Pinocchio lizards and other lizards bearing rostral appendages
Open access at hps://www.salamandra-journal.com
© 2022 Deutsche Gesellscha für Herpetologie und Terrarienkunde e.V. (DGHT), Germany
15 May 2022 ISSN 0036–3375
SALAMANDRA 58(2): 123–138 SALAMANDRA
German Journal of Herpetology
Pinocchio lizards and other lizards bearing rostral appendages −
the peculiar habitus of the draconine agamid Harpesaurus tricinctus
with highlights on its ecological implications and convergence
with its New World equivalent, the dactyloid Anolis proboscis
I I, T K, W B
1) Institut de Systématique, Évolution, Biodiversité (ISYEB), Muséum national d‘Histoire naturelle, Sorbonne Université,
École Pratique des Hautes Études, Université des Antilles, CNRS, CP 30, 57 rue Cuvier 75005 Paris, France
2) Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, 53113 Bonn, Germany
3) Natural History Museum, University of Oslo, P.O. Box 1172 Blindern, 0138 Oslo, Norway
* Corresponding author: I I, ORCID-ID: 0000-0003-1235-1505, e-mail: firstname.lastname@example.org
Manuscript received: 3 November 2021
Accepted: 27 February 2022 by P W
Abstract. Harpesaurus tricinctus is an Asian agamid lizard described in from Java, Indonesia, and since then known
only from its holotype located at the Paris Natural History Museum (MNHN-RA), supposedly a male, characterized by a
long sickle-shaped rostral appendage. Ecological data are virtually lacking since no other specimen have ever been found.
Here we review its morphology as compared with its congeners. We also review the morphology of other lizards and some
snakes with rostral appendages and discuss their possible functional implications. We identied a South American dacty-
loid lizard, Anolis proboscis, the males of which show by their general habitus and a long, sword-shaped rostral append-
age a striking convergence with the possibly extinct H. tricinctus. Anolis proboscis was also for long considered a rare and
little-known lizard but recent eld work allowing new observations provided additional data on its ecology. We suggest to
use these new data on its habitat requirements which might be comparable to those of the externally so similar but alleg-
edly extinct Indonesian agamid which might have survived in the canopy of tropical humid forests in one of the Greater
Sunda Islands of Indonesia.
Key words. Squamata, rostral appendage-bearing (Pinocchio) lizards, convergent evolution, functional implications, trop-
ical rain forest, arboreality.
Head ornamentation, here the presence of rostral or su-
praorbital protuberances, is present in numerous ani-
mal groups. In avian reptiles, such bony structures like
those seen in some bird groups, e.g. the rhinoceros birds
(Buceroti dae), have a wide range of elaborate bony struc-
tures surmounting the prominent bill with dierent
shapes and colors (K ). Various theories suggest-
ed functions for the casques of hornbill. Some investiga-
tions even attributed a possible acoustic function for that
casque structure (A et al. ). Rostral and su-
praorbital outgrowths are also present in anuran amphib-
ians as well as in squamate reptiles. In frogs of the families
Megophryidae and Cyclorhamphidae for instance, it has,
according to C et al. () and M et al. (),
a somatolytic camouaging function, but some species of
the pelodryadid genus Litoria T, have sexually
dimorphic eshy spikes at their snout tips which are in-
terpreted as partly somatolytic and partly also responsible
for mate recognition (M J , Oliver
et al. a, b). Also in squamates, rostral appendages can
play a role in camouaging, or, particularly when conned
to males only, can have a function in sex-specic species
recognition, thus being sexually selective (O S-
In this paper, we focus on rostral appendages in lizards
while nasal appendages are rarely observed in snakes. Apart
from some viperids, such appendages (not only specialized
rostral shields) occur in the Neotropical Philodryas baroni
B, (Dipsadidae), as well as in the Oriental Gonyo-
soma boulengeri (M, ) and G. hainanense
P, Z, H, B W, (Colubri-
dae) as well as in some vine snakes (Colubridae), particu-
larly prominent in Ahaetulla pulverulenta (A. M. C. D-
, B A. H. A. D, ), but also devel-
I I et al.
oped to a lower degree in other congeners, e.g. A. anoma la
(A, ), A. laudankia D, N,
D M, , A. oxyrhyncha (B, )
or A.sahya drensis M, A, G, P,
V S, . Rare cases are the paired
antenna-like rostral appendages in the homa lopsid snake
Erpeton tentaculatum L, which serve as sen-
sitive mechanoreceptors under water (W , F
, C et al. ), or the acutely pointed ros-
tral scale in the male sea snake Emydocephalus annulatus
K, and congeners used to stimulate the female
during copulation (H , G , F ).
A special case is the Malagasy pseudoxyrhophid genus
Langaha B, (G , ), bearing
a long even sexually dimorphic, leaf-like nasal appendage
whose function is still not yet fully understood (K
, T ).
Apart from pathological nasal protuberances (e.g. M-
-S et al. for a lacertid lizard), the elon-
gated paired, funnel-shaped nasal scalation of the gekko-
nid Rhinogekko misonnei ( W, ) (M et al.
) and the nasal protuberance of the male gharial Gavia-
lis gangeticus (G, ) (B et al. ) are men-
tioned here just for completeness’ sake. In lizards, rostral
appendages or protuberances are conned to the iguanian
families (Iguanidae sensu lato, Agamidae and Chamaeleo-
nidae) which will be discussed in more detail below.
A particular wide variety in head ornamentation and
appendages can be found among Asian agamid lizards
(S et al. , O S-F ). e her-
petological collection of the National Museum of Natural
History in Paris (MNHN-RA) houses a unique specimen of
a bizarre Asiatic draconine lizard bearing a long, falciform
(sickle-shaped) rostral appendage (Fig. ). e lizard was
described in a new genus as Arpephorus tricinctus A. H. A.
D, (Arpephorus being preoccupied by a beetle
and replaced by Harpesaurus B, ). is is the
only known specimen, and because the species has never
been found again since its description (A. H. A. D
in A. M. C. D A. H. A. D, ), it is re-
garded as likely extinct. e ve other species of the genus
Harpesaurus, viz. H. beccarii (D, ), H. borne ensis
(M, ), H. brooksi (P, ), H. ensicauda
F. W, , and H. modiglianii V, are
all known from just a very few specimens. With the ex-
ception of H. borneensis (L et al. , M
), their ecology remains a mystery. e allocation of
the nomen brooksi, described in a new monotypic genus as
aumatorhynchus brooksi to Harpesaurus was proposed
already by V () who was followed by W-
(), K (), L et al. (), M-
(), and U et al. (). Only M ()
suspected a closer anity to the genus Aphaniotis.
We here compare this unique and only known specimen
of H. tricinctus with all other lizards bearing a rostral ap-
pendage (mostly only in males) and show that one lizard
species from a distinct, geographically remote, but likewise
iguanian family (Dactyloidae), viz. Anolis proboscis P
O, , shows a striking convergence in habitus
and color pattern. Such a spectacular convergence between
a Neotropical iguanid and an Asian agamid in perch use
and limb lengths had been already indicated for the Carib-
bean Anolis F, and the Southeast Asian y-
ing agamid Draco L, by O K ().
In both genera interspecic competition shaped eco-
behavioral and morphological dierentiation. ere are
numerous cases of such morphologically similar species
“pairs” among iguanian families living on separate conti-
nents, e.g. the lizard genera Sauromalus A. H. A. D,
(Iguanidae) and Uromastyx M, (Agami-
dae), Basiliscus L, (Iguanidae) and Hydrosau-
rus K , (Agamidae), or Phrynosoma W,
(Phrynosomatidae) and Moloch G, (Agami-
dae). Almost a textbook example among snakes, the im-
pressive convergence of the Neotropical Corallus caninus
(L, ) (Boidae) and Australo-Melanesian Mo-
relia viridis (S, ) (Pythonidae) (E
K ). All these examples show great similarities
in their convergent habitat selection (see also P et
Figure 1. Harpesaurus tricinctus in color. Reproduction of the holotype on vellum (MNHN vellum collection).
Pinocchio lizards and other lizards bearing rostral appendages
Since ecological data on other species of Harpesaurus
are scarce or even completely missing (see below), we sug-
gest to use the available data on the ecology of A. proboscis
to provide inferences on habitats in which a rediscovery
of H. tricinctus in Indonesia nearly decades aer its de-
scription might be more likely or even possible.
Results and Discussion
History of the unique
Harpesaurus tricinctus specimen
In the “Catalogue Méthodique de la Collection des Rep-
tiles”, A. H. A. D in A. M. C. D A. H.
A. D (: –) described the monotypic new
aga mid genus Arpephorus characterized by its long ros-
tral sickle-shaped appendage that reminded him of the
form of a harp (Latin arpe = harp) and by a distinct tym-
panum whereas the genera Ceratophora G, and
Lyrio cepha lus M, , likewise equipped with na-
sal protuberances, have a hidden tympanum. e unique
specimen, holotype by monotypy, which is credited as hav-
ing been collected on Java, is described as a new species,
Arpe phorus tricinctus, based on the three broad, light yel-
lowish bands on its back. In a second paper which A. H.
A. D () published alone, he indicated that he
previously provided the diagnosis of his new genus Arpe-
phorus in the “Catalogue Méthodique” published in April
(A. M. C. D A. H. A. D ), thus
approving that the rst-cited publication from the same
year was actually earlier. A. H. A. D (), rede-
scribing both the genus and species a second time, stat-
ed that the unique available specimen was acquired sev-
eral years earlier (no precise date indicated) as originating
from Java and that it was present in the Muséum national
d’Histoire naturelle (Paris) collections (MNHN-RA
[former tag number ]); it was discovered desiccated
and pinned dry in a box with insects [the hole made by the
pin is visible on the back of the lizard; Fig. ]. Note that in
a recent checklist on amphibians and reptiles of Java and
Bali, K et al. () erroneously located that holo-
type at the Natural History Museum in London (UK). No
collector name was indicated and none could be found in
any of the MNHN-RA catalogues. Further no indication of
any such lizard collected from Java (even without collec-
tor name) prior to was found in any available MNHN
catalogues that we have checked (handwritten sh & rep-
tile catalogues –, –; handwritten reptile
gis catalogue –). Additionally, both handwritten
and catalogues indicated “Java” as the location for
the lizard and the later catalogue also said that an illustra-
tion on vellum had been made to illustrate the species (see
Fig. ; MNHN vellum collection). is lizard was designed
and painted by M-F B (–). It is
“arranged” with two other colored drawings of B
among the MNHN vellums, Portfolio , No. . is is the
same drawing used by the engraver L. L to make the
plate VII illustrating the original publication.
As Arpephorus A. H. A. D, was preoccu-
pied by Arpephorus S, (Coleoptera), a new
genus, Harpesaurus B, , was created to ac-
commodate the species previously referred to the former
genus (B ). e genus Harpesaurus is char-
acterized by its tympanum concealed by a scaly plate (in the
type-species H. tricinctus and H. borneensis but not in other
species generally included in the genus) and a snout ending
in a long sometimes compressed (round in cross section in
H. brooksi), cutaneous (not scaly) appendage. Body and tail
are compressed, covered above with equal smooth scale of
moderate size while ventral and tail scales are keeled. Ex-
cept in H. ensicauda which lacks a dorsal crest, a more or
less developed dorsal crest is oen present on back and tail.
No gular sac, no transverse gular fold, nor femoral or pre-
anal pores are present (modied aer B ).
B () gave the following diagnosis for
H.tricinctus: “Elongated rostral appendage longer than the
head, formed of a unique compressed falciform scale curved
upwards, surrounded at the base by a few large scales. Up-
per head-scales small, slightly tubercular; a large promi-
nent triangular scale on the snout; canthus rostralis form-
ing a serrated ridge. Gular scales tubercular. Dorsal crest
a low serrated ridge. Dorsal scales smooth, equal, forming
regular transverse series; ventral scales keeled. Scales on the
limbs keeled. Tail transversally compressed, crested above,
the crest a little higher than the dorsal; caudal scales keeled,
the inferior spinose. Brown, with three broad transverse
yellow bands on the body, the anterior narrowest on the
scapular region”. Snout–vent length of the unique known
specimen (holotype by monotypy) is and mm and
Figure 2. Detail of the back of the holotype of H. tricinctus show-
ing the black hole where the insect pin passed to conserve the
dried lizard before it was discovered.
I I et al.
tail length and mm according to B ()
and B (), respectively (Fig.).
Since its description in , the species has never been
found again. Moreover, its alleged geographic origin from
Java cannot be conrmed since no collector name is as-
sociated with the specimen. e reproductive mode of
H.tricinctus cannot be veried; although oviparity is most
common among Asiatic draconine agamids, viviparity is
known only from the two species of the Sri Lankan endem-
ic genus Cophotis P, , viz. C. ceylanica P,
and C. dumbara S, R, R-
, A, O, R S-
, (S et al. , M-
-A et al. , see H
B ), and in one of the congeners of H. tricinctus,
viz. H. borneensis (L et al. , M ,
K et al. ).
e other species of Harpesaurus
e genus Harpesaurus as recognized today comprises
six species conned to the Greater Sunda Islands. Next to
H.tricinctus (likely originating from Java) these are: H. bec-
carii from Sumatra, H. borneensis from Borneo (Sarawak
and Kalimantan), H. brooksi from Sumatra, H. ensicau-
da from Nias (Mentawai archipelago), and H. modiglia-
nii again from Sumatra. All of them are known by a small
number of individuals only, in one of them (H. ensicauda)
even the single type specimen is lost, and ecological data, if
any, are extremely scarce. So, nearly all these species belong
to the group of “extinct, obscure or imaginary lizard spe-
cies” as recently dened by M et al. (). We rapidly
indicate their main characters below:
Harpesaurus beccarii is characterized by a double rostral
appendage in the male, viz. a lower median recurved one,
and a second straight process above it. It is bright green
in life with two short white, oblique stripes on the side of
head and neck. Its body is laterally compressed and has
separate nuchal and dorsal crests composed of large trian-
gular scales; the caudal crest composed of lanceolate scales
is separated from the dorsal crest. e strongly compressed
tail forms in its rst half a little sail supported by elongat-
ed neural spines of the proximal tail vertebrae. e rough
drawing reproduced from D () by R ()
is rather imprecise in respect to all these characters. Photo-
graphs of a living male can be found in B () and
M (), both showing the same individual.
Figure 3. Holotype of Harpesaurus tricinctus (MNHN-RA 0623), drawing from B (1988: 49).
Pinocchio lizards and other lizards bearing rostral appendages
Harpesaurus borneensis was originally described as the
only species of a monotypic genus Hylagama M,
, based on a single female specimen. While W
() and M () maintained this generic alloca-
tion, M () considered Hylagama as a synonym
of Harpesaurus. K’s () new species Harpesaurus
thelescorhinus was believed to be identical with M’
() species (M , L et al. , M-
). According to the drawing of the type speci-
mens given by K (), the male holotype of his new
species had a very long cylindrical (not compressed) curved
rostral appendage, exceeding the head length, while the fe-
male paratype was hornless. L et al. () re-ex-
amined K’s two type specimens deposited in the Chica-
go Natural History Museum (CNHM = FMNH). e male
holotype proved to have a much smaller rostral appendage,
reaching just the eye when laid backwards, while the fe-
male was hornless, as is also M’ () female holo-
type of his H. borneensis. However, L et al. ()
were able to discover more females in Sarawak which had
also (though much shorter) rostral appendages. So, the
presence or absence of rostral appendages is either variable
in this species or casts doubt on their conspecity as ongo-
ing examination will check (D et al. ). Moreover,
two of these females were found to be highly gravid with
two rather large, well-developed fetuses each. More ecolog-
ical data were provided by K et al. ().
Harpesaurus brooksi was originally described as the sole
representative of a new genus aumatorhynchus erected
by P (), the synonymy of which with Harpe-
saurus was claimed by V (), W
(), K () and M (). It is the second
(of three) Sumatran species of Harpesaurus, characterized
by a cylindrical rather than laterally compressed rostral ap-
pendage, a small dorsal and no caudal crest.
Harpesaurus ensicauda has no broad bands on the
body but its tail is strongly compressed with a crest above
(W , R ). It is only known from its
male holo type (W , W ) which is
today obviously lost. e only morphological characteri-
zation is that of W () himself, referred to by
R (). Its rostral appendage is shorter than the head,
curved backwards and surrounded at the base by two en-
larged scales. It has no dorsal crest, but a strongly com-
pressed tail with denticulate projections on its upper mar-
Harpesaurus modiglianii, originally also known only
from its single holotype collected in , has recently been
rediscovered by P et al. () aer years. ey
recorded the nd of two male specimens not far from the
type locality in northern Sumatra, one encountered dead
on the forest oor, and another one alive, which was re-
leased aer photography. Both males agreed well with the
type specimen in having also a single rostral appendage.
ree characters make H. tricinctus unique within this
rather diverse genus, the monophyly of which has still to be
demonstrated (D et al. in prep.): () its tympanum is
hidden by a scaly plate (like in H. borneensis), () no other
agamid species has such an extended proboscis-like rostral
appendage and also () none possesses its typical threefold
banded black and yellow body pattern.
Other rostral appendage-bearing Iguania
Rostral appendages occur across the entire clade of igua-
nian lizards, particularly in many species of African and
Malagasy chameleons, ranging from broad, at, single or
double scale-covered lobes to as many as six pointed, an-
nulated horns on the snout tip. ese rostral projections,
sometimes forming annulated horns which can be com-
bined with likewise annulated preorbital horns, are either
used as visual signals, sometimes also as weapons in ago-
nistic male contests on tree branches (E , J-
et al. , S-F ), or as cues for sexual
recognition (R , P , B K
), functions which will be discussed below.
