iiiiiiiiiiiiiiiJIlfiillliiii

i

\
i

1

I
(

i
I

I.
t
K

Jl\ll

HARVARD UNIVERSITY

LIBRARY

OF THE

Museum of Comparative Zoology

^J

UNIVERSITY OF KANSAS PUBLICATIONS
MUSEUM OF NATURAL HISTORY

The University of Kansas Publications, Museum of Natural History, beginning
wdth volume 1 in 1946, was discontinued with volume 20 in 1971. Shorter research
papers formerly published in the above series are now published as Occasional
Papers, Museum of Natiual History. The Miscellaneous Publications, Museum of
Natural History, began with number 1 in 1946. Longer research papers are pub-
lished in that series. Monographs of the Museum of Natural History were initiated
in 1970. All manuscripts are subjected to critical review by intra- and extramural spe-
cialists; final acceptance is at the discretion of the publications committee.

Institutional libraries interested in exchanging pubHcations may obtain the Occa-
sional Papers and Miscellaneous Publications by addressing the Exchange Librarian,
University of Kansas Library, Lawrence, Kansas 66044. Individuals may purchase
separate numbers of all series. Prices may be obtained upon request addressed to
Publications Secretary, Museum of Natural History, University of Kansas, Law-
rence, Kansas 66044.

O

?\?

University of Kansas
Museum of Natural History

Miscellaneous Publication No. 57
February 2, 1973

A Systematic Review of the Teiid Lizards,

Genus Bachia, with Remarks on

Heterodactylus and Anotosaura

By

James R. Dixon

Chief Curator

Texas Cooperative Wildlife Collection

Texas A <b- M University

College Station, Texas 77843

University of Kansas

Lawrence

1973

University of Kansas Publications, Museum of Natural History

Editor: Linda Trueb
Managing Editor: William E. Duellman

Miscellaneous Publication No. 57

pp. 1-47; 15 figures

Published February 2, 1973

Museum of Natural History

University of Kansas

Lawrence, Kansas 66044

U.S.A.

Printed By

University of Kansas Printing Service

Lawrence, Kansas

CONTENTS

INTRODUCTION 4

Historical Resume 4

The Status of Ophiognomon Cope 6

The Status of Anotosaura Amaral and Heterodactijlus Spix 7

Methods and Materials 8

Acknowledgments 11

EVOLUTIONARY TRENDS IN THE GENUS BACHIA 12

BACHIA GRAY 1845 14

KEY TO SPECIES 16

SYSTEMATIC ACCOUNTS 17

The Bachia bresslaui Group 17

The Bachia dorbigmji Group 20

The Bachia heteropa Group 32

The Bachia flavescens Group 37

SUMMARY 42

RESUMEN 43

SPECIMENS EXAMINED 43

LITERATURE CITED 45

INTRODUCTION

Bachia is one of three genera in the
microteiid section of the family Teiidae
having no external ear; the other two
genera are Anotosaura and Heterodac-
tylns. The 15 species of Bachio are sim-
ilar in body shape, size, and frequently
in color and color pattern. The body
and tail are extremely elongated and the
limbs relatively short in all species. The
species of Bachia exhibit differences in
external structural features, size of limbs,
and numbers of fingers and toes.

Species of Bachia are inhabitants of
semi-arid, subtropical and tropical habi-
tats where there is an abundance of
ground litter (forest floor) or soft sandy
loam soils with abundant cactus litter
(benches above Rio Maraiion at 500 m).

When I initiated a study of the lizard
genus PhyUodactyhis of South America
at Lima, Peru, in 1968, I was immedi-
ately confronted with many lizards that
I could not identify. One of these was
a series of worm-like lizards (Bachia)
from the Amazonian slopes of the Rio
Maraiion. Further examination of the
samples taken from the Rio Maraiion
flood plain revealed the existence of pos-
sible hybridization between two species,
Bachia intermedia and B. harhouri
(Dixon, 1972). The absence of hind
limbs in one species, degenerate limbs in
the other, the presence or absence of
certain head scales, the arrangement of
other cephalic squamation, aroused my
interest in the possible evolutionary
pathways, function and behavior of these
species. Therefore, I initiated a study of
all of the available material of the known
species of the genus, with the hope that
a detailed study of the various popula-
tions would reveal why some species
possess prefrontals, supraoculars, inter-
parietals, internasals, and four toes on
each limb, whereas others lack these fea-
tures or possess various combinations of
them.

The study, at times, became entan-
gled in purely taxonomic problems in-

volving literature; there were other
problems, such as loss of type material,
absence of comparative material, and
loss of specimens through the mails.
Nevertheless, the final result was the
accumulation of data on some 600
specimens of Bachia from throughout its
known distribution, and the formulation
of several interesting evolutionary trends.

Historical Resume

Writings pertaining to the species of
the teiid genus Bachia are, in the modern
sense, chaotic. Total taxonomic confu-
sion has prevailed since the earliest in-
adequate description of the species
flavescens by Bonnaterre ( 1789 ) , which,
at best is poorly described; this descrip-
tion is based on an illustration given by
Lacepede in 1789. Prior to 1940, thirteen
generic names were proposed for various
species of Bachia. Short notes on the
distribution and/or taxonomy of several
species of Bachia are scattered through
the literature of the past 75 years. Some
of the more pertinent notes are those of
Peracca (1896), Loveridge (1933),
Parker (1928, 1935a), Crawford (1931),
Daniel (1955), Brongersma (1956),
Cunha (1958) and Vanzolini (1961a).

Boulenger (1885, 1887) did not rec-
ognize the current generic name Bachia
Gray ( 1845 ) and grouped ten species
into three genera — Scolecosaurus Bou-
lenger (1885), Cophias Fitzinger (1843)
and Ophiognomon Cope (1868). Con-
currently he placed seven generic names
into the synonymies of these genera.
Boulenger did not know that three of
the species were based upon Lacepede's
(1789) figure on plate 32. Thus, he was
the first to perpetuate a mistake that has
permeated the literature until 1970,
when Peters and Donoso-Barros reduced
the three names to strict synonyms.

Carman (1892) showed that the ge-
nus Cophias was preoccupied and that
Bachia Gray (1845) was the earliest avail-
able name for the genus. Ruthven (1925)

DIXON: SYSTEMATIC REVIEW OF BACHIA

discussed the squamation of those spe-
cies recognized in the genus Cophias by
Boulenger (1SS5). He divided the spe-
cies into two groups of Bachia based
upon Boulenger 's (1885) division of the
genus Cophias into two sections. One
group was characterized by quadran-
guhir dorsals, whereas the other was dis-
tinguished by hexagonal dorsals; Ruth-
ven called the first group "cophias," and
the second "dorbignyi." Like Boulenger,
Ruthven failed to describe the species
correctly; he stated that Bachia Uneata
and Bachia hetewpa have quadrangular
dorsals, whereas in fact, they do not.
Each of the mistakes in identification
and/ or characters of the species from the
earliest descriptions has been perpetu-
ated in part to the present time.

Burt and Burt (1931) added another
species (Bachia barbouri) and attempted
to rectify some of the mistakes made by
earlier authors. They peipetuated Bou-
lenger's ( 1903 ) error by stating that
Bachia Uneata had quadrangular dorsals,
and therefore placed it in the "cophias"
group where it does not belong. Burt
and Burt ( 1931 ) were the first to correct
the systematic position of Bachia heter-
opa. Other authors failed to notice that
in Boulenger's ( 1887 ) addenda and cor-
rigenda to the "Catalogue of the Lizards
. . ." he redescribed heteropa from an
examination of the type, and described
a new species, Lepidosternon boettgeri,
for the species he had called heteropa
in 1885.

Vanzolini ( 1961b ) described an ad-
ditional species (Bachia scolecoides) and
correctly allocated the genus Scolecosati-
rus Boulenger ( 1885 ) and Apatelus
Amaral ( 1935 ) to the synonymy of
Bachia, thereby adding five additional
species to the genus. He summarized the
major characters of the species, and in-
dicated that two species were intermedi-
ate between Ruthven's (1925) "cophias"
and "dorbignyi" groups, thus questioning
the advisability of recognizing Ruthven's
species groups.

Thomas ( 1965a ) discussed the status
of Bachia alleni, B. trinitatis and B. pal-

lidiceps in northeastern South America.
He showed that trinitatis and alleni are
conspecific, relegated alleni parviceps to
the synonymy of alleni alleni, and dis-
cussed the status of Bachia cnvieri
(Fitzinger, 1826). He maintained that
Barbour (1914, 1933) was correct in
suggesting that, because Bachia cuvieri
was from an unknown locality and the
type lost, the name should be held as
incertae sedis pending the availability of
additional material. Burt and Burt (1931)
indicated that the name applied to a
population of Scolecosaurus alleni from
the island of Grenada, West Indies, be-
cause the characters mentioned briefly in
the original description could be associ-
ated with that population. Subsequently
Barbour ( 1933 ) indicated that Burt and
Burt did not have all the facts. On the
basis of his examination of the three
specimens available to Dumeril and
Bibron (1839), Barbour decided that
three species were represented by their
redescription of Chalcis cuvieri. Barbour
(1933) stated further that Dumeril and
Bibron ( 1839 ) gave South America, spe-
cifically "Colombie," as the range of
Scolecosaurus cuvieri. At that time New
Grenada or Colombia included what is
now Venezuela, Colombia, Panama, Ec-
uador, Peru, and Bolivia. According to
Barbour, one specimen represented
Scolecosaurus alleni (MNHN 2836), one
Scolecosaurus trinitatis (MNHN 2837),
and one Scolecosaurus pallidiceps
(MNHN 7051). If any one of the speci-
mens represented Scolecosaurus cuvieri,
it would be the specimen relegated to
Scolecosaurus alleni because of the me-
dial contact of the prefrontals ( Barbour,
1933).

The only character of any value in
the original description of Brachypus
cuvieri by Fitzinger (1826) is the pres-
ence of four toes on anterior and pos-
terior limbs. Bachia panoplia, B. Sco-
lecoides, and B. alleni have four toes on
each limb. All three species are easily
separated on the basis of a number of
characters; only re-examination of the
holotype will definitely establish the

6

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

identity of Bachia ciwieri. Since the type
is lost (formerly in the Naturhistorisches
Museum, Wien), I propose that the cur-
rently recognized species of "four toed"
Bachia be retained at their specific level,
because their holotypes are extant, avail-
able, and their specific names are in
current use. Bachia ctivieri will remain
incertae sedis until the type is redis-
covered.

Peters and Donoso-Barros (1970)
followed Vanzolini's (1961b) arrange-
ment, leaving the genera Scolecosaurus
and Apatehis in the synonymy of Bachia.
They recognized 19 species and four
subspecies of Bachia, and for the first
time formulated a key for all currently
recognized taxa. Peters and Donoso-
Barros (1970) pointed out that the trivial
names favescens, cophias and tridactyhis
are all based upon the figure of plate 32
of Lacepede (1789). All three have the
same "Iconotype" and the latter two are
strict synonyms of flavescens (Bonna-
terre). Boulenger (1885) recognized Co-
phias tridactyhis and Cophias flavescens,
but indicated that tridactyhis possesses
an interparietal whereas flavescens does
not. The type description of Chalcides
flavescens (Bonnaterre, 1789) is brief,
with a poor illustration, and no indica-
tion of the t>'pe of head scales. Boulenger
(1885) further confused the issue by
stating that Cophias flavescens has an
undivided hind limb, whereas Bonna-
terre (1789) definitely stated that the
hind limb of Chalcides flavescens bears
three, clawed toes. It seems obvious that
Boulenger's description of Cophias
flavescens is based upon specimens from
the Guianas (Bachia monodactyhis), as
indicated in his list of specimens exam-
ined following the description.

Daudin (1802:267) described Chal-
cides tridactyhis and C. monodactyhis
in the same paper, and apparently had
specimens on hand of one or both forms.
Even though he based the description of
C. tridactyhis on the figure on plate 32
of Lacepede (1789), he indicated that
C. tridactyhis possessed an interparietal.
Of some 600 specimens of Bachia exam-

ined, I have found only one specimen
that possesses three toes on each foot
and an interparietal. Daudin must have
either seen the specimen illustrated by
Lacepede or corresponded with him
about its essential characters and incor-
porated these into his description of C.
tridactyhis.

Regardless of the actual circum-
stances leading to Daudin's (1802) de-
scription of C. tridactyhis, a population
does exist that essentially fits his descrip-
tion. Since the name Bachia flavescens
(Bonnaterre, 1789) applies to a popula-
tion having three toes on each foot, and
the only extant specimen having three
toes on each foot also possesses an inter-
parietal, I assume the specimen illus-
trated by Lacepede had an interparietal.
If my assumptions are correct, then the
trivial name tridactyhis is indeed, a strict
synonym of flavescens, as pointed out by
Peters and Donoso-Barros (1970).

The trivial names that I assign to
Bachia are listed in table 1.

The Status of Ophioii,nomon Cope

Cope (1868) described the microteiid
genus Ophioii^nomon, t\'pe species tri-
sanale, and distinguished it from Bachia
by the absence of the frontonasal scale.
He later (1874) described the genus
Propus, type species verniiformis, and
separated it from Ophio^nonwn by the
absence of a hind limb, more numerous
scales and mutual contact of the second
pair of chin shields. Peters (1872) de-
scribed the genus Hapahlcpis, t\pe spe-
cies ahendrothi: apparently he was un-
aware of Cope's (1868) description of
Ophiognomon. Boulenger (1885) noted
the similarities of the three genera and
placed Propus and Hapalolepis as strict
synonyms of Ophio<i,nomon. This ar-
rangement has been followed to the pres-
ent time.

I have examined the holotypes of all
three species, and 69 additional speci-
mens. I find few difl^^erences in the vari-
ation in scutellation exhibited by the 72
specimens of OpJiiognonion and 450
specimens of Bacliia. The only character

DIXON: SYSTEMATIC REVIEW OF BACHIA

Table 1. Alphabetical Synonymy of the Species of Bachia.

Trivial Name, Original Generic Name, Author, Date Current Name

abencJrothii (Hapololepis) Peters, 1871 B. trisanale

alleni (Scolecosaunis) Barbour, 1914 B. heteropa

anomaki (Bachia) Roiix, 1929 B. heteropa

harJ)ouri (Bachia) Burt and Burt, 1931 B. barbouri

bicolor (Heteroclonium) Cope, 1896 B. bicolor

boett^eri (Lepidosternon) Boulenger, 1885 B. favescens

bresslaui (Apatehis) Amaral, 1935 B. bresslaui

cophias (Chamaesaura) Schneider, 1801 B. favescens

cuvieri (Brachypus) Fitzinger, 1826 incertae sedis

daudini (Colobus) Merrem, 1820 B. monodactijlus

dorbignyi (ChaJcides) Dumeril and Bibron, 1839 B. dorbignyi

flavescens (Chalcides) Bonnaterre, 1789 B. flavescens

heteropa (Chalcides) Lichtenstein, 1856 B. heteropa

intermedia (Bachia) Noble. 1921 B. intermedia

lineata (Bachia) Boulenger, 1903 B. heteropa

marcclae (Bachia) Donoso-Barros and Garrido, 1964 B. heteropa

monodactyhis (Chalcides) Daudin, 1802 B. monodactylus

pallidiceps (Brachypus) Cope, 1862 B. pallidiceps

panoplia (Bachia) Thomas, 1965 B. panoplia

parkeri (Bachia) Ruthven, 1925 B. vionodactyhis

peruanus (Cophias) Werner, 1901 B. peruana

schlegeli (Chalcides) Dumeril and Bibron, 1839 B. favescens

scolecoides (Bachia) Vanzolini, 1961 B. scolecoides

falpa (Bachia) Ruthven, 1925 B. taJpa

tridactylus (Chalcides) Daudin, 1802 B. favescens

trilineatus (Chalcides) Peters, 1872 B. monodactyhis

trinitatis (Scolecosaurus) Barbour, 1914 B. heteropa

trisanale (Ophiognomon) Cope, 1868 B. trisanale

vermiformis (Propus) Cope, 1874 B. trisanale

that may be of importance in separating
the two genera is the absence of the
frontonasal scale in Ophiognomon. How-
ever, certain specimens of Bachia peru-
ana either lack or have a very small
frontonasal scale, with the nasal scales
enlarged and in contact behind the ros-
tral. In Ophiognomon, the frontonasal
space has been filled by the expansion of
the nasal and frontal scales. This trend
is also evident in Bachia, thereby lending
support to the idea that Ophiognomon
and Bachia are congeneric.

All essential characters of scutellation
of Ophiognomon (scales around mid-

body, dorsals, ventrals, gulars, presence
or absence of contact betsveen labial and
parietal and between second pair of chin
shields) fall within the variation exhib-
ited by Bachia. Thus, I propose that
Ophiognomon Cope (1868) be placed in
the synonymy of Bachia Gray (1845).

The Status of Anotosaura Amaral and
Heterodactylus Splx

A survey of the major museums in
Europe and the United States revealed
the existence of only 15 specimens of
Heterodactylus. This genus is comprised
of two species, H. imbricatus Spix ( 1825)

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

and H. hindii Reinhardt and Lutken
(1861); neither type seems to be extant.
IBoulenger (1885) gave detailed descrip-
tions of both species, and I have exam-
ined 14 of 15 specimens (11 imbricahis,
3 hmdii).

Amaral (1933) indicated that his new
genus Anotosaura, type species coUaris,
was distinct from HeterocJactijlus in hav-
ing the nostril in the center of the nasal
scale, presence of prefrontals, smooth
dorsal scales, absence of occipital scales,
and relatively short limbs. I have been
unable to examine the unique specimen
of A. coUaris, but apparently all three
genera (Anotosaura, Bachia, Heterodac-
tijlus) arc closely related (Table 2). The
teeth on the dentary of Hetewdactylus
are bicuspid or tricuspid (linear) and
are considerably larger than those of
Bachia. In addition, Heterodoctylus has
lanceolate scales on the body and tail
that are four to five times longer than
wide, strongly keeled, and quite different
than most species of Bachia. Only Bachia
bresslaui, B. panoplia and B. scolecoides
approach the latter condition. All three
species are smaller in snout-vent length
than Heterodactyhis, and lack the small
occipital and nuchal scales present in
Heterodactylus (Fig. 1). Bachia pano-
plia has lanceolate, keeled, ventral body
scales, whereas Heterodactyhis, Anoto-
saura, Bachia bresslaui, and B. sco-
lecoides have quadrangular, smooth ven-
trals. All three species of Bachia show a
reduction in the number of toes on the
limbs (4-4, panoplia, scolecoides: 1-1
bresslaui) whereas Heterodactylus lias a
digital number of 4-5 or 5-5. However,
one may find as much scale variation
among species of Bachia as between
Bachia and Heterodactylus.

