Biological Conservation 19 (1980-81) 85-97

DISTRIBUTION A N D STATUS OF MANATEES

T R I C H E C H U S SPP. NEAR THE MOUTH OF THE
A M A Z O N RIVER, BRAZILt

DARYL P. DOMNING

Department of Anatomy, College of Medicine, Howard University. Washington, DC 20059, USA

ABSTRACT

Interviews with hunters and collection 4[ skul& indicate that Trichechus inunguis
occurs throughout the region o f the Amazon River estuaries f?om Amapd to the
mainland of" Pard, including llha de Maraj6 and islands on its Atlantic coast. T.
manatus has a disjunct distribution in Brazil, apparently occurring both on the coast
o f Amapd north o f Cabo Norte and in the Rio Mearim ( Maranhdo) as well as further
to the southeast; it seems to have been exterminatedj)'om the Atlantic coast o f Para
and is absent f r o m the Maraj6 region. Continued subsistence hunting can best he
controlled by destruction o f illegal camboas (fi, nee-like traps) which catch manatees
at high tide. A reas deserving study are the coasts and inland lakes of eastern A mapd,
the onO'place in the worm where two sirenian species might still be.[bund in sympato'
or immediate proximity, amt the lower Rio Mearim in Maranh~o, whieh may still
contain a sizable population o f T . manatus. Ecological studies in these areas shouM
have particular relevance to hypotheses ~[ sirenian evolutiona O' interaetions, and
manatee reserves shouM he established in both areas.

INTRODUCTION

The Order Sirenia appears to have reached its maximum diversity in the Miocene,
with about a dozen known genera, but subsequently declined and is represented
today by only two living genera and four species. The reduced diversity of the
modern sirenian fauna is a handicap to understanding the biology of the more
diverse Tertiary sirenians and the evolutionary patterns of sirenians in general.
Observations of sympatric or contiguously-occurring living forms would be of great

t Contribution No. 9 from the Department of Aquatic Mammal Biology. INPA, Manaus, Brazil.
85
Biol. Conserv. 0006-3207/81/0019-0085/$02'25 ~i Applied Science Publishers Ltd, England, 198 I
Printed in Great Britain
86 D A R Y L P, D O M N I N G

value to studies of sirenian evolutionary biology, but only one area in the world
today provides any hope of finding two sirenian species living in sympatry.
This area is the mouth, or mouths, of the Amazon river system in Brazil. The
Amazonian manatee, Trichechus inunguis (Natterer 1883), is distributed in fresh
waters throughout the Amazon basin, including the lower reaches of the Amazon
(Rio Amazonas) itself (Bertram & Ricardo Bertram, 1973). The range of the West
Indian manatee Trichechus manatus Linnaeus 1758, apart from North and Central
America and the Caribbean Islands, includes fresh and salt waters of Colombia,
Venezuela, the Guianas, and the coasts of the Brazilian Northeast (that is, on both
sides of the Amazon estuaries) (von Pelzeln, 1883; Bertram & Ricardo Bertram,
1973; Banks da Rocha, 1971a,b) as well as the lower Orinoco River (Mondolfi,
1974). Despite repeated statements in the literature, which originate from a report by
von Humboldt & Wiegmann (1838), there is no good evidence that T. inunguis ever
occurred in the Orinoco basin, nor does it appear likely (in view of the locations of
rapids in both rivers) that manatees of either species exist in the vicinity of the
Casiquiare Canal or other natural connections between the Amazon and Orinoco
drainage systems. Thus the Amazon mouth is the only area where the two species
might be sympatric.
Goeldi & Hagmann (! 904) reported T. inunguis from llha de Maraj6. Goeidi (in
Dilg, 1909: plate 13) presented a map of manatee distribution in South America,
showing T. inunguis in the Amazon basin and T. manatus on the coast of Amap~i
north of Cabo Norte. He believed that the manatees on the coast south of the
Amazon were T. inunguis, but cited no specimens or other data to support this
(erroneous) conclusion, which was nonetheless repeated by subsequent writers.
Pereira (1944), though asserting the existence of manatees throughout the area of
the present study, including the Rio Mearim, not only cited no evidence for his
conclusions but considered all New World manatees referable to T. manatus!
Finally, Cabrera (1961) suggested that the coastal manatees of the Brazilian
Northeast might be T. manatus, in contrast to the fluviatile inunguis. This
identification has been confirmed for the present-day situation by Banks da Rocha
(1971a,b) and for the 16th and 17th centuries by Whitehead (1977, 1978). De
Carvalho & Toccheton (1969) and Pine (1973) reported a specimen of T. inunguis
(M PEG 1518; see below) from near Bel6m. Apart from these, however, no published
locality records of either species seem to be available from the Atlantic coasts of
Parh or Amapfi, the lower parts of the Rios Amazonas or Tocantins, or Ilha de
Maraj6, leaving in doubt whether their ranges overlap in this area. (Hybridisation is
unlikely as T. inunguis has 56 chromosomes whereas T. manatus has 48; Loughman
et al., 1970; White et al., 1976.)
The results reported below were gathered in a survey of sites on the coasts of
Amap~ and Pardi, the lower Rio Mearim in Maranh~o, and the eastern part of llha de
Maraj6; that is, the areas where the identity of any existing manatee populations
seemed to be most in doubt. The main objective was to determine the present
 DISTRIBUTION AND STATUS OF MANATEES, AMAZON RIVER, BRAZIL 87

