Content uploaded by David Prötzel
Author content
All content in this area was uploaded by David Prötzel on Oct 31, 2020
Content may be subject to copyright.
342
F G et al.
Open access at hp://www.salamandra-journal.com
© 2020 Deutsche Gesellscha für Herpetologie und Terrarienkunde e.V. (DGHT), Mannheim, Germany
30 October 2020 ISSN 0036–3375
SALAMANDRA 56(4): 342–354 SALAMANDRA
German Journal of Herpetology
Rediscovery, conservation status and genetic relationships
of the Malagasy chameleon Furcifer voeltzkowi
F G, D P, F E, N A. R,
R N. R, T G, K G,
J F M V
1) Zoologische Staatssammlung München (ZSM-SNSB), Münchhausenstr. 21, 81247 München, Germany
2) NABU Niedersachsen e.V., Projektbüro, Kerschensteinerweg 3, 31757 Rinteln, Germany
3) Mention Zoologie et Biodiversité Animale, Faculté des Sciences, Université d’Antananarivo, Antananarivo 101, Madagascar
4) Museum Mensch und Natur, Schloss Nymphenburg, 80638 München, Germany
5) Zoologisches Institut, Technische Universität Braunschweig, Mendelssohnstr. 4, 38106 Braunschweig, Germany
Corresponding author: F G, e-mail: glaw@snsb.de
Manuscript received: 30 April 2020
Accepted: 1 September 2020 by J K
Abstract. e chameleon Furcifer voeltzkowi (B, ) from northwestern Madagascar was considered to be a
synonym of Furcifer rhinoceratus for many decades and was resurrected only recently based on studies of the morphology
and osteology of a few male specimens, which were collected more than years ago. However, basic data on this species
remain unavailable, including its conservation status, life history, colouration in life, morphology of the female, genital mor-
phology of the male, phylogenetic anities, and genetic dierentiation from F. labordi and other congeners. During a tar-
geted expedition, we rediscovered F. voeltzkowi in its natural habitat close to its type locality, allowing us to ll some gaps of
knowledge. Furcifer voeltzkowi is a sexually dimorphic species. e life colouration of males is largely green, whereas that of
females is highly variable and can be extremely colourful. Both, morphology and life colouration of males and females show
close similarities to F. labordi from west and southwest Madagascar, but also a number of distinct dierences (e.g. a larger
size of F. voeltzkowi and a smaller rostral appendage in both sexes), enabling a clear distinction of the two species in both
sexes. DNA sequences of the nuclear CMOS gene and two mitochondrial markers (S rRNA and ND) also conrm that
F. voeltzkowi is a distinct species and sister to F. labordi (pairwise dierences in the S gene .–.). We estimate the dis-
tribution range (extent of occurrence) of F. voeltzkowi to comprise ca. . km and suggest that it qualies as Endangered
Bab(iii) under the IUCN Red List Criteria as its populations are expected to be severely fragmented, and there is continuing
decline in the extent and quality of its habitat. Similar to F. labordi, F. voeltzkowi might be an extremely short-living chamele-
on with a post hatching life span of only several months in the rainy season. e assumed short life might also partly explain
why this splendid species got “lost” for many decades, since most roads in its habitat are not accessible in the wet season.
Key words. Squamata, Chamaeleonidae, morphology, molecular genetics, rediscovery, conservation.
Introduction
Our planet is probably facing the beginning of an enor-
mous extinction of species, oen referred to as the “sixth
mass extinction”, the “Holocene extinction” or the “Anthro-
pocene extinction”. In contrast to the ve earlier catastroph-
ic extinction periods in the Earth’s history the current loss
of biodiversity is caused by human activity (W V-
, K , C et al. ). Ac-
cording to the IUCN Red List () species ( ani-
mals and plants) have been ocially classied as Extinct
since the beginning of modern times about years ago.
However, the number of recent extinctions documented by
the Extinct category on the IUCN Red List is likely to be a
signicant underestimate (IUCN ). e recently intro-
duced IUCN category Possibly Extinct is already applied to
more species () than the category Extinct (IUCN ),
demonstrating the poor state of knowledge and the risk of
“silent” (unrecognized) extinctions for many species. Nu-
merous researchers and conservationists are convinced that
these ocial gures do not appropriately mirror the dra-
matic extent of the present biodiversity loss, and if current
estimates of extinction rates (e.g., P et al. ) are cor-
rect, they would translate into hundreds of species being
driven to extinction every year. ere is widespread agree-
ment that our knowledge on the total species number on
343
Rediscovery of Furcifer voeltzkowi
Earth is very rudimentary, and that the taxonomy of most
organism groups is in urgent need of revision (L et
al. ). e lack of reliable knowledge is not restricted to
microorganisms and insects – the continued existence of
numerous vertebrate species is uncertain as well.
Rediscoveries of “lost” species are very important as they
provide crucial data for conservation measures and also
bring some hope amidst the biodiversity crisis. In the
NGO Global Wildlife Conservation initiated a program to
rediscover poorly known species, which had not been re-
corded for many decades or even centuries. For this pro-
gram a list of “ most wanted lost species” was produced.
As of March four of them were successfully rediscov-
ered: Jackson’s climbing salamander (Bolitoglossa jacksoni),
the silver-backed chevrotain (Tragulus versicolor), Wallace’s
giant bee (Megachile pluto), and the velvet pitcher plant
(Nepenthes mollis), see https://www.globalwildlife.org/
search-for-lost-species/. e chameleon species Furcifer
voeltzkowi, which had not been recorded for more than
years, was also a target species of this program.
Furcifer voeltzkowi has a convoluted taxonomic histo-
ry. e original description as Chamaeleon voeltzkowi by
B () was based on a single male and comple-
mented by data and photographs of additional specimens
years later (B ). Based on the distribution
of morphological characters in a limited number of
specimens H () synonymized C. labordi and
C. voeltzkowi with Chamaeleo rhinoceratus and insisted on
this conclusion in a subsequent paper (H ).
M () listed labordi and voeltzkowi as subspecies
of rhinoceratus in his taxonomic checklist, though without
providing a justication for this change. In their revision
of the species group B D () resur-
rected C. labordi, but le C. voeltzkowi in the synonymy of
C. rhinoceratus. K B () split the genus
Chamaeleo into four genera and placed several species in
the resurrected genus Furcifer, including F. rhinoceratus and
F. labordi. G () noted that the synonymy of voeltz-
kowi with F. rhinoceratus was in need of revision and nally
S et al. () resurrected F. voeltzkowi based on com-
parisons of external morphology and micro-CT data with
F. labordi, but the continued existence of this species re-
mained uncertain.
