'J0ij.na.iiSNi_NviN0SHiiws^S3 1 a va a I i_li b rari es smithsonianjnstit

O “'..W ~ O 'Z ’ O

JBRARIES SMITHSONIAN'^^INSTITUTION NOIinilXSNI ’^NVINOSHIIWS ^S3 I

m — >^iixst^ rn

NOIiniliSNI NVINOSHIIWS S3 I B Vd 8 n~LI B RAR I ES^SMITHSONIAN~INSTIT

i ^ ^ i ^

</5 ^31 t/> ^ • C/> 07

o
2

2 vj-YDi^^ > 5 > XlOmstjX S

LIBRARIES^^SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SHillMs‘^S3 1

(/) X <0 “ . CO

r 5

'JOIXnilXSNl'^NVINOSHillAIS^SB I B VB 8 11 “"lI B RAR I ES^SMITHSOn'|AN'"INSTIT
r* , z r- 2 [I ^

03

'<> K

- %%A

m '' >5^' ^ \.jvAsj^ rn x^osh^ ^ m xicio

JBRARIES SMITHSONIAN~INSTITUTION^NOIi.nili$NI~NVINOSHimS S3 IBM
> ^ 2 » <o z w

> X4«s^ 5 5'

MOIJinillSNI NVIN0SHiIIMs‘^S3 I B VB 8 ll^LI B RAR I ES^SMITHSONIAN INSTITl

5 \ ^ = CO -*

CO

5^- H g

. O Xi^os»i^ «. o

.IB RAR I ES SM1THS0NIAN”’iNSTITUTI0N N0linilXSNl"'NVIN0SHilWS^S3 I B\
z r- 2 r- ,.. 2

NoiiniiiSNi NViNosHims‘”s3iavaan libraries'^smithsonian instiH

> ^ _

oiiniiisNi^NViNOSHxms^sa I dVd a n^Li B rar i es'^smithsonian institu

“ ifi :z w ^

o '' .W' 2 o 2 ^ o

Z ■ -J Z ^ 2J

IBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHillAIS S3iav

OlinilXSNirNVINOSHJ.lWs'^Sa 1 y va a n“LI BRAR I Es'^SMITHSONiAN“lNSTm

Z ^ w z g Z

< ^ I s

iW £

> S > X^Oiixst^ ^

I B RAR I ES^^SMITHSONIAN INSTITUTION^NOlinillSNI NVIN0SHill/VS^S3 I d V
c/> c/> c/)

OliniliSNrNVINOSHillAlS^Sa I d Vd a n ”^LI B R AR 1 ES^SMITHSONIANllNSTITl
r- , Z r- 2

___ m ^ m Xii^os

IBRARIES SM!THSONIAN"'lNSTITUTION‘^NOIiniliSNl”NVINOSHimS S3ldV
> ^ z t ^ ^ 5 c/) .

? ^ I ~

« (li. il) ^ i If

OliniIiSNl_NVmOSHilWS^S3 I dVd a n^LI B RAR I ES^SMITHSONIAN INSTITL

</) X </5

o •• w “ o z o

Z ’ «j 2! *j Z

IBRARIES SMITHSONIAN INSTITUTION NOUfliliSNI NVINOSHilWS SBIdV

z r* Z r- z

2 ,.s /CSS^^X 2 xs5s^ ™ P

~ xc;?^®*3^oy i-^ ^ h vrf _

NOlinillSNI NVINOSHJLIINS S3 I dVd a n~LI B RAR I ES^SMITHSONIAN^INSTITI

Annals of Carnegie Museum

Volume 41

^4 ft

Article 1

SCOLECODONTS FROM WELL CORES
OF THE MAQUOKETA SHALE, UPPER ORDOVICIAN,
ELLSWORTH COUNTY, KANSAS

E. R. Eller

Curator of Geology and Invertebrate Fossils
Carnegie Museum, Pittsburgh, Pennsylvania

Dr. William K. Pooser, Research Geologist of the Cities Service Oil
Company, has graciously presented to the Carnegie Museum a suite
of 18 scolecodonts for study. The specimens, from Ellsworth County,
Kansas, were taken from well cores between 3336 and 3413 feet below
the surface and are thought to be from the Maquoketa shale, late
Ordovician. Scolecodonts from Ordovician rocks in a number of areas
have been studied by various workers: Hinde (1879), Stauffer (1933),
Eller (1942, 1945, 1946), Kozlowski (1956), and Zofia Kielan-
Jaworowska (1961, 1962). No single species described in this paper
can be said to be closely related to any other described Ordovician
forms. All the specimens are very small, being less than a half-milli-
meter in size. If the surface outcrops, together with cores from a
number of wells of the Maquoketa formation were examined, it is
likely that scolecodonts would be found in relative abundance. De-
tailed studies at numerous localities and stratigraphic horizons could
be beneficial to biostratigraphers.

Issued Dec. 31, 1969

Genus Lumbriconereites Ehlers, 1868

Lumbriconereites inconditus, new species
Maxilla I, figures 1-3

Irregular lateral margins are distinctive characteristics of the nar-
row and elongate jaw. A series of twelve conical, sharp-pointed to
rounded, medium-sized, backward-directed denticles extends along

x5t;^Hso'/v7^x

f ia;I 3 2 1‘:70 )

2

Annals of Carnegie Museum

VOL. 41

the inner margin nearly to the posterior end. The first denticle, or
fang, is long, sharp-pointed, and somewhat hooked. Most of the den-
ticles are fairly uniform in size and all are oblique to the under side
of the jaw. Part of the posterior end of the jaw is missing but it prob-
ably tapered to a small rounded extremity. The margins of the jaw
are irregularly curved and form small flange-like projections. About one-
half of the upper surface is occupied by a deep fossa, which extends
from about the mid-area of the jaw to the posterior end. The margins
of the fossa are slightly thickened and rounded. Both the upper and
under sides of the jaw are irregularly convex and concave.

A number of species of Ltimhriconer cites are similar in a general
way to Lumbriconer cites inconditus. They differ mostly in the outline
of the margins and the size of the fossa.

Lumhriconereites dissimilaris, new species
Maxilla I, figures 4, 5

The jaw is elongate and narrow. On first examination the narrow-
ness of the specimen suggested that part of the jaw along the outer
margin is missing. Since the margin of the fossa seems to be complete
and regular, the impression is probably due to the irregular curving
of the margin at various places. Along the inner margin a series of
eighteen large, blunt or rounded denticles extends the full length of
the jaw. They are oblique to the surface of the jaw. The fang, or first
denticle, is large, narrow, and sharp-pointed, and is directed forward.
The second denticle is blunt and is also directed toward the anterior.
The remaining denticles are more or less perpendicular to the inner
margin and decrease in size uniformly to small ones at the posterior
end. A large rounded flange at about the middle of the jaw curves in
slightly to form a shallow bight. A small, shallow fossa occupies about
one-third of the flange and extends to the narrow, rounded, posterior
extremity. The margins of the fossa, especially the outer one, are
thickened and rounded. The surfaces of the jaw are convex except for
a narrow area on the under side of the jaw at the posterior half, which
is concave.

Except for the narrowness of the jaw at the anterior half, and the
small fossa, Lumbriconereites dissimilaris is typical of the genus and
can be compared to a number of forms in a general way. It is, how-
ever, dissimilar to any other species.

1969

SCOLECODONTS FROM UpPER OrDOVICIAN, KaNSAS

3

Lumbriconereites machinosus, new species
Maxilla I, figures 6, 7

When viewed from the under side the surface of the jaw is irreg-
ularly concave and convex. Along the inner margin a series of eleven
conical-to-rounded denticles extends for more than three quarters of
the length of the jaw. The first denticle or fang is large, long, and
points nearly in a forward direction. The second tooth is similar but
smaller and is separated from the first and third denticles by a space.
Beginning with the third, the denticles are large, wide, rounded,
short, and very close together. The teeth increase in size to about the
middle of the jaw where they then decrease gradually in size toward
the posterior. The denticles are slightly oblique to the under surface
of the jaw. From the fang the irregularly curved outer margin forms
a small shank and a bight and then recurves to a narrow, knob-like
posterior extremity. A large deep fossa occupies more than one-half of
the upper surface of the jaw. The margins of the fossa are well-
rounded and thickened. A narrow elongate flange is present on the
inner margin of the jaw. The surface of the jaw anterior to the fossa
is convex.

The right jaw of Lumbriconereites machinosus is similar to those in
a number of specimens of the genus. These jaws differ mostly in the
type, arrangement, and obliqueness of the denticles. There is a slight
resemblance between Lumbriconereites machinosus and Lumbricon-
ereites perpentatus Hinde (1822: fig. 68). Except for the denticles
and their curved arrangement and a wider flange, Lumbriconereites
proclivis Eller (1942) is like Lumbriconereites machinosus. Two
forms, Lumbriconereites tuberosus Eller (1945) and Lumbriconereites
jenningsi Eller (1945), are like Lumbriconereites machinosus in cer-
tain ways but differ in the type and arrangement of the denticles and
in other details of structure. Lumbriconereites celsus Eller ( 1961 ) is
similar, generally, to Lumbriconereites machinosus, especially in the
form and position of the first and second denticles. They are unlike in
the position of the shank and in other characteristics.

Lumbriconereites attactus, new species
Maxilla 1, figures 8, 9

The jaw is elongate and rounded at the anterior and posterior ends.
A series of 17 long, narrow, conical, closely arranged, sharp-pointed,

4

Annals of Carnegie Museum

VOL. 41

backward-directed denticles extends nearly the full length of the jaw.
The first denticle is larger than the remaining teeth, which decrease in
size only slightly to the posterior. A large, curved shank extends nearly
the full length of the jaw. The margin of the shank may be broken.

On the upper surface a large, shallow fossa occupies about three-
quarters of the jaw. The margins of the fossa are slightly thickened
and rounded. The upper surface of the jaw is convex and the under
surface is nearly flat.

This form does not compare very closely with other species of the
genus. The left jaw of Lumhriconereites hibhardi Eller (1940) re-
sembles Lumhriconereites attactus in a general way. The two species
differ mostly in the shape and arrangement of the denticles and the
shape of the shank. Hinde (1879) described a form as Eunicites din-
tonensis that is probably a side-view of a Liimbriconereites species.

It is similar to the side-view of Liimbriconereites attactus but differs
in the width of the posterior end and the length of the denticles.

Genus Paleoenonites Eller, 1942
Paleoenonites discrepans, new species
Maxilla II, figure 10

This form varies considerably from other species of the genus be-
cause of the position of the shank. It begins at the mid-area of the jaw
instead of in the anterior part. The jaw is elongate with an acute
posterior extremity. Along the inner margin a series of eight conical, j
sharp-pointed, backward-directed denticles extends for about two- '
thirds the length of the jaw. The first denticle or fang is large, pointed, i
and separated from the second tooth by a space. The remaining den-
ticles are nearly uniform in size and are slightly oblique to the under
surface of the jaw. A shallow-to-deep irregular fossa occupies more j
than half the upper surface of the jaw. The margins of the fossa are
thickened and rounded. The outer margin curves gently from the
anterior and then curves in to form a large, angular shank that curves
slightly to the posterior end. The upper surface of the jaw is irregu-
larly concave. The under surface is irregularly concave and convex. |

Paleoenonites discrepans has characteristics found in a number of
species of the genus but does not compare very closely with any of I
them in the position of. the shank and fossa and the absence of den-
ticles at the posterior end. The form might be grouped with Paleoen-

1969

SCOLECODONTS FROM UpPER OrDOVICIAN, KaNSAS

5

onites acutus Eller ( 1945 ) , Paleoenonites circumscriptiis Eller ( 1945 ) ,
Faleoenonites armigerus Eller ( 1955 ) , and Paleoenonites davisae Eller
(1945).

Paleoenonites infundibulus, new speeies
Maxilla II, figure 25

In outline, the jaw is subtriangular, wide anteriorly, and the lateral
margins taper to an acute posterior extremity. Along the inner margin
a series of fourteen sharp-pointed, conical, backward-directed den-
ticles extends the full length of the jaw. The first denticle, or fang, is
large, forward-directed, and only slightly hooked. From the fang the
denticles increase in size very slightly to the mid-area and then de-
crease in size gradually to the posterior end. The anterior margin is
fairly straight, but at the outer end curves abruptly with the curved
outer margin to form a small, round, shank-like projection. The sur-
face of the jaw is irregularly convex and concave. Since matrix covers
the upper side of the jaw the fossa cannot be seen and described.

A number of species of Paleoenonites resemble Paleoenonites in-
fundibulus. If the fossa of the specimen could be observed, this form
might prove to be the same as Paleoenonites acutus Eller (1945).
The denticles and the shank, or area at the anterior end, are slightly
different in the specimens of the two species. Paleoenonites accuratiis
Eller (1942) resembles Paleoenonites infundibulus except at the pos-
terior end. With the exception of the number of denticles and the con-
tour of the outer margin, Paleoenonites latissimus Eller ( 1942 ) is sim-
ilar to Paleoenonites infundibulus. Two species, Paleoenonites kopfi
Eller (1940) and Paleoenonites fornicatus Eller (1940), are com-
parable to Paleoenonites infundibulus. A form, Oenonites radula
Hinde (1882), from Gotland, is similar to Paleoenonites infudibidus,
especially in the shape of the fang and denticles and their arrange-
ment. Eller ( 1964 ) described a form, Paleoenonites lacinatus, that re-
sembles Paleoenonites infundibulus, but only in a general way.

Genus Ildraites Eller, 1936
Ildraites infractus, new species
Maxilla I, figures 11, 12

Part of the posterior end of this specimen is missing. The jaw is
elongate and subtriangular in shape. Along the nearly straight inner

6

Annals of Carnegie Museum

VOL. 41

margin a series of eleven conieal, blunt and sharp-pointed, baekward-
directed denticles extends the full length of the jaw. In the complete
specimen there might have been two or three more teeth in the series.
The first denticle is fairly heavy, pointed, slightly curved, and directed
forward. The second denticle is blunt and thickened. All the teeth are
slightly oblique to the under-surface of the jaw. The outer margin
curves in from the rounded anterior end and the fang to form a large,
triangular-shaped shank with a fairly acute point. A rather shallow
bight is formed by the shank and the nearly straight outer margin of
the posterior area of the jaw. On the upper side of the jaw a large,
fairly deep fossa occupies about one-half of the surface. The margins
of the fossa seem to be slightly thickened and rounded. The concavity
and convexity of the under-surface of the jaw are barely perceptible.
The upper side is definitely convex.

Ildraites infractiis cannot be compared very closely with most forms
described under this genus. The narrowness of the jaw, the curve of
the outer margin, the forward-directed fang, and the shape and size of
the second denticle make the form unlike other species. Stauffer ( 1933)
described Eiinicites granus, which, except for the two small denticles
following the fang, appears to be similar to Ildraites infractiis. Ildraites
howenensis Eller ( 1941 ) closely resembles Ildraites infractus. They dif-
fer in the position of the first denticle and the size and shape of the
second tooth. Loranger (1963) figured Ildraites howenensis from Al-
berta. It may be compared to Ildraites infractus. A form, Ildraites emi-
nulus Eller (1963), resembles Ildraites infractus in a general way. The
differences usually occur in the anterior part. The fang and the arrange-
ment of the anterior denticles are dissimilar.

Ildraites pooseri, new species
Maxilla I, figures 13, 14

This form has a very small fang, or first denticle, and there is not
much space between it and the second denticle. The jaw is wide an-
teriorly but decreases abruptly at about the midpoint to form a narrow
posterior area and lateral margins that are nearly parallel. On the
gently curved inner margin a series of eleven, rather large, blunt,
conical, backward-directed denticles extends nearly to the truncate
posterior extremity. The fang is narrow, well hooked, and slightly
oblique to the under surface of the jaw. The first four denticles follow-

1969

SCOLECODONTS FROM UpPER OrDOVICIAN, KaNSAS

7

ing the fang are not as blunt as the remaining ones, and not as oblique
to the surface of the jaw. From the first denticle the outer margin
curves gently and then curves in slightly to form a heavy truncate
shank. A wide, shallow bight emphasizes the shortness of the shank.
The fossa is narrow and irregular in shape and extends from the an-
terior end of the shank to the posterior extremity of the jaw. In the
area of the shank the fossa is wide, then narrows abruptly for most
of the length of the jaw and again becomes wide at the posterior. The
margins of the fossa are thickened and rounded. The upper surface
is irregularly convex. The under surface is mostly concave except for
narrow ridges extending lengthwise to the jaw.

Stauffer (1939) described a form, Arahellites dauphinensis, that,
with the exception of the shank, is slightly similar to Ildraites pooseri
in general shape. Ildraites howelli Eller (1941; Sylvester, 1959) re-
sembles Ildraites pooseri in a number of characteristics but differs in
the shape of the shank and fossa. Ildraites pooseri is like Ildraites
bowenensis Eller (1941; Loranger, 1963) in a number of ways. The
two differ in the shape of the fossa and in the space between the first
and second denticles. There is a slight resemblance between Ildraites
eminulus Eller (1963) and Ildraites pooseri.

Ildraites cyclus, new species
Maxilla I, figure 22

The jaw is elongate and narrow, especially in the posterior area.
Along the nearly straight inner margin a series of fourteen rounded,
closely arranged denticles extends for about three-fourths of the
length of the jaw. The first denticle, or fang, is large and is followed
by three small teeth of uniform size. The fifth denticle is larger, and
the teeth increase in size to about the mid-area, where they begin to
decrease uniformly in size to the posterior. The denticles are slightly
oblique to the under side of the jaw. From the fang the outer margin
is rounded and then incurved to form a large, semi-rounded shank.
The posterior margin of the shank is nearly straight and forms an
angular bight with the straight outer margin. A deep fossa occupies
the area of the shank and the narrow posterior end. The margins of
the fossa are slightly thickened and rounded. The under surface of
the jaw is mostly convex except at the anterior end.

A form, Ildraites exquisites Eller (1942), is similar to Ildraites
cyclus except for the size and shape of the shank and the type and

8

Annals of Carnegie Museum

VOL. 41

Explanation of Plates

Figures magnified about 50 times. Numbers in parentheses are Carnegie Museum

catalogue designations of the respective type specimens.

Figs. 1, 2, 3. Lumhriconereites inconditus, new species, Maxilla I (30299). Sub-
surface level, 3336 ft.

Figs. 4, 5, Lumbriconereites dissimilaris, new species. Maxilla I (30300). Sub-
surface level, 3410 ft.

Figs. 6, 7, Lumhriconereites maehinosus, new species. Maxilla I (30301): Sub-
surface level, 3340 ft.

Figs. 8, 9. Lumbriconereites attactus, new species. Maxilla I (30302). Subsurface
level, 3414 ft.

Fig. 10. Paleoenonites discrepans, new species. Maxilla II (30303). Subsurface
level, 3414 ft.

Figs. 11, 12. Ildraites infractus, new species. Maxilla I (30305). Subsurface
level, 3410 ft.

Figs. 13, 14. Ildraites pooseri, new species. Maxilla I (30306). Subsurface level,
3414 ft.

1969

SCOLECODONTS FROM UpPER OrDOVICIAN, KaNSAS

9

10

Annals of Carnegie Museum

VOL. 41

Figs. 15, 16. Leodicites interstr ictus, new species, Maxilla II (30309). Subsurface
level, 3414 ft.

Fig. 17. Staurocephalites apicatus, new species. Maxilla II (30310). Subsurface
level, 3412 ft.

Figs. 18, 19. Nereidavus novenus, new species. Maxilla I (30313). Subsurface
level, 3425 ft.

Figs. 20, 21. Staurocephalites ellstuorthensis, new species. Maxilla II (30311).
Subsurface level, 3412 ft.

Fig. 22. lldraites cijclus, new species. Maxilla I (30307). Subsurface level, 3410

ft.

Figs. 23, 24. lldraites limatus, new species. Maxilla I (30308). Subsurface level
3412 ft.

Fig. 25. Paleoenonites infundibulus, new species, Maxilla II (30304). Subsurface
level, 3340 ft.

Fig. 26. Staurocephalites invisitatus, new species, Maxilla II (30312). Subsurface
level, 3340 ft.

Figs. 27, 28. Marphysaites multiformis, new species, Carrier? (30314). Sub-
surface level, 3340 ft.

Fig. 29. Marphysaites claviger, new species. Carrier (30315). Subsurface level,
3410 ft.

Fig. 30. Siluropelta crassecaulis, new species. Mandible (30316). Subsurface
level, 3412 ft.

1969

SCOLECODONTS FROM UpPER OrDOVICIAN, KaNSAS

27

12

Annals of Carnegie Museum

VOL. 41

arrangement of the denticles, lldraites fritzae Eller (1942) has the
narrow, elongate, posterior end, like lldraites cyclus. The shape of the
bight and the type and arrangement of the teeth are different, how-
ever.

lldraites limatus, new species
Maxilla I, figures 23, 24

In shape the jaw is elongate and subtriangular. Along the slightly
curved inner margin a series of nine, long, conical, sharp-pointed,
backward-directed denticles extends the length of the jaw. The pos-
terior end of the jaw is missing and it is possible that there may have
been two or more additional denticles in the series. The fang, or first
denticle, is large, well hooked, and slightly oblique to the under, sur-
face of the jaw. When viewed from the under side the impression is
that there is not as much space between the first and second denticles
as there actually is. The second denticle is small and is followed by
teeth that increase in size to about the middle of the jaw and then
decrease slightly in size to the posterior. The outer margin curves
gently from the fang and then curves in slightly to form a medium-
sized, fairly acute shank. A shallow, crescent-shaped bight on the outer
margin forms the curved posterior margin. A rather small fossa occu-
pies the posterior part of the jaw. The margins of the fossa are slightly
thickened and rounded. The under surface of the jaw is irregularly
convex and concave, and the upper side is mostly convex.

A number of forms of the genus lldraites may be compared with
lldraites limatus in certain details, but none of the forms resembles it
closely. Lumhriconereites expansus Stauffer (1939) and Eunicites
grandis Stauffer (1939) have similar characteristics. They differ mostly
in the type and arrangement of the denticles, lldraites bowenensis
Eller (1941; Loranger, 1963), resembles lldraites limatus closely ex-
cept for the width at the anterior end and the size of the fossa. Except
for the denticles and the shape of the anterior area, lldraites howelli
Eller (1941); Sylvester, 1959), is similar to lldraites limatus in a gen-
eral way. There is a slight resemblance between lldraites patulus
Eller (1942) and lldraites limatus. The narrowness of the anterior
area and the type of denticles of lldraites rarus Eller (1945) and
lldraites limatus are comparable, lldraites appressus Eller (1955) is
like lldraites limatus except for the number and arrangement of the

1969

SCOLECODONTS FROM UpPER OrDOVICIAN, KaNSAS

13

denticles, lldraites eminulus Eller (1963) and Ildraites limatus are
fairly similar except for the space between the first and second den-
ticles and the width of the anterior and posterior ends. Ildraites tin-
exspectatus Eller (1964) has characteristics akin to Ildraites limatus.
They differ mostly in the size of the fossa.

Genus Leodicites Eller, 1940
Leodicites interstrictus, new species
Maxilla II, figures 15, 16

A small part of the shank, unfortunately, is missing from this form.
The jaw is elongate and subtriangular in outline. Along the nearly
straight inner margin a series of eleven blunt, conical denticles extends
nearly to the narrow, rounded, posterior extremity. The denticles are
uniform in shape and decrease gradually in size to the posterior. All
the teeth are directed backwards and appear to overlap each other.
The anterior margin is broadly rounded from the first denticle and
then incurved slightly to form a shank that is probably nearly straight.
A deep crescent-shaped bight on the outer margin emphasizes the
length of the shank. A large fossa is present on the upper side of the
jaw. The margins of the fossa are slightly thickened and rounded. In
the center of the fossa a convex ridge extends from the posterior to
about the center of the jaw. This convexity is reflected by a concave
area on the under side of the jaw.

Although Leodicites interstrictus is typical of the genus it is not
very similar to other forms. There is a slight resemblance to Leodicites
variedentatus Eller (1940), Leodicites reimanni Eller (1941), and
Leodicites altilis Eller (1955). The form differs mostly in details of
its general shape and in the type and arrangement of the denticles.

Genus Staurocephalites Hinde, 1879
Staurocephalites apicatus, new species
Maxilla II, figure 17

The jaw is narrow and elongate with a pointed anterior end and a
rounded posterior extremity. A nearly straight inner margin bears a
series of eight large, conical sharp-pointed, backward-directed den-
ticles, which extends to the posterior. The first denticle is slightly
larger than the remaining teeth, which decrease in size uniformly. A

14

Annals of Carnegie Museum

VOL. 41

large fossa extends from the first denticle to the posterior area. The
inner margin of the fossa is thickened; the outer margin is thin. The
under surface of the jaw is irregularly concave and convex. A narrow
ridge extends the full length of the jaw on the under side.

Staurocephalites clentatus Stauffer (1933), Staurocephalites hep-
penstallae Eller (1945), and Staurocephalites mccallae Eller (1945)
resemble Staurocephalites apicatus in a general way. They differ
mostly in the number of denticles and the shape of the anterior end.

Staurocephalites ellsworthensis, new species
Maxilla II, figures 20, 21

In outline the jaw is elongate and in shape semiovate, with the
broader end at the posterior. On the slightly curved inner margin a
series of ten large, conical, sharp-pointed, backward-directed denticles
extends nearly the full length of the jaw. The first denticle is slightly
larger than the remaining teeth, which decrease in size uniformly to
the posterior. A narrow fossa extends from opposite the first denticle
nearly to the posterior end of the jaw. The inner, or upper, margin
of the fossa is thickened and rounded; the outer margin is thin. The
anterior end of the jaw is missing but it probably terminated in an
acute ending. The posterior end of the jaw is broad and truncate. The
under side of the jaw is irregularly convex and concave. The upper
surface is mostly convex.

There is a general resemblance of Staurocephalites dentatus ( Stauf-
fer (1933), Staurocephalites mccallae Eller (1945), and Staurocepha-
lites heppenstallae EUer (1945) to Staurocephalites ellsworthensis.
They differ mostly in the anterior area and in the number and type
of denticles.

Staurocephalites invisitatus, new species
Maxilla II or III, figure 26

There is a section missing from the posterior end of the jaw, but it
is probably very small since the last denticle is minute in size. The
jaw is small and subtriangular in shape. Along the inner margin a
series of seven, very large to very small, conical, sharp-pointed den-
ticles extends the full length of the jaw. The first denticle points
slightly forward and the remaining teeth are nearly perpendicular to

1969

SCOLECODONTS FROAf UpPER OrDOVICIAN, KaNSAS

15

the margin. A narrow, deep fossa occupies the full length of the jaw.
At the anterior end of the fossa are two small spurs. The margins of
the fossa are slightly thickened and rounded. The upper surface of
the jaw is mostly concave and the under surface is convex.

Staurocephalites invisitatus is unlike most species of the genus.
Except for the length and size of the denticles, Staurocephalites kayi
Eller (1945) slightly resembles Staurocephalites invisitatus.

Genus Nereidavus Grinnell, 1877
Nereidavus novenus, new species
Maxilla 1, figures 18, 19

When viewed from different sides, the outline of the jaw varies in
shape. It is generally elongate and narrow. On the inner margin a
series of nine conical denticles extends for nearly three-quarters of
the length of the jaw. The fang is long, narrow, and is directed for-
ward. It is followed by a smaller, similar tooth that is in close contact
with the fang. The next four denticles are fairly uniform in size and
point anteriorly. The remaining three teeth are larger and directed
backward. A deep, oval fossa occupies about two-fifths of the posterior
area of the jaw. The margins of the fossa are thickened and rounded.
The upper side of the jaw is convex and the lower side is irregularly
concave and convex.

Usually the fossa and the posterior end of the right jaw of
Nereidavus species are irregular in shape and contour. Flanges are
usually present. This form lacks the usual central ridge in the fossa
which is reflected on the under side by a depressed area. The form
and arrangement of the denticles are similar in a number of Nereidavus
species. The right jaw of Nereidavus invisibilis Eller (1940) is similar
to that of Nereidavus novenus in shape but differs in the denticles
and in the complexity of the fossa. Nereidavus hamulus Eller (1941)
has an uncomplicated fossa like that of Nereidavus novenus. All four
species are unlike in the shape of the posterior end, and in number
and arrangement of the denticles.

Genus Marphysaites Eller, 1945
PMarphysaites multiformis, new species
Carrier? figures 27, 28

It is not certain whether this specimen is a carrier or actually part
of a jaw apparatus. Although there is a general resemblance of the

16

Annals of Carnegie Museum

VOL. 41

form to carriers of modern and fossil species no illustration has been
found in whieh the carriers were structurally as complex as this one.
Unlike this specimen, the illustrations show a nearly straight inner
margin in artieulation with the margin of the other carrier. It is
difficult to know which is the upper or under surface of the specimen.
In shape the form is elongate, with irregularly curved margins. A por-
tion of the posterior end is probably missing. The anterior end is
truncate and slightly ineurved, appearing to provide a natural place
for articulation with a maxilla. If figure 28 is the upper side, the lateral
rounded margins are bent inward and form a tube-like area. The pur-
pose of this hollow may be to accommodate a muscle or a part of the
maxilla. On the opposite side is a deep fossa-like area with a eurved
lip-like anterior margin. The surfaee of the speeimen is irregularly
convex and concave with round and elongate depressions and ridges.

Marphysaites claviger, new species
Carrier, figure 29

In outline the carrier is elongate and narrow with parallel margins.
The anterior end is truncate with a slightly incurved margin. The
posterior end may be broken. A ridge, whieh is forked near the
anterior, extends the full length of the specimen to the posterior where
it curves to the left. Adjacent to the ridge on the left side is a long
depressed area. At the anterior end of the form are three depressions.

Marphysaites claviger is unlike other forms of the genus. It differs
mostly in the sculpture of the surface and in the shape of the anterior
end. Eller (1945) described Marphysaites aptus as a mandible based
on the resemblance to species of the modern genus Marphysa. In
1964, however, Eller described two species, Marphysaites junctus and
Marphysaites gomphoides, as carriers since the anterior margin of these
species appeared to be the probable articulating surface with the
posterior end of a maxilla I. The carrier Marphysaites claviger resem-
bles some of the carriers of the modern genus Leodice Savigny.

Genus Siluropelta Eisenack, 1939
Siluropelta crassecaulis, new species
Mandible, figure 30

In shape the mandible is wide with an irregular ( probably broken )
outer and posterior margin. The inner margin is slightly incurved
instead of being nearly straight as in most species. In the articulation

1969

SCOLECODONTS FROM UPPER OrDOVICIAN, KaNSAS

17

there is probably an overlapping with an adjacent mandible. The
anterior margin curves in slightly and at the outer margin forms a
long, thick, rounded shaft or spine that terminates in a knob-like
ending. A small projection is present at the inner margin of the
anterior end. The surface of the mandible is mostly concave with
some ridges and raised areas.

Siluropelta cmssecaulis does not resemble other species of the genus

very closely.

References Cited

Eller, E. R.

1940. New Silurian scolecodonts from the Albion beds of the Niagara Gorge,
New York. Ann. Carnegie Mus., 28(2): 9-46, pis. i-vii,

1941. Scolecodonts from the Windom, middle Devonian of western New York.
Ann. Carnegie Mus., 28(16): 323-340, pis. xxxvii, xxxviii.

1942. Scolecodonts from the Erindale, upper Ordovician, at Streetsville, Ontario.
Ann. Carnegie Mus., 29(11): 241-270, pis. i-iv.

1945. Scolecodonts from the Trenton Series (Ordovician) of Ontario, Quebec,
and New York. Ann. Carnegie Mus., 30(12): 119-212, pis. i-vii.

1946. New scolecodonts from the Kagawong of Manitoulin Island, Ontario.
Proc. Pennsylvania Acad. Sci., XX: 71-75, pl.l, figs. 1-13.

1955. Additional Scolecodonts from the Potter Farm formation of the Devonian
of Michigan. Ann. Carnegie Mus., 33(21): 347-386, pis. 22-27.

1961. Scolecodonts from well samples of the Dundee, Devonian of Michigan.
Ann. Carnegie Mus., 36(4): 29-48, pis. 1, 2.

1963. Scolecodonts from the Dundee, Devonian of Michigan. Ann. Carnegie
Mus., 36(15): 173-180, pi. 1.

1964. Scolecodonts of the Delaware limestone, Devonian of Ohio and Ontario.
Ann. Carnegie Mus., 36(21): 229-275, pis. 1-5.

Hinde, a. J.

1879. On annelid jaws from the Cambro-Silurian, and Devonian Formations in
Canada and from the Lower Carboniferous in Scotland. Quart. Jour. Geol.
Soc. London 35: 370-383, pis. xviii-xx.

1882. On annelid jaws from the Silurian Strata of the Isle of Gotland. Bihang
till kongl. Svenska Velenskaps-Akademiens Handlengar, 7:1-28, pis. i-iii.
Kielan-Jaworowska, Zofia

1961. On two Ordovician Polychaete jaw apparatus. Acta Palaeont. Polonica,

6: 237-259, pis. i-vii.

1962. New Ordovician genera of Polychaete jaw apparatuses. Acta Palaeont.
Polonica, 6: 291-332, pis. 1-xiii.

Kozlowski, Roman

1956. Sur quelques appareils masticateurs des Annelides Polychetes Ordovician.
Acta Palaeont. Polonica, 1: 165-210.

Loranger, D. M.

1963. Devonian microfauna from northeastern Alberta, Canada. Thesis, PhD.
degree. University of London, published by the author, Calgary, Canada,
pp. 1-13.

Stauffer, C. R.

1933. Middle Ordovician Polychaeta from Minnesota. Bull. Geol. Soc. America,
44: 1173-1218. pis. 59-61.

1939. Middle Devonian Polychaeta from the Lake Erie district. Jour. Paleont.,
13: 500-511, pis. 57, 58.

Sylvester, R. K.

1959. Scolecodonts from central Missouri. Jour. Paleont. 33: 33-49, pis. 5, 6,

Back issues of many Annals of Carnegie Museum articles
are available, and a few early complete volumes and parts
are listed at half price. Orders and inquiries should be
addressed to; Publications Secretary, Carnegie Museum,
4400 Forbes Avenue, Pittsburgh, Pa. 15213.

Article 2

Annals of Carnegie Museum

Volume 41

REPORTS ON THE MARGARET M. CARY-CARNEGIE MUSEUM
EXPEDITION TO BAJA GALIEORNIA, MEXICO, 1961

5. Two new subspecies of Hesperiidae (Lepidoptera)
from the Cape region, Baja California Sur, Mexico

Lee D. Miller^

The Allyn Foundation, Sarasota, Florida,
and Research Associate, Section of Insects and Spiders,
Carnegie Museum, Pittsburgh, Pennsylvania

and

C. Don MacNeill

Division of Natural Sciences, Oakland (California) Museum

This is the -fifth of a series of papers based on the Margaret M. Cary -Carnegie
Museum Expedition to Baja California, Mexico, 1961, to appear in the Annals of
Carnegie Museum. For an account of the itinerary and description of localities,
see the first paper in this series [Richard M. Fox, Ann. Carnegie Mus., 36 (16):
181-192]. Except as noted, all specimens including type series are in the collection
of Carnegie Museum.

In his review of the Hesperiidae of Baja California one of us
(MacNeill, 1962) recognized the existence of endemic subspecies of
three Cape region hesperiids Cogia hippalus (Edwards), Polites
sahuleti (Boisduval), and Paratrytone melane (Edwards) — but did
not name them because there was not enough material. As a result of
the collections made by the Cary-Camegie Expedition, good series
are now at hand to describe subspecies of Cogia hippalus and Polites
sabuleti. The montane Paratrytone melane population was not taken.

1 Sarasota Bank Building, Sarasota, Fla. 33578.
Issued Dec. 31, 1969

20

Annals of Carnegie Museum

VOL. 41

Cogia hippalus peninsularis, new subspecies
Figures 1 (male holotype), 2 (female paratype),

7 (male genitalia)

male: Head, thorax, and abdomen brown above, paler below. Palpi dark brown
above and pale gray below. Antennae brown above, shaft yellow below, ringed
with brown; club completely encircled with brown basad of nudum. Legs scaled
with light brown. Upper surface of wings much like C. h. hippalus (Edwards),
but the spots on the forewings are larger, particularly those in spaces Ms-Cui
and Cui-Cu2, which often are as broad as they are long. On the under surface
of the hindwings the dark bands are very broad and often connected in space
Cui-Cu2. This connection is never present in specimens of the nominate sub-
species. There is no whitish bloom marginad of the discal band on the hindwings
beneath in C. h. peninsularis. This band is characteristic of C. h. hippalus. The
fringes of the forewings are grayish-brown with dark-brown checkering at the
ends of the veins, paler gray near the anal angle. The fringes of the hindwings
are whitish, completely interrupted by dark-brown checkering. The male genitalia
are indistinguishable from those of C. h. hippalus. Length of the forewing of the
male holotype is 19.0 mm.; the forewings of the eighteen male paratypes range
from 19.0 to 23.0 mm., averaging 20.2 mm.

female: Generally differs from C. h. hippalus in the same particulars as does
the male. The forewing cell spots of one specimen are divided like Arizona speci-
mens of the nominate subspecies. In all other specimens of C. h. peninsularis
these spots are conjoined. The dark bands of the hindwings beneath, while
broader than those of mainland material, are not as broad as in the males and
never connected across space Cui-Cu2. Lengths of the forewings of the five
female paratypes range from 19.0 to 23.0 mm., averaging 21.2 mm.

Described from 24 specimens, 19 males, and five females, from the Cape region
of Baja California Sur and the adjacent island of Espiritu Santo.

holotype male: Mexico: Baja California Sur: Arroyo San Bartolo, Nov. 3,
1961, Cary-Carnegie Expedition (Lee D. Miller), $ genitalia slide M-1208.

paratypes: Mexico: Baja California Sur: La Paz: Guaycura Hotel grounds,
Nov. 6, 1961, 1^. Baja California Sur: Rancho Vinarama, Oct. 23, 1951, 15 1$.
Baja California Sur: Rancho El Novillo, Oct. 28, 1961, 15. Baja California Sur:
Rancho Palmarito, Oct. 27, 1961, 25; Oct. 31, 1961, 15 1$; Nov. 5, 1961, 15.
Same locality and date as holotype, 1 5 . Baja California Sur: roadway 5 km. south
of Rancho Buenavista, Oct. 25, 1951, 15 1$. Baja California Sur: Boca de la
Sierra, Nov. 17, 1951, 1 5 ; Nov. 18, 1961, 1 5 . Baja California Sur: 4 mi. south
of Arroyo Candelaria, Oct. 26, 1961, 15 1 $ ; Nov. 24, 1961, 15. All foregoing
collected by Cary-Carnegie Expedition. Baja California Sur: 15 mi. south of La
Paz, Aug. 31 1959 (K.W. Radford and F.G. Werner), 25. Baja California Sur:
Candelaro Bay: Espiritu Santo Island, June 30, 1964 (R. Bandar), 3 5 1$.

The holotype male, eleven male and three female paratypes are
deposited in Carnegie Museum. Three male and one female paratypes
are deposited in the California Academy of Sciences. Four male and
one female paratypes are deposited in the collection of C. Don

MacNeill.

1969

Miller-MacNeill : Mexican Hesperiidae

21

This subspecies is not the one mentioned by MacNeill in a previous
paper (1962:98). The specimen he recorded in 1962 was a male
from the Sierra de la Victoria, whereas the present subspecies seems
to be restricted to the lowlands. It is probable that the population
MacNeill cited represents yet another peninsular C. hippalus sub-
species.

C. h. peninsularis is rather widespread throughout the Cape region
in the better watered places. It does not occur in the most arid desert,
and even where it is present it is never abundant. This butterfly has
about the same habits as typical members of C. hippalus in Arizona.

Figs. 1, 2. Cogia hippalus peninsularis, new subspecies. Fig. 1, upper surface
of holotype Arroyo San Bartolo, Baja California Sur, Nov. 3, 1961. Fig. 2,
upper surface of paratype $ , Rancho Vinarama, Baja California Sur, Oct. 23, 1961.

Figs. 3-6. Polites sahuleti margaretae, new subspecies. Fig. 3, upper surface,
and fig. 4, under surface of holotype $ , S.E. shore of La Paz harbor, Baja Cali-
fornia Sur, Dec. 6, 1961. Fig. 5, upper surface, and fig. 6, under surface of para-
type $ , same locality and date.

22

Annals of Carnegie Museum

VOL. 41

Polites sabuleti margaretae, new subspecies
Figures 3, 4 (male holotype), 5, 6 (female paratype),

8 (male genitalia)

male: Head, thorax, and abdomen dark brown clothed with fulvous and brown
scales above and golden-fulvous ones below. Palpi fulvous above and pale
ochreous below. Antennae brown, heavily intermixed with fulvous except on the
nudum. Legs heavily scaled with golden-fulvous in fresh specimens. Upper sur-
face of wings marked about as in P. s. sabuleti (Boisduval), but the fuscous
borders are broader, darker, and more clearly defined. Stigma slender, its width
about equal throughout its length so that its baso-costal margin is as angular-
sigmoid as its anal margin. The fulvous and fuscous areas of the upper surface
are darker than in the nominate subspecies, more closely approximating the shades
seen in P. s. tecumseh (Grinnell), but the fulvous is richer with reddish tones.
The under surface is much darker than in any previously described subspecies.
The forewings have the fuscous border of the upper surface completely reproduced
on the under side with no apical fulvous shading as in other subspecies, and
a well defined basal dark area connected to the marginal border along the inner
margin. The hindwings below are rich chocolate brown heavily overscaled in fresh
specimens with yellow scales, particularly in the area of the anal veins. The other
veins are lightly, but distinctly, indicated by yellow scaling, and the yellow discal
spots are much reduced, resulting in a narrow, but clearly defined, discal band
that is unique in populations of F. s. sabuleti. The basal yellow loop on the under
surface of the liindwings is obsolescent and represented only by yellow bars on
either side of the cell. The male genitalia are identical with those of the nominate
subspecies. Length of forewing of the male holotype is 12.0 mm. The 32 male
paratypes have forewing lengths between 11.0 and 13.0 mm., averaging 12.1 mm.

female: Head, body, and appendages as in the male. The upper surface of
the wings is similar to that of F. s. sabuleti, but of a deeper color, and the fulvous
markings are reduced. The contrast between the dark borders and fulvous discal
markings is greater in F. s. margaretae than in the other nominate subspecies. The
under surface is as that described for the male. The forewing lengths of the 23
female paratypes range from 12.5 to 15.0 mm., averaging 13.8 mm.

Described from 56 specimens, 33 males and 23 females, from the southeastern
shore of La Paz harbor, Baja California Sur.

HOLOTYPE male: Mcxico: Baja California Sur: southeastern shore of La Paz
harbor, Dec. 6, 1961, Cary-Carnegie Expedition (Lee D. Miller), $ genitalia
slide M-1206.

paratypes: Same locality as holotype, Nov. 8, 1961, 3 1$; Nov. 10, 1961,

1^7$; Dec. 5, 1961, 12$ 5$; Dec. 6, 1961, 16$ 10$. All collected by Cary-
Carnegie Expedition.

The holotype male, 25 male and 15 female paratypes are deposited
in Carnegie Museum. Four male and four female paratypes are de-
posited in the California Academy of Sciences. Four male and four
female paratypes are deposited in the collection of C. Don MacNeill.

We take great pleasure in naming this beautiful skipper for Mrs.
Margaret M. Cary of Philadelphia, Pennsylvania, whose generosity
made possible the Cary-Carnegie Expedition and whose work on the

1969

Miller-MacNeill : Mexican Hesperiidae

23

American Sphingidae has so greatly enriched our knowledge of that
group.

This is the subspecies of P. sahuleti, discussed but not named by
MacNeill (1962: 109), from the Cape region. It is by far the darkest
and most distinctly marked of any of the subspecies and may be
readily distinguished by the slender angled stigma of the males above
and by the narrow discal band, pale veins, chocolate-brown ground
color, and obsolete basal loop on the under surface of the hindwings.

The population described is from a narrow strip of the coastal
grassland just inland from the Salicornia flats, but seaward of the more
typical desert scrub. This area is intermittently flooded by storms and
abnormally high tides, but the usual tides do not reach it. The uni-
dentified grass with which P. S. margaretae is associated is apparently
moderately tolerant to salt water (there are some salt deposits on the
ground where the grass is growing) but cannot survive the daily
flooding of the normal intertidal zone; we saw no examples of the
grass in the tidal basin proper.

The butterflies seldom stray far from the grass but visit the associ-
ated flowers regularly. The flight of these skippers is erratic and close
to the ground. This is particularly true of the females, which are
most commonly found in the clumps of grass. The males are very
alert, frequently darting at and chasing others of their own species.

P. s. margaretae is notably local and was found in no other coastal
area sampled by the expedition. It should be sought elsewhere in
suitable habitats in the Cape region, however.

Fig. 7. Cogia hippalus peninmlaris, new subspecies, $ genitalia of holotype.
Fig. 8, Polites sahuleti margaretae, new subspecies, $ genitalia of holotype.

24

Annals of Carnegie Museum

VOL. 41

References Cited

MacNeill, C.D.

1962. A preliminary report on the
tera). Proc. California Acad.

Hesperiidae of Baja California (Lepidop-
ScC (4) 30:91^116.

Back issues of many Annals of Carnegie Museum articles
are available, and a few early complete volumes and parts
are listed at half price. Orders and inquiries should be
addressed to: Publications Secretary, Carnegie Museum,
4400 Forbes Avenue, Pittsburgh, Pa. 15213.

Article 3

Annals of Carnegie Museum

Volume 41

REPORTS ON THE MARGARET M. CARY-CARNEGIE MUSEUM
EXPEDITION TO BAJA CALIFORNIA, MEXICO, 1961
6. The Subfamily Ennominae ( Geometridae : Lepidoptera)

Frederick H. Rindge

Curator, Department of Entomology
The Ameriean Museum of Natural History

This is the sixth of a series of papers based on the Margaret M. Cary-Carnegie
Museum Expedition to Baja California, Mexico, to appear in the Annals of
Carnegie Museum. For an account of the itinerary and description of localities,
see the first paper in this series [Richard M. Fox, Ann. Carnegie Mus. S6 (16):

The Expedition caught a total of 354 specimens representing 45
different species and subspecies that belong to the Ennominae. These
were caught on the peninsula and the mainland. All, with one excep-
tion, have been placed at least to genus. The one exception is a single,
rather worn female from Arroyo San Bartolo, Baja California, Novem-
ber 1, 1961 (genitalic slide F.H.R. No. 12,611). Apparently it belongs
in the Cleorini; a male is necessary for accurate placement. This
species is not included in the text of this paper.

In the Territory of Baja California Sur 32 species were taken. Of
these, five are uncertain as to species. Of the remainder, three are
endemic to the peninsula. Two of these three are species of nearctic
genera and the third is a new subspecies of a widely ranging neo-
tropical species. The faunal relationships of the Ennominae of Baja
California Sur are strongly nearctic, as 22 of the species are northern,
whereas only three belong to present-day neotropical species.

Two new species, one a peninsular endemic and the other occurring
around the Gulf of California in both Sonora and Baja California,
are described in this paper. Two new subspecies of the widely ranging
Thyrinteina arnobia (Stoll) are also described. One is from western
Mexico and the other is an endemic population in the Territory of
Baja California Sur.

Issued Dec. 31,, 1969

U 'i

25

26

Annals of Carnegie Museum

VOL. 41

A word of caution is necessary regarding some of the generic
and specific names used in this paper. A number of the names can
be relied on with confidence; these are ones that have been included f
in modern revisionary studies (Rindge, 1959a, 1959b, 1961, 1966) I'
or are species that occur in the United States. Many of the remaining j
names should be considered as being tentatively applied, pending !
revisionary studies of the groups concerned. |

Abbreviations of museum names used frequently in this article i
are: AMNH, the American Museum of Natural History; CM, Carnegie
Museum.

Subfamily Ennominae

Ixala klotsi Sperry I

Ixala klotsi Sperry, 1940: 146. ^

A single battered specimen is placed here, based primarily on a I
study of the genitalia. The species was described from Arizona. '

BAJA CALIFORNIA SUR: Bahia de Palmas, Oct. 24, 1961, 1$ (CM). I

Pterospoda kinoi, new species
Figs. 1, 2, 7, 8

An undescribed species of Pterospoda occurs around the Gulf of j
California, and it is apparently seasonally dimorphic. The spring
generation is large and has a pinkish brown upper surface, whereas
the summer generation is smaller and the males, in particular, have
much more ochraceous wings.

MALE, SPRING GENERATION: Head with vertex creamy white, becoming dark
gray above front; front brownish gray; palpi brownish gray or grayish brown.
Thorax pinkish gray or brownish gray above, with collar tending to be slightly
darker; whitish below; legs grayish white, fore and middle legs dark gray on
outer surfaces. Abdomen dull pinkish gray above, with scattered brownish gray
scales; paler below.

Upper Surface of Wings: Forewings grayish brown, suflFused with pink, with
veins and subterminal area pinkish; cross lines obsolescent; t. a. and median
lines, when present, nebulous and proceeding across wing at right angle to inner
margin; discal dot black, small; t. p. line, when present, slightly more sinuous
than other lines; s. t. line weakly represented, black, outwardly angulate in cells,
shaded outwardly by white, with angulations broadly filled in with pink; terminal
line absent, intravenular spots black; fringe concolorous with wing. Hind wings
concolorous with forewings; maculation obsolescent except for s. t. line, similar
to that of forewing; discal dot absent. (See fig. 1.)

Under Surface of Wings: All wings pale gray, more or less irrorate with dark
gray scales; costa of forewing, and some veins, pale brownish gray; without
maculation except for discal dot on forewings, and with nebulous, dark gray
subterminal band on forewings of some specimens.

1969

Rindge: Mexican Ennominae

27

Length of Forewing: 18 to 19 mm.; holotype, 18 mm.

female, spring generation: Similar to male, or more suffused with dark gray

scaling.

Length of Fore wing: 14 to 19 mm.; allotype, 14 mm.

MALE, summer GENERATION: Similar to male of spring generation but smaller,
with front and upper surfaces of body and wings ochraceous, overlain with dark

Figs. 1, 2. Pterospoda kinoi, new species. 1. Holotype, male. 2. Paratype, male.
Fig. 3. Morina daedalea, new species, holotype.

Fig. 4. Anacamptodes cerasta Rindge, holotype.

Fig. 5. Thyrinteina arnobia tephra, new subspecies, holotype.

Fig. 6. Thyrinteina arnobia picta, new subspecies, holotype. All figures xl.5.

28

Annals of Carnegie Museum

VOL. 41

gray and grayish brown scales; terminal area of all wings dark gray or grayish
brown; faint traces of pink in siibterminal area and on veins. Under surface with
definite subterminal band on forewings. (See fig. 2.)

Length of Forewing: 14 to 15 mm.

FEMALE, SUMMER GENERATION: Upper surfaces of thorax and abdomen and all
wings dark grayish brown above; upper surface without maculation except for
discal dot on forewings and for reddish brown subterminal band on all wings,
sometimes obsolescent, and for traces of s. t. line. Under surface grayish brown,
with outer portion of wings darker.

Length of Forewing: 15 to 16 mm.

MALE GENITALIA: Uncus elongate, curving; gnathos weakly sclerotized, anterior
margin broad, flat; valves slender, apically curving, inner face with numerous
slender setae; costa of valve with large, sclerotized process at base, extending
about one-third of length of costa, roughly triangular in outline, but with apex
curved inwardly, and with median ridge extending from apex to base of valve;
transtilla represented by inwardly projecting lobes meeting at mid-line; anellus

Figs. 7, 8. Genitalia of Pterospoda kinoi, new species. 7. Holotype, male. 8. Allo-
type, female.

1969

Rindge: Mexican Ennominae

29

small, elliptical; aedeagus very long and slender, longer than combined lengths
of uncus, tegumen, and saccus; vesica armed with row of heavy spines occupying
about one-half length of aedeagus.

FEMALE genitalia: Sterigma elliptical, bluntly pointed anteriorly and flattened
posteriorly, membranous, with small lunate sclerotized piece medially; ductus
bursae sclerotized, elongate, weakly constricted medially, anterior end partly ro-
tated; ductus seminalis arising dorsally from near posterior end of corpus bursae;
latter large, increasing in width anteriorly, and with ventral swelling, central
portion thickly set with numerous inwardly projecting spinose processes.

TYPES: Holotype, male: Mexico: Sonora: Bahia San Carlos: February 18, 1963
(P. H. Arnaud, Jr.), AMNH; allotype, female, same data but February 19, 1963
(AMNH). The genitalia of the holotype are mounted on slide F.H.R. No.
11,590; and of the allotype on F.H.R. No. 11,715. Paratypes, all from Mexico:
same data as holotype, February 17, 19, 1963, three males (AMNH); Gulf of
California: Isla San Marcos (27°13'N., 112°50'W.), March 28, 1962 (C. F.
Harbison), six males and five females (San Diego Natural History Museum);
Baja California Sur: Arroyo San Bartolo, November 13, 1961 ( Cary-Carnegie
Museum Expedition), three males and one female (CM); Baja California Sur:
La Paz: Guaycura Hotel grounds, October 20, 1961 (Cary-Carnegie Museum
Expedition ) , one female ( CM ) .

remarks: a total of 21 specimens (13 males and eight females) and eight
genitalic dissections (four males and four females) have been studied.

There is no doubt that the moths from these three localities belong
to a single species, as the genitalia are so similar. The only question
is whether the population on the peninsula should be accorded sub-
specific status or not. Until adequate material is obtained, captured
throughout the year, it is thought advisable to place these differently
colored moths as seasonal forms of the same species.

Pterospoda kinoi is closely related to P. opuscularia (Hulst), and
can be distinguished therefrom by its darker color and less definitely
defined maculation, as well as by the genitalia. The male structures
of the present species can be recognized by the non-spinose costal
process, and by the row of numerous spines in the vesica of the
aedeagus. The female genitalia of kinoi lack the heavily sclerotized
ostial ring of opuscularia, and have a much longer ductus bursae and
a ventral swelling of the corpus bursae not present in opuscularia.

This species is named in honor of Father Eusebio Francisco Kino,
the great explorer and missionary of Baja California and Sonora.

Syrrhodia decrepitaria Hiibner
Syrrhodia decrepitaria Hiibner, 1823: 29, figs. 371, 372.

Two specimens were taken in Baja California Sur. This species
occurs in much of South America, extending north as far as southern
Texas. It is apparently not rare in Mexico.

30

Annals of Carnegie Museuivi

VOL. 41

BAJA CALIFORNIA suR: La Paz: Guaycura Hotel grounds, Dec. 2, 1961, 1^;
Arroyo San Bartolo, Nov. 13, 1961, 1$. (1^ CM, 1^ AMNH).

Thysanopyga carfinia (Dmce)

Pachydia carfinia Dmce, 1893 (1891-1900): 136; 1893 (1881-1900),
pi. 53, figs. 25, 26.

Dmce described carfinia from speeimens taken in Veracruz, Mexico,
Guatemala, and Chiriqui, Panama. The species is known to occur in
Chiapas and Oaxaca, Mexico, and in British Honduras.

SINALOA: 16 miles north of Mazatlan, Oct. 28, 1961, 1$ (CM).

Thysanopyga sp.

A study of the types, plus a designation of lectotypes, is necessary
in the complex of small species that includes nicetaria Guenee, cermala
Druce, and oroanda Dmce, before a meaningful usage of names is
possible. An examination of the genitalia shows that a number of
species are more easily recognized by the genitalia than by maculation;
the latter can be quite variable within a species.

SONORA: Navojoa, Oct. 21, 1961, 1$.

SINALOA: Mazatlan: at motel, Oct. 24, 1961, 1 ; 5 miles \vest of Concordia,

Nov. 2, 1961, 2^,5$; 16 miles north of Mazatlan, Oct. 28, 29, 1961, 65, 10$.

The specimens are in the following collections: 12$ CM; SJ*,

4$ AMNH.

Chloraspilates bicolor aria Packard
Chloraspilates hicoloraria Packard, 1876:212, pi. 13, fig. 40,

This species ranges from Texas to southern Nevada and south-
eastern California; in Mexico it is known from Coahuila, Chihuahua,
and Sonora. The single specimen from the trip is a small, worn and
faded male that has been determined largely on the basis of a study
of its genitalia.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 26, 1961, 15
(CM).

1969

Rindge; Mexican Ennominae

31

Fernaldella fimitaria angelata Wright
Fernaldella fimitaria angelata Wright, 1923: 114.

Wright’s type material of three males and one female came from
Angeles Bay, Lower California (holotype and paratype), and from
Guaymas, Sonora (allotype and paratype). The second paratype
referred to has been examined and dissected, and its genitalia agree
with the notes on the holotype genitalia kindly provided me by Dr.
C. D. MacNeill. This population is apparently correctly placed as
fimitaria, and not partita.

BAJA CALIFORNIA SUR: La Paz: Giiaycura Hotel grounds, Oct. 20 to Dec. 5,
1961, 5^, 2$; Bahia de Palmas, Oct. (without date) 1961, 1^. (4^, 1$ CM;
2^ 19 AMNH).

Semiothisa punctolineata (Packard)

Macaria puncto-lineata Packard, 1873: 64.

Packard described this species from Texas. It extends from that
state across the southeastern portion of the United States to Florida.
Specimens of this species, taken in Sinaloa, Sonora, and Baja California
Sur by members of the expedition, are in the following collections:
7d',3? CM;5d',5? AMNH.

SINALOA: 5 miles west of Concordia, Nov. 2, 1961, 6(^, 2$; 16 miles north of
Mazatlan, Oct. 28, 1961, 3 5 ; 18 miles north of Mazatlan, Oct. 29, 1961, 1 ^ , 1 $ .

Sonora: Navojoa, Oct. 21, 1961, 2^; Guaymas, Oct. 20, 1961, 29.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct 20, 1961, 29;
Bahia de Palmas, Nov. 16, 1961, 1 9 .

Semiothisa errata McDunnough

Semiothisa errata McDunnough, 1939: 252.

Two specimens of this moth, described from Arizona, were taken.
BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20, Dec. 2, 1961,
1^, 19. (15 CM; 19 AMNH).

Semiothisa parcata ( Grossbeck )

Sciagraphia parcata Grossbeck, 1908: 26.

This species is common in Arizona. It also occurs in Nevada, Sonora,
and Baja Galifornia Norte.

SONORA: Sonoyta, Oct. 17, 18, 1961, 35, 39; Guaymas, Oct. 20, 1961, 19.
BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20, 1961, 1 9 .

32

Annals of Carnegie Museum

VOL. 41

The specimens are in the following collections: 2 cT , 3 ? CM; 1 cf , 2 ?
AMNH.

Semiothisa colomta Grote i!

Semiothisa colorata Grote, 1883: 7.

!|

S. colorata is one of the commoner species in the semi-arid portion i
of the southwestern United States, as it extends from southern Califor- j
nia to western Texas. It is also known from Sonora and Baja California
Norte. I

SONORA: Sonoyta, Oct. 17, 18, 1961, 2 5, 7$. i

I

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20, Nov. 6, 1961, I

2$. !

The specimens are in the following collections: Icf, 5$ CM; Id, '
4 ? AMNH. I

Semiothisa sp., Psublacteolata (Hulst)

Sciagraphia suhlacteolata Hulst, 1887: 189.

Hulst described this species from Dayton, Ohio; the few specimens
that have been available for study were taken in Texas. The single
specimen taken on the expedition is smaller and darker than the single
Texas female before the author. The genitalia of these two moths are
very similar. Pending receipt of adequate material, the specimen is
tentatively referred to this species.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 31, 1961, 1 $ .
(CM).

Semiothisa pictipennata (Hulst)

Macaria pictipennata Hulst, 1898: 162.

This species occurs from western Texas to California, and it is known
from Baja California Norte.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20 to Dec. 5,
1961, 7 5, 4$. (3 5,2$ GM; 4 5,2$ AMNH).

1969

Rindge: Mexican Ennominae

33

Semiothisa nigricomma Warren
Semiothisa nigricomma Warren, 1904: 127.

Warren described this species from Guadalajara, Jalisco. It extends
north into southern Texas.

SINALOA: 16 miles north of Mazatlan, Oct. 28, 1961, 2$; 18 miles north of
Mazatlan, Oct 29, 1961, 4^,3$. (2^,3$ CM; 2^,2$ AMNH).

Semiothisa sp.

Six specimens were taken of a species that is allied to nigricomma;
this moth is also known from southern Arizona.

BAJA CALIFORNIA suR: La Paz: Guaycura Hotel grounds, Nov. 2, 5, 1961, 1^,
2$; Arroyo San Bartoio, Nov. 13, 1961,15,29.(1^,2$ CM; 15, 2$ AMNH).

Semiothisa sirenata McDunnough
Semiothisa sirenata McDunnough, 1939: 254.

This species was described from Arizona; it also occurs in southern
California, southern Nevada, and Sonora.

SONORA: Navojoa, Oct, 21, 1961, 35; Sonoyta, Oct. 17, 1961, 2$,

BAJA CALIFORNIA SUB: Bahia de Palmas, Oct. 24, 1961, 1 5 ; La Paz: Guaycura
Hotel grounds, Oct. 20, Nov. 6, 1961, 2 $ .

The specimens are in the following collections: 3^, 2$ CM; IJ*,
2? AMNH,

Semiothisa cyda ( Druce )

Eubolia cydii Dmce, 1892 (18914900): 177; 1893 (1881»1900), pi 58,
fig. 4 (lectotype).

Semiothisa cyda: Rindge, 1959b: 8, figs. 2, 6, 10 (male and female
genitalia ) .

This species is very common and widely ranging throughout the
southwestern United States and much of Mexico. See Rindge (1959b)
for a complete list of the states involved.

SONORA: Sonoyta, Oct 17, 18, 1961, 65* 14$.

BAJA CALIFORNIA SIM: La Paz: Guaycura Hotel grounds, Oct. 20 to Dec. 6,
1961, 255* 28$.

The specimens are in the following collections: 15 21? CM;
16d^, 21? AMNH.

34

Annals of Carnegie Museum

VOL. 41

Semiothisa melanderi Sperry

Phasiane s-signata form ballandrata Wright, 1923: 115.

Semiothisa melanderi Sperry, 1948: 58. Rindge, 1959b: 13, figs. 4, 8, 11
(male and female genitalia).

This species looks almost exactly like S. cijda. Specimens of the two
were captured together at La Paz. Good differences are present in the
ventral plate of the male, as well as in the genitalia of both species.
It has been known heretofore from Arizona and Baja California Norte.

Wright’s older “form” name is not used, following Article 45 of the
International Code of Nomenclature.

SONORA: Navojoa, Oct. 21, 1961, 1^, 3$; Guaymas, Oct. 20, 1961, 1$.

4$, 12$.

BAJA CALIFORNIA SUR: La Paz: Guayciira Hotel grounds, Oct. 19 to Dec. 6, 1961,
4 5, 12$.

The specimens are in the following collections: 2d', 10? CM; 3d
6? AMNH.

Semiothisa hypaethrata (Grote)

Phasiane hypaethrata Grote, 1881: 167.

This species is known from southern Arizona and southwestern
New Mexico.

SONORA: Sonoyta, Oct. 17, 1961, 2$; Guaymas, Oct. 20, 1961, 15.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 19 to Nov. 6,
1961, 3 5, 4$; Bahia de Palmas, Oct. 24, 1961, 1$.

The specimens are in the following collections: 3d, 3? CM; Id,
4? AMNH.

Semiothisa sp.

Tliree worn females represent a species unknown to me; they have
the seventh sternite sclerotized.

BAJA CALIFORNIA SUR: San Jose del Cabo, Nov. 17, 1961, 2$ ; Bahia de Palmas,
Nov. 16, 1961, 1$. (2$ CM; 1$ AMNH.)

Prophasiane sp.

A single female was taken that is related to mendicata Hulst.

BAJA CALIFORNIA SUR: Bahia de Palmas, Nov. 17, 1961. 1$ (CM).

1969

Rindge: Mexican Ennominae

35

Itame sobriaria Barnes and McDunnough, new status

Itame graphidaria sobriaria Barnes and McDunnough, 1917: 239, pi.
24, fig. 9.

This name was proposed as an Arizona subspecies of the Texan
graphidaria Hulst. The two are similar in appearance but the genitalia
of both sexes are quite distinct. This species is also known from New
Mexico and Chihuahua.

BAJA CALIFORNIA suR: La Paz: Guaycura Hotel grounds, Oct. 20 to Nov. 10,
1961, S$,49. (2$, 29 CM; 1$,29 AMNH).

Glaucina eupetheciaria osiana (Druce)

Eupithecia (?) osiana Druce, 1893 (1891-1900): 147; 1893 (1881-1900),
pi. 55, figs. 5, 6.

Glaucina eupetheciaria osiana: Rindge, 1959a: 294, pi. 24, figs. 8, 9;
text fig. 24.

A number of specimens of both sexes were taken in the Central
Gulf Coast area, and these are tentatively assigned to the Sonoran
subspecies of eupetheciaria. Unfortunately most of the specimens were
rubbed, so that an accurate comparison with Sonoran examples is not
feasible. The Baja California examples seem to be paler than osiana
but more material is needed before this can be ascertained for certain.

SONORA: Navojoa, Oct. 21, 15.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20, 31, Nov.
5, 8, Dec. 6, 1961, 55,4$; Bahia de Palmas, Oct. 24, Nov. 16, 1961, 8$; San
Jose del Cabo, Nov. 17, 1961, 1 $ .

The specimens are in the following collections: dcf, 6? CM; 2cr,
7? AMNH.

Glaucina anomala Rindge

Glaucina anomala Rindge, 1959a: 341, pi. 27, figs. 10, 11; text figs. 31,
70, 103.

Four specimens of this anomalous species were taken in the Central
Gulf Coast area near the sea. It has been previously known from
southeastern Arizona and the lower portions of the Sierra Madre
Occidental of southern Sonora.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20, 23, 1961,
15,1$; Bahia de Palmas, Nov. 12, 16, 1961, 25. (15, 1$ CM; 2 5 AMNH).

36

Annals of Carnegie Museum

VOL. 41

Synglochis perumbraria Hulst

Synglochis perumbraria Hulst, 1896: 352. Rindge, 1959a: 343, pL 27,
figs. 12, 13; text figs. 33^35, 71, 104.

This species occurs throughout the southwestern United States; in
Mexico it is known to occur in Coahuila and Baja California Norte,
The five specimens taken are the first ones examined from Sonora,

SONORA: Sonoyta, Oct. 17, 18, 1961, 2^, 2$; 10 miles east of San Luis, Oct
17, 1961, 1$. (1^,2$ CM; 1^,1$ AMNH).

Eubarnesia ritaria ritaria ( Grossbeck )

Cedaria ritaria Grossbeck, 1910: 207.

Eubarnesia ritaria ritaria: Rindge, 1959a: 346, pi. 27, figs. 14, 15; text
figs. 36, 72, 105.

One specimen was taken near the Arizona border in Sonora. Nomi-
nate ritaria is found in Arizona and western Texas.

SONORA; Sonoyta, Oct. 17, 1961, 1^ (CM).

Eubarnesia ritaria arida Rindge

Eubarnesia ritaria arida Rindge, 1959a: 347, pL 27, figs. 16, 17; text
fig. 36.

This paler subspecies is known to occur in southern California,
northern Sonora, and Baja California Norte. The present records
extend its distribution much farther south on the peninsula.

BAJA CALIFORNIA SUR: Bahia de Palmas, Oct. 24, Nov. 16, 1961, 3^; La Paz:
Guaycura Hotel grounds, Oct. 26, Nov. 1, 6, 8, 1961, 2^, 3$. (3^, 1$ CM;
2$, 2$ AMNH).

Morina daedalea, new species
Figs. 3, 9, 10

This species can be recognized by the presence of prominent discal
spots on all wings above, and by the genitalia of both sexes. It is
endemic to Baja California Sur.

male: Head with vertex white or ochraceous white; front ochraceous around
large truncate tubercle; palpi ochraceous. Thorax ochraceous white above, paler
below; legs ochraceous brown, with all tarsi darker. Abdomen ochraceous above,
with white and ochraceous brown scales; paler below.

1969 Rindge: Mexican Ennominae 37

Upper Surface of Wings: Forewings grayish white, with ochraceous and brown
scaling; cross lines dark brown, narrow; t. a. line crossing costa at right angle,
then swinging basad and paralleling costa to inner margin; discal spot large, dark
brown, circular; t. p. line extending from about vein Rs and proceeding, with
two basal bends, to inner margin just basad of its middle; large patch of ochra-
ceous brown scales distad of t. p. line above inner margin, and continuing anteri-
orly as shade band; subterminal and terminal areas with some brown scaling in
cells; dark intravenular dots present along wing margin; fringe concolorous with
wing. Hind wings white, with scattered brown scales, particularly along outer
and anal margins; discal dot dark grayish brown, prominent; traces of intradiscal,
extradiscal and subterminal lines present at anal margin and extending part way
across wing; terminal line dark, narrow, obsolescent.

Under Surface of Wings: Grayish white, with variable number of dark brown
and ochraceous scales; without maculation except for large, brownish black discal
dots, and narrow, dark terminal line.

Length of Forewing: 10 to 12 mm.; holotype, 10.5 mm.

female: Similar to male, as far as can be seen.

Length of Forewing: 10 mm. (allotype).

male genitalia: Uncus elongate, triangular, with sides weakly concave;
gnathos with large median enlargement bluntly pointed or rounded; valves with
sacculus broad, lightly sclerotized, and with raised ridge having from six to 12
heavy short, well-spaced spines; anellus with large, sclerotized, rounded anterior
part, becoming narrowed medially and with posterior margin rounded; aedeagus
longer than combined lengths of uncus, tegumen, and saccus, posterior end bifid,
with narrow, heavily sclerotized tine-like process and wider, hollow, less heavily
sclerotized projection, and with anterior end broadly swollen; vesica unarmed.

Figs. 9, 10. Genitalia of Morina daedalea, new species. 9. Holotype, male. 10.
Allotype, female.

38

Annals of Carnegie Museum

VOL. 41

FEMALE GENITALIA: Sterigiual area broad, sclerotized, anterior margin bilobate;
ductus bursae short, broad, curved, and with large swelling on left side; ductus
seminalis arising from posterior end of corpus bursae on right side; corpus bursae
globular, membranous, with large dorsal signum having transverse ridge.

types: Holotype, male, and allotype, female, Guaycura Hotel grounds. La Paz,
Baja California Sur, Mexico, October 31, 1961, Cary-Carnegie Museum Expedi-
tion (CM). The genitalia of the holotype are on slide F.H.R, No. 12,601, and of
the allotype on F.H.R. No. 14,580. Paratypes: same data as types, November 5
and December 1, 1961, two males (CM; AMNH).

remarks: The male genitalia of daedalea can be separated from those of M,
coniferarum Grossbeck by the much sparser spining on the sacculus and by the
bifurcate apex of the aedeagus. In the female genitaha, the present species can
be recognized by the wider sclerotized sterigma and by the thicker, curved, and
swollen ductus bursae.

The description of the adults may have to be somewhat modified when addi-
tional material comes to hand.

Anavitrinella pampinaria (Guenee)

Boarmia pampinaria Guenee, 1857: 245.

This species occurs from coast to coast in the United States, and
south into Mexico.

SONORA: Navojoa, Oct. 21, 1961, 1^ (CM).

Anavitrinella sp.

This species is yellowish brown, and has a curved t. p. line; it is
placed in Anavitrinella on the basis of the genitalic structures. No
name is being given to it at this time, pending a revisionary study of
this genus.

BAJA CALIFORNIA SUR: Arroyo San Bartolo, Nov. 1, 13, 1961, 1^, 2$; La Paz:
Guaycura Hotel grounds, Nov. 6, 8, 1961, 2 $ ; Bahia de Palmas, Nov. 17, 1961,
1$. (4$ CM; 15,1$ AMNH.)

Anacamptodes pseudoherse Rindge

Anacamptodes pseudoherse Rindge, 1966: 218, pi. 24, fig. 8; text figs.
27, 47.

The two specimens that were taken have been made paratypes.
This species is widely distributed over much of Mexico.

SINALOA: Los Mochis, Oct. 22, 1961, 1$ (CM).

BAJA CALIFORNIA SUR: Arroyo San Bartolo, Nov. 13, 1961, 15 (CM).

1969

Rindge: Mexican Ennominae

39

Anacamptodes cerasta Rindge
Fig. 4

Anacamptodes cerasta Rindge, 1966: 221, pi. 24, fig. 11; text figs. 28, 48.

This species is endemic in the Territory of Baja California, and has
been captured only at La Paz.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20, 23, 31, Nov.
5, 8, Dec. 1, 6, 1961, 75, 4$. (45,2$ CM; 35,2$ AMNH).

Thyrinteina arnobia picta, new subspecies
Fig. 6

Thyrinteina arnobia phala Rindge, 1961: 135 (partim). Comstock and
Vazquez, ‘1963” [1964]: 263, fig. 18.

In this population, found in western Mexico and Guatemala, the
wings of the males are predominately brown.

male: Head, thorax and abdomen similar to those of nominate subspecies, but
with last two tending to have more dark scaling.

Upper Surface of Wings: Forewings, ground color gray, heavily and evenly
overlain with brown and dark brown scales, the latter tending to be concentrated
in median area. Hind wings concolorous with forewings, or with ground color
more prominent in sub terminal and terminal areas; maculation of all wings hke
that of nominate subspecies.

Under Surface of Wings: All wings unicolorous dark brown or dark grayish
brown; either without maculation or with t. p. and extradiscal lines weakly
represented.

Length of Forewing: 16 to 19 mm.; holotype, 19 mm.

female: Similar to that of phala but tending to be slightly more suffused with
dark scales, and with s. t. line shghtly stronger.

Length of Forewing: 27 mm. (allotype).

male genitalia: Similar to those of phala.

FEMALE genitalia: Not examined.

types: Holotype, male, Agua del Obispo, Guerrero, Mexico, July, 1932 (C. C.
Hoffmann; paratype of pnala Rindge) (AMNH); allotype, female, Puerto Vallarta,
Jalisco, Mexico, emerged from pupa September 15, 1957 (J. A. Comstock)
(AMNH). The genitalia of the holotype are on slide F.H.R. No. 8987. Para-
types, all from Mexico: same data as holotype, one male (AMNH); “La Granja,”
Chiapas, July, 1930 (C. C. Hoffmann; paratype of phala Rindge), one male
(AMNH); 5 miles west of Concordia, Sinaloa, November 2, 1961 ( Cary-Carnegie
Museum Expedition), two males (AMNH, CM); 16 miles north of Mazatlan,
Sinaloa, October 28, 1961 (Cary-Carnegie Museum Expedition), one male, CM.
(45, 1$ AMNH; 25 CM).

40

Annals of Carnegie Museum

VOL. 41

remarks: Seven specimens and one genitalic dissection were examined. Typi-
cal phala is characterized by the white upper surface of the wings in the male
( Rindge, 1961, pi. 22, fig. 2). It occurs from southern Texas through eastern
Mexico and at least as far south as British Honduras. In Mexico it is known from
the states of San Luis Potosi, Veracruz, Tabasco, and Yucatan, and from Alta
Verapaz in Guatemala. Included in the type series were four dark-brown speci-
mens from Guerrero and Chiapas. With the receipt of additional material from
western Mexico it becomes apparent that these brown specimens form a distinct
population in that part of the country.

Thyrinteina arnobia tephra, new subspecies
Fig. 5

In Baja California Sur there occurs a population in which the upper
surface of the wings of the males is pale gray, overlain with dark
scales, and is thus quite distinct from the other two known popula-
tions from Mexico.

MALE: Head, thorax, and abdomen similar to those of nominate subspecies, but
with abdomen tending to have more dark scales.

Upper Surface of Wings: Forewings with ground color pale gray, lightly over-
lain with grayish brown and dark-brown scales. Hind wings concolorous with,
or slightly paler than, forewings; maculation of all wings like that of nominate
subspecies.

Under Surface of Wings: All wings varying in color from gray to dark-grayish
brown, and overlain with grayish-brown scales; maculation varying from having
t. p. and extradiscal lines present to being obsolescent.

Length of Forewing: 15 to 20 mm.; holotype, 16 mm.

female: Similar to female of phala.

Length of Forewing: 26 to 32 mm.; allotype, 28 mm.

MALE GENITALIA: Similar to those of phala but with projections of juxta slightly
longer and more slender.

FEMALE genitalia: Not examined.

types: Holotype, male, Arroyo San Bartolo, Baja Galifornia Sur, Mexico, No-
vember 15, 1961 (GM); allotype, female, same data, November 1, 1961 (GM).
The genitalia of the holotype are on slide F.H.R. No. 12,573. Paratypes, all from
Baja Galifornia Sur: same data as holotype, various dates between November 1
and 15, 1961, eight males (GM; AMNH); Guaycura Hotel grounds. La Paz,
various dates between October 20 and December 2, 1961, 11 males (CM;
AMNH); Bahia de Palmas, various dates between October 24 and November 17,
1961, five males and two females (CM; AMNH); San Jose del Cabo, November
17, 1961, one male (CM); Puerto Chileno, November 25, 1961, one female (CM).
All of the above specimens were taken on the Cary-Camegie Museum expedition.
(1S$, 19 CM; 12^, 29 AMNH.)

REMARKS: Thirty specimens and two genitalic dissections have been studied.
This population is quite distinct from any of the described ones, and can be
recognized by the light-gray coloration of the upper surface of the wings of the
males.

1969

Rindge: Mexican Ennominae

41

Stergamataea delicata dolliata Grossbeck
Stergamataea dolliata Grossbeck, 1908: 30.

This population was described from Arizona; it also occurs in west-
ern Texas. The single specimen taken by the members of the expe-
dition tends to be slightly darker and more mottled than are most
Arizona males. More material is needed before we can ascertain
whether or not these diEerences are constant.

BAJA CALIFORNIA SUR: Arroyo San Bartolo, Nov. 15, 1961, 1$ (CM).

N ematocampa sp., Preticulata Butler
Nematocampa reticulata Butler, 1881: 323.

Butler described reticulata from the Amazons of Brazil, and Druce
[1892 (1891-1900): 43] gave Mexican records in the states of Sinaloa,
Veracruz, and Tabasco. A careful genitalic study and revision of the
group is necessary before names can be safely applied in this genus.

SINALOA: 5 miles west of Concordia, Nov. 2, 1961, 1^ (CM).

Steuaspilates meskearia albomacularia ( Hy. Edwards )

Azelina albomacularia Hy. Edwards, 1882: 130.

The four specimens that were taken are smaller in size (length of
forewing for the males, 12-13 mm., for the females, 12 mm. ) than are
most of the examples from Arizona and southern Galifornia. However,
some of the individuals from the latter state are as small. The range
of color variation of the upper surface of the wings in albomacularia
is rather extensive, and the La Paz specimens are well within this
range.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Oct. 20, 26, Nov. 8,
Dec. 5, 1961, 2^,2$. (1^,1$ CM; 1^,19 AMNH).

Pero lignata (Warren)

Pergama lignata Warren, 1897: 499.

Specimens bearing this name are from South America, Costa Rica,
Guatemala, and southern Mexico.

SINALOA: 5 miles west of Concordia, Nov. 2, 1961, 1 ^ ; 16 miles north of Ma-
zatlan, Oct. 28, 1961, 1 $ ; 18 miles north of Mazatlan, Oct. 29, 1961, 1 $ . ( 1 ^ ,
19 CM; 19 AMNH).

42

Annals of Carnegie Museum

VOL. 41

Pero melissa (Druce)

Azelina melissa Druce, 1892 (1891-1900): 63; 1892 (1881-1900), pi.
47, fig. 12.

Druce described this species from Jalapa, Veracruz, Mexico.

BAJA CALIFORNIA SUR: Arroyo San Bartolo, Nov. 13, 15, 1961, 2$. (1$ CM;
1$ AMNH).

Philtraea elegantaria (Hy. Edwards)

Zerene elegantaria Hy. Edwards, 1881: 121.

This species was described from Arizona.

BAJA CALIFORNIA SUR: La Paz: Guaycura Hotel grounds, Dec. 2, 1961, 1^;
Bahia de Palmas, Nov. 17, 1961, 1 $ ; Puerto Chileno, Nov. 24, 1961, 1 $ ; San
Jose del Cabo, Nov. 17, 1961, 1$; Arroyo San Bartolo, Nov. 13, 1961, 2$;
Sierra Laguna: Rancho San Bernardo, Nov. 13, 1961, 2$. (1^,4$ CM; 3$
AMNH.)

Apicia melenda Druce

Apicia melenda Druce, 1892 (1891-1900): 39; 1892 (1881-1900), pi.
44, figs. 23, 24.

