Why on earth?: Evaluating hypotheses about the physiological functions of human geophagy

Geophagy has been hypothesized to be an adaptive behavior, either as a means to allay nutrient deficiency or to protect against ingested pathogens and toxins. Others have proposed that geophagy is non-adaptive, occurring either to allay hunger or as an epiphenomenon of nutrient deficiencies. This paper evaluates these hypotheses using 482 published cultural-level accounts of human geophagy and 330 accounts of geophagy among 297 species of mammals, birds, and reptiles. Information was extracted from reports of human geophagy to permit statistical analysis; reports of non-human geophagy were tabulated. Human geophagy did not parallel changes in nutrient requirements, occurred most frequently among children and pregnant women and in tropical areas (where pathogen densities are highest), and was associated with ingestion of toxic substances and gastrointestinal distress. Earth ingested by humans was craved and carefully selected and prepared; it had high clay content, but few bioavailable mineral nutrients. In primates, geophagy was associated with both protection from toxins and obtaining nutrients, whereas in other vertebrates it was associated mainly with obtaining nutrients. Our results indicate that human geophagy is best explained as providing protection from dietary chemicals, parasites, and pathogens, whereas animal geophagy may involve both micronutrient acquisition and protection.     

Differentiation in mineral constituents in elephant selected versus unselected water and soil resources at Central African bais (forest clearings)

Natural forest clearings (bais) in the Central African rain forest attract large numbers of mammals. Little is known about the factors influencing bai use by forest species, though geophagy and hydro-mineral resources are assumed to be important attractants. In the present study, clay and mineral concentrations in water and soil were examined at 15 bais. Water samples from elephant excavated pits showed significantly higher concentrations of most minerals sampled relative to surface waters. But mineral portfolios varied markedly between bais. Geophagy sites were less differentiated from control soil samples, leading to the interpretation that geophagy may not structure bai visitation. Monthly sampling of pit water at one bai suggested higher dry season mineral concentrations, which may relate to seasonal wildlife visitation patterns. The complexity and variability in bai-specific mineral resources suggest there is not a single determining factor (or mineral) driving bai use. The protection of bai mosaics should be a conservation priority in order to ensure access to the portfolio of minerals likely required by endangered species such as the African forest elephant Loxodonta africana cyclotis.     

Geophagy in new world monkeys (Platyrrhini): ecological and geographic patterns

Geophagy has been recorded in an increasing number of New World monkeys (Platyrrhini) over recent years, permitting a tentative analysis of ecological patterns. While geophagy has now been recorded in species representing all 4 platyrrhine families and a majority of genera, there is a marked tendency for it to occur in the larger-bodied Pitheciidae and Atelidae. Howlers (Alouatta) are responsible for almost a third of reports, which are concentrated in the more frugivorous species, Alouatta belzebul and Alouatta seniculus. Geophagy may also be relatively common in the spider monkeys (Ateles) and the pitheciids, which are specialised frugivores and seed predators, respectively. An overview of the available data points to a marked Amazonian bias, allowing for geographical differences in the number of species and field studies. This pattern is demonstrated most emphatically by Alouatta, for which there are almost as many reports as field studies in the Amazon basin, in stark contrast with Central American sites, which have a long tradition of fieldwork, but no published records of geophagy. There are also relatively few reports from the Brazilian Atlantic forest. Despite the growth in reports, and the patterns identified here, the functional aspects of geophagy in the platyrrhines still remain unclear.     

Geophagy by the Indian short-nosed fruit bat, <Emphasis Type="Italic">Cynopterus sphinx</Emphasis> (Pteropodidae) while foraging on <Emphasis Type="Italic">Madhuca latifolia</Emphasis> (Sapotaceae) in Tamil Nadu,South India

