ZIERITZ, A. 2010. Variability, function and phylogenetic significance of unionoid shell characters....

VARIABILITY, FUNCTION AND

PHYLOGENETIC SIGNIFICANCE OF

UNIONOID SHELL CHARACTERS

by

ALEXANDRA ZIERITZ

ST CATHARINE’S COLLEGE

SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSPHY

UNIVERSITY OF CAMBRIDGE, JUNE 2010

SUPERVISOR:

DR. DAVID CHRISTOPHER ALDRIDGE

iii

PREFACE

This dissertation is submitted for the degree of Doctor of Philosophy. It is the result of my

own work and includes nothing which is the outcome of work done in collaboration except

as stated in the following:

Dr. Aldridge (PhD supervisor) provided general scientific guidance and advice, and helpful

comments on previous drafts of all chapters.

Dr. Hoffman and Prof. Amos supervised AFLP analysis. They also provided helpful advice and

comments regarding statistical analysis of AFLP data, and helpful comments on a previous draft of

Chapter 3.

Dr. Bogan provided some helpful comments and discussion on fossil unionoids, and helpful

comments on a previous draft of Chapter 5.

Dr. Harper and Prof. Checa provided invaluable expertise and discussion on bivalve

microsculptures. Dr. Harper furthermore provided supervision during scanning electron

microscopy and helpful comments on a previous draft of Chapter 6.

No part of this dissertation has been or is being concurrently submitted for a degree,

diploma or other qualification at any other university.

Permission is granted to consult or copy the information contained herein for the purpose

of private study, but not for publication.

This thesis does not exceed the limit prescribed by the Degree Committee of Biology as

stated in the Memorandum to Graduate Students.

Alexandra Zieritz

CONTENTS

Title page iPreface iiiContents vAcknowledgements ixSummary xiGlossary xiii Chapter 1 - Introduction 1

1.1 A brief introduction to the uniqueness of unionoids 41.1.1 Unionoid phylogeny, diversity and distribution 41.1.2 Importance and conservation status 71.1.3 Life cycle and morphological variation 7

1.2 Importance of understanding morphological patterns in unionoid shells

8

1.2.1 Systematics, phylogeny and evolution 91.2.2 Ecology, conservation and environmental reconstruction 101.2.3 Reconstruction of population parameters 11

1.3 Aims and questions addressed in this thesis 111.3.1 “Gross shell morphology”: size, shape, inflation and

thickness 12

1.3.2 Umbonal sculpture 121.3.3 Periostracal microprojections 13

1.4 Style of thesis 13 Chapter 2 - Identification of ecophenotypic trends within three European

freshwater mussel species (Bivalvia: Unionoida) using traditional and modern morphometric techniques

15

2.1 Abstract 172.2 Introduction 172.3 Materials and methods 19

2.3.1 Sampling of individuals 192.3.2 Sampling of habitat parameters 212.3.3 Growth measurements, and age and sex determination of

individuals 21

2.3.4 Morphological analysis 212.3.4.1 Fourier shape analysis 212.3.4.2 Analysis of traditional shell and anatomical

measurements 22

Variability, function and phylogenetic significance of unionoid shell characters A. Zieritz

vi

2.3.5 Statistical analysis 232.4 Results 24

2.4.1 Habitat parameters 242.4.2 Species composition 242.4.3 Growth 252.4.4 Morphological analysis 26

2.4.4.1 Morphological trends in U. pictorum of five paired sites and influence of habitat and non-habitat factors

26

2.4.4.2 Morphological trends in all three unionoid species at Abingdon

28

2.5 Discussion 302.5.1 Patterns across all species 302.5.2 Patterns within single species 322.5.3 Utility of the patterns observed 33

Chapter 3 - Phenotypic plasticity and genetic isolation-by-distance in the

freshwater mussel Unio pictorum (Mollusca: Unionoida) 35


3.3.1 Sampling 393.3.2 Morphological analysis 403.3.3 Genetic analysis 413.3.4 Quantification of the genotyping error rate 433.3.5 Genetic data analysis 43

3.4 Results 443.4.1 Morphological analysis 443.4.2 Population structure 46

3.5 Discussion 493.5.1 Genetic population structure 503.5.2 Phenotypic plasticity of shell form 51

Chapter 4 - Sexual, habitat-constrained and parasite-induced dimorphism in

the shell of a freshwater mussel (Anodonta anatina, Unionidae) 55


4.3.1 Sampling, sex determination and trematode infection 604.3.2 Size and growth rates 614.3.3 Shell shape, thickness and density 61

Glossary

vii

4.4 Results 634.4.1 Comparison of the five study populations 634.4.2 Sex ratios 644.4.3 Size and growth rates 644.4.4 Shell shape, thickness and density 654.4.5 Trematode parasitism 70

4.5 Discussion 714.5.1 Size and hermaphroditism 714.5.2 Sexual dimorphism in sagittal shell shape 724.5.3 Sexual dimorphism in relative shell width 724.5.4 Sexual dimorphism in shell thickness and density 734.5.5 Trematode parasite-induced dimorphism 744.5.6 Habitat-constrained dimorphism 744.5.7 Application of the patterns observed 75

Chapter 5 - Variability and a new model for character evolution of umbonal sculptures in the Unionoida

77


5.3.1 Classification of beak sculpture types 825.3.2 Development of model of character evolution, and

identification of homologies and homoplasies 83

5.4 Results 845.4.1 Morphological types of unionoid beak sculpture 845.4.2 Intermediate forms and implications for character evolution

of unionoid beak sculpture 88

5.5 Discussion 895.5.1 (In)validity of previous models of character evolution 895.5.2 Plesiomorphic character state and the fossil record 90

5.5.2.1 Smooth vs. sculptured plesiomorphic character state 905.5.2.2 Most likely plesiomorphic beak sculpture type 91

5.5.3 A new model of beak sculpture character evolution in the Unionoida

91

5.5.3.1 Implications for unionoid phylogeny and evolution 945.5.3.2 Convergences and implications for probable functional

morphologies 94

Chapter 6 - Variability, function and phylogenetic significance of periostracal

microprojections in palaeoheterodont bivalves 97

6.1 Abstract 996.2 Introduction 99


viii

6.3 Materials and methods 1046.4 Results 105

6.4.1 Occurrence and morphology of periostracal microprojections in Palaeoheterodonta

105

6.4.2 Variation in spike morphology, abundance and distribution across the shell

108

6.4.3 Mineralisation status of periostracal microprojections 1106.4.4 Trigonioid ‘bosses’ 1126.4.5 Influence of periostracal thickness on presence/absence of

microprojections 112

6.5 Discussion 1136.5.1 Occurrence of periostracal microprojections across the

unionoid phylogeny 113

6.5.2 Morphological variation and possible functional morphologies of structures

115

6.5.3 Comparison to spikes of other bivalve groups and implications for bivalve phylogeny

118

Chapter 7 - Conclusions 119

7.1 What can we learn from a unionoid shell? 1227.1.1 Systematics, phylogeny and evolution 124

7.1.1.1 “Gross shell morphology”: size and form 1247.1.1.2 Umbonal sculpture 1257.1.1.2 Periostracal microprojections 126

7.1.2 Ecology, conservation and environmental reconstruction 1267.1.2.1 “Gross shell morphology”: size, shape, inflation,

thickness, density and adductor scar sizes 127

7.1.2.2 Umbonal sculpture 1287.1.2.3 Periostracal microprojections 129

7.1.3 Determination of sex and trematode loads 1297.1.3.1 “Gross shell morphology”: size, shape, inflation and

thickness 129

7.2 Future directions 1307.2.1 Testing for consistency of the patterns observed 1317.2.2 Open questions 1317.2.3 Application of methods to non-unionoid taxa 132

Appendix 133

Table A.1 Umbonal sculpture types of unionoid taxa examined from shell material, photographs and/or drawings

135

Bibliography 149

ix

ACKNOWLEDGEMENTS

First and foremost I thank my supervisor David Aldridge for his guidance and

support over the past three and a half years. I am particularly grateful to him for giving me

the freedom to pursue my own scientific interests and for encouraging me throughout. My

secondary advisors Elizabeth (“Liz”) Harper and Richard Preece provided invaluable

additional malacological expertise and were always there when I needed them. Special

thanks go to Richard for showing me around the museum collection and to Liz for

spending hours with me on the SEM, teaching me about bivalve microstructure and life.

My subsistence costs during this PhD were predominantly covered by studentships

of the Austrian Federal Ministry of Science and Research and the Cambridge European

Trust. Additional financial support was kindly provided by the Ministry of Lower Austria,

the Siegfried-Ludwig Fund, the Balfour Fund and the St Catharine’s College. Genetic

analyses of mussel populations (Chapter 3) were funded by a research grant of the

Conchological Society of Great Britain & Ireland.

I could not have asked for a more pleasant and stimulating working environment

than the Aquatic Ecology Group. Thank you so much for all your help and support, both in

the field and in the office – Anna McIvor, Beccy Mant, Du Lina, Gawsia Chowdhury,

Holly Barclay, Line zu Ermgassen, Matthew Oreska and Nicole Spann. Numerous

additional people from the Department of Zoology and other Cambridge University

departments have dedicated some of their valuable time to teach me. Bill Amos, Joe

Hoffman and the whole Molecular Ecology Group gave me a “molecular” home for

several months and introduced me into AFLP analysis. John Parker offered much

appreciated advice on how to develop a model of character evolution. I also wish to thank

Andrea Manica for his statistical advice, Keith Gray for teaching me how to prepare shells

for SEM, André Sartori for his help in the field and inspiring bivalve discussions, our

technical assistants Ian Goldstone and Ben Taylor, our graduate secretary Linda Wheatley,

and Clair Castle and Jane Acred for being the most helpful librarians I have ever met.

Generous travel grants by the Malacological Society of London, the Freshwater

Mollusk Conservation Society and the St Catharine’s College gave me the opportunity to

present and discuss my work with scientists from other parts of the world. Some of these,

and in particular, Antonio Checa, Art Bogan and James Crampton have offered some


x

greatly appreciated expert’s advice and insight. Daniel Graf, Kevin Cummings and Keith

Walker aided in identification of tricky specimens.

Investigation of interspecific variation in shell morphology would not have been

possible without the generous donations of shell material by following people and

institutions: The Bivalve Tree-of-Life Project, Dan Hua (Freshwater Mollusk Conservation

Center, VA, USA), Nathan Eckert (Virginia Department of Game and Inland Fisheries,

VA, USA), Manuel Lopez-Lima (University of Porto), Dr. Ellinor Michel and Dr.

Jonathan Todd (both Natural History Museum, London), Keith Scriven (Maerdy and

Mawddach Hatchery) and Conor Wilson (Queen’s University Belfast).

I also wish to thank Abingdon Boat Marina, Thames & Kennet Marina, Harleyford

Estate & Marina, Racecourse Marina Windsor and Saxon Moorings, and all the friendly

people there for allowing access to the marina study sites.

Finally, I am extremely grateful to my parents Michael and Renate, and my whole

family for their love, support, letters, emails and visits. Many thanks also go to my friends

both at home and in Cambridge, who always knew how to put a smile on my face and

helped me through this whole process. Last but definitely not least, I want to thank my

partner Roman, who has been putting up with me for over a decade now, developed a

computer program called “schlexihexi” that saved me days of copy-paste, and knows

exactly when it is time to treat me with some chocolate and an episode of ‘Columbo’.

xi

SUMMARY

Freshwater mussels of the order Unionoida show a wide variability in shell features

but an understanding of which factors determine which trends in shell morphology is poor.

This thesis investigates inter- and intraspecific patterns in unionoid shell characters and

their potential use for reconstructing (1) environments, (2) characteristics of populations,

and (3) evolutionary trends and phylogenies.

Investigation of morphological patterns within three unionoid species from two

habitat types (marinas and river) of the River Thames, UK, elucidated consistent

ecophenotypic trends in maximum shell size, relative adductor size and shape of the dorso-

posterior shell margin. These shell characters may thus have broad ecological significance

and could have considerable utility to palaeontologists, taxonomists and conservation

biologists.

Molecular analyses using Amplified Fragment Length Polymorphisms suggested

that pronounced differences in shell morphology between populations of the same species

were caused by phenotypic plasticity. Observed genetic differences along the River

Thames, on the other hand, were consistent with a pattern of isolation by distance and

probably reflect limited dispersal via host fish species upon which unionoid larvae are

obligate parasites.

While relative shell width was a poor indicator of environment, this character was

significantly influenced by allometric growth, sexual dimorphism and trematode

parasitism. Detailed investigations on Anodonta anatina revealed that differences in the

degree of sexual dimorphism between populations may reflect the overarching effect of

habitat on morphology. In addition, other non-habitat related dimorphic patterns in sagittal

shape and thickness of shells were observed.

Interspecific morphological trends and their potential use for reconstructing

phylogenies were investigated with regard to two types of shell sculptures. First, a new

model of character evolution of umbonal sculptures in the Unionoida was developed by

examination of over 150 extinct and modern species. Second, investigation of shell

surfaces from specimens of all extant palaeoheterodont (unionoid + trigonioid) families

using scanning electron microscopy revealed the presence of three types of periostracal

microprojections. These possibly aid in the stabilisation and/or orientation of the mussel

within the sediment. Observations on both umbonal and periostracal sculptures indicated

considerable phylogenetic value of these two shell features.

xiii

GLOSSARY

Adductor muscle: One of usually two large muscles (one anterior, one posterior) that contract to close the shell and maintain it in that condition.

Alae (adj. alate): Anterior and/or posterior winglike projections of the valves that extend dorsally above the hinge line; also called wings; bivalves exhibiting alae are also called symphynote.

Allometric growth: The variation in the relative rates of growth of various parts of the body.

Annual ring: Compact line of temporarily arrested growth or rest period appearing on the shell surface as a raised or darker comarginal line; also called annulus.

Annulus (pl. annuli): See annual ring.

Anterior: Front or forward; head end.

Apomorphy (adj. apomorphic): A derived characteristic of a clade; any feature novel to a species and its descendants.

Beak: See umbo.

Beak sculptures: See umbonal sculptures.

Character state: The specific “value” taken by a character in a specific taxon (e.g. for character ‘colour’, character states ‘red’, ‘green’…).

Clade: A monophyletic group of two or more species.

Comarginal: Parallel to the shell margin; also sometimes referred to as concentric.

Concentric: See comarginal.

Crown group: A group consisting of living representatives, and their ancestors, back to the most recent common ancestor of that group.

Convergence (adj. convergent): Independent evolution of similar traits in unrelated lineages; also called homoplasy.

Corrugated: Marked by wrinkles or ridges and grooves.

Ctenidium: A thin, platelike paired organ within the mantle cavity which serves as an organ of respiration and food-gathering in unionoids, and at least partly, as marsupium in female unionoids; usually suspended by tissue or cilial junctions from the dorsal region of the animal and typically comprising a W-shaped doubly folded lamella on each side of the visceral mass; each side typically consisting of an inner and outer demibranch.


xiv

Demibranch: See ctenidium.

Deposit feeding: Feeding type during which organic particles are harvested from sediments; see also suspension feeding.

Divaricate: Branching, usually in reference to external sculpture.

Distal: Situated away from point of attachment or origin; terminal end.

Dorsal: The hinge side of a bivalve; opposite of ventral.

Edentulous: Without hinge teeth.

Eutrophic: Nutrient-enriched and consequently highly productive (water body).

Exhalant siphon/aperture: Aperture or siphon that controls water outflow from the mantle cavity.

Filter feeding: Feeding type involving the filtering of organic particles from water by the gills, after which appropriately sized particles are transported to the mouth; see also suspension feeding, deposit feeding.

Foot: Muscular organ at the ventral part of the visceral mass; used by contraction and expansion for locomotion, burrowing and/or anchoring a bivalve.

Gill: See ctendidium.

Glochidium (pl. glochidia): The bivalve larvae of freshwater mussels in the superfamily Unionoidea which are generally parasitic on fish.

Hermaphrodite (adj. hermaphroditic): Sexually mature animal in which male and female gametes are produced by the same individual, either simultaneously or in sequence; see also protandric and protogynous.

Hinge: Dorsal border of the articulated valves, including the ligament, hinge teeth, and other structures that function to permanently unite the two valves; also called hinge line.

Hinge dentition: See hinge teeth.

Hinge line: See hinge.

Hinge teeth: A series of calcified dorsal interlocking teeth and sockets that allow alignment of the valves to be maintained during opening and closing.

Homology (adj. homologous): Similarity between characteristics of organisms that is due to their shared ancestry.

Homoplasy (adj. homoplasic): See convergence.

Infauna (adj. infaunal): Organisms which live in soft sediment and are large enough to displace sediment.

Glossary

xv

Inhalant siphon/aperture: Aperture or siphon, usually posterior, that controls water intake into the mantle cavity.

Interstitium (adj. interstitial): Pore spaces between the grains of rock and sediments.

Lateral: Situated at or extending to the side.

Lentic: Standing water environment.

Ligament: Elastic structure that connects the two bivalve shells at the hinge line and functions as a spring to open the valves when the adductor muscles relax.

Lotic: Running water environment.

Mantle: Fleshy outer tissue surrounding the organs of a molluscan body and secreting the periostracum and shell; consisting of two lobes in a bivalve, one lining each shell, and at the ventral edge several folds that may have different functions or features.

Mantle cavity: Chamber between the mantle lobes and interior visceral mass.

Marsupium (pl. marsupia): In unionoids, a brood pouch for eggs and developing glochidia, formed by the complete ctenidia or a restricted portion of the ctenidia.

Monophyletic: Containing a common ancestor and all of its descendants (pertaining to a taxonomic group); defined by synapomorphies.

Muscle scar: Impression on the shell interior that indicates the attachment position of a muscle.

Nonindigenous: Not native to an area.

Obesity: See relative shell width.

Ontogeny (adj. ontogenetic): The developmental history of an organism.

Oocyte: Immature egg cell.

Oogonia: Earliest recognizable form of the egg.

Palaeoenvironment: The environment of a former period of geologic time.

Pedal: By means of the foot.

Periostracum: Outermost layer of the shell; secreted by the mantle.

Plesiomorphy (adj. plesiomorphic): Ancestral state of a character in an evolutionary analysis.

Polyphyletic: Taxon that does not contain the most recent common ancestor of its members.

Posterior: Hind or rear; anal end.


xvi

Prismatic: Shell microstructural variety consisting of parallel columnar prisms of calcium carbonate.

Protandric/protandrous: A form of hermaphroditism in which the male phase precedes the female phase during the life cycle of the same individual; see also hermaphrodite.

Protogynous: Condition in a sequential hermaphrodite in which female gonads mature before male gonads.

Radial: In this thesis referring to external sculptural features that originate at a central point at the umbo and fan outward toward the margins.

Random genetic drift: The process of change in the genetic composition of a population due to chance or random events rather than by natural selection.

Relative shell length: Degree of elongation across the antero-posterior axis.

Relative shell width: Degree of inflation across the lateral shell axis.

Rib: An elongated sculptural element that is raised above the surrounding shell surface; also called ridge.

Ridge: See rib.

Riverine: Characteristic of rivers.

Rugae: See umbonal sculptures.

Sagittal: A sagittal plane is an imaginary plane that travels vertically from the top to the bottom of the body, dividing it into left and right portions.

Sculpture: Ornament or markings on the shell surface.

Semi-infaunal: Partially infaunal.

Sexual dimorphism: Condition in which males and females of the same species are morphologically different.

Shell height: Maximum sagittal shell diameter perpendicular to shell length (usually across the dorso-ventral axis).

Shell length: Maximum sagittal shell diameter (usually across the antero-posterior axis).

Shell width: Maximum lateral shell diameter.

Siphon: Posterior extension (usually two) of the mantle, made tubular by either tissue fusion or ciliary junctions of the mantle folds, through which water, waste products and gametes are directed in and out of the body.

Sister clades: Taxa with a common ancestor and no additional descendents.

Spermatozoa: The male reproductive cell.

Glossary

xvii

Suspension feeding: Feeding type during which organic particles are harvested from the water column.

Symphynote: See alate.

Synapomorphy (adj. synapomorphic): A shared, derived, taxon-defining trait or characteristic.

Teeth: See hinge teeth.

Transverse: Situated or lying across; crosswise.

Umbo (pl. umbos or umbones; adj. umbonal): The raised portion of the dorsal margin of a shell that reflects the early growth stage (= oldest part of the shell); formed by the embryonic shell around which the rest of the shell develops distally in a concentric manner; also called beak.

Umbonal sculptures: Sculptures on the umbo; also called beak sculptures or rugae.

Valve: The right or left half of a bivalve shell.

Ventral: The underside or foot-side of a bivalve; opposite of dorsal.

Visceral mass: Region of the bivalve body containing most of the digestive, excretory, circulatory, and nervous systems, that is suspended dorsally between the gills and that usually terminates ventrally as the foot.

Wings: See alae.

CHAPTER 1

INTRODUCTION

3

CHAPTER 1

INTRODUCTION

Freshwater mussels of the order Unionoida (also known as pearly mussels or naiads)

show an extremely wide range in shell characters. The notable differences between

unionoid species occur in shell size, form, sculpture and hinge dentition (Fig. 1.1), and

microstructure. In addition to this interspecific morphological variability, shell

morphology often varies considerably within single unionoid species (Fig. 1.2).

Figure 1.1. A-F. Some examples for interspecific variation in unionoid shell characters. Scale bars = 3 cm. Shells displayed with dorsal margins facing to the head and anterior margins to the (A-C) right and (D-F) left of the page, respectively. A. Physunio superbus (Lea, 1841), Indonesia. B. Lamellidens marginalis (Lamarck, 1819), India. C. Cyclonaias tuberculata (Rafinesque, 1820), USA. D. Parreysia tavoyensis (Gould, 1843), India. E. Theliderma cylindrica (Say, 1817), USA. F. Pleiodon ovata (Swainson, 1823), Senegal. All specimens deposited at the University Museum of Zoology Cambridge.

Understanding the factors that drive such inter- and intraspecific morphological

patterns could be helpful to various fields of research, including the understanding of

bivalve evolution, conservation of rare and threatened species, and the interpretation of

palaeoenvironments. In contrast to their marine relatives (e.g. Stanley, 1970; Vermeij,

1993), freshwater bivalves have so far received comparatively little attention in this

respect. In particular, the functional morphologies of these shell characters and the extent

to which morphological patterns are associated with differences in genotype remain poorly


4

understood. It is the aim of this thesis to fill some of these gaps in our understanding of

unionoid shell morphologies and the mechanisms involved in determining interspecific

and intraspecific variation.

Figure 1.2. A-I. Some examples for intraspecific variation in shell shape of Unio pictorum (Linnaeus, 1758) in the UK. Scale bars = 3 cm. Left valves displayed with anterior margins facing to the left and dorsal margin to the head of the page. A. Cut-off Channel near Hilgay, Norfolk. B. River Meden, Nottinghamshire. C. Lake Llangorse, Wales (NMW.1920.393.001). D. Welford Reservoir, Northamptonshire. E. Canal near Bath, Somerset. F. Kennet and Avon Canal, Somerset. G. Wateringbury Stream, Kent. H. River Lee, Essex. I. River Frome, Dorset (NMW.1955.158). Specimens from (A) private collection, (B,D,E,G,H) the Natural History Museum London, (C,I) the National Museum of Wales and (F) the University Museum of Zoology Cambridge.

1.1. A BRIEF INTRODUCTION TO THE UNIQUENESS OF UNIONOIDS

1.1.1. UNIONOID PHYLOGENY, DIVERSITY AND DISTRIBUTION

Unionoids are filter feeding freshwater bivalves with a distribution spanning all

continents except Antarctica. Besides their wide spatial distribution, this group of

freshwater mussels is also an ancient clade, with the oldest known representatives of their

extant crown group dating from the Triassic (Haas, 1969b; Watters, 2001). Though the

phylogenetic origin of the Unionoida is currently unknown, they possibly descended from

Chapter 1 - Introduction

5

ancient members of the marine Trigonioida (Newell & Boyd, 1975; Giribet & Wheeler,

2002).

Current unionoid diversity is still poorly recorded but latest global estimates range

from about 800 species (Bogan, 2008; Bogan & Roe, 2008) to approximately 900 species

(Graf & Cummings, 2006b, 2007b). Based on phylogenetic analyses by Graf & Cummings

(2006b; Fig. 1.3A) using combined morphology + mitochondrial (COI) DNA + nuclear

ribosomal (28S) DNA data, Graf & Cummings (2007b) represents the most recent attempt

of a global checklist of freshwater mussel species. These authors list a total of 840

described unionoid species, which they classify into 166 genera, six families and two

superfamilies (Table 1.1). The large number of incertae sedis taxa (i.s., i.e. "of uncertain

placement”) in Graf & Cummings’ (2007b) classification reflects our incomplete

understanding of unionoid systematics and phylogeny, which is further evidenced by

contradictory phylogenetic cladograms obtained by other authors. For example, analyses

based on mitochondrial (COI) DNA (Bogan & Hoeh, 2000) and combined mitochondrial

(COI) DNA + morphology data (Hoeh, Bogan & Heard, 2001) (Fig. 1.3B) recovered a

polyphyletic status of both the “Unionidae” and the freshwater oysters (“Etheriidae”). The

respective positions of the Hyriidae and Margaritiferidae within the Unionoida are further

issues currently under intense scientific debate (see Graf & Cummings, 2006b; Hoeh et al.,

2009; Graf & Cummings, 2010a).

1according to Graf & Cummings (2006b, 2007b) member of the Etheriidae

Figure 1.3. A-B. Two contrasting phylogenetic cladograms of the Unionoida (both redrawn from Hoeh et al. (2009)). A. Combined evidence phylogeny based on morphology + mtDNA (COI) + nuclear ribosomal DNA (28S) data (Graf & Cummings, 2006b: fig. 4). B. Combined summary tree based on mtDNA (COI) (Bogan & Hoeh, 2000) and morphology + mtDNA (COI) (Hoeh, Bogan & Heard, 2001).


6

Table 1.1. Unionoid systematics after Graf & Cummings (2007b).

Superfamily UNIONOIDEA Genera

Family UNIONIDAE Subfam. UNIONINAE

Tribe UNIONINI Acuticosta, Arconaia, Cafferia, Cuneopsis, Inversiunio, Lanceolaria, Lepidodesma, Nodularia, Rhombuniopsis, Schistodesmus, Unio

Tribe ANODONTINI Alasmidonta, Anemina, Anodonta, Anodontoides, Arcidens, Cristaria, Lasmigona, Pegias, Pseudanodonta, Pyganodon, Simpsonaias, Simpsonella, Sinanodonta, Strophitus, Utterbackia

Subfam. AMBLEMINAE Tribe AMBLEMINI Amblema Tribe LAMPSILINI Actinonaias, Arotonaias, Cyprogenia, Cyrtonaias,

Delphinonaias, Disconaias, Dromus, Ellipsaria, Epioblasma, Friersonia, Glebula, Hamiota, Lampsilis, Lemiox, Leptodea, Ligumia, Medionidus, Obliquaria, Obovaria, Potamilus, Ptychobranchus, Toxolasma, Truncilla, Venustaconcha, Villosa

Tribe PLEUROBEMINI Cyclonaias, Elliptio, Elliptoideus, Fusconaia, Hemistena, Lexingtonia, Plethobasus, Pleurobema, Popenaias, Uniomerus

Tribe QUADRULINI Amphinaias, Megalonaias, Plectomerus, Quadrula, Quincuncina, Theliderma, Tritogonia

Tribe GONIDEINI Gonidea i. s. Mesoamerican Ambleminae Barynaias, Martensnaias, Micronaias, Nephritica, Nephronaias,

Pachynaias, Psoronaias, Psorula, Reticulatus, Sphenonaias i.s. UNIONIDAE

Western Palearctic: Leguminaia, Microcondylaea, Potomida, Pseudodontopsis Afrotropical: Brazzaea, Coelatura, Grandidieria, Mweruella, Nitia,

Nyassunio, Prisodontopsis, Pseudospatha Indotropical: Arcidopsis, Caudiculatus, Chamberlainia, Contradens,

Ctenodesma, Diaurora, Discomya, Elongaria, Ensidens, Haasodonta, Harmandia, Hyriopsis, Inversidens, Lamellidens, Lamprotula, Modellnaia, Oxynaia, Parreysia, Physunio, Pilsbryoconcha, Pressidens, Prohyriopsis, Protunio, Pseudobaphia, Pseudodon, Ptychorhynchus, Rectidens, Scabies, Schepmania, Solenaia, Trapezoideus, Unionetta

Family MARGARITIFERIDAE Margaritifera

Superfamily ETHERIOIDEA

Family HYRIIDAE Subfam. HYRIINAE

Tribe HYRIINI Prisodon Tribe CASTALIINI Callonaia, Castalia, Castaliella Tribe RHIPIDODONTINI Diplodon, Rhipidodonta Tribe HYRIDELLINI Cucumerunio, Echyridella, Hyridella, Virgus

Subfam. VELESUNIONINAE Alathyria, Lortiella, Microdontia, Velesunio, Westralunio

Family ETHERIIDAE Acostaea, Bartlettia, Etheria, Pseudomulleria

Family MYCETOPODIDAE Subfam. MYCETOPODINAE Mycetopodella, Mycetopoda Subfam. ANODONTITINAE Anodontites, Lamproscapha Subfam. LEILINAE Leila Subfam. MONOCONDYLAEINAE Diplodontites, Fossula, Haasica, Iheringella, Monocondylaea,

Tamsiella

Family IRIDINIDAE Subfam. IRIDININAE Chelidonopsis, Mutela, Pleiodon Subfam. ASPATHARIINAE Aspatharia, Chambardia, Moncetia


7

1.1.2. IMPORTANCE AND CONSERVATION STATUS

Our insufficient understanding of unionoid evolution and diversity is particularly

unfortunate given that freshwater mussels are amongst the most critically endangered

groups of animals worldwide (Lydeard et al., 2004; Strayer, 2006). In fact, in North

America alone, about 36 unionoid species (10%) are already presumed extinct (Neves et

al., 1997). Most important causes for their decline include habitat loss and degradation,

loss of host fishes (see 1.1.3) and the introduction of nonindigenous species (Bogan, 1993;

Williams et al., 1993; Neves et al., 1997). As a result, two of the six unionoid species

inhabiting freshwater habitats in the UK, are considered to be a national conservation

priority, appearing on the UK Biodiversity Action Plan priority list (Aldridge, 2004).

At the same time, Unionoida have important ecosystem functions such as particle

filtration and processing, nutrient release and sediment mixing (Vaughn & Hakenkamp,

2001). As a consequence, freshwater mussels are increasingly recognised as keystone

fauna (Aldridge, Fayle & Jackson, 2007), and their decline can profoundly affect

ecosystem processes in aquatic habitats (Geist & Auerswald, 2007). In addition, unionoids

are used by people, both as a food source and for pearl production (Anthony & Downing,

2001). Finally, these shell-bearing organisms can be used as environmental indicators,

thereby providing critical data for monitoring habitat changes (e.g. Mutvei & Westermark,

2001).

1.1.3. LIFE CYCLE AND MORPHOLOGICAL VARIATION

Besides a practical interest due to their conservation status, and ecological and

economic importance, freshwater mussels are particularly interesting from a biological

perspective. Unionoids exhibit a unique life cycle incorporating both parental care (i.e.

brooding) and, in most species, obligate larval parasitism upon freshwater fishes (Kat,

1984; Wächtler, Mansur & Richter, 2001) (Fig. 1.4). The parasitic stage, lasting from a

few weeks to several months, is believed to have evolved as a means of distribution of

these otherwise relatively immobile animals (Kat, 1984). As a further consequence of their

hosts’ mobility, however, habitat conditions in which the juvenile mussel excysts from the

host will be rendered highly unpredictable. This unpredictability of the “new” habitat(s) of

a given mussel’s offspring is believed to be a major source of the wide morphological

variability characteristic of many unionoid species.


8

Figure 1.4. A-E. The unionoid life cycle. A. Schematic drawing of a typical unionoid life cycle. Sperm released by an adult male via his exhalant siphon is infiltrated by the female through her inhalant siphon. In the female gills (marsupia), fertilised eggs develop into mature larvae (glochidia), which are released into the water column. Glochidia attach and encyst into the gills, fins or scales of freshwater fishes. After this parasitic phase, the then fully transformed juvenile mussel drops off the fish and begins its life in the sediment. B. Glochidial shell of Anodonta anatina (Linnaeus, 1758). C. Teeth on the external surface of the glochidial hook in A. anatina. D. Glochidia of Sinanodonta woodiana (Lea, 1834) encysted into fins of the Golden Line Fish (Sinocyclocheilus grahami Regan, 1904) (photograph by Holly Barclay); shell lengths approx. 330 µm. E. 2 year-old juvenile of Margaritifera margaritifera (Linnaeus, 1758); shell length approx. 2 mm.

1.2. IMPORTANCE OF UNDERSTANDING MORPHOLOGICAL PATTERNS IN UNIONOID SHELLS

The morphological makeup (phenotype) of a given mussel is a result of two main

mechanisms: (1) its genetic composition (genotype) and (2) environmental or other non-

genetic factors triggering changes in morphology.

On the species and higher taxonomic levels, ‘genetic inheritance’ results in similar

traits of related taxa due to common ancestry (‘homologies’). Similarities in life habit,

environmental conditions and/or other factors, on the other hand, can lead to independent


9

evolution of similar morphological solutions (‘homoplasies’ or ‘convergences’) in

unrelated taxa.

Equally, morphological differences within single species can be a result of either

differences in the genotype or due to plasticity of the phenotype (Via et al., 1995).

Intraspecific patterns in shell morphology lacking a genetic basis may directly or indirectly

be caused by differences in, for example, sex, levels of parasitic infestation or

environmental conditions. The ‘habitat’ factor is regarded of particular importance to

intraspecific patterns in unionoid shell morphologies. However, though such intraspecific

ecomorphotypes may simply reflect phenotypic plasticity, these could also be a result of

selective survival of only those genotypes that are most adequately equipped for the

respective environmental conditions. The unique unionoid life cycle potentially allows

both natural selection and phenotypic plasticity to operate, but the relative importance of

these two mechanisms is currently unknown.

In conclusion, shell morphology may reflect both the (phylo)genetic history and

other characteristics, such as life habit and habitat, of a given species, population or

individual. Whereas a morphological characteristic that has predominantly been influenced

by inheritance can tell us about a mussel’s genetic composition and/or phylogenetic

position, morphological convergences between or within species can contain information

on the character’s function. An accurate understanding of inter- and intraspecific patterns

in shell morphology can thus be valuable for various fields of research, some of which are

discussed in the following.

1.2.1. SYSTEMATICS, PHYLOGENY AND EVOLUTION

Early unionoid taxonomists (e.g. Lea, 1870; Simpson, 1900; Simpson, 1914; Modell,

1942) based their classifications predominantly on shell traits and, in particular, shell form,

sculpture and hinge dentition. By including anatomical, life history and molecular

characters in their phylogenetic analyses, other and subsequent authors (e.g. Ortmann,

1912; Prashad, 1931; Haas, 1969a, b; Davis, 1983, 1984), however, recognised widespread

convergences in shell morphologies of unrelated unionoid taxa. In addition, the large

extent of intraspecific variation, most notably in shell form and sculpture, has in the past

led to considerable synonymy (e.g. Rossmässler, 1835-1837; Küster, 1848; Locard, 1890)

and still causes taxonomic confusion. The high degree of interspecific convergence and


10

intraspecific variability renders these shell characters unsuitable for addressing questions

regarding unionoid taxonomy and phylogeny.

The hard part of unionoid and other bivalves generally represents, on the other hand,

the only source of information available for the study of fossil taxa. A good understanding

of present homologies, convergences and the extent of intraspecific variation in shell

morphologies is therefore a prerequisite for resolving ancient freshwater mussels’

systematics and phylogeny. Knowledge on the evolution of a given shell character, and

thereby, the morphological characteristics of basal (plesiomorphic) and derived

(apomorphic) character states, can be particularly useful in this respect. At the same time,

this information can help us answer unresolved questions regarding modern unionoids’

evolution.

1.2.2. ECOLOGY, CONSERVATION AND ENVIRONMENTAL RECONSTRUCTION

Identifying habitat characteristics associated with particular morphological traits of a

given unionoid species, population or individual, can provide valuable information about

its ecology. This knowledge can in further course help identify habitat requirements of

endangered species and assist in their management. In addition, comparing

ecomorphotypes of recent shells with similar fossilised ones can enable the reconstruction

of palaeoenvironments. The application of unionoids to these purposes has so far been

limited due to a lack of an adequate understanding of which shell morphologies are

triggered by which environmental conditions.

At species and higher taxonomic levels, certain convergences in shell morphologies

of unrelated taxa sharing particular ecological characteristics, are relatively well

understood. For example, ‘symphynote’ species, characterised by a dorsal, “wing-shaped”

extension of the shell, termed ‘alae’ or ‘wings’, can typically be found in standing/slow-

flowing, fine sediment water bodies. Alae have independently arisen several times within

the Unionoida, are believed to prevent sinking into soft substrates and are commonly

associated with a laterally compressed form, reduced dentition and a lack of sculpture of

the shell. Unionoid species with well developed shell sculpture (such as macroscopic

‘ridges’ or ‘pustules’), on the other hand, are often associated with fast-flowing habitats

(Watters, 1994).

Whereas identification of interspecific convergences requires knowledge of the

extent of genetic inheritance of the respective shell trait, this does not apply to consistent


11

ecophenotypic trends within a species. The often extreme degree of intraspecific variation

in unionoid shell morphology could thus be particularly valuable for the aforementioned

fields of research. Despite the considerable amount of literature on habitat-shell

morphology associations within unionoid species (e.g. Wetherby, 1882; Ortmann, 1920;

Agrell, 1948), most of these studies suffer from one or more caveats, including (1) a lack

of statistic evidence, and (2) failure to consider influence of non-habitat factors that might

influence shell morphology.

1.2.3. RECONSTRUCTION OF POPULATION PARAMETERS

In addition to those concerning a mussel’s environmental surroundings, other factors

can influence the morphology of a given specimen (Seed, 1980). These include (1) sexual

dimorphism, (2) allometric growth during ontogeny, and (3) morphological alteration due

to parasitic infestation (Roper & Hickey, 1994; Scholz, 2003). While it is crucial to

consider the relative influence of such non-habitat factors when attempting to elucidate

ecophenotypic patterns, knowledge on how morphology changes with various parameters

can also be helpful in its own right. Identification of sexual shell dimorphisms can, for

example, facilitate quick sex determination in the field and/or the reconstruction of ancient

populations’ sex ratios.

The importance of the respective non-habitat and habitat factors on particular

morphological trends in freshwater mussel species are still poorly understood. Sexual shell

dimorphism in Unionoida is known predominantly from the North American tribe

Lampsilini, whereas patterns in European groups remain unresolved. The extent to which

allometric growth and/or parasitic infestation influence morphology within unionoid

species remains almost completely unresolved.

1.3. AIMS AND QUESTIONS ADDRESSED IN THIS THESIS

The aim of this thesis is to improve our understanding of the factors that determine

inter- and intraspecific patterns in unionoid shell morphology. Such an understanding can

ultimately be used in the reconstruction of evolutionary trends and phylogenies,

environments and habitat requirements of species, and other characteristics of individuals

and populations such as their sex ratios or parasitic loads. Broadly, three types of shell

characters were studied: (1) gross shell morphology (i.e. size, shape, inflation and

thickness), (2) umbonal (‘beak’) sculpture, and (3) periostracal microsculpture.


12

1.3.1. “GROSS SHELL MORPHOLOGY”: SIZE, SHAPE, INFLATION AND THICKNESS

Due to a high degree of interspecific convergences and intraspecific variability, the

shell characters of ‘size’ (at a given age), ‘shape’, ‘inflation’ and ‘thickness’ are

considered to be of poor taxonomic and phylogenetic value. However, an accurate

understanding of which factors trigger such trends within a given species could yield

considerable information.

Our lack of knowledge in this respect is reflected by conflicting observations of

habitat-morphology associations reported by previous authors. In particular, descriptions

of how shell form changes with certain habitat parameters and consequentially, the

functional explanations given, are often contradictory. Reasons for such inconsistencies

may include inappropriate measurements of habitat, failure to consider influence of non-

habitat factors and random genetic drift, and inadequate choice of morphological features.

Applying traditional and modern morphometric techniques, Chapter 2 investigates

intraspecific trends in gross shell morphology and internal characters within three

European unionoid species. The main objectives were to (1) detect consistent, truly

ecophenotypic intraspecific trends in unionoid shell shape, (2) determine the importance of

various habitat and non-habitat factors on an individual’s morphology, (3) elucidate

internal morphological characteristics associated with certain changes in shell morphology,

and (4) discuss probable functional advantages of shell ecomorphotypes found.

In Chapter 3, I use Amplified Fragment Length Polymorphisms (AFLPs) to ask

whether intraspecific morphotypes identified in Chapter 2 are due to genetic differences or

phenotypic plasticity. Chapter 4 investigates the importance of non-habitat parameters in

determining intraspecific trends in shell morphology. In particular, the roles of sexual and

parasite-induced dimorphism are investigated.

1.3.2. UMBONAL SCULPTURE

Sculptures restricted to the early shell region (‘umbonal sculptures’, ‘beak

sculptures’ or ‘rugae’) have long been used for identification of unionoid species and,

occasionally, genera. Specificity of this shell character at higher taxonomic levels and

consequently, its phylogenetic value and adequacy for classification of fossil taxa, is still

under considerable scientific debate. To a large part, this is due to a lack of understanding

of unionoid beak sculpture character evolution. Knowledge on synapomorphic,

plesiomorphic and convergent character states can help address unresolved questions on


13

unionoid phylogeny and systematics, but might additionally help formulate likely

functional hypotheses of this shell character. Ultimately, this could give novel insights on

ecological aspects of this poorly understood, but crucial life stage.

Based on examination of over 150 ancient and modern species, Chapter 5 discusses

patterns of interspecific variability and presents a new model of character evolution of

umbonal sculptures in the Unionoida.

