ArticlePDF Available

Colour polymorphism in Salamandra salamandra (Amphibia: Urodela), revealed by a lack of genetic and environmental differentiation between distinct phenotypes


Abstract and Figures

The existence of two or more distinctly coloured phenotypes among individuals of an interbreeding population is known as colour polymorphism. In amphibians, this phenomenon is pervasive among anurans, but rare or absent among salamanders and caecilians, respectively. Here, we examine whether various distinct phenotypes of Salamandra salamandra in North Spain, used as a basis to describe the subspecies S. s. bernardezi and S. s. alfredschmidti, indeed warrant separate taxonomic status or that these co-occur and belong to a single taxon. Based on a sample of 1147 individuals from 27 local populations, six phenotype classes were designated. Although two phenotypes that are attributable to S. s. alfredschmidti show some degree of geographical restriction, these co-occur with those representing typical S. s. bernardezi. A fifth phenotype class could not be unambiguously attributed to either subspecies due to an overlap in previously suggested diagnostic characteristics. Mitochondrial (cytochrome b) and nuclear (β-fibrinogen) DNA analyses revealed S. s. alfredschmidti to be nested within several subclades of S. s. bernardezi, without displaying unique lineages. Furthermore, no significant divergence was recovered by means of niche overlap analyses. As a result, we revoke the subspecies status of S. s. alfredschmidti, which should be regarded as a junior synonym of S. s. bernardezi. The current findings confirm the existence of colour polymorphism in S. salamandra and the family Salamandridae, which provides exciting possibilities for future research.
Content may be subject to copyright.
CIBIO/InBIO, Centro de Investigac
ao em Biodiversidade e Recursos Gen
eticos, Instituto de Ci^
encias Agr
arias de Vair~
Universidade do Porto, Vair~
ao Portugal;
Department of Pathology, Bacteriology and Avian Diseases, Faculty of Veterinary
Medicine, Ghent University, Merelbeke Belgium;
Departamento de Biolog
ıa de Organismos y Sistemas, Universidad de Oviedo,
Oviedo Spain;
Unidad Mixta de Investigaci
on en Biodiversidad (UMIB), CSIC-Universidad de Oviedo-Principado de Asturias,
Mieres, Spain
Colour polymorphism in Salamandra salamandra (Amphibia: Urodela),
revealed by a lack of genetic and environmental differentiation between distinct
The existence of two or more distinctly coloured phenotypes among individuals of an interbreeding population is known as colour polymorphism. In
amphibians, this phenomenon is pervasive among anurans, but rare or absent among salamanders and caecilians, respectively. Here, we examine
whether various distinct phenotypes of Salamandra salamandra in North Spain, used as a basis to describe the subspecies S. s. bernardezi and
S. s. alfredschmidti, indeed warrant separate taxonomic status or that these co-occur and belong to a single taxon. Based on a sample of 1147 individu-
als from 27 local populations, six phenotype classes were designated. Although two phenotypes that are attributable to S. s. alfredschmidti show some
degree of geographical restriction, these co-occur with those representing typical S. s. bernardezi.Afth phenotype class could not be unambiguously
attributed to either subspecies due to an overlap in previously suggested diagnostic characteristics. Mitochondrial (cytochrome b) and nuclear (b-bri-
nogen) DNA analyses revealed S. s. alfredschmidti to be nested within several subclades of S. s. bernardezi, without displaying unique lineages.
Furthermore, no signicant divergence was recovered by means of niche overlap analyses. As a result, we revoke the subspecies status of S. s. alfred-
schmidti, which should be regarded as a junior synonym of S. s. bernardezi. The current ndings conrm the existence of colour polymorphism in
S. salamandra and the family Salamandridae, which provides exciting possibilities for future research.
Key words: Colour polymorphism microgeographical variation niche taxonomy mtDNA nuDNA
Colour polymorphism describes the presence of two or more dis-
tinct phenotypes in a single interbreeding population, of which
the rarest is too frequent to simply represent the result of recur-
rent mutation (Huxley 1955). Because divergent phenotypes are
often readily observed and registered in their natural environ-
ment, polymorphic species have often been used as models to
study the fundamental processes that affect genetic variation. The
mechanisms that maintain colour polymorphism are, however,
complex, and comprise both intrinsic and extrinsic factors (e.g.
Hoffman and Blouin 2000; Bond 2007; Noonan and Comeault
2009; Richards-Zawacki et al. 2012). For instance, phenotypic
maintenance in the guppy Poecilia reticulata (Peters 1859), a
polymorphic model species, is inuenced by at least apostatic
predation, sexual selection, sensory bias and disruptive correla-
tional selection (Gray and McKinnon 2007 and references
therein). Whereas genetic colour polymorphism is ubiquitous
among birds, habitat diversity, mate choice and behaviour act as
strong determinants for the relative abundance of different phe-
notypes within populations (Roulin 2004; Roulin et al. 2004). In
general, prolonged maintenance of several phenotypes within a
single species might result in incipient speciation and eventually
the evolution of reproductive isolation (Gray and McKinnon
2007; Fisher-Reid et al. 2013).
In amphibians, colour polymorphism is generally expressed by
differences in background and eye colour, as well as dorsal
patterns (e.g. Hoffman and Blouin 2000; McKnight and Nelson
2007). Anuran species are frequently polymorphic due to which
frogs have often been used as models to study colour polymor-
phism (Hoffman and Blouin 2000; Rudh and Qvarnstr
om 2013).
Contrarily to anurans, few salamanders and none of the caecilian
species have been described to exhibit colour polymorphism
(Wells 2007; Wollenberg and Measey 2009; Petranka 2010),
although a large body of literature is available regarding the main-
tenance of several phenotypes in at least one salamander species,
Plethodon cinereus (Green 1818) (e.g. Highton 1959; Fitzpatrick
et al. 2009; Fisher-Reid et al. 2013; Venesky et al. 2015). As all
other polymorphic salamander species have proven to be plethod-
ontids (Garc
ıs et al. 2000; Petranka 2010), the phenomenon
is considered conned to the family Plethodontidae (but see e.g.
Wu et al. 2010). Indeed, previous discoveries of multiple
morphsor variantswithin non-plethodontid salamander species
have consistently been followed by taxonomic revisions, based on
subsequent evidence which revealed distinct evolutionary histories
and/or allopatric occurrence of such morphs(e.g. Nussbaum
et al. 1995; Carranza and Wade 2004; Nishikawa et al. 2013).
Nevertheless, colour polymorphism seems to occur in populations
of Salamandra salamandra (Linnaeus 1758), a member of the
Salamandridae, although only anecdotal descriptions are available
concerning the presence of this phenomenon (Eiselt 1958; Malk-
mus 1991; Barrio and Fonoll 1997; Pasmans and Keller 2000).
Salamandra salamandra comprises approximately 13 sub-
species distributed across most of Europe, although intraspecic
differentiation is most pronounced in the Iberian Peninsula
(Thiesmeier and Grossenbacher 2004). In this region, Pleistocene
climate oscillations coupled with the Iberian physiographic
heterogeneity drove cyclic patterns of range contractions and
expansions, during which allopatric divergence took place in gla-
cial refugia (Steinfartz et al. 2000; Garc
ıs et al. 2003).
These allopatric events likely resulted in the distinct phenotypes
observed across the Iberian Peninsula, which led to the
description of at least 10 subspecies in this area (Montori and
Herrero 2004; Thiesmeier and Grossenbacher 2004). As such,
S. salamandra is highly polytypic, although colour pattern
Corresponding authors: Wouter Beukema (,
Guillermo Velo-Ant
on (
Contributing authors: Alfredo G. Nicieza (, Andr
Lourencßo (
Accepted on 7 November 2015
©2016 Blackwell Verlag GmbH J Zool Syst Evol Res doi: 10.1111/jzs.12119
variation is generally continuous at local geographical scales
(Boulenger 1911) due to which cases of polymorphism have not
been explicitly acknowledged (but see e.g. Eiselt 1958; Malkmus
1991). Nevertheless, syntopy of up to four diagnosable pheno-
types varying in background colour and presence/absence of
colour pattern has been reported from several populations of
Salamandra salamandra bernardezi in northern Spain
(Villanueva 1993; Barrio and Fonoll 1997; Pasmans et al. 2004;
Beukema 2006). It remains ambiguous whether the observed
variation comprises colour polymorphism (Pasmans and Keller
2000), or actually indicates the presence of multiple taxa (K
and Steinfartz 2006). Indeed, despite the presence of typical
S. s. bernardezi individuals (sensu Wolterstorff 1928), Salaman-
dra salamandra alfredschmidti was described based on
differences in their colour patterns and mitochondrial D-loop
sequences in respect to a highly restricted sample of other
S. salamandra subspecies (K
ohler and Steinfartz 2006). Cur-
rently, S. s. alfredschmidti is considered to occupy a small
enclave (Tendi and Marea valleys, central Asturias, northern
Spain) within the distribution of S. s. bernardezi.
Here, we combine comprehensive phenotypic data, mitochon-
drial and nuclear DNA analyses and niche overlap tests to explore
whether S. s. alfredschmidti and S. s. bernardezi indeed show
geographical and evolutionary divergence or whether both taxa
conform a single taxon characterized by colour polymorphism.
These analyses are coupled with a comprehensive geographical
coverage of both S. s. alfredschmidti and S. s. bernardezi distri-
butions. We ask specically whether (1) both subspecies are well
diagnosable from each other and show geographical separation,
(2) genetic divergence can be identied between them and (3)
niche divergence might underlie diversication.