Bachia panoplia and B. scolecoides
seem to be closely related to Heterodac-
tylus, with B. bresslaui a slightly more
advanced form, modified for burrowing.
However, without concrete osteological
and myological data to corroborate these
relationships, I propose retaining Heter-
odactylus and Anotosaura as separate
genera until an anatomical study can be

Heterodactylus imbricatus BMNH K.")-21t)l

Figure 1. Lateral, dorsal, and ventral view of

the head, dorsal view of the limbs and ventral

view of the pelvic region of Heterodactylus

imbricatus.

completed. This suggested study is pres-
endy underway by Dr. William Presch,
University of California, Berkeley. The
general distribution of the three genera
is given in figure 2.

Methods and Materials

The methods utilized in the analysis
of variation of squamation in populations
of Bachia consisted of standard tech-
niques for determining the mean, stand-
ard deviation and standard error. Char-
acters examined for statistical analyses
were as follows: 1) number of scales
around middle of body, counted midway
between axilla and groin (=SAB); 2)
number of dorsal scales from the occiput
to above vent (= dorsals); 3) number of
ventral scales between pectoral shield
and preanal shields ( = ventrals ) ; 4 )
number of ventral scales between collar
fold (immediately behind angle of jaw)
and pectoral shield (regulars); and 5)
number of longitudinal rows of quad-
rangular ventral scales. The number of
supraoculars was counted for each side
and written 2-2, 2-3, or however they
varied. Similar counts were made of
supercilliaries, labials, toes, femoral and/
or preanal pores. The number of preanal
shields usually varied according to how

DIXON: SYSTEMATIC REVIEW OF BACHIA

9

Table 2. A comparison of the essential external features of the microteiid genera
Anotosaura, HetewdactyJus and Bochia (data on Anotosaura taken from Amaral,
(1933). Abbreviations are H, L, Q, K, S — hexagonal, lanceolate, quadrangular,

keeled, smooth; — = absent; + = present.

Character Anotosaura Heterodactylus Bachia

Maximum snout-vent length 37 mm 114 mm 107 mm

Tooth type ? tricuspid simple cone

Body scale type H,S L,K L,H,Q,K,S,

Femoral pores 2-2 3-3 to 5-5 0-0 to 2-2

Preanal pores 2-2 1-1 to 2-2 0-0 to 2-2

Forelimb toes 5-5 4-4 to 5-5 1-1 to 4-4

Hind limb toes 5-5 5-5 0-0 to 4-4

Prefrontals + — -\

Interparietal -\- -\ H

Enlarged nuchals — + —

Chin shields 3 pr 2 pr 2 pr

Pairs of chin shields in

medial contact 3 1 or 2 1 or 2

Pairs of chin shields reaching

oral border 3rd — — or 2nd

Supraoculars 3-3 3-3 0-0 to 3-3

Superciliaries 2-2 3-3 to 5-5 1-2 to 3-4

Scales around midbody 23 28-36 22-47

Dorsals 25 34-41 33-56

Ventrals 18 24-27 24-43

Gulars 6 6 6-10

Supralabials 6 5 to 6 4-4 to 6-6

Infralabials ____ . 5 4 to 6 4-4 to 6-6

Preanal shields 9 3-5 3-6

External ear — — —

Collar fold + +, weak -f

enlarged scales were arranged (i.e.,
longitudinal or transverse). When three
shields are present, the division is longi-
tudinal; when four or five are present
the division is transverse, and the middle
shield is usually divided into halves or
into thirds.

The shape of the body scale was uti-
lized to determine group relationships.
All populations leaving lanceolate,
keeled, imbricate scales are placed in the
bresslaiii group, and those with all quad-
rangular, smooth, juxtaposed scales are
placed in the favescens group. Numbers
of hexagonal, smooth, and imbricate

scales present on the dorsum and sides
of the body distinguish the dorbigmji
and hetewpa groups. In this study, the
lanceolate scale is one that is usually
keeled, four to five times longer than
wide, and with strongly pointed proxi-
mal and distal ends. The hexagonal scale
is smooth, and two to three times as long
as wide with somewhat rounded or ob-
tusely pointed ends. The quadrangular
scale is smooth and usually rectangular.
The presence or absence of certain
scales were noted and used to determine
interpopulational relationships. Such ob-
\'ious scales as prefrontals, frontonasal,

10

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

Slolult mill)

l_i_J L-

FiGURE 2. Distribution of the closely related microtciid genera Anotosaura, Bachia and Heterodac-

tijlus in South America and Panama.

DIXON: SYSTEMATIC REVIEW OF BACHIA

11

interparietal, supraoculars, superciliaries,
and their size, position, or absence are
useful in determining generalized and
advanced conditions for certain popula-
tions. The reduction in number of toes,
and the size or absence of limbs are also
used in assessment of generalized or ad-
vanced conditions.

Reduction in size and number of toes
is difficult to determine. The feet are
very small in BacJiia, and any reduction
in size or loss of toes can only be deter-
mined by the use of a high power,
binocular dissecting microscope. The
terminal ends of degenerate toes vary
from a single apical scale, a divided
apical scale, or two to three distinct api-
cal scales covering minute toes. When-
ever a single apical scale is present, the
limb itself is much reduced, styloform,
or tubercular in appearance.

The position and size of chin shields
are also utilized in determining inter-
population relationships. The first pair
of chin shields is always in broad contact
medially. The second pair may or may
not be in contact medially and in most
species, usually does not reach the oral
border. In only one specimen is the sec-
ond pair in contact medially and also
reaching the oral border.

Little variation is evident in the num-
ber of labials, but contact of the fourth
or fifth supralabial with the parietals is
an important character in some species.
The general shape of the snout is uni-
form for most species, but B. talpa and
B. bicolor have depressed and somewhat
protiaiding rostrals. The bodies of all
Bachia are somewhat compressed dorso-
ventrally, extremely long, and the fore-
and hind limbs positioned far apart. The
tail is usually one and one-third to two
times longer than the body. No ratios
were taken on the length of the body
compared to the length of the head or
limbs because of the poor preservation
of many of the specimens.

All major museums in the United
States, Europe, and South America were
canvassed for their holdings of Bachia
and Heterodactylus. Some 651 specimens

of Bachia and Heterodactylus were lo-
cated, and 520 of these were examined.
Three partial skeletons were examined,
one skull, and one cleared and stained
specimen.

Specimens were borrowed from the
following museums and individuals:
American Museum of Natural History
(AMNH), Academy of Natural Sciences,
Philadelphia (ANSP), British Museum
(Natiual History) (BMNH), California
Academy of Science (CAS), Field Mu-
seum of Natural History (FMNH), Flor-
ida State Museum, University of Florida
(UF), University of Kansas Museum of
Natural History (KU), Los Angeles
County Museum of Natural History
(LACM), Rijksmuseum van Natuurlijke
Historic, Leiden (RMNH), Louisiana
State University Museum of Zoology
(LSUMZ), Museum of Comparative Zo-
ology, Harvard (MCZ), Merriam Mu-
seum, University of Texas, Arlington
(MM), Museum of Vertebrate Zoology
(MVZ), Museu de Zoologia, Universidade
da Sao Paulo (MZUSP), Museum Na-
tional d'Histoire Naturelle, Paris
(MNHN), Texas Cooperative Wildlife
Collection (TCWC), University of Mich-
igan Museum of Zoology (UMMZ),
United States National Museum
(USNM), James A. Peters (JAP), Rich-
ard Thomas (RT).

Acknowledgments

I thank Howard Campbell, William
Duellman, Alice Grandison, Hany
Greene, Jean Guibe, Marinus Hoogmoed,
Alan Leviton, Edmond Malnate, Hymen
Marx, James Peters, Douglas Rossman,
Richard Thomas, P. E. Vanzolini, David
Wake, Charles Walker, Ernest Williams,
John Wright and Richard Zweifel for
permitting me to examine specimens in
their care.

I also thank William E. Duellman,
William Presch and P. E. Vanzolini for
criticizing the manuscript. This paper
would not have been completed without
the aid of Fred Henchicks, James Kroll
and Carl Lieb.

12

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

EVOLUTIONARY TRENDS IN THE GENUS BACHIA

The generalized scale pattern in
Bachia is one in which the dorsal scales
are lanceolate, keeled, and imbricate,
and the ventrals are usually quadrangu-
lar and smooth. The head shields consist
of a normal complement of nasals,
frontonasal, frontal, paired parietals, in-
terparietal, 3 supraoculars and 3 super-
ciliaries. The limbs are moderately de-
veloped for walking, with four toes on
each limb. In slightly specialized spe-
cies, the scale pattern consists of wider,
smoother hexagonal scales, which are
less pointed and imbricate. There has
been no change in the number of head
shields, or toes, but the limbs are some-
what shorter and less developed. In more
specialized species, the body scales be-
come more quadrangular in shape in
some groups; head shields are succes-
sively lost with a corresponding increase
in the size of the nasals, frontal, and
frontonasal. The trend towards reduc-
tion is marked by disappearance of the
prefrontals first, followed by loss of the
interparietal, supraoculars, toes and
eventually, the frontonasal shield and the
hind limb.

The reduction in number of head
shields, toes, and eventual loss of the
frontonasal and hind limb are adaptive
modifications for fossorial habits. The
successive loss of head shields and toes
tends to follow distinct and independent
patterns within each of the species
groups (hresslaui, dorhigniji, heteropa,
flavescens). Although each group has its
complement of generalized species, there
is a definite trend towards fossorialness
in at least one of its species or sub-
species.

However, one must not lose sight of
the genetic variabilit}' in the various pop-
ulations of Bachia that might give rise to
variations yet to be explored by evolu-
tionary pathways. One population may
lose its supraoculars, interparietal and
prefrontals prior to loss or reductions in
size and number of toes and limbs;
whereas another population may retain

its full complement of head shields but
have loss or deduction in numbers of toes
and limbs. Modification of body scales
{i.e., becoming quadrangular and juxta-
posed, allowing for greater flexibility in
body movement) and atlas-axis-cervical
vertebrae {i.e., for stronger vertical thrust
in head movements for burrowing
through leaf litter) may be the primary
method of evolutionary modification for
fossorial existence in Bachia. Unfortu-
nately, few specimens are available for
examination of the latter character. The
rate of evolutionary change in one set of
characters for one set of environmental
conditions may or may not aff^ect the
same rate of change in other characters
{i.e., modification of neck vertebrae for
more vertical thrust, but retention of
limbs and toes with unmodified pelvic
girdle ) , as it were, carrying out one evo-
lutionary modification but not "placing
all of its eggs in one basket."

Based upon body scale patterns (i.e.,
lanceolate, hexagonal, quadrangular,
(Fig. 3a) Bachia seems to have evolved
from a Heterodactijhis-hke ancestor. The
hresslaui group (hresslaui, scolecoides,
panopUa) all show some relationship to
Heterodactijhis {i.e., lanceolate, keeled
dorsals), with B. scolecoides appearing
more closely related to Heterodactijhis
than the other two species of the group.
All three species share body scale type,
presence of an interparietal, and femoral
pores in common with Heterodactijhis.

The ancestor to this group is the
probable progenitor to all species of
Bachia. Two species appear to be supra-
terranean, leaf litter inhabitants (scole-
coides, panoplia), whereas the presence
of a reduced head scale number and de-
generate limbs in the third (hresslaui)
suggest that it is fossorial. Neither life
history information nor large series of
individuals are available for study.

On the basis of the presence of an
interparietal, prefrontals, and number of
toes, the nearest relative of the hresslaui
group would be the heteropa group.

DIXON: SYSTEMATIC REVIEW OF BACHIA

13

A

Dorsal

Ventral

1

— <

1 —

1

1

<

\

' \

—

1 —

Bresslaui
Group

Dorbignyi
Group

Heteropa
Group

la

11.

. 1^, ' ' 1,1': '

14

15

ft

D^ 1 , ' I I ' ' ' ' 1-1

/ ^ /

5. intermedia

6. dorbignyi

7. barbouri

8. talpa

9. huallagana

10. peruana

11. trisanale
Heteropa Group (H)

12. heteropa

13. pallidiceps
Flavescens Group (F)

14. flavescens

15. monodactylus

Figure 3. A. Scale pattern of the four species groups of Bachia. B. A proposed
phylogenetic arrangement of the genus Bacliia.

Flavescens
Group

Bresslaui Group (B)

1. scolecoides

2. panoplia

3. bresslaui
Dorbignyi Group (D)

4. bicolor

However, the type of body scales in the
heteropa group is quite unHke that in
the bresslaui group. The dorsal scales of
the heteropa group are hexagonal,
smooth and imbricate, and the lateral
and ventral scales are quadrangular,
smooth and juxtaposed. The dorbignyi

group has dorsal and lateral hexagonal
scales, smooth and imbricate, and only
the ventrals are quadrangular and juxta-
posed.

Collectively, the dorbignyi group
shows a general reduction in number of
head scales and toes, and a tendency

14

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

toward fossorialness. Bachia hicolor ap-
pears to be the most generalized species
of the group, with B. trisanale the most
advanced. However, there are variations
within each species toward fossorial
adaptations that mask tlieir interspecific
phylogenetic relationships.

Most species of the dorhigmji group
have lost the prefrontals, supraoculars
and interparietal (except harhomi), and
in one case (trisonale) the frontonasal.
In addition, the latter group shows a uni-
form trend toward reduction in number
of toes on the forelimb.

The Jieteropa group seems to have
given rise to the favescens group, shar-
ing several characters of the head and
body squamation. The primary differ-
ences between these groups is the pres-
ence of all quadrangular body scales and
complete loss of the prefrontals in the
latter group. My hypothesis of the phy-
logenetic relationships of Bachia and
Heterodactylus is shown in figure 3b.

Bachia Gray 1845

Chalcides Bonnaterie, 1789:66 [Type species. —
Chalcides ftavescens Bonnaterre, 1789, (pre-
occupied by Chalcides Laurenti, 1768)].
Chalcis Merrem, 1820:75 [Type species. — Chal-
cides tridoctylus Daudin, 1802. (preoccu-
pied by Chalcis Fabricus, 1787)].
Colohus Men-em, 1820:76 [Type species. — Seps
monodactiihis Daudin, 1802, (preoccupied
by Colohus Illiger, 1811)].
Brachypus Fitzinger, 1826:20 [Type species. —
Brachypus citvieri Fitzinger, 1826, (pre-
occupied by Brachypus Swainson, 1824)].
Cophias Fitzinger, 1813:20 [Type species. —
Chalcides dorhignyi Dumeril and Bibron,
1839, (preoccupied by Cophias Merrem,
1820)].
Microdactijhts (Tschudi MS) Agassiz, 1844:28,
not of Fitzinger, 1843, (preoccupied by
Microdactijhts GeofFroy, 1809).
Bachia Gray, 1815:58 [T>'pe species. — Chal-
cides dorhignyi Dumeril and Bi])ron, 1839].
Ophiognomon Cope, 1868:100 [Type species. —

Ophiognomon trisanale Cope, 1868].
Hapalolepis Peters, 1871:399 [Type species. —
Chalcides (Hapalolepis) ahendrothii Peters,
1871].
Propus Cope, 1874:70 [Type species. — Propus
verminformis Cope, 1874, (preoccupied by
Propus Oken, 1816)].
Herpetochalcis Boettger, 1883:150 [Type spe-

cies.— Chalcides heteropus Lichtenstein,
1856].

Scolecosaurus Boulenger, 1885:416 [Type spe-
cies.— Brachypus cuvieri Fitzinger, 1826].

Sesquipes Cope, 1896:466 [Type species.—
Chalcides heteropus Lichtenstein, 1856].

Heteroclonium Cope, 1896:466 [Type species.
— Heteroclonium hicolor Cope, 1896].

Anisoclonium Cope, 1900:561 [Type species.—
none designated].

Apatelus Amaral, 1935:249 [T>pe species. —
Apatelus hresslaui Amaral, 1935, (preoccu-
pied by Apatelus Mulsant and Rey, I860)].

Generic Definition. — The microteiid
genus Bachia comprises those popula-
tions of lizards having elongate snake-
like bodies and tails, and short limbs;
external ear absent; eyes reduced in size;
lower eyelid with unsegmented window;
ventral scales quadrangular, juxtaposed
(except one species); hind limb fre-
quently reduced to a styloform femur
and tibia, often without visible toes and
entire hind limb may be completely ab-
sent (for variation in squamation see
table 3).

The osteology of the genus Bachia
demonstrates adaptive modifications for
burrowing. The following osteological
description is based upon four members
of the dorhigmji group (barbouri, hi-
color, intermedia, trisanale).

Camp (1923) stated that the follow-
ing conditions prevail in Bachia: Zygo-
sphenal articulation supplementing ordi-
nary zygapophyseal type is rudimentary
or absent; three cervdcal ribs; interclav-
icle reduced to longitudinal bar; simple
clavicles; none to one sternal fontanelle;
scapular fenestra absent; first and second
scapulo-coracoid fencstrae present and
closed; 13 pairs of parasternal ribs; hyoid
arch lacking second epibranchial; second
ceratobranchial present.

