4 Cellulose microfibril angle No major insights into genicular evolution arose from the birefringence experiments in Ch. 5, but many interesting avenues for future research did present themselves. The most surprising result from these experiments was how quickly genicular cells in Calliarthron tuberculosum begin to fray when stressed in tension, and this opens up the possibility that the dynamic shifts in stiffness that are observed in the stress-strain curve of this species could be due to cellular organization within genicular tissues rather than cellulose alignment. The high fatigue resistance of another Calliarthron species, C. cheilosporioides, has been attributed to the 127  capability of corallinoid cells to break one by one rather than propagate cracks throughout the entire geniculum (Denny et al. 2013). It is therefore entirely possible that the weak cell-to-cell connection in genicula of this group has other mechanical consequences. For example, the extended yield that is often observed in corallinoid genicula in tension could be the result of cells breaking prior to complete tissue failure; these cells would no longer be resisting force, which would cause overall resistance to decrease. Extension would continue to occur, but at a faster rate relative to the force being applied, so that the whole tissue would seem to become less stiff. Actual material stiffness could be much more consistent with increased strain, which would be seen if I was able to account for the actual surface area under stress continuously throughout tensile testing. Weak genicular cell adhesion is unique to corallinoid genicula, and genicula of both lithophylloids and metagoniolithoids are expected to behave more like solid, homogenous tissues under stress (see histology in Ch. 2 and mechanical modelling in Ch. 3). Performing similar tensile tests under polarized microscopy for lithophylloid or metagoniolithoid species would not only give insight into the cellulose network of these groups, it would help us understand the unique effect of cellular organization in corallinoids. Using birefringence to calculate cellulose microfibril angle is not uncommon, but it is indirect and requires several assumptions about crystallinity and total cellulose content. In the future, I would like to explore other more direct ways to measure cellulose angle, such as x-ray diffraction (Abraham and Elbaum 2013, Hu et al. 2017) or atomic force microscopy (Kafle et al. 2014, Zhang et al. 2015). If tensile testing under these imaging methods is not possible, they could at least be utilized in ground truthing measurements of microfibril angle calculated in this study. Birefringence, x-ray diffraction, or atomic force microscopy could also be performed on genicula from other articulated clades. 128  References Abraham, Y., & Elbaum, R. (2013). Quantification of microfibril angle in secondary cell walls at subcellular resolution by means of polarized light microscopy. New Phytol. 197: 1012-1019.  Aguirre, J., Perfectti, F., & Braga, J. C. (2010). Integrating phylogeny, molecular clocks, and the fossil record in the evolution of coralline algae (Corallinales and Sporolithales, Rhodophyta). Paleobiology 36: 529-533.  Altaner, C. M., & Jarvis, M. C. (2008). Modelling polymer interactions of the “molecular Velcro” type in wood under mechanical stress. J. Theor. Biol. 253: 434-445.  Altschul, S. F., Madden, T. L., Schaffer, A. A, Zhang, J., Zhang, Z., Miller, W., & Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25(17): 3389-3402.  Anderson, L. M., & Martone, P. T. (2014). Biomechanical consequences of epiphytism in intertidal macroalgae. J. Exp. Biol. 217: 1167-1174.  Armstrong, S. L. (1988). Mechanical properties of the tissues of the brown alga Hedophyllum sessile (C. Ag.) Setchell: variability with habitat. J. Exp. Mar. Biol. Ecol. 114: 143-151.  Bailey, J. C. (1999). Phylogenetic positions of Lithophyllum incrustans and Titanoderma pustulatum (Corallinaceae, Rhodophyta) based on 18S rRNA gene sequence analyses, with a revised classification of the Lithophylloideae. Phycologia 38: 208-216.  Bailey, J. C., & Chapman, R. L. (1998). A phylogenetic study of the Corallinales (Rhodophyta) based on nuclear small-subunit rRNA gene sequences. J. Phycol. 34: 692-705.  Baskin, T. I., Meekes, H. T. H. M., Liang, B. M., & Sharp, R. E. (1999). Regulation of Growth Anisotropy in Well-Watered and Water-Stressed Maize Roots. II. Role of Cortical Microtubules and Cellulose Microfibrils. Plant Physiol. 119(2): 681-692.  Berthelsen, A. K., & Taylor, R. B. (2014). Arthropod mesograzers reduce epiphytic overgrowth of subtidal coralline turf. Mar. Ecol. Prog. Ser. 515: 123-132.  Bilan, M. I., & Usov, A. I. (2001). Polysaccharides of Calcareous Algae and Their Effect on the Calcification Process. Russ. J. Bioorg. Chem. 27: 2-16.  Bittner, L., Payri, C. E., Maneveldt, G. W., Couloux, A., Cruaud, C., de Reviers, B., & Le  Gall, L. (2011). Evolutionary history of the Corallinales (Corallinophycidae, Rhodophyta) inferred from nuclear, plastidial and mitochondrial genomes. Mol. Phylogenet. Evol. 61: 697-713.  129  Boller, M. L., & Carrington, E. (2006). The hydrodynamic effects of shape and size change during reconfiguration of a flexible macroalga. J. Exp. Biol. 209: 1894-1903.  