1. Diet and Primate Evolution
Many characteristics of modern primates, including
our own species, derive from an early ancestor’s practice
of taking most of its food from the tropical canopy
by Katharine Milton
A
s recently as 20 years ago, the especially those consumed in industrial- siderable enhancement of the visual
canopy of the tropical forest was ly advanced nations, bear little resem- apparatus ( including depth perception,
regarded as an easy place for blance to the plant-based diets anthro- sharpened acuity and color vision),
apes, monkeys and prosimians to Þnd poids have favored since their emer- thereby helping primates travel rapidly
food. Extending an arm, it seemed, was gence. Such Þndings lend support to the through the three-dimensional space of
virtually all our primate relatives had to suspicion that many health problems the forest canopy and easily discern the
do to acquire a ready supply of edibles common in technologically advanced presence of ripe fruits or tiny, young
in the form of leaves, ßowers, fruits and nations may result, at least in part, from leaves. And such pressures favored in-
other components of trees and vines. a mismatch between the diets we now creased behavioral ßexibility as well as
Since then, eÝorts to understand the re- eat and those to which our bodies be- the ability to learn and remember the
ality of life for tree dwellers have helped came adapted over millions of years. identity and locations of edible plant
overturn that misconception. Overall, I would say that the collected parts. Foraging beneÞts conferred by
My own Þeld studies have provided evidence justiÞably casts the evolution- the enhancement of visual and cognitive
considerable evidence that obtaining ad- ary history of primates in largely di- skills, in turn, promoted development of
equate nutrition in the canopyÑwhere etary terms. an unusually large brain, a characteris-
primates evolvedÑis, in fact, quite diÛ- The story begins more than 55 mil- tic of primates since their inception.
cult. This research, combined with com- lion years ago, after angiosperm forests As time passed, primates diverged
plementary work by others, has led to spread across the earth during the late into various lineages: Þrst prosimians,
another realization as well : the strat- Cretaceous (94 to 64 million years ago). most of which later went extinct, and
egies early primates adopted to cope At that time, some small, insect-eating then monkeys and apes. Each lineage
with the dietary challenges of the arbo- mammal, which may have resembled a arose initially in response to the pres-
real environment profoundly inßuenced tree shrew, climbed into the trees, pre- sures of a somewhat diÝerent dietary
the evolutionary trajectory of the pri- sumably in search of pollen-distributing niche; distinct skills are required to be-
mate order, particularly that of the an- insects. But its descendants came to rely come an eÛcient forager on a particu-
thropoids (monkeys, apes and humans). substantially on edible plant parts from lar subset of foods in the forest cano-
Follow-up investigations indicate as the canopy, a change that set the stage py. Then new dietary pressures placed
well that foods eaten by humans today, for the emergence of the primate order. on some precursor of humans paved
Natural selection strongly favors traits the way for the development of mod-
that enhance the eÛciency of foraging. ern humans. To a great extent, then, we
Hence, as plant foods assumed increas- are truly what we eat.
KATHARINE MILTON is professor of
anthropology at the University of Califor-
ing importance over evolutionary time
M
nia, Berkeley. After earning her doctor- (thousands, indeed millions, of years), y interest in the role of diet in
ate from New York University in 1977, selection gradually gave rise to the suite primate evolution grew out of
she spent three years conducting Þeld of traits now regarded as characteristic research I began in 1974. While
studies on the foraging behavior of howl- of primates. Most of these traits facili- trying to decide on a topic for my doc-
er and spider monkeys. She joined the tate movement and foraging in trees. toral dissertation in physical anthro-
Berkeley faculty in 1980. Milton is now For instance, selection yielded hands pology, I visited the tropical forest on
studying the dietary behavior and dis-
well suited for grasping slender branch- Barro Colorado Island in the Republic
eases of indigenous human populations
living in the tropical forests of Brazil. es and manipulating found delicacies. of Panama. Studies done on mantled
