HARVARD UNIVERSITY

Ernst Mayr Library

of the Museum of

Comparative Zoology

, Mcz

Ubrary

^^^ 4 2006

PROCEEDINGS

of the

San Diego Society of Natural History

JUL 2 1 WJ

Founded 1874

Numher 32

30 June- 1097

New Specimens of Metanoiamys, Pauromys, and Simimys (Rodentia:

Myoniorpha) fVoni the Uintan (middle liocene) of San Diego County, California,

and Comments on the Relationships of Selected Paleogene Myoniorpha

Stephen L. Walsh

Departmeni of Paleontology. San Diego Natural History Museum. San Diego, California 92112

ABSTRACT. — The first known niandibular and maxilhirj' specimens of the middle Eocene hasal eoniyid Mi'hinoiamys agorus have been recovered
from the Friars Fomiation (early Uintan) of San Diego County. Although tlie genus was previously diagnosed as having a horizontal dorsal border of
the mandible anterior to p4. Ibis character was based on a specimen thai actually pertains to Micropitnimys sp. cf. M. minulus. Undoubted mandibles
of M. iigonis are deeply concave anterior to p4. as in all other known eomyids. However, Metanoiamys is relatively primitive in that it retains a small
P3 (or DP3). has fourth premolars distinctly smaller than the first molars, and has a masseteric fossa that extends only to below the p4-m I contact.

Pauromys is confidently identified from the Eocene of southern California for the first time, and is currently known in the region only from the lower
member of the Stadium Conglomerate (late eariy Uintan), where it is represented by the new species P. litlegraveni . An emended diagnosis ^1i Pauromys
emphasizing the very small fourth upper and lower premolars suggests that P. schaubi Wood and at least part of the hypodigms of P. exallos Emry and
Korth, P. texensis Walton, and P. simplex Walton should be excluded from the genus. Since Pauromys has lost P3 and has morphologically "myodont"
molars, it is here removed from the Sciuravidae and classified as Myodonta, incertae sedis.

The abundant southern California late Uintan and Duchesnean rodent Simimys has been assumed to have lost P4, leading some workers to link this
genus to the Muroidca. However, several new maxillary specimens show that San Diego County late Uintan populations of Simimys pos.sessed a tiny,
peglike P4 (and/or DP4). The presence of this tooth in the original Sespe Fomiation species S. simplex, S. velu.s. andS. muriiiusis uncertain, so specific
assignment of the San Diego County material must await recovery of more complete maxillary specimens from the Sespe. The totality of the available
evidence suggests that Sniiimys should be classified as Dipodoidea, incertae sedis.

INTRODUCTION

Eocene rodents from southeni California were first described by
R. W. Wilson and Chester Stock in the 1930s and 1940s (see Golz
and Lillegraven, 1977, for a complete bibliography). These early
collections were obtained from the Sespe Fomiation of Ventura
County (representing the later part of the Uintan and the Ducliesneati
North American Land Mammal "Ages," NALMAs) and from what is
now known as the Friars Fomiation of San Diego County (early
Uintan). See Krishtalka et al. (1987) for a discussion of the Uintan
and Duehesneati NALMAs. Later work on the Eocene rodents of
southeni Califomia was undertaken by Wood ( 1 962 ), Lindsay ( 1 968 ),
Lillegraven and Wilson ( 1975), Lillegraven (1977), Chiment ( 1977),
Emry (1981), Walsh (1987, 1991a,b), Mason (1988), Kelly (1990,
1992), Kelly el al. (1991), Kelly and Whistler (1994), and Chiment
and Konh (1996).

During the past several years, extensive paleontological monitor-
ing of grading operations in San Diego County has been conducted
by PaleoServices, Inc., and the San Diego Natural History Museum.
These salvage efforts have produced important new collections of
Eocene mammals, various aspects of which were discussed by Walsh
(1996). Walsh et al. (1996) provided a reinteipretation of the Friars
Formation and Poway Group (Kennedy and Moore 1971) that is
critical to understanding the stratigraphic provenance of the Uintan
mammals of southwestern San Diego County. The known local

stratigraphic ranges of several rodent taxa I discuss are shown in
Fig. 1. Such data can help illuminate speculations about ancestral
relationships between certain taxa. The purpose of this paper is to
describe important new specimens of three genera of myomorph
rodents from the Uintan of San Diego County and to discuss the
phylogenetic position of these and other selected Paleogene
myomorphs.

METHODS

Specimens will often be introduced in the form "XXXX/
YYYYY," where "XXXX" represents the locality number atid
"YYYYY" the speciinen number. Some of these localities, various
Eocene local faunas, and geographic collecting districts of San
Diego County were discussed by Golz and Lillegraven (1977),
Walsh (1996), and Walsh et al. (1996). Measurements of teeth were
made on an Ehrenreich Photo-Optical Industries "Shopscope" to the
nearest 0.0 1 tiini. Measurements of cheek teeth of Metanoiamys and
Microparamys were taken according to the method used by
Lillegraven (1977:227). Note that the endpoints for tneasuring the
anterior and posterior widths of the lower cheek teeth of
Metanoiamys used by Chitiient and Korth ( 1996, following Chiment
1977) yield slightly smaller values than the endpoints used by
Lillegraven (1977). Measurements of cheek teeth of Pauromys and
Simimys were taken according to the method used by Lillegraven and

Stephen L. Walsh

36
39

4 0-

4 1-

4 2-

4 3-

4 4-

4 5-

4 6-

47
Ma.

t3 O O O O O C

r-> O

O^-^ upper member q

O O O O O "^

o o o _o o o

O O O O

Miramar Sandstone Mttr.

Mission Valley Formation

O O upper member *^^

O O O O o^-^o o
O O lower member 0-_o

upper tongue

lower tongue

H^ 3 Z
h- * —

H

C/l

E

(0
CO

W

5

is

R

ig

->

O

U

-^

tn

t^

w

1

3

o

c
2
1)

ii

5

o

^

Figure 1. Known lilhostratigraphic ranges in southwestern San Diego
County of several rodent taxa disciLssed in this paper. Dashed upper part of
the range of Simimys sp. represents the inferred extent of its bioehronozone.
Estimated numerical ages of various units of the Friars Fonnation and Poway
Group are based on infonnation in Walsh ( 1996), Walsh et al. (1996), and an
unpublished new late Duchesnean( V ) mammal assemblage from the upper
member of the Ponierado Conglomerate. Ep-lc, lower conglomerate member
of the Pomerado Conglomerate.

Wilson ( 1975). Tooth tenninology used for Pauromys and Simimys is
largely that of Lillegraven and Wilson (1975). Abbreviations are as
follows:

N, sample size.

OR, obsei^ed range of variation.

M, arithmetic mean.

SD, standard deviation.

CV, coefficient of variation.

P and M, upper premolar and molar, respectively.

i, p, and m, lower incisor, premolar, and molar, respectively,

D and d, upper and lower deciduous teeth, respectively.

L and R, left and right, respectively.

AP, anteroposterior length of upper and lower teeth.

W, maximum transverse width (for upper premolars of Pauromys

and Simimys).
WTRI, width of Irigotiid.
WTAL, width of talonid.
AW, anterior width of upper cheek teeth.
PW, posterior width of upper cheek teeth.
AMNH, American Museum of Natural History, New York.

CM, Carnegie Museum of Natural History, Pittsburgh.

CU, University of Colorado Museum of Geology, Denver

LACM (CIT), original collections of the California Listitute of

Technology, now housed at the Natural History Museum of Los

Angeles County, Los Angeles.
SDSNH, San Diego Society of Natural History.
UCMP, University of California Museum of Paleontology.
USNM, United States National Museum.
YPM, Yale Peabody Museum.

SYSTEMATIC PALEONTOLOGY

Class Mammalia Linnaeus, 1758

Order Rodeiitia Bowditch, 1 82 1

Suborder Myomorpha Brandt, 1855

Infraorder Geomorpha Thaler, 1966

Family Eoinyidae Deperet and Douxami, 1902

Metanoiumys Chiment and Korth, 1996

Known Distrihuuon. — Uintan to Duchesnean (Middle Eocene)
of western North America.

Type Species. — Melanoiamys agorus Chiment and Korth, 1996.

Included Species.— Metanoiumys agorus Chiment atid Korth,
1996, M. marinus Chitnent and Korth, 1996, M. fantasma (Lindsay,
1968), M.fugitivus (Storer, 1984), and M. lacus (Storer, 1987).

Emended Diagnosis. — Mouse-sized basal eomyid rodent with
generally sciuravid-like cheek teeth (see Chiment and Korth, 1996,
for details). Dorsal surface of matidible anterior to p4 deeply con-
cave as in later eomyids, unlike sciuravids. Ridges defining the
tnasseteric fossa converge and tenninate anteriorly below the p4-ni 1
contact, not below p4 as in later eomyids. Linear dimetisions of P4/
p4 about 10-15% smaller than Ml/ml, notsubequal to or larger than
Ml/ml as in later eomyids. Small P3 retained, not lost as in most
later eomyids. No autapomorphies known.

Discussion. — See Korth (1994) for a review and classification of
Nortli American Eomyidae. Dawson (1977) and Chiment (1977)
showed that specimens from the Sespe Fonnation of California
described by Lindsay (1968) as Namawmys fantasma were not
congeneric with the type species Nanuilomrs lloxdi Black, 1965.
Subsequently, the uimamed genus was infonnally referred to as
"Nanuitomy.s" (Storer 1984, 1987, Kelly 1992, Walton 1993).
Chiment and Korth (1996) erected the genus Metanoiamys to ac-
cotnmodate several species oVNamatomys" from the middle Eocene
of North America.

In addition to their newly described type species M. agorus,
Chiment and Korth named another new species, M. marinus. based
on a stnall sample of isolated teeth from the latest Uintan or
Duchesnean Camp San Onofre local fauna in northwestern San
Diego County (.see Golz and Lillegraven 1977, Kelly 1990, and
Walsh 1996 for discussions of this local fauna). They transferred the
late Uintan "N." fantasma (Hartman Ranch local fauna) to
Metanoiamys as well. Walsh (1991b) referred the SDSNH material
of M. agorus to "cf. Pauromys sp.,'" but the subsequent discovery in
1993 of unequivocal material of Pauromys from San Diego leaves no
doubt that Metanoiamys is indeed generically distinct. Finally, Kelly
(1992) described additional isolated teeth o( Metanoiamys from the
Sespe Fonnation, under the designation of "Namatomys" sp., cf.
"Nl" fantasma. Chiment and Korth (1996) suggested that Kelly's
sample probably pertained to M. marinus. In addition to its occur-
rence in California, Metanoiamys has been reported from strata of
possibly earliest Uintan age in the Turtle Butte Member of the
Bridger Formation (Evanoff et al. 1994), and from the late Uintan

New Specimens nl Mcluiunaniys. Paiinniixs. and Siiiiinixs Inirn the Uintaii (niJLklle Eucenel of San Diegi) Ciiunly

and Duchesnean of Saskatchewan (Storer 1984, 1987). Certain speci-
mens from the Uintan of Texas assigned by Walton (1993) to
Puiiromys may also be referable to Mekiiwicimys (see below).

Chiment and Korth ( 1996, fig. la) based part of the diagnosis of
Mctanouimys on a mandible with il and alleged dp4 (IICMP V-
71211/96399; Figs. 2B, 2D). The cheek tooth in this mandible,
however, is a permanent, not decidnons, premolar, and the sjiecimen
actually pertains to the slightly larger rodent Microparamxs .sp. cf.
M. miiuitus, isolated teeth of which were previously described by
Lillegraven (1977). The mandible of UCMP 96399 differs from
those of Metaiwiamys agorus (e.g., SDSNH 50590; Figs. 2A, 2C) in
its deeper ramus, larger incisor (Table 1), and notably stronger
masseteric ridges. The masseteric fossa in UCMP 96399 extends
anteriad to below the m 1 -m2 contact, while the anterior edge of the
most anterior ridge that defines the fossa extends to below the middle
of 111 1 . The anterior extent of the masseteric fossa in UCMP 96399
matches that in Muropammys triciis and M. duhiiis (Wilson 1940a,
Dawson 1966). The masseteric fossa, however, does not extend as far
forward as in the undoubted mandibles of Metanoiamys described
below.

On the basis of UCMP 96399, Chiment and Korth (1996) in-
cluded in the diagnosis of Metanoiamys the presence of a distinct
ridge extending across the horizontal diastema from immediately
anterior to p4 to just posteriad of the incisor alveolus ( Fig. 2B). This
ridge seems to be a primitive "protrogomorph" character. A similar
ridge is present in the holotype of Microparaims incus (LACM
ICIT] 1122; Wilson 1940a), the holotype of' Reitliroparamys
delicarissiwiis (AMNH 12561; Wood 1962), several SDSNH speci-
mens of Pseuclotoinus littoralis and P. caUfoniicus, and various
species of Paramys. It is also present but weaker in Sciuravus and
Pareumys. The ridge is absent, however, in the three known man-
dibles oi Melanolamys agonis described below (e.g., SDSNH 50590,
Fig. 2A), and the diastema is markedly concave, not horizontal as in
UCMP 96399.

The p4 in UCMP 96399 is unlike that of Metanoiamys agonis but
typical of Micropanimys sp. cf. M, minutus in having a weaker
protoconid, a weaker mesostylid, a lower but more continuous and
more posteriorly situated posterolophid, and a relatively smaller
entoconid and hypoconid that are located at the extreme posterior
comers of the crown. Tlie p4 in UCMP 96399 is also larger in
absolute size (AP 1. 13 mm, WTAL 1.04 mm) than all but one known
p4 of Metanoiamys agonis (UCMP 109876). That the premolar in
UCMP 96399 is a pennanent p4 is demonstrated by its occlusal
outline's being coinpressed anteroposteriorly relative to imdoubted

Table 1. Mandibular depth below ml, and dorsoventral lengths
and transverse widths of lower incisors of Microparamvs,
Metanoiamys. Pauromys, and Simimys from the Uintan of San
Diego County, California. Measurements in mm.