Rostral appendages are moreover present in several oth-
er Asiatic agamids and some Neotropical iguanians. In the
former they are considerably more variable in shape, num-
ber and squamation than those of the few proboscis-bear-
ing anoles (W ).
Other agamid lizards with rostral appendage
Also, the small tree-dwelling draconine agamids of the Su-
matran genus Aphaniotis P, have two species
with rostral appendages in males: A. acutirostris M-
, and an undescribed species from the Menta-
wai Islands southwest of Sumatra, Aphaniotis sp. ( R
, L ). e appendages are small, scaly
ere is another small draconine agamid lizard from
Sumatra (?) and Java, where the males have also a rostral
appendage which is, however, very tiny and not compara-
ble with those found in the genus Harpesaurus or in the Sri
Lankan draconine radiation dealt with below. It is Pseudo-
cophotis sumatrana (H, ) a likewise extreme-
ly rare arboreal lizard originally assigned to the genus Co-
photis. A rather good line drawing of the male holotype
from Leiden collections is given by R (). An-
other male specimen stored at the Senckenberg Museum
in Frankfurt/Main (SMF , former SMF a) is re-
ported by M () and gured in H
B (). ere are also two females known. e
minute rostral appendage looks like a rudiment and has
likely no functional meaning in an ecological or behavioral
context. A further male specimen was recently recovered
by D et al. ().
One Asiatic draconine agamid clade (K et al.
) endemic to Sri Lanka includes two lineages where
the members have rostral appendages. One is Lyriocepha-
lus scutatus (L, ), with a knob-like rostral ap-
pendage, neither comparable to that of H. tricinctus nor
to any other rostral appendage-bearing lizard. L. scuta-
I I et al.
tus is an oviparous arboreal species with a round, bulbous
scale complex forming its rostral appendage (S et
al. ). It has a developed gular sac (similar to anoline
throat fans), orange-yellowish in life, whereas its back is
green with bluish anks, and more prominent in the male
which also possesses a nuchal hump. It is found in well-
shaded natural forests and plantations (less in home gar-
dens) in the wet and intermediate zones below m a.s.l.
(K A ).
e second lineage of this clade endemic to Sri Lan-
ka is Ceratophora G, , composed of six oviparous
species lacking dorsal crests. ey are characterized by
the presence of a simple rostral appendage in the form of
a modied rostral scale or a complex rostral appendage
comprising several scales, sometimes together with post-
rostrals; the appendage is prominent in males (except in
some male specimens of C. erdeleni). e genus comprises
Ceratophora aspera G, [low vegetation dwell-
ing species with horn-shaped cylindrical appendage cov-
ered with pointed scales, absent in juveniles and dimorphic
in adults], C. erdeleni P M-
A, [subarboreal species with rudimentary or
absent appendage], C. karu P M-
, [ground-dwelling species with rudimentary or
absent appendage], C. stoddartii G, [subarboreal
species with pointed horn-shaped appendage restricted to
rostral scale, absent in juveniles and dimorphic in adults],
C. tennentii G, [arboreal species with a later-
ally compressed leaf-shaped appendage, elliptical and cov-
ered with granular scales present in juveniles and mono-
morphic in adults], and the recently described C. uku-
welai K, P, A, S-
, B, M, G S,
[rare and elusive species found on the forest oor in dense
forest patches, bearing a long, complex and rough rostral
appendage in males]. In general, horns of male Ceratopho-
ra are larger than female horns and are movable (P-
M-A ). It seems that
the leaf-shaped appendage of C. tennentii plays a role in
crypsis whereas sexual selection seems involved in other
species of the genus. e appendage is present in juveniles
and monomorphic (no sexual dimorphism) in adults of
C.tennentii (W et al. ). In C. aspera and C. stod-
dartii, it is horn-shaped, eectively absent in juveniles and
dimorphic (sexual dimorphism present) in adults (J-
et al. ). ese dierences suggest that the append-
age evolved independently in C. tennentii vs. C. aspera and
C. stoddartii. While the phenotypic similarities between
C. aspera and C. stoddartii might be consistent with ho-
mology, such homology is however refuted by the wide
separation of these two species in the phylogeny obtained
by J et al. (). eir combined phenotypic and
phylogenetic evidence suggests that rostral appendages
evolved independently at least three times in Ceratophora
(see also P M-A ,
S et al. , K et al. ). Another
possible explanation for the existence of such ornaments
is that rostral appendages in Ceratophora improve crypsis
by breaking up the lizard’s outline (S , S-
-F O ). is hypothesis would predict that
the appendage should be present in both juvenile and adult
lizards, and would not require that it is sexually dimor-
phic among adults. is may apply to C. tennentii, a slow-
moving arboreal lizard of the cloud forests (S
, P M-A ).
e crypsis hypothesis is unlikely to apply in C. aspera and
C. stoddartii because in these taxa the rostral appendage
is sexually dimorphic (J et al. ). e absence
of a rostral appendage in juveniles and its sexually dimor-
phic nature in two species of the genus rather suggest that
it may function as a sexually selected ornament. However,
although the appendage is horn-like in both taxa, there are
some important dierences concerning allometry relative
to sex: a strong positive allometry was identied in males
only. is suggests that positive sexual selection may be
acting to increase the size of the appendage in male C. as-
pera which have smaller snout–vent lengths than females.
Both male and female C. aspera are cryptically colored
against the leaf litter in the lowland forest where they live
(P M-A ). e
dimorphic rostral appendage in this species may most like-
ly function in mate or rival recognition (J et al.
). While C. aspera is widely distributed in the lowland
moist forests in Sri Lanka’s south-western wet zone, all the
other species in the genus are restricted to areas of undis-
turbed cloud forest between and m above sea lev-
el (P M-A ).
ere is one more group of Asian agamids not belong-
ing to the subfamily Draconinae, where particularly the
males bear a kind of rostral excrescence, viz. the sailn aga-
mas of the genus Hydrosaurus K, . It is not really
an appendage, rather it is a longitudinally directed hump
on the snout (D et al. ) which shapes, however,
the head silhouette of these big-growing lizards and may
therefore play a role in the social behavior of these lizards.
Chameleons with rostral appendage
e second group of acrodont iguanians with numerous
species adorned with rostral appendages are the chamele-
ons, highly derived, primarily arboreal lizards character-
ized by bulbous, independently moving eyes, specialized
grasping feet, prehensile tails and a unique, body-long pro-
trusible tongue. ey are concentrated in the Afro-Mala-
gasy realm with a few species entering the southwestern
Palearctic and Oriental regions, and are biologically di-
verse (T H ). Of the eleven genera cur-
rently recognized, one ground-dwelling (Palleon G ,
H R, ) and two tree-
dwelling genera (Calumma G, and Furcifer F-
, ) in Madagascar (G ), and likewise
one ground-dwelling (Rhampholeon G, ) and
two tree-dwelling (Kinyongia T, T B,
and Trioceros S, ) genera in Africa (T-
) contain species with rostral appendages.
Pinocchio lizards and other lizards bearing rostral appendages
In Madagascan tree-dwelling species (Calumma, Furci-
fer) these appendages are either unpaired, compressed or-
naments which are partly moveable and so, partly sti
and supported by a bony rostral process, or forked, like-
wise bone-supported scaly, diverging processes (B-
D , B , G V
). Also in many of the African tree-dwelling species
(Kinyongia, Trioceros) the males (sometimes also the fe-
males) bear rostral appendages. ese can be dierently
shaped single, unpaired scaly projections with or with-
out internal bony support, i.e. exible or sti, or they can
form a paired bone-supported structure (T ).
In several species of Trioceros, the head ornamentation
consists of annulated horns, most oen a single rostral
one, supplemented by two preorbital ones. ese are sup-
ported by bony cones covered by a keratinized, annulated
sheath, comparable to the horns of cavicorn (i.e. bovid)
mammals. In Trioceros conirostratus (T, ), a
seemingly rudimentary movable hornlike appendage on
the snout tip (T ) strongly resembles that of
Pseudocophotis sumatrana males (see above). One Calum-
ma species group, however, viz. the C. gallus group, has
a long sword-like rostral appendage (G V
, P ) (Fig. ) resembling in shape that of
Harpe saurus tricinctus (see Fig.) but with a scaled struc-
ture like in Anolis proboscis (Figs , c). It is a unique or-
namentation among its closer relatives of the C. nasutum
and C. boettgeri groups.
In some African and Madagascan species rostral ap-
pendages of males such as bony projections of the snout
(e.g. Calumma brevicorne (G, ), or annulated
horns typical of the genus Trioceros (e.g. Trioceros jack-
soni (B, ), T. johnstoni (B, ),
T. montium (B, ), T. oweni (G, )) are
used as weapons for intraspecic contests (e.g. E
, S-F ), but they function also for spe-
cies recognition in sympatric species, as shown already by
R (). He had demonstrated that of the ve East Af-
rican species with three annulated horns none of them is
sympatric with a second species bearing the same constel-
lation of horns. According to his table, the same is true for
other kinds of rostral appendages. at these head orna-
ments of the males serve actually as distinguishing cues for
mate selection has experimentally been proven by P
() who removed the exible nasal lobe in C.nasu tum
(A. M.C. D B, ) females and thus pro-
voked dierent reactions of the conspecic males.
