Myth, Marula, and Elephant: An Assessment of Voluntary Ethanol
Intoxication of the African Elephant (Loxodonta africana)
Following Feeding on the Fruit of the Marula Tree
School of Biological Sciences, University of Bristol,
Woodland Road, Bristol BS8 1UG, United Kingdom
Accepted 6/1/2005; Electronically Published 2/6/2006
Africa can stir wild and fanciful notions in the casual visitor;
one of these is the tale of inebriated wild elephants. The sug-
gestion that the African elephant (Loxodonta africana)becomes
intoxicated from eating the fruit of the marula tree (Sclerocarya
birrea) is an attractive, established, and persistent tale.Thisidea
now permeates the African tourist industry, historical travel-
ogues, the popular press, and even scholastic works. Accounts
of ethanol inebriation in animals under natural conditions ap-
pear mired in folklore. Elephants are attracted to alcohol, but
there is no clear evidence of inebriation in the field. Extrap-
olating from human physiology, a 3,000-kg elephant would
require the ingestion of between 10 and 27 L of 7% ethanol
in a short period to overtly affect behavior, which is unlikely
in the wild. Interpolating from ecological circumstances and
assuming rather unrealistically that marula fruit contain 3%
ethanol, an elephant feeding normally might attain an ethanol
dose of 0.3 g kg?1, about half that required. Physiologicalissues
to resolve include alcohol dehydrogenase activity and ethanol
clearance rates in elephants, as well as values for marula fruit
alcohol content. These models were highly biased in favor of
inebriation but even so failed to show that elephants can or-
dinarily become drunk. Such tales, it seems, may result from
“humanizing” elephant behavior.
*This paper was prepared as an overview of a symposium session presented at
“Animals and Environments,” the Third International Conference for Com-
parative Physiology and Biochemistry, Ithala Game Reserve, KwaZulu-Natal,
South Africa, 2004(http://www.natural-events.com/ithala/default-follow_2.asp).
†Corresponding author; e-mail: email@example.com.
Physiological and Biochemical Zoology 79(2):363–369. 2006. ? 2006 by The
University of Chicago. All rights reserved. 1522-2152/2006/7902-5055$15.00
Africa is intoxicatinginitsvistasandwildnessandinitsvarieties
of bottled liqueurs. First-time visitors frequently imagine rav-
enous beasts behind each bush, a fancy brought on by the
undiluted and visceral attraction of wild Africa and, perhaps,
by an indulgence with the liqueurs. Commercial brewing and
distilling are important industries in South Africa, and among
the more popular liqueurs is Amarula, made from the fruit of
the marula tree, whose bottle is adorned with marula fruit and
a rampant bull elephant. (The producers of Amarula [Distell
Group Ltd., Stellenbosch, South Africa] make no claim as to
marula intoxication of elephants but correctlydescribethelocal
folklore that elephants are insatiably attracted to the marula
tree, known locally as the “elephant tree”; http://www
ference for Comparative Physiology and Biochemistry, in the
Ithala Game Reserve, South Africa, we were prompted toreflect
on these issues by the persistently related anecdote of drunken
elephants. Even a superficial perusal of promotional material
from the African safari and tourist industry reveals a widely
held belief that evidence exists for ethanol intoxication in the
African elephant and most especially that this elephant (Lox-
odonta africana) becomes intoxicated from eating the ripe and
fermenting fruit of the marula tree (Sclerocarya birrea). This
belief is persistent. For example, the French naturalist Dele-
gorgue, describing his experiences in thevicinityoftheiMfolozi
River in present-day KwaZulu-Natal around 1839, reports how
his Zulu guides told him of the strangely aggressive behavior
shown by elephant bulls after feeding on marula fruits. He
writes, “The elephant has in common with man a predilection
for a gentle warming of the brain induced by fruit which has
been fermented by the action of the sun” (Alexander and Webb
1990; p. 275). Naturally, at a meeting with physiologists, there
were some attempts to calculate whether this was a likely phe-
nomenon, most befuddled by attempts at empirical proof with
the Amarula. In this paper, we explore and analyze the likeli-
hood of natural intoxication in the African savannah elephant.