Mandibular
depth

i 1 11 width/

length width length

Microparamvs sp. cf. M. minimis

UCMP V-'7 1 2 11/96399
Metanoiamys ui^onis

SDSNH 3784/49385

SDSNH 3784/.'>0590
Pauromys lillei;ra\eni

SDSNH 3691/46501
Siniimvs sp.

SDSNH 3564/47100

SDSNH 3564/47807

SDSNH 3870/54610

SDSNH 4019/60065

3.8

1,65

0.83

0.50

3.5

_

3.5

1.02

0.59

0.58

3.4

1.20

0.68

0.57

3.3

1.21

0.66

0.55

3.4

—

—

—

3.4

1.25

0.66

0.53

3.3

—

—

—

dp4s of Microparamys sp. ct . M. niiniitns. Given these observations,
one other p4 listed by Chiment and Korth (1996:117) pertains to
Microparamys ^p.cf. M. w(;i;//i(.v (UCMPV-72 157/109615;AP 1.23
mm, WTAL 1.07 mm.). Fortunately, the first known mandibles and
maxillary fragments of Metanoiamys have recently been collected
from San Diego, allowing a revised diagnosis and a reevaluation of
Ihc phylogenetic position of this rodent.

Metanoiamys agonis Chiment and Korth, 1996

Figs. 2A, 2C, 3, 5

Holotype.— yjCM? 106432, an isolated dp4 (not p4 as indicated
by Chiment and Korth 1996).

Type Z.o«7//n'.— UCMPV-72 157, "Poway Pipeline One," Rancho
de los Peiiasquitos district, San Diego County, upper tongue of Friars
Foniiation.

Known Distribution.— EarMest Uintan to late eariy Uintan of San
Diego County, California. Friars Formation, Member B of Santiago
Fonnation, and lower member of Stadium Conglomerate. All UCMP
localities assigned by Chiment and Korth (1996) to the Mission
Valley Fonnation actually pertain to the upper tongue of the Friars
Fonnation (Walsh 1996, Walsh et al. 1996).

Diagnosis. — See dental diagnosis of M agorus given by Chiment
and Korth (1996:118).

Referred Material.— ^GSHH 49385 (mandible fragment with
p4^m3) and SDSNH 50590 (associated mandible with il + p4-ml
and maxilla fragment with Ml -3), both from SDSNH Loc. 3784,
Friars Formation. SDSNH 49630 (maxillary fragment with alveolus
for P3 and complete P4-M1), from SDSNH Loc. 3789, Friars
Fonnation. SDSNH 56201 (mandible fragment with il + p4-m2),
from SDSNH Loc. 3824, conglomerate tongue of Friars Fonnation.
In addition, about 1683 isolated teeth are known from 46 different
SDSNH localities in the Friars Fonnation, 226 isolated teeth are
known from Member B of the Santiago Formation (SDSNH Locs.
3440, 3443, 3448, 3450, 3465, and 3571; Mesa Drive local fauna of
Walsh 1996), and 3 isolated teeth are known from the lower member
of the Stadium Conglomerate (SDSNH Loc. 3691, Murray Canyon
local faima of Walsh 1996).

Mandible and Lower Incisor. — Three mandibular fragments of
M. agonis are now available, two of which are complete enough to
show the sciurognathous condition of the angle (SDSNH 49385,
50590). The relatively weak ridges defining the masseteric fossa
converge and tenninate below the p4-ml contact (Fig. 2C). The
ventral ridge is slightly stronger than the dorsal ridge. The diastema
is deeply concave in SDSNH 50590 (Fig. 2C). This region is dam-
aged in SDSNH 49385 and 56201 but seems to have been distinctly
concave in these specimens as well. On SDSNH 50590 and 56201,
there is a single mental foramen on the lateral side of the mandible,
ventral and well anterior to p4 (this area is broken away on 49385). A
small keel is present on the ventral side of the mandible below the
diastema on SDSNH 50590 (the other two mandibles are damaged in
this area). On SDSNH 49385 and 50590, the coronoid begins to
ascend from the body of the ramus at the level of the m2 talonid, and
both mandibles possess two small foramina between ni3 and the
ascending ramus. There is a distinct "pocket" in the dorsal surface of
the mandible po.sterior to in3, bounded laterally by the ascending
ramus and medially by a distinct ridge running posteriad from the
posterior root of m3 . The depth of both measurable mandibles below
ml is 3.5 mm (Table 1). Unlike the condition in Pauromys and
Simimys (see below), there is no promirtent anteroposteriorly elon-
gate bulge on the medial face of mandible below p4-ml (Fig. 3 A).

Only SDSNH 50590 preserves a complete lower incisor (Figs.
2A, 2C, 3A). It is smaller and not as tranversely compressed as that
of Microparamys sp. cf. M. minutus (Table 1). The ventral face is
distinctly convex, the lateral face is slightly convex, and the medial

Stephen L. Walsh

Figure!. Stercopholographs ot (A), SDSNH 50590, L inaiidiblc vt Mctuiioiuinys ai;unis wnb il + p4-ml, occlusal view; (B), UCMP 96399, R mandible
of Microparamys sp. cf. M. minutus with il + p4, occlusal view; (C), SDSNH 50590, lateral view; (D), UCMP 96399, lateral view. Scale bars = 5 mm.

face is nearly flat. Tlie enamel is ungrooved, extending about one-
third of the way onto the lateral face and only a very short distance
onto the medial face. A small narrow pulp cavity is present.

Lower Premolars. — SDSNH 49385 preserves p4-m3 (Fig. 3B)
and confinns the original description of the lower tooth row based on
isolated teeth (Chiment and Korth 1996). Measurements of cheek

teeth oi Metanoiamys agorus preserved in mandibular and maxillary
fragments are given in Table 2. Chiment and Korth (1996:117)
identified eight lower premolars of M. agorus as deciduous, nine as
peniianent, and 23 as "deciduous or permanent" (although their table
I listed 17 dp4s and 2 1 p4s). Some lower premolars of M. agorus are
indeed difficult to allocate, but most fall into two distinct morpho-

New Specimens of Meuiiwidiiiys. Piiiiniinys. M\d Siiiiiiiiys Iroin the Uintan (niicidle Eocene) of San Diego Counly

^^fl

w y

■a^

9

^^^^^H r^^Sr '<' ^^^^H

1

Figure 3. Stercopliolographs of (A), SDSNH 50590, L mandible of Metanouimys agonis witfi i 1
of Melanoiamys agorus with p4-m3, occlusal view. Scale bars = 5 mm.

p4-ml, medial view, (B), SDSNH 49385, L mandible

logical categories. One niorph is a relatively small anteroposteriorly
elongated tooth with a relatively narrow talonid and usually with a
tiny anteroconid. This niorph is represented by the holotype speci-
men UCMP 106432, which Chiment and Korth (1996, fig. IC)
identified as p4. The other morph is a relatively equant tooth, with a
relatively wide talonid and usually with a weaker or absent
anteroconid. This morph is represented by UCMP 109949, which
Chiment and Korth (1996, fig. IB) identified as dp4. From the
following evidence, however, the more equant morph actually per-
tains to p4, and the aiiteroconid-bearing elongate morph to dp4.

As seen in Micropanimys, Sciiiraviis (Dawson 1968; Lillegraven
1977), Parewnys (Lillegraven 1977), a variety of ischyromyid ro-
dents (Wood 1962), and the eomyids Adjidaitmo and Yoderunys
(Wood 1937, fig. 49; Entry and Korth 1993, figs. 1-5; 1-6), dp4'is
always a relatively long and narrow tooth, more molarifomi than p4,
usually with a stronger anterior cingulid and/or anteroconid. The
same pattern holds for Metanokunys, in which the anteroconid on
dp4 complements the anterocone on DP4 and the relatively long dp4
complements the relatively long DP4. The relatively elongate morph
usually has resorbed roots, as expected for a deciduous tooth, whereas
the more equant morph more often has intact, relatively robust roots,
as expected for a pennanent tooth. Conclusive evidence that the
more equant morph represents the pemianent p4 in Mehmoiamys is
seen in SDSNH 49385, 50590, and 56201, all mandibles having
relatively equant, lightly worn premolars in place with equally woni

or more heavily worn molars. In addition, SDSNH 56201 contains
an equant premolar with long, robust roots and no sign of an
unerupted premolar below it. Accordingly, the lower premolars of
Metanoiainys agorus studied by Chiment and Korth (1996) are
reidentified as follows (asterisks indicate UCMP specimens in-
cluded in Table 3 and Figure 4):

dp4: UCMP 99201, 101288*, 101564*, 106060*, 106260*,
106800*, 106905*, 106925*, 109914, 109765*, 110170*.

p4: UCMP 96262*, 96389, 101138*, 101157*, 101202,
101207*, 101280, 104588*, 106254*, 106427*, 106449*, 106450,
106823*. 106922*, 109615, 109635*, 109762, 109876*, 109949*,
110221, 110254*.

p4 ordp4: UCMP 101190, 101139, 105964, 106265, 106827,
106832, 110220.

Given the removal of two relatively large p4s of Microparamys
sp. cf. M. minuttis from the original hypodigm of M. agorus, and
given that most teeth identified by Chiment and Korth ( 1 996) as dp4s
are actually p4s (and vice versa), the .statistics for the lower premolars
of M. agorus provided by these authors are not meaningful. There-
fore, I remeasured AP and WTAL on the confidently reidentified
complete lower premolars in the original hypodigm. To these were
added measurements of a new sample of lower premolars of M.
agorus from several SDSNH localities in the upper tongue of the
Friars Fonnation. The resulting statistics for this combined sample

Stephen L. Walsh

A

= Microparamys sp, cf, M- mmutus dp4s

o

= Microparamys sp, cf, M. minutus p4s /

1.20-

A

= Metanoiamys agorus dp4s j/^ O

•

= Metanoiamys agorus p4s / Q

1.15-

(D

= UCMP 96399 y/^ "^ O O

e

= UCMP 109615 y^ ° °

1.10-

o/ o8*o °° °
/ ,JP 0 o o

E
F

1.05-

/<Sfo " .

I
1-

1 .00-

^.••.•O o A A

y/^» " ■ " •

.95 -

/ • • •• ^

Z

2

.90 -
.85 -

/

"^•^ •Na lAi^ A
. • . ■ ^ A^A

^ I^ ^ A

.80 -

A mJ^\ a "^
^ A AJ^iJ^A

^ M A

.75 -

** .

.70 -

» 1

*l 1 1 1 1 1 1 1

.8

5 .90 ,95 1,00 1,05 1,10 1,15 1,20 1,25 1,30 1

ANfTEROPOSTERIOR LENGTH (mm)

Figure 4. Anteroposterior length versus talonid width for permanent and
deciduous lower premolars of Metanoiamys agorus and Microparamys sp,
cf, M. minulus. Note slight overlap in size ranges of each category and the
distinctly smaller average size of the Metanoiamys premolars. Specimen
numbers of teeth represented are given in Tables 3 and 4.

are given in Table 3.

A pooled sample of lower premolars of Microparamys sp. cf. M.
minutus from several SDSNH localities from all lithostratigraphic
subdivisions of the Friars Formation was also measured (Table 4) to
quantify the metric differences between the two species better.
Bivariale plots of the data used to compile Tables 3 and 4 are
illustrated in Fig. 4. Although a slight overlap is evident, the four
categories cluster at distinctly different regions on the graph. Note
that for both species p4 tends to be more equant (plot closer to the
line of equal length and width) than dp4.

Maxillae and Upper Premolars. — Two maxillary fragments of
M. agorus are now available. SDSNH 50590 (not figured) contains
Ml -3 and confinns the original description of the upper molars
based on isolated teeth (Chiment and Korth 1996) but preserves no
other relevant characters. SDSNH 49630 (Fig. 5) preserves an empty

alveolus at the P3 locus plus complete, unworn P4-M1. SDSNH
49630 is barely complete enough to show that the posterior edge of
the inferior zygomatic root originated anterolateral to P3. In contrast
to the condition in Pauromys and Simimys (see below), there was no
distinct knob for the insertion of the masseter lateralis supert'icialis
immediately anterior to P3, but not enough of the maxilla is pre-
served to determine whether such a knob was present immediately
ventral to the infraorbital foramen as in Paradjiclaumo and
Aidolithomys (Wood 1974:74).

Although the empty P3 alveolus in SDSNH 49630 is relatively
large, it seems to be somewhat damaged, and P3 (and/or DP3) was
presumably single-rooted, with a small crown of unknown morphol-
ogy. The presence of DP3 in juveniles of A/, agorus is established by
the occurrence of small, often very subtle, anterior appression facets
on 24 of the 80 DP4s in SDSNH collections for which the status of
this feature can be detennined. (e.g., SDSNH 43030, 43175, and
51002). Contra Chiment and Korth (1996:117), two DP4s in the
original hypodigm also have such a facet (UCMP 99202 and 106669).
That adults of M. agorus possessed a tooth in the P3 locus is shown
by small anterior appression facets on 47 of 104 detemiinable P4s in
SDSNH collections (e.g., SDSNH 39221, 43186, and 51007). Om-
tra Chiment and Korth ( 1996: 117), four P4s in the original hypodigm
also have such a facet (UCMP 96256, 96328, 96478, and 101 177).
Whether the tooth in the P3 locus in adult animals is a retained DP3
or a peniianent P3 is uncertain.