Borowitzka, M. A., & Larkum, A. W. D. (1987). Calcification in algae: Mechanisms and the role of metabolism. CRC Crit. Rev. Plant. Sci. 6: 1-45.  Borowitzka, M. A., & Vesk, M. (1978). Ultrastructure of the Corallinaceae. I. The Vegetative Cells of Corallina officinalis and C. cuvierii. Mar. Biol. 46: 295-304.  Borowitzka, M. A., & Vesk, M. (1979). Ultrastructure of the Corallinaceae. II. The vegetative cells of Lithothrix aspergillum Gray. J. Phycol. 15: 146-153.  Brahim, M. B., Mabrouk, L., Hamza, A., Mahfoudi, M., Bouiain, A., & Aleya, L. (2014). Bathymetric variation of epiphytic assemblages on Posidonia oceanica (L.) Delile leaves in relation to anthropogenic disturbance in the southeastern Mediterranean. Environ. Sci. Pollut. Res. 21: 13588-13601.  Brodie, J., & Zuccarello, G. C. (2007). Systematics of the Species Rich Algae: Red Algal Classification, Phylogeny and Speciation. In Hodkinson, T. R. & Parnell, J. A. N. [Eds.] Reconstructing the Tree of Life: Taxonomy and Systematics of Species Rich Taxa (Systematics Association Special Volume 72). London: CRC Press, pp. 323-336.  Carrington, E., Grace, S. P., & Chopin, T. (2001). Life history phases and the biomechanical properties of the red alga Chondrus crispus (Rhodophyta). J. Phycol. 37: 699-704.  Cases, M. R., Stortz, C. A., & Cerezo, A. S. (1992). Methylated, sulphated xylogalactans from the red seaweed Corallina officinalis. Phytochem. 31(11): 3897-3900.  Cases, M. R., Stortz, C. A., & Cerezo, A. S. (1994). Structure of the ‘corallinans’ – sulfated xylogalactans from Corallina officinalis. Int. J. Biol. Macromol. 16(2): 93-97.  Cave, I. D. (1968). The anisotropic elasticity of the plant cell wall. Wood Sci. Technol. 6: 284-292.  Cave, I. D., & Walker, J. C. F. (1994). Stiffness of wood in fast-grown plantation softwoods: the influence of microfibril angle. Forest Prod. J. 44: 43-48.  Clarkston, B. E., & Saunders, G. W. (2012). An examination of the red algal genus Pugetia (Kallymeniaceae, Gigartinales), with descriptions of Salishia firma gen. and comb. nov., Pugetia cryptica sp. nov. and Beringia wynnei sp. nov. Phycologia 51: 33-61.   Cosgrove, D. J., & Jarvis, M. C. (2012). Comparative structure and biomechanics of plant primary and secondary cell walls. Front. Plant. Sci. 3: 1-6.  130  Cronshaw, J., Myers, A., & Preston, R. D. (1958). A chemical and physical investigation of the cell walls of some marine algae. Biochim. Biophys. Acta. 27(1): 89-103.  de Carvalho, R. T., Salgado, L. T., Filho, G. M. A., & Leal., R. N. (2017). Biomineralization of calcium carbonate in the cell wall of Lithothamnion crispatum (Hapalidiales, Rhodophyta): Correlation between the organic matrix and the mineral phase. J. Phycol. 53: 642-651.    Demes, K. W., Carrington, E., Gosline, J., & Martone, P. T. (2011). Variation in the anatomical and material properties explains differences in hydrodynamic performances of foliose red macroalgae (Rhodophyta). J. Phycol. 47: 1360-1367.  Denny, M. W. (1985). Wave forces on intertidal organisms: A case study. Limnol. Oceanogr. 30: 1171-1187.  Denny, M. W. (1988). Biology and the Mechanics of the Wave-Swept Environment. Princeton, NJ: Princeton University Press, 344 pp.  Denny, M. W. (1995). Predicting Physical Disturbance: Mechanistic Approaches to the Study of Survivorship on Wave-Swept Shores. Ecol. Monogr. 65(4): 371-418.  Denny, M. W., & Gaylord, B. (2002). The mechanics of wave-swept algae. J. Exp. Biol. 205: 1355-1362.  Denny, M. W., & Hale, B. B. (2003). Cyberkelp: an intergrative approach to the modelling of flexible organisms. Philos. Trans. R. Soc. Lond. B Biol. Sci. 358: 1535-1542.  Denny, M. W., & King, F. A. (2016). The extraordinary joint material of an articulated coralline alga. II. Modeling the structural basis of its mechanical properties. J. Exp. Biol. 219: 1843-1850.  Denny, M. W., Mach, K., Tepler, S., & Martone, P. T. (2013). Indefatigable: An erect coralline alga is highly resistant to fatigue. J. Exp. Biol. 216: 3772-3780.  Denny, M. W., Miller, L. P., Stokes, M. D., Hunt, L. J. H., & Helmuth, B. S. T. (2003). Extreme water velocities: topographical amplification of wave-induced flow in the surf zone of rock shores. Limnol. Oceanogr. 48: 1-8.  Doolittle R. F. (1994). Convergent evolution: the need to be explicit. Trends Biochem. Sci. 19(1): 15-18.  Ducker, S. C. (1979). The Genus Metagoniolithon Weber-van Bosse (Corallinaceae, Rhodophyta). Aust. J. Bot. 27: 67-101.  Eaton, J. W., Bateman, D., Hauberg, S., & Wehbring, R. (2016). GNU Octave version 4.2.0 manual: a high level interactive language for numerical computations.  131  URL: http://www.gnu.org/software/octave/doc/interpreter/  Eder, M., & Lutz-Meindl, U. (2010). Non-invasive LC-PolScope imaging of biominerals and cell wall anisotropy changes. Protoplasma 246: 49-64.  Feldman, C. R., Brodie, E. D. Jr., Brodie, E. D. III, Pfrender, M. E. (2012). Constraint shapes convergence in tetrodotoxinresistant sodium channels of snakes. Proc. Natl. Acad. Sci. U.S.A. 109(12): 4556-4561.  Frei, E., & Preston, R. D. (1961). Variants in the structural polysaccharides of algal cell walls. Nature 192: 939-943.  Geitmann, A. (2010). Mechanical modeling and structural analysis of the primary plant cell wall. Curr. Opin. Plant Biol. 13: 693-699.  