ETHOLOGY AND BEHAVIORAL ECOLOGY OF SIRENIA. 9783030907419, 9783030907426, 3030907414

Table of contents :
Introduction to the Series
Preface
Contents
1 What Can We Infer About the Behavior of Extinct Sirenians?
1.1 Habitat and Locomotion
1.2 Interactions with Other Animals
1.2.1 Parturition and Care of Young
1.2.2 Mating Systems
1.2.3 Agonistic Encounters
1.3 Feeding
1.3.1 Diet
1.3.2 Vibrissae
1.3.3 Rostral Deflection
1.3.4 Cropping Mechanism
1.3.5 Tusks
1.3.6 Intraoral Food Transport
1.3.7 Mastication
1.4 Sensory Perception
1.5 Concluding Remarks
References
2 Morphological and Sensory Innovations for an Aquatic Lifestyle
2.1 Introduction
2.2 Morphological Innovations for an Aquatic Life
2.2.1 Thermoregulation
2.2.2 Hydrostasis
2.2.3 Buoyancy
2.2.4 Swimming Behavior, Kinematics, and Performance
2.2.5 Feeding Innovations
2.3 Sensory Innovations for an Aquatic Life
2.3.1 Central Nervous System Characteristics
2.3.2 Peripheral Nervous System Characteristics—Structure and Function of Sensory Systems
2.3.3 Vision
2.3.4 Auditory System
2.3.5 Vocalizations
2.3.6 Gustation
2.3.7 Olfaction
2.4 Conclusions
References
3 Diving and Foraging Behaviors
3.1 Introduction
3.2 Diving
3.2.1 Anatomical Adaptations to Diving
3.2.2 Diving Physiology
3.2.3 Surfacing Behavior
3.2.4 Dive Duration
3.2.5 Dive Depths
3.2.6 Diving Behavior
3.3 Foraging
3.3.1 Methodology
3.3.2 Feeding Modes
3.3.3 Diet
3.3.4 Coprophagy
3.3.5 Ingestion of Non-food Items
3.3.6 Food Acquisition and Processing
3.3.7 Duration of Feeding
3.3.8 Impacts of Feeding on Vegetation
3.4 Conclusions
References
4 Social and Reproductive Behaviors
4.1 Introduction
4.2 Group Size and Dispersion
4.2.1 Group Size and Dispersion in Dugongs
4.2.2 Group Size and Dispersion in Manatees
4.3 Movements in Relation to Social Behavior
4.4 Mating Systems
4.4.1 Lek Mating in Dugongs
4.4.2 Mating Herds: Scramble Promiscuity
4.5 Mother-Young Behavior
4.5.1 Birthing
4.5.2 Suckling and Synchronous Breathing
4.6 Communication
4.6.1 Tactile and Chemosensory Communication
4.6.2 Vocal Communication: Mechanisms and Basic Acoustic Traits
4.6.3 Vocal Communication: Contexts and Variability
4.7 Other Social Behavior
4.8 Natal Philopatry, Dispersal, Matrilines, and Related Topics
4.8.1 Behavioral Observations from Florida Manatees in the Field
4.8.2 Support from Genetic Studies
4.8.3 Role of Tradition and Fission–Fusion Dynamics
4.9 Conclusions and Future Directions
References
5 Movement Behavior of Manatees and Dugongs: I. Environmental Challenges Drive Diversity in Migratory Patterns and Other Large-Scale Movements
5.1 Introduction
5.2 Methods for Studying Large-Scale Movement Behavior of Sirenians
5.3 Florida Manatee: Long-Distance Migrant on the Thermal Fringe of Sirenian Ranges
5.3.1 Regional Variation in Seasonal Movement Patterns in the Southeastern United States
5.3.2 Migratory Behavior and the Role of Temperature
5.3.3 Long-Distance Movements and the Role of Forage
5.3.4 Unusual Extralimital Movements by Florida Manatees
5.4 Antillean Manatee: Diversity of Aquatic Habitats Begets Large Variation in Movement Behavior
5.4.1 Seasonal Movement Patterns in Coastal Populations of the Antillean Manatee
5.4.2 Seasonal Movement Patterns in Riverine Populations of the Antillean Manatee
5.4.3 Seasonal Movement Patterns of the Antillean Manatee in Coastal-Riverine Systems
5.5 Amazonian Manatee: From Bottlenecks to Treetops in a Flood-pulse System
5.5.1 Seasonal Migrations and Changing Water Levels
5.5.2 Migratory Timing is Critical to Traverse Shallow Bottlenecks
5.6 African Manatee: Enigmatic Recluse in a Vast Continent
5.6.1 Year-Round Residency in Coastal Populations of the African Manatee
5.6.2 Seasonal Movement Patterns in Inland Populations of the African Manatee
5.6.3 Seasonal Movement Patterns of African Manatees in Coastal-Riverine Systems
5.7 Dugong: Seagrass Community Specialist in Coastal Marine Waters
5.7.1 Do Dugongs Migrate?
5.7.2 Movement Response of Dugongs to Periodic Declines in Forage
5.7.3 Individual Large-Scale Movements in Dugongs
5.8 Long-Term Range Fidelity
5.8.1 Interannual Fidelity to Seasonal Ranges and Refugia
5.8.2 Natal Philopatry
5.9 Conclusions: Commonalities and Contrasts Across Sirenian Species
5.10 Future Research Directions
References
6 Movement Behavior of Manatees and Dugongs: II. Small-Scale Movements Reflect Adaptations to Dynamic Aquatic Environments
6.1 Introduction
6.2 The Actors and Their Needs
6.3 Methods for Studying Sirenian Movement Behavior at Small Spatio-Temporal Scales
6.3.1 Radio-Tracking in the Field and via Satellite
6.3.2 Acoustic Tracking
6.3.3 Depth and Environmental Loggers
6.3.4 Instrumentation to Record 3-Dimensional Movements, Sounds, and Video
6.4 Space Use, Home Range, and Within-Season Movements
6.4.1 Florida Manatees
6.4.2 Antillean Manatees
6.4.3 Amazonian Manatees
6.4.4 African Manatees: Coastal Lagoon Systems
6.4.5 Dugongs
6.5 Diel Movement Patterns
6.5.1 Florida Manatees
6.5.2 Antillean, Amazonian, and African Manatees
6.5.3 Dugongs
6.6 Movements in Relation to Tidal Cycles
6.6.1 West Indian and African Manatees
6.6.2 Dugongs
6.7 Movements to Freshwater Sources by Manatees
6.8 Travel Corridors
6.8.1 West Indian and Amazonian Manatees
6.8.2 Dugongs
6.9 Flight Behavior from Threats
6.9.1 Manatees (All Species)
6.9.2 Dugongs
6.10 Movement Behavior Related to Reproduction
6.10.1 Manatees (All Species)
6.10.2 Dugongs
6.11 Concluding Remarks
References
7 Historical and Current Interactions with Humans
7.1 Introduction
7.2 Sirenian Morphology, Maternal and Feeding Behaviors Are Reflected in Human Mythology and Culture
7.3 Sirenian Morphology, Accessibility and Behavioral Predictability Make Them Attractive to Eat and Easy to Hunt
7.4 Sirenian Behavior and Habitat Use Increase the Risk of Interactions with Fishing Operations
7.5 Sirenian Behavior and the Conservation Challenges of Sirenian Tourism
7.6 Sirenian Morphology, Behavior and Anatomy Influence Their Interactions with Vessels
7.7 Behavioral Responses to the Effects of Human Activities and Degradation of Sirenian Habitats
7.7.1 Feeding
7.7.2 Calving and Nursing Behaviors
7.7.3 Behavioral Thermoregulation Increases the Links Between Florida Manatees and People
7.7.4 Movements
7.8 Conclusion
References
8 How Might Climate Change Affect the Ethology and Behavioral Ecology of Dugongs and Manatees?
8.1 Introduction
8.2 Changes to the Physical Parameters of Sirenian Habitats
8.2.1 Temperature Increases
8.2.2 Sea Level Rise
8.2.3 Changes in Water Chemistry and Quality
8.2.4 Increase in Intensity and Frequency of Extreme Weather Events
8.2.5 Changes in Rainfall Patterns
8.3 Changes to the Plant Communities on which Sirenians Depend
8.3.1 Seagrass Meadows
8.3.2 Mangrove Communities
8.3.3 Tidal Marshes and Estuarine Lagoons
8.3.4 Freshwater Wetlands
8.4 Co-stressors
8.4.1 Harmful Algal Blooms
8.4.2 Infrastructure
8.4.3 Land Clearing
8.4.4 Human Food Insecurity
8.5 Changes to Sirenian Ethology and Behavioral Ecology
8.5.1 Diving
8.5.2 Foraging
8.5.3 Social and Reproductive Behaviors
8.5.4 Movements and Habitat Use
8.5.5 Social Learning
8.6 Conclusions
References
Index

Citation preview

Ethology and Behavioral Ecology of Marine Mammals Series Editor: Bernd Würsig

Helene Marsh Editor

Ethology and Behavioral Ecology of Sirenia

Ethology and Behavioral Ecology of Marine Mammals Series Editor Bernd Würsig , Department of Marine Biology, Texas A&M University at Galveston, Galveston, TX, USA

The aim of this series is to provide the latest ethological information on the major groupings of marine mammals, in six separate books roughly organized in similar manner. These groupings are the 1) toothed whales and dolphins, 2) baleen whales, 3) eared seals and walrus, 4) true seals, 5) sea otter, marine otter and polar bear, and 6) manatees and dugong, the sirens. The scope shall present 1) general patterns of ethological ways of animals in their natural environments, with a strong bent towards modern behavioral ecology; and 2) examples of particularly well-studied species and species groups for which we have enough data. The scope shall be in the form of general and specific reviews for concepts and species, with an emphasis especially on data gathered in the past 15 years or so. A final 7th book was added since the beginning of this series, on “The Evolving Human Factor” to explore the effects that humans had, are having and will have (unless we change our ways) on these magnificent mammals of the seas. The editors and authors are all established scientists in their fields, even though some of them are quite young.

More information about this series at https://link.springer.com/bookseries/15983

Helene Marsh Editor

Ethology and Behavioral Ecology of Sirenia

Editor Helene Marsh James Cook University Townsville, QLD, Australia

ISSN 2523-7500 ISSN 2523-7519 (electronic) Ethology and Behavioral Ecology of Marine Mammals ISBN 978-3-030-90741-9 ISBN 978-3-030-90742-6 (eBook) https://doi.org/10.1007/978-3-030-90742-6 © Springer Nature Switzerland AG 2022 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover illustration: The close relationship between a mother and her calf is the most enduring feature of the social behavior of sirenians. On the left: A Florida manatee calf suckles from the left axillary teat of its mother. Photo by D.Schrichte/manateepics.com; On the right: Dugong mother and calf in Coral Bay, Western Australia. Photo by Samantha Lawrence. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

In his foreword to Marsh et al. (2011), John Robinson wrote: ‘On the surface of it, it seems preposterous to confuse mermaids and sirens with a 500 kg marine mammal with a face that only another sirenian might love’. Ten years later, this book confirms that the bond between a female sirenian and her nursing calf is long and strong and that the knowledge transmitted during the period of calf dependency is likely to be significant to individual survival. In future, deeper understanding of this knowledge may demonstrate the conservation significance of the mother love embodied in the mermaid legend and the ‘siren songs’ between mother and calf.

The top photograph of a Florida manatee and her calf is by David Schrichte. The bottom photograph of dugongs in Palau is by Mandy Etpison. Both are reproduced with permission. The reference is to Marsh H, O’Shea TJ, Reynolds JE III (2011) The ecology and conservation of Sirenia: dugongs and manatees. Cambridge University Press, 521pp.

Introduction to the Series

We—multiple topic editors and authors—are pleased to provide a series on ethology and behavioral ecology of marine mammals. We define ethology as “the science of animal behavior,” and behavioral ecology as “the science of the evolutionary basis for animal behavior due to ecological pressures.” Those ecological pressures include us, the humans. We determine, somewhat arbitrarily but with some background, that “marine mammals” habitually feed in the sea, but also include several mammals that went from saltwater oceans back into rivers, as see the chapter by Sutaria et al., first book on “Odontocetes.” Polar bears represent a somewhat outlier “marine mammal,” as they are quite at home in the sea, but can also feed on terrestrial mammals, birds, berries, lichens, and mosses. In six books, we include toothed whales (the odontocetes); baleen whales (the mysticetes); sea lions and fur seals (the otariids) as well as the walrus; true seals (the phocids); the special cases of the sea otter and polar bear; and manatees and the dugong (the sirens). Each of our chosen editors and their chapter authors have their own schedules, so the series will not arrive in the order given above, but within the five years of 2019 through 2023, all six marine mammal books on “Ethology and Behavioral Ecology of Marine Mammals” will see light of day, and you, the readers, will be able to ascertain their worth and their promise, as to current knowledge and to accumulating data while our fields of science advance. Since the first book on odontocetes came out in 2019, we added a seventh final book, on “The Human Factor,” with chapters on past assaults on marine mammals, continuing assaults on the marine and other environments, dawning of awareness of assaults, and perhaps ways that we humans can and must do better. Several of us simply felt that to detail modern science of marine mammal ethology and behavioral ecology was not enough—we need to be aware of the amazingly destructive Anthropocene epoch in which we live, and try to improve, for all of nature (and therefore also for us). While topics of human influence run throughout each of the first six books, a concentration on human actions and potential solutions is needed. Not all mammals that occur in marine waters are represented, nor all that have gone back to fresh water. Thus, there is nary a mention of marine-feeding bats, marine-feeding river otters, those aspects of beluga whales that foray way up into vii

viii

Introduction to the Series

major rivers, seals living in land-locked lakes at times thousands of kilometers from the ocean, and other species that occasionally make the marine environment or— as generally accepted marine mammals—adjacent fresh water systems, their home. Such are the ways of a summary, and we apologize that we do not fully encompass all. As series editor, I have been a science partner to all major taxonomic entities of this series, but this only because have been in the marine mammal field for about 50 years now, with over 65 graduate students who—in aggregate—have conducted research on all seven continents. In no manner do I pretend to have kept up with all aspects of diverse fields of modern enquiry. It is a special privilege (and delight) to have multiple up-to-date editors and their fine authors involved in this modern compilation, and I am extremely grateful (and humbled) for this. Still learning, and ever-so. Each chapter is reviewed by the book editors, peer reviewed by other scientists as chosen by the editors and perused and commented on by me. If you learned something new and imparted that to your colleagues, students, or your own mentors, then the series and sections of it shall have been worthwhile. Tortolita Desert, Arizona December 2021

With respect and best wishes Bernd Würsig

Preface

When series editor Bernd Würsig invited me to edit a book on the ethology and behavioral ecology of sirenians, I wondered whether there was enough material to justify a stand-alone volume on that topic. There are only four extant species of manatees and dugongs, and all are challenging to observe as they mostly occur in turbid waters and surface cryptically. All species are on the IUCN Red List of Threatened Species, mostly occur in developing countries and have “faces that only another sirenian might love”.1 Consequently, most research has been motivated by conservation concerns. Behavioral studies have rarely been a priority. This book demonstrates that my concern was unjustified. There is a substantive body of research relevant to this topic, and the results were in urgent need of synthesis. I have certainly learned a lot through editing this volume and hope you will too as you read it. From a biological perspective, sirenians are more different than most other marine mammals. As the only herbivorous mammals that spend all their lives in the water, they are grouped in a separate order in the clade Paenungulata. Thus, manatees and dugongs are more closely related to elephants and hyraxes than to other marine mammals. There are three recent genera: Hydrodamalis (one species), and the Dugong (one species) are in the Family Dugongidae; the three species of manatee, genus Trichechus in the family Trichechidae. Hydrodamalis gigas, (the giant Steller’s sea cow), once ranged widely across the coastal waters of the North Pacific but was hunted to extinction within three decades after the relict population was discovered by “Western science” in the Commander Islands in the eighteenth century. The range of the dugong, Dugong dugon, spans some 40 Indo-West Pacific countries from east Africa to Vanuatu. Manatees occur on both sides of the Atlantic Ocean: the African manatee, Trichechus senegalensis, in 21 tropical countries in West Africa; the West Indian manatee, Trichechus manatus, mirrors this distribution on the other side of the Atlantic. The West Indian manatee has two sub-species: the Florida manatee, Trichechus manatus latirostris, which occurs in the southeastern USA and the Bahamas, and the Antillean manatee, Trichechus manatus manatus, with a range 1

Quote from John Robinson’s foreword to Marsh et al. (2011). ix

x

Preface

across 19 countries from Mexico to Brazil. The Amazonian manatee, Trichechus inunguis, exists in four countries in the Amazon basin. Even though sirenian habitats are limited to relatively shallow waters where there is sufficient light to support the plant communities on which they depend for food, their habitats are diverse: The dugong is strictly marine, the West Indian and African manatees occur in coastal, riverine, and lake habitats, while the Amazonian manatee exists only in freshwater habitats. Unlike the extinct giant Steller’s sea cow, all extant species are restricted to the sub-tropics and tropics due to their limited tolerance of colder water. Although they are in separate families, manatees and dugongs look remarkably alike. The most obvious difference is in the shape of the tail: Manatees have a round tail like that of beavers (Castor sp.); the tail of a dugong somewhat resembles that of a cetacean. The build of a manatee is more robust than that of a dugong, the latter looks like a manatee that goes to the gym! This book has eight chapters. Chapters 1 (Domning) and 2 (Marshall et al.) set the scene. The order Sirenia has a rich fossil record, and the three recent genera have adapted in radically different ways to environmental changes over the past 10 million years, especially with regard to their feeding adaptions. Chapter 1 explores what we can infer about the behavior of fossil sirenians from their skeletal remains. Chapter 2 explains the novel innovations of manatees and dugongs for life as aquatic herbivores. Their large body size confers thermal advantages and protection from predation. Their morphology also enables easy transitions from the benthic substrate, where their food is often located, to the surface where air is inhaled. Their perception of the aquatic environments is largely through somatosensation (touch and hydrodynamic reception) and hearing, although vision and taste (chemoreception) are also important to some degree. These morphological and sensory traits determine many aspects of sirenian ethology and behavioral ecology. Chapters 3–6 describe the behaviors of manatees and dugongs: diving and feeding (Chap. 3: Keith-Diagne et al.), social and reproductive behaviors (Chap. 4: O’Shea et al.), and movements (Chaps. 5 and 6: Deutsch et al.). The diving achievements of all sirenians are modest, mostly reflecting the distributions of the food communities on which they depend. Although all sirenian diets are plant-based, it is likely that they all eat some animals as well. African manatees are arguably best described as omnivores. In contrast to dugongs, which apparently meet their water requirements from their food, all manatees like to drink fresh water. Sirenian social and reproductive behaviors lack much complexity or diversity and social groupings are transient, apart from the close relationship between a cow and her suckling calf. Home ranges overlap. Socially transmitted knowledge (tradition) is important to Florida manatees and perhaps all species, particularly when movements are necessary for survival. The development and deployment of animal-borne GPS telemetry has enabled the movement behaviors of sirenians to be studied at a range of spatial scales, despite the difficulties of observing them directly. Consequently, Chip Deutsch and his collaborators required two chapters to synthesize knowledge of sirenian movement behaviors. Chapter 5 reviews sirenian movements across large spatial scales; Chapter 6 considers their movements and habitat use across diel and seasonal temporal scales.

Preface

xi

Individual manatees and dugongs can be marathon swimmers, undertaking longdistance journeys over hundreds of kilometers. There is considerable variation in large-scale movement behaviors among and within individuals. The environments inhabited by manatees and dugongs are spatially heterogeneous and dynamic over a range of time scales. Sirenians must negotiate trade-offs among key activities within these fluctuating environments while minimizing exposure to predators including hunters and other anthropogenic threats. In his Introduction to the Series, Bernd Würsig defines behavioral ecology as “the science of the evolutionary basis for animal behavior due to ecological pressures” and points out that these ecological pressures include us, the humans. We live in the Anthropocene, and human existence is now the biggest influence on the environment. Chapters 7 (Ponnampalam et al.) and 8 (Marsh et al.) consider the implications of current and future environments on the interactions between sirenians and people. Many sirenian habitats overlap with sites of high human use, and people have for millenia used knowledge of the predictability of sirenian habitat use and behaviors to capture them for their meat and other products. Manatees and dugongs nurse their calves over prolonged periods via axillary teats, fostering the widespread belief that sirenians are the basis of mermaid myths with associated folklore and magic. The cultural values of both dugongs and manatees are very high, especially for indigenous peoples. Dugongs and manatees are exposed to modern anthropogenic threats including habitat loss, vessel strike and the increased incidence of harmful algal blooms. Climate change stressors are already affecting the subtropical and tropical coastal, estuarine, and riverine habitats of sirenians with consequential changes to their ethology and behavioral ecology. Climate stressors are predicted to increase over the coming decades and will be exacerbated by anthropogenic stressors, especially habitat loss and human food insecurity. The cumulative impacts on all sirenian habitats will be locally variable, but changes in habitat extent, continuity and food plants are likely to be widespread. Knowledge of how sirenians alter their behaviors in response to these changes will be central to designing strategies to increase their resilience to the climatic changes to their habitats. Major questions remain. Do dugongs practice lek mating as well as scramble promiscuity, and if so is lek mating associated with “vocalization hotspots?” How great is the manatees’ physiological need for freshwater? What factors determine food quality for each species of sirenian? How do we determine the carrying capacity of sirenian habitats? How do sirenians navigate over large distances? How quickly can an individual update its cognitive environmental map, especially if it is no longer dependent on its mother? How do sirenians transmit information among conspecifics? The answers to these questions will determine capacity of sirenians to alter their behavior in response to climate change. I predict that knowledge of the behavioral ecology of sirenians will become central to future conservation efforts. I am delighted to thank the many people who have assisted with this book, especially the 23 authors from seven countries and all major continents, a spread that fittingly represents the huge collective range of sirenians. The corresponding authors were remarkable in their persistence in “herding cats in a pandemic” and

xii

Preface

their capacity to find photographers and artists that allowed us to reproduce their extraordinary works. Sirenians are notoriously difficult to photograph, and it is both wonderful and informative for this book to have such a fine spread of illustrations. I am extremely grateful to the reviewers and editors whose thoughtful comments improved the book, especially to Éva Lörinczi and her highly professional team at Springer. Series editor Bernd Würsig has been unfailingly supportive in gently and positively providing wise advice. Last but certainly not least, I thank my husband, Lachlan Marsh, for his unstinting love and support. Townsville, Australia March 2022

Helene Marsh

All sirenians use their sensory hairs to obtain information about their environnment. These hairs occur all over the body but are most developed on the oral disk. In the top image, a dugong uses its oral disk to find sparse seagrass shoots (Photo by Ahmed Shawky). In the bottom image, two Florida manatees use their oral disks to explore the skin of a third (Photo by D.Schrichte/manateepics.com). Both images reproduced with permission.

Contents

1 What Can We Infer About the Behavior of Extinct Sirenians? . . . . . . Daryl P. Domning

1

2 Morphological and Sensory Innovations for an Aquatic Lifestyle . . . Christopher D. Marshall, Diana K. Sarko, and Roger L. Reep

19

3 Diving and Foraging Behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lucy W. Keith-Diagne, Margaret E. Barlas, James P. Reid, Amanda J. Hodgson, and Helene Marsh

67

4 Social and Reproductive Behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Thomas J. O’Shea, Cathy A. Beck, Amanda Hodgson, Lucy Keith-Diagne, and Miriam Marmontel 5 Movement Behavior of Manatees and Dugongs: I. Environmental Challenges Drive Diversity in Migratory Patterns and Other Large-Scale Movements . . . . . . . . . . . . . . . . . . . . . . 155 Charles J. Deutsch, Delma Nataly Castelblanco-Martínez, Rachel Groom, and Christophe Cleguer 6 Movement Behavior of Manatees and Dugongs: II. Small-Scale Movements Reflect Adaptations to Dynamic Aquatic Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Charles J. Deutsch, Delma Nataly Castelblanco-Martínez, Christophe Cleguer, and Rachel Groom 7 Historical and Current Interactions with Humans . . . . . . . . . . . . . . . . . 299 Louisa S. Ponnampalam, Lucy Keith-Diagne, Miriam Marmontel, Christopher D. Marshall, Roger L. Reep, James Powell, and Helene Marsh

xv

xvi

Contents

8 How Might Climate Change Affect the Ethology and Behavioral Ecology of Dugongs and Manatees? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 Helene Marsh, Camille Albouy, Eduardo Arraut, Delma Nataly Castelblanco-Martínez, Catherine Collier, Holly Edwards, Cassandra James, and Lucy Keith–Diagne Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407

Chapter 1

What Can We Infer About the Behavior of Extinct Sirenians? Daryl P. Domning

Abstract Several aspects of the behavior of fossil sirenians can be inferred from their skeletal remains. Their transition from terrestrial walking to obligate swimming is relatively well documented by their postcranial skeletons. The salinity of their aquatic habitats, as well as their diets, is determinable from stable isotopes in their tooth enamel. Deflection and width of the front parts of their skulls, respectively, reflect where in the water column they fed, and how selective they were in feeding. Specializations of tusks and other teeth also offer hints about diet, intraoral food transport, and mastication. Sizes of the infraorbital and mental foramina may reflect the importance of their prehensile and tactile vibrissae. The three Recent sirenian genera have divergently adapted in radically different ways, especially in feeding adaptations, to environmental changes of the last 10 million years. Fossils shed little light on vision, chemical senses, or touch, apart from the facial vibrissae, but future study of their ear bones could reveal much about the evolution of sirenian hearing. Keywords Ballast · Fossil sirenians · Habitat · Intraoral food transport · Locomotion · Mastication · Oripulation · Rostral deflection · Swimming · Teeth · Vibrissae The fossil record of vertebrates is not usually considered to be a source of detailed behavioral data. Some broad generalizations, however, are possible, and specifically for the sea cows, there are several major categories of behavior on which fossil bones can shed at least dim light.

1.1 Habitat and Locomotion As descendants of land mammals, sirenians obviously underwent a major transition to an aquatic lifestyle early in their history: so early, in fact, that no fully terrestrial ancestors have yet been identified. Proboscideans are generally considered their D. P. Domning (B) Department of Anatomy, College of Medicine, Howard University, Washington, DC 20059, USA e-mail: [email protected] © Springer Nature Switzerland AG 2022 H. Marsh (ed.), Ethology and Behavioral Ecology of Sirenia, Ethology and Behavioral Ecology of Marine Mammals, https://doi.org/10.1007/978-3-030-90742-6_1

1

2

D. P. Domning

closest living relatives, although other Paenungulata, Afrotheria, or Tethytheria are possible candidates for their actual sister group (cf. Gheerbrant et al. 2005). This implies an Old World origin for the sirenia. A skull fragment of a very primitive, Early or Middle Eocene supposed sirenian has been reported from North Africa (Benoit et al. 2013); however, its postcranial skeleton and mode of locomotion are unknown. The next most primitive sirenian, Prorastomus (also known by just a skull; Savage et al. 1994; Table 1), from the early Middle Eocene of Jamaica (ca. 47 Ma), was aquatic enough to have at least followed the warm North Atlantic shoreline to the New World, if it did not actually swim the then-narrower low-latitude Atlantic Ocean (Fig. 1.1). Somewhat later (late Middle Eocene, ca. 42 Ma), and also from Jamaica, is Pezosiren (Fig. 1.2), the earliest sirenian known from fossils that adequately show the form of its body (Domning 2001a). It was a low-slung quadruped about 2.1 m long, with a relatively short neck, barrel-shaped trunk, short but strong legs, toes built for land rather than flippers, a firm sacroiliac joint that could support its weight on land, and a substantial (but not powerfully muscled) tail. The tall neural spines of its anterior thoracic vertebrae indicate a strong nuchal ligament that could support its heavy head and neck out of water. Although clearly amphibious, it bears distinct marks of having spent most of its time in the water (more so than a modern Hippopotamus, which rests in the water but forages on land): strongly retracted nasal openings, lack of paranasal air sinuses, and in particular some 20 pairs of swollen (pachyostotic)

Fig. 1.1 Approximate positions of the major landmasses in the Middle Eocene. The Tethys Seaway (along the shores of which the early sirenians arose at an earlier time) was open at both ends, allowing dispersal of sirenians to the Caribbean via warm circumtropical currents (arrows). The Caribbean was also closer to the Tethys Sea in the Early Eocene than later in the earth’s history, and warm waters extended above latitudes of the landmasses now known as Europe. (Map drawn by Adella Edwards, reprinted from Marsh et al. (2011) with permission)

1 What Can We Infer About the Behavior of Extinct Sirenians?

3

Fig. 1.2 Artist’s model of possible appearance of Pezosiren portelli. (Model courtesy of locolobo.org, drawing by Gareth Wild, reproduced from Marsh et al. (2011) with permission)

ribs composed of bone nearly as dense (osteosclerotic) as those of modern sea cows (Buffrénil et al. 2010). These ribs serve as ballast, helping to provide neutral buoyancy, especially while feeding on bottom-growing plants (Domning and Buffrénil 1991). Judging from details of its vertebrae and hindquarters, Pezosiren swam by simultaneous pelvic paddling (kicking both hind legs backward together) along with dorsoventral pelvic undulation, foreshadowing the up-and-down tail propulsion of all later sea cows. Corroborating the interpretation of aquatic habits based on sirenians’ gross osteology is the evidence from stable isotopes in their tooth enamel, which show that these Eocene sea cows were already dwelling and feeding in saltwater, albeit of varying salinity (e.g., Clementz et al. 2009). Modern manatees (Trichechidae, Subfamily Trichechinae; Table 1) are euryhalic except for Trichechus inunguis, the most highly derived species, which is strictly freshwater. Although trichechines have a very limited fossil record (Domning 1982, 1997, 2005; Perini et al. 2020), the extinct trichechid subfamily Miosireninae was evidently marine, so the Family Trichechidae as a whole most likely was primitively marine as well. T. inunguis has fewer (14–16 pairs) and relatively slender ribs, while the other two living species (T. manatus and T. senegalensis) have more (respectively, 16–19 and 17–18 pairs) and broader and heavier ribs (especially T. manatus) (Domning and Hayek 1986), while the extinct Miosiren had 20 pairs of extremely massive ribs (Sickenberg 1934). This may reflect the freshwater habitat of T. inunguis and its consequent lesser need for ballast.

4

D. P. Domning

Sobrarbesiren (Middle Eocene of Spain; Diaz-Berenguer et al. 2018) and Protosiren (e.g., P. smithae, early Late Eocene of Egypt [ca. 37 Ma]; Domning and Gingerich 1994; Zalmout and Gingerich 2012) represent a diverse but poorly sampled Eocene adaptive radiation of amphibious sirenians that were more derived than Pezosiren. They show progressive weakening of the weight-bearing limbs, to the extent that P. smithae has been called “aquatic quadrupedal” rather than “amphibious quadrupedal” (Diaz-Berenguer et al. 2018). This trend continued with the development of propulsion using a powerful tail, followed by rapid loss of external hind limbs in the Late Eocene. Subsequent sirenians were all exclusively aquatic. As part of their increasing efficiency at swimming, the Family Dugongidae (Table 1.1) had also evolved dolphin-like triangular flukes and a caudal-oscillation mode of propulsion, like cetaceans, by at latest the Early Oligocene. The flukes are indicated in fossil skeletons by widening of the caudal vertebrae posterior to the peduncle, as seen in the Recent Dugong. Manatees (Trichechus), in contrast, have retained a primitively beaver-like, rounded caudal fin in which the vertebrae steadily decrease in width toward the tip. Their mode of swimming is classified as caudal dorsoventral undulation, which is suited to a less active lifestyle than the dugong’s (Kojeszewski and Fish 2007). Steller’s sea cow, Hydrodamalis gigas, was unique among sirenians in its great body size (clearly advantageous in its cold climate), but also unique among tetrapods in having lost entirely the phalanges of the front limb: its claw-like hand skeleton comprised only carpals and metacarpals, as observed and stated by Steller (1751). Table 1.1 The families of Sirenia. References cited emphasize more recent literature and reviews. The online bibliography of Sirenia should be consulted for a more comprehensive listing of additional sources Family

Time of occurrence

Distribution

References

Prorastomidae

Early middle to late Eocene

West Indies, North America, north and west Africa

Savage et al. (1994); Domning (2001a); Gheerbrant et al. (2005); Hautier et al. (2012); Benoit et al. (2013)

Protosirenidae

Early middle to late Eocene

Western Atlantic, Mediterranean, Indian Ocean regions

Domning and Gingerich (1994); Domning (2001b); Gheerbrant et al. (2005); Bajpai et al. (2009)

Dugongidae

Middle Eocene to recent

Mediterranean, Europe, North Africa, western Atlantic–Caribbean, Indian and Pacific oceans

Domning and Furusawa (1995); Domning (2001b); Gheerbrant et al. (2005)

Trichechidae

Oligocene to recent

Europe, South America, Domning (1982, 2001b, western 2005); Gheerbrant et al. Atlantic–Caribbean, North (2005) America, west Africa

1 What Can We Infer About the Behavior of Extinct Sirenians?

5

This reflected a major shift in uses of the flipper, from mainly steering and clasping food items or other animals, to harvesting seaweed and fending off from rocks. These inferences are corroborated by reconstructions of the forelimb muscles from their bony attachments (Domning 1978: 124–129). A Late Miocene ancestor of Hydrodamalis, Dusisiren dewana, displays an earlier stage in reduction of the phalanges (Takahashi et al. 1986). This feature corroborates Steller’s account, which some later anatomists had difficulty believing, as evidenced by their attempts at reconstructing the animal’s appearance (e.g., Kleinschmidt 1951). Even harder to accept has been the testimony of Steller and other eyewitnesses that H. gigas was so buoyant that it was never seen to entirely submerge. Although loss of diving ability has not been (and perhaps cannot be) confirmed from the H. gigas skeleton alone, it is actually reasonable to assume when neck muscle attachments, other anatomical and physiological considerations, and arguments from selective value are taken into account (Domning 1978: 129–132). Its body size increase compared to its ancestors, including a proportionately larger gut and possibly greater lung gas volume, meant that the bones made up proportionately less of the body, thereby having less capacity to act as an “inbuilt weight-belt.” Keeping part of the back out of the water would reduce heat loss to the water. The thick blubber layer would have done the same, while further enhancing the animal’s buoyancy. Floating would also have reduced wave drag while swimming at the surface, allowed access to shallower water to forage and elude predators, reduced skin area accessible to parasites, and permitted birds to remove them. On the other hand, obligate rather than facultative buoyancy would have diminished access to deeper-growing plants, exposed the back to subzero air temperatures, and exposed any open wounds to further damage by birds. On the whole, the adaptations of Hydrodamalis would seem to repay further analysis.