The inconsistent occurrence of anterior appression facets on P4/
DP4 of Metamnamys agorus is probably attributable to ( 1 ) death of
some individuals prior to full emption of the premolars and (2) slight
individual variations in the relative positions of DP3/DP4 and P3/P4.
This conclusion is supported by the evidence seen in isolated premolars
of S(7MwiY<i/)<)uwve/LS7s (see Wilson 1940b, Lillegraven 1977). Even
though this species is known to possess a DP3 replaced by a pennanent
P3, only 23 of 33 detemiinable DP4s and 17 of 46 detemiinable P4s in
SDSNH collections have a detectable anterior facet.

It should be noted that eight of the M3s referred by Chiment and
Korth (1996:118) to Metanoiamys agorus actually pertain to
Microparamys sp. cf. M. minutus (UCMP 96258, 96267, 96402,
101281, 106861, 109565, 109815, and 110219). These teeth are
larger than M3s of Metanoiamys agorus, often have crenulated
enamel and taller paracones, and are generally more anteroposteriorly
elongated. Finally, UCMP 99343 does not pertain to M. agorus but is
a heavily damaged M3 probably referable to Sciuravus.

Discussion. — The presence of a P3 alveolus in SDSNH 49630,
together with the presence of frequent appression facets on the
anterior faces of isolated DP4s and P4s, clearly shows that the early
Uintan species Metanoiamys agorus retained the primitive rodent
condition of possessing two upper premolars. Lindsay (1968) de-
scribed P4s of the late Uintan Sespe Fomiation species M. fantasma
as lacking an anterior appression facet. Of the 12 detemiinable P4s
of this species from UCMP V-5814, however, 7 have small but

Table 2. Measurements (mm) of cheek teeth of Metanoiamys agorus in maxillary and mandibular fragments.

P4

Ml

M2

M3

A?

AW

PW

AP

AW

PW

AP

AW

PW

AP

AW

PW

SDSNH 49630
SDSNH 50590

1.03

1.04
p4

1.03

1.19
1.09

1.17
1,16

ml

1.22
1.21

1.10

m2

—

0.92

1.02
m3

—

AP

WTRl

WTAL

AP

WTRI

WTAL

AP

WTRI

WTAL

AP

WTRI

WTAL

SDSNH 50590
SDSNH 49385
SDSNH 56201

0.99
090
1.00

0.77
0.73
0.78

1.00
0.91
1.00

1.13
1.12
1.09

1,01
1.02
0.90

1.09
1.10
0.91

1.14
1.15

1.08

1.13
111

1.08

1.01

0.95

New Specimens of Mckinoiaiiiys. I'aiironiys, and Siiiiiniys I'nini the Llintan (niidillc Eocene) cil .San Diego County

Figure 5. Stcreophotograpli in occlusal view of SDSNH 49630, R maxillary fragment of Mclaiutiumys Uf^orKS with alveolus for P3 (or DP3) + P4-M1.
Scale bar = I mm.

di.stinct anterior facets, indicating the presence of P3 and/or DP3
(UCMP 77648-77650, 77652, 91685-91686, and 11 6482). The sole
DP4 from this locahty also has a faint anterior facet (UCMP 79504).
No P4s or DP4s are present in the recently collected Tapo Canyon
and Brea Canyon samples of isolated teeth of Metcmoiamys de-
scribed by Kelly (1992). Finally, of the five P4s and one DP4 of
Metcmoiamys marinus from V-72088, none has an unequivocal ante-
rior facet. More complete material must be collected from the
superposed local faunas of the Sespe Fonnation for the persistence
of P3/DP3 in this genus from late Uintan to Duchesnean time to be
evaluated.

When the first species of Melanoiamys was described from the
late Uintan of southern California (under the name Namatomys
fantasma), Lindsay ( 1968) suggested it may have been derived from
the early Uintan rodent Sciuravus powayensis. As discussed by
Storer (1987), however, the origin of Metanoiamys is more likely to
be found among Wasatchian-Bridgerian species of the sciuravid
Knightomxs. Indeed, upper molars from the Wasatchian of New
Mexico assigned by Flanagan (1986) to K. reginensis appear to be
quite similar in size and morphology to those of M. agorus. and the
collection of complete mandibles and maxillaries of K. reginensis is
awaited.

On the basis of its presumed lack of P3, Storer (1987) assumed
that the Uintan Metanoiamys could not be ancestral to the aberrant

Chadronian eomyid Yoderimys, which retains P3 (Wood 1974). He
therefore suggested that the stem eomyid must be of Bridgerian age.
The documented presence of P3 in Metanoiamys, however, now
removes all objections to this genus as representing the basal eomyid
niorphotype. Another primitive sciuravid character of Metanoiamys
is the distinctly smaller size of p4 and P4 relative to ml and Ml. In
Yoderimys and later eomyids, the premolars are subequal to or larger
than the first molars (Wood 1974, Emry and Korth 1993). Despite
these primitive retentions, the dorsally concave mandibular diastema
is a derived character linking Metanoiamys with later eomyids (e.g.,
Aidolitliomys and yiejadjidaiinun'Wood 1974). More complete ma-
terial will be required to detemiine if Metanoiamys had already
evolved the sciuromorphous zygomasseteric structure typical of later
members of the family (e.g.. Wood 1974).

Infraorder Myodonta Schaub, 1958

Myodonta ineertae sedis

Pauromvs TroxeW, 1923

Txpe Species. — Paiiromys perditiis Troxell, 1923.
Included Species. — Pauromys perditiis Tro.xell, 1923, Pauromys
sp. [unnamed Powder Wash species; Dawson 1968], and P. lillegra-

Table 3. Statistics for lower premolars of Metanoiamys agorus.
based on reidentified specimens from original UCMP hypodigm,
combined with sample from several SDSNH localities in the upper
tongue of the Friars Formation (Rancho Pehasquitos and Carmel
Mountain Ranch districts)."

Table 4. Statistics for lower premolars of Microparamys sp. cf. M.
minutus, based on sample from several SDSNH localities in all
lithostratigraphic subdivisions of the Friars Formation."

dp4

P4

AP

WTAL

AP

dp4

p4

N

OR

M

SD

CV

22

0.99-1.27

1.13

0.080

7.1

22

0.81-1.02

0.89

0.082

6.9

34

1.07-1.31

1.18

0.065

5.5

AP

WTAL

AP

WTAL

34
1.00-1.19

N
OR

M

48

0.90-1.07

0.99

0.044

4.5

47

0.70-0.90

0.80

0.046

5.7

45

0.91-1.15

1.02

0.055

5.4

45

0.85-1.10

0.95

0.049

5.2

1.08
0.051

4.7

SD
CV

"Specimens measured, dp4s: Loc. 3254;
3391: 32600. Loc. 3414: 41319. Loc.

; 26901. Loc. 3373
3482: 37624. Loc.

: 31564. Loc.
3483: 38785.

"SDSNH specimens measured, dp4s: Loc. 3254: 26909, 26919, 26960,
26983, 26991, 27125, 27684, 27722, 27736, 27055. Loc. 3391: 32562,
32571, 32572, 32584, 32612. Loc. 3482: 37654. Loc. 3483: 39013-
39015, 39017-39020, 39022, 39023, 39025-39029. 39031-39033,
39035-39038. Loc. 3771 : 4727 1 , 47274, 47275.

p4s; Loc. 3254: 26897, 26898, 26900, 26964, 27735. Loc. 3373: 31423,
31429, 31439, 31433, 31447. Loc. 3391: 32564, 32589, 32601. 32608,
32618, 32621. Loc. 3482: 37655, 37658. Loc. 3483: 39039, 39041-
39046, 39051-39053, 39055, 39057. Loc. 3771: 47277.

38787, 38790, 38792. Loc. 3494: 37402. Loc. 3591: 43962. Loc. 3611:
45920-45922. Loc. 3617: 43101, 43102. Loc. 3621: 58037, 58038. Loc.
3655: 46156, 46157. Loc. 3656: 46298, 46299.

p4s: Loc. 3254: 26903. Loc. 3373: 31413, 31544, 31545. 31549, 31565.
Loc. 3380: 42441. Loc. 3483: 38793, 38794, 38798-38801, 38803-
38805. Loc. 3611: 45176. Loc. 3612: 45563, 45565 Loc. 3883: 55691.
55692. Loc. 3616: 4301 1. Loc. 3621: 58039-58041, 58043. 58044. Loc.
3655: 46158, 46159, 46161. 46301. Loc. 3658: 45839, Loc. 3893:
55176,55313.

Stephen L. Walsh

veni sp. nov.

Known Distribution. — Early Bridgerian of the Green River For-
mation, Utah. Early or late Bridgerian of the Bridger Foniiation,
Wyoming. Early Uintanof the AdobeTown Member of the Washakie
Fonnation, Wyoming (M. R. Dawson, pers. comm.). Late early
Uintan of the lower member of the Stadium Conglomerate, Califor-
nia. Questionably known from the Bridgerian of Nevada (Emry and
Korth 1989), and the Uintan of Texas (Walton 1993) and Utah
(Walsh 1996, table 4).

Emended Diagnosis. — Mouse-sized basal inyodont rodent. Mo-
lars of generalized myodont construction. Dorsal surface of man-
dible virtually horizontal anterior to p4 but without distinct ridge as
in Sciiiravus, Micropciramys, and other ischyromyids. Ridges defin-
ing the masseteric fossa extend anteriorly to below the ml talonid.
Derived with respect to Armintomys in the loss of P3 and the
reduction of P4/p4. Primitive with respect to Simimys, Elymys,
Pappocrici'todon, and Nonomys in retaining a three-rooted P4 (al-
though with an anteroposterior length only about 50% of Ml ), a p4
(although with an anteroposterior length only about 50% of ml), and
having a relatively wider ml trigonid. No autapomorphies known.

Pauromys lillegraveni sp. nov.
Figs. 6, 7A

Holot^'pe. —SDSNH 46501, L mandible fragment with i 1 + p4-
m3.

Type Locality. — SDSNH Loc. 3691, "Murray Canyon 1," lower
member of Stadium Conglotnerate.

Known Distribution. — Lower member of the Stadium Conglom-
erate (late early Uintan), San Diego County, California.

Etymology. — Species named for Dr Jason A. Lillegraven, for his
contributions to vertebrate paleontology.

Diagnosis. — Lower p4 comparable in absolute and relative size to
that oi' P. perditus, averaging distinctly smaller than that of P. sp. from
Powder Wash. Lower p4 roots completely fused at level of alveolar
border, not separate as in P perditus. Lower p4 with single trigonid
cusp (no metaconid), unlike P. perditus and P. sp. from Powder Wash.
Lower p4 with no ectolophid or mesoconid, unlike P. sp. from Powder
Wash. Entoconids of m 1 -2 conical, not anteroposteriorly compressed
as in P. perditus. Length of m3 averaging shorter than m 1 and m2,
apparently unlike P. perditus and P. sp. from Powder Wash.

Referred Matericd. SDSNH Loc. 3691: SDSNH 47708-47709,
p4s. 46530-46532, mis. 46533, m3. 47710, P4. 46502, maxilla frag-
ment with P4-M1. 46535, MI. 46503 and 46534, M2s. 46536, M3.

SDSNH Loc. 3701: SDSNH 48011-48012, mis. 48016, m2.
48013, Ml.

SDSNH Loc. 3731: SDSNH 56685, P4. 56686, Ml. 56687, M2
fragment.

Mandible and Lower Incisor. — The holotype mandible shows the
sciurognathous condition of the angle (Fig. 6B). The relatively weak
ridges defining the masseteric fossa converge and terminate below
the m 1 talonid (Fig. 6C).The ventral ridge is slightly stronger than the
dorsal ridge. A single mental foramen is present at the dorsoventral
midpoint of the lateral side of the mandible, directly below the
anterior edge of the root of p4. Tlie diastema anterior to p4 is virtually
horizontal but lacks the distinct ridge present in Microparamys,
Sciuravus, and a variety of ischyromyid rodents (Fig. 6A). The
coronoid process begins to ascend from the dorsal border of the
ramus at about the middle of m3, as in Simimys (and not at the level of
the middle of m2 as in Metanoiamy.s). As in Simimys (see below),
there is a single foramen between m3 and the ascending ramus
(obscured by matrix in Fig. 6A). An apparently similar foramen in
Plesiosminthus myarion was identified by Schaub (1930, fig. 9) as an
opening for the mandibular canal. Unfortunately, the other two
known mandibles of Pauromys (YPM 13601 and CM 19568) are
damaged in this area Also as in Simimys, there is a prominent

anteroposteriorly elongated bulge on the medial face of the mandible
below ml, p4, and the posterior part of the diastema (Fig. 6B). With
two minor exceptions, the mandible of SDSNH 4650 1 agrees closely
in all comparable characters with CM 19568, the edentulous man-
dible from Powder Wash assigned by Dawson ( 1968) lo Pauromys sp.
The ventral keel noted by Dawson (1968) on CM 19568 is better
developed than on SDSNH 46501, while the mental foramen on CM
19568 occurs slightly higher on the mandible than it does in SDSNH
46501.

SDSNH 46501 preserves a complete lower incisor (not figured).
The tip was broken off during the screen-washing process and was
picked out separately but has not been reattached. The ventral face is
distinctly convex, the lateral face is slightly convex, and the medial
face is very slightly convex. The enamel is ungrooved, extending
about one-fourth of the way onto the lateral face and only a very
short distance onto the medial face. A small narrow pulp cavity is
present. The lower incisor of Pauromys lillegraveni is similar in
proportions to that of Metanoiamys agorus but distinctly larger, and
very similar in proportions and size to the lower incisor of Simimys
sp. (Table 1).