Druce described melenda from southern Mexico (Jalisco, Guerrero,
Tabasco, and Veracruz), Guatemala, and Panama.

SINALOA: 5 miles west of Concordia, Nov. 2, 1961, 15 (CM).

Apicia vibicaria ( Cramer )

Phal. Geomet. vibicaria Cramer, 1775: 112, pi. 71D.

This species is of wide occurrence in South and Central America,
going as far north as southern Mexico.

SINALOA: 16 miles north of Mazatlan, Oct. 28, 1961, 1$ (CM).

Halesa meticulaia ( Guenee )

Drepanodes meticulata Guenee, 1857 : 68.

This widely ranging species occurs from Brazil and Peru, through
northern South America and Central America, into Mexico. In the last
country, specimens have been examined from the states of Chiapas,
Oaxaca, Hidalgo, and Veracruz.

1969

Rindge: Mexican Ennominae

43

SINALOA: 5 miles west of Concordia, Nov. 2, 1961, 4$; 16 miles north of
Mazatlan, Oct. 28, 1961, 1 $ .

SONORA: Navojoa, Oct. 21, 1961, 15 .

The specimens are in the following collections: 4cf CM; 2cr AMNH.

Drepanodes epionata Guenee
Drepanodes epionata Guenee, 1857 : 68.

Guenee described this species from Haiti, and it is known from the
other islands of the Greater Antilles. Specimens are also placed under
this name from Central and northern South America; and it is known
from southern Texas.

SINALOA: 18 miles north of Mazatlan, Oct. 29, 1961, 15 (CM).

References Cited
Barnes, W., and J. McDunnough

1917. New species and varieties of Geometridae. Contributions to the natural
history of the Lepidoptera of North America. Decatur, Illinois, the

Review Press, 3: 213-298, pis. 16-33.

Butler, A. G.

1881. On the Lepidoptera of the Amazons, collected by Dr. James W. H. Trail
during the years 1873-1875. Trans. Ent. Soc. London, pt. 3: 315-349.

Comstock, John Adams, and Leonila Vazquez Garcia

1964. Estudios de los ciclos biologicos en Lepidopteros Mexicanos. II. Hetero-
cera. An. del Inst, de Biologia, 1963, 34: 217-273.

Cramer, P.

1775-1776. Plaaten van Uitlandsche Kapellen (Planches des Papillons Exo-
tiques). Amsterdam, S. J. Baalde & Utrecht, B. Wild, 1: xxx + 156 pp.,
pis. 1-96.

Druce, Herbert

1881-1900. Biologia Centrali-Americana. Insecta. Lepidoptera-Heterocera. Lon-
don, 3: pis. 1-101.

1891-1900. Op. cit. London, 2; 1-622.

Edwards, Henry

1881. Descriptions of some new species of Heterocera. Papilio, 1: 115-121.

1882. Descriptions of new species of N. Am. Heterocera. Papilio, 2: 123-132.

Guenee, A.

1857. Histoire naturelle des insectes. Species general des lepidopteres. Paris,
9: lvi-1-514 pp.

Grossbeck, John A.

1908. New moths of the family Geometridae. Jour. New York Ent. Soc., 16:
19-31.

1910. New species and one new genus of Geometridae. Ibid., 18: 199-207, pi. 6.

44

Annals of Carnegie Museum

VOL. 41

Grote, a. R.

1881. New moths from Arizona, with remarks on Catocala and Heliothis.
Papilio, 1: 153-168.

1883. New species and notes on structure of moths and genera. Canadian Ent.,
15: 3-13.

Hubner, J.

1823. Zutrage zur Sammlung exotischer Schmetterlinge. Augsburg, 2: 300 pp.,
figs. 371-372.

Hulst, Geo. D.

1887. New species of Geometridae, No. 3. Ent. Amer., 2: 185-192.

1896. A classification of the Geometrina of North America, with descriptions of
new genera and species. Trans. Amer. Ent. Soc., 23: 245-386, pis. 10, 11.

1898. Descriptions of new genera and species of the Geometrina of North
America. Canadian Ent., 30: 158-164.

McDunnough, J.

1939. New species of Ceometridae with notes, II. Canadian Ent, 71: 249-258.
Packard, A. S.

1873. Descriptions of new American Phalaenidae. Fifth Kept. Peabody Acad.
Sci., pp. 52-81.

1876. A monograph of the geometrid moths or Phalaenidae of the United
States. In Hayden, F. V., Report of the United States Geological Survey
of the territories. Washington, 10: 607 pp., 12 pis.

Rindge, Frederick H.

1959a. A revision of Glaucina, Synglochis, and Eubarnesia ( Lepidoptera, Geo-
metridae). Bull. Amer. Mus. Nat. Hist., 118: 259-366, figs. 1-106, pis.
23-27.

1959b. Descriptions of and notes on North American Geometridae (Lepidop-
tera), No. 4. Amer. Mus. Novitates, no. 1968: 1-17, figs. 1-13.

1961. A revision of the Nacophorini (Lepidoptera, Geometridae). Bull. Amer.

Mus. Nat. Hist., 123: 87-154, figs. 1-46, pis. 18-23.

1966. A revision of the moth genus Anacamptodes (Lepidoptera, Geometridae).
Ibid., 132: 175-244, figs. 1-53, pis. 22-25.

Sperry, John L.

1940. Two apparently undescribed species of Geometridae from the south-
west. Canadian Ent., 72: 145-147.

1948. Southwestern Geometrid notes and new species. I. Bull. Brooklyn Ent.
Soc., 43: 54-60.

Warren, William

1897. New genera and species of Thyrididae, Epiplemidae, and Geometridae,
from South and Central America and the West Indies, in the Tring
Museum. Novitates Zook, 4: 408-507.

1904. New American Thyrididae, Uraniidae, and Geometridae. Ibid., 11:
1-173, 493-582.

Wright, W.

1923. Expedition of the California Academy of Sciences to the Gulf of Cali-
fornia in 1921. IX. The geometrid moths. Proc. California Acad. Sci.,
fourth series, 12: 1 13-115.

I. The Cuban Subspecies by Albert Schwartz

II. The Bahamian Subspecies by Clarence J. McCoy

III. Discussion by Albert Schwartz

^thso/v7^

uoi 9 1970

>7 •'7 3

w

VoL. 41

Annals of Carnegie Museum

Art. 4

Volume 41

Annals of Carnegie Museum

Article 4

!

A SYSTEMATIC REVIEW OF AMEIVA AUBERI COCTEAU
(REPTILIA, TEIIDAE) IN CUBA AND THE BAHAMAS

1. THE CUBAN SUBSPECIES

Albert Schwartz^

Research Associate, Section of Amphibians and Reptiles
Carnegie Museum, Pittsburgh, Pennsylvania

The ground lizard Ameiva auberi Cocteau, of Cuba and the Isla de
Pinos, was described in 1838 or 1839 (Smith and Grant, 1958)^ in Ra-
mon de la Sagra's monumental ''Historia fisica, politica y natural de la
Isla de Cuba.” Until Hecht (1954) considered the Bahamian Ameiva
thoracica Cope as conspecific with the Cuban species, the latter was
regarded as monotypic. The last revisers of the genus Ameiva (Bar-
bour and Noble, 1915) had little material of A. auberi (only fifteen
specimens from widely scattered localitites on both Cuba and the Isla
de Pinos, in no case more than five specimens from any single locality )
so that it is not surprising that intraisland variation in A. auberi went
unnoted. In fact, other than the recent collections made by myself and
parties between 1957 and 1960, there is remarkably little material of
A. auberi in American museums. It is the purpose of the present paper
to discuss the variation in A. auberi and to comment on the history and
affinities of the species.

I have had available for study 668 Ameiva auberi specimens col-
lected by myself and assistants in Cuba and the Isla de Pinos under
the sponsorship of National Science Foundation grants G-3865 and
G-6252. These specimens have all been deposited in the American
Museum of Natural History (AMNH). I have also studied lizards in
the Carnegie Museum (CM); the Museum of Natural History, Univer-
sity of Kansas (KU); the Museum of Comparative Zoology, (MCZ);
the Museum National d’Historie Naturelle (MNHN); the University of
Florida/Florida State Museum (UF); the University of Illinois Muse-
um of Natural History (UIMNH); the Museum of Zoology, University

1. Miami-Dade Junior College, Miami, Florida 33167.

2. Bibliography follows section III.

Issued September 18, 1970

45

SMITHSONIA'

46

Annals of Carnegie Museum

VoL. 41

of Michigan (UMMZ); and the United States National Museum
(USNM). For the loans of these lizards I wish to thank Charles M.
Bogert and George W. Foley, Neil D. Richmond and Clarence J.
McCoy, Jr., William E. Duellman, Ernest E. Williams, Jean Gnibe,
Walter Auffenberg and Lewis D. Ober, Hobart M. Smith, the late
Norman E. Hartweg, Charles F. Walker and Dale Hoyt, the late Doris
M. Cochran and James A. Peters. The large quantity of fresh material
available to me has been invaluable. These new and carefully docu-
mented specimens have aceumulated through the efforts of several
eapable and interested field companions, namely: Edwin B. Eriekson,
John R. Feick, William H. Gehrmann, Jr., Robert S. Howard, Ronald
F. Klinikowski, David C. Leber, Dennis R. Paulson, Peter F. Pepe,
Donald R. Price' Frank C. Sentz, Jr., James D. Smallwood, Barton L.
Smith, Willard M. Stitzell, James R. Talada, Richard Thomas, and
George R. Zug. Of these men, Messrs. Klinikowski and Leber were
recipients of National Seienee Foundation Undergraduate Research
Participation grants in 1959 and 1960. I wish especially to acknowledge
the help of Armando Gareia and Porfirio Azeuy in western Cuba. Spee-
imens of A. aiiheri from eastern Cuba colleeted by Major Chapman
Grant and in the UIMNH were under study by Dr. Hobart M. Smith.
He relinquished his interest in these lizards, and for this eourtesy 1 am
very grateful.

In addition to examining specimens in Ameriean eollections, 1 have
been able, through the courtesy and cooperation of Hector Sague D.
and Orlando H. Garrido, to examine the excellent collections of A.
auheri made throughout Cuba by Sr. Garrido for the Institute de
Biologia, Academia de Ciencias de Guba ( IB ) . Also, Sr. Garrido made
available to me his notes on color and pattern, as well as eolor slides of
many of the speeimens he had collected, and they have proved inval-
uable. Through the kindness of Mario S. Buide I have been able to
study a fine series of Ameiva from southern Las Villas Province. Some
of these specimens are in Sr. Buide’s collection (MSB) and the remain-
der are in the Albert Schwartz Field Series (ASFS) and in the Na-
tional Musuem of Canada (NMC). Some lizards, originally in the IB
collection, have also been deposited in the ASFS, the CM, and the
NMC. A total of 1067 specimens of Ameiva auheri is available to me.

Scale eounts were taken by William H. Gehrmann, Jr. on several lots
of material collected in 1957 and 1958. Indeed the present paper at its
inception was to be a joint effort between Mr. Gehrmann and myself,
but other duties prevented him from continuing with the study. I am

1970

Ameiva auberi: I. The Cuban Subspecies

47

grateful for his early assistance in compiling data and for allowing me
to use his counts.

Ameiva auberi is widespread on Cuba and the Isla de Pinos. The
lizard occurs also on many of the offshore keys, including the Laberinto
de las Doce Leguas off the south coast of Camagiiey Province and the
Archipielago de los Canarreos off Punta del Este of the Isla de Pinos.
A. auberi is not purely coastal in distribution, occurring in the interior
of all Cuban provinces and throughout the the Isla de Pinos. It does not
occupy areas that are either extremely mesic or moderately to extremely
high. Substrates are usually sandy or rocky ( diente de perro limestone
for example), but gravelly soils (like the Malpais gravel on the Isla de
Pinos or the gravelly serpentine savannas of Camagiiey Province) are
occupied in some regions. In the areas near Soledad, Las Villas Prov-
ince, and near San Miguel de los Banos in Matanzas Province, A.
auberi occurs in old abandoned earthen fields that have none of the
characters specified above.

Nowhere is A. auberi known to occur in the mountains except in the
Sierra de los Organos-Sierra del Rosario massif in Pinar del Rio Prov-
ince, unless the occurrence of the species near Palma Soriano in Oriente
Province is interpreted as being within the Sierra Maestra ( if so inter-
preted, the elevation is low — about 1000 feet — and hardly to be con-
strued as within the high uplands of this range), and at La Poa in the
Cuchillas de Toa. In the Sierra de los Organos, the lizards occur in cul-
tivated areas and in rather arid upland ( 1000 feet and slightly higher )
woods on well-drained and gravelly soils. In this region, A. auberi is
not associated with the luxuriant forests on the mogote sides and sum-
mits, but rather with valley floors and drier woods.

Certainly the habitats occupied by the largest number of A. auberi
are coastal, including offshore cays and islands. In such areas, the liz-
ards occur in beach growth of seagrape and coconuts, in coastal scrub,
and xeric coastal woodlands. On some of the cays A. auberi occurs on
open and hard-packed sandy soils which, in the heat of midsummer,
become uncomfortably hot for the collector at midday.

Inspection of the map (Fig. 1) shows that although A. auberi occurs
virtually from one end of Cuba to the other, there are large areas where
these lizards have not been collected. Much of the area whence A.
auberi is unknown is forested uplands, but some of it is also apparently
suitable for occupation by these lizards. The hiatuses may well be the
result of lack of collecting in some instances, but in others the lizards
simply appear to be absent. For instance, despite several weeks collect-

48

Annals of Carnegie Museum

VoL. 41

ing over the years in southern and central Camagiiey Province, no
Ameiva aiiberi were secured in apparently suitable situations there.
Visits to the head of the Bahia de Cochinos in Las Villas Province re-
vealed no lizards in an area of open beach and dry lowland woods —
habitats which in southern Oriente would have been aswarm with
Ameiva. Such incidents, the list of which might be much expanded, in-
dicate that the apparently spotty distribution of A. aiil)eri may well be
more real than a glance at the map reveals.

I have collected or studied as-yet-unpreserved specimens of all sub-
species except three, and have extensive notes on coloration and pat-
tern in life. Such information is absolutely esssential when studying liz-
ards whose coloration is fugitive in preservative and whose pattern
may be obliterated by prolonged periods in formalin or alcohol. Bor-
rowed material in most cases has served merely to confirm the exist-
ence of these lizards in the general regions where I have taken them,
and to confirm as well the differences between subspecies in life. The
excellent recent collections in the Institute de Biologia, Academia de
Ciencias, La Habana, made by Sr. Garrido, have revealed the presence
of Ameiva in three other areas. In only one instance ( a single example
from the Sierra de Jatibonico, Las \hllas Province) have I examined an
old specimen from an area which is not represented in fresh collec-
tions, and of course the treatment of this one individual has been ex-
tremely tentative. The racial status of A. auheri from two other areas
(extreme northeastern Pinar del Rio Province; north coast of Camagiiey
Province north of the Sierra de Cubitas), despite fresh material from
both areas, has been left in abeyance. These cases will be discussed in
the text.

Nomenclatorial History

Smith and Grant (1958) discussed the authorship of the names pro-
work was published in both Spanish and French editions. The Spanish
posed by Cocteau and Bibron in de la Sagra’s “Historia . . . de Cuba.” The
edition, which was the earlier published, in twelve folio volumes, con-
tains the description of Ameiva auheri in volume 4. Volume 4 of both
Spanish and French editions was also published in two parts. From the
evidence presented by Smith and Grant, part I of volume 4 of the Span-
ish edition antedates its homologue in the French edition, the former
appearing in 1838 or 1839, and die latter by 1841. As far as authorship
is concerned. Smith and Grant ( p. 216 ) stated that Cocteau was the sole
author of all material before page 120 of part I of volume 4 of both edi-
tions, and that all names after page 120 are to be attributed jointly to

1970

Ameiva auberi: I. The Cuban Subspecies

49

Cocteau and Bibron. Since the original description of Ameiva auberi is
on page 51 of the ( earlier ) Spanish edition, the name is presumably to be
attributed solely to Cocteau. Additional evidence that A. auberi was de-
scribed only by Cocteau is that the name is followed, at its original place
of proposal, by the name of Cocteau only, whereas later names (i.e.,
Epierates angulifer and Dromicus angulifer) are followed by “nobis,”
thereby suggesting joint authorship. Although the evidenee is not com-
pletely incontrovertible, it seems appropriate to eonsider that Coeteau
was the author of Ameiva auberi.

One other name has been proposed for Cuban Ameiva: Ameiva tri-
lineata Gray, 1845. This name was based upon four syntypes in the
British Museum (Natural History) colleeted by W. S. Maeleay. Pertinent
color data included in the original description are that the species is
“olive, with three pale streaks,” and that the tail has a “white streak on each
side.” Gray, although also listing Ameiva auberi Cocteau, had obviously
not examined material of the latter form and was surely unaware that
his A. trilineata demonstrated a pattern that is commonplace in juve-
niles and subadults of several populations of A. auberi, and that also per-
sists into the adults of some races. Maeleay {fide Conde, 1958b: 261) lived
in Cuba from 1827 to 1836, residing at Guanabacoa to the east of La
Habana in Habana Provinee. Presumably the syntypes of A. trilineata
were secured in the region near Guanabacoa. In this region at least
juvenile and subadult lizards have a trilineate pattern. Consequently
I regard Ameiva trilineata Gray a strict synonym of Ameiva auberi
Cocteau.

The monotypic status of Amiva auberi was maintained until Hecht
(1954:133), in discussing the Bahamian fauna, used the eombination
Ameiva atd)eri thoraciea Cope for the ground lizards of these islands.
The similarities between A. auberi and A. thoraciea are obvious upon
casual inspection; in fact, Sutcliffe (1952:5) identified as A. thoraciea a
ground lizard from Cayo Largo, of the Archipielago de los Canarreos off
the eastern tip of the Isla de Pinos. This identification was made on the
basis of agreement in lepidosis and coloration of the Cayo I.argo speci-
men with Barbour and Noble’s diagnosis of A. thoraciea, in eontrast to
their diagnosis of A. auberi. Although 1 have not examined this partic-
ular individual, a series of these lizards from Cayo Largo shows that
they indeed represent a distinetive subspecies of A. auberi. The point
here, however, is that auberi and thoraciea are very close in both pattern
and sealation details, and I have no doubt that they are eorreetly consid-
ered as conspecific. In the present paper I have made no effort to com-

50

Annals of Carnegie Museum

VoL. 41

pare the Cuban and Bahamian subspecies of A. auberi since the Baha-
mian lizards are under study by Clarence J. McCoy (see part II of this
article). I have examined fairly extensive material from various Baha-
mian islands and am aware of the variation in both pattern and scalation
of these lizards in the Bahamas.

Cocteau (1838 or 1839:54) discussed the coloration of the material
before him at the time of the description of A. auberi. It is obvious that
he had more than a single specimen since he contrasted the juvenile with
the adult coloration. The former was stated as being a deep-colored
green above with five (median dorsal, two dorsolateral, two lateral)
whitish or yellowish lines. The lateral fields (between the dorsolater-
al and lateral pale lines ) are dark brown. In adults, the lines disappear
and the general dorsal coloration becomes a more or less dull ashy
green. The coloration of the sides is changed into lenticular black spots,
which form a not particularly clear, undulating black longitudinal band
(the lateral fields). The ventral surfaces in the adult are greenish yellow.
The figure of Ameiva auberi in the Atlas of the de la Sagra work shows a
dorsally unlined lizard with the back grayish green, lateral fields bounded
above and below by dirty white lines, scattered large black blotches
in a grayish tan lateral field, a gray belly, and a greenish-gray throat.
No population of A. auberi with which I am familiar is characterized
by the above combination of color and pattern.

Since it is mandatory to restrict the type locality of A. aul)eri Cocteau,
I have corresponded with Jean Guibe of the Museum National d’His-
toire Naturelle, in which it might be assumed the original material of
A. auberi is deposited. In response to my query. Dr. Guibe replied that
there are four specimens (MNHN 1112, 2647, 1788, and 4178) that
came from Guba and were given to the collection by Ramon de la Sagra.
These four lizards are considered in the eatalogue to be the syntypes of
A. auberi. However, none of them corresponds with the measurements
given by Gocteau (1838 or 1839:51). Cocteau’s most pertinent mea-
surements are a total length of 260 mm. and a tail of 170 mm., giving a
snout-vent length of 90 mm.

Dr. Cuibe further stated that MNHN 4178 is probably the specimen
later described in detail by Dumeril and Bibron (1839:111 et seq.),
but as noted above, it does not agree with the data given by Coeteau.

The problem thus resolves itself into the precise status of these four
specimens. The evidence for their being syntypes seems ineonclusive.
Since none of the four lizards agrees in dimensions or in pattern with
Coeteau’s detailed data, plate, and description, and since often in that

1970

Ameiva auberi: I. The Cuban Subspecies

51

period of zoological research the ultimate importance of typical material
was unrecognized — with the result that the designation of these four
lizards as syntypes may have been made much later than the appearance
of both the de la Sagra and Dumeril and Bibron works — I have disre-
garded these four lizards (of which I have examined MNHN 1788 and
2647) as indicators of the provenance of the specimens on which the
name A. auberi was based.

The situation is thus anomalous. The “syntypes” themselves do not
agree with the details as given by Cocteau, and the Cocteau description
and plate do not agree with any known population of A. auberi. How-
ever, the following comments are pertinent.

Cocteau (1838 or 1839:51) stated that Ameiva auberi was named
for Don Pedro Auber of the Jardin botanico de la Habana. Since these
lizards even now are moderately abundant in the La Habana region, it
would seem not improbable that Auber collected the original specimens
in this general area. The two Paris specimens that 1 have examined do
not agree in any way with La Habana material but do agree in scale
counts, coloration, and pattern with recent material from the southern
third of the Isla de Pinos. Conde (1958a:78) stated that among many
other contributors to de la Sagra’s work was A. H. Lanier (the French
consul at Trinidad, Cuba), who explored the Isla de Pinos in 1831 (Jen-
nings, 1917:22). Since Lanier made a collection of plants for de la Sagra
on the Isla de Pinos it is not improbable that he may have collected a
few other natural-history specimens there also. Thus, the two Paris
specimens I examined may well have come from the Lanier collections
from Isla de Pinos, but there is no assurance that these two (or the
remaining two ) were before Cocteau when he named Ameiva auberi.

Everything considered, I have placed most emphasis on two facts
stated by Cocteau : ( 1 ) that the species was named for Auber, who re-
sided in La Habana, and ( 2 ) that no mention was made of a black throat
on any of the syntypes, especially the large individual for which meas-
ments and detailed description were given. Since the recently taken ma-
erial from Habana Province lacks a black throat ( a very striking feature
where it occurs ) , in at least this one feature the fresh material resembles
Cocteau’s description. I therefore restrict the type locality of Ameiva
auberi to the vicinity of La Habana, Habana Province, Cuba. If, in the
future, the status of the supposed syntypes can somehow be established
without doubt, and they ( or some of them ) are considered to have orig-
inated on the southern portion of the Isla de Pinos, a nomenclatorial shift
alone will be required. The northern coastal population in Habana Pro-

52

Annals of Carnegie Museum

VoL. 41

vince will require a new name, and the name I have proposed in the pres-
ent paper for the southern Isla de Pinos population will become A. a.
auberi.

With the above restriction, the population of Ameiva auberi in north-
ern Habana and Matanzas provinces becomes the nominate race. Al-
though heretofore I have begun the systematic treatment of Ameiva
( from Hispaniola ) with a discussion of the nominate form, I depart from
this more logical procedure here. It seems preferable in this case to be-
gin the discussions of the various subspecies with that of the race at the
extreme western end of the island and proceed eastward. Such a method
permits the placement of A. a. auberi in its proper geographic position
relative to the subspecies to the east and west.

Ameiva auberi may be defined as follows (data from both Cuban
and Bahamian populations combined):

Characteristics and Variation

( 1 ) a moderate-to-large species of the genus Ameiva with snout-vent length
to 136 mm. in males and 115 mm. in females; (2) dorsal caudal scales keeled and
straight; ( 3 ) ventrals in 8 to 14 transverse rows ( although the modal count
is overwhelmingly 10, and counts of 8, 9, 11, 12, and 14 occur rarely and never
characteristically in any subspecies) and in 28 to 36 longitudinal rows; (4)
fourth toe subdigital scales from 61 to 94 — combined counts for both fourth toes;
(5) femoral pores 20 to 35 — both series combined; (6) fifteenth caudal verticil
with 22 to 37 scales; (7) dorsal pattern variable but consisting of either (a)
a median and two dorsolateral pale lines enclosing between them the dorsal fields,
which are some shade of brown, varying from pale-sandy or tannish gray to brown
or rich reddish brown, or ( b ) more or less unicolor tans to browns dorsally
without (or only faintly indicated) median and dorsolateral lines, and having
the lateral fields black or brown, with varying amounts and patterns of red,
brick-red, tan, or brown included in the lateral fields; (8) throat varying from
cream or yellow to solid black, the black often extending posteriorly onto the
abdomen for various distances; and (9) hemipenis extending to about the sixth
to eighth caudal verticil, sulcate surface naked, sulcus bifurcating apically, the
branches ending in two poorly defined apical discs, non-sulcate surface entirely
flounced, the flounces extending to the margins of the sulcate surface, a small
smooth triangular area on the non-sulcate side dividing the flounces for about
one third the length of the organ into two fields of flounces which correspond
to the apical discs.

In the following subspecies accounts, data have been given on size,
rows of transverse and longitudinal ventrals, total number of fourth-toe
scales, total number of femoral pores, scales in the fifteenth caudal ver-
ticil, and coloration and pattern. Emphasis has been placed on the last
categories, and on size ( snout-vent length ) . The meristic data from scale
counts, although showing means which are often statistically significant
inter se (see Table 4), offer no sure way of identifying any individual

1970

Ameiva auberi: I. The Cuban Subspecies

53

Specimen, but on the other hand offer differences between various pop-
ulations. It should be recognized from the beginning of the accounts that
such rankings as “high,” ‘moderate,” and “low” in regard to fourth-toe
scales, femoral pores, and fifteenth verticil scales are arbitrary rankings
according to the means of these counts. However, such scale characters
are an additional way of quantifying the interpopulation differences and
are moderately useful in defining subspecies.

The remarkable consistency in counts of longitudinal and transverse
rows of ventrals renders them of no value in differentiating the subspe-
cies. Despite transverse counts of 8, 9, 11, 12, and 14, the modal con-
dition in all populations is very strongly 10 (see, however, the account
of the subspecies from the northern Las Villas caijeria) . The modal
number of longitudinal rows of ventrals in all Cuban subspecies varies
from 31 to 34. The modal condition in any population is not significant
(with the exception that only one Cuban population has a mode of 31
and only two have modes of 34 ) , since the range of variation in any giv-
en sample almost matches the entire range for this character ( 28 to 36 )
in all Cuban specimens studied. For example, in a sample of 85 speci-
mens from between the Bahia de Santiago and the Bahia de Guantana-
mo, the range of longitudinal rows of ventrals is 30 to 36, and the sample
of 57 specimens from the northern portion of the Isla de Pinos has longi-
tudinal counts of from 29 to 34. In the former instance, the mode is 32,
in the latter 31. It is obvious that this count is hardly valuable for diag-
noses, but data are nevertheless presented in each subspecies account.

Attention should be directed to the throat coloration. Ten Cuban pop-
ulations from ( at times ) widely separated regions are characterized by
having black throats. In most cases, the black throat and chest color are
fairly constant within a sample from a particular region. Males have the
throat and chest black ( although in two of the black-throated popula-
tions — near Trinidad in Las Villas, and in northern Matanzas — the
throat is not completely black but has a black median blotch ) . Although
it would be reassuring to be able in the case of these black-throated sub-
species to state that “all adult males above a snout-vent length of such-
and-such have black throats,” such a statement is seldom completely
true. In most cases some large and obviously adult males lack any indi-
cation of a black throat and chest, and this character may also appear
randomly in smaller males far below the maximally sized individuals.
Very rarely do females of these black-throated races have the throats
I as completely black as do the black- throated males. Throats of females
resemble in intensity of pigment those of adult males that are just acquir-

54

Annals of Carnegie Museum

VoL. 41

ing the fully black throat and chest. Small subadults and juveniles of
both sexes lack any indication of black throats. The details of the con-
dition of black throats in each sex in those subspecies that are black-
throated are discussed individually in each account. The above qualifica-
tions should be understood in advance.

In several populations ( northern Oriente coast at Gibara, Cayo Can-
tiles in the Archipielago de los Canarreos, Cayuelo de la Vela in the
Archipielago de Sabana off the northern coast of Las Villas) there are
some peculiarities of throat pigmentation: at Gibara, there apparently
occurs a non-black-throated population of a subspecies that elsewhere
along this coast is black-throated, whereas in the two island chains of
the Archipielago de los Canarreos and Sabana, individual islands have
black-throated populations or occasional black-throated lizards in oth-
erwise non-black-throated areas. My treatment of these three anoma-
lous instances has been arbitrary, primarily because in two cases (Gi-
bara, Cayo Cantiles) I have seen little or no living material. In these
two instances I have included the peculiarly colored lizards with their
adjacent relatives, but this treatment is not necessarily correct. In the
third case (Cayuelo de la Vela), I have chosen to distinguish the pop-
ulation there nomenclatorially. In each instance I have given my rea-
sons for the course I have pursued. I have not mentioned here the sit-
uation on the southern portion of the Isla de Pinos, wherein the popula-
tion may or may not have black throats, although there, if the throat is
not black, that color is not replaced by red or orange, but is merely pale
blue. Details of the southern Isla de Pinos lizards will be discussed later.

Aside from the anomalous populations mentioned above, all those
subspecies that have the throat colored (in contrast to black) have it
so regularly, without exception (subadults and females may have the
throats somewhat paler than adult males ) . Thus, in over 60 specimens
of the widespread race in Pinar del Rio, the throat in all is uniformly
yellow, and no specimen shows any indication of black pigment. From
these circumstances, I deduce that some populations have the potential
for ontogenetic development of black throats, whereas others lack it
completely. This basic assumption has brought about the description of
colored- and black-throated races that at times closely approximate
each other geographically, and that at times also have relatively re-
stricted distributions. Comments will be made in the appropriate
places on such situations.

Recent workers on the related teiid genus Cnemidophorus have
employed the number of dorsal granules at midbody as a character

1970

Ameiva auberi; I. The Cuban Subspecies

55

useful for differentiation of species, and, to a more limited degree, of
subspecies. Although in my previous studies on West Indian Ameiva,
I have not used this count, in an effort possibly to give added data on
subspecific differentiation in A. auberi, a set of pilot counts was taken
on specimens from the two exti'emes of Cuba — Cabo de San Antonio
and east of Guantanamo — plus specimens from the southern coast of
Pinar del Rio. The midbody granules vary from 89 to 112 in the lot
counted. The series from Cabo San Antonio and east of Guantanamo
have virtually identical extremes (91 to 112), whereas the lower counts
(89 to 98) are from southern Pinar del Rio. These data suggest that, as
in Cnemidophortis, some subspecies of A. auberi might be partially de-
finable on the basis of number of dorsal granules, but just as in other
counts in this species, the dorsal granule differences are by no means
absolute.

Maximum size of the various subspecies of A. aul)eri shows some
striking differences. Sometimes there has been comment on the validity
of using this character to differentiate races of teiid lizards, the premise
being that the collector may simply not encounter fully adult speci-
mens in the course of his activities. 1 feel strongly that such an assump-
tion is incorrect. The very large size of specimens of A. aid)eri from
Pinar del Rio Province, for instance, whence long series are available,
stands in contrast to the much smaller maximum size of a large number
of specimens from southern Oriente between the Bahia de Santiago
and the Bahia de Guantanamo. In the former region, males reach a
snout-vent length of 136 mm. ( with many males with snout-vent lengths
in excess of 100 mm.), whereas in the latter region no male reaches a
snout- vent length greater than 78 mm. Such differences hardly seem
attributable to artifacts of collecting and sampling. To me the differ-
ences represent inherent genetic differences in size (and bulk) of the
members of the populations involved. In none of the five large sub-
species (one in Pinar del Rio, two in northern Las Villas, two in south-
ern Las Villas) exceeding 100 mm. snout-vent length in maximum
size, are males represented by a single unusually large or “giant” male
and a larger suite of much smaller males. In all cases there is graded
series from small to very large males, with the largest specimen merely
the end of a long and evenly graded series of males of increasing snout-
vent lengths. From these observations, I consider that size is a perfectly
valid character for differentiation of subspecies in Ameiva.

56

Annals of Carnegie Museum

VoL. 41

Systematic Review

Ameiva auberi cacuminis, new subspecies

holotype: AMNH 83028, an adult male, from Ensenada de Cajon, Pinar del
Rio Province, Cuba, taken December 22, 1958, one of a series collected by Ar-
mando Garcia, Ronald F. Klinikowski, Donald R. Price, Frank C. Sentz, Jr., and
Albert Schwartz. Original number 7072.

PARATYPES (all from Pinar del Rio Province, Cuba): AMNH 96311-96320, same
data as holotype; USNM 51790, Cabo de San Antonio, May 23, 1914, J. B. Hen-
derson and P. Bartsch; USNM 51795, Cabo de San Antonio, May 23, 1914,
J. B. Henderson and P. Bartsch; USNM 51851-51852, Cabo de San Antonio, May
26, 1914, J. B. Henderson and P. Bartsch; MCZ 10868, Cabo de San Antonio,
April, 1915, V. D. Rodriguez; IB 135, La Tumba, January 19, 1966, O. H.
Garrido.

diagnosis: A Cuban subspecies of A. aul)eri characterized by a combination of
small size (males to 80 mm., females to 60 mm. snout-vent length), moderate
number of fourth-toe subdigital scales, femoral pores, and scales in the fifteenth
caudal verticil; dorsum brown with an inconspicuous median yellow line in adults,
brown with prominent lemon-yellow line in subadults and juveniles, lateral fields
black with irregular red edges; and throat dusky gray.

DISTRIBUTION: Kuown only from the extreme tip of Cabo de San Antonio, Pinar
del Rio Province, Cuba; all distributions shown in Fig. 1.

DESCRIPTION OF HOLOTYPE: All adult male with the following measurements and
counts: snout-vent length 72 mm., tail 167 mm.; ventrals in 32 longitudinal and
10 transverse rows; fourth-toe subdigital scales 41 and 42 (total 83); femoral pores
14 and 14 ( total 28 ) ; 27 scales in the fifteenth caudal verticil. Dorsal ground color
brown in life, with the median and two dorsolateral lines inconspicuously yellow;
lateral fields vaguely outlined above by the dorsolateral yellowish lines, solid black
with irregular red edges. Throat dusky gray; venter dull grayish blue. Tail grayish
blue above, slightly brighter below, with ventrolateral white stripe moderately con-
spicuous along the basal third of tail. Limbs brown above, with some very vague
darker marbling.

VARIATION: The series of 17 specimens of A. a. cacuminis has the following
counts: longitudinal ventrals 30-33 (mode 32); rows of transverse ventrals 10 in
all specimens; fourth-toe seales 66-83 (mean 75.1); femoral pores 24-29 (mean
25.6); fifteenth verticil 24-29 (mean 26.4). The largest male has a snout-vent
length of 80 mm., the largest female (the only adult female in the series) 60 mm.

The longest series of paratypes was collected by myself and party in
1958. In life, these specimens were recorded as being brown dorsally,
with the median dorsal line lemon yellow in juveniles and subadults,
and less conspicuously yellow in adults. In all there is no indication
of a black throat, the throats being dusky gray. The lateral fields are
solid black without any interior red dotting; red pigment occurs only
along the edges of the lateral fields as a series of ragged and irregular
scallops or blotches. The USNM, MCZ, and IB paratypes (of which
USNM 51851 is the largest male and USNM 51795 is the largest female)

1970

Ameiva auberi: I. The Cuban Subspecies

57

agree well with the fresh material, except that one specimen (USNM
51851 ) has a few isolated red dots in the lateral fields, and the edges of
the fields are somewhat more extensively red than in all other specimens.
No lizard has any indication of black on the throat or chest.

REMARKS: The series from the type locality was collected in xeric for-
est along the coast of the Ensenada de Cajon. Few adults were seen,
although juveniles were quite abundant in the woods and along the
beach front.

Ameiva auberi denticola, new subspecies

holotype: AMNH 79202, an adult male, from north shore, Ensenada de
Corrientes, Pinar del Rio Province, Cuba, taken July 10, 1957, one of a series col-
lected by John R. Feick, Armando Garcia, and William H. Gehrmann. Original
number 2618.

paratypes (all from Pinar del Rio Province, Cuba): AMNH 79200-79201,
79203-79231, same data as holotype; AMNH 78391, 79189-79199, same locality as
holotype, July 5, 1957, J. R. Feick, A. Garcia, W. H. Gehrmann; MCZ 55552,
Ensenada de Corrientes, July 4, 1956, K. F. Koopman; AMNH 83026 ( 13 speci-
mens), north shore, Ensenada de Corrientes, about 40 km. W. Cayuco, August 19,
1958, A. Garcia, A. Schwartz, G. R. Zug; AMNH 83027 (7 specimens), north
shore, Ensenada de Corrientes, about 45 km. W. Cayuco, August 19, 1958, A.
Garcia, A. Schwartz, G. R. Zug; AMNH 79187-79188, western coast, Cabo Corri-
entes, July 5, 1957, J. R. Feick, W. H. Gehrmann; AMNH 79232, western coast,
Cabo Corrientes, July 10, 1957, A Garcia,

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
large size (males to 96 mm., females to 78 mm. snout-vent length), moderate
number of fourth-toe subdigital scales, high number of femoral pores and scales
in the fifteenth caudal verticil; dorsum olive-tan, unicolor in adults and with the
median and dorsolateral lines yellowish in juveniles and subadults, lateral fields
variable but most often with the black extremely restricted to a series of discrete
blotches in the lateral field area and with consequent extensive reddish areas along
the margins of the fields; and throat and chest black.

DISTRIBUTION: The Peninsula de Guanahacabibes, from the western shore of
Cabo Corrientes west to about 45 km. west of Cayuco, Pinar del Rio Province,
Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 96 mm., tail 187 mm., tip regenerated; ventrals
in 32 longitudinal and 10 transverse rows; fourth-toe subdigital scales 39 and 38
(total 77); femoral pores 15 and 16 (total 31); 29 scales in the fifteenth caudal
verticil. Dorsum olivaceous tan, with no indications of median or dorsolateral lines;
lateral fields much reduced, consisting of a linear series of small and rather
disjunct black blotches bordered above and below with reddish. Lower sides
olivaceous, somewhat mottled with greenish and not set off from the lateral
fields by pale lateral lines. Throat and chest black, the black extending to about
the level of the fourteenth row of ventrals, although more diluted, and restricted
to the margins of the ventrals posteriorly; venter bright blue laterally, lighter and
grayer centrally. Limbs concolor with dorsum on upper sides, forelimbs dark

58

Annals of Carnegie Museum

VoL. 41

blue-gray ventrally, hindlimbs blue ventrally. Tail olivaceous blue above, purer
blue below; ventrolateral pale caudal stripe absent.

variation: The series of 59 specimens of A. a. denticola has the following
counts: longitudinal ventrals 31-34 (mode 32 or 33); rows of transverse ventrals
10, 12 (2 specimens), or 14 [(1 specimen) (mode 10)]; fourth-toe scales 67-86
(mean 76.4); femoral pores 23-31 (mean 27.1); fifteenth verticil 24-33 (mean
28.0). The largest male has a snout-vent length of 96 mm. and the largest female
78 mm.

The long series of adult paratypes agrees with the description of the
holotype in all details except the condition of the lateral fields. Although
the modal condition is as described for the holotype — a series of dis-
junct black blotches bordered above and below by dull reddish areas
— at times the lateral fields may be more complete and have a discrete
reddish border both above and below, enclosing a more or less irregular
and sinuous black “line” from the neck to the groin. Occasional speci-
mens have the lateral fields solid black, or solid black with a very few
isolated reddish dots. The ventrolateral caudal stripe is absent in adults.

Subadults and juveniles are lineate dorsally, the lines yellowish and
not especially contrasting to the olive ground color. The lineate condition
is pronounced only in juveniles, and by the time a snout-vent length of
about 70 mm. is reached, most subadults have the dorsum unicolor and
patternless like that of the much larger adults.

Since A. a. denticola is a black-throated race, special comment is re-
quired. All adult and subadult specimens (37 males and 21 females)
were graded into three categories: ( 1 ) throat and chest completely black;
(2) throat and chest only partially black; and (3) throat and chest
without black. All specimens (15) in category 1 are males, ranging in
snout-vent lengths from 64 mm. to 96 mm. (thus including the largest
male). Category 2 includes 12 males and 9 females. The males range
in snout-vent length from 70 mm. to 93 mm. (and are thus bracketed
in size by the males of category 1), and the females from 56 mm. to
75 mm. (thus not including the largest female). Category 3 includes 11
males with snout-vent lengths from 66 mm. to 90 mm., and 12 females
from 58 mm. to 78 mm. (thus including the largest female). From these
data it is obvious that most males (27 of 37) have some black on the
throat and chest, but that black pigment is not restricted to large speci-
mens. In females, only nine of 21 specimens have black on the throat and
chest, and in denticola the black pigment in females is never as extensive
as it is in males. These peculiarities are difficult to analyze. All speci-
mens are summer-collected, so it is possible that here (and in the
other black-throated races) the black-throated coloration is somehow

1970

Ameiva auberi; I, The Cuban Subspecies

59

correlated with season (see comments below) or with sexual aetivity.
If the latter is the ease, it seems strange that a small male with a
snout-vent length of only 64 mm. would be sexually mature. My sug-
gestion is that the blaek ehest and throat may appear in males of the
black-throated subspeeies at any size onee the early juvenile condition
is past. Thus, some individuals become black-throated at a small size
and others achieve this state only with a much larger size. This im-
plies that males without a black chest and throat at a larger size (for
instanee, 90 mm., the largest non-black-throated male in the present
lot of specimens) have still to reaeh this eondition.

The situation in females is comparable, although in denticola there
is no evidenee that females are ever completely black-throated and
blaek-ehested. The fact that the largest female of the series lacks any
black pigment suggests that this lizard has still not reached that con-
dition but would ultimately have done so.

In arguing against the premise that possession of black throats is
a seasonal phenomenon, I note here that in other black-throated races
where speeimens have been eolleeted in both winter and summer,
there is no difference in frequency of black-throated lizards by season.

COMPARISONS: The raees cacuminis and denticola can be distin-
guished by three features: (1) the former is brown dorsally, the lat-
ter olivaeeous tan; (2) the former is gray-throated, the latter black-
throated and black-ehested; (3) the former has a solid black lateral
field, the latter has the lateral field usually very much reduced and
fragmented.

Of all the raees of A. auheri, only denticola has sueh a wide varia-
tion of number of transverse rows of ventrals, although the modal
eondition of 10 rows is strong. Only two lizards (AMNH 79215 and
79221) have eounts of 12 rows, and one (AMNH 79219) has a
eount of 14. These variants are purely easual and have no taxonomic
significance.

remarks: The area whence most of the specimens of A. a. denti-
cola were secured is a xeric region of sandy soils and diente de perro
I ( dog’s tooth — hence the name denticola ) eroded limestone on the
' Peninsula de Guanahaeabibes. The large number of specimens of
I denticola attests to the abundance of ground lizards in this particu-

I larly favorable habitat. The series from 40 km. west of Cayuco was

j taken in monte (lowland broadleaf forest) in a more mesie situation

woods oceupied by cacuminis at the tip of the peninsula.

60

Annals of Carnegie Museum

VoL. 41

The differentiation of A. auberi into two subspecies on the Penin-
sula de Guanahacabibes is difficult to explain, but the same pheno-
menon will be observed later on the much shorter and more attenuate
Peninsula de Hicacos in Matanzas Province. Actually, all records for
denticola are on the southern coast about the Ensenada de Corrientes,
whereas the records for cacuminis are on the northern coast. In 1960,
David Leber and I collected between Cayuco and the Valle de San
Juan in apparently suitable habitats. The road in this region is about
equidistant from the north and south shores of the peninsula. No
Ameiva were seen. It is possible that cacuminis and denticola repre-
sent north- and south-^hore subspecies on the peninsula, rather than
basal and distal subspecies. The lack of specimens from the more cen-
tral portions ( latitudinally ) of the Peninsula or the more distal re-
gion (longitudinally) precludes any definite resolution of the problem.

Certainly the Peninsula de Guanahacabibes is a significant region
zoogeographically. From it have been described one endemic species
( Anolis quadriocellifer Barbour and Ramsden ) and several endemic sub-
species of more widely ranging reptiles (Leiocephalus carinatus zaijasi
Schwartz, Leiocephalus stietigaster stictigaster Schwartz, Leiocephalus
macropus koopmani Zug, Dromicus andreae peninsulae Schwartz and
Thomas). The peninsula is generally, despite its relatively poor rainfall
of 115 to 125 cm. per year (Marrero, 1951:442), well wooded, and at
the time of my last visit in 1960, was more or less uncultivated, especially
to the west. There are several sinkhole lakes, such as that at Vallecito
de San Juan, which are surrounded by luxuriant broadleaf trees with
epiphytes and lianas. Such interior wooded regions contrast strongly
with the more bleak and arid southern coastal portion about the Ensen-
ada de Corrientes. To the east, the peninsula abuts upon the flat, lake-
studded plains of southwestern Pinar del Rio (see Marie-Victorin and
Leon, 1944:132-33 and 302-04 for typical photographs of this interest-
ing region). These plains are inhabited by the following subspecies of
A. auberi.

Ameiva auberi procer, new subspecies

holotype: AMNH 78390, an adult male, from San Vicente, Pinar del Rio Prov-
ince, Cuba, taken June 24, 1957 by Albert Schwartz. Original number 1857.

PARATYPES (all from Pinar del Rio Province, Cuba): AMNH 79153, same
locality as holotype, June 20, 1957, native collector; AMNH 79165, same locality
as holotype, June 28, 1957, native collector; AMNH 79167, same locality as holo-
type, June 30, 1957, native collector; AMNH 79168-79172, same locality as
holotype, July 5, 1957, native collector; AMNH 79173-79179, same locality as
holotype, July 6, 1957, native collector; AMNH 79180, same locality as holotype.

1970

Ameiva auberi: I. The Cuban Subspecies

61

July 8, 1957, native collector; AMNH 79181, same locality as holotype, July 9,
1957, native collector; AMNH 79182-79186, same locality as holotype, July 10,
1957, native collector; AMNH 83005 (3 specimens), same locality as holotype,
August 7, 1958, native collector; AMNH 83024 (9 specimens), same locality as
holotype, August 8, 1958, native collector; AMNH 83029 (2 specimens), same
locality as holotype, December 26, 1958, R. F. Klinikowski; AMNH 79150-79152,
mountains north of San Vicente, June 19, 1957, D, R. Paulson; AMNH 79166,
mountains north of San Vicente, June 29, 1957, native collector; AMNH 83004
(2 specimens), 0.5 mi. N. San Vicente, August 5, 1958, G. R. Zug; AMNH
76553-76554, mountains behind (N.W. ) San Vicente, December 28, 1956, W. H.
Gehrmann; MGZ 55570-55571, near ^hhales, July 9-10, 1956, K. F. Koopman.

ASSOCIATED SPECIMENS (all from Pinar del Rio Province, Cuba): AMNH 96323,
8.2 mi. W. Cayuco; AMNH 96321-96322, 6.9 mi. W. Cayuco; AMNH 83025 (2
specimens), 1 mi. N.W. Cayuco; AMNH 83944, 13 mi. S.W. Isabel Rubio; AMNH
79160-79162, 11.3 mi. S.E. Isabel Rubio; AMNH 79163-79164, 2.9 mi. E. Isabel
Rubio; AMNH 79154-79159, 7.6 mi. E. Isabel Rubio; IB 1358, San Waldo,
4 km. N. Cortes, road between Cortes and Isabel Rubio (not mapped); MCZ
10833, Guane; USNM 27368-27371, El Guama; MCZ 7938, San Diego de los
Banos; IB 141, Dayaniguas; MCZ 4388, Bahia Honda; USNM 27662-27663, 27676-
27683, Cabanas.

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
very large size (males to 136 mm., females to 115 mm. snout-vent length), low
number of fourth-toe subdigital scales, moderate number of femoral pores, high
number of scales in the fifteenth caudal verticil; dorsum brown, unicolor in adults
and subadults but juveniles showing median and dorsolateral yellow lines, lateral
fields with more black than red, the red pigment usually forming a series of dorsal
and ventral scallops along the upper and lower borders of the field with each
lower scallop about two to three times the area of each dorsal scallop; and throat
and chest yellow.

DISTRIBUTION: Pinar del Rio Province, from the vicinity of Cayuco in the south-
west, north at lower elevations (about 1000 feet) in at least the southern and
central portions of the Sierra de los Organos and Sierra del Rosario, as far east as
San Diego de los Banos and Dayaniguas on the south coast, and north on the
coast to Bahia Honda and Cabanas.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 119 mm., tail 267 mm.; ventrals in 34 longitudinal
and 10 transverse rows; fourth-toes missing; femoral pores 14 and 14 (total 28);
27 scales in the fifteenth caudal verticil. Dorsal ground color brown in life, with
no indications of median or dorsolateral lines; lateral fields prominent and bold,
consisting of a black sinuous band with both dorsal and ventral brick-red scallops,
the ventral scallops about three times the size of the dorsal scallops; ventral scal-
lops separated by ventrad extensions of black pigment and thus more or less dis-
crete. Lower sides blue, mottled with irregular and somewhat tigroid blackish
blotches; blue of sides continuous onto venter, involving chest (where blue is ad-
mixed with yellow) and all enlarged ventral scales except the paramedian rows.
Anterior faces of hindlimbs and sides of tail bluish gray, with a white ventrolateral
stripe on the proximal third of the tail; cheeks, throat, sublabials ( and chest,
where admixed with bluish gray) yellow. Limbs grayish brown much marbled

62

Annals of Carnegie Museum

VoL. 41

with black above, and pale blue-gray to yellowish below,

variation: The series of 62 specimens of A. a. procer has the following counts:
longitudinal ventrals 31-35 (mode 33); rows of transverse ventrals 8 (3 speci-
mens), 9 (1 specimen) or 10 (mode 10); fourth-toe scales 63-84 (mean 71.9);
femoral pores 21-31 (mean 26.0); fifteenth verticil 24-31 (mean 27.3). The largest
male has a snout-vent length of 136 mm., the largest female 115 mm.

The large lot of A. a. procer agrees well with the description of color-
ation and pattern of the holotype. The major variation is in the lateral
fields. The holotype represents the modal condition. Occasional indivi-
duals have the red pigment so much expanded that the black areas are
restricted to irregular blotches in an otherwise brick-red field. In all
fresh specimens, the throats (and chests, unless they are additionally
overlaid with grayish) were yellow, and no specimen shows any ten-
dency toward the black throat of denticola to the southwest. Specimens
from the vicinity of Cayuco, however, show a tendency toward having
the throats orange rather than yellow.

The unlined dorsum is a constant feature of adults. In juveniles and
males to snout-vent lengths of about 80 mm. and females to about 87
mm., the usual juvenile pattern of a median and two dorsolateral yel-
lowish lines is displayed.

COMPARISONS: A. a. procer differs from both cacuminis and denticola
in having a yellow throat ( in contrast to dusky gray and black throats ) ,
and in being much larger than either of the two more western sub-
species. In fact, procer is the largest of any Cuban subspecies of A.
auberi. The brown dorsum of procer contrasts with the olive-tan dor-
sum of denticola but is like the brown back of cacuminis.

Although as usual the modal number of transverse rows of ventrals
is 10 in procer, three specimens have counts of 8 rows, and one has an
irregular count of 9.

Remarks: A. a. procer occupies a wide variety of situations, from the
flat, grassy, lake-studded plains between Isabel Rubio and Cayuco in
the southwest to the southern slopes and interior of the Sierra de los
Organos and Sierra del Rosario to elevations of 1000 feet. In these
mountain ranges, as noted in the introduction, Ameiva occupies the
valley floors and the drier uplands, and not the heavily wooded and
mesic mogote sides and tops or associated rich forests. A. a. procer is
common in the vicinity of San Vicente, occurring there along the road-
sides and in cane fields and truck gardens. Most of the series of para-
types were collected by native boys while cutting sugar cane. In the
mountains to the north and northeast of San Vicente, the lizards are
less abundant in the rather more arid woods that occupy gravelly soils.

1970

Ameiva auberi: I. The Cuban Subspecies

63

Specimens from 7.6 miles east of Isabel Rubio were taken in riparian
growth along a small stream in a pasture, and those from 13 miles
soutliwest of Isabel Rubio were collected along the margin of a lake in
a dry pasture. The specimens from 1 mile northwest of Cayuco were
taken from a well-shaded encinal with palmetto undergrowth.

Barbour (1914:309) reported the occurrence of A. auberi at Herra-
dura and Pinar del Rio city, but no specimens are available from either
loeality. The latter record almost certainly pertains to A. a. procer,
which may also occur at Herradura, although that area might be oc-
cupied by the following subspecies.

There is no evidence of intergradation between procer and denticola,
although it is to be expected between the westermost record of procer
(8.2 mi. W. Cayuco) and the easternmost record of denticola (western
coast, Cabo Corrientes), a distance of about 16 kilometers. It is possible
that denticola occupies only the arid coastal section of the Peninsula de
Guanahacabibes, and procer the more mesic plains about Cayuco, and
that the zone of intergradation between the two forms is rather narrow
and restricted to the region where these two habitats come in contact
with one another. The relationships of A. a. procer to the race to the
northeast will be discussed under that form.

Ameiva auberi paulsoni, new subspecies

holotype: AMNH 83012, an adult male, from 1 km. N. Las Canas, Pinar del
Rio Province, Cuba, taken August 7, 1958, one of a series collected by Albert
Schwartz and George R. Zug. Original number 6329.

PARATYPES (all from Pinar del Rio Province, Cuba): AMNH 83013-83023,
same data as holotype; AMNH 96325-96329, 1 km. E. Las Canas, June 23, 1960,
D. C. Leber, A Schwartz; AMNH 78386-78389, 1 km. N. La Coloma, June 25,
1957, J. R. Feick, W. H. Gehrmann, D. R. Paulson; AMNH 83006-83011, 1 km.
N. La Coloma, August 7, 1958, A. Schwartz, G. R. Zug; AMNH 96324, 1 km.

[ N. La Coloma, June 23, 1960, D. C. Leber,

j DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
i moderate size (males to 92 mm,, females to 75 mm. snout-vent length), low num-
I ber of fourth-toe subdigital scales, femoral pores, and scales in the fifteenth caudal
verticil; dorsum reddish brown with bright lemon-yellow median and dorsolateral
lines in adults, lateral fields black with a series of small red dots or scallops along
1 only the ventral border of the fields; and throat cream colored.

distribution: Known only from the vicinity of Las Canas and La Coloma on
, the south-central coast of Pinar del Rio Province, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements and
j counts: snout-vent length 92 mm., tail (incomplete and regenerated) 123 mm.;

! ventrals in 34 longitudinal and 10 transverse rows; fourth-toes missing; femoral
I pores 12 and 13 (total 25); scales in the fifteenth caudal verticil damaged. Dorsal
j ground color reddish brown in life, with bright lemon-yellow median and dorsolat-
I eral lines, the latter extending onto the basal portion of tail as a dorsolateral pair

64

Annals of Carnegie Museum

VoL. 41

of yellowish streaks; lateral fields black with a series of discrete reddish dots along
their lower borders only, and extending behind the hindlimbs onto the sides of
the tail as a broad black lateral stripe; lower sides tigroid blue-gray and darker
gray; venter very pale blue centrally and darker grayish blue laterally. Throat
cream; chest cream with some very pale bluish suffusions and a few isolated black-
ish flecks on the third and fourth transverse rows of ventral scales. Upper surface
of forelimbs and hindlimbs reddish-brown, with black tigroid markings. Tail deep
blue above, with dorsolateral yellowish lines, bordered below by the black lateral
streak, which in turn is bordered below by a bold white ventrolateral stripe;
under side of tail pale blue.

VARIATION: The series of 28 A. a. paulsoni has the following counts: longitudinal
ventrals 30-34 (mode 32); rows of transverse ventrals 10 in all specimens; fourth-
toe scales 61-77 (mean 68.6); femoral pores 20-28 (mean 24.2); fifteenth verticil
24-27 (mean 25.6). The largest male has a snout-vent length of 92 mm., the
largest female 75 mm.

In all details of pattern and coloration, the series of A. a. poiilsoni is
very consistent. The striking configuration of the lateral fields is espe-
cially noteworthy and is a regular feature of the series. No lizard shows
any tendencies toward having the throat any color (including black)
other than cream. The persistence of the lineate dorsal pattern into full
adulthood in both sexes is an outstanding character.

COMPARISONS: A. a. paulsoni can be differentiated by its cream-
colored throat from the races cacuminis ( dusky gray ) , denticola ( black ) ,
and procer (yellow). A. a. paulsoni is larger than cacuminis, about the
same size as denticola, and distinctly smaller than its relative procer
to the north, west, and east. The lateral field pattern of a row of reddish :
dots along the ventral border of the fields is characteristic of paulsoni
and occurs in no other subspecies. In addition, the reddish-brown dor-
sal coloration and persistent dorsal lines distinguish paulsoni from
the other western subspecies. A. a. paidsoni has the lowest mean of
fourth-toe scales of all Cuban subspecies of A. atd)eri, and the means of
femoral pores and fifteenth verticil scales are also low.

REMARKS: The region whence A. a. paulsoni is known is occupied by
two endemic subspecies of other wide-ranging Cuban species of rep- |
tiles — Anolis sagrei bremeri Barbour and Tropidophis melanurus
dysodes Schwartz and Thomas. La Coloma lies in a region of low and
rather moist palm savannas on the Llanura Costera del Sur de Pinar
del Rio (Marrero, op cit. :430 et seq.). The Las Canas lizards were f
taken on wet sand about a garbage dump adjacent to a borrow-pit
pond and rather near the coast. The Llanura Costera del Sur is a region
of low average rainfall (115 to 140 cm. annually), but the general ]
aspect of the countryside is not arid. Stands of palms ( including Sahal
parviflora, Cameraria retusa, and Colpothrinax wrighti ) and occasion-

1970

Ameiva auberi; I. The Cuban Subspecies

65

al pines occur in wide areas of open grassy flats, which give an almost
cultivated and pastoral aspect to the region.

The complete distribution of A. a. paidsoni doubtless is more exten-
sive than indicated by the three adjacent localities whence it has been
taken. Although I have collected on the southern Pinar del Rio coast
to the east of the known range of paidsoni at Herradura, and to the
west at Punta de Cartas and San Juan y Martinez, no Ameiva were
observed at either locality. The localities for paidsoni are about 42 kilo-
meters airline from the nearest locality of procer to the west (7.6 mi. E.
Isabel Rubio), about 46 kilometers airline from the nearest procer
locality to the northeast (San Diego de los Banos), and about 38 kilo-
meters from the nearest procer to the east ( Dayaniguas ) .

Ameiva auberi auberi Cocteau

Ameiva auberi Cocteau, 1838 or 1839, in de la Sagra, Historia fisica,

PoUtica y natural de la Isla de Cuba: 51.

Ameiva trilineata Gray, 1845, Catalogue of specimens of lizards in

British Museum: 19.

TYPE LOCALITY: Cuba; restricted to the vicinity of La Habana, Habana Prov-
ince, Cuba.

diagnosis: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 92 mm., females to 88 mm. snout-vent length), moderate
number of fourth-toe subdigital scales, and high number of femoral pores and
scales in the fifteenth caudal verticil; dorsum rusty brown to brown in adults,
juveniles with a vivid lemon-yellow median dorsal stripe, lateral fields with much
red pigment, the black areas usually forming a reticulum; and throat orange.

distribution: The north coast of Cuba, from Marianao and La Habana on the
west, east to the vicinity of Canasi, Matanzas Province, on the east, but see dis-
cussion below.

variation: The series of 33 A. a. auberi has the following counts: longitudinal
ventrals 32-35 (mode 32); rows of ventrals 10 in all specimens; fourth-toe scales
68-89 (mean 76.3); femoral pores 23-31 (mean 26.6); fifteenth verticil 26-31
(mean 28.2). The largest male has a snout-vent length of 92 mm., the largest
female 88 mm.

In living material, the dorsal ground color of A. a. auberi varies from
rusty brown to brown; the back is unicolor and without longitudinal
lines. In juveniles there is a vivid yellow median dorsal line and some-
what less obvious dorsolateral yellow lines, all of which persist in var-
ious degrees of intensity into subadults. The lateral fields are reddish,
and usually this pigment is so extensive that the black in the lateral"
field is reduced to a black reticulum or to a linear series of practically
discrete black blotches. The lower sides are blue (PI. 26 L 1; all color
designations from Maerz and Paul, 1950) and this color extends onto

66

Annals of Carnegie Museum

VoL. 41

the ventral scales, the more central rows of which are much paler. The
throat and chin are orange (PI. 10 J 9, PL 9 B 12), somewhat lighter
(almost yellow) on the cheeks, and the axillae and chest are likewise
orange. The orange chest coloration may extend as a pinkish wash over
the blue of the venter as far as the vent, onto the under side of the
hindlimbs and tail, and onto the inner surface of the forelimbs. Juve-
niles have the bellies pink, the throats yellow to pinkish, and tails vivid
blue, whereas the tail in adults is brown above and grayish blue to
purplish below. The ventrolateral pale tail stripe is whitish to grayish,
and not especially prominent in adults.

COMPARISONS: A. a. auheri differs from all other described subspecies
in having the throat orange, and often having an orange wash over the
venter and underside of the hindlimbs. No specimens show any ten-
dency toward having a black throat or chest, although there are often
a few scattered gray scales in the otherwise brightly colored throat. In
dorsal coloration, lack of persistent dorsal stripes, and restriction of the
black in the lateral fields, auheri most closely resembles procer to the
west. The very large size and yellow throat of procer contrast with the
smaller size and orange throat of auheri.

REMARKS: A. a. auheri occurs along the north coast of Habana and
Matanzas provinces. I have encountered no specimens in any situations
except coastal habitats — seagrape stands and on beaches at the base
of the sea cliffs that abound in this stretch of coastline. There are spec-
imens in collections from La Habana and its suburbs ( Marianao, Regia )
that indicate that ground lizards are in part urban in this area, and may
occur some short distance inland from the ocean proper. Barbour’s
(1914:309) reports of A. auheri from two localities in Habana Province
~™ Cojimar and Madruga — are unsubstantiated by specimens. The first
locality is within the known range of A. a auheri (between La Habana
and Boca de Jaruco) and the second lies on the Habana-Matanzas bor-
der, inland about 60 kilometers from the coast. I have seen no lizards
from this general region and the assignment of Ameiva from Madruga
to any specific subspecies must await material. 1 suspect that they will
be found to be black-throated and black-chested like the subspecies
next to the east of A, a. auheri.

SPECIMENS examined: Cuba, Habana Province, Marianao, 1 (USNM 28007);
La Habana, 4 (USNM 58486-58488, AMNH 65599); Regia, 1 (USNM 128105);
2 mi. E. Playa de Guanabo, 2 (AMNH 78021-78022); Boca de Jaruco, 2 (AMNH
78027-78028); 2 mi. E. Boca de Jaruco, 9 (AMNH 78023-78026, 82946-2 speci-
mens, AMNH 96330 — 3 specimens); Santa Cruz del Norte, 4 (IB 137-140); Playa
Jibacoa, 6 ( IB 1296-1301); Matanzas Province, 5.5. mi. N.E. Canasi, 5 (AMNH

1970

Ameiva auberi: I. The Cuban Subspecies

67

82944 — 5 specimens); 6.6 mi. N.E. Canasi, 2 (AMNH 82945 — 2 specimens).

There are eight specimens from northeastern Pinar del Rio Province
that are anomalous: USNM 27617-27618 and USNM 27668-27671
from Mariel, and AMNH 82964-82965 from 5 km. north of Mariel. The
first two lots are old specimens and are much discolored. The last two
specimens were collected by myself. As far as can be determined from
the old specimens, the males from Mariel are characterized by having
at least a black chest (males with snout- vent lengths from about 70
mm. to 87 mm.). One of the recently collected males (snout-vent
length 73 mm. ) also has indications of a black chest. The single female
appears not to have had any black chest or throat pigment. The throats
of the two recently collected males ( snout- vent lengths 80 and 73 mm. )
were recorded as dirty cream and grayish. The dorsa of adults were
apparently some shade of brown, and the subadults show some indica-
tion of lines. Since procer with a yellow throat occurs to the west at
Cabanas, and auberi with an orange throat to the east at Havana, it
seems that there is interposed between these two races a black-chested
subspecies (neither of the two adjacent populations ever has any
indication of a black chest or throat). I withhold any subspecific de-
signation on the specimens from Mariel until additional fresh material
from the northeastern portion of Pinar del Rio Province and the extreme
northwestern portion of Habana Province becomes available.

Ameiva auberi pullata, new subspecies

holotype: AMNH 82953, an adult male, from 13 km, N.E. Matanzas, Matanzas
Province, Cuba, taken June 13, 1958, one of a series collected by Albert Schwartz
and G. R. Zug. Original number 5147.

paratypes (all from Matanzas Province, Cuba): AMNH 82949-82952, 82954-
82963, same data as holotype; AMNH 83747-83748, same locality as holotype,
June 18, 1959, A. Schwartz, B. L. Smith; AMNH 83947-83948, 6 km. N.E. Matan-
zas, June 13, 1958, A. Schwartz, G. R. Zug; IB 136, nr. Matanzas, March 23,
1965, J. de la Cruz; AMNH 32306-32308, Cardenas, January 11, 1927, J. A.
Weber.

ASSOCIATED SPECIMENS: AMNH 82966-82969, 1.6 mi. N. San Miguel de los
Banos, Matanzas Province, Cuba.

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
small size (both males and females to snout-vent length of 81 mm., but see dis-
cussion below), moderate number of fourth-toe subdigital scales, high number of
femoral pores, and moderate number of scales in the fifteenth caudal verticil;
dorsum brown, adults unicolor or with faint yellow median and dorsolateral lines,
juveniles prominently lined with yellow, lateral fields with black restricted to a
series of vertical bars (often absent anteriorly), and red areas almost completely
replaced by tan-to-brown, practically concolor with dorsum; throat with a black
median spot, and chest black.

68

Annals of Carnegie Museum

VoL. 41

DISTRIBUTION: From the vicinity of the city of Matanzas east to Cardenas (but
excluding the distal half of the Peninsula de Hicacos), and presumably inland as
far as the vicinity of San Miguel de los Banos ( see discussion ) in Matanzas Prov-
ince, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements and
counts: snout-vent length 78 mm., tail 163 mm.; ventrals in 33 longitudinal and ]
10 transverse rows; fourth-toe subdigital scales 38 and 38 (total 76); femoral j

pores 14 and 14 (total 28); 25 scales in fifteenth caudal verticil. Dorsum brown I

with median and dorsolateral yellow lines very faint but present; lateral fields ob- j
scure and composed of about 8 to 10 very reduced black vertical bars or dots on j
the posterior half of the trunk, the balance of the fields completely obscured by j

tan pigment. Chest black, pigment extending onto anterior and ventral faces of !

forelimbs; throat with a rather extensive, but incomplete anteriorly and laterally, i
black spot. Belly very pale blue; lower sides dark gray-blue with few very vague j
tigroid markings; upper side of limbs brown with dark-gray tigroid markings. !

Tail brown above, blue-gray below with the ventrolateral pale stripe much re- j

duced and restricted to the anterior third of the tail.

VARIATION: The series of 24 A. a. pullata has the following counts: longitudinal
ventrals 31-34 (mode 32); rows of transverse ventrals 10 or 12 (3 specimens);
fourth-toe scales 64-86 (mean 75.0); femoral pores 23-31 (mean 26.3); fifteenth
verticil 24-29 (mean 26.8). The largest male has a snout-vent length of 81 mm.,
the largest female 81 mm. (see below).

Considering only the speeiinens from between Matanzas and Car-
denas, and excluding those from San Miguel de los Banos, the para-
types agree very well with the description of the holotype in coloration j
and pattern. In life, the median and dorsolateral stripes occur in most jl
adults but are never so prominent as they are in subadults and juve-
niles. The lateral fields are variable but regularly have more tan (rath- i

er than red) and the black area is much restricted or even limited to
disjunct series of bars or irregular blotches. The black chest and throat i
— the latter never completely black but having rather a black spot i
posteriorly — occur only in males. As in the other black-throated sub-
species, the size at which the black throat occurs in males is variable !
in the sample. The smallest (of 13) male with the complete black chest j:
and throat spot has a snout-vent length of 68 mm., the largest 78 mm.

All other males with snout-vent length from 66 mm. to 70 mm., have '
the chest black and an incipiently black spot on the throat, with the
exception of one small male, with a snout-vent length of 66 mm., which
lacks any black on the throat or chest. In females, the maximal black jl
chest and throat-spot condition is equivalent to the intermediate con-
dition in males — a black chest and an incipient black throat spot.
Females in this category range in snout- vent lengths from 62 mm. to ■
72 mm. Females with snout-vent lengths of 52 mm. to 63 mm. have the
chest and throat without black pigment. (i

1970

Ameiva auberi: I. The Cuban Subspecies

69

The four specimens from San Miguel de los Banos present a differ-
ent picture. The largest male and the largest female pullata (both 81
mm.) are drawn from this small San Miguel sample. Otherwise, the
largest pullata (from the coastal sample) male measures 78 mm. and
the largest female 72 mm. Thus the two San Miguel females (snout-
vent lengths 76 and 81 mm. ) are larger than all coastal females and the
two San Miguel males are large (snout-vent lengths 76 mm. and 81
mm.) and at the upper extreme of the coastal males. The larger San
Miguel female also has the tlu'oat and chest as black as coastal pullata
males — a condition not known for coastal females — and the larger
San Miguel male also has the throat more extensively black than any
coastal pullata. Everything considered, I doubt that the San Miguel
series is correctly regarded as A. a. pullata, and doubtless represents a
local interior subspecies. However, absence of an adequate series from
interior Matanzas Province precludes further comment.

COMPARISONS: A. a. pullata requires comparison only with the other
black-throated population thus far described — A. a. cleiiticola. In the
latter, the throat and chest are wholly black, rather than having only a
black spot as in pullata. A. a. clenticola reaches a larger size, and has
black lateral fields with red edges, rather than restricted black mark-
ings with extensive tan edging in the lateral field region. The two races
are separated by the subspecies procer and auberi. From auberi, the
race closest to the west (the closest records of auberi and pullata are
6.6 mi. N.E. Canasi and near Matanzas — about 27 kilometers airline
but actually somewhat farther since the Bahia de Matanzas inter-
venes), pullata differs especially in having the black chest and throat
spot, whereas auberi has the throat orange. A. a. pullata also is bordered
on the north by the next race from the Peninsula de Hicacos and prob-
ably also meets another subspecies from the northeastern corner of Las
Villas Province. Comparisons with these races will be made later.

REMARKS: The AMNH type material of A. a. pullata (with the ex-
ception of the specimens from Cardenas, which were not taken
recently) was collected in grassy areas on the landward side of the
extensive beach between Matanzas and the base of the Peninsula de
Hicacos, and on the beach itself, in scanty gi'owth of herbs and grasses.
The specimens from San Miguel de los Banos were taken in a cut-over
Canaveral and in grass along its edges. At the latter locality, the lizards
were not abundant, whereas along the beaches they were exceptionally
common.

70

Annals of Carnegie Museum

VoL. 41

Ameiva auberi abducta, new subspecies

holotype: AMNH 96331, an adult male, from Punta Hicacos, Matanzas Prov-
ince, Cuba, taken June 30, 1960, by David C. Leber, Original number 886L

PARATYPES: AMNH 96332-96360, 8.9 mi. N.E. Varadero, Matanzas Province,
Cuba, June 30, 1960, R. F, Klinikowski, D. C. Leber, J. D. Smallwood; IB
133-134, Dupont, Matanzas Province, Cuba, July 6, 1966, O. H. Garrido.

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 87 mm., females to 68 mm. snout-vent length), moderate
number of fourth-toe subdigital scales and femoral pores, high number of scales
in the fifteenth caudal verticil; dorsum brownish with persistent pale yellowish
tan median and dorsolateral lines in adults, lateral fields usually inconspicuous
with black restricted to discrete vertical bars or blotches or absent entirely, lateral
field red replaced by brownish; and throats yellowish orange.

DISTRIBUTION: Known only from the distal half of the Peninsula de Hicacos,
Matanzas Province, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 85 mm,, tail 181 mm., tip missing; ventrals in 33
longitudinal and 10 transverse rows; fourth-toe subdigital scales -/36; femoral
pores 12 and 13 (total 25); 28 scales in fifteenth caudal verticil. Dorsum dull
brownish with median and dorsolateral lines present and yellowish tan; lateral
fields obscure and composed of a black reticulum with extensive invasion of
brownish, concolor with dorsum. Throat yellowish orange; venter and chest dull
bluish. Lower sides dark bluish gray with some vague darker gray tigroid mark-
ings. Tail brownish above, dull blue-gray below, ventrolateral caudal stripe ob-
scure. Limbs brownish above with vague darker markings.

VARIATION: The series of 30 A, a. ahclucta has the following counts: longitudinal
ventrals 31-34 (mode 32); rows of transverse ventrals 8 (3 specimens) or 10;
fourth-toe scales 64-84 (mean 75.2); femoral pores 21-30 (mean 25.4); fifteenth
verticil 24-29 (mean 27.6). The largest male has a snout-vent length of 87 mm.,
and the largest female has a snout-vent length of 68 mm.

The series of paratypes agrees well with the description of color and
pattern of the holotype. The persistence of the yellowish tan median
and dorsolateral lines is consistent in all adults, although in all cases
these lines are not particularly contrasting with the rather drab brown-
ish dorsal ground color. Major variation occurs in the condition of the
lateral fields. These may be somewhat more distinct than in the holo-
type, depending upon the precise shade of the invading tan or brown-
ish pigment and the extent to which it has restricted the back pigment.
Some ahducta lack any sort of black markings in the lateral fields what-
soever, whereas in others the black areas are a less intense gray. Juve-
niles are not appreciably more brightly colored than adults, although
the dorsal lines may be somewhat more apparent.

COMPARISONS: A. a. ahducta requires comparison with three other
subspecies: paulsoni (which is also dorsally lined in adulthood), auberi
(which has an orange throat), and pullata (which is geographically

1970

Ameiva auberi: I. The Cuban Subspecies

71

adjacent). All other described races are distinguishable by throat color
or dorsal pattern. From paulsoni, abducta differs in throat color ( yellow-
ish-orange versus cream), in having the dorsal ground color paler
(brownish versus reddish brown) with consequent loss of intensity of
the dorsal lines (which are bright lemon-yellow in paulsoni and yel-
lowish tan in abducta). From auberi, which has a brightly colored
throat, abducta differs in having a lineate and paler (brownish versus
rusty brown-to-brown ) dorsum, and a less brightly colored throat. The
ranges of auberi and abducta are separated by that of pullata.

From geographically adjacent pullata, abducta differs in having a
yellowish orange throat in contrast to a black chest and black-spotted
throat. Although both abducta and pullata are striped dorsally, the
striping in pullata is much fainter and less consistent an adult feature
than in abducta.

REMARKS: The Peninsula de Hicacos extends in a northeastern direc-
tion from the north coast of Matanzas Province, and is about 20 kilo-
meters long. The outer extreme of the peninsula is rather arid, is cov-
ered with extensive scrub and dry woods, and has extensive sandy
beaches. Ground lizards were abundant in this latter habitat, which
they shared with Leiocephalus raviceps klinikowskii Schwartz, and in
the scrub itself. The closest records of abducta and pullata (Dupont
and 13 km. N.E. Matanzas ) are separated by only 28 kilometers airline,
but the latter locality (for pullata) is on the Cuban mainland, and
much of the basal portion of the Peninsula de Hicacos is presently con-
verted to resorts and hotels and thus not available for extensive collect-
ing. Presumably abducta and pullata intergrade between these two re-
ported localities. Buide (1966:8), in discussing the peculiarity of dis-
tribution of L. f. klinikowskii, suggested that this subspecies had evolv-
ed on the peninsula when the latter was insular. The differentiation of
A. a. pullata and A. a. abducta may be a comparable situation.

Ameiva auberi gemmea, new subspecies

holotype: AMNH 82972, an adult male, from mouth of Rio de Sierra Morena,
near Playa Ganuza, Las Villas Province, Cuba, taken June 20, 1958, one of a
series collected by Albert Schwartz and George R. Zug. Original number 5297.

PARATYPESi AMNH 82970-82971, 82973-82981, same data as holotype;
AMNH 82983-82985, 0.5 km. S. Playa de Sierra Moreno, Las Villas Province,
Cuba, June 22, 1958, G. R. Zug.

diagnosis: A Cuban subspecies of A. auberi characterized by a combination of
large size (males to 105 mm., females to 97 mm. snout-vent length), moderate
number of fourth-toe subdigital scales, femoral pores, and scales in the fifteenth
caudal verticil; dorsum reddish brown with median and dorsolateral lines yellow-

72

Annals of Carnegie Museum

VoL. 41

ish to brown in adults, lateral fields prominent and extensively blaek with two
ineluded rows of diserete brown spots and with seattered bright yellow scales;
and throats yellow to orange.

DISTRIBUTION: Knowii only from the northwestern corner of Las Villas Province,
Cuba, near the Las Villas-Matanzas boundary.

DESCRIPTION OF HOLOTYPE: All adult male with the following measurements and
counts: snout-vent length 97 mm., tail 182 mm.; ventrals in 33 longitudinal and
10 transverse rows; fourth-toe subdigital scales 32 and 32 (total 64); femoral
pores 13 and 14 (total 27); 24 scales in fifteenth caudal verticil. Dorsum reddish
brown with median and dorsolateral lines yellowish and fairly prominent; lateral
fields black with two longitudinal series of irregularly placed but discrete and
more or less round brown dots, the more ventral of the two series additionally
dotted with scattered bright yellow scales; lateral fields bounded below by a bold
greenish-grayish longitudinal stripe from above the forelimb to the groin. Lower
sides gray with a distinct darker-gray reticulum; belly clear grayish centrally,
darker grayish blue laterally on two outermost dows of ventrals. Tail brownish
above, pale-gray below, with a bold whitish ventrolateral stripe for about half the
length of the tail. Limbs brown above with vague darker reticulations. Throat
yellowish, with orange (PI. 10 E 11) suffusions.

VARIATION: The series of 15 A. a. gemmea has the following counts: longitudinal
ventrals 31-34 (mode 32); rows of transverse ventrals 10 in all specimens; fourth-
toe scales 64-80 (mean 73.8); femoral pores 23-30 (mean 25.9); fifteenth verticil
24-30 (mean 25.9). The largest male has a snout-vent length of 105 mm., the
largest female 97 mm.

The reddish brown dorsum with yellowish-to-brown dorsal lines is a
common feature of the entire series. The lines in juveniles are yellow to
greenish yellow and are even more bold than in the adults. The exten-
sively black lateral fields with their two (or at times, three) rows of
discrete, more or less circular, brown spots and additional bright yel-
low scales are also of regular occurence, at least in adults, and the lateral
fields are bordered below by a well-defined greenish or gray lateral
stripe, which continues onto the tail as a bold whitish ventrolateral
stripe. The throat color varies from flesh (in juveniles) to yellow or
orange (PL 10 K 2 to Pi. 10 E 11) in adults of both sexes, and the
cheeks are usually concolor with the throat. The belly varies from very
pale blue (usually) to gray, or even faintly purplish in juveniles.

COMPARISONS: A. a. gemmea is unlike any other subspecies of A.
auheri in having scattered yellow scales in the lateral fields which
themselves are distinctive in being extensively black with the double
series of discrete brown spots. By these characters gemmea may be
distinguished from all other subspecies. In throat color gemmea re-
sembles auheri and ahdueta most closely, but the striped dorsum, de-
tails of lateral field coloration and pattern, and dorsal color will dis-
tinguish gemmea from these two races. The conspicuous lateral fields

1970

Ameiva auberi: I. The Cuban Subspecies

73

of gemmea are in strong contrast to those of the geographically close
ahducta.

REMARKS: A. a. gemmea is known to occupy only a small area in ex-
treme northwestern Las Villas Province, but its range is almost surely
more extensive. At the type locality, the ground lizards were collected
in brush and Aeaeia on muddy ground in and about a fresh-water
swamp, altogether a rather unusual situation for Ameiva. The range of
gemmea is removed some 58 kilometers airline from the most eastern
record for the next adjacent race to the west, A. a. pullata, at Cardenas.

