Evolution of the Sirenia
The
order Sirenia is closely associated with a large group of hoofed mammals known
as Tethytheria, which includes the extinct orders Desmostylia (hippopotamus-like marine mammals) and Embrithopoda (rhinoceros-like mammals). Sirenians probably split off from these
relatives in the Palaeocene (65-54 mya) and quickly took to the water, dispersing to the New World. This outline attempts to order all the
species described from the fossil record in chronological order within each of
the recognized families of Prorastomidae, Protosirenidae, Dugongidae, and
Trichechidae. This outline began as an
exercise in preparation for my Ph. D. preliminary exams and is primarily based
on decades of research and peer-reviewed literature by Daryl P. Domning, to
whom I am eternally grateful.
Ancestral
line: Moeritherium
Order Proboscidea
Elephantidae (elephants
and mammoths)
Mastodontidae
Deinotheriidae
Gomphotheriidae
Ancestral line:
Paleoparadoxia
Order Desmostylia (only known extinct Order of marine mammal)
Order Sirenia Illiger,
1811
Prorastomidae
Protosirenidae
Dugongidae
Trichechidae
With
only 5 species in 2 families known to modern man, you might be surprised to
learn that the four extant species represent only a small fraction of the
sirenians found in the fossil record. As
of 2006, ~50 species have been described and placed in 4 families, including
the modern species. Sirenians, as well
as the seagrasses (their primary food item) probably originated in the the Tethys Sea area, the ancient sea that separated
Gondwanaland from Laurasia. Morphologically and molecularly they are
grouped with proboscideans (elephants) and
desmostylians in a taxon called Tethytheria. Sirenians probably share an Old World origin
with these and other orders such as the Embrithopoda
and Hyracoidea (hyraxes). After returning to the sea, they quickly
“dispersed throughout the Tethyan relm and have remained pantropical
ever since with the notable exception of the lineage leading to Steller’s
seacow (Hydrodamalis), which adapted to the temperate and cold waters of the
North Pacific. This persistent tropicality very likely accounts for the depleted diversity
of the Sirenian, given the global cooling of the last few million years.” (Domning 2001b).
Sirenians
first appear in the fossil record about 50 mya in the
Middle Eocene Epoch (image left is (c) Christopher Scotese,
http://www.scotese.com), early Cenozoic Era.
The Cenozoic is the most recent of three major subdivisions of animal
history; a convenient Epoch Key is found in the footer of this document (PDF
and DOC versions). Prior to the
Cenozoic, there were the Paleozoic Era (age of the inverts, fishes, &
amphibians) and the Mesozoic Era (age of the reptiles & dinosaurs). The Cenozoic Era is short, relative to the
two previous Eras, spanning only about 65 million years, from the end of the
Cretaceous Epoch (marked by the extinction of non-avian dinosaurs) to the
present. The Cenozoic Era is often called
the Age of Mammals, but it could just as easily be called the Age of the Birds,
Insects, Flowering Plants, or Teleost Fish.
Nomenclature omitted from this outline (because I
could not place them within a family based on my readings): Sirenavus hungaricus Kretzoi,
1941 possibly referred to the Prorasomidae (Domning
1978) and Anisosiren pannonica Kordos, 1979 are both from the Middle Eocene Hungary, Paralitherium tarkanyense Kordosm 1977 Late Eocene Hungary, Ishatherium subathuensis Sahni
and Kumar, 1980 (Domning, Morgan & Ray 1982); nomen nudum Trachypleurotherium
(Domning 1982a.); Florentinoameghinia mystica
Simpson, 1932a from the Early Eocene of Patagonia, referred to the Sirenia by Sereno (1982), is considered Mammalia incertae sedis (Domning 2001a); Lophiodolodus chaparralensis is best regarded as Mammlaia incertae sedis, possibly Sirenia (Domning 1982a); Thalattosiren petersi (Abel, 1904) may represent merely immature Metaxytherium (Domning 1994); Prohalicore?
1.
Family Prorastomidae Cope, 1889 (Middle Eocene): Oldest fossil records of sirenians, the
taxonomic family of prorastomids, were amphibious quadrupeds that resembled
Paleocene and Eocene condylarths, but had aquatic
specializations such as retracted nasal openings, absence of paranasal air sinuses, and dense and swollen ribs. The most primitive prorastomids had adequate
support of body weight through the sacroiliac joint to provide for terrestrial
locomotion. Shared characteristics
included: (1) undeflected,
laterally compressed, anterior skull and mandible; and
(2) well-developed hind legs.
Prorastomids had almost straight (undeflected)
rostra, whereas all later sirenians had more or less down turned snouts (for
bottom feeding).
1.1. 