Genet, M., Stokes, A., Salin, F., Mickovski, S. B., Fourcaud, T., Dumail, J. -F. & van Beck, R. (2005). The influence of cellulose content on tensile strength in tree roots. Plant Soil 278: 1-9.  Girault, R., Bert, F., Rihouey, C., Jauneau, A., Morvan, C., & Jarvis, M. (1997). Galactans and cellulose in flax fibres: putative contributions to the tensile strength. Int. J. Biol. Macromolec. 21: 179-188.  Green, P. B. (1980). Organogenesis – a biophysical view. Ann. Rev. Plant Physiol. 31: 51-82.  Hale, B. (2001). Macroalgal materials: foiling fracture and fatigue from fluid forces. PhD thesis, Stanford University, Stanford, CA.  Harder, D. L., Speck, O., Hurd, C. L., & Speck, T. (2004). Reconfiguration as a Prerequisite for Survival in Highly Unstable Flow-Dominated Habitats. J. Plant Growth Regul. 23: 98-107.  Harder, D.L., Hurd, C.L., & Speck, T. (2006). Comparison of mechanical properties of four large, wave-exposed seaweeds. Am. J. Bot. 93: 1426-1432.  Harun, R., & Danquah, M. K. (2011). Enzymatic hydrolysis of microalgal biomass for bioethanol production. Chem. Eng. J. 168: 1079-1084.  Heiss, C., Klutts, J. S., Wang, Z., Doering, T. L., & Azadi, P. (2009). The structure of Cryptococcus neoformans galactomannan contains beta-D-glucuronic acid. Carbohydr. Res. 344: 915-920.  Hind, K. R., & Saunders, G. W. (2013). A molecular phylogenetic study of the tribe Corallineae (Corallinales, Rhodophyta) with an assessment of genus-level taxonomic features and descriptions of novel genera. J. Phycol. 49: 103-114.  132  Hind, K. R., Gabrielson, P. W., Jensen, C. P., and Martone, P. T. (2016). Crusticorallina gen. nov., a nongeniculate genus in the subfamily Corallinoideae (Corallinales, Rhodophyta). J. Phycol.  52: 929-941.  Hu, K., Huang, Y., Fei, B., Yao, C., & Zhao, C. (2017). Investigation of the multilayered structure and microfibril angle of different types of bamboo cell walls at the micro/nano level using a LC-PolScope imaging system. Cellulose 24:4611-4625.  Huntley, S. K., Ellis, D., Gilbert, M., Chapple, C., & Mansfield, S. D. (2003). Significant Increases in Pulping Efficiency in C4H-F5H-Transformed Poplars: Improved Chemical Savings an Reduced Environmental Toxins. J. Agric. Food Chem. 51: 6178-6183. Iyer, K. R. K., Neelakantan, P., & Radhakrishna, T. (1968). Birefringence of native cellulosic fibers. I. Untreated cottom and ramie. J. Polym. Sci. A-2 6: 1747-1758.   Johansen, H. W. (1969). Patterns of genicular development in Amphiroa (Corallinaceae). J. Phycol. 5: 118-123.  Johansen, H. W. (1981). Coralline Algae, a First Synthesis. 1st ed. Boca Raton: CRC Press, 239 pp.  Joll, L. M., & Phillips, B. F. (1984). Natural diet and growth of juvenile western rock lobsters Panulirus Cygnus George. J. Exp. Mar. Biol. Ecol. 75(2): 145-169.  Kafle, K., Xi, X., Lee, C. M., Tittmann, B. R., Cosgrove, D. J., Park, Y. B., & Kim, S. H. (2014). Cellulose microfibril orientation in onion (Allium cepa L.) epidermis studied by atomic force microscopy (AFM) and vibrational sum frequency generation (SFG) spectroscopy. Cellulose 21: 1075-1086.  Kamiyama, T., Suzuki, H., & Sugiyama, J. (2005). Studies of the structural change during deformation in Cryptomeria japonica by time-resolved synchrotron small-angle X-ray scattering. J. Struct. Biol. 151: 1-11.  Kato, A., Baba, M., & Suda, S. 2011. Revision of the Mastophoroideae (Corallinales, Rhodophyta) and polyphyly in nongeniculate species widely distributed on Pacific coral reefs. J. Phycol. 47: 662-672.  Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., & Drummond, A. (2012). Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28(12): 1647-1649.  Kitzes, J. A., & Denny, M. W. (2005). Red algae responds to waves: morphological and mechanical variation in Mastocarpus papillatus along a gradient of force. Biol. Bull. 208: 114-119.  133  Kloareg, B., & Quatrano, R. S. (1988). Structure of the cell walls of marine algae and ecophysiological functions of the matrix polysaccharides. Oceanogr. Mar. Biol. Annu. Rev. 26: 259-315.  Koehl, M. A. R., & Wainwright, S. A. (1977). Mechanical adaptations of a giant kelp. Limnol. Oceanogr. 22: 1067-1071.  Kohler, L., & Spatz, H. -C. (2002). Micromechanics of plant tissues beyond the linear-elastic range. Planta 215: 33-40.  Kraemer, G. P., & Chapman, D. J. (1991). Biomechanics and alginic acid composition during hydrodynamic adaptation by Egregia menziesii (Phaeophyta) juveniles. J. Phycol. 27: 47-53.  Kundal, P. (2011). Generic Distinguishing Characteristics and Stratigraphic Ranges of Fossil Corallines: An Update. J. Geol. Soc. India 78: 571-586.  Lahaye, M. (2001). Developments on gelling algal galactans, their structure and physico-chemistry. J. Appl. Phycol. 13: 173-184.  Le Gall, L., Payri, C. E., Bittner, L., & Saunders, G. W. (2010). Multigene phylogenetic analyses support recognition of the Sporolithales ord. nov. Mol. Phylogenetics Evol. 54: 302-305.  Le Gall, L., & Saunders, G. W. (2010). DNA barcoding is a powerful tool to uncover algal diversity: a case study of the Phyllophoraceae (Gigartinales, Rhodophyta), in the Canadian flora. J. Phycol. 46: 374-389.  Littler, M. M., & Littler, D.S. (1980). The Evolution of Thallus Form and Survival Strategies in Benthic Marine Macroalgae: Field and Laboratory Tests of a Functional Form Model. Am. Nat. 116: 25-44.  