The Meratus Mountains of south-east Borneo are biogeographically interesting due to their isolation and potential endemicity. However, Meratus' birds have never been quantitatively surveyed or collected, so we know little of their community structure or taxonomy. In May 2017, we visited Mt. Besar in the Meratus range to assess its avifauna and collect specimens for phylogeographic study. Here, we report on Mt. Besar's avifauna and provide preliminary ND2-sequence assessments of their relationships. We recorded 89 species, of which we collected 68 species; 17 represented distributional extensions. Mt. Besar's avian community is depauperate compared to Borneo's main mountains, and several of its species exhibit unusual elevational abundance patterns. We attribute these findings to the range's small area, low elevation, isolation from other mountains, depleted surrounding native lowland forest, and exploitation for food and pet-trading. ND2 comparisons indicate that at least 12 Meratus species exhibit population structure that merits additional phylogeographic study.
The island of Borneo is well known for its impressive avian diversity, which includes at least 373 breeding landbird and 52 endemic species, depending on classification (Dickinson & Remsen 2013, Dickinson & Christidis 2014, Sheldon et al. 2015, Eaton et al. 2016c, Clements et al. 2017). This remarkable diversity hinges on the island's wide range of habitats, from low-elevation mixed-dipterocarp, peatswamp and kerangas forests to montane moss forests and ericaceous heath. Although most of Borneo's avian diversity inhabits the lowlands, the island's montane forests are particularly important in terms of biogeographic history, not only of the island itself but the Greater Sundas as a whole (de Bruyn et al. 2014, Sheldon et al. 2015). Borneo's mountains include the main central range, which runs north-east to southwest across most of the island and features several peaks above 2,000 m in elevation, one of which, Mt. Kinabalu, reaches 4,095 m (Fig. 1). There are also numerous smaller ranges and peaks that are isolated from the main range by areas of low elevation. As a group, Borneo's mountains—especially those on the eastern side of the island—are thought to have played a key role in fomenting and preserving much of South-East Asia's avian diversity (de Bruyn et al. 2014, Sheldon et al. 2015). The mountains supported rainforest continuously during dramatic climatic changes in the Oligocene, Pliocene and Pleistocene, when colder temperatures extirpated rainforest over much of the rest of Sundaland in favour of seasonal forest and even grasslands (Bird et al. 2005, Morley 2012, Sheldon et al. 2015). However, despite the biogeographic importance of Borneo's mountains, remarkably little is known concerning the evolution and ecology of birds there, especially those occurring in mountains outside the Malaysian states of Sarawak and Sabah. This is particularly true of birds in the Meratus range of south-east Borneo. These mountains have rarely been visited by ornithologists or birdwatchers (Davison 1997, Eaton et al. 2016a), and their birds have never been surveyed quantitatively or collected for taxonomic study. As a result, we know little concerning the biogeographic relationships or community composition of the Meratus avifauna.
The Meratus Mountains are a low-elevation range skirting the south-east corner of Borneo in the Indonesian province of South Kalimantan (Fig. 1). They run north-northeast and are c.300 km in length and 65 km wide (Witts et al. 2012), with a highest point at Mt. Besar (1,901 m). The mountains consist mainly of an early Cretaceous volcanic arc and ophiolite sequence (Moss & Wilson 1998) but, like Mt. Kinabalu, they are relatively young, having arisen in the middle to late Miocene as a result of collision and subduction of continental plates on Borneo's south-eastern side (Witts et al. 2012). Their potential as a Pleistocene refugium and area of high endemicity has long been appreciated (MacKinnon et al. 1996, Gathorne-Hardy et al. 2002, Quek et al. 2007, Raes et al. 2009). Most forest in the Meratus Mountains is intact compared to the surrounding lowlands, but it is largely unprotected from logging and development, except the southern section, which lies within the Martapura-Pleihari Reserve (Holmes & Burton 1987, Davison 1997).