distributions of the two species in this region and identify areas of possible sympatry.
The major motivation for locating such areas was their potential value for studies
aimed at testing the hypotheses of sirenian evolutionary interactions proposed by
Domning (1977). Secondary aims were to determine the status of the manatee
populations and the outlook for their conservation, and to collect incidental data on
manatee natural history.

MATERIALS AND METHODS

From 6 May to 8 June 1978 I travelled through coastal regions of Parfi, Ilha de
Maraj6, Amapfi and Maranhfio (Fig. I) to interview fishermen and other residents
and to collect bones and other data relating to manatee distribution and status.
Useful information was obtained from about 28 persons, including 14 who claimed
to have personally killed manatees in this region. Three captive manatees from
known localities in the study area were examined out of the water, and diagnostic
portions of four skulls were collected: the latter are deposited in the collection of the
Departamento Peixe-Boi, Instituto Nacional de Pesquisas da Amaz6nia, Manaus
(nos. INPA-PB 178-180, 182). Another specimen (skull, skeleton, and skin) from
the study area is in the mammal collection of the Museu Paraense 'Emilio Goeldi'~
Bel6m ( M P E G 1518). Three additional specimens (IN PA-PB 129~ 156~ 185) from
localities marginal to the study area are also mentioned below.
Trichechus manatus and T. inunguis can be easily distinguished by a variety of
skeletal characters (Hartlaub, ! 886: Hatt, 1934: Domning, 1978 and in prep.). These
include the more complex molar pattern of T. inunguis and the conformation of the
latter's occipital region, which allows the skull to stand upright on the occiput when
placed on a flat surface, in contrast to most T. manatus. The latter species has 17 to
19 pairs of ribs, whereas T. inunguis has 14 to 16 (Goeldi & Hagmann, 1904; de
Carvalho & Toccheton, 1969; Hatt, 1934). The most conspicuous and reliable
external character is the presence of nails on the flipper margin in T. manatus and
their absence in T. inunguis. Reports by hunters that manatees lack nails are
unreliable for identification of T. inunguis, but unsolicited statements that manatees
had nails are good indications of T. rnanatus. The presence of white or pink belly
patches is usually thought to be a constant character of T. inunguis in contrast to
other manatees, but in reality T. inunguis shows a range of variation from complete
absence of ventral markings to elaborate series of patches extending from the throat
to the tail. Moreover, hunters' perceptions of'white marks on the belly' often include
mere countershading, whereas very small patches may be readily overlooked.
Therefore extreme caution must be used in identifying species from hunters"
testimony unsupported by diagnostic specimens.
In general, questions asked of informants included at least the following: Do
manatees occur in the area? How many kinds of manatees are recognised ? Do they
 u~
0 ..0 0
_.J

Z
r~
<
<

~°

,<
E
o.

C~

9!~oW ep
o
,-,,i

-0
o~
I--.- _o~ .~D ~

ouo!xeW oNIIl
soaod sop O~lllq

ouo,^o~o~,1
o~ .<
o.~

, 0 0

;.=~
<
,<
Q_

<

-._

NvIn©
 DISTRIBUTION AND STATUS OF MANATEES, AMAZON RIVER, BRAZIL 89

have white marks or areas on the belly or nails on the flippers? To establish the first-
hand nature of the information, the interviewee was always asked if he had
personally killed or seen a manatee: this generally seemed to be answered honestly.
Persons appearing to have more extensive knowledge were also questioned about
manatee habits, food plants, hunting practices, and related topics.

DISTRIBUTION

Territory of Amaph
The fishing village of Amapfi was briefly visited, with inconclusive results.
Residents claimed that manatees were now seldom caught. Two hunters reported
killing adult manatees in the vicinity of the Rio Amapfi Grande, in ' M a y or June
1977' (a male) and 'about 5 years ago' (c. 1973), respectively. Two others killed
several each, between 15 and 30 years ago at l lha de Maracfi offthe Amapfi coast, but
another longtime resident of IIha de Maracfi assured me that none had been killed
there in years. No diagnostic specimens could be obtained, but when I enquired
about the occurrence of manatees with nails, one resident said he had seen a calf with
nails that had been caught in the region of Cunani about 2 years before. On 28 July
1978, R. C. Best collected limb bones of a male manatee said to have had nails
( I N P A - P B ! 85) from Taperebo, Rio Oiapoque. The overall shape of the radius-ulna
and the ratio of length of the fourth metacarpal to that of the radius (about 0.57: cf.
Hatt, 1934) agree best with T. manatus. Manatees of undetermined species are said
to occur on the coast near Cabo Norte and in lakes in eastern Amapfi, especially
Lago N o v o and Lago Comprido. A live female T. inunguis 1.65 m long, which had
been caught near Macapfi circa 1969, was examined in the latter city; this animal had
a single, very small white belly patch only ~ 10 cm long. A resident of Ilha Caviana in
Parfi claimed to have seen a manatee 2 m long with 'white on the belly' which was
caught 'about 8 years ago' (c. 1970) at Ilha do Juru~ south of Cabo Norte. It thus
appears that T. manatus occurs north of Cabo Norte and T. inunguis to the south,
with a possible zone of contact or sympatry near Cabo Norte and in the interior
lakes.

llha de Maraj6
I travelled up the Rio Arari, through Lago Arari and Canal das Tartarugas, and
along the Atlantic coast to Ilha Caviana, Ilha Mexiana, and Ilha das Pacas, and
interviewed residents at several localities. Canal Perigoso, separating the latter three
islands, was the most active centre of manatee hunting visited in the study, and was
widely recognised as such by inhabitants of the region; at least 13 manatees were
killed there in 1977, and three had been caught (plus one which escaped) in 1978 as of
May. The three 1977 animals whose skulls were collected ( I N P A - P B 178-180) were
all T. inunguis.
90 D A R Y L P. D O M N I N G

A hunter on the Rio Arari caught two manatees above Cachoeira do Arari in
1977: a male calf of 14-16 kg in January or February (skull fragments and other
bones collected, INPA-PB 182, T. inunguis); and a female on 25 September. The
latter was still living in captivity and proved to be a T. inunguis 2-05 m long, with a
very small (6.5cm long) white belly patch.