Littler, M. M., Littler, D. S., & Taylor, P. R. (1995). Selective Herbivore Increases Biomass of Its Prey: A Chiton-Coralline Reef-Building Association. Ecology 76: 1666-1681.  Littler, M. M., Taylor, P. R., & Littler, D. S. (1986). Plant defense associations in the marine environment. Coral Reefs 5: 63-71.  Losos, J. B. (2011). Convergence, adaptation, and constraint. Evolution 65(7): 1827-1840.  Lubchenco, J., & Cubit, J. (1980). Heteromorphic Life Histories of Certain Marine Algae as Adaptations to Variations in Herbivory. Ecology 61: 676-687.  Mach, K. J. (2009). Mechanical and biological consequences of repetitive loading: crack initiation and fatigue failure in the red macroalgae Mazzaella. J. Exp. Biol. 212: 961-976.  134  Mach, K. J., Tepler, S. K., Staaf, A. V., Bohnhoff, J. C., & Denny, M. W. (2011). Failure by fatigue in the field: a model of fatigue breakage for the macroalga Mazzaella, with validation. J. Exp. Biol. 214: 1571-1585.  Maneveldt, G. W., & Keats, D. W. (2008). Effects of herbivore grazing on the physiognomy of the coralline alga Spongites yendoi and on associated competitive interactions. African Journal of Marine Science 30: 581-593.  Martone, P. T. (2006). Size, strength, and allometry of joints in the articulated coralline Calliarthron. J. Exp. Biol. 209: 1678-1689.  Martone, P. T. (2007a). Biomechanics of flexible joints in the calcified seaweed Calliarthron cheilosporioides. PhD thesis, Stanford University, Stanford, CA.  Martone, P. T. (2007b). Kelp versus coralline: Cellular basis for mechanical strength in the wave-swept seaweed Calliarthron (Corallinaceae, Rhodophyta). J. Phycol. 43: 882-891.  Martone, P. T., & Denny, M. W. (2008a). To bend a coralline: Effect of joint morphology on flexibility and stress amplification in an articulated calcified seaweed. J. Exp.Biol. 211: 3421-3432.  Martone, P. T., & Denny, M. W. (2008b). To break a coralline: mechanical constraints on the size and survival of a wave-swept seaweed. J. Exp. Biol. 211: 3433-3441.  Martone, P. T., Estevez, J. M., Lu, F., Ruel, K., Denny, M. W., Somerville, C., & Ralph, J. (2009). Discovery of Lignin in Seaweed Reveals Convergent Evolution of Cell-Wall Architecture. Curr. Biol. 19: 169-175.  Martone, P. T., Kost, L., & Boller, M. (2012). Drag reduction in wave-swept macroalgae: Alternative strategies and new predictions. Am. J. Bot. 99: 1-10.  Martone, P.T., Navarro, D. A., Carlos, A. S., & Estevez, J. M. (2010). Differences in polysaccharide structure between calcified and uncalcified segments in the coralline Calliarthron cheilosporioides (Corallinales, Rhodophyta). J. Phycol. 46: 507-515.  Masarin, F., Cedeno, F. R. P., Chavez, E. G. S., de Oliveira, L. E., Gelli, V. C., & Monti, R. (2016). Chemical analysis and biorefinery of red algae Kappaphycus avarezii for efficient production of glucose from residue of carrageenan extraction processes. Biotechnol. Biofuels 9:122.  Mine, I., & Okuda, K. (2003). Extensibility of isolated cell walls in the giant tip-growing cells of the xanthophycean alga Vaucheria terrestris. Planta 217: 425-435.  Mine, I., & Okuda, K. (2007). Fine structure of cell wall surfaces in the giant-cellular xanthophycean alga Vaucheria terrestris. Planta 225: 1135-1146. 135   Natarajan, C., Hoffman, F. G., Weber, R. E., Fago, A., Witt, C. C., & Storz, J. F. (2016). Predictable convergence in hemoglobin function has unpredictable molecular underpinnings. Science 354(6310): 336-339.  Navarro, D. A., Ricci, A. M., Rodriguez, M. C., & Stortz, C. A. (2011). Xylogalactans from Lithothamnion heterocladum, a crustose member of the Corallinales (Rhodophyta). Carbohydr. Polym. 84: 944-951.  Navarro, D. A., & Stortz, C. A. (2002). Isolation of xylogalactans from the Corallinales: influence of the extraction method on yields and compositions. Carbohydr. Polym. 49: 57-62.  Navarro, D. A., & Stortz, C. A. (2008). The system of xylogalactans from the red seaweed Jania rubens (Corallinales, Rhodophyta). Carbohydr. Res. 343: 2613-2622.  Nelson, W. A., Sutherland, J. E., Farr, T. J., Hart, D. R., Neill, K. F., Jeong Kim, H., & Yoon, H. S. (2015). Multi-gene phylogenetic analyses of New Zealand coralline algae: Corallinapetra Novaezelandiae gen. et sp. nov. and recognition of the Hapalidiales ord. nov. J. Phycol. 51(3): 454-468.  Newman, R. H., & Davidson, T. C. (2004). Crystallina forms and cross-sectional dimensions of cellulose microfibrils in the Florideophyceae (Rhodophyta). Bot. Mar. 47: 490-495.  Niklas, K. J. (2004). The Cell Walls that Bind the Tree of Life. BioScience 54(9): 831-841.  Nishiyama, Y. Structure and properties of the cellulose microfibril. J. Wood. Sci. 55: 241-249.  Okazaki, M., Furuya, K., Tsukayama, K., & Nisizawa, K. (1982). Isolation and Identification of Alginic Acid from a Calcareous Red Alga Serraticardia maxima. Bot. Mar. 25: 123-131.  Okazaki, M., & Tazawa, K. (1989). Alginic Acid in Corallinaceae (Cryptonemiales, Rhodophyta). Korean J. Phycol. 4(2): 213-219.  Oldenbourg, R. (2007). Analysis of microtubule dynamics by polarized light. Methods Mol. Med. 137: 111-123.  Padilla, D. K. (1984). The importance of form: Differences in competitive ability, resistance to consumers and environmental stress in an assemblage of coralline algae. J. Exp. Mar. Biol. Ecol. 79: 105-127.  