Prasinohaema Greer 1974

Biogeography and evolution of Prasinohaema Prasinohaema scurrula sp. nov. is thus far only known from montane areas of the South-fold Mountains in western Papua New Guinea. While herpetofaunal surveys in these mountains over the last 20 years have resulted in the description of many new montane frog species (e.g., Richards & Oliver 2010; Oliver et al. 2012; Iannella et al. 2014; Oliver & Richards 2019; Günther & Richards 2021), P. scurrula sp. nov. is only the fourth new species of montane lizard recognised from this area over the same time period, with the other three being Cyrtodactylus medioclivus Oliver, Richards & Sistrom, 2012, Lobulia marmorata Slavenko, Tamar, Tallowin, Kraus, Allison, Carranza & Meiri, 2021 and Papuascincus eldorado Slavenko, Richards, Donnellan, Allison & Oliver, 2024. Of these, three belong to the Lobulia group. This contrasting number of frog versus lizard descriptions is likely a cumulative outcome of lower lizard diversity at high elevations (Tallowin et al. 2017), a greater research focus on frogs, and potentially more difficulty finding often ecologically cryptic lizard taxa. Two of the three afore-mentioned species (C. medioclivus and P. eldorado) have wide distributions that span several degrees of longitude but within narrow elevational bands. This pattern of distribution emphasises that for at least some montane taxa there has been connectivity within climatically similar elevational bands spanning large areas of the South-fold Mountains. Lineage diversification and ecological divergence in skinks of the Lobulia group appears to be closely associated with mountain uplift and segregation across elevational gradients (Slavenko et al. 2020, 2022). Prasinohaema scurrula sp. nov. does not appear to be a truly ‘alpine’ species, with no records above 3,000 metres a.s.l. However, it is clearly a montane species and occurs at the highest elevations at which Prasinohaema sensu stricto has been recorded. The sister genus to Prasinohaema sensu stricto, Nubeoscincus, contains two species, with some of the highest elevational records (~ 3,000 –4,000 m a.s.l.) of any skink in New Guinea, and indeed the world (Greer et al. 2005; Slavenko et al. 2022). This emphasises the overall close association between the Lobulia group, and the Prasinohaema - Nubeoscincus lineage in particular, and mountains. Our divergence dates for Prasinohaema sensu stricto (~16 Mya) also support models and analyses suggesting that extensive uplift in at least some parts of what is now New Guinea’s Central Cordillera date back well into the Miocene (Quarles van Ufford & Cloos 2005; Tallowin et al. 2018; Slavenko et al. 2020, 2022; Toussaint et al. 2021). When compared against its congeners P. scurrula sp. nov. has a relatively robust build and less extensive lamellae on the digits, suggesting more terrestrial habits. However, the paratype was collected high up in a Pandanus tree, suggesting arboreality. The two other observations we have of this species in the wild were of lizards on or close to the ground, but at forest edges or clearings. The morphology of P. scurrula sp. nov. is therefore reminiscent of arboreal lizards such as the ‘twig anole’ ecomorph, which is characterised by short limbs and tails and small lamellae (Williams 1983; Losos 2009), rather than trunk dwelling species. In New Guinea’s higher mountains (> 2,500 m a.s.l.) largely terrestrial skink lineages such as Carlia Gray, 1845 spp., Emoia spp. and other radiations of sphenomorphin skinks currently assigned to the genus Sphenomorphus Fitzinger, 1843 are relatively rare or absent. In contrast arboreal and semi-arboreal species such as Lobulia spp., Prasinohaema sensu stricto spp. (including the new species described here), and some members of Ornithuroscincus dominate. The elevational distributions of these different skink radiations may provide evidence to support the argument that vertebrate lineages in the tropics will often show increasing arboreality as elevation increases (Scheffers et al. 2013). The basis of this argument is that there is generally little overlap in temperature regimes between low- and high-elevation ground and soil microhabitats (Scheffers & Williams 2018), meaning that lowland terrestrial species are unlikely to be able to extend to high elevations in the tropics (Scheffers et al. 2013). In contrast, arboreal tropical species are likely to have broader physiological tolerances and wider distributions (Withers 1981; Adolph 1990; Wikramanayake & Dryden 1993). Based on these observations, Scheffers & Williams (2018) concluded that arboreal lineages are more likely to have broader niches, higher dispersal capabilities (Scheffers et al. 2017) and increased diversification rates (Moen & Wiens 2017). Greater physiological tolerance in the more arboreal Lobulia group is one potential explanation for the greater success of this lineage in montane habitat. In turn, ecological release could have allowed members of the