Transitional Vertebrate Fossils FAQ
Part 2B

Part 2A

Contents

Part 2C

Lagomorphs

• Barunlestes (see above) The possible Asian rodent/lagomorph ancestor.
• Mimotoma (Paleocene) -- A rabbit-like animal, similar to Barunlestes, but with a rabbit dental formula, changes in the facial bones, and only one layer of enamel on the incisors (unlike the rodents). Like rabbits, it had two upper incisors, but the second incisor is still large and functional, while in modern rabbits it is tiny. Chuankuei-Li et al. (1987; also see Szalay et al., 1993) think this is the actual ancestor of Mimolagus, next.
• Mimolagus (late Eocene) -- Possesses several more lagomorph-like characters, such as a special enamel layer, possible double upper incisors, and large premolars.
• Lushilagus (mid-late Eocene) -- First true lagomorph. Teeth very similar to Mimotoma, and modern rabbit & hare teeth could easily have been derived from these teeth.
• After this, the first modern rabbits appeared in the Oligocene.

Known species-to-species transitions in lagomorphs:

• The mid-Tertiary lagomorph Prolagus shows a very nice "chronocline" (gradual change over time), grading from one species to the next. Gingerich (1977) says: "In Prolagus a very complete fossil record shows a remarkable but continuous and gradual reorganization of the premolar crown morphology in a single lineage."
• Lundelius et al. (1987) mention transitions in Pleistocene rabbits, particularly from Nekrolagus to Sylvilagus, and from Pratilepus to Aluralagus. Note that both these transitions cross genus lines. Also see the lagomorph paper in Chaline (1983). Some of these transitions were considered to be "sudden appearances" until the intervening fossils were studied, revealing numerous transitional individuals.

Condylarths, the first hoofed animals

• Protungulatum (latest Cretaceous) -- Transitional between earliest placental mammals and the condylarths (primitive, small hoofed animals). These early, simple insectivore- like small mammals had one new development: their cheek teeth had grinding surfaces instead of simple, pointed cusps. They were the first mammal herbivores. All their other features are generalized and primitive -- simple plantigrade five-toed clawed feet, all teeth present (3:1:4:3) with no gaps, all limb bones present and unfused, pointy-faced, narrow small brain, eyesocket not closed.

Within a few million years the condylarths split into several slightly different lineages with slightly different teeth, such as oxyclaenids (the most primitive), triisodontines, and phenacodonts (described in other sections). Those first differences amplified over time as the lineages drifted further and further apart, resulting ultimately in such different animals as whales, anteaters, and horses. It's interesting to see how similar the early condylarth lineages were to each other, in contrast to how different their descendants eventually, slowly, became. Paleontologists believe this is a classic example of how 'higher taxa" such as families and orders arise.

Says Carroll (1988, p.505): "In the case of the cetaceans [whales] and the perissodactyls [horses etc.], their origin among the condylarths has been clearly documented....If, as seems likely, it may eventually be possible to trace the ancestry of most of the placental mammals back to the early Paleocene, or even the latest Cretaceous, the differences between the earliest ancestral forms will be very small -- potentially no more than those that distinguish species or even populations within species. The origin of orders will become synonymous with the origin of species or geographical subspecies. In fact, this pattern is what one would expect from our understanding of evolution going back to Darwin. The selective forces related to the origin of major groups would be seen as no different than those leading to adaptation to very slightly differing enviromments and ways of life. On the basis of a better understanding of the anatomy and relationships of the earliest ungulates, we can see that the origin of the Cetacea and the perissodactyls resulted not from major differences in their anatomy and ways of life but from slight differences in their diet and mode of locomotion, as reflected in the pattern of the tooth cusps and details of the bones of the carpus and tarsus." (p. 505)

Species-to-species transitions among the condylarths:

• The most common fossil mammal from the lower Eocene is a little primitive weasel-looking condylarth called Hyopsodus. It was previously known that many very different species of Hyopsodus were found at different sites, with (for example) very different tooth size. In 1976, Gingerich analyzed the tooth size of all the known fossils of Hyopsodus that could be dated reliably and independently. He found that "the pattern of change in tooth size that emerges is one of continuous gradual change between lineages, with gradual divergence following the separation of new sister lineages." When tooth size is charted against time, it shows the single lineage smoothly splitting into four descendant lineages. (This was one of the first detailed & extensive studies of speciation.)
• By 1985, Gingerich had many more specimens of Hyopsodus and of several other Eocene condylarth lineages as well, such as Haplomylus. For example: "Haplomylus speirianus ...gradually became larger over time, ultimately giving rise to a new species Haplomylus scottianus... Hyopsodus latidens also became larger and then smaller, ultimately giving rise to a still smaller species, Hyopsodus simplex." These analyses were based on hundreds of new specimens (505 for Haplomylus, and 869 for Hyposodus) from Clark's Fork Basin in Wyoming. Note, however, that several other species from the same time showed stasis (particularly Ectocion, which was previously reported to show change, but in fact stayed much the same), and that not all species transitions are documented. So transitions are not always found. But sometimes they are found.