Geophagy has been reported in many mammals and birds. However, in pteropodid bats, this behaviour has not been reported. Here, we document geophagy behaviour by the pteropodid bat, Cynopterus sphinx, during foraging in the fruiting trees of Madhuca latifolia (Sapotaceae). These bats removed single fruits from the foraging tree and carried it to the feeding roost for consumption. At around 1900 h, we observed three individuals of C. sphinx gleaning over the ground surface by circling flights. The bats landed with their wings spread apart and started licking the soil with head movements. Their ventral body surface rested completely on the ground while consuming the soil. A total of 91 feeding bouts of soil consumption were recorded within 1 h; however, thereafter, this behaviour was not observed. In order to understand the function of geophagy, we quantitatively analysed the composition of minerals in the soil, as well as in the M. latifolia fruits (partially ripe) along with its secondary metabolites from the study site. Our results revealed that the partially ripe fruits contained high levels of secondary metabolites (alkaloids, tannins, coumarins and saponins). The soil contained higher levels of Mg, Ca, Fe, Na and K minerals than the fruits. These findings suggest that bats may gain essential minerals that are low in their fruit diet by consuming soil. In addition, the observed minerals that play a role in detoxification of the secondary metabolites present in the partially ripe fruits that might be detrimental to the consuming bats. Frugivory, nectarivory and folivory are well understood in C. sphinx; however, geophagy behaviour has not been reported in this species so far. This may represent an ‘adaptive behavioural plasticity’ in the foraging behaviour of the observed C. sphinx population.     

The Adaptive Significance of Geophagy for Japanese Macaques (Macaca fuscata) at Arashiyama,Japan

We gathered data on the amount, composition, and rate of ingestion of foods and soils by the provisioned Japanese macaques (Macaca fuscata fuscata) at Arashiyama, Japan. Behavioral observations spanned one year on 8 adult females, using focal animal sampling. We analyzed a subsample of their foods for nutritional and toxic secondary compound content. We also analyzed soils eaten by the macaques for several physical-chemical properties and tested their adsorption affinity to tannins and alkaloids. Geophagy occurred at a high rate of 2.97 g/indiv./day with an elevated frequency in the afternoon. About two-thirds of their foods (by fresh weight) were provisioned items, which are extremely rich in proteins and soluble carbohydrates. The soils that they ingested were generally poor in mineral elements, the bio-availability of which was low. The soils had a high adsorption capacity for plant alkaloids but were poorly absorptive for tannins. They were rich in clay minerals of proven buffering capacity. Geophagy at Arashiyama may improve the health of macaques via buffering gastric upset. We discuss the results from the viewpoint of several hypotheses on geophagy.     

土壤pH值对秦岭川金丝猴食土行为的影响

食土行为广泛存在于灵长类动物中,关于灵长类的食土行为有很多功能性的假说,包括补充矿物质、有毒物质吸附、抗腹泻、抗体内寄生虫感染及肠胃酸性中和等假说。为了解秦岭川金丝猴取食黏土的影响因子和功能假说,于2013年9月至2015年1月采用扫描观察法和瞬时记录法收集食性数据,记录取食黏土个体的性别年龄及取食位置,并采集秦岭川金丝猴取食的黏土和对照黏土样品。于2016年07月采用电位法测定土壤样品的pH值。食性结果表明不同性别年龄个体,从1岁的婴幼猴到成年个体都存在取食黏土的行为;秦岭川金丝猴取食黏土的pH显著高于对照样品的pH值。因此,本研究结果与肠胃酸性中和假说更为契合,秦岭川金丝猴取食偏碱性的黏土,有利于中和取食大量芽和叶摄入的酸性物质。     

Assessing the function of geophagy in a Malagasy rain forest lemur

Geophagy has been observed in nearly every long‐term study of folivorous primates. Yet despite frequent observations of this behavior, conclusive explanations for soil consumption remain ambiguous. This study tests the most frequently proposed hypotheses for geophagy using data collected on the geophagic behavior of the Milne‐Edwards’ sifaka (Propithecus edwardsi) living in two forests with varying levels of disturbance within Ranomafana National Park, Madagascar. Field data were collected from December 2002 to November 2003, during which time soil samples were collected for mineral analyses from 102 sites selected for geophagy and 42 control sites at which no geophagy was recorded. Results indicate that control soils differ significantly between the two study sites and that this difference is primarily attributable to varying levels of habitat disturbance. The frequency of soil consumption also does not vary significantly by sex or between logged and unlogged habitats. Soil consumption significantly correlated with fruit/seed consumption overall, but to a lesser degree in logged compared with unlogged sites. Clay minerals found in soils likely prevent absorption of dietary toxins in the gut and/or may mediate gut pH. This provides strong evidence for the protection hypothesis for geophagy, which may be especially important in areas where logging, or other forms of habitat disturbance, has been experienced. Abstract in French is available with online material.     

