Fan-throated lizards have undergone rapid diversification given that the dated tree of Blankers et al. (2012) indicates that Otocryptis and Sitana diverged beginning ~12 million years ago. In Peninsular India, the three Sitana clades are geographically separated with some overlap between the Sitana spinaecephalus clade and the Sarada clade. Each clade harbors multiple species that exhibit varying degrees of genetic divergence. Traditional morphological characters used in lizard taxonomy, such as the number of scales around the body, ventral scales, labials, and lamellae scales, were not useful for separating most species. This overlap in morphological characters may be due to their occupancy of similar open habitats in the subcontinent and/or shared ancestry. However a combination of meristic and morphometric characters were useful to diagnose species.
Genetic relationship and distributionnext section
It is already established that Sitana is the sister taxon of Otocryptis wiegmanni from Sri Lanka (Macey et al., 2000). The ND2 sequences compared with other genera from Peninsular India and Sri Lanka shows that the minimum sequence divergence (uncorrected distance) is between Lyriocephalus and Cophotis (17.9%) and the maximum divergence is between Draco blanfordii and Sitana (38.1%). The three species in the new genus Sarada gen. nov. are highly divergent (20-24%) from all species in the genus Sitana. Species in Sarada gen. nov. are geographically restricted in distribution and are sympatric with members of Sitana spinaecephalus clade at multiple locations in their distribution (Fig. 8). Sitana spinaecephalus sp. nov. is the most widespread species, adapted to both lowland and high elevations in Gujarat and Maharashtra. Even though the Peninsular Indian landscape looks continuous, there are several hill ranges and rivers that could have played a significant role in the diversification of Sitana and Sarada. Based on the current distribution of the lineages, some of the hill ranges and rivers in Peninsular India seem to have served as a barrier to gene flow. For example, in the case of Sitana spinaecephalus sp. nov. and Sitana laticeps sp. nov., the Godavari river along with the mountain ranges (Trimbak Hills & Balaghat Hill Range) may have served as a barrier for these species. The Krishna, Bhima and Godavari rivers along with the mountain ranges (Mahadeo Range) may have played a role in Sarada deccanensis comb. nov. and Sarada darwini sp. nov. distributions. Sarada superba sp. nov. appears to be a species restricted to the high elevation plateaus of Satara, Maharashtra. During our sampling, we only found this species on plateaus; searches at elevations below 1000 meters did not yield any individuals. Interestingly, and not surprisingly, species that are closely related to Sitana bahiri from Sri Lanka are found on Rameshwaram Island and one inland location in Tamil Nadu. These locations are just across the narrow Gulf of Mannar that separates Tamil Nadu from Sri Lanka. Studies show that India and Sri Lanka had terrestrial connections in the recent past (most recently ca. 17,000 years ago) (see Voris 2000), which may have allowed these species share genes until quite recently (e.g. Guptha et al., 2015). Although we sampled extensively in peninsular India we did not cover the entire range of Sitana distribution. The genetic relationship of species from the Sitana sivalensis complex with the other Sitana spp. needs to be investigated. The only fine scale phylogeographic study in Peninsular India is on Geckoella, a predominantly forest dwelling lizard, which suggests that there are several undescribed species from the hills in the dry zones of Peninsular India (Agarwal and Karanth, 2015). Sitana and Sarada gen. nov. predominantly occur in open habitats in the dry zones of Peninsular India. Even though the Peninsular Indian landscape looks continuous, there are several hill ranges and rivers that could have played a significant role in the diversification of Sitana and Sarada.
