The oldest evidence for the use of arrow poison globally was long thought to come from Egypt, dating to 4,000 years ago. It was a black, toxic residue on bone arrowheads from a tomb at the Naga ed Der archaeological site.

New evidence from southern Africa is challenging this.

New research has found poison on stone arrow tips from South Africa dating to 60,000 years ago. It is the oldest direct evidence for hunting with poisoned arrows.

This adds to what is already known about the know-how of ancient African bowhunters. These abilities may have contributed to our species’ long and flourishing evolution in the region, and ultimately the successful spread of Homo sapiens out of Africa.

Hunter-gatherers in southern Africa

The evidence comes from Umhlatuzana Rock Shelter, in South Africa’s KwaZulu-Natal province. The site was partly excavated in the 1980s to preserve archaeological material that could be damaged during the construction of the N3 highway between the cities of Durban and Pietermaritzburg.

Two of the authors (Marlize Lombard and Anders Högberg) at Umhlatuzana rock shelter, where the poisoned arrow tips were excavated. Gerrit Dusseldorp, Author provided (no reuse)

Umhlatuzana is recognised as an important Stone Age site where hunter-gatherers lived at least 70,000 years ago. It is one of only a few sites in southern Africa where people continued to live until just a few thousand years ago.

In southern Africa, people have a long history of hunting with poisoned arrows. For example, a team of South African and Swedish archaeologists found residues on arrow tips dating to between a few centuries and 1,000 years ago, that revealed how different arrow poison recipes were used.

Recently, three bone arrowheads stored in a poison-filled bone container were reported from Kruger Cave in South Africa dating to almost 7,000 years ago. This pushed back direct molecular evidence of arrow poison use to about 3,000 years before the Egyptian poisoned arrows.

Traces of poison have previously been found on a stick and in a lump of beeswax dating to between 35,000 and 25,000 years ago at Border Cave in KwaZulu-Natal. These were seen as indirect suggestions of early hunting poisons.

As a researcher in cognitive and Stone Age archaeology, I studied some of the artefacts from Umhlatuzana almost 20 years ago, finding use traces and adhesive residues on some of the quartz backed microliths (small, shaped stone tools) from 60,000 years ago. This showed that they were probably used as arrow tips.

Now, Sven Isaksson in the archaeology laboratory at Stockholm University has been able to identify molecular traces of toxic plant alkaloids (chemical substances), known to be an arrow poison, on a handful of these artefacts.

Left: Front and back of a 60,000-year-old poisoned arrow tip from Umhlatuzana. Middle: Two micrographs of the sharp, top edge that shows impact scars and the direction in which the arrow tip was fixed in the shaft. Right: A 2,000-year-old arrow head with stone tip fixed in the same way as the arrow tip from Umhlatuzana. Marlize Lombard, Author provided (no reuse)

Poison from indigenous plants

This latest research revealed the presence of buphandrine and epibuphanisine toxic alkaloids on five out of ten analysed arrow tips from Umhlatuzana. The same alkaloids were also found on bone arrowheads collected by Swedish travellers in the region 250 years ago. This tells us that the same arrow poison was used for many millennia in southern Africa.

Both alkaloids can be found in several southern African species of Amaryllidaceae, a family of flowering plants growing from bulbs. But only what is colloquially known as gifbol (poison bulb, Boophone disticha) is well-recorded as the source of an arrow poison. The plant’s bulb contains a toxic juice (exudate).

Finding these specific alkaloids on five out of the ten quartz arrow tips studied cannot be coincidental. Ancient hunter-gatherers would have been familiar with the toxic properties of the gifbol exudates. For example, by about 77,000 years ago, people of the same region also understood the insecticidal and larvicidal properties of some aromatic leaves that were used for bedding. So they probably would not have kept the gifbol substance in their living space.

Substances with buphandrine and epibuphanisine molecules are not used commercially or in archaeological conservation, ruling out accidental modern contamination of the arrow tips.

Gifbol bulbs can survive for a century or more, despite drought cycles and fire regimes. The plant is indigenous to South Africa, thriving in grassland, savanna and Karoo vegetation. It is widespread throughout the southern, eastern and northern regions of South Africa, growing within a day’s walk from Umhlatuzana Rock Shelter today. For various reasons, it’s likely that it was also available to the inhabitants of the site thousands of years ago.

The toxic chemicals in the bulb last a long time. They don’t decompose easily, even in wet environments, and they interact well with mineral surfaces like stone arrow tips. That’s probably why they survived for 60,000 years at Umhlatuzana.

Illustration of gifbol being prepared for use in hunting. Gemini generated image, Author provided (no reuse)

Implications of the world’s oldest known poisoned arrow tips

The quartz arrow tips with gifbol poison now represent the first direct evidence for hunting with poisoned arrows in southern Africa, and globally – at 60,000 years ago.

It demonstrates that these ancient bowhunters possessed a knowledge system enabling them to identify, extract and apply toxic plant exudates effectively. They must have also understood prey ecology and behaviour to know that the delayed effect of poison shot into an animal would weaken it after some time. That would make it easier to run down, a technique known as persistence hunting.

Such out-of-sight, long-distance action is a convincing indicator of complex cognition that requires response inhibition (being able to delay an action for a reason). Because poison is not a physical force, but functions chemically, the hunters must also have relied on advanced planning, abstraction and causal reasoning.

Thus, apart from providing the first direct evidence of hunting with poisoned arrows, the findings contribute to the understanding of human adaptation, techno-behavioural complexity and modern human behaviour in southern Africa.