BREVIORA

Museum of Comparative Zoology

Mths etesess,,,
— A) 111 el =

US ISSN 0006-9698

CAMBRIDGE, MASss. 10 JuLty 2013 NUMBER 534

SIZE-ASSORTATIVE PAIRING AND SOCIAL MONOGAMY IN A NEOTROPICAL
LIZARD, ANOLIS LIMIFRONS (SQUAMATA: POLYCHROTIDAE)

ALEXIS HARRISON

ApstrActT. Social monogamy, the formation of stable male-female pairs, is uncommon among reptiles and is
particularly rare among squamates, in which only a handful of cases has been reported. Only one case of persistent
pair formation has ever been reported 1n anoles, for Anolis limifrons, at a single site in Costa Rica. Detailed studies of
A. limifrons at other sites, however, have not shown evidence of pair formation. I revisited the site where pairing was
originally reported to observe pair behavior in this species and to measure morphological traits of paired and
unpaired animals. I confirmed that male-female pairs are commonly encountered in the wild, although a smaller
proportion of the animals observed in this study were found in pairs than previously reported. I also found evidence
for size-assortative pairing; larger males tended to be found with larger females and smaller males were found with
smaller females. I did not find any differences in the morphology of paired and unpaired animals. Although social
monogamy has not been widely reported in squamates, I suggest that more examples of this phenomenon will be
described as the social behaviors of poorly known species are increasingly subject to study.

Key worps: Anolis limifrons; mating system; assortative mating; pair formation; monogamy

Social monogamy, the persistent associa-
tion between an adult male and an adult
female during the breeding season, is rela-
tively common among birds and mammals
(Wittenberger and Tilson, 1980), although
genetic or mating monogamy is more rare
(Petrie and Kempenaers, 1998). Among
reptiles however, social monogamy and pair
formation are very rare (Uller and Olsson,

Museum of Comparative Zoology—Department of
Herpetology, Harvard University, 26 Oxford Street, Cam-
bridge, Massachusetts 02138, U.S.A.; e-mail: asharris@
fas.harvard.edu

2008), and genetic fidelity is almost un-
known. Squamates in particular are usually
both socially and genetically promiscuous
(Bull, 2000; Uller and Olsson, 2008).

There are a few notable exceptions to this
generalization (Bull, 2000). The best known
case is that of the sleepy lizard, Tiliqua
rugosa, a long-lived skink native to southern
and central Australia (Bull, 2000). This
species is active through spring and early
summer, when it forages on vegetation, eggs,
nestlings, and carrion across a broad home
range that can overlap with the home range

© The President and Fellows of Harvard College 2013.

2 BREVIORA

of other individuals of both sexes. Individu-
als form stable pairs that appear to be
socially and genetically monogamous, both
within a single breeding season and across
multiple breeding seasons (Bull, 2000).
Paired animals are often observed in close
proximity to each other, and when they are
experimentally separated, both males and
females actively work to relocate their
partners using chemosensory cues (Bull ef
al., 1993). Pairs of animals can persist over
several breeding seasons; one pair was
observed together over 10 consecutive years
(Bull, 1994; Bull et al, 1998). However, even
in this species not all animals are observed
exclusively in pairs: 40% of females and 18%
of males were found with more than one
partner in a season during radio-tracking
surveys (Bull et a/., 1998). It appears that
males that pursue a polygynous strategy do
not benefit: females paired with polygamous
males were more likely to have multiple-
paternity litters, as revealed by microsatellite
paternity analyses (Bull et al, 1998). It is
unclear whether pair formation is equally
beneficial to females and, if so, how.
Australian skinks in the genus Egernia
have also been shown to form socially
monogamous pairs in nature (E. stokesii:
Gardner et al., 2002; E. cunninghami: Stow
and Sunnucks, 2004; E. whitii: Chapple and
Keogh, 2005). In Egernia saxatilis, not only
do males and females form monogamous
pairs, they live in close proximity to their
subadult offspring in a situation that mirrors
the “‘nuclear family” found in other verte-
brates (O'Connor and Shine, 2003). The
Tasmanian snow skink, Niveoscincus micro-
lepidotus, also forms pairs that persist, on
average, for 29 days during the breeding
season (Olsson and Shine, 1998). Aside from
skinks, at least two species of Chameleon,
Chamaeleo hoehnelii and C. jacksoni, have
also been observed in stable pairs in the field
(Toxopeus et al., 1988). In C. hoehnelii, pairs

No. 534

persisted for an average of 85 days, and 30-
40% of all animals were observed in pairs. In
C. jacksoni, pairs persisted for an average of
63 days. About half of females were observed
in pairs, whereas about a third of males were
paired.