Pezosiren portelli Domning, 2001 (from Jamaica)1: Oldest know sirenian in the fossil record;
named and described by Domning in 2001.
“This animal was fully capable of locomotion on land, with four
well-developed legs, a multivertebral sacrum, and a
strong sacroiliac articulation that could support the weight of the body out of
water as in land mammals. Aquatic
adaptations show, however, that it probably spent most of its time in the water. Its intermediate form thus illustrates the
evolutionary transition between terrestrial and aquatic life. Similar to contemporary primitive cetaceans,
it probably swam by spinal extension with simultaneous pelvic paddling, unlike
later sirenians and cetaceans, which lost the hindlimbs
and enlarged the tail to serve as the main propulsive organ. Together with fossils of later sirenians
elsewhere in the world, these new specimens document one of the most marked
examples of morphological evolution in the vertebrate fossil record.” Domning, D. P. 2001. The earliest known fully quadrapedal
sirenian. Nature 413:625-627.
1.2. 
An un-named genus and species of Prorastomidae from
the Late Early or early Middle Eocene, Jamaica Guys Hill Member, Chapelton Formation (Domning et al., 1995, 1996 as reported
in Domning 2001a)
1.3. Prorastomus sirenoides Owen,
1855 (Middle Eocene from Jamaica):
Reduced sacroiliac articulation, but could probably still support its
body weight out of the water. Image at right is artist’s conception of Prorastomus sirenoides submitted
by Daryl P. Domning
2.
Family Protosirenidae
Sickenberg 1934: Middle Eocene Protosirenids, the next most
primitive sirenian family, retained well-developed (through reduced) hind
limbs, but had weak sacroiliac joints (interpreted as unable to support body
via hind limbs on land?). The genus
Protosiren represents one of the more widely distributed Eocene genera, ranging
across the southern part of the eastern Tethys Sea from North Africa (Egypt) to
South Asia (Indo-Pakistan). Compare
image below to Eocene map on previous page.
Protosiren is known from skulls and partial skeletons (Abel, 1907;
Domning and Gingerich, 1994; Gingerich et al., 1994, 1995, Zalmout et al.,
2003). They were mainly aquatic animals
that probably spent little if any time on land.
Shared characteristics included:
(1) well-developed hind legs; (2) more or less down-turned rostrum suitable for
bottom feeding; (3) broadened mandibular symphysis
indicative of less selective grazing; and (4) reduction and eventual loss of
incisors and canines (except tusks).
Both prorastomids and protosirenids were evidently extinct by the end of
the Eocene, ~38 mya.
2.1. Genus Protosiren
Abel, 1907 (from Egypt, North America?, Europe, Asia)
2.1.1.
P. eothene (Zalmout et al. 2003, from early middle Eocene deposit
in Pakistan)
2.1.2.
P. fraasi Abel, 1907 (Early Middle Eocene, India, Egypt,
Hungary?)
2.1.3.
P. sattaensis (Gingerich et al. 1995; late middle Eocene, 39-40 Ma,
Pakistan)
2.1.4.
P. smithae (Domning & Gingerich, 1994; direct descendent from
P. fraasi;
Late Middle Eocene, Egypt)
2.1.5.
P. minima Desmarest,
1822 (Middle-late Eocene, France –
not mentioned in Zalmout et al. 2003, current classification unknown?)
2.2. A new genus and species of protosirenid
from Florida and North Carolina Avon Park, Inglis,
and Castle Hayne formations and possibly another from
Florida Late Eocene Crystal River Formation (Domning 2001a)
3.
Family Dugongidae Gray, 1821 (Middle and Late Eocene to Recent): Early members of the taxonomic family of
dugongids were fully aquatic, with only vestigial hind limbs (i.e., loss of
functional hind limbs). Dugongids went
on to become the most diverse and successful sirenian family, and also may have
given rise to the trichechid family. The
Oligocene Epoch (38-23 mya) was dominated by the
dugongids, including the genera Halitherium, Metaxytherium,
Caribosiren,
(ALL subfamily Halitheriinae), Crenatosiren, Dioplotherium, and probably Corystosiren and Rytiodus, (ALL
subfamily Dugonginae) which have Early Miocene (and
thereafter) fossil records.
Subfamily Halitheriinae Carus, 1868: Halitheriines are not known in the Caribbean after the
Miocene, which ended 5 mya.
3.1. Genus Eotheroides Palmer, 1899 (Synonymous with Eotherium Owen 1875, nec Leidy
1853, Eosiren Andrews 1902, Archaeosiren Abel 1913, Masrisiren Kretzoi
1941): Mid Eocene, Mediterranean. Contained 5 nominal species?
Domning 1978.
3.1.1.