To date, only two surveys of Meratus birds have been published. Davison (1997) visited these mountains between 30 September and 10 October 1996. He identified 80 species above 800 m, expanding the known range of 21 montane specialists (sensu Wells 1985). Eaton et al. (2016a) visited the mountains on 7–10 July 2016. They observed, audio-recorded and photographed birds, identified 91 species, expanded the known range of several species, and discovered what they believed to be new species of Cyornis and Zosterops. Both expeditions approached the range from the west via Loksado (c.200 m) and the village of Kadayang (c.500 m). Davison (1997) reached close to the top of Mt. Karokangan (1,686 m), while Eaton et al. (2016a) reached the peak of Mt. Besar (1,901 m). Both noted that forest was heavily disturbed below 900 m, where rugged limestone outcrops appear (Davison 1997), and mildly disturbed above this elevation. The discoveries made by these two expeditions highlight the potential importance of the Meratus Mountains to understanding the evolution of avian diversity on Borneo.
Louisiana State University Museum of Natural Science (LSUMNS) and the Museum Zoologicum Bogoriense, Indonesian Institute of Sciences (MZB), conducted a joint expedition to Mt. Besar on 1–19 May 2017. The primary purpose was to obtain molecular and morphological specimens for use in ongoing studies of Sundaic avian biogeography (e.g., Lim et al. 2011, Lim & Sheldon 2011, Chua et al. 2017, Lim et al. 2017, Moyle et al. 2017). We were also interested in the elevational distribution of birds in the Meratus Mountains, and how this distribution compares with montane communities examined elsewhere on Borneo (e.g., Harris et al. 2012, Burner et al. 2016). Here, we report on the expedition, approximate distribution of birds in the Mt. Besar region, and findings of preliminary molecular phylogeographic comparisons.
Expedition.—Our team comprised three researchers from LSUMNS (SBS, RCB, FHS), a researcher (TH) and field technician from MZB (Suparno), and a variety of local guides and porters. Like Davison (1997) and Eaton et al. (2016a), we approached Mt. Besar from Loksado in the Hulu Sungai Selatan regency of South Kalimantan province. We travelled from Loksado to the village of Kadayang (02°44′47″S, 115°33′20″E; 500 m), where we mistnetted birds on 1–6 May 2017. Near Kadayang, virtually all forest had been converted to (mostly overgrown) rubber, plots of hill rice, and gardens. On 6 May, we established camp on the Mt. Besar summit trail at a site known locally as ‘Matupa’ (02°43′30″S, 115°35′11″E; 1,150 m). We remained at this site until 19 May, setting mist-nets along the main ridge at 1,100–1,200 m and along another ridge at 1,350 m (02°43′22″S, 115°35′35″E). As noted by earlier expeditions, rubber plantation and shifting cultivation gave way to less disturbed forest at c.900–1,000 m. The forest above 1,000 m was largely intact, although there were some cleared areas of varying age, and footpaths were heavily trafficked by hunters armed with air guns (pellet guns) and persons seeking forest products.
Elevational records.—Avian species were sampled using mist-nets and by opportunistic observation based on visual and aural detections. Team members noted all species heard or seen during the trip, but most of our effort was directed towards mist-netting and specimen preparation.
Comparative analysis.—For morphometric comparisons we measured bill length, width and depth (all at the rear of the nares), as well as tail length, diagonal tarsus and flat wing-chord length.
We compared mitochondrial ND2 sequences of 41 species from Mt. Besar with sequences from populations in other parts of South-East Asia as available (Appendices 1–2). ND2 was used because it is the most commonly sequenced avian gene and numerous sequences were available in GenBank. When possible, we compared two individuals from each population, and included outgroups for perspective. Total genomic DNA from preserved tissues was extracted using DNEasy® Blood and Tissue Kit (Qiagen) following the manufacturers' protocol. PCR amplifications were performed in 25 μl reactions using Taq DNA Polymerase (New England BioLabs Inc.) and ND2 primers L5215 (Hackett 1996) and HTrpC (STRI). Amplification consisted of 34 cycles at a denaturing temperature of 95°C, an annealing temperature of 54°C and an extension temperature of 72°C. We visualised the PCR products in 1.5% agarose gel stained with SYBR® Safe DNA Gel Stain (Invitrogen). Samples were sequenced at Macrogen USA (Rockville, MD). The ND2 sequences were assembled in Geneious 8.1.9 (Biomatters) and aligned using MUSCLE (Edgar 2004) implemented in Geneious. We generated mitochondrial gene trees using the maximum likelihood (ML) framework in RAxML 8 (Stamatakis 2014), with 1,000 bootstrap replicates, through the CIPRES Science Gateway (Miller et al. 2010).
