# Baboon Sleeping Site Preferences and Relationships to Primate Grouping Patterns

> Where does a baboon troop choose to spend the night? Drawing on troops across Africa, including his own in Botswana’s Okavango and Savuti, William Hamilton found a clear order: cliffs and caves first, then tall, strongly branched trees, with open woodland a last resort. Safety from night-hunting leopards seems to drive the choice, and the author suggests that where safe sleeping sites are scarce, primates may live in larger groups.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/1982-iii-baboon-sleeping-site-preferences/

## Citation and access

- **Authors:** William J. Hamilton III
- **Year published:** 1982
- **Journal or source:** American Journal of Primatology
- **Full citation:** Hamilton, W. J. III. 1982. Baboon sleeping site preferences and relationships to primate grouping patterns. American Journal of Primatology 3(1): 41-53.
- **DOI:** 10.1002/ajp.1350030104
- **Access status:** Publisher link only
- **Where to find it:** https://doi.org/10.1002/ajp.1350030104

## Tags

- **Species:** Mammals > Chacma baboon; Mammals > Gelada; Mammals > Guinea baboon; Mammals > Hamadryas baboon; Mammals > Olive baboon; Mammals > Yellow baboon
- **Region:** Africa > Botswana; Africa > Ethiopia; Africa > Kenya; Africa > Namibia; Africa > Senegal; Africa > South Africa; Africa > Tanzania; Africa > Uganda
- **Behaviour and ecology:** Behaviour; Feeding; Habitat selection; Predation risk; Predator avoidance
- **Methods and technology:** Direct observation; Literature review
- **Data type:** Existing data; New field data; Review
- **Observer qualification:** Not needed
- **Discipline:** Behaviour and ecology; Palaeoanthropology
- **Year published:** 1982

## Detailed keywords

- **Species:** Studied: baboons (Papio spp.): chacma baboon (Papio ursinus), Papio cynocephalus, Guinea baboon (Papio papio), hamadryas baboon (Papio hamadryas), Papio anubis, and gelada baboon (Theropithecus gelada). Predators discussed: leopard (seen chasing baboons at a roost, and attacking troops at night), spotted hyena, cats in general; humans as observers causing disturbance. Comparison species: patas monkey (Erythrocebus patas), chimpanzee (Pan troglodytes) and red colobus monkey (Colobus badius), macaques and langurs, and early and recent humans (hominids, Bushmen and other hunter-gatherers). Food: scale insects on mopane. Sleeping trees: Acacia nigrescens, the palm Hyphaene ventricosa, mopane (written in the paper as “Mopane calophospermum”) and dead acacia snags.
- **Location and habitat:** Troops summarised from across Africa: South Africa (northern Transvaal, Natal, Cape Point, Kruger Park), Botswana (Okavango Swamp, Savuti Marsh and the flooded Savuti Channel, Serondella), Namibia (the Namib, open cliffs), Kenya (Amboseli, Gilgil), Tanzania (Mikumi, Gombe), Senegal, Ethiopia and Uganda. Habitats: cliffs and caves, emerging trees in continuous forest, closed canopy forest, open woodland, riparian forest, mopane forest, palm groves, tropical forest and savanna.
- **Topics:** sleeping sites (nocturnal roosts); preference order of sleeping site types (caves and cliffs, emerging trees, closed canopy, open woodland); site categories defined (caves over 5 m deep, open cliffs within 20° of vertical, wooded cliffs, emerging trees, closed canopy, open woodland, rock overhangs no deeper than 5 m); predation avoidance and escape by lateral movement in the canopy; nocturnal leopard predation; displacement between troops at sleeping sites; use of less preferred roosts when foraging far away (mopane scale insect outbreak); passive flight versus active defence; caves and defensibility; hominid rock shelters and human evolution; group size and primate social structure; distribution of sleeping sites relative to food resources. Inconsistencies in the paper: (1) the text says caves were present and chosen equally with cliffs at four sites, but Table I marks caves as a preferred site for only two troops (northern Transvaal, Marais 1939, and Cape Point S troop); (2) the text describes the mopane forest roost as closed canopy and refers to Fig. 2, but the mopane forest is shown in Fig. 3 (Fig. 2 shows the emerging Acacia nigrescens roost); (3) Hall is cited as 1962a in Table I and 1962 in the text and references; Stoltz and Saayman (table and text) appear as Stolz and Saayman in the references.
- **Statistics or models:** Counts of troop choices where two site types were both available, with P values (P < 0.001) for cliffs over emerging trees (12 to 0), emerging trees over closed canopy (23 to 0) and emerging trees over open woodland (22 to 0). The statistical test is not named. Troops are the unit of analysis.
- **Data type:** Review of published and unpublished (personal communication) observations, plus the author’s own field observations in Botswana (Okavango, 1972 to 1980; Savuti, 1980) and Namibia. The observations were not collected systematically to study sleeping site choice; the author presents the results as hypotheses.