A recent symposium at the Society for Integrative and Com-
S. Morris, D. Humphreys, and D. Reynolds
parative Biology New Orleans meetings (“In Vino Veritas: The
Comparative Biology of Ethanol Consumption,” January 5–9,
2004) examined the proposition of Dudley (2000, 2002) that
the predilection of humans for imbibing ethanol may be in-
herited from frugivorous primate ancestors (Dudley and Dick-
inson 2004) who developed an attraction to the ethanol that
may accumulate in overripe fruit (Dudley 2002). This intrigu-
ing idea remains to be fully tested (e.g., Dominy 2004; Levey
2004). Many of the accounts of “drunken animals” are anec-
dotal, mired in folklore and myth and possibly confused with
other forms of intoxication. Monkeysapparentlyavoidoverripe
fruit, although this is based on human definitions of acceptable
ripeness (Milton 2004). While bats seem to avoid fruit if it
contains more than 1% ethanol, the attractiveness of fruit with
an ethanol content close to that naturally within fruit has yet
to be tested (Sa ´nchez et al. 2004). Thus, while a pattern is hard
to discern, frugivory in elephants may be similarly alcohol
A dearth of data has led to a tendency to recount anecdotes
and a reliance on the few published accounts (see also Dennis
1987; Dudley 2000). There are almost no reliable studies on
ethanol intoxication for wild mammals (Dudley 2004) and few
for birds (e.g., Fitzgerald et al. 1990). Elephants clearly do have
a taste for alcohol (Siegel and Brodie 1984), which can have
deleterious effects on their behavior (e.g., http://news.bbc.co
.uk/2/hi/south_asia/3423881.stm). There are numerous ac-
counts of elephants becoming dangerously intoxicated afterac-
cessing stores of wine or beer in Assam and Bengal in India
(e.g., “Elephants Rampage, Trample 5 in India,” Los Angeles
Times, January 1, 1985, sec. 1). The legend of the elephant and
the marula fruit was explored by Siegel (1989) without a clear
outcome. Yet the popular press seems convinced; for example,
“What does it take to get an elephant drunk? Not much, ap-
parently: A few bites of the fruit of the marula tree seems to
do the trick … the fruit that leads elephants to shake down
marula trees” (Graves 2002). Can marula fruit really drive el-
ephants to such exertions? So compelling, it seems, is the idea
of drunken elephants that the suggestion appears in scholastic
works (e.g., p. 196 of Sukumar 2003) and embellishes press
reports (Yoon 2004).
Nonetheless, various animals may seek out alcohol (Dominy
2004) and use it in sensing and selecting between different
qualities of fruit. How likely, then, is it that African elephants
can truly be intoxicated from eating marula fruit? Do elephants
even eat the fruit of the marula tree, is an elephant capable of
becoming intoxicated by alcohol, and is there sufficient natu-
rally occurring alcohol in the fruit?
The Marula as an Elephant Food Item
The marula is a dioecious tree in deciduoussavannahsofsouth-
ern Africa and is a member of the mango family (Anacardi-
aceae) with an important role in local tribal cultures (Coates
Palgrave 1993). The tree is considered to have healing qualities
(e.g., Eloff 2001; Grace et al. 2003; Steenkamp 2003; Ojewole
2004), and the fruits are a source of proteinandvitamins.These
fruits are traditionally fermented by the local people toproduce
marula wines. The marula is of economic importance (e.g.,
Shackleton 2002), but what of the ecology? Many species rely
on Sclerocarya birrea as a food source or as a host, in the case
of some parasitic plants and the larvae of some species of Lep-
idoptera (Kroon 1999). Several herbivorous mammals browse
the marula tree, including the giraffe and the rhinoceros, but
the African elephant has a close affiliation with the species
(Gadd 2002; Jacobs and Biggs 2002a, 2002b).
Elephants can both browse and graze (Clemens and Maloiy
1982). During the growing season, food is more abundant and,
early on, of higher nutritional value, but during the dormant
season, trees often shed their leaves. It is during the start of
this period of lower-quality food that fruits and seedpods can
The large extent to which elephants destructively feed on the
bark and branches of marula (S. birrea) was a major limiting
factor on the marula tree population in three South African
reserves (Gadd 2002). Lewis (1987) similarly reported that el-
ephants in Zimbabwe had marula as an important part of the
diet and that the elephants were an important distributor of
Estimates of daily food intake of elephants range between
1% and 2% of the body mass (Table 5.1 in Sukumar 2003),
and feeding requires as much as 60%–75% of the waking hours
(Wyatt and Eltringham 1974; Poole 1997). In the dry season,
browse makes up 58.6% of daily food intake, herbs (included
as a browse material) are 12.8%, and forage is 28.6%. This
changes in the wet season, with 20.7% of the daily intake made
up of browse, 22.2% of herbs, and 57.1% of forage (collated
by Sukumar ). Fruits are normally not the only items
that elephants consume in a day, and their availability will be
seasonal. However, during fruiting periods, elephants may
spend longperiods ataparticulartreeorclusteroftrees,literally
gorging themselves with ripe fruit (Sukumar 2003). At these
times, fruit could become the main constituent of their diet.
