# Disruption of an ant-plant mutualism shapes interactions between lions and their primary prey

> A tiny invasive ant is changing how lions hunt in Kenya. By killing the native ants that protect whistling-thorn trees, it lets elephants open up the savanna, leaving lions less cover to ambush zebra. Lions have not declined, but their kills have shifted towards buffalo.

Summary by Kersey Lawrence, Original Wisdom. Updated 7 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2024-kamaru-palmer-ants-lions-zebra/

## Citation and access

- **Authors:** Douglas N. Kamaru, Todd M. Palmer, Corinna Riginos, Adam T. Ford, Jayne Belnap, Robert M. Chira, John M. Githaiga, Benard C. Gituku, Brandon R. Hays, Cyrus M. Kavwele, Alfred K. Kibungei, Clayton T. Lamb, Nelly J. Maiyo, Patrick D. Milligan, Samuel Mutisya, Caroline C. Ng’weno, Michael Ogutu, Alejandro G. Pietrek, Brendon T. Wildt, Jacob R. Goheen
- **Year published:** 2024
- **Journal or source:** Science 383: 433–438
- **Full citation:** Kamaru, D.N., Palmer, T.M., Riginos, C., Ford, A.T., Belnap, J., Chira, R.M., Githaiga, J.M., Gituku, B.C., Hays, B.R., Kavwele, C.M., Kibungei, A.K., Lamb, C.T., Maiyo, N.J., Milligan, P.D., Mutisya, S., Ng’weno, C.C., Ogutu, M., Pietrek, A.G., Wildt, B.T., and Goheen, J.R. 2024. Disruption of an ant-plant mutualism shapes interactions between lions and their primary prey. Science 383: 433–438.
- **DOI:** 10.1126/science.adg1464
- **Access status:** Publisher link only
- **Where to find it:** https://doi.org/10.1126/science.adg1464

## Tags

- **Species:** Mammals > African buffalo; Mammals > African elephant; Invertebrates > Big-headed ant; Mammals > Lion; Mammals > Plains zebra
- **Region:** Africa > Kenya
- **Tracking:** Kill sites; Tracking; Tracks and signs identification
- **Behaviour and ecology:** Behaviour; Feeding; Habitat selection; Hunting; Predation; Predation risk
- **Methods and technology:** Akaike Information Criterion (AIC); Analysis of variance (ANOVA); Global Positioning System (GPS) collars; GPS cluster analysis; Path analysis; Resource selection function; Utilization distribution
- **Data type:** Existing data; New field data
- **Observer qualification:** Not reported
- **Discipline:** Behaviour and ecology; Wildlife tracking
- **Year published:** 2024

## Detailed keywords

- **Species:** Studied: lion (Panthera leo), six GPS-collared lionesses from distinct prides; plains zebra (Equus quagga), their main prey; African buffalo (Syncerus caffer), the second most commonly killed wild ungulate; African elephant (Loxodonta africana), as the browser opening up the trees; invasive big-headed ant (Pheidole megacephala) and native acacia ants (Crematogaster spp.); whistling-thorn tree (Vachellia drepanolobium). Other megaherbivores excluded by fencing: giraffe (Giraffa camelopardalis), black rhinoceros (Diceros bicornis) and white rhinoceros (Ceratotherium simum, printed “Cerototherum”).
- **Location and habitat:** Ol Pejeta Conservancy, Laikipia, Kenya: whistling-thorn tree savanna on a 364-km² landscape with big-headed ant invasion fronts advancing about 50 m a year; glades from old livestock corrals, water sources and human settlements. A fenced reserve.
- **Topics:** invasive species; ant-plant mutualism; foundation species; elephant browsing and tree cover; visibility and hiding cover for predators; lion hunting success; spatial distribution of zebra kills; predation risk and landscapes of fear; zebra density; prey switching from zebra to buffalo; lion population stability
- **Statistics or models:** Experimental plots in three blocks (megaherbivore exclusion versus open, invaded versus uninvaded) analysed with two-way analysis of variance (ANOVA); resource selection functions for time-varying zebra density surfaces; telemetry-derived utilization distributions for lion activity; GPS cluster algorithm to find kills; 21 nested path models tested with d-separation, Fisher’s C and small-sample Akaike Information Criterion (AICc); linear regressions of kill proportions on whistling-thorn cover and year.
- **Data type:** New field data: a three-year visibility experiment (from 2017), GPS collars on six lionesses, and 55 zebra kill sites found from GPS clusters and confirmed in the field within 5 to 10 days. Existing data: kills found opportunistically by anti-poaching patrols at Ol Pejeta from 2003 to 2020, lion counts over 13 years, zebra and buffalo densities from 2014 to 2020, and whistling-thorn cover in 2000, 2005, 2010, 2016 and 2020. Inconsistency in the paper: of the 21 nested path models, 14 were viable and three (17, 19 and 9) were best supported, which would leave 11 other viable models, but the text refers to “the 10 remaining viable nested path models”.