Lower Cheek Teeth. — The lower cheek teeth of Pauromys
lillegraveni are siinilar in proportions and general morphology to
those of /?/)cr(//mi- (see Troxell 1923, fig. UDawson 1968, figs. 39-
40). Three p4s are known, all of which are much smaller than ml
(Table 5). The p4 in the holotype mandible is well worn (Fig. 6A), but
the two isolated p4s are unworn, and both show only a single small
trigonid cusp. This condition differs from that in Pauromys perditus
and P. sp. from Powder Wash, in which both the protoconid and
metaconid are reportedly present on p4 (Dawson 1968:358). All three
teeth are rounded and subtriangular in occlusal outline. The median
valley is anteriorly concave and has no accessory cuspules or lophids.
The hypoconid and entoconid are subequal in size, conical, and
connected to one another by an anteriorly concave hypolophid. A
short posterior cingulid is present only on SDSNH 47709. There is
only a single trunk root of p4 visible above the alveolar border in
SDSNH 46501, although a slight constriction on the labial face of the
root suggests the possibility of a partial ventral bifurcation. Neverthe-
less, the condition of the p4 root in SDSNH 4650 1 differs from that
seen in the type mandible of P. perditus, in which two distinct roots
are present immediately below the crown (Dawson 1968, fig. 40).

Six complete mis are known, although SDSNH 48011 is too
worn to show any significant details of the crown. Distinct appression
facets on the anterior faces of the five isolated mis preclude their
identification as Simimys. The metaconid is always essentially coni-
cal and lacks a metalophulid. A weak lingual lophid extends from the
posterior face of the metaconid to the metastylid region on all five
determinable mis. A weak transversely compressed metastylid is
present at the posterior end of the lingual lophid on all four determin-
able mis. The entoconid is usually slightly elongated transversely
and is subequal to the metaconid. A moderately strong
postprotocristid is present on all five detenninable mis. In all three
unworn mis, the postprotocristid extends only about halfway up the
posterior face of the metaconid. The unworn mis have a weak but
transversely elongate anteroconid, which assumes the morphology
of an anterior cingulid with wear There is no preprotocristid con-
necting the protoconid with the anterior cingulid. There are no
prominent lingual or labial mesolophids on any ml (unlike many
teeth of Simimys), so the term mesoconid is used for the entirety of
the small, usually slightly transversely elongate ridge present in the
labial half of the median valley. The mesoconid is connected to the
hypoconid by a short posterior ectolophid in three of four determin-
able mis; in the remaining ml, the mesoconid is small, conical, and
isolated. The hypoconid is subequal to the protoconid and entoconid
and is conical to slightly anteroposteriorly compressed. A weak
hypolophid connects the hypoconid with the hypoconulid in all five
detemunable mis; there is never a direct connection between the

New Specimens of MekiiKiiniiiys. Pauroiiiys, ami Siniiniys Irom llie Uintan (middle Eocene) ot San Diego County

Figure 6. Stereophotographs of SDSNH 46501 , holotype L mandible of Pauromys lillegraveiu with i 1 + p4-m3: (A), occlusal view; (B), medial view; (C),
lateral view. Scale bar = 5 mm.

hypocoiiid and entoconid. The hypoconulid is a weak to moderate
swelling of the short posterior cingulid. A very narrow labial shelf is
present in all three deteniiinable mis.

Two m2s are known, differing from ml in that the trigonid is
subequal in width to the talonid, the anterior cingulid is much longer
transversely, a weak preprotocristid connects the anterior cingulid
with the protoconid, a metastylid is absent, and a direct hypolophid
connection between the entoconid and hypoconid is present on
SDSNH 48016.

At least two m3s are known. Both are anteroposteriorly shorter

than ml -2. Both lack a preprotocristid, have a much smaller ento-
conid than in ml -2, and have a trigonid wider than the talonid. An
isolated m3 from SDSNH Loc. 3701 (SDSNH 4799 1 ) differs from
the above in3s in that it is anteroposteriorly longer, has a
preprotocristid, and a strong posterior ectolophid. It may or may not
pertain to Pauromys.

Maxillary. — A single maxillary fragment is available (SDSNH
46502, Fig. 7A). A weak ridge is present at the medial edge of the
maxillary. The postero-medial edge of the fragment is probably
broken along the palatine-maxillary suture. The antero-medial edge

10

Stephen L. Walsh

Figure 7. Stereophotographs of (A), SDS>fH 46502, L maxilla fragment of Paiiromys lillegraveni with P4-M1, occlusal view. Scale bar = 1 mm. (B),
SDSNH 492 10, ventral view of partial skull of Sciuravus powayensisshov/ing features of the anterior zygomatic root. Note knob for origin of masseter lateralis
supertlcialis located anterolateral to P3. Scale bar = 5 mm.

of the fragment is apparently broken across the maxillary, and there
is no indication of the posterior end of the left incisive foramen.
Anterior to P4, a small damaged knob of bone seetns to be present in
the same relative position as the prominent knob for the attachment
of the masseter lateralis superficialis in Simimys (see below). A
similar knob also seems to be present in the maxilla of Paummys sp.
illustrated by Dawson (1968, fig. 41) from Powder Wash, hi
Pauromys the knobs are located more directly anterior to P4 than in
Sciuravus nitidus (Dawson 1961, plate III) and S. powayeiisis (Fig.
7B). Immediately anteromedial to the datiiaged bony knob on
SDSNH 46502 is a prominent depression in the maxilla, siinilar in
location and orientation to but deeper than that seen in Sciuravus
powayensis. Not enough of the maxilla is preserved on SDSNH
46502 to reveal whether the anteroventral edge of the inferior zygo-
matic root was distinctly ridged as in Sciuravus and Pauromys sp.
from Powder Wash. No upper incisors oi Pauromys lillegraveni have
been identified.

Upper Cheek Teeth. — Three P4s are known. Two of them are

isolated and imwom, while one is present in a maxilla fragment and
somewhat worn (SDSNH 46502). The teeth are three-rooted, with
subequal anterolabial and posterolabial roots and a larger lingual
root. The anterolabial root juts out from the base of the crown in an
anterodorsal direction, as implied for Pauromys sp. by Dawson
(1968, fig. 41). Tlie crown is anteroposteriorly coinpressed, oval in
occlusal outline, and quite small relative to MI. A distinct anterior
cingulum is present on SDSNH 47710 but not on SDSNH 56685.
The protocone is roughly conical and is connected to the
anterolingual base of the paracone by a distinct preprotocrista. The
paracone is subequal to the protocone. A postprotocrista extends
posteriad from the protocone apex to merge into the posterior cingu-
lum, which ends at the posterolingual base of the metacone. There is
no liypocone. The metacone is slightly compressed anteroposteriorly
and slightly larger than the protocone and paracone. SDSNH 47710
has a distinct metaconule fused to the anterolingual base of the
metacone, but a metaconule is absent in SDSNH 56685.

Ml -2 are best distinguished with reference to SDSNH 46502 (L

New Specimens o'i Metaiiouiinys. Puuroinys. and Siniiiiiys fniiii llic Uinlaii (middle Eoeeiie) ol San Diegi) County

maxillary fragment with P4-M1; Fig. 7A)and SDSNH 46503 (LM2;
not figured). The latter tooth is probably associated with the maxilla,
as both were found in the same 30 kg-batch of screen- washed matrix,
have similar crown colors, are worn to the same degree, and show a
similar dissolution of parts of their labial faces. In general, these
teeth are quite similar in crown and root morphology to the M 1 -2s of
Simimys described by Lillegraven and Wilson ( 1975).

Three teeth are identified as Mis. As in Siniiniys, the anterior
width is slightly less than the posterior width. SDSNH 46535 is
complete and iinwoni. It has a strong anterior cingiilum and no
aiiterocone. The protocone is strong and conical, sending a strong
preprotocrista to the anterolingual base of the paracone. There is no
protoconule. The paracone is conical and subequal to the protocone.
The hypocone is strong, conical, and taller than the protocone. The
posthypocrista extends from the hypocone apex to become the
posterior cingulnm, which then extends to the anterolabial base of
the metacone. A centrally located mesocone is connected to the
hypocone by a short prehypocrista (posterior mure of some authors).
A minute mesostyle is present. The metacone is slightly larger and
taller than the paracone.

Two M2s are known; they differ from M 1 in being slightly wider
anteriorly than posteriorly, and their metacones and hypocones are
shorter than their paracones and protocones, respectively. A
postprotocrista (anterior mure of some authors) is present in both
M2s, connected to the mesocone. In SDSNH 46534, the prehypo-
crista connects with the lingual base of the metacone.

A single isolated M3 may pertain to P. lillegraveiii (SDSNH
46536) but it is heavily worn and cannot be confidently distinguished
from Metanoiamys. Measurements of the cheek teeth of Pauromys
lillegraveni are given in Table 5.

Discussion. — Walton (1993) noted that the highly reduced p4 in
the type specimen oi P. perdilus may be diagnostic of the genus. Tliis
condition is also present in P. lillegraveni, corroborating the biologi-
cal reality of this character. Accordingly, the inorphological concept
of Pauromys adopted here is narrower than that used by most recent
workers, and I exclude from the genus certain species previously
referred to Pauromys.

Pauromys schauhi was named by Wood (1959) on the basis of
AMNH 11722, a mandible fragment with the roots of p4 and
complete ml -2, from the Twin Buttes Member of the Bridger
Formation. Wood's ( 1959, tig. 1 ) illustration shows that the p4 of "P."
schauhi must have been substantially larger than that of either P.
pcrditus or P. lillegraveni, while the ml trigonid was not as narrow
relative to the talonid as in the latter two species. The cheek teeth in
AMNH 11722 are also similar in size, proportions, and general
morphology to those of Melanoianns and certain species of
Apato.sciiirarus and Knightomys (e.g., Flanagan 1986, figs. 3-5).
Until more complete material is recovered, "P." schauhi is here
tentatively excluded froiii Pauromys.

Dawson (1968) described a large sample of isolated teeth from
the Green River Formation of Utah (early Bridgerian Powder Wash
locality). She conservatively assigned the specimens to Pauromys
sp., and suggested that larger samples from the Bridger Basin might
grade morphologically into the Powder Wash sample. Nevertheless,
Dawson noted that the available Powder Wash p4s were larger and
had better developed trigonids (i.e., were more primitive) than the p4
in the type of P. perdilus. This situation might constitute a stage-of-
evolution argument for a relatively younger age for the poorly
constrained type locality of P. perdilus (Dry Creek). Unfortunately,
the latter can be restricted only to Bridger B or C (Dawson 1968:354,
Gazin 1976:10).

Nelson ( 1974) assigned six isolated teeth from the Fowkes For-
mation of Wyoming (later Bridgerian) to Pauromys sp. Although the
two molars illustrated by Nelson ( 1974, fig. 10) are morphologically
consistent with this generic assignment, there are no fourth premolars
in the sample. Korth (1984) assigned nine isolated teeth from the
Wind River Fonnation of Wyoming (Wasatchian) to Pauromys sp.
Unlike the situation in undoubted species of Pauromys, however, the
p4 illustrated by Korth (1984, tig. 29B) is only about 15% shorter
than ml. Finally, Flanagan (1986) assigned a single isolated upper
molar from the San Jose Fonnation of New Mexico (Wasatchian) to
Pauromys sp. More complete material of the species described by
Nelson ( 1974), Korth (1984), and Flanagan (1986) must be collected
for their assignment to Pauromys to be coiToborated.

Table 5. Measurements of cheek teeth of Pauromys lillegraveni.

P4

Ml

M2

M3

AP

W

AP

AW

PW

AP

AW

PW

AP

AW

SDSNH 47710

0.70

0.71

—

—

_

—

—

—

SDSNH 56685

0.69

0.70

—

—

—

—

—

—

—

—

SDSNH 46502

0.66

0.78

1.17

0.96"

0.99"

—

—

—

—

—

SDSNH 46535

—

—

1.22

1.08

1.15

—

—

—

—

—

SDSNH 48013

—

—

1.18

1.15

1.18

—

—

—

—

—

SDSNH 46503

—

—

—

—

—

1.17"

1.17"

1.13

—

—

SDSNH 46534

—

—

—

—

—

1.25"

1 .09"

1.02"

—

—

SDSNH 46536

~

P4

"

ml

"

"

m2

0.97

0.97
m3

AP

WTAL

AP

WTRI

WTAL

AP

WTRI

WTAL

AP

WTRI

WTAL

SDSNH 47708

0.57

0.53

—

—

—

—

—

—

SDSNH 47709

0.69

0.68

—

—

—

—

—

—

—

—

—

SDSNH 46501

0.55

0.62

1.17

0.90

0.94

1.21

1.05

1.08

1.09

0.97

0.87

SDSNH 46530

—

—

1.17

0.81

0.94

—

—

—

—

—

—

SDSNH 46531

—

—

1.20

0.79

0.92

—

—

—

—

—

—

SDSNH 46532

—

—

1.29

0.79"

0.92"

—

—

—

—

—

—

SDSNH 48011

—

—

1.18

0.82

0.99

—

—

—

—

—

—

SDSNH 48012

—

—

1.15

0.76

0.90

—

—

—

—

—

—

SDSNH 48016

—

—

—

—

—

1.21

0.97

1.05

—

—

—

SDSNH 46533

-

—

—

—

—

—

—

—

1 .09"

1 .03"

0.78"

"Value a minimum owing to slight damage.

12

Stephen L. Walsh

Eniry and Korth (1989) named the species Pauwmys e.xallos on
the basis of eleven isolated teeth from the Sheep Pass Fomiation of
Nevada (Bridgerian). It is uncertain whether this generic assignment
is correct. The p4 (USNM 336401; Entry and Korth 1989. fig. 4C)
referred to "P." exullos is subequal in size to tlie molars assigned to
this species and is relatively much larger than the p4s in P. perditus
and P. lillegmveni. Therefore, if USNM 336401 is in fact conspecific
with the other teeth assigned by Eniry and Korth to "P." e.xallos, then
this species is clearly excludable from Pauromys. hi my opinion,
however, USNM 336401 probably pertains to Microparamys
sambiaus (e.g., compare with the p4 in UCMP 96399 and numerous
p4s of M. sp. cf M. miimtus and M. woodi in SDSNH collections).
Siinilarly, an M3 figured by Emry and Korth ( 1989, fig. 4B; USNM
336429) has the posterior elongation typical oi Microparamys and is
distinctly larger than USNM 417474 and 404695, isolated M3s that
are more securely assignable to "P." exallos. While it is still possible
that "P." e.xallos is a valid species of Pauromys possessing greatly
reduced but as yet uncollected premolars, this generic assignment
has not been confidently established.