Digestive Anatomy and Physiology of Elephants
Anecdotal reports suggest that marula fruit may ferment to
alcohol within the digestive system. Elephants are nonrumi-
nants and have a hindgut fermentation system consisting of an
enlarged cecum and colon to facilitate cellulose digestion
(Clemens and Maloiy 1982; for review, see Clauss and Hummel
2005). Sugars within the diet are metabolized by the microflora
to volatile fatty acids (Church 1988), making them unavailable
to alcoholic fermentation. Rees (1982) suggested a gut passage
time of between 21.4 and 46 h and Sukumar (2003) an even
shorter time of 12 h. These short times and the digestive ef-
ficiency of elephants (Clauss and Hummel 2005) would allow
Elephant and Marula
Figure 1. Blood alcohol content as milligrams per 100 mL in healthy
human males when fasted (solid symbols) or fed (opensymbols).Dosage
was 0.30 g kg?1body mass. Data abstracted from Jones et al. (1997).
10 mg 100 mL
≈ 0.1 g kg
very little further alcoholic fermentation of the fruit sugars to
take place. However, any endogenous marula yeasts (see next
section) are also ingested and could, together with the endog-
enous flora, support continued fermentation of fruit sugars
once the digesta enters the hindgut, although this would also
be rapidly fermented to fatty acids and methane. Thus, some
limited ethanol production may occur on diets rich in appro-
priately ripe fruit, although cecal fermentation also depends on
diet (Maloiy and Clemens 1991). However, elephants drink up
to 225 L of water each day (Poole 1997). Regular or recent
drinking will dilute the gut contents, but elephants can go
without drinking for up to 4 d. Elephants usually drink at night
and at midday, directly after the peak feeding periods of early
morning and late evening (Guy 1976; Clemens and Maloiy
1982, referring to Douglas-Hamilton 1973; Katugaha et al.
Can the Fruit of the Marula Become Sufficiently Alcoholic?
Elephant intoxication relies on the assumption that the marula
fruits contain a significant amount of alcohol. If they do, yeast
spores must find their way onto the skin of the fruits and begin
the fermentation process while the fruits are still on the branch,
because it is from there that the elephants feed upon them.
Siegel (1989) appears in no doubt that the alcoholic content
of the fruit can become as high as 7%, but there appear to be
no other data to support this high level. Empirical studies of
alcohol content in naturally fermenting fruit are surprisingly
few (Levey 2004). Yeast, including Saccharomyces, occur in a
variety of fruits (e.g., Tang et al. 2003) but will produce ethanol
only when protected from O2within the body of large fruits
(Levey 2004). Microorganisms isolated from ripe marula fruits
include at least three species of Hansenula (Okagbue andSiwela
2002); the presence of members of the Saccharomycetaceae,
and thus spontaneous fermentation, in marula fruit is at least
possible. Eriksson and Nummi (1982), Dudley (2002), Dominy
(2004, table 1), and Sa ´nchez et al. (2004) report a range of
ethanol from 0.04% to 0.72% in the fruit of 18 species, con-
siderably below the 7% of Siegel (1989). However, Dudley
(2004) reported the pulp of overripe palm fruits (Astrocaryum
standleyanum) to contain up to 4.5% ethanol.
