The puzzle of the pepperbark tree

Is the pepperbark doomed to die out in the Kruger National Park? Pepperbark detectives KAYLEE VAN DEN BOSCH, GLYNIS CRON, DAVE THOMPSON & ED WITKOWSKI went searching for answers to the mystery of why it is failing to reproduce

The puzzle of the pepperbark tree 

Is the pepperbark doomed to die out in the Kruger National Park? Pepperbark detectives KAYLEE VAN DEN BOSCH, GLYNIS CRON, DAVE THOMPSON & ED WITKOWSKI went searching for answers to the mystery of why it is failing to reproduce

Glynis Cron and Kaylee van den Bosch carry out hand pollination on some pepperbark trees in their Kruger Park research population. Photo by Mike Goodman


The dusky orange sun was rising over the bushveld as we drove out of our camp in the Kruger National Park on our way to begin monitoring its pepperbark trees (Warburgia salutaris).

After a while, we had to park and continue on foot – trying to match the energetic pace of our field guides.

It was May 2019 and we were excited to be investigating the mystery of why pepperbark trees seem to fail to reproduce in the park. It would take us two challenging years – including Covid interfering with planned field trips and many other bumps along the way.


The pepperbark is listed as endangered on the IUCN Red List and was recognised as being of conservation concern in South Africa almost a century ago in 1926. Its numbers have declined both because of habitat loss and because of unsustainable harvesting of the bark, which is highly valued in southern Africa for use in traditional medicine.

Various conservation initiatives have aimed to combat the species’ decreasing numbers. Such initiatives have particularly included cultivation and distribution of pepperbark seedlings to nurseries, traditional healers and communities to try to reduce wild harvesting.

As well as being endangered, surprisingly little is known about the pepperbark’s reproductive ecology – how its reproduction is affected by ecological variables.


The Kruger Park supports one of the largest protected populations of pepperbark trees in South Africa – but even so, there are concerns over the species’ future in the park.

Curiously, you will not see young pepperbark seedlings in the Kruger Park, unlike in other smaller populations of pepperbark elsewhere in South Africa. Ultimately, very few of the seeds that begin to develop reach maturity.

Left: Glynis Cron tries to spot flowers in a Kruger Park pepperbark tree – evidence of flowering would prove much more elusive than the researchers expected. Photo by Mike Goodman


With little known about the intricacies of the pepperbark’s reproductive ecology, our challenge and goal was to try to understand why these trees were reproducing so unsuccessfully.

We aimed to track the pepperbark’s reproduction – from flowering, pollen production and transfer to fruit production and seed germination – so we could use this knowledge to understand why so few pepperbark fruit are produced in the protected Kruger Park population.



In South Africa, the pepperbark’s habitats range widely – from humid ravines and coastal forests in KwaZulu-Natal through eastern Mpumalanga up to drier and hotter areas in the northern Kruger Park.


The bark, roots and leaves of this species contain complex peppery chemical compounds that have antibacterial properties. As a result, these parts of the tree are used to treat a wide range of ailments, including colds, inflammation, fever, headaches, skin irritations, sores and ulcers.


The pepperbark has various names across South Africa including: 

* isibhaha (isiZulu)

* shibaha (Tsonga)

* manaka (Venda)

* molaka (Sotho)

* muranga (Shona)


The tree’s medicinal properties are recognised in its specific name, salutaris, meaning ‘health-giving’.


The pepperbark tree is also found in Namibia, Botswana, Zimbabwe and Malawi and extends into east African countries such as Kenya and Tanzania.

A day-flying moth (Amata simplex, left) feeds on pepperbark nectar at the Kruger National Park and a night-time moth (Dysgonia angularis, right) feeds on nectar during the evening at the Eston pepperbark orchard. Photos by Kaylee van den Bosch



As we searched the groups of pepperbark trees in the Kruger Park, we were surprised to see that most trees did not have any flowers, even though this was their prime flowering season.

Trees were in flower in only six of the groups of pepperbark trees. From these, we managed to identify only 42 trees in flower. The flowers themselves are inconspicuous. They are small, about five to seven millimetres in diameter, and a dull green colour.


The pepperbark’s flowers are visited by a range of insects during the day, including wasps, beetles and butterflies. We suspected that the flowers could also have night-time visitors.