Expedition.—We recorded 89 species on Mt. Besar (Appendix 1), of which 68 were collected. Of the species recorded, 17 were new for the Meratus Mountains (marked with double asterisks in Appendix 1). Most notable among these was Bornean Spiderhunter Arachnothera everetti, the first record of this species outside Borneo's main mountain chain (Mann 2008, Brickle et al. 2010, Burner et al. 2016, Moyle et al. 2017). Its occurrence in the Meratus Mountains mirrors Bornean Leafbird Chloropsis kinabaluensis, which until it was discovered in the Meratus Mountains by Eaton et al. (2016a) was also thought to be restricted to Borneo's main mountain chain (Mann 2008, Moyle et al. 2011, 2017). Two of our ‘new’ Meratus records, Black-backed Dwarf Kingfisher Ceyx erithaca and Pin-tailed Parrotfinch Erythrura prasina, were anticipated by Davison (1997) from discussions with villagers. Some usually easily detected species recorded by Davison (1997) were not found by us or by Eaton et al. (2016a), most notably Helmeted Hornbill Buceros vigil, Oriental Magpie-Robin Copsychus saularis, White-rumped Shama C. malabaricus and Grey-cheeked Bulbul Alophoixus bres. These species are highly sought by hunters and pet traders (Eaton et al. 2016b), and probably have been extirpated from the area we visited.
Elevational records.—Elevational ranges of most species detected on Mt. Besar were within the limits known from other parts of Borneo. However, some patterns of species occurrence appeared unusual. Several species that are common at the same elevations in similar forest in other parts of Borneo were absent or found in low numbers on Mt. Besar. These include species that are generally common and conspicuous at 500 m, such as Greybellied Bulbul Pycnonotus cyaniventris, Spectacled Bulbul P. erythropthalmos, Yellow-bellied Bulbul Alophoixus phaeocephalus, Fluffy-backed Tit-Babbler Macronus ptilosus, Rufousfronted Babbler Cyanoderma rufifrons, Rufous-crowned Babbler Malacopteron magnum and Short-tailed Babbler Pellorneum malaccense. Other ‘missing’ species were Rufouswinged Philentoma Philentoma pyrhoptera, Green Iora Aegithina viridissima, Blyth's Paradise Flycatcher Terpsiphone affinis and Pale Blue Flycatcher Cyornis unicolor. Another unusual feature of the bird community at 500–1,200 m was that its commonest species (based on capture and incidental observations) was Hill Blue Flycatcher Cyornis banyumas, which is generally uncommon and restricted to a narrow submontane elevational band in other parts of Borneo. Other slope specialists that are principally montane elsewhere in Borneo but occurred in unusually large numbers at 500 m on Mt. Besar were Grey-throated Babbler Stachyris nigriceps and Temminck's Babbler Pellorneum pyrrogenys. Their abundance near Kadayang was remarkable. At higher elevation, c.1,300–1,400 m, some usually scarce species were remarkably common, e.g., Bornean Stubtail Urosphena whiteheadi. In contrast, a species that is generally common and easy to mist-net in the main Bornean mountain range, Chestnut-hooded Laughingthrush Ianthocincla treacheri, was rarely seen and only occasionally heard (see Discussion).
