The Poison Hunters of Umhlatuzana: Researchers Find the Earliest Direct Evidence of Toxic Weaponry
- 4 days ago
- 7 min read
Recent Study Reveals 60,000+-Year-Old Toxic Alkaloids on Microliths from Umhlatuzana Rock Shelter, South Africa
Black and white, salt and pepper, peanut butter and jelly, so many things that work well together, you know what else works well together, arrows and poison.
Sure, you don’t need to combine these things, but somehow putting them together tends to make them so much better. Maybe not peanut butter and jelly, I don’t like that combo, but I do like the combination of arrows and poison. And I am not the only one, poisoned weapons have been a hallmark of advanced hunter-gatherer technology for millennia. Though evidence for it is scarce, the deeper into ancient history you go.
In a recent study, Sven Isaksson, Anders Högberg, and Marlize Lombard microchemically and biomolecularly identified the oldest traces of toxic plant alkaloids on backed microliths from Umhlatuzana Rock Shelter; in other words, they identified the oldest direct evidence of poison on stone tools. What type of poison, what type of stone tools? All your questions shall be answered.
History of Poisoned Arrows
When humans began consuming meat, this brought with it multiple evolutionary advantages, including resistance to various diseases, enhanced life expectancy, and neuronal growth. These benefits likely contributed to the development of weapon-assisted hunting, including bowhunting. Evidence of bowhunting has been found in the Pleistocene of southern Africa, dating back to ~ 70–60 thousand years (ka). This evidence, in the form of bone arrowheads, closely resembles Holocene examples of poisoned bone arrowheads. The use of poisoned hunting weapons marks a remarkable innovation in meat acquisition strategies and highlights the development of technical and cognitive complexity.
Before the study by Isaksson and his colleagues, the earliest direct evidence of poisoned arrows was dated to the mid-Holocene. Direct evidence is crucial here, as direct evidence means we found a poisoned arrow; indirect evidence means we found the poison but no arrow, or we found arrows but no evidence of poison, so we can only infer the existence of the other. So, the earliest direct evidence of poisoned arrows is from Egypt, dated to around 4431 to 4000 BP, and a compound poison dated to around 6700 BP at Kruger Cave in South Africa.
The earliest indirect evidence comes from ca. 24 ka and 35 ka, respectively. The first is a ‘poison applicator’ found at Border Cave in KwaZulu-Natal (KZN), South Africa (of which the source-plant remains highly debated), and the other, in the same site, is a lump of beeswax mixed with plant poison.
Now, alongside bone points, hunters throughout the African Holocene also made use of backed microliths as arrow tips. Such artifacts have been recovered from various sites, including the Umhlatuzana Rock Shelter in KZN, which is dated to >60 ka, and have been interpreted as arrowheads.
What makes these microliths special is that they had something adhering to them, but what was this something? Blood, curation glue, someone’s peanut butter and jelly-stained fingerprints? Or perhaps poison?
To find out a bit more, sleuthing was necessary.
Finding Umhlatuzana’s Poison
Umhlatuzana was discovered in 1982 by Dr. R. R. Maud during a geological survey of the proposed Johannesburg-Durban N3 highway. Located around 35 km west of Durban on the Kerkmans 915 farm. It was named for the nearby Umhlatuzana River, located just 100 m below the site. It was excavated in a short time between 5 May and 4 July 1985 by Dr. Jonathan Kaplan with two objectives: the first, to rescue the site’s material, which was endangered by the construction of the highway, and the second, to investigate hunter-gatherer and farmer relations. Thankfully, the new highway did not destroy the site, and new excavations in 2018 could be conducted.
The new excavations’ initial aim was to conduct a high-resolution geoarchaeological study to clarify the site formation processes before re-excavation, and a revised stratigraphic division took place. Upon re-excavation of the site, various microliths were found, among them were the 10 pieces of backed quartz, which were selected for the study.
These microliths exhibited macrofractures consistent with projectile use as well as residue traces consistent with adhesives containing resin or water-soluble gum mixed with ocher. Optically stimulated luminescence dating revealed the layer they came from to be 60 ± 3.5 ka old.
Using GC-MS analysis, the researchers detected plant-derived toxic alkaloids (buphanidrine and epibuphanisine) on 5 of the 10 tested microliths. These same compounds have been found on 250-year-old poisoned bone arrowheads and on modern Boophone disticha bulb extracts, which are a known arrow poison in southern Africa.