Ameiva auberi extraria, new subspecies

holotype: AMNH 82982, an adult male, from Cayo Bahia de Cadiz, Las Villas
Province, Cuba, taken June 21, 1958, one of a series collected by Albert Schwartz
and George R. Zug. Original number 4380.

PARATYPES: AMNH 96250-96280, same data as holotype.

ASSOCIATED SPECIMENS (all froiii Las Villas Province, Cuba): IB 41, Cayo
Monos de Jutia; IB 64-71, Cayo Lanzanillo; IB 36-40, Cayo Carenero; IB 61-63,
Cayo las Tocineras; IB 42-50, Cayo Tio Pepe.

DIAGNOSIS: A Cuban subspecies of A. a. auberi characterized by a combination
of small size (males to 79 mm., females to 68 mm. snout-vent length), moderate
number of fourth-toe subdigital scales, low number of femoral pores, and mod-
erate number of scales in the fifteenth caudal verticil; dorsum yellowish brown to
reddish with only a pale-yellow median line which does not extend to the groin
but is diffuse or absent entirely posteriorly; lateral fields at best only faintly indi-
cated by a darker brownish streak, and usually absent completely; and throat
gray to diffuse dull orange.

DISTRIBUTION: The Archipielago de Sabana off the northern coast of Las Villas
Province, Cuba; known from Cayo Bahia de Cadiz, Cayo Monos de Jutia, Cayo
Lanzanillo, Cayo Carenero, Cayo las Tocineras, and Cayo Tio Pepe; specimens
from Cayo Carenero tend slightly to be intergradient with the subspecies on
Cayuelo de la Vela.

DESCRIPTION OF HOLOTYPE: All adult male with the following measurements and
counts: snout-vent length 75 mm., tail 179 mm.; ventrals in 34 longitudinal and

10 transverse rows; subdigital scales 37 and 40 (total 77); femoral pores 11 and

11 (total 22); 29 scales in fifteenth caudal verticil. Dorsum yellowish brown, un-
lined except for a vague remnant of the median line which is very pale yellow
and incomplete posteriorly. Lateral fields absent, their position indicated only by
a vague grayish band along each side. Tail brownish above, grayish blue below,
and without a ventrolateral pale stripe. Throat grayish orange, belly blue. Cheeks
pale yellowish; lower sides dull gray. Upper surfaces of limbs concolor with dor-
sum and with very vague grayish reticula.

VARIATION: I have grouped together as A. a. extraria those speeimens
of ground lizard from the Arehipielago de Sabana whose throats are
gray (in which case the sublabials and infralabials are orange) to dull
orange. Lizards from Cayuelo de la Vela (adjacent to Cayo Carenero)
I regard as a distinct form, restricted to that cay, because of their hav-

74

Annals of Carnegie Museum

VoL. 41

ing black throats and chests. Of the cays inhabited by extraria, one
(Cayo Bahia de Cadiz) lies near the western end of the chain of islets,
near the Matanzas-Las Villas line, whereas the balance of the cays lie
about 75 kilometers southeast of Bahia de Cadiz and to the east of the
town of La Isabela. The series of 55 A. a. extraria has the following
counts : longitudinal ventrals 28-34 ( mode 32 ) ; rows of transverse
ventrals 8 or 10 ( mode 10, but see comments below ) ; fourth-toe scales
68-88 (mean 76.3); femoral pores 21-27 (mean 23.8); fifteenth ver-
ticil 23 - 30 ( mean 26.5 ) . The largest male has a snout-vent length of
79 mm. and is from Cayo Monos de Jutia. The largest female has a
snout-vent length of 68 mm. and is from Cayo Bahia de Cadiz.

The dorsal color varies from yellowish brown to reddish. The for-
mer is the modal adult coloration, the latter that of young subadults
and juveniles. The dorsolateral stripes are absent ( as are also the lateral
fields), and the median dorsal stripe is pale yellow and never com-
plete, disappearing about two-thirds of the way between the neck and
the groin. Even the smallest juvenile ( snout-vent length 45 mm. ) lacks
the dorsolateral stripes and the lateral fields. The median dorsal stripe
in this juvenile agrees in intensity and extent with that of full adults.
In juveniles, the tail is not bright blue but concolor with the dorsum.
The throat coloration varies from gray (regularly with orange sub-
labials and infralabials) to diffuse grayish orange. The differences in
throat pigmentation in the topotypical sample may well be correlated
with age or sex, but my field notes are not specific in this regard. Sr.
Garrido’s notes on the IB material indicate that a male from Cayo
Monos de Jutia and one from Cayo Tio Pepe had solid orange throats,
whereas males from Cayo Carenero, Cayo las Tocineras, and Cayo
Lanzanillo have gray throats with orange sublabials and infralabials.
Considering the geographic relationships of these cays to each other,
it seems preferable to regard all populations as A. a. extraria rather
than attempt to distinguish two subspecies on the basis of throat color.
At least lizards from the various cays present a generally uniform dor-
sal aspect, although even this, as might be expected, shows some minor
variations in intensity of pigmentation. The absence of lateral fields
and subdued-to-absent dorsal lineation are common to all samples.

There has been some evolution within the Archipielago de Sabana,
as far as A. a, extraria is concerned. This is shown by pertinent scale
differences between the several populations regarded as extraria. Most
noteworthy is the number of transverse ventral rows in the Cayo Tio
Pepe sample. Of the nine lizards from this cay, eight have only 8 trans-

1970

Ameiva auberi: I. The Cuban Subspecies

75

verse rows, the ninth having the more orthodox 10. No other Cuban
sample studied has such a high incidence of this low count, and in fact
no other extraria has a count of 8. There is also a distinct tendency for
topotypical extraria (Cayo Bahia de Cadiz in the west) to have more
fourth-toe scales (71 to 88), whereas lizards from the group of eastern
cays have slightly lower counts (68 to 77 ) . Despite these peculiarities,
1 feel that all should at least temporarily be considered A. a. extraria.
There is such an abundance of cays along this northern Las Villas
coast, whose Ameiva are as yet unsampled, and the two areas whence
specimens are available are so far removed from one another, that to
consider all material as extraria seems the conservative course at this
time.

COMPARISONS: A. a. extraria is geographically closest, of the named
subspecies, to A. a. gemrnea, but the two subspecies are very different
in pigmentation and pattern. The bold and decorated lateral fields of
gemmea differ strongly from the virtually non-existent lateral fields of
extraria. The throat color of extraria is darker (gray to dull orange)
than that of gemmea ( yellow to orange ) , and gemmea reaches a much
larger size than extraria (105 mm. versus 79 mm. in males). The black
throats and chests of denticola and pullata distinguish those races from
extraria, and the cream throat of paulsoni and the yellow throat of
procer distinguish these subspecies. Those races which have gray-to-
orange ( although never as dull an orange as extraria ) throats are cacu-
minis, auberi, and abducta. All are somewhat larger than extraria,
none lacks lateral fields, and none in adults has only a median line
without dorsolateral lines. The dorsal pigmentation of extraria is paler
than that of any subspecies thus far described. In number of femoral
pores, extraria ranks next to the lowest of all Cuban subspecies.

REMARKS: At the type locality, A. a. extraria was quite abundant.
The vegetation was arid forest and scrub, with some mangroves along
the margins of the cay. In general Cayo Bahia de Cadiz is fairly high
and rocky, rather than flat and sandy.

Although Cayo Bahia de Cadiz lies only about 20 kilometers off the
coast and on the fringing bank, whose depth between the coast and the
cay does not exceed 1.25 fathoms, the striking differences between
gemmea and extraria suggest that either these two subspecies have
been long separated from one another or that extraria has not been
derived directly from the mainland gemmea but from a more east-
ern population. The occurrence of extraria on cays off the coast at La
Isabela also suggests that neither gemmea nor the subspecies on the

76

Annals of Carnegie Museum

VoL. 41

adjacent coast and at Sagua la Grande has been the ancestor of extra-
ria. Resemblances between extraria and subspecies farther to the east
are greater than between extraria and any of the more western races
(including gemmea, abducta, and pullata). Elaboration of these re-
marks will follow in a later section.

Ameiva auberi extorris, new subspecies
holotype: IB 52, an adult male, from Cayuelo de la Vela, Las Villas Province,
Cuba, taken March 11, 1965, one of a series collected by Orlando H. Garrido.
PARATYPES: IB 51, IB 56-57, ASFS V11373-11375, sama data as holotype.
diagnosis: A Cuban subspecies of A. auberi characterized by a combination of
small size (males to 76 mm. snout-vent length, females unknown), low number of
fourth-toe subdigital scales, moderate number of femoral pores, and low number
of scales in the fifteenth caudal verticil; dorsum dull brown, completely patternless
and without lateral fields; and throat with a black spot and chest black,

distribution: Known only from Cayuelo de la Vela, Archipielago de Sabana,
off the north coast of Las Villas Province, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements and
counts: snout- vent length 73 mm,, tail 172 mm.; ventrals in 33 longitudinal and
10 transverse rows; fourth-toe subdigital scales -/35; femoral pores 13 and 13
( total 26 ) ; 26 scales in fifteenth caudal verticil. Dorsum ( as preserved, but not
strongly different from color notes) brown, completely without any indication of
dorsal lines or lateral fields. Throat gray, with a black posterior spot extending
onto the mesoptychium; first three transverse rows of ventrals on chest black.
Venter dull blue-gray. Limbs brown, concolor with dorsum. Tail dull gray above,
blue-gray ( slightly darker than venter ) below, without lines.

VARIATION: The series of seven male A. a. extorris has the following counts:
longitudinal ventrals 31-33 (mode 33); rows of transverse ventrals 10 in all speci-
mens; fourth-toe scales 68-72 (mean 69.5); femoral pores 24-28 (mean 25.3);
fifteenth verticil 24-27 ( mean 25.4 ) . The largest male has a snout-vent length of
76 mm.; females are unknown.

Little need be said about pattern and color variation in the short
series of extorris. The series consists exclusively of males with snout-
vent lengths ranging between 64 mm. and 76 mm., and the smaller in-
dividuals are exactly like the holotype as far as dorsal pattern and col-
oration are concerned. The entire aspect of the lizards is one of ex-
treme drabness, without markings of any kind.

Males with snout-vent lengths between 68 mm. and 76 mm. have the
thi'oat and chest pattern as described for the holotype. The three re-
maining males with snout- vent lengths between 64 mm. and 69 mm.
lack the throat spot but have some black scales on the mesoptychium
and chest.

COMPARISONS: A. a. extorris requires comparison with the black-
throated and black-chested subspecies heretofore named {denticola,
pullata) and with the geographically adjacent extraria. Compared

1970

Ameiva auberi: L The Cuban Subspecies

77

with the subspecies denticola and pullata, extorris is much more drab
and lacks dorsal lines and lateral fields. Additionally, the entire throat
is black in denticola, whereas pullata resembles extorris in having only
a black throat spot. As might be surmised, extorris most closely re-
sembles its neighbor extraria, but extorris carries the reduction of dor-
sal and lateral markings even further, is generally darker than extraria,
and has the black throat spot and chest which extraria lacks. A. a.
extorris has the next-to-the-lowest mean of number of fourth-toe scales
of all the Cuban subspecies, whereas extraria ranks ninth in this
character.

EEMAKKS: Although I debated naming the population on Cayuelo de
la Vela as a distinct subspecies, the presence of a black throat and chest
seems sufficiently distinctive, despite the apparently restricted distri-
bution of the subspecies, to separate these lizards from their relatives
on the other cays. Interestingly, a small male (IB 52) extraria from
Cayo Carenero, the cay closest to de la Vela and separated from it by
a narrow channel only two fathoms deep, has a few black scales on the
throat, whereas all other Carenero lizards have throats typical of ex-
traria. Apparently at least the Carenero lizards still retain the poten-
tiality for black throatedness, and isolation of a population of extraria
on de la Vela has resulted in the expression of this character on that
single island.

Ameiva auberi garridoi, new subspecies

holotype: IB 96, an adult male, from Ciiatro Bocas, Sagiia la Grande, Las
Villas Province, Cuba, taken May 2, 1966, one of a series collected by P. Saavedra.

PARATYPES (all from Las Villas Province, Cuba): IB 95, CM 40638--40639, same
data as holotype; IB 107-108, IB 110-112, NMC 10066 (3 specimens), CM
40640, same locality as holotype, March 21, 1966, Jorge de la Cruz; ASFS VI 1376,
Finca El Dorado, Sagua la Grande, May, 1966, P. Saavedra; ASFS VI 1377-1 1379,
IB 116-119, Finca El Dorado, March 9, 1965, O. H. Garrido; IB 79, IB 103,
Cayo Boca de Sagua, Isabela de Sagua, May, 1966, P. Saavedra.

DIAGNOSIS: A Cuban subspecies of A. atiheri characterized by a combination of
large size (males to 110 mm., females to 70 mm. snout-vent length), high number
of fourth-toe subdigital scales, very high number of femoral pores, and high num-
ber of scales in the fifteenth caudal verticil; dorsum brown and unlined in adults
but with a median and two dorsolateral white lines in subadiilts and juveniles;
lateral fields black, bordered above and below by a marginal row of large brown-
to-tan scallops, which in adults tend to limit the black lateral field color into al-
most a series of vertical lateral bars; and throat lemon-yellow.

DISTRIBUTION: Known only from the region between the city of Sagua la Grande
and La Isabela, in northern Las Villas Province, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements and
counts: snout-vent length 88 mm., tail 185 mm.; ventrals in 33 longitudinal and
10 transverse rows, siibdigital scales 39 and 40 (total 79), femoral pores 12 and

78

Annals of Carnegie Museum

VoL. 41 i'

I

13 (total 25); 27 scales in fifteenth caudal verticiL Dorsum (as preserved) me- i
dium brown and unlined; top of head concolor with dorsum. Lateral fields black I
with two rows of brown scallops, one dorsal and the other ventral, limiting the
black lateral field pigment into an irregularly shaped but well-defined lateral figure; i
lower sides more or less tigroid blue and blue-grey; tail blue-gray above, grayer |
below, with the white ventrolateral longitudinal line fairly prominent on the basal
half of the tail. Throat and mesopytchium clear, bright lemon-yellow. Chest yel-
low, admixed with blue-gray; venter yellowish medially on about first eighteen i
transverse rows of ventrals, marbled blue and yellow to cream elsewhere. Limbs ’
blue-gray, vaguely marbled with darker gray, especially on hindlimbs; transverse
white thigh line prominent.

VARIATION: The series of 23 A. a. garridoi has the following counts: longitudinal
ventrals 31-34 (mode 33); rows of transverse ventrals 10 in all specimens; fourth
toe scales 68-87 (mean 78.3); femoral pores 25-32 (mean 27,6); fifteenth verticil ;
26-30 (mean. 27.9). The largest male has a snout-vent length of 110 mm., the j
largest female 70 mm. i

Adults resemble the description of the holotype closely. The median |
and dorsolateral lines are seldom in evidence, and when visible, they ;
are barely discernible in the preserved material. The lateral fields
regularly have a dorsal and ventral row of brown scallops, which may
ultimately so restrict and limit the black lateral field pigment that the I
latter becomes a series of vertical bars (ASFS V11376, IB 103) especially
anteriorly. The throat is bright lemon-yellow in both sexes, although ;
apparently the throat color of females is somewhat paler than that of |
males. One male ( IB 95, snout-vent length 105 mm., and thus not quite
the largest male) has a few black scales on the mesoptychium, and the '
entire venter has some black pigment dispersed on all the ventrals, |
although the ventral color is still more blue than black. The largest male
(IB 103) shows no evidence of this darkening, nor do any other speci-
mens, regardless of sex or size. The white ventrolateral tail line is at least
indicated basally in even the largest specimens. '

Females and subadult males and juveniles resemble the adult males
in dorsal color and in the configuration of the lateral fields, but there
are a median and two dorsolateral white lines. The lateral fields are f
also delimited below by a pale line. )

COMPARISONS: A. a. garridoi most closely resembles A. a. procer to
the west in Pinar del Rio Province. Both subspecies have yellow throats
and both are quite large, although procer exceeds garridoi in size ( 136
mm. versus 110 mm.). Both races have the lateral fields with dorsally
and ventrally scalloped edges. Restriction of black lateral field pigment
is much more pronounced in procer than in garridoi, however, and the ,
scallops in procer are brick-red rather than brown as in garridoi. A. a. '
procer has a significantly lower mean (71.9) of fourth-toe scales than

1970

Ameiva auberi: I. The Cuban Subspecies

79

does garridoi (mean 78.3), which ranks third among the Cuban races. In
femoral pores, garridoi (mean 27.6) ranks first among the Cuban sub-
species, whereas procer has a lower mean (26.0) and ranks tenth. The
mean of fifteenth caudal verticil scales are high in both subspecies ( 27.3
and 27.9).

A. a. garridoi differs from the other subspecies in having a lemon-
yellow throat {cacuminis, denticola, paulsoni, auberi^ pullata, ahducta,
gemmea, extraria, extorris), in having a unicolor and unlined dorsum
(cacuminis, paulsoni, pullata, ahducta, gemmea), and in having
lateral fields well expressed ( ahducta, extraria, extorris ) . The differences
between garridoi and closely adjacent (but insular rather than main-
land) extraria and extorris are striking in size, coloration, and pattern.
The nearest mainland subspecies to the west is gemmea, from which
garridoi differs in many details, not the least of which is the absence of
the yellow decorative spots in the lateral fields.

REMARKS: I have not seen living nor freshly collected examples of
A. a. garridoi. All specimens of this subspecies have been collected by
Orlando H. Garrido (for whom the subspecies is named), Jorge de la
Cruz, and P. Saavedra. Sr. Garrido (in litt., March 17, 1966), after
having collected the first specimens of garridoi, noted the distinct char-
acters of this population as he collected material at El Dorado and
along the banks of the Rio Sagua la Grande. I am very grateful to Sr.
Garrido for allowing me to describe this distinctive subspecies and it
gives me great pleasure to name it in his honor. Doubtless A. a. garridoi
will be found to have a wider distribution than the few locality records
presently indicate.

Ameiva auberi atrothorax, new subspecies

HOLOTYPE: AMNH 78035, an adult male from Finca Morales, 8 mi. N.W. Trini-
dad, Las Villas Province, Cuba, taken July 27, 1957, one of a series collected by
William H. Gehrmann, Jr., and John R. Feick. Original number 3077.

PARATYPES (all from Las Villas Province, Cuba); AMNH 78034, 78036-78040,
same data as holotype; AMNH 78029-78032, Trinidad, July 28, 1957, native col-
lector; AMNH 78033, Trinidad, July 26, 1957, W. H. Gehrmann; AMNH 78044,
Trinidad, July 29, 1957, native collector; AMNH 1607, Trinidad, July 30, 1892,
F. W. Chapman; AMNH 78045, Finca la Pastora, 2 km. NW Trinidad, July 30,
1957, native collector; IB 100-102, Finca la Pastora, March 8, 1966, O. H. Gar-
rido; AMNH 96361-96368, 8 mi. N.W. Trinidad, July 14, 1960, D. C. Leber,
R. F. Klinikowski; USNM 138474-138484, 10 mi. N.W. Trinidad, September 10,

1956, J. D. Hardy; AMNH 78041-78043, Punta Casilda, near Trinidad, July 29,

1957, W. H. Gehrmann, A. Schwartz; ASFS VI 1380-1 1382, Casilda, September
1, 1966, T. Naranjo; IB 126-127, ASFS V11383, CM 40641-40642, Cayaguazan,
Casilda, March 6, 1966, O. H. Garrido; NMC 10065 (2 specimens), Los Bias-

80

Annals of Carnegie Museum

VoL. 41

mones, Casilda, January 20, 1966, T. Naranjo; IB 128-130, Playa de Maria Aguilar,
March, 1965, T. Naranjo.

ASSOCIATED SPECIMENS (all from Las Villas Province, Cuba): USNM 136096,
UMMZ 65323 (2 specimens), MCZ 22720, MCZ 12363-12364, MCZ 32675, MCZ
22278, MCZ 84918-84919, Soledad; AMNH 82986 (2 speeimens), 2.9 mi. S.E.
Soledad; AMNH 78072, Guajimico, 16 mi. S.E. Soledad; AMNH 78071, 5 km.
S.E. Paso Caballo; MCZ 22721, western edge. Sierra de Trinidad.

DIAGNOSIS: A Cuban subspecies of A. auheri characterized by a combination of
large size (males to 126 mm., females to 93 mm. snout-vent length), high number
of fourth-toe subdigital scales and femoral pores, and very high number of seales
in fifteenth caudal verticil; dorsum tan, either unicolor or with moderately prom-
inent greenish to yellow median and dorsolateral lines, lateral fields with exten-
sive tan markings ( concolor with dorsum ) and black restricted to a series of irreg-
ular blotehes or vermiculations; chest black, and throat with a posterior black
patch.

DISTRIBUTION: South-ccntral Las Villas Province, Cuba, from Paso Caballo and
Soledad on the west to Trinidad and Casilda on the east.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts; snout-vent length 106 mm., tail 256 mm.; ventrals in 33 longitudinal
and 10 transverse rows; fourth-toe subdigital scales 42 and 40 (total 82); femoral
pores 15 and 15 (total 30); 31 seales in the fifteenth caudal verticil. Dorsum
tan, without a median line but with dorsolateral lines greenish and fairly prom-
inent. Lateral fields obscure, consisting of an irregular blaek retieulum encroached
upon by extensive tan markings, concolor with dorsum. Lower sides mottled
greenish yellow. Venter bright blue (Pi. 33 K 4) laterally and pale blue centrally.
Limbs tan dorsally, with some vague grayish markings; hindlimbs fleeked pale
blue (Pi. 33 G 5). Tail tan above, grayish blue below; pale ventrolateral caudal
stripe obseure. Chest black (including first 12 transverse rows of ventrals); throat
with a large black patch or blotch, continuous across mesoptychium with black
chest. Face suffused with reddish, most prominent on lips, around tympanic
opening, and sides of neck.

variation: The series of 65 A. a. atrothorax has the following counts: longitu-
dinal ventrals 31-35 (mode 34); rows of transverse ventrals 10, 11 (1 specimen),
or 12 (4 specimens); fourth-toe seales 72-91 (mean 79.0); femoral pores 21-32
(mean 26.8); fifteenth verticil 25-33 (mean 29.5). The largest male has a snout-
vent length of 126 mm., the largest female 93 mm.

A. a. atrothorax is a somewhat variable race as far as dorsal pattern
is concerned. The dorsal ground color is always tan. The median and
dorsolateral pale lines may be expressed or not in full adults, and their
color is variable, from greenish (in the Trinidad region) to yellow
(Paso Caballo). The lateral fields are likewise variable, although the
black reticulum and much invasion of tan pigment into the lateral field
area is the modal condition. At times the black pigment forms a sinuous
and rather irregular line, or a series of rather vertically oriented short
bars or spots. Only in juveniles is the lateral field extensively black. In
adults from near Soledad the lateral fields are more prominent than

1970

Ameiva auberi: I. The Cuban Subspecies

81

they are in Trinidad specimens. The lower sides have been regularly
recorded in life as greenish yellow or bluish green.

The status of the black chest and throat pigmentation in both sexes
may be summarized as follows. Thirty males (snout-vent length 65
mm. to 126 mm. ) and thirty females ( snout-vent lengths 55 mm. to 93
mm. ) were separated into four categories : ( 1 ) black chest and black
throat patch fully developed; (2) black chest fully developed and
small black throat patch; (3) black chest fully developed and only
faint indication of black throat patch; (4) no black on chest or throat.
Thirteen males (snout- vent lengths 72 mm. to 126 mm., and thus in-
cluding the largest male) fall in the first category and thus have the
complete complement of dark pigment. Ten males (snout-vent lengths
75 mm. to 108 mm. ) have the chest black and the throat patch partially
developed. Four males (snout-vent lengths 70 mm. to 90 mm.) fall into
category 3, and three males (snout-vent lengths 65 mm. to 82 mm.)
lack any black pigment and fall into category 4. In summary, almost
all males (one exception, snout-vent length 82 mm.) above 71 mm.
have at least the chest black; the degree of development of the throat
patch is variable but it is usually well expressed at snout-vent lengths
above about 90 mm. Only one female (snout-vent length 60 mm.) falls
in category 1, and only four females (snout-vent lengths 61 mm. to 93
mm.) fall into category 2. Eight females (snout-vent lengths 63 mm.
to 87 mm. ) are regarded as being in category 3, and 17 females ( snout-
vent lengths 55 mm. to 80 mm. ) lack any black pigment and are in
category 4. In summary, development of black chest and throat in fe-
males seldom reaches the extreme it does in males. Females in excess
of 79 mm. are fairly consistent in having some black chest and throat
pigment, whereas females below 68 mm. usually lack pigment on either
chest or throat. The significance of these variations in each sex is no
more open to solution here than in the previously discussed black-
throated and black-chested subspecies.

Although the juvenile dorsal color is tan as it is in adults, juveniles
are purplish ventrally. This ventral coloration may persist into some
subadult females as dull pink.

COMPARISONS: In having a black chest and black throat spot (in
contrast to an all-black throat), A. a. atrothorax is like A. a. piillata and
A. a. extorris. The southern Las Villas subspecies reaches a much larger
size ( 126 mm. in male atrothorax, 81 mm. in male pullata, 76 mm. in
male extorris) than the two similar subspecies, has a tan dorsum
(brown in pullata), and has the lateral fields present and prominent

82

Annals of Carnegie Museum

VoL. 41

(absent in extorris). The very high mean number of scales in the fif-
teenth caudal verticil in atrothorax (29.5) contrasts with the much
lower means in pullata (26.8) and extorris (25.4). Similar mean differ-
ences in fourth-toe scales (79.0 in atrothorax, 75.0 in pullata, 69.5 in
extorris) are significant. The races atrothorax and pullata resemble
each other in dorsal pattern ( either unicolor or with faint longitudinal
lines), although the lines are usually greenish in atrothorax and yel-
low in pullata.

A. a. atrothorax differs from all other subspecies except denticola, in
having a black chest and a black tliroat patch. A. a. atrothorax differs
from denticola in having a black throat patch rather than the throat
entirely black.

REMARKS: In my experience, Ameiva is not especially common in
the vicinity of Trinidad. Specimens were collected on the outskirts of
the city along stone walls in short grass, and at Punta Casilda only a
very few were seen along the boulder- strewn beach. The series from
the type locality was taken in xeric scrub woods. The substrate was
limestone outcroppings and rocks, imbedded in red earth (see
Schwartz, 1959d:15 for description). Several sandy scrub areas along
the coast between Trinidad and Cienfuegos were visited, but despite
apparently suitable habitat, no Ameiva were observed. The fact that
only a single specimen was collected in two days at Paso Caballo is
significant.

Hardy (1956:323) reported the occurrence of A. auberi on Cayo
Macho de Tierra, about 10 miles southeast of the town of Casilda. No
specimens were collected and the racial status of this population re-
mains in doubt. These lizards may be assignable to atrothorax or may
represent an endemic race on this small offshore islet.

The known range of A. a atrothorax is removed about 120 kilometers i
from that of the closest described subspecies, A. a. garridoi, in northern
Las Villas Province. Although much of the area between these two
races seems suitable for A. auberi, no specimens are known from this
intervening region. In fact, no Ameiva have been collected along the
south coast of Cuba between atrothorax and Cabo Cruz to the east,
and, with the exception of the population next to be described at
Juragua on the southern Las Villas coast, there are no Ameiva known
to the west until one reaches Dayaniguas in Pinar del Rio Province.
Thus A. a. atrothorax and the Juragua subspecies stand alone on the
coast of central southern Cuba.

1970

Ameiva auberi: I. The Cuban Subspecies

83

Ameiva auberi peradusta, new subspecies

holotype: IB 85, an adult male, from Juragua, Las Villas Province, Cuba, taken
May 15, 1966, one of a series collected for Orlando H. Garrido.

PARATYPEs: IB 86-90, CM 40643-40646, same data as holotype; ASF S V11166-
V11170, NMC 9584 (4 specimens), MSB 374, 376-378, 380-381, 383-384, 387,
same locality as holotype, April 27, 1965, M. S. Buide.

diagnosis: a Cuban subspecies of A. auberi characterized by a combination of
large size (males to 105 mm., females to 76 mm. snout-vent length), very high
number of fourth-toe subdigital scales, high number of femoral pores and scales
in fifteenth caudal verticil; dorsum grayish black in adults, without longitudinal
lines, lateral fields terra cotta, v^th or without (usually) some included irregular
markings or blotches; sides of head, neck, all of throat, under sides of both fore-
limbs and hindlimbs, and almost all of venter, rich, lustrous black.

DISTRIBUTION: Knowu only from the type locality on the west side of the Bahia
de Cienfuegos, Las Villas Province, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements and
counts: snout-vent length 105 mm., tail 210 mm.; ventrals in 33 longitudinal and
10 transverse rows; fourth-toe subdigital scales 44 and 43 (total 87); femoral
pores 13 and 13 (total 26); 29 scales in the fifteenth caudal verticil. Dorsum dull
grayish black, top of head dark grayish brown; no dorsal or dorsolateral lines.
Lateral fields terra cotta in life with a few scattered and irregularly shaped black
spots or flecks within the terra cotta band. Sides of head and neck, all of throat,
under side of forelimbs and all of venter rich, lustrous black; under side of hind-
limbs dark gray. Infralabials and sublabials blackish gray; snout reddish. Fore-
limbs darker than hindlimbs above, the former concolor with the dorsum, the latter
paler brownish gray with some scattered flecking. Lower sides blue-black, with
some isolated blue flecks on the more lateral black rows of ventrals. Tail gray-
brown above, blue-gray below; ventrolateral pale caudal stripe indistinct on
basal third of tail, absent on distal two-thirds.

VARIATION: The series of 28 A. a. peradusta has the following counts: longi-
tudinal ventrals 32-36 (mode 33); rows of transverse ventrals 10 or 11 ( 1 speci-
men); fourth-toe scales 70-87 (mean 79.9); femoral pores 23-33 (mean 27.2);
fifteenth verticil 26-32 (mean 28.9). The largest male is the holotype with a
snout-vent length of 105 mm.; the largest female has a snout-vent length of 76 mm.

The series of A. a. peradusta is remarkably consistent in all details of
color and pattern. The lateral fields may be less red than those describ-
ed for the holotype, and in many preserved specimens the lateral fields
are no longer evident. Such individuals lack any black flecks or blotches
in the lateral field areas, and the dorsum and sides are unicolor grayish
black to brownish gray. The black throat, venter, cheeks, and neck are
constant features of males, regardless of size. Even the smallest male
available ( CM 40644, snout-vent length 75 mm. ) is equally as black as
the largest males. There are only three females. One of these (CM
40646, snout-vent length 71 mm.) has the throat and chest solid black,
and there is incipient black and gray smudging over the rosy venter

84

Annals of Carnegie Museum

VoL. 41

and on the underside of the forelimbs. This female is dark brown dor-
sally with three greenish, almost fluorescent, lines. The lateral fields are
unmarked brown, almost terra cotta, and the hindlimbs are dotted with
black. A second female (CM 40645, snout-vent length 72 mm.) resem-
bles males dorsally, lacks lateral fields, and has the black restricted to
the throat and chest, with some black suffusion on the balance of the
venter. In both sexes the pale ventrolateral caudal stripe and the trans-
verse thigh line are inconspicuous.

COMPAEISONS: A. a. peradusta is so very distinctive that it requires
comparison with no other Cuban subspecies. The black cheeks, neck,
throat, and venter distinguish it from all other known populations of
A. auberi in Cuba. Its closest geographic ally is atrothorax, from which
peradusta can be distinguished by the characters that separate it from
all other subspecies.

REMARKS: I am indebted to Sr. Garrido for his excellent notes on A.
a. peradusta and to Sr. Buide for the gift of specimens in his private
collection. Both these men were most impressed with the differences
between peradusta and other Ameiva they had collected, and certainly
peradusta is one of the most distinctive of all the subspecies of A.
auberi, both Cuban and Bahamian.

Juragua lies on the west side of Bahia de Cienfuegos. Although
the mouth of the bay is very narrow, it effectively separates peradusta
from atrothorax, which occurs in the vicinity of Paso Caballo on the
east side of the mouth of the bay. Almost certainly, peradusta is a
direct derivative of atrothorax, a derivative that has been isolated on
the west side of the bay. As noted in the discussion of atrothorax, there
are no Ameiva known to the west of peradusta until the occurrence of
procer at Dayaniguas in Pinar del Rio Province. Possibly A, a. pera-
diista will be found to be widespread along the southern coast between
the Bahia de Cienfuegos and the Bahia de Cochinos. At this time there
are no specimens from this entire southern area. Other than the ad-
jacent atrothorax, the nearest known population to peradusta is gem-
mea, some 1 10 kilometers to the north along the northern Las Villas
coast.

Ameiva auberi llanensis, new subspecies

holotype: MCZ 59321, an adult male, from just south of the west end of Sierra
de Cubitas, Camagiiey Province, Cuba, taken August 21, 1959, by Ramon Molina,
Ernest E, Williams, and Rodolfo Ruibal.

paratypes (all from Camagiiey Province, Cuba): AMNH 78069, 2 km. S.E.
Banao, August 15, 1957, J. R. Feick; AMNH 78070, 2 km. S.E. Banao, August
18, 1957, J. R. Feick; AMNH 96409, 0.1 mi. S.E. Banao, August 3, 1960, D. C.

1970

Ameiva auberi: I. The Cuban Subspecies

85

Leber; AMNH 96406-96408, 2.7 mi. S.E. Banao, August 3, 1960, R. F. Klinikow-
ski, D. C. Leber; AMNH 96404-96405, 8.5 mi. S.E. Banao, August 3, 1960, D. C.
Leber; MCZ 57934-57935, halfway between Camagiiey and Banao, July 16,
1957, R. Molina and R. Ruibal; MCZ 7277, 84917, Camagiiey, 1908, T. Barbour.

diagnosis: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 90 mm., females to 78 mm. snout-vent length), moderate
number of fourth-toe subdigital scales, very low number of femoral pores, and
high number of scales in fifteenth caudal verticil; dorsum brownish with fairly
prominent faintly yellow median and dorsolateral lines, the dorsal fields (between
the median and dorsolateral line on each side) blotched or marbled with black
(especially posteriorly), lateral fields black with heavy brown spotting; and throat
pale orange.

DISTRIBUTION: The serpentine savannas from the western end of the Sierra de
Cubitas south to the vicinity of the city of Camagiiey, Camagiiey Province, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 90 mm., tail 211 mm.; ventrals in 33 longitudinal
and 10 transverse rows; fourth-toe subdigital scales 42 and 39 (total 81); femoral
pores 12 and 13 (total 25); 28 scales in the fifteenth caudal verticil. Dorsum
brown with pale median and dorsolateral stripes evident but only fairly prom-
inent; dorsal fields between median and dorsolateral stripes marked ( especially
on posterior half of trunk, but visible anteriorly as well) with fine filigreed dusky
gray marblings. Lateral fields broad and black, the black much restricted by
about three longitudinal series of brown, subcircular spots, and bounded below
by a prominent pale lateral line. Lower sides blue, marbled with dark gray. Ven-
ter pale blue; throat pale (pale orange in life?). Tail brown above, pale grayish
blue below; pale ventrolateral caudal stripes well developed on basal third of tail.

VARIATION: The series of 13 A. a. llanensis has the following counts: longitudinal
ventrals 31-34 (mode 34); rows of transverse ventrals 10; fourth-toe scales 68-81
(mean 73.5); femoral pores 21-26 (mean 23.8); fifteenth verticil 25-30 (mean
27,7). The largest male has a snout-vent length of 90 mm., the largest female
78 mm.

The most distinctive dorsal feature of A. a. llanensis is the filigreed
to boldly marbled dorsal fields. This is a common attribute of all spec-
imens, regardless of size, but it is variable in intensity. Two old speci-
mens from the city of Camagiiey (MCZ 7277 and 84917) show the
dorsal markings most vividly, but perhaps this is due to the length of
time these lizards have been in preservative, with consequent lighten-
ing of the dorsal brown pigments, thereby rendering the dark markings
more apparent. The lateral fields are always bold and very heavily
spotted with two or three series of large ovoid or subcircular brown
spots, which at times fuse with one another. In fresh specimens, the
dorsal coloration is brownish. The median and dorsolateral lines are
faint yellow and persist in adults. The venter is a very pale blue (al-
most gray) centrally, and the lower sides and lateral-most rows of
ventrals are darker blue to greenish. The throat is pale orange (PI. 11

86

Annals of Carnegie Museum

VoL. 41

B 8 ) grading to pale yellow on the sides of the throat and on the sub-
labials. The longitudinal stripe below the lateral fields is clear greenish
in life.

Although there are no juvenile specimens, small subadults (snout-
vent lengths 61 to 65 mm., both sexes represented ) have the blotched
dorsal fields as adults do.

COMPARISONS: The blotched dorsal fields separate A. a. llanensis
from all other subspecies of A. ciuberi, including the orange-throated
races auberi, abducta, gemmea and extraria. In lateral field pigmenta-
tion, llanensis most closely resembles gemmea, although these two sub-
species are not close geographically.

REMARKS: In the serpentine savannas of Camagiiey Province, Ameiva
appears to be distinctly uncommon. Only eight specimens were secured
by myself and parties in two lengthy visits to this region. All were
taken on sandy soil or the gravelly and rocky hillocks that dot the area,
and that are frequented as well by Leiocephalus sfictigaster ophipla-
codes Schwartz. The vegetation where the Ameiva were collected was
regularly short, sparse grass.

The known range of A. a. llanensis encompasses only a small portion
of the serpentine savannas that extend from central Camagiiey to north-
western Oriente and are bounded on the north by the Sierra de Cubitas.
A. a. llanensis is well removed from Ameiva to the west, with atrotho-
rax the closest known subspecies at a distance of about 200 kilometers
airline. Despite collecting at several stations between those two regions,
no Ameiva has been observed.

ADDITIONAL SPECIMENS

It is appropriate here to mention ten specimens from Las Villas and
Camagiiey provinces which are not racially assigned at this time. These
include one subadult female (AMNH 6516) from the Sierra de Jati-
bonico near the Camagiiey-Las Villas province line, and a series of
nine specimens (AMNH 96410-96418) from 11.9 mi. N.W. Banao,
Camagiiey Province. The first specimen is long preserved, but it is
distinct from the two closest subspecies near the north coast — garridoi
on the west and llanensis on the east — although neither of these is
known to occur close to the Sierra de Jatibonico. The locality is nearer
geographically to the range of atrothorax on the south, but the lizard
does not appear to represent that subspecies. I leave it temporarily un-
assigned subspecifically.

The series from northwest of Banao was collected by myself and
party in 1960. This locality is north of the Sierra de Cubitas ( in which

1970

Ameiva auberi: I. The Cuban Subspecies

87

range Ameiva is not known to occur), and the lizards were taken along
a road through mesic lowland forest a quite different habitat from
that occupied by llanensis to the south of the Cubitas. My color notes
are not adequate to diagnose these specimens; they do differ from all
other Las Villas and Camagiiey races (except gemmea and garridoi) in
having pale yellow throats. The specimens resemble gemmea in
lateral field and dorsal patterns but seem to lack the scattered yellow
scales in the lateral fields, which is so distinctive a feature in gemmea.
Since gemmea is separated from these Camagiiey specimens by the
interposed range of garridoi, it seems unlikely that they represent the
former subspecies. They might also be assignable to garridoi, but con-
sidering the distance separating them from garridoi (about 250 kilo-
meters), I doubt that they are correctly to be considered garridoi
either. Additional material from this region will be necessary before the
status of this population can be ascertained.

Ameiva auberi sublesta, new subspecies
holotype: AMNH 96393, an adult male, from Playa Bonita, east end, Cayo
Sabinal, Camagiiey Province, Cuba, taken July 31, 1960, one of a series collected
by Ronald F. Klinikowski, David C. Leber, and James D. Smallwood. Original
number 9585.

PARATYPES: AMNH 96392, 96394-96403, same data as holotype.

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
small size (males to 81 mm., females to 67 mm. snout-vent length) high number
of fourth-toe subdigital scales, femoral pores, and scales in the fifteenth caudal
verticil; dorsum tannish gray with gray-to-cream median line which expands and
fades posteriorly; dorsolateral lines obscure to absent, lateral fields with black
restricted to isolated small blotches or obsolescent grayish vertical bars because
of intrusion of dark-brown edges of lateral fields; and throat brick red.

distribution: Known only from Cayo Sabinal off the northern coast of
Camagiiey Province, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 81 mm., tail broken and partially regenerated;
ventrals in 33 longitudinal and 10 transverse rows; fourth-toe subdigital scales 39
and 41 (total 80); femoral pores 12 and 13 (total 25); scales in fifteenth caudal
verticil uncountable. Dorsum dull tannish gray without dorsolateral light lines;
median dorsal line gray, expanding on midback and fading out posteriorly. Tail
gray above, dull bluish gray below; ventrolateral pale stripe obscure. Lateral fields
obscure, reduced to a few black or dark-gray flecks in an obsolescent brown stripe,
not set off from either dorsal or lateral pigmentation by pale lines. Limbs dull
brownish with very vague grayish markings. Belly dull bluish; throat brick red.

VARIATION : The series of 12 A. a. sublesta has the following counts : longitudinal
ventrals 32-34 (mode 33); rows of trai^verse ventrals 8 (1 specimen) -10; fourth-
toe scales 69-84 (mean 77.6); femoral pores 25-28 (mean 26.3); fifteenth verti-

88

Annals of Carnegie Museum

VoL. 41

cil 26-30 (mean 28.0). The largest male has a snout-vent length of 81 mm., the
largest female 67 mm.

The tannish gray dorsal coloration and gray-to-cream expanding and
fading median dorsal line are common features of the series. In some
specimens the lateral fields are more solidly black than in the holotype
and have about two longitudinal series of subcircular brown dots. In
other individuals of the same size these dots are coalesced and restrict
the black pigment to spots of flecks on an otherwise brown ground. At
best, the dorsolateral lines are faint and lack the distinctness of the
median line; thus the lateral fields are not outlined in pale above. Both
sexes have the throats brick red; the throats of males are more deeply
pigmented (PL 3 I 12) than are those of females (PL 3 A 10). There
seems to be no ontogenetic change in dorsal pattern and color, since a
juvenile female with a snout-vent length of 45 mm. is comparable in
these features to the adults.

COMPARISONS: Of the described races, A. a. siihlesta most closely
resembles the more western A. a. extraria. The brick-red throat of siih-
lesta (the only subspecies with this feature) differs from the dull
orange or gray tliroat of extraria, and the dorsal ground color of ex-
traria is darker ( yellowish brown to reddish ) than the tannish gray of
siihlesta. The median dorsal line that expands and fades out posteriorly
is characteristic of a complement of eastern subspecies of A. aiiheri ( and
occurs as well in extraria). It is obvious that the relationships of siih-
lesta are with the eastern Camagiiey-Oriente subspecies rather than
with races on the mainland to the west. Comparisons with these more
eastern subspecies will be made below.

REMARKS: As just uoted, A. a. suhlesia is doubtless derived from the
eastern Camagiiey subspecies about to be described. The physical re-
lationships of Cayo Sabinal to the mainland and the balance of the
northern Camagiiey cayeria have been discussed elsewhere (Schwartz,
1964a: 216). The lizards were moderately abundant on Cayo Sabinal,
and were collected in coastal scrub, along the beach, and on the mud
in and about coastal mangroves. At most localities they were accom-
panied by Leiocephalus stictigaster parasphex Schwartz.

1970

Ameiva auberi: L The Cuban Subspecies

89

Ameiva auberi citra, new subspecies

holotype: AMNH 96375, an adult male, from 2 mi. W. Playa Santa Lucia,
Camagiiey Province, Cuba, taken July 27, 1960, one of a series collected by
Ronald F. Klinikowski and James D. Smallwood. Original number 9532.

PARATYPES (all from Camagiiey Province, Cuba); AMNH 96374, 96376-96386,
same data as holotype; AMNH 96387-96391, 3 mi. W. Playa Santa Lucia, July
27, 1960, R. F. Klinikowski, J. D. Smallwood; AMNH 83749-83754, Playa Santa
Lucia, June 29, 1959, R. F. Klinikowski, B. L. Smith; AMNH 83755-83760, Playa
Santa Lucia, June 30, 1959, R. F. Klinikowski, B. L. Smith; AMNH 96369-96373,
1 mi. E. Playa Santa Lucia, July 26, 1960, D. C, Leber, A. Schwartz.

diagnosis: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 93 mm., females to 72 mm. snout-vent length), high
number of fourth-toe subdigital scales, low number of femoral pores, and mod-
erate number of scales in the fifteenth caudal verticil; dorsum dull tan, median
dorsal stripe lemon-yellow to pale tan, expanding and fading posteriorly, dorso-
lateral lines absent or restricted to anterior half of trunk in adults, lateral fields
variable from black with two or three longitudinal series of brick-red spots to
completely obsolete, even in juveniles, and throat gray to faintly orange.

DISTRIBUTION; Known only from the vicinity of Playa Santa Lucia, on the
north coast of Camagiiey Province near the Oriente border, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 92 mm., tail 190 mm.; ventrals in 33 longitudinal
and 10 transverse rows; fourth-toe subdigital scales 39 and 38 (total 77); femoral
pores 12 and 12 (total 24); 27 scales in fifteenth caudal verticil. Dorsum dull tan
with a pale-tan median dorsal line which expands and fades out posteriorly; dor-
solateral lines faint, tan, and restricted to anterior half of body. Tail dull grayish
tan dorsally, bluish ventrally; ventrolateral pale caudal stripe moderately prom-
inent on basal half of tail. Lateral fields black with a series of two brick-red spots
along the upper and lower margins, the spots restricting the interior black to a
series of irregular markings; lateral fields bordered below by a gray longitudinal
stripe. Lower sides gray with a few darker vermiculations. Limbs dull tan above
with a few vague darker marblings. Throat grayish orange; venter bluish.

variation: The series of 31 A. a. citra has the following counts: longitudinal
ventrals 31-35 (mode 32 or 33); rows of transverse ventrals 8 (2 specimens) • —
10; fourth-toe scales 67-89 (mean 77.2); femoral pores 20-29 (mean 24.3); fif-
teenth verticil 24-29 (mean 25.8). The largest male has a snout-vent length of
92 mm., the largest female 72 mm.

The most variable feature of A. a. citra is the condition of the lateral
fields. In some individuals the lateral field is a solid black band with a
minimum of brick-red “frosting” scales along its edges. At the other
extreme is the condition described for the holotype. Some specimens
also seem to lack the lateral fields entirely except for a reddish-brick-to-
brownish area along each side. None of these conditions is correlated
with size, and all occur in adults, subadults, and juveniles. The throats
vary between gray and orange (PI. 9 B 7) and are never bright.

90

Annals of Carnegie Museum

VoL. 41

Although my notes do not so indicate, presumably the brighter color is
that of the males. The median dorsal stripe is pale tan to lemon-yellow,
with the brighter shades in subadults. The belly varies from purplish
in juveniles to bluish in adults and is not bright.

COMPARISONS: A. a. citra, in having a gray-to-orange throat, is easily
differentiated from those subspecies with throats that are black or have
a black spot ( denticola, pullata, extorris, atrothorax ) , and those that have
the throat colored other than orange ( procer, paulsoni, abducta, garridoi,
sublesta). Of these, sublesta is the closest geographically. From all
described subspecies except extraria and sublesta, citra differs in having
the median dorsal line expanded and disappearing posteriorly. The
median stripe is pale tan to lemon-yellow in citra, gray to cream in sub-
lesta, and pale yellow in extraria. The dorsal ground color is yellowish
brown to reddish in extraria, tannish gray in sublesta, and dull tan in
citra.

REMARKS: A. a. citra is obviously closely allied to its neighbor on Cayo
Sabinal, A. a. sublesta; the ranges of the two are separated only by a
narrow channel. As pointed out previously, sublesta and citra show a
style of dorsal pattern that occurs from northeastern Camagiiey eastward
along much of the Oriente coast.

At and about Play a Santa Lucia, A. a. citra was moderately abundant,
occurring in coastal scrub, along the mangrove border, in small man-
groves on a rocky substrate, along the beaches (where the lizards were
distinctly less common than in habitats behind the beach ) , and the arid
woods near Punta Practices where they were taken in Eugenia thickets.

Although the range of A. a. citra is apparently circumscribed, the race
is presumably distributed farther to the east, at least along the Oriente
coast. The Oriente style of dorsum does not occur on the mainland to
the west of Playa Santa Lucia, since the specimens from the north coast
of Camagiiey north of the Sierra de Cubitas already noted possess
longitudinally lined dorsa without suppression of the dorsolateral lines
characteristic of the more central and western subspecies. The closest
named population to the west of citra is llanensis (removed 95 kilo-
meters airline), and the unnamed population north of the Sierra de
Cubitas lies at about the same distance from citra.

Despite two long visits to Playa Santa Lucia and extensive collecting
to the south of that resort, no Ameiva were seen in what appeared to be
suitable habitats. Much of the country to the south of Playa Santa Lucia
is cut-over xeric forest, a habitat these lizards often occupy.

1970

Ameiva auberi: I. The Cuban Subspecies

91

Ameiva auberi granti, new subspecies

holotype: AMNH 83784, an adult male, from Baracoa, east side, Bahia de
Miel, Oriente Province, Cuba, taken August 8, 1959, one of a series collected by
Filo Hernandez and Ronald F. KHnikowski. Original number 8178.

PARATYPES (all from Oriente Province, Cuba): AMNH 83785-83788, same data
as holotype; USNM 139779-139783, Banes, May, 1957, C. Grant; AMNH 83790,
Baracoa, August 11, 1959, native collector; USNM 29849, Baracoa, February 2,
1902, W. Palmer; MCZ 11896-11901, Baracoa, September, 1915, V. J. Rodriguez;
MCZ 11206-11207, Baracoa, date unknown, V. J. Rodriguez; AMNH 83789, 9 km.
W., 1 km. S. Baracoa, bank of Rio Bay, August 10, 1959, B. L. Smith; IB 1243-
1245, La Poa, road to La Florida, Sabanilla, Baracoa (not mapped), March 18,
1967, O. H. Garrido; AMNH 83935, Bahia de Taco, December 21, 1959, A.
Schwartz; AMNH 83946-83950, 5 mi. N.W. Bahia de Taco, December 25, 1959,
P. F. Pepe and J. R. Talada; USNM 80412, Rio Moa, Puerto Moa, March 8, 1930,
L. H. Parish and W. Perrygo; MCZ 59322-59323, ca. 9 km. S.E. Moa, July 21,
1959, R. Molina and R. Ruibal; UMMZ 98018, Preston, date unknown, C. Clark;
UIMNH 64448-64462, 64464, Banes, April, 1956, C. Grant.

ASSOCIATED SPECIMENS (all from Oriente Province, Cuba): USNM 80415, IB
17, IB 19-30, ASFS V11384-V11388, Gibara.

diagnosis: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 93 mm., females to 82 mm. snout-vent length), mod-
erate number of fourth-toe scales and femoral pores, and low number of scales in
the fifteenth caudal verticil; dorsum reddish tan in adults, dark brown in sub-
adults and juveniles, median dorsal stripe yellow, expanding and fading posteriorly
and continuing onto tail as a distinctly defined gray median caudal zone, dor-
solateral lines yellow and often irregular or fragmented, outlining lateral fields
which are extensively black with small red scallops or isolated dots, and chest and
throat black ( see, however, discussion ) .

distribution : The north coast of Oriente Province, Cuba, from Banes ( and
possibly as far west as Gibara) on the west to the vicinity of Baracoa on the east.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 93 mm., tail 166 mm., incomplete; ventrals in 32
longitudinal and 10 transverse rows; fourth-toe subdigital scales 36/-; femoral
pores 13 and 14 (total 27); 26 scales in fifteenth caudal verticil. Dorsum reddish
tan with paler tan median line which expands and fades posteriorly and becomes
a sharply distinct gray dorsal band ( about four scales wide ) on the upper sur-
face of the tail; sides of tail black, admixed with tan scales basally, the black
fading to darker gray distally; caudal ventrolateral pale stripe conspicuous basally
but becoming dim distally. Lateral fields extensively black, bounded above by a
rather irregular or slightly sinuous tan line, the fields with an upper and lower
series of red spots ( at times joined to the lateral green stripe below the lateral
fields). Sides greenish and heavily tigroid with dark gray. Limbs reddish tan
above with rather heavy gray vermiculations. Throat and chest (to about level of
fourteenth transverse row of ventrals) black; venter bluish gray, slightly brighter
on under side of tail.

variation: The series of 57 A. a. granti (including material from Gibara) has
the following counts: longitudinal ventrals 31-34 (mode 32); rows of transverse

92

Annals of Carnegie Museum

VoL. 41

ventrals 9 (1 specimen) -10; fourth-toe scales 63-92 (mean 75.9); femoral
pores 21-31 (mean 25.7); fifteenth verticil 22-29 (mean 25.1), The largest male
has a snout-vent length of 93 mm., the largest female of 82 mm.

The material from Gibara, here assigned tentatively to granti, will be
discussed separately below. Except for the Gibara specimens, A. a.
granti is fairly constant in all chromatic characters throughout its rela-
tively long range. The dark dorsum with yellow ( to tan in subadults and
juveniles) expanding and fading middorsal line, which in turn gives
rise to the sharply delineated gray upper surface of the tail, bounded
laterally by the black sides of the tail, is a common feature. The ventro-
lateral white caudal stripe is also prominent and serves to delimit sharply
the black lateral faces of the tail below. As usual, the lateral fields are
somewhat variable, but they usually have irregular edges and some sort
of scalloping or spotting. Very rarely are they solid black as in some ‘
southern Oriente populations. The tail is blue in young adults and :
juveniles.

Ameiva a. granti shows the by-now rather familiar black chest of the
black-throated races of A. auheri; however, except for the Gibara speci-
mens, the situation in male granti is diagrammatic. All (26) males with
snout-vent lengths above 67 mm. have the throat and chest solid black,
and at times the black pigment extends far posteriorly onto the belly.
Only a single subadult male (60 mm. snout-vent length) has the chest
black and a dark gray (rather than black) throat. The situation in
females is not so clear. The females fall into three categories. Only two
specimens have a completely black throat and chest ( category 1 ) . These
have snout-vent lengths of 58 mm. and 77 mm. Eight females have a
black chest and gray throat (category 2). These range in snout-vent
length from 55 mm. to 82 mm. and include the largest female. Seven
females that lack any black pigment on either chest or throat ( category
3) have snout-vent lengths of 58 mm. to 72 mm. Thus the females that
lack black pigment are bracketed on both sides by females with solid
black or gray areas. The occurrence of solid black chests and throats in
two females is unusual. Both of these were taken in December. The 58
mm. female without black pigment was collected in April. April-taken
females with gray throats and black chests include two specimens with
snout-vent lengths of 64 mm. and 72 mm.

Although I include the Gibara series with A. a. granti, I do so with
some misgivings. I have not seen Gibara lizards in life, but the excellent
lot, collected by Sr. Garrido, from that area is fresh. Sr. Garrido has
also supplied me with a color diagram and color slides of these Gibara
lizards. Dorsally, these lizards do not violate my concept of granti except

1970

Ameiva auberi: I. The Cuban Subspecies

93

that the median dorsal stripe is very pale tan, almost cream, and the dor-
solateral pale lines are not fragmented. The lateral fields agree in detail
with granti from farther east. My hesitancy in regarding them (with
complete certainty) as granti arises from the throat condition: of the
series of 18 fresh Gibara lizards, all but one have pale orange throats,
whereas that exceptional lizard (a male with a snout-vent length of
67 mm), has a black chest and very large black throat spot (but the
entire throat is not black as in granti males at 67 mm.). The single old
Gibara specimen (USNM 80415, a female, snout-vent length 64 mm.)
agrees with the fresh material in lacking a black throat.

I suspect that these lizards are either still another (orange-throated)
subspecies on the northern Oriente coast or represent extreme inter-
grades between granti to the east and an orange-throated subspecies
farther west. The nearest unquestioned granti locality is Banes, 50 kilo-
meters to the southeast. The nearest known population along the coast
to the northwest is citra at Playa Santa Lucia, about 68 kilometers dis-
tant. Since citra has a gray-to-orange throat, it is possible that the
Gibara lizards are granti x citra intergrades, but this section of Oriente
coast is so poorly known that it seems imprudent to regard them as such
at this time.

COMPARISONS: The only previously described subspecies of A. auberi
with solid black chests and throats are denticola, far to the west in Pinar
del Rio Province, and per adust a in Las Villas Province. A. a. denticola
and A. a, granti differ in many features, but the most striking are the
expanding and fading median dorsal line and the gray band, bordered
by black sides, on the upper surface of the tail in granti. A. a. granti
differs from peradusta in lacking the black cheeks and neck and com-
plete venter, and in being patterned dorsally. These two subspecies also
differ markedly in mean number of fifteenth verticil scales (25.1 in
granti, 28.9 in peradusta); granti ranks low and peradusta high in this
character. The dorsal chromatic details (coupled with the darker red-
dish tan dorsum) will also serve to differentiate granti from all previ-
ously described subspecies. The dorsal coloration and pattern of extraria
is somewhat comparable to that of granti, but the inconspicuous and
brown lateral fields in extraria are distinct from the conspicuous lateral
fields of granti.

REMARKS: A. a. granti is based in part on the long series collected by
Major Chapman Grant at Banes. It is indeed a pleasure to associate his
name with such a very distinctive subspecies, of whose existence he was
well aware ( Grant, 1960 ) . Recently taken specimens have been secured

94

Annals of Carnegie Museum

VoL. 41

on diente de perro limestone with some trees and shrubs, Bahia de
Miel), in shoreline vegetation both along the coast (northwest of Bahia
de Taco) and in mesic riverine growth (Rio Bay).

To the south, much of the range of granti is bordered by the Oriente
massifs, and these mountain masses separate granti from the races on the
soulh coast of the province, although I suspect that granti intergrades
with the easternmost southern Oriente subspecies between Baracoa and
Cabo Maisi. The granti from La Poa indicate that these lizards penetrate
the northern slopes and foothills of the Cuchillas de Toa southwest of
Baracoa to some extent. There are no specimens from the serpentine
savannas in the region about Holguin; it would be most interesting to
learn what subspecies occurs in this interior region — either granti
(expected on the basis of proximity of range) or llanensis (expected
because of similarity and continuity of habitat ) . The Gibara lizards may
represent a marginal sample, in part intergradient with granti, of a
distinctive subspecies inhabiting these interior Oriente plains.

Ameiva auberi sabulicolor, new subspecies

holotype: AMNH 83941, an adult male, from 2.8 mi. E. Imias, Oriente Prov-
ince, Cuba, taken August 21, 1959, one of a series collected by Ronald F. Klini-
kowski. Original number 8506.

PARATYPES (all from Oriente Province, Cuba); AMNH 83940, 83942-83943,
same data as holotype; AMNH 83951, 4.6 mi. N. Cajobabo, December 28, 1958,
J. R. Talada; AMNH 83938-83939, 4 mi. W. Baitiquiri, August 21, 1959, R. F.
Klinikowski and B. L. Smith; IB 59-60, ASFS VI 1389, Baitiquiri, April 5, 1965,
O. H. Garrido; AMNH 96457-96465, 33.8 mi. S.E. Guantanamo, August 27, 1960,
R. F. Klinikowski and D. C. Leber; AMNH 96466-96468, 33.8 mi. S.E. Guan-
tanamo, August 30, 1960, D. C. Leber; UMMZ 110184, United States Naval Base,
February 20, 1954, T. M. Uzzell, Jr.; UMMZ 110203 (2 specimens). United
States Naval Base, May 16, 1954, R. E. Etheridge; MCZ 68920, United States
Naval Base, date unknown, R. V. Lando; UF 21879, United States Naval Base,
October 18, 1965, J. C. Dickinson, Jr., and T. H. Patton; UIMNH 64431-64447,
United States Naval Base, December, 1957, C. Grant; USNM 81665-81667, United
States Naval Base, Cuzco Valley, August 16, 1930, P. Bartsch; USNM 59137-
59140, United States Naval Base, Hospital Point, date unknown, J. B. Henderson
and P. Bartsch; USNM 59166, United States Naval Base, Fisherman’s Point,
March 14, 1917, J. B. Henderson and P. Bartsch; USNM 81654, north side, Bahia
de Guantanamo, August 14, 1930, P. Bartsch; USNM 59136, Guantanamo, March
18, 1917, J. B. Henderson and P. Bartsch; USNM 59163-59165, Guantanamo,
date unknown, J. B. Henderson and P. Bartsch; USNM 139784-139787, Guan-
tanamo, December, 1957, C. Grant; USNM 59192, Guantanamo Bay, date un-
known, J, B. Henderson and P. Bartsch; USNM 81821, Boqueron, August 19,
1930, P. Bartsch.

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
small size (males to 80 mm., females to 66 mm. snout-vent length), low number

1970

Ameiva auberi: L The Cuban Subspecies

95

of fourth-toe subdigital scales, high number of femoral pores, and low number of
scales in. the fifteenth caudal verticil; dorsum pale tan (sandy) to pale brown
with median line pinkish anteriorly and gray posteriorly, fading and expanding
posteriorly and continuing onto tail as a distinctly defined median gray caudal
zone, dorsolateral lines tan and more or less straight, outlining lateral fields which
are solid black (rarely with only a bare indication of reddish stippling along their
margins), and chest and throat black.

distribution: The southeastern coast of Oriente Province, Cuba, from the
vicinity of the city of Guantanamo and Boqiieron, east to Cajobabo, and probably
as far as Cabo Maisi.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 77 mm., tail 150 mm., partially regenerated; ven-
trals in 33 longitudinal and 10 transverse rows; fourth-toe subdigital scales 34
and 35 (total 69), femoral pores 14 and 15 (total 29), 25 scales in fifteenth caudal
verticil. Dorsum very pale sandy tan with a pale (pinkish anteriorly, gray pos-
teriorly ) median stripe, which fades and expands posteriorly and becomes a
sharply distinct bluish gray dorsal zone ( about four scales wide ) on the upper
surface of the tail; sides of tail black, admixed with grayish scales basally, the
black fading to darker gray distally; ventrolateral caudal pale stripe prominent
on unregenerated portion of tail. Lateral fields solid black, bounded above and
below by fairly prominent pale tan (above) or pale gray (below) lines, and with
a faint stippling of reddish-to-tan scales, especially along the posterior margins
of the fields. Sides grayish, marbled with darker gray vermiculations. Limbs sandy
gray above with vague darker gray marblings. Throat, chest (to about level of
fourteenth transverse row of ventrals), and anterior faces of forelimbs solid black.
Venter pale bluish gray, paler on underside of tail.

variation: The series of 47 A. a. sabulicolor has the following counts: longi-
tudinal ventrals 30-34 (mode 32); rows of transverse ventrals 10; fourth-toe scales
62-81 (mean 72.8); femoral pores 23-31 (mean 26.7); fifteenth verticil 22-29
(mean 25.6). The largest male has a snout-vent length of 80 mm., the largest
female 66 mm.

A. a. sabulicolor is one of a series of three races along the southern
coast of Oriente Province between Cabo Maisi and Cabo Cruz. Its habi-
tation of the most arid portion of this southern coast is reflected in its
pallid coloration, varying from sandy or pale tan to pale brown. In
juveniles and subadults the dorsum is somewhat darker brown than in
full adults. The pink, grading to gray, median dorsal stripe and the tan
dorsolateral stripes are constant features. In young specimens the lateral
fields are solid black, and in adults there is only the barest indication of
reddish-to~tan stippling along the very edges of the fields, especially in
the posterior half. At times the lower sides have an additional longitudi-
nal zone of tan paralleling the line that borders the lateral fields ven-
trally.

The black chest and throat are once again variable features. Specimens
of sabulicolor were divided into three categories : ( 1 ) black chest and

96

Annals of Carnegie Museum

VoL. 41

throat; (2) black chest and some black on throat; (3) pale chest and
throat. Of 38 males, 19 ( snout-vent lengths 55 mm. to 80 mm. and thus
including the largest male) fall into category 1. Category 2 includes
eleven males with snout-vent lengths of 52 mm. to 74 mm., and category
3 includes eight males with snout-vent lengths from 50 mm. to 74 mm.
Most (29 of 35) males with snout-vent lengths greater than 55 mm. have
black chests and some indication of black on the throats. No females fall
into category 1. In category 2 there are four females (snout-lengths
between 50 mm. and 65 mm. ) and in category 3 ten females ( snout-vent
lengths 52 mm. to 67 mm.). The smallest female has a black chest and
some black on the throat, whereas many of the larger females ( including
the largest) lack any black pigment. The situation is inconclusive.

COMPARISONS: A. a. sabulicolor most resembles the races granti and
citra from the northern coast of Camagiiey and Oriente provinces. A. a.
sabulicolor differs from both these northern subspecies in being much
paler dorsally, and it differs additionally from citra in having a black
throat and chest rather than a gray-to-orange throat. Sabulicolor differs
as well in having solid black lateral fields (with at best some faint red-
dish stippling posteriorly), whereas both citra and granti regularly
have some markings within the field itself. Both extraria and sublesta
agree with sabulicolor in having an expanding and fading median dorsal
line, but the throat in extraria is gray-to-diffuse orange, and brick-red in
sublesta.

The southeastern coast of Oriente is typical of the arid tropics, and
lies in the rain shadow of the Sierra de Burial and the Cuchillas de Toa
and associated uplands. Although A. a. sabulicolor is primarily coastal,
it does occur inland as far as the city of Guantanamo and north of the
village of Cajobabo in the extreme eastern portion of its range. This
latter locality is on the southern slopes of the Sierra de Burial and,
although somewhat more mesic than the coast itself, is nonetheless arid.
The elevation is probably less than 1000 feet. Ameiva was taken along
the coast in such habitats as coastal scrub and arid woods, but in general
the race appears not to be exceptionally abundant. The series from the
United States Naval Base is the result of many years’ collecting, and no
single collector has secured large numbers of specimens there.

The range of A. a. sabulicolor is separated from that of A. a. granti by
the interior Oriente massifs. As noted previously, I suspect that sabuli-
color inhabits the xeric Cabo Maisi, and intergradation between it and
granti occurs between Cabo Maisi and the region about Baracoa. Since
the north Oriente coast is mesic, it seems likely tliat sabulicolor (which

1970

Ameiva auberi: I, The Cuban Subspecies

97

inhabits the xeric southern coast) occurs on Cabo Maisi, rather than
granti (which occupies the mesic north coast).

Ameiva auberi ustulata, new subspecies

holotype: AMNH 83778, an adult male, from 7.8 mi, E. Siboney, Oriente
Province, Cuba, taken July 21, 1959, one of two collected by Ronald F. Klini-
kowski and Barton L. Smith. Original number 7779.

PARATYPES (all from Oriente Province, Cuba): AMNH 83779, same data as
holotype; USNM 29800, San Luis, Febmary 15, 1902, W, Palmer; USNM 29796-
29799, El Cobre, February, 1902, W. Palmer; AMNH 83763-83764, 6.5 km. S.
Palma Soriano, July 11, 1959, R. F. Klinikowski; AMNH 96439, 4 km. N. Santiago
de Cuba, August 12, 1960, R. F. Klinikowski; IB 4-9, ASFS V11390-V11391,
Rancho Club, 4 km. N. Santiago de Cuba, March 25, 1965, O. H, Garrido; AMNH
20385-20387, Santiago de Cuba, December, 1920, G. Bosnia; MCZ 6920, 84920,
Santiago de Cuba, February 22, 1904, W. Robinson; UMMZ 90730 (2 specimens),
Santiago de Cuba, Morro Castle, February, 1939, A. Vanderhorst; AMNH 83762,
Aguadores, July 10, 1959, R. F. Klinikowski; AMNH 83765-83766, Aguadores,
July 11, 1959, R. F. Klinikowski; AMNH 83767-83775, Aguadores, July 15, 1959,
R. F. Klinikowski, B. L. Smith; AMNH 96440-96450, Aguadores, August 13,
1960, R. F. Klinikowski,' D. C. Leber, J. D. Smallwood; AMNH 83776 (23 speci-
mens), Playa Juragua, 3.7 mi. E. Siboney, July 16, 1959, R. F. Klinikowski, A.
Schwartz, B. L. Smith; AMNH 96451-96456, Playa Juragua, 3.7 mi. E. Siboney,
August 24, 1960, R. F. Klinikowski, D. C. Leber; AMNH 83780-83781, 18.2 km.
E. Siboney, July 25, 1959, R. F. Klinikowski, B. L. Smith; AMNH 83782, 26.6
km. E. Siboney, July 25, 1959, R. F. Klinikowsky; AMNH 96419-96426, 9.4 mi.
W. Laguna de Baconao, August 9, 1960, R. F. Klinikowski, J. D. Smallwood;
AMNH 83783 (6 specimens). Laguna de Baconao, 21.8 mi. E Siboney, August 3,

1959, R. F. Klinikowski, A. Schwartz, B. L. Smith; AMNH 96437-96438, 0.7 mi.
E. Laguna de Baconao, August 11, 1960, J. D. Smallwood; AMNH 96432-96436,
1.8 mi. E. Laguna de Baconao, August 11, 1960, R. F. Klinikowski, J. D. Small-
wood; AMNH 96431, 2.8 mi. E. Laguna de Baconao, August 11, 1960, J. D.
Smallwood; AMNH 96430, 3.7 mi. E. Laguna de Baconao, August 11, 1960, J. D.
Smallwood; AMNH 96427-96429, 1.6 mi. NE Laguna de Baconao, August 11,

1960, R. F. Klinikowski, J. D. Smallwood.

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
small size (males to 78 mm., females to 68 mm. snout-vent length), low number
of fourth-toe subdigital scales, high number of femoral pores, and moderate
number of scales in the fifteenth caudal verticil; dorsum brown with a yellow-to-
gray median line, not (or only slightly) expanding posteriorly, remaining distinct,
and forming a narrow dorsal stripe on the base of the tail which expands to form
a broad dorsal caudal stripe, bounded laterally by black sides; dorsolateral lines
gray to tan and prominent, outlining the solid black lateral fields above; lateral
fields outlined below by a prominent tan-to-pale-gray lateral stripe, and chest
and throat black.

DISTRIBUTION: Southern coast of Oriente Province between the Bahia de San-
tiago and the Bahia de Guantanamo, and occurring inland to the northwest of
Santiago de Cuba at San Luis, El Cobre, and Palma Soriano, Cuba.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements

98

Annals of Carnegie Museum

VoL. 41

and counts: snout-vent length 78 mm.; tail 126 mm., distal half regenerated; ven-
trals in 33 longitudinal and 10 transverse rows; fourth-toe subdigital scales 41 and
40 (total 81); femoral pores 13 and 14 (total 27); 27 scales in fifteenth caudal
verticil. Dorsum brown with a prominent yellow median line which becomes gray
at midbody, expands only slightly posteriorly, and forms a narrow, gray stripe
(involving three scales) on the base of the tail, and then expands to form a
rather broad, gray dorsal stripe on the unregenerated portion of the tail; caudal
stripe bordered laterally by black sides of tail, which are basally much marked
with tan to brown; ventrolateral caudal stripe white and very bold on unregen-
erated portion of tail. Lateral fields solid black bordered above by tan dorsolateral
stripes and below by pale gray lateral stripes; a few small circular reddish spots
in the lateral fields above the forelimb insertion and above the groin. Limbs
brown above with vague gray marbling; posterior face of thighs with a bold
white horizontal stripe bordered both above and below by coarse black-to-dark-
gray vermiculations. Belly very pale blue; throat black, but not completely so.

VARIATION: The series of 85 A. a. iistulata has the following counts: longitudinal
ventrals 30-36 (mode 32); rows of transverse ventrals 10; fourth-toe scales 64-85
(mean 72.8); femoral pores 22-32 (mean 26.2); fifteenth verticil 22-31 (mean
25.8). The largest male has a snout-vent length of 78 mm., the largest female
68 mm.

The description of the coloration and pattern of the holotype suffices
for most of the series of A. a. ustiilata, which, considering the large
amount of material, is remarkably constant. The median dorsal line in
many specimens shows no posterior expansion and is merely a median
dorsal line, prominent and of the same width throughout its entire length.
If there is posterior expansion, it is never so striking as in the races to
the north and east. The lateral fields are solid black and very seldom
show any included markings, the type being unusual in this respect. The
two specimens from south of Palma Soriano have the lateral fields with
scalloped edges. This may be the result of the influence of the next sub-
species to the west (although the Palma Soriano specimens are not
coastal ) , or the scalloped edges may represent the effect of some interior
subspecies, as yet unsampled, from the valley of the Rio Cauto. The
Palma Soriano specimens were taken from the rather low pass that
crosses the Sierra Maestra between Palma Soriano and Santiago de
Cuba. The posterior thigh markings in the holotype of ustulata are not
shown so boldly in all specimens, but there is always at least a fairly
clear white transverse bar with some darker markings above and below.
There is little or no difference in pigmentation and pattern between
juveniles and adults, although all pattern elements seem a bit better
defined in young individuals.

When all specimens of ustulata are categorized, the following data on
black throats and chests may be summarized. In category 1, which

1970

Ameiva auberi: I. The Cuban Subspecies

99

includes all specimens with black throats and chests, are included 29
males with snoutwent lengths from 55 mm. to 74 mm. In category 2
( chest black or with small black throat patch ) are 15 males with snout-
vent lengths from 55 mm. to 78 mm., and thus including the holotype —
the largest male. In category 3 ( no black on throat or chest ) are 15 males
with snout-vent lengths of 52 mm. to 74 mm. Most males (39 of 45)
with snout-vent lengths in excess of 60 mm. have some black on the
chest or throat. A single female ( snout-vent length 58 mm. and thus not
the largest female) falls into category 1. Only three females (snout-vent
lengths 54 to 67 mm. ) fall into category 2, whereas 26 females ( snout-
vent lengths 48 to 68 mm. ) lack any black on chest or throat. In general,
only rarely do female ustulata achieve any black pigmentation on throat
or chest, although occasional specimens do so.

COMPARISONS: A. a. ustulata is obviously related to the more eastern
sahtilicolor but differs from it in being darker dorsally (pale tan and
pale brown versus brown), in having the median dorsal line not fading
and expanding posteriorly, and in having a rather clear-to-bold trans-
verse thigh stripe. The condition of the median dorsal line in ustulata
will differentiate it from the other eastern subspecies {granti, citra, sub-
lesta ) and the more western extraria, which have an expanding and fad-
ing median stripe. Of these, extraria, stihlesta, and citra do not have
black throats and chests. In some ways ustulata is rather like the far
western paulsoni, but the black throat, gray median stripe, and black-
sided and gray-surfaced tail of the former contrast with the cream throat,
yellow lines, and unicolor tail of the latter. In the east, the shade of the
median dorsal line differentiates citra (lemon-yellow to pale tan) from
sabulicolof ( pinkish to tan ) and ustulata ( pale yellow to gray ) , but not
from granti, which also has the line pale yellow to gray.

REMARKS: A. a. ustulata is one of the few subspecies of A. auberi that
occurs any significant distance inland and at any appreciable height.
Comments have already been made on the specimens from Palma Sori-
ano. The specimens from San Luis and El Cobre are from interior
localities, and El Cobre is in a moderately high basin in the extreme
eastern end of the Sierra Maestra (or in the Sierra del Cobre as it is
locally known). San Luis lies in the Valle Central between the valley
of the Rio Cauto to the west and the Cuenca de Guantanamo to the
east The elevation of San Luis is only 171 meters (Marrero 1951:618).
The Valle Central lies north of the Sierra Maestra (and north of the
Sierra de Boniato and the Sierra de la Gran Piedra, which are Sierra

100

Annals of Carnegie Museum

VoL. 41

Maestra affiliates) and south of the Sierra de Nipe and the eastern '
Alturas de Baracoa.

The major portion of the range of A. a. ustulata lies to the south of ^
the Sierra de la Gran Piedra and in its coastal rain shadow. The coast
between the Bahia de Santiago and the Bahia de Guantanamo is arid,
but somewhat less so than the coast to the east of Guantanamo. Ground
lizards were collected in coastal scrub and in moderately moist gallery ’
forest along the intermittent water courses near the coast. At Playa
Juragua, Aguadores, and Laguna de Baconao, Ameiva was abundant in
shady woods, as well as along the borders of mangroves at the last-
named locality.

The Bahia de Guantanamo separates the ranges of A. a. ustulata and
A. a. sahulicolor, and the Bahia de Santiago separates the range of A. a.
ustulata from the race to the west. The effect of these two bays upon
populations of reptiles along the southern Oriente coast has been demon-
strated by Schwartz (1960) in Leiocephalus raviceps and by Thomas '
and Schwartz ( 1966 ) in Sphaerodactijlus torrei. It is not surprising that
A. atiberi, which also has an essentially coastal distribution in this area, |
has differentiated into three distinct subspecies in southern Oriente.

Ameiva auberi hardyi, new subspecies |

holotype: USNM 138468, an adult male, from Ocujal, Oriente Province, Cuba, ^

taken August 31, 1956, one of a series collected by Jerry D. Hardy. j

PARATYPES (all from Oriente Province, Cuba): USNM 138468-138471, same
data as holotype. AMNH 83761, Cabo Cruz, July 5, 1959, B. L. Smith; IB 10-14 ^

ASFS V11392-V11394, Verreon, Cabo Cruz, March 31, 1965, O. H. Garrido;
USNM 138473, 5 mi. N. Cabo Cruz, September 6, 1956, J. D. Hardy; USNM
81678, Rio Puercos, August 29, 1930, P. Bartsch; USNM 138472, Jucaral (not

mapped), September 6, 1956, J. D. Hardy; AMNH 83777, 18.6 mi. W. Santiago I

de Cuba, July 17, 1959, B. L. Smith.

DIAGNOSIS: A Cuban subspecies of A. auheri characterized by a combination of
moderate size (males to 93 mm., females to 76 mm.), high number of fourth-toe
subdigital scales, moderate number of femoral pores, and low number of scales |
in the fifteenth caudal verticil; dorsum tan with pale-yellow-to-cream dorsal and
dorsolateral lines, the median line expanding (although often not broadly) pos-
teriorly and remaining distinct, lateral fields black with prominent red scallops
and often with an included series of red spots, and throat orange.

DISTRIBUTION: Southwestern coast of Oriente Province, Cuba, from the vicinity
of Cabo Cruz on the west presumably to the Bahia de Santiago on the east, but
specimens are lacking from the area immediately to the west of the bay.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 93 mm., tail 201 mm.; ventrals in 32 longitudinal
and 10 transverse rows; fourth-toe subdigital scales 39 and 38 (total 77); femoral ,
pores 12 and 12 (total 24); 24 scales in the fifteenth caudal verticil. Dorsum |

1970

Ameiva auberi: I. The Cuban Subspecies

101

tan with a prominent median cream dorsal line, narrow anteriorly and wider (but
not extremely so) posteriorly, extending onto the upper surfaces of the tail as a
narrow dorsal blue stripe basally, and somewhat broader (four scales) posteriorly;
a pair of dorsolateral tan lines above the lateral fields. Lateral fields extensively
black and with large, reddish scallops ventrally, and irregular, reddish markings
along the upper border; lower border of lateral fields with a faint grayish longi-
tudinal line. Sides of tail black, sharply limiting the blue upper surface of the tail;
ventrolateral caudal pale stripe prominent on the basal half of the tail. Lower
sides extensively dark gray with some paler gray punctuations. Venter blue,
under side of tail paler blue; throat very pale blue-gray in preserved specimen,
but presumably pale orange in life. Upper side of limbs tan with little indication
of darker marbling; no white transverse stripe along posterior border of thighs.

VARIATION: The series of 15 A. a. hardyi has the following counts: longitudinal
ventrals 30-34 (mode 33); rows of transverse ventrals 10; fourth-toe scales 73-83
(mean 77.8); femoral pores 21-30 (mean 24.8); fifteenth verticil 24-29 (mean
25.6). The largest male has a snout- vent length of 93 mm., the largest female
76 mm.

I have seen but two examples of A. a. hardyi in life. One of these
(AMNH 83777) is a juvenile on which I have no color data. The other
(AMNH 83761) is from the extreme western end of the range at Cabo
Cruz. Since the holotype and paratypes from between Cabo Cruz and
west of Santiago are not black-throated, and since neither of the two
specimens I have collected are black-throated (their throats were pale
orange) I am assuming that all material from this stretch of coast is
chromatically identical. Color slides and diagrams of specimens from
Verreon, supplied by Sr. Garrido, confirm that the lizards at that locality
are orange-throated. In general, material from farther west (Cabo Cruz
and Rio Puercos) has the median dorsal line more expanded than most
specimens from the type locality, but even lizards from the Ocujal area
have the dorsal line more expanded posteriorly than do specimens of
ustulata. The lateral fields invariably are marked with reddish along
both borders; one individual has an included longitudinal series of red-
dish dots in addition to the marginal scallops. Another specimen has a
rather restricted series of ventral scallops. Even in the smallest specimen
(snout-vent length 41 mm.) there are red markings within the lateral
fields. No specimen shows any indication of black on the throat or chest.

