HARVARD UNIVERSITY e Library of the Museum of Comparative Zoology i (aes iy ECOLOGY AND BEHAVIOR OF THE MANATEE (TRICHECHUS MANATUS) IN FLORIDA SPECIAL PUBLICATIONS This series, published by the American Society of Mammalo- gists, has been established for papers of monographic scope con- cerned with some aspect of the biology of mammals. Correspondence concerning manuscripts to be submitted for publication in the series should be addressed to the Editor for Special Publications, Hugh H. Genoways, Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsyl- vania 15213. Copies of special publications may be ordered from the Secre- tary-Treasurer of the Society, Duane~A- Schlitter, Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213. Price of this issue $10.00 COMMITTEE ON SPECIAL PUBLICATIONS JAMES N. Layne, Editor Archbold Biological Station, Route 2, Box 180, Lake Placid, Florida 33852. J. KNox JONES, JR., Managing Editor The Museum, Texas Tech University Lubbock, Texas 79409 CONSULTING EDITORS FOR THIS ISSUE Kar_ W. KENYON JOSEPH CurTIS MOORE ECOLOGY AND BEHAVIOR OF THE MANATEE (Trichechus manatus) IN FLORIDA By DANIEL S. HARTMAN DEPARTMENT OF CONSERVATION CORNELL UNIVERSITY IrHAcA, NEw York 14853 (Present address: RFD 1, Bethel, Maine 04217) SPECIAL PUBLICATION NO. 5 THE AMERICAN SOCIETY OF MAMMALOGISTS PUBLISHED JUNE 27, 1979 ili MUS. COMP. ZOOL LIBRARY act 1 980 HARVARD UNIVERSITY Library of Congress Catalog Card No. 79-52633 . © 1979 by The American Society of Mammalogists iv FOREWORD ANATEES are members of an obscure order of aquatic mam- mals known as the Sirenia and are believed to be descendants of the same ancestor from which the elephants evolved. Living representatives of the Sirenia include three species of manatees (West Indian, Trichechus manatus; Amazonian, T. inunguis; and West African, T. senegalensis) and their marine relative, the dugong (Dugong dugon). A fifth species, a toothless 25-foot kelp-feeder from the Bering Sea named Steller’s sea cow (Hydrodamalis gigas), was exterminated by fur sealing expeditions within a quarter cen- tury of its discovery in 1741. Present-day sirenians are tropical or subtropical in distribution. Esteemed for their succulent flesh, they are harpooned and netted wherever unprotected and, as a result, have been reduced to rare or endangered status throughout most of their range (Bertram and Bertram, 1973). Little is known of the ecology and behavior of sirenians. The paucity of information on their natural history can be attributed to their phlegmatic, retiring habits and to their inaccessibility in coastal shoals and turbid rivers. Most of the literature on the be- havior of manatees and dugongs is founded on studies of captive animals. Field reports are mostly speculative and fragmentary, being limited to observations from above the surface of the water. Manatees have been effectively used as agents of aquatic weed control (Allsopp, 1960, 1961, 1969; Anonymous, 1961, 1964, 1973, 1974; Sguros, 1966; MacLaren, 1967). They also have been sug- gested as a potential source of protein to help alleviate world food requirements (Anonymous, 1917; Pirie, 1967; C. Bertram and G. Bertram, 1968). Until more data are available on the reproductive physiology and population dynamics of manatees, however, it 1s premature to envision managing and utilizing them on a large scale for weed clearance or meat production. In the United States, sirenians are represented by a single species of manatee (Trichechus manatus Linnaeus) confined, with rare ex- ceptions, to peninsular Florida and the coast of Georgia. The be- havior of these manatees, apparently isolated in the North Tem- perate Zone, is distinguished by seasonal cold-induced congregations in warm-water refugia. Some of these refugia are Vv fed by limpid springs, the only places in the world, to my knowl- edge, where sirenians can be observed underwater with relative ease. It was this consideration that prompted my research. This study is a contribution of the New York Cooperative Wild- life Research Unit and is based on a doctoral dissertation submitted to Cornell University. I am grateful to my former advisor, James N. Layne, for the idea of the study and to the members of my graduate committee at Cornell, Daniel Q. Thompson, William C. Dilger, and David Pimentel, for their suggestions and criticisms. In addition, I would like to thank Joseph C. Moore for reviewing my original proposal; Daniel B. Ward, Dana Griffin, David Hall, Joanne Gaudsmith, Jack van Breedveld, J. Stephen Davis, and Harold Humm for identifying aquatic plants; Stephen G. Zam for analyzing a manatee fecal sample; Carter R. Gilbert for identifying a species of shark; and David K. Caldwell, Earl S. Herald, and Kenneth S. Norris for their counsel. Among the many Florida residents who contributed information or donated their time to work directly with me, I am particularly obligated to Harold Watson, Ed Collinsworth, Charlie Barnes, Bonnie Bonsall, Brownie Searle, Sharell and Richard Howze, and the late Gary Morrison. I wish especially to acknowledge the co- operation and friendship of Jim Macbeth, Margaret Cole, Violet Stewart, and Tom McQuarrie. I am also indebted to Robert M. Ingle who placed the services of the Florida Board of Conservation at my disposal and to Rudi Wolter who generously provided me with an outboard-powered runabout for the duration of my work. Buddy Powell who assisted me throughout the study deserves sep- arate thanks. Lastly, my greatest debt is to my wife, Maggie, whose continuous encouragement and good cheer made research bearable under the often merciless West Florida sun. This study was supported by the National Geographic Society; the U.S. Fish and Wildlife Service, Department of the Interior; the Theodore Roosevelt Memorial Fund of the American Museum of Natural History; and The Society of The Sigma Xi. Vi CONTENTS Mitr OCLUCEI © Tee es el ee ee ee See oe a ee 1 al itea ts aan LB g oe Brew ater Seis Fo te eS ee ] ERO CE CUTE See ee eee ness es Ate ae ee ee el MES 12 Characteristics of the Winter Populations in the Headwaters of the Crystaliand FlomosassaRIVErs 2228 22s ee es See ee ee 15 INEUETTT bb Cr Sieseereenes eae Sei hl) cea Re Ee eas le al Beek oe a 15 ING ca@OmPOSIGONEANG SEX] RaAtlO = pe ser ee eee ees Soke ee ee 16 Coldrinducedi@ongregations: eee == sss se asses eee eee ee ee ee ee 17 Headwaters of the Crystal and Homosassa Rivers-___----------------------- 17 BS Weg S PLING Sp RUT eee Seen a eee See Nes Ae Sea enews ees ees 25 INFO ETI E TES eae aera ee ne AIL ENS ME eNews eee ee ee 28 IN oat mS peresae ee ans eee seen aren Se ea Stra ee eee 28 PIR AVie IROU CE Spee nee nt aN Bert ee eel a ante ee ra ce Be es 32 RSX Ol aACO Iga A CUVICY = ee ree ee Peel ree See eee eae eae 35 abitatwOetenmimants: so sens ot. oes ee ee Wee oe ee 36 WO) ANT yy ge NC UNV Gy pee eet 2 teens eae ee Se pe Sh A AR on atest eee SE 41 \RGroyal 15 LORS ees ON Pee es SE, neigh Ae een eae eye 2 ee PEA ee 44 Kood=Rlantsiand Preferences 2222s. n se 44 ee Gin Cal COO Dye arora sree nes See nt eee cee eee 55 Interactions with Other Animals __________-__-_-____________--____ 59 Interest inanimate! @ObjectSna= 2s ee ee 60 AT ASHCE Spal Cle @ OMNES all Si see a eee 62 TTC O Daas ites pee see ae ete eee a ee ae ee es ee 62 External Associates seiet. wie ae UN bn i ae ce dee Se ae 62 Maintenance Behavign sass ae nee een eee eee 64 TEOCOIMNOLIO Tiere eee en een hee Sree ae Sie es PES pre 64 I AL INNING ere er Nee ae A i ON Ne Ne van A oe Boe Us |RSS a ea i a ea 82 HE CCI) p eaeretenee rs tee eee seam ae rile Oh oe ee aes ee SR 85 WombonteActivitiess.. st a 86 BliminativerDehavione ee ee ee 93 Social eB ehiayi@ Tee aa re les ee) Be eee ne ee, 95 WOGAli zation Simeermmiere cance Petcare ret Cok Lt ade ET ee 98 SexualWBehavio reeeeiewee seen sn oul Pe ea eee, ea 8 Oe a 100 la ypeeemene orl nie aes tere ane elon.) be Cea ESS 108 Motheravoungeb chavio tees a0.) = see eee Soe eee eee eee 110 Vii leaning 28205 2 2 At 2 ee 115 Sights ters Sule son eee SS oe wet ieee ees era nee ee eee 116 UNG UNG te cet ee ee ee ee Oe 118 MIRAS TEX oa ree SORE ek Re a De ee eee 118 Sin elllt = eee oi ye ot ie ee Sa ed i eel 119 Populations) main cswesss se. sae —ene meee eee ne ee 120 Birthi Rater. 2 een a een ee Si 120 AgeatiWieanin =e ise Ai 121 ING Chat SeXUall gs Mia GUN IG] iy ee ee ee ee 122 Ong évity 225-6. 2 = eS eee 122 Mortality Factors, = 22222222 <2 2 ee eee 123 Man-Manatee Relations-.2-—.. = ee eee 126 Responserofs Manttos Manatees eases = awe nnn ee 126 Response ofsManatees) toy Miami estes see ene see eee eee eee 127 ID ISGWSSIO We ee ee re ee 131 SS UIT 1 ay ee 138 leéiteraturet@itedt 22. 82 2 See ee ee eo ee 142 Dn GEG 4h ake as rs ae le oe Se es SS 2 151 viil INTRODUCTION HIS study was conducted primarily in Citrus County on the 1g central west coast of Florida. Research was focused on the headwaters of the Crystal and Homosassa rivers where clear springs of constant temperature afforded ideal conditions for above- and underwater observations of periodic cold-induced con- gregations of manatees during the winter months. The study was conducted from 1 October 1967 to 31 March 1969. Following completion of the study, I returned twice to Florida to engage in further manatee research. In the winter of 1970-71, I spent a month at Crystal River and a month at Blue Springs Park on the St. Johns River, Volusia County. From November 1972 to January 1974, I conducted a study of the manatee’s distribution and status in the United States, operating from headquarters at Crystal River. Many new discoveries relating to the ecology and behavior of manatees were made during these subsequent visits to Florida. These discoveries, most of which were made outside Citrus County, have been incorporated in the present paper. HABITATS ITRUS County lies within the terraced Coastal Lowlands sub- division of the Coastal Plain Province (Vernon, 1951). The immediate coastal belt consists of relatively undisturbed mangrove keys, salt marsh islets, and labyrinthine waterways. The region is underlain by Tertiary limestones and is a discharge area of the artesian ground water system of central Florida. The general study area extended from the Chassahowitzka River on the south to the Withlacoochee River on the north, and lay rough- ly between latitudes 28°40’ and 29°00’ N (Fig. 1). The floor of the Gulf of Mexico in this area is a gently rolling shelf that slopes gradually seaward. At low water, depths of less than 2 meters occur as far as 10 kilometers offshore. The floor is broken by oyster reefs, shell deposits, limestone outcrops, and sandbars (Vernon, 1951). Although many bars are exposed at low tide, depths of up to 3 meters occur in the interbar waters and in the cuts through the oyster reefs (Dawson, 1955). In contrast, the inshore or “back- 1 2 Spec. Publ. Amer. Soc. Mamm. 5 swamp’ waters that flow through the network of islands are ex- ceptionally shallow, averaging approximately 0.5 meter in depth at mean low tide (Dawson, 1955). According to the Tide Tables of the U.S. Coast and Geodetic Survey, the diurnal tidal range is 1 meter. Maximum tidal fluctua- tions of more than 2 meters have been recorded. Tides tend to be higher in summer under prevailing west and southwest winds, and lower in winter when east and northeast winds predominate. Salinity values within the study area range from 0.0 parts per thousand (ppt) in the rivers to normal sea water (35.0 ppt) off- shore. At the mouths of rivers salinity characteristics are, of course, estuarine. Dawson (1955) recorded a mean salinity of 16.0 ppt in Crystal Bay at the mouth of the Crystal River over the period September 1951 through August 1952. The coastal waters off Citrus County support extensive subma- rine meadows of the seagrasses Thalassia testudinum, Syringodium filiforme, and Ruppia mantima. Halophila engelmanni and Diplanthera wright occur in lesser abundance (Phillips, 1960a). In addition, Phillips (1960) found 46 taxa of algae, of which 25 were epiphytic. Thorne (1954) cited Acetabulum, Caulerpa, Gracilaria, Halimeda, Hypnea, Penicillus, Polysiphonia, Sargassum, and Udotea as typical as- sociates of the marine spermatophytes and Melobesia farinosa as a common epiphyte on Thalassia. The distribution of the seagrasses seems to depend primarily on salinity. Thalassia, Syringodium, and Halophila favor offshore marine conditions; Diplanthera and Ruppia thrive in inshore bays and estuaries of diluted salinity (McNulty et al., 1972). The coastline of Citrus County consists of an intricate maze of thousands of islands, which, like the margin of the mainland, are part of a vast salt marsh that extends along the coast well beyond the confines of the study area. The flora of the marsh is dom- inated by needle rush (Juncus roemerianus), cordgrass (Spartina pa- tens), and saw grass (Cladiwm jamaicense). Frequent hardwood ham- mocks characterized by cabbage palm (Sabal palmetto), red-bay (Persea borboma), magnolia (Magnolia virginiana), wax-myrtle (My- rica cerifera), swamp holly (Ilex cassine), yaupon holly (lex vomitoria), and red-cedar (Juniperus silicicola) interrupt the flat monotony of the land. The marsh is dissected by many shallow coves, small lagoons, and meandering tidal creeks. A few kilometers inland, coastal flatwoods replace the savannah. Hartman—Manatee in Florida 3 Crystel ws Bey a Fic. 1. Map of the study area. This forest surmounts a limestone escarpment that is directly con- nected with the artesian water of the Floridan aquifer (Vernon, 1951). From the escarpment arise three spring-fed rivers—from north to south, the Crystal, the Homosassa, and the Chassahow- itzka (Figs. 1 and 2). The physical, chemical, and biological properties of the large springs feeding these rivers have been analyzed by Furguson et al. 4 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 2. Aerial views of manatee habitats in Citrus County: A, headwaters of the Crystal River (arrow indicates site of Main Spring). (1947), Odum (1957), Wetterhall (1965), and Mann and Cherry (1969). The total daily output of the springs fluctuates with the tides and with seasonal rainfall. Of the three artesian rivers, the Crystal discharges the greatest volume of water. The combined output of its springs averages 2.25 million kiloliters (600 million gallons) of water per day, more than four times that of the Homo- sassa’s waters and more than six times the flow of the Chassahow- itzka (Mann and Cherry, 1969). The northern boundary of Citrus County is formed by the With- lacoochee River, which flows into the Gulf 10 kilometers north of the Crystal River. Midway between the mouths of the two rivers the Florida Power Corporation has dredged an intake and a dis- charge canal for a generating plant. An uninterrupted spoil bank bordering the intake canal projects 4 kilometers into the Gulf from the mainland. The entrance to the Cross Florida Barge Canal lies a kilometer south of the mouth of the Withlacoochee. Hartman—Manatee in Florida 5 Fic. 2 (cont.) B, headwaters of the Homosassa River. Coastal forest is dominated by cabbage palm (Sabal palmetto), bald cypress (Taxodium distichum), live oak (Quercus virginiana), red maple (Acer rubrum), red-cedar (Juniperus silicicola), magnolia (Mag- nolia virginiana), red-bay (Persea borbonia), and wax-myrtle (Myrica cerifera). Of the five major stream types in Florida recognized by Beck (1965), spring-fed rivers such as the Crystal, Homosassa, and Chas- sahowitzka are classified as calcareous streams, as distinct from sand-bottomed streams like the Withlacoochee. The Withlacoo- chee, in fact, differs strikingly from the other rivers in Citrus County. Whereas the Crystal, Homosassa, and Chassahowitzka rivers are all less than 13 kilometers in length, the Withlacoochee reaches more than 100 kilometers into the Florida interior and 1s 6 Spec. Publ. Amer. Soc. Mamm. 5 fed as much by surface runoff as by artesian water. Also, whereas the three spring-fed rivers of the study area are mostly broad, straight, and shallow, the Withlacoochee is comparatively narrow, winding, and deep. It averages 25 to 30 meters in width and_has a controlled depth of 3 meters at mean low water. A channel for small craft has been cut through the bars from its mouth to off- shore waters. Twenty kilometers upstream from the Gulf, a flood control dam presents an insurmountable barrier to manatees. Un- like the relatively clear waters of the spring-fed rivers, those of the Withlacoochee are darkened by tannic acid. Nonetheless, aquatic vegetation grows profusely along the banks of the river below the dam. The dominant species of submerged vascular plants found in the Withlacoochee were Hydrilla verticillata, Elodea densa, Cera- tophyllum demersum, Mynophyllum spicatum, and Ruppia maritima. Po- tamogeton pusillus and P. illinoensis occurred in isolated patches. Natant vegetation on the Withlacoochee consisted principally of Exchhorma crassipes and Salvimia rotundifolia. Of the three spring-fed rivers in Citrus County, the Chassahow- itzka is the smallest, shortest (9 kilometers), and shallowest. Except for a channel 2 to 3 meters in depth in its lowest reaches, it is exceedingly shoal, with depths rarely greater than | meter at low tide. No access channel has been dredged from offshore waters to the mouth of the river. The Homosassa River is longer (12 kilometers) and deeper than the Chassahowitzka. It consists of a chain of bays linked by nar- rows. Before reaching the Gulf its waters merge with a series of branch waterways that are actually tidewater extensions. A natural offshore channel 2 to 5 meters deep leads to its mouth. The depth of the river at mean low water ranges from | to 3 meters. The main boil at the river’s headwaters is the feature of a tourist at- traction. The Halls River, a spring-fed tributary entering 1 kilo- meter downstream, is unnavigable. The Crystal River enters the Gulf of Mexico 15 kilometers north of the Homosassa. The Crystal varies in breadth from 100 to 200 meters and flows 11 kilometers from its headwaters to the Gulf. From its mouth a channel 18 meters wide has been dredged through the oyster bars offshore to permit the passage of small craft. The channel’s controlled depth at mean low water is 2 me- ters. Maximum depths of the river range from 2 to 4.5 meters at low tide. Six-and-a-half kilometers upstream, a shallow tidal creek, Hartman—Manatee in Florida dl the Salt River, diverts a portion of the Crystal River’s backflow to the southwest and connects circuitously with the Homosassa River. At its head, the Crystal River expands into Kings Bay. During normal tidal cycles, maximum flow in the main channel of the Crystal River is approximately 112 cubic meters per second (m*/s) (Mann and Cherry, 1969). Velocities higher than 6 kilome- ters per hour (km/hr) have been recorded in gaps at the mouth of the Crystal and in the Salt River (Dawson, 1955). During hurri- canes, the Crystal River’s flow has been estimated at more than 280 m*/s (Mann and Cherry, 1969). The mineral content of water in the Crystal River is subject to wide variation. The concentration of dissolved solids, principally sodium chloride, is dependent on both seasonal and diurnal tidal cycles. During a two-and-a-half year period from 1964 to 1966, the concentration of dissolved minerals in the river fluctuated be- tween 300 and 15,000 milligrams per liter (mg/l) (Mann and Cher- ry, 1969). The focus of my study was on Kings Bay, the headwaters of the Crystal River (Fig. 2A). The bay, an area of roughly 165 hectares, is bordered by the town of Crystal River. Commerical fishermen and charter boatmen have operated out of Crystal River for more than 60 years. It is only within the last 15 to 20 years, however, that the shorelines of Kings Bay and the river itself have been altered by canal excavations, landfills, and the building of sea walls. The center of the town is situated near the northeast corner of Kings Bay. Much of the remainder of the bay is surrounded by real estate subdivisions. During the study, dredges were often at work somewhere on the bay or up one of its canals. In some places silt produced by dredging reduced underwater visibility to 1 meter, whereas, in the larger springs, visibility normally exceeded 30 me- ters. The consensus of local residents was that the bay and river were much clearer before housing development. The floor of Kings Bay is permeated by hundreds of springs (Fig. 3), the largest of which, the Main Spring, is the principal site of the manatee congregations. Where limestone is not exposed, the bay floor is of sand and muck, with a mean depth of between 2 and 3 meters. The shallow, relatively clear waters of Kings Bay are conducive to the growth of submerged aquatics. Species identified included Ceratophyllum demersum, Myriophyllum spicatum, Potamogeton pectina- 8 Spec. Publ. Amer. Soc. Mamm. 5 a cS ° 5 5 km/hr) probably discourage manatees from occupying certain tributaries of the St. Johns River. Aerial reconnaissance revealed that, in regions where tidal waters funnel through narrow channels, fast currents pose a navigational problem for manatees. Animals migrating via the Intracoastal Waterway on the Atlantic seaboard were never seen swimming against currents that exceeded 6 km/hr. In the Intracoastal Water- way north of the Haulover Canal (Brevard County), a manatee was clocked swimming 1.5 km/hr against a current estimated at 4 to 5 km/hr. Most of the drawbridge tenders on the Intracoastal Waterway felt that manatees were more disposed to swim along the shore than in the center of a channel, especially when faced with a swift current. Sightings from the air substantiated this opinion. In the Sebastian Inlet (Brevard and Indian River counties), currents in excess of 11 km/hr have been recorded (McCall et al., 1970). Cur- rents of such velocity surely deter manatees or force them to crowd the shore when swimming against the tide. Perhaps the animals wait for a change in the tide to slip in or out between the ocean and the Intracoastal Waterway. At Crystal River the activities of manatees were little affected by the generally modest flow of the waters. The current in the river 40 Spec. Publ. Amer. Soc. Mamm. 5 (2 to 3 km/hr) was not sufficient to influence the course set by an animal. Manatees, for example, did not hug the shoreline where the rate of flow is reduced. Nor was there any indication that man- atees increased their swimming efforts when countering a current; they simply took longer to cover a given distance when competing against a current than when cruising with it. Off Shell Island at the narrow mouth of the Crystal River where tidal flow was swiftest (6 km/hr), guides claimed to have seen manatees buck the current as they entered or left Crystal Bay; more often the animals chose alternate gaps where flow was reduced (4 to 5 km/hr). If undeterred by swift currents, manatees will apparently ascend rivers as far as the depths will allow. Within the study area, man- atees have been spotted as much as 20 kilometers up the Withla- coochee River. According to Moore (19516), animals have also been seen 100 kilometers from the Gulf at the junction of the Suwannee and Santa Fe rivers (Gilchrist, Lafayette, and Suwannee counties). Manatees in the upper St. Johns River (Volusia, Lake, and Putnam counties) are 230 kilometers from the ocean. In Nicaragua, man- atees have been observed 130 kilometers up the larger waterways (Townsend, 1904). Storms.—Wind and rain had no significant effect on manatee activity in Kings Bay. During storms of exceptionally high winds and heavy rain, animals conducted their daily routines uninter- rupted. They never responded to strong surface turbulence by seeking shelter on the leeward side of land masses. In more ex- posed coastal regions, however, water turbulence may play a role in the movements of manatees. According to Kingdon (1971), shel- ter from rough conditions is a factor determining the distribution of dugongs on the coast of East Africa. There may be a similar relation between turbulence and the predilection of manatees for inshore areas where they find protection from the heavy wave action engendered by storms. Out of 61 sightings made from the air over coastal waters, 92 per cent were on the inshore side of mainland or barrier beaches and only eight per cent on the sea- ward side. That storms can prove fatal to manatees was confirmed in 1972 when a live calf washed up on shore at Vanderbilt Beach (Collier County) during a strong northwester. Sun.—It was suggested by a resident of Crystal River that the sun influences the choice of resting sites. My observations did not support this theory. On sunny afternoons a difference in water Hartman—Manatee in Florida 41 temperature existed between the bottom and the surface, but this difference was minimized by the intermixing of currents generated by surfacing manatees. Furthermore, animals bottom-rested in precisely the same location regardless of whether it was sunny or overcast and usually in the midst of silt churned up by their tails. While observing cold-induced congregations of manatees in the turbid Miami River, Moore (195la) noted that animals remained longer at the surface with backs exposed as the sun warmed the air. At Crystal River, however, there was no evidence that manatees resting suspended with backs awash were sunbathing, for this ac- tivity occurred as often on cloudy days or at night as during sunny days. _ Turbidity.—Manatees showed no preference for murky or limpid water. Animals were never seen to seek camouflage or protection in turbidity of their own making. Vegetation.—The activities of manatees in Kings Bay were con- siderably influenced by vegetative growth. In many sections of the bay, the animals were forced to detour around dense clumps of Hydrilla, Ceratophyllum, and Myriophyllum. Near the Main Spring animals roamed under galleries formed by Hydrilla, sometimes plunging headlong through walls of the waterweed and occasion- ally becoming deeply enmeshed. At the head of the river, animals ranged through the margins of the Myriophyllum beds, emerging bedecked with stalks. Normally, manatees did not hesitate to sur- face under mats of vegetation, respiring through the debris. Un- usually dense mats of Hydrilla, however, discouraged manatees from surfacing and compelled them to seek openings to breathe. DAILY ACTIVITY HE results of this study indicate that manatees are essentially ae arhythmic. Individual animals observed in the headwaters of the Crystal River failed to manifest a predictable periodicity of activity. The frequency of feeding, resting, and other activities showed no consistent differences related to time of day, as sug- gested by the behavior of the manatees in Fig. 14. The activities of these animals was observed without interruption for periods ranging from 10.5 to 12 hours. The movements of the animals over the same period are traced in Fig. 15. Total distances covered 42 Spec. Publ. Amer. Soc. Mainm. 5 [__] Bottom-resting | Feeding =| Swimming SSS Playing LZ Courtship A ae. = — iii B > Pee = Sra < C GS eso SS accompanie juvenile acc. by adult 0600 0800 1000 1200 1400 1600 1800 HOUR OF DAY Fic. 14. Diurnal activities of manatees in Kings Bay and the upper reaches of the Crystal River: A, estrous female between 0704 and 1900 hours on 9 February 1969; B, adult female with calf between 0654 and 1832 hours on 14 January 1969; C, adult female between 0724 and 1813 hours on 6 January 1969; D, juvenile male between 0638 and 1828 hours on 23 January 1969. in Figs. 15A through 15D are 1.5, 4, 5.5, and 12.5 kilometers, respectively. Night observations of manatees revealed no change in their activities. Manatees occupied most of their time feeding, resting, idling, cruising, and socializing. Adult animals tended to feed six to eight hours a day in sessions that usually lasted from one to two hours. Resting consumed as few as two and as many as 12 hours a day. During cold snaps, animals were known to have spent the time from dawn to dusk resting on the bottom at the Main Spring. However, six to 10 hours resting, distributed through a 24-hour period in sessions of two to four hours, was more typical. The amount of time spent in social interactions was highly variable. In summer, solitary animals may go for days without coming in con- tact with other manatees. In contrast, social encounters within win- Hartman—Manatee in Florida 43 Fic. 15. Movements of manatees in Kings Bay and the upper reaches of the Crystal River: A, estrous female between 0704 and 1900 hours on 9 February 1969; B, adult female with calf between 0654 and 1832 hours on 14 January 1969; C, adult female between 0724 and 1813 hours on 6 January 1969; D, juvenile male between 0638 and 1828 hours on 23 January 1969. ter congregations were frequent. From the shore I have watched manatees play for as long as three and a half hours, feed and rest for six hours, then resume playing. Sociality peaks in an estrous herd where a cow and her male escorts may remain together for more than a month. 44 Spec. Publ. Amer. Soc. Mamm. 5 FOOD HABITS HE preceding discussion indicated that the movements of man- T atees in Florida are influenced by several factors, including depth, salinity, tides, currents, and access to warm water during winter cold spells. The distribution of manatees is also determined by the availability of food (Hartman, unpublished manuscript). Manatees are opportunistic feeders and have been observed graz- ing on a wide variety of plants both aquatic and terrestrial, fibrous and nonfibrous, vascular and nonvascular. They are even known to feed on fish. This section considers food preferences and feed- ing ecology. Food Plants and Preferences Vascular vegetation.