(FIGS 2, 45A, B, 46A, 60B; TABLE 2)

Type material: Holotype, BNHS 4191, an adult male, SVL 27.9 mm, from Kirundadu, Kodagu (formerly Coorg), India; paratypes, BNHS 4192, OMNH Am 11412, two adult males, from Kudremukh, Chikmagalur, Karnataka.

Variation: Measurements of seven samples, including the holotype, are given in Table 2.

Colour in life: Some colour variation is observed in individuals of the same population within a small geographical area. The dominant colour form is yellowish brown with light-brownish lines (BNHS 4477, Fig. 45A), or discontinuous light-brown lines making a spotted appearance on a light-yellowish white background (BNHS 4476); however, the degree of brown colour is reduced in certain specimens, and the most extreme colour is golden yellow without markings (BNHS 4479, Fig. 45B), other than faint spots towards the snout and light-grey cross bands on limbs (which are more visible in preservation). In almost all of the colour forms (except SDB 1106), at least a few faint dorsal lines and cross bands on both limbs are visible in life and preservation: BNHS 4478 has a uniform light-brownish yellow dorsum with numerous minute brown spots, and faint brown cross bands on both limbs in life; the dorsum turns greyish brown with light-brown spots in preservation. In all colour forms the iris is light brown encircled with a bluish green outer ring (Fig. 45A, B).

Other material studied: BNHS 4476 and BNHS 4477, two adult males, from Kudremukh; BNHS 4478, an adult male, from Mercara; BNHS 4479, an adult male, from Muthodi; BNHS 4480, an adult male, from Sakleshpur; BNHS 4532, an adult male from Jog falls (Table 2).

Diagnosis: Philautus luteolus can be distinguished from known congeners by the following combination of characters: (1) medium adult male snout–vent length (SVL 26.8 ± 1.7 mm, N = 7); (2) snout pointed; (3) canthus rostralis rounded; (4) dorsum yellow or yellowish brown, usually with four–six faint brownish discontinuous lines from snout to vent; (5) loreal and tympanic regions golden yellow or yellowish brown.

Because of the typical dorsal coloration, P. luteolus cannot be confused with any of the Western Ghats Philautus except P. travancoricus. Philautus luteolus differs from P. travancoricus by its pointed snout (vs. oval snout), canthus rostralis rounded (vs. indistinct), dorsum uniformly granular (vs. shagreened), lateral side yellowish (vs. light brown), and absence of a dark-brown streak on either side of the snout (Figures 45A and B, 60B; vs. presence of dark-brown streak on either side of snout, Figs 59B, 60A, 61, 62).

Description of the holotype: A detailed morphological description and illustrations were published in Kuramoto & Joshy (2003). Measurements are given in Table 2.

Distribution and natural history: Jog Falls, Mavingundi, Kudremukh-Malleshwaram, Sakleshpur, Kempholay, Kirundadu, Madenadu, Mercara, and Muthodi in Karnataka (Fig. 46A, Table 1). This species was predominantly found in disturbed habitat near coffee estates adjacent to secondary forest or wayside vegetation. Individuals were observed either on leaves or stems of shrubs about 1 m above the ground. In Mercara, males started calling from under the leaf litter at dusk (about 18:00 h) in low frequencies, and then climbed onto the vegetation where they continued calling during the night-time until 22:00 h.

Remarks: Gururaja et al. (2007a) described a new taxon, Philautus neelanethrus, from Shimoga District in Karnataka state. We have examined the type series (BNHS 4510 and BNHS 4511, adult males) and could not find any substantial morphological difference with P. luteolus. The morphological distinctions mentioned in the original publication, ‘lack (indistinct) of tympanum and supratympanic fold, snout length equal/ subequal to eye diameter, and a distinct blue ring on the outer margin of the eye’, are all present in our collection of P. luteolus. Furthermore, we compared the published sequence of the 16S rRNA gene of P. neelanethrus with our P. luteolus sequence from near the type locality. Apart from a few likely sequencing errors at the very start of the P. neelanethrus sequence (i.e. constant in all Philautus, except P. neelanethrus), the sequence of the remaining 520 bp only differed in a single substitution in a variable region of this gene fragment. We therefore consider that there is currently no evidence for recognizing P. neelanethrus as distinct, and regard it as a new synonym of P. luteolus.