Canarium (Burseraceae) is a genus of c. 100 species, centred on Malaysia, comprising mainly large primary or secondary forest trees growing at low altitudes. There are monographs by Leenhouts (1959a; 1959b) for Malesia (the ever-wet tropics of Southeast Asia and Papua New Guinea) and Leenhouts (1955) for the Pacific. Several Canarium species are a valuable source of edible nuts, oil and resin, sometimes timber, but most have soft wood, and are popular wayside trees (TABLE 1). Most are tall, several species have buttresses, the leaves are pennate and the fruits are generally plum-shaped, with stones and a fleshy pericarp, and blue-black when ripe (Leenhouts 1959b). There are three sub-genera: Canarium, Pimella and Canariellum, with the last in east Queensland, New Caledonia and adjacent islands only. There are a plethora of local names, but the only widespread ones are kanari and kendondong in Malesia.
Canarium is cultivated in Melanesia as far as Vanuatu (Gosden 1992) with one species (C. harveyi) even found in Samoa, western Polynesia, [TABULAR DATA FOR TABLE 1 OMITTED] where it may be a modern introduction (Whistler 1991). C. baileyanum nuts are used by Australian Aborigines (Yen 1995: 838). Some species seem almost confined to rainforest, but C. hirsutum Willd. ssp. hirsutum occurs infrequently in secondary forest in Malesia and C. sumatranum Boerl. & Koord. is commoner in secondary forest below 500 m in Sumatra. Two of the 15 Sumatran species are found above 1400 m altitude: C. littorale Bl. (rarely up to 2000 m) and C. hirsutum ssp. hirsutum, occasionally found up to 1800 m. Two species have been reported from lowland freshwater swamps in Singapore (Corner 1978).
Fossil archaeological and pollen finds are listed on TABLE 1 and site locations on FIGURE 1. The Seraba, Sepik-Ramu Valley, fruits are large, and they have been labelled by Gosden (1993) as 'domesticated'. Yen (1991: 84) is unsure but suggests (Yen 1995: 843) that the continuity of sequences indicates early domestication. This, and Sri Lankan evidence, indicates that hunter-gatherers were using wild resources of the lowland rainforest as far back as the Late Pleistocene and trees may have been selectively conserved before they were planted. The commonest species present archaeologically in the Lapira homeland area is C. indicum (Yen 1991: 84) but there are four domesticated species in the New Guinea region, and another centred on the western Solomon Islands (Yen 1995: 836). A Canarium-type pollen, closely resembling modern reference material of C. littorale, has been discovered in several Sumatran pollen diagrams.
The archaeobotanical record
No finds of Canarium have been reported from the lower levels of Niah Cave, Sarawak, the Southeast Asian site thought to have been occupied earliest in the Late Pleistocene by Homo sapiens ssp. sapiens (Bellwood 1985), but Deraniyagala (1991: 18) seems to suggest that the nut-bearing C. zeylanicum might have been used by hominids in Sri Lanka from about 34,000 b.p. onward.
The earliest known occupation of Melanesia dates to c. 40,000 b.p. (Groube et al. 1986) and the oldest archaeobotanical record is for a large-fruited form of C. indicum from the Sepik-Ramu area of Papua New Guinea (Yen 1990: 262; Gorecki pers. comm.) said to date to 14,000 b.p. Kajale (1989) found C. zeylanicum endocarps in Sri Lanka dating to 12,500-10,000 b.p. associated with remains of bananas and Spriggs (1993) does not rule out the possibility of Pleistocene agriculture in the Pacific. Gosden (1992: 56) described Canarium with bananas as components of the Melanesian yam-taro complex, which also includes other tree crops, such as sago and breadfruit, not found in Sri Lanka. Charred Canarium fruit stones from Vietnamese Hoabinhian sites have been directly dated c. to 9000-8000 b.p. (Gorsdorf & Viet 1995: figure 3) but remains are present back to the Late Pleistocene at Con Moong Cave [ILLUSTRATION FOR FIGURE 1 OMITTED].
