Furcifer timoni, Glaw, Frank, Köhler, Jörn & Vences, Miguel, 2009

Glaw, Frank, Köhler, Jörn & Vences, Miguel, 2009, A distinctive new species of chameleon of the genus Furcifer (Squamata: Chamaeleonidae) from the Montagne d’Ambre rainforest of northern Madagascar, Zootaxa 2269, pp. 32-42 : 33-40

publication ID

https://doi.org/ 10.5281/zenodo.190909

DOI

https://doi.org/10.5281/zenodo.6224500

persistent identifier

https://treatment.plazi.org/id/03FAF57F-117A-4170-D5BA-FB8E4B64345D

treatment provided by

Plazi

scientific name

Furcifer timoni
status

sp. nov.

Furcifer timoni sp. nov.

Figs. 1–5 View FIGURE 1 View FIGURE 2 View FIGURE 3 View FIGURE 4 View FIGURE 5

Holotype. ZSM 2103/2007 ( FGZC 1095), adult male, from Montagne d'Ambre National Park (12°30' E, 49°11' S), ca. 750–800 m a.s.l., Antsiranana Province, northern Madagascar, collected on 25 February 2007, by P. Bora, H. Enting, F. Glaw, A. Knoll, J. Köhler and A. Razafimanantsoa.

Paratypes. ZSM 255/2004 ( FGZC 495), ZSM 256/2004 ( FGZC 496), ZFMK 87585 (originally ZSM 257/2004 [ FGZC 499]), UADBA uncatalogued ( FGZC 497), UADBA uncatalogued ( FGZC 498), five adult and gravid females, from Montagne d'Ambre National Park (same general area as holotype locality), Antsiranana Province, northern Madagascar, collected 20–23 February 2004, by F. Glaw, M. Puente, R. Randrianiaina & A. Razafimanantsoa; ZSM uncatalogued ( FGZC 1884) [still in UADBA], adult and gravid female, Montagne d'Ambre National Park, 850 m a.s.l. (similar geographical coordinates as holotype), Antsiranana Province, northern Madagascar, collected 26 February 2008, by N. D'Cruze, F. Glaw, Z. Nagy & A. Razafimanantsoa.

Diagnosis. The new species differs from all other Furcifer species except F. bifidus and F. balteatus by a whitish lateroventral band that is composed of scales which are arranged in a rosette-like manner. It differs from F. balteatus by smaller size (male SVL 88 mm, female SVL 95–96 mm versus male SVL up to 175 mm, female SVL up to 145 mm), shorter rostral appendages of the males (6 mm versus up to 10 mm), general colouration, and the position and extension of an oblique light band on the flanks that runs from the dorsal crest (or slightly below) posterioventrally without reaching the lateroventral band (versus running from the upper flanks to the whitish lateroventral band). It differs from F. bifidus by distinctly smaller size (SVL of adult male 88 mm versus up to 200 mm), much shorter rostral appendages of the males (6 mm versus up to 24 mm, Fig. 5 View FIGURE 5 ), a lower number of scales (4 vs. 6-8) between nostril and tip of rostral appendage, and by colouration. In addition, males of Furcifer timoni differ from all other Furcifer species except those of the Furcifer bifidus group - F. bifidus , F. balteatus , F. m i n o r (Günther, 1879), F. willsii (Günther, 1890) and F. petteri (Brygoo & Domergue, 1966) - by the presence of two bony rostral appendages and from adult males of all five species within the F. bifidus group by a shorter length of the rostral appendages (6 mm vs. 5–24 mm).

Description of the holotype. Adult male, in good condition, both hemipenes completely everted; SVL 88.0 mm (length from tip of rostral appendage to vent 93.0 mm); tail 121 mm; axilla-groin distance 48.8 mm; eye horizontal diameter 8.7 mm. Head with two ossified, scaled rostral appendages, length 5.9 mm, projecting beyond the upper lip by 4.7 mm, rostral appendages laterally compressed, almost parallel, but slightly diverging, distance between anterior tips 9.2 mm; rostral crest well developed; supra-orbital crest rounded in lateral view and formed by single, rather smooth row of tubercles; lateral crest moderately distinct, smooth in lateral view; temporal crest recognizable behind eye, almost parallel with lateral crest, both crests fusing posteriorly; parietal crest absent; no traces of occipital lobes; no traces of gular crest. Dorsal crest present, consisting of a single row of 12 pointed tubercles 1 mm or less in height on the anterior dorsum; body laterally compressed with homogeneous scalation; no distinct ventral crest; axillary pits not obvious; limbs and tail with homogeneous scalation, tail without dorsal crest of pointed tubercles, feet without tarsal spines.