Despite these reports of pairing behavior
in a handful of species, social monogamy is
thought to be extremely rare in squamates.
Numerous studies of a wide variety of species
support the notion that most squamates are
polygamous (reviewed in Stamps, 1983).

Why Monogamy?

Little is known about why some species of
lizards associate in pairs while most do not
(Bull, 2000). Three non—mutually exclusive
hypotheses may account for social monoga-
my across animal taxa: (1) males remain with
females to provide parental care (e.g., food
or protection); (2) males guard females
because the potential benefit of matings with
other females is outweighed by the loss of
paternity if other males mate with the focal
female; and (3) an individual may remain in
the presence of a mate because it benefits
directly from their presence (e.g., male may
fend off harassment by other males, or both
partners may improve the chance of spotting
predators; Bull, 2000). Parental care is
typically rudimentary or absent in squamates
(Gans, 1996); therefore, parental care is
unlikely to explain most examples of social
monogamy in lizards. A more likely expla-
nation 1s that either the males, the females, or
both directly benefit from pairing with a
single partner. One plausible scenario is that
monogamy evolves as a consequence of mate
guarding in species where it is difficult for
males to successfully defend multiple fe-
males, such as when females occur at low
densities and occupy a broad home range
(Emlen and Oring, 1977; Bull, 2000). Alter-
natively, social monogamy may evolve when

2013

the direct benefits of pairing are especially
high, such as when predation risk is substan-
tially reduced by an extra set of eyes.

When monogamy does evolve, the process
by which individuals form pairs becomes
highly important because an_ individual’s
fitness may be closely tied to the quality of
its mate. Each individual should therefore
strive to pair with the highest quality mates.
A common pattern in animals is for the
largest males to pair with the largest females
and smaller males to pair with smaller
females, a pattern known as size-assortative
pairing (SAP; Crespi, 1989). SAP can be a
product of three processes: mutual mate
emorce® for large “body ‘size*(e.g:, beetles:
Harari et al, 1999; spiders: Masumoto,
1999); physical constraints (e.g., beetles:
Brown, 1993; fish: Bisazza, 1997); or mate
availability—when, for some reason, indi-
viduals that are similar in size are more likely
to encounter each other and pair by chance
(e.g., limpets: Pal et al., 2006). One way that
mate availability could lead to a pattern of
size assortative mating would be that, on
reaching sexual maturity, an animal pairs
with the first unpaired, sexually receptive
animal they encounter. Older and larger
animals are likely to be paired already and
smaller animals are not yet sexually mature,
so they are most likely to pair with an animal
of similar size.

It may be possible to differentiate among
these processes based on differences between
paired and unpaired animals. For example, if
one sex is more abundant than the other,
paired animals should be larger than un-
paired animals in the more abundant sex if
mutual mate choice for body size is operat-
ing. In contrast, if physical constraints are
responsible for SAP, then the size of
unpaired animals should be related to the
size distribution of potential mates—large
animals may remain unpaired if large mates
are rare. Studying SAP can therefore provide

LIZARD ASSORTATIVE MATING 3

insight onto the process of pair formation. In
addition, when SAP occurs, it can also have
important implications for social behavior,
population genetics, and even, potentially,
speciation if assortative mating produces a
division in the gene pool (Crespi, 1989;
Kawecki, 1997; Nagel and Schluter, 1998:
Harari et al, 1999: Bessa-Gomes et al.,
2003).