Osteology. Baig et al. (2012) noted that generic or suprageneric level variation is observed in various bony elements of agamids. However, he reported that variation within genera is low. The majority of agamids have 21 or 22 trunk vertebrae (Moody, 1980). Reduction of the trunk vertebrae to 18, 19 or 20 has occurred in phenotypically different agamids viz. the slow moving (terrestrial) Moloch, fast moving facultative bipeds (terrestrial) Sitana and Otocryptis and the slow moving short-limbed (arboreal) Cophotis ceylanica and Ceratophora stoddartii (Moody, 1980). Interestingly our results shows that the new genus Sarada gen. nov. also has one reduced trunk vertebra compared to the genus Sitana. Additionally, all the examined skeletons of Sitana and Sarada in this study lacked the fifth toe. The function of hooked fifth metatarsal as a ‘heel bone’ is discussed in detail by Russell and Rewcastle (1979). This trait seems to have been conserved in both Sitana and Sarada gen. nov. All of the species in the genus Sitana and Sarada gen. nov. are terrestrial and are facultative bipeds which may be one of the reasons this trait is conserved in all known Fan-throated lizards. Drastic reduction in finger phalanges (formula 2:2:3:3:2) is found in Moloch (Moody, 1980). In the case of Sitana spinaecephalus sp. nov. and Sitana ponticeriana the formula is 2:3:4:4:3 and in Sarada deccanensis comb. nov. and in Sarada darwini sp. nov. there is an additional phalange in the fourth finger (2:3:4:5:3).
Hemipenial morphology and external morphology. Hemipenial morphology may be a useful character in Sitana taxonomy as some species have unique micro ornamentation. Hemipenial morphology has been identified as a useful line of evidence for species delimitation in many lizards (e.g. Arnold 1986; Glaw et al., 2012; Nunes et al., 2012). Iguanian and anguimorphan lizard taxa with visual communication and sexually dimorphic epigamic characters usually have less diverse genital structures (Bӧhme and Ziegler, 2009). On the other hand, taxa with less developed visually epigamic characters and highly developed chemical intersexual communications usually have diverse genital structures (Bӧhme and Ziegler, 2009). Interestingly, this hypothesis does not seem to hold true in the case of Sitana, which have one of the most elaborate epigamic characters along with remarkable hemipenial diversity. Recent studies on hemipenial morphology of agamid lizards from Sri Lanka suggests that there is low variation in the shape of hemipenis among genera (Maduwage et al., 2008; Maduwage and Silva, 2012) and most of the Sri Lankan agamids have epigamic characters. Sitana bahiri from Sri Lanka has a bilobed hemipenis which has very different micro-ornamentation compared to its sister species Sitana ponticeriana and Sitana visiri sp. nov. (See Fig. 3 in Maduwage and Silva, 2012).
Variations in dewlap size are an important key in identification of Sitana, which is also evident in the raw data and PCA results. The members of Sarada deccanensis clade are not only genetically very distinct from other Sitana species but also morphologically distinct.
Understanding species diversity is an essential step for conservation planning, especially when cryptic species exhibit narrow distributions. Herpetofaunal studies in Peninsular India have been largely focused on the global Biodiversity Hotspot, the Western Ghats, a mountain chain running along the west coast of India (Das, 2002). However, recent explorations of areas outside the Western Ghats of Peninsular India have resulted in the discovery of many species new to science (Bauer and Das, 2000; Agarwal et al., 2012, 2013; Datta-Roy et al., 2013). Similarly, our study of Sitana also uncovered many new species from the dryer parts of Peninsular India. We thus suggest that there is an urgent need to document biodiversity in these ignored landscapes of Peninsular India. Since Sitana in most of India were considered to be a single species, it is currently placed in the category ‘Least Concern’ by IUCN (2015). Species with restricted distributions are likely to be more vulnerable to extinction than the widespread ones. Therefore, developing a better understanding of the diversity and distribution of Peninsular Indian lizards is necessary to plan any successful conservation efforts in the dry zones. Our study, which reveals hitherto unknown diversity of Sitana across India, is a step towards understanding this poorly studied group of agamids. Although our sampling efforts were quite robust covering most of the biogeographic regions of India, due to time constraints there is poor fine scale data on the distribution of the species diversity. Focused and long-term studies are essential to address this problem. Although the taxonomy of Fan-throated lizards is much improved, due to the lack of proper distribution data and other natural history observations, we consider all the species as ‘Data Deficient’ regarding IUCN status.