Materials and Methods
A total of 95 locations, mostly across the northern Iberian Peninsula,
were visited in order to either collect distribution data of different pheno-
types, to gather tissue samples for genetic analyses or to assemble a data
set of geographical occurrence records to calibrate niches of S. s. alfred-
schmidti and S. s. bernardezi. Accordingly, phenotypic data were gath-
ered at 27 of these 95 sites, tissue samples of several S. salamandra
subspecies were collected at 34 of the 95 locations (Table 1, Fig. 1a, b),
and geographical coordinates for 85 of the 95 sites (the remaining 10
coordinates fell within the same grid cells as already selected coordinates)
were used for niche calibration. For a comprehensive overview summa-
rizing all these data, see Fig. 1 and Table 1 and S1.
Distribution assessment and phenotype delimitation
Dorsal photographs were taken to record background colour and pattern
of all encountered salamanders from 27 sites spread across central and
eastern Asturias, comprising the entire known distribution of S. s. alfred-
schmidti and the eastern half of the S. s. bernardezi distribution (Fig. 1c).
Despite the fact that four phenotype groups were proposed for S. sala-
mandra in this region by Barrio and Fonoll (1997) and Pasmans and Kel-
ler (2000), we erected six phenotype groups to deal with all variation
documented by dorsal photographs (Table 2; Fig. 2). As phenotype group
6 was erected to contain individuals not assignable to any other category,
we did not provide a graphical example of this group in Fig. 2 (see also
below). For subsequent genetic and niche overlap analyses, phenotype
groups 1 and 2 were regarded to represent typical S. s. bernardezi fol-
lowing Wolterstorff (1928). As individuals from phenotype group 3 could
be attributed to S. s. bernardezi as well as S. s. alfredschmidti due to
overlapping diagnostic characteristics (Wolterstorff 1928; K
ohler and Ste-
infartz 2006), we regarded this group as an intermediate class which was
not ascribed to either subspecies. Phenotype groups 4 and 5 were classi-
ed as S. s. alfredschmidti following K
ohler and Steinfartz (2006).
Phenotype group 6 was erected to include individuals that were not attri-
butable to any other of the phenotype groups.
DNA extraction and amplication
Tissue samples of the subspecies S. s. alfredschmidti/S. s. bernardezi (40),
S. s. bejarae/S. s. gallaica (9) and S. s. longirostris (1) were collected in
the eld at 34 sites across Spain and Portugal (Table 1, S1). The latter two
were used in the analyses as outgroups. As no specimens were collected,
reference material in the form of tissue samples was deposited in the per-
sonal collection of GVA. Individuals belonging to phenotype groups 1 and
2 were classied as S. s. bernardezi in all genetic analyses, while those
belonging to groups 4 and 5 were ascribed to S. s. alfredschmidti. Pheno-
typic assignment of these individuals was made based on their colour pat-
terns. Genomic DNA was extracted from fresh tissue samples using
Genomic DNA Tissue Kit (EasySpin), following the protocol of the manu-
facturer. Quantity and quality of DNA extract products were assessed on a
0.8% agarose gel. A section of ca. 1400 bp of the mitochondrial genome
including the complete cytochrome b gene b (cyt b) and a fragment of ca.
700 base pairs of the intron of the nuclear gene b-brinogen (bFib) were
amplied and sequenced for each sample. The cyt b fragment was ampli-
ed using primers Glu14100L (forward, 50GAA AAA CCA AYG TTG
TAT TCA ACT ATA A 30) and Pro15500H (reverse, 50AGA ATT YTG
GCT TTG GGT GCCA 30) (Zhang et al. 2008), while the bFib gene was
amplied using BFIB_F (forward, 50TGG GAC TGG CAG TTG TTT AG
30) and BFIB_R (reverse, 50TGA TTC ACG AGT TTG TTG CTC 30)
(Pereira et al. unpublished). The alignment of cyt b was trimmed to avoid
missing data and resulted in a nal alignment of ca. 1100 bp. Each poly-
merase chain reaction (PCR) had a total volume of 1011 ll: 5 llof
MyTaqTM HS Mix 2X (Bioline), 3 ll of distilled H
O, 0.5 ll of each pri-
mer from a primer solution of 10 lM and 12ll of DNA extract (~50 ng/
ll). A negative control was employed to identify possible contaminations.
For cyt b gene, cycling conditions were as follows: initial denaturation at
94°C for 5 min, followed by 40 cycles of 40 s at 94°C, 40 s of annealing
at 51°C, 72°C for 2 min 30 s, ending with a nal extension of 5 min at
72°C. PCR conditions for bFib gene were as follows: initial denaturation at
94°C for 5 min, followed by 40 cycles of 30 s at 94°C, 30 s of primer
annealing at 59°C, elongation at 72°C for 45 s, nishing with a nal exten-
sion of 5 min at 72°C. PCR product quality and quantity was assessed by
visual inspection in a 2% agarose gel. Sequencing of PCR products was
outsourced to Macrogen Inc. (Amsterdam, Netherlands) and Beckman
Coulter Inc. (Grenoble, France). The same primers used in PCR were
employed for sequencing, except for cyt b, where instead of Pro15500H we
used an internal forward primer (available upon request). All the obtained
chromatograms were veried, aligned and corrected by eye using GENEIOUS
PRO v4.8.5 (
Phylogenetic analyses
Phylogenetic relationships were analysed using Bayesian analyses con-
ducted in BEAST v1.7.5 (Drummond et al. 2012). JMODELTEST v.2.1.4 (Dar-
riba et al. 2012) was used to test for the best tting model of nucleotide
substitution, under Bayesian information criteria correction (BIC;
HKY+G). A lognormal relaxed clock and a coalescence constant size
model were used as tree priors. Markov chain Monte Carlo (MCMC) anal-
yses were run in three independent runs of 100 million generations, with
a sampling frequency of 1000 generations and discarding 25% trees as
burn-in. Parameter convergence was veried by examining the effective
sample sizes (ESSs) using TRACER v1.6 and used the remaining trees to
obtain the subsequent maximum clade credibility summary tree with pos-
terior probabilities for each node using TREEANNOTATOR v1.7.5 (distributed
with the BEAST package). Phylogenetic relationships at nuclear bFib
were analysed using a haplotype network. Heterozygous sequences within
the nuclear bFib fragment were phased using the PHASE algorithm as
implemented in DNASP 5 (Librado and Rozas 2009). Phase probabilities
parameter was set at 0.7, and all other settings were set by default. TCS
v.1.21 (Clement et al. 2000) was used to construct the haplotype network
and applying default settings (probability of parsimony cut-off: 95%).
Niche overlap
Bioclimatic data at a 300resolution consisting of 19 temperature- and pre-
cipitation-related parameters (Hijmans et al. 2005) were downloaded from
the WORLDCLIM database ( These parameters were
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
clipped in ARCGIS 10.1 using a rectangular area comprising the distribution
of both S. s. alfredschmidti and S. s. bernardezi, ranging between N
43.9642.69 to W 8.40 4.44 (Fig. 1b). In order to calibrate niches,
these parameters were combined with a data set of georeferenced occur-
rences gathered during eld visits between 2004 and 2014, which were
supplemented with literature sources and additional personal observations.
Ten populations with the presence of individuals attributable to phenotype
groups 4 and 5 (Fig. 1c, Table S1) were used in combination with two lit-
erature records (Villanueva 1993; Pasmans and Keller 2000), one personal
record (David Buckley) and 17 personal records of GVA and AN to con-
struct a nal database of 28 S. s. alfredschmidti occurrences (Table S1).
For S. s. bernardezi, occurrence data for 17 populations without the pres-
ence of individuals from phenotype groups 4 and 5 were gathered
(Fig. 1c), in addition to 40 personal observations of GVA, AN and WB
resulting in a total of 57 occurrences (Table S1). All records used for
niche calibration were characterized by a resolution of at least 300. Niches
were calibrated and overlap was measured on a 2D representation of
environmental space in R 2.14.1, represented by the rst two axes of a
principal component analysis (PCA) using the PCA-env ordination
approach presented by Broennimann et al. (2012). The PCA was used to
summarize information contained in the climatic parameters. Environmen-
tal space thereby consists of a grid of r9rcells, with a standard resolu-
tion of 100. Based on occurrence records, smoothed occurrence densities
were constructed for each entity using a Gaussian kernel (standard band-
width) density function applied to all cells. PCA-env selects orthogonal
and linear combinations of environmental parameters explaining as much
variance as possible, which were subsequently summarized into the rst
two components. Hence, different positions of speciessmoothed occur-
rence densities represent dissimilar occupation of environmental space.
PCA-env is calibrated on the entire environmental space of the study area
rather than using only climatic values corresponding to occurrence
records. Subsequently, actual niche overlap was calculated based on over-
lap of the occurrence densities using the Dmetric of Schoener (1968),
which varies between 0 (no overlap) and 1 (complete overlap).
We used two tests to measure the degree of niche overlap between
S. s. alfredschmidti and S. s. bernardezi (Warren et al. 2008). First, the
Table 1. Salamandra salamandra tissue samples used in this study, including subspecies identity, locality data, geographical coordinates, voucher
codes and GenBank accession numbers for each locus sequenced. See Fig. 1a, b for a graphical overview of the visited locations.
