Rumphius's Cannabis: A Seventeenth-Century Naturalist and the Oldest Plant Dispersal Nobody Followed Up On
- Éloïse
- Feb 22
- 12 min read
Updated: 1 day ago
V1.12 | Éloïse | 23.02.2026
Georg Eberhard Rumphius arrived in Indonesia on Ambon Island in 1654 as a merchant for the Dutch East India Company and almost immediately began cataloguing the plants of the Spice Islands. He would spend the rest of his life there, eventually producing the Herbarium Amboinense, a six-volume catalogue of some 1,200 species that remains a foundational reference for the flora of the Moluccas. The Europeans called him Plinius Indicus, the Pliny of the Indies and made him a member of the Academy of Sciences in Vienna.
The work nearly didn't survive. Rumphius went blind from glaucoma in 1670 and spent the remaining three decades of his life dictating descriptions to assistants, among them his wife Suzanna and his son Paul August, who also made the botanical illustrations and the only known portrait of his father. Suzanna was killed, along with a daughter, when a wall collapsed on them during the earthquake and tsunami of February 1674. In 1687 a fire destroyed his library, his manuscripts and the original illustrations for the Herbarium. He and his remaining assistants rebuilt the work, completing it in 1690, only for the ship carrying the manuscript to the Netherlands to be sunk by the French. It survived because Governor Camphuys, an amateur astronomer who had personally reviewed the text, had insisted on keeping a copy in Ambon. The manuscript finally reached the Netherlands in 1696, only to be embargoed by The Company, who were afraid it contained sensitive commercial intelligence about the spice trade. Rumphius died in 1702 and the embargo was lifted in 1704, but it took decades more to find a publisher. Johannes Burman finally brought out the Latin edition in 1741, thirty-nine years after the death of a man who had been blind for the last thirty-two years of his life.
The Herbarium Amboinense is well known to botanists. The cannabis chapter is not. I have yet to find anything more than a passing reference to it in the context of cannabis during my research. Yet Chapter XXXIV of Volume 5, spanning pages 208 to 211 with the accompanying Plate LXXVII, is a detailed ethnobotanical account of cannabis in the eastern Indonesian archipelago describing three forms of the plant, multiple preparation methods, medicinal uses observed first-hand and a Malay cultural framework for intoxication.
I think it is one of the most important primary sources in cannabis history and its implications for the dispersal of the plant across the Austronesian world have been missed.
What Rumphius Recorded
Rumphius describes cannabis cultivated in gardens on Ambon, maintained from Javanese seed that had to be restocked every two generations because the plants could not sustain themselves locally. He records the Malay name Ginji, distinguishes male from female plants with precise morphological descriptions (the male six to seven feet tall, the female reaching ten) and documents a range of uses. Cannabis was smoked with tobacco, mixed with betel, prepared as an edible compound called Majub with nutmeg and opium, steeped as a tea and used medicinally: the root chewed for gonorrhea, the leaves boiled with cloves and nutmeg for respiratory complaints. He personally witnessed people smoking it in Ambon and recorded how the effects varied by temperament. He documented the Malay concept of Hayal, a state of intoxicated transcendence that the Malays described as "befitting their Kings."
The medicinal uses came from firsthand observation. Hitu Muslims were using plants from Rumphius's own garden and he recorded their preparations in detail. The Hayal concept places cannabis use within a Malay framework of royal prerogative. The preparations parallel those documented across India and the Middle East but are recorded here from the easternmost edge of the known cannabis world.
The most consequential passage, though, concerns a third variety.
Ginji Papoua
After describing the male and female cannabis grown in Ambon's gardens from Javanese seed, Rumphius writes:
In Ambon there is also a third species, called Ginji Papoua, short of stem, with small narrow leaves heaped together in bundles, between which the seed with its pods stick out, brought from the Papuan and Sula Islands, but mostly so named because its foliage is so curled and wrinkled like the hair of Papuans.
A morphologically distinct form: compact, bundled leaf structure, originating from the Papuan and Sula Islands at the far eastern reaches of the Indonesian archipelago. It has its own name. It is different enough that the botanist Hasskarl, examining Rumphius's description in 1866, made it the type of a new variety: Cannabis sativa var. crispata.