Cetaceans (whales, dolphins)

Just several years ago, there was still a large gap in the fossil record of the cetaceans. It was thought that they arose from land-dwelling mesonychids that gradually lost their hind legs and became aquatic. Evolutionary theory predicted that they must have gone through a stage where they had were partially aquatic but still had hind legs, but there were no known intermediate fossils. A flurry of recent discoveries from India & Pakistan (the shores of the ancient Tethys Sea) has pretty much filled this gap. There are still no known species-species transitions, and the "chain of genera" is not complete, but we now have a partial lineage, and sure enough, the new whale fossils have legs, exactly as predicted. (for discussions see Berta, 1994; Gingerich et al. 1990; Thewissen et al. 1994; Discover magazine, Jan. 1995; Gould 1994)

• Eoconodon or similar triisodontine arctocyonids (early Paleocene) Unspecialized condylarths quite similar to the early oxyclaenid condylarths, but with strong canine teeth (showing first meat-eating tendencies), blunt crushing cheek teeth, and flattened claws instead of nails.
• Microclaenodon (mid-Paleocene) -- A transitional genus intermediate between Eoconodon and the mesonychids, with molar teeth reorganizing in numerous ways to look like premolars. Adapted more toward carnivory.
• Dissacus (mid-Paleocene) -- A mesonychid (rather unspecialized Paleocene meat-eating animal) with molars more like premolars & several other tooth changes. Still had 5 toes in the foot and a primitive plantigrade posture.
• Hapalodectes or a very similar mesonychid (early Eocene, around 55 Ma) -- A small mesonychid with very narrow shearing molars, a distinctively shaped zygomatic arch, and peculiar vascularized areas between the molars. Probably a running animal that could swim by paddling its feet. Hapalodectes itself may be just too late to be the whale ancestor, but probably was a close relative of the whale ancestor. Says Carroll (1988): "The skulls of Eocene whales bear unmistakable resemblances to those of primitive terrestrial mammals of the early Cenozoic. Early [whale] genera retain a primitive tooth count with distinct incisors, canines, premolars,, and multirooted molar teeth. Although the snout is elongate, the skull shape resembles that of the mesonychids, especially Hapalodectes...."
• Pakicetus (early-mid Eocene, 52 Ma) -- The oldest fossil whale known. Same skull features as Hapalodectes, still with a very terrestrial ear (tympanic membrane, no protection from pressure changes, no good underwater sound localization), and therefore clearly not a deep diver. Molars still have very mesonychid-like cusps, but other teeth are like those of later whales. Nostrils still at front of head (no blowhole). Whale- like skull crests and elongate jaws. Limbs unknown. Only about 2.5 m long. This skull was found with terrestrial fossils and may have been amphibious, like a hippo.
• Ambulocetus natans (early-mid Eocene, 50 Ma) -- A recently discovered early whale, with enough of the limbs and vertebrae preserved to see how the early whales moved on land and in the water. This whale had four legs! Front legs were stubby. Back legs were short but well-developed, with enormous broad feet that stuck out behind like tail flukes. Had no true tail flukes, just a long simple tail. Size of a sea lion. Still had a long snout with no blowhole. Probably walked on land like a sea lion, and swam with a seal/otter method of steering with the front feet and propelling with the hind feet. So, just as predicted, these early whales were much like modern sea lions -- they could swim, but they could also still walk on land. (Thewissen et al., 1994)
• Rodhocetus (mid-Eocene, 46 Ma) -- Another very recent (1993) fossil whale discovery. Had hind legs a third smaller than those of A. natans. Could probably still "waddle" a bit on land, but by now it had a powerful tail (indicated by massive tail vertebrae) and could probably stay out at sea for long periods of time. Nostrils had moved back a bit from the tip of the snout.
• Basilosaurus isis, Protocetes, Indocetus ramani and similar small-legged whales of the mid-late Eocene (45-42 Ma) -- After Rodhocetus came several whales that still had hind legs, but couldn't walk on them any more. For example, B. isis (42 Ma) had hind feet with 3 toes and a tiny remnant of the 2nd toe (the big toe is totally missing). The legs were small and must have been useless for locomotion, but were specialized for swinging forward into a locked straddle position -- probably an aid to copulation for this long-bodied, serpentine whale. B. isis may have been a "cousin" to modern whales, not directly ancestral. Another recent discovery is Protocetes, a slightly more advanced whale from the late Eocene. It was about 3m long (dolphin sized), and still had primitive dentition, nostrils at end of snout, and a large pelvis attached to the spine; limbs unknown. Finally Indocetus is known from only fragmentary remains, but these include a tibia. These late Eocene legged whales still had mesonychid-like teeth, and in fact, some of the whale fossils were first mis-identified as mesonychids when only the teeth were found. ( See Gingerich et al. (1990) for more info on B. isis.)
• Prozeuglodon (late Eocene, 40 Ma) Another recently discovered whale, found in 1989. Had almost lost the hind legs, but not quite: still carried a pair of vestigial 6- inch hind legs on its 15-foot body.
• Eocetus, & similar "archeocete whales" of the late Eocene These more advanced whales have lost their hind legs entirely, but retain a"primitive whale" skull and teeth, with unfused nostrils. They grew to larger body size (up to 25m by the end of the Eocene), an had an elongate, streamlined body, flippers, and a cartilaginous tail fluke. The ear was modified for hearing underwater. Note that this stage of aquatic adaptation was attained about 15 million years after the first terrestrial mesonychids.
• Dorudon intermedius -- a late Eocene whale probably ancestral to modern whales.

