Callulina kreffti Nieden, 1911
Original Published Description:
Krefft's Secret Frog (English), Krefft's Warty Frog (English)
Species of this genus are known to exude a gummy substance from dermal glands when disturbed. Callulina kreffti Pupil is horizontal. The tongue is wide, elliptical, margin complete, free posterior. The tips of the fingers and toes are strongly broadened, similar to the genus Callula. The outer metatarsals are connected.
Named after Dr. P. Krefft who discovered them.
The length in the biggest examined specimen is 41 mm from the tip of the snout to the anus. Except from this specimen, a female whose ovaries contain conspicuously large eggs of a diameter up to 3 mm, several smaller specimens between 20-26 mm long, are from Dr. Kreffts collection.
Body is very stout. Snout is short, triangular if seen from above, with a rounded tip, as long as the horizontal eye diameter. Interorbital space is approximately 1½ times as wide as the upper eyelid. Nares much further away from the anterior eye margin than from the tip of the snout. Finger and toes are short, with strongly broadened tips. The first finger is half as long as the second. Toes are nearly without webbing, only the 4th and 5th being connected by skin webbing between the first toe segment. Subarticular tubercles are very conspicuous. At the tarso-metatarsal joint there are two broad, “flatrounded” metatarsal tubercles, which are in contact in the middle line of the foot. With adpressed hind limbs, the tibio-tarsal joint reaches approximately to the middle of the fore leg and the eye. Skin is very rich of glands, on all sides of the animal very granular, in most specimens the back is stronger granulated than the vent, but the opposite condition does occur.
The colour is very different in all examined specimens. The strongly granulated ones are blackish grey above, and whitish or yellowish below with unconspicuous dark marks [mottling?]. Another specimen is greyish brown with greenish toning, that is most obvious on the back. Again another one is rust-brown, in the anterior part of the back greenish grey. The two specimens mentioned lastly do have a somewhat longer and a shorter spot on each side in the region of the shoulder girdle, and in the middle of the back several, partly inconspicuous spots arranged in 2 or 3 horizontal rows. The ZMB additionally houses one specimen collected by Prof. Vosseler in Amani as well, and one specimen collected by the already mention head of the station, Martienssen, collected in Tanga; the latter being characterised by especially strongly developed glands on the back.
A very small soft omosternum is present. Sternum is absent. Articular processes of the sacral vertebra strongly are broadened. Terminal phalanges of fingers and toes are T-shaped.
Habitat and Ecology
It lives on the forest floor, hiding under rotting logs or stones. It can also climb to at least 2m in the vegetation, and has been found in the leaf axils of both wild and cultivated bananas.
These toads have been found in the stomachs of Crotaphopeltis tornieri.
Evolution and Systematics
To test the distinctiveness of Callulina species and infer their evolutionary relationships we constructed two alignments of sequence data for all six described species: (1) all Callulina individuals for which 12S and 16S were available (2) all individuals for which 12S, 16S and cytb were available. The two datasets comprised (1) 18 individuals with 762 unambiguously aligned characters, of which 533 were constant, 89 variable, and 140 informative under parsimony (only 1 gap present, alternative coding of this gap made no difference in resulting trees), and (2) 14 individuals with 1185 unambiguously aligned characters, of which 884 were constant, 98 variable, and 203 informative under parsimony. Phrynomantis was used as outgroup as in previous analyses, following Loader et al., (2009). Analysis of Alignments 1 and 2 yielded single most parsimonious (MP) trees of 422 and 870 steps, respectively, and the trees (not shown) are not notably different to those recovered in likelihood and Bayesian analyses. Maximum likelihood (ML) and Bayesian analyses were conducted with the GTR+I+G model. Both analyses supported the monophyly of Callulina, and rejected suboptimal topologies in most tests. For Alignment 1 both analyses demonstrate the genetic distinctiveness of the named species, with highly supported reciprocally monophyletic groupings recovered for each, and suboptimal topologies significantly different in MP tests (apart from C. stanleyi - see Figure 8a). The truncate finger/toe species (C. dawida, C. laphami, C. shengena) form a paraphyletic assemblage outside a clade of all expanded finger/toe species. The sister species grouping between C. shengena and C. laphami is weakly supported. The expanded finger/toe clade of C. kreffti, C. kisiwamsitu, and C. stanleyi is weakly supported, though the sister pair of the morphologically similar C. kisiwamsitu and C. stanleyi is strongly supported. Both analyses of Alignment 2 recover multiple individuals of each named species (where included) together in maximally supported clades and significantly different from suboptimal topologies. Different from Alignment 1, C. shengena is recovered as the sister to all other species instead of grouping with C. laphami, though this is weakly supported. Analyses of Alignment 2 (as with Alignment 1) recovered a truncate toed paraphyletic assemblage outside a clade of all expanded finger/toe species. Constrained topologies including a monophyletic North and South Pare radiation (C. stanleyi, C. lapami, C. shengena), or a monophyletic South Pare (C. stanleyi, C. shengena) radiation are both statistically significantly suboptimal for both alignments using MP and ML (p= <0.05). Pairwise distances show high genetic distinctiveness of all named Callulina species (>4.2%, mean of 8.4% among all species).
IUCN Red List Category and Justification of Conservation Status
Callulina kreffti is currently considered ‘Least Concern’ because of its seemingly wide distribution and tolerance to habitat modification; however, Loader et al. 2010 suggest changes. The IUCN rests their listing on the assumption that several populations outside of the East Usambara are referable to C. kreffti (e.g. Ukaguru, Nguu, Nguru, Uluguru, Mahenge and Udzungwa; IUCN, 2009). Based on the pattern of local endemism in Callulina revealed in this study, similar patterns in other amphibians (e.g. Blackburn, 2008), and other taxon groups (Stanley et al., 1998; Bowie et al., 2004; Fjeldså et al., 2006), this assumption must be questioned. Loader et al. (2010) recommend that Callulina occurring outside of the Usambara, Pare, and Taita Hills should, in the first instance, be considered Callulina sp. and not identifiable as a named species without careful consideration of morphological, molecular and/or acoustic data.
Despite being somewhat adaptable, it is probably adversely affected by ongoing forest loss, especially for small-scale agriculture. Its habitat in the East Usambaras has recently come under serious threat as a result of the activities of illegal gold miners.