Further evidence for the role of head ornamentation in
male chameleons as distinguishing optical cues in syntop-
ic species has been provided by B K ().
ey demonstrated that the natural invasion of a two-
horned species (T. montium) into the area of the T. quadri-
cornis group in West Cameroon lead rst to a complete re-
duction of the horns in T. q. eisentrauti (M, ),
accompanied by an enlargement of the dorsal and caudal
sail, plus the replacement of conical gular scales by scaly
skin lobes. Subsequently, when C. montium conquered also
the distribution area of T. q. quadricornis (T, ),
the same morphological change started in the latter: re-
duced horn size, increased size of the dorsal and tail sails,
and accentuation of the gular crest which is weakly devel-
oped in T. montium. Only the third taxon, C. q. gracilior
(B K, ), living in the most remote and
elevated area of West Cameroon, did not experience pen-
etration of T. montium into its range so that it kept its rela-
tive longer four horns and lower dorsal and tail sails, thus
resembling the silhouette of the allotopic T. montium (see
the gs. in B K ).
Figure 4. Calumma gallus (male) from Madagascar showing its proboscis. Photo: F G.
I I et al.
Iguanid lizards with rostral appendage
Horn-like protuberances on the snout occur also in two
big-growing iguanids of the genera Iguana L,
and Cyclura H, , both named also in respect to
these attributes, viz. in I. iguana rhinolopha W,
of Central America (but see also B et al. )
and in C. cornuta (B, ) occurring on the
Caribbean island of Hispaniola. In both cases, the rostral
protuberances are comparatively small and seem to serve
for mate recognition purposes because Iguana iguana
(L, ) shows them only in its rhinolopha form
which is widely coexisting with the black iguanas of the
genus Ctenosaura W, . Similarly, Cyclura cor-
nuta is the only species of this genus with otherwise strictly
allopatric insular species which coexists with a sympatric
congener on Hispaniola, viz. C. ricordii (A. M. C. D
B, ). at these large lizards rely on epigamic
signals, such as horns, crests, gular sacs etc., in their mat-
ing system, is also supported by genital morphological evi-
dence (Z B ). erefore, the dierences
in head ornamentation described above are again best ex-
plained by character displacement developed in lizards liv-
ing in syntopy (B et al. ).
Dactyloid lizards with rostral appendage
A rostral appendage or proboscis at the front of the snout,
like that of Harpesaurus tricinctus, is also a distinctive fea-
ture of some species of the highly diverse genus Anolis D-
, (Iguania, Dactyloidae) collectively called probos-
cis-bearing anoles. Only three species of proboscis-bearing
anoles have been so far described: Anolis laevis C, ,
A. phyllorhinus M C, , and A. probos-
cis P O, . All are rare arboreal forest liz-
ards with distributions encompassing the Pacic Chocó in
Ecuador (A. proboscis), Amazonian Peru (A. laev is), and
Brazilian central Amazonia (A. phyllorhinus). ey were
represented by only nine voucher specimens in (W-
). In , P O, based on the rar-
ity of all these proboscis-bearing species, known only from
males, even suggested that they might correspond to spe-
cies previously described based on females only. Howev-
er, there are few species known only from females in that
lizard group and that hypothesis is no longer tenable ac-
cording to recent observations. Among them, A.pro boscis
is particularly convergent with H. tricinctus not only by the
size and shape of its rostral appendage but also by its habi-
tus and a similar color pattern of broad dorsal alternating
dark and light bands (Fig. ). e convergence for the col-
oring is particularly pronounced in certain details such as
for example the presence in both species of colored ocelli
like spot in the hollow of the armpit in both species (Figs,
, ). Below we compare proboscis-bearing anole species
and suggest the use of recent ecological data obtained from
this morphologically convergent proboscis-bearing anole
lizard to more precisely dene the potential habitat in
which to search for nding the rare Javan agamid species.
e Anolis radiation comprises ve main clades that
dene groups of species oen referred to as ‘species se-
ries’ (C Q ). Most Amazonian
Anolis are currently assigned to the punctatus group, which
includes ca. taxa, some of which exhibit wide ranges
in South America. A rostral proboscis can be observed
in three species of the genus. Based on the possession of
the proboscis, W () grouped the three taxa
(A. laevis, A. proboscis and A. phyllorhinus) in the laevis
species group despite they present highly disjunct distri-
bution ranges (see above). Based on morphological char-
acters, A. proboscis has been consistently grouped with
taxa in the heteroderma species group (P , N-
et al. , P et al. , C Q-
). Morphological comparisons by Y-M
et al. () challenged the hypothesis of a close relation-
ship between A. phyllorhinus and A. proboscis. By contrast,
Figure 5. Among all proboscis bearing lizards, Anolis proboscis is the most strongly convergent with the Asiatic agamid H. tricinctus
and their resemblance is striking. Pichincha: Mindo, Ecuador (QCAZ 10541). Photo: S R. R.
Pinocchio lizards and other lizards bearing rostral appendages
other morphological examinations suggested A. phyllo-
rhinus as being related to the punctatus group (R
et al. , Y-M et al. ). Although genetic
data were currently neither available for A. proboscis nor
for A. laevis, recent phylogenetic studies based on com-
bined molecular and morphological evidence suggest that
A. proboscis is closely related to the heterodermus species
group (P , N et al. , P et al. ,
C Q ). Morphological compari-
sons, in turn, led W () to suggest A. laevis to be
closely related to A. heterodermus A. H. A. D, .
In a recent paper, P et al. () infer the relation-
ships of the proboscis-bearing species and demonstrate
that A. phyllorhinus is the sister taxon of A. punctatus D-
, in all analyses. e former species is more closely
related to the broadly sympatric A. punctatus than to the
remaining proboscis-bearing species, showing that ros-
tral appendages have evolved independently at least twice
in the anole family. P et al. () subsequently showed
that the three proboscis-bearing anoles belong to three dis-
tinct lineages and thus the proboscis has evolved at least
three times separately, as it is also the case in the Sri Lan-
kan agamid Ceratophora species. Interestingly, with the ex-
ception of the proboscis and its reddish dewlap coloration,
A. phyllo rhinus (with proboscis) diers from A. punctatus
(without proboscis) by only a few quantitative morpho-
logical traits. e proper identication of females, which
lack both the proboscis and a developed dewlap, is indeed
dicult (R et al. ). Marked structural dif-
ferences in the proboscises of A. phyllorhinus, A. proboscis
and A. laevis (see W ) (Fig. a–c) also sup-
port the view that these structures are not homologous
(Y-M et al. ). Anolis laevis has an append-
age composed of a single scale (like in H. tricinctus; Fig. ),
the rostral plate produced into a exible appendage (P-
O ) but A. phyllorhinus and A. proboscis
are distinct for they have the snout projection composed of
several rows of small scales (see M et al. ) (see
Figs a,c). e elongated, eshy appendage of A. probos-
cis diers from that of A. phyllorhinus in having a serrated
edge and a median dorsal row of scales which is produced
into a serrated prominent dorsal crest. e former also
possess a very strongly compressed tail with sharp verte-
bral angle like in H. tricinctus.
Among all proboscis-bearing lizards, one anole spe-
cies, A. proboscis, is particularly remarkable since its habi-
tus and color pattern are strongly convergent with its Asi-
atic agamid counterpart H. tricinctus. Below we summa-
rize the ecological data obtained on the three proboscis
anoles and particularly the highly convergent A. probos-
cis. Our hypothesis is that similar habitus and coloration
of phylo genetically distinct lizards and snakes oen re-
ect a similar ecology and habitat. us we suggest using
the recent data obtained on the ecology of A. proboscis to
infer the places where to search for H. tricinctus on Java
and, as the type locality could not be ascertained because
there is no date nor a collector name, also on other Sunda
Ecology of proboscis-bearing anoles
e Brazilian Anolis phyllorhinus was known only from
two male specimens until eldwork in the Amazon a dec-
ade ago led to the capture of eight additional specimens,
including a female (R et al. ), and even more
specimens recently (M et al. ). e moderate-
sized proboscis of this anole diers from that of A. pro-
boscis since it is shorter and dorsoventrally more enlarged,
almost equal to the head size in length; however both are
scaly (Fig. a,c). Tail length is much longer than body
length and a maximum snout–vent length of mm was
noted for a male (R et al. ).
Observations during that Brazilian expedition provide
habitat data for nine specimens, indicating that they oc-
cur at varying heights on medium- to large-diameter trees
in primary forest; once spotted, most lizards moved high-
er into the tree. R et al. () considered their
sample size as too small to indicate microhabitat perch
preference. ey observed that A. phyllorhinus explores
virtually all structural habitats on the trunk of thin to
thick trees from the ground level to the canopy. Although
most of their specimens were rst sighted at low heights
in trees, they climbed quickly when disturbed, eventually
jumping on to branches and leaves. Females have no ros-
tral appendage. Most specimens were brown colored when
rst sighted. ese last specimens were basking at sun-
ny patches on tree trunks, indicating that A. phyllorhinus
Figure 6. (a) Anolis phylorrhinus, (b) A. laevis, (c) A. proboscis,
all aer W, 1979.