Selective pressures also favored con- howler monkeys (Alouatta palliata) in
86 SCIENTIFIC AMERICAN August 1993 Copyright 1993 Scientific American, Inc.

2. the 1930s at that very locale had inad- must seek out a variety of complemen- cies in tropical forests, which include
vertently helped foster the impression tary nutrient sources, a demand that hundreds of tree species. An animal
that primates enjoyed the Òlife of Ri- greatly complicates food gathering. that concentrated on eating food from a
leyÓ in the canopy. For instance, most arboreal primates single species would have to exert great
Yet, during my early weeks of follow- focus on ripe fruits and leaves, often eÝort going from one individual of that
ing howlers, I realized they were not supplementing their mostly herbivorous species to another. What is more, trees
behaving as expected. Instead of sitting intake with insects and other animal exhibit seasonal peaks and valleys in
in a tree and eating whatever happened matter. Fruits tend to be of high quality the production of the fruits and young
to be growing nearby, they went out of (rich in easily digested forms of carbo- leaves primates like to eat, again mak-
their way to seek speciÞc foods, mean- hydrate and relatively low in Þber), but ing reliance on a single food species
while rejecting any number of seeming- they provide little protein. Because all untenable.
ly promising candidates. Having found a animals need a minimal amount of pro-
F
preferred food, they did not sate them- tein to function, fruit eaters must Þnd rom an evolutionary perspective,
selves. Instead they seemed driven to additional sources of amino acids. Fur- two basic strategies for coping
obtain a mixture of leaves and fruits, thermore, the highest-quality items in with these many problems are
drawn from many plant species. the forest tend to be the most scarce. open to a nascent plant eater. In one,
The old easy-living dogma was clearly Leaves oÝer more protein and are more morphology reigns supreme: over long
far too simplistic. I decided on the spot plentiful than fruit, but they are of low- time spans, natural selection may favor
to learn more about the problems howl- er quality (lower in energy and higher the acquisition of anatomic specializa-
ers and other anthropoids face meeting in Þber) and are more likely to include tionsÑespecially of the digestive tractÑ
their nutritional needs in the tropical undesirable chemicals. that ease the need to invest time and
forest. I hoped, too, to discern some of The need to mix and match plant energy searching for only the highest-
the strategies they had evolved to cope foods is further exacerbated by the large quality dietary items. That is, morpho-
with these dietary diÛculties. distance between trees of the same spe- logical adaptations enable animals to
The challenges take many forms. Be-
cause plants cannot run from hungry
predators, they have developed other
defenses to avoid the loss of their edi-
ble components. These protections in-
clude a vast array of chemicals known
as secondary compounds (such as tan-
nins, alkaloids and terpenoids). At best,
these chemicals taste awful ; at worst,
they are lethal.
Also, plant cells are encased by walls
made up of materials collectively re-
ferred to as Þber or roughage: substanc-
es that resist breakdown by mammalian
digestive enzymes. Among the Þbrous
constituents of the cell wall are the
structural carbohydratesÑcellulose and
hemicelluloseÑand a substance called
lignin; together these materials give
plant cell walls their shape, hardness
and strength. Excessive intake of Þber is
troublesome, because when Þber goes
undigested, it provides no energy for
the feeder. It also takes up space in
the gut. Hence, until it can be excreted,
it prevents intake of more nourishing
items. As will be seen, many primates,
including humans, manage to extract a
certain amount of energy, or calories,
from Þber despite their lack of Þber-
degrading enzymes. But the process is
time-consuming and thus potentially
problematic.
The dietary challenges trees and vines
pose do not end there. Many plant foods
lack one or more nutrients required by
animals, such as particular vitamins or
amino acids (the building blocks of pro-
YOUNG CHIMPANZEES SEEK FRUIT as part of a diet that consists primarily of ripe
tein), or else they are low in readily di- fruits supplemented by leaves and some animal prey. Obtaining the foods needed
gestible carbohydrates (starch and sug- for adequate nutrition in the tropical forest turns out to be signiÞcantly more diÛ-
ar), which provide glucose and there- cult for primates than was once believed. The author contends that the solutions
fore energy. Usually, then, animals that adopted by primates millions of years ago strongly inßuenced the subsequent evo-
depend primarily on plants for meet- lution of the primate order. The drawings on the opposite page depict some typical
ing their daily nutritional requirements plant foods available to arboreal animals in the tropical forest.