Genetic and morphological comparisons.—Of the 41 Meratus species whose ND2 we compared, 23 were not obviously distinct from populations sampled elsewhere on Borneo, i.e., their phylogeographic trees had little branching structure, and their ND2 divergence values averaged less than 1%: Plaintive Cuckoo Cacomantis merulinus, Rufous Piculet Sasia abnormis, Malaysian Pied Fantail Rhipidura javanica, White-throated Fantail R. albicollis, Spotted Fantail R. perlata, Ashy Drongo Dicrurus leucophaeus, Bornean Treepie Dendrocitta cinerascens, Ashy Bulbul Hemixos flavala, Mountain Leaf Warbler Phylloscopus trivirgatus, Yellow-breasted Warbler Seicercus montis, Rufous-tailed Tailorbird Orthotomus sericeus, Scaly-crowned Babbler Malacopteron cinereum, Brown Fulvetta Alcippe brunneicauda, Pellorneum pyrrogenys, Ianthocincla treacheri, Rufous-chested Flycatcher Ficedula dumetoria, Little Pied Flycatcher F. westermanni, Eyebrowed Jungle Flycatcher Vauriella gularis, Chloropsis kinabaluensis, Chestnut-crested Yuhina Yuhina everetti, Arachnothera everetti, Little Spiderhunter A. longirostra and Purple-naped Spiderhunter A. hypogrammicum. In six cases, we did not possess an adequate number of specimens or sequences to speculate on patterns of population structure: Bornean Barbet Psilopogon eximius, Grey-chinned Minivet Pericrocotus solaris, Blyth's Shrike-Babbler Pteruthius aeralatus, Urosphena whiteheadi, Mountain Tailorbird Phyllergates cucullatus and Indigo Flycatcher Eumyias indigo. However, the remaining 12 species exhibited phylogeographic patterns worthy of further work. We depict these patterns in Fig. 2 and describe them briefly below, but caution that population genetic parameters have not been examined in detail. Also, note that mitochondrial trees sometimes indicate incorrect population structure (see Bornean examples in Lim et al. 2017, Manthey et al. 2017). The rooted trees in Fig. 2, therefore, are intended only as preliminary guides for future investigation.
Bornean Whistler Pachycephala hypoxantha.—Meratus individuals form a clade separate from other Bornean individuals, but divergence between the populations is not large (c.1%).
Ochraceous Bulbul Alophoixus ochraceus.—The Meratus population differs by 4.5% from both A. o. ruficrissus (Sabah specimens) and A. o. fowleri (Sarawak specimens), which are also distinct genetically from each other (Chua et al. 2017). All three populations differ in plumage (Table 1). The Sabah and Sarawak populations lack yellow pigment in their plumage, causing them to appear distinct from other populations of A. ochraceus. Indeed, the Meratus individuals look more like Sumatran than northern Bornean birds, raising the possibility of multiple invasions of Borneo. Confusion is also wrought by paraphyly of A. ochraceus and Grey-cheeked Bulbul A. bres as traditionally constituted; A. ochraceus of Sabah is sister to A. bres of Java (Fuchs et al. 2015, Collar 2017). A. ochraceus and A. bres populations among all of the Greater Sundas and the Malay Peninsula should be compared before their scientific names are revised.
Lemon-bellied White-eye Zosterops chloris.—This is one of the taxa identified as a new species by Eaton et al. (2016a). ND2 comparisons with 45 other species of white-eyes, including the genera Zosterops, Lophozosterops, Speirops and Chlorocharis (mostly from Moyle et al. 2009), reveal the Meratus population to be most closely related to Z. chloris (ND2 divergence 2.69%). This finding makes general biogeographic sense, because Z. chloris occurs from Wallacea west across some islands of the Java Sea to Karimata and Belitung (van Balen 2017). The occupied islands include Kalambau and Solombo Besar near the south-east corner of Borneo (Clements et al. 2017). However, Z. chloris generally occurs at low elevations and in coastal habitats, whereas the Meratus population is montane. As such, the Meratus population appears to fill a potential ecological gap left by the absence of montane white-eye species that occupy mountains in other parts of Borneo: Black-capped White-eye Z. atricapilla, Mountain Black-eye Chlorocharis emiliae and Pygmy White-eye Oculocincta squamifrons. None of these species seems to occur in the Meratus range, leaving the high-elevation forest open for invasion and occupation by Z. chloris. We have not yet performed a morphological analysis for lack of specimens from other populations, but plumage of the Meratus birds appears much like that in photographs of other populations (Fig. 3). Given the large number of Z. chloris populations across the Java Sea and Wallacea, extensive study will be required to determine more precise phylogeographic relationships of the Meratus birds.