I I et al.
practice behavioral thermoregulation. When spotted, most
specimens were in position of territorial survey: pointing
downwards, head elevated, and limbs extended. Upon no-
ticing an observer, the green color immediately changed
to brown (those already brown clearly even changed to a
darker tone), and most tried to escape; climbing the tree,
displaying, head bobbing and ashing the bright red dew-
lap (R et al. ). at kind of green natural col-
oration fading to a brownish coloration when threatened
was recently described for Harpesaurus modiglianii from
Sumatra (P et al. ), and such a behavior can be
considered as convergent in several unrelated arboreal liz-
ards. Note that males of Anolis phyllorhinus and A. puncta-
tus from Aripuanã are also identical in color and size, but
have strikingly dierent dewlaps: very large and creamy-
white in A. punctatus whereas all males of A. phyllorhi-
nus have a small and bright red dewlap (R et al.
). e female lacks the characteristic male proboscis
and, except of a few consistent dierences, is morphologi-
cally similar to females of A. punctatus, a species closely re-
lated and broadly sympatric with A. phyllorhinus. e most
recent data obtained from that species are consistent with
the hypotheses that the male proboscis probably evolved as
a sexual signal and that visual signals for intra- and inter-
specic recognition are redundant in anoles (R
et al. ).
Anolis laevis is another distinctive ‘‘proboscis anole’’
known only from the type specimen originating from a
mountain trail between Moyobamba and Balsa Puerto
in the Department of San Martin, Central Peru (P
Y-M ). It has not been seen from Peru
since its description by C in based on a single pre-
served male specimen (L et al. ). Its small probos-
cis (Fig. b) is just minimally developed (W ).
e original description by Cope () indicates a total
length of mm for a tail of mm.
e third proboscis anole, Anolis proboscis (see Fig. ,
c), was reported from its type locality in Ecuador at Cu-
nunuco (and from Las Tolas, from Lloa-Mindo, and from
Tulipe: Y-M et al. ). All these localities are
on the western slopes of the Andes in Distrito Metropoli-
tano de Quito, province of Pichincha. P Y-M-
() were unable to verify the presence of A. pro-
boscis in Peru, and they suspect these are mistaken records.
e most likely Ecuadorian endemic A. proboscis was an
enigma until recently. Known for decades from only six
specimens, all being males and last one collected in ,
nearly four decades passed without a reported sighting un-
til , when ecotourists on a birdwatching trip discov-
ered one lizard crossing a road near the type locality (A-
V ). Subsequently, ve more speci-
mens, including the rst females, were located in two new
areas in and , extending the species’ known dis-
tribution –km northward (Y-M et al. ).
Eleven more specimens were found near the type locality
in and (P , L et al. , P et al.
e snout–vent length of ve specimens ranges from
. to . mm (. ± . mm). e dorsal crest is covered
with – spiny scales from the neck to above the cloaca
(Y-M et al. ). L et al. () indicated
a maximum snout–vent length of . mm for males and
. mm for females based on examination of males and
females. Male has three dorsal lighter bands which do
not meet ventrally – one is anterior of front legs, one poste-
rior of front legs and one about midway between front and
hind legs (see Fig. in Y-M et al. ).
Anolis proboscis is an extremely slow moving species
which is cryptic in pattern and coloration. It appears to
spend most of its time in dense vegetation high o the
ground where it is almost impossible to observe. Indeed,
several times when watched with binoculars, lizards were
lost due to a slight movement – once as a result of a sneeze,
another when the wind blew the vegetation about – and
the lizard could not be relocated (L et al. ). Anolis
proboscis is clearly a twig anole. In all respects, L et al.
() considered it as extremely similar to Greater Antil-
lean twig anoles. Morphologically, it has a slender body (as
evident by relatively low residual mass), short legs and tail,
and a narrow head, all features that characterize twig anoles
(L ) but also their Asian morphological counter-
part H. tricinctus. Its tail is weakly prehensile, as is that
Figure 7. Holotype of H. tricinctus (MNHN-RA 0623) showing the proboscis much longer than head and not covered with scales,
having only enlarged scales at its base.
Pinocchio lizards and other lizards bearing rostral appendages
of Greater Antillean twig anoles (e.g., W R
, H T ). e sexes of A. proboscis
dier in possession of the horn, and in color and pattern
(see Fig. ; Y-M et al. ). One aspect of the
crypticity of A. proboscis is that it shows rarely display be-
havior (L et al. ). is is, however, consistent with
a general trend among West Indian anoles, in which twig
anoles tend to display less than other ecomorphs. Feeding
behavior did not dier markedly between sexes (L et
al. ) although of males, but only of females,
had consumed beetles, whereas of males, but of
females, had consumed dipterans (see Fig. ). is also
shows that a rostral appendage, only present in males, has
no function as a lure to attract insects or other prey. Anolis
proboscis is a montane species living in a narrow range of
altitudinal levels, from m to about – m with
cool dawn air temperatures of .°C (L et al. ).
Y-M et al. () indicate that the species is very
dicult to observe in the wild. Four eld trips to Las To-
las between July and September , covering a dis-
tance of about km allowed sampling of only four speci-
mens. At Las Tolas, the ratio of A. proboscis observations
versus other sympatric species (Anolis aequatorialis W-
, and A. gemmosus O’S, ) was
:. ey noted that the species’ rarity may be due to sev-
eral factors: () naturally low populations; () high micro-
habitat specicity (e.g., canopy); and () sampling method
eect since the species is cryptic and a typical shrub/ar-
boreal species. According to available data the species was
only observed from a small area ( km²) and lives in a
narrow altitudinal range of only m in areas with se-
vere fragmentation and habitat destruction. Most likely it
does not prefer to be associated with disturbed habitat but
is more visible there compared to deep forest where it cer-
tainly occurs preferentially (I ). It was most of-
ten observed, day and night, on twigs or branches with di-
ameters – cm, rarely on trunks, whereas it stays on leaves
at night, on vines or branches and twigs.
Anolis proboscis oen occurs high in the trees. e ros-
tral horn notwithstanding, it is indistinguishable from
Greater Antillean anoles of the “twig” ecomorph class in
morphology, ecology, and behavior. e possession of a
proboscis by males only could suggest a role in sexual se-
lection (Fig. ). e horn in life is so and highly exi-
ble (see Fig. ) and thus unsuitable for use as a weapon in
male–male combat; hence, the proboscis most likely serves
as an intraspecic communication signal and may be in-
volved in mate choice or territorial displays. Q et
al. () provided the rst data on the role of that append-
age in social interactions. By using a semi-natural environ-
ment where males and females were placed, they describe
social interactions of this species during male–female
courtship and mating interactions, as well as three male–
male agonistic interactions. ey also describe four types
of displays by males, many involving the rostral append-
age. ey found that the rostral appendage is used as an
ornament in social displays but not as a weapon in combat.
at research reports also the ontogeny of the rostral ap-
pendage. Most interesting is the behavior termed “probos-
cis ourishing”: a display composed of stereotyped lateral
movements of the head that appear to be a way for the male
to present the rostral appendage to the female counterpart.
In addition, the paper reports the rst captive-born A. pro-
boscis among which males are born with a tiny rostral ap-
pendage. is study also conrms that the proboscis is ac-
tively lied before any bite attempt and is not, under any
circumstance, used as a weapon against other males as pre-
viously suggested (L et al. , P et al. ). In
Figure 8. A pair of Anolis proboscis in their natural habitat. Photo by M V.
I I et al.
contrast, the authors suggest that the movement of the pro-
boscis could be performed to facilitate feeding behaviors or
even other behaviors related to courtship as the proboscis
was lied when males stimulate the female’s nape (similar
to what was described in other anoles). It is now clear that
the rostral appendage has no direct use in physical com-
bat. Given its cryptic morphology and behavior, it is not
surprising that A. proboscis is so rarely observed (L
et al. ).
Suspected ecology of the extinct
We now have summarized all available data on probos-
cis-bearing lizards to try an extrapolation of the ecology
of H.tricinctus compared to what we know on convergent
species. However, sexual dimorphism for horns, if not pre-
sent in H. tricinctus, would be an important evolutionary
dierence with its convergent dimorphic Anolis proboscis.
Obviously, H. tricinctus can be considered as a typical ar-
boreal canopy dwelling species.
Functional aspects of lizard rostral appendages
Several functional or ecological purposes for the rostral
appendage have been suggested (including its use to warm
more rapidly in the morning or as a lure to attract prey). It is
now demonstrated that the males do not use them as weap-
ons in intrasexual combat. However, the rostral appendage
of A. proboscis is an extremely exible structure, bent when
it comes into contact, even gently, with other objects (see
Fig. ). It was further suggested that the proboscis simply
evolved as a way to make a male look larger (see Fig. ).
