BREVIORA

Museum of Comparative Zoology

OS: CS

<2
BONN

ie

Ann en

a
sss
ple

US ISSN 0006-9698

CAMBRIDGE, MASs.

5S DECEMBER 2014

NUMBER 539

NEAVES’ WHIPTAIL LIZARD: THE FIRST KNOWN TETRAPLOID
PARTHENOGENETIC TETRAPOD (REPTILIA: SQUAMATA: TEITDAE)

CHARLES J. CoLe,! HARRY L. TAyLor,?2 DIANA P. BAUMANN,? AND PETER BAUMANN?

ABSTRACT.

The first known tetraploid amniote that reproduces through parthenogenetic cloning by individual

females is named and described. The species originated through hybridization between Aspidoscelis exsanguis
(triploid parthenogen) X Aspidoscelis inornata (diploid bisexual or gonochoristic species) in the laboratory. We
compared multivariate morphological variation between two lineages that arose from separate F; hybrid zygotes in
one clutch and among several generations in those lineages. The tetraploid species is also compared with its ancestral
taxa, with two hybrids of A. exsanguis X A. inornata that were found in nature at two localities that are 100 km apart
in southern New Mexico, and with three laboratory hybrid males. This will facilitate identification of field-caught

tetraploids in the future.

KEY WORDS:

INTRODUCTION

Among amniotes, true parthenogenesis
(initiation and completion of embryogenesis

‘Division of Vertebrates (Herpetology), American
Museum of Natural History, 200 Central Park West,
New York, New York 10024, U.S.A.; e-mail: cole@
amnh.org.

>Department of Biology, Regis University, Denver,
Colorado 80221, U.S.A.; e-mail: htwhiptail@gmail.com
* Stowers Institute for Medical Research, 1000 East 50th
Street, Kansas City, Missouri 64110, U.S.A.; e-mail:
dpb@stowers.org

“Howard Hughes Medical Institute and Stowers Insti-
tute for Medical Research, Kansas University Medical
Center, Kansas City, Missouri 64110, U.S.A.; e-mail:
peb@stowers.org

Aspidoscelis neavesi; new species; tetraploid; parthenogenesis; clonal lineages

in the absence of spermatozoa) occurs as the
normal mode of reproduction in only a few
species, all of which are reptiles (several
lizards and apparently one snake; reviewed
in Dawley and Bogart, 1989; Lutes ef al.,
2011; Neaves and Baumann, 2011). Many of
these are whiptail lizards (Aspidoscelis), of
which the unisexual (all-female) lineages
receive formal recognition as species because
they are of unique ancestry (reviewed by
Reeder et al., 2002) and individuals repro-
duce by parthenogenetic cloning (Lutes et
al., 2010, 2011). Additionally, rare instances
of hybridization and fertilization of a par-
thenogenetic female’s cloned eggs have re-
sulted in triploid clonal species of unique

© The President and Fellows of Harvard College 2014.

to

ancestry (e.g., Aspidoscelis exsanguis; re-
viewed by Reeder et al., 2002). Although a
few tetraploid hybrid individuals were re-
ported in the past (Lowe ef al., 1970a;
Neaves, 1971), no tetraploid clonal lineages
were known until Lutes et a/. (2011) reported
the laboratory origin of the species for which
we provide a name, morphological descrip-
tion, and both intraspecific and interspecific
comparisons here.

Aspidoscelis is a genus of North American
whiptail lizards that includes several unisex-
ual (all-female), parthenogenetic species as
well as gonochoristic species (reviewed by
Reeder et al., 2002). All of the unisexual
species ultimately had a hybrid ongin, the
females clone themselves, and the primary
lineages bear formal binomials, although
derived clonal lineages with postformational
mutations usually do not. Although the
International Code of Zoological Nomencla-
ture (ICZN, 1999) prohibits naming hybrids
(Article 1.3.3), this restriction applies to
individual animals that are hybrids between
two species, not self-perpetuating clonal
entities that are unique evolutionary lineages.
In the case of unisexual species of Aspidosce-
lis, historically, the F,; hybrid females estab-
lished continuing lineages by cloning them-
selves parthenogenetically, and the offspring
continued to reproduce as did their mother.
This quantum leap in evolution (the switch
from spermatozoan-dependent to spermato-
zoan-independent reproduction) occurred in
each case in just one generation (reviewed by
Reeder et a/l., 2002). In such lineages, individ-
uals of the F, and subsequent generations are
not hybrids but clones of their single parent.
In recognition of this, ICZN (1999) adopted
Article 17.3, validating the naming of parthe-
nogenetic entities of hybrid origin.

Some clonal lineages of hybrid origin are
morphologically cryptic species, but many
have distinctive morphologies in size, color
pattern, and/or scalation. Consequently, sev-

BREVIORA

No: 539

eral were diagnosed, described, and named
well before scientists knew that unisexual,
clonal lizards exist (e.g., Ameiva tesselata [Say
in James, 1823:50-51], known today as
Aspidoscelis tesselata). Each clone with a
distinctive ploidy and distinctive combination
of ancestral genomes has been named, reflect-
ing its unique historical origin (reviewed in
Reeder ef al., 2002). Additionally, some
clones with distinctive scalation and/or color-
ation derived as a consequence of postforma-
tional mutations have been named even
though they share a common hybrid origin
with other clones (e.g., Aspidoscelis maslini
Fritts, 1969, versus other clones of the
Aspidoscelis cozumela complex; Taylor et al.,
2005). Coauthors of the present paper differ
in philosophy and practice on this point, but
in general, we do not favor naming post-
formational clones but prefer to treat them as
a complex of derivative forms under one
specific epithet (e.g., the A. tesselata com-
plex), which clearly reflects their relation-
ships. This is similar to recognizing that
extensive genetic variation occurs within
named species of gonochoristic taxa.

In this paper we name, diagnose, and
describe a unique unisexual species of hybrid
origin. It is the first tetraploid vertebrate
known to clone itself parthenogenetically
and it is reproductively isolated from all
other species (Lutes et al., 2011). The name,
given below, will provide for efficient, effec-
tive, and unambiguous communication, par-
ticularly for tracking data among various
publications, as this model organism 1s being
used for considerable research on_ basic
biological processes (e.g., molecular aspects
of meiosis; Lutes et al., 2010, 2011) and DNA
sequencing, which involves listing in on-line
data information services such as GenBank,
RefSeq, and UniProt.

Although the presently known reproduc-
ing lineages of this tetraploid species origi-
nated in the laboratory by hybridization

2014

between the triploid parthenogenetic 4A.
exsanguis (Q) and Aspidoscelis inornata (°C;
a diploid bisexual species), this same combi-
nation of chromosomes was found in a
tetraploid female (MCZ 101991) from Ala-
mogordo, Otero County, New Mexico, in
1967 by William B. Neaves, and the female
laid eggs in captivity (Neaves, 1971). At the
time, however, the female oviposited in dry
sand, and the eggs became desiccated and
were discarded, although otherwise they
looked normal. A second relevant specimen
(UCM 29196) was found near Mesilla, Dona
Ana County, New Mexico (Taylor et al.,
1967). This one was a male and was
identified as representing the all-female A.
exsanguis. However, we hypothesize that this
specimen is also a tetraploid of hybrid origin
between A. exsanguis and A. inornata, because
eggs of whiptail lizards that receive a Y-
bearing spermatozoan produce males, regard-
less of ploidy level (Cole et al., 1969; Lowe et
al., 1970a; Taylor et al., 2001). We borrowed
and examined both of these specimens from
New Mexico and included them in our
comparisons below.

In addition to describing the unique
tetraploid species in this paper, we present
morphological data for individuals of sev-
eral lineages. These were cloned from F,
zygotes of a single pair of parents (A.
exsanguis X A. inornata). We also compare
samples of two well-represented lineages
with each other (including generation-to-
generation comparisons) and compare the
tetraploids to samples of their parental
species and specimens of similar hybrid
ancestry found in nature.

MATERIALS AND METHODS

Specimens. Methods for maintaining the
lizards in a captive colony were described by
Lutes et al. (2010, 2011). Identity of individ-
uals is tracked as follows: the individual

NEAVES’ WHIPTAIL LIZARD 3

mother of each egg clutch is noted (as is the
father, if applicable), hatchlings are noted as
to clutch, and throughout life, individuals
are photographed periodically and tracked
as to which enclosures they occupy at any
time. Lineage membership is confirmed using
microsatellite DNA analysis (e.g., Lutes ef
Gi ZOEY,

Morphological Characters Examined. These
are listed in Appendix 1, and specimens
examined are in Appendix 2. Nomenclature
for epidermal scales follows Smith (1946). Sex
was determined by dissection and examina-
tion of primary sexual characters, a history of
Oviposition, or examination of external sec-
ondary characters.