Walton (1993) named two new species of Pauromys from the
Uintan of Texas, P. te.xensis (early and late Uintan) and P. simplex
(late Uintan). From her figures and some casts, it appears that some
specimens of P. le.xensis and P. simplex may be referable to genera
other than Pauromys. For example, TMM 41745-54 and TMM
41745-102 (Walton 1993, figs. 6D-E), are more similar in mor-
phology and relative size to dp4 and p4 of Metanoiamys af^orus
than to p4 oi Pauromys perditus or P. lillegraveni. TMM 41745- 145
(Walton 1993, fig. 6A), a probable DP4, closely resembles the DP4
of Melanoiamvs in its trapezoidal occlusal outline and size relative
to the molars. Similarly, TMM 41745-383, identified by Walton
(1993, fig. 60) as an M2? of an indeterminate sciuravid, has the
typical quadrate occlusal outline of a P4 of Metanoiamys. On the
basis of their relatively large size, if the DP4 and dp4 referred by
Walton (1993, fig. 8J-K) to P. simplex are in fact conspecific with
the rest of the hypodigm, then this species is definitely not assign-
able to Pauromys. Similarly, the lower molars and M2 off! simplex
illustrated by Walton ( 1993, figs. 8D, 8G, 8H) are quite similar to
those of Metanoiamys agorus in size, occlusal outline, and simple
construction of the median valley. M. R. Dawson (pers. comm.)
also suggests that some of the Texas specimens are referable to an
undescribed, relatively advanced sciuravid genus known from the
late Uintan of Badwater, Wyoming. As noted by Walton (1993),
collection of mandibles and maxillae of the Texas species will be
necessary to establish the identity of these teeth confidently.

Pauromys has traditionally been assigned to the Sciuravidae, which
is presumably a paraphyletic taxon characterized in part by the primi-
tive retention of two upper premolars. The general morphology of the
molars of Pauromys, however, is very similar to that seen in Simimys
and other imdoubted Myodonta. The absence of P3 is also a derived
character shared with other myomorphs. Thus, I regard Pauromys as a
primitive member of the Myodonta, slightly more derived than
Armintomys. Although the zygomasseteric structure of Pauromys is
unknown, this genus seems likely to have descended from a small-
bodied hystricomorphous rodent similar to Armintomys but with
ungrooved incisors The presumed hystricomorphy of Pauromys is
consistent with the position of the anterior end of the masseteric fossa
in this genus (below the ml talonid) being identical to that in the
hystricomorphous rodent Simimys (see below).

Several authors have commented on the possibility of a close
relationship between Pauromys and Simimys (Wilson 1949, Dawson
1968, Lindsay 1968, Walton 1993). Other workers have proposed
that Simimys was an immigrant from Asia, without a North American
ancestry (Vianey-Liaud 1985). With the discovery of Pauromys
lillegraveni, the first hypothesis seems more attractive. As discussed
above, there are detailed similarities in the mandible, lower incisor,
ventral zygomatic root, and molars between the late early Uintan P.

lillegraveni and the late Uintan Simimys sp. in San Diego. The latter
genus is known to differ from the former only in its more derived loss
of p4, narrower ml trigonid, reduced P4 (or DP4), and greater
diastemal concavity. Pauromys lillegraveni possesses no known
autapomorphies that exclude its being directly ancestral to Simimys.

Superfamily Dipodoidea Weber, 1904
Dipodoidea incertae sedis
Simimys (Wilson, 1935a, b)

Type Species. — Simimys simplex (Wilson, 1935a).

Included Species. — Simimys simplex (Wilson, 1935a) and
Simimys landeri Kelly, 1992.

Distrihution. — Known only from the late Uintan and Duchesnean
(late middle Eocene) of southern California.

Emended Diagnosis. — Mouse-sized basal dipodoid rodent.
Cheek teeth of generalized myodont construction. Zygomasseteric
structure hystricomorphous. Derived with respect to Pauromys in the
reduction of P4 to a single-rooted peg, the loss of p4, and the
narrowing of the ml trigonid. Similar in most respects to
Pappocricelodon, but with weaker or absent anteroconids and
anterocones on m 1 and M 1 and a separate neurovascular infraorbital
canal. Dentally similar to Elymys, but with stronger mesolophs and
mesolophids and a relatively more anteroposteriorly elongate ml.
Primitive with respect to Plesiosmintbus and later zapodids in having
ungrooved upper incisors, having only a moderately concave man-
dibular diastema, having the ridges that define the masseteric fossa
extending anteriorly only to below the ml talonid, and having the
posterior margins of the incisive foramina extending posteriorly only
to the level of the bony knob anterior to P4. Possible autapomorphy :
Strong taiob present on ventral zygoma anterior to P4 for origin of
masseter lateralis superficialis.

Discussion. — Simimys is among the most interesting of Califor-
nia Eocene rodents. Its first historical appearance in the region helps
characterize the beginning of the late Uintan (Walsh 1996), and it is
by far the most abundant micromammal at virtually all well-sampled
late Uintan sites. The genus was named by Wilson (1935a, b) on the
basis of several lower dentitions from the Sespe Formation inVentura
County. Wilson (1935a) originally recognized two species: S. sim-
plex (the type species, from the Duchesnean locality CIT 150), andS.
vetus (from the late Uintan locality CIT 207). Later, from the late
Uintan locality CIT 180, Wilson (1949) named a new species that he
questionably assigned to the genus, S?. murinus. On the basis of the
damaged holotype skull of this species (LACM [CIT] 3529), Wilson
( 1949: 19) stated "the area immediately in front of Ml is somewhat
damaged, but P4 appears to be absent. If it is present, the root must be
very slender indeed."

Lillegraven and Wilson (1975) described a large sample of iso-
lated teeth of S(w//?;v,sfrom the Santiago Formation of northwestern
San Diego County (UCMP Loc, V-72088) and compared it with the
original Sespe specimens described by Wilson. Tliey documented
extensive morphological variation in the V-72088 sample and ob-
served that it encompassed all of the feattires previously considered
diagnostic of Wilson's three named species. They therefore regarded
S. vetus and S. murinus as junior synonyms of S. simplex and
assigned all known southern California specimens of the genus to the
latter species. Finally, Kelly (1992) named the large species Simimys
landeri on the basis of isolated teeth from the later Duchesnean Simi
Valley Landfill local fauna. This species may also occur in the
questionably Duchesnean "Sweetwater" Formation of southwestern
San Diego County on the basis of specimens reported by Walsh
( 1991a) as "Myomorpha, unidentified genus and sp."

As will be discussed in more detail below, different interpreta-
tions of the relative significance of "dipodoid hystricomorphy" and

New Specimens of Metamnamys. Pauromys. and Simimys from the Uintan (middle Eocene) of San Diego County

13

Figure 8. Stereophotographs of SDSNH 54610, R mandible of S/mimvx sp. with il + ml-2: (A), occlusal view; (B), medial view; (C), lateral view. Scale
bar = 5 mm.

the presumed loss of P4 in Simimys have been piiniaiHIy responsible
for the controversy over whether the genus should be regarded as
inuroid, dipodoid, or neither. The specimens descinbed here shed
new hght on the possible affinities of this genus.

Simimys sp.
Figs. 8-9

Referred Specimens. Santiago Formation. Member C. — Jeff's
Discovery local fauna: SDSNH Loc. 3276: SDSNH 43640, maxilla
fragment with Ml. SDSNH Loc. 3564: SDSNH 47100, mandible
fragment with il + ml-2; 47101 and 52224, mandible fragments
with ml-2; 47468 and 47807, mandible fragments with ml; 54275
and 56695, maxilla fragments with P4; 47102, maxilla fragment
with P4-M1; 49848, maxilla fragment with Ml-2; 54116, maxilla
fragment with P4 alveolus + M 1 -2; 54930, maxilla fragment with P4

14

Stephen L. Walsh

alveolus + Ml -3. An additional 681 isolated teeth are known from
SDSNH Locs. 3276 and 3560-3564.

Rancho del Ore local fauna: 53 isolated teeth from SDSNH Locs.
3433, 3436, 3441, 3444, 3449, 3451, 3463, and 3466.

Stadium Conglonienitc. Upper Member. — Stonecrest local fauna:
47 isolated teeth from SDSNH Locs. 3530 and 3536.

Mission Valley Formalion.—VCMP Loc. V-6893/95866, man-
dible fragment with damaged m 1 -3. SDSNH 3870/54610, mandible
fragment with il + inl-2. SDSNH 4019/60065, mandible fragment
with il + ni2-3. In addition to the specimens studied by Lillegraven
and Wilson (1975), about 1597 isolated teeth are known from
SDSNH Locs. 3273, 3383, 3426, 3428, 3429, 3539, 3741, and 3822,
and hundreds of uncatalogued isolated teeth are known from SDSNH
Locs. 3870, 4019, and 4020.

Pomerado Coni^lomerate. Lower Member. — Eastview local
fauna: SDSNH Loc. 3755: SDSNH 56166, maxilla fragment with
alveolus for P4 + Ml-2. SDSNH 56167, maxilla fragment with
alveolus for P4 + Ml -3. An additional 55 isolated teeth are known
from Locs. 3493 and 3755.

Mandible and Lower Incisor. — See Lillegraven and Wilson
( 1975) for a detailed description of the molars of Simimys. Measure-
ments of cheek teeth in all available inandibular and maxillary
specimens of Simimrs from San Diego County are given in Table 6.
Mandibles of 5. simplex and S. veins were first described from the
Sespe Fomiation by Wilson (1935a); none of these specimens pre-
serves the angle or incisor. Several more complete mandibles are
now available from San Diego County (e.g., SDSNH 54610, Fig. 8).
The depth of these mandibles below ni 1 averages 3.35 mm (Table 1 ).
SDSNH 54610 and 60065 are complete enough to show the
sciurognathous condition of the angle. As noted by Wilson (1935a),
the ridges defining the masseteric fossa converge and tenninate
anteriorly below the ml talonid. As seen on SDSNH 47100 and
54610, the dorsal surface of the mandible anterior to ml is moder-
ately concave, distinctly more so than in Pauromys lillegraveni, but
not as much as in Metanoiamys agorus. A single mental foramen is
present on all detenninable specimens anterior and ventral to ml,
slightly higher than halfway up the lateral side of the mandible. As in
Pauromys, there is an anteroposteriorly elongated bulge on the
medial surface of the mandible below the diastema and ml. On
SDSNH 54610, the ascending ramus leaves the dorsal surface of the

mandible at the level of the m3 talonid, and, as seen in Fig. 8 A, there
is a small foramen immediately lateral to the posterior alveolus of
m3, as in Panromys lillegraveni and Plesiosminthus (Schaub 1930,
fig. 9; Wilson 1960:81).

The lower incisor is preserved in several mandibles. The ventral
face is distinctly convex, the lateral face is slightly convex, and the
medial face is very slightly convex, almost flat The enamel is
ungrooved, extending about one-third of the way onto the lateral face
and only a very short distance onto the medial face. A small narrow
pulp cavity is present. As noted, the lower incisor of Simimys sp. is
very similar in proportions and absolute size to that of Pauromys
lillegraveni (Table 1).

Maxillary and Upper Premolar. — The only specimens to show a
significant part of the ventral zygomatic root of the maxillary are
SDSNH 56166 and 47102 (Figs, 9A,B), which do not differ in
preserved morphology from LACM (CIT) 3529, the holotype skull
of Simimys mnrinns. The posterior edge of the ventral zygomatic
root lies lateral to P4, as in Pappocricetodon (Wang and Dawson
1994). The zygomatic plate is slightly more inclined than in Pappo-
cricetodon. As noted by Wilson (1949), there is a prominent knob for
the origin of the masseter lateralis superficialis immediately anterior
to P4. This structure is well shown in SDSNH 47102, 56166, and
56167. SDSNH 56166 also shows a distinct depression in the maxil-
lary immediately anteromedial to this knob, as also reported for
LACM (CIT) 3529 (Wilson 1949: 19). The extreme posterior margin
of the incisive foramen is preserved on SDSNH 56166 (Fig. 9A),
and, as in LACM (CIT) 3529 (WiLson 1949: 19), it ends posteriorly at
the level of the knob for the origin of the masseter lateralis
superficialis, rather than at the level of P4 as in Plesiosminthus and
other zapodids (Engesser 1979, Korth 1980).

P4 (or DP4) is preserved in place in SDSNH 47 102 (Fig. 9B). It is
single-rooted, with a tiny peglike crown that is slightly wider than long
(Table 6). There are no accessory cuspules or cingula. Given the tiny
size of P4 (or DP4) in Simimys (Table 6), it is probable that isolated
premolars would pass easily through a 30-mesh screen (0.6-mni
openings). This hypothesis is supported by the fact that no isolated
premolars were picked from the +30-mesh concentrates that resulted
from screen-washing about 22,000 kg of matrix from SDSNH Locs.
3273, 3383, and 3564. The only P4/DP4s that were recovered from
these sites are still embedded in small pieces of maxillary.

Table 6. Measurements of cheek teeth of Simimys sp. in maxillary and mandibular fragments.