1.2 Interactions with Other Animals 1.2.1 Parturition and Care of Young There is no direct indication of whether prorastomids or other amphibious sirenians gave birth on land or in the water. However, the extreme slenderness and fragility of a juvenile Pezosiren femur (105 mm long, with mid-shaft diameters of only 9 × 6 mm) casts doubt on the ability of juveniles to support their body weight on land until quite some time after birth. Whether born on land or in the water, therefore, they might have started to swim at an early age; but they must have rested on their bellies like seals whenever they hauled out, until the shafts of their long bones had thickened enough to support them in the manner of adults. Whatever provision the parents made in the meantime for care of their offspring and their protection from predators is open to speculation.

6

D. P. Domning

1.2.2 Mating Systems Social arrangements of extinct sea cows also remain speculative beyond what we might extrapolate from observations on the living species, as Anderson (2002) has done in great detail. Judging from the Recent sirenians alone, these social systems may always have been quite diverse, as they are in terrestrial herbivores. The same goes for other aspects of natural history such as circadian behavior, movements, and migrations, which may lie forever beyond the reach of the paleontologist. On the borderline between the extinct and extant realms are the few eyewitness reports of the behavior of Hydrodamalis. For example, in contrast to scramble promiscuity in manatees, and in some dugong populations, and alleged lek mating by dugongs in eastern Shark Bay in Western Australia, Anderson (2002) noted Steller’s observations that it formed monogamous pair bonds and presented an ecological rationale supporting this mating system. Anderson’s analyses are a rich source of testable hypotheses about mating in other prehistoric sirenians, if anyone is clever enough to devise suitable tests.

1.2.3 Agonistic Encounters Anderson (2002) also emphasized the potential role of tusks in mating and other social interactions, a role which has also recently been documented in a study of scars inflicted on the skin of dugongs by the tusks of other dugongs (Lanyon et al. 2021). However, this behavior is unlikely to leave its mark in the fossil record. What are frequently observed on fossil sirenian (and other marine mammal) bones, are scars made by shark teeth, whether the result of predation or scavenging is unknown. An interesting question is the extent to which the sirenians might have actively defended themselves from sharks or other predators like killer whales or crocodilians. There are anecdotal reports of a bull dugong disemboweling a shark with its tusks to protect the calves in its herd (Promus 1937) or killing a crocodile by repeatedly jumping out of the water and landing on top of it (Sunter 1937), although, again, corroboration is lacking and fossil evidence still more so. As for the tusks themselves, Anderson (2002) acknowledges their multifunctionality (for social functions, whether intra- or interspecific, and for use in foraging) and ponders which uses have been primary versus secondary in evolution. He argues (p. 78) that “[e]volution and retention of tusks exclusively, or even primarily, as foraging structures would be unique among mammalian herbivores.” But then, no other mammalian herbivores have been exclusively aquatic. My own thinking is that tusks are teeth and teeth are primitively for feeding, but this function has never excluded simultaneous use for other purposes that may present themselves. This will be expanded upon below.

1 What Can We Infer About the Behavior of Extinct Sirenians?

7

1.3 Feeding 1.3.1 Diet We know from the living species that the sirenians have the collective capacity to feed on almost any plant types anywhere in the water columns of their marine or freshwater habitats. Occasionally, they even feed on plants close to, or overhanging, the water, even though individual species, especially the dugong, are more specialized (Chap. 3). Most fossil sirenians were tropical and marine, and thus (like modern dugongs) presumably fed on seagrasses in preference to algae (a supposition confirmed by stable isotopes; Clementz et al. 2009), though the cooler-water, kelp-eating hydrodamalines were an exception (Domning 1978).

1.3.2 Vibrissae A more applicable generalization is that sirenians have always relied heavily on their muscular lips and vibrissae for sensing, grasping, and ingesting plants, which Reep et al. (2001) call “oripulation” (in effect, manipulation by mouth instead of hands; see Chap. 2). This is reflected in the fossil record by the increased size and/or number of the infraorbital and mental foramina: openings on the side of the snout and lower jaw, respectively, that carry branches of the trigeminal nerve, which transmit touch sensations from the vibrissae. Sirenians more derived than the Eocene prorastomids and protosirenids show marked enlargement of the infraorbital foramen in particular, suggesting that the array of vibrissae (and, implicitly, the facial muscles that move them, and the blood vessels that supply the muscles) had taken on greater importance in the animals’ behavior.

1.3.3 Rostral Deflection A still more direct and sensitive indicator of feeding behavior is the form of the rostrum, the enlarged bony snout that characterizes all sirenians. In the earliest known form, Prorastomus (Savage et al. 1994: Figs. 1, 6), this projects horizontally forward from the braincase as in land mammals. But in all later sea cows, the rostrum is turned down to varying degrees, so that its palatal surface forms a plane deflected from the occlusal plane of the cheek teeth. As explained in Chaps. 2 and 3, the angle of deflection correlates with the extent to which the extant manatees rely on bottomgrowing plants for their diet: 15–40° (mean of 20 = 25.8°) in Trichechus senegalensis, which feeds on overhanging, shoreline, emergent and floating aquatic plants, as well as benthic ones; 25–36° (mean of 35 skulls = 30.4°) in T. inunguis, which feeds almost entirely on floating vegetation; and 24–52° (mean of 72 = 38.2°) in T.

8

D. P. Domning

manatus, which has as broad a dietary range as T. senegalensis but with more access (especially in Florida) to marine habitats with benthic seagrasses (Domning 1982; Marsh et al. 2011). In marked contrast, the exclusively bottom-feeding Dugong has rostral deflections in the neighborhood of 67–72° (in 3 D. dugon). The sirenian mouth is consequently subterminal, and where the deflection is as great as in Dugong, it opens nearly straight down (see Chap. 2), as is desirable for an essentially obligate bottom-feeder. The deflection is accordingly seen as a good indicator of where in the water column fossil sirenians fed.

1.3.4 Cropping Mechanism Another relevant variable seen in the upper and lower jaws is the width of the masticating surfaces at the front of both jaws, particularly that of the mandible. In Prorastomus, the latter surface is extremely narrow, scarcely wider than the parallel left and right tooth rows borne by the very elongate mandibular symphysis (Savage et al. 1994: Figs. 6, 8). Together with the rostral masticating surface, which is also relatively narrow, this forms an almost forceps-like mechanism well suited for selective browsing amid diverse stands of floating as well as benthic plants. Later sirenians, in contrast, such as the modern dugong, which is a seagrass community specialist (Marsh et al. 2018), generally show mandibular surfaces that are proportionately much broader. An analogous contrast has been documented in terrestrial herbivores, between narrow-snouted browsers and broad-snouted grazers (Janis and Ehrhardt 1988).

1.3.5 Tusks Most fossil sirenians had first upper incisors that were enlarged to various degrees to form tusks. These were primitively small and conical in most Eocene forms and do not show much potential as weapons of defense or offense. In most later dugongids the tusks came to be larger, with deep roots extending most, or all of the length of the premaxillary symphysis, and accordingly more likely to have had social uses along with more forceful application as digging tools. In a few specialized Oligocene and Miocene dugongines like Rytiodus and Corystosiren, the tusks became flattened and bladelike with a posterior, self-sharpening cutting edge held up by thin enamel that covers the medial surface (Fig. 1.3). These cutting tools (rather resembling a box-cutter, with a short cutting edge attached to a large handle embedded deep in the bone) are interpreted as having been used to sever the tough, fibrous rhizomes of large seagrasses like Thalassia, whereas rhizomes of the smallest seagrasses (e.g., Halophila) can be harvested efficiently even without tusks (Domning and Beatty 2007). The similarly self-sharpening tusks in Dugong dugon, however, are large but sexually dimorphic: developed in both sexes, but erupting only in males and

1 What Can We Infer About the Behavior of Extinct Sirenians?

9

Fig. 1.3 Replicas of fossil sirenian tusks: a Metaxytherium floridanum, left tusk in premaxilla, medial view. b Crenatosiren olseni, right tusk, medial view. c Dioplotherium manigaulti, left tusk, medial view. d Corystosiren varguezi, distal part of right tusk, lateral view showing broad wear surface, partly restored. The tusks were self-sharpening because of the different thickness of the enamel on different sides of the tusk. Scale bar = 15 cm. Reproduced from Domning and Beatty (2007) with permission

post-reproductive females, and apparently no longer important in feeding but only in social interactions like fighting and mating (Marsh 1980; Domning and Beatty 2007; Lanyon et al. 2021; Chap. 4).

1.3.6 Intraoral Food Transport Prorastomus retained full batteries of anterior teeth (incisors and canines) that lined the edges of the rostral and mandibular masticating surfaces. These more-or-less peglike teeth, which evidently did not occlude with one another (only with the food), doubtless formed part of the cropping mechanism. But in addition, they (especially the more posterior ones, including most of the premolars) seem likely to have aided in intraoral food transport. The tongue is short in sirenians, not reaching to the front of the mouth, so these anterior teeth might have served like a ratchet to keep ingested vegetation moving toward the back of the mouth. This became all the more necessary as the rostral deflection became steeper.

10

D. P. Domning

But coincident with this development, something else occurred: In almost all sirenians, the incisors (excepting the first upper incisor tusk, if present) and the canines were steadily lost over the course of the Eocene and Early Oligocene, leaving the anterior masticating surfaces toothless, but covered with horny pads. While these pads are useful for grasping and crushing plant material, they also (at least in Dugong) form a surprisingly effective transport mechanism (Chap. 2). The surfaces of these pads are formed by short, closely packed bristles that are angled posteriorly (Lanyon and Sanson 2006a). When the upper and lower pads are rubbed together, leaves placed between them are moved steadily up the sloping surfaces toward the cheek teeth, as though on a conveyor belt. I suspect that this proved so efficient a means of intraoral transport that it rendered superfluous the much larger and clumsier incisors, canines, and anterior premolars.

1.3.7 Mastication This brings us to the cheek dentition, which primitively in sirenians comprised five premolars and three molars in each jaw quadrant. Deciduous teeth occupied all the premolar positions (except possibly the first). These were replaced during growth, as in most mammals, and the permanent ones were more peg-like, like the more anterior teeth. Whether these five premolars (instead of the four found in typical placental mammals) represent a primitive condition retained from Mesozoic mammals, or a derived condition unique to sirenians, remains in doubt. What we do know, however, is that from almost their first appearance in the fossil record sirenians started losing teeth, beginning with the permanent fifth premolars. This loss resulted in that tooth’s deciduous precursor (which resembled the true molars in having a low crown with two transverse ridges formed by distinct cusps) being retained into adulthood, thereby increasing the functional molariform battery from three to four. Thereafter, the more peg-like anterior teeth (all the permanent and most of the deciduous ones) were gradually eliminated in the course of evolution, starting at the front. By the Late Oligocene and continuing down through the Pliocene, a typical adult sirenian (of which nearly all the known ones were dugongids) had only a first upper incisor tusk on each side, separated by a long diastema (toothless gap) from the cheek tooth battery, which comprised the molariform fifth deciduous premolar followed by three molars (Domning 1982). These four teeth came into wear in sequence, so they display a strong front-to-back gradient of decreasing wear. By maturity, the fifth premolar is usually worn out and lost, followed by loss of one or more of the molars in old age. What does all this tell us about behavior? First, that for most of the last 28 million years nearly all sirenians had settled on a tried-and-true oral mechanism for processing seagrass in a very stable tropical to subtropical marine ecological niche and there was no reason to experiment very much with this mechanism (except for the tusks, as discussed below). Thus in contrast to most fossil mammals, sirenian teeth are not particularly useful for distinguishing species or higher taxa: The molars

1 What Can We Infer About the Behavior of Extinct Sirenians?

11

mostly look pretty much alike, and identification and classification have to be based mainly on the structure of the skull itself. But then, within the last 10 million years or even less, all sirenians with dentitions like those just described either went extinct or evolved into something quite different. This was doubtless due in various ways to the significant global cooling and other environmental changes that occurred during this period. In any case, all three of the lineages that survived into historic times evolved feeding adaptations drastically different from each other and from any that came before. Hydrodamalis: In the North Pacific, climatic cooling, consequent marine floral shift from a tropical seagrass-dominated to a temperate kelp-dominated ecosystem, and (in North America) tectonic uplift and draining of coastal: embayments (e.g., the Central Valley of California), led to the evolution of a cool-adapted lineage culminating in Steller’s sea cow (H. gigas) (Domning 1978; Takahashi et al. 1986; Domning and Furusawa 1995). This genus was characterized not only by large body size (for heat retention) and the loss of phalanges noted above (producing a boathook-like forelimb, useful for fending off from rocks on high-energy shorelines), but also total loss of teeth (permitted by a non-fibrous algal diet) and mastication performed solely with the horny rostral pads. Trichechus: In South America, temporary conversion of the western Amazonian region into a closed basin with interior drainage appears to have isolated a population of Miocene trichechids, while also fertilizing the waters of this basin with runoff from the rising, rapidly-eroding Andes Mountain range. This nourished a freshwater ecosystem dominated by floating true grasses, which contained not only nutrients but also highly abrasive siliceous phytoliths that radically increased the rate of wear on the teeth of the sirenians. They adapted by evolving the endless horizontal replacement of small, cheap, disposable molars that characterizes Trichechus. Subsequent breakthrough of Amazonian drainage to the Atlantic allowed dispersal of these manatees to North America and West Africa, with subsequent speciation, and may even have equipped them to outcompete Caribbean dugongids having less wear-resistant teeth (Domning 1982, 1997, 2005; Domning and Hayek 1984). Dugong: Probably the latest, the least dramatic, but certainly the most complex and mysterious of the evolutionary innovations for feeding seen in modern sea cows was the set of dental modifications found in the Indo-Pacific dugong: 1.

2.

Most conspicuously, the cheek tooth battery has degenerated by functional loss of the enamel crowns of the teeth, which initially form but are very thin and wear off quickly. In compensation, the last two molars in each quadrant retain open, ever-growing roots (root hypsodonty), so that at maturity the animal has only pegs of dentine with flat occlusal surfaces (Marsh 1980): teeth that seem only better than no teeth at all, and whose very functionality has even been questioned (Lanyon and Sanson 2006a, b). In immature specimens, there is a tiny upper incisor that never erupts, lying anterior to the definitive (presumably first incisor) tusk (Marsh 1980). While this

12

3.

4.

D. P. Domning

seems most likely to be a deciduous first incisor, such a precursor to the permanent tusk has never been observed in any fossil sirenian. Dugongs have four pairs of shallow empty pits on the mandibular masticating surface, which are assumed to be the vestigial alveoli of the lower incisors and canines. These are covered by the lower masticating horny pad. In many adult D. dugon, however, one or more of these pits are deep and contain vestigial teeth that never erupt (e.g., Lyman 1939; Lanyon and Sanson 2006b: Fig. 1.1). Similar deep pits containing unerupted teeth are never seen in fossil sirenians that have lost the erupted lower incisors and canines. As noted above, the tusks in D. dugon are sexually dimorphic and apparently used only in males, and then only in social interactions like fighting and mating (Marsh 1980; Domning and Beatty 2007; Chap. 4). No such tusk dimorphism has been documented for any fossil sirenian (Sorbi et al. 2012).

All four of these features suggest that some heterochronic developmental process has been at work in the evolution of the dugong’s dentition. The arrested development of the female tusk and the enamel crowns of the molars, the vestigial and non-erupting upper and lower incisors, and the persistently open roots of the posterior molars may have been achieved at least in part by a neotenic mechanism (Domning 1995). All of these features together suggest the hypothesis (Domning 1995) that Dugong cheek teeth represent a desperate evolutionary attempt to salvage some chewing capacity under conditions of severely increased tooth wear that was beginning to cut into the animals’ reproductive lifespan—a plausible explanation given that lack of food reduces dugong reproductive output today (Chap. 8). The only plausible cause of this in a marine environment over a huge area (the Indo-Pacific seas) would seem to be increased runoff of abrasive siliciclastic sediment into seagrass beds due to accelerated erosion resulting from global eustatic sea level drop during a glacial period. Faced with increased wear of its molar crowns, the dugong evidently resorted to root hypsodonty of molars 2 and 3 as an alternative to total loss of a functional cheek dentition. However, the resulting abandonment of molars having enamel cutting edges made the eating of fibrous material (i.e., large rhizomes) no longer energetically feasible, so the species shifted its diet to less fibrous seagrasses and possibly a nutritional strategy based more on cell contents than on cell walls. The dugong’s present preference for the most delicate, easily chewed and digestible seagrasses (in particular, Halophila) may also be reflected in one more skeletal feature. The ribs of D. dugon, while as dense (osteosclerotic) as those of other sirenians, are noticeably slenderer (less pachyostotic): in fact, among the slenderest ribs of known fully aquatic sirenians. This reduction in ballast could facilitate resurfacing from deep diving, and unlike other sirenians, dugongs may dive as deep as 36 m (Marsh et al. 2011:157; Chap. 3) in areas where large beds of Halophila (the deepest-growing seagrass; York et al. 2015) are found at that depth. The above hypothesis would imply that the dugong’s dental degeneration occurred only after the beginning of the Pleistocene glaciations. Unfortunately, D. dugon itself has a negligible fossil record that might be used to test this idea. As it happens, however, it has a close sister taxon (arguably a congener; still undescribed

1 What Can We Infer About the Behavior of Extinct Sirenians?

13

because it is known only from one skull in a private collection) that dates from near the beginning of the Pleistocene, around 2 million years ago, and still has fully enameled molars, just as this hypothesis predicts. The most surprising thing about this specimen is that it was found not in the Indo-Pacific region, but in Florida— suggesting dispersal across the Atlantic and around the Cape of Good Hope, possibly at the same time manatees crossed to West Africa. Lest the foregoing leave the impression that the dental adaptations of all the Sirenia are well understood, attention must be paid to Miosiren, a strange Miocene trichechid from Belgium and England. Its dentition is in itself only moderately surprising (large, conical upper tusks; retained permanent premolars 3 and 4; conventional-looking enameled molars except for a reduced, peg-like third molar). But its palate and the solid bony walls that support the palate are astonishingly thick (~4 cm) (Beatty et al. 2012: Fig. 1.3), suggesting that in comparison with any other sirenian, it generated enormous occlusal forces when chewing whatever it was that it ate. It certainly seems over-engineered for eating ordinary seagrass. A diet of molluscs has been suggested and not ruled out, but stable isotopes have not so far been able to shed light on its diet, whether vegetable or animal. Calcareous algae are a possibility, or the thickened skull bones may merely have served for ballast. In any case, there are still striking mysteries to be solved in sirenian paleontology.

1.4 Sensory Perception Compared with the foregoing topics, there are few clues to sirenian sensory systems in the fossil record. There are no osteological signs of changes in sirenian vision over their evolutionary history. Neither, as far as we can tell, have the chemical senses of taste and smell changed very much. All sirenians retain cribriform plates (the sievelike bony structures at the front of the braincase through which pass the olfactory nerves). This suggests that sirenians do have some sense of smell, which is more than can seemingly be said for modern cetaceans, in which the cribriform plate is greatly reduced (in baleen whales) or absent (in toothed whales). But the chemical senses have scarcely been studied even in the living sea cows (see Chap. 2). Florida manatees reportedly lack a vomeronasal organ, but retain taste buds, some olfactory epithelium and a rudimentary olfactory bulb (Mackay-Sim et al. 1985; Marriott et al. 2013; Barboza and Larkin 2020). As described above and in Chap. 2, touch is an important sense in sirenians. Not only do they have well-developed tactile vibrissae on their snouts, but the sparse hairs all over their bodies are likewise sinus hairs, which seem to constitute an analog to the lateral-line systems of fishes (Reep et al. 2002; Gaspard et al. 2017; Chap. 2). Again, though, these have left no trace in the fossil record. Although they do not echolocate, sirenians do have sensitive hearing (Chaps. 2, , 7). Indeed, the best chance of tracing the evolution of sirenian senses is offered by the bones of the ear, which are well represented throughout the history of the order, starting with the earliest, most primitive of them all: the Early or Middle Eocene

14

D. P. Domning

sirenian reported by Benoit et al. (2013), which consists precisely of a petrosal bone. Unfortunately, the ears of fossil sea cows have hardly been studied from a functional viewpoint, but there is potential for future investigators. Even insights into underwater vocal communication may be conceivable.

1.5 Concluding Remarks There are several clear ways forward to gain further insights into the “paleoethology” of sirenians, two in particular: collecting more complete specimens of species still inadequately known and employing new techniques of analyzing the fossils already in hand. Under the first heading comes traditional paleontological fieldwork to discover new forms of sirenians previously unknown (as is happening every year), and to recover better material of known taxa. An outstanding example of a specimen already in hand and needing basic description is the manus of Steller’s sea cow: the first known skeleton that appears to preserve these bones was reportedly excavated in 2017 on Bering Island. It should have urgent priority for study by Russian morphologists. Under the second heading are studies of the inner ear by CT scanning; stable isotope and other forms of chemical analysis of teeth and bones, and biomechanical studies such as principal-components analyses of skull architecture to reconstruct the stresses caused by tusk use and mastication. Beyond these, discoveries in the decades to come will be limited only by technology and the imagination of new generations of investigators. Such progress would include inferences from fossil endocasts of sirenian brains, which have been studied since Owen (1875), especially by Edinger (1933, 1975) and most recently by Kerber and Moraes (2021). Since I lack the knowledge of neuroanatomy needed to interpret behavior from such specimens, however, I must leave this study to others.

References Anderson PK (2002) Habitat, niche, and evolution of sirenian mating systems. J Mamm Evol 9:55–98. https://doi.org/10.1023/A:1021383827946 Bajpai S, Domning DP, Das DP et al (2009) A new middle Eocene sirenian (Mammalia, Protosirenidae) from India. Neues Jahrb Geol Pal Abh 252:257–267. https://doi.org/10.1127/00777749/2009/0252-0257 Barboza M, Larkin IV (2020) Gross and microscopic anatomy of the nasal cavity, including olfactory epithelium, of the Florida manatee Trichechus manatus latirostris. Aquat Mamm 46:274–284. https://doi.org/10.1578/AM.46.3.2020.274 Beatty BL, Vitkovski T, Lambert O et al (2012) Osteological associations with unique tooth development in manatees (Trichechidae, Sirenia): a detailed look at modern Trichechus and a review of the fossil record. Anat Rec 295:1504–1512. https://doi.org/10.1002/ar.22525 Benoit J, Adnet S, El Mabrouk E et al (2013) Cranial remain from Tunisia provides new clues for the origin and evolution of Sirenia (Mammalia, Afrotheria) in Africa. PLoS ONE 8:e54307. https://doi.org/10.1371/journal.pone.0054307

1 What Can We Infer About the Behavior of Extinct Sirenians?

15

de Buffrénil V, Canoville A, D’Anastasio R et al (2010) Evolution of sirenian pachyosteosclerosis, a model-case for the study of bone structure in aquatic tetrapods. J Mamm Evol 17:101–120. https://doi.org/10.1007/s10914-010-9130-1 Clementz MT, Sorbi S, Domning DP (2009) Evidence of Cenozoic environmental and ecological change from stable isotope analysis of sirenian remains from the Tethys-Mediterranean region. Geology 37:307–310. https://doi.org/10.1130/G25533A.1 Diaz-Berenguer E, Badiola A, Moreno-Azanza M et al (2018) First adequately-known quadrupedal sirenian from Eurasia (Eocene, Bay of Biscay, Huesca, northeastern Spain). Nat Sci Rep 8:1–13. https://doi.org/10.1038/s41598-018-23355-w Domning DP (1978) Sirenian evolution in the North Pacific Ocean. Univ Calif Publ Geol Sci 118:xi + 176 Domning DP (1982) Evolution of manatees: a speculative history. J Paleont 56:599–619 Domning DP (1995) What do we know about the evolution of the dugong? In: Mermaid symposium: first international symposium on Dugong and Manatees. November 15–17, 1995, Toba, Mie, Japan. Abstracts. Toba (Japan), Toba Aquarium, pp 23–24 Domning DP (1997) Sirenia. In: Kay RF, Madden RH, Cifelli RL et al. (eds) Vertebrate paleontology in the Neotropics: the Miocene fauna of La Venta, Colombia. Smithsonian Inst., Washington & London, pp 383–391 Domning DP (2001a) The earliest known fully quadrupedal sirenian. Nature 413:625–627. https:// doi.org/10.1038/35098072 Domning DP (2001b) Sirenians, seagrasses, and Cenozoic ecological change in the Caribbean. In: Miller W III, Walker SE (eds) Cenozoic palaeobiology: the last 65 million years of biotic stasis and change. Palaeogeogr Palaeoclimatol Palaeoecol 166(Special Issue):27–50 Domning DP (2005) Fossil Sirenia of the West Atlantic and Caribbean region. VII. Pleistocene Trichechus manatus Linnaeus, 1758. J Vert Pal 25:685–701. https://doi.org/10.1671/0272-463 4(2005)025[0685:FSOTWA]2.0.CO;2 Domning DP, Beatty BL (2007) Use of tusks in feeding by dugongid sirenians: observations and tests of hypotheses. Anat Rec 290:523–538. https://doi.org/10.1002/ar.20540 Domning DP, de Buffrénil V (1991) Hydrostasis in the Sirenia: quantitative data and functional interpretations. Mar Mamm Sci 7:331–368. https://doi.org/10.1111/j.1748-7692.1991.tb00111.x Domning DP, Furusawa H (1995) Summary of taxa and distribution of Sirenia in the North Pacific Ocean. Island Arc 3:506–512. https://doi.org/10.1111/j.1440-1738.1994.tb00129.x Domning DP, Gingerich PD (1994) Protosiren smithae, new species (Mammalia, Sirenia), from the late Middle Eocene of Wadi Hitan Egypt. Contr Mus Pal Univ Michigan 29:69–87 Domning DP, Hayek LC (1984) Horizontal tooth replacement in the Amazonian manatee (Trichechus inunguis). Mammalia (Paris) 48:105–127. https://doi.org/10.1515/mamm.1984.48. 1.105 Domning DP, Hayek LC (1986) Interspecific and intraspecific morphological variation in manatees (Sirenia: Trichechus). Mar Mamm Sci 2:87–144. https://doi.org/10.1111/j.1748-7692.1986.tb0 0034.x Edinger T (1933) Über Gehirne tertiärer Sirenia Ägyptens und Mitteleuropas sowie der rezenten Seekühe. Abh Bayer Akad Wiss, math-natw Abt (n.s.) 20:5–36 Edinger T (1975) Paleoneurology 1804–1966: an annotated bibliography. Adv Anat Embryol Cell Biol 49:1–258 Gaspard JC III, Bauer GB, Mann DA et al (2017) Detection of hydrodynamic stimuli by the postcranial body of Florida manatees (Trichechus manatus latirostris). J Comp Physiol A 203:111–120. https://doi.org/10.1007/s00359-016-1142-8 Gheerbrant E, Domning DP, Tassy P (2005) Paenungulata (Sirenia, Proboscidea, Hyracoidea, and relatives). In: Rose KD, Archibald JD (eds) The rise of placental mammals: origins and relationships of the major extant clades. Johns Hopkins, Baltimore, pp 84–105 Hautier L, Sarr R, Tabuce R et al (2012) First prorastomid sirenian from Senegal (western Africa) and the old world origin of sea cows. J Vertebr Paleontol 32:1218–1222. https://doi.org/10.1080/ 02724634.2012.687421

16

D. P. Domning

Janis C, Ehrhardt D (1988) Correlation of relative muzzle width and relative incisor width with dietary preferences in ungulates. Zool J Linnean Soc 92:267–284. https://doi.org/10.1111/j.10963642.1988.tb01513.x Kerber L, Moraes MH (2021) Endocranial morphology of an early/middle Miocene South American dugong and the neurosensorial evolution of sirenians. J Mamm Evol. https://doi.org/10.1007/s10 914-021-09555-8 Kleinschmidt A (1951) Über ein Skelet und eine Rekonstruktion des äusseren Habitus der Riesenseekuh, Rhytina gigas Zimmermann 1780. Zool Anz 146:292–314 Kojeszewski T, Fish FE (2007) Swimming kinematics of the Florida manatee (Trichechus manatus latirostris): hydrodynamic analysis of an undulatory mammalian swimmer. J Exp Biol 210:2411– 2418. https://doi.org/10.1242/jeb.02790 Lanyon JM, Sanson GD (2006a) Degenerate dentition of the dugong (Dugong dugon), or why a grazer does not need teeth: morphology, occlusion and wear of mouthparts. J Zool 268:133–152. https://doi.org/10.1111/j.1469-7998.2005.00004 Lanyon JM, Sanson GD (2006b) Mechanical disruption of seagrass in the digestive tract of the dugong. J Zool 270:277–289. https://doi.org/10.1111/j.1469-7998.2006.00135.x Lanyon JM, Athousis C, Sneath HL et al (2021) Body scarring as an indicator of social function of dugong (Dugong dugon) tusks. Mar Mamm Sci 37:1–12. https://doi.org/10.1111/mms.12788 Lyman CP (1939) A vestigial lower incisor in the dugong. J Mammal 20:229–231 Mackay-Sim A, Duvall D, Graves BM (1985) The West Indian manatee (Trichechus manatus) lacks a vomeronasal organ. Brain Behav Evol 27:186–194. https://doi.org/10.1159/000118729 Marriott S, Cowan E, Cohen J (2013) Somatosensation, echolocation, and underwater sniffing: adaptations allow mammals without traditional olfactory capabilities to forage for food underwater. Zool Sci 30:69–75. https://doi.org/10.2108/zsj.30.69 Marsh H (1980) Age determination of the dugong (Dugong dugon (Müller)) in northern Australia and its biological implications. In: Perrin WF Myrick, AC Jr (eds) Age determination of toothed whales and sirenians. Reports International Whaling Commission 3(Special Issue):181–201 Marsh H, O’Shea TJ, Reynolds JE III (2011) Ecology and conservation of the Sirenia: dugongs and manatees. Cambridge University, Cambridge (U.K.) (Conservation Biology Series No. 18):xvi + 521 Marsh H, Grech A, McMahon K (2018) Dugongs: Seagrass Community Specialists. In: Larkum A, Kendrick G, Ralph P. (eds) Seagrasses of Australia. Springer, Cham. https://doi.org/10.1007/ 978-3-319-71354-0_19 Owen R (1875) On fossil evidences of a sirenian mammal (Eotherium aegyptiacum, Owen) from the Nummulitic Eocene of the Mokattam Cliffs, near Cairo. Q J Geol Soc London 31:100–105. https://doi.org/10.1144/GSL.JGS.1875.031.01-04.05 Perini FA, Nascimento ER, Cozzuol MA (2020) New species of Trichechus Linnaeus, 1758 (Sirenia, Trichechidae), from the upper Pleistocene of southwestern Amazonia, and the evolution of Amazonian manatees. J Vertebr Paleontol 39:e1697882. https://doi.org/10.1080/02724634.2019. 1697882 Promus J (1937) Netting dugong. Walkabout 3:40–41 Reep RL, Stoll ML, Marshall CD et al (2001) Microanatomy of facial vibrissae in the Florida manatee: the basis for specialized sensory function and oripulation. Brain Behav Evol 58:1–14. https://doi.org/10.1159/000047257 Reep RL, Marshall CD, Stoll ML (2002) Tactile hairs on the postcranial body in Florida manatees: a mammalian lateral line? Brain Behav Evol 59:141–154. https://doi.org/10.1159/000064161 Savage RJG, Domning DP, Thewissen JGM (1994) Fossil Sirenia of the West Atlantic and Caribbean region. V. The most primitive known sirenian, Prorastomus sirenoides Owen, 1855. J Vertebr Paleontol 14:427–449. https://doi.org/10.1080/02724634.1994.10011569 Sickenberg O (1934) Beiträge zur Kenntnis tertiärer Sirenen. I. Die eozänen Sirenen des Mittelmeergebietes. II. Die Sirenen des belgischen Tertiärs. Mém Mus Roy Hist Nat Belgique 63:1–352