Pairing Behavior in A. limifrons

A surprising candidate for pair formation
and SAP in lizards is a Costa Rican
population of Anolis limifrons, a slender
arboreal lizard that is abundant in a variety
of habitats from southern Mexico to Panama
(Savage, 2002). This is very unusual behavior
for an anole, a genus in which males typically
gain access to multiple females by defending
a territory and excluding other males (Tri-
vers, 1976; Andrews, 1985; Jenssen and
Nunez, 1998; Losos, 2009).

Pair formation in A. limifrons was first
reported by Talbot (1979), who found that
70% of adults were found in male-female
pairs; that is, a single male and a single
female were found within 2 m of each other
with no other lizards present. Mark-recap-
ture data showed that these pairs persisted
for 4-6 months, approximately the adult
lifespan for this species. Detailed observa-
tions of pair behavior revealed that individ-
uals in these pairs display to each other
frequently (Fig. 1) and move in tandem for
distances up to 20 m (Talbot, 1979).

Stable pairs have not been observed in
other populations of A. Jlimifrons despite
extensive study; in fact, other populations
demonstrate typical resource-defense polyg-
yny (Andrews and Rand, 1983; Andrews and
Stamps, 1994). Pair formation has also never
been studied in other anoles, though it has
been suggested for two species beside A.
limifrons on the basis of the proximity of

4 BREVIORA

A male A. limifrons displays to a female

Figure 1.
on an adjacent perch. These animals were observed in
close proximity and were interacting. No other lizards
were observed in the area, suggesting that these animals
form a pair.

sleeping males and females (A. occultus:
Gorman, 1980; A. cuvieri: Rios-Lopez and
Puente-Colon, 2007). This study had three
objectives: to observe pairs of A. limifrons at
the site where Talbot collected his data, to
determine if such pairs demonstrate size-
assortative pairing, and to compare the
morphology of paired and unpaired animals.

MATERIALS AND METHODS

Field Data

Field data were collected during a 17-day
period from 19 April to 5 May 2007 at La
Selva Biological Reserve in the Heredia
province of Costa Rica between 7:30am
and 5:30pm. Individual A. /imifrons were
spotted by walking slowly along established
trails while visually surveying vegetation, a
standard method for conducting herpetolog-
ical surveys (Doan, 2003). When an individ-

No. 534

ual was spotted, two observers positioned
themselves approximately 5 m away from the
subject. One observer recorded the behavior
of the focal animal while the other scanned
the area to identify other nearby individuals.
If no additional lizards or displays were
observed during the first 15 minutes of
observation, the lizard was considered “‘sol-
itary’ and was captured. If the initial lizard
displayed, or if another lizard was spotted
nearby, the observation was extended to
30 minutes. Following the procedure of
Talbot (1979), two lizards were considered
a pair if they were observed within 2 m of
each other, and no other lizards were seen
within 5 m during the observation period. In
most cases, paired males and females were
less than half a meter apart. On some
occasions, more than one lizard was ob-
served within 2 m of the initial animal. In all
of these cases, a single female was found in
the presence of multiple males; multiple
females were never found in such a group.
It seems likely that these groups represented
a pair and one or more intruders but,
because it was impossible to determine which
male or males were intruding, these cases
were excluded from the final analysis (but see
supplementary materials to see how the
inclusion of these animals would affect the
observed pattern). In some cases, more than
one individual was identified during an
observation, but not all lizards were success-
fully captured. If a lizard was spotted but not
captured, no animals from the observation
were included in subsequent analysis.

After 30 minutes of observation, attempts
were made to capture all lizards by hand or
by noose. Each lizard captured was mea-
sured, photographed with the dewlap ex-
tended, marked with a unique pattern of
colored ink dots on the ventral surface, and
released at the site of capture. The data
recorded for each individual were sex, snout-
vent length (svl, a standard variable for

2013

estimating body size in lizards), forelimb
length, hindlimb length, tail length, and the
length, depth and width of the head, to the
nearest tenth of a millimeter. Males less than
32 mm svl and females less than 35 mm svl
were considered juveniles (following Talbot
1979) and were excluded from further
analysis (but see supplementary materials to
see how inclusion of these animals would
affect the analysis). All measurements were
taken with digital calipers by the same
individual. The surface area of the dewlap
was measured for all males in ImageJ
(Abramoff et al, 2004) and scaled by
reference to a l-cm grid.