P4

Ml

M2

M3

AP

W

AP

AW

PW

AP

AW

PW

AP

AW

SDSNH 47102

0.39

043

1.29

1.06

1.19

—

—

SDSNH 54275

0.39

0.44

—

—

—

—

—

—

—

—

SDSNH 56695

0.40

0.42

—

—

—

—

—

—

—

—

SDSNH 49848

—

—

1.20

1.01

1.08

1.19

1.07

1.03

—

—

SDSNH 541 16

—

—

1.30

1.09

1.18

1.31

1.21

1.13

—

—

SDSNH 54930

—

—

141

1.20

1.29

1.37

1.30

1.27

0.98

1.02

SDSNH 56166

—

—

1.40

1.12

1.21

1.39

1.26

1.23

—

—

SDSNH 56167

1.29

1.06
ml

1.13

1.31

1.21
ml

1.10

1.08

1.04
mi

AP

TRI

TAL

AP

TRI

TAL

AP

TRI

TAL

SDSNH 47100

1.26

0.71

0.97

1.32

1.09

1.08

SDSNH 47101

—

—

0.95

1.29

0.93

1.11

—

—

—

SDSNH 47468

1.27

0.69

0.94

—

—

—

—

—

—

SDSNH 47807

1.34

—

1.08

—

—

—

—

—

SDSNH 52224

1.28

0.68

0.91

—

—

—

—

—

SDSNH 54610

1.34

_

—

1.34

1.01

1.09

—

—

—

SDSNH 60065

—

-

—

1.39

0.98

1.05

1.26

1.01

0.91

New Specimens of Mdaiuyiuiiiyx, Puiiivinys. and Siiiiiiiiys troni the Uintan (middle Eocene) dI San Diego County

15

Figure 9. Stercophotographs in occlusal view of (A), SDSNH 56166, L maxillary fragment of Simimys sp. with Ml -2 and matrix-filled alveolus for P4.
Note strong knob anterior to P4 for origin of M. lateralis superficialis. Scale bar = 2 mm. (B), SDSNH 47102. R maxillary fragment of Si/n/mv5 sp. with P4-
Ml. Scale bar = I mm.

Frequency of Anterior Appression Facets on Ml. — No anterior
appression facets were noted on any of the 15 M is ofSimtmxs from the
Mission Valley Formation described by Lillegraven andWilson ( 1975).
However, 18 of 75 determinable Mis fronfsDSNH Loc. 3273 have
small anterior facets (e.g., SDSNH 33000 and 33226), as do 9 of 47
detenninable Mis from SDSNH Loc. 3383 (e.g., SDSNH 41561).
Such facets are less frequently observed on specimens from the Jeff's
Discovery local fauna, in which only 9 of 132 detenninable isolated
M Is have detectable facets. By analogy with Metanoiamys agoriis and
Sciuravus powayensis, it seems probable that most or all San Diego
County late Uintan populations of Simimys possessed P4 (or DP4) and
that the infrequency of anterior appression facets on M 1 is caused by
the death of some individuals prior to full tooth eruption, and/or slight
variations in the relative position of these teeth.

In their examination of at least 100 Mis of Simimys from the
latest Uintan and/or Duchesnean Camp San Onofre local fauna
(UCMP Loc, V-72088), Lillegraven and Wilson (1975) noted a
possible anterior wear facet on only one tooth. The apparent rarity of
Ml facets from V-72088 may suggest a decreasing frequency of
occurrence of P4/DP4 through time. Interestingly, M. R. Dawson
(pers. comm. ), in studying a large sample of isolated teeth of Simimys
from the Tapo Canyon and Brea Canyon local faunas of the Sespe
Formation, has not identified a single M 1 with an anterior appression
facet.

It is uncertain whether San Diego County late Uintan populations
of Simimys ( 1 ) possessed and then shed DP4 without replacement.

(2) retained DP4 into adulthood, or (3) replaced DP4 with a perma-
nent P4. The first hypothesis would partly explain the infrequency of
appression facets on Ml, since the absence of a P4 would allow a
facet to form only during the brief youth of the animal. This hypoth-
esis is unlikely, however, because several maxillae with moderately
to well-worn molars (e.g., SDSNH 47102, 541 16, 56167) possess
either a premolar or its intact alveolus, showing the presence of a
tooth in this locus in adulthood.

Implications for Species-Level Taxonomy. — The new specimens
of Simimys described here require a conceptual reevaluation of the
specific taxonomy of this genus. Potential alternatives are as follows;

1. If all three Sespe populations represented by the holotypes of
S. simplex, S. vetiis, and 5. muriniis had in fact retained P4 (or DP4),
then the San Diego County specimens would still be assignable to
the senior synonym, S. simplex.

2. If all three Sespe populations represented by the holotypes of
S. simplex, S. vetus, and S. miirinus had in fact lost P4/DP4, then the
San Diego County specimens could pertain to a new, more primitive
species characterized by the retention of P4/DP4.

3. If the Duchesnean S. simplex had in fact lost P4/DP4, but the
late Uintan populations represented by the holotypes of S. vetus and/
or 5. murinus retained P4/DP4, then one of the latter names could be
resurrected, and the San Diego County specimens could be assigned
to the appropriate Sespe species.

Evaluation of these possibilities must awah the collection of

16

StephLMi L. Walsh

EOMYINAE YODERIMYINAE

other Plesiosminthus, other

zapodids Schaubemys Nonomys muroids

5c, 7d
Pappo- ^^

icetodon ^^^6c, 10e

Elymys

DIPODOIDEA- 16b -\- »« MUROIDEA- 17b

UNNAMED NODE- 5b, lOd, lid, 15b, 19b

6b. 10c. 11c, 12b

MYODONTA- 7c, 8b, 9b

MYOMORPHA- lb, 2b, 3b, 4b

"SCIURAVIDAE"

Characters

1-19 all
have state "a"

1. M3 length: a. Greater than or equal to M1-2. b Less than or equal to M1-2.

2. nn3 length: a Greater than or equal to m1-2. b. Less than or equal to m1-2.

3 Incisor enamel: a. Pauciserial b Uniserial or trending that way. c. Specialized uniserial.

4. One or more foramina between m3 and ascending ramus: a Absent, b Present

5 Mandibular diastema: a Essentially hohzontal. b Moderately concave c. Deeply concave.

6. Masseteric fossa ends below: a m1-2 contact, b. ml talonid. c ml trigonid. d. p4-m1 contact, e. p4.

7 Zygomasseteric structure: a. Protrogomorphous b Sciuromorphous. c Hystricomorphous d. Myomorphous.

8. M1-2: a. Essentially square in occlusal outline, b. Slightly AP-elongated.

9. Lower molar mesolophids: a Weak b Variably elongated and progressively complex.

10. P4 : a 10-15% smaller than Ml b Subequal to or larger than M1 c About 50% smaller than Ml. d. 1-rooted peg. e.

11. p4: a. 10-15% smaller than ml. b. Subequal to or larger than ml c About 50% smaller than ml. d. Lost

12 P3: a. Present, b. Lost.

13 Cheek tooth pattern: a. Relatively simple, b Relatively complex.

14 Upper incisors: a Ungrooved. b. Grooved.

15. ml trigonid: a. Slightly narrower than talonid b Distinctly narrower than talonid.

16. Separate neurovascular intraorbital foramen: a Absent, b. Present.

17 Anterocone, anteroconid: a. Absent or weak b Moderately to well-developed.

18. Knob for origin of mis.: a. Absent or weak b Strong.

19. Posterior ends of Incisive foramina located; a. Far anterior to P4. b. Opposite knob for m.l.s. c. Opposite P4.

Lost.

Figure 10. Cladogram showing hypothesized relationships of various Paleogene myomorph rodents. Geomyoidea and Dipodidae omitted for simplicity.
See text for discussion.

more complete maxillae from the Sespe Foniiation. Until then, I
conservatively identify all late Uintan San Diego County specimens
of the genus as "Simimys sp.""

Discussion. — Wilson (1949) noted three characters in which
Simimys differed from dipodoids: (1) P4 apparently absent, (2)
distinct anterior head of masseter lateralis superficialis, and (3)
zygomatic plate slightly broadened and inclined. He also noted three
characters in which Simimys differed from muroids; { 1 ) zygomatic
plate essentially horizontal, (2) infraorbital canal not particularly
muroid, and (3) absence of a well-developed anterocone or
anteroconid. Wilson (1949) concluded that "Simimys can perhaps be

viewed as a more or less primitive survivor into the late Eocene of a
stalk which was ancestral to both cricetids and the Dipodoidea, but in
which enough progress had been made in skull structure and dental
formula so that it is a muroid rather than a dipodoid rodent."

Largely on basis of the presumed absence of P4, Lindsay ( 1968)
felt that Simimys should be regarded as an early cricetid and derived
this genus from Sciuravus via Metanoiamys. Wood (1974) criticized
Lindsay's (1968) phylogeny, pointing out that it implied evolution of
the zygoma from protrogomorphous to sciuromorphous to
hystricomorphous, all in the late [now regarded as middle] Eocene of
southern California. Lindsay (1977) acknowledged the difficulties

New Specimens of Metaiumimys, Pcmromys. and Sunimys trom the Uintan (middle Eocene) of San Diego County

17

inherent in liis previous phylogeny and presented t'tirther evidence to
support the assigntnent of Sinumys to tlie Cricelidae. Lillegraven and
Wilson (1975) felt that assigivnient of Simimys to the Zapodidae
(Dipodoidea) was preferable to assigntnent to the Cricetidae
(Muroidea), on the basis of a character apparently unique to the
Dipodoidea: presence of separate neurovascular and itifraorbital
canals (see Eniry 1981). Wood (1980) recognized a new inonotypic
family Siniiinyidae in order to fonnalize Wilson's ( 1 949) view on the
relationships of Simimys. However, Wood assigned the Siniiinyidae
to the Dipodoidea rather than to the '.'Muroidea, as proposed by
Wilson. Eiury (1981) clarified the infraorbital anatoniv of Siniimxs
as previously discussed by Lillegraven and Wilson ( 1975) and Lind-
say (1977) and assigned the genus to Muroidea, incertae seciis, in
part on the basis of the presutned loss of P4.

The new material described above clearly shows that at least one
late Uintan population of Simimys retained P4, and initil the Sespe
populations can be definitely shown to lack P4, this genus should not
be excluded as a potential ancestor of later dipodoid rodents. Whether
or not Simimys should itself be regarded as a dipodoid is still
debatable. Although it had a separate neurovascular foramen as in
later dipodoids, it is possible that this is a primitive character that
originated relatively early in myodont history, only to be lost by later
muroids (the presence of a separate neurovascular foramen in the
"dentally muroid" genus Nonomys may be consistent with this
hypothesis; see below). Simimys also has an accessory opening for
the mandibular canal between m3 and the ascending ramus, but the
presence of this character in Pauromys suggests it is primitive for the
Myodonta, and cannot be used to allocate ta.xa to either the
Dipodoidea or Muroidea (both the zapodid Plesiosmintlms and the
presumed early cricetid Puppocricefodon also have one or more
foramina in this area; see below). Unlike later zapodids, Simimys
retains the primitive condition of having ungrooved upper incisors.
In addition, although the posterior margins of the incisive foramina
of Simimys extend further posteriorly than those of sciuravids, they
do not reach the level of P4 as in Plesiosminthus and other zapodids.
A potential autapomorphy for Simimys is the strong knob anterior to
P4 for the origin of the masseter lateralis superficial is, which is
apparently not comparably developed in any other known Paleogene
myomorph or extant dipodoid.

DISCUSSION OF SELECTED
PALEOGENE MYOMORPH GENERA

A brief review of selected Paleogene myomorph rodents is
presented here to provide perspective on the material of
Meranoiumys. Pauromys. and Simimys discussed above, and to high-
light pertinent character distributions shown in Fig. 10.

Armmtomys. — Armintomys was described by Dawson et al.
(1990) on the basis of a partial skull from the eariiest Bridgerian (late
early Eocene) of Wyoming. This genus is hystricomorphous, retains
both upper premolars, has slightly anteroposteriorly elongate but
otherwise primitive sciuravid-like molars, and lacks a distinct knob
for the origin of the masseter lateralis supert'icialis. Dawson et al.
(1990) named the new inonotypic family Annintomyidae and ques-
tionably assigned it to the Dipodoidea. Wang and Dawson ( 1994:250)
later noted that Armintomys lacked the derived neurovascular canal
of Simimys and later Dipodoidea and suggested it might represent the
sister group of dipodoids + cricetids. Although the grooved upper
incisor of ArmiiUomys is a character shared with Plesiosmintlms and
extant zapodids, given the ungrooved incisors of the more dentally
derived myodonts Pauromys and Simimys, the condition in
Armintomys is perhaps best assumed to represent a precocious
convergence (Fig. 10). The presence in Armintomys of an incisor
microstructure transitional between pauciserial and uniserial
(Dawson et al. 1990) is a potential autapomorphy for either the

Myodonta or Myomorpha, and the latter interpretation is favored in
Fig. 10.

Elymys.—Elymys was named by Eniry and Korth (1989) from
the Bridgerian of Nevada on the basis of a maxilla fragment with P4-
M3 and several isolated teeth. Elymys is clearly more derived than
Pauromys and is similar taSimimys in the apparent absence of p4 and
the reduction of P4 to a single-rooted peg. Entry and Korth (1989)
noted the similarities in molar morphology among El\m\s,
Plesiosminthus, and Simimys and suggested the possibility of an
ancestral relationship between Elymys and Simimys. This hypothesis
represents a different scenario involving a North American origin of
Simimys, and the discovery of more complete material of Elymys is
awaited. Critical evidence in this regard will be the presence or
absence in Elymys of a separate neurovascular canal, a foramen
between m3 and the ascending ramus, a distinct knob for the origin
of the masseter lateralis supert'icialis. and the relative positions of the
masseteric fossa and incisive foramina.