Copyright 1993 Scientific American, Inc. SCIENTIFIC AMERICAN August 1993 87

3. macaque orangutan
lemur
marmoset
gibbon
HOLOCENE
PLEISTOCENE PROSIMIANS NEW WORLD MONKEYS OLD WORLD MONKEYS LESSER APES GREAT
0
PLIOCENE
10
MIOCENE
20
MILLIONS OF YEARS AGO
30 OLIGOCENE
40
EOCENE
50
PRECURSOR SMALL, INSECT-EATING MAMMAL
60 PALEOCENE MOVES INTO THE TREES
OF PROSIMIANS
70 LATE
CRETACEOUS
EVOLUTIONARY TREE of the primate order is rooted in the dants of this mammal (artistÕs representation to left of tree)
late Cretaceous, when a small , insect-eating mammal climbed adapted to a new dietary niche in the canopy, they developed
into the trees to take advantage of feeding opportunities pre- traits now regarded as characteristic of primates, such as a
sented by the spread of angiosperm forests. As the descen- rounded snout and nails (instead of claws). These descendants
depend on plant parts that are ubiqui- energy for body tissues or are delivered pious amounts of Þber does not confer
tous, such as on mature leaves (which to the liver for conversion into glucose. the same beneÞts as it does for the di-
are readily available but not of particu- Some researchers think the colobine gestively specialized colobines.
larly high quality). forestomach may also aid in the de- Another morphological change that
Colobine monkeys, one of the Old toxiÞcation of harmful secondary com- can facilitate survival on lower-qual-
World primate groups in Africa and pounds in plant foods. ity plant parts is to grow larger over
Asia, oÝer an excellent example of this EÛciency of nutrient extraction from time. Compared with small animals, big
strategy. Unlike the typical primate di- Þbrous foods is enhanced in another ones must consume greater absolute
gestive tract (including that of humans), way in colobine monkeys. As cellulolytic amounts of food to nourish their more
with its simple acid stomach, that of bacteria die, they pass out of the fore- extensive tissue mass. But, for reasons
colobines includes a compartmental- stomach into the second compartment, that are imperfectly understood, the big-
ized, or sacculated, stomach function- a simple acid stomach similar to our ger animals can actually attain adequate
ally analogous to that of cows and oth- own. Here special enzymes (lysozymes) nourishment by taking in less energy
er ruminants. This anatomic specializa- cleave the bacterial cell walls. In con- per unit of body mass. This relative-
tion enables colobines to process Þber sequence, protein and other nutritious ly lower energy demand means larger
extremely eÛciently. materials that compose the cellulolytic animals can meet their energy require-
Chewed leaves ßow through the bacteria become available for digestion ments with lower-quality foods. Grow-
esophagus into the forestomach, one by the monkeys. (In a sense, then, once ing bigger has been only a limited op-
of the two stomach compartments in leaves are chewed and swallowed, colo- tion for most primates, however. If ar-
colobines. In this alkaline forestomach, bine monkeys do not interact directly boreal animals grow too massive, they
microbes known as cellulolytic bacteria with their food ; they live on products risk breaking the branches underneath
do what digestive enzymes of the mon- of the fermentation process and on the their feet and falling to the ground.
keys cannot do: degrade Þber. In a pro- nutrients provided by the fermenters.)
T
cess known as fermentation, the bacte- In contrast to colobines, humans and he second basic strategy open to
ria break down the cellulose and hemi- most other primates pass Þber basical- plant eaters is more behavioral
cellulose in plant cell walls, using those ly unchanged through their acid stom- than morphological. Species can
substances as an energy source to fuel ach and their small intestine (where opt to feed selectively on only the high-
their own activities. As the bacteria con- most nutrients are absorbed ) and into est-quality plant foods. But because
sume the Þber, they release gases called the hindgut (the cecum and colon). Once quality items are rare and very patchily
volatile fatty acids. These gases pass Þber reaches the hindgut, cellulolytic distributed in tropical forests, this strat-
through the stomach wall into the colo- bacteria may be able to degrade some egy requires the adoption of behaviors
bine bloodstream, where they provide of it. But, for most primates, eating co- that help to minimize the costs of pro-
88 SCIENTIFIC AMERICAN August 1993 Copyright 1993 Scientific American, Inc.

4. chimpanzee ets. Howler and spider monkeys, which meant food had a longer distance to
diverged from a common ancestor, are travel and that signiÞcantly more bulk
alike in that they are about the same could be retained.