1.3.3. PERIOSTRACAL MICROPROJECTIONS

Periostracal microprojections, observable only by high magnification (electron)

microscopy, have been little studied in palaeoheterodont (i.e. unionoid and trigonioid)

bivalves. However, recent studies on other bivalve groups (Glover & Taylor, 2010; Checa

& Harper, in press) indicate that these structures are probably of considerable phylogenetic

value. In Chapter 6, I analysed specimens covering all six unionoid families and the only

extant genus of the Trigonioida. This enabled me to provide the first comprehensive

review of periostracal microprojections in the Palaeoheterodonta, and to discuss their

possible functional morphologies and importance in phylogenetic reconstruction.

1.4. STYLE OF THESIS

Chapters 2-6 are written in paper-style and slightly altered versions are either

published (Zieritz & Aldridge, 2009; Zieritz et al., 2010; Zieritz et al., in press) or

currently under review (Zieritz & Aldridge, in review; Zieritz, Bogan & Aldridge, in

review) by a peer-reviewed scientific journal.

CHAPTER 2

IDENTIFICATION OF ECOPHENOTYPIC TRENDS WITHIN THREE EUROPEAN

FRESHWATER MUSSEL SPECIES (BIVALVIA:UNIONOIDA) USING TRADITIONAL AND

MODERN MORPHOMETRIC TECHNIQUES

“Rafinesque collected the Unionidae extensively in Kentucky and

published a large number of genera, minor groups, and

species…His figures are more like those made by children, or the

caricatures drawn by aboriginal tribes, than the creations of an

intelligent naturalist…”

Simpson (1900)

17

CHAPTER 2

IDENTIFICATION OF ECOPHENOTYPIC TRENDS WITHIN THREE EUROPEAN FRESHWATER MUSSEL SPECIES (BIVALVIA:

UNIONOIDA) USING TRADITIONAL AND MODERN MORPHOMETRIC TECHNIQUES

2.1. ABSTRACT

Most species of freshwater mussels (Unionoida) show a wide variability in shell

form and size but an understanding of which factors determine unionoid morphology is

poor. We identified ecophenotypic trends in shell and internal characters within three

unionoid species from two habitat types (marinas and river) of the River Thames, UK,

using traditional and modern morphometric techniques. In marinas, all species grew to

larger maximum sizes than in the river, which might be a result of higher temperatures and

phytoplankton densities in marinas. Unio pictorum in marinas was more elongated than in

the river and Fourier shape analysis revealed a trend from dorsally arched river specimens

to straight dorsal and pointed posterior margins in marina individuals. The degree of shell

elongation and shape of dorso-posterior margin were not associated with sediment

composition, but were associated with the different hydrological characters of the two

habitat types. Relative shell width was a poor indicator of collection site and influenced by

allometric growth. Unlike U. pictorum, a difference in shell elongation of marina and river

mussels could not be detected in Unio tumidus and Anodonta anatina. However, all three

species showed the same trends regarding the shape of the dorso-posterior shell margin.

This shell character may thus have broad ecological significance and could have

considerable utility to palaeontologists, taxonomists and conservation biologists.

2.2. INTRODUCTION

Freshwater mussels of the order Unionoida display a wide range of intraspecific

morphological variability (Fig. 1.2), which has in the past led to considerable synonymy

(Küster, 1848; Lea, 1870; Locard, 1890). Besides the obvious value for taxonomic

research, identifying habitat characteristics associated with given phenotypes of a species

can provide valuable information about its ecology, assist in the management of


18

endangered species (Bogan, 1993), and enable the reconstruction of palaeoenvironments.

In addition to those concerning a mussel’s habitat, other factors such as its sex or

ontogenetic stage might influence the morphology of a given specimen (Seed, 1980;

Tevesz & Carter, 1980). However, the importance of the respective non-habitat and habitat

factors on particular morphological trends in freshwater mussel species and the

mechanisms involved are still poorly understood.

The relationship between morphotype and habitat has been studied in unionoids for

over 100 years (Hazay, 1881; Buchner, 1910; Israel, 1910; Haas & Schwarz, 1913;

Ortmann, 1920; Grier & Mueller, 1926; Bloomer, 1938). However, such descriptions,

usually based on measurements of the three shell dimensions (length, height and width),

are often contradictory. For example, in their study on 18 unionoid species at the

Mississippi River system, USA, Grier & Mueller (1926) found that in a slow flowing, fine

sediment river-lake, some species displayed more elongated shells (decreased shell height

to length ratio) than in the main channel, whereas others showed the exact opposite

pattern. Sometimes even populations within the same species are apparently “reacting” in

an opposing manner. Cvancara (1972) found Lampsilis radiata (Gmelin, 1791) to grow

relatively longer shells in finer substrates of Long Lake, Minnesota, whereas in Lake Erie,

Hinch, Bailey & Green (1986) observed shells of the same species to be more elongated in

sand than in mud.

Three factors may explain the apparent contradictions in the relationship between

shell form and habitat. First, the characters under study might simply not be associated

with habitat. For example, unionoids are known to switch from an interstitial deposit

feeding to a suspension feeding mode of life at a certain point in their early life (Yeager,

Cherry & Neves, 1994), which might be accompanied by a change in allometric growth.

Other non-habitat factors potentially influencing growth and morphology in unionoids are

a shift of metabolism at sexual maturity, sexual dimorphism or genetic differences of

geographically distant populations possibly caused by random genetic drift. Although

infestation by certain parasites has shown to induce abnormal shell growth in New Zealand

unionoids (Roper & Hickey, 1994), no such growth-altering parasites are currently known

from European waters (but see Chapter 4). Second, inappropriate measurements of habitat

may have been made with the key environmental determinant being overlooked. Finally,

the choice of morphological features (typically the three shell dimensions) may not

provide sufficient description of morphology. Ecologically more significant trends are

likely to be elucidated by analysing more accurate morphological descriptions, such as the

Chapter 2 – Ecophenotypic trends in freshwater mussels

19

whole outline of a shell. Statistical analysis of the degree of difference of, for example,

sagittal shell outlines has been made feasible only recently by the development of

advanced morphometric methods such as Fourier shape analysis. These tools are widely

used by palaeontologists but, unfortunately, have found only sporadic application in

ecological studies on recent freshwater unionoids (Scholz, 2003).

Intraspecific shell morphotypes that are consistently correlated to particular habitat

conditions could potentially be adaptations to the same. For example, short shells might be

advantageous under high current velocities. Alternatively, a given shell shape might

simply be a result of certain adaptive internal characteristics of the respective mussel

(Kauffman, 1969; Stanley, 1970); for example, it might be that, in a given species, larger

feet are advantageous in providing better anchorage in fast flows; this could result in

relatively large anterior parts of the shell. Thus, detecting changes in internal characters

associated with those in shell forms could help explain why certain shell forms are present

in certain habitats.

In the present study, we combined an ecological study design (i.e. five replicate

paired sites) with traditional and advanced morphometric techniques aiming to statistically

estimate morphological patterns within three European unionoid species. The important

study aims were to: (1) compare traditional and advanced morphometric methods in

detecting morphological patterns in the species; (2) elucidate internal morphological

characteristics associated with certain changes in shell morphology; (3) determine the

importance of various habitat factors (i.e. sediment, water movement, temperature, food

availability) and non-habitat factors (i.e. ontogenetic factors (age and size), sex,

geographic distance of populations) on an individual’s morphology; and (4) estimate

consistence of morphological patterns over replicates and species.

2.3. MATERIALS AND METHODS

2.3.1. SAMPLING OF INDIVIDUALS

From May to September 2007, freshwater mussel populations at five marina and five

adjacent river sites of the River Thames from Abingdon to Old Windsor were surveyed

(Fig. 2.1). Marinas were directly connected to the river, and paired sampling sites were no

more than 800 m apart. Marinas were characterised by areas in the approximate range 10-

80 ha and maximum water depths in the approximate range 1.5-4 m, and were constructed

at least 30 years ago (i.e. longer than the longevity of any of the mussel species studied;


20

Aldridge (1999b)). The River Thames in the stretch studied has an average width of 50 m,

and flow is regulated by a series of weirs, such that mean summer and winter discharge

values are approximately 40 m³s-1 and 80 m³s-1 (data from the UK Environment Agency),

respectively .

Figure 2.1. Sampling area and the five sampling locations at the River Thames. At each location one marina and one river site was sampled. Species sampled for morphological analysis at respective locations were Unio pictorum (grey circles), Unio tumidus (filled circle) and Anodonta anatina (open circle). Map © Crown Copyright/database right 2008. An Ordnance Survey/EDINA supplied service.

Relative mussel densities and species abundances at each site were estimated by

counting the number of individuals per species present in 15 replicate dredges

(approximately 5 m in length, dredge aperture 46 x 21 x 7.5 cm, mesh size = 2.5 cm).

To elucidate (dis-)similarities in morphological trends, at least 15 Unio pictorum

(Linnaeus, 1758) individuals were collected by hand at each marina and river site and

brought back to the laboratory for subsequent morphological analysis. Additionally, the

two other unionoids present in the study area, Anodonta anatina (Linnaeus, 1758) and

Unio tumidus Philipsson, 1788, were sampled at one of the five locations (Abingdon;

Fig. 2.1). Overall, a total of 181 U. pictorum, 39 U. tumidus and 39 A. anatina specimens

were investigated regarding their morphological characters. To minimise morphological

differences between populations resulting from differences of ontogenetic stages of their

individuals, all mussels collected were more than 4 years old based on external annual

rings (Haskin, 1954), and sagittal shell areas between populations of the same species were

not significantly different (ANOVA: U. pictorum: F9,181 = 1.868, P = 0.060; U. tumidus:

F1,38 = 0.236, P = 0.630; A. anatina: F1,38 = 0.735, P = 0.397). All specimens analysed have

been deposited at the University Museum of Zoology Cambridge.


21

2.3.2. SAMPLING OF HABITAT PARAMETERS

At each site, one measurement of water temperature was made using a digital

thermometer. Five replicate measurements of chlorophyll a concentrations were also taken

using an in vivo fluorometer (Aquafluor Handheld Fluorometer, Turner Designs,

Sunnyvale, California). Measurements at marina and river pairs were made within a period

of 1h.

At each site, five random sediment cores were taken down to a depth of 10-15 cm.

Sediment samples were dried to constant weight and sieved through a sieving tower of six

sieves with a mesh size in the range 2-31.5 mm. Sediment fractions <2 mm grain size were

analysed in a Malvern Mastersizer 2000 giving relative volume-proportions of grain size

classes to the 0.01% level. Assuming a similar specific density for all grain sizes, the

weighing of the total respective <2 mm- and >2 mm-grain size proportions allowed the

conversion of volume-proportions to weight-proportions. Finally, median grain sizes of

samples were calculated from their cumulative weight percentage plots. The organic

matter content of each sediment sample was estimated by calculating weight loss on

ignition (24h at 600°C).

2.3.3. GROWTH MEASUREMENTS, AND AGE AND SEX DETERMINATION OF INDIVIDUALS

The age of each U. pictorum specimen was estimated by counting annual winter

rings displayed on the shell (Aldridge, 1999b). For shape-independent analysis of growth,

lengths and heights of individuals at each of these annuli were measured with a digital

calliper, and sagittal shell areas (SaA) at each year of age were estimated by applying the

standard formula for an ellipse. Square root of sagittal shell area (SaA1/2) per age plots

were produced for each site by calculating the mean SaA1/2 at each year. Growth

parameters were determined using the Walford plot model (Walford, 1946) assuming von

Bertalanffy growth curves (von Bertalanffy, 1938). All specimens were sexed by

microscopic inspection of gonadal tissue sensu Heard (1975).

2.3.4. MORPHOLOGICAL ANALYSIS

2.3.4.1. Fourier shape analysis

To compare two-dimensional shell outlines of the individuals, we used Fourier shape

analysis sensu Crampton & Haines (1996). This multivariate morphometric tool analyses


22

the whole sagittal outline of the shell, making it ideal for bivalves that lack the sufficient

number of homologous landmarks needed for landmark-based morphometric techniques.

In Fourier shape analysis, an outline contour is decomposed into a number of basic

waves, termed harmonics, with each of them in turn explained by two respective Fourier

coefficients. The more harmonics, and therefore Fourier coefficients, calculated, the more

accurately the description of the outline will be. Each shell outline is so described by a set

of Fourier coefficients that can be statistically treated as any usual variable. Because

Fourier coefficients contain no size information, a standardisation of the outlines to the

same size is not necessary.

For analysis, digital photographs of right valves of all specimens were taken. After

enhancing contrast between shell and background, shell outlines were digitised with

program ImageJ (Rasband, 2008). These digitised outlines were then used for Fast Fourier

Transform using the HANGLE software (Crampton & Haines, 1996). Before calculating

Fourier coefficients, a smoothing normalisation of 10 was applied for reduction of high-

frequency pixel noise resulting from the digitisation process. Preliminary analyses showed

that the first eight harmonics explained the shell outlines with sufficiently high precision,

whereby the first harmonic does not contain any shape information and, thus, is discarded

from the analysis by the software. This resulted in a set of 14 Fourier coefficients per

individual. Finally, HMATCH software was used to rotate all outlines treated in one

statistical analysis (e.g. all outlines of one species) for maximum overlap. The sets of

Fourier coefficients obtained this way were subsequently used in statistical analysis

described below.

2.3.4.2. Analysis of traditional shell and anatomical measurements

Figure 2.2 illustrates shell and anatomical measurements taken. For traditional shell

shape analysis the three shell dimensions length, height and width were measured at each

specimen. Additionally, several anatomical measurements were taken to elucidate changes

in internal characters associated with those in shell forms. Sagittal cross-sectional areas of

the adductors and unfused “siphons” (apertures) were estimated by measuring their dorso-

ventral and antero-posterior radii and applying the standard formula for the area of an

ellipse. Gills were dissected from the mussel and areas measured. Excised tissue (dissected

foot and remaining soft parts separately) was dried to constant mass (60°C for 48h).

Sagittal shell area (SaA) and shell volume (V) were estimated by applying the standard


23

formulas for an ellipse and ellipsoid, respectively, using the measurements of the three

shell dimensions.

Figure 2.2. Shell and anatomical measurements taken at each specimen. Abbreviations: aA, anterior adductor area; eS, exhalant ‘siphon’ area; F, dry foot weight; H, shell height; iG, inner gill area; iS, inhalant “siphon” area; L, shell length; oG, outer gill area; pA, posterior adductor area; W, shell width. Additionally, the dry weight of all soft tissues of the animal was measured.

2.3.5. STATISTICAL ANALYSIS

Statistical analyses were performed using PAST (Hammer, Harper & Ryan, 2001;

Hammer & Harper, 2006) and MINITAB for Windows, Version 14 (Minitab Inc.).

Altogether eight separate principal component analyses (PCA) were carried out

using either sets of 14 Fourier coefficients or five internal morphological ratios for four

respective groups of populations: U. pictorum from all ten sites together, and each of the

three unionoid species from the Abingdon location separately. The number of principal

components (PCs) to be retained was determined by use of the broken stick model of the

scree plot. Only the first two PCs were used for graphical representation in scatter plots.

Synthetic outlines of “extreme shell forms” were drawn using HCURVE software sensu

Crampton & Haines (1996).

To estimate importance of non-habitat factors in relation to the habitat factor on

morphological trends, three three-way-analyses of covariance (ANCOVAs) were

performed on relative shell length, relative shell width, and the first PC obtained by PCA

on Fourier coefficients on all ten U. pictorum populations using factors habitat, location

and either (1) sex, (2) covariate size, or (3) covariate age. Eta-squared values calculated

this way are a measure of proportion of variance explained by each factor or covariate,

respectively.


24

2.4. RESULTS

2.4.1. HABITAT PARAMETERS

Marinas were approximately 0.5°C warmer than adjacent river sites (paired: t =

3.627, P = 0.022, N = 5; Table 2.1). No significant differences could be observed between

marina and river sites in chlorophyll a concentration (Wilcoxon: z = 1.753, P = 0.080, N =

5), median grain size (paired: t = 1.989, P = 0.117, N = 5) and percent organic matter

content of the sediment (paired: t = 1.867, P = 0.135, N = 5).

Table 2.1. Chlorophyll a, median grain sizes, organic matter content of sediment (means ± SD) and water temperature at the ten sites. Capital letters indicate the ‘location’; lower case the ‘habitat’ of the site: A, Abingdon; C, Caversham; M, Marlow; W, Windsor; O, Old Windsor; m, marina; r, river.

Site Temperature (°C) Chl a [μg L-1] Q50 (grain size, µm) % Organic matter Am 13.9 7.6 ± 0.8 4433 ± 3001 4.23 ± 0.37 Ar 13.2 6.6 ± 0.3 3780 ± 5823 4.46 ± 2.65 Cm 14 5.0 ± 0.1 9191 ± 6705 3.89 ± 1.32 Cr 13.1 5.8 ± 0.2 13724 ± 11847 3.06 ± 1.66 Mm 13.5 8.4 ± 0.8 5162 ± 6997 13.26 ± 7.36 Mr 13 5.8 ± 1.3 11453 ± 9321 2.71 ± 2.05 Wm 13.7 24.4 ± 1.1 1160 ± 2037 5.87 ± 5.79 Wr 13.5 5.8 ± 0.5 12845 ± 3471 1.56 ± 0.68 Om 13.7 434.5 ± 11.0 1379 ± 1525 8.31 ± 6.22 Or 13.5 5.7 ± 0.3 1668 ± 3404 5.97 ± 2.05

2.4.2. SPECIES COMPOSITION

Figure 2.3. Relative abundances of the three unionoid species present at the ten sites. Numbers above each column correspond to overall number of specimens dredged at each site. Abbreviations: m, marina site; r, river site.


25

Mussel densities (as estimated by number of mussels per 15 replicate dredges) were

not significantly different between the two habitat types (all species combined; paired: t =

0.043, P = 0.968, N = 5). Although the same three freshwater mussel species were present

at all ten sites, A. anatina was more abundant in the marinas, comprising 39% of the

mussel population compared to only 12% in the rivers, whereas U. tumidus was more

dominant in the river sites than in the marinas (53% versus 19%; Fig. 2.3). U. pictorum

was the most abundant species and comprised similar proportions in the marina and river

(42% and 35%, respectively) (pooled marina versus river populations: 2χ = 56.679, d.f. =

2, P < 0.01).

2.4.3. GROWTH

Figure 2.4. Average shell size (given as square root of sagittal shell area (SaA1/2) per age for pooled marina (full circles) and river (empty circles) Unio pictorum of ten populations of the River Thames.

Growth constants, i.e. the rates at which the asymptotic size is approached,

calculated by the Walford plot were not significantly different between marina and river U.

pictorum (growth constant, mean ± SD; marina = 0.88 ± 0.02; river = 0.85 ± 0.02; paired: t

= 1.92, P = 0.128, N = 5). However, shell size (measured as sagittal shell area; SaA1/2) per

age plots of pooled marina versus river populations indicate that, from the age of

approximately 5 years, individuals from the marina were generally larger than same aged

specimens living in the river (Fig. 2.4), resulting in significantly larger maximum sizes of

marina than river U. pictorum (SaA1/2 of largest individual per population; mean ± SD;

marina = 51.9 ± 1.9 mm; river = 44.4 ± 2.5 mm; paired: t = 6.43, P = 0.003, N = 5).

Annual growth rings of U. tumidus and A. anatina were not distinctive enough to allow for

reliable calculation of growth parameters using the Walford plot model (Walford, 1946).


26

However, in both species, maximum shell sizes were larger in the respective marina

populations (A. anatina: SaA1/2max = 70.4 mm at marina, SaA1/2

max = 56.0 mm at river; U.

tumidus: SaA1/2max = 53.8 mm at marina, SaA1/2

max = 51.2 mm at river).


2.4.4.1. Morphological trends in U. pictorum of five paired sites and influence of habitat and non-habitat factors

Figure 2.5 displays scores of 181 U. pictorum specimens for (A) relative shell

elongation and obesity, and the first two PCs obtained by analyses using (B) five internal

morphological ratios and (C) 14 Fourier coefficients, respectively. All three scatter plots

show a clear clustering of river versus marina individuals along their first respective axes,

indicating that, across all five paired sites, U. pictorum from the same habitat were

morphologically more similar to each other than those from the same location.

Synthetic outlines (Fig. 2.5C) and photographs (Fig. 2.6) visualise sagittal shapes of

“extreme” habitat forms. Marina shells typically displayed a pointed posterior and straight

dorsal margin, whereas those from the river tended to be dorsally arched. Additionally,

marina U. pictorum were comparatively more elongated than river forms, which is shown

both by modern (Fig. 2.5C) and traditional (Fig. 2.5A) morphometric techniques. These

trends in sagittal shell shape were accompanied by relatively higher dry soft tissue weight

and larger adductor muscles in the marina mussels compared to those in the river

(Fig. 2.5B). Scatter along subsequent PC axes retained by the broken stick model, but not

displayed in Figure 2.5 (i.e. PC3 of Fourier coefficients explaining 11% of variance),

reflected variation within populations and did not discriminate marina from river mussels.

Eta-squared values obtained by three three-way-ANCOVAs indicate that the shape

of the dorso-posterior margin (PC1 of Fourier shape analysis) was mainly influenced by

the habitat the mussel was living in, and that neither ontogenetic stage nor sex of an

individual had any considerable effect in this respect (Fig. 2.7). The degree of shell

elongation was significantly influenced by factors habitat and location but also the

interaction of these two factors. This indicates that ecophenotypic trends in this shell

character differed between locations. Although shell obesity was significantly influenced

by several habitat and non-habitat factors/covariates and their interactions, this character

was especially associated with the habitat-location interaction factor (Fig. 2.7). This is also

indicated in Figure 2.5A, demonstrating that mussels of the Marlow and Windsor marinas

displayed generally lower relative shell width values than those from all other sites.


27

Figure 2.5. A-C. Shell morphological characters of 181 Unio pictorum specimens of ten populations of the River Thames. A. Relative shell length versus relative shell width: SaA, cross sectional shell area; L, shell length; W, shell width. B. Principal component (PC) scores for first two PC axes obtained by PC analysis (PCA) on five internal morphological ratios: Adductors, (summed anterior and posterior adductor area)/(sagittal shell area); Foot, (foot dry weight)/(soft tissue dry weight); Gills, (summed inner and outer gill area)/(sagittal shell area); Siphons, (summed inhalant and exhalant “siphon” area)/(sagittal shell area); Soft, (soft tissue dry weight)/(shell volume). C. PC scores for first two PC axes obtained by PCA on 14 Fourier coefficients: Synthetic shell outlines of ”extreme” morphotypes are displayed with anterior margin facing to the right and dorsal margin to the head of the page.


28

Figure 2.6. A-B. Photographs of ‘extreme’ (A) marina (Windsor), and (B) river (Abingdon) morphotypes of Unio pictorum.

Figure 2.7. A-C. Proportions of total variance in the shell morphologic parameters relative shell length (rel. L; given as L/SaA1/2), relative shell width (rel. W; given as W/SaA1/2), and shape of dorso-posterior margin (PC1 Fourier; given as PC1 scores obtained by principal component analysis on 14 Fourier coefficients) of ten Unio pictorum populations, respectively, as a result of variation in the factors habitat (Hab), location (Loc) and (A) sex (Sex), (B) covariate sagittal shell area (Size; i.e. measurement for size), or (C) covariate age (Age). Only significant interactions are shown as vertical bars.

2.4.4.2. Morphological trends in all three unionoid species at Abingdon

In all three unionoid species at Abingdon, scatter along the first two axes of PCAs on

14 Fourier coefficients show similar patterns with regard to a pronounced clustering of

individuals from the same habitat, and a trend from relatively straight dorsal and pointed

posterior margins in marina mussels to downwardly curved dorso-posterior margins in

river forms (Fig. 2.8G,H,I). However, the elongation of marina forms was more

pronounced for U. pictorum than for A. anatina and could not be observed in U. tumidus

(Fig. 2.8A,B,C,G,H,I). Marina specimens of A. anatina show a trend to more pronounced

wing development than river individuals (Fig. 2.8I). Finally, all three species showed the

same trend for relatively larger adductor muscles and higher soft tissue weight per shell

volume in marina mussels compared to those from the main channel of the River Thames

(Fig. 2.8D,E,F). Subsequent significant principal components retained but not displayed in

Figure 2.8 (i.e. PCA on Fourier coefficients; U. pictorum: PC3 11%; U. tumidus: 12% of


29

variance) reflected variation within populations and did not discriminate marina from river

mussels.

Figure 2.8. A-I. Shell morphological characters of three unionoid species at one marina (full circles) and one river site (empty circles) at the River Thames at Abingdon. Small circles, individual specimens; large circles, centroids. A, D, G. Unio pictorum (N = 45); B, E, H. Unio tumidus (N = 39); C, F, I. Anodonta anatina (N = 39). A, B, C. Relative shell length versus relative shell width, and principal component scores for first two PC axes obtained by PCAs on (D, E, F) five internal morphological ratios and (G, H, I) 14 Fourier coefficients. Abbreviations: SaA, sagittal shell area; L, shell length; W, shell width; Adductors, (summed anterior and posterior adductor area)/(sagittal shell area); Foot, (foot dry weight)/(soft tissue dry weight); Gills, (summed inner and outer gill area)/(sagittal shell area); Siphons, (summed inhalant and exhalant ‘siphon’ area)/(sagittal shell area); Soft, (soft tissue dry weight)/(shell volume). Synthetic shell outlines of “extreme” morphotypes are displayed with anterior margin facing to the right and the dorsal margin to the head of the page.


30

2.5. DISCUSSION

The main aim of the present study was to identify ecophenotypic trends in unionoid

shell and internal morphology that could find application in taxonomy, palaeontology, and

conservation of rare species. We revealed a number of consistent morphological patterns,

both within and between species, which characterise mussels collected from river or

marina sites.

2.5.1. PATTERNS ACROSS ALL SPECIES

In general, marina-collected specimens grew larger, had relatively higher soft tissue

dry weight, larger adductors, were less dorsally arched and had a more pointed posterior

than river-collected specimens. Shell size at a given age in freshwater mussel species has

often been shown to be associated with changes in water temperature and/or food

(especially phytoplankton) availability (Grier, 1920; Negus, 1966; Reigle, 1967; Ghent,

Singer & Johnson-Singer, 1978). In the present study, in marinas, all species grew to larger

maximum sizes than those in the river. This could be explained by the higher water

temperatures in marinas, which can support a longer growing season and faster metabolism

(Mann, 1965). Mussel growth might additionally have been enhanced by higher

phytoplankton densities that would be expected in the standing, relatively warm water

bodies of marinas compared to the flowing, colder river. Lack of a complete dataset on

seasonal patterns may explain why we were unable to detect significant differences in

chlorophyll a levels between the two habitats. The higher quality of growing conditions in

the marinas is also supported by the mussels’ higher soft tissue dry weight per shell cavity

volume ratios, comprising a measurement commonly used as bivalve body condition index

(Crosby & Gale, 1990).

The observation of relatively large adductors in marina specimens compared to those

in the river may be a side-effect of differences in relative soft tissue mass at the two

habitats. Under such a scenario, adductor size may therefore be considered as non-

adaptive. Adductors are known to play an important role in bivalve burrowing and

anchorage (Trueman, 1966a, b, 1968), and larger and therefore stronger adductors would

be expected to be more valuable in stronger water movement. Our observation of

proportionately larger muscles in slower flowing habitats is clearly in conflict with this

theory and other observations (Sell, 1907-1908; Balla & Walker, 1991). It is quite possible

that internal anatomy is not independent of shell morphology, and so adductor size might


31

also be affected by the shape of the shell. However, in contrast to our observations, in the

Australian hyriid unionoid Alathyria jacksoni Iredale, 1934, larger, more powerful

adductors were found to be associated with dorsal arching (Balla & Walker, 1991).

Without experimental data, it is not possible to identify whether adductor size is adaptive

in its own right, or whether it reflects physical constraints by shell morphology. Although

the reasons for larger adductors in marina sites remain unresolved, it is striking that the

pattern was found across all species. As such, it could be indicative of lentic environments

for A. anatina, U. pictorum and U. tumidus.

Table 2.2. Comparison of three shell shape characters: (1) ‘relative elongation’, (2) ‘obesity’, and (3) ‘shape of dorso-posterior margin’, regarding their suitability as ecophenotypic shell form characters.

Consistent across replicate populations

Consistent across different species

Not largely influenced by non-habitat factors

(1) Elongation YES NO YES (2) Obesity NO NO NO (3) Dorso-posterior margin YES YES YES

The second trend consistent across all three species of this study was that from more

pointed posterior margins in marina specimens to more dorsally arched river mussels

(Table 2.2), which was not influenced by the non-habitat factors studied (i.e. age, size, and

geographic location). Although a specific habitat parameter associated with this change in

shell shape could not be identified, the results obtained in the present study indicate that,

rather than sediment composition, hydrological parameters such as type of water

movement (e.g. lotic versus lentic) or mean/maximum current velocities might be

determining the shape of the dorso-posterior margin of unionoids in the River Thames and

its marinas. Similar observations have been made in other unionoid families, including

Hyriidae (Balla & Walker, 1991) and Margaritiferidae, and Carboniferous Anthracosiidae

(Eagar, 1948, 1978). These shape differences might have adaptive significance. In a series

of studies, Eagar (1948, 1971, 1974, 1977, 1978) argued that dorsal arching results in

relatively heavy shells and modifies the hinge structure in such a way that the pedal gape is

increased, allowing the foot to extend further into the sediment. These features would

increase initial probing force, anchorage and the stability of the bivalve when subject to

lifting forces resulting from turbulent water. In addition to Eagar’s arguments, the changes

in shell shape might cause a shift of the centre of gravity of the shell and/or a change in

extent of sediment scouring (Menard & Boucot, 1951; Watters, 1994), possibly enhancing


32

the effects mentioned above. Nevertheless, the differences in shell shapes could have no

adaptive value whatsoever but be caused by a non-functional reaction of the mussel to the

environment (e.g. via distortion of the shell secreting mantle margin as a result of water

movement). Intraspecific variation in life position such as burrowing depth or angle

(Amyot & Downing, 1991; Scholz, 2003) could further influence shell growth and should

be considered in this respect. Although the functional significance of posterior pointing

remains unresolved, the consistency of this shell form trend indicates that the shape of the

dorso-posterior margin could be a powerful ecophenotypic character in freshwater

mussels, and be valid for a wide range of species and habitats.

2.5.2. PATTERNS WITHIN SINGLE SPECIES

Some shell form trends, such as that regarding the degree of wing development in

A. anatina, were observed only in one of the three species studied. Wings (alae) are

considered to act as secondary ligaments or hinge teeth making up for the reduced

dentition of most symphynote species (Savazzi & Peiyi, 1992), whereas, as a result of their

well developed dentition, members of the genus Unio are not known to increase their

dorsal margin surface in such a manner. Watters (1994) argued that enlargement of wings

within a given species might enhance the “snowshoe-effect” (i.e. preventing sinking and

providing increased stability in soft or unstable substrates), which is a probable

explanation for our observations of bigger alae in marina specimens. The degree of shell

elongation, on the other hand, was markedly different between populations in the two

habitats for U. pictorum but could not be detected in the other two unionoid species.

Similar observations on inconsistent ecophenotypic trends of unionoid species are

common in the literature. For example, several studies reported that, in the genus Unio,

more elongated shells were associated with slower flowing habitats (Bridgeman, 1875;

Hey, 1882; Israel, 1910; March, 1910-1911; Tudorancea, 1972), although contradictory

observations have been made on Margaritifera margaritifera (Linnaeus, 1758) (Altnöder,

1926). Such inconsistencies of intraspecific patterns might root in both ecological and

behavioural differences between individuals or populations. For example, microhabitat can

affect the degree of embedding in the sediment (Agrell, 1948), which could lead to

differences in physical forces acting on the mussels and, in turn, influence their shell

formation. Additionally, dominance of A. anatina and U. tumidus at marina and river sites,

respectively, indicates that these two species show stronger habitat specificity than the


33

more generalist U. pictorum. On this basis, generalists would be expected to show a higher

degree of phenotypic variability, which could explain our failure to detect ecophenotypic

trends in shell height to length ratios in A. anatina and U. tumidus.

By contrast to patterns in sagittal shell shape, relative shell width (obesity) of

U. pictorum proved to be considerably influenced by allometric growth. Similar alterations

in growth pattern with size and/or age have commonly been described in both marine

(Ohba, 1959; Johannessen, 1973) and freshwater mussels (Ball, 1922; Scholz, 2003).

Nevertheless, relative shell width has often been used as an ecophenotypic character,

which has even led to the formulation of the ‘Law of Stream Distribution’ (Ortmann,

1920) with numerous studies supporting (Grier, 1920; Ball, 1922; Grier & Mueller, 1926;

Reigle, 1967; Cvancara, 1972; Anderson & Ingham, 1978; Hinch, Bailey & Green, 1986;

Hinch, Kelly & Green, 1989) as well as contradicting the same (March, 1910-1911; Grier,

1920; Bailey & Green, 1988; Hinch & Bailey, 1988). An additional cause for this apparent

contradiction in pattern of obesity is likely to stem from the non-independence of the shell

measurements (length, width and height) used in traditional morphometric studies. A

consideration of allometric growth patterns can overcome such problems.

2.5.3. UTILITY OF THE PATTERNS OBSERVED

Knowing the range of morphological variation within a species is doubtlessly

important for unionoid taxonomy. Furthermore, understanding the factors certain

morphological trends are associated with can help reconstructing palaeoenvironments

(Good, 2004; Scholz & Hartmann, 2007a, b; Scholz, Tietz & Büchner, 2007) and aid in

conservation of rare or threatened species. These are often known predominantly from

shell material, with their habitat requirements poorly understood. Perhaps a comparison of

morphology with commoner congenerics can be informative, provided that features are

consistent across species (e.g. degree of arching). This may help understand reasons for a

decline (e.g. habitat loss) and also identify habitat where surveys for rare species can be

focused.

At least in temperate latitudes where annual growth rings are usually well

established, maximum sizes of shells are likely to be a good indicator of mean/maximum

water temperatures and phytoplankton densities at the habitat, and could thus be used for

palaeoenvironmental reconstructions and estimation of condition of the freshwater mussel

population at the site. Adductor scar sizes in shells could further help estimate the


34

condition of an individual, although other factors, such as current velocity, should be taken

into account as well. According to Claassen (1998), shell shape is perhaps the best method

for reconstructing habitat with freshwater molluscs. The results obtained in the present

study indicate that the degree of dorsal arching is one of the most generally valid

characters in unionoids and, thus, modern rather than traditional morphometric techniques

should be used in future studies of this kind. The shape of the posterior dorsal margin

appears to be associated with hydrological character of the habitat and could thus be used

in both palaeontological reconstructions and identification of habitat requirements of (rare)

species.

CHAPTER 3

PHENOTYPIC PLASTICITY AND GENETIC ISOLATION-BY-DISTANCE IN THE

FRESHWATER MUSSEL UNIO PICTORUM(MOLLUSCA: UNIONOIDA)

“In 1892 Arnould Locard, one of the great lights of the new school,

stated that there were 208 species of Unios and 250 Anodontas in

France alone. Life is too short and valuable to be wasted in any

attempt at deciphering such nonsense, and I have not even

cumbered the pages of this work with a list of these new species.

Those interested can find them in the works of Westerlund and

Kobelt.”

Simpson (1900)

37

CHAPTER 3

PHENOTYPIC PLASTICITY AND GENETIC ISOLATION-BY-DISTANCE IN THE FRESHWATER MUSSEL UNIO PICTORUM

(MOLLUSCA: UNIONOIDA)

3.1. ABSTRACT

Freshwater mussels (Unionoida) show high intraspecific morphological variability,

and some shell morphological traits are believed to be associated with habitat conditions. It

is not known whether and which of these ecophenotypic differences reflect underlying

genetic differentiation or are the result of phenotypic plasticity. Using 103 amplified

fragment length polymorphism (AFLP) markers, we studied population genetics of three

paired Unio pictorum populations sampled from two different habitat types (marina and

river) along the River Thames. We found genetic differences along the Thames which

were consistent with a pattern of isolation by distance and probably reflect limited

dispersal via host fish species upon which unionoid larvae are obligate parasites. No

consistent genetic differences were found between the two different habitat types

suggesting that morphological differences in the degree of shell elongation and the shape

of dorso-posterior margin are caused by phenotypic plasticity. Our study provides the first

good evidence for phenotypic plasticity of shell shape in a European unionoid and

illustrates the need to include genetic data in order properly to interpret geographic

patterns of morphological variation.

3.2. INTRODUCTION

Many species of freshwater mussels of the order Unionoida display great variability

in shell morphology (Fig. 1.2), some of which appear to be associated with differences in

habitat (Chapter 2; Ortmann, 1920; Agrell, 1948; Eagar, 1978; Watters, 1994; Zieritz &

Aldridge, 2009). These so-called ecomorphotypes may arise through two mechanisms.

First, they may simply reflect phenotypic plasticity, the ability to change phenotype in

response to variation in the environment (Via et al., 1995). The alternative mechanism

involves genetic variability. If a species produces very large numbers of offspring, it is

possible that those that manage to settle in any given habitat represent a biased subset of


38

all offspring, specifically those able to do best under those conditions. The unionoid life

cycle potentially allows for both mechanisms to operate. Thus, the larvae of almost all

unionoid species are obligate fish parasites, making the habitat in which the juvenile

mussel excysts from the host highly unpredictable, and hence putting a premium on the

ability to thrive under diverse habitats. Equally, females brood eggs in marsupia which can

be fertilised by several males via the inhalant current, potentially producing genetically

diverse offspring (e.g. Kat, 1984). Which of these two mechanisms dominates in nature

remains open to debate.

Few genetic studies have explicitly investigated the role of genetic versus phenotypic

variability on morphological variation in unionoids, and these have typically used

relatively conservative DNA markers that lacked adequate resolution to detect genetic

differences among populations (e.g. Soroka & Zdanowski, 2001). Nevertheless, previous

publications have generated mixed support for the two hypotheses. For example, reciprocal

transplant experiments on two unionoid species between Canadian lakes showed that shell

height to length ratios usually change as a function of the environment, indicating

phenotypic plasticity of shell form. On the other hand, overall growth rates appear to be

under genetic control (Hinch, Bailey & Green, 1986; Hinch & Green, 1989).

Transplant experiments offer a direct test of whether phenotypic plasticity operates,

but are often undesirable ecologically. Alternative tests are based on searches for a

correlation between phenotype and genotype. For example, Buhay et al. (2002) and

Machordom et al. (2003), studied morphologically distinct Epioblasma and Margaritifera

“subspecies” respectively, but failed to find differences based on allozymes and/or

mitochondrial genes. On the other hand, Serb, Buhay & Lydeard (2003) found that

mtDNA sequence data supported the validity of several Quadrula taxa that had originally

been classified on the basis of shell morphology but were later lumped with other species

on the basis that this simply reflected phenotypic plasticity. Moreover, in their study on

genetics of central European freshwater pearl mussels (Margaritifera margaritifera

Linneaus, 1758), Geist & Kuehn (2005) observed that, while in some cases

morphologically atypical mussels showed a strong genetic divergence to other populations,

in other populations a link between genetic status and shell shape was not evident.

Such inconsistent findings hint at a deeper problem. In looking for a correlation

between phenotype and genotype it is natural to select individuals from opposite ends of

the observed morphological range. These will often be drawn from habitats that are not

only divergent in ecological properties, but also widely separated geographically.

Chapter 3 – Phenotypic plasticity in Unio pictorum

39

Consequently, the resulting samples may appear genetically different simply through a

long-term lack of gene flow, rather than the presence of discrete lineages adapted to each

habitat. Since most of the work on unionoid population genetics has so far focused on

systematics and phylogeny (e.g. Davis & Fuller, 1981; Serb, Buhay & Lydeard, 2003;

Källersjö et al., 2005), conservation genetics (e.g. Mulvey et al., 1997; Buhay et al., 2002;

Geist & Kuehn, 2005) and evolutionary history (e.g. Nagel, 2000; Huff et al., 2004;

Elderkin et al., 2007), morphological differences of populations in these studies were

usually accompanied with geographic distance. This makes it difficult to assess if genetic

differences between such populations are merely a result of isolation by distance (i.e.

evidence for phenotypic plasticity) or actually reflect differences in morphology (i.e.

evidence for genetically induced morphological differences).

Here, we attempt to distinguish between phenotypic plasticity and genetic adaptation

by exploiting an unusual system in which large habitat differences can be found over very

short geographic distances. Unio pictorum (Linnaeus, 1758) is an abundant species in the

River Thames and can be found living both in the river itself, and in adjacent marinas (Fig.

3.1). By sampling from paired sites along the river, we are able to ask whether any genetic

differences are due primarily to a simple isolation by distance model, or to the existence of

two genetically distinct morphs, one adapted to the marina habitat and one to the river.

Genetic analysis was conducted using Amplified Fragment Length Polymorphisms

(AFLPs) (Vos et al., 1995; Blears et al., 1998), a technique that reveals large numbers of

variable traits capable of resolving finescale differentiation within a population (Mueller &

Wolfenbarger, 1999).

3.3. MATERIAL AND METHODS

3.3.1. SAMPLING

From May to October 2007, a total of 146 U. pictorum specimens were collected by

hand from three river and three adjacent marina sites of the River Thames (Fig. 3.1). A

foot tissue sample was removed from each mussel and stored in 96% ethanol for

subsequent genetic analysis. All specimens and genetic samples have been deposited at the

University Museum of Zoology Cambridge.

Marinas were directly connected to the river and the paired sampling sites were no

more than 800 m apart, whereas geographic distances between the three sampling locations

ranged from about 30-100 km. The River Thames in the stretch studied has an average


40

width of ca. 50 m, and maximum depth of ca. 4 m. Flow is regulated by a series of weirs,

such that mean summer and winter discharges approximate 40 to 80 m3 s-1, respectively

(data from UK Environment Agency). Marina sites were typical lentic systems with no

flow, surface areas of 10 to 80 ha and maximum depths up to 4 m. The river sites typified a

UK lowland lotic system. Chapter 2 and Zieritz & Aldridge (2009) showed that water

temperatures and phytoplankton densities in River Thames marinas were consistently

higher than in the adjacent main river channel.

Figure 3.1. Sampling locations along the River Thames. Sample sizes are shown for each marina and river site (see methods section for details). Map © Crown Copyright/database right 2008, an Ordnance Survey/EDINA supplied service.


Shell length (L; maximum diameter) and shell height (H; maximum diameter

perpendicular to L) were measured in each individual to ±0.1 cm using a vernier calliper.