BelOm area
An adult male T. inunguis (M PEG 1518), having 14 pairs of ribs and no nails, was
caught at Icoraci on 20 August 1958 (de Carvalho & Toccheton, 1969; Pine, 1973).
Another manatee, caught in a fishing net at Ilha do Arapiranga circa March 1978,
was still in captivity and proved to be a male T. inunguis 1.74m long, with no belly
patches. Two other captive T. inunguis in Bel+m, at the Museu Paraense 'Emilio
Goeldi' and the Bosque Rodrigues Alves, have no locality data.

Atlantic coast of Pard

Elderly fishermen in S~o Jo~o de Pirabas and Salin6polis agreed that manatees
were absent from the area. A 78-year-old man said there had been none there in his
lifetime. A 63-year-old man said a couple were seen when he was about 15 (that is, c.
1930), and another said one was killed some 20 years ago. Inquiries made on my
behalf by Dr W. L. Overal in Braganqa and Viseu produced the same negative result.
This region has been densely settled and developed for many decades, and despite
the persistence of a town and a river named Peixe-Boi (manatee), the manatee
population seems to have been largely exterminated. Specimens from archaeologi-
cal sites could serve to identify the species which formerly occurred here; but the
Museu Goeldi's large collection of faunal remains from coastal middens in Par/l
consists almost entirely of fish (Dr Mhrio Sim6es, pers. comm.), with no manatee
bones except for one possible rib fragment.

Lower Amazon river and tributaries

These areas were not included in the interview survey. All known manatee
specimens from the interior of the Amazon basin are T. inunguis; however, few
specimens have been collected from the region below Santar6m. The Museu Goeldi
possesses seven skulls of T. inunguis from near Taperinha, just below Santar~m,
collected in 1910-1933 by G. Hagmann. Skull and bone fragments resembling T.
inunguis (IN PA- PB 129), from an adult male manatee killed in August 1974 and said
to have had a white belly, were collected near the mouth of the Rio Paru by J.
Lovisek. Manatees reportedly occur in the lower parts of the Rio Xingu and Rio
Tocantins; I presume these to be T. inunguis though I have not been able to examine
specimens from these localities.

Maranh6o
Having failed to find recent evidence of manatees on the coast of Par~, I went on
to Maranh~o in an effort to determine how close to the south side of the Amazon
 DISTRIBUTION AND STATUS OF MANATEES, AMAZON RIVER, BRAZIL 91

mouth they could still be found. Although manatees, evidently T. manatus, occurred
near Silo Luis in the early 17th century (Walter, 1967; Whitehead, 1977), none had
been reported more recently. The coastal region was not visited for lack of time, but
an excursion up the lower Rio Mearim established the continued presence of the
species. Only an indeterminate rib fragment (INPA-PB 156), from an animal killed
circa 1975 1976, could be collected; however, a resident of the village of Coivara
( = Boqueirfio) stated that in about 1973 he had helped kill a male manatee in that
area which had had nails. As the latter detail was volunteered before I or anyone else
present had mentioned or asked about nails, I consider this reliable evidence of the
presence of T. manatus.

FOOD PLANTS

As the hunters interviewed generally claimed to have little specialised knowledge of

manatees and their habits, they could provide few ecological or behavioural data
apart from names of local plants eaten by manatees. In Amapfi, these were said to be
principally cai-seca, a vinelike herb with white flowers (Rhabdadenia biflora (Jacq.)
M. Arg., Apocynaceae), andparaturd, a type of'beach grass', which I was unable to
collect. At llha Caviana, principal food plants reported were intertidal 'beach
grasses' such as barba de bode (Fimbristylis capillaris (L.) Kuntz, Cyperaceae),
.junco ( Cyperus articulatus L. var. nodosus, Cyperaceae), capita de botdo ( Cyperus
luzulae Retz), tiririca (Dichromena ciliata Vahl, Cyperaceae), and grama da praia
(Panicum polycomum Trin, Gramineae), as well as fruits and leaves of aninga
(Montrichardia arborescens (L.) Schott, Araceae) and fruits of genipapo (Genipa
americana L., Rubiaceae). One informant, however, denied that the Cyperus spp.
were eaten. Here, as in Maranhfio and elsewhere in Amazonia, manatees are also
generally acknowledged to eat murur~, which denotes non-gramineous floating
aquatics in general but especially Eichhornia (Pontederiaceae).