Museum abbreviations for specimens exam-
ined: AMNH, American Museum of Natural
History, New York, New York; MCZ,
Museum of Comparative Zoology, Harvard
University, Cambridge, Massachusetts:
MSB, Museum of Southwestern Biology,
University of New Mexico, Albuquerque,
New Mexico; SIMR, Stowers Institute for
Medical Research, Kansas City, Missour1;
UCM, University of Colorado Museum,
Boulder, Colorado.

Multivariate Statistical Analyses. A\l-
though we had 177 specimens available for
this study, we could use only specimens with
complete data for the suite of 10 meristic
characters analyzed. Therefore, as required
by the procedures, 65 specimens were ex-
cluded because of one or more damaged or
missing characters. This left us with the
following numbers of specimens for principal
components analyses (PCAs) and canonical
variate analyses (CVAs): tetraploids (90 of
130); A. exsanguis (10 of 25); A. inornata (10
of 20); field-caught, presumed or known 4n
hybrids (1 of 2); and laboratory F, tetraploid
hybrid males (1 of 3).

We based our comparisons of lineages,
generations, and taxa on PCAs of meristic
characters (Appendix 1) followed by CVAs

- BREVIORA

of the principal components. As an alterna-
tive to using the raw meristic data for CVAs,
the benefits and rationale for using a PCA
intermediary were detailed by Jombart et al.
(2010). We used samples of the species as a
priori classified groups and stepwise selection
of principal components for inclusion in
CVA models. Principal components were
added if F-to-enter probabilities were <0.05
and did not exceed 0.06 when other compo-
nents were included in the model. Statistical
procedures and tests were performed with
SPSS® and NCSS® software.
We used pairwise Mahalanobis D? distanc-
es to quantify the relative meristic resem-

blances among the tetraploids and progenitor
- species A. exsanguis and A. inornata. Distanc-
es were calculated from D2 = [q(N — g)(Na+
Nb)V/[(N — g — g+1)NaNb] X Fab. In this
equation (Neff and Marcus, 1980), a and b
indicate the two groups being compared, g
is the number of predictors (characters), N
is the total number of individuals across all
groups, g is the total number of groups, Na
is the number of individuals in group a, Nb
is the number of individuals in group b, and
Fab is the F-statistic comparing groups a
and b. F-values were from the matrix of F-
values and probabilities generated when the
last qualifying principal component had
been included in the CVA model by
stepwise variable selection (provided as
output by SPSS and SYSTAT® statistical
software).

Mahalanobis D2? distances are sensitive to
multivariate outliers (Tabachnick and Fidell,
2013). Therefore, we checked each a priori
group for multivariate outliers by evaluating
D2 distances from each specimen to the
centroid of the remaining cases in that group
(provided as output by SPSS). Specimens
with D2 values exceeding a critical chi-square
value at P = 0.001 and degrees of freedom
defined by the number of principal compo-
nents included in the CVA model would be

No. 539

identified as outliers (Tabachnick and Fidell,
2013), but none was found in this study.

We used ¢ tests to test tetraploid lineages
of AMNH R-176077—176148 (= SIMR 4921
lineage) and the lineage of MCZ R-192209
(= SIMR 4919) for significant differences
and one-way analyses of variance (ANO-
VAs) to check for differences among lineage/
generation combinations of tetraploids. We
also used one-way ANOVAs to check for
differences among samples of tetraploids, A.
exsanguis, and A. inornata, including canon-
ical variates 1 and 2 (CV1 and CV2). We
followed ANOVAs that indicated the pres-
ence of significant differences with Tukey
multiple comparison tests to identify the
significantly different groups.

THE NEW SPECIES

Aspidoscelis neavesi, new species
Neaves’ Whiptail Lizard
Figures. 12-2

Holotype. MCZ R-192219 (= SIMR 8093),
a cloned adult female of the F, laboratory-
reared generation that also cloned herself at
the SIMR. She hatched on August 13, 2008,
and her mother was MCZ R-192209 (=
SIMR 4919). See Appendix 2 for a partial list
of her offspring.

Paratypes. See Appendix 2, Specimens
Examined. Each individual of A. neavesi
other than the holotype is a paratype, except
for two that are still alive.

Diagnosis. A species of the Aspidoscelis
sexlineata species group as reviewed by Lowe
et al. (1970b). The species is distinguished
from all others in the genus by the following
combination of characters: abruptly enlarged
mesoptychials; enlarged postantebrachials; 2
frontoparietals; usually 3 parietals; 4 suprao-
culars each side; unisexual (only females
exist); body with 6 longitudinal pale stripes
that fade and can disappear in large adults;
hatchlings basically unspotted but adults

2014

Upper. Adult holotype of A. neavesi
(MCZ R-192219 [= SIMR 8093]) in life, January 11,
2011, SVL = 77 mm. Lower. Adult A. exsanguis (SIMR
13209) of Alamogordo, Otero County, New Mexico,
stock, SVL = 76 mm.

Figure’ 1.

with pale dots and spots on body; maximum
snout-vent length (SVL) about 80 mm;
tetraploid number of chromosomes about
91, with 4 haploid sets of the sexlineata
species group, including the slightly modified
triploid karyotype of some A. exsanguis from
Alamogordo (Lutes et a/., 2011), and the 3
largest chromosomes being metacentric.
Description of Holotype. Paired data pre-
sented in the form x-y are for scale counts on
the left-right sides of the body. SVL, 78 mm;
rostral large, visible from above, wider than
high; nostril low, posterior to center of nasal;
nasals with a long median suture behind the
rostral; frontonasal hexagonal; pair of irreg-
ularly hexagonal prefrontals with a long
median suture; frontal hexagonal, longer
than wide, wider anteriorly than posteriorly;
pair of irregularly pentagonal frontoparietals
with a long median suture; 3 irregular-sized
and irregular-shaped parietals in a transverse

NEAVES’ WHIPTAIL LIZARD 5

Figure 2. Ontogenetic changes in colors and pattern
for two young individuals of A. neavesi for comparison
with Fig. 1, upper. Upper. MCZ R-192243 (= SIMR
10400), SVL = 36 mm. Lower. SIMR 9575, SVL =
62 mm. Both photographed on January 11, 2011.

series, the medial one basically heptagonal;
irregular-sized and -shaped occipitals, poste-
rior to but in contact with parietals, much
smaller than parietals, much larger than
dorsal granules. Scales in contact with outer
perimeter of parietal and interparietal scales
(PSC; Appendix 1) 15; back of head covered
with small granules, smaller than middorsal
ones on body; supraoculars 4, Ist and 4th
smallest, 2nd largest, separated from super-
ciliaries by 1 or 2 rows of granules (LSG, 13-
12; Appendix 1), except Ist supraocular
broadly contacts first 2 superciliaries. Last
supraocular separated from frontoparietals
by row of small scales (circumorbital semi-
circles, 3-5); postnasal 1 on each side; loreal |
on each side, large, somewhat rectangular;
preocular 1 on each side, with distinct ridge
(keel). A row of 3 suboculars, the 2 anterior
ones with a suborbital ridge continuing
anteriorly onto the preocular, the 2nd
subocular longest. Postoculars irregular and
varied in size and shape; superciliaries 6-6,

6 BREVIORA

the 3rd (left) or 2nd and 3rd (right) longest,
the anterior ones elongate, posterior ones
basically quadrangular. A few somewhat
enlarged supratemporals; somewhat enlarged
scales anterior to ear opening; central region
of temple with small, roundish granules. Ear
opening large, surrounded by small scales
forming a smooth edge; external auditory
meatus short, tympanum clearly visible:
large supralabials 6-6, followed by small
ones; suture between Sth and 6th below
center of eye; the 3rd, 4th, and Sth longest.
Lower eyelid with semitransparent disc of 5
enlarged palpebrals; pupil shape basically
round to somewhat oval horizontally, with
small irregularity on lower edge.