size, have a simple, unsacculated stom- Collectively, these results implied that
ach, are totally arboreal and eat an al- howlers could survive on leaves because
HOMINIDS most exclusively plant-based diet, con- they were more adept at fermenting Þ-
APES
sisting for the most part of fruits and ber in the cecum and colon. They pro-
leaves. But my Þeldwork showed that cessed food slowly, which gave bacte-
FORESTS IN the foundation of the howler diet in the ria in the capacious hindgut a chance to
FIRE USE SOME AREAS Barro Colorado forest was immature produce volatile fatty acids in quanti-
GIVE WAY leaves, whereas the foundation of the ty. Experiments I later carried out with
TO SAVANNA
spider monkey diet was ripe fruits. Richard McBee of Montana State Uni-
Most of the year howlers divided their versity conÞrmed that howlers may ob-
daily feeding time about equally between tain as much as 31 percent of their re-
new leaves and fruits. But during season- quired daily energy from volatile fatty
al low points in overall fruit availabili- acids produced during fermentation.
ty, they ate virtually nothing but leaves. In contrast, spider monkeys, by pass-
In contrast, spider monkeys consumed ing food more quickly through their
ripe fruits most of the year, eating only shorter, narrower colons, were less eÛ-
small amounts of leaves. When fruits cient at extracting energy from the Þ-
became scarce, spider monkeys did not ber in their diet. This speed, however,
simply Þll up on leaves as the howlers enabled them to move masses of food
did. Their leaf intake did increase, but through the gastrointestinal tract each
they nonetheless managed to include day. By choosing fruits, which are high-
considerable quantities of fruit in the ly digestible and rich in energy, they at-
diet. They succeeded by carefully seek- tained all the calories they needed and
ANGIOSPERM FORESTS
SPREAD ACROSS THE EARTH ing out all fruit sources in the forest ; some of the protein. They then supple-
they even resorted to consuming palm mented their basic fruit-pulp diet with
nuts that had not yet ripened. a few very select young leaves that sup-
These observations raised a number plied the rest of the protein they re-
gave way to true primates, beginning of questions. I wanted to know how quired, without an excess of Þber.
with the prosimians. Our own genus, howlers obtained enough energy dur- Hence, howler monkeys never devote
Homo, emerged during the Pliocene. Ex- ing months when they lived exclusively themselves exclusively to fruit, in part
act dates of radiations are debatable. on leaves. As already discussed, much because their slow passage rates would
of the energy in leaves is bound up in probably prevent them from processing
Þber that is inaccessible to the diges- all the fruit they would need to meet
curing these resources. The strategy tive enzymes of primates. Further, why their daily energy requirement. And the
would be greatly enhanced by a good did howlers eat considerable foliage amount of fruit they could consume cer-
memory. For example, an ability to re- even when they had abundant access tainly would not provide enough pro-
member the exact locations of trees that to ripe fruits? By the same token, why tein. Conversely, spider monkeys must
produce desirable fruits and to recall did spider monkeys go out of their way eat fruit because their digestive tract is
the shortest routes to those trees would to Þnd fruit during periods of scarcity ; ill equipped to provide great amounts of
enhance foraging eÛciency by lower- what stopped them from simply switch- energy from fermenting leaves; eÛcient
ing search and travel costs. So would ing to leaves, as howlers did? And how fermentation requires that plant matter
knowledge of when these trees were did spider monkeys meet daily protein be held in the gut for some time.