The brewing of marula beer requires leaving the juice to
ferment naturally for 3–4 d with endogenous yeasts; no other
strains are incorporated, as in commercial brewing, which may
et al. 2004). Thus, on one view, it may be possible that the
alcoholic content of a fallen marula fruit with naturally fer-
mented sugars could perhaps reach the lower end of the range
of alcohol of marula beer (3%). Temperature is important in
the process of fermentation. Sclerocarya birrea is endemic to
southern Africa (Nerd and Mizrahi 2000) and has a fruiting
period from January to March (ambient temperatures of 26?–
28?C). This range is at the high end of temperatures used in
commercial fermentation, and preferential temperatures used
for production of marula wines are around 15?C (Fundira et
Ethanol Intoxication in Wild Elephants
The amount of alcohol required to intoxicate an elephant can,
potentially, be determined by extrapolation. In humans, ap-
proximately 20% of ingested ethanol is absorbed immediately
in the stomach, with further and slower absorptioninthestom-
ach and duodenum, although this dependsonfastingstate(e.g.,
Cortot et al. 1986; Jones et al. 1997; Fig. 1). Ethanol quickly
distributes itself throughout the body water space, reaching
peak concentrations under fasting conditions at approximately
1 h after consumption (Fig. 1). The ethanol is metabolizedover
the next few hours. There appear to be no data for the relative
activity of alcohol dehydrogenase (ADH) in elephants, but in
three bird species, Prinzinger and Hakimi (1996) found the
highest ADH activity in the frugivorous species. Further evi-
dence that ADH activity is related to thepropensitytofrugivory
is scarce, and consequently, Dudley (2004) cites evidence from
Drosophila, in which the mechanisms of inebriation are similar
to those in vertebrates (e.g., Moore et al. 1998; Wolf and He-
berlein 2003). Flies lacking functional ADH are rendereddrunk
by an alcohol concentration of 5%, and there appears to be a
correlation between developmental ethanol exposure, enzyme
activity, and ethanol resistance (e.g., Ashburner 1998; Fry 2001;
Fry et al. 2004). Given the evidence of elephants feeding on
fruit likely to contain at least some ethanol, there is no reason
to assume that their ADH levels are especially low. However,
should ADH activity prove to be especially low, then there
remains the possibility that relatively lower ethanol doses may
S. Morris, D. Humphreys, and D. Reynolds
be intoxicating; but this would necessarily contradict the re-
quired doses for elephant intoxication provided by Siegel and
Brodie (1984; see next section). The likelihood of inebriation
in elephants can also be interpolated from their ecological cir-
cumstances, and here we present models for both extrapolation
Model 1: Extrapolation by Scaling from Homo sapiens to
Adult African elephants range from 5,500 to 6,000 kg in males
and from 2,500 to 3,000 kg in females (Owen-Smith 1988;
Shoshani 1992; Poole 1997). In humans, a blood alcohol con-
tent (BAC) of approximately 0.15 g 100 mL?1is likely to pro-
duce overt intoxication effects. Thus, a 70-kg human would
need to consume approximately 90 mL of ethanol (12 mL of
ethanol results in a BAC of 0.02 g 100 mL?1) to acquire this
level of BAC (Garriott 1988, 2003). Proportional scaling to the
mass of elephants would produce a figure of 3.87 L of ethanol,
or 155 L of 7% ethanol. However, human and elephant me-
tabolism should scale to a power of ∼0.75 (M0.738;Kleiber1961).
Failure to account for scaling contributed in part to the con-
troversial massive overdosage and death of an elephant into
whichWest et al. (1962) injected297mgofLSD(Spinage1994).
Scaling from humans and accounting for metabolic rate would
have reduced this to less than 4 mg. Thus, on the basis of
cellular metabolic rate, the amount of alcohol required to in-
toxicate elephants may be relatively less, reducing the 55 L of
7% ethanol to closer to 10 L. There are no directly relevant
data for rates of alcohol metabolism.
Inebriation of elephants was determinedbySiegelandBrodie
(1984) to occur at a BAC of 0.05–0.1 g 100 mL?1. The highest
likelihood for intoxicationcanbeestimatedusingbloodvolume
alone, since in humans the degree of impairment is directly
proportional to BAC (Garriott 1988, 2003). With an approx-
imate blood volume of 7% of body mass (Cameron et al. 1999),
a 70-kg person possesses 4.9 kg of blood. In elephants, the
blood mass is less, approximately 3.5% of body mass (Shoshani
1992), and thus a 3,000-kg elephant circulates 105 kg of blood.
A 90-mL dose of ethanol raises human BAC to 0.15 g 100
mL?1, a level for which an elephant would proportionately
require 1,900 mL of ethanol (cf. 0.05–0.1 g 100 mL?1; Siegel
and Brodie 1984).
Marula trees produce up to 8,000 fruits, each 42 g and ap-
proximately 3.5 cm in diameter (Lewis 1987). Elephants are
capable of consuming around 1%–2% of their body mass per
day (Clemens and Maloiy 1982; Owen-Smith 1988). Thus, a
3,000-kg elephant might therefore eat at least 30 kg of fruit in
one day (assuming it consumed only marula fruit; see next
section), or approximately 714 individual fruits.