Padilla, D. K. (1989). Algal structural defenses: form and calcification in resistance to tropical limpets. Ecology 70: 835-842.  Patterson, M. R., Harwell, M. C., Orth, L. M., & Orth, R. J. (2001). Biomechanical properties of the reproductive shoots of eelgrass. Aquat. Bot. 69: 27-40. 136   Pennings, S. C., & Paul, V. J. (1992). Effect of Plant Toughness, Calcification, and Chemistry on Herbivory by Dolabella auricularia. Ecology 73: 1606-1619.  Popper, Z. A. (2008). Evolution and diversity of green plant cell walls. Curr. Opin. Plant Biol. 11: 286-292.   Preston, J. M. (1933). Relations between the refractive indices and the behavior of cellulose fibres. J. Chem. Soc. Faraday Trans. 29: 65-71.  Ratnasingham, S., & Herbert, P. D. (2007). BOLD: The Barcode of Life Data System. Mol. Ecol. Notes 7(3): 355-364.  Rees, D. A., & Conway, E. (1962). The Structure and Biosynthesis of Porphyran: a Comparison of some Samples. Biochem. J. 84: 411-416.  Rosler, A., Perfectii, F., Pena, V., & Braga, J. C. (2016). Phylogenetic relationships of Corallinaceae (Corallinales, Rhodophyta): taxonomic implications for reef-building corallines. J. Phycol. 52(3): 412-431.  Ryden, P, Sugimoto-Shirasu, K., Smith, A. C., Findlay, K, Reiter, W. -D., & McCann, M. C. (2003). Tensile properties of Arabidopsis Cell Walls Depend on Both a Xyloglucan Cross-Linked Microfibrillar Network and Rhamnogalacturonan II-Borate Complexes. Plant Physiol. 132(2): 1033-1040.  Saunders, G. W. (2008). A DNA barcode examination of the red algal family Dumontiaceae in Canadian waters reveals substantial cryptic species diversity. I. The foliose Dilsea-Neodilsea complex and Weeksia. Botany 86: 773-789.  Speed, M. P., & Arbuckle, K. (2017). Quantification provides a conceptual basis for convergent evolution. Biol. Rev. 92: 815-829.  Starko, S., Claman, B., & Martone, P. T. (2015). Biomechanical costs of branching in flexible wave-swept macroalgae. New Phytol. 206: 133-140.   Starko, S., Mansfield, S., & Martone, P. T. (2018). Cell wall chemistry and tissue structure drive shifts in material properties of a perennial kelp. Euro. J. Phycol. In press.  Starko, S., & Martone, P. T. (2016). Evidence of an evolutionary-developmental trade-off between drag avoidance and tolerance strategies in wave-swept intertidal kelps (Laminariales, Phaeophyceae). J. Phycol. 52: 54-63.  Stayton, C. T. (2015). The definition, recognition, and interpretation of convergent evolution, and two new measures for quantifying and assessing the significance of convergence. Evolution 69(8): 2140-2153. 137   Steneck, R. S. (1982). A Limpet-Coraline Alga Association: Adaptations and Defenses Between a Selective Herbivore and its Prey. Ecology 63: 507-522.  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., Hacker, S. D., & Dethier, M. D. (1991). Mechanisms of competitive dominance between crustose coralline algae: an herbivore-mediated competitive reversal. Ecology 72: 938-950.  Steneck, R. S., & Watling, L. (1982). Feeding capabilities and limitation of herbivorous molluscs: A functional group approach. Mar. Biol. 68: 299-319.  Stern, D. L. (2013). The genetic causes of convergent evolution. Nature Rev. Genet. 14: 751-764.  Stockton, B., Evans, L. V., Morris, E. R., & Rees, D. A. (1980). Circular dichroism analysis of the block structure of alginates from Alaria esculenta. Int. J. Biol. Macromol. 2: 176-178.  Stortz, C. A., Cases, M. R., & Cerezo, A. S. (1997). Red seaweed galactans: methodology for the structural determination of corallinan, a different agaroid. In R. R. Townsend & A. T. Hotchkiss Jr. [Eds.] Techniques in glycobiology. New York: Marcel Dekker, pp. 567-593.  Sugimoto, K., Williamson, R. E., & Wasteneys, G. O. (2000). New Techniques Enable Comparative Analysis of Microtubule Orientation, Wall Texture, and Growth Rate in Intact Roots of Arabidopsis. Plant Physiol. 124(4): 1493-1506.  Sugiyama, J., Vuong, R., & Chanzy, H. (1991). Electron Diffraction Study on the Two Crystalline Phases Occurring in Native Cellulose from an Algal Cell Wall. Macromolecules 24: 4168-4175.  Taiz, L. (1984). Plant Cell Expansion: Regulation of Cell Wall Mechanical Properties. Ann. Rev. Plant Physiol. 35: 585-657.  Takano, R., Hayashi, J., Hayashi, K., Hara, S., & Hirase, S. (1996). Structure of a Water-soluble Polysaccharide Sulfate from the Red Seaweed Joculator maximus Manza. Bot. Mar. 39: 95-102.  Toole, G. A., Smith, A. C., & Waldron, K. W. (2002). The effect of physical and chemical treatment on the mechanical properties of the cell wall of the alga Chara corallina. Planta 214: 468-475.  Toole, G. A., Kacurakova, M., Smith, A. C., Waldron, K. W., & Wilson, R. (2004). FT-IR study of the Chara corallina cell wall under deformation. Carbohydr. Res. 339: 629-635. 138   Tsekos, I. (1999). The sites of cellulose synthesis in algae: Diversity and evolution of cellulose-synthesizing enzyme complexes. J. Phycol. 35: 635-655.  Tsekos, I., Okuda, K., & Brown, R. M. Jr. (1996). The formation and development of cellulose-synthesizing linear terminal complexes (TCs) in the plasma membrane of the marine red alga Erythrocladia subintegra Rosenv. Protoplasma 193(1-4): 33-45.  