Hemipenis ( Fig. 3 View FIGURE 3 ) large, total length 17.4 mm, morphology strongly differing from that of the closely related species F. bifidus , F. minor , F. w i l l s i and F. p e t t e r i (see drawings and descriptions in Ramanantsoa 1978 and Brygoo & Domergue 1969). Truncus covered with calyces, especially in asulcal view, calyces absent around the sulcus spermaticus. Sulcus spermaticus poorly recognizable. Apex bilobed, formed by two big lobes each covered with two rotulae and one field of short papillae; a further field of long papillae on each side at the base of the apex in sulcal view.

Colour after almost two years in preservative generally blackish brown, especially on flanks, dorsal surfaces of upper arms and legs and lateral parts of the proximal tail. Throat, sides of head, lower arms, lower legs, distal parts of tail and dorsalmost 5–6 mm of flanks grey to purple. A white band starting at the angle of the mouth runs along the upper lip and fades below the eye. The ventrolateral light band between fore- and hindlimb is faded to brownish but still recognizable. An indistinct whitish midventral line is recognizable. Several blue scales on the dorsal plate of the head are the only colourful remnants of the life colouration. Hemipenis uniformly whitish.

In life when unstressed ( Fig. 1 View FIGURE 1 ), dorsal and lateral colouration of head, body, limbs, and tail uniformly lime green, with a whitish-beige oblique stripe on the anterior flanks, starting several millimeters below the seventh dorsal tubercle and ending several millimeters above a whitish lateroventral band of rosette-like scales that runs along the flanks between the insertion of fore- and hindlimbs. A white band along the whole upper lip starting at mouth angle. Eye ball with a characteristic colouration: upper half green, lower half with a light brown outer ring and a blue inner ring, eye opening encircled by a complete, narrow yellowish-orange ring. Lateral crest light brown. Head dorsally with numerous blue scales. Ventral surfaces of head yellowishgreen, body ventrally greyish-green, inner surface of limbs light grey.

Variation of the paratypes. For measurements and character conditions of the type specimens see Table 1. Females of F. timoni lack the rostral appendages of the male and have a distinctly less swollen tail base. Furthermore, the relative tail length of the male holotype is longer than in the females ( Table 1) which were all gravid with fully developed, yellowish eggs. The colouration of the female paratypes in preservative is almost uniformly blackish. The light ventrolateral bands and the whitish midventral line are more distinct than in the holotype. Bluish spots are recognizable on the heads.

ZSM 2103/2007 ZSM 255/2004 ZSM 256/2004 ZFMK 87585

Status holotype paratype paratype paratype Sex male (adult) female (gravid) female (gravid) female (gravid) SVL 88.0 96.0 94.5 95.5 Total length 214 194 195 198

head length 28.7 27.3 26.7 27.5 snout length 20.8 19.6 18.9 19.3 eye diameter 8.7 7.8 7.0 7.6

number of dorsal tubercles ca. 12 ca. 10 ca. 12 ca. 9 egg number in body cavity --- 14 10 13

During the day the ground colouration of apparently unstressed gravid females in life is green with a pattern of green or brown, mainly transversal bands ( Fig. 4 View FIGURE 4 ) on the flanks and sides of the head. Thinner, regularly shaped and regularly spaced transversal bands are also present on tail and legs. A red band starting from middorsum or slightly below runs posterioventrally until the middle of the flanks. The ventrolateral band between fore- and hindlimbs is reddish ( Fig. 4 View FIGURE 4 ). The posterior part of the head is covered by a large more or less triangular red spot which is strongly bordered posteriorly but fades less distinct anteriorly. The head (including the red spot) is covered with many blue scales and similar blue scales, which are partly arranged in longitudinal rows, are scattered along the flanks, decreasing in density posteriorly. There is no distinct white band along the lip, but a whitish spot on the upper lip is recognizable below the eye. The eye ball is similar but less distinctly coloured than in males, usually without distinct blue colour. At night when roosting the body is uniformely green with only indistinct transversal banding pattern, but with distinct diagonal red band on the flanks and triangular red spot on the head. Only few bluish scales are recognizable on the head and no distinctly blue scales on the flanks. The eye ball is brown in the lower half, but without blue colour. When the throat is inflated a reddish reticulated pattern is visible.