COMPARISONS: The combination of orange throat and expanded median
dorsal line separates hardyi from all other subspecies of A. auberi except
possibly citra on the north coast of Camagiiey and extraria on the
northern cayeria in Las Villas. Ameiva a. hardyi may be distinguished
from the latter in having all dorsal pattern elements much more distinct,
in having a darker dorsal ground color, and in having prominent lateral
fields. Ameiva a. hardyi differs from citra in having prominent lateral

102

Annals of Carnegie Museum

VoL. 41

fields and the tail with black sides restricting a dorsal blue stripe. Hardyi
differs in its larger size from its neighbor ustulata on the east (even
though the series of ustulata is six times the size of the hardyi sample,
none of the former in any way approaches the larger size of the 15
specimens of hardyi) in having an orange rather than black throat, and
in other details of dorsal pattern and pigmentation.

REMARKS: A. a. hardyi is limited in distribution to the xeric coast south
of the Sierra Maestra system between the Cabo Cruz region and Santi-
ago de Cuba ( the eastern portion of the range is hypothetical since there
are no specimens from close to the Bahia de Santiago). At the type
locality. Cooper (1958:13) reported securing specimens of A. auberi
in the southern foothills of the Sierra Maestra as he and his party climbed
toward Pico Turquino. The Ocujal region itself is arid. The AMNH
specimen from Cabo Cruz was collected in a beach scrub habitat, and
the lizard from 18.6 mi. west of Santiago de Cuba was from a similar
situation. Many of the specimens of A. a. hardyi are from the vicinity of
the type locality and were collected through the efforts of Jerry D.
Hardy, for whom the subspecies is named.

Although there are no specimens of hardyi from close to the Bahia de
Santiago to the west, it is possible that hardyi and ustulata intergrade
somewhere about the western side of the bay. As noted in the discussion
of ustulata, the two specimens from south of Palma Soriano differ some-
what from most ustulata. One, the largest specimen of that race, is a
black-chested male. Both have extensive reddish markings in the lateral
fields. If these specimens are hardyi x ustulata intergrades, then hardyi
must somewhere avoid the eastern periphery of the Sierra Maestra to
reach the Palma Soriano region. As previously stated, it is also possible
that these two lizards are intergradient between ustulata and some sub-
species to the north, rather than being hardyi x ustulata intergrades.

Ameiva auberi secta, new subspecies

holotype: AMNH 82997, an adult male, from Playa de Rocas, between Bibi-
jagiia and Jiicaro, Isla de Pinos, Habana Province, Cuba, taken July 24, 1958,
one of a series collected by George R. Ziig. Original number 6095.

PARATYPES ( all from the Isla de Pinos, Habana Province, Cuba ) : AMNH
96299-96302, same data as holotype; AMNH 82996, just W. Nueva Gerona, east
base. Sierra de las Casas, July 22, 1958, G. R. Zug; KU 55139, Nueva Gerona,
June 6, 1953, T. H. Eaton; MCZ 10919, 10922-10923, Nueva Gerona, 1915, T.
Barbour and W. S. Brooks; USNM 27982-27983, Nueva Gerona, July 6, 1900, W.
Palmer and J. H. Riley; USNM 28041, Nueva Gerona, July 8, 1900, W. Palmer
and J. H. Riley; KU 55145-55146, 3 mi. E. Nueva Gerona, June 6, 1953, T. H.
Eaton; AMNH 82994 (5 specimens), Bibijagua, July 20, 1958, A. Schwartz, G. R.

1970

Ameiva auberi: I. The Cuban Subspecies

103

Ziig; AMNH 82998, Playa de Herradura, just N. Bibijagua, July 24, 1958, G. R.
Zug; AMNH 82988 (2 specimens), 2 mi. S. Bibijagua, July 8, 1958, A. Schwartz,
G. R. Zug; AMNH 78352-78355, 2 km. N. Santa Fe, December 24, 1957, E. B.
Erickson, A. Schwartz, G. R. Zug; AMNH 78356-78361, Santa Fe, December 26,
1957, E. B. Erickson, W. H. Stitzell; ASF'S V11395-V11397, Santa Fe, April 13,
1965, O. H. Garrido; AMNH 78362-78365, 1.5 mi. W. Santa Fe, December 26,
1957, E. B. Erickson, W. H. Stitzell, G. R. Zug; AMNH 78366-78373, east shore,
Ensenada de la Siguanea, nr. Siguanea, December 27, 1957, E. B. Erickson, A.
Schwartz, W. H. Stitzell, G. R. Zug; AMNH 82989, east shore, Ensenada de la
Siguanea, nr. Siguanea, July 9, 1958, G. R. Zug; IB 142, IB 153-155, Hotel Colony,
Siguanea, April 16, 1965, O. H. Garrido; MCZ 10949-10951, Los Indios, 1912,
G, A. Link; AMNH 82990 (10 specimens), Playa de Buena Vista, July 12, 1958,
A. Schwartz, G. R. Zug.

ASSOCIATED SPECIMENS: Isltt de Pinos, 6 (CM 877-878, 881-882, 884, 886);
?Isla de Pinos, 1 (CM 1088).

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 92 mm., females to 77 mm. snout-vent length), low
number of fourth-toe subdigital scales, moderate number of femoral pores, and
very low number of scales in the fifteenth caudal verticil; dorsum brown to red-
dish brown with prominent median and dorsolateral longitudinal yellow lines,
lateral fields usually extensively black with scattered and isolated red dots, and
throat black.

DISTRIBUTION: The northern two-thirds of the Isla de Pinos, north of the
Cienaga de Lanier.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 82 mm., tail 212 mm,; ventrals in 32 longitudinal
and 10 transverse rows; fourth-toe subdigital scales 35 and 36 (total 71); femoral
pores 12 and 13 (total 25); 26 scales in the fifteenth caudel verticil. Dorsum
reddish brown with median and dorsolateral lines yellow and very bold. Lateral
fields black with two longitudinal series of discrete red dots, and bordered below by
a bold, grayish yellow line. Lower sides pale bluish gray with dark vermicula-
tions. Limbs reddish brown above, heavily marbled with very dark brown; a
clear white transverse line on the posterior face of the thigh, bordered above and
below by wide black bands. Throat black, with very httle black pigment extend-
ing onto first two or three transverse rows of ventrals. Tail grayish brown above,
with a pair of whitish dorsolateral lines on its basal half; sides of tail dark gray
or black, not clearly demarcating a dorsal paler longitudinal band; ventrolateral
white caudal stripes distinct on basal half of tail. Belly dull grayish blue, under
side of tail dark blue-gray.

VARIATION: The series of 57 A. a. secta has the following counts: longitudinal
ventrals 29-34 (mode 31); rows of transverse ventrals 8 (5 specimens) -10;
fourth-toe scales 63-86 (mean 69.6); femoral pores 21-31 (mean 24.8); fifteenth
verticil 22-28 (mean 24.9). The largest male has a snout-vent length of 92 mm.,
the largest female 77 mm.

The dorsal color of A. a. secta varies from brown to rich reddish
brown, and the median and dorsolateral lines are yellow and prominent
at all ages. The lateral fields are somewhat variable as is customary.

104

Annals of Carnegie Museum

VoL. 41

They are always extensively black and usually have some included red
markings, often, as described in the type, a series of discrete red spots.
In some individuals the lateral fields are almost solid black, with some
faint reddish stippling along the upper and lower margins, and some
have the margins of the fields scalloped with reddish. The ventrolateral
caudal pale stripes are bold, but only occasional individuals show any
indications of white dorsolateral caudal stripes as does the holotype.
There is often a bold black lateral line on the tail, delimited below by
the ventrolateral caudal stripe and above by the dorsolateral caudal
stripe (if present). The bold white transverse thigh stripe, surrounded
on both sides by black patches, is fairly constant.

The series of A. a. secta was separated into three categories on the
basis of chest and throat pigmentation: (1) black chest and throat; (2)
black spot on chest and small black patch on throat; (3) no black on
chest or throat. Of 32 males, 17 ( snout-vent lengths 54 mm. to 92 mm.,
including the largest and smallest males) fall into category 1. Eleven
males (snout-vent lengths 61 mm. to 91 mm.) are grouped in category
2, and only four males (snout-vent lengths 57 mm. to 65 mm.) are in
category 3. Thus, most males (28 of 32), with snout-vent lengths of 54
mm. or more, have the chest black and some black on the throat. Of 27
females, two (snout-vent lengths 56 mm. and 75 mm.) are as heavily
pigmented on tliroat and chest as most of the males, and fall into cate-
gory 1. Four females (snout-vent lengths 57 mm. to 77 mm. and thus
including the largest female ) are included in category 2. Twenty females
with snout- vent lengths from 52 mm. to 67 mm. lack any black on throat
or chest and are in category 3. Once again, the situation with black
throats and chests is anomalous. It can be stated at least that even small
females at 56 mm. may achieve the totally black throat and chest of most
males, but many larger females seem not to have reached this condition.

COMPARISONS: Of the Cuban mainland races, A. a. secta most closely
resembles (as might be expected) A. a. paulsoni, which is geographic-
ally the most closely adjacent race. The two may be distinguished from
each other by the cream throat and lateral field pattern (black with
ventral margin scallops only) of paulsoni in contrast to the black throat
and more variably patterned lateral fields of sectq. The close relationship
of the two races is shown by the low means of fourth toe scales ( 68.6 in
paulsoni, 69.6 in secta ) .

Ameiva auberi secta is unique among Cuban subspecies in having a
boldly lined brown or reddish brown dorsum without an expanding
median dorsal line, and a black throat.

1970

Ameiva auberi: I. The Cuban Subspecies

105

REMARKS: A. a. secta is widespread in many habitats, both interior
and coastal, on the northern two-thirds of the Isla de Pinos. It was col-
lected near Santa Fe on Malpais gravel in pinelands. coastally along
beaches among Ipomaea (near Siguanea), in grass and cut mangrove
trash (Buena Visita), and in broadleaf forest (Bibijagua). On two
occasions we observed birds preying upon A. a. secta. Both episodes
were diurnal and occurred at Playa de Buena Visita in cut-over man-
groves adjacent to the coast. The predators were the Sparrow Hawk
(Falco sparverius sparverioides) and Cuban Pygmy Owl (Glaucidium
siju vittatum). In both cases, the birds swooped down to the ground
and returned to perch clutching a subadult Ameiva.

The relationships of A. a. secta with the southern Isla de Pinos sub-
species will be discussed below.

Ameiva auberi marcida, new subspeeies

holotype: AMNH 82991, an adult male, from Jacksonville, Isla de Pinos,
Habana Province, Cuba, taken July 14, 1958, one of a series collected by Albert
Schwartz and George R. Zug. Original number 5840.

PARATYPES (all from the Isla de Pinos, Habana Province, Cuba): AMNH
96281-96298, same data as holotype; AMNH 78374-78375, Puerto Frances,
December 30, 1957, E, B. Erickson, A. Schwartz, W. H. Stitzell, G. R. Zug;
AMNH 82999 (9 specimens), Caleta Grande, July 26, 1958, A. Schwartz, G. R.
Zug; AMNH 82992, 7 mi. E. Jacksonville, July 15, 1958, G. R. Zug; AMNH
82993 (8 specimens), Caleta de Carapachibey, July 18, 1958, A. Schwartz, G. R.
Zug; IB 147-152, 162-163, CM 40647-40648, ASFS V11398-V11399, NMC 10064
(2 specimens), Guayacanal, April 17, 1965, O. H. Garrido; AMNH 82995 (17
specimens), Punta del Este, July 21, 1958, A. Schwartz, G. R. Zug; AMNH 82987
(14 specimens), Paso de Piedras, ca. 20 km, S.W. Santa Fe, July 2, 1958, A.
Schwartz, G. R. Zug; IB 146, Cayo Piedras, April 14, 1965, O. H. Garrido.

diagnosis: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 87 mm,, females to 77 mm. snout-vent length), moderate
number of fourth-toe subdigital scales and femoral pores, and low number of
scales in the fifteenth caudal verticil; dorsum brown to tan with heads gray, me-
dian and dorsolateral lines tan, the median line very obscure, lateral fields black
with brown included edging and consequent reduction of the black pigment, and
throats pale blue (or with some black pigment; see discussion).

distribution: The southern third of the Isla de Pinos, south of the Cienaga
de Lanier, but including the Paso de Piedras which joins the north and south
portions of the island across the Cienaga.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 77 mm., tail 76 mm., terminal 20 mm. regenerated;
ventrals in 30 longitudinal and 8 transverse rows; fourth-toe subdigital scales 37
and 35 (total 72); femoral pores II and 10 (total 21); 27 scales in the fifteenth
caudal verticil. Dorsum tan with head distinctly gray and set off from dorsal
coloration; median and dorsolateral stripes paler tan and median line not espe-
cially bold against ground color. Lateral fields much reduced because of en-

106

Annals of Carnegie Museum

VoL. 41

croachment of brownish pigment along their edges, so that blaek pigment is
restrieted to a short, sinuous line in the posterior half of the lateral field area.
Lower sides gray, not blotched or vermiculate with darker. Tail dull grayish
above, sides paler, and ventrolateral pale caudal stripe obscure. Throat pale blue
and without any black pigment; belly and under side of tail pale blue. Limbs
tan above and without darker reticulations or markings. A white transverse stripe
on posterior face of thigh, surrounded by tannish gray above and below.

VARIATION; The series of 91 A. a. marcida has the following counts; longitu-
dinal ventrals 30-34 ( mode 32 ) ; rows of transverse ventrals 8 ( 6 specimens ) , 9
(1 specimen) or 10; fourth-toe scales 63-83 (mean 75.4); femoral pores 20-30
( mean 24.8 ) ; fifteenth verticil 22-29 ( mean 25.0 ) . The largest male has a snout-
vent length of 87 mm., the largest female 77 mm.

The dorsal color of A. a. marcida varies from tan to brown. Adults are
regularly the paler color and additionally have the head distinctly grayer
than the back. The three dorsal lines are tan; the median stripe is always
more obscure than the dorsolateral pair, and often becomes very faint
or disappears on the posterior half of the trunk. The lateral fields are
variable, but in most specimens there is more or less extensive brown
pigment included along the margins of the fields, so that the more cen-
tral black pigment is reduced in extent, often to as little as a sinuous or
irregular line along the lateral field area. Juveniles have the dorsal lines
more prominent than adults, but in subadults the lines have the faded
appearance of fully adult specimens. The tails regularly lack the more
prominent black or dark gray sides of the northern Isla de Pinos race
( although there may be a rather narrow lateral dark area along the sides
of the tail).

The throat and chest situation in A. a marcida is extremely peculiar.
The specimens were divided into three categories : ( 1 ) black chest and
throat; ( 2 ) black chest and black patch on throat; and ( 3 ) no black on
chest or throat. Of 52 males, only six specimens (snout-vent lengths 65
mm. to 75 mm.) fall into categoiy 1. Fifteen males (snout-vent lengths
62 mm. to 87 mm. and thus including the largest males ) are included in
category 2, and 31 males (snout-vent lengths 55 mm. to 85 mm.) lack
any black throat or chest pigment, and have the throats and chests uni-
color pale blue. In summary, slightly more than half the marcida males
(31 of 52 specimens) lack any black pigment on either the chest or
throat, but within the snout-vent-length range of these males are in-
cluded both males with partially black throats and males with solid
black throats and chests. The largest males have the chests black and
the throats partially so.

No females are included in category 1. Only two females (snout-vent
lengths 53 mm. and 72 mm. ) are in category 2, and the balance of the

1970

Ameiva auberi: L The Cuban Subspecies

107

females (35 specimens), with snout- vent lengths from 48 mm. to 80 mm.,
lack any black throat pigment. In females, the vast majority of speci-
mens lack black throats and chests. One of the two black- throated
females (the larger) is from the Paso de Piedras, and might be inter-
preted as showing some genetic influence of secta from the north (in
which race only a very few females are comparable to males in chest
and throat pigmentation). But the smaller female is from Puerto Frances,
at the extreme western tip of the island, and thus far removed from any
direct secta influence. The only interpretation I can offer of the entire
chest and throat picture of marcida is that the southern Isla de Pinos
population is in the process of losing the black pigmentation on the
throat and chest, but that many adults of both sexes do not attain the
pigmentless condition.

COMPARISONS: Marcida is most pertinently compared with the more
northern secta. Differences of throat pigmentation have just been
reviewed. The majority of marcida can be separated from the majority
of adult secta in that the former have pale blue rather than black throats.
The dorsal coloration and pattern of the two races is surprisingly dif-
ferent; marcida is in general faded and lacks the bold median and
dorsolateral lines of secta as well as the prominent lateral fields with red
markings. No other subspecies has a pale blue throat. Superficially there
are resemblances between marcida and extraria: both are rather pale
and have a poorly developed median line. The throat color in extraria
is gray to diffuse dull orange.

A. a. marcida is widely distributed on the southern portion of the
Isla de Pinos, from Puerto Frances in the west to Punta del Este in the
east. The specimens from Paso de Piedras are clearly marcida, and inter
gradation between marcida and secta presumably occurs to the north
of the Paso de Piedras, since only in this region is there an obvious
possibility of contact between the northern and southern populations of
A. auberi. The northern and southern portions of the Isla de Pinos are
almost completely separated by the Cienaga de Lanier. On the southern
part of the island, marcida was collected in beach scrub (Puerto
Frances), in palm woods and broadleaf forest (Jacksonville and Punta
del Este), and in rich, moist broadleaf forest (Paso de Piedras).

Ameiva auberi zugi, new subspecies

holotype: AMNH 83003, an adult male, from Cayo Largo, Archipielago de
los Canarreos, Habana Province, Cuba, taken July 30, 1958, one of a series col-
lected by Albert Schwartz and George R. Zug. Original number 6221.

PARATYPES (all from the Archipielago de los Canarreos): AMNH 96303-96310,
same data as holotype; AMNH 83000 (9 specimens), Cayo Matias, July 18, 1958,

108

Annals of Carnegie Museum

VoL. 41

A, Schwartz, G. R, Zug; USNM 81718-81719, Cayo Matias, September 22, 1930,
P. Bartsch; AMNH 83001 (6 specimens), Cayo Hicacos, July 28, 1958, G. R. Zug;
AMNH 83002 (3 specimens), Cayo Avalos, July 29, 1958, G. R. Zug; USNM
81766, Cayo Avalos, September 21, 1930, P. Bartsch; IB 75, 78, ASFS V11400-
VI 1401, Cayo Largo (del Sur), April 28, 1966, O. H. Garrido; IB 80-82, CM
40649-40650, Cayo Largo (del Sur), May 20, 1965, O. H. Garrido.

ASSOCIATED SPECIMENS: Cuhtt, Hcihana Province, Archipielago de los Canar-
reos, Cayo Cantiles, AMNH 93852, IB 72-74; Cayo Cantiles, Punta del Negrito,
IB 1340; Cayo Cantiles, La Aguada de la Laguna del Ingles, IB 1330-1336,
1338-1339, ASFS V11402-V11406.

DIAGNOSIS: A Cuban subspecies of A. atiheri characterized by a combination of
small size (males to 82 mm., females to 65 mm. snout-vent length), low number
of fourth-toe subdigital scales and femoral pores, and very low number of scales
in the fifteenth caudal verticil; dorsum pale sandy to tan, all dorsal lines pale
yellowish tan, not especially contrasting with the dorsal ground color and fading
posteriorly, lateral fields extensively black, with tan edging or at times included
discrete tan dots, and throats white to pale blue ( except on Cayo Cantiles; see
discussion ) ,

DISTRIBUTION: Known from five cays (Matias, Hicacos, Avalos, Cantiles, Largo)
in the Archipielago de los Canarreos, east of Punta del Este, Isla de Pinos.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 66 mm., tail 51 mm., broken; ventrals in 33 longi-
tudinal and 10 transverse rows; fourth-toe subdigital scales 37 and 36 (total 73);
femoral pores 13 and 13 (total 26); 26 scales in the fifteenth caudal verticil.
Dorsum pale sandy with median and dorsolateral lines pale yellowish tan and
barely distinguishable from dorsal ground color, all three lines fading out on
posterior half of trunk; head concolor with dorsum. Lateral fields black with a
dorsal and ventral series of tan and irregular dots or scallops restricting the black
pigment to a central irregularly-edged band; lateral fields bordered ventrally by a
moderately bold pale tan lateral line; lower sides pale sandy, grading to blue
ventrally, and without dark markings or vermiculations. Tail grayish tan dorsally,
with a moderately well defined lateral black stripe which fades out distally; ven-
trolateral caudal white stripe fairly conspicuous. Throat pale blue; venter pale
blue laterally, white medially; under side of tail pale blue. Limbs concolor with
back and without prominent darker vermiculations or markings.

VARIATION: The series of 55 A. a. zugi has the following counts: longitudinal
ventrals 30-34 ( modes 32 and 33 ) ; rows of transverse ventrals 8 ( 1 specimen ) —
10; fourth-toe scales 64-79 (mean 72.8); femoral pores 21-28 (mean 24.9);
fifteenth verticil 22-28 (mean 24.9). The largest male is from Cayo Cantiles and
has a snout-vent length of 82 mm.; the largest females are from Cayo Largo and
Cayo Cantiles and have a snout-vent length of 65 mm.

As might be expected, there is some variation between specimens of
zugi from the various cays. Material from Cayo Cantiles is the most
aberrant. Specimens from the more western islands (Matias, Hicacos,
Avalos), which are much closer to the Isla de Pinos, are somewhat
darker dorsally and are intermediate in this character between marcida
and zugi, but are closer to the latter subspecies. Specimens from Cayo

1970

Ameiva auberi: I. The Cuban Subspecies

109

Largo are obviously the most extreme and have a distinctly bleached or
faded appearance. In no adults from any cay are the dorsal lines con-
spicuous, and these regularly fade out posteriorly to become lost in the
dorsal ground color. The lateral fields are black centrally and have some
tan-to-sandy markings along the upper and lower margins. In one
specimen from Cayo Avalos, there is no black pigment on the lateral
fields, these areas being concolor with the back. In general, the throats
vary from white to very pale blue, and the venter is white centrally and
pale blue laterally. The white transverse thigh stripe is usually present
and surrounded by grayish tan. The tails are grayish tan to grayish
dorsally in adults but are blue in juveniles.

The series from Cayo Cantiles, recently collected by Sr. Garrido, is
most interesting in that the series contains five individuals that have
some black pigment on the throat and chest. Most extreme is a male
(IB 1334, snout-vent length 70 mm. and thus not the largest male) that
has a solid-black throat spot and some black on the mesoptychium and
chest. A second male (IB 72, snout-vent length 80 mm., again not the
largest male) has black on the posterior throat and mesoptychium and
some black on the chest. The next two males (IB 1335, ASFS V11406,
snout-vent lengths 81 mm. and 82 mm. ) have some black on the mesop-
tychium and chest and scattered black scales on the throat. The final
male (ASFS V11405, snout-vent length 68 mm.) has some black pig-
ment on the mesoptychium only. No specimen of zugi from any other
cay shows any indication of having black throat or chest pigment.
Although the situation with the Cantiles lizards may well be comparable
to that of extorris and extraria — i.e., assumption that the Cantiles
Ameiva are a local derivative from a more widespread form (zugi) —
I have have not named the Cantiles lizards as a distinct subspecies. I
did not do so because (1) the population on Cantiles cannot be inter-
preted as being black-throated since this feature occurs in only a small
portion of the population, and (2) in contrast to extorris and extraria,
the Cantiles lizards are '"surrounded” both to the east and west along the
linear Archipielago de los Canarreos by more orthodox pale throated
zugi. If one constructs a series (based upon the occurrence of black
throats and chests ) of the Isla de Pinos-associated subspecies, the series
is secta-marcida-zugi. This is almost surely a natural evolutionary se-
quence as well. A. a. secta is predominantly black-throated, marcida has
a lesser proportion of black-throated individuals, and in zugi, black-
throatedness occurs only as a rarity on one cay within the chain. Just as
marcida seems to be in the process of losing its capabilities for black-
throatedness, so in zugi this potentiality has been lost almost completely

110

Annals of Carnegie Museum

VoL. 41

except for occasional individuals, and apparently only on Cayo Cantiles.
It seems best, under these circumstances, to consider the Cantiles lizards
as slightly aberrant zugi. My opinions on this matter might well be
changed if it could be demonstrated that Ameiva on the cays (Avalos,
Matias, Hicacos) between Cantiles and Punta del Este also had some
black-throated individuals, but the fairly extensive material (21 speci-
mens) available from these cays does not show any such tendencies.

comparisons: The only other Cuban subspecies comparable to zugi
in dorsal coloration, pattern, and size is extraria from the northern
cayeria of Las Villas Province. Extraria has the throat gray-to-dull
orange, never white or pale blue, and has brown and inconspicuous
lateral fields. Compared with marcida of the adjacent southern coast
of the Isla de Pinos, zugi is distinctly paler and has the dorsal lines
much fainter. The throat in marcida is pale blue or black. The means of
fourth-toe scales in eoctraria (76.3) and zugi (72.8) are significantly
distinct, whereas the mean of marcida (75.4) is closer to that of zugi,
but still significantly different.

REMARKS: The Archipielago de los Canarreos is a series of cays which
extends roughly from Punta del Este eastward to Cayo Largo for a
distance of about 145 kilometers. The keys have been generally dis-
cussed by myself (Schwartz, 1959a: 117-118) in the description of the
endemic subspecies Leiocephalus cubensis pamhasileus. No Ameiva
were collected on Cayo Campos or Cayo del Rosario, both of which I
visited with George Zug, for whom the cays' race is named. On Cayo
Largo, A. a. zugi was secured on sun-baked mudflats adjacent to man-
groves at the western end of the island, and on Cayo Matias and Cayo
Avalos ground lizards were secured in the xerophytic vegetation that
covers many of these cays. Cayo Cantiles, however, is rocky and covered
with practically impenetrable jungle, and we secured only one Ameiva
there along the very rocky and hazardous shoreline. Sutcliffe (1952:5-6)
reported the occurrence of Ameiva (as Ameiva thoracica) on Cayo
Largo and on Cayo Majas Chico, a small satellite of Cayo Largo. No
specimens are available from the latter cay but I assume that they are
assignable to A. a. zugi.

A. a. zugi is obviously a derivative of A. a. marcida on the southern
part of the Isla de Pinos. Presumably the former race is widespread on
other islands of the Archipielago de los Canarreos, but the apparent
absence of Ameiva on some cays we visited is paralleled by the absence
of L. cid)ensis on some other cays. Leiocephalus was not collected on
Matias, Avalos, Cantiles, Rosario, or Largo, whereas Ameiva was not

1970

Ameiva auberi: I. The Cuban Subspecies

111

collected on Campos or Rosario. Island size surely is not a factor in such
vagaries of distribution, since both Campos and Rosario are much larger
than M atlas and Avalos where Ameiva has been collected.

Ameiva auberi galbiceps, new subspecies
holotype: AMNH 78058, an adult male, from southernmost point of large cay,
3 km. N.W. Cayo Cachiboca, Laberinto de las Doce Leguas, Camagiiey Province,
Cuba, taken August 24, 1957, one of a series collected by William H. Gehrmann
and John R. Feick. Original number 4320.

paratypes ( all from the Laberinto de las Doce Leguas, Camagiiey Province,
Cuba): AMNH 78049-78057, 78059-78064, same data as holotype; USNM
81707, Cayo Caballones, September 9, 1930, P. Bartsch; USNM 81705-81706,
Pilot Point, Cayo Anclitas, September 9, 1930, P. Bartsch; AMNH 78046-78048,
Cayo Cachiboca, August 25, 1957, W, H. Gehrmann, J. R. Feick; AMNH 78065-
78068, unnamed cay, 20° 37' N. latitude, 78° 32' W. longitude, August 25, 1957,
W. H. Gehrmann, J. R. Feick; USNM 81700, cay 78° 33' W. longitude (= same
cay as AMNH 78065-78068), September 7, 1930, P. Bartsch; MCZ 57936, Cayo
Cabeza del Este, August 15-16, 1957, R. Molina and R. Ruibal; USNM 81701-
81703, southeast end, Doce Leguas (= Cayo Cabeza del Este), September 7,
1930, P. Bartsch; USNM 81704, “Doce Leguas," September 8, 1930, P. Bartsch.

DIAGNOSIS: A Cuban subspecies of A. auberi characterized by a combination of
moderate size (males to 87 mm., females to 75 mm. snout-vent length), mod-
erate number of fourth-toe subdigital scales and femoral pores, high number of
scales in the fifteenth caudal verticil; dorsum tan, often with the dorsal fields
with some vague grayish marbling especially in their posterior halves, head and
neck distinctly yellow-green ( almost chartreuse ) , dorsal lines yellowish tan, me-
dian line fading posteriorly, lateral fields black with two or three longitudinal
included series of reddish tan spots, and throats yellow-green.

DISTRIBUTION: Known from six cays in the Laberinto de las Doce Leguas off
the southern coast of Camagiiey Province, Cuba.

description of holotype: An adult male with the following measurements
and counts: snout-vent length 87 mm., tail 168 mm.; ventrals in 33 longitudinal
and 10 transverse rows; fourth-toe subdigital scales 38 and 37 (total 75); femoral
pores 12 and 13 (total 25); 28 scales in the fifteenth caudal verticil. Dorsum
tan with median and dorsolateral lines yellowish tan, the median line visible only
far anteriorly, but the dorsolateral lines fairly prominent above the lateral fields;
head and neck distinctly yellowish green and in striking contrast to the dorsal
color; dorsal fields with very vague grayish marbling, especially posteriorly. Lat-
eral fields extensively black with about three longitudinal series of reddish tan
spots which are fused vertically to give a series of more or less vertical reddish
tan bars, although the spots often maintain their individual identities and are not
fused; lateral fields bordered below by a pale lateral line. Lower sides blue-gray
without prominent tigroid markings. Tail tan above, and with the pale ventro-
lateral caudal stripe fairly prominent basally. Limbs tan above, marbled with gray
vermiculations; pale transverse thigh stripe obscure and surrounded by grayish
tan areas both above and below. Throat yellow-green; belly pale blue centrally,
darker blue laterally; under side of tail pale blue to whitish.

variation: The series of 29 A. a. galbiceps has the following counts: longi-

112

Annals of Carnegie Museum

VoL. 41 I

tudinal ventrals 31-35 (mode 33); rows of transverse ventrals 10-12 (1 specimen);
fourth-toe scales 70-81 (mean 74.7); femoral pores 21-28 (mean 25.3); fifteenth
verticil 25-30 (mean 27.8). The largest male is from the large cay 3 km. N.W.
Cayo Cachiboca and has a snout-vent length of 87 mm. The largest female is from
the same cay and has a snout-vent length of 75 mm.

Throughout its range, A. a. galbiceps is fairly constant in coloration
and pattern. The yellow-green head, neck, and throat are regular |
features; no other subspecies of A. aiiheri has such a brightly marked |
head or a yeUow-green throat. Many specimens (both adults and sub- I
adults) have the dorsal fields with rather distinct grayish marbling, but '
these markings are very faint in the holotype and in some other speci- |
mens. The lateral fields may have the series of reddish tan spots dis- |
Crete or the spots may be fused, or partially fused, with one another as |
in the holotype. The black area in the lateral fields is seldom extremely
reduced or restricted, however, except in the specimen from Cayo |
Caballones at the western end of the Doce Leguas. The dorsal lines are |
yellowish tan and are fairly prominent at least anteriorly. The median j
line fades posteriorly and is lost in the dorsal ground color. I

comparisons: A. a. galbiceps in dorsal pattern resembles A. a.
llanensis from the serpentine savannas of Camagiiey Province. Both have !
some grayish-to-brownish markings in the dorsal fields and are the only !'
subspecies that have this feature. The yellow-green head and throat of !
galbiceps is a distinctive feature which occurs in no other subspecies; the
throat in llanensis is pale orange. The yellow throat of procer far to the
west on the Cuban mainland is similar to that of galbiceps, but the jj
latter race has the throat pigmentation distinctly more green, is much j
smaller, and has persistent dorsal lines in the adult. li

REMARKS: The series of A. a. galbiceps was collected principally i:
through the efforts of Gehrmann and Feick. The lizards were moderately H
abundant on all the cays visited and were at once recognized as some- |
thing quite distinctive from any populations that had been sampled on i
the mainland. The absence of A. auberi material from much of the '
province of Camagiiey, especially the southern coast, makes comparison ■,
of galbiceps with mainland populations difficult. It is interesting that
no A. auberi have been collected by us or others throughout much of
Camagiiey Province, despite apparently suitable habitats in very many
areas. The dorsal field markings in galbiceps and llanensis suggest that
they are related, but the ranges of the two are widely separated today.

It is possible that llane7isis occurs much farther to the south than present
records indicate, however. The scrub habitat of galbiceps is in contrast
to the savanna habitat of llanensis.

1970

114

Annals of Carnegie Museum

VoL. 41

TABLE 1

Subspecies of A. auberi Ranked by Snout- vent
Length (in millimeters) of Largest Male

Subspecies

Largest

Largest

Male

Female

procer

136

115

atrothorax

126

93

garridoi

no

70

gemmea

105

97

peradusta

105

76

denticola

96

78

granti

93

82

secta

93

77

auberi

92

88

hardtji

92

76

paulsoni

92

75

citra

92

72

llanensis

90

78

marcida

87

77

galbiceps

87

75

abducta

87

68

ziigi

82

65

pullata

81

81

mblesta

81

67

sabulicolor

80

66

cacuminis

80

60

extraria

79

68

ustulata

78

68

extorris

76

—

TABLE 2

Subspecies of A. auberi Ranked According

TO Mean Number of Femoral Pores

Subspecies

N

Femoral Pores

garridoi

23

27.6 (25-32)

peradusta

28

27.2 (23-33)

denticola

59

27.1 (23-31)

atrothorax

65

26.8 (21-32)

sabulicolor

47

26.7 (23-31)

auberi

33

26.6 (23-31)

pullata

24

26.3 (23-31)

sublesta

12

26.3 (25-28)

ustulata

85

26.2 (22-32)

procer

62

26.0 (21-31)

gemmea

15

25.9 (22-30)

( Continued next page )

1970

Ameiva auberi; I. The Cuban Subspecies

115

TABLE 2 ( Continued )
Subspecies of A. auberi Ranked According
TO Mean Number of Femoral Pores

granti

54

25.7 (21-31)

cacuminis

17

25.6 (24-29)

ahducta

30

25.4 (21-30)

galbiceps

29

25.3 (21-28)

extorris

7

25.3 (24-28)

zugi

55

24.9 (21-28)

hardiji

15

24.8 (21 -30)

secta

57

24.8 (21-31)

marcida

91

24.8 (20-30)

citra

31

24.3 (20-29)

paulsoni

28

24.2 (20-28)

extraria

55

23.8 (21-27)

llanensis

13

23.8 (21-26)

TABLE 3

Subspecies of A. auberi Ranked According to Means of Number

OF Scales in Fifteenth Caudal Verticil; N as in Table 2

Subspecies

Fifteenth verticil

atrothorax

29.5

(25-33)

peradusta

28.9

(26-32)

auberi

28.2

(26-31)

denticola

28.0

(24-33)

sublesta

27.9

(26-30)

garridoi

27.9

(26-30)

galbiceps

27.8

(25-30)

llanensis

27.7

(25-30)

abducta

27.6

(24-29)

procer

27.3

(24-31)

pullata

26.8

(24-29)

extraria

26.5

(23-30)

cacuminis

26.4

(24-29)

gemmea

25.9

(24-30)

citra

25.8

(24-29)

ustulata

25.8

(22-31)

paulsoni

25.6

(24-27)

sabulicolor

25.6

(22-29)

hardyi

25.6

(24-29)

extorris

25.4

(24-27)

granti

25.1

(22-29)

marcida

25.0

(22-29)

secta

24.9

(22-28)

zugi

24.9

(22-28)

i

116

Annals of Carnegie Museum Vol. 41

TABU

1

o

e

o

'S

to

1

e

o

3

3

a

a

a

cacuminis

16

75.1 ± 3.00

X — ™

+

—

denticola

41

76.4 ±; 1.38

X +

—

procer

56

71.9 ± 1.35

X

—

+

paulsoni

22

68.6 ± 1.84

X

+

+

auberi

26

76.3 ± 2.46

X

—

pullata

21

75.0 ± 1.80

X

ahducta

21

75.2 ± 2.14

gemmea

12

73.8 ± 2.36

garridoi

22

78.3 ± 2.19

ext r aria

44

76.3 ± 0.87

atrothorax

51

79.0 ± 1.18

peradusta

27

79.9 ±: 1.64

citra

21

77.2 ±; 2.18

granti

48

75.9 ± 1.69

salndicolor

43

72.8 ± 1.32

ustulata

60

72.8 ± 1.10

hardtji

10

77,8 ± 2.12

secta

49

69.6 ± 1.34

marcida

72

75.4 ± 1,23

zugi

46

72.8 ± 1.02

galbiceps

20

74.7 It 1.60

Subspecies arranged from west to east, three subspecies (liar ^
extorris, suhlesta) not included because of small sample sit!
for each sample indicated in first column, means and two sta ^

ahducta

1970

Ameiva axjberi: I. The Cuban Subspecies

117

piRisoN OF Foumth-toe Subdigital Scale Counts in Cuban Ameiva auheri

K

g

%«

o

e

eo

a.

1

'3

O

8

Q

1

a

citra

granti

.o

§

ustulati

hardyi

secta

:s

o

Q

s

s

tlB

—

+

—

—

—

—

—

+

—

cacuminis

"

+

+

—

+

+

—

+

—

—

denticola

-

+

+

+

+

+

—

+

—

procer

-

+

+

+

+

+

+

+

—

+

+

+

paulsoni

—

+

—

+

+

—

+

auheri

-

+

+

—

—

—

+

—

pullata

+

+

—

—

+

—

ahducta

1“

+

+

—

—

—

—

+

—

gemmea

r

—

+

+

+

+

garridoi

i

+

+

—

+

+

—

+

—

+

extraria

1

X

—

+

+

+

+

+

+

+

atrothorax

X

—

—

+

+

—

+

+

+

+

peradusta

X

—

+

+

+

—

+

—

citra

X

+

+

+

+

granti

X

+

+

+

—

sabuUcolor

X

+

+

+

—

ustulata

X

+

—

+

hardyi

X

+

+

+

secta

X

+

—

marcida

1

X

zugi

1

X

galbiceps

rj f means in second column, A “-f-” in table indicates that the
>S3species involved differ significantly (non-overlap of two
i d errors of mean ) ; a indicates no statistical difference.

I

I

Ameiva auberi: I. The Cuban Subspecies

117

116

Annals of Carnegie Museum

VoL. 41

1970

Table ‘

Fourth-toe Subdigital Scale Counts in Cuban Ameiva auberi

C

H

0

2

3

S.

■g

3

a

procer

paulsoi

auberi

pullata

1

i

0

0

■«

<3

citra

granti

"S

3

secta

3

Sb

cacuminis

16

75.1 ±

3.00

X

-

-

+

~

-

_

-h

-

-

-

-

-

+

—

—

—

cacuminis

denticola

41

76.4 ± 1.38

X

+

+

-

-

_

+

-1-

-

-

+

-f-

-

+

—

—

—

denticola

procer

56

71.9 ± 1.35

X

-

+

-

-

+

+

-f

-t-

+

-

-

+

-

+

—

—

procer

paulsoni

22

68.6 ±

1.84

X

+

-f

1

-f

-h

-f

+

+

+

+

—

-f

+

+

paulsoni

auberi

26

76.3 ± 2.46

X

-

-

-

-

-f

-

—

—

—

-f

+

—

+

—

auberi

pullata

21

75.0 ±

1.80

X

-

-

-f

+

-

-

—

—

—

+

—

—

—

pullata

abducta

21

75.2 ±

2.14

X

-

+

+

-

-

-

-

-

+

—

—

—

abducta

gemmea

12

73.8 ±

2.36

-

+

-F

-

-

-

-

—

+

—

—

—

gemmea

garridoi

22

78.3 ±

2.19

-

-

-

-

-

+

+

—

+

—

+

—

garridoi

extraria

44

76.3 ±

0.87

X

-f

+

-

-

+

+

—

+

—

+

—

extraria

atrothorax

51

79.0 ±

1.18

X

-

-

+

+

+

—

+

+

+

+

atrothorax

peradusta

27

79.9 ±

1.64

X

-

-

+

+

-

+

+

+

+

peradusta

citra

21

77.2 ±

2.18

X

-

+

+

—

+

—

-f

—

citra

granti

48

75.9 ±

1.69

X

-f-

-f

—

+

—

-f

—

granti

sahulicolor

43

72.8 ±

1.32

X

—

+

+

+

—

—

subulicolor

ustulata

60

72.8 ±

1.10

X

+

+

+

—

—

ustulata

hardtji

10

77.8 ±

2.12

X

+

—

-f

—

luirdiji

secta

49

69.6 ±

1.34

X

+

-f

+

secta

marcida

72

75.4 ±

1.23

X

+

—

marcida

zugi

46

72.8 ±

1.02

X

zugi

galbiceps

20

74.7 ±

1.60

X

galbiceps

Subspecies arranged from west to east, three su si®

extorris, sublcsta) not included because of
for each sample indicated in first column, means a

tanJ N

i’^des inv ‘Column. A in table indicates that the
** of rne'^^ ) significantly (non-overlap of two
ean), a indicates no statistical difference.

II. THE BAHAMIAN SUBSPECIES
Clarence J. McCoy

Associate Curator, Section of Amphibians and Reptiles,

Carnegie Museum, Pittsburgh, Pennsylvania

Introduction

Ameiva thoracica Cope was described in 1862 from a series of speci-
mens collected on New Providence Island in the Bahamas by Dr. H. C.
Wood, Jr. In Stejneger’s early revision of the Bahamian land-reptile |i

fauna (in Shattuck, 1905),^ A. thoracica was listed from Eleuthera II

Island, Andros Island, and Great Abaco Island, in addition to New
Providence. Barbour and Noble (1915) correctly questioned the Great
Abaco record, and Barbour ( 1916 ) reported thoracica from several
localities in the Exuma Cays. Cochran (1934) recorded specimens col-
lected by Bartsch, supposedly from the Ragged Islands and Crooked
Island. By the time of Barbour’s ‘‘Third List of Antillean Reptiles and
Amphibians’" (1937), Ameiva thoracica was firmly established as ''wide-
spread in the northern and central portions of the Bahama Archipelago.”
Despite the relatively long history of taxonomic recognition of Ameiva
thoracica, infraspecific variation in this common Bahamian species has
never been closely studied. Only the astute Tliomas Barbour, who
probably saw more thoracica than any other herpetologist of his day,
wrote “with much larger series Ameiva thoraeica may be shown to break
up into several races” (Barbour and Shreve, 1935).

In July, 1965, Neil D. Richmond of Carnegie Museum participated in
the Fourth Expedition of the Bahamas Biological Survey, under the
auspices of the American Museum of Natural History and the Office
of Naval Research [Project No. 552(07)]. Among the specimens col-
lected during this expedition were series of Ameiva auberi from widely
scattered localities in the northern Bahamas, including the Berry Islands,
Andros Island, the Exuma Cays, Long Island, the Ragged Islands, and
Green Cay. With this fresh material available for study it became obvi-
ous that Bahamian auberi exhibits a surprising range of variation in
color and pattern. Preliminary observations on these specimens were
extended to include examination of color and scale-count characters
in auberi from throughout the Great Bahama Bank. The inter-population
differentiation so immediately apparent in color patterns was substan-
tiated and quantified by the data on scutellation and size. The variants
occur as internally consistent geographic units, which are here given
taxonomic recognition as subspecies.

1. Bibliography follows section III.

118

1970

Ameiva auberi: II. The Bahamian Subspecies

119

It is not surprising that the well-marked Bahamian subspecies of
Ameiva auberi have previously escaped detection. The most important
and obvious characters lie in fugitive elements of color and pattern which
are inevitably destroyed by preservation. The quantitative treatment and
statistical comparsion of scale counts are taxonomic tools only recently
applied in studies of the macroteiid genera Ameiva and Cnemidophorus,
and have not previously been used in treatments of A. auberi. Moreover,
despite the accumulation of Bahamian Ameiva in many collections, ade-
quate series of fresh and well-preserved specimens from many parts of
the species' Bahamian range have been lacking. During this study I
examined 480 specimens from 41 cays and islands representing nine
major island groups on the Great Bahama Bank, including the Bimini
Islands, the Berry Islands, Andros Island, New Providence Island,
Eleuthera Island, Cat Island, the Exuma Cays, Long Island, and the
Ragged Islands.

I am grateful to the following curators and institutions (identified
throughout by the abbreviations in parentheses ) for permission to
examine species of Ameiva auberi in their care: Charles M. Bogert and
Richard G. Zweifel, American Museum of Natural History (AMNH);
Edmond V. Malnate, Academy of Natural Sciences, Philadelphia
(ANSP); Albert Schwartz, Miami, Florida (ASFS); Walter Auffenberg,
Florida State Museum (UF); and Doris M. Cochran and James A.
Peters, United States National Museum (USNM). Specimens in the
collections of Carnegie Museum are designated by the abbreviation CM.
My special thanks are due Wayne King, George B. Rabb, Neil D. Rich-
mond, and Albert Schwartz for permission to use unpublished field notes
on Bahamian Ameiva, and to Richmond, Schwartz, and M. Graham
Netting for helpful criticism of the manuscript.

Gazetteer of Localities

The confusing multiplicity of names available for the innumerable
cays and islands of the Bahamas has caused some difficulty in interpre-
ting locality data in museum catalogues and in the published literature.
The following alphabetical gazetteer includes the names of all cays and
islands of the Great Bahama Bank from which specimens of Ameiva
auberi have been examined. The larger island groups are labeled on the
accompanying map (Fig. 1), and reference to the map and this gazet-
teer should permit general location of all islands mentioned in the
systematic accounts. More detailed maps and charts, from which the
place names here used were taken, are available in Kline ( 1964 ) and on
U.S. Coast and Geodetic Survey chart 1002, “Straits of Florida.”

120

Annals of Carnegie Museum

VoL. 41

ANDROS ISLAND (23° 43' - 25° 11' N., 77° 30' - 78° 28' W. ) — A very large island,
approximately 110 miles long, north to south, by a maximum of 45 miles wide,
lying in the western arm of the Great Bahama Bank. Andros is crossed just
south of center by three tidal passages. North Bight, Middle Bight, and South
Bight, which divide this part of the island into a complex system of smaller
cays.

BELL ISLAND (24° 19' N., 76° 32' W. ) — A small circular cay between Halls Pond
Cay and Cambridge Cay in the northern third of the Exuma Cays.

BITTER GUANA CAY (24° 08' N., 76° 23' W. ) — A small elongate cay just off the
north end of Great Guana Cay in the middle of the Exuma Cays chain.

bond's cay (25° 29' N., 77° 42' W. ) — A long cay near the southeastern corner
of the Berry Islands.

CAT ISLAND (24° 07' - 24° 44' N., 75° 15' - 75° 46' W.) — A relatively large,
long island in an isolated position between Eleuthera Island and San Salvador
Island, near the eastern edge of the northern Bahamas.

CAVE CAY (23° 50' N., 76° 14' W. ) — A small cay between Big Farmers Cay and
Darby Island, north of Great Exuma Island in the Exuma Cays chain.

CHUB CAY (25° 24' N., 77° 50' W. ) — The western portion of the largest cay in
the southwestern segment of the Berry Islands, barely separated from Frazer’s
Hog Cay by Frazer’s Creek.

CISTERN CAY (26° 46' N., 77° 52' W. ) — A small island closely adjacent to the
western side of Great Harbour Cay, in the northern Berry Islands.

CROOKED ISLAND (22° 40' - 22° 50' N., 74° 02' - 74° 21' W.) — A large island
in the southern Bahamas, across the Crooked Island Passage from Long Island.

devil’s cay (25° 36' N., 77° 42' W. ) — A small rectangular cay between Holmes
Cay and Little Harbour Cay, near the center of the arc of the Berry Islands.

east BIMINI ISLAND (25° 24' N., 79° 15' W.) — Easternmost of the three main
islands in the Bimini Island group, at the northwestern tip of the Great Bahama
Bank.

EASTER CAY (25° 24' N., 79° 15' W. ) — The recurved tip of North Bimini Island
which lies adjacent to East Bimini Island, Bimini Island group.

ELEUTHERA ISLAND (24° 37' - 25° 35' N., 76° 08' - 76° 47' W.) — A very long
crescent-shaped island north of Cat Island and east of New Providence Island,
at the northern tip of the east arm of the Great Bahama Bank.

ELIZABETH ISLAND (23° 31' N., 75° 42' W. ) — A small cay that forms part of the
outer bar of Elizabeth Harbour, Great Exuma Island.

FLAMINGO CAY (22° 53' N., 75° 51' W.) — A relatively high and well-wooded
small cay near the northern end of the Ragged Island chain.

Frazer’s hog cay (25° 24' N., 77° 50' W. ) — The eastern tip of theTargest island
in tlie southern part of the Berry Island group, barely separated from Chub
Cay by Frazer’s Creek.

GIBSON CAY (24° 18' N., 77° 39' W. ) — A small cay in the mouth of the Middle
Bight, east side of Andros Island.

GREAT EXUMA ISLAND (23° 28' - 23° 48' N., 75° 41' - 76° 02' W.) — Largest of
the Exuma Cays, at the southern tip of the chain.

GREAT GUANA CAY (24° 09' N., 76° 20' W. ) — A very long, narrow cay, one of
the largest of the Exuma Cays, near the middle of the chain.

1970

Ameiva auberi: II. The Bahamian Subspecies

121

GREAT RAGGED ISLAND (22° 12' N., 75° 43' W. ) — Largest cay in the Ragged
Island group, at the extreme southern tip of the Great Bahama Bank.

GREEN CAY (24° 02' N., 77° 11' W. ) — A small cay isolated on a westward exten-
sion of the east arm of the Great Bahama Bank, approximately midway between
the southern part of Andros Island and the center of the Exuma Cays chain.

GUN CAY (25° 35' N., 79° 19' W.) — A small cay in the Bimini Island group,
south of the main islands.

HOG CAY (22° 15' N., 75° 44' W. ) ■ — A small cay across the harbor from Great
Ragged Island, in the Ragged Island Group.

HOG ISLAND (25° 06' N., 77° 20' W. ) — An elongate small cay that lies parallel
to the northeast shore of New Providence Island, opposite Nassau harbor. Some-
times called Paradise Island.

LITTLE EXUMA ISLAND (23° 27' N., 75° 37' W. ) — A relatively large cay adjacent
to the southern end of Great Exuma Island, at the southern tip of Exuma Cays
chain. This island is frequently shown connected to Great Exuma (which it is
not), and is not separately named on many charts.

LITTLE HARBOUR CAY (25° 34' N., 77° 42' W. ) — A fair-sized cay between
Devil's Cay and Frozen Cay in the center of the Berry Islands.

LITTLE RAGGED ISLAND (22° 10' N., 75° 42' W. ) — A Small uninhabited cay which
lies very close to the south end of Great Ragged Island, at the southern end of
the Ragged Islands.

LITTLE SAN SALVADOR ISLAND (24° 40' N., 75° 55' W. ) — A Small island in the
channel between Eleuthera Island and Cat Island.

LONG ISLAND (22° 50' - 23° 42' N., 74° 49' - 75° 21' W.) — A very long and
narrow island south of the Exuma Cays chain, on the southeastern corner of
the Great Bahama Bank.

MANGROVE CAY (24° 15' N., 77° 37' W. ) — A large triangular island that sepa-
rates the South Bight from the Middle Bight of Andros Island at their eastern
ends.

NEW PROVIDENCE ISLAND (25° 03' N., 77° 15' - 77° 34' W.) — A fairly large, oval
island at the northern tip of the eastern arm of the Great Bahama Bank, be-
tween Andros Island and Eleuthera Island.

NORTH BIMINI ISLAND (25° 44' N., 79° 18' W. ) — -The westernmost of the three
main islands in the Bimini Island group.

NURSE CAY (22° 29' N., 75° 50' W. — A small cay north of Great Ragged Island,
near the middle of the Ragged Island chain.

RACCOON CAY (22° 21' N., 75° 49' W.) -—A small cay between Nurse Cay and
Hog Cay, in the southern third of the Ragged Island chain.

REIDS CAY (24° 18' N., 77° 54' W. ) — A small cay centered in the North Bight
of Andros Island.

ROSE ISLAND (25° 06' N., 77° 13' W. ) — An elongate cay in the series between
the eastern tip of New Providence Island and the northwest point of Eleuthera
Island.

SAMPSON CAY (24° 15' N., 76° 29' W. ) — A small cay between Thomas Cay and
Bitter Guana Cay, in the central part of the Exuma Cays chain.

SHIP CHANNEL CAY (24° 49' N., 76° 49' W. ) — A long cay near the northern end
of the Exuma Cays chain.

122

Annals of Carnegie Museum

VoL. 41

SOUTH BIMINI ISLAND (25° 42' N., 79° 18' W. ) — Southernmost of the three larg-
est islands in the Bimini Island group,

SOUTH CAT CAY (25° 32' N., 79° 18' W. ) — A small cay south of Gun Cay, at
the southern end of the Bimini Island group.

WARDERicK WELLS CAY (24° 23' N., 76° 37' W. ) — A small cay between Halls
Pond Cay and Hawksbill Cay in the northern third of the Exuma Cays chain.

Characters Studied

In analyzing geographic variation in Bahamian populations of Ameiva
auheri I have used a combination of characters, including scutellation,
color pattern, and size differences. Four standard scale counts selected
for analysis were: (1) number of dorsal granules counted in a line
around midbody (GAB); (2) total number of femoral pores; (3) num-
ber of scales in the fifteenth complete caudal verticil; (4) total number
of subdigital scales on the fourth toes of both hind feet. Summaries of
these data including mean values, ranges of variation, standard errors
of the means, and sample sizes, in the eight Bahamian subspecies are
presented in Table 2. In Tables 3 through 6 scutellation data for the
eight subspecies are compared. In these latter tables a + ” symbol at
the intersection of vertical and horizontal lines for two subspecies being
compared denotes a statistically significant difference, based on non-
overlap of the ranges described by two standard errors on either side
of the mean counts for the character in the two forms. A ” indicates
overlap of these ranges, and consequently the probable absence of a
significant difference. The four characters of scutellation vary discor-
dantly, and geographically adjacent races are frequently distinguished
by significant differences in two or more characters.

MIDBODY GRANULES ( GAB )

Northern populations from the Bimini Islands, Berry Islands, and
Andros Island have the lowest GAB counts, but are not significantly
different from each other. The populations of New Providence Island,
Eleuthera Island, Gat Island, the Exuma Gays, and Long Island have I
significantly higher GAB counts, and form a second overlapping group
without internal differentiation. The Bagged Island population has GAB
counts that overlap those of the northern populations. (Tables 2 and 3).

FEMORAL PORES

The populations of the Bimini Islands and Berry Islands have similar
average femoral pore counts, at the upper limit of the range observed in
Bahamian populations of A. auheri. The Andros Island sample has sig-
nificantly lower femoral pore counts, but overlaps the New Providence
and Eleuthera populations in this character. Ship Ghannel Gay and Gat

1970

Ameiva auberi: II. The Bahamian Subspecies

123

Island share counts significantly higher than the New Providence and
Eleuthera group, and overlap the Bimini-Berry group. There is a sharp
separation between the Ship Channel Cay population and more south-
erly forms, which are united by much lower femoral pore counts.
(Tables 2 and 4).

FIFTEENTH CAUDAL VERTICIL SCALES

The Bimini Islands and Berry Islands populations are similar in this
character, and trenchantly differentiated from the Andros population
which has a much lower mean count. The New Providence, Eleuthera,
Cat Island, and Ship Channel Cay samples share overlapping counts at
the upper limit for Bahamian populations. There is a sharp discontin-
uity, partially bridged by the Long Island population, between the
Ship Channel Cay population and the remainder of the Exuma Cay
samples. Ragged Island lizards have even lower numbers of scales in
the fifteenth caudal whorl, overlapping only the population of Andros
Island in this character. (Tables 2 and 5).

FOURTH TOE SUBDIGITALS

Counts of subdigitals in lizards from the Bimini and Berry Islands
populations are similar, and among the lowest found in the Bahamian
races of A. auberi. The Andros Island population counts are only
slightly, but significantly, higher than those of the northern populations.
New Providence, Eleuthera, and Cat Island samples share even higher
counts, significantly different from the Andros sample. The Ship Channel
Cay population is differentiated from geographically adjacent popula-
tions in this character, but overlaps the Andros and Ragged Island
populations. Counts are slightly higher in the southern Exuma Cays and
Long Island populations, and lower in the Ragged Islands lizards.
(Tables 2 and 6).

COLOR PATTERN

The basic dorsal pattern of Bahamian Ameiva auberi, common to
hatchlings and juveniles of all subspecies, consists of a series of five
narrow, light, longitudinal stripes separated by darker fields three or four
times wider than the stripes. Geographic variation in color pattern is
expressed in the relative completeness of the stripes and their persistence
through ontogeny, and in the color of the intervening fields.

The middorsal stripe is continuous from the parietal scales to the
base of the tail but never appears on the free part of the tail. At maxi-
mum development the middorsal stripe passes forward between the
orbits, where it is intensified and bordered by black marks, producing a
brilliant black-yellow-black mark on the parietals. In some populations

124

Annals of Carnegie Museum

VoL. 41

the middorsal stripe persists throughout life; in others it fades and the
entire area between the dorsolateral stripes becomes unicolor. The
paired dorsolateral stripes may origininate on the supraciliaries, above
the ear openings, or farther back on the sides of the neck. They may
persist throughout life or fade at adulthood, but are always more
persistent than tlie middorsal stripe. Tlie lateral stripe is variable in ;;
width, and sharply defines the lower edge of the dorsolateral field above,
but may be either sharp-edged below or gradually blended with the "
color of the lower sides.

The dorsal fields, between the middorsal stripe and the dorsolateral jj
stripes, are gray, brown, or reddish, with or without dark edges out-
lining the adjacent shapes. The dorsolateral fields are usually black at |i
hatching, although in some populations they are light-centered, or even jj
spotted with lighter colors. In some populations old individuals grad- i
ually resorb the black pigment, producing at first an emarginate field,
then a punctate field, and eventually deshoying the black field entirely.
The dorsolateral field frequently extends onto the tail as a black stripe.
Tlie lateral field, below the lateral stripe, is usually diffuse and poorly
defined, but in one population it is intensely black and relatively sharp-
edged.

Adult males of nearly all Bahamian populations exhibit extensive
blackening of the throat, mesoptychium, chest, and undersides of the
forelegs. The extent of the black ventral pigmentation is geographically
variable, and often serves to differentiate adjacent subspecies. Sexual
dimorphism in color pattern is further expressed in the potential for
ontogenetic pattern changes. Females are usually much smaller than i
males (Table 1), and retain juvenile elements of the pattern into adult- |
hood. The ontogenetic pattern modifications produced by the loss of !
stripes and changes in the color of the fields not only start earlier in I
males, but progress much further in the adult males. Females of all i
populations tend to retain the juvenile pattern with relatively little j
ontogenetic modification. i

TABLE 1

Maximum Size in Bahamian Ameiva aiiheri

Subspecies

Total Examined

Largest $

Largest $

richmondi

69

116.6

80.4

multilineata

54

119.7

71.3

kingi

32

99.4

72.2

thoracica

123

116.4

76’0

felis

46

97.0

61.7

focalis

22

109.0

83.0

ohsoleta

83

107.0

70.9

hilateralis

42

87.0

75.0

1970

Ameiva auberi: II. The Bahamian Subspecies

125

Finally, size has been used as a character differentiating some Baha-
mian subspecies of Ameiva auberi. Absolute size differences are difficult
to quantify and interpret, but in populations from which adequate sam-
ples are available, differences in the potential size attained by adults
are obvious. The maximum sizes, snout-vent length (SVL), observed
in males and females of the various subspecies are listed in Table 1.
Adult females are remarkably smaller than adult males in all Bahamian
populations of Ameiva auberi.

Systematic Review

Ameiva auberi thoracica Cope

Ameiva thoracica Cope, 1862, Proc. Acad. Nat. Sci. Philadelphia, 14:64.
Ameiva auberi thoracica, Hecht, 1954, Year Book Amer. Phil. Soc.,
(1954): 133-135.

HOLOTYPE: ANSP 9158, an adult male, from New Providence Island, Bahama
Islands, collected by Dr. H. C. Wood, Jr.

diagnosis: a Bahamian subspecies of Ameiva auberi characterized by a com-
bination of large size (males to 116 mm., females to 74 mm. SVL), high number
of dorsal granules at midbody, moderate number of femoral pores, very high
number of scales in the fifteenth caudal verticil, and high number of fourth-toe
subdigital scales; middorsal stripe barely persistent, dorsolateral stripes continuous
to supraciliaries, dorsal fields olive, dorsolateral fields black and becoming emargi-
nate with age; throat, mesoptychium, and two-thirds of ventrals black, chinshields
white (adult males).

SPECIMENS examined: Ncw Providence Island: ANSP 9149-9180 (holotype and
paratypes); AMNH 76284-76288 + 6 untagged, CM 20379, Nassau; AMNH
76279-76283 -f- H untagged, near Nassau Village; ASFS 10640-10641, 2.8 mi.
W Oakes Field; UF 14686, Mermaid Pool; ASFS 6960, V10319-V10324, V10343-
V10348, Yamacraw Beach. Hog Island: CM 20434, across from Nassau (New
Providence Island). Rose Island: AMNH 76255-76258. Eleuthera Island (locali-
ties arranged from north to south): ASFS 17215, 6.2 mi. NW Gregory Town;
ASFS 17216-17217, 1.4 mi. NW Gregory Town; AMNH 69226-69228, Hatchet
Bay; ASFS 17464-17465, 17470-17482, Hatchet Bay Plantation; ASFS 17414,
17510, Alice Town; ASFS 17463, ocean side, opposite Alice Town; ASFS 17123-
17135, 3 mi. NW James Cistern; ASFS 17566, 6.3 mi. SE James Cistern; USNM
31992, The Bluff; ASFS V6827, approx. 4 mi. NNW Rock Sound; ASFS V6824-
V6826, Southeast Point. Little San Salvador Island: MCZ 39645, Salt Pond Cay.

DISTRIBUTION: New Providence and Eleuthera islands and connecting cays, and
Little San Salvador Island, Bahama Islands.

REDESCBiPTiON OF HOLOTYPE: An adiilt male with the following measurements
and counts: snout-vent length 114 mm., tail 267 mm. (tip missing); dorsal gran-
ules at midbody (GAB) 112; femoral pores 14 and 14 (total 28); 32 scales in
the fifteenth caudal certicil; fourth-toe siibdigital scales 40 and 42 (total 82). The
dorsal ground color has faded to a uniform light gray, and the middorsal stripe
is only faintly visible on the neck and anterior part of the back. The head is also
a faded gray color dorsally. Dorsolateral light stripes extend from the temporal

126

Annals of Carnegie Museum

VoL. 41

areas to the groin. The dorsolateral black fields are broad and ragged-edged, but
solid black centrally and continuous from just behind the ears, well onto the
posterior half of the tail. The areas immediately below these black fields are a
slightly lighter gray, and definite light stripes extend from these light areas to
the ears. Below the lateral light stripes and above the ventrals the gray ground
color is stippled with black pigment. The dorsal surfaces of the fore- and hind-
legs are uniform light gray. Ventrally the chinshields, throat, chest, under sides
of the forelegs, and central ventrals to the pre-anal scales are black. Scales on the
lateral, posterior, and anterior surfaces of the thighs are black with lighter centers,
and the underside of the tail is dark The holotype is slightly soft, somewhat
faded, and has the ventral body wall cut free as an anterior flap. Otherwise, the
specimen is in good condition.

Inexplicably Barbour and Loveridge ( 1929 ) wrote that the types of Ameiva
thoracica Cope were not to be found, either in the Museum of Comparative
Zoology or the Academy of Natural Sciences of Philadelphia. The series ANSP
9149-9180 unquestionably represents the syntypes of A. thoracica, and ANSP
9158, which most closely approximates the measurements and data given by
Cope, is identified as the holotype.

VARIATION: See tables. The type series from New Providence Island (ANSP
9149-9180) has the following counts: GAB 103-120 (110.6 zt .77); femoral pores
24-34 (27. 3 ±.41); fifteenth caudal verticil scales 29-37 (32. 2 ±.28); fourth-
toe subdigitals 74-94 (82.3 ± .72).

Ameiva auberi thoracica is a strongly lineate subspecies. In hatchlings
and subadults the middorsal stripe is continuous to the frontoparietals,
and somewhat intensified and black-edged on the head. The dorso-
lateral stripes extend forward to the light-edged supraciliaries, and the
lateral stripes are continuous to the ear openings. The middorsal stripe
may be lost in the largest males, but persists in smaller males and all
females. The dorsolateral and lateral stripes persist throughout life in
all individuals. The dorsal fields, reddish brown in hatchlings, gradually
fade to light brown or olive in older animals. The dorsolateral fields are
intensely black and sharp-edged in all except the oldest males, in which
they usually become emarginate or even reduced to a row of spots
through loss of the black pigment. Tlie black dorsolateral fields have
vague brick-red to dull tan edges in life. The lateral stripes, pale cream
to gray in life, usually retain sharp lower edges throughout ontogeny.

The ventral black coloration of adult males first appears at a snout-
vent length of approximately 60 mm. (ASFS V10347) and gradually
extends and intensifies until the throat, mesoptychium, chest, frequently
the under sides of the forelegs, and most of the ventrals are solid glossy
black. On the ventrals, however, black pigment is confined to the cen-
tral four or six rows of scales, and the outer ventral scale rows are spotted
and mottled with bluish pigment. The chinshields are almost always
white, in striking contrast to the intense black of the throat granules.

1970

Ameiva auberi: II. The Bahamian Subspecies

127

In the largest males (AMNH 76281, 105 mm. SVL) the chinshields are
also blackened along the inner edges.

Hatchlings have reddish brown dorsal fields, a tan-to-yellowish mid-
dorsal stripe, pale cream or yellow dorsolateral stripes, and a grayish
green tail.

The series from Eleuthera Island has the folowing scale counts: GAB
99-129 (109.8 =b .16); femoral pores 23-30 (26.6 ± .31); fifteenth caudal
verticil scales 28-34 ( 31.2 it .24); fourth-toe subdigitals 74-93 (81.2 zb
.66). Despite the significant difference in mean number of fifteenth
caudal verticil scales, and some subtle differences in color pattern, the
Eleuthera Island sample (which consists mostly of hatchlings and juve-
niles) is referred to thoracica. Elements of the color pattern shared by
Eleuthera and New Providence populations are the distinct and anter-
iorly extensive stripes, the reddish-brown dorsal and black dorsolateral
fields, and the striped interparietal and supraciliaries. In contrast to
New Providence lizards, Eleuthera individuals show early loss of the
middorsal stripe (ASFS V6824, male, 76 mm. SVL), less extensive black
ventral coloration in adult males, extreme reduction and eventual loss of
the black pigment in the dorsolateral fields (USNM 31992), and blue
rather than green tails as hatchlings. Nevertheless, in view of the
divergence of adjacent populations on Andros Island, Cat Island, and
Ship Channel Cay, internal variation of thoracica seems relatively
insignificant.

Evidence from populations inhabiting the islets and cays connecting
New Providence and Eleuthera, which could illuminate the relation-
ships of the end populations, is inconclusive. Four specimens from Rose
Island (AMNH 76255-76258) have characters common to both popula-
tions, although the adult male ( 92 mm. SVL ) has reduced throat color
suggestive of Eleuthera males.

A single specimen from Little San Salvador Island (MCZ 39645) is
referred to thoracica, rather than to the subspecies of nearby Cat Island,
on the basis of the unmarked solid black dorsolateral fields and a mid-
dorsal stripe continuous onto the parietals.

coMPARisiONs: Although Ameiva auberi thoracica includes a relatively
large range of variation, the subspecies has adequate internal agreement
when compared with adjacent races. The population of Andros Island
has significantly fewer dorsal granules, fifteenth caudal verticil scales,
and fourth-toe subdigital scales. These populations share elements of
the color pattern, but the rich bronze-red dorsal fields of Andros Island
lizards is not matched by thoracica. To the east there are similarly

128

Annals of Carnegie Museum

VoL. 41

trenchant differences between thoracica and the Cat Island subspecies.
Cat Island lizards have higher GAB counts, more femoral pores and
are probably significantly smaller as adults. In addition, the dorsolateral
fields of thoracica are solid black or emarginate rather than internally
spotted as in the Cat Island population. Some of the differences between
New Providence thoracica and the Cat Island population may be par-
tially bridged by adequate samples of adult lizards from Eleuthera
Island.

Ameiva auberi thoracica finds its closest relative in the population of i:
Ship Channel Cay, which is almost surrounded on the north by the
range of thoracica. There are significant differences here in femoral pores
and fourth-toe subdigital scales, but size and other characters of scala-
tion are similar. Although the color patterns are grossly alike there are
substantial differences in the extent of the ventral coloration in males, in
the color of the snout, and in the tendency to lose dorsal striping. :

Ameiva auberi felis, new subspecies ■

holotype: cm 20440, an adult male, from The Bight, Cat Island, Bahama n
Islands, taken on June 14, 1941, by George R. Campbell. '

PARATYPES: All from Cat Island (localities arranged from north to south), as
follows: ASFS V2174, Orange Creek, December 1, 1963, D. R. Paulson; AMNH
76267-76273 + 13 untagged, Bennett’s Harbour, March 24, 1963, Van Voast Ex-
pedition; ASFS V2161-V2163, V2237-V2238, Tea Bay, Nov. 30, 1963, D. R.
Paulson; AMNH 76266, Tea Bay Settlement, March 21, 1953, Van Voast Expedi-
tion; ASFS V2113-V2117, hills above The Bight, Nov. 28, 1963, D. R. Paulson; |
AMNH 76259-76265, McQueens, 1-2 mi. E of the village, Jan, 24, 1953, Van ;
Voast Expedition; ASFS VI 11 16, 0.5 mi. N Port Howe, Nov. 2, 1966, native
collector.

DIAGNOSIS: A Bahamian subspecies of Ameiva auberi characterized by a com-
bination of moderate size (males to 97 mm., females to 62 mm. SVL), high num-
ber of dorsal granules at midbody, high number of femoral pores, moderate :
number of scales in the fifteenth caudal verticil, and high number of fourth-toe '
subdigital scales; dorsal stripes becoming obsolete with age, dorsal fields olive-
gray, dorso-lateral fields black spotted with lighter color; throat, mesoptychium,
and anterior few rows of ventrals black in adult males.

distribution: Cat Island, Bahama Islands.

DESCRIPTION OF THE HOLOTYPE: An adult male with the following measure-
ments and counts: snout-vent length 72.3 mm., tail incomplete; dorsal granules
at midbody 109; femoral pores 13 and 13 (total 26); 30 scales in the fifteenth ;
caudal verticil; fourth-toe subdigital scales 39 and 40 (total 79). Dorsal ground
color dark olive, middorsal and dorsolateral stripes completely absent. Dorsolateral ■
black fields continuous from above forelegs to groin, and punctuate with num-
erous, almost confluent light spots, producing a distinctly reticulate appearance.
Lateral light stripe medium gray, sharp-edged above and below, and bordered |i
below by a dark gray field shading into the dark ventrals. Dorsal surfaces of limbs ,

1970

Ameiva AUBERi: II. The Bahamian Subspecies

129

olive. Central part of throat, mesoptychium, and anterior three rows of ventrals
black, the black extending to the under surfaces of the forelegs as black stippling
on dark gray. Lateralmost few rows of ventrals light-spotted, remainder of ven-
tral surface dark gray, probably as a result of overexposure to strong formalin.
The holotype appears to be generally darkened by improper preservation.

VARIATION: See tables. The presence of reddish spots in the dark-
brown-to-black dorsolateral fields of felis is unique among Bahamian
races of Ameiva auberi. These spots are visible in hatchlings, and may
grow and coalesce with adjacent spots and with the border of the dorso-
lateral field to form a ragged or even reticulate field in older individuals.
The majority of animals exhibit distinctly spotted dorsolateral fields.
The middorsal stripe is tan-orange to tan, continuous onto the inter-
parietal in post-hatchlings (ASFS V2117, 42 mm. SVL), but completely
obliterated by the brown-to-medium-brown dorsal fields in the largest
males. The dorsolateral stripes are also tan-orange in life and more
persistent than the middorsal stripe. In even the smallest individuals
examined the dorsolateral stripes end above the foreleg insertions. The
lateral stripes are wide and buffy in life, and bordered below by lateral
fields of purplish blue. The lateral fields may be irregularly spotted,
giving a checkered appearance to the lower sides. The ventral color in
life is pinkish blue or slightly purplish marked with black in adult males.
The black ventral markings first appear at approximately 55 mm. SVL as
a black smudge in the center of the throat, a black mesoptychium, and
blackening of the anterior few rows of chest scales. Fully developed, the
black covers most of the throat (but usually not the chinshields ) , the
mesoptychium and under sides of the forelegs, and the anterior third of
the ventral scales. The tail is bright blue to blue-green in life, and several
hatchlings retain a bright cerulean, but distinctly striped, tail in pre-
servative.

COMPARISONS. Ameiva auberi felis stands alone among the Bahamian
races of auberi in the presence of light-spotting in the dorsolateral fields.
From the adjacent population of the Exuma Cays felis is strongly
divergent in all characters of scalation, and probably has closest affinity
with the Eleuthera Island population of thoracica. Eleuthera Island
thoracica and felis have a tendency toward ontogenetic loss of the mid-
dorsal stripe, similar GAB counts, fifteenth caudal verticil scale counts,
and fourth-toe subdigital scale counts. They are distinguished by a
significant difference in femoral-pore counts, and by the completely
different color of the dorsolateral fields, which early in life fade into zones
of black spots in Eleuthera Island thoracica.

130

Annals of Carnegie Museum

VoL. 41

Ameiva auberi kingi, new subspecies

holotype: cm 40915, an adult male, from Gibson Cay, in the mouth of South
Bight, Andros Island, Bahama Islands, taken on July 4, 1965, by Neil D.
Richmond.

PARATYPEs: CM 40916-40920, same data as holotype.

ASSOCIATED SPECIMENS: All from Andros Island (localities arranged from north
to south) as follows: UF 5982, 17923-17924, approx. 9 mi. at 240° from Stafford
Creek Settlement; ASFS 10287-10293, Fresh Creek; UF 17927-17930, Andros
Town; UF 17920-17921, east tip of Reid’s Cay, North Bight; MCZ 42008, Man-
grove Cay; MCZ 42003, south side of South Bight; UF 17931-17933, 2 mi. at 55°
from mouth of Deep Creek; MCZ 42006-42007, Cistern Creek.

DIAGNOSIS: A Bahamian subspecies of Ameiva auberi characterized by a com-
bination of moderate size (males to 99 mm., females to 72 mm. SVL), low number
of dorsal granules at midbody, moderate numbers of femoral pores and fourth-toe
subdigital scales, and very low number of scales in the fifteenth caudal verticil;
middorsal stripe persistent in adults, dorsal fields reddish brown, dorsolateral
fields black; chinshields, throat, mesoptychium, and anterior fourth of ventrals
black in adult males.

DISTRIBUTION: Audros Island, Bahama Islands.

DESCRIPTION OF HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 99.4 mm., tail 258 mm. (complete); dorsal gran-
ules at midbody 97; femoral pores 13 and 13 (total 26); 27 scales in the fifteenth
caudal verticil; fourth-toe subdigital scales 36 and 37 (total 73). Dorsal fields
a rich, reddish brown, shading to medium brown on top of the head and tail.
Middorsal stripe gray, extending from the frontoparietal almost to the middle of
the back. Dorsolateral light stripes intense, extending from the temporal regions
to near the groin. Dorsolateral black fields vivid and sharp-edged, bearing narrow
stripes of lighter color adjacent to the bordering light stripes. Lateral light stripes
broad, subtended by lines of irregular black spots above the ends of the ventrals.
Chinshields, throat, chest, under sides of forelegs, and approximately a quarter of
the anterior ventrals black. Posterior ventrals bluish laterally shading to white at
mid-venter; pre-anal scales, under surfaces of thighs, and lower side of tail very
light gray with wliite mottling. Tail brown above, with black and white stripes,
light gray below. Dorsal surfaces of limbs gray-brown, stippled with black.

VARIATION: See tables. The type series from Gibson Cay (CM 40915-40920)
has the following counts: GAB 88-103 (average 96.8); femoral pores 22-26
(24.2); fifteenth caudal verticil scales 27-31 (28.3); fourth-toe subdigital scales
73-77 (74.7).

Ameiva auberi kingi is characterized by reddish brown, almost bronze,
dorsal fields in combination with solid black dorsolateral fields. The
middorsal stripe is buff to yellowish in life, and continuous anteriorly
beyond the parietal scales where it is bordered by black marks on each
side, forming a distinctive black-white-black mark on the head of pre-
served specimens. The dorsolateral stripes are continuous from the
canthals, over the supraciliaries (also yellow), to the base of the tail.
The stripes become intensely yellow near the groin. The black dorso-

1970

Ameiva auberi: II. The Bahamian Subspecies

131

lateral fields are edged with brown to brick-red stripes in life, and the
dorsal fields have dark edges in hatchlings. The venter is dirty white to
gray, and adult males have intense black ventral markings which cover
the chinshields, throat, mesoptychium, under sides of the forelegs, and
a few anterior rows of ventrals. The greatest extent of ventral blackening
is exhibited by a male from Gibson Cay ( CM 40918, 87 mm. SVL ) that
has the anterior four ventral rows, as well as the centers of the next three
rows, entirely blackened.

The series of specimens from Fresh Creek (ASFS 10287-10293) shows
several distinct differences in color pattern from the Gibson Cay sample.
In the Fresh Creek series the vivid white middorsal stripe is broadly
edged with black, from the parietals to about the midpoint of the back.
In the smaller specimens {ca. 50-60 mm. SVL) the dorsal fields are
light-centered and black-edged. The dorsolateral fields, however, are
solid black, without evidence of light centers. These seven specimens
average 104.9 dorsal granules at midbody (95-115); 27.3 femoral pores
(25-32); 28.4 scales in the fifteenth caudal verticil (26-30); and 80.4
fourth-toe subdigital scales ( 74-89 ) .

The huge land mass of Andros Island is largely unexplored herpeto-
logically. Available samples of Ameiva auberi from Andros are inade-
quate for analysis of intra-island variation, but there are suggestions that
representative series from all parts of the island may reveal the presence
of more than one recognizable race.

COMPARISONS: Ameiva auberi kingi occupies a position of intermediacy
in scale counts between thoracica and the populations of the Berry
and Bimini Islands to the north. In GAB counts, kingi, with the Berry
and Bimini Island populations, forms a compact group distinct from all
other Bahamian races of Ameiva auberi. Femoral pore counts, on the
other hand, align kingi closely with the thoracica-felis group. In fifteenth
caudal verticil and fourth-toe subdigital scale counts kingi is significantly
different from the geographically adjacent populations, and in both these
characters, overlaps the Ragged Islands population far to the south.

The color pattern of kingi shows a strong resemblance to thoracica in
the development of the stripes anteriorly on the parietals, frontals, and
supraciliaries. The general tendency for intensification of the middorsal
stripe on the head reaches a maximum in the Fresh Creek population of
kingi. The combination of solid black dorsolateral fields and reddish
brown dorsal fields distinguishes kingi not only from thoracica, but from
all other Bahamian populations of Ameiva auberi as well. Further points
of difference between kingi and geographically adjacent populations are

132

Annals of Carnegie Museum

VoL. 41

found in the development of the black ventral coloration in adult males.
The extent of blackening of the ventral scales is much less in kingi than
in thoracica, but kingi, unlike thoracica, has black chinshields. The com-
bination of divergent scale counts with the reddish dorsal and black
dorsolateral fields makes kingi one of the most distinctive Bahamian
races of Ameiva auberi.

Ameiva auberi multilineata, new subspecies

holotype: cm 41196, an adult male, from Bond’s Cay, Berry Islands, Bahama
Islands, taken July 26, 1965, by Neil D. Richmond.

PARATYPES; CM 41195, 41197-41199, same data as holotype.