To estimate relative influence of habitat and location on the degree of shell elongation

(H:L-ratio), we constructed a general linear model (GLM; Minitab 15 for Windows). For

this, H was the response variable with L and geographic distance (measured as distance

from most upstream site, i.e. Abingdon Marina, to ±10 m) fitted as covariates and habitat

as a factor with two levels. To find the minimum adequate model, the full model was fitted

and then simplified by sequentially removing non significant predictor variables (Crawley,

2002).

For a more detailed analysis of shell morphology, we used Fourier shape analysis, as

developed and explained by Crampton & Haines (1996). This method decomposes xy-


41

coordinates of an outline into a number of harmonics, each of which is in turn explained by

two Fourier coefficients, which can be analysed statistically like any other traits. Digital

photographs of all specimens were taken and digitised using the program IMAGEJ

(Rasband, 2008). The digitised outlines were then subjected to fast Fourier transformation

using the program HANGLE, applying a smoothing normalisation of 20 to eliminate high-

frequency pixel noise. Preliminary analysis indicated that the first 10 harmonics described

the outlines with sufficiently high precision. Discarding of the first harmonic, not

containing any shape information, by the program resulted in a set of 18 Fourier

coefficients per individual. After rotating outlines to maximum overlap by program

HMATCH, principal component analysis (PCA) was performed on the 18 Fourier

coefficients using program PAST (Hammer & Harper, 2006). Number of principal

components to be retained was determined by use of the broken stick model of the scree

plot. Synthetic outlines of “extreme shell forms” were drawn using program HCURVE as

explained in Crampton & Haines (1996).

To test for statistical significance of habitat and location on sagittal shell shape,

GLMs of all significant principal components were carried out, fitting habitat (as a factor

with two levels) together with geographic distance as predictors, and sequentially

dropping insignificant factors from the model (Crawley, 2002). Morphological distances in

overall sagittal shell outline between the six populations were estimated by performance of

discriminant function analysis (DFA) and subsequent calculation of Mahalanobis’s D²

distances of each population pair using Minitab 15 for Windows. Age and sex were not

included as predictors in GLMs because sexual dimorphism and allometric growth during

ontogeny exert a negligible influence on relative shell elongation and overall sagittal shell

shape in the six populations studied (Chapter 2; Zieritz & Aldridge, 2009). Similarly, the

degree of shell inflation (relative shell width) was not considered, in this case because, in

contrast to shell elongation and sagittal shell shape, this morphological character is

strongly influenced by ontogenetic growth in unionoids (Chapter 2; Zieritz & Aldridge,

2009).

3.3.3. GENETIC ANALYSIS

Total genomic DNA was extracted from a small piece of foot tissue using a high salt

method (Watts, 2001). AFLP genotyping was based on the protocol of Vos et al. (1995)

and is described in detail by Dasmahapatra, Hoffman & Amos (2009). Briefly, 100-400 ng


42

of genomic DNA was first digested using TaqI (5 U in a 10 µl volume at 65 °C for 2

hours) and then with EcoRI (5 U in a 20 µl volume at 37 °C for 2 hours). TaqI and EcoRI

and adapters (Ajmone-Marsan et al., 1997) were then ligated onto the digested DNA using

T4 DNA ligase (1 U in a 50 µl volume at 37 °C for 3 hours), and the resulting products

diluted by a factor of ten in 10 mM Tris HCL and EDTA (0.1 mM, Ph 8.0). For the pre-

amplification, 5 µl of ligation mix was added to a 50 µl PCR reactions containing Tris-HCl

(10 mM, pH 8.3), MgCl2 (1.5 mM), KCl (50 mM), dNTPs (0.2 mM), Taq polymerase (1

U) and 50 ng each of the TaqI-C and EcoRI-A pre-amplification primers (the primer

sequences were 5’-GATGAGTCCTGACCGAC-3’ and 5’-GACTGCGTACCAATTCA-3’

respectively). Following 30 pre-amplification cycles (30 s at 94 °C, 60 s at 50 °C and 60 s

at 72 °C) the products were diluted 10 times with 10 mM Tris HCl and EDTA (0.1 mM,

pH 8.0).

Table 3.1. Primer combinations used for the AFLP selective amplification and numbers of monomorphic AFLP loci, and unambiguously scoreable and poorly resolved AFLP polymorphisms generated for 146 U. pictorum individuals.

TaqI primer (5'-3') EcoRI primer (5'-3') mono-morphic*

poly-morphic scored

poly-morphic poorly resolved*

GATGAGTCCTGACCGACAC GACTGCGTACCAATTCACA 16 10 13 GATGAGTCCTGACCGACAG GACTGCGTACCAATTCACA 27 24 11 GATGAGTCCTGACCGACAG GACTGCGTACCAATTCATG 36 6 38 GATGAGTCCTGACCGACCA GACTGCGTACCAATTCAGC 28 14 71 GATGAGTCCTGACCGACGA GACTGCGTACCAATTCAGC 25 10 11 GATGAGTCCTGACCGACGA GACTGCGTACCAATTCATG 15 7 33 GATGAGTCCTGACCGACTG GACTGCGTACCAATTCAAC 23 15 11 GATGAGTCCTGACCGACTG GACTGCGTACCAATTCAGC 15 17 26

* As with all AFLP analyses these values are somewhat arbitrary since inevitably there are bands which occur above the resolution of the gel end and cannot be scored.

For the selective amplification, 2.5 µl of the diluted pre-amplification product was

added to a 12.5 µl reaction containing Tris-HCl (10 mM, pH 8.3), MgCl2 (1.5 mM), KCl

(50 mM), dATPs, dTTP and dGTP (0.2 mM each), dCTP (0.04 mM), α33P-dCTP, Taq

polymerase (0.2 U), TaqI selective primer (30 ng) and EcoRI selective primer (5 ng).

Samples were subjected to 13 selective amplification cycles (30 s at 94 °C, 60 s at 65 °C,

reducing by 0.7 °C each cycle, and 60 s at 72 °C), followed by a further 23 cycles (30 s at

94 °C, 60 s at 56 °C and 60 s at 72 °C). Eight different selective primer combinations were

used (Table 3.1). PCR products were resolved by electrophoresis on standard 6%

polyacrylamide sequencing gels and detected by autoradiography. Exposed X-ray films


43

were assessed and if required, a second exposure was made for an adjusted time period.

Gels were scored manually.

3.3.4. QUANTIFICATION OF THE GENOTYPING ERROR RATE

Although AFLPs tend to be reproducible due to the use of highly specific restriction

endonucleases coupled with stringent PCR conditions (Vos et al., 1995; Mueller &

Wolfenbarger, 1999; Bonin et al., 2004), genotyping errors can nevertheless accrue, with

potential sources of error including DNA contamination (Dyer & Leonard, 2000),

restriction artefacts (Polisky et al., 1975), human error (Bonin et al., 2004) and variation in

DNA quality among samples. Consequently, we estimated the genotyping error rate for

our dataset by independently re-extracting, re-genotyping and blind-scoring 14 (9.6%) of

the samples following Hoffman & Amos (2005). To ensure broad coverage of the dataset

in this regenotyping exercise, two to three individuals were selected at random from each

of the six populations. The error rate per reaction was then quantified as the number of

mismatching genotypes divided by the number of polymorphic bands compared (Bonin et

al., 2004).

3.3.5. GENETIC DATA ANALYSIS

All gels were independently scored by two observers and genotypes recorded as ‘1’

= band present or ‘0’ = band absent. Data were entered into a spreadsheet. We used the

program AFLP-SURV v1.0 (Vekemans, 2002) to calculate estimated heterozygosity

values for each population and pairwise FST values between the six populations (Weir &

Cockerham, 1984) following the approach of Lynch & Milligan (1994). This program was

also used to conduct a permutation test for overall genetic differentiation using 10, 000

permutations of the dataset.

In order to test whether any of the AFLP loci scored could be subject to divergent

selection relating to habitat type and thus, potentially associated with shell morphology,

we used the Dfdist program package (Beaumont & Balding, 2004; Beaumont, 2008)

following the approach of Beaumont & Nichols (1996). Dfdist implements the Bayesian

method of Zhivotovsky (1999) to estimate allelic frequencies from the proportion of

recessive phenotypes (absent bands) and then compares FST values estimated for each

locus against a theoretical null distribution of genetic differentiation conditional upon

heterozygosity in a subdivided population using the coalescent and an island model of


44

migration. Loci that fall outside specified confidence limits can be identified as having

significantly elevated or reduced FST values relative to expectations under the assumption

of selective neutrality. This analysis was conducted separately on four population pairs: (1)

Pooled marina vs. river, (2) Abingdon marina vs. Abingdon river, (3) Marlow marina vs.

Marlow river and (4) Old Windsor marina vs. Old Windsor river.

For each analysis, parameters a = 0.25 and b = 0.25 were used for the beta-

distributed prior of the Bayesian allele frequency estimator and the trimmed mean FST was

computed by removing the 30% highest and lowest of the observed FST values. A null

distribution was generated based on 50 000 simulated loci, the parameter 4Nµ set to 0.5

and the target-neutral FST determined using the program pv2. Simulation models were run

using different small baseline FST values until the “correct” FST for the simulations was

obtained, i.e. half the neutral points were greater than/less than the median and no trend in

the proportion of p-values greater than 0.5 with increasing heterozygosity was observable

(Beaumont, 2008). The robustness of the results to variations in the simulation parameters

was further evaluated by repeating the analyses with varying values of 4Nµ (i.e. 0.2 and

1.0). Loci with FST values above the 99% quantile were inferred as being potentially under

directional selection. This approach bypasses the joint problem that trimmed mean FST (as

recommended by Caballero, Quesada & Rolan-Alvarez (2008) was slightly negative in all

cases, while a zero FST cannot be used because it prevents the coalescent from determining

a common ancestor (see Miller et al., 2007).

3.4. RESULTS


We began by constructing general linear models to explore which factors are

significantly associated with shell shape, summarised in Table 3.2. When shell height was

the response variable, habitat but not geographic distance was retained as a significant

predictor, indicating a strong morphological difference between mussels from the marina

and linked river habitats. This difference is also evidenced by the pronounced clustering of

marina and river individuals in a log height-length plot (Fig. 3.2). To learn about other

aspects of shape we summarised the 18 Fourier coefficients using a PCA analysis and then

fitted four further GLMs, one each using the first four principal components as the

response variable (Table 3.2). These four components were selected using the broken stick

model and together explain 57.1% of the total variance in sagittal shell shape.


45

Table 3.2. Results of general linear models (GLMs) of shell height and the four significant principal components obtained by Fourier shape analysis, comparing the six U. pictorum populations. Values correspond to the minimal adequate models (Crawley, 2002).

Response variable Source d.f. SS F P

Shell height Habitat 1 0.621 39.04 <0.0001 R² = 0.91 Geographic distance - - - n.s. Shell length 1 19.421 1219.98 <0.0001 Error 143 2.276 PC1 (Fourier) Habitat 1 0.033 133.03 <0.0001 R² = 0.49 Geographic distance 1 0.002 9.34 0.003 Error 143 0.035 PC2 (Fourier) Habitat - - - n.s. R² = 0.02 Geographic distance - - - n.s. Error - - PC3 (Fourier) Habitat - - - n.s. R² = 0.28 Geographic distance 1 0.010 57.44 <0.0001 Error 143 0.025 PC4 (Fourier) Habitat - - - n.s. R² = 0.00 Geographic distance - - - n.s. Error - -

Figure 3.2. Log shell length vs. shell height scatterplots of six U. pictorum populations sampled along the River Thames. Upper case indicates the location, lower case the habitat of the site: A, Abingdon; M, Marlow; O, Old Windsor; m, marina; r, river.

PC1 scores were significantly influenced by both habitat type and geographic

distance (Table 3.2). However, whereas fitting the GLM without the term ‘geographic

distance’ resulted in a decrease of only 2.6% of variance explained by the model, 46% of

the variance explained was lost following exclusion of the term ‘habitat’. This strong

habitat-shell shape association is also reflected by the pronounced clustering of river and


46

marina individuals in Fig. 3.3, which shows that marina mussels tended to have more

pointed posterior margins whereas river shells displayed more arched dorso-posterior

margins. While the second and fourth principal components were not significantly

influenced by any of the two factors, PC3 was significantly correlated with geographic

distance but not habitat (Table 3.2). With respect to extreme shell outlines along this axis

(Fig. 3.3), this suggests that the posterior region of U. pictorum from the Marlow and Old

Windsor populations was broader than those from the two Abingdon sites.

Figure 3.3. Principal component scores for the first four PC axes obtained by PCA on 18 Fourier coefficients. Synthetic shell outlines of “extreme” morphotypes are displayed with the anterior margin facing to the right and the dorsal margin to the top of the page.

3.4.2. POPULATION STRUCTURE

By genotyping a total of 146 individuals at eight different selective primer

combinations, we obtained 103 AFLP loci that could be scored unambiguously across

most of the samples (Table 3.1). The AFLP dataset analysed consisted of 13, 917 binary

characters representing the presence and absence of bands. The genotyping error rate was

2.4% (35 mismatches observed out of 1442 comparisons), which is broadly consistent with

a range of previously reported values for studies using AFLPs (Bensch & Åkesson, 2005;

Bonin, Ehrich & Manel, 2007).


47

Table 3.3. Number of polymorphic loci (out of 103 scored) and expected heterozygosity values for the six U. pictorum populations. Upper case indicates the location, lower case the habitat of the site: A, Abingdon; M, Marlow; O, Old Windsor; m, marina; r, river.

Population Number of polymorphic loci Expected heterozygosity ± SE Am 94 0.3274 ± 0.0149 Ar 91 0.3312 ± 0.0155 Mm 94 0.3400 ± 0.0147 Mr 97 0.3389 ± 0.0137 Om 100 0.3556 ± 0.0142 Or 94 0.3379 ± 0.0148

Table 3.4. (a) Genetic distance matrix (FST-values bottom left corner, P-values upper right corner), (b) Mahalanobis’ D² distance matrix obtained by DFA on 18 Fourier coefficients (F-values bottom left corner, P-values upper right corner). Upper case indicates the location, lower case the habitat of the site: A, Abingdon; M, Marlow; O, Old Windsor; m, marina; r, river.

(a) Genetic distance Am Ar Mm Mr Om Or Am 0.8864 0.0058 0.0012 <0.0001 <0.0001 Ar 0 0.0142 <0.0001 <0.0001 0.0012 Mm 0.0093 0.0082 0.1643 0.0092 0.0033 Mr 0.0118 0.0179 0.0017 0.0009 <0.0001 Om 0.0267 0.0251 0.0109 0.0153 0.0205 Or 0.0273 0.0149 0.0117 0.0221 0.009 (b) Morphological (Mahalanobis’ D² distance) Am Ar Mm Mr Om Or Am <0.0001 <0.0001 0.0004 0.0008 <0.0001 Ar 11.4 <0.0001 0.0208 <0.0001 0.0180 Mm 12.8 14. 5 0.180 1 0.0090 Mr 7.4 4.5 8.6 0.0064 0.7664 Om 9.9 16.1 3.6 7.2 0.0091 Or 8.5 5.5 7 2.5 7.8

Estimated heterozygosity values were similar in all six populations (Table 3.3).

Pairwise FST- and P-values among the six populations (Table 3.4a) show that almost all of

the populations were significantly different from each other except Abingdon marina from

Abingdon river and Marlow marina from Marlow river. In contrast, pairwise comparisons

of overall sagittal shell outline measured using Mahalanobis’ D² distances (Table 3.4b)

reveal that all population pairs except Marlow and Old Windsor marina and Marlow and

Old Windsor river were significantly different from each other. Moreover, the

geographically closest pairs of populations (i.e. the two respective populations of each

location) were also genetically the closest, whereas comparison of geographically most

distant ones (i.e. Abingdon vs. Old Windsor) resulted in comparatively high FST values,

indicating genetic isolation by distance (Fig. 3.4). Although there are too few sites to test

this pattern statistically using a Mantel test to control for non-independence, discrimination


48

by type of comparison (Fig. 3.4; marina vs. river, marina vs. marina and river vs. river)

reveals that this isolation-by-distance trend was largely driven by the paired river – marina

and marina – marina comparisons. In contrast, the river – river comparisons all show

rather similar genetic distances and the trend they yield is negative. Thus, if an isolation by

distance pattern does exist, it appears stronger in the marina – river and marina – marina

comparisons than in the river – river comparisons.

Figure 3.4. Geographic vs. genetic distance (FST values obtained by 103 AFLP loci) between six U. pictorum populations sampled along the River Thames.

We next asked whether shell shape differences between the river and marina habitats

could be associated with individual AFLP loci, as might be the case if one or a small

number of loci were influenced by selection. To do this, we compared heterozygosity and

FST values estimated for each locus against a theoretical null distribution (Fig. 3.5). For the

comparison of pooled marina vs. pooled river mussels, in only one of the three simulations

(i.e. when using 4Nµ = 1.0) one locus was significantly above the 99% CI of the

theoretical null distribution (grey circle in Fig. 3.5A). The only other evidence for possible

selection between the two habitats was found at the Marlow populations, where two

different loci consistently showed significantly elevated FST values (Fig. 3.5C). However,

none of these three loci were found to be significantly different between the two habitats at

the remaining two locations, Abingdon and Old Windsor (Fig. 3.5B,C). This lack of

consistency provides further support for phenotypic plasticity.


49

Figure 3.5. A-D. Plots of FST vs. heterozygosity for 103 AFLP markers and four respective pairs of U. pictorum populations from the River Thames: (A) pooled marina vs. pooled river specimens (baseline FST = 0.003), (B) Abingdon marina vs. Abingdon river specimens (baseline FST = 0.0005), (C) Marlow marina vs. Marlow river specimens (baseline FST = 0.001), and (D) Old Windsor marina vs. Old Windsor river specimens (baseline FST = 0.015). Solid lines indicate the mean and 99% CI, and dotted lines represent 95% CI for selectively neutral loci as determined by simulation. Black circles correspond to loci falling outside of the 99% CI in all simulation models performed per population pair using different 4Nµ values, grey circles indicate loci outside the 99% CI in one of the simulation models. Loci with significantly elevated FST are labelled with their identifying code.

3.5. DISCUSSION

The question of if and how much of the intraspecific morphological variation in

freshwater mussels (Unionoida) is attributable to genotype as opposed to phenotypic

plasticity has been the subject of considerable debate. Here we provide evidence for

phenotypic plasticity of two shell shape characters, the degree of shell elongation and

shape of the dorso-posterior margin, across six River Thames populations of U. pictorum.

Furthermore, a genetic isolation-by-distance pattern over moderate geographic distances


50

(i.e. 30-100 km), indicates that the AFLP markers offered sufficient genetic resolution to

detect subtle differences in genetic composition within the populations under study.

3.5.1. GENETIC POPULATION STRUCTURE

Overall our data suggest a trend of increasing genetic distance with increasing

geographic distance over a scale of about 100 km, which is presumably driven by

appreciable numbers of larvae (glochidia) settling out close to where they were born.

However, an isolation by distance pattern was stronger across marina – marina

comparisons of sites than across river – river comparisons. One possibility is that such a

pattern could arise entirely from chance. Alternatively however, it might reflect a key

aspect of the unionoid life cycle in which the prevalent means of dispersal is thought to be

via parasitic larvae that attach to fish (Kat, 1984). U. pictorum exploits a variety of

different host fish species (Berrie & Bioze, 1985; Aldridge, 1997; Blažek & Gelnar, 2006).

In the River Thames, Berrie & Bioze (1985) observed ten fish species to be infested by

glochidia of the two British Unio species, but the three-spined stickleback (Gasterosteus

aculeatus Linnaeus, 1758) and perch (Perca fluviatilis Linnaeus, 1758) together carried

about 90% of all Unio glochidia. While both fish species are widespread and common

within the River Thames (UK Environment Agency, pers. comm.), the relatively small-

sized three-spined stickleback is more commonly associated with lentic localities such as

marinas (Maitland & Campbell, 1992) from which populations will show predominantly

localised dispersal (Bolnick et al., 2009). Localised dispersal is also implied from a study

of lentic (lake) perch (Bergek & Björklund, 2007) where cryptic barriers to dispersal

restricted gene flow. Conversely, host fishes, especially perch, within the river can be

expected to be both actively and passively more mobile (Tudorache et al., 2008). If the

dominant host fishes within the marinas show relatively low dispersal compared with the

river, this would explain the weaker isolation by distance observed for riverine mussel

samples, with their vagile hosts dispersing excysted juvenile mussels some distance from

their parents. Thus, the striking difference in IBD trends between marina and river

populations demonstrates that the evolutionary consequences of a parasite-host interaction

(genetic divergence of parasite populations) could be dependent on the habitat where the

interaction takes place.

A pattern of genetic isolation-by-distance has been observed in other species of

freshwater mussel, for example, the North American species Quadrula quadrula


51

Rafinesque, 1820 (Berg et al., 1998), but is by no means universal. Thus, other studies

have failed to find evidence of isolation by distance across similar or even greater

geographic distances in, for example, Lexingtonia dolabelloides (Lea, 1840) (Grobler et

al., 2006), M. margaritifera (Machordom et al., 2003; Bouza et al., 2007) and Amblema

plicata (Say, 1817) (Elderkin et al., 2007). Reasons for such differences between species

and/or water bodies are likely to be complicated and, besides differences in the vagility of

their fish hosts, probably include confounding patterns of pre- and postglacial colonisation

that may “overshadow” recent dispersal events and anthropogenic influences relating to

fish stocking measures.

3.5.2. PHENOTYPIC PLASTICITY OF SHELL FORM

By comparing geographically distant, replicate population pairs of two respective

habitats, the present study has shown that individuals sampled from the same habitat were

morphologically more similar but genetically more distinct from one another compared

with those from different but geographically adjacent habitats. This suggests that the

degree of relative elongation, i.e. shell height to length ratio, and the shape of the dorso-

posterior margin, two shell characters that are strongly and consistently different between

marina and river sites, are unlikely to be under genetic control in U. pictorum. Further

evidence for phenotypic plasticity of shell shape in U. pictorum is provided by the fact that

not a single locus could be identified that showed convincing evidence of natural selection

among the different habitats we sampled, and thus, could potentially be associated with

shell shape differences of ecomorphotypes.

Our findings support observations by Hinch, Bailey & Green (1986), who, on the

basis of reciprocal transplant experiments, found that relative shell height to length growth

in the North American unionoid Elliptio complanata (Lightfoot, 1786) was determined by

phenotypic plasticity. Besides providing molecular evidence for the same pattern in the

European species U. pictorum, we also show that the shape of the dorso-posterior margin,

which is a more consistent ecophenotypic shell character in unionoids than relative shell

elongation (Chapter 2; Zieritz & Aldridge, 2009), is also probably not under genetic

control. This knowledge fills a gap in the scarce molecular evidence for phenotypic

plasticity of shell morphology in unionoids, which has so far consisted mainly of studies

reporting a lack of match of intraspecific morphological and genetic patterns (Buhay et al.,

2002; Machordom et al., 2003; Molina, 2004; Geist & Kuehn, 2005). However, since most


52

of these previous studies were not primarily focused on determining the basis of

phenotypic plasticity, they failed to include a joint analysis of morphology, genetics and

geographic distance of populations, and therefore generally fail to make a convincing case.

Evidence for phenotypic plasticity of shell morphology has been found in a number

of other mollusc taxa that occupy heterogenous habitats and have high dispersal potential,

most notably marine gastropods with planktonic larvae (e.g. Littorina striata (King &

Broderip, 1832), de Wolf, Backeljau & Verhagen, 1998; Nodilittorina australis (Gray,

1826), Yeap, Black & Johnson, 2001; Nacella concinna (Strebel, 1908), Hoffman et al.,

2010). Conversely, local adaptation in other gastropod species might be capable of driving

genetic divergence even within continuous populations (e.g. Tregenza & Butlin, 1999;

Doebeli & Dieckmann, 2003). Potential examples for such ecological nonallopatric

speciation include Littorina saxatilis (Olivi, 1792) (Johannesson & Johannesson, 1996;

Panova, Hollander & Johannesson, 2006; Conde-Padín et al., 2007) and Nucella lapillus

(Linnaeus, 1758) (Guerra-Varela et al., 2009). However, extensive gene flow is predicted

to prevent local adaptation (Johannesson, 2003) so that these species are usually direct

developers with restricted larval dispersal. Moreover, despite some evidence for divergent

selection acting on marine broadcast spawners with long-lived pelagic larvae (Luttikhuizen

et al., 2003; de Aranzamendi et al., 2008), the dispersal of unionoid larvae by their fish

hosts apparently leads to a relatively even redistribution of larval genotypes across

different habitats. Consequently, morphological adjustments are possible only through a

plastic phenotype.

The unionoid shell ecophenotypes observed might have adaptive significance which,

along with their possible functional morphologies, are discussed in detail in Chapter 2,

Zieritz & Aldridge (2009) and references therein. In short, hydrological parameters such as

water movement (e.g. mean and/or maximum current velocities) are probably the main

factors determining the sagittal shape of these unionoids’ shells. Dorso-posterior arching

of river morphotypes, for example, results in heavier weight and increased pedal gape of

the shell, allowing the foot to extend further into the sediment. These features would

increase initial probing force, anchorage, and the stability of the bivalve when subject to

lifting forces resulting from turbulent water (Eagar, 1978). Alternatively, differences in

shell shapes could have no adaptive value whatsoever but be caused by a non-functional

reaction of the mussel to the environment (e.g. via distortion of the shell secreting mantle

margin as a result of water movement). If the observed phenotypic plasticity of shell form


53

is adaptive (i.e. results in an improvement in growth, survival or reproduction; Stearns,

1989), nonadaptive or even maladaptive remains to be tested.

Finally, our study potentially has implications for mussel classification. Previous

work has debated the extent to which morphotypes reflect phenotypic plasticity versus

genuine differentiation leading to speciation (e.g. Mulvey et al., 1997; Serb, Buhay &

Lydeard, 2003). This problem is exacerbated by the problem that contrasting morphologies

were often sampled from geographically distant sites, increasing the chance that

morphological and genetic isolation will be correlated (Davis et al., 1981; Davis, 1983,

1984; Baker et al., 2003). A further issue is that, over and above any phenotypic plasticity,

some molluscan shell characteristics are known to be under genetic control (e.g. smoothed

versus ribbed shell sculpture in the gastropod Oncomelia hupensis Gredler, 1881; Davis &

Ruff, 1974). Although limited in range, our study illustrates the value of looking at

genetics and morphology over a range of geographic scales in order to interpret which

factors are most important in driving divergence and provides what we believe is the first

good evidence for phenotypic plasticity of shell shape in U. pictorum. Further similar

studies on other unionoid species using similar markers will be needed in order to establish

the generality of what we have found.

CHAPTER 4

SEXUAL, HABITAT-CONSTRAINED AND PARASITE-INDUCED DIMORPHISM IN THE

SHELL OF A FRESHWATER MUSSEL (ANODONTA ANATINA, UNIONIDAE)

“I’m not obese, it’s all mussel!”

Slogan by Holly Barclay

57

CHAPTER 4

SEXUAL, HABITAT-CONSTRAINED AND PARASITE-INDUCED DIMORPHISM IN THE SHELL OF A FRESHWATER MUSSEL

(ANODONTA ANATINA, UNIONIDAE)

4.1. ABSTRACT

We investigated to what extent sex, trematode infection and habitat effects determine

shell dimorphisms in five populations of the freshwater mussel Anodonta anatina.

Significant sexual shell width dimorphism was displayed by three of the five study

populations. Here, shells of females were significantly wider than males, probably as a

result of altered shell growth to accommodate marsupial gills. In two of these populations

female shells were additionally significantly thinner than those of males, which could be a

result of resource depletion by offspring production. No significant dimorphic patterns

were found in two other populations of the same species. The different degree of sexual

dimorphism between populations may reflect the overarching effect of habitat on

morphology. Thus, populations in the most favourable habitat might exhibit faster growth

rates, attain larger maximum sizes and produce more offspring, which results in more

swollen gills and consequently more inflated shells of gravid females compared to less

fecund populations. None of the populations showed any evidence for sexual dimorphism

in any of the shell characters ‘overall size’, ‘growth rate’, ‘sagittal shape’ and ‘density’. In

addition to sexual dimorphisms, infestation by trematode parasites significantly altered

sagittal and lateral shell shape of A. anatina in one of the populations, with infected

specimens being wider and more elongated.

4.2. INTRODUCTION

The interpretation of diverse patterns in the morphology of animal species has long

been a matter of scientific interest. Understanding which factors determine morphological

characteristics of individuals within a species can provide useful information concerning

the biology and ecology of species, populations or even individuals. Taxa producing hard

parts that persist in sedimentary deposits can additionally be used to reconstruct

characteristics of ancient species, populations and palaeoenvironments. Shell morphology


58

of freshwater mussels (Unionoida) could be of particular value in these respects.

Unionoids are common and widespread, inhabiting all continents except the Antarctic

(Graf & Cummings, 2006b). Their oldest known representatives stem from the Triassic

(Watters, 2001), and their typically high degree of intraspecific variation in shell

morphology potentially contains a range of information about an individual’s biological

and ecological characteristics. However, due to a lack of understanding of which factors

cause which trends in shell form, such intraspecific morphological differences in unionoids

are rarely used for redrawing characteristics of both the present and the past.

Probably the best studied, though still insufficiently understood, type of intraspecific

variation in unionoid shell morphology is that associated with habitat conditions (i.e.

ecophenotypic) (see Chapter 2; Chapter 3; Zieritz & Aldridge, 2009; Zieritz et al., 2010;

and references therein). Here we investigate two other, largely ignored sources of

intraspecific variation in freshwater mussels; sexual dimorphism and parasite induced

change in shell morphology.

Numerous striking examples of sexual dimorphism are known from the animal

kingdom and in particular mammals (e.g. Thom, Harrington & Macdonald, 2004), birds

(e.g. Owens & Hartley, 1998), fish (e.g. Kitano, Mori & Peichel, 2007), reptiles (e.g.

Brecko et al., 2008) and insects (e.g. Stillwell & Fox, 2007). All these are examples for

well developed sexual differences in external structures and are caused by (1) preference

of one sex for particular traits of the other sex, (2) intra-sexual competition (sexual

selection) or (3) niche divergence between the sexes (Shine, 1989; Parker, 1992;

Andersson, 1994). Though none of these three forms of sexual dimorphism would be

expected in broadcast spawners such as freshwater mussels, sexual dimorphism in this

group may arise as a direct result of differences in their sexes’ internal anatomy.

Protandric or protogynous consecutive hermaphroditism results in size dimorphism

of sexes in several marine bivalve families (e.g. Veneridae (Loosanoff, 1936), Astartidae

(Saleuddin, 1965), Galeommatidae (Morton, 1976)). While hermaphroditism is rare in

unionoids (e.g. Bloomer, 1934; McIvor & Aldridge, 2007), protandry and consequential

sexually dimorphic size has been documented in the unionid Elliptio complanata

(Lightfoot, 1786) (Downing et al., 1989).

Sexual dimorphism may also arise from the unique reproductive mode of unionoids.

Their larvae mature in certain parts of the female gills (marsupia), and in some of these

cases the resulting swelling of gravid marsupia is believed to alter growth in female shells.

A particularly pronounced sexual shape dimorphism is exhibited by most members of the

Chapter 4 – Shell dimorphism in Anodonta anatina

59

tribe Lampsilini (subfamily Ambleminae; phylogeny after Graf & Cummings, 2006b) (e.g.

Kirtland, 1834; Heard & Guckert, 1970; Kotrla & James, 1987). Lampsilines are unique in

using only the posterior portions of their outer gill demibranchs as marsupia (heterogenous

condition), which enlarge both laterally and ventrally when gravid, resulting in female

shells with more evenly rounded posterior margins and more inflated ventro-posterior

regions than males. More subtle patterns of shape dimorphism have been found in species

with gills marsupial across their entire lengths (homogenous condition) (e.g. Kirtland,

1834; Heard & Guckert, 1970).

European freshwater mussels of the unionid subfamily Unioninae have homogenous

marsupia and are generally believed not to display any considerable sexual shell

dimorphism. The occasional observations on sex-related differences in shell morphology,

and in particular shell inflation (width), are mostly anecdotal (e.g. Noll, 1869; Hazay,

1881; Sterki, 1903). Brander (1954) described male shells of Pseudanodonta complanata

(Rossmässler, 1835) as more elongated and ovally shaped than the more trapezoid females.

Contrary to this, Siebold (1837) found females of Anodonta anatina (Linnaeus, 1758) and

Anodonta cygnea (Linnaeus, 1758) to be more elongated than males. Other unionid shell

characters that have been considered to be sexually dimorphic include shell colour (Israel,

1910), thickness (Jass & Glenn, 2004) and composition of shell material (Cameron,

Cameron & Paterson, 1979).

Infection by parasites is another factor that may cause alteration in shell size and

shape. Freshwater mussels are known to host a number of parasites including copepods

(e.g. Saarinen & Taskinen, 2004), mites (e.g. Mitchell & Pitchford, 1953), chironomids

(e.g. Forsyth & McCallum, 1978), fish (e.g. Aldridge, 1999a) and trematodes (e.g. Jokela,

Uotila & Taskinen, 1993). The effects of parasitism on unionoid morphology are

understudied but shell growth in A. anatina has been shown to be slowed during heavy

infection by the bucephalid trematode Rhipidocotyle fennica Gibson, Valtonen et

Taskinen, 1992 (Taskinen & Valtonen, 1995). Additionally, trematodes can be associated

with reduced reproductive output (Gangloff, Lenertz & Feminella, 2008), which may

affect the development of dimorphic characters.

This study represents the first attempt to investigate if and to what extent sex,

trematode infection and indirect habitat effects induce shell dimorphisms in five

populations of the European unionoid A. anatina. A modern morphometric method

(Fourier shape analysis) was employed to describe shell shape. Morphological characters

under study were (1) size and growth rates, (2) relative elongation, (3) overall sagittal shell


60

shape, (4) relative width, (5) thickness and (6) density of the shell. Possible reasons for the

dimorphic patterns observed and why certain sexual differences are present in some but

not in other populations are discussed.


4.3.1. SAMPLING, SEX DETERMINATION AND TREMATODE INFECTION

From May 2007 to September 2009, five A. anatina populations were investigated

for dimorphic patterns in shell morphology (Fig. 4.1). To test if habitat conditions

contributed to dimorphic trends, we selected five sites that differed in both environmental

conditions and characteristics of their A. anatina populations (such as population densities

and growth rates). The River Thames in the stretch studied is a typical UK lowland lotic

system whereas the marina sites are lentic habitats with almost no flow. Quidenham Mere

is a eutrophic lake system.

Figure 4.1. Locations of the five UK Anodonta anatina populations of the study. Abbreviations: Q, Quidenham Mere; AM, Abingdon Boat Marina (River Thames); AR, River Thames, Abingdon/Culham; CM, Thames & Kennet Marina (River Thames); CR, River Thames, Caversham. Map © Crown Copyright/database right 2009. An Ordnance Survey/EDINA supplied service.

A. anatina abundances at each site were estimated by counting the number of

individuals present in 15 replicate dredges (approx. 5 m in length, dredge aperture 46 cm x

21 cm x 7.5 cm, mesh size = 2.5 cm). In addition, 15-40 individuals per population,

depending on A. anatina abundance at the site, were sampled by hand and taken to the

laboratory for further morphological investigation.


61

In the laboratory, the sex of each individual was determined by macroscopic

inspection of the demibranchs for swelling of the outer gills, indicating

female/hermaphroditic sex, and examination of gonadal fluid under a light microscope.

Mussels presenting oogonia, oocytes or larvae (glochidia) of any developmental stage

were assigned as females; occurrence of spermatozoa and/or sperm-morulae indicated

male sex (Heard, 1975). Individuals exhibiting both male and female gonad characteristics

were classified as hermaphrodites. State of trematode parasitism (‘parasitised’ or ‘not

parasitised’) was also determined by examining gonadal fluid under a light microscope. An

individual mussel was considered parasitised if digenean trematode sporocysts and/or

cercariae were present in one or more of three gonadal samples.

4.3.2. SIZE AND GROWTH RATES

Age of 15-35 specimens per site was estimated by counting annual winter rings

(Aldridge, 1999b). Maximum length at each annulus was measured with a digital calliper

to ± 0.1 mm, and length-age plots were produced for each population by calculating the

mean shell length at each year. Growth parameters were determined using the Walford plot

model (Walford, 1946) assuming von Bertalanffy growth curves.

Differences between populations in size and growth rates were statistically tested

using the measurements of shell length (L; across the maximum sagittal diameter) and

shell length at the third annulus (L3) for each aged mussel. The same two measurements

were used to test for sexual size dimorphism (consecutive hermaphroditism) and

differences in growth rates between sexes within each of the five populations.

4.3.3. SHELL SHAPE, THICKNESS AND DENSITY

For analysis of shell shape the three shell dimensions length (L), height (H;

maximum sagittal diameter perpendicular to L) and width (W; maximum lateral diameter)

were measured to ± 0.1 mm using a digital calliper. Using these measurements, sagittal

area (SaA) and surface area (SuA) of each mussel were estimated by applying the formulas

for an ellipse and ellipsoid, respectively. Shell dry weight (DW) and shell volume (V;

measured as water displaced by the shell using method of Rodhouse (1977)) were recorded

to ± 0.01 cm³. These parameters were subsequently used in general linear models (GLMs;

Minitab 15 for Windows) to test for statistical significance of sexual dimorphism in degree

of shell elongation, inflation, thickness and density. For this purpose, GLMs were carried


62

out on H, W, V and DW, fitting L, SaA1/2, SuA and V as covariates, respectively, population

as a factor with five levels, sex as a factor with two levels, and the covariate*sex,

covariate*population and sex*population interaction factors. Insignificant factors were

then sequentially dropped from the model (Crawley, 2002). For those parameters that

proved to be significantly influenced by sex, the sex*covariate and/or the sex*population

interaction factor, GLMs were then carried out for each population separately in order to

elucidate which population(s) exhibited significant sexual dimorphism in the respective

shell character.

For reasons explained in the results section, trematode parasite-induced alteration in

shell morphology was tested only in the Quidenham Mere population regarding all four

characters as described above. For this purpose, GLMs were carried out including the

respective covariate, both trematode (parasitism) and sex as factors with two levels, and

the covariate*sex-, covariate*trematodes- and sex*trematodes-interaction factors.

To examine if sexually dimorphic and/or parasite-induced patterns were present with

respect to overall sagittal shape of the shell, we carried out Fourier shape analyses as

developed and explained by Crampton & Haines (1996) on two groups of populations: (1)

all five populations combined and (2) the Quidenham Mere population alone. This method

decomposes xy-coordinates of an outline into a number of harmonics, each of which is in

turn explained by two respective Fourier coefficients, which can be statistically treated as

any usual variable. For this analysis digital photographs of specimens were taken, and

outlines were digitised using the program IMAGEJ (Rasband, 2008). The xy-coordinates

of digitised outlines were then subjected to fast Fourier transformation using program

HANGLE (Crampton & Haines, 2007) applying a smoothing normalisation of 20 to

eliminate high-frequency pixel noise. Preliminary analysis indicated that the first ten

harmonics described the outlines with sufficiently high precision. Discarding of the first

harmonic, which does not contain any shape information, resulted in a set of 18 Fourier

coefficients per individual. After rotating outlines to maximum overlap by program

HMATCH (Crampton & Haines, 2007), principal component analysis (PCA) was

performed on the 18 Fourier coefficients using program PAST (Hammer & Harper, 2006).

Number of significant principal components was determined by the broken stick model of

the scree plot. Subsequently, GLMs were carried out on significant PC axes obtained by

PCA on (1) all five populations including factors population and sex and the

sex*population interaction factor, and (2) Quidenham Mere population including factors

sex and trematodes, and their interaction.


63

4.4. RESULTS

4.4.1. COMPARISON OF THE FIVE STUDY POPULATIONS

Abundance of A. anatina estimated by 15 replicate dredges (median count of

individuals per dredge) was significantly different at the five sites (Kruskal-Wallis: H =

20.33, d.f. = 4, P < 0.001). In detail, this species was significantly more abundant at

Quidenham Mere (Q) than at Caversham Marina (CM) and Caversham River (CR), with

the latter site additionally being significantly less densely populated than Abingdon Marina

(AM) (Bonferroni-corrected Mann-Whitney pairwise comparisons significant at <5%).

Table 4.1. Abundance (given as count of individuals per dredge), and size, age and growth parameters for female (F) and male (M) Anodonta anatina from the five study populations (Pop.). Lmax is the observed maximum length in the field. Growth constant (k) calculated from the Walford plot. Abbreviations: X , mean; SD, standard deviation.

Pop. Abundance Sex n age [yrs] L [mm] L3 [mm] Lmax[mm] k X ± SD X ± SD X ± SD X ± SD Q 2.67 ± 2.29 all 38 5.3 ± 2.2 85.3 ± 16.4 63.6 ± 7.4 114.6 0.26 F 17 5.0 ± 1.9 96.4 ± 16.4 63.4 ± 8.2 114.6 0.18 M 16 5.8 ± 2.5 84.1 ± 16.5 63.8 ± 6.8 104.4 0.27 AM 1.87 ± 1.88 all 40 5.7 ± 1.4 80.4 ± 10.2 62.1 ± 5.9 99.1 0.23 F 20 5.8 ± 1.5 82.3 ± 11.2 61.1 ± 5.8 99.1 0.23 M 20 5.6 ± 1.3 78.65 ± 9.1 63.0 ± 6.0 94.3 0.24 AR 0.87 ± 1.88 all 34 5.2 ± 0.8 63.7 ± 8.5 53.7 ± 7.1 80.0 0.32 F 15 5.4 ± 0.9 63.7 ± 9.7 52.4 ± 6.9 80.0 0.31 M 17 4.9 ± 0.7 63.75 ± 8.1 55.0 ± 7.3 76.6 0.28 CM 0.53 ± 0.52 all 19 4.3 ± 0.7 60.4 ± 7.8 54.1 ± 5.5 81.3 0.16 F 11 4.3 ± 0.7 61.0 ± 10.0 55.5 ± 64.0 81.3 0.18 M 7 4.3 ± 0.8 59.3 ± 4.0 52.3 ± 35.9 64.5 0.35 CR 0.07 ± 0.26 all 15 2.4 ± 0.5 49.7 ± 6.4 57.9 ± 1.7 59.8 - F 5 2.4 ± 0.5 51.6 ± 4.9 56.7 ± 1.2 57.6 - M 8 2.4 ± 0.5 50.2 ± 7.3 58.7 ± 1.6 59.8 -

The five populations were also significantly different from each other in shell length

(Table 4.1; L; ANOVA: F4, 141 = 43.72, P < 0.0001) with Q and AM A. anatina being

significantly larger than at all other sites, and Abingdon River (AR) and CM being

significantly larger than CR (Bonferroni-corrected Tukey pairwise comparisons significant

at <5%). A similar pattern was observed at the third annulus (L3; ANOVA: F4, 112 = 13.17,

P < 0.0001), with Q and AM being significantly larger than CM and AR. Length-age plots

(Fig. 4.2) suggest that from approximately the third year of age, mussels were growing

fastest and to largest maximum sizes at Q, followed by AM, CM and AR. No individuals


64

older than three years old were found at CR and thus growth parameters could not be

determined (Table 4.1).