Additional photographic records. Photographs of one or two males of unknown origin were published in a Japanese chameleon book ( Kimura 2003: 114) and might refer to F. timoni . Two additional males from Montagne d'Ambre were photographed by H.-P. Berghof and P. Schönecker (e.g., Fig. 1 View FIGURE 1 c), respectively. All these photographs strongly resemble the holotype of Furcifer timoni in morphology and principal colouration and strongly confirm that it represents a distinctive new species and not a subadult or aberrant specimen of the green colour morph of Furcifer bifidus as shown in Martin & Wolfe (1992: 139). Similarly, the photographs of females taken from the pet trade ( Kimura 2003 and own photographs of a specimen taken in the 1990s) resemble those from Montagne d'Ambre. Unstressed gravid females appear to be largely green, with a network pattern of brown bands which are principally arranged similarly among different females.

Etymology. The senior author wishes to dedicate this new chameleon species to his son Timon Robert Glaw.

Distribution, habitat and habits. Furcifer timoni is only known from the Montagne d’Ambre National Park, between 750 and 900 m altitude. The discovery of this distinctive new Furcifer species in Montagne d’Ambre was very surprising since this area has been repeatedly and intensively surveyed for reptiles over many years ( Mocquard 1895; Ramanantsoa 1974; Andreone 1991; Kauffmann 1994; Raxworthy & Nussbaum 1994; Glaw & Vences 1994; D'Cruze et al. 2008). Apart from the type locality the species possibly occurs in the Marojejy National Park ( Fig. 6 View FIGURE 6 ) as well, since photographs of a female taken by G. Gomboc at Marojejy (between Camp 1 and Camp 2) resemble females of F. timoni and might represent this or a very closely related species. It differs from the studied type specimens (see Table 1) by a dorsal crest composed of more (ca. 20) tubercles (see Fig. 6 View FIGURE 6 ). Similar to the situation at Montagne d'Ambre, all previous herpetological surveys in the Marojejy National Park since 1968 (see summary in Raselimanana et al. 2000) did not reveal any record of a chameleon resembling F. timoni , indicating that both the Montagne d'Ambre and the Marojejy populations had been overlooked, possibly due to them occupying a cryptic niche hidden high up in the tree canopy.

At Montagne d'Ambre, Furcifer timoni was exclusively found in mid-altitude primary rainforest during the rainy season, in close syntopy with Furcifer petteri . All specimens we captured in the years 2004–2008 were found in the period 20–26 February, roosting at night in the vegetation. The only male seen and captured during the same period was captured from a position more than 3 m above the ground. In contrast, gravid females were relatively common in late February, usually sitting on branches ca. 1.5–3 m above the ground. We suspect that they were just descending from the trees to bury their eggs into the ground. Data on egg numbers of preserved females are given in Table 1.

Available names. Among the Malagasy chameleons with paired bony rostral appendages in males, there is only one taxon, Dicranosaura bifurca var. crassicornis Gray, 1864 that needs to be discussed as a possible earlier name for Furcifer timoni (the other synonyms of Furcifer bifidus are replacement names due to errors and therefore not available). Brygoo (1971) considered the possibility that crassicornis could be a synonym of F. balteatus rather than F. bifidus , but the type specimens were apparently never identified or studied. Unfortunately, the original description of this taxon is very short, imprecise and difficult to interpret and we here repeat the whole description ( Gray 1864: 479):

" Var. crassicornis B. M.

One of the males, with the horns only partly developed, has them very thick and trigonal at the base, so as nearly to reach across the nose. In another young male, about the same size, they are compressed and far apart at the base, as in the type specimens."