Statistical Analysis

The morphological traits of paired and
solitary animals were compared using an
analysis of covariance. These traits included
svl, fore- and hindlimb length, tail length,
head length, width and depth, and dewlap
area; body size was included as a covariate in
the analysis of all traits except body size.
Males and females were treated separately.
Because no tests approached significance, no
correction for multiple tests was used.

The correlation between body size for
members of a pair was calculated using
Pearson’s correlation coefficient. All statis-
tics were calculated in SPSS.

RESULTS

Among 150 animals observed and cap-
tured, 40 individuals were found in male-
female pairs; 33 males and 15 females were
observed alone; 22 males and 8 females were
observed in groups of more than two
animals; and 22 males were observed in
male-male pairs. Ten males were observed
with another animal, but the other individual
was not captured and the sex could not be
definitely assigned. When animals whose
pair status could not be assigned were

LIZARD ASSORTATIVE MATING 3

excluded (e.g., animals in groups, male-male
pairs, pairs where one animal was not
captured), 57.1% of females and 37.7% of
males were observed in pairs (44.9% average
for both sexes). Individuals in male-female
pairs were often interacting during the time
that they were observed, although copulation
was never witnessed. Visual displays were
performed by both males and females (males
displayed in 8/20 observations, whereas
females displayed in 2/20). Often one animal,
typically the male, followed the other up and
down a perch, and from one perch to
another (5/20 observations), often moving
slowly and frequently stopping until the
partner was quite close.

During the course of three observations, a
male and female in close proximity were
approached by a second male, who the first
male proceeded to chase away. While these
males were thus distracted, a third male
would suddenly appear and approach the
female while displaying. In one case, the
female approached the third male and
watched him display. In the other two cases,
the female retreated from the third male.
Copulation was never observed during the
course of these intrusions.

Male and female body size were positively
correlated for paired animals (Fig. 2; R =
0.50, P = 0.039), and these results were
qualitatively similar when smaller animals
were included or when animals found in
groups with one female and several males
were included (see supplementary materials);
however, neither body size (Fig. 2), body
dimensions, or dewlap area differed between
paired and solitary animals (see supplemen-
tary materials).

DISCUSSION

Pair formation and social monogamy are
rarely observed in reptiles, particularly squa-
mates; yet, during the course of this study,

6 BREVIORA

40 5

39 - A

a7
E |
E
>
Yn 36 °
re
oO
=
35 |
34 - é
33 Z Up at T = T 1
36 37 38 39 40 41 42
Female SVL (mm)
Figure 2. Correlation between male and female

body size for paired animals. Each pair is symbolized
by a grey diamond.

male-female pairs were observed quite fre-
quently. In most cases, members of a pair
were within 50 cm or less of each other and
were frequently observed displaying to each
other. Pairs also showed coordinated move-
ment, in which one animal (usually the male)
followed the female vertically or horizontally
through the habitat. These observations,
combined with Talbot’s (1979) observations
of pair persistence over 4-6 months, indicate
that these animals are not in association by
chance—they are in fact associating in pairs.
This social behavior is highly atypical for
lizards in general and for anoles in particular,
which more typically demonstrate resource-
defense polygyny (Losos, 2009). This behav-
1or may also be atypical for A. /imifrons,
which has been shown to defend polygynous
territories in other parts of its range. These
data also show one of the first reptilian
examples of SAP (size-assortative pairing).
We found no morphological differences
between paired and unpaired animals, sug-
gesting that mutual mate choice is an unlikely

No. 534

explanation for SAP. Two other hypotheses
to account for SAP, physical constraints and
mate availability, were neither supported nor
rejected by these data, but remain a fruitful
avenue for future research efforts.

Existence and Prevalence of Pairs

The rates of pairing observed in this study
were lower than those observed by Talbot:
45% vs. 70% of all observed individuals,
respectively. The lower percentage in this
study may have resulted because some paired
individuals may have been observed when
their partner was not visible, or pairs may
have been excluded from analysis due to the
presence of a temporary intruder. Unpaired
animals might also have been mistaken for
paired animals because of temporary prox-
imity to a member of the opposite sex.

