Content uploaded by Julia Riley
Author content
All content in this area was uploaded by Julia Riley on Apr 13, 2021
Content may be subject to copyright.
Social and spatial patterns of two Afromontane crag lizards
(Pseudocordylus spp.) in the Maloti-Drakensberg Mountains,
South Africa
JULIA L. RILEY,
1,2
*JAMES H. BAXTER-GILBERT
3
AND MARTIN J. WHITING
4
1
Department of Botany and Zoology, Stellenbosch University, Stellenbosch, Western Cape, 7600,
South Africa (Email: julia.riley87@gmail.com);
2
Department of Biology, Dalhousie University, Halifax,
Nova Scotia, Canada;
3
Centre for Invasion Biology, Department of Botany and Zoology, Stellenbosch
University, Stellenbosch, Western Cape, South Africa; and
4
Department of Biological Sciences,
Macquarie University, Marsfield, New South Wales, Australia
Abstract Understanding the evolution of vertebrate sociality requires comparative data on social associations
across the vertebrate phylogeny. In the case of group-living lizards (i.e. species that live in stable social aggrega-
tions often associated with a shared resource), most work has focused on the Egerniinae in Australia, resulting in
a taxonomic and geographic skew to our understanding of reptile sociality. The African cordylid lizards (Cordyli-
dae) are also a promising system to study the evolution of sociality because grouping behaviour varies across the
clade. Here, we studied the conspecific grouping behaviour of two crag lizards, Pseudocordylus langi and P. melan-
otus subviridis that occur at high elevations in the Maloti-Drakensberg Mountains of South Africa. To better
understand their social organisation and mating system, we also present data on their spatial distribution, sexual
dimorphism, and bite force. Both Pseudocordylus spp. were sexually dimorphic in morphology (males had larger
heads than females of similar body size), colouration (males were more colourful) and female P. langi had a
weaker bite force than males. Both P. langi and P. m. subviridis were associated with rocky habitat on the moun-
tainside (e.g. cliffs, rock buttresses, and rock outcrops) and both were spaced apart and rarely in groups (79% of
P. langi and 90% of P. m. subviridis were observed alone). Based on our findings, we hypothesise that both Pseu-
docordylus spp. have a territorial social structure and a polygynous mating system. This novel natural history
information about crag lizards supports the assertion that Cordylidae is another model system for examining the
evolution of sociality.
Key words: aggregation, bite force, polygyny, sexual dimorphism, social behaviour, squamate.
INTRODUCTION
Social life is fundamental to the emergence of com-
plex behaviours across many animals, including
birds, mammals, and invertebrates (Ward & Webster
2016). In a recent synthesis on social evolution,
sociality was defined as co-operative group living
(Rubenstein & Abbot 2017). More broadly, sociality
can be thought of as the degree to which individuals
within a species interact with one another and the
complexity of these social interactions (i.e. the num-
ber of individuals in a social unit, the nature of their
interactions, as well as their mating and care systems)
all of which vary greatly across the animal kingdom
(Whiting & While 2017; Kappeler 2019). Coopera-
tion, cognitive ability, and complex vocal repertoires
are all behaviours that may have evolved as an out-
come of a species’sociality (Byrne & Bates 2007;
Sussman & Cloninger 2011). Lizards are an
emerging taxon that, over the last few decades, have
been documented to exhibit a wide range of sociality.
To date, some form of social grouping has been doc-
umented in 94 species from 22 taxonomic families
(Whiting & While 2017). An important dimension of
sociality is social organisation, which refers to the
size, demographic composition, stability, and genetic
structure of a social unit (Kappeler 2019). An Aus-
tralian lineage of skinks, the subfamily Egerniinae,
may be the most extensively studied taxonomic group
of reptiles to date in terms of their sociality. This
group contains species that are largely solitary and
promiscuous, others that form long-term monoga-
mous pairs, and even species that live in large com-
munal family-groups (Chapple 2003; While et al.
2015; Whiting & While 2017). These recent insights
into the diversity of lizard social life, coupled with
the presumption that the ancestral state (i.e. traits of
basal common ancestors) of squamates was solitary
(O’Connor & Shine 2003) and the fact that lizards
are amendable to both behavioural observation and
experimental manipulation, suggests they may be a
*Corresponding author.
Accepted for publication February 2021.
© 2021 Ecological Society of Australia doi:10.1111/aec.13030
Austral Ecology (2021) ,–
source of valuable insights into to the diversity and
evolution of animal sociality.
The cordylid lizards of Africa are a promising model
for understanding the evolution of sociality because
they represent a different, albeit closely related, clade
to the comparatively well-studied social egerniinae
skinks (Pyron et al. 2013). A well-supported phylogeny
exists for Cordylidae (Stanley et al. 2011) and studies
have shown that at least four species live in stable
aggregations and more are anecdotally reported to be
gregarious (Fig. 1, Appendix S2). One of these species
is Lang’s Crag Lizard (Pseudocordylus langi), for which
anecdotal observations in the wild have noted this spe-
cies lives in ‘colonies’at high elevations in the Maloti-
Drakensberg Mountains (Broadley 1962). Depending
on the source, the terms used to describe aggregations
of this species vary between ‘small groups’and ‘colo-
nies’, however, definitions of these terms or further
description of this species’behaviour were not detailed
in the original account (Broadley 1962). A closely
related, sympatric lizard within this region is the Drak-
ensberg Crag Lizard (Pseudocordylus melanotus sub-
viridis), which has similarly been anecdotally reported
to live in ‘colonies’(Bates et al. 2014). In general, the
term ‘colony’can be used to refer to a group of ani-
mals that live and interact closely with one another –it
is often applied to eusocial insects or seasonal breeding
congregations of birds or pinnipeds. A more detailed
study of the aggregations of these two crag lizards
would help to clarify their nature (i.e. determining if
aggregations of P. langi are similar to other group-liv-
ing lizards) and increase our knowledge of sociality
across the Cordylidae phylogeny. Information such as
this would greatly inform future research regarding the
evolution of sociality within the Cordylidae, as well as
predictions about the nature of other cordylid lizards’
sociality, for example, whether they could be family-
living and potential phylogenetic constraints.
We investigated whether, and to what degree, both
species (P. langi and P. m. subviridis) aggregate. Our
focus was to collate data on the natural history of
these lizards in order to gain insights into their
potential social systems. Known group- and family-
living lizard aggregations often (1) consist of a mating
pair and their offspring, and (2) share a limited
resource with one another, which forms an initial
base for their interactions and preferential associa-
tions (Whiting & While 2017). Thus, we focused on
surveying the gregariousness (i.e. the frequency of
grouping) and social organisation (defined as the size
and demographic composition of a group; Kappeler
2019) of crag lizard conspecific groups, as well as
determine if lizards share retreat sites, which would
establish an understanding of the potential proximate
causes behind aggregations.