—Captive and semicaptive manatees have been known to consume many types of vascular aquatic plants (Table 2). In addition to aquatic vegetation, captive animals have been fed lawn grass (Poa sp.); dandelions (Taraxacum officinale); sow-thistles (Sonchus oleraceus); palmetto fronds (Sabal palmetto); most of the garden vegetables, cultivated legumes, and pasture grasses; and many commercial fruits (Chapman, 1875; Murie, 1880; Crane, 1881; Coates, 1940; Gunter, 1941; Allsopp, 1961; personal observations). Bread, fish, and meat also are consumed (Crane, 1881; Mohr, 1957; Powell, 1978). In the wild, manatees, as other herbivores, probably select plants for their palatability, digestibility, and nutritional value (Heinsohn and Birch, 1972). Allsopp (1969) found that manatees used for weed control in Guyana generally preferred submerged, floating, and emergent vegetation in that order. I found a similar tendency among manatees in Florida. Crystal River manatees fed almost exclusively underwater, virtually ignoring natant and emergent plants as a source of food. In more turbid rivers, manatees fed on submerged plants when available and, in their absence, on floating vegetation, mostly water hyacinth (Ezchhornia crassipes). In those waterways that were destitute of aquatic spermatophytes, manatees resorted to grazing on algae and bank growth. Within the study area, manatees were relatively indiscriminate in selection of food and generally ate whatever submerged vege- Hartman—Manatee in Florida 45 TABLE 2 VASCULAR AQUATIC VEGETATION CONSUMED BY TRICHECHIDS UNDER CaPTIVE AND SEMICAPTIVE CONDITIONS. Genera Source Submergents Cabomba Allsopp, 1969 Ceratophyllum Chapman, 1875; Allsopp, 1969 Elodea Allsopp, 1969 Myriophyllum Allsopp, 1969 Najas Browder, 1967 Nitella Allsopp, 1969 Potamogeton Brown, 1878; Townsend, 1904; Allsopp, 1969 ’ Ruppia Allsopp, 1969 Utricularia Browder, 1967; Allsopp, 1969 Vallisneria Chapman, 1875; Allsopp, 1969 Zostera Townsend, 1904 Natants Azolla Allsopp, 1969 Eichhornia Allsopp, 1969 Lemna Allsopp, 1969 Mimosa Allsopp, 1969 Nelumbium Allsopp, 1969 Nelumbo Allsopp, 1969 Neptunia Allsopp, 1969 Nymphaea Allsopp, 1969 Paspalum Allsopp, 1969 Pistia Allsopp, 1969 Salvinia Allsopp, 1969 Victoria Allsopp, 1969 Emergents Alternanthera Allsopp, 1969 Hymenachne Allsopp, 1969 Ipomoea Allsopp, 1969 Leersia Allsopp, 1969 Luziola Allsopp, 1969 Montrichardia Allsopp, 1969 Panicum Browder, 1967 Sagittaria Brown, 1878; Browder, 1967; Allsopp, 1969 Typha Browder, 1967; Allsopp, 1969 46 Spec. Publ. Amer. Soc. Maiam. 5 tation was at hand. In fresh and brackish waters of the Crystal, Homosassa, and Withlacoochee rivers, manatees fed on the six dominant species of submerged plants—H)ydrilla verticillata, Vallis- neria neotropicalis, Ceratophyllum demersum, Myriophyllum spicatum, Ruppia maritima, and Diplanthera wrightu. Hydrilla comprised the bulk of the diet in the Crystal River head- waters but was not necessarily preferred to its associates Vallisnenia and Ceratophyllum. In areas where Ceratophyllum and Myriophyllum both grew, the animals showed a predilection for Ceratophyllum. Manatees also preferred Ruppia and Myriophyllum where they grew in association with Potamogeton pectinatus. Potamogeton, however, was occasionally grazed. Najas guadalupensis, highly localized in the rivers, was ingested incidentally. Potamogeton illinoiensis, P. pusillus, and Zannichellia palustris were found in traces in the waterways of Citrus County, but were never seen to be cropped by manatees. From interviewees it was learned that manatees fed on Elodea den- sa, which was largely restricted to the Withlacoochee River. In salt water, manatees have been alleged to feed on seagrasses (Barrett, 1935; Krumholz, 1943; Charnock-Wilson, 1968). This was confirmed during aerial surveys over the Gulf and Atlantic coasts of Florida. I observed manatees from the air feeding on Syringodium filiforme, Thalassia testudinum, and Diplanthera wrightu (Table 3). Animals appeared to favor Syringodium where it grew in mixed stands with Thalassia. For the most part, Diplanthera prefers intertidal water and was not found in association with Thalassia and Syringodium, which occur only below low water levels. Diplanth- era usually was consumed near the shore of turbid waterways that did not support beds of Thalassia or Syringodium. There are two additional species of seagrasses in Florida, Halo- phila baillons and H. engelmanni. Both are strictly limited in distri- bution (Phillips, 1960a). Halophila engelmanni was found growing in association with Thalassia at three manatee feeding sites, but it occurred only in traces close to the substrate and was apparently ignored by manatees. Floating vegetation held little attraction for manatees in Citrus County, although I found that water hyacinth was their staple food in many rivers that lack submerged plants. At Crystal River, man- atees were observed feeding on water hyacinth less than a half dozen times. All other natant vegetation, including water lettuce Hartman—Manatee in Florida 47 TABLE 3 DeEpPTH, SALINITY, AND LOCATION OF SITES AT WHICH MANATEES FEEDING ON MARINE SPERMATOPHYTES. WERE OBSERVED Water depth Salinity Plant Guay) (ppt) Body of water County Thalassia 1.0-1.6 31 Tampa Bay Hillsborough 0.9-1.0 26 Charlotte Harbor Charlotte 1.0 29 Turtle Bay Charlotte 1.2-1.5 31 San Carlos Bay Lee 0.9-1.2 27 Lostmans River Monroe 1.0-1.4 34 Norris Cut Dade Syringodium 1.0-1.6 31 Tampa Bay Hillsborough 1.1 30 Indian River Brevard 1.4 23 Indian River Brevard 1.4 19 Indian River Brevard Diplanthera 0.8-1.0 8 Withlacoochee River Citrus 0.7 11 Salt River Citrus 0.9-1.2 31 San Carlos Bay Lee 0.5 35 Indian River North Volusia (Pistia stratiotes), water lilies (Nymphaea mexicana), and water ferns (Azolla caroliniana), was disregarded by the animals. In the St. Johns River, manatees were seen to feed regularly on water hyacinth and alligatorweed (Alternanthera philoxeroides), but showed no interest in water lettuce. Water ferns (Salvinia rotundi- folia) were consumed only accidentally while manatees were feed- ing on water hyacinth. Citrus County manatees completely ignored emergent vegeta- tion and bank growth. Animals observed in the St. Johns River near Blue Springs Park (Volusia County) paid no attention to spat- ter-dock (Nuphar advena), shunning even its tender submerged seedlings. At other locations, however, manatees have been known to graze on littoral aquatics and phraetophytes (Barrett, 1935; Moore, 195la; C. Bertram and Bertram, 1968; Hartman, unpub- lished manuscript). Table 4 lists the bank grasses and forbs that manatees have been seen to feed upon in Florida. Apparently the foliage of mangroves does not appeal to manatees. Bangs (1895) reported that they fed on mangrove leaves when other food was unavailable, but no one interviewed during this study had seen manatees eating mangrove leaves. One interviewee claimed that the animals occasionally ate the floating seedlings of red mangrove 48 Spec. Publ. Amer. Soc. Mamm. 5 TABLE 4 BANK GRASSES AND FORBS COLLECTED FROM SITES AT WHICH MANATEES HAVE BEEN OBSERVED FEEDING. Type Body of water County Typha domingensis (young shoots) Tomoka River Volusia Distichlis spicata Tomoka River Volusia Hillsboro Canal Palm Beach Dania Cut-off Canal Broward Panicum purpurascens St. Johns River Volusia Panicum sp. Imperial River Lee Paspalum vaginatum Little Manatee River Manatee Eau Gallie River Brevard Paspalum sp. Intracoastal Waterway Volusia Earman River Palm Beach Spartina alterniflora St. Johns River Duval Tolomato River St. Johns Intracoastal Waterway Flagler Vigna repens Eau Gallie River Brevard (Rhizophora mangle). It is noteworthy that manatees in the vicinity of Blue Springs Park never were seen to feed on the leaves of overhanging trees and shrubs or on Spanish moss (Tillandsia us- neotdes). Although manatees generally avoid fibrous vegetation, those that congregate in the discharge of a power plant in Palm Beach County have eaten the fronds of coconut palms (Cocos nucifera) fed to them by local residents. The death of a young manatee at the Miami Seaquarium was attributed to an impacted intestine, the result of eating palm fronds (Warren Zeiller, personal communication). Algae.—In captivity, manatees have been reported to accept both fresh-water and marine algae, including Chara, Spirogyra, and Ulva (Chapman, 1875; Coates, 1939; Allsopp, 1969). In the wild, how- ever, it appears that algae do not figure significantly in the diet of manatees unless vascular vegetation is unavailable. In the Crystal River headwaters, manatees seldom fed on nonvascular plants and then only incidentally while foraging on the bottom for detritus. At such times, animals devoured clumps of Enteromorpha, Oscilla- toria, and Navicula. At the mouth of the Crystal River, manatees showed no interest in Sargassum, and from the air I never saw animals feeding on beds of marine algae. There is evidence that, Hartman—Manatee in Florida 49 TABLE 5 ATTACHED ALGAE COLLECTED FROM MARINAS, DOCKYARDS, AND JETTIES WHERE MANATEES HAVE BEEN OBSERVED FEEDING. Type Body of water County Chlorophyceae (green algae) Caulerpa prolifera Ft. Pierce Inlet St. Lucie Caulerpa racemosa Ft. Pierce Inlet St. Lucie Caulerpa sertularioides Ft. Pierce Inlet St. Lucie Chaetomorpha brachygona Intracoastal Waterway Broward Enteromorpha sp. Intracoastal Waterway Flagler Halimeda tuna Ft. Pierce Inlet St. Lucie Ulva lactuca Matanzas River St. Johns Tolomato River St. Johns San Sebastian River St. Johns Rhodophyceae (red algae) Agardhiella tenera Ft. Pierce Inlet St. Lucie Botrycladia pyriformis Ft. Pierce Inlet St. Lucie Champia parvula Ft. Pierce Inlet St. Lucie Gracilaria mammillaris Matanzas River St. Johns Grateloupia filicina Matanzas River St. Johns Hypnea cervicornas Ft. Pierce Inlet St. Johns Hypnea spinella Ft. Pierce Inlet St. Lucie Laurencia microcladia Ft. Pierce Inlet St. Lucie Wrangelia argus Ft. Pierce Inlet St. Lucie Cyanophyceae (blue-green algae) Lyngbya sp. Indian River Brevard Xanthophyceae (yellow-green algae) Vauchena sp. Matanzas River St. Johns under normal conditions, dugongs also shun algae and feed ex- clusively on spermatophytes (Gohar, 1957; Jarman, 1966; Hein- sohn and Birch, 1972). However, Heinsohn and Spain (1974) found that dugongs began to supplement their diet with brown algae following severe damage to their sea grass pastures by a cyclone. In turbid waterways with impoverished vascular flora, manatees supplement their diet by feeding on algae. On the east coast of Florida, it is well known that manatees graze on algae that grow on jetties, pilings, floats, docks, mooring lines, and the hulls of boats. Table 5 presents a list of attached algae collected from ma- rinas, dockyards, and jetties where manatees have been seen feed- 50 Spec. Publ. Amer. Soc. Mamm. 5 TABLE 6 ALGAL ASSOCIATES AND EPIPHYTON OF MARINE AND ESTUARINE SPERMATOPHYTES GRAZED BY MANATEES AT OBSERVED FEEDING SITES. Type Non-epiphytes Chlorophyceae (green algae) Caulerpa paspaloides Caulerpa prolifera Spirogyra sp. Ulva lactuca Rhodophyceae (red algae) Acanthophora muscoides Acanthophora spicifera Chondria lhttoralis Chondria sedifolia Dasya sp. Gracilaria compressa Gracilaria verrucosa Hypnea cervicornis Hypnea musciformis Polysiphonia spp. Spyridia filamentosa Phaeophyceae (brown algae) Dictyota dichotoma Cyanophyceae (blue-green algae) Lyngbya sp. Epiphytes Rhodophyceae (red algae) Ceramium subtile Ceramium sp. Polysiphonia macrocarpa Phaeophyceae (brown algae) Ectocarpus sp. Cyanophyceae (blue-green algae) Lyngbya sp. Bacillariophyceae (diatoms) Anhnanthes longipes Amphora spp. Grammatophora marina Body of water Crystal Bay Tampa Bay Crystal Bay Salt River Turtle Bay San Carlos Bay Indian River Turtle Bay Tampa Bay Crystal Bay Turtle Bay Tampa Bay Turtle Bay Indian River Crystal Bay ‘Tampa Bay Turtle Bay San Carlos Bay Tampa Bay Salt River Turtle Bay San Carlos Bay Indian River Crystal Bay Indian River Indian River Crystal River Indian River Indian River Indian River Indian River Indian River County Citrus Hillsborough Citrus Citrus Charlotte Lee Brevard Charlotte Hillsborough Citrus Charlotte Hillsborough Charlotte Brevard Citrus Hillsborough Charlotte Lee Hillsborough Citrus Charlotte Lee Brevard Citrus Brevard Brevard Citrus Brevard Brevard Volusia Brevard Volusia Hartman—Manatee in Florida 51 TABLE 6 (Continued) Type Body of water County Licmophora sp. Tampa Bay Hillsborough Mastogloia sp. Indian River Volusia Melosira nummuloides Indian River Volusia Nawvicula sp. Tampa Bay Hillsborough Nitzschia sp. Indian River Brevard Rhabdonema adriaticum Indian River Volusia Striatella sp. Tampa Bay Hillsborough Synedra hennedyana Indian River Brevard Synedra sp. Indian River Brevard ing. The animals seem to prefer green algae. They have been seen to feed on mixed associations of red and green algae but have only been found eating green algae where colonies consisting of a single species occur. In Everglades National Park, manatees may eat the alga Chara hornemanni that, according to Tabb et al. (1962), flour- ishes in the quiet removes of Whitewater Bay when the salinity is less than 30 ppt. Manatees cannot avoid ingesting apinlnde algae when feeding on spermatophytes. Nonepiphytic algae appear to be taken only accidentally while animals graze. In the fresh waters of the study area, the dominant algae associated with submerged spermato- phytes were Enteromorpha, Spirogyra, Cladophora, and Oscillator. Manatees made no effort to avoid feeding on such Aufwuchs. In estuarine and marine waters, manatees adventitiously consume a wide variety of algae associated with the seagrasses Ruppia, Di- planthera, Thalassia, and Syringodium (Table 6). In addition to the algae listed, a number of unidentified diatoms and dinoflagellates as well as a red and a blue-green alga were found on blades of Thalassia collected from manatee feeding sites. Feeding methods.—Manatees feed in a variety of manners de- pending on the growth form of the food source and the part of the plant being consumed. When feeding on submerged plants that produce simple leaves from basal clusters or rhizomes, such as the seagrasses and Vallisneria, manatees behave as grazers. In these circumstances, the body is generally tilted cephalad, the mouth and flippers often touching the substrate, especially when short growth is being cropped. It appears that manatees eat only the leaves of Thalassia, Syrin- godium, Diplanthera, and Ruppia, and there was no evidence that 52 Spec. Publ. Amer. Soc. Mamm. 5 they preferred young to mature plants. When feeding on Vallis- neria, however, they seemed to prefer the leaf bases, basal sheaths, and young shoots to the mature blades. The rhizomes were not touched. Where food plants attained considerable height in the water or floated on its surface, manatees tended to act more like browsers. In many areas of Kings Bay, for example, Myriophyllum and Hy- drilla formed dense “jungles” of stalks that reached 2 to 4 meters to the surface. Manatees fed haphazardly on these stems, attacking them at any point from crown to base. Stems, leaves, and flowers were consumed indiscriminately. Manatees routinely fed on Hydrilla by cropping marginal growth or by immersing themselves in clumps. Animals sometimes became buried in masses of Hydrilla, hollowing out a cave as they fed. In such cases they were forced to back out in order to surface but usually returned to the same hole to continue feeding. Occasionally they had difficulty disengaging themselves from the tangle of vege- tation and had to plunge through it to escape. Manatees also fed on rafts of floating Hydrilla uprooted by their activity. Feeding manatees normally concentrated on a particular clump of Hydrilla but sometimes roved from one to the next. Strands of vegetation often dangled from the mouths of animals making their rounds. I once watched four manatees drift in tandem out of a canal snatching at Hydrilla as the current swept them along. The animals “browsed” Ceratophyllum from the top of one clump to another, favoring the dense, young, uppermost “coontails.” A juvenile male once fed with nose awash on the uprooted, broken, decaying stems and leaves of Ceratophyllum that had collected near the surface in a mat of Hydrilla. Observations of manatees feeding on water hyacinth were made chiefly on the St. Johns River at Blue Springs Park (Volusia Coun- ty). Here, rafts of water hyacinth collect beside the shore or among beds of spatterdock. Manatees fed on these rafts, often working their way slowly along the banks of the river. The animals grazed at the periphery of a raft or fully immersed in the tangle of roots and stalks. Upon surfacing under a “ceiling” of plants, their heads covered with hyacinths, animals sometimes gave the impression of exhaling with added force in an apparent effort to clear their nos- trils of obstructing vegetation. When small clumps of water hy- acinth broke free from the rafts and were carried across the water Hartman—Manatee in Florida 53 by wind or current, manatees often intercepted them as they drift- ed overhead. Manatees approached water hyacinths from below the surface or with their heads and backs awash. Reaching a plant, they would evert their lip pads, grasp a stalk or blade above the surface, then pull the whole unit underwater. Occasionally an animal projected its entire head above water in order to seize a leaf. While feeding, manatees either remained exposed at the surface or submerged, sometimes sinking to the bottom to finish eating. In several in- stances, two or three animals were seen sharing the same plant below the surface. Underwater, the manatee held the water hyacinth in front of its mouth with the tips of its flippers and delicately cropped each leaf. In general, only the leaves were consumed. They were clipped just above the inflated portion of the petiole. Stalks and roots were normally rejected. When an animal had eaten all of the leaves, it released the plant and rose to the surface to grasp another. On occasion, entire plants, roots included, were consumed. In those few instances when manatees were observed to feed on water hyacinth at Crystal River, they appeared first to investigate the clumps by nibbling, then dragged them under water and de- voured them completely—leaves, petioles, and roots. A few times manatees were seen investigating and nibbling clumps of mori- bund water hyacinth without actually eating the plants. When feeding on floating mats of alligatorweed in the St. Johns River, manatees usually remained at the surface chewing with their heads awash. The stems, leaves, and roots of this plant were eaten indiscriminately. Additional sources of food.—Manatees consume a variety of inver- tebrate periphyton when feeding in fresh or salt water. The in- vertebrate fauna associated with manatee food plants includes: am- phipods; isopods; tiny shrimp, crayfish, and crabs; insect larvae; bivalves; snails; leeches; nematodes; platyhelminths; polychaete worms; tunicates; hydroids; bryozoans; anemones; and brittle stars. Gammarid and caprellid amphipods were found to be par- ticularly common in aquatic vegetation. Arthropods occur in the vegetation in such abundance as to suggest that they contribute an important source of protein to the manatee’s diet. Manatees were occasionally observed to forage on the bottom, nibbling the surface of the substrate and pausing to chew detritus. Spec. Publ. Amer. Soc. Mamm. 5 54 uoneotunuWo0s [euosiad ‘PIPI9H “SA uoneAslasgo yeuosiad ‘uewnsieY IP6T ‘t93UuND I881 ‘oueID 1881 ‘euety O88I ‘auNW GL8T ‘ueudeyD 201N0g “SINONANI SQHOFHOIYT GNV SALVNFW SQHOFTHOIN LT, ‘SAALVNVJ AALLAVT) AO NOILAWASNO’) GOO] ATIVG LATAVL SP-LG L OS-OF OI (sure180]1y) Aep aad pauinsuos junoury sn] wauiedy a[0.1v9S9 aatpua 20n3] JAIpua 20ni13] 20nn2] ALO]IIPTIM queid OFI hall OFG 861 (s19]9WINUaD) yisuaT (sue130]1¥) ISIIM aayeury] sindunua sSnyDUDUL * snyouDU * smdunua sindunut * snypubU * SnyDUDUL * saivads Sas Hartman—Manatee in Florida 55 It often appeared that they were feeding not only on decaying organic matter but also on a mixture of sand and mud, perhaps in quest of trace elements or other essential minerals. Several times I saw animals ingest manatee feces. More often, they simply nuz- zled and chewed the dung, then allowed the mash to filter out of their mouths. Most extraordinary is the recent documentation of piscivorous behavior in manatees on the northern coast of Jamaica (Powell, 1978). It appears that sirenians in that area systematically visit gill nets set by local fishermen and deflesh the small scaled fish (ca- rangids and scombrids) caught in the mesh. Daily food consumption.—During each of the winters of the study, manatees gradually denuded whole stands of Hydrilla, Ceratophyl- lum, and Myriophyllum in Kings Bay, but, because the attendance of animals in the bay was irregular and because other vegetation- removing factors (for example, herbicides, weed harvesters, dredges, anchors, outboard propellers, and waterfowl) constantly intruded, it was impossible to quantify daily food consumption per animal. The only available data on daily food consumption are from captive individuals (Table 7). Captives have been recorded to eat up to one fourth their body weight per day in wet greens or as much as | kilogram of vegetation per 5 centimeters of body length. For comparison, a pair of captive dugongs daily ate 1 ki- logram of seagrasses (wet weight) per 7 centimeters of body length (Jones, 1967). Feeding Ecology Feeding sessions.—Manatees fed in discrete sessions during which they concentrated on one species of plant. Feeding sessions nor- mally lasted between 30 and 90 minutes but continued for more than two hours if animals were exceptionally hungry. From the shore I once clocked a bull who fed for 75 minutes of which 70 were spent feeding and the remaining five surfacing and stationing himself in front of the food. Calves fed with less concentration and for shorter periods than did adults. They often had to wait for their mothers to finish feeding. As animals became sated, they frequently interrupted feeding to rest. Eventually they would cease feeding altogether and move off to other pursuits. Depths.—The depths at which manatees feed are determined by 56 Spec. Publ. Amer. Soc. Mamm. 5 tides, by the depth at which plants grow, and by the growth forms of the food species. Table 3 includes the depths at which manatees were seen feeding on various seagrasses. In Kings Bay, animals were observed feeding from the surface to a depth of 4 meters. In the lower river where submerged vegetation is strictly littoral, man- atees fed on the narrow bank close to shore at depths of approx- imately a meter. In the Salt River and neighboring estuarine waters, animals occasionally grazed on flats as shoal as 80 centi- meters. In the St. Johns River, near Blue Springs Park (Volusia County), manatees normally fed on water hyacinth in water a me- ter or more in depth, but were occasionally observed feeding near the shore in water only 50 centimeters deep. In Guyana, manatees were observed to feed on bank growth in water shallower than 50 centimeters (Datakaran Jeetlall, personal communication). Animals cropped Thalassia, Syringodium, Diplanthera, Ruppia, and Vallisnerra on or near the bottom, depending on the size of the plants. At manatee feeding sites, Thalassia and Syringodium were found growing to heights of 60 centimeters, Diplanthera and Ruppia to 20 and 15 centimeters, respectively, and Vallisneria to 50 cen- timeters. Manatees usually fed on Ceratophyllum and Myriophyllum at middle depths of 1 to 3 meters in Kings Bay. They grazed on Hydrilla from the bottom to the surface, sometimes with their backs awash. Animals never descended below 4 meters to feed on Hy- drilla, although it grew at depths of 10 meters in the Main Spring. There are vast areas of manatee habitat in which shallow water renders potential food plants inaccessible. Syringodiwm and Di- planthera grow profusely in the Indian River north of Titusville (Brevard County), but only Syringodium is found at depths acces- sible to manatees; Diplanthera grows from the shoreline to a depth of only 26 centimeters (Kerry Clark, personal communication). In the St. Johns watershed, Ruppia, Zannichellia, Najas, and the algae Chara and Nitella are normally found in water 20 to 60 centimeters deep, unavailable to manatees. Selection of feeding sites and timing of movements.—Although at one time or another manatees were seen feeding in every accessible corner of Kings Bay, they were generally selective in their choice of feeding sites. It was not unusual for an animal to return re- peatedly to a preferred feeding area until its resources had been depleted or until another site had found favor. For example, an Hartman—Manatee in Florida 57 adult female returned to Kings Bay at least four times within two weeks to feed on precisely the same patch of Vallisneria. During one week in October, two cows with calves returned daily to feed on a single bed of Hydrilla. A bait-house owner on the Withlacoo- chee River reported that a group of manatees appeared daily for three weeks to feed below his dock on Elodea. From the air, man- atees were repeatedly seen feeding in the same general coastal areas, further suggesting that they have preferred feeding sites. There is evidence that manatees time their movements to coin- cide with the availability of food. In the Banana River (Brevard County), I have seen 10 or more manatees congregate in a cove to feed on floating mats of Syringodium and Diplanthera blown inshore by the wind. During the rainy season in southern Florida, manatees move up the canals to the flood control dams to feed on water hyacinth and other vegetation pouring through the gates (Hart- man, unpublished manuscript). Similar behavior has been report- ed in the Demerara River of Guyana where manatees collect at the outlets of drainage canals to intercept floating plants sluiced through the locks (W. H. L. Allsopp, personal communication). It has even been recorded that Amazonian manatees congregate to feed on fruit that drops to the water from overhanging trees (Nunes Pereira, 7n G. Bertram and C. Bertram, 1973). In some areas the feeding habits of manatees are dependent on the season. In the winter, growth of aquatic vegetation is termi- nated or much reduced. Leaf kill of Thalassia and Syringodiwm in the offshore waters of Citrus County was not determined, but in inshore waters Diplanthera suffered extensive leaf kill in winter. Ruppia generally survived in the rivers, and there was always an adequate supply of submerged plants in the warm headwaters. In many segments of the Intracoastal Waterway, however, algae die off each winter, and manatees must find alternative sources of food. In estuarine waters, the distribution and abundance of Ruppia and Diplanthera fluctuate seasonally with the rains (Phillips, 196 1a; Tabb et al., 1962). During periods of high salinity, Diplanthera re- places Ruppia, and manatees presumably shift feeding from one to the other. Heinsohn and Spain (1974) reported that the diet and move- ments of dugongs were influenced by cyclone-induced damage to 58 Spec. Publ. Amer. Soc. Mamm. 5 seagrass beds. Although a reasonable possibility, no comparable influence on the feeding behavior of manatees has been docu- mented. Freshwater needs.