In addition to Camp's (1923) obser-
vations, the following osteological fea-
tures are characteristic of Bachia: 45
presacral vertebrae; three pairs of sternal
ribs; one pair of xiphisternal ribs; full
complement of pelvic elements in ter-
restrial species, reducing to single fusion
of ilium, ishium and pubis in fossorial
species lacking hind limbs; phalangeal
formulae highly variable depending on

DIXON: SYSTEMATIC REVIEW OF BACHIA

15

C/3

1—1

CD

lO

o

t-

00

t^

CM

CO

in

t^

CM

^

Ol

1

I-H

xfl

1—1

CO

r— 1

1—1

1

CM

CM

CM

1 ^1

>

o

CD

1

d

d

d

d

d

d

1

d

d

d

' d

+1

+1

+1

+1

+1

11

+1

+1

+1

+1

+1

+1

00

00

lO

oq

t^

in

00

1—1

^

CTi

in

CO

C/2

•l-H

xo

t^

1

1 — 1

00

CM

in

CO

t-

1

00

1 — 1

I-H

1 ^

(U

C/5

1-H

d

1

1—5

1— ^

rH

I— i

I— 5

d

1

1— i

CM

i-H

' ^

Q

Q

Q

Q

Q

Q

Q

Q

Q

Q

Q

Q

2

a)

C/2

CD

CD

CD)

CD

CD

CD

CDl

CD

CD

CD

CD

>

o

00

lO

l>

1—1

o>

CM

t^

O

CO

l^

O

CO

CO in

ai

CO

t-^

CO

t^

ai

t-^

in

CM

l>

t>

1>^

I-H

CO in

CO

CO

CO

CO

CO

CO

CO

CO

^

CO

CO

CM

CO

CO CO

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX IX

Q

cq

00

00

05

o

CM

CM

00

CO

00

CM

CO

^

^ 2J

Tf

CO

CO

CO

rr

TT

t

CO

Tf

CO

1

CO

CO

1

CO CO

1 1

?j

CD

CD

t-!-

Tfl

^

t-!-

in

CO

r— 1

CO

-^

4l

o

CO CO

CO

CO

CO

CO

CO

CO

CO

CO

'^

CO

CO

CM

CO

CO CO

MH

1/5 >H

Gi

t^

00

lO

a^

1—1

b-

Tf

CO

CM

CO

'^

.gfe

CM

CO

1

1—1

"^

1—1

Tf

1—1

l—l

I

CM

CM

CO

1 ^

d

d

1

d

d

d

d

d

d

1

d

d

d

' o

o tj

e sp
ndar

+1

+1

+1

+1

+

+

+1

+1

+

+

+1

+1

'^

1— (

CO

o

00

05

CD

1—1

Tf

CM

in

*^

45 03

1— I

I— I

1

CO

r—t

CO

CD

CO

00

1

00

05

CO

1 ^

-4-> +J

V3

w

rH

i-H

1

1— i

oi

1— i

1— i

I-H

d

1

1-H

1-H

I— ^

' 1-1

o -0

13

S2

Q

Q

Q

Q

Q

Q

Q

Q

Q

Q

Q

Q

o

CD

CD

CD

CD

CD

CD

CD

CD

CD

CD

CD5

CD

Q

15 r-"

CO

^H

o

O

t-

05

CM

Tf

CD

CO

CO

CD

CO

CO CO

^.2

05

d

00

oi

ai

oi

ai

ai

-^

05

d

1-H

in

i> ai

TT

lO

^

-*

^

lO

^

^

in

"^

in

Tf

Tf

TJH -^

Oj --H

>X

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX IX

■S ^

(M

CM

05

Ol

CO

lO

"^

CM

CD

1—1

in

05

00

00 CO

s ^

lO

lO

tT

lO

lO

in

in

in

in

in

in

^

T

^ in

1 1

> ^

1

00

b^-

CD

CO

d>

i>-

CD

CO

i>-

i>-

00

CO

CD l>

^

Tt^

^

'^

^

lO

TT

^

in

'^

^

CO

^

Tf TJH

— rt

tT

lO

lO

Cvl

CO

1—1

t-

Tf

CO

l>

CM

<=^

S t^

Ol

^

1

I— I

(M

1—1

"^

rH

.— I

1

CM

I-H

CN

1 f>^.

d

d

d

d

d

d

d

d

d

d

d

o

>^

+

1

+

1

+

+

+

+

1

+

1

+1

o

Tt<

Tt^

I— 1

05

b-

00

CD

05

b-

CM

I-H

00

"S >^

PQ

CT>

CO

1

^-

05

1—1

CO

CO

t-

1

00

in

05

1 <^

>.-Q

T3

CD

.— i

1—

d

I— H

1-H

i-H

d

1-H

1-H

d

1-H

1^

1—'

o

Q

Q

Q

Q

Q

Q

Q

Q

Q

P

Q

Q

f %

<

en

CD

CD)

CD

CD

CD

CD

:n

CD

CD

CD

CD)

ound
folio

t~-

t-

o

00

c:5

CO

1— t

P

1—1

o

CM

Tf

05

p Ol

t^

^

-^

00

00

-^

CD

1—5

in

CD

CO

ocf

C<

I-H C35

CO

^

CO

CM

cq

CM

CM

CO

CM

CM

CM

CM

CM

CO CM

^

CO

rt

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

IX

"X

IX

IX IX

5fi

o

l>

lO

1 — 1

1—1

CO

o

in

CD

t-

o

i—H

in

CM CO

0

t

^

CO

CO

CO

CM

CO

CO

^^

Ol

CO

CO

CM

CO CO

1 1

13

CD

CO

CO

CO

1>

CO

Tf

CO

CO

in

CO

4

^

2 <=^

CO

^

CO

CM

CM

CM

CM

(M

CM

CM

CM

(M

CM

CO CM

•5

O

:?:

ITJ

CM

(M

t-

,_^

CO

00

CO

00

CO

CM

CM

OC

CO o

i-H

.— 1

lO

cq

xo

^^

CD

CO

t>

00

I— ^

in

OS

g

?o

p

^

C/3

Cr

e

^

0:

;;?

CO

Or

1

S
J

C
C.

i-C

**•

-5:

5

' 1

c:)

o
o

16

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

fossorial adaptations in any particular
species or populations of a species; atlas
and axis modified for burrowing in the
species examined, with atlas composed of
three separate and distinct units (single
median hypocentrum articulating with
paired, unf used neural arches ) ; axis pos-
sessing posteriorly projecting hypopohy-
sis upon which long^us colli muscles insert;
neural spine of axis notched posteriorly
to receive neural spine of third cervical
vertebra; cranial elements somewhat
modified for burrowing with broad pre-
maxillary in contact with frontal, sepa-
rating paired nasals; postorbital and
postfrontal unfused and generally paral-
leling each other, not in close contact
with jugal; postfrontal extremely long
and thin, extending from orbit to pos-
terior edge of fused parietals; quadrate
short and stout; stapes large and plate-
like; 10 scleral ossicles around each eye;
7-9 premaxillary teeth, 16-20 maxillary
teeth, and 22-26 dentary teeth (totaled
for both sides ) .

Content. — As defined here, the genus
Bachia contains 15 species, of which four
are polytypic, containing a total of 12
subspecies; the genus is divided into four
species groups.

Distribution. — Bachia occurs in the
lowland tropics and semi-arid foothills
from northwestern Panama southward to
Gran Chaco, Paraguay. The genus occurs
on Grenada, Grenadines, Trinidad, and
Tobago, but is absent from the Pacific
lowlands of South America (Fig. 2). Most
records of occurrence are at elevations of
less than 500 m, but one species is known
to occur at elevations exceeding 1,500 m
in the Tulumayo Valley of southcentral
Peru.

Key to Species

The following key to the species of
Bachia is based primarily on the various
scale characters of the groups presented
previously. The number of toes is not
reliable in some species of Bachia; there-
fore, the number of toes is used only
once in a primary couplet.

1. All midbody scales quadrangular 2

All midbody scales not quadrangular
4

2. Interparietal scale absent 3

Interparietal scale present

fiavescens

3. Hind limb as long or longer tlian
preanal shields, scales around mid-
body 26 or more monodactylus

Hind limb less than half the length
of preanal shields, scales around
body 24 species Q

4. Ventral scales quadrangular, smooth
5

Ventral scales lanceolate, keeled
panoplia

5. Dorsal scales lanceolate, keeled - 6
Dorsal scales hexagonal, smooth __ 7

6. Prefrontals present, fore and hind

limb toes 4-4 scolecoides

Prefrontals absent, fore and hind
limb toes 1-1 hresslaui

7. Lateral body scales hexagonal 8

Lateral body scales quadrangular
15

8. Interparietal scale absent 9

Interparietal scale present _ barhouri

9. Frontonasal scale usually present
10

Frontonasal scale usually absent ___.
trisanale

10. Second pair of chin shields not in
contact medially 11

Second pair of chin shields in con-
tact medially 13

11. Three toes on the forelimb 12

Four toes on the forelimb - bicolor

12. Second pair of chin shields separate
third and fourth infralabials and
reach oral border, five temporal
scales talpa

Second pair of chin shields do not
separate third and fourth infralabials
nor reach oral border, three temporal
scales huaUaiiana

DIXON: SYSTEMATIC REVIEW OF B AC HI A

17

13. Two supercilliaries, a supralabial

touching the parietal 14

Three supercilharies, no supralabial
touching parietal intermedia

14. Fourth supralabial touching parietal
peruana

Fifth supralabial touching parietal
dorhiii^mji

15. Two supercilliaries, SAB number
usually less than 24 (x = 22.9)
pallidiceps

Three supercilliaries, SAB number

usually more than 24 (x = 28.4)

heteropa

SYSTEMATIC ACCOUNTS

The BacJiia bresslaui Group

The bresslaui group is composed of
three species, bresslaui, scolecoides and
panoplia, none with recognized races.
This group occupies the Amazonian
drainages of the Rio Amazonas (pano-
plia), Rio Teles Pires (scolecoides), and
the Rio Juruena (bresslaui), Brasil.
Bachia panoplia is known only from the
immediate vicinity of Manaus, Brasil, but
may occur in one or more of the nearby
river drainages that empty into the Rio
Amazonas. All three species are allo-
patric with each other, but panoplia is

sympatric with monodactijlus of the
flavescens group.

The bresslaui group comprises those
populations having lanceolate, keeled,
imbricate dorsal and lateral body scales
(including the ventrals in panoplia), with
c|uadrangular, juxtaposed ventrals; 2-2
femoral pores and 1-1 preanal pores in
males; interparietal, supraoculars, and
superciliaries present; prefrontals pres-
ent (fused to anterior supraocular in
bresslaui); digits normally four on each
limb (one in bresslaui) .

Bachia bresslaui (Amaral)
Figure 4

Apatelus bresslaui Amaral, 1935:250 [Holotype
IB525, from the interior of tlie state of Sao
Paulo, Brasil, collector unknown].

Bachia bresslaui — Vanzolini ( 1961b:200).

Range. — Known only from Utiariti,
Rio Papagaio (tributary of Rio Juruena)
Mato Grosso, Brasil (and possibly the
state of Sao Paulo, Brasil, Vanzolini,
1966) (Fig. 5).

Diagnosis. — Dorsal and lateral body
scales lanceolate, keeled, imbricate; ven-
tral scales quadrangular, juxtaposed; all
tail scales lanceolate, keeled, imbricate;
prefrontals absent (fused to anterior
supraocular); interparietal, frontonasal,
nasals, parietals present; supraoculars
2-2; superciliaries 2-2; second pair of

B bresslaui

B scolecuides AMNH S6;i

B. panuplia AMNH till^TS

FiGUBE 4. Lateral, dorsal, and ventral views of the head, dorsal view of tlie limbs and ventral views
of the pelvic region of B. bresslauii, B. scolecoides, and B. panoplia.

18

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

Figure 5. Distribution of B. dorbignyi, B. peruana, B. Jiuallagana, B. intermedia, B. hressJaui, B.
scolecoides and B. panoplia in west central South America.

chin shields in contact medially, not
reaching oral border; femoral pores 1-1
or 2-2; preanal pores 1-1; anterior and
posterior limbs reduced with a single
apical scale resembling a toe; preanal
shields 6 (not including 2 small lateral
scales); SAB 33-35; dorsals 47-49; ven-
trals 38; gulars 9; maximum snout-vent
length 75 mm.

Variatior}. — I have examined one
(MZUSP 10300) of the two known speci-
mens. Based on the original description
(Amaral 1935), Vanzolini's (1916b) re-
definition of the type, and Vanzolini's
(1966) description of the second known
specimen, there is little evidence of vari-
ation in this species.

Remarks. — This species is the most
advanced member of the bresslaiii group,
based on the reduction of toes, size of

limbs, and head scales. Although no in-
formation on the natural history is avail-
able, the nature of its digits suggests a
fossorial existence.

Bachia scolecoides Vanzolini
Figure 4

Bachia scolecoides Vanzolini, 1961b:202 [Holo-
type.— MZUSP 3293, from Rio Teles Pires,
Mato Grosso, Brasil; Mr. H. Sick collector].

Ratline. — Known only from the type
locality (Fig. 5).

Diagnosis. — Dorsal and lateral bod)'
scales lanceolate, keeled and imbricate;
ventral scales quadrangular and juxta-
posed; all tail scales similar to dorsals;
prefrontals present but without medial
contact; interparietal, frontonasal, na-
sals and parietals present; supraoculars

DIXON: SYSTEMATIC REVIEW OF BACIHA

19

2-2; superciliaries 3-3; second pair of
chin sliields in medial contact bnt not
reacliing oral border-; femoral pores 2-2;
preanal pores 1-1; anterior and posterior
limbs moderately developed with four
toes on each foot; preanal shields 3-4;
SAB 36-40; dorsals 47-52; ventrals 36-42;
gulars 8; maximum snout-vent length 78
mm.

Variation. — Little variation was noted
in the type series (Vanzolini, 1961b).
Females xaried in snout-vent length from
50 to 69 mm males from 48 to 78 mm.
The number of supraoculars (2-2), super-
ciliaries (3-3) and toes (4-4) were con-
stant in the sample. The transverse rows
of gular scales are usually 8 (9 in one
specimen); enlarged supralabials 5-5, in-
fralabials 4-4; labials not in contact
parietals: SAB are 36-40 (x 37.7); dorsals,
47-52 (x 49.6); ventrals, 36-42 (x 39.0).

Coloration. — The dorsal color pattern
of body and tail consists of three black
to dark brown lines (one median, two
paravertebral) separated by light tan to
yellowish brown interspaces. The sides
are dark brown to black with the venter
lighter than the sides. The dark dorsal
lines are present on the head, but may
be broken into a series of irregular lines
or spots.

Remarks. — This species is known
only from the type series ( 15 specimens).
Natural history information is lacking.
Additional collecting along and between
the Rio Teles Pires and Rio Juruena may
reveal intermediates or hybrids between
bresslaui and scolecoides.

Bachia panoplia Thomas

Figure 4

Bachia panoplia Thomas, 1965:18 [Holotype
AMXH 64877, from Manaos, .\mazonas,
Brasil; E. T. Gilliard collector].

Range. — Known only from the type
localit)' (Fig. 5), and Timbo, Comisaria
de Vaupes, Colombia.

Diagnosis. — Dorsal, lateral and ven-
tral body and tail scales lanceolate,
keeled and imbricate; prefrontals with
medial contact; interparietal, frontonasal,

nasals and parietals present; supraoculars
3-3; superciliaries 3-3; second pair of
chin shields in medial contact but do not
reach oral border; femoral pores 2-2; pre-
anal pores 1-1, absent in females; anterior
and posterior limbs moderately devel-
oped with four toes on each limb; pre-
anal shields 5 to 6; SAB 43-47; dorsals
48-52; ventrals 36-38; gulars 8; maximum
snout-vent length 85 mm.

Variation. — Little variation was noted
in the number of head shields. Only one
specimen had a 3-4 combination in su-
perciliaries. Four toes on the forelimb
were constant and three toes on the hind
limb occurred twice in 10 specimens.
The gular rows were eight in all speci-
mens. The preanal shields are 5-6 (x 5.3)
(one small scale on each side not
counted); SAB are 43-47 x 44.7, SD
1.34 ±0.45; dorsals, 48-52 (x 50.1, SD
1.10±0.37); ventrals, 36-38 (x 36.8, SD
0.78±:0.26).

Coloration. — Dorsal color pattern usu-
ally consists of a wide olive brown, mid-
dorsal stripe (9-11 scale rows wide),
extending from nape to tail; outer edge
of dorsal stripe occasionally dark olive,
forming an obscure dorsolateral dark
line; sides of body somewhat darker than
dorsum; venter pale yellow, or tan; head
with a large olive, lobate figure from
parietals to anterior edge of prefrontals;
posterior edge of each dorsal scale edged
with dark olive, giving an obscure "band-
ing" effect to each scale row.

Remarks. — This species is known only
from the type series, four additional
specimens from the type locality, and
one specimen from Colombia. No natural
history information is available. The
specimen from Colombia was removed
from the stomach of a snake, Pseudohoa
coronata. Additional collecting along the
southern rivers flowing north into the
Amazon may reveal sympatry between
panoplia and other members of the
group. Bachia monodactyltis has been
taken at Manaos but actual sympatry
between panoplia and monodactylus has
not been recorded. Based on the size of
the limbs and number of toes, B. pano-

20

MISCELLANEOUS PUBLICATION MUSEUM OF NATUR.\L HISTORY

plia appears to be a leaf litter inhabitant
while monodactylus is primarily fosso-

ri

al.

The Bachia dorbignyi Group

The dorbignyi group is composed of
eight species, one with three subspecies
(dorbignyi, barbouri, bicolor, hiiaUagana,
intermedia, peruana, talpa, trisanaJe tri-
sanale, trisanale ahendrothi, trisanale
vermiforme). This group contains the
largest number of species, of which the
majority are allopatric. Two species, bi-
color and talpa, are restricted to northern
and central valleys of Colombia, and
separated from other species of the group
by the northern Andes. The remainder
of the species ( except trisanale ) occur in
isolated river valleys along the eastern
edge of the Peruvian and Bolivian Andes.
Bachia trisanale is widespread in the
lowlands of Peru and Ecuador.

This group comprises those popu-
lations of Bachia having hexagonal,
smooth, imbricate scales on the dorsal
and lateral surfaces of the body and tail;
quadrangular, juxtaposed scales on the
ventral surfaces of the body; hexagonal
scales on the ventral surface of the tail,
although occasionally the proximal ven-
tral surface may have a few quadrangu-
lar scales; a styloform hind limb (except
barbouri and trisanale, where it may be
absent or tubercular); usually three toes
on each forelimb ( four in bicolor, two in
barbouri, one to three in trisanale);
supraoculars absent (occasionally one
present in intermedia); superciliaries
usually 2-2 (occasionally one to three);
usually fourth or fifth supralabial in con-
tact with parietal (no contact in inter-
media); one or two preanal pores in
males (sometimes very small, indistinct
in peruana); second pair of chin shields
usually not in contact with oral border
(in contact in talpa, barbouri, and occa-
sionally bicolor); second pair of chin
shields in medial contact (no contact in
talpa and huallagana), and occasionally
in trisanale); frontonasal shield present
(absent in trisanale, reduced in size or
absent in peruana); interparietal absent

(present in barbouri); prefrontals absent.

Bachia intermedia Noble

Figure 6

Bachia intermedia Noble, 1921:142 [Holotype.
— MCZ 14630, from Perico, Departamento
Cajamarca, Peru, G. K. Noble collector].

Range. — Known only from the Rio
Chinchipe and Rio Maranon valleys
from Perico on the north and Bellavista
on the south ( Fig. 5 ) .

Diagnosis. — Dorsal and lateral body
scales hexagonal, smooth and imbricate;
ventral scales quadrangular, juxtaposed;
tail scales similar to dorsals; nasals,
frontonasal, parietals present; nasals not
in contact medially behind rostral; pre-
frontals, interparietal, and supraoculars
absent; no supralabial /parietal contact;
forelimbs small with three toes; hind
limb styloform with one or two apical
scales resembling toes; superciliaries usu-
ally three; preanal pores 1-1 in males,
absent in females; SAB 28-35 (x 31.0);
dorsals 46-52 (x 49.4); ventrals 33-38 (x
35.7); gulars 7-9 (x 8.2); maximum
snout- vent length 67 mm.