In this paper, we report on the rediscovery of Furcifer
voeltzkowi in northwestern Madagascar and aim to ll
some of the major gaps in the knowledge of this species.
We analyze its molecular phylogenetic relationships and the
divergences from its sister species, describe the rst known
female and provide basic data on its habitat, life-history and
conservation status.
Materials and methods
Our rediscovery expedition was conducted at the end of
the rainy season from March to April in the Ma-
hajanga region in northwestern Madagascar. We searched
at and around the following localities for chameleons
and other reptiles and amphibians: () unprotected dry
forests near the Antsanitia hotel and the adjacent river
(-.°, .°, m a.s.l.; – March), () a
secondary forest near the river (-.°, .°,
m a.s.l., March), () around Betsako, the type local-
ity of Furcifer monoceras (-., ., ca. m a.s.l.,
coordinates according to R et al. ;
March), ()Katsepy, mostly in the garden of Hotel Mad-
ame Chabaud (-.°, .°, m a.s.l.; March
to April), () unprotected dry forest around the light-
house near Antrema reserve (-.°, .°,
m a.s.l.; March), the approximate type locality of Fur-
cifer voeltzkowi, and () area around limestone caves east
of Mahajanga (-.°, .°, ca. m a.s.l.; –
April). Specimens used for morphological comparisons are
deposited in the Zoologische Staats samm lung München
(ZSM), the Senckenberg Museum, Frankfurt am Main
(SMF), and the Museum national d’Histoire naturelle Par-
is (MNHN). We examined one adult male (ZSM /)
and one adult female (ZSM /) of F. voeltzkowi and
compared them with several specimens of Furcifer labordi,
which were found dead during a study in the Ki rin dy for-
est near Morondava (ZSM –/). e terminology
and description scheme of the hemipenes follows P
et al. (). Morphological terminology and abbreviations
of diagnostically relevant characters are given according
to S et al. (): (SVL) Snout–vent length, from the
snout tip to the cloaca; (TaL) tail length, from cloaca to tail
tip; (TL) total length, as the sum of TaL and SVL; (HW)
head width, width between the posterior part of the two or-
bital crests in dorsal view; (LCL) lateral crest length, from
the back of the eye ring to the back bone where the angle
changes; (TCL) temporal crest length, from the end of lat-
eral crest to the top of casque; (PCL) parietal crest length;
(LRA) length of rostral appendage, from snout tip to tip of
rostral appendage; (WRA), widest width of rostral append-
age; (CH) casque height, vertical line from the end of lateral
crest; (CN) casque to neck, from the casque tip vertically
to the neck skin; (DC) dorsal crest presence (+) or absence
(–); (DCc) dorsal crest continuous to tail crest presence (+)
or absence (–); (NSL) number of supralabial scales, counted
from the mouth slit to the snout tip. e following measure-
ments were size-corrected using the ratios: (RTaL) TaL to
SVL; (RHW) HW to HL; (RLCL) LCL to SVL; (RTCL) TCL
to SVL; (RPCL) PCL to SVL; (RLRA) LRA to SVL; (RWRA)
WRA to SVL; (RCH) CH to SVL; (RCN) CN to SVL.
Total genomic DNA was extracted following a standard
salt extraction protocol using proteinase K digestion in a
concentration of mg/ml (B et al. ). For mo-
lecular analysis, we complemented DNA sequences from
GenBank obtained in previous studies (R et al.
, T et al. ) with new sequences for Furcifer
voeltzkowi. Primers and PCR protocols were as employed
in previous studies (T et al. ). PCR products were
sequenced directly using an automated DNA sequencer
(ABI XL, Applied Biosystems) and quality-checked
in CodonCode Aligner (Codon Code Corporation). Our
analysis is based on DNA sequences of fragments of the
344
F G et al.
two mitochondrial genes NADH dehydrogenase subu-
nit (ND) and S ribosomal RNA (S), and of the nu-
clear locus oocyte maturation factor Mos (CMOS). Newly
generated sequences were deposited in GenBank with the
following accession numbers: MT–MT and
MT–MT. We used MEGA (K et al.
) for sequence alignment, for calculating uncorrected
p-distances between sequences, and for phylogenetic analy-
sis. Because for several species (F. antimena, F.belalandaen-
sis) sequences from reliably identied individuals were only
available for ND, and exploratory analyses of the available
S sequences yielded poor phylogenetic support, we relied
on ND only to reconstruct the mitochondrial phylogenet-
ic relationships among the target taxa. Model testing based
on the Bayesian Information Criterion in MEGA suggest-
ed a HKY+G model best tting the data. We performed
maximum likelihood (ML) analyses in MEGA with SPR
branch swapping and bootstrap pseudoreplicates. For
CMOS, sequences were aligned in MEGA and analyzed
manually. Because two sequences were of poor quality in
the beginning, a total of bp were discarded given that
these were completely invariable in six sequences for which
they were fully represented; consequently only bp were
analyzed and since only two substitutions were found in the
gene fragment, we drew a CMOS haplotype network man-
ually to represent allelic variation.
Results
Molecular dierentiation and genetic divergences
e newly obtained S sequences of the three F. voeltzkowi
specimens from Katsepy (total alignment length bp)
were identical to each other, but diered from previously
published and newly obtained S sequences of F.labordi
by – substitutions (.–. uncorrected pairwise dis-
tance).
e newly obtained ND sequences (of ve individu-
als of F. labordi from Kirindy and three F. voeltzkowi from
Katsepy) were aligned with sequences of the related species
F. labordi, F. antimena, F. viridis, F. major and F. belalan da-
en sis from GenBank. A sample of F. verru cosus was used
as the outgroup. In the ND fragment, F.voeltz kowi and
F. labordi diered by – substitutions (.–. uncor-
rected pairwise distance). e inferred phylogenetic tree
(Fig. ) based on a total of bp of ND reconstructed
F. voeltzkowi as sister species to F. labordi. e tree also
suggests that the ND sequence HF of F.anti mena
in T et al. () most likely represents a sample con-
fusion with F. labordi. On the other hand, the identical
ND sequences of F. antimena (KX) and F. belalan-
daensis (FJ) probably represent mitochondrial in-
trogression of F. antimena into F. belalandaensis according
to own, unpublished data.
In the nuclear CMOS gene, three alleles (haplotypes)
were identied in the bp analyzed, diering by – mu-
tations from each other. No haplotype sharing between
F.voeltzkowi and F. labordi was detected (Fig. ).