Thus, accepting that the unlikely estimate of 7% ethanol in
marula fruit (Siegel 1989) can be achieved, if the required 1.9
L of ethanol were diluted to aconcentrationof7%,theresulting
volume would be ∼27 L. Further assuming, rather unrealisti-
cally, that the alcohol remained in the bloodstream throughout
the day, then each fruit would need to contain 38 mL of the
7% ethanol solution to produce intoxication. This volume is
impossible to accommodate inside the average fruit, which has
a volume of ∼22 mL, including the large kernel (∼1.7-cm di-
ameter; Glew et al. 2004). Local marula wine production re-
quires at least 200 fruits to make 1 L of alcoholic beverage
These calculations were highly biased in favor of “drunken”
elephants. Any alcoholic effects would be mitigated by the me-
tabolism of the alcohol during the day, and the model also
assumed all the fruits ingested would be overripe and in an
advanced stage of fermentation (cf. Dudley 2004). Because of
the bias in the model toward inebriation, it seems impossible
that all, if any, of these conditions can be met; thus, drunken
elephants seem highly improbable.
Model 2: Interpolating from Ecological Reality to
Relative densities of the marula tree Sclerocarya birrea ranged
from 2% to 35% over 28,000 hectares in three game reserves
in South Africa (Gadd 2002); of these, 38% were fruiting. To-
gether with data for likely elephant movement, these values
allow a projection of the number of trees and fruit an elephant
will be able to access. This model requires a solitary animal,
maximizing the marula fruit component of the diet but having
drunk little or no water for a number of days.
An approximate density of marula trees can be derived as
13.66 ha?1(Gadd 2002). A complete understanding of the el-
ephant foraging strategy has yet to be attained (seefactorslisted
by Sukumar , p. 217), but assuming equal distribution
of trees at ∼13 ha?1, an optimized model can be devised. In
this scenario, the elephant can visit homogeneouslyspacedtrees
within that 1 ha by covering a distance of approximately 430
m. Guy (1976) suggested that elephants spend between 16.4%
and 19% of the day walking, and employing the average dis-
tance traveled (∼11.5 km d?1; maximum 38.6 km; Wyatt and
Eltringham 1974), an elephant could visit up to 345 trees a day.
Of these trees, 38% could potentially be in fruit (Gadd 2002),
giving a figure of 131 trees that the elephant visits being in
fruit. Fruit production by the marula tree gave an average yield
of 36.8 kg (Shackleton 2002). Thus, if just 1% of this fruit were
accessible (i.e., ripe), the elephant could access 48 kg of fruit
in one day, which is close to the 1%–2% of body mass that
the elephants must consume each day. If an elephant consumes
2% of its body mass in fruit, and assuming 3% of this fruit to
be alcohol, the model grants access to an ethanol dose of 0.6
g kg?1. In fed humans, a dose of 0.3 g kg?1resulted in a
maximum BAC of 15 mg 100 mL?1, and even accounting for
scaling, the 0.6 g kg?1is insufficient to permit inebriationbased
on the 50–100 mg 100 mL?1estimate of Siegel and Brodie
Elephant and Marula
(1984). To achieve this dose estimate, this model omits thetime
required for actually harvesting the fruit, assuming instead that
this occurs while the elephant is slowly moving. In addition,
elephants eat a varied diet and not entirely alcoholic marula
Ethanol assimilation is not 100% efficient or instantaneous.
Furthermore, should an elephant consume purely fruit as its
forage material in addition to browse (43%), which is unlikely
(Sukumar 2003), the fruit would constitute approximately half
the total daily diet. With a 3% ethanol content and assuming
100% absorption efficiency with zero ADH activity, this would
result in ∼0.34 g kg?1body mass assimilated. Furthermore, the
kernel of the marula would occupy 2.6 cm3(Glew et al. 2004).
Thus, the potential ethanol entering the elephant is reduced by
a further 11%, to ∼0.3 g kg?1, and is substantially less than
that required to reach the figure of Siegel and Brodie (1984).