ASSOCIATED SPECIMENS: All from the Berry Islands, as follows: Cistern Cay:

CM 41201-41205. Devil’s Cay: AMNH 76299-76303 -f 3 untagged. Frazer’s Hog
Cay: AMNH 76289-76293 + 15 untagged, ASFS V10665-V10666. Little Harbour
Cay: AMNH 76294-76298. Chub Cay: ASFS V10649-V10656.

DIAGNOSIS: A Bahamian subspecies of Ameiva auberi characterized by a com-
bination of very large size (males to 120 mm., females to 71 mm. SVL), low
number of dorsal granules at midbody, fairly high number of femoral pores,
moderate number of scales in the fifteenth caudal verticil, and very low number
of fourth-toe subdigital scales; all dorsal stripes persistent and continuous onto
the head, lateral stripes sharp-edged and distinct, dorsal and dorsolateral fields
light reddish brown with dark edges; chinshields, throat, mesoptychium, and
most of the central ventrals black in adult males.

distribution: The Berry Island group, Bahama Islands. Definitely known from
Bond’s Cay, Chub Cay, Cistern Cay, Devil’s Cay, Frazer’s Hog Cay, and Little
Harbour Cay.

DESCRIPTION OF THE HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 97.6 mm., tail 237 mm. (complete); dorsal granules
at midbody (GAB) 105; femoral pores 15 and 14 (total 29); 31 scales in the
fifteenth caudal verticil; fourth-toe subdigital scales 36 and 40 (total 76). Dorsal
ground color reddish brown, top of head gray. Middorsal stripe distinct and sharp-
edged from the interparietal almost to the sacrum, edged with dark brown. Dorsal
fields light-centered, the dark brown edges bordering the adjacent stripes. Dor-
solateral stripes bright gray-white, continuous from the supraciliaries onto the tail, '
and broadly and sharply edged on both upper and lower sides with dark brown.
Dorsolateral field light red-brown centrally, dark brown to black at the edges.
Lateral stripe sharp-edged above and below, distinctly lighter than the ventro-
lateral field, continuous to the ear. Sides below the lateral stripes gray, shading
into the black ventrals. Chinshields, throat, mesoptychium, under sides of forelegs,
chest, and almost all ventrals (both central and lateral) black. Lateralmost few
rows of ventrals spotted with blue. Under sides of thighs and pre-anal scales light
gray mottled with dark gray, and spotted with blue. Under side of tail grayish
white. Dorsal surfaces of limbs brown, mottled with dark gray or black.

VARIATION: See tables. The type series from Bond’s Cay (CM 41195-41199)
has the following counts: GAB 97-107 (average 104.2); femoral pores 27-31
(29.0); fifteenth caudal verticil scales 28-31 (28.9); fourth-toe subdigital scales
72-77 (75.2).

1970

Ameiva auberi: II. The Bahamian Subspecies

133

The most distinctive feature of Ameiva auberi multilineata, as the name
implies, is the many-lined appearance of adults. This effect is produced
in part by the persistence of the dorsal light stripes, and in part by the
dark borders located in the edges of the light dorsal fields. In life the
dorsal and dorsolateral fields are brick-red to red-brown in the centers,
and dark brown at the edges. Even in hatchlings (CM 41160, 37.7 mm.
SVL), the light centers of the dorsolateral fields are apparent. The
middorsal stripe in multilineata is unusually vivid, about twice the width
of the dorsolateral stripes, and distinctly grayish rather than yellow. The
lateral stripe is pale gray, and the lower sides are slightly darker gray.
Black ventral coloration appears in males at about 65 mm. SVL, and
eventually involves the chinshields, throat, mesoptychium, chest, under
sides of the forelegs, and part of the ventrals. In the very largest males
more than half of the ventrals are blackened, but the outermost rows
of ventrals remain dark gray with bluish spots and mottling.

The short series of specimens from Little Harbour Cay (AMNH
76294-76298) shows a tendency toward development of light-spotting
in the dorsolateral fields, and to some extent in the dorsal fields. The
spots give a dilute or faded appearance to the dorsal pattern of the adult
female, and in a lesser degree to that of the adult males. The stripes are
persistent and black-edged, however, and other details of color pattern
are normal for the subspecies.

A series of magnificent specimens from Devil’s Cay (AMNH 76299-
76303 + 3 untagged) includes three males between 110 and 120 mm.
SVL — the largest Bahamian Ameiva known. In these gigantic specimens
the dorsal and dorsolateral fields are faded to a uniform light brown,
and the middorsal stipe is visible only on the anterior half of the body.
The most striking aspect of the pattern of these animals is the vivid
yellow, black-edged dorsolateral stripes diagnostic of multilineata. The
black ventral coloration involves slightly more than half of the ventrals
in these largest males.

COMPARISONS: Ameiva auberi multilineata is the largest Bahamian
race, approached in size only by the Bimini Island subspecies and by
thoracica. The Bimini Island population and multilineata share scale-
count characters that differentiate them from all other Bahamian races
of Ameiva auberi. In GAB counts, multilineata and the Bimini popula-
tion overlap only with kingi; their fifteenth caudal verticil scale counts
distinguish them from kingi, thoracica, and felis but not from other
races; and in fourth-toe subdigital counts they differ significantly from all
other Bahamian subspecies of auberi. Femoral pore counts separate
multilineata from all races except felis.

134

Annals of Carnegie Museum

VoL. 41

The northernmost two subspecies of Ameiva auberi, multilineata and [

the Bimini Island subspecies, share a significant number of features of '

color pattern and scutellation and are strongly differentiated from all r

other Bahamian races. The subtle characters that serve to distinguish |i

these two populations from each other are outlined in the following
discussion of the Bimini Island subspecies. The reddish brown dorsal |

fields of these northern populations separate them from all Bahamian
races except kingi, and the light-centered, dark-edged dorsolateral fields j|
are found only in this northern group. The persistence of the dorsal i
striping is unmatched in any other subspecies, and the reduction of ven-
tral blackening in adult males distinguishes the northern pair of races
from adjacent forms. Together, these two populations form a unit sharply
divergent from all other Bahamian races of Ameiva auberi, not only in
color pattern but also in size and scutellation.

Ameiva auberi richmondi, new subspecies

holotype: cm 34140, an adult male, from near Lyons, North Bimini Island, i

Bahama Islands, taken March 15, 1956, by M. Graham Netting and Neil D. [

Richmond.

PARATYPEs: All from North Bimini Island, as follows: CM 34139, 34141-34145, '

same data as holotype; CM 32589, Mosquito Point, March 23, 1954, M. G. Netting
and N. D. Richmond; CM 34112-34114, between Lyons and Lerner Marine
Laboratory, March 10, 1956, M. G. Netting and N. D. Richmond; GM 34130, :

Lerner Marine Laboratory, March 12, 1956, M. G. Netting and N. D. Richmond;

UF 16591-16597, Paradise Point, April 19, 1960, W. King; ASFS VI 1603, between
Alicetown and Paradise Point, March 6, 1967, Richard Thomas; UF 16598-16601,
Alicetown, April 18, 1960, W. King.

ASSOCIATED SPECIMENS: All from the Bimini Islands, as follows: East Bimini
Island: CM 34182-34185, extreme southern tip. Easter Cay: CM 34171-34172.
Gun Gay: UF 17934-17941. South Bimini Island: AMNH 68745-68749 -f 7 I
untagged, ASFS X4784, CM 32556, 34116, 34131-34132, west end; CM 32558- ,

32559, 32580-32581, Black Land; CM 34127-34128, 33180, Water Tower; ASFS I
X4689, western side; UF 22421, 2 mi. S Alicetown (North Bimini Island) at 141°.
South Cat Cay: CM 34163-34164. I;

diagnosis: A Bahamian subspecies of Ameiva auberi characterized by a combi- J
nation of large size (males to 117 mm., females to 80 mm., SVL), low number
of dorsal granules at midbody, very high number of femoral pores, moderate
number of scales in the fifteenth caudal verticil, and very low number of fourth-
toe subdigital scales; dorsal stripes persistent but black edging eventually lost,
lateral stripe not well-defined, dorsal and dorsolateral fields light-centered ( reddish
brown to tan ) ; throat, mesoptychium, chest, and first four or five ventral rows '
black in adult males. j

distribution: The Bimini Island group, Bahama Islands. Definitely known i
from East Bimini Island, Easter Cay, Gun Gay, North Bimini Island, South Bimini
Island, and South Cat Cay. '

1970

Ameiva auberi: II. The Bahamian Subspecies

135

DESCRIPTION OF THE HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 91.1 mm., tail 178 mm. (regenerated); dorsal
granules at midbody (GAB) 108; femoral pores 15 and 15 (total 30); 30 scales
in the fifteenth caudal verticil; fourth-toe subdigital scales 37 and 37 (total 74).
Dorsal ground color medium reddish brown, head slightly lighter and more
yellowish. Middorsal stripe light gray, continuous from interparietal almost to
sacrum, not noticeably dark-edged. Dorsolateral light stripes from temporal areas
to groin, light gray and distinct, and edged on the anterior half of the body with
black flecking both above and below. Dorsolateral fields reddish brown, lighter
than the dorsal fields, with concentrations of black stippling along the upper and
lower edges anteriorly. Lateral stripe light gray, dark-edged above and sharp but
fading into a dark gray area below and above the ventrals. Dorsal surfaces of limbs
medium brown. Inner half of chinshields, throat, chest, and approximately one-third
of anterior central ventrals black. Posterior central ventrals blue-gray, spotted with
white; lateral ventrals dark gray, spotted with blue. Under sides of thighs gray-
blue, spotted with white and pre-anal seales white. Tail indistinctly striped, brown
above and gray below.

VARIATION: See tables. The type series from North Bimini Island has the fol-
lowing counts: GAB 91-108 (99.8 ± .89); femoral pores 28-34 (30.0 ±.36);
fifteenth caudal verticil scales 28-31 (29. 6 ±.22); fourth-toe subdigitals 68-75
(71.7 ± .40).

Ameiva auberi richmondi is another strongly lineate race from the
northwestern corner of the Great Bahama Bank. Hatchlings of riehmondi
have vivid cream dorsal and dorsolateral stripes, usually somewhat
expanded but never black-edged on the head. The stripes remain bril-
liant cream to yellow in subadults and adult females but are consider-
ably diluted and reduced to paler yellow or yellow-gray in adult males.
In adults the dorsolateral stripes are continuous only onto the neck, and
the lateral stripes to the forelegs. The dorsal and dorsolateral fields are
black or dark brown at the edges in hatchlings, the former with broad
red-brown central fields, the latter with very narrow lighter centers. As
the lizards grow, the light central zones of the dark fields enlarge and
eventually obliterate the dark edges. The dorsal field becomes completely
light long before the dorsolateral field is unicolor. The lateral stripe is
cream to gray, sharp-edged above, but blending gradually into the
darker color below.

In an adult male paratype from North Bimini Island (ASFS V11603,
98 mm. SVL), the middorsal stripe is light gray and visible from the
posterior edge of the parietal scales almost to the sacrum. The light
gray dorsolateral stripes originate in the temporal areas and extend
nearly to the groin. The dorsal and dorsolateral fields are almost
exactly the same shade of light brown, and their black edges persist only
as scattered melanophores along the edges of the dorsolateral stripes.
The lateral stripes, and the sides below, are uniform light gray.

136

Annals of Carnegie Museum

VoL. 41

The tail of hatchlings is bright blue; that of adults is gray above, and
white marked with bluish mottling below, and is not black-striped along I
the side. I

Suffusion of the gular area with black pigment begins at an early age 1
in richmondi, and all males over 65 mm. SVL have extensively black-
ened throats and mesoptychia. The usual maximum extent of the black j
includes the throat, mesoptychium, chest, under sides of the forelegs, and :
approximately one-third to one-half of the anterior ventrals. The lateral j
ventrals are blue-spotted, and the pre-anal and thigh scales are dirty |
white. In a common variant type, represented in series from East Bimini
Island, South Bimini Island, and Easter Cay, the black of the ventral ji

scales is restricted to the first two or three transverse rows. Tlie remain- ;i

der of the ventrals is white, mottled with light gray. |

In virtually all aspects of color and scutellation, the populations of the i
various islands and cays of the Bimini Islands form a close-knit and ■
relatively invariable group. '

COMPARISONS: The close similarity of Ameiva auberi richmondi with i
multilineata and their trenchant differences from the other Bahamian
races of Ameiva auberi have been mentioned. It remains only to point
out the characters that differentiate these two subspecies of the north-
western arm of the Great Bahama Bank. These races are almost identical
in all counts except femoral pore scales, which are significantly more
numerous in richmondi. Here, admittedly, the difference is small, but it
is based on large samples from each island group, and the standard j.

errors of the mean values are reassuringly small. Size is another distin- j

guishing characteristics of these races. Male richmondi attain a maximum
snout- vent length of 117 mm., but they do not compare in bulk to the
120 mm. males of midtilineata.

There are several subtle, but constant characters of color pattern that ‘

distinguish richmondi from multilineata. Paramount among these is the :

early loss, in richmondi, of the black edges of the dorsolateral fields,
producing uniform-colored rather than the dark-edged fields of multi- I
lineata. In richmondi the dorsal stripes of adults are less extensive
anteriorly, and the lateral stripe less well defined, in comparison with
multilineata. Finally, the ventral black pattern of male richmondi rarely j:
includes the chinshields, which are always dark in multilineata.

In sum, the subspecies richmondi and 7nultilineata are remarkably j
similar and obviously share a common ancestry. Nevertheless, there I
are constant characters of both color and scutellation that can be used to
differentiate them. Furthermore, their relative isolation from one another ,

1970

Ameiva auberi: II. The Bahamian Subspecies

137

at opposite ends of the northwestern arm of the Great Bahama Bank
leads me to regard their subtle but consistent differences as adequate
evidence of subspecific divergence.

Ameiva auberi focalis, new subspecies

holotype: cm 41147, an adult male, from Ship Channel Cay, Exuma Cays,
Bahama Islands, taken on July 23, 1965, by Neil D. Richmond.

PARATYPEs: All from Ship Channel Cay, as follows: CM 41139-41146, same data
as holotype; ASFS V10718-V10730, taken on July 6, 1966, by J. B. Brown, A.
Schwartz, and R. Thomas.

diagnosis: A Bahamian subspecies of Ameiva auberi characterized by a combi-
nation of fairly large size (males to 107 mm., females to 83 mm. SVL), very high
number of dorsal granules at midbody, moderate number of femoral pores and
fifteenth caudal verticil scales, and low number of fourth-toe subdigital scales;
middorsal stripe lost in adults, dorsal fields olive, dorsolateral fields black, chin-
shields and throat blue, black ventral markings reduced to a bar in the center of
the mesoptychium and another on the anterior two or three rows of ventrals,
snout pink.

DISTRIBUTION: Known only from the type locality. Ship Channel Cay, but pos-
sibly occurs on several of the northernmost Exuma Cays, Bahama Islands.

DESCRIPTION OF THE HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 110.3 mm., tail ea. 239 mm (complete); dorsal
granules at midbody 112; femoral pores 14 and 14 (total 28); 30 scales in the
fifteenth caudal verticil; fourth-toe subdigital scales 37 (one toe missing). Dorsal
ground color, top of head and dorsal sides of limbs olive gray. Middorsal stripe
faintly indicated as a lighter gray area. Dorsolateral light stripes also faint and
only feebly lighter than the ground color. Dorsolateral dark fields slightly emargi-
nate, but vivid black and continuous from the foreleg insertions to the groins.
Below the dorsolateral black fields are slightly lighter gray areas, blending into
the gray-blue ventral color. None of the stripes extends onto the neck, which is
entirely unlined, shading gradually from the ventral blue-gray to the dorsal
olive gray. Mental scale, rostral scale, and anterior upper and lower labials lighter
than other head scales. Entire ventral surface from chinshields to pre-anals light
gray-blue, with a small patch of black scales on the mesoptychium and black
stippling on the anterior two transverse rows of ventral scales. Central posterior
ventrals, femoral-pore-scale rows, and under side of tail white with blue-gray
smudging. Tail weakly but distinctly striped laterally, olive-gray dorsally.

VARIATION (See tables): Ameiva auberi focalis is unique among the
Bahamian races of the species in the extreme reduction of the black
ventral coloration of adult males. The maximum development consists
of a black bar across the mesoptychium, a second bar on the anterior
two and one-half rows of ventral scales, and a speckling of black throat
granules. The under sides of the forelegs, chinshields, and throat are
never solid black. The entire ventral surface is usually gray-blue in
males, white in females. Adult males occasionally lack any trace of the
usual black ventral pigmentation (CM 41146).

138

Annals of Carnegie Museum

VoL. 41

The middorsal stripe is pale lemon-yellow in life, continuous from the
frontal scale to the sacrum in the smallest specimens examined (ASFS
V10729-V10730, ca. 65 mm. SVL), but is not black-edged on the head.
The middorsal stiipe is persistent in adult females, but may be com-
pletely lost in large males (CM 41140, 107 mm. SVL). The dorsal fields
are olivaceous in life, and the dorsolateral stripes are tan to yellowish,
vaguely continuous to the supraciliaries in small individuals. In old
adults the dorsolateral stripes are conspicuously brighter than the mid-
dorsal stripe.

The intense black dorsolateral fields retain their consistency through-
out life, or may be slightly invaded by their narrow brick-red edging.
The lateral stripes and lateral fields are only poorly differentiated,
greenish in life, and blend gradually with the blue-gray ventral color.
The tail is gray above in adults, lighter than the olive dorsal fields,
black-striped laterally, and gray-blue below the stripe, shading to white
midventrally.

comparisons: Although immediately distinguishable by the reduction
of black ventral markings in adult males, Ameiva auberi focalis shares a
combination of characters of the geographically adjacent subspecies. In
scale counts focalis is not significantly different from thoracica in gran-
ules around midbody and fifteenth caudal verticil scales but has a sig-
nificantly higher number of femoral pores and a significantly lower
number of fourth-toe subdigital scales. From the populations farther
south in the Exuma Cays chain, focalis is significantly different in both
fifteenth caudal verticil scales and fourth-toe subdigital scales, differences
which are, in both cases, partially bridged by the Long Island popula-
tion. There is broad overlap with the southern populations in number of
granules at midbody, but the number of femoral pores is significantly
different. Tlius focalis overlaps the range of variation of the adjacent
population on the north in two characters and of the population on the
south in three characters. Ameiva auberi foealis is distinguished from the
adjacent subspecies on the north by two characters, and from the popu-
lation on the south by one character.

Elements of the color pattern of foealis may also be traced to the
neighboring populations. The pink nose and tendency for early loss of
the middorsal stripe are shared with the populations in the Exuma Cays.
Retention of the solid black dorsolateral fields throughout life, lack of a
well-defined, conspicuously lighter lateral stripe, and olive rather than
gray dorsal fields are characteristic of Ameiva auberi thoraeiea.

The cays immediately to the south of Ship Channel Cay along the

1970

Ameiva auberi: II. The Bahamian Subspecies

139

chain of the Exuma Cays have been poorly explored herpetologically.
Scanty evidence suggests that there may be an actual hiatus in the
range of Ameiva auberi between Ship Channel Cay and the northern-
most island inhabited by the Exuma Cays subspecies (A. Schwartz,
in lift. ) . The details of the distribution of Ameiva auberi in the northern
Exuma Cays are in need of investigation to clarify the exact range of
focalis and the relationship of focalis to the Exuma Cays subspecies.

Ameiva auberi obsoleta, new subspecies

holotype: cm 43976 (formerly ASFS V10805), an adult male, from Adderley’s,
Long Island, Bahama Islands, taken on October 12, 1966, by Richard Thomas.

paratypes: All from Long Island (localities arranged from north to south) as
follows: ASFS V8598, Cape Santa Maria, January 24, 1964, C. R, Warren; ASFS
V10806W10810, Adderley s, October 12, 1964, Richard Thomas, ASFS V10868,
same locality and collector, October 14, 1966; ANSP 24169, MCZ 42309-42314,
Simms, 1936, R. McLean; ASFS V8553, Grays Settlement, January 27, 1965, C. R.
Warren, ASFS V8594-V8597, same locality and collector, January 23-30, 1965;
ASF'S V10827-V10829, 4.9 mi. SE Clarence Town, October 13, 1966, Richard
Thomas; ASFS V10830-V10834, 2.8 mi. SW Mortimers, October 13, 1966, Richard
Thomas.

associated SPECIMENS: All from the Exuma Cays (localities arranged from
north to south) as follows: Warderick Wells Cay: AMNH 76244-76249. Bell
Island: AMNH 76254. Sampson Cay: CM 4110I-4I103, 41117. Bitter Guana Gay:
AMNH 76252-76253, CM 41070. Great Guana Gay: CM 41055-41057, 41060-
41062, 41064-41066. Cave Cay: CM 41023-41025, 41040-41045. Great Exuma
Island: ASFS V6993-V7000, 5.4 mi. SE Rolleville; ASFS V10795-V10797, 9 mi.
NW George Town. Elizabeth Cay: ASFS V7100. Little Exuma Island: ASFS
V7044, V7093, 5.7 mi. SE The Ferry; ASFS VI0794, I mi. ESE Forbes Hill; ASFS
V10792-V10793, ca. 2 mi. SE Forbes Hill; ASFS V10768-V10773, ca. 0.5 km. W
Williams Town.

DISTRIBUTION: The southern Exuma Cays and Long Island, Bahama Islands.
Definitely known from Warderick Wells Cay, Bell Island, Sampson Cay, Bitter
Guana Cay, Great Guana Cay, Cave Cay, Great Exuma Island, Elizabeth Cay, and
Little Exuma Island, all in the Exuma Cays group, and from several localities on
Long Island. The specimens reported from Crooked Island (Cochran, 1934) belong
with this subspecies (see below).

DESCRIPTION OF THE HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 86.6 mm., tail 225 mm. (complete); dorsal granules
at midbody (GAB) 103; femoral pores 13 and 13 (total 26); 32 scales in the
fifteenth caudal verticil; fourth-toe subdigital scales 40 and 40 (total 80). Dorsal
ground color and top of head medium gray-brown, middorsal stripe faded. Dorso-
lateral light stripes narrow but bright, extending from the forelegs to the tail.
Dorsolaterial fields intensely black, slightly emarginate. Lateral light stripes
dirty gray-white, sharp-edged both above and below. Ventrolateral fields black
with a series of large, discrete, light spots. Dorsal surfaces of limbs dark brown.
Chinshields, throat, under side of forearms, chest, and two-thirds of central ventrals
black. Posterior third of central ventrals off-white smudged with gray. Lateral

140

Annals of Carnegie Museum

VoL. 41

I

ventrals tessellated (light blue on black). Pre-anal scales, under sides of thighs, I
and under sides of tail dirty white. Dorsal surface of distinctly striped tail gray-
brown. Middorsal field brown to tan, dorsolateral stripes buff, lateral stripes pale
gray to buff, and tip of snout orange in life ( R. Thomas, field notes).

VARIATION ( See tables ) : The type series from Long Island has the following i'
counts: GAB 96-117 (107.07 ± 1.1) femoral pores 22-30 (25.7 ± .37); fifteenth I
caudal-verticil scales 27-33 (30.5 zt .30); fourth-toe subdigital scales 72-90 ,

(83.6 ±1.1). !

The most striking and immediately obvious characteristic found in '
all populations of Ameiva auberi ohsoleta is the early loss of the mid-
dorsal stripe and eventual fading of all lineate parts of the dorsal pattern, i
In the Long Island series the middorsal stripe is only faintly visible in
the smallest specimen (ASFS V8598, 40 mm, SVL) and is completely '
obliterated in all larger individuals. The resulting broad middorsal zone !
between the dorsolateral stripes is light brown to gray. The faint buff ;
dorsolateral stripes are continuous from the forelegs to the sacrum even !
in large specimens. The intense black dorsolateral fields are emarginate ;
in very old specimens, but retain their integrity in the type series. The i
lateral stripe is light gray, and the dark gray field below it is light- |
spotted.

The black ventral coloration of males appears at an early age. One
68-mm. male has the chest, throat, mesoptychium, and even the under
sides of the forelegs densely blackened. In older individuals the black
also covers the chinshields. Unique characteristics of obsoleta are the '
pointed posterior margin of the black patch on the ventrals, and the |
brilliantly tessellated light-blue-on-dark-blue pattern of the lateral ven- '
tral scales. The tail is bright blue in hatchlings, and vividly striped in all :
adults.

Despite its extensive geographic range northward through the Exuma
Cays chain to Warderick Wells Cay, obsoleta is relatively invariable in
color-pattern characteristics. Samples from the northern Exuma Cays
exhibit tendencies toward longer retention of the middorsal stripe, and !
more brilliant striping. For example, the middorsal stripe is faintly
visible in large males from Sampson Cay (CM 41102, 90 mm. SVL, and
CM 41117, 95 mm. SVL). In general, the specimens from Sampson Cay
are more brilliantly striped than animals from the Long Island popu-
lation. Persistent stripes also appear in the series from Warderick WeUs '
Cay (AMNH 76244-76249) and in a single specimen from Bitter Guana
Cay (CM 41070). The general tendency of obsoleta toward reduction
of the striped pattern reaches its ultimate expression in a series of speci-
mens from Cave Cay (CM 41023-41025, 41040-41045). In the smallest

1970

Ameiva auberi: II. The Bahamian Subspecies

141

animals of this series, males between 65 and 75 mm. SVL, the dorsal
field is a uniform gray-brown, the dorsolateral stripes are faintly visible
above solid, but emarginate, black dorsolateral fields, and the lateral
stripes appear as vague lighter gray areas above the ventrals. In males
there is progressive reduction of all stripes and fields, culminating in a
virtually uniform gray-brown dorsum separated from the lighter gray
sides by a line of black stippling representing the dorsolateral field. The
appearance is that of a nearly unicolor lizard. Similiar patterns, resulting
from extreme reduction of the dorsolateral fields, are found in occasional
specimens of Ameiva auberi thoracica from Eleuthera Island, and rarely
in the populations of the northern Ragged Islands (Cochran, 1934).

In addition to the variable reduction of the dorsal pattern in obsoleta,
there are several constant features of the pattern which also obtain
throughout the range of the subspecies. The most important of these is
the ventral pattern of adult males. The combination of the reduced
black patch on the ventrals, in the shape of a posteriorly pointed V, with
the brilliant light and dark blue tessellation of the lateral ventrals,
occurs in all populations. A second constant ' character of obsoleta is the
presence of a contrastingly colored snout, either noticeably lighter, or of
a hue different from that of the sides of the head. In specimens from
Long Island the snout is orange in life; in Warderick Wells Cay and
Cave Cay animals it is pink in life, and in the specimens from Sampson
Cay, it was light gray in life.

Scale-count data were analyzed separately for each island in the
Exuma Cays chain in an attempt to detect clinal variation along this
extensive archipelago, but no significant clinal pattern was found.

It is appropriate to mention at this point the published record of
Ameiva thoracica ( = auberi) from Crooked Island, across the Crooked
Island passage from Long Island ( Cochran, 1934 ) . I have examined the
two specimens on which this record is based, USNM 81497 and 81498,
collected by Dr. Paul Bartsch in 1930. These adult males have 106 and
110 dorsal granules at midbody, 23 and 25 femoral pores, 30 scales
(each) in the fifteenth caudal verticil, and 72 and 79 fourth-toe sub-
digital scales. In almost all elements of the color pattern they are identical
with males of comparable size from Long Island, a notable exception
being the faint but distinct middorsal stripe of both individuals. Because
of the virtual meristic identity of these two specimens with Ameiva auberi
obsoleta from Long Island, and in view of the failure of recent expedi-
tions to find a population of Ameiva auberi on Crooked Island, I am
dubious of the validity of the Bartsch record. Of course there is a pos-

142

Annals of Carnegie Museum

VoL. 41

sibility that a natural population of auberi occurs on Crooked Island,
but it requires verification.

COMPARISONS: Ameiva auberi obsoleta occupies a central position
geographically on the Great Bahama Bank, and in many apects of color
and scutellation obsoleta is central to the other Bahamian races of
Ameiva auberi. The basic color pattern of oliveTo-gray dorsal fields and
black dorsolateral fields separated by cream-to-yellow stripes is repeated
in the Bagged Island population to the south, and in focalis, thoracica,
and other more northerly populations. Obsoleta diflFers from both focalis
and thoracica in the early obsolescence of the striping and the unique
ventral pattern of adult males, and from felis in the unspotted dorso-
lateral black fields. From the populations of the Ragged Islands, obsoleta
may be distinguished by its tessellated lateral ventrals, less discrete
lateral stripes, and spotted lateral fields. Moreover, the Ragged Island
lizards generally have a persistant middorsal stripe and more brownish
dorsal fields.

In scale counts obsoleta finds close similiarity with focalis and thora-
cica to the north. Obsoleta is distinct from the northern populations in
femoral pores, and overlaps in the other three scale-count characters
examined. From the adjacent Ragged Islands subspecies, obsoleta is
distinguished by significant differences in mean numbers of dorsal
granules at midbody, fifteenth caudal verticil scales, and fourth-toe
subdigital scales. Finally, size is a useful character in differentiating
obsoleta from geographically adjacent races of Ameiva auberi. Obsoleta,
with size maxima of 107 mm. and 71 mm. in males and females respec-
tively, is smaller than focalis ( 109 mm. and 80 mm. ) and larger than the
Ragged Island subspecies (87 mm. and 75 mm.).

Ameiva auberi bilateralis, new subspecies

HOLOTYPE: CM 40985, an adult male, from the south end of Great Ragged
Island, Bahama Islands, taken July 10, 1965, by Neil D, Richmond.

PARATYPES: All from Great Ragged Island, as follows: CM 40984, 40988-40995,
ASFS V11363-V11364, same data as holotype; CM 40955-40959, Gun Point, July
9, 1965, Neil D, Richmond.

ASSOCIATED SPECIMENS: All from the Ragged Islands, as follows: Little Ragged
Island: CM 40982. Hog Cay: CM 40943-40945. Raccoon Cay: USNM 81475.
Nurse Cay: CM 40922-40937. Flamingo Cay: USNM 81451-81454.

DIAGNOSIS: A Bahamian subspecies of Ameiva auberi characterized by a combi-
nation of small size (males to 87 mm., females to 75 mm, SVL), moderate number
of dorsal granules at midbody, very low number of femoral pores, low number of
scales in the fifteenth caudal verticil, and moderate number of fourth-toe sub-
digital scales; middorsal stripe usually persistent, dorsal fields brown, dorsolateral
fields black and sharp-edged, lateral stripe clear white, broad and vivid, with a

1970

Ameiva auberi: II. The Bahamian Subspecies

143

black lateral field below and above the ends of the ventrals; throat, chinshields,
mesoptychium, and anterior half of ventrals black in adult males.

DISTRIBUTION: The Ragged Island group, Bahama Islands. Definitely known from
Raccoon Cay, Nurse Cay, Great Ragged Island, Little Ragged Island, Flamingo
Cay, and Hog Cay.

DESCRIPTION OF THE HOLOTYPE: An adult male with the following measurements
and counts: snout-vent length 87 mm., tail 174 mm. (complete); dorsal granules
at midbody 100; femoral pores 13 and 13 (total 26); 29 scales in the fifteenth
caudal verticil; fourth-toe subdigital scales 37 and 37 (total 74). Dorsal ground
color and top of head light brown, middorsal stripe visible as a faint trace posterior
to neck. Dorsolateral stripes whitish, narrow, not black-edged above. Dorsolateral
fields sharp-edged, solid black, continuous from above forelegs onto tail. Lateral
stripes dead white, well-defined and sharp-edged, almost as wide as dorsolateral
fields. A second black field below each lateral stripe, the two separated by the broad
white lateral stripe. All stripes continuous only from foreleg insertion; none starts on
head. Chinshields, throat, chest, mesoptychium, under sides of forelegs, and almost
half of central ventrals black. Posterior central ventrals smudged with gray, lateral
ventrals bluish smudged with black. Under sides of thighs light gray spotted with
bluish, dorsal sides of limbs dark brown stippled with black. Tail brown above,
with a strong black lateral stripe enclosed in vivid white stripes. Under side of
tail gray.

VARIATION (See tables): The type series from Great Ragged Island (CM 40955-
40959, 40984-40985, 40988-40995, ASFS V11363-V11364) has the following
counts: GAB 97-116 (104.5 ±1.0); femoral pores 20-27 (24.1 ± .27); fifteenth
caudal verticil scales 26-32 (28.9 ± .22); fourth-toe subdigital scales 71-82 (76.3
± .53).

Ameiva auberi bilateralis is characterized by the broad, sharp-edged
and very white lateral stripes, bordered by intense black dorsolateral
and lateral fields. Hatchlings (CM 40994-40995, ca. 32 mm. SVL) have
a brilliantly striped pattern, including a broad middorsal stripe continu-
ous from neck to tail, slightly narrower dorsolateral stripes originating
on the neck, and lateral stripes fully as wide and bright as the middorsal
stripe. In adult females the dorsolateral and lateral stripes retain their
brilliance, unlike the faded middorsal stripe. The middorsal stripe of
adult males is also faded, but remains visible as a gray area in the
brownish dorsal fields even in the largest individuals. The dorsolateral
fields are intense black in individuals of both sexes and all ages. Old
males have slightly emarginate dorsolateral fields. An exception is seen
in males from Flamingo Cay (USNM 81451-81454), which have the
black fields reduced to a line of spots, and other lineate elements of the
pattern faded.

The black ventral coloration of males is well developed in individuals
as small as 60 mm. SVL (CM 40927) and ultimately extends to cover
the chinshields, throat, mesoptychium, and a small triangular area of

144

Annals of Carnegie Museum

VoL. 41

anterior ventrals. The remainder of the central ventrals is frequently
smudged with black in larger males, and the lateral ventrals are dark
blue, with faint suggestions of a tessellated pattern.

The series from Nurse Cay (CM 40922-40937) presents several dif-
ferences from the Great Ragged Island sample. Nurse Cay lizards have
a greater tendency toward loss of the middorsal stripe, and a less pro-
nounced lateral light stripe. The dorsolateral black field of the adult
males is considerably reduced, as in the Flamingo Cay population. Adult
males from Nurse Cay have extensive blackening of the anterior half
of the central ventrals.

Ameiva auberi bilateralis is probably the smallest of the Bahamian
races, smaller even than the diminutive felis. Although less than 40
specimens were examined from throughout the range of bilateralis, the
samples contained many adults and a good representation of all size
classes. The largest male examined was only 87 mm. snout to vent, and
the largest female 75 mm. snout to vent.

COMPARISONS: Bilateralis is very similar to the adjacent subspecies
obsoleta in many features of color pattern, but is distinguished from
obsoleta, indeed from all other Bahamian subspecies of Ameiva auberi,
by the brilliant white lateral stripes and development of unmarked black
fields below the lateral sti'ipes. Other color characteristics that distin-
guish bilateralis from obsoleta are the lack of a boldly tessellated pattern
on the lateral ventrals, and a tendency toward longer retention of the
striped elements of the pattern. Despite the similarities in the color
patterns of obsoleta and bilateralis, these subspecies are strongly differ-
entiated in scutellation. Bilateralis is significantly different from obsoleta
in dorsal granules at midbody, fifteenth caudal verticil scales, and
fourth-toe subdigital scales, and overlaps the range of obsoleta only in
femoral-pore-scale counts. Bilateralis occupies a unique position among
the southern Bahamian races. Rather than continuing the generally
gradual clinal reduction in all scale counts, which obtains from New
Providence through Long Island, bilateralis has sharply lower counts
which overlap only the populations of Andros Island and the Berry
Island and Bimini Island groups in the northern Bahamas. Bilateralis is
distinguished from these northern populations by many characters of
the color pattern and by size.

The Green Cay Population

A small series of specimens of Ameiva auberi (CM 40171-40179) was
taken on Green Cay on July 18, 1965, by Neil D. Richmond. These lizards
constitute something of a zoogeographic enigma, since they resemble

1970

Ameiva auberi: IL The Bahamian Subspecies

145

most closely Ameiva auberi bilateralis rather than the geographically
adjacent populations: obsoleta of the Exuma Cays chain and kingi of
Andros Island.

The Green Cay lizards have a strongly lineate color pattern, with all
dorsal stripes persistent throughout life. The middorsal stripe is con-
tinuous from the neck to the tail, the narrow dorsolateral stripes originate
on the neck, and the lateral stripe is fairly sharp and white. The dorsal
fields are gray-brown, the dorsolateral fields are black and emarginate in
older adults. The lateral fields are dark gray, neither black as in
bilateralis nor spotted as in obsoleta. Adult males have the chinshields,
throat, mesoptychium, and anterior two or three rows of ventrals black-
ened, and the remaining central ventrals black-smudged. The lateral
ventrals are light blue, the under sides of the thighs and tail are white
to light gray. The tail is boldly striped.

In scutellation the Green Cay population is almost identical with
bilateralis in dorsal granules at midbody, femoral pores, and fourth-toe
subdigital scales, and has a slightly higher count of fifteenth caudal
verticil scales. Of course the sample is too small to be statistically
reliable, but meristically the Green Cay population is obviously very
similar to, if not identical with, bilateralis. Until a larger sample is
available it seems best to withhold formal assignment of the Green
Cay population to one of the described subspecies. Nevertheless, the
similarity of these lizards to bilateralis suggests that the population has
been directly derived from the Ragged Islands.

Discussion

The land reptile fauna of the Bahama Islands has been the subject
of much discussion and speculation because, as a totally derivative
fauna in a geologically recent archipelago, it promises to yield concrete
data on basic problems such as the potential for overwater dispersal
and the evolutionary rates of various species. The earlier literature
on the relationships of the land biota of the Bahamas has been sum-
marized by Rabb and Hayden ( 1957 ) , who also presented a useful
analysis of the complex and as yet incompletely understood geological
history of the Bahamas. In a more recent contribution Schwartz ( 1968a)
has elaborated on the faunistic relationships and history of the Bahamian
amphibians and reptiles. Schwartz's scheme explains the present dis-
tribution and origins of the faunas, and helps to place Ameiva auberi
in perspective with regard to the other reptile species of the Bahamas.
Schwartz postulates early invasion of the Bahamas by Cuban and
Hispaniolan faunas, subsequent extinction of this fauna over most of the

146

Annals of Carnegie Museum

VoL. 41

Great Bahama Bank with limited survival on peripheral and southern
refuges, and the eventual re-invasion of the Great Bahama Bank in late
Pleistoeene times. The colonists of this second major wave have invaded
the newly emergent lands and old refuges to varying extents. It is with
this final group of immigrants that I associate the Bahamian populations
of Arneiva aiiheri.

The pattern of distribution of Ameiva auheri on the Great Bahama
Bank strongly suggests very recent arrival from Guban sources. Auheri
has managed to invade most of the central islands of the bank, but
is restricted peripherally. Gat Island represents the maximum extent
of eastward dispersal, and although the island lies on the edge of the
Great Bank, it is only a short distance across a narrow water barrier
from Eleuthera Island. Auheri has also reached the Bimini Islands. This
might be regarded as an indication of a longer history in the archipelago,
but Ameiva auheri richmondi of the Bimini Islands is extremely similar
to A. a. multilineata of the adjacent Berry Islands, which suggests that
richmondi may be newly arrived on the Biminis. Despite having reached
the peripheral Gat Island and Bimini Islands, and isolated interior islands
such as Green Gay and Flamingo Gay, the general distribution of
auheri on the Great Bahama Bank bespeaks recency. Auheri has failed
to extend its range to any of the more isolated large islands at the
eastern periphery, such as Gonception Island, Bum Gay, and San Sal-
vador Island, and apparently has not crossed the Grooked Island
Passage to the south (but see discussion under A. a ohsoleta, above).
Most significantly, auheri has crossed none of the relatively narrow
water barriers between the northern islands of the Great Bahama Bank
and the large islands of the Little Bahama Bank to the north.

The exact timing of the invasion, or invasions, of the Great Bahama
Bank by Guban Ameiva auheri is not clear. The only concrete evidence
lies in the fossils of Ameiva auheri from New Providence Island reported
by Etheridge (1966). These specimens, identical with recent species,
came from cave deposits assigned to the pre-Pamlico portion of the
Wisconsin glacial stage of the Pleistocene. Of the nineteen species
of amphibians and reptiles that comprise the Great Bahama Bank
herpetofauna, no less than thirteen have been successful in extending
their ranges beyond the islands of the Great Bank proper. Seven species
have invaded the Little Bahama Bank to the north, using the Great
Bahama Bank as a stepping stone, and ten of the thirteen species have
either reached eastern peripheral islands, or have crossed the Grooked
Island Passage to the south. The remaining species include the proba-

1970

Ameiva auberi: II. The Bahamian Subspecies

147

bly old, Cat Island endemic turtle Chnjsemijs felis, and four species of
lizards (including Ameiva auberi) regarded as conspecific witli Cuban
stocks. This fairly clear evidence of recent immigration to the Great
Bahama Bank is somewhat clouded by the relatively great morphological
radiation of the Bahamian Ameiva auberi. The total of eight sub-
species of auberi here recognized far exceeds the number of recogniz-
able geographic races in other species of the reptile fauna. By compar-
ison Leiocephalus carinatus, which has a much larger geographic range,
has evolved no more than four Bahamian races (Babb, MS.). Sphaero-
dactylus deeoratus has a distribution on the Great Bahama Bank very
similar to that of Ameiva auberi, with the single exception that deeoratus
has reached Bum Cay. A modern revision of deeoratus admits only three
recognizable races on the Great Bahama Bank (Thomas and Schwartz,
1966), although it must be noted that Sphaerodactylus are the master
mariners among lizards, and thus the Bahamian populations may be
of exceedingly recent origin. Other Great Bank reptiles and amphibians
follow this general pattern of relatively little fragmentation into geo-
graphic races. The eight races of Ameiva auberi have apparently evolved
as isolates during the gradual and successive invasion of various parts
of the Great Bahama Bank by over-water dispersal.

The obvious recency of immigration of auberi to the Bahamas, and the
remarkable morphological radiation which has produced a constellation
of well-marked subspecies, underscores the latent variability and
evolutionary plasticity of this group of lizards. Both Ameiva and the
related genus Cnemidorphorus contain many notoriously labile species.
In topographically diverse land masses and in island archipelagoes this
lability is expressed in the formation of numerous relatively divergent
geographic races. The only comparable situation outside the family
Teiidae of which I am aware is the explosive radiation of some species
of the lacertid genus Laeerta on islands and archipelagoes in the Med-
iterranean Sea.

148

Annals of Carnegie Museum

VoL. 41

Fig. 1. Map of the Bahama Islands, showing distributions of the Bahamian sub-
species of Ameiva auberi, as follows ( north to south ) : ri - — richmondi
( Bimini Islands ) ; mu — multilineata ( Berry Islands ) ; ki - — kingi ( Andros
Island); th — thoracica (New Providence, Eleuthera, and Little San
Salvador Islands); ie — felis (Cat Island); io ■— focalis (Ship Channel
Cay); ob — obsoleta (Exuma Cays from Warderick Wells Cay south, Long
Island ) ; bi — bilateralis ( Ragged Islands ) .

1970

Ameiva auberi: IL The Bahamian Subspecies

149

d

W

m

<

H

o

uo

h

Z

p

O

U

w

p

<

u

cn

s

z

o

h

<

5

>

mcioosoococD

1/5 lO CO O ^ U- CO

05005'^oqc\10c<l

t^»«O5O5Q0O5G0

QO^-4^4'eqoo

CDSDCDt-U-t-.t-lO-

coOcoioOf-H-^oq

coiOTOincoco'^m

+1 +1 +1 +1 +i +i +i +1

osocowoino-^
CD 03 iq O cq CD CD 05
d c4 CD d P-? CD 05 ID

C<1C<IC3C0'^C<|CC|(»

«©ioei>-(^c<iQOco

gggggggg

cococqt-'^co'^oi

cocococococococo

C<! C<l OJ Ol C<l Ol 0-1 oi

OOo'o5pHCOCDQO'-h
f-iedoicciccic'l phdi

+i +i +1 +i +i +i 71 +i

OC<IOT05^C0'— 1"^
oi lo oq lo CD o] q q
d d 00 d p-4 rH d oo'
cocooqcocococoDi

cDOOr-H,-ioi.-Hin

IDIOCO^^C^IOOCO

gggggggz

locooiooqp-iou-

CO CO CO CO CO CO CO oi

cDAAAAiT-cqo

oi cq oi oi oq cd

CO ® 05 ro o' iOOU-
oq Di CO D-i CO oq cq oq

71 71 71 71 71 71 71 71

OOCDoOOCO'^O

q r- q t- 0-1 q o q
d d CD CD d d d d
CO cq Di CM CM eq cq cq

03 oq f-H CD D1 lO t-
'^locoo'^cqt-cq

gggggggg

OO'-HIOO^OIOCO
o rH >-H oq Di Di oq 1-H

p

a

i-i

O

X

-o

O

05U-C000O00CD1O

OT»00050a3a503

O »0 CO D1 CO

CO q oi « q q q q

71 71 71 71 71 71 71 71

OOt'-IO'O'OOO

00 t- q q q q q q
d d 05 d d c-i CO

G50050-H^00

150

Annals of Carnegie Museum

VoL. 41

TABLE 3

Comparison of GAB Counts in Bahamian Ameiva auheri

AS

a

c

o

•S

0)

s

'•S

05

cj

or

O

§

o

richmondi

49

98.8+1.2

X

—

—

+

+

+

+

+

multilineata

52

100.7+1.6

X

— •

+

+

+

+

kingi

31

99.9+2.5

X

+

+

+

+

—

thoracica

106

109.2+1.8

X

—

—

—

+

felis

42

111.7 + 1.8

X

—

+

+

focalis

22

112.6+3.4

X

+

+

ohsoleta

75

107.7+1.2

X

+

hilateralis

27

103.6+2.0

X

Subspecies arranged from north to south. For explanation of table see footnote,
pages 116-117.

TABLE 4

Comparison of Femoral Pore Counts in Bahamian Ameiva auheri

AS

03

a

-2

C

0)

-S

o

•2

03

2

AS

*§)

a

•2

Co

s

s

'I

c

C

richmondi

56

30.0+.40

X

+

+

+

+

+

+

+

multilineata

50

28.7+.50

X

+

—

• —

+

+

kingi

30

26.7+.80

X

—

+

+

+

+

thoracica

111

26.8+.60

X

+

+

+

+

fells

41

28.3+.60

X

—

+

+

focalis

22

29.0+.50

X

■T

+

ohsoleta

81

25.0+.40

X

—

hilateralis

35

24.1+.60

X

Subspecies arranged from north to south. For explanation of table see footnote,
pages 116-117.

1970 Ameiva auberi: II. The Bahamian Subspecies 151

TABLE 5

Comparison of Fifteenth Caudal- Verticil
Scale Counts in Bahamian Ameiva auberi

-2

c

3

c

G

3

Co

•2

3

c

CJ

s

•+«1

o

"o

richmondi

62

30.2±.4()

X

—

+

+

+

— ,

—

+

multilineata

52

30.5±.40

X

+

+

+

—

—

+

kingi

29

28.3±.60

X

+

+

—

+

thoracica

113

31.6±.4()

X

—

—

+

+

felis

44

3L6±.40

X

—

+

+

focalis

22

3L2±.6()

X

+

+

obsoleta

82

30.0±.4()

X

+

Inlateralis

38

28.7±.40

X

Subspecies arranged from north to south. For explanation of table see footnote,
pages 116-117.

TABLE 6

Comparison of Fourth-Toe Subdigital Scale
Counts in Bahamian Ameiva auberi

§

3

•2

.2 "3

2

2

-2

3

V.

o

•22

■c

s

M-.,

o

'-t—

richmondi

55

72.7±.80

X

+

+

+

+

+

+

multilineata

52

72.9±L0

X

T"

+

+

+

+

+

kingi

30

76.5±1.6

X

+

+

+

thoracica

109

81.1±L2

X

+

+

felis

40

8L3±L2

X

+

—

+

focalis

20

76.7±L2

X

+

obsoleta

73

79.6±.80

X

“F

bilateralis

36

75.9+1.0

X

Subspecies arranged from north to south. For explanation of table see footnote,
pages 116-117.

III. DISCUSSION

Albert Schwartz^

Research Assoeiate, Section of Amphibians and Reptiles
Carnegie Museum, Pittsburgh, Pennsylvania

With the description of 24 Cuban subspecies and eight Bahamian
subspecies of Ameivci aiiberi Cocteau, the species has now been shown
to be extremely polytypic and genetically plastic. Such diversity may
be surprising to some readers, but actually it is to be expected. The
geographically correlated variation of the three Hispaniolan species
(Ameiva chnjsolaema, A. taeniura, and A. Uneolata), whose study
preceded tlie present one on A. auberi, has served in a way as a prec-
edent. The study of the Hispaniolan species indicated that on at
least one of the Greater Antilles the genus Ameiva has differentiated
on the subspecific level in response to edaphic and geographical
isolating factors. Although Cuba does not present the topographic
and physiographic diversity of Hispaniola, it is nonetheless far from
a flat, featureless and topographically uniform plain over which
Ameiva auberi is neatly and evenly distributed. The response of some
Cuban reptiles (notably Anolis equestris in Oriente Province;
Schwartz, 1964b) to often-dramatic ecological differences and geo-
graphical isolating features, by the development of distinct subspecies,
is well known. Considering also that: Cuba is an elongate island
(about 1200 kilometers long with a maximum width of from 35 to
200 kilometers); there are several extensive chains of cays associated
both with it and the Isla de Pinos; Cuba varies ecologically, from
extremely arid coasts to mesic forested mountainous highlands with
elevations to 2005 meters (Pico Turquino); and that there are many
ranges and massifs with lesser but still respectable heights and areas,
it is not surprising that Ameiva auberi varies greatly in several char-
acters in a geographically correlated manner. As far as variation in
the Bahamian archipelago is concerned, this too is expected, consid-

1 Miami-Dade Junior College, Miami, Florida 33167

152

1970

Ameiva auberi; III. Discussion

153

eriiig both the plasticity of A. auberi on Cuba and the (at least
presently) completely disjunct nature of many of the Great Bank
islands that are inhabited by the species.

Apparently the closest relative of A. auberi is A. dorsalis Gray from
Jamaica. In Barbour and Noble's revision of Ameiva (1915:425), dor-
salis, thoracica, and auberi are keyed out together. Examination of
specimens of A. dorsalis shows that the Jamaican species has many
resemblances to A. auberi insofar as scutellation is concerned. Both
have 10 transverse rows of ventral scales and comparable numbers
of longitudinal rows of ventral and fourth-toe subdigital scales. There
are, however, trenchantly different features of scutellation in A. dor-
salis (greater number of femoral pores, much smaller dorsal granules,
much less enlarged preantebrachials, differences in scutellation of the
temporal region) which I regard as being of specific rather than sub-
specific value. Thus, A. dorsalis is to be regarded as a species distinct
from A. auberi. That dorsalis is related to auberi is attested by the
gross similarity of dorsal pattern (a median dorsal pale longitudinal
area, pointed anteriorly, which gradually expands as it progresses
posteriorly to embrace the base of the tail) to that of some of the
eastern Cuban subspecies of A. auberi, namely sabidicolor and granti.
Such resemblances may well be fortuitous, but on the other hand no
other West Indian Ameiva has this type of median dorsal marking.
The proximity of Jamaica to the southern Oriente coast (about 150
kilometers) seems to reinforce these resemblances as indicators of
relationship.

Barbour and Noble (1915:422), in their schema representing the
relationships of the West Indian and continental species of Ameiva,
suggested that auberi and dorsalis were distantly and independently
derived from A. undulata of Middle America, that thoracica is closely
related to auberi, and that dorsalis is directly related to A. chrijso-
laema of Hispaniola. At the moment I do not agree with this latter
relationship. Certainly dorsalis is much more closely related to auberi
than to chrysolaema. Barbour and Noble also suggested that A. taeni-
ura {=A. lineolata sensu Barbour and Noble) is a distant relative of
auberi. This relationship seems more plausible, although it is now
generally accepted (Cochran, 1941:273-274) that taeniura is more
closely related to A. polops of St. Croix, and thence to A. maijnardi
in the southeastern Bahamas and A. tvetmorei in Puerto Rico. These
three species have the dorsal caudal scales with oblique keels, although
this character is least definitely shown in taeniura.

154

Annals of Carnegie Museum

VoL. 41

It seems likely that dorsalis-auberi (ineluding thoracica) and pos-
sibly chrysolaemci represent a single stock of western Greater Antilles
ground lizards, of which dorsalis has been long isolated on Jamaica.
Ameiva aiiberi has been long in residence on Cuba and the Isla de
Pinos, and is a recent arrival in the Bahamas. A. dorsalis merits spe-
cific status, whereas thoracica does not.

Savage (1966:742-743) has suggested that A7neiva and Cnemido-
phoriis represent, respectively, South American and Middle American
derivatives of a common ancestral stock. These two genera evolved
separately in areas north and south of the Isthmian Link. In the late
Cenozoic, with the closing of the Isthmus of Panama, Ameiva invaded
CenLal America and Cnemidophonis South America, with limited
evolution in the newly invaded area in each case. If such a history
is essentially correct (and the continental data support it), a relation-
ship between A. tindidata and auberi-dorsalis-{?chrysolaema) seems
less likely. Final and “permanent” closure of the Isthmian Link took
place in the Pliocene, although prior to that time there had been a
series of stepping-stone islands scattered between South America and
Mexico. It is possible that a South American Ameiva stock reached
Middle America sometime earlier, during the Cenozoic, and thence
reached Cuba (and later, Jamaica, and still later, the Bahamas). In
fact, the diversity of Greater Antilles Ameiva (contrast for example
such species as A. ehrysoJaema and A. lineolata on Hispaniola, or even
such dissimilar species as A. maijnardi and A. auberi, occurring on
islands in or near the same archipelago) makes it seem very unlikely
that this high degree of differentiation on the species level has oc-
curred since Pliocene, or later, invasion of the Greater Antilles from
the continent.

On the other hand, considering the diversity of Lesser Antilles
Ameiva (see Baskin and Williams, 1966, for an incisive discussion),
it seems more likely that all Antillean Ameiva are related to one or
more South American stocks. It also seems likely that the Antilles as
a faunal region has been colonized by leap-frogging and island-
hopping, primarily by South American stocks. Such a history, being
independent of the Isthmian Link events, would allow sufficient time
for Ameiva to have reached many of its Antillean outposts. Although
absent in the Gayman Islands and some of the northern and south-
eastern Bahamas, Ameiva is one of the most widespread reptilian
genera in the Antilles, exceeded in total distribution only by Anolis
and Sphaerodactylus. This history would also allow sufficient time for

1970

Ameiva auberi: III. Discussion

155

the various major “kinds” or “styles” of Ameiva to have evolved. And
it would not demand that the great specific and subspecific diversity
of the western Caribbean Ameiva be accomplished more or less
synchronously since the Pliocene.

Along these lines, it is pertinent that presumably the Great Bank
in the Bahamas was inundated throughout the Pliocene and variably
or partially submerged during the Pleistocene (Babb and Hayden,
1957:8). This implies that all the Bahamian subspecies of A. auberi
have had a very recent origin (at least post-Pliocene, and possibly
even more recent) since the Great Bank was generally not available
earlier for colonization from Cuba. The same statement is true
of some of the Cuban subspecies; extraria, extorris, zugi, and galbiceps
occupy low cays and islets that were inundated during the Pliocene.
These cays and islets could not have been invaded from the Cuban
mainland or the Isla de Pinos until after they emerged from the
ocean. The total history of A. auberi in the Bahamas and Cuba
indicates, by the very recency of some of the land, masses occupied
by quite distinctive subspecies, that there had long been a resident
stock of Ameiva (presumably remotely derived from South America)
on Cuba. There diversification occurred, and at an early date, by
overseas transport, Ameiva arrived in Jamaica (probably from eastern
Cuba). At a much later date, Ameiva reached the Bahama Islands'
Great Bank. I suggest the derivation of dorsalis from auberi, rather
than the reverse, because dorsalis is relatively constant in its pattern
features throughout Jamaica, suggesting that it is the derived rather
than basic species of the auberbdorsalis pair.

Given a long period of residence of A. auberi on Cuba and the
ecological, physiographic, and geographic diversity of Cuba as well,
the subspeciation of A. auberi there is expected. The eastern sub-
species (of which it has been noted that sabidicolor and granti are
rather like A. dorsalis in dorsal pattern) seem to have retained the
original style of dorsal pattern of the basic stock. The expanding and
fading median dorsal line, in fact, is a general feature of A. auberi
from northeastern Camagiiey to Cabo Cruz (although ustulata on the
south-central Oriente coast has diverged from this basic Oriente pat-
tern) in the subspecies sublesta, citra, granti, sabidicolor, and hardyi.
These races presently (with the exception of sublesta and citra) are
associated with coastal regions bounded by interior mountains. It
seems possible that all the eastern subspecies from sublesta to hardiji
represent radial derivatives from one ( or several ) eastern mountainous

156

Annals of Carnegie Museum

VoL. 41

areas that served as refuges in Plioeene-Pleistocene time. Subsequent
differentiation along the coast has resulted in the development of the
large and orange-throated southwestern hardiji (whereas the more east-
ern subpecies mtiilata, sabiilicolor, and granti are black-throated) and
the geographically intermediate ustulata, which lacks the expanding
and fading dorsal line. The extreme peripheral members of this group,
citra and the insular but closely adjacent sublesta, are subspecies that i
are farthest removed from any eastern massif but presumably were
derived from such areas (see Fig. 1 for a graphic rendition of these
and the following statements).

The central Camaguey subspecies Uanensis is allied most closely to
galbiceps from the Cayos de las Doce Leguas. The apparent absence
of Ameiva from most of Camagiiey Province has been previously
mentioned (Schwartz, Section I), and it has been admitted that this
absence may be more apparent than real. The Doce Leguas cays are
low and surely would have been inundated with any appreciable rise
in sea level. Thus colonization of these islands must have taken place
since their last emergence from the sea. The occurrence on the Doce
Leguas cays of two endemic subspecies of lizards, as well as other
reptiles (one of which, Sphaerodactijhis argus, is purely Jamaican in
origin) indicates that sufficient time has elapsed since the appearance
of these cays above the ocean surface to allow not only the arrival
of Cuban reptiles there, with subsequent racial differentiation, but
the waif arrival of a Jamaican reptile also. Presumably, despite the i
present absence of Ameiva from the adjacent Camagiieyan coast,
galbiceps and Uanensis represent common derivatives of an ancestral
Cuban form. The patterned dorsal fields and non-black throats ally
these two races.

Turning to the extreme western end of Cuba and the Isla de Pinos,
there are two major groups of subspecies involved: (1) the very large
and yellow-throated procer with a relatively wide distribution, and
(2) the smaller races cacuminis, denticola, patdsoni^ secta, marcida,
and ziigi. Of the second complex, paulsoni-sectamiarcida-ziigi repre-
sent a southern Pinar del Rio-Isla de Pinos-Archipielago de los
Canarreos group. The floral and faunal relationships of Pinar del
Rio to the Isla de Pinos have been pointed out repeatedly (Schwartz,
1959a: 138; 1959b: 61; 1959c: 118), and it is not surprising that the
subspecies of Aiiieiva in these two regions show similarities. Of the
four races, paulsoni and zugi (usually with non-black throats, and zugi
with its extreme obsolescent dorsal pattern) are the most divergent. It

1970

Ameiva auberi: III. Discussion

157

must be recalled, however, that marcida on the southern Isla de Pinos
is apparently in the process of losing the black-throated condition, and
that one population of otherwise pale-throated ziigi (Cayo Candles)
contains some black-throated individuals. There seems little doubt that
zugi has been derived along the Archipielago de los Canarreos from
the adjacent and partially black- throated marcida. Possibly this group
of four subspecies represents radial dispersal from an Isla de Pinos
haven, paulsoni having arrived in its present south Cuban coastal
area by overseas transport or by overland dispersal while the Isla de
Pinos and Cuba were connected, and marcida having differentiated
from secta on the southern third of the Isla de Pinos (because of
the intervening Cienaga de Lanier) at a more recent date, with zugi
a derivative of marcida on the Archipielago. Even slight rises in sea
level would bring about division of the Isla de Pinos into two separate
islands, and this may have helped reinforce the genetic distinction of
secta and marcida from each other.

The distribution of procer, centering as it does in the Sierra de los
Organos, would indicate that it has had a long, independent history
in this range, which was part of the Cuban archipelago during
Pliocene-Pleistocene time. Since restoration of the surrounding low-
lands, procer has expanded its range into the lower areas surround-
ing the Sierra de los Organos and has reached as far to the southwest
as the base of the Peninsula de Guanahacabibes and as far east as
Cabanas.

The two Guanahacabibes subspecies, denticola and cacuminis, have
been discussed in detail previously ( Schwartz, Section I ) . It is difficult
to associate denticola with its adjacent neighbor procer, and denticola
is better regarded as a western derivative of the paulsoni (and thus
indirectly of secta ) stock, with cacuminis as either a terminal or north-
ern peninsular coast population with non-black throats.

To the east of the range of procer, there is a series of northern
coastal subspecies from Habana Province to Las Villas Province {auberi-
pullata-gemmea-garridoi) of which one (pullata) has a black chest
and throat patch. In addition to these subspecies, ahducta occupies the
Peninsula de Hicacos, and extraria and extorris occur on the Archipiel-
ago de Sabana, off the northern Las Villas coast. The same style of
throat and chest coloration and pattern as pullata occurs in the south-
ern Las Villas atrotliorax, whereas the latter’s closest geographic rela-
tive is peradusta, a subspecies that carries ventral blackening to an
extreme in the species. A. a. atrotliorax occupies that part of the

158

Annals of Carnegie Museum

VoL. 41

southern coast immediately adjacent to the high Sierra de Trinidad,
whereas the north coast races occupy in general the coasts adjacent to
more interior uplands — the Escaleras de Jaruco, the Sierra Morena,
and the Sierra de Jatibonico. A single subspecifically unidentifiable
specimen from northeastern Las Villas is from the Sierra de Jatibonico,
and the small series from north of the Sierra de Cubitas completes the
records of Ameiva from this northern coast between Habana and
central Camagiiey provinces. The occupation of lowlands immediately
adjacent to interior uplands on both north and south coasts (of these
subspecies, peradusta, which I consider an atrothorax derivative, is the
farthest removed from any immediately adjacent montane mass) of
central Cuba suggests that in each case the more or less widespread
Ameiva had been forced to retreat during extensive Pleistocene inun-
dation to a series of disjunct islands, whence it subsequently expanded
with emergence of surrounding lowlands. If such is the case, then
auheri and pidlata may have been derived from an Escaleras de Jaruco
population, with the peripheral a])duct(i evolving on an offshore island
which is today the Peninsula de Hicacos, gemmea and garridoi from j
the Sierra Morena (or possibly garridoi from the Sierra de Jatibonico),
and atrothorax (and, more removed on the west side of the Bahia de
Cienfuegos, peradiista) from the Sierra de Trinidad. The differences
pointed out between coastal ])idlata and the few specimens from the
interior of Matanzas Province at San Miguel de los Banos may indi-
cate that these latter have been derived from still another isolated ,
population on the Bejucal-Madruga-Limonar anticline, in which range
San Miguel lies. The similarity in chest and throat pattern between
the north coast puUata and the south coast atrothorax may be inter-
preted (in this area) as the retention of a primitive throat pattern by
two presently disjunct populations which previously may have been
continuous across the area now occupied by the La Habana-Matanzas
Red Plain (Llanura roja la Habana-Matanzas; see Marerro, 1951:474
et seep), and across which these two races may have restablished
contact.

The general association of Cuban A. auheri with lowland coastal
regions allied to nearby mountains, and the contention that these ad-
jacent mountains were insular refugia during the period when Cuba ||
was an archipelago, support the differentiation of A. auheri into a
series of races, these races having been at least partially differentiated j
during the period of inundation of the lowlands. The fact that the
center of distribution of one race (proeer) is still primarily in the

1970

Ameiva auberi: III. Discussion

159

interior uplands lends added support to this theory. The only serious
exceptions to the montane refiigia are llanensis in Camagiiey Province,
and caciiminis and denticola on the Peninsula de Guanahacabibes. The
latter two races, as has been pointed out, may be more recent western
derivatives of paulsoni (and its Isla de Pinos relatives), whereas
llanensis occupies an interior region adjacent to the Sierra de Cubitas.
If the unnamed population north of the Sierra de Cubitas is presumed
to represent the original residual Cubitas population, then llanensis
can hardly also be so considered. In southern Camagiiey Province is
the Sierra de Najasa. This range is geographically intermediate between
the distribution of llanensis and the Doce Leguas subspecies galhiceps,
and possibly llanensis represents the Najasa isolate that has subse-
quently invaded the north as far as the Sierra de Cubitas and also
reached the Laberinto de las Doce Leguas.

As I mentioned above, it is obvious that some insular subspecies of
A. miheri (extmria, extorris, ziigi, galhiceps, and possibly siihlesta)
have had more recent development and differentiation from their main-
land relatives. These races are in most cases closely related to the
subspecies on the adjacent mainland: i.e., snhlesta to citra, zugi to
marcida, and galhiceps to llanensis. The differences between extraria
and gemmea, however, are much greater than those between the above
racial pairs. Considering the long and linear extent of the north
coast caijeiia from Piinta Hicacos in the west to Cayo Sabinal in the
east, and considering that Cayo Sabinal is inhabited by the westermost
member of the Oriente complex of races, it is possible that extraria
(and extorris, which is a local derivative of extraria) has been derived
by serial colonization and invasion of the northern cayeria from the
east, rather than from the coast directly to the south. The resemblances
between extraria and the eastern subspecies are such that extraria
may well be more closely related to suhlesta and citra than the more
geographically dose gemmea and garridoi.

If the above interpretation of the history of Ameiva in Cuba is
correct, then the suggestion that the distribution of A. aiiheri, com-
posed of widely disjunct populations separated by at times broad
areas where no ground lizards are known, is a relictual one is reinforced.
Those areas from which no Ameiva are known are thus interpreted as
regions that A. auheri has not yet penetrated. These regions generally
are also farthest removed from montane refugia. Although A. auheri
is absent today from some regions like the highlands of the Sierra
Maestra and extreme eastern Oriente because of ecology and elevation,

160

Annals of Carnegie Museum

VoL. 41

the absence of the species from other apparently suitable areas may be
explained by this suggested history of the species in Cuba and the
Isla de Pinos.

Considering the Bahamian subspecies as a whole (see McCoy,
section II, for details), the consistency of presence of black throat
and chest pigment (reduced but still present in focalis) suggests that
the parent Cuban population of the Bahamian stock was black-
throated, and logically would be from the northern Cuban coast. The
closest present point of the Bahama Islands and Cuba is along the
southwestern arm of the Creat Bahama Bank paralleling the Camagiiey
caijena between Cayo Coco and Cayo Cuajaba. Here the Great Bank
is separated from Cuba by the narrow (50 kilometers) but deep
(locally to over 1000 fathoms) Old Bahama Channel. At this closest
point of the Cuban insular platform and the Great Bahama Bank,
there are but two tiny and remote Bahamian islets ( Guinchos Cay and
Cay Lobos) that are not known to harbor Arneiva. To the east, along
the southeastern margin of the Great Bank and about 110 kilometers
north of Cabo Lucrecia in Oriente, lie the southernmost Bahamian
islands with any considerable land mass, the Ragged Island group.
Arneiva a. hilateralis, one of the most distinctive of the Bahamian
subspecies, occupies these islands.

It might be expected that hilateralis would be like one of the north
Cuban coastal subspecies. Arneiva auberi granti, by virtue of its occur-
rence from Banes eastward to Baracoa, would be a logical Cuban
possibility. A. a. granti is also black-throated, which would satisfy
one of the pigmentary requirements of a Cuban ancestor for the
Bahamian lizards. Along the northern Cuban coast, other black-
throated subspecies are pullata and extorris. The former lies far to
the west in Matanzas Province and the latter has a limited distribution
on one of the cays of the Archipielago de Sabana. The two other
eastern Cuban north-coast races, sublesta and eitra, have colored
throats.

Despite the proximity of granti and hilateralis, these two subspecies
have little in common other than throat and chest color. A. a. hilateralis
is a pale but distinctly lined subspecies with broad, sharp-edged and
white lateral stripes bordering black lateral fields. The dorsal stripe is
present but faded, maintains its intensity and width for the length
of the dorsum, and does not become progressively fainter and wider
as is proceeds posteriorly. A. a. granti on the other hand is a rich
reddish brown above, has a yellow and contrasting median dorsal

1970

Ameiva auberi: III. Discussion

161

line which expands and becomes paler posteriorly, narrow and often
fragmented dorsolateral lines, and lateral fields with scallops or
internal spotting. A. a. bilateralis is generally faded in aspect with
boldly contrasting solid black lateral fields, whereas A. a. gra7'iti is a
richly colored and brightly patterned lizard. Two subspecies could
hardly be more different in pattern and color than these two. As far as
meristic data are concerned, bilateralis and granti have identical
means (75.9) of fourth-toe scales and comparable means of femoral
pores (24.1 and 25,7), but bilateralis has a higher mean number (28.7)
of fifteenth verticil scales than does granti (25.1). All things considered,
Cuban granti and Bahamian bilateralis seem not to be closely related.

Of the north Cuban coast subspecies, bilateralis most closely resem-
bles extraria and extorris. Of this pair, extorris is black-throated but
is known to occur on only one islet. All three subspecies are faded in
appearance, but extorris has an unpatterned dorsum and extraria has
a yellow median dorsal line that fades posteriorly. Finally, both extraria
and extorris lack lateral fields, quite in contrast to the prominence of
these features in bilateralis. Of the three subspecies, the means of
fourth-toe scales in extraria (76.3) and bilateralis (75.9) are compara-
able, whereas that of extorris is much lower (69.5). The series of ex-
torris is small, however. Femoral pore means (24.1 in bilateralis, 23.8
in extraria, 25.3 in extorris) are variable, with bilateralis intermediate
between the two Cuban subspecies, and means of fifteenth verticil
scales (28.7 in bilateralis, 26.5 in extraria, 25.4 in extorris) show that
bilateralis exceeds both Cuban subspecies in this count. It is interest-
ing to note that, of the Bahamian subspecies, bilateralis ranks next to
last in fifteenth verticil means (range of Bahamian means 28.3 to 31.6).
Means of fifteenth verticil counts in all Cuban subspecies vary from
24.9 to 29.5. Thus, in general Cuban subspecies as a group have fewer
scales in the fifteenth caudal verticil than do Bahamian subspecies as
a group. The low fifteenth-verticil mean of bilateralis suggests that,
of all the Bahamian races, it may indeed be most closely related to the
Cuban stock.

Everything considered, the differences between the pertinent north
Cuban coastal subspecies of A. auberi and the most southern of the
Bahamian subspecies, bilateralis, suggest that the Ameiva invasion of
the Great Bank may well have been along that section of Cuba that
most closely approximates the Great Bank today, i.e., the northern
cayeria between Punta Hicacos in Matanzas and the Bahia de Nuevitas.
At least there are some similarities between some of the subspecies

162

Annals of Carnegie Museum

VoL. 41

along this coast and bilatercilis. Although the Ragged Islands are elosest
geographically to the Oriente coast, there is little similarity between
hilateralis and granti.

This postulated relationship is of eourse questionable. Resemblances
between extraria and hilateralis may be due to responses to small-islet
environments and more arid conditions. It should be recalled that
granti occupies one of the wettest and most densely forested portions
of Cuba, whereas extraria inhabits a series of cays that are open, more
or less arid, and lack dense forests. It is pertinent that ziigi, a sub-
species of the exposed and dry islands in the Archipielago de los
Canarreos, is also pale. Remarkably, of all the Cuban subspecies,
hilateralis most closely resembles the far-removed zugi. Of the pale and
faded Cuban subspecies, only zugi regularly retains the black lateral
fields. This surely is an instance of convergence.

McCoy ( section II ) has commented succinctly on the evolution and
radiation of A. auheri within the Bahama Islands themselves. In the
Bahamas, Ameiva occupies solely the Great Bahama Bank and has not
succeeded in invading the Little Bahama Bank or in crossing the
Crooked Island Passage. It has reached Cat Island, which is peripheral
to the Great Bank proper. The remarkable vagility of A. auheri is
demonstrated by its ability to reach some remote islets on the bank.
Green Cay and the Bimini Islands are two striking cases in point. Of
the Bahamian subspecies, kingi, richmondi, and multilineata seem to
form a rather eompaet group of predominantly lineate subspecies that
occupy islands to the west of the Tongue of the Ocean dividing the
Great Bank into two nearly equal portions. To the east of the Tongue
of the Ocean, thoracica, felis, and focalis form a less integrated group,
with felis most divergent in dorsal and lateral pattern. A. a. ohsoleta
might also be included with this assemblage. A. a. hilateralis stands
alone in its strikingly contrasting dorsal pattern, although it too is
probably related to the eastern Great Bank subspecies, most closely
to the adjacent ohsoleta. There seems little reason to reject a hypothesis
that the Bahamian subspeeies represent two basie stocks (but not
likely two different invasions) that were early separated by the Tongue
of the Ocean. After this early division, distinctive subspecies on each
island or island group evolved.

A. a. felis requires speeial comment. This subspecies occupies the
only island (Cat) peripheral to the Great Bank (although there must
have been a former connection, in contrast to several remote Bahamian
Islands that have never been connected) inhabited by A. auheri. Me-

1970

Ameiva auberi: III. Discussion

163

Coy (section II) stated that felis stands alone among the Bahamian
races in that it possesses spotted lateral fields. Patterned lateral fields
are of regular occurrence in many Cuban subspecies, however. The
occurrence of this character in felis, so far removed geographically
from Cuba and separated from it by the two intervening subspecies
ohsoleta and hilateralis is especially intriguing. Considering the isolated
position of Cat Island in reference to the Great Bank and its member-
ship in the eastern seaward volcanic arc, the resemblance of felis to
some Cuban subspecies in this character may be of added significance.
It seems possible that the Cat Island A. auberi have had a somewhat
different history from the balance of the Bahamian subspecies. Detailed
studies of two other Bahamian reptiles (Sphaerodactijlus cleeoratus,
Thomas and Schwartz, 1966:7; Thomas, 1968; Anolis distiehus,
Schwartz, 1968b) have shown that the Cat Island populations are
peculiar and neither readily assignable to adjacent subspecies nor
sufficiently and constantly divergent to be regarded as nomencla-
turally distinct. In both cases the Cat Island populations show peculiar
admixtures of characteristics of populations from adjacent islands. The
contrast between the situation in S. decoratus and A. distiehus on one
hand, and A. auberi on the other, is quite striking in this regard, since
felis is an exceptionally distinct subspecies. Apparently the reptile
fauna of Cat Island has had a varied history, and presumably this is
due to the location of Cat, peripheral to the Great Bank mass itself,
and possibly to a geological persistence it may have enjoyed.

In summary, I interpret the history of Ameiva auberi as follows.
A pre-auberi stock arrived on Cuba from South America, by island-
hopping and overseas transport, at some early date. Evolution of this
stock into subspecies began on Cuba, and differentiation was aided by
the successive partial inundations that occurred during Cuba’s geologi-
cal history, resulting in archipelagos variously formed and later vari-
ously recombined. Sometime in its history, Ameiva reached Jamaica,
probably from the southern Oriente coast, and in Jamaica arose A.
dorsalis. During the Pliocene-Pliestocene, Cuba was once more a series
of islands, which the present montane masses roughly outline. On
these islands, Ameiva continued its local evolution, and with the
recession of the seas and final emergence of the Cuban lowlands, the
lizards were able once more to recolonize not only the resultant low-
lands but also to reach the Great Bahama Bank from the northern
Cuban coast in the region between Las Villas and Camagiiey. With
establishment of Ameiva in the Cuban lowlands from montane refugia,

164

Annals of Carnegie Museum

VoL. 41

various subspecies have evolved because of isolation and intensifica-
tion of differences that had their inceptions while the lizards occupied
their Pleistocene refugia. On the Bahama Islands, two (or perhaps
three) basic stocks have resulted in two major divisions of the sub-
species of those islands, one east of, the other west of, the Tongue of
the Ocean. I interpret all extant subspecies of A. auheri as being quite
recent in development. Such an interpretation is mandatory since much
of the land area now occupied by A. auheri has been permanently
available only since the Pleistocene.

1970

Ameiva auberi: HI. Discussion

165

o 3

^ 0 sS

bO

cd C3
>

0

^ U

0

0

c „ „ _

QJ ;r o

O ^ > o

■a ® M ~

CD J

^ 't

“ ^

QJ Sh

^ r< "S S

0

0

a xn

0 ’-^

^ G g

a O

0 ;> C/2

Ҥ I

“ U

£f> (D

J « S
2^0
G

0 ^
Cm

0) cd

a ^

p

p < ^ ^

0 ctf

^ p

>^ cd

S eg

0

h <u

S CJ o

5 ^ P be

.2 ^ cs

^ -TS H ^
2 p . ‘p

-cs .S'

■n c3 0 ,n
0 > -JH O

^ S c ii
rt .S <I

>. bO

^ -cm •;: 2

. fi o ^

tn (D

S O

qZ <D C
_Q 2 o
« p
P 00 c/2 -Eh

- s

t— I JH tsi O

CO "-—I

0 N Sm O

^ ;p ^ ^

* (11 ^

'13 pp ^

^ c ^

2 - ^
p 0 p

-P P3 .

P bo3i

u

s ^ s

S P

a

^ §

CD <D

- 1 ■§ ^

- . 4->>

. OO C/0

bX) ^ c^

^ S 8

effected serially via the coastal cayeria.

166

Annals of Carnegie Museum

VoL. 41

References Cited

Barbour, Thomas

1914. A contribution to the zoogeography of the West Indies, with especial
reference to amphibians and reptiles. Mem. Mus. Comp. Zool., 44(2):
209-359, 1 pi.

1916. Additional notes on West Indian reptiles and amphibians. Proc. Biol. Soc.
Washington, 29: 215-220.

1937. Third list of Antillean reptiles and amphibians. Bull. Mus. Comp. Zool.,
82(2): 77-166.

Barbour, Thomas, and Arthur Loveridge

1929. Typieal amphibians and reptiles. Bull. Mus. Comp. Zool., 69(10): 205-
360.

Barbour, Thomas, and G. K. Noble

1915. A revision of the lizards of the genus Ameiva. Bull. Mus. Comp. Zool.,
59(6): 417-479.

Barbour, Thomas, and Benjamin Shreve

1935. Coneerning some Bahamian reptiles, with notes on the fauna.

Proc. Boston Soc. Nat. Hist., 40(5): 347-366’.

Baskin, Jonathan N., and Ernest E. Williams

1966. The Lesser Antillean Ameiva (Sauria, Teiidae) reevaluation, zoogeo-
graphy and the effects of predation. Stud. Fauna Curacao and other Carib-
bean Islands, 23(89): 144-178, 2 figs., 2 pis.

Buide, Mario S.

1966. Reptiles de la Peninsula Hicacos. Poeyana, ser. A., no. 21: 1-12.

Cochran, Doris M.

1934. Herpetological collections from the West Indies made by Dr. Paul Bartsch
under the Walter Rathbone Bacon Scholarship, 1928-1930. Smithsonian
Misc. Coll., 92(7): 1-48.

1941. The herpetology of Hispaniola. Bull. U. S. Natl. Mus., 177: i-vii, 1-398,
120 figs., 12 pis.

Cocteau, J. T.

1838 in de la Sagra, Historia fisica, politica y natural de la Isla de Cuba. Paris,
or 4(1): 1-124.

1839.

Conde, Jose Alvarez

1958a. Historia de la botanica en Cuba. La Habana, Publ. Junta Nacional de
Arqueol. y Etnol., pp. i-xvi, 1-353, numerous unnumbered plates.

1958b. Historia de la zoologia en Cuba. La Habana, Publ. Junta Nacional de
Arqueol. y Etnol., pp. i-xv, 1-369, numerous unnumbered plates.

Cooper, John E.

1958. Ascent to Cueva del Aura. Maryland Nat., 28(1-4): 3-16, numerous un-
numbered plates.

Cope, Edward Drinker

1862. Synopsis of the species of Holcosus and Ameiva, with diagnoses of new
West Indian and South American Colubridae. Proc. Acad. Nat. Sci.
Philadelphia, 14: 60-82.

1970

Ameiva auberi: III. Discussion

167

Dumeril, a. M. C., and G. Bibron

1839. Erpetologic Generale. Paris, 5: i-viii, 1-854.

Etheridge, Richard

1966. Pleistocene lizards from New Providence. Quart. Jour. Florida Acad. Sci.
28(4): 349-358.

Grant, Chapman

1960. Differences in shade of some reptiles from the north and south coasts of
Orientc, Cuba. Herpetologica, 16(3): 174.

Hardy, Jerry D., Jr.