New data on males of Furcifer voeltzkowi
Morphology: With . mm snout–vent length and
mm total length, the recently collected male (ZSM
/) is larger than the studied F. labordi and all previ-
ously known males of F. voeltzkowi, which had a maximum
total length of mm (see S et al. and Table).
Although F. voeltzkowi males are larger than F. labordi,
their rostral appendage length is consistently smaller (ab-
solutely and relatively, see Table ).
Colouration in life (based on photographs of two males,
as shown in Figure , and observations in the natural habi-
tat): Basic colouration of body, extremities and tail bright
green; a white lateral stripe on enlarged tubercle scales that
are in a line can occur from the neck across ¾ of the length
of the body; the black interstitial skin can create a net-like
pattern, especially on the head; if stressed, dark green ver-
tical stripes appear on the body and the tail, also on the
casque and radially on the eyelids; further bright green/
yellowish dots contrast to the darker stripes; a black “C”
occurs in the neck region and another mirrored “C” later-
ally on the body; throat can be speckled with black spots
and the stripes on the casque and the anterior part of the
body can intensify to black; eyelids can be crossed by a
black stripe; supra- and infralabial scales can be brown/
black coloured giving the mouth a “ dirty-smeared” look;
rostral appendage is of the same colour as the head. Ac-
cording to our limited observations, the display coloura-
tion is moderate and not as conspicuous as in males of oth-
er chameleon species.
Hemipenis morphology (based on ZSM /): Both
hemipenes are fully everted (length . mm), the general
form of the hemipenis is subcylindrical and symmetrical
with a slightly bilobed apex; calyces on the truncus are dis-
tinct and clearly reduced on the sulcal side and towards
the upper truncus and pedicel, size of calyces rather small;
calyx ridges are smooth and not serrated; one pair of large
rotulae that are nely denticulated, tips of the rotulae get-
ting larger towards sulcal side; papillary eld of medium
sized, unpaired papillae on the sulcal side basal to the ro-
tulae; no other ornaments (e.g. eshy papillae, horns, pe-
dunculi) were observed. Since the hemipenis morphology
of F.labordi is unknown, a comparison of both species in
this respect is currently not possible.
Description of the rst known female
of Furcifer voeltzkowi
Snout–vent length . mm, tail length . mm, total length
. mm; rostral appendage very rudimentary and hard-
ly recognizable, tip of snout rounded; small and rounded
casque, elevated . mm from neck; lateral crest well devel-
oped, slightly curved; temporal crest indistinct, not meet-
ing highest point of casque; parietal crest present, length
. mm; occipital lobe absent; obvious axillary pits; gular
crest and ventral crest present, consisting of small conical
scales – still of white colour in preservation; dorsal crest and
345
Rediscovery of Furcifer voeltzkowi
Figure 1. A) Maximum Likelihood phylogenetic tree of Furcifer voeltzkowi, F. labordi, and several related species, based on DNA
sequences (695 bp) of the mitochondrial gene for ND. Furcifer verrucosus was used as outgroup. Numbers at nodes are bootstrap
support values in percent from an analysis with 1000 pseudoreplicates. e grey-shaded GenBank sequence originally assigned to
F.antimena probably is based on a misidentied F.labordi, whereas the sequence identity between F.antimena and F. belalandaensis
is probably due to mitochondrial introgression. B) Haplotype network constructed from DNA sequences of the nuclear CMOS gene
(234 bp) of three F.voeltzkowi and six F.labordi. C) Map with the four known locality records of F. voeltzkowi of which only Katsepy
is conrmed by DNA sequences, three locality records of F. labordi conrmed by DNA sequences, and the range map from the IUCN
Red List assessment for F. labordi (DOI: 10.2305/IUCN.UK.2011-2.RLTS.T8765A12929754.en) modied by S et al. (2018) to ex-
clude records referred to F. voeltzkowi. e base map follows the Madagascar Vegetation Mapping Project carried out from 2003–2006
(M S 2007); green is humid forest (rainforest), reddish tones are deciduous forest and spiny forest-thicket.
tail crest absent; tail length shorter than body length; pho-
lidosis homogeneous except for a few larger, circular scales
on body. Overall colouration of the preserved specimen is
a dark purple without any patterning except for the white
ventral stripe, continuing on the ventral side of the tail and
the inner surface of the extremities. e female contained
346
F G et al.
seven well-developed cream-whitish eggs (size given as
maximum length x maximum width in mm), four in the
right oviduct (. × ., . × ., . × ., . × .), and
three in the le oviduct (. × ., . × ., . × .).
Colouration in life (based on photographs, e.g. in Fig-
ure and observations of several females in the natural
habitat): Distinct sexual dichromatism, displaying colour-
ation of females is more colourful and more conspicuous
than in males; in a relaxed state body with bright green
ground colouration with dark green vertical stripes across
the body and tail, extremities and tail of same colour as
the body; dorsal part of the body and the casque can be
reddish-brown, also the supra- and infralabial scales; a row
of two or three dark to bright red spots is present dorso-
laterally along the anterior body; when stressed, females
radically change their colouration (Fig. ): the dark green
vertical stripes change to black; a broad lateral stripe oc-
curs running from the cheek to the basis of the tail and
develops a distinct violet colouration, contrasting well the
two or three red spots; the remaining body colouration, ex-
tremities and dorsal part of the tail and head are speckled
in black and white; throat with a red interstitial skin that is
shown when hissing with the mouth open.
Morphological dierences between female
Furcifer voeltzkowi and F. labordi
Due to the general similarities between the females of
F.voeltzkowi (Fig. ) and its sister species F. labordi (Fig.)
their dierences are described in the following, based on
the adult specimens ZSM / of F. voeltzkowi and
ZSM / and ZSM / of F. labordi.
Just like the males, the single studied female of F.voeltz-
kowi is distinctly larger than F. labordi with SVL of
. mm and TL of . vs. SVL of .–. mm and
TL of . mm, and observations of additional females
(not measured) in their habitat conrm the larger size of
F. voeltz kowi. Although F. voeltzkowi is the larger species,
it appears to be more slender as can be seen in the relative
head width (RHW) with . vs. .–. in F. labordi.
Also, the casque height is lower in F. voeltzkowi with .
(RCN) vs. .–.. Female F. labordi show a small ros-
tral appendage of .–. mm (RLRA .–.), whereas it
consists of only two scales in the studied female F. voeltz-
kowi (not quantiable). However, photographs of additional
F. voeltzkowi females show that the rostral appendage can
also be of similar size as in F. labordi. e temporal and pa-
rietal crest is shorter in F. voeltzkowi with . (RTCL) vs.