In the Oligocene, whales split into two lineages:

1. Toothed whales:
   • Agorophius (late Oligocene) -- Skull partly telescoped, but cheek teeth still rooted. Intermediate in many ways between archaeocetes and later toothed whales.
   • Prosqualodon (late Oligocene) -- Skull fully telescoped with nostrils on top (blowhole). Cheek teeth increased in number but still have old cusps. Probably ancestral to most later toothed whales (possibly excepting the sperm whales?)
   • Kentriodon (mid-Miocene) -- Skull telescoped, still symmetrical. Radiated in the late Miocene into the modern dolphins and small toothed whales with asymmetrical skulls.
2. Baleen (toothless) whales:
   • Aetiocetus (late Oligocene) -- The most primitive known mysticete whale and probably the stem group of all later baleen whales. Had developed mysticete-style loose jaw hinge and air sinus, but still had all its teeth. Later,
   • Mesocetus (mid-Miocene) lost its teeth.
   • Modern baleen whales first appeared in the late Miocene.

Perissodactyls (horses, tapirs, rhinos)

Here we come to the most famous general lineage of all, the horse sequence. It was the first such lineage to be discovered, in the late 1800's, and thus became the most famous. There is an odd rumor circulating in creationist circles that the horse sequence is somehow suspect or outdated. Not so; it's a very good sequence that has grown only more detailed and complete over the years, changing mainly by the addition of large side-branches. As these various paleontologists have said recently: "The extensive fossil record of the family Equidae provides an excellent example of long-term, large-scale evolutionary change." (Colbert, 1988) "The fossil record [of horses] provides a lucid story of descent with change for nearly 50 million years, and we know much about the ancestors of modern horses."(Evander, in Prothero & Schoch 1989, p. 125) "All the morphological changes in the history of the Equidae can be accounted for by the neo-Darwinian theory of microevolution: genetic variation, natural selection, genetic drift, and speciation." (Futuyma, 1986, p.409) "...fossil horses do indeed provide compelling evidence in support of evolutionary theory." (MacFadden, 1988)

So here's the summary of the horse sequence. For more info, see the Horse Evolution FAQ.

• Loxolophus (early Paleocene) -- A primitive condylarth with rather low-crowned molars, probably ancestral to the phenacodontid condylarths.
• Tetraclaenodon (mid-Paleocene) -- A more advanced Paleocene condylarth from the phenacodontid family, and almost certainly ancestral to all the perissodactyls (a different order). Long but unspecialized limbs; 5 toes on each foot (#1 and #5 smaller). Slightly more efficient wrist.