HUNTING METHODS AND UTILISATION OF MANATEES

In most parts of Amazonia manatees are usually hunted with a harpoon, most
commonly from a canoe. On the Atlantic coast of Maraj6, however, hunters take
advantage of the rise and fall of the tides in employing two other methods. The
shores in many places are covered with 'meadows' of Fimbristylis, Cyperus,
Dichromena, Panicum, and other sedges and grasses, which are submerged at the
highest tides and grazed by manatees. With the first method, a three-legged wooden
platform or montar 2-2-5 m high is erected in one of these meadows (Fig. 2). At high
tide the hunter waits atop the platform until a manatee passes close enough to
harpoon; no bait is used, in contrast to the otherwise similar technique used in, for
example, Nigeria (Sikes, 1974). With the second method, up to several hundred
92 DARYL P. DOMNING

Fig. 2. Wooden platform or montar used for harpooning manatees on Ilha Caviana. Note meadow of
Fimbristylis, Cyperus, and other sedges and grasses fed on by Trichechus inunguis at high tide.
 DISTRIBUTION AND STATUS OF MANATEES, AMAZON RIVER, BRAZIL 93

metres of shoreline are enclosed by a fence of wooden stakes about 1.5 m high, called
a c a m b o a or curral (Fig. 3). At high tide the fence is completely submerged, and
manatees, fish, and other aquatic animals passing over it to feed in the intertidal zone
are trapped when the tide recedes. As this device works automatically and requires
no skill or patience on the part of the hunter, or even his presence, it is far more
destructive to the manatee population than are individual harpooners; of the 13
manatees reported killed in Canal Perigoso in 1977, 11 were caught in camboas (up
to four in a single c a m b o a during the season).

Fig. 3. Wooden fence or camboa enclosing part of grassy shoreline of llha das Pacas in Canal Perigoso.
At least five Trichechus inunguis were caught in this camboa in 1977 78.

Fortunately, use of camboas is said to be declining, for various reasons. An

informant estimated that three men could build one in about two weeks that would
last up to two years; but such structures are vulnerable to storms and drifting logs
and require upkeep. Some fishermen say they are too destructive to fish populations;
others, that the increased availability of nylon nets allows them to go where the fish
are in preference to relying on the stationary camboa. One man was said to have
stopped building camboas because cattle would get in and eat the grass; another,
because an aggrieved neighbour had performed black magic at the site of his c a m b o a
94 DARYL P. DOMNING

to spoil his luck. Such camboas are also illegal (see below). Their use on the Amapfi
coast is said to have ended.
Utilisation of manatee carcasses is largely confined to the meat and fat, which are
consumed locally. Sausages may be made from the intestines, and the distal parts of
the flippers are sometimes used in a dish called manhiqoba. Mixira, fried meat
preserved in its own fat, is sometimes made for storage or local sale. In most of
Amazonia, the hide is now normally discarded; but throughout the area of this
study, dried pieces are often saved. Softened into a paste by boiling, they are used as
plasters for wounds, hernias, and pulled muscles. Manatee fat is said to be good for
rheumatism. The bones are reputed to have medicinal value, but no one I talked to
seemed to know what they were supposed to cure. One hunter on the Rio Mearim
described how a section of rib with a slot cut in one end was sometimes used as'pliers'
for breaking fish bones.
Strangely, reports of kills prior to 1977 in Canal Perigoso (presently the most
active centre of hunting) were few and vague, the most definite pertaining to a male
manatee about 2 m long caught at Ilha Caviana 'about ten years ago' (c. 1968); the
two most active hunters I interviewed denied having killed any before 1977.
Although I hesitate to accept this at face value, the wife of one longtime resident of
llha Caviana admitted being unacquainted with manatees until 1977. She said they
are not killed every year because the local people are not very keen on eating
manatee meat, believing that it is not healthy and that pregnant women should not
eat it. One hunter sold much of the meat of the two manatees he harpooned in 1977
to the local landowner, and another who caught four in a c a m b o a took them alive to
Macapfi for sale.

