All is not lost: Herpetofaunal “extinctions” in the Fiji Islands

By Adam G. Clause*, Nunia Thomas-Moko, Sialisi Rasalato and Robert N. Fisher

Abstract Invasive mammals are implicated in the decline or extinction of numerous insular vertebrate species worldwide, yet rediscoveries of supposedly extinct vertebrates occur regularly. In particular, recent records of secretive amphibian and taxa in the Fiji Islands show that earlier claimed extirpations of Fijian wildlife were erroneous. We add to this growing body of evidence by documenting the Fiji barred treeskink trossula (: Scincidae) from Vanua Levu island, Fiji where it was widely considered extirpated. Regional literature, coupled with this new record, emphasizes the conservation importance of remote forest blocks in Fiji as refugia against non-native predatory mammals. Moreover, a clear need exists for additional survey work in Fiji to document the contemporary distribution of endemic and endangered herpetofaunal species across the archipelago.

*Corresponding Author E-mail: [email protected]

Pacific Science, vol. 72, no. 3 January 26, 2018 (Early view) Introduction

This species [Fiji Barred Treeskink, Emoia trossula] now has a patchy distribution in the

Fiji Islands and is almost certainly extinct on the main islands of Viti Levu and Vanua Levu. The two specimens in the Hamburg Museum (ZMH R01976-77) that are recorded from Viti Levu were collected early in the nineteenth century. It is suggested that the introduction of the mongoose in 1887 may have led to this extinction. This theory would appear to be indirectly supported by the fact that E. nigra, a primarily terrestrial of about the same size as E. trossula, is also absent from Viti Levu but common on some other islands on which E. trossula occurs.

------Brown and Gibbons 1986

Biodiversity loss on oceanic islands worldwide is often attributed, at least in part, to introduced mammalian predators (Case and Bolger 1991, Blackburn et al. 2004). Mongoose

Herpestes spp., rats Rattus spp., and feral cats Felis catus are especially damaging and pernicious invaders (Towns and Daugherty 1994, Towns et al. 2006, Hays and Conant, 2007). However, vertebrate rediscoveries occur at an average global rate of roughly three species annually, and are increasing in frequency (Scheffers et al. 2011). These rediscoveries have led many authors to call for more restraint in declaring a species extinct or extirpated, particularly for secretive taxa that occupy remote or poorly surveyed habitats (Ladle et al. 2011, Scheffers et al. 2011, Caviedes-

Solis et al. 2015). Detection problems associated with herpetofauna (Durso et al. 2011) make their populations especially prone to being overlooked, sometimes for many decades (e.g.,

Watling et al. 2010; Caut et al. 2013; Solano-Zavaleta et al. 2017).

In the Fiji Islands of the South Pacific, non-native mammals have been blamed as central players in the decline or extirpation of many native land-dwelling herpetofauna (Gibbons 1984,

Zug 1991, Morrison et al. 2004, Fisher et al. 2012, Morley and Winder 2015). Nonetheless, on

2 Fiji’s largest island, Viti Levu, isolated forest tracts have largely resisted penetration by mongoose and rats (Olson et al. 2006), despite the presence of these mammals on the island for over 100 years (Pernetta and Watling 1978, Simberloff et al. 2000). The mountainous interiors of

Fiji’s second-largest (Vanua Levu) and third-largest (Taveuni) islands also retain large expanses of relatively undisturbed forest (Olson et al. 2009).

Here, we announce the existence of the Fiji Barred Treeskink Emoia trossula in western

Vanua Levu, review the taxonomic and distribution literature for this species, and review the literature on recent herpetofaunal rediscoveries in Fiji. Our purpose is to summarize the and distribution of E. trossula, evaluate the validity of claimed herpetofaunal extirpations/extinctions in Fiji, and comment on the conservation value of relatively undisturbed, poorly surveyed forests throughout the archipelago.

Materials and Methods

In December 2015, we led an expedition to the poorly-studied Nabou forest in

Nasarawaqa, Letuku District, Bua Province, on Vanua Levu island. This forest is located approximately 10 km inland from the coastline of the island, along the foothills of its rugged mountainous interior. It is contiguous with an intact, roadless expanse of interior forest some

15,000 ha in size. Our work in Nabou forest focused on rope-aided herpetofaunal surveys of canopy epiphytes. In total, our team spent four days surveying the forest, during which we climbed into the canopy of 12 mature trees.