Yen (1985: 320; 1990: 262, 268) suggested that edible species of Canarium may have been introduced to the Bismarcks and the Solomons from New Guinea where they are naturally widespread. This presumes movement before the end of the Pleistocene. Gosden (1992: 59-60) went further, stating that Canarium was domesticated in the Pleistocene and dispersed over a considerable area of Melanesia by the mid Holocene, a 'domestication of the environment' (Groube 1989) in the form of forest clearance and genetic manipulation of plants. Yen (1995: 843) tentatively concurs. Palynology and analysis of microfossil charcoal from the Markham Valley of eastern Papua New Guinea (Garrett-Jones 1979) shows that grassland became established near Lake Wanum (35 m elevation) around 5300 b.p., after forest disturbance began about 8500 years ago. So there is some support for part of this theory. More recent evidence from Lake Hordorli (780 m elevation) in Irian Jaya indicates that disturbance taxa and microfossil charcoal began to increase around 10,500 b.p.; this 'may include a component from regrowth in clearings in the lowland forests' (Hope & Tulip 1994: 394). Other sites, e.g. Haeapugua in the Tari Basin (Haberle 1993) of the Southern Highlands Province of Papua New Guinea (1650 m) and Kelela Swamp (1400 m), in the upper Baliem Valley of Irian Jaya (Haberle et al. 1991), are probably too high for Canarium.
Canarium nuts are comparatively non-perishable when carbonized. While collection of wild nuts elsewhere in the world may have led to the selective conservation or extension in area of particular plant species, e.g. hazel (Corylus) in western Europe (cf. Smith 1970; Simmons et al. 1981), it did not lead to domestication. In the case of hazel, it is thought that people may have accidentally promoted the spread of a useful plant until it was realized that it would expand its area naturally after burning. Hazel seems to have been fairly abundant in the early post-glacial, whereas Canarium, as a rainforest tree, is likely to have been thinly dispersed in the forest. There is no evidence that hazel was transported ever considerable distances as Canarium seems to have been. These points may explain why Canarium was domesticated, and hazel was not.
Some species of Canarium are resinous; resin, damar, is a traditional export product of tropical forests (cf. Meilink-Roelofsz 1962). The wood is of lesser value, as it is often soft, there are much better timber trees in the lowland forests of Southeast Asia and Melanesia, but it is used in house-building.
Canarium species usually out-cross (Yen 1991) but seem to be low pollen producers; the pollen might be expected to increase relative to that of other forest trees if the trees were selectively conserved during more general forest destruction. This is what is so intriguing about the Sumatran evidence.
The pollen morphology of Canarium
The pollen morphology of eight Indo-Malayan genera in Burseraceae was examined by Mitra et al. (1977; taxonomic errors amended by Leenhouts (1978)).
Six of the 16 Sumatran species of Canarium were considered: C. denticulatum BI., C. dichotomum (Bl.) Miq., C. hirsutum Willd. (as C. longiflorescens Elmer ex Mill.), C. littorale Bl., C. patentinervium Miq. and C. pilosum Benn. The pollen grains were reported as trizonocolporate (tricolporate), prolate spheroidal, oval in meridional view, 25-50 microns in size and with lalongate ora (pores) and a surface patterning ranging from obscure - psilate - striate - rugulose. Pollen from Pachylobus edulis G. Don (Dacryodes edulis H.J. Lam) and Dacryodes rostrata H.J. Lam was oval in meridional view and of medium size. Eight species of Dacryodes occur in Sumatra, mostly at low altitude; D. laxa (Benn.) H.J. Lam is present in primary forest at 900-1200 m in north Sumatra.