POPULATION STATUS AND CONSERVATION

It is impossible to estimate actual numbers of manatees still occurring in this region.

Hunting pressure varies from light to fairly heavy: on the Atlantic coast of Amapfi
perhaps no more than two or three a year are taken; in Canal Perigoso at least 13
were caught in 1977 and three in early 1978; at Ilha Ganhofio, at least one each in
1977 and 1978; on the Rio Arari, two in 1977: in the Baia de Maraj6 near Bel6m, at
least four or five in 1977-78; on the Rio Mearim, possibly one or two a year. Some of
these may be underestimates, and it is safe to assume that hunting at a comparable or
greater rate occurs throughout the Maraj6 region as it does in the rest of Amazonia.
Population trends likewise cannot be documented with available information.
Some hunters at Canal Perigoso thought manatees were less abundant than
formerly, but only because progressive silting-up of the coastline had caused them to
feed farther from the hunters' homes.
Of a total of 19 animals of known sex reported taken in the study area, 15 were
males, so there does not appear to be any disproportionately high mortality of
 DISTRIBUTION AND STATUS OF MANATEES, AMAZON RIVER, BRAZIL 95

females. (It is noteworthy, and possibly significant, that of 15 T. manatuscaptured in

Surinam by Dekker (1974), 14 were likewise males.) Only three (one male, two of
unknown sex) of 22 for which hunters gave size estimates were described as calves.
Although manatees have been protected by law in Brazil since 1967, effective
enforcement is impractical. In addition to this blanket protection, however,
construction o f c a m b o a s and similar structures by persons not officially licensed as
breeders of aquatic animals is prohibited by decree of the fisheries service (Portaria
S U D E P E No. 310, Article 12, 23 July 1973). As camboas, in contrast to montars,
are both conspicuous and time-consuming to construct, they could easily be found
and destroyed by periodic inspection patrols during the January-June high-water
season when nearly all manatee hunting occurs. As camboas appear to play a minor
and already-diminishing role in the local economy, such measures would place
minimum strain on both enforcement agencies and local inhabitants while being of
maximum benefit to the manatee population.
In addition, at least two parts of the study area should be given high priority as
permanent manatee reserves. The area of eastern Amap~i including Cabo Norte and
the adjacent coasts and inland lakes appears to be the only remaining part of the
world where two sirenian species might occur in close proximity, and it certainly
deserves further study in the near future. The lower Rio Mearim in Maranh~o
appears to harbour a population of T. manatus (one resident claimed to have seen
two manatees a few hours before our interview); it certainly has large areas of superb
manatee habitat, with vast and undisturbed floating meadows in lakes and channels
off the main river. As suitable habitat is more restricted in the coastal lagoons and
rivers of the Brazilian Northeast (Piaui to Bahia), the Rio Mearim may well have the
largest T. manatus population remaining in Brazil. The Rio Mearim appears very
similar in vegetation and other features to many parts of the Amazon basin, and
would be of special value for comparative studies of the two manatee species in
comparable habitats.

