BY
WILLIAM A. CLEMENS, JR.
The family Apatemyidae has a long geochronological range in
North America, beginning in the Torrejonian land-mammal age, but
is represented by a relatively small number of fossils found at a few
localities. Two fossils of Orellan age, found in northeastern Colorado
and described here, demonstrate that the geochronological
range of the Apatemyidae extends into the Middle Oligocene. Isolated
teeth of Sinclairella dakotensis Jepsen, part of a sample of a
Chadronian local fauna collected by field parties from the Webb
School of California, are also described.
I thank Mr. Raymond M. Alf, Webb School of California, Claremont,
California, and Dr. Peter Robinson, University of Colorado Museum, Boulder,
Colorado, for permitting me to describe the fossils they discovered. Also Dr.
Robinson made available the draft of a short paper he had prepared on the
tooth found in Weld County, Colorado; his work was facilitated by a grant
from the University of Colorado Council on Research and Creative Work. I
also gratefully acknowledge receipt of critical data and valuable comments
from Drs. Edwin C. Galbreath, Glenn L. Jepsen, and Malcolm C. McKenna
who is currently revising the Paleocene apatemyids and studying the phylogenetic
relationships of the family. The prefixes of catalogue numbers used
in the text identify fossils in the collections of the following institutions: KU,
Museum of Natural History, The University of Kansas, Lawrence; Princeton,
Princeton Museum, Princeton, New Jersey; RAM-UCR, Raymond Alf Museum,
Webb School of California, Claremont, California (the permanent repository
for these specimens will be the University of California, Riverside); and UCM,
University of Colorado Museum, Boulder, Colorado. The system of notations
for teeth prescribed for use here is as follows: teeth in the upper half of the
dentition are designated by a capital letter and a number; thus M2 is the notation
for the upper second molar; teeth in the lower half of the dentition are
designated by a lower-case letter and a number; thus p2 is the notation for the
lower second premolar.
The type of the species, Princeton no. 13585, was discovered in
Chadronian strata of the upper part of the Chadron Formation
cropping out in Big Corral Draw, approximately 13 miles south-southwest
of Scenic, in southwestern South Dakota (Jepsen, 1934,
p. 291). Detailed descriptions of the type specimen are given in
papers by Jepsen (1934) and Scott and Jepsen (1936). Isolated
teeth of Chadronian age referable to Sinclairella dakotensis have
been discovered subsequently at a locality in Nebraska and fossils
of Orellan age, also referable to S. dakotensis, have been collected
at two localities in Colorado. The sample from each locality is
described separately.
Material.—RAM-UCR nos. 381, left M1; 598, left m2; 1000, right m1; 1001,
right m2; 1079, right m2; 1674, right M2; and 3013, left m2.
Locality and stratigraphy.—These Chadronian fossils were discovered by
Raymond Alf and members of his field parties in several harvester ant mounds
built in exposures of the Chadron Formation in Sec. 26, T 33 N, R 53 W, Sioux
County, Nebraska (Alf, 1962, and Hough and Alf, 1958). This is UCR
locality V5403. The collectors carefully considered the possibility that some
of the fossils found in the ant mounds were collected from younger strata by
the harvester ants and concluded this was unlikely (Alf, personal communication).
Description and comments.—The cusps of RAM-UCR no. 381, a left M1,
are sharp and the wear-facets resulting from occlusion with the lower dentition
are small. The paraconule is a low, ill-defined cusp on the anterior margin
of the crown; a metaconule is not present. A smooth stylar shelf is present
labial to the metacone. The crown was supported by three roots. There are
no interradicular crests.
The crown of RAM-UCR no. 1674, a right M2, is heavily abraded and
many morphological details of the cusps have been destroyed. Low interradicular
crests linked the three roots of the tooth with a low, central prominence.
As was the case with RAM-UCR no. 381, no significant differences
could be found in comparisons with illustrations of the teeth preserved in
Princeton no. 13585.
