Inner Melanesia - the Bismarcks & Solomons
Introduction
Melanesia, the ‘Black Islands’ as the Latin translates, is named in reference to its people, who are darker than the Austronesians with whom they share their island world. The name could equally be given for the region’s rich, dark soils, however, for it is one of incredible fertility and biodiversity. New Guinea, which constitutes the vast majority of the region’s land-area, holds up to 6% of the Earth’s species, whilst harbouring the richest flora of any island on the planet (3) (7). Beyond its shores, this natural wealth does not end, from the jungle hills of New Britain to the ancient, Gondwanan forests of remote New Caledonia, far to its south. It is a region of political and cultural complexity as well, divided across multiple nations and home to more unique languages than any other region in the world (5). There are, in short, more habitats, more cultures, more species—more islands—than any one article can cover. Here, we will focus on a particular subset, on what may be dubbed ‘Inner Melanesia’, the Bismarck Archipelago and the Solomon Islands.
Figure 1. Map of the Inner Melanesia region, highlighting the Kilu Cave site on Buka Island.
Politically separate, geographically sundered, that they in any way constitute a historical or ecological unity may seem questionable. Yet there runs between them a deep connection, particularly in the study of their ancient extinctions: they, together, constitute the utmost extremity of humanity’s great Pleistocene trek. When the first wave of colonists drew across the seas of Wallacea, they did not stop on the jungle-plains of Sahul. They continued, over the Bismarck Sea, colonising New Britain and New Ireland by at least 33,000 BC, and the Solomons only shortly after (2). They were the most remote settlement of humanity anywhere in the world. But here, at last, they stopped. The leap across the vast Coral Sea to the Vanuatu and beyond remained a bridge too far, and one that would not be crossed till the coming of a new people, over 30,000 years later. This, then, is the line between Inner and Outer Melanesia—30,000 years of human history, settlement and impact. 30,000 years of extinction.
Figure 2. Beach looking out at Lamasbie Bay in New Ireland, the northernmost part of the Bismarck Archipelago
Terms of use: This image is licensed under a Creative Commons Attribution-ShareAlike 4.0 International. It is attributed to Konoagila. The image is unedited and can be found here.
The New British Isles
In terms of avian diversity, the Bismarck Archipelago is an outstanding pearl of Oceania, second only to New Guinea. New Britain, the largest island in the chain, is unsurprisingly also the richest in species, hosting over 125 species of native land-birds. A total of 14 genera are entirely endemic to the archipelago, 3 of which are passerines, the rest belonging to various other families (14). Among this modern avifauna are 3 genera of cuckoos, 2 families of owls, 4 genera of kingfishers, as well as rails, pigeons, parrots, hawks and falcons, not to mention representatives of characteristically Australasian groups such as megapodes, flowerpeckers, 2 genera of monarch flycatchers, both endemic to Australasia, and even cassowaries (14). In general, the rates of avian endemism are lower than in the Solomons, let alone Outer Melanesia, by virtue of the greater proximity to New Guinea. The archipelago was never connected by land to the continent, not even at the lowest glacial sea-levels, nor were the main islands ever connected to each other. Nevertheless, there exists a substantial continuity between the archipelagic fauna and that of the mainland, with New Britain serving as the primary bridge. Many species there, though shared with New Guinea, are found nowhere else in insular Melanesia (14).