Figure 4.2. Average shell length at age plots for the five A. anatina study populations. Males and females are combined for each site.

4.4.2. SEX RATIOS

Out of the 146 specimens sampled, eight individuals could not be sexed reliably and

two specimens, one from each AR and Q, respectively, were identified as hermaphroditic.

These ten individuals were excluded from further analysis of sexual dimorphism. Sex

ratios were not significantly different between populations (Table 4.1; χ² = 1.74, d.f. = 4, P

= 0.784) and met a 1:1-ratio (altogether 68 females : 68 males).

4.4.3. SIZE AND GROWTH RATES

No significant differences in average length (L; paired: T = 1.74, N = 5, P = 0.157)

and maximum length (Lmax; paired: T = 2.05, N = 5, P = 0.110) of female and male shells

were found, indicating that no sexual size dimorphism was present in the study populations

(Table 4.1). Likewise, indifferent shell lengths at the third annual ring (L3; paired: T = 0.68,

N = 5, P = 0.533) and growth rates (k; paired: T = 1.35, N = 4, P = 0.27) of male and

female shells indicated growth rates of sexes were similar.


65

4.4.4. SHELL SHAPE, THICKNESS AND DENSITY

Table 4.2. Results of general linear models (GLMs) of shell height, the three significant principal components obtained by Fourier shape analysis on sagittal shell shape, width, volume and dry weight. Values correspond to the minimal adequate models (Crawley, 2002).

Response variable Source F d.f. P height length 1789.5 1 <0.0001 R² = 0.96 population 17.9 4 <0.0001 sex - - n.s. population*length - - n.s. sex*length - - n.s. population*sex - - n.s. PC 1 (Fourier) population 29.3 4 <0.0001 R² = 0.46 sex - - n.s. population*sex - - n.s. PC 2 (Fourier) population 14.3 4 <0.0001 R² = 0.29 sex - - n.s. population*sex - - n.s. PC 3 (Fourier) population - - n.s. sex - - n.s. population*sex - - n.s. width sagittal area1/2 841.5 1 <0.0001 R² = 0.96 population 4.5 4 0.002 sex 10.8 1 0.001 population*sagittal area1/2 3.0 4 0.019 sex*sagittal area1/2 - - n.s. population*sex - - n.s. volume surface area 1071.3 1 <0.0001 R² = 0.98 population 4.6 4 0.002 sex 17.1 1 <0.0001 population*surface area 8.1 4 <0.0001 sex*surface area 8.4 1 0.005 population*sex - - n.s. weight volume 25018.6 1 <0.0001 R² = 1.0 Population 5.3 4 0.001 sex - - n.s. population*volume - - n.s. sex*volume - - n.s. population*sex - - n.s.

Table 4.2 summarises results of general linear models testing if factors population

and sex significantly influenced the appearance in the four shell morphological characters

(1) relative shell elongation (by fitting height against length), (2) relative shell width (by

fitting width against sagittal area1/2), (3) shell thickness (by fitting volume against surface

area), and (4) shell density (by fitting dry weight against volume). All morphological


66

characters under study, including overall sagittal shape of the shell as determined by

Fourier shape analysis, showed significant differences between populations (Table 4.2,

Fig. 4.3).

Figure 4.3. Scatterplots of PC1 and PC2 values obtained by principal component analyses on 18 Fourier coefficients of five A. anatina populations. Synthetic outlines of extreme sagittal shapes obtained using method of Haines & Crampton (2000), displayed with anterior margin facing to the right and dorsal margin to the head of the page. Values for the third significant principal component, explaining 10.4% of the variation in shell outline, are not displayed.

Factor sex was retained as a significant predictor only for relative shell width and

thickness (Table 4.2), which were therefore analysed for each population separately.

Though females of all populations tended to have relatively wider shells than males

(Fig. 4.4), GLMs revealed that this sexual difference in relative shell width was significant

only in populations AM and CR (Table 4.3). Shell thickness, on the other hand, was

significantly influenced by sex only in populations Q and AM (Table 4.3), with females

exhibiting thinner shells than males (Fig. 4.5).

67

Figure 4.4. Scatterplots of five A. anatina populations showing patterns in degree of shell inflation (width vs. sagittal area1/2): filled symbols, females; open symbols, males; circles, not infected by trematode parasites; triangles, infected by trematode parasites; solid regression lines, female; dashed regression lines, male.

68

Table 4.3. Results of general linear models (GLMs) of shell width and volume in five A. anatina populations. Values correspond to the minimal

adequate models (Crawley, 2002).

Q AM AR CM CR Response variable Source F d.f. P F d.f. P F d.f. P F d.f. P F d.f. P

width sagittal area1/2 498.3 1 <0.0001 149.2 1 <0.0001 144.3 1 <0.0001 238.6 1 <0.0001 216.7 1 <0.0001 sex - - n.s. 10.0 1 0.003 - - n.s. - - n.s. 13.4 1 0.004 sex*sagittal area1/2 - - n.s. - - n.s. - - n.s. - - n.s. - - n.s. volume surface area 1310.1 1 <0.0001 174.4 1 <0.0001 325.2 1 <0.0001 128.2 1 <0.0001 349.2 1 <0.0001 sex 6.9 1 0.013 7.3 1 0.010 - - n.s. - - n.s. - - n.s. sex*surface area - - n.s. 17.0 1 <0.0001 - - n.s. - - n.s. - - n.s.


69

Figure 4.5. Scatterplots of five A. anatina populations showing patterns in shell thickness (volume vs. surface area): filled circles and solid regression lines, females; open circles and dashed regression lines, males.


70

4.4.5. TREMATODE PARASITISM

Table 4.4. Results of general linear models (GLMs) shell height, the three significant principal components obtained by Fourier shape analysis on sagittal shell shape, width, volume and dry weight in the A. anatina population of Quidenham Mere, UK. Values correspond to the minimal adequate models (Crawley, 2002). Response variable Source F d.f. P height length 332.5 1 <0.0001 R² = 0.98 sex - - n.s. trematodes 8.5 1 0.007 sex*length - - n.s. trematodes*length 7.7 1 0.010 sex*trematodes - - n.s. PC 1 (Fourier) sex - - n.s. R² = 0.21 trematodes 8.65 1 0.006 sex*trematodes - - n.s. PC 2 (Fourier) sex - - n.s. trematodes - - n.s. sex*trematodes - - n.s. PC 3 (Fourier) sex - - n.s. trematodes - - n.s. sex*trematodes - - n.s. width sagittal area1/2 206.0 1 <0.0001 R² = 0.95 sex 6.0 1 0.021 trematodes 5.4 1 0.028 sex* sagittal area1/2 - - n.s. trematodes* sagittal area1/2 6.4 1 0.017 sex*trematodes - - n.s.

Trematodes were found in individuals of three of the five A. anatina populations; Q

(27% infected), AM (5%) and CM (31%). However, only at population Q were numbers of

infected and uninfected mussels sufficient to test statistically for significance of trematode-

induced alteration of shell morphology. GLMs revealed a significant difference in sagittal

and lateral shell shape of uninfected and infected mussels (Table 4.4). Thus, trematodes

was retained as a factor significantly explaining both relative elongation and overall

sagittal shell shape (as analysed by Fourier shape analysis), indicating that specimens

infected with trematodes tended to grow more elongated shells than uninfected ones (Fig.

4.6). Additionally, inclusion of factor trematodes in the GLM on shell width revealed that

relative shell inflation at population Q was not only significantly influenced by trematode

infection but also by sexual dimorphism (Table 4.4). Infected specimens tended to be

relatively wider than uninfected ones (Fig. 4.4A). Thus, accounting for alteration of shell


71

width growth due to infection by trematodes revealed that besides populations at AM and

CR, sexual width dimorphism was also significant at population Q.

Figure 4.6. Scatterplots of PC1 and PC2 values obtained by principal component analyses on 18 Fourier coefficients of four groups of A. anatina specimens of Quidenham Mere, UK: filled symbols, females; open symbols, males; circles, not infected by trematode parasites; triangles, infected by trematode parasites. Synthetic outlines of extreme sagittal shapes obtained using method of Haines & Crampton (2000), displayed with anterior margin facing to the right and dorsal margin to the head of the page. Values for the third significant principal component, explaining 14.9% of the variation in shell outline, are not displayed.

4.5. DISCUSSION

This study revealed the presence of sexual dimorphism with respect to shell inflation

and thickness of three and two A. anatina populations, respectively. No significant

dimorphic patterns were found in two other populations of the same species. Furthermore

and for the first time, we found evidence for trematode parasite-induced alteration of

sagittal and lateral shell shape of unionoids. The fact that some, but not all, populations

studied showed patterns of dimorphism can be explained by the constraint of habitat on the

expression of shell morphology.

4.5.1. SIZE AND HERMAPHRODITISM

We were not able to detect any significant sex-related size or age differences in the

five populations studied and thus, our data does not indicate the presence of sequential


72

hermaphroditism in these populations of A. anatina. This is further supported by the fact

that only two out of 138 sexed specimens were identified as hermaphroditic.

4.5.2. SEXUAL DIMORPHISM IN SAGITTAL SHELL SHAPE

None of the A. anatina populations studied showed any sex-related difference with

regard to sagittal shape of shells. This is not surprising given that when gravid, marsupia in

this and other Unioninae primarily expand laterally (i.e. along the shell width axis) and not

ventrally as in lampsilines, known for their well developed sexual dimorphism in sagittal

shell shape (e.g. Jass & Glenn, 2004). Although some authors of the 19th century claimed

the presence of sex-related differences regarding shell elongation of European unionids

(e.g. Kirtland, 1834; Siebold, 1837; Buchner, 1900; Brander, 1954), most of these

observations are merely anecdotal. Fallacious unionid taxonomy and doubtful sex

identification at that time further add to the fact that such documentations should be treated

with caution. More recently, however, more subtle differences in sagittal shell shape were

observed in several Indotropical unionids. Agrawal (1974) and Hamai (2004) found the

anterior part of female Parreysia wynegungaensis (Lea, 1859) and Inversidens japonensis

(Lea, 1859) respectively, to be more developed than in males. Unfortunately, these authors

did not discuss possible explanations for the patterns they observed.

4.5.3. SEXUAL DIMORPHISM IN RELATIVE SHELL WIDTH

There has been considerable anecdotal documentation of differences in shell width

of male and females in several unionoid species (e.g. Kirtland, 1834; Hazay, 1881; Sterki,

1903; Hayashi, 1935) but this study provides the first statistically significant evidence for

shell width dimorphism in a European unionoid. As expected, in all study populations,

females were more inflated than males. It is likely that this is a result of altered width

growth of females during the period of gravidity. For example, gravid marsupial gills in

the long-term brooding Pyganodon cataracta (Say, 1817) swell to nearly thirty times their

non-brooding width (Tankersley & Dimock, 1992). However, sexual width dimorphism

was significant in only three of the five populations studied and similar inconsistencies in

dimorphic patterns across populations of the same species have been reported in the past

(Weisensee, 1916; Brander, 1954). Possible reasons for such interpopulational

discrepancies have, however, not been discussed before, leaving one to question why

dimorphic patterns occur in some populations but not in others.


73

Our study provides some evidence for an indirect effect of environmental condition

on sexual shell width dimorphism in a unionoid. Thus, two of the three populations

exhibiting significant dimorphic patterns in this character (i.e. Q and AM) were those with

the highest population abundances, fastest growth rates and largest maximum sizes. All

these parameters indicate that habitat conditions at these two sites were more favourable

for A. anatina than at the remaining sites. Under such “good” habitat conditions, a

female/hermaphroditic mussel could be expected to be able to be more fecund, resulting in

more swollen gills and consequently more inflated shells.

Favourable conditions for growth are also indicated by the large shell length at the

third annulus in the third population displaying significant shell width dimorphism (CR).

The low density of mussels at this site, coupled with unusually young maximum age in this

population (< 4 years) could be the result of a catastrophic event rather than reflecting

unfavourable conditions. In this population, some of the gravid specimens were only two

years old, which is especially young for maturity in anodontines (Heard, 1975) but may

help explain the sexual dimorphism recorded. Thus, interpopulational variation of life

history traits such as size/age of maturity, incubation length and/or timing of glochidial

release (e.g. Bauer, 2001; Hochwald, 2001; Haag & Stanton, 2003) might also influence

female shell bulging and should be considered when depicting sexual dimorphic patterns

in unionoid populations.

4.5.4. SEXUAL DIMORPHISM IN SHELL THICKNESS AND DENSITY

Shell thickness and density have rarely been considered as potentially sex-

influenced. Nevertheless, in two of the five A. anatina populations a significant sex-related

difference in shell thickness was found, with female shells tending towards thinner shells

than males.

It seems likely that this type of sexual dimorphism could be triggered by costs of

producing and nourishing offspring, depleting resources that otherwise would be used for

shell production. Brooding female unionoids are known to contribute both nutrients

(Wood, 1974) and calcium (Silverman, Steffens & Dietz, 1985; Silverman, Kays & Dietz,

1987) to their eggs and glochidia, formed in water channels of their marsupia. Reduced

nutrient and/or calcium availability during the gravid period of females have been argued

to slow down or even cause negative growth (Downing & Downing, 1993; McIvor &

Aldridge, 2007). Though we did not detect any differences in perimeter growth rates of


74

male and female mussels, thickness growth could be affected in a similar way. What is

more, investing energy in growing larger on the expense of growing thick could be

advantageous given that fecundity has repeatedly been shown to increase with size in

unionoids (e.g. Bauer, 1998; Hochwald, 2001; Haag & Stanton, 2003).

4.5.6. TREMATODE PARASITE-INDUCED DIMORPHISM

Based on one A. anatina population of Quidenham Mere, our study suggests that

unionoid shell shape can be significantly influenced by trematode parasitism. This

represents the first evidence of parasite-induced shape alteration in a European unionoid

though previous authors have shown that overall growth rates can be slowed down in

mussels due to parasitism by trematodes (Taskinen & Valtonen, 1995; Taskinen, 1998).

Other studies have found that trematode parasitism can additionally reduce reproductive

output and physiological condition of unionoids (Jokela, Uotila & Taskinen, 1993;

Gangloff, Lenertz & Feminella, 2008). While reduced fecundity might be expected to

result in narrower shells, our data suggest that infected mussels of both sexes tended to be

relatively more inflated than uninfected ones. One possible explanation for this pattern

could be that reproduction of trematodes within the mussel’s gonad might cause bulging of

the mussel’s soft tissue, ultimately resulting in more inflated shell growth. Further work is

needed to understand the effects of trematodes on shell swelling and indeed on our

observation that infected A. anatina were especially elongated.

4.5.6. HABITAT-CONSTRAINED DIMORPHISM

While this study has found pronounced patterns in shell dimorphism in A. anatina,

which can be attributed to sex and parasite infection, no pattern was universal across all

five populations in terms of its presence and/or degree of expression. This variation in

expression of morphology can be explained by habitat constraints (e.g. Chapter 2; Zieritz

& Aldridge, 2009) which may in some circumstances have a dominating effect on

morphology. Similar patterns of environmentally-constrained differences between

populations in the degree of, for example, sexual dimorphisms have been observed in other

animal taxa such as gastropods (Goodwin & Fish, 1977) and beetles (Stillwell & Fox,

2007).


75

4.5.7. APPLICATION OF THE PATTERNS OBSERVED

Freshwater mussels are some of the most imperilled fauna in the world’s freshwaters

(Bogan, 1993; Lydeard et al., 2004). Understanding the trade-offs between reproductive

potential and habitat-imposed constraints may assist in identifying optimal habitats for

particular species and selecting appropriate management activities.

In some populations, shell shape may provide a tool for the rapid identification of

sex in the field, which may be especially useful in propagation programmes, where only

gravid females are typically required. Interspecific differences in male and female shell

morphologies may reveal species-specific patterns in brooding, fecundity and size at

maturity. An understanding of the biotic and abiotic factors that influence shell shape may

be particular valuable in inferring habitat requirements for very rare, extinct or fossil

species, where little or no data can be collected from living populations. Our data also

suggest that a change in morphology of a population over time could reflect a change in

parasite load, which could prove a useful tool for managers of closely-monitored

populations, such as those used in pearl cultivation or those of high conservation value.

CHAPTER 5

VARIABILITY AND A NEW MODEL FOR CHARACTER EVOLUTION OF UMBONAL

SCULPTURES IN THE UNIONOIDA

“The fossils are giving me indigestion over where the generic lines

are drawn and how long the genera persist in the fossil record.”

Arthur Bogan (pers. comm.)

79

CHAPTER 5

VARIABILITY AND A NEW MODEL FOR CHARACTER EVOLUTION OF UMBONAL SCULPTURES IN THE UNIONOIDA

5.1. ABSTRACT

‘Umbonal sculptures’ (also known as ‘beak sculptures’ or ‘rugae’) of freshwater

mussels (Unionoida) are restricted to the early shell region and have long been used for

species identification. Specificity of beak sculptures to higher taxonomic levels, and their

value for phylogenetic reconstructions and classification of fossil taxa is still under

considerable scientific debate. This is due to our lack of understanding of character

evolution and consequently, knowledge of synapomorphies, plesiomorphies and

convergences in beak sculpture character states. Based on examination of over 150

species, covering five of the six unionoid families, this study presents a new model of

character evolution of umbonal sculptures in this bivalve group. Presence of ‘V-shaped’

sculpture in Triassic unionoids and fossil members of their closest extant relatives, the

Trigonioida, indicates that this type represents the original character state of unionoid beak

sculpture. From this plesiomorphic ‘V-shaped’ type, ‘pseudo-radial’ sculpture probably

arose in one line of development, whereas ‘W-shaped’ and ultimately ‘double-looped’

sculpture was developed in other unionoid groups. All remaining beak sculpture types

observed in the study material (i.e. ‘pseudo-concentric’, ‘single-looped’, ‘nodulous’,

‘wrinkled’ and ‘smooth’) represent derivates of the ‘V-shaped’, ‘W-shaped’ and/or

‘double-looped’ type. Implications of this new model of character evolution for unionoid

phylogeny and functional morphology of beak sculpture are discussed.

5.2. INTRODUCTION

Most juvenile freshwater mussels (Unionoida) exhibit surface protrusions, which at a

certain life stage (usually within the first years of age) either completely cease to be

produced or change into a different type of sculpture. Morphological characteristics of

these so called ‘umbonal sculptures’ (also ‘beak sculptures’ or ‘rugae’) have long been

recognised to be species-specific and are thus widely used as an identification tool (e.g.

Marshall, 1890; Simpson, 1900; Kennard, Salisbury & Woodward, 1925; Bloomer, 1937;


80

Dolmen & von Proschwitz, 1999; Killeen, Aldridge & Oliver, 2004; Watters, Hoggarth &

Stansbery, 2009). While some authors discussed unifying characteristics of these structures

at the genus and occasionally higher taxonomic levels (Frierson, 1909b, a; Ortmann,

1912), distribution of beak sculpture morphotypes across the unionoid phylogeny and

consequently, character evolution of beak sculpture remains poorly understood.

Knowing if a given character state of unionoid beak sculpture is plesiomorphic

(basal), apomorphic (derived) or homoplasic (convergent) may help determine

phylogenetic positions of taxa and resolve open questions on these bivalves’ systematics

and phylogeny. This is particularly important with regard to the fossil record, for which

taxonomic classification is based largely on beak sculpture and a few other shell characters

(e.g. Haas, 1969b; Sha, 1993; Guo, 1998a).

By recognising ecological similarities of convergent but unrelated taxa, a good

understanding of present homologies and homoplasies of beak sculptures can additionally

help identify probable functional morphologies of this shell character. Several possible

functions have been attributed to umbonal sculptures, including predator deterrence

(Clarke, 1986), anchoring (Watters, 1994) and shell reinforcement (Checa & Jiménez-

Jiménez, 2003a), but function remains poorly resolved. Being most pronounced during the

early postlarval period of unionoid life, understanding the functional morphology of

umbonal sculptures could give vital insights on ecological aspects of this crucial but poorly

understood life stage.

Attempts have been made to reconstruct the appearance of different types of rugae in

the course of unionoid evolution, but proposed models of character evolution are largely

contradictory (Fig. 5.1). Simpson (1900) believed the original character state of unionoid

beak sculpture to be ‘radial’ (Fig. 5.1A). This author argued that curving and coalescation

of one or more of the middle pairs of radial bars would result in a ‘divaricate’ sculpturing,

which would ultimately have developed in ‘concentric’ and ‘double-looped’ types,

exhibited by what he regarded as the most derived unionoid species. Largely agreeing with

Simpson’s model of character evolution, Modell (1942) (Fig. 5.1C) considered the

phylogenetic signal of umbonal sculptures of utmost importance and in fact based a whole

phylogenetic tree predominantly on this shell character. Ortmann (1912) (Fig. 5.1B), on

the other hand, not only disagreed with the phylogenetic importance of rugae above the

genus level but also believed that all types of umbonal sculpturing originated from a

‘concentric’ type. He considered the ‘divaricate’ (zig-zag) and ‘radial’ sculpture as the

Chapter 5 – Umbonal sculptures of Unionoida

81

most extreme conditions of two lines of development via intermediate ‘double-looped’ and

‘V-shaped’ stages, respectively.

Figure 5.1. A-C. Previously suggested models for character evolution of unionoid umbonal sculptures, simplified after descriptions by (A) Simpson (1900), (B) Ortmann (1912) and (C) Modell (1942).

Over the past 30 years or so, modern techniques including the use of molecular

markers, have drastically altered our view on unionoid phylogeny and evolution, which,

nevertheless remains a matter of intense scientific debate (see 1.1.1; Graf & Cummings,

2006b; Hoeh et al., 2009; Graf & Cummings, 2010a). Character evolution of umbonal

sculpture has not been revisited in this new context but this shell trait has repeatedly been

included in modern phylogenetic analyses (e.g. Hoeh, Bogan & Heard, 2001; Graf &

Cummings, 2006b). Based largely on one of the three aforementioned models of character

evolution (i.e. Simpson, 1900; Ortmann, 1912; Modell, 1942), these attempts have,

however, been rather unsuccessful. Following Ortmann’s (1912) model of character

evolution, Graf & Cummings (2006b), for example, classified beak shell sculpture into

three character states (1) ‘simple concentric’ or ‘absent’, (2) ‘radial’ and (3) ‘double-

looped’ or ‘zigzag’. Using this system of character state classification, Graf & Cummings’

(2006b) phylogenetic analysis resulted in a relatively low consistency index of only 0.222

for beak sculpture, indicating an unexpectedly high rate of change in this shell feature’s

morphology during unionoid evolution. Alternatively, the three character states used by


82

these authors might be inadequate in describing the morphological appearances of

unionoid umbonal sculptures.

Another caveat of almost all previous studies dealing with umbonal rugae

morphology is their general disregard of the morphogenetic aspect of sculpture formation.

According to Checa & Jiménez-Jiménez (2003b), bivalve ribs can be classified into three

categories, which differ in their fabricational morphology and consequently, are likely to

be of considerable phylogenetic value. (1) ‘Radial’ ribs, characteristic for e.g. cardiids, are

produced by particular portions of the mantle specialised for rib secretion and

consequently, are represented as helicospirals that converge towards the umbo (Checa,

2002; Checa & Jiménez-Jiménez, 2003b). (2) ‘Comarginal’ (‘concentric’) ribs are secreted

by periodic extension of the mantle all along its length and thus, always appear completely

parallel to the margin (Checa, 2002; Checa & Jiménez-Jiménez, 2003b). This ribbing

pattern is displayed for example in tellinids. (3) The ‘oblique’ (‘divaricate’) pattern is less

commonly found in bivalves. Ribs of this type have directions that are intermediate

between radial and comarginal, and are produced by an elaborate contact-guidance

behaviour of the mantle (Checa, 2002). Watters (1994) argued that all types of unionoid

beak sculpture are divaricate, with all other forms such as ‘double-looped’, ‘single-looped’

and ‘barred’ representing derivations of the same.

This study uses the beak sculpture from over 150 unionoid species, covering all

families but the Etheriidae, to (1) develop an up-to-date model of character evolution using

recent advances in understanding of bivalve phylogeny and rib formation, (2) identify

convergence of ecologically similar species, and on that basis, (3) discuss phylogenetic

significance and probable functional morphologies of these structures.


5.3.1. CLASSIFICATION OF BEAK SCULPTURE TYPES

Where possible, beak sculpture morphotype was determined directly on shell

material of the University Museum of Zoology Cambridge (UMZC) and private collections

(p.c.). To maximise number of taxa included in the analysis, beak sculpture morphology

was additionally examined on the basis of photographs and/or drawings from following

publications and websites: Unionoida from the Nearctica (Marshall, 1890; Coker et al.,

1921; Clarke, 1973, 1981, 1985; Frank & Lee, 1998; Grabarkiewicz & Todd, 2010;

Klocek, Bland & Barghusen, 2010), Neotropica (Mansur & Pereira, 2006; Mansur &


83

Pimpão, 2008; Pimpão, Rocha & de Castro Fettuccia, 2008; Cummings & Mayer, 2009;

Kohl, 2010), Afrotropica (von Martens, 1897; Pilsbry & Bequaert, 1927; Haas, 1929;

Scholz & Glaubrecht, 2004; Graf & Cummings, 2006a, 2007a), Palearctica (Germain,

1922; Kennard, Salisbury & Woodward, 1925; Killeen, Aldridge & Oliver, 2004; Araujo

et al., 2009), Indotropica (Savazzi & Peiyi, 1992; Fischer, 2007), Australasia (McMichael

& Hiscock, 1958; Playford & Walker, 2008) and worldwide (Modell, 1942; Haas, 1969b;

Graf & Cummings, 2010b). Further photographs of Indotropical unionids and Nearctical

margaritiferids were kindly provided by Ms Du Lina (Kunming Institute of Zoology,

China) and Ms Mary Sollows (University of New Brunswick, Canada), respectively.

Overall, beak sculpture of 156 modern unionoid species (about 300 specimens), covering

80 of the 161 currently described unionoid genera (according to Graf & Cummings,

2007b), was examined on the basis of shell and/or picture material (Table 5.1; Appendix

Table A.1).

For some taxa, only a small number of shells/pictures of well preserved umbonal

surface areas could be obtained. In these cases, additional information on beak sculpture

morphology was gathered from descriptions in the aforementioned and additional

publications (Simpson, 1900, 1914; Connolly, 1939; Modell, 1949; Dell, 1953; Parodiz &

Bonetto, 1963; Modell, 1964; Bonetto, 1966; Haas, 1969a; Ponder & Bayer, 2004). This

was the case particularly for taxa with apparently smooth umbos, i.e. Velesunioninae,

Mycetopodidae and Iridininae.

5.3.2. DEVELOPMENT OF MODEL OF CHARACTER EVOLUTION, AND IDENTIFICATION OF HOMOLOGIES AND HOMOPLASIES

Umbonal sculpture of each individual analysed was classified into at least one of

nine easily distinguishable morphological types. Single specimens that exhibited

intermediate forms and/or more than one form simultaneously were assigned to several

beak sculpture types. Based on this data on main and intermediate types of rugae observed

in the unionoid taxa examined, a model of beak sculpture evolution within the Unionoida

was subsequently developed. This model was further used as basis to identify homologies

and homoplasies (convergences) of sculpture morphotypes across the unionoid phylogeny.

This was done on the two currently best accepted cladograms on interfamiliar relationships

of the Unionoida, i.e. (1) Graf & Cummings (2006b) and (2) Bogan & Hoeh (2000) +

Hoeh, Bogan & Heard (2001). Although beak sculpture morphology had been included as

a character in both Graf & Cummings’ (2006b) and Hoeh, Bogan & Heard’s (2001)


84

phylogenetic analyses, non-independence of our analysis from these two studies can be

considered negligible for two reasons: (1) both Graf & Cummings’ (2006b) and Hoeh,

Bogan & Heard’s (2001) phylogenies are based on a large number of DNA sequences and

morphological data, and (2) in both cases, classification of beak sculpture character states

differs from the one applied in the present study.

5.4. RESULTS

5.4.1. MORPHOLOGICAL TYPES OF UNIONOID BEAK SCULPTURE

Beak sculptures of the more than 150 unionoid species analysed could be classified

into nine well distinguishable types (Fig. 5.2; Table 5.1):

(1) ‘V-shaped’ sculpture, characterised by one central “V” and at least one pair of

oblique lateral ribs, was found in some Afrotropical and Indotropical incertae sedis

unionids (e.g. Parreysia triembola (Benson, 1855); Fig. 5.2A) and members of the

Hyriinae (e.g. Prisodon corrugatus (Lamarck, 1819); Fig. 5.2B). Although no V-sculpture

could be found in any specimens of the family Iridinidae examined, Haas (1969a: p.583)

described this beak sculpture type from the iridinid species Mutela hargeri (Smith, 1908).

(2) In two of the four hyriid tribes and a few Indotropical unionids, predominantly of

the genus Parreysia, an almost radial beak sculpture is exhibited. However, in all of these

cases, ribs either (a) originate from several points near the umbo (e.g. Diplodon chilensis

(Gray, 1828); Fig. 5.2C) and/or (b) show at least one V- or Λ-shaped

convergence/divergence (e.g. Castalia ambigua Lamarck, 1819; Fig. 5.3A). Since such

irregularities would not occur in truly radial ribs sensu Checa (2002) and Checa &

Jiménez-Jiménez (2003b), we propose the term ‘pseudo-radial’ rather than ‘radial’ for this

kind of unionoid beak sculpture.

(3) ‘W-shaped’ (’zigzag’) sculpture, representing two or more series of connected

“V”s, was identified in several unionine, quadruline (e.g. Quadrula quadrula (Rafinesque,

1820); Fig. 5.2D) and incertae sedis unionids (e.g. Scabies crispata (Gould, 1843); Fig.

5.2E).

85

Table 5.1. Distribution of umbonal sculpture types in the Unionoida. Systematics and numbers of extant genera/species after Graf & Cummings (2007b). Numbers of analysed genera/species refer to specimens analysed from shell material, photographs and/or drawings (see Appendix Table A.1 for a complete list of individuals analysed, including their assigned sculpture type(s), and source, type and locality of material examined). As explained in the text, further information was gathered from descriptions in the literature. Abbreviations: c, pseudo-concentric (bars); d, double-looped; n, nodulous/pustulous; r, pseudo-radial; s, single-looped; sm, smooth; v, V-shaped; w, W-shaped; wr, wrinkled/corrugated.

According to description in 1McMichael & Hiscock (1958), 2Modell (1964), 3Haas (1969a) and 4Haas (1969b). *no data available

Family Subfamily Tribe Genera

analysed/extant Species

analysed/extant v r w d n wr c s sm Unionidae Unioninae Unionini 6 / 11 12 / 43

Anodontini 13 / 15 27 / 64 Ambleminae Quadrulini 5 / 7 5 / 27

Pleurobemini 5 / 10 8 / 102 Amblemini 1 / 1 1 / 3 Lampsilini 11 / 25 23 / 148 Gonideini3,4 0 / 1 0 / 1 i.s. Mesoamerica3,4 0 / 10 0 / 61

i.s. Palearctica - 2 / 4 3 / 5 i.s. Afrotropica - 6 / 8 16 / 38 i.s. Indotropica - 18 / 32 32 / 182

Margaritiferidae - - 1 / 1 2 / 12 Hyriidae

Hyriinae

Hyriini 1 / 1 1 / 3 Castaliini 1 / 2 2 / 10 Rhipidodontini 1 / 3 5 / 27 Hyridellini 3 / 4 6 / 14

Velesunioninae1,2,3,4 - 0 / 5 0 / 17 Etheriidae* - - 0 / 4 0 / 4 Mycetopodidae2,3,4

Mycetopodinae - 0 / 2 0 / 4 Anodontitinae - 1 / 2 1 / 18 Leilinae - 0 / 1 0 / 2 Monocondylaeinae - 0 / 6 0 / 12

Iridinidae Iridininae2,3,4 - 3 / 3 5 / 17 Aspathariinae - 2 / 3 7 / 26

Overall 80 / 161 156 / 840


86

Figure 5.2. A-O. Examples of the major types of unionoid beak sculpture: (A,B) V-shaped, (C) pseudo-radial, (D,E) W-shaped, (F,G) double-looped, (H,I) nodulous, (J) wrinkled, (K) pseudo-concentric, (L,M) pseudo-concentric bars, (N) single-looped, and (O) smooth. Left valves shown with anterior margin facing to the left. Scale bars = 3 mm. A. Parreysia triembola (Unionidae, i.s.), India; UMZC. B. Prisodon corrugatus (Hyriidae, Hyriini), Brazil; UMZC. C. Diplodon chilensis (Hyriidae, Rhipidontini), Chile; UMZC. D. Quadrula quadrula (Unionidae, Quadrulini), USA; UMZC. E. Scabies crispata (Unionidae, i.s.), India; UMZC. F. Alasmidonta marginata (Unionidae, Anodontini), Canada; UMZC. G. Villosa iris (Unionidae, Lampsilini), USA; UMZC. H. Unio pictorum (Unionidae, Unionini), UK; p.c.. I. Pseudanodonta complanata (Unionidae, Anodontini), UK; p.c.. J. Unio tumidus (Unionidae, Unionini), UK; p.c.. K. Anodonta cygnea (Unionidae, Anodontini), UK; p.c.. L. Elliptio arctata (Unionidae, Pleurobemini), USA; UMZC. M. Cristaria plicata (Unionidae, Anodontini), China; UMZC. N. Uniomerus tetralasmus (Unionidae, Pleurobemini), USA; UMZC. O. Delphinonaias delphinulus (Unionidae, Lampsilini), Mexico; UMZC.


87

(4) Almost all tribes of the Unionidae, the Margaritiferidae and one genus

(Aspatharia) of the iridinid subfamily Aspathariinae show a ‘double-looped’ sculpture

(e.g. Alasmidonta marginata Say, 1818; Fig. 5.2F; and Villosa iris (Lea, 1829); Fig. 5.2G).

A considerable degree of variation was observed in this beak sculpture type, which can

predominantly be attributed to differences in thickness of the ribs, in the angle of the two

loops and in the angle at the point of convergence of the two loops.

(5) Partially or completely ‘nodulous’ protrusions were found in members of almost

all families and subfamilies. These protuberances are often arranged in two or more

radiating rows (e.g. Unio pictorum (Linnaeus, 1758); Fig. 5.2H; Pseudanodonta

complanata (Rossmässler, 1835); Fig. 5.2I; Parreysia corrugata (Müller, 1774); Fig. 5.3B;

Coelatura aegyptiaca (Cailliaud, 1827); Fig. 5.3D; Unio tumidus Philipsson, 1788; 5.3F)

but can also appear irregularly distributed (e.g. Diplodon multistriatus (Lea, 1831); Fig.

5.3C; Tritogonia verrucosa (Rafinesque, 1820); Fig. 5.3E).

(6) Ribs in the shape of irregular zigzags, loops or corrugations were classified as

‘wrinkled’ (‘corrugated’) and were exhibited by almost all unionid subfamilies and tribes

(e.g. U. tumidus; Fig. 5.2J).

(7) In several unionid, margaritiferid and aspathariine specimens, ribs of nearly

concentric outline but never strictly following growth rings were found and classified as

‘pseudo-concentric’ (e.g. Anodonta cygnea (Linnaeus, 1758); Fig. 5.2K). The ventral part

of such ribs sometimes tends away from this curved structure towards a more linear “bar”

shape (e.g. Elliptio arctata (Conrad, 1834); Fig. 5.2L). In some species, these ‘pseudo-

concentric bars’ are completely free standing (e.g. Cristaria plicata (Leach, 1815); Fig.

5.2M).

(8) Another type of sculpture that has been misleadingly described as “concentric”

was exhibited by only a few genera of unionids (e.g. Toxolasma, Uniomerus). Loops of

such ‘single-looped’ rugae are sometimes orientated somewhat oblique to the growth

direction and usually originate from one point at the umbo (e.g. Uniomerus tetralasmus

(Say, 1831); Fig. 5.2N).

(9) The Mycetopodidae, the hyriid subfamily Velesunioninae and most members of

the iridinid subfamily Iridininae are all characterised by completely ‘smooth’ beaks. Apart

from these three taxa, specimens examined belonging to three species of the family

Unionidae (i.e. Pseudospatha tanganyicensis (Smith, 1880), Actinonaias pectorosa

(Conrad, 1834) and Delphinonaias delphinulus (Morelet, 1849); Fig. 5.2O) did not exhibit

any apparent beak sculpture.


88

5.4.2. INTERMEDIATE FORMS AND IMPLICATIONS FOR CHARACTER EVOLUTION OF UNIONOID BEAK SCULPTURE

(1) At least at one point of the shell of most ‘pseudo-radial’ forms, radiating bars

either converge or diverge to/from each other, forming a V- or Λ-shaped pattern (e.g.

C. ambigua; Fig. 5.3A). ‘Pseudo-radial’ sculpture may therefore be regarded as a derivate

of the ‘V-shaped’ and possibly also the ‘W-shaped’ type, potentially arising from a change

in angle at the points of convergence/divergence of the V-bars.

(2) In a few cases (e.g. U. tumidus; Fig. 5.3F), beak sculpture apparently represents

an intermediate form between ‘W-shaped’ and ‘double-looped’.

(3) ‘Nodulous’ protrusions were found to be associated with ‘V-shaped’ (e.g.

P. corrugata; Fig. 5.3B; D. multistriatus; Fig. 5.3C), ‘W-shaped’ (C. aegyptiaca;

Fig. 5.3D; T. verrucosa; Fig. 5.3E), ‘W-shaped/double-looped’ (e.g. U. tumidus; Fig. 5.3F)

and ‘double-looped’ sculpture.

(4) Several species presented irregular ‘wrinkles’ or corrugations associated with

either ‘V-shaped’, ‘W-shaped’ or ‘double-looped’ sculpture (e.g. Lasmigona compressa

(von Martens, 1860); Fig. 5.3G).

(5) Gradual alteration from ‘double-looped’ to ‘pseudo-concentric (barred)’ rugae

from the umbonal to the distal shell part within a specimen was commonly observed (e.g.

Elliptio dilatata (Rafinesque, 1820); Fig. 5.3H).

(6) ‘Single-looped’ rugae were present only in those higher taxa that also exhibited

‘double-looped’ sculpture, suggesting that this sculpture developed from the ‘double-

looped’ type (Table 5.1). While a gradual reduction, and ultimately omission of one of the

two loops from umbonal to distal shell parts was observed in several unionid species (e.g.

Pegias fabula (Lea, 1838); Fig. 5.3I), none of the shells examined displayed an umbonal

sculpture that represented an unambiguously intermediate form between the ‘double-

looped’ and the ‘oblique single-looped’ type of, for example, U. tetralasmus (Fig. 5.2N). It

is therefore unclear if the ‘regular single-looped’ type (of e.g. P. fabula; Fig. 5.3I) and the

‘oblique single-looped’ type (of e.g. U. tetralasmus (Fig. 5.2N)) represent

morphogenetically identical or different forms of beak sculpture.

(7) Although no specimens displayed clear intermediate stages between ‘V-shaped‘

and ‘W-shaped’ sculpture, both these forms were found within single genera (e.g.

Parreysia), indicating that these are very closely “related” forms.


89

Figure 5.3. A-I. Intermediate types of unionoid beak sculpture. Left valves shown with anterior margin facing to the left. Scale bars = 3 mm. A. Castalia ambigua (Hyriidae, Castaliini), Brazil; V-shaped – pseudo-radial. B. Parreysia corrugata (Unionidae, i.s.), India; V-shaped – pseudo-radial – nodulous. C. Diplodon multistriatus (Hyriidae, Rhipidontini), Brazil; V-shaped – pseudo-radial. D. Coelatura aegyptiaca (Unionidae, i.s.), Egypt; W-shaped – nodulous. E. Tritogonia verrucosa (Unionidae, Quadrulini), USA; W-shaped – nodulous. F. Unio tumidus (Unionidae, Unionini), UK; W-shaped – double-looped – nodulous. G. Lasmigona compressa (Unionidae, Anodontini), USA; double-looped – wrinkled. H. Elliptio dilatata (Unionidae, Pleurobemini), USA; double-looped – pseudo-concentric bars. I. Pegias fabula (Unionidae, Anodontini), USA; double-looped – single-looped. All specimens UMZC. Abbreviations: cb, concentric bars; d, double-looped; n, nodulous; r, pseudo-radial; s, single-looped; v, V-shaped; w, W-shaped; wr, wrinkled.

5.5. DISCUSSION

5.5.1. (IN)VALIDITY OF PREVIOUS MODELS OF CHARACTER EVOLUTION

Confirming observations by Watters (1994), beak sculptures of all shells examined

in this study can be assigned to the oblique (divaricate) type of sculpture formation.

Although some unionoids exhibit seemingly “radial” or “concentric (i.e. comarginal)” ribs,

and despite ubiquitous use of these terms by previous authors, none of the sculptures

examined in this study can be characterised as such in a morphogenetic sense. In fact, such


90

‘pseudo-radial’ and ‘pseudo-concentric’ sculptures should be regarded as derived forms of

oblique ribs. On this basis, previous models for character evolution of unionoid umbonal

sculptures proposed by Simpson (1900), Ortmann (1912) and Modell (1942), which

presume an original radial or concentric (comarginal) state, respectively, have to be

rejected. Occurrence of both ‘V-shaped’ and ‘W-shaped’ sculptures in very closely related

taxa, e.g. species within the genera Parreysia and Coelatura, provides a further argument

falsifying Ortmann (1912), who regarded these beak sculpture types as part of two separate

lineages of divergence (Fig. 5.1B).