On this basis, it is highly unlikely that an elephant would be-
Both the extrapolative and interpolative models were con-
structed to maximize the likelihood of inebriation, but neither
was able to convincingly support this conclusion. When fruit
are available, elephants may feed on them to the exclusion of
other items (White et al. 1993) and feed to excess (Sukumar
2003), but even so, inebriation is highly unlikely. Other vari-
ables include access to fruit and the rate of alcohol clearance
in elephants. Lone individuals (usually bulls; Katugaha et al.
1999) will have relatively greater access to marula fruit. While
it must be concluded that elephants likely cannot become suf-
ficiently intoxicated by the fruits of the marula, physiological
knowledge of how elephants deal with alcohol is practically
nonexistent. Unregulated behavior of elephants in the field and
as featured in Zulu accounts may be due to an intoxicant other
than alcohol. For example, the bark of marula is home to beetle
tips (e.g., Mebs et al. 1982). Marula fruit are extraordinarily
high in some vitamins, including nicotinic acid (niacin/vitamin
B3; SEPASAL [Survey of Economic Plants for Arid and Semi-
Arid Lands], Royal Botanic Gardens, Kew, http://www.rbgkew
.org.uk/ceb/sepasal/birrea.htm), and large doses may cause
overt effects. Furthermore, unexpectedly aggressive behavior is
most often reported for bull elephants, and if the marula fruit
is a prized food item, then the observed behavior may simply
be defense of that valued food source.
Field data are few, but a study by Mathonsi (1999) found
that approximately 90% of the marula fruits were passed with-
out being “squeezed, squashed or processed in any way.” While
keeping the fruit whole might allow fermentation to persist,
protected within the fruit body, it seems very unlikely that the
elephants truly access all of the fruit content. Furthermore,
elephants show a distinct preference for fruit still on the tree
rather than those ripening on the ground. It is improbable that
enough of the fruit on the tree could simultaneously achieve
sufficiently alcoholic status, although an elephant may select
those that do. Assuming that all other model factors are in
favor of inebriation, then intoxication would minimallyrequire
that the elephant avoids drinking water and consumes a diet
of only marula fruit at a rate of at least 400% normalmaximum
food intake and with a mean alcohol content of at least 3%.
On our analysis, this seems extremely unlikely.
Elephants display many behavioral characteristics viewed as
positive traits in humans, often causing us to identifywiththem
in anthropocentric ways. Thus, like beauty, it seems that the
tipsy pachyderm may exist in the “eye of the beholder,” a view
bolstered perhaps by a mutual desire for the fruits of themarula
We thank the reviewers, who helped straighten up some be-
fuddled points and kindly pointed us in the direction of further
historical anecdotes. We also thank the delegates to the Ithala
2004 meeting in Africa, who further fermented the tale of
drunken elephants and stirred us to this analysis, and the many
hundreds of elephants we have observed throughout Africa,
some very angry but not one of which has ever appeared to
Alexander S. and C. de B. Webb, eds. 1990. Adulphe Dele-
gorgue’s Travels in Southern Africa. Translated by F. Webb.
Vol. 1. University of KwaZulu-Natal Press, Pietermaritzburg.
Ashburner M. 1998. Speculations on the subject of alcohol
dehydrogenase and its properties in Drosophila and other
flies. Bioessays 20:949–954.
Cameron J., R. Grant, and J. Skofronick. 1999. Physics of the
Body. 2nd ed. Medical Physics, Madison, WI.
Church D.C. 1988. The RuminantAnimal:DigestivePhysiology
and Nutrition. Prentice Hall, Englewood Cliffs, NJ.
Clauss M. and J. Hummel. 2005. The digestive performance of
Mammal Rev 35:174–187.
Clemens E.T. and G.M.O. Maloiy. 1982. The digestive physi-
ology of three East African herbivores: the elephant, rhi-
noceros and hippopotamus. J Zool (Lond) 198:141–156.
Coates Palgrave K. 1993. Trees of Southern Africa. Struik, Cape
Cortot A., G. Jobin, F. Ducrot, C. Aymes, V. Giraudeaux, and
R. Modigliani. 1986. Gastric emptying and gastrointestinal
absorption of alcohol ingested after a meal. Dig Dis Sci 31:
S. Morris, D. Humphreys, and D. Reynolds
Dennis J.V. 1987. If you drink, don’t fly: fermented fruit and
sap can inebriate birds! Birder’s World 1:15–19.
Dominy N.J. 2004. Fruits, fingers, and fermentation: the sen-
sory cues available to foraging primates. Integr Comp Biol
Douglas-Hamilton I. 1973. On the ecology and behaviour of
the Lake Manyara elephant. E Afr Wildl J 11:401–403.