In December 2016, we performed a literature review to gather all published information on E. trossula. We then queried the online VertNet specimen database to assemble all vouchered records for E. trossula from Fiji, excluding Rotuma. Currently, the binomen E. trossula is restricted to populations from Fiji excluding Rotuma (Zug 2013), a concept we follow here (but

3 see Gill [1995], Hamilton et al. [2010], and Allison et al. [2013]). We subsequently surveyed the literature for all herpetofaunal rediscoveries in Fiji since the year 2000. <>

Results

On 17 December 2015, at 19:00 h shortly after the cessation of a rainstorm, we captured an adult male E. trossula (Figure 1) in Nabou forest, Vanua Levu island. The lizard was taken among intact old-growth lowland rainforest (16.70177° S, 178.80713° E; WGS 84) at 190 m elevation. The specimen (California Academy of Sciences #CAS 261882) measured 83 mm snout-to-vent length (SVL) and 110 mm tail length, with 51 mm of the tail being a post-autotomy regeneration. We found the lizard 16 m high in an epiphytic Drynaria rigidula fern in the forest canopy. Elsewhere in the forest we observed, but were unable to capture, a second very large skink perched high on a tree trunk that might have been E. trossula.

Zug (1991) and Morrison (2003) provide keys to and descriptions of the non-snake squamate fauna of Fiji. These authors indicate that three traits diagnose E. trossula from all described Fijian skink species: (1) large adult body size, with snout-to-vent length >65 mm; (2) dorsal habitus olive brown to gray, usually with interrupted dark transverse bands and pale spots;

(3) 42–56 lamellae under fourth toe. The Nabou forest E. trossula specimen possesses all three diagnostic traits (Figure 1).

Our literature review indicates that the taxonomic status of E. trossula has recently become unstable. Originally lumped within Emoia samoensis (A. Duméril, 1851), the species E. trossula was first recognized as a distinct taxon by Brown and Gibbons (1986). Subsequently, population clusters formerly attributed to E. trossula were elevated to full species status on

Rarotonga, Cook Islands (Zug et al. 2011), Tonga (Zug et al. 2012), and Rotuma (Zug 2012).

Specimens from Futuna tentatively attributed to E. trossula (Gill 1995) were later re-allocated to

4 E. samoensis by Zug (2012) with reservations (but see Allison et al. 2013). Furthermore,

Hamilton (2010) indicated the possible presence of additional cryptic species within Fijian populations of E. trossula.

Including our Vanua Levu record, E. trossula has been vouchered from 21 islands in Fiji

(Figure 2) (Brown 1991, Zug 1991, VertNet 2016). However, the most recent E. trossula voucher is over 30 yr old for all islands except Ovalau, Taveuni and now Vanua Levu. This raises questions as to the contemporary persistence of E. trossula across the majority of its range.

Additional survey work is needed to confirm the species’ continued existence on islands that lack a recent voucher. The IUCN Red List of Threatened Species classifies E. trossula as Endangered

(Allison et al. 2013), further emphasizing the importance of our Vanua Levu discovery. <>

This new Vanua Levu E. trossula record is one of several recent finds of conservation relevance in Fiji, as revealed in our literature review. The Fiji ground frog Cornufer vitianus

(formerly Platymantis vitianus; see Brown et al. [2015] for a review) was rediscovered on Vanua

Levu in the Waisali Reserve in 2003 (Morrison, 2004). Subsequently, C. vitianus was documented persisting in five additional populations across Vanua Levu (Osborne et al. 2013). In

2009, C. vitianus was also rediscovered in the isolated Nakauvadra Range on the island of Viti

Levu (Thomas 2009). Prior to these discoveries, C. vitianus was considered extirpated from one or both islands (Morrison 2003; Zug et al. 2004). Furthermore, Fijian iguanas in the genus

Brachylophus were recently documented on both Viti Levu and Vanua Levu (Fisher et al. 2012,

Fisher unpub. data), representing the first reports from these islands in over 30 years (Gibbons

1984). Although no published source ever asserted the extirpation of these iguanas on Viti Levu, they were considered to have declined markedly due to introduced mongoose, rats, feral cats, and

5 habitat destruction (Gibbons 1984; Fisher et al. 2012). On Vanua Levu, however, iguanas had recently been assumed extirpated (Fisher et al. 2012).

Discussion

The unstable history of E. trossula, as detailed previously, leads us to consider the taxonomy of this species to be poorly resolved. However, because we vouchered our record with a whole-body specimen, genetic tissue sample, and photos in life, we are confident that future workers will be able to accurately reassign our record taxonomically if needed. We emphasize the importance of this three-technique vouchering strategy, and of reporting diagnostic traits based on original literature, when documenting novel observations of herpetofauna in Fiji and the South Pacific.