1 What Can We Infer About the Behavior of Extinct Sirenians?

17

Sorbi S, Domning DP, Vaiani SC et al (2012) Metaxytherium subapenninum (Bruno, 1839) (Mammalia, Dugongidae), the latest sirenian of the Mediterranean basin. J Vertebr Paleontol 32:686–707. https://doi.org/10.1080/02724634.2012.659100 Steller GW (1751) De bestiis marinis. Novi Comm Acad Sci Petropolitanae 2:289–398 Sunter GH (1937) Adventures of a trepang fisher. London, Hurst & Blackett Ltd., pp 1–288 Takahashi S, Domning DP, Saito T (1986) Dusisiren dewana, n. sp. (Mammalia: Sirenia), a new ancestor of Steller’s sea cow from the Upper Miocene of Yamagata Prefecture, northeastern Japan. Trans Proc Pal Soc Japan (n.s.) 141:296–321. https://doi.org/10.14825/prpsj1951.1986.141_296 York PH, Carter AB, Chartrand K et al (2015) Dynamics of a deep-water seagrass population on the Great Barrier Reef: annual occurrence and response to a major dredging program. Sci Rep 5:13167. https://doi.org/10.1038/srep13167 Zalmout IS, Gingerich PD (2012) Late Eocene sea cows (Mammalia, Sirenia) from Wadi Al Hitan in the Western Desert of Fayum, Egypt. University of Michigan Papers on Paleontology 37:xiii + 158

Chapter 2

Morphological and Sensory Innovations for an Aquatic Lifestyle Christopher D. Marshall, Diana K. Sarko, and Roger L. Reep

Abstract Sirenians have evolved novel innovations relative to other terrestrial and marine mammals for life as aquatic herbivores. The study of their natural history and adaptations provides insights into the range of possibilities in mammalian evolution, including their ethology and behavioral ecology. Their large body size accommodates an expanded digestive system necessary to process the large amounts of food ingested, and it also confers thermal advantages and protection from predation. Other thermoregulatory adaptations include a divergent blubber arrangement, dense heavy skin, and a series of counter-current heat exchangers to balance both heat loss and heat gain. Due to their herbivorous niche in relatively shallow environments, sirenians have evolved an unusual skeletal system and arrangement of their lungs, diaphragm, and digestive system, and a re-arrangement of the thoracic and abdominal cavities, that enable easy transitions from the benthic substrate, where their food is often located, to the surface where air is inhaled. Their hydrostasis allows for precise control over buoyancy, which in turn reduces energetic costs of movement. Their mode of food acquisition involves both the sensory and motor functions of facial vibrissae. This muscular-vibrissal complex is capable of numerous varied and detailed movements due to the hypertrophy of muscles into a muscular hydrostat or shortened elephantine trunk-like muzzle. Their pachyosteosclerotic bones, that function so well as part of their buoyancy control and hydrostasis system, are also brittle, like a ceramic material, and prone to fracture. Sirenian perception of the aquatic environment is largely through somatosensation (touch and hydrodynamic reception) and hearing, although vision and taste (chemoreception) are also important to some degree. Sirenians are one of a few mammalian groups in which all hairs on the body are sensory hairs that mediate exquisitely sensitive hydrodynamic reception C. D. Marshall (B) Texas A&M University, Galveston Campus, 200 Seawolf Parkway, Galveston, TX 77553, USA e-mail: [email protected] D. K. Sarko Southern Illinois University School of Medicine, 1135 Lincoln Drive, Carbondale, IL 62901, USA R. L. Reep University of Florida, College of Veterinary Medicine, 2015 SW 16th Avenue, Gainesville, FL 32611, USA © Springer Nature Switzerland AG 2022 H. Marsh (ed.), Ethology and Behavioral Ecology of Sirenia, Ethology and Behavioral Ecology of Marine Mammals, https://doi.org/10.1007/978-3-030-90742-6_2

19

20

C. D. Marshall et al.

that is analogous to the function of the lateral line system in fish and amphibians. This mode of reception likely plays a role in sirenian spatial orientation and navigation. Novel structures in the brain (Rindenkerne) are likely responsible for information processing of touch and hydrodynamic reception and provide a substrate for multi-modal sensory perception capabilities. Rindenkerne may also play a key role in mediating interesting behaviors such as synchronous breathing of groups when sleeping. These innovations are important in terms of sirenian conservation and form the basis for much of the sirenian ethology and behavioral ecology observed. Sirenians are a special group of mammals with unusual and interesting morphological and sensory innovations for aquatic life that we are just beginning to understand and explore. Keywords Body size · Buoyancy · Counter-current heat exchange · Hemidiaphragms · Hydrodynamic reception · Hydrostasis · Lateral line · Metabolic rate · Muscular-vibrissal complex · Oripulation · Sensory hairs · Sirenian · Vocalizations

2.1 Introduction Evolution allows us to explore the possibilities in which animals survive. For mobile animals this involves an intricate and ongoing interplay among factors that influence finding food, movement, communicating, reproducing, thermoregulating, and avoiding predation and disease. Success in these activities depends on adaptive behavior, and behavior is shaped and constrained by anatomical and functional innovations: “Structure without function is a corpse; function without structure is a ghost” (Vogel and Wainwright 1969). When we observe a manatee or dugong swimming, feeding, or resting, we are witnessing behavior that is the result of millions of years of shaping through adaptations related to aquatic herbivory, buoyancy and locomotion, thermoregulation in the aquatic environment, communication with conspecifics, and the need to conserve energy. We are also seeing the result of interplay between these adaptations, and those of the peripheral and central nervous systems, that determine what is perceived, and the variety of possible behavioral responses. In this chapter, we present behavioral, anatomical, and physiological data that describe these adaptations, and we suggest how they inform hypotheses about the range and trajectory of sirenian ethology and behavioral ecology. As explained in Chap. 1, sirenians have a long evolutionary history that spans 50 MA, equivalent to that of cetaceans. The first sirenians were quadrupeds that slowly transitioned from land to aquatic life and became obligate marine mammals. Sirenians represent a novel evolutionary line of mammals, divergent from terrestrial mammals and from other aquatic mammals. They possess a suite of morphological and sensory innovations for aquatic life. Even their basic innovations for life in an aquatic environment, such as body organization, thermoregulation, buoyancy, and hydrostasis, are novel compared to other marine mammals. Life in the water

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

21

requires a re-configuration of terrestrial sensory systems, and many aspects of the sirenian central and peripheral nervous systems are unusual compared to those of most terrestrial and other marine mammals. Their role as aquatic herbivores requires sirenians to process large quantities of relatively low-quality food, compared to carnivores. Large body size facilitates the accommodation of a large digestive tract and is also advantageous for thermal efficiency. Although any mammal living in water must deal with thermoregulatory and locomotor demands, additional sirenian innovations relate to feeding as marine mammal herbivores, digestion, and the sensory capabilities to find those food resources. In this chapter, we address evidence that distinguishes between the two extant sirenian families, the Trichechidae (manatees) and Dugongidae (dugongs). Despite the fact that both groups are marine mammal herbivores, manatees and dugongs are quite different in their trophic ecology, feeding mechanisms, and in many instances their natural history, ethology, and behavioral ecology. Overall, manatees tend to be generalists whereas dugongs are benthic specialists (Chap. 3). Manatees variously inhabit quiet, riverine, estuarine, and coastal areas, whereas dugongs are obligatorily marine and venture further away from coastal areas than manatees. The exception is the Amazonian manatee (Trichechus inunguis), which stays within freshwater riverine systems of the Amazonian tributaries (Marsh et al. 2011). Although similar in fusiform body shape, manatee bodies are built differently from dugongs. Manatees tend to be longer and more massive. From the frontal perspective, manatees are wider than tall, whereas dugongs are taller than wide. Again, Amazonian manatees differ in being the smallest sirenian (Marsh et al. 2011). Manatees propel themselves through water using a caudal paddle, whereas dugongs generate thrust using flukes. These simple morphological observations contribute to different lifestyles, behaviors, and ecology.

2.2 Morphological Innovations for an Aquatic Life 2.2.1 Thermoregulation One of the most basic adaptations for any endothermic vertebrate living in an aquatic habitat is to grow to a large size. Even sea otters, the smallest of marine mammals, are among the largest otters in terms of mass. Thermal inertia through increased body size is a powerful thermoregulatory innovation. Since the thermal conductivity of water is 25 times that of air (Pabst et al. 1999; Marshall 2002), maintaining core body temperatures in an aquatic environment can be challenging. Aquatic life frees mammals from the constraints of gravity, allowing larger body sizes to be attained. As an animal gets larger, and the linear body dimension increases, surface area increases proportionately to the second power while volume increases proportionately to the third power. Therefore, volume increases faster than surface area as an animal becomes larger. Since heat is lost at the interface with the environment, increasing the volume to surface area ratio through body growth decreases heat loss

22

C. D. Marshall et al.

and is an effective way to maintain body temperature (Innes et al. 1990; Elsner 2007). In sirenians, large body size serves two basic functions: thermoregulatory efficiency, and the ability to house a large gastrointestinal system that is necessary for processing the large amounts of food associated with aquatic herbivory. Cold-adapted marine mammals use blubber, a dense pelage, or both (Liwanag et al. 2012) to minimize heat loss. Sirenians and cetaceans have secondarily lost their pelage, which provides thermal insulation in pinnipeds, sea otters and polar bears. In sirenians, the remaining hairs are all specialized sensory hairs (Reep et al. 2001, 2002, 2011; see Sect. 2.3). Instead, sirenians and cetaceans invest in a blubber layer. Skin is comprised of an outer epidermis, a middle dermis, and a deeper hypodermis. Blubber is the enlargement of the hypodermis and associated adipose tissue and provides less insulation than fur (Scholander et al. 1950); fat transmits heat three to five times faster than a dry, high quality pelage (Schmidt-Nielsen 1990; Pabst et al. 1999). However, the insulating properties of blubber depend strongly upon both its thickness and its lipid content, which impacts its insulative properties. Thick blubber increases the distance between the body cavity and the water. The lipid content of the adipose tissue in many marine mammals is seasonally variable and species-specific but can vary between 9 and 82% of wet blubber weight (Lockyer et al. 1985; Worthy and Edwards 1990). Blubber that is thick and comprised of high-density lipids with low perfusion of blood to the tissues increases its insulative properties (Lockyer et al. 1985; Worthy and Edwards 1990; Kvadsheim and Aarseth 2002). Due to their tropical to sub-tropical distribution, large body size is typically enough for manatees to stay within their thermal neutral zone. Florida manatees (Trichechus manatus latirostris), which are at the northern-most limit of their range, tend to be much larger in body size, specifically their girth, compared to their Caribbean subspecific relatives (Trichechus manatus manatus; Rommel et al. 2003) and Amazonian manatees (Marsh et al. 2011). However, in Florida, where water temperatures may drop below 16–18 °C seasonally, manatees migrate to warm-water refugia (Laist and Reynolds 2005; Deutsch et al. 2006; Chaps. 5 and 6). If warm water is not attainable, body condition can be compromised (Ward-Geiger 1997) and manatees may ultimately exhibit cold stress syndrome (Bossart et al. 2002). Manatees tend to be less active than dugongs, likely have a lower metabolic rate (Gallivan and Best 1980; Irvine 1983; Miculka and Worthy 1995; Lanyon et al. 2006) and have a poor thermoregulatory capacity (Gallivan et al. 1983; Marsh et al. 2011), making them more susceptible to cold stress syndrome. In fact, Horgan et al. (2014) consider that dugongs may not be susceptible to cold stress syndrome, a conclusion that is contested by Owen et al. (2013). In addition to body size, sirenian skin also provides insulative properties. In sirenians, the hypodermis includes a thin layer of blubber, a thin sheet of muscle (cutaneous trunci), and a second, deeper, but still thin, inner layer of blubber (Bonde et. al. 1983; Eros et al. 2007). Manatee blubber tends to be thicker relative to that of dugongs (Nichols 2005). Blubber layers in dugongs, and likely manatees, are not particularly lipid-dense, but in combination with the dermis, they afford sirenians integumentary insulation similar to warm-water dolphins of similar body size (Horgan et al. 2014). However, dugongs living at high latitudes of their range (e.g., Okinawa, the Arabian Gulf, and southern portions of Australia)

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

23

are also limited by colder water and have been observed to engage in behavioral thermoregulation (Zeh et al. 2018; Chaps. 5 and 6). Another thermoregulatory mechanism is to recycle heat produced by body functions (e.g., muscle contraction and digestion) using counter-current heat exchangers (CCHE). In marine mammals, these are particularly well developed in locations that have little-to-no blubbers, such as the dorsal fins, pectoral limbs, and flukes. In a CCHE, the arterial supply (central arteries) is surrounded by numerous veins (circumarterial veins). Typically, heat from the warm arterial blood is transferred to cool venous blood returning to the heart. The result is that arterial blood reaching the periphery is significantly cooler and therefore less heat is lost to the environment. Conversely, cool venous blood traveling from the periphery is warmed and therefore does not significantly reduce core body temperature. Florida manatees, and likely all sirenians, possess CCHEs in the axillary region, where the brachial plexus and major blood supply to the forelimbs are located (Murie 1872; Fawcett 1942), in the region supplying the tail (Rommel and Caplan 2003), and around the intra-abdominal testes to maintain their temperature below body temperature for spermatogenesis (Rommel et al. 2001). A study of heat flux in two captive Florida manatees (Erdsack et al. 2018), in which 42 anatomical sites were measured, demonstrated that heat flux was significantly greater at the least insulated body parts (the flippers, fluke, and the head) than on the trunk. However, most heat fluxes occurred in the axilla region, and heat loss was greater in summer than in winter. Counter-current heat exchangers can be bypassed by dilation of the central artery, which compresses and collapses the circumarterial veins. Venous blood flow is then re-routed to veins further away from the central artery and closer to the surface of the skin. This allows excess heat to be dumped into the environment if needed. So, although CCHEs are often thought of as a mechanism to retain heat, they can be used to dump heat as well. They are an amazing thermoregulatory innovation. These thermoregulatory adaptations contribute to the ability of manatees to thrive in the aquatic environment despite generally having a low metabolic rate. However, sirenians do not have extensive reserves, so their behavior is geared toward limiting expenditure of energy. This need contributes to their typically slow-paced movements. Manatees are often pushed to their physiological thermal limits in Florida, where periodic cold weather necessitates choices between resting to conserve energy or foraging to obtain more energy resources. This is likely true for any sirenian living at higher latitudes.

2.2.2 Hydrostasis Sirenians exhibit a suite of innovations for maintaining their equilibrium, or horizontal trim, in the water. This is known as hydrostasis (Domning and Buffrénil 1991). All sirenians spend much of their time traveling between their food sources and the surface to breathe. As explained in Chap. 3, food sources vary from being

24

C. D. Marshall et al.

predominantly floating surface meadows (Amazonian manatees) to almost exclusively benthic(dugongs). Adaptations for buoyancy and trim can be found in changes in the skeletal, respiratory, integumentary, and digestive systems. It is advantageous for most sirenians to be slightly negatively buoyant to facilitate benthic foraging, although Amazonian manatees may be an exception since they feed predominantly at the surface (Domning 1980). All sirenians have a derived anatomical arrangement of the lungs and diaphragm compared to other marine and terrestrial mammals. In the typical terrestrial mammalian condition, the diaphragm divides the body into front and back (anterior versus posterior) portions. However, in sirenians, the diaphragm lies in a horizontal plane, divides the body into upper and lower (dorsal and ventral) portions, and does not attach to the sternum (Rommel and Reynolds 2000; Eros et al. 2007). Whereas in most mammals the diaphragm separates the heart from the abdominal cavity, in sirenians a transverse septum (a separate structure) separates the heart from the abdominal cavity (Rommel and Reynolds 2000; Eros et al. 2007). The diaphragm forms two anatomically and functionally distinct hemidiaphragms that adjust trim and attitude in the water. As a result, the thoracic cavities where the lungs are located span most of the animal’s body length (Rommel and Reynolds 2000; Eros et al. 2007) and greatly change the possibilities of hydrostasis (Domning and Buffrénil 1991). Rommel and Reynolds (2000) hypothesized that the muscularized hemidiaphragms allow for independent control of the volume of each pleural cavity, thereby allowing sirenians to adjust their buoyancy, roll, and pitch in the water column. Further, they suggested that methane gas generated by the large digestive tract may be compressed due to the contractile action of the hemidiaphragms and the abdominal muscles, providing an additional mechanism for buoyancy control. These mechanisms likely underlie the observed fine adjustments in buoyancy, pitch, and roll that are seen during slow swimming, rising to the surface to breathe, and sinking back down toward the substrate (Hartman 1979).

2.2.3 Buoyancy Innovations for buoyancy and hydrostasis in sirenians include the skeleton and skin. Thick blubber can be a potential problem in many cold-adapted marine mammals due to increased buoyancy. Materials that are less dense than water (blubber, lungs) float; materials that are denser than water (bone, muscle) sink (Pabst et al. 1999; Marshall 2002). Sirenians are typically denser than water and tend to sink, although this may not have been the case in the now extinct Steller’s sea cow (Hydrodamalis gigas) (Marsh et al. 2011; Chap. 1). Manatees and dugongs possess pachyosteosclerotic ribs. These bones are thickened (pachyostotic) and the trabecular bone is replaced with dense compact (osteosclerotic) bone (Buffrénil and Schoevaert 1989; Domning and Buffrénil 1991) that is also highly mineralized (Yan et al. 2006; Clifton et al. 2008). The arrangement of this dense ribcage around the body’s center of mass creates ballast that is countered by the unusual arrangement of the lungs and diaphragm.

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

25

These traits for precise hydrostasis and buoyancy make manatees and dugongs more vulnerable to boat strikes (Reynolds and Marshall 2012). This is because sirenian long bone microstructure results in material properties in which the bone behaves as a quasi-brittle solid, similar to ceramic material (Clifton et al. 2008), a feature which has consequences for the capacity of a sirenian to survive blunt trauma such as vessel strike (Chap. 7). Mammalian long bones typically contain a marrow-filled cavity that is the primary center for red and white blood cell and platelet production (collectively called hemopoiesis). A consequence of using long bones as ballast is the loss of this function. Instead, hemopoiesis in West Indian manatees (and presumably all sirenians) occurs in the vertebrae, which are less dense than the long bones (Bazzini et al. 1986). In addition, sirenian skin also contributes to hydrostasis and buoyancy. The skin of manatees is ~2.5 thicker for its body mass than that of other mammals and is reinforced with a dense collagen network (Nill et al. 1999). The mass and density of manatee skin are denser than some small cetaceans and produce negative buoyancy (Nill et al. 1999; Kipps et al. 2002). Collectively, sirenian skin, pachyosteosclerotic bones, long lungs with independent hemidiaphragms, and a large digestive tract, form a functional complex for hydrostasis. This capability reduces energy expenditure when moving between benthic habitats and the surface. Manatees spend hours feeding and breathing in shallow water, moving slowly from the bottom to top to breathe, and back to the bottom at near-neutral buoyancy. This is critical due to their low metabolic rate. Dugongs can also forage in this manner, but also inhabit deeper water in which they dive to the bottom to forage and swim back to the surface to breathe (Chap. 3).

2.2.4 Swimming Behavior, Kinematics, and Performance Hydrostasis and postcranial musculoskeletal innovations constrain and define how sirenians swim. The sirenian postcranial skeleton is greatly modified compared to the dog, the species that is commonly used as an anatomical baseline for mammals (Evans and De Lahunta 2013). Only rudiments of the pelvis remain and the hindlimbs have been lost (Buchholtz et al. 2007). Sirenians possess modified forelimbs that are supported by bones of the forelimb and manus (Bonde et al. 1983; Marshall 2002; Eros et al. 2007). The sirenian forelimb is flattened and the humerus is short. The flattened bones form the basis of the pectoral flipper shape and are hydrodynamic. These flippers can generate thrust but are typically used to assist in maneuvering and stability behaviors, as well as hydrostasis. Manatees are unusual in that they are one of only a few taxa that have six cervical vertebrae; seven are typical among mammals (Domning 2000; Buchholtz et al. 2007). The cervical vertebrae are anteriorly-posteriorly compressed, an arrangement that shortens the neck and aids in streamlining the body. The cervical and thoracic vertebrae of sirenians are greater in height (dorsal-ventral) than in length (anterior-posterior), which stiffens the anterior body (Domning and Buffrénil 1991;

26

C. D. Marshall et al.

Buchholtz et al. 2007). Thrust is an important function of the postcranial musculoskeletal system; it can be increased through high mechanical advantage via muscle attachment to vertebral attachments and by generating large forces through large muscle masses. Compared to the dog, the spinous and transverse processes of sirenian thoracic and lumbar vertebrae are long, which increases the surface area for attachment of large epaxial and hypaxial muscles and increases mechanical advantage. However, compared to odontocetes, only the vertebral transverse processes of sirenian lumbar (caudal portion of dorsal vertebrae sensu Buchholtz et al. 2007) and caudal vertebrae are relatively enlarged (Buchholtz et al. 2007); their vertebral column likely functions in a manner similar to cetaceans, but with relatively decreased power and a smaller range of motion. Increased force from muscle contraction can be generated by increasing the number of muscle fibers in parallel (i.e., increased cross-sectional area). Muscle force is directly proportional to the physiological crosssectional area of that muscle (i.e., the cross-section perpendicular to the fibers). In sirenians, the centra (body) of the lumbar and caudal vertebrae are longer than they are high, which allows a greater range of motion in the dorsal-ventral direction. The central, and more so the posterior vertebral column, functions as a variably flexible beam. This beam is bent dorsally by epaxial muscles and ventrally by hypaxial muscles (Pabst et al. 1999). These morphological and biomechanical traits in manatees manifest in a different type of locomotion than found in other marine mammals, especially cetaceans and dugongs. Manatees exhibit an undulatory type of locomotion (Hartman 1979; Kojeszewski and Fish 2007), rather than the oscillatory locomotion observed in cetaceans and dugongs. Aquatic locomotion in which the axial skeleton is used to produce thrust is characterized by the degree of vertebral column bending used to produce that thrust. Thus, anguilliform (eel-like) locomotion uses a sinusoidal wave that encompasses the majority of the vertebral column. Progressively less of the vertebral column is used in sub-carangiform, carangiform, and thunniform (tuna-like) locomotion. Thunniform axial bending only involves the peduncle and caudal fin. It is useful to apply these terms to marine mammal locomotion. Manatees tend to use subcarangiform locomotion, whereas dugongs (and more so, cetaceans) limit the flexing of the axial skeleton as is more typical of carangiform, and possibly thunniform, locomotion (Fish 2001). The undulatory swimming mode of manatees is accomplished by a dorso-ventrally oriented traveling wave passing posteriorly along the body and caudal paddle-shaped tail (Kojeszewski and Fish 2007), where the force generated is imparted to the water as thrust. In captivity, Florida manatees swimming underwater were reported to reach velocities of 0.06–1.14·ms− 1 (0.2–4.1 km/hr) (Kojeszewski and Fish 2007). This is similar to data for wild manatees, which are reported to swim between 0.6 and 13.3 ms−1 but typically cruise at slow speeds (0.8 to 1.9 ms−1 ) (2.9–6.8 km/hr) (Hartman 1979). Dugongs have been reported to cruise at 2.67 ms−1 (9.6 km/hr) or more (Jarman 1966). In the wild, both manatees and dugongs are reported to swim up to 6 ms−1 (21.6 km/hr) if fleeing from a predator (dugongs), or if disturbed (Hartman 1979; Nishiwaki and Marsh 1985; Chap. 6). Although there are fewer locomotion data for dugongs compared to manatees, it is clear that dugongs are more streamlined than manatees (increased fineness ratio;

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

27

Fish and Hui 1991), are more active and mobile, and have a fluke-like tail similar to dolphins. The morphological and functional differences among the tails, or propulsors, of manatees versus dugongs, are interesting. In general, thrust increases as the span of a paddle or fluke increases, but the drag also increases with greater surface area in proportion to increased thrust (Webb 1978; Kojeszewski and Fish 2007). Therefore, the flukes of cetaceans, and presumably dugongs, can achieve faster speeds with less drag by having a higher aspect ratio (Fish and Hui 1991; Kojeszewski and Fish 2007).

2.2.5 Feeding Innovations For sirenians, many innovations involve acquiring food resources. Sirenians spend 5–8 h a day foraging, consume a wide variety of vegetation, and consume up to ~10% of their body weight per day (Hartman 1979; Best 1981; Bengtson 1983; Etheridge et al. 1985; Ledder 1986; Hurst and Beck 1988; Preen 1992, 1995; Marsh et al. 2011; Chap. 3). There are several morphological and behavioral trends for feeding among sirenians that reflect their varying foraging ecologies. Food preferences are suspected but have not been systematically studied, especially in manatees (Chap. 3); likewise for the role of taste, as mentioned in the section on gustation below. Basic anatomical trends, such as rostral deflection, are important in understanding sirenian foraging behavior and ecology, as explained in Chaps. 1 and 3. Snout deflection is indicative of their preferred feeding location within the water column (Domning 1980; Velez-Juarbe et al. 2012). Amazonian manatees have among the least deflected rostra of modern sirenians. These freshwater manatees feed primarily at the surface upon floating vegetation (Poaceae; Best 1981; Rosas 1994; Chap. 3). Until recently it was thought that African manatees also primarily consume floating vegetation. However, recent work by Keith Diagne (2014) has shown that they also consume fish and freshwater mollusks, demonstrating that they are more versatile in their feeding habits than previously thought. Both of these manatee species inhabit turbid habitats where submerged aquatic plants are not widely supported (Best 1981; Keith Diagne 2014; Chap. 3). Concomitantly, their rostra are the least deflected (~25–42° and 15–40°, respectively; Chap. 1). Dugong rostra are the most deflected of modern sirenians (70°) and this corresponds to their benthic foraging niche (Domning 1982: Chap. 1). Both West Indian manatees (Ledder 1986; Lefebvre et al. 2000; Marsh et al. 2011) and dugongs may consume above- and below-ground biomass, although excavating behavior is more typical of dugongs (Heinsohn et al. 1977; Anderson and Birtles 1978; Preen 1992, 1995; Aragones 1994; Chap. 3). Rhizomes are a rich source of carbohydrates in many seagrasses (Sheppard et al. 2007), and obtaining these nutrients is likely an aim of excavation behavior. In contrast, West Indian manatees consume 60+ species of aquatic vegetation (Hartman 1979; Chap. 3) including seagrasses, brackish and freshwater submerged aquatic vegetation, and some terrestrial grasses (Hartman 1979; Chap. 3). Since they feed anywhere in the water column (benthic, mid-water, floating, emergent, and terrestrial; Domning

28

C. D. Marshall et al.

1980), it is not surprising that their rostral deflection falls between that of Amazonian manatees and dugongs (29–52°: Chap. 1).

2.2.5.1

Muscular-Vibrissal Complex—The Sirenian Feeding Apparatus

Another important sirenian innovation for foraging is the gathering apparatus. Following Hiiemae’s (2000) “process model,” feeding is broken down into five steps: (1) ingestion, (2) transport to teeth, (3) mastication and manipulation, (4) transport to pharynx, and (5) deglutition (swallowing). Although many marine mammals alter this model, sirenians follow the model fairly closely (Marshall and Pyenson 2019), using novel innovations of the snout and muzzle. Manatees and dugongs are tactile animals, perceiving the world with their sense of touch, typically through the muzzle and mouth. Manatees and dugongs possess groups of vibrissae (sensory hairs on the face), or bristle fields around the mouth (perioral bristle fields; Marshall et al. 1998a, 2003; Reep et al. 1998; Fig. 2.1). Sirenian bristles are short, thick, and organized in a series of six fields located on both the broad and expanded upper and lower lip margins (Reep et al. 1998; Marshall et al. 2003). These bristle fields are used to both sense and manipulate vegetation (sensory and motor functions). The orofacial muscles of all sirenians are hypertrophied (Domning 1977, 1978; Marshall et al. 1998b) and form a muscular hydrostat (Kier and Smith 1985), a type of supporting structure that does not rely upon the antagonistic actions of muscles in a lever-based, hardened-skeletal system. Instead, such support in muscular hydrostats is provided by the fluid within the muscles themselves. Common examples of muscular hydrostats include tentacles of squid, trunks of elephants, and tongues and lips of mammals. Due to the lack of a hard support element (e.g., long bones), a muscular hydrostat has greater freedom of movement, and those movements are intricate, highly coordinated, and complex compared to a lever-based skeletal support system (Kier and Smith 1985; Marshall 2016). The facial muscles of sirenians, which form the muscular hydrostat (or shortened trunk), Fig. 2.1 Perioral bristle fields of living sirenians. Note the differences in distribution of the U1 bristle fields between manatees and dugongs but the homologous positions of the U2-U4 and L1-L2 fields (from Marshall et al. 2003)

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

29

enclose, insert and manipulate the six perioral bristle fields and are responsible for their movement. These muscles also define the perioral anatomy, such as the oral disk and supradisk, and are responsible for the “flare” behavior that everts the large perioral bristles (Marshall et al. 1998a, b). The dorsal and ventral buccal branches of the facial nerve (CN VII) innervate the facial musculature and are involved in bristle eversion and other movements that constitute feeding behavior (Reep et al. 1998; Marshall et al. 1998a). The motor neurons that supply CN VII, and innervate the orofacial muscles in all mammals, are located within the brainstem and within the facial motor nucleus (Marshall et al. 2005). As in the somatosensory system (see below), variation of subnuclear organization within the facial motor nucleus, such as number of subnuclei, consistency of subnuclear boundaries, and size and distribution of neuronal pools, can be used to assess functional importance. Anatomical and functional divisions of orofacial musculature for feeding are well known among terrestrial mammals. Direct experimental evidence from pigs (Sus scrofa; Marshall et al. 2005) can be used to infer function from histological studies in Florida manatees (Marshall et al. 2007). In pigs, the upper lip (orbicularis oris) is divided anatomically, neuroanatomically, and functionally (Marshall et al. 2005). The anatomical and functional separation of the upper lip into rostral and caudal components corresponds to segregations of rostral and caudal neuronal pools of motor neurons. Furthermore, the motor neuron pools of all pig facial muscles are well organized and display a somatotopic organization within the facial motor nucleus. The segregation and somatotopy of motor neuron pools innervating orofacial muscles indicate functional independence and importance in terms of freedom of movement. Such functional data can be used to infer a similar function in the facial motor nucleus of Florida manatees. Florida manatee facial motor nucleus also exhibits distinct subnuclei that are somatotopically arranged (Marshall et al. 2007), suggesting precise control over orofacial musculature, and increased degrees of freedom of movement for feeding. A model of how specific muscle activity is used in conjunction with perioral bristles for manipulating vegetation (Marshall et al. 1998b) demonstrates that the acquisition of plant material by sirenians is unique among mammals. This muscular-vibrissal complex is an effective and forceful plant-gathering apparatus capable of handling a variety of plant morphologies (Marshall et al. 2000). Bristles are used to crop or clip the blades of aquatic vegetation. Dugongs use their bristles to excavate the root system (below-ground biomass), and rhizomes, from the benthic substrate (Figs. 2.1, 2.2). This action leaves a signature “feeding trail” on the sea floor (Anderson and Birtles 1978; Preen 1995; Chap. 3). Florida manatees also occasionally dig pits to consume rhizomes (Ledder 1986; Lefebvre et al. 2000; Marsh et al. 2011). The gathering apparatus of manatees is divergent from dugongs and its prehensilelike behavior (not shared by dugongs) allows them to handle a much wider diversity of vegetative morphologies (Fig. 2.3; Marshall et al. 1998a, 2000, 2003). The gathering apparatus (muscular-vibrissal complex) allows all sirenians to ingest food and transport that food to the teeth. During studies at Homosassa Springs (Florida, USA) on the feeding behavior of Florida manatees (Marshall et al. 1998a), manatees were observed untying knots in ropes that helped keep research equipment in place;

30

C. D. Marshall et al.

Fig. 2.2 Dugong on feeding platform constructed in Toba Aquarium to investigate the use of perioral bristles for cropping and excavating seagrasses (from Marshall et al. 2003)

a testament to the dexterity of the vibrissal-muscular complex and its prehensile behavior. Sirenians differ from other marine mammals in that they masticate their food. Some oral processing (chopping) occurs in pinnipeds, odontocetes, and sea otters, but once food is in the oral cavity, manatees and dugongs spend time masticating plants to break open their cellular components, rather than swallowing food whole (Marshall and Goldbogen 2016; Marshall and Pyenson 2019). Mastication reduces particle size, increases surface area, and ruptures the tough plant cell walls. In dugongs, the rugose upper and lower keratinized palatal pads likely perform much of this function for several reasons. The seagrasses that comprise most of the dugong diet are easily disrupted (Lanyon and Sanson 2006a); some evidence suggests that dugongid cheekteeth are not functional and that all food processing occurs through the palatal pads which also transport food to the teeth (Lanyon and Sanson 2006a, b). Although trichechids also possess keratinized palatal pads, they are less robust and cover a smaller surface area compared to dugongs (Marsh et al. 1999). There is an interesting biomechanical and functional trade off between tooth and palatal pad functional morphology, that likely impacts sirenian trophic ecology and foraging behavior, that needs further exploration.