Bold-striped Tit-Babbler Mixornis bornensis.—Genetic comparisons of the Meratus population indicate it is closest to individuals from Sarawak and western Sabah (Meligan range) (ND2 divergence 0.55%) than to those in the rest of Sabah and mid-eastern Kalimantan (Berau) (ND2 divergence 3.33%). However, this species displays greater mtDNA than nuclear DNA structure on Borneo (Lim et al. 2017). It also has complex relationships with Pin-striped Tit-Babbler M. gularis (Cros & Rheindt 2017), its closest relative outside of Borneo, decreasing the reliability of conclusions based on a few mtDNA comparisons of a limited sample of specimens.
Plumage colour variation in Ochraceous Bulbul Alophoixus ochraceus and Grey-cheeked Bulbul A. bres from the Meratus Mountains and other parts of Borneo.
White-necked Babbler Stachyris leucotis.—The Meratus, Sarawak and Sabah populations appear to form distinct clades. ND2 distances are: Meratus—Sarawak 1%, Meratus—Sabah 3% and Saba—Sarawak 3%. The substantial divergence separating Sabah's population from the others is consistent with many other species (Lim et al. 2010, 2011, Lim & Sheldon 2011, Lim et al. 2017).
Grey-throated Babbler Stachyris nigriceps.—The Meratus population appears distinct from western Sarawak and north-eastern populations (ND2 divergence 4%; also see Chua et al. 2017).
Grey-headed Babbler Stachyris poliocephala.—Like White-necked Babbler S. leucotis, this lowland species exhibits marked divergence among Meratus, Sarawak and Sabah populations (ND2 divergence 2.3%) (Lim et al. 2011).
Moustached Babbler Malacopteron magnirostre.—The population-genetic structure of Borneo's Malacopteron species has never been assessed. Meratus and Sarawak populations appear to be closely related and quite distinct from birds in eastern Sabah (ND2 divergence 5.81%). The pattern in this species suggests a connection between western and southern Bornean populations and, again, a distinct history for the north-east population (Lim et al. 2010, 2011, Lim & Sheldon 2011, Lim et al. 2017).
Black-capped Babbler Pellorneum capistratum.—The Meratus population is distinct from others in Borneo (average ND2 divergence 7.2%), and Sabah and Sarawak populations are distinct from one another as well (ND2 divergence 11.2%) (Lim et al. 2011).
Eyebrowed Wren-Babbler Napothera epilepidota.—Few tissues of this species are available for genetic assessment. However, the three Bornean populations we compared are remarkably different from one another genetically, in the order of 7%. We did not obtain sufficient specimen material to judge morphological differences.
Hill Blue Flycatcher Cyornis banyumas.—This taxon is one of two possibly new species mentioned by Eaton et al. (2016a). ND2 comparisons indicate that the Meratus population is monophyletic with C. banyumas and divergent by 3.27% from other Bornean populations that have been sampled (i.e., in Sabah and on Mt. Mulu in eastern Sarawak). Morphometric comparisons indicate that Meratus individuals are larger overall in size and have larger bills than those from Borneo's central mountain chain (Table 2). The orange breast of Meratus males also grades into a whiter belly and their black chin is much reduced compared to male C. banyumas from Mt. Mulu and Sabah. As with Z. chloris, thorough phylogeographic comparisons across the entirety of Sundaland will be required to establish the phylogeographic position of the Meratus population.
Yellow-breasted Flowerpecker Prionochilus maculatus.—Our single Meratus specimen is relatively distinct genetically from other Bornean populations and closer to a Peninsular Malaysian bird. Better sampling will be required to tell if this unexpected relationship is well supported and what it implies for Bornean populations.
Avifaunal characteristics .—Compared to Borneo's central mountain chain, the Meratus forest bird community has distinctive features. At 500 m, some common lowland species are absent (e.g., Pycnonotus erythropthalmos, Macronus ptilosus and Pellorneum malaccense), whereas species normally uncommon at this elevation are abundant (e.g., Stachyris nigriceps, Pellorneum pyrrogenys and Cyornis banyumas). At higher elevations, 1,100–1,400 m, some species that are usually common are uncommon (e.g., Ianthocincla treacheri), and others that are typically uncommon are remarkably common (e.g., C. banyumas, Urosphena whiteheadi). Moreover, at this higher elevation, many montane species are completely absent, including numerous Bornean endemics such as Whitehead's Trogon Harpactes whiteheadi, Whitehead's Broadbill Calyptomena whiteheadi and Whitehead's Spiderhunter Arachnothera juliae.