It is also possible that the pairing behavior
is not typical of A. limifrons and occurs
occasionally at this location because of
unknown environmental or social factors.
In other well-studied populations of A.
limifrons, pairing between males and females
has never been reported, despite intensive
study (Andrews and Rand, 1983; Andrews
and Stamps, 1994). Rather, these popula-
tions exhibit a sedentary polygamous mating
system based on territory defense, a system
typical of anoles. A possible explanation is
that A. limifrons observed at sites outside La
Selva actually belong to a different species,
which has diverged in social behavior.
Recent morphological work has suggested
that A. limifrons might actually consist of
three or more distinct species (Kohler and
Sunyer, 2008). If A. limifrons at La Selva
comprise a unique lineage, their distinct
evolutionary history could have influenced
the evolution of their social behavior in a
number of ways, via shifts in habitat use,
population density, or predation risk, to
name a few possibilities.

2013

Another possibility is that A. /imifrons has
a high degree of behavioral plasticity in their
mating behavior and that local conditions
determine what strategy they pursue. This
would be consistent with the polygyny
threshold model, which suggests that indi-
vidual mating decisions may change when
the distribution of resources shifts to allow
higher densities or stable aggregations of
females, or both, that males can monopolize
(Altmann eft al, 1977). Although data on
forest characteristics were not collected as
part of this study, it appeared that pairs
occur more frequently in primary forest,
whereas clusters of animals were found in
great abundance in disturbed habitat. This
potential pattern could be a productive
direction for future studies.

Characteristics of Pairs

Although the data show SAP in A.
limifrons, 1 did not find support for the
mutual mate choice hypothesis, the physical
constraints hypothesis, or the mate availabil-
ity hypothesis. Paired and unpaired animals
did not differ in any of the morphological
variables that were considered, including
body size; head, limb, and tail dimensions;
or dewlap area, suggesting that mate choice
is not operating on these traits (see supple-
mentary materials for details). Mutual mate
choice may be operating on traits that we did
not consider, or a pattern of mate choice in
the traits that were examined could have
been obscured by the shortcomings of our
survey methods or confounding ecological
processes that also act on body size, such as
differential mortality.

Mating constraints seem unlikely to be
responsible for SAM in A. limifrons because
pairs were observed that were somewhat size-
mismatched (and that deviate substantially
from the observed correlation), but I cannot
rule out this possibility entirely because

LIZARD ASSORTATIVE MATING i)

physical constraints may only be relevant to
the most extreme mismatches, and I did not
observe all possible size combinations in the
population. Captive breeding experiments
with highly size-mismatched individuals
could be conducted to test this hypothesis
explicitly. Likewise, no spatial or temporal
discontinuity was observed in the distribu-
tion of size classes that would support a mate
availability mechanism as an explanation for
SAM. However, the spatial distribution of
body sizes was not explicitly examined, and
the duration of this study was insufficient to
uncover temporal patterns of variation in
body size. Further studies that explicitly
examine the spatial and temporal distribu-
tion of body size of paired and unpaired
animals may clarify this issue. In short, the
process that is responsible for the pattern of
SAP in A. limifrons remains obscure; uncov-
ering the process of pair formation in this
species will require substantial future efforts.

More Questions than Answers

The existence of socially monogamous
pairs in A. limifrons does not necessarily
imply that these pairs are also genetically
monogamous. Molecular studies of parent-
age will be necessary to characterize the
genetics of this unusual social behavior.
Moreover, we still have little conception of
how pairs are formed in the wild or what
benefits accrue to pair members. There is a
wealth of opportunity for future studies in
the field and in the lab on these questions.

The unusual behavior of A. /imifrons also
begs the question: Could social monogamy
be more common in lizards than previously
thought? The behavior of most species of
anole is poorly known from observations in
the field; indeed, the behavior of very few
lizard species has been studied in the wild.
Species that have been overlooked in previ-
ous studies are precisely the ones that are

8 BREVIORA

predicted to exhibit social or genetic monog-
amy; for example species with wide home
ranges without established territories or terri-
torial behavior, low densities, and cryptic
appearance and habits (Emlen and Oring,
1977). The unusual social behavior of A.
limifrons described in this study, combined
with the fact that it is relatively abundant, easily
observed, and geographically widespread,
could make this a useful species for future
studies of pairing behavior in squamates.

ACKNOWLEDGMENTS

Field assistance was provided by the
Reverend P. Humphrey, M. D., and C.
Tongier. Comments on the manuscript were
kindly provided by D. L. Mahler, E. H. Kay,
Y. Stuart, J. Losos, and two anonymous
reviewers. This work was funded in part with
a grant from the Organization for Tropical
Studies (OTS). Permission to work on A.
limifrons came from OTS and the Minesterio
del Ambiente y Energia (MINAE) of Costa
Rica. This work was authorized under Har-
vard University Animal Care Protocol 26-11.

LITERATURE CITED

ABRAMOFF, M. D., P. J. MAGELHAES, AND S. J. Ram.
2004. Image processing with ImageJ. Biophotonics
International 11(7): 36-42.

ALTMANN, S. A., S.S. WAGNER, AND S. LENINGTON. 1977.
Two models for the evolution of polygyny. Behav-
ioral Ecology and Sociobiology 2: 397-410.

ANDREwsS, R. M. 1985. Mate choice by females of the
lizard, Anolis carolinensis. Journal of Herpetology
19: 284-289.

ANpREws, R. M., AND A. S. RANb. 1983. Limited
dispersal of juvenile Anolis limifrons. Copeia 1983:
429-434.

Anprews, R. M., AND J. A. Stamps. 1994. Temporal
variation in sexual size dimorphism of Anolis
limifrons in Panama. Copeia 1994: 613-622.

BessA-Gome_s, C., J. CLOBERT, S. LEGENDRE, AND A. P.
Mo .ter. 2003. Modelling mating patterns given
mutual mate choice: the importance of individual
mating preferences and mating system. Journal of
Biological Systems 11: 205-219.

No. 534

Bisazza, A. 1997. Sexual selection constrained by
internal fertilization in the livebearing fish Xenotoca
eiseni. Animal Behavior 54: 1347-1355.

Brown, W. D. 1993. The cause of size-assortative
mating in the leaf beetle Trirhabda canadensis.
Behavioral Ecology and_ Sociobiology 33(3):
151-157.

BuLL, C. M. 1994. Population dynamics and pair fidelity
in sleepy lizards, pp. 159-174. In L. J. Vitt, and E.
R. Pianka eds. Lizard Ecology: Historical and
Experimental Perspectives. Princeton, New Jersey,
Princeton University Press.

BuL._, C. M. 2000. Monogamy in lizards. Behavioral
Processes 51: 7-20.

BuLL, C. M., G. S. BEDFoRD, AND B. A. SCHULTz. 1993.
How do sleepy lizards find each other? Herpetolo-
gica 49: 294-300.

BuLL, C. M., S. J. B. Cooper, AND B. C. BAGHUuRST. 1998.
Social monogamy and extra-pair fertilization in an
Australian lizard, Tiliqua rugosa. Behavioral Ecol-
ogy and Sociobiology 44: 63-72.

CHAPPLE, D. G., AND J. S. KeEoGH. 2005. Complex
mating system and dispersal patterns in a social
lizard, Egernia Molecular Ecology 14:
1215-1227.

Crespl, B. J. 1989. Causes of assortative mating in
arthropods. Animal Behavior 38: 980-1000.

Doan, T. M. 2003. Which methods are most effective for
surveying rain forest herpetofauna? Journal of
Herpetology 37: 72-81.

EMLEN, S. T., AND L. W. OrING. 1977. Ecology, sexual
selection, and the evolution of mating systems.
Science 197: 215-223.

Gans, C. 1996. An overview of parental care among the
Reptilia. Advances in the Study of Behavior 25:
145-157.

GARDNER, M. G., C. M. BuLL, AND S. J. B. Cooper.
2002. High levels of genetic monogamy in the
group-living Australian lizard Egernia_ stokesii.
Molecular Ecology 11: 1787-1794.

GorMAN, G. C. 1980. Anolis occultus, a small cryptic
canopy lizard: are there pair bonds? Caribbean
Journal of Science 15: 3-4.

Harari, A. R., A. M. HANDLER, AND P. J. LANDOLT.
1999. Size-assortative mating, male choice, and
female choice in the curculionid beetle Diaprepes
abbreviatus. Animal Behavior 58: 1191-1200.

JENSSEN, T. A., AND S. C. Nunez. 1998. Spatial and
breeding relationships of the lizard, Anolis caroli-
nensis: evidence of intrasexual selection. Behaviour
135: 981-1003.

Kaweckl, T. J. 1997. Sympatric speciation via habitat
specialization driven by deleterious mutations.
Evolution 51: 1751-1763.

whitil.

2013

KOHLER, G., AND J. SUNYER. 2008. Two new species of
anoles formerly referred to as Anolis limifrons
(Squamata: Polychrotidae). Herpetologica 64:
92-108.

Losos, J. B. 2009. Lizards in an Evolutionary Tree
Berkeley and Los Angeles, University of California
Press.

Masumoto, T. 1999. Size assortative mating and
reproductive success in the funnel-web spider,
Agelena limbata. Journal of Insect Biology 12:
353-361.

NAGEL, L., AND D. SCHLUTER. 1998. Body size, natural
selection, and speciation in sticklebacks. Evolution
52: 209-218.

O’Connor, D., AND R. Sune. 2003. Lizards in ‘nuclear
families’: a novel reptilian social system in Egernia
saxatilis (Scincidae). Molecular Ecology 12:
743-752.

Oxsson, M., AND R. SHINE. 1998. Chemosensory mate
recognition may facilitate prolonged mate guarding
by male snow skinks, Niveoscincus microlepidus.
Behavioral Ecology and Sociobiology 43: 359-363.

Pa, P., J. ERLANDSSON, AND M. SKoLp. 2006. Size-
assortative mating and non-reciprocal copulation in
a hermaphroditic intertidal limpet: test of the mate
availability hypothesis. Marine Biology 148:
1273-1282.

PeTriE, M., AND B. KEMPENAERS. 1998. Extra-pair
paternity in birds: explaining variation between
species and populations. Trends in Ecology and
Evolution 13(2): 52-58.

LIZARD ASSORTATIVE MATING 9

Rios-Lopez, N., AND A. R. PUENTE-COLON. 2007. Anolis
cuvieri (Puerto Rico giant anole): reproductive
ecology. Herpetological Review 38: 73-75.

SAVAGE, J. M. 2002. The Amphibians and Reptiles of
Costa Rica: A Herpetofauna between Two Conti-
nents, between Two Seas Chicago, University of
Chicago Press.

STAMPS, J. A. 1983. Sexual selection, sexual dimorphism,
and territoriality, pp. 169-204. In R. B. Huey, E. R.
Pianka, and T. W. Schoener eds. Lizard Ecology:
Studies of a Model Organism. Cambridge, Massa-
chusetts, Harvard University Press.

Stow, A. J., AND P. SuNNucks. 2004. High mate and site
fidelity in Cunningham’s skinks (Egernia cunnin-
hami) in natural and fragmented habitat. Molecular
Ecology 13: 419-430.

TALBoT, J. J. 1979. Time budget, niche overlap, inter-
and intraspecific aggression in Anolis humilis and A.
limifrons from Costa Rica. Copeia 1979: 472-481.

Toxopeus, A. G., J. P. Kruyit, AND D. HILLENtus. 1988.
Pair-bonding in chameleons. Naturwissenschaften
75: 268-269.

Trivers, R. L. 1976. Sexual selection and resource-
accruing abilities in Anolis garmani. Evolution 30:
253-269.

ULLeR, T., AND M. Oxsson. 2008. Multiple paternity in
reptiles: patterns and processes. Molecular Ecology
17: 2566-2580.

WITTENBERGER, J. F., AND R. L. TiLson. 1980. The evolution
of monogamy: hypotheses and evolution. Annual
Review of Ecology and Systematics 11: 197-232.

Note: Supplementary material referenced in this paper is available online at www.mcz.harvard.edu/Publications/.

™