more arboreal Lobulia group to diversify and exploit vacant terrestrial niches once they colonised montane areas. We suspect this may be the case both for the true ‘alpine’ species—such as Alpinoscincus spp. and Nubeoscincus spp., which occupy rocky outcrops, clearings and grasslands above the tree line (Greer et al. 2005)—and in the heliothermic and basking Papuascincus spp. To test this hypothesis further a more formal analysis of the evolutionary trajectory, historical biogeography and correlates of major ecological shifts in the Lobulia group would be needed, combined with more rigorous studies on the ecophysiology of New Guinea’s montane skinks. By better documenting the species diversity of this radiation we are hopefully improving the framework needed for such an analysis. The description of P. scurrula sp. nov. also emphasises the many residual issues associated with the taxonomy of Prasinohaema. Greer (1974) originally defined the genus largely based on the presence of a prehensile tail with a glandular tip and green blood plasma due to extremely high concentrations of biliverdin (Greer & Raizes 1969; Austin & Jessing 1994). However, Greer recognised that the genus shows variation in reproductive mode and the condition of the secondary palate. With the advent of molecular phylogenetics, it has become evident that the genus Prasinohaema is not monophyletic and that key traits—including green blood plasma—probably evolved several times independently (Rodriguez et al. 2018; Slavenko et al. 2022), meaning a revised diagnosis of the major lineages of so-called “green-blooded” skinks is needed. Additionally, Rodriquez et al. (2018) provided evidence of unrecognised taxa in both Prasinohaema sensu stricto and Prasinohaema sensu lato virens. It has also been known for decades that Prasinohaema sensu lato semoni likely comprises a species complex (G. Shea pers. obs.), and we are aware of additional morphologically distinctive populations of named species that have as yet not been subject to genetic analysis. As noted above the status of P. flavipes paniaiensis remains unresolved, but it seems likely that this form is not conspecific with P. flavipes; indeed it shares several similarities with P. prehensicauda. Our genetic data build on this background by further emphasising deep genetic divergences between populations ascribed to the nominal taxa P. flavipes and P. prehensicauda and highlighting a need for detailed taxonomic investigations of these taxa. This in turn highlights the broader pattern that skinks are both the most diverse, and the most poorly known, of Melanesia’s major reptile radiations (Slavenko et al. 2023). Addressing these knowledge gaps is important to improve our understanding of Melanesian biogeographic history, and to better elucidate centres of diversity and endemism across the region. Nomenclature of skink scales The large variation in definitions of skink scales, particularly on the head and around the eyes, has created issues when comparing states across different publications. We provide here a thorough overview of different scale definitions used in skinks, particularly in the Australasian region, and stabilise the nomenclature. An accompanying labelled diagram of the scale definitions we used can be found in Figure 2. Taylor (1936) defines the supraciliaries as being the line of small scales bordering the lateral margin of the supraocular scales, from the first scale contacting the prefrontal to the last scale wedging between the last supraocular and parietal, and overlapping the uppermost postsubocular, and this definition was followed by us and, among others, Shea & Greer (2002), Zug (2004), Das & Austin (2007), Kraus (2007, 2018, 2020), Shea (2017) and Shea & Allison (2021) in the Indopapuan region. Others (e.g., Hardy 1977; Horner 2007; Grismer et al. 2011; Hedges & Conn 2012) exclude the posteriormost supraciliary, either leaving it unnamed (Horner 2007), or referring to it as an anterior postocular (Hardy 1977), an upper secondary postocular (Hedges & Conn 2012), an anterior primary temporal (Grismer et al. 2011) or an anterior pretemporal (Greer 1983; Greer et al. 2005; followed by Slavenko et al. 2022, 2024). Greer’s (1983) creation of the name pretemporals for the posteriormost supraciliary and posteriormost postsubocular of Taylor’s nomenclature was originally due to the combined fate of these two scales in some Australian Lerista Bell, 1833 species, in which they are fused into a single scale. These scales are not fused in New Guinea skinks, making the term pretemporals superfluous in this geographical context. Other definitional variation exists at the anterior end of the row of supraciliaries. Hardy (1977), working on New Zealand skinks, excluded the first supraciliary from the count, identifying this as an upper preocular. Taylor (1936) identified as presuboculars the row of scales bordering the dorsal margins of the supralabial scales, from the first scale posterior to the s