Variation. — The head squamation is
fairly consistent. One of 63 specimens
has a small azygous scale between the
posterior medial contact of the parietals;
numbers of superciliaries are 3-3 in 56
specimens, 2-3 (3) and 2-2 (4); those
specimens with two superciliaries have a
small supraocular wedged between the
second and third superciliary, but fail-
ing to reach the orbital border; second
pair of chin shields in medial contact in
all specimens, but do not reach oral bor-
der; transverse rows of gular scales num-
ber seven in one specimen, eight (51),
nine (11); supralabials and infralabials
are 6/5 in all specimens; preanal shields
number three in 3 specimens, four (59),
five ( 1 ) ; longitudinal rows of quadran-
gular ventrals are 4-9 (x 6.2); number of
toes on forelimb are 3-3 in 62 specimens,
4-4 in one; hind limb always styloform
with single apical scale that may be
divided.

Coloration. — Dorsal color pattern a
broad, yellowish tan stripe about 10 scale

DIXON: SYSTEMATIC REVIEW OF BACHIA

21

Bachia intermedia LSl'M/ 193S3

Bachia barhoiiri TCWC 28908

B. bicolor USNM 8-1971

B. talpa UMMZ .',17

Figure 6. Lateral, dorsal, and ventral views of the head, dorsal view of the limbs of
B. intermedia, B. barbouri, B. bicolor and B. talpa.

rows wide from occiput to near tip of
tail, with three dark brown to black hnes
superimposed over the stripe; the me-
dian black line usually straight edged
and continuous, paravertebral lines usu-
ally a series of dashes; all stripes and
lines continue anteriad to frontonasal
scale; sides of head, body and tail black
to dark brown; venter dark brown, but

occasionally lighter than sides of body.

Remarks. — Loveridge (1933) gives
some individual variation data for a se-
ries of intermedia from the type locality
vicinity. Dixon ( 1972 ) found natural hy-
brids between intermedia and barbouri.
Neither species, following examination of
additional specimens, revealed additional
hybrids.

22

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

Bachia intermedia has been found on
dry foothill slopes and in slightly moist
conditions along river valleys. It has
been taken beneath fallen cacti, mesquite
trees, strangler figs, and grass clumps.
This species lays two eggs, probably dur-
ing November, December, and January,
judging from the egg maturity of females
taken during this period.

Burt and Burt (1931) indicate that
the series of intermedia paratypes they
examined had tvvo toes on each hind
foot. Actually the terminal scale is di-
vided and only one degenerate toe is
present beneath the scale.

Bachia barbouri Burt and Burt

Figure 6

Bachia harhouri Burt and Burt, 1931:318 [Hol-
otype.— AMNH 28437, from Perico, De-
partamento Cajamarca, Pen'i; G. K. Noble
collector].

Range. — Known only from the Rio
Chinchipe, Rio Marafion and Rio Utcu-
bamba \'allevs of northwestern Peru
(Fig. 9).

Diagnosis. — Dorsal and lateral body
scales hexagonal, smooth and imbricate;
ventral scales quadrangular, juxtaposed;
tail scales similar to dorsals; nasals,
frontonasal, interparietal, parietals pres-
ent; prefrontals and supraoculars absent;
frontonasal large, separating nasal scales;
fifth supralabial in contact with parietal
in 30 percent of the sample; forelimbs
small with two toes; hind limb usually
absent but occasionally a very small tu-
berculc, present, preanal pores 1-1 in
males, absent in females; SAB varying
from 26-31 (x 28.8); dorsals 46-52 (x
49.0); ventrals 34-39 (x 36.7); gulars 8-10
(x 8.9); maximum snout-vent length 68
mm.

Variation. — The head shields of this
species are remarkably constant. Pres-
ence of the interparietal, two supercili-
aries, six supralabials, five infralabials,
and second pair of chin shields reaching
oral border does not varj' in 57 speci-
mens. One of 57 specimens lacks medial
contact between second pair- of chin

shields; interparietal completely splits
parietals in 70 percent of sample; fifth
supralabial touches parietal on one side
in 6 specimens, both sides in 10, neither
side in 41; toes on forelimbs are 2-2 in 55
specimens, 1-2 in one, and 2-3 in one;
hind limbs absent in 56 specimens, a
small tubercule present in one. Preanal
shields number three in 3 specimens,
four (44), five (10); transverse rows of
gular scales number eight in 13 speci-
mens, nine (33), ten (11); number of
ventral longitudinal scale rows is 6-12
(x 8.7).

Coloration. — The dorsal head pattern
of B. barbouri consists of a large arrow-
shaped, black mark, beginning on the
anterior edge of frontal and extending to
the posterior edge of parietals; open end
of arrow facing posteriorly; usually me-
dian black spot or short line inserted be-
tween open end of arrow and nape;
dorsolateral and lateral surface of head
black; occasionally black spot in center
of frontonasal; rostral usually white but
occasionally black line extending around
rostral from nasal to nasal.

Dorsal pattern from nape to near tip
of tail consists of a cream to white stripe
about 10 scale rows wide, with three
longitudinal black lines superimposed
over light stripe; black lines usually dis-
continuous, broken into a linear series of
dots or dashes; sides black, venter black
or occasionally dark brown.

Remarks. — During 26-28 November
1968, John Wright and I collected a
series of barbouri from the vicinity of
Bellavista and Bagua Grande, Peru. One
specimen was taken beneath a fallen
Cerceus above the Rio Maranon flood
plain; 12 were found beneath drift debris
on the flood plain. Eleven others were
found beneath leaf litter and fallen cacti
near a small stream. An additional seven
were found under decaying agave rela-
tively high on a hillside. Richard Thomas
(pers. com.) found one in a termite nest.
Gravid females taken contained two
eggs in various stages of development.
One egg, ready to be deposited, meas-
ured 4.7 x 15.4 mm.

DIXON: SYSTEMATIC REVIEW OF B AC HI A

23

Natural liybridization between ])or-
bouri and intermedia was noted by
Dixon (1972). The hybrids were taken
beneatli rotten, moist logs in xeric serul)

\egetation.

Bachia bicolor (Cope)
Figure 6

Heterodonmm bicolor Cope, 1896:466 [Syn-
t\pes. — AMNII 9544-45 from \icinity of
Bogota, Departamento Cundinamarca, Co-
lombia; collector unknown].

Heterodonmm bicolor Cope, 1899:9 [Identical
description published posthumously by
Percy Moore].

Bachia fo/co/or— Ruth\en, 1922:63.

Range. — This species is known only
from northern and central Colombia, and
exti-eme western Venezuela (Fig. 7).
Literature records from Cuenta, Honda
(Ruthven, 1925), Tucurinca (Ruthven,
1922), San Gil (Dunn, 1944), Colombia
and Sierra de Perija, near Tukuko (Ale-
man, 1953), Venezuela are probably re-
ferable to Bachia bicolor.

Diagnosis. — Dorsal and lateral body
scales hexagonal, smooth and imbricate;
ventral scales quadrangular, juxtaposed;
tail scales similar to dorsals; nasals,
frontonasal, parietals present; interpari-
etal, prefrontals and supraoculars absent;
usually no contact between supralabials
and parietals; two superciliaries; preanal
pores 1-1 in males, absent in females;
forelimb small with four toes; hind limb
stvloform with one to three apical scales
resembling toes; SAB 27-31 (x 28.9);
dorsals 46-53 (x 49.7); ventrals 34-40 (x
37.1); gulars 7-9 (x 8.0); maximum
snout-vent length 75 mm.

Variation. — No variation was ob-
served in the absence of the interparietal,
prefrontals or number of forelimb toes.
Only one of 21 specimens had a small
supraocular on each side in addition to
the superciliaries. The fourth supralabial
in contact with the parietal on both sides
in 2 specimens, one side in one, and
neither side in 18. The second pair of
chin shields are never in medial contact
and reach oral border in about 50 per-

cent of the sample; supralabials and in-
fralabials 5/5 in all; preanal shields three
in 19 specimens, four (2); longitudinal
rows of quadrangular ventral scales are
4-S(x6.95).

Coloration. — Bachia bicolor has three
distinct dorsal color patterns, although
the head color pattern is rather constant.
The head is brown to blackish with two
thin yellowish tan to cream lines begin-
ning above eye on outer edge of frontal,
passing posteriorly along outer edge of
parietals to nape. Dorsal body pattern
consists of two or three black or brown-
ish black lines one half to one scale row
wide, extending from the nape to near
tip of tail; median dark line occasionally
absent; median line (when present) and
paravertebral lines straight edged, irreg-
ular, or broken into a linear series of
short dashes; dorsal lines separated by
light tan to tan brown stripe about two
scale rows wide; light stripes ( fused into
a single light stripe five scale rows wide
when median dark line absent) are fre-
quently freckled with dark brown; dorso-
lateral light stripe usually uniform in
color; upper side of body usually as dark
as dark dorsal lines, fading into brownish
color of venter; chin and throat usually
uniform brown.

Remarks. — Bachia bicolor was first
described in a footnote by Cope (1896).
Percy Moore apparently overlooked this
description when he published Cope's
(1899) posthumous paper on "Contribu-
tions to the herpetology of New Granada
and Argentina." The latter paper con-
tains a better "original" description and
excellent figures of bicolor. Cope's (1899)
figure of the preanal plate and adjacent
body scales does not show the hind limb,
unless the two large scales shown repre-
sent the limbs. However, Cope's text
states that a hind limb is present and is
1.3 mm in length.

Ruthven (1922) reports a series of
bicolor came from beneath logs and
rocks in a dry forest. Valdivieso and
Tamsitt ( 1963 ) found a specimen be-
neath a rock in a cow pasture.

24

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

Figure 7. Distribution of B. bicolor and B. taJpa in Colombia and Venezuela.

Bachia talpa Ruthven
Figure 6

Bachia talpa Ruthven, 1925:101 [Holotype.—
UMMZ 54771 from Valledupar, Sierra Santa
Marta, Colombia; A. G. Ruthven collector].

Range. — Known only from extreme
northern Colombia, along the west, south

and east sides of the Sierra Santa Marta
(Fig. 7).

Diagnosis. — Dorsal and lateral body
scales hexagonal, smooth and imbricate;
ventral scales quadrangular, juxtaposed;
tail scales similar to dorsals; nasals,
frontonasal, parietals present; interpari-
etal, prefrontals, supraoculars absent;

DIXOX: SYSTEMATIC REVIEW OF BACHIA

25

fourth supralabial occasionally contacts
parietal; preanal shields three; preanal
pores absent; superciliaries two; fore-
limbs small with three toes on each foot;
hind limb tubercular with one apical
scale that mav be divided; SAB 25-27 ( x
26.0); dorsals 47-51 (x 49.3); ventrals
36-38 (x 37.3); gulars 9; maximum snout-
vent length 65 mm.

Variation. — Three specimens, para-
types, were examined. Apparently no
specimens exist in museum collections
other than the type series. By incorpo-
rating the information from the original
description with my data, the number of
forelimb toes, superciliaries, preanal
shields, labials, and gulars are constant.
Numbers of longitudinal rows of quad-
rangular ventrals are 8-10 (x 9.1); sec-
ond pair of chin shields never in medial
contact, but reach oral border in all spec-
imens; apical scale of hind limb tubercle
may be single or divided.

Coloration. — Color patterns of dor-
sum consists of three distinct, longitu-
dinal dark brown to black lines (one
median, two paravertebral); dark lines
are one half to one scale row wide and
extend from nape to near tip of tail; sides
of body dark brown, upper edge almost
black and forming a somewhat distinct
line, giving an appearance of five dark
lines on dorsum; lower sides slightly
lighter in color than sides; venter dark
brown to almost black and darker than
side of body; dark dorsal lines separated
by yellowish tan, tan or light brown
stripes one and one half to two scale
rows wide; all lines and stripes extend
cephlad to level of frontonasal; side of
head dark brown; snout light brown;
throat and chin brown.

Remarks. — Ruthven (1922) reports
the type series came from beneath logs in
a dry forest, and dry parts of flood plain
forests.

Bachia huallagana new species

Figure 8

Holotype.—VSNM 192997, adult
male.

Type Locality. — San Jose de Tocache

( = Tocache Nuevo) 80° 12' S, 76° 27'
W\ 450 m., on Rio Huallaga, Provincia
Mariscal Caceres, Departamento San
Martin, Perii, collected by Wade C. Sher-
brooke, 2 February 1967.

Paratypes.—USNM 192996, 192998-
193004 topotypes; USNM 192993-95,
193005-07, Tingo Maria 670 m, Departa-
mento Huanuco, Peru; AMNH 56566,
upper Biabo Vallev, Departamento San
Martin, Peru; UF 28143-44, 27.4 km SSE
Aucayacu 650 m, Departamento Hua-
nuco, Peru (Fig. 5).

Diagnosis. — A member of the dor-
bignyi group with hexagonal, smooth
and imbricate scales on the dorsal and
lateral surfaces of body; ventral scales
quadrangular and juxtaposed; tail scales
similar to dorsals; interparietal, prefron-
tals and supraoculars absent; nasals,
frontonasal, frontal and parietals present;
nasals not in medial contact behind ros-
tral; superciliaries 2-2; forelimb small
with three toes on each foot; hind limb
tubercular; second pair of chin shields
never in medial contact nor reaching
oral border; usually fourth supralabial
in contact with parietal; preanal shields
usually four; preanal pores 2-2 in males,
1-1 or absent in females; SAB 24-30 (x
26.1); dorsals 47-54 (x 49.2); ventrals
35-42 (x 37.2); gulars 7; labials usually
5/4; maximum snout- vent length 73 mm.

Description of Holotype. — Rostral
slightly more than twice as wide as high;
frontonasal narrow anteriorly, one-half
as broad as posterior end; supralabials
five on each side; first supralabial in-
dented dorsally by nostril lying between
it and nasal; fourth supralabial highest,
in contact with parietal on each side;
fifth supralabial elongate and somewhat
triangular in shape; frontal in brief con-
tact with posterior edge of nasals; pari-
etals two, large; supraoculars and pre-
frontals absent; superciliaries two on
each side, anterior one largest; loreal
rectangular, in contact with frontal, su-
perciliary, preocular, nasal and second
and third supralabials; preocular one on
each side; suboculars two on each side,
anterior one elongate and narrow, pos-

26

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

B. huallagana UF 28143

B. dorbignyi UMMZ 6S082

B. peruana AMNH 56561

B. tnsanalc ANSP 9637

Figure 8. Lateral, dorsal and ventral \ie\vs of the head, dorsal \'iew of the limbs of
B. huallagana, B. dorhignyi, B. peruana, and B. trimnale.

terior one small, almost granular; post-
ocular fused to fourth supralabial;
temporals three on each side; mental
trapezoidal; postmental large, septangu-
lar, followed by two pair of enlarged
chin shields, second pair widely sepa-
rated medially and almost reaching oral
border; infralabials four; union between
third and fourth reduced because of in-
trusion of second chin shield; tympanum

absent; collar fold present; enlarged
pectoral shields two; enlarged preanal
shields three, middle one divided in half;
transverse rows of dorsal scales above
pectoral shields and above preanal
shields three; transverse dorsal rows from
occiput to above vent 49; scales around
midbody 27; six ventral rows quadrangu-
lar, remainder hexagonal, smooth and
imbricate; transverse gular scale rows

DIXON: SYSTEMATIC REVIEW OF BACIIIA

27

from collar fold to pectoral shield seven,
transverse scale ro\\'s between pectoral
and preanal shields 36; transverse scale
rows from vent to tip of tail 98; forelimb
with five longitudinal scale rows from
body to toes; three clawed toes beneath
three apical scales on forelimb; hind limb
tubercular with a longitudinally divided
apical scale; preanal pores 2-2.

Dorsal and lateral head surfaces
blackish brown with two broken yellow-
ish tan lines between last superciliary
and collar fold; dorsum brownish black
with two well defined dorsolateral black
lines extending from nape to tip of tail;
median, irregular blackish line extending
from nape to about middle of tail; dark
dorsal lines superimposed over yellowish
tan ground color about 10 scale rows
wide; ground color between paraverte-
bral dark lines heavily mottled with dark
brown; broad lateral blackish stripe ex-
tends from collar fold to tip of tail, be-
coming slightly lighter in color ventro-
laterally; venter brown to blackish
brown.

Measurements in mm: Total length
149.0; snout-vent length 60.0; head
length 7.1; head width 5.2; forelimb
length 2.3; hind limb length 0.5; mid-
body width 4.2; axilla-groin length 45.0.

Variation. — Details of color pattern
and cephalic squamation of paratypes
are similar to holotype. The third supra-
labial is divided in one specimen, cre-
ating six supralabials, of which the fifth
has parietal contact. One specimen has
the first and second supralabials fused
on one side creating four supralabials,
the third in contact with a parietal. One
specimen has supralabial in contact with
parietal only on one side and another not
in contact on either side. Number of
superciliaries are 2-2 in 17 specimens, 1-2
in one. Number of gulars (7) and longi-
tudinal rows of quadrangular ventrals
(6) are constant in the sample. Preanal
shields vary from 3 to 5, three occurring
4 times, four (13), five (1). Preanal
pores are 2-2 in males, 1-1 or absent in
females; hind limb always tubercular
and forelimb with 3 toes on each foot.

For statistical analyses of SAB, dorsals
and ventrals see table 3.

Comparisons. — Bachia hnaUag,ana dif-
fers from all species in the dorhignyi
group (except talpa and hicolor) by ab-
sence of medial contact between second
pair of chin shields. It differs from hi-
color by having three forelimb toes rather
than four, and males with two preanal
pores on each side rather than one. The
new species differs from talpa by the
presence of four preanal pores and no
contact between second pair of chin
shields and oral border, rather than no
preanal pores and broad contact between
second pair of chin shields and oral bor-
der. Bachia huallagana differs from talpa
and hicolor by having a relatively short
rostral that barely extends beyond lower
jaw while talpa and hicolor have de-
pressed heads with an expanded rostral
extending well beyond the lower jaw.

Remarks. — Fred Thompson (pers.
com. ) indicated that specimens were
found within decaying logs about one
meter in diameter in tropical rainforest
at Aucayacu. Wade Sherbrooke (pers.
com.) stated that his series from the
Tingo Maria area were found in the soil
and humus beneath decaying logs in the
Agraria de la Selva University botanical
garden. Some of Sherbrooke's specimens
were maintained in captivity for a short
while. During daylight periods they re-
mained beneath the soil, but after dark
they emerged and crawled about on the
surface, indicating nocturnal tendencies.
Sherbrooke indicated that the majority of
the topotypic series were taken by fol-
lowing a bulldozer clearing a road to an
airstrip.