5.5.2. PLESIOMORPHIC CHARACTER STATE AND THE FOSSIL RECORD

5.5.2.1. Smooth vs. sculptured plesiomorphic character state

All unionoid beak sculptures are of the divaricate type, which involves a

sophisticated contact-guidance mechanism of the mantle and has evolved only a few times

within the whole Bivalvia (Checa, 2002; Checa & Jiménez-Jiménez, 2003a). Against this

background, several facts indicate a common ancestor of the extant Unionoida exhibiting a

sculptured as opposed to a smooth beak:

(1) Oblique/divaricate sculpture is also present in many members of the closest

extant relatives of the Unionoida, the marine Trigonioida (e.g. Cox, 1969; Hoeh et al.,

1998). It is unlikely that oblique rib formation would have evolved independently in these

two very closely related groups. Other authors (e.g. Carter, Campbell & Campbell, 2006),

however, argued that Unionoida evolved from the Pachycardiidae, a Permian/Triassic

family with smooth beaks and currently unresolved systematic position (Haas, 1969b;

Newell & Boyd, 1975). This fact and the presence of apparently smooth beaks in some

Triassic unionoids, such as a “Mycetopoda-like” specimen described by Wanner (1921),

might indicate a smooth common ancestor of the Unionoida.

(2) If the plesiomorphic character state was a smooth umbo, beak sculpture would

have evolved several times within the Unionoida. Switching off or losing the ability for

sculpture production when “not needed” would involve fewer changes in genotype and

would thus be more likely to happen than repeated independent evolution of oblique

sculpture within a single bivalve order.

(3) Finally, morphologies of beak sculptures that would have evolved independently

from each other in case of a ‘smooth’ plesiomorphic scenario, are sometimes strikingly


91

similar (e.g. ‘V-shaped’ and/or ‘pseudo-radial’ sculpture of some Hyriinae and

Indotropical Unionidae).

5.5.2.2. Most likely plesiomorphic beak sculpture type

Presuming a sculptured plesiomorphic character state, the following facts indicate

that the basal condition was of the ‘V-shaped’ type:

(1) Of all unionoid beak sculptures observed in this study, the ‘V-shaped’ type is the

only one also exhibited by several ancient trigonioid genera, such as Asiatotrigonia,

Iotrigonia, Korobkovitrigonia, Orthotrigonia and Vaugonia (Cox, 1969; Cooper, 1991;

Checa & Jiménez-Jiménez, 2003a).

(2) Plesiomorphy of ‘V-shaped’ sculpture for the Unionoida is further supported by

similar sculpture patterns in early representatives of this group. For example, shells from

the North American Triassic, some of the oldest fossil records generally accepted to

belong to this group of bivalves, commonly exhibit ‘V-shaped’ and/or ‘pseudo-radial’

sculpture (Simpson, 1896; Wanner, 1921; Reeside, 1927; Modell, 1957; Good, 1989,

1998). Strikingly similar ‘V-shaped’, ‘W-shaped’ and ‘pseudo-radial’ sculptures can also

be found in the Cretaceous Trigonioidoidea of Eurasia (Gu, 1998; Guo, 1998b, a; Sha,

2007). Trigonioidoids are believed to be an extinct superfamily of the Unionoida (Sha,

2006) and may represent an independent parallel-evolution from a primitive ‘V-shaped’

type towards ‘W-shaped’ and ‘pseudo-radial’ sculpture (Guo, 1998a). It should be noted,

however, that rugae resembling the ‘double-looped’ type of modern unionoids are

exhibited by another relatively early (probably Jurassic) group of freshwater mussels, the

vetulonaians (Branson, 1935; Watters, 2001).

5.5.3. A NEW MODEL OF BEAK SCULPTURE CHARACTER EVOLUTION IN THE UNIONOIDA

Assuming a plesiomorphic ‘V-shaped’ umbonal sculpture and based on our

observations on intermediate forms of beak sculptures, we propose a new model of

character evolution of umbonal rugae in the Unionoida (Fig. 5.4), on which basis

homologies and homoplasies can be identified (Figs 5.5A,B). Two lines of development

originate from a primitive ‘V-shaped’ sculpture, which is retained in some Afrotropical

and Indotropical unionids, and members of the hyriid subfamily Hyriinae and iridinid

subfamily Iridininae: (1) ‘Pseudo-radially’ sculptured umbos developed at least twice, i.e.

in the Hyriinae and Indotropical Unionidae. (2) In the remaining Unionidae, but also the

Margaritiferidae and aspathariine Iridinidae, ‘W-shaped’ and ultimately ‘double-looped’


92

sculpture is exhibited. Assuming strictly unidirectional evolution from the ‘W-shaped’ to

the ‘double-looped’ character state, this change would have occurred at least five times

within the Unionoida (Fig. 5.5A,B). However, reciprocal evolution from ‘double-looped’

to ‘W-shaped’ beak sculpture cannot be excluded, particularly since these two sculpture

types differ only in the speed at which the double-looped waves travel in the longitudinal

direction (i.e. slower in the ‘double-looped’ than in the ‘W-shaped’ pattern). The

possibility of bi-directional evolution from ‘W-shaped’ to ‘double-looped’ beak sculpture

and vice versa is indicated by a double arrow in Figure 5.4 and merging of w- and d-

symbols in Figure 5.5.

‘Pseudo-concentric’ and possibly also ‘single-looped’ forms represent derivates of

the ‘double-looped’ type, and evolved several times within the Unionoida. Equally, the

‘nodulous’ and ‘wrinkled’ character states arose independently numerous times and may

be derived from ‘double-looped’, ‘W-shaped’ and ‘V-shaped’ sculpture. It is rarely

possible to determine which of these three types was the ancestral ribbing pattern to a

given ‘nodulous’ or ‘wrinkled’ form, as the underlying patterning is usually not retained in

specimens exhibiting these sculpture types. Lastly, ‘smooth’ forms can potentially develop

from any other character state.

Figure 5.4. A new model of character evolution of unionoid umbonal sculptures. In addition, completely smooth beaks may develop at any stage. Continuous arrows indicate evolutionary connection between two beak sculpture types evidenced by presence of *intermediate stages between two sculpture types, or #both sculpture types in species of the same genus. Dashed arrows indicate possible – but unproven – connection between two beak sculpture types. For discussion of directionality of character evolution see main text.

93

1 includes i.s. Palearctical, Afrotropical and Indotropical Unionidae (see Table 5.1) 2 member of i.s. Indotropical Unionidae in Graf & Cummings’ (2007b) classification (see Table 5.1) 3 according to Graf & Cummings (2006b, 2007b) member of the Etheriidae

Figure 5.5. A,B. Presence/absence of beak sculpture types in unionoid taxa, plotted on two phylogenetic cladograms with contrasting interfamiliar relationships (subfamiliar classification after Graf & Cummings (2007b)): A. Combined evidence phylogeny based on DNA sequence and morphology data (beak sculpture one of 59 morphological characters) after Graf & Cummings (2006b: fig. 4); missing taxon: i.s. Mesoamerican Ambleminae. B. Combined summary tree based on COI (Bogan & Hoeh, 2000) and combined COI + morphology data (beak sculpture one of 28 morphological characters) (Hoeh, Bogan & Heard, 2001); missing taxa: i.s. Mesoamerican Ambleminae, and i.s. Palearctical, Afrotropical and Indotropical Unionidae except Coelatura. Letters indicate occurrence of beak sculpture types in each taxon. Circles and according letters indicate synapomorphies of beak sculpture types. Abbreviations: c, pseudo-concentric (bars); d, double-looped; n, nodulous/pustulous; s, single-looped; r, pseudo-radial; sm, smooth; v, V-shaped; w, W-shaped; wr, wrinkled/corrugated.


94

5.5.3.1. Implications for unionoid phylogeny and evolution

The new model of beak sculpture evolution developed in this study (Fig. 5.4) has

profound implications on our view on homologies vs. homoplasies of umbonal sculptures

within the unionoids. For example, while Graf & Cummings (2006b) considered the

smoothness of umbos of the Velesunioninae as the plesiomorphic state, results of the

present study indicate synapomorphy of this character state for this hyriid subfamily. On

the other hand, morphological similarity of ‘V-shaped’ and ‘pseudo-radial’ sculpture of

some Indotropical unionids (e.g. Harmandia) and Neotropical hyriids (e.g. Castalia) are

considered as superficial (i.e. homoplasic) by Graf & Cummings (2006b). Our

observations, however, indicate the ‘V-shaped’ character state to be the plesiomorphic one

Knowledge on morphological characteristics of primitive and derived beak

sculptures can further be used to hypothesise on the phylogenetic position of unresolved

taxa, such as Parreysia and Coelatura. Thus, the primitive ‘V-shaped’ or ‘W-shaped’

sculpture exhibited by these genera suggests a basal position within the Unionidae.

Another example is the poorly understood phylogeny within the Ambleminae, consisting

of the tribes Amblemini, Lampsilini, Pleurobemini, Quadrulini and Gonideini (Table 1.1).

Several phylogenetic analyses (Lydeard, Mulvey & Davis, 1996; Campbell et al., 2005;

Graf & Cummings, 2006b), found the Quadrulini to be the basal tribe to all other tribes of

this subfamily. Based on the primitive ‘W-shaped’ sculpture present in most quadruline

specimens analysed in the present study, a basal quadruline tribe would conform with the

present model of character evolution. However, other authors obtained cladograms that did

not place the Quadrulini as the sole basal tribe of the Ambleminae (Bogan & Hoeh, 2000;

Graf & Foighil, 2000; Hoeh, Bogan & Heard, 2001). It will be interesting to see if future

phylogenetic studies on such unresolved taxa will support or undermine our proposed

model of character evolution.

5.5.3.2. Convergences and implications for probable functional morphologies

The pronounced difference between juvenile and adult shell sculpture implies that

rugae are unlikely to play an important role in adults but may be extremely important in

juvenile stages. Other differences between juvenile and adult shell morphology have been

observed, for example, in microsculpture (Chapter 6; Zieritz et al., in press) and lime

content (Coker et al., 1921). Such morphological discrepancies of the two life stages are

likely to be associated with differences in their ecology and/or behaviour, e.g. feeding


95

habit, burrowing depth, and susceptibility to dislodgement and predation (Coker et al.,

1921; Yeager, Cherry & Neves, 1994). In addition, there is some evidence of juveniles

requiring different environments than adults do (Hruska, 1999; Geist & Auerswald, 2007),

but due to their small size and low abundances, optimal habitat conditions for individual

species are generally unknown.

Understanding the functional morphology of beak sculpture could potentially be

helpful for reconstructing differences in juvenile habitat requirements of unionoid species.

A number of possible functions has been proposed for rugae, but most likely functional

morphologies are (1) anchoring, (2) shell reinforcement, and/or (3) burrowing (Watters,

1994; Checa & Jiménez-Jiménez, 2003a). All these three hypotheses are supported by our

observation that species characteristic of fast flowing, coarse sediment habitats often

exhibit ‘wrinkled’ or ‘nodulous’ umbos (e.g. Unio spp. (Killeen, Aldridge & Oliver, 2004),

Quadrula spp. (Parmalee & Bogan, 1998)). Repeated convergences to completely

‘smooth’ beaks occurred mainly in either (a) taxa characteristic of standing or slow

flowing habitats with soft-sediment (e.g. D. delphinulus, Velesunioninae), or (b)

“lanceolate” taxa which, in general, are deep and occasionally fast burrowers (e.g.

P. tanganyicensis, most Mycetopodidae and Iridininae). Whereas both these types of

convergences conform with an anchoring and/or reinforcing function of rugae, absence of

umbonal sculpture in several deeply burrowing unionoid species contradicts a burrowing

function of umbonal rugae.

CHAPTER 6

VARIABILITY, FUNCTION AND PHYLOGENETIC SIGNIFICANCE OF

PERIOSTRACAL MICROPROJECTIONS IN PALAEOHETERODONT BIVALVES

“Mrs Jalin: George?

Mr Jalin: Yes, Gladys.

Mrs Jalin: There's a man at the door with a moustache.

Mr Jalin: Tell him I've already got one. All right, all right. What's he want

then?

Mrs Jalin: He says do we want a documentary on molluscs.

Mr Jalin: Molluscs?!

Mrs Jalin: Yes.

Mr Jalin: What's he mean, molluscs?

Mrs Jalin: MOLLUSCS!! GASTROPODS! LAMELLIBRANCHS!

CEPHALOPODS!

Mr Jalin: Oh molluscs, I thought you said bacon.”

Monty Python

99

CHAPTER 6

VARIABILITY, FUNCTION AND PHYLOGENETIC SIGNIFICANCE OF PERIOSTRACAL MICROPROJECTIONS IN PALAEOHETERODONT

BIVALVES

6.1. ABSTRACT

Microprojections in the Palaeoheterodonta (Trigonioida and Unionoida) have been

little studied but could be important characters for resolving questions on phylogeny and

ecology of these bivalves. By investigating 26 unionoid and one trigonioid species using

scanning electron microscopy, we identified three types of periostracal microprojections.

(1) Microridges were present only in one respective species of the two unionoid families

Mycetopodidae (Anodontites trapesialis) and Iridinidae (Chambardia bourguignati), and

may represent a synapomorphy for the mycetopodid-iridinid clade. In A. trapesialis,

microridges were additionally equipped with (2) flag-like projections (microfringes),

possibly a synapomorphic character for the Mycetopodidae. Examination of partially

bleached specimens indicated that both microridges and microfringes are predominantly or

purely organic. In contrast, previously undescribed (3) spicule-like spikes represent

calcifications within the periostracum. These were found in 20 of the 27 species and four

of the six unionoid families. Particularly large and abundant spikes were exhibited in

umbonal (juvenile) parts of the shell, and species characteristic of fast flowing habitats.

These structures may thus serve in protecting the periostracum and shell underneath,

and/or stabilising life position by increasing shell friction. Microfringes and microridges,

on the other hand, possibly aid in the orientation of the mussel within the sediment.

6.2. INTRODUCTION

The outermost layer of bivalve shells, the periostracum, is a multilayered, quinone-

tanned protein (Beedham, 1958), primarily serving as a substrate for growth of the

calcified shell and allowing the necessary conditions for crystallisation to be achieved and

maintained (e.g. Taylor & Kennedy, 1969; Wilbur & Saleuddin, 1983; Checa, 2000). A

further periostracal function is that of a waterproof covering, preventing corrosion of the

underlying shell by acidic or poorly buffered waters (Taylor & Kennedy, 1969; Tevesz &


100

Carter, 1980) and therefore of particular importance in freshwater mussels (Unionoida).

Finally, Harper (1997) has shown that differences in periostracal thickness can constrain

formation of various shell morphologies and in particular, sculpture.

Unionoids comprise an estimated 840 species from six families and inhabit

freshwater habitats on all continents except Antarctica (Graf & Cummings, 2007b). Their

adults are suspension feeders and, with exception of the freshwater oysters (Etheriidae),

spend most of their lives in a (partially) buried position, though most species are also

capable of considerable horizontal movement across the sediment (Balfour & Smock,

1995; Amyot & Downing, 1997). Most taxonomic classifications consider the marine,

monogeneric order Trigonioida as the closest living relatives of the Unionoida, and group

these two bivalve orders into the subclass Palaeoheterodonta (e.g. Bieler & Mikkelsen,

1996; Hoeh et al., 1998; Graf & Cummings, 2006b). While structure, composition,

mineralogy and calcification processes of palaeoheterodont shells are relatively well

understood (e.g. Taylor & Kennedy, 1969; Taylor, Kennedy & Hall, 1969; Tevesz &

Carter, 1980; Carter, 1990; Checa, 2000; Checa & Rodríguez-Navarro, 2001; Marie et al.,

2007), surprisingly little is known about surface ‘microprojections’ in trigonioids and

unionoids. These structures usually are of merely a few µm in size and thus, can be made

visible only by high magnification microscopy.

Periostracal microsculptures are known to occur in several marine bivalve orders.

Some of these structures are not calcified (e.g. in Arcoida) or are secreted over the exterior

of the periostracum and thus, are not periostracal in origin (e.g. "periostracal" hairs of

some Mytiloidea) (Bottjer & Carter, 1980). Nevertheless, though initially believed to be a

purely organic layer (Gray, 1825), calcifications “within” the periostracum have since been

reported from members of various bivalve orders: Praenuculida, Mytiloida,

Modiomorphida, Trigonioida, Carditoida, Anomalodesmata, Veneroida and Myoida

(Aller, 1974; Carter & Aller, 1975; Newell & Boyd, 1975; Carter, 1978; Bottjer & Carter,

1980; Carter, 1990; Schneider & Carter, 2001; Glover & Taylor, 2010; Checa & Harper, in

press). Such calcifications are known to take up various shapes and sizes, up to more than

400 µm long needles in venerids (Glover & Taylor, 2010). However, detailed

investigations of the morphology and formation of these calcified structures are restricted

to Aller (1974), Glover & Taylor (2010) and Checa & Harper(in press). Thus, for the

majority of the taxa mentioned above, we do not know if these structures are actually

periostracal in origin or part of the calcified shell underneath.

Chapter 6 – Periostracal microprojections in Palaeoheterodonta

101

With respect to the Unionoida, we are aware of merely a single paper using high

magnification (electron) microscopy to describe periostracal microprojections in two

species of Anodontites, belonging to the unionoid family Mycetopodidae (Callil &

Mansur, 2005). Based on their observations on this one genus, these authors classified

periostracum projections into four size-related types: (1) macroscopically visible ‘folds’,

(2) mesoscopically visible ‘corrugations’, (3) ‘microridges’, and (4) ‘microfringes’ and

‘spikes’. Structures of type 1 and 2 have been described from several South American,

Chinese and African taxa of the mycetopodid (e.g. Anodontites), unionid (e.g. Cuneopsis)

and iridinid (e.g. Pleiodon) families (Marshall, 1926). However, apart from the

aforementioned mycetopodid taxa, nothing is known about occurrence of periostracal

microridges, microfringes and/or spikes in any other palaeoheterodont family and species.

What is more, we do not know anything about the mineralisation status (i.e. calcified or

organic) and formation process of these structures.

As a result of this lack of knowledge, periostracal microsculptures have so far not

been considered in phylogenetic analyses of Palaeoheterodonta (e.g. Hoeh, Bogan &

Heard, 2001; Graf & Cummings, 2006b). Including presence/absence data of homologous

types of microprojections in such studies might, however, help resolving questions

regarding taxonomic relationships and character evolution within this group of bivalves,

both issues currently under considerable debate (see Graf & Cummings, 2006b; Hoeh et

al., 2009; Graf & Cummings, 2010a). Furthermore, understanding the functional

morphologies of sculptures could potentially lead to new insights on habitat preferences

and behavioural characteristics of certain bivalve species and/or life stages. This

knowledge is of particular importance for developing conservation measures of the many

endangered freshwater bivalve taxa (Bogan, 1993; Williams et al., 1993; Lydeard et al.,

2004). All this necessitates, however, a basic knowledge on occurrence, morphology and

distribution of periostracal microprojections in the palaeoheterodont bivalves; data we are

currently lacking.

This study aims to fill this gap of knowledge. Incorporating species of all extant

palaeoheterodont families, we provide a first dataset on (1) morphology and mineralisation

status of their periostracal microprojections, (2) presence/absence of the different types of

microprojections across their phylogeny, and (3) inter- and intraspecific variability of

morphology, abundance and distribution of their microprojections. Potential implications

on palaeoheterodont phylogeny, character evolution and functional morphologies of the

structures are discussed.


102

(Right page) Table 6.1. Classification, collection locality and measurements of microprojections of the specimens examined. Phylogeny after Graf & Cummings (2007b). Abbreviations in brackets indicate source of specimens if not from private collection of one of the authors: UMZC, University Museum of Zoology Cambridge; NHML, Natural History Museum London. Other abbreviations: b, boss; mf, microfringe; mr, microridge; n.d., no data; s, spike; SD, standard deviation; X , mean.

103

Taxon Locality Sculpture umbonal (<1 cm shell length) distal (>2 cm shell length) density diameter covering density diameter covering [1/100µm²] [µm; X ± SD] [%] [1/100µm²] [µm; X ± SD] [%] Order Trigonioida

Family Trigoniidae Neotrigonia lamarcki (Gray, 1838) Moreton Bay, Australia b 0.9 5.7 ± 0.9 24 n.d. n.d. n.d.

Order Unionoida Family Unionidae

Subfamily Unioninae s.s. Anodonta anatina (Linnaeus, 1758) River Thames, UK s 3.6 2.3 ± 0.2 15 n.d. n.d. n.d. Anodonta cygnea (Linnaeus, 1758) River Great Ouse, UK s 2.8 2.2 ± 0.1 11 7.1 1.1 ± 0.1 7 Pseudanodonta complanata (Rossmässler, 1835) River Waveney, UK s 4.9 1.7 ± 0.3 11 n.d. n.d. n.d. Sinanodonta woodiana (Lea, 1834) Lake Dianchi, China s 4.0 1.8 ± 0.4 10 n.d. n.d. n.d. Unio pictorum (Linnaeus, 1758) River Thames, UK s 5.0 2.3 ± 0.3 20 3.2 0.9 ± 0.1 2 Unio tumidus Philipsson, 1788 River Thames, UK s 5.7 3.5 ± 0.5 56 n.d. n.d. n.d.

Subfamily Ambleminae Actinonaias pectorosa (Conrad, 1834) Clinch River, VA, USA s 6.1 1.6 ± 0.2 12 n.d. n.d. n.d. Elliptio complanata (Lightfoot, 1786) Potomac River, MD, USA s n.d. n.d. n.d. 6.1 1.4 ± 0.2 10 Epioblasma brevidens (Lea, 1831) unknown locality, USA s 3.4 2.6 ± 0.4 18 n.d. n.d. n.d. Lampsilis ovata (Say, 1817) Virginia Tech Hatchery, USA s 26.9 0.8 ± 0.1 13 n.d. n.d. n.d. Ligumia recta (Lamarck, 1819) Clinch River, TN, USA s 7.5 0.8 ± 0.1 4 n.d. n.d. n.d. Villosa iris (Lea, 1829) Virginia Tech Hatchery, USA s 4.8 2.2 ± 0.2 19 3.9 1.9 ± 0.3 11

incertae sedis unionids Coelatura horei (Smith, 1880) Lake Tanganyika, Tanzania s 3.7 2.2 ± 0.2 12 n.d. n.d. n.d. Coelatura leopoldvillensis (Putzeys, 1898) unknown locality (NHML) s 7.4 1.7 ± 0.1 17 9.2 0.8 ± 0.1 5 Lamellidens marginalis (Lamarck, 1819) Tangua Haor, Bangladesh s n.d. n.d. n.d. 7.0 0.8 ± 0.1 4 Nitia teretiuscula (Philippi, 1847) River Nile, Egypt s 28.3 1.2 ± 0.2 31 15.5 0.6 ± 0.1 5 Parreysia caerulea (Lea, 1831) Hakaluki Haor, Bangladesh s 3.5 2.7 ± 0.2 20 1.9 2.4 ± 0.4 9 Potomida littoralis (Cuvier, 1798) River Tua, Portugal - - - - - - - Pseudospatha tanganyicensis (Smith, 1880) Lake Tanganyika, Tanzania - - - - - - -

Family Margaritiferidae Margaritifera margaritifera (Linneaus, 1758) Ballindery Hatchery, UK s 5.3 3.6 ± 0.5 53 n.d. n.d. n.d.

Family Hyriidae Velesunio ambiguus (Philippi, 1847) Australia (UMZC) - - - - - - - Velesunio wilsonii (Lea, 1859) Queensland, Australia - - - - - - -

Family Etheriidae Etheria elliptica Lamarck, 1807 unknown locality s n.d. n.d. n.d. 14.4 0.7 ± 0.1 6

Family Mycetopodidae Anodontites trapesialis (Lamarck, 1819) São Francisco River, Brazil mr - - - - - - mf n.d. n.d. n.d. 27.7 1.5 ± 0.3 2

Family Iridinidae Chambardia bourguignati (Bourguignat, 1885) Tanzania (NHML) mr - - - - - - Pleiodon spekii (Woodward 1859) Lake Tanganyika (NHML) s 3.3 3.0 ± 0.4 24 n.d. n.d. n.d.


104


Shell samples with well preserved periostracum from 27 species, representing at

least one species from each of the six unionoid families and one species of the Trigonioida,

were collected or obtained from various sources (Table 6.1). During the initial stages of

this study, it became apparent that, if present at all, microsculptures were more abundant in

the umbonal (juvenile) region of the shell and thus, preferably young specimens were

studied. However, in some cases only large individuals could be obtained. These were

fragmented, and fragments with well preserved periostracum were selected from as close

to the umbo as possible. One to three individuals per species were analysed for occurrence

of periostracal microprojections using scanning electron microscopy (SEM).

To investigate if periostracal projections were calcified or purely organic, one to two

shells of Anodonta anatina, Unio tumidus, Lampsilis ovata and Anodontites trapesialis

were immersed in commercial bleach for 10-120 minutes (exact time given in respective

Figure legends) until the periostracum was partly digested, and subsequently washed in

distilled water.

One valve of A. anatina, Sinanodonta woodiana, Unio pictorum and U. tumidus was

prepared for visualisation of periostracal projections in transverse sections. For this

purpose, shells were embedded in resin, radially sectioned with a circular rock saw,

polished with a series of grit sizes, etched for 15 seconds in 5% HCl and subsequently

washed in distilled water.

Last, to examine a possible association of periostracal thickness and

presence/absence of microprojections, shell fragments of 12 unionoid species from five

families were selected for determination of periostracum thickness values.

All samples except those that had been bleached and those prepared for

determination of periostracal thickness were cleaned in an ultrasonic bath for

approximately 20 seconds. All samples were then dried at room temperature and

subsequently fixed on stubs with carbon adhesive tape, super glue or silver glue. After

sputter-coating the samples with gold the shell surface area was scanned, and images were

obtained using a JEOL 820 scanning electron microscope.

Measurements were made on image files obtained by SEM using program ImageJ

(Rasband, 2008). Images were scaled using scale bars obtained by SEM. Maximum and

minimum thickness of the periostracum was measured in one picture of a fragment in

transverse perspective per species. Densities and average sizes (diameter and height) of


105

discrete periostracal microprojections (i.e. spikes, bosses and microfringes) were estimated

for one specimen of each species exhibiting the respective structures. Dependent on the

availability of preserved periostracum, this was done on either (1) only the umbonal region

(i.e. <1 cm shell length), (2) only the distal region (i.e. >2 cm shell length) or (3) both the

umbonal and distal region. For this purpose, microprojections were counted on a randomly

selected area of one SEM image, and abundance per 100 µm² shell area calculated. On the

same images, 10 structures of each type present were selected randomly, their basal

maximum diameter measured, and average and standard deviation of diameters calculated.

Average proportion of shell area covered by the given microsculpture was estimated using

average basal diameter and density values.

All SEM specimens examined have been deposited at the Field Museum Chicago

(Neotrigonia lamarcki), the University Museum of Zoology Cambridge (Velesunio

ambiguus (Philippi, 1847): UMZC 2010.25) or the Natural History Museum London (all

remaining species: NHML Aq Zoo-2010-115).

6.4. RESULTS

6.4.1. OCCURRENCE AND MORPHOLOGY OF PERIOSTRACAL MICROPROJECTIONS IN PALAEOHETERODONTA

Periostracal microprojections present in the studied shell material can generally be

classified into three types: (1) microridges (Fig. 6.1A), (2) flap-like microfringes (Fig.

6.1A,B), and (3) spicule-like periostracal projections (spikes) (Fig. 6.2).

Figure 6.1. A-B. Examples of microfringes and microridges in two unionoid families. A. Anodontites trapesialis (Mycetopodidae). B. Chambardia bourguignati (Iridinidae). Abbreviations: mf, microfringe; mr, microridge.


106

Figure 6.2. A-H. Examples of spike appearance in four unionoid families. A. Anodonta cygnea (Unionidae, Unioninae); low density of spikes with pointed tips. B. Sinanodonta woodiana (Unionidae, Unioninae). C. Unio tumidus (Unionidae, Unioninae); tendency towards spikes distribution in comarginal rows. D. Actinonaias pectorosa (Unionidae, Ambleminae). E. Parreysia caerulea (Unionidae); outer periostracum apparently folding over spike. F. Margaritifera margaritifera (Margaritiferidae); high density of spikes with rounded (possibly abraded) tips. G. Etheria elliptica (Etheriidae). H. Pleiodon spekii (Iridinidae). Abbreviations: op, outer periostracum (fold); s, spike.


107

Microfringes were observed only in the one mycetopodid species studied

(A. trapesialis) and appeared as flag-like projections with bases of approximately 1-2 µm

in maximum length and <0.1 µm in thickness (Fig. 6.1A, Table 6.1). Commonly these

sculptures were associated with somewhat radially orientated microridges, upon which

they sit (Fig. 6.1A). Microridges lacking microfringes were additionally observed in one

iridinid species, i.e. Chambardia bourguignati (Fig. 6.1B). For more detailed

morphological description of both microfringes and microridges in A. trapesialis and

Anodontites elongatus (Swaison, 1823) see Callil & Mansur (2005).

Spikes, on the other hand, were found in 20 of the 27 species studied, with the

exception of the trigonioid Neotrigonia lamarcki, the unionids Potomida littoralis and

Pseudospatha tanganyicensis, the hyriids Velesunio wilsonii and Velesunio ambiguus, the

mycetopodid A. trapesialis and the iridinid C. bourguignati (Table 6.1). At the family

level, spikes were therefore observed in at least one member of the Unionidae,

Margaritiferidae, Etheriidae and Iridinidae (Fig. 6.2A-H, Table 6.1). No spikes could be

found in any of the specimens belonging to the unionoid families Hyriidae (two species)

and Mycetopodidae (one species; confirming observations by Callil & Mansur (2005)).

Figure 6.3. A-C. Bosses and associated structures in Palaeoheterodonta. A. Neotrigonia lamarcki (Trigonioida). B-C. Lampsilis ovata (Unionoida) bleached for 90 minutes. Abbreviations: b, boss; ip, inner periostracum; p, periostracum.


108

Finally, shell surfaces of specimens of the trigonioid N. lamarcki were covered with

regularly distributed, pronounced bosses, apparently associated with the prismatic layer

underneath (Fig. 6.3A). As described by Taylor, Kennedy & Hall (1969) and Newell &

Boyd (1975), these structures probably represent the initial calcification centres of prisms.

6.4.2. VARIATION IN SPIKE MORPHOLOGY, ABUNDANCE AND DISTRIBUTION ACROSS THE SHELL

Spikes typically appeared in a conical shape with a somewhat circular to elliptical

base ranging from less than 1 µm to approximately 5 µm in diameter, and protruded less

than 1 µm to approximately 2 µm above the periostracum (Fig 6.2A-H). Pointedness of

tips varied both between and within species. Anodonta cygnea had, for example,

predominantly rather pointed spikes (Fig. 6.2A), whereas Margaritifera margaritifera

spikes appeared uniformly blunt across the whole shell area (Fig. 6.2F). With regard to

variation within single individuals, a more blunted appearance was especially common in

older parts of shells.

Maximum spike diameter usually varied considerably within a single specimen

(Table 6.1) and a range of different sizes could even be seen on only a few µm² of shell

area (e.g. Fig. 6.2A,B,F). Nevertheless, in all six species for which available shell material

allowed us to compare umbonal and distal regions, average spike diameter was larger in

the umbonal compared to the distal part of the shell (Table 6.1). This implies a tendency of

spikes to become smaller towards the shell margin. Besides this variation within species

and individuals, some interspecific differences in spike size seemed striking. For example,

L. ovata and Ligumia recta spikes were on average less than 1 µm in diameter, despite the

fact that juvenile shells were analysed (Table 6.1). Equally, on the four, though relatively

distal fragments of two freshwater oyster (Etheria elliptica) specimens studied, only spikes

< 1 µm in diameter and height, were found (Fig. 6.2G, Table 6.1). In contrast, some of the

largest spikes with basal diameters of about 4-4.5 µm were observed on shells of

M. margaritifera (Fig. 6.2F), Unio tumidus (Fig. 6.2C) and Pleiodon spekii (Fig. 6.2H)

(Table 6.1).

Besides this variation in spikes size, our data suggest that spike count per unit area

differs considerably between species. Partly this appears to be inversely related to average

spike size, i.e. spikes of relatively small diameters regularly appeared at relatively high

densities (Table 6.1; e.g. in L. ovata, Coelatura leopoldvillensis and Actinonaias

pectorosa). Nevertheless, in some species, high spike densities were observed despite large


109

spike sizes (Table 6.1; e.g. M. margaritifera, U. tumidus). In addition to this interspecific

variation, spike density was also highly variable within the same specimen. In some

species, spikes were more abundant at the umbonal region when compared to more

recently formed shell areas (Table 6.1; i.e. U. pictorum, V. iris and Parreysia caerulea),

while other species showed the opposite pattern (Table 6.1; i.e. A. cygnea and

C. leopoldvillensis).

Figure 6.4. A-B. Occurrence of comarginal growth bands with and without spikes in Anodonta cygnea (Unionidae, Unioninae). A. Low magnification of shell surface (inset shown in B). B. High magnification of bands with and without spikes. Abbreviation: s, spike.

By combining our data on average spike diameters and densities we were able to

investigate trends in the proportion of the shell segment covered by spikes. On a species

level, percentage of umbonal areas covered by spikes was over five times higher in

U. tumidus and M. margaritifera than in L. recta and S. woodiana (Table 6.1). As with

spike size, a clear pattern emerged when looking at distribution across single shells.

Interestingly, in all the six specimens where both umbonal and distal shell fragments were

analysed, relative spike coverage was always higher in the umbonal compared to the distal

region (Table 6.1). What is more, at later parts of the shell a pattern of comarginal bands

lacking spikes intercepted by bands carrying spikes was found in several species (Fig.

6.4A,B). Such spike-free growth bands were observed in A. cygnea, V. iris, Epioblasma

brevidens and C. leopoldvillensis. Late shell fragments of even larger sized individuals of


110

some species (e.g. U. pictorum, A. cygnea) completely lacked spikes, indicating a decrease

of spike density from early to late shell parts.

A consistent pattern regarding spike density or size distribution along the anterior-

posterior axis could not be found. However, in some individuals such as the one available

L. recta specimen, spikes were more abundant at the posterior compared to the anterior

region of the shell.

Last, in the palaeoheterodont species studied, spikes usually appear to be randomly

distributed across the shell surface (Fig. 6.2A,B,D,F,G,H). However, in one individual of

U. tumidus, a tendency to a distribution in comarginal rows was observed (Fig. 6.2C). We

were not able to detect any consistency of this distribution pattern however, i.e. spikes in

other areas of the same individual were distributed randomly.

6.4.3. MINERALISATION STATUS OF PERIOSTRACAL MICROPROJECTIONS

Though mostly of a somewhat circular shape, in some cases, spikes exhibited an

almost hexagonal outline (Fig. 6.5), suggesting that these represent singular or twinned

aragonitic crystals. Further evidence for a calcified nature of these structures is provided

by spike skeletons that remained after partly bleaching organic shell parts of A. anatina,

U. tumidus and L. ovata specimens (Fig. 6.6B). On the other hand, no projections but only

the bases of microfringes were to be seen after bleaching of A. trapesialis shell fragments

known to carry microridges and –fringes (Fig. 6.6A). This implies that apart from a

possibly mineralised core of microfringes, these structures consist of organic material

which was oxidised during the bleaching process.

Figure 6.5. Periostracal spikes in Unio pictorum (Unionidae, Unioninae). Abbreviation: s, spike.


111

Furthermore, high magnification pictures of spikes of several specimens suggest that

these apparently calcified structures are covered by a thin sheet of periostracum (possibly

the outer periostracum layer), resulting in a wrinkled appearance (Fig. 6.2B,E). This

hypothesis is supported further by some pictures of spikes in transversal sections (Fig.

6.7), where the spike appears to be situated in between the inner and the outer

periostracum.

Figure 6.6. A-B. Periostracal microsculptures in unionoids after partial bleaching of organic parts of the shell. A. Anodontites trapesialis (Mycetopodidae) bleached for 120 minutes. B. Anodonta anatina (Unionidae, Unioninae) bleached for 10 minutes. Abbreviations: ip, inner periostracum; mf, microfringe; s, spike.

Figure 6.7. A-B. Section through periostracal microsculptures of two unionoid species. A. Sinanodonta woodiana. B. Unio pictorum. Abbreviations: ip, inner periostracum; op, outer periostracum; pr, prism; s, spike.


112

6.4.4. TRIGONIOID ‘BOSSES’

The one trigonioid studied, N. lamarcki, possessed large (i.e. average diameter of 5.7

µm), regularly-spaced bosses (Table 6.1, Fig. 6.3A). These bosses were largest and best

developed in troughs between ribs. The fact that one boss always originates from the

centre of one prism (Fig. 6.3A) indicates that these structures represent the calcification

centres of prisms. Although no bosses could be observed in any untreated unionoid

specimens, similar structures appeared after partial bleaching of organic shell parts in

L. ovata (Fig. 6.3B,C) and three other unionoid species that were examined in this respect.

Unionoid bosses equally represent calcification centres of prisms but never protrude above

the periostracum as trigonioid bosses do.

6.4.5. INFLUENCE OF PERIOSTRACAL THICKNESS ON PRESENCE/ABSENCE OF MICROPROJECTIONS

No correlation could be observed between presence/absence of microprojections and

thickness of periostracum across 12 palaeoheterodont species analysed (Table 6.2). Thus,

spikes were present in both species with extremely thin (e.g. S. woodiana) and extremely

thick (e.g. M. margaritifera) periostraca.

Table 6.2. Minimum and maximum periostracum thickness values measured for 15 palaeoheterodont species. Sorting order from lowest to largest maximum thickness. Abbreviations: b, boss; mf, microfringe; mr, microridge; s, spike.

Species Type(s) of

microprojection present

Approximate periostracum thickness [µm]

Source (if not this study)

Sinanodonta woodiana s 0.5 – 1 Unio pictorum s 0.5 – 2 Pleiodon spekii s 1.5 – 2.5 Potomida littoralis - 2 – 3 Checa (2000) Anodonta anatina s 2.5 – 3.5 Lampsilis ovata s 4 – 5 Unio tumidus s 4.5 – 6 Pseudospatha tanganyicensis - 4.5 – 6.5 Velesunio ambiguus - 5 – 6.5 Anodontites trapesialis mr, mf 4 – 7 Neotrigonia spp. b 3 – 12* Harper (1997) Chambardia bourguignati mr 10 – 12.5 Etheria elliptica s 14* Harper (1997) Coelatura horei s 10 – 15.5 Margaritifera margaritifera s 16.5 – 20.5

*values represent mean values (for four species in the case of Neotrigonia spp.)


113

6.5. DISCUSSION

6.5.1. OCCURRENCE OF PERIOSTRACAL MICROPROJECTIONS ACROSS THE UNIONOID PHYLOGENY

In the 27 palaeoheterodont taxa studied, three different types of periostracal

microprojections were identified. Non-calcified ‘microridges’ were present only in one

mycetopodid (A. trapesialis) and one iridinid (C. bourguignati) species. In A. trapesialis,

these were additionally accompanied by flag-shaped ‘microfringes’. Though calcified

periostracal ‘spikes’ were present in most shells examined, we were not able to find these

structures in the trigonioid N. lamarcki and six species from four of the six unionoid

families, including all hyriid and mycetopodid species analysed.

A further type of microprojection was observed only in the one trigonioid species

studied. These ‘bosses’ protrude from the prismatic layer of the shell itself and are thus

possibly not periostracal in origin.

The main objective of this study was to provide a first overview on periostracal

protrusions exhibited across the palaeoheterodont phylogeny. As a consequence, only a

limited number of specimens per species (usually one) could be investigated, which

precluded investigation of intraspecific variability of these structures. Nevertheless, despite

the limited and preliminary nature of our dataset, there appears to be some phylogenetic

signal in the distribution of features recorded.

Bosses, found only in N. lamarcki, most likely represent the calcification centres of

the aragonitic prisms (as previously suggested by Taylor, Kennedy & Hall (1969) and

Newell & Boyd (1975)), which ultimately “pierce” through the overlaying periostracum.

Although similar structures can be found after bleaching of the organic periostracum of

unionoid shells, bosses of freshwater palaeoheterodonts examined were never as well

developed and always covered by the periostracum. This and the fact that Newell & Boyd

(1975) also found bosses in several fossil trigonioid taxa indicates that this character could

be synapomorphic for the Trigonioida.

Despite the highly debated and insufficient understanding of most interfamilial

relationships in the Unionoida, most modern phylogenies consider the Mycetopodidae and

Iridinidae as closely related (Fig. 6.8). The fact that microridges were present exclusively

in members of these two families could therefore indicate a homologous nature of this

morphological trait, which potentially evolved only after divergence of the mycetopodid-

iridinid branch (Fig. 6.8A; phylogeny after Graf & Cummings (2006b)) or the iridinid-


114

mycetopodid-etheriid branch (Fig. 6.8B; phylogeny after Bogan & Hoeh (2000) and Hoeh,

Bogan & Heard (2001)). Similarly, exclusiveness of microfringes to the mycetopodids

could indicate evolution of this shell character just before this family branched off from

the remaining unionoids. Our dataset can be extended by observations on one additional

mycetopodid species by Callil & Mansur (2005), who found similar microridges and

microfringes in A. elongatus. As these conclusions are drawn from just two iridinid and

two mycetopodid species of only one genus, respectively, it will be important and

interesting to determine whether or not other genera of these unionoid families exhibit the

same morphological traits.

1 according to Graf & Cummings (2006b, 2007b) member of the Etheriidae

Figure 6.8. A-B. Presence/absence of microprojections in the palaeoheterodont families plotted on two contrasting phylogenetic cladograms (both redrawn from Hoeh et al. (2009)). A. Combined evidence phylogeny after Graf & Cummings (2006b: fig. 4). B. Combined summary tree based on COI (Bogan & Hoeh, 2000) and combined evidence data (Hoeh, Bogan & Heard, 2001). Letters indicate occurrence of microprojections in each taxon: b, bosses; mf, microfringes; mr, microridges; s, spikes. Grey circles and associated letters indicate synapomorphies of the four characters.

