Boyer, Douglas M.; Gingerich, Philip D.
Plesiadapis cookei is a large-bodied plesiadapiform euarchontan (and potential stem primate) known from many localities of middle Clarkforkian North American Land Mammal age, late Paleocene epoch, in the Clarks Fork Basin of northwestern Wyoming. Most specimens are gnathic and dental remains, but one specimen, University of Michigan Museum of Paleontology (UM) 87990, is a relatively complete skeleton preserving the skull, much of the axial skeleton, forelimbs, and hind limbs. Description, measurement, and illustration of the UM 87990 skeleton of P. cookei provides new information enabling analysis of proportions within the skeleton and comparisons across a range of primates and related mammals.
The skull of UM 87990 includes much of the upper and lower dentition, rostrum, braincase, and basicranium. The dental formula is 18.104.22.168 / 22.214.171.124. Composition of the auditory bulla as petrosal, ectotympanic, or entotympanic, is uncertain due breakage and some suture-like structures rimming the promontorium. P. cookei had a very small posterior carotid foramen and probably lacked a functioning internal carotid artery.
The axial skeleton of UM 87990 is represented by five cervical, 12 thoracic, and six lumbar vertebrae. The sacrum is complete with three vertebrae. There are 17 caudal vertebrae preserved, with at least three anterior positions missing. Thus there were at least 20 caudal vertebrae. Caudal vertebral proportions indicate the tail of P. cookei was relatively long. We expect the full plesiadapid vertebral formula to have been 126.96.36.199.24, matching the formula in Tupaia and the median formula in primates. Seven sternebrae are preserved, possibly representing the complete series. Ribs are slender and lack any notable anteroposterior expansion.
The pectoral girdle and forelimb of UM 87990 include clavicles, much of one scapula, humeri, ulnae, radii, and a number of carpal bones. Two sets of metacarpals, one set slightly longer than the other, were found in association with UM 87990. One set must have come from a partial skeleton of Uintacyon (UM 88187) that was preserved next to UM 87990. Manual proximal phalanges have long, low flexor sheath ridges. Intermediate phalanges have narrow, deep shafts, and distal articular surfaces suggesting flexed distal interphalangeal joints. Distal phalanges are all relatively long and distinctly falciform.
The pelvic girdle and hind limb of UM 87990 include well-preserved innominates, femora, tibiae, incomplete fibulae, and most tarsal bones (navicular and entocuneiform excepted). Metatarsals are identified based on their size and form in comparison to metatarsals described for other plesiadapids. Pedal phalanges are similar to manual phalanges but can be distinguished on the basis of size and a more neutrally flexed distal interphalangeal joint.
A principal components analysis of body proportions places P. cookei and other plesiadapiforms close to Tupaia, sciurids, and callitrichid primates. P. cookei is not particularly close to the lorisine slow climber Nycticebus, nor is it close to the colugo Cynocephalus. The postcranial skeleton of P. cookei suggests that it was a forest-dwelling arboreal climber primarily adapted, and possibly constrained, to large diameter vertical and horizontal supports. Headfirst descent of large supports was likely accomplished using claw-clinging with a reversed (supinated) foot. P. cookei would have been more cautious and less scansorial in its movements than smaller-bodied plesiadapids.
The body weight of P. cookei is predicted to be 1,799 to 2,052 g, based on long bone lengths and diameters of UM 87990. Brain weight is predicted to be close to 5 g based on a partial endocast and three-dimensional model. Comparison of observed with expected brain size for a mammal the size of P. cookei yields an encephalization quotient in the range of 0.257 to 0.281 and an encephalization residual in the range of −1.958 to −1.831. Both indicate that the brain of P. cookei was just over one-quarter of the size expected for an average living mammal of its body weight. A smaller than expected brain is a common feature of many Paleogene mammals.
We interpret UM 87990 to be a male individual of P. cookei , because the innominate has a relatively wide superior pubic ramus like that of male individuals in a range of rodents and primate species. UM 87990 has a fully erupted adult dentition, little tooth wear, and long-bone epiphyseal fusion limited to the elbow. We have no independent means of determining sexual maturity, and thus can only interpret UM 87990 as an advanced and nearly full-grown subadult, or as a full-grown young adult.
Cladistic analysis of craniodental characters and a geographic character indicates that Plesiadapidae is monophyletic, with the carpolestid Elphidotarsius forming the sister taxon of the group. Pronothodectes is paraphyletic but its species are the most basal within Plesiadapidae. Chiromyoides, and Platychoerops are monophyletic. Plesiadapis itself is polyphyletic, with P. cookei recovered as the sister taxon of Platychoerops. Our reconstructed phylogeny for the family is largely congruent with that of Gingerich (1976) based on stratophenetic methodology, but less resolved. A stratocladistic analysis run by adding a stratigraphic character to the craniodental and geographic character set is more resolved than our strict consensus using morphology and geography alone, and it is also more divergent from Gingerich’s (1976) hypothesis in certain ways. Some notable results include recovery of Nannodectes intermedius as the common ancestor of all non-Pronothodectes plesiadapids and recovery of Platychoerops antiquus as ancestral to all later Platychoerops species.
On a broader scale, cladistic analysis of higher-level taxa, with scorings for Plesiadapidae revised based on optimization of the ancestral node at the species level, indicates that plesiadapids and carpolestids exhibit a greater number of identical character states than previously thought, though some previously proposed synapomorphies were refuted. Even so, analysis of combined data from dentition, cranium, and postcrania still robustly support a link between plesiadapids, saxonellids, and carpolestids (Plesiadapoidea) and does not contradict previous hypotheses suggesting a special relationship of plesiadapoids to euprimates (Euprimateformes).
University of Michigan Deep Blue - digital version
Publisher: University of Michigan
Month of Publication: October
Year of Publication: 2019
Location: Ann Arbor, MI
Volume Number: 38
# of Pages: 287