This species is named for the major
river drainage (Rio Huallaga) from
which the majority of specimens have
been taken.

Bachia dorbignyi (Dumeril and Bibron)

Figure 8

Chalcides dorbignyi Dumeril and Bibron,
1839:462 [Holotype.— MNHN 2841 from
Santa Cruz, "Chile"; collector unknown
(corrected to Santa Cruz de la Sierra, De-

28

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

partamento Santa Cruz, Bolivia, by Vanzo-

lini, 1961b)].
Bachia dorhignyi — Gray, 1845:58.
Cophias dorbigntji — Boulenger, 1885:417.
Bachia dorbigniji — Gamian, 1892:2.

Raniie. — Known from Rio Manoel
Carreia, extreme western Brasil (Vanzo-
lini, 1961b), Grand Chaco Region, north-
ern Paraguay ( Hellmich, 1960 ) , low and
moderate elevations of Bolivia, and ex-
treme southeastern Perii ( Fig. 5 ) .

Diagnosis. — Dorsal and lateral body
scales hexagonal, smooth and imbricate;
ventral scales quadrangular, juxtaposed;
tail scales similar to dorsals; nasals,
frontonasal, frontal, parietals present,
intei-parietal, prefrontals, supraoculars
absent; rostral frontonasal with short
contact, separating nasal scales; fifth su-
pralabial contacts parietal (never the
fourth); forelimbs with 3 toes on each
foot; hind limb styloform with one apical
scale that may be divided; superciliaries
2-2; preanal pores 2-2 in males, 1-1 or
absent in females; preanal shields usually
4; SAB 23-28 (x 24.6); dorsals 50-55 (x
52.9); ventrals 37-42 (x 39.9); gulars 6-7;
maximum snout-vent length 80 mm.

Variation. — Little variation was ob-
served in the cephalic squamation. Su-
perciliaries 2-2 in 50 specimens, 2-3 (2),
3-3 (1); fifth supralabial contacts parietal
on both sides in 45 specimens, one side
(4), neither side (4); second pair of chin
shields in medial contact in 49 speci-
mens, separated by small azgyous scale
(4); gular scales seven in 49 specimens,
six (4); preanal shields three in 2 speci-
mens, four (51); longitudinal quadran-
gular ventral rows four in 7 specimens,
five (9); six (37); toes on forelimbs 2-3 in
1 specimen, 3-3 (52); hind limb invari-
ably styloform with single apical scale;
supralabials and infralabials always 6/5.

Coloration. — The dorsal head pattern
of B. dorhignyi consists of a large arrow-
shaped mark pointing anteriad, extend-
ing from anterior edge of frontal to pos-
terior edge of parietals; small black line
or series of small spots between open
ends of arrow and nape connecting to
median dark line on body; dorsolateral

and lateral surfaces of head brownish
black to black; two yellowish tan lines
extend from parietals to anterior end of
frontal, fusing to form light colored
frontonasal and rostral; dorsolateral sur-
face of head and arrow-shaped mark
frequently spotted with tan; chin and
throat unifonn light to dark brown.

Dorsum with broad yellowish tan or
cream stripe, 10 scale rows wide, from
nape to near tail tip; three brownish
black to black lines (median and two
paravertebral) superimposed over light
stripe from nape to near tail tip; edges of
dark lines straight and seldom broken
into series of dashes or spots; sides of
body dark brown to black, venter brown
to brownish black and usually a little
lighter than sides.

Remarks. — Natural history informa-
tion is lacking for this species. Ruthven
(1922) recorded dorhignyi from the Si-
erra Santa Marta of Colombia, but later
(1925) described the same sample as a
new species (taJpa). Werner (1900) con-
sidered dorhignyi and peruana to be
closely related, but Ruthven (1925) and
Burt and Burt (1931) suggested possible
synonymy. Vanzolini (1961b) suggested
that peruana was probably not a strict
synonym of dorhignyi, but intimated a
conspecific relationship.

I find a significant difference (3 SE)
between the two species in the number
of dorsals and ventrals, and no overlap
in the number of labials and the specific
labial in contact with the parietal. Even
though the two populations are allo-
patric, I consider them to be species.

Bachia peruana (Werner)
Figure 8

Cophias peruana Werner, 1901:5 [Holotype. —
DM 1698 ( destroyed during World War II)
from Chanchamayo, Peru; collector un-
known].

Bachia peruana — Ruth\en, 1925:105.

Range. — Known from the western
valleys of the middle tributaries of the
Rio Ucayali and the flood plain of Rio
Ucayali from Chanchamayo on the south
to Orellana on the north, Peru (Fig. 5).

DIXON: SYSTEMATIC REVIEW OF BACHIA

29

Diagnosis. — Dorsal and lateral body
scales hexagonal, smooth and imbricate;
ventral scales quadrangular, juxtaposed;
tail scales similar to dorsals; nasals,
frontal, parietals and frontonasal present
(frontonasal occasionally absent); pre-
frontals, interparietal and supraoculars
absent; frontonasal usually reduced (oc-
casionally absent) allowing contact be-
tween nasal scales in about 25 percent
of sample; fourth supralabial always
contacting parietal (never fifth); fore-
limbs small with three toes on each foot;
hind limb styloform with single apical
scale; superciliaries 2-2; preanal pores
1-1 (frequently indistinct), absent in fe-
males; preanal shields usually 4; SAB
23-26 (x 25.1); dorsals 53-56 (x 54.6);
ventrals 41-43 (x 42.0); gulars 6-7; maxi-
mum snout-vent length 107 mm.

Variation. — The cephalic squamation
shows little variation. One of 38 speci-
mens has 3-3 superciliaries; fourth supra-
labial always in contact with parietal on
both sides; second pair of chin shields
always in medial contact but never reach
oral border; forelimbs with 3-3 toes in 37
specimens, 2-3 ( 1 ) ; preanal shields three
in one, four (37); gulars six in 5 speci-
mens, seven (33); longitudinal quad-
rangular ventral scale rows are 4-6 (x
5.4); see Table 3 for statistical analyses
of other body scales.

Coloration. — Dorsal head surface sim-
ilar to B. barbouri, except frontonasal
scale usually densely spotted with
brown; body and tail coloration like bar-
bouri but dark lines usually straight,
continuous and with dense freckling
along each side.

Remarks. — The relationship of peru-
ana to dorbignyi is discussed in the re-
marks section of dorbignyi. Bachia
peruana shows a marked frontonasal re-
duction (occasionally lost) in the south-
ern part of its range, allowing the nasal
scales medial contact behind the rostral.
When the frontonasal is lost, peruana
superficially resembles trisanale, a spe-
cies in which the frontonasal is always
absent. However, a significant difference
of 3 SE in numbers of dorsals and ven-
trals separates the two species.

Richard Thomas (pers. com.) found
this species under rotting logs and in
loose soil beneath decaying vegetation in
coffee groves.

Bachia trisanale (Cope)

New Combination
Figure 8

Ophiognomon trisanale Cope, 1868:100 [Holo-
type. — ANSP 9637 from Napo or Alto
Maranon, "Ecuador"; Mr. J. Orton collec-
tor].

Chalcides abendrothi Peters, 1871:399 [Syn-
types. — ZMB (10 without numbers) from
Sarayacu, Provincia Napo, Ecuador; collec-
tor unknown]. New synonymy.

Pwpus venniformis Cope, 1874:70 [Holotype.
—ANSP 11353 from Nauta, Peru; Mr. J.
Orton collector]. New synonymy.

Ophiognomon abendrothi — Boulenger, 1885:
421.

Ophiognomon vermiforme — Boulenger, 1885:
421.

Range. — This species is widely dis-
tributed in the lowlands of eastern Ec-
uador and Peru, and probably occurs in
southern Colombia and western Brasil
(Fig. 9). Literature records from Cono-
naco (Peracca, 1897) and Igarape Cham-
puia (Vanzolini, 1961c), Peru, are refer-
able to B. trisanale.

Content. — Three subspecies.

Key to .subspecies. —

1. Three toes on forelimb, hind limb tu-
bercular 2

One toe on forelimb, hind limb ab-
sent t. vermiformis

2. Second pair of chin shields in medial

contact t. abendrothi

Second pair of chin shields not in me-
dial contact t. trisanale

Diagnosis. — Dorsal and lateral body
scales hexagonal, smooth and imbricate;
ventral scales quadrangular, juxtaposed;
tail scales similar to dorsals; nasal scales
in broad contact behind rostral; fronto-
nasal, prefrontals, interparietal and su-
praoculars absent; nasals, frontal and
parietals present; forelimb toes one to
three; hind limb tubercular or absent;
superciliaries one to two on each side;

30

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

Figure 9. Distribution of B. barhouri and B. trisanale in Peru and Ecuador.

DIXON: systematic: review of bach I a

31

preanal pores 2-2 in males, 1-1 or absent
in females; SAB 23-30 (x 26.2); dorsals
47-55 (x 50.6); ventrals 34-42 (x 37.7);
preanal shields 3-5 (x 3.2); gnlars 7-(S
(x 7.1); maximnm snout-vent length 79
mm.

Variation. — Two of 72 specimens
lacked contact between fourth supra-
ocular and parietal, four specimens had
contact onh' on one side. The fronto-
nasal, prefrontals, interparietal and su-
praoculars are absent in all specimens.
Gular rows vary slightly with eight rows
occurring 3 times, seven (69); longitu-
dinal (juadrangular ventral rows are 4-7
(x 5.8); preanal shields three in 69 speci-
mens, four (6), or five (5).

Coloration. — Dorsal head surface uni-
form slate brown, blackish brown, or
black, with or without pair of yellowish
tan lines from above eye to posterior
edge of parietals; light lines may be
present as series of spots, dashes, or ir-
regular blotches; lateral and dorsal head
surfaces often obscurely mottled with
brown and blackish brown; chin and
throat similar to head color.

Dorsum of body and tail similar to
bicolor, usually lacking a definite median
dark line; area between paravertebral
dark lines heavily freckled or mottled
with dark brown to black; median dark
line, when present, usually terminates
above hind limbs; the two paravertebral
dark lines extend to tail tip, upper side
of body black, fading to blackish brown
ventrally; venter dark brown, slightly
lighter in color than sides of body.

Remarks. — Bachia ahendrothi was
originally described as having a smaller
hind limb (tubercular) and more trans-
verse body scale rows than tri.sana[e
(Peters, 1871). However, examination of
72 specimens from throughout the range
of both species indicates complete over-
lap in the number of dorsals.

Hind limb length in populations orig-
inally assigned to trisanale from Ecuador
and northern Peru, is 33-60 ( x 45.0 ) per-
cent of the preanal shield length, while
those populations assigned to ahendrothi
from central and southern Peru have a

hind limb length of 5-53 (x 34.2) percent
of the preanal shield length. The num-
bers of dorsal and ventral scales respec-
tively, vary from 47-55 (x 50.3), 34-42
(x 37.3) for the northern populations,
while these characters for the central and
southern populations are 48-55 (x 51.1),
34-42 (x 38.2). Numbers of SAB are
similar, 23-30 (x 27.1) to the north, 23-30
(x 26.2) in the south.

The only useful character that sepa-
rates the two populations is the presence
or absence of medial contact between
the second pair of chin shields. Only one
specimen of 21 examined from south of
7° S lacks medial contact, and only one
of 50 from north of 7° S has medial con-
tact. Thus, I recognize ahendrothi as a
subspecies of trisanale by having the
second pair of chin shields in medial
contact.

Vanzolini (1961c) pointed out the
similarities between the two species and
suggested several hypotheses as to their
relationships. He concluded that the two
species should be recognized pending
further investigations of geographic vari-
ation. Vanzolini noted that the type of
trisanale could not be found and that
Cope's ( 1868 ) count of the midbody
scale rows (20) may be in error. There
are 23 midbody scale rows rather than
20.

Since the original description of
vermiformis by Cope (1874), no addi-
tional specimens representing the taxon
have been found. However, in a series
of 13 specimens of trisanale from the
vicinity of Iquitos, Peru, several speci-
mens show a marked reduction in the
number of toes on the forelimb. The toes
vary from 3-3 to 2-1 and the hind limb is
represented by a small tubercle. The
variation in the number of forelimb toes
and the near absence of the hind limb
suggests that the Iquitos population may
represent intergrades between trisanale
and vermiformis. Similar variation is
present in populations of monodactijlus
from Surinam. Of the 13 specimens, eight
have forelimb toes numbering 3-3, one
3-2M, two 3-2, one 2-2, and one 2-1.

32

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

The type locality of vermiformis,
Nauta, Peru, is 95 airline kilometers SSW
Iquitos. I would expect to find vermi-
formis sympatric with trisamile through-
out this short distance of uniform habitat
if vermiformis truly represented a dis-
tinct species. However, the squamation
and color pattern of the type of vermi-
formis falls within the range of variation
of trisanale, except for the medial con-
tact of the second pair of chin shields.
One specimen from Iquitos has the sec-
ond pair of chin shields in medial con-
tact, further suggesting that these two
species are conspecific, therefore I regard
vermiformis as a subspecies of trisanale.

Data taken from field tags indicate
that most specimens of this species were
found beneath decaying logs in primary
or secondary forests along streams flow-
ing into the Amazon Basin. Richard
Thomas (pers. com.) found this species
in loose soil around rotting palm trunks
and beneath a board within a residential
area of Yarinacocha, Peru.

The Bachia heteropa Group

The Bachia heteropa group comprises
populations from Panama, Colombia,
West Indies, and Venezuela having 6 to
12 rows of hexagonal dorsal scales; 13 to
21 rows of quadrangular lateral and ven-
tral scales; 4 digits on the forelimbs; in-
terparietal and frontonasal shields pres-
ent; prefrontal shields present or absent.

The complex is composed of two
species, paUidiceps and heteropa, the
latter with a rassenkreis of populations
formerly recognized as six species (heter-
opa, alleni, trinitatis, lineata, marcelae,
anomala). Bachia alleni and B. trinitatis
were considered conspecific and B. a.
parviceps relegated to the synonymy of
B. a. alleni by Thomas (1965a). Bachia
lineata and B. marcelae were considered
conspecific by Donoso-Barros (1968) and
B. anomala relegated to the synonymy of
lineata by Burt and Burt (1933). Shreve
(1947) suggested that lineata and ano-
mala might not be conspecific.

I examined the holotype of B. hlairi

and found no differences between it and
populations of B. paUidiceps of Panama
and Colombia. Hence I refer hlairi to the
synonymy of paUidiceps. The status of
B. cuvieri was previously discussed; it
may belong to this group. In addition, I
find that certain head shield and digital
characters that were formerly used to
separate species of this group (i.e., het-
eropa, lineata, alleni) show an inter-
mediate condition or, are of questionable
value at the species level (i.e., presence
or absence of prefrontals, loss of toes on
hind limb ) .

Thomas (1965a) indicated that B. a.
trinitatis might represent an intermediate
population between a mainland race re-
lated to alleni ( = heteropa ) and the
population oi alleni alleni (=/?. alleni) on
Grenada Island. He based this assump-
tion on the separation of the prefrontals
(split by contact of the frontonasal and
frontal shields). All specimens from
Grenada have prefrontals in medial con-
tact, 15.4 percent of the Trinidad sample
have medial contact and the mainland
sample lacks medial contact. Populations
of Bachia nearest the Rio Orinoco delta
tend to have prefrontals more reduced in
size and more lateral in position. Popu-
lations of Bachia immediately west of the
Paria Peninsula, Venezuela, have the pre-
frontals either fused to the anterior su-
praocular, or reduced to small azygous
scales wedged between the nasal and
anterior supraocular scales. In the west-
ern part of the range of B. heteropa, the
number of supraoculars are smaller and
reduced to two, and the prefrontals are
fused to the loreal scale. Correlated with
this shift in head scales, the frontonasal
and frontal shields are enlarged and
occupy the prefrontal space. The longi-
tudinal rows of body scales are reduced
in number (scales are larger), and the
hind limbs are smaller with a corre-
sponding reduction in the number of
digits. The hind limb is moderately de-
veloped in B. h. alleni with four digits,
slightly reduced with smaller toes in B.
h. trinitatis, further reduced in size with
loss of toes (three present) in heteropa

DIXON: SYSTEMATIC REVIEW OF BACHIA

33

hetewpa, and c\'cntuall>' a clegciK-rate
hind limb with two toes in /?. lineota and
h. marccUie (Fig. 10). All populations
seem to be conspecific with slight modi-
fication in some for a more fossorial ex-
istence. The oldest available name for
the species is heteropa Lichtenstein
(1S56).

Populations of Bachia (represented
b\' pallid iceps) in northern Colombia
and Panama haw retained distinct pre-
frontal shields (not in contact medially),
have the digits of hind limb reduced to
three ( four in one specimen ) , and super-
ficially resemble B. li. trinitatis from the
Paria Peninsula of Venezuela. However,
the former population has lost the me-
dian black stripe of the dorsum, has a
reduced number of scales around the
middle of the body, and has a higher
number of longitudinal dorsal and ven-
tral scales.

The generalized head scale pattern

(presence of prefrontals) in pallid iceps
must have been retained during the
spread of populations of Bachia in the
Cenozoic. Populations associated with
mesic conditions have retained a gener-
alized condition ( i.e., well developed
limbs and toes, presence of prefrontals)
favoring supraterranean existence in for-
est leaf litter. Those populations (heter-
opa, lineata, marcelae) associated with
the "rain shadow" or arid scrub forests
of northern Venezuela have become
more adapted to a fossorial existence
{i.e., reduction in size of hind limb, loss
of toes, loss of head shields).

Bachia heteropa (Lichtenstein)
Figure 11

Chalcides hetero})us Lichtenstein, 1956:17
[Holotype.— SMF 39900 from La Guaira,
Venezuela; collector unknown].

Bachia lineata Boulenger, 1903:432 [Holotype.
— BMNH 1962.7.29.90 from Duaca, Estado

Figure 10. Progressive reduction in size and loss of prefrontal scales and hind limb toes in the

Bachia Heteropa group in northern Venezuela. A. B. heteropa alleni. B. and C. B. heteropa trinitatis

(Trinidad and mainland). D. B. heteropa heteropa x heteropa trinitatis. E. B. heteropa heteropa.

F. B. heteropa lineata. Direction of arrow indicates trend from mesic to xeric habitat.

34

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

B. heteropa KU 117099

B. pallidiceps KU 9(58";

Figure U. Lateral, dorsal, and ventral views of the head, dorsal view of the limbs of B. heteropa

and B. pallidiceps.