Due to their distinct morphology and mineralised nature, spikes, observed in all

unionoid families except hyriids and mycetopodids, should be considered as a different,

possibly non-homologous character state to the predominantly un-mineralised

microfringes. As recently discussed by Hoeh et al. (2009) and Graf & Cummings (2010a),

the position of the Hyriidae in the unionoid phylogeny is of particular uncertainty. Several

analyses based on morphology (Graf, 2000; Hoeh, Bogan & Heard, 2001), combined

morphology + mtDNA (COI) (Roe & Hoeh, 2003), and combined morphology + mtDNA

(COI) + nuclear ribosomal DNA (28S) (Graf & Cummings, 2006b) (Fig. 6.8A) support a

close relationship of hyriids and “etherioids” (i.e. Etheriidae + Mycetopodidae +


115

Iridinidae). Alternative phylogenetic analyses based on COI sequences (Bogan & Hoeh,

2000; Hoeh et al., 2002; Walker et al., 2006), and combined morphology + COI data

(Hoeh, Bogan & Heard, 2001) (Fig. 6.8B) indicate that the Hyriidae is the most basal

family of unionoids. Against this background, absence of spikes in the hyriids could

potentially be interesting. That is, if the second phylogenetic scenario was correct, the

ability to produce spikes might have evolved only after the split of hyriids from the rest of

the unionoids.

On the other hand, based on the data available, spikes are also absent in the

Mycetopodidae, regarded as one of the most derived unionoid families. Thus, we failed to

detect spikes in A. trapesialis, and though Callil & Mansur (2005) mention the presence of

“spikes” in this species and A. elongatus, closer inspection of their figures suggests that

what they observed in fact represented merely the tips of microfringes in a situation where

the remaining part of the fringe was covered by extraneous material. Spike production

might therefore have been secondarily “switched off” in the mycetopodids. Interestingly,

all species exhibiting microridges (that is the mycetopodid Anodontites spp. and iridinid

C. bourguignati) also lacked spikes, which might imply that these two structures are never

produced at the same time. However, spikes were found in the second iridinid species

under study. Similarly, in the Unionidae both species with and without spikes were found,

though it should be noted that the phylogenetic positions of both spike-free unionid species

(P. littoralis and P. tanganyicensis) within this family are unclear (incertae sedis). In any

case, these inconsistencies in spike formation within evolutionary entities show that

expression of this character has been secondarily lost on several occasions, indicating that

spike production is probably easy to be “switched off”.

Further studies on other palaeoheterodont species will be needed to support or

contradict our preliminary hypotheses on character evolution of palaeoheterodont

microprojections. Indeed, in older shells with partly damaged periostracum,

microprojections were often very difficult to find and thus, future studies might well

identify these structures in species we now describe as “free of microprojections”.

6.5.2. MORPHOLOGICAL VARIATION AND POSSIBLE FUNCTIONAL MORPHOLOGIES OF STRUCTURES

The morphological range of palaeoheterodont spikes appeared as rather uniform,

especially when compared to spikes in the Anomalodesmata, which are known to exhibit

morphologies ranging from pointed-conical spikes to flat plaques (Checa & Harper, in


116

press). Nevertheless, a certain amount of variation was observed in the degree of

pointedness, size and abundance of spikes, both between species and across the shell of

single specimens belonging to the Palaeoheterodonta.

A trend of pointed spikes at newly produced parts of the shell towards gradually

more blunt spikes closer to the umbonal area was observed in several specimens and

indicates that originally pointed spikes become abraded by the surrounding sediment.

Further support for this hypothesis is provided when comparing spikes of the freshwater

pearl mussel M. margaritifera with that of the swan mussel A. cygnea. The former species

is known to live in oligotrophic streams and rivers with coarse substrate and often high

current velocities (Altnöder, 1926; Killeen, Aldridge & Oliver, 2004). These habitat

features would be expected to result in a fast abrasion and blunt appearance of spikes

across the whole of the shell. The spikes of the A. cygnea specimen, which was collected

from a slow-flowing, muddy habitat, typical for this species (Stone et al., 1982; Killeen,

Aldridge & Oliver, 2004), on the other hand, were almost continuously “fresh” and

pointed. Consequently, one function of spikes of the (semi)infaunally living unionoids

could be a mechanical protection of the (inner) periostracum, which is of particular

importance to freshwater mussels. Additionally or alternatively, a spiky shell surface will

also increase friction of the shell within the sediment. Thus, Aller (1974) hypothesised that

spikes in the infaunal marine anomalodesmatan genus Laternula function either as active

burrowing aids or stabilisers of life position (by increasing friction force).

Several additional observations support two of the aforementioned functions of

palaeoheterodont spikes, i.e. (1) protection of the periostracum and underlying shell, and

(2) stabilising the mussel’s life position. Firstly, a particularly large proportion of shell

area was covered by comparatively large and densely packed spikes in U. tumidus and

M. margaritifera. Both these species prefer fast flowing stream and river habitats (Killeen,

Aldridge & Oliver, 2004), and an increased friction would improve anchoring of the shell

in the sediment and thus, withstanding dislodgement by high current velocities. A. cygnea

and S. woodiana, on the other hand, occupy slower flowing rivers and/or lakes (Killeen,

Aldridge & Oliver, 2004) and appeared at the other end of the scale regarding spike size

and shell area covered by spikes. Only very small spikes were also found in E. elliptica,

which can perhaps be explained by this species’ unique life habit. Freshwater oysters are

cemented to hard substrate by one valve (Yonge, 1962), which results in the reduction of

shell abrasion by sediment and consequently, the hypothetical advantage a shell with

increased friction would bring. Last, an absolutely smooth periostracum was presented by


117

the African unionid P. tanganyicensis. Although almost nothing is known about the

ecology of this bivalve, it has been observed to be an extremely fast burrower (E. Michel,

personal communication); secondary loss of spike production might aid this capability.

In addition to differences between species, spike size varied considerably within

single specimens. Partly this is probably due to variability in ‘spike age’ (developmental

stage of spikes). Nevertheless, we did find a consistent trend of spikes becoming smaller

and occupying a smaller proportion of the surface area from umbonal to distal shell parts.

On a functional level, this pattern would equal greater shell friction for young individuals

compared to older ones. Juvenile unionoids are known to live interstitially as deposit

feeders up to a certain life stage, when they change to a semi-infaunal, suspension feeding

life habit (Yeager, Cherry & Neves, 1994). Higher spike density would thus increase

friction and consequently, stability of the juvenile within the sediment. Additionally or

alternatively, a spiky surface would be expected to alter flow patterns around the shell,

which could be of particular importance to small/juvenile mussels. Last, the burrowing

habit of trigonioids and unionoids makes the umbonal area the shell part most liable to

erosion (Trueman, 1966a). Thus, a higher abundance and size of spikes in this shell region

would be expected if spikes serve as a protection of the shell underneath.

Interestingly, Callil & Mansur (2005) found a similar pattern of higher abundance of

microfringes in juvenile compared to later parts of A. trapesialis and A. elongatus shells.

These authors suggest an association of these structures with increased sensitivity to water

flow, possibly aiding the juvenile in orientating itself within the sediment. Given their

extremely thin (<0.1 µm) and mostly uncalcified morphology, mycetopodid microfringes

are unlikely to offer any notable frictional improvements. Their function is thus likely to

differ from that of other unionoids’ spikes. Although Callil & Mansur (2005) do not give

any suggestion for functional morphology of microridges, their morphological

characteristics, i.e. parallel rows of ridges, may be effective in aiding the orientation of

mussels.

Apart from the functional explanations given above, presence/absence and

development of certain types of periostracal microprojections might be associated with

other morphological characteristics of the shell, in particular with thickness and

composition of the periostracum. For example, spike production might be ceased below a

certain threshold of periostracal thickness. However, we failed to detect any obvious

correlation of occurrence of microprojection and thickness of periostracum (Table 6.2).


118

6.5.3. COMPARISON TO SPIKES OF OTHER BIVALVE GROUPS AND IMPLICATIONS FOR BIVALVE PHYLOGENY

Periostracal calcifications such as the unionoid spikes described in this study have

been reported from several other bivalve groups, most notably the Veneroida (Glover &

Taylor, 2010), Anomalodesmata (Checa & Harper, in press) and Myoida (Carter, 1978).

According to Glover & Taylor (2010) and Checa & Harper (in press), present evidence

suggests that calcification of the periostracum has occurred independently in several

bivalve clades. Checa & Harper (in press) hypothesise, however, that spikes of the closely

related Anomalodesmata and Palaeoheterodonta might be homologous structures. The

latter theory is supported by our observation that, similarly to those in anomalodesmatans,

unionoid spikes are probably formed immediately below the outer periostracum. On the

other hand, morphology of anomalodesmatan spikes differs somewhat from their

palaeoheterodont relatives. In particular, anomalodesmatan spikes are generally much

larger (i.e. about 20 µm in diameter; Checa & Harper, in press) and often arranged in

comarginal rows (e.g. in Laternula flexuosa Reeve, 1863 (Aller, 1974)). Furthermore, we

could not identify spikes in the Trigonioida, generally considered as a sister clade of the

Unionoida (Newell & Boyd, 1975; Healy, 1989; Hoeh et al., 1998; Watters, 2001). If

anomalodesmatan and unionoid spikes were homologous structures, this could therefore

mean a loss of character development in the trigonioids. Alternatively, spikes could be a

synapomorphic character of the Unionoida.

The matter is further complicated by the unresolved formation process of the

protruding trigonioid bosses. Though we observed bosses directly at the surface of

N. lamarcki shells, it is not clear that these protrusions were not initially covered by a

pellicle-like outer periostracum. Trigonioid bosses may in fact develop in those parts of the

prisms embedded within the periostracum, which could indicate a common ability of

intraperiostracal calcification, synapomorphic to Palaeoheterodonta, Anomalodesmata and

possibly other heterodont bivalve taxa. An improved understanding of the production

mode of spikes and protruding bosses in the Palaeoheterodonta will resolve these open

questions.

CONCLUSIONS

121

CHAPTER 7

CONCLUSIONS

In this thesis I aimed to better understand the factors which determine the

morphology of freshwater mussels (Unionoida). The morphological makeup (phenotype)

of a given freshwater mussel’s shell can be influenced by a range of factors and

mechanisms. These include (1) the specimen’s genetic composition, (2) its life habit and

surrounding environmental conditions, and (3) other characteristics of the individual such

as its sex, age, size and level of parasitic infestation. The results of this thesis show that all

these factors may play a part, but that their relative importance varies considerably

between different shell features, different populations and different species. By

understanding the factors that determine shell morphology, we have an opportunity to use

inter- and intraspecific patterns for the reconstruction of environments and habitat

requirements of species, determination of sex ratios or parasitic loads of individuals and

populations, and gain new insights into evolutionary trends and phylogenies.

Due to the great intraspecific variability in shell ‘size’, ‘shape’, ‘inflation’ and

‘thickness’, these characters were investigated with respect to patterns within single

unionoid species (Chapter 2-4). Chapter 5 and 6 took a broader look at unionoid shell

morphology to consider the taxonomic value and functional importance of macroscopic

(umbonal sculpture) and microscopic (periostracal microprojections) shell features.

Historically, considerable attention has been paid to unionoid shell morphology and

its functional and taxonomic significance. However, the absence of clear patterns and

numerous conflicting interpretations mean that the last major contribution to this field was

made over 30 years ago (Eagar, 1978). The opportunity to gain new insights into the

factors that determine unionoid shell morphology has been made possible by the

availability of modern statistical and analytical techniques. These tools have enabled me to

reconcile the apparent contradictions in the literature and identify new features of

taxonomic importance.

Fourier shape analysis provided a new and more powerful tool for the investigation

of shell shape in freshwater mussels, yielding consistent patterns across taxa which could

not be detected using traditional morphometric methods. Amplified fragment length

polymorphism (AFLP) analysis provided an opportunity to investigate intraspecific

variation in genotype within a small geographic scale and revealed that phenotypic


122

plasticity was of overriding importance in this instance. Scanning electron microscopy,

combined with access to extensive and well catalogued shell collections enabled me to

conduct studies of macro- and microscopic shell sculptures and identify new features of

considerable taxonomic and phylogenetic value.

7.1. WHAT CAN WE LEARN FROM A UNIONOID SHELL?

Based on results of this thesis and previous publications, Table 7.1 summarises the

potential utility of 12 unionoid shell characters for taxonomic classification and

phylogenetic reconstructions, environmental reconstructions, and the determination of sex

and parasite loads. The relative importance of these features for different applications is

considered below.

123

Table 7.1. Factors influencing intraspecific trends in 12 unionoid shell characters with an assessment of their value in four fields of application: GOOD, generally suitable for application; LIMITED, application limited (e.g. to only certain genera or species); BAD, unsuitable for application; I.D., insufficient data; N.D., no data. Cells highlighted in grey indicate interactions studied in this thesis with respect to interspecific morphological variation (light grey) and intraspecific variation (dark grey), respectively.

Shell character

Systematics & Phylogeny

Environmental reconstructions (factor habitat)

Sex determination

(factor sex)

Reconstruction of trematode load

(factor trematodes)

Size (at given age) LIMITED (e.g. Parmalee & Bogan, 1998) GOOD* (2,4) LIMITED*

(e.g. Jass & Glenn, 2004) LIMITED*

(e.g. Taskinen, 1998)

Dorso-posterior margin shape (arched vs. pointed)

LIMITED (e.g. Parmalee & Bogan, 1998) GOOD* (2,3,4) BAD* (2) BAD

Wing development LIMITED (e.g. Killeen, Aldridge & Oliver, 2004) LIMITED* (2,4) BAD BAD

Elongation LIMITED (e.g. Parmalee & Bogan, 1998) LIMITED* (2,3,4) BAD LIMITED* (4)

Inflation (obesity) LIMITED (e.g. Parmalee & Bogan, 1998) LIMITED* (2,3,4) LIMITED* (2,4) LIMITED* (4)

Thickness LIMITED (e.g. Parmalee & Bogan, 1998) I.D.* (4) LIMITED* (4) BAD

Density N.D. I.D.* (4) BAD BAD

Adductor (scar) size BAD GOOD* (2) N.D. N.D.

Umbonal sculpture GOOD (5) GOOD (5) N.D. N.D.

Periostracal spikes i.d. (6) GOOD (6) N.D. N.D.

Microridges GOOD (6) N.D. N.D. N.D.

Microfringes GOOD (6) N.D. N.D. N.D.

* indicate presence of a statistically significant influence of the respective factor on the respective shell character found in chapters of this thesis or other publications (chapter numbers and references in brackets)


124

7.1.1. SYSTEMATICS, PHYLOGENY AND EVOLUTION

Despite recent methodological and analytical advances, including the use of

molecular markers, unionoid phylogeny is still poorly understood. Under particularly

intense debate are the phylogenetic positions of the Hyriidae within the Unionoida (e.g.

Graf & Cummings (2006b) vs. Hoeh et al. (2009)), several tropical genera within the

Unionidae (e.g. Graf & Cummings, 2006b) and the Quadrulini within the unionid

subfamily Ambleminae (e.g. Campbell et al. (2005) vs. Hoeh, Bogan & Heard (2001)).

These unresolved questions are, to a great extent, the result of our lack of understanding of

which shell features are suitable for the use in reconstructions of bivalve phylogenies. That

is, a shell character is of high phylogenetic value if it does not exhibit a high degree of

intraspecific variation and if interspecific patterns in this shell feature reflect bivalve

evolution rather than convergences of unrelated taxa. What is more, with respect to those

characters that currently are used in phylogenetic analyses, we usually lack a proper

understanding of character evolution and thus, cannot discriminate between

phylogenetically informative (homoplasic) and uninformative (convergent) character

states.

7.1.1.1. “Gross shell morphology”: size and form

To a limited degree, gross shell morphology can be useful for taxonomic

classification and phylogenetic reconstructions. Shell features that are to some extent

informative and reliable at the species and higher taxonomic levels include ‘shell size’

(e.g. Anodonta cygnea (Linnaeus, 1758) attains much larger maximum shell sizes than

Anodonta anatina (Linnaeus, 1758)) and ‘presence/absence of wings’ (e.g. wings are

present in Anodonta spp. but not in Unio spp.) (e.g. Parmalee & Bogan, 1998; Killeen,

Aldridge & Oliver, 2004). However, due to the high degree of interspecific convergences

and intraspecific variability in shell ‘size’, ‘form’ and ‘thickness’, the use of these shell

characters in higher taxonomic classifications or phylogenetic reconstructions is generally

regarded as problematic (e.g. Ortmann, 1912; Prashad, 1931; Haas, 1969a, b; Davis, 1983,

1984).

The fact that shell ‘shape’ of a unionoid is a poor indicator of its genetic composition

was demonstrated in Chapter 3. Thus, genetic comparisons of three paired Unio pictorum

(Linnaeus, 1758) populations from two habitats failed to identify consistent genetic

differences between two shell ecomorphotypes. This suggests that pronounced

Chapter 7 - Conclusions

125

morphological differences in the degree of shell elongation and the shape of the dorso-

posterior margin in this unionid species are caused by plasticity of the phenotype and

provides the first good evidence for phenotypic plasticity of shell shape in a European

freshwater mussel. Genetic differences along the River Thames, on the other hand, were

consistent with a pattern of isolation by geographic distance and probably reflect limited

dispersal via host fish species upon which unionoid larvae are obligate parasites.

7.1.1.2. Umbonal sculpture

In contrast to the extreme phenotypic plasticity in, for example, shell ‘size’ and

‘shape’, the morphological appearance of ‘umbonal sculptures’ within a species is much

more conservative. As a result, ‘umbonal sculpture’ is one of the few shell features used in

modern phylogenetic analyses of the Unionoida (e.g. Hoeh, Bogan & Heard, 2001; Graf &

Cummings, 2006b). What is more, these sculptures are one of the few features available

for taxonomic classification of both modern and fossil taxa, which in turn is the basis for

improving our surprisingly poor knowledge on early unionoid evolution. Unfortunately,

these fields of research suffer from our lack of understanding of homoplasies and

homologies of umbonal sculpture types across the unionoid phylogeny. Thus, previously

suggested models of character evolution of beak sculptures (Simpson, 1900; Ortmann,

1912; Modell, 1942) are contradictory and based on outdated assumptions on unionoid

evolution and sculpture formation.

Based on examination of over 150 modern and extinct species, Chapter 5 presented a

new model of character evolution of umbonal sculptures in the Unionoida. Challenging all

previous models of character evolution, my observations indicated that the plesiomorphic

characters state of unionoid beak sculpture is ‘V-shaped’. ‘Pseudo-radial’ sculpture

probably arose in one line of development, whereas ‘W-shaped’ and ultimately ‘double-

looped’ sculpture was developed in other unionoid groups. All other forms of umbonal

sculpture present in this bivalve group represent derivates of one of these four main types.

The study confirmed the suitability of umbonal sculptures for reconstructing

phylogenies and classifying unknown specimens to higher unionoid taxonomic levels.

Furthermore, examination of the distribution of beak sculpture types across the unionoid

phylogeny revealed new insights into unresolved questions on freshwater mussel

evolution. For example, presence of ‘V-shaped’ and ‘pseudo-radial’ sculpture in the

Hyriidae and some incertae sedis unionid genera (e.g. Coelatura and Parreysia) suggested

a basal position of these taxa within the Unionoida. A basal position of the Quadrulini


126

within the unionid subfamily Ambleminae was supported by the ‘W-shaped’ umbonal

sculpture of many quadruline species examined. On the other hand, I showed that some

beak sculpture types, such as ‘nodulous’, ‘wrinkled’ and ‘smooth’, have evolved

independently numerous times within the Unionoida and thus, represent phylogenetically

uninformative, convergent character states.

7.1.1.3. Periostracal microprojections

In Chapter 6, the phylogenetic significance of a further, previously unrecognised

shell character was investigated for the first time in the Unionoida and their presumed

closest living relatives, the Trigonioida. Such periostracal microprojections had been

described in other bivalve groups (e.g. Aller, 1974; Glover & Taylor, 2010; Checa &

Harper, in press).

Investigation of shell surfaces of specimens covering all extant palaeoheterodont

(unionoid + trigonioid) families using scanning electron microscopy, revealed three types

of periostracal microprojections; microridges, microfringes and spicule-like spikes.

Examination of the distribution of these three distinct morphologies across the

palaeoheterodont phylogeny indicated a considerable value for the reconstruction of

bivalve evolution. Thus, microridges were present only in one respective species of the

two unionoid families Mycetopodidae (Anodontites trapesialis (Lamarck, 1819)) and

Iridinidae (Chambardia bourguignati (Bourguignat, 1885)), and may represent a

synapomorphy for the mycetopodid-iridinid clade. In A. trapesialis, microridges were

additionally equipped with flag-like microfringes, which are possibly a synapomorphic

character for the Mycetopodidae.

Calcified spikes were widespread across the unionoid phylogeny but were not found

in any hyriid and mycetopodid species examined. Given the relatively well established,

derived position of the Mycetopodidae within the Unionoida, it is likely that spike

production has secondarily been lost or switched off in this family. Absence of spikes in

the Hyriidae, on the other hand, could represent the plesiomorphic unionoid character

state. As in the case of beak sculptures (7.1.1.2), this would suggest a basal position of the

Hyriidae within the Unionoida.

7.1.2. ECOLOGY, CONSERVATION AND ENVIRONMENTAL RECONSTRUCTION

Most (palaeo)environmental reconstructions using unionoid shells have so far been

based on relative abundance data of species with known habitat requirements (i.e. species


127

assemblages) (e.g. Matteson, 1960; Morey & Crothers, 1998; Radley & Barker, 1998).

Naturally, these exclude most endangered and extinct taxa, restricting utility of such

assemblage analyses to the reconstruction of only relatively recent palaeoenvironments.

Knowledge on inter- and intraspecific trends in shell morphologies that are reliable

indicators of certain environmental conditions and exhibited across a wide taxonomic

range can overcome this problem. At the species and higher taxonomic levels, this can be

attained by identification of unifying characteristics in habitat requirements of unrelated

taxa exhibiting convergent morphologies. Currently, our knowledge in this respect is

restricted to only a small number of unionoid shell morphologies, in particular wing

development (i.e. symphynote species usually indicate soft sediment habitats) and adult

shell sculpture (i.e. well sculptured species indicate fast-flowing habitats) (Watters, 1994).

Information about the environment can also be obtained using intraspecific trends in

morphology that are consistently associated with changes in habitat condition. Probably

the most widely used unionoid shell character in this respect has so far been maximum or

average ‘shell size’ of populations, which is generally regarded to be positively correlated

with temperature and/or food availability (e.g. Grier, 1920; Mann, 1965). Additionally, the

extreme variability in shell form of many unionoid species could be particularly useful for

redrawing characteristics of the flow or sediment conditions of the habitat. Unfortunately,

reconstruction of environments and habitat requirements of species on the basis of

intraspecific patterns in unionoid shell shape has so far been limited by the lack of

identification of consistent ecophenotypic trends that are not considerably confused by

non-habitat factors such as sexual dimorphism or ontogenetic growth (Tevesz & Carter,

1980; Claassen, 1998).

7.1.2.1. “Gross shell morphology”: size, shape, inflation, thickness, density and adductor scar sizes

Partially confirming observations by Grier (1920) and Mann (1965), investigation of

morphological patterns within three unionid species from two habitat types (marinas and

river) of the River Thames, UK, in Chapter 2 showed that larger maximum shell sizes (and

probably also relatively large sizes of adductor muscle scars) can be indicative of warmer

temperature and/or greater phytoplankton abundances. What is more, the study elucidated

an intraspecific ecophenotypic trend in shell form that was consistently exhibited across

replicate populations and species. This trend from dorsally arched river specimens to

straight dorsal and pointed posterior margins in marina individuals was always associated


128

with differences in the hydrological condition of the habitat and could have an adaptive

value of stabilising the mussel’s life position in fast flowing habitats. Similar observations

by Eagar (1948, 1971, 1974, 1977, 1978) on margaritiferid unionoids and Carboniferous

anthracosiid species suggest that this ecophenotypic trend is present across a wide range of

freshwater mussel taxa. Intraspecific differences in the shape of the dorso-posterior margin

of unionoid shells could thus be used, for example, in the determination of habitat

requirements of endangered species and reconstruction of flow regimes of ancient

freshwater habitats.

Relative ‘elongation’ and ‘wing development’ were two further shell characters that

showed significant habitat-associated trends. However, these patterns were restricted only

to single species, limiting their potential utility for environmental reconstructions to only

these species (or possibly genera) where a particular trend has been elucidated.

Shell ‘inflation’ should generally be regarded as a poor indicator of habitat since it

was shown to be considerably influenced by allometric growth, sexual dimorphism and

trematode parasites (Chapter 2 and 4). These observations not only help explain

contradictory observations in ecophenotypic trends in this character by previous authors

(e.g. Ortmann, 1920) but also demonstrate the importance of considering non-habitat

factors when attempting to elucidate truly ecophenotypic trends.

Statistically significant differences between populations in two further shell

characters, i.e. ‘thickness’ and ‘density’ of the shell, was demonstrated in Chapter 4

(4.4.4). However, which habitat conditions are associated with these changes in shell

morphology and the potential value of these characters for environmental reconstructions

remains unclear.

7.1.2.2. Umbonal sculpture

The new model of character evolution of unionoid rugae developed in Chapter 5

elucidated interspecific convergences in sculpture types. For example, though species with

relatively faint or completely ‘smooth’ umbos were found in several unrelated taxa, almost

all of them were characteristic of lentic habitats. Species with well developed rugae, and

particularly of the ‘wrinkled’ or ‘nodulous’ type, on the other hand, were typically

inhabitants of turbulent habitats. These three types of umbonal sculptures could thus be

used to for assigning a species to a typically standing or flowing habitat.

Provided a good understanding of their function(s) is available, beak sculptures

could ultimately be used for determination of habitat requirements of juveniles of different


129

unionoid species. Such knowledge could be of particular value for developing optimal

habitat recreation measures or conditions in artificial propagation of endangered

freshwater mussels, thereby increasing the usually extremely low rate of survival in the

first months of the postparasitic life stage (Wächtler, Mansur & Richter, 2001). However,

though my observations indicated that umbonal rugae probably serve in increasing

reinforcement and/or anchoring of the juvenile shell in the sediment, these functional

hypotheses remain to be tested in experiments.

7.1.2.3. Periostracal microprojections

Particularly large and abundant periostracal spikes were observed in species

characteristic of fast flowing, coarse sediment habitats and possibly serve to protect the

periostracum and underlying shell, and/or to stabilise life position by increasing shell

friction (Chapter 6). Average spike size and abundance of a given unionoid shell of

unknown origin may thus be used to redraw information about flow and/or sediment

conditions of the habitat it once inhabited.

Microfringes and microridges possibly aid in the orientation of the mussel within the

sediment, but information on the patterns of these microsculptures across different habitats

currently remains too sparse to enable any ecophenotypic generalisations to be made.

7.1.3. DETERMINATION OF SEX AND TREMATODE LOADS

7.1.3.1. “Gross shell morphology”: size, shape, inflation and thickness

Intraspecific patterns in some shell features were considerably affected by other

factors than the habitat, rendering such characters unsuitable for application in

environmental reconstructions. Such morphological trends may, on the other hand, help in

the reconstruction of other characteristics of an individual, population or species. This is

particularly true for changes in unionoid shell morphology that are induced by sexual

dimorphism and parasitic infestation. Apart from the well known differences in sagittal

shell shape between male and female lampsilines (e.g. Kotrla & James, 1987), both these

types of intraspecific shell dimorphism are currently poorly understood in the Unionoida.

Chapter 4, which investigated five A. anatina populations, represents the first study

revealing statistically significant sexual and parasite-induced dimorphic trends in the shells

of a European unionid species.


130

More specifically, differences between sexes of A. anatina were observed in shell

‘inflation’ and ‘thickness’, and probably are a result of altered female shell growth due to

accommodation of marsupial gills and resource depletion by offspring production,

respectively. My results further indicated that the extent of morphological differences

between sexes was determined by the environmental conditions to which the respective

population is subjected. This observation helps explain the so far inconclusive reports on

sexual shell dimorphism in the unionid subfamily Unioninae and answers a long debated

question; why some populations are apparently dimorphic and others are not. In general,

sexual dimorphic trends were more pronounced in populations from “high quality” habitats

compared to those living at less favourable conditions. Thus, only in populations of high

reproductive output could differences in shell width and/or thickness between sexes be a

useful tool, providing rapid sex determination in the field or enabling determination of a

fossil population’s sex ratio.

The same study also showed for the first time that trematode parasites can

significantly alter both inflation and sagittal shape of freshwater mussel shells. The

underlying reasons and mechanisms involved in these parasite-induced changes in shell

morphology are currently unclear. Nevertheless, provided a better understanding of

trematode-mussel interactions, this or similar dimorphic patterns could potentially become

a useful tool for detecting parasite infections, that can lead to castration or even death of

unionoids (e.g. Jokela, Uotila & Taskinen, 1993; Gangloff, Lenertz & Feminella, 2008;

Grizzle & Brunner, 2009). As such, monitoring for changes in morphology within

threatened populations of unionoids may provide important indications of changes in

harmful parasite burdens.

7.2. FUTURE DIRECTIONS

This thesis provides several new insights into the suitability of 12 unionoid shell

characters for phylogenetic and environmental reconstruction, and determination of sex

and trematode loads. In doing so, it also raises many new questions which are worthy of

further investigation.


131

7.2.1. TESTING FOR CONSISTENCY OF THE PATTERNS OBSERVED

Investigations on intraspecific ecophenotypic trends and the degree of phenotypic

plasticity in shell morphology were based on populations of just one river catchment area,

and involved only three and one common unionoid species, respectively. In order to

establish the generality of what was found in these studies, similar experiments on other

unionoid species involving populations of various river catchments and lake habitats are

needed. The distribution of genetic diversity of a given mussel species/population would,

for example, be expected to be strongly dependent on the mobility of the fish species it is

parasitising. This could be tested by looking at several unionoid species using different

host fish species.

The dimorphic patterns in shell morphology observed and discussed in Chapter 4

were based merely on a small number of populations of only a single unionoid species.

Further studies involving populations from a wide range of habitats and additional species

are recommended in order to test if the hypotheses developed here hold true. If and to what

extent the degree of sexual dimorphisms is influenced by the “quality” of the habitat and

thus, mussel fecundity at the site, could be further investigated on the basis of

measurements of glochidia production per female.

Although a large number of species were included in the analysis of umbonal

sculptures in Chapter 5, some unionoid families were underrepresented or, in the case of

the Etheriidae, completely omitted due to a lack of available material. Consideration of

additional material could help test the robustness of my model and could help to illuminate

the taxonomy of fossil taxa.

The main purpose of Chapter 6 was to provide an initial overview on periostracal

microprojections present in the palaeoheterodont families. Further studies are needed to

confirm or reject the synapomorphies and other hypotheses proposed in this study.

7.2.2. OPEN QUESTIONS

One of the main objectives of this thesis has been the identification of consistent

ecophenotypic trends in unionoid shell features. The next big challenge in this respect will

be to test the adaptive significance of the patterns observed and the functional hypotheses

developed in this thesis. This will require an experimental approach. For example, the

adaptive significance of the intraspecific ecomorphotypes of marina and river habitats of

the River Thames could be investigated by means of reciprocal transplant experiments.


132

Flume experiments could be used to test the hypothesis that arched “river” shells are able

to withstand higher current velocities than “marina” morphotypes. Equally, laboratory

experiments on juvenile unionoids could investigate the functional significance of

unionoid umbonal sculptures.

To what extent sex and parasites influence unionoid shell morphology remains

poorly understood. This is particularly so for most Afrotropical and Indotropical unionid

species, for which we often lack even the most basic biological data, such as the time of

brooding or which parts of the female gills are used as marsupia. With respect to European

unionoids, it would be interesting to test if and to what extent other freshwater mollusc

parasites, such as parasitic fish (e.g. Bitterling (Rhodeus spp.)) or mites (Unionicolidae),

alter unionoid shell growth.

As discussed in Chapter 6 (6.5.3), investigation of the formation process of

trigonioid bosses and unionoid spikes is needed to resolve the question if these structures

represent a common ability to periostracal calcification in the two palaeoheterodont orders.

Study of intraspecific variation in both unionoid microprojections and umbonal sculpture

including specimens from a range of habitats will possibly elucidate ecophenotypic trends

useful for reconstruction of habitat requirements and environments of juvenile mussels.

7.2.3. APPLICATION OF METHODS TO NON-UNIONOID TAXA

While this thesis has focused on the use of new tools to better understand

morphology in unionoid mussels, the approach could also provide new and important

insights into other taxa. Many of the patterns observed in my studies may also hold true for

other freshwater and marine bivalves, while a closer inspection of aquatic gastropods is

likely to reveal consistent intra- and interspecific patterns in shell shape which can be

attributed to genotype, habitat type, sex and parasite load. Indeed, the marine limpets

(Patella spp.) and periwinkles (Littorina spp.) provide classic examples of consistent

ecophenotypes that have been linked with trade-offs between factors such as reproductive

capacity, risk of displacement by wave action, and vulnerability to predation (e.g. Vermeij,

1973; Johannesson, 2003). Revisiting these examples using more powerful descriptive and

analytical tools may generate an enhanced understanding in these already intensively

studied systems. Moreover, as tools continue to become increasingly powerful (e.g. 3D-

imaging), we can expect to continue to gain an increasingly greater understanding of the

factors that determine morphology of a particular organism.

APPENDIX

135

APPENDIX

Table A.1. Umbonal sculpture types of unionoid taxa examined from shell material, photographs and/or drawings. Additionally, the type, source and locality of the material are given. Systematics after Graf & Cummings (2007b). Abbreviations: Umbonal sculpture type: c, pseudo-concentric (bars); d, double-looped; n, nodulous/pustulous; r, pseudo-radial; s, single-looped; sm, smooth; v, V-shaped; w, W-shaped; wr, wrinkled/corrugated. Source: CUMZ, Cambridge University Museum of Zoology; KIZ, Kunming Institute of Zoology (China). Locality: Geographical subregion(s) of species/genus according to Graf & Cummings (2007b): AF, Afrotropica; AU, Australasia; IN, Indotropica; NA, Nearctica; NT, Neotropica; PA, Palearctica.

Taxon Umbonal sculpture type Type of

material examined

Source Locality v r w d n wr c s sm

Superfam. UNIONOIDEA Family UNIONIDAE

Subfam. UNIONINAE Tribe UNIONINI

Acuticosta chinensis (Lea, 1868) x sketch Modell (1942) unknown (IN) Cafferia caffra (Krauss, 1848) x shell CUMZ unknown (AF) Cafferia caffra (Krauss, 1848) x x sketch Modell (1942) unknown (AF) Cuneopsis celtiformis (Heude, 1874) x sketch Modell (1942) unknown (IN) Cuneopsis pisciculus (Heude, 1874) sketch Modell (1942) unknown (IN)

Nodularia douglasiae (Griffith & Pidgeon, 1834) x x photograph Savazzi & Peiyi (1992) Poyang Lake, China (PA,IN)

Nodularia douglasiae (Griffith & Pidgeon, 1834) x drawing Haas (1969b) unknown (PA,IN) Nodularia douglasiae (Griffith & Pidgeon, 1834) x x x sketch Modell (1942) unknown (PA,IN) Nodularia schoedei (Haas, 1930) x sketch Modell (1942) unknown (IN) Rhombuniopsis sp. x drawing Haas (1969b) unknown (IN) Unio delphinus* Spengler, 1793 x photograph Araujo et al. (2009) Iberian Peninsula Unio gibbus* Spengler, 1793 x photograph Araujo et al. (2009) Iberian Peninsula Unio mancus Lamarck, 1819 x x x sketch Modell (1942) Italy (PA) Unio pictorum (Linnaeus, 1758) x shell private collection Rv. Thames, UK (PA)

* species status not recognised by Graf & Cummings (2007b)

136

TAXON v r w d n wr c s sm Material Source Locality

Unio pictorum (Linnaeus, 1758) photograph Killeen, Aldridge & Oliver (2004) UK (PA)

Unio ravoisieri* Deshayes, 1847 x x photograph Araujo et al. (2009) Iberian Peninsula Unio terminalis Bourguignat, 1852 drawing Germain (1922) Syria (PA) Unio terminalis Bourguignat, 1852 x x sketch Modell (1942) unknown (PA) Unio tigridis Bourguignat, 1852 x sketch Modell (1942) unknown (PA) Unio tumidiformis* Castro, 1885 x photograph Araujo et al. (2009) Iberian Peninsula Unio tumidus Philipsson, 1788 x x x shell private collection Rv. Thames, UK (PA) Unio tumidus Philipsson, 1788 shell CUMZ unknown (PA)

Unio tumidus Philipsson, 1788 photograph Killeen, Aldridge & Oliver (2004) UK (PA)

Unio tumidus Philipsson, 1788 photograph Killeen, Aldridge & Oliver (2004) UK (PA)

Tribe ANODONTINI

Alasmidonta arcula (Lea, 1838) drawing Clarke (1981) Altamaha Rv., USA (NA)

Alasmidonta heterodon (Lea, 1829) drawing Clarke (1981) Connecticut Rv., USA (NA)

Alasmidonta marginata Say, 1818 x shell CUMZ Canada (NA)

Alasmidonta marginata Say, 1818 photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Alasmidonta marginata Say, 1818 drawing Clarke (1981) West Okoboji Lake, USA (NA)

Alasmidonta marginata Say, 1818 x drawing Marshall (1890) Huron Rv., USA (NA) Alasmidonta marginata Say, 1818 x sketch Modell (1942) unknown (NA)

Alasmidonta undulata (Say, 1817) drawing Clarke (1981) Monocacy Rv., USA (NA)

Alasmidonta undulata (Say, 1817) x sketch Modell (1942) unknown (NA)

Alasmidonta varicosa (Lamarck, 1819) drawing Clarke (1981) Massachusetts, USA (NA)

Alasmidonta viridis (Rafinesque, 1820) x x shell CUMZ USA (NA) * species status not recognised by Graf & Cummings (2007b)

137

TAXON v r w d n wr c s sm Material Source Locality Alasmidonta viridis (Rafinesque, 1820) photograph Klocek, Bland &

Barghusen (2010) Chicago, USA (NA)

Alasmidonta viridis (Rafinesque, 1820) drawing Clarke (1981) Illinois, USA (NA) Alasmidonta viridis (Rafinesque, 1820) x x x sketch Modell (1942) unknown (NA) Anodonta anatina (Linnaeus, 1758) x x x shell private collection Rv. Thames, UK (PA) Anodonta anatina (Linnaeus, 1758) shell private collection Rv. Great Ouse, UK

(PA) Anodonta anatina (Linnaeus, 1758) photograph Killeen, Aldridge & Oliver

(2004) UK (PA)

Anodonta anatina (Linnaeus, 1758) photograph Araujo et al. (2009) Iberian Peninsula (PA) Anodonta cygnea (Linnaeus, 1758) x shell private collection Rv. Great Ouse, UK

(PA) Anodonta cygnea (Linnaeus, 1758) photograph Killeen, Aldridge & Oliver

(2004) UK (PA)

Anodonta kennerlyi Lea, 1860 x x drawing Clarke (1973) Canada (NA) Anodontoides ferussacianus (Lea, 1834) x shell CUMZ USA (NA) Anodontoides ferussacianus (Lea, 1834) photograph Klocek, Bland &

Barghusen (2010) Chicago, USA (NA)

Anodontoides ferussacianus (Lea, 1834) x photograph Grabarkiewicz & Todd (2010)

Rv. Maumee, USA (NA)

Anodontoides ferussacianus (Lea, 1834) x drawing Marshall (1890) Albany County, USA (NA)

Arcidens confragosus (Say, 1829) drawing Clarke (1981) Mississippi Rv., USA (NA)

Cristaria plicata (Leach, 1815) x shell CUMZ Shanghai, China (PA,IN)

Cristaria plicata (Leach, 1815) x photograph Savazzi & Peiyi (1992) Poyang Lake, China (PA,IN)

Lasmigona complanata (Barnes, 1823) x shell CUMZ USA (NA) Lasmigona complanata (Barnes, 1823) x photograph Grabarkiewicz & Todd

(2010) Rv. Maumee, USA (NA)

Lasmigona complanata (Barnes, 1823) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Lasmigona complanata (Barnes, 1823) x drawing Clarke (1985) Ohio, USA (NA)

138

TAXON v r w d n wr c s sm Material Source Locality Lasmigona compressa (Lea, 1829) x x shell CUMZ USA (NA)

Lasmigona compressa (Lea, 1829) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Lasmigona compressa (Lea, 1829) x x drawing Clarke (1985) Iowa, USA (NA) Lasmigona costata (Rafinesque, 1820) x x shell CUMZ USA (NA)

Lasmigona costata (Rafinesque, 1820) x photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Lasmigona costata (Rafinesque, 1820) x photograph Grabarkiewicz & Todd (2010)


Lasmigona costata (Rafinesque, 1820) x x drawing Marshall (1890) USA (NA) Lasmigona costata (Rafinesque, 1820) x x drawing Clarke (1985) Ohio, USA (NA) Lasmigona decorata (Lea, 1852) x drawing Clarke (1985) N. Carolina, USA (NA) Lasmigona holstonia (Lea, 1838) x drawing Clarke (1985) Tennessee, USA (NA) Lasmigona subviridis (Conrad, 1835) x drawing Clarke (1985) N. Carolina, USA (NA) Lasmigona subviridis (Conrad, 1835) x drawing Marshall (1890) Erie Canal, USA (NA) Pegias fabula (Lea, 1838) x x shell CUMZ USA (NA) Pseudanodonta complanata (Rossmässler, 1835) x x shell private collection UK (PA)

Pseudanodonta complanata (Rossmässler, 1835) photograph Killeen, Aldridge & Oliver (2004) UK (PA)

Pseudanodonta complanata (Rossmässler, 1835) photograph Killeen, Aldridge & Oliver (2004) UK (PA)