To investigate which insects visit the pepperbark trees after we had to be back in camp with the gates closed, we set up camera traps. The footage revealed that after sunset, moths visit the flowers.

We were later able to confirm this with our own evening observations of cultivated pepperbark trees on Malcolm and Les Stainbank’s farm near Eston in KwaZulu-Natal. It was also interesting to see how prolific and healthy fruit production and seedling establishment was in these trees.


The outcome at the Kruger Park, by contrast, underperformed even our low expectations of how many fruit the pepperbark trees might produce naturally. By December, we had to recognise that the natural fruit set was so low that only 3% of tagged flowers were developing into mature fruit.

We even tried hand pollinating some of the flowers but still did not succeed in producing any mature fruit from these naturally growing pepperbarks. However, when we tried hand pollinating some cultivated pepperbarks in the plantation at Eston – something which is not commonly done – this method was successful.


Among the Kruger Park pepperbark trees, a number of flowers were aborted shortly after fertilisation. This indicated that pollination had succeeded but that there was some barrier to flowers or fruit developing after fertilisation. Even those flowers that had been naturally pollinated by insects and were protected from potential seed predators (such as wasps and flies) by gauze bags had aborted their fruit during development.

When we compared this to our study of the cultivated pepperbark population at Eston, we spotted two illuminating points:

1. Cross-pollinating the trees could and did result in mature fruit; but
2. No fruit was produced after flowers were self-pollinated by hand.

We used camera traps to observe which insects visit the pepperbark trees at night – this trap was set up on the pepperbark plantation on Malcolm and Les Stainbank’s farm at Eston, KZN. Photo by Kaylee van den Bosch


Our next step was to assess the viability of the pollen and the growth of pollen tubes in both the Kruger Park and Eston pepperbark populations. To our surprise, there was little difference between the Kruger Park and Eston trees. Pollen viability was similarly high in both. All three pollination treatments – self-, cross- and natural-open pollination – successfully produced pollen tubes.

We had not yet found what was inhibiting fertilisation and fruit set in the Kruger Park trees. We had to keep searching.

Kaylee van den Bosch observes the Eston pepperbark plantation – the advantages and disadvantages of naturally growing versus cultivated pepperbark were not always clear cut. Photo by Malcolm Stainbank


In the early days of our study in 2019, we had pinpointed just 42 pepperbark trees in the Kruger Park that produced any flowers. We had tagged them but less than half (45%) of these trees had a few immature fruit developing on them.

One group was clearly the most productive. Seven out of its 10 flowering trees were producing young fruit through natural, open pollination. Yet when we went to monitor fruiting in December 2019, fruit had matured on only two of these trees.

But there was still information that we could glean from this meagre harvest.


We collected the fruit to assess production of seeds, to what extent insect larvae had preyed on the seeds and whether these seeds would germinate.

Left: Among the mysteries that we needed to solve was why fruit set in pepperbark trees was low and what factors affected germination. Photo by Kaylee van den Bosch

We ran germination trials comparing the Kruger Park seeds to others from the cultivated pepperbark trees in Eston and from more pepperbarks in Leshiba Nature Reserve in the Soutpansberg. We were excited to find that the Kruger Park seeds were viable – and, surprisingly, had the highest percentage germination (61%) of the three populations.

Careful nurturing produced the first ‘wild’ seedlings from the Kruger Park in many decades. The seedlings were delivered to delighted staff at the Skukuza Nursery for further cultivation.


We still needed to solve the mystery of what dispersed the pepperbark’s fruit. Pete and Kathryn Haneko allowed us to set up camera traps at a small natural population in the Leshiba Nature Reserve. The footage showed us that baboons and vervet monkeys both eat pepperbark fruit. These primates perform the vital task of removing the fruit’s pulp and seeds from its hard shell. They suck out the pulpy contents, which are very much like that found in a grenadilla.

The seeds do not have a hard seed coat so they would most probably not survive the journey through the baboon’s gut. But it seems very likely that the primates spit out some seeds before swallowing the pulp, suggesting that they are important agents of seed dispersal.