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

31

Fig. 2.3 Grasping apparatus of manatees viewed frontally. a–e: Amazonian manatee; f–j: West Indian manatee; and k–o: West African manatee. Each column shows the grasping behavior of the U2 bristle fields and the sweeping movement of the single L1 field. See Fig. 2.1 for positions of bristle fields

32

C. D. Marshall et al.

Herbivores experience extensive tooth wear compared to carnivores. Among sirenians, trichechids experience relatively more tooth wear because their diet includes floating meadows of true grasses and emergent terrestrial grasses that contain silica phytoliths, which are abrasive. Florida manatees, in particular, suffer excessive wear that drastically reduces the number of functioning teeth in many adults. This is likely due to the fact that much of the food available to them in Florida (unlike most areas inhabited by T. manatus) consists of seagrasses growing on quartz-sand bottoms (in contrast to the less abrasive carbonate sand in much of the Caribbean). In addition to its cold winters, this is another way in which Florida constitutes marginal habitat for manatees. (Goulbourne et al. [In Press]). In contrast, dugongs feed almost entirely on less abrasive seagrasses (Marsh et al. 2011). Manatees possess only cheek-teeth, whereas dugongs have incisor tusks in addition to cheek-teeth. Incisors of varying sizes have been a recurring feature in dugongid evolution, perhaps as a tool to excavate rhizomes, though extant dugongs do not appear to rely upon them for this purpose (Domning and Beatty 2007), and there is evidence that they have social functions (Chaps. 1 and 4). The distinction between premolars and molars in manatees has been lost and these teeth are referred to as cheek-teeth. Manatees typically have 6–8 erupted and functional cheek-teeth in each dental arcade. As they wear, these teeth migrate horizontally from the posterior region of the tooth-row to the anterior region (Domning and Hayek 1984). The bony septa between the roots of each tooth are reabsorbed in front of a tooth root and then re-deposited behind it. This allows the teeth to move through the bone of the mandible over time. By the time cheek-teeth reach the anterior-most location, little-to-no crown remains. The roots of these now nonfunctional teeth are resorbed, and the tooth falls out. New molars erupt at the posterior end of the tooth-row and migrate anteriorly as replacements. The number of replacement teeth is indeterminate. In contrast, dugongs do not possess such a conveyor-belt mechanism to resist an abrasive diet. Instead, adult dugongs possess open-rooted, simple peg-like molars (three per dental arcade) that consist of dentin covered by cementum, a material that is softer than the enamel, found in manatee teeth. The last two molars in dugongs are hypsodont molars that erupt slowly over their lifetime (Mitchell 1978; Marsh 1980; Lanyon and Sanson 2006a, b); this is a common mechanism among herbivores to resist tooth wear. Dugongs have a finite number of molars over their lifetime. Such differences in dentition between manatees and dugongs are a result of varying selection pressures of the material properties of their herbivorous diet. Dugongs are constrained to consume vegetation that is less abrasive compared to manatees. The supernumerary tooth replacement system of manatees allows them to consume a greater diversity of vegetation, including abrasive foods that contain silica, such as true grasses.

2.2.5.2

Active Touch Using the Facial Hairs and Bristles

Sirenians also use their facial vibrissae to investigate novel objects. Most often the oral disk is used during initial exploration (Marshall et al. 1998a, 2003). Although

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

33

the mechanism differs slightly, the bristle-like hairs of the oral disk are everted in a ‘flare’ response, whereby facial muscles act to expand and flatten the oral disk, causing protrusion of the bristle-like hairs prior to contact. The perioral bristles (Fig. 2.1) are often used during the subsequent investigation (e.g., of anchor lines), much like the use of macrovibrissae by rodents, while the bristle-like hairs (hairs intermediate in size and thickness between perioral bristles and body hair) are used like microvibrissae (Brecht et al. 1997). Acuity of active touch was tested in captive Antillean manatees through their capacity to discriminate targets consisting of grooves of different widths (Bachteler and Dehnhardt 1999), and in Florida, manatees using targets with ridges and grooves of different widths (Bauer et al. 2012). In both cases, the manatees investigated targets with their facial area, using the bristle-like hairs on their oral disc, as well as bristles in the perioral areas. These studies demonstrate that West Indian manatees’ active touch capabilities have good acuity, comparable to Asian elephants (Elephas maximus) using their trunk tips (Dehnhardt et al. 1997), and to human index finger sensitivity (Morley and Goodwin 1983). Memory of this tactile task persisted over long periods: the Florida manatees had 100% correct performance after up to 22 months without rehearsal (Bauer et al. 2012). Sirenians also engage in tactile behaviors in social contexts including cavorting, mating herd jostling, pushing, hugging, touching with flippers, mouthing, and nonaggressive body contact (Hartman 1979; Reynolds 1981; Marshall et al. 1998a; Chap. 4). Solitary tactile behaviors include self-touching or grooming and environmental exploration using the flippers and face (Hartman 1979; Marshall et al. 1998a). A recent study of captive and semi-captive Antillean manatees quantified tactile behaviors in three contexts: social, environmental exploration, and self-maintenance (Lucchini et al. 2021). Tactile behaviors constituted 14% of the activity budget. Environmental tactile behaviors represented the majority of observed tactile behaviors, and social behaviors occurred more frequently than self-maintenance behaviors. There were no differences based on age, with the exception of “hanging” behavior, which was more frequent in juveniles. Females engaged in tail touches more often than males, and males engaged in body contact with the tank wall more often than females. Although sirenian social structure is ephemeral, except for cow-calf pairs, aggregations of 2–3 individuals are common (Reynolds 1981; Chap. 4) and represent a rich, rather unexplored territory for further quantitative study of how component behaviors (e.g., mouthing) play a role in social interactions in the wild.

2.2.5.3

Digestive System

The digestive system comprises the digestive tract (oral cavity, pharynx, esophagus, stomach, small and large intestine) and accessory structures (teeth, tongue, salivary glands, liver, gall bladder, and pancreas). Its function is varied but includes the ingestion of food, mechanical and chemical breakdown of food, absorption of nutrients, and excretion of waste products. All sirenians are hindgut fermenters and possess

34

C. D. Marshall et al.

similar digestive components that are unique to these herbivores. The sirenian digestive system is massive and long (Marsh et al. 1977; Reynolds and Rommel 1996). The stomach of West Indian manatees and dugongs is simple (one chamber) with a prominent muscular ridge that projects into the stomach lumen, partially dividing it into anterior and posterior regions (Ketchington 1972; Reynolds and Rommel 1996). The dugong stomach is responsible for 50% of the post-oral breakdown (Lanyon and Sanson 2006b). It is likely that the stomach of all sirenians functions for water resorption and particle size reduction (Marsh et al. 2018). All sirenians have a discrete digestive accessory gland (gastric or cardiac gland) that protrudes from the greater curvature of the stomach, which contains the cells that secrete acid and enzymes (Marsh et al. 1977; Reynolds and Rommel 1996). Although the organization of the stomach is simple, this segregation of the digestive acid and enzyme secreting cells from the rest of the stomach is unusual among mammals (Reynolds and Rommel 1996). It is speculated that this segregation, together with secretion of mucus by the cardiac gland, may protect these cells, and therefore the rest of the stomach, from the abrasive nature of their herbivorous diet. Additional secreted mucus may also serve to lubricate the digesta (Marsh et al. 1977; Reynolds and Rommel 1996). The small and large intestine of sirenians may exceed 20 and 30 m in length, respectively (Marsh et al. 1977; Bonde et. al. 1983; Reynolds and Rommel 1996; Eros et al. 2007); the small intestine of dugongs is about 50% the length of its large intestine (Lanyon and Sanson 2006a). Both manatees and dugongs possess duodenal ampullae, a pair of blind pouches that extend from the expanded portion of the duodenum and secrete an acidic mucous. They are unusual structures among mammals. A large cecum with two diverticula is present at the junction of the small and large intestines. The proximal large intestine and cecum are the most probable sites for cellulose breakdown by microbes (hindgut fermentation) in manatees and dugongs (Reynolds and Rommel 1996). Both manatees and dugongs have extended passage times of digesta ranging from 6 to 10 days (Lanyon and Marsh 1995; Reynolds and Rommel 1996; Larkin et al. 2007), which contribute to a high digestion efficiency (Burn 1986; Lomolino and Ewel 1984; Lanyon and Sanson 2006a; Worthy and Worthy 2014). The heat generated from cellulolysis in the large sirenian gut may also contribute to greater efficiency in body thermoregulation, especially in the large-girthed Florida manatees that are at the northern extreme of their usable range (Glaser and Reynolds 2003). Seed dispersal through sirenian fecal matter is known to occur in dugongs (Tol et al. 2017). Dispersal is facilitated by high consumption rates, slow digesta passage time, and long-distance movement patterns. Florida manatees may have recently been responsible for dispersing Valisneria (eelgrass) seeds in Kings Bay (USA), which spread at a much faster than anticipated rate after being planted. A combination of traits allows sirenians to succeed as fully aquatic herbivores, in spite of their need to ingest large quantities of food on a regular basis, and the low metabolic rates (at least in manatees) that result from relatively low-quality food resources compared to carnivores. These traits include prehensile lips with bristles, molar adaptations, keratinized oral pads, cardiac gland specializations, a long digestive tract, slow digesta passage time resulting in high digestive efficiency, and hindgut fermentation. These traits manifest themselves as long feeding bouts,

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

35

and ties to habitats where aquatic vegetation (freshwater, marine, and/or brackish) is plentiful.

2.3 Sensory Innovations for an Aquatic Life In addition to morphological innovations for acquiring resources, sirenians possess sensory innovations that, among other functions, help them to find those resources. Generally, mammalian sensory systems are thought of as involving five senses: touch, hearing, sight, taste, and smell. Sirenians perceive their world principally through an enhanced sense of touch, or somatosensation (Bauer et al. 2018). Since vision is not as important a sense in the aquatic environment, particularly in turbid habitats, selection pressure for vision has been reduced, and as a consequence sirenian eyes and visual cortices are relatively small compared to their somatosensory systems. Sirenians possess novel innovations in both their central nervous system (brain and spinal cord) and peripheral nervous system to detect a multitude of tactile, but also other sensory, environmental cues. Studies of sirenian brains are best considered within the context of comparative mammalian brain structure, function, and evolution, a field that has yielded a wealth of information regarding brain-behavior traits (Fig. 2.4).

2.3.1 Central Nervous System Characteristics Comparative neurobiologists historically have investigated brain size, as well as the pattern of gyri and sulci on the outside of the brain, to understand brain function and evolution. Both of these metrics are unusual among sirenians. However, studies of cerebral cortex organization have revealed important functional data that are likely more informative than simple metrics of size and gyration.

2.3.1.1

Encephalization Quotient and Brain Size

Relative brain size was often viewed as a proxy for intelligence, due more to the ease of gathering data on brain size than to any empirical justification. There is also the difficulty of measuring intelligence, and agreement on criteria that apply across taxa. Since the 1970s considerations of brain size have often focused on correlations with ecological factors rather than intelligence, and recent studies have focused on functional subdivisions of the brain instead of whole brain size. Relative brain size in its simplest form is expressed as the ratio of brain weight to body weight. More often, for comparative purposes, it is expressed as an encephalization quotient (EQ; Jerison 1973), which normalizes the brain/body weight ratio to a reference curve, usually the average trend line for brain/body weight across mammalian taxa. In this scheme, taxa with brain weights equivalent to the value expected for an average mammal of

36

C. D. Marshall et al.

Fig. 2.4 A collection of mammalian brains. Note the large size but relatively lissencephalic surface appearance of the manatee brain. Photo credit: Wally Welker, comparative mammalian brain (http:// neurosciencelibrary.org/index.html)

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

37

that body weight have an EQ = 1.0; EQs > 1.0 indicate larger than expected relative brain size, and EQs < 1.0 indicate smaller than expected relative brain size. Sirenians as a group (including manatees, dugongs, and the extinct Steller’s sea cow) exhibit low encephalization quotients (O’Shea and Reep 1990). Relative to mammals as a whole, sirenian EQs range from 0.116 to 0.382, much lower than expected. Among sirenians, the most robust data exist for Florida manatees, wherein adult brains weighing 364 g on average produce an EQ of 0.275, meaning that manatee brain size is ~27% of the size expected for an average mammal of that body size. Considering the significant influence of the aquatic environment on a range of important anatomical and physiological variables, it is reasonable to compute sirenian EQs with reference to other aquatic mammals rather than to mammals as a whole. The EQ of Florida manatees is 0.254 compared to cetaceans (0.305 compared to mysticetes; 0.171 compared to odontocetes), and 0.506 compared to pinnipeds (O’Shea and Reep 1990). This indicates that manatees have low EQ independent of the influence of the aquatic environment. In seeking to understand the low EQ observed in sirenians, it is helpful to recall that there is a natural tendency to view brain weight as a dependent variable with respect to body weight. Although there is undoubtedly some degree of genetic/developmental coupling between brain size and body size, many factors could drive changes in body size without changing brain size. For example, consider blubber deposition in cetaceans and pinnipeds, or reduced body weight in connection with flight in birds. In the case of sirenians, there has been strong selection pressure for large bodies, including large digestive systems, to accommodate continual processing of large amounts of aquatic plants. Diet probably restricts metabolic rate, which is low in manatees, about one fifth the rate expected for their body size (Irvine 1983), and large body size also mitigates the effects of heat losses associated with low metabolic rate. Thus, the specialized niche of sirenian aquatic herbivory involves selection for large body size and low metabolic rate in manatees. Large adult body size in sirenians occurs subsequent to an extended period of postnatal growth beyond the period of brain growth (O’Shea and Reep 1990). Therefore, this may be a factor in establishing the low adult EQ. This hypothesis could be tested by plotting EQ for earlier stages of sirenian postnatal development and comparing this dynamic EQ with similar measures from other taxa. Despite their low EQ, it should be noted that the absolute size of sirenian brains is large compared to all mammals, and EQ should not be a metric used to explain behavioral traits.

2.3.1.2

Gyration Index

One of the most easily recognized features of sirenian brains is the relatively smooth surface of the cerebral hemispheres (Fig. 2.5). The condition of smooth hemispheres is called lissencephaly and is typical of the adult brains of rodents and bats (which comprise 70% of all mammals; Vaughn et al. 2011), and the fetal brains of all mammals. In contrast, adult brains having numerous convolutions, as in carnivores, cetaceans, and many primates, are said to be gyrencephalic (Welker 1990).

38

C. D. Marshall et al.

Fig. 2.5 a lateral and b dorsolateral perspectives of the external morphology of a Florida manatee brain. Note the large size but smooth (lissencephalic) appearance of the cerebral hemispheres. In panel a: ob = olfactory bulb, lower arrow denotes olfactory tract, upper arrow denotes the prominent lateral fissure; in panel b: bottom arrow denotes lateral fissure, upper arrow denotes a longitudinally oriented fissure. (from Comparative Mammalian Brain Collection, Manatee Brain Site, www.manateebrain.org)

Not surprisingly, there are intermediate degrees of cortical folding across taxa, and this range of variation is encapsulated by the gyration index (GI), which quantifies the amount of hidden, versus exposed, cortical surface (Zilles et al. 1988). Highly lissencephalic brains have GIs near 1.0, whereas highly convoluted cetacean brains may exhibit GIs greater than 4.0. As with relative brain size, gyration index has often been viewed as a proxy for intelligence. However, gyration patterns are orderspecific and within each lineage, there is a positive correlation of brain size with GI (Pillay and Manger 2007). For example, the ungulate, carnivore, and rodent patterns of gyri are each distinct, and the least weasel (a carnivore; Mustela nivalis) has a very small, convoluted brain, whereas the muskrat (a rodent; Ondatra zibethicus) brain is lissencephalic, though it is larger in absolute size than that of weasels (Welker 1990). Adult sirenian brains are distinct in being the largest mammalian brains that are lissencephalic (GI = 1.06; Reep and O’Shea 1990). All other mammalian brains over 300 g in weight are substantially convoluted (Welker 1990; Pillay and Manger 2007).

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

39

Why are sirenian brains lissencephalic, unlike other large mammalian brains? As evidence of the ever-present desire to associate gross brain traits with intelligence, we may note that an eminent comparative neuroanatomist once made the following statement: Amongst the whole series of placental mammals, there is no other animal in which the brain presents features so extraordinary and so bizarre as in the Sirenia. The only parallel which can be found for the peculiar cases presented by the manatee and the dugong is that presented in the brains of idiots. (Elliott Smith 1902)

This statement was not altogether surprising because some human brain abnormalities include types of lissencephaly. However, we have learned much about the developmental bases for these abnormalities since Elliott Smith’s time, including genes that regulate the microtubule cytoskeleton during neuronal division, migration, and maturation (Kerjan and Gleeson 2007). The adult manatee cerebral cortex exhibits the typical six layers seen in many mammalian brains, including those of ungulates, carnivores, and primates (but not cetacean brains, which only have four of those typically six cortical layers; Reep et al. 1989; Marshall and Reep 1995); there is no evidence that arrested cell division or migration influence development of lissencephaly in manatees. Regarding neuronal maturation, Golgi studies indicate that manatee cortical neurons exhibit a rich variety of dendritic branching and axonal arborizations, comparable to other taxa (Reyes et al. 2015, 2016). That is, they have the typical number of synapses and neural connectivity that is consistent with gyrencephalic brains. In manatees, and presumably dugongs, the white matter underlying the cortical gray matter is relatively thick (Reep and O’Shea 1990). This represents an abundance of axons grouped in bundles, many making connections between nearby and distant cortical areas. Collectively, axons in bundles have biomechanical properties that likely influence the folding of the gray matter (Essen 1997; Mota and HerculanoHouzel 2012). Thus the relative timing between establishment of the gray matter, versus neuronal connections between cortical areas, is one developmental variable that could plausibly influence the degree of gyration. If the white matter axons are established relatively early, this could inhibit gyration of the overlying gray matter. As in the use of EQ as a metric, sirenians’ brains have evolved very differently compared to all other mammals, and simple metrics such as GI should not be used to explain behavioral traits. The sirenian brain represents a departure and new direction in mammalian brain evolution.

2.3.1.3

Rindenkerne

Although in general, the sirenian cerebral cortex exhibits many features that are consistently found across mammalian species, evolutionary innovations can generate unusual specializations. One such cortical specialization is known as “Rindenkerne” (cortical nuclei) and is unique to sirenians. Rindenkerne are neuronal aggregates

40

C. D. Marshall et al.

present in layer VI of several putative somatic sensory and auditory areas of sirenian cerebral cortex (Reep et al. 1989; Marshall and Reep 1995). They were first discovered in dugongs (Dexler 1913). Rindenkerne aresimilar to “barrels,” which are cortical representations of vibrissae found in the primary somatosensory cortex of many other species (Woolsey and Van der Loos 1970; Johnson 1980; Rice 1995). However, Rindenkerne are found in layer VI (Dexler 1913; Reep et al. 1989), a deep output zone, whereas barrels are found in layer IV (Woolsey and Van der Loos 1970; Johnson 1980; Rice 1995), an input zone, suggesting a difference in connectivity and function. Like barrels, Rindenkerne are thought to be related to sensory hairs on the manatee’s face and body (Reep et al. 1989). Localization to layer VI suggests that if Rindenkerne are associated with sensory hair function, they may do so in the capacity of sensorimotor control (e.g., facial vibrissae eversion, oripulation (the use of the perioral bristles and lips in grasping, in contrast to manipulation, which connotes the use of the manus, or hand; Reep et al. 2001), feeding behavior, and tactile exploration/object recognition (Reep et al. 1989).

2.3.1.4

Functional Assignments to Manatee Cerebral Cortex

Comparative neuroanatomists investigate the structure and function of the cerebral cortex by looking at patterns of neurons from the surface to the deepest part of the gray matter. Mammals typically have six cortical layers below which we find white matter—axons grouped in bundles that make connections between and beyond cortical areas. This pattern of neurons, layers of neurons, and the type and number of neurons present is referred to as cortical cytoarchitecture. The cortical cytoarchitecture of the brain of Florida manatees has been mapped, and presumptive functions have been assigned histologically (Reep et al. 1989; Marshall and Reep 1995). However, to definitively assign and confirm functions to cortical areas in sirenians (e.g., visual cortex, auditory cortex, sensory cortex, etc.), electrophysiological studies would need to be conducted. While such invasive studies are not possible to conduct on threatened species such as sirenians, histological and histochemical assessments are very informative when placed within a comparative neurobiological context based on what is known in other species (Johnson 1990; Johnson et al. 1994). Brain sections stained for cytochrome oxidase were particularly useful in identifying primary sensory areas (somatosensory, auditory, and visual) along with functional subdivisions within primary somatosensory cortex (Fig. 2.6; Sarko and Reep 2007). Based on comparative neurobiology, and the relatively conserved functional organization of somatosensory cortex across species, a lateral-to-medial progression from the face, to flipper and body trunk, and finally fluke, were assigned to the manatee primary somatosensory cortex (Sarko and Reep 2007). Cytochrome oxidase staining also revealed that Florida manatee primary somatosensory cortex occupied approximately 25% of the total cortical area (Sarko and Reep 2007). This relatively large area is comparable to that of other somatosensory specialists such as naked mole-rats (Heterocephalus glaber; which, like manatees, also have sensory

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

41

Fig. 2.6 Functional assignments of cortical cytoarchitectural regions in Florida manatee brain 83– 74. Left lateral view of a whole brain. MI = motor cortex, SI = primary somatosensory cortex, SII = secondary somatosensory cortex, Aud = auditory cortex, Vis = vision cortex

hairs covering their face and body) which have 31% of their total cortical area occupied by the primary somatosensory cortex (Catania and Remple 2002). Mean Florida manatee neuronal density for presumptive primary somatosensory cortical areas DL1 and CL2 were 35,617 neurons/mm3 , falling within what was predicted based on Florida manatee brain mass compared to afrotherians and xenarthrans, but inhibitory interneuron density was relatively low (Reyes et al. 2015). This further confirms the importance of the sense of touch in Florida manatees, and these results can likely be inferred to all sirenians. These intriguing functional assignments also imply that manatee auditory cortex contains Rindenkerne. If Rindenkerne are in fact associated with sensory hairs, it follows that auditory cortex in manatees has extensive overlap with somatosensory function, which may include multisensory integration related to hydrodynamic stimulation of hairs (Gerstein et al. 1999; Sarko and Reep 2007). In other words, sound— particularly the low-frequency sounds that travel best underwater—may not only be heard but also felt. Overall, the primary auditory and visual cortices combined occupied less cortical area than the primary somatosensory cortex, a result that fits hypotheses based on the relative behavioral importance of each sensory modality for Florida manatees (Sarko and Reep 2007). Golgi staining in portions of presumptive primary somatosensory and primary visual cortex demonstrated greater dendritic branching of pyramidal neurons in somatosensory, compared to visual cortex in manatees, but no evidence of the tritufted and star-like neurons that characterize cetacean brains (Reyes et al. 2016). These data support the hypothesis that the senses of touch and hearing are more important in Florida manatees, and presumptively all sirenians than vision.

42

2.3.1.5

C. D. Marshall et al.

Central Nervous System—Subcortical Structure and Function

Sensory portions of the manatee central nervous system are organized along the typical mammalian plan, whereby brain regions exhibit specializations, expansions, and reductions, in accordance with similar features of the sensory periphery. With regard to anatomy, known physiology, and behavior in manatees, somatic sensation and hearing are the dominant sensory modalities; taste appears to be moderate, and olfaction and vision are much reduced. In manatees, and presumably dugongs, the brainstem regions that are the first stage in processing information related to somatic sensation are similar to those in other mammals. Across mammalian taxa, these include structures such as the cuneate-gracile nucleus, which receives input from axons supplying mechanoreceptors in the trunk, forelimb, and hindlimb; the trigeminal nuclear complex that receives input from the face and head; and Bischoff’s nucleus, which receives input from the tail region. The Bischoff’s and cuneate-gracile complex are large, and collectively exhibit three distinct regions that (from medial to lateral) are hypothesized to represent the fluke, trunk, and forelimb flipper (Sarko et al. 2007b). The cuneate-gracile complex exhibits numerous internal lobules (best visualized using axonal staining), which are indicative of functional compartmentalization (similar to facial motor nucleus discussed above). Throughout the central nervous system in a wide range of mammals, somatic sensory information is topographically organized. This is often associated with boundaries between regions within a nucleus, or between smaller groups of neurons (as described for facial motor nucleus and facial muscles, above). These boundaries are usually created by bundles of myelinated axons that terminate in specific compartments. Increased segregation through increased boundaries infers greater functional importance and sensory resolution of information processing. The trigeminal system (Cranial Nerve V) is large and elaborate in manatees. Its fibers are both motor and sensory, but in the face, it receives sensory information from vibrissae and conveys that information to the cerebral cortex via a series of neural nuclei. The principal sensory nucleus, and to a lesser extent the spinal trigeminal nucleus (all parts of the trigeminal system), exhibit numerous internal lobules or compartments. In some mammalian taxa, clusters of neurons called barrelettes are seen in the trigeminal complex. Although barrelettes were not identified as discrete cell groups in the manatee brain, the internal lobules may function similarly by providing for independent processing of information from circumscribed regions of the face. A particularly interesting possibility is that each of the six perioral sensory hair fields (U1-U4 and L1-L2; see Fig. 2.1), and the field of bristle-like hairs on the oral disk, may be associated with a single lobule (Sarko et al. 2007b) indicating highly refined sensation. The major source of peripheral mechanoreceptive input to the brainstem appears to be the sensory hairs, with their extensive innervation by a total of ~210,000 axons (Reep et al. 2002). A study of the manatee integument (Graham 2005) found few mechanoreceptors in the skin, aside from Pacinian corpuscles around the nostrils, so it may be that the sensory hairs have assumed the role of all or most of the mechanoreception associated with the manatee integument.

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

43

2.3.2 Peripheral Nervous System Characteristics—Structure and Function of Sensory Systems 2.3.2.1

Microanatomy

While the central nervous system comprises the brain and spinal cord, all other components of the mammalian nervous system are collectively referred to as the peripheral nervous system. Sensory hairs, and other mechanoreceptors, are part of the peripheral nervous system that interacts with the external environment and convey sensory information back to the brain. Follicle-sinus complexes (F-SCs), the technical name for sensory hairs, are one of seven major hair types (Zelena 1994). F-SCs are classified according to their following microanatomical attributes: a circumferential blood sinus, a thick connective tissue capsule, and dense innervation (Rice et al. 1986). Mammalian F-SCs are generally restricted and named with respect to their regional distribution on the body. They are often concentrated on the mystacial region of the face where they can most effectively be used for sensory navigation, prey detection, object recognition, and tactile exploration (Dykes 1975; Ling 1977; Brecht et al. 1997; Dehnhardt et al. 1998, 2001). Sirenians are unusual in that they exhibit these specialized sensory hairs not only on the face, but also on the entire body. This was first noted by Dosch (1915) and later reported in dugongs (Bryden et al. 1978; Kamiya and Yamasaki 1981) and Antillean manatees (Sokolov 1986). However, most of these assessments lacked comprehensive, systematic characterization and specific details, including specification of body regions from which follicles were sampled, that would have facilitated replication and cross-species comparisons. The land-dwelling hyraxes, one of the closest living relatives to sirenians, also exhibit sensory hairs on the postcranial body, interspersed among pelage hairs (Sarko et al. 2015). Across the sirenian species studied to date, Florida manatee F-SCs have been characterized in the most detail (Reep et al. 2001, 2002). All hairs on the manatee body and face were proven to be F-SCs, with F-SCs 30 times more densely distributed on the face than on the body (Reep et al. 1998, 2001, 2002). On the face, the largest bristles (from the U2 perioral field; see Fig. 2.1) exhibit the most classical and pronounced F-SC structure. Branches of CN V innervate all facial F-SCs (Reep et al. 1998). The infraorbital branch (a branch of V2 ) of the trigeminal nerve divides to innervate the upper bristles of the muscular-vibrissal complex. The inferior alveolar branch of the mandibular division of the trigeminal nerve (V3 ) supplies the bristles of the lower bristle pad. The micro-anatomy of F-SCs is complex (Fig. 2.7): a detailed description of their functional morphology has been described by Reep et al. (2001) and many comparisons have been made with other marine mammal F-SCs (e.g., Marshall et al. 2006; Czech-Damal et al. 2011; Marshall et al. 2014; McGovern et al. 2014; Mattson and Marshall 2016; Jones and Marshall 2019; Sprowls and Marshall 2019). Manatee facial bristles and hairs are used for tactile exploration/object recognition (relying primarily on bristle-like hairs of the oral disk) and oripulation behaviors (relying

Fig. 2.7 F-SC structure from a Florida manatee large U2 perioral bristle (left, see Fig. 2.1), a bristle-like hair from the oral disk (BLH, center), and a postcranial hair (right; from Sarko et al. 2007a). BM = basement membrane; DVN = deep vibrissal nerve; HP = hair papilla; ICB = inner conical body; OCB = outer conical body, RRC = rete ridge collar; RS = ring sinus

44 C. D. Marshall et al.

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

45

primarily on perioral bristles of the U2 and L1 fields (Hartman 1979; Marshall et al. 1998a; Bachteler and Dehnhardt 1999; Marshall et al. 2000). Such behaviors would certainly be facilitated by the well-innervated perioral and oral disk F-SCs, with microanatomical structures that are well-suited to detect and relay a wealth of tactile information while manatees navigate and explore their underwater environments. Not only do sirenians have sensory hairs over the entire body, but F-SCs are the only type of hair they possess. In Florida manatees, there are ~2000 hairs on the face and head and ~3300 on the postcranial body (Reep et al. 2001; 2002). The distribution of hairs on the face is ~30 × denser than on the postcranial body. On average the postcranial hairs are spaced 25 mm apart, enough to allow each hair an independent field of movement (Reep et al. 2002). Individual hairs on the head and face have thicker shafts, larger and more complex follicles, greater variety of mechanoreceptors, and more innervation than those on the postcranial body. The F-SCs in the postcranial body appears to function in a manner analogous to the lateral line system present in fish, serving to detect underwater features such as water currents and approaching animals (Reep et al. 2002, 2011; Gaspard et al. 2017).