Inner Melanesia is unique in Island Oceania for its unusually developed mammalian fauna. Besides bats, of which numerous species are found in the Bismarcks and Solomons as elsewhere in the Pacific, there exists an assemblage of rodents. These are all of the family Muridae, deriving in origin from New Guinea, where the diversity is vastly greater. The five genera found upon the islands—Hydromys, Melomys, Solomys, Pogonmys and Uromys—are all shared with the Papuan mainland, though all specific species are endemic. Of these, all save Solomys occur in the Bismarck Archipelago, with New Britain constituting the peak of rodent diversity (14). This indigenous assemblage has since been supplemented, first by Rattus exulans, the Polynesian rat, brought by the ancestors of said people from Indonesia, and later by R. rattus, the black rat, brought by Europeans. Interestingly, a third non-native rodent appears to have been transported in prehistoric times by humans: R. praetor, the large New Guinea spiny rat, remains of which are attested not merely from the Bismarcks and Solomons but from all of Melanesia (14). In addition to the placental rodent fauna, there is the point of marsupials. Archaeological evidence indicates that the northern common cuscus, Phalanger orientalis, may have been introduced to the Bismarck Archipelago from its native range in New Guinea more than 20,000 years ago (4). This appears to have occurred through animals carried for food or trade—a practice which has remained in effect in the Melanesian region into modern times. Entering the Holocene, this introduced assemblage appears to have been supplemented by yet further species, including the Brown’s pademelon, Thylogale browni, which appears in fossil records on both New Britain and New Ireland from around 7kya (4). Though the actions of ancient, Pleistocene Australasians cannot in any way be proved based on the recorded activities of historical people, it is interesting to note that P. orientalis, this same species, was deliberately reintroduced by tribesmen to Long Island, north of New Guinea, about 300 years ago, following an eruption that destroyed nearly all life on the isle (4). Though the precise consequences of these ancient introductions cannot be modelled with accuracy, comparative data from other, more recent island ecosystems subjected to similar introductions of alien mammals, including cuscuses and rodents, give a probable indication. Extinctions—first local, in the less remote and sheltered regions, then total—would most probably have been both swift and severe. There are, however, certain indications that these processes may not have unfolded quite in the Bismarcks as elsewhere in historical times, a point which will be discussed in greater detail below.
Despite our generally tenuous fossil record, some clear glimpses into the lost avifauna are afforded us. This fauna included on New Ireland an undescribed flightless rail of medium size, an enormous flightless swamphen of the genus Porphyrio, taller though less stoutly built than the Takahē of New Zealand (P. hochstetteri/mantelli) and two rails which remain extant to this day but have since become locally extinct in New Ireland—the Spotless crake (Porzana tabuensis) and the Australasian swamphen (P. porphyrio/melanotus). Other extinct taxa include an undescribed species or subspecies of cockatoo in the genus Cacatua, substantially larger than any species currently extant there, as well as Corvus crows larger than the Torresian crow (C. orru) which today are New Ireland’s only species.
Figure 3. Late Pleistocene scene on New Ireland. From left to right the animals shown are: Homo sapiens, Cacatua sp., Porphyrio sp., a ribbon-tailed drongo (Dicrurus megarhynchus), a D'Albertis' python (Leiopython albertisii), pied coucals (Centropus ateralbus), an introduced grey cuscus (Phalanger orientalism) and Bismarck hanging parrots (Loriculus tener).
Terms of use: Artwork by Hodari Nundu and Commissioned by The Extinctions
As a land settled for over 40,000 years, though biogeographically linked to the rest of Oceania the historical circumstances of Melanesia’s human settlement are nevertheless greatly different. A time-gap of some 37,000 years is hardly insignificant. For one thing, it is interesting to note that, while the patterns of vanished taxa here are similar to those elsewhere—eagles, large parrots and flightless rails seem especially susceptible to human impact—the process of extirpation appears far more gradual. Less than 2% of avian fossils in New Ireland predate 15,000 BC, meaning over 98% of remains post-date human settlement by some 20,000 years. That we nevertheless see a substantial quantity of extinct taxa (>20%), broadly conforming to the expected trends of human impact, indicates that we are seeing the same pattern as elsewhere, yet greatly protracted. This is further illustrated by the fact that extinct species, though constituting 31% of Pleistocene remains, still take up 21% of Holocene sites as well. Chronological decline is clear—the site with the fewest extinct taxa is also the most recent—yet the process is spread over millennia (14) (2).