Falcon, Venezuela; Mr. Wayman collector].
New synonymy.

Scolecosatirus alleni Barbour, 1914:315 [Holo-
type. — MCZ 7793 from St. George's, Gre-
nada; G. M. Allen collector]. New syn-
onymy.

Scolecosaurus trinitatis Barbour, 1914:316 [Hol-
otype. — MCZ 8947 from Caparo, Trinidad;
A. B. Carr collector]. New synonymy.

Bachia heteropa — Ruthven, 1925:105.

Bachia anomala Roux, 1929:31 [Holotype. —
MB 9912 from El Mene, Distrito Acosta,
Estado Falcon, Venezuela; H. G. Kugler
collector]. New synonymy.

Bachia marcelac Donoso-BaiTos and Garrido,
1964:3 [Holotype.— D-B 6312061 from
Bosque La Luz, Estado Barinas, Venezuela;
R. Donoso-Barros collector]. New syn-
onymy.

Range. — The islands of Grenada,
Grenadines, Tobago, Trinidad, and the
arid coastal zone of northern Venezuela,
from El Mene east to Paria Peninsula,
and south to the subtropical interior near
La Luz (Fig. 12).

Literature records from the Port of
Spain, Trinidad (Burt and Burt, 1931),
7.5 mi N San Juan, Caura Valley, Prince-
ston-Moruga Road, Manzanilla Bay,

Trinidad (Thomas, 1965a); Monagas,
Venezuela (Thomas, 1965a); are refer-
able to Bachia h. trinitatis. Those from
Grenada, Grand Etang (Burt and Burt,
1931), 0.2 mi E Willis, 1 mi. SE Vin-
cennes, 1 mi. N Vincennes, Bailey
Bacolet Stock Farm, North Shore of
W^esterhall Bay, 0.8 mi S Sauteurs, Mt.
Alexander (Thomas, 1965a); Grenadines;
Cannovan Island (Thomas, 1965a); To-
bago; 4 mi NE Penbroke, 4 mi N Mt. St.
George (Thomas, 1965a), are referable
to Bachia h. alleni; from La Luz, Barinas,
Venezuela (Donoso-Barros and Garrido,
1964) to Bachia h. marcelae.
Content. — Five subspecies.
Key to Subspecies. — If prefrontals
appear to be present but quite lateral in
position, the specimen may be an inter-
Urade between h. trinitatis and h. heter-
opa. Gheck the number of toes on the
hind limb for correct allocation.

1. Prefrontals present, four toes on hind

limb 2

Prefrontals absent, three or less toes
on hind limb 3

DIXON: SYSTEMATIC REVIEW OF BACH I A

35

Figure 12. Distribution of B. heteropa and B. paUiidiceps in nortlieastem South America and Panama.

2. Prefrontals not in medial contact

heteropa tiinitatis

Prefrontals in medial contact

heteropa ailen'i

3. Hind limb with two distinct toes 4

Hind limb with three distinct toes
heteropa heteropa

4. Seven rows of scales in front of pec-
toral shields heteropa Uneata

Five rows of scales in front of pectoral
shields heteropa marcelae

Diagnosis. — Five to 12 longitudinal
rows of hexagonal, smooth and imbricate
dorsal body scales; 16 to 22 longitudinal
rows of quadrangular, juxtaposed lateral
and ventral scales; tail squamation sim-
ilar to body scales in subspecies heter-
opa, Uneata and marcelae, mostly hexag-
onal in aUeni and trinitatis; interparietal,
frontonasal, nasals and parietals present;
two or three supraoculars and supercili-
aries; second pair of chin shields not usu-
ally in medial contact nor reaching oral
border; no labial /parietal contact; four
toes on forelimbs; two to four toes on

hind limbs; males with one to two pre-
anal pores, absent in females; SAB 24-31
(x 28.4); dorsals 38-49 (x 41.6); ventrals
24-36 (x 27.0); gulars 7-8 (x 7.5); maxi-
mum snout- vent length 64 mm.

Variation. — The general variation of
this species was discussed in the intro-
duction to the group. I have not found
intermediate specimens between Uneata
and heteropa, though I expect them to
occur between Caracas and the Paria
Peninsula. The reduction from three to
two toes may or may not be controlled
by a single dominant gene; however, this
type of reduction is widespread in B.
nionodactyhts, with many specimens
showing various combination of toes on
the hind limbs. The progressive loss of
the prefrontal shields is illustrated in
figure 10. A reduced number of gular
scale rows follows the general evolu-
tionary trend outlined in the introduction
to the group. The number of gular rows
are eight in all h. aUeni, reduced to seven
in 57.5 percent on h. trinitatis, seven in
all h. heteropa and h. Uneata, and five
in h. marcelae.

36

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

The number of preanal shields is usu-
ally four in all subspecies of heteropa,
with three and five occurring once in 80
specimens. The number of SAB varies
slightly between subspecies with oUeni
having 27-31 (x 29.0), trinitotis 26-31 (x
28.5), heteropa 28-30 (x 29.3), Uneata
24-26 (x 25.3), marcelae 25 (3 speci-
mens ) . Number of dorsal scale rows are:
alleni 39-45 (x 40.9), trinitatis 38-45 (x
41.3), heteropa 45-47 (x 45.3), Uneata
45-49 (x 47.2), marcelae 39-42 (x 40.7).
Number of ventrals: alleni 24-29 (x
26.4), heteropa 27-30 (x 28.3), Uneata
24-26 (x 25.3), (not known for marce-
lae). The number of supralabials and
infralabials is usually 6/6 or 6/5 for all
subspecies. The number of supraoculars
is always three in alleni and trinitatis,
usually two in heteropa but occasionally
three in intergrades between heteropa
and trinitatis, and always two in Uneata
and marcelae. The number of super-
cilaries is always three in all subspecies
except 3/4 occurring once in 80 speci-
mens.

One specimen is intermediate be-
tween heteropa and trinitatis; it has the
second pair of chin shields in medial
contact, a unique condition in B. heter-
opa. The medial contact may be due to
recombination of genes between inter-
grading populations, or some genetic in-
fluence from an unknown population of
Bachia in the area. The problem will be
resolved only by additional sampling in
the area of contact between the two
populations.

Coloration. — Each subspecies tends to
vary slightly from the basic body pattern
of three longitudinal dorsal dark lines
alternating with four wider, light inter-
space stripes, extending from the pari-
etals onto the tail. The sides of the body
are dark brown to blackish and the ven-
ter light brown to yellowish tan in all
subspecies. Bachia h. alleni tends to have
the median dark line obscure, with the
paravertebral dark lines broken into a
series of dashes or spots and the light
interspaces heavily mottled or freckled
with dark brown. Bachia h. trinitatis has

retained the three dark lines, although
occasionally broken or irregular. The
two paravertebral light interspace stripes
are usually freckled with dark brown,
whereas the two dorsolateral light inter-
space stripes are uniform tan to yellowish
brown. The three dorsal dark lines are
usually evident in the remaining sub-
species, occasionally irregular, with the
light interspace stripes less freckled or
mottled. All subspecies tend to have the
upper edge of the lateral dark stripe
darker than the stripe, giving an illusion
of five dark stripes on the body.

Remarks. — Barbour (1914) indicated
that B. h. alleni is a burrower, but did
not say that he actually found specimens
in burrows. He stated that the type se-
ries of B. h. trinitatis came from beneath
fallen cacao pods. Parker (1935b) stated
that a stomach of trinitatis contained a
small earwig, a dipterous larva, and a
mass of mycetophilid larvae.

The size and development of the
limbs and toes suggests that B. h. Uneata
and marcelae are predominately fosso-
rial, heteropa partially fossorial, and al-
leni and trinitatis supraterranean leaf
litter inhabitants.

Bachia pallidiceps (Cope)
Figure 11

Brachypus pallidiceps Cope, 1862:356 [Holo-
type.— ANSP 4324 (lost) from Rio Truando
Region, Colombia; L. Michler collector].

Scolecosatims hiairi Dmin, 1940:115 [Holotype.
— ANSP 21773 from Puerto Armuelles,
Chiriqui, Panama; E. R. Dunn collector].
New synon\my.

Bachia pallidiceps— Vanzohni, 1961b:204.

Range. — Northwestern Panama to
west-central Colombia (Fig. 12). A lit-
erature record for Laguna de Pita, Pan-
ama (Dunn, 1940) is referable to this
species.

Diagnosis. — Dorsum of body with
7-14 longitudinal row^s of hexagonal,
smooth imbricate scales; lateral and ven-
tial body surfaces with 10-16 rows of
(juadrangular, juxtaposed scales; tail and
boch' scales similar; nasals, frontonasal,
parietals, interparietal, prefrontals pres-

DIXON: SYSTEMATIC REVIEW OF BACHIA

37

ent; prefrontals not in medial contact;
two supraoculars and superciliaries; sec-
ond pair of chin shields not in medial
contact nor reaching oral border; no
supralabial/parietal contact; four clawed
toes on forelimbs; hind limbs with three
toes (four in one specimen); preanal
shields four; preanal pores 1-1 in males,
absent in females; SAB 21-25 (x 22.9);
dorsals 43-48 (x 45.3); ventrals 30-34
(x 31.6); gulars 7; maximum snout-vent
length 73 mm.

Variation. — Little variation was noted
in this species, other than body squama-
tion (see table 3 for statistical analyses).

Coloration. — The dorsal color pattern
varies from almost uniform dark brown
to black ground color without lateral
stripes (blairi Dunn 1940), to a dark
brown or black median stripe about four
scale rows wide, bordered by a golden
tan to yellowish line one to two scale
rows wide. The dorsolateral light stripe
(line) begins at the nape and continues
onto the tail, fading into a series of bro-
ken lines from proximal half of tail to
near the tip. The venter is black or dark-
brown, throat and chin dark gray. The
head is black or dark brown above, with
or without faint light brown spots on the
parietals.

Remarks. — Dunn (1940) stated that
his specimen of blairi (i= paUidiceps)
was taken beneath debris along the bank
of a small stream. No other natural his-
tory information is available.

Bachia paUidiceps resembles B. h.
trinitatis in the position of the prefrontal
. and occasional presence of four toes on
I the hind limb, but it differs by having
fewer SAB, supraoculars, and supercili-
aries, and more dorsals and ventrals. In
addition, paUidiceps lacks freckled light
stripes and has a black to dark brown
venter.

Thomas (1965b) stated that Bou-
lenger (1887) may have been correct in
regarding paUidiceps as a synonym of
heteropa, and suggested a racial rela-
tionship based upon prefrontal contact.
However, Thomas ( 1965b ) had not seen
specimens of heteropa, but instead, based

his conclusions on Boulenger's (1887)
redescription of the type. Beebe's ( 1945 )
illustration of heteropa referred to by
Thomas (1965b) is actually an intergrade
between heteropa and trinitatis, and pos-
sesses three supraoculars in addition to
prefrontals. Regarding supraoculars,
Thomas ( 1965b ) correctly inferred spe-
cific differences between paUidiceps (two
supraoculars) and heteropa (three supra-
oculars). However, by recognizing Une-
ata and marcelae as subspecies of heter-
opa, paUidiceps could be recognized as
a race of heteropa, disregarding other
characters that support specific recog-
nition.

The Bachia favescens Group

The Bachia ffavescens group is com-
prised of two species (flavesceiis, mono-
dactyhis), each with two races, and pos-
sibly a third undescribed species from
Colombia. The group occupies the Gui-
ana Shield and certain areas in the lower
and middle Amazon Basin in Brasil and
Colombia.

The group includes those populations
having quadrangular, juxtaposed scales
covering the entire body and tail; pre-
frontal shields absent; interparietal pres-
ent or absent; one or two supraoculars;
two or three superciliaries; two to three
digits on the forelimb; one to three digits
on the hind limb, or only a minute tu-
bercle present.

I mentioned the confusion concerning
the names favescens, cophias, and tri-
dactyla in the introduction to this paper.
Until Peters and Donoso-Barros' ( 1970 )
investigation of the status of the names,
all were used for about 100 years.
Brongersma (1946) distinguished be-
tween tridactyla and schlegeU (Dumeril
and Bibron) and retrieved schJegeU from
Boulenger's (1885) synonymy of fnc/ac-
tyla; Brongersma (1946) also showed
that Boulenger's (1887) description of
boettgeri was identical to that of schle-
geU. Vanzolini ( 1950 ) discovered that
the name tridactyla (Daudin) was pre-
occupied by Chalcides tridactyla (a

38

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

skink) and consequently substituted the
earliest available name, schlegeli (Du-
meril & Bibron, 1839). However, Vanzo-
lini had not seen Brongersma's (1946)
paper and later (Vanzolini, 1961) sug-
gested some other name be used but did
not specify which one. Burt and Burt
(1931) showed that flavescens (Bonna-
terre, 1789) was the oldest described
form of Bachia, and that Boulenger
(1885) had applied the name of those
populations with a single toe on the hind
foot. Because flavescens was described
as having three toes on the hind foot,
Burt and Burt ( 1931 ) suggested that
Daudin's (1802) description of tridac-
tyla was identical with flavescens. They
indicated that the name cophias ( Schnei-
der, 1801 ) was applicable to those popu-
lations having one toe on the hind foot.
Thus, Schneider's species, cophias, be-
came established for those populations
of Bachia having quadrangular body
scales and one toe on the hind limb. The
latter name was in common usage until
Peters and Donoso-Barros (1970) pointed
out that the descriptions of cophias,
flavescens, and tridactyla were based
upon Lacepede's ( 1789 ) iconot>'pe, and
that flavescens was the earliest available
name for the populations with three toes
and quadrangular scales, and that mono-
dactyhis ( Daudin, 1802 ) should be used
for the populations with single toes.

I find only minor differences between
Bachia flavescens and B. schlegeli. I have
examined the holotype of schlegeli, one
additional specimen from the British
Museum, and the onlv known specimen
oi flavescens (UMMZ 65170). The only
difFerence between flavescens and schle-
geli is the number of toes on the hind
limb {.schlegeli has two, flavescens,
three). I regard the latter characters as
suspect, because the number of toes on
the hind foot is highly variable within
and between populations of this group.
A similar situation exists between mono-
dactyhis and parkeri, and the only char-
acter of value is the presence of one
supraocular in parkeri, two in monodac-
tylus. The presence or absence and vari-
ation in number of supraoculars in other
species of Bachia indicates that this
character is probably not reliable at the
species level. I consider schlegeli as a
subspecies of flavescens, and parkeri a
subspecies of monodactylus.

Bachia flavescens (Bonnaterre)
Figure 13

Chalcidcs flavescens Bonnaterre, 1789:67 [Hol-
otype.— unknown; type locality and collec-
tor unknown].

Chalcides cophias Schneider, 1801:209 [Holo-
type.— unknown; type locality and collector
unknown].

Chamaesaura tridacUjlus Daudin, 1802:367

B. flavescens UMMZ G'llTO

B. monodactylus FMNH 2:!80T

Bachia sp i FlaviM.fn-- Groupi MM R .S 17

FiGxmE 13. Lateral, dorsal, and ventral views of the head, dorsal \iew of the limbs of B. flavescens,

B. monodactylus and Bachia sp.

DIXON: SYSTEMATIC REVIEW OF BACH! A

39

[Holotype. — unknown; type locality and col-
lector unknown].

Chalcides schlegeli Dumcril and Bibron, 1839:
457 [Holotype.— RMNH 3580 from "Cal-
cutta" (in error); collector unknown]. New
s\nonyiny.

Cophias bocttgeri Boulenger, 1887:517 [Holo-
type.—SMF 39900 from Central America?,
collector unknown]. New synonymy.

Badiia flavesccns — Burt and Burt, 1931:317.

Ramie. — Known onlv from Kartabo
and Moraballi, Guyana (Fig. 14). The
Britisli Museum has one additional speci-
men (1903.5.29.2) from Ireng Valley,
Guyana, that I have not seen.

Content. — Two subspecies.

Key to subspecies. —
1. Three toes on the hind foot

flavesccns

Two toes on the hind foot __.^ schlegeli

Diagnosis. — Body and tail with quad-
rangular, juxtaposed scales; nasals, pari-
etals, interparietal, frontonasal and fron-
tal present; two supraoculars; three su-
perciliaries; three toes on forelimb, two
to three toes on hind limb, all toes
clawed; males with two preanal pores on
each side, absent in females; SAB 30-32
(x 31.0); dorsals 46-48 (x 47.3); ventrals

33-34 (x 33.3); gulars 7; preanal shields
3 or 4; maximum snout-vent length 67
mm.

Variation. — There is more variation
in the shape and size of the interparietal
of this species than one might expect
from an examination of only four speci-
mens, and only two of these from definite
localities. In two specimens the inter-
parietal splits the parietals and contacts
the frontal. In one, the interparietal is
about 75 percent of the parietal length,
allowing contact of the parietals at their
medial anterior ends. One specimen has
a smaller interparietal, about 60 percent
of the parietal length and allowing
greater contact of the parietals at their
medial anterior ends.

The reduction in interparietal size
suggests that loss of the scale may occur
in some individuals. Other than the in-
terparietal, the only important variation
is reduction from three to two toes on
the hind limb. The only specimen of /.
flavesccns examined has three toes on the
hind limbs and was taken on the west
side of the Rio Essequibo, Guyana, while
all specimens of /. schlegeli have two

-r-

69

66 63

^v

57

54

51

~\

-

;'' VENEZUELA

f
1

\

\
1

>

^FIAVESCENS

•i

^

fr-

1 ',

? B- / •

•

\

3-

1 '^'"'2S^

•

MONODACTYLUS

0-

1,

i

^

^-x ?

1

BRAZIL

•\

.^^

^ ^

"\

y \

'^

=L 1 _:i \ —

-V- -■

•^ J

1

P

'

3-

FiGtJRE 14. Distribution of B. flavesccns, B. monodactyhts and Bachia sp. in northeastern South

America.

40

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

toes on the hind hmb and occur on the
east side of the Rio Essequibo. Although
only 24 km separates the two localities,
the river may act as an effective barrier,
preventing unrestricted gene flow be-
tween the two subspecies.

Coloration. — Both subspecies of fia-
vescens have a dorsal coloration consist-
ing of three obscure dark brown lines
(one median, two dorsolateral), with ir-
regular edges. Occasionally, the dorsum
is heavily pigmented with dark brown
and the lines are no longer discernible,
and the yellowish-brown interspaces re-
duced to two linear rows of small ob-
scure spots. The head is usually brown
to blackish-brown, without invasion of
yellowish-brown ground color. The tail
may or may not have the dorsal dark
lines extending onto it. The sides are
dark brown to black and grade abruptly
into the yellowish-tan venter.