—There is growing evidence that manatees re- quire fresh water. They may seek it to rid themselves of barnacles and marine ectoparasites, but their attraction to hyposaline water seems more directly related to osmoregulation. Their apparent need for fresh water was first noticed in 1681 by William Dampier who wrote: “Manatees that live in the sea do commonly come once or twice in 24 hours to the mouth of any freshwater river that is near their place of abode” (Dampier, 1906). A need for fresh water also may have been shared by Steller’s sea cows, which were drawn to the mouths of brooks on Bering Island (Steller, 1751). Phillips (1964) reported that a captive manatee housed in a saltwater tank drank fresh water regularly from a hose. (All captive manatees, to my knowledge, have been kept in fresh water or provided with a source of fresh water.) Phillips also proposed that manatees travel up rivers specifically to drink fresh water. This theory may account for the overwhelming prevalence of manatee sightings in fresh- water and estuarine habitats. Many interviewees claimed that heavy rains often are followed by the appearance of manatees at the heads of tidal rivers. In the canals of southern Florida, manatees are commonly seen just below the flood control dams, presumably drinking the fresh water sluiced through the spillways. Groups of manatees are found almost daily below the spillway on the Fakha Union Canal (Collier County). In Ft. Lauderdale, I once watched a manatee swim up to the base of a flood control dam and fan its mouth as if drinking. It appears that manatees are drawn to virtually any source of fresh water. They have been observed drinking sulfur water flow- ing from a pipe into the Tolomato River (St. Johns County). They were attracted to the fresh water discharging from a high pressure hose used at a construction site on the Intracoastal Waterway (St. Johns County). In years past, manatees focused their activity around a storm drain at the City Yacht Basin in Daytona Beach. They were also known to gather around the discharge of the mu- nicipal sewer in St. Augustine. At least a dozen interviewees had observed manatees drink fresh water from hoses hanging off piers. One of the few manatees recently seen in the Lower Keys turned up near a source of fresh water at a marina in Key West. According Hartman—Manatee in Florida 59 to a report in the November 1972 issue of Conservation News (Flor- ida Department of Natural Resources), manatees in the Titusville Yacht Basin are attracted to boats that are being washed in order to drink the freshwater runoff. Manatees also may drink from artesian springs at their point of discharge into salt or brackish waters. It may not be far-fetched to suggest that manatees drink fresh water at the surface during heavy downpours. I have personally observed manatees feeding in fresh water and in water as salty as 35 ppt (Table 3). This feat is most singular when one considers that the vascular vegetation on which manatees graze contains over 85 per cent water (Rickett, 1924; Phillips, 1960a). Feeding in both fresh and salt water must place consid- erable demand on the water balance mechanisms of manatees. The manatee kidney, in fact, bears little resemblance to the kidneys of other mammals, including the dugong (Owen, 1838; Murie, 1874; Beddard, 1897; Petit, 1925; Hill, 1945; Quiring and Harlan, 1953; Batrawi, 1957), but it is not known if manatees excrete less water under conditions of hypersalinity. INTERACTIONS WITH OTHER ANIMALS Nt paid little heed to the indigenous fauna of Kings Bay. They took no interest in crabs (Callinectes sapidus), coo- ters (Chrysemys floridana), sting rays (Dasyatis sabina), or large species of fish such as longnose gar (Lepisosteus osseus), sharp-nosed sharks (Rhizoprionodon terraenovae), and tarpon (Tarpon atlanticus) and passed unperturbed through dense schools of migratory marine fish. On occasion, however, manatees were startled by commotions at the surface. The animals were sometimes frightened by jacks (Caranx hippos) splashing after needlefish (Strongylura marina), by coots (Fulica americana) taxiing directly overhead, or by pelicans (Pelecanus occidentalis) plunging nearby. Ordinarily the reaction was to flee a few meters, then return to investigate the source of the disturbance. Manatees resting on the bottom were pestered by fish, which pecked their hides apparently in quest of microorganisms. The manatees rarely reacted when their backs were pecked by smaller 60 Spec. Publ. Amer. Soc. Mamm. 5 fish such as gray snappers (Lutianus griseus), pinfish (Lagodon rhom- boides), and bluegills (Lepomis macrochirus), but winced whenever such fish pecked in their orbits. They flinched and slapped inef- fectually with their flippers when pecked by large fish such as jacks and sheepshead (Archosargus probatocephalus). One morning a single sheepshead molested a group of seven manatees who, in their ef- forts to avoid being pecked, rolled on their sides and upside down but were finally forced to leave their resting site. At the same lo- cation the next day, however, a sheepshead pecked but failed to irritate a young cow. Manatees also encountered otters (Lutra canadensis) and bottle- nosed dolphins (Tursiops truncatus) at Crystal River. I watched such encounters from the runabout. Only one otter-manatee confron- tation was observed. Three otters crossed the path of a bull as it was cruising through a cut; neither otters nor manatee hesitated or changed course. Dolphin-manatee encounters were seen on four occasions. Each time, two or more dolphins passed casually through a group of manatees and were apparently ignored by the sirenians. The reaction of manatees to alligators (Alligator massissippiensis) was never observed. Alligators, particularly larger individuals, had been all but exterminated in Citrus County at the time of the study. INTEREST IN INANIMATE OBJECTS ANATEES frequently reconnoitered the bottom, pausing to M nudge and nibble rocks and organic debris as well as bottles, cans, and other foreign articles. An animal often satisfied its cu- riosity by taking an object into its mouth and tonguing it or gently grinding on it with its teeth. The following incidents at the Main Spring exemplify the interest of manatees in inanimate objects. An adult male picked up an 8-centimeter piece of palmetto bark and carried it for a few meters. An old bull picked up an 18- centimeter forked piece of waterlogged driftwood, swam with it for a short distance, then let it fall. A cow nudged a 40-centimeter log along the sand. Her calf rolled a beer can with his muzzle and later pushed a piece of sunken wood across the bottom for several Hartman—Manatee in Florida 61 meters before losing interest. The same calf was also seen to pick up a piece of limestone and carry it to the surface before releasing it. A gravid cow chewed briefly on a small rock, audibly sloshing it in her mouth. Animals were also seen nibbling and transporting odds and ends of glass, plastic, and rubber found on the bottom. A juvenile male chewed on the head of a Coca Cola bottle, eventually picked it up by its neck and dropped it. Another young male chewed on my rubber flippers for five minutes as I stood on a ledge overlooking the spring. A juvenile cow once nibbled the edge of my polyeth- ylene clipboard as it lay on the ledge. On another occasion a young male picked my thermometer up from the bottom, munched on it momentarily, then dropped it. Once when I dangled the ther- mometer on a string in front of a juvenile male, he approached it, everted his lip pads, tucked the thermometer end-first into his mouth and chewed on it until I jerked it free. The manatees responded with curiosity to a 4-meter steel rein- forcing rod that I dangled with strings from the surface in an effort to measure them. The more inquisitive animals nibbled and rubbed on it. Eventually a cow caught one of the strings under a flipper and broke the line. When my runabout was anchored at the Main Spring, manatees often nudged and nibbled the anchor and anchor rope. An inor- dinately tame young female regularly investigated the boat by nuz- zling its hull and the propeller on the outboard (Fig. 27). Occa- sionally she grasped the anchor rope to her belly with her flippers, nibbling and rubbing on it (Fig. 191). Another tame female had a habit of chewing on the hydrophone cable. Manatees occasionally nuzzled floating objects such as buoys, branches, palmetto fronds, beer cans, dead coots, and pieces of paper. A juvenile female once nudged a strip of floating facial tissue and surfaced with it on her nose. 62 Spec. Publ. Amer. Soc. Mamm. 5 PARASITES AND COMMENSALS Endoparasites CCORDING to Jones and Johnson (1967), sirenians are parasit- A ized by digenetic trematodes and ascarid nematodes. Dailey and Brownell (1972) listed 10 species of trematodes and two species of nematodes found in the stomach, intestines, and caecum of dugongs. The paramphistomid fluke Chiorchis fabaceus has been obtained from the intestines of Trichechus manatus, T. senegalensis, and T. inunguis (Sokoloff and Caballero, 1932; Price, 1932; Baylis, 1936). Another fluke, Opzsthotrema cochleotrema, has been found in the alimentary system of 7. manatus (Dailey and Brownell, 1972). Numerous intestinal roundworms (Plicalolabia hagenbecki) were found in a young male T. manatus that died in 1969 at Busch Gardens in Tampa (Radhakrishnan and Bradley, 1970). Hill (1945) found a small fish inside the prepuce of a male du- gong and suggested a commensal association. As no fresh manatee carcasses were available for autopsy during this study, examination for endoparasites was confined to fecal analysis. A single fecal sample from the floor of Kings Bay was subjected to three standard techniques—iodine-stained direct smear, formol-ether sedimentation, and zinc sulphate flotation. A number of unidentifiable nematode eggs were present. External Associates The hide of manatees supports an assortment of plant and an- imal associates. Many are difficult to categorize as parasites or com- mensals. The only previously identified sirenian ectoparasite is a copepod, Harpacticus pulex, discovered on two captive manatees at the Miami Seaquarium (Humes, 1964). Local bacterial and fungal infections are encountered among captive animals (personal ob- servations), but there is no evidence that the organisms involved infest manatees in the wild. The Crystal River animals were without visible infections other than occasional pus-filled tumors. Water mold grew harmlessly in minor wounds. The hides of some individuals bore incrusted patches, the cause of which was perhaps a parasite. The crusts did not seem to irritate the animals but, once acquired, remained despite the constant epi- dermal sloughing characteristic of manatee hide. The incrusted areas were ridged and grooved and seemed to shelter a greater number of organisms than the smooth surfaces of the skin. Hartman—Manatee in Florida 63 A few animals supported extensive growths of a blue-green alga (Lyngbya martensiana) on their backs, suggesting that they had spent most of their time in rivers and estuaries. A red alga (Compsopogon coeruleus) grew from the tips of hairs and vibrissae on some man- atees. The backs of animals arriving in Kings Bay from the Gulf were often coated with a brown scum composed largely of the diatoms Zygnema and Nawvicula. In fresh water the diatoms died, and the scum eventually wore off. The presence of the scum in- dicates that some manatees pass considerable time outside rivers. Occasionally animals arrived in the Crystal River headwaters with balanid barnacles on their backs and tails. The barnacles grad- ually died in fresh water and dropped off, leaving their hosts tem- porarily mottled. That barnacles settle and mature on manatees is another indication that the animals spend protracted periods away from fresh water. It was on the basis of barnacles and algal growth that Moore (1956) believed manatees living predominantly in the Miami River could be distinguished from those living in Biscayne Bay. During the summer and autumn, manatees arriving from the Gulf were often accompanied by remoras (Echeneis naucrates). One animal was seen hosting over 20 fish. The remoras tended to re- main with the same individual, avoiding manatees that had been in fresh water longer. The fish are evidently intolerant of hypo- salinity for they abandoned their hosts a few days after entering fresh water, presumably to return to the Gulf. The remora-man- atee association is further evidence that manatees make extensive use of marine environments. Epidermal organisms were collected by scraping the hides of manatees with the open ends of plastic vials. Scraping was concen- trated on the back and tail and in the folds and recesses under flipper and chin. No lice or mites were found. Smooth-skinned manatees were free of all but microscopic protozoans, nematodes, copepods, and ostracods. Samples from backs covered with Lyngbya were more productive, yielding, in addition to the above, amphi- pods, isopods, and dipteran larvae. Tiny gastropods and small leeches, churned up from the bottom, appeared on manatees from time to time but soon disappeared. In general, the macroscopic associates of manatees seem to be free-living and ectophoretic rather than parasitic. Incrusted and algae-covered hides provide them with transport and a suitable microhabitat. The microscopic inhabitants of the skin apparently 64 Spec. Publ. Amer. Soc. Mamm. 5 range from casual commensals to obligate parasites. The euryhaline habits of manatees may be responsible for the comparative scarcity of parasites and commensals; potential associates would be sub- jected to rigorous osmotic stresses. MAINTENANCE BEHAVIOR Locomotion Ae manatee’s aquatic stability seems to stem from the elongate shape and dorsal position of its lungs and from the pachyos- totic condition of its skull and ribs. The heavy bone structure ap- parently serves as ballast and contributes to the animal’s neutral buoyancy. Virtually weightless underwater, the manatee swims with agility and grace. Its movements are ponderous only when compared with those of the Cetacea. Manatee hydrobatics and body postures include somersaults, half gainers, back dives, head and tail stands, barrel rolls, and upside-down gliding. These antics, however, are incidental to routine locomotion and will be dealt with individually under later headings. The present section con- siders the locomotory processes manifested in normal swimming and maneuvering. Use of the tail—The hydrodynamics of swimming in the Sirenia are virtually unstudied. A cruising manatee propels itself by un- dulating its tail dorsoventrally like a cetacean (Figs. 16 and 17). The propulsive forces in cetacean locomotion are discussed and diagrammed by Slijper (1962), Lang (1966), Gray (1968), and Her- tel (1969). In the Cetacea, thrust is developed by movement of both the tail and the body; the body is flexible and undulates with each stroke of the tail. I observed no such body motion among manatees, although it may occur at high speeds. It is apparent, however, that each stroke of the tail displaces the body vertically, the degree of pitch increasing with the power of the stroke (Fig. NZ). There are also basic structural differences between manatees and cetaceans, the outcome of which is that manatees are compar- atively inefficient swimmers and are unable to reach or sustain high speeds. Manatees are less streamlined than cetaceans, which undoubtedly results in reduced laminar flow and increased drag Hartman—Manatee in Florida 65 ie 5 10 y 12 17 22 f os ee 32 ‘\ = ‘a ; Fic. 16. Lateral view of a cruising manatee. Drawings are from an 8-mm film sequence and encompass one complete undulation of the tail. Arrows indicate di- rection of movement. The heavy lines passing through the animal’s approximate center of gravity represent the long axis of the body and illustrate the slight devia- tion from the horizontal plane (light lines) engendered by each stroke. Numbers refer to the number of the film frame. during swimming. Manatees also lack the peduncular musculature found in the Cetacea and cannot generate comparable thrust. Manatees seem to beat the tail about a pivotal point near its base compared with cetaceans, in which the fulcrum of thrust appears to be above the anal aperture. In full stroke the tip of a manatee’s tail describes an arc of only 130 degrees as opposed to an arc of over 200 degrees described by the fluke of a dolphin (pictured in Slijper, 1962). Moreover, the manatee’s tail is flexible and has a 66 Spec. Publ. Amer. Soc. Mamm. 5 35 40 45 Fic. 17. Posterior view of a cruising manatee. Drawings were made from an 8- mm film sequence and encompass one full undulation of the tail. Coordinates illustrate body’s relative displacement along a vertical axis. Number indicates num- ber of film frame. slight positive dihedral, which results in deflection of the caudal and lateral margins during downswings. Manatees initiate move- ment from a stationary position by an upswing of the tail followed by a downswing, repeated until the undulatory rhythm is estab- lished. Hartman—Manatee in Florida 67 A B C Fic. 18. Diagrammatic representation of the use of the tail as a rudder. Pos- terior view of a cruising manatee in forward glide (A), preparing to turn to the right (B), and preparing to turn to the left (C). Roll does not always accompany a turn. It should be mentioned here that dugongs are more streamlined than are manatees and show other structural modifications, in- cluding a fluke, that suggest a greater specialization for aquatic life than is found in trichechids. The tail of the manatee also serves as a rudder (Fig. 18). Ac- cording to Kinzer (1966), the tail may be at an angle as great as 45 degrees to the long axis of the body when a manatee is turning in a circle. Cruising animals can steer, bank, and roll by means of the tail alone. Immobile or idling manatees frequently adjust the angle of the tail to counteract roll and, to some degree, yaw. At the same time, the tail may be elevated or depressed to adjust the longitu- dinal axis of the body. Stroke rate.-—Stroke rate and the amplitude of undulations of the tail vary with the state of activity. Defining a stroke as one complete undulation of the tail, I determined the stroke rates of adult and subadult manatees while idling, cruising, and fleeing. Adult man- atees were seen to stroke 18 to 20 times per minute when idling and 24 to 36 times per minute while cruising. Idling involves sev- eral moderate strokes followed by a glide. Idling manatees swim with less regular strokes than when cruising and either dangle their flippers or adduct them at the wrists under their chests. When cruising, in contrast, manatees tend to maintain a constant rhythm of even strokes, the tail moves through a greater arc, and the flippers are held back flush against the body. Maximum stroke rate and amplitude are associated with flight. Fleeing manatees were observed to stroke their tails 45 to 50 times per minute. 68 Spec. Publ. Amer. Soc. Mamm. 5 Ventral view nel Fic. 19. Selected manatee attitudes illustrating various uses of the flippers (drawn from photographs). Stroke rate also varied with size. Calves were found to have a higher stroke rate than did older animals engaged in the same activity. The number of strokes per minute for cruising calves ranged from 36 to 44 as compared with 24 to 36 for adults. To keep pace, calves must maintain a higher stroke rate than their Hartman—Manatee in Florida 69 mothers. The cruising speeds of cows with calves were clocked at Crystal River, and it was discovered that mothers, apparently to accomodate their offspring, swim more slowly than do other adults. One cow with a pre-yearling calf was clocked cruising ap- proximately 1 km/hr slower than a cow with a yearling calf. Speeds.—Fishermen from British Honduras claim that manatees attain speeds up to 48 km/hr (Charnock-Wilson, 1968). Jarman (1966) estimated that cruising dugongs swam at 9.6 km/hr. At Crystal River, I clocked manatee speeds with a stopwatch while following animals in the runabout over a 30-meter course. This clocking method, although crude, involved 10 different manatees and suggested that swimming velocities range from 2 to 25 km/hr, depending on the activity. Idling manatees swam at 2 to 3 km/hr, cruising animals at 3 to 7 km/hr, and animals in flight at 18 to 25 km/hr. The sprints of fleeing manatees were short, usually 20 to 30 meters and never more than 100 meters. I obtained one esti- mate of manatee speeds from the air; three animals moving along the beach at Sanibel Island (Lee County) cruised at approximately 3.5 km/hr. Use of the flippers.—The manatee flipper is a highly maneuverable appendage (Fig. 19). Only the forearm and hand are free from the torso, but the elbow and wrist permit remarkable flexion (Flow- er and Lydekker, 1891). Davilliers (1938) stated that the manus can be rotated 180 degrees in a parasagittal plane. Lateral reach, on the other hand, is limited by the downward displacement of the forelimb. Manatees differ strikingly from the cetaceans in the use of their flippers for locomotion. While cruising, manatees do not utilize their flippers as hydrofoils as is typical of cetaceans (Felts, 1966). Instead, the flippers are normally held motionless against the sides. Moore (1956) noted an adult manatee that occasionally swam using flippers as well as tail, alternately stroking with right and left flip- per. I did not observe such use of the flippers during my study. It should be noted that, unlike manatees, cruising dugongs may use their flippers. Prater (1928) claimed that when swimming, du- gongs employ the flippers to turn and to maintain balance. Newborn manatees are reported to swim exclusively with their flippers (Moore, 1956, 1957). The youngest calf observed at Crystal River was two to three months of age, and it swam entirely with its tail. 70 Spec. Publ. Amer. Soc. Mamm. 5 The flippers are normally used only for precise maneuvering and for minor corrective movements to stabilize, position, and orient an animal while it is feeding, idling, investigating, or so- cializing. In such situations, the flippers may be worked simulta- neously or independently. On the bottom, the flippers function as the sole source of motion. Manatees use the tps of their flippers to balance on the bottom and to propel themselves forward or backward, barely grazing the substrate with their bodies. Animals “walk” on the bottom with alternate movements of the flippers. When greater speed is re- quired, the flippers are manipulated in unison, pulling the animal ahead in a series of lunges until enough momentum is gained to launch off the bottom into a full cruise. Off the bottom, the flippers are usually used in conjunction with the tail during such activities as feeding or cavorting. In maneu- vering to the surface, the flippers are feathered to complement the rudder action of the tail; in turning, they themselves may be em- ployed as rudders (Fig. 19F). Flippers and tail also coordinate as a unit to control pitch. The flippers sometimes serve as the only instruments of loco- motion, initiating or retarding forward or backward movement. By means of forward or reverse feathering of the flippers, man- atees can approach or back away from objects. The primary use of flippers is to turn an animal to the right or left on its longitudinal axis. Due to the structure and rigidity of the manatee’s spinal column and, in particular, of its cervical vertebrae, the head cannot be turned to the side. An animai must resort to “coming about” with the aid of its flippers in order to face whatever has drawn its attention. To turn abruptly to the left, for example, a manatee paddles backward with its left flipper while pulling for- ward with the right. Flippers are always feathered to present maximum surface area on the power stroke and minimum surface area on the return. Figure 20 diagrams a forward stroke. The flipper is brought for- ward from the neutral or resting position, its leading (radial) edge exposing the least surface to water resistance. At its fullest exten- sion the flipper is rotated 90 degrees so that its broad palmar surface delivers the stroke. On a backward stroke, the flipper 1s drawn back with its ulnar edge leading, but again rotated so that the palmar surface draws water. Hartman—Manatee in Florida 71 Fic. 20. Lateral and frontal views of flipper positions during suspended resting (A) and during a typical forestroke (B, C, D). Arrows indicate direction of move- ment. Drawings are from an 8-mm film sequence. 72 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 21. Ellipse patterns described by the flippers during various movements as seen from above: A, forward stroke; B, backward stroke; C, coming about. Rec- tangles symbolize flipper cross sections and show orientation of forearm at various points in the orbit; triangles indicate palmar surfaces. Arrows indicate direction of movement. “x” indicates the approximate point of emergence of flipper from the torso. The path of a flipper during either a forward or backward stroke describes a narrow ellipse (Fig. 21). Power strokes are delivered closer to the body than return strokes. The shape of the ellipse varies depending on an individual’s reach, on the strength of the stroke, and on the use to which it is being put. When an animal is coming about, the flippers tend to describe wider, more erratic ovals than during direct forward paddling. Propelled only by their flippers, manatees are capable of surfac- ing perpendicularly from the bottom. This, in fact, is the ordinary mode of ascent for young calves, which rise to the surface at a steep angle by working the flippers synchronously or, as observed by Moore (1956), by alternating strokes. Young calves seem to have a buoyancy problem and must make a considerable effort to rise and dive. Unlike more mature animals, resting calves depend al- most entirely on their flippers for the climb to the surface and return to the bottom. Normally, resting manatees rise and descend in the water in a horizontal position with little motion of tail or flippers. The flippers are occasionally used to aid the lift off from the bottom. G. Bertram and C. Bertram (1964) suggested that resting manatees are able to ascend and descend effortlessly by means of muscular compression and relaxation of the lungs; rising is effected by expanding the air space in the lungs and sinking by compressing it. Hartman—Manatee in Florida 73 Breathing The breathing behavior of manatees is probably a conditioned reflex dating from the day of birth when a mother, calf on back, is reputed to introduce her offspring to a breathing rhythm by dunking it into the water (Moore, 1951a). Except during the social ritual of “kissing,” only one breath is taken during a surfacing. The manatee opens its nostrils for air at the exact moment of reaching the surface and closes them on submersion with equally precise timing. Exhalation is followed instantaneously by inhala- tion. Occasionally, a manatee starts to exhale before its nose breaks water, but no animal was ever seen to blow out air as it submerged. This agrees with the observations of Scholander and Irving (1941), who reported that a large captive manatee always dove with full inspiration. Scholander and Irving also found that at rest an adult manatee exchanges about 50 per cent of the air in its lungs as opposed to 80-90 per cent in the bottle-nosed dolphin and 10-20 per cent in man. When disturbed in the act of breathing, manatees interrupted the respiratory sequence and dove, apparently without filling the lungs with air. In such situations animals were always observed to resurface immediately. From time to time manatees experienced difficulty breathing and surfaced with a snort. Wheezing inhalations often accom- panied the snorts and suggested that the respiratory tracts were partially obstructed. It seemed that the hoarse “blows” were made - in an effort to clear the nasal passages. Snorts were frequently repeated in successive surfacings until the congestion was relieved. During snorts or forceful exhalations, spouts of spray | to 2 meters high were sometimes visible. It appeared that these vapor plumes resulted simply from water being blown off the nostrils. Variation in surfacing behavior.—When resting, idling, or feeding, manatees tended to surface gently and to breathe quietly. The sequence of a typical surfacing from the bottom is portrayed in Fig. 22. Approaching the surface, the manatee elongates its body and elevates its nostrils above the water. Inhalation is accompanied by a slight upthrust of the tail, which arches the back and com- mences the descent while the nose is still exposed. The animal then sinks slowly to the bottom, usually to the very spot from which it 74 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 22. Surfacing sequence from bottom-resting position. Inhalation (top drawing) is accompanied by a slight upstroke of the tail, which arches the back and commences the descent while the nose is still exposed above the surface. Hartman—Manatee in Florida 75 rose. Sometimes a manatee followed a quiet surface with a stretch and/or steep dive before settling to the bottom. The upthrust of the tail has the additional advantage of tilting back the head and raising the nostrils higher above the surface during inhalation. This would reduce the chance of inhaling water, especially under turbulent conditions. Manatees do, in fact, expose more of the head above the water line when surfacing in rough water, as first observed by Moore (1951la). Young calves appear to be positively buoyant at the surface and expose not only the nostrils but the crown of the head and sometimes the back. On occasion, a manatee lifted its entire head out of water during a surfacing. The purpose of this peculiar behavior was not clear, but it was decidedly not a means of surveying the above-water surroundings (see section on sight). Cruising or socially active manatees surfaced in a fashion that bore no resemblance to the surfacing behavior of less active ani- mals. Stages in the surfacing of a cruising manatee are dia- grammed in Fig. 23. As can be seen, a swimming manatee climbs gradually to the surface, momentarily ceases its rhythmic tail strokes, exposes its nostrils, then plunges head-first to lower depths, normally breaking water with its back and tail. When sur- facing, cruising manatees elevated their noses particularly high in order to clear their own wake. The breathing of active manatees was forceful and distinctly audible. Respirations during sexual interactions were especially noisy. Ventilative breathing.—Rhythmic patterns in the breathing of cap- tive manatees and dugongs have been described by Parker (1922), Scholander and Irving (1941), and Kenny (1967). Parker found that upon rising for air after a protracted submergence, manatees would remain near the surface and breathe two to four times in comparatively rapid succession before sinking to the bottom. Scho- lander and Irving noted a similar phenomenon in an adult man- atee whose respiratory frequency increased from less than one breath per minute after submergences of one to two minutes to two to three breaths per minute following submergences of 12 to 15 minutes. In Everglades National Park, Moore (1951a) found that the respiration of an adult manatee was characterized by relatively short “breathing periods” between “long, submerged resting pe- riods.” 76 = Spec. Publ. Amer. Soc. Mamm. 5 ee = “ v Fic. 23. Sequence of a surfacing while cruising. Arrows indicate direction of tail’s movement. The rhythmic respirations that accompany bottom-resting are apparently a means of recovering oxygen depleted by prolonged submergence and of replenishing air in preparation for the next dive. According to Kooyman (1973), it is typical of all slow-breath- ing marine mammals to remain at the surface after a dive and to ventilate several times to eliminate carbon dioxide and to renew oxygen. Figure 24 illustrates the ventilative breathing of manatees ob- served bottom-resting at Crystal River. Generally two to three breaths are taken in less than a minute when ventilating. The data Hartman—Manatee in Florida 77 mumanenen SUSPENDED RESTING SEES IDLING | SURFACING BOTTOM RESTING CRE Re el LL ta "| E Hee 0 10 20 30 40 50 60 70 80 90 100 N10 TIME IN MINUTES Fic. 24. Consecutive submergence periods of four resting manatees: A, male calf approximately 2 meters long; B, juvenile male approximately 2.4 meters long; C, adult female approximately 3 meters long; D, adult male approximately 3.1 meters long. contradict Parker (1922), who concluded that the larger the mana- tee the more breaths it takes while ventilating. Ventilative breathing was observed only when manatees were resting on the bottom or feeding under dense mats of vegetation. Prolonged submergence characterized both activities. Normally manatees did not compress their breathing into discrete bouts; rather they took single breaths at relatively equal intervals over a period of time. Notice, for example, how the respiratory sequence of the calf (A) changes when it stops idling and begins to bottom- rest (Fig. 24). When involved in social interactions, manatees did not have a respiratory rhythm but tended to breathe irregularly whenever their activity brought them to the surface. Synchronous surfacing.—When two or more manatees were resting or swimming together, they usually surfaced synchronously to breathe. According to Jarman (1966), groups of dugongs also break water in unison. Synchronous surfacing also occurs among many species of cetaceans (Slijper, 1962; Caldwell, Caldwell, and Rice, 1966; personal observations). 78 Spec. Publ. Amer. Soc. Mamm. 5 TABLE 8 FREQUENCY AND SYNCHRONIZATION OF BREATHING OF COws AND ACCOMPANYING CALVES WHILE IDLING AND RESTING. Percentage of Mean number Mean number of calfs breaths in Hours of breaths synchronous breaths unison with its Individuals observed per hour per hour mother’s breaths cow I 21.5 d 9 calf I 2 23.5 20). 2 cow II 21.0 . calf II : 23.5 Pe D 92 cow III : 18.7 _ calf II : 31.0 ve = cow IV 26.7 calf IV : 31.0 za 88 cow V 20.0 calf V- ; 30.0 18.0 2 The functional value of synchronous surfacing is conjectural. Among manatees, it appears to be more than a manifestation of social facilitation. It surely does not afford an advantage in pro- curing food nor in discouraging predators. It is possible that sur- facing in unison is an extension of behavior learned at birth when a mother synchronizes her movements with those of her newborn calf or in some manner impresses her offspring to remain at her side. The incidence of synchronous surfacing between mothers and their calves is shown in Table 8. Eighty two to 92 per cent of a calf’s breaths are in unison with the breathing of its mother. Whatever its function or motivation, synchronous surfacing en- sures closer side-by-side contact between animals, decreasing the odds of accidental separation, a constant hazard in turbid water, especially for a calf. Length of respiration.