likely to bear ripe fruits. Reliance on needs with their fruit-rich diet? ( Recall
B
memory, with its attendant beneÞts, that fruits are a poor source of protein.) y luck, I had chosen to study two
might then select for bigger brains hav- Because howler and spider monkeys species that fell at opposite ends
ing more area for storing information. are much alike externally, I speculated of the continuum between slow
Of course, these two basic evolution- that some internal feature of the two and rapid passage of food. It is now
ary strategiesÑthe morphological and speciesÑperhaps the structure of the clear that most primate species can be
behavioralÑare not mutually exclusive, gut or the eÛciency of digestionÑmight ranked somewhere along this continu-
and species vary in the extent to which be inßuencing these behaviors. And, in- um, depending on whether they tend to
they favor one or the other. As a group, deed, studies in which I fed fruits and maximize the eÛciency with which they
however, primates have generally de- leaves to temporarily caged subjects re- digest a given meal or maximize the vol-
pended most strongly on selective feed- vealed that howler monkeys digested ume of food processed in a day. This re-
ing and on having the brain size, and food more slowly than did spider mon- search further shows that even with-
thus the wit, to carry oÝ this strategy keys. Howlers began eliminating colored out major changes in the design of the
successfully. Other plant-eating orders, plastic markers embedded in foods an digestive tract, subtle adjustments in
in contrast, have tended to focus heavi- average of 20 hours after eating. In con- the size of diÝerent segments of the
ly on morphological adaptations. trast, spider monkeys began eliminat- gut can help compensate for nutritional
I gained my Þrst insights into the ing these harmless markers after only problems posed by an animalÕs dietary
evolutionary consequences of selective four hours. Examining the size of the choices. Morphological compensations
feeding in primates in the mid-1970s, digestive tract in the two species then in the digestive tract can have their
when I noticed that howler monkeys revealed how these diÝerent passage drawbacks, however, because they may
and black-handed spider monkeys (Ate- rates were attained. In howler monkeys make it difÞcult for a species to alter
les geoffroyi )Ñtwo New World primate the colon was considerably wider and its dietary habits should environmental
speciesÑfavored markedly diÝerent di- longer than in spider monkeys, which conditions change suddenly.
Copyright 1993 Scientific American, Inc. SCIENTIFIC AMERICAN August 1993 89

5. HIGH FIBER CONTENT
LARGE DISTANCES BETWEEN LIKE TREES Cell walls of plant parts, especially mature leaves,
Trees bearing a favored food are often distributed patchily can contain much fiber (inset), which
is resistant to digestion
LIMITED AVAILABILITY
Many favored items are available only
part of the year, some for only hours
CHEMICAL DEFENSES
INCOMPLETE NUTRITION Potential foods often contain chemicals that
Few plant foods are both high in energy (calories) are ill tasting or poisonous or that interfere
and high in critically needed protein, vitamins and minerals with digestion of other foods
These digestive Þndings fascinated monkeys comb the forest for fruit by very selectively, favoring the highest-
me, but a comparison of brain size in dividing into small, changeable groups. quality plant partsÑfor instance, even
the two species yielded one of those Òeu- Expanded mental capacity would have primates that eat leaves tend to choose
rekasÓ of which every scientist dreams. I helped them to recognize members of very immature leaves or only the low-
examined information on the brain sizes their particular social unit and to learn Þber tips of those leaves.
of howler and spider monkeys because the meaning of the diÝerent food-relat-
H
the spider monkeys in Panama seemed ed calls through which troop members aving uncovered these links be-
ÒsmarterÓ than the howlersÑalmost hu- convey over large distances news of pal- tween dietary pressures and evo-
man. Actually, some of them remind- atable items. Howler monkeys, in con- lution in nonhuman primates, I
ed me of my friends. I began to wonder trast, would not need such an extensive became curious about the role of such
whether spider monkeys behaved dif- memory, nor would they need so com- pressures in human evolution. A review
ferently because their brains were more plex a recognition and communication of the fossil record for the hominid
like our own. My investigations showed system. They forage for food as a cohe- familyÑhumans and their precursorsÑ
that, indeed, the brains of howler and sive social unit, following well-known provided some intriguing clues.