E. aegyptiacum Owen, 1875:
Middle Eocene, near Cairo, Egypt, later discoveries from the late Eocene
marine beds of Fayum.
3.2. Genus Prototherium de Zigno, 1887 (Mid
to Late Eocene, Mediterranean)
3.2.1.
P. veronense de Zigno, 1875 Late Eocene dugongid from Italy.
(includes Protosiren dolloi Abel 1904 = Mesosiren Abel 1906) The most primitive sirenian in which the juvenile
dentition is adequately known, with 5 being the primitive number of premolars
in sirenians, the last 4 of which are presumably homologous to those in other
placental mammals (Domning 1982a). There
is no fossil record of the juvenile dentitions of the manatees, making it all
the more striking to find in the Recent Trichechus
a pattern almost identical to that of P. veronese.
3.2.1.1. “P.” intermedium Bizzotto, 1983
3.3. Genus Eosiren Andrews, 1902 (Mid to Late Eocene, Mediterranean)
3.3.1.
E. abeli Sickenberg, 1934 (?synonymous w/
E. imenti, see
Domning 1994; Early Oligocene, Egypt)
3.3.2.
E. libyca (Eotheroides libycum?) Andrews, 1902 (see Domning
1994; Late Eocene)
3.3.3.
E. stromeri (Eotherium stromeri?) Sickenberg, 1934
3.4. Genus Halitherium Kaup, 1838 (extends to
Late Oligocene in Europe, and apparently the western Atlantic and Caribbean,
where it gave rise to the genus Metaxytherium)
3.4.1.
H. schinzii Kaup, 1838
3.4.2.
H. christolii Fitzinger, 1842 (here includes H. abeli Spillman,
1959 and H. pergense
Toula, 1899)
3.4.3.
H. antillense Matthew (1916):
Early to Late Oligocene, Puerto Rico Juana Diáz and San Sebastián
formations. May be synonymous w/ Metaxytherium)
3.4.4.
H. alleni
3.5. Genus Caribosiren Reinhart,
1959
3.5.1.
Caribosiren turneri Reinhart,
1959 (Late Oligocene, Puerto Rico)
3.6. Genus Metaxytherium
de Christol, 1840 (prominently represented in the Miocene
& Pliocene, sympatric with Crenatosiren and Dioplotherium in northern Florida, and sympatric with Dusisiren and Dioplotherium in Baja; the taxon Hesperosiren has
been placed in the synonymy of Metaxytherium)
3.6.1.
M. krahuletzi Depéret, 1895 (Early Miocene in Europe, probably direct
ancestor to all other Old World Metaxytherium;
relationship to New World Metaxytherium is
unclear)
3.6.2.
M. medium Desmarest,
1822 (Europe, direct descendent of M. krahuletzi) 2
3.6.3.
M. serresii Gervais, 1847 (Europe, direct descendent of M. medium) 2
3.6.4.
M. subapenninum Bruno, 1839 (Europe, direct descendent of M. serresii; synonymous
w/ M. forestii Capellini, 1872)
2
3.6.5.
M. crataegense (= M. calvertense Kellogg, 1966, = M. riveroi, American Atlantic,
dispersed to eastern Pacific Middle Miocene Baja California, Mexico, and Orange
County California (Aranda-Manteca et al. 1994).
Photo above submitted by Bonnie Abellera, Florida Fish & Wildlife
Conservation Commission, ~15 million year old skeleton was found in a quarry in
northwest Florida and donated to the state in 1929; currently
on display in the Geology office building at Florida State University.
3.6.6.
M. floridanum Hay,
1922 (American Atlantic). Metaxytherium floridanum skeleton and artist
conception below are on dislplay at the Smithsonian
Museum of Natural History in Washington, D.C. Photo submitted by Caryn
Self-Sullivan
3.6.7.
M. arctodites (Aranda-Mantea Francisco J. et al. 1994; direct link to new subfamily, the
Hydrodamalinae)
Subfamily Dugonginae Gray, 1821 (here includes Rytiodontinae Abel, 1914): Late Oligocene to Late Pliocene. The subfamily Dugonginae are the only dugongines
known from the Caribbean region’s Pliocene deposits. By this time the halitheriines
have disappeared from the Caribbean fossil record. Corystosiren is
present in Early Pliocene, and Xenosiren (descendent
from Dioplotherium)
may have been contemporaneous with it (Corystosiren). The ONLY Late Pliocene dugongid
know in the Caribbean region is a Dugong–like dugongine
with large tusks (# 8 Un-named, below). Its
immediate ancestry is unclear, but may have stemmed from one of the
un-described small dugongines mentioned above. By the end of the Pliocene, 1.8 mya, dugongids seem to have died out in the western
Atlantic altogether. Dugonginae
had the tendency to evolve large and often blade-like tusks, in contrast to the
smaller, sub-conical tusks of Oligocene and Miocene halitheriines,
and may have fed preferentially on the rhizomes of more robust sea grasses. This sub-family probably arose in the
Caribbean-West Atlantic region and underwent an adaptive radiation and pantropical dispersal in the Late Oligocene and Early
Miocene.