Mental trapezoid with convex anterior
edge: postmental basically pentagonal; 7
pairs of chinshields (= sublabials of some
authors) curving posteriorly and dorsally to
the lower labials, only those of anterior pair
in contact at midline: chinshields, from 2nd—
7th pair separated in part or completely from
infralabials by interlabial scales (7-7); 7
enlarged anterior infralabials, followed by
small scales: 3rd infralabial (left side) or 4th
(right side) largest.

Gulars small, flat, rounded, juxtaposed to
slightly imbricate, somewhat larger on the
anterior part of the throat, smaller posteri-
orly, from level of ear openings (GUL, 17:
Appendix 1). Mesoptychial scales (on ante-
rior edge of distinct gular fold) abruptly
enlarged, slightly imbricate, smooth, about
13 enlarged ones across throat. Scales on
nape and side of neck similar to dorsal and
lateral body scales but smaller.

Dorsal and lateral scales irregularly gran-
ular, in indistinct transverse and oblique
rows; ventrals large, usually somewhat
rhomboidal, usually wider than long, imbri-
cate, smooth, mostly in 8 longitudinal and 28
transverse rows (axilla to groin), the anterior
rows on chest interrupted by a _ small
triangular area of smaller scales. Number

No. 539

of dorsal granules around midbody 63;
preanal area with 4 clearly enlarged, smooth,
irregularly shaped, juxtaposed or slightly
imbricate scales plus smaller ones.

Femoral pores 18-19, usually each pore
surrounded by three small scales (medial one
largest); midventrally, 2 scales separate the
femoral pore series of each side.

Scales on dorsal and lateral aspects of
tail basically rectangular, obliquely keeled,
slightly mucronate, somewhat imbricate, in
transverse rows, keels forming longitudinal
ridges (TBS, 22; Appendix 1); scales under
tail wider, smooth, more imbricate; scales
on regenerated part of tail small, irregular,
keeled, in transverse rows; tail round in cross
section.

Scales on upper and anterior surfaces of
upper arm, on upper and anterior surfaces of
forearm, on anterior and lower surfaces of
thighs, and on lower surfaces of lower legs
large, smooth, imbricate. Scales on lower
and posterior surfaces of upper arm, on
posterior and ventral surfaces of forearm, on
posterior and upper surfaces of thighs, and
on upper, anterior, and posterior surfaces of
lower legs small, granular, juxtaposed (but
postantebrachials on forearms enlarged, an-
gular, and irregular in shape). Lamellae on
ventral surface of 4th finger 16-15. Lamellae
on ventral surface of 4th toe 33-32, usually in
single row on fingers but often paired on
toes: fingers and toes somewhat laterally
compressed; palms and soles with small,
irregular, juxtaposed, flat scales (some tu-
bercular); upper surfaces of hands and feet
with large, imbricate, smooth, flat scales.

Color and Pattern of Holotype in Life. On
January 11, 2011, with an adult SVL of
77 mm and age of about 29 months (Fig. 1,
upper), coloration was as follows: dorsum
reddish-brown with small pale cream to tan
light dots and somewhat larger spots; 6
extremely subtle gray to tan stripes as seen
from above, more evident if one looks over

2014

the lizard from low in the rear, as in aiming a
rifle; paravertebral light stripes wavy or
zigzag, other light stripes basically straight;
top of head brown; arms reddish-brown
with few dark brown to black markings; legs
similar to arms but with more _ black
markings; very few pale tan dots dorsally
on thighs, not as evident as in most A.
exsanguis; some light tan spots present
dorsally on rump, not as evident as in most
A. exsanguis; dorsal surface of tail mostly
reddish-brown on base, becoming greenish-
brown distally.

Chin, throat (up to lower edge of ear
opening), and chest pale blue; tan beneath
arms and on chest; very pale blue to gray on
abdomen, underside of legs, and underside of
base of tail; light tan with gray smudges on
underside of medial and distal tail.

Color and Pattern of Holotype in Preser-
vative (70% Ethanol). The following changes
occurred after preservation. The light dots
and spots on the body are pale gray or
cream. Only four largely faded gray light
stripes are barely evident, more visible
posteriorly than anteriorly. No light dots or
spots visible on thighs. Dorsal surface of tail
reddish-brown, becoming brown posteriorly,
then reddish-brown distally.

Ontogenetic Development of Color Pattern
in Life. The following notes on three typical
individuals of A. neavesi of different ages were
taken on January 11, 2011. The youngest,
MCZ R-192243 (= SIMR 10400; Fig. 2,
upper), was 16 days old at a SVL of 36 mm.
Dorsum brown (not reddish-brown as in
adults) with 6 bold, conspicuous light
stripes (yellow, but the lateral stripe tends
toward cream); paravertebral light stripes
wavy or zigzag, others basically straight;
dorsum essentially unspotted (except on
limbs), with few hints of tiny pale dots in
the dorsolateral and lateral dark fields; top
of head light brown; dorsal tail at base
similar to posterior body but becoming

NEAVES’ WHIPTAIL LIZARD 7

conspicuous bluish-green; arms dorsally
dark brown with yellow spots; legs similar
to arms but also with short yellow stripes;
ventral surfaces very light tan, but underside
of tail is bluish-green.

The next youngest, SIMR 9575 (Fig. 2,
lower), was noted at about 9 months old at a
SVL of 62 mm and with colors and pattern
intermediate between the youngest and oldest
individuals (as described for the holotype).
Dorsal ground color becoming reddish-
brown; light stripes becoming less conspicu-
ous, especially anteriorly, becoming gray and
significantly faded; lateral light stripe tannish-
cream; top of head darker brown; spots on
arms and legs now tan; distal tail now
brownish-green; ventral surfaces as on youn-
ger individuals, without blue.

The next oldest lizard, MCZ R-192218 (=
SIMR 8092) is a sister of the holotype (in the
same clutch) and was about 29 months old
when noted, with a SVL of 68 mm, about
9 mm shorter than the holotype at the time.
She was similar to the holotype except as
follows: dorsum and sides with considerably
more light spots (mostly tan); tail becoming
brownish-green posterior to base; chest,
abdomen, and ventral surfaces of legs and
tail light gray to pale blue.

The ontogenetic changes can be summa-
rized as follows: dorsal ground color changes
from brown to reddish-brown; dorsal light
stripes change from being conspicuous and
yellow or cream to become tan or gray and
extremely faded; upon hatching, dorsal body
is essentially unspotted, but later develops
light spots, which in largest adults become
less conspicuous; top of head changes from
light brown to brown; tail changes from
conspicuous bluish-green to greenish-brown:
arms and legs change from dark brown to
reddish-brown and tend to lose the light
spots and stripes; ventral surfaces change
from very light tan to gray, pale blue, or
darker blue.

8 BREVIORA

Karyotype. Lutes et al. (2011) described
and illustrated the karyotype. It consists of
the three haploid sets of chromosomes of
A. exsanguis (as they were cloned) plus a
haploid set of A. inornata. The karyotype of
A. exsanguis involved is slightly modified
from the ideal theoretical triploid condition.
The modifications are identical to those
found in the maternal triploid A. exsanguis
that were collected in Alamogordo, New
Mexico, in 2003-2005, and their laboratory
offspring, as follows: (A) one of the largest
macrochromosomes (a metacentric with a
subterminal secondary constriction on one
arm) had apparently undergone centric
fission into two smaller telocentric chromo-
somes: and (B) two microchromosomes were
apparently missing. These modifications were
not found in the original field-caught tetra-
ploid hybrid or in the A. exsanguis collected
with it (Neaves, 1971), which indicates that
two or more karyotypic clones of A. exsanguis
have been present in Alamogordo since 1967.
This is not particularly unusual for partheno-
genetic whiptails (Lowe et al., 1970b: Cole.
1979).

Reproduction. Individuals reproduce by
parthenogenetic cloning (Lutes ef al/., 2010,
2011).

Etymology. The specific epithet, a noun in
the genitive singular case, honors Dr. William
B. Neaves, who was awarded a Ph.D. at
Harvard University. Dr. Neaves’ graduate
studies on unisexual whiptail lizards (Neaves
and Gerald, 1968, 1969: Neaves, 1969, 1971)
provided important early insights into the
molecular genetics, origins, and speciation of
parthenogens through hybridization, as well
as the origin of a tetraploid hybrid lizard of A.
exsanguis X A. inornata that he discovered in
the field in Alamogordo, Otero County, New
Mexico, which was the inspiration for the
present laboratory hybridization project.