spider monkeys do diÝer, even though arboreal pathways over a much smaller Australopithecus, the Þrst genus in
the animals are about the same size. home range. our family, emerged in Africa more than
(Same-sized animals generally have like- If I was correct that the pressure to 4.5 million years ago, during the Plio-
sized brains.) The spider monkey brain obtain relatively diÛcult-to-Þnd, high- cene. As is true of later hominids, they
weighs about twice that of howlers. quality plant foods encourages the de- were bipedal, but their brains were not
Now, the brain is an expensive organ velopment of mental complexity (which appreciably larger than those of todayÕs
to maintain; it usurps a disproportion- is paid for by greater foraging eÛcien- apes. Hence, selection had not yet be-
ate amount of the energy (glucose) ex- cy), I would expect to Þnd similar diÝer- gun to favor a greatly enlarged brain in
tracted from food. So I knew natural se- ences in brain size in other primates. our family. The fossil record also indi-
lection would not have favored develop- That is, monkeys and apes who concen- cates Australopithecus had massive mo-
ment of a large brain in spider monkeys trated on ripe fruits would have larg- lar teeth that would have been well suit-
unless the animals gained a rather pro- er brains than those of their leaf-eat- ed to a diet consisting largely of tough
nounced beneÞt from the enlargement. ing counterparts of equal body size. To plant material. Toward the end of the
Considering that the most striking dif- pursue this idea, I turned to estimates Pliocene, climate conditions began to
ference between howler and spider mon- of comparative brain sizes published change. The next epoch, the Pleistocene
keys is their diets, I proposed that the by Harry J. Jerison of the University of (lasting from about two million to 10,-
bigger brain of spider monkeys may California at Los Angeles. To my ex- 000 years ago), was marked by repeat-
have been favored because it facilitated citement, I found that those primate ed glaciations of the Northern Hemi-
the development of mental skills that species that eat higher-quality, more sphere. Over both epochs, tropical for-
enhanced success in maintaining a diet widely dispersed foods generally have a ests shrank and were replaced in many
centered on ripe fruit. larger brain than do their similar-sized areas by savanna woodlands.
A large brain would certainly have counterparts that feed on lower-quality, As the diversity of tree species de-
helped spider monkeys to learn and, more uniformly distributed resources. creased and the climate became more
most important, to remember, where As I noted earlier, primates typical- seasonal, primates in the expanding sa-
certain patchily distributed fruit-bear- ly have larger brains than do other vanna areas must have faced many new
ing trees were located and when the mammals of their size. I believe the dietary challenges. In the Pleistocene the
fruit would be ready to eat. Also, spider diÝerence arose because primates feed last species of AustralopithecusÑwhich
90 SCIENTIFIC AMERICAN August 1993 Copyright 1993 Scientific American, Inc.

6. READILY AVAILABILITY
ACCESSIBLE PROTEIN FIBER CHEMICAL ON A GIVEN
CALORIES DEFENSES TREE
FLOWERS Moderate Moderate Low to Variable Fewer than
to high moderate three months
FRUITS High Low Moderate Low Fewer than
three months
YOUNG Low High Moderate Moderate Half the year
LEAVES
MATURE Low Moderate High Moderate Almost
LEAVES year-round
MANY CHALLENGES can deter primates in the tropical forest from obtaining the
calories and mix of nutrients they need from plant foods (left). Because most such
foods are inadequate in one way or another, animals must choose a variety of items
each day. The chart at the right loosely reßects the relative abundance of desirable
( green) and problematic ( yellow ) components in a mouthful of common foods. It
also indicates the typical availability of these foods on any given tree.
by then had truly massive jaws and mo- tions in living primates between larger dea (apes and humans), an increase in
larsÑwent extinct. Perhaps those spe- brains and a high-quality dietÑthis in- body size combined with decreased di-
cies did so, as my colleague Montague crease also points to the conclusion that etary quality leads to a slow-moving,
W. Demment of the University of Cali- the behavioral solution was to concen- fairly sedentary and unsociable ape. Yet
fornia at Davis speculates, because they trate on high-quality foods. In fact, I sus- our Homo ancestors apparently were
were outcompeted by the digestively pect early humans not only maintained mobile and sociableÑmore resembling
specialized ungulates (hoofed animals). dietary quality in the face of changing the lively, social and communicative
The human, or Homo, genus emerged environmental conditions but even im- chimpanzee. Unlike orangutans and go-
during the Pliocene. The Þrst species proved it. rillas, chimpanzees feed preferentially
of the genus, H. habilis, was similar in Expansion of the brain in combination on high-quality, energy-rich ripe fruits.