3.7. Genus Crenatosiren Domning,
1991 (Late Oligocene, SE USA)
3.7.1.
C. olseni Reinhart, 1976
3.8. Genus Dugong Lacépède, 1799
3.8.1.
Dugong dugon Müller,
1776. This is the only extant
dugong. Its modern distribution includes
37 countries, from SW Africa and Madagascar around the Indian Ocean coastline,
including the Red Sea, Gulf of Mannar between Indian and Sri Lanka, and Shark
Bay, Western Australia; to the Indo-Pacific islands, Torres Strait; and in the
Pacific it’s found south to Hervey and Moreton Bay
and north to Okinawa.
3.9. Genus Dioplotherium Cope, 1883
3.9.1.
D. manigaulti Cope,
1883 (Late Oligocene, Florida, divergently specialized and contemporary with Crenatosiren)
3.9.2.
D. allisoni (Kilmer, 1965):
Early Miocene, Brazil, California and Baja; Middle Miocene, Argentina; in
California this species was sympatric with H.
gigas ancestor, Dusisiren Kilmer, 1965
3.10.Genus Xenosiren Domning, 1989
3.10.1. X. yucateca Domning, 1989 (Late Miocene/Early Pliocene,
Yucatan, a direct descendant of Dioplotherium)
3.11.Genus Corystosiren Domning,
1990
3.11.1. C. varguezi Domning,
1990: Early Pliocene in Yucatan and
Florida, only, but lineage probably in the region throughout the Miocene;
unique in its extremely massive skull roof
3.12.Genus Rytiodus Lartet, 1866 (Europe,
North Africa, and in Brazil sympatric with D.
allisoni and Metaxytherium)
3.12.1. R. capgrandi Lartet,
1866
3.13.Genus Bharatisiren (India)
3.13.1. B. kachchhensis Bajpai, Singh, and Singh 1987: Probably arose in the
Caribbean-western Atlantic region and underwent adaptive radiation and pantropical dispersal in the late Oligocene and early
Miocene (Bajpai and Domning 1997)
3.14.New undescribed Genus (Late
Pliocene, Florida, close to living Indo-Pacific Dugong, latest dugongid known in West
Atlantic-Caribbean region.
Subfamily Hydrodamalinae Palmer, 1895 (1833):
Distinction due to new adaptive direction; departed from bottom-feeding,
increased body size, expanded into temperate to cold climates; adaptations
attributed to general cooling in the Pacific and replacement of seagrasses with
kelps. See “Sea Cow Family Reunion”, Daryl P. Domning, Natural History, April 1987.
3.15.Genus Dusisiren Domning,
1978
3.15.1.
D. jordani (Kellogg, 1925):
Middle Miocene (10-12 mya), California,
ancestor of D. dewana;
sympatric with Metaxytherium and Dioplotherium in
California. Length of one specimen = 4.3
m.
3.15.2.
D. dewana Takahashi, Domning, and Saito, 1986 (Middle Miocene (9-10
mya), Japan, perfect morphological and chronological
“link” between Dusisiren
and Hydrodamalis; see Domning 1987)
3.16.Genus Hydrodamalis
Retzius, 1794
3.16.1.
H. cuestae Domning, 1978 (here includes H. spissa Furusawa,
1988. Upper Pliocene Pismo Formation
California, California & Baja; immediate ancestor to Steller’s seacow;
largest sirenian skull on record; H. cuestae
represents the largest sirenians at up to 10 meters, they are larger than the
remnant population of H. gigas in the
Commander Islands)
3.16.2.
H. gigas (Zimmermann,
1780)3: Steller’s seacow; Late
Pleistocene-Recent. Extant in the North Pacific Commander Islands
during modern times; extirpated by humans for food. Last known kill in 1768. See note 4 for details.
Top:
Steller’s sea cow skeleton mounted in the Smithsonian Museum of Natural
History in Washington, D.C. Bottom: Steller’s seacow mounted next to modern dugong
for size comparison. Photos submitted by
Caryn Self-Sullivan
4.
Family Trichechidae Gill, 1872 (1821): This family arose
in the Late Eocene or Early Oligocene, about 38 mya, possibly
from within Dugongidae. During the Pliocene
(5-1.8 mya), trichechids in the form of Ribodon are known
in the Caribbean/Western Atlantic region from North Carolina and had apparently
expanded their range outside South America (where the family originated) by the
late Pliocene and with the end of the Pliocene Epoch, had become the only
surviving western Atlantic-Caribbean sirenians.