Having followed the pepperbark trees from flowering and fruiting through to seedlings, we had established three crucial points:

1. Why so few fruit: The pepperbark relies on pollinators to carry pollen between trees. Flowers cannot be fertilised by pollen from flowers on the same tree (self-incompatible). With often only a few individuals in some groups of Kruger Park pepperbark trees flowering, there is limited potential for them to cross-pollinate successfully with genetically different individuals.

Right: Kruger Park pepperbark trees produced relatively few fruits and seeds but had the highest germination rate in the three groups monitored – these seedlings went to the Skukuza Nursery to be grown on. Photo by Kaylee van den Bosch

2. Why so few seeds: The abortion of many of the young fruit showed that there may be some mechanism preventing fruit development after fertilisation. In semi-arid areas such as the northern Kruger Park where the pepperbark trees grow, water availability is a major factor limiting plant growth and sexual reproduction.

This might also contribute to the high rate of fruit abortion. So even though fruiting coincides with the rainy season, rainfall is often erratic – a few high-volume rainfall events interspersed with a few low-volume showers and periods of intra-seasonal drought. Under conditions where resources are limited, the early abortion of fruit would allow the trees to allocate their resources to storage organs and vegetative growth.

3. Too many clones: Another important piece of evidence in unravelling this puzzle is that root suckering has been seen in the Kruger Park pepperbark trees. Recent genetic analyses have shown that about 25% of the total population in the Kruger Park are clones. Fruit abortion could also be a result of self-pollination between these clones. This vegetative reproduction helps the species to persist but a population of genetically diverse individuals that can adapt to environmental changes through sexual reproduction is vital for long-term survival.

We are still left with more intriguing questions to be answered such as understanding the impact of the age of these trees on their reproductive capacity, monitoring how flowering and fruiting change over the years and the impact of rainfall patterns. We hope that future studies will be carried out to answer these questions, so that we can continue to resolve the pepperbark puzzle.

Kaylee van der Bosch (kayleerobyn@gmail.com) has an MSc in environmental science and a strong interest in sustainability and the preservation of natural systems. She is currently a climate-change consultant, working on a range of projects including carbon footprints, climate-change strategy and carbon-credit projects.

Glynis Cron is an associate professor at the University of the Witwatersrand whose interests include plant systematics, reproductive ecology, pollination, and conservation.

Dr Dave Thompson is the biodiversity scientist at the SAEON Ndlovu node, with interests in plant population dynamics and community-level response to disturbance in grassy ecosystems. His research has also expanded into pollination ecology, population genetics, river flow, crop phenology, mammal behaviour and sustainable resource use.

Ed Witkowski is a professor in plant ecology at University of the Witwatersrand with interests in conservation, especially threatened plant species, and the ecology and management of African savanna trees.

We thank the NRF, the Botanical Education Trust and SAEON for funding and SANParks for permission to collect pollinator and plant material.

Further reading

A. Maroyi, ‘The genus Warburgia: A review of its traditional uses and pharmacology’, Pharmaceutical Biology, 52(3), pp.378–391, (2014)

K. Glennon, J.J. le Roux & D. Thompson, ‘Genetic insights into pepper‑bark tree (Warburgia salutaris) reproduction in South Africa’, Conservation Genetics, 24, pp.883-891 (2023):

K. van den Bosch, E.T.F. Witkowski, D. Thompson & G.V. Cron, ‘Reproductive ecology offers some answers to the pepperbark tree persistence puzzle in the Kruger National Park, South Africa’, Global Ecology and Conservation, 41: e02330


Submit a Comment

Your email address will not be published. Required fields are marked *

Spring Flower Watch: Ingwehumbe Nature Reserve
A new field guide for Overberg Renosterveld
Spring Flower Watch: Harmony Flats Nature Reserve
Spring Flower Watch: Tienie Versfeld Nature Reserve
1 2 19


Open Monday to Friday 10h00 to 14h00. Closed on weekends and public holidays.

December holidays:  Office will be closed from the 25th December and reopen 1st working day of the new year

Contact Us

We are experiencing intermittent faults with our landlines. If you can't get through on our landline +27 (0) 21 797 2090 , please phone or send a message to our alternate WhatsApp number: +27 (0) 65 922 6163.







Pin It on Pinterest

Share This