Comments. The specimen selected as the
holotype is an F, generation lizard that

No. 539

produced cloned offspring. The F, lizards
were true hybrids with two parents of
different species, but individuals of subse-
quent generations were not hybrids; they had
only one parent from which they were
cloned. It is a paradox of convention for
unisexual lizards of hybrid origin that the
animals of the F, generation and beyond are
considered to represent a named species, but
the F,; female hybrids of which they are a
clone are not considered to be members of a
species.

Specimens of A. neavesi examined for the
present report are members of three lineages
derived from two clutches of F; hybrid eggs
of A. exsanguis X A. inornata that were
produced at the SIMR from three F,; females
that cloned themselves (Lutes et a/., 2011).
These cloned lineages ultimately were a
product of one female’s ovaries and one
male’s testes. We treat the lineages as one
species (following Cole, 1990), rather than
three, although some authors have suggested
that theoretically each lineage that stems
from a hybrid zygote should be treated as a
separate species (Frost and Wright, 1988:
Frost and Hillis, 1990). There were genetic
differences in microsatellite loci detected
among the F, hybrids owing to fertilizations
by different spermatozoa from A. inornata
and a derived mutant allele of MS 14 that
was found in an F3 female (SIMR 9706:
Lutes et al., 2011). This kind of genetic
variation is found within gonochoristic
species as Well. Despite such genetic differ-
ences, individuals of the different lineages are
very similar to each other in morphology (see
below). Nevertheless, reproduction varied
widely within the three lineages discussed in
this paper. For example, at the time of this
writing the lineage of AMNH R-176077-
176148 (= SIMR 4921 lineage) had pro-
duced about 250 offspring, whereas the
lineage of MCZ R-192211 (= SIMR 5983
lineage) had produced only about 54.

2014

TABLE 1. DESCRIPTIVE STATISTICS FOR 10 MORPHOLOGICAL
CHARACTERS IN A POOLED SAMPLE OF THE F3, F3, AND F4
GENERATIONS OF THREE LINEAGES OF ASP/DOSCELIS NEAVESI.

Character? Mean + 1 SE (range) Sample Size (JN)
SPV 4.6 + 0.06 (3-6) 113
GAB 63.6 + 0.18 (58-69) 125
FP 38.0 + 0.14 (32-41) 124
SDL-T 31.8 + 0.10 (29-35) 123
SDL-F 15.0 + 0.07 (13-17) 122
GUL 16.8 + 0.13 (13-20) 113
COS 9.1 + 0.10 (6-12) 121
TBS 18.6 + 0.18 (15-24) 125
PSC 14.7 + 0.12 (11-18) 118
LSG 25.9 + 0.20 (21-35) 116

aSee Appendix 1 for detailed descriptions of charac-

ters.

Normally, new species are named not
because they are newly formed but because
they became recently known to science. In
this case, A. neavesi really is a new species; its
date of origin was August 12, 2008, when the
first F, individual hatched.

INTRASPECIFIC COMPARISONS

Scalation. Basic quantitative morphologi-
cal data for 125 specimens of A. neavesi are
provided by descriptive univariate statistics
for 10 meristic scalation characters (Appen-
dix 1; Table 1). The following additional
characters were examined also.

Condition of enlarged ventral preanal
scales: in Type I there are 3 enlarged scales,
2 bordering the vent and | anterior to these.
In Type II there are 2 enlarged scales, 1
bordering the vent and 1 anterior to it. In
Type III the pattern is different from both
Type I and I. Of 124 specimens examined,
40% had Type I, 8% had Type II, and 52%
had Type III.

Number of rows of enlarged ventral plates
across belly at midbody: there were 8 in all
124 specimens examined.

Number of ventral scales separating the
series of femoral pores: of 123 specimens

NEAVES’ WHIPTAIL LIZARD ?

examined, 87% had 2 scales, 12% had 3
scales, and 1% (1 individual) had 4 scales.

Condition of postantebrachial scales on the
forearm: all of the 124 specimens examined
had enlarged scales.

Condition of mesoptychial scales across
the throat: all 124 specimens examined had
abruptly enlarged scales, but 3 specimens
had some irregular small scales in the middle
of the series.

Statistical Comparisons of Two Lineages.
Here we address the question of whether there
are differences between the two best repre-
sented lineages, that of MCZ R-192209 (=
SIMR 4919 lineage) and AMNH R-176077—
176148 (= SIMR 4921 lineage). These two
lineages differed significantly in 2 of the 10
meristic characters (SPV and FP; Table 2),
but we included all 10 characters in a PCA
of the two lineages. There were no multi-
variate outliers among the 88 specimens
with complete data in these two lineages.
We used the 10 principal components
generated by the PCA as potential candi-
date variables for a CVA, of which PC1 and
PC7 were selected by the stepwise selection
criteria (see Materials and Methods) for
inclusion in the CVA model (Table 3).

The two lineages differed significantly
in CV1 (Table 2). However, this statistical
difference was counterbalanced by evidence
that possible biological significance is open
to question. For example, the Wilks’ lambda
value (0.840) from the CVA indicated that
only about 16% of the total variation was
explained by meristic differences between the
two lineages. Additionally, the small eigen-
value of 0.190 for CV1 indicated that the
discriminant function was weak (Table 3),
and only 69.3% of the specimens were
classified correctly to their respective lineages
by the CVA model (Table 4).

It was not surprising that the difference in
CV1 was statistically significant because
these two lineages differed statistically in

TABLE 2.

BREVIORA

SIMR Lineage>

Character@ 4921 (N = 56)
SPV 4.4 + 0.07 (3-5) 48
FP 38.0 + 0.22 (3240) 38.8
GAB 64.0 + 0.24 (58-68) 63.4
SDL-T 31.9 + 0.15 (29-35) 32.1
SDL-F 15.1 + 0.11 (13-17) [5.2
GUL 16.7 + 0.16 (13-20) 17.0
COS 9.0 + 0.14 (6-11) 9.4
TBS 18.6 + 0.26 (15-23) 19.2
PSC 14.7 + 0.18 (11-17) 14.7
LSG 25.9 + 0.26 (22-30) 26.4
CV1 —0.326 + 0.121 (—2.677 to 1.761) 0.570

No. 539

COMPARISON OF MEANS OF MORPHOLOGICAL CHARACTERS OF TWO LINEAGES OF ASPIDOSCELIS NEAVESI.

Is the Difference

Significant?
4919 (N = 32) Y/N tg6 P
+ 0.12 (3-6) Yes . =—2.468 0:02
+ 0.18 (36-41) Yes .-2479) O02
+ 0.41 (59-69) No 1.225. 2
+ 0.19 (30-35) No -—0.940 0.35
+ 0.12 (13-16) No —0.845 0.40
+ 0.23 (15-19) No -1.090 0.28
+ 0.21 (8-12) No -—1.498 0.14
+ 0.35 (16-24) No . —1.3384-"30si7
+ 0.25 (12-18) No -—0.073 0.94
+ 0.37 (23-33) No. 17491 - 30:28
+ 0.204 (—2.323 to 2.406) Yes —4.040 <0.001

«Mean + SE and range limits are shown for 10 univariate characters described in Appendix 1 and CV1. Only
specimens with complete data for all 10 univariate characters are included. All variances were homogeneous

(Levene’s Test probabilities: 0.05—0.93).

>Lineages are described in Appendix 2. SIMR 4921 lineage is AMNH R-176077-176148, and the lineage of SIMR

4919 is the lineage of MCZ R-192209.

two univariate characters (see above; Ta-
ble 2). But is there a possibility that evolu-
tionary divergence of the lineages 1s occurring
at a low level of resolution? There appears to
be a trend in a progressive separation of the
two lineages across generations, as seen in a
decrease in mean CV1 from the F; to the Fy,
generations in the AMNH lineage and an
increase In mean CV1 from the F; to the F,
generations in the MCZ lineage. Meristic
divergence, as measured by CV1, reached a
level of separation at the F3; generation of the
AMNH lineage and Fy, generation of the
MCZ lineage that became a statistically
significant difference in CV1 between the
two lineages (Table 5). There were too few
F, specimens with complete data in the MCZ
lineage to include in this lineage-generation
combination of comparisons. The possibility
that morphological divergence 1s taking place
between lineages in this newly formed parthe-
nogenetic species is worth monitoring, and
this hypothesis can be tested by including
more individuals and future generations in the
analysis.