Ribodon
and its descendant Trichechus are
characterized by supernumerary molars, which continue to be replaced horizontally
throughout the animal’s life; this is an adaptation to eating the abrasive true
grasses (Gramineae), which constitute the principal
diet of manatees in South American rivers, their ancestral home. Nutrient runoff from the rising Andes during
the Late Tertiary is assumed to have greatly increased the abundance of these
true grasses and the biomass of freshwater plants in general (note: seagrasses are not true grasses, but
angiosperms). Trichechids may have been
restricted to coastal rivers and estuaries of South America until the Late
Miocene. Here they fed on freshwater
plants, while dugongids inhabited the West Atlantic and Caribbean marine waters
and exploited seagrass meadows. “The
Mio-Pliocene Andean orogeny (5-10 mya)
dumped large quantities of silt and dissolved nutrients into many South
American rivers, stimulating growth of aquatic macrophytes,
particularly true grasses. Manatees
adapted to this newly abundant but abrasive food source by evolving
supernumerary molars continually replaced throughout life, as in the
Mio-Pliocene form Ribodon
from Argentina.” DPD 1982. Prior to the Andean orogeny,
most of the western and central Amazon basin drained in the Pacific and was separated
(as it remains today) from the Magdalena (Potamorsiren), Orinoco, and La
Plata basins by drainage divides. The
Miocene orogeny closed off the Pacific entrance,
temporarily creating (in Mio-Pliocene times) an intially
brackish lake system with interior drainage.
Subfamily Miosireninae Abel, 1919:
“The Miosireninae are the sister group of the
Trichechidae and are now placed in that family.
The Trichechidae in this broader sense appear to have arisen somewhat
later than previously supposed (late Eocene or early Oligocene rather than
middle Eocene) and are rooted will within the Dugongidae instead of being
derived separately from the Protosirenidae” (Domning 1994); no further
literature is available on these genera)
4.1. Genus Anomotherium Siegfried,
1965
4.1.1.
A. langewieschei Siegfried, 1965
4.2. Genus Miosiren Dollo, 1889: Middle Miocene,
Europe, aberrant, possibly molluscivorous. Previously placed (with its own subfamily, Miosireninae) within the Dugongidae, though a separate
descent from protosirenids was considered an alternate
possibility (Domning 1978).
4.2.1.
M. kocki Dollo, 1889
4.2.2.
M. canhami (a and b are probable synonyms)
Subfamily Trichechinae Gill, 1872 (1821)
4.3. Genus Sirenotherium Possible Early Miocene trichechid described by
Paula Couto (1967) on the basis of 2 teeth and some
postcranial fragments from Brazil.
4.3.1.
Sirenotherium pirabense is allocated
to Trichechidae or Dugongidae incertae sedis (Domning 1982a)
4.4. Genus Potamosiren Reinhart,
1951: Colombia, Middle Miocene (Friasian); earliest known probable trichechid. Another species not listed below from one
tooth probably represents a large species of manatee either contemporaneous and
sympatric with or immediately derived from P.
magdalenensis; and another possible species from
an atlas(Domning 1982a).
4.4.1.
P. magdalenensis Reinhart, 1951 (here
includes “Metaxytherium” ortegense Kellogg 1966, earliest
and most primitive known trichechine or true
manatee. Earliest specimens from Middle
Miocene (Friasian 10-15 mya)
Colombia, synonymous with “Metaxytherium”
ortegense; lacked the horizontally-replaced
supernumerary teeth characteristic of all later manatees, suggesting that
siliceous true grasses (Gramineae) had not yet become
an important part of its diet)
4.5. Genus Ribodon Ameghino, 1883: Late Miocene/Early Pliocene, ~ 5-6 mya, Argentina and North Carolina; only one species known; probably
gave rise to Trichechus; evolution of
unlimited horizontal tooth replacement appears in Ribodon and continues in modern manatees 4
4.5.1.
R. limbatus Ameghino, 1883 Argentina
4.6. Genus Trichechus
Linnaeus4, 1758 (Modern manatees6): Plio-Pleistocene – Recent (1.8 mya): Diversification of the New World Trichechus into the ancestors of
the modern Amazonian and West Indian species, followed by dispersal of members
of the latter stock to West Africa, probably occurred in the Late Pliocene to
Pleistocene. Trichechids, probably
initially distributed mostly in brackish and/or fresh water expanded their
niche following the Later Tertiary Andean orogeny
into fresh water and gave rise to the Amazonian manatee. After the end-Pliocene extinction of the last
Caribbean dugongids, West Indian manatees invaded the salt-water ecological
vacuum.