.–. and . (RPCL) vs. .–.. No dierences
could be found in pholidosis, but the colouration in life dif-
fers between females. In a relaxed state both species have a
green body colouration with green or light brown vertical
stripes (Figs B, C). In F. labordi, there is a series of ca. –
orange-brown spots along the vertebral column inbetween
these stripes and the upper side of the head is orange-brown
as well; further there is only one red spot laterally on the
neck. In F. voeltzkowi there are two or even three distinct
Table 1. Morphological characters of Furcifer labordi and F. voeltzkowi (combined data from S et al. 2018 and new measurements presented herein). All measurements in mm.
Abbreviations of variables are indicated in the Materials and methods; additional abbreviations: M = male, F = female, sa = subadult.
Collection nº Species Sex SVL TaL RTaL TL HW RHW LCL RLCL TCL RTCL PCL RPCL LRA RLRA WRA RWRA CH RCH CN RCN DC DCc NSL
ZSM 223/2018 F. voeltzkowi M 122.1 142.0 1.16 264.1 14.6 0.120 11.3 0.093 13.5 0.111 21.4 0.175 7.4 0.061 5.7 0.047 14.9 0.122 7.9 0.065 + + 14
SMF 16375 F. voeltzkowi M 101.3 133.5 1.32 234.8 11.4 0.113 10.5 0.104 12.2 0.120 20 0.197 7.2 0.071 4.6 0.045 10.3 0.102 7.4 0.073 + + 13
SMF 16377 F. voeltzkowi M 116.1 128.6 1.11 244.7 14.5 0.125 12.8 0.110 12.6 0.109 20.2 0.174 6.5 0.056 5.6 0.048 13.9 0.120 8.1 0.070 + + 14
SMF 16378 F. voeltzkowi sa M 88.4 100.3 1.13 188.7 10.4 0.118 9.8 0.111 5.7 0.064 13 0.147 6.7 0.076 4.3 0.049 8.9 0.101 5.6 0.063 + + 13
ZSM 213/2018 F. labordi M 90.4 94.9 1.05 185.3 14.1 0.156 8.1 0.090 11.4 0.126 16.1 0.178 8.2 0.091 4.3 0.048 12.2 0.135 9.8 0.108 + + 20
ZSM 210/2018 F. labordi M 97.5 103.2 1.06 200.7 12.9 0.132 9.6 0.098 12.4 0.127 17.1 0.175 9.7 0.099 4.9 0.050 18.5 0.190 7.9 0.081 + + 17
ZSM 211/2018 F. labordi M 90.5 112.5 1.24 203.0 13.2 0.146 8.9 0.098 13.4 0.148 19.5 0.215 10.8 0.119 5.4 0.060 21.3 0.235 9.4 0.104 + + 15
MNHN 5469 F. labordi M 103.2 118.0 1.14 221.2 12.0 0.116 10.6 0.103 13.8 0.134 19.4 0.188 9.3 0.090 6.2 0.060 15.2 0.147 9.1 0.088 + + 17
ZSM 875/2000 F. labordi sa M 86.7 89.4 1.03 176.1 10.0 0.115 6.0 0.069 10.4 0.120 14.5 0.167 5.8 0.067 3.8 0.044 11.5 0.133 7.8 0.090 + + 13
ZSM 782/2000 F. labordi sa M 71.9 73.7 1.03 145.6 8.0 0.111 5.9 0.082 5.8 0.081 10.6 0.147 4.0 0.056 3.1 0.043 6.8 0.095 3.6 0.050 + + 14
ZSM 222/2018 F. voeltzkowi F 75.9 73.9 0.97 149.8 9.2 0.121 6.1 0.080 5.1 0.067 8.3 0.109 – – – – 9.2 0.121 1.0 0.013 – – 14
ZSM 214/2018 F. labordi F 60.5 65.1 1.08 125.6 8.8 0.145 5.3 0.088 5.4 0.089 7.8 0.129 0.8 0.013 1.4 0.023 9.1 0.150 1.2 0.020 – – 14
ZSM 215/2018 F. labordi F 58.0 cut – – 8.9 0.153 4.6 0.079 5.4 0.093 7.3 0.126 1.2 0.021 1.8 0.031 9.0 0.155 1.3 0.022 – – 14
ZSM 25/1921 F. ‘voeltzkowi’ sa M 81.9 101.1 1.23 183.0 9.4 0.115 8.7 0.106 7.8 0.095 13.7 0.167 4.7 0.057 4.6 0.056 6.4 0.078 5.3 0.065 + + 14
347
Rediscovery of Furcifer voeltzkowi
Figure 2. Males of Furcifer voeltzkowi from Katsepy near Mahajanga in dierent colour states: A) slightly stressed; B, C) relaxed (sub-
adult); D, E) displaying. Note in (A) the injuries on the head probably resulting from intraspecic conicts. Males dier from those
of the sister species F. labordi by larger size, a lower casque, a shorter rostral appendage, darker skin around the scales and a row of
white and distinctly enlarged tubercles along the anks.
348
F G et al.
Figure 3. Females of Furcifer voeltzkowi from Katsepy near Mahajanga in dierent colour states: A) stressed/gravidity colouration; B,
C)rather relaxed; D, E) stressed. e females dier from those of the sister species F. labordi by larger size, oen by a rudimentary, poorly
recognizable rostral appendage, and several details of their colouration (e.g. 2–3 red dorsolateral spots versus only one in F. labordi).
349
Rediscovery of Furcifer voeltzkowi
red spots in a line laterally on the body and a pattern of al-
ternating white and brown crossbands is present across the
vertebral column and continued onto the head. When dis-
playing or showing gravidity F. labordi turn violet or blue
with black vertical stripes and black spots except for the se-
ries of orange-brown spots along the dorsal line. Stressed F.
voeltzkowi have a similar black pattern, but they only show
violet in a broad lateral stripe, contrasting the red spots; the
remaining body and the extremities are of a black ground
colour with intense white spotting.
