Calumma fallax ( Mocquard, 1900 )

Prötzel, David, Scherz, Mark D., Ratsoavina, Fanomezana M., Vences, Miguel & Glaw, Frank, 2020, Untangling the trees: Revision of the Calumma nasutum complex (Squamata: Chamaeleonidae), Vertebrate Zoology 70 (1), pp. 23-59 : 48-52

Identity and re-description of Calumma fallax ( Mocquard, 1900)

Syntypes: MOCQUARD (1900a, 1900b) did not specify the specimen numbers of the types, but stated that they were one male and one female, collected in Forêt d’Ikongo by Grandidier ( MOCQUARD, 1900a). By inference from collection data, we follow BRYGOO (1971), GEHRING et al. (2011) and KLAVER (2019) in accepting only two of potentially six specimens as syntypes, MNHN 1899.317, an adult male, and MNHN 1899.318, an adult female, from the locality Forêt d’Ikongo, collected in 1898 – 1899 by Guillaume Grandidier.

Lectotype designation: We designate MNHN 1899.317 , the adult male syntype, as the lectotype of the species, the remaining syntype, MNHN 1899.318 , becoming the paralectotype.

Assignment to genetic clade (based on a comparison of the lectotype with the male specimens assigned to clade H, ZSM 693/2003, ZSM 286/2010, ZSM 685/2003, and ZSM 694/2003): Calumma fallax belongs to clade H, according to the high casque (2.1 mm in the lectotype vs 1.3 – 2.5 mm), rounded, oval rostral appendage with a length of (2.7 mm vs 1.8 – 4.3 mm), heterogeneous scalation in head region with diameter of largest scale in temporal region of 1.6 mm vs 0.8 – 1.5 mm, distinct parietal crest ending in the casque, temporal crest present consisting of 1 or 2 tubercles (2 vs 1 – 2); osteology of the skull is almost identical, e.g. width of the frontoparietal fenestra with 16.4% of skull length vs 13.9 – 15.4%.

Referred material: The specimens MNHN 1890.430 , MNHN 1890.431 , MNHN 1890.432 , all three adult males, and MNHN 1888.24 , adult female, are non-type specimens. In addition, in anticipation of our conclusions on the taxonomic identity of the species, we here refer the following specimens to C. fallax as it is here re-defined : ZSM 685 /2003 ( FG / MV 2002-0291 ) , ZSM 693 /2003 ( FG / MV 2002-0317 ) and ZSM 694 /2003 ( FG / MV 2002-0318 ), all three adult males, collected in Ranomafana NP, Vohiparara, near Kidonavo bridge (about 21.22°S, 47.37°E, about 1000 m a.s.l.), Vatovavy-Fitovinany Region, eastern Madagascar, on 16– 20 January 2003 by F. Glaw, M. Puente, L. Raharivololoniaina, M. Thomas, D . R. Vieites ; ZSM 134 /2005 ( FGZC 2508 ), female, collected in Andohahela NP (24.5440°S, 46.7141°E, 1548 m a.s.l.), Anosy Region , southeastern Madagascar, on 27 January 2005 by F. Glaw, M. Vences, P. Bora ; ZSM 286 /2010 ( FGZC 4588 ), adult male, collected east of Tsinjoarivo , between camps 2 and 1 (19.7103°S, 47.8182°E, 1465 m a.s.l.) on 23 April 2010 by F. Glaw, J. Köhler, P.-S. Gehring, J.L. Brown, E. Rajeriarison ; ZSM 313 /2006 ( ZCMV 2930 ), adult female, collected in Ranomafana NP, Vohiparara river and stream/swamp (about 21.25°S, 47.40°E, about 1100 m a.s.l.) on 20 February 2006 by M. Vences, E. Rajeriarison, Y. Chiari, E. Balian ; ZSM 476 /2010 ( FGZC 4352 ), adult female, collected in Anjozorobe region , Mananara Lodge (18.4629°S, 47.9381°E, 1298 m a.s.l.), Analamanga Region, eastern Madagascar, on 6 April 2010 , ZSM 479 /2010 ( FGZC 4575 ), adult female, collected east of Tsinjoarivo , camp 1 (19.6800°S, 47.7706°E, 1607 m a.s.l.) on 19 April 2010, both by F. Glaw, J. Köhler, P.-S. Gehring, M. Pabijan, K. Mebert, E. Rajeriarison, F. Randrianasolo, S. Rasamison ; ZSM 149 /2016 ( FGZC 5226 ) and ZSM 150 /2016 ( FGZC 5225 ), both adult females, collected in Mandraka (18.9122°S, 47.9144°E, 1235 m a.s.l.), Analamanga Region, eastern Madagascar, on 5 January 2016 by F. Glaw, D. Prötzel, L. Randriamanana ; ZSM 258 /2016 ( FGZC 5291 ), adult female, collected in Mandraka (18.9133°S, 47.9145°E, 1260 m a.s.l.) on 3 August 2016 by F. Glaw, D. Prötzel, J. Forster, N. Raharinoro .

Diagnosis (based on the type series and the referred material, see above; osteology based on micro-CT scans of MNHN 1899.317, MNHN 1890.430, ZSM 693/2003, and ZSM 286/2010, all four males): Calumma fallax is characterised by (1) a medium size (male SVL 42.9 – 50.6 mm, female SVL 40.8 – 50.7 mm; male TL 90.9 – 107.3 mm, female TL 77.3 – 99.8 mm), (2) a long (1.8 – 4.3 mm in males, 1.7 – 3.2 mm in females) and distally rounded rostral appendage, (3) rostral scale not integrated into the rostral appendage, (4) prominent rostral crest forming a concave cup on the snout, (5) lateral crests present, (6) temporal crest generally present, (7) cranial crest generally absent, (8) parietal crest generally present but short, (9) a distinctly raised casque in males with a height of 1.3 – 2.5 mm, (10) a dorsal crest of 6 – 11 cones in males, generally absent in females (one specimen with five cones), (11) 10– 16 supralabial scales with a straight upper margin, (12) absence of axillary pits, (13) diameter of the largest scale in the temporal region of the head 0.8 – 1.8 mm, (14) a frontoparietal fenestra in the skull, (15) parietal and squamosal generally in contact, (16) parietal bone width at midpoint 6.7 – 15.7% of skull length, (17) a generally greenish, greyish, or brownish body colouration, (18) a typically blue or grey nose in non-stressed colouration, (19) a green cheek colouration, (20) three blue dorsoventral stripes on the body and a white lateral stripe, and (21) a diffuse brown strip crossing the eye.

C. fallax can be distinguished from all species of the C. boettgeri complex (see above) by the absence of occipital lobes; from C. gallus by different length, shape and colour of its rostral appendage (see above); from all other species of the C. nasutum group without occipital lobes (except for C. ratnasariae , see below) by the presence of a frontoparietal fenestra.

In addition, it can be distinguished from C. vatosoa by the presence of a rostral appendage (vs absence); from C. vohibola by longer rostral appendage (RRS 4.2 – 8.5% vs 0.2 – 3.1%), supralabials with a straight upper margin (vs serrated), parietal crest generally present (vs absent); from C. nasutum as here redefined by general absence of cranial crest (vs present), a shorter frontal (39.4 – 50.4% of skull length vs 51.2 – 82.1%), blue rostral appendage (vs brown), and three blue lateral blotches (vs four brown blotches with light spots); from C. radamanus by rostral scale not integrated into the rostral appendage (vs generally integrated), parietal crest generally present (vs absent), supralabials with a straight upper margin (vs serrated), parietal and squamosal in contact or closely approaching (vs widely separated), and width of parietal at midpoint (6.7 – 15.7% vs 16.1 – 22.4%); from C. emelinae sp. nov. by general presence of parietal crest (vs general absence), higher casque in males (1.3 – 2.5 mm vs 0.5 – 1.1 mm), dorsal crest consisting of cones (vs spines) in males, and larger temporal scale in males (0.8 – 1.6 mm vs 0.7 mm); from C. tjiasmantoi sp. nov. by fewer supralabials (10 – 15 vs 15 – 17), larger diameter of temporal scale (1.0 – 1.8 mm vs 0.6 – 0.8 mm), and slightly narrower postparietal process.

Re-description of the lectotype ( Fig. 4E ): Adult male, with mouth slightly opened, in good state of preservation, hemipenes not everted. SVL 42.9 mm, tail length 53.2 mm, for other measurements, see suppl. Table 1 ; distinct rostral ridges that render the dorsal surface of the snout a concave cup, laterally compressed dermal rostral appendage of oval tubercle scales that projects straight forward over a length of 2.7 mm with a diameter of 3.4 mm, oblong but taller than wide; 11 infralabial and 11 supralabial scales, all relatively large; supralabials with a smooth dorsal margin; distinct lateral crest running horizontally; distinct temporal crest consisting of two tubercles per side; no cranial crest; distinct parietal crest; no occipital lobes; highly elevated (2.1 mm) and rather acute casque; dorsal crest present, consisting of 11 cones that decrease in height posteriorly; no traces of gular or ventral crest. Body laterally compressed with fine homogeneous scalation and distinctly larger scales on extremities and head region, largest scale in temporal region with diameter of 1.6 mm and in cheek region of 1.4 mm; no axillary or inguinal pits.

Skull osteology of the lectotype ( Fig. 14A ): Skull length 12.0 mm; snout-casque length 14.6 mm; narrow paired nasals completely separated from each other by the anterior tip of frontal that meets the premaxilla; prefrotal fontanelle and naris fused; prominent prefrontal with laterally raised tubercles exceeding more than the half of the prefrontal fontanelle; frontal and parietal smooth without any tubercles; frontal with a width of 2.7 mm (22.5% of skull length) at border to prefrontal extending to 4.4 mm (36.7%) at border to postorbitofrontal; large frontoparietal fenestra with a width of 2.4 mm (20.0%); curved parietal tapering strongly from a width of 3.9 mm (32.5%) at the border to postorbitofrontal to a width at midpoint of 0.8 mm (6.7%) and broadening slightly poterodorsally, where it is in weak contact with the squamosals; squamosals thin without any tubercles. For further measurements, see Table 2 .

Hemipenial morphology, based on the lectotype (no micro-CT scan available): small sized calyces (hemipenial character A); two pairs of finely denticulated rotulae of different size, on sulcal side large with about 16 tips, on asulcal side small with about 8 tips (B); papillary field of few, unpaired papillae (C); pair of medium sized cornucula gemina (D), only visible when hemipenis fully everted.

Variation: The osteology of the male ZSM 693/2003 differs from the other specimens in having the parietal and squamosal bone not connected; probably this is due to its juvenile state. For variation in measurements, see Table 1 . Sexual dimorphism: Body size (SVL and TL) is slightly larger in males than females. Tail length is longer in males than in females (RTaSV 102 – 124% vs 89 – 104%). Relative rostral appendage length does not differ. Dorsal crest is more pronounced in males than females.

Colouration in life ( Fig. 15 ): Weak sexual dichromatism, males slightly more colourful. In both sexes grey/ beige body colouration with three bright blue dorsoventral stripes on the body that can be crossed by a broad white lateral stripe; extremities and tail of same colour as the body, tail in males can be diffusely annulated; rostral appendage grey or blue; cheek region can be bright green; a diffuse brown stripe may cross the eye.

Etymology: A Latin adjective meaning ‘deceptive’ or ‘fallacious’ in the neutral nominative, with unclear justification.

Distribution ( Fig. 9 ): Calumma fallax as redefined here, occurs in eastern Madagascar from Andohahela in the south to Mandraka about 650 km further north (coordinates, see above), from an elevation of 922– 1781 m a.s.l.

Figures

Fig. 4. Preserved type specimens of (A) Calumma nasutum (lectotype, MNHN 6643C), adult male; (B) C. radamanus (holotype, SMF 22132), adult male; (C) C. emelinae sp. nov.(holotype, ZSM 618/2009), adult male; (D) C. tjiasmantoi sp. nov.(holotype, ZSM 735/2003), adult female; (E) C. fallax (lectotype, MNHN 1899.317), adult male; (F) C. ratnasariae sp. nov. (ZSM 35/2016), adult male.

Fig. 14. Micro-computed tomography scans of the skulls of species with a frontoparietal fenestra in dorsal and lateral view. (A) Calumma fallax (MNHN 1899.317), male lectotype; (B) C. fallax (MNHN 1840.430), male; (C) C. fallax (ZSM 286/2016), male of genetic clade H, assigned to C. fallax; (D) C. ratnasariae sp. nov. (ZSM 35/2014), male holotype. Diagnostic characters are encircled in red. Scale bars: 2.0 mm.

Fig. 15. Calumma fallax from different locations. (A) Adult male (not collected) from Tsinjoarivo in relaxed state; (B, C) adult male (not collected) from Ranomafana NP, relaxed, photos:A. Laube/T. Negro; (D, E) both adult females (ZSM 149/2016 and ZSM 258/2016) from Mandraka, relaxed.

Fig. 9. Distribution map of six species of the Calumma nasutum group in Madagascar. Localities based on collected specimens and/or DNA sequences of GEH- RING et al. (2012) and new sequences shown in Fig. 2. Black stars within the symbols indicate type localities. Colours correspond to those used in Fig. 2.