Most authors characterize E. trossula as semi-arboreal (Brown and Gibbons 1986, Zug

1991, Morrison 2003), or arboreal (Zug 2013). Pregill and Steadman (2014) observed an E. trossula 6 m high in a tree. However, our E. trossula record is the first from the upper forest canopy, and is the highest ever documented (16 m). We hypothesize that this penchant for scansorial behavior has contributed to the species’ persistence in the face of invasion by predatory mammals. Our field teams observed both rats and mongoose in and around Nabou forest.

Brown and Gibbons (1986) were the first to claim the status of E. trossula as “almost certainly extinct” from Viti Levu and Vanua Levu islands, Fiji. This hypothesis was repeated by many subsequent authors (Case and Bolger 1991, Zug 1991, Morrison 2003, Morrison 2005,

Allison et al. 2013, Pregill and Steadman 2014). Moreover, both Zug et al. (2011) and Zug

(2013) indicated the existence of E. trossula on these two islands. However, all of these published sources either provide no evidence to support their assertions, or offer supporting

6 citations tracing back to Brown and Gibbons (1986). Crucially, neither Brown and Gibbons

(1986) nor any subsequent author actually substantiates the historical presence of E. trossula on

Vanua Levu. No specimen or photo voucher exists from that island. Thus, the record we report here is the first for E. trossula from Vanua Levu. It represents a discovery (not a re-discovery), and reveals a history of unsupported assertions of presence/extirpation that trace back to Brown and Gibbons (1986).

The IUCN Red List Categories and Criteria Version 3.1 specifies that a taxon can be presumed extinct only when “exhaustive surveys in known and/or expected habitat, at appropriate times (diurnal, seasonal annual)…have failed to record an individual.” The criteria further specify that these surveys “should be over a time frame appropriate to the taxon’s life cycle and life form.” We support these criteria, and suggest that their adoption could help limit future incorrect claims of extinction or extirpation. In the case of the Fiji Islands, the IUCN criteria might have prevented the widespread belief in C. vitianus, Brachylophus iguana, and E. trossula extirpations that contemporary data have shown were erroneous.

Our survey work and literature review indicates that the two largest islands in Fiji retain significant biodiversity value, despite their lengthy history of human habitation and long- established populations of introduced mammalian predators. A fuller picture of the richness and local distribution of this biodiversity will necessitate much additional survey effort. We remain in an “age of discovery” in the Fiji Islands. Not only with respect to the herpetofaunal assemblage of large forest blocks on the main islands, but also across the 330+ additional islands of the group (Morrison 2005; Fisher et al., 2017).

7 Conclusions

We do not wish to dismiss the negative effects of introduced mammalian predators on the native Fijian herpetofauna. Nonetheless, it is becoming apparent that suspicions of extirpations due to invasive mammals have been exaggerated in Fiji. This sentiment was expressed as early as Pernetta and Watling (1978), and Didham et al. (2005) warned against invoking any single cause to explain species extinction. Ultimately, claims of extirpation might warrant re- examination using the IUCN Red List criteria for assessing extinction. In the past 15 years, both the Fiji ground frog and Brachylophus iguanas have been rediscovered on Vanua Levu and Viti

Levu islands (Morrison et al. 2004, Thomas 2009, Osborne et al. 2013, Fisher et al. 2012, Fisher unpub. data). These finds reversed earlier fears that these species were extirpated or nearly extirpated from those islands, although they remain rare and of major conservation concern.

Taken together with the E. trossula record from Vanua Levu described herein, these discoveries support the concept promoted by Olson et al. (2006) that large, intact tracts of forest on Fiji’s main islands are biodiversity refugia, sheltering native taxa against heavy predation pressure from rats and mongoose. Olson et al. (2006, 2009) also indicate that the conservation importance of remote forest blocks on Fiji’s main islands should not be discounted, and call for renewed preservation efforts in these areas; our results are consistent with this suggestion. Much additional survey work, involving both terrestrial and arboreal sampling, is also needed in these forests and across the Fijian archipelago (Morrison 2005). Such surveys hold great promise for revealing populations of secretive Fijian wildlife in vital need of protection.

8 Acknowledgments

We thank Aporosa Rokovala and Ilaisa Mataqeledra for invaluable logistical support.

Additional permission was facilitated by Emori Dakai, Elizabeth Erasito, Sanjana Lal, Sitiveni

Lalibuli, Tuiviti Nakalavo, Jone Niukula, Asaeli Ravasakula, Eliki Senivasa, Jese Volau, and

Isireli Vunibaka. Funding was provided by the International Iguana Foundation and a University of Georgia Presidential Fellowship. Specimen collection/export/import was approved by permit

#RA24/13 issued by the Fiji Ministry of Education, National Heritage, Culture and Arts; UGA

IACUC AUP #A2012 10-004-Y1-A0; CITES permits #16FJ/NV200095 and #15US66618B/9; and USFWS cleared eDec #2016963116. Special thanks to Lauren Scheinberg for accessioning our vouchers. We are grateful to George R. Zug and two anonymous reviewers for providing comments that improved an earlier version of this manuscript.