We also collected data on sexual dimorphism in
these species, since the degree to which it is
expressed can reflect a species’social structure and
mating system (Whiting & While 2017). Family-liv-
ing and long-term, stable social aggregations, as well
as monogamous mating systems, typically occur in
sexually monomorphic lizards because there is less
sexual conflict (Table 1). Conversely, territorial
lizards are often polygynous and males are often
more colourful, have larger heads than females, and
have greater bite force in response to male–male con-
test competition (Fox et al., 2003). In general, mon-
tane cordylid lizards vary in their degree of sexual
size dimorphism (SSD) between species (Mouton &
Van Wyk 1993; Van Wyk & Mouton 1998; Mouton
et al. 2005; Costandius & Mouton 2006; Bates
2007). Previous work has shown that male-biased
body and head size dimorphism occurs in P. m. sub-
viridis (Mouton & Van Wyk 1993). Male P. langi
tend to be larger than females, yet differences in head
size and colouration have not been adequately
described in wild populations (Bates 2007). We
examined individuals of both species and quantified
their morphological traits, documented colouration,
and measured bite force. Thus, if there is further evi-
dence for sexual dimorphism in both crag lizards,
this would suggest that some aspect of contest com-
petition is playing a role in their social interactions.
By combining these multiple lines of natural history
information, we aimed to provide insights into the
social system of two crag lizards, which will con-
tribute to our understanding of both lizard and verte-
brate sociality.
METHODS
Study Species
There are five described crag lizards (Psuedocordylus spp.;
Stanley et al. 2011), of which four taxa (species and sub-
species of this genus) reside in the central and northern
regions of the Maloti-Drakensberg (Bates et al. 2014) (see
supplementary materials for more details). Pseudocordylus
langi is found in Lesotho and the Free State and KwaZulu-
Natal provinces of South Africa (Bates 2007). They are a
medium-sized, viviparous cordylid lizard (maximum
recorded snout-vent length, SVL, is 106 mm; Reissig 2014)
and are found within crevices and on cliffsides at the
escarpment edge and rocky buttresses of the Maloti-Drak-
ensberg Mountain range. Their habitat is limited as they
only occur over 2600 m a.s.l. (Reissig 2014). Pseudocordylus
langi is endemic to southern Africa and globally listed as
‘Near Threatened’(Bates & Cunningham 2017). Its persis-
tence is potentially threatened by their limited ability to dis-
perse, particularly in the face of global climate change, and
may be affected by over-collecting and disturbance at some
hiking trails in the Maloti-Drakensberg (Bates et al. 2014).
Pseudocordylus m. subviridis is also a medium-sized, vivi-
parous cordylid lizard (maximum recorded SVL is
140 mm; Bates 2007). They are more abundant than P.
doi:10.1111/aec.13030 © 2021 Ecological Society of Australia
2J.L.RILEYET AL.
langi and range widely across Lesotho and south-eastern
South Africa (Bates 2007; Bates et al. 2014). They are a
saxicolous lizard inhabiting rock outcrops, mountainsides
and rock crevices, and can be found in several high eleva-
tion areas. As such, within suitable habitat in the Maloti-
Drakensberg, they co-occur with P. langi (Broadley 1962).
Fieldwork
In October 2019, we conducted our fieldwork within the
central and northern regions of the Maloti-Drakensberg
Transfrontier Park, which are part of the uKhahlamba-
Drakensberg World Heritage Site. There were two aspects
to our fieldwork: (1) capture and measurement of lizards to
examine sexual dimorphism, and (2) transects observing
lizard spatial and social patterns.
Documenting Crag Lizard Phenotypic Traits
Before our transect surveys, from 14 to 15 October, we
captured lizards by lassoing (i.e. a loop of fishing line at the
end of a pole). Once captured, we measured each lizard’s
head length, width, and height using digital callipers
Fig. 1. Phylogeny of Cordylidae (adapted from Stanley et al.2011) with species that exhibit year-round stable aggregations
(orange squares following the species’name) and species that anecdotally have been observed in groups (purple triangles). We
represent if the species is grouping for ecological or social factors using a black ‘e’or ‘s’, respectively. We also highlight spe-
cies that have are known or hypothesised to have a territorial social structure (blue circles), including the two Pseudocordylus
spp. in this study (indicated with a black asterisk). The other species, to the best of our knowledge, lack data on their social-
ity, and a detailed account of the literature this figure is based on can be found in our supplementary material (Appendix S2).
© 2021 Ecological Society of Australia doi:10.1111/aec.13030
SOCIAL AND SPATIAL PATTERNS OF CRAG LIZARDS 3
(0.01 mm), as well as SVL using a clear plastic ruler
(1 mm). We used a cut-off of 80 cm to delineate between
adults and juveniles of both species (Mouton & Van Wyk
1993; Bates 2007). We were able to determine the sex of
both species visually using descriptions in Mouton and Van
Wyk (1993) and Bates (2007) for P. m. subviridis and field
guides for P. langi (Alexander & Marais 2007; Reissig
2014). In general, male crag lizards are brighter in coloura-
tion and have obviously wider heads and larger bodies than
females and juveniles (Mouton & Van Wyk 1993; Alexan-
der & Marais 2007; Reissig 2014). We also verified our
visual assessments by sexing males through hemipenal ever-
sion. We took photographs to document each individual’s
colouration, but appreciate that visually scoring colour is
subjective and biased by human perception. It will fail to
take into account any colouration such as ultraviolet, which
is not detectable by humans. However, our aim was simply
to document wholescale visual differences between the
sexes and not take into account spectral (chromatic and
achromatic contrast) characteristics. Bite force (N) was
measured using an isometric force transducer connected to
a Kistler charge amplifier (type 5995; Kistler Inc.
Wintherthur, Switzerland). Lizards were coerced to bite on
two parallel plates (fixed at a distance of 1 mm), by strok-
ing both sides of their jaw simultaneously with our index
finger and thumb (Baxter-Gilbert & Whiting, 2018). Each
individual was tested five consecutive times, and the maxi-
mum bite force was used in analyses (Anderson et al. 2008;
Baxter-Gilbert & Whiting, 2018). Lizards were released at
their location of capture within 24 h. One day elapsed
between our capture period and the beginning of our tran-
sect surveys to allow lizards to re-acclimatise to their sur-
roundings.
Transect Surveys to Observe Social and Spatial Patterns
We observed P.langi and P. m. subviridis along the Sentinel
Trail (a hiking trail) in the Maloti-Drakensberg Mountains
of South Africa. We used the hiking trail as a transect
(Fig. 2) that we surveyed once per species to avoid repeat
sampling of the same individuals (see below for more
details). Along each transect, we used binoculars to visually
observe lizards within 5 m of either side of the trail and
recorded each lizard’s location using a GPS (3–5 m),
demographics (i.e. male, female or juvenile), the habitat
they were sighted on, their behaviour (i.e. basking, moving,
hiding within a refuge or foraging), and whether or not they
were observed in a group.
We were able to determine each species’demographics
(i.e. male, female, or juvenile) visually using binoculars, so
Table 1. Sexual dimorphism of family-living lizards or lizards with kin-biased associations. The definition of sexual dimor-
phism is a distinct difference in size or appearance between the sexes, in addition to the sexual organs themselves. In this
table, we are specifically interested in if lizards are visually distinguishable from one another, and do not consider fine differ-
ences (e.g. slight differences in head measurements researchers might use to identify sex through analyses after fieldwork) as
sexual dimorphism
Family Species
Sexually
Dimorphic Description References
Agamidae Intellagama lesueurii
†
Yes The abdomens and chests of large
males are covered with orange–
red to reddish-black colouration
that is absent in females
Piza-Roca et al. (2019);
Thompson (1993)
Liolaemidae Liolaemus leopardinus No –Brito (2017)
Scincidae Liopholis kintorei No
§
–McAlpin et al. (2011); Chapple
(2003); Dennison (2015)
Scincidae Liopholis whitii No –Chapple and Keogh (2006),
While et al. (2009)
Scincidae Tiliqua rugosa
‡
No
§
–Bull and Pamula (1996), Bull
(2000), Bull and Lindle (2002)
Scincidae Bellatoris major No –Osterwalder et al. (2004),
Shea (1999)
Scincidae Bellatoris frerei No –Fuller et al. (2005)
Scincidae Egernia kingii No
§
–Masters and Shine (2003)
Scincidae Egernia cunninghami No
§
–Stow and Sunnucks (2004)
Scincidae Egernia saxatillis No –O’Connor & Shine (2003)
Scincidae Egernia striolata No
§
–Duckett et al. (2012)
Scincidae Egernia stokesii No –Gardner et al. (2001)
Scincidae Gnypetoscincus
queenslandiae
No
§
–Sumner (2006); Sumner et al.
(1999)
Scincidae Corucia zebrata No –Hagen et al. (2013)
Xantusiidae Xantusia vigilis No –Davis et al. (2011)
†
Rather than family-living (residing in long-term stable, social aggregations with relatives), conspecifics have kin-biased
social associations within a territorial social system.
‡
Largely solitary-living, but pairs up with long-term mates before and during the breeding season.
§
Species exhibits dimorphism in morphology, which through analyses, can be used to identify sex. But, the degree to which
this can be observed by eye is subjective.
doi:10.1111/aec.13030 © 2021 Ecological Society of Australia
4J.L.RILEYET AL.
lizards were not captured during transect sampling. We
familiarised ourselves with the visual differences between
species and sexes during the capture period (see details
above). Lizards that were classified at juveniles from visual
observations were obviously smaller than adults, exhibited
juvenile colouration, and had less-developed bulbous jowls
(Appendices S3 and S5). Males were visually differentiated
from females by their pronounced, bulbous jaw muscles
(i.e. larger and wider heads), and brighter colouration (e.g.
head and flanks were yellow, orange, and orange–red in
P. m. subviridis, and yellow, green, and turquoise in P.
langi; Appendices S3 and S4).
Lizards were recorded as within a group if they met three
criteria: (1) they were less than 3 m from one another, (2)
they were in visual (i.e. line of sight) or physical contact (i.e.
touching each other), and (3) not exhibiting aggressive
behaviour towards each other (e.g. biting, chasing, or pos-
turing with head-bobbing, push-ups, or badge displays; Fox
et al. 2003). Lizards in social groups may still act aggres-
sively towards one another (Riley et al. 2017), but they are
more likely to exhibit aggressive behaviour towards non-
group members (Fox et al. 2003). Further, our third crite-
rion did not play a large role in our study, as we only
observed one aggressive encounter (i.e. head-bobs and
chasing) during the transect surveys, which occurred
between an adult male P. langi, that appeared to be guard-
ing a water spring on a cliff face, and a juvenile conspecific.
Also, although our first criterion (i.e. a 3 m limit) is subjec-
tive, we based it on our previous observations of other cor-
dylids in the field in which individuals either show tolerance
to familiar group members close to a shared resource or dis-
play at rivals. Also, our criteria for designating if lizards
were within a group or not are also similar to other studies
of lizard sociality (Whiting & While 2017) and are intended
to reflect that lizards were aware and tolerant of one
another. There were a few cases where lizards were less than
3 m from a conspecific but were unaware of each other due
to the presence of rocks or grasses obstructing their view; in
these cases, we recorded these lizards as solitary and visually
estimated the distance between them (to the nearest
10 cm).
If lizards were observed in a group, we recorded if indi-
viduals were in visual or physical contact (see definitions
above), if individuals were sharing a shelter site (i.e. rock or
cliff crevice, burrow), and a visual-estimation of the dis-
tance (nearest 10 cm) between individuals within the
group. Also, if they retreated during observations, we noted
what habitat they retreated to (i.e. rock or cliff crevice, bur-
row, space between two rocks, or a space within an artificial
structure like a fence or brickwork), and if their retreat site
was shared with other group members. This allowed us to
establish if lizard groups were associated with a particular
resource (i.e. shelter site).
Along the Sentinel Trail, the transects we hiked were spe-
cies-specific. We were able to visually-distinguish between
species using their distinct dorsal patterning (Bates 2007).
These two lizards co-occur along the hiking trail after 2915
a.s.l. (Fig. 2), but P. m. subviridis also occurs at lower eleva-
tions, so we decided to survey each species separately and
focus on conspecific groups. On 17 October 2019, we walked
our P. m. subviridis transect, from the Sentinel Trail parking
lot to the base of the chain ladders (4.38 km in length). We
sampled P. langi along a 1.42 km transect on 19 October
2020. This transect began at the base of the chain ladders
(2954 m a.s.l) and extended down the hiking trail to the
point where we observed our last P. langi (2922 m a.s.l.). We
also included eight additional sightings of P. langi collected
while climbing the chain ladders (maximum elevation of
3014 m a.s.l.) on the previous day (18 October 2020). These
transects spanned an elevational gradient from 2543–3014 m
a.s.l., and all occurred within the lizard’s active period (i.e.
mid-morning to early afternoon). Daily average ambient
temperatures were also similar (20 to 21°C), and we passed a
similar number of hikers on our transects (ranging between
35 to 44 people).
As opportunities presented themselves, we also noted the
grouping behaviour of both species in two additional loca-
tions. These are separate to the transects we describe
above, so no repeated sampling of lizards occurred. On 20
October 2020, we ascended the mountain again and
searched the Phofung Plateau for lizards. We also hiked the
same pathway as outlined in Broadley (1962) along the
Organ Pipes Pass near Cathedral Peak in the Maloti-Drak-
ensberg on 23 October 2019. Approaching Organ Pipes
Pass, we began noting the presence of both P. m. subviridis
(n=50 total) after reaching an elevation of 2358 m a.s.l.,
and P. langi (n=13 total) after reaching 2743 m a. s. l.
We have included a description of our observations from
these additional surveys below.
Statistical Analyses
Behavioural and spatial observations were summarised and
compared between species qualitatively. We used R version
3.5.0 to test for differences in morphological traits and bite
force between males and females for each species (R Core
Team 2018). All morphological traits and SVL were log
10
-
transformed before analyses to ensure a scalar linear rela-
tionship (Lailvaux et al., 2004; Baxter-Gilbert et al. 2020).
First, we used a Type I analysis of variance (ANOVA) to
test for a difference in SVL between sexes, which was per-
formed using the R package ‘car’and the function ‘aov’
(Fox & Weisberg 2019). Then, we analysed differences in
head morphometrics (i.e. head width, length, and height)
between sexes using an analysis of covariance (ANCOVA)
that also included SVL to control for sex-specific differ-
ences in body size (using the function ‘Anova’specifying
‘type=”III”’ from the R package ‘car’; Fox & Weisberg
2019). Before performing the ANOVAs, a Levene’s test
(using the ‘Levene Test’function in the R package ‘car’) was
performed to check for homogeneity of variance. If variance
did not fit this assumption, then we used a heteroscedastic-
ity-corrected coefficient (also termed a ‘White-corrected’or
‘White-Huber’; White 1980) covariance matrix to run
ANOVAs (Type II) or ANCOVAs (by specifying ‘white.ad-
just =TRUE’in the ‘Anova’function of the R package
‘car’; Fox & Weisberg 2019). Additionally, we used a prin-
ciple components analysis (PCA) to summarise measure-
ments of SVL, head width, head length and head height.
We used the function ‘prcomp’from the R package ‘factoex-
tra’to perform the PCAs (Kassambara & Mundt 2020). All
means are reported 1 standard deviation and sum-
marised from raw data.
© 2021 Ecological Society of Australia doi:10.1111/aec.13030
SOCIAL AND SPATIAL PATTERNS OF CRAG LIZARDS 5
Ethical Statement
Research methods were approved by the Stellenbosch
University Animal Ethics Committee (Protocol # ACU-
2019-6766) and were permitted in the Free State (Permit
No. 201910000003314) and KwaZulu-Natal (Ezemvelo
KZN Wildlife, Permit No. OP 3486/2019).
RESULTS
Observations of Aggregations
Along the Sentinel Trail, 10% (n=12/115) of P. m.
subviridis were observed in groups. In total, we
observed 45 females, 44 males, 24 juveniles, and two
of unknown sex and age class. Only groups of two
individuals were observed, and the demographics of
groups varied (Table 2). Of these groups, most
lizards were basking on rock outcrops, but one group
was observed refuging together within a rock crevice.
P. m. subviridis within groups ranged from 0.15 to
2.00 m from one another (average of 0.9 0.7 m,
n=6). Only one group was found sharing a rock cre-
vice, all other groups were observed in visual contact
with one another while basking on the same rock
(n=5). All individuals, when disturbed, retreated to
separate locations (other than the group that was
already sharing a crevice). Similarly, we did not
observe any P. m. subviridis groups on the Pofung
Plateau, while 16% (8/50) were in groups along
Organ Pipes Pass where we observed 21 females, 20
males and 9 juveniles (Table 2).
Along the Sentinel Trail, 21% (n=15/72) of P.
langi were observed in groups. In total, we observed
27 females, 19 males, 23 juveniles, and three of
unknown sex and age class. We observed six groups
of two individuals and one group of three individuals,
all of which occurred while lizards were basking on
cliffs. The demographics of groups varied greatly
among observations (Table 2). Within groups, P.
langi ranged from 0.1 to 3.0 m from one another (av-
erage of 1.4 1.1 m, n=7), and were observed in
visual contact on the same cliff face. Of groups where
we observed their retreats, two out of five groups
retreated to the same location. On the Pofung Pla-
teau and Organ Pipes Pass, P. langi were never
observed in groups (Table 2).
Spatial Observations
The nearest neighbour distance between individual
P. m. subviridis was, on average, 46.5 57.2 m (me-
dian =22.7 m), and ranged between 2.8 to 321.3 m
for individuals that were solitary (n=103). For P.
langi, the nearest neighbour distance between lizards
was 22.6 32.2 m (median =8.1 m), and ranged
between 1.0 to 122.3 m for individuals that were
solitary (n=57).
Morphology and Bite Force
We captured 12 P. m. subviridis (5 females, 3 juve-
niles, and 4 males) and 15 P. langi (4 females, 5
juveniles, and 5 males). Male P. m. subviridis were
significantly larger than females (F
1,7
=8.60,
p=0.02). From raw data, the SVL of female P. m.
subviridis averaged 84.80 2.17 mm, while male
SVL was on average 97.25 10.05 mm (Fig. 3a).
Juvenile P. m. subviridis averaged 75.67 1.15 mm
in SVL. Male P. langi (93.00 9.80 mm) tended to
Fig. 2. Locations of (a) Pseudocordylus m. subviridis and (b) Pseudocordylus langi along their respective transects (indicated by
the white path). The black inset box in the left map indicates the border of the right map. In both species, the lighter fill indi-
cates individuals observed alone, and the darker fill shows individuals that were observed in a group. The map’s details have
been blurred to protect the exact location data.
doi:10.1111/aec.13030 © 2021 Ecological Society of Australia
6J.L.RILEYET AL.
be larger than females (82.00 1.41 mm; Fig. 3b),
and this difference was significant (F
1,8
=6.12,
p=0.04). Juvenile P. langi averaged
72.25 8.34 mm in SVL.
Pseudocordylus m. subviridis head length, but not
their head width or height, was significantly related
to their SVL (length: F
1, 6
=9.71, p=0.02, width:
F
1, 6
=5.14, p=0.06, height: F
1, 6
=0.33, p=
0.59). Male P. m. subviridis heads (22.66 2.98
mm) were wider than females (16.93 1.35 mm),
yet this trend was not significant (F
1, 6
=3.46,
p=0.11). Head length (female =22.97 0.77 mm,
male =29.69 3.19 mm; F
1, 6
=8.81, p=0.03)
and head height (female =8.11 0.52 mm, male =
10.31 0.68 mm; F
1, 6
=9.91, p=0.02) was signif-
icantly larger in males than females. Our PCA shows
clear differentiation in head measurements between
P. m. subviridis females, juveniles, and males
(Appendix S1, Fig. 4a).
With respect to P. langi, head width (fe-
male =15.29 0.77 mm, male =21.08 3.03 mm;
F
1, 7
=40.44, p<0.01) and head length (fe-
male =20.87 0.26 mm, male =216.32 3.17 mm;
F
1, 7
=27.98, p<0.01) differed between sexes. Yet,
head height did not differ between females (8.16
0.65 mm) and males (9.77 1.66 mm) (F
1, 7
=
0.06, p=0.82). Pseudocordylus langi head morphol-
ogy was significantly related to their SVL (length:
F
1, 7
=53.34, p<0.01, width: F
1, 7
=46.04,
p<0.01, height: F
1, 7
=22.99, p<0.01). Our PCA
shows differentiation in head measurements between
P. langi males and females, as well as males and
juveniles, but not females and juveniles
(Appendix S1, Fig. 4b).
Maximum bite force was not related to P. m. sub-
viridis SVL (F
1, 4
=0.01, p=0.94), but it was signifi-
cantly related to P. langi SVL (F
1,6
=26.16,
p<0.01). Male P. m. subviridis maximum bite force
(36.37 5.54 N) tended to be greater than female
maximum bite force (17.53 3.36 N), but this dif-
ference was not significant (F
1, 4
=5.09, p=0.09).