2.3.2.2

Innervation

The impressive array of sensory hairs on the manatee’s body and face is associated with a large amount of innervation, which varies for each F-SC based on region and function. Florida manatee F-SCs on the postcranial body exhibited the lowest amount of innervation at 20–50 axons per F-SC (Reep et al. 2002). With approximately 3000 F-SCs on the postcranial body, this would amount to 60,000–150,000 axons for postcranial F-SCs alone (Reep et al. 2002, 2011). The number of axons per facial F-SC was higher, ranging from 34 to 254 (Reep et al. 2001). Supradisk and chin F-SCs had the fewest axons per FSC, ranging from 34 to 48 and overlapping with postcranial body F-SC ranges (Reep et al. 2001). Bristle-like hairs (BLHs) located on the oral disk, and important in tactile exploration/object recognition, were innervated by 49–74 axons per F-SC (Reep et al. 2001). Among the perioral bristles, U1 F-SCs had the fewest axons per F-SC at 61–84 axons per FSC (Reep et al. 2001). U3, U4, and L2 F-SCs had approximately 100 axons per F-SC, whereas U2 and L1 F-SCs (the largest perioral bristles, and critical to feeding and oripulation behaviors) had 200 or more axons per F-SC (Reep et al. 2001). Altogether, this amounts to approximately 110,000 axons innervating F-SCs of the manatee face (Reep et al. 2001), a significant neuronal investment that is comparable to sensory specializations found in other terrestrial species (e.g., the tactile star of star-nosed moles, Condylura cristata, with approximately 100,000 axons (Catania and Kaas 1997); and marine mammal species such as harbor seals, Phoca vitulina, with an estimated 143,000 axons innervating the mystacial vibrissae (Jones and Marshall 2019)). This level of neural investment implies that the stimuli detected by sirenian sensory hairs are significant and contribute to behavior in important ways.

46

2.3.2.3

C. D. Marshall et al.

Mechanoreception

In addition to quantifying innervation for manatee F-SCs, it is also important to characterize the types of nerve endings from this innervation within the F-SCs, to gain a better understanding of what types of stimuli each hair might be able to detect, and ultimately what manatees might be capable of perceiving, as the central nervous system integrates dynamic information from many hairs. Although microanatomical structures of F-SCs are relatively consistent across species, innervation patterns vary according to behavioral demands and evolutionary selection pressure. Immunofluorescence revealed that each Florida manatee’s F-SC is innervated by two deep vibrissal nerves supplying nerve endings to various portions of the F-SC (Sarko et al. 2007a). Manatee F-SCs have a range of C, Aδ, and Aβ fiber type innervations, including terminations as Merkel-Neurite Complexes, club, and longitudinal lanceolate endings at the ring sinus level (Sarko et al. 2007a). Because Merkel endings are considered to be low-threshold, slowly adapting mechanoreceptors attuned to directionality, we concluded that manatee F-SCs have prioritized directionality detection of hair deflection through dense innervation investment. This Merkel ending innervation could subserve the detection of hydrodynamic stimuli and lateral line function of postcranial hairs (Reep et al. 2002; Sarko et al. 2007a; Gaspard et al. 2013, 2017). Merkel endings and lanceolate endings are responsive to a wide range of frequencies, and might also be responsive to sound detection when a vibrissa is deflected at a certain frequency (Stephens et al. 1973; Gottschaldt and Vahle-Hinz 1981; Hyvarinen 1989, 1995), which may, in turn, support the high degree of overlap thought to be present in manatee auditory and somatosensory areas cortically (Sarko and Reep 2007). Manatee skin has sparsely distributed mechanoreceptors (Meissner and Pacinian corpuscles) within the skin adjacent to manatee F-SCs and a dense and wellinnervated vascular network with especially well-innervated arteriovenous shunts (Sarko et al. 2007a). This may serve to regulate blood flow to the epidermis of manatees as a thermal regulatory mechanism. This rich innervation likely confers the capacity to sense changes in, and to regulate, the relative blood pressure within the F-SC sinuses. The regulation of F-SC blood pressure is likely critical for optimizing tactile sensation in an underwater environment. We hypothesized that the extensive innervation of the hair shafts of manatee U2 and L1 F-SCs (Fig. 2.1) serves to detect forces from applied stress without material displacement, facilitating oripulative behaviors by allowing manatees to assess and regulate the force applied to follicles via sensorimotor feedback and adjustments (Sarko et al. 2007a).

2.3.2.4

Behavioral Studies of Hydrodynamic Reception in Manatees

The investment in elaborate sensory hairs with multiple mechanoreceptors, together with the central nervous system machinery to analyze their input, as described above, strongly suggests that sirenians derive critical information from direct touch and hydrodynamic stimuli that manifest in important behavioral responses. Woodie

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

47

Hartman and John Reynolds were the first to suggest that the body hairs of manatees might be used in hydrodynamic reception (Hartman 1979; Reynolds 1979). In his field studies of Florida manatees during the late 1960s and early 1970s, Hartman observed groups of manatees resting together on the bottom of Crystal River. Every 10–15 min an individual would slowly float to the surface to breathe, then sink back down to rest again. The other animals would follow this cue so that the whole group appeared to rise and fall together, with their eyes closed. This is known as synchronous behavior and has been reported in dugongs as well (Jarman 1966). Hartman conjectured that deflection of the body hairs in response to water movements made by the first individual might be the means by which this collective behavior occurred. He also suggested that this system of body hairs might be the basis for detecting the movements of other manatees during non-resting periods. Ed and Laura Gerstein studied manatee hearing over several years using as subjects Stormy and Dundee, two manatees resident at Lowry Park Zoo in Tampa, Florida. They suggested that the body hairs mediate detection of low-frequency sounds (Gerstein et al. 1999). When acoustic stimuli lower than 400 Hz frequency were played, Stormy consistently rotated his body and bent his head down. This posturing was consistent with maximizing the exposure of the body hairs to these low-frequency stimuli. The Gersteins referred to this as a vibrotactile response, to distinguish it from the typical responses involving the auditory system, which did not include posturing. Systematic studies of hydrodynamic reception have involved Hugh and Buffett, two manatees housed at Mote Marine Laboratory in Sarasota, Florida. In the first study (Gaspard et al. 2013), Hugh and Buffett were trained to station their chin at an underwater bar. A sphere about the size of a ping pong ball was attached to a long rod and was located a few centimeters in front of their faces. Small sinusoidal oscillations of the ball were controlled by a computer-driven motor, generating a dipole hydrodynamic stimulus. A go/no-go behavioral paradigm with a staircase method was used to generate systematic sequences of stimuli. Both manatees could detect very small water movements (~1 um) over the range of vibrations from 5 to 150 Hz. We do not actually know if the manatees were detecting particle displacement, velocity, or acceleration because these are simple mathematical transformations of each other. The role of the facial vibrissae (as distinct from skin mechanoreception) was assessed by restricting the vibrissae with masks of variable-sized mesh, which allowed a differential number of hairs to protrude. When masks with large holes were used, so that ~50% of the hairs could move, performance was slightly poorer than normal, but as the number of protruding hairs was reduced, detection thresholds rose. This indicates a loss of sensitivity as the number of freely moving vibrissae was reduced. In a second study (Gaspard et al. 2017), this vibrating ball was located about 20 cm from the side of the manatee’s body, targeting the sensory hairs in this region. Three locations (forward, middle, and rear third) of the postcranial body were tested at the same frequencies used in the facial experiments. Again, both Hugh and Buffett could detect small water movements. Because the three locations yielded similar thresholds, the data were combined. Detection thresholds were about one standard deviation higher for postcranial hairs than for facial hairs. To evaluate the contribution of the postcranial sensory hairs to hydrodynamic sensitivity, a 50 cm square was

48

C. D. Marshall et al.

shaved on the side of each manatee and the tests were repeated. Shaving reduced sensitivity by a factor of three to four at 75 Hz, the one frequency at which detection was tested. Together, these studies on facial and postcranial sensory hairs suggest that all the sensory hairs on the body are capable of responding to water movement. A large number of mechanoreceptors, nerve fibers, and the large sizes of the brain regions devoted to processing this information, are evidence for the importance of hydrodynamic reception in the lives of manatees.

2.3.2.5

Manatee Sensory Hairs—A Mammalian Lateral Line?

Hydrodynamic reception, the ability to sense water movements, is evidently of great value because a wide variety of invertebrate and vertebrate taxa exhibit spatially elongated hair-like structures associated with mechanoreceptors that respond to hydrodynamic stimuli (Budelmann 1989; Bleckmann 1994; Leitch and Catania 2012; Mercado 2014). Of particular interest is the lateral line system of fishes and amphibians, which detects hydrodynamic stimuli in the frequency range of 1–150 Hz (Bleckmann 1994; Coombs and Montgomery 1999), the same range in which the manatee hair system operates. In fish, the lateral line system has been shown to mediate behavioral orientation to water currents (Montgomery et al. 1997), such as those generated by a stream or by moving animals, and it is also capable of detecting underwater objects (Hassan 1989). Stationary underwater objects change the flow field generated by an animal moving through water, and this conveys information about object size and distance that is detected by the lateral line system of fish through analysis of the velocity distribution of the flow field over the entire body (Hassan 1989; Bleckmann 1994; Windsor 2014). If the distributed system of sensory body hairs in Florida manatees is used in a similar way, the large body size of manatees may facilitate this capability by providing a larger detector array. Thus, manatees, and possibly all sirenians, may be able to detect and localize a range of sizes of fixed objects in the underwater environment. Such a capability would be of obvious use in orientation and navigation.

2.3.3 Vision In general, vision is not the most useful sense in aquatic environments. Vision can be limited by physical issues related to refraction index, attenuation of light by water, and by reflectance of particles within the water (Marshall 2017). Most sirenians live in turbid water environments that restrict the presence and usefulness of visual cues. It is therefore unsurprising that both peripheral and central nervous system structures related to vision are reduced, with dedication of metabolically expensive neural resources better spent elsewhere (i.e., somatosensory perception). In behavioral experiments with captive Florida manatees, Bauer et al. (2003) demonstrated limited underwater visual acuity in two captive manatees. The minimum angle of

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

49

resolution was 21 min of arc for one subject; this is comparable to a cow and worse than a legally blind human. The other subject performed even more poorly and is likely functionally blind. Additional behavioral experiments demonstrated that manatees possess a color vision that may be dichromatic (Griebel and Schmid 1996). The retina is rod-dominant but also includes two types of cone photoreceptors, which appear to be attuned to the blue and green wavelengths of the light spectrum (Cohen et al. 1982; Newman and Robinson 2006), a discrimination ability that suits their underwater environment and diurnal activities. Manatees are able to discriminate brightness as well as fur seals, despite the contrasting demands of herbivorous versus predatory niches (Bush and Ducker 1987; Griebel and Schmid 1997). Color and brightness discrimination may help manatees, and perhaps all sirenians, to detect edible and preferred plants for consumption, a key focus and critical concern are given how much vegetation sirenians consume daily. Overall, sirenian eyes are small relative to their body size (Griebel and Schmid 1997; Marsh 2009). The eyes of Florida manatees are approximately 1.9 cm in diameter (Griebel and Schmid, 1997). Florida manatees have a relatively low number of retinal ganglion cells (Mass et al. 2012), which are important sensory cells of the eye. Florida manatees (Harper et al. 2005) and Antillean manatees (Ambati et al. 2006) are the only known species to have vascularization throughout their corneas, which would typically be a pathological state resulting from trauma or injury (Harper et al. 2005). Every other species studied to date, including dugongs, possesses avascular corneas, which provide optical clarity and improve vision. It has been suggested that corneal vascularization in manatees may serve a protective function against their freshwater (hypotonic) environments, as opposed to dugongs, which inhabit only saltwater environments (Ambati et al. 2006). Although the ciliary body of the manatee eye lacks muscle fibers, which would theoretically prevent accommodation, angioarchitecture (i.e., pattern of blood vessels) related to the aqueous humor may alter vascular pressure as an alternative method to allow accommodation (West et al. 1991; Natiello et al. 2005; Natiello and Samuelson 2005). Underwater, the eyes of Amazonian manatees (Piggins et al. 1983) and dugongs (Dexler and Freund 1906; Petit and Rochon-Duvigneaud 1929) are farsighted, or hyperopic. Collectively, these findings suggest that sirenians use vision at intermediate or longer distances to locate large objects in the environment, for example, a green patch of underwater vegetation. Consistent with this hypothesis, Hartman (1979) and Marshall et al. (1998a) noted that wild manatees appear to utilize vision for their initial orientation to an environment, then when approaching objects of interest, they shift to direct tactile contact for further exploration.

2.3.4 Auditory System Sirenians lack pinnae and have only a remnant of the external auditory meatus that leads to the tympanic membrane (Husar 1978; Ketten et al. 1992). As in other aquatic

50

C. D. Marshall et al.

mammals, tissues in the head that are of the correct density may enable direct conduction of sound to the tympanic membrane. In manatees, such tissues form a channel that is optimized for detecting sounds that occur at angles of 45–90° relative to the long axis of the body (Chapla et al. 2007). The zygomatic process of the squamosal bone is closely aligned with the tympanic membrane and contains lipid-rich deposits, so it may function as a major channel by which sound is conducted directly to the tympanic membrane. In comparison with cetaceans, the middle ear ossicles are quite massive in both manatees and dugongs, both absolutely and relative to the mass of the tympanoperiotic complex (Dexler and Freund 1906; Ketten et al. 1992; Chapla et al. 2007; Chap. 7). True seals also exhibit a robust ossicular chain (Nummela 1995). In manatees, the massive middle ear ossicles, and the large size and thickness of the tympanic membrane, may be adaptations for hearing at frequencies above ~ 15 kHz (Chapla et al. 2007). Neuroanatomical structures related to hearing are well developed in manatees. The auditory/vestibular nerves (Cranial Nerve VIII) are robust in size, indicating that a large number of axons enter the brain from the cochlea and vestibular apparatus. In the brainstem, the dorsal and ventral cochlear nuclei, superior olivary complex, and the nucleus of the lateral lemniscus and lateral lemniscus itself, all structures related to hearing, are robust in size (Sarko et al. 2007b). The inferior colliculus receives much of the lemniscal input and is a major feature of the midbrain. The inferior colliculus (a hearing-related structure) is significantly larger than the visually related superior colliculus. The IC/SC ratio has been viewed as a first-order index of auditory versus visual dominance, and the manatee ratio is 1.47 (Johnson and Kirsch 1993), indicating the dominance of audition over vision. However, in species such as starnosed moles that rely more on somatosensation than on vision, the superior colliculus is dominated by somatosensory rather than visual inputs (Crish et al. 2003), and this may also be the case for sirenians. A substantial brachium of the inferior colliculus courses to the thalamic medial geniculate nucleus, which is very large and exhibits subdivisions, in contrast to the small lateral geniculate nucleus which subserves vision (Sarko et al. 2007b).

2.3.4.1

Hearing Performance

Traditional hunters of both manatees and dugongs have long considered that they have acute hearing (Chap. 7). Systematic studies of manatee hearing began with Ted Bullock, a major figure in the history of comparative neurobiology, who tested the hearing of an Amazonian manatee at INPA in Manaus, Brazil, in 1977 (Bullock et al. 1980), and in collaboration with Tom O’Shea tested the hearing of manatees at Sea World, Florida in the early 1980s (Bullock et al. 1982). These experiments were done with the subjects’ heads out of the water, using evoked potentials to test responses to auditory stimuli. They found maximal sensitivity at 1.0–1.5 kHz. In the 1990s, Ed and Laura Gerstein and colleagues established the underwater audiograms of Stormy and Dundee, two captive manatees housed at Lowry Park Zoo in Tampa, Florida (Gerstein et al. 1999). This study was the first in which manatees were trained

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

51

to respond behaviorally. In this two-alternative forced-choice task, the subjects were trained to station at an underwater hoop, and a light indicated when a trial was beginning. The manatees were trained to touch one target if they heard a sound and to touch the other target if they heard nothing. The sound stimuli were tonal pulses of 500 ms duration. The researchers found that the range of best hearing was 6–20 kHz, with maximal sensitivity at 16–18 kHz, and high frequency hearing up to 46 kHz. Comparison with other marine mammals indicated that manatees have comparable or better hearing sensitivity than pinnipeds, and odontocete cetaceans, at frequencies up to 20 kHz. Between 20 and 46 kHz manatees exhibit greater sensitivity than pinnipeds, but less sensitivity than odontocetes (Gerstein et al. 1999). The Mote Marine Laboratory group (Sarasota, Florida, USA), led by Gordon Bauer of New College, found that Hugh and Buffet had similar audiograms as Stormy and Dundee (Gaspard et al. 2012). All four manatees showed greatest sensitivity to sounds in the range of 8–20 kHz. Hugh could hear sounds up to 38 kHz. However, Buffet was able to hear much higher frequencies, up to 72 kHz. These findings on four manatees suggest that there may be significant individual variation in manatee hearing ability at frequencies greater than 20 kHz. In the natural environment, the hearing occurs in the context of ambient sounds generated by a variety of sources. Critical ratios are estimates of the ability to hear in noise, measured as the decibel difference between the tonal signal and background sound. Manatees are able to hear well in noise with critical ratio values ranging from 18.3 to 34.1 dB for tonal frequencies from 4 to 32 kHz (Gaspard et al. 2012). They are particularly sensitive in the range of the second and third harmonics of the tonal components of typical manatee vocalizations (fundamental ~2–4 kHz). These more variable upper harmonics may be one factor allowing for detection of differences between calls, thus allowing identification of individuals. This is consistent with the observed peak hearing sensitivity at 8–20 kHz (Gaspard et al. 2012; Gerstein et al. 1999). The broadband aspects of phonations (O’Shea and Poché 2006) would also make them more localizable. The auditory temporal processing rate of manatees, measured as the ability of the nervous system to map amplitude-modulated tones, was determined using an envelope-following, evoked potential technique. Neural response measurements made in the thick dermal layer of the cranium indicated that subjects had high auditory temporal processing rates, up to 600 Hz, which likely aids spatial localization of sound sources (Mann et al. 2005). Studies of localization on captive manatees in tanks have been done. Chapter 7 describes these findings and discusses how they translate to the wild, where open water and variations in depth occur. This is particularly important considering the threat of vessel traffic to sirenians, particularly Florida manatees.

2.3.5 Vocalizations The main frequency range of vocalization often coincides with that of the maximum sensitivity of hearing across a wide variety of vertebrates (Chen et al. 2016). Woodie

52

C. D. Marshall et al.

Hartman noted that manatees tend to vocalize most during play, sexual activity, and when alarmed. The most frequent communication is between a mother and her calf (Hartman 1979; Chap. 4). Bengtson and Fitzgerald (1985) found that calls among groups of manatees occurred most often during social interactions such as cavorting. Importantly, their findings agreed with those of Hartman (1979), and Schevill and Watkins (1965), that manatees do not appear to use calls for echolocation during navigation, because calls made during swimming were no more frequent than calls made when at rest. Manatee vocalizations typically consist of short (~300 ms) chirps, with a fundamental frequency of 1–4 kHz. Several harmonic bands occur in the range of 3– 20 kHz (Bengtson and Fitzgerald 1985; Nowacek et al. 2003; O’Shea and Poché 2006; Grossman et al. 2014). A study of Amazonian manatees (Sousa-Lima et al. 2002) found that adult females produce sounds of shorter duration and higher frequency than adult males. Juveniles produce even shorter sounds of higher frequency, and calves produce the shortest, highest frequency sounds. Manatees may use these variations to discern the age class of individuals making calls, and the sex of adults that vocalize. More significantly, individuals appear to produce “signature” calls that uniquely identify them to others. This is particularly important for mother-calf pairs that become visually separated in turbid environments. O’Shea and Poché (2006) identified several categories of manatee vocalizations based upon features including frequency modulation profile and the presence of subharmonic bands. They found variations in fundamental frequency, number of frequency bands, the band having the greatest energy, frequency modulation characteristics, and subharmonics. These features provide the raw material for producing variation in individual vocalizations (O’Shea and Poché 2006; Mann et al. 2006). Importantly, they also found that for individual manatees, call structure is stable over years, further supporting the hypothesis that individuals have signature calls. Manatees appear to produce sound by transmitting air from the lungs through the upper respiratory tract, where the frequency band structure of the sound is generated using modified vocal folds in the laryngeal region (Grossman et al. 2014; Landrau-Giovannetti et al. 2014). The sound produced then travels from the larynx into the nasal region, where the overlying space and soft tissue acts like a drum, or resonance cavity, amplifying and transmitting the sound into the water. Consistent with this scenario, observable swelling of the nasal region occurs simultaneously with vocalizations. Dugongs produce a variety of phonations in a similar frequency range as manatees (3–18 kHz). However, the length of these vocalizations may be considerably longer, over two sec in the case of trills (Anderson and Barclay 1995). Do sirenians produce ultrasonic vocalizations? Early studies suggested that they do not, as mentioned above. However, these studies employed hydrophones that were not sensitive in the higher ultrasonic range, and the recordings may have been done in situations where sirenians were less likely to vocalize in this range. Recent findings (Ramos et al. 2020), that wild Antillean manatees generate ultrasonic vocalizations with frequency bands up to 88 kHz, suggest that other sirenians may also utilize this high frequency part of the vocal spectrum. We still have much to learn about the behavioral significance of sirenian ultrasonic hearing and vocalizing (see Chap. 4).

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

53

2.3.6 Gustation Gustation, the sensory perception of taste, is of course closely related to the feeding functional morphology of sirenians (see Sect. 2.2.5 above). Manatee taste buds are present in foliate papillae located on the sides of the root of the tongue (Levin and Pfeiffer 2002; Barboza and Larkin 2020a). Largely because of this restricted distribution of taste buds in manatees, it has generally been thought that taste must be relatively unimportant. However, there have long been suggestions that taste may be involved in detection of hormones released by conspecifics, detection of salinity changes, or avoidance of toxic plants (Best 1981). Marshall et al. (1998a) reported that Florida manatees had a negative reaction to some plant roots that retained mud and soil and suggested that this may have been due to taste-mediated aversion. The role of taste in sirenian food preference has not been systematically investigated. More recent findings support a significant role for taste in manatees. A quantitative study (Barboza and Larkin 2020a) found that manatees have more taste buds (~11,000) than many other mammals for which similar information is available, including dolphins (Komatsu and Yamasaki 1980; Yamasaki et al. 1980), and comparable to the number of taste buds in bovines and rhesus monkeys (Davies et al. 1979; Mack et al. 1997). Based upon these data, a revised view is that manatees have a robust number of taste buds, but these are concentrated in a specific region, the sides of the root of the tongue. This focal concentration of taste buds suggests that they may serve a specialized function, likely related to mastication and transport of the resultant mixture of water and food particles. All sirenian tongues are relatively immobile, therefore any gustatory cues would need to be transported to the sides of the tongue. In Florida manatees, this role appears to be played by a series of narrow grooves on either side of the tongue that leads from the upper oral cavity to the root of the tongue, where the taste buds are located. This anatomical design would allow for the mixture to be tasted before the bolus was transported to the back of the mouth, thus allowing for quicker expulsion of non-desired food material. The same research group discovered that male and female manatees possess anal glands (Bills et al. 2013). Anal glands are used for scent marking in a wide variety of species, in the context of territorial marking and reproduction. Because manatees are not territorial, these researchers suggested that in manatees the anal glands may be used to communicate reproductive status. This hypothesis is supported by the observation that the glands appear to be enlarged during the warmer months when manatees are more actively engaged in reproductive behavior. Furthermore, there are anecdotal reports of male manatees investigating underwater objects after female manatees have rubbed against them with the ventral side of their body where the anal glands are located. Thus, the signal generated by the anal glands may be detected by taste buds. This makes sense in the aquatic environment where airborne signals detected via olfaction would be far less effective. Clearly much more work on this sensory modality is warranted.

54

C. D. Marshall et al.

2.3.7 Olfaction As obligate marine mammals, sirenians have little apparent use for olfactory cues and exhibit commensurately reduced olfactory systems. Indeed, sirenians have a relatively small amount of olfactory epithelium in the caudal nasal cavity compared to terrestrial mammals (Barboza and Larkin 2020b), small olfactory bulbs and lateral olfactory tracts, and cortically exhibit an “aborted” rhinal fissure and reduced olfactory cortex (Elliott Smith 1902; Reep et al. 1989; Reep and O’Shea 1990). Manatees and dugongs also lack a vomeronasal organ (Mackay-Sim et al. 1985; MartinezMarcos 2009), present in many semi-aquatic and terrestrial species to facilitate chemical signal detection and pheromone-driven mating behaviors. Sirenians may therefore rely more on taste for chemoreception underwater. This could involve detection of hormones released by conspecifics, detection of salinity changes, or avoidance of toxic plants (Best 1981).

2.4 Conclusions Selection for large body size, and key re-organization of the internal body plan, were key factors in the evolution of sirenians to function as aquatic herbivores. The large body accommodates an expanded gastrointestinal system necessary for digesting large quantities of relatively low-quality herbivorous food, facilitates thermoregulation within the constraint of a low metabolic rate, accommodates increased bone density needed for hydrostatic control, obviates predation, affects locomotory dynamics, and likely increases the usefulness of hydrodynamic stimuli. Unusual blubber and multiple vascular counter-current heat exchangers combine with large body sizes to aid thermoregulation. The need to ingest large quantities of food per day puts a premium on feeding efficiency, which is aided by the development of the facial muscular-vibrissal complex. In manatees, continuous tooth replacement has expanded the variety of usable food resources to include true grasses that are high in silica. This in turn has allowed for expansion of the geographic range of manatees into Florida, which is at the northern limit of their ability to tolerate cold weather (as well as tooth-abrading quartz sand mixed with their seagrass intake). Other novel innovations include independent hemidiaphragm that allow for enhanced control of buoyancy and body trim in these generally slow-moving mammals. The presence of sensory hairs on the entire body has driven the development of specializations in regions of the central nervous system that process somatic sensory information. The muscular-vibrissal system, and oral disk, play important roles in tactile investigation of the environment and in interactions with conspecifics. The presence of body-wide sensory hairs enabled the development and usefulness of hydrodynamic reception. This specialized function of the somatic sensory system has facilitated success in low-light aquatic environments in ways we are just beginning to appreciate.

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

55

Several pieces of evidence point to possible interactions between the somatic sensory and auditory modalities at the level of the cerebral cortex, and investigations along these lines hold the promise of revealing novel mechanisms of perception. Chemosensation is poorly understood in sirenians, but recent findings suggest that it may play an important role in taste discrimination and signaling between conspecifics. A clear understanding of these morphological and sensory innovations for an aquatic lifestyle is necessary to further understand the evolution and mechanisms underlying sirenian ethology and behavioral ecology.

References Ambati BK, Nozaki M, Singh N et al (2006) Corneal avascularity is due to soluble VEGF receptor-1. Nature 443:993–997 Anderson PK, Barclay RMR (1995) Acoustic signals of solitary dugongs: physical characteristics and behavioral correlates. J Mammal 76:1226–1237. https://doi.org/10.2307/1382616 Anderson PK, Birtles A (1978) Behaviour and ecology of the Dugong, Dugong dugon (Sirenia): observations in Shoalwater and Cleveland Bays, Queensland. Wildl Res 5:1–23 Aragones LV (1994) Observations on dugongs at Calauit Island, Busuanga, Palawan, Phillipines. Wildl Res 21:709–717 Bachteler D, Dehnhardt G (1999) Active touch performance in the Antillean manatee: evidence for a functional differentiation of facial tactile hairs. Zoology (Jena) 102:61–69 Barboza MLB, Larkin IV (2020a) Functional morphology of the taste buds of Florida manatee, Trichechus manatus latirostris. Mar Mamm Sci 1–14.https://doi.org/10.1111/mms.12692 Barboza MLB, Larkin IV (2020b) Gross and microscopic anatomy of the nasal cavity, including olfactory epithelium, of the Florida manatee (Trichechus manatus latirostris). Aquat Mamm 46:274–284. https://doi.org/10.1578/AM.46.3.2020.274 Bauer GB, Colbert DE, Gaspard JC et al (2003) Underwater visual acuity of Florida manatees (Trichechus manatus latirostris). Int J Comp Psychol 16:130–142 Bauer GB, Gaspard JC III, Colbert DE et al (2012) Tactile discrimination of textures by Florida Manatees (Trichechus manatus latirostris). Mar Mamm Sci 28:456–471. https://doi.org/10.1111/ j.1748-7692.2012.00565.x Bauer GB, Reep RL, Marshall CD (2018) The tactile senses of marine mammals. Int J Comp Psychol 31. https://doi.org/10.46867/ijcp.2018.31.02.01 Bazzini MD, Reynolds JE III, Essman RA (1986) Erythropoiesis and granulopoiesis in the West Indian manatee, Trichechus manatus (Mammalia: Sirenia). Cells Tissues Organs 126:150–152 Bengtson JL (1983) Estimating food consumption of free-ranging manatees in Florida. J Wildl Manage 47:1186–1192 Bengtson JL, Fitzgerald SM (1985) Potential role of vocalizations in West Indian manatee. J Mamm 66:816–819. https://doi.org/10.2307/1380821 Best RC (1981) Foods and feeding habits of wild and captive Sirenia. Mammal Rev 11:3–29 Bills ML, Samuelson DA, Larkin IV (2013) Anal glands of the Florida manatee, Trichechus manatus latirostris, a potential source of chemosensory signal expression. Mar Mamm Sci 29:280–292. https://doi.org/10.1111/j.1748-7692.2012.00564.x Bleckmann H (1994) Reception of hydrodynamic stimuli in aquatic and semiaquatic animals. Progr Zool 41:1–115 Bonde RK, O’Shea TJ, Beck CA (1983) Manual of Procedures for the salvage and necropsy of carcasses of the West Indian manatee (Trichechus manatus). National Technical Information Service, Springfield, VA. Doc. No. PB83-255273, 175 p

56

C. D. Marshall et al.

Bossart GD, Meisner RA, Rommel SA et al (2002) Pathological features of the Florida manatee cold stress syndrome. Aquat Mamm 29:9–17 Brecht M, Preilowski B, Merzenich MM (1997) Functional architecture of the mystacial vibrissae. Behav Brain Res 84:81–97. https://doi.org/10.1016/S0166-4328(97)83328-1 Bryden MM, Marsh H, MacDonald BW (1978) The skin and hair of the dugong, Dugong dugon. J Anat 126:637–638 Buchholtz EA, Booth AC, Webbink KE (2007) Vertebral anatomy in the Florida manatee, Trichechus manatus latirostris: a developmental and evolutionary analysis. Anat Rec 290:624–637. https:// doi.org/10.1002/ar.20534 Budelmann BU (1989) Hydrodynamic receptor systems in invertebrates. In: Coombs S, Gorner P, Munz H (eds) The mechanosensory lateral line, neurobiology and evolution. Springer, New York, pp 607–632 Bullock TH, Domning DP, Best RC (1980) Evoked brain potentials demonstrate hearing in a manatee (Trichechus inunguis). J Mamm 61:130–133. https://doi.org/10.2307/1379969 Bullock TH, O’Shea TJ, McClune MC (1982) Auditory evoked potentials in the West Indian Manatee Sirenia: Trichechus manatus. J Comp Physiol 148:547–554 Burn DM (1986) The digestive strategy and efficiency of the West Indian manatee, Trichechus manatus. Comp Biochem Physiol 85:139–142. https://doi.org/10.1016/0300-9629(86)90475-5 Bush H, Ducker G (1987) Das visuelle leistungsvermögen der seebären (Arctocephalus pusillus und Arctocephalus australis). Zool Anz 219:124–197 Catania KC, Kaas JH (1997) Somatosensory fovea in the star-nosed mole: behavioral use of the star in relation to innervation patterns and cortical representation. J Comp Neurol 387:215–233 Catania KC, Remple MS (2002) Somatosensory cortex dominated by the representation of teeth in the naked mole-rat brain. PNAS 99:5692–5697. https://doi.org/10.1073/pnas.072097999 Chapla ME, Nowacek DP, Rommel SA et al (2007) CT scans and 3D reconstructions of Florida manatee (Trichechus manatus latirostris) heads and ear bones. Hear Res 228:123–135. https:// doi.org/10.1016/j.heares.2007.01.029 Chen J, Jono T, Cui J et al (2016) The acoustic properties of low intensity vocalizations match hearing sensitivity in the Webbed-Toed Gecko, Gekko subpalmatus. Plos One 11:e0146677.https://doi. org/10.1371/journalpone.0146677 Clifton KB, Yan J, Mecholsky JJ et al (2008) Material properties of manatee rib bone. J Zool (lond) 274:150–159. https://doi.org/10.1111/j.1469-7998.2007.00366.x Cohen JL, Tucker GS, Odell DK (1982) The photoreceptors of the West Indian manatee. J Morphol 173:197–202. https://doi.org/10.1002/jmor.1051730207 Coombs S, Montgomery JC (1999) The enigmatic lateral line system. In: Fay RR, Popper AN (eds) Comparative hearing: fish and amphibians. Springer, New York, pp 319–362 Crish SD, Comer CM, Marasco PD et al (2003) Somatosensation in the superior colliculus of the star-nosed mole. J Comp Neurol 464:415–425. https://doi.org/10.1002/cne.10791 Czech-Damal NU, Liebschner A, Miersch L et al (2011) Electroreception in the Guiana dolphin (Sotalia guianensis). Proc Royal Soc B 279:663–668. https://doi.org/10.1098/rspb.2011.1127 Davies RO, Kare MR, Cagan RH (1979) Distribution of taste buds on fungiform and circumvallate papillae of the bovine tongue. Anat Rec 195:443–446. https://doi.org/10.1002/ar.1091950304 De BV, Schoevaert D (1989) Donnèes quantitatives et observations histologiques sur la “pachyostose” du squelette du dugong, Dugong dugon (Muller) (Sirenia, Dugongidae). Can J Zool 67:2107–2119. https://doi.org/10.1139/z89-300 Dehnhardt G, Friese C, Sachser N (1997) Sensitivity of the trunk of Asian elephants for texture differences of actively touched objects. Zeitschrift Fur Saugetier 62:37–39 Dehnhardt G, Mauck B, Bleckmann H (1998) Seal whiskers detect water movements. Nature 394:235–236 Dehnhardt G, Mauck B, Hanke W et al (2001) Hydrodynamic trail-following in harbor seals (Phoca vitulina). Science 293:102–104. https://doi.org/10.1126/science.1060514