—From studies of captive manatees, Parker (1922) found that larger animals took longer breaths than did smaller manatees. Data collected at Crystal River support this find- ing and indicate that not only a manatee’s size but also its state of activity have direct bearing on the duration of breathing (Table 9). Length of respiration was 0.1 to 0.7 seconds shorter for calves than for juveniles and 0.1 to 0.5 seconds shorter for juveniles than for adults. Moore (1951a) estimated that the average respirations of Hartman—Manatee in Florida 79 TABLE 9 LENGTH OF RESPIRATION (IN SECONDS) IN RELATION TO Bopy SIZE AND STATE OF ACTIVITY. AVERAGES ARE WEIGHTED. “LIGHT ACTIVITY’ INCLUDES RESTING AND IDLING; “MODERATE ACTIVITY” INCLUDES FEEDING, CRUISING, AND PLAYING; “STRENUOUS ACTIVITY” INVOLVES FLEEING. Light activity Moderate activity Strenuous activity Number Number Number of obser- of obser- of obser- Age group vations Mean Range vations Mean Range vations Mean Range pales 46. 25 20-218 14 De 1.922455 49 19 1.4-2.8 (1.8-2.3 meters) : ee: ‘ aS finn i aaa vents: 51 3.2 26-42 21 23 13-26 6 2.0 ‘ (2.3-2.7 meters) it see : Ct : 1.7-2.8 Adult (us meters) 167 3.8 2.5-5.2 74 3.3 2.64.0 9 2.5 2.3-2.7 a resting calf a meter long lasted two seconds, half a second less than the average recorded for calves at Crystal River. The differ- ence can perhaps be explained by the small size of Moore’s animal, almost a meter shorter than the smallest calf encountered at Crystal River. Due to their smaller lung capacity, calves undoubtedly exchange less air when breathing than do larger manatees. When a cow and calf surfaced synchronously, the full breathing cycle of the calf was sometimes completed before its mother had begun to inhale. Table 9 shows that the length of each breath is reduced as a manatee increases its activity. Manatees at rest may take over a second longer to breathe than animals in flight. Length of exhalation and inhalation.—The relation of exhalation and inhalation times to size and state of activity is shown in Table 10. The data indicate that it takes adult manatees roughly 0.5 sec- onds longer than calves to exhale and 0.5 seconds longer than calves to inhale. The exhalation and inhalation times of juveniles fall midway between those for calves and adults. Analyses of vari- ance (significant at the .10 level) suggested that the breathing of active manatees is shortened as a result of quick, forceful exhala- tions and that the length of inhalation is little affected by an increase in activity. Exhalation occupies approximately 55 per cent of total breathing time when manatees are at rest but only 43 per cent when they are in flight. 80 Spec. Publ. Amer. Soc. Mamm. 5 TABLE 10 LENGTH OF EXHALATION AND INHALATION (IN SECONDS) IN RELATION TO Bopy SIZE AND STATE OF ACTIVITY. AVERAGES ARE WEIGHTED. “LIGHT ACTIVITY” INCLUDES RESTING, IDLING, AND FEEDING; “MODERATE ACTIVITY” INCLUDES CRUIS- ING AND PLAYING; “STRENUOUS ACTIVITY” INVOLVES FLEEING. Light activity Moderate activity Strenuous activity Number Number Number of obser- of obser- of obser- Age group vations Mean Range vations Mean Range _ vations Mean Range Exhalation Calves Mineeotaimeten) 20 13 0.9-1.8 10 10 0.7-1.6 2 0.7. -0.6-0.7 Juveniles (2.3-2.7 meters) 19 1.9 1.5-2.3 14 1.5 1.1-2.0 5 12 0.8-1.5 Adults (emotes) 65 2.1 1.3-3.0 14 1.8 1.3-2.1 4 1.2 1.0-1.3 Inhalation Calves (eels meters) 14 10 0.5-1.3 8 0.9 0.6-1.3 4 0.9 0.6-1.1 Juveniles (2.3-2.7 meters) 19 15 0.9-2.1 16 1.2 0.8-1.9 3 1.3 0.8-1.5 Adults 65 1.6 1.0-2.3 12 1.5 1.0-2.2 3 1.5 1.4-1.6 (2.7-3.1 meters) Intervals between breaths. —According to Parker (1922), large man- atees can remain submerged longer than small animals. The av- erage submergence period of a captive calf less than 2 meters long was approximately four and a half minutes as compared to 12 minutes for an adult 3 meters long. Data from the present study corroborate these findings but indicate that the size of a manatee is related to the length of time it remains submerged only when comparing calves with older animals (Table 11). Juveniles do not appear to remain submerged for briefer periods than adults, at least while bottom-resting or idling. There is evidence that the submergence periods of very young calves are exceptionally short. In Everglades National Park, Moore (1951a) observed a resting calf approximately 1 meter in length and found that the intervals between its breaths averaged less than a minute. This compares with an average submergence period of 3.4 minutes recorded at Crystal River for resting calves, the small- est of which was nearly 2 meters long. D. Caldwell (1955) found that the frequency of breathing in the 81 Hartman—Manatee in Florida (s1ayour [°g—-1"3) 00¢6-6F rae TZ 8EE-ES 091 #9 69-61 0¢ IF OLF-19 G6 88 synpy = = e = (stayau 1°S-§°3) PEI-68 FOr al 866-61 S8I 6I 6S-F1 96 $E& 069-8 &1€ 6¢ saquaan{ (stayaut €°Z-8"T) OII-é& 88 9¢ 096-2 FIT 18 Gr-6 06 LI SCE-F8 106 8F OA 1120) asury uray suonea adury uray suonea asuey uray suonea asuey uray suonea dnoi3 a3y -198qO Jo -198qO Jo -19sqo jo -19Sqo jo Joaquin Jaquinyy Jaquinn Joaquin AWANR aIVIZpO| Aqanoe Wwysrq papuadsng wi01j0q UG Bunsay ' “ONISINAD) SAATOANT (ALIAILOY ALVAAGOP,, SONILOOY ANV ‘ONIGAAT ‘ONITG] SAGNTONT ALIAILOY LHOIT,, ‘ALHOIAM ALY SAQVAAAY “ALIAILOY AO ALVLS GNV AZIS AGOG OL NOILV1AY NI (SANOOAS NI) ONADAAWANS JO HLONAT Il ATaVL 82 Spec. Publ. Amer. Soc. Mamm. 5 spotted porpoise (Stenella plagiodon) increased as its activity in- creased. It appears that the same is true of sirenians. Moore (195 1a) clocked the intervals between breaths of a young manatee during a 10-minute pursuit by speedboat and found that the ani- mal was submerged for substantially shorter times than those re- corded for an adult that was resting on the bottom. Jarman (1966) found that dugongs surface every three minutes when cruising in contrast to every five to 10 minutes when feeding. A direct cor- relation between length of submergence and state of activity was also observed among manatees at Crystal River (Table 11). The data indicate that manatees remain submerged for progressively shorter periods as their activity becomes more strenuous. For lack of sufficient data, submergence periods during flight or at the culmination of intense social interactions are omitted from the ta- ble. It is fair to predict, however, that intervals between breaths during these most strenuous of activities would be even briefer than those recorded for cruising manatees. It seems that manatees also increase respiration in response to cold. In the Alafia River (Hillsborough County), bottom-resting animals were observed to breathe every minute or thereabouts in water 16 to 19°C—four to five times as often as their Crystal River counterparts, which rest in water 22 to 24°C. I am tempted to conclude that the reduced submergence intervals of the manatees in the Alafia River were a reflection of increased oxygen expen- diture in the face of cold stress. The longest submergence time recorded during the study was 11 minutes 46 seconds for a juvenile male resting on the bottom. The briefest intervals between breaths occurred when manatees were ventilating after a protracted dive (Fig. 24) and while engaged in “kissing” at the surface (Fig. 32). I have seen manatees breathing as often as eight times per minute while kissing but feel that this behavior should be divorced from the present discussion, inas- much as I consider that respiration during kissing is an involuntary response not directly related to metabolic needs. Resting Two basic resting postures were employed by manatees observed in this study: 1) hanging suspended near the surface and 2) lying prostrate on the bottom. In both positions, the animals lapsed into Hartman—Manatee in Florida 83 sf wef Fic. 25. Suspended-resting postures (drawn from photographs). Above, adult male with extruded penis. Below, preyearling calf beside its mother. a somnolent state with their eyes closed and their bodies motion- less. The eyes were opened only to surface. Manatees allow the tail and flippers to dangle in typical sus- pended resting (Fig. 25). Animals usually adopted this posture just below the surface but occasionally broke water with their backs. While lolling at the surface, animals were sometimes seen to roll on their backs to bask for a moment with chest and flippers pro- jecting above the water. Young calves evidently lack proper ballast and have a unique suspended resting posture that suggests the center of gravity is shifted cephalad (Fig. 25). Moore (195la) observed a meter-long 84 Spec. Publ. Amer. Soc. Mamm. 5 calf that appeared to have positive buoyancy and difficulty main- taining its equilibrium. When resting suspended, manatees are liable to drift and must regularly compensate for their displacement by making minor cor- rective movements with tail and flippers. Turbulent water tended to discourage suspended resting, the animals opting for the bottom where conditions were more placid. Suspended resting was frequently interspersed with idling and was a more transitory activity than bottom-resting. It sometimes served as a substitute for bottom-resting on cold days when the upper layers of water were warmer than those at the bottom. Sus- pended resting was often a prelude to bottom-resting. A special case of suspended resting is found among bottom-resting animals when they take ventilative breaths at the surface between rests (see section on breathing). When resting on the bottom, manatees are supported on the muzzle, stomach, and tail (Fig. 22). The flippers are held along the sides or flexed at the wrists and adducted under the chest. The tail usually rests flat on the bottom but may be curled under at its tip. It was common for bottom-resting manatees to roll on their sides or backs momentarily. One juvenile male was seen to lie immobile on his back for periods exceeding a minute. When alerted by a disturbance, bottom-resting manatees poised themselves on the tips of their flippers, ready to shove off at the least provocation. Animals were seen to bottom-rest, apparently without prefer- ence, on a variety of substrates (including sand, mud, limestone, oyster bars, and beds of aquatic plants). I have even observed them “asleep” burrowed into clumps of dense vegetation. Resting calves sometimes lay on the backs or tails of their mothers. I once ob- served a bull resting on a limestone ledge supported by his muzzle and tail with his midsection suspended over a wide crevice. The swollen bellies of cows in advanced pregnancy forced them to rest at an angle to the bottom so that their heads or tails were off the substrate. One gravid female was seen resting on a clump of Ay- drilla tilted forward at a 45-degree angle to the bottom. Manatees generally chose shoals of 1 to 3 meters for bottom- resting, but occasionally rested at greater depths. At the site of a large spring in one canal, they bottom-rested at a depth of 4 me- ters. In the vicinity of the Main Spring, animals showed a predi- lection for two sandy areas kept free of vegetation by their activity. Hartman—Manatee in Florida 85 Feeding Use of the mouth.—The upper lips of a manatee are bilobed and covered with stiff bristles. Their use in feeding has been described by Chapman (1875), Garrod (1879), Mohr (1957), and Allsopp (1961). The lobes are everted, projecting the bristles into the food source, then closed laterally, forcing the bristles to grasp the vege- tation, tuck it into the cleft between the lobes and funnel it to the mouth. The process is rapidly repeated and suggests fanning (Chapman, 1875). Manatees chew incessantly while they manipulate their lip pads. The sound of their teeth grinding is audible underwater. The . masticatory rate is approximately two chews per second. Feeding animals masticate as they rise to surface, pause to breathe, then resume chewing on the descent. Occasionally at Crystal River a manatee would begin its ascent to breathe by jerking its head back to break rooted plant stalks in its mouth. Use of the flippers.—Use of the flippers to facilitate feeding de- pends upon the type of vegetation being consumed. In the St. Johns River where water hyacinth is the principal food, manatees were observed to pull the plants below the surface with their lips, then to hold the hyacinth in front of their mouths with the tips of their flippers. At Crystal River where the preferred food is sub- merged and rooted, manatees did not use their flippers to convey food to their mouths or to assist ingestion in any way. Feeding animals employed their flippers solely to position themselves and to sweep encumbering vegetation away from their heads. Manatees normally raise themselves on their flippers to feed on bank growth. Bangs (1895) wrote of animals dragging their bodies partially out of water with the flippers to reach mangrove leaves. At the Crandon Park Zoo (Dade County), manatees have been seen to raise their heads and shoulders above the surface to crop lawn grass (Gordon Hubble, personal communication). Many interview- ees in Florida had seen manatees feeding on the banks of rivers with heads and backs awash and supported on the flippers. In Guyana, manatees regularly graze on phraetophytes with their bodies half out of water and their flippers exposed on the mud (Datakaran Jeetlall, personal communication). Gohar (1957) reported that for feeding purposes dugongs use their flippers to dig out and collect seagrasses into little piles on 86 Spec. Publ. Amer. Soc. Mamm. 5 the bottom, but I observed no evidence of comparable behavior among manatees. Comfort Activities Stretching.—The bodies of manatees are remarkably supple, as reflected in their stretch postures (Fig. 26). In one, the back is arched concavely and in another, convexly. Variant postures com- bining elements of both the convex and concave arches also are illustrated. In all stretch postures the eyes were closed, the chin drawn in on the chest, and the flippers adducted tightly across the chest. Animals usually uttered a prolonged groan while stretching. Stretching was closely associated with resting. Resting sessions were regularly interrupted or terminated by stretches. Animals stretched on the bottom or on rising to or descending from the surface. Manatees usually arched their backs convexly then con- cavely in sequence. A stereotyped stretch-roll often followed the stretching se- quence. The animal rolled on its back in a concave arch, then backdived to the bottom, often plunging headfirst into the sand or silt where it lay momentarily, its muzzle buried, in a partial head- stand or upsidedown. While arching their backs convexly, males sometimes extruded the penis. Doubled over thus, they occasionally stroked the penis with one or both flippers in what looked to be incipient mastur- bation. Stroking, however, was awkward and unrhythmic and rare- ly lasted more than a few seconds. Ejaculation was never observed. Extrusion and retraction of the penis was often repeated several times in succession. Occasionally, males swam about and rested with the penis extruded (Fig. 25). In one instance a bull’s penis, extruded during a stretch, remained tumescent and pulsing for nearly five minutes of suspended resting and was only retracted when the animal finally sank to the bottom. In another case a young male maintained an erect penis for 16 minutes during which time he alternately bottom-rested and courted passing cows. The erect penis was always directed craniad and did not deviate from the longitudinal axis of the body. Scratching.—Manatees make use of their flippers to scratch but are restricted by the limited reach of the appendages and must confine scratching to the chest, ventral neck, and head regions. Hartman—Manatee in Florida 87 Fic. 26. Various stretch postures. 88 Spec. Publ. Amer. Soc. Mamm. 5 Irritations on the remainder of the body are relieved by rubbing on objects. The manatee operates its flippers singly or in unison when scratching. The radial edge of the flipper is normally used to rub the creases under the chin and the tip and palmar surface to rub the head, chest, opposite flipper, and axil. I once saw a bull use the tip of his right flipper to scratch at a hook caught in his left flipper. Scratching tended to be concentrated in the axils of the flippers, in the folds under the chin, and in the orbits of the eyes (Fig. 19D). Microorganisms may have been responsible for the itch; epidermal scrapings from afflicted areas yielded microscopic nematodes. In Kings Bay, manatees regularly rubbed on logs, poles, cement blocks, and limestone outcrops. Sandy areas bared by manatee ac- tivity surrounded these objects. A broken milk bottle firmly im- planted in the sand was used for rubbing and may have cut the bellies of some animals. Local crabbers complained that manatees rubbed on their crab traps, turning them over or imbedding them in the mud. Manatees also rubbed on buoy lines and anchor ropes, making repeated passes that recalled a cat brushing against a person’s leg. Once, several manatees rubbed along the string suspending a ther- mometer hung in their midst. One especially tame young cow used to swim up to my boat to nibble and rub her head on the anchor, anchor rope, hull, and propeller blades (Fig. 27). At times she grasped the anchor rope to her chest and rubbed awkwardly on it (Fig. 191). At a yacht basin in Palm Beach County, manatees have been seen to rub their backs on the hulls of boats (to remove bar- nacles, it was claimed by those who made the observation). The limestone ledges at the Main Spring were a favorite rubbing site for some animals. One mature bull was particularly fond of rubbing there. He was once seen to spend an hour rubbing on the rock, slamming into it, rebounding, tumbling, turning, and twist- ing to make contact with all parts of his body, and working himself into a ferment. On another occasion, a young cow rolled sideways, wedged herself in a limestone crevice, and rubbed back and forth on her side. A log below the Main Spring was another popular rubbing site. Manatees gathered here to rub, wallow in the sand, and scratch on a cement block nearby. Usually the animals dragged their bellies Hartman—Manatee in Florida 89 Fic. 27. Juvenile female nuzzling propeller and hull of runabout (drawn from photographs). and tails to and fro across the log. Among males there was one instance of possible masturbation during such behavior. A juvenile bull with penis extruded courted a cow near the log but was thwart- ed; in an apparent displacement reaction, the young male rubbed 90 Spec. Publ. Amer. Soc. Mamm. 4 his genital region briefly against the wood. One bull stroked the surface of his tail to the right and left across the log for more than five minutes, then rolled over and repeated the act. At the cement block manatees rubbed their noses, chins, and orbits. Bottom-rest- ing animals occasionally rubbed their muzzles in the sand and rolled on their sides to rub their heads. A young male once lay on his back rubbing in the sand for almost a minute. An adult female was once discovered standing on her head at the foot of an iron signpost, rubbing the underside of her tail on the pole. Other animals were frequently observed rubbing their bellies, sides, backs, heads, and tails at the base of the same pole. Rust smears on the hide were a telltale indication that an individual had recently engaged in this activity. Mouth cleaning.—Prominent among the self-care activities of the manatee is mouth cleaning. The animal opens its mouth wide, wrinkles back its nose, and everts the lobes of its upper lips in an apparent effort to free irritating particles of vegetation lodged in the mouth (Fig. 28). Sometimes the mouth was closed so that the lower lip protruded over the upper lobes. If this manipulation failed to loosen entrapped particles, the animals resorted to the fanning, masticatory movements of feeding, and/or to rubbing the tip of a flipper back and forth across the gums. The flipper on the side of the irritation was usually used, although on rare occasions, both flippers were employed simultaneously. if the annoyance per- sisted, the manatee was apt to rub frantically and even swat its lip pads (Fig. 19C). I once watched a bull become almost frenzied in its attempts to extricate plant debris from its mouth. Mouth cleaning took place at any time, interrupting any activity. It was not more prevalent during or after feeding sessions and was observed in calves as well as adults. Mouth cleaning was generally engaged in for only a few seconds, although occasionally it occu- pied an animal for several minutes. Often during the cleaning process broken bits of vegetation could be seen falling from the mouth. In behavior evidently unrelated to mouth cleaning, manatees sometimes nibbled the tips of their flippers in a manner suggesting that the irritation was on the manus. Rooting.—Manatees sometimes paused to root in the sand or mud while reconnoitering the bottom. They appeared to “chew” the substrate, drawing a mixture of sand and water into their Hartman—Manatee in Florida 91 Fic. 28. Mouth cleaning. Above, adult male everting lips; note bristles. Below, adult female rubbing mouth parts with her flipper (photos by Russ Kinne and James Powell, Jr.). 92 = Spec. Publ. Amer. Soc. Mamm. 5 mouths, then expelling it. The activity, audible underwater and reminiscent of feeding, sounded as if sand was being sifted through the mouth. In the process, the flippers made abortive movements to push sand away from the mouth. The flippers were sometimes flailed about so violently as to stir up clouds of sand around the head. Eventually the action of mouth and flippers scooped a pit out of the sand. A juvenile cow was once observed “standing” on her head while she rooted in a pit; her muzzle was buried up to the eyes in the loosened sand. After surfacing she would return to the same depression and start rooting again. This continued for 20 minutes. On another occasion I observed a calf rooting for more than an hour. One juvenile cow that was rooting along the sand collided head on with a bull that was rooting in the opposite direction. Both then wallowed in the silt they had kicked up. Shortly thereafter, the cow wedged herself into a wide limestone crevice and rooted in the sand at its base. The function of rooting is not clear. It may be a means of clean- ing the mouth or perhaps simply affords a pleasurable sensation. If silt and sand are ingested, rooting may provide manatees with ballast or with needed minerals. Sneezing.—Twice in the course of the study manatees appeared to sneeze. In each instance the animal opened its mouth and re- coiled, exploding a mass of bubbles from the nostrils and startling its companions. The sneezes were followed at once by a surfacing. Eliminative Behavior Micturition and defecation.—Manatees have no special postures or behavior associated with urination and defecation. The elimination of waste took place at any time, even while resting. The frequency of urination was not determined. Defecation was practically con- tinuous. Feces are passed slowly and often trail partially extruded from the anal aperture. Manatee feces are soft, mealy, ochroid cylinders that are some- times grooved. They are tapered at the ends if not broken (Fig. 29). One fecal specimen was 28 centimeters long, an unusual length because the cylinders usually break on discharge before Hartman—Manatee in Florida 93 Fic. 29. Manatee fecal specimens. reaching that size. The diameter of most feces is 5 to 7 centimeters, the maximum diameter observed being 7.3 centimeters. Manatee excrement was not consumed by fish or other animals and lay on the bottom until it disintegrated, usually within a day or two. Flatus.—The herbivorous diet of manatees generates consider- able gas in the alimentary tract, and the animals were constantly relieving flatulence. Apparently gas is never expelled through the mouth. Regurgitation.—Manatees periodically regurgitated a yellow- green mash composed of partially digested plant material. Animals often chewed while the mash was being brought up and passed through the mouth. Neither eructation nor visible stomach con- tractions were seen to accompany regurgitation. Calves regurgi- tated as well as adults. Regurgitation occured preponderantly during resting, at least 94 Spec. Publ. Amer. Soc. Maram. 5 200 175 150 125 SIGHTINGS 100 OF 75 NUMBER 50 25 1 2 3 4 5 6 / 8 9 10 NUMBER OF MANATEES AT EACH SIGHTING Fic. 30. Frequency distribution of sightings of single and various-sized groups of manatees (a cow with calf is considered a single animal). Data are from 335 aerial sightings outside the cold-induced congregations. Hartman—Manatee in Florida 95 an hour after the last feeding session. Once, a juvenile cow regur- gitated after ingesting manatee excrement, but there appeared to be no correlation between the two acts. SOCIAL BEHAVIOR HE manatee is a mildly social, essentially solitary animal. Results a: of tracking manatees on their daily rounds indicated that each animal was fully independent and generally passed its time alone. Outside the cold-induced congregations, 53 per cent of all manatee sightings from the air were of solitary individuals; groups of two, three, and four were seen 17 per cent, 13 per cent, and 6 per cent of the time, respectively (Fig. 30). These statistics are contradicted by more recent data from aerial surveys conducted by the National Fish and Wildlife Laboratory. According to How- ard Campbell, leader of the project, groups of two or more man- atees were seen more often than single animals, suggesting that manatees are not “essentially solitary.” I found the only firm association among manatees to be that between a cow and its calf, which comprises the manatee family unit. All other associations, (with the exception of the estrous herd) were casual, temporary groupings. Groups consisted randomly of juveniles and adults of both sexes. Animals grouped to cavort, migrate, rest, or feed, but the social bond was highly unstable, sometimes lasting only minutes. Few animals were known to have remained together so long as a day. Bulls were more active socially than cows and were responsible for most intraspecific contact. Cows were generally passive, limiting their social overtures to nonsexual situations. Gravid cows were exceptionally sluggish and generally nonsocial. Resting manatees that preferred to be left undisturbed rolled or swam away from solicitous companions. Bottom-resting animals occasionally cuffed intruders with their flippers. Social interactions occurred in both sexual and nonsexual con- texts. The distinction was not sharply defined because the patterns of all intraspecific contact were basically the same. These included mouthing, nuzzling, nudging, and embracing. There were no dis- plays. The preponderance of mouthing (here defined as any caress with the mouth or lip pads) suggested that manatees possess a 96 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 31. Juvenile male mouthing and embracing peer. Adult female in fore- ground (photo by James A. Sugar, (c) National Geographic Society). chemoreceptive sense (see section on taste). Mouthing was concen- trated on the backs and sides of a recipient and was often accom- panied by an embrace with the flippers (Fig. 31). Nuzzling and nudging were minor variations of mouthing. Muzzle to muzzle contact was generally restricted to stereotyped “kisses” at the sur- face (Fig. 32). For lack of a better understanding of their function, kissing and related forms of nonsexual interaction are discussed beyond in the section on play. Manatees exhibited social facilitation when in groups. Responses were mutually stimulating. If one animal fed or rested, its associ- ates often did likewise. The herd instinct, however, appeared to be vestigial. Flight reactions, communicated vocally, generally pro- duced a response from other manatees in the vicinity, but the vocalizations of an alarmed animal seemed less group-oriented than simply a reflection of its own emotional state. At the approach of potential danger, squeals that might be interpreted as warnings or alerts were not always emitted. The only predictable alarm re- sponse was between a cow and her calf and this rarely attracted the attention of other manatees in the area. Hartman—Manatee in Florida 97 Fic. 32. Two juvenile females “kissing” at the surface (drawn from photograph). Cruising manatees, furthermore, did not maintain formations. Animals sometimes swam temporarily in eschelons, abreast of one another or in tandem, but these formations did not appear to be intentional. As far as is known, manatees do not assist distressed or wounded individuals. Within groups at Crystal River, there was no evidence of communal defense or mutual aid. The behavior of a mother toward her calf may be an exception. Frank Rivell, an old guide on the Tomoka River (Volusia County), told me of having seen a cow butt her dead calf for two days in an apparent effort to keep it above the surface. I never witnessed the reaction of manatees to an afflicted com- panion although Steller (1751) wrote of harpooned sea cows being accompanied in their struggles by fellow animals. During the pres- ent study a local resident fishing in Kings Bay inadvertently hooked a manatee. He told me that a second individual remained close by the hooked animal until the line was cut. In the absence of more conclusive proof, this activity is, to me, as much an indication of curiosity as it is of succorant behavior. Aggression in manatees was confined to collisions between bulls for a position next to an estrous female. Agonistic contact was never observed outside the estrous herd, but there were occasional hints of an age-dependent hierarchy. For example, on one occasion a juvenile male that I was scratching seemed to flee at the approach 98 Spec. Publ. Amer. Soc. Mamm. 5 of an adult bull. Normally, however, animals seeking my attention underwater never competed for priority. At rubbing stations, an- imals of the same size tolerated one another, but subadults usually yielded to adults. There appears to be no territoriality among manatees. I obtained no evidence of an animal defending or guarding an area or ex- hibiting agression toward conspecifics in a territorial context. My observations support Moore’s (1956) conclusion that manatees have independent, overlapping, often identical ranges. Vocalizations Manatee vocalizations were first recorded by Schevill and Wat- kins (1965), who found most of the calls to be squeaky and ragged, lasting 0.15 to 0.5 seconds and composed of two or more frequen- cies that were not harmonically related. Most fundamental tones were between 2.5 and 5 kHz. Evans and Herald (1970) found that the sounds produced by a young male Amazonian manatee were basically similar, but of higher fundamental frequency (6 to 8 kHz). Manatee calls heard underwater at Crystal River consisted mostly of high-pitched squeals, chirp-squeaks, and screams. These noises were produced with mouth and nostrils closed and with no accom- panying escape of air. According to Phillips (1964), manatees also vocalize out of water; a captive female squeaked as she was being hoisted from one tank to another at the Miami Seaquarium. Manatees were normally silent at Crystal River, even in groups. Hours of recordings with a hydrophone gave the impression that manatees have an impoverished vocabulary. They seemed to con- vey information by varying the intensity and duration of their calls rather than by emitting mood-specific, harmonically distinct noises. They apparently make sounds only under conditions of fear, ag- gravation, protest, internal conflict, male sexual arousal, and play. Unlike cetacean phonations, manatee vocalizations appear to be nonnavigational; to lack ultrasonic signals, pulsed emissions, or directional sound fields; and to be more impulsive than commu- nicative. The relative importance of vocalizations in intraspecific com- munication is unclear. To what extent, for example, are individual vocalizations a monologue or a dialogue? The human ear cannot Hartman—Manatee in Florida 99 pinpoint the source of a phonation, so in most social interactions it was impossible to attribute a sound to a specific animal. In some cases, however, I was able to identify the individuals producing sounds and to speculate on their function. Frightened animals generally screamed in alarm. I gained the impression that juvenile manatees vocalized more often than adults under conditions of stress, perhaps because juveniles still retain the tendency, acquired as calves, to call repeatedly for mother (see below). Older manatees were rarely heard to emit an alarm call. Occasionally bottom-resting animals squealed in apparent irri- tation when disturbed by other manatees. Cows often squealed in seeming annoyance when embraced by bulls. Bulls thwarting the homosexual overtures of other males squealed in what appeared to be vexation. At the height of sexual frenzy in an estrous herd, vocalizations were almost constant from cow and bulls alike. It is possible that the bulls’ squeals were rivalry signals emitted in a competitive con- text. Bulls also squeaked in homosexual embraces, appearing to reinforce the abnormal pair bond. Manatees also vocalized during play. I presume that many of the calls associated with this activity were expressions of pleasure and inducements to further contact. Manatees occasionally emitted sounds when alone. From time to time animals chirp-squeaked and squealed in apparent pleasure while rubbing on logs or other objects. A few squealed when I caressed their backs. One animal emitted a single, high, almost inaudible squeak while being scratched. Manatees generally inves- tigated divers with caution, sometimes emitting “snort chirps” or squeals that I interpreted to be displacement vocalizations result- ing from approach-avoidance conflicts. In addition to squeaking and squealing, manatees groaned, par- ticularly while stretching. Bulls interspersed groans with squeaks during collisions in pursuit of estrous cows. Animals that ap- proached me to be caressed emitted soft, barely audible grunts. The only predictable vocal exchange between manatees was the alarm duet between a cow and her calf. The appearance of a boat or divers ordinarily threatened a cow with offspring. She would react by calling the calf to her side with repeated screams. Each scream triggered an immediate reply from the calf. Once together, the pair usually fell silent and investigated the source of danger. 