Distribution
e two unambiguous distribution records of F. voeltzkowi
as summarized and mapped by S et al. () include
() the type locality “Antema [sic] an der Bembatukabai”
based on B (), which probably refers to the
coastal village Antrema (ca. -.°, .°, m
a.s.l.) or to the adjacent forest, and ()the coastal village
Soalala (ca. -.°, .°, m a.s.l.) based
on B (). Our observations conrm the as-
sumption of S et al. () that the locality Katsepy
(a coastal village in the Bombetoka bay) listed by B-
() as locality of Furcifer labordi, actually refers to
F. voeltzkowi. Photographic records of F. voeltzkowi from
two additional localities, which were originally identi-
ed as F. labordi, were recently published on iNaturalist
(https://www.inaturalist.org/taxa/-Furcifer-voeltz-
kowi) by M M. One of these records, made on
November , shows roosting juvenile and apparently
subadult males near the type locality Antrema (-.,
., precision of coordinates m), less than km
from the coast. e second record, made on November
, shows a roosting juvenile in the Tsiombikibo forest
(-., ., precision of coordinates m, es-
timated altitude ca. m a.s.l.), approximately half-way
between Katsepy and Soalala and at km distance from
the coast. An additional record without exact locality data
most likely referring to F.voeltzkowi was reported from the
newly protected area Antrema as Furcifer labordi (Associa-
tion Reniala ). Surprisingly, we could not nd any in-
dividuals in the relatively intact private dry forest below the
lighthouse near the type locality Antrema. In accordance
with R et al. () we also failed to record
F. voeltz kowi at all surveyed localities east of the Betsiboka
river. B (: ) noticed an unconrmed voeltz-
kowi record from Lakato ( km from Moramanga) in east-
ern Madagascar by A B (), which,
however, must be erroneous for biogeographic reasons.
Figure 4. Furcifer labordi from Kirindy, western Madagascar. A) adult male, in relaxed state; B) subadult male, stressed colouration;
C) adult female, relaxed; D) adult female in stressed/gravidity colouration.
350
F G et al.
Based on the reliable and precisely known records, the
elevational range of F. voeltzkowi is between sea level and
m above sea level. is pattern matches with its sister
species F. labordi, which also inhabits low elevation sites
(< m above sea level).
Conservation status
e available data support the assumption of S et al.
() that the distribution range of F. voeltzkowi might
extend at least km along the coast between Katsepy at
the Betsiboka river and Soalala at the Baly Bay, an area that
is characterized by moderately high reptile species rich-
ness (J et al. ). However, it remains unclear if
its distribution is largely continuous throughout its range
or scattered in isolated populations (which is more likely)
and if it is restricted to a narrow band along the coast or
also occurs more inland. It is also unclear if records of F.
tuzetae from coastal forests of Belambo and Sahamalaza
(R ), – km northeast of the Bet-
siboka, refer to F. voeltzkowi or another species. For these
reasons, its extent of occurrence (EOO) cannot be reliably
assessed at present, but might comprise ca. km² ac-
cording to the currently available data. us it is clear that
F. voeltz kowi is not a microendemic species and almost
certainly occurs in the newly protected area Antrema, of-
cially known as “Réserve de Ressources Naturelles du
Site Bioculturel d’Antrema” (G et al. ). Al-
though Antrema suers from anthropogenic pressures
such as slash-and-burn agriculture and wildres (G-
et al. ) its protection as a reserve might mitigate
these threats in the future. e observed high density of
F. voeltzkowi in the secondary vegetation at the border of
the village Katsepy demonstrates that the species is not
dependent on intact dry forests and is probably adapta-
ble to a certain degree of habitat degradation. We hypoth-
esize that one reason why this brightly coloured species
was overlooked for many decades is its assumed short life
span of only few months in the rainy season, which makes
it dicult and unlikely to record this species, since most
roads in its distribution range are not accessible in the wet
season.
We recommend that the extinction risk of this species
is assessed for the IUCN Red List of reatened Species.
According to our current state of knowledge F. voeltzkowi
might qualify as Endangered Bab(iii) under the Red List
Criteria of the IUCN () as the extent of occurrence is
estimated to be less than . km², the populations are
expected to be severely fragmented, and there is continu-
ing decline in the extent and quality of its habitat. How-
ever, with its relatively large extent of occurrence and an
ability to survive in secondary habitats, F. voeltzkowi does
not appear to be a species in urgent need of targeted con-
servation, but because its remaining habitat continues to
decline and fragment, it is likely to require conservation
in the long-term to prevent them from becoming more
threatened.
Natural history
We observed three adult males and numerous (ca. )
adult females of F. voeltzkowi at the end of the rainy sea-
son (between March and April ) on the property
of the hotel “Madame Chabaud”, which is a mosaic of gar-
den, open secondary forest and plantations. All individuals
were found roosting at night at estimated heights of – m
above the ground on branches of trees. e males showed
distinct signs (bite marks and smaller injuries) of interspe-
cic ghting and subjective signs of ageing, suggesting that
they were approaching the end of their life. On the oth-
er hand, the observation of small juveniles and subadult
males in November (published by M. M on
iNaturalist, see above) suggests that juveniles might have
hatched in October. ese data are still rather rudimen-
tary, but suggest that F. voeltzkowi might have a very short
lifecycle similar to its sister species F. labordi (K et
al. , E et al. ) although probably less ex-
treme since the prolonged rainy season in northwestern
Madagascar might support a slightly longer post hatching
life. Most observed females did not show obvious signs of
ageing and had well-developed eggs in their body, which
could be easily felt through the skin, suggesting that these
chameleons were still in the peak of the egg-laying peri-
od. According to the owner of the hotel, a much higher
density of F. voeltzkowi can be observed in February and
March, when the rainy season is at peak (Ms. C
pers. comm.).
Discussion
Specic distinctness of Furcifer voeltzkowi
According to the concept of integrative taxonomy deci-
sions on species delimitation should be based on dierent
and independent lines of evidence (e.g., P et al. ,
M et al. , H et al. ). Our re-
sults conrm the morphological dierences between males
of F. voeltzkowi and F. labordi observed by S et al.
() and revealed additional dierences in the morphol-
ogy and colouration of females. Furthermore, we found
substantial and concordant genetic dierentiation between
F. voeltzkowi and F. labordi in two mitochondrial and one
nuclear gene segments. us, the combination of four dif-
ferent lines of evidence (morphology, colouration, mito-
chondrial and nuclear DNA sequences) strongly supports
the specic distinctness of F. voeltzkowi.
Morphological dierentiation of
F. voeltzkowi and F. labordi
Apart from the larger size female F. voeltzkowi are relatively
similar to F. labordi, but dier, for example, by the number
of red spots on each body side (mostly two, but rarely three
in F. voeltz kowi versus only one in F. labordi). Similar to
the dierently coloured rostral appendages in the Ca-
351
Rediscovery of Furcifer voeltzkowi
lumma nasutum group (P ) these spots might
function as a signal for species recognition. Although both
species do not live syntopically, at the current knowledge,
such signals could have function as drivers for speciation
in the past, as was shown for anole lizards (I et al.