Subspecies Locality Lat Long Voucher cyt b bFib
S. s. bernardezi Villaviciosa 43.44 5.49 GVA3794 KT799695 KT799741, KT799742
S. s. bernardezi Pesanca 43.26 5.33 GVA4601 KT799711 ND
S. s. bernardezi Ma~
nangas 43.39 4.80 GVA4675 KT799712 KT799761, KT799762
S. s. bernardezi Pimiango 43.39 4.53 GVA4687 KT799713 KT799763, KT799764
S. s. bernardezi Buferrera 43.27 4.98 GVA4692 KT799714 KT799765, KT799766
S. s. bernardezi FaR
ıo 43.43 5.57 GVA4707 KT799716 ND
S. s. bernardezi Tuiza 43.02 5.91 GVA3532 KT799686 KT799733, KT799734
S. s. bernardezi Somiedo 43.09 6.20 GVA1921 KT799684 ND
S. s. bernardezi Somiedo 43.15 5.98 GVA3626 KT799688 ND
S. s. bernardezi Mondo~
nedo 43.39 7.30 GVA3673 KT799689 KT799735, KT799736
S. s. bernardezi Soto 43.54 6.06 GVA3711 KT799690 ND
S. s. bernardezi Oviedo 43.35 5.84 GVA3755 KT799693 ND
S. s. bernardezi Oviedo 43.35 5.84 GVA3770 KT799694 KT799739, KT799740
S. s. bernardezi Oviedo 43.35 5.84 GVA3954 KT799701 KT799747, KT799748
S. s. bernardezi Oviedo 43.35 5.84 GVA3719 KT799691 KT799737, KT799738
S. s. bernardezi Oviedo 43.35 5.84 GVA3724 KT799692 ND
S. s. bernardezi Oviedo 43.35 5.84 GVA3976 KT799702 ND
S. s. bernardezi Oviedo 43.35 5.84 GVA3992 KT799703 KT799749, KT799750
S. s. bernardezi Salas 43.40 6.20 GVA3836 KT799699 ND
S. s. bernardezi Serra do Xistral 43.42 7.52 GVA3873 KT799700 ND
S. s. bernardezi La Cuevona de Cuevas 43.43 5.07 GVA5102 KT799728 ND
S. s. bernardezi Restriello 43.29 6.18 GVA4988 KT799726 ND
S. s. bernardezi N Borin
es 43.40 5.31 GVA4971 KT799696 ND
S. s. bernardezi/S. s. alfredschmidti P
uron 43.37 4.69 GVA4053 KT799708 KT799759, KT799760
S. s. bernardezi/S. s. alfredschmidti Llerandi 43.31 5.21 GVA4697 KT799715 KT799767, KT799768
S. s. bernardezi/S. s. alfredschmidti Sueve 43.44 5.19 GVA4722 KT799717 KT799769, KT799770
S. s. bernardezi/S. s. alfredschmidti R
ıo Tendi 3 43.20 5.15 GVA3546 KT799687 ND
S. s. bernardezi/S. s. alfredschmidti Caldevilla 43.34 -5.23 GVA3816 KT799697 ND
S. s. bernardezi/S. s. alfredschmidti Caldevilla 43.34 5.23 GVA3815 KT799696 KT799743, KT799744
S. s. bernardezi/S. s. alfredschmidti La Roza 43.30 5.24 GVA3822 KT799698 KT799745, KT799746
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 3 43.32 5.39 GVA4944 KT799718 ND
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 4 43.30 5.40 GVA4947 KT799719 ND
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 5 43.29 5.42 GVA4950 KT799720 KT799771, KT799772
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 5 43.29 5.42 GVA4951 KT799721 KT799773, KT799774
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 6 43.27 5.41 GVA4958 KT799722 ND
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 6 43.27 5.41 GVA4959 KT799723 ND
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 6 43.27 5.41 GVA4968 KT799724 ND
S. s. bernardezi/S. s. alfredschmidti R
ıo Marea 6 43.27 5.41 GVA4969 KT799725 KT799775, KT799776
S. s. bernardezi/S. s. alfredschmidti Nava 43.34 5.48 GVA4051 KT799706 KT799755, KT799756
S. s. bernardezi/S. s. alfredschmidti Nava 43.34 5.48 GVA4052 KT799707 KT799757, KT799758
S. s. gallaica/S. s. bejarae Bo~
nar 42.86 5.29 GVA4114 KT799709 ND
S. s. gallaica/S. s. bejarae Abadim 41.56 7.98 GVA1848 KT799681 ND
S. s. gallaica/S. s. bejarae Cubillos del Sil 42.59 6.56 GVA5001 KT799727 ND
S. s. gallaica/S. s. bejarae Mindelo 41.32 8.72 GVA1863 KT799682 KT799729, KT799730
S. s. gallaica/S. s. bejarae Mirandela 41.52 7.18 GVA4003 KT799705 KT799753, KT799754
S. s. gallaica/S. s. bejarae Mux
ıa 43.10 9.12 GVA1009 KT799680 ND
S. s. gallaica/S. s. bejarae Viana do Castelo 41.71 8.81 GVA1891 KT799683 ND
S. s. gallaica/S. s. bejarae Sedano 42.68 3.73 GVA4248 KT799710 ND
S. s. gallaica/S. s. bejarae Nuez 41.77 6.51 GVA4000 KT799704 KT799751, KT799752
S. s. longirostris San Pablo de Buceite 36.49 5.40 GVA1948 KT799685 KT799731, KT799732
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
Colour polymorphism in Salamandra salamandra 3
identity or equivalency test assesses whether niches of two taxa are iden-
tical; the occurrences of both subspecies were pooled, two random sets
of occurrences with the same original sample sizes were extracted, and
the overlap scores were determined. This procedure was repeated 100
times in order to create a null distribution of overlap scores, which was
compared to the actual overlap. When the actual overlap value falls
beyond 95% of the simulated values, the hypothesis of niche identity is
rejected. Second, the background or similarity was used to assess whether
niches of the two subspecies are more similar than expected by chance
based on the geographical regions (environmental background) in which
they occur (as opposed to solely the actual occurrence points used in the
rst test). Again, 100 randomizations were created by placing the kernel
density of occurrences at random within the background of entity A,
which was compared to the background of entity B and vice versa. When
the actual overlap value is signicantly (p <0.05) higher or lower than
expected from the null distribution based on a two-tailed test, the null
hypothesis that the two entities are not more similar to each other can be
A total of 1147 individuals from 27 local populations (42
individuals per population) were assigned to the six phenotypic
groups. The relative occurrence of these groups decreased gradu-
ally (Fig. 3), with individuals assigned to group 1 being the most
common (n=466) and those belonging to group 6 the fewest in
number (n=22). Individuals assigned to groups 1, 2 and 3
occur throughout the studied area, although the relative occur-
rence of the former two groups decreases greatly in the area
formed by the mid-altitude valleys from the central Asturian
Basin (between Pe~
namayor Mountains and the River Sella),
while that of groups 45 increases (Fig. 1c). Individuals belong-
ing to groups 4 and 5 were found in 10 of the 27 visited popula-
tions, ranging from Urbi
es in the east, towards Sueve in the
north and Pur
on in the west (Fig. 1c). All populations but one
ıo Marea 2) shows co-occurrence of individuals from groups 1
and 2 and those from groups 4 and 5. The relatively high occur-
rence of group 6 in the area south of Llanes is associated with
the presence of individuals characterized by highly restricted and
occasionally irregular yellow markings.
Phylogenetic analyses
Bayesian analyses of mtDNA sequences showed S. s. lon-
girostris as sister to a well-supported clade (BPP =0.98) that
includes all the remaining studied subspecies (Fig. 4). This
main clade is divided into two subclades: a well-supported
subclade that includes all S. s. gallaica and S. s. bejarae
samples and a moderately supported clade comprising all
S. s. bernardezi and S. s. alfredschmidti samples that is further
Fig. 1. (a) Range of Salamandra salamandra on the Iberian Peninsula (light grey grid) including the distribution of S. s. bernardezi (hatched) and
hitherto assumed occurrence of S. s. alfredschmidti in black; localities at which tissue samples were collected are indicated by dark grey squares. (b)
Detail comprising the ranges of the aforementioned taxa, tissue sample locations and occurrences (white dots) used for niche modelling. (c) Distribu-
tion of phenotype groups in central and eastern Asturias. Population numbers correspond to those in Table S1.
Table 2. Background colour and pattern of the Salamandra salamandra phenotypic groups native to central and eastern Asturias, Spain
1 23456
Yellow Yellow Yellow Yellow/light brown Brown Yellow or black
Pattern Broad and continuous
dorsal and lateral
black stripes
Thin and
discontinuous dorsal
and lateral black
Either only dorsal
black stripe or also
with vestigial lateral
Yellow or orange
coloured head
region and tiny
irregular lighter
ecks covering the
Lateral and/or
dorsal black stripes
and occasionally
lighter parotoids
Not corresponding
to groups 15
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
subdivided into ve groups. Samples of individuals attributed
to groups 4 and 5, and ascribed to S. s. alfredschmidti during
analyses, are intermixed with S. s. bernardezi individuals of
groups 1 and 2 in two of the ve monophyletic sublineages
instead of constituting a monophyletic clade (Fig. 4). Four
haplogroups were identied in the nuclear haplotype network
(Fig. 4). While S. s. longirostris,S. s. bejarae and S. s. gal-
laica group together, S. s. bernardezi and S. s. alfredschmidti
are intermixed in two of the three remaining haplogroups
(Fig. 4).