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

57

Deutsch C, Edwards H, Barlas M (2006) Florida manatee foraging behavior around a winter warm-water refuge. Final grant report to the Wildlife Foundation of Florida. FWC/FWRI file FCWT0405–12, p 42 Dexler H (1913) Das Hirn von Halicore dugong Erxl. Gegenbaurs Morphologisches Jahrb 45:97– 190 Dexler H, Freund L (1906) Contributions to the physiology and biology of the dugong. Am Nat 40:49–72 Domning DP (1977) Observations on the myology of Dugong dugon (Müller). Smithsonian contributions to zoology, pp 1–57.https://doi.org/10.5479/si.00810282.226 Domning DP (1978) The myology of the Amazonian manatee, Trichechus inunguis (Natterer) (Mammalia: Sirenia). Acta Amazon 8:1–81. https://doi.org/10.1590/1809-439219780821s005 Domning DP (1980) Feeding position preference in manatees (Trichechus). J Mammal 61:544–547. https://doi.org/10.2307/1379851 Domning DP (1982) Evolution of manatees: a speculative history. J Paleontol 56:599–619 Domning DP (2000) The readaptation of Eocene sirenians to life in water. Hist Biol 14:115–119. https://doi.org/10.1080/10292380009380559 Domning DP, Beatty BL (2007) Use of tusks in feeding by dugongid sirenians: observations and tests of hypotheses. Anat Rec 290:523–538. https://doi.org/10.1002/ar.20540 Domning DP, de Buffrénil V (1991) Hydrostasis in the Sirenia: quantitative data and functional interpretations. Mar Mamm Sci 7:331–368. https://doi.org/10.1111/j.1748-7692.1991.tb00111.x Domning DP, Hayek LA (1984) Horizontal tooth replacement in the Amazonian manatee (Trichechus inunguis). Mammalia 48:105–128. https://doi.org/10.1515/mamm.1984.48.1.105 Dosch F (1915) Bau und entwicklung des integuments der sirenen. Jenaische Zeitschr 53:805–854 Dykes RW (1975) Afferent fibers from mystacial vibrissae of cats and seals. J Neurophysiol 38:650– 662. https://doi.org/10.1152/jn.1975.38.3.650 Elliott Smith G (1902) Descriptive and illustrated catalogue of the physiological series of comparative anatomy in the Museum of the Royal College of Surgeons of England, 2nd edn. Royal College of Surgeons, London. https://doi.org/10.5962/bhl.title.100654 Elsner R (2007) Living in water: solutions to physiological problems. Reynolds JE III. Rommel SA. Biology of marine mammals. Smithsonian Institution Press, Washington, DC, pp 73–116 Erdsack N, Phillips SRM, Rommel SA et al (2018) Heat flux in manatees: an individual matter and a novel approach to assess and monitor the thermal state of Florida manatees (Trichechus manatus latirostris). J Comp Physiol B 188:717–727. https://doi.org/10.1007/s00360-018-1152-7 Eros C, Marsh H, Bonde R et al (2007) Procedures for the salvage and necropsy of the dugong (Dugong dugon), 2nd edn. Great Barrier Reef Marine Park Authority, Research Publication No. 85, 98 pp Etheridge K, Rathbun GB, Powell JA et al (1985) Consumption of aquatic plants by the West Indian manatee. J Aquat Plant Manag 23:21–25 Evans HE, De Lahunta A (2013) Miller’s anatomy of the dog-e-book. In: Elsevier health sciences Fawcett DW (1942) A comparative study of blood-vascular bundles in the Florida manatee and in certain cetaceans and edentates. J Morph 71:105–133. https://doi.org/10.1002/jmor.1050710106 Fish FE (2001) A mechanism for evolutionary transition in swimming mode by mammals. In: de Buffrénil V, Mazin JM (eds) Secondary adaptation of tetrapods to life in water. Pfeil Verlag, München, pp 261–287 Fish FE, Hui CA (1991) Dolphin swimming-a review. Mamm Rev 21:181–195.https://doi.org/10. 1111/j.1365-2907.1991.tb00292.x Gallivan GJ, Best RC (1980) Metabolism and respiration of the Amazonian manatee (Trichechus inunguis). Physiol Zool 53:245–253. https://doi.org/10.1086/physzool.53.3.30155787 Gallivan GJ, Best RC, Kanwisher JW (1983) Temperature regulation in the Amazonian manatee Trichechus inunguis. Physiol Zool 56:255–262. https://www.jstor.org/stable/30156057 Gaspard JC, Bauer GB, Reep RL et al (2012) Audiogram and auditory critical ratios of two Florida manatees (Trichechus manatus latirostris). J Exp Biol 215:1442–1447. https://doi.org/10.1242/ jeb.089201

58

C. D. Marshall et al.

Gaspard JC, Bauer GB, Reep RL et al (2013) Detection of hydrodynamic stimuli by the Florida manatee (Trichechus manatus latirostris). J Comp Physiol A Neuroethol Sens Neural Behav Physio 199:441–450. https://doi.org/10.1007/s00359-013-0822-x Gaspard JC, Bauer GB, Mann DA, Boerner K, Denum L, Frances C, Reep RL (2017) Detection of hydrodynamic stimuli by the postcranial body of Florida manatees (Trichechus manatus latirostris). J Comp Physiol A 203(2):111–120. https://doi.org/10.1007/s00359-016-1142-8 Gerstein ER, Gerstein L, Forsythe SE et al (1999) The underwater audiogram of the West Indian manatee (Trichechus manatus). J Acoust Soc Am 105:3575–3583. https://doi.org/10.1121/1. 424681 Glaser K, Reynolds JE III (2003) Mysterious manatees. University Press of Florida, Gainesville, FL Gottschaldt KM, Vahle-Hinz C (1981) Merkel cell receptors: structure and transducer function. Science 214:183–186. https://doi.org/10.1126/science.7280690 Goulbourne L, Gonzalez-Socoloske D, Domning DP (In press) Interspecific differences in the molar occlusal surfaces of manatees (Trichechus spp.). Michigan Academician (Michigan Academy of Science, Arts & Letters) Graham AR (2005) Histological examination of the Florida manatee (Trichecus manatus latirostris) integument. M.S. Thesis, University of Florida, College of Veterinary Medicine, Gainesville, FL, USA Griebel U, Schmid A (1996) Color vision in the manatee (Trichechus manatus). Vision Res 36:2747– 2757. https://doi.org/10.1016/0042-6989(95)00323-1 Griebel U, Schmid A (1997) Brightness discrimination ability in the West Indian manatee (Trichechus manatus). J Exp Biol 200:1587–1592. https://doi.org/10.1242/jeb.200.11.1587 Grossman CJ, Hamilton RE, De Wit M et al (2014) The vocalization mechanism of the Florida manatee (Trichechus manatus latirostris). Online J Biol Sci 14:127–149. https://doi.org/10.3844/ ojbsci.2014.127.149 Harper JY, Samuelson DA, Reep RL (2005) Corneal vascularization in the Florida manatee (Trichechus manatus latirostris) and three-dimensional reconstruction of vessels. Vet Ophthalmol 8:89–99. https://doi.org/10.1111/j.1463-5224.2005.00351.x Hartman DS (1979) Ecology and behavior of the manatee (Trichechus manatus) in Florida. Special publication, American Society of Mammalogists, 153 p Hassan ES (1989) Hydrodynamic imaging of the surroundings by the lateral line of the blind cave fish Anoptichthys jordani. In: Coombs S, Gorner P, Munz H (eds) The mechanosensory lateral line, neurobiology and evolution. Springer, New York, pp 217–227 Heinsohn GE, Wake J, Marsh H et al (1977) The dugong (Dugong dugon (Müller)) in the seagrass system. Aquaculture 12:235–248. https://doi.org/10.1016/0044-8486(77)90064-3 Hiiemae KM (2000) Mammalian feeding. In: Schwenk K (ed) Feeding: form, function and evolution in tetrapod vertebrates. Academic Press, San Diego, pp 411–448 Horgan P, Booth D, Nichols C et al (2014) Insulative capacity of the integument of the dugong (Dugong dugon): thermal conductivity, conductance and resistance measured by in vitro heat flux. Mar Biol 161:1395–1407 Hurst LA, Beck CA (1988) Microhistological characteristics of selected aquatic plants of Florida with techniques for the study of manatee food habits. U.S. Fish and Wildlife Service, Biological Report No. 88, 145 p Husar SL (1978) Dugong dugon. Mamm Species 88:1–7 Hyvarinen H (1989) Diving in darkness: whiskers as sense organs of the ringed seal. J Zool (lond) 218:663–678. https://doi.org/10.1111/j.1469-7998.1989.tb05008.x Hyvarinen H (1995) Structure and function of the vibrissae of the ringed seal (Phoca hispada L). In: Kastelein RA, Thomas JA, Nachtigall PA (eds) Sensory systems of aquatic mammals. De Spil Publishers, Woerden, pp 429–445 Innes S, Worthy GAJ, Lavigne DM et al (1990) Surface areas of phocid seals. Can J Zool 68:2531– 2538. https://doi.org/10.1139/z90-354

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

59

Irvine AB (1983) Manatee metabolism and its influence on distribution in Florida. Biol Cons 25:315–334. https://doi.org/10.1016/0006-3207(83)90068-X Jarman PJ (1966) The status of the dugong (Dugong dugon Müller); Kenya, 1961. East Afr Wildl J 4:82–88. https://doi.org/10.1111/j.1365-2028.1966.tb00885.x Jerison HJ (1973) Evolution of the Brain and Intelligence. Academic Press, New York Johnson JI (1980) Morphological correlates of specialized elaborations in somatic sensory cerebral neocortex. In: Ebbesson SOE (ed) Comparative neurology of the telencephalon. Plenum Press, New York, pp 423–447 Johnson JI (1990) Comparative development of somatosensory cortex. Cerebral cortex. In: Jones EG, Peters A (eds) Comparative structure and evolution of cerebral cortex, vol 8B, Part II. Plenum Press, New York, pp 335–450 Johnson JI, Kirsch JA (1993) Phylogeny through brain traits: interordinal relationships among mammals including primates and chiroptera. In: MacPhee RDE (ed) Primates and their relatives in phylogenetic perspective. Plenum, New York, pp 293–331 Johnson JI, Kirsch JA, Reep RL et al (1994) Phylogeny through brain traits: more characters for the analysis of mammalian evolution. Brain Behav Evol 43:319–347. https://doi.org/10.1159/000 113643 Jones A, Marshall CD (2019) Does vibrissal innervation patterns and investment predict hydrodynamic trail following behavior of harbor seals (Phoca vitulina)? Anat Rec 302:1837–1845. https://doi.org/10.1002/ar.24134 Kamiya T, Yamasaki F (1981) A morphological note on the sinus hair of the dugong. In: Marsh H (ed) The dugong: Proceedings of a seminar/workshop held at James Cook University, 8–13 May 1979, James Cook University, North Queensland, Australia, pp 111–113 Keith Diagne LW (2014) Phylogenetics and feeding ecology of the African manatee, Trichechus senegalensis. University of Florida Dissertation, Gainesville, FL, USA, 263 p Kenchington RA (1972) Observations on the digestive system of the dugong, Dugong dugon (Erxleben). J Mamm 53:884–887. https://doi.org/10.2307/1379226 Kerjan G, Gleeson JG (2007) Genetic mechanisms underlying abnormal neuronal migration in classical lissencephaly. Trends Genet 23:623–630. https://doi.org/10.1016/j.tig.2007.09.003 Ketten DR, Odell DK, Domning DP (1992) Structure, function and adaptation of the manatee ear. In: Thomas JA, Kastelein RA, Supin AY (eds) Marine mammal sensory systems. Springer, Boston, MA, pp 77–95. https://doi.org/10.1007/978-1-4615-3406-8_4 Kier WM, Smith KK (1985) Tongues, tentacles and trunks: the biomechanics of movement in muscular-hydrostats. Zool J Linn Soc 83:307–324. https://doi.org/10.1111/j.1096-3642.1985. tb01178.x Kipps EK, Mclellan WA, Rommel SA et al (2002) Skin density and its influence on buoyancy in the manatee (Trichechus manatus latirostris), harbor porpoise (Phocoena phocoena), and bottlenose dolphin (Tursiops truncatus). Mar Mamm Sci 18:765–778. https://doi.org/10.1111/j.1748-7692. 2002.tb01072.x Kojeszewski T, Fish FE (2007) Swimming kinematics of the Florida manatee (Trichechus manatus latirostris): hydrodynamic analysis of an undulatory mammalian swimmer. J Exp Biol 210:2411– 2418. https://doi.org/10.1242/jeb.02790 Komatsu S, Yamasaki F (1980) Formation of the pits with taste buds at the lingual root in the striped dolphin, Stenella coeruleoalba. J Morph 164:107–119. https://doi.org/10.1002/jmor.1051640202 Kvadsheim PH, Aarseth JJ (2002) Thermal function of phocid seal fur. Mar Mamm Sci 18:952–962. https://doi.org/10.1111/j.1748-7692.2002.tb01084.x Laist DW, Reynolds J III (2005) Influence of power plants and other warm-water refuges on Florida manatees. Mar Mamm Sci 21:739–764. https://doi.org/10.1111/j.1748-7692.2005.tb01263.x Landrau-Giovannetti N, Mignucci-Giannoni AA, Reidenberg JS (2014) Acoustical and anatomical determination of sound production and transmission in West Indian (Trichechus manatus) and Amazonian (T. inunguis) manatees. Anat Rec 297:1896–1907. https://doi.org/10.1002/ar.22993 Lanyon JM, Marsh H (1995) Digesta passage times in the dugong. Austral J Zool 43:119–127. https://doi.org/10.1071/ZO9950119

60

C. D. Marshall et al.

Lanyon JM, Sanson GD (2006a) Mechanical disruption of seagrass in the digestive tract of the dugong. J Zool (lond) 270:277–289. https://doi.org/10.1111/j.1469-7998.2006.00135.x Lanyon JM, Sanson GD (2006b) Degenerate dentition of the dugong (Dugong dugon), or why a grazer does not need teeth: morphology, occlusion and wear of mouthparts. J Zool 268:133–152. https://doi.org/10.1111/j.1469-7998.2005.00004.x Lanyon JM, Newgrain K, Alli TSS (2006) Estimation of water turnover rate in captive dugongs (Dugong dugon). Aquat Mamm 32:103–108. https://doi.org/10.1578/AM.32.1.2006.103 Larkin, IV, Fowler VF, Reep RL (2007) Digesta passage rates in the Florida manatee (Trichechus manatus latirostris). Zoo Biol 26:503–515. https://doi.org/10.1002/zoo.20150 Ledder DA (1986) Food habits of the West Indian manatee, Trichechus manatus latirostris in south Florida. M.S. thesis, University of Miami, Coral Gables, FL. 138 p Lefebvre LW, Reid JP, KenworthyWJ PJA (2000) Characterizing manatee habitat use and seagrass grazing in Florida and Puerto Rico: implications for conservation and management. Pac Conserv Biol 5:296–298. https://doi.org/10.1071/PC000289 Leitch DB, Catania KC (2012) Structure, innervation and response properties of integumentary sensory organs in crocodilians. J Exp Biol 214:4217–4230. https://doi.org/10.1242/jeb.076836 Levin MJ, Pfeiffer CJ (2002) Gross and microscopic observations on the lingual structure of the Florida Manatee Trichechus manatus latirostris. Anat Histol Embryol 31:278–285. https://doi. org/10.1046/j.1439-0264.2002.00387.x Ling JK (1977) Vibrissae of marine mammals. In: Harrison JB (ed) Functional anatomy of marine mammals, vol 3. Academic Press, London, pp 387–415 Liwanag HEM, Berta A, Costa DP et al (2012) Morphological and thermal properties of mammalian insulation: the evolution of fur for aquatic living. Biol J Linn Soc 106:926–939. https://doi.org/ 10.1111/j.1095-8312.2012.01900.x Lockyer CH, McConnell LC, Waters TD (1985) Body condition in terms of anatomical and biochemical assessment of body fat in north Atlantic fin and sei whales. Can J Zool 63:2328–2338. https:// doi.org/10.1139/z85-345 Lomolino MV, Ewel KC (1984) Digestive efficiencies of the West Indian manatee (Trichechus manatus). Fl Sci 47:176–179. https://www.jstor.org/stable/24319928 Lucchini K, Umeed R, Guimaraes L et al (2021) The role of touch in captive and semi-captive Antillean manatees (Trichechus manatus manatus). Behaviour. https://doi.org/10.1163/1568539X-bja 10069 Mack A, Singh A, Gilroy C et al (1997) Porcine lingual taste buds: a quantitative study. Anat Rec 247:33–37. https://doi.org/10.1002/(SICI)1097-0185(199701)247:1%3c33:: AID-AR5%3e3.0.CO;2-G Mackay-Sim A, Duvall D, Graves BM (1985) The West Indian manatee (Trichechus manatus) lacks a vomeronasal organ. Brain Behav Evol 27:186–194. https://doi.org/10.1159/000118729 Mann DA, Colbert DE, Gaspard JC et al (2005) Auditory temporal resolution of the manatee (Trichechus manatus latirostris) auditory system. J Comp Physiol A 191:903–908. https://doi. org/10.1007/s00359-005-0016-2 Mann DA, O’Shea TJ, Nowacek DP (2006) Nonlinear dynamics in manatee vocalizations. Mar Mamm Sci 22:548–555. https://doi.org/10.1111/j.1748-7692.2006.00036.x Marsh H (1980) Age determination of the dugong (Dugong dugon (Muller)) in northern Australia and its biological implications. In: Perrin WF, Jr Myrick AC (eds) Age determination of toothed whales and sirenians. Proceedings of the international conference on determining age of odontocete cetaceans (and sirenians) La Jolla, Calif. 5–19 Sep 1978. International Whaling Commission Special Issue No. 3, pp 181–201 Marsh H (2009) Dugong: Dugong dugon. In: Perrin WF, Würsig B, Thewissen JGM (eds) Encyclopedia of marine mammals. Academic Press, San Diego, pp 332–335 Marsh H, Heinsohn GE, Spain AV (1977) The stomach and duodenal diverticula of the dugong (Dugong dugon). In: Harrison RJ (ed) Functional anatomy of marine mammals, vol 3. Academic Press, London, pp 271–294

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

61

Marsh H, Beck CA, Vargo T (1999) Comparison of the capabilities of dugongs and West Indian manatees to masticate seagrasses. Mar Mamm Sci 15:250–255. https://doi.org/10.1111/j.17487692.1999.tb00798.x Marsh H, O’Shea TJ, Reynolds III JE (2011) Ecology and Conservation of the Sirenia. Cambridge University Press Conservation Biology No. 18, New York, 521 pp Marsh H, Grech A, McMahon K (2018) Dugongs: seagrass community specialists. In: Larkum AWD, Kendrick GA, Ralph PJ (eds) Seagrasses of Australia. Springer, Cham, pp 629–661. https://doi.org/10.1007/978-3-319-71354-0_19 Marshall CD (2002) Marine mammal functional morphology. In: Perrin WF, Würsig B, Thewissen JGM (eds) Encyclopedia of marine mammals. Academic Press, San Diego Marshall CD (2016) Morphology of the bearded seal (Erignathus barbatus) muscular-vibrissal complex: a functional model for phocid subambient pressure generation. Anat Rec 299:1043– 1053. https://doi.org/10.1002/ar.23377 Marshall J (2017) Vision and lack of vision in the ocean. Curr Biol 27:R494–R502. https://doi.org/ 10.1016/j.cub.2017.03.012 Marshall CD, Goldbogen J (2016) Marine mammal feeding mechanisms. In: Castellini MA, Mellish JA (eds) Functional anatomy of marine mammals, vol 3. CRC Press, pp 95–117 Marshall CD, Pyenson ND (2019) Feeding in aquatic mammals: an evolutionary and functional approach. In: Bels V, Whishaw IQ (eds) Feeding in vertebrates: evolution, morphology, behavior, biomechanics. Springer, pp 743–785 Marshall CD, Reep RL (1995) Manatee cerebral cortex: cytoarchitecture of the caudal region in Trichechus manatus latirostris. Brain Behav Evol 45:1–18. https://doi.org/10.1159/000113381 Marshall CD, Huth GD, Edmonds VM et al (1998a) Prehensile use of perioral bristles during feeding and associated behaviors of the Florida manatee (Trichechus manatus latirostris). Mar Mamm Sci 14:274–289. https://doi.org/10.1111/j.1748-7692.1998.tb00716.x Marshall CD, Clark LA, Reep RL (1998b) The muscular hydrostat of the Florida manatee (Trichechus manatus latirostris) and its role in the use of perioral bristles. Mar Mamm Sci 14:290–303. https://doi.org/10.1111/j.1748-7692.1998.tb00717.x Marshall CD, Kubilis PS, Huth GD et al (2000) Prehensile use of perioral bristles by Florida manatees during feeding upon several species of aquatic vegetation. J Mamm 81:649–658 Marshall CD, Maeda H, Iwata M et al (2003) Orofacial morphology and feeding behaviour of the dugong, Amazonian, West African and Antillean manatees (Mammalia: Sirenia): functional morphology of the muscular–vibrissal complex. J Zool (lond) 259:1–16. https://doi.org/10.1017/ S0952836902003205 Marshall CD, Hsu RH, Herring SW (2005) Somatotopic organization of perioral musculature innervation within the pig facial motor nucleus. Brain Behav Evol 66:22–34. https://doi.org/10.1159/ 000085045 Marshall CD, Amin H, Kovacs K et al (2006) Microstructure and innervation of the vibrissal follicle-sinus complex in the bearded seal, Erignathus barbatus (Pinnipedia: Phocidae). Anat Rec 288A:13–25. https://doi.org/10.1002/ar.a.20273 Marshall CD, Vaughn SD, Sarko D et al (2007) Topographical organization of the facial motor nucleus in Florida manatees (Trichechus manatus latirostris). Brain Behav Evol 70:164–173. https://doi.org/10.1159/000104307 Marshall CD, Rozas K, Kot B et al (2014) Innervation Patterns of Sea Otter (Enhydra lutris) Mystacial follicle-sinus complexes: Support for a specialized somatosensory system. Front Neuroanat 8:121. https://doi.org/10.3389/fnana.2014.00121 Martínez-Marcos A, Halpern M (2009) Evolution of olfactory and vomeronasal systems. In: Binder MD, Hirokawa N, Windhorst U (eds) Encyclopedia of neuroscience. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-29678-2_3135 Mass AM, Ketten DR, Odell DK et al (2012) Ganglion cell distribution and retinal resolution in the Florida manatee, Trichechus manatus latirostris. Anat Rec 295:177–186. https://doi.org/10. 1002/ar.21470

62

C. D. Marshall et al.

Mattson EE, Marshall CD (2016) Follicle microstructure and innervation vary between pinniped micro- and macrovibrissae. Brain Behav Evol 88:43–58. https://doi.org/10.1159/000447551 McGovern KA, Davis RW, Marshall CD (2014) Are vibrissae of northern elephant seals viable sensory structure for prey capture? Anat Rec 298:750–760. https://doi.org/10.1002/ar.2306100 Mercado E III (2014) Tubercles: what sense is there? Aquat Mamm 40:95–103. https://doi.org/10. 1578/AM.40.1.2014.95 Miculka TA, Worthy GAJ (1995) Metabolic capabilities and the limits to thermoneutrality in juvenile and adult West Indian manatees (Trichechus manatus). Published Abstract, Eleventh Biennial Conference on the Biology of Marine Mammals, Orlando Mitchell J (1978) Incremental growth layers in the dentine of dugong incisors (Dugong dugon (Müller)) and their application to age determination. Zool J Linn Soc 62:317–348. https://doi. org/10.1111/j.1096-3642.1978.tb01043.x Montgomery JC, Baker CF, Carton AG (1997) The lateral line can mediate rheotaxis in fish. Nature 389:960–963 Morley JW, Goodwin AW (1983) Tactile discrimination of gratings. Exp Brain Res 49:291–299 Mota B, Herculano-Houzel S (2012) How the cortex gets its folds: an inside-out, connectivity-driven model for the scaling of mammalian cortical folding. Front Neuroanat 6:1–14. https://doi.org/10. 3389/fnana.2012.00003 Murie J (1872) On the form and function of the manatee (Manatus americanus). Tran R Soc Lond 8:127–202. https://doi.org/10.1111/j.1469-7998.1872.tb08571.x Natiello M, Samuelson D (2005) Three-dimensional reconstruction of the angioarchitecture of the ciliary body of the West Indian manatee (Trichechus manatus). Vet Ophthalmol 8:367–373. https://doi.org/10.1111/j.1463-5224.2005.00406.x Natiello MP, Lewis P, Samuelson D (2005) Comparative anatomy of the ciliary body of the West Indian manatee (Trichechus manatus) and selected species. Vet Ophthalmol 8:375–385. https:// doi.org/10.1111/j.1463-5224.2005.00408.x Newman LA, Robinson PR (2006) The visual pigments of the West Indian manatee (Trichechus manatus). Vision Res 46:3326–3330. https://doi.org/10.1016/j.visres.2006.03.010 Nichols C (2005) Temporal change in dugong body condition and seagrass in Moreton Bay, a temperature environment. Dissertation, The University of Queensland, 55 p Nill EK, Pabst DA, Rommel SA (1999) Does the thick skin of the Florida manatee provide ballast? Am Zool 39:14A-114A Nishiwaki M, Marsh H (1985) Dugong Dugong dugon (Müller, 1776). In: Ridgway SH, Harrison RJ (eds) Handbook of marine mammals, vol 3. Academic Press, London, pp 1–31 Nowacek DP, Casper BM, Wells RS et al (2003) Intraspecific and geographic variation of West Indian manatee (Trichechus manatus spp.) vocalizations (L). J Acoust Soc Am 114:66–69. https://doi. org/10.1121/1.1582862 Nummela S (1995) Scaling of the mammalian middle ear. Hear Res 85:18–30. https://doi.org/10. 1016/0378-5955(95)00030-8 O’Shea TJ, Poché LB (2006) Aspects of underwater sound communication in Florida manatees (Trichechus manatus latirostris). J Mamm 87:1061-1071.https://doi.org/10.1644/06-MAMM-A066R1.1 O’Shea TJ, Reep RL (1990) Encephalization quotients and life-history traits in the Sirenia. J Mamm 71:534–543 Owen HC, Flint M, Limpus CJ et al (2013) Evidence of sirenian cold stress syndrome in dugongs Dugong dugon from Southeast Queensland, Australia. Dis Aquat Org 103:1–7 Pabst SA, Rommel SA, McLellan WA (1999) The functional morphology of marine mammals. In: Reynolds JE III, Rommel SA (eds) Biology of marine mammals. Smithsonian Press, Washington DC, pp 15–72 Petit G, Rochon-Duvigneaud A (1929) L’oeil et la vision de L’Halicore dugong ERXL. Bull Soc Zool De Fr 54:129–138

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

63

Piggins D, Muntz WRA, Best RC (1983) Physical and morphological aspects of the eye of the manatee Trichechus inunquis Natterer 1883: (Sirenia: Mammalia). Mar Behav Physiol 9:111–130. https://doi.org/10.1080/10236248309378588 Pillay P, Manger PR (2007) Order-specific quantitative patterns of cortical gyrification. Eur J Neurosci 25:2705–2712. https://doi.org/10.1111/j.1460-9568.2007.05524.x Preen AR (1992) Interactions between dugongs and seagrasses in a subtropical environment. PhD Thesis, James Cook University, Townsville Preen A (1995) Impacts of dugong foraging on seagrass habitats: observational and experimental evidence for cultivation grazing. Mar Ecol Progr Ser 124:201–213 Ramos EA, Maust-Mohl M, Collom KA et al (2020) The Antillean manatee produces broadband vocalizations with ultrasonic frequencies. J Acoust Soc Am 147:EL80-86. https://doi.org/10. 1121/10.0000602 Reep RL, O’Shea TJ (1990) Regional brain morphometry and lissencephaly in the sirenia. Brain Behav Evol 35:185–194. https://doi.org/10.1159/000115866 Reep RL, Johnson JI, Switzer RC et al (1989) Manatee cerebral cortex: cytoarchitecture of the frontal region in Trichechus manatus latirostris. Brain Behav Evol 34:365–386. https://doi.org/ 10.1159/000116523 Reep RL, Marshall CD, Stoll ML et al (1998) Distribution and innervation of facial bristles and hairs in the Florida manatee (Trichechus manatus latirostris). Mar Mamm Sci 14:257–273. https://doi. org/10.1111/j.1748-7692.1998.tb00715.x Reep RL, Marshall CD, Stoll ML et al (2001) Microanatomy of facial vibrissae in the Florida manatee: the basis for specialized sensory function and oripulation. Brain Behav Evol 58:1–14. https://doi.org/10.1159/000047257 Reep RL, Marshall CD, Stoll ML (2002) Tactile hairs on the postcranial body in Florida manatees: a mammalian lateral line? Brain Behav Evo 59:141–154. https://doi.org/10.1159/000064161 Reep RL, Gaspard JC, Sarko D et al (2011) Manatee vibrissae: evidence for a “lateral line” function. Ann N Y Acad Sci 1225:101–109. https://doi.org/10.1111/j.1749-6632.2011.05992.x Reyes LD, Stimpson CD, Gupta K et al (2015) Neuron types in the presumptive primary somatosensory cortex of the Florida manatee (Trichechus manatus latirostris). Brain Behav Evol 86:210–231. https://doi.org/10.1159/000441964 Reyes LD, Harland T, Reep RL et al (2016) Golgi analysis of neuron morphology in the presumptive primary somatosensory cortex and visual cortex of the Florida manatee (Trichechus manatus latirostris). Brain Behav Evol 87:105–116. https://doi.org/10.1159/000445495 Reynolds JE III (1979) The semisocial manatee. Nat Hist 88:44–53 Reynolds JE III (1981) Aspects of the social behavior and herd structure of a semi-isolated colony of West Indian manatees, Trichechus manatus. Mammalia 45:431–451. https://doi.org/10.1515/ mamm.1981.45.4.431 Reynolds JE III, Rommel SA (1996) Structure and function of the gastrointestinal tract of the Florida manatee, Trichechus manatus latirostris. Anat Rec 245:539–558. https://doi.org/10.1002/(SIC I)1097-0185(199607)245:3%3c539::AID-AR11%3e3.0.CO;2-Q Reynolds III JE, Marshall CD (2012) Vulnerability of sirenians. In: Hines E, Reynolds JE III, Aragones L et al (eds) Sirenian conservation: issues and strategies in developing countries. University of Florida Press Gainesville, FL, pp 12–19 Rice FL (1995) Comparative aspects of barrel structure and development. In: Jones EG, Diamond IT (eds) Cerebral cortex, vol 11. The barrel cortex of rodents. Plenum Press, New York, pp 1–75 Rice FL, Mance A, Munger BL (1986) A comparative light microscopic analysis of the sensory innervation of the mystacial pad. I. Innervation of vibrissal follicle-sinus complexes. J Comp Neurol 252:154–174. https://doi.org/10.1002/cne.902520203 Rommel SA, Caplan H (2003) Vascular adaptations for heat conservation in the tail of Florida manatees (Trichechus manatus latirostris). J Anat 202:343–353. https://doi.org/10.1046/j.14697580.2003.00170.x