). Gravid females of F. voeltzkowi can show a spectac-
ular stress colouration, including pattern of black-, blue-
and white (Fig.). is colouration becomes very distinct
when females catch sight of a male, and is similar to that of
F.labordi, and – to a lesser extent – also to that of the Fur-
cifer lateralis complex. All these species are short living and
have rapid life histories with about – months from hatch-
ing to egg deposition (B a, b, K et al.
). is implements that there is only a short period
for mating and the courtship of males must be as ecient
as possible. In general, gravidity colouration functions as
a signal to males that they should not even approach; the
more conspicuous it is the better is the signal – even over
a great distance. As a result, the males can recognize grav-
id females earlier and can focus on receptive females. is
might work as a selective advantage for the species as both
sexes save energy during the mating season.
Another interesting dierence between F. voeltzkowi
and F. labordi is the size of the rostral appendage. In fe-
male F. voeltzkowi, it is rather variable, ranging from al-
most completely absent to small but distinct, as it is typical
for F. labordi. e reasons for this variability are unknown,
but could be due to individual hormonal dierences. A
rostral appendage length dierence is also evident be-
tween the males of both species, with F. labordi having
longer rostral appendages than F. voeltzkowi (Table ). It
remains to be claried in future studies, if there is a gen-
eral intraspecic correlation between rostral appendage
length of males and females in certain chameleon groups
with bony rostral appendage, which could contribute to
elucidate the evolution of ornamentation in chameleons
(to be studied in a forthcoming paper by S and col-
leagues). In certain species with a very pronounced male
rostral appendage (F. rhinoceratus, F. antimena) females
have a relatively distinct rostral appendage, whereas in
species with a shorter male rostral appendage (F. voeltz-
kowi, F. angeli) those of females are very rudimentary. In
any case, there seems to be a relatively strong and poor-
ly understood intraspecic variability in rostral append-
age length of males of Furcifer species with a single bony
rostral appendage (F. rhinoceratus, F. antimena, F. labordi,
F.voeltzkowi, F. angeli).
Biogeography
A large geographical gap of more than km separates
the currently known distribution ranges of the two ap-
parent sister species, F. labordi and F. voeltzkowi (Fig. ).
Knowledge of the reptile and amphibian communities in
this gap area is rather poor, especially in unprotected coast-
al forest rudiments, and new locality records could thus
lead to signicant range extensions of both species in the
future. A better geographical sampling is also necessary be-
fore a reliable biogeographic interpretation of the distribu-
tion of these chameleons can be drawn. For instance, their
occurrence may be inuenced by major river barriers (e.g.,
F. labordi apparently has not been found north of the Tsi-
ri bi hina river so far) or by speciation in watershed refugia
(see W et al. , V et al. ).
Potential of life history studies
on Furcifer voeltzkowi
Life history strategies in chameleons are very dierent,
ranging from long-living species, which can live for more
than nine years (T et al. ), to short-living species
like F. labordi (K et al. , E et al. ).
Furcifer labordi has an extreme and unique life history with
a posthatching lifespan of just – months, which is appar-
ently the shortest among tetrapods (K et al. ).
is annual chameleon lives mostly as an egg, at least in
the driest southernmost part of its range (K et al.
). A recent study on F. labordi in the less arid Kirindy
forest in west Madagascar revealed a similar but less ex-
treme life cycle (E et al. ). e available data
on the life history of F. voeltzkowi are still very rudimen-
tary, but support the assumption, that its general lifecycle
might be similar to its sister species F. labordi, although the
less arid climate in northwestern Madagascar might allow
for a prolonged lifespan. Due to the assumed correlation of
life history traits with climate in the F. labordi/F. voeltzkowi
species complex, future comparative studies on the life his-
tory of F. voeltzkowi might be an excellent model system
to increase our knowledge on the evolution of extreme life
histories and their constraints in tetrapods.
Necessary eorts to rediscover
further Malagasy chameleons
Herpetological research in recent decades has led to an
enormous progress of knowledge on Malagasy reptiles.
For instance, many new species of chameleons have been
described (e.g., G et al. , , C et al.
, F et al. , G et al. , , P
et al. , , S et al. ) and several endemic
reptile species not seen for more than a century have been
rediscovered, including the blindsnake Xenotyphlops gran-
didieri, which is the only representative of the family Xeno-
typhlopidae (W et al. , V et al. , W-
et al. ), and the scincid lizards Paracontias roth-
schildi and P. minimus (K et al. ). With the re-
discovery of Furcifer voeltzkowi we demonstrate that even
a brightly coloured chameleon can become a “lost species”
although it does not occur in a remote mountain massif,
but just a few kilometres from the large town of Mahajan-
ga. However, despite the substantial progress, numerous
Malagasy species are still known from only a single or very
few specimens, collected many decades ago and their con-
352
F G et al.
tinued existence is questionable: Furcifer monoceras is only
known from the adult male holotype collected years ago
(B, ) and characterized by a very long rostral
appendage not known from any other Malagasy chamele-
on. It was only recently resurrected (S et al. ), but
our short survey around the type locality Betsako did not
reveal any hints of its continued existence. Furcifer tuzetae
is another large chameleon species ( mm total length)
so far only known from the holotype, which was collect-
ed in in southwest Madagascar (B et al. ).
Not seen for many decades, a photograph of this species
has recently appeared on the internet (e.g., at http://www.
chameleondatabase.com/portfolio-item/furcifer-tuzetae/).
e dwarf chameleon Brookesia lambertoni, known from
only two type specimens probably collected in the year
at the unidentied locality “Fito” (B ) awaits its
rediscovery as well. Another example is the large snake
Pseudoxyrhopus ankanaensis, which is only known from
the holotype collected in (R N
) and is currently the only reptile species from Mada-
gascar considered as Critically Endangered (Possibly Ex-
tinct) on the IUCN Red List. However, the only diagnostic
characters of this species are based on the asymmetrical
head scalation of the holotype, suggesting that this individ-
ual might well represent an aberrant specimen of P. microps
rather than a distinct species.
In all these cases the lack of basic knowledge is evident
and is a crucial impediment to clarify the conservation sta-
tus of these species. is problem is not restricted to Mala-
gasy reptiles, but true for many lizards (M et al. )
and other species throughout the world. We therefore
greatly acknowledge the “Search for lost species Initiative”
of Global Wildlife Conservation and similar programs,
which signicantly contribute to ll the enormous gaps in
our knowledge of the extant biodiversity on our planet.