黄山短尾猴食土行为

从行为生态学角度,依据黄山短尾猴食土行为在年龄、性别、社会等级序位、食土量、食土频率、食土持续时间中的分布和变化规律,结合土壤基本理化性质测定和分析,探讨了黄山短尾猴食土行为。黄山短尾猴对土壤的摄取是寻找和有目的的选择,有些场所是其"喜好"或"常去之处",食土场所通常是以多个体多次取食挖掘而形成的洞穴形式存在。取食土壤颜色黄色或黄棕色,酸性土壤,富含铁、钙、镁等矿物元素,粘土比例较高。社群各年龄段及性别个体均参与食土行为。平均食土行为持续时间在年龄和性别之间无显著差异;平均食土频率和食土量在成年雌性或雄性个体的社会等级序位之间无显著性差异;而成年雌性的平均食土频率和食土量,显著高于其它年龄或性别组,与雌性正值怀孕末期及产仔哺乳期,需要补充大量铁等矿物元素,以维持生理所需和体力消耗有关。典型的植食性动物黄山短尾猴有规律地取食粘土的行为,支持了食土行为具有"食物解毒作用"假说。人工投喂和食土行为的相关性表现为提高了食土频率,可能与粘土能够缓解高热量、低纤维人工食物造成的胃肠不适有关。     

Geophagy at termitaria by bearded sakis (Chiropotes satanas) in Southeastern Brazilian Amazonia

Geophagy is rare in primates, but has been recorded in a variety of platyrrhine genera. The first observation of geophagy in bearded sakis (Chiropotes satanas) is reported here. Two bouts were recorded during long-term monitoring of a free-ranging group inhabiting a small island in a reservoir in eastern Brazilian Amazonia. On both occasions, soil was obtained systematically from a termitarium in the crown of an Eschweilera coriacea tree. Both events occurred during the late dry season. Although there is no evidence of any significant shift in foraging behavior during this period, it is possible that soil ingested provided an essential mineral supplement to a diet limited by the small size of the island.     

Geophagy in masked titi monkeys (Callicebus personatus melanochir) in sBrazil

Geophagy was observed 14 times during an 11-month field study on the ecology and behavior of masked titi monkeys (Callicebus personatus melanochir). Geophagy took place in 12 cases on the surface of a leaf-cutting ant mound (Atta spec.). Analysis of chemical composition and natural features of the soils consumed by the monkeys showed elevated concentrations of several elements. In contrast, a lower concentration of carbohydrates was found in the ant mound sample compared to random non-ant mound soil samples. No difference in the concentration of clay elements, pH-value or percentage of organic matter was found between ant mound samples and randomly collected soil samples. Comparison of used versus not used samples failed to reveal significant clues about function of geophagy. However, chemical and physical properties of soils ingested support previous hypothesis that this behavior may aid in the adsorbtion of plant toxins.     

Geophagy in chacma baboons: patterns of soil consumption by age class, sex, and reproductive state

Despite baboons' widespread distribution across Africa, geophagy among all subspecies has been poorly documented. We used video camera traps and soil analyses to investigate geophagy in chacma baboons (Papio cynocephalus ursinus) inhabiting the Western Cape of South Africa. During an 18‐month study, from August 2009 to January 2011, we continually monitored the largest and most frequently visited geophagy sites with camera traps for 545 days and captured soil consumption at one or more sites on 266 of those days (49%). In 3,500 baboon visits to geophagy sites, video camera traps captured 58.6 hr of geophagy. From these data, we evaluated site preference based on time spent consuming soil among these four geophagy sites. One hundred and seventy days of soil consumption data from the most frequently visited geophagy site allowed us to look for demographic trends in geophagy. Selected consumed soils from geophagy sites were analyzed for mineral, physical, and chemical properties. The baboons spent more time consuming white alkaline soils with high percentages of clay and fine silt, which contained higher concentrations of sodium than non‐white acidic soils that contained higher concentrations of iron. Our data indicate that pregnant chacma baboons spent more time consuming soil at monitored geophagy sites than baboons of any other age class, sex, or reproductive state. Based on analytical results, the soils consumed would be effective at alleviating gastrointestinal distress and possibly supplementing minerals for all age/sex classes, but potentially for different age/sex requirements. Am. J. Primatol. 74:48–57, 2012. © 2011 Wiley‐Liss, Inc.     