64

C. D. Marshall et al.

Rommel S, Reynolds JE III (2000) Diaphragm structure and function in the Florida manatee (Trichechus manatus latirostris). Anat Rec 259:41–51. https://doi.org/10.1002/(SICI)1097-018 5(20000501)259:1%3c41::AID-AR5%3e3.0.CO;2-Q Rommel SA, Pabst DA, McLellan WA (2001) Functional morphology of venous structures associated with the male and female reproductive systems in Florida manatees (Trichechus manatus latirostris). Anat Rec 264:339–347. https://doi.org/10.1002/ar.10022 Rommel SA, Reynolds III JE, Lynch HA (2003) Adaptations of the herbivorous marine mammals. In: Mannetje L, Ramirez-Aviles L, Sandoval-Castro C et al (eds) Matching herbivore nutrition to ecosystems biodiversity. Proceedings of the VIth International symposium on the nutrition of herbivores, Merida, México, 19–24 October 2003, Universidad Autonoma de Yucatan, Mexico, pp 287–308 Rosas FCW (1994) Biology, conservation and status of the Amazonian manatee Trichechus inunguis. Mamm Rev 24:49–59. https://doi.org/10.1111/j.1365-2907.1994.tb00134.x Sarko DK, Reep RL (2007) Somatosensory areas of manatee cerebral cortex: histochemical characterization and functional implications. Brain Behav Evol 69:20–36. https://doi.org/10.1159/000 095028 Sarko DK, Reep RL, Mazurkiewicz JE et al (2007a) Adaptations in the structure and innervation of follicle-sinus complexes to an aquatic environment as seen in the Florida manatee (Trichechus manatus latirostris). J Comp Neurol 504:217–237. https://doi.org/10.1002/cne.21446 Sarko DK, Johnson JI, Switzer RC et al (2007b) Somatosensory nuclei of the brainstem and thalamus in Florida manatees. Anat Rec 290:1138–1165 Sarko DK, Rice FL, Reep RL (2015) Elaboration and innervation of the vibrissal system in the rock hyrax (Procavia capensis). Brain Behav Evol 85:170–188. https://doi.org/10.1159/000381415 Schevill WE, Watkins WA (1965) Underwater calls of Trichechus (manatee). Nature 205:373–374 Schmidt-Nielsen K (1990) Animal physiology: adaption and environment, 4th edn. Cambridge University Press, Cambridge, p 602 Scholander PF, Walters V, Hock R et al (1950) Body insulation of some arctic and tropical mammals and birds. Biol Bull 99:225–236. https://doi.org/10.2307/1538740 Sheppard JK, Lawler IR, Marsh H (2007) Seagrass as pasture for seacows: landscape-level dugong habitat evaluation. Estuar Coast Shelf Sci 71:117–132. https://doi.org/10.1016/j.ecss.2006.07.006 Sokolov VE (1986) Manatee-morphological description. Nauka Press, Moscow Sousa-Lima RS, Paglia AP, Da Fonseca GAB (2002) Signature information and individual recognition in the isolation calls of Amazonian manatees, Trichechus inunguis (Mammalia: Sirenia). Anim Behav 63:301–310. https://doi.org/10.1006/anbe.2001.1873 Sprowls C, Marshall CD (2019) Vibrissal innervation patterns and investment in California sea lions (Zalophus californianus). J Morph 280:1617–1627. https://doi.org/10.1002/jmor.21053 Stephens RJ, Beebe IJ, Poulter TC (1973) Innervation of the vibrissae of the California sea lion, Zalophus californianus. Anat Rec 176:421–441. https://doi.org/10.1002/ar.1091760406 Tol SJ, Jarvis JC, York PH et al (2017) Long distance biotic dispersal of tropical seagrass seeds by marine mega-herbivores. Sci Rep 7:4458. https://doi.org/10.1038/s41598-017-04421-1 Van Essen DC (1997) A tension-based theory of morphogenesis and compact wiring in the central nervous system. Nature 385:313–318 Vaughan T, Ryan J, Czaplewski N (2011). Mammalogy, 5th edn. Jones & Bartlett Learning Velez-Juarbe J, Domning DP, Pyenson ND (2012) Iterative evolution of sympatric seacow (Dugongidae, Sirenia) assemblages during the past ∼26 million years. Plos One 7:e31294. https:// doi.org/10.1371/journal.pone.0031294 Vogel S, Wainwright SA (1969) A functional bestiary. Addison-Wesley, Reading, Massachusetts Ward-Geiger LI (1997) Blubber depth and body condition indices in the Florida manatee (Trichechus manatus latirostris). Dissertation, University of South Florida, p 10 Webb PW (1978) Hydrodynamics: nonscombrid fish. In: Hoar WS, Randall DJ (eds) Fish physiology: Locomotion, vol VII. Academic Press, New York, pp 189–237

2 Morphological and Sensory Innovations for an Aquatic Lifestyle

65

Welker WI (1990) Why does cerebral cortex fissure and fold? A review of determinants of gyri and sulci. In: Jones EG, Peters A (eds) Cerebral cortex, comparative structure and evolution of cerebral cortex, Part II. Plenum Press, New York, pp 3–136 West JA, Sivak JG, Murphy CJ et al (1991) A comparative study of the anatomy of the iris and ciliary body in aquatic mammals. Can J Zool 69:2594–2607. https://doi.org/10.1139/z91-366 Windsor SP (2014) Hydrodynamic imaging by blind Mexican cavefish. In: Bleckmann H, Mogdans J, Coombs SL (eds) Flow sensing in air and water. Springer, Heidelberg, pp 103–125 Woolsey TA, Van der Loos H (1970) The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units. Brain Res 17:205–242. https://doi.org/10.1016/0006-8993(70)90079-x Worthy GAJ, Edwards EF (1990) Morphometric and biochemical factors affecting heat loss in a small temperate cetacean (Phocoena phocoena) and a small tropical cetacean (Stenella attenuata). Physiol Zool 63:432–442. https://doi.org/10.2307/30158506 Worthy GAJ, Worthy TA (2014) Digestive efficiencies of ex situ and in situ West Indian Manatees (Trichechus manatus latirostris). Physiol Biochem Zool 87:77–91. https://doi.org/10.1086/ 673545 Yamasaki F, Komatsu S, Kamiya T (1980) A comparative morphological study on the tongues of manatee and dugong (Sirenia). Sci Rep Whales Res Inst 32:127–144 Yan J, Clifton KB, Mecholsky JJ et al (2006) Fracture toughness of manatee rib and bovine femur using a chevron-notched beam test. J Biomech 39:1066–1074. https://doi.org/10.1016/j.jbiomech. 2005.02.016 Zeh D, Heupel M, Hamann M et al (2018) Evidence of behavioural thermoregulation by dugongs at the high latitude limit to their range in eastern Australia. J Exp Mar Biol Ecol 508:27–34. https:// doi.org/10.1016/j.jembe.2018.08.004 Zelena J (1994) Nerves and mechanoreceptors. Chapman and Hall, London Zilles K, Armstrong E, Schleicher A et al (1988) The human pattern of gyrification in the cerebral cortex. Anat Embryol 179:173–179. https://doi.org/10.1007/BF00304699

Chapter 3

Diving and Foraging Behaviors Lucy W. Keith-Diagne, Margaret E. Barlas, James P. Reid, Amanda J. Hodgson, and Helene Marsh

Abstract Manatees and dugongs live in tropical and semi-tropical regions around the world. Their preferred habitats are seagrass beds, rivers, lakes, and estuaries. Manatees live in both freshwater and marine systems although habitat preferences vary across the three species, while the dugong is entirely marine. Sirenians are shallow water divers, and their dive durations are short compared to most other marine mammals. The maximum recorded manatee dive duration is 24 min, with the maximum recorded duration of a dugong dive being about half that. Even though the durations of dugong dives are shorter than those of manatees, current data indicate that dugongs dive deeper than manatees. Dive depths for manatees generally do not exceed 5 m, other than during occasional travel over deeper water; however, this may be an artifact of water depth in areas where diving data were recorded, or where manatees live. In some parts of their range, dugongs are found over deep-water seagrass beds and dives have been recorded to more than 30 m. All extant sirenians eat diverse plant-based diets: collectively they have been documented feeding on at least 55 genera of marine and freshwater plants. Although not confirmed for the Amazonian manatee, it is likely that all extant sirenians eat animal, as well as plant, matter. West Indian and African manatees have been documented eating marine and freshwater fish and invertebrates, and for African manatees, these food resources are a regular part of their diet, arguably making them omnivores. Dugongs have been L. W. Keith-Diagne (B) African Aquatic Conservation Fund, BP 80, 23015 Joal, Senegal e-mail: [email protected] M. E. Barlas Florida Fish and Wildlife Conservation Commission, Fish and Wildlife Research Institute, 100 Eighth Ave SE, St. Petersburg, FL 33701, USA J. P. Reid U.S. Geological Survey Sirenia Project, Wetland and Aquatic Research Center, 7920 NW 71st St., Gainesville, FL 32653-3071, USA A. J. Hodgson Harry Butler Institute, Murdoch University, South St., Murdoch, WA 6150, Australia H. Marsh College of Science and Engineering, James Cook University, Townsville, QLD 4811, Australia © Springer Nature Switzerland AG 2022 H. Marsh (ed.), Ethology and Behavioral Ecology of Sirenia, Ethology and Behavioral Ecology of Marine Mammals, https://doi.org/10.1007/978-3-030-90742-6_3

67

68

L. W. Keith-Diagne et al.

recorded targeting invertebrates at the high latitude limits of their range in winter. All manatees like to drink fresh water, in contrast to dugongs, which live entirely in marine systems and apparently meet the water requirements from their food. Keywords Breathing · Depth · Diet · Diving · Foraging · Habitat · Respiration · Seagrass · Sirenians

3.1 Introduction The feeding biology of sirenians is the basis of both their popular name, sea cows, and most defining characteristic, aquatic herbivory (Marsh et al. 2011). Thus, their positions in food chains are unlike that of any other marine mammal. In comparison with most other marine mammals, dugongs and manatees are shallow divers, spending much of their lives accessing their largely stationary food sources (Marsh et al. 2011). As explained in the Preface to this book and in Chap. 1, the evolutionary origins of Sirenia are quite separate from those other marine mammals. Like cetaceans, they evolved from a terrestrial ancestor that returned to the ocean. However, unlike cetaceans, and due to their plant-based diets, they never evolved to occupy pelagic waters. Instead, the four species of extant sirenians are subtropical and tropical marine mammals that mostly occur in shallow coastal and riverine environments where vascular plants are available. In this chapter, we describe the diving and feeding behaviors of wild manatees and dugongs, and the methods used to obtain this information. As for other aspects of sirenian ethology and behavioral ecology, most research has involved the Florida manatee (Trichechus manatus latrirostris) and the dugong (Dugong dugon), with much less research on Antillean (Trichechus manatus manatus) and African manatees (Trichechus senegalensis), although in recent years, as outlined below, there has been more work on African manatees. The diving and feeding behaviors of wild Amazonian manatees (Trichechus inunguis) are the least studied. As explained below, much of the research on diving and feeding behavior has been incidental to work on more applied problems, particularly the impact of watercraft on Florida manatees, and estimating abundance in dugongs.

3.2 Diving 3.2.1 Anatomical Adaptations to Diving As explained in Chap. 1, pachyosteosclerotic rib bones, which are solid, dense, and provide ballast, have been a physiological feature of all sirenians as far back as the Eocene. Along with their dense, heavy skin, their heavy ribs enable manatees and

3 Diving and Foraging Behaviors

69

dugongs to control their position in the water column hydrostatically and maintain neutral buoyancy at shallow depths, which allows them to descend without active swimming (see Kipps et al. 2002 for manatees; Chap. 2). In contrast, dives are a hydrodynamic function, achieved by tail thrusts. Manatees propel themselves with undulating movements of their horizontal tail fluke, using it in conjunction with their highly maneuverable pectoral flippers to steer (Hartman 1979), while dugongs propel themselves with movements closer to those of cetaceans using caudal oscillations of the tail (Marsh et al. 2011).

3.2.2 Diving Physiology The only research on the diving physiology of sirenians, published in the 1980s, was on the Amazonian manatee. Several studies indicated that the heart rate of unrestrained, captive Amazonian manatees remained the same during dives, but their breathing became more rapid (Gallivan et al. 1986). Manatees recovered their oxygen debt after 3–4 short dives with breaths in between dives (Gallivan and Best 1980; Gallivan et al. 1986), and, in a pattern similar to cetaceans but different from phocid seals, who recover their gas stores after each dive (Gallivan et al. 1986), several shorter dives generally followed a longer dive (Gallivan and Best 1980). The resting heart rate of the Amazonian manatees was 30–40 beats per minute (bpm), which increased to 50 bpm while at the surface, and then returned to normal 30–40 bpm when the manatees submerged again (Gallivan et al. 1986). Predictably, heart rates rose when manatees swam, ate, or were exposed to stimuli (Gallivan et al. 1986). Due to their relatively constant heart rate and ability to recover quickly after long dives, Gallivan et al. (1986) suggested that Amazonian manatees maintain aerobic metabolism during most dives, although this has not been confirmed for any sirenian.

3.2.3 Surfacing Behavior When surfacing between dives, sirenians generally exhibit a very low profile above the water surface (Marsh et al. 2011) and spend a very short time there, with only their paired nostrils above the surface when breathing. A single breath is taken via the nostrils each time they surface with exhalation followed instantaneously by an inhalation (Hartman 1979). West Indian manatees have sometimes been observed to begin exhaling prior to completely raising their nostrils above the water’s surface, thereby causing bubbles and spray, but they have not been observed exhaling while submerging (Hartman 1979). After longer dives (e.g., 12–15 min), which result in greater oxygen depletion and accumulation of metabolic by-products, individuals will remain near the surface and take several breaths in rapid succession (e.g., 2–4 times within 1 min) before the next prolonged dive. This behavior presumably eliminates carbon dioxide and renews oxygen stores (Parker 1922; Scholander and Irving

70

L. W. Keith-Diagne et al.

1941; Moore 1951; Hartman 1979). This breathing pattern is typically observed when Florida manatees are resting or foraging in deeper water, where they spend long periods underwater, rather than when they are engaged in activities such as traveling or socializing near the surface (Hartman 1979). The time taken for a single breath differs for Florida manatees of different age classes and activity states. Adults with greater lung capacity take longer to complete the actions of exhaling and inhaling than smaller animals with smaller lung capacity (Parker 1922; Hartman 1979). Activity state also has a direct impact on the duration of a single breath; resting Florida manatees may take over a second longer to breathe than those that are swimming (Hartman 1979). Active Florida manatees can shorten their breathing time with forceful exhalations, but inhalation time appears to remain the same (Hartman 1979). However, when disturbed during breathing, individuals will stop inhaling or exhaling and dive (Hartman 1979). There are few data on the surfacing patterns of wild Amazonian or African manatees. Powell (1996) documented surfacing intervals of ≥4 min for resting African manatees and 2–4 min for active manatees. Hodgson (2004) obtained the most accurate and precise estimates of dugong respiration intervals between dives using blimp-cam video-recordings of dugongs in shallow (0.9–4 m), clear water in Moreton Bay, Queensland. She conducted focal follows of 33 individual dugongs and 35 cow-calf pairs. The overall mean of 1100 respiration intervals was 2 ± se 1 s with a range of 1–5 s. On average, dugong calves spent less time with their nostrils above the surface (1 ± se 1 s) than their mothers (2 ± se 1 s), even though the cow-calf pair often surfaced synchronously. Dugong calves often respire multiple times at the surface and commonly move to a position over their mother’s back, or cross over the mother while surfacing and submerging (Hodgson 2004). Some manatee calves also ride on their mother’s back (Hartman 1979; Reynolds 1981). These positions likely provide hydrodynamic advantages for sirenian calves. As for dolphins (Weihs 2004), the drafting force created by the mother counteracts the drag experienced by the calf and reduces the energetic cost of surfacing and submerging (Hodgson 2004). Florida manatee groups including cow-calf pairs, animals traveling together, and mating herds, often surface synchronously, thereby maintaining associations between individuals (Hartman 1979; Reynolds 1981; see Chaps. 2 and 4). In contrast, the only report of synchronous surfacing for dugong groups, other than cow-calf pairs, comes from surveys of fishers in Kenya who reported that groups of dugongs appeared to surface in synchrony when moving together into their feeding grounds (Jarman 1966).

3.2.4 Dive Duration Maximum sirenian dive times are relatively short compared with those of deep diving marine mammals, such as Cuvier’s beaked whale (137 min; Schorr et al. 2014) or southern elephant seals (120 min; Hindell et al. 1992). Dive times vary between taxa,

3 Diving and Foraging Behaviors

71

size class, and activity within each taxon (Hartman 1979; Reynolds 1981; Hodgson 2004). A single Florida manatee dive has been reported to last for up to 24 min, although dive durations are typically less than 10 min (Scholander and Irving 1941; Hartman 1979; Reynolds 1981; Edwards et al. 2007). The capacity of manatees to make long dives has been attributed to their low metabolic rate (Scholander and Irving 1941). Preliminary data suggest that the metabolic rate of dugongs is higher than predicted, based on its phylogeny (Lanyon et al. 2006). The maximum recorded duration of a dugong dive (12.3 min; Chilvers et al. 2004) is much less than that recorded for Florida manatees. The length of time a manatee remains underwater decreases with increased activity; a Florida manatee that is traveling or socializing generally breathes more often than an individual at rest (Hartman 1979; Reynolds 1981; Edwards et al. 2007), because energy expenditure increases with greater activity (Reynolds 1981). Additionally, dive duration for individual Florida manatees engaged in the same activity is not the same for different age classes (i.e., calves, subadults, adults); calves breathe more frequently than larger individuals (Parker 1922; Reynolds 1981). Subadults remain submerged for similar periods to adults, at least while resting (Hartman 1979). The information on the dive durations of Amazonian and African manatees is sparse. Gallivan and Best (1980) reported a mean dive time of less than 2 min for captive Amazonian manatees, but individuals were also recorded diving for over 10 min. Van den Bergh (1968) recorded a submergence time of 7 min for a captive African manatee, but longer dives have been observed in the wild (Keith Diagne pers. obs.). The duration of dugong dives varies both between individuals within studies and between studies. At least some of this variation is likely an artifact of the methodological difficulties associated with: (1) recording dive duration visually in deep and/or turbid water; and (2) defining a dive from a Time-Depth Recorder (TDR) profile. Other sources of variation undoubtedly include environmental conditions, especially water depth. Nonetheless, most mean dive durations of dugongs are short (1.5 m deep, per hour) of the dugong is variable (Chilvers et al. 2004). Individual means ranged from 6.6 to 22.5 dives per hour with an overall mean across all the telemetered dugongs of 11.8 ± se 1.2 dives per hour. Whiting (2002a) reported similar values based on visual observations, but dive rates extrapolated from dive times measured by most other researchers are much higher (see Marsh et al. 2011). Anderson (1982) speculated that foraging behavior, rather than oxygen deficits, determine the duration of dugong dives, and certainly, the maximum dive durations recorded are much shorter than the aerobic dive limits of Amazonian manatees (19–22 min) estimated by Gallivan et al. (1986). The fact that dugongs exhibit shorter dive times in shallow water suggests that it is more energetically efficient to rise the short distances to the surface to breathe than to feed until there is an oxygen deficit. In deeper water, dugongs likely maximize the time spent foraging or resting between surfacings.

3.2.5 Dive Depths Even though dugong dive durations are shorter than those of manatees, dugongs have been recorded diving deeper than manatees. However, this may be an artifact of water depths in the areas where diving data were recorded or where manatees live. In a Tampa Bay habitat use study, Florida manatees dove as deep as was possible in the area in which they were tagged (Edwards et al. 2016). Many of the dives followed the bottom contour, with the deepest at 16.2 m, which corresponded to local bathymetry (Edwards et al. 2016). This indicated that the manatees were diving to the deepest part of the bay, which included the shipping channel. Based on water temperatures, there appeared to be no thermal advantage to diving deeply (Edwards et al. 2007). However, diving may be energetically efficient, as is the case for other marine mammals. Amazonian manatees have been reported diving to depths of 1–4 m (Marmontel et al. 2012), but no further diving studies have been undertaken. From 52 African manatee sightings in coastal lagoons and rivers during surveys in Gabon, manatees were sighted in depths ranging from 0.7 to 5.0 m, with an average depth of 1.75 m across sightings (Keith and Collins 2007). Similarly, during his study in Côte d’Ivoire (Ivory Coast), Akoi documented tagged manatees at depths ranging from 0.53–4.15 m (Akoi 2004). From the little work done so far on African manatee diving capability, they seem to be shallow divers, with preferred depths similar to those recorded for Amazonian manatees. Alternatively, their diving may also be related to the depths of their habitats, rather than their capability, but this has not yet been studied in these species. In contrast, dugongs have been recorded diving to depths of more than 30 m. Using TDRs, Hagihara (2015) recorded a maximum dive depth of 31.5 m and Sheppard et al. (2006) recorded a dugong diving to 36.5 m. Dugong dive depths have also been inferred by recording dugong feeding trails during seagrass surveys, and

3 Diving and Foraging Behaviors

73

Lee Long et al. (1996) recorded dugong feeding trails to 33 m. There are no records of dugongs diving beyond 40 m. TDR studies indicate that most dugong dives are relatively shallow. Chilvers et al. (2004) recorded a mean maximum depth per dive across 15 individual dugongs of 4.8 ± se 0.4 m (analysis restricted to dives >1.5 m); the corresponding statistic for Hagihara (2015) was similar (4.6 ± se 0.04 m) across four individuals.

3.2.6 Diving Behavior Early research on sirenian diving behavior focused on visual observations. The only data for African manatees were obtained using this method, whereas advances in technology have provided greater insight into vertical movements of individual Florida manatees and dugongs. TDRs and Digital Acoustic Recording Tags (DTAGs) on satellite-linked radio-tagged Florida manatees have enabled detailed diving behavior to be described in relation to habitat features (e.g., depth of water and presences of seagrass) and manatee activity (e.g., resting, traveling, foraging). However, research using these tools has concentrated on studying manatee responses to watercraft and assessing the risk of vessel strike (Edwards et al. 2016; Martin et al. 2016; Udell et al. 2019; Chap. 7), rather than on describing diving behavior per se. Dive profiles from TDRs show Florida manatees making consecutive dives to the bottom while traveling, even in areas >14 m deep in Tampa Bay (Edwards et al. 2016) (Fig. 3.1a). Nine individuals spent on average 78% of their time at ≤1.25 m below the surface when away from their winter, warm-water refuge (Edwards et al. 2016; also see Chap. 6). Their depths showed a diel pattern with deepest mean hourly depths during the morning (~1.4–1.6 m, CI = 0.10–0.31) and shallowest at night (~0.8–1.1 m, CI = 0.10–0.25). Manatees were documented traveling fast most often in the early morning hours, possibly due to their thermoregulatory need to quickly return to warm power plant waters after spending time in cooler water away from the plant at night (Edwards et al. 2016). The difference in the proportion of time spent on the surface by time of day may reflect the manatee behavior of swimming more deeply or using deeper water areas, especially during morning trips back to the power plant’s warm water (Edwards et al. 2016). Tide likely influenced these results. Manatees in Tampa Bay are only able to access shallow nearshore seagrass beds during higher tidal maxima, which typically occurs at night (1600–0500; Edwards et al. 2016). Additionally, swimming is energetically expensive, and marine mammals are known to modify their method of locomotion to reduce energy costs (Williams et al. 2000; Williams 2001; Zeh et al. 2018). In winter, Florida manatees fast as they rest in the warm waters of power plant discharge canals. They then travel out and back to feeding sites in cold water, only to return to fasting upon return to the power plant (Edwards et al. 2016; Chap. 6). Therefore, minimizing the amount of energy expended getting to and from foraging grounds is important. Drag is greater at the surface of the water than below it, so staying submerged and using energy-efficient behaviors may reduce the overall cost

74

L. W. Keith-Diagne et al.

Fig. 3.1 Dive profiles of sirenians showing their capacity for successive deeper dives when traveling. a Florida manatee TTB108 crossing Tampa Bay, Florida (west shoreline to Tampa Electric Company’s Big Bend Power Station) from 04:47:55 to 16:47:15 EST, January 31, 2004. Dashed line depicts water temperature recorded by time-depth recorder (from Edwards et al. 2016 with permission); b Sample of a dugong named Bumkaman’s dive profile (4.5 h) during his 284 km large-scale movement from Burrum Heads to Great Keppel Island, Hervey Bay, Queensland. Note repeated dives to >21 m. From Sheppard et al. (2006) with permission

3 Diving and Foraging Behaviors

75

of the long-distance movement (Williams et al. 2000; Williams 2001). For example, a dive profile of a manatee crossing one of the shipping channels in Tampa Bay while returning to the power plant (Fig. 3.1a), illustrates that under some circumstances, manatees make consecutive dives to the bottom, even in the deepest areas of the bay, possibly to conserve energy. Sheppard et al. (2006) noted similar behavior in dugongs, as discussed below. Details of manatee dives varied based on region, due to different bathymetrical features and tidal cycles, as tide and season do not restrict access to many Florida seagrass beds, and in summer, the temperature does not restrict Florida manatee movement (see Chaps. 5 and 6). Amazonian manatee diving behavior has not been studied, and there is also little information about African manatee diving behavior. African manatees are mostly found in shallow, nearshore habitats, at river confluences, or mouths of rivers and lagoons at the sea, which may provide them with options to move quickly to safe locations if disturbed (Powell 1996; Dodman et al. 2008; Mayaka et al. 2019). They are generally timid and will dive and leave an area in response to boats and humans (Akoi 2004; Keith and Collins 2007). As with Florida manatees, a range of techniques has been used to study dugong diving behavior. Early studies were based on visual observations of a captive dugong (Kenny 1967) and wild dugongs from boats or shoreline vantage points (Anderson and Birtles 1978; Anderson 1982; 1998; Marsh and Rathbun 1990; Whiting 2002a). All field observations made the untested assumption that the surfacing of individual animals could be reliably detected despite the often-turbid water. From the mid1990s, researchers have used increasingly sophisticated TDRs to record the diving behavior of satellite-linked radio-tagged dugongs (Chilvers et al. 2004; Sheppard et al. 2006; Hagihara 2015; Hagihara et al. 2011; 2014; 2018). Many of these TDR studies aimed to estimate the proportion of time dugongs were detectable by observers conducting aerial surveys in light aircraft in a range of environmental conditions rather than dugong diving per se. Studies generally obtain data for a large number of dives by relatively few dugongs (e.g., 40,000 dives from 15 dugongs, Chilvers et al. 2004; 150,000 dives from four dugongs, Hagihara 2015) because of the difficulty in catching dugongs, and the need to recover most TDRs to download the archived data. Such studies have been carried out in six locations in Australia (Torres Strait, and Moreton, Hervey, and Shoalwater Bays, Queensland; Gulf of Carpentaria, Northern Territory; Shark Bay, Western Australia). Hagihara et al. (2018) overcame the TDR recovery problem by using TDRs programed to detach and float to the surface after 60 days and then transmit data via satellite to a collection center from which they could be downloaded remotely. They used this technique on six dugongs in Torres Strait, a region that supports extensive seagrass beds to 40 m deep. We are not aware of any published studies on dugongs using three dimensional sensors or accelerometers. Despite their short respiration intervals, TDR recordings indicate that dugongs spend a high proportion of their time close to the surface. Both Chilvers et al. (2004) and Hagihara et al. (2014) found that dugongs in various locations spend more than a third to over half of their time within 1.5 m of the surface, although this proportion varies among locations with different bathymetries (Hagihara et al. 2014; 2018).

76

L. W. Keith-Diagne et al.

Not surprisingly, dugongs spend relatively more time away from the surface when they are over seagrass beds in deeper water (5 to < 20 m), presumably to maximize their feeding time. However, in water >20 m deep, time spent within 1.5 m of the surface is similar to very shallow water in Moreton Bay and Torres Strait. It is not known whether dugongs spend longer at the surface recovering after deeper dives, as manatees do. Dugongs spend less time close to the water surface while making large-scale directed movements (>15 km, sensu Sheppard et al. 2006) suggesting that they travel mid-water rather than at the surface (Chilvers et al. 2004). Sheppard et al. (2006) recorded the diving behavior of telemetered dugongs making large-scale movements in Queensland coastal waters. Like the Florida manatees in Tampa Bay (Edwards et al. 2016; Fig. 3.1a), several dugongs (Fig. 3.1b) made repeated consecutive dives to 27–36.5 m. However, Sheppard et al. (2006) could not determine if these animals were tracking the bottom. They speculated that the dugongs might be using these dives for navigation, a conclusion supported by Hagihara’s (2015) detailed studies of the diving behavior of three telemetered dugongs moving locally in, and between, Moreton Bay and the adjacent oceanic waters. The currents are always smaller close to the bottom due to bottom friction (pers. comm., Eric Wolanski, James Cook University, Australia; written communication to Helene Marsh, 2021). As explained in Chap. 2, their highly developed somatosensatory capacity enables sirenians to detect such hydrodynamic differences. Nonetheless, dugongs cannot always track the bottom when moving between locations because occasional animals (likely vagrants) have been recorded moving across ocean trenches that are much deeper than their diving capability (Hobbs et al. 2007; Hill-Lewenilovo et al. 2018). Sheppard et al. (2006) also speculated that deep diving during large-scale movements (Fig. 3.1b) might be a tactic to avoid predatory large sharks. There is limited evidence to support this suggestion. Studies in Shark Bay, Western Australia (Heithaus et al. 2002) demonstrate that tiger sharks (Galeocerdo cuvier) mainly use shallow inshore seagrass habitats, where dugongs choose microhabitats that allow them to escape from, rather than avoid predators (Wirsing et al. 2007a, b). TDR records demonstrate that dugongs dive throughout the diel cycle (Chilvers et al. 2004). Nonetheless, the proportion of time spent close to the surface varies consistently with time of day in the three locations studied by Hagihara et al. (2018), with the proportion of time spent close to the surface of the water being least in daylight hours (0800–1600), similar to Florida manatees (Edwards et al. 2016). Chilvers et al. (2004) attempted to infer dugong behavior from the shapes of two-dimensional dive profiles provided by TDRs. Hagihara (2015) pointed out the problems with this approach and showed that it is not possible to identify the phases of a dive or infer the behavior of shallow diving animals such as dugongs from dive profiles, especially using dive software developed for deeper diving animals. It can even be difficult to define what constitutes a dive in shallow water. Consequently, Hagihara et al. (2011) used a modeling approach to maximize the likelihood of correctly identifying a dive and to minimize errors associated with depth records, the resolution of the TDR (which has improved over time), wave action, and the orientation of the dugong’s body. The last is an important source of error for a

3 Diving and Foraging Behaviors

77

tailstock mounted TDR on an adult dugong 2.4–3 m long in water only a few meters deep. The variable ways that dugongs surface and submerge means their tailstock does not necessarily reach the water surface in between each dive cycle (Anderson and Birtles 1978). Hagihara (2015) used eight dive metrics measured from TDR profiles, logistic regression models, and information on bathymetry, tide, and the inferred distribution of seagrass to differentiate probabilistically between: (a) midwater dives that did not provide the telemetered dugong with access to the sea floor, and (b) dives that enabled the animal to access the seafloor in the presence and absence of seagrass. The models showed that dives classified as mid-water dives had shorter bottom times, a greater degree of vertical displacement (presumably a result of active tail movements during the bottom phase), and slower ascent rates than seafloor dives. In locations supporting seagrass communities, dugongs transited quickly between the surface and the seafloor, presumably to maximize feeding time. Dugongs undertaking seafloor dives in locations that were assumed to be without seagrass transited more slowly between the surface and the bottom but could spend considerable time on the bottom, presumably resting. Dugongs in Moreton Bay rested at the bottom, mid-water, and on the surface (Hodgson 2004). Surface resting was most commonly observed at the reef lagoon edge in New Caledonia (Hagihara et al. 2018) and in Shark Bay (Anderson 1982). Hodgson (2004) noted that overall, dugongs spend a considerable portion of their daily time budget (18%) rising to the surface and submerging to resume their previous activity.