Remarks. — The presence of three su-
perciliaries in B. flavescens would be the
only character separating it from inono-
dactijlus if some populations of flavescens
lost the interparietal. Some individuals
of monodactylus have two toes on the
hind foot, one specimen having two toes
on one limb and three on the other.
There is complete overlap in the range
of body squamation between species. As
additional material becomes available,
the variation exhibited may show inter-
mediacy between the two species and
thus suggest that they are conspecific.
However, both species have been taken
from vicinity of Kartabo, Guyana, and
show no intermediacy at that locality.

Bachia monodactylus (Daudin)
Figure 13

Chalcides monodactylus Daudin, 1802:370
[Holotype. — lost, type locality and collector
unknown].

Colobtis daudini Merrem, 1820:76 [Replace-
ment name for Chalcides monodactylus
Daudin].

Chalcides trilineatiis Peters, 1872:775 [Holo-
type.— ZMB no number (lost) from South
America; collector unknown].

Bachia parked Ruthven, 1925:103 [Holotype. —
UMMZ 60813 from Chenapowu River "on

the upper Potato River," Guyana; collector
unknown]. New synon>aiiy.
Bachia monodactylus — Peters and Donoso-Bar-
ros, 1970:81.

Range. — Colombia to Cayenne and
south to Manaus, Brasil, South America
(Fig. 14). Literature records for Moreira,
Brasil (Cunha, 1958), Kartabo, Guyana
(Beebe, 1945), and Ananas Mountain,
Surinam (Brongersma, 1946) are refer-
able to Bachia m. monodactylus.

Content. — Two subspecies.

Key to subspecies. —

1. Two supraoculars present

m. monodactylus

One supraocular present _ tn. parkeri
Diagnosis. — Body and tail with quad-
rangular, juxtaposed scales; interparietal
absent; nasals, frontonasal, frontal, pari-
etals present; one or two supraoculars;
two superciliaries; second pair of chin
shields without medial contact nor reach-
ing oral border; no supralabial touching
parietal; forelimb with three clawed
toes; hind limb with single toe and limb
somewhat degenerate, or hind limb with
two definitive toes or combinations of
one toe on one limb, two on the other;
males with two preanal pores on each
side, absent in females; SAB 26-33 (x
29.2); dorsals 46-55 (x 49.6); ventrals
33-39 (x 35.5); gulars 7-8 (x 7.1); pre-
anal sliields 3 or 4; maximum snout-vent
length 80 mm.

Variation. — The number of toes on
the hind foot of B. monodactylus is
highly variable. The toes vary from an
almost styloform terminal scale, to two
distinct toes with an occasional third toe
on one of the legs. Specimens from the
central and eastern part of the species
range (Surinam, Cayenne) may have one
or two toes on each foot and occasionally
one on one foot, two on the other. Sam-
ples from northern Guyana have a sim-
ilar arrangement, whereas those from
central and southern Guyana have a
single terminal scale resembling a toe.
Specimens from the vicinity of Manaos,
Brasil, are similar to the latter except one
specimen with two toes on each hind
foot. By combining the number of toes

DIXON: SYSTEMATIC REVIEW OF BACHIA

41

on each hind foot and dividing by two,
the average number of toes per hind foot
for specimens from Guyana are 1.41,
Brasil 1.50, Surinam 1.62, and Cayenne
1.83.

The supralabials and infrahibials are
usually 5-5, but are occasionally 5-4 or
6-6. The preanal plate is normally com-
posed of three scales (34 specimens) but
occasionallv varies to four (10) or five
(2). SAB 'are 26-33 (x 29.2); dorsals
47-53 (x 49.6); ventrals 33-39 (x 35.5);
number of gular scales normally seven
(41 specimens), but occasionally eight
(5).

There is no significant difference be-
tween the number of SAB, dorsals, ven-
trals, and gulars of populations of B.
monodactyhis from Guyana, Brasil, Suri-
nam, and Cayenne. The means of SAB,
dorsals and ventral scales respectively,
are Guyana (28.4, 49.1, 35.2) Brasil
(26.8, 49.0, 34.4) Surinam (30.9, 50.4,
36.6) Cayenne (29.7, 49.7, 35.4).

The only character separating the
western populations of monodactyhis in
Colombia and extreme western Guyana
from those farther east and south is the
presence of a single supraocular in the
former. All other squamation characters
fall within the range of variation of the
total population. The shift from two
supraoculars to one is accomplished by
the expansion of the frontal scale, cre-
ating a reduction in size and fusion of
one of the supraoculars (usually the an-
terior one) with either the anterior
superciliary or the frontal. This type of
fusion has been observed in populations
of Bachia intermedia (dorhignyi group).

Bachia monodactyhis parkeri is
known only from the type series and is
presently allopatric from the nominate
race by only 18 airline kilometers.

Coloration. — The color pattern of
both subspecies consists of three distinct,
dorsal dark brown to black lines extend-
ing from the collar fold onto the tail;
the dark lines are separated by ground
color stripes of yellowish-tan to light
brown. The sides of the body are dark
brown to almost black, with the upper

edges frequently black, giving an allu-
sion of five blackish stripes or lines on
the upper surface of the body. Occasion-
ally the paravertebral dark lines are
broken into a zig-zag pattern with each
indentation yellowish-tan; the light
stripes are frequently freckled or heavily
mottled with dark brown. The head is
usually dark brown with two light stripes
reaching anteriorly to the edge of the
frontal. The venter is yellowish-tan to
brown and may or may not grade into
the dark sides.

Remarks. — Beebc (1945) stated that
specimens of this species taken in Guy-
ana during the dry season were found
under logs, jungle debris, and as deep as
30 cm in the soil while workers were
digging jungle pits. They were also
taken from the nests of leaf-cutter ants
{Attas sp.). During the rainy season
monodactyhis were found during early
morning hours crawling over wet leaves.
Specimens taken by M. S. Hoogmoed
(pers. com.) in Surinam were found in-
side of decaying tree trunks, under logs
and near openings in hills.

Beebe (1945) stated that stomachs
of some monodactyhis contained grubs,
ants, sowbugs, and termites. He found
one specimen of B. monodactyhis in the
stomach of a coatimundi.

Beebe (1945) also reported that one
specimen laid an egg 4.1 X 11.5 mm be-
neath a leaf. The shell was leathery with
slight longitudinal striae on the surface.

A single specimen of the flavescens
group from Timbo, Colombia, (Fig. 14)
cannot be allocated to any of the cur-
rently recognized populations of B.
monodactyhis. This male (MM R-3477,
Fig. 13) having a snout- vent length 50
mm, resembles monodactyhis by having
two supraoculars and two superciliaries,
but difli^ers in several other features. The
specimen represents the most fossorial
member of the flavescens group. The
hind limb is reduced to a small tubercle
with one minute apical scale (Fig. 15).
The forelimbs are short with two apical
scales resembling toes. One preanal pore

42

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTOIIY

on each side and preanal plate composed
of two large scales. The SAB (23) is
outside the range of monodactijlus (26-
33), whereas the dorsals (53) and ven-
trals (39) arc the highest extremes for
monodactijlus.

This specimen may represent the ex-
treme end of a cline from Manaos, Brasil,
to the Rio Vaupes, via the Rio Amazonas,
Rio Negro, and Rio Vaupes. Supporting
this hypothesis are specimens of mono-
docttjlus from Brasil having a degenerate
hind limb with one toe and the lowest
number of SAB (26) for the species.
However, the distance separating Ma-
naos and Timbo is an airline distance of
1300 kilometers, representing a formi-
dable allopatric gap.

flavescens

Direction of loss or
reduction of limbs & scales

Figure 15. Ventral view of the pelvic region of
members of the Bacliia flavescens group show-
ing progressive loss of body scales and hind
Hmb reduction.

SUMMARY

Bochia is one of three genera in the
family Teiidae lacking external ears; the
other genera are Anotosaura and Heter-
odactylus. The fifteen species of Bach'ia
exhibit differences in the type of body
scales, other external structural features,
coloration, and in certain features of
osteology. Most species of Bachia are
distributed in the semi-arid foothills and
tropical lowlands surrounding the Ama-
zon Basin, with some taxa occurring out-
side of the Basin on some of the Carib-
bean Islands, north of the Colombian
Andes to Costa Rica, but not on the
Pacific lowlands of South America. There
are areas of sympatry between species of
different species groups and only one
instance of sympatry between species of
the same group.

Evolutionary trends within the genus
are related to fossorial adaptations
within species and species groups. The
characters that showed the greatest
amount of specialization ( i.e., loss of
limbs, toes, various combinations of head
scales in sequence; modification of body
scales for greater flexibility and of neck
vertebrae for stronger vertical thrusts)
were utilized in an analysis of inter- and
intraspecific relationships. The basic
evolutionary trend is toward fossorialism

in each species group, and each group is
independently evohdng toward fossorial
adaptations at different rates and by dif-
ferent combinations of character modi-
fications.

The systematic statuses of the genera
Bacliia, Ophiognomon, Anotosaura and
Heterodactylus are clarified for the first
time. The possible origin of Bachia and
a definition of the genus is presented.
The variability in coloration, types of
body scales, size and number of limbs
and toes, and cephalic scjuamation, has
resulted in thirty names being proposed
for various species of the genus. Five
nominal species are herein relegated to
the synonymy of Bachia hcteropa (Lich-
tenstein, 1856); these arc Bachia lineata
Boulenger, 1903, Scolecosaurus allcni
Barbour, 1914, Scolecosaurus trinitatis
Barbour, 1914, Bachia anomala Roux,
1929, Bachia marcelae Donoso-Barros
and Garrido, 1964; to the synonymy of
Bachia trisanaJe (Cope, 1868), Hapalo-
lepis ahendrotJiii Peters, 1871 and Propus
vermifonnis Cope, 1874; to the syn-
onymy of Bachia flavescens (Bonnaterre,
1789), Chalcides schlegeli Dumeril and
Bibron, 1839; to the synonymy of Bachia
monodactijlus ( Daudin, 1802), Bachia
parkeri Ruthven, 1925.

DIXON: SYSTEMATIC REVIEW OF BACHIA

43

RESUxMEN

Bachia cs imo de los trcs gencros en
la familia Tciidac que caiecen de oido
externo; los otros generos son Anofosaura
y Heterodactyhis. Las quince cspecies
de Bachia niuestran difeiencias en el
tipo de escanias corporales, otras carac-
teiisticas externas, coloracion, y en cier-
tas extrueturas osteologicas. La mayoria
de las especies de Bachia se distribuyen
en el pie de monte semiarido y en las
bajas elevaciones tropicales que rodean
la cuenea amazonica, con algunas espe-
cies distribuidas en algunas de la islas
del Caribe, norte de los Andes Colom-
bianos hasta Costa Rica, pero no se en-
cuentran en las bajas elevaciones de las
costa del Pacifico en Sur America. Hay
areas simpatridas entre especies de difer-
entes grupos, y solo un caso de simpatria
entre especies del mismo grupo.

Las tendencias evolutivas se descifra-
ron por las adaptaciones intra- e intere-
specificas y entre grupos de especies, a
la vida subterranea. En el analisis de
las relaciones intra- e interespecificas se
usaron aquellos caracteres que mostraron
mayor especializacion (perdida de ex-
tremidades, artejos, varias combinaciones
en las secuencias de las escamas cefali-
cas; modificacion de las escamas cor-
porales para mayor flexibilidad, y de las

vertebras cervicales para excavar mcjor).
La tendencia evolutiva basica en cada
grupo esta evolucionando a diferentes
niveles y mediante diferentes combina-
ciones de modificacion es de los carac-
teres hacia este tipo de adaptacion.

Se clarifica por primera vez la posi-
cion sistematica de los generos Bachia,
Ophiognomon, Anotosaura y Heterodac-
tyJus. Se presenta el posible origen de
Bachia, asi como una definicion del ge-
nero. Debido a la variacion, en coloracion,
tipos de escamas corporales, tamaiio y mi-
emro de extremidades y artejos, escamas
cefalicas, en el pasado se ban propuesto
treinta nombres para varias especies del
genero. Se relegan cinco especies nomi-
nales al sinonimo de Bachia heteropa
(Lichtenstein, 1856); estas son Bachia lin-
eata Boulenger, 1903, ScoJecosaiirus alleni
Barbour, 1914, Scolecosaurus trinitatis
Barbour, 1914, Bachia anomaJa Roux,
1929, y Bachia marcelae Donoso-Barros
y Garrido, 1964; al sinonimo de Bachia
trisanale (Cope, 1868), Hapalolepis
abendwthii Peters, 1871 y Propus vermi-
fonnis Cope, 1874; al sinonimo de Bachia
flavescens (Bonnaterre, 1789), Chalcides
sch]e<ie]i Dumeril v Bibron, 1839; al sino-
nimo de Bachia monodactyhis (Daudin,
1802), Bachia parked Ruthven, 1925.

SPECLMENS EXAMINED

The arrangement of the localities for
each of the specimens examined is as
follows: alphabetically by country, state
(department or province), and locality;
alphabetically by the first letter in the
abbreviations for the museums, and nu-
merically after each museum abbrevi-
ation. Specimens lacking precise locality
data are listed last in the most restricted
political unit possible. Where more than
one specimen is included under one mu-
seum number, the number of specimens
is given in parentheses after the museum
number.

Bachia sp.
COLOMBIA: Vaupes: Timbo, near Mitu
on Rio Vaupes, MM R-3477.

Bachia harbotiri

PERtJ: Amazonas: 19 km S Bagua, LACM
493.3.3-39; 3 km S junction Bagua and Bagua
Grande road, LACM 49347-50, TCWC 28899,
2890.5-08; 3.5 km SE Bagua Grande, LACM
49344-46, TCWC 28903-04; 8 km WSW Bagua
Grande, MYZ 82444-45, 82449-50; 15.5 km W
Bagua Grande, LACM 49340-43, TCWC 28900-
02; 1.2 km NE Milagro, LSUMZ 19375-77; 5
km SW Milagro, LSUMZ 19378; 14 km NE
Milagro, LSUMZ 19379; 5.2 N Rio Maranon
bridge, LACM 49,351, TCWC 28909. Caja-
marca: Bellavista, MCZ 14740-41, 14843-44,
14746, 14748-49, 14751, 14754-55, 14758,
USNM 120788; 6 km N Bellavista, LSUMZ
19373-74; Perico, AMNH 28437, MCZ 32770.

Bachia hicolor
COLOMBIA: Atlantico: Barranquilla,
BMNH 1961.1862-63; MCZ 58784; Cienega

44

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

de Amajahuevo, FMNH 165776. Bolivar: Car-
tagena, BMNH 1928.1.5.1, FMNH 165774-75.
Ctindinamarca: Bogota, AMNH 9544-45 (prob-
ably in error). Magdalena: Bolivar, UMMZ
54767; Loma Larga, UMMZ 54776 (3); Mag-
dalena, MCZ 29325; Palomina, MCZ 6559;
Finca El Aranar, Bonoa, FMNH 165777; Rio
Frio, MCZ 25053, 29715; II km W Santa
Marta, CAS 116302; no specific locality MNHM
2842. Norte de Santander: Cucuta, USNM
84971. Stir de Santander: S of Bucaramanga,
AMNH 91765; Bucaramanga, AMNH 92994-96.

Bachia hresslani
BRASIL: Mato Grosso: Utiariti, Rio Papa-
gaio, MZUSP 10300.

Bachia dorhigniji

BOLIVIA: Santa Cruz: El Bcni: Esperanza,
BMNH 1920.11.29.10; Rio Sirutu UMMZ
63782; Tumupasa, AMNH 22525; no specific
locality BMNH 1904.10.29.17. La Paz: Bar-
ranca, on Rio Madidi, BMNH 98.6.9.9; Buena-
vista, BMNH 1927.8.1. 16S-170, UMMZ 60513
(2), 60515 (3), 60520 (2), 60587 (5), 60589
(4), 60590 (2), 63083, 63781 (9), 68082 (3),
68083 (2), 68084-85, 68086 (2) USNM 94095,
FMNH 35588-89; Santa Cruz MNHN 2841
(type ) .

PERtJ. Madre de Dios: near Avispes,
FMNH 154831; La Union, Carabaya, BMNH
1907.5.7.2; Itahauania, FMNH 81368; Mare-
apata, FMNH 62124, 62940.

Bachia flavescens flavescens
GUYANA: Kartabo UMMZ 65170.

Bachia flavescens schlegeli
GUYANA: Moraballi, on Rio Essequibo

BMNH 19.30.10.10/167. No specific locality,

SMF 39900 (type of boettgeri).

"INDIA": Calcutta (in error) RMNH 3580

(type of schlegeli).

Bachia heteropa alleni

GRENADA: Saint George, AMNH 38968,
CAS 39462-63, USNM 43218-19; Amnandale
Estate, USNM 79190; No specific locality,
BMNH 96.6.5.1-2; USNM 67217, 67219-21.

GRENADINES: Bequia Island, Spring Es-
tate, AMNH 90516.

SOUTH AMERICA: No specific locality,
BMNH 53.2.4.68, 1961.1859; MNHM 2836.

TOBAGO: Easterfield road, NE of Mason
Hall in the vicinity of milestone 2-3/4, USNM
167514; No specific locality, FMNH 55392.

Bachia heteropa heteropa
VENEZUELA: Sucre: 7.6 km S Junction
Casanay, Capipito-Maturin road, KU 117099.

Bachia heteropa heteropa X heteropa trinitatis
VENEZUELA: Sucre: Capipito, (AMNH;
Beebe field No. 30061?).

BacJiia heteropa lineata
VENEZUELA: Caracas, RMNH 3579;
Duaca, BMNH 1962.7.29.90; El Mene, north of
Rio Tocuya, BMNH 1928.12.12.8; Pauji,
MZUSP 3010-11; San Esteban, Ravina de
Palmas, UMMZ 55880.

Bachia heteropa trinitatis
TRINIDAD: Upper Arima Valley, AMNH
81486; Blanchiseuse, MCZ 55675; Brickfield,
FMNH 49873-79, 49881-84, 49886-90, 49893,
49896-97; Caparo, AMNH 1608, 38815, CAS
39481-82; Chacachacare, MCZ 79117; Maracas
Valley, MCZ 79819; Morne Bleu, AMNH 72852;
Mt. Saint Benedict, FMNH 25013, 40000; Mt.
Tucuche, MCZ 32521; Points— Pierre, MCZ
49062-63; Saint Augustine, FMNH 40447,
55390-91; Saint Mary's Gasparee Island, MCZ
79118; Tucker Valley, AMNH 64458 (3), 64522
(4); No specific locality, FMNH 10900.