A partial explanation here may be the more continental affinities of New Ireland, contra places such as New Caledonia or Hawai’i. Depredation by introduced mammals, in pre-European times largely rodents, has been one of the greatest causes of Avian decline on remote islands. Yet as described, the Bismarcks already possessed an assemblage of endemic mammals, with even remote New Ireland hosting one, Melomys rufescens, the Black-tailed mosaic-tailed rat. The native fauna of the archipelago therefore necessarily had a certain degree of pre-adaptation to mammalian pressures (14). Coming at the question from a different angle, it may be noted that when we view places like New Caledonia, our assessment of losses and survivors is temporally biased. That is, we presume that, by virtue of a species having continued to the present day, it therefore necessarily constitutes one of the animals that ‘passed’ human settlement. Yet humans have only inhabited New Caledonia for a few millennia. Even setting aside the case of Melanesia, experience from Europe, Africa, Australia, and numerous other places shows the casualties of human impact can be very slow to accumulate. That is to say, just because a given New Caledonian bird has survived thus far following human settlement, this does not necessarily mean human pressures would not eventually do it in, even in a hypothetical scenario where Europeans had never arrived. The upshot of this is that the protracted pattern of extinctions observed in the inner Melanesian fossil record may yet play out on more recently settled landmasses such as New Caledonia, and the apparent contrast of gradual versus abrupt faunal turnovers may be an artifact of observation.
Adding to this, only 2% of avian fossils from New Ireland date back to before 15kya. The fact is that this postdates our first records of human settlement in the region by some 20,000 years, with our earliest finds of human habitation in the Bismarcks dated to approximately 35kya (2) (13). The implications are evident: whilst we can paint some rough picture of the nature and pattern of extinctions following this point, we can say nothing about the initial impacts of human arrival. Large terrestrial birds and reptiles are attested from numerous neighbouring landmasses, from the Sahulian mainland to the west, to Vanuatu, Fiji and New Caledonia in southern Melanesia, into New Zealand and even Lord Howe island to the southeast (14) (8). Certain of the groups distributed across these islands are found both east and west of Inner Melanesia, including notably the meiolaniid turtles and mekosuchine crocodiles, as well as large ratite birds. Of the latter, the multiple species of moa are famously recorded from New Zealand, whilst mainland Sahul presents the emus and cassowaries. In fact, the dwarf cassowary (Casuarius bennetti) occurs today on New Britain, whilst remains of the southern cassowary (Casuarius casuarius) are known prehistorically from the island. These, however, have been suggested to be the results of ancient introductions by humans, as with the cuscuses, since cassowaries have very little natural ability to disperse over water (14). That remains, not even of C. bennetti, but of C. casuarius, are known from the small offshore Arawe islands perhaps adds further weight to this notion.
Figure 4. Painting of the dwarf cassowary, Casuarius bennetti, which occurs in the Bismarck Archipelago even to this day. Its relative, the southern cassowary (C. casuarius), is attested from the fossil record.
Terms of use: The copyright of this image has expired and it is in the public domain
Nevertheless, their presence shows if nothing else that New Britain at least, and by extension perhaps New Ireland, is capable of supporting such large animals. That the particular species of cassowaries presently found there are not endemic nor even necessarily truly native, does not mean no equivalent creatures filled their niche before. Nature abhors a vacuum. It seems indeed almost overwhelmingly likely, considering the size of the islands and the biogeographical neighbourhood, that some form of large terrestrial fauna, plausibly including meiolaniids and/or mekosuchines, would have pre-dated human arrival. Groups such as meiolaniids in particular display impressive propensities towards long-distance oversea dispersal, even to remote islands, raising the speculation. Other plausible candidates for such “ghost lineages” might include large monitors such as are found throughout island Australasia, or perhaps even certain elements of the Sahulian marsupial megafauna, though in this we enter pure guesswork. Whether any evidence of such a fauna will ever arise depends entirely on an improvement in our fossil-records.