4.6.1.
T. inunguis (Natterer in von Pelzeln 1883): Amazonian manatee. Probably evolved when the Amazon basin was
isolated from both Pacific and Atlantic.
Perhaps entered the basin prior its closure
during the Andean orogeny from the Pacific (Ribodon) or later
during intermittent connections or stream capture events on the other sides of
the basin (Trichechus). In adapting to the new environment, T. inunguis evolved more rapidly than
did the populations remaining in the coastal regions, and hence exhibits more derived characters including
smaller and more complex molars, loss of nails, increased diploid chromosome
numbers (54 vs. 48 in T. m.).
4.6.2.
T. manatus Linnaeus
1758: West Indian manatee; appears about
1.3 mya, Early Pleistocene in Florida, Late
Pleistocene - subrecent in Jamaica, Louisiana, Ohio, Arkansas,
Florida, South Carolina, North Carolina, Maryland, New Jersey
4.6.2.1.
T. m. manatus Linnaeus,
1758: Antillean manatee
4.6.2.2.
T. m. bakerorum (Domning 2005):
North America Pleistocene
4.6.2.3.
T. m. latirostris (Harlan, 1824): Florida manatee
4.6.2.4.
cf. Trichechus sp. Plio-Pleistocene,
western Amazon basin of Brazil, which resembled the modern
T. m. rather than T. i.
4.6.2.5.
T. m.
unnamed subspecies (120,000-125,000 ya; SE USA, Florida, North Carolina, Louisiana ? is this now named T. m. bakerorum ?)
4.6.3.
T. senegalensis
Link 1795: West African manatee - dispersed
across the Atlantic from South America to Africa in the Late Pliocene or
Pleistocene; distributed from Senegal to Angola, this is the least studied extant species.
Notes
1. See National Geographic Magazine April 2003 for
popular account and extraordinary images of Pezosiren
portelli.
2. In the Pliocene, these were the only sirenians left in
the Mediterranean; they were extinct by the end of the Pliocene – representing
the last sirenians in the European-North African segment of the former Tethyan realm (Domning 1982). Large shark fossils (Carcharodon megalodon)
are found almost exclusively in association with M. serresii bones in the Sahabi Formation, Libya; the sirenian bones bear small
scratches or grooves, which could have been made by shark teeth, indicating
that shark predation and/or scavenging played an important role in this system.
3. Hydrodamalis
gigas is commonly known as Steller’s
sea cow5 and is lumped with modern sirenians because it was extant
during modern times. A remnant
population (~2000 animals living near what are today known as Copper and Bering
Islands) of this giant sea-cow was discovered and described by Georg Wilhelm
Steller in 1741.
4. Unlimited horizontal tooth replacement has been found
only in Ribodon and Trichechus. This evolutionary oddity appears in the
marsupial Peradorcas concinna (convergent
evolution) but not in other sirenian or mammal, not even elephants despite a
widespread misunderstanding. Elephants
have neither an increased number of teeth nor true horizontal replacement. Like dugongs and many mammals, elephants have
forward movement of cheek teeth, AKA “mesial drift”, and the eruption of successive teeth is spaced out over
time, but they have no supernumerary molars and do not replace teeth
continuously throughout their life like manatees (Domning and Hayek 1984). Dugongs have tusks and also exhibit root
hypsodonty.
Photo from Helsinki Museum, submitted by
Ari Lampinen, University of Jyvaskyla, Finland, online at
http://www.cc.jyu.fi/~ala/ilmasto/steller.htm. Note: H. gigas had lost its finger bones, but this
mount included man-made ones, which have since been removed. There is also
a full skeleton on display at the Smithsonian in Washington, DC. Other skeletons have been reported to me from
Kamchatka. Steller’s sea cow is the only
known Sirenian to have lived in extremely cold, sub-polar waters. For an excellent reconstruction of how
Steller’s sea cow came to be, see Domning 1987, listed below.