Acknowledgements
We are grateful to the Malagasy authorities for granting research
and export permits (research conducted under permit Nº//
MEEF/SG/DGF/DSAP/SCB.Re, specimens exported under
CITES permit NºC-EA/MG) and the German authorities
for granting import permits. We are greatly indebted to our two
sponsors, to Global Wildlife Conservation (“Search for Lost Spe-
cies” grant to FG) and to C Z (chameleon asset
management ltd). Both sponsors contributed of the costs for
the expedition. We also thank C Z for his per-
sonal eorts in participating in our rediscovery expedition and
A R for his great support in the eld.
Last not least we are very grateful to the organization MICET for
logistic support as well as to P-S G and R-
J for reviewing the manuscript.
References
A, J. R. B (): Compte rendu de Mission à
Madagascar (Octobre –Janvier ). – Bulletin du Mu-
séum national d’Histoire Naturelle, Paris, (ser. ), 35: –.
Association Reniala (): Plan d’Aménagement et de Gestion
de la nouvelle aire protegée d’Antrema. – Unpublished report
(available on the internet).
B, F. (a): Contribution à l’étude de la croissance post-
embryonnaire de Chamaeleo lateralis Gray, . – Annales de
l’Université de Madagascar, Sciences, 7: –.
B, F. (b): Le cycle reproducteur chez la femelle de Cha-
maeleo lateralis Gray, . – Annales de l’Université de Mada-
gascar, Sciences, 7: –.
B, O. (): Katalog der Reptilien-Sammlung im Muse-
um der senckenbergischen naturforschenden Gesellscha in
Frankfurt am Main. I. (Rhynchocephalen, Schildkröten, Kro-
kodile, Eidechsen, Chamäleons ). I–X, –.
B, O. (): Reptilien und Amphibien von Madagascar,
den Inseln und dem Festland Ostafrikas. – pp. –, Tafeln
–, in: V, A.: Reise in Ostafrika in den Jahren
–, Wissenschaliche Ergebnisse, dritter Band, syste-
matische Arbeiten. – Schweizerbartsche Verlagsbuchhand-
lung, Stuttgart.
B, M. W., O. H, J. F. Y. B T. B
(): Single-locus and multilocus DNA ngerprint. – pp.
– in: H , A. R. (ed.): Molecular Genetic Anal-
ysis of Populations: A Practical Approach. – Oxford: IRL
Press.
B, E. R. (): Reptiles Sauriens Chamaeleonidae. Genre
Chamaeleo. – Faune de Madagascar, 33: –.
B, E. R. (): Reptiles Sauriens Chamaeleonidae. Genre
Brookesia et complément pour le genre Chamaeleo. – Faune de
Madagascar, 47: –.
B, E. R., R. B C. A. D (): Notes
sur les Chamaeleo de Madagascar. C. tuzetae n. sp., nouvelle
espèce du sud-ouest (Reptilia, Squamata, Chamaeleontidae).
– Bulletin du Muséum national d’Histoire Naturelle, Paris,
(ser.), , 21: –.
B, E. R. C. A. D (): Les Caméléons a ros-
tre impair et rigide de l’ouest de Madagascar. – Mémoires du
Muséum National d’Histoire Naturelle, Série A, Zoologie 52:
–.
C, G., P. R. E R. D (): Biological an-
nihilation via the ongoing sixth mass extinction signaled by
vertebrate population losses and declines. – Proceedings of
the National Academy of Sciences of the USA, 114: E–
E.
C, A., A. M, F. G, D. J. H, A. L
M. V (): Description of a new pygmy chameleon
(Chamaeleonidae: Brookesia) from central Madagascar. –
Zootaxa, 3490: –.
E, F., P. M. K C. K (): Highly va-
riable lifespan in an annual reptile, Labord’s chameleon (Fur-
cifer labordi). – Scientic Reports, 7: .
F, A. M., C. M. I, H. A. R, E. E.
L C. J. R (): Detecting cryptic specia-
tion in the widespread and morphologically conservative car-
pet chameleon (Furcifer lateralis) of Madagascar. – Journal of
Evolutionary Biology, 25: –.
G, P.-S., M. P, F. M. R, J. K, M.
V F. G (): A Tarzan yell for conservation: a
new chameleon, Calumma tarzan sp. n., proposed as a agship
species for the creation of new nature reserves in Madagascar.
– Salamandra, 46: –.
353
Rediscovery of Furcifer voeltzkowi
G, P.-S., F. M. R, M. V F. G ():
Calumma vohibola, a new chameleon species (Squamata:
Chamaeleonidae) from the littoral forests of eastern Madagas-
car. – African Journal of Herpetology 60: –.
G, F. (): Taxonomic checklist of chameleons (Squamata:
Chamaeleonidae). – Vertebrate Zoology, 65: –.
G, F., J. K, T. M. T M. V ():
Rivaling the world’s smallest reptiles: discovery of mini-
aturized and microendemic new species of leaf chamele-
ons (Brookesia) from northern Madagascar. – PLoS One, 7:
e.
G, F., J. K M. V (): A distinctive new spe-
cies of chameleon of the genus Furcifer (Squamata: Chamae-
leo nidae) from the Montagne d’Ambre rainforest of northern
Madagascar. – Zootaxa, 2269: –.
G, S. M., M. J. R S. W (eds.)
(): Les aires protégées terrestres de Madagascar : Leur his-
toire, description et biote / e terrestrial protected areas of
Madagascar: eir history, description, and biota. – Associa-
tion Vahatra, Antananarivo.
H, O., Z. T. N F. G (): Island evolution
and systematic revision of Comoran snakes: why and when
subspecies still make sense. – PLoS One, 7: e.
H, D. (): e dierentiation within the genus
Chamaeleo Laurenti . – Beaufortia, 8: –.
H, D. (): Notes on Chameleons I. Comparative cy-
tology: aid and new complications in Chameleon-taxonomy.
– Beaufortia, 9: –.
I, T, A. H, D. L. M, M. R
C, R. E. G, A. H, Y. E. S J. B.
L (): Comparative tests of the role of dewlap size in
Anolis lizard speciation. – Proceedings of the Royal Society of
London B: Biological Sciences, 283: .
IUCN (): IUCN Red List Categories and Criteria: Version ..
– IUCN, Gland, Switzerland and Cambridge, UK.
IUCN (): e IUCN Red List of reatened Species. –
https://www.iucnredlist.org, assessed April .
J, R. K. B., M. F. T, P. B, N. C, J. L.