The Solomons
The Solomons, like the Bismarcks, are a rich archipelago for birdlife. 102 species of landbirds alone are recorded from Bougainville, the largest island in the chain, with the general avian fauna dominated by hawks, rails, doves, parrots and kingfishers (14). From the last millennia of the Pleistocene until the voyages of the Austronesians, the Solomon Islands constituted the very furthest extent of human settlement into Oceania. Not all of the Solomons were settled at once, and the initial Pleistocene migration seems to have been confined to the northern portion of the Archipelago. At this time, low sea levels had united all the isles from Bougainville in the north to the Nggela Islands in the south—a landmass dubbed variously Greater Bougainville or Greater Bukida (16). Even so, the crossing from New Ireland, the nearest island in the Bismarcks, would still have been well over 100 kilometres, making it plausibly the longest oversea voyage in human history up till that point (16). Travelling overland or near the shores of Greater Bougainville, Pleistocene settlers would have been able to reach nearby Guadalcanal and San Cristobal with little apparent difficulty. Nevertheless, no evidence exists of human presence in the southeastern Solomons prior to the Middle Holocene, a few millennia before the arrival of the Austronesian-speaking Lapita people. Here, however, we once more strike against the issue of taphonomy and site visibility, which is no less a problem here than in the Bismarcks. The only known pre-Holocene site from all the Solomons is the Kilu Cave, located in a raised limestone formation on Buka, a small island just above Bougainville, and thus in the extreme north of the archipelago. Here, a Pleistocene layer preserves intermittent occupation from 29-20kya, providing a crucial, yet necessarily limited, picture into the region’s earliest history (16). And, likewise as with the Bismarcks, we find this paucity of remains reflected in our limited understanding of the archipelago’s ancient fauna.
Figure 5. View of the Olu Malau or Three Sisters Islands, Makira-Ulawa Province, Solomon Islands.
Terms of use: This image is licensed under a Public Domain-Merket 1.0 Universal. It is attributed to RH D 22. The image is unedited and can be found here.
As in the Bismarcks, an early introduction attested at the Kilu site is P. orientalis, the common northern cuscus, of which a bone is dated to circa 9kya (17). Where this really constitutes the earliest occurrence of the animal in the Solomons is perhaps debatable, yet it gives, at least, a clear indication that the practice of ferrying useful species long distances oversea did not end with New Britain. Another possible tell of early human “environmental engineering” is not an animal at all, but a plant: Canarium indicum, known commonly as galip nut, an endemically Melanesian species of great horticultural importance in the region. Its utilisation is documented on Buka island as early as 9kya, and though native aboriginally to parts of the region, its use in the more easterly Solomon isles was likely a result of archaic human introduction (17). The rainforests of the Solomons would have presented poor yields of edible plants to the first settlers, raising questions about whether forms of primitive horticulturalism were practiced even by the very earliest colonists. Indeed, there is evidence from the Kilu Cave site to indicate that the deliberate management of stands of Colocasia, the ancestor of the domestic taro, may have begun already in the Pleistocene (16).
Interestingly, the site at Kilu also preserves the evidence of two extinct, endemic species of rats: Solomys spriggsarum and Melomys spechti. Both seem to vanish from the records some time around 5.5kya, indicating a plausible window of extinction. Besides these species of rodents, the two layers of the site also preserve the remains of bats of varying sizes, as well as skinks, snakes, and 18 species of landbirds (17) (14). Of these latter, 11 have since ceased occurring on Buka, whilst 7 constitute totally extinct taxa, all as-yet-undescribed, with 3 species pertaining to entirely extinct genera. All but one of the landbird bones from the Pleistocene layer represent extinct taxa, whilst 71% of the Holocene bones likewise pertain to species now either extinct or extirpated from the island (14). Among the 7 extinct birds are a medium-sized night heron of the genus Nycticorax and a large Megapodius scrubfowl, comparable in size to the New Caledonian pile-builder megapode, M. molistructor. There are also three extinct rails, the first two ascribed with differing degrees of certainty to Gallirallus and Pareudiastes/Gallinula, whilst the last is a massive swamphen of the genus Porphyrio. The last two taxa are extinct doves of uncertain genus-affiliations (14).