RAM-UCR nos. 598, 1001, 1079, and 3013 all appear to be m2's. The
talonids of these teeth are not elongated, their trigonids have quadrilateral outlines,
and the paraconids are small but prominent, bladelike cusps. The trigonid
of RAM-UCR 1000 is elongated and the paraconid is a minute cusp; the
tooth closely resembles the m1 of the type of Sinclairella dakotensis.
Material.—KU no. 11210 (fig. 1), a fragment of a left maxillary containing
P4 and M1-2.
Locality and stratigraphy.—The fossil was found in the center of the W-1/2,
Sec. 21, T 11 N, R 53 W, Logan County, Colorado, " ... in the bed below
Agnotocastor bed, Cedar Creek Member...." (Ronald H. Pine, 1958,
field notes on file at the University of Kansas). The bed so defined is part of
unit 3 in the lower division of the Cedar Creek Member, as subdivided by
Galbreath (1953:25) in stratigraphic section XII. The fauna obtained from
unit 3 is of Orellan age.
Fig. 1. Sinclairella dakotensis Jepsen, KU no. 11210, fragment of left maxillary
with P4 and M1-2; Orellan, Logan County, Colorado; drawings by Mrs. Judith
Hood: a, labial view; b, occlusal view; both approximately × 9.
Description and comments.—P4 of KU no. 11210 has a large posterolingual
cusp separated from the main cusp by a distinct groove, which deepens posteriorly.
The posterolingual cusp is supported by the broad posterior root. P4
of the type specimen of Sinclairella dakotensis is described (Jepsen, 1934, p.
392) as having an oval outline at the base of the crown, and a small, posterolingual
cusp. A chip of enamel is missing from the posterior slope of the main
cusp of the P4 of KU no. 11210. The anterior slope of the main cusp is
flattened, possibly the result of wear, and there is no evidence of a groove like
that present on the P4 of the type specimen.
Only a few differences were found between the molars preserved in KU no.
11210 and their counterparts in the type specimen. A stylar shelf is present
labial to the metacone of M1 of KU no. 11210, but, unlike the type, its surface
is smooth and there is no evidence of cusps. Of the three small stylar
cusps on the stylar shelf of M2 the smallest is in the position of a mesostyle.
The M2 lacks a chip of enamel from the lingual surface of the hypocone. Unlike
the M2 of Princeton no. 13585, in occlusal view the posterior margin of the
M2 of KU no. 11210 is convex posterior to the metacone. The anterior edge
of the base of the zygomatic arch of KU no. 11210 was dorsal to M2. The
shallow oval depression in the maxillary dorsal to M1 might be the result of
post-mortem distortion.
The molars preserved in KU no. 11210 and their counterparts in the type
specimen do not appear to be significantly different in size (table 1) or morphology
of the cusps. The only difference between the two specimens that
might be of classificatory significance is the difference in size of the posterolingual
cusp of P4. At present the range of intraspecific variation in the morphology
of P4 has not been documented for any species of apatemyid. The
evolutionary trend or trends of the apatemyids (McKenna, 1960, p. 48) for
progressive reduction of function of p4 probably were paralleled by similar
trends in the evolution of the P4. If so, the intraspecific variation in the morphology
of P4 could be expected to be somewhat greater than that of the upper
molars, for example. The morphological difference between the P4's of the
type of Sinclairella dakotensis and KU no. 11210 is not extreme and does not
exceed the range of intraspecific variation that could be expected for this element
of the dentition. The close resemblances in
size and morphology between the M1-2 of Princeton
no. 13585 and KU no. 11210 also favor identification
of the latter as part of a member of an
Orellan population of Sinclairella dakotensis.
Fig. 2. Sinclairella dakotensis
Jepsen, UCM no. 21073, right M2; Orellan, Weld County, Colorado;
drawing by Mrs. Judith Hood: occlusal view, approximately × 9.
Material.—UCM no. 20173 (fig. 2), is a right M2.