B, L. C, F. A, A. A, R. R.
A, M. A, P. B, L. D.
B, E. F. H, F. G, R. A. G, C.
H-T, M. H, V. K, N. H. R-
, J. R, D. R, H. R-
, N. A. R, J. R,
J.-B. R, H. R, J. C. R-
, H. H. R, J. E. R-
, H. R, A. P. R, A. R-
, F. M. R, C. J. R, E. R-
, F. R, P. P. D, A. D. Y
M. V (): Extinction risks and the conservation of
Madagascar’s reptiles. – PLoS One, 9: e.
K, K. B., L. N. A, S. F. F C. J.
R (): A unique life history among tetrapods:
An annual chameleon living mostly as an egg. – Proceedings
of the National Academy of Sciences of the United States of
America, 105: –.
K, C. W. B (): Phylogeny and classication of
the Chamaeleonidae (Sauria) with special reference to hemi-
penis morphology. – Bonner zoologische Monographien, 22:
–.
K, J., M. V, M. E F. G (): System-
atics of limbless scincid lizards from northern Madagascar:
morphology, phylogenetic relationships and implications for
classication (Squamata: Scincidae). – Organisms, Diversity &
Evolution, 10: –.
K, E. (): e Sixth Extinction: An Unnatural History.
– Henry Hold and Company.
K, S., G. S K. T (): MEGA: Molecular
Evolutionary Genetics Analysis version . for bigger datasets.
– Molecular Biology and Evolution, 33: –.
L, B. B., E. C. M, M. K. R J. J. W ():
Inordinate fondness multiplied and redistributed: the number
of species on Earth and the new pie of life. – Quarterly Review
of Biology, 92: –.
M, S., A. M. B, A. A, F. C-H, L.
C, G. C, I. D, T. M. D, F. G, L. L. G-
, M. H, F. K, M. LB, D. M, Z.
T. N, C. C. N, P. O, O. S. G. P,
D. P-D, G. S, R. S, O. J. S. T-
, O. T-C, J.-F. T, P. U, P. W-
, Y. W, T. Z U. R (): Extinct, obscure
or imaginary: e lizard species with the smallest ranges. –
Diversity & Distributions, 24: –.
M, R. (): Chamaeleonidae. – Das Tierreich, 83: I–X.
–.
M, A, R. V, A. P, D. J. H S.
C (): An integrative taxonomic revision of the
Cape Verdean skinks (Squamata, Scincidae). – Zoologica
Scripta, 40: –.
M, J. P. S () (eds): Atlas of the vegetation of Mada-
gascar. – Kew Publishing, Kew, UK, pp.
P, J. M., A. M, I. D R M. V ():
e integrative future of taxonomy. – Frontiers in Zoology, 7:
.
P, S. R. (): Observations on the natural histories of six
Malagasy Chamaeleontidae. – Zeitschri für Tierpsychologie,
34: –.
P, S. L., C. N. J, R. A, T. M. B, J. L. G-
, L. N. J, P. H. R, C. M. R J. O. S
(): e biodiversity of species and their rates of extinction,
distribution, and protection. – Science, 344: .
P, D., M. D. S, F. M. R, M. V F.
G (): Untangling the trees: Revision of the Calumma
nasutum complex (Squamata: Chamaeleonidae). – Vertebrate
Zoology, 70: –.
P, D., M. V, M. D. S, D. R. V F.
G (): Splitting and lumping: An integrative taxonom-
ic assessment of Malagasy chameleons in the Calumma guibei
complex results in the new species C. gehringi sp. nov. – Verte-
brate Zoology, 67: –.
R, A., J. E, F. M. R M. V
(): Amphibian and reptile records from around the Betsi-
boka delta area in North-Western Madagascar. – Herpetology
Notes, 8: –.
R, A. P. (): Herpétofaune des forêts sèches
malgaches. – Malagasy Nature, 1: –.
R, C. J. R. A. N (): A review of the
Madagascan snake genera Pseudoxyrhopus, Pararhadinaea,
and Heteroliodon (Squamata: Colubridae). – Miscellaneous
354
F G et al.
Publications, Museum of Zoology, University of Michigan,
182: –.
R, C. J., M. R. J. F R. A. N ():
Chameleon radiation by oceanic dispersal. – Nature, 415: –
.
S, M. D., J. K, A. R, F. G M.
V (): A new dwarf chameleon, genus Brookesia,
from the Marojejy massif in northern Madagascar. – Zoosys-
tematics and Evolution, 95: –.
S, M., Y. C, M. D. S, D. P F. G
(): Rising from the ashes: resurrection of the Malagasy
chameleons Furcifer monoceras and F. voeltzkowi (Squama-
ta: Chamaeleonidae), based on micro-CT scans and external
morphology. – Zootaxa, 4483: –.
T, G., F. G F. A (): Longevity in Calum-
ma parsonii, the world’s largest chameleon. – Experimental
Gerontology, 89: –.
T, K. A., T. M. T M. V (): Large-scale
phylogeny of chameleons suggests African origins and Eocene
diversication. – Proceedings of the Royal Society of London
B, 280(): .
V, M., K. C. W, D. R. V & D. C. L
(): Madagascar as a model region of species diversica-
tion. – Trends in Ecology and Evolution, 24: –.
V, N., J. M, M. M, S. D, W. R. B,
R. T, M. V, A. W, C. C S. B. H
(): Blindsnake evolutionary tree reveals long history on
Gondwana. – Biology Letters, 6: –.
W, D. B. V. T. V (): Are we in the midst of
the sixth mass extinction? A view from the world of amphib-
ians. – Proceedings of the National Academy of Sciences of the
United States of America, 105 (Suppl. ): –.
W, V., V. M F. A (): Rediscovery
of the enigmatic blind snake genus Xenotyphlops in northern
Madagascar, with description of a new species (Serpentes: Ty-
phlopidae). – Zootaxa, 1402: –.
W, J. E., S. S, O. H, M. F, V.
W, M. V, Z. T. N, S. B. H, J. K
F. G (): Morphological variation and taxonomic reas-
sessment of the endemic Malagasy blind snake family Xenoty-
phlopidae (Serpentes, Scolecophidia). – Spixiana, 36: –.
W, L., S. M. G & J. U. G (): Biogeo-
graphic evolution of Madagascar’s microendemic biota. – Sci-
ence, 312: –.
Supplementary data
e following data are available online:
Supplementary Video 1. Furcifer voeltzkowi in the natural habitat
(FG and DP): https://youtu.be/Exen2NIb_bw