Mitra et al. (1977) who studied C. asperum Benth. twice - as C. clemantis Merr. (an error for C. clementis) placed it in a Canarium-type pollen sub-group, and as C. thyrsoidieum Perkins in a Santiria-type sub-group. C. asperum has not been reported from Sumatra.
These indications that Canarium-type pollen can overlap pollen morphological sub-groups suggest circumspection is needed in identifications. The Santiria-type pollen illustrated in Maloney (1979) is very different from the Canarium-type pollen (which was mislabelled as Sapotaceae comp.) and matches modern reference material from C. littorale Bl. excellently (except that some of the fossil pollen grains are rugulate or reticulate and not psilate: these were not separated from each other during pollen-counting).
The fossil record of Canarium-type pollen
Most published Southeast Asian pollen diagrams show selected taxa only; where Canarium occurs, it is a minor contributor grouped with the forest pollen types rather than a discrete pollen curve. The pollen frequencies from all the north Sumatran sites analysed so far (except the undated and unpublished Sibisa Swamp) (Pea Bullok, Pea Sim-sim, Pea Sijajap and Tao Sipinggan) are shown on FIGURE 2. Estimated sample ages have been calculated using the 41 radiocarbon dates, but ignoring the statistical errors, so the chronology is illustrative rather than precise.
Canarium littorale Bl., which the fossil pollen most resembles, has (Leenhouts 1959b) four vernacular names in the Tapanuli region of north Sumatra where most of the pollen sites are located. So it does, or did, occur there: the region, now almost entirely deforested, has been so since the 1930s judging by the 1:100,000 Dutch maps of the area (HIND 1063, sheet 37, 3rd edition). Canarium was not collected in a vegetation survey made in 1973-74 (Maloney 1983).
The most detailed Late Quaternary-Holocene pollen diagrams from other island Southeast Asian sites are those of Stuijts (1993), who did not distinguish a Canarium-type pollen, nor has it been reported in diagrams from central Sumatra (Morley 1982; Newsome & Flenley 1988), Thailand (Maloney 1991), island Melanesia (Hope & Spriggs 1982; Haberle 1996), New Caledonia (Stevenson & Dodson 1995) or, even in the most recent levels, Polynesia (Ellison 1994; Flenley 1994; Kirch & Ellison 1994; Parkes & Flenley 1990; Parkes et al. 1992), although it may be in the grouped tree category of incompletely published diagrams.
Morley (1981) distinguished a Burseraceae type, in which he included Canarium, Dacryodes and Santiria, from lowland peninsular Malaysian Tasek Beta, and Morley (1982) a Santiria comp. (comparatus = comparable to), in which he grouped Santiria and Dacryodes, at Danau Padang (c. 950 m) in central Sumatra. He did not refer to Mitra et al. (1977), and reference pollen of Canarium littorale Bl. was unavailable to him. Maloney (1979) distinguished Santiria comp. along similar lines to Morley (1982). The north Sumatran sites are at 1300-1400 m altitude, so the pollen is likely to be mostly from Santiria, not Dacryodes. Supiandi & Furukawa (1986) recognized and illustrated a Santiria pollen type from Holocene lowland east Sumatran pollen sites, but did not report Canarium.
Canarium-type pollen in the north Sumatran fossil record
Canarium-type pollen in five radiocarbon-dated cores from highland north Sumatra is shown on FIGURE 2. Canarium-type pollen first appears in Pea Bullok Core A, 25-30,000 years ago before the most intense period of Pleistocene cold. A single pollen grain was present in a sample dating to about 21,000 b.p., and another two in two samples from around 20,500-20,400 b.p. Thereafter the pollen occurred at trace levels (less than 1%) at all except one depth from 19,400 b.p. to the present, although it contributed 4% of the total dry-land pollen and pteridophyte spores in the uppermost sample (this may be about 2200 years old, has a microfossil charcoal content of 12% total pollen, and Panicoid grass phytoliths containing carbon inclusions).