GAP: There are almost no known perissodactyl fossils from the late Paleocene. This is actually a small gap; it's only noticeable because the perissodactyl record is otherwise very complete. Recent discoveries have made clear that the first perissodactyls arose in Asia (a poorly studied continent), so hopefully the ongoing new fossil hunts in Asia will fill this small but frustrating gap. The first clue has already come in:

• Radinskya yupingae (late Paleocene, China) -- A recently discovered perissodactyl-like condylarth. (McKenna et al., in Prothero & Schoch, 1989.)
• Hyracotherium (early Eocene, about 55 Ma; previously "Eohippus") -- The famous "dawn horse", a small, doggish perissodactyl, with an arched back, short neck, omnivore teeth, and short snout. 4 toes in front and 3 behind. Compared to Tetraclaenodon, has longer toes, interlocking ankle bones, and slightly different tooth cusps. Probably evolved from Tetra. in about 4-5 my, perhaps via an Asian species like Radinskya. Note that Hyrac. differed from other early perissodactyls (such as tapir/rhino ancestors) only by small changes in tooth cusps and in body size.
• Hyracotherium vassacciense (early Eocene) -- The particular species that probably gave rise to the equids.
• Orohippus (mid-Eocene, ~50 Ma) -- Small, 4/3 toed, developing browser tooth crests.
• Epihippus (late Eocene, ~45 Ma) -- Small, 4/3 toed, good tooth crests, browser.
• Epihippus (Duchesnehippus) -- A later subgenus with Mesohippus-like teeth.
• Mesohippus celer (latest Eocene, 40 Ma) -- Three-toed on all feet, browser, slightly larger
• Mesohippus westoni (early Oligocene) -- A slightly later, more advanced species.
• Miohippus assiniboiensis (mid-Oligocene) -- This species split off from early Mesohippus via cladogenetic evolution, after which Miohippus and Mesohippus overlapped for the next 4 my. Distinctly larger, slightly longer skull, facial fossa deeper and more expanded, subtly different ankle joint, variable extra crest on upper cheek teeth. In the early Miocene (24 My) Miohippus began to speciate rapidly. Grasses had just evolved, & teeth began to change accordingly. Legs, etc., started to change for fast running.
• Kalobatippus (late Oligocene) -- Three-toed browser w/foot intermediate between Mio. & Para.
• Parahippus (early Miocene, 23 Ma) -- Three-toed browser/grazer, developing "spring foot". Permanent establishment of the extra crest that was so variable in Miohippus. Stronger tooth crests & slightly taller tooth crowns.
• 'Parahippus' leonensis (mid-Miocene, ~20 Ma) -- Three-toed browser/grazer with the emphasis on grazer. Developing spring-foot & high-crowned teeth.
• 'Merychippus' gunteri (mid-Miocene, ~18 Ma) -- Three-toed grazer, fully spring-footed with high-crowned teeth.
• Merychippus primus (mid-Miocene, ~17 Ma) -- Slightly more advanced.
• Merychippus spp. of mid-late Miocene (16-15 Ma) -- 3-toed grazers, spring-footed, size of small pony. Diversified into all available grazer niches, giving rise to at least 19 successful three-toed grazers. Side toes of varying sizes, very small in some lines. Horsey hoof develops, leg bones fuse. Fully high-crowned teeth with thick cement & same crests as Parahippus. The line that eventually produced Equus developed as follows: M. primus, M. sejunctus, M. isonesus (these last two still had a mix of primitive, hipparion, and equine features), M. intermontanus, M. stylodontus, M. carrizoensis. These last two looked quite horsey, with quite small side toes, and gave rise to a set of larger three-toed and one-toed horses known as the "true equines". Crystal clear, right?

SMALL GAP: It is not known which Merychippus species (stylodontus? carrizoensis?) gave rise to the first Dinohippus species (Evander, in Prothero & S 1988).