Turner, S. R., and Somerville, C. R. (1997). Collapsed Xylem Phenotype of Arabidopsis Identifies Mutants Deficient in Cellulose Deposition in the Secondary Cell Wall. Plant Cell 9: 689-701.  Turvey, J. R., & Simpson, P. R. (1966). Polysaccharides from Corallina officinalis. Proc. Int. Seaweed Symp. 5: 323-328.  Turvey, J. R., & Williams, E. L. (1970). The structure of some xylans from red algae. Phytochem. 9(11): 2383-2388.  Usov, A. I. (1992). Sulfated polysaccharides of the red seaweeds. Food Hydrocolloid. 6: 9-23.  Usov, A. I., & Bilan, M. I. (1998). Polysaccharides of algae. 52. The structure of sulfated xylogalactan from the calcareous red alga Bossiella cretacea (P. et R.) Johansen (Rhodophyta, Corallinales). Russ. J. Bioorg. Chem. 24: 123-129.  Usov, A. I., Bilan, M. I., & Klochkova, N. G. (1995). Polysaccharides of algae. 48. Polysaccharide composition of several calcareous red algae: Isolation of alginate from Corallina pilulifera P. et. R. (Rhodophyta, Corallinaceae), Bot. Mar. 38: 43-51.  Vincent, J. (1990). Structural Biomaterials, revised edn. Princeton, NJ: Princeton University Press. 244 pp.  Vogel S. (1984). Drag and Flexibility in Sessile Organisms. Am. Zool. 24(1): 37-44.  Vreeland, V., & Kloareg, B. (2000). Cell wall biology in red algae: divide and conquer. J. Phycol. 36: 793-797.  Woelkerling, W. J. (1988). The Coralline Red Algae: An Analysis of the Genera and Subfamilies of Nongeniculate Corallinaceae. New York: Oxford University Press, 268 pp.  Yendo, K. (1904). A study of the genicula of Corallinae. J. Coll. Sci. Imp. Univ. Tokyo 19: 1-45.  Yoon, H. S., Hackett, J. D., & Bhattacharya, D. (2002). A single origin of the peridinin and fucoxanthin containing plastids in dinoflagellates through tertiary endosymbiosis. Proc. Natl. Acad. Sci. U.S.A. 99: 11724-11729. 139  Appendices  Appendix A: Genicular dimensions and breaking force    Species Area (mm2) Length (mm) Force (N) Cheilosporum sagittatum 0.04 ± 0.01 0.21 ± 0.01 2.2 ± 0.1 Calliarthron tuberculosum 0.39 ± 0.05 0.49 ± 0.01 8.3 ± 0.9 Corallina officinalis var. officinalis 0.44 ± 0.05 0.43 ± 0.02 12.82 ± 0.87 Johansenia macmillanii 0.50 ± 0.04 0.45 ± 0.01 9.76 ± 0.73 Lithothrix aspergillum 0.13 ± 0.01 0.30 ± 0.01 1.73 ± 0.08 Amphiroa anceps 0.64 ± 0.06 0.92 ± 0.08 5.71 ± 0.50 Amphiroa gracilis 1.10 ± 0.09 5.51 ± 0.08 1.10 ± 0.49 Metagoniolithon stelliferum #1 0.13 ± 0.01 6.53 ± 0.26 0.83 ± 0.05 Metagoniolithon stelliferum #2 0.41 ± 0.03 3.40 ± 0.26 1.08 ± 0.07 Metagoniolithon chara 0.19 ± 0.02 1.31 ± 0.16 1.06 ± 0.07  Table A.1. Area, length, and breaking force of genicula in species tested in Ch. 2 (mean ± s.e.m). All measurements apply specifically to joints found within the most basal 2 cm of each plant - dimensions may vary for more distal joints.  140  Appendix B: Mathematical details of MatLab bending model  The model calculates external bending moments (Mexternal) of each geniculum as: 𝑀𝑀external = Fdragδ Eq. B. 1 where, Fdrag is the drag force applied in a perpendicular direction, and δ is the lever arm as measured by the distance between force application and the centre of the geniculum. An erect frond will bend until external moments of each geniculum equal internal moments (Minternal), measured as the sum of all elemental moments and calculated as: 𝑀𝑀internal =  � zEƐ2r1r2 cos2 θ𝑑𝑑θπ2−π2Eq. B. 2 where, z is distance of any elemental area of the geniculum to the neutral axis, E is tissue stiffness, Ɛ is tissue strain resulting from the applied force, r1 and r2 are radii of the geniculum, and θ is the angle of an elemental area relative to the geniculum centre (for details, see Martone and Denny 2008a). The neutral axis is the area within a tissue in bending that does not experience stress, with tissue on one side of the axis being under tension and tissue on the other side being under compression. The exact location of the neutral axis depends on a tissue’s tensile and compressive stiffness. The model used species averages of tensile stiffness from Ch. 2, and estimated compressive stiffness as being 4 times lower (see Martone and Denny 2008a). The model output a total frond deflection based on the sum of bending of all genicula within an average virtual frond. Due to differences in tissue construction between joint types, two different model variations were used. Corallinoid genicula were modelled as cables made of independently moving cells, based on data suggesting that genicular cells in this group have minimal lateral 141  connection to one another (Martone and Denny 2008a, Denny and King 2016). Lithophylloid and metagoniolithoid genicular tissue appears much more cohesive (Ch. 2), so these species were modelled as solids. While solid genicula are expected to curve in bending, the individual cells of corallinoid genicula are expected to follow the shortest straight line possible between their calcified anchor points (Fig. B.1A). The key mathematical distinction involves how strain is calculated in each model, and relevant equations are outlined below.   Fig. B.1. Genicula bending in long section before intergenicular contact (A, B), at the moment of contact (C, D), and after contact (E, F). Both cable (A, C, E) and solid (B, D, F) models are shown. Light grey represents uncalcified genicular tissue, dark grey represents the ends of adjacent calcified intergenicula. See Martone and Denny 2008a for additional derivations and details. 