Figure 6. Image of a Solomon Islands skink, Corucia zebrata, an arboreal, prehensile-tailed lizard endemic to the archipelago which is the largest extant species of skink.
Terms of use: This image is licensed under a Attribution-ShareAlike 3.0 Unported. It is attributed to JSutton93. The image is unedited.
A final area worthy of note is Woodlark or Muyua Island, located in the Solomon Sea, which pertains to neither of the great archipelagos, nor to the Papuan mainland, yet has connections with all three. Unlike either the Bismarcks or the Solomon archipelago, Woodlark Island has an endemic species of cuscus, P. lullulae. Its closest relative is P. orientalis, the northern common cuscus, but its position as a genuinely endemic taxon on such a remote landmass, more than 200 km from the Papuan mainland, is interesting, and demonstrates a facility towards natural oversea dispersal in the phalangers (6) (10). Even more intriguing, however, and far more enigmatic, is another possible endemic of Woodlark. In 1905, fragmentary remains of an apparent gavialid crocodilian—the gharial lineage—were uncovered and reported by Charles De Vis, who dubbed it ‘Gavialis papuensis’. For the better part of the 20th century, De Vis’ discovery languished uncited and uninvestigated in the literature, until in 1982 when a follow-up study finally attempted to throw some light on the specimen (9) Consisting only of a mandible, vertebrae and two dermal scutes, the animal could not be assigned with confidence to any one genus yet was found to resemble most closely not members of the gharial-genus Gavialis, but instead the extinct Ikanogavialis. This was, however, a tentative association at best, made on the basis of a single character of the mandible, and would not much reduce the mystery surrounding this animal. Only two genera of gavialids are known from the prehistory of Australasia, the Miocene-era Harpacochampsa and the Plio-Pleistocene Gunggamarandu, a tomistomine related to the extant false gharial of Southeast Asia (11). Though the presence of this group in Pleistocene Melanesia is therefore not altogether improbable, the possible association with Ikanogavialis does little to clarify matters, as this is a Miocene-era genus known only from South America, being removed therefore from the Woodlark animal by both millions of years and the width of the Pacific. A more recent relevant expert, observing the specimen in person, raised again the possibility of its pertaining to Gavialis (11), but in lieu of an actual study on the matter, or of further fossil material, all that can be said with relative certainty is that the Woodlark animal is, in fact, a gavialid, whose more certain affinities cannot yet be determined.
All the same, the combined existence of the enigmatic, probably Pleistocene Woodlark gavialid and the Australian false gharial-relative Gunggamarandu from the Plio-Pleistocene boundary raises the tantalising possibility that this clade may have a deeper and more widespread history in Australasia and Oceania than we have hitherto known. The fossil records of the Melanesian archipelagos are, as detailed above, patchy and often downright absent, and it cannot therefore be precluded that further relatives of ‘G.’ papuensis occurred throughout these islands, too.
Conclusion
One can scarcely survey our records from the prehistory of Inner Melanesia without concluding that the image painted is of a decidedly more impressionistic slant than often elsewhere. Poverty, remoteness, post-glacial flooding and the twinned obstacles of political instability and local inaccessibility have long been severe limiting factors on our knowledge of Melanesia’s past. This is unlikely to change in the immediate future. All the same, those glimpses we do have, insufficient as they may be, hint of a narrative both fascinating and familiar. It has always been the biogeographical fate of Melanesia to exist as something of an intermediary between the continental ecosystems of New Guinea and Sahul, and the truly insular world of the South Pacific. In their mammalian fauna and numerous other ecological links with Papua, they carry a “mainland” strain absent from the more far-flung isles, yet in their evidently sizeable, if now scarcely remembered assemblage of flightless birds, peculiar endemics and unusual survivals, they remain distinctly “insular”. Inner Melanesia is not today as remote as it once was, and those same processes of globalisation and modernisation which are slowly transforming the Papuan mainland and the further Pacific are now gradually, but definitely, making their inroads. Deforestation is accelerating, motivated particularly in the Solomons by a growing trade with China (1), and overfishing, as in so many corners of the world, is a pressing concern (12). It may be hoped that these processes, for all their downsides, will also in time allow an opening of the island world, such that further studies and more detailed surveys can shine a much-needed light upon the peculiar ecosystems of its past and present. Until then, what remains is a vista both clouded and dimly viewed, yet perhaps all the more tantalising for that.