Colin McCarthy (in litt., 22 July and 24 Aug. 2009) kindly provided us the following information: "The situation regarding the types of Dicranosaura bifurca var. crassicornis Gray, 1864 is unclear. We do not appear to have distinguished the types of this variety in our collection. What we have are the four specimens of ' Chamaeleo bifurcus ' mentioned in Gray's (1845) lizard catalogue to which I assume could be added BMNH 1863.4.16.1 which appears to have been the only other specimen of this taxon added to the collections before Gray's paper of 1864. So the relevant specimens for consideration are:

BMNH xxv.12a Madagascar [SVL: 170 mm, tail length: 233 mm, rostral appendage (length, ventral edge): 26.2 mm, rostral appendage length (projection): 23.4 mm]

BMNH xxv.12b Madagascar / Pres: J.E. Gray [SVL: 97 mm, tail length: 119 mm, rostral appendage (length, ventral edge): 5.7 mm, rostral appendage length (projection): 5.4 mm]

BMNH xxv.12c Madagascar / Pres: J.E. Gray [SVL: 105 mm, tail length: 128 mm, rostral appendage (length, ventral edge): 6.6 mm, rostral appendage length (projection): 6.1 mm]

BMNH RR1935.12.8.1 (formerly xxv.12d) Madagascar / Pres: J.E. Gray [SVL: 125 mm, tail length: 174 mm, rostral appendage (length, ventral edge): n/a, rostral appendage length (projection): n/a

BMNH 1863.4.16.1 Madagascar / Purch: Mr Stevens [SVL: 147 mm, tail length: 212 mm, rostral appendage (length, ventral edge): n/a?, rostral appendage length (projection): n/a]

Just looking at the specimens, at least one of the possible candidates for crassicornis appears to be BMNH 1863.4.16.1. This has very short horns and I think that a very faded label on the jar may say 'crassicornis' - though it is difficult to be sure. However this specimen is rather larger than BMNH xxv.12b and BMNH xxv.12c (which are relatively close in size to each other).

I think we can rule out BMNH xxv.12a since, as Gray 1845 states, it has 'elongated nose horns' and BMNH RR1935.12.8.1 (formerly xxv.12d) is Gray's 'female without horns'."

We fully agree with this interpretation and after examination of these specimens in the Natural History Museum on 30 September 2009 we consider only the three specimens (BMNH xxv.12b [subadult male], BMNH xxv.12c [subadult male], and BMNH 1863.4.16.1. (formerly xxv.12d) [probably adult male] as syntypes. To remedy the uncertainty in the usage of this name, we hereby designate BMNH 1863.4.16.1 as lectotype of Dicranosaura bifurca var. crassicornis Gray, 1864 . This probably adult male (total length 359 mm, tail base moderately swollen), apparently labelled as " crassicornis " has very short and probably malformed rostral appendages, but otherwise agrees with males of Furcifer bifidus . It differs from the male holotype of Furcifer timoni by much larger size (147 mm vs. 88 mm SVL, total length 359 mm vs. 214 mm), more tubercles of the dorsal crest (ca. 25 vs. ca. 12) and a broader ventrolateral band of rossette-like scales (maximum width 4.8 mm vs. 2.2 mm). The remaining two paralectotypes are smaller than the lectotype. They superficially resemble the holotype of F. t i m o n i in size and general morphology, but differ by a broader ventrolateral band of rosette-like scales (maximum width 3.7-3.8 mm vs. 2.2 mm) and the scalation of the rostral appendage which consists of more scales than in F. t i m o n i (7 vs. 4 scales between nostril and tip of rostral appendage). A higher number (6-8) of scales between nostril and tip of rostral appendage is also typical for adult (BMNH xxv.12a, drawings in Brygoo 1971; 1978) and subadult (ZMB 7551) males of F. bifidus and we therefore consider the two paralectotypes as subadult males of F. bifidus .

ZSM

Bavarian State Collection of Zoology

ZFMK

Zoologisches Forschungsmuseum Alexander Koenig

UADBA

University dAntananarivo, Department de Biologie Animale

Kingdom

Animalia

Phylum

Chordata

Class

Reptilia

Order

Squamata

Family

Chamaeleonidae

Genus

Furcifer

GBIF Dataset (for parent article) Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF