Trioceros kinangopensis, Stipala, Jan, Lutzmann, Nicola, Malonza, Patrick K., Wilkinson, Paul, Godley, Brendan, Nyamache, Joash & Evans, Matthew R., 2012

Stipala, Jan, Lutzmann, Nicola, Malonza, Patrick K., Wilkinson, Paul, Godley, Brendan, Nyamache, Joash & Evans, Matthew R., 2012, Mountains in the central highlands of Kenya, Zootaxa 3391, pp. 1-22 : 15-17

Trioceros kinangopensis sp. nov.

( Figures 2 A, 6A/ B/ C/ E)

Holotype. NMK 3071/12, adult male, Kinangop Peak (3500m), Aberdare Mountains National Park, Kenya, collected by Jan Stipala and Joash Nyamache on 24th February 2007.

Paratypes. 7 males and 5 females. NMK 3071/1, 3–5, 7, 11, males, NMK 3071/ 2, 6, 8–9, females, locality same as holotype; NMK 2536/1, male, NMK 2536/2, female, vicinity of Kinangop Peak.

Diagnosis. A small, robust-bodied chameleon that differs from other members of the bitaeniatus group in having a combination of the following characters: tail shorter than SVL in both sexes; heterogeneous body scalation (fine body scales with small scattered tubercles and two lateral rows of enlarged tubercles on each flank); low straight casque; short gular crest (orange in males); scales on the temporal region moderately enlarged and graduating in size posteriorly to merge with the body scales; snout with sloping profile and smooth canthus rostrales.

Description of holotype. Adult male, SVL = 58mm, tail = 51mm (total length = 109mm). Head with prominent tubercular head crests, the lateral and parietal crests forming a low triangular casque. The scales on the casque and between the canthal crests are weakly convex and between the supraorbital crests are more strongly convex. The parietal crest forks anteriorly, consisting of two ridges of strongly convex tubercles that meet with the posterior edge of the supraorbital crests. The temporal region is covered in weakly convex scales, larger than on the flanks but smaller than the tubercles on the fore-limbs. A row of larger scales borders the posterior margin of the orbit. The gular region consists of several rows of enlarged, strongly convex tubercles separated from the gular crest by fine scales. The gular crest is made up of short, conical tubercles that continue onto the belly as a much shorter ventral crest that terminates at the vent. The scales on the eye turrets are weakly heterogeneous, fine towards the outer margin with slightly larger, more convex scales around the eye opening.

Body scalation is strongly heterogeneous and consists of fine background scales scattered with slightly larger, convex tubercles. There are two prominent rows of large tubercles on each flank. The lower row runs between the limbs. The upper row starts above the forelimb level with the lateral crest and extends the length of the body, continuing onto the tail as a series of smaller tubercles. The dorsal crest is well-developed and consists of a single row of conical scales that extends the length of the body and continues on to the tail almost to the tip, the scales decreasing in size posteriorly. The scales of the dorsal crest are heterogeneous in size, forming groups of four scales that increase in size posteriorly within each group, the last two scales significantly larger than the ones before them. The upper surfaces of the limbs are covered in numerous enlarged, weakly convex tubercles, although smaller in size than those of the two lateral rows. A large convex tubercle is present on the flank above the point where the forelimb meets the body. The ventral surfaces of the limbs and tail are covered in fine homogenous scales. The hemipenes are not everted.

Colour in preservation. The holotype is uniform black.

Variation in paratypes. SVL and tail length measurements of all type specimens are given in Table 6 . Mean SVL: males = 54.3mm ± 4.9 (n = 8); females = 56.0mm ± 6.4 (n = 5). Largest male: 117mm total length (SVL = 63mm, tail = 54mm), largest female: total length = 112mm (SVL = 61mm, tail = 51mm).

The size and convexity of the scales in the two lateral rows is variable. In some individuals the tubercles are flat and very large in both rows, in others individuals they are moderately convex, the tubercles of the upper row somewhat smaller and more strongly convex, although always larger than the scattered tubercles on the flanks and limbs. In one individual the gular crest between the legs forms a double row of conical tubercles.

The hemipenes are not everted in any of the male specimens.

Colour in life. Males have a background colour of grey, although in some individuals this is replaced with turquoise on the lower half of the body, head, tail and limbs. The flanks, limbs and tail are banded with a lighter yellow-green. Two broad stripes on the cheeks and the lateral rows of enlarged tubercles are also a light yellow-green. The enlarged tubercles on the sides of the throat are bright yellow and the gular crest is a distinctive orange. The eye turrets are grey-turquoise with lighter yellow-green tubercles surrounding the opening.

Females have a background colour of dark brown, a more orange-brown on the dorsal keel and the top of the head. The flanks, legs and tail are banded with grey or pale yellow. The cheek stripes and lateral rows of tubercles are cream or white. The enlarge tubercles on the side of the throat are yellow and the gular crest pale yelloworange. The eye turrets grey with green tubercles surrounding the opening to the eye.

Etymology. T. kinangopensis sp. nov. is named after the collecting locality, Kinangop Peak at the southern end of the Aberdare Mountains, Kenya.

Distribution. T. kinangopensis sp. nov. appears to be endemic to the Kinangop Peak area at the southern end of the Aberdare Mountains. Specimens were collected between 3500–3600m. At lower elevations (3000–3200m) we found only T. hoehnelii . Trioceros kinangopensis appears to be absent from afroalpine vegetation in the central Aberdare Mountains between Mutubio and Kiandogoro gates (3000–3150m). We surveyed 13km of roadside vegetation, which included pure tussock-grass moorland, low ericaceous scrub and tall stands of St. John's Wort ( Hypericum revolutum ) and found 50 specimens of T. hoehnelii but no T. kinangopensis sp. nov.. Although Andren (1976) reported T. schubotzi (= T. kinangopensis sp. nov.) from the northern peaks in the Aberdare Mountains, a photograph labelled as T. schubotzi clearly shows T. hoehnelii . SVL/ tail length ratios in the same paper are within the range of T. hoehnelii , which is relatively longer-tailed than either T. schubotzi or T. kinangopensis sp. nov.. Nevertheless, the northern peaks reach a similar altitude to Kinangop Peak (approximately 4000m) and it is possible that T. kinangopensis sp. nov. may occur there.

Ecology. Trioceros kinangopensis sp. nov.. appears to be restricted to the afroalpine zone. Specimens were found in low ericaceous shrubs, which occur in patches in a habitat dominated by tussock-grass with scattered clumps of Alchemilla and megaphytic Lobelia and Senecio spp..

Conservation status. The results of our field surveys suggest that T. kinangopensis sp. nov. may be restricted to elevations above 3500m on Kinangop Peak and its range may be only 10km 2. Although the distribution of T. kinangopensis sp. nov. is completely within the Aberdare National Park, burnt woody remains of ericaceous shrubs suggest that extensive fires have affected the entire peak. Fires are reported to occur annually in the afroalpine zone on Mt. Kenya and may be quite extensive, sometimes burning large areas (> 10km 2)( Coe 1967, Bongo Woodley, pers. comm.). There are no published reports on the frequency of fires in the afroalpine zone on Kinangop Peak, and the impact of fires on chameleons populations in the afroalpine zone has not been studied. Fires typically result in a total loss of surface vegetation and are likely to lead to the extinction of local chameleon populations in the short-term. The frequency of fires and the ability of chameleons to re-colonise burnt areas may result in chameleons being absent from much of their potential range. There may also be other long-term negative effects of fires due to population fragmentation and loss of genetic diversity through extreme population fluctuations. Further research is recommended to learn more about the distribution, ecology and the impact of fires on T. kinangopensis sp. nov. and T. schubotzi . Although little in known about T. kinangopensis sp. nov., it has a very restricted distribution and lives in a habitat that is subject to fires. Therefore we suggest that under IUCN guidelines (IUCN 2010) it may be reasonable to categorise T. kinangopensis sp. nov. as either Endangered (E) or Critically Endangered (CR).

Figures

FIGURE 2. Male head morphology. A— male (NMK L / 3071 / 12), Kinangop Peak; B— T. schubotzi (NMK L / 2325), Mt. Kenya; C— T. nyirit, Cherangani Hills (NMK L / 3166 / 1). T. nyirit differs from Kinangop Peak specimens in having a prominent rostral process and elevated casque. T. schubotzi differs from the Kinangop Peak specimens in having an angular snout profile, present as a short rostral process in many individuals, and enlarged scales across the entire temporal region.

Tables

TABLE 6. Morphometric measurements (in millimetres) of the type series of T. kinangopensis sp. nov.

Type	Locality	Museum No.	Sex	SVL	Tail length	Head length	Head width
holotype	Kinangop peak	NMK3071/12	male	58	51	17.1	8.7
paratype	"	NMK3071/1	"	52	48	15.8	8.1
"	"	NMK3071/2	"	53	48	16.2	9.0
"	"	NMK3071/3	"	49	41	15.6	8.4
"	"	NMK3071/5	"	63	54	18.7	8.7
"	"	NMK3071/7	"	48	39	17.6	9.1
"	"	NMK3071/11	"	57	49	16.7	8.7
"	"	NMK2536/1	"	54	46	16.4	8.6
"	"	NMK2536/2	female	61	51	19.2	10.4
"	"	NMK3071/6	"	60	48	18.4	9.3
"	"	NMK3071/8	"	47	39	14.6	8.1
"	"	NMK3071/9	"	56	45	17.0	8.8