This is likely due to our limited sample size. For P.
langi, maximum bite force was significantly lower in
females than males (F
1, 6
=15.97, p=0.01). Maxi-
mum bite force of P. langi females was, on average,
13.28 2.04 N and for males was 26.58 8.42 N
(Fig. 5b).
Colouration
Both species’colouration differed between sexes.
This has previously been well-documented in P. m.
subviridis (Mouton & Van Wyk 1993; Bates 2007).
This species has smooth dorsal scales with small,
wide granular black scales patterning their dorsal sur-
face. Males and females have black mottling and
striping on their backs, due to stippling from these
raised micro-scales (Appendix S3). On males, the
background to this dorsal patterning can vary
between light green to yellow, whereas in females the
background is often yellow-brown to olive-brown.
The central, ventral scales of both males and females
are white. The male’sflanks are very colourful (vary-
ing from yellow, orange, and orange–red). Females
also can have small amounts of colouration on their
sides, but the colouration is not as deeply saturated
as males (i.e. light yellow and orange instead of
Table 2. Group size and composition for the aggregations of two crag lizards (Pseudocordylus langi and P. melanotus sub-
viridis) we observed along the Sentinel Trail and Organ Pipes Pass, and on the Pofung Plateau in the Maloti-Drakensberg
Mountains, South Africa
Species
Number of individuals in
groups/Total observations Group size Group composition
Sentinel Trail (Fig. 2)
P. m. subviridis 12/115 (6 groups) 2 (n=6) 1F and 1J (n=3)
1F and 1M (n=2)
2J (n=1)
P. langi 15/72 (7 groups) 2 (n=6)
3(n=1)
2F (n=2)
1F and 1J (n=2)
1F and 2J (n=1)
1F and 1M (n=1)
1J and 1M (n=1)
Pofung Plateau
P. m. subviridis 0/8 ––
P. langi 0/7 ––
Organ Pipes Pass
P. m. subviridis 8/50 (3 groups) 2 (n=2)
4(n=1)
1F and 1M (n=2)
4J (n=1)
P. langi 0/13 ––
© 2021 Ecological Society of Australia doi:10.1111/aec.13030
SOCIAL AND SPATIAL PATTERNS OF CRAG LIZARDS 7
deeper orange to red). No juveniles we observed had
colouration on their sides or distal ventral scales.
Some males also had light blue to turquoise coloura-
tion on their heads, central ventral scales, and on
their arms.
Pseudocordylus langi also differed in the colouration
between sexes (Appendix S4). Their dorsal surface
was very smooth and was patterned with black stripes
and blotches. This dark patterning was often inter-
spersed with light crossbars. The background colour
of their back varied between yellow, olive, and green-
ish-grey, and often the males appear brighter in their
colouration than females. This species’also have two
to six bright blue, almost iridescent, spots extend from
the neck posteriorly along the back of the body in two
lines (Appendix S5). The ventral colouration varies
from light grey to blue, and males often had more satu-
rated blue on their venter. Males also had yellow to
green colouring on the distal ventral sides, red-blueish
centrally on their undersides, and had a blueish-grey
tinge to the base of their heads. Interestingly, both spe-
cies had a black throat patch that also has been noted
by Bates (2007), which we observed being extended
during social signalling (Appendix S6).
DISCUSSION
Our observations suggest that P. m. subviridis and P.
langi are not group-living, which is further supported
by their sexual dimorphism in body size, colouration
and bite force –traits often consistent with territorial-
ity and polygynous mating systems in lizards. Both
Pseudocordylus spp. were typically associated with
specific habitat types that are clumped on the land-
scape-scale, which resulted in the localised presence
Fig. 3. Differences between sexes in snout-vent length (mm) of (a) Pseudocordylus m. subviridis and (b) P. langi captured along
the Sentinel Trail, Maloti-Drakensberg Mountains, South Africa. Raw data is plotted. P-values are shown between sex-specific
comparisons and significant differences are denoted using an asterisks.
Fig. 4. Biplot of the first and second principles components assessing differences in snout-vent length (SVL), and head
measurements (HW =head width, HL =head length, and HH =head height) in (a) Pseudocordylus m. subviridis and (b) P.
langi among juveniles (represented using triangles and a beige colour), males (represented using squares and a blue colour),
and females (represented using circles and a red colour).
doi:10.1111/aec.13030 © 2021 Ecological Society of Australia
8J.L.RILEYET AL.
of lizards in patches along our transects. For P. m.
subviridis their preferred habitat was typically rock
outcrops on mountain slopes, while P. langi were
almost exclusively seen on cliffsides and rock but-
tresses. The habitat type preferences we observed for
both species support what has been previously
described (Broadley 1962; Bates 2007; Reissig 2014;
Bates et al. 2014). Habitat did not appear to be lim-
ited, and thus we would not expect lizards to be dri-
ven to aggregate due to a lack of resources (i.e. the
ecological constraints hypothesis; Hatchwell & Kom-
deur 2000). Our behavioural observations support
this; within areas of suitable habitat, lizards were
spaced apart, and a low percentage of lizards were
grouping (i.e. individuals were less than 3 m to one
another, in visual or physical contact, and not
exhibiting aggressive behaviour). Specifically, 10% of
P. langi and 21% of P. m. subviridis were observed in
groups and rarely shared refuges. Also, groups were
small –the majority were pairs. Most commonly, P.
m. subviridis groups were either a male and female or
a female and juvenile, while P. langi groups were
composed of an adult and juvenile(s). We also
observed a similar lack of aggregative behaviour on
the Phofung plateau and along the Organ Pipes Pass,
which further supports our finding that both species
are not group-living.
The behaviour we observed in both Pseudocordylus
spp. differed from group- and family-living lizards
that typically bask near their shelters (i.e. rock or tree
crevices, plants, or burrows) in small to large groups
that are in close visual and physical contact with one
another (e.g. cordylids like Armadillo Lizards, Ouro-
borus cataphractus, and egerniinae skinks like the
Cunningham’s Skink, Egernia cunninghami, and the
Great Desert Skink, Liopholis kintorei; Whiting &
While 2017). Instead, their fine-scale spacing was
more akin to Australian Water Dragons (Intellagama
lesueurii; Piza-Roca et al. 2019) and Augrabies Flat
Lizards (Platysaurus broadleyi; Whiting 1999, Whiting
et al. 2003); whereby individuals are regularly spaced
within suitable habitat. In these examples, regular
social interactions between individuals do occur;
including aggressive encounters during territorial dis-
putes, as well as social tolerance for shared shelter
sites in Platysaurus broadleyi when habitat is limited
(Schutz et al. 2007) and kin-based social associations
in the case of I. lesueurii (Piza-Roca et al. 2019).
Thus, although we expect the social systems of crag
lizards to differ from lizards that have been described
as group-living and/or gregarious (Whiting & While
2017), we are not suggesting they are asocial. Rather,
below, we provide a hypothesis as to the nature of
their sociality. Further, and based on our observa-
tions of P. langi and P. m. subviridis, we assert that
although these species have previously been described
as living within ‘colonies’(Broadley 1962; Bates et al.
2014; Reissig 2014), this term was likely originally
used in reference to the clumped distribution of
lizards on the landscape rather than as a description
of the nature of their sociality.