Dudley R. 2000. Evolutionary origins of human alcoholism in
primate frugivory. Q Rev Biol 75:3–15.
———. 2002. Fermenting fruit and the historical ecology of
ethanol ingestion: is alcoholism in modern humans an evo-
lutionary hangover? Addiction 97:381–388.
———. 2004. Ethanol, fruit ripening, and the historicalorigins
of human alcoholism in primate frugivory. IntegrCompBiol
Dudley R. and M. Dickinson. 2004. The comparative biology
of ethanol consumption: an introduction to the symposium.
Integr Comp Biol 44:267–268.
Eloff J.N. 2001. Antibacterial activity of marula (Sclerocarya
birrea (A. rich.) Hochst. subsp. caffra (Sond.)Kokwaro)(An-
acardiaceae) bark and leaves. J Ethnopharmacol 76:305–308.
Eriksson K. and H. Nummi. 1982. Alcohol accumulation from
ingested berries and alcohol metabolism in passerine birds.
Ornis Fenn 60:2–9.
Fitzgerald S.D., J.M. Sullivan, and R.J. Everson. 1990.Suspected
ethanol toxicosis in 2 wild cedar waxwings. Avian Dis 34:
Fry J.D. 2001. Direct and correlated responses to selection for
larval ethanol tolerance in Drosophila melanogaster. J Evol
Fry J.D., C.M. Bahnck, M. Mikucki, N. Phadnis, and W.C.
Slattery. 2004. Dietary ethanol mediates selection on alde-
hyde dehydrogenase activity in Drosophila melanogaster. In-
tegr Comp Biol 44:275–283.
Fundira M., M. Blom, I.S. Pretorius, and P. vanRensburg.2002.
Selection of yeast starter culture strains for the production
of marula fruit wines and distillates. J Agric Food Chem 50:
Gadd M. 2002. The impact of elephants on the marula tree
Sclerocarya birrea. Afr J Ecol 40:328–336.
Garriott J.C. 1988. Medicolegal Aspects of Alcohol Determi-
nation in Biological Specimens. PSG, Littleton, MA.
———. 2003. Medical-Legal Aspects of Alcohol. 4th ed. Law-
yers and Judges, Tucson, AZ.
Glew R.S., D.J. VanderJagt, Y.-S. Huang, L.-T.Chuang,R.Bosse,
and R.H. Glew. 2004. Nutritional analysis of the edible pit
of Sclerocarya birrea in the Republic of Niger (daniya,
Hausa). J Food Compos Anal 17:99–111.
Grace O.M., H.D.V. Prendergast, A.K. Jager, and J. van Staden.
2003. Bark medicines used in traditional healthcare in
KwaZulu-Natal, South Africa: an inventory. S Afr J Bot 69:
Graves E. 2002. Spirits: out of Africa. Food and Wine (June).
Guy P.R. 1976. Diurnal activity patterns of the elephant in the
Sengwa area, Rhodesia. E Afr Wildl J 14:285–295.
Jacobs O.S. and R. Biggs. 2002a. The status and population
structure of the marula in the Kruger National Park. S Afr
J Wildl Res 32:1–12.
———. 2002b. The impact of the African elephant on marula
trees in the Kruger National Park. S Afr J Wildl Res 32:13–
Jones A.W., K.A. Jonsson, and S. Kechagias. 1997. Effect of
high-fat, high-protein, and high-carbohydrate meals on the
pharmacokinetics of a small dose of ethanol. Br J Clin Phar-
Katugaha H.I.E., M. de Silva, and C. Santiapillai. 1999. A long-
term study on the dynamics of the elephant (Elephas max-
imus) population in Ruhuna National Park, Sri Lanka. Biol
Kleiber M. 1961. The Fire of Life: An Introduction to Animal
Energetics. Wiley, New York.
Kroon D.M. 1999. Lepidoptera of Southern Africa: Host Plants
and Other Associations: A Catalogue. Lepidopterists’Society
of Africa, Jukskei Park.
Levey D.J. 2004. The evolutionary ecology of ethanol produc-
tion and alcoholism. Integr Comp Biol 44:284–289.
Lewis D. 1987. Fruiting patterns, seed germination, and dis-
tribution of Sclerocarya caffra. Biotropica 19:50–56.