## Synopsis

### 1. What they did

In Laikipia, Kenya, whistling-thorn trees cover huge areas, protected from elephants by native acacia ants that live in their swollen thorns. An invasive big-headed ant is killing off these acacia ants. The authors asked how this affects lions and their main prey, plains zebra, at Ol Pejeta Conservancy.

They measured visibility for three years in experimental plots with and without big-headed ants, fenced or open to elephants and other megaherbivores. They GPS-collared six lionesses from different prides, found 55 zebra kills by visiting clusters of GPS positions, mapped zebra density and lion activity, and used path analysis to test which links explained where zebra were killed. They also examined 18 years of kill records from anti-poaching patrols.

### 2. What they found

- Where big-headed ants had invaded and elephants could browse, visibility after three years was 2.67 times higher (48.2 m versus 18.1 m), because elephants broke and killed the undefended trees.
- Across the landscape, invaded locations had 13.45 m greater visibility than uninvaded ones.
- Zebra kills were less likely where visibility was higher: at typical visibility, the probability of a zebra kill was 2.87 times higher in uninvaded than in invaded savanna.
- Higher visibility did not change zebra density or lion activity; it reduced kills by removing the cover lions use to ambush zebra.
- Lion activity was similar in invaded and uninvaded areas, and lion numbers (about 53 adults and subadults) stayed stable for 13 years.
- From 2003 to 2020, zebra fell from 67% to 42% of lion kills, while buffalo rose from 0% to 42%, as whistling-thorn cover declined. Zebra and buffalo densities showed no directional change from 2014 to 2020.

### 3. What they concluded

By breaking the ant-tree partnership, a tiny invasive ant lets elephants open up the savanna, which makes zebra harder for lions to catch. Lions appear to have compensated so far by switching to buffalo, more dangerous prey. Disrupted mutualisms may be an overlooked driver of predator-prey relationships, and whether lions can keep adjusting as the ants spread remains an open question.

### 4. How tracking was or was not used

Kill sites were found by visiting clusters of GPS positions from the collared lions, and the prey species at each was confirmed in the field within 5 to 10 days. Visibility and the presence of big-headed ants were recorded at each kill site. Long-term kill records came from kills found by anti-poaching patrols. No trailing is described.

### 5. Tracker or observer certification

Not reported. The paper does not say who identified prey at kill sites or how their skill was checked.

### 6. Other technologies, skills or innovations

- Electrified megaherbivore exclusion plots.
- Path analysis with d-separation tests to compare competing explanations.
- Data and code are publicly available on Dryad.

### 7. Implications for conservation (from the paper)

- One of the world’s worst invasive species can reshape a savanna and its predator-prey relationships, with effects that spread across whole landscapes. Prey switching by lions to buffalo may hold numbers steady for now, but the future is uncertain as the ants advance.

### 8. Implications for tracking (Original Wisdom's observations)

_These are Original Wisdom's observations, not the authors'._

- The study rests on kill-site investigation, a core tracking skill: identifying prey remains at GPS clusters. Reading cover and visibility at a kill adds to understanding how a hunt succeeded.

## Cite this summary

Lawrence, K. 2026. Summary of: Disruption of an ant-plant mutualism shapes interactions between lions and their primary prey (Kamaru and Palmer et al., 2024). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2024-kamaru-palmer-ants-lions-zebra/