(a) (b) (c)
(d) (e) (f)
(g) (h) (i)
(j) (k) (l)
(m) (n) (o)
Fig. 2. Phenotypes of Salamandra salamandra in northern Spain, in series of three. Photographs by WB unless stated otherwise. Phenotype 1: Moreda,
central Asturias (a), R
ıo Tendi, eastern Asturias (Mario Riedling, b), La Caridad, western Asturias (c). Phenotype 2: La Caridad, western Asturias (d),
on, eastern Asturias (e and f). Phenotype 3: R
ıo Marea (Philip Gerhardt, g), Urbi
es, central Asturias (h), R
ıoInerno, central Asturias (i). Phenotype
4: R
ıo Tendi, eastern Asturias (Serg
e Bogaerts, j), R
ıo Tendi, eastern Asturias (Frank Deschandol, k), Urbi
es, central Asturias (l). Phenotype 5: R
Tendi, eastern Asturias (Frank Deschandol, m), R
ıo Tendi, eastern Asturias (Sebastian Voitel, n), R
ıo Tendi, eastern Asturias (Mario Riedling, o)
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
Colour polymorphism in Salamandra salamandra 5
Niche overlap
The niches of individuals from groups 1 and 2 (S. s. bernardezi)
and that of groups 4 and 5 (S. s. alfredschmidti) occupy similar
positions in environmental space (Fig. 5a, b, respectively). While
the niche of S. s. bernardezi is clearly wider, the actual overlap
with S. s. alfredschmidti was moderate (D=0.25). The equiva-
lency test showed that both niches were not identical as the
actual overlap fell beyond the null distribution, leading to a p
value of 1 (Fig. 5c). The actual overlap did fall within the null
distributions created during the similarity tests, both when com-
paring the S. s. bernardezi niche to that of S. s. alfredschmidti
and vice versa (Fig. 5d, e). In both cases, the actual overlap was
therefore not found to be signicantly higher or lower than
expected in respect to these null distributions (Fig. 5d, e). Visual
inspection of the niches in environmental space revealed that
the niche of S. s. alfredschmidti represents a subset of the
S. s. bernardezi niche.
Through the combination of phenotypic, genetic and environ-
mental data, we found that individuals of S. s. alfredschmidti
and S. s. bernardezi are not signicantly differentiated. In addi-
tion, we conrmed the usefulness of integrative analyses to
tackle taxonomical issues (Haig and Winker 2010; Wielstra et al.
2012; Torstrom et al. 2014).
Phenotypic and subspecic diagnosability
Although we presented evidence on the occurrence of at least
six diagnosable phenotypic groups in central and eastern Astur-
ias (see also Barrio and Fonoll 1997; Pasmans and Keller
2000), we acknowledge that phenotypic attribution of individu-
als in some cases can be problematic; especially, groups 13
seem to display a continuum regarding the extent of their dor-
sal and lateral black stripes, which motivated Pasmans and Kel-
ler (2000) to treat these as subgroups of a single phenotypic
class. It should be pointed out that the extent of dorsal and lat-
eral stripes among individuals of S. s. bernardezi might change
during ontogeny (Pasmans and Keller 2000; Bogaerts 2002),
although this point has never been studied comprehensively
(see also Beukema et al. 2009; Beukema 2011). In contrast,
phenotypes 4 and 5 show clearly distinct colours and patterns
due to the, respectively, lack of a striped pattern, occasionally
in addition to a yellow or orange head region and small light
ecks all over the body (phenotype 4) and a brown instead of
yellow background coloration with dorsal and sometimes lateral
black stripes as well as lighter parotoids (phenotype 5). Up to
three highly distinct phenotypes therefore seem to occur within
Asturian populations of S. salamandra, which in turn show a
considerable variation.
While the differentiation between the vast majority of pheno-
types analysed herein was unambiguous, an initial attempt to
attribute these to distinct subspecies was far from being clear-cut
as diagnostic traits between S. s. alfredschmidti and
S. s. bernardezi overlap. Wolterstorff (1928) described
S. s. bernardezi based on a sample of 22 individuals, which were
different in their hue of yellow background colour, and the
extent and demarcation of dorsal and lateral black stripes. In
turn, K
ohler and Steinfartz (2006) included individuals character-
ized by dirty to greyish-yellow coloration, absent lateral stripes
and overall irregularly demarked stripes in the diagnosis of
S. s. alfredschmidti, under the rationale that S. s. bernardezi
showed sharply delimited dorsal and dorsolateral stripes. How-
ever, individuals displaying both sharply and irregularly demar-
cated black stripes, those lacking dorsolateral stripes and those
displaying various hues of yellow background colour occur
throughout the populations of S. s. bernardezi analysed herein
(Fig. 1c). Moreover, although at least groups 4 and 5 do com-
prise phenotypes that are highly distinct from classical
S. s. bernardezi, these are nearly without exception intermixed
with individuals of phenotype groups 1, 2 and 3 (Table S1; Pas-
mans et al. 2004). A geographical basis for the occurrence of
S. s. alfredschmidti therefore seems to be lacking even when
restricting this subspecies to the former two phenotype groups.
Polytypism versus polymorphism
Perhaps the most remarkable nding of the current study com-
prises the fact that individuals attributed to S. s. alfredschmidti
solely on the basis of colour phenotypes do not represent a
monophyletic unit. Rather, these were found to be interspersed
within several subclades of S. s. bernardezi according to both
mitochondrial and nuclear data. Similarly, niche divergence
between these subspecies is absent as the niche of
S. s. bernardezi completely overlaps that of S. s. alfredschmidti
in environmental space, while their niche centres (assumed to
correspond to the environmental optimum; Austin 1985) closely
match. These data do not suggest that environmental variation
currently maintains divergence, which is not unexpected as
S. salamandra occurs continuously from sea level up to at least
2000 m in the Cantabrian Mountains (Mart
ınez-Rica and Rein
nales 1988). We therefore do not see sufcient grounds to
acknowledge S. s. alfredschmidti as a separate subspecies (see
also below), and regard the existence of several discrete pheno-
types in S. s. bernardezi as a classic case of colour polymor-
phism. In other words, S. salamandra is both polytypic as the
species encompasses a high number of well-diverged subspecies
(Montori and Herrero 2004; Thiesmeier and Grossenbacher
2004), but shows geographically restricted colour polymorphism
as well. To the best of our knowledge, this is the rst explicitly
conrmed case of this phenomenon in the family Salamandridae.
The occurrence of both polytypism and polymorphism in a
single species is rare; however, these two phenomena are not
mutually exclusive. Colour polymorphism is associated with
accelerated speciation rates, due to which an initially polymor-
phic species can end up as polytypic when phenotypes diverge
and receive taxonomic recognition (Gray and McKinnon 2007;
Fig. 3. Number of recorded Salamandra salamandra individuals belong-
ing to each phenotypic group
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
Hugall and Stuart-Fox 2012; Fisher-Reid et al. 2013). Among
salamanders, polytypism is relatively more common than poly-
morphism (e.g. Petranka 2010). Occurrence of the former among
at least several temperate salamander species can be explained
by processes of isolation and subsequent recolonization during
the Pleistocene glacial cycles, when currently recognized sub-
species diverged in local (micro)refugia for varying amounts of
time. Accordingly, species of the genera Ensatina,Lissotriton
and Salamandra encompass large numbers of recently derived
subspecies which form broad secondary contact zones (Steinfartz
et al. 2000; Garc
ıs et al. 2003; Pereira and Wake 2009;
Vences et al. 2014; Pabijan et al. 2015). Development of consid-
erable morphological divergence during short periods of isolation
(leading to subspecic recognition) is, however, exception rather
than rule among salamanders, as most species remain morpho-
logically conserved despite possessing high levels of intraspecic
genetic structure (but see Wake et al. 1983; Arntzen et al. 2015).
This situation seems to hold truth for S. s. bernardezi, which dis-
plays an overall conserved morphology despite its considerable
genetic heterogeneity (current results; see also Garc
ıs et al.
2003; Velo-Ant
on et al. 2007). Nevertheless, colour polymor-
phism did also evolve in this subspecies, although the drivers
that lead to this situation remain unknown. Specically, we
currently cannot infer whether colour polymorphism arose within
populations independently or that the present situation is the
result of complete intermixing between classicalS. s. bernar-
dezi and individuals attributed to phenotypes 4 and 5. In case of
the latter scenario, the area characterized by high occurrence of
phenotype groups 4 and 5 located between Urbi
es, Pe~
and the River Sella (Fig. 1) might have functioned as a former
microrefugium, after which individuals dispersed eastwards
ıs et al. 2003). As the overall lack of genetic diver-
gence and occurrence of these phenotypes in several subclades
of S. s. bernardezi, however, does not support such a hypothesis,
we stress the need for increased sampling and more elaborate
molecular analyses to shed light on the origin of colour polymor-
phism in S. s. bernardezi. It should additionally be noted that
Boulenger (1911), Eiselt (1958) and Malkmus (1991) gave anec-
dotal descriptions regarding the presence of multiple phenotypes
within populations of S. s. gallaica (e.g. through the presence of
both striped and spotted dorsal patterns).
Maintenance of colour polymorphism
Colour polymorphism within salamander populations is known
to have a genetic basis (Highton 1959), although at least assor-
tative mating (Acord et al. 2013), apostatic predation (Fitz-
patrick et al. 2009), the chytrid Batrachochytrium dendrobatidis
Fig. 4. Genetic relationships between S. s. bernardezi (blue) and S. s. alfredschmidti (red) and the outgroups used in this study (S. s. gallaica/bejarae
in green, and S. s. longirostris) displayed by a bFib haplotype network inferred by TCS under the 95% criterion showing four haplogroups, two of
which show shared haplotypes between S. s. bernardezi and S. s. alfredschmidti (a). The size of each haplotype symbol is proportional to its fre-
quency, and lines represent mutational steps separating observed haplotypes. Also shown is a Bayesian consensus phylogram based on mtDNA data
cytochrome b. Posterior probability values are shown below each node (b). Colours are concordant with the nuclear haplotype network (a). Asterisks
denote individuals sequenced for bFib.