Observations of termitarium geophagy by Rylands’ bald-faced saki monkeys (<Emphasis Type="Italic">Pithecia rylandsi</Emphasis>) in Madre de Dios,Peru

Geophagy, or soil consumption, has been documented in diverse animal taxa, including many primates. Physiological functions such as mineral supplementation, detoxification of secondary compounds, and antacid properties are possible causes for this behavior. We report on observations of geophagy at arboreal termitaria by free-ranging Pithecia rylandsi at La Estación Biológica Los Amigos (EBLA) in Perú between 2008 and 2015. Characteristics of geophagy events, including saki monkey behavior at the termitaria, were recorded and geochemical analyses were conducted on consumed termitaria, nearby topsoils, and unvisited termitaria. We observed 76 feeding bouts at 26 different termitaria by two groups of P. rylandsi during 1125 observational hours (0.07 bouts/obs. h). Geophagy occurred throughout the year, but rates peaked in January during the rainy season. All age and sex classes visited both active and inactive mounds. Feeding bouts were brief (171 ± SD 154 s), and no differences were observed in time spent feeding at active or inactive termitaria. Analyses showed that consumed soils contained higher concentrations of phosphorous, potassium, calcium, and magnesium than did topsoil. Consumed soils also contained a higher total cation exchange capacity than topsoil. Our analysis of consumed versus control termitaria revealed no differences in their chemical composition. We discuss these results in the context of the two primary hypotheses proposed for geophagy in pitheciins: mineral supplementation and toxin adsorption. Our data are consistent with the interpretation that P. rylandsi consume soils from arboreal termitaria to aid in adsorption of toxins found in immature seeds, which are a year-round component of their diet.     

Geophagy amongst rhesus macaques on Cayo Santiago,Puerto Rico

Soil mining and eating (geophagy) behavior of rhesus macaques (Macaca mulatta) on Cayo Santiago, Puerto Rico, is described and assessed with respect to the chemical, geochemical, and mineralogical composition of the ingested materials. The samples forming the uneaten (control) and eaten (matrix and blocky) groups of soils come from the top and flanks of a marine terrace underlain with volcanic tuff on Cayo Santiago, off the east shore of Puerto Rico. Both the uneaten and geophagy samples were analyzed to determine particle size distributions, clay and primary mineralogy, and soil chemical and geochemical compositions. Primary minerals such as orthoclase and plagioclase feldspar in the clay fraction is higher in the control group than in the ingested samples. Both the control and matrix plus blocky samples have moderate to abundant amounts of kaolinite and halloysite (both silicon:aluminum = 1:1 type clay minerals) that may be important as a stimulus to geophagy behavior. The pH, total salts, and phosphorus levels in both the control and geophagy samples show considerable overlap with little clear indication of causal factors. Analysis of the geochemical data showed no clear cut elemental differences to suggest elemental supplementation as a possible explanation for mining and eating of tropical soil. It is possible that rhesus macaques ingest clay to obtain kaolinite/halloysite minerals which may alter the taste of their provided food, and may act as pharmaceutical agents to alleviate intestinal ailments such as diarrhea.     

Food availability and breeding season as predictors of geophagy in Amazonian parrots

Geophagy, the consumption of soil, is common in many species, but the drivers of geophagy are not well understood. The best‐studied example of avian geophagy is the parrots of the western Amazon Basin, but even here, there is debate over what drives the behaviour. There are two possible explanations: (1) extra nutritional demands of reproduction drive an increase in geophagy, which would predict that geophagy should be highest during the breeding season, and (2) consumption of naturally toxic plant foods increases the need for the toxin protection effects of soil, which would predict that geophagy should be highest when food availability is low and animals are forced to consume more toxic foods. We used long‐term data from lowland Amazonia to compare seasonal fluctuations in rainfall, food availability, parrot breeding and parrot geophagy, and conducted novel tests of these hypotheses. Our analyses of annual patterns suggested that seasonal changes in rainfall drive plant fruiting, the resulting food availability patterns drive the timing of parrot breeding, and breeding drives seasonal patterns of geophagy. Surprisingly, chicks of the largest psittacine species fledged as food supplies approach their annual lows, suggesting that future climatic changes that alter peaks in food availability could have unexpected impacts on the reproduction of large psittacines in this system. Our tests found no evidence to support the toxin‐protection hypothesis. Instead, we found that the peak of geophagy occurred during species’ breeding seasons, which strongly supports the supplemental nutrients hypothesis. Our findings join a growing body of biochemical, physiological, behavioural, ecological and biogeographical evidence suggesting that a need for sodium is driving soil consumption in this classic model system of avian geophagy.     