1956. Notes on the Cuban iguana. Herpetologica, 12(4) : 323-324.

Hecht, Max K.

1954. The comparison of recent and fossil amphibian, reptilian, and mammalian
faunas in the Bahamas. Year Book Amer. Phil. Soc., 1954: 133-135.

Jennings, O. E.

1917. A contribution to the botany of the Isle of Pines, Cuba, based upon the
specimens of plants from that island contained in the herbarium of the
Carnegie Museum under date of October, 1916. Ann. Carnegie Mus.,
11(1): 19-290, 23 pis.

Kline, Harry (Editor)

1964. The yachtsman’s guide to the Bahamas. Tropic Isle Publishers, Coral
Gables, Florida, 296 pp.

Maerz, a., and M. Rea Paul

1950. A dictionary of color. New York, McGraw-Hill Book Co., pp. i-vii, 1-23,
137-208, 56 pis.

Marie-Victorin, PYiere, and Frere Leon

1944. Itineraires botaniques dans Pile de Cuba (deuxieme serie). Contrib. Inst.
Bot. Univ, Montreal, 50; 1-410, 257 figs.

Marrero, Levi

1951. Ceografia de Cuba. La Habana, Alfa, pp. i-xvi, 1-736, 478 figs,

Rabb, George B.

MS. A study of variation in iguanid lizards of the Leiocephalus carinatus com-
plex. Thesis, University of Michigan, 1957.

Rabb, George B., and Ellis B. Hayden, Jr.

1957. The Van Voast-American Museum of Natural History Bahama Islands
Expedition record of the expedition and general features of the islands.
Amer. Mus. Novitates, 1836: 1-53, 15 figs.

Savage, Jay M.

1966. The origins and history of the Central American herpetofauna. Copeia,
1966, (4): 719-766, 26 figs.

Schwartz, Albert

1959a. Variation in lizards of the Leiocephalus cubensis complex in Cuba and
the Isla de Pinos. Bull. Florida State Mus., 4(4): 97-143, 10 figs.

1959b. The status of Eleutherodactylus pinarensis and a new species of the genus
from western Cuba. Herpetologica, 15(2): 61-69, 2 figs.

1959c. A new species of toad, Bufo cataulaciceps, from the Isla de Pinos and
western Cuba. Proc. Biol. Soc. Washington, 72: 109-120, 3 figs.

1959d. A new species of frog of the Eleutherodactylus ricordi group from cen-
tral Cuba. Amer. Mus. Novitates, 1926: 1-16, 2 figs.

168

Annals of Carnegie Museum

VoL. 41

1960. Variation in the Cuban lizard Leiocephalus raviceps Cope. Proc. Biol.
Soe. Washington, 73: 67-82, 10 figs.

1964a. New subspecies of Leiocephalus from Cuba. Quart. Jour. Florida Acad.
Sci., 27(3): 211-222, 4 figs.

1964b. Anolis equestris in Oriente Province, Cuba. Bull. Mus. Comp. Zool.,
131(12): 403-428, 7 figs.

1968a. The geckos (Sphaerodactylus) of the southern Bahama Islands. Ann.
Carnegie Mus., 39(17): 227-271.

1968b. Geographic variation in Anolis distichus Cope ( Lacertilia, Iguanidae) in
the Bahama Islands and Hispaniola. Bull. Mus. Comp. Zook, 137(2):
255-310.

Shattuck, George Burbank (editor)

1905. The Bahama Islands. New York, MacMillan Co., xxxii -|- 630 pp.

Smith, Hobart M., and Chapman Grant

1958. The proper names for some Cuban snakes: an analysis of dates of publica-
tion of Ramon de la Sagra’s Historia Natural de Cuba, and Fitzinger’s
Systema Reptilium. Herpetologica, 14(4): 215-222.

Sutcliffe, Robert

1952. Results of the Catherwood-Chaplin West Indies Expedition, 1948. Part VI.
Amphibia and Reptilia. Notulae Naturae, 243: 1-8.

Thomas, Richard

1968. Notes on Antillean geckos (Sphaerodactylus). Herpetologica, 24(1):
46-60, 4 figs.

Thomas, Richard, and Albert Schwartz

1966. The Sphaerodactylus decoratus complex in the West Indies. Brigham
Young Univ. Sci. Bull., Biol. Ser., 7(4): 1-26, 20 figs.

Back issues of many Annals of Carnegie Museum articles
are available, and a few early complete volumes and parts
are listed at half price. Orders and inquiries should be
addressed to: Publications Secretary, Carnegie Museum,
4400 Forbes Avenue, Pittsburgh, Pa. 15213.

b 7 '7 3

PQ>% 2^

Volume 41

Annals of Carnegie Museum

Article 5

REPORTS ON THE MARGARET M. CARY— CARNEGIE MUSEUM
EXPEDITION TO BAJA CALIFORNIA, MEXICO, 1961.

7. The Family Hesperiidae ( Lepidoptera )

Lee D. Miller^

Research Associate, Section of Insects and Spiders,
Carnegie Museum, Pittsburgh, Pennsylvania

This is the seventh of a series of papers based on the Margaret M. Cary —
Carnegie Museum Expedition to Bafa California, Mexico, 1961, to appear in the
Annals of Carnegie Museum. For an account of the itinerary and description of
localities, see the first paper in this series [Richard M. Fox, Ann. Carnegie Mus., 36
(16): 181-192], Except as noted, all specimens including type series discussed in
the following study are in the collection of Carnegie Museum.

Introduction

The Hesperiidae of Baja California have been reviewed] by MacNeill
(1962). This paper adds locality records and confirms the occurrence
in the Cape region of some poorly known skippersL.,Jwo|^additipna4

species are added to the list of peninsular skippers, the rare Tlwryhes

Valeriana (Plotz) and Pyrgus o.^oileus (Linne). Two new peninsular
subspecies were described in a preceding paper (Miller and MacNeill,
1969). The list of hesperiids is by no means complete; there is still an

undescribed subspecies of Paratrytone melane (Edwards) in the high

mountains south of La Paz; Prenes californica A. G. Weeks is still an

unknown butterfly; possibly there are other species of Erynnis in the

mountains. And the spring skipper fauna has been poorly sampled.

The Expedition personnel collected 1558 hesperiids of 87 species. Of
these, 411 specimens of 81 species were taken on the mainland, and
1147 specimens of 26 species collected on the peninsula. Four of these
species are represented by separate mainland and peninsular subspecies.

In addition to the Expedition material, I have examined a small
number of skippers from the San Diego Natural History Museum, and
these are reported under the appropriate species below.

^Allyn Museum of Entomology, 712 Sarasota Bank Building, Sarasota, Fla. 33577.
Issued September 25, 1970

169

/

170 Annals of Carnegie Museum vol. 41

I wish to express my appreciation to the late Dr. Richard M. Fox
and to Harry K. Clench of the Section of Insects and Spiders, Carnegie
Museum, for their encouragement and suggestions. Dr. C. Don MacNeill
identified some specimens for which we had no comparative material,
and generally was most helpful. Thanks go to Dr. C. F. Harbison, San
Diego Natural History Museum, for the loan of specimens. Finally, the
greatest thanks are due Mrs. Margaret M. Cary of Philadelphia, Penn-
sylvania, for making this study possible by sponsoring the Expedition.

Zoogeography, Cape Region Hesperiidae

The occurrence outside the Cape region of the Hesperiidae thus far
recorded from there is summarized in Table 1. From these data it
becomes apparent that not all the skippers in the region were derived
from the same geographic source, nor were they simultaneous arrivals
at the tip of the peninsula. The 28 species show five general distribu-
tional patterns, which are summarized below.

1. Basically temperate species showing affinities to the California
mountains. There are two of these species known from the tip of the
peninsula at present, both in the high mountains of the Sierra de la
Laguna. When these mountains become better collected, particularly
in the spring, the number will probably increase. Roth species are
represented in the Cape region by endemic subspecies, Erynnis tristis
pattersoni Rums, and an unnamed subspecies of Paratrytone melane
(see MacNeill, 1962: 110-111). These two primarily temperate skippers
have also managed to penetrate the Mexican highlands, with melane
extending as far south as the Cerro de la Muerte in Costa Rica. The
differ entation shown by these butterflies in the Cape region indicates
long isolation from their northern relatives, and their temperate habitat
on the peninsula suggests that they arrived at a time when the peninsula
was moister and perhaps cooler than at present. Truxal (1960: 168)
discusses the post-Miocene extension of the Arcto- and Madro-Tertiary
geofloras into the peninsula, and it seems likely that the introduction
of tristis and melane into the Cape region was synchronous with this
event.

2. Sonoran desert species neither found in southern California nor
significantly penetrating the tropics. There are five of these species:
Thoryhes valeriana (Plotz), Systasea zampa (Edwards), Pyrgus
philetas Edwards, Nastra neamathla (Skinner and Williams) and
Lerodea dysaules Godman. The lack of differentiation of the Cape
region populations of these species indicates some interconnection with

1969

Miller: Mexican Hesperiidae

171

TABLE 1

Distribution of Hesperiidae Species of the Cape Region

Showing Endemicity

OF Cape

Fauna

Species

Distribution Outside Cape Region

BCN

sc

CM

AM ASD

MH

SM

SA

Chioides catillus

a

a

a

a

b

Urhanus proteus

a?

a

a

a

a

a

a

a

dorantes

?

b

b

b

b

b

Thorybes valeriana

a

a?

Cogia hippalus

b?

b

b

b

b

b

b

Staphylus ceos

a

a

a

Systasea zampa

a

a

Chiomara asychis

b

b

b

b

Erynnis funeralis

a

a

a

a

a

a

a

a

tristis

b

b

b

b

b

Pyrgus communis

a

a

a

a

a

a

b

b

oileus

a

a

a

a-b

philetas

a

a

Heliopet es domicella

a?

a

a

a

b

laviana

a?

a

a

a

b

Pholisora catullus

a

a

a

a

a

a

Nastra neamathla

a?

a

Copaeodes aurantiaca

a

a

a

a

Hylephila phyleus

a

a

a

a

a

a

a

a-b

Polites sahuleti

b

b

b

b

b

Atalopedes campestris

a

a

a

a

a

a

a

a

Paratrytone melane

b

b

Amblyscirtes tolteca

?

p

p

Lerodea eufala

a

a

a

a

?

P

?

a-b

dysaules

a?

a

a

Calpodes ethlius

a?

a

a

a

a

a

a

a

Panoquina panoquinoidesa.

a

a

b

b

a = subspecies same as, b=: subspecies different from, those found in the Cape
region. Abbreviations for places outside the Cape region are: BCN, Baja California
Norte; SC, southern California; CM, California mountains; AM, Arizona moun-
tains; ASD, Arizona-Sonora desert; MH, Mexican highlands; SM, southern Mexico;
SA, South America.

mainland populations, or isolation of relatively short duration. In either
event, these skippers are probably later, perhaps even post-Pleistocene,
arrivals. All are lowland butterflies.

3. Sonoran desert species that have penetrated California but not the
tropics. There are four of these species: Staphylus ceos (Edwards),
Pholisora catullus (Fabricius), Copaeodes aurantiaca (Hewitson), and
Polites sahuleti (Edwards). One of these is represented by an endemic
Cape region subspecies, Polites sahuleti margaretae Miller and Mac-
Neill, but the peninsular populations of the other three species are
indistinguishable from their mainland counterparts. P. sahuleti inhabits
the coastal grasslands on the peninsula, while the others are of more
general occurrence in the lowlands. All are probably fairly recent
arrivals on the peninsula. P. sahuleti is a well-known polytypic species.

ill

172 Annals of Carnegie Museum vol, 41 «

!!

whereas the others have no presently recognized geographic isolates.

4. Sonoran desert, and humid or subhumid, tropical species not found 1
in southern California. There are eight species in this group: Chioides
catillus (Cramer), Cogia hippalus (Edwards), Chiomara asychis li
(Stoll), Pyrgus oileus (Linne), Heliopetes domicella (Erichson), H. |j
laviana (Hewitson) and Amblyscirtes tolteca Scudder. This group is I
second only to Group 1 in degree of endemism. Three of the species
have endemic Cape region subspecies, Urbanus dorantes calafia (Wil- i
liams), Cogia hippalus peninsularis Miller and MacNeill, and Chiomara
asychis pelagica (Weeks), and the peninsular populations of catillus and
laviana may be distinguished from their mainland relatives by differ- ,
ences too slight to warrant naming them. Fives species, catillus, laviana, |
domicella, asychis and oileus, are also represented in South America !
by different subspecies, while tolteca has two subspecies in Mexico: the i,
nominate one in southern Mexico and prenda Evans in Sonora. Most, |
if not all, these species are earlier arrivals in the Cape region, probably
post-Miocene and certainly pre-Pleistocene.

5. Widely distributed tropical and subtropical species. There are

eight species in this category lUrbanus proteus (Linne), Erynnis fune-
ralis (Scudder and Burgess), Pyrgus communis (Grote), Hylephila
phyleus (Drury), Atalopedes campestris (Boisduval), Lerodea eufala '
(Edwards) Calpodes ethlius (Stoll), and Panoquina panoquinoides
(Skinner). None of these has an endemic Cape region subspecies, and
in only one species (panoquinoides) can the peninsular material be ^
distinguished from either Mexican or Californian specimens. P. pano- *
quinoides is unique in the group as a colonial butterfly intimately
associated with coastal or even salt-marsh grasses. Five of the other I
members of the group, proteus, phyleus, campestris, eufala, and i
ethlius, are strong-flying tropical skippers that have been taken as far I
north as the northern United States, but they probably are not resident :
there. These species arrived in the Cape region, probably in recent i
times, either over water or via land, through the Sonoran desert.
C. ethlius, though, is so intimately associated with cannas wherever
it is found that it is impossible to ascertain whether it actually arrived
in the Cape region under its own power or was imported by man with
his ornamental cannas. No specimens were taken on the peninsula
except in canna gardens. The other two species, E. funeralis and P.
communis, are widespread butterflies which certainly arrived recently,
probably by way of the Sonoran desert. 1

1969

Miller: Mexican Hesperiidae

173

In summary: all members of Group 1 are represented by endemic
subspecies derived from the California mountains, and they probably
date from the post-Miocene floral invasion of the peninsula; all members
of Group 2 are recent arrivals from the Sonoran desert; the members
of Group 3 arrived fairly recently from the Sonoran desert, although
sabuleti may be an earlier immigrant; the members of Group 4 are
early arrivals from the Sonoran desert, as evidenced by their distinct-
ness from mainland populations; and the Group 5 skippers are recent
arrivals of widely distributed, strong flying species, one of which,
ethlius, may have been introduced by man.

Systematic Account

In the following account, no attempt has been made to separate the
peninsular from the mainland skippers, except by separately listing the
records from the various Mexican states. I have listed the material that
was available from the San Diego Museum of Natural History separately
at the end of the discussion.

Subfamily Pyrrhopyginae
Mysoria affinis (Herrich-Schaffer), 1869.

This skipper is known from the western lowlands of Mexico and
occurs infrequently in Guatemala (Evans, 1951: 69). It was not taken
on the peninsula; indeed, no pyrrhopygine is known from there.

SINALOA: 16 mi. north of Mazatlan, 1 $ 29-x.

Subfamily Pyrginae

Phocides pigmalion belus Godman and Salvin, 1893.

Evans's description (1952: 13) of this Mexican subspecies mentions

. . costal spot over cell spot vestigial." While this description applies
to males, the females have the costal spot well developed. This skipper
is not common in collections.

SINALOA: Mazatlan, 2$ 1$ 24-x, 1$ 26-x.

Proteides mercurius mercurms (Fabricius), 1787.

The single specimen taken by the mainland party is in all respects
referable to the mainland subspecies of this skipper. The species is
general throughout the Neotropics.

SINALOA: Urias, 2 mi. south of Mazatlan, 1 $ 31-x.

Epargyreus exadeus complex.

Although this species was recorded from Baja California, MacNeill
(1962: 95) has discussed the problems inherent in such a record. The
three specimens collected by the mainland party show another of the

174

VOL. 41

Annals of Carnegie Museum

problems — exact species determination in the exadeus complex is diffi-
cult, if not altogether impossible at present. When I was at the British
Museum in 1964 I examined the series of Epargyretis arranged by Evans
and found that it was confused. The entire genus, and particularly the I
exadeus complex, is impossible to identify with certainty using Evans’s
keys ( 1952 ) . A careful and exhaustive revision of this genus is necessary
before any certain identifications may be made.

The males before me apparently represent two species: the one from
19 miles east of Concordia, Sinaloa, keys to sociis Hiibner, but the j
genitalia are like exadeus Cramer. The one from 2 miles east of Con- !
cordia keys to spina Evans. The female appears to be of spina.

SINALOA: 2 mi. east of Concordia,, 1^ 25-x; 17 mi. east of Concordia, 1$ 1-xi; ,

19 mi. east of Concordia, 1 $ 25-x. i

Fohjgonus leo leo (Cmelin), 1790. i

This species, which occurs generally throughout the Neotropics, was
represented by a single specimen taken by the mainland party. This
specimen is similar to a long series from Yucatan and definitely is i
referable to the nominate subspecies.

SINALOA: 17 mi. east of Concordia, 1^ 1-xi.

Chioides catilliis alhofasciata (Hewitson), 1867. 1

This species in its several subspecies is found throughout much of ;
tropical America. The subspecies alhofasciata is recorded from Arizona ,
through Central America and into Colombia (Evans, 1952: 56). This
species is not an inhabitant of the very wet tropical forests. It prefers
drier, more open situations than do other “long-tails.”

C. c. albofasckita is an avid flower visitor, showing definite preferences ;
for zinnias in cultivated gardens, the “desert willow” (Chilopsis linearis),
and the flowers of large Compositae in the arroyos. These butterflies
are very pugnacious and will engage others in mock combat at any time. j|

This species was taken both on the peninsula and in Sinaloa, but it is
surprisingly absent from collections made in Sonora. 1 have never seen
catillus as abundant as it was in Baja California Sur. The series from
the peninsula differs from the mainland representatives in the char-
acteristics described by MacNeil (1962: 95), but the differences are not !
of a magnitude to warrant description of the Baja California population
as an entity distinct from the mainland alhofasciata.

BAJA CALIFORNIA SUR: La Paz, Hotel Guaycura grounds, 1^ 21-x; Rancho
Rosarito, 1 S 2.3-x; Rancho Vinorama, 2 $ 23-x; Rancho El Novillo, 1 $ 28-x;

I

1969

Miller: Mexican Hesperiidae

175

Rancho Palmarito, 1^1$ 27-x, 1 $ 31-x, 1 $ 4-xi, 1 $ 30-xi; Arroyo Hondo, near
El Triunfo, 1 $ 24-x; Mesa Puerta Azul, 1 $ 1-xi; Arroyo San Bartolo, 4 $ 2-xi, 4 $
3-xi, 1^ [14?]-xi; Boca de la Sierra, 1,$ 24-xi; San Jose del Cabo, 1^ 22-xi,
3^2$ 23-xi, 2$ 7 $ 25-xi; Arroyo Candelaria, 2$ 1 $ 24-xi, 1 $ .

SINALOA: 48 mi. northwest of Culiacan, 1 1$ 23-x; Urias, 2 mi. south of

Mazatlan, 1 ^ 31-x.

Two additional specimens have been seen from the collection of the San Diego
Natural History Museum.

BAJA CALIFORNIA SUR: La Paz, 1$ 13-ix-1959; Colonia Calles, 1^ 15-ix-1959
(both C. F. Harbison).

Chioides zilpa zilpa ( Butler ) , 1874.

I can see no difference between the specimens the mainland party
took in Sonora and Sinaloa and others from Central America. The
specimens are not referable to the northern subspecies namba Evans.
C. 2:. zilpa is found from Mexico to northern South America.

SONORA: 12 mi. south of Hermosillo, 1^ 1$ 20-x.

SINALOA: 16 mi. north of Mazatlan, 1$ 28-x; Urias, 2 mi. south of Mazatlan,

1 $ 31-x; 8 mi. west of Concordia, 1 $ 2-xi; 19 mi. east of Concordia, 1 $ 25-x,
1$ 1-xi.

Aguna asander asander (Hewitson), 1867.

The nominate subspecies is found generally throughout the Neo-
tropics; there are two West Indian subspecies. This skipper is not found
on the peninsula.

SINALOA: 17 mi. east of Concordia, 2$ 1-xi.

Aguna metophis (Latreille),1824.

This rare species is recorded from Mexico, Nicaragua, Panama, Ven-
ezuela, and Brazil (Evans, 1952: 63). A single specimen was taken
by the mainland party.

SINALOA: Urias, 2 mi. south of Mazatlan, 1 $ 31-x.

Typhedanus undulatus (Hewitson), 1867.

This species is found generally throughout the hotter parts of the
Neotropics; it has not been taken on the peninsula.

SINALOA: 48 mi. northwest of Culiacan, 1^ 23-x; Mazatlan, 1^ 24-x; Urias,

2 mi. south of Mazatlan 1 $ 31-x; 19 mi. east of Concordia, 1 $ 25-x, 1 $ 1-xi.

Polythrix octomaculata octomaculata (Sepp), 1848.

176

Annals of Carnegie Museum

VOL. 41 '!

The nominate subspecies is found generally throughout the conti-
nental Neotropics. It has not been recorded from Baja California.

SINALOA: 16 mi. north of Mazatlan, 1^ 29-x.

Cadatractus arizonensis (Skinner), 1905.

The species is known from Arizona, Mexico and Central America.
Rindge (1948: 306) records a specimen of “Heteropia cydd' from Baja
California which MacNeill (1962: 96) feels may be referable to the
present species. No Codatractiis were taken on the peninsula by the
expedition.

SINALOA: 16 mi. north of Mazatlan, 2 9 28-x.

Urbanus proteus proteus ( Linne ) , 1758.

The nominate subspecies is known from the entire continental Neo-
tropics, and is common as far north as the southeastern United States.
U. proteus is a common species in the better-watered parts of the
peninsula.

These butterflies are avid flower visitors and show some pugnacity
toward others of their own kind, but not to the same degree as does
C. catilhis aJhofasciata.

BAJA CALIFORNIA SUR: La Paz, Hotel Giiaycura grounds, 2$ 1$ 21-x, 15 2$
22-x, 1 5 4-xi, 1 $ 8-xi; Rancho Rosarito, 1 $ 23-x; Rancho Vinorama, 1 $ 23-x;
Rancho Palmarito, 1 $ 30-xi, 2 $ 4-xii; Arroyo Hondo, near El Triunfo, 1 $ 1-xi;
Mesa Piierta Azul, 1 $ 12-xi; Arroyo San Bartolo, 15 12 3-xi; Bahia de Palmas,
15 3 2 20-xi; San Jose del Cabo, 1 2 23-xi, 15 2 2 25-xi; Puerto Chileno, 3 5
2 2 22-xi, 2 5 26-xi; Arroyo Candelaria, 2 2 24-xi; 4 mi, south of Arroyo Cande-
laria, 15 12 24-xi,

Urbanus dorantes dorantes ( Stoll ) , 1790.

In its nominate subspecies this species is one of the commonest hes-
periids throughout the American tropics, but the long series from the
Cape region of Baja California shows the peninsular material to
represent a distinct subspecies. There are also Antillean subspecies.
SONORA: Hermosillo, 3 5 2 2 20-x.

SINALOA: 16 mi. north of Mazatlan, 15 28-x, 4 5 29-x; Mazatlan, 3 5 24-x,
2 5 12 26-x, 15 31-x; Urias, 2 mi. south of Mazatlan, 3 5 12 31-x; 5 mi. west
of Concordia, 1 2 2-xi; 2 mi. east of Concordia, 15 3 2 25-x; 17 mi. east of
Concordia, 2 5 3 2 1-xi; 19 mi. east of Concordia, 15 12 25-x.

Urbanus dorantes calafia ( Williams ) 1926.

This subspecies was synonymized to nominate dorantes by Evans

1969

Miller: Mexican Hesperiidae

177

(1952:92), but he saw no peninsular specimens. The Baja California
material is extremely distinct from mainland examples, particularly
in the extensive olive-gray overscaling of the upper surface and the
grayer under surface lacking the purplish sheen of nominate dorantes.
A more detailed description of calafia is given by MacNeill (1962:97).

These butterflies are avid flower visitors and are found almost any-
where in the Cape region where there is green vegetation. During the
morning and late afternoon these butterflies are seen in the open, but
during the heat of the day they are active only in shady spots. On
two occasions (Mesa Puerta Azul, 1-xi, and Boca de la Sierra, 24-xi) we
encountered swarms of calafia during the hottest part of the day under
trees, feeding at flowers and flying about. No specimens were seen in
open sunlight except when they were flying from the shade of one
tree to the shade of another. I observed a similar phenomenon with
the mainland subspecies in Costa Rica during July and August, 1963.
Inasmuch as high temperature, greater amount of incident light, and
lower relative humidity all were conditions during the times that we
observed these habits, it is impossible to determine which factor gov-
erns the dorantes habit of seeking shade in midday. Perhaps all factors
were in operation. None of the other skippers showed this disposition to
escape the sun of the hot part of the day while remaining active. This
was the skipper most often attracted to light on the peninsula, and
it seems to be somewhat crepuscular, flying much later than most other
skippers.

BAJA CALIFORNIA SUR: La Paz, Hotel Guaycura grounds, 1 1$ 20-x, 2^

4$ 22-x, 2$ 6-xi, 7-xi, 1^ 8-xi; southeast shore of La Paz harbor,

2 5 2 $ 10-xi; Rancho Vinorama, 1 5 23-x; Racho El Novillo, 5 5 4 $ 28-x;

Rancho Palmarito, 3 5 1? 27-x, 15 30-x, 3$ 31-x, 15 19 4-x, 2 9 5-xi, 3 5
2 9 30-xi, 1 9 4-xii; Arroyo Hondo, near El Triunfo, 12 5 8 9 24-x; Mesa Puerta
Azul, 2 5 24-x, 12 5 6 9 1-xi, 9 5 19 12-xi; Arroyo San Bartolo, 4 5 19 2-xi,

4 5 3 9 3-xi, 15 19 12-xi; Bahia de Palmas, 15 12-xi, 15 16-xi, 3 5 19

20-xi; roadway 5 kilometers south of Rancho Buenavista, 10 5 4 9 25-x; Mira-
flores, 3 9 25-x; Boca de la Sierra, 4 5 8 9 13-xi, 2 5 2 9 24-xi; Arroyo San
Bernardo, 15 19 18-xi; Rancho San Bernardo, 500-600 m., 1 5 17-xi; Arroyo El
Rincon, 1 9 26-xi; San Jose del Cabo, 2 5 19 22-xi; 4 mi. SW of San Jose del
Cabo, 2 $ 24-xi; Puerto Chileno, 3 5 3 9 22-xi, 1 5 26-xi; Cabo San Luca, 1 9

23- xi; Arroyo Candelaria, 2 5 24-xi; 4 mi. south of Arroyo Candelaria, 2 5 4 9

24- xi.

Ten additional specimens have been examined from the collection
of the San Diego Museum of National History.

BAJA CALIFORNIA SUR: San Bartolo, 9 5 24-xi-1956; Todos Santos, 19 10-ix-
1958 (C. F. Harbison).

178

Annals of Carnegie Museum

VOL. 41

Urbanus teleus Hubner, 1821.

This skipper is found throughout the eontinental Neotropics but has
not been recorded from the peninsula.

SONORA: 16 mi. south of Guaymas, 1 $ 21-x.

SINALOA: 16 mi. north of Mazatlan, 1^ 29-x; Urias, 2 mi. south of Mazatlan,
2$ 31-x; 19 mi. east of Concordia, 1 $ 1-xi; 17 mi. east of Concordia, 1 $ l-xi.

Urbanus simplicius (Stoll), 1790.

This species has often been confused with the following one, but
Tilden (1965) has given the characters for separating them. My ovm
experience in Costa Rica indicated that simplicius is more character-
istic of wetter environments than is procne. The present species is
known from Mexico to Paraguay and Argentina.

SINALOA: Mazatlan, 1^ 24-x; 19 mi. east of Concordia, 1$ 25-x.

Urbanus procne (Plotz), 1881.

This is the species generally referred to in the literature as sim-
plicius from the United States (see Tilden, 1965). U. procne has the
wider range of the two, being commonly found from southern Texas
to Argentina, and it is more tolerant of dry conditions than is simplicius.
The present species has not been recorded from Baja California.

SONORA: 12 mi. south of Hermosillo, 1,$ 20-x; 16 mi. south of Guaymas, 1^
21-x; 16 mi. south of Navojoa, 1 $ 22-x.

SINALOA: 46 mi. north of Los Mochis, 2$ 22-x; 10 mi. southeast of Los Mochis,
5 5 23-x; 11 mi. north of Mazatlan, 15 28-x; 16 mi. north of Mazatlan, 15 24-x;
Mazatlan, 1 $ 24-x; Urias, 2 mi. south of Mazatlan, 1 5 31-x; 2 mi. east of Con-
cordia, 1 5 25-x.

Urbanus chales (Godman and Salvin), 1893.

This species is thus far known only from western Mexico (Evans,
1952: 98). It is not common in collections.

SINALOA: 19 mi. east of Concordia, 15 25-x, 1-xi.

Astraptes fulgerator azul (Reakirt), 1866.

A. /. azul is the subspecies of this attractive skipper found in Mexico,
Central America, and western South America. It has been recorded
from southern Texas. This species is another that has not been taken
in Baja California.

SINALOA: Mazatlan, 15 24-x; 19 mi. east of Concordia, 2 5 26-x.

1969

Miller: Mexican Hesperiidae

179

Achalarus albociliatus albociliatus (Mabille), 1877.

The nominate subspecies occurs from Mexico to Costa Rica, but it
has not been recorded from the peninsula.

SINALOA: 16 mi. north of Mazatlan, 1^ 29-x; Mazatlan, 19 24-x; Urias, 2 mi.
south of Mazatlan, 1 ^ 1 $ 31-x; 17 mi. east of Concordia, 1 $ 1-xi.

Achalarus toxeus (Plotz), 1882.

This '‘species” is in a confused state. Evans (1952: 128, pi. 21) de-
scribes and presents five different genitalic configurations (poorly
drawn ) of the valva with the comment, “The cuiller of the clasp varies
considerably, as specified in the list of the B. M. material. But the dif-
ferences seem to be individual rather than sub-specific.” In my opinion
these differences are possibly specific, but the material available is not
sufficient for definitive determinations. Significant results might be
obtained by rearing through a few broods of these butterflies to see
whether or not various configurations of the valva might occur in sib-
lings. The material at hand from Sonora and Sinaloa either agrees with
the figure of the genitalia given in Lindsey, Bell and Williams (1931:
pi. 5, fig. 7, as coyote) or the valva is more pointed than in that figure.
From Evans’s figures it is difficult to determine which of his groups
are represented in our material, but probably groups D and E are.
All the males have well-defined costal folds and apical hyaline spots;
the female lacks the apical spots.

SONORA: Hermosillo, 1 5 20-x.

SINALOA: Mazatlan, 2^ 24-x; Urias, 2 mi. south of Mazatlan, 1$ 31-x.
Thorybes valeriana (Plotz), 1882.

Figs. 1, 2 ( d ), 3 ( d genitalia).

Two ragged males of this species (determined by C. Don MacNeill)
were taken on the peninsula and represent the first records from Baja
California. These skippers are too worn to be compared profitably with
mainland material and must be referred simply to valeriana at present.
Evans (1952: 132) synonymizes the Dyar names mysie and uvydixa
to valeriana, but MacNeill {in lift.) expresses some doubt as to the
validity of Evans’s conclusion. Much more material is necessary before
any decisions can be made on these names.

BAJA CALFORNiA SUR: 4 mi. south of Arroyo Candelaria, 2$ 26-x.

180

Annals of Carnegie Museum

VOL. 41

Figs, 1-3. All Thonjbes valeriana (Plotz), $, Arroyo Candelaria, Baja California
Sur, Mexico, 26-X-1961. Fig. 1, upper side : Fig. 2, under side; Fig. 3, $ genitalia.

Cabares potrillo potrillo (Lucas), 1857.

The nominate subspecies is found from southern Texas to Costa Rica,
and in the West Indies. It is unrecorded from the peninsula.

SINALOA: 48 mi. northwest of Culiacan, 2$ 1 $ 23-x; 16 mi. north of Mazatlan,
1$ 29-x.

Celaenorrhinus fritzgaertneri fritzgaertneri Bailey, 1880.

This species occurs on the continent from southern Texas to Costa
Rica (Evans, 1952: 165).

SINALOA: 17 mi. east of Concordia, 1$ 1-xi.

Cogia hippalus hippalus (Edwards), 1882.

Two specimens of this subspecies were taken by the mainland party.
They are distinct from the subspecies penimularis represented in the
Cape region and described previously (Miller and MacNeill, 1969:
20-21). MacNeill (1962) mentions the possibility that the nominate
subspecies may occur on the peninsula in the San Pedro Martir massif,
Baja California Norte, but it has not been recorded to date.

SINALOA: 16 mi. north of Mazatlan, 15 28-x; Urias, 2 mi. south of Mazatlan,
15 31-x

1969

Miller: Mexican Hesperiidae

181

Cogia hippalus penimularis Miller and MacNeill, 1969.

This subspecies is restricted to the Cape region, as far as known.
The records and characteristics are given in the original description
(Miller and MacNeill, 1969: 20-21).

Cogia calchas (Herrich-Schaffer), 1869.

This species is one of the commonest in the Neotropics, but it is un-
recorded from the peninsula. I have found it more commonly in dis-
turbed areas, rather than in forest.

SINALOA: 17 mi. east of Concordia, 1 $ 1-xi.

Arteurotia tractipennis tractipennis Butler and Druce, 1872.

This species is widespread on the Mexican and Central American
mainland.

SINALOA: 16 mi. north of Mazatlan, 3 ($ 1$ 29-x.

Nisoniades rubescens (Moschler), 1876.

This species is widely distributed throughout the moister parts of
the Neotropics, but it has not been reported from the peninsula.

SINALOA: Urias, 2 mi. south of Mazatlan, 1$ 31-x; 19 mi. east of Concordia,
1$ 1-xi.

Fellicia costimacula arina Evans, 1953.

The subspecies arina is the Central American representative of this
widely spread Neotropical skipper. All the specimens taken on the
Expedition were captured on the mainland.

SINALOA: Urias, 2 mi. south of Mazatlan, 1^ 31-x; 5 mi. west of Concordia,

1$ 2-xi.

Pellicia dimidiata dimidiata Herrich-Schaffer, 1870.

This is another widespread tropical species that has not been re-
corded from Baja California.

SINALOA: Urias, 2 mi. south of Mazatlan, 1$ 31-x; 5 mi. west of Concordia,
1$ 2-xi.

Noctuana noctua bipuncta (Plotz), 1884.

This, too, is a mainland species; the subspecies bipuncta is known
from Mexico to Nicaragua.

SINALOA: 19 mi. east of Concordia, 1^ 1-xi.

182

Annals of Carnegie Museum

VOL. 41

Noctuana stator (Godman and Salvin), 1899.

This is another mainland species that does not reach the peninsula.
SINALOA: 19 mi. east of Concordia, 2$ 1$ 25-x.

Bolla phylo pullata (Mabille), 1878.

This Central American butterfly is represented in the Expedition
collection by a single female which is provisionally placed here.
SINALOA: 19 mi. east of Concordia, 1$ 25-x.

Bolla clytius (Godman and Salvin), 1897.

This uncommon species is thus far known only from western conti-
nental Mexico and Honduras (Evans, 1953: 83).

SONORA: 12 mi. south of Hermosillo, 2$ 20-x; 36 mi. north of Guaymas, 2$
20-x.

SINALOA: 46 mi. north of Los Mochis, 1 $ 22-x; 48 mi. northwest of Culiacan,
1^1$ 23-x.

Staphyliis ceos (Edwards), 1882.

This species is basically a member of the Sonoran desert fauna that
has also penetrated Baja California. I can detect no difference between
the peninsular and mainland populations.

On the peninsula ceos was rather widely distributed, though never
abundant. It generally sought the shadiest places available and seldom
ventured far from them, at least during the hours of hot sunlight — I
never saw it in the early morning or late afternoon. This species may
be taken occasionally at water.

BAJA CALIFORNIA SUR: Raucho El Novillo, 2$ 28-x; Rancho Palmarito, 1^
27-x, 1^ 4-xii; Arroyo San Bartolo, 1 ($ 2$ 2-xi; Bahia de Palmas, 1^ 16-xi;
roadway 5 kilometers south of Rancho Buenavista, 1 9 25-x; Boca de la Sierra,
1 9 13-xi, 2 $ 24-xi; Arroyo Candelaria, 1 $ 26-x; 4 mi. south of Arroyo Candel-
aria, 1 ^ 19 24-xi.

SONORA: 16 mi. south of Guaymas, 1 $ 21-x.

Staphtjliis mazans mazans (Reakirt), 1866.

These specimens seem to be referable to typical mazans (type-
locality: near Voracruz, Mexico), not to ascaphaliis ( Staudinger ) . The
various subspecies, if subspecies they are, of mazans seem to replace
ceos in the more humid parts of Mexico.

SINALOA: 48 miles northwest of Culiacan, 2 9 23-x; 16 mi. north of Mazatlan,
3 $ 29-x; 5 mi. west of Concordia, 1 9 2-xi; 19 mi. east of Concordia, 2$ 19
25-x.

1969

Miller: Mexican Hesperiidae

183

Mylon pelopidas (Fabricius), 1793.

This is a moderately common skipper throughout the more humid
parts of the Neotropics. It is not found on the peninsula.

SINALOA: 5 mi. west of Concordia, 1 $ 2-xi.

Mylon lassia (Hewitson), 1868.

This species is found on the mainland from Mexico to northern South
America.

SINALOA: 19 mi. east of Concordia, 1$ 1-xi.

Antigonus erosus ( Hiibner ) , 1812.

This is a widespread continental Neotropical species.

SINALOA: 48 mi. northwest of Culiacan, 2$ 23-x.

Systasea zampa (Edwards), 1876.

Hesperia zampa Edwards, 1876. Trans. Amer. Ent. Soc., 5: 207 (type-locality:
South Apache, Arizona).

Antigonus evansi Bell, 1941. Ent. News, 52: 165; fig. 2 (type-locality: Babo-
quivari Mts., Arizona). New synonomy.

Figs. 4, 5 (Neotype d ), 6 (Neotype d genitalia).

Edwards (1876: 207) described Hesperia zampa from a single ex-
ample collected at “South Apache, Arizona” and retained it in his own
collection. Later he sent this specimen and several others to Speyer
for study (Brown, 1964). This shipment of butterflies was lost when
the S. S. Pomerania went down in the English Channel, so it is appar-
ent that the type of zampa was lost.

Holland (1931: pi. 46, fig. 1) wrote the next chapter in the history
of this name by figuring a specimen of S. pulverulenta (Eelder) as the
“type” of zampa. As noted above, that specimen cannot be the authen-
tic holotype of zampa, and was not even from the Edwards collection.
It bears only the printed labels, “Arizona” and “Holland/Collection.”
The Holland designation of the identity of zampa, however, was ac-
cepted by many subsequent authors, notably Bell (1941: 164) and
Evans (1953: 160). Both placed zampa in the synonymy of pulveru-
lenta, justifiably if the specimen figured by Holland were the holotype
of zampa. Bell (1941: 165) then described the second species of the
genus occurring in North America as Antigonus evansi.

Inasmuch as Bell and Evans have successfully separated these two
species, I am reluctant to upset current nomenclature. I have discussed

184

Annals of Carnegie Museum

VOL. 41

the status of the present species with several colleagues who have had
extensive field experience in Arizona, and while all of us have seen
hundreds of examples of this skipper, none of us has seen an unques-
tionably authentic specimen of pulverulenta from that state. The few
specimens that bear data other than the vague “Arizona” of earlier
collectors are specimens taken by O. C. Poling that are labeled as from
the Baboquivari Mountains. Poling, a professional collector, also col-
lected extensively in the Davis Mountains, Texas (where pulverulenta
occurs), and he was not overly careful with his data. So I suspect that
at least some of the specimens he labeled as from Arizona were act-
ually collected in Texas. Since I can find no authentic pulverulenta
from Arizona, and since there are two specimens of “evansf from
Arizona in the Edwards collection, I reluctantly designate one of the
latter as the neotype of Hesperia zampa Edwards. This specimen bears
labels as follows: first, a handwritten label, “Zampa cT/Edw. Ariz^.”;
second, a printed label, “Edwards/Collection”; and third, a partly
printed, partly handwritten label, “Neotype/Hesperia/zampa cT/W.
H. EDWARDS/ designated by/Lee D. Miller, 1965.” The designation
of this specimen has been done with the knowledge and encourage-
ment of F. Martin Brown, who is engaged in a complete study of the
Edwards types.

The present species is well differentiated by Bell (1941: 164-166)
and Evans (1953: 160), as evonsi, and these works should be consulted
for a complete listing of the distinguishing characteristics.

H. zampa is a desert or subdesert species, whereas pulverulenta
occurs in less arid situations. We found the present species around
watering places where it flew close to the ground, rarely visited flowers
and lighted with the wings half open. These skippers were never
particularly abundant.

The Expedition records are as follows:

BAJA calfornia sur: Rancho Palmarito, 2$ 27-x, 31-x, 3 5 1$ 4-xi,

3 5 1 $ 5-xi, 8 5 2 2 30-xi.

SONORA: 12 mi. south of Hermosillo, 3 5 20-x; 10 mi. northwest of Caborca,
15 12 19-x.

Systasea pulverulenta (Felder), 1869.

Fig. 7 {d genitalia).

Evans (1953: 160) lists specimens from Arizona, Texas, Mexico, and
Guatemala but, as indicated in the discussion of zampa, the records of

1969

Miller: Mexican Hesperiidae

185

Figs. 4-7. Systasea zampa (Edwards), neotype “Arizona,” upper side. Fig. 5,
same, under side. Fig. 6, same, $ genitalia. Fig. 7, S. pulverulenta (Felder), S
genitalia.

the present species from Arizona are apparently based on erroneous
data.

All specimens taken by Expedition personnel were collected on the
mainland.

SONORA: 16 mi, south of Navojoa, 1^ 22-x.

SINALOA: 16 mi. north of Mazatlan, 1$ 29-x; Urias, 2 mi. south of Mazatlan,
1 $ 31-x; 4 mi. southeast of Mazatlan, 1 $ 25-x.

Zopyrion sandace Godman and Salvin, 1896.

This species is found in the moderately humid parts of Mexico and
Guatemala.

SINALOA: 48 mi. mrthwest of Culiacan, 1$ 23-x; 16 mi. north of Mazatlan,

1 3 29-x; Urias, 2 mi. south of Mazatlan, 9 ^ 3 $ 29-x; 17 mi. east of Concordia,

2 3 1-xi.

186

Annals of Carnegie Museum

VOL. 41

Achylodes thraso tamemind (Edwards), 1870.

I am retaining the Edwards name tamenund for the Texas and Mexi-
can material, although Evans (1953: 177) synonymized it to the nomi-
nate race. It is an exclusively mainland subspecies, and there are three
West Indian ones.

SINALOA: 16 mi. north of Mazatlan, 2,^ 2$ 29-x.

Achylodes selva Evans, 1953.

This species seems to be restricted to the higher, subtropical moist
forests of the Neotropics.

SINALOA: 19 mi. east of Concordia, 1^ 25-x.

Timochares ruptifasciata ruptifasciata (Plotz), 1884.

This is another mainland Mexican species.

siNAOLA: Mazatlan, 1^ 28-x.

Anastrus sempiternus sempiterniis (Butler and Druce), 1872.

This species is another from the mainland.

SINALOA: Mazatlan, 1,$ 26-x, 15 27-x, 2 5 29-x.

Anastrus toUmus tolimus (Plotz), 1884.

This, too, is a species found on the mainland only.

SINALOA: Mazatlan, 1$ 27-x, 15 3-xi.

Chiomara asychis georgina (Reakirt), 1868.

This subspecies is the one represented in the southernmost United
States, Mexico, and Central America. Evans (1953: 198) considers the
Baja California material to be the same subspecies, as does MacNeill
(1962: 100), but the peninsular specimens are consistently so different
from mainland ones that the Weeks name pelagica (see below) is resur-
rected for them.

SONORA: Hermosillo, 15 19 20-x; 12 mi. south of Hermosillo, 3 5 20-x; 36
mi. north of Guaymas, 1 5 20-x.

SINALOA: 46 mi. north of Los Mochis, 2 5 22-x; 19 mi. north of Los Mochis,
1 5 22-x; 17 mi. east of Concordia, 1 5 1-xi; 22 mi. east of Concordia, 2 5 25-x.

Chiomara asychis pelagica (Weeks), 1891.

The peninsular asychis are very different from the mainland repre-
sentatives of this species, hence I am restoring the Weeks name
pelagica to them. This subspecies most closely resembles Argentinian

1969

Miller: Mexican Hesperiidae

187

and Bolivian specimens of C. a. autander (Mabille) and bears little
resemblance to the Mexican and Central American georgina. The differ-
ences between georgina and pelagica are summarized in Table 2.

TABLE 2

Chiomara asychis: Differentiating Characters of the Subspecies
georgina (Mexican Mainland) and pelagica (Baja California)

georgina

1. $9 paler above.

2. Discal spot in forewing space
Cu2-2A always white.

3. Hindwing white discal band
broad and well developed.

4. Sexual dimorphism slight.

5. $ hind wing below with little
basal and discal brown mottling.

pelagica

1. $9 darker above.

2. Discal spot in forewing space Cu2-2A
often gray, never white in 9 .

3. Hindwing white discal band narrow,
often broken and obsolete in 9 .

4. Sexual dimorphism marked.

5. 9 hind wing below with prominent
basal and discal brown mottling.

These butterflies were common almost everywhere on the peninsula
where collecting was done. They do not visit flowers as avidly as do
some of the other skippers and are more likely to be seen sunning
themsleves on the tips of twigs with their wings outstretched. They are
not particularly pugnacious, although they may on occasion attack
almost anything that moves in their vicinity. The silhouette of these
butterflies at rest is characteristic: the wings are held outstretched and
the tips of the fore wings droop, giving the butterflies a convex
appearance.

BAJA CALIFORNIA SUR: La Paz, Hotel Guaycura grounds, 21-x, 1$ 6-xi;

28-x; Rancho Palmarito, 3^1$ 27-x, 1$ 30-x, 4^2$ 31-x, 4$ 4-xi, 1$ 5-xi,
9 ^ 6 $ 30-xi, 5 $ 4-xii; Arroyo Hondo, near El Triunfo, 3 $ 24-x; Mesa Puerta
Azul, 1 ^ 1 $ 12-xi; Arroyo San Bartolo, 2$ 1 $ 2-xi, 2 $ 3-xi, 2 $ 12-xi,
1 $ 14-xi; roadway 5 kilometers south of Rancho Buenavista, 3 9 25-x; Arroyo San
Bernardo, 1 $ 18-xi; Arroyo El Rincon, 2 9 26-xi; San Jose del Cabo, 5$ 5 9
22-xi, Puerto Chileno, 2$ 2 9 22-xi, 1 $ 26-xi; Arroyo Candelaria, 4 $ 24-xi;
4 mi. south of Arroyo Candelaria, 3^1$ 26-x, 2^3? 24-xi.

Two additional specimens have been examined in the collection of the
San Diego Museum of Natural History.

BAJA CALFORNiA SUR: San Bartolo, 1^ 24-xi-1959; Las Parras, 14-xi-1961
(C. F. Harbison).

Gesta gesta invisus (Butler and Druce), 1872.

The present subspecies is represented on the mainland from southern

Texas to Costa Bica.

SINALOA: 16 mi. north of Mazatlan, 1$ 29-x; Mazatlan, 24-x, 1^ 2$ 28-x;
19 mi. east of Concordia, 4$ 1 $ 25-x,

188

Annals of Carnegie Museum

VOL, 41

Erynnis funeralis (Scudder and Burgess), 1870.

I am following Burns ( 1964 ) in attributing specific rank to funeralis,
but most authors have considered it a subspecies of zarucco ( Lucas ) .

This species is one of the most distinctive elements of the Sonoran
desert fauna, and it is common in Baja California. In sharp contrast to
most of the skippers, funeralis is to be found even in the very dry desert
and in the brightest, hottest sunshine. It is very pugnacious and will
engage in mock combat with almost anything that moves. It visits flowers
frequently, showing a particular preference for the flowers of large
yellow composites.

BAJA CALIFORNIA suR: La Paz, Hotel Guayciira grounds, 2$ 2$ 6-xi, 1^ 8-xi;
Rancho Rosarito, 5 $ 3 $ 23-x; southwest shore of La Paz harbor, 2 $ 8-xi, 2 $

1 $ 10-xi, 1 $ 5-xii, 1 $ 6-xii; Rancho Vinorama, 1 $ 23-x; Rancho El Novillo,

2$ 28-x; Rancho Pahnarito, 1^ 27-x, 1^ 31-x, 2$ 4-xi, 5$ 30-xi; Mesa
Puerta Azul, 2 $ 1-xi; Arroyo San Bartolo, 2 $ 2-xi, S $ 1 $ 3-xi, 1 $ 12-xi; Bahia
de Palmas, 1 9 12-xi; road 5 kilometers south of Rancho Buenavista, 1 19 25-x;

Miraflores, 1 9 25-x; La Cienaga I, 1270 m., 1 ^ 19 16-xi; Puerto Chileno, 1 $
22-xi; Cabo San Lucas, 1 $ 23-xi; Arroyo Candelaria, 1 $ 24-xi, 1 9 26-x; 4 mi.
south of Arroyo Candelaria, 1 9 24-xi.

BAJA CALIFORNIA NORTE: 21 lui. south of Mcxicali, 1 $ 17-x.

SONORA: 10 mi. northwest of Caborca, 2S 19-x; 12 mi south of Hermosillo, 2$

2 9 20-x; 26 mi. north of Gu4ymas, 1 9 20-x.

SINALOA: 48 mi. northwest of Culiacan, 1 $ 23-x.

Ten additional specimens have been examined from the collection of the San
Diego Museum of Natural History.

BAJA CALIFORNIA SUR: Las Parras, 1 9 19-xi-1952; Colonia Calles, 1 9 15-ix-1959.
BAJA CALIFORNIA NORTE: Valle de la Trinidad, 2$ 16-iii-1936; Santo Tomas,
2$ 19 20-iii-1935, 1$ 16-ix-1935; Santa Agueda, 1^ 16-ix-1935 (all C. F,
Harbison ) .

Erynnis tristis pattersoni Burns, 1964.

This subspecies was originally described from seven males collected
in the Sierra de la Laguna (sometimes called Sierra de la Victoria).
The genitalia of the two males taken by the Expedition look more like
those of E. tristis tatius (Edwards) than those of the nominate sub-
species. These two specimens were taken at the top of a windswept
grassy hill at an elevation of 1270 meters. Oaks were plentiful nearby;
these are postulated as the food plant by Burns (1964: 143). Since the
type series was taken in May and the Expedition material was collected
in November, this species is at least double-brooded in the Cape region.
BAJA CALIFORNIA SUR: La Cicucga, I, Sierra Laguna, 1270 m., 2$ 15-xi,

1969

Miller: Mexican Hesperiidae

189

Pyrgus communis albescens Plotz, 1884.

This is another of the Sonoran desert species that has become success-
fully established throughout the peninsula. Its habits are quite similar
to those shown by other communis populations throughout the United
States and Mexico.

BAJA CALIFORNIA SUR: Southeast shore of La Paz harbor, 1$ 10-xi; Rancho

Palmarito,, S$ 29 27-x, S$ 49 31-x, 7^5 9 4-xi, 4^2$ 5-xi, 5^2$ 20-xi;
Mesa Puerta Azul, 1 $ 1-xi; Arroyo San Bartolo, 1 $ 12-xi; Bahia de Palmas,

1 $ 20-xi; Rancho San Bernardo, 1 $ 17-xi; Caduano, 1 $ 25-xi; Arroyo Candelaria,

2 $ 24-xi; 4 mi, south of Arroyo Candelaria, 1 $ 26-x.

Three additional specimens have been examined from the collection of the San
Diego Museum of Natural History.

BAJA CALiFORiA SUR: Coiuondu, 2$ 16-xii-1956.

BAJA CALIFORNIA NORTE: Santa Maria, 1(5 23-viii-1953 (C. F. Harbison).

Pyrgus oileus oileus (Linne), 1767.

This subspecies is widespread throughout Mexico, Central America,
and the Antilles, but it has not been previously recorded from Baja
California. A single male in the Expedition material from the Cape
region represents the first record from the peninsula of this species.

BAJA CALIFORNIA SUR: Rancho Palmarito, 15 3()-xi.

SONORA: 16 mi. south of Navojoa, 1 9 22-x.

SINALOA: 48 mi, northwest of Culiacan, 2$ 4 9 23-x; 16 mi. north of Mazatlan,
1 $ 28-x, 1 $ 29-x; Mazatlan, 15 19 24-x; Urias, 2 mi. south of Mazatlan,
15 29 31-x; 19 mi. east of Concordia, 55 5 9 25-x; 17 mi. east of Concordia, 19
29-x, 3 9 1-xi; 2 mi. east of Concordia, 2 5 25-x.

Pyrgus philetas Edwards, 1881.

I agree entirely with AlacNeill (1962: 101), who has reinstated
philetas as a distinct species. Evans (1953: 222) had placed it as a sub-
species of oileus. The present species is characteristic of the Sonoran
and Chihuahuan desert and is found in the Cape region of Baja
California.

This skipper was found wherever there was water, though it was sel-
dom abundant. It visits flowers and is particularly fond of those of the
“desert willow.” Many other specimens were collected at mud puddles.
P. philetas seems to be more widespread throughout the Cape region
than albescens, but is not as common as that species in areas where
both are present.

BAJA CALIFORNIA SUR: Raucho Palmarito, 2 5 27-x; 19 30-x, 15 2 9 31-x, 3 9
4-xi, 2 5 19 5-xi; Rancho Vinorama, 1 5 23-x; Arroyo Hondo, near El Triunfo, 1 5

190

Annals of Carnegie Museum

VOL. 41

24- x; Mesa Puerta Azul, 1 $ 1-xi; Arroyo San Bartolo, 1 $ 2-xi, 1 $ 3-xi; Bahia
de Palmas, 1 $ 20-xi; roadway 5 kilometers south of Rancho Buenavista, 3 $

25- x; Boca de la Sierra, 2 $ 13-xi, 1 $ 24-xi; Arroyo San Bernardo, 2 $ 18-xi;
Rancho San Bernardo, 500-600 m., 1 $ 17-xi; La Cienaga I, Sierra Laguna, 1270
m., 1 19 16-xi; Arroyo El Rincon, 1 ^ 19 26-xi; San Jose del Cabo, 1 5 2 9
25-xi; Arroyo Candelaria, 2 9 26-x; 1 $ 24-xi; 4 mi. south of Arroyo Candelaria,

1 $ 26-x, 2 $ 24-xi.

SONORA: 12 mi. south of Hermosillo, 2 3 2 9 20-x.

SINALOA: Urias, 2 mi. south of Mazatlan, 2 3 31-x.

Heliopetes clomicella domicella (Erichson), 1848.

This subspecies is found infrequently from Texas and Arizona south
to northern South America. A few specimens have been recorded previ-
ously from the peninsula.

This butterfly was never common and was taken only in several
arroyos, although at Rancho Palmarito I saw a specimen or two that
eluded me. Generally they were seen flying about in moderate shade
or visiting low flowers. Unlike many of their relatives, these butterflies
are very inconspicuous in the field. They are not combative, in contrast
to most Mexican Heliopetes.

The single female from south of Hermosillo is a striking aberration,
differing from the ordinary female in the following particulars: the
apical-submarginal row of white spots on the forewing, both above and
below, is fused into a band that is connected to the discal white band,
leaving a subtriangular black patch at the end of the forewing cell.
The hind wing pattern is more or less normal, perhaps showing the
white discal band a bit more developed.

BAJA CALIFORNIA SUR: Arioyo San Bartolo, 1 3 3-xi, 1 3 12-xi; roadway 5 kilom-
eters south of Rancho Buenavista, 1 3 25-v; Arroyo San Bernardo, 1 3 18-xi.

SONORA: La Zorra, 10 mi. northwest of Caborca, 1 ($ 1 9 19-x; 12 mi. south of
Hermosillo, 3 ^ 19 20-x.

SINALOA: 16 mi. north of Mazatlan, 19 28-x; 5 mi. west of Concordia, 19 2-xi.

Heliopetes macaira macaira Reskirt, 1866.

This species has not been found on the peninsula. The nominate sub-
species is moderately common from Texas to Panama.

SINALOA: 48 mi. northwest of Culiacan, 2 3 23-x; Mazatlan, 2 9 24-x; Urias,

2 mi. south of Mazatlan, 2 3 31-x,

Heliopetes laviana loviana (Hewitson), 1868.

This widespread tropical species is common in the Cape region of
Baja California.

1969

Miller: Mexican Hesperiidae

191

Almost wherever there is water, natural or otherwise, laviana is
abundant. These butterflies are avid flower visitors, preferring the
flowers of low composites or leguminous plants. Otherwise the males
will establish territories and attempt mock combat with other members
of their species, but they are not as pugnacious as some of the “long-
tails.”

BAJA CALIFORNIA SUR: Rancho Rosarito, 2$ 1$ 23-x; Rancho El Novillo, 1$
28-x; Rancho El Salto, 1$ 28-x; Rancho Palmarito, 2$ 27-x, 3^ 2$ 31-x, 2$
4-xi; 21 ^ 4 $ 30-xi; Arroyo Hondo, near El Triimfo, 2 $ 24-x; Arroyo San Bartolo,
1 $ 12-xi; roadway 5 kilometers south of Sancho Buenavista, 1 $ 25-x; Miraflores,
1 $ 25-x; Boca de la Sierra, 1 $ 28-xi; Arroyo El Rincon, 1 $ 26-xi; Puerto Chi-
leno, 1 $ 22-xi; Arroyo Candelaria, 1 $ 24-xi; 4 mi. south of Arroyo Candelaria,
2$ 24-xi.

SINALOA: 46 mi. north of Los Mochis, 9^ 22-x; 19 mi. north of Los Mochis,
1 $ 23-x; 48 mi. northwest of Culiacan, 1 $ 23-x; 16 mi. north of Mazatlan, 1 $
24tx; 17 mi. east of Concordia, 1 $ 1-xi; 19 mi. east of Concordia, 1 9 25-x; 19
mi. east of Concordia, 1 9 25-x; 17 mi. east of Concordia, 1 $ 1-xi.

Pholisora catullus (Fabricius), 1793.

This species is primarily North American, but is also known from
northern Sonora and from scattered localities along the Baja California
peninsula where it is not common.

BAJA CALIFORNIA SUR: La Paz, Hotel Guaycura grounds, 2$ 21-x, 19 22-x;
Rancho Palmarito, 1 $ 27-x; roadway 5 kilometers south of Rancho Buenavista,
1 9 25-x; Arroyo El Rincon, 1 $ 26-xi; 4 mi. south of Arroyo Candelaria, 1 $ 26-x.

Three additional specimens have been examined from the collection of the San
Diego Museum of Natural History.

BAJA CALIFORNIA NORTE: Vallc dc la Trinidad, 1^ 16-iii-1936; Santo Tomes,
2$ 2()-iii-1935 (C. F. Harbison).

Callimormus satiirniis ( Herrich-Schaffer ) , 1869.

This skipper is common throughout the humid portions of the Neo-
tropics.

SINALOA: 16 mi north of Mazatlan, 1 9 29-x; 11 mi. north of Mazatlan, 1 S 27-x,
1 $ 28-x; 19 mi. east of Concordia, 1 19 25-x, 1 9 1-xi.

Subfamily Hespkjiinae

Monca telatci tyrtaeus (Plotz), 1883.

This subspecies is moderately common in Mexico and Central Amer-
ica. It is not known from the peninsula.

SINALOA: Mazatlan, 2$ 24-x.

192

Annals of Carnegie Museum

VOL. 41

Nastra neamathla (Skinner and Williams), 1923.

This species was first recorded from the peninsula by MacNeill
(1962: 103-104), who reported specimens from Canyon San Pedro in
the Sierra de la Laguna of the Cape region. The specimens listed below
indicate that neamathla may actually be widely distributed throughout
the Cape region, but it is apparently never common. With the abun-
dance of Lerodea e. eufala (Edwards), however, it may have been
overlooked by most collectors, since the two have rather similar habits.

BAJA CALIFORNIA suR: Rancho Palmarito, 1^ 4-xii; Arroyo San Bartolo, 15 1$

2-xi; San Jose del Cabo, 1 $ 28-xi.

Cymaenes odilia trehhis (Mabille), 1891.

Two specimens of this species were taken by the mainland party*
SINALOA: 5 ini. west of Concordia, 15 2-xi; 19 mi. east of Concordia, 1$ 1-xi.

Vehilius inca (Scudder), 1872. ;

This species is common in the continental Neotropics.

SINALOA: 17 mi east of Concordia, 15 1-xi; 19 mi. east of Concordia, 1$ 25-x,

15 1$ 1-xi.

Lerema Iiimina ( Herrich-Schaffer ) , 1869. j

Evans (1955: 163) records this species only from Guatemala, Costa
Rica, Panama, and Colombia, but I have seen specimens from Mexico.
The specimen listed below is the first I am aware of from so far north |
on the west coast.

SINALOA: Mazatlan, 1 5 28-x.

Lerema aecius accius (Abbot and Smith), 1797.

This subspecies is common in the southern United States and Mexico
and occurs more infrequently in Central America. It has not been
reported from the peninsula.

SONORA: Hermosillo, 15 1 $ 20-x.

SINALOA: Mazatlan, 15 29-x; 2 mi. east of Concordia, 15 1$ 25-x; 19 mi. east
of Concordia, 1 5 25-x.

Vettius fantasos fantasos (Stoll), 1780.

This species is common in the humid parts of the Neotropics, but
seems to be absent from arid or semi-arid regions.

SINALOA: 48 mi. northwest of Culiacan, 15 1$ 23-x.

1969

Miller: Mexican Hesperiidae

193

Ancyloxypha arene (Edwards), 1871.

This species rejlilaces numitor (Fabricius) in the southwestern United
States, Mexico, and Central America. It has not been recorded from the
peninsula, but there is the barest possibility it may occur in suitable
habitats in the northern part near the mouth of the Colorado River.

SONORA: 16 mi. south of Navojoa, 1^ 22-x; 16 mi. south of Guaymas, 2$ 21-x.

SINALOA: 48 mi. northwest of Culiacan, 1 ($ 1$ 23-x; Mazatlan, 1^ 31-x.

Copaeodes aurantiaca (Hewitson), 1868.

This species and the next tend to replace one another, aurantiaca
being far more abundant in the very arid regions, while minima is
commoner in more humid environments. The present species is found
in the southwestern United States and western Mexico, including the
peninsula.

C. aurantiaca was present in almost all areas of the peninsula visited
by the Expedition, even in some of the driest arroyos. It does not visit
flowers as readily as most of the skippers, but when it does, it prefers
large yellow composites. Most of the time, however, it perches on a
rock or a twig and waits for other butterflies, with whom it engages in
mock combat.

BAJA CALIFORNIA SUR: La Paz, Hotel Guaycura grounds, 1^ 21-x, 15 22-x;
5 mi. east of La Paz, 2 $ 9-xi; southeast shore of La Paz harbor, 15 19 8-xi;
Rancho Rosarito, 3 5 19 23-x; Rancho Vinorama, 1 9 23-x; Rancho El Novillo, 2 5
28-x; Rancho Palmarito, 3 5 19 27-x, 3 5 31-x, 19 4-xi, 3 5 19 5-xi,
25 19 30-xi, 15 19 4-xii; Mesa Puerta Azul, 3 5 24-x, 15 25-x, 15 1-xi,
2 5 2-xi, 1 5 12-xi; Arroyo San Bartolo, 1 5 2-xi, 2 5 3-xi; Miraflores, 1 5 25-x;
Boca de la Sierra, 2 5 2 9 24-xi, 4 5 28-xi; Rancho San Bernardo, 1 5 14-xi; San
Jose del Cabo, 1 5 22-xi; Puerto Chileno, 2 5 22-xi; 4 mi. south of Arroyo Cande-
laria, 2 5 24-xi.

SONORA: 12 mi. south of Hermosillo, 15 2 9 20-x; 36 mi. south of Guaymas,
15 20-x.

SINALOA: Mazatlan, 1 9 24-x.

Two additional specimens have been examined from the collection of the San
Diego Natural History Museum.

BAJA CALIFORNIA NORTE: Valle dc la Trinidad, 1 5 16-iii-1936; Santa Agueda,
1 5 16-ix-1935.

Copaeodes minima (Edwards), 1870.

This species occurs in moist to semi-arid areas from the southern
United States to Costa Rica. It has not been reported from the peninsula.

SINALOA: Mazatlan, 19 24-x, 4 5 2 9 26-x, 19 29-x, 15 31-x; Urias, 2 mi.
south of Mazatlan, 1 9 31-x.

194

Annals of Carnegie Museum

VOL. 41

Hylephila phyleus phyleus (Drury), 1770.

This species, one of 'the most widespread in the Neotropics, penetrates
the temperate regions to the north and south. It is common on the
peninsula.

H. phyleus is an avid flower visitor and almost any cultivated flower
garden on the peninsula swarmed with these butterflies. Even when they
are visiting flowers, these butterflies are always ready to engage another
butterfly, a bird, or a butterfly collector in mock combat. They are easily
captured when they are visiting blossoms, but otherwise these skippers
are extremely wary.

BAJA CALIFORNIA suR: La Paz, Hotel Guayciira grounds, 1- 20-x, 8^ 2$ 21x,
1 $ 7-xi; southeast shore of La Paz harbor, 8 $ 8-xi, 2$ 3 $ 10-xi, 1 $ 5-xii;
Rancho Palmarito, 12 31-x, 2$ 4-xi, 12 5-xi; Mesa Puerta Azul, 1^ 24-x;
Arroyo San Bernardo, 1 17“Xi; San Jose del Cabo, 2 2 23-xi, 3 2 25=xi; Arroyo

Candelaria, 1 5 2 2 26-x.

SONORA: 5 mi. west of Riito, 1 2 17-x; 16 mi. south of Guaymas, 1 $ 21-x, 1 $
22-x.

SINALOA: Mazatlan, 1^ 26-x; Urias, 2 mi. south of Mazatlan, 1^ 12 31-x;
5 mi west of Concordia, 1 5 2-xi,

Two other specimens from the collection of the San Diego Natural History
Museum have been examined.

BAJA CALIFORNIA SUR: Commondu, 12 16-xi-1956,

BAJA CALIFORNIA NORTE: El Barril, 12 14-iii-1947 (C. F. Harbison),

Yvretta cams subreticuhta (Plotz), 1883.

A single small, rather dark female of this subspecies was taken by
the mainland party. I am indebted to Dr. C. Don MacNeill for identify-
ing the specimen. Thus far suhreticulata is known from western Mexico
and Guatemala; it does not appear to be common.

SINALOA: Mazatlan, 12 24-x.

This subspecies is widespread throughout the American tropics but
has not been recorded from the peninsula.

SINALOA: 16 mi, north of Mazatlan, 12 29-x; Mazatlan, 3^ 24-x, 12 27-x,
2$ 28-x; Urias, 2 mi. south of Mazatlan, 2$ 31-x; 22 mi. east of Concordia, 1^
25-x.

Polites sabuleti chusca ( Edwards ) , 1873.

This is the subspecies found in the Sonoran desert. It almost certainly
occurs in Baja California Norte, since the specimens taken by the main-
land party were collected just across the Colorado River from that state.

SONORA: 5 mi. west of Riito, 1^22 17-x.

1969

Miller: Mexican Hesperudae

195

Polites sabuleti rmrgaretae Miller and MacNeill, 1969.

This subspecies is known from only a few scattered coastal localities
in the Cape region of Baja California Sur. The records and characteris-
tics are given in the original description (Miller and MacNeill, 1969:
22-23).

Wallengrenia otho (Abbot and Smith), 1797, subspecies.

There are no records of otho from the peninsula. The mainland speci-
mens more nearly resemble curassavica (Snellen) than nominate otho,
but Evans (1955: 333) mentions only otho from Mexico. The nearest
records cited for curassavica are from Panama. Much work needs to be
done on the otho complex both on the mainland and in the Antilles,
before we shall have a reasonable understanding of the subspecies
involved.

SINALOA: Urias, 2 mi. south of Mazatlan, 1 $ 1-xi; 19 mi. east of Concordia,
1$ 31-x.

Pompeius pompeius (Latreille), 1824.

This species is one of the commonest skippers in disturbed areas of
the continental Neotropics. It is not known from the peninsula.

SINALOA: Mazatlan, 2$ 1$ 24-x; Urias, 2 mi. south of Mazatlan, 1$ 31-x;
17 mi. east of Concordia, 1 $ 25-x.

Atalopedes campestris (Boisduval), 1852.

This species is widespread throughout the United States, Mexico and
Central America. It was common particularly in cultivated areas of Baja
California Sur, where its habits are sustantially the same as in the
United States.

BAJA CALIFORNIA SUR : La Paz, Hotel Guaycura grounds, 3 5 2 $ 22-x, 2 $ 6-xi,
1 $ 7-xi, 1 5 8-xi; southeast shore of La Paz harbor, 1 5 8-xi, 6 5 3 $ 10-xi;
Rancho Palmarito, 1 $ 30-xi; Mesa Puerta Azul, 1 5 1-xi; San Jose del Cabo, 1 $
22-xi, 3 5 1 $ 25-xi.

SINALOA: 19 mi. north of Los Mochis, 15 22-x.

Mellana eulogius (Plotz), 1883.

This species is widespread throughout the Neotropics.

SINALOA: 16 mi. north of Mazatlan, 15 29-x; Urias, 2 mi. south of Mazatlan,
15 31-x.

Atrytonopsis ovinia zaovinia Dyar, 1913.

No Atrytonopsis has been recorded from the peninsula. The present
subspecies is restricted to the Sonoran desert.

SONORA: Hermosillo, 15 1$ 20-x.

196

Annals of Carnegie Museum

VOL. 41

Amblyscirtes tolteca Scudder, 1872, Psubspecies.

The nominate subspeeies is widely distributed in southern Mexico,
and the subspecies prenda Evans is thus far known only from Tucson,
Arizona; and Guaymas, Sonora. Rindge (1948 : 309) recorded this skip-
per from the peninsula, but the specimen was not available for study
by MacNeill, who concluded (1962: 111) that tolteca was “a doubtful
inhabitant” of the peninsula. Expedition personnel collected a dozen
specimens in scattered localities in the Cape region, so this species is
definitely present in the peninsula. All these specimens are worn, how-
ever, and I have no specimens at hand of prenda, so I am unable to say
which subspecies is represented by the Cape material. The specimens
are generally a bit smaller than a series at hand from Piste, Yucatan,
but there is no evidence of the grayish flush Evans (1955: 389) employs
to characterize prenda.

A. tolteca is not a common species, and its secretive habits would
account for its being overlooked. It has a rapid flight and always re-
mains close to the ground. It is not a pugnacious butterfly and is so
unobtrusive that the collector must be especially observant to find
specimens. The only flowers visited were those that grew close to the
ground, such as coralvine.

BAJA CALIFORNIA suR: La Paz, Hotel Guayciira grounds, 1 ($ 19-x, 1$ 21-x;
Rancho Palmarito, 27-x, 15 31-x, 3 5 1? 5-xi; Puerto Chileno, 15 22-xi;

Cabo San Lucas, 1 $ 23-xi, 1 $ 24-xi; 4 mi, south of Arroyo Candelaria, 1 5
[26-xi] (26-x).

Lerodea eufala eufala ( Edwards ) , 1869.

This species is common throughout the American tropics and sub-
tropics. It is abundant in Baja California Sur.

These butterflies are avid flower visitors, preferring the cultivated
varieties, and ‘‘desert willow,” coralvine, and large yellow composite
blossoms. While these butterflies are combative, they are not as aggres-
sive as H. phyleus or C. aurantiaca. Along with those two species
eufala is one of the most abundant hesperiines on the peninsula.

BAJA CALIFORNIA SUR: La Paz, Hotel Cuaycura grounds, 15 2$ 19-x, 15

20-x, 3 5 1$ 21-x, 12 5 17$ 22-x, 15 4-xi, 2 5 1$ 6-xi, 4 5 7-xi, 15 8-xi,
2 5 1 $ 10-xi; southeast shore of La Paz harbor, 9 5 8-xi, 9 5 2 $ 10-xi,
2 5 1 $ 5-xii; Rancho Rosarito, 1 $ 23-x; Rancho El Novillo, 1 5 28-x; Rancho
Palmarito, 7 5 27-x, 3 5 1$ 31-x, 2 5 2$ 4-xi, 25 4$ 5-xi, 65 3 5 30-xi, 3$
4-xii; Arroyo Hondo, near El Triunfo, 15 1 $ 24-x; Mesa Puerta Azul, 15 3 $
24-x, 2 5 1 $ 1-xi; Arroyo San Bartolo, 2 5 2 $ 2-xi, 1 5 3-xi, 2 5 2 $ 12-xi,
1 5 14-xi; Rancho Ensenada de Palmas, 1 5 16-xi; Bahia de Palmas, 1 5 16-xi,

1969

Miller: Mexican Hesperiidae

197

1 S 20"Xi; Boca de la Sierra, 1 $ 24-xi, 1 $ 28-xi; Arroyo San Bernardo, 1 $ 18-xi;
Rancho San Bernardo, 1 $ 14-xi; La Cienaga I, 1 5 16-xi; San Jose del Cabo, 1 $
22-xi, 9 ^ 5 $ 23-xi, 14$ 49 25-xi; Puerto Chileno, 3 $ 22-xi; Cabo San Lucas,

2 $ 23-xi; Arroyo Candelaria, 1 $ 24-xi; 4 mi. south of Arroyo Candelaria, 1 $
26-x, 4^1$ 24-xi.

SONORA: 5 mi. west of Riito, 1^ 17-x; 10 mi. northwest of Caborca, 1^ 19-x;
Hemiosillo, 1 $ 20-x; 12 mi. south of Hermosillo, 3 9 20-x; 22 mi. north of Ciudad
Obregon, 1 $ 21-x; Guaymas, 1 $ 21-x.

SINALOA: 16 mi. north of Mazatlan, 1$ 29-x; Mazatlan, 1$ 24-x, 1$ 29-x,

1$ 31-x.

Lerodea dysaules Godman, 1900.

This species is known only from Baja California Sur and western
Mexico. It is closely related to L. arabus (Edwards) and is probably
a subspecies of that species. There is a wide range of individual varia-
tion in the specimens before me. All males have faintly delimited spots
on the under surface of the hindwing in at least spaces Ms-Cui and
Cui-Cu2. The spotting of the forewing, however, is far more variable:
the cell spot is obsolescent in one of the males, a female from Baja
California, and the female from Sinaloa, and the spots in general are
variously developed, from prominent to almost obsolete, particularly
those in spaces M.3"Cui and Cu2“2A. The variation is without regard
to the configuration of the male genitalia.

These butterflies are rather secretive skippers, little given to com-
bativeness. They fly low, along the ground, and are attracted only to
flowers that are a foot or less in height.

BAJA CALIFORNIA SUR: La Paz, Hotel Guaycura grounds, 20-x, 1$ 21-x, 1$
21-x; southeast shore of La Paz harbor, 1 $ 8-xi; Rancho Rosarito, 1 ^ 1 $ 23-x.
SONORA: Hermosillo, 1^ 20-x.

SINALOA: Mazatlan, 1$ 28-x.

Two more specimens have been examined from the collection of the San Diego
Natural History Museum.

BAJA CALIFORNIA SUR: Comondu, 1$ 1$ 16-xi-1952 (C. F. Harbison).

Calpodes ethlius ( Stoll ) , 1782.

This species is abundant throughout many parts of the Neotropics,
seemingly limited by the distribution of its foodplant, Canna. Since
it has been recorded on the peninsula only from the environs of settle-
ments where many ornamental cannas are grown year in ad year out,
I am inclined to the opinion that ethlius is not native to Baja California,
but rather was introduced there along with ornamental cannas. C.

198

Annals of Carnegie Museum

VOL. 41

ethlius is, however, a well-known migrant (see Welling, 1964), and j
it is possible that stray specimens have often reached Baja California ;i
on their own. But not until the advent of cultivated cannas was there
opportunity for this butterfly to establish colonies there. In any event, J
this species is certainly not part of the natural peninsular fauna and
is a recent arrival. i

BAJA CALIFORNIA suR: La Paz, Hotel Giiaycura grounds, 15 22-x; Bahia de
Palmas, 1 5 1 $ 16-xi, 1$ 29 20-xi.

SINALOA: Mazatlan, 2 5 27-x, 15 1? 29-x.

Panoquina panoqiiinoides errans ( Skinner ) , 1892.

This subspecies was well characterized by MacNeill (1962: 113- i
114). It is known from California, most of the peninsula, and western
coastal Mexico.

These butterflies are always associated with coastal grasses and are i
seldom found far from them. They are most active late in the afternoon, i
when the males will take positions on grass stems and engage one i
another in mock combat or will hover around females. The females :
generally fly deep in the grass and are not so combative as the males.
These skippers seldom visit flowers, but flowers were not abundant on
the tidal flats where we observed these butterflies.

BAJA CALIFORNIA SUR: La Paz, Hotel Guayciira grounds, 1? 22-x; southeast
shore of La Paz harbor, 215 5? 8-xi, 14 5 10 9 lO-xi, 10 5 10 9 5-xii, ,

2 5 6-xii; San Jose del Cabo, 15 22-xi, 4 5 2.3-xi, 5 5 25-xi, 19 26.xi.

SINALOA: Mazatlan, 25 19 26-x, 2 5 1 9 28-x, 15 29 9 x, 1 5 31-x.

An additional specimen from the collection of the San Diego Natural Histor>'
Museum has been examined.

BAJA CALIFORNIA NORTE: Calmalli, 15 24-vii-1953,

Panoquina ocola (Edwards), 1863. t

This species is widespread in the tropics and subtropics; it reaches
the Antilles, but has not been recorded from the peninsula.

SONORA: 16 mi. south of Guaymas, 15 19 21-x.

SINALOA: 19 mi. north of Los Mochis, 1 5 22-x; 10 mi. southeast of Los Mochis,

15 2 9 23-x; 16 mi. north of Mazatlan, 3 5 29-x; Mazatlan, 1 9 26-x, 1 9 28-x;
Urias, 2 mi. south of Mazatlan, 1 5 31-x; 17 mi. east of Concordia, 15 19 25-x,

15 1-xi.

Panoquina pauper pauper (Mabille), 1878.

A single specimen of this rather uncommon tropical species was
taken by the mainland party.

SINALOA: 11 mi. north of Mazatlan, 19 28-x.

1969

Miller: Mexican Hesperiidae

199

Nyctelius nyctelius nyctelius (Latreille), 1824.

This species is widespread throughout western Mexico, Central and
South America, and the Antilles. It is not known from the peninsula.

siNALAO: 10 mi. southeast of Los Mochis, 1 $ 23-x; Urias, 2 mi. south of Mazat-
lan, 1 $ 31-x; 17 mi. east of Concordia, 1 $ 1-xi.

Thespeius dalman (Latreille), 1824.

This species has a wide range in the continental Neotropics.

SINALOA: 17 mi. east of Concordia, 1$ 1-xi.

References Cited

Bell, E. L.

1941. Two new species of Hesperiidae from North America. Ent. News, 52:
163-169; figs. 1-3.

Brown, F. M.

1964. The W. H. Edwards types of Hesperiidae lost on the “S. S. Pomerania” in
1878. Ent. News, 75: 24-25.

Burns, J. M.

1964. Evolution in skipper butterflies of the genus Erynnis. Univ. California
Publ. Ent, 37: 1-216; pi. 1, figs. 1-24.

Edward:, W. H.

1876. Description of new species of diurnal Lepidoptera found within the
United States and British N. A. Trans. American Ent. Soc., 5: 202-208.

Evans, W. H.

1951. A catalogue of the American Hesperiidae ... in the British Museum
(Natural History). Part 1. Introduction and Group A. Pyrrhopyginae.
London, Trustees British Mus.: vii-x -|- 92 pp.; pis. 1-9.

1952. Op. cit. Part II (Groups B, C, D). Pyrginae, Section 1. v -|- 178 pp.; pis.
10-25.

1953. Op. cit. Part III (Groups E, F, G). Pyrginae, Section 2. v -f- 246 pp.;
pis. 26-53.

1955. Op. cit. Part IV (Groups H to P). Hesperiinae and Megathyminae, v +
499 pp.; pis. 54-88.

Fox, R. M.