Both Pseudocordylus spp. were sexually dimorphic in
morphology and colouration with a male-bias for lar-
ger size and ‘brighter’colouration. Our limited sam-
ple size prevented us from making comparisons of the
morphology between species. Yet, an interesting ave-
nue for future research would be comparing the mor-
phology of crag lizards occurring within the Maloti-
Drakensberg Region, and assessing if that may relate
to their species-specific habitat use. For example, is
the morphology of P. langi constrained in some way
(i.e. head or body height, limb or foot morphology)
by their predominant use of vertical cliffs and narrow
crevices? In this study, we restricted our comparison
of crag lizard morphology between sexes within a spe-
cies. The majority of the differences we detail herein,
Fig. 5. Differences between sexes in maximum bite force (N) of (a) Pseudocordylus m. subviridis and (b) P. langi captured at along
the Sentinel Trail, Maloti-Drakensberg Mountains, South Africa. Raw data is plotted. P-values are shown between sex-specific
comparisons and significant differences are denoted using an asterisks.
© 2021 Ecological Society of Australia doi:10.1111/aec.13030
SOCIAL AND SPATIAL PATTERNS OF CRAG LIZARDS 9
especially in their colouration, are visually distinguish-
able from a distance with binoculars.
Pseudocordylus m. subviridis females were smaller in
SVL, head length, and head height than males. We
did not find a significant sex-difference in P. m. sub-
viridis head width or bite force, but non-significance
is likely due to our limited sample size. Other studies
have established that females have smaller heads and
body size than males (Mouton & Van Wyk 1993),
and our PCA demonstrated clear differentiation in
morphology between demographic groups. Further,
female P. m. subviridis colouration could be consid-
ered as ‘drab’in comparison with males, because
they do not have the same bright, intense orange–red
venters or blue–green heads as males. Similarly, P.
langi females were smaller in SVL and their heads
were also smaller in head width and length than
males, but not head height. Bates (2007) also found
that female P. langi were smaller in SVL than males,
but did not find sex-specific differences in coloura-
tion in preserved museum specimens. We assert that
in the wild, female P. langi are less colourful than
males. Although, the degree of sex-specific coloura-
tion differences in P. langi is lower (i.e. more subtle)
than in other crag lizards. The differences we
observed in P. langi morphology also translated into
a difference in bite force, with females having a lower
maximum bite force than males. This sexual dimor-
phism suggests there may be a territorial basis to
these species’social structure. The greater size and
bite force observed in males, may suggest males are
more territorial than females. During our observa-
tions we saw that males of both species had a ten-
dency to perch or bask on elevated locations (e.g. on
a cliffside or at the top of a rock or outcrop) with a
raised body posture (Fox et al. 2003), appearing to
be monitoring and maintaining a territory. Also,
females were more often observed in groups with
juveniles or males, and we never observed two males
in a group; suggesting a lack of tolerance for one
another. Our observations on these crag lizard’s
social organisation, spatial patterns, and sexual
dimorphism have led us to hypothesise that both
Pseudocordylus spp. have a territorial social system
and a polygynous mating system (for a similar asser-
tion for P. m. subviridis see Mouton & Van Wyk
1993), yet more behavioural observations and experi-
ments are needed to confirm this.
Although informative, our study is limited because
it was conducted at a single point in time (October
2019) and at, largely, one study site. From this, it is
not possible to determine if group-living is facultative
in either species. Our observations that crag lizards
were predominately observed alone does not discount
the possibility that these Pseudocordylus spp. could be
group-living under a different set of circumstances.
One factor that can affect social behaviour is a
species’reproductive behaviour, cycle, and mode
(Halliwell et al. 2017). For example, some lizards have
delayed dispersal when habitat is limited and risks of
infanticide are high (O’Connor & Shine 2003), which
can increase the association between parent(s) and off-
spring and set the stage for family-living. Also, Timber
Rattlesnakes (Crotalus horridus) group with relatives
while gestating their young (Clark et al.2012).Inour
study, adults were most-often observed in groups with
juveniles and occurrence of these groups may increase
during the breeding season or following birth. One of
the female P. m. subviridis we captured gave birth and
other females appeared gravid (i.e. a distended abdo-
men and high mass). So, our sampling may have over-
lapped with female parturition at this site. Although,
this does not follow the reproductive cycle described
in Flemming (1993), where P. m. melanotus (a closely
related subspecies; Bates 2007) was reported to give
birth in January (during the Austral summer), which
is also typical of most southern African lizards. Our
understanding of P. langi and P. m. subviridus sociality
would benefit from repeated surveys across seasons or
years to establish the proper timing of parturition and
its potential effect on social organisation.
Based on our findings it would appear that social
group-living is not the ancestral state of the Pseudo-
cordylus clade, as we hypothesise that the basal species
of within this genus (P. langi) has a territorial social
system and a polygynous mating system. Future
research is needed to test our hypothesis, and it could
involve (1) observing grouping behaviour of lizards
repeatedly across the year, (2) conducting experi-
ments that provide individuals with unlimited shelters
to tease apart potential drivers of grouping, like eco-
logical or social factors (as per Visagie et al. 2005;
Schutz et al. 2007), and (3) carrying out behavioural
assays in the field or lab to study the territorial
response of a resident lizard to an intruder lizard (as
per Fox & Baird 1992; Whiting 1999). The presence
of stable group-living in other cordylid lizards (Van
Wyk 1992; Mouton 2011) suggests it may either have
evolved multiple times or been lost within this clade,
but we need data on more species in order to test this
using ancestral state reconstruction. Nevertheless, we
have mapped cases of confirmed social grouping to
indicate the spread across the cordylid phylogeny
(Fig. 1) to help future researchers select taxa to exam-
ine the diversity of sociality within this group, and the
degree to which sociality is plastic or fixed.
Overall, African cordylid lizards have a number of
benefits that make them a suitable model for study-
ing sociality –a known phylogeny, overt and simple
behaviour that is easily observable in open habitats,
and a number of natural- and life-history characteris-
tics that may select for family-living (i.e. viviparity, a
K-selected life history, longevity; Whiting & While
2017). Our collection of natural history observations
doi:10.1111/aec.13030 © 2021 Ecological Society of Australia
10 J.L. RILEY ET AL.
provides insights into the social behaviour and spatial
ecology of two cordylids, P. langi and P. m. subviridis,
and we hope our study prompts further research of
the diversity and evolution of sociality within this
group.
ACKNOWLEDGEMENTS
We thank Professor Michael Cherry for logistical and
theoretical support and guidance throughout this
project, as well as Professor Le Fras Mouton for his
expertise and advice about cordylid lizards. We
would like to thank the management and staff at Wit-
sieshoek Mountain Lodge for their hospitality and
support, as well as Associate Professor John Measey
for the use of his bite force measuring equipment.
We would also like to thank three anonymous
reviewers and Associate Professor Michael Gardner
for their constructive comments on this manuscript.
CONFLICT OF INTEREST
All authors declare they have no conflict of interest.