Many aspects of the displays of male anoles, and of males
of other species of lizard, serve to make individuals look
larger, such as ventrolateral attening of the body, erection
of crests on the neck and back, engorgement of the throat,
and extension of the dewlap (J , L ).
In addition, during aggressive displays to other males, sev-
eral species of anole will turn sideways, straighten their
bodies, and extend their tongues far out of their mouths,
perhaps for the same purpose to maximize their apparent
body length (S M ). In general, larg-
er males have a big advantage in male–male combat (re-
viewed in L ), so any characteristic that makes
one male look larger than another may be useful. Alter-
natively, females may be more attracted by males with an
extended appendage, perhaps because it made those males
look larger, or perhaps for other reasons; note that, accord-
ing to S-F O (), larger males don’t have
always advantages. at entire hypothesis could be experi-
mentally tested in captive animals with e.g., robots or arti-
cial proboscis of dierent lengths and/or colors placed on
the snout of males and females.
A number of chameleon species possess so rostral
projections instead of bony horns (e.g., Calumma spp.,
Rhampholeon spp., Kinyongia spp.), as do a number of aga-
mid lizards in the genera Lyriocephalus and Ceratophora
(S et al. ), but the appendages in these lizards
are equally poorly understood. In all of these taxa, species
exhibit sexual dimorphism in horn size, shape, number,
or presence, and some Ceratophora have been described
moving their horns in a manner similar to that reported by
L et al. () in A. proboscis. e dewlap of A. probos-
cis is not particularly large. is small size may have sev-
eral explanations. Certainly, the dewlap of A. proboscis is
smaller than those of the three other sympatric anoles. e
small size of the dewlap of A. proboscis thus may serve as a
species recognition cue, a means for lizards to distinguish
conspecics from sympatric non-conspecics (L
C , N et al. ). Alternatively, anoles
with smaller dewlaps or no dewlaps at all oen have oth-
Figure 9. e exibility of the rostral appendage of Anolis proboscis can be recognized e.g. during food intake (here a male hunting
a spider). Photo: M V.
Pinocchio lizards and other lizards bearing rostral appendages
er signals, such as colorful body patterns or permanently
erect crests (W R , F H
, L C ) or even the ability for vocaliza-
tion, e.g., A. vermiculatus C, , A. salvinii B-
, (its synonym A. vociferans pointing on this
ability) and others (M J , B et al.
One nal aspect of A. proboscis’s snout appendage de-
serves mention: it is moveable. Observations of L et al.
() and P et al. () concur with those of Q
et al. (), who observed males raising their horns to a °
angle. P et al. () also suggested that males changed
the inclination of their horns during social encounters.
Anyway the orientation of the appendage changed during
the course of behavioral observations. Ecomorphological-
ly, A. proboscis and A. phyllorhinus are dierent in many
ways (L et al. ). Compared with A. proboscis,
A. phyllo rhinus has longer legs, a longer tail, and is green
in color. Ecologically, it does not seem to be a twig special-
ist. L et al. () argued that these ecomorphological
dierences suggest that possession of a proboscis, whether
homologous or convergent, is not associated with a par-
ticular ecomorphological lifestyle. However, when associ-
ated with other characters like body shape and color pat-
tern, it certainly indicates some similarities in habitat use.
Our examination of proboscis-bearing lizards clearly
shows that there is a strong convergence between Harpe-
saurus tricinctus and Anolis proboscis concerning habitus
and color pattern as well as proboscis size and shape. Both
are well in accordance with the Greater Antillean ‘‘twig’’
ecomorph anoles, which are generally cryptically colored,
diminutive species with short limbs, narrow heads, and a
short, prehensile tail (L et al. ). However, the later
point was not demonstrated for H. tricinctus.
One might suspect, in accordance with the observations
made on proboscis anoles that females of H. tricinctus have
been mixed in the past with females of another sympat-
ric species of the genus. is is unlikely since there is no
other species of the genus known from Java (if Java is really
the origin of that species). K et al. () showed that
there is a kind of competition between two principal mech-
anisms for increasing conspicuousness, either by increas-
ing the ornament’s color or brightness contrast against
the background and to increase the size of the ornament,
which is clearly the case for H. tricinctus and A.proboscis.
Both have a cryptic, banded color pattern and no gular
fold or only a weakly developed dewlap. A negative rela-
tionship across species between color contrast against the
background and dewlap or proboscis size in males, but not
in females, suggest that males of dierent species use in-
creasing color contrast and dewlap/proboscis size as alter-
native strategies for eective communication and social
life. K et al. () also showed that each component
in a signal (such as color or size) may be inuenced by dif-
ferent selection pressures.
Apart from the dierent body and tail but similar head
shape of the chameleon Calumma gallus (G, )
(Fig. ), the habitus of Harpesaurus tricinctus is strongly
convergent with that of the canopy-dwelling Anolis probos-
cis. e presence of a proboscis in anoles is sexually dimor-
phic and most likely excludes a role in thermoregulation
or as a lure to attract prey. Its role could be to increase the
observed size of males vis à vis females and/or other con-
specic males and also to function as an intraspecic sig-
nal. Potential absence of sexual dimorphism of the rostral
appendage in H. tricinctus, would be an important dier-
ence with A. proboscis. All possibilities exist among Asian
arboreal proboscis-bearing agamid lizards.
We here suggest that H. tricinctus should be searched on
Java in deep montane moist forest (above m eleva-
tion), most likely on tree trunks and branches around the
canopy but also at night when sleeping on low elevation
branches. Anyway, as Java cannot be ascertained because
there is no collector nor collect locality attached to the
holotype specimen, searching on other Sunda Islands also
seems realistic to us. Temperature can be cool at dawn
and below °C. As such lizards may be impossible to be
observed in deep forest, we suggest to prospect in freshly
deforested timber and agricultural areas where they will
be easier to spot at remaining recently created forest edges
easy to access (see I ). e species certainly will
allow reduced time for basking and thus will be dicult to
observe. It will capture its prey by a sit and wait strategy,
and together with its cryptic coloration, few movements
will make eye sightings delicate. According to its banded
color pattern, the species certainly will use mosses and
lichens in the montane cloud forest for insolation in cold
weather condition as well as in higher temperature. Its
habitat should present a high vegetative cover with abun-
dant shade, moving sun patches and low amount of bare
We would like to thank K Q for the loan of a speci-
men of Anolis proboscis in NMNH (USNM) collections, R
B (MNHN) for his information on some older illustrations,
and O T C (Escuela de Biología, Ponticia Uni-
versidad Católica, Quito, Ecuador), as well as M V
(Zoological Institute, University of Braunschweig, Germany), for
providing photographs of Anolis proboscis. Likewise, we thank
F G (Bavarian State Collection Munich (ZSM)) for the
photograph of Calumma gallus.
A, G. D., D. C. H M. C (): A
possible acoustic function for the casque structure in hornbills
(Aves: Bucerotidae).– Journal of Zoology,233(1): –.
A, A. C. V (): Anolis proboscis (Sauria: Poly-
chrotidae), una lagartija rara pero no extinta. – Politécnica
(Biol. Ser. ), 27: –.
I I et al.
B, S., L. N. A S. M (): A note on
the protuberance of knob on the snout of male gharial [Gavia lis
gangeticus (G)]. – Journal of the Bombay Natural History
Society, 74: –.
B, W. (): Rediscovery of the Sumatran agamid lizard
Harpesaurus beccarii Doria, , with rst notes on a live spec-
imen. – Tropical Zoology, 2: –.
B, W., R. H W. B (): Die Stimme der
Lacertidae, speziell der Kanareneidechsen (Reptilia: Sauria). –
Bonner zoologische Beiträge, 36(3/4): –.
B, W. C. J. J. K (): Zur innerartlichen Gliederung
und zur Artgeschichte von Chamaeleo quadricornis Tornier,
(Sauria: Chamaeleonidae). – Amphibia-Reptilia, 1(3/4):
B, G. A. (): Catalogue of lizards in the British Mu-
seum (Natural History). Vol. . Taylor & Francis, London.
B, M., B. V, D. S, U. K, M. N.
M, P. H, J. C. D, E. C, G. G, J.
G, N. B, M. J F. G (): A story
of nasal horns: two new subspecies of Iguana Laurenti,
(Squamata, Iguanidae) in Saint Lucia, St. Vincent and the
Grenadines (southern Lesser Antilles). – Zootaxa, 4608: –
B, E. R. (): Faune de Madagascar XXXIII. Reptiles Sau-
riens Chamaeleonidae, Genre Chamaeleo. – O.R.S.T.O.M./
C.N.R.S, Paris, France.
B, E. R. (): Les types d’Agamidés (Reptiles, Sauriens)
du Muséum national d’Histoire naturelle. Catalogue critique. –
Bulletin du Muséum national d’Histoire naturelle, ème série,
10, section A (n° , suppl.): –.