• Dinohippus (late Miocene, 12 Ma) -- One-toed grazer, spring-footed. Very equine feet, teeth, and skull, with straighter teeth & smaller fossae. First was D. spectans, followed by D. interpolatus and D. leidyanus. A slightly later species was D. mexicanus, with even straighter teeth and even smaller fossae.
• Equus (Plesippus), also called the "E. simplicidens" group (Pliocene, ~4 My) -- Three closely related species of one-toed spring-footed high-crowned grazers. No fossae and very straight teeth. Pony size, fully "horsey" body -- rigid spine, long neck, long legs, fused leg bones with no rotation, long nose, flexible muzzle, deep jaw. The brain was a bit larger than in early Dinohippus. Still had some primitive traits such as simple teeth & slight facial fossae, which later Equus species lost. These "simple Equus" species quickly diversified into at least 12 new species in 4 different groups. During the first major glaciations of the late Pliocene (2.6 Ma), certain Equus species crossed to the Old World. Worldwide, Equus took over the niche of "large coarse-grazing plains runner".
• Equus (Hippotigris) (Pleistocene) -- Subgenus of modern 1-toed spring-footed grazing zebras.
• Equus (Equus) (Pleistocene) -- Subgenus of modern 1-toed spring-footed grazing horses & donkeys. [note: very rarely a horse is born with small side toes, indicating that some horses retain the genes for side toes.]

• Loxolophus, see above
• Tetraclaenodon, see above
• Homagalax (early Eocene) -- Very like its sister genus Hyracotherium, but had cross-lophs on teeth. Note that these early perissodactyls differed only in slight details of the teeth.
• Heptodon (late early Eocene) -- A small early tapiroid showing one more tooth cusp change. Split into two lineages:
  1. Helaletes (mid-Eocene) which had a short proboscis, then Prototapir (late Oligocene), much like modern tapirs but without such a flexible snout, then Miotapirus (early Miocene), an almost- modern tapir with a flexible snout, then Tapirus (Pliocene) the modern tapir.
  2. Hyrachyus (late Eocene), a tapiroid with increased shearing function in its teeth. Led to the late Eocene hyracodontids such as Hyracodon (rhino-tapiroids, or "running rhinos") that show increasing development of high-crowned teeth and larger body size. They led to Caenopus (early Oligocene), a large, hornless, generalized rhino which led to the modern horned rhinos of the Miocene & Pliocene. Our living genera first appear in the Pliocene, about 4 Ma.

Species-species transitions:

• Horses: Gingerich (1980) documented speciation from Hyracotherium grangeri to H. aemulor. Prothero & Schoch (1989) mention some intermediate fossils that link late Orohippus to Mesohippus celer. MacFadden (1985) has documented numerous smooth transitions among the three-toed horses, particularly among Merychippus and the various hipparions. Hulbert (in Prothero & Schoch, 1989) showed that Dinohippus smoothly grades into Equus through successive Pliocene strata. Simpson (1961) describes gradual loss of the side toes in Pliohippus through 3 successive strata of the early Pliocene.
• Rhinos: Wood (1954) said of the rhino fossils "whenever we do have positive paleontological evidence, the picture is of the most extreme gradualism" (quoted in Gingerich, 1977), and Kurten (1968) describes a smooth transition between Dicerorhinus species.

Elephants

• Minchenella or a similar condylarth (late Paleocene) -- Known only from lower jaws. Has a distinctive broadened shelf on the third molar. The most plausible ancestor of the embrithopods & anthracobunids.
• Phenacolophus (late Paleocene or early Eocene) -- An early embrithopod (very early, slightly elephant-like condylarths), thought to be the stem-group of all elephants.
• Pilgrimella (early Eocene) -- An anthracobunid (early proto-elephant condylarth), with massive molar cusps aligned in two transverse ridges.
• Unnamed species of proto-elephant (early Eocene) -- Discovered recently in Algeria. Had slightly enlarged upper incisors (the beginnings of tusks), and various tooth reductions. Still had "normal" molars instead of the strange multi-layered molars of modern elephants. Had the high forehead and pneumatized skull bones of later elephants, and was clearly a heavy-boned, slow animal. Only one meter tall.
• Moeritherium, Numidotherium, Barytherium (early-mid Eocene) -- A group of three similar very early elephants. It is unclear which of the three came first. Pig-sized with stout legs, broad spreading feet and flat hooves. Elephantish face with the eye set far forward & a very deep jaw. Second incisors enlarged into short tusks, in upper and lower jaws; little first incisors still present; loss of some teeth. No trunk.
• Paleomastodon, Phiomia (early Oligocene) -- The first "mastodonts", a medium-sized animals with a trunk, long lower jaws, and short upper and lower tusks. Lost first incisors and canines. Molars still have heavy rounded cusps, with enamel bands becoming irregular. Phiomia was up to eight feet tall.

GAP: Here's that Oligocene gap again. No elephant fossils at all for several million years.