It is also the earliest documented cannabis from anywhere near Papua. How did it get to the Papuan and Sula Islands and how long had it been there? I cannot find this question addressed anywhere in the literature.
The Conventional Account
The standard story of cannabis in this part of the world is one of late arrival. Cannabis originates in Central Asia, spreads westward to Europe and eastward to India through ancient trade and migration, reaches Southeast Asia through Indian maritime networks sometime in the medieval period and arrives in the far eastern archipelago and the Pacific essentially in living memory. The Wikipedia article on cannabis in Papua New Guinea states that "Cannabis is believed to have been introduced to PNG by Australian and American expatriates in the 1960s-1970s."
Rumphius's Ginji Papoua predates that timeline by three centuries. He is describing a cannabis variety from the Papuan and Sula Islands in the 1690s, one distinct enough from the Javanese-seeded Ambon plants to warrant its own name. A plant that has developed distinct local morphology and acquired a vernacular name has been somewhere for a while. What Rumphius describes looks like an established regional variety and it points toward a dispersal route that has received almost no attention in the cannabis literature.
The Austronesian Vector
The answer might lie at the other end of the Indian Ocean, in Madagascar.
Sometime in the last centuries BCE, Austronesian-speaking peoples from Island Southeast Asia sailed outrigger canoes across the Indian Ocean and settled Madagascar, more than 6,000 kilometers from their homeland. It is one of the longest maritime migrations in human prehistory. Pearce and Pearce (2010) argue it was not an isolated voyage but part of an established transoceanic spice trade, the Cinnamon Route, which followed the Equatorial Current from the Spice Islands to East Africa, with Madagascar colonized as a rest and supply station along the way. The Malagasy language is Austronesian. The culture retains Austronesian architecture, music, funerary practices and crop plants. Genetic studies confirm the connection. What is less discussed is what the settlers brought with them botanically and this is where it gets interesting for cannabis.
At Lake Tritrivakely in the central highlands of Madagascar, cannabis pollen (or possibly Humulus, which is palynologically difficult to distinguish) appears in a sediment core at an interpolated date of around 2200 BP, roughly 200 BCE. A second site, Lake Kavitaha, shows cannabis pollen around AD 500. Cannabis has no natural range in Madagascar. Scholars like Burney and Blench treat its pollen in these cores as a proxy for human presence: someone carried the plant to Madagascar. The Austronesians, who were simultaneously carrying rice, taro, banana, coconut and sugarcane across the same ocean, are the most likely candidates. Crowther et al. (2016) documented these cultigens as the "first archaeological signature of the westward Austronesian expansion." Cannabis may have been part of the same package — part of what Kirch calls the "transported landscape," the ensemble of crops, animals and cultural technologies that Austronesian colonizers carried into every new settlement.
If it was, then we need to rethink what cannabis in the eastern archipelago means. Its presence in Papua and the Sula Islands is typically treated as something requiring a modern explanation, a recent introduction from the west. But if Austronesian voyagers were already carrying cannabis across the Indian Ocean two thousand years ago, then finding it at the eastern edge of their maritime world is exactly what we would expect.
The question changes from "how did cannabis get to Papua?" to "how long has cannabis been part of Austronesian material culture?"
The Homeland Debate
The implications of this depend partly on where the Austronesians themselves came from, which is one of the most contested questions in Pacific archaeology.
The dominant model, Peter Bellwood's "Out of Taiwan" hypothesis, proposes that Austronesian-speaking peoples originated in Taiwan around 6,000 years ago and expanded rapidly southward through the Philippines into Island Southeast Asia, an express train to the Bismarcks and beyond. Under this model the people of the eastern Indonesian archipelago are relatively recent arrivals from the north.