INTERSPECIFIC COMPARISONS

Coloration. Here we compare colors and
patterns of A. neavesi with its ancestral
species, A. inornata and A. exsanguis. Lutes
et al. (2011) illustrated adults of the three
species in color (their fig. 1A:9911). Aspidos-
celis inornata is so different from the others
that it is easily identified before capture in
the field. Its dorsum is dark brown with 6—7
conspicuous yellow stripes, no spots, a bright
blue tail, blue on the ventral surfaces (darker
in males than females), and only minor
changes in color tones during ontogenetic
development. The strong difference between
A. inornata and the other two species is
interesting because A. inornata 1s one of the
three ancestral species of A. exsanguis
(having provided at least 33% of the overall
genome of A. exsanguis; Dessauer and Cole,
1989; Reeder et al., 2002); consequently, it
provided at least 50% of the overall genome
of A. neavesi. The strong similarities between
A. exsanguis and A. neavesi reflects matricli-
nous inheritance (i.e., having a _ greater

2014

TABLE 3. LOADINGS: CORRELATIONS BETWEEN
CHARACTERS AND EITHER PRINCIPAL COMPONENTS (PC) OR A
CANONICAL VARIATE (CV1) FROM MULTIVARIATE ANALYSES

OF TWO LINEAGES OF ASPIDOSCELIS NEAVESI.

Character PC] Pe7
Principal components SPV 0.651 0.086
analysis FP O:65il 0.1335

GAB 0.640 —0.533

COS 0.388 0.346

SDL-T 0.429 (0.254

PSC 0.266 —0.216

GUL 0.363 0.329

LSG 0.185 0.201

SDL-F 0.119 0.169

TBS 0.202 0.047
Eigenvalues 1.884 0.719
Proportion of variation (%) 18.8 12
Canonical variate analysis Character CVI

PCy? 0.756

PCI® 0.588

PC6 0.072

PC9 0.046

PC8 0.046

PC2 —0.038

PC3 =,056

PCIO: -..=0.032

PC5 —0.024

PCs 0.009
Eigenvalue 0.190
Proportion of intergroup 100

variation (%)

aUnivariate characters are described in Appendix |
and the lineages of AMNH R-176077-176148 (= SIMR
4921 lineage) and MCZ R-192209 (= SIMR 4919
lineage) in Appendix 2.

bOf the 10 principal components, only PCl and PC7
were selected as having discrimination value for the
CVA.

resemblance to the maternal ancestral spe-
cies) following the most recent episode of the
three hybridization events in the ancestry of
A. neavesi (see below).

Detailed notes on color and pattern
comparing A. neavesi and A. exsanguis were
made on preserved specimens. Young indi-
viduals of A. exsanguis (based on MSB 95084
and MSB 95086 [= SIMR 9982 and 9989,
respectively], with SVLs of 53 and 59 mm,

NEAVES’ WHIPTAIL LIZARD 1]

TABLE 4. CLASSIFICATION RESULTS OF A CANONICAL
VARIATE ANALYSIS OF TWO SIMR LINEAGES OF
ASPIDOSCELIS NEAVESI.

Lineage 4921, Lineage 4919,

N No. (%) No. (%)

Original grouped

specimensé

Lineage 4921 56 40 (71.4) 16 (28.6)

Lineage 4919 32 10 (31.3) 22 (68.8)
Jackknifed

classification

Lineage 4921 56 40 (71.4) 16 (28.6)

Lineage 4919 32 11 (34.4) 21 (65.6)

aA priori groups are in the first column, and the row
for each a priori group shows the number of individuals
assigned to each group by the CVA model (Table 3).
Overall classification success was 70.5% for the original
grouped specimens and 69.3% for the jackknifed
classification. Lineages are described in Appendix 2.
The SIMR 4921 lineage is AMNH R-176077—-176148
and the SIMR 4919 lineage is of MCZ R-192209.

respectively) differ from similar-sized A.
neavesi only in having the emerging light
spots become more random in distribution,
including appearing on the light stripes.
Large adults of A. exsanguis (e.g., as MCZ
R-192216 [= SIMR 7190], at 78 mm SVL)
differ from adults of A. neavesi by having the
following characteristics: dorsal light spots
larger and more conspicuous, especially
across rump; dorsal thighs with large,
abundant, conspicuous light spots (these
disappear in large A. neavesi); medial region
of tail dorsally greenish-brown (brown in
large A. neavesi); more cream, less blue on
ventral surfaces than A. neavesi (compare
Figs. | upper and lower).

Comparisons in preservative between A.
neavesi and two male tetraploid hybrids of A.
exsanguis X A. inornata collected in the field
follow. Natural hybrid MCZ R-101991 with
a SVL of 69 mm is similar in all respects to
A. neavesi of similar size and noted in life
(e.g., SIMR 9575 at 62 mm SVL), except that
on the natural hybrid, the paravertebral light
stripes converge posterior to the neck and

12 BREVIORA

TABLE 5.

No. 539

DESCRIPTIVE STATISTICS AND TESTS OF SIGNIFICANCE AMONG SIMR LINEAGE-GENERATION COMBINATIONS FOR

CANONICAL VARIATE | (CV1) IN ASPIDOSCELIS NEAVESI.

Descriptive Statistics

Generation@ 4921-F, 4921-F; 4921-F, 4919-F; 4919-F,
N 4 30 23 16 14
Mean + SE (range) =—054 2.0772" =036 = 0.181 —(0.20 =-0.14 0:52 0:36 0:62-=7023
(—2.41 to 1.36) (—2.:68 to 1.76) (—1.70to 0.85) (—2.32 to 2.41) ((—O7 Ite 8s)

Lineage> Probabilities (P) That CV1 Means Are the Same

4921-F, 1.0

4921-F;3 0.999 1.0

4921-F, 0.989 0.956 1.0

4919-F; 0.734 0.230 0.373 1.0

4919-F, 0.639 0.016 0.035 0.999 1.0

aSample size, mean + SE and range of variation of CV1 are shown.
>Lineages are described in Appendix 2. The SIMR 4921 lineage is AMNH R-176077-176148, and the SIMR 4919

lineage is of MCZ R-192209.

continue immediately adjacent to each other
as a gray vertebral area, then separate again
at the hip. In life, this natural hybrid also
had brilliant blue ventral surfaces (Neaves,
1971). Natural hybrid UCM 29196, also with
a SVL of 69 mm, was similar except that the
paravertebral stripes were like those of A.
neavesi, including some zigzag, contrary to
Taylor et al. (1967). Three large male hybrids
of A. exsanguis X A. inornata formed at
SIMR (MCZ R-192210 [= SIMR_ 5134];
MCZ R-192238 [= SIMR 9682]; and MCZ
R-192239 [= SIMR 9683]) were similar to
large A. neavesi, but two had larger and more
conspicuous dorsal spots, and they had
considerable blue ventrally.

In summary, male hybrids usually are very
similar to female hybrids, to A. neavesi, and
to A. exsanguis, especially when active
individuals are observed in the field. If
collected, the male hybrids stand out in
particular because of the deeper blue ventral
surfaces and because of their hemipenes
(absent in A. exsanguis). Alternatively, fe-
male hybrids or specimens of A. neavesi, if
collected, would be more easily misidentified
as A. exsanguis. Nevertheless, by comparing
specimens carefully with Figure 1, the color

notes, the karyotype, and the morphological
data presented here, one can consistently
distinguish between A. neavesi and A. ex-
sanguis.

Scalation and Multivariate Statistics. As
with the intraspecific comparisons (see
above), a PCA was used to develop uncor-
related characters (principal components)
from the raw meristic data, to be used in a
follow-up CVA (Tables 6, 7). There were no
multivariate outliers among the 110 speci-
mens with complete data in these three taxa.