• Gomphotherium (early Miocene) -- Basically a large edition of Phiomia, with tooth enamel bands becoming very irregular. Two long rows cusps on teeth became cross- crests when worn down. Gave rise to several families of elephant- relatives that spread all over the world. From here on the elephant lineages are known to the species level.
• The mastodon lineage split off here, becoming more adapted to a forest browser niche, and going through Miomastodon (Miocene) and Pliomastodon (Pliocene), to Mastodon (or "Mammut", Pleistocene).

Meanwhile, the elephant lineage became still larger, adapting to a savannah/steppe grazer niche:

• Stegotetrabelodon (late Miocene) -- One of the first of the "true" elephants, but still had two long rows of cross-crests, functional premolars, and lower tusks. Other early Miocene genera show compression of the molar cusps into plates (a modern feature ), with exactly as many plates as there were cusps. Molars start erupting from front to back, actually moving forward in the jaw throughout life.
• Primelephas (latest Miocene) -- Short lower jaw makes it look like an elephant now. Reduction & loss of premolars. Very numerous plates on the molars, now; we're now at the modern elephants' bizarre system of one enormous multi-layered molar being functional at a time, moving forward in the jaw.
• Primelephas gomphotheroides (mid-Pliocene) -- A later species that split into three lineages, Loxodonta, Elephas, and Mammuthus:
  1. Loxodonta adaurora (5 Ma). Gave rise to the modern African elephant Loxodonta africana about 3.5 Ma.
  2. Elephas ekorensis (5 Ma), an early Asian elephant with rather primitive molars, clearly derived directly from P. gomphotheroides. Led directly to:
     • Elephas recki, which sent off one side branch, E. hydrusicus, at 3.8 Ma, and then continued changing on its own until it became E. iolensis.
     • Elephas maximus, the modern Asian elephant, clearly derived from
     • E. hysudricus. Strikingly similar to young E. hysudricus animals. Possibly a case of neoteny (in which "new" traits are simply juvenile features retained into adulthood).
  3. Mammuthus meridionalis, clearly derived from P. gomphotheroides. Spread around the northern hemisphere. In Europe, led to M. armeniacus/trogontherii, and then to M. primigenius. In North America, led to M. imperator and then M. columbi.

Species-species transitions among the elephants:

• Maglio (1973) studied Pleistocene elephants closely. Overall, Maglio showed that at least 7 of the 17 Quaternary elephant species arose through smooth anagenesis transitions from their ancestors. For example, he said that Elephas recki "can be traced through a progressive series of stages...These stages pass almost imperceptibly into each other....In the late Pleistocene a more progressive elephant appears which I retain as a distinct species, E. iolensis, only as a matter of convenience. Although as a group, material referred to E. iolensis is distinct from that of E. recki, some intermediate specimens are known, and E. iolensis seems to represent a very progressive, terminal stage in the E. recki specific lineage."
• Maglio also documented very smooth transitions between three Eurasian mammoth species: Mammuthus meridionalis --> M. armeniacus (or M. trogontherii) --> M. primigenius.
• Lister (1993) reanalyzed mammoth teeth and confirmed Maglio's scheme of gradual evolution in European mammoths, and found evidence for gradual transitions in the North American mammoths too.

Sirenians (dugongs & manatees)

GAP: The ancestors of sirenians are not known. No sirenian-like fossils are known from before the Eocene.

• Early Eocene -- fragmentary sirenian fossils known from Hungary.
• Prorastomus (mid-Eocene) -- A very primitive sirenian with an extremely primitive dental formula (including the ancient fifth premolar that all other mammals lost in the Cretaceous! Could this mean sirenians split off from all other mammals very early on?) The skull is somewhat condylarth-like. Had distinctive sirenian ribs. Not enough of the rest of the skeleton was found to know how aquatic it was.
• Protosiren (late Eocene) -- A sirenian with an essentially modern skeleton, though it still had the very primitive dental formula. Probably split into the two surviving lineages:
  1. Dugongs: Eotheroides (late Eocene), with a slightly curved snout and small tusks, still with the primitive dental formula. Perhaps gave rise to Halitherium (Oligocene) a dugong-ish sirenian with a more curved snout and longer tusks, and then to living dugongs, very curved snout & big tusks.
  2. Manatees: Sirenotherium (early Miocene); Potamosiren (late Miocene), a manatee-like sirenian with loss of some cheek teeth; then Ribodon (early Pliocene), a manatee with continuous tooth replacement, and then the living manatees.

Part 2A

Contents

Part 2C