100 = Spec. Publ. Amer. Soc. Mamm. 4 The cows resumed vocalizations only as they prepared to flee, pre- sumably to insure that their calves remained close beside them. Recordings indicated that calves called more often than their mothers. Calves vocalized even when undisturbed and in close proximity to their mothers. The cows usually responded unless resting. The distress squeals of startled calves always brought their mothers to investigate. Mothers and calves also appeared to in- crease vocalizations in turbid water. I was impressed by the fre- quency of sound emission whenever I dove with cows and their young in water in which visibility was a few meters or less. This suggests that animals maintain contact under conditions of im- paired vision by becoming more vocal. I believe that manatees are able to recognize one another by sound—that is, by subtle variations in the pitch, frequency, and timbre of individual vocalizations. It was obvious, for example, that cows were selectively responsive to the calls of their calves and vice versa. Bottle-nosed porpoises have been found to discriminate be- tween the whistles of various conspecifics, and sonagraphic analysis of the calls has revealed minor differences unique to the individual (Caldwell et al., 1972). It seems reasonable to assume similar “voice prints” characterize individual manatees. Sexual Behavior Estrous herd.—The only cohesive association between manatees, besides the cow-calf family unit, is found in the estrous herd com- posed of a cow in heat accompanied by courting bulls. These groups may remain together for periods ranging from a week to more than a month. As many as 17 bulls once were observed fol- lowing an estrous cow—virtually the entire male population of Kings Bay at the time. Juvenile males joined and left the herd constantly, but a nucleus of mature bulls was always present and in persistent pursuit of the cow. The urgency of male courtship within an estrous herd waned when the cow was not in motion—when feeding or bottom-resting, for example. At such times, bulls themselves were induced to feed or rest, only resuming their sexual overtures when the cow was again active. Courting bulls had a tendency to huddle around an estrous cow that was bottom-resting and to face in the same direc- tion as she. Hartman—Manatee in Florida 101 Fic. 33. Courtship activity. Three adult males mouthing and embracing estrous female (drawn from photograph). For the most part, the courtship of bulls was relentless, forcing the cow into continuous flight. The bulls mouthed and embraced her back (Fig. 33), rode her when she surfaced (Figs. 34 and 35), and rolled upside down in an effort to approach her abdomen. The cow invariably rolled away from them, presenting her back to their advances. If a bull succeeded in nibbling or grasping her underside, she escaped with a violent jackknife, which usually in- cited her escorts to more frenzied embraces. I have seen bulls knocked to the surface by the powerful thrusts of a cow’s tail. I myself was once thrown clear of the water when a cow directly below me jackknifed in response to a bull mouthing her genital region. Sexually aroused bulls sometimes left the herd momentar- ily to court passing anestrous cows or to transfer their sexual drive to homosexual encounters. At the peak of courtship, estrous herds are frequently observed in shallow water a meter or less in depth. It is a popular conception in Florida that under these circumstances manatees are mating. It appears that, to the contrary, estrous cows seek shoals to evade their male escorts. A cow 1s less vulnerable in shallow water because it is more difficult for bulls to approach her underside. I have reason to believe that cows in estrus may even resort to stranding 102 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 34. Male riding back of female during surfacing (photo by Russ Kinne). in an effort to escape the unrelenting pursuit of the bulls. Once, during frenetic sexual activity in shallows off a sand bar in Kings Bay, a member of an estrous herd beached momentarily on the bar with its back awash. I was unable to identify the beached ani- mal, but the following incidents lead me to think it may well have been the cow. From the air my pilot and assistant saw an adult manatee beached in 20 centimeters of water on the shore of a spoil island in the Indian River (Brevard County). In the shallow water beside the stranded animal were at least five other manatees milling about. There was little question that the group was an estrous herd, and it appeared that the beached manatee was the female driven in to shore by the accompanying bulls. As the plane circled, the stranded manatee lifted itself on its flippers, floundered around, reversed direction, and headed toward the water. The New Smyr- na Beach News (13 September 1973) reported the stranding of a mantee on an outgoing tide and its rescue by local residents. The Marine Patrol Officer who supervised the rescue operation said the beached animal was an adult female and that 10 to 15 manatees had been seen beyond the surf just before the stranding. To my knowledge, copulation never took place in shoals. I have been reliably informed, however, that in Belize a pair of copulating manatees was found awash on their sides in knee-deep water (Colin Bertram, personal communication). The one copulation I observed Hartman—Manatee in Florida 103 Fic. 35. Male riding back of female during surfacing. This posture is sometimes assumed by two males engaged in homosexual activity. occurred near the bottom in water 2.5 meters deep. Immediately before this observation, the cow in question was sexually active over a “deep” and may have mated at a depth of 4—5 meters. Her activity at that time was obscured by silt. The parallels between the mating behavior of manatees and el- ephants (Loxodonta africana and Elephas maximus) are striking and possibly reflect their common origin. Elephant cows are promis- cuous and cow-bull associations are transitory (Buss and Smith, 1966; Elapata, 1969; McKay, 1973). According to Short (1966), intervals of sexual inactivity are intermixed with periods of intense activity when as many as four matings may take place within two hours. Manatees exhibit similar sexual behavior; while receptive, estrous cows are polyandrous. I observed copulatory behavior on 7 December 1968. For 45 minutes a female, pursued by five bulls (three adults and two ju- veniles), engaged in intense reproductive activity in a 5-meter 104 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 36. Copulatory embrace. “deep” near the Main Spring in Kings Bay. Although interactions were obscured by the roiled water, mating probably took place. When the herd finally emerged in clear water, the cow paused in flight to allow successive copulations by two of the adult and one of the juvenile bulls. The nearest male would roll on his back, swim up from the rear, mouth the genital area of the cow, then slide fully under her, firmly grasping her from beneath, extruding his penis and effecting intromission (Fig. 36). Pelvic thrusts were not evident. During copulation, the other bulls mouthed and embraced the cow from above and from the side. The cow was passive throughout and sank to the bottom in each embrace so that the males lay on their backs while mating. Between copulations the cow swam short distances before again pausing. Intervals between matings lasted less than a minute. The copulatory embraces lasted 15 to 30 seconds. Postcopulatory behavior among the bulls was identical to precopulatory behavior. They continued to embrace and pursue the cow. Within 10 minutes of the final mating, the cow made vigorous but unsuccessful attempt to evade her escorts. In her efforts to escape, she plunged headlong into clumps of Hydrilla. The males continued to pursue the female in the same area of the bay for most of the day, and the herd was there again the next morning. With the exception of the above instance, estrous cows were never seen to make receptive gestures to their male escorts. It Hartman—Manatee in Florida 105 appears, therefore, that a particular stage of the estrous cycle is responsible for a cow’s receptivity and that the intensity of male courtship is of lesser importance. Perhaps cows are receptive only during estrus proper but engage the attention of bulls during the other phases of the cycle. Among elephants, multiple copulations suggested to Buss and Smith (1966) that proboscideans may have successive periods of estrus leading to conception. Like the ele- phant, the manatee may be a monovular, polyestrous species that undergoes a number of sterile estrous cycles before conceiving (Short, 1966). The fact that estrous herds sometimes involve cows with calves supports this hypothesis, assuming that ovulation does not occur during lactation. The actual duration of estrus in man- atees is unknown. Male aggression.—Male competition for the nearest position to an estrous cow appears to be the sole cause of aggression among man- atees. As courtship activity intensified in an estrous herd, the ma- ture bulls maintained closer station to the cow and, rolling on their sides, collided belly-to-belly with one another, briefly embraced, then thrashed apart in violent contortions. Bulls turning to em- brace the cow sometimes collided sideways with oncoming rivals. Squeaks and groans accompanied the collisions. As the pitch of sexual activity heightened, the frequency of col- lisions increased. In their excitement, bulls often tore through sub- merged vegetation, causing mats of uprooted plants to float to the surface. Competition eventually reached a peak, irrespective of the cows receptivity, then gradually subsided. Juvenile males tended to remain at the periphery of the herd and rarely competed with older bulls. Collisions never resulted in injury. In the long run, the strongest, most aggressive bull is presumably closest to the cow during her receptive period and thus first to inseminate her. Clashes between bulls over anestrous cows were short-lived and infrequent. Social interactions outside the estrous herd.—Sexual interactions be- tween male and female manatees were not limited to the estrous herd. In fact, sexual encounters outside estrous herds were habit- ual. These encounters also appear to be characteristic of dugong- ids. Describing social interactions of the now-extinct northern sea cow, Steller wrote that the females turned away from the males and fled from their incessant pursuit (translation in Goodwin, 1946). 106 Spec. Publ. Amer. Soc. Mamm. 5 Sexual encounters among manatees were invariably initiated by bulls, both juveniles and adults, but always thwarted by cows. Bulls appear to have strong sexual appetites and spent much of their time courting any anestrous cows fortuitously encountered. The bulls were indescriminate in their overtures and mouthed and em- braced subadult and adult females as well as cows with calves. Bulls often devoted attention to several cows at a time. Sometimes several bulls converged simultaneously on a single cow. Anestrous cows rebuffed and fled from all sexual advances. Like estrous females, they were particularly defensive of the venter and presented the dorsum to every male overture. This behavior was first described by Moore (1956), who wrote of an adult female that predictably turned her genital aperture 180 degrees away from an aggressive male. In response to a male rolling fully on his back to mouth her underside, a cow would roll upside down, discouraging a belly-to- belly embrace. Once, a bull, nudging a cow that had rolled on her side to avoid his embrace, pushed her into a limestone ledge. The majority of bull-cow encounters outside of the estrous herd were casual and brief. On a few occasions, however, sexually aroused males persisted in their overtures for hours and even drove cows into hiding. I once watched two juvenile males pester an adult cow for three hours until she stymied them by concealing. herself in a massive clump of Hydrilla. The excitement generated by sexually aroused males was per- vasive and stimulated other males in the immediate area to sexual activity. The pursuit of cows became more intense and was fre- quently interspersed with homosexuality. Groupings dispersed and reformed in a variety of male-male and male-female combinations. At the height of sexual activity, cows in flight from bulls were seen to tear through vegetation and plunge into the mud. The general agitation subsided in time as the animals dispersed to feed or rest. Homosexuality Homosexual and masturbational activities have been seen among captive male manatees (Earl Herald, personal communication; personal observations). In the wild, manatees also engage in mutual masturbation. In typical homosexual behavior, two bulls embrace, merge genital apertures, then extrude and rub penes. It is likely that ejaculation occurs although I never saw emission of semen. Homosexual activity was never observed among females. Homosexual embraces in no way resembled the copulatory em- Hartman—Manatee in Florida 107 Fic. 37. Male homosexual embraces (drawn from photographs). brace. Males embraced one another head to tail or from the side (Fig. 37). No bull ever assumed the position of a receptive female. In the embrace, frequently preceded by “kisses” at the surface, the two males tumbled to the bottom where they remained tightly clasped, thrusting and wallowing. Homosexual embraces lasted up to two minutes before the participants disengaged to surface. Often more than two bulls were involved simultaneously in a 108 Spec. Publ. Amer. Soc. Mamm. 5 homosexual match. I once watched four adult males as they kissed at the surface, then firmly interlocked in a group embrace. The two outside males clasped the central participants and rubbed their genitalia against any exposed abdominal surface. One of the cen- tral bulls, constrained by the embraces of the three other males, eventually jackknifed out of their grasp. Homosexual behavior was contagious and persisted sometimes for hours as more and more males were drawn into the activity. The individuals involved in the homosexual embraces constantly changed as new groups formed, dissolved, and reformed. Even- tually, with the masturbational drive presumably consummated, the excitement of homosexual activity tapered off. Aroused by homosexual encounters, bulls were apt to mouth and embrace any other individual in the vicinity, whether a male or a female. Other males, however, were not always receptive to these advances and might only engage in a perfunctory embrace or flee, squeaking in apparent annoyance. I once saw a bull em- brace the back of another only to be walloped by the latter’s tail as it fled. Juvenile males engaged in homosexual embraces with both adult and other juvenile bulls. Two young males were once observed nibbling each other’s venters before embracing, but in their em- brace there was less penis-to-penis contact than awkward thrusting against the side and belly of the other animal. Additional males arrived, made incipient gestures of joining the embrace, then swam away. Sometimes homosexual behavior occurred spontaneously. A young bull that was feeding on Hydrilla finally yielded to the per- sistent overtures of another juvenile. The two embraced, sinking into the vegetation. On disengaging, the juvenile returned to feed but the other pressed his advances. The first individual plunged into the Hydrilla and became temporarily enmeshed in an attempt to evade the other. Most often, homosexual behavior resulted from the frustration of repeated thwarting by cows. Competitive belly- to-belly collisions while jockeying for position next to an estrous cow were apt to evolve into homosexual embraces. Play Several types of behavior occur in manatees that may tentatively be defined as play in the context of Loizos (1966). Juvenile males Hartman—Manatee in Florida 109 occasionally exhibited behavior suggestive of sexual activity but lacking the full complement of patterns found in adult reproduc- tive interactions. Young males gathered to mouth and embrace accommodating cows, either juveniles or adults, who, instead of thwarting their advances, appeared to welcome them, allowing bel- ly-to-belly embraces and even tolerating mouthing of the genital regions. For their part, the males neither extruded their penes, clashed competitively, nor engaged in more than incipient homo- sexuality. Some of these play sessions lasted for hours during which activity gained momentum, reached a peak, then gradually subsid- ed. Day after day in the main boil of the Springs O’Paradise, I watched juvenile males caress cows in this playful manner. As many as six juveniles were seen to mouth and embrace a cow simulta- neously. Two cows accepting caresses at once came together from time to time to kiss one another. Cows with calves were also ca- ressed. The calves “stood by” or exchanged random kisses with their mothers or other cows. One mother that had consistently thwarted all advances by young males, finally succumbed to their caresses and was drawn into the play. The arrival of mature bulls usually disrupted the “games,” provoking the cows to abandon play in their effort to rebuff the more serious intentions of the bull. Sexual undercurrents did not appear to run through all manatee play. Much intraspecific contact consisted of gentle nibbles, kisses, and embraces that were reciprocal, age and sex independent, and decidedly nonsexual. In such encounters, cows accepted the ad- vances of bulls but still remained intolerant of contact with their venters. For example, I once observed an adult cow mouthing the back and sides of an old bull lying on the bottom. The bull recip- rocated, but whenever he mouthed the cow’s underside, she thrashed away. Typically, two or three animals were involved in nonsexual ex- changes, although larger groups sometimes formed for mutual kissing, mouthing, embracing, bumping, nudging, and chasing. Calves also participated, frolicking with older manatees as well as with their peers. Once a calf “charged” an idling juvenile from the rear, riding up on its back and causing it to thrust away in surprise. Calf and juvenile then rolled upside down and swam briefly on their backs with muzzles and flippers protruding above the surface. Several times I observed what appeared to be play intention 110 Spec. Publ. Amer. Soc. Mamm. 4 movements. A juvenile cow seemed to be inviting bottom-resting companions to play as she rolled on her back and grazed their sides. Similarly, a young bull swimming on his back made several lazy approaches to bottom-resting cows. Another young male chewed on the tip of a cow’s tail as she bottom-rested but elicited no response. Occasionally manatees played by themselves. An adult bull broke from the playful embraces of two juvenile males, slid along the bottom on his belly, then rolled on his back, skimming the sand and plunging through vegetation. Calves amused themselves by twisting, tumbling, and barrel-rolling through the water. While its mother was resting 20 meters away, one calf completed several minutes of play by “rocketing” to the surface so that his chest and flippers broke water. The ‘function of manatee play remains unclear. Moore (1956) suspected the kiss to be relict behavior inherited from terrestrial ancestors and involving odor of breath. He proposed that it was a form of greeting, perhaps assisting in individual recognition. Yet, when two or more manatees meet, there is no greeting ritual. The animals give little more than a passing glance at one another or approach head-on only to veer aside at the last instant without contact. I believe manatee play may serve as social reinforcement. The reciprocal nibbles and embraces of play may serve to solidify a rudimentary social bond between animals. Undoubtedly the high incidence of tactile contact in play provides the animals with plea- surable sensations. In so far as I could tell, play only occurred when the animals were fully fed and rested and free of environ- mental pressures such as human harassment. Mother-Young Behavior Probably to reduce the danger of predation, a cow about to give birth apparently seeks the safety and seclusion of a sheltered back- water where she bears her calf and nurses it through its vulnerable first days. Evidence for this was my discovery of a dead newborn calf whose carcass washed ashore at the end of a narrow canal 3 kilometers long. Shelter-seeking behavior among cows that are nearing their time of delivery may be characteristic of female sirenians in general. Hartman—Manatee in Florida 111 According to a dugong fisherman in Mozambique, calves are born in estuaries and not in offshore waters (Hughes and Oxley-Oxland, NOMA): Parturition in the manatee has never been witnessed. A head- first presentation, like that of cetaceans, is the likely mode of de- livery. The heavy head of the foetus probably settles in the lower front of the uterus and the tail toward the cervix (Shyper, 1962). Support for this theory is MacMillan’s (1955) observation of a du- gong cow giving birth to a calf head-first on a sand-bank at low tide. This is the only record of a sirenian giving birth out of water. The behavior of captive manatee cows toward newborn calves has been described by Barbour (1937), Moore (195la, 1957), and Phillips (1964). Immediately after delivery the mother is reported to lift her calf above the surface on her back and then to dunk it repeatedly until it has established a breathing rhythm of its own. Within half a day of birth calves are apparently capable of swim- ming and surfacing under their own power, unassisted by their mothers. For the first days of its life the newborn calf of a captive cow swam entirely with its flippers and rode occasionally on the back of the female (Moore, 1957). Calves normally suckled when their mothers were idling or rest- ing. Cows made no movements to accomodate their nurslings. Cows suspended near the surface often raised their heads to breathe while being suckled. Calves nursed suspended or on the bottom or as their mothers sank from the surface to the bottom (Fig. 38). One calf was observed to grasp its mother’s teat as she slowly cruised. To suckle, a calf approaches its mother’s side from the rear, rolls toward her, and grasps the axil of a flipper in its mouth. Not infrequently one can see the lips of a nursing calf move with a slight, regular sucking action. Nursing periods lasted about two minutes. The mean length of 16 nursing intervals involving six different calves was 126 seconds with a range of 106 to 146 sec- onds. Pre-yearling calves, even the younger ones, showed no ten- dency to grasp the teat longer than did yearlings. Calves sometimes released their grasp momentarily to breathe. One mother was seen to break away from her nursing calf but immediately allowed it to resume feeding. On another occasion three cavorting manatees bumped a cow and calf, interrupting nursing. Manatee cows evidently lactate as long as they are accompanied 112 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 38. Cow suckling yearling calf (photo by James A. Sugar, (c) National Geographic Society). by offspring. As they mature, however, calves become less depen- dent on their mothers for milk and increasingly turn their atten- tion to grazing. In six hours of continuous observation I saw one yearling nurse only three times at intervals of 96 and 132 minutes. Sometimes yearlings go even longer without feeding, then, as if to compensate, suckle and graze several times within a few minutes. In the space of five and a half hours a yearling male devoted eight and a half minutes to suckling and 17 minutes to grazing, inter- spersing the two activities at intervals of 262, 14, 6, 11, and four minutes. It appears that sirenian calves begin to nibble on vegetation at an early age. According to Heinsohn (1972), young dugongs start grazing by three months of age. Kenchington (1972) found the stomach of a newborn dugong calf distended with seagrasses. Moore (1957) recorded a 38-day-old captive manatee feeding on lettuce. The youngest calf observed at Crystal River, known to be no more than three months of age, was both nursing and grazing when first encountered. Hartman—Manatee in Florida 113 Apparently cow and calf maintain intimate contact for a period of one to two years. The calf constantly reinforces the bond by mouthing and kissing its mother. As previously noted, a conspic- uous feature of mother-young behavior is synchronous surfacing. Only when one wandered away from the other did a cow and calf fall out of unison in their surfacing. In the absence of its mother, a calf sometimes fell into the breathing rhythm of a neighboring animal. The cow-calf association is markedly close while the pair is cruis- ing, when the risk of becoming separated is undoubtedly the great- est. The calf swims beside its mother, just behind a flipper, crossing occasionally above or beneath her to swim on her opposite side. At no time were mothers seen to assist their young in swimming as has been reported among cetaceans (Tavolga and Essapian, 1957), although this behavior may occur between a mother and a newborn calf. When not cruising, cow and calf showed less intimate association. The calf still tended to follow the lead of its mother—resting when she rested, feeding when she fed—but they were more apt to wan- der away from each other for short periods. While their mothers bottom-rested, calves sometimes ventured to explore, feed, or bot- tom-rest independently for periods up to an hour. Calves also “played” with other manatees, mouthing and grazing the sides of juveniles and adults who, in turn, responded with gentle nibbles, nudges, and embraces. There is little evidence that the cow and calf ever separate beyond the range of effective communication. One six- to eight-month-old calf was extremely independent and strayed afield as much as 50 meters, yet cow and calf managed easily to locate one another, apparently by means of vocal cues. Another female moved more than 60 meters away from her calf but returned to it quickly when alerted by its squeals. I feel certain that manatees communicate underwater over even greater distances than those involved in the above observations. The defensive behavior of a menaced cow with calf was restricted to alarm calls and flight. A cow answered scream for scream any distress calls from its calf, hurried to it, and led it away from the source of anxiety. For example, on one occasion a cow called her calf to her side as it was about to surface too near a boat. In another 114 Spec. Publ. Amer. Soc. Mamm. 5 case, a bull rising from the bottom collided accidentally with a calf who screamed in surprise, igniting a duet with its mother and bringing her to investigate. Cows with calves never were seen to engage in agonistic displays. Mothers never threatened divers that were frightening their calves. Once, however, a cow appeared to dissuade her calf from inves- tigating a diver by turning it aside with gentle nudges. When a cow with calf was the focus of an estrous herd, the position of the calf beside its mother was frequently usurped by courting males, forc- ing the calf to keep station beside peripheral bulls. During intense courtship, the calf was apt to drop far behind the herd into the company of straggling escorts. If a calf became isolated in the imbroglio of sexual activity, its mother, perhaps alerted by loss of vocal contact with her offspring, doubled back to find it. A cow persistently pestered by bulls sometimes fled without summoning her calf to her side. The deserted calf either waited until called by its mother or swam squealing in the direction of her departure. When cows with calves consorted together, calves often had to wait for their mothers to finish feeding or resting. At such times calves milled around and occasionally amused themselves by em- bracing, nudging, kissing, and lazily pursuing one another. With the exception that they did not indulge in the sexual in- teractions typical of older animals, calves exhibited much the same behavior as adults. They rooted in the sand, toyed with inanimate objects, rubbed on logs and ledges, and cavorted with other man- atees. Their hydrobatics were replicas of adult behavior and in- cluded basking at the surface upside down. Male calves even ex- truded their penes from time to time in what seemed to be an involuntary response relating to stretching and not to sexual ex- citation. Hartman—Manatee in Florida 115 SENSES HE following description of manatee sensory capacities is based AG largely on circumstantial evidence. Findings relating to the senses of whales and porpoises are included to illustrate the con- vergent adaptations of sirenians and cetaceans. Hearing Sound appears to be the principal sensory modality of both si- renians and cetaceans (Evans, 1967). At Crystal River, there was no question but that manatees have exceptional acoustic sensitivity and that sound was a major directional determinant in social in- teractions. Guided apparently by vocal cues alone, lagging bulls tended to swim directly toward estrous herds that were beyond visual range. Manatees were also extremely efficient at localizing surface noises. Even in highly turbid water, animals arrived at once to investigate the splashing of divers. Keeping in mind that sound pressure is 60 times greater in water than in air, the manatee’s actual range of effective audibility un- derwater is uncertain. From the shore I attracted animals 15 me- ters away by gently splashing the water with my hand. One animal, passing roughly 40 meters from my boat, altered its course to in- vestigate the banging of an oar against the gunwale. From the air, manatees were seen to change bearing and “zero in” on other manatees from distances as great as 50 meters. As noted above, a cow responded to the squeals of her calf from a distance of more than 60 meters. Reflex actions induced by different man-made sounds provided additional proof of the manatee’s sensitive hearing. The click of a camera shutter within a few meters of their heads caused man- atees to wince. Once, a bull that was lolling beside me flinched when I blew water out of my snorkel. Not infrequently animals were irritated by the high-pitched wheezing of regulators on SCU- BA divers. The sharp sound of an outboard motor changing gears 10 meters away caused several animals to wince in unison. I once saw a cow flinch at the splash of a diver entering the water 30 meters from her. On the other hand, an ultrasonic transmitter, beeping at a fre- quency of 70,000 cycles per second (c/sec) did not elicit reactions from animals even when held next to their external ear openings. 