As shown in Fig. 10, a general trend in myoinorphs is the
reduction of M3 relative to Ml -2. The type specimen of Elvmxs,
however, has an M3 that is approximately subequal to Ml -2 (Emry
and Korth 1989, fig. 5A-C, table 6), which seems anomalous for a
myomorph relatively derived dentally. Whether this condition repre-
sents individual variation, a retained primitive sciuravid character, or
a secondary reenlargement is unclear. If the latter is correct, it could
represent an autapomorphy for Elynivs.

Plesiosminthus and Schauhemys. — Plesiosminthus is a late
EoceneC?) to Miocene zapodid known primarily from Europe and
Asia, with one species (P. clivosus) currently recognized from North
America (Wilson 1960, Korth 1980, Green 1992, Korth 1994).
Schauhemys is an Oligocene-Miocene North American zapodid
(Wilson 1960, Korth 1980, Korth 1994). Strong similarities in dental
morphology between Plesiosminthus and Simimys have been noted
by several authors (e.g., Wilson 1949). Mandibles of Plesiosminthus
described by Schaub ( 1930, fig. 9) and Galbreath (1953, fig. 26) and
mandibles of Schauhemys described by Wilson (1960, fig. 131)
appear to be similar to those of Simimys in their diastemal concavity
(more pronounced in Plesiosminthusand Schauhemys), their posses-
sion of an anteroposteriorly elongated bulge on the medial side of the
mandible, in the departure of the ascending ramus from the alveolar
border at the level of m3, and in their possession of a foramen
between m3 and the ascending ramus (the latter condition is variable
in Schauhemys; see Wilson 1960:84). Known maxillae of
Plesiosminthus and Schauhemys are similar to those of Simimys sp.
in their possession of a single-rooted peglike P4 (although the crown
of this tooth seems to be somewhat more complex in the fonner two
genera) . Plesiosminthus and Schauhemys differ from Simimys in that
the masseteric fossa extends farther anteriad to below the m 1 trigo-
nid (Galbreath 1953:99; Korth 1980, fig. 2), the upper incisors are
grooved, there is apparently no distinct knob anterior to P4 for the
origin of the masseter lateralis superficialis (Wilson 1960, figs. 126,
128; Green 1977, fig. 31; Korth 1980, fig. 3A), and the posterior
margins of the incisive foramina extend back to the level of P4. Since
almost all of the characters in which Simimys differs from
Plesiosminthus and Schauhemys appear to be primitive, it seems
likely that Simimys was close to the line of direct ancestry of the
fonner genera. As noted, the very strong knob anterior to P4 may
represent an autapomorphy for Simimys that would exclude it from
being an actual ancestor of the younger taxa.

Pappocricetodon. — Pappocricetodon was named by Tong (1992)
to include two species from the late middle Eocene and late Eocene
of China. Wang and Dawson ( 1994) named a third .species from the
middle Eocene of China. Pappocricetodon antiquus, and regarded
the genus as the earliest known and most morphologically primitive
member of the Cricetidae. Pappocricetodon antiquus is similar to
Simimys sp. in general molar morphology, mandibular structure, and

18

Stephen L. Walsh

in the possession of a small alveolus for a single-rooted P4 (or DP4).
This species also has one or more small foramina between m3 and
the ascending ramus (M. R. Dawson, written comm., 1996).
Pappocricetodon antiquiis differs from Simimys sp. in its lack of a
distinct kiiob for the origin of the masseter lateralis superficialis, in
having stronger anterocones and anteroconids on upper and lower
first molars, and most importantly in its lack of a neurovascular
infraorbital canal separate from the masseteric infraorbital canal.

The existence of Pappocricetodon antiquiis in China could be
interpreted as evidence favoring the immigration of Simimys to
North America (e.g., Vianey-Liaud 1985), but the direction of this
putative immigration is unclear for two reasons. Although the stron-
ger anterocones and anteroconids in Pappocricetodon are probably
derived over the condition in Simimys (Wang and Dawson 1994), the
polarity of the other characters in which the two genera differ is
unclear. Second, the precise age of P. antiquiis relative to Simiitiys is
uncertain. Given the -"'Ar/^Ar date of 42.83 ± 0.24 Ma for Simimys-
bearing strata of the Mission Valley Formation (J. D. Obradovich;
Walsh et al. 1996), the first historical appearance of this genus in
southern California can probably be constrained to about 43-44 Ma.
This date may play a key role in detennining whether Simimys pre-
or postdates the middle Eocene records of Pappocricetodon antiquus
in China. Unfortunately, the locality oi P. antiquiis is a fissure-filling
and cannot yet be assigned a numerical age (M. R. Dawson, pers.
comm.).

Nonomys. — The peculiar rodent Nonomvs was named by Emry
and Dawson (1972; 1973) from Chadronian deposits of Wyoming
and referred to the Cricetidae (note that the Chadronian NALMA is
now regarded as late Eocene; e.g., Prothero and Swisher 1992).
Emry ( 198 1 ) described additional material of Nonomys and pointed
out that it had a neurovascular canal separate from the masseteric
infraorbital canal (as in dipodoids) but also had typical muroid
characters such as strong anterocones and anteroconids and lacking
P4. Emry (1981) emphasized the dental features of Nonomys and
placed it in the Muroidea, incertae sedis. In contrast, Wang and
Dawson (1994) gave more weight to the separate neurovascular
canal, assigned Nonomxs to the Dipodoidea, and regarded it as the
sister taxon of Simimys. Implicit in the latter decision is the conclu-
sion that the loss of P4 and the development of strong anterocones
and anteroconids must have been achieved independently in
Nonomxs and muroids. Given the available evidence, superfamilial
assignment of Nonomys depends entirely on one's philosophy of
character weighting, i.e., can one relatively coiTiplex zygomasseteric
structure outweigh two dental characters that might be more subject
to homoplasy? There is no "correct" answer to this methodological
question, and both alternatives to the phylogenetic position of
Nonomxs are shown in Fig. 10. Interestingly, R. J. Emry (pers.
comm., 1996) indicates that the two mandibles of Nonomys
simplicidens figured by Emry ( 198 1 , fig. 1 ) possess a small foramen
between m3 and the ascending ramus, as in Pauromys, Simimys,
Plesiosrtiinthus, and Pappocricetodon.

PROPOSED RELATIONSHIPS OF SELECTED
PALEOGENE MYOMORPHA

A generalized hypothesis of cladistic relationships for selected
Paleogene myomorphs is shown in Fig. 10, based upon the tradi-
tional view that myomorphs were derived from sciuravids [e.g.,
Jaeger ( 1988) and Wang and Dawson ( 1994), but see Vianey-Liaud
(1985) and Luckett and Hartenberger (1985) for different scenarios].
The character distributions were obtained from the literature and the
new material described above. Polarity deteniiinations and the re-
sulting topology represent my own preferred synthesis of the exist-
ing consensus — see especially Wang and Dawson (1994) and Korth
( 1994) — but no attempt was made to perform a numerical parsimony
analysis. Such a study will be more meaningful when better speci-

mens of Armintomys. Pauromys. Elymys, and Nonomxs are discov-
ered. Note that the Geomyoidea and Dipodidae are excluded from
Fig. 10 for simplicity. No attempt is made to place the presumed
geomorpli Gnphomxs Wilson, 1940b, or the presumed dipodoid
Simiacritomxs Kelly, 1992, on the cladogram, as these two genera
are known only from relatively fragmentary material.

Tlie decision as to which nodes certain taxon names should
occupy is somewhat arbitrary. Some might prefer to restrict the
Myodonta to the most recent common ancestor of the Dipodoidea
and Muroidea, and all of its descendants, in which case Armintomys
and Pauromxs would not be members of Myodonta. A similar
approach would exclude Simimys and Elymys from the Dipodoidea.
See de Queiroz and Gauthier (1990) for a discussion of the logic
behind such decisions. A detailed consideration of the semantics of
this problem is important, but beyond the scope of this paper.

Character states lb and 2b (M3/m3 length • Ml-2/ml-2 length)
together represent a trend in early inyomorphs that is clearly derived
over the condition in sciuravids and ischyromyids, although excep-
tions are seen in some taxa, and these characters are unknown in
Armintomys. Character state 3b (incisor enamel uniserial or trending
that way) is assumed to be an autapomorphy for the Myomorpha, |
based on Armintomys (Dawson et al. 1990). The presence of this
character should be corroborated in Pauromys and Simimys. Note
that Wahlert (1968), Wahlert and von Koenigwald ( 1985), and Emry
and Korth (1993) reported a derived type of uniserial enamel in
Adjidaumo, Paradjidaumo, Yoderimys, and various other eomyids
(character state 3c) . It will obviously be important to detemiine if
the same condition exists in the basal eomyid Metanoiamys.

Character state 4b (presence of one or more foramina between
m3 and the ascending ramus) may be an autapomorphy for the
Myomorpha, as it occurs in Metanoiamys, Pauromys, Simimys,
Plesiosminthus, and Pappocricetodon but is absent in Sciuraviis,
Pareumvs, and ischyromyids. These forainina are also present in at
least some extant zapodids (e.g., Zapus and Napaeozapiis) and some
(but not all) extant cricetids. The functional significance of these
foramina is unknown, and their phylogenetic validity requires fur-
ther study.

The various derived states of Character 5 (condition of the
mandibular diastema) appear to have arisen more than once from the
primitive horizontal condition. The earliest known geomorph
(Metanoiamys) already has a deeply concave diastema. In Pauromys
it is still virtually horizontal, becomes moderately concave in
Simimys, Pappocricetodon, and Nonomys, and independently be-
comes deeply concave in zapodids and later muroids.

Various states of Character 6 (anterior extent of the masseteric
fossa) show much homoplasy. In Metanoiamys, the masseteric fossa
already reaches as far anteriad as the p4-ml contact; it reaches to
below p4 in all known later eomyids. Anterior advancement of the
fossa progressed more slowly in myodonts, since it extends only to
below the ml talonid in Pauromys, Simimys, and Pappocricetodon.
Further advancement of the fossa to below the m 1 trigonid appar-
ently took place independently in Plesiosminthus and later zapodids,
and in Nonomys and later muroids. Character state 7b (sciuromor-
phy) is assumed to be an autapomorphy for the Geomorpha with
respect to all other myomorphs (e.g., Wahlert 1985), although the
zygomasseteric structure of Metanoiamys is unknown. Character
state 7c (hystricomorphy) is assumed to be an autapomorphy for the
Myodonta (Dawson et al. 1990), and is therefore predicted to be
present in Pauromys and Elymys.

The typical "myodont" molar morphology is represented largely
by the combination of character states 8b (anteroposteriorly elon-
gated upper molars) and 9b (progressively elongated mesolophids).
Characters 10 and 1 1 (relative size and occurrence of P4 and p4)
show two distinct trends. In most sciuravids, P4/p4 is slightly smaller
(10-15%) than Ml/ml. This primitive condition persists in Meta-
noiamys, but P4/p4 become subequal to or larger than M 1/m 1 in the

New Specimens of Metcinoiainys, Pauromys. and Siiiiiinys from the Uinlaii (midille Eoeeiie) of San Diego County

19

Yoderimyinae and later eomyids (Wood 1974, tigs. 29c, 35; Emry
and Korth 1993:1047). The opposite trend is .seen in niyodonts,
where P4/p4 are highly reduced in Pauromys, P4 is further reduced
to a one-rooted peg and p4 is lost in Simimys ami Elxmxs, and P4 is
finally lost in most inuroids. P3 is primitively retained in
Mclauouiinys luid Yoderiinys, independently lost in later eomyids
and Paiironiys (character 12).

Character state 16b (separate neurovascular infraorbital foramen
present) is assumed to be an autaponiorphy for the Dipodoidea (Einry
198 1 ), but, as noted above, its presence in the possible nnmikl Nonnmys
is problematical (Wang and Dawson 1994 vs. Einry 1981). Two
possible phylogenetie positions for Nonomys are therefore shown in
Fig. 10. In addition, more complete material of Pauromys must be
collected to demonstrate that a separate neurovascular foramen did not
originate even earlier in myodont evolution than shown here.

ACKNOWLEDGMENTS

Collection of the specimens discussed herein was made possible hy the
cooperation and financial support of Camiel Mountain Ranch, Gatlin Develop-
ment, McMillin Connnunitics, the California Department of Transportation,
and H. G. Fenton Materials Compajiy. R. A. Cenitti and R. L. Clark expertly
prepared most of the delicate mandibular and ina.xillary material. C. R Majors
liel[X'd to collect fijuro/Hv.s-bearing matri.x from SDSNH Loc. 3691. J. D.
Archibald provided access to screen-washing facilities and the use of a
measuring microscope. I thank J. Alexander (AMNH), M. R. Dawson (CM), S.
McLcod (LACM), R. J. Emry (USNM), J. H. Hutchison (UCMP), P. Robin.son
and R Murphey (CU), and A. H.Walton and M. C.Winans {Univer>,ity of Texas
at Austin) for the loan of specimens and casts. A. H. Walton also provided a
copy of her unpublished cladistic analysis of dental characters of various
sciuravid and niyomorph rodents. Finally, 1 thank M. R. Dawson. R. J. Emry,
and W. W. Korth for their careful reviews of the manuscript, and T A, Deniere
and P. Unitt for their editorial improvements. However, none of these individu-
als necessarily agree with any of my interpretations.

LITERATURE CITED

Black, C. C. 1965. Fossil mammals from Montana. Part 2. Rodents from
the early Oligocene Pipestone Springs local fauna. Annals of
Carnegie Museum 38: 1-48.

Chiment, J.J. 1977. A new genus of eomyid rodents from the later Eocene
(Uintan) of southern California. Master's thesis, San Diego State
University.

Chiment, J. J., and W.W. Korth. 1996. A new genus of eomyid rodent
(Mammalia) from the Eocene (Uintan-Duchesnean) of southern
California. Journal of Vertebrate Paleontology 16:1 16-124.

Dawson, M. R. 1961. The skull of Sciuravus niridus, a middle Eocene
rodent. Postilla 53.