142   .1 Cable model The model calculates strain for every elemental area within the geniculum as bending occurs (for details, see Martone and Denny 2008b), and does so differently depending on whether it occurs before or after intergenicula make contact (Fig. B.1). In general, strain (Ɛ) is calculated as: Ɛ =  𝐿𝐿𝐿𝐿o− 1 Eq. B. 3 where, Lo is the pre-bent length of the geniculum and L is the post-bent length. Before intergenicular make contact, L in the cable model is calculated as: 𝐿𝐿cable,pre−contact = 𝐿𝐿o + 2m Eq. B. 4 where, m is the additional stretched length on one end of the geniculum. This is calculated as a straight line using the triangle defined in Fig. B.1A: m = z sin �ф2� Eq. B. 5 where, ф is the bending angle of the whole geniculum and z is the distance between any elemental area of the geniculum and the neutral axis. Prior to intergenicular contact, z is calculated as: zpre−contact = r2 sinθ − η Eq. B. 6 where, r2 is the short radius of the geniculum (i.e. the one over which it is assumed bending will most easily occur), θ is the angle of the elemental area relative to the geniculum centre, and η is perpendicular distance between the neutral axis and the genicular midline. For derivation of this equation, see Appendix A of Martone and Denny 2008a. Combining equations B.4, B.5, and B.6 and inputting the result into Eq. B.3 gives a calculation for pre-contact strain in the cable model: 143  Ɛcable,pre−contact =  2(r2 sin θ − η) sin �ф2�𝐿𝐿o Eq. B. 7 To determine the point at which the model switches from pre-contact to post-contact strain, contact angle must be calculated. When contact occurs, the neutral axis shifts all the way to the contact point and causes the entirety of the geniculum to be in tension. Contact angle is calculated is in the cable model with the following: βcable =  arcsin 𝐿𝐿𝑜𝑜 − 2x2(η + y) Eq. B. 8 where, y is the radius of the intergeniculum and βcable is half the contact angle (ф=2β, Fig. B.1C). Once contact has occurred, we use the triangle defined in Fig. B.1E to calculate k, half the new genicular length between intergenicula: k = z ∗ sin �ф2� Eq. B. 9 where, z is now measured as the distance to the intergenicular contact point due to the shift in the neutral axis. It is now calculated as: zpost−contact = y + r2sinθ Eq. B. 10 For derivation of this equation, see the appendix of Martone and Denny 2008a. Lcable is then calculated as the following: 𝐿𝐿cable,post−contact = 2(y + r2sinθ) sin �ф2� + 2x Eq. B. 11 where, x is half the length of the genicular tissue hidden by intergenicular lips. Inputting Eq. B.11 into Eq. B.3 gives a calculation for post-contact strain: Ɛcable,post−contact = 2(y + r2sinθ) sin �ф2� + 2x𝐿𝐿o − 1 Eq. B. 12 144  To account for pre-contact strain, we add together equations B.7 and B.11, applying the pre-contact strain equation to all angles up until contact (β) and applying the post-contact equation to angles beyond β: Ɛcable,total = 2(r2sinθ − η)sinβ + 2(y + r2sinθ) sin �ф2 − β� + 2x𝐿𝐿o − 1 Eq. B. 13   .2 Solid model For lithophylloid and metagoniolithoid genicula, new length is calculated as an arc: 𝐿𝐿solid = (R + z)ф + 2x= (𝑅𝑅 + r2 sinθ − η)ф + 2x Eq. B. 14 where, R is the radius of curvature to and x is the length of the intergenicular lips (Fig. B.1B). R is calculated from the neutral axis: 𝑅𝑅 = 𝐿𝐿o − 2xфEq. B. 15 Combining equations B.14 and B.15 and inputting them into Eq. B.3 gives us an equation for pre-contact strain: Ɛsolid,pre−contact =  ф(r2 sin θ − η)𝐿𝐿o Eq. B. 16 In the solid model, β is calculated by treating the neutral axis as an arc (Fig. B.1D): βsolid = 𝐿𝐿o − 2x2(η + y) Eq. B. 17 After the contact angle has been reached and the neutral axis is shifted to the contact point, R disappears, and z is calculated using Eq. B.10. Strain is then calculated using the following: Ɛsolid,post−contact = (y + r2sinθ)ф + 2x𝐿𝐿o − 1 Eq. B. 18 145   To account for pre-contact strain, we add together equations B.16 and B.18, applying the pre-contact strain equation to all angles up until contact (β) and applying the post-contact equation to angles beyond β: Ɛsolid,total = β(r2sinθ − η) + (ф − β)(y + r2sinθ) + 2x𝐿𝐿o − 1 Eq. B. 19 The differences in both strain and contact angle calculations between the cable and solid models ultimately result in genicula being modeled as stiffer in the solid model compared to the cable model (see Martone and Denny 2008b).    .3 Model accuracy Model accuracy was tested by comparing real bending data with modeled data of the same fronds. The solid bending model performed well at all forces tested (0.05N, 0.20N, 0.98N). Average real percent bent and average modelled percent bent differed by less than 15% for all lithophylloid and metagoniolithoid species at 0.05N, with the model typically over-estimating flexibility. Accuracy improved at higher forces as values converged on 100% bent. The cable bending model sometimes underestimated bending of corallinoid species by 20-40% at 0.05N, however, the solid model predicted even less bending, so the cable model was maintained. Discrepancies are at least partially due to using average values of material stiffness for each species when modelling individual fronds.    