Pyganodon cataracta (Say, 1817) x x drawing Clarke (1973) Canada (NA) Pyganodon implicata (Say, 1829) x drawing Marshall (1890) USA (NA) Pyganodon grandis (Say, 1829) x shell CUMZ USA (NA,NT) Pyganodon grandis (Say, 1829) x x shell private collection unknown (NA,NT)

Pyganodon grandis (Say, 1829) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Pyganodon grandis (Say, 1829) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Pyganodon grandis (Say, 1829) x drawing Clarke (1973) USA (NA,NT) Pyganodon lacustris* (Lea, 1852) x photograph private collection USA


139

TAXON v r w d n wr c s sm Material Source Locality Simpsonaias ambigua (Say, 1825) x x shell CUMZ USA (NA)

Simpsonaias ambigua (Say, 1825) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Simpsonaias ambigua (Say, 1825) x sketch Modell (1942) unknown (NA) Sinanodonta woodiana (Lea, 1834) x shell CUMZ China (PA,IN) Sinanodonta woodiana (Lea, 1834) x photograph KIZ China (PA,IN)

Sinanodonta woodiana (Lea, 1834) photograph Killeen, Aldridge & Oliver (2004) UK (PA)

Sinanodonta woodiana (Lea, 1834) x photograph Savazzi & Peiyi (1992) Poyang Lake, China (PA,IN)

Sinanodonta woodiana (Lea, 1834) x sketch Modell (1942) unknown (PA,IN) Strophitus undulatus (Say, 1817) x shell CUMZ (NA)

Strophitus undulatus (Say, 1817) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Strophitus undulatus (Say, 1817) x drawing Marshall (1890) Albany County, USA (NA)

Utterbackia imbecilis (Say, 1829) x shell CUMZ USA (NA)

Utterbackia imbecilis (Say, 1829) x photograph Grabarkiewicz & Todd (2010)


Utterbackia imbecilis (Say, 1829) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Utterbackia imbecilis (Say, 1829) x x drawing Coker et al. (1921) USA (NA)

Subfam. AMBLEMINAE Tribe QUADRULINI

Amphinaias pustulosa (Lea, 1831) x x shell CUMZ USA (NA)

Megalonaias nervosa (Rafinesque, 1820) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Quadrula quadrula (Rafinesque, 1820) x shell CUMZ USA (NA)

Quadrula quadrula (Rafinesque, 1820) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Quadrula quadrula (Rafinesque, 1820) x x sketch Modell (1942) unknown (NA) Theliderma cylindrica (Say, 1817) x x x x shell CUMZ USA (NA)

140

TAXON v r w d n wr c s sm Material Source Locality Tritogonia verrucosa (Rafinesque, 1820) x x x x shell CUMZ Ohio, USA (NA)

Tribe PLEUROBEMINI

Cyclonaias tuberculata (Rafinesque, 1820) x shell CUMZ Rv. Huron, USA (NA)

Cyclonaias tuberculata (Rafinesque, 1820) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Elliptio arctata (Conrad, 1834) x shell CUMZ USA (NA) Elliptio buckleyi (Lea, 1841) x sketch Modell (1942) unknown (NA)

Elliptio complanata (Lightfoot, 1786) x x drawing Marshall (1890) Champlain Canal, USA (NA)

Elliptio dilatata (Rafinesque, 1820) x x shell CUMZ USA (NA)

Elliptio dilatata (Rafinesque, 1820) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Fusconaia flava (Rafinesque, 1820) x x x shell CUMZ USA (NA) Pleurobema sintoxia (Rafinesque, 1820) x x shell CUMZ USA (NA)

Pleurobema sintoxia (Rafinesque, 1820) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Uniomerus tetralasmus (Say, 1831) x shell CUMZ USA (NA)

Uniomerus tetralasmus (Say, 1831) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Uniomerus tetralasmus (Say, 1831) x sketch Modell (1942) unknown (NA)

Tribe AMBLEMINI

Amblema plicata (Say, 1817) x shell CUMZ USA (NA)

Amblema plicata (Say, 1817) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Amblema plicata (Say, 1817) x x sketch Modell (1942) unknown (NA)

Tribe LAMPSILINI

Actinonaias ligamentina (Lamarck, 1819) x photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Actinonaias pectorosa (Conrad, 1834) x shell private collection Clinch Rv., USA (NA) Delphinonaias delphinulus (Morelet, 1849) x shell CUMZ Yucatan (NT) Epioblasma brevidens (Lea, 1831) x shell private collection unknown, USA (NA)

141

TAXON v r w d n wr c s sm Material Source Locality Epioblasma torulosa (Rafinesque, 1820) x shell CUMZ USA (NA)

Lampsilis cardium Rafinesque, 1820 photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Lampsilis fasciola Rafinesque, 1820 x shell CUMZ USA (NA) Lampsilis fasciola Rafinesque, 1820 x x shell private collection USA (NA)

Lampsilis ovata (Say, 1817) x shell private collection Virginia Tech Hatchery, USA (NA)

Lampsilis radiata (Gmelin, 1791) x photograph Grabarkiewicz & Todd (2010)


Lampsilis siliquoidea (Barnes, 1823) x shell CUMZ USA (NA)

Lampsilis siliquoidea (Barnes, 1823) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Lampsilis teres (Rafinesque, 1820) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Lampsilis teres (Rafinesque, 1820) x sketch Modell (1942) unknown (NA)

Ligumia nasuta (Say, 1817) x drawing Marshall (1890) Canal, NY State, USA (NA)

Ligumia recta (Lamarck, 1819) x sketch Modell (1942) unknown (NA) Ligumia subrostrata (Say, 1831) x shell CUMZ USA (NA) Obliquaria reflexa Rafinesque, 1820 drawing Coker et al. (1921) USA (NA) Ptychobranchus fasciolaris (Rafinesque, 1820) x x sketch Modell (1942) unknown (NA) Toxolasma cylindrellus (Lea, 1868) x shell CUMZ USA (NA) Toxolasma parvus (Barnes, 1823) x shell CUMZ USA (NA) Toxolasma paulus (Lea, 1840) x photograph Frank & Lee (1998) Florida, USA (NA) Truncilla donaciformis (Lea, 1828) x x shell CUMZ Ohio, USA (NA)

Truncilla donaciformis (Lea, 1828) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Venustaconcha ellipsiformis (Conrad, 1836) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Villosa fabalis (Lea, 1831) x photograph Grabarkiewicz & Todd (2010)


Villosa iris (Lea, 1829) shell private collection Virginia Tech Hatchery, USA (NA)

142


Villosa iris (Lea, 1829) x shell CUMZ Rv. Shiawassee, USA (NA)

Villosa iris (Lea, 1829) x shell CUMZ Tennessee, USA (NA) Villosa iris (Lea, 1829) x shell CUMZ USA (NA)

Villosa iris (Lea, 1829) photograph Klocek, Bland & Barghusen (2010) Chicago, USA (NA)

Villosa taeniata (Conrad, 1834) x shell CUMZ USA (NA)

incertae sedis UNIONIDAE Western Palearctic

Leguminaia saulcyi (Bourguignat, 1852) x shell CUMZ Rv. Tigris (PA) Leguminaia wheatleyi (Lea, 1862) x shell CUMZ Rv. Tigris (PA) Potomida littoralis (Cuvier, 1798) x x photograph Araujo et al. (2009) Iberian Peninsula (PA) Potomida littoralis (Cuvier, 1798) drawing Germain (1922) Syria (PA)

incertae sedis UNIONIDAE Afrotropical

Brazzaea sp. x drawing Haas (1969b) unknown (AF) Coelatura aegyptiaca (Cailliaud, 1827) x x shell CUMZ Rv. Nile (AF) Coelatura aegyptiaca (Cailliaud, 1827) x shell CUMZ “Suez” (AF) Coelatura aegyptiaca (Cailliaud, 1827) photograph Graf & Cummings (2010b) unknown (AF) Coelatura aegyptiaca (Cailliaud, 1827) x sketch Modell (1942) unknown (AF) Coelatura bakeri (Adams, 1866) x sketch Modell (1942) unknown (AF) Coelatura bakeri (Adams, 1866) x photograph Pilsbry & Bequaert (1927) Lake Albert (AF)

Coelatura bakeri (Adams, 1866) x photograph Scholz & Glaubrecht (2004) Lake Albert (AF)

Coelatura bakeri (Adams, 1866) x x drawing von Martens (1897) Lake Albert (AF) Coelatura hauttecoeuri (Bourguignat, 1883) x photograph Graf & Cummings (2007a) northern Africa (AF) Coelatura hauttecoeuri (Bourguignat, 1883) x sketch Modell (1942) unknown (AF)

Coelatura hauttecoeuri (Bourguignat, 1883) x photograph Scholz & Glaubrecht (2004) Africa (AF)

Coelatura horei (Smith, 1880) x x shell private collection Lake Tanganyika, Tanzania (AF)

143


Coelatura horei (Smith, 1880) x photograph Pilsbry & Bequaert (1927) Lake Tanganyika, Tanzania (AF)

Coelatura hypsiprymna (von Martens, 1897) x shell CUMZ Lake Malawi, Malawi (AF)

Coelatura leopoldvillensis (Putzeys, 1898) x shell private collection Lake Nymagena, Tanzania (AF)

Coelatura leopoldvillensis (Putzeys, 1898) x photograph Graf & Cummings (2006a) Congo River Estuary, Angola (AF)

Coelatura lobensis (Frierson, 1913) x x drawing Haas (1969b) unknown (AF)

Coelatura ratidota (Charmes, 1885) x photograph Scholz & Glaubrecht (2004) Tanzania (AF)

Coelatura stuhlmanni (von Martens, 1897) x photograph Pilsbry & Bequaert (1927) Lake Edward (AF)

Coelatura stuhlmanni (von Martens, 1897) x photograph Scholz & Glaubrecht (2004) Lake Edward (AF)

Coelatura stuhlmanni (von Martens, 1897) x drawing von Martens (1897) Lake Albert (AF) Grandidiera burtoni (Woodward, 1859) x x sketch Modell (1942) unknown (AF) Grandidiera burtoni (Woodward, 1859) x x drawing Haas (1969b) unknown (AF) Nitia acuminata (Adams, 1866) x photograph Pilsbry & Bequaert (1927) Lake Albert (AF) Nitia teretiuscula (Philippi, 1847) x shell CUMZ Rv. Nile, Egypt (AF)

Nyassunio nyassaensis (Lea, 1864) x photograph Scholz & Glaubrecht (2004)

Lake Tanganyika, Tanzania (AF)

Nyassunio ujijiensis (Crosse, 1881) x photograph Scholz & Glaubrecht (2004)

Lake Tanganyika, Tanzania (AF)

Pseudospatha tanganyicensis (Smith, 1880) x shell private collection Lake Tanganyika, Tanzania (AF)

Pseudospatha tanganyicensis (Smith, 1880) x photograph Pilsbry & Bequaert (1927) Lake Tanganyika, Tanzania (AF)

incertae sedis UNIONIDAE Indotropical (& Eastern Palearctic)

Caudiculatus caudiculatus (von Martens, 1866) x sketch Modell (1942) unknown (IN) Contradens sp. x drawing Haas (1969b) unknown (IN) Elongaria orientalis (Lea, 1840) x x sketch Modell (1942) unknown (IN) Harmandia castelneaui* Rochebrune, 1882 x photograph Graf & Cummings (2010b) Cochinchina, Vietnam


144


Harmandia somboriensis Rochebrune, 1882 x photograph Graf & Cummings (2010b) Cochinchina, Vietnam (IN)

Hyriopsis cumingii (Lea, 1852) x shell CUMZ China (IN) Hyriopsis schlegelii (von Martens, 1861) x sketch Modell (1942) unknown (PA) Inversidens japanensis (Lea, 1859) x x x sketch Modell (1942) unknown (PA) Lamellidens marginalis (Lamarck, 1819) x shell CUMZ India (IN)

Lamellidens marginalis (Lamarck, 1819) x shell private collection Tangua Haor, Bangladesh (IN)

Lamellidens marginalis (Lamarck, 1819) x sketch Modell (1942) unknown (IN) Lamprotula leai (Gray, 1834) x x sketch Modell (1942) unknown (IN) Parreysia caerulea (Lea, 1831) x x shell CUMZ Calcutta (IN) Parreysia caerulea (Lea, 1831) x x x shell CUMZ India (IN)

Parreysia caerulea (Lea, 1831) x shell private collection Hakaluki Haor, Bangladesh (IN)

Parreysia caerulea (Lea, 1831) x x photograph Fischer (2007) Rv. Ganges, India (IN) Parreysia corbis (Hanley, 1856) x x shell CUMZ unknown (IN) Parreysia corrugata (Müller, 1774) x x shell CUMZ India (IN) Parreysia corrugata (Müller, 1774) x shell CUMZ Bengal (IN) Parreysia corrugata (Müller, 1774) x shell private collection Bangladesh (IN) Parreysia corrugata (Müller, 1774) x sketch Modell (1942) unknown (IN) Parreysia favidens (Benson, 1862) x x x photograph Fischer (2007) Rv. Ganges, India (IN) Parreysia lima (Simpson, 1900) x x photograph Fischer (2007) Rv. Ganges, India (IN) Parreysia occata (Lea, 1860) x shell CUMZ unknown (IN) Parreysia occata (Lea, 1860) x photograph Fischer (2007) Rv. Ganges, India (IN) Parreysia rajahensis (Lea, 1841) x x x shell CUMZ India (IN) Parreysia smaragdites (Benson, 1862) x x shell CUMZ unknown (IN) Parreysia tavoyensis (Gould, 1843) x x shell CUMZ unknown (IN) Parreysia triembola (Benson, 1855) x shell CUMZ India (IN) Physunio superbus (Lea, 1841) x shell CUMZ Celebes, Indonesia (IN) Physunio superbus (Lea, 1841) x x x sketch Modell (1942) unknown (IN) Pilsbryoconcha compressa (von Martens, 1860) x x x shell CUMZ unknown (IN)

145

TAXON v r w d n wr c s sm Material Source Locality Pilsbryoconcha exilis (Lea, 1838) x sketch Modell (1942) unknown (IN) Pressidens exanthematicus (Küster, 1861) x sketch Modell (1942) unknown (IN) Protunio messageri (Bavay & Dautzenberg, 1901) x x sketch Modell (1942) unknown (IN) Pseudodon inoscularis (Gould, 1844) x x sketch Modell (1942) unknown (IN) Pseudodon omiensis (von Heimburg, 1884) x sketch Modell (1942) unknown (PA) Pseudodon vondembuschianus (Lea, 1840) x x sketch Modell (1942) unknown (IN) Ptychorhynchus murinum (Heude, 1883) x sketch Modell (1942) unknown (IN) Scabies crispata (Gould, 1843) x x shell CUMZ India (IN) Scabies crispata (Gould, 1843) x x shell CUMZ India (IN) Scabies crispata (Gould, 1843) photograph KIZ China (IN) Scabies crispata (Gould, 1843) x x sketch Modell (1942) unknown (IN) Scabies crispata (Gould, 1843) x drawing Haas (1969b) unknown (IN) Scabies nucleus (Lea, 1856) x x x shell CUMZ China (IN) Trapezoideus exolescens (Gould, 1843) x sketch Modell (1942) unknown (IN) Unionetta sp. x photograph KIZ China (IN)

Family MARGARITIFERIDAE

Margaritifera margaritifera (Linnaeus, 1758) x photograph Mary Sollows Canada (NA,PA)

Margaritifera margaritifera (Linnaeus, 1758) x photograph Kennard, Salisbury & Woodward (1925) UK (NA,PA)

Margaritifera margaritifera (Linnaeus, 1758) Killeen, Aldridge & Oliver (2004) UK (NA,PA)

Margaritifera margaritifera (Linnaeus, 1758) x x sketch Modell (1942) unknown (NA,PA) Margaritifera monodonta (Say, 1829) x sketch Modell (1942) unknown (NA)

Superfam. ETHERIOIDEA Family HYRIIDAE

Subfam. HYRIINAE Tribe HYRIINI

Prisodon corrugatus (Lamarck, 1819) x shell CUMZ Brazil (NT)

Prisodon corrugatus (Lamarck, 1819) x photograph Graf & Cummings (2010b) Brazil (NT)

146


Prisodon corrugatus (Lamarck, 1819) x photograph Pimpão, Rocha & de Castro Fettuccia (2008) Catalão, Brazil (NT)

Prisodon corrugatus (Lamarck, 1819) photograph Kohl (2010) Amapa State, Brazil (NT)

Prisodon corrugatus (Lamarck, 1819) x drawing Haas (1969b) unknown (NT)

Prisodon (Triplodon) chodo* x x photograph Mansur & Pimpão (2008) Amazonas Basin, Brazil

Triplodon stevensii* (Lea, 1871) photograph Cummings & Mayer

(2009) Yuruari Rv., Venezuela

Triplodon sp. x drawing Haas (1969b) unknown (NT)

Tribe CASTALIINI

Castalia ambigua Lamarck, 1819 x shell CUMZ Brazil (NT)

Castalia ambigua Lamarck, 1819 x x photograph Graf & Cummings (2010b) Rio Yaquerana, Peru (NT)

Castalia ambigua Lamarck, 1819 photograph Kohl (2010) Paraguay (NT)

Castalia stevensi* (Baker, 1930) photograph Cummings & Mayer (2009) Rio Yuruari, Venezuela

Castalia psammoica (d’Orbigny, 1835) x photograph Kohl (2010) Argentina (NT) Castalia sp. x drawing Haas (1969b) unknown (NT)

Tribe RHIPIDODONTINI

Diplodon chilensis (Gray, 1828) x shell CUMZ Chile (NT) Diplodon chilensis (Gray, 1828) x x sketch Modell (1942) unknown (NT)

Diplodon deceptus* (Simpson, 1914) photograph Mansur & Pereira (2006) Sinos River Basin, Brazil

Diplodon flucki Morrison, 1943 x photograph Cummings & Mayer (2009)

Rio Siapa, Venezuela (NT)

Diplodon fluctiger (Lea, 1859) x sketch Modell (1942) unknown (NT) Diplodon multistriatus (Lea, 1831) x x shell CUMZ Brazil (NT) Diplodon rhuacoicus (d’Orbigny, 1835) x photograph Kohl (2010) Uruguay (NT) Diplodon rhuacoicus (d’Orbigny, 1835)

x sketch Modell (1942) unknown (NT)


147


Tribe HYRIDELLINI

Echyridella menziesii (Gray, 1843) x shell CUMZ New Zealand (AU) Echyridella menziesii (Gray, 1843) x photograph Graf & Cummings (2010b) unknown (AU)

Hyridella australis (Lamarck, 1819) x photograph McMichael & Hiscock (1958) Australia (AU)

Hyridella drapeta (Iredale, 1934) x photograph McMichael & Hiscock (1958) Australia (AU)

Hyridella glenelgensis (Dennant, 1898) x photograph Playford & Walker (2008) various (AU)

Hyridella narracanensis (Cotton & Gabriel, 1932) x photograph McMichael & Hiscock (1958) Australia (AU)

Virgus beccarianus (Tapparone Canefri, 1883) x photograph McMichael & Hiscock (1958) Australia (AU)

Virgus beccarianus (Tapparone Canefri, 1883) x sketch Modell (1942) unknown (AU)

Family MYCETOPODIDAE Subfam. ANODONTITINAE

Anodontites trapesialis (Lamarck, 1819) shell pivate collection São Francisco River, Brazil (NT)

Family IRIDINIDAE Subfam. IRIDININAE

Chelidonopsis hirundo (von Martens, 1881) x photograph Pilsbry & Bequaert (1927) Kisangani, DR Congo (AF)

Mutela dubia (Gmelin, 1791) x photograph Pilsbry & Bequaert (1927) Aba River, Nigeria (AF)

Mutela nilotica* (Cailliaud, 1823) x photograph Pilsbry & Bequaert (1927) Lake Albert Mutela rostrata (Rang, 1835) x shell CUMZ W-Africa (AF) Pleiodon ovata (Swainson, 1823) x shell CUMZ Senegal (AF) Pleiodon spekii (Woodward, 1859) x shell private collection Lake Tanganyika,

Tanzania (AF) Subfam. ASPATHARIINAE

Aspatharia chaiziana (Rang, 1835) x photograph Haas (1929) DR Congo (AF) Aspatharia divaricata (von Martens, 1897) x photograph Graf & Cummings (2007a) northern Africa (AF)


148

TAXON v r w d n wr c s sm Material Source Locality Aspatharia pfeifferiana (Bernardi, 1860) x photograph Pilsbry & Bequaert (1927) Stanleyville (AF) Aspatharia rugifera (Dunker, 1858) x sketch Modell (1942) unknown (AF) Chambardia bourguignati (Bourguignat, 1885) x shell private collection Tanzania (AF) Chambardia nyassaensis (Lea, 1864) x shell CUMZ Africa (AF) Chambardia wahlbergi (Krauss, 1848) x sketch Modell (1942) unknown (AF)

BIBLIOGRAPHY

151

BIBLIOGRAPHY

Agrawal, H.P. 1974. Sexuality of relative growth in the freshwater mussel, Parreysia wynegungaensis (Lea) (Pelecypoda: Unionidae). Indian Journal of Zoology 2: 29-31.

Agrell, I. 1948. The shell morphology of some Swedish unionides as affected by ecological conditions. Arkiv för Zoologi 41A: 1-30.

Ajmone-Marsan, J., Valentini, A., Cassandro, M., Vecchiotti-Antaldi, G., Bertoni, G. & Kuiper, M. 1997. AFLPTM markers for DNA fingerprinting in cattle. Animal Genetics 82: 418-426.

Aldridge, D.C. 1997. Reproductive ecology of bitterling (Rhodeus sericeus Pallas) and unionid mussels. PhD thesis, University of Cambridge, UK. 154 pp.

Aldridge, D.C. 1999a. Development of European bitterling in the gills of freshwater mussels. Journal of Fish Biology 54: 138–151.

Aldridge, D.C. 1999b. The morphology, growth and reproduction of Unionidae (Bivalvia) in a fenland waterway. Journal of Molluscan Studies 65: 47-60.

Aldridge, D.C. 2004. Conservation of freshwater unionid mussels in Britain. Journal of Conchology Special Publication 3: 81-90.

Aldridge, D.C., Fayle, T. & Jackson, N. 2007. Freshwater mussel abundance predicts biodiversity in UK lowland rivers. Aquatic Conservation: Marine and Freshwater Ecosystems 17: 554-564.

Aller, R.C. 1974. Prefabrication of shell ornamentation in the bivalve Laternula. Lethaia 7: 43-56.

Altnöder, K. 1926. Beobachtungen über die Biologie von Margaritana margaritifera und über die Ökologie ihres Wohnorts. Archiv für Hydrobiologie 17: 423-447.

Amyot, J.-P. & Downing, J.A. 1991. Endo- and epibenthic distribution of the unionid mollusc Elliptio complanata. Journal of the North American Benthological Society 10: 280-285.

Amyot, J.-P. & Downing, J.A. 1997. Seasonal variation in vertical and horizontal movement of the freshwater bivalve Elliptio complanata (Mollusca: Bivalvia). Freshwater Biology 37: 345-354.

Anderson, R.V. & Ingham, R.E. 1978. Character variation in mollusc populations of Lampsilis Rafinesque, 1820. Transactions of the Illinois State Academy of Science 71: 403-411.

Andersson, M. 1994. Sexual selection. Princeton University Press: Princeton, NJ. 624 pp.


152

Anthony, J.L. & Downing, J.A. 2001. Exploitation trajectory of a declining fauna: a century of freshwater mussel fisheries in North America. Canadian Journal of Fisheries and Aquatic Sciences 58: 2071-2090.

Araujo, R., Reis, J., Machordom, A., Toledo, C., Madeira, M.J., Gómez, I., Velasco, J.C., Morales, J., Barea, J.M., Ondina, P. & Ayala, I. 2009. Las náyades de la península Ibérica. Iberus 27: 7-72.

Bailey, R.C. & Green, R.H. 1988. Within-basin variation in the shell morphology and growth rate of a freshwater mussel. Canadian Journal of Zoology 66: 1704-1708.

Baker, A.M., Bartlett, C., Bunn, S.E., Goudkamp, K., Sheldon, F. & Hughes, J.M. 2003. Cryptic species and morphological plasticity in long-lived bivalves (Unionoida : Hyriidae) from inland Australia. Molecular Ecology 12: 2707-2717.

Balfour, D.L. & Smock, L.A. 1995. Distribution, age structure, and movement of the freshwater mussel Elliptio complanata (Mollusca: Unionidae) in a headwater stream. Journal of Freshwater Ecology 10: 225-268.

Ball, G.H. 1922. Variation in freshwater mussels. Ecology 3: 93-121.

Balla, S.A. & Walker, K.F. 1991. Shape variation in the Australian freshwater mussel Alathyria jacksoni Iredale (Bivalvia, Hyriidae). Hydrobiologia 220: 89-98.

Bauer, G. 1998. Allocation policy of female freshwater mussels. Oecologia 117: 90-94.

Bauer, G. 2001. Life history variation on different taxonomic levels of naiads. In: Bauer, G. & Wächtler K. (eds.). Ecology and Evolution of the Freshwater Mussels Unionoida. Berlin: Springer-Verlag. pp. 83-91.

Beaumont, A.R. 2008. Dfdist package. Available at http://www.rubic.rdg.ac.uk/~mab/stuff/ (accessed April 2009).

Beaumont, M.A. & Balding, D.J. 2004. Identifying adaptive genetic divergence among populations from genome scans. Molecular Ecology 13: 969–980.

Beaumont, M.A. & Nichols, R.A. 1996. Evaluating loci for use in the genetic analysis of population structure. Proceedings of the Royal Society of London, Series B 263: 1619-1626.

Beedham, G.E. 1958. Observations on the mantle of the Lamellibranchia. Quarterly Journal of Microscopical Science 99: 181-197.

Bensch, S. & Åkesson, M. 2005. Ten years of AFLP in ecology and evolution: Why so few animals? Molecular Ecology 14: 2899-2914.

Berg, D.J., Cantonwine, E.G., Hoeh, W.R. & Guttman, S.I. 1998. Genetic structure of Quadrula quadrula (Bivalva: Unionidae): little variation across large distances. Journal of Shellfish Research 17: 1365-1373.

Bergek, S. & Björklund, M. 2007. Cryptic barriers to dispersal within a lake allow genetic differentiation of Eurasian perch. Evolution 61: 2035-2041.

Bibliography

153

Berrie, A.D. & Bioze, B.J. 1985. The fish hosts of Unio glochidia in the River Thames. Verhandlungen der Internationalen Vereinigung für theoretische und angewandte Limnologie 22: 2712-2716.

Bieler, R. & Mikkelsen, P.M. 1996. Bivalvia – a look at the Branches. Zoological Journal of the Linnean Society 148: 223–235.

Blažek, R. & Gelnar, M. 2006. Temporal and spatial distribution of glochidial larval stages of European unionid mussels (Mollusca: Unionidae) on host fishes. Folia Parasitologica 53: 98–106.

Blears, M.J., De Grandis, S.A., Lee, H. & Trevors, J.T. 1998. Amplified fragment length polymorphism (AFLP): a review of the procedure and its applications. Journal of Industrial Microbiology & Biotechnology 21: 99-114.

Bloomer, H.H. 1934. On the sex, and sex-modification of the gill, of Anodonta cygnea. Proceedings of the Malacological Society of London 21: 20-29.

Bloomer, H.H. 1937. On distinguishing the shell of Anodonta cygnea from that of Anodonta anatina. Journal of Conchology 20: 321-327.

Bloomer, H.H. 1938. The British species of Anodonta Lamarck, and their varieties. Journal of Conchology 21: 33-48.

Bogan, A.E. 1993. Freshwater bivalve extinctions (Mollusca: Unionoida): A search for causes. American Zoologist 33: 599-609.

Bogan, A.E. 2008. Global diversity of freshwater mussels (Mollusca, Bivalvia) in freshwater. Hydrobiologia 595: 139–147.

Bogan, A.E. & Hoeh, W.R. 2000. On becoming cemented: evolutionary relationships among the genera in the freshwater bivalve family Etheriidae (Bivalvia: Unionoida). In: Harper, E.M., Taylor J.D. & Crame J.A. (eds.). The evolutionary biology of the Bivalvia. London: The Geological Society of London. pp. 31-46.

Bogan, A.E. & Roe, K. 2008. Freshwater bivalve (Unioniformes) diversity, systematics, and evolution: status and future directions. Journal of the North American Benthological Society 27: 349-369.

Bolnick, D.I., Snowberg, L.K., Patenia, C., Stutz, W.E., Ingram, T. & Lau, O.L. 2009. Phenotype-dependent native habitat preference facilitates divergence between parapatric lake and stream stickleback. Evolution 63: 2004-2016.

Bonetto, A.A. 1966. Especies de la subfamilia Monocondylaeinae en las aguas del sistema del Rio de La Plata (Moll. Mutelacea). Archiv für Molluskenkunde 95: 3-14.

Bonin, A., Bellemain, E., Bronken Eidesen, P., Pompanon, F., Brochmann, C. & Taberlet, P. 2004. How to track and assess genotyping errors in population genetic studies. Molecular Ecology 13: 3261-3273.


154

Bonin, A., Ehrich, D. & Manel, S. 2007. Statistical analysis of amplified fragment length polymorphism data: a toolbox for molecular ecologists and evolutionists. Molecular Ecology 16: 3737-3758.

Bottjer, D.J. & Carter, J.G. 1980. Functional and phylogenetic significance of projecting periostracal structures in the Bivalvia (Mollusca). Journal of Paleontology 54: 200-216.

Bouza, C., Castro, J., Martínez, P., Amaro, R., Fernández, C., Ondina, P., Outeiro, A. & San Miguel, E. 2007. Threatened freshwater pearl mussel Margaritifera margaritifera L. in NW Spain: low and very structured genetic variation in southern peripheral populations assessed using microsatellite markers. Conservation Genetics 8: 937-948.

Brander, T. 1954. Über Geschlechtsdimorphismus bei europäischen Unionazeen. Archiv für Molluskenkunde 83: 163-172.

Branson, C.C. 1935. Fresh-water invertebrates from the Morrison (Jurassic?) of Wyoming. Journal of Paleontology 9: 514-522.

Brecko, J., Huyghe, K., Vanhooydonck, B., Herrel, A., Grbac, I. & Van Damme, R. 2008. Functional and ecological relevance of intraspecific variation in body size and shape in the lizard Podarcis melisellensis (Lacertidae). Biological Journal of the Linnean Society 94: 251-264.

Bridgeman, J.B. 1875. A variety caused by a locality. Quarterly Journal of Conchology 1: 70.

Buchner, O. 1900. Beiträge zur Formenkenntnis der einheimischen Anodonten unter besonderer Berücksichtigung der württembergischen Vorkommnisse. Jahreshefte des Vereins für vaterländische Naturkunde in Württemberg 56: 60-223.

Buchner, O. 1910. Ueber individuelle Formverschiedenheiten bei Anodonten. Jahreshefte des Vereins für vaterländische Naturkunde in Württemberg 65: 50.

Buhay, J.E., Serb, J.M., Dean, C.R., Parham, Q. & Lydeard, C. 2002. Conservation genetics of two endangered unionid bivalve species, Epioblasma florentina walkeri and E. capsaeformis (Unionidae: Lampsilini). Journal of Molluscan Studies 68: 385-391.

Caballero, A., Quesada, H. & Rolán-Alvarez, E. 2008. Impact of amplified fragment length polymorphism size homoplasy on the estimation of population genetic diversity and the detection of selective loci. Genetics 179: 539-554.

Callil, C.T. & Mansur, M.C.D. 2005. Ultrastructural analysis of the shells of Anodontites trapesialis (Lamarck) and Anodontites elongatus (Swaison) (Mollusca, Bivalvia, Etherioidea) from the Mato Gross Pantanal Region, Brazil. Revista Brasileira de Zoologia 22: 724-734.

Cameron, C.J., Cameron, I.F. & Paterson, C.G. 1979. Contribution of organic shell matter to biomass estimates of unionid bivalves. Canadian Journal of Zoology 57: 1666-1669.

Bibliography

155

Campbell, D.C., Serb, J.M., Buhay, J.E., Roe, K.J., Minton, R.L. & Lydeard, C. 2005. Phylogeny of North American amblemines (Bivalvia, Unionoida): prodigious polyphyly proves pervasive across genera. Invertebrate Biology 124: 131-164.

Carter, J.G. 1978. Ecology and evolution of the Gastrochaenae (Mollusca, Bivalvia) with notes on the evolution of the endolithic habitat. Peabody Museum Bulletin 41: 1-92.

Carter, J.G. 1990. Evolutionary significance of shell microstructures in the Palaeotaxodonta, Pteriomorpha and Isofilibranchia (Bivalvia: Mollusca). In: Carter, J.G. (ed.) Skeletal biomineralization: patterns, processes and evolutionary trends. New York: Van Nostrand Reinhold. pp. 136–271.

Carter, J.G. & Aller, R.C. 1975. Calcification in the bivalve periostracum. Lethaia 8: 315-320.

Carter, J.G., Campbell, D.C. & Campbell, M.R. 2006. Morphological phylogenetics of the early Bivalvia. In Malchus, N. & Pons J.M., (eds.). Organic Diversity and Evolution Supplement. Abstracts and Posters of the “International Congress on Bivalvia” at the Universitat Autònoma de Barcelona, Spain, 22-27 July 2006, 20-21.

Checa, A.G. 2000. A new model for periostracum and shell formation in Unionidae (Bivalvia, Mollusca). Tissue & Cell 32: 405-416.

Checa, A.G. 2002. Fabricational morphology of oblique ribs in bivalves. Journal of Morphology 254: 195-209.

Checa, A.G. & Harper, E.M. in press. Spikey bivalves: intra-periostracal crystal growth in anomalodesmatans. Biological Bulletin.

Checa, A.G. & Jiménez-Jiménez, A.P. 2003a. Evolutionary morphology of oblique ribs of bivalves. Palaeontology 46: 709-724.

Checa, A.G. & Jiménez-Jiménez, A.P. 2003b. Rib fabrication in Ostreoidea and Plicatuloidea (Bivalvia, Pteriomorphia) and its evolutionary significance. Zoomorphology 122: 145-159.

Checa, A.G. & Rodríguez-Navarro, A. 2001. Geometrical and crystallographic constraints determine the self-organization of shell microstructures in Unionidae (Bivalvia: Mollusca). Proceedings of the Royal Society of London, Series B 268: 771-778.

Claassen, C. 1998. Shells. Cambridge University Press: Cambridge. 266 pp.

Clarke, A.H. 1973. The fresh water mollusks of the Canadian Interior Basin. Malacologia 13: 1-509.

Clarke, A.H. 1981. The tribe Alasmidontini (Unionidae: Anodontinae), part I. Pegias, Alasmidonta, and Arcidens. Smithsonian Contributions to Zoology 326: 1-101.

Clarke, A.H. 1985. The tribe Alasmidontini (Unionidae: Anodontinae), part II. Lasmigona and Simpsonaias. Smithsonian Contributions to Zoology 399: 1-75.


156

Clarke, A.H. 1986. The mesoconch: a record of juvenile life in Unionidae. Malacology Data Net 1: 21-36.

Coker, R.E., Shira, A.F., Clark, H.W. & Howard, A.D. 1921. Natural history and propagation of fresh-water mussels. Bulletin of the Bureau of Fisheries (U.S.) 37: 75-181.

Conde-Padín, P., Carvajal-Rodríguez, A., Carballo, M., Caballero, A. & Rolán-Alvarez, E. 2007. Genetic variation for shell traits in a direct-developing marine snail involved in a putative sympatric ecological speciation process. Evolutionary Ecology 21: 635-650.

Connolly, M. 1939. A monographic study of South African non-marine Mollusca. Annals of the South African Museum 33: 1-660.

Cooper, M.R. 1991. Lower Cretaceous Trigonioida (Mollusca, Bivalvia) from the Algoa Basin, with a revised classification of the order. Annals of the South African Museum 100: 1-52.

Cox, L.R. 1969. Superfamily Trigoniacea. In: Moore, R.C. (ed.) Treatise on invertebrate paleontology. Lawrence, Kansas: Geological Society of America & University of Kansas Press. pp. N471-N488.

Crampton, J.S. & Haines, A.J. 1996. Users' manual for programs HANGLE, HMATCH and HCURVE for the Fourier shape analysis of two-dimensional outlines Institute of Geological and Nuclear Sciences Science Report. 1-28.

Crampton, J.S. & Haines, A.J. 2007. Programs HANGLE, HMATCH and HCURVE for the Fourier shape analysis of two-dimensional outlines.

Crawley, M.J. 2002. Statistical computing, an introduction to data analysis using S-plus. John Wiley and Sons Ltd: Chichester.

Crosby, M.P. & Gale, L.D. 1990. A review of bivalve condition index methodologies with a suggested standard method. Journal of Shellfish Research 9: 233-239.

Cummings, K.S. & Mayer, C.A. 2009. The Freshwater Mussels (Bivalvia: Etherioidea) of Venezuela. Available at http://www.inhs.illinois.edu/animals_plants/mollusk/SA/Ven.html (accessed 2010).

Cvancara, A.M. 1972. Lake mussel distribution as determined with SCUBA. Ecology 53: 154-157.

Dasmahapatra, K.K., Hoffman, J.I. & Amos, W. 2009. Pinniped phylogenetic relationships inferred using AFLP markers. Heredity 103: 168-177.

Davis, G.M. 1983. Relative roles of molecular genetics, anatomy, morphometrics and ecology in assessing relationships among North American Unionidae (Bivalvia). In: Oxford, G.S. & Rollinson D. (eds.). Protein polymorphism: adaptive and taxonomic significance. London and New York: Academic Press. pp. 193-222.

Bibliography

157

Davis, G.M. 1984. Genetic relationships among some North American Unionidae (Bivalvia): Sibling species, convergence, and cladistic relationships. Malacologia 25: 629-648.

Davis, G.M. & Fuller, S.L.H. 1981. Genetic relationships among recent Unionacea (Bivalvia) of North America. Malacologia 20: 217-253.

Davis, G.M., Heard, W.H., Fuller, S.L.H. & Hesterman, C. 1981. Molecular genetics and speciation in Elliptio and its relationships to other taxa of North American Unionidae (Bivalvia). Biological Journal of the Linnean Society 15: 131-150.

Davis, G.M. & Ruff, M. 1974. Oncomelania hupensis (Gastropoda: Hydrobiidae): hybridization, genetics, and transmission of Schistosoma japonicum. Malacological Review 6: 181-197.

de Aranzamendi, M.C., Sahade, R., Tatian, M. & Chiappero, M.B. 2008. Genetic differentiation between morphotypes in the Antarctic limpet Nacella concinna as revealed by inter-simple sequence repeat markers. Marine Biology 154: 875-885.

de Wolf, H., Backeljau, T. & Verhagen, R. 1998. Spatio-temporal genetic structure and gene flow between two distinct shell morphs of the planktonic developing periwinkle Littorina striata (Mollusca: Prosobranchia). Marine Ecology progress series 163: 155–163.

Dell, R.K. 1953. The freshwater Mollusca of New Zealand. Transactions of the Royal Society of New Zealand 81: 221-237.

Doebeli, M. & Dieckmann, U. 2003. Speciation along environmental gradients. Nature 421: 259-264.

Dolmen, D. & von Proschwitz, T. 1999. Searching for the swan mussel Anodonta cygnea in Norway: And the use of characters to distinguish it from the duck mussel A. anatina. Fauna 52: 82-88.

Downing, J.A., Amyot, J.-P., Pérusse, M. & Rochon, Y. 1989. Viceral sex, hermaphroditism and protandry in a population of the freshwater bivalve Elliptio complanata. Journal of the North American Benthological Society 8: 92-99.

Downing, W.L. & Downing, J.A. 1993. Molluscan shell growth and loss. Nature 362: 506.

Dyer, A.T. & Leonard, K.J. 2000. Contamination, error, and nonspecific molecular tools. Phytopathology 90: 565-567.

Eagar, R.M.C. 1948. Variation in shape of shell with respect to ecological station. A review dealing with Recent Unionidae and certain species of the Anthracosiidae in Upper Carboniferous times. Proceedings of the Royal Society of Edinburgh, Section B 63: 130-148.

Eagar, R.M.C. 1971. Variation in shape of shell in relation to palaeoecological station in some non-marine Bivalvia of the Coal Measures of south-east Kentucky and Britain. Compte rendu te Congres International de Stratigraphie et de Geologie du Carbonifere, Krefeld 2: 387-413.


158

Eagar, R.M.C. 1974. Shape of shell of Carbonicola in relation to burrowing. Lethaia 7: 219-238.

Eagar, R.M.C. 1977. Shape of shell in relation to weight of Margaritifera margaritifera. Journal of Conchology 29: 207-218.

Eagar, R.M.C. 1978. Shape and function of the shell: a comparison of some living and fossil bivalve mussels. Biological Reviews of the Cambridge Philosophical Society 53: 169-210.

Elderkin, C.L., Christian, A.D., Vaughn, C.C., Metcalfe-Smith, J.L. & Berg, D.J. 2007. Population genetics of the freshwater mussel, Amblema plicata (Say 1817) (Bivalvia: Unionidae): Evidence of high dispersal and post-glacial colonization. Conservation Genetics 8: 355-372.

Fischer, W. 2007. The Bivalvia of the Ganga River. Available at http://ipp.boku.ac.at/private/wf/Bivalvia_India/Bivalvia_Ganga_River.html (accessed 2010).

Forsyth, D.J. & McCallum, I.D. 1978. Xenochironomus canterburyensis (Diptera: Chironomidae), an insectan inquiline commensal of Hyridella menziesi (Mollusca: Lamellibranchia). Journal of Zoology 186: 331-334.

Frank, B. & Lee, H.G. 1998. The Jacksonville Shell Club. Available at http://www.jaxshells.org/ (accessed 2010).

Frierson, L.S. 1909a. Notes on Oriental Unionidae. Nautilus 24: 97-98.

Frierson, L.S. 1909b. Remarks on the subfamilies Hyriinae and Unioninae. Nautilus 22: 106-107.

Gangloff, M.M., Lenertz, K.K. & Feminella, J.W. 2008. Parasitic mite and trematode abundance are associated with reduced reproductive output and physiological condition of freshwater mussels. Hydrobiologia 610: 25-31.