Because a PCA reflects all of the intra-
group and intergroup variation in one
pooled sample (1.e., without classification of
individuals into a priori groups), we wanted
to see if the PCA would objectively cluster
the specimens into three separate groups
representing A. exsanguis, A. inornata, and
A. neavesi. The three species were clearly
separated by the first two principal compo-
nents (PC1 and PC2), which accounted for
approximately 44% of the total meristic
variation (Table 7; Fig. 3). However, A.
neavesi did not differ significantly from A.
exsanguis for PCl, and A. neavesi did not
differ significantly from A. inornata for PC2
(Table 6). Scores for PC1 and PC2 placed

NEAVES’ WHIPTAIL LIZARD 13

2014

‘SIOJORIVYS J}VLIBAIUN (QT [[B OJ RIeP dJa;dWIOd YIM sudUIdads astudu0d sajdwuirga
‘COO = © 1¥ WOdJOJJIP APURITTUSIS JOU JIB 19}]9] [eIIdeO owIPS dy} SuLIVYs sdnois Spasn o19M
S}so} Uostedtuos gdyjnu Aaynyp |] xtpusddy ur poqiiosap siajovsieyo ayeUeAIUN QT pure ‘TAD ‘LAO ‘ZOd ‘1Od JOJ UMOYS oe sy] odULI pure AS F uvdple

d (1S'€ 91 Or’ €—) L090 = OF'0 € (LZ7'€ 91 SO'T—) €60°0 + Hr'0 V (87 €— 91 O€ S—) 77710 + SED CAO
) 66 S— 91 Ip O1—) SLV'0 = €8'L= d (€L'7 9} 8¢°T—) IOT'0 + 98°0 V (v8'0 ©} 60°I—) €€7'0 + 900 [AO
€ (€L°7 91 pI'7T—) v0S'0 + 00 d (S8'I 91 66°I—) 790'0 + 97'0- V (p97 91 EPL) OIL0 = ITZ COd
q (98°0— ©} Ol’ P—) 9ZE'0 = EL'T— V (971 91 OL I—) 8700 = LZ7'0 V (L9'0 ©1 £0'0) 890°0 = €£°0 lOd
VY 6I-eD 690 = C9! V (0c-€1) p10 + 891 V SIS) 600 = CELI AS)
VY O1-CD 60 + PTI Gii-cl) 800 = Sl V (vI-€l) 910 + O'ET SETS
VV Ul) Wo 06 V (ZI-9) T1'0 + 16 V (1-2 v0. = 76 SOD
d (ZZ-S) pL'0 + 761 d (b7-SD 170 + L'8I V (S7-61) 95°0 ¥ LIZ Sa.
2 (L-@S) SLI = 6'09 d (69-85) 170 + L'€9 V (I1L-89) Sv'0 + L'69 qv)
d (6€-LI) LST = EZ qd (co-co) 170 = 1 9C V (ve-87) 89°0 + OTE Sal
16-57) 190 = S'8c A (So-60) C10 = OE V UC 80).Sc.0 = 86 EIS
d (St +D) LET + LSI V (SI-ID p10 + Lvl V WIV 1e0:= si Sd
aH (Olo) Cr OSC 8 V (9-€) 90'0 + O'P Vcr 9l0 + 7P AdS$S
dq OF 7) 970 = OSI V UCC a0 76él VY OG6) LI0 = Col dH
(Ol = N) DIpusoUl “PY (06 = N) Sadvau “py (Ol = N) SINBUDSXA ‘PY elOJOVIVYD
qdnoip

‘VLVNUYONI ‘P UNV SINONVSXT “P *SAIOAdS UOLINADOUd SLI GNV ISTAVAN SITADSOdIdS f AO SAALOVAVHO TVOIDOTOHdYOW ONOWYV SNVAW JO NOSIYVdWO)) °9 ATEV

14 BREVIORA

TABLE 7.

No. 539

LOADINGS: CORRELATIONS BETWEEN CHARACTERS AND EITHER PRINCIPAL COMPONENTS (PC) OR CANONICAL

VARIATES (CV) FROM MULTIVARIATE ANALYSES OF ASPIDOSCELIS NEAVESI AND ITS PROGENITOR SPECIES, A. EXYSANGUIS AND
A. INORNATA.

Character>

Principal components analysis@ FP

SPV
PSG
SDL-T
LSG
GAB
TBS
COS
SDL-F
GUL

Eigenvalue
Proportion of variation (%)

Character
PCI
PC6
PC2
PC3
PC4
PCF
PC9
PCS
PC10

Canonical variate analysis

Eigenvalue
Proportion of intergroup variation (%)

PCI

0.782
=O5351
—0.726

0.249
0.458
0.022
—0.060
0.207
0.358

2.620
26.2

CV1
0.672
0.060
—0.046

0.099
—0.049
=0.022
—0.016

0.027
—0.056

6.359
76.5

Fe? PCS PC4 PES PC6
0.076 O.F71 -—0.099 0.075 0.309
0.082. =0:195 0.349 0.020 —0.154
0.263 0.116 0.191 0.093 0.280
0.700 —0.245 0.033 0.183 0.154 0.109
0.707 0.302 0.033 0.069 —0.462
0.670 —0.270 0.095 -—0.041 -—0.202
0.591 —0.404 0.125 0.490 0.361
0.401 0.686 0.293 —0.340 0.279
—0.447 0.245 0.634 0.434 —0.185
—0.044. —0.472 0.494 —0.568 0.072
1.801 1.164 0.960 0.911 0.712
18.0 11.6 9.6 9.1 TA
CV2
—0.134
0.043
—0.686
0.176
0.165
0.152
—0.102
0.098
0.065
1.948
23,3

4Loadings are not shown for PC7, PC9, and PC10 that summarized, respectively, only 6.2%, 3.9%, and 3.3% of the
meristic variation; PC8 was not included in the CVA model.

>Univariate characters are described in Appendix 1.

the laboratory hybrid male (MCZ R-192239
[= SIMR 9683]) well inside the cluster of A.
neavesi. In contrast, the field-caught pre-
sumptive hybrid male (UCM 19196) was
outside the clusters of the three species but
closest to the cluster of A. exsanguis (Fig. 3).

Next we used our samples of A. exsanguis,
A. inornata, and A. neavesi as a priori
(preidentified) groups for a CVA, and the
hybrid males (MCZ R-192239 and UCM
29196) were included as unclassified individ-
uals for assignment to the most similar a
prior! group. Our objectives were (1) to
determine how well the three species could

be discriminated with meristic characters, (2)
to quantify the morphological resemblance
of A. neavesi to each of its progenitor species,
and (3) to determine the species alliance in
multivariate space of the tetraploid hybrid
males of A. exsanguis X A. inornata.

Our samples of 10 A. exsanguis and 10 A.
inornata are from the same locality as the
female hybrid (MCZ 101991) that was
collected in 1967 in Alamogordo, New
Mexico, and reported by Neaves (1971). Of
129 individuals in the four lineages of A.
neavesi, 39 were excluded from statistical
analyses because they could not be scored for

2014

Samples

O A. exsanguis
-4 O A. inornata
© A. neavesi :
%* Field-caught hybrid male
@ Lab hybrid male (centered-near zero)

PC1 (26.2%)

Figure 3. Scatterplot of principal component scores
of 90 specimens of A. neavesi and 10 each of its
progenitor species, A. exsanguis and A. inornata. Sample
centroids are centered on the 50% confidence ellipses of
group scores, and axis percentages are percentages of
variance explained by PC] and PC2 (Table 7). The field-
caught hybrid male (UCM 29196) is clearly visible, but
the laboratory hybrid male (MCZ R-192239 [= SIMR
9683]) is buried within the cluster of A. neavesi.

one or more meristic characters, mostly
owing to damage incurred when specimens
were dissected. For the 10 univariate meristic
characters, A. neavesi resembled A. exsanguis
in FP, SPV, and PSC; resembled A. inornata
in LSG and TBS; and was significantly

NEAVES’ WHIPTAIL LIZARD 15

different from both progenitor species in
SDL-T, GAB, and SDL-F (Table 6).

We included all 10 univariate characters in
a PCA, and 9 of the 10 principal components
were selected for inclusion in the CVA model
(Table 7). Wilks’ lambda (0.046) from the
CVA indicated that about 95% of the total
variation was explained by meristic differenc-
es among the three species. Large eigenvalues
(Table 7) and an overall classification success
of 99.1% (Table 8) were evidence of a strong
discrimination. There was only one misclassi-
fication (Table 8)—specimen AMNH R-
176081 (= SIMR 10310) from the F, gener-
ation of the AMNH lineage of A. neavesi was
misclassified to the A. exsanguis group based
on a marginal probability of 0.51. This
assignment was automatically made by the
CVA model because the probability for
assigning this specimen to the A. neavesi
group (0.49) was smaller. The assignment
of the unclassified laboratory hybrid male
(MCZ R-192239 [= SIMR 9683]) to the A.
neavesi group was robust (P = 0.999). In
contrast, the unclassified field-caught hybrid
male (UCM 29196) was assigned to the A.
exsanguis group with a high level of proba-
bility (P = 0.989).