116 Spec. Publ. Amer. Soc. Mamm. 5 It is possible that the manatee cannot hear such high-pitched sounds. The fundamental tones of manatee vocalizations range from 600-5000 c/sec with the highest conspicuous harmonic at 16,000 c/sec (Schevill and Watkins, 1965). Cetaceans seem to be sensitive to a wider range of tones than manatees and have re- sponded to sounds between 150 and 153,000 c/sec (Sliyper, 1962). Sirenians are also able to detect above-water sounds. Oke (1967) described the conditioned response of a captive dugong to a gong. I have witnessed a captive manatee respond to its keeper’s voice by swimming in the direction of the sound. At Crystal River, the roar of a low-flying jet fighter caused a resting male to flinch vio- lently and dive from the surface to the bottom. A group of man- atees bottom-resting at 3 meters was frightened out of the Main Spring by the voices of fishermen whose boat drifted 10 meters away. On a number of occasions I was able to distract manatees from their activities with shouts from the shore a few meters away. Sight The manatee’s vision, used by its early ancestors for a terrestrial existence, has become adapted to life under water. The orbits are lined with oil glands that bathe the corneas, possibly protecting the eyes from the harmful effects of brine. Strands of mucus some- times accumulate in the orbits but do not seem to bother the ani- mals. The manatee blinks frequently to keep its eyes lubricated. The nictitating membrane closes, followed by contraction of the muscular walls of the orbit. It is apparent from the behavioral reactions of captives that many of the smaller toothed whales use vision extensively (D. Cald- well and M. Caldwell, 1972). Like cetaceans, manatees make use of their eyes in clear water, but their vision is presumably of little value under murky conditions. According to Walls (1967), siren- ians have made fewer ocular modifications for aquatic life than cetaceans: the eyes are highly sensitive but visual acuity is poor and visual stimuli of no consequence. The findings of this study contradict Walls. In Kings Bay, the preferred method of environmental exploration by manatees was visual. The animals typically reacted to an unfamiliar auditory stimulus by investigating its source with their eyes. The animals either rose to face the source head-on or balanced on the bottom with their bodies cocked to the side so they could look up. In the Hartman—Manatee in Florida 117 latter case they repeatedly shifted position so that first one eye was exposed and then the other. Manatees approached objects head-on without moving their heads from side to side, suggesting that they posses binocular ste- reoscopic vision. It is obvious, at least, that the visual fields of the two eyes overlap. The rigid neck structure prevents a manatee from turning the head from side to side to scan and forces it to “come about” with its flippers to confront an object. Visual signals alone sufficed to alert manatees. While I lay mo- tionless on the surface, resting animals were often oblivious to my presence, noticing me only when they opened their eyes to surface. Their visual range underwater is as great as a diver’s through a face mask. Under exceptionally clear conditions, animals saw and approached me from distances of up to 35 meters. This refutes the supposition that manatees are myopic. In fact, poor depth perception at close range is evidenced by their tendency to bump heads and eyes, causing them to wince, while rubbing on cement blocks and other objects. There is no reason to suspect that the vision of manatees at night is poorer than that during the day. Their tapeta lucida shine pink in the dark, indicative of good vision in low light intensities (Walls, 1967). The animals rarely shied from a flashlight beam. Kellogg and Rice (1966) found that the bottle-nosed dolphin was sensitive to above-water movements but could not discriminate vi- sual forms from water to air. It is questionable whether the man- atee has the same capacity. At times I was convinced that manatees investigating my boat were alerted by my movements to and fro, but my motion was betrayed by auditory stimuli, which may have been responsible for their attention. As evidence that manatees are either disinterested in or unable to detect above-water motion, no amount of gesticulation from the shore perturbed animals in a canal 3 meters away. Several species of odontocetes are thought to have good vision out of water (Kritzler, 1952; Slijper, 1962; Kellogg and Rice, 1966). The behavior of captive manatees in the Botanic Gardens of Georgetown, Guyana, provide evidence that they also can see with their heads out of water. These animals, accustomed to accepting handouts of plants, raise their heads above the surface at the feet of visitors, then swim with obvious discrimination towards persons that proffer grass or other herbage. At Crystal River, on the other hand, there was no indication that manatees gained information 118 Spec. Publ. Amer. Soc. Mamm. 5 by raising their heads above water. They never raised their eyes above the surface in response to an above-water stimulus. Touch A high incidence of body contact and mouthing in social inter- actions is characteristic of both manatees and dolphins and sug- gests that the sense of touch is very important to both (Tavolga and Essapian, 1957; D. Caldwell and M. Caldwell, 1972). The pro- pensity of manatees to rub on objects and to solicit caresses from divers is further evidence that the epidermis is highly sen- sitive to touch. The scattered dorsal hairs seem to be sensitive to currents in the water and are conceivably receptive to low fre- quency vibrations or pressure waves. Layne and Caldwell (1964) suggested that the Amazon dolphin (Inia) may use the hairs on its rostrum to locate food. Assuming a subminimal auditory cue is not involved, the stimulus for bottom-resting manatees to surface in near unison appears to be tactile; nearby manatees, often faced in the opposite direction, appear to “feel” the slight currents pro- duced by the ascent of a lead animal and are thus stimulated to rise. In turbid water manatees cruising in tandem or in eschelon formation may follow a lead animal by means of the currents gen- erated by its undulations. The bristles on the lip pads serve a tactile function when man- atees feed, when they mouth one another, and when they explore the bottom by nibbling. Taste The prevalence of mouthing in social encounters suggests that manatees have specialized sense receptors on their tongues and respond to gustatory cues. The sophistication of their taste appa- ratus, however, is purely conjectural. Do manatees, for example, possess a chemical sense by which they can recognize odor gradi- ents in the water? Can they obtain directional information from salinity gradients? Do males differentiate anestrous from estrous cows by a change in the taste of their hides? Is a waterborne pher- omone involved in sexual identity? Is there a smell-taste basis to individual recognition? The fact that odors dissipate more slowly in water than in air would tend to favor evolution of a chemoreceptive sense. Fur- thermore, there is growing evidence that toothed whales perceive Hartman—Manatee in Florida 119 odors in water. Yablokov (1957, 1961) discovered preanal glands in male belugas (Delphinapterus) and sperm whales (Physeter) and found “olfactory pits” on the surface of the tongue of several odon- tocetes. D. Caldwell et al. (1966) suggested that sperm whales “lay a trail” in the water, imparting chemical information to other whales and maintaining contact between members of a pod. D. Caldwell and M. Caldwell (1972) cited evidence of a strong gus- tatory sense among captive odontocetes and suggested that taste centers on the tongue may function in chemoreception. It seems reasonable to assume that a communication system utilizing che- moreception also exists among sirenians. Smell Observations at Crystal River indicate that the manatee’s sense of smell, if still functional, is less important than was formerly assumed. Sirenians have a fairly well-developed olfactory organ which, it has been contended, may be used for smelling above the surface (Allsopp, 1961; Slipper, 1962; Harrison and King, 1965). Moore (1956) suggested that the manatee’s habit of “kissing” at the surface involved ritual smelling of one another’s breath or other source of odor. My own observations do not support this inter- pretation. The increased frequency of breathing associated with kissing seemed to me to be a reflex stimulated by exposure of the nose above water. Furthermore, kissing occasionally took place underwater where the nostrils were tightly closed at all times. Whatever manatees recognize or communicate while kissing at the surface, I suspect that the message is transmitted predominantly, if not exclusively, through the mouth, not through the nose. One incident suggests that airborne odors may be relevant to manatees. Within a period of 15 seconds, I once saw an adult male take five breaths where bubbles of flatus from a female were break- ing on the surface. This behavior may not have been related to smell, however, since on no other occasion were animals ever seen “scenting” or making any smelling gestures at the surface. When I was stationed directly upwind of them in my boat, there was no indication that the animals were ever alerted to my presence as a result of smell. Flight reactions could always be traced to auditory and visual signals. Sirenians seem to be similar to cetaceans in having a poor or absent olfactory sense. It is claimed that the sense of smell is com- 120 Spec. Publ. Amer. Soc. Mamm. 45 pletely lacking in odontocetes and rudimentary in mysticetes (Howell, 1930; Sliper, 1962). Yablokov (1961) regarded the un- reduced olfactory region in the cerebral cortex of the common dolphin (Delphinus) as evidence of a functional “sense of smell” probably associated with chemoreception. POPULATION DYNAMICS Birth Rate T appears that manatees in Florida breed year-round. As evi- dence for this, Moore (1953, 1956) cited the fact that young calves have been seen at all times of year. Births at Crystal River provided further evidence that manatees are nonseasonal breed- ers. Five cows calved during the study, three in spring or summer and two in winter. The dates between which births occurred were as follows: 3 February to 26 February; 14 March to 20 July; 25 March to 21 September; 25 May to 26 August; and 20 November to 2 December. Howard Campbell (personal communication) re- cently has gathered data, based on the recovery of newborn calf carcasses, which suggest there may be an increase in natality in the spring. Perhaps manatees in Florida, confronted with winter tem- peratures potentially dangerous to their calves (see section on mor- tality factors), are gradually evolving a reproductive cycle in tune with the North Temperate climate. The gestation period of manatees is not certain. A captive cow calved 152 days after her capture, indicating a gestation of at least five months (Moore, 1951a). Harrison and King (1965) suggested a gestation interval of 12 months. The following observations made at Crystal River during this study indicate an even longer gestation. On 20 October 1967, an adult female became the focus of an estrous herd. She was escorted by bulls until 5 November 1967. Between 20 November and 2 December 1968, she delivered a fe- male calf whose carcass was eventually found by the shore of a canal. Assuming that conception took place during her period of estrus in 1967 and that her calf was born about 25 November 1968, gestation lasted from 385 to 400 days, or approximately 13 months. Weight and measurements (in centimeters) of the dead calf were Hartman—Manatee in Florida 121 as follows: weight, 42.0 kg; total length, 139.7; navel to genital aperture, 22.9; genital aperture to anus, 5.6; length of flipper, 26.7; maximum width of flipper, 8.4; maximum width of tail, 33.5. The postpartum-preconception interval for manatees evidently varies from one to two years, the time required to wean a calf. Cows with calves come into estrus but may not be impregnated (see section on estrous herd). However, females that loose their calves may breed soon after. One cow, for example, which lost her calf shortly after its birth was in estrus 10 to 15 days later. Thus, barring infant mortality, females breed no less than every two years and more likely every two and a half years. In support of this contention, a cow that lost her calf on 25 November 1968 was seen with a new preyearling calf on 11 November 1970 and with another preyearling on 4 December 1972. A cow’s fertile life span and total reproductive capacity are unknown. Norris (1960) and Jarman (1966) stated that dugongs occasion- ally have twins. An incident at Crystal River confirmed that the manatee is not always uniparous. During the winter of 1970-71, a female provided evidence of both twinning and foster parent- hood among manatees. On 11 November, I encountered a cow accompanied by two preyearling calves of identical size. The family was met again a week later. On 19 December, I came upon the same cow accompanied by three preyearlings. The new calf was slightly larger than the twins. Apparently the cow had not only given birth to twins but had also adopted an orphaned calf. I fol- lowed the foursome underwater for half an hour during which each of the twins suckled once. Interactions between the orphan and the cow were similar to those between the cow and its own offspring, leading me to suspect that the orphan was also being nursed. Since it is believed that a cow with a newborn calf immediately raises her offspring above the water on her back to prevent its drowning, there would have to be an interval between delivery of twins that would allow the mother time to introduce the first calf to a surfacing rhythm before the birth of the second. Age at Weaning Mohr (1957) proposed that, with teeth well developed at birth, calves could feed on plants at birth and tolerate separation from 22 Spec. Publ. Amer. Soc. Mamm. 5 their mothers at an early age. Captive calves have been known to nibble algae and grass when only a few weeks old (Phillips, 1964). One calf born and raised in captivity was eating lettuce within 38 days of birth (Moore, 1957). At Crystal River, a calf that was born in July or August was feeding extensively on vegetation the fol- lowing October, but it is not known precisely when it began this behavior. The literature suggests that sirenian calves accompany and are suckled by their mothers from one to two years (MacMillan, 1955; Norris, 1960; Walker, 1964; Harrison and King, 1965; Heinsohn, 1972). Evidence accumulated at Crystal River supports this claim. Of six calves present in Kings Bay the first winter of this study, the two youngest accompanied their mothers for at least a year and a half. The size or age of a calf, however, is an unreliable indication of impending separation from its mother. At the conclusion of the study, one yearling calf (+20 months old) was only 35 centimeters shorter than its mother yet less independent than another calf that was six to eight months old. Age at Sexual Maturity According to Heinsohn (1972), male and female dugongs attain reproductive maturity at approximately two years of age. The age of the manatee at sexual maturity has been estimated at three to four years (Walker, 1964; Harrison and King, 1965). At Crystal River, observations of the size, growth rate, and behavioral attri- butes of juveniles suggested that sexual maturity is not reached before three and possibly as late as five years. This is strictly an estimate as the study was too brief to follow the maturation of a calf to adulthood. Longevity There is no information on the length of life of sirenians in the wild, although Betz (1968) thought that manatee longevity might exceed 50 years. Reports of captive manatees throw little light on the subject. An anecdote in True (1884) mentions a manatee con- fined in a pond by Spanish colonists for 26 years. The longevity record among recent captives is held by a manatee born in captivity 28 years ago and presently housed in Bradenton, Florida. Hartman—Manatee in Florida 123 Mortality Factors Although there is, as yet, no concrete evidence to substantiate the claim, it has been postulated that sirenians are preyed upon by aquatic predators. MacMillan (1955) stated that dugongs panic when killer whales (Orcinus orca) are near. According to several sources (Allsopp, 1961; G. Bertram and C. Bertram, 1968, 1973; Datakaran Jeetlall, personal communication), manatees in the rivers of South America may be vulnerable to attack from piranhas (Serrasalmo). Sharks and crocodilians also have been implicated as potential enemies of manatees and dugongs, especially of calves at parturition (MacMillan, 1955; Jarman, 1966; G. Bertram and C. Bertram, 1968, 1973; Kingdon, 1971). However, natives from Mafia Island (Tanzania) insist that crocodiles never attack dugongs and point out that the sirenians are regularly found in shark-in- fested waters (Dollman, 1933). In Florida, crocodiles (Crocodylus acutus), alligators, and several species of large shark occur in man- atee habitat, but no manatees were seen to bear scars suggesting attacks by any of these predators. From South America there is a report of jaguars (Panthera onca) preying on the Amazonian man- atee (Nunes Pereira, in G. Bertram and C. Bertram, 1973). Pleurisy and bronchial pneumonia from cold exposure have been held responsible for the deaths of captive manatees (Beddard, 1897; Townsend, 1904; Sguros, 1966) and may be the cause of fatalities in the wild during unusually low temperatures. I believe the death of a newborn calf found washed ashore at Crystal River after two nights of subfreezing temperatures was the result of cold exposure. Young calves have a comparatively large surface to vol- ume ratio and are poorly insulated, rendering them more suscep- tible to cold than are older animals. Complications during birth may also result in death. In January 1973, the carcasses of a cow and newborn calf were found together on Sewall Point in the Indian River (Martin County). According to Don Rodgers of the Fish and Wildlife Service, the afterbirth was still intact and protruding from the cow’s genital opening and the calf appeared to have drowned or died of starvation. Measure- ments (in centimeters) of the cow were as follows: total length, 305; navel to genital aperture, 68; genital aperture to anus, 26; length of flipper, 48; maximum width of flipper, 17; maximum width of tail, 66. Measurements of the calf (female) were: total length, 111; 124 Spec. Publ. Amer. Soc. Mamm. 5 navel to genital aperture, 18; genital aperture to anus, 5; length of flipper, 18; maximum width of flipper, 6; maximum width of tail, 20. Two unusual incidents of mortality among sirenians are worthy of mention. The deaths of seven manatees in the Fort Myers area in the spring of 1963 coincided with the deaths of other aquatic animals and with an outbreak of red tide in the vicinity (Layne, 1965). However, there was no conclusive relation between these events. On the coast of India, Jones (1967) reported the discovery of a dead dugong with the tail of a sting ray imbedded in its un- derside, although death could not with certainty be attributed to the sting. By far the most serious threat to the survival of manatees is man. In Florida, the propellers of power craft are the major cause of manatee deaths. Half of the 14 manatee carcasses reported to me from across the state bore severe propeller wounds on the back or head. Virtually all of the animals in Citrus County carried scars inflicted by propellers and skegs. Other important sources of mortality to manatees in Florida are vandalism, poaching, and habitat alteration (Hartman, unpub- lished manuscript). John Reynolds (personal communication) has also called attention to flood control gates which have been known to crush and drown manatees and are a particularly serious threat to the animals in the canal network of southeastern Florida. At least two chemicals have been implicated in the deaths of sirenians: Oke (1967) cited copper sulphate as the possible killer of a captive dugong, and arsenic was discovered in the tissues of a calf found dead in the Crystal River headwaters in 1976 (James Powell, Jr., personal communication). In the St. Johns River, man- atees are presumed to feed on water hyacinth coated with the herbicide 2,4-D, but long-term effects are unknown. During my study, various herbicides, notably copper sulfate and sulfuric acid, were dumped in the rivers of Citrus County in experiments to control aquatic weeds. Manatees were exposed to only dilute con- centrations, and no immediate effects were observed. Polluted water may be responsible for a peculiar condition re- cently noticed among Florida manatees. In June, 1973, residents of Everglades City (Collier County) reported seeing manatees so bloated they could not submerge. The Miami Herald (1 March 1974) carried an article on a bloated manatee found in Broward Hartman—Manatee in Florida 125 County and transferred to the Miami Seaquarium for study. The cause of the bloat has not been diagnosed, but if similar to bloat in ruminants, it may have resulted from the ingestion of bacteria, the fermentation of which in the stomach produced a froth that blocked the alimentary canal preventing the escape of gas. The bacteria were presumably in the water or in plants that were eaten. Bloat evidently offsets the manatee’s neutral buoyancy and pre- vents the animals from diving. I was unable to establish a relation- ship between the appearance of bloated manatees and local water contamination. Another potential source of peril to manatees is the accidental ingestion of hooks, spinners, or other fishing lures caught in the vegetation on which the animals are feeding. Lodged in the diges- tive tract, barb and hook could cause a severe, possibly fatal, ulcer or blockage. No such cases have been documented; that manatees encounter fishing devices, however, was evidenced by hooks and tears in the skin of several animals. It also appears that manatees occasionally drown after becoming entangled in fish or turtle nets. In Guyana, the greatest threat to manatees is accidental capture in fishing nets and subsequent slaughter for food (W. H. L. Allsopp, personal communication). In Florida, I encountered several commercial fishermen who claimed to have had manatees drown in their mullet nets. Entanglement in buoy lines, anchor ropes, underwater cables, and monofilament is another danger for manatees. In the winter of 1971-72, a mother manatee attending the cold-induced congre- gations in Blue Springs Run (Volusia County) was found to have a trotline looped tightly around the axil of her left flipper (James Powell, Jr., personal communication). The flipper had begun to atrophy, and her calf was never seen to suckle from the left teat. Otherwise, the line in no way seemed to handicap the animal or hinder its normal motor patterns. The preyjous summer, person- nel of the Lake Woodruff National Wildlife Refuge reported a manatee towing a pole in the Norris Dead River (Volusia County). A manatee trailing a lobster buoy and line was seen on several occasions in Palm Beach County during the fall of 1973. 126 Spec. Publ. Amer. Soc. Mamm. 5 MAN-MANATEE RELATIONS Response of Man to Manatees LORIDIANS living in contact with manatees tend to have a dis- F passionate or benign attitute toward them. The animals are unobtrusive and interfere little with man’s activities. The general attitude toward manatees is reflected in the views of Citrus County residents, most of whom accept the animals with interest and gra- titude. Manatees are among the attractions that draw divers to the Crystal River springs, thus contributing to the tourist economy of the community. More importantly, their usefulness as agents of aquatic weed control in the spring-fed rivers is welcomed. On the other hand, guides and fishermen complain of breaking propellers on the backs of manatees, and mullet fishermen begrudge mana- tees the holes torn in their gill nets. However, the incidence of such cases is rare and is not treated too seriously. Maliciousness toward manatees does occur. Attacks by vandals, especially in canals on the east coast of Florida, have increased substantially in recent years (Hartman, unpublished manuscript). The following incidents in Citrus County illustrate the problem. In the winter of 1966-67, an animal was observed passing up a little-used canal off Kings Bay with the head of a garden rake imbedded in its back. Several years ago a manatee was shot through the head as it surfaced in the Withlocoochee River. I have person- ally seen bored fishermen hook animals intentionally and “play” them for sport. The most constant sources of harassment to the Crystal River manatees were power boats and divers. The travel routes utilized by manatees were also the principal thoroughfares for boats. The approach of boats often caused animals to interrupt activities or alter their course. Boat traffic was especially heavy on weekends and harassment was compounded by tourists and divers patrolling in search of manatees. Divers daily invaded the Main Spring with SCUBA and snorkel equipment and when manatees were present, bothered the animals by trying to “ride” them. Invariably the man- atees were chased from the spring. The activity of divers, however, was generally restricted to the spring and rarely disrupted the activities of manatees elsewhere in Kings Bay. Turbid water at the Hartman—Manatee in Florida 127 head of the Homosassa River discouraged divers, and the manatees there remained unmolested. Response of Manatees to Man Manatees are inquisitive animals. They were frequently seen to investigate boats that were anchored or drifting nearby. On several occasions during dredging operations at the mouth of the Suwan- »nee River, workmen watched manatees investigate their barge, attracted to it, no doubt, by the noise. Stories of extraordinarily tame manatees are legion. At a marina in Miami there was a fa- mous manatee that would roll over to have its belly scratched with a deck brush. At the City Yacht Pier in St. Augustine, the dock- master uses a paint scraper to clean barnacles off the back of a large manatee that appears regularly. Frank Rivell, captain of a “jungle cruise” on the Tomoka River (Volusia County), claims to have petted manatees over the gunwales of his tour boat. Response to boats.