The Pea Bullok Core B pollen diagram differed somewhat. Canarium-type pollen was ever-present at trace levels from around 10,600 to 3400 b.p. Thereafter, during a time of forest clearance when microfossil charcoal occurred more consistently and there were phytoliths which had taken up carbon, it was absent. It reappeared in the uppermost level, which, in this instance, may be only about 900 years old. This sample also contained phytoliths with carbon inclusions and microfossil charcoal particles comprised 153% of total pollen. Carbon containing phytoliths had been present from around 4700 b.p.; there is evidence for earlier vegetation disturbance (Maloney 1996).
Canarium-type pollen is not present in the Pea Sim-sim record until about 10,700 b.p. (the oldest evidence for human occupation in Sumatra (Bronson & Asmar 1976) dates to c. 10,000 b.p.). Then it is found in all except one sample until 3500 b.p., usually as more than trace percentages, but fades from the record during forest-clearance phases which followed until 1500 b.p. It recurs in four samples which should date from about this time-period until 500 b.p., despite the occurrence of burning, as indicated by the presence of microfossil charcoal and phytoliths with carbon inclusions (which first appear at about 2600 b.p.). Canarium-type pollen reached a peak at 4% around 7800 b.p., when swamp forest was possibly being disturbed by people, with other peaks between c. 4200-3500 b.p. when open vegetation indicating Urticaceae/Moraceae pollen was commoner.
So Canarium-type pollen begins to emerge as a possible warm indicator, while recurrence after intensive forest clearance hints at conservation during felling or at planting.
Canarium-type pollen is present from the beginning of sediment deposition at Tao (Lake) Sipinggan. It reaches a significant peak around 11,100 b.p. (over 4% of total dry-land pollen and pteridophyte spores) when disturbance-indicating Urticaceae/Moraceae pollen was also quite common. Another peak of about 6% of total dry-land pollen and pteridophyte spores follows at c. 9750 b.p. when other shifts in the pollen curves suggest that vegetation disruption was taking place; there is a very large increase in tree-fern spores in the following sample. An earthquake occurred in this area in 1921 (Braak 1929); but the major fault movements are claimed to have ceased about 17,000 years ago (Hehanussa et al. 1987). Later tectonic activity cannot be ruled out though, and the vegetation disturbance cannot be ascribed unequivocally to a human cause.
Canarium-type pollen has a frequency of more than 2% in samples estimated c. 6358, 5100, 4150, 3365 and 2450 b.p. It disappears from the record (except in a sample estimated to date to about 450 b.p.) after 1700 b.p. when grasses came to prominence and there may have been wet rice cultivation around the lake and dry rice cultivation on the crater slopes (cf. Maloney 1996).
The pollen evidence from north Sumatra suggests that Canarium was probably not domesticated or selectively conserved during the periods of most intense forest destruction on the Toba Plateau from about 3500 b.p. onwards but the reappearance of the taxon at the top of the diagrams hints at planting. Some earlier peaks might indicate conservation, although a convincing case cannot be made for this. If it were not for the fact that Canarium littorale, which the pollen type most resembles, can grow at altitudes up to 2000 m, this taxon would be an important warm indicator. Its almost continuous curve from the opening of the Holocene suggests that it is normally a warm indicator, as it is in Vietnam (Gorsdorf & Viet 1995). It also seems to like wet conditions; that, as much as temperature, may explain its general absence in full glacial times.
Clearly, there is a now a need to investigate the pollen record of Canarium in Melanesian contexts to see if it lends stronger support for selective conservation or domestication. If the pollen is found there and in deposits from the rest of Melanesia, it might provide a further means of tracing the human dispersal of the taxon within radiocarbon datable contexts, but macrofossil finds must remain the main source of evidence.
Acknowledgements. The writer thanks two anonymous referees who noted errors in the original draft and suggested improvements. Any continuing defects are the responsibility of the author.