AUTHOR CONTRIBUTION
Julia Riley: Conceptualization (lead); Formal analysis
(lead); Funding acquisition (lead); Investigation
(equal); Methodology (lead); Project administration
(lead); Writing-original draft (lead); Writing-review &
editing (equal). James Baxter-Gilbert: Conceptual-
ization (supporting); Funding acquisition (support-
ing); Investigation (equal); Methodology (supporting);
Writing-review & editing (equal). Martin Whiting:
Conceptualization (supporting); Funding acquisition
(supporting); Methodology (supporting); Writing-re-
view & editing (equal).
REFERENCES
Alexander G. & Marais J. (2007) A Guide to the Reptiles of
Southern Africa. Struik Nature, Cape Town, South Africa.
Anderson R. A., McBrayer L. D. & Herrel A. (2008) Bite
force in vertebrates: opportunities and caveats for use of a
nonpareil whole-animal performance measure. Biol. J.
Linn. Soc. 93, 709–20.
Bates M. F. (2007) An analysis of the Pseudocordylus melanotus
complex (Sauria: Cordylidae). Unpublished PhD thesis,
Stellenbosch University, Stellenbosch, Western Cape,
South Africa.
Bates M. F., Branch W. R., Bauer A. M. et al. (2014) Atlas
and Red List of the Reptiles of South Africa, Lesotho and
Swaziland. South African National Biodiversity Institute,
Cape Town.
BatesM.F.&CunninghamM.J.(2017)Pseudocordylus langi.
The IUCN Red List of Threatened Species, 2017,
e.T18514A110321779. https://dx.doi.org/10.2305/IUCN.
UK.2017-1.RLTS.T18514A110321779.en
Baxter-Gilbert J. H., Riley J. L., Fr
ere C. H. & Whiting M. J.
(2020) Shrinking into the big city: influence of genetic and
environmental factors on urban dragon lizard morphology
and performance capacity. Urban Ecosyst. https://doi.org/
10.1007/s11252-020-01065-4.
Baxter-Gilbert J. H. & Whiting M. J. (2018) Street fighters: Bite
force, injury rates, and density of urban Australian water
dragons (Intellagama lesueurii). Austral Ecol. 44, 255–64.
Brito E. S. (2017) Group living, parental care, age structure, and
genetic relatedness in Liolaemus leopardinus, a high-elevation
lizard from the Andes of Chile. Unpublished Doctoral
Dissertation, Oklahoma State University. Stillwater,
Oklahoma, USA.
Broadley D. G. (1962) The herpetofauna of the Cathedral Peak
area of the Natal Drakensberg. J. Herpetol. Assoc. Rhodesia
19, 20–2.
Bull C. M. (2000) Monogamy in lizards. Behav. Process. 51,
7–20.
Bull C. M. & Lindle C. (2002) Following trails of partners in the
monogamous lizard, Tiliqua rugosa. Acta Ethol. 5, 25–8.
Bull C. M. & Pamula Y. (1996) Sexually dimorphic head sizes
and reproductive success in the sleepy lizard Tiliqua rugosa.
J. Zool. 240, 511–21.
Byrne R. W. & Bates L. A. (2007) Sociality, evolution, and
cognition. Curr. Biol. 17, R714–R723.
Chapple D. G. (2003) Ecology, life-history, and behavior in the
Australian scincid genus Egernia, with comments on the
evolution of complex sociality in lizards. Herpetol. Monogr.
17, 145–80.
Chapple D. G. & Keogh S. J. (2006) Group structure and
stability in social aggregations of White’s skink, Egernia
whitii.Ethol. 112, 247–57.
Clark R. W., Brown W. S., Stechert R. & Greene H. W.
(2012) Cryptic sociality in rattlesnakes (Crotalus horridus)
detected by kinship analysis. Biol. Lett. 8, 523–5.
Constandius E., & Mouton P. L. F. N. (2006) Sexual size
dimorphism in montane cordylid lizards: a case study of
the dwarf crag lizard, Pseudocordylus nebulousus.African
Zoology 41, 103–12.
Davis A. R., Corl A., Surget-Groba Y. & Sinervo B. (2011)
Convergent evolution of kin-based sociality in a lizard. P.
Roy. Soc. B. 278, 1507–14.
Dennison S. (2015) Social organisation and population genetics of
the threatened great desert skink, Liopholis kintorei.
Unpublished Honours thesis. Macquarie University,
Sydney, New South Wales, Australia.
Duckett P. E., Morgan M. H. & Stow A. J. (2012) Tree-
dwelling populations of the skink Egernia striolata aggregate
in groups of close kin. Copeia 2012, 130–4.
Flemming A. F. (1993) The female reproductive cycle of the
lizard Pseudocordylus m. melanotus (Sauria: Cordylidae). J.
Herpetol. 27, 103–7.
Fox J. & Weisberg S. (2019) An R Companion to Applied
Regression, 3rd edn. Thousand Oaks, Sage, California.
Fox S. F. & Baird T. A. (1992) The dear enemy phenomenon
in the collared lizard, Crotaphytus collaris, with a cautionary
note on experimental methodology. Anim. Behavr. 44,
780–2.
Fox S. F., McCoy J. K. & Baird T. A. (2003) Lizard Social
Behavior. Johns Hopkins University Press, Baltimore,
Maryland.
© 2021 Ecological Society of Australia doi:10.1111/aec.13030
SOCIAL AND SPATIAL PATTERNS OF CRAG LIZARDS 11
Fuller S. J., Bull C. M., Murray K. & Spencer R. J. (2005)
Clustering of related individuals in a population of the
Australian lizard, Egernia frerei.Mol. Ecol. 14, 1207–13.
Gardner M. G., Bull C. M., Cooper S. J. B. & Duffield G. A.
(2001) Genetic evidence for a family structure in stable
social aggregations of the Australian lizard Egernia stokesii.
Mol. Ecol. 10, 175–83.
Hagen I. J., Herfindal I., Donnellan S. C. & Bull C. M. (2013)
Fine scale genetic structure in a population of the
Prehensile Tailed Skink, Corucia zebrata.J. Herpetol. 47,
308–13.
Halliwell B., Uller T., Holland B. R. & While G. M. (2017)
Live bearing promotes the evolution of sociality in reptiles.
Nat. Commun. 8, 2030.
Hatchwell B. & Komdeur J. (2000) Ecological constraints, life
history traits, and the evolution of cooperative breeding.
Anim. Behav. 59, 1079–86.
Kappeler P. M. (2019) A framework for studying social
complexity. Behav. Ecol. Sociobiol. 73, 13.
Kassambara A. & Mundt F. (2020) factoextra: Extract and
Visualize the Results of Multivariate Data Analyses. R
package version 1.0.7. https://CRAN.R-project.org/packa
ge=factoextra.
Lailvaux S. P., Herrel A., Vanhooydonck B., Meyers J. J. &
Irschick D. J. (2004) Performance capacity, fighting tactics
and the evolution of life–stage male morphs in the green
anole lizard (Anolis carolinensis). P. Roy. Soc. B. 271, 2501–8.