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
Colour polymorphism in Salamandra salamandra 7
(Venesky et al. 2015) and possibly climate (Fisher-Reid et al.
2013) play signicant roles in maintaining different phenotypes.
Moreover, alleles coding for striped colour patterns are domi-
nant over unstriped patterns among amphibians in general, mak-
ing uniform phenotypes among polymorphic populations
generally less abundant (ONeill and Beard 2010). In
S. s. bernardezi, the latter factor could perhaps explain the rela-
tively low occurrence of phenotype group 4, although pheno-
type maintenance in this subspecies is undoubtedly much more
complex. Populations of S. s. bernardezi are, in contrast to
nearly all other populations of S. salamandra, characterized by
pueriparous reproduction (Velo-Ant
on et al. 2015). As such,
Pasmans and Keller (2000) suggested that the transition from
larviparous to pueriparous reproduction might be associated
with the decreased surface activity and the loss of aposematic
colours, leading to a darker background colour or loss of pat-
tern. Pueriparity has indeed been associated with the loss of
yellow coloration in several Salamandra taxa, although it
remains to be investigated whether these traits show correla-
tional selection or whether these play a role in maintaining phe-
notypes of S. s. bernardezi. On the other hand, S. s. bernardezi
is assumed to display pueriparity throughout most of its
distribution, generally without showing a decrease in apose-
matic coloration. It seems likely that at least apostatic predation
could have a signicant effect in maintaining different pheno-
types as well, due to the fact that largely yellow individuals
starkly contrast with those of, for example, phenotype group 5.
However, at this point, we can merely speculate on the neces-
sarily complex factors that uphold colour polymorphism in this
subspecies. Future experimental trials and increased eld
research are needed to assess the differences in survival
between the herein established phenotypes.
Taxonomical implications
Populations of S. s. alfredschmidti do not represent a distinct
geographical or genetic unit, as these are without exception inter-
spersed (phenotypically, genetically, ecologically and geographi-
cally, in various degrees) by S. s. bernardezi. Recognizing
S. s. alfredschmidti as subspecies renders S. s. bernardezi para-
phyletic and impairs subspecic diagnosability. Consequently,
we explicitly reject subspecies status for S. s. alfredschmidti and
regard this taxon as a junior synonym of S. s. bernardezi. In this
decision, we took the long-standing notion that subspecies should
be erected for the sake of convenience into account (Mayr 1982;
Fitzpatrick 2010). Intraspecic taxonomy of S. salamandra is
highly confused, especially within the Iberian Peninsula (Eiselt
1958; Speybroeck et al. 2010), due to which there is an obvious
need to move towards a comprehensive systematic revision. Tax-
onomic rearrangements like the current work are an essential part
of this process and will hopefully provide a basis and shift focus
towards eco-evolutionary studies aiming to explore the exciting
high degree of phenotypic variation observed in S. salamandra.
We thank D. Donaire-Barroso, J. Beukema, S. Bogaerts, R. Fonoll,
F. Pasmans and A. Villanueva for providing literature, discussions and
sharing eld observations during the past decade. Special thanks to
S. Bogaerts, F. Deschandol, P. Gerhardt, M. Riedling and S. Voitel for
Fig. 5. Niches of S. s. bernardezi (a) and S. s. alfredschmidti (b; dark shading) in 2D environmental space, composed of the two-rst axes of a princi-
pal component analysis summarizing information of bioclimatic parameters. The solid and dashed contour lines illustrate 100% and 50% of the back-
ground environment. Panels ce show histograms displaying the null distributions consisting of 100 randomizations (grey bars) in respect to the actual
niche overlap (D=0.25; red arrow). From left to right, the histograms show tests of niche equivalency (c), niche similarity of S. s. bernardezi to
S. s. alfredschmidti (d) and vice versa (e). Signicance of the tests is shown below.
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
their photographic contributions and to D. Buckley for constructive com-
ments on the manuscript. N. Sillero provided spatial distribution data of
S. salamandra on the Iberian Peninsula. Unpublished primers were devel-
oped by R.J. Pereira supported by the European Science Foundation
(Frontiers of Speciation Research, Exchange Grant 3318). This work is
funded by FEDER funds through the Operational Programme for Com-
petitiveness Factors COMPETE and by National Funds through FCT
Foundation for Science and Technology under the PTDC/BIA-EVF/
3036/2012 and FCOMP-01-0124-FEDER-028325 (to GVA) and
through Ministerio de Econom
ıa y Competitividad (Spanish Govern-
ment): MINECO/CGL2012-40246-C02-02 (to AGN). GVA and AL are
supported by FCT (IF/01425/2014 and PD/BD/106060/2015, respec-
tively). All procedures performed in studies involving animals were in
accordance with the ethical standards of the institution or practice at
which the studies were conducted.
Acord MA, Anthony CD, Hickerson C-AM (2013) Assortative mating in
a polymorphic salamander. Copeia 4:676683.
Arntzen JW, Beukema W, Galis F, Ivanovi
c A (2015) Vertebral number
is highly evolvable in salamanders and newts (family Salamandridae)
and variably associated with climatic parameters. Contrib Zool 84:85
Barrio CL, Fonoll R (1997) Sobre una poblaci
on de salamandras
Salamandra salamandra con pigmentaci
on an
omala. Bol Assoc
Herpetol Esp 8:3336.
Beukema W (2006) Filling in the gap in the distribution of Salamandra
salamandra alfredschmidti K
ohler & Steinfartz 2006, and remarks on
the reproduction of the Rio Tendi Valley salamanders in Asturias,
Spain. Amphibia 5:2023.
Beukema W (2011) Ontogenetic pattern change in amphibians: the case
of Salamandra corsica. Acta Herpetol 6:169174.
Beukema W, de Pous P, Brakels P (2009) Remarks on the morphology
and distribution of Lyciasalamandra luschani nikensis with the
discovery of a new isolated population. Zeitschr Feldherp 16:115126.
Bogaerts S (2002) Farbkleidentwicklung bei einigen Feuersalamandern.
Amphibia 1:410.
Bond AB (2007) The evolution of color polymorphism: crypticity,
searching images, and apostatic selection. Annu Rev Ecol Evol Syst
Boulenger EG (1911) A contribution to the study of the variations of the
spotted salamander (Salamandra maculosa). Proc Zool Soc London
Broennimann O, Fitzpatrick MC, Pearman PB, Petitpierre B, Pellissier L,
Yoccoz NG, Thuiller W, Fortin M-J, Randin CF, Zimmermann NE,
Graham CH, Guisan A (2012) Measuring ecological niche overlap
from occurrence and spatial environmental data. Glob Ecol Biogeogr
Carranza S, Wade E (2004) Taxonomic revision of Algero-Tunisian
Pleurodeles (Caudata: Salamandridae) using molecular and
morphological data. Revalidation of the taxon Pleurodeles nebulosus
(Guichenot, 1850). Zootaxa 488:124.
Clement M, Posada D, Crandall KA (2000) TCS: a computer program to
estimate gene genealogies. Mol Ecol 9:16571659.
Darriba D, Taboada GL, Doallo R, Posada D (2012) jModelTest 2: more
models, new heuristics and parallel computing. Nat Methods 9:772.
Drummond AJ, Suchard MA, Xie D, Rambaut A (2012) Bayesian
phylogenetics with BEAUti and the BEAST 1.7. Mol Biol Evol
Eiselt J (1958) Der Feuersalamander Salamandra salamandra (L.).
age zu einer taxonomischen Synthese. Abh Ber f Naturkd
Magdeburg 10:77154.
Fisher-Reid MC, Engstrom TN, Kuczynski CA, Stephens PR, Wiens JJ
(2013) Parapatric divergence of sympatric morphs in a salamander:
incipient speciation on Long Island? Mol Ecol 22:46814694.
Fitzpatrick JW (2010) Subspecies are for convenience. Ornithol Monogr
Fitzpatrick BM, Shook K, Izally R (2009) Frequency-dependent selection
by wild birds promotes polymorphism in model salamanders. BMC
Ecol 9:12.
ıs M, Good DA, Parra-Olea G, Wake DB (2000) Biodiversity
of costa rican salamanders: implications of high levels of genetic
differentiation and phylogeographic structure for species formation.
Proc Natl Acad Sci USA 97:16401647.
ıs M, Alcobendas M, Buckley D, Wake DB (2003) Dispersal
of viviparity across contact zones in Iberian populations of re
salamanders (Salamandra) inferred from discordance of genetic and
morphological traits. Evolution 57:129143.
Gray SM, McKinnon JS (2007) Linking color polymorphism
maintenance and speciation. Trends Ecol Evol 22:7179.
Green J (1818) Descriptions of several species of North American
Amphibia, accompanied with observations. J Acad Nat Sci
Philadelphia 1:348359.
Haig SM, Winker K (2010) Avian subspecies: summary and prospectus.
Ornithol Monogr 67:172175.
Highton R (1959) The inheritance of the color phases of Plethodon
cinereus. Copeia 1959:3337.
Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very
high resolution interpolated climate surfaces for global land areas. Int J
Climatol 25:19651978.
Hoffman EA, Blouin MS (2000) A review of colour and pattern
polymorphisms in anurans. Biol J Linn Soc 70:633665.
Hugall AF, Stuart-Fox D (2012) Accelerated speciation in colour-
polymorphic birds. Nature 485:631634.
Huxley JS (1955) Morphism in birds. In: Portmann A, Sutter E (eds),
Proc 11th Int Ornithological Congress. Birkh
auser, Basel, pp 309328.
ohler G, Steinfartz S (2006) A new subspecies of the re salamander,
Salamandra salamandra (Linnaeus, 1758) from the Tendi valley,
Asturias, Spain. Salamandra 42:1320.