I find other models more persuasive and the genetic evidence increasingly supports them. The "Polynesian motif," a unique suite of four mutations in mitochondrial DNA that defines Oceanic populations, is found throughout Wallacea and coastal Melanesia but is almost absent west of Wallace's Line. It is not found in Taiwan, the Philippines, or China. Richards, Oppenheimer and Sykes estimate it originated in Wallacea around 17,000 years ago and a Y-chromosome equivalent (the M38 mutation) dates to roughly 11,500 years ago in the same region. Soares et al. (2008) used complete mitochondrial genome sequencing to demonstrate that the direction of gene flow was from Island Southeast Asia to Taiwan, not the reverse. The evidence from commensal animals is equally clear: Matisoo-Smith and Robins traced the Pacific rat (Rattus exulans), which the Lapita peoples carried into every island they colonized, to its greatest haplogroup diversity in the Spice Islands, specifically Halmahera. A major worldwide study of pig domestication (Larson et al. 2005) showed that the pig carried into the Pacific was unrelated to the Taiwanese wild boar and unrelated to pigs in the Philippines, the supposed transit route.
These findings align with models that place Austronesian origins in the archipelago itself.
Wilhelm Solheim's "Nusantao" concept, developed from the 1970s onward, locates the origin of Austronesian maritime culture in the southern Philippines and eastern Indonesia. Stephen Oppenheimer's "Out of Sundaland" hypothesis argues along similar lines, pointing to rising sea levels at the end of the last glacial period as the driver of population dispersals outward from the Indonesian archipelago. Pearce and Pearce (2010) build on this evidence to argue that the Spice Islands functioned as an ancient maritime trading hub, with Spice Island mariners following warm currents out of the West Pacific Warm Pool to colonize Madagascar, New Zealand and Hawaii from the same Wallacean base. These remain minority positions in Pacific archaeology, but the genetic and archaeological evidence continues to complicate the Out of Taiwan consensus and the debate is far from settled.
When it comes to cannabis, the distinction is important. Under these models, the people of the eastern Indonesian archipelago are among the oldest maritime populations in the region, not recent arrivals from Taiwan but an ancient seafaring culture with deep roots in Wallacea. If cannabis was already part of their translocated landscape when they sailed to Madagascar (and the pollen data suggests it was), then its presence in Papua and the Sula Islands may be very old indeed, reflecting an association between this plant and these maritime peoples that goes back much further than any Indian trade network.
Speculation, to be sure - but the speculation has structure. I am connecting a primary source from the 1690s to pollen data from the opposite end of the Austronesian world and reading both through genetic evidence that increasingly points to Wallacea as the origin of Austronesian expansion. As far as I can determine, this connection has not been made before.
The Australian Question
There is a cannabis variety in Australia known as Australian Bastard Cannabis (ABC), a possibly feral (or guerrilla grown?) form of which seems to have been first documented near Sydney in the 1970s. It is compact, with tiny curled and wrinkled leaves that lack the typical cannabis serration and a grow in a bushy habit that makes it almost unrecognisable as cannabis. Its origins are debated. Some attribute it to Indian hemp germ-plasm sent by the botanist William Roxburgh from Calcutta in 1802 and 1803. Others point to Joseph Banks's earlier hemp trials. Others yet suggest it "arose from Indian cannabis introduced into Australia in the nineteenth century crossing with '70s imports from regions such as Thailand or Papua New Guinea."
We see morphological parallels with Rumphius's Ginji Papoua. Both are compact plants with unusual leaf morphology that makes them difficult to identify as cannabis. And the geography between them is connected by one of the oldest trade routes in the region. Makassan trepangers from Sulawesi sailed annually to Arnhem Land to harvest sea cucumbers, a trade documented from the 1650s and possibly much older. C.C. Macknight, who wrote the foundational study of this trade in 1976, found physical evidence of Makassan campsites along the northern Australian coast. Tobacco was already being exchanged with Yolngu communities through these networks. Psychoactive plant material was moving across the Arafura Sea well before any European arrived.
The morphological descriptions of Ginji Papoua and ABC are similar: Rumphius describes "curled and wrinkled" foliage, which is ABC's defining trait. These could be different mutations producing superficially similar results, or they could be related. Only genetic analysis would resolve it. We raise the comparison because it seems worth raising: a seventeenth-century Papuan cannabis variety and a supposedly modern but largely unknown Australian, possibly feral variety, separated by a few hundred kilometers of sea and connected by centuries of documented maritime trade that sound similar, appearance wise. It's a long shot, but there might be something there.