9 Literature Cited

Allison, A., A. M. Hamilton, A.M., and O. Tallowin. 2013. Emoia trossula. The IUCN Red List

of Threatened Species 2013: e.T196624A2467661.

http://dx.doi.org/10.2305/IUCN.UK.2013-1.RLTS.T196624A2467661.en

Blackburn, T. M., P. Cassey, R. P. Duncan, K. L. Evans, and K. J. Gaston. 2004. Avian extinction

and mammalian introductions on oceanic islands. Science 305:1955–1958.

Brown, W. C. 1991. Lizards of the Genus Emoia (Scincidae) with observations on their evolution

and biogeography. Mem. Cal. Acad. Sci. 15:1–94.

Brown, W. C., and J. R. H. Gibbons. 1986. Species of the Emoia samoensis group of lizards

(Scincidae) in the Fiji Islands, with descriptions of two new species. Proc. Cal. Acad.

Sci., Fourth Ser. 44:41–53.

Case, T. J., and D. T. Bolger. 1991. The role of introduced species in shaping the distribution and

abundance of island . Evol. Ecol. 5:272–290.

Caut, S., M. Holden, M. J. Jowers, R. Boistel, and I. Ineich. 2013. Is Bocourt’s Terrific Skink

really so terrific? Trophic myth and reality. PLoS ONE 8:e78638.

Caviedes-Solis, I. W., L. F. Vázquez-Vega, I. Solano-Zavaleta, E. Pérez-Ramos, S. M. Rovito, T.

J. Devitt, P. Heimes, O. A. Flores-Villela, J. A. Campbell, and A. Nieto Montes de Oca.

2015. Everything is not lost: Recent records, rediscoveries, and range extensions of

Mexican hylid frogs. Mesoamer. Herpet. 2:230–241.

Didham, R. K., R. M. Ewers, and N. J. Gemmell. 2005. Comment on “Avian extinction and

mammalian introductions on oceanic islands.” Science 307:1412.

Durso, A. M., J. D. Willson, and C. T. Winne. 2011. Needles in haystacks: Estimating detection

probability and occupancy of rare and cryptic snakes. Biol. Cons. 144:1508–1515.

10 Fisher, R. N., T. Grant, and P. S. Harlow. 2012. Brachylophus bulabula. The IUCN Red List of

Threatened Species 2012: e.T174471A1414101.

http://dx.doi.org/10.2305/IUCN.UK.2012.RLTS.T174471A1414101.en

Fisher, R. N., J. Niukula, D. Watling, and P. S. Harlow. 2017. A new species of iguana

Brachylophus Cuvier 1829 (Sauria: Iguania: Iguanidae) from Gau Island, Fiji Islands.

Zootaxa 4273:407–422

Gibbons, J. 1984. Iguanas of the South Pacific. Oryx 18:82–91.

Gill, B. J. 1995. Notes on the land reptiles of Wallis and Futuna, South-west Pacific. Rec. Auk.

Inst. Mus. 32:55–61.

Hamilton, A. M., G. R. Zug, and C. C. Austin. 2010. Biogeographic anomaly or human

introduction: A cryptogenic population of tree skink (Reptilia: Squamata) from the Cook

Islands, Oceania. Biol. J. Linn. Soc. 100:318–328.

Hays, W. S. T., and S. Conant. 2007. Biology and impacts of Pacific Island invasive species. 1. A

worldwide review of effects of the small Indian mongoose, Herpestes javanicus

(Carnivora: Herpestidae). Pac. Sci. 61:3–16.

Ladle, R. J., P. Jepson, A. C. M. Malhado, S. Jennings, and M. Barua. 2011. The causes and

biogeographical significance of species’ rediscovery. Front. Biogeog. 3:111–118.

Morley, C. G., and L. Winder. 2015. Vulnerability of to predation by introduced

mongoose in the Fiji Islands. Pac. Sci. 69:313–317.

Morrison, C. 2003. A field guide to the herpetofauna of Fiji. Institute of Applied Sciences,

University of the South Pacific, Suva, Fiji.

———. 2005. Distribution and diversity of Fiji’s terrestrial herpetofauna: Implications for forest

conservation. Pac. Sci. 59:481–489.