5. Hydrodamalis
gigas, formerly classified as Rytina gigas, was first seen by modern
humans when Captain Vitus Bering and his comrades discovered an uninhabited
island (later named Bering Island) in 1741. Bering and
his two ships, St. Peter and St. Paul, were on their way home to Kamchatka
following an expedition to map the coast of Alaska for Tsar Peter I the Great
of Russia. The ships were separated during a storm and
Captain Bering, the St. Peter, and her crew were stranded on the island. Although Bering died on the island during the winter of
1741, Georg Wilhelm Steller (a German-born naturalist), and about half of the
ship's crew survived. Steller described
a giant sea cow and its habits, but was vague in his accounts of abundance and
distribution. He said he found it numerous and in
herds, leaving future researchers to guess at exactly how many. Stejneger (1887) estimated the number at less than 1500
and hypothesized that they were the last survivors of a once more numerous and
widely distributed species which had been spared because man had not yet
reached their last resort. Upon the
survivors' return to Kamchatka in 1742, new hunting expeditions were formed
almost every year. They returned to Bering Island
where they spent 8-9 months hunting fur-animals and eating sea cow meat to
survive. Indeed, many of the expeditions are reported
to have wintered on Bering Island for the express purpose of collecting sea cow
meat for the remainder of their 3-4 year journey to the Aleutian Islands and
America. The last sea cow was reported killed in 1768, just 27 years after the island had been discovered by
modern man. From Steller's description,
these huge herbivores are believed to have numbered around 1500-2000 in the
Bering Island and Copper Island areas of the North Pacific (circa 1741). The largest animals were 4-5 fathoms long (1 fathom = 6
feet), 3.5 fathoms thick around, and weighed 200 puds or 80 short hundredweight
(up to about 8,000 pounds). They had no teeth, but two
flat white bones, the one above fixed to the palate, and the one below on the
inside of the lower jaw. Both were furrowed and had
raised ridges with which they masticated kelp. The sea cows were found in herds close to
shore. They drifted just below the surface of the water;
a single animal resembled an overturned boat. Steller
and Waxell both noted large midsections and very small heads.
6. The evolutionary history of trichechids is not as well
defined as it is for the dugongids.
Rather than interpret what I’ve read, I’ve selected an excerpt from the
expert below. Enjoy!
Evolution of
manatees: a speculative history, by Daryl
P. Domning
Journal of
Paleontology 56(3):599-619 May 1982
“Conclusions”
A.
Primitive sirenians (Protosiren or related form) were present in the New World by the
Middle Eocene and presumably colonized coastal rivers and estuaries of the
then-isolated South American continent.
The descendents of these colonists (the Trichechidae), though paralleling
contemporary dugongids in reduction of dental formula, remained morphologically
conservative in other respects down to Potamosiren in the Middle Miocene. They probably continued to occupy a fluvioestuarine niche in contrast to the dugongids in
adjacent marine waters; whereas most Oligocene to Miocene Caribbean dugongids
had strongly downturned snouts evidently adapted to
feeding on marine seagrass beds, Tertiary trichechids (like protosirenids) had
slight rostral deflections more suited to a diet of
floating or emergent aquatic plants.
B.
Mio-Pliocene Andean orogeny,
with resultant erosion and runoff of dissolved nutrients in South American
river systems, greatly increased the productivity of these waters and the
abundance of floating macrophytes, especially grasses
(Gramineae).
Trichechids adapted to this major new food resource, first by evolving
horizontally-replaced supernumerary molars (in Ribodon) and later by reducing
the size of the molars (to increase length of enamel ridge per unit of occlusal area), complicating their enamel folds, and
increasing the number of teeth in the toothrow (in Trichechus).
C.
Pliocene trichechids gained entrance to the
interior of the Amazon basin, at that time temporarily isolated by drainage
reorientation resulting from the Andean orogeny. Cut off from marine waters, they adapted to
feed on the floating meadows of nutrient-rich Amazonian lakes, and gave rise to
the most derived living species, T.
inunguis.
D.
The onset of continental glaciation and the
eustatic fall in sea level during the Pliocene cause worldwide rejuvenation of
drainage systems and increased silt runoff into nearshore
marine waters. Bottom-feeding brachyodont dugongids probably underwent selective pressure
for more wear-resistant teeth; in the Indopacific
this resulted in the evolution of root hypsodonty in Dugong. West Atlantic
dugongids might have evolved comparable adaptations, but for the presence of
trichechids already possessing highly resistant dentitions. Ribodon and primitive Trichechus
then spread into marine waters as far as North America, replacing the
dugongids, broadening their feeding niche to include seagrass meadows, and
giving rise to the modern T.
manatus. Trichechus very similar to T. manatus dispersed to West Africa,
where as T. senegalensis, they
continue to occupy a fluvioestuarine niche.
References
Aranda-Manteca, F. J.; Domning, D. P;
Barnes, L. G. 1994. A
new middle Miocene Sirenian of the genus Metaxytherium
from Baja California and California: Relationships and paleobiogeographic
implications. Proceedings of the San Diego Society of
Natural History 0(29):191-204.
Bajpai, S.; and Domning, D. P. 1997. A new dugongine sirenian
from the early Miocene of India. Journal of Vertebrate
Paleontology 17 (1):219-228.
Domning, D. P.