## Synopsis

### 1. What they did

Where an animal spends the night can matter as much for survival as where they feed. The author asked whether baboons pick their sleeping sites to stay safe from predators, and whether the way good sleeping sites are spread across a landscape could help explain why some primates live in large groups and others in small ones.

He grouped baboon sleeping sites into categories (Fig. 1): caves (over 5 m deep), open cliffs (rock faces within 20° of vertical), wooded cliffs, emerging trees that rise a little above the forest and have strong branches for moving sideways, closed canopy that baboons can cross without coming to the ground, open woodland where they cannot reach the next tree without coming down, and rock overhangs (no deeper than 5 m).

He then gathered descriptions of sleeping sites for many baboon troops across Africa from the published literature, from colleagues, and from his own observations in Botswana (1972 to 1980) and Namibia (Table I). For each troop he recorded which site types were available and which the troop used most. The observations were not originally collected to study sleeping sites, so he presents the findings as hypotheses.

### 2. What they found

The preference order

- Where caves were present (a small number of sites), they were chosen about as often as cliff faces.

- Where open cliffs were present but caves were not, cliffs were preferred over emerging trees (12 troops to none).

- Emerging trees were preferred over closed canopy (23 to none) and over open woodland (22 to none) when both were available.

- The only exception was F troop in Botswana, which chose scattered palm trees over shorter closed canopy.

- No baboons were reported to use wooded cliffs or rock overhangs for sleeping, and free-ranging baboons have never been reported sleeping on the ground.

Details from the field

- At one mixed site in the Okavango, independent two to three year old juveniles mostly slept in isolated palms, while adults and mothers with infants used the emerging trees and palms that touched each other.

- At Savuti Marsh, one troop slept in dead acacia snags standing in the flooded Savuti Channel, a shallow river about 20 m wide. Some baboons swam out to them and others crossed on fallen logs. Leopards were seen chasing baboons arriving at this roost.

- After night attacks by leopards, troops were seen changing roosts.

- Many sleeping sites were open to wind and weather, such as leafless treetops and bare cliffs.

- When disturbed at night in trees, baboons moved sideways into the next trees without coming down. On open cliffs in Namibia, they moved less when disturbed by spotted hyenas and people, gave alarm calls and released stones, and never abandoned the cliffs at night.

- When Okavango Swamp troops met at sleeping sites (N = 18 in the paper), the losing troop always moved from an emerging Acacia nigrescens roost to a closed canopy roost.

- During an outbreak of scale insects (a favoured food) in mopane forest, one troop slept seven times in the mopane forest, a closed canopy site, when this food took up over 50% of their foraging time. On 42 other days of feeding there, they returned to their two preferred emerging tree roosts. They did not use closed canopy roosts when the insects were not available.

### 3. What they concluded

- Baboons seem to choose sleeping sites mainly for safety from predators, especially leopards at night. The preferred sites let them escape upward or sideways where predators cannot easily follow.

- Caves may be favoured for a different reason: a few individuals could block the entrance against predators. This remains speculative, because no one has reported how baboons respond when predators enter their caves.

- Rock overhangs, used by Bushmen and other hunter-gatherers, are not used by baboons, probably because they can only be defended actively. When early humans first used such shelters, they may have been able to defend themselves together, perhaps linked to a shift towards group hunting.

- Where good sleeping sites are few and far apart relative to food, groups may gather at them, which may allow large groups to develop. In tropical forests, where suitable sleeping trees are many and evenly spread, small groups become possible.

- Sleeping sites are a constraint that makes some grouping patterns possible, not a cause of social structure.

- Baboons use less suitable sites when they forage far from good ones, so poorer feeding conditions should mean more use of poorer sites. Vulnerability to leopards should be highest at the least preferred sites, but no data were available to test this.

### 6. Other technologies, skills or innovations

- A simple classification of sleeping site types by their structure (Fig. 1), which can be applied across habitats and species.

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

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

- The preference order (cliffs and caves first, then taller emerging trees with strong side branches) gives trackers a guide to where baboon roosts are most likely in a landscape, and where leopards may hunt baboons at night.

## Cite this summary

Lawrence, K. 2026. Summary of: Baboon Sleeping Site Preferences and Relationships to Primate Grouping Patterns (III, 1982). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/1982-iii-baboon-sleeping-site-preferences/