The spatial relationships of the three
species in the CVA are shown in Figure 4.
The positions of the three clusters on the

TABLE 8. CLASSIFICATION RESULTS OF A CANONICAL VARIATE ANALYSIS OF ASPIDOSCELIS NEAVESI AND ITS PROGENITOR
SPECIES, A. EYSANGUIS AND A. INORNATA.
Original Grouped Specimens# N No. A. neavesi (%) No. A. exsanguis (%) No. A. inornata (%)
A. neavesi 90 89 (98.9) i @ ee 0
A. exsanguis 10 0 10 (100) 0
A. inornata 10 0 0 10 (100)
Laboratory hybrid male® ] ] --- —

Field-caught hybrid males ] —

1 =

aA priori groups are in the first column, and the row for each a priori group shows the number of individuals
assigned to each group by the CVA model (Table 7). Overall classification success was 99.1% for the three a priori

groups. A jackknifed classification gave the same results.

>This individual (MCZ R-192239 [= SIMR 9683]) was included in the CVA as unclassified.
¢This individual (UCM 29196) was included in the CVA as unclassified.

16 BREVIORA No. 539

—
=
>
_~
@ ~
NN 2
~"
N
=
—
: ers)
— STOUpS
2 syesens
oO A. exsanguis
O A inomate
- =
<4 © A. neaves
% Fieid-caught hybrid male
@ 22> -ydric male
:
&
;
< -$ -< ~ 4
ni «gn de
CV1 (76.5%

papa aneet of aoa variate scores
fl Og

exs ty A. inornata. The CVA m S
shown in Table oF | a centroids are aa on ae
50% confidence ellipses of group scores, and axis
percentages are contributions of canonical vaniates
CV1 and CV2 to the disc

rmination (Table 7). The
laboratory hybrid male is MCZ R-192239 (= SIMR
9683). and the field-caught hybrid male is UCM 29196.

CV1 axis (which expresses the greater pro-
portion of variation) indicate that A. neavesi
is more similar to A. exsanguis than A.
inornata in multivariate space. This was
confirmed by pairwise Mahalanobis D?
distances (smaller distances = greater resem-
blance) of 23.6 between A. neavesi and A.
exsanguis. 75.6 between A. neavesi and A.
inornata, and 84.9 between A. exsanguis and
A. inornata. Therefore. A. neavesi is matri-
clinous in resembling the maternal ancestral
species. All three species differed significant-
ly in mean CV1 (Table 6). but A. neavesi
clearly resembled A. inornata on the CV2
axis, as substantiated by the absence of a
significant difference in CV2 between these
two groups (Table 6). The hybrid males are
peripheral to the respective groups to which
they were assigned (Fig. 4). but their classi-
fication probabilities near unity leave little

doubt that these assignments based on
meristic characters are robust.
ACKNOWLEDGMENTS

We thank Rick Kupronis and his team of
dedicated reptile technicians at SIMR, David
Jewell, Alex Muensch, Christina Piraquive.
and Kristy Winter. for excellent husbandry
and herpetocultural skills. We also thank
Jillian Kupronis and Aracely Lutes for
contributions to genotyping analysis. The
following individuals at museum repositories
provided efficient curating of the specimens:
Margaret Amold. David Dickey. Darrel
Frost, David Kizirian, and Christopher Rax-
worthy (AMNH): Jonathan Losos, Joe Mar-
tinez, and José Rosado (MCZ): and Tom
Giermakowski and Howard Snell (MSB).

We are also grateful to Jonathan Losos and
Jose Rosado of the MCZ and to Mariko
Kageyama and Christy McCain of the UCM
for lending us specimens that were needed for
this research. Additionally. Charles Painter,
New Mexico Department of Game and Fish,
helped with permit matters. and Carol Town-
send (AMNH) assisted in many ways.

Most of all. we are deeply indebted to
William B. Neaves for spearheading the
research on parthenogenetic lizards at the
SIMR and for guidance, encouragement,
and assistance in our work. This work was
funded by the Stowers Institute for Medical
Research and the Howard Hughes Medical
Institute.

APPENDIX |

Abbreviations for morphological charac-
ters used are as follows: COS, number of
circumorbital semicircle scales (total of both
sides of head: following Wright and Lowe.
1967): GAB, number of dorsal granules
(scales) around midbody,. following Wright
and Lowe (1967): FP. number of femoral
pores (for intraspecific comparisons we used

2014

the total of both legs summed, but for
interspecific Comparisons we increased the
sample size by using the right leg only, but
the: lett. the; right.-was..not. clear); (GUL,
number of gular scales, following Cole et al.
(1988); LSG, number of lateral supraocular
granules (total of both sides of head, whether
in 1 or 2 rows) between the supraoculars and
superciliaries, counting forward from an
imaginary line extended from the suture
between the 3rd and 4th supraoculars toward
the superciliaries, following Walker ef al.
(1966); PSC, total number of scales in
contact with outer perimeter of parietal and
interparietal scales, following Cole ef al.
(2010); SDL-F, number of subdigital lamel-
lae on the right 4th finger but using the left if
the right is missing, following Taylor et ai.
(2001); SDL-T, number of subdigital lamel-
lae on the right 4th toe but using the left if
the right is missing, following Cole ef ail.
(1988); SPV, number of granules (scales)
between the paravertebral light stripes, fol-
lowing Wright and Lowe (1967); SVL, snout-
vent (body) length in millimeters; TBS,
number of enlarged dorsal scales around
dorsal aspect of base of tail; the count is
made while holding the hind legs at the hip
perpendicular to the body and counting on
an imaginary line along the posterior edges
of the legs, but not including lateral granules
on the tail.

APPENDIX 2
Specimens Examined

Aspidoscelis exsanguis. MCZ R-100419;
MSB 95084 (= SIMR 9982); MSB 95086
(= SIMR 9989); MCZ R-192263 (SIMR
11575); MCZ R-192287-192288 (SIMR
13218-13219); and MCZ R-192289-192292
(SIMR 13273-13276). In addition, one spec-
imen illustrated (Fig. 1, lower, SIMR 13209)
has not reached a permanent repository, as it
is still alive. All specimens were either caught

NEAVES’ WHIPTAIL LIZARD 17

in the city park, Alamogordo, Otero County,
New Mexico, or reared at SIMR from stock
obtained at that locality.

Aspidoscelis inornata: MCZ_ R-100425;
MCZ R-192214 (SIMR 6636); MCZ R-
192215 (SIMR 6755); MCZ R-192221 (SIMR
8453); MCZ R-192222 (SIMR 8678); MCZ
R-192223 (SIMR_ 8680); MCZ R-192228
(SIMR 8860); MCZ R-192230 (SIMR 9098);
MCZ R-192232 (SIMR 9427); and MCZ
R-192264 (SIMR 11678). All specimens were
either caught in the city park, Alamogordo,
Otero County, New Mexico, or reared at
SIMR from stock obtained at that locality.

Aspidoscelis neavesi: All of these specimens
are paratypes except for SIMR 9575 and
SIMR 9706 (see below). For economy of
space, we do not list the correlated SIMR
catalog numbers for these specimens, which
are included on an individual basis in the
catalogs at the AMNH and MCZ. The two
lineages of which specimens were compared
by PCA and CVA stem from F, hybrid
females SIMR 4921 and MCZ R-192209,
although those used in the PCA and CVA are
only the ones with all data available for the
complete suite of 10 characters, and the F,
hybrid females were not included. Specimens
from the AMNH R-176077-176148 series (=
SIMR 4921 lineage) are the following: F>
generation, AMNH_ R-176077-176085; F;
generation, AMNH R-176086—176120; and
F, generation, AMNH_ R-176121-176148.
Specimens from the MCZ R-192209 lineage
are the following: F, generation, MCZ
R-192218-192219, MCZ R-192224—-192227,
MCZ R-192229, MCZ R-192236-192237,
MCZ R-192243, and MCZ R-192256; F3
generation, MCZ R-192233-192234, MCZ
R-192241-192242, MCZ R-192244—-192246,
MCZ R-192248-192255, MCZ R-192258-
192260, MCZ R-192262, MCZ R-192265,
MCZ_ R-192267-192268, MCZ R-192276,
and MCZ R-192279; F4 generation, MCZ
R-192261, MCZ R-192266, MCZ R-192269-

18 BREVIORA

192275, MCZ R-192277-192278, and MCZ
R-192280—-192286. Individuals from another
lineage, the MCZ R-192211 (= SIMR 5983)
lineage, excluding the F, female are MCZ R-
192229, MCZ_ R-192236-192237, MCZ R-
192254, and MCZ R-192256. In addition, two
specimens noted for a few characters have not
reached a permanent repository: SIMR 9575
(an F, individual of the MCZ R-192209
lineage, used for color notes, is still alive)
and SIMR 9706 (an F3 individual of the MCZ
R-192209 lineage, noted for a microsatellite
mutation [Lutes et a/., 2011] is still alive). The
last two individuals are not paratypes. The
individual ancestors of these lineages, A.
exsanguis (SIMR 71) and A. inornata (SIMR
69), were both reared at SIMR from stock
obtained in the city park, Alamogordo, Otero
County, New Mexico.