—In general, manatees reacted to a boat under power as to a potential danger. The sound of an approaching motor usually sent animals that were near the surface diving to the bottom. If surprised in shallows (2 meters or less in depth), animals escaped to deeper water. Once they reached the bottom in deep water, they were not as easily intimidated and were more likely to watch a boat pass over than to flee. I have motored over animals bottom-resting at 3 meters barely disturbing them. _Manatees that were overtaken unawares at the surface by a boat “spooked,” thrashing the surface with their tails and churning up _the water in flight. In their panic animals frequently plunged head- long through vegetation. I once surprised a cow and calf who be- came momentarily separated in flight. On another occasion I saw a fleeing juvenile collide with a calf. The alarm of a manatee scared by a boat was often contagious. One afternoon as I was chasing a juvenile male in order to clock his flight speed, the surface erupted suddenly with frightened manatees, some as far as 100 meters from my boat. When pursued by boat, manatees usually accelerated to a fast cruising speed but were slow to change course until about to be overtaken at which time they were apt to veer sharply away from the boat or somersault to reverse direction. Manatees sometimes demonstrated exceptional furtiveness in evading pursuit. Twice 128 Spec. Publ. Amer. Soc. Mamm. 5 Fic. 39. Author scratching juvenile female (photo by James A. Sugar, (c) Na- tional Geographic Society). during the study animals apparently resorted to concealment tac- tics. An adult cow fled into a dense stand of Hydrilla and rested on the bottom under cover. A cow with calf that I had pressed for 10 minutes sought seclusion in the heart of a bed of Myriophyllum. While in hiding, manatees seemed to breathe with unusual stealth, but there was no indication that they remained submerged for abnormal lengths of time. On a few occasions, I managed to follow manatees closely in the runabout without disturbing them. The animals seemed unaware of the boat and the sound of its motor. One cow even collided with the bow on surfacing. Perhaps the noise of a motor at close range interferes with or disrupts sound exchange between animals or in some way disorients them. Response to divers. —TYhe manatees at Crystal River exhibited wide variation in their response to divers. Most were wary and would not permit close approach. A few of the older animals, evidently accustomed to divers, paid them little heed. Manatees that were unfamiliar with divers tended to investigate them with cautious cu- riosity, then fled. Timid animals were often torn between flight Hartman—Manatee in Florida 129 Fic. 40. Above, juvenile female “kissing” author; below, juvenile female nib- bling author’s hand (photos by James A. Sugar, (c) National Geographic Society). 130 Spec. Publ. Amer. Soc. Mamm. 5 and curiosity. This approach-avoidance conflict resulted in inten- tion movements and ambivalence. The animals would rise from the bottom to within arm’s length of a diver, then dive in retreat, squealing constantly and repeating the performance several times before swimming away. A handful of Crystal River animals were attracted to divers and actively solicited caresses from them. At times as many as a half dozen manatees sought my attention and followed me wherever I swam. Manatees remained immobile, often with closed eyes, when caressed by a diver. Some uttered intermittent squeaks. A few rolled on their backs and presented their bellies to be scratched (Fig. 39). Animals routinely resisted caresses in the region of their mouths and noses and tended to evade two-handed embraces, which they seemed to identify with the clasps of their own species. One exceptionally tame juvenile cow gave the impression that she returned to the Main Spring as much to be scratched by divers as to linger in the warm water. I once saw her lolling under the caresses of seven divers. She was also attracted to my runabout as I motored into the spring. More than once, she swam to the boat, nudged the hull at the waterline, and allowed me to pet her from over the gunwale. In my encounters with her underwater, she was extremely for- ward in her attentions. At the surface she nuzzled my face mask in a manner suggesting the manatee kiss (Fig. 40). She also nibbled my dry suit and fingers (Fig. 40) and clutched my body with her flippers, drawing me below the surface. Once she embraced my head and tore off my mask. Throughout these encounters she uttered faint grunts audible only when my head was next to hers. Cows with calves were no more cautious of divers than other manatees. One mother regularly sought the attention of divers in the Main Spring. Her calf, at first unapproachable, gradually fol- lowed its mother’s example and within a few days was responding to caresses. Once while the pair were feeding at the surface on water hyacinth, they accepted some from my hand underwater. Similar behavior was witnessed in Blue Springs Run on the St. Johns River (Volusia County) where several manatees, including mothers and their calves, were persuaded to accept water hyacinth, alligatorweed, coontail, and para grass (Panicum purpurascens) when offered to them underwater by divers. Hartman—Manatee in Florida 131 I felt that in time many of the manatees at Crystal River grew accustomed to my presence and showed an acceptance of me that was not apparent with other divers, but whether they could distin- guish me or whether my quiet approaches simply made them less apprehensive was never clear. The aversion of manatees to the average diver stemmed from thoughtless harassment by divers. They startled animals with their sudden motions and loud splashing. Manatees were especially frightened when divers left the surface and plunged to the animals’ level in the water. SCUBA divers were generally given a wide berth, the manatees evidently associating the sound of a regulator with being pestered. Animals that were being chased or “ridden” by divers turned their backs and jackknifed away when con- strained. No manatee ever made a threatening gesture toward a diver. A few animals accepted abuses by divers as the price for resting in the Main Spring. In late winter of 1969, an estrous herd ap- peared daily at the spring regardless of the commotion caused by divers. Members of the herd, seemingly in anticipation of distur- bance by divers and in preparation for a fast escape, faced their exit route when bottom-resting. The tamer animals, however, were quick to respond to considerate treatment and would allow close approach. One day a bull rubbed against the dangling legs of a friendly diver. The number of divers visiting Crystal River is increasing every year, and the appearance of manatees in the Main Spring is be- coming less frequent. In the winter of 1972-73 manatees were seen in the spring only on mornings of subfreezing temperatures and fled at the arrival of the first boat. DISCUSSION IRENIANS are fully aquatic mammals, a distinction shared only with the Cetacea. The two orders have separate phylogenetic origins but show a basic convergence in morphological and loco- motory adaptations. There also appear to be similarities in their diving physiology (Scholander and Irving, 1941; Irving, 1966; Harrison and Kooyman, 1971; Blessing, 1972) and, as mentioned 132 Spec. Publ. Amer. Soc. Mamm. 5 earlier, in their sensory capacities. A comparison of the behavioral repertoire of manatees and whales in relation to their ecology and evolution provides, I feel, an appropriate conclusion to this study. Sirenians are believed to share a common subungulate origin with the proboscideans (Simpson, 1932; Romer, 1966). Fossil evi- dence suggests that the Sirenia evolved from primitive terrestrial herbivores early in the Tertiary. The Cetacea are also products of early land mammals and appear to have originated at about the same time as the Sirenia. The ancestry of the cetaceans is obscure, but they seem to have derived from insectivore-creodont stock just before the divergence of the carnivore and ungulate lines (Kulu, 1972). According to Howell (1930), the sedentary, herbivorous life of the Sirenia has not stimulated rapid evolutionary change, with the result that manatees and dugongs retain certain terrestrial fea- tures and are less specialized for aquatic life than cetaceans. Their widely separate lineages are reflected in the basic habits of sirenians and cetaceans. Manatees and dugongs are sluggish herbivores that require neither speed nor “intelligence,” because they have no need to catch prey. Most cetaceans, by comparison, are lively carnivores that, to obtain food, must constantly confront problem-solving situations that require some measure of insight. In general, carnivores tend to have a wider range of behavioral capacities than do herbivores, as will be evident in the following comparison. Relatively little is known about the natural history of cetaceans, but it appears that a full spectrum of social organization may be found among members of the order ranging from the solitary lives and small family groups of Balaenoptera musculus, B. borealis, Ba- laena, Kogia, Ina, and Platanista to large schools of Balaenoptera physalus, Eubalaena, Delphinus, Prodelphinus, Lagenorhynchus, Orcael- la, Sousa, Stenella, Steno, and Tursiops that at times number in the hundreds or thousands (Layne, 1958; Sliper, 1962; Pilleri, 1970; Saayman and Tayler, 1973). The size of many cetacean groupings seems to fluctuate with the breeding season and the availability of food. There is little doubt, however, that social ties among ceta- ceans are stronger than those among manatees. The herd instinct seems fundamental to most whales, dolphins, and porpoises but appears vestigial in manatees. Social facilitation is highly evolved in most cetaceans (M. Caldwell and D. Caldwell, 1972) but rudi- mentary in manatees. Formation swimming has been observed 1n the narrow-snouted porpoise (Stenella) and beluga (Delphinapterus) Hartman—Manatee in Florida 133 (Kleinenberg et al., 1964; Evans and Bastian, 1969) but does not occur in manatees. Hump-backed dolphins (Sousa) perform court- shiplike interactions in a greeting context (Saayman et al., 1973), whereas manatees appear to have no greeting ceremony. Another indication that the Cetacea are more tightly social than the Sirenia is evident in epimeletic (care-giving) behavior. Reports of mysticetes and odontocetes supporting injured, distressed, or dead companions and especially of mothers raising dead or afflict- ed calves to the surface are not uncommon (McBride and Hebb, 1948; McBride and Kritzler, 1951; Hubbs, 1953; Moore, 1955; Brown and Norris, 1956; Slijper, 1962; Andersen, 1969; Pilleri and Knuckey, 1969). In most instances the mother was joined in her efforts by other animals. Pilleri (1971b) presented evidence of a female La Plata dolphin (Pontoporia) attempting to free her calf from a net. It should be mentioned, though, that cases of cetaceans failing to give aid to wounded companions have also been recorded (Slijper, 1962). Among manatees there is only one report of assis- tance given to a distressed individual—that of a lone cow butting her dead calf to the surface. According to Sliper (1962), male leaders are found in schools of pilot whales (Globicephala) and bottle-nosed whales (Hyperoodon). Conversely, several species of cetaceans (Eschrichtius, Delphinapterus, Delphinus, Orcinus, and Monodon) are believed to consort in lead- erless schools of mixed age and sex at certain times of the year and in separate age and sex-differentiated schools at other times (Tom- ilin, 1935; Kleinenberg et al., 1964; Rice and Wolman, 1971). There is also evidence that, like manatees, at least some cetaceans may not form stable or cohesive groups. Saayman and Tayler (1973) found that hump-backed dolphins (Sousa) consort with dif- ferent companions in temporary associations. According to Evans and Bastain (1969), territoriality probably does not exist among wild cetaceans, which is in keeping with the behavior of manatees. However, social dominance, virtually absent in manatees, has been observed in both captive and free-ranging cetaceans. In captivity, bottle-nosed and white-beaked dolphins (Lagenorhynchus) have established microterritories and peck orders based largely on size, and there is evidence for social hierarchies among wild whales (Slijper, 1962; Tavolga, 1966; M. Caldwell and D. Caldwell, 1972; Tayler and Saayman, 1972). In Tursiops aduncus, it appears that dominant males cooperate in their behavior to co- ordinate group activities (Tayler and Saayman, 1972). Sliper 134 Spec. Publ. Amer. Soc. Mamm. 5 (1962) hypothesized that violent interactions among cetaceans oc- cur more in contexts of social dominance than in contexts of sexual competition. In manatees, sexual rivalry appears to be the only stimulus for aggression. Visual signals related to threat and sub- mission have been observed in dolphins (D. Caldwell and M. Cald- well, 1972) but evidently do not occur among manatees. Fundamental differences in the behavioral attributes of ceta- ceans and sirenians also are obvious in their play. The antics of porpoises and dolphins, both captive and free, suggest a sense of play far more refined than the stereotyped, sex-oriented play ses- sions of manatees. Play in dolphins appears to be an integral part of their behavior, consummating an important psychological need. Many of the smaller toothed whales occupy hours of each day engaging in hydrobatics and chasing and playing with one another. Turstops has even been known to direct play toward other animals (Tayler and Saayman, 1972). Manatees seem to engage in intra- specific play only coincidentally when the right combination of animals happens to meet. At such times young males chase and embrace females in a near replica of sexual behavior. Manatees have never been observed to play with other animals. Various species of cetaceans are known to ride waves, leap free of the water, spin and somersault in midair, and lobtail (Slijper, 1962; Evans and Bastian, 1969). Manatees have never been seen to breach with the exception of one occasion when an animal 1s reported to have jumped completely out of water to escape capture in a net (Scholander and Irving, 1941). Many of the odontocetes appear to be not only playful but im- itative and inventive. In captivity they are highly responsive to training and have learned a complex variety of tricks, games, and behavioral sequences, many of which they perform without rein- forcement and seemingly for pleasure (McBride and Hebb, 1948; McBride and Kritzler, 1951; Kritzler, 1952; Brown and Norris, 1956; Norris and Prescott, 1961; Layne and Caldwell, 1964; Cald- well and Caldwell, 1964, 1966, 1968; Caldwell et al., 1965; Brown et al., 1966; Tavolga, 1966; M. Caldwell and D. Caldwell, 1972; Tayler and Saayman, 1973). Captive manatees, on the other hand, have shown no evidence of perceptual learning and are only known to perform a few simple conditioned responses in antici- pation of feeding (personal observations). The behavior of a cap- tive male that would roll over and slap his flipper on his chest for Hartman—Manatee in Florida 135 a reward was typical. Dolphins have been known to amuse them- selves for hours playing with floating objects while the manatee’s interest in toying with floating or submerged objects is perfunctory. Both manatees and cetaceans appear to have strong investigative and exploratory drives. However, scouting behavior which has been observed in Turstops (Caldwell et al., 1965) does not seem to occur in manatees. Manatee vocalizations were compared briefly with those of ce- taceans (page 98). Manatees are normally silent and evidently have no need for ultrasound. Their vocal repertoire of squeaks and squeals appears remarkably simple when compared with the wide range of clicked, burst-pulsed, and pure tone signals produced by cetaceans for purposes of echo-ranging and communication (Nor- ris, 1969; Busnel and Dziedzic, 1966). There is evidence that Tur- siops truncatus can discriminate frequencies from | to 36 kHz (Her- man and Arbiet, 1972), whereas the spectrum of recorded frequencies for manatees is only 0.6 to 16 kHz (Schevill and Wat- kins, 1965). It seems that whales and dolphins produce both sonic and ultrasonic emissions in connection with procuring prey, main- taining contact within pods, and expressing a wide variety of emo- tional states (Evans and Bastian, 1969). Manatee phonations seem basically emotive and less socially oriented than the calls of ceta- ceans. The supracranial air sinuses appear to play an important role in the production of sound by cetaceans (Lawrence and Sche- vill, 1956; Kellogg, 1961), but there is no suggestion that sirenians produce noise through the nasal passages. Mass panic and strandings, notably among pilot whales (Globr- cephala), killer whales (Orcinus), false killer whales (Pseudorca), and sperm whales (Physeter), are believed to be related to disruption of sonar and subsequent disorientation in shallow water (Dudok van Heel, 1966). Stranding in manatees appears to result from failure to retreat with an outgoing tide or from evasive action intentionally undertaken by an estrous female to escape pursuit by male escorts. Among the Cetacea there appear to be both arhythmic and rhythmic species. Captive Amazon and bottle-nosed dolphins have been found to intersperse sleep and activity throughout a 24-hour period, although in the latter there is evidence for a periodicity in the frequency of vocalizations (M. Caldwell et al., 1966; Powell, 1966; M. Caldwell and D. Caldwell, 1972). In the wild, Amazon dolphins seem to be active day and night (Layne, 1958), and it 136 Spec. Publ. Amer. Soc. Mamm. 5 appears that gray whales do not rest at night, at least while mi- grating (Cummings ef al., 1968). There is also evidence that sperm whales are equally active day and night (Slyper, 1962). However, Saayman et al. (1973) discovered that the feeding of free-ranging Tursiops aduncus was influenced by a diel cycle and that social in- teractions and whistle phonations in captive Turszops showed a daily rhythm with peaks in early morning and late afternoon. Manatees, it will be recalled, have shown no evidence of a circadian rhythm. All cetaceans on which reproductive data have been accumulated appear to have breeding cycles (McBride and Kritzler, 1951; Ta- volga and Essapian, 1957; Kleinenberg et al., 1964; Tavolga, 1966; Evans and Bastian, 1969; M. Caldwell and D. Caldwell, 1972). Ac- cording to Slijper (1962), seasonal breeding in whales and dolphins often is timed with the end of migrations and the availability of food. It also seems that dugongs have a mating and calving season (Anderson, 1902; Phillips, 1927; MacMillan, 1955; Heinsohn, 1972). However, there is little indication of an annual breeding cycle in manatees. The reason for this is obscure, but it may have to do with the fact that manatees always have access to food and, in the tropics at least, are not subject to seasonal changes in envi- ronmental stress. Prolonged courtship evidently precedes copulation in both ce- taceans and sirenians. Only in the Cetacea, however, does court- ship appear to be characterized by tenderness and active partici- pation of the female (Slijper, 1966). Saayman et al. (1973) found that during courtship in Tursiops aduncus the behavioral roles of male and female were interchangeable. Moreover, courtship be- havior in the hump-backed whale (Megaptera), gray whale (Eschrich- tius), and certain odontocetes seems to involve more activity and ritual than is found in manatees (Tavolga and Essapian, 1957; Sliper, 1962; Sauer, 1963; Evans and Bastian, 1969). Precopula- tory behavior in captive bottle-nosed dolphins includes mutual pos- turing, stroking, rubbing, nuzzling, mouthing, jaw clapping, and yelping (Tavolga and Essapian, 1957). By comparison, courtship activity in manatees 1s limited to chasing, mouthing, and embraces that are initiated by males and never reciprocated by females. It appears that male manatees do not direct their vocalizations as much to the female as to other males; the female, in turn, squeals mostly in annoyance. Tayler and Saayman (1972) found that mating in Turseops adun- Hartman—Manatee in Florida 137 cus 18 a system of rotating consort relationships that does not in- volve the formation of permanent pair bonds. Manatees do not pair up to mate; rather, estrous females are escorted by many males and are promiscuous. Harems and bachelor herds are reported in sperm whales (Slijper, 1962) but are not found in manatees. A strong sexual drive among males seems to be characteristic of both the Sirenia and Cetacea. This may be related to the difficulty of broadcasting the occurrence of estrus in water. In manatees at least, it appears as if male sexual appetite has evolved in conjunc- tion with an extended estrous period to increase the chance of a female being accompanied while in heat. Homosexual behavior has been observed among wild manatees and masturbation in cap- tive manatees and bottle-nosed dolphins (M. Caldwell and D. Cald- well, 1972; personal observations). Horizontal copulation is apparently the rule among cetaceans; however, vertical copulation has been observed in mysticetes and the sperm whale (Slijper, 1962; D. Caldwell et al., 1966), and above- water copulation has been recorded in the Gangetic dolphin (Plat- anista) (Pilleri, 1971a). As far as known, manatees mate only in a horizontal position. Intromission in the bottle-nosed dolphin lasts 2 to 10 seconds (Tavolga and Essapian, 1957) as compared with 15 to 30 seconds in the manatee. Sirenians and cetaceans are generally uniparous. Slow matura- tion of the young and a strong tie between mother and calf are characteristic of both manatees and whales. In both groups, moth- ers bear and suckle their young underwater and raise the newborn to the surface for air. In certain dolphins, however, the care of a young animal is sometimes shared by other cows or “aunts” (Shjper, 1962). There is no evidence of comparable behavior in manatees; the mother raises her offspring unassisted. There is reason to believe that cetacean cows are aggressive in defense of their calves (Slijper, 1962), whereas manatee mothers react pas- sively in times of danger. Manatee calves pump milk from the pectoral teats of their moth- ers by sucking for periods up to two minutes. Cetacean mothers squirt milk into the mouth of the calf from pelvic nipples. In Tur- stops the calf has been seen to suck one to nine times per feeding session, each suckling lasting for only a few seconds (Tavolga and Essapian, 1957). The reason for behavioral differences in manatees and cetaceans 138 Spec. Publ. Amer. Soc. Mamm. 5 are speculative, but it appears that manatees have less need for tight sociality than do pelagic mammals. Manatees live within the confines of rivers and coastal channels where there are more op- portunities to find mates, where the dissemination of odor and sound is more restricted, where the chance of becoming lost is reduced, and where the risk of predation is lessened. Statements in Tate (1947) and Jarman (1966) suggest that dugongs are as loosely organized as manatees and have not evolved stronger social bonds as an adaptation to life in coastal shoals. It is also worth noting that river dolphins (/nza, Lipotes, Platanista, and Pontoporia) and harbor porpoises (Phocoena) tend to be solitary or to live in much smaller groups than their pelagic relatives. Cetaceans that inhabit the open ocean seem to have evolved a highly developed social life as a means to facilitate capture of prey and to counteract the threats of separation and predation. It is known that schooling functions in cooperative fishing (Slyper, 1962; Kleinenberg et al., 1964; Saayman et al., 1973). Coordinated herding of food fish has been reported in a number of delphinids, and I have observed such behavior in Tursiops. A tight communal organization also seems to play a role in defense against predators (Sliper, 1962). Hostile interactions between sharks and dolphins have been reported (Wood et al., 1970; Irvine et al., 1973) and instances recorded of sharks (Carcharodon) and killer whales (Or- cinus) preying on cetaceans (Kleinenberg et al., 1964; Rice and Wolman, 1971; Arnold, 1972; Baldridge, 1972). Manatees, on the other hand, appear to lack predators and have no need for co- operation in the procurement of food, obviating the need for mu- tual aid in times of stress and accounting in part, it seems, for their relatively unstructured social life. SUMMARY HE ecology and behavior of the manatee, Trichechus manatus Linnaeus, were studied in Florida from October 1967 through March 1969, during the winter of 1970-71, and from November 1972 to January 1974. Research was focused on the coast of Citrus County in west-central Florida, particularly on the headwaters of the Crystal River, where clear springs of constant temperature (23.7°C) provided convenient conditions for above- Hartman—Manatee in Florida 139 and underwater observations of periodic cold-induced congrega- tions of manatees during the winter months. Summers were mostly devoted to interviews and aerial surveys and to observations of manatees in salt water. The following were among the major find- ings of the study. 1. Seventy manatees were identified in Citrus County in the winters of 1967-68 and 1968-69. The population consisted of 13 calves, 15 juveniles, and 42 adults, and included 31 males and 32 females among animals of known sex. 2. Drops in the air temperature below 10°C induced congre- gations of manatees in the headwaters of the Crystal and Homo- sassa rivers. A concomitant drop in the temperature of the Gulf coastal waters apparently reinforced this primary stimulus. Man- atees were observed in water as cold as 13.5°C. 3. Populations of manatees appear to be concentrated in select estuarine and riverine habitats on the central west coast of Florida. There is evidence of long-range offshore migrations between pop- ulation centers. Migrations appear to be both seasonal, in response to changes in the air temperature, and nonseasonal. 4. The activities of manatees seem to be arhythmic. 5. In their movements manatees followed established travel routes. They preferred channels that were 2 meters or more in depth and generally shunned waterways less than a meter deep. The animals usually swam at depths of 1 to 3 meters. Salinity, tides, and currents affect the activity of manatees. 6. In the rivers and estuaries of Citrus County, manatees fed almost entirely on submerged vascular plants. Their staple food species were Hydrilla verticillata, Vallisneria neotropicallis, Ceratophyl- lum demersum, Myriophyllum spicatum, Ruppia maritima, and Diplanth- era wrightu. In salt water, manatees preferred Syringodium filiforme and Thalassia testudinum. In the absence of submerged vascular vegetation, manatees were seen to feed on Eichhornia crassipes. In waterways devoid of submerged or floating spermatophytes, man- atees turned to algae and bank growth. Manatees usually fed in discrete sessions during which they focused their attention on one species of plant. There is evidence that the animals time their movements to coincide with the availability of food. They also ap- pear to require fresh water for osmoregulation. 7. The manatee uses its tail not only as an organ of propulsion but as a rudder by means of which it is able to control roll, pitch, 140 Spec. Publ. Amer. Soc. Mamm. 5 and yaw. The flippers are used in precise maneuvering and in minor corrective movements to stabilize, position, and orient the animal. Cruising manatees do not utilize their flippers as hydro- foils. Cruising speeds ranged from 3 to 7 km/hr; fleeing manatees reached speeds of 25 km/hr. 8. Lengths of breaths and the intervals between breaths are cor- related with age and state of activity. The surfacing behavior of resting manatees bore no resemblance to that of cruising animals. Manatees tended to surface synchronously when in groups. 9. Manatees rested by hanging suspended near the surface or lying prone on the bottom. In both positions animals lapsed into a somnolent state with eyes closed and bodies motionless. 10. Comfort activities included stretching, scratching, rubbing, mouth cleaning, and rooting. 11. Manatees periodically regurgitated a yellow-green mash composed of partially digested plant material. 12. The manatee is a weakly social, essentially solitary animal. The family unit consists of a cow with a calf. All other associations, with the exception of an estrous herd, are temporary and loosely organized. Groups are randomly made up of juveniles and adults of both sexes. Social interactions occur in both sexual and nonsex- ual contexts. The repertoire of intraspecific contact includes mouthing, nuzzling, nudging, and embracing. There are no dis- plays. Manatees exhibit social facilitation, but there is no evidence of communal defense or mutual aid. There was virtually no indi- cation of a social hierarchy. 13. Manatees were normally silent but emitted high-pitched squeals, chirp-squeaks, and screams under conditions of fear, ag- gravation, protest, internal conflict, male sexual arousal, and play. Their vocalizations seem to be nonnavigational; to lack ultrasonic signals, pulsed emissions, or directional sound fields; and to be more impulsive than communicative. The only predictable vocal exchange between manatees was the alarm duet between a cow and her calf. 14. A cow in heat was accompanied by courting bulls for periods of less than a week to more than a month. During this time the courtship of bulls was relentless, but the cow appeared to be re- ceptive only at brief intervals. While receptive, the cow was pro- miscuous. Copulation was effected when a bull rolled on his back and grasped the cow to his belly from underneath. Male compe- Hartman—Manatee in Florida 141 tition for the position next to an estrous cow seemed to be the sole source of aggression among manatees. Bulls also made sexual ad- vances toward anestrous cows but were invariably thwarted. Male manatees engaged in various types of homosexual behavior. 15. Manatees indulged in what appeared to be play. The ani- mals exchanged gentle nibbles, kisses, and embraces that were age and sex independent and decidedly nonsexual. Juvenile males oc- casionally instigated interactions with cows that suggested sexual activity, but lacked the full complement of patterns found in adult reproductive behavior. 16. Presumably to reduce the danger of predation, cows are believed to seek the safety of a sheltered backwater to give birth. Nursing periods lasted approximately two minutes. The defensive behavior of a menaced cow with a calf was restricted to alarm calls and flight. 17. Manatees in Florida appear to have no definite breeding season. Barring the death of an infant, cows probably breed every two to two and a half years. Gestation apparently lasts about 13 months. A cow suckles her calf from its birth to the dissolution of the parent-offspring bond, a period of one to two years. One cow at Crystal River provided evidence of both twinning and foster parenthood. 