B, E. R. C. A. D (): Les caméléons à rostre
impair et rigide de l’ouest de Madagascar. – Mémoires du Mu-
séum national d’Histoire naturelle, nouvelle série (A : Zoolo-
gie), 52(2): –.
C, M. D. R. K. Q (): Phylogeny of the
Dactyloa clade of Anolis lizards: New insights from combining
morphological and molecular data. – Bulletin of the Museum of
Comparative Zoology, 160: –.
C, K. C., D. B. L, D. G (): Function of
the appendages in tentacled snakes (Erpeton tentaculatus). –
Journal of Experimental Biology, 213(3): –.
C, E. D. (): Report on the Reptiles brought by Professor
James Orton from the middle and upper Amazon and western
Peru. – Journal of the Academy of Natural Sciences of Philadel-
phia, N.S. (), 8: –.
C, C. A. G. M. F N (): A new species
of smooth horned frog, genus Proceratophrys Miranda-Ribeiro
(Amphibia: Anura: Cyclorhamphidae), from the Atlantic rain-
forest of eastern Bahia, Brazil. – Zootaxa, 2060: –.
D, W., P. D. C, U. M, A. G-T
A. K (): Dragons in neglect: Taxonomic revision of
the Sulawesi sailn lizards of the genus Hydrosaurus Kaup,
(Squamata: Agamidae). – Zootaxa, 4747(2): –.
D, W., E. D, T. K, K. L, U. M-
, M. M W. B (): Nomenclatural and
morphological notes on the rare agamid lizard Pseudocophotis
sumatrana (Hubrecht, ) (Squamata: Agamidae: Draconi-
nae). – Raes Bulletin of Zoology, 69: –. doi: ./
R, N. (): e reptiles of the Indo-Australian archipel-
ago. I. Lacertilia, Chelonia, Emydosauria. – E. J. Brill, Leiden.
D, G. (): Note Erpetologiche – Alcuni nuovi sauri raccolti
in Sumatra dal Dr. O. Beccari. – Annali del Museo civico di sto-
ria naturale di Genova, 2(ser. ): –.
D, A. H. A. (): Note sur un nouveau genre de la famille
des Reptiles Iguaniens acrodontes. – Revue et magasin de zoo-
logie pure et appliquée, (e série), 3: –.
D, A. M. C. A. H. A. D (): Catalogue Mé-
thodique de la Collection des Reptiles du Muséum. I. – Gide et
Baudry, Paris, France.
E, D. L. (): e Evolution of Animal Weapons. – Annual
Review of Ecology, Evolution, and Systematics, 39: –.
E, D. S. J. K (): Parallel selective pressures
drive convergent diversication of phenotypes in pythons and
boas. – Ecology Letters, 19(7): –.
F, H. S. R. W. H (): Ecological and ethological
parameters in Anolis bahorucoensis, a species having rudimen-
tary development of the dewlap. – Amphibia-Reptilia, 8: –.
F, H. (): Barbels and barbel-like tentacular structures in
sub-mammalian vertebrates: a review. – Hydrobiologia, 403:
G , F. (): Taxonomic checklist of chameleons (Squamata:
Chamaeleonidae). – Vertebrate Zoology. 65(2): –.
G, F. M. V (): A eld guide to the amphibians and
reptiles of Madagascar. – Vences & Glaw Verlags GbR, Köln,
Ge rma ny.
G, J. (): Sur le dimorphisme sexuel des espèces du genre
Langaha (Ophidiens). – Comptes Rendus de l’Académie des sci-
ences, 226: –.
G, J. (): Révision du genre Langaha (Ophidiens). Le di-
morphisme sexuel, ses conséquences taxonomiques. – Mé-
moires de l’Institut scientique de Madagascar, Série A, 3: –
G, M. L. (): Functions of cephalic scales of the sea snake
Emydocephalus annulatus. – Journal of Herpetology, 30: –
H, D. E. (): Comments on tentacle structure and function
in Erpeton tentaculatum (Serpentes: Colubridae). – Journal of
Herpetology, 7: –.
H, J. W. B (): A review of the genus Pseudo-
calotes (Squamata: Agamidae), with description of a new species
from West Malaysia. – Amphibia-Reptilia, 21(2): –.
H, J. W. B (): On the validity of two re-
cently described species of the endemic Sri Lankan genus Co-
photis Peters, (Squamata: Agamidae) from the Knuckles
Forest Range. – Salamandra, 43: –.
H, S. B. R. T (): A new species of Anolis (Sau-
ria, Iguanidae) from the Sierra de Neiba, Hispaniola. – Herpe-
tologica, 45: –.
I, I. (): How habitat disturbance benets geckos: conser-
vation implications. – Comptes Rendus Biologies, 333: –.
J, E. G. D. P., W. A. D. M A. D S
(): Habitat utilization of endangered rhino horned lizard
(Ceratophora stoddartii) (Sauria: Agamidae) in the Horton
Plains National Park, Sri Lanka. – Journal of Entomology and
Zoology Studies, 6(4): –.
J, T. A. (): Evolution of anoline lizard display behavior.
– American Zoologist, 17: –.
J, G. R., M. S. L T. D. S (): Morpholo-
gy and allometry suggest multiple origins of rostral appendages
in Sri Lankan agamid lizards. – Journal of Zoology, 289: –.
Pinocchio lizards and other lizards bearing rostral appendages
K, D. M. S. S. A. A. T. A (): Be-
havioral ecology and microhabitat use by Lyriocephalus scuta-
tus (Linnaeus, ): a monotypic genus in Sri Lanka (Reptilia:
Agamidae: Draconinae) with notes on the taxonomy. – Russian
Journal of Herpetology, 20(1): –.
K, S., N. A. P, C. A, T. S-
, A. V. B , M. M, V. A. G A. D
S (): A new species of the genus Ceratophora Gray,
(Reptilia: Agamidae) from a lowland rainforest in Sri Lanka, with
insights on rostral appendage evolution in Sri Lankan agamid liz-
ards. – Amphibian & Reptile Conservation, 14(3): – (e).
K, A.C. (): e Hornbills – Bucerotiformes. Vol. – Oxford
University Press, Oxford.
K, F. W. (): A new Bornean lizard of the genus Harpesaurus.
– e Sarawak Museum Journal, 1978: –.
K, D. A., T. J. O, I. D, A. D, N. A D. S-
-F (): Ornament size and color as alternative strate-
gies for eective communication in gliding lizards. – Journal of
Evolutionary Biology, 29(9): –.
K, K. L. (): Feeding behavior of the Madagascar leaf-
nosed snake Langaha madagascariensis (Serpentes: Colubridae:
Pseudoxyrhophiinae), with an alternative hypothesis for its bi-
zarre head structure. – African Journal of Herpetology, 54(2):
K, T., Y. K, M. Y. H K. N ():
Phylogenetic position of a bizarre lizard. Harpesaurus implies
the co-evolution between arboreality, locomotion, and repro-
ductive mode in Draconinae (Squamata: Agamidae). – System-
atics and Biodiversity, 18(7): –. doi:./.
K, M. D., A. H, L. B. P, R. N, D. A-
, N. F, E. H A. A (): Cre-
ation of an amphibian and reptile atlas for the Indonesian is-
lands of Java and Bali reveals gaps in sampling eort. – Herpe-
tology Notes, 14: –.
L, B., S. Y. C U. M (): Notes on Harpe-
saurus borneensis (Mertens, ), a live bearing agamid lizard
from the lowlands of Borneo. – Russian Journal of Herpetology,
L, S. Z. S. (): Historical biogeography of Sumatra and
Western Archipelago, Indonesia: Insights from the ying lizards
in the genus Draco (Iguania: Agamidae). University of Califor-
nia, Berkeley, USA, PhD esis.
L, J. B. (): Lizards in an Evolutionary Tree: Ecology and
Adaptive Radiation of Anoles. – University of California Press,
L, J. B. L.-R. C (): Examination of factors aecting
dewlap size in Caribbean anoles. – Copeia, 1998: –.
L, J. B., M. L. W, L. D. M, O. T-C-
, K. E. C, E. W. S, A. E. N, F. A-
V A. H (): Notes on the Natural History of
the Little-Known Ecuadorian Horned Anole, Anolis proboscis.
– Breviora, 531: –.
M-A, K., A. S T. A
(): Description of a second species of Cophotis (Reptilia:
Agamidae) from the highlands of Sri Lanka. – Lyriocephalus,
M, U. (): Agamid Lizards of Southern Asia Draconi-
nae & Leiolepidinae / Agamen des südlichen Asien Draconi-
nae & Leiolepidinae. – Terralog b. Edition Chimaira & A. C.
S. Glaser, Frankfurt am Main, Germany.
M, M. (): Life in the water: ecology of the jacarerana
lizard, Crocodilurus amazonicus. – e Herpetological Journal,
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 imagi-
nary: the lizard species with the smallest ranges. – Diversity and
Distributions, 24(2): –.