ALLEN, J. & C. GOSDEN (ed.). 1991. Report on the Lapita Homeland Project. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University. Occasional Papers in Prehistory 20.
BELLWOOD, P. 1985. Prehistory of the Indo-Malaysian archipelago. New York (NY): Academic Press.
BRAAK, C. 1929. Earthquakes, in L.M.R. Rutten (ed.), Science in the Netherlands East Indies: 75-9. Amsterdam: Schletens & Giltay.
BRONSON, B. & T. ASMAR. 1976. Prehistoric investigations at Tiako Panjang Cave, Sumatra, Asian Perspectives 18: 128-45.
CORNER, E.J.H. 1978. The freshwater swamp-forest of south Johore and Singapore. Singapore: Botanic Gardens Parks & Recreation Department.
COX, P.A. & S.A. BANACK (ed.). 1991. Islands, plants, and Polynesians: an introduction to Polynesian ethnobotany. Portland (OR): Dioscorides Press.
DERANIYAGALA, S.U. 1991. Man and environment during the Pleistocene in Sri Lanka, Bulletin of the Indo-Pacific Prehistory Association 10: 12-22.
ELLISON, J. 1994. Paleo-lake and swamp stratigraphic records of Holocene vegetation and sea-level changes Mangaia, Cook Islands, Pacific Science 48: 1-15.
FLENLEY, J.R. 1994. Pollen in Polynesia: the use of palynology to detect human activity in the Pacific islands, in J.G. Hather (ed.), Tropical archaeobotany: applications and new developments: 202-14. London: Routledge.
FREDERICKSEN, C., M. SPRIGGS & W. AMBROSE. 1993. Pamwak rockshelter: a Pleistocene site on Manus Island, Papua New Guinea, in Smith et al.: 144-52.
GARRETT-JONES, S.E. 1979. Evidence for changes in Holocene vegetation and lake sedimentation in the Markham Valley, Papua New Guinea. Unpublished Ph.D thesis, Australian National University, Canberra.
GLOVER, I.C. 1979. Prehistoric plant remains from South east Asia, with special reference to rice, in M. Taddei (ed.), South Asian archaeology 1977: 7-37. Naples: Institute Universitario Orientale. Seminario de Studi Asiatica I.
GORMAN, C.F. 1969. Hoabinhian: a pebble-tool complex with early plant associations in Southeast Asia, Science 163: 671-3.
1971. Excavations at Spirit Cave, North Thailand, Asian Perspectives 13: 79-107.
GORSDORF, J. & N. VIET. 1995. Berlin 14C dates of archaeological sites in Vietnam, Radiocarbon 37: 221-5.
GOSDEN, C. 1992. Production systems and the colonization of the western Pacific, World Archaeology 24(1): 55-69.
1993. Understanding the settlement of the Pacific Islands in the Pleistocene, in Smith et al.: 131-6.
GROUBE, L. 1989. The taming of the rainforests: a model for Late Pleistocene forest exploitation in New Guinea, in Harris & Hillman (ed.): 292-304.
GROUBE, C. et al. 1986. A 40,000 year-old human occupation site at Huon Peninsula, Papua New Guinea, Nature 324: 453-5.
HABERLE, S. 1993. Pleistocene vegetation change and early human occupation of a tropical mountainous environment, in Smith et al.: 109-22.
1996. Explanations for palaeoecological changes an the northern plains of Guadalcanal, Solomon Islands: the last 3200 years, Holocene 6: 333-8.
HABERLE, S., G.S. HOPE & Y. DEFRETES. 1991. Environmental change in the Baliem Valley, montane Irian Jaya, Republic of Indonesia, Journal of Biogeography 18: 25-40.
HARRIS, D.R. & G.C. HILLMAN. 1989. Foraging and farming: the evolution of plant domestication: London: Unwin Hyman.