Geist, J. & Auerswald, K. 2007. Physicochemical stream bed characteristics and recruitment of the freshwater pearl mussel (Margaritifera margaritifera). Freshwater Biology 52: 2299-2316.

Geist, J. & Kuehn, R. 2005. Genetic diversity and differentiation of central European freshwater pearl mussel (Margaritifera margaritifera L.) populations: implications for conservation and management. Molecular Ecology 14: 425-439.

Germain, L. 1922. Mollusques terrestres et fluviatiles de Syrie. Part 2: Pelecypodes. J.-B. Baillere et Fils: Paris.

Ghent, A.W., Singer, R. & Johnson-Singer, L. 1978. Depth distributions determined with SCUBA, and associated studies of the freshwater unionid clams Elliptio complanata and Anodonta grandis in Lake Bernard, Ontario. Canadian Journal of Zoology 56: 1654-1663.

Bibliography

159

Giribet, G. & Wheeler, W. 2002. On bivalve phylogeny: a high-level analysis of the Bivalvia (Mollusca) based on combined morphology and DNA sequence data. Invertebrate Biology 121: 271-324.

Glover, E.A. & Taylor, J.D. 2010. Needles and pins: acicular crystalline periostracal calcification in venerid bivalves (Bivalvia: Veneridae). Journal of Molluscan Studies 76: 157-179.

Good, S.C. 1989. Nonmarine mollusca in the Upper Triassic Chinle Formation and related strata of the western Interior: systematics and distribution. In: Lucas, S.G. & Hunt A.P. (eds.). Dawn of the Age of Dinosaurs in the American South-west. Albuquerque, New Mexico: New Mexico Museum of Natural History. pp. 233-248.

Good, S.C. 1998. Freshwater bivalve fauna of the Late Triassic (Carnian-Norian) Chinle, Dockum, and Dolores Formations of the Southwest United States. In: Johnston, P.A. & Haggart J.W. (eds.). Bivalves: an eon of evolution - palaeobiological studies honoring Norman D. Newell. Calgary: University of Calgary Press.

Good, S.C. 2004. Paleoenvironmental and paleoclimatic significance of freshwater bivalves in the Upper Jurassic Morrison Formation, Western Interior, USA. Sedimentary Geology 167: 163-176.

Goodwin, B.J. & Fish, J.D. 1977. Inter- and intraspecific variation in Littorina obtusata and L. mariae (Gastropoda: Prosobranchia). Journal of Molluscan Studies 43: 241-254.

Grabarkiewicz, J. & Todd, C. 2010. Freshwater Mussels of the Maumee Drainage. Available at http://www.farmertodd.com/musselguide/ (accessed 2010).

Graf, D.L. 2000. The Etherioidea revisited: A phylogenetic analysis of hyriid relationships (Mollusca: Bivalvia: Paleoheterodonta: Unionoida). Occasional Papers of the Museum of Zoology University of Michigan: 1-21.

Graf, D.L. & Cummings, K.S. 2006a. Freshwater mussels (Mollusca: Bivalvia: Unionoida) of Angola, with description of a new species, Mutela wistarmorrisi. Proceedings of the Academy of Natural Sciences of Philadelphia 155: 163-194.

Graf, D.L. & Cummings, K.S. 2006b. Palaeoheterodont diversity (Mollusca: Trigonioida + Unionoida): what we know and what we wish we knew about freshwater mussel evolution. Zoological Journal of the Linnean Society 148: 343-394.

Graf, D.L. & Cummings, K.S. 2007a. Preliminary review of the freshwater mussels (Mollusca: Bivalvia: Unionoida) of northern Africa, with emphasis on the Nile. Journal of the Egyptian German Society for Zoology 53D: 89-118.

Graf, D.L. & Cummings, K.S. 2007b. Review of the systematics and global diversity of freshwater mussel species (Bivalvia: Unionoida). Journal of Molluscan Studies 73: 291-314.


160

Graf, D.L. & Cummings, K.S. 2010a. Comments on the value of COI for family-level freshwater Mussel systematics: a reply to Hoeh, Bogan, Heard & Chapman. Malacologia 52: 191-197.

Graf, D.L. & Cummings, K.S. 2010b. The MUSSEL Project — Home Page. Available at http://mussel-project.ua.edu (accessed 2010).

Graf, D.L. & Foighil, D.O. 2000. The evolution of brooding characters among the freshwater pearly mussels (Bivalvia : Unionoidea) of North America. Journal of Molluscan Studies 66: 157-170.

Gray, J.E. 1825. Conchological observations, being an attempt to fix the study of conchology on a firm basis. Journal of Zoology 1: 204-223.

Grier, N.M. 1920. Morphological features of certain mussel-shells found in Lake Erie, compared with those of the corresponding species found in the drainage of the Upper Ohio. Annales of the Carnegie Museum 13: 145-182.

Grier, N.M. & Mueller, J.F. 1926. Further studies in correlation of shape and station in fresh water mussels. Bulletin of the Wagner Free Institute of Science 1: 11-28.

Grizzle, J.M. & Brunner, C.J. 2009. Infectious Diseases of Freshwater Mussels and Other Freshwater Bivalve Mollusks. Reviews in Fisheries Science 17: 425-467.

Grobler, P.J., Jones, J.W., Johnson, N.A., Beaty, B., Struthers, J., Neves, R.J. & Hallerman, E.M. 2006. Patterns of genetic differentiation and conservation of the Slabside Pearlymussel, Lexingtonia dolabelloides (Lea, 1840) in the Tennessee drainage. Journal of Molluscan Studies 72: 65-75.

Gu, Z. 1998. Evolutionary trends in non-marine Cretaceous bivalves of northeast China. In: Johnston, P.A. & Haggart J.W. (eds.). Bivalves: an eon of evolution - palaeobiological studies honoring Norman D. Newell. Calgary: University of Calgary Press. pp. 267-276.

Guerra-Varela, J., Colson, I., Backeljau, T., Breugelmans, K., Hughes, R.N. & Rolán-Alvarez, E. 2009. The evolutionary mechanism maintaining shell shape and molecular differentiation between two ecotypes of the dogwhelk Nucella lapillus. Evolutionary Ecology 23: 261-280.

Guo, F. 1998a. Origin and phylogeny of the Trigonioidoidea (Non-Marine Cretaceous Bivalves). In: Johnston, P.A. & Haggart J.W. (eds.). Bivalves: an eon of evolution - palaeobiological studies honoring Norman D. Newell. Calgary: University of Calgary Press. pp. 277-289.

Guo, F. 1998b. Sinonaiinae, a new Subfamily of Asian Non-Marine Cretaceous Bivalves. In: Johnston, P.A. & Haggart J.W. (eds.). Bivalves: an eon of evolution - palaeobiological studies honoring Norman D. Newell. Calgary: University of Calgary Press. pp. 291-294.

Haag, W.R. & Stanton, J.L. 2003. Variation in fecundity and other reproductive traits in freshwater mussels. Freshwater Biology 48: 2118-2130.

Bibliography

161

Haas, F. 1929. Die von der Zweiten Deutschen Zentral-Afrika-Expedition 1910-11 mitgebrachten Süßwassermuscheln. Senckenbergiana 11: 110-116.

Haas, F. 1969a. Superfamilia Unionacea. In: Mertens, R. & Henning W. (eds.). Das Tierreich (Berlin). Berlin: de Gruyter & Co. pp. 1-663.

Haas, F. 1969b. Superfamily Unionacea. In: Moore, R.C. (ed.) Treatise on invertebrate paleontology. Lawrence, Kansas: Geological Society of America & University of Kansas Press. pp. N411-N470.

Haas, F. & Schwarz, E. 1913. Die Unioniden des Gebietes zwischen Main und deutscher Donau in tiergeographischer Hinsicht. Abhandlungen der Königlich Bayerischen Akademie der Wissenschaften, mathematisch-physikalische Klasse 26 1-34.

Haines, A.J. & Crampton, J.S. 2000. Improvements to the method of Fourier shape analysis as applied in morphometric studies. Palaeontology 43: 765-783.

Hamai, I. 2004. Sexuality of relative growth in the freshwater mussel, Inversidens japonensis (Lea). Science reports of Tohoku Imperial University (Biology) 167: 163-176.

Hammer, Ø. & Harper, D.A.T. 2006. PAST version 1.57. Available at http://folk.uio.no/ohammer/past/ (accessed April 2009).

Hammer, Ø., Harper, D.A.T. & Ryan, P.D. 2001. PAST: Paleontological statistics software package for education and data analysis. Palaeontologica Electronica 4: 9 pp.

Harper, E.M. 1997. The molluscan periostracum: an important constraint in bivalve evolution. Palaeontology 40: 71-97.

Haskin, H.H. 1954. Age determination in molluscs. Transactions of the New York Academy of Sciences 16: 300-304.

Hayashi, K. 1935. On the convexity of the shell of Anodonta. Venus 5: 23-25.

Hazay, J. 1881. Die Mollusken-Fauna von Budapest. Malakozoologische Blätter 4: 132-208.

Healy, J.M. 1989. Spermiogenesis and spermatozoa in the relict bivalve genus Neotrigonia: relevance to trigonioid relationships, particularly with Unionoidea. Marine Biology 103: 75-85.

Heard, W.H. 1975. Sexuality and other aspects of reproduction in Anodonta (Pelecypoda: Unionidae). Malacologia 15: 81-103.

Heard, W.H. & Guckert, R.H. 1970. A re-evaluation of the recent Unionadea (Pelecypoda) of North America. Malacologia 10: 333-355.

Hey, W.C. 1882. Freshwater mussels in the Ouse and the Fosse. Journal of Conchology 3: 263.


162

Hinch, S.G. & Bailey, R.C. 1988. Within- and among-lake variation in shell morphology of the freshwater clam Elliptio complanata (Bivalvia: Unionidae) from south-central Ontario lakes. Hydrobiologia 157: 27-32.

Hinch, S.G., Bailey, R.C. & Green, R.H. 1986. Growth of Lampsilis radiata (Bivalvia: Unionidae) in sand and mud: a reciprocal transplant experiment. Canadian Journal of Fisheries and Aquatic Sciences 43: 548-552.

Hinch, S.G. & Green, R.H. 1989. The effects of source and destination on growth and metal uptake in freshwater clams reciprocally transplanted among south central Ontario lakes. Canadian Journal of Zoology 67: 855-863.

Hinch, S.G., Kelly, L.J. & Green, R.H. 1989. Morphological variation of Elliptio complanata (Bivalvia: Unionidae) in differing sediments of soft-water lakes exposed to acidic deposition. Canadian Journal of Zoology 67: 1895-1899.

Hochwald, S. 2001. Plasticity of life-history traits in Unio crassus. In: Bauer, G. & Wächtler K. (eds.). Ecology and Evolution of the Freshwater Mussels Unionoida. Berlin: Springer-Verlag. pp. 127-141.

Hoeh, W.R., Black, M.B., Gustafson, E., Bogan, A.E., Lutz, R.A. & Vrijenhoek, R.C. 1998. Testing alternative hypotheses of Neotrigonia (Bivalvia: Trigonioida) phylogenetic relationships using Cytochrome C Oxidase Subunit I DNA sequences. Malacologia 40: 267-278.

Hoeh, W.R., Bogan, A.E., Cummings, K.S. & Guttman, S.I. 2002. Evolutionary relationships among the higher taxa of freshwater mussels (Bivalvia: Unionoida): inferences on phylogeny and character evolution from analyses of DNA sequence data. Malacological Review 31/32: 117-141.

Hoeh, W.R., Bogan, A.E. & Heard, W.H. 2001. A phylogenetic perspective on the evolution of morphological and reproductive characteristics in the Unionoida. In: Bauer, G. & Wächtler K. (eds.). Ecology and Evolution of the Freshwater Mussels Unionoida. Berlin: Springer-Verlag. pp. 257-280.

Hoeh, W.R., Bogan, A.E., Heard, W.H. & Chapman, E.G. 2009. Palaeoheterodont phylogeny, character evolution, diversity and phylogenetic classification: a reflection on methods of analysis. Malacologia 51: 307-317.

Hoffman, J.I. & Amos, W. 2005. Microsatellite genotyping errors: detection approaches, common sources and consequences for paternal exclusion. Molecular Ecology 14: 599-612.

Hoffman, J.I., Peck, L.S., Hillyard, G., Zieritz, A. & Clark, M.S. 2010. No evidence for genetic differentiation between Antarctic limpet Nacella concinna morphotypes. Marine Biology 157: 765-778.

Hruska, J. 1999. Nahrungsansprueche der Flussperlmuschel und deren halbnatuerliche Aufzucht in der tschechischen Republik. Heldia 4: 69-79.

Huff, S.W., Campbell, D., Gustafson, D.L., Lydeard, C., Altaba, C.R. & Giribet, G. 2004. Investigations into the phylogenetic relationships of freshwater pearl mussels

Bibliography

163

(Bivalvia: Margaritiferidae) based on molecular data: implications for their taxonomy and biogeography. Journal of Molluscan Studies 70: 379-388.

Israel, F.B. 1910. Die Najadeen des Weidagebietes. Nachrichtenblatt der Deutschen Malakozoologischen Gesellschaft 42, Supplement 4: 49-58.

Jass, J. & Glenn, J. 2004. Sexual dimorphism in Lampsilis siliquoidea (Barnes, 1823) (Bivalvia: Unionidae). American Malacological Bulletin 18: 45-47.

Johannessen, O.H. 1973. Population structure and individual growth of Venerupis pullastra (Montagu) (Lamellibranchia). Sarsia 52: 97-116.

Johannesson, B. & Johannesson, K. 1996. Population differences in behaviour and morphology in the snail Littorina saxatilis: Phenotypic plasticity or genetic differentiation? Journal of Zoology 240: 475-493.

Johannesson, K. 2003. Evolution in Littorina: ecology matters. Journal of Sea Research 49: 107-117.

Jokela, J., Uotila, L. & Taskinen, J. 1993. Effects of the castrating trematode parasite Rhipidocotyle fennica on energy allocation of fresh-water clam Anodonta piscinalis. Functional Ecology 7: 332-338.

Källersjö, M., von Proschwitz, T., Lundberg, S., Eldenäs, P. & Erséus, C. 2005. Evaluation of IST rDNA as a complement to mitochondrial gene sequences for phylogenetic studies in freshwater mussels: an example using Unionidae from north-western Europe. Zoologica Scripta 34: 415-424.

Kat, P.W. 1984. Parasitism and the Unionacea (Bivalvia). Biological Reviews of the Cambridge Philosophical Society 59: 189-208.

Kauffman, E.G. 1969. Form, function and evolution. In: Moore, R.C. (ed.) Treatise on invertebrate paleontology. Part N. Mollusca 6. Bivalvia. Lawrence, Kansas: Geological Society of America & University of Kansas Press. pp. N129-205.

Kennard, A.L.S., Salisbury, A.E. & Woodward, B.B. 1925. Notes on the British Post-Pliocene Unionidae, with more especial regard to the means of identification of fossil fragments. Proceedings of the Malacological Society of London 16: 267-285.

Killeen, I., Aldridge, D.C. & Oliver, G. 2004. Freshwater Bivalves of Britain and Ireland. FSC Publications: Shrewsbury.

Kirtland, J.P. 1834. Observations on the sexual characteristics of the animals belonging to Lamarck's family of Naiades. American Journal of Science and Arts 26: 117-220.

Kitano, J., Mori, S. & Peichel, C.L. 2007. Sexual dimorphism in the external morphology of the threespine stickleback (Gasterosteus aculeatus). Copeia 2: 336-349.

Klocek, R., Bland, J. & Barghusen, L. 2010. A Field Guide to the Freshwater Mussels of Chicago Wilderness. Available at www.fieldmuseum.org/chicagoguides (accessed 2010).


164

Kohl, M. 2010. Freshwater Molluscan Shells. Available at http://mkohl1.net/FWshells.html (accessed 2010).

Kotrla, M.B. & James, C.F. 1987. Sexual dimorphism of shell shape and growth of Villosa villosa (Wright) and Elliptio icterina (Conrad) (Bivalvia: Unionidae). Journal of Molluscan Studies 53: 13-23.

Küster, H.C. 1848. Die Flussperlmuscheln (Unio et Hyria). Bauer und Raspe: Nürnberg.

Lea, I. 1870. A synopsis of the family Unionidae. Fourth Edition: Philadelphia, Pennsylvania.

Locard, A. 1890. Contributions á la faune malacologique française. 14. Catalogue des espėces françaises appartenant aix genres Pseudanodonta et Anodonta connues jusqu' á ce jour. Annales de la Société Linnéenne de Lyon 36: 50-285.

Loosanoff, V. 1936. Sexual phases in the quohog. Science 83: 287-288.

Luttikhuizen, P.C., Drent, J., Van Delden, W. & Piersma, T. 2003. Spatially structured genetic variation in a broadcast spawning bivalve: quantitative vs. molecular traits. Journal of Evolutionary Biology 16: 260–272.

Lydeard, C., Cowie, R.H., Ponder, W.F., Bogan, A.E., Bouchet, P., Clark, S.A., Cummings, K.S., Frest, T.J., Gargominy, O., Herbert, D.G., Hershler, R., Perez, K.E., Roth, B., Seddon, M., Strong, E.E. & Thompson, F.G. 2004. The global decline of nonmarine mollusks. BioScience 54: 321-330.

Lydeard, C., Mulvey, M. & Davis, G.M. 1996. Molecular systematics and evolution of reproductive traits of North American freshwater unionacean mussels (Mollusca: Bivalvia) as inferred from 16S rRNA gene sequences. Philosophical Transactions (Series B): Biological Sciences 351: 1593-1603.

Lynch, M. & Milligan, B.G. 1994. Analysis of population genetic structure with RAPD markers. Molecular Ecology 3: 91-99.

Machordom, A., Araujo, R., Erpenbeck, D. & Ramos, M.A. 2003. Phylogeography and conservation genetics of endangered European Margaritiferidae (Bivalvia: Unionoidea). Biological Journal of the Linnean Society 78: 235-252.

Maitland, P.S. & Campbell, R.N. 1992. Freshwater Fishes of the British Isles. Harper Collins: London.

Mann, K.H. 1965. Heated effluents and their effects on the invertebrate fauna of rivers. Proceedings of the Society for Water Treatment and Examination 14: 45-53.

Mansur, M.C.D. & Pereira, D. 2006. Limnic bivalves of the Sinos river basin, Rio Grande do Sul, Brazil (Bivalvia, Unionoida, Veneroida And Mytiloida). Revista Brasileira de Zoologia 23: 1123-1147.

Mansur, M.C.D. & Pimpão, D.M. 2008. Triplodon chodo, a new species of pearly fresh water mussel from the Amazon Basin (Mollusca : Bivalvia : Unionoida : Hyriidae). Revista Brasileira de Zoologia 25: 111-115.

Bibliography

165

March, M.C. 1910-1911. Studies in morphogenesis of certain Pelecypoda. Manchester Memoirs 4: 1-18.

Marie, B., Luquet, G., De Barros, J.-P.P., Guichard, N., Morel, S., Alcaraz, G., Bollache, L. & Marin, F. 2007. The shell matrix of the freshwater mussel Unio pictorum (Paleoheterodonta, Unionoida). FEBS Journal 274: 2933-2945.

Marshall, W.B. 1890. Beaks of Unionidae inhabiting the vicinity of Albany, N.Y. Bulletin of the New York State Museum 2: 169-189.

Marshall, W.B. 1926. Microscopic sculpture of pearly freshwater mussel shells. Proceedings of the United States National Museum 67: 1-14.

Matteson, M.R. 1960. Reconstruction of prehistoric environments through the analysis of molluscan collections from shell middens. American Antiquity 26: 117-120.

McIvor, A.L. & Aldridge, D.C. 2007. The reproductive biology of the depressed river mussel, Pseudanodonta complanata (Bivalvia: Unionidae), with implications for its conservation. Journal of Molluscan Studies 73: 259-266.

McMichael, D.F. & Hiscock, I.D. 1958. A monograph of freshwater mussels (Mollusca: Pelecypoda) of the Australian region. Australian Journal of Marine and Freshwater Research 9: 372-508.

Menard, H.W. & Boucot, A.J. 1951. Experiments on the movement of shells by water. American Journal of Science 249: 131-151.

Miller, N.J., Ciosi, M., Sappington, T.W., Ratcliffe, S.T., Spencer, J.L. & Guillemaud, T. 2007. Genome scan of Diabrotica virgifera virgifera for genetic variation associated with crop rotation tolerance. Journal of Applied Entomology 131: 378–385.

Mitchell, R.D. & Pitchford, G.W. 1953. On mites parasitizing Anodonta in England. Journal of Conchology 23: 365-370.

Modell, H. 1942. Das natürliche System der Najaden. Archiv für Molluskenkunde 74: 161-191.

Modell, H. 1949. Das natürliche System der Najaden. 2. Archiv für Molluskenkunde 78: 29-48.

Modell, H. 1957. Die fossilen Najaden Nordamerikas. Archiv für Molluskenkunde 86: 183-200.

Modell, H. 1964. Das natürliche System der Najaden. 3. Archiv für Molluskenkunde 93: 71-126.

Molina, R.A. 2004. Morphological and genetic description of the freshwater mussel, Elliptio complanata (Lightfoot, 1786) in the Cape Fear River system, NC. PhD thesis, North Carolina State University


166

Morey, D.F. & Crothers, G.M. 1998. Clearing up clouded waters: Palaeoenvironmental analysis of freshwater mussel assemblages from the Green River shell middens, western Kentucky. Journal of Archaeological Science 25: 907-926.

Morton, B. 1976. Secondary brooding of temporary dwarf males in Ephippodonta (Ephippodontina) oedipus sp. nov. (Bivalvia: Leptonacea). Journal of Conchology 29: 31-39.

Mueller, U.G. & Wolfenbarger, L.L. 1999. AFLP genotyping and fingerprinting. Trends in Ecology and Evolution 14: 389-394.

Mulvey, M., Lydeard, C., Pyer, D.L., Hicks, K.M., Brim-Box, J., Williams, J.D. & Butler, R.S. 1997. Conservation genetics of North American freshwater mussels Amblema and Megalonaias. Conservation Biology 11: 868-878.

Mutvei, H. & Westermark, T. 2001. How environmental information can be obtained from naiad shells. In: Bauer, G. & Wächtler K. (eds.). Ecology and Evolution of the Freshwater Mussels Unionoida. Berlin: Springer-Verlag. pp. 367-379.

Nagel, K.O. 2000. Testing hypotheses on the dispersal and evolutionary history of freshwater mussels (Mollusca: Bivalvia: Unionidae). Journal of Evolutionary Biology 13: 854-865.

Negus, C.L. 1966. A quantitative study of growth and reproduction of unionid mussels in the River Thames at Reading. Journal of Animal Ecology 35: 513-532.

Neves, R.J., Bogan, A.E., Williams, J.D., Ahlstedt, S.A. & Hartfield, P.D. 1997. Status of aquatic mollusks in the southeastern United States: a downward spiral of diversity. In: Benz, G.W. & Collins D.E. (eds.). Aquatic Fauna in Peril: The Southeastern Perspective. Special Publication No. 1. Decatur, GA: Southeast Aquatic Research Institute, Lenz Design and Communications. pp. 43–86.

Newell, N.D. & Boyd, D.W. 1975. Parallel evolution in early trigonioidean bivalves. Bulletin of the American Museum of Natural History 154: 55-162.

Noll, F.C. 1869. Unsere Flußmuscheln (Najaden). Ihre Entwicklung und ihre Beziehungen zur übrigen Thierwelt. Bericht über die Senckenbergische naturforschende Gesellschaft in Frankfurt am Main 70: 33-44.

Ohba, S. 1959. Ecological studies in the natural population of a clam, Tapes japonica, with special reference to seasonal variations in the size and structure of the population and to individual growth. Biological Journal of Okayama University 5: 13-42.

Ortmann, A.E. 1912. Notes upon the families and genera of the Najades. Annals of the Carnegie Museum 8: 222-364.

Ortmann, A.E. 1920. Correlation of shape and station in freshwater mussels (Naiades). Proceedings of the American Philosophical Society 59: 268-312.

Owens, I.P.F. & Hartley, I.R. 1998. Sexual dimorphism in birds: why are there so many different forms of dimorphism? Proceedings of the Royal Society B-Biological Sciences 265: 397-407.

Bibliography

167

Panova, M., Hollander, J. & Johannesson, K. 2006. Site-specific genetic divergence in parallel hybrid zones suggests nonallopatric evolution of reproductive barriers. Molecular Ecology 15: 4021-4031.

Parker, G.A. 1992. The evolution of sexual size dimorphism in fish. Journal of Fish Biology, Supplement B 41: 1-20.

Parmalee, P.W. & Bogan, A.E. 1998. The freshwater mussels of Tennessee. The University of Tennessee Press: Knoxville. 328 pp.

Parodiz, J.J. & Bonetto, A.A. 1963. Taxonomy and zoogeographic relationships of the South American naiades (Pelecypoda: Unionacea and Mutelacea). Malacologia 1: 179-213.

Pilsbry, H.A. & Bequaert, J.C. 1927. The aquatic mollusks of the Belgian Congo: with a geographical and ecological account of Congo malacology. Bulletin of the American Museum of Natural History 53: 69-602.

Pimpão, D.M., Rocha, M.S. & de Castro Fettuccia, D. 2008. Freshwater mussels of Catalão, confluence of Solimões and Negro rivers, state of Amazonas, Brazil. Check List 4: 395–400.

Playford, T.J. & Walker, K.F. 2008. Status of the endangered Glenelg River Mussel Hyridella glenelgensis (Unionoida: Hyrudae) in Australia. Aquatic Conservation-Marine and Freshwater Ecosystems 18: 679-691.

Polisky, B., Greene, P., Garfin, D.E., McCarthy, B.J., Goodman, H.M. & Boyer, H.W. 1975. Specificity of substrate recognition by the EcoRI restriction endonuclease. Proceedings of the National Academy of Sciences of the United States of America 72: 3310-3314.

Ponder, W.E. & Bayer, M. 2004. A new species of Lortiella (Mollusca : Bivalvia : Unionoidea : Hyriidae) from northern Australia. Molluscan Research 24: 89-102.

Prashad, B. 1931. Some noteworthy examples of parallel evolution in the molluscan faunas of South-eastern Asia and South America. Proceedings of the Royal Society of Edinburgh 51: 42-53.

Radley, J.D. & Barker, M.J. 1998. Palaeoenvironmental analysis of shell beds in the Wealden Group (Lower Cretaceous) of the Isle of Wight, southern England: an initial account. Cretaceous Research 19: 489-504.

Rasband, W. 2008. ImageJ. Image processing and analysis in Java. Available at http://rsbweb.nih.gov/ij (accessed April 2009).

Reeside, J.B. 1927. Two new unionid pelecyopds from the Upper Triassic. Journal of the Washington Academy of Science 17: 476-478.

Reigle, N.J. 1967. An occurrence of Anodonta (Mollusca, Pelecypoda) in deep water. American Midland Naturalist 78: 530-531.


168

Rodhouse, P.G. 1977. An improved method for measuring volume of bivalves. Aquaculture 11: 279-280.

Roe, K.J. & Hoeh, W.R. 2003. Systematics of freshwater mussels (Bivalvia: Unionoida). In: Lydeard, C. & Lindberg D. (eds.). Molecular Systematics and Phylogeography of Mollusks. Washington: Smithsonian Inst. Press. pp. 91-122.

Roper, D.S. & Hickey, C.W. 1994. Population structure, shell morphology, age and condition of the freshwater mussel Hyridella menziesi (Unionacea: Hyriidae) from seven lake and river sites in the Waikato River system. Hydrobiologia 284: 205-217.

Rossmässler, E.A. 1835-1837. Iconographie der Land- und Süßwasser-Mollusken mit vorzüglicher Berücksichtigung der europäischen noch nicht abgebildeten Arten. Arnoldische Buchhandlung: Dresden und Leipzig.

Saarinen, M. & Taskinen, J. 2004. Aspects of the ecology and natural history of Paraergasilus rylovi (Copepoda, Ergasilidae) parasitic in unionids of Finland. Journal of Parasitology 90: 948-952.

Saleuddin, A.S.M. 1965. The gonads and reproductive cycle of Astarte sulcata (Da Costa) and sexuality in A. elliptica (Brown). Proceedings of the Malacological Society of London 36: 229-257.

Savazzi, E. & Peiyi, Y. 1992. Some morphological adaptations in freshwater bivalves. Lethaia 25: 195-209.

Schneider, J.A. & Carter, J.G. 2001. Evolution and phylogenetic significance of cardoidean shell microstructure (Mollusca, Bivalvia). Journal of Paleontology 75: 607–643.

Scholz, H. 2003. Taxonomy, ecology, ecomorphology, and morphodynamics of the Unionoida (Bivalvia) of Lake Malawi (East-Africa). Beringeria 33: 1-86.

Scholz, H. & Glaubrecht, M. 2004. Evaluating limnic diversity: Toward a revision of the unionid bivalve Coelatura Conrad, 1853 in the Great Lakes of East Africa and adjacent drainage systems (Mollusca, Bivalvia, Unionidae). Mitteilungen des Museums für Naturkunde in Berlin, Zoologische Reihe 80: 89-121.

Scholz, H. & Hartmann, J.H. 2007a. Fourier analysis and the extinction of unionoid bivalves near the Cretaceous-Tertiary boundary of the Western Interior, USA: Pattern, causes, and ecological significance. Palaeogeography, Palaeoclimatology, Palaeoecology 255: 48-63.

Scholz, H. & Hartmann, J.H. 2007b. Paleoenvironmental reconstruction of the Upper Cretaceous Hell Creek formation of the Williston Basin, Montana, USA: Implications from the quantitative analysis of Unionoid bivalve taxonomic diversity and morphologic disparity. Palaios 22: 24-34.

Scholz, H., Tietz, O. & Büchner, J. 2007. Unionoid bivalves from the Miocene of Berzdorf (eastern Germany): taxonomic remarks and implications for palaeoecology and

Bibliography

169

palaeoclimatology. Neues Jahrbuch für Geologie und Paläontologie. Abhandlungen 244: 43-51.

Seed, R. 1980. Shell growth and form in the Bivalvia. In: Rhoads, D.C. & Lutz R.A. (eds.). Skeletal Growth of Aquatic Organisms. New York: Plenum Press. pp. 23-67.

Sell, H. 1907-1908. Biologische Beobachtungen an Najaden. Archiv für Hydrobiologie 3: 179-180.

Serb, J.M., Buhay, J.E. & Lydeard, C. 2003. Molecular systematics of the North American freshwater bivalve genus Quadrula (Unionidae: Ambleminae) based on mitochondrial ND1 sequences. Molecular Phylogenetics and Evolution 28: 1-11.

Sha, J.G. 1993. Trigonioides from early Cretaceous Hekou Formation of Hekou Basin, Ninghua, Fujian, China, with special remarks on classification of Trigonioididae. Acta Palaeontologica Sinica 32: 285-302.

Sha, J.G. 2006. Ontogenetic variations of the Early Cretaceous non-marine bivalve Trigonioides (T.) heilongjiangensis and their phylogenetic significance. In Malchus, N. & Pons J.M., (eds.). Organic Diversity and Evolution Supplement. Abstracts and Posters of the “International Congress on Bivalvia” at the Universitat Autònoma de Barcelona, Spain, 22-27 July 2006, 69-70.

Sha, J.G. 2007. Cretaceous trigonioidid (non-marine Bivalvia) assemblages and biostratigraphy in Asia with special remarks on the classification of Trigonioidacea. Journal of Asian Earth Sciences 29: 62-83.

Shine, R. 1989. Ecological causes for the evolution of sexual dimorphism: a review of the evidence. The Quarterly Review of Biology 64: 419-461.

Siebold, C.T.v. 1837. Über den Unterschied der Schalenbildung der männlichen und weiblichen Anodonten. Archiv für Naturgeschichte 3: 415-416.

Silverman, H., Kays, W.T. & Dietz, T.H. 1987. Maternal calcium contribution to glochidial shells in freshwater mussels (Eulamellibranchia: Unionidae). Journal of Experimental Zoology 242: 137-146.

Silverman, H., Steffens, W.L. & Dietz, T.H. 1985. Calcium from extracellular concretions in the gills of freshwater unionid mussels is mobilized during reproduction. Journal of Experimental Zoology 236: 137-147.

Simpson, C.T. 1896. Description of four new Triassic Unios from the Staked Plains of Texas. Proceedings of the United States National Museum 18: 381-385.

Simpson, C.T. 1900. Synopsis of the naiades, or pearly fresh-water mussels. Proceedings of the United States National Museum 22: 501-872.

Simpson, C.T. 1914. A Descriptive Catalogue of the Naiads, or Pearly Fresh-water Mussels. Part I-III. Bryant Walker: Detroit, Michigan.


170

Soroka, M. & Zdanowski, B. 2001. Morphological and genetic variability of the population of Anodonta woodiana (Lea, 1834) occurring in the heated Konin lakes system. Archives of Polish Fisheries 9: 239-252.

Stanley, S.M. 1970. Relation of shell form to life habits in the Bivalvia (Mollusca). Memoirs of the Geological Society of America 125: 1-296.

Stearns, S.C. 1989. The evolutionary significance of phenotypic plasticity. BioScience 39: 436-445.

Sterki, V. 1903. Notes on the Unionidae with their classification. American Naturalist 37: 103-113.

Stillwell, R.C. & Fox, C.W. 2007. Environmental effects on sexual size dimorphism of a seed-feeding beetle. Oecologia 153: 273-280.

Stone, N.M., Earll, R., Hodgson, A., Mather, J.G., Parker, J. & Woodward, F.R. 1982. The distributions of three sympatric mussel species (Bivalvia: Unionidae) in Budworth Mere, Cheshire. Journal of Molluscan Studies 48: 266-274.

Strayer, D.L. 2006. Challenges for freshwater invertebrate conservation. Journal of the North American Benthological Society 25: 271-287.

Tankersley, R.A. & Dimock, R.V., JR. 1992. Quantitative analysis of the structure and function of the marsupial gills of the freshwater mussel Anodonta cataracta. Biological Bulletin 182: 145-154.

Taskinen, J. 1998. Influence of trematode parasitism on the growth of a bivalve host in the field. International Journal for Parasitology 28: 599-602.

Taskinen, J. & Valtonen, E.T. 1995. Age-, size- and, sex-specific infection of Anodonta piscinalis (Bivalvia: Unionidae) with Rhipidocotyle fennica (Digenea: Bucephalidae) and its influence on host reproduction. Canadian Journal of Zoology 73: 887-897.

Taylor, J.D. & Kennedy, W.J. 1969. The influence of the periostracum on the shell structure of bivalve molluscs. Calcified Tissue Research 3: 274-283.

Taylor, J.D., Kennedy, W.J. & Hall, A. 1969. The shell structure and mineralogy of the Bivalvia. Introduction. Nuculacea-Trigonacea. Bulletin of the British Museum (Natural History), Zoology. Supplement 3: 1-125.

Tevesz, M.J.S. & Carter, J.G. 1980. Environmental relationships of shell form and structure in Unionacean bivalves. In: Rhoads, D.C. & Lutz. R.A. (eds.). Skeletal Growth of Aquatic Organisms. New York: Plenum Press. pp. 295-322.

Thom, M.D., Harrington, L.A. & Macdonald, D.W. 2004. Why are American mink sexually dimorphic? A role for niche separation. Oikos 105: 525-535.

Tregenza, T. & Butlin, R.K. 1999. Speciation without isolation. Nature 400: 311-312.

Bibliography

171

Trueman, E.R. 1966a. Bivalve mollusks: Fluid dynamics of burrowing. Science 152: 523-525.

Trueman, E.R. 1966b. The fluid dynamics of the bivalve molluscs Mya and Margaritifera. Journal of Experimental Biology 45: 369-382.

Trueman, E.R. 1968. The locomotion of the freshwater clam Margaritifera margaritifera (Unionacea: Margaritiferidae). Malacologia 6: 401-410.

Tudorache, C., Viaene, P., Blust, R., Vereecken, H. & De Boeck, G. 2008. A comparison of swimming capacity and energy use in seven European freshwater fish species. Ecology of Freshwater Fish 17: 284-291.

Tudorancea, C. 1972. Studies on Unionidae populations from the Crapina-Jijila complex of pools (Danube zone liable to indulation). Hydrobiologia 39: 527-561.

Vaughn, C.C. & Hakenkamp, C.C. 2001. The functional role of burrowing bivalves in freshwater ecosystems. Freshwater Biology 46: 1431-1446.

Vekemans, X. 2002. AFLP-SURV version 1.0. Distributed by the author. Bruxelles, Belgium: Laboratoire de Génétique et Ecologie Végétale, Université Libre de Bruxelles.

Vermeij, G.J. 1973. Morphological patterns in high-intertidal gastropods - adaptive strategies and their limitations. Marine Biology 20: 319-346.

Vermeij, G.J. 1993. A natural history of shells. Princeton University Press: Princeton, NJ. 216 pp.

Via, S., Gomulkiewicz, R., Scheiner, S.M., Schlichting, C.D. & Van Tienderen, P.H. 1995. Adaptive phenotypic plasticity: consensus and controversy. Trends in Ecology and Evolution 10: 212-217.

von Bertalanffy, L. 1938. A quantitative theory of organic growth. Human Biology 10: 181-213.

von Martens, E. 1897. Beschalte Weichthiere Deutsch-Ost-Afrikas. Dietrich Reimer (Ernst Vohsen): Berlin.

Vos, P., Hogers, R., Bleeker, M., Reijans, M., van de Lee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J. & Kuiper, M. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 4407-4414.

Wächtler, K., Mansur, M.C.D. & Richter, T. 2001. Larval types and early postlarval biology in naiads (Unionoida). In: Bauer, G. & Wächtler K. (eds.). Ecology and Evolution of the Freshwater Mussels Unionoida. Berlin: Springer-Verlag. pp. 93-125.

Walford, L.A. 1946. A new graphic method of describing the growth of animals. Biological Bulletin 90: 141-147.


172

Walker, J.M., Curole, J.P., Wade, D.E., Chapman, E.G., Bogan, A.E., Watters, G.T. & Hoeh, W.R. 2006. Taxonomic distribution and phylogenetic utility of gender-associated mitochondrial genomes in the Unionoida (Bivalvia). Malacologia 48: 265-282.

Wanner, H.E. 1921. Some faunal remains from the Trias of York County, Pennsylvania. Proceedings of the Academy of Natural Sciences of Philadelphia 73: 25-37.

Watters, G.T. 1994. Form and function of unionoidean shell sculpture and shape (Bivalvia). American Malacological Bulletin 11: 1-20.

Watters, G.T. 2001. The evolution of the Unionacea in North America, and its implications for the worldwide fauna. In: Bauer, G. & Wächtler K. (eds.). Ecology and Evolution of the Freshwater Mussels Unionoida. Berlin: Springer-Verlag. pp. 281-307.

Watters, G.T., Hoggarth, M.A. & Stansbery, D.H. 2009. The Freshwater Mussels of Ohio. The Ohio State University Press: Columbus.

Watts, P.C. 2001. Extraction of DNA from tissue: High salt method. Available at http://www.genomics.liv.ac.uk/animal/RESEARCH/ISOLATIO.PDF (accessed April 2009).

Weir, B.S. & Cockerham, C.C. 1984. Estimating F-statistics for the analysis of population structure. Evolution 38: 1358–1370.

Weisensee, H. 1916. Die Geschlechtsverhältnisse und der Geschlechtsapparat bei Anodonta. Zeitschrift für wissenschaftliche Zoologie 115: 262-335.

Wetherby, A.G. 1882. On the geographical distribution of certain fresh-water mollusks of North America and the probable causes of their variation. American Journal of Science 23 (3rd series): 203-212.

Wilbur, K.M. & Saleuddin, A.S.M. 1983. Shell formation. In: Saleuddin, A.S.M. & Wilbur K.M. (eds.). The Mollusca, Volume 4: Physiology. London and New York: Academic Press. pp. 235-287.

Williams, J.D., Warren, M.L., Cummings, K.S., Harris, J.L. & Neves, R.J. 1993. Conservation status of freshwater mussels of the United States and Canada. Fisheries 18: 6-22.

Wood, E.M. 1974. Development and morphology of the glochidium larva of Anodonta cygnea (Mollusca: Bivalvia). Journal of Zoology 173: 1-14.

Yeager, M.M., Cherry, D.S. & Neves, R.J. 1994. Feeding and burrowing behaviors of juvenile rainbow mussels, Villosa iris (Bivalvia, Unionidae). Journal of the North American Benthological Society 13: 217-222.

Yeap, K.L., Black, R. & Johnson, M.S. 2001. The complexity of phenotypic plasticity in the intertidal snail Nodilittorina australis. Biological Journal of the Linnean Society 72: 63-76.

Bibliography

173

Yonge, C.M. 1962. On Etheria elliptica Lam. and the course of evolution, including assumption of monomyrianism, in the family Etheriidae (Bivalvia: Unionacea). Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 244: 423-458.

Zhivotovsky, L.A. 1999. Estimating population structure in diploids with multilocus dominant DNA markers. Molecular Ecology 8: 907–913.

Zieritz, A. & Aldridge, D.C. 2009. Identification of ecophenotypic trends within three European freshwater mussel species (Bivalvia: Unionoida) using traditional and modern morphometric techniques. Biological Journal of the Linnean Society 98: 814–825.

Zieritz, A. & Aldridge, D.C. in review. Sexual, habitat-constrained and parasite-induced dimorphism in the shell of a freshwater mussel (Anodonta anatina, Unionidae).

Zieritz, A., Bogan, A.E. & Aldridge, D.C. in review. Variability and a new model for character evolution of umbonal sculptures in the Unionoida.

Zieritz, A., Checa, A.G., Aldridge, D.C. & Harper, E.M. in press. Variability, function and phylogenetic significance of periostracal microprojections in unionoid bivalves. Journal of Zoological Systematics and Evolutionary Research.

Zieritz, A., Hoffman, J.I., Amos, W. & Aldridge, D.C. 2010. Phenotypic plasticity and genetic isolation-by-distance in the freshwater mussel Unio pictorum (Mollusca: Unionoida). Evolutionary Ecology 24: 923-938.

ZIERITZ, A. 2010. Variability, function and phylogenetic significance of unionoid shell characters....

Documents

Transcript of ZIERITZ, A. 2010. Variability, function and phylogenetic significance of unionoid shell characters....