M, J. I. G. R. J (): e structure of the male
proboscis in the New Guinean treefrogs, Litoria prononimnia
and Litoria havina (Anura: Hylidae). – Australian Journal of
Zoology, 63: –.
M R. (): Über das im Senckenbergischen Museum be-
ndliche Exemplar von Cophotis sumatrana Hubrecht (Rept.,
Lac.). – Senckenbergiana, 3: –.
M, T. H. T. A. J (): Description and signicance
of vocalizations by Anolis grahami (Sauria: Iguanidae). – Co-
peia, 1979(3): –.
M, S. M. (): Phylogenetic and historical biogeographical
relationships of the genera in the family Agamidae. – University
of Michigan, Ann Arbor, USA: Unpublished PhD esis.
M, N., S. A. S, H. F S. B (): Ad-
ditional information on Misonne’s swollen-nose gecko, Rhino-
gecko [sic] misonnei de Witte, (Squamata, Geckonidae
[sic]) in Iran. – Amphibian & Reptile Conservation, 5(1): –.
M, L. J. C. L., F. P. W D. P (): New record
and updated geographic distribution of Anolis phyllorhinus My-
ers & Carvalho, . – Herpetozoa, 31: –.
M, M., A. H, A. F E. N. S (): A
new Megophrys Kuhl and van Hasselt (Amphibia: Megophryi-
dae) from southwestern Sumatra, Indonesia. – Zootaxa,
N, K. E., B. I. C, C. G J. M. S ():
It is time for a new classication of anoles (Squamata: Dactyloi-
dae). – Zootaxa, 3477: –.
N, K. E., L. J. H J. B. L (): Evolution of
Anolis lizard dewlap diversity. – PLoS One, 2: e. doi:./
O, P. M., R. G, M S. J. R (a):
Systematics of New Guinean treefrogs (Litoria: Pelodryadidae
[sic]) with erectile rostral spikes: an extended description of
Litoria pronimia and a new species from the Foja Mountains. –
Zootaxa, 4604(2): –.
O, P. M., S. J. R S. C. D (b): Two
new species of treefrog (Pelodrydidae [sic]: Litoria) from south-
ern New Guinea, elucidated by DNA barcoding. – Zootaxa,
O, T. J. D. A. K (): Habitat partitioning and morpho-
logical dierentiation: the Southeast Asian Draco lizards and
Caribbean Anolis lizards compared. – Oecologia, 175: –.
O, T. J. D. S-F (): Ornament evolution in dragon
lizards: multiple gains and widespread losses reveal a complex
history of evolutionary changes. – Journal of Evolutionary Biol-
ogy, 19(3): –.
P, S. R. (): Observations on the natural history of six
Malagasy Chamaeleontidae. – Zeitschri für Tierpsychologie,
I I et al.
P, N., K. O. W, L. J. V, D. B. F
E. R. P (): How do lizard niches conserve, diverge or
converge? Further exploration of saurian evolutionary ecology.
– BMC Ecology and Evolution, 21(1): –.
P, J. A. V. G. O (): A third leaf-nosed species of the
lizard genus Anolis from South America. – Breviora, Museum
of Comparative Zoology, 62: –.
P, R. K. M-A (): A re-
vision of the endemic Sri Lankan agamid lizard genus Cerato-
phora Gray, , with description of two new species. – Journal
of South Asian Natural History, 3(1): –.
P, S. (): Phylogeny of anoles. – Herpetological Monographs,
P, S. (): Finding Anolis proboscis. – Anolis Newsletter, VI:
P, S., F. A, I. M. L, T. L. K, J. A. C-
, L. N. G, N. J. B, B. M. A E. W. S
(): Morphology, phylogeny, and behavior of Anolis pro-
boscis. – Breviora, Museum of Comparative Zoology, 530:
P, S., A. N-M O , O. T-C, K.
Q, J. A. V, B. T, L. N. G, M. J. R,
G. K, F. A-V I. L (): A phyloge-
netic, biogeographic, and taxonomic study of all extant species
of Anolis (Squamata; Iguanidae). – Systematic Biology, 66(5):
P, S. C. Y-M (): Another new species of
green Anolis (Squamata: Iguania) from the Eastern Andes of
Peru. – Journal of Herpetology, 42(3): –.
P, I., M. T. R, P. R. M-S C. C
(): Phylogenetic relationships of Amazonian anole lizards
(Dactyloa): taxonomic implications, new insights about pheno-
typic evolution and the timing of diversication. – Molecular
Phylogenetics and Evolution, 82: –.
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(1): –.
P, C. A., A. A. T. A, D. H, S. S,
J. J. B, U. M I. I (): Rediscovery of
Modigliani’s nose-horned lizard, Harpesaurus modi glianii Vin-
ciguerra, (Reptilia: Agamidae) aer years without any
observation. – Taprobanica, 9(1): –.
Q, D. R., A. M, O. T-C, A. E.
N, F. A-V I. T. M (): Use of a ros-
tral appendage during social interactions in the Ecuadorian
Anolis proboscis. – Journal of Natural History, 51(27–28): –
R, A. S. (): A suggested function of the ornamentation of
East African forest chameleons. – Copeia, 196(4): –.
R, M. T., V. X, G. S D. P (): New
specimens of Anolis phyllorhinus (Squamata, Polychrotidae):
the rst female of the species and of proboscid anoles. – Papéis
avulsos de zoologia, 42(16): –.
S, V. A. M. P. K., K. B. R, D. R. N. S.
R, N. B. A, N. L. O, J. M. A. S. R-
V. A. P. S (): A new species of
the genus Cophotis (Squamata: Agamidae) from Sri Lanka. –
Russian Journal of Herpetology, 13(3): –.
S II, J. A., J. R. M, R. P A. L
(): Rostral horn evolution among agamid lizards of the ge-
nus Ceratophora endemic to Sri Lanka. – Molecular Phyloge-
netics and Evolution, 22(1): –.
S, K. G. C. M (): Tongue display in anoles and
its evolutionary basis. – pp. – in: L, J. G. C. M
(eds): Anolis Newsletter IV. – National Museum of Natural
History (Smithsonian), Division of Amphibians and Reptiles,
Washington, D.C., USA.
S, F. R. (): Notes on lizards of the genus Cerato-
phora. – Loris, 15: –.
S-F, D. (): Chameleon behavior and color change. –
pp. – in: T, K. A. A. H (eds): e biology
of chameleons. – University of California Press, Berkeley, Los
S-F, D. M. T. J. O (): Sexual selection, natural
selection and the evolution of dimorphic coloration and orna-
mentation in agamid lizards. – Proceedings of the Royal Society
of London, B, 27: –.
T, C. (): Chameleons of Africa. An atlas, including the
Chameleons of Europe, the Middle East and Asia. Edition Chi-
maira, (Contributions to Natural History, volume ), Frank-
furt am Main, Germany.
T, J. (): Field observations on the behavioural ecology of
the Madagascan leaf-nosed snake, Langaha madagascariensis. –
Herpetological Conservation & Biology, 7(): –.
T, K. A. A. E. H (eds) (): e Biology of chame-
leons. – University of California Press, Berkeley and Los Ange-
les, California, USA.
U, P., P. F, R. A J. H (eds) (): e Rep-
tile Database. Available at http://www.reptile-database.org, ac-
cessed May .
V, D. (): Descrizione di una nova species di Harp-
esaurus di Sumatra. – Annali del Museo civico di storia naturale
di Genova, 56: –.
W, H. (): Liste der rezenten Amphibien und Reptilien:
Agamidae. – Das Tierreich, 86: –.
W, M. J., D. W. N R. S (): Sexual di-
morphism in conspicuousness and ornamentation in the enig-
matic leaf-nosed lizard Ceratophora tennentii from Sri Lanka. –
Biological Journal of the Linnean Society, 116(3): –.
W, E. E. (): South American anoles: the species groups.
. e proboscis anoles (Anolis laevis group). – Brevio ra, Mu-
seum of Comparative Zoology, 449: –.
W, E. E. A. S. R (): Anolis insolitus, a new dwarf
anole of zoogeographic importance from the mountains of the
Dominican Republic. – Breviora, Museum of Comparative Zo-
ology, 326: –.
W, E. E. A. S. R (): Species recognition, dewlap
function, and faunal size. – American Zoologist, 17: –.
W, R. M. (): e integumentary tentacles of the snake
Erpeton tentaculatum. Structure, function, evolution. – Herpe-
tologica, 33: –.
Y-M, M. H., M. A. U, M. A S. R.
C (): Redescripción de Anolis proboscis Peters &
Orcés (Reptilia: Polychrotidae), con el descubrimiento de las
hembras de la especie y comentarios sobre su distribución y
taxonomía. – ACI Avances en Ciencias e Ingenierías, sección
B, 2(3): B–B.
Z, T. W. B (): Genitalstrukturen und Paarungs-
biologie bei squamaten Reptilien, speziell den Platynota, mit
Bemerkungen zur Systematik. – Mertensiella, 8: –.