HEHANUSSA, P.E., S. FUJII & T. YOKOYAMA. 1987. New dates of fluvio-lacustrine deposit around Lake Toba, Indonesia, International Project on Palynology Newsletter 4: 17-20.
HOPE, G.S. & M.J.T. SPRIGGS. 1982. A preliminary pollen sequence from Aneityum Island, southern Vanuatu, Bulletin of the Indo-Pacific Prehistory Association 3: 88-94.
HOPE, G.S. & J. TULIP. 1994. A long vegetation record from lowland Irian Jaya, Indonesia, Palaeogeography, Palaeoclimatology, Palaeoecology 109: 385-98.
KAJALE, M.D. 1989. Mesolithic exploitation of wild plants in Sri Lanka: archaeological study at the cave site of Beli-Lena, in Harris & Hillman: 269-81.
KIRCH, P.V. 1982. Second millenium BC arboriculture in Melanesia: archaeological evidence from the Mussau Islands, Economic Botany 47: 225-40.
KIRCH, P.V. & J. ELLISON. 1994. Palaeoenvironmental evidence for human colonization of remote Oceanic islands, Antiquity 68: 310-21.
KIRCH, P.V. & D.E. YEN. 1982. Tikopia: the prehistory and ecology of a Polynesian outlier. Honolulu (HI): B.P. Bishop Museum. Bulletin 238.
LEENHOUTS, P.W. 1955. The genus Canarium in the Pacific. Honolulu (HI) B.P. Bishop Museum. Bulletin 216.
1959a. Revision of the Burseraceae of the Malaysian area in a wider sense. Xa. Canarium Stickm., Blumea 9(2): 275-647.
1959b. Burseraceae, Flora Malesiana Series I, 5: 209-96.
1978. The pollen morphology of Burseraceae: a taxonomic comment, Grana 17: 175-7.
MALONEY, B.K. 1979. Man's influence on the vegetation of north Sumatra: a palynological study. Unpublished Ph.D thesis, University of Hull.
1991. Palaeoenvironments of Khok Phanom Di: the pollen, pteridophyte spore and microscopic charcoal record, in C.F.W. Higham & R. Bannanurag (ed.), The excavation of Khok Phanom Di: The biological remains: 7-134. London: Society of Antiquaries. Research reports 48.
1996. New perspectives on possible early dry land and wet land rice cultivation in highland north Sumatra. Hull: University of Hull, Centre for South-East Asian Studies. Occasional Paper 29.
MARSHALL, B. & J. ALLEN. 1991. Excavations at Panakiwuk Cave, New Ireland, in Allen & Gosden: 59-91.
MEILINK-ROELOFSZ, M.A.P. 1962. Asian trade and European influence in the Indonesian archipelago between 1500 and about 1630. The Hague: Mouton.
MITRA, K., M. MONDAL & S. SAHA. 1977. The pollen morphology of Burseraceae, Grana 16: 75-9.
MORLEY, R.J. 1981. The palaeoecology of Tasek Hera, a lowland swamp in Pahang, west Malaysia, Singapore Journal of Tropical Geography 2: 49-56.
1982. A palaeoecological interpretation of a 10,000 year pollen record from Danau Padang, central Sumatra, Indonesia, Journal of Biogeography 9: 151-90.
NEWSOME, J. & J.R. FLENLEY. 1988. Late Quaternary vegetational history of the Central Highlands of Sumatra II: Palaeopalynology and vegetational history, Journal of Biogeography 15: 363-86.
PARKES, A. & J.R. FLENLEY. 1990. Hull University Moorea Expedition, 1985: final report. Hull: University of Hull, School of Geography and Earth Resources. Miscellaneous Series 37.
PARKES, A., J.T. TELLER & J.R. FLENLEY. 1992. Late Quaternary environmental history of the Lake Vaihiria drainage basin, Tahiti, French Polynesia, Journal of Biogeography 19: 431-47.