Maloiy G.M.O. and E.T. Clemens. 1991. Aspects of digestion
and in vitro fermentation in the caecum of some East African
herbivores. J Zool (Lond) 224:293–300.
Mathonsi S. 1999. The myth of intoxication by marula fruit in
the diet of the African elephant. Ecol J 1:56.
Mebs D., F. Bruning, N. Pfaff, and H.D. Neuwinger. 1982.
Preliminary studies on the chemical properties of the toxic
principle from Diamphidia nigroornata larvae, a source of
Bushman arrow poison. J Ethnopharmacol 6:1–11.
Milton K. 2004. Ferment in the family tree: does a frugivorous
dietary heritage influence contemporary patterns of human
ethanol use? Integr Comp Biol 44:304–314.
Moore M.S., J. DeZazzo, A.Y. Luk, T. Tully, S.M. Singh, and
U. Heberlein. 1998. Ethanol intoxication in Drosophila: ge-
netic and pharmacological evidence for regulation by the
cAMP signaling pathway. Cell 93:997–1007.
Morrissey W.F., B. Davenport, A. Querol, and A.D.W. Dobson.
2004. The role of indigenous yeasts in traditional Irish cider
fermentations. J Appl Microbiol 97:647–655.
Nerd A. and Y. Mizrahi. 2000. Introduction of marula, an un-
exploited fruit tree from southern Africa, to theIsraeliNegev.
Isr J Plant Sci 48:217–222.
Ojewole J.A.O. 2004. Evaluation of the analgesic, anti-inflam-
matory and anti-diabetic properties of Sclerocarya birrea (A.
rich.) Hochst.: stem-bark aqueous extract in mice and rats.
Phytotherapy Res 18:601–608.
Elephant and Marula
Okagbue R.N. and M. Siwela. 2002. Yeasts and related micro-
organisms isolated from ripe marula fruits (Sclerocarya caf-
fra) in Zimbabwe. S Afr J Sci 98:551–552.
Owen-Smith R.N. 1988. Megaherbivores: The Influenceof Very
Large Body Size on Ecology. Cambridge University Press,
Poole J. 1997. The Elephants. 2nd ed. Colin Baxter Photog-
Prinzinger R. and G.A. Hakimi. 1996. Alcohol resorption and
J Ornithol 137:319–327.
Rees P.A. 1982. Gross assimilation, efficiency and food passage
time in the African elephant. Afr J Ecol 20:193–198.
Sa ´nchez F.,C.Korine, B.Pinshow,andR.Dudley.2004.Possible
roles of ethanol in the relationship between plants and fru-
givores: first experiments with Egyptian fruit bats. Integr
Comp Biol 44:290–294.
Shackleton C. 2002. Growth and fruit production of Sclerocarya
birrea in the South African lowveld. Agrofor Syst 55:175–180.
Shoshani J. 1992. Elephants: Majestic Creatures of the Wild.
Rodale, Emmaus, PA.
Siegel R.K. 1989. Intoxication: Life in Pursuit of Artificial Par-
adise. Dutton, New York.
Siegel R.K. and M. Brodie. 1984. Alcohol self-administration
by elephants. Bull Psychon Soc 22:49–52.
Spinage C. 1994. Elephants. Poyser, London.
Steenkamp V. 2003. Traditional herbal remedies used by South
African women for gynaecological complaints. J Ethnophar-
Sukumar R. 2003. The Living Elephants: Evolutionary Ecology,
Behaviour and Conservation. Oxford University Press,
Tang A.M.C., K.D. Hyde, and R.T. Corlett. 2003. Diversity of
fungi on wild fruits in Hong Kong. Fungal Divers 14:165–
West L.J., C.M. Pierce, and W.D. Thomas. 1962. Lysergic acid
diethylamide: its effect on a male Asiatic elephant. Science
White L.T.J., C.E.G. Turin, and M. Fernandez. 1993. Group
composition and diet of forest elephants, Loxodonta africana
cyclotis Matschie 1900, in the Lope ´ Reserve, Gabon. Afr J
Wolf F.W. and U. Heberlein. 2003. Invertebrate models of drug
abuse. J Neurobiol 54:161–178.
Wyatt J.R. and S.K. Eltringham. 1974. The daily activity of the
elephant in the Rwenzori National Park Uganda. E Afr Wildl
Yoon C.K. 2004. “Of drunken elephants, tipsy fish and scotch
with a twist.” New York Times, March 23, late edition–final,