Several interacting factors probably contribute to these patterns. First, the Meratus range is relatively small in area, low in elevation and isolated from other mountain ranges. Simply as a function of size, elevation and location, i.e., by the expectations of island biogeography, the Meratus Mountains should lack a complete montane avifauna. Second, native lowland forest surrounding the Meratus Mountains has been replaced mostly by plantations: oil palm at low elevation, rubber and mixed-use forest at higher elevation (up to 900 m). Thus, the lowland forest avifauna that would normally contribute to the Meratus community at 500 m and above has largely been depleted. Montane species such as Stachyris nigriceps and Cyornis banyumas may be filling this low-elevation species void. There may also be a Massenerhebung effect driving montane species to lower elevation (Bruijnzeel et al. 1993). Third, the local human population appears to depend heavily on small birds for protein. Virtually every villager we encountered in the forest was carrying an air gun (pellet gun) and was shooting small birds and mammals for food. Unlike shotgun ammunition, air gun ammunition is inexpensive and, thus, these guns can be used economically to hunt animals that individually yield small amounts of food. The guns are also quiet when fired, and do not disturb wildlife. We observed hunters climbing fruiting fig trees and shooting numerous barbets at a sitting. This type of hunting pressure might explain the paucity of pigeons, trogons, laughingthrushes, drongos, Alophoixus bulbuls and other mid-sized species in the area we visited. Finally, hornbill-ivory hunters and pet traders appear to have extirpated Buceros vigil, Copsychus saularis, C. malabaricus and Alophoixus bres in the Kadayang area, and may be influencing numbers of other popular cagebird species (e.g., laughingthrushes and leafbirds).
Size comparison between six specimens of Meratus and five specimens of Sabah and Sarawak populations of Hill Blue Flycatcher Cyornis banyumas, including males and females. T-test values show significance of differences between the two groups. All measurements in mm.
Biogeographic issues .—Knowledge of Sundaic rainforest bird phylogeography is confined mainly to studies of populations in Malaysian Borneo (Sabah and Sarawak). Although restricted geographically, these studies provide hypotheses against which to judge the results of our preliminary genetic comparisons of Meratus bird populations. Studies in Malaysian Borneo indicate that lowland populations of birds in Sarawak are often more closely related to those in western Sundaland (Malay Peninsula and Sumatra) than to populations in Sabah (Sheldon et al. 2009, Lim et al. 2010, 2011, Lim & Sheldon 2011, Lim et al. 2017). This pattern, combined with molecular estimates of divergence dates, suggests that Sabah's lowland birds were isolated (one or more times) from populations in western Sundaland, probably in rainforest refugia during global glacial events early in the Pleistocene (e.g., 1–2 million years ago) when most of central Sundaland was subaerial and covered by dry or seasonal forest (Sheldon et al. 2015). Similarly, studies suggest that Bornean populations of montane taxa, such as Chloropsis kinabaluensis, Bornean Forktail Enicurus leschenaulti borneensis and Arachnothera everetti, were isolated in rainforest refuges in north-east Borneo from congeneric populations in western Sundaland early in the Pleistocene (Moyle et al. 2005, 2011, 2017). During the more recent Pleistocene, glaciation events apparently had a different effect on Sundaic habitats; instead of reducing rainforest cover in Sundaland, they increased it, causing greater habitat connectivity among islands and the mainland (Cannon et al. 2009, Raes et al. 2014). Genetic comparisons suggest that, concomitant with this recent increase in rainforest coverage, several western Sundaic bird populations invaded Borneo and occupied the western lowlands, pushing or restricting endemic Bornean populations to the east or into the mountains. These dynamics caused the marked parapatry witnessed today between populations in the lowlands of Sarawak and Sabah (e.g., between White-rumped Shama Copsychus malabaricus suavis in Sarawak and White-crowned Shama C. m. stricklandii in Sabah) and between populations in the lowlands and mountains of Borneo (e.g., White-crowned Forktail Enicurus leschenaulti frontalis in the lowlands and Bornean Forktail E. l. borneensis at higher elevation) (Sheldon et al. 2015, Moyle et al. 2017). Using these biogeographic scenarios as a foundation, we can put our preliminary phylogeographic comparisons of Meratus birds into context and identify the requirements of future investigations.