11 Morrison, C., A. Naikatini, N. Thomas, I. Rounds, B. Thaman, and J. Niukula. 2004.

Rediscovery of an endangered frog Platymantis vitianus, on mainland Fiji: Implications

for conservation and management. Pac. Cons. Biol. 10:237–240.

Olson, D., L. Farley, W. Naisilisili, A. Raikabula, O. Prasad, J. Atherton, and C. Morley. 2006.

Remote forest refugia for Fijian wildlife. Cons. Biol. 20:568–572.

Olson, D., L. Farley, A. Patrick, D. Watling, M. Tuiwawa, V. Masibalavu, L. Lenoa, A. Bogiva, I.

Qauqau, J. Atherton, A. Caginitoba, M. Tokota'a, S. Prasad, W. Naisilisili, A. Raikabula,

K. Mailautoka, C. Morley, and T. Allnutt. 2009. Priority forests for conservation in Fiji:

Landscapes, hotspots and ecological processes. Oryx 44:57–70.

Osborne, T., A. Naikatini, C. Morrison, and N. Thomas. 2013. The distribution of the Fiji frogs,

Platymantis spp.: New records and ramifications. Pac. Cons. Biol. 19:175–183.

Pernetta, J. C., and D. Watling. 1978. The introduced and native terrestrial vertebrates of Fiji.

Pac. Sci. 32: 223–244.

Pregill, G. K., and D. W. Steadman. 2014. The prehistory of terrestrial reptiles and birds in the

Central Lau Group, Fiji. Bull. Fl. Mus. Nat. Hist. 53:1–25.

Scheffers, B. R., D. L. Yong, J. B. C. Harris, X. Giam, and N. S. Sodhi. 2011. The world’s

rediscovered species: Back from the brink? PloS One 6:1–8.

Simberloff, D., T. Dayan, C. Jones, and G. Ogura. 2000. Character displacement and release in

the small Indian mongoose, Herpestes javanicus. Ecology 81:2086–2099.

Solano-Zavaleta, I., N. M. Cerón de la Luz, and A. G. Clause. 2017. Solving a 50-year mystery:

Rediscovery of Mesaspis antauges (Squamata: Anguidae). Zootaxa 4303:559–572.

Thomas, N. 2009. Herpetofauna of the Nakauvadra Range, Ra Province, Fiji. Pages 43–51 in C.

Morrison and S. Nawadra, eds. A rapid biodiversity assessment of the Nakauvadra

12 Highlands, Ra Province, Fiji. RAP Bulletin of Biological Assessment 57. Conservation

International, Arlington, VA, USA.

Towns, D. R., and C. H. Daugherty. 1994. Patterns of range contractions and extinctions in the

New Zealand herpetofauna following human colonisation. New Zeal. J. Zool. 21:325–

339.

Towns, D. R., I. A. E. Atkinson, and C. H. Daugherty. 2006. Have the harmful effects of

introduced rats on islands been exaggerated? Biol. Inv. 8:863–891.

Watling, D., A. Wynn, and G. R. Zug. 2010. Rediscovery of the Taveuni blind snake. Oryx

44:165–166.

Zug, G. R. 1991. The lizards of Fiji: Natural history and systematics. Bishop Mus. Bull. Zool.

2:1–136.

———. 2012. A new species of treeskink (Squamata: Scincidae: Emoia samoensis species

group) from Rotuma, South-central Pacific. Proc. Biol. Soc. Wash. 125:74–84.

———. 2013. Reptiles and amphibians of the Pacific Islands: A comprehensive guide.

University of California Press, Berkeley and Los Angeles, California.

Zug, G., D. Watling, and C. Morrison. 2004. Cornufer vitianus. (errata version published in

2015) The IUCN Red List of Threatened Species 2004: e.T17518A84680701.

http://dx.doi.org/10.2305/IUCN.UK.2004.RLTS.T17518A7118149.en

Zug, G. R., A. M. Hamilton, and C. C. Austin. 2011. A new Emoia samoensis group lizard

(Squamata: Scincidae) from the Cook Islands, South-central Pacific. Zootaxa 2765:47–

57.

Zug, G. R., I. Ineich, G. Pregill, and A. M. Hamilton. 2012. Lizards of Tonga with description of

a new Tongan Treeskink (Squamata: Scincidae: Emoia samoensis group). Pac. Sci.

66:225–237.

13 Figure 1. Adult male Emoia trossula (CAS 261882) from Nabou forest on Vanua Levu island,

Fiji. Both photographs illustrate the same individual. Photos by A.G. Clause.

14 Figure 2. Island-level distribution of Emoia trossula. Triangle indicates new Vanua Levu record, and circles indicate existing records.

15