Parasite transmission risk from geophagic and foraging behavior in chacma baboons

Numerous behavioral and ecological factors are associated with parasite transmission. One factor explored in human research, but absent from nonhuman primate research, is parasite transmission from soil ingestion. Human studies suggest geophagy, the regular and deliberate consumption of soil, increases risk of soil‐transmitted helminth (STH) infection. Geophagy, which is prevalent in nonhuman primates, has several positive associations: gastrointestinal distress alleviation, possible mineral supplementation, and bacterial infection prevention. Our objective was to determine whether STH transmission was possible from deliberate or accidental soil ingestion, in a troop of chacma baboons (Papio hamadryas ursinus) that engaged in geophagy, foraged on the terrestrial substrata, and had a Trichuris sp. sample prevalence of 100%. We collected and analyzed 80 soil samples from geophagy and ground foraging sites on and around Wildcliff Nature Reserve, South Africa. Forty soil samples were collected from sites where soil was consumed, and 40 were collected from sites where soil was not consumed. At geophagy sites, the number of Trichuris sp. eggs recovered varied significantly between areas used and unused by the baboons, suggesting behavior is an important factor. In contrast, at foraging sites, there was only a tendency to recover more eggs at used than unused areas, and we propose egg recovery was influenced by fecal contamination that occurred throughout foraging stands. The difference in egg recovery between used areas at geophagy sites and used areas at foraging sites was not significant. These preliminary findings suggest both geophagy and foraging sites are a potential source of STH infection for this troop. Whether geophagy benefits outweigh the potential cost of parasite reinfection for this baboon troop is unknown, so we encourage future research on the influence that host foraging behavior may have on parasite reinfection. Am. J. Primatol. 74:940‐947, 2012. © 2012 Wiley Periodicals, Inc.     

Use of Mineral Licks by White-Bellied Spider Monkeys (<Emphasis Type="BoldItalic">Ateles belzebuth</Emphasis>) and Red Howler Monkeys (<Emphasis Type="BoldItalic">Alouatta seniculus</Emphasis>) in Eastern Ecuador

Geophagy occurs in all primate groups and is particularly common in species that consume greater quantities of plant material, i.e., leaves, fruit. The function of geophagy is not fully understood and likely varies over space and time, perhaps in connection with changes in diet. Central to a better understanding of geophagy in primate ecology is knowledge of the occurrence of such behavior among different species and seasons. We used camera traps triggered by heat and motion to document the use of mineral licks by primates over a 3-yr period at a lowland forest site in eastern Ecuador (Tiputini Biodiversity Station). Such mineral licks can be important sources of minerals, nutrients, and other compounds for a wide range of species in Amazonian forests. Although 10 species of primates are known from the study site, we obtained photographs of only 2 species, Ateles belzebuth (white-bellied spider monkey) and Alouatta seniculus (red howler) at 2 of 4 saladeros surveyed. From late December 2004 through early January 2008, we recorded 192 photographs with a total of 318 Ateles belzebuth representing ≥66 separate visits. Comparable numbers for Alouatta seniculus were 80, 121, and 37. We recorded both species visiting a mineral lick at the same time on ≥7 occasions. Use of mineral licks varied across months; we recorded more visits from November through February, the drier period at Tiputini. Visits also varied by hour, with no visits before 0830 or after 1630; Ateles belzebuth showed a stronger mid-day peak in visits. Average visit length (calculated as the time between the first and last photographs of a given visit) was similar between the 2 species but median visit length was more than twice as long for Ateles belzebuth (15 min) as for Alouatta seniculus (6 min). Results indicate that mineral licks are important in the ecology of these species, but further studies are needed to determine the precise benefit(s) obtained and how benefits may vary with diet and other factors.     

Parrot claylick distribution in South America: do patterns of “where” help answer the question “why”?

Geophagy is well known among some Neotropical parrots. The clay apparently adsorbs dietary toxins and/or provides supplemental nutrients. We used location data and 23 environmental layers to develop a predictive model of claylick distribution using Maxent software. We related species characteristics to claylick use and examined how parrot assemblages using claylicks changed with distance from the centre of claylick distribution. Fifty‐two parrot claylicks were reported from an area of ca 4 million km² but over 50% were restricted to a 35 000 km² region of southeast Peru and northern Bolivia. Claylicks were strongly associated with moist forest on younger (<65 millions of yr) geological formations and exposed river banks. The predictive model of claylick distribution matched our reported range well, with precipitation of warmest quarter, land cover, temperature seasonality, and distance from the ocean being most important predictors of claylick presence. Twenty‐six of the region's 46 parrot species visited claylicks. Species differed greatly in their lick use, but body size, dietary breadth, abundance and other traits were poor predictors of lick use. We are confident that our survey identified the distribution of major parrot claylicks in South America, although less conspicuous parrot geophagy may occur elsewhere. We suggest that claylick distribution reflects both underlying geology (allowing claylick formation in only some regions) and the physiological need for geophagy among parrots in different parts of the continent. Data on the latter are inconclusive, but we argue that parrot claylick distribution supports the contention that geophagy is related more to sodium deficiencies than to protection from dietary toxins.     