REYNOLDS, T.E.G. 1992. Excavations at Banyan Valley Cave, Northern Thailand: a report on the 1972 season, Asian Perspectives 31: 77-97.
ROE, D. 1992. Investigations into the Prehistory of the Central Solomons, in J.C. Galipaud (ed.), Poterie Lapita et peuplement: 91-102. Noumea: ORSTOM.
SIMMONS, I.G., G.W. DIMBLEBY & C. GRIGSON. 1981. The Mesolithic, in I.G. Simmons & M.A. Tooley (ed.), The environment in British prehistory: 82-124. London: Duckworth.
SMITH, A.G. 1970. The influence of Mesolithic and Neolithic man on British vegetation: a discussion, in D. Walker & R.G. West (ed.), Studies in the vegetational history of the British Isles: 81-96. Cambridge: Cambridge University Press.
SMITH, M.A., M. SPRIGGS & B. FANKHAUSER (ed.). 1993. Sahul in review: Pleistocene archaeology in Australia, New Guinea and Island Melanesia. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University. Occasional Papers in Prehistory 24.
SPRIGGS, M. 1991. Nissan, the island in the middle. Summary report on excavations at the north end of the Solomons and the south end of the Bismarcks, in Allen & Gosden: 222-43.
1993. Pleistocene agriculture in the Pacific: why not? in Smith et al.: 137-43.
SPRIGGS, M. & S. WICKLER. 1989. Archaeological research on Erromango: recent data on southern Melanesian prehistory, Bulletin of the Indo-Pacific Prehistory Association 9: 68-91.
STEVENSON, J. & J.R. DODSON. 1995. Palaeoenvironmental evidence for human settlement of New Caledonia, Archaeology in Oceania 30: 36-41.
STUIJTS, I.-L.M. 1993. Late Pleistocene and Holocene vegetation of west Java, Indonesia, Modern Quaternary Research in Southeast Asia 12: 1-188.
SUPIANDI, S. & H. FURUKAWA. 1986. A study of floral composition of peat soil in the Batang Hari river basin of Jambi, Sumatra, Southeast Asian Studies 24(2): 113-32.
SWADLING, P., N. ARAHO & B. IVUYO. 1991. Settlements associated with the inland Sepik-Ramu sea, Bulletin of the Indo-Pacific Prehistory Association 11: 92-112.
SWADLING, P., B.H. SCHAUBLIN, P. GORECKI & F. THEISLER. 1988. The Sepik-Ramu. Boroko: Papua New Guinea National Museum.
TRAN VAN BO et al. 1970. Dong vat va thuc vat O Dong-dau [Animal and plant remains at Dong-dau], Khao c'o hoc 7/8: 133-9.
WHISTLER, W.A. 1991. Polynesian plant introductions, in Cox & Banack (ed.): 41-66.
WICKLER, S. 1990. Prehistoric Melanesian exchange and interaction: recent evidence from the northern Solomon Islands, Asian Perspectives 29: 135-54.
YEN, D.E. 1977. Hoabinhian horticulture: the evidence and the questions from northwest Thailand, in J. Allen, J. Golson & R. Jones (ed.), Sunda and Sahul: prehistoric studies in Southeast Asia, Melanesia and Australia: 567-99. London: Academic Press.
1985. Wild plants and domestication in Pacific Islands, in V.N. Misra & P. Bellwood (ed.), Recent advances in Indo-Pacific prehistory: 315-26. New Delhi: Oxford & IBH.
1990. Environment, agriculture and the colonisation of the Pacific, in D.E. Yen & J.M.J. Mummery (ed.), Pacific production systems: 258-77. Canberra: Department of Prehistory, Research School of Pacific Studies, Australian National University. Occasional Papers in Prehistory 18.
1991. Polynesian cultigens and cultivars: the questions of origin, in Cox & Banack: 67-95.
1995. The development of Sahul agriculture with Australia as bystander, Antiquity 69: 831-47.…