3.3 Foraging 3.3.1 Methodology Several techniques have been used to study the foraging behavior and ecology of sirenians. Field observations, fine-scale tracking data with benthic habitat sampling, tissue stable carbon and nitrogen isotope ratios, and plant fragment analysis of stomach contents, have all been used to document the foraging strategies and diet of West Indian manatees (Reynolds 1981; Ledder 1986; Hurst and Beck 1988; Ames et al. 1996; Reich and Worthy 2006; Slone et al. 2013). For Amazonian manatees, observational studies of feeding are extremely difficult due to their large and remote range and the dark, tannic waters that dominate their habitats. Therefore, most studies on their diet have focused on microhistological analysis of fecal material found in manatee habitats, and samples from the digestive tracts of dead manatees (Colares and Colares 2002; Guterres-Pazin et al. 2014). The African manatee’s diet has been studied in the field by documenting signs of feeding (torn leaves and stems, plants uprooted) on mangroves, shoreline grass species, and reedy plants such as Typha australis and Phragmites species (Powell 1996; Keith Diagne 2014). In the laboratory, methodologies for determining their diet have included stable isotope techniques

78

L. W. Keith-Diagne et al.

and dissection of stomach contents and fecal samples (Powell 1996; Keith Diagne 2014; Takoukam 2019). Field studies of dugong foraging behavior have mostly been undertaken as part of general behavioral studies, for example, visual observations (Anderson and Birtles 1978; Anderson 1998) and Hodgson’s (2004) research using a ‘blimp-cam’. Most information on the dugong’s diet has come from the analysis of gut contents using either direct observation (e.g., Johnstone and Hudson 1981; Heinsohn and Birch 1972; Marsh et al. 1982; Preen 1995; see Marsh et al. 2011 for a complete list) or Near Infra-red Reflectance Spectroscopy (André and Lawler 2003). Chilvers et al. (2004) attempted to identify feeding behavior from TDR dive profiles, but Hagihara (2015) pointed out the serious limitations of this approach for a shallow diving animal such as a dugong, as discussed above. Recently, Rasheed and his colleagues have developed techniques for using low-level aerial photography, next-generation photogrammetry, and machine learning to study dugong feeding trails in intertidal seagrass habitats at meadow scales (Rasheed et al. 2017).

3.3.2 Feeding Modes Sirenians use two different feeding modes: excavating and cropping (sensu Wirsing et al. 2007c), but the relative importance of the two feeding modes differs in the two genera. Manatees mainly use cropping (Fig. 3.2a–c) because they feed on emergent and floating plants as well as benthic flora; as obligate bottom feeders, dugongs mainly excavate (Fig. 3.2d). Manatees and dugongs crop leaves when plant roots and rhizomes are not accessible because of the structure of the plant or the hardness of the sediment. Both manatees and dugongs excavate whole plants with accessible roots and rhizomes, presumably because of the nutritional advantages in eating both the above- and below-ground components (see Marsh et al. 2011 for a comprehensive discussion). When feeding on seagrasses growing in compacted sediments, or on large seagrasses such as Amphibolus antarctica or Enhalus acoroides, dugongs feed by cropping seagrass leaves (Anderson 1982; Nakanishi et al. 2008). Stomach contents analysis suggests that dugongs do not eat the rhizomes of Thalassia hemprichii and E. acoroides (Erftemeijer and Djunarlin 1993; André et al. 2005; Domning and Beatty 2007), presumably because they are either too fibrous to process or extend too deep in the sediment (e.g., 6–12 cm) for a feeding dugong to disturb (Domning and Beatty 2007). Wirsing et al. (2007c) considered that excavation is a risky foraging tactic because the associated clouds of sediment not only advertise the presence of dugongs to a predator but also reduce a dugong’s capacity to be vigilant to the approach of predators (Fig. 3.3c), an assumption that may be unjustified if dugongs rely on sensory modalities other than vision (see Chap. 2). They studied dugong feeding behavior in Shark Bay, where they assumed with limited supporting evidence that dugongs mostly crop on A. antarctica, except during the summer months when the low biomass tropical species are more available. During summer, the time dugongs

3 Diving and Foraging Behaviors

79

Fig. 3.2 The varied feeding behavior of modern sirenians: a Florida manatees feeding on the emergent grass Spartina (USGS photo); b an Amazonian calf eats water hyacinth in a semi-captive enclosure at the community-based Rehabilitation Center at the Amanã Sustainable Development Reserve, Brazil (Hilda Chávez-Perez photo); c an African manatee in Angola flees from a shoreline where it had been feeding on water hyacinth (Miguel Morais photo); d a dugong bottom feeding on seagrass in Egypt (Ahmed Shawky photo). Reproduced with permission

allocated to excavation was inversely related to the abundance of tiger sharks, rather than the availability of these seagrass species, a behavioral tactic that Wirsing et al. (2007a, b) interpreted as a predator avoidance strategy. Excavating dugongs (and manatees) leave characteristic feeding scars in the sediment (Fig. 3.3a); occasionally flipper marks are also visible when animals are ‘pec walking’ (Fig. 3.3b, see also Chap. 6). They generate feeding plumes in the water column (Fig. 3.3c). These signs have been used to study dugong feeding behavior in intertidal areas at the meadow scale (e.g., Rasheed et al. 2017).

3.3.3 Diet All extant sirenians eat diverse diets and it is likely that none eats only plants, although that has not been confirmed for the Amazonian manatee. West Indian manatees consume a wide variety of forage items including submerged, floating, and emergent

80

L. W. Keith-Diagne et al.

Fig. 3.3 Signs of excavating dugongs: a feeding trails at Yule Point, Queensland, Australia (Len McKenzie photo); b flipper prints in a sparse seagrass meadow at Urangan Queensland, Australia (Len McKenzie photo) c clouds of sediment produced by a feeding dugong in New Caledonia (Christophe Cleguer). Reproduced with permission

plants, and algae in marine, estuarine and freshwater habitats (Best 1981; Reynolds 1981; Quintana-Rizzo and Reynolds 2007). The plant species consumed vary by region, habitat type, and availability. West Indian manatees eat more than 60 species of freshwater, marine, and terrestrial food plants (Hartman 1979; Reynolds and Odell 1991). They also occasionally consume fish (Powell 1976; Caicedo-Herrera et al. 2020), clams (Magor 1978), and tunicates (O’Shea et al. 1991; Courbis and Worthy 2003). Amazonian manatees have been documented feeding on 49 species of plants, but likely feed on many more (Colares and Colares 2002; Guterres-Pazin et al. 2014). African manatees have been documented eating leaves, stems, roots, seeds, and fruits from over 90 species of plants (Villiers and Bessac 1948; Husar 1978; Powell 1996; Reeves et al. 1988; Akoi 2004; Dodman et al. 2008; Keith Diagne 2014; Mayaka et al. 2019; Takoukam 2019). This high diversity of plant species is likely related to the varied habitats of African manatees throughout their large range, which includes rivers and lakes in tropical rainforests, savannahs, and the Sahel, as well as mangrove channels, coastal estuaries, and lagoons. They also consume fish and mollusks (Powell 1996; Reeves et al. 1988; Dodman et al. 2008; Keith Diagne 2014). Dugongs consume at least 17 genera of seagrasses, marine algae, and invertebrates (Marsh et al. 2011). Figure 3.4 illustrates the different genera of food resources consumed by each species.

3 Diving and Foraging Behaviors

81

Fig. 3.4 Food resources consumed by sirenians by number of genera. ‘Observed’ signifies genera detected by visual observation or by using stable isotopes, stomach or fecal analyses; ‘inferred’ signifies genera identified via other means such as interviews with local people. FW = freshwater. Graph drawn by Adella Edwards from data synthesized by Lucy Keith Diagne. Reproduced with permission

82

L. W. Keith-Diagne et al.

In a review of West Indian manatees across their range, Quintana-Rizzo and Reynolds (2007) noted that Florida manatees regularly eat the submerged species Ceratophyllum sp., Vallisneria sp. Hydrilla verticillate, Najas sp., Potamogeton sp., Egeria densa, and the floating and emergent species Eichhornia crassipes, Pistia stratiotes, Sagitaria sp., Panicum sp., and Phragmites (Quintana-Rizzo and Reynolds 2007). Many of these genera are also consumed by African manatees. Occasionally, Florida manatees have been recorded eating atypical plants. O’Shea (1986) made detailed observations of wild manatees foraging on highly nutritious live oak (Quercus virginiana) acorns at Blue Spring, Volusia County, Florida. Frequently eaten marine and estuarine forage plants include Thalassia testudinum, Syringodium filiforme, Halodule wrightii, three species of Halophila (H. engelmanni, H. decipiens, H. johnsonii), various marine algae, and the emergent plants Rhizophora mangle, Spartina alterniflora, and Distichlis spicata (Best 1981; Quintana-Rizzo and Reynolds 2007). In a study in southeast Florida, Lefebvre et al. (1999) found more instances of manatees feeding on H. wrightii than S. filiforme and suggested that manatees may prefer H. wrightii, even though the short-shoot biomass of S. filiforme during the study was at least twice that of H. wrightii in all seasons (Kenworthy 1992). Seagrasses in this study were restricted to depths 5 m depth, and most dugong sightings are in waters 5–20 m deep (Hagihara et al. 2018). Satellite tracking studies have contributed substantially to our knowledge of dugong movement behavior, as detailed in Supplementary Material Table 5. In the years preceding and including 2018, 215 individuals were tracked, mostly along the Australian coast, except for a few studies carried out in East Indonesia, New Caledonia, and the Red Sea.

5.7.1 Do Dugongs Migrate? The dugong is listed as migratory under Appendix II of the Convention on the Conservation of Migratory Species of Wild Animals, which is intended to protect species that cross international boundaries (CMS 2020), as dugongs do. Although capable of long-distance movements, regular round-trip migrations of entire regional dugong populations, as observed in manatees, are uncommon. Broad-scale aerial surveys conducted across different seasons and latitudes suggest that dugong distribution is relatively consistent over the annual cycle (reviewed in Marsh et al. 2011), although partial migration and multidirectional movements of varying distances (Sheppard et al. 2006) may occur in some places and at some times, particularly after seagrass loss. Such year-round residency has been documented or inferred in the following regions through aerial surveys (and sometimes limited satellite tracking) at more than one time of year or, rarely, year-round (e.g., Moreton Bay): New Caledonia (Cleguer et al. 2015, 2017); the Northern Territory of Australia (Bayliss and Freeland 1989; Whiting 2008; Cardno 2014; Udywer et al. 2019); and Queensland, Australia, including Torres Strait (e.g., Marsh et al. 2015), the northern GBR (e.g., Marsh and Saalfeld 1989; Marsh et al. 2020a), the southern GBR and Cleveland Bay (e.g., Marsh et al. 2020b), and Moreton Bay (Preen 1992; Lanyon 2003; Zeh et al. 2018). Although a paucity of data limits the strength of inferences about seasonal movement patterns in much of the dugong’s vast range, year-round residency is likely to be the common pattern in most tropical regions (Table 5.1). As noted in Box 5.1, and discussed further in Sect. 5.7.2, periodic seagrass loss caused by extreme weather events plays a more important role in driving dugong large-scale movements than do seasonal influences in most tropical parts of their range. In some locations, dugongs adapt to seasonal changes in their environment by making local adjustments in their space use rather than undertaking long-distance migrations. For example, Moreton Bay on Australia’s southeast coast, at the high latitudinal limit of the dugong’s range, supports hundreds of dugongs throughout the

196

C. J. Deutsch et al.

year (Lanyon 2003; Lanyon et al. 2019), but genetic data suggest that the residency of individuals in this region may be variable (Cope et al. 2015). Tracking studies have shown that dugongs modify their movements and habitat use of the bay during winter in response to seasonal changes in water temperature, alternately foraging on seagrass flats in cooler bay waters and moving offshore to thermoregulate in warmer oceanic waters immediately outside the bay (Preen 1992; Zeh et al. 2018). Similar mesoscale temperature-induced movements between disparate habitats were observed in Hervey Bay, north of Moreton Bay, where six animals tracked in winter made 80 km trips lasting 2–6 days across the bay to warmer oceanic waters off Sandy Cape, despite the apparent lack of suitable seagrass meadows in that area (Sheppard et al. 2006). Some of them made the return trip to inshore foraging areas during the winter. Chap. 6 describes this scale of movements in more detail. Shark Bay, Australia, and the Arabian/Persian Gulf encompass important dugong habitats at similar latitudes south and north of the equator, respectively, representing the approximate latitudinal limits of the species’ range. They are the only places where seasonal migration of dugong populations has been documented or seems likely (see below) (Table 5.1). Anecdotal sightings of dugongs well beyond the southern limit of the species’ accepted range on the east coast of Australia (New South Wales) during the austral summer suggest that seasonal long-distance movements occur there too, but as in many other areas of the dugong’s range, it has not yet been determined whether these sightings represent exploratory ranging movements in the warmer months or seasonal migrations (Allen et al. 2004).

5.7.1.1

Shark Bay, Australia

Shark Bay is a vast (~23,000 km2 ) water body in Western Australia that provides vital habitat for thousands of dugongs. Aerial survey and satellite tracking studies demonstrate that seasonal distribution patterns of dugongs within Shark Bay reflect changes in water temperature; the animals seek out offshore waters warmed by the Indian Ocean when winter water temperatures in the bay fall below 18 °C (Holley et al. 2006), the dugongs’ apparent lower thermal threshold (Sheppard et al. 2006; Marsh et al. 2011). Holley (2006) equipped 18 dugongs with satellite-linked tags, four of which were tracked for sufficiently long periods (4.9–8.7 months) to document large-scale seasonal movements. Although these individuals remained within Shark Bay, they moved seasonally between distinct core use areas separated by 120–150 km for three males and by 35 km for one female. Movements between summer and winter areas of high use were associated with changes in sea surface temperature, with the males’ winter home ranges located closer to warmer oceanic waters in the western gulf of Shark Bay; at the onset of summer, these males returned to the eastern gulf of the bay (Fig. 5.7a). Significant shifts in seasonal distribution of dugongs were also inferred from sightings data by comparing a summer aerial survey (Holley et al. 2006) with previous boat-based observations and aerial surveys conducted in winter (Prince et al. 1981; Anderson 1982, 1986; Marsh et al. 1994; Preen et al. 1997). Compared to the winter season when dugongs were concentrated more in the deeper,

5 Movement Behavior of Manatees and Dugongs …

197

Fig. 5.7 Large-scale dugong movements in Australia. Depiction of some of the longest movements (blue lines) of individual dugongs documented using satellite-linked tracking devices. Black circles indicate locations of tagging. Single-arrowed lines show movement paths without evidence of return during the tracking period, whereas double-arrowed lines show round-trip movements. Insets (a), (b), and (c) provide close-ups of dugong movements in Shark Bay, Torres Strait, and the Hervey Bay-Moreton Bay region, respectively. ‘W’ and ‘S’ denote identified winter and summer grounds, respectively, for dugongs in Shark Bay. Dashed lines delineate the southernmost year-round limits of the dugong range on the west and east coasts of Australia. P.C. Bay = Princess Charlotte Bay

western part of Shark Bay, the summer distribution appeared to shift to the eastern and southwestern portions of the bay. Change in water temperature is thought to be the main driver of dugong seasonal migrations within Shark Bay, but apparently, not all dugongs respond to temperature changes in the same way. This is inferred from the finding that 4–14% of dugongs seen during aerial surveys of Shark Bay were in waters less than 18 °C (Marsh et al. 1994; Preen et al. 1997) and by observations of dugongs here in waters less than 15 °C (Wirsing et al. 2007).

5.7.1.2

Arabian/Persian Gulf

Temperature-induced seasonal migration of dugongs in the Arabian/Persian Gulf seems likely based on the available evidence. The summer distribution in this region has been well-studied (Preen 1989, 2004) but the winter distribution is less clear and raises the possibility of long-distance seasonal movements within the Gulf. The only large-scale aerial survey of the region conducted during winter covered just onethird of the summer survey extent, so an adequate regional-scale seasonal comparison is not possible. Nevertheless, the winter survey showed a striking difference in dugong distribution in the waters of Saudi Arabia, Bahrain, and Qatar compared to the summer surveys. Almost all recorded dugongs occurred in a single massive herd,

198

C. J. Deutsch et al.

composed of two groups (577 and 97 animals), in Gulf waters between Qatar and Bahrain (Preen 1989, 2004). Preen (2004) suggested that dugongs undertake seasonal movements within the survey area and perhaps migrate up to 400 km between summer and winter use areas within the southern Gulf region. Large winter aggregations of dugongs have since been found at nearly the same location during boat-based surveys (Marshall et al. 2018). The studies suggest that behavioral thermoregulation is the most likely reason for the occurrence of these aggregations around benthic thermal springs (Preen 2004) because mean sea surface temperature was 18 °C in January (Marshall et al. 2018). Researchers could not ascertain the origin of the dugongs forming the herds or how far they had traveled.

5.7.2 Movement Response of Dugongs to Periodic Declines in Forage Some of the most intense episodic pressures on seagrass habitats are caused by storms, cyclones, floods, and heat waves (Babcock et al. 2019). These events can decimate extensive areas of seagrass and, therefore, impact the health and survival of dugongs (Preen and Marsh 1995; Wooldridge 2017; Marsh et al. 2018). Behavioral responses of dugongs to these changes have been inferred from large-scale population shifts along the coast. For example, analysis of a time series of dugong aerial surveys conducted in Western Australia detected a 40% increase in the abundance of dugongs in Shark Bay (from ~10,000 to ~14,000), and at least an 80% decline in abundance of smaller populations to the north at Ningaloo Reef and Exmouth Gulf, over a 5-year period (Gales et al. 2004). Gales et al. (2004) hypothesized that this shift in abundance was partly a result of dugongs undertaking long-distance movements (~300 to 650 km) from the Ningaloo-Exmouth regions south into Shark Bay after Tropical Cyclone Vance had adversely impacted seagrass habitats in the northern regions, although the evidence remains circumstantial. A similar movement response occurred on the other side of the continent, in Hervey Bay, Queensland, after more than 1,000 km2 of seagrass were lost following two floods and a cyclone in 1992 (Preen et al. 1995). An estimated 73% decline in the regional dugong population ensued (Preen and Marsh 1995). While many dugongs died of starvation 6–8 months after the flooding event, others survived by either: (1) remaining in the bay and consuming any remnant seagrass and other low-quality food such as macroalgae or invertebrates; or (2) emigrating from the affected area, apparently swimming up to hundreds of kilometers to reach other foraging grounds. This population exodus was almost certainly induced by the extensive loss of foraging habitat, and it likely accounts for the unprecedented number of live-stranded and dead dugongs reported during the following year in New South Wales, up to 900 km south of Hervey Bay and well beyond the southern limit of their accepted range (Preen and Marsh 1995; Allen et al. 2004).

5 Movement Behavior of Manatees and Dugongs …

199

Severe weather events also occurred along the urban coast of Queensland in the summer of 2010–11. These included the strongest La Niña event since 1973, three tropical cyclones, and ensuing major floods, all negatively impacting important dugong habitats. A large-scale aerial survey conducted after these events detected a substantial decline in the abundance of dugongs in the southern GBR waters (Sobtzick et al. 2012). A possible explanation for this decline is that dugongs had moved large distances to other, less affected coastal areas not covered by the aerial survey, or to offshore areas of deep-water seagrass which recover more quickly from perturbations (Rasheed et al. 2014). Further south, the size and density of the dugong populations of Hervey Bay and Moreton Bay were largely unaffected by these weather events, although the proportion of dugong calves detected during aerial surveys was negatively associated with various features of La Niña episodes (Fuentes et al. 2016). Dugongs do not always respond to catastrophic habitat loss with long-distance movements. A severe marine heat wave along the Western Australian coast in the late summer of 2010–11 caused significant and long-lasting damage to marine communities and habitats, including the large-scale loss of seagrass, mainly the temperate species Amphibolis antarctica, in Shark Bay (Kendrick et al. 2019; Strydom et al. 2020). Aerial surveys across the region after the heat wave found no evidence for major redistribution of dugongs from Shark Bay to northern regions with intact seagrass habitats (Bayliss et al. 2018). This was likely because the region of seagrass loss was quickly colonized by small, fast-growing tropical seagrass species (such as Halodule uninervis) readily eaten by dugongs (Kendrick et al. 2019). Although dugong abundance appears to have increased from 2007 to 2018 in NingalooExmouth Gulf, corresponding numbers in Shark Bay appear to have also increased slightly or at least remained stable. This suggests that there was no large northward emigration of dugongs due to the seagrass dieback event (Bayliss et al. 2018), probably because the loss was not an important dugong food source. The connectivity and extent of movements among the dugong populations along the Western Australian coast require further investigation.

5.7.3 Individual Large-Scale Movements in Dugongs All studies on the use of space by satellite-tagged dugongs show considerable variation in movement patterns among individuals (Box 5.2). While some dugongs remain relatively sedentary during their tracking period, others tracked during the same period show highly mobile behavior, swimming relatively large distances in a few days. In this section, we highlight long-distance movements of individuals tracked with satellite-linked tags. Marsh and Rathbun (1990), who pioneered the use of PTTs on dugongs, found that one of the six males they had tagged in north Queensland made three trips between Cleveland and Upstart Bays, a one-way linear distance of 143 km in as little as two days, whereas the other five animals remained relatively sedentary (remaining within 22 km of capture site). Sheppard et al. (2006) found that of the 70 dugongs they tracked from different locations in Australia (64

200

C. J. Deutsch et al.

in Queensland and six in the Northern Territory), 44 made large-scale movements (defined as trips >15 km); the maximum distance moved during a trip was 560 km by a female traveling from Hinchinbrook Island to Princess Charlotte Bay (Fig. 5.7). The maximum distance a tagged dugong moved from its capture site was 625 km (Sheppard et al. 2006), and the movements of one female were reported to have spanned 860 km of coastline over a 7-month tracking period (Preen 2001). Other examples of large-scale movements by dugongs on the east coast of Queensland include four dugongs swimming distances of 278 to 338 km from Moreton Bay to Hervey Bay over 5–9 days (Zeh et al. 2016; Fig. 5.7c), and a dugong that traveled 513 km from Hervey Bay to Shoalwater Bay (Sheppard et al. 2006) (Fig. 5.7). In Torres Strait, dugongs captured from different locations in the central part of the strait made repeated return movements across the strait to Boigu island, close to the Papua New Guinea coast, covering more than 100 km on each trip (Gredzens et al. 2014; Cleguer et al. 2016) (Fig. 5.7b). Of the five dugongs tracked by Bayliss and Hutton (2017) in the Kimberley region, Western Australia, one adult female traveled 325 km over 14 days (averaging 23.2 km per day). Similarly, a young adult male moved a cumulative distance of 1,160 km over 78 days (averaging 14.9 km per day) yet was only 85 km from its capture point when its tag detached (Bayliss and Hutton 2017). Three of the six dugongs equipped with satellite tags by Campbell et al. (2010) in the West Kimberley also undertook large-scale movements, with the highest cumulative unidirectional distance exceeding 400 km over a 6-week period (Fig. 5.7). Dugongs making moves of > 15 km (up to 67 km) were also reported in habitats that are much more spatially constricted, such as small tropical lagoonal systems (de Iongh et al. 1998; Cleguer et al. 2020). Dugongs have been known to undertake long-distance movements across deep ocean trenches. For example, dugongs have been observed at the Aldabra Atoll, some 400 km north of Madagascar and over 600 km from the African continent (Van de Crommenacker 2013). It is unknown whether the dugongs at Aldabra (estimated at a minimum of only 11–14 animals) are resident or transient, perhaps visiting this island as a stepping stone in their movements across the Western Indian Ocean. A subadult female dugong was found stranded on Fiji in the Pacific, more than 600 km from the nearest known dugong population and beyond the eastern edge of the species’ range (Hill-Lewenilovo et al. 2019). These and other reports of oceanic, extralimital longdistance movements of vagrant dugongs (e.g., Whiting et al. 2005; Hobbs et al. 2007) may illustrate how the species has likely expanded its range through evolutionary time.

5.8 Long-Term Range Fidelity Next, we turn our attention to patterns of movement over long temporal scales, reflecting interannual fidelity (or lack thereof) of individuals to seasonal or yearround ranges. That is, do they return to the same sites and areas year after year, or do their ranges shift in response to environmental pressures or for other reasons? This

5 Movement Behavior of Manatees and Dugongs …

201

is a particularly challenging question because it is difficult to track sirenians for long periods of time (Reid et al. 1995) and photo-identification is not a viable option for most populations; genetic approaches can sometimes inform this issue, as discussed above (see Chap. 4). The long-term deployment of tracking devices on dugongs, for example, has been hampered by logistical constraints, including unreliability of tag attachment, tag battery life, and difficulty in recapturing to replace the tag. As a result, most dugong tracking studies have been limited to a particular season within a year (e.g., Gredzens et al. 2014; Cleguer et al. 2016, 2020; Udyawer et al. 2019). We end this section with a review of evidence for natal philopatry in sirenians, the tendency of individuals to remain in or return to the ranges that they used as dependent calves (Waser and Jones 1983); this topic is also considered in Chapter 4.

5.8.1 Interannual Fidelity to Seasonal Ranges and Refugia 5.8.1.1

Florida Manatees

Manatees in Florida experience two main seasons that drive large-scale movements and habitat use patterns: the cold winter season (December to February) and the warm season (April to October), separated by variably cool transitional months (November and March). As noted in the above section, Florida manatees typically undertake rapid directed movements along narrow migratory corridors between disjunct winter and warm-season ranges. We use the term interannual site fidelity to describe the movement pattern of an individual returning to a specific site or seasonal range from one year to the next. Everglades National Park biologist Joe Moore was the first to scientifically document site fidelity in manatees by sketching scar patterns of individuals that used a warm-water power plant discharge in the Miami River during winter (Moore 1956). Of ten animals with distinct marks, Moore identified seven of them in more than one winter, including several 5 years apart. Since the late 1960s, researchers have photo-identified thousands of manatees at winter aggregation sites around Florida (Hartman 1979; Beck and Reid 1995), with individuals resighted many times over multiple years, even several decades. Although there is some movement among sites within regions, the overriding pattern is one of strong, long-term fidelity to specific warm-water sites or groups of sites within an area (e.g., Reid et al. 1991; Deutsch 2000). At the artesian spring systems of Crystal and Homosassa Rivers, for example, the year-to-year return rate was calculated to be about 90% (Powell and Rathbun 1984; Rathbun et al. 1990, 1995). Winter fidelity to the crystal-clear waters of Blue Spring was similarly high (O’Shea and Hartley 1995), with annual resighting probability of adults estimated to be 0.95 (Langtimm et al. 1998). These spring systems comprise the principal warm-water habitats within their regions and so high fidelity to these sites is both adaptive and expected. In contrast, there are several major warm-water sites along the extensive coast of the Atlantic region, and manatee use patterns are more complex and dynamic.

202

C. J. Deutsch et al.

Summarizing movement patterns derived from resightings of distinct individuals, Reid et al. (1991) documented movements among winter aggregation sites on the east coast within and across years. They concluded that many of the apparent changes in warm-water sites were likely due to stopovers during migration. A telemetry study of 71 individuals over 12 winter seasons along the Atlantic coast helped to elucidate important aspects of these warm-water habitat use patterns (Deutsch 2000). Tagged manatees used a network of warm-water refuges (16 in this study) that varied among individuals from 500 km apart. They visited an average of 4.3 (min–max = 2–9) known thermal sites over their tracking periods—some sites just during migration—and every manatee used at least one industrial effluent every winter (Deutsch 2000) (Fig. 5.8d). Individuals demonstrated high site fidelity to one or two warm-water refuge areas (areas may include more than one aggregation site) across multiple years, consistently returning to the same southernmost destination(s) in each winter (Fig. 5.8). Despite this tenacity to traditionally used sites, some manatees also

Fig. 5.8 Interannual fidelity of tagged adult Florida manatees to warm-water aggregation sites in winter along the Atlantic coast, 1987–1995. (a) Example of a female showing high fidelity to one primary refuge area in southeast Florida over seven winters. (b) Female showing high fidelity to two primary refuge areas in southeast Florida over six winters. (c) Female showing substantial variability from year to year in overwintering locations across central-east to southeast Florida. (d) Aerial photo of a manatee aggregation in the FPL Riviera Beach power plant discharge canal (RB) (photograph by John E. Reynolds III, courtesy of Mote Marine Laboratory and Florida Power and Light Co.). The relative area of each circle represents the percentage of days at that refuge out of the total number of days located at warm-water sites during that winter (n, given below the year); a large, single circle equals 100% and the smallest circles are < 5%. A cross denotes presence at a refuge in years without sufficient data for analysis. Distances between sites on the y-axis are not to scale, but the total distance from Brunswick, Georgia to Biscayne Bay, Florida is 680 km. Reprinted from Deutsch (2000), with permission

5 Movement Behavior of Manatees and Dugongs …

203

displayed flexibility in the amount of time spent at different sites from one winter to the next (Fig. 5.8c). This points to inherent behavioral plasticity that can allow individuals to adapt to changing environmental conditions, including power plant operational changes, within certain constraints (Deutsch 2000). Less information on site fidelity and movements has been derived from photoidentification research during the warm season because manatees are dispersed during this time of year and so most field effort has been conducted during the winter months. Interannual fidelity to warm-season ranges has been reported, however, for photoidentified manatees in Sarasota Bay (Koelsch 1997) and the Banana River (Shane 1983; Reid et al. 1991). Long-term tracking of manatees with PTTs on the US Atlantic coast led to the first quantification of interannual fidelity to seasonal ranges in sirenians (Deutsch et al. 2003). Individuals showed remarkable consistency in their migratory patterns across years: 92% of year-to-year comparisons showed the same seasonal movement pattern and 82% of individuals showed the same seasonal movement pattern across all years of tracking (Fig. 5.9a). This consistency was due to the strong philopatry that individuals showed to their seasonal ranges from one year to the next. In both winter and warm seasons, the center of an individual’s core use area(s) within a region remained within 5 km (median) of that from the previous year and was under 10 km for three-quarters of the manatees (Deutsch et al. 2003). Perseverance and success in retagging efforts by field researchers were key in demonstrating that the observed tenacity to seasonal core areas was maintained over multiple years, including ten individuals tracked for 4–9 warm seasons and six adults tracked for 6–8 winters (Deutsch et al. 2003). Although manatees typically returned to the same winter ranges year after year, the amount of time spent in different core areas often varied across years and, in some cases, manatees shifted their overwintering region (Figs. 5.8, and 5.9a). The higher interannual variability in range use during the winter compared to the warm season was most likely due to year-to-year variation in winter severity and power plant operations.

5.8.1.2

Antillean and Amazonian Manatees

Few West Indian manatee individuals have been radio-tracked for multiple years outside of Florida, so there is relatively little information on interannual fidelity to seasonal ranges or sites. But the available long-term data suggest that most individuals, especially females, remain faithful to small core areas. This strong degree of site fidelity, for example, was evident in manatees tracked in Guerrero Lagoon (Chetumal Bay, Mexico) and in Southern, Northern, and Placencia Lagoons (Belize) (MoralesVela and Medrano-González 1999; Auil et al. 2007). Likewise, most of the manatees tracked for at least one year in Puerto Rico occupied the same areas year-round and displayed strong site fidelity to particular freshwater sources that were close to the capture site (Slone and Reid IN PRESS). In addition to the tracking data showing that manatees in Puerto Rico are generally quite sedentary, strong differences in

204

C. J. Deutsch et al.

Fig. 5.9 Long-term site fidelity of West Indian manatees to seasonal movement patterns and ranges as shown by plots of north–south movements versus date for individuals with multiple years of tracking. (a) Florida manatees tracked along the Atlantic coast of the southeastern United States (see Fig. 5.3 for geographic reference), mostly with PTTs; dashed vertical lines denote the start of winter (Dec 1). Listed are the manatee ID, sex, tracking duration of total tracking period, total range over the tracking period, and mean annual range. The bottom plot shows an individual that shifted its overwintering region to the north by ~ 300 km in the second year of tracking. Modified from Deutsch et al. (2003); permission not required.

5 Movement Behavior of Manatees and Dugongs …

205

b

Fig. 5.9 (b) The longest continuous tracks of any sirenian are of rehabilitated orphaned Antillean manatees after their release on the northeast coast of Brazil with VHF tags. Listed are the name, sex, tracking duration, number of areas with residence (i.e., ≥ 10% of locations in 0.1° bin), percentage of locations within those areas, total range over the tracking period, and maximum annual range. Note that in some years, the annual ranges for “Lua” and “Astro” were quite small (