Librado P, Rozas J (2009) DnaSP v5: a software for comprehensive
analysis of DNA polymorphism data. Bioinformatics 25:14511452.
Linnaeus C (1758) Systema Naturae per Regna Tria Naturae, Secundum
Classes, Ordines, Genera, Species, cum Characteribus, Differentiis,
Synonymis, Locis. 10 edn. L. Salvii, Stockholm.
Malkmus R (1991) Einige Bemerkungen zum feuersalamander Portugals
(Salamandra salamandra gallaica- Komplex) (Amphibia, Urodela:
Salamandridae). Zool Abh Staatl Mus Tierk Dresden 46:165190.
ınez-Rica JP, Rein
nales A (1988) Altitudinal distribution of
amphibians and reptiles in the Spanish Pyrenees. Pirineos 131:5782.
Mayr E (1982) The Growth of Biological Thought. Belknap P. of
Harvard U.P, Cambridge.
McKnight ML, Nelson NA (2007) Life history and color variants in a
matriline of Oklahoma Salamander (Eurycea tynerensis). Southeast Nat
Montori A, Herrero P (2004) Caudata. In: Garc
ıs M, Montori A,
Herrero P (eds), Amphibia, Lissamphibia, Museo Nacional de Ciencias
Naturales. CSIC, Madrid, pp 43275.
Nishikawa K, Khonsue W, Pomchote P, Matsui M (2013) Two new
species of Tylototriton from Thailand (Amphibia: Urodela:
Salamandridae). Zootaxa 3737:261279.
Noonan BP, Comeault AA (2009) The role of predator selection on
polymorphic aposematic poison frogs. Biol Lett 5:5154.
Nussbaum RA, Brodie ED, Datong Y (1995) A Taxonomic Review of
Tylototriton verrucosus Anderson (Amphibia: Caudata:
Salamandridae). Herpetologica 51:257268.
ONeill EM, Beard KH (2010) Genetic basis of a color pattern
polymorphism in the Coqui Frog Eleutherodactylus coqui. J Hered
Pabijan M, Zieli
nski P, Dudek K, Chloupek M, Sotiropoulos K, Liana
M, Babik W (2015) The dissection of a Pleistocene refugium:
phylogeography of the smooth newt, Lissotriton vulgaris, in the
Balkans. J Biogeogr 42:671683.
Pasmans F, Keller H (2000) Morphological variation in neighbouring
populations of Salamandra salamandra bernardezi in northern Spain.
Zeitschr Feldherp 7:7784.
Pasmans F, Bogaerts S, Keller H (2004) Note on the distribution of
Salamandra salamandra cf. bernardezi in Asturias, northern Spain.
POD@RCIS 5:5860.
Pereira RJ, Wake DB (2009) Genetic leakage after adaptive and
nonadaptive divergence in the Ensatina eschscholtzii ring species.
Evolution 63:22882301.
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
Colour polymorphism in Salamandra salamandra 9
Peters WCH (1859) Eine neue vom Herrn Jagor im atlantischen Meere
gefangene Art der Gattung Leptocephalus, und
uber einige andere
neue Fische des Zoologischen Museums. Monatsb K
on Preuss Akad
Wiss Berlin 1859:411413.
Petranka JW (2010) Salamanders of the United States and Canada.
Smithsonian Books, Washington DC.
Richards-Zawacki CL, Wang IJ, Summers K (2012) Mate choice and the
genetic basis for colour variation in a polymorphic dart frog:
inferences from a wild pedigree. Mol Ecol 21:38793892.
Roulin A (2004) The evolution, maintenance and adaptive function of
genetic colour polymorphism in birds. Biol Rev Camb Philos Soc
Roulin A, Bize P, Ravussin P-A, Broch L (2004) Genetic and
environmental effects on the covariation between colour polymorphism
and a life-history trait. Evol Ecol Res 6:12531260.
Rudh A, Qvarnstr
om A (2013) Adaptive colouration in amphibians.
Semin Cell Dev Biol 24:553561.
Schoener TW (1968) The Anolis lizards of Bimini: resource partitioning
in a complex fauna. Ecology 49:704726.
Speybroeck J, Beukema W, Crochet PA (2010) A tentative species list of
the European herpetofauna (Amphibia and Reptilia) an update.
Zootaxa 2492:127.
Steinfartz S, Veith M, Tautz D (2000) Mitochondrial sequence analysis
of Salamandra taxa suggests old splits of major lineages and
postglacial recolonizations of Central Europe from distinct source
populations of Salamandra salamandra. Mol Ecol 9:397410.
Thiesmeier B, Grossenbacher K (2004) Salamandra salamandra
(Linnaeus, 1758) Feuersalamander. In: Thiesmeier B,
Grossenbacher K (eds), Handbuch der Reptilien und Amphibien
Europas. Aula, Wiebelsheim, pp 10591132.
Torstrom SM, Pangle KL, Swanson BJ (2014) Shedding subspecies: the
inuence of genetics on reptile subspecies taxonomy. Mol Phylogenet
Evol 76:134143.
on G, Garc
ıs M, Gal
an P, Cordero Rivera A (2007) The
evolution of viviparity in holocene islands: ecological adaptation
versus phylogenetic descent along the transition from aquatic to
terrestrial environments. J Zool Syst Evol Res 45:345352.
on G, Santos X, Sanmart
ın-Villar I, Cordero-Rivera A, Buckley
D (2015) Intraspecic variation in clutch size and maternal investment
in pueriparous and larviparous Salamandra salamandra females. Evol
Ecol 29:185204.
Vences M, Sanchez E, Hauswaldt JS, Eikelmann D, Rodr
ıguez A,
Carranza S, Donaire D, Gehara M, Helfer V, L
otters S, Werner P,
Schulz S, Steinfartz S (2014) Nuclear and mitochondrial multilocus
phylogeny and survey of alkaloid content in true salamanders of the
genus Salamandra (Salamandridae). Mol Phylogenet Evol 73:208216.
Venesky MD, Hess A, DeMarchi JA, Weil Murone J, Hickerson C-AM,
Anthony CD (2015) Morph-specic differences in disease prevalence
and pathogen-induced mortality in a terrestrial polymorphic
salamander. J Zool 295:279285.
Villanueva A (1993) Hallazgo de una nueva coloracion de Salamandra
salamandra bernardezi en Asturias. Bol Assoc Herpetol Esp 4:1415.
Wake DB, Roth G, Wake MH (1983) On the problem of stasis in
organismal evolution. J Theor Biol 101:211224.
Warren DL, Glor RE, Turelli M (2008) Environmental niche equivalency
versus conservatism: quantitative approaches to niche evolution.
Evolution 62:28682883.
Wells KD (2007) The Ecology and Behaviour of Amphibians. The
University of Chicago Press, Chicago.
Wielstra B, Beukema W, Arntzen JW, Skidmore AK, Toxopeus AG,
Raes N (2012) Corresponding mitochondrial DNA and niche
divergence for crested newt candidate species. PLoS One 7:e46671.
Wollenberg KC, Measey GJ (2009) Why colour in subterranean
vertebrates? Exploring the evolution of colour patterns in caecilian
amphibians. J Evol Biol 22:10461056.
Wolterstorff W (1928) Vollmolche-geb
arende Feuersalamander aus
Oviedo. Bl Aquar Terrarienkd 39:132133.
Wu Y, Wang Y, Jiang K, Chen X, Hanken J (2010) Homoplastic
evolution of external colouration in Asian stout newts (Pachytriton)
inferred from molecular phylogeny. Zool Scr 39:922.
Zhang P, Papenfuss TJ, Wake MH, Qu L, Wake DB (2008) Phylogeny
and biogeography of the family Salamandridae (Amphibia: Caudata)
inferred from complete mitochondrial genomes. Mol Phylogenet Evol
Supporting Information
Additional Supporting Information may be found in the online
version of this article:
Table S1. Overview of locations used for population and
genetic analyses and niche modelling.
doi: 10.1111/jzs.12119
©2016 Blackwell Verlag GmbH
... The fire salamander Salamandra salamandra (Linnaeus, 1758) is a temperate amphibian largely distributed in Europe 18 . Owing to its remarkable phenotypic variation in traits encompassing external morphology, colour patterns, and reproductive strategy, the geographic variation and population structure of S. salamandra have been investigated in several portions of its range, using various combinations of phenotypic and genetic traits 19,20 . As a consequence, the taxonomy of fire salamanders has been long discussed and repeatedly revised. ...
... In this case, evidence for admixture were substantial and appeared mostly asymmetric, from south to north. Finally, with K=5 ( Figure 2F) samples drawn from the alpine arc (19)(20)(21) were assigned to a distinct cluster. Among them, sample19 (i.e. the one in closer geographic contiguity to the remaining samples), was the only one showing evidence of mixed ancestry. ...
... Two main groups of closely related haplotypes were observed, separated from one another by 25 mutational steps. One haplogroup(green in Figure 2A; haplotype series S)was geographically restricted to the south of the Italian peninsula (samples 1-8), whereas the other haplogroup (red in Figure 2A; haplotype series N) was widespread from the alpine arc to samples in south-central Italy (samples [8][9][10][11][12][13][14][15][16][17][18][19][20][21]. Syntopy between both haplogroups was only found at the geographically intermediate sample 8. Within the southern haplogroup, to sub-groups of haplotypes were found, one restricted to samples from the Calabria region, and the other restricted to more northern samples (dark and light green in Fig. 2A, respectively). ...