Locality and stratigraphy.—The tooth was discovered
at the Mellinger locality, Sec. 17, T 11
N, R 65 W, Weld County, Colorado. The Mellinger
locality is in the Cedar Creek Member,
White River Formation, and its fauna is considered
to be of Orellan age (Patterson and McGrew,
1937, and Galbreath, 1953).
Description and comments.—UCM no. 21073,
which is more heavily abraded than KU no. 11210,
shows no evidence of a stylar cusp either anterolabial
to the metacone or in the position of a mesostyle. A small stylar cusp
is present anterolabial to the paracone. A notch that appears to have been cut
through the enamel of the posterolabial corner of the crown could have received
the parastylar apex of M3. A similar notch is not present on the M2 of KU no.
11210 nor indicated in the illustrations of the M2 of Princeton no. 13585. The
coronal dimensions of UCM no. 21073 (table 1) do not appear to differ significantly
from those of the M2's of KU no. 11210 and the type specimen of
Sinclairella dakotensis.
With the discovery of Orellan apatemyids the geochronological
range of the family in North America is shown to extend from the
Torrejonian through the Orellan land-mammal ages. The discoveries
reported here enlarge the Oligocene record of apatemyids
to include not only the type specimen of Sinclairella dakotensis, a
skull and associated mandible from South Dakota, but also seven
isolated teeth, representing at least two individuals, from a Chadronian
fossil locality in Nebraska and one specimen from each of
two Orellan fossil localities in northeastern Colorado. Simpson
(1944:73, and 1953:127) presented tabulations of the published records
of American apatemyids and suggested the data indicated the
populations of these mammals were of small size throughout the
history of the family. The few pre-Oligocene occurrences of apatemyids
described subsequently (note McKenna, 1960, figs. 3-10,
and p. 48) and occurrences described here tend to reinforce Simpson's
interpretation. This interpretation may have to be modified
to some degree, however, when current studies of collections of pre-Oligocene
apatemyids are completed (McKenna, personal communication).
Although information concerning the evolutionary trends of American
apatemyids has been published, no data on the morphological
variation in a population are available in the literature. An adequate
basis for evaluating the significance of the morphological differences
between the P4's of Princeton no. 13585 and KU no. 12110 coupled
with the similarities of their M1-2's is lacking. In the evolution of
American apatemyids the P4 underwent reduction in size and, apparently,
curtailment of function. This history suggests the range
of morphological variation of P4 in populations of Sinclairella dakotensis
could be expected to be greater than that of the molars
and encompass the morphological differences between the P4's of
Princeton no. 13585 and KU no. 12110. The difference in age of
the Chadronian and Orellan fossils does not constitute proof that
they pertain to different species. Although the identification is admittedly
provisional until more fossils including other parts of the
skeleton are discovered, the Orellan fossils described here are referred
to Sinclairella dakotensis.
Table 1.—Measurements (in millimeters) of Teeth of Sinclairella
dakotensis Jepsen.
| | P4 | M1 | M2 |
|---|
| | length | width | length[1] | width[1] | length[1] | width[1] |
| Princeton no. 13585[2] | 2.1 | 1.1 | 4.0 | 3.7 | 3.4 | 4.7 |
| RAM no. 381 | | | 4.1 | 3.5 | | |
| RAM no. 1674 | | | | | 3.4 | 4.2 |
| KU no. 11210 | 2.4 | 1.6 | 3.9 | 3.5 | 3.8 | 4.1+ |
| UCM no. 21073 | | | | | 3.6 | 4.1 |
| | m1 | m2 |
| length | width | length | width |
| Princeton no. 13585[3] | 3.5 | 2.4 | 3.7 | 2.8 |
| RAM no. 1000 | 3.5 | 2.2 | | |
| RAM no. 598 | | | 3.8 | 2.6 |
| RAM no. 1001 | | | 3.6+ | 2.6 |
| RAM no. 1079 | | | 4.0 | 2.8 |
| RAM no. 3013 | | | 3.6 | 2.8 |