CONCLUSIONS

1. Trichechus inunguis is found throughout the lower Amazon and its

tributaries, including Ilha de Maraj6 out to its Atlantic coast and coastal
islands, and from the northern (Amap~) to the southern (Par~) shore of the
Amazon estuaries. This entire area is hydrologically and phytogeographi-
cally part of the Amazon basin, with little or no influence of salt water even
on coasts open to the Atlantic; however, unconfirmed reports suggest that T.
inunguis may sometimes also be found in salt water.
2. Trichechus manatus probably occurs on the AmapS. coast somewhere north
of Cabo Norte and in the Rio Mearim in Maranh~.o; it appears, however, to
have been exterminated from the Atlantic coast of Par~, assuming that it,
96 DARYL P. DOMNING

and not T. inunguis, formerly occurred there. Its distribution on the

Brazilian coast is disjunct, the Amapfi population being separated from
those in Maranhfio and the Northeast by populations of T. inunguis
inhabiting the mouths of the Amazon and llha de Maraj6. The two species
may coexist in eastern Amapfi.
. We may tentatively conclude that any zone of sympatry between T. manatus
and T. inunguis is very restricted, if it exists at all, and that for practical
purposes the species have mutually exclusive distributions. This has
important implications in the context of the evolutionary hypotheses of
Domning (1977), and poses several interesting theoretical questions, which
will be discussed elsewhere.
. Sizes and trends of manatee populations in the study area are unknown.
lllicit subsistence hunting continues, especially in Canal Perigoso on the
coast of llha de Maraj6. This hunting can be most effectively controlled by
systematic destruction of camboas, which would require minimal effort by
enforcement authorities and probably cause minimal dislocation of the local
economy.
. Manatee reserves should be immediately established to include the Amap~
coasts near Cabo Norte and adjacent inland lakes, as well as the lower Rio
Mearim valley in Maranhfio.

ACKNOWLEDGEMENTS

This study is part of an ongoing research programme on the biology of the

Amazonian manatee at the lnstituto Nacional de Pesquisas da Amaz6nia, Manaus,
Brazil. The programme is supported by the Conselho Nacional de Desenvolvimento
Cien tifico e Tecnol6gico (CN Pq) of Brazil, and by grants from the Vancouver Public
Aquarium, British Columbia, Canada, and the Fauna Preservation Society of Great
Britain. I thank Drs W. E. Kerr and W. Junk of INPA for organizational aid; Dr
L. M. Scaff for making available to me the facilities of the Museu Goeldi; Dr P. B.
Cavalcante for plant identifications; and Drs F. Novaes, O. Santos, and M. Sim6es
for other assistance. Dr Raimundo Dias, Dr Camilo Viana, Sr Liberato Castro, Sr
and Sra Pedro Dantas, Dr A. Farias, Dr M. B. da Rocha Araujo, Sr U. J.
Guimar~es, and Sr J. J. de Miranda Coelho (mayor of Amap~) all provided
invaluable aid in organizing and carrying out field work. Dr Scarf, Sr L. A. do
Santos (lnstituto Brasileiro de Desenvolvimento Florestal~ Macap;i)~ Sr M. T. de
Almada (Bosque Rodrigues Alves, Bel~m), Sr L. Castro, and Sr S. Vasconcelos
permitted examination of captive manatees in their custody. Most of all I thank Dr
W. L. Overal of the Museu Goeldi and Sr Jos6 Mascarenhas of Macap~ for their
tireless exertions in furtherance of this study. I thank Drs R. L. Brownell, Jr, B.
lrvine, K. Rails, and G. Rathbun for helpful criticisms of the manuscript.
 DISTRIBUTION AND STATUS OF MANATEES, AMAZON RIVER, BRAZIL 97