Full-text available
Discordance between mitochondrial and nuclear patterns of population genetic structure is providing key insights into the eco-evolutionary dynamics between and within species, and their assessment is highly relevant to biodiversity monitoring practices based on DNA barcoding approaches. Here, we investigate the population genetic structure of the fire salamander Salamandra salamandra in peninsular Italy. Both mitochondrial and nuclear markers clearly identified two main population groups. However, nuclear and mitochondrial zones of geographic transition between groups were located 600 km from one another. The overall pattern of genetic variation, together with morphological and fossil data, suggest that a rampant mitochondrial introgression triggered the observed mitonuclear discordance, following a post-glacial secondary contact between lineages. Moreover, at a shallower level of population structure, we observed evidence of asymmetric introgression of nuclear genes between two sub-groups in southern Italy. Our results clearly show the major role played by reticulate evolution in shaping the structure of Salamandra salamandra populations and, together with similar findings in other regions of the species’ range, contribute to identify the fire salamander as a particularly intriguing case to investigate the complexity of mechanisms triggering patterns of mitonuclear discordance in animals.
... Disease mitigation and conservation breeding programmes require robust decision-making based upon multiple complementary priorities, including the genetic representativeness of at risk populations (Thomas et al., 2019); therefore, narrowing existing knowledge gaps within S. salamandra subspecific relationships, and delimiting genetically distinct intraspecific lineages is a conservation priority. Furthermore, molecular techniques that are sensitive enough to resolve recent relationships would be particularly useful for this genus, in order to generate robust hypotheses about the evolution of ecologically relevant adaptive traits such as colouration (Beukema et al., 2016;Burgon et al., 2020) reproductive mode (Buckley et al., 2007;Velo-Antón et al., 2012) and morphological traits (Alarcón-Ríos et al, 2020. Here, our goal is to clarify the evolutionary relationships of Salamandra with a phylogenomic approach, mainly focusing on subspecies-level lineages. ...
... One comprised two northern Iberian subspecies, S. s. fastuosa and S. s. bernardezi, as well as the Apenninic S. s. gigliolii. Some S. s. bernardezi populations display a strikingly polymorphic colouration and were previously described as subspecies "alfredschmidti" (Köhler and Steinfartz, 2006), later subsumed within bernardezi (Beukema et al., 2016). Our tree supports the placement of "alfredschmidti" populations within bernardezi, again suggesting that "alfredschmidti" as initially defined is not a genetically distinctive unit (Beukema et al., 2016;Burgon et al., 2020). ...
... Some S. s. bernardezi populations display a strikingly polymorphic colouration and were previously described as subspecies "alfredschmidti" (Köhler and Steinfartz, 2006), later subsumed within bernardezi (Beukema et al., 2016). Our tree supports the placement of "alfredschmidti" populations within bernardezi, again suggesting that "alfredschmidti" as initially defined is not a genetically distinctive unit (Beukema et al., 2016;Burgon et al., 2020). must have occurred, and we suspect that this introgression is a widespread phenomenon across these clades. ...
Full-text available
The salamander genus Salamandra is widespread across Europe, North Africa, and the Near East and is renowned for its conspicuous and polymorphic colouration and diversity of reproductive modes. The phylogenetic relationships within the genus, and especially among the highly polymorphic species S. salamandra, have been very challenging to elucidate, leaving its real evolutionary history and classification at species and subspecies levels a topic of debate and contention. However, the distribution of diversity and species delimitation within the genus are critically important for identifying evolutionarily significant units for conservation and management, especially in light of threats posed by the pathogenic chytrid fungus Batrachochytrium salamandrivorans that is causing massive declines of S. salamandra populations in central Europe. Here, we conducted a phylogenomic analysis from across the taxonomic and geographic breadth of the genus Salamandra in its entire range. Bayesian, maximum likelihood and network-based phylogenetic analyses of up to 4,905 ddRADseq-loci (294,300 nucleotides of sequence) supported the distinctiveness of all currently recognised species (Salamandra algira, S. atra, S. corsica, S. infraimmaculata, S. lanzai, and S. salamandra), and all five species for which we have multiple exemplars were confirmed as monophyletic. Within S. salamandra, two main clades can be distinguished: one clade with the Apenninic subspecies S. s. gigliolii nested within the Iberian S. s. bernardezi/fastuosa; the second clade comprising all other Iberian, Central and East European subspecies. Our analyses revealed that some of the currently recognized subspecies of S. salamandra are paraphyletic and may require taxonomic revision, with the Central- and Eastern-European subspecies all being poorly differentiated in the analysed genomic markers. Salamandra s. longirostris – sometimes considered a separate species – was nested within S. salamandra, consistent with its subspecies status. The relationships identified within and between Salamandra species provide valuable context for future systematic and biogeographic studies, and help elucidate critical evolutionary units for conservation and taxonomy.
... Within S. salamandra, pueriparity has been described in three of the ten main subspecies. Salamandra s. bernardezi in northern Spain ( Fig. 1) is considered pueriparous (Alarcón-Ríos et al., 2020a;Buckley et al., 2007;Velo-Antón et al., 2015), but genetic diversity and divergence within the subspecies is very high (Beukema et al., 2016;Lourenço et al., 2019), and for large parts of its range there are no direct observations of reproductive mode ( Fig. 2A). Salamandra s. fastuosa, found to the east of S. s. bernardezi ( Fig. 1 and Fig. 1. ...
... We amplified a CytB fragment of ca. 1100 bp, following the protocol described in (Beukema et al., 2016) and outsourced DNA sequencing to Genewiz Inc. (Leipzig, Germany). ...
The ability to bear live offspring, viviparity, has evolved multiple times across the tree of life and is a remarkable adaptation with profound life-history and ecological implications. Within amphibians the ancestral reproductive mode is oviparity followed by a larval life stage, but viviparity has evolved independently in all three amphibian orders. Two types of viviparous reproduction can be distinguished in amphibians; larviparity and pueriparity. Larviparous amphibians deliver larvae into nearby ponds and streams, while pueriparous amphibians deliver fully developed juveniles and thus do not require waterbodies for reproduction. Among amphibians, the salamander genus Salamandra is remarkable as it exhibits both inter- and intraspecific variation in the occurrence of larviparity and pueriparity. While the evolutionary relationships among Salamandra lineages have been the focus of several recent studies, our understanding of how often and when transitions between modes occurred is still incomplete. Furthermore, in species with intraspecific variation, the reproductive mode of a given population can only be confirmed by direct observation of births and thus the prevalence of pueriparous populations is also incompletely documented. We used sequence capture to obtain 1,326 loci from 94 individuals from across the geographic range of the genus, focusing on potential reproductive mode transition zones. We also report additional direct observations of pueriparous births for 20 new locations and multiple lineages. We identify at least five independent transitions from the ancestral mode of larviparity to pueriparity among and within species, occurring at different evolutionary timescales ranging from the Pliocene to the Holocene. Four of these transitions occurred within species. Based on a distinct set of markers and analyses, we also confirm previous findings of introgression between species and the need for taxonomic revisions in the genus. We discuss the implications of our findings with respect to the evolution of this complex trait, and the potential of using five independent convergent transitions for further studies on the ecological context in which pueriparity evolves and genetic architecture of this specialized reproductive mode.
... According to Arnold (2002), the two lineages share a contact zone, which -based on colour pattern phenotypes -appears to stretch from eastern Germany to the Czech Republic and Poland. To date, only a few studies have attempted to quantify the distribution of fire salamanders' colour phenotypes (Klewen 1985, Beukema et al. 2016, Najbar et al. 2018, Burgon et al. 2020) and even fewer have related it to the distribution of genetic lineages (Veith 1992, Beukema et al. 2016, Najbar et al. 2018, Burgon et al. 2020. ...
... According to Arnold (2002), the two lineages share a contact zone, which -based on colour pattern phenotypes -appears to stretch from eastern Germany to the Czech Republic and Poland. To date, only a few studies have attempted to quantify the distribution of fire salamanders' colour phenotypes (Klewen 1985, Beukema et al. 2016, Najbar et al. 2018, Burgon et al. 2020) and even fewer have related it to the distribution of genetic lineages (Veith 1992, Beukema et al. 2016, Najbar et al. 2018, Burgon et al. 2020. ...
Full-text available
Two evolutionary lineages of the fire salamander occur in central Europe: the typically striped subspecies Salamandra salamandra terrestris (Bonnaterre, 1789) and the typically spotted Salamandra salamandra salamandra (Linnaeus, 1758). In the Czech Republic, fire salamanders have traditionally been viewed as belonging to the S. s. salamandra evolutionary lineage. Nevertheless, the colour pattern of some individuals in the westernmost part of the Czech Republic resembles that of S. s. terrestris in having parallel continuous bands along the back. In this study, we investigated whether in the Czech Republic the presence of striped fire salamander phenotype could be associated with the genotype of S. s. terrestris. We sequenced the mitochondrial D-loop and two nuclear markers, Rag2 and PDGFRα, of 61 fire salamander individuals from the Czech Republic. To describe the geographical distribution pattern of the striped and spotted fire salamander phenotype in the Czech Republic, we evaluated colour phenotypes of 398 individuals from ten localities distributed so as to cover the whole country. We found no evidence of presence of genotypes corresponding to the S. s. terrestris lineage. We did, however, find that the striped phenotype is found mostly in the northwest of the Czech Republic, where both the striped and the intermediate phenotype occur significantly more frequently than in the rest of the country, where the spotted phenotype seems dominant. This finding indicates that Czech and Polish populations of S. salamandra show a degree of phenotypic pattern variation comparable to that observed in German populations, although at a local level the frequencies of the striped and spotted phenotype vary. It would be interesting to test whether a genetic toolkit responsible for the colour pattern is shared via genetic introgression between populations, or whether the striped phenotype of Czech fire salamanders evolved independently.