VENEZUELA: Orinoco, RMNH 3578,
MNHM 2837. Sucre: CeiTo Azul, 10 km N
Macuro, KU 133498-500, 133501 (2).

WEST INDIES: No specific locality,
BMNH 1961.1860.

Bachia huallagana
PERtJ: Hudnuco: 27.4 km SSE Aucayacu,
UF 28143-44; Tingo Maria, USNM 192993-95,
193005-07; San Martin: San Jose de Tocache,
USNM 192296-3004; upper Biabo Valley,
AMNH 56566.

BacJiia intermedia
PERU: Cajamarca: Bellavista, USNM
120787; 9 km S Jaen, LSUMZ 19381; 5 km
SSE Jaen, MVZ 92867; Perico, AMNH 227.30,
284.38-59, BMNH 1946.8.31.94, CAS 54615,
54627, MCZ 14701-26; 28 km N Santa Cruz,
LSUMZ 19382-416.

Bachia monodactijlus monodactijlus

BRASIL: Amazonas. Manaos, AMNH 64869-
71, MZUSP, 471, 10912, 8353; Km 50, between
Manaos and Itacotiara, MZUSP 8352. Para:
Ilhade Marajo, BMNH 1923.11.9.88; Rio Cu-
mca, MCZ 3782.

CAYENNE: No specific locality, BMNH
53.11.22.16a-b, MNHM 2839, 2840a-b, 1899.76,
1903.229, 1928.109-110, 2838.

GUYANA: Bartica, AMNH 21269; Esse-
quibo, BMNH 1970.714; Kaburi, MCZ 81179;
Kamakusa, AMNH 25084-85; near mouth of
Guyani River, BMNH 1930.10.10.166; Macca-
suma, BMNH 87.1.22.8-9; New River, BMNH
1939.1.1.76-80; Pickersgill, USNM 85012; Po-
taro, BMNH 1970.713; Potaro Road, 82 mi S
Bartica, BMNH 1934.11.1.106; Rupununi Dis-
trict, north of Acarahy Mts., KU 69801; No
specific locality, AMNH 60911.

SURINAM: Albina, RMNH 13411; Broko-
pondo, RMNH 13413; Coppername, RMNH
15216; Maratakka, RMNH 13414; Marowijne

DIXON: SYSTEMATIC REVIEW OF BACHIA

45

Basecamp, RMNH 13412a-b; near Papatam,
RMNH 15217a-h; Sipaliwini, near Sipaliwini
Ri\'er, RMNH 16377; Sipaliwini, 5 km NW
airstrip, RMNH 16434a-b; Sipaliwini, 4 km E
airstrip, RMNH 15183; No specific locality,
RMNH 3581.

VENEZUELA: Caracas, BMNH 51.7.17.4
(probably in error).

Bachia monodactijlus parkeri
COLOMBIA: Vaiipes: Rio Tiquia USNM

65437.

GUYANA: Chenapowu River, on the upper

Potaro River, BMNH 1946.8.42-43, MCZ 21685;

Demerara, Lama Creek, FMNH 23807.

Bachia pallidiceps

COLOMBIA. Antioquia: Villa Arteaga,
FMNH 78135. Choco: Quesada River, Atrato
region, AMNH 18230; Quibdo, BMNH 1909.
10.,30.37.

PANAMA. Chiriqui: Puerto Armulles,
ANSP 21773. Darien: Rio Tuira at Rio Mono,
KU 96868-78; paradise Camp on Rio Morti,
FMNH 170025-26; Yavisa, MCZ 37750.

SOUTH AMERICA. No specific locality,
MNHM 7051.

Bachia panoplia

BRASIL: Amazonas: Manaos, AMNH
64872-80, 64882, MZUSP 10910-11.

COLOMBIA: Vaupes: Timbo, MM R-
3621.

Bachia peruana

PERIJ: Ayacucho: Between Petaccocha
and San Jose, along the Rio Santa Rosa,
LSUMZ 25505-08. Hudnuco: Rio Llulla
Pichis, AMNH 104278. Junin: Chanchamayo
(Valley), AMNH 56560-63, FMNH 40585; Pe-
rene (Valley), AMNH 23212, 23214-15; San
Ramon, FMNH 134399. Loreto: Colonia Cal-
leria on Rio Calleria, CAS 93215-32, 104365;
Orellana, AMNH 56564; Roaboya, AMNH
56557-59, 56565.

Bachia scolecoicles

BRASIL: Mato Grosso: Rio Teles Pires,
AMNH 86955, MZUSP 3289-94.

Bachia talpa
COLOMBIA: El Cesar: Valledupar, UMMZ
54769-71. La Guajira: Fonseca, UMMZ 54768.
Magdalena: Valencia, UMMZ 54772-73.

Bachia trisanale ahendrothi

PERtJ: Junin: Chanchamayo Valley, MCZ
82898. Loreto: Balta, LSUMZ 25502-03; Orel-
lana, USNM 1271,36; Rio Ucayali Valley, near
mouth of Rio Cushabatay, AMNH 56576,
56581; Sarayacu, BMNH 1946.8.2.14 (syntype),
81.5.13.26-27, MCZ 21069; Selvas del Rio,
FMNH 40019; Yarinacocha, FMNH 45473,
55976-78, LSUMZ 25501, 25504; Pucallpa,
FMNH 55979; no specific locality, AMNH
56582.

Bacliia trisanale trisanale

ECUADOR: Napo or upper Maraiion,
ANSP 9637 (type); "oriente," UMMZ 82875.
Chimhorazo: Riobamba, AMNH 14574 (in er-
ror). Morona-Santiago: Macas, AMNH .38809;
2 mi E Sucua, USNM (JAP 2202). Napo: Lago
Agrio, KU 126812-31, 126848, UMMZ 129344;
Puerto Libre, 122194-95; Santa Cecilia, KU
126811; near Tena UMMZ, 84738. Pastaza:
Limon, USNM (JAP 6900).

PERtJ: 5 km NE Aramargo, RT 286-87
Rio Comania, Maranon Valley, AMNH 56580
Quaracayo Pongo, AMNH 56578. Amazonas
15 km CW^ Chiriaco, RT 318-20, 416-17.

Bachia trisanale vermiformis
PERtJ: Loreto: Nauta, ANSP 11.353 (type).

Bachia trisanale trisanale X trisanale
vermiformis

PERtJ: Loreto: Iquitos, MCZ 123709;
Mishana, TCWC 38117-18; Moropon, TCWC
36715, 39028; Rio Itaya, AMNH 56567-75,
56577, 56579.

LITERATURE CITED

Alemax, G. C.

1953. Contribucion al estudio de los rep-
tiles y batracios de la Sierra Perija.
Mem. Soc. Cienc. nat. la Salle,
13 (15): 205-225.

Amaral, a.

1933. Estudos sobre Lacertilios neotropics.

I. Novos generos e especies de la-
gartos do Brasil. Mem. Instit. Bu-
tantan, (1932)7:5.3-74.

19.35. Estudos sobre Lacertilios neotropics.

II. Novo genero e especie do lar-
garto do Brasil. Ibid, 9:249-250.

Barbour, T.

1914. A contribution to the zoogeography
of the West Indies, with special ref-
erence to the amphibians and reptiles.
Mem. Mus. Comp. Zool., 44:209-359.

1933. Notes on Scolecosaurus. Copeia, 1933
(2): 74-77.

Beebe, W.

1945. Field notes on the lizards of Kartabo,
British Guiana, and Caripito, Vene-
zuela. 3. Teiidae, Amphisbaenidae
and Scincidae. Zoologica, 30 ( 1 ) : 2-
32.

46

MISCELLANEOUS PUBLICATION MUSEUM OF NATURAL HISTORY

BONNATERRE, P. J.

1789. Tableau encyclopedique et metho-
dique des trois Regnes de la Nature,
Erpetologie. Paris, pp. 1-71.

BOULENGER, G. A.

1885. Catalogue of the lizards in the Brit-
ish Museum (Natural History). Lon-
don, Vol. II, pp. 1-497.

1887. Catalogue of the lizards in the Brit-
ish Museum (Natural History). Lon-
don, Vol. Ill, pp. 1-575.

1903. Descriptions of new lizards in the
collection of the British Museum.
Ann. Mag. Nat. Hist., 7:429-35.

Brongersma, L. D.

1946. Some notes on species of the genera
Bachia and ScoJccosaurtis. Zool.
Meded. Leiden, 26:237-245.

1956. On some reptiles and amphibians
from Trinidad and Tobago, BWI, I-
II. Proc. ned. Akad. Wet., 596
(2): 165-188.
Burt, C. E. axd M. D. Burt

1931. South American lizards in the collec-
tion of the American Museum of
Natural History. Bull. Amer. Mus.
Nat. Hist., 61:315-323.

1933. A preliminary check list of the lizards
of South America. Trans. Acad. Sci.,
St. Louis, 28(1):1-104.

Camp, C. L.

1923. Classifications of the lizards. Bull.
Amer. Mus. Nat. Hi,st., 48:289-481.

Cope, E. D.

1862. Catalogues of the reptiles obtained
during the explorations of the Parana,
Paraguay, Vemiejo and Uruguay
rivers, by Capt. Thos. J. Page, U.S.N.,
and those procured by Lieut. N.
Michler, U.S. Trop. Eng., Com-
mander of the expedition conducting
survey of the Atrato River. Proc.
Acad. Nat. Sci., Philadelphia, 14:346-
359.

1868. An examination of the Reptilia and
Batrachia obtained by the Orton E.x-
pedition to Ecuador and the Upper
Amazon, with notes on other species.
Ibid, 20:96-140.

1874. Description of some species of rep-
tiles obtained by Dr. John F. Brans-
ford, Assistant Surgeon U.S.N., while
attached to the Nicaraguan Surveying
Expedition in 1873. Ibid, 26:70-72.

1896. On the hemipenes of the Sauria.
Ibid, 48:461-67.

1899. Contributions to the herpetology of
New Grenada and Argentina, with
descriptions of new forms. Philadel-
phia Mus., Sci. Bull., (1):1-19.

Crawford, S. C.

1931. Field keys to the lizards and amphib-
ians of British Guiana. Ann. Carnegie
Mus., 21:11-42.

CUNHA, O. R. DA

1958. I. Lacertilios da Amazonia. Sobre a
ocrrencia do genero Bachia Gray,
1845, na Amazonia Brasileira ( Lacer-
tilia-Teiidae ) . Bol. Mus. Goeldi
(Zool.) 11:1-12.
Daxiel, H.

1955. Algunos aspectos de la lucha bio-
logica. IV, Los Reptiles y la Agri-
cultura una seipiente bicefala. Como
se han clasificado las serpientes. Rev.
Fac. nac. Agron. Medellin, 16 (48)3-
168.
Dauoin, F. M.

1802. Histoire naturelle, gencrale et par-
ticuliere des reptiles. Paris, Vol.
4:367-70.
Dixox, J. R.

1972. Natural hybrids between the South
American microteiid species Bachia
barboiiri and Bachia intermedia, J.
Herp., 5:205-07.
Doxoso-Barros, R.

1968. The lizards of Venezuela (check list
and key). Carib, J. Sci., 8:105-122.
AXD R. Garrido

1964. Nuevo Teiidae de Venezuela, Bachia
marcelae nov. sp. Publ. Ocas. Mus.
Cience. Nat., Caracas (Zool.) 8:1-7.

DUMERIL, A. M. C. AXD G. BlBROX

1839. Erpetologie Generale ou histoire
naturelle complete des reptiles. 5:1-
854. Paris.
Duxx, E. R.

1940. New and noteworthy herpetological
material from Panama. Acad. Nat.
Sci., Phila., 92:105-122.

1944. Los generos de anfibios y reptiles de
Colombia, II. Reptiles, Orden de las
Saurios. Caldasia, 11:73-110.

FiTZIXCER, L.

1826. Neue Classification der Reptilien nach
ihren naturlichen Ver^vandtschaften.,
V^ien, pp. 1-66.
1843. Systema Reptilium. Wien, pp. 1-106.
Garma.x, S.

1892. On the reptiles collected by George
Baur near Guayaquil, Ecuador. Bull.
Essex Inst., 19:1-12.
Gray, J. E.

1845. Catalogue of the specimens of lizards
in the collection of the British Mu-
seum, London, pp. 1-289.
Hellmich, W.

1960. Die Sauria des Gran Chaco und sei-
ner Randgebiete. Abh. bayer Akad.
Miss.-Math.-Nat. K.L. N.F., 101:1-
131.

DIXON: SYSTEMATIC REVIEW OF BACHIA

47

Lacepede, B. G.

1789. Histoire naturelle des qiiadrupcdes
oxipares et des Serpents, I & II, 1178
pp., Paris.

LiCHTEXSTEIX, H.

1856. Nomenclator Reptiliuni et Amphibi-
oruni. Musei Zoologici Berolinensis,
pp. 1-48.

LOVERIDGE, A.

1933. On Bacliia intermedia Noble and
Bachia harbouri Burt. Copeia, 1933
(1):42.
Merrem, B.

1820. Versuch eines Systems der Aniphib-
ien. Marburg, pp. 1-191.

Noble, G. K.

1921. Two new lizards from northwestern
Peru. New York Acad. Sci., 29:141-
143.
Parker, H. W.

1928. Notes on Reptiles and Batrachians
from Mato Grosso and Eastern Bo-
livia. Ann Mag. Nat. Hist. (10)2:96-
99.

1935a. The frogs, lizards and snakes of Brit-
ish Guiana. Proc. Zool. Soc. London,
1935 (2):505-530.

1935b. The lizards of Trinidad. Trop. Agric.,
(1936)12(3):65-70.

Peracca, M. G.

1896. Rettili ed Anfibi raccolti nel Darien
ed a Panama dal Dott. E. Festa. Boll.
Mus. Torino, (253): 12.

1897. Intomo ad una piccola raccolta di
rettili di conoacco (peru Orientale).
Ibid, 12(284):1.

Peters, W.

1871. Ueber eine von Hrn. Dr. Robert
Abendroth in dem Hochlande von
Peru gemachte Samnilung von Am-
phibien. Monatsber. K. Preuss. Akad.
Wiss. Berlin, 1871:387-404.

1872. Ueber einige Arten der herpetolo-
gischen Sammlung des Berliner zo-
ologischen Museums. Ibid, 1871:644-
652.

ReINHARDT, J., AND C. LuTKEX

1861. Bidrag til det. vestindiske Origes og
navnligen til de dansk- vestindiske
Oers Herpetologie. \'idensk. Meddel.
nat. Foren ( 1861 ): 153-290.
Roux, J.

1929. Sur deux Reptiles nouveaux du Vene-
zuela. Verb. Nat. Ges. Basel., 40:29-
34,

RUTHVEX, A. G.

1922. The amphibians and reptiles of the
sierra Nevada de Santa Marta, Co-
lombia. Univ. Mich. Mus. Zool., Misc.
Publ., (8):l-69.

1925. Lizards of the genus Bachia. Proc.
Boston Soc. Nat. Hist., 28(3)101-109.
Schneider, J. G.

1801. Historiae Amphiborimi, II. Croci-
dilos, Scincos, Chamaesaurus, Boas,
Pseudoboas, Elapes, Angues, Am-
phisbaenas et Caecilias. Jena, pp. 1-
374.

SlIREVE, B.

1947. On Venezuelan reptiles and amphib-
bians collected by Dr. G. Kugler.
Bull. Mus. Comp. Zool., Hai^vard,
99(5):517-537.

SpIX, J. B. DE

1825. Animalia Nova sive Species Novae
Lacertaium Brasiliam. Monachi, pp.
1-26.
Thomas, R.

1965a. The smaller teiid lizards ( Gtjmnoph-
thalmus and Bachia) of the south-
eastern Caribbean. Prov. Biol. Soc.
Wash., 78:141-154.

1965b. A new species of Bachia (Sauria:
Teiidae) from Brasil. Heipetologica,
21(l):18-22.

Valdivieso, D., axd J. R. Tamsitt

1963. Records and observations on Colom-
bian reptiles. Ibid, 19:28-39.

Vaxzolixi, p. E.

1950. Bachia tridactyla ( Daudin, 1802 ) re-
placed by Bachia schlegeli ( Dumeril
& Bibron, 1839). Copeia, 1950(2):
51.

1961a. Redescricao de Scolecosaunis trinitatis
(Sauria: Teiidae) Pap. Dep. Zool. Sec.
Agric. Sao Paulo, 14:183-185.

1961b. Bachia: Especies Brasileiras e conceito
generico (Sauria: Teiidae). Ibid,
14:193-209.

1961c. On Ophiognomon trisanale and aben-
drotlii (Sauria: Teiidae). Ibid,
14:249-254.

1966. Sobre o segundo exemplar de BacJiia
hresslaui (Sauria: Teiidae). Ibid,
19:189-192.

Werner, F.

1901. Reptilien und Batrachier aus Peru
und Bolivien. Abb. Mus. Dresden,
9(2):1-14.

AVAILABLE MISCELLANEOUS PUBLICATIONS

IN HERPETOLOGY

UNIVERSITY OF KANSAS PUBLICATIONS

9. Handbook of amphibians and reptiles of Kansas. By Hobart M. Smith. 2nd
[revised] edition. Pp. 1-356, 253 figures in text. March 10, 1956. Paper bound
$1.50 postpaid.

52. Reproductive cycles in Uzards and snakes. By Henry S. Fitch. Pp. 1-247, 16 fig.
ures in text. June 19, 1970. Paper bound $5.00 postpaid.

53. Evolutionary relationships, osteology, and zoogeography of leptodactyloid frogs.
By John D. Lynch. Pp. 1-238, 131 figures in text. June 30, 1971. Paper bound
$7.00 postpaid.

55. Middle American lizards of the genus Ameiva (Teiidae) with emphasis on geo-
graphic variation. By Arthur C. Echtemacht. Pp. 1-86, 28 figures in text. De-
cember 14, 1971. Paper boimd $3.00 postpaid.

57. A Systematic review of the teiid lizards. Genus Bachia, with remarks on Heter-
odactylus and Anotosaura. By James R. Dixon. Pp. 1-47, 15 figures in text.
February 2, 1973, Paper bound $1.50 postpaid.

MONOGRAPHS

1. The hyhd frogs of Middle America. By William E. Duellman. Pp. i-xi, 1-753,
324 figures in text, 72 plates (32 in color). December 15, 1970. Cloth bound in
two volumes $25.00 postpaid.

p

u^^:me

ioakbindins Co.; Inej

300 Summer Shwf

Boston 10, A^ass."

3 2044 093 361 624

Date Due