1963. Reports on the Margaret M. Cary and Carnegie Museum Expedition to
Baja California, Mexico, 1961. 1. Introduction. Itinerary and localities.
Ann. Carnegie Mus., 36(16): 181-192; fig. 1.

Holland, W. J.

1931. The butterfly book (rev. ed.). Garden City, New York, Doubleday, Doran
and Co.: xii -|- 424 pp.; pis. 1-78.

Lindsey, A. W., E. L. Bell, and R. C. Williams, Jr.

1931. The Hesperioidea of North America. Denison Univ. Bull. Jour. Sci. Labs.,
26: 1-142; pis. 1-33.

MacNeill, C. D.

1962. A preliminarv report of the Hesperiidae of Baja California (Lepidoptera).
Proc. California Acad. Sci., (4)30: 91-116.

200

Annals of Carnegie Museum

VOL. 41

Miller, L. D. and C. D. MacNeill

1969. Reports on the Margaret M. Cary-Carnegie Museum Expedition to Baja
California, Mexico, 1961. 5. Two new subspecies of Hesperiidae from the
Cape region, Baja California Sur, Mexico, Ann. Carnegie Mus., 41(2):
19-24; figs. 1-8.

Rindge, F. H.

1948. Contributions toward a knowledge of the insect fauna of Lower Califor-
nia. No. 8. Lepidoptera: Rhopalocera. Proc. California Acad. Sci., (4)24:
289-311.

Tilden, J. W.

1965. Urbanus procne and Urbanus simplicius (Hesperiidae). Jour. Lepidop-
terists’ Soc., 19: 53-55.

Truxal, F. S.

1960. The entomofauna with special reference to its origin and afiBnities. In
Symposium: the biogeography of Baja California and adjacent seas. Part
III. Terrestrial and freshwater biotas. Syst. Zook, 9: 165-170.

Welling, E. C.

1964. Butterfly migrations in southeastern Mexico during 1963 and 1964. Jour.
Lepidopterists’ Soc., 18: 229-230.

>7. 7 3
' H P 6^

Volume 41

Annals of Carnegie Museum

Article 6

PALEONTOLOGY AND GEOLOGY OF THE
BADWATER CREEK AREA, CENTRAL WYOMING
Part 5. The cylindrodont rodents

Craig C. Black ^

Section of Vertebrate Fossils,
Carnegie Museum, Pittsburgh, Pennsylv

Introduction

In this, the fifth in a series of papers dealing witlr^^tebrates^^3m
Badwater Creek, Wyoming, four species of cylindriodont rodents are
described, one as a new species of Pseudocylindrodon. All specimens
are from the Hendry Ranch Member of the Tepee Trail Formation
(Tourtelot, 1957) and are of late Eocene or Uintan age. There are
approximately one hundred and fifty isolated cylindrodont teeth in
the Badwater collections at Carnegie Museum and at the University of
Colorado Museum. The large majority of these specimens are from a
single locality and appear to represent a single species, but because
of the lack of associated dentition, it is possible that one or two speci-
mens that represent different populations have been erroneously
grouped together. Such an error, if it exists at all, is probably minor,
however, as most of the specimens appear to fall quite readily into
four discrete categories.

Wood ( 1949 ) was the first to describe rodents from the late Eocene
deposits along Badwater Creek, but he did not mention the presence
of cylindrodonts in the fauna. However, he described and figured
(1949, p. 559, fig. 11) an upper incisor which he referred to as cricetid
indet. This incisor (Amherst College Museum 10004) is probably that
of a cylindrodont rather than a cricetid and it appears to be referable
to the new species of Fseudocylindrondon described below. The upper
incisor of Pseudocylindrodon medius and of P. neglectus is oval in

1 Present address: Museum of Natural History, University of Kansas, Lawrence,
Kansas 66044.

Issued September 25, 1970

201

202

Annals of Carnegie Museum

VOL. 41

cross-section with a rounded anterior face and has the enamel limited
to the anterior of the tooth. This is the condition described for the
Amherst specimen referred to as a cricetid by Wood.

Abbreviations used in this paper are: CM, Carnegie Museum; UCM,
University of Colorado Museum; a-p, anteroposterior; tr, transverse
width, the first measurement given is the width of the upper teeth or
the anterior width of the lower teeth, the second measurement is the
posterior width of the lower teeth; N, number; O.R., observed range;
M, mean; s, standard deviation; V, coefficient of variation.

ACKNOWLEDGEMENTS: This work was supported by NSF grants
GB-1266 and GB-4089 and by grants from the Childs Frick Corporation
and the Gulf Oil Corporation. I would like to thank Dr. Peter Bobinson
for the loan of the University of Colorado Museum specimens and Dr.
Mary Dawson for many discussions of problems dealing with early
Tertiary rodents. Figures 1-26 are by Donald Bodkey, figures 27-35 by
A. Lee Ambrose.

Since 1962, when we began our work along Badwater Creek, we have
received all manner of assistance from Mr. and Mrs. James Hendry and
their son. Bob, owners of the Clear Creek Cattle Co., on whose prop-
erty most of our quarries are located. Their help and their friendship
were decisive factors in making this project a success and I wish to offer
our deepest thanks and appreciation for all they have done.

Systematic Review
Order Bodentia
Family Cylindrodontidae
Genus Pseudocylindrodon Burke, 1935
Pseudocylindrodon tobeyi^, new species
Figures 1-24

type: cm 21349, LM i or 2.

hypodigm: dP4, CM 21301; Ph CM 14633, 15294, 21302, 21303, 21304;

CM 14636, 14637, 15295, 15298, 15302, 15304, 15309, 15312, 15313,

16076, 16764, 21305-21321; M^, CM 15293, 15296, 15297, 15306, 15310, 21322-
21332; dP4, CM 21341; P4, CM 15355, 16001, 16074, 16763, 21340, 21341;
Mi-2, cm 14634, 14635, 14638, 15299, 15300, 15303, 15311, 16765, 16835, 21342-
21356, 21358, 21359, 21370; Ms, CM 14632, 15305, 16017, 21357, 21360-21365.

HORIZON AND LOCALITIES: Late Eocene, Wood and Rodent localities (Black
and Dawson, 1966).

DIAGNOSIS: Larger than Pseudocylindrodon medius and P. neglectus; cheek
teeth not as high crowned as in P. medius and P. neglectus; M^-^ not as wide in

1969

Badwater Area: The Cylindrodont Rodents

203

relation to length as in other species of Pseudocylindrodon; only partially re-
duced; hypoconid on P4-M3 not as enlarged and not pulled anteriorly as strongly
as in P. medius and P. neglectus; Ms not greatly reduced.

description: The deciduous is more triangular in occlusal outline
than because of the enlarged buccal half of the anterior cingulum.
The cusps are quite prominent on dP^ with the lophs poorly developed.
The protoconule and metaconule are large and there is a small but
distinct hypocone present. The permanent fourth premolar is as large
as and and is similar in structure. P^ differs from the molars
primarily in having a much shorter anterior cingulum which never
passes internally more than halfway along the anterior face of the
tooth. The posterior cingulum is also shorter than it is on and
M^. No hypocone can be distinguished. On all the P^s in the sample
the metaconules are large and the metaloph is incomplete, failing to
reach the internal face of the protocone. In one specimen, CM 21303,
there is a short, thin crest from the metaconule to the posterior cingulum.
Mesostyles are prominent on all P^s.

It is impossible to separate the first from the second upper molars
when dealing with isolated teeth and the same is true for the lower
first and second molars. The descriptions given here, therefore, apply
to both. As in Oligocene cylindrodonts there is considerable variation
in occlusal outline of M^-M^ which is dependent upon the stage of
wear. Generally, are somewhat wider than long but the impres-

sion is of a rather square occlusal outline. The internal border of the
teeth is broad although there is no distinct hypocone. The union of the
posterior cingulum with the protocone, however, is broad and the
postero-internal corner of the tooth is consequently squared off. The
anterior cingulum does not pass internally to the protocone, but covers
only about two-thirds of the anterior face of the tooth. The mesostyles
are usually large although in two specimens they are absent. In no
instance is the buccal end of the central valley closed by the mesostyle
or posterior slope of the paracone. There is considerable variation in the
size of the metaconule which is occasionally almost undistinguishable
within the metaloph, but more normally is quite large. The metaloph is
either constricted at the protocone, fusing into the internal face well
below the top of the cusp, or occasionally it is incomplete. In some
specimens there is a short crest from the metaconule to the posterior
cingulum (fig. 5) but more often the posterior valley is not bridged

1 Named for _Mrs. David Ruhala, formerly Miss Tobey Hall, vi^ho was a cheerful, able, and
dedicated assistant with the Carnegie Museum field parties for three summers while we worked
the Badwater area.

204

Annals of Carnegie Museum

VOL. 41

by such a crest. The protoloph and metaloph diverge only slightly as
they pass from the protocone to the buccal margin of the tooth.

Figs. 1-21. Pseudo cylindrodon tobeyi, isolated teeth, all approximately x7. 1. RdP^,
CM 21301. 2. RP4, CM 21303. 3. LP^ CM 21302. 4. RM" 2^ cM 15304. 5. LMi
or 2 CM 15312. 6. LMior2^ CM 21321. 7. RMior2^ CM 15302. 8. RM^, CM
21322. 9. RM3, CM 15310. 10. LM^, CM 21323. 11. RdP4, CM 21341. 12. LP4,
CM 21340. 13. LP4, CM 15355. 14. LMi or 2, CM 21349, Type. 15. RMi or 2, CM
14634. 16. RMi or 2, CM 16835. 17 RMi or 2, CM 21356. 18. LMi or 2, CM
21343, 19. RMs, CM 15299. 20. LMs, CM 21345. 21. LMs, CM 21367.

1969

Badwater Area; The Cylindrodont Rodents

205

The third upper molars are roughly square in oeclusal outline and
are considerably smaller than and M^. The protoloph and metaloph
pass more or less directly across the tooth, not diverging as much as
they do in the Oligocene species. A metaconule is still distinct on a few
M^s (fig. 10) but more commonly it is absent. In several specimens
there is a loph from the metaloph or metaconule running posteriorly to
the posterior cingulum. One or two specimens are not at all reduced,
resembling and to a remarkable degree (fig. 8). There is no
posterior interdental wear facet on these teeth, however, indicating
they are certainly third molars.

The deciduous lower fourth premolar is quite long and narrow. The
trigonid basin is wide and closed by the anterior cingulum and by a
heavy posterior protoconid arm. The hypoconid is not enlarged or
pulled forward as in the permanent P4, hence the buccal valley is wide.
The hypolophid is strong and there is a wide valley between the
hypolophid and posterolophid.

There is considerable variation in the structure of P4 among the speci-
mens here assigned to P. toheiji but the sample size is small and there
is nothing in the sample of molars that would suggest that more than
a single species is present. The amount of variation is displayed in two
specimens, both taken from one quarry (figs. 12 and 13). P4 is generally
as long as Mi and M2 but not as wide. The protoconid and metaconid
are appressed, with only a small trigonid basin intervening. There is
no trace of an anterior cingulum, so that the trigonid basin opens onto
the anterior face of P4. The hypoconid is swollen and pulled anteriorly,
resulting in a postero-internal direction for the buccal valley. The
posterolophid is quite strong on all specimens and joins the entoconid
just below the apex of that cusp, with the posterior valley thus closed
off internally. The hypolophid is completely absent in one specimen
(fig. 12), while it is a strong, complete crest uniting with the posterior
end of the ectolophid in another ( fig. 13 ) . There are no stages of partial
hypolophid development in the sample. When the hypolophid is present
the posterior valley becomes isolated as a shallow basin after slight
wear. The central, or talonid, basin opens internally as the entoconid
and metaconid are separated by a deep cleft.

The first and second lower molars are identical as far as can be deter-
mined from a collection of isolated teeth. They are also quite uniform,
showing little morphologic variation within the sample. Mi -Mo are
rectangular in occlusal outline, with the anterior half of the teeth some-

206

Annals of Carnegie Museum

VOL. 41

what narrower than the posterior. The trigonid basin is of moderate
size in an unworn Mi or M2 and it is elosed both anteriorly and pos-
teriorly. The hypolophid is always well-developed as is the posterolo-
phid. The central valley is generally open internally with no metastylid
present, while the posterior valley is closed internally after moderate
wear. The hypoconid is swollen and pulled forward as in the Oligocene
species.

M3 resembles Mi and M2 in most details but differs from them in
being more variable in size. Also, in some specimens of M3, the central
valley is closed lingually and the posterior half of the tooth is somewhat
reduced. This is effected through shortening of the hypolophid and
posterolophid. One specimen, CM 21367, (fig. 21) is greatly reduced
and approaches the condition seen in an unworn M3 of Cylindrodon.
However, this tooth is much lower crowned than any M3 known of
Cylindrodon.

DISCUSSION: In view of its age, Pseiidocylindrodon tobeyi is amazingly
similar to both P. medius and P. neglectus of the early Oligocene of
Montana. There are a number of minor differences in dental morphology
between the three, but P. tobeyi certainly bears out Burke’s conviction
(1935, p. 12) that “the ancestor of Cylindrodon in the Upper Eocene
had progressed beyond the stage represented in Pareuniys{?) troxelli
m!’ and confirms his belief that Pareumys could not have been ancestral
to Cylindrodon or Pseiidocylindrodon. By Uintan time there was at least
one lineage whose cheek teeth were considerably higher crowned than
in any species of Pareumys and this lineage could have been ancestral
to Cylindrodon of the early Oligocene.

Pseiidocylindrodon tobeyi differs from Pareumys not only in having
considerably higher crowned cheek teeth but in a number of other
features as well. The metaloph in Pareumys is incomplete (Wilson,
1940), whereas in Pseiidocylindrodon tobeyi it is usually complete. The
metalophids on the lower molars are always complete and close the
trigonid basin, while in Pareumys, the metalophids are progressively
shortened from Mi to M3. The hypoconids of Pseiidocylindrodon are
swollen and pulled anteriorly, while in Pareumys they are more as in
My sops or various ischyromyids.

Differences between Pseiidocylindrodon tobeyi and the Oligocene
species of Pseiidocylindrodon are less striking and are usually ones of
degree of development of structures rather than differing modes of
development. There is less unilateral hypsodonty of P^-M^ in P. tobeyi

1969

Badwater Area: The Cylindrodont Rodents

207

than in P. medius and P. neglectus, and the Badwater species is some-
what lower crowned. P. tobeyi is larger than the Oligocene species, and
it is also larger than Cylindrodon. The valleys of the lower cheek teeth
of P. tobeyi are not closed lingually as completely as they are in P.
medius and P. neglectus. The hypoconid does not swing anteriorly as
far nor is it as swollen in P. tobeyi as it is in P. tnedius and P. neglectus.
Finally, the M3/3s of P. tobeyi are not as greatly reduced as those of
the Oligocene species of Pseudocylindrodon.

Through a slight reduction in size and a further emphasis on increase
in crown height, P. tobeyi could have given rise to Cylindrodon, and
with less increase in crown height it could have given rise to Pseudo-
cylindrodon medius. If the Badwater species was not directly ancestral
to the Oligocene cylindrodonts, it was certainly much closer to the
Mysops-Cylindrodon lineage than was any species of Pareumys.

Wilson (1940, p. 102, pi. 1, fig. 8) described and illustrated a single
tooth from the Eocene Sespe deposits which he referred to as Pareu-
mys(?) sp. This specimen was stated to differ from Pareumys in greater
height of crown and completely enclosed posterior basin, and Wilson
suggested the specimen might have affinities to the Oligocene genera
Pseudocylindrodon and Cylindrodon. In addition to the characters men-
tioned by Wilson, the hypoconid on this specimen is more swollen and
is pulled further anteriorly than it is in the other California specimens
of Pareumys described in the same paper. This tooth resembles the

TABLE 1

MEASUREMENTS

IN MILLIMETERS, P.

tobeyi.

NEW SPECIES

N

M

s

V

O.R.

P4 a-p

4

1.94

tr

4

2.19

Ml o*” 2 a-p

27

1.92

.09

4.69

1.75-2.10

tr

27

2.11

.14

6.64

1.80-2.40

CO

14

1.78

.10

5.62

1.50-1.90

tr

14

1.74

.12

6.90

1.50-2.00

P4 a-p

6

1.94

tr metalophid

6

1.23

hypolophid

6

1.85

M 1 or 2 a-p

22

1.96

.11

5.61

1.80-2.20

tr metalophid

22

1.69

.14

8.28

1.45-1.95

hypolophid

22

1.95

.10

5.13

1.75-2.10

Ms a-p

7

2.06

tr metalophid

7

1.81

hypolophid

7

1.74

2C8

Annals of Carnegie Museum

VOL. 41

lower molars of Pseudocylindrodon tohetji to a great extent and it may
represent a species of Pseudocylindrodon in the California late Eocene
which is at the same level of development as the population at
Badwater.

^Pseudocylindrodon sp.

Figures 25-30

material: RP4, CM 15568; LM^, CM 16896; M1-2, CM 14813, 15314, 15578,
17504, UCM 24773, 24774; Ms, CM 15315, UCM 24765, 24828. I

HORIZON AND LOCALITIES: Late Eoceiie, localities 5, 5A, 6, and 7 (Black and 1
Dawson, 1966).

description: The is unilaterally hypsodont but not as high crowned
as the upper fourth premolars of Pseudocylindrodon toheyi. The pro- j
toloph and metaloph are complete with no distinct conules present. |
These lophs are rather low, joining the protocone considerably below ‘
the top of that cusp. There is no hypocone or mesostyle. The anterior 1
cingulum is long, while the posterior cingulum is short and does not |]
reach the buccal margin of the tooth.

The tooth here considered to be an M^, CM 16896, has an unusual
pattern unknown in any other cylindrodonts. Both the protoloph and
the metaloph fail to reach the protocone. Instead they are both joined |
to a short ridge that passes between them from the protocone. The
protocone itself is hardly distinguishable as a cusp, rather the entire j
lingual ridge of the tooth is thickened. The anterior and posterior cin- j
gula are of equal length and neither reaches the buccal border of the
tooth.

Ml and M2 are rectangular in occlusal outline. They are somewhat
lower crowned than the molars of P. toheyi, but are higher crowned
than any Pareumys molars. The posterior protoconid arm is not fused ^
to the posterior slope of the metaconid and in some specimens it is
directed posteriorly so that the trigonid basin is open to the rear (figs.

27 and 28). The hypolophid is strong in all specimens and the hypo-
conulid is distinct. The central and posterior valleys are usually open j
lingually, but they are partially closed in CM 15578. The hypoconids are
not greatly swollen and there is only slight indication of the antero-buc-
cal torsion of the hypocone seen in other species of Pseudocylindrodon.

On M;i the posterior protoconid arm is very short, with the trigonid
basin open posteriorly. The central valley is open, but the posterior
valley is closed. The posterior half of the tooth is onh' slightly reduced
through shortening of the posterior cingulum.

1969

Badwater Area: The Cylindrodont Rodents

209

DISCUSSION: It is possible that these nine specimens do not all repre-
sent the same species. The seven lower teeth from four different local-
ities all seem reasonably alike morphologically. The association of the
and with the others is very questionable. Both these teeth,
however, appear to be cylindrodont in morphology, and are of the
proper size to be associated with the lower teeth. Nevertheless, no
cylindrodont known possesses some of the features seen in the and
M^, while the lower teeth are similar to lower molars known for other
species of Pseudocylindrodon and Pareumys.

The lower teeth appear to be intermediate between Pareumys and
Pseudocylindrodon. They differ from Pareumys and resemble Pseudo-
cylindrodon in: (1) being slightly higher crowned; (2) having some
swelling and anterior torsion of the hypocone. They differ from Pseudo-
cylindrodon and resemble Pareumys in having: (1) open central and
posterior valleys; (2) short posterior protoconid arm with open trigonid
basin.

The material is too scrappy and distinctions still too hazy to warrant
description of a new taxon. This taxon, if indeed these specimens repre-
sent a single taxon, is more primitive than Pseudocylindrodon tobeyi
and more advanced than the Badwater or any other species of Pareumys.

TABLE 2

MEASUREMENT IN MILLIMETERS, ?Pseudoctjlindrodon SP.

a-p

tr

tr

CM 15568 RP4

1.95

2.70

CM 18896 LM3

1.90

2.15

CM 15314 LMi or 2

2.20

2.00

2.30

CM 15578 LMi or 2

2.40

2.30

2.60

CM 14813 RMi or 2

2.25

' 2.10

2.35

CM 17504 RMi or 2

2.10

1.70

1.90

UCM 24773 LMi or 2

2.40

2.20

2.45

CM 15315 LMs

2.35

2.10

UCM 24828 LMs

2.20

2.10

Genus Pareumys Peterson, 1919
Pareumys sp.

Figures 31-32

material: cm 15989 and CM 15990, both left Mi or M2.

HORIZON AND LOCALITY: Late Eocene, locality 6 (Black and Dawson, 1966).

description: Both molars are low crowned and have low lophids,
much lower than those of Pseudocylindrodon tobeyi. The hypolophids

210

Annals of Carnegie Museum

VOL. 41

Figs. 22-35. Badwater cylindrodonts, all approximately x7. 22-24, Pseudocylindro-
don tobeyi. 22. RMs, CM 15308. 23. RP4, CM 21339. 24. CM 15301.

25-30, ?Pseudocylindrodon sp. 25. RP^, CM 15568. 26. LM^, CM 16896. 27.
LMj ,, CM 15578. 28. LMi or 2, UCM 24773. 29. LM, 3, CM 15314. 30. LMs,
CM 15315. 31-32, Pareumys sp. 31. LMi or 2, CM 15990. 32. LM^ CM 15989.
33-34, PCylindrodont sp. 33. RP4, CM 21371. 34. LP4, CM 21369. 35. RMs, CM
21366.

1969

Bad WATER Area: The Cylindrodont Rodents

211

are somewhat lower than those of Pareumys milleri or P. grangeri. The
two Badwater teeth are appreciably longer than wide, another Pareu-
mys character in contrast to Pseudocylindrodon. The hypoconids of
Pareumys sp. are not swollen nor do they show any anterior torsion as
they do in the higher-crowned cylindrodonts. The posterolophids are
low and the hypoconulids indistinct. The central and posterior valleys
are broad and shallow. There is no metastylid. On CM 15990 the
posterior protoconid arm passes into the metaconid slope isolating the
trigonid basin from the talonid. These two teeth are slightly larger than
either the first or second molars of Pareumys milleri or P. grangeri and
the cusps and crests of the Badwater specimens are somewhat lower
than in the Uinta species. In all other respects, however, they are quite
close to CM 2938, the type of P. milleri.

There is not sufficient material available from this Badwater popula-
tion of Pareumys to allow specific determination. It is impossible to tell
whether this population is consistently larger than P. milleri or merely
a slightly larger variant of that species.

TABLE 3

MEASUREMENTS IN MILLIMETERS

a-p

tr

tr

CM

15989 LMi or 2

2.55

2.35

2.40

CM

15990 LMi or 2

2.55

2.30

2.30

PCylindrodontid sp. small species
Figures 33-35

material: RP4, CM 21371; LP4, CM 21369; RM3, CM 21366.

HORIZON AND LOCALITY: Late Eocenc, Wood locality (Black and Dawson, 1966).

DESCRIPTION: All teeth are extremely small, much smaller than those
known for any other cylindrodontid. On the metaloph is complete
but there is a second crest from the metaconule to the posterior cin-
gulum (fig. 33). The tooth is unilaterally hypsodont. The anterior
cingulum is quite short, a character commonly found on P^s of other
cylindrodonts. There is no mesostyle and there is no distinct hypocone.
On the lower premolar there is almost no separation between the
protoconid and metaconid. There is no hypolophid. The hypoconid is
swollen and pulled slightly forward. The ectolophid does not reach
the hypoconid, leaving the central basin of the tooth open into the
buccal valley. The M3 has no posterior protoconid arm, so there is no
trigonid basin. There is just an open slope from the anterior margin of

212

Annals of Carnegie Museum

VOL. 41

the tooth into the talonid basin. There is a strong hypolophid (fig. 35) ||
on Ms and the hypoconid is enlarged and has a slight anterior torsion, i:

These three teeth are tentatively referred to the Cylindrodontidae I
because they display several characters common to that family and not '
found in other late Eocene rodent families: (1) unilateral hypsodonty
of P^; (2) enlargement of hypoconid and its anterior direction on P4 I;
and Ms; and (3) strong hypolophid on Ms. There is, of course, no cer-
tainty that these three should be associated. However, because of their
occurrence together at one locality, their uniform small size, and their
similarity to other cylindrodonts, the association seems possible.

The incomplete ectolophid on P4 and the complete absence of a :
posterior protoconid arm on Ms are unique features unknown in other 1
cylindrodonts, although some third lower molars of Pareumys have the |
posterior protoconid arm greatly reduced. The three teeth are relatively |
higher crowned than those of Pareumys, however, and the hypoconid ji
development differs from that in Pareumys.

It is of interest that the three specimens are all from a single locality. | j
This small species has not been found at any of the other localities 1
where large samples also have been collected using washing techniques, i
It is hoped that future work will give us more information about this | :
unique, tiny cylindrodont. !

TABLE 4

MEASUREMENTS IN MILLIMETERS

a-p

tr

tr

CM 21371 RP4

1.25

1.50

CM 21369 LP4

1.10

.75

1.15

CM 21366 RM3

1.40

1.20

1.25

Distribution of Badwater Cylindrodonts
The sample of cylindrodont teeth studied in this report is not large, i
Nevertheless, a pattern of distribution of species between localities in i
the Hendry Ranch Member emerges that is probably real. The samples j
of teeth from localities 5, 5A, 6, and Wood were recovered from con- J
centrate resulting from the washing of at least four tons of matrix from '
each locality. The concentrate from each locality appears to be equally
rich both in the number of total mammalian specimens and in the num-
ber of mammalian taxa represented. There is still a small amount of
unsorted concentrate from 5A and much larger amounts from 5, 6, and
Wood. However, it is unlikely that sorting of this concentrate will

1969

Badwater Area: The Cylindrodont Rodents

213

change the distribution pattern of cylindrodont species at these local-
ities, even though it will certainly alter the number of individuals
and perhaps of the percentage of cylindrodonts in relation to the total
fauna at each locality. The cylindrodont samples from localities 7 and
Rodent were obtained through surface prospecting, and consequently
may not reflect a true picture of species diversity at these two localities.
I doubt this, however, as the general pattern for all localities seems
to suggest an association of localities 5, 5A, 6, and 7 on the one hand,
and Wood and Rodent on the other. Locality 7 is thus grouped with
three localities from which large samples of matrix have been processed,
and the Rodent locality is also grouped with a locality from which a
large sample has been taken.

Two cylindrodont species found at Badwater are known from such
small samples that little faith could be placed on any generalizations
based on them. Pareumys sp., the most primitive of the Badwater species,
is known from only two specimens recovered a locality 6. The very small
cylindrodont is known from only three specimens from the Wood
locality. This species is certainly more advanced than the Pareumys
species.

The really striking distribution pattern is provided by Pseudocylin-
drodon tobeyi, which is known to occur only at the Wood and Rodent
localities, with only a half-dozen of the nearly one hundred teeth
coming from the latter locality. With the concentrate resulting from
the processing of at least twelve tons of mati'ix sorted, no specimens
of P. tobeyi are known from localities 5, 5A, 6 or 7. That P. tobeyi is
totally absent, or at the very least, exceedingly rare at these localities
is, I believe, certain. At localities 5, 5A, 6, and 7, on the other hand,
we find ?P seudocylindrodon sp. a somewhat more primitive species than
P. tobeyi. ?P seudocylindrodon sp. is not common at these localities,
however, as P. tobeyi is at the Wood locality.

Whether this difference in faunal composition, which is seen in the
cylindrodont rodents between these six localities, is due to age, ecology,
or some other unforeseen sampling factor is unanswerable at present.
It is interesting to note that Parectypodus lovei ( Sloan, 1966 ) is known
from localities 5, 5A, and 6 from twenty specimens, but that only one
specimen is known from the Wood locality. (Sloan, in his publication,
did not mention the five specimens from 6 and the one from Wood.
These have been recovered from the concentrate since 1966 ) . As more
of the Badwater fauna is described, a clearer understanding of faunal

214

Annals of Carnegie Museum

VOL. 41

differences should result. At present only eleven of the species listed by
Black and Da\vson ( 1966 ) have been reviewed, and additional material
is already known for five of the seven species discussed by Robinson
(1968). This additional material in no way modifies the systematics
Robinson presented but does alter the locality distribution for some taxa.

On the basis of the cylindrodonts alone there is a slight suggestion ^
that the Wood and Rodent locality faunas are either somewhat younger i
than those from 5, 5A, 6, and 7 or that there was an ecological difference
reflected in the depositional environment.

References Cited
Black, C. C. and Mary R. Dawson

1966. Paleontology and geology of the Baclwater Creek area, central Wyoming.
Part 1. History of field work and geological setting. Ann. Carnegie Mns.,

38 (13): 297-307, fig. 1, table 1.

Burke, J. J.

1935. Fossil rodents from the Uinta Eocene Series. Ann. Carnegie Mns., 23
23 (2): 5-12, figs. 1-4.

Peterson, O. A.

1919. Report upon the material discovered in the Upper Eocene of the Uinta
Basin by Earl Douglas in the years 1908-1918 and O. A. Peterson in 1912.
Ann. Carnegie Mns., 12 (2): 40-128, figs. 1-19, pis. 34-42.

Robinson, Peter

1968. The paleontology and geology of the Badwater Creek area, central
Wyoming. Part 4. Late Eocene primates from Badwater, Wyoming, with
a discussion of material from Utah. Ann. Carnegie Mns., 39 (19): 307-
326, figs. 1-24, tables 1-7.

Sloan, Robert E.

1966. Paleontology and geology of the Badwater Creek area, central Wyoming.
Part 2. The Badwater multitiiberciilate. Ann. Carnegie Mns., 38 (14):
309-315, figs. 1-5, table 1.

Tourtelot, H. a.

1957. The geology and vertebrate paleontology of upper Eocene strata in the
northeastern part of the Wind River Basin, Wyoming. Part 1. Geology.
Smithsonian Misc. Coll., 134 (4): 1-27, figs. 1-7, 1 pi.

Wilson, R. W.

1940. Fareumijs remains from the later Eocene of California. Carnegie Inst.
Washington Pnbk, 514: 97-108, 2 pis.

Wood, A. E.

1949. Small mammals from the uppermost Eocene (Dnchesian) near Badwater,
Wyoming. Jonr. Paleont., 23 (5): 556-565, figs. 1-24, tables 1-3.

Back issues of many Annals of Carnegie Museum articles
are available, and a few early complete volumes and parts
are listed at half price. Orders and inquiries should be
addressed to: Publications Secretary, Carnegie Museum,
4400 Forbes Avenue, Pittsburgh, Pa. 15213.

Volume 41

Annals of Carnegie Museum

Article 7

PALEONTOLOGY AND GEOLOGY OF THE
BADWATER CREEK AREA, CENTRAL WYOMING
Part 6. The leporid Mytonolagus ( Mammalia, Lagomorpha )

Mary R. Dawson
Associate Curator of Vertebrate Fossils
Carnegie Museum, Pittsburgh, Pennsylvania

Introduction

One of the best represented small mammals in upper Eocene
localities along Badwater Creek is Mytonolagus. Specimens of this
leporid have been found there at all localities of this level, not only
isolated teeth but also maxillae, partial jaws, and a few postcranial
parts.

The genus Mytonolagus, the only described North American late
Eocene leporid, was first reported from Uintan (Myton) and
Duchesnean (Randlett) aged deposits in the Uinta Basin (Burke,
1934), where the species M. petersoni occurs. Later, Wood (1949)
reported the genus from along Badwater Creek and named the new
species M. wyomingensis on the basis of his Wyoming material. Thir-
teen isolated teeth from one locality (Wood Locality of Black and
Dawson, 1966: fig. 1; this and other localities shown there) constituted
Wood’s sample. A later collection from a different upper Eocene local-
ity along Badwater Creek (Gazin, 1956:8), probably Locality 5, 6, or
7, yielded a ro.axilla with P^-M^ that was referred to M. wyomingensis.
Other reports of Mytonolagus since Burke’s original description are
RusselFs (1965:8-9) reference of material from Swift Current Creek,
Saskatchewan, to the Uinta Basin species, M. petersoni, and Black’s
mention (1967:62) of the genus in the Shoddy Springs (Montana)
local fauna.

New material from Wood Locality adds to the record of M. wyom-
ingensis from its type locality. Even more abundant, however, are
specimens of Mytonolagus from Localities 5 and 6, located to the west
of Wood Locality. These specimens differ from those from the type

Issued September 25, 1970

215

^THSb7v7^

UUT9 1970 )

216

Annals of Carnegie Museum

VOL. 41

locality in a number of characters, most of which indicate a more
primitive level of development than in M. wyomingensis. Locality 7
also has yielded remains of Mytonolagus, although less abundant, in
a level of development comparable to that at localities 5 and 6. Speci-
mens from localities 5, 6, and 7 are all referred to here as Mytonolagus
near M. petersoni. Finally, Rodent Locality has produced specimens
referable to M. wyomingensis, although these show a few differences,
possibly advances, from Wood Locality specimens.

The following abbreviations have been used: ACM, Amherst Col-
lege Museum; CM, Carnegie Museum; NMC, National Museum of
Canada; ROM, Royal Ontario Museum; UCM, University of Colorado
Museum; USNM, United States National Museum.

ACKNOWLEDGEMENTS: This study is part of a project supported by
National Science Foundation grants GR-1266, GB-4089, and GB-7901.
Comparative studies of Asian lagomorphs were made possible by a
grant from the Marsh Fund, National Academy of Sciences, and a
Smithsonian Foreign Currency Grant. Some of the field work costs
and all the publication costs have been defrayed by a grant from the
Gulf Oil Corporation to the Section of Vertebrate Fossils. Illustrations
were prepared by A. Lee Ambrose.

Systematics

Mytonolagus wyomingensis
Figures 1-12, Table 1

localities: Wood (type) Locality (63 teeth in CM; 13 teeth in ACM) and
Rodent Locality (25 teeth in CM).

EMENDED DIAGNOSIS: Species of Mytonolagus with buccal root on showing
some sign of separation into two parts. Hypostria on P^-M^ more persistent than
in M. petersoni. On M^'^ limit of enamel deposition reached buccally before tooth
is worn out; buccal pattern eliminated at relatively earlier wear stage than in M.
petersoni. Trigonid of P3 narrower transversely than talonid, has anterolingual
groove in early wear; following wear trigonid and talonid united medially between
buccal and lingual folds, and finally lingual fold worn away. On P4-M3 postero-
lophids present in early wear; lingual groove between trigonid and talonid persists
nearly as far as buccal groove.

DESCRIPTION: Only isolated teeth are known, but they provide a more
complete sample and show a greater range of wear stages than in the
sample available to Wood.

No unworn or little-worn specimens of are represented. The worn
pattern (fig. 1) shows three lobes transversely oriented. The central
lobe is largest and protrudes anteriorly. The groove between central

1969

Badwater Area: The Leporid Mytonolagus

217

TABLE 1

MEASUREMENTS

IN MILLIMETERS

OF Mytonolagus wyomingensis

Number

Observed Range

Mean

P2 a-p

2

1.15-1.22

1.19

width

1.49

1.49

P3 a-p

4

1.70-1.90

1.80

width

2.44-3.46

2.88

P4 a-p

3

1.83-2.03

1.95

width

2.41-3.05

2.69

a-p

1

1.83

1.83

width

2.85

2.85

M2 a-p

2

1.36-1.70

1.53

width

2.17-2.37

2.27

a-p

3

0.81-0.95

0.86

width

1.36-1.73

1.56

Ps a-p

4

1.76-2.03

1.88

width trigonid

1.22-1.42

1.35

width talonid

1.63-1.83

1.72

a-p

1

1.42

1.42

width trigonid

1.46

1.46

width talonid

1.19

1.19

^Rodent Locality; all others from Wood Locality.

lobe and the small buccal lobe persists less far than that between
central and lingual lobes. Following greater wear (fig. 2) this buccal
lobe loses its separation from the central lobe, with a two-lobed tooth
resulting. The single root bends posteriorly at its distal end.

There is considerable variation among specimens of from Wood
Locality, apparently both age and individual, in development of the
anterior loph. In all relatively little-worn teeth, however, the anterior
loph does not reach the buccal wall of the tooth, and two valleys, one
buccal and one lingual to the central lobe, thus open onto the anterior
face of the tooth (fig. 3). The central lobe does not contact the an-
terior loph. Wood’s type specimen of M. wyomingensis, ACM 10022, is
a tooth, broken anterobuccally, that he described as P^. On this tooth
it appears that the central lobe contacts the anterior loph. This condi-
tion is not duplicated in unbroken specimens of P^ from the type
locality, but can be found on P^. It seems likely that Wood’s type speci-
men is actually P^. Following wear, P^ becomes widened transversely
(fig. 4), and in the most-worn known specimens has all pattern elimi-
nated except the crescentic valley lingual to the central lobe. There
is one buccal root, usually grooved medially and variable in showing
one or two pulp cavities at its distal end.

218

Annals of Carnegie Museum

VOL. 41

Only two specimens of are known from Rodent Locality, but
these show some advance over those from the type locality in having a
central lobe that bends more sharply anterobuccad, and two distinct
pulp cavities in the buccal root.

On the buccal root is grooved medially, and has two pulp cavities.
The pattern in a moderately worn specimen (fig. 5) shows a shallow
hypostria, central lobe that contacts the well-developed anteroloph,
and posterior (Wood’s metacone), or posterior and anterior (Wood’s
paracone), buccal cuspules. As with P^, the one known P^ from Rodent
Locality is advanced beyond those from the type locality in the more
pronounced anterobuccal swing of the central lobe and in the greater
separation of the buccal root into two parts.

Figs. 1-6. Mytonolagus wyomingensis, occlusal views of teeth, x 10. fig. 1: CM
21613, left P2. fig. 2: CM 15233, left P2. fig. 3: CM 15219, right P3. fig. 4: CM
21610, right P^ fig. 5: CM 21614, right P^ fig. 6: CM 15439, left M^.

1969

Badwater Area: The Leporid Mytonolagus

219

On little-worn (fig. 6) the central lobe has a more pronounced
posterobuccal protrusion than on (fig. 7). Following wear, the
pattern becomes reduced to crescent and hypostria, and finally the
buccal part of the hypostria forms an isolated lake. Deposition of the
enamel is such that in well-worn specimens the tooth is past the enamel
limit on the buccal side.

The three known specimens of from Wood Locality show that
the tooth is reduced and compressed anteroposterioiiy. The least-worn
specimens (fig. 8) show the tooth to be composed primarily of the
trigon. A deep buccal fold is present within the trigon and a very small
posterior protrusion represents the talon. With wear, all pattern
except a lingual groove, comparable to the hypostria on M^‘^, is present
between trigon and talon. In the most worn stage known (fig. 9), the
tooth is a cylinder, enamel is absent posterobuccally, and the only
trace of pattern is a small enamel-lined lake, the remnant of the
buccal end of the hypostria.

No unworn specimens of P3 are known, but in the least- worn speci-
mens available, trigonid and talonid are separate columns; the trigonid
has an anterolingual fold, and the talonid is wider than the trigonid
(fig. 10). Following greater wear the fold is lost, the trigonid becomes
relatively wider, though still narrower than the talonid, and the two
columns become joined by a medial bridge, resulting in lingual and
buccal folds. Finally, the lingual fold loses its connection with the
lingual wall and forms an isolated lake, which is worn away follow-
ing still greater wear.

Well developed posterolophids occur on P4-M2 in early stages of
wear. The lingual groove between trigonid and talonid persists nearly
as far down the tooth as does the buccal fold. None of the specimens
shows a lingual bridge between the columns, although this might form
in a very well-worn tooth. The talonid is slightly narrower trans-
versely than the trigonid and both taper buccally.

The only known M3 is from Rodent Locality. This tooth has separate
trigonid and talonid, with the posterolophid indicated by a posterior
expansion of the tooth and a posterobuccal groove.

Only two specimens of deciduous teeth are known. DP^ (fig. 11)
has three transversely oriented lobes, which join one another postero-
buccally. The buccal lobe is very small and the lingual lobe forms most
of the posterior wall. DP^ of M. wyomingensis (fig. 12) is less worn
than that of M. petersoni, CM 11932, having the hypostria still con-

220

Annals of Carnegie Museum

VOL. 41

nected to the lingual side of the tooth. The anterior arm of the central
lobe joins the anterior lobe bucally at this stage of wear, whereas in
the more worn dP^ of M. petersoni, anterior and central lobes are still
independent bucally.

Figso7-12, Mytonolagus wyomingensis, occlusal views of teeth, x 10. fig. 7: CM
15221, right M2, fig. 8: CM 16012, right M3, fig. 9: CM 16013, left M3, fig. 10:
CM 15224, right Ps. fig 11: CM 21654, right dP4. fig. 12: CM 21612, left dP4.

DISCUSSION: In his description and discussion of Mytonolagus wyom-
ingensis, Wood (1949:561-565) considered some of this species’ charac-
ters to be more primitive and some more advanced than in M. peter-
soni, but he concluded that M. wyomingensis is ''a form close to, and
perhaps even more primitive than, the Uinta Mytonolagus petersonir
The characters considered primitive by Wood, which were taken from

1969

Badwater Area: The Leporid Mytonolagus

221

ACM 10022, interpreted by him as but here as were position of
the metacone, and the poorly developed hypocone. The now-known
specimens of both P^ and P^ show that these teeth in M. wyomingensis
are actually more advanced than in M. petersoni. On P^ and P^ the
hypostria, separting "protocone” and “hypocone” parts of the lingual
lobe, is more persistent than in M. petersoni, and P^ has better devel-
oped anterior buccal cuspules (paracone and metacone of Wood). It
is not clear whether Wood considered the formation of two buccal
roots on the premolars as a primitive or an advanced feature of
M. wyomingensis, but it appears to be advanced. In the most prinimtive
known late Eocene lagomorph, the Asian Lushilagus lohensis, both
P® and P have a single buccal root (Li, 1965:30). This condition
occurs also in Mytonolagus petersoni. The subdivision of the buccal
root into two parts, which begins to occur in M. wyomingensis, is a
stage of advance beyond M. petersoni and leads toward the condition
of two buccal roots on P^""^ such as occurs in the early Oligocene
leporids Palaeolagus temnodon and Megalagus hrachyodon. In these
forms on P^ the anterior buccal root is fused to the large lingual root,
and the two buccal roots of P^ are free. Loss of buccal roots on pre-
molars and molars that occurred in the more progressive lines of
leporids during the Oligocene was accompanied by increased hypos-
odonty and atrophy of more buccal parts of the upper cheek teeth. In
the molars also M. wyomingensis shows advance beyond M. petersoni,
especially in longer persistence of the hypostria and earlier elimination
of buccal parts of the molar pattern. The leporid from Badwater Lo-
calities 5, 6, and 7, described below, seems to be a good intermediate
between M. wyomingensis and the more primitive M. petersoni.

Mytonolagus near M. petersoni
Figures 13-27, Tables 2, 3

localities: 5 (70 teeth, 3 maxillae, 2 jaw fragments in CM), 5 A (52 teeth,
1 jaw fragment in CM; 21 teeth in UCM), 6 (37 teeth, 6 maxillae, 1 jaw in
CM), 7 (11 teeth in CM); 5, 6, or 7, USNM 21090, left maxilla with P3M2.

DESCRIPTION: This leporid is represented by more complete upper
and lower jaw material than is M. petersoni from the Uinta Basin. The
parts than can be compared show agreement between the two forms
in most features. The zygoma protrudes laterally about in line with
the anterior margin of P^, which is slightly more posterior than in
Palaeolagus temnodon. There are two mental foramina on the jaw

222

Annals of Carnegie Museum

VOL. 41

(fig. 13), the more posterior of which is in line with the talonid of
P4, as in M. petersoni. The more anterior foramen is relatively far for-
ward below the diastema, in an area not preserved in known speci-
mens of M. petersoni. The anterior mental foramen is more posteriorly
situated on the diastema in P. temnodon. The anterior border of the
masseteric fossa reaches to below Ms, with a knob and fairly well-
developed ridge forming the anteroventral border of the fossa. On the
medial surface of the jaw the swelling over the shaft of the incisor
terminates posteriorly in a line below the anterior of M2, about as
in M. petersoni.

Figs. 13-14. Mytonolagus near M. petersoni. fig. 13: CM 21615, lateral view of
left jaw, X 2/i fig. 14: CM 19742, occlusal view of left P^-M^, x 10.

1969

Badwater Area: The Leporid Mytonolagus

223

Considering the proportions of the cheek teeth to one another (figs.
14, 15, 16) in both M. petersoni and Mytonolagus near M. petersoni,
is slightly smaller than Mb with the reduction of especially
marked posterobuccally. Palaeolagus temnodon has a relatively more
reduced M^. In the lower jaw M1.2 are subequal in M. petersoni and
Mytonolagus near M. petersoni, or M2 may be slightly larger. Both
these molars are larger than P4. In P. temnodon, P4-M0 are more nearly
equal in size.

15

Figs. 15-16. Mytonolagus near M. petersoni, occlusal views of teeth in jaws, x 10.
fig. 15: CM 14403, right P4-M2. fig. 16: CM 19770, right Mi-s.

The groove of the upper incisor is slightly medial of center, and the
medial portion protrudes farther anteriorly than the lateral.

The first upper cheek tooth, P- (fig. 17), consists of three lobes,
transversely oriented as in M. wyomingensis. Even in the most worn
specimen available, three lobes are still present as they are in the very
worn type specimen of M. petersoni, and it appears that the two-lobed

224

Annals of Carnegie Museum

VOL. 41

shape found in worn of M. ivyomingensis would not occur here. The
roots are fused into a wider posterior and narrower anterior part or
may separate into two parts at the distal end.

As in M. petersoni and M. wyomingensis, exhibits variation in
transverse width of the anterior loph. The lingual groove is barely
indicated, even in relatively unworn teeth, and thus is similar to that
in M. petersoni, but less well developed than in M. wyomingensis. The
single buccal root has one pulp cavity, as in M. petersoni.

The lingual hypostria is only rudimentary on P^ (fig. 18), shallow
even in relatively little-worn teeth. The central lobe remains distinct
from the anterior lobe until well worn. The tooth is less well developed
anterobuccally than in M. wyomingensis, has less buccad development
of the anterior lobe, and less development of the antero-buccal cusp.
The buccal root is grooved but has only a single pulp cavity.

The central lobe of and posterobuccal part of that tooth are better
developed than on M^. In relatively little worn examples of both
teeth, the central lobe has a contact with the lingual shelf and a
posterobuccal protiusion which is better developed and extends
farther on than on M^. The hypostria curves strongly posterobuccad
at this stage of wear. After greater wear the central lobe loses its con-
nection to the lingual shelf and the buccal tip of the hypostria forms an
isolated lake. These teeth differ from those of M. petersoni in two
principal features. First, the hypostria extends closer to the enamel
limit than in M. petersoni. Second, the central lobe in this Badwater
form shows some advance: on MMt is more concave buccally and its
posterior connection is oriented more posterobuccally than directly
posteriorly, as in M. petersoni; on the central lobe has a stronger
posterior protrusion than in M. petersoni. Even in very well worn
teeth, traces of pattern are preserved buccal to the central lobe. These
traces do not occur in M. wyomingensis in comparable stages of wear.
Enamel persists on the buccal side of the molars even in very well
worn stages in contrast to M. wyomingensis, in which the buccal
enamel is absent in well worn teeth.

It appears that considerable reduction has already affected (figs.
19-21), but the tooth is distinctly more robust and less reduced than
in M. wyomingensis. When unworn, the trigon shows a strong V-shape
with a rounded remnant of the central lobe in the buccal opening
of the V. Posterior to this is the protrusion of the talon, developed
lingually but not bucally. In the most worn specimen available there

1969

Badwater'Akea: The Leporid Mytonolagus

225

is a hypostria, lingually between trigon and talon, and an isolated
lake. The only known, very worn of M. petersoni, CM 11937, is
somewhat larger than most specimens of this tooth of Mytonolagus
near M. petersoni.

Figs* 17-21* Mytonolagus near M. petersoni, occlusal views of teeth, x 10. fig. 17 :
CM 18245, right P^. fig. 18: CM 21653 left P4, fig. 19: UCM 24748, left M^.
fig. 20: UCM 24781, left M^. fig. 21: CM 21620, right M^.

The lower incisor, as in M. wyomingensis and M. petersoni, is rela-
tively narrower transversely in its ventral part than is that of Palae-
olagus temnodon.

In the least- worn specimen of P3 (fig. 22) a small posterolophid is
still indicated by posterolingual fold, and the trigonid has a well-

226

Annai.s of Carnegie Museum

VOL. 41

developed anterolingual notch. The presence or absence of a connec-
tion between trigonid and talonid is variable, some specimens having
a connection near the lingual side, while in others, seemingly in com-
parable stages of wear, trigonid and talonid remain separate. The
trigonid and talonid are nearly equal in width ( fig. 23 ) , as opposed to a
narrower trigonid in M. wyomingensis. There is never a deep lingual
fold as in M. wyomingensis. In very well worn teeth (fig. 24) the pat-
tern consists of a buccal fold and an isolated lake lingual to the fold,
a remnant of the more lingual part of the trigonid-talonid fold. The
only know Pss of M. petersoni are very worn, retaining only the buccal
fold between trigonid and talonid.

The pattern on Pt-M^ generally resembles that in M. wyomingensis,
with a well-developed posterolophid present in early wear. These teeth
are hypsodont but M2 shows toward its distal end a constriction of the
rooted portion. The lingual groove between trigonid and talonid is
more persistent relative to the buccal groove than in M. petersoni.
Following wear the talonid extends an anterolingual process toward
the trigonid, but in known specimens the columns do not join as they
do in the very worn type of M. petersoni.

Little-worn M3 has a two-cusped trigonid, two cusped talonid, and
well-developed, cuspate, posterolophid. With wear the cusps are
flattened, and trigonid and talonid join lingually. Next, trigonid and
talonid join buccally also, first with an enamel-ringed lake between
them that is finally worn away. In the most worn Ms available, the
posteroexternal groove between talonid and posterolophid is also worn
away. The tooth is relatively larger, especially anteroposteriorly, than
that of M. wyomingensis.

All deciduous teeth except dP^ are represented. Little worn dP^ (fig.
35) has a shallow hypostria and a long anterior lobe, which extends
to the buccal wall anterior to the central lobe. The hypostria is re-
moved with wear. DP^ has a hypostria that forms an isolated lake
following wear. Its crescentric valley lingual to the central lobe con-
nects to the buccal wall in earlier stages of wear but loses the connec-
tion later. An isolated lake is buccal to the central lobe. Out of the four
known dPss, none are unworn. The least worn shows that the tooth had
three lobes anteroposteriorly and a protruding posterolophid. The two
anterior lobes join first (fig. 26). In greatly worn specimens the three
lobes are joined. As on dPs, the talonid of dP4 has a protruding postero-

1969

Badwater Area: The Leporid Mytonolagus

227

lophid. Following wear the two lobes of dP4 (fig. 27) become joined
lingually by a bridge.

Figs. 22-27, Mytonolagus near M. petersoni, occlusal views of teeth, x 10. fig. 22:
CM 15440, left Ps. fig. 23: CM 16810, left Pa. fig. 24: CM 15186, right Ps. fig. 25:
CM 15175, right dPA fig. 26: CM 21607, left dPs. fig. 27: CM 21652, right dP4.

DISCUSSION: The leporid from localities 5, 6, and 7 differs from M.
petersoni mainly in its relatively longer hypostria on the upper molars,
longer lingual fold between trigonid and talonid in the lower molars,
and in evidence of a more crescentic shape of the central lobe on
It and M. petersoni differ from M. ivyomingensis in having a more
persistent buccal lobe on P^, having on P^"^ a single root with little or
no subdivision into two parts and less development of the hypostria,
retaining into later wear stages the folds buccal to the central lobe and
buccal enamel on and having less reduced M%. The differences

228

Annals of Carnegie Museum

VOL. 41

between M. petersoni and Mytonolagus from localities 5, 6, and 7
suggest a more advanced level of development in the latter. These two
forms seem more similar to one another than either is to M. wyom-
ingensis, but differences of the Badwater form from the Uinta Basin
species suggest that the former should be referred to as Mytonolagus
near M. petersoni.

TABLE 2

MEASUREMENTS IN MILLIMETERS OF TEETH IN JAWS OF MytOnolugUS
NEAR M. petersoni

CM 19742

P3 a-p 1.97

width 2.58

P4 a-p 2.10

width 3.05

Ml a-p 2.17

width

M2 a-p 1.97

width

P4 a-p

width trigonid
width talonid
Ml a-p

width trigonid
width talonid
M2 a-p

width trigonid
width talonid
Ms a-p

width trigonid
width talonid

CM 14403 CM 21615 CM 19770

1.83

3.63

1.83

3.36

1.76

3.02

2.17

2.10

1.56

2.37

2.30

2.17

2.14

1.66

1.86

2.27

2.30

2.37

1.70

1.83

1.63

1.36

1.15

TABLE 3

MEASUREMENTS

IN MILLIMETERS OF TEETH

OF Mytonolagus near

M. petersoni

Number

Observed Range

Mean

P2 a-p

12

0.88-1.36

1.09

width

1.18-1.76

1.54

P3> a-p

5

1.59-1.76

1.72

width

2.37-2.71

2.54

P4 a-p

9

1.73-1.93

1.84

width

2.03-3.52

2.63

M^ a-p

12

1.63-2.00

1.85

width

2.10-3.86

2.80

M2 a-p

6

1.56-1.90

1.74

width

2.14-3.25

2.64

M® a-p

6

0.95-1.36

1.23

width

1.36-2.10

1.73

Ps a-p

6

1.70-2.30

1.97

width trigonid

1.15-1.56

1.37

width talonid

1.12-1.66

1.35

Ms a-p

9

1.36-1.73

1.57

width trigonid

1.25-1.56

1.37

width talonid

1.02-1.36

1.19

1969

Badwater Area: The Leporid Mytonolagus

229

Mytonolagus and other late Eocene lagomorphs

Within the upper Eocene localities along Badwater Creek there are
now known two and possibly three levels of development of Myton-
olagus. The more westerly localities (5, 6, and 7) have produced the
most primitive, a form slightly more advanced than M. petersoni.
The more advanced M. wyomingensis occurs in the more easterly
Wood Locality, and it, or a slightly more advanced form, is repre-
sented also at Rodent Locality. So far as known no M. wyomingensis
is present in localities 5, 6, and 7, and no M. petersoni in Wood or
Rodent localities. These forms, along with the Uinta Basin leporid M.
petersoni, represent gradual stages of development within the late
Eocene.

In addition to the late Eocene occurrences of Mytonolagus in the
Uinta Basin and Badwater Creek, the genus has been reported from
the Shoddy Springs local fauna of southwestern Montana (Black,
1967:62) and from Swift Current Creek, Saskatchewan (Russell,
1965:8-9). The Shoddy Springs Mytonolagus is represented by only a
few teeth, which are smaller than those of M. petersoni but more
similar to it in occlusal pattern than to M. wyomingensis. The Swift
Current Creek lagomorph was referred by Russell (1965:8-9) to M.
petersoni. Comparisons with the Uinta Basin specimens show, however,
that in the Saskatchewan form the hypostria on the upper molar of
NMC 8653 extends closer to the enamel limit than in M. petersoni.
It differs from both M. petersoni and M. wyomingensis in having on
P4-M2 of ROM 1679 the trignonid shorter anteroposteriorly relative
to the talonid, which is more rounded lingually and reaches less far
linguad, and in having a relatively longer lingual groove between
trigonid and talonid. No little-worn P3S are known for M. petersoni,
but comparisons with P3 of Mytonolagus near M. petersoni and M.
wyomingensis show that on P3 of the Saskatchewan form both the
anterolingual fold and the more posterior lingual fold are situated more
posteriorly than are the comparable folds in the specimens of Mytono-
lagus. Perhaps the more posterior lingual fold in the Saskatchewan
forms marks the posterolophid of P3. The Saskatchewan form is not
yet known enough to serve as the basis for speculation. However, it is
probably not referable to M. petersoni, and possibly not even to
Mytonolagus. The characters of its lower cheek teeth are somewhat
similar to those of some relatively primitive lagomorphs like Desmato-

230

Annals of Carnegie Museum

VOL. 41

lagus vetustus of the Mongolian Oligocene. In Shamolagus medius of
the Mongolian late Eocene the folds on P,; seem similar in position to
those on P;^ of the Saskatchewan form. The Saskatchewan lagomorph
suggests greater diversity in late Eocene North American lagomorphs
than has been suspected, and although the evidence is not yet sufficient,
it seems to have a few characters found in some of the Asian late Eocene
lagomorphs and possibly in some later Tertiary North American lago-
morphs that have been referred to as "'DesnuitolagusT The late Eocene
lagomorphs of China and Mongolia (Burke, 1941; Li, 1965) form at
least two groups, that of Lushilagtis and its relative Shamolagus, and
that of Gobiolagus. Of these, the former group is the more generalized
and probably led toward Desmatolagus of the Oligocene. Gobiolagus is
specialized in its strongly developed unilateral hypsodonty and rela-
tively small talonids, and seems to have persisted into the Oligocene.
Mytonolagus cannot be tied closely to either of these groups and
appears to be more advanced toward later North American leporids
in the proportions of its cheek teeth. The Saskatchewan form is more
advanced in the proportions of one tooth to another than are mem-
bers of the Asian Lushilagus-Shamolagus-Desmatolagus group. How-
ever, the Saskatchewan form shows a little greater resemblance to that
group than does Mytonolagus.

References Cited

Black, C, C.

1967. Middle and late Eocene mammal communities: A major discrepancy.
Science, 156 (3771): 62-64.

Black, C. C., and M. R. Dawson

196^ Paleontology and geology of the Badwater Creek area, central Wyoming.
Part 1. History of field work and geological setting. Ann. Carnegie Mus.,
38 (13): 297-307.

Burke, J. J.

1934. Miftonolap;us, a new leporine genus from the Uinta Eocene series in Utah.
Ann. Carnegie Mus., 23 (9): 399-420.

1941. New fossil Leporidae from Mongolia. Amer. Mus. Novitates, 1117: 1-23.
Gazin, C. L.

1956. The geology and vertebrate paleontology of upper Eocene strata in the
northeastern part of the Wind River basin, Wyoming Part 2. The mam-
malian fauna of the Badwater area. Smithsonian Miscell. Coll., 131 (8):
1-35

Li, Chuan-Kuei

1965. [Eocene leporide of North China.] Vert. Palasiatica, 9 (1): 23-29 (Chi-
nese), 29-36 (English).

Russell, Loris S.

1965. Tertiary mammals of Saskatchewan. Part 1. The Eocene fauna Royal
Ontario Mus., Univ. Toronto, Life Sciences, Contrib. 67: 1-33.

Wood, A. E.

1949. Small mammals from the uppermost Eocene (Duchesnian) from Bad-
water, Wyoming. Jour Paleont., 23 (5): 556-565.

r '

h

li

Volume 41 Annals of Carnegie Museum Article 8

I FOSSIL FISHES FROM SOUTHWESTERN PENNSYLVANIA
! Part I; Fishes from the Duquesne Limestones

I (Conemaugh, Pennsylvanian)

i

' Richard Lund^

Research Associate, Section of Vertebrate Fossils,

Carnegie Museum, Pittsburgh, Pennsylvania

I Introduction

I The excellent exposures of Conemaugh, Monongahela, and Dunkard
' group rocks in and around Pittsburgh, Allegheny County, Pennsylva-
fi nia, have been the subject of a few sporadic searches for vertebrate
I fossils. Beds containing fish scales and teeth were reported as early as
I 1876 (Stevenson, 1876: 78, 326). P. E. Raymond (1910) noted the
I presence of bone in the Ames limestone and underclay and in the
I Birmingham shale. E. C. Case ( 1915) described some tetrapod remains
I from Pitcairn, Allegheny County, and summarized earlier vertebrate
I fossil finds from southwestern Pennsylvania. Moran (1952) and Romer
I (1952) discussed all vertebrate fossil finds previous to 1952 from the
Pennsylvanian of Allegheny County as well as from the Dunkard of
I West Virginia and Ohio. Kent (1967) and Schweinfurth (1967) repeat-
I edly mentioned the occurrence of fish remains in the fresh-water
I limestones of southwestern Pennsylvania. Fossil fishes below the Cone-
I maugh are known from two famous localities: Linton, Ohio (New-
I berry, 1873, 1875), and Cannelton, Pennsylvania (Westoll, 1944),

I within 70 miles of Pittsburgh, but are seriously in need of restudy.
acknowledgments: I am grateful to Carla Hartz and Lee Ambrose,
who have ably and enthusiastically illustrated this paper; W. S. Steph-
ens, District Engineer, State Highway Department, for access to this
locality; the Penn Central Company for permission to prospect along
1 their right-of-way; Drs. Craig C. Black and Mary Dawson, Carnegie
Museum; Dr. Keith S. Thomson, Yale Peabody Museum; and Dr. Bobb

^Department of Earth and Planetary Sciences, University
Pa. 15213; Scientific Contribution No. DEPS-69-210.

Issued September 25, 1970

231

of Pittsburgh, Pittsburgh,

232

Annals of Carnegie Museum

VOL. 41

Schaeffer, American Museum of Natural History, whom I thank for
reading and criticizing the manuscript. This work has been supported
by a grant from the Pittsburgh Foundation to the Section of Vertebrate
Fossils, Carnegie Museum.

Geologic Setting

The Ames marine horizon of the central Conemaugh group marks a
change in depositional history in this area from alternating near-shore
marine and fresh-water sediments of the lower Conemaugh, to the
predominantly, if not exclusively, fresh-water beds of the upper Cone-
maugh, Monongahela, and Dunkard groups. Fossils useful for correla-
tion above the Ames limestone are plants, ostracods, and insect wings.
Zonation of the Pennsylvanian-Permian section of western Pennsylvania
is based chiefly upon plant successions (Darrah, 1934), coupled with
classic applied stratigraphic principles. Little use has been made of the
invertebrates, and no use of the vertebrates.

The presence of cyclical deposits throughout the Pennsylvanian and
Permian of southwestern Pennsylvania indicates repetition of varied
ecological and depositional conditions. A typical cyclothem in the Pitts-
burgh area consists of several elements. A clay zone, representing soil
formation, contains scattered plant material. A carbonate zone which
may contain a few teeth and denticles of Orthacanthus, and an occa-
sional microsaur tooth, may occur in the clay. These carbonates thus
seem to represent local flooding or standing-water conditions. Fresh-
water limestone may be developed within the clay zone, usually locally.
Overlying the clay is either coal or a trace of coal, with plant material
only. The coal is usually succeeded by a jet-black, carbonaceous and
pyritic shale, either massive or very thin-bedded, and of fresh-water
origin. Massive shales may have occasional scattered fish material,
while thin, evenly-bedded shales may contain many fish scales and
occasional articulated vertebrates, as well as pieces of plant stem and
leaf. This thin black-shale zone grades upward into gray, flaky shale,
irregularly bedded, containing plants, insect wings, and conchostrac-
ans, seemingly representative of very shallow-water, mudflat facies.
Above this zone lies a thick zone of plant-containing, light-gray silts
and interfingering sands, upon which another clay zone is developed.

Fish remains occur, broadly speaking, in two different sedimentary
types representing strikingly different ecologies. The most readily
observable remains occur in the jet-black, carbonaceous siltstones and
shales that frequently accompany coals or coal zones. Bones, scales.

1969

Fossil Fishes From Southwestern Pennsylvania

233

teeth, and spines of fish, filled with pyrite, are accompanied by copro-
litic material and, much more rarely, by articulated specimens or
portions of specimens. Fish scales occur irregularly throughout the
black shales and may be locally concentrated. Tetrapod remains are
rare to nonexistent. Fresh- water limestones may contain local concen-
trations of fish and tetrapod remains, under somewhat specialized sets
of depositional circumstances that are not well understood. Fossil
vertebrate specimens, furthermore, are almost never found articulated
in these limestones. While it is not unusual to find small numbers of
fish teeth and scales in fresh-water limestones, only rarely are sufficient
concentrations of material found to be of any paleontologic use.

Stratigraphically useful invertebrate fossils are rare. But many hori-
zons contain fossil-fish material, and it may be possible to correlate
units on the basis of their vertebrate faunas when the latter are known
well enough. This paper is a preliminary report on a few fresh-water
fossil fish from one of these horizons, the Duquesne limestone.

The following abbreviations are used throughout: AMNH, American
Museum of Natural History; CM, Carnegie Museum; DL, Duquesne
limestone; CMSL, Carnegie Museum Stratigraphic Locality, Section
of Invertebrate Fossils; FMNH, Field Museum of Natural History.

DUQUESNE CLAY AND LIMESTONES

The Duquesne clay occurs between the top of the Grafton shales
and the base of the Duquesne coal (Johnson, 1929). In places where
the Grafton shales are cut out, the Duquesne clay lies directly on the
Ames limestone (Johnson, 1929: 5). The clay averages about eight
feet in thickness throughout Allegheny County and contains, in its
upper portion, occasional well-preserved terrestrial plant fossils stand-
ing erect (CMSL 366).

A carbonate zone occurs near the center of this clay, generally poorly
developed and unfossiliferous, but persistent. North of the bluffs of the
Allegheny and Ohio rivers, the carbonate zone appears in three layers
of fresh-water limestone. They are dark gray to black, weathering to
tan on a fresh break, and have a peculiar, foul, swamp-like odor when
crushed or in contact with acid (White, 1878). Broad slabs exposed
along McKnight Road, Ross Township, show many unrecognizable
fragments of bone on their surfaces.

To my knowledge, the Duquesne limestones occur in only one area
south of the Allegheny River, namely the west end of Mount Washing-

234

Annals of Carnegie Museum

VOL. 41

River, opposite Herr’s Island; 3. Riverview Park, 100 feet below Allegheny
Observatory,

ton, within the city of Pittsburgh. In this area, limestones 2, 4, and 6
outcrop for almost a mile along the bluffs of the Monongahela River,
and again for a short distance west of the south end of the Fort Pitt
Tunnels (fig. 1). Limestone 2a, 3, and 5, however, seem to occur only
for about 200 yards of outcrop east of the north portal of the Fort Pitt
Tunnels. They represent small bays that extended southward from
extensive marly lakes. The Duquesne limestones from this area have
produced an extremely rich and varied vertebrate microfauna, part of
which is described here, as well as many fragments of larger vertebrates.

1969

Fossil Fishes From Southwestern Pennsylvania

235

The invertebrate fauna is limited to ostracods, conchostracans, and
Spirorbis. The outcrop of these limestones at the north portal of the
Fort Pitt Tunnels is in a tall, slowly sagging, slump block.

Significant quantities of varied vertebrate material occur within and
between the two top limestone units and in the lowest limestone. The
sedimentary character of these units is worthy of more study than
presently can be devoted to them. The upper unit (DL 2) grades into
gray calcareous siltstone above as well at the east end of the outcrop.
Scattered fish scales and teeth are found near the base of this siltstone.
The limestone is distinctly conglomeratic, composed of dark and light
lime pebbles and plates of siltstone in a calcareous and silt ground
mass. The particle sizes are clearly larger toward the base of the unit,
and most of the bone occurs in the lower inch of this bed. Bone occurs
in all components of the rock unit, but is most common in the calcareous
silt ground mass. There are associations of bony elements that indicate
parts of individuals, but no articulated components. Non-carbonized
but finely comminuted plant material is present throughout the unit.
Fragmented tetrapod remains are common.

A very thin shale break provides an irregular plane of separation
between DL 2 and DL 2a. This break, between ^3 2 inch and % inch
thick, has yielded scales, teeth, a partially articulated fish, and parts of
a reptile.

Duquesne limestone 2a varies as greatly in composition as it does in
thickness. In the thinnest parts, it is composed of thin plates of very
light and very dark shale with calcareous cement. This unfossiliferous
sediment may grade into either of two other types within a few inches.
The more common type is a dark-brown limestone interrupted by
frequent and seemingly random seams of calcareous silt. Small fossils
are more prevalent in the silty fractions than in the limestones, while
macrofossils are found in the limestones or crossing both silty and lime-
stone portions. The remaining lithologic type is a massive, very dense,
very black limestone with finely disseminated crystalline calcite. Macro-
fossils are lacking but fragmented microvertebrate fossils are common.
Tetrapod material is almost totally absent in the lower of the two lime-
stone units, and is limited to microsaur bones.

Duquesne limestone 6, averaging almost one foot in thickness, is a
massive unit high in clay, locally high in iron compounds, and contain-
ing some gypsum crystals as well as an irregular mesh of calcite tubules,
possibly from burrow fillings. Orthacanthus and Xenacanthus teeth,

236

Annals of Carnegie Museum

VOL. 41

2a

1 r

1969

Fossil Fishes From Southwestern Pennsylvania

237

paleoniscoid scales, and small tetrapod bones are irregularly distributed
throughout and may be locally concentrated near the base of the unit.
Lungfish remains are present but rare.

Transport of bone seems to have been significant in DL 2, but mini-
mal in DL 2a and DL 6. Reworking of the unconsolidated substrate by
burrowing organisms may have been the most important factor in
disarticulating specimens in these units.

Deformation of bones prior to lithification can be seen in certain
specimens from all fossiliferous limestone units. Small microsaurian
vertebrae are generally skewed to one side. A large portion of an
Orthacanthus spine was bent to the point of shattering individual
longitudinal bony spicules. A large Sagenodus cf. S. periprion operculum
was crushed from above, fragmented, and spread out to occupy almost
twice its intact area. These examples, however, are exceptions, for most
bones, large and small, show no trace of the effects of fossilization.

The distinctly fresh-water nature of the deposits is indicated by the
presence of Orthacanthus sp., Sagenodus periprion, and another dip-
noan, as well as by the several tetrapod remains.

The assignment of the name Duquesne limestone to these beds
(Johnson, 1929: 60) recognizes their proximity to the overlying
Duquesne coal and/or black shale. This stratigraphic proximity is de-
ceiving, for the limestones occur in the midst of a clay representing a
soil zone, a gap in deposition of unknown duration. At least one layer
of erect, fossil land plants occurs between the definitive Duquesne coal
horizon and the underlying limestone-coal sequence (fig. 2), The six-
inch coal overlying the limestone sequence at the Fort Pitt Tunnels is
not the Duquesne coal of Johnson (1929), which is represented by a
trace at the base of the Birmingham shales at this place.

The elapsed time between the faunas of the Duquesne limestone and
< — ——————

Fig. 2. Outcrop of Duquesne limestone 100 feet east of north portal of Fort Pitt
Tunnels prior to removal of layer 2a. At the top is a zone containing upright plant
fruiting bodies. 1. Coal zone. Six inches of well-developed coal, underlain by soft
gray siltstone. 2. Soft, laminated gray siltstone grades downward and laterally
into a graded limestone conglomerate. Fossil vertebrates chiefiy near base. A thin
shale break at bottom contains occasional bones. 2a. Limestone interfingered with
shale. Variable in composition, very rich in bones. Underlain by clay-shale. 3. Thin,
silicified layer, with occasional pleuracanth teeth, underlain by clay. 4. Thick,
limy clay, with some fine silicate sand. Very few fossil vertebrates. Ostracods,
Spirorbis, pleuracanth present. Underlain by clay. 5. Variably developed limy
clay. No known fossils. Slight clay break underneath. 6. Thick, persistent layer.
Limy clay mixed with areas richer in lime. Fossils in scattered concentrations near
bottom of layer. Significant quantity of tetrapod material as well as a good repre-
sentation of fish.

238

Annals of Carnegie Museum

VOL. 41

the overlying Duquesne shale may be significant. The time elapsed,
however, in deposition of each of the two principal fossil-bearing lime-
stones (DL 2 and DL 2a), is geologically very short. Neither bed shows
any interruption or change in the nature of deposition vertically. Parti-
cle-size distribution suggests that DL 2 is a single, graded bed. Larger
pieces of tetrapod bone are abraded and occur at the base of the bed, i
suggesting transportation of these chunks along with the plates of |
hardened silt. The small bones of fish and microsaurs are, for the most ;
part, unworn and uninjured in spite of their great fragility. The remains j
of these aquatic vertebrates, found in the lime-silt ground mass, prob-
ably represent an in situ fauna. The peculiarities of this bed can be j
compared to a dried swamp bed flooded with silt-laden water. i

DL 2a contained a major portion of the skull of one individual of ;
Sagenodus cf. S. periprion (CM 19127). The bones were disarticulated *
but closely associated, and in some cases extended from the top to the
bottom of the stratum. The bed itself is quite limited in area and ;
probably represents the product of a single season of deposition of .
calcium carbonate in a shallow and algal-filled body of water .

Certain vertebrate elements in the fauna of the Duquesne limestones
lend themselves readily to identification; others, such as the isolated
paleoniscoid scales, cannot presently be treated.

Systematics I

Class Acanthodii j

Family Acanthodidae !

Genus Acanthodes Agassiz I

Acanthodes cf. A. marshi Eastman, 1902 |

Certain fin spines found in the Pennsylvanian and Permian of south
western Pennsylvania have been referred by Romer ( 1952:49) to Acan-
thodes cf. A. marshi Eastman, Neither spines nor scales are ornamented. ,
Although spines and scales have been found in association ( CM 21549,
Fishpot, Monongahela group, upper Pennsylvanian), not enough is
known of this form to render a valid comparison with Acanthodes
marshi Eastman.

Class Elasmobranchii I

Family Xenacanthidae
Genus Orthacanthus Agassiz, 1843

Onhacanthus Agassiz, 1837-1843.

Diplodus Agassiz, 1843. '

1969

Fossil Fishes From Southwestern Pennsylvania

239

Xenacanthus Newberry, 1856.

Orthacanthus Fritsch, 1889.

Xenacanthus Olson, 1946.

Xenacanthus Hotton, 1952.

Xenacanthus Romer, 1952.

Acondylacanthus Romer, 1952.

Orthacanthus Gluckman, 1964.

diagnosis: See von Fritsch, 1889: 100.

Orthacanthus cornpressus (Newberry)

Figure 3

Diplodus cornpressus Newberry, 1875.

Xenacanthus cornpressus Hotton, 1952.

description: Several hundred teeth, ranging in height from /2 mm. to 3 cm.,
plus parts of two fin spines, have been found within the limited volume (5 cubic
feet) of rock dissolved from DL 2a. These teeth, while varying in size and pro-
portions of the two main cusps, are perfectly uniform in characteristics. Each tooth
has two large, blade-like serrated cusps, and a tiny central, generally unserrated,
cusp upon a prominent rounded base. The basal surface of the tooth is cupped
for attachment to the jaw, and a prominent boss is present lingual to the cusps on
the denticulated surface. The two major cusps are almost of equal size in teeth
that seem to be from near the symphysis, while on other teeth there is progressive
proportional difference in the height of the two cusps. This feature, together with
strong angling of the smaller cus, seems to be indicative of teeth located far back
along the jaws. The spines (fig. 3) rae rounded in cross-section, thick, composed
of longitudinal bony fibers reflected externally as fine striae, and bear a double
row of small, sharply downcurved denticles proximally along a cupped rear margin.
No complete spines have been found.

DISCUSSION: Hotton (1952) suggested that only one genus {Xenacan-
thus) of North American pleuracanth shark is known. Gluckman ( 1964:
211-212) reproduced the superb illustrations of von Fritsch (1889).
These illustrations (von Fritsch, 1889, figs. 173-175) clearly indicate
that the teeth named Diplodus (Agassiz, 1843) are associated with the
spine named Orthacanthus (Agassiz, 1843, pi. 45, figs. 7-9), while the
spine named Pleuracanthus (Agassiz, 1843, pi. 45, figs. 4-6), which is
a senior synonym for Xenacanthus (Byrich, 1848), is associated with
distinctly different teeth built upon the same general plan as those ol
Orthacanthus ( Diplodus ) .

Almost all reported North American pleuracanth teeth are of the
Diplodus type, corresponding to the genus Arthacanthus. North Amer-
ican spines of the genus Xenacanthus have been uniformly assigned

240

Annals of Carnegie Museum

VOL. 41

Figs. 3-4. Fig. 3. Orthacanthus sp. Fragment of spine. CM 19131. Unit is 1 cm.
Fig. 4. Xenacanthus sp. spine. CM 19133. Unit is 1 mm.

(incorrectly) to the genus Orthacanthus (Newberry, 1873, pi. 40, fig, 4,
for example ) . True Orthacanthus spines have been assigned incorrectly
to the “genus” Acondylacanthus (Romer, 1952: 51). North American
Xenacanthus teeth are rarely reported and tend to be overlooked
prineipally because of their small size.

In DL 2a, two robust spines of Orthacanthus accompany many teeth
of comparable size. Orthacanthus teeth are also known from DL 2, 3,
and 6.

Genus Xenacanthus (Beyrich, 1848)
Pleuracanthus Agassiz, 1843.

Orthacanthus Newberry, 1873.

1969

Fossil Fishes From Southwestern Pennsylvania

241

Xenacanthus sp.

Figure 4

One almost complete spine, plus a fragment of a second spine of this
genus, is known from DL 2a. Teeth associated with spines of this size
and delicacy have not been found in the same bed. However, Xenacan-
thus teeth have been retrieved from DL 6 and are quite common in the
Dunkard series of southwestern Pennsylvania.

The spine (fig. 4) is quite thin, evenly tapered to the tip, and bears
a row of depressed, recurved denticles on each side nearly throughout
its length. The spine is nearly oval in cross section distally where the
denticle rows are borne midlaterally, and is more highly arched proxi-
mally along its presumed dorsal surface than its somewhat flattened
ventral surface. The denticle rows are ventrolaterally placed near the
proximal end.

The teeth of Duquesne limestone Xenacanthus are diminutive, and
differ strikingly from those of Orthacanthus. The two major cusps are
subequal in size, approximately twice the length of the central cusp,
thin, and circular in cross-section throughout their length. The two
major cusps each bear six to eight thin vertical ridges along their distal
halves from Orthacanthus in lacking a prominent labial process. The
lingual “button’' is quite variable in development. Little angular devia-
tion of the cusps from each other is present. There is no difficulty in
distinguishing the teeth of Xenacanthus from teeth in the same size
range of Orthacanthus. The spine of Xenacanthus arcuatus ( Newberry,
1873, pi. 40, fig. 4) from Linton, Ohio, Upper Allegheny series, differs
from the Duquesne specimen in having a concave ventral face. No
differences in spine shape are known between Conemaugh and Dunk-
ard Xenacanthus ( to be reported on at a later date ) . It is not presently
possible to make species assignments based on the small samples
available.

Order Hybodontida
Family Hybodontidae
Genus Hybodus Agassiz
Hybodus allegheniensis, new species
Figure 5

Hybodont shark: Romer, 1952, p. 49.

Agassizodus variabilis Romer, 1952 : 49.

definition: A small, fresh- water hybodont shark, distinguishable from others by
the following characters: the teeth bear a robust central crown flanked by 2 to 3

242

Annals of Carnegie Museum

VOL. 41

Fig. 5. Hybodus sp. A. Spine fragment, lateral view. B. Spine fragment, posterior
view. CM 19134. C. Tooth, lingual view. CM 19135. D. Denticle dorsal view.
E. Denticle, lateral view. CM 19136. Units = 1 mm.

progressively smaller crowns on each side; all crowns bear vertical, radiating cren-
ulations, and a central ridge unites the cusps; tooth bases are produced inward
into a strong roughened ridge. Denticles are large with a smooth, sHghtly convex
internal surface, and an external portion consisting of two major lateral cusps and
one to two smaller, central cusps, projecting backwards. The dorsal-fin spine is

1969

Fossil Fishes From Southwestern Pennsylvania

243

greatly compressed, slightly curved and strongly tapering, ornamented with a few
strong, narrow, longitudinal ridges laterally. Denticles along the posterior edge are
directed basally, but alternately point to left and right, as do the teeth of a saw.

HORIZON AND LOCALITY: Duqiicsne limestone 2a, Conemaugh, Pennsylvanian,
200 feet east of north portal, Fort Pitt Tunnels; Pittsburgh, Allegheny County.

TYPE SPECIMEN; CM 19134, distal portion of a dorsal spine, Duquesne lime-
stone 2a.

REFERRED SPECIMENS: CM 19136, dermal denticles; CM 19135, teeth, from the
Duquesne Limestone.