... Main cases of male-male interactions in the genus Salamandra reported in the scientific literature and thematic books. b as S. salamandra alfredschmidti (synonymy byBeukema et al., 2016;Burgon et al., 2021). c as S. salamandra europaea (synonymy byVelo-Antón and Buckley, 2015). ...
Full-text available
Di Nicola M.R., Zabbia T., Mezzadri S., Cerullo A., Bruni G. (2022). Male-male interactions in Alpine Salamanders, Salamandra atra atra (Laurenti, 1768), with an overview of the main cases reported for the genus Salamandra (Amphibia: Salamandridae). Herpetology Notes, 15: 601-604.
... As also indicated earlier, many toxic amphibians combine their toxicity with aposematic coloring to warn potential predators to keep distance [43]. Well-known examples are the poison dart frogs in the genus Phyllobates (Dendrobatidae) from Central and South America [44], and the fire salamander Salamandra salamandra (Salamandridae) from southern and central Europe [45]. These amphibians combine the production of the highly toxic alkaloids curare and samandarin, respectively, in their skin secretions with vividly bright skin coloration. ...
... 1100 bp) of the cytochrome b and adjacent tRNAs (hereafter cytb). We used the primers Glu14100L and Pro15500H 44 to amplify cytb following a previously described protocol 28 . We outsourced DNA sequencing to Genewiz Inc. (Leipzig, Germany), and inspected and aligned resulting chromatograms using GENEIOUS version 11.1.4 ...
Full-text available
Explicitly accounting for phenotypic differentiation together with environmental heterogeneity is crucial to understand the evolutionary dynamics in hybrid zones. Species showing intra-specific variation in phenotypic traits that meet across environmentally heterogeneous regions constitute excellent natural settings to study the role of phenotypic differentiation and environmental factors in shaping the spatial extent and patterns of admixture in hybrid zones. We studied three environmentally distinct contact zones where morphologically and reproductively divergent subspecies of Salamandra salamandra co-occur: the pueriparous S. s. bernardezi that is mostly parapatric to its three larviparous subspecies neighbours. We used a landscape genetics framework to: (i) characterise the spatial location and extent of each contact zone; (ii) assess patterns of introgression and hybridization between subspecies pairs; and (iii) examine the role of environmental heterogeneity in the evolutionary dynamics of hybrid zones. We found high levels of introgression between parity modes, and between distinct phenotypes, thus demonstrating the evolution to pueriparity alone or morphological differentiation do not lead to reproductive isolation between these highly divergent S. salamandra morphotypes. However, we detected substantial variation in patterns of hybridization across contact zones, being lower in the contact zone located on a topographically complex area. We highlight the importance of accounting for spatial environmental heterogeneity when studying evolutionary dynamics of hybrid zones.
... Across Europe, several genetically differentiated lineages of S. salamandra hybridize in contact zones of varying widths, usually associated with strong mito-nuclear discordances (e.g., García-París et al. 2003;Pereira et al. 2016;Bisconti et al. 2018), and sometimes coincident with sharp ecological transitions. In such cases, parental populations have evolved differences in color pattern (Beukema et al. 2016;Burgon et al. 2020), morphology (Alarcón-Ríos et al. 2020, and reproductive modes (García-París et al. 2003;Velo-Antón et al. 2007;Lourenço et al. 2019) that are suggestive of local adaptation and are taxonomically recognized as subspecies. The integration of population and landscape analyses with the study of hybrid zones in Salamandra can help to infer the relative role of landscape features and ecological factors in reproductive isolation. ...
Landscape features shape patterns of gene flow among populations, ultimately determining where taxa lay along the continuum between panmixia to complete reproductive isolation. Gene flow can be restricted, leading to population differentiation in two non-exclusive ways: “physical isolation”, in which geographic distance in combination with the landscape features restricts movement of individuals promoting genetic drift, and “ecological isolation”, in which adaptive mechanisms constrain gene flow between different environments via divergent natural selection. In central Iberia, two fire salamander subspecies occur in parapatry across elevation gradients along the Iberian Central System mountains, while in the adjacent Montes de Toledo Region only one of them occurs. By integrating population and landscape genetic analyses, we show a ubiquitous role of physical isolation between and within mountain ranges, with unsuitable landscapes increasing differentiation between populations. However, across the Iberian Central System, we found strong support for a significant contribution of ecological isolation, with low genetic differentiation in environmentally homogeneous areas, but high differentiation across sharp transitions in precipitation seasonality. These patterns are consistent with a significant contribution of ecological isolation in restricting gene flow among subspecies. Overall, our results suggest that ecological divergence contributes to reduce genetic admixture, creating an opportunity for lineages to follow distinct evolutionary trajectories.
Investigation of the ecological and evolutionary mechanisms governing the origin and diversification of species requires integrative approaches that often have to accommodate strong discordance among datasets. A common source of conflict is the combination of morphological and molecular characters with different evolutionary rates. Resolution of these discordances is crucial to assess the relative roles of different processes in generating and maintaining biodiversity. Anuran amphibians provide many examples of morphologically similar, genetically divergent lineages, posing questions about the relative roles of phylogeny and ecological factors in phenotypic evolution. We focused on three circum-Mediterranean anuran genera (Hyla, Alytes and Discoglossus), characterizing morphological and environmental disparity and comparing diversity patterns across biological levels of organization. Using a comparative phylogenetic framework, we tested how shared ancestry and climatic factors come together to shape phenotypic diversity. We found higher morphological differentiation within Hyla and Alytes than in Discoglossus. Body size and limb morphology contributed most to inter- and intraspecific morphological variation in Hyla and Alytes, but there was no strong phylogenetic signal, indicating that shared ancestry does not predict patterns of phenotypic divergence. In contrast, we uncovered a significant association between morphology and climatic descriptors, supporting the hypothesis that morphological disparity between species results from adaptive evolution.
Full-text available
Warning signals are predicted to develop signal monomorphism via positive frequency‐dependent selection (+FDS) albeit many aposematic systems exhibit signal polymorphism. To understand this mismatch, we conducted a large‐scale predation experiment in four countries, among which the frequencies of hindwing warning coloration of the aposematic moth, Arctia plantaginis, differ. Here we show that selection by avian predators on warning colour is predicted by local morph frequency and predator community composition. We found +FDS to be the strongest in monomorphic Scotland and lowest in polymorphic Finland, where the attack risk of moth morphs depended on the local avian community. +FDS was also found where the predator community was the least diverse (Georgia), whereas in the most diverse avian community (Estonia), hardly any models were attacked. Our results support the idea that spatial variation in predator communities alters the strength or direction of selection on warning signals, thus facilitating a geographic mosaic of selection. A geographic mosaic of selection by predators could explain the paradoxical maintenance of warning signal variation, but direct ecological evidence is scarce and focused on tropical systems. We monitored local avian predators and attacks on 4000 + moth models representing red, yellow or white warning colour morphs in a temperate moth system with natural variation in local morph frequencies. We found positive frequency‐dependent selection to be strongest in monomorphic populations and the direction and strength of selection to be significantly associated with local predator community composition and diversification, which can explain not only geographic variation (polytypism) but also local polymorphism when coupled with gene flow.
Full-text available
Research on the taxonomy of European amphibians and reptiles has increased noticeably over the last few decades, indicating the need for recognition of new species and the cancellation of others. This paper provides a critical review of recent changes and draws up a tentative species list.
Full-text available
Ontogenetic, post-metamorphic pattern development is a rarely studied topic in amphibian science. As there are indications that the pattern of Salamandra corsica might expand over time, digital image analyses were applied in order to measure several phenol typical variables which were related to the snout vent length. Results show a significant increase of patches which change to irregular shapes while SVL increases. Digital image analysis is identified as a suitable tool to explore pattern shape and change in general, while the documented pattern development in S. corsica might be one of the first quantified cases of post-metamorphic ontogenetic pattern change in amphibians.
Full-text available
During an excursion to the southern Lycian coast (southwestern Turkey) from the 12 th February to the 21 st of February 2007 observations were made on three populations of Lyciasalamandra luschani finikensis, while a fourth population was discovered on the Çatal Tepesi between 190 and 200 m a. s. l. Notes on the vegetation communities of the habitat were made at all four locations. Thirty individuals were measured at the Çatal Tepesi. This population is isolated by the Alakır River from the karstic limestone Bey Dağları Platform on which all subspecies of Lyciasalamandra luschani occur. The previous known locality near Turunçova is most likely continuous up to a visited locality on the central Gölcük Tepesi at 800 m a. s. l., which concerns a minor
The taxonomic status of Algero-Tunisian Pleurodeles was reanalysed in the light of new molecular and morphological evidence. Mitochondrial DNA sequences ( 396 bp of the cytochrome b and 369 of the 12S rRNA) and the results of the morphometric analysis, indicate that Algero-Tunisian P. poireti consists of two genetically and morphologically distinct forms. One restricted to the Edough Peninsula, and another one covering all the rest of its distribution in Algeria and Tunisia. The name P. poireti ( Gervais, 1835) is restricted to the population of the Edough Peninsula, while P. nebulous (Guichenot, 1850) correctly applies to all other populations in the distribution. P. poireti originated approximately 4.2 Myr ago, probably as a result of the Edough Peninsula being a Pliocene fossil island, allowing both forms of Algero-Tunisian Pleurodeles to diverge both genetically and morphologically.