Masters C. & Shine R. (2003) Sociality in lizards: family
structure in free-living King’s Skinks Egernia kingii from
southwestern Australia. Aust. Zool. 32, 377–80.
McAlpin S., Duckett P. & Stow A. (2011) Lizards
cooperatively tunnel to construct a long-term home for
family members. PLoS One 6, e19041.
Mouton P. L. F. N.. (2011) Aggregation behaviour of lizards
in the arid western regions of South Africa. Afr. J.
Herpetol. 60, 155–70.
Mouton P. L. F. N., Gagiano C. & Sachse B. (2005)
Generation glands and sexual size dimorphism in the Cape
crag lizard, Pseudocordylus microlepidotus.African J. Herpetol.
54, 43–51.
Mouton P. L. F. N. & Van Wyk J. (1993) Sexual dimorphism
in cordylid lizards: a case study of the Drakensberg crag
lizard, Pseudocordylus melanotus.Can. J. Zool. 71, 1715–23.
O’Connor D. & Shine R. (2003) Lizards in ‘nuclear families’:a
novel reptilian social system in Egernia saxatilis (Scincidae).
Mol. Ecol. 12, 743–52.
Osterwalder K., Klingenb€
ock A. & Shine R. (2004) Field
studies on a social lizard: home range and social
organization in an Australian skink, Egernia major. Austral
Ecol. 29, 241–9.
Piza-Roca C., Strickland K., Kent N. & Frere C. H. (2019)
Presence of kin-biased social associations in a lizard with
no parental care: the eastern water dragon (Intellagama
lesueurii). Behav. Ecol. 30, 1406–15.
Pyron R. A., Burbrink F. T. & Wiens J. J. (2013) A phylogeny
and revised classification of Squamata, including 4161
species of lizards and snakes. BMC Evol. Biol. 13, 1–54.
R Core Team (2018) R: A Language and Environment for
Statistical Computing. R Foundation for Statistical
Computing, Vienna, Austria. https://www.R-project.org/.
Reissig J. (2014) Girdled Lizards and Their Relatives. Chimera,
Frankfurt, Germany.
Riley J. L., Noble D. W. A., Byrne R. W. & Whiting M. J.
(2017) Early social environment influences the behaviour
of a family-living lizard. R. Soc. Open. Sci. 4, 161082.
Rubenstein D. R. & Abbot P. (2017) The evolution of social
evolution. In: Comparative Social Evolution (eds D.
Rubenstein & P. Abbot), Cambridge University Press,
Cambridge, UK.
Schutz L., Stuart-Fox D. & Whiting M. J. (2007) Does the
lizard Platysaurus broadleyi aggregate because of social
factors? J. Herpetol. 41, 354–9.
Shea G. (1999) Morphology and natural history of the Land
Mullet Egernia major (Squamata: Scincidae). Austral. Zool.
31, 351–64.
Stanley E. L., Bauer A. M., Jackman T. R., Branch W. R.,
Mouton P. L. F. N. (2011) Between a rock and a hard
polytomy: rapid radiation in the rupicolous girdled
lizards (Squamata: Cordylidae). Mol. Phylogenet. Evol.
58, 53–70.
Stow A. J. & Sunnucks P. (2004) High mate and site fidelity in
Cunningham’s skinks (Egernia cunninghami) in natural and
fragmented habitat. Mol. Ecol. 13, 419–30.
Sumner J. (2006) Higher relatedness within groups due to
variable subadult dispersal in a rainforest skink,
Gnypetoscincus queenslandiae. Austral Ecol. 31, 441–8.
Sumner J., Moritz C. & Shine R. (1999) Shrinking forest
shrinks skink: morphological change in response to
rainforest fragmentation in the prickly forest skink
(Gnypetoscincus queenslandiae). Biol. Conserv. 91, 159–67.
Sussman R. W. & Cloninger C. R. (2011) Origins of altruism
and cooperation. Springer, New York.
Thompson M. B. (1993) Estimate of the population structure
of the eastern water dragon, Physignathus lesueurii (Reptilia:
Agamidae), along riverside habitat. Wildl. Res. 20, 613–9.
Van Wyk J. H. (1992) Life history and physiological ecology of the
lizard, Cordylus giganteus. Unpublished PhD thesis.
University of Cape Town. Cape Town, Western Cape,
South Africa.
Van Wyk J. H., & Mouton P. L. F. N. (1998) Reproduction
and sexual dimorphism in the montane viviparous lizard,
Pseudocordylus capensis (Sauria: Cordylidae). South African
J. Zool. 33, 156–65.
Visagie L., Mouton P., Le F. N. & Bauwens D. (2005)
Experimental analysis of grouping behaviour in cordylid
lizards. Herpetol. J. 15, 91–6.
Ward A. J. W. & Webster M. (2016) Sociality: the behaviour of
group-living animals. Springer International Publishing,
Switzerland.
While G. M., Chapple D. G., Gardner M. G., Uller T. &
Whiting M. J. (2015) Egernia lizards. Curr. Biol. 25, R593–
R595.
While G. M., Uller T. & Wapstra E. (2009) Within-population
variation in social strategies characterize the social and
mating system of an Australian lizard, Egernia whitii.
Austral Ecol. 34, 938–49.
White H. (1980) A heteroskedastic consistent covariance
matrix estimator and a direct test of heteroskedasticity.
Econometrica 48, 817–38.
Whiting M. J. (1999) When to be neighbourly: differential
agonistic responses in the lizard Platysaurus broadleyi.
Behav. Ecol. Sociobiol. 46, 210–4.
Whiting M. J., Nagy K. A. & Bateman P. W. (2003) Evolution
and Maintenance of Social Status Signalling Badges:
Experimental Manipulations in Lizards. In: Lizard Social
Behavior (eds S. F. Fox, J. K. McCoy & T. A. Baird) pp.
47–82.Johns Hopkins University Press, Maryland.
Whiting M. J. & While G. M. (2017) Sociality in Lizards. In:
Comparative Social Evolution (eds D. Rubenstein & P.
Abbot), Cambridge University Press, Cambridge, UK.
doi:10.1111/aec.13030 © 2021 Ecological Society of Australia
12 J.L. RILEY ET AL.
SUPPORTING INFORMATION
Additional supporting information may/can be found
online in the supporting information tab for this article.
Appendix S1. Component loadings for the
principle component analyses (PCA) for head mor-
phology of (A) Pseudocordylus m. subviridis and (B)
P. langi.
Appendix S2. A summary of grouping behaviour
that has been reported in Cordylidae and that Fig-
ure 1 is based on.
Appendix S3. A range of colouration that can be
observed in Psuedocordylus melanotus subviridis that
varies between age classes and sexes of this species.
Appendix S4. A range of colouration that can be
observed in Psuedocordylus langi that varies between
sexes of this species.
Appendix S5. Close-up view of the blue spots
along a Psuedocordylus langi individual’s back, as well
as a clear view of the males enlarged head and jaw
muscles.
Appendix S6. A male Psuedocordylus langi with his
throat extended and the black patch in view.
© 2021 Ecological Society of Australia doi:10.1111/aec.13030
SOCIAL AND SPATIAL PATTERNS OF CRAG LIZARDS 13