146   Appendix C: Estimation of cellulose thickness in genicular tissue  Cellulose mass is calculated as: Mcell. =  Pcell.Mgen. Eq. C. 1 where, Mcell. is mass of cellulose, Pcell. is proportion of cellulose, and Mgen. is mass of the genicular tissue. Mgen. is calculated as: Mgen. = Vgen.Dgen. Eq. C. 2 where, Vgen. is volume of genicular tissue and Dgen. is density of genicular tissue. By combining Eq. C.1 and Eq. C.2, we get: Mcell. = Pcell.Vgen.Dgen. Eq. C. 3 Volume of cellulose is calculated as: Vcell. = Mcell.Dcell. Eq. C. 4 where, Dcell. is density of cellulose. By inserting Eq. C.3 into Eq. C.4 we get: Vcell. = Pcell.Vgen.Dgen.Dcell. Eq. C. 5 If we assume that the volume ratio of cellulose to genicular tissue is the same as the thickness ratio of cellulose to genicular tissue, i.e.: Vcell.Vgen. = tT Eq. C. 6 where, t is “effective thickness” or thickness of the cellulose within the tissue, and T is the total thickness of the tissue, we can rearrange this equation and insert it into Eq. C.5 to get: 147  t = T ∗ Pcell.Dgen.Dcell. Eq. C. 7 It is here that several assumptions had to be made. The first assumption was that all the glucose within the genicular tissue is from cellulose, giving us an estimate of 6.23% by dry weight (see Ch. 3). This was converted to an estimate of % wet weight, using estimates of whole frond weight loss from total desiccation measured by Patrick Martone (unpublished data) – as most of the frond is calcified intergenicular tissue, this estimate is likely an underestimate of the % weight loss that occurs in genicula. The second assumption made was that the cellulose present is crystalline and therefore birefringent. While crystalline cellulose has been documented in corallines (Newman and Davidson 2004), the proportion of amorphous cellulose is not known, so this estimate may be high. The value of Dcell. used for calculations in this study was 1.599 g cm-3, the value for crystalline cellulose type Iβ (Sugiyama et al. 1991), which is the dominant crystalline type found in Florideophycean red algae (Newman and Davidson 2004).  The value of Dgen. was calculated using density measurements of whole fronds and intergenicular tissue, and measurements of the proportion of genicular tissue by volume in whole fronds. Density of intergenicula and whole fronds was calculated by Patrick Martone (unpublished data), by weighing tissues and estimating their volume by submersing them in water. Based on this work, density of whole fronds (Dfrond) is 1.405 g cm-3, while density of intergenicular tissue alone (Dint.) is 1.287 g cm-3. The proportion of genicular tissue in whole fronds was calculated using morphometric data collected in Ch. 3; volume of each geniculum and intergeniculum within 2 cm long in length were calculated as: V = πr2L Eq. C. 8 148  where, V is the volume of either the geniculum or intergeniculum, r is the radius, and L is the length. Volumes of all genicula and intergenicula were added up to get a total frond volume, then genicular volume was divided by total volume to get the proportion of genicular tissue. Based on measurements taken on 15 segments, the average proportion of genicular tissue (Pgen.) is 0.07. Density of genicular tissue can then be calculated as: Dgen. = Dfrond − Dint.Pgen. Eq. C. 9 which gives us a value of 1.79 g cm-3. While Pcell., Dcell., and Dgen. are all constants in this system for the calculation of t, T (the thickness of the genicular tissue) changes with strain. As length increases, both width and thickness decrease. What’s more, they do not decrease in a uniform manner across the length of the geniculum. Width and thickness are unchanged at the ends of the geniculum where cells are anchored into the calcified intergenicular tissue, and the geniculum “waists” inwards at an increasing rate until the midpoint between the two ends, where tissue is the thinnest. In order to appropriately calculate thickness of the geniculum at any given point, we have to consider this behaviour. We did this by assuming that the ratio of width at the edges of genicula in tension (w) to the width at the middle (w’) was equal to the ratio of thickness at the edges (T) to thickness at the middle (T’, Fig. C.1).    149   Fig. C.1. Schematic of the top-down and side views of a longitudinal genicular section in tension. The width and thickness at the end of the geniculum (w, T) remains constant with increased strain, as the ends of the genicular cells are anchored in calcified tissue. The width and thickness at the centre of the geniculum (w’, T’) decreases with increasing strain. In this study, we assumed that T’/T was equal to w’/w, and used this equality to estimate T’.   Width measurements were taken in ImageJ (NIH Image, http://rsb.info.nih.gov/ij) using the same images for which retardance values were measured. Edge thickness was assumed to be 0.3 mm at all strains, as this was the starting thickness of the longitudinal sections.  T’ for the middle of genicula at each strain was calculated as: T′ = 0.3 ∗ �w′w � Eq. C. 10 Using T=0.3 mm for edge measurements and calculating values of T’ for middle measurements of genicula at each strain, I was able to use Eq. C.7 to estimate the thickness of cellulose in order to estimate microfibril angle.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0370962/manifest

Comment

Related Items