18. The greatest dangers to the manatee in Florida are the pro- pellers of power boats and attacks by vandals. There was no evi- dence of manatees being preyed upon by aquatic predators. 19. Manatees have exceptional acoustic sensitivity; sound is doubtless the major directional determinant in social interactions. Manatees also make extensive use of their eyes; in clear water their preferred method of environmental exploration is visual. The prevalence of mouthing in social encounters suggests that mana- tees possess a chemoreceptive sense by which they can recognize odors in the water. 20. Manatees usually dove from the surface to avoid being hit by power boats. Most of the animals at Crystal River were wary and would not allow close approach by divers. A handful of ani- mals, however, had become inured to the presence of divers and actively solicited caresses from them. 21. Members of the Sirenia and Cetacea are fully aquatic and show convergence in morphology, locomotion, and diving physi- ology. The behavioral attributes of the two orders, however, are 142 Spec. Publ. Amer. Soc. Mamm. 5 largely divergent. Manatees are lethargic littoral herbivores, sub- ject to minimal environmental stress. Without need for association in obtaining food or defense against predators, they manifest a rudimentary sociality. The majority of cetaceans, on the other hand, are animated carnivores whose activities demand insight, group cooperation, and the production of ultrasound. Porpoises and whales, especially pelagic species, appear to have structured societies whose organization is reflected in intricate vocal reper- toires and in highly developed patterns of play, courtship, herding, and defense. Behavioral similarities between manatees and ceta- ceans (at least certain species) include lack of territoriality and ab- sence of a diel rhythm. LITERATURE CITED A.isopp, W. H. L. 1960. The manatee: ecology and use for weed control. Nature, 188:762. . 1961. Putting manatees to work. New Scientist, 12:548—549. . 1969. Aquatic weed control by manatees—its prospects and problems. Pp. 344-351, in Man-made lakes (L. E. Obeng, ed.), Ghana Univ. Press, Ac- cra. ANDERSEN, S. 1969. Epimeletic behaviour in captive harbour porpoise, Phocoena phocoena (L.). Pp. 203-205, in Investigations on Cetacea (G. Pilleri, ed.), vol. 1, Univ. Berne, Switzerland. ANDERSON, J. 1902. Zoology of Egypt: Mammalia. Hugh Rees, Ltd., London, 374 pp. fig eS iate 1917. A new food animal. J. Hered., 8:339-345. 1961. Some notes on the use of the manatee (Trichechus) for the control of aquatic weeds. FAO Fish. Biol. Tech. Paper, 13:1-7. . 1964. Use of manatees to control aquatic weeds. Comm. Fish. Rev., 26:107— 108: . 1973. Some prospects for aquatic weed management in Guyana. Nat. Sci. Res. Council Guyana and Nat. Acad. Sci. U.S.A., Workshop on Aquatic Weed Mgmt. and Utilization, 39 pp. . 1974. An international centre for manatee research. Nat. Sci. Res. Council, Georgetown, Guyana, 34 pp. ARNOLD, P. W. 1972. Predation on harbor porpoise, Phocoena phocoena, by a white shark, Carcharodon carchanas. J. Fish. Res. Bd. Canada, 29:1213-1214. BatpripGE, A. 1972. Killer whales attack and eat grey whale. J. Mamm., 53:898— 900. Bancs, O. 1895. The present standing of the Florida manatee, Trichechus latirostris (Harlan), in the Indian River waters. Amer. Nat., 29:783-—787. Barsour, T. 1937. Birth of a manatee. J. Mamm., 18:106-107. Hartman—Manatee in Florida 143 BarreTT, O. W. 1935. Notes concerning manatees and dugongs. J. Mamm., 16:216-220. Batrawl, A. 1957. The structure of the dugong kidney. Publ. Marine Biol. Stn. Al-Ghardagqa (Egypt), 9:51-68. Bay is, H. A. 1936. Some parasitic worms from the British Cameroons. Ann. Mag. Nat. Hist., 17:257-272. Beck, W. M. 1965. The streams of Florida. Bull. Florida State Mus., 10:91—126. BEDDARD, F. E. 1897. Notes upon the anatomy of a manatee (Manatus inunguis) lately living in the Society’s gardens. Proc. Zool. Soc. London, pp. 47-53. BERTRAM, C. K. RICARDO, AND G. C. L. BERTRAM. 1968. The Sirenia as aquatic meat-producing herbivores. Symp. Zool. Soc. London, 21:385-391. BERTRAM, G. C. L., AND C. K. R1cARDO BERTRAM. 1964. Manatees in the Guianas. Zoologica, 49:115-120. . 1968. Bionomics of dugongs and manatees. Nature, 218:423-426. . 1973. The modern Sirenia: their distribution and status. Biol. J. Linnean Soc., 5:297-338. Betz, J. J. 1968. Sea cow deception. Sea Frontiers, 14:204-—209. Buessinc, M. H. 1972. Studies on the concentration of myoglobin in the sea-cow and porpoise. Comp. Biochem. Physiol., 41A:475-480. Browper, J. 1967. Can man save the manatee? Florida Nat., 40:3-5, 34. Brown, A. E. 1878. The Sirenia. Amer. Nat., 12:291-298. Brown, D. H., D. K. CALDWELL, AND M. C. CALDWELL. 1966. Observations on the behavior of wild and captive false killer whales, with notes on associated behavior of other genera of captive delphinids. Los Angeles County Mus. Contrib. Sci., 95:1-32. Brown, D. H., And K. S. Norris. 1956. Observations of captive and wild cetaceans. J. Mamm., 37:311-326. BusneL, R.-G., AND A. Dztrpzic. 1966. Acoustic signals of the pilot whale Globi- cephala melaena and of the porpoises Delphinus delphis and Phocoena pho- coena. Pp. 607-646, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los Angeles. Buss, I. O., AnD N. S. SmirH. 1966. Observations on reproduction and breeding behavior of the African elephant. J. Wildlife Mgt., 30:375-388. Cann, A. R. 1940. Manatees and the Florida freeze. J. Mamm., 21:222-223. CALDWELL, D. K. 1955. Notes on the spotted dolphin, Stenella plagiodon, and the first record of the common dolphin, Delphinus delphis, in the Gulf of - Mexico. J. Mamm., 36:467-470. CALDWELL, D. K., AnD M. C. CaLpwe._. 1968. The dolphin observed. Nat. Hist., 77:58-65. . Senses and communication. Pp. 466-502, in Mammals of the sea: biology and medicine (S. H. Ridgway, ed.), Charles C. Thomas, Springfield, Il- linois. CaLpweELtL, D. K., M. C. CALDWELL, AND D. W. Rice. 1966. Behavior of the sperm whale, Physeter catodon L. Pp. 677-717, in Whales, dolphins, and por- poises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los An- geles. CALDWELL, D. K., AND F. B. GoLLrey. 1965. Marine mammals from the coast of Georgia to Cape Hatteras. J. Elisha Mitchell Sci. Soc., 81:24-32. CaLpweELL, M. C., anp D. K. CaLpweLi. 1964. Experimental studies on factors involved in care-giving behavior in three species of the cetacean family Delphinidae. Bull. So. California Acad. Sci., 63:1—20. 144 Spec. Publ. Amer. Soc. Mamm. 5 ——. 1966. Epimeletic (care-giving) behavior in Cetacea. Pp. 755-789, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los Angeles. . 1972. Behavior of marine mammals. Pp. 419-465, zn Mammals of the sea: biology and medicine (S. H. Ridgway, ed.), Charles C. Thomas, Spring- field, Illinois. CALpwELL, M. C., D. K. CALDWELL, AND W. E. Evans. 1966. Sounds and behavior of captive Amazon freshwater dolphins, Inia geoffrensis. Los Angeles County Mus. Contrib. Sci., 108:1—24. CaLpweELL, M. C., D. K. CALDWELL, AND J. B. SIEBENALER. 1965. Observations on captive and wild Atlantic bottlenose dolphins, Turstops truncatus, in the northeastern Gulf of Mexico. Los Angeles County Mus. Contrib. Sci., 91:1-10. CALDWELL, M. C., N. R. HALL, AnD D. K. CaLpwe.i. 1972. Ability of an Atlantic bottlenosed dolphin to discriminate between, and respond differentially to, whistles of eight conspecifics. Proc. 8th Ann. Conf. Biol. Sonar Diving Mammals, pp. 57-65. CHAPMAN, H. C. 1875. Observations on the structure of the manatee. Proc. Acad. Nat. Sci. Philadelphia, pp. 452-462. CHARNOCK-WILSON, J. 1968. The manatee in British Honduras. Oryx, 9:293-—294. Coates, G. W. 1939. Baby mermaid; a manatee at the aquarium. Bull. New York Zool. Soc., 42:140-148. . 1940. Manatees at the aquarium. Bull. New York Zool. Soc., 43:99—100. Crane, A. 1881. Notes on the habits of the manatees (Manatus australis) in captivity in the Brighton Aquarium. Proc. Zool. Soc. London, pp. 456-460. CuMMINGS, W. C., P. O. THompson, AND R. Cook. 1968. Underwater sounds of migrating gray whales, Eschrichtius glaucus (Cope). J. Acoust. Soc. Amer., 44:1278-1281. Daitey, M. D., ano R. L. BRowneELL, JR. 1972. A checklist of marine mammal parasites. Pp. 528-589, in Mammals of the sea: biology and medicine (S. H. Ridgway, ed.), Charles C. Thomas, Springfield, Illinois. Dampier, W. 1906. Dampier’s voyages. E. Grant Richards, London, vol. 1, 612 pp. DavituerRs, C. 1938. Sur la biologie du lamantin en captivite. Mammalia, 2:84— 88. Dawson, C. E. 1955. A study of the oyster biology and hydrography at Crystal River, Florida. Bull. Inst. Marine Sci. (Univ. Texas), 4:279-302. DotimaN, G. 1933. Dugongs from Mafia Island and a manatee from Nigeria. Nat. Hist. Mag., 4:117-125. Dupok van Hee, W. H. 1966. Navigation in Cetacea. Pp. 597-606, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berke- ley and Los Angeles. Evapata, S. A. I. 1969. The sexual behavior of wild elephants in Ceylon. Loris, 11:246-247. ErpMan, D. S. 1970. Marine mammals from Puerto Rico to Antigua. J. Mamm., 51:636-639. Evans, W. E. 1967. Vocalization among marine mammals. Pp. 159-186, in Marine bio-acoustics (W. N. Tavolga, ed.), vol. 2, Pergamon Press, New York. Evans, W. E., AND E. S. Heratp. 1970. Underwater calls of a captive Amazon manatee, Trichechus inunguis. J. Mamm., 51:820-823. Evans, W. E., AND J. BAsTIAN. 1969. Marine mammal communication: social and ecological factors. Pp. 425-475, in The biology of marine mammals (H. T. Anderson, ed.), Academic Press, New York and London. Hartman—Manatee in Florida 145 Frets, W. J. L. 1966. Some functional and structural characteristics of cetacean flippers. Pp. 255-276, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los Angeles. FLower, W. H., AND R. LypEKKER. 1891. An introduction to the study of mammals, living and extinct. Adam and Charles Black, London, 763 pp. Furcuson, G. E., C. W. LincnaM, S. K. Love, AND R. O. VERNON. 1947. Springs of Florida. Bull. Florida Geol. Survey, 31:1-196. Garrop, A. H. 1879. Notes on the manatee (Manatus americanus) recently living in the Society’s garden. Trans. Zool. Soc. London, 10:137-145. Gonar, H. A. F. 1957. The Red Sea dugong. Publ. Marine Biol. Stn. Al-Ghardaga (Egypt), 9:3-50. Goopwin, G. G. 1946. The end of the great northern sea cow. Nat. Hist., 55:57— 61. Gray, J. 1968. Animal locomotion. W. W. Norton and Co., New York, 479 pp. Gunter, G. 1941. Occurrence of the manatee in the United States with records from Texas. J. Mamm., 22:60-64. —. 1942. Further miscellaneous notes on American manatees. J. Mamm., 23:89-90. HamILTon, W. J., JR. 1941. Notes on some mammals of Lee County, Florida. Amer. Midland Nat., 25:686-691. Harrison, R. J., AND J. E. Kinc. 1965. Marine mammals. Hutchinson and Co., Ltd., London, 192 pp. Harrison, R. J., AND G. L. Kooyman. 1971. Diving in marine mammals. Oxford Univ. Press, 16 pp. Hartman, D.S. Manuscript. Distribution, status, and conservation of the manatee in the United States. Unpublished, 247 pp. HeEInsoun, G. E. 1972. A study of dugongs (Dugong dugon) in northern Queens- land, Australia. Biol. Conserv., 4:205-213. Hernsouwn, G. E., AND W. R. Bircu. 1972. Foods and feeding habits of the dugong, Dugong dugon (Erxleben), in northern Queensland, Australia. Mammalia, 36:414-422. HEINsOoHN, G. E., AND A. V. Spain. 1974. Effects of a tropical cyclone on littoral and sub-littoral biotic communities and on a population of dugongs (Du- gong dugon (Muller)). Biol. Conserv., 4:143—-152. - HERMAN, L. M., AND W.R. ArBeiT. 1972. Auditory frequency discrimination from 1-36 kHz in Tursiops truncatus. Proc. 8th Ann. Conf. Biol. Sonar Diving Mammals, pp. 79-87. Herte.t, H. 1969. Hydrodynamics of swimming and wave-riding dolphins. Pp. 31-63, in The biology of marine mammals (H. T. Andersen, ed.), Aca- demic Press, New York and London. Hitt, W. C. O. 1945. Notes on the dissection of two dugongs. J. Mamm., 26:153- 175. Howe ., A. B. 1930. Aquatic mammals: their adaptations to life in the water. Charles C. Thomas, Springfield, Illinois, 338 pp. Husss, C. L. 1953. Dolphin protecting dead young. J. Mamm., 34:498. Hucues, G. R., AND R. OxLEY-OxLAND. 1971. A survey of dugong (Dugong dugon) in and around Antonio Ennes, northern Mocambique. Biol. Conserv., 3:299-301. Humes, A. G. 1964. Harpacticus pulex, a new species of copepod from the skin of a porpoise and a manatee in Florida. Bull. Marine Sci. Gulf Caribbean, 14:517-528. 146 Spec. Publ. Amer. Soc. Mamm. 5 Irving, B., R. S. WELLS, AND P. W. GitBerT. 1973. Conditioning an Atlantic bot- tlenosed dolphin, Tursiops truncatus, to repel various species of sharks. J. Mamm., 54:503-505. IrvinG, L. 1966. Elective regulation of the circulation in diving animals. Pp. 381— 396, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. Calli- fornia Press, Berkeley and Los Angeles. . 1969. Temperature regulation in marine mammals. Pp. 147-174, in The biology of marine mammals (H. T. Anderson, ed.), Academic Press, New York and London. . 1973. Aquatic mammals. Pp. 47-96, in Comparative physiology of ther- moregulation (G. C. Whittow, ed.), vol 3, Academic Press, New York and London. Jarman, P. J. 1966. The status of the dugong (Dugong dugon Muller) in Kenya. East African Wildlife J., 4:82—88. Jones, J. K., JR., AND R. R. JoHNson. 1967. Sirenians. Pp. 366-373, im Recent mammals of the world: a synopsis of families (S. Anderson and J. K. Jones, Jr., eds.), Ronald Press Co., New York. Jones, S. 1967. The dugong: its present status in the seas round India with ob- servations on its behavior in captivity. Internat. Zoo Yearbook, 7:215- 220. KANWISHER, J., AND G. SUNDNES. 1966. Thermal regulation in cetaceans. Pp. 297— 409, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. Cali- fornia Press, Berkeley and Los Angeles. KeLtocc, W. N. 1961. Porpoises and sonar. Univ. Chicago Press, 177 pp. KeLtoce, W. N., AND C. E. Rice. 1966. Visual discrimination and problem solving in a bottlenose dolphin. Pp. 731-754, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los Angeles. KENCHINGTON, R. A. 1972. Observations on the digestive system of the dugong, Dugong dugon (Erxleben). J. Mamm., 53:884—887. Kenny, R. 1967. The breathing pattern of the dugong. Australian J. Sci., 29:372—- Bes Kincpon, J. 1971. East African mammals: an atlas of evolution in Africa. Academic Press, New York and London, 446 pp. Kinzer, J. 1966. Beobachtungen tiber das Verhalten des Lamantin Trichechus sen- egalensis (Link, 1795) in Gefangenschaft. A. Sauget., 31:47—52. KLEINENBERG, S. E., A. V. YABLOKOV, B. M. BEL’KOVICH, AND M. N. TARASEVICH. 1964. Beluga (Delphinapterus leucas): investigation of the species. Akad. Nauk SSSR, Moscow, 456 pp. [Translated by the Israel Program for Sci- entific Translations, Jerusalem, 1969, 375 pp.] Kooyman, G. L. 1973. Respiratory adaptations in marine mammals. Amer. Zool., 13:457-468. Krirzier, H. 1952. Observations on the pilot whale in captivity. J. Mamm., 33:321-334. Krumuo1z, L. A. 1943. Notes on manatees in Florida waters. J. Mamm., 24:272— Pall) Kutu, D. D. 1972. Evolution and cytogenetics. Pp. 503-527, in Mammals of the sea: biology and medicine (S. H. Ridgway, ed.), Charles C. Thomas, Springfield, Illinois. Lane, T. G. 1966. Hydrodynamic analysis of cetacean performance. Pp. 410-432, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los Angeles. LawreENcE, B., AND W. E. ScHEVILL. 1956. The functional anatomy of the delphinid nose. Bull. Mus. Comp. Zool., 114:103—151. Hartman—Manatee in Florida 147 Layne, J. N. 1958. Observations of freshwater dolphins in the upper Amazon. J. Mamm., 39:1-22. . 1965. Observations on marine mammals in Florida waters. Bull. Florida State Mus., 9:131-181. Layne, J. N., AND D. K. CALDWELL. 1964. Behavior of the Amazon dolphin, Jnza geoffrensis (Blainville), in captivity. Zoologica, 49:81—-108. Loizos, C. 1966. Play in mammals. Pp. 1-9, 7 Play, exploration, and territory in mammals (P. A. Jewell and C. Loizos, eds.), Symp. Zool. Soc. London, no. 18. MacLaren, J. P. 1967. Manatees as a naturalistic biological mosquito control meth- od. Mosquito News, 27:387-393. MacMi1aNn, L. 1955. The dugong. Australian Geogr. Walkabout Mag., 21:17—20. Mann, J. A., AND R. N. Cuerry. 1969. Large springs of Florida’s “Sun Coast” Citrus and Hernando Counties. Leaflet U. S. Geol. Survey, Tallahassee, 9:1-21. McBripr, A. F., AND D. O. Hess. 1948. Behavior of the captive botilenose dolphin, Tursiops truncatus. J. Comp. Physiol. Psych., 41:111—123. McBripr, A. F., AND H. Kritz_er. 1951. Observations on pregnancy, parturition, and post-natal behavior in the bottlenose dolphin. J. Mamm., 32:251— 266. McCa _t, D., J. G. Cook, J. A. LASATER, AND T. A. NEVIN. 1970. A survey of salinity levels in the Indian River-Banana River complex. Bull. Env. Contam. Toxicol., 5:414-421. McKay, G. M. 1973. Behavior and ecology of the Asiatic elephant in southeastern Ceylon. Smithsonian Contrib. Zool., 125:1-113. McNutty, J. K., W. N. Linpatt, JR., AND J. E. Sykes. 1972. Cooperative Gulf of Mexico estuarine inventory and study, Florida: phase I, area description. NOAA Tech. Rep., Circ. Nat. Marine Fish. Serv., 368:1—126. Mixter, G. S. 1918. Mammals and reptiles collected by Theodor de Booy in the Virgin Islands. Proc. Smithsonian Inst., 54:507-511. Monk, E. 1957. Sirenen oder Seektithe. Die neue Brehm-Buicherei, 197:1-61. Moorg, J. C. 195la. The status of the manatee in the Everglades National Park, with notes on its natural history. J. Mamm., 32:22-36. —. 1951b. The range of the Florida manatee. Quart. J. Florida Acad. Sqi., 14:1-19. —. 1953. Distribution of marine mammals to Florida waters. Amer. Midland Nat., 49:117-158. —. 1955. Bottle-nosed dolphins support remains of young. J. Mamm., 36:466— 467. —. 1956. Observations of manatees in aggregations. Amer. Mus. Novit., 1811:1-24. . 1957. Newborn young of a captive manatee. J. Mamm., 38:137-138. Murig, J. 1874. On the form and structure of the manatee (Manatus americanus). Trans. Zool. Soc. London, 8:127—202. —. 1880. Further observations on the manatee (Manatus). Trans. Zool. Soc. London, 11:19-48. Norris, C. E. 1960. The dugong. Loris, 8:296—-300. Norris, K. S. 1969. The echolocation of marine mammals. Pp. 391-423, zn The biology of marine mammals (H. T. Andersen, ed.), Academic Press, New York and London. Norris, K. S., AND J. H. Prescotr. 1961. Observations on Pacific cetaceans of Californian and Mexican waters. Univ. California Publ. Zool., 63:291- 402. 148 Spec. Publ. Amer. Soc. Mamm. 5 Ovum, H. T. 1957. Primary production measurements in eleven Florida springs and a marine turtle-grass community. Limnol. Oceanogr., 2:85—97. Oxr, V. R. 1967. A brief note on the dugong at Cairns Oceanarium. Internat. Zoo Yearbook, 7:220-221. Owen, R. 1838. On the anatomy of the dugong. Proc. Zool. Soc. London, pp. 28- 46. Parker, G. H. 1922. The breathing of the Florida manatee (T7richechus latirostris). J. Mamm., 3:127-135. Petit, G. 1925. Recherches anatomiques sur l'appareil genito-urinaire male des sireniens. Archives de Morphologie Generale et Experimentale, Paris, 326 pp. PHILLIPS, C. 1964. The captive sea. Chilton Co., Philadelphia, 284 pp. Puiurps, R. C. 1960a. Observations on the ecology and distribution of the Florida seagrasses. Florida State Bd. Conserv. Prof. Papers Ser., Dil 72. 1960b. The ecology of marine plants of Crystal Bay, Florida. Quart. J. Florida Acad. Sci., 23:328-337. Puitiips, W. W. A. 1927. Guide to the mammals of Ceylon, Part 7: Sirenia. Ceylon J. Sci., 14:51-55. Pitter, G. 1970. Observations on the behaviour of Platanista gangetica in the Indus and Brahmaputra Rivers. Pp. 27—60, in Investigations on Cetacea -(G. Pilleri, ed.), vol. 2, Univ. Berne, Switzerland. . 1971a. Observation on the copulatory behaviour of the Gangetic dolphin, Platanista gangetica. Pp. 31-33, in Investigations on Cetacea (G. Pilleri, ed.), vol. 3, Univ. Berne, Switzerland. . 1971b. Epimeletic (nurturant) behaviour by the La Plata dolphin, Ponto- poria blainviller. Pp. 74-76, in Investigations on Cetacea (G. Pilleri, ed.), vol 3, Univ. Berne, Switzerland. PILLeRI, G., AND J. KNuckEy. 1969. Behaviour patterns of some Delphinidae ob- served in the Western Mediterranean. Z. Tierpsychol., 26:48—72. Pirie, N. W. 1967. Orthodox and unorthodox methods of meeting world food needs. Sci. Amer., 216:27-—35. PoweL, B. A. 1966. Periodicity of vocal activity of captive Atlantic bottlenose dolphins: Tursiops truncatus. Bull. So. California Acad. Sci., 65:237-244. PoweLt, J. A., JR. 1978. Evidence of carnivory in manatees (Trichechus manatus). J. Mamm., 59:442. Prater, S. H. 1928. The dugong or sea cow (Halicore dugong). J. Bombay Nat. Hist. Soc., 33:84-99. Price, E. W. 1932. The trematode parasites of marine mammals. Proc. U.S. Nat. Mus., 81:1-68. Quirinc, D. P., anp C. F. Hartan. 1953. On the anatomy of the manatee. J. Mamm., 34:192-203. RADAKRISHNAN, C. V., AND R. E. BRADLEY. 1970. Some helminths from animals at Busch Gardens Zoological Park. Assoc. Southeast. Biol. Bull., 17:58—59. Ray, C. E. 1960. The manatee in the Lesser Antilles. J. Mamm., 41:412—413. Rice, D. W., AND A. A. WotMAN. 1971. The life history and ecology of the gray whale (Eschrichtius robustus). Spec. Publ. Amer. Soc. Mamm., 3:1—142. Ricketr, H. W. 1924. A quantitative study of the flora of Green Lake, Wisconsin. Trans. Wisconsin Acad. Sci. Arts Let., 21:381—-414. Romer, A. S. 1966. Vertebrate paleontology. Univ. Chicago Press, 468 pp. SaayMaN, G. S., AND C. K. TayLer. 1973. Social organization of inshore dolphins (Tursiops aduncus and Sousa) in the Indian Ocean. J. Mamm., 54:993-996. Hartman—Manatee in Florida 149 SaayMaNn, G. S., C. K. TayLer, AND D. Bower. 1973. Diurnal activity cycles in captive and free-ranging Indian Ocean bottlenose dolphins (Tursiops aduncus Ehrenberg). Behaviour, 44:212—233. Sauer, E. G. F. 1963. Courtship and copulation of the gray whale in the Bering Sea at St. Lawrence Island, Alaska. Psychol. Forsch., 27:157-174. ScHEVILL, W. E., aNnb W. A. Watkins. 1965. Underwater calls of Trichechus (man- atee). Nature, 205:373-374. SCHOLANDER, P. F., AND L. IrRvinc. 1941. Experimental investigations on the res- piration and diving of the Florida manatee. J. Cell. Comp. Physiol., 17:169-191. Scuros, P. 1966. Research report and extension proposal submitted to the Central and Southern Florida Flood Control Board on use of the Florida manatee as an agent for the suppression of aquatic and bankweed growth in es- sential inland waterways. Dept. Biol. Sci., Florida Atlantic Univ., Boca Raton, 57 pp. SHort, R. V. 1966. Oestrous behavior, ovulation and the formation of the corpus luteum in the African elephant. East African Wildlife J., 4:56—-68. Simpson, G. G. 1932. Fossil Sirenia of Florida and the evolution of the Sirenia. Bull. Amer. Mus. Nat. Hist., 59:419—503. SLyPER, E. J. 1962. Whales. Basic Books, New York, 475 pp. . 1966. Functional morphology of the reproductive system in Cetacea. Pp. 277-319, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los Angeles. Soxkotorr, D., AND E. CABALLERO. 1932. Una nueva especie de trematodo parasite del intestino del manati: Schizamphistoma manati. Anal. Inst. Biol. (Mexi- co), 2:163-167. STELLER, G. W. 1751. De Bestiis marinis. Novi Comm. Acad. Sci. Petropolitanae, 2:289-398. Tass, D. C., D. L. DuBrow, AND R. B. MANNING. 1962. The ecology of northern Florida Bay and adjacent estuaries. Florida Bd. Conserv. Tech. Ser., 39:1-— 81. Tate, G. H. H. 1947. Mammals of Eastern Asia. MacMillan Co., New York, 366 PRE Tavoica, M. C. 1966. Behavior of the bottlenose dolphin (Turstops truncatus): social interactions in a captive colony. Pp. 718-730, in Whales, dolphins, and porpoises (K. S. Norris, ed.), Univ. California Press, Berkeley and Los Angeles. Tavouca, M. C., ano F. S. Essapian. 1957. The behavior of the bottle-nosed dolphin (Turszops truncatus): mating, pregnancy, parturition and mother- infant behavior. Zoologica, 42:11-31. Taycer, C. K., AND G. S. SAAYMAN. 1972. The social organization and behaviour of dolphins (Tursiops aduncus) and baboons (Papio ursinus): some compar- isons and assessments. Ann. Cape Prov. Mus. Nat. Hist., 9:11—49. . 1973. Imitative behaviour by Indian Ocean bottlenose dolphins (Turszops aduncus) in captivity. Behaviour, 44:286-298. THorRNE, R. F. 1954. Flowering plants of the waters and shores of the Gulf of Mexico. Pp. 193-202, in Gulf of Mexico: its origin, waters, and marine life. U.S. Fish and Wildlife Serv., Fish. Bull. 89. Tomiuin, A. G. 1935. Maternal instinct and sexual attachment in whales. Bull. Soc. Nat. (Moscow), Sec. Biol., 44:351-361 (Russian with English summary). . 1951. On the thermal regulation in cetaceans. Priroda, 6:55—58 (Russian with English summary). 150 Spec. Publ. Amer. Soc. Mamm. 5 TOWNSEND, C. H. 1904. Notes on the manatee or sea-cow. 8th Ann. Rep. New York Zool. Soc., pp. 85-87. TRuE, F. W. 1884. The sirenians or sea cows. Fisheries and Fisheries Ind. of the U.S., Sec. 1:114-128. VeRNON, R. O. 1951. Geology of Citrus and Levy Counties, Florida. Bull. Florida Geol. Survey, 33:1—256. Wa ker, E. P. 1964. Mammals of the world. Johns Hopkins Press, Baltimore, vol. 2, pp. 647-1500. Watts, G. L. 1967. The vertebrate eye and its adaptive radiation. Hafner Publ. Co., New York, 785 pp. WETTERHALL, W. S. 1965. Reconnaissance of springs and sinks in west-central Florida. Florida Geol. Survey, Rep. Invest., 39:1-42. Woop, F. G., JR., D. K. CALDWELL, AND M. C. CALDWELL. 1970. Behavioral inter- actions between porpoises and sharks. Pp. 264-277, in Investigations on Cetacea (G. Pilleri, ed.), vol. 2, Univ. Berne, Switzerland. YaBLoKov, A. V. 1957. On the organs of chemical perception and glands of special purpose among certain toothed whales. Coll. Students’ Sci. Papers, Mos- cow State Univ., pp. 19-20. [Translated by Alberta Freidus.] . 1961. The “sense of smell” of marine mammals. Trudy Soveshch. Ikhtiol. Komis. Akad. Nauk SSSR, 12:87—93 (Russian with English summary). INDEX Acoustical sensitivity, 115-116 Aggression, 97, 105, 134 Alarm calls, 96, 99-100, 127 Aquatic vegetation, 2, 6-11, 41 as food, 44-53 Associates: invertebrate, 62—64 vertebrate, 59-60, 63 Barnacles, 23, 58, 63 Behavior: aggressive, 97, 105, 134 comfort movements, 89—93 eliminative, 93, 95 feeding, 42, 51-53, 55-59, 85 flight, 96, 99-100, 113 homosexual, 99, 101, 106—108, 137 influence of habitat factors on, 36-41, 55-56 interaction with other animals, 59-60 interest in inanimate objects, 60-61 play, 43, 108-110, 113, 134-135 resting, 82-84 sexual, 37, 99-105, 136-137 social, 95-114 surfacing, 73-75 territorial, 98, 133-134 Birth rate, 120-121, 137 Bloating, 124-125 Blue Springs Park, 1, 25-27, 35, 39, 47-48, 52-53, 56, 125 Breathing, 73-82 duration, 78-79 effect of cold on, 82 exhalation-inhalation times, 79 intervals, 80-82 synchronous surfacing, 77-78, 113 variations in surfacing, 73-75 ventilating, 75—77 Breeding season, 120 Calves, 16, 35, 69, 72, 77-84, 95, 99-100, 111-114, 121-124 Cetacea, 23-24, 64-65, 69, 73, 80- 81> 1005 113. 15-120) 131= 138 Chemoreception, 118-119 Citrus County: coastal rivers in, 3—7 flora of, 2-3, 6=11 physiography, 1-3 Comfort movements, 86—93 Commensals, 62—64 Cold-induced congregations, 2a Copulation, 102-104, 137 Courtship, 100-105, 136 Currents: influence on manatees, 39—40 17- Daily activity, 41-43 Defecation, 93 Depth: influence on manatees, 36-38 Distribution in U.S., v, 1 Dredging, 6—7, 10, 15 Drowning, 125 Dugong dugon, see dugong Dugong, v, 31, 32, 49, 57-58, 59, 62, 67, 69, 75, 82, 85-86, 111, 2S NO. 21225123124 132, 136, 138 Elephants, 103, 105 Eliminative behavior, 93, 95 Estrous herd, 43, 100-105, 137 Feces, 93 ingestion of, 55 151 152 Spec. Publ. Amer. Soc. Mamm. 5 Feeding, see food habits Flatus, 93 Flippers, 69-72, 85 Food habits: carnivorousness, 55 coprophagy, 55 daily requirements, 55 feeding methods, 51-53, 85 freshwater needs, 58—59 ingestion of detritus, 53, 55, 90— 93 ingestion of invertebrates, 53 plant preferences, 44-51 selection of sites, 56—58 sessional feeding, 55 timing of movements, 56-58 Fosterage, 121 Gestation, 120-121 Hearing, 115-116 Homosexuality, 99, 101, 106-108, 137 Hydrodamalis, see Steller’s sea cow Identification of individuals, 14 Interactions: interspecific, 59-60 intraspecific, 95-114 Locomotion, 64—72 speeds, 69 tail stroke rate, 67 use of flippers, 69-72 use of tail, 64-67, 70 Longevity, 122 Manatee: age at sexual maturity, 122 age classes, 16 as tourist attraction, 126 effect of cold on, 17-27, 123 measurements, 16, 120-121, 123-124 metabolic rate of, 24, 82 numbers in Citrus County, 15-16 osmoregulation, 58-59 sex ratio at Crystal River, 16-17 skeleton, 64, 70 skin, 62, 118 species of, v weed control by, v, 126 Man-manatee relations, 126-131 Masturbation, 89-90, 106 Maternal-young relations, 95, 110— 114 communication, 96, 113—114 nursing, 111-112, 137 synchronous surfacing, 77-78, ig} weaning, 112-113, 121-122 Mortality, 123-125 Movements: daily, 41-43 exploratory activity, 35-36, 60- 61 migrations, 23, 28-32, 35-36 travel routes, 32—35 Mouth cleaning, 90 99-100, Natality, 120 Navigation, see orientation Olfaction, see smell Orientation, 24, 31—32 Osmoregulation, 59 Parasites: external, 62—64 internal, 62 Parturition, 110-111 Penis, 86, 89-90, 106-108 Phylogeny, 132 Play, 43, 108-110, 113, 134 Postpartum estrus, 121 Radio tracking, 14 Refugia, v-wi, 17, 23 Crystal and Homosassa Rivers, 17-25 Blue Springs Run, 25-27 Regurgitation, 93, 95 Relationship with man, 124-131 Respiration, see breathing Response to boats, 127-128 Response to divers, 128-131 Resting, 82-84 Rooting, 90-93 Salinity: influence on manatees, 31, 38— 39 Scars, 14, 124 Scratching, 86—90 Sensory capacities, 115-120 Sexual behavior, 100-108 Sexual maturity, 122 Sight, 116-118 Smell, 119-120 Sneezing, 93 Social facilitation, 96 Social interactions, 95—114 Steller’s sea cow, v, 58, 97, 105 Storms: influence on manatees, 40 Stranding, 37-38, 101-102, 135 Stretching, 86 Sun: influence on manatees, 40-41 Swimming, see locomotion Hartman—Manatee in Florida 153 Tactile sensitivity, see touch Tagging, 14 Tail, 64-69 Taste, see chemoreception Thermoregulation, 23-24 Touch, 118 Trichechus, see manatee Turbidity: influence on manatees, 41 Twinning, 121 Vegetation: influence on manatees, 41 Vision, see sight Vocalizations, 96, 98-100, 105, 108, 113-114, 116, 135 Water: influence of currents, 39-40 influence of depth, 36-38, 55- 56 influence of salinity, 31, 38-39 influence of temperature, 18-25, 27 influence of tides, 38 influence of turbidity, 41, 49 influence of turbulence, 40 Weaning, 112-113, 121-122 Winter populations, 17-27 age composition, 16, 26 cold tolerance, 22—23 number of manatees, 15-16, 26 sex ratio, 16-17, 26 vit) Wh en f Ain or i cS in Bene Ve QL737.S63 H33 pees ae behavior SF the Mane 93272 il I) (ni DATE DUE NOW-A.6 onna aes HIGHSMITH #45230 Printed in USA ee Sa Sermtaeese ae Sa == are : - = = 2 oS Sees eect = 2 me = SS ok = 5 - E = ae ee te SSS = : SESS = ese. res : SSS ~ — SES Ss = Se Soe eee Se