1978. Sirenia. Chapter 28 in Evolution of
African Mammals, Edited by Vincent J. Maglio and H.
B. S. Cooke. Harvard
University Press, Cambridge, Massachusetts, London, England.
Domning, D. P.
1982a.
Evolution of manatees: a
speculative history. Journal of
Paleontology 56(3):599-619.
Domning, D. P.
1982b. Fossil
Sirenia from the Sahabi Formation. Garyounis
Scientific Bullentin Special Issue, No. 4:29-32.
Domning, D.P. 1987. Sea cow
family reunion. Natural History (April 1987)
96(4):64-71.
Domning, D. P. 1989.
Fossil Sirenia of the west Atlantic and
Caribbean region. III. Xenosiren yucateca,
GEN. ET SP. NOV. Journal of Vertebrate Paleontology 9(4):429-437.
Domning D. P. 1994.
A phylogenetic analysis of
the Sirenia. In Contibutions
in Marine Mammal Paleontology Honoring Frank C. Whitmore, Jr. Edited by Berta and T. A. Deméré. Proceedings of the San Diego Society of
Natural History. 29:177-189.
Domning, D. P.
1997. Fossil Sirenia of the west Atlantic and Caribbean region. VI. Crenatosiren olseni
(Reinhart, 1976). Journal
of Vertebrate Paleontology 17 (2):397-412.
Domning, D. P. 1999.
Fossils explained 24: Sirenians (seacows). Geology Today
15(2): 75-79.
Domning, D. P. 2001a.
Sirenians, seagrasses, and Cenozoic
ecological change in the Caribbean. Palaeogeography, Palaeoclimatology,
Palaeoecology 166(1-2):27-50.
Domning, D. P.
2001b.
Evolution of the Sirenia and Desmostylia. Pages 151-167 in: J.-M. Mazin
& V. de Buffrénil, eds., Secondary Adaptation of Tetrapods to Life in Water. Verlag Dr.
Friedrich Pfeil, München.
Domning, D. P. 2001d.
The earliest known fully quadrupedal sirenian. Nature 413:
625-627.
Domning, D. P.
2005. Fossil Sirenia of the West
Atlantic and Caribbean region. VII. Pleistocene Trichechus manatus Linnaeus, 1758. Journal of Vertebrate Paleontology
25(3):685–701.
Domning, D. P; and Gingerich, P. D. 1994. Protosiren smithae,
new species (Mammalia, Sirenia), from the late middle Eocene of Wadi Hitan, Egypt. Contributions From
the Museum of Paleontology University of Michigan 29(3):69-87.
Domning, D. P.; Gingerich, P. D.;
Simons, E. L.; Ankel-Simons, F. A. 1994. A new early
Oligocene Dugongid (Mammalia, Sirenia) from Fayum Province, Egypt.
Contributions From the Museum of Paleontology
University of Michigan 29(4):89-108.
Domning, D. P., Hayek, L.-A. C. 1984.
Horizontal tooth replacement in the Amazonian manatee
(Trichechus inunguis). Mammalia 48(1):105-127.
Domning, D. P.; Morgan G. S.; Ray,
C. E. 1982.
North American Eocene sea cows (Mammalia: Sirenia). Smithsonian Contributions
to Paleobiology Number 52. Smithsonian Institution
Press, Washington, D.C.
Domning, D. P.; Pervesler,
P.
2001. The osteology
and relationships of Metaxytherium krahuletzi DEPERET, 1895 (Mammalia: Sirenia). Abhandlungen der Senckenbergischen Naturforschenden
Gesellschaft 553:1-89.
Gingerich, P. D.; Domning, D. P.; Blane, C. E.; Uhen, M. D. 1994. Cranial morphology of Protosiren fraasi (Mammalia, Sirenia)
from the middle eocene of
Egypt: A new study using computed tomography.
Contributions From the Museum of Paleontology
University of Michigan 29(2):41-67.
Savage, R. J. G.; Domning, D. P., Thewissen, J. G. M. 1994. Fossil Sirenia of the west
Atlantic and Caribbean region. V. The most primitive known sirenian, Prorastomus sirenoides
Owen, 1855. Journal of Vertebrate
Paleontology 14(3):427-449.
Williams, M. E.; Domning, D. P. 2004. Pleistocene or post-Pleistocene manatees in
the Mississippi and Ohio River valleys.
Marine Mammal Science 20(1):167-176.
Zalmout, I. S.; Ul-Haq,
M.; Gingerich, P. D. 2003. New species of Protosiren (Mammalia, Sirenian) from the
early middle Eocene of Balochistan (Pakistan). Contributions from the Museum of
Paleontology, The University of Michigan
31(3):79-87. August 15, 2003.