References
1. Al Jazeera (2018) Chinese demand fuels Solomon Islands mass deforestation, Al Jazeera. Available at: https://www.aljazeera.com/news/2018/10/18/chinese-demand-fuels-solomon-islands-mass-deforestation (Accessed: 10 February 2024).
2. Allen, J. (1996). The Pre-Austronesian Settlement of Island Melanesia: Implications for Lapita Archaeology
3. Barrows et al. (2009). Making the most of Papua New Guinea’s biodiversity: Establishment of an integrated set of programs that link botanical survey with pharmacological assessment in “The Land of the Unexpected”
4. Heinsohn, T. E. (2010). Marsupials as introduced species: Long-term anthropogenic expansion of the marsupial frontier and its implications for zoogeographic interpretation. In Haberle, S. et al (eds), Altered ecologies: Fire, climate and human influence on terrestrial landscapes, pp. 133-167. Canberra, ACT: ANU E Press.
5. Kik et al. (2021). Language and ethnobiological skills decline precipitously in Papua New Guinea, the world’s most linguistically diverse nation
6. Leary, T., Singadan, R., Menzies, J., Helgen, K., Wright, D., Allison, A. & Hamilton, S. 2016. Phalanger lullulae. The IUCN Red List of Threatened Species 2016: e.T16846A21951419. https://dx.doi.org/10.2305/IUCN.UK.2016-2.RLTS.T16846A21951419.en. Accessed on 10 February 2024.
7. Leret-Cámara et al. (2020). New Guinea has the world’s richest island flora
8. Lichtig, A. & Lucas, S. G. (2018). THE ECOLOGY OF MEIOLANIA PLATYCEPS, A PLEISTOCENE TURTLE FROM AUSTRALIA
9. Molnar, R. E. 1982. A longirostrine crocodilian from Murua (Woodlark), Solomon Sea. Memoirs of the Queensland Museum 20, 675-685.
10. Raterman, D., Meredith, R. W., Ruedas, L. A., Springer, M. (2006). Phylogenetic relationships of the cuscuses and brushtail possums (Marsupialia:Phalangeridae) using the nuclear gene BRCA1. Australian Journal of Zoology 54(5):353-361. DOI: 10.1071/ZO05067
11. Rio, J. P. & Mannion, P. D. (2021). Phylogenetic analysis of a new morphological dataset elucidates the evolutionary history of Crocodylia and resolves the long-standing gharial problem. PeerJ. 2021; 9: e12094. doi: 10.7717/peerj.12094
12. Solomon Islands. WWF. Available at: https://www.wwfpacific.org/about/solomon_islands_/ (Accessed: 10 February 2024).
13. Steadman et al. (1999). Prehistoric birds from New Ireland, Papua New Guinea: Extinctions on a large Melanesian island
14. Steadman, D. (2006). Extinction and Biogeography of Tropical Pacific Birds
15. Swadling et al. (2008). Prehistoric stone artefacts from Enga and the implication of links between the highlands, lowlands and islands for early agriculture in Papua New Guinea
16. Walter, R. and P. Sheppard. (2009). A review of Solomon Island archaeology. In P. Sheppard, T. Thomasand G. Summerhayes (eds), Lapita: Ancestors and descendants, pp. 35–72. New Zealand Archaeological Association Monograph Series 28. New Zealand Archaeological Association, Auckland.
17. Wickler, S. (1990). Prehistoric Melanesian Exchange and Interaction: Recent Evidence from the Northern Solomon Islands. Asian Perspectives, Vol. 29, no. 2.