REFERENCES

BANKS DA ROCHA, N. (1971a). Mem6ria sobre u m exemplar de Trichechus manatus L., 1758, capturado
em Goiana (Pernambuco). Archos Mus. n a c , Rio de J., 54, 101 3.
BANKS DA ROCHA, N. ( 1971 h). Nora pr6via sobre a ocorr6ncia de sir6nios no Nordeste. Anais Inst. Cidm'.
Biol. (Recife), !(I), 133.
BERTRAM,G. C. L. & BERTRAM.C. K. RICARDO(1973). The modern Sirenia : Their distribution and status.
Biol. J. Linn. Soc., 5, 297-338.
CABRERA, A. (1961). Catalogo de los mamiferos de America del Sur, I1. Revta Mus. argent. Ciem'. nat.
Bernardino Rivadavia, Cienc. Zool., 4, 309-732.
DE CARVALHO, C. T. & TOCCHETON, A. J. (1969). Mamiferos do nordeste do Para, Brasil. Revta Biol.
trop., 15, 215-26.
D~KKER, D. (1974). On the natural history of manatees ( Trichechus manatus manatus) from Suriname for
the A m s t e r d a m Zoo. Aquatic Matures., 2(2), 1 3.
DILG, C. (1909). Beitr/ige zur K e n n m i s der Morphologie und postembryonalen Entwicklung des
Sch/idels bei Manatus inunguis Natt. Morph. Jh., 39. 83- 145.
DOMNING, D. P. (1977). An ecological model for Late Tertiary sirenian evolution in the North Pacific
Ocean. Syst. Zool., 25, 352-62.
DOMNING, D. P. (1978). The myology of the A m a z o n i a n manatee, Trichechus inunguis (Natterer)
(Mammalia: Sirenia). Acta Amazonica, 8(2), Suppl. 1, 1-80.
GOELDI, E. A. & HAGMANN, G. (1904). P r o d r o m o de u m catalogo critico, c o m m e n t a d o da collefa6 de
mammiferos no Museu do Parfi. (1894.1903). Bol. Mus. para. Emilio GoeMi Hist. nat. Ethnogr_ 4,
38-106, 119-22.
HARTLAUB, C. (1886). Beitr~ige zur Kenntniss der Manatus-Arten. Zool. Jh. I, Abt. Syst. Geogr. Fiol.
Thiere, 1, 1-112.
HATT, R. T. (1934). A manatee collected by the American M u s e u m Congo Expedition, with observations
on the Recent manatees. Bull Am. Mus. nat. Hist., 66, 533 66.
LOUGHMAN,W. D., FRYE, F. L. & HERALD, E. S. (1970). The c h r o m o s o m e s of a male manatee, Trichechu.s
inunguis. Int. Zoo Yb., 10, 151-2.
MONDOLFI, E. (1974). T a x o n o m y , distribution and status of the manatee in Venezuela. Meres Soc. Cienc.
nat. "La Salle', 34, 5 23.
PEREIRA, M. N. (1944). O peixe-boi da Amaz6nia. Bolm Minist. Agric. ind. Com. Rio de J., 33(5), 21 95.
PINE, R. H. (1973). M a m m a l s (exclusive of bats) of Bel6m, Par~, Brazil. Acta Amazonica, 3(2), 47 79.
SIKES, S. (1974). How to save the mermaids. Oryx, 12, 465-70.
VON HUMBOLDT,A. & WmGmANN, A. F. A. (1838). Ueber den Manati des Orinoko. Wiegmann's Arch.
Naturgesch., 4, 1 18.
VON PELZELN, A. (1883). Brasilische Sfiugethiere. Resultate von Johann Natterers Reisen in den Jahren
1817 bis 1835. Verh. zool. bot. Ges. Wien, Beiheft, 33, 1-140.
WALTER, J. (1967). Frei Crist6vdo de Lisboa: Histbria dos animais e 6rvores do Maranhdo. Lisbon, Arq.
Hist. Ultramar.
WnrrE, J. R., HARKNESS, D. R., ISAACKS,R. E. & DUFFIELD, D. A. (1976). Some studies on blood of the
Florida manatee, Trichechus manatus latirostris. Comp. Biochem. Physiol. A, 55, 413 17.
WrII~HEAD, P. J. P. (1977). The former southern distribution of New World manatees ( Trichechus spp.).
Biol. J. Linn. Soe., 9, 165-89.
WmTEHEAD, P. J. P. (1978). Registros antigos da presenqa do peixe-boi do Caribe (Trichechus manatus)
no Brazil. Acta Amazonica, 8, 497 506.