One of the most interesting discoveries of this study is the occurrence of Chloropsis kinabaluensis and Arachnothera everetti in the Meratus Mountains (also see Eaton et al. 2016a). Previously these species were known only from north-eastern portions of the main Bornean mountain range, e.g., Mulu, Crocker Range and Menyapa Mountains (Mann 2008, Brickle et al. 2010, Burner et al. 2016, Moyle et al. 2017). Their presence in the Meratus Mountains suggests that early Bornean Pleistocene rainforest refuges occurred not just in Sabah (Sheldon 2016) but also in mountains and adjacent lowlands near the coast across much of eastern Borneo. This hypothesis is supported by paleo-habitat data (Morley 2012), paleo-habitat modelling (Cannon et al. 2009, Lim et al. 2011, Manthey et al. 2017) and phylogeographic studies (Quek et al. 2007, Chua et al. 2015).
Some of our preliminary ND2 comparisons suggest that lowland species in the Meratus Mountains are more closely related to western Sarawak or even western Sundaic populations than to north-east Bornean populations (e.g., Malacopteron magnirostre, Stachyris leucotis and Prionochilus maculatus). This pattern could result from the same phenomenon causing similarity between Sarawak and western Sundaic populations: invasion of Borneo from the west during recent global glacial events when the islands and mainland were connected by suitable habitat. Recent invasion of Borneo might also explain the genetic difference between montane birds in the Meratus Mountains and those in northern Borneo (e.g., Alophoixus ochraceus, Pellorneum pyrrogenys, Napothera epilepidota and Cyornis banyumas). Invasion from Sumatra is possible for all four of these species; the last three could have arrived from Java. A faunal connection between Java and south-east Borneo is especially likely, given proximity. Several open- or dry-habitat taxa are shared between Java and Borneo, e.g., Red-breasted Parakeet Psittacula alexandri, Savanna Nightjar Caprimulgus affinis, Copsychus saularis and Scarlet-headed Flowerpecker Dicaeum trochileum (e.g., Mees 1996, Phillipps & Phillipps 2014). More importantly for Meratus populations, preliminary genetic comparisons of some taxa from Borneo's central mountain chain, e.g., swiftlets (Collocalia), spiderhunters (Arachnothera) and forktails (Enicurus), suggest they are closest to Javan populations (Moyle et al. 2005, 2008, 2011). Thus, it is reasonable to expect that some birds in the Meratus Mountains will have at least as much, or greater, affinity with Javan populations than with other Bornean populations.
Overall, understanding the biogeographic significance of genetic and morphological divergence in both lowland and montane populations of the Meratus Mountains will require phylogeographic comparisons with populations of Java, Sumatra and even Peninsular Malaysia, as well as the rest of Borneo. For Zosterops chloris, it will require comparisons with populations across the Java Sea and Wallacea.
We thank the State Ministry of Research and Technology (RISTEK research license numbers 388/SIP/FRP/E5/Dit.KI/XI/2016 and 389/SIP/FRP/E5/Dit.KI/XI/2016) and the Ministry of Forestry for permission to undertake research in south-east Kalimantan. We also thank the Research Center for Biology, Indonesian Institute of Science (RCB-LIPI), and the Museum Zoologicum Bogoriense for their invaluable support; and MZB technician Suparno, Lindsay Burner and the people of Kadayang village for much assistance in the field. John Klicka helped with specimen identification. David Wells, Geoffrey Davison and James Eaton commented on the manuscript, and James Eaton kindly provided photographs of Cyornis banyumas and Zosterops chloris. David Beadle permitted us to use his Internet Bird Collection photo of Zosterops chloris. The project protocol was approved by LSU's Institutional Animal Care and Use Committee (#16-039). Financial support was provided by the US National Science Foundation grant 1241059 and Louisiana State University Museum of Natural Science. All sequences generated from this project have been deposited in GenBank (accession numbers: MG546332–MG546514; MG590020–MG590022).
List of ND2 sequence-samples compared for this study, classified according to Clements et al . (2017).