A considerably broader spine fragment, equally severely compressed
but showing more rapid tapering and no left-right alternation of denticle
direction, comes from the Mason shale at Sample, Allegheny County.
This spine (CM 21565) is regarded as a pectoral spine of the same
"species” of hybodont shark.

DISCUSSION: The teeth originally referred to Agassizodus variabilis
by Romer ( 1952 ) were, in fact, not related to the marine Pennsylvanian
fish bearing the original name. The teeth and spine fragments men-
tioned by Romer are those of a fresh-water fish, and lack the broad,
rounded cusps and the characteristic bradyodont structure of typical
Agassizodus teeth.

Order Dipnoi

Family Ctenodontidae Woodward, 1891
Genus Sagenodus Owen, 1867
Sagenodiis cf. S. periprion
Figures 6-9

Sagenodus cf. S. periprion have previously been described from the
uppermost Conemaugh and the Greene of this region by Romer (1952:
52 ) . Remains from DL 2a consist of two opercula, two tabulars, a par-
ietal, parasphenoid, numerous cheek and snout bones, ribs, and right
pterygoid and vomerine tooth plates, all pertaining to one individual
(CM 19127). This specimen has an associated vomerine tooth (fig. 6),
the first such tooth reported for this genus.

The forms of the cranial bones are apparently unchanged from the
upper Conemaugh through the Dunkard and are relatively undis-
tinguishable from the Texas Permian Sagenodus periprion (Romer,
1952: 54).

The vomerine tooth ( fig. 6 ) is remarkable in its similarity to the form
illustrated in figure 9 from the Duquesne horizon. Differences between
these vomerine teeth are almost exclusively a reflection of the differ-
ences in size of the forms. The similarity between these teeth is in

244

Annals of Carnegie Museum

VOL. 41

Fig. 6. Sagenodus cf. S. peripron. CM 19127. Left vomerine tooth. A. Anterior
view. B. Ventral view. Unit is 1 mm.

1969

Fossil Fishes From Southwestern Pennsylvania

245

Fig. 7. Sagenodus cf. S. periprion. A. CM 19127. Right pterygoid, ventral view.
Unit is 1 cm. B. CM 19469. Right pterygoid, ventral view. Unit is 1 mm.

strong contrast to the differences between the tooth plates of large
(fig. 7A) and small (figs. 7B, 8) forms. The tiny vomerine teeth pro-

246

Annals of Carnegie Museum

VOL. 41

vided the first indication that a series of very tiny Sagenodus were
preserved in the Duquesne limestone. This was confirmed only after
a number of intermediate-size specimens were found in Monongahela-
and Dunkard-group rocks. The latter specimens are currently under
study.

EARLY GROWTH: The Duqucsnc limestone Sagenodus provide an in-
sight into the ontogeny of the jaws of this form. The most immature
specimen, CM 19465, bears four rows of teeth on a flat, simple, bony
plate. The four rows contain five, five, five, and three, teeth respective-
ly, beginning with the most anterior row, which measures 86 mm. in
length. Each tooth in each row is almost double the size of the tooth
mesial to it.

In CM 19471, the first row of teeth is missing, but the specimen bears
a ventral ridge of bone that presumably lay mesial to Meckel’s carti-
lage. The tooth rows show some fusion with growth of the individual
teeth, so although there are five, four, and one, cusps respectively, the
innermost cusps have been almost completely incorporated into the
bases of the cusps next lateral.

CM 19467, which shows a well-developed ventral ridge, also bears
a first cusp-row of 1 mm. length. It has four rows, bearing three, two,
two, and two, cusps respectively. All traces of the tiniest mesial teeth
have vanished, while the remaining teeth in individual rows have grown
into hefty, wellrounded cusps united at the base. The most anterior
tooth row maintains a sharply bladed series of cusps. In all three speci-
mens where the first tooth row measures more than 1 mm., or where
the spenial bone is well-developed vertically, there are four tooth rows,
with a maximum of three cusps in each row.

CM 19468 (fig. 7) is the largest of the small Duquesne limestone
Sagenodus, and the only specimen with five tooth rows. Evidence from
Dunkard group specimens indicates that two additional cusps are
added to each row before additional tooth rows are added to the
splenial element.

One right palatal element is known that corresponds in size range
with these splenials ( CM 19469; fig. 7B ) . It is a broad, rounded element
in a poor state of preservation which bears teeth or tooth bases indicat-
ing the presence of six rows. The most anterior row is compressed and
blade-like, while teeth in succeeding rows decrease in compression,
in tooth size, and in basal proximity. The superficial glossy layer of the
teeth seems to extend at least across the anterior two valleys between
tooth ridges. The remarkable thinness of tooth walls, dramatically

1969

Fossil Fishes From Southwestern Pennsylvania

247

A

Fig. 8. Sagenodus cf. S. periprion. CM 19468. Right splenial. A. Dorsal view.
B. Labial view. Unit is 1 mm.

illlustrated in this specimen, corresponds to the condition in several of
the splenials.

248

Annals of Carnegie Museum

VOL. 41

The blade-like anterior ridge of the palate and splenial evidently
were piercing and chopping ridges. The large, rounded, and yet thin-
walled teeth of the posterior ridges seem incapable of masticating
anything firmer than soft plant tissue or soft-bodied invertebrates. A
generally vertical action of the splenial-dental ridges between those of
the palate seems to be the only possible way of masticating.

The vomerine teeth assigned to this group ( fig. 9 ) have been placed
here on the basis of two observations. The first is that they closely
resemble the vomerine tooth of Sagenodus cf. S. periprion (fig, 6), to
which the tooth rows on the jaws also bear some resemblance. The
second is the assignment of vomerines illustrated by Berman ( 1968 ) to
Gnathorhiza, which, if correct, strongly suggests that vomerines such
as that in figure 12 belong with a genus more closely related to
Gnathorhiza.

Vomerine teeth of this size generally lack strong, bony bases. The
bases, where seen, consist of highly spongy bone, spicular in construc-

Fig. 9. Sagenodus cf. S. periprion. CM 19431. Right vomerine tooth. Anterior
view. Unit is .25 mm.

1969 Fossil Fishes From Southwestern Pennsylvania 249

tion and probably endochondral in origin. The teeth themselves vary
only in the degree of development of secondary cusps, which are not
constant in number, proportion, or size. Few cusps, gradually increas-
ing in size away from the midline, plus a slight anterior convexity,
describe this form. Only two features distinguish this vomerine from
that of large Sagenodus: its minuscule size and a lack of evident wear
on the sides of the more mesial cusps.

At the growth stages represented by the small Sagenodus specimens,
there is no evident tooth wear. Addition of tooth rows and cusps with
age is indicated in the Duquesne limestone Sagenodus where numbers
of cusps per row is decreased by the incorporation of the smallest teeth
into the bases of adjacent ones. Growth of individual teeth seems in-
dicated, not only by this observation but by the tendency of larger
splenials to have the teeth united at their bases, producing rows of
cusps.

The small specimens of Sagenodus provides a remarkable insight into
the ontogeny of this form.

Family Lepidosirenidae Bonaparte, 1841
Genus Monongahelab new genus
Type species: Monongahela stenodonta, new species

provisional diagnosis: Lungfish, related to Gnathoriza, and known only from
toothed elements recognizable on the basis of the following criteria: pterygoid
element with four laterally divergent tooth ridges, the anterior three ridges diver-
gent from an apex, the most posterior ridge originating lateral to the apex from
the side of the ridge next anterior; all ridges slightly convex anteriorly; palatal
bone broad and thin, not extending under lateral ends of posterior three ridges,
and with a long horizontal symphysis; mandibular element with three tooth ridges
divergent laterally, the anterior ridge longest, transverse ridge shortest, and orig-
inating posterior to apex of other two ridges; bone vertically developed with a
posterior coronoid enlargement.

Monongahela stenodonta^, new species
Figures 10-15

HORIZON: Duquesne limestone, Conemaugh group, upper Pennsylvanian.

locality: North face of Mount Washington, 200 feet east of the north portal
of the Fort Pitt Tunnels, Allegheny County, Pittsburgh, Pennsylvania.

type: cm 19450, right pterygoid element, from DL 2a.

REFERRED SPECIMENS: Mandibles CM 19412-19419, CM 19434-19442, CM
19445-19448, CM 19459-19461, CM 19473-19476, CM 19478-19484, AMNH

1 Name is taken from the site where the first representatives were found, the bluffs at the mouth
of the Monongahela River, Pittsburgh, Pennsylvania.

^ From the extreme narrowness of the tooth rows.

250

Annals of Carnegie Museum

VOL. 41

4665; pterygoids, CM 19420-19430, CM 19449-19458, AMNH 4664; partial skull
with two pterygoids and mandible, CM 19447; two mandibles and one pterygoid
of a single individual, CM 19485; vomerines, CM 19433, 19487, 19488.

PROVISIONAL diagnosis: Lungfish of genus Monongahela, known chiefly from
jaw elements. Jaw elements with cusps strongly flattened along axes of tooth
ridges; cusps always united into distinct, separate ridges, with cusps becoming
progressively less discrete with age. Pterygoid element with four ridges (a,b,c,d,
anterior to posterior, fig. 12), angle ab plus be usually equalling 90 degrees, angle
cd with a mean of 30 degrees, range of 23-40 degrees, the angle decreasing with
increasing individual age. Ridges a,b,c are all slightly convex anteriorly, and do
not add cusps with increasing size, ridge d almost straight and adding cusps with
increasing size. Lower jaw with three ridges (a,b,c anterior-posterior), angle ab
with a mean of 48.6 degrees, range 25-65 degrees, median of 45 degrees; angle

Fig. 10. Monongahela stenodonta. CM 19413. Right splenial. A. Dorsal view. B.
Lateral view. Unit is 1 mm.

1969

Fossil Fishes From Southwestern Pennsylvania

251

be with a mean of 70.2 degrees, range 100-130 degrees, median of 115 degrees.
Vomerines with over 10 small cusps and blade-like.

geologic range: Upper Freeport limestone, Allegheny series ( = Westphalian
of European section, Desmoinesian of Western North America), Pennsylvanian,
through the lower Washington limestone member of the Washington formation,
Dunkard group, Wolfcampian ( ? ) , Permian ( ? ) .

DISCUSSION: The jaws of small individuals of this species were very
common in DL 2a, and were present in DL 2. One fragmented jaw was
found in DL 6. Preservation was generally excellent; very few frag-
mentary jaws were found. Size variation is between 1 mm. total lower-

Fig. 11. Monongahela stenodonta. CM 19450, Type. Right pterygoid. A. Ventral
view. B. Lateral view. Unit is 1 mm.

252

Annals of Carnegie Museum

VOL. 41

jaw length and 9 mm. total jaw length. One pterygoid (CM 19455)
shows growth and wear indicative of a specimen much older than the
others. Most of the remains seem to represent young specimens, pos-
sibly young-of-the-year.

The pterygoid plates (fig. 11) show an interesting pattern of growth
of the various ridges, which sheds light on changes in feeding during
the life of the fish. The three anterior-cusp ridges are extremely sharp-
edged in the small specimens with indivdual cusps quite discrete
although not isolated from each other. The number of cusps per tooth-
ridge stays constant through growth of the animal until all cusps are
obliterated by wear ( fig. 13 ) . The fourth cusp row consists of one rela-
tivley rounded cusp in the smallest individuals. Cusps are added with
growth until the adult condition of the tooth plate is reached. At this
point, the posterior ridge is of significant size, may contain up to seven
cusps, and addition of cusp ceases. The three anterior ridges of the
pterygoid are discrete in the young and meet at an apical cusp con-
tributed by the central ridge. The posterior ridge is subsidiary to the
ridge next anterior, originating on the side of that ridge (fig. 11).
Growth of the ridges results in early union of the first three ridges at
the apex, and considerably later in the incorporation of the fourth
ridge.

There is little wear in juvenile Monongahela pterygoids. Slight wear
facets are developed only at the tips of the cusps, and extremely sharp
edges are maintained in valleys and all sides. There is no indication of
shearing of upper against lower jaw and no means of crushing food
items. The mode of feeding is almost certainly that of an active predator
with teeth designed to pierce and hold prey.

C

Fig. 12. Occlusal pattern in Monongahela. Solid lines represent pterygoid tooth
ridges, dashed lines represent splenial tooth ridges. Ridges A,B,C, and D (ptery-
goid only) labelled from anterior to posterior.

1969

Fossil Fishes From Southwestern Pennsylvania

253

With the achievement of a subadult size, as evidenced by growth of
the fourth pterygoid ridge, aspects of wear become evident. Cusps
gradually are obliterated by growth, from within and by abrasion, to
form unified ridges. Ridges a and c, (see fig. 12 for key) united at the
apex, are separated from b by a valley (fig. 13) produced by wear of
the mandibular element, and a notch tends to be worn into ridge c
by the mandible, generally lateral to the junction of c and d. The sub-
adult shows a plate with one sharp, arcuate mesial ridge, composed of
ridge a, the mesial part of c, and ridge d, plus two converging ridges
separated from it, ridge b, and the lateral part of ridge c. Upon occa-
sion, a specimen may be found where ridge c is notched mesial to its
juncture with ridge d ( CM 19455 ) .

The change in ridge shape and relationship occurs when the jaw
is relatively large. Therefore it is postulated that the change occurs
at a subadult stage. The change results in a style of feeding better
adapted to cutting and chopping. Both animal and vegetable material
could be obtained and chewed with this mechanism.

The vomerine teeth attributed to this species (fig. 14) are rarely

Fig. 13. Monongahela stenondonta. CM 19455. Right pterygoid. Ventral view.
Unit is 1 mm.

254

Annals of Carnegie Museum

VOL. 41

_ i

/

: : ■ ,

Fig. 14. Monongahela stenodonta. CM 19433. Left vomerine. A. Posterior view.

B. Ventral view. Unit is .5 mm.

found, as they are tiny. They are blade-like, slightly convex anteriorly, I
and grow by addition of successively larger cusps laterally. Newer
vomerine cusps are slightly serrated along their mesial edges. The older I

cusps show wear only on the mesial sides of their tips. As the vomerine i

teeth are not opposed by any known toothed elements, it is inferred
that wear was produced by the action of tooth against prey, which |
suggests prey with a resistant external layer. The ostracods found in |
the same deposits would be within the size-range of prey for most of
the Monongahela specimens represented.

The mandibular, or splenial, elements of Monongahela (fig. 10) do
not differ noticeably from those of the Permian genus Gnathorhiza
(Berman, 1968, fig. 2), except in size. The smallest specimen found, jj

CM 19460, shows that: the anterior ridge a includes the apical cusp, j

ridge b (see Fig. 12 for key) originates from the cusp next posterior |j
to the apex, and ridge c includes only the remainder of the cusps on a |

1969

Fossil Fishes From Southwestern Pennsylvania

255

line from the apical cusp. The separation between ridges is lost very
early.

The range of angles found between ridges of the mandible is much
greater than those of the relatively unvarying pterygoid. Bilateral
variation is strong in CM 19485. The degree of latitude in lower jaw
angles virtually prohibits accurate shearing action of the upper against
the lower jaw and reinforces the belief that the feeding was accom-
plished by chopping and piercing. In the young, cusps served to hold,
pierce, and cut prey, while older individuals lost the need for cusps,
either as a result of change in the predator-prey size relationship or a
change in diet.

The presence of single-bladed splenial elements in populations of
Permian Gnathorhiza has been noted by Berman ( 1968 ) . Single-bladed
splenial elements comprise 19.5 percent of the total number of Mon-
ongahela lower jaws, and comprise a minimum of 16.1 percent of in-
dividuals sampled in this study. The size-range of single-bladed splen-
ials is great. CM 19464 (fig. 14) is the largest splenial of any sort found.

These peculiar elements differ from normal Monongahela splenials
in few ways. Most important, the transverse and posterior tooth blades
are not present. The anterior blade is proportionally equal, in length,
to the same element of a normal splenial. Shearing wear along the
posteromesial half of the blade is evident (fig. 15) in all but the small-
est specimens, as Berman (1968, fig. 2D) has illustrated for Gnathor-
hiza. No other acute differences are known and no palates that match
this lower jaw have been found. Several specimens give some insight
into a developmental peculiarity of high frequency and surprisingly
great duration. CM 19461, a very small single-bladed splenial, displays
one cusp oriented at an angle equivalent to the transverse blade, and
one cusp oriented at an angle equivalent to the posterior blade. These
two cusps are quite small, and are at the posterior end of the large
anterior blade. CM 19464 bears one excedingly small, rounded cusp
on its posterolateral edge. CM 19463 bears a slight posterolateral ridge.
CM 19444 and CM 19476 each display two rudimentary cusps of the
transverse ridge, and CM 19481 displays a rudimentary transverse
ridge.

The single-bladed splenial is evidently produced through a reduction
in the number of teeth in the posterior two ridges coupled with a lack
of growth or exceedingly limited growth of the teeth that were present.

256

Annals of Carnegie Museum

VOL. 41

Fig. 15. Monongaliela stenodonta. CM 19464, Right splenial, single bladed. Lin-
gual view. Unit is 1 mm.

1969

Fossil Fishes From Southwestern Pennsylvania

257

The anterior ridge is normal in shape, size, and position, but the
splenial lacks a prominent “heel” to stop jaw travel by contact with the
“heel” of the upper jaw. An abnormal, though functional, shearing
action can thus take place between the anterior ridges of upper and
lower jaws.

No palate has been found that would correspond to a single-bladed
splenial in form or function, either in the Duquesne limestone or in the
Permian Lenders formation ( Berman, 1968 ) . Berman has inferred that
this type of splenial is simply an abberant lower jaw. When a specific
change in dentittion of this magnitude, reflecting a relatively major
change in jaw function and almost certainly in diet, can be demon-
strated in high proportion over a considerable span of time and space,
it is obviously an important biologic and ecologic factor. As the single-
bladed splenial seems to arise through a developmental abnormality,
it seems unwise to grant it separate taxonomic status at present.

Conclusions

Monongahela is clearly related to the Permian Gmthorhiza (fig. 16),
differing chiefly in the interrelationships of the pterygoid tooth rows.
Derivation of the Gnathorhiza palate may have been accomplished by
a mesial shift in the point of origin of the most posterior tooth row,
plus a spreading of the space between rows. Detailed treatment of
this problem will appear in a later paper on some lower Dunkard
Monongahela.

The Monogahela-Gnathorhiza group shows definite similarities to
the very young Sagenodus described above. The blade-like anterior
tooth row, and the tendency towards a blade-like second row in
Sagenodus is quite reminiscent of Monongahela. The presence of only
one more tooth row in Sagenodus jaws than in Monongahela jaws of
the same size is an important addition to the similarities. At present,
however, there are obvious barriers to a clear relationship between
Sagenodus and Monongahela. Similarities exist only at the earliest
ontogenetic stages of Sagenodus. As Sagenodus grows, it adds cusps
and tooth rows, until at the adult condition it reveals basically different
chewing motions, dietary adaptations, and ontogenetic patterns from
Monongahela. Body form and skull bones are also quite different.
Nevertheless, it is possible to conceive of an ancestral condition of
Sagenodus from which both Monongahela and Sagenodus might be
derived.

Contrary to the assertions of Berman (1968), there is a striking

258

Annals of Carnegie Museum

VOL. 41

Figs. 16-17. Fig. 16 Gnathorhiza dikeloda. FMNH UF 45. Right pterygoin. Ven-
tral view, Unit is 1 mm. Fig. 17. Protopterus annectans. Right pterygoid. Ventral
view. Unit is 1 mm.

similarity between the Gnathorhiza-Monongahela group and the recent
lepidosirenians in cranial morphology ( Olson, 1951 ) , in habit ( Romer

1969

Fossil Fishes From Southwestern Pennsylvania

259

and Olson, 1954; Sterba, 1963), and in tooth-plate structure, develop-
ment, and function. Two differences are conspicious.

The palate of Protopterus (fig. 17) has but three tooth blades, con-
trasted with four in Paleozoic forms. Further, the proportionally longest
tooth blade is the most posterior rather than the most anterior blade. It
has been shown, however, that the most posterior pterygoid tooth row
of Monongahela is very short in small individuals and that it is late in
developing. Further, suppression of splenial tooth rows is not unusual
in M. stenodonta and Gnathorhiza, and is not an improbable mechanism
by which the Protopterus pattern may have been derived. It may be
pointed out that two small palates of M. stenodonta (CM 19420 and
I942I) apparently lack a posterior-tooth row. Differences in propor-
tional length of tooth rows between the Paleozoic and the Recent are
to be expected.

The pterygoid bone of Protopterus is a vertically expanded element
with little flattening under the tooth rows. The pterygoid bones of
the Paleozoic genera are thin, flattened horizontal plates, somewhat
extended under the tooth rows. The Recent fish appear to have
strengthened the bony support for the palate considerably, a difference
that is not anticipated in the Pennsylvanian fossils.

The diet of the Recent forms varies between small vertebrates and
shelled invertebrates like snails and small bivalves (Sterba, 1963, p.
846). The tooth-blades, admirably suited to such a diet, are them-
selves substantially identical in form to those of the Paleozoic genera,
and widely divergent from the crushing dentition of the ceratodont line
and Epieeratodus.

Summary

The non-actinopterygian portion of the Duquesne limestone fish
fauna consists of an acanthodian (microphagous?), two xenacanthid
sharks ( large and small predator ) , a hybodont shark ( crushing-inverte-
brate feeder), a large dipnoan (scavenger-predator-herbivore) and a
small dipnoan (crustacean feeder-predator). The actinopterygian por-
tion of the fauna contains small, abundant secondary and tertiary
consumers only.

Monongahela is a dipnoan approximating the ancestors of Gnathor-
hiza, and closely approximating the ancestor of the modern Lepidosi-
renidae. Although an ancestor for Monongahela is not known, the
young stages of Sagenodus seem to be morphologically close to the
necessary conditions.

260

Annals of Carnegie Museum

VOL. 41

References Cited

Agassiz, L.

1837-1843. Recherches sur les Poissons Fossiles. Vol. III.

Berman, D. S.

1968. Lungfish from the Lenders Formation (lower Permian, Texas) and the
Gnathorhiza-hepidoskenid ancestry questioned. Jour. Paleont., 42 (3):
827-835.

Beyrich Ej

1848. Monatsb. Verh. k Preuss. Akad. Wiss. 1848,

Case, E. C.

1915. The Permo-Carboniferous red beds of North America and their verte-
brate fauna. Carnegie Inst. Washington Publ. 207: 1-176.

Darrah, W. C.

1934. Stephanian in America (abstr. ). Proc. Geol. Soc. America 1933: 451.
Eastman, C. R.

1902. Some Carboniferous Cestraciont and Acanthodian shark. Bull. Mus.
Comp. Zool. Harvard, 39: 35-99.

Fritsch, a, von

1889. Fauna der Gaskohle. Bd. Ill, hft. IV, Fr. Rivnae, Prague.

Gluckman, L. G.

1964. Elasmobranchii. In Obruchev, D. V. (ed.), Osnovy Paleontologii, (Fund-
amentals of paleontology. Agnathes, Pieces.), vol. 11, 522 p., 197 figs.,
14 pis, -f- figs.

Hotton, N.

1952, Jaws and teeth of American Xenacanth sharks. Jour. Paleont., 26 (3):
489-500.

Johnson, M. E.

1929, Geology of the Pittsburgh Quadrangle. Topographic and Geologic Atlas
of Penna., no. 27, Penna. Geol. Surv. Ser. 4: 1-236.

Kent, B. H.

1967. Geologic map of the Hackett Quadrangle, Washington Co., Penna. U.S.
Geol. Surv. Map GQ-630.

Moran, W. E.

1952. Location and stratigraphy of known occurrences of fossil tetrapods in the
upper Pennsylvanian and Permian of Pennsylvania, West Virginia and
Ohio. Ann. Carnegie Mus., 33 (1): 1-44.

Newberry, J. S,

1873. Report of the Geological Survey of Ohio. Vol. 1, pt. 2. Paleontology.

Description of fossil fishes: 247-358, pis. 24-40.

1875. Report of the Geological Survey of Ohio. Vol. II, pt. 2. Paleontology.
Description of fossil fishes: 1-64, pis. 54-59.

Olson, E. C.

1946. Fresh and brackish water vertegrate-bearing deposits of the Pennsyl-
vanian of Illinois. Jour. Geol., 54: 281-305.

1951. Fauna of the upper Vale and Choza. 3. Lungfish of the Vale. 4. The skull
of Gnathorhiza dikeloda Olson. Fieldiana Geol., 10: 104-124.

Raymond, P, E.

1910. A preliminary list of the fauna of the Allegheny and Conemaugh Series
of western Pennsylvania. Ann. Gamegie Mus., 7: 144-158.

Romer, a. S.

1952. Late Pennsylvanian and early Permian vertebrates of the Pittsburgh- West
Virginia region. Ann. Carnegie Mus., 33 (2): 47-112.

Romer, A, S., and E. C. Olson

1954. Aestivation in a Permian lungfish. Breviora, Mus. Comp. Zook, 30: 1-6.
SCHWEINFURTH, S. P.

1967, Geologic map of the California Quadrangle, Washington and Fayette
Cos., Penna. U.S. Geol, Survey Map GQ-648.

1969

Fossil Fishes From Southwestern Pennsylvania

261

Sterba, G.

1963. Freshwater fishes of the world. Viking Press, New York: 1-878.
Stevenson, J. J.

1876. Report of progress in the Greene and Washington districts of the bitum-
inous coal-fields of western Pennsylvania. Second Geol. Surv. Penna., K:
1-419.

Stromer, E.

1938. Die Wustenfisch Ceratodus Ag. 1833 und seine mezo- und kanozoischen
Verwandter. Neues Jahrb. Min. Geol. Pal Beil 80 (13): 248-2632.
Westoll, T. S.

1944. The Haplolepidae, a new family of late Carboniferous Bony Fishes. Bull.
Amer. Mus. Nat. Hist. 83 ( 1 ) : 1-22.

White, J. C.

1878. Report of progress in the Beaver River district ... of western Pennsyl-
vania. Second Geol. Surv. Penna., Vol. Q: 1-337.

Back issues of many Annals of Carnegie Museum articles
are available, and a few early complete volumes and parts
are listed at half price. Orders and inquiries should be
addressed to: Publications Secretary, Carnegie Museum,
4400 Forbes Avenue, Pittsburgh, Pa. 15213.

Volume 41

Annals of Carnegie Museum

Article 9

THE INFLUENCE OF PREHISTORIC MAN ON
THE DISTRIBUTION OF THE BOX TURTLE

Kraig Adler

Department of Biology, University of Notre Dame
Notre Dame, Indiana

!

I Introduction

The Eastern Box Turtle {Terrapene Carolina) has a wide distribution
throughout eastern North America from Maine to Florida, and west to
Texas. It is a common and conspicuous element of the fauna, but its
I distribution at the northern edge of its range has been difficult to deter-

I mine, in part because this turtle is so often kept as a pet and then released

1 beyond its natural range. This report is a re-evaluation of distributional
I records for this species from Wisconsin to Maine based on the large body
I of archeological data now available from this area. Accurate determina-
1 tion of the present range is prerequisite for a discussion of the various
: factors, including man, that have altered the distribution of this turtle,

j ACKNOWLEDGMENTS: I thank Neil D. Richmond (Carnegie Museum),
J Charles M. Bogert and George W. Foley (American Museum of Natural

I History), Ernest E. Williams (Harvard University), Albert A. Barden

(University of Maine), and Sherman A. Minton (Indiana University) for
supplying distributional data from their respective states. I am particularly
indebted to John E. Guilday (Carnegie Museum), William A. Ritchie
p (New York State Museum), and J. V. Wright (National Museum of
I Canada) for sending archeological specimens, dates, and information.
Ij Paul S. Martin (University of Arizona), James B. Griffin (University of
I Michigan), George R. Zug (Smithsonian Institution), and Guilday have

« kindly reviewed early drafts of the MS. Charles F. Wray (West Rush,

: New York) and Peter P. Pratt (State University College, Oswego, New

I York) sent me data on early historic sites in western New York. Deposi-
tories referred to are the Museum of Comparative Zoology (MCZ) and
The University of Michigan (UMMZ).

Issued September 25, 1970

264

Annals of Carnegie Museum

VOL. 41

This research was completed during tenure of a National Institutes of
Health Traineeship (NIH 2 T1 GM 989-07).

Fossil Records

Fossil material of this turtle is extensive, but is in large part only from
two areas (Florida and Texas), both in the southern part of the range
(Auffenberg, 1958, 1967; Milstead, 1969). Very few records are avail-
able from the northern part of the range that could give a clue to the
distribution of this species during glacial and interglacial periods. Turtles
like T, Carolina are known from the upper Pliocene and Pleistocene of
Florida (Auffenberg, 1958, 1967), and the Pleistocene of New Mexico,
Texas, Kansas, Arkansas, and Missouri (Milstead, 1967). Although the
box turtle still exists in most of Florida, the records from southern Texas,
western Kansas, and eastern New Mexico, which are beyond the present
range, suggest a more southern and western distribution limit during the
Pleistocene. During this period or earlier, the Texas populations were
doubtless in contact with others to the south in Mexico that now have
become morphologically distinct and disjunct (T. Carolina mexicana,
T. c. ijucatana, T. coahiiila; Milstead, 1967). Terrapene is well known
from archeological sites in Yucatan dating from the Pleistocene (Lange-
bartel, 1953).

Fossil T. Carolina are reported from only five localities in northeastern
North America: two each in Pennsylvania (Fig. 1) and Virginia, and one
in Maryland. Both Virginia localities are of late Wisconsin age: Natural
Chimneys (Augusta Co.; Guilday, 1962b) and Early’s Pits (Wythe Co.;
Guilday, 1962a). The Maiyland locality is Pleistocene in age (Talbot Co.;
Milstead, 1965). The oldest Pennsylvania record is from Port Kennedy
Cave (Montgomery Co.; Milstead, 1965), which is of Yarmouth age.
Guilday et al. ( 1966) suggest a minimum date of about 1770 ± 200 years
B. C. for the Bootlegger Sink Terrapene (York Co., Pennsylvania). The
Bootlegger fauna contains both temperate and boreal elements and is
undoubtedly the result of a long period of deposition. The box turtle,
judging from its low fluoride content, was probably a late addition to the
deposit. In contrast, the extensive excavations at New Paris Cave no. 4
(Bedford Co., Pennsylvania), dating 9345 ±: 1000 B. C., failed to disclose
this species (Guilday et ah, 1964), and it did not turn up at Frankstown
or Cumberland caves, both Pleistocene assemblages (N. D. Richmond,
pers. comm.). These records, although few in number, establish the
presence of Terrapene as far north as southern Pennsylvania during at
least some periods of the Pleistocene.

1969

Prehistoric Man's Influence on Box Turtle

265

Fig. 1. Distribution of the Eastern Box Turtle {Terrapene Carolina) in northeastern North America. Stippling indicates the
modern distribution; single-specimen extralimital records are indicated by X. Archeological and fossil records are indicated by
circles. Cireles bisected by a vertical line represent multicomponent sites with occupations of two widely different ages.
Numbers identifying sites refer to Table 1.

266

Annals of Carnegie Museum

VOL. 41

Archeological Records

Fossil and archeological records are listed separately, since prehistoric
man occasionally transported turtles, including box turtles, and this
means of dispersal must be taken into account. Archeological records for
T. Carolina in northeastern North America (Adler, 1968; table I, fig. 1)
represent material from refuse pits or burial mounds. I have seen most
of this material, and have particularly examined critical records.

TABLE 1

Archeological Records for Terrapene Carolina
IN Ontario, New York, and Pennsylvania

SITE

ONTARIO

COUNTY

DATES

1 Sidey-Mackay

Simcoe

1400- A. D.; 1591 A. D.

2 Lake Medad

Halton

1600 A. D.

3 East Gore

Northumberland

date undetermined

4 East Sugar Island Northumberland

NEW YORK

2000-1500 B. G.; 300 A. D.

5 Kleis

Erie

1620-1640 A. D.

6 Oakfield

Genesee

*1509 ± 150 A. D.

7 Dutch Hollow

Livingston

1590-1615 A. D.

8 Sackett

Ontario

*1140 ± 150 A. D.

9 Morrow

Ontario

*630 ± 100 B. G,; *563 ± 250 B. C.

10 Bristol Hills

Ontario

date undetermined

11 Kipp Island* 2-3

Seneca

*310 ± 100 A. D.; *630 ± 100 A. D.

12 Lamoka Lake

Schuyler

*3433 ± 250 B. G.; *2575 ± 400 B, C,

13 Frontenac Island

Gayuga

*2980 It 260 B. G.; *2013 ± 80 B. G.;
*1723 It 250 B.G.

14 Felix

Onondaga

500 A. D.

15 Garoga

Fulton

1550 A. D. (*1330 it 100 A. D.)

16 Zimmerman
Rockshelter

Greene

2200 B. G.; 1400 A. D.

17 Dutchess Quarry

Orange

2500-1500 B. G.; 1000-1400 A. D.

18 Vosberg

Dutchess

*2780 ± 80 B. G.

19 Stony Brook 2

PENNSYLVANIA

Suffolk

*974 ± 250 B. G.; *944 ± 250 B. C.

20 Ibaugh

Lancaster

1400- A. D.

21 Eschelman

Lancaster

1400-1625 A. D.

22 Sheep

Rockshelter

Huntingdon

date undetermined

23 Varner

Greene

1000-1400 A. D.

24 Quaker State

Venango

100 B. G.-900 A. D.

C-14 dates are preceded by an asterisk; others are estimates based on associated archeological
material. In some cases, these dates may be accurate only for the Terrapene-hearing component.
Dates are from Ritchie (1965) and Ritchie, Guilday, and Wright (pers. comm.). Numbers
correspond to sites in Fig. 1; details for Michigan and Ohio records are in Adler (1968).

Box turtles (or “land tortoises, Cistudo” as they often appear in some
of the archeological and older zoologieal literature ) have been found at
archeologieal sites in the northeast dating over the past five thousand
years. The oldest record is from Lamoka Lake in central New York (site
dated 3433 to 2575 B. C. ). From then until about 1700 A. D. (Kleis,

1969

Prehistoric Man’s Influence on Box Turtle

267

Oakfield, Dutch Hollow, and Garoga sites ) they were locally common in
western New York, southeastern New York, and southern Ontario, areas
largely beyond the present range of the species. In Pennsylvania all
archeological records fall within, or only a few miles beyond, the present
range (Fig. 1). Farther west, there are a few archeological specimens
from beyond the present range in Ohio and Michigan, but in each case
they were used as bowls and thus could have been introduced into these
areas via inter-tribal trade. The two extralimital records from Michigan
are each represented by a single fragment. These reeords and some
material from Wisconsin originally misidentified as T. Carolina are dis-
cussed elsewhere (Adler, 1968).

Box turtle artifacts are also known from four sites in southern Ontario
(Table 1). In all cases, however, the specimens have been made into
bowls or rattles, so the possibility exists that these may have been intro-
duced via trade. However, one of these localities (Lake Medad; Bleak-
ney, 1958) is very near the place where the single modern Canadian
specimen was found ( see below ) .

The archeological records indicate the widespread occurrence of the
box turtle in northeastern North America since about 3000 B. C., and
suggest a reduction of its range in New York and possibly Ontario sinee
1700 A. D. The possible causes of this reduction will be discussed in
more detail below.

Present Distribution

Conant (1958) gives the most recent map for the range of the box
turtle. After careful review of the literature and new records, and after
consideration of each of these in view of the abundant archeological in-
formation now available (Adler, 1968; table 1), I have drawn up a new
map (Fig. 1). There are significant differences between this and Conant’s
map, and these differences are briefly discussed below.

NEW ENGLAND

The box turtle ranges through the maritime provinces of New England
to and including the southern tip of Maine. There are no records for
Vermont. The only two records for New Hampshire are old (Huse,
1901 ), and both in the maritime part of the state. There is a single, old
record from Maine (York Co., A. A. Barden, pers. comm.). Surveys in
other parts of the state, including offshore islands, have not found it.
However, box turtles are known from two islands off the Massachusetts
coast: Martha’s Vineyard (personal record) and Nantucket Island (MCZ

268

Annals of Carnegie Museum

VOL. 41

46174 ) . Both islands were apparently separated from the adjacent main-
land about 2000 B. C., according to botanical and zoological evidence
(Ogden, 1958; Starrett, 1958). These two islands are located 5 and 20
miles, respectively, from the Cape Cod mainland. If these turtles did
not swim these distances or were not transported by humans, this species
has been present on these offshore islands (and presumably on the ad-
jacent mainland) for nearly 4000 years.

NEW YORK, PENNSYLVANIA AND ONTARIO

Within historic times, the box turtle has never been listed as common
in New York state except in the southeastern corner. Mearns ( 1898 ) said
it was common along the lower Hudson Biver, from the Fishkill Moun-
tains and Storm King Mountain ( Putnam Co. ) southwards. There are
numerous records from Long Island (Latham, 1916) and Staten Island
(Davis, 1912). Farther up the Hudson, there are two old records from
near Albany (Eights, 1835; Baur in Taylor, 1895). It has apparently
disappeared in that area since, as Bishop (1923) recorded no further
specimens. Tlie only historic record for western New York is a single
specimen from Allegany State Park (Cattaraugus Co.; Eaton, 1945),
although Bishop and Alexander (1927) did not find it in their extensive
survey of the Park. Werner ( 1959 ) reported a specimen from Ironsides
Island (Jefferson Co.), an island in the headwaters of the St. Lawrence
River.

In Pennsylvania, T. corolina is not known from the northern half of
the state, except along Lake Erie and near the Ohio Border. Since this
turtle is not common at high elevations in mountainous regions, it seems
unlikely that the western New York and Ontario populations of prehistor-
ic times were derived from eastern New York, although the “Water Level
Route” across central New York could have been used as an invasion
route. It seems more likely that these populations reached this area via
the low plains along the shore of Lake Erie, since the archeological
records are more consistent with this hypothesis.

There is only one modern record for Ontario (and Canada), previously
unreported, from 5.2 miles east of Alberton: “there was nothing about
the turtle that might have led me to suspect that it had been a captive”
(C. M. Bogert, pers. comm.). The proximiU^ of this locality to an archeo-
logical site for T. Carolina, namely Lake Medad, lends some support to
the belief that this is a natural record.

1969

Prehistoric Man’s Influence on Box Turtle

269

OfflO, INDIANA AND MICHIGAN

The distribution of the box turtle in Ohio and Indiana is apparently
not as continuous as currently thought. I can find only two seemingly
authentic records for a broad belt of land from the Pennsylvania border
to and including east-central Indiana. Dexter’s (1955) record for Portage
County, Ohio, was for a specimen actually collected in Carroll County
farther south, and within the natural range of the species. However, his
earlier record (Dexter, 1948) of a single specimen from Twin Lakes
(Portage Co., Ohio) appears authentic. T. Carolina is known from several
Cleveland parks ( Cuyahoga Co. ) but these records are discounted since
the species is absent from other areas beyond the city limits. T. Carolina
triunguis ( a form found generally west of the Mississippi River ) has also
been taken in these same parks. It and T. c. Carolina have both laid eggs
there, and newly-hatched juveniles of both were found in the parks in
1959 (C. Strong, pers. comm.). Box turtles are absent on the Lake Erie
islands (Langlois, 1964), although they are abundant on the adjacent
Ohio mainland.

The only apparently authentic record for east-central Indiana is a
single specimen collected in Adams Co. in 1953 (UMMZ 108078). Ex-
tensive collecting by many persons in this region and in the part of Ohio
mentioned above has failed to turn up additional specimens. Actually,
there is no fossil or archeological evidence to show that this turtle was
ever native to this region. Today, this area is largely under cultivation,
and the moist forests that T. Carolina typically frequents have been re-
duced or eliminated. It is possible that the drying conditions of the
Xerothermic Interval contributed to the demise of this turtle in these
parts of Ohio and Indiana, if it ever occurred there at all.

In Michigan, the box turtle is known from the southwestern half of
the lower peninsula. The most northern record along Lake Michigan is a
specimen from Benzie County (UMMZ 108890) collected in 1953. There
is only one record farther north, a single specimen collected during four
summers of field work at Douglas Lake (Cheboygan Co.; Ellis, 1917),
site of The University of Michigan’s biological station. Continuous col-
lecting in this region since that time has failed to turn up an additional
specimen (F. H. Test, pers. comm.).

The northern distribution of the box turtle in Michigan (nearly to 45°
North latitude) is of particular interest because it reaches only to about
43° North latitude on the Wisconsin side of Lake Michigan. Presumably,
the temperature-moderating influence of the lake (caused by the high

270

Annals of Carnegie Museum

VOL. 41

specific heat of water ) provides a more suitable climate on the eastern
shore. The effective growing season is extended well into the fall on the
eastern shore, which has resulted in a ffourishing agricultural “fruit belt”
in this region. In fall and winter, the temperature of the prevailing west-
erlies is often raised as much as 20° F. before these winds reach the
eastern shore (Wills in Hambridge, 1941). In spring, the warm south-
westerlies are cooled as they pass over the lake. The main effect is the
delay of plant development — and perhaps emergence of terrestrial
animals — until such time as the frequency of killing frosts is much re-
duced. Only a few of the terrestrial and aquatic vertebrates found on both
sides of the lake show a more northerly distribution in Michigan. I do
not have sufficient information on birds to comment on their distributions
or times of migration on both sides of the lake. None of the fish or mam-
mals shows a more northerly range limit in Michigan, except for a mole,
Scaloptis aquaticus (see Burt, 1964), but there are several examples
among terrestrial amphibians and reptiles. The most striking instances are
the box turtle, spotted turtle (Clenmiys guttata), and Fowler’s Toad
(Bufo woodhousei) . In fact, the northern limits of the ranges of these
three forms are almost exactly the same east of the Mississippi River, even
to minute details of distribution ( compare ranges in Conant, 1958, Logier
and Toner, 1961; Adler, 1968). In each of these, the distribution around
Lake Michigan is alike, the range in Michigan extending nearly to the
Upper Peninsula in the western part of the state along the lake. It is
possible that these three forms have some similar sensitive point in their
life histories that accounts for the resemblance in their northern distribu-
tions in this region. During the spring, they live in rather different habi-
tats: Biifo in temporary pools and streams, Clernmys in peiTnanent ponds
and marshes, and Terrapene on land. The moderating inffuence of Lake
Michigan would be expected to have rather different effects in each of
these habitats. In the fall, however, the habitats of the three are similar,
in that all are terrestrial. All three hibernate on land, including Clernmys
(Netting, 1936). If a common limiting factor like temperature exists,
the effects on the three species might be expected to be more uniform
in the fall when all exist in similar habitats. I suggest that the absence
of early killing frosts in the fall (in western Michigan as opposed to
Wisconsin) might allow them a longer period in which to find a suitable
site to bury themselves and successfully hibernate. Box turtles, for ex-
ample, bury themselves in a slow, stepwise fashion requiring several days
or weeks to complete. A killing frost or two in rapid succession could

1969

Prehistoric Man’s Influence on Box Turtle

271

kill such an animal before it could dig deep enough to effectively
hibernate.

ILLINOIS AND mSCONSIN

In Illinois, Smith ( 1961 ) records T. Carolina from the southern half of
the state. There are no fossil, archeological, or modern records for the
northern half.

The presence of this species in Wisconsin has been a matter for debate.
Some authors ( for example, Dickinson, 1965 ) have confused the eastern
and western box turtle (T. ornata), and their records are thus useless.
Higley (1889) reported two specimens of T. Carolina from Walworth
County, and Cahn (1929) mentions one from nearby Waukesha County.
Both appear valid. It is doubtful that Terrapene were carried about and
released in Higley ’s day as much as they are today.

DISCUSSION OF MODERN DISTRIBUTION

The modern limit of distribution for T. Carolina from Wisconsin to
Maine, as here understood (Fig. 1), is uneven. Its northward presence
along the eastern shore of Lake Michigan may be due to the tempering
effects of the lake, and its presence in maritime Maine and New Hamp-
shire may be due to similar moderating effects of the North Atlantic.
The absence of the box turtle in large portions of Ohio and Indiana, and
in northern Illinois, may be due to effects of the dry Xerothermic Interval,
although there is no fossil or archeological evidence to show that box
turtles ever inhabited this area. The present limited distribution of this
species in New York and Ontario, and its more widespread occurrence
there in prehistoric times, seem not to be explained by any of these in-
fluences. The probability that some other factor is involved, namely man,
is discussed below.

Distributional Change in New York and Ontario

We must first consider the possibility that introduction of the archeo-
logical specimens of T. Carolina was via inter-tribal trade, and that the
species was never native to western New York or Ontario. This possibility
is considered remote since specimens are present in large numbers at
many widespread localities. In at least half these places they were used
for food purposes, since unworked shells and limb bones, many of them
charred, have been found in several New York middens. While it is
possible that some of the worked material was introduced, it is doubtful
that the Indians of this region would have imported turtles for food when

272

Annals of Carnegie Museum

VOL. 41

deer and other game species were plentiful locally. Indeed, the Indians
of this area, particularly those of Late Woodland times, were noted for
their independence and virtual lack of trade with outside tribes ( Ritchie,
1965). It is true that seashells, shark teeth, Ohio flint, and other objects
were imported via trade, but there were no substitutes for these items
in western New York. Thus, the specimens of this turtle recovered from
burials and middens are assumed to have been collected from resident
populations.

Influence of Climate

Changes in the ranges of organisms over time periods are often ex-
plained on the basis of climatic factors, especially when there are co-
incident changes in the ranges of different forms. A unique change in a
single species’ range, like that which has occurred in the instance of the
box turtle, is more difficult to explain. Today, the box turtle is found in
tliree regions (Wisconsin, Michigan and New England; Fig. 1) at more
northerly latitudes ( above 42° ) than those in which it occurs in New
York. There are no obvious climatic factors, either now or since the
Pleistocene, that would simultaneously explain the absence of these
turtles in New York state, where they once occurred, and their presence
elsewhere at more northern latitudes. The possible tempering effects of
large bodies of water may account for the presence of box turtles so far
north in Michigan and New England. Why then are these turtles absent
today around other large bodies of water such as Lake Ontario (where
box turtles once occurred)? Box turtles may well have entered New
York during the warm, moist Climatic Optimum, as Guilday (in Ritchie,
1965) suggests. The later Xerothermic Interval, with its dry, prairie-like
conditions, may have been instrumental in eliminating this turtle from
certain parts of the Midwest, but probably did not affect populations
farther east in New York. This turtle persisted in New York long after
the termination of the Xerothermic Interval.

More gradual climatie changes also do not seem to be implicated as a
cause for extinction in western New York. A comparison of pollen profiles
from western New York (where box turtles disappeared) with those
from New England (where the box turtle persists) suggests more dra-
matic changes in climate in the latter area. The time-span of interest is
contained within the C-3 pollen zone, a sequence of deposition extending
from about 100-0 B.C. to the present time (Davis, 1965). C-14 dates are
not available from all profiles, nor is the C-3 zone equivalent in age at
all sites under consideration. Nevertheless, the C-3 profiles are approxi-

1969

Prehistoric Man’s Influence on Box Turtle

273

mately of equal age and if they are not truncated, should give a view of
floral ( and climatic ) conditions over this time period. An increase in the
percentage of non-arboreal pollen, such as Plantago and Compositae, is
assumed to indicate the effect of European forest clearance and, thus,
a fairly complete, non-truncated C-3 zone. With these limitations, I
have compared profiles in Vermont (Davis, 1965) and eastern New York
(Cox, 1959) with those from west-central New York (Cox, 1959). None
of these profiles appears to be truncated.

Since T. Carolina is an inhabitant of open, deciduous woods ( typically
beech-maple; Stickel, 1950 ) , particular emphasis is placed on the relative
abundance of Fagus and Acer. Although hemlock is not commonly as-
sociated with Terrapene habitats in regions south of New England, this
tree is a good indicator of moisture. It is present in all profiles considered
here and is used to give a relative idea of moistness. In all profiles ex-
amined, the abundance of Fagus steadily decreases in C-3. In west-
central New York (Bull Head Pond, Mud Lake-Tully, Perch Lake,
McLean Bog), Fagus pollen gradually decreases in abundance, eventual-
ly reaching 35 per cent to 70 per cent of its maximum abundance in C-3.
In eastern New York and Vermont (Mud Lake-Jordanville, Consaulus
Bog, Chestertown Bog, Brownington Pond), Fagus also decreases in
abundance, reaching 15 per cent to 40 percent of its maximum abundance
in C-3. Acer pollen is less abundant in all these profiles but shows
similar trends from west to east. Tsuga is more abundant and, in general,
decreases in frequency during the C-3 zone.

These data suggest a warming and drying trend throughout the New
York-Vermont region during the last 2000 years, but the trend is more
rapid and pronounced in the East. If the box turtle is sensitive to such
changes in climate, one would expect it to be most responsive in eastern
New York and New England. Nevertheless, box turtles persist in the
East and it follows that they would be expected to survive the apparently
less drastic changes that occurred in western New York. Thus, it appears
that such climatic changes were not directly responsible for elimination
of the box turtle in western New York.

Influence of Man

A more plausible explanation for the box turtle’s disappearance in the
Lake Ontario region involves the relationship between these turtles and
the Indians of this area. The box turtle was important to the Iroquois and
certain cognate tribes, and the time and place of its elimination is closely
correlated with the chronology of prehistoric cultural development in the

274

Annals of Carnegie Museum

VOL. 41

Lake Ontario region ( Ritchie, 1954 ) . Of all the Indian tribes of eastern
North America, those of New York and Ontario had the longest and most
intensive interest in box turtles. Judging from the burials and middens in
New York state, predation on this species was intense and included both
adult and juvenile specimens. At Archaic and Early Woodland sites
(dating from 3500 to 0 B.C.), these turtles were used for food and as
ceremonial objects as well, but mostly for the former purpose. At most
sites where turtles were recovered, box turtle was the species most com-
monly eaten, although the amount of meat per specimen is far less than
in several of the larger aquatic species. This use is in contrast to speci-
mens from Middle and Late Woodland sites (0-1600 A.D.) where the
majority were used in ceremonies.

The importance of the box turtle to Indians in the Lake Ontario region
is particularly noticeable in the Iroquoian culture. Terrapene was the
symbol of the earth or earth-bearer to the Iroquois and cognate Algon-
kian tribes. Consequently, it was much revered by them and was included
among ceremonial objects in most burials, even those of women and
children (Ritchie, 1954). Box turtles were used in many dances as
ornaments and rattles (Ritchie, 1954), and in game-hunting to invoke
the aid of the demons to capture prey (Fenton, 1941). In contrast, other
Indian tribes of the time considered the turtle unimportant in religion
or even despised the animal. Among the Indians in the western Great
Lakes region, box turtles were rarely used in ceremonies; in many of
these areas the aquatic Blanding’s Turtle ( Emycloidea blandingi ) served
this purpose (Adler, 1968). The Illinois and Miamis believed the earth
was held up by an otter rather than by a “tortoise.” Turtle eggs and
small turtles were even considered a sign of bewitchment to the Hurons
of Michigan and extreme western Ontario, although some turtles (but
not Terrapene) were made into rattles and others occasionally used for
food (Kinietz, 1940).

These Indians may also have affected box turtle populations in an-
other, less-direct way. The popular belief that eastern North America was
a continuous unbroken forest when the first Europeans arrived has been
challenged by several recent authors ( see Day, 1953 ) . There can be little
doubt that Indians, including the Iroquois, significantly altered these
primeval forests. The Indians created large clearings for villages and
agricultural fields. Their seasonal migrations and periodic relocation of
villages extended their influence even farther. Women continuously
foraged for firewood beyond the clearings and the men regularly fired
the woods to drive game, to improve visibility for war purposes, and to

1969

Prehistoric Man’s Influence on Box Turtle

275

increase the supply of grass seeds. Such fires are typically fatal to terres-
trial turtles and would be especially critical for juveniles with their less-
protective shells. More important, perhaps, were the extensive clearings
of land for cultivation of maize. The extent of these plots was often
enormous, sometimes encompassing thousands of acres around a single
village. The military expeditions into Iroquois country in the 18th Cen-
tury often discovered large quantities of corn and grain. Sullivan’s
expedition in 1779 destroyed at least 160,000 bushels of corn in a series
of small villages ( Day, 1953 ) . Thus, Terrapene were denied the use of a
large portion of their former habitat since these clearings were usually
made in forested valleys and lowlands.

The use of box turtles by New York Indians ended abruptly. As late
as 1650 A.D., box* turtles ( and rarely the painted turtle, Chrysemys picta )
were used as ceremonial objects (at the Kleis and Dutch Hollow sites,
for example), but by contact times the Indians had switched to the
snapping turtle (Chelydra serpentina), a large aquatic form. This species
was relatively common in archeological material dating from the earliest
times but was then used solely for food purposes until it came into cere-
monial use about 1700 A.D. The earliest record of worked Chelydra
comes from a single burial at the Pen Site (Onondaga Co., New York;
P. P. Pratt, pers. comm.), a site that was destroyed by Frontenac’s in-
vasion, about 1685-1696. Morgan ( 1904 ) illustrates a snapping turtle
rattle used about 1800. The Senecas of western New York still use
Chelydra rattles today (Fenton, 1941, pis, 17, 20, 23, 24).

The timing of this substitution cannot be precisely determined. Box
turtles have been found in graves at Dutch Hollow (1590-1615 A.D.)
and several nearby sites dating from 1550 to 1687 A.D. ( Rochester Junc-
tion, Dann, Power House, Feugle, Adams, and Cameron, according to
C. F, Wray, pers. comm.; none of these is mapped in Fig. 1 but they are
near locality 7 ) . It is perhaps significant that the frequency of box turtles
in these graves is much lower than at many earlier sites in this area. At
Dutch Hollow, only 7 of 272 graves had box turtles (Ritchie, 1954) and
at Rochester Junction, only 4 of 120 (Wray, pers. comm.). In addition,
there is no evidence in these later excavations that box turtles were used
as food, although they were commonly eaten as late as Oakfield and
Zimmerman times (1400-1500 A.D.). This suggests growing scarcity of
this turtle from 1500 to 1700 A.D. Some of the records of box turtles in
the most recent gravesites could represent specimens that had actually
been collected many years previously and made into artifacts.

276

Annals of Carnegie Museum

VOL. 41

It is possible that European mans agricultural activities had some
additional effect since the first pioneers in western New York arrived
about 1800 (Gordon, 1940). I consider this event less likely to be the
cause of the turtles’ disappearance than Indian influence, for two reasons.
First, the box turtle populations were apparently already decreasing in
size before 1700. Second, European cultivation in New England had
seemingly not affected the box turtle populations there, at approximately
the same latitude and at the edge of the species’ range.

Amount of Predation

To anyone familiar with the high population densities of T. Carolina
in many regions it may seem naive to consider it possible that prehistoric
man could be responsible for this turtle’s elimination in a region as large
as western New York state and parts of Ontario. But several facts should
be considered. First, the Lake Ontario region represented, originally, the
range limit for this species and, as is true for most organisms, the densi-
ties were probably reduced at the edge of the range where habitat is
presumably less favorable. The box turtle was apparently never wide-
spread in western New York, but rather localized in distribution and
restricted to lowland areas. This is coupled with primitive man’s steady,
selective predation for 5000 years, becoming most intensive during recent
Iroquoian times; there was a marked Indian population increase about
700-900 A.D. with the development of maize agriculture (J. B. Griffin,
pers. comm.). I do not think it is mere coincidence that box turtles dis-
appeared from precisely that portion of eastern North America where
human predation was most intense, especially when these events are
correlated temporally as well as geographically.

Fenton (1940) has estimated the population sizes of the various Iro-
quoian tribes at contact times, from which we can get a rough estimate
of the anqount of predation on box turtles for burial artifacts. These
population figures are undoubtedly underestimates since European dis-
eases and disruption of primitive economic patterns preceded the first
observers (Day, 1953). The Five Nations of the Iroquois Confederacy,
which occupied a part of western New York state, numbered at least
7000 persons when the first Europeans arrived (Fenton, 1940). If we
assume that the population size was stable (including a stable age
distribution) and further, that an average Indian lived 30 years, this
means that l/30th of the population, or 233 individuals, died each year.
Although box turtles were buried even with women and children, not all

1969

Prehistoric Man’s Influence on Box Turtle

277

graves contain shells. Possibly many shells did not survive the ravages of
decay (Ritchie, 1954). Nevertheless, except for the most recent burials,
some individual graves had a dozen or more shells. If there was an
average of one shell per grave, this means a predation of only 233 box
turtles per year in this region, a relatively small number. However, this
number represents only one type of predation on the box turtle. If we
add to this (1) the predation of other contemporary tribes (Hurons,
30,000 persons; Neutrals, 12,000; Eries, 8,000-10,000, at contact times),
at least in part of whose territory box turtles once lived; (2) the added
predation among Iroquoian and cognate tribes for food, medical, cere-
monial, and hunting purposes; (3) the sustained and selective predation
on box turtles for some 5000 years; and (4) perhaps most important, the
destruction of much of the turtle’s original habitat, then the full effect
of this predation and disruption of habitat seems evident.

Thus, present evidence suggests that the box turtle, an easily collected
and especially sought-for species, was eliminated in western New York
by the Iroquois Indians, a situation similar to the earlier Pleistocene ex-
tinctions of large mammals apparently perpetrated by man ( Martin and
Wright, 1967 ) . The porcupine ( Erethizon dorsatum ) may likewise have
been eliminated in Tennessee by Indians who prized the quills (Parma-
lee and Guilday, 1966). In contrast, the rice rat {Oryzomys palustris)
took advantage of the spread of Indian maize agriculture to extend its
distribution northward about 1000 A.D. (Goslin, 1951; Smith, 1965).
All these cases point to the importance of primitive man in altering
postglacial distributions of certain species.

Summary

In northeastern North America, the range of the box turtle ( Terrapene
Carolina) currently extends to central Pennsylvania, the lower Hudson
Valley of New York state, most of Massachusetts, and into southern
Maine. From 3500 B.C. until 1650 A.D., however, the range was more
extensive in this area, extending up the Hudson Valley and including
most of the western part of New York state and possibly southern On-
tario. Evidence suggests that prehistoric man, primarily Iroquoian
Indians, was responsible for elimination in these last-named areas.

Today, the box turtle is found in three areas at more northerly latitudes
than those in which it occurs in New York: Wisconsin, Michigan, and
New England. Each of these three areas had resident Indian populations,
but unlike those in New York state, these Indians used box turtles only
rarely. In western New York and southern Ontario, predation for food

278

Annals of Carnegie Museum

VOL. 41

and ceremonial purposes began as early as 3433 dz 250 years B.C. (Lam-
oka Lake site) and became most intense after 700 A.D. with the develop-
ment of the Iroquoian culture. Burials, including those of women and
children, sometimes contain up to a dozen specimens. Added to this was
the predation for food, medical, ceremonial, and hunting purposes. Per-
haps more important was the Indians’ destruction of the turtles’ habitat,
primarily for cultivation of maize. As late as 1650 A.D. (Kleis and Dutch
Hollow sites) box turtles were still in use, but by 1700, snapping turtles
(Cheltjdra serpentina) had replaced them in burials.

References Cited

Adler, Kraig

1968. Turtles from archeological sites in the Great Lakes region. Michigan
Archaeol., 14(3-4): 147-163.

Auffenberg, Walter

1958. Fossil turtles of the genus Terrapene in Florida. Bull Florida State Mus.,
Biol ScL, 3(2): 53-92.

1967. Further notes on fossil box turtles of Florida. Copeia, 1967(2’): 319-325.
Bishop, Sherman C.

1923. Notes on the herpetology of Albany County, New York. Copeia, (125):
117-120.

Bishop, Sherman C., and W. P. Alexander

1927. The amphibians and reptiles of Allegany State Park. New York State
Mus. Handbook, (3): 31-141.

Bleakney, Sherman

1958. The significance of turtle bones from archeological sites in southern On-
tario and Quebec. Canadian Field-Nat., 72(1): 1-5.

Burt, William H.

1964. The mammals of Michigan. Ann Arbor, Univ. Michigan Press.

Cahn, Alvin R.

1929. The herpetology of Waukesha County, Wisconsin. Copeia, (170): 4-8.
CoNANT, Roger

1958. A field guide to reptiles and amphibians. Boston, Houghton-Mifflin Co.
Cox, Donald D.

1959. Some postglacial forests in central and eastern New York as determined
by the method of pollen analysis. New York State Mus. Sci. Serv., Bull.
(377): 1-52.

Davis, Margaret B.

1965. Phytogeography and palynology of northeastern United States, p. 377-
401. In H. E. Wright, Jr. and D. G. Frey (eds. ), The Quaternary of the
United States, Princeton, New Jersey, Princeton Univ. Press.

Davis, William T.

1912. Miscellaneous observations on the natural history of Long Island, N. Y.
Proc. Staten Island Inst. Arts Sci., 3: 113-115.

Day, Gordon M.

1953. The Indian as an ecological factor in the northeastern forest. Ecology,
34(2): 329-346.

1969

Prehistoric Man’s Influence on Box Turtle

279

Dexter, Ralph W.

1948. More records of reptiles from Portage County, Ohio. Copeia, 1948(2):
139.

1955. Vertebrate fauna on Kent State University campus. Biologist, 37(2-4):
84-88.

Dickinson, W. E.

1965. Amphibians and reptiles of Wisconsin. Pop. Sci. Handbook Ser., Milwau-
kee Publ. Mus., (10): 1-45.

Eaton, Theodore H. Jr.

1945. Herpetological notes from Allegany State Park, New York. Copeia,
1945(2): 115.

Eights, James

1835. Papers on natural history. Zodiac, 1 : 62, 132.

Ellis, Max M.

1917. Amphibians and reptiles of the Douglas Lake region. 19th Ann. Rep.
Michigan Acad. Sci.: 45-63.

Fenton, William N.

1940. Problems arising from the historic northeastern position of the Iroquois.
Smithsonian Misc. Coll., 100: 159-251.

1941. Masked medicine societies of the Iroquois. Smithsonian Rep. for 1940:
397-430.

Gordon, Robert B.

1940. The primeval forest types of southwestern New York. Bull. New York
State Mus., (321): 1-102.

CosLiN, Robert M.

1951. Evidence of the occurrence of the rice rat in prehistoric Indian village
sites in Ohio. Bull. Ohio Indian Relic Coll. Soc., (26): 19-21.

Guilday, John E.

1962a. Notes on Pleistocene vertebrates from Wythe County, Virginia. Ann.
Carnegie Mus., 36: 77-86.

1962b. The Pleistocene local fauna of the Natural Chimneys, Augusta County,
Virginia. Ann. Carnegie Mus., 36: 87-122.

Guilday, John E., Harold W. Hamilton, and Allen D, McCrady

1966. The bone breccia of Bootlegger Sink, York County, Pennsylvania. Ann.
Carnegie Mus., 38: 145-163.

Guilday, John E., Paul S. Martin, and Allen D. McCrady

1964. New Paris no. 4: A Pleistocene cave deposit in Bedford County, Penn-
sylvania. Bull. Natl. Speleol. Soe., 26(4): 121-194.

Hambridge, Gove (ed. )

1941. Climate and man. Yearbook U. S. Dept. Agr.: i-xii, 1-1248.

Higley, W. K.

1889. Reptilia and Batrachia of Wisconsin. Trans. Wisconsin Acad. Sci., Arts
Lett., 7: 155-176.

Huse, W. H.

1901. The Testudinata of New Hampshire. Proc. Manchester Inst. Arts Sci.,
2: 47-51.

Kinietz, W. Vernon

1940. The Indians of the western Great Lakes, 1615-1760. Occas. Contr. Mus.
Anthropol., Univ. Michigan, (10): i-xiv, 1-427.

Langebartel, Dave A.

1953. The reptiles and amphibians, p. 91-109. In R. T. Hatt et ah. Faunal and
archaeological researches in Yueatan Caves. Cranbrook Inst. Sci. Bull.,
(33): 1-119.

280

Annals of Carnegie Museum

voi. 41

Langlois, Thomas H.

1964. Amphibians and reptiles of the Erie Islands. Ohio Jour. Sci., 64(1): 11-
25.

Latham, Roy

1916. Notes on Cistudo Carolina from Orient, Long Island. Copeia, (34): 65-67.
Logier, E. B. S., and G. C. Toner

1961. Cheek list of the amphibians and reptiles of Canada and Alaska. Roy.
Ontario Mus., Life Sci. Div., Contr. (53): 1-92.

Martin, P. S., and H. E. Wright, Jr. (eds)

1967. Pleistocene extinctions. New Haven, Yale Univ. Press.

Mearns, E. a.

1898. A study of the vertebrate fauna of the Hudson highlands. Bull. Amer.
Mus. Nat. Hist., 10: 303-352.

Milstead, William W.

1965. Notes on the identities of some poorly known fossils of box turtles (Terra-
pene). Copeia, 1965(4): 513-514.

1967. Fossil box turtles (Terrapene) from central North America and box turtles
of eastern Mexico. Copeia, 1967(1): 168-179.

1969. Studies on the evolution of box turtles (genus Terrapene) . Bull. Florida
St. Mus., Biol. Sci., 14(1): 1-113.

Morgan, L. H.

1904. League of the Ho-De'-No-Sau-Nee or Iroquois. Second edition. New York,
Dodd, Mead & Co. Vol. I, 338 p.

Netting, M. G.

1936. Hibernation and migration of the spotted turtle, Clemmys guttata. Copeia,
1936(2): 112.

Ogden, J. Gordon, HI

1958. Wisconsin vegetation and climate of Martha’s Vineyard. Doctoral disser-
tation, Yale University, New Haven.

Parmalee, Paul W., and John E. Guilday

1966. A recent record of porcupine from Tennessee. Jour. Tennessee Acad. Sci.,
41(3): 81-82.

Ritchie, William A.

1954. Dutch Hollow, an early Historic Period Seneca site in Livingston County,
New York. Res. Trans. New York St. Archeol. Assn., 13(1): i-vi, 1-98.
1965. The archeology of New York state. Garden City, New York, Nat. Hist.
Press, 357 p.

Smith, Philip W.

1961. The amphibians and reptiles of Illinois. Bull. Illinois Nat. Hist. Surv.,
28(1): 1-298.

1965. Recent adjustments in animal ranges, p. 633-642. In H. E. Wright, Jr.
and D. G. Frey (eds.). The Quaternary of the United States. Princeton,
New Jersey, Princeton Univ. Press.

Starrett, Andrew

1958. Insular variation in mice of the Microtus pennsylvanicus group in south-
eastern Massachusetts. Doctoral dissertation, Univ. Michigan, Ann Arbor.

Stickel, Lucille F.

1950. Populations and home range relationships of the box turtle, Terrapene c.
Carolina (Linnaeus). Ecol. Monogr., 20: 351-378.

Taylor, W. E.

1895. The box tortoises of North America. Proc. U. S. Natl. Mus., 17: 573-588.
Werner, William E., Jr.

1959. Amphibians and reptiles of the Thousand Islands Region, New York.
Copeia, 1959(2): 170-172.

Back issues of many Annals of Carnegie Museum articles
are available, and a few early complete volumes and parts
are listed at half price. Orders and inquiries should be
addressed to: Publications Secretary, Carnegie Museum,
4400 Forbes Avenue, Pittsburgh, Pa. 15213.

. 73

Pfc ^>4-2-^

Volume 41 Annals of Carnegie Museum Article 10

EUPROOPS FROM THE "UFFINGTON SHALE”
OF COLUMBIANA COUNTY, OHIO

James L. Murphy

Department of Geology,

Case- Western Reserve University
Cleveland, Ohio

A nearly complete specimen of Euproops laevicula Raymond and the
prosoma of a specimen of E. danae (Meek and Worthen) have been
found in two large ironstone concretions in the dark fissile roof shales
overlying the upper Freeport ( No. 7) coal exposed in a strip mine, SE%,
SE/4 Section 19, Center Township, Columbiana County, Ohio. These are
the first xiphosuran remains reported from the Pennsylvanian system
of Ohio, although several specimens have been described from the
Conemaugh and Allegheny groups of Pennsylvania, and their occur-
rence is of considerable stratigraphic interest.

Class ARACHNIDA
Subclass MEROSTOMATA Woodward
Order Xiphosura Gronovius
Family Euproopidae Eller
Genus Euproops Meek
Euproops danae (Meek and Worthen)

Plate 1, figure 1

Bellinurus danae Meek and Worthen, 1865, Proc. Acad. Nat. Sci. Phila-
delphia, 2: 44. Meek and Worthen, 1866, Illinois Geol. Surv., 2: 395,
pi. 32, figs. 2, 2a.

Prestwichia danae Meek, 1867, Amer. Jour. Sci., 43: 257. Packard, 1886,
Mem. Natl. Acad. Sci., 3(16): 146, pi. 5, figs. 3, 3a; pi. 6, figs. 1, 2a.
Euproops danae Meek, 1867, Amer. Jour. Sci., 43: 395. Woodward, 1868,
Geol. Mag., 5: 2. White, 1885, Geol. Natl. Hist. Surv. Indiana, 13th.

Issued September 25

282

Annals of Carnegie Museuim

VOL. 41

Ann. Kept. (1883); 170, pi. 39, fig. 1. Raymond, 1943, Illinois State

Mus. Sci. Papers., 3(3): 4-6, pi. 1, figs. 1-2; pL 2, figs. 1-2. Raymond,

1944, Rull. Mus. Comp. ZooL, 94(10): 484-486.

repository: Carnegie Museum (CM 33060)

measurements: Maximum length of prosoma, 11.5 mm. Maximum width of
prosoma at the base of the genal spines, 23.0 mm.

DISCUSSION: This species is so well known that no formal diagnosis is
necessary (Raymond, 1944: 485). Although incomplete, consisting only
of the prosoma, the specimen can be readily identified, for it possesses
the rapidly tapering cardiac lobe characteristic of E. danae. E. packardi
Willard and Jones has a similar cardiac lobe, but the prosoma is relative-
ly wider. E. packardi is also described as having a rather strong posterior
notch at the base of the cardiac lobe (Willard and Jones, 1935: 1277).
This feature could not be discerned on a cast of the holotype from the
Carnegie Museum collections (CM 23459).

The width/length ratio of the upper Freeport specimen is less than
that of typical specimens of E. danae. The very short genal spines and
the relatively small size of this specimen also suggest that it represents
a young individual. The prosoma is unusually well preserved, only
slightly wrinkled, with the ocelli prominent. The transverse bar at the
midpoint of the cardiac lobe is also preserved.

Euproops laevicida Raymond
Plate 1, figure 2

Euproops laevicida Raymond, 1944, Rull. Mus. Comp. Zook, 94(10);

490, fig. 3.

repository: Carnegie Museum (CM 33061)

description: Cardiac lobe gradually tapering, prosoma narrow with a width/
length ratio of 2.0. Thoracetron with nearly smooth lateral lobes, axial lobe with
faint transverse furrows and nodes on the last several rings. Spines on thoracetron
not preserved; genal spines short.

The prosoma of this specimen is partly obscured by the impression of a second
specimen, though most of the features of the prosoma can be discerned. Diag-
nostic features of the second specimen cannot be determined.

measurements: Maximum length of prosoma, 11 mm. Maximum length of
specimen (without telson), 20 mm.

DISCUSSION: The obscure transverse ridges on the axial and lateral lobes
of the thoracetron are characteristic of this species. Euproops laevicula
can be distinguished from the similar E, parkardi of the Allegheny group
of Pennsylvania by the tapering cardiac lobe; that of E. packardi is like

1969

Euproops From the “Uffington Shale’

283

the abruptly contracted lobe of E. dame. Euproops packardi also has a
much wider prosoma and comparatively conspicuous transverse ridges

PLATE 1

Fig. 1. Euproops danae (Meek and Worthen); CM 33060. Prosoma from iron-
stone concretion. Upper Freeport coal roof shale, strip mine high wall, SEli,
SE/4 Section 19, Center Township, Columbiana County, Ohio.

Fig. 2. Euproops laevicula Raymond; CM 33061. A nearly complete specimen from
the same horizon and locality.

284

Annals of Carnegie Museum

VOL. 41

on the lateral lobes of the thoracetron. The Ohio specimen may repre-
sent a slightly more mature individual than the holotype of E. packardi
(CM 23459), for it has a longer thoracetron. Euproops danae can be
distinguished by the presence of prominent transverse ridges on the
thoracetron and by the abruptly tapering cardiac lobe.

Euproops darrahi Raymond of the Mason shale of Pennsylvania is
probably a junior synonym of E. packardi.

This is only the second reported specimen of E. laevicula. It is of
particular interest because of the close agreement with Raymond’s
original description. In a group like the Xiphosura, one should be ex-
tremely cautious in proposing new species. Considerable variation may
be expected between individuals of the same species, fossil specimens are
rare and usually distorted by crushing, and there is the further diESculty
of molt stages and sexual dimorphism. Raymond might in fact be sus-
pected of a needless proliferation of species based upon very slender
evidence, arbitrary criteria, and often only single specimens. These two
new specimens can readily be assigned to existing species within Ray-
mond’s scheme of classification. The conclusion to be drawn is that
Raymond’s classification, though artificial, is internally consistent, fairly
conservative, and may suffice until a thorough, modem reappraisal of
this group is undertaken.

The Upper Freeport Roof Shale Fauna

Associated fossils at the xiphosuran locality include the estherid
Cyzicus sp., which has been found in both the black fissile shale and the
ironstone nodules. The black shale also yields abundant ganoid (Haplo-
lepis) and crossopterygian fish scales. Ostracods {Candona and Guts-
chickia) are also common in the nodules. A single blattoid wing has
been identified by C. J. Durden as Stygetohlatta. A conspecific specimen
has been found in the Mason shale of Pennsylvania (C. J. Durden,
personal comm.. Sept. 5, 1967).

Examination of other outcrops of the upper Freeport roof shales in
Madison, Center, Hanover, and West townships reveals that the black
fissile shale bed is widespread, with abundant ganoid scales and
occasional Xenacanthus teeth. One fragmentary insect wing, identified
by C. J. Durden as either Mylacris or Orthomylacris sp., has been found
by R. Max Card, Lisbon, Ohio, in an exposure along U. S. Route 30,
NE/4 24, Hanover Township, Columbiana County, Ohio. At this locality
fossiliferous ironstone concretions occur sparingly, containing a few
fresh-water ostracods. Estheriids (Leaia) occur in a section measured

1969

Euproops From the “Uffington Shale’

285

by Stout and Lamborn (1924: 258) at West Point, Madison Township,
and in a strip mine in West Township, though in both instances the
unit is a silty shale overlying the black fissile shale. Normal marine fossils
have not been found overlying the coal. With the exception of a Perry
County locality cited by Sturgeon (1958: 92), where the brackish-water
forms Lingula and Orbiculoidea occur sparingly, no brachiopods or
molluscs are known from the beds immediately overlying the upper Free-
port coal.

STRATIGRAPHIC DISCUSSION: The stratigraphic problem has been con-
fused by recognition of a marine Uffington shale member that does not
exist (Sturgeon, 1958: 92). Sturgeon failed to note that the marine Uf-
fington shale of the type area, Morgantown, West Virginia, was long ago
found to have been based upon a misidentified outcrop of the Mahoning
coal and Brush Creek shale member. Nor did he note Price's (1917)
redescription of the Uffington as a non-marine, plant-bearing shale im-
mediately overlying the upper Freeport coal. A definite marine member
overlying the upper Freeport coal would certainly deserve formal strati-
graphic recognition, but the presence of two brackish-water brachiopod
species at the Perry County locality is a poor foundation on which to
erect a new marine member. The Columbiana County localities cited by
Sturgeon ( Stout and Lamborn, 1924: 238, 258 ) were originally described
only as “fossiliferous,” and collecting at these localities has shown that
Sturgeon's belief that these fossils are marine is ill-founded.

If a marine zone were discovered over the upper Freeport coal, it
could not be called Uffington, for that name belongs to the non-marine,
plant-bearing unit of Price. Price’s Uffington shale, incidentally, though
recognized in the USGS lexicon, is probably deserving of no more than
bed rank. The same is true of the black, fissile shale which is usually
found between the coal and the Uffington shale or between the coal and
the Mahoning sandstone member when the Uffington is absent. This
black shale unit is presumably equivalent to units 2 and 3 of Cross’s
typical Dunkard cyclothem (Beerbower, 1961: 1031). Typical Uffington
shale, as described by Price, and the black fissile roof shale bed are both
present in Athens County, Ohio, but neither can be considered important
enough to be termed a member.

References Cited

Beerbower, J. R.

1961. Origin of cyclothems of the Dunkard Group (upper Pennsylvanian-lower
Permian) in Pennsylvania, West Virginia, and Ohio. Bull. Geol. Soc.
Amer., 72: 1029-1050.

286

Annals of Carnegie Museum

VOL. 41

Price, W. A.

1917. The UflBngton Shale of northern West Virginia — absence of marine
fauna. 807-816 in Hennen, R. V., Geology of Braxton and Clay Counties.
West Virginia Geol. Survey, 19: 1-883.

Raymond, P. E.

1944. Late Paleozoic Xiphosurans. Bull. Mus. Comp. Zool., 44: 475-508.
Stout, Wilber, and R. E. Lamborn

1824. Geology of Columbiana County, Ohio. Ohio Geol, Survey, Bull. 28: 1-408.
Sturgeon, M. T., et al.

1958. Geology and mineral resources of Athens County, Ohio. Ohio Geol.
Survey, Bull. 57: 1-600.

Willard, Bradford, and T. Husband Jones

1935. A new xiphosuran from the Allegheny of Pennsylvania. Pa. Acad. Sci.
Proc., 9:126-131.

Back issues of many Annals of Carnegie Museum articles
are available, and a few early complete volumes and parts
are listed at half price. Orders and inquiries should be
addressed toi Publications Secretary, Carnegie Museum,
4400 Forbes Avenue, Pittsburgh, Pa. 15213.

/

^ S3 -

UllSNI NVIN0SHilWS^S3 I MVa 9 n LI B RAR I ES SMITHSONIAN“lNSTlTUTIOt

I E

CO 1|& CO

t 2

ARIES SMITHSONIAN INSTITUTION NOliniliSNI NVIN0SH1IWS^S3 I U VH 8 I
CO r: CO — (rt

_ O o

lillSNl"^NVIN0SHiIlfllS^S3 I BVa 8 n"'LI B R AR I ES^SMITHSONIAN~‘lNSTITUTIOr
>2 r~ z r“

AR 1 ES SMITHSONIAN~INSTITUTION‘”NOIXniIiSNl“NVINOSHilWs'”s3 I B VB 8 I

S < V s s

±

CO

o

> >vH:a^ 2 >• 2 Xi^vosii^ >

iIJ.SNI_NVIN0SHimS^S3 I 8 V8 8 ll^LI B RAR I ES^SMITHSONIAN^INSTITUTION

— - CO — _

2 ^ -J 2

^R I ES SMITHSONIAN INSTITUTION NOIlflillSNI NVINOSHimS S3 I B VB 8 I

CO
70
>

m S

IXIXSNI^NVIN0SHilWs"s3 1 aVB 8 n“LI B RAR I Es“sMITHS0n1an"|NSTITUTI0I

- - X ^ ^

z Ai^W^A -H . z —I /o^eWfcW

2
CO

o

ARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVIN0SHiIl/MS^S3 I d Vy 8 T

^ 2 ^ ^ v CO

^ li! S •' ^

O _ '<«iu^5X o

XIXSNI NVIN0SHXWS_S3 I ava 8 n"'LI B RAR I ES^SMITHSONIAN"'lNSTITUTIO^

_ > ? r- Z r-

illSNI NVINOSHimS SBIdVdan^LIBRARIES SMITHSONIAN INSTITUTION

2 ^ w W 2

2 ». ^ ^ ><U5V/Vi\ ^

C/3

o
V ?

(\R I ES^^SMITHSONIAN'^INSTITUTION^^NOliniliSNI NVIN0SHiIlN$^S3 I HVd 8 11
c/) 5 ^ 5 ^

jl i lOl i lOI I 2

QO

iliSNI NVINOSHillAIS S3ldVHan LIBRARIES SMITHSONIAN INSTITUTION
r- V Z r* 2 ■”

CD

:d
>

?□

.oy —

m ^ m

\RIES SMITHSONIAN~INSTITUTION‘”NOIinXliSNl“NVINOSHilWS SHlHVaan
<2 z \ v> z

S I I I

^ o 4^^ mj/ X 2

*v '- z, t S.»

> '■O'Ainy g ' > ' S 'vSoflS'

iIXSNI NVIN0SHXriMs‘^S3iavaan^LIBRARIEs‘^SMlTHS0NIAN INSTITUTION

ARIES SMITHSONIAN INSTITUTION NOIXOXIXSNI NVINOSHXIWS S3iavaar

e/j Z CO * Z </> *■■

ARIES SMITHSONIAN INSTITUTION NOIXOXIXSNI NVINOSHXIWS S3iavaar