The stability of these alkaloids allowed them to be preserved for thousands of years, something that is not always possible with other components. This, together with the impact damage of the tools, provided the researchers with the earliest direct evidence of poisoned arrowheads.
And how did this poison work, and why was it necessary in the first place?
60,000+ Year Old Poison
Unlike other hunters around the world, who may have fashioned arrows that kill on impact or at least gravely wound the animal. Southern African hunters did not manufacture arrows that killed on impact. Rather, the arrow was designed to hit the animal and dislodge, leaving only the head behind while the shaft fell to the ground.
By itself not very effective in killing anything larger than a shrew. So, poison is necessary. The poison identified on the microliths from Umhlatuzana was derived from the B. disticha plant remains. These plants have been recovered from South African Holocene contexts (ca. 4000–500 BP), where the less toxic leaves or the scales of the outer bulb were used as a preservative. However, they also have a more deadly side. Also locally called ‘gifbol’ in Afrikaans, meaning ‘poison bulb,’ the plant excretes a milky-white substance that can be used as a poison.
Carl Peter Thunberg, an early European traveler who ventured into the hinterlands of the Cape of Good Hope (1772–1774), wrote that the B. disticha plant was used by the indigenous hunters mainly to poison the arrow shafts used to hunt game such as springbok. On his travels, he also collected some arrows, two of which were included in the current study as a control, and guess what they tested positive for: buphanidrine and epibuphanisine, and were thus consistent with B. disticha bulb exudate.
While B. disticha contains many toxic compounds, buphanidrine stands out due to its chemical stability, which allowed it to be preserved for more than 60,000 years. Although there are no direct archaeological or palaeobotanical data that show the plant grew near Umhlatuzana during Marine Isotope Stage 4 (71 to 57 ka), it is indigenous to KZN.
Additionally, its ability to survive for a century or more despite frequent droughts and fire cycles would have likely made it widespread across South Africa’s southern, northern, and eastern regions. This suggests it could very well have been encountered by the local hunter-gatherers 60 ka ago, who would have been familiar with its properties.
In addition to how easy it likely was to procure, the milky bulb extract could be used with minimal processing. To use as a poison, the hunter-gatherers would simply need to sun-dry the extract until it turned into a gum-like substance, which could then be applied to arrow tips.
If they wanted to, they could also reduce the substance by cooking or heating it over a fire and mix it with other plant- or animal-derived substances. Once done, you have a lethal toxin. Tests have shown that even small quantities can be lethal to rodents within 20 to 30 minutes. In humans, the poison can induce nausea, muscular flaccidity, visual impairment, stertorous breathing, coma, increased pulse, or feeble pulse, respiratory paralysis, dyspnea, hyperemia, and edema of the lungs.
In short, it would not be a pleasant experience to be poisoned. Once the desired animal was hit, it likely would experience similar symptoms. The hunter-gatherers would then commence to track the animal (for hours or days) as it slowly grew weaker and eventually died.
The research not only identified the oldest direct evidence of poisoned weapons in the world, but also highlighted the sophisticated cognitive abilities and complex knowledge systems necessary to identify and extract plant toxins and understand the delayed effects they would have on the hunter-gatherers’ prey.
In a recent study, the oldest direct evidence of poisoned weapons in the form of microlithic arrowheads was identified. Dating back over 60,000 years, these microliths from Umhlatuzana Rock Shelter contained traces of toxic alkaloids from the Boophone disticha plant. The plant would have caused various symptoms in the hunted animal, slowly letting it die as hunters tracked it. The discovery highlights the remarkable cognitive sophistication of these early hunter-gatherers.
What poisonous plants do you know in your area of the world? And when do you think humans first began using poisons?
Originally Posted on Medium
Reference
Isaksson, S., Högberg, A., & Lombard, M. (2026). Direct evidence for poison use on microlithic arrowheads in Southern Africa at 60,000 years ago. Science advances, 12(2), eadz3281. https://doi.org/10.1126/sciadv.adz3281
Kaplan, J. (1990). The Umhlatuzana rock shelter sequence: 100 000 years of Stone Age history. Southern African Humanities, 2(11), 1–94.
Sifogeorgaki, I., Klinkenberg, V., Esteban, I., Murungi, M., Carr, A. S., van den Brink, V. B., & Dusseldorp, G. L. (2020). New excavations at Umhlatuzana Rockshelter, KwaZulu-Natal, South Africa: a stratigraphic and taphonomic evaluation. African Archaeological Review, 37(4), 551–578.
Comments