Avian fruit preferences across a Puerto Rican forested landscape: pattern consistency and implications for seed removal

Avian fruit consumption may ensure plant reproductive success when frugivores show consistent preference patterns and effectively remove and disperse seeds. In this study we examined avian fruit preferences and their seed-removal services at five study sites in north-central Puerto Rico. At each site, we documented the diet of seven common fruit-eating avian species from February to September 1998. Using foraging observations and area-based estimates of fruit abundance, we examined preference patterns of birds. We found that 7 out of 68 fleshy-fruited plant species were responsible for most of the fruit diet of birds. Seventeen plant species were preferred and four of them were repeatedly preferred across several study sites and times by at least one avian species. Preferred plant species comprised a small percentage of fleshy fruits at each site (<15% in four out of five study sites), but showed extended phenology patterns. The quantity of seeds removed by frugivore species was not strictly related to preferences. Some frugivores showing no preference could effectively remove more seeds from plants at some locations than species exhibiting constancy in their patterns of preference. Only two frugivores, Euphonia musica and Vireo altiloquous, removed most of the seeds of plants for which they exhibited repeated preference across the landscape. Preference patterns, particularly those exhibiting consistency in space and time for plant species having prolonged fruiting periods, may have important mechanistic consequences for the persistence, succession, and regeneration of tropical plant communities.     

Do Tropical Frugivores Compete for Food?

In this paper I ask two questions: Is fruit ever limiting tovertebrate frugivores and, if so, do frugivores presently competefor food, either with closely- or distantly-related species?A brief review of the fruiting strategies of tropical plantsindicates that fruit can occasionally be superabundant, butit is often produced at low rates and in low quantities. Variationin fruit abundance results from several biotic selective pressures,including variation in the density, diversity, and reliabilityof potential dispersal agents. To judge from the size structure,dietary similarities, habitat preferences and foraging behaviorsof taxonomically-restricted guilds of frugivorous birds andmammals, members of these guilds have competed for food in thepast and must occasionally do so today. Unusual climatic conditionscan occasionally "upset" phenological patterns and can createfood shortages that promote competition among closely-relatedspecies of frugivores. Avian and mammalian frugivores, however,probably seldom compete with each other for food in present-daytropical ecosystems. A major reason for this is that many tropicalplants have evolved fruits that are attractive to only a limitedsubset of frugivores (e.g., only birds or only bats). Plantsapparently "perceive" qualitative differences in the dispersalservices of birds and mammals and attempt to attract membersof one group but not the other.     

Geophagy among primates: adaptive significance and ecological consequences

We review geophagy, or soil ingestion, in primates. This behaviour is widespread and is presumed to be important to health and nutrition. Primates may engage in geophagy for one or a combination of reasons. Here we present, and make a preliminary assessment of, six nonexclusive hypotheses that may contribute to the prevalence of geophagy. Four hypotheses relate to geophagy in alleviating gastrointestinal disorders or upsets: (1) soils adsorb toxins such as phenolics and secondary metabolites; (2) soil ingestion has an antacid action and adjusts the gut pH; (3) soils act as an antidiarrhoeal agent; and (4) soils counteract the effects of endoparasites. Two hypotheses pertain to geophagy in supplementing minerals and/or elements: (5) soils supplement nutrient-poor diets and (6) soils provide extra iron at high altitudes. In addition to these hypotheses, geophagy may satiate olfactory senses, serve as a famine food and finally may have no function at all. We draw together a large body of information from various sources to assess these hypotheses and suggest some tests to understand the function of geophagy. Our review suggests that primates engage in geophagy for a number of reasons that are nonexclusive. We conclude that mineral supplementation, adsorption of toxins, treatment of diarrhoea and pH adjustment of the gut seem the most plausible reasons why primates engage in geophagy. Copyright 2000 The Association for the Study of Animal Behaviour.