What Comes Next
The Herbarium Amboinense is freely available on the Biodiversity Heritage Library. The cannabis chapter has been translated into English in Beekman's 2011 Yale edition. The primary source is there for anyone to read.
The follow-up research is where the gaps are. We see four priorities:
Fieldwork in the Sula Islands and Indonesian Papua. If Ginji Papoua or its descendants still survive, they need to be sampled, described and genetically characterized. This should be a priority for landrace conservation fieldwork.
Comparison of Rumphius's descriptions with modern Southeast Asian landrace morphology. Plate LXXVII shows narrow-leaflet plants consistent with tropical drug-type cannabis. Placing this seventeenth-century illustration alongside contemporary landrace classification would establish a morphological baseline for the region.
Genetic comparison between ABC and eastern Indonesian cannabis. Medicinal Genomics has done limited genomic work on ABC. Testing for genetic relatives in the eastern archipelago would address the possible connection directly.
The Austronesian cannabis dispersal hypothesis. The Madagascar pollen data is currently discussed only in the context of human settlement chronology, with cannabis as a proxy for human presence. The dispersal history of the plant itself, as a cultigen carried across the Austronesian maritime world, deserves attention on its own terms.
Rumphius's Chapter XXXIV belongs in the cannabis literature alongside Garcia ab Orta, Linnaeus and Lamarck. The Cannabis indica naming priority alone would justify the attention. The Ginji Papoua is the more consequential observation: a distinct, named cannabis variety from the edge of the Austronesian world, documented three centuries ago and then left alone. The research needs to catch up with the primary source.
The full text of Rumphius's cannabis chapter is analyzed on our wiki page: Herbarium Amboinense. The original Latin and Dutch text is available at the Biodiversity Heritage Library.
References cited:
Solheim, W.G. II (1984–85). "The Nusantao hypothesis." Asian Perspectives 26(1): 77–88.Beekman, E.M. (2011). The Ambonese Herbal, vols. 1–6. New Haven: Yale University Press.
Blench, R. (2007). "New palaeozoogeographical evidence for the settlement of Madagascar." Azania 42: 69–82.
Burney, D.A. (1987). "Late Holocene vegetational change in central Madagascar." Quaternary Research 28(1): 130–143.
Crowther, A. et al. (2016). "Ancient crops provide first archaeological signature of the westward Austronesian expansion." PNAS 113(24): 6635–6640.
Jarvis, C.E. (2019). "Georg Rumphius' Herbarium Amboinense (1741–1750) as a source of information for Carl Linnaeus." Gardens' Bulletin Singapore 71 (Suppl. 2): 87–107.
Kirch, P.V. (2000). On the Road of the Winds: An Archaeological History of the Pacific Islands Before European Contact. Berkeley: University of California Press.
Larson, G. et al. (2005). "Worldwide phylogeography of wild boar reveals multiple centers of pig domestication." Science 307: 1618–1621.
Macknight, C.C. (1976). The Voyage to Marege': Macassan Trepangers in Northern Australia. Melbourne: Melbourne University Press.
Matisoo-Smith, E. and Robins, J.H. (2004). "Origins and dispersals of Pacific peoples: evidence from mtDNA phylogenies of the Pacific rat." PNAS 101(24): 9167–9172.
Merrill, E.D. (1917). An Interpretation of Rumphius's Herbarium Amboinense. Manila: Bureau of Printing.
Oppenheimer, S. (2004). "The 'Express Train from Taiwan to Polynesia': on the congruence of proxy lines of evidence." World Archaeology 36(4): 591–600.
Pearce, C.E.M. and Pearce, F.M. (2010). Oceanic Migration: Paths, Sequence, Timing and Range of Prehistoric Migration in the Pacific and Indian Oceans. Dordrecht: Springer.
Rumphius, G.E. (1747). Herbarium Amboinense, vol. 5, pp. 208–211. Amsterdam: Changuion, Uytwerf & Catuffe.
Soares, P. et al. (2008). "Climate change and postglacial human dispersals in Southeast Asia." Molecular Biology and Evolution 25(6): 1209–1218.
Comments