F, laboratory hybrids of A. exsanguis X A.
inornata. MCZ R-192209-192211 and MCZ
R-192238-192239. These individuals were
reared at SIMR from parental stock that
was caught in the city park, Alamogordo,
Otero County, New Mexico.

F, hybrids of A. exsanguis X A. inornata
that were found in nature: MCZ R-101991, a
female, from the city park, Alamogordo,
Otero County, New Mexico, and (presumed
hybrid) UCM 29196, a male, from 2 mi W,
1 mi S Mesilla, Dona Ana County, New
Mexico.

LITERATURE CITED

Co.e, C. J. 1979. Chromosome inheritance in parthe-
nogenetic lizards and evolution of allopolyploidy in
reptiles. Journal of Heredity 70: 95—102.

Cote, C. J. 1990. When is an individual not a species?
Herpetologica 46: 104-108.

CoLe, C. J., C. H. Lowe, AND J. W. WriGuT. 1969. Sex
chromosomes in teid whiptail lizards (genus
Cnemidophorus). American Museum Novitates 2395:
1-14.

CoLE, C. J., H. C. DESSAUER, AND G. F. BARROWCLOUGH.
1988. Hybrid origin of a unisexual species of
whiptail lizard, Cnemidophorus neomexicanus, in

No. 539

western North America: new evidence and a review.
American Museum Novitates 2905: 1-38.

Core, C.. J.,. 1. M. HArpy, “Ho © DEssaAviers tee
TAYLOR, AND C. R. TOWNSEND. 2010. Laboratory
hybridization among North American whiptail
lizards, including Aspidoscelis inornata arizonae X
A. tigris marmorata (Squamata: Teiidae), ancestors
of unisexual clones in nature. American Museum
Novitates 3698: 1-43.

DaAwLey, R. M., AND J. P. BoGart (eds.). 1989. Evolution
and ecology of unisexual vertebrates. New York
State Museum Bulletin 466: 1-302.

DESSAUER, H. C., AND C. J. Coie. 1989. Diversity
between and within nominal forms of unisexual
teiid lizards, pp. 49-71. In Dawley, R. M., and J. P.
Bogart, eds. Evolution and Ecology of Unisexual
Vertebrates. New York, New York State Museum
Bulletin 466.

Fritts, T. H. 1969. The systematics of the parthenoge-
netic lizards of the Cnemidophorus cozumela com-
plex. Copeia 1969: 519-535.

Frost, D. R., AND D. M. Hits. 1990. Species in
concept and practices: herpetological applications.
Herpetologica 46: 87-104.

Frost, D. R., AND J. W. WriGutT. 1988. The taxonomy
of uniparental species, with special reference to
parthenogenetic Cnemidophorus (Squamata: Teii-
dae). Systematic Zoology 37: 200-209.

ICZN. 1999. International Code of Zoological Nomen-
clature. Fourth Edition London, The International
Trust for Zoological Nomenclature.

JAMES, E. 1823. Account of an Expedition from Pittsburgh
to the Rocky Mountains, Performed in the Years
1819 and '20. Volume 2. Philadelphia, H. C. Carey
and I. Lea.

JoMBART, T., S. DEVILLARD, AND F. BALLoux. 2010.
Discriminant analysis of principal components: a
new method for the analysis of genetically struc-
tured populations. BMC Genetics 11: 94.

Lowe, C. H., J. W. WriGuT, C. J. COLE, AND R. L. BEzy.
1970a. Natural hybridization between the tetid
lizards Cnemidophorus sonorae (parthenogenetic)
and Cnemidophorus tigris (bisexual). Systematic
Zoology 19: 114-127.

Lowe, C. H., J. W. WriGcut, C. J. CoLe, AND R. L. BEzy.
1970b. Chromosomes and evolution of the species
groups of Cnemidophorus (Reptilia: Tetidae). Sys-
tematic Zoology 19: 128-141.

Lutes, A., W. B. NeaAves, D. P. BAUMANN, W.
WIEGRAEBE, AND P. BAUMANN. 2010. Sister chromo-
some pairing maintains heterozygosity in partheno-
genetic lizards. Nature 464: 283-286.

Lutes, A., D. P. BAUMANN, W. B. NEAVES, AND P.
BAUMANN. 2011. Laboratory synthesis of an inde-

2014

pendently reproducing vertebrate species. Proceed-
ings of the National Academy of Sciences 108:
9910-9915.

NEAveEs, W. B. 1969. Adenosine deaminase phenotypes
among sexual and parthenogenetic lizards in the
genus Cnemidophorus (Teiidae). Journal of Experi-
mental Zoology 171: 175-183.

NEAvEsS, W. B. 1971. Tetraploidy in a hybrid lizard of the
genus Cnemidophorus (Teidae). Breviora 381: 1—25.

NEAVES, W. B., AND P. BAUMANN. 2011. Unisexual
reproduction among vertebrates. Trends in Genetics
27: 81-88.

NEAvEs, W. B., AND P. S. GERALD. 1968. Lactate
dehydrogenase isozymes in parthenogenetic teiid
lizards (Cnemidophorus). Science 160: 1004-1005.

NEAvVES, W. B., AND P. S. GERALD. 1969. Gene dosage at
the lactate dehydrogenase b locus in triploid and
diploid teiid lizards. Science 164: 557-559.

Nerr, N. A., AND L. F. Marcus. 1980. A Survey of
Multivariate Methods for Systematics. New York,
American Museum of Natural History.

REEDER, T. W., C. J. CoLE, AND H. C. DEssAuER. 2002.
Phylogenetic relationships of whiptail lizards of the
genus Cnemidophorus (Squamata: Teiidae): a test of
monophyly, reevaluation of karyotypic evolution,
and review of hybrid origins. American Museum
Novitates 3365: 1-61.

SmitH, H. M. 1946. Handbook of Lizards. Ithaca, New
York, Comstock Publishing Co.

NEAVES’ WHIPTAIL LIZARD 19

TABACHNICK, B. G., AND L. S. FIDELL. 2013. Using
Multivariate Statistics. 6th ed. Upper Saddle River,
New Jersey, Pearson Education Inc.

TAyLor, H. L., J. M. WALKER, AND P. A. Mepica. 1967.
Males of three normally parthenogenetic species of
telid lizards (genus Cnemidophorus). Copeia 1967:
737-743.

TAYior, H. L., C; J. Core, L. M. Harpy, H.C. DESSAUER,
C. R. Townsenp, J. M. WALKER, AND J. E.
Corpes. 2001. Natural hybridization between the
teiid lizards Cnemidophorus tesselatus (parthenoge-
netic) and C. ftigris marmoratus (bisexual): assess-
ment of evolutionary alternatives. American Museum
Novitates 3345: 1-64.

Tay or, H. L., J. M. WALKER, J. E. CORDES, AND G. J.
MANNING. 2005. Application of the evolutionary
species concept to parthenogenetic entities: com-
parison of postformational divergence in two clones
of Aspidoscelis tesselata and between Aspidoscelis
cozumela and = Aspidoscelis maslini (Squamata:
Teiidae). Journal of Herpetology 39: 266-277.

WALKER, J. M., H. L. TAYLor, AND T. P. MASLIN. 1966.
Evidence for specific recognition of the San Esteban
Whiptail lizard (Cnemidophorus estebanensis). Co-
peia 1966: 498—SOS.

WriGut, J. W., AND C. H. Lowe. 1967. Hybridization in
nature between parthenogenetic and bisexual spe-
cies of whiptail lizards (genus Cnemidophorus).
American Museum Novitates 2286: 1—36.

77