Dawson, M. R. 1966. Additional late Eocene rodents (Mammalia) from
the Uinta Basin, Utah. Aimals of Carnegie Museum 38:97- 1 14.

Dawson, M. R. 1968. Middle Eocene rodents (Mammalia) from north-
eastem Utah. Annals of Carnegie Museum 39:327-370.

Dawson, M. R. 1974. Paleontology and geology of the Badwater Creek
area, central Wyoming. Part 8. The rodent Micropanimys (Mamma-
lia). Annals of Carnegie Museimi 45: 145-150.

Dawson, M. R. 1977. Late Eocene rodent radiations: Nortli America,
Europe, and Asia. Geobios, Memoire Special 1 : 195-209.

Daw.son, M. R., L. Krishtalka, and R. K. Stucky. 1990. Revision of the
Wind River faunas, early Eocene of central Wyoming. Part 9. The
oldest known hystricomorphous rodent (Mammalia: Rodentia). An-
nals of Carnegie Museum 59:135-147.

De Queiroz, K., and J. Gauthier 1990. Phylogeny as a central principle in
taxonomy: Phylogenetie definitions of taxon names. Systematic
Zoology 39: 307-322.

Emry, R.J. 1 98 1 . New material of the Oligocene muroid rcxlent Nonomys,
and its bearing on muroid origins. American Mu.seum Novitates 2712.

Emry, R. J., and M. R. Dawson. 1972. A unique cricetid (Rodentia,
Mammalia) from the early Oligocene of Natrona County, Wyoming.
American Museum Novitates 2508.

Emry, R. J., and M. R. Dawson. [91i. Nonomys. new name for the cricetid
(Rodentia, Mammalia) genus Nanomys Emry and Dawson. Journal

of Paleontology 47:1003.

Emry, R. J., and W. W. Korth. 1989. Rodents of the Bridgerian (Middle
Eocene) Elderberry Canyon local fauna of eastern Nevada.
Smithsonian Contributions to Paleobiology 67.

Emry, R. J., ;uid W. W. Korth. 1993. Evolution in Yoderimyinae
(Eomyidae; Rodentia), with [lew material from the White River
Formation (Chadronian) at Flagstaff Rim, Wyoming. Journal of
Paleontology 67: 1047- 1057.

Engesser, B. 1979. Relationships of some insectivores and rodents from
the Miocene of North America and Europe. Bulletin of the Carnegie
Museum of Natural History 14.

Evanoff, E., R Robinson. P C. Murphey, D. G. Kron, D. Engard, and R
Monaco. 1994. An early Uintan fauna from Bridger E. Journal of
Vertebrate Paleontology 14, supplement to no. 3:24A.

Flanagan, K. M. 1986. Early Eocene rodents from the San Jose Forma-
tion, San Juan Basin, New Mexico. Contributions to Geology, Uni-
vcrsiry of Wyoming, Special Paper 3:197-220.

Galbreath, E.C. 1953. A contribution to the Tertiary geology and paleon-
tology of northeastern Coloratlo. University of Kansas Paleontologi-
cal Contributions, Vertehrata, 4.

Gazin, C. L. 1976. Mammalian fainial zones of the Bridger iniddle
Eocene. Smithsonian Contributions to Paleobiology 26.

Golz, D. J., and J. A. Lillegraven. 1977. Summary of known occurrences
of terrestrial vertebrates from Eocene strata of southern California.
University of Wyoming Contributions to Geology 15:43-65.

Green, M. 1977. Neogene Zapodidae (Mammalia: Rodentia) from South
Dakota. Journal of Paleontology 5 1 :996- 1 0 1 5.

Green, M. 1992. Comments on North American fossil Zapodidae (Roden-
tia: Mammalia) with reference to Mcgasminlhus, Plesiosmintlius,
and Schaubemys. Occasional Papers of the Museum of Natural
History of the University of Kansas 148.

Jaeger, J.-J. 1988. Rodent phylogeny: New data and old problems. Pp.
177-199 in M. J. Benton (ed.). The Phylogeny and Cla.ssification of
the Tetrapods, Volume 2: Manniials. Systematics Association Spe-
cial Volume no. 35B, Clarendon Press, Oxford, England.

Kelly, T. S. 1990. Biostratigraphy of the Uintan and Duchesnean land
mammal assemblages from the middle member of the Sespe Fonna-
tion, Simi Valley, California. Los Angeles County Natual History
Museum Contributions in Science 419.

Kelly, T S. 1992. New Uintan and Duchesnean (middle and late Eocene)
rodents from the Sespe Fomiation, Simi Valley, California. Bulletin
of the Southern California Academy of Sciences 91:97-120.

Kelly, T. S., E. B. Lander, D. P Whistler, M. A. Roeder, and R. E.
Reynolds. 1991. Preliminary report on a paleontologic investigation
of the lower and middle members, Sespe Fomiation, Simi Valley
Landfill, Ventura County, California. PaleoBios 13.

Kelly, T S., and D. P Whistler. 1994. Additional Uintan and Duchesnean
(middle and late Eocene) mammals from the Sespe Fomiation, Simi
Valley, California. Los Angeles County Natural History Museum
Contributions in Science 439.

Kennedy, M. P., and G. W. Moore. 197 1 . Stratigraphic relations of Upper
Cretaceous and Eocene fomiations, San Diego coastal region, Cali-
fornia American A.ssociation of Petroleum Geologists Bulletin
55:709-722.

Korth, W. W. 1980. Cricetid and zapodid rodents from the Valentine
Fomiation of Knox County, Nebraska. Annals of Carnegie Museum
49:307-322.

Korth, W. W. 1984. Earliest Tertiary evolution and radiation of rodents
in North America. Bulletin of the Carnegie Museum of Natural
History 24.

Korth, W.W. 1994. The Tertiary Record of Rodents in North America.
Plenum, New York.

Krishtalka, L., R. M. West, C. C. Black, M. R. Dawson, J. J. Flynii, W. D.
Tumbull, R. K. Stucky, M. C. McKcnna, T M. Bown, D. J. Golz, and
J. A. Lillegraven. 1987. Eocene (Wa.s;itchian through Duchesnean)
biochronology of Nortli America. Pp. 77-117 ui M. O. Wtxidbume
(ed.) Cenozoic Mammals of North America: Getx;hronology and
Biostratigraphy. University of Califomia Press, Berkeley.

Lillegraven. J. A. 1977. Small rodents (Mammalia) from Eocene deposits
of San Diego County, Califomia. Bulletin of the American Museum
of Natural History 158:221-262,

Lillegraven, J. A., and R. W. Wilson. 1975. Analysis of Simimys simplex, an
lux-ene rodent ('.'Zapodidae). Journal of Paleontology 49:856-874.

Lindsay, E. H. 1968. Rodents from the Hartman Ranch local fauna.

20

Stephen L. Walsh

California. PaleoBios 6.

Lindsay, E. H. 1977. Simimys and origin of the Cricctidac (Rodentia:
Muroideal. Geobios 10:597-623.

Luckett, W. P., and J. L. Harlenberger. 1985. Evolutionary rclatioaships
among rodents: Comments and Conehisions. Pp. 685-71 1 in W. P.
Luckett and J. L. Hartenberger (eds.). Evolutionary Relationships
Among Rodents: A Multidisciplinary Analysis, NATO Advanced
Science Series, ser A, Life Sciences, Volimie 92. Plenum, New York.

Mason, M. A. 1988. Mammalian paleontology and stratigraphy of the
early to middle Tertiary Sespc and Titus Canyon formations, south-
ern California. Ph.D. dissertation. University of California, Berke-
ley.

Nelson, M. E. 1974. Middle Eocene rodents (Mammalia) from south-
western Wyoming. University of Wyoming Contributions to Geol-
ogy 13:1-10.

Prothero, D. R., and C. C. Swisher, 111. 1992. Magnetostratigraphy and
geochronology of the terrestrial Eocene-Oligoeene transition in
North America. Pp. 46-73 in D.R. Prolhero and W. A. Berggrcn
(eds.). Eocene-Oligocene Climatic and Biotic Evolution. Princeton
University Press, Princeton, N.J.

Sehaub, S. 1930. Fo.ssile Sicislinae. Eclogae Geologicae Helvetiae
23:616-637.

Sehaub, S. 1958. Simplicidentata (= Rodentia). Pp. 659-8l8()i J. Pivcteau
(ed,). Traite de Paleontologie, Volume 6. Ma.sson, Paris.

Storer, J. E. 1984. Mammals of the Swift Current Creek Local Fauna
(Eocene: Uintan, Saskatchewan). Saskatchewan Museum of Natural
History Contributions 7.

Storer, J. E. 1987. Dental evolution and radiation of Eocene and early
Oligocene Eoniyidae (Mammalia, Rodentia) of NorthAmerica, with
new material from the Diichesncan of Saskatchewan. Dakolerra
3:108-117.

Tong, Y.-s. 1992. Poppocricetodoit, a pre-Oligocene cricetid genus (Ro-
dentia) from central China. Vertebrata PalAsiatiea 30: 1- 16 |in Chi-
nese, English sunniiaryl-

Troxell, E. L. 1923. Pauromys perdUiis. a small rodent. American Journal
of Science, series 5, 5:155-156.

Vianey-Liaud, M. 1985. Po.ssible evolutionary relationships among
Eocene and lower Oligocene rixlenLs of Asia, Europe, and North
America. Pp. 277-309 ;/( W. R Luckett and J. L. Hartenberger (eds.).
Evolutionary Relationships Among Rodents: A Multidisciplinary
Analysis. NATO Advanced Science Series, ser A, Life Sciences,
Volume 92. Plenum, New York.

Wahlert, J, H. 1 968. Variability of rodent incisor enamel as viewed in thin
section, and the microstnicture of the enamel in fossil and recent
rodent groups. Breviora 309.

Wahlert, J. H. 1978. Cranial foramina and relationships of the Eomyoidea
(Rodentia, Geomorpha). Skull and upper teeth of Kivisiisimys.
American Museum Novitates 2645.

Wahlert, J. H. 1985. Skull morphology and relationships of geomyoid
rodents. American Museum Novitates 2812.

Wahlert, J. H., and W. von Koenigswald. 1985. Specialized enamel in
incisors of eoniyid rodents. American Museum Novitates 2832.

Walsh, S. L. 1987. Mammalian paleontology of the southern outcrops of
the Mission Valley Fomiation, San Diego County, California. Under-

graduate thesis. Department of Geology, San Diego State University.

Walsh, S. L. 1991a. Late Eocene mannnals from the Sweetwater Fomia-
tion, San Diego County. California. Pp. 149-159 in P. L. Abbott and
J. A. May (eds.). Eocene Geologic History San Diego Region.
Pacific Section, Society of Economic Paleontologists and Mineralo-
gists 68.

Walsh, S. L. 1991b. Eocene mammal faunas of San Diego County. Pp.
161-178 m P. L. Abbott and J. A. May (eds.). Eocene Geologic
History San Diego Region. Pacific Section, Society of Economic
Paleontologists and Mineralogists 68.

Walsh, S. L. 1 996. Middle Eocene mammal faunas of San Diego County,
California. Pp. 75- 1 19 in D. R. Prothero and R. J. Entry (eds.). The
Terrestrial Eocene-Olig(x;ene Transition in Nortli America. Cam-
bridge University Press, Cambridge, England.

Walsh, S. L., D. R. Prothero, and D. J. Luiidquist. 1996. Stratigraphy and
paleomagnetism of the middle Eocene Friars Formation and Poway
Group in .southwestern San Diego County, California. Pp. 120-154
111 D R. Prothero and R. J. Emry (eds. I. The Terrestrial Eocene-
Oligocene Transition in NorthAmerica. Cambridge University Press,
Cambridge, England.

Walton, A. H., 1993. Pauromys and other small Sciuravidae (Mammalia:
Rodentia) from the middle Eocene of Texas. Journal of Vertebrate
Paleontology 13:243-261.

Wang. B.-Y., and M. R. Dawson. 1994. A primitive cricetid (Mammalia:
Rodentia) from the Middle Eocene of Jiangsu Province, China.
Annals of Carnegie Museum 63:239-256.

Wilson, R. W., 1935a. Cricctine-like rodents from the Sespe Eocene of
California. National Academy of Sciences Proceedings 2 1 :26-32.

Wilson, R.W. 1935b. Simimys. a new name to replace fH/Hv.vo/M Wilson,
preoccupied — A correction. National Academy of Sciences Pro-
ceedings 2 1 : 1 79- 1 80.

Wilson, R, W, 1940a, Califoniian paramyid rodents. Carnegie Institution
of Washington Publication 514:59-83.

Wilson, R. W. 1940b. Two new Eocene rodents from California. Carnegie
Institution of Washington Publication 514:85-95.

Wilson, R. W. 1949. Additional Eocene rodent material from southern
California. Carnegie Institution of Washington Publication 584:1-
25.

Wilson, R. W. 1960. Early Miocene rodents and insectivores from north-
eastern Colorado. University of Kansas Paleontological Contribu-
tions, Vertebrata, 7.

Wood, A. E. 1937. The mammalian fauna of the Wliite River Oligocene.
Part 2. Rodentia. Transactions of the American Philosophical Soci-
ety 28:155-269.

Wood, A. E. 1 959. A new sciuravid rodent of the genus Pauromys from the
Eocene of Wyoming. American Museum Novitates 1978.

Wood, A. E. 1962. The early Tertiary rodents of the family Paramyidae.
American Philosophical Society Transactions (n.s.) 52, part 1 .

Wood, A. E. 1974. Early Tertiary vertebrate faunas Vieja Group Trans-
Pecos Texas: Rodentia, Texas Memorial Museum Bulletin 21,

Wood, A. E. 1980. Tlie Oligocene rodents of North America. American
Philosophical Society Transactions 70, part 5.

3 2044 093' 361 1

86