body size to Australopithecus but had a with growth in body size and a reduc- Likewise, the reduction in the molars
notably larger brain. This species was tion in the dentition supports the no- and premolars shows that the texture of
replaced by the even larger-brained H. tion of a high-quality diet for a couple of foods we ate had somehow been altered
erectus and then, in the Pleistocene, by reasons. When one examines present- such that the dentition no longer had so
H. sapiens, which has the biggest brain day orangutans and gorillas, it becomes much work to do. In other words, either
of all. In parallel with the increases in clear that in our superfamily, Hominoi- these early humans were eating diÝer-
brain size in the Homo genus, other
anatomic changes were also occurring.
The molar and premolar teeth became
smaller, and stature increased.
To me, the striking expansion of brain ALKALINE
size in our genus indicates that we be- FORESTOMACH
CONTAINING
came so successful because selection CELLULOLYTIC
ampliÞed a tendency inherent in the pri- BACTERIA ACID
Colobus guereza Cercopithecus STOMACH
mate order since its inception: that of
pygerythrus
using brain power, or behavior, to solve
dietary problems. Coupled with the ana-
tomic changesÑand with the associa-
ACID
STOMACH
DIGESTIVE TRACT of colobine monkeys,
such as that in Colobus guereza (left),
is specialized : the stomach consists of CECUM
two distinct compartments instead of the
single chamber found in vervet monkeys COLON
(right) and most other primates. One of
those compartmentsÑthe forestomachÑ
is designed to extract more energy from SMALL INTESTINE
Þber than would normally be obtainable.
Colobine monkeys can thus survive on
a more Þbrous diet than can other pri-
mates of similar size.
Copyright 1993 Scientific American, Inc. SCIENTIFIC AMERICAN August 1993 91

7. ent (less Þbrous, easier-to-chew) foods cut through tough hides and to break retain a mental map of plant food sup-
than was Australopithecus, or they were bones for marrow. To incorporate meat plies but also having knowledge of how
somehow processing foods to remove into the diet on a steady basis and also to procure or transform such supplies.
material that would be hard to chew to amass energy-rich plant foods, our In addition, survival now required an
and digest. Indeed, stone tools found ancestors eventually developed a truly ability to recognize that a stone tool
with fossil remains of H. habilis indicate novel dietary approach. They adopted could be fashioned from a piece of a
that even the earliest members of our a division of labor, in which some indi- rock and a sense of how to implement
genus were turning to technology to viduals specialized in the acquisition of that vision. And it required the capacity
aid in the preparation of dietary items. meat by hunting or scavenging and oth- to cooperate with others (for instance,
The probability that hominids per- er individuals specialized in gathering to communicate about who should run
sisted in seeking energy-rich foods plants. The foods thus acquired were ahead of a hunted zebra and who be-
throughout their evolution suggests an saved instead of being eaten on the hind), to defer gratiÞcation (to save food
interesting scenario. As obtaining cer- spot ; they were later shared among the until it could be brought to an agreed
tain types of plant foods presumably entire social unit to assure all members site for all to share) and both to deter-
became more problematic, early hu- of a balanced diet. mine oneÕs fair portion and to ensure
mans are thought to have turned in- Survival of the individual thus came that it was received. Such demands un-
creasingly to meat to satisfy their pro- to depend on a number of technologi- doubtedly served as selective pressures
tein demands. One can readily envi- cal and social skills. It demanded not favoring the evolution of even larger,
sion their using sharp stone ßakes to only having a brain able to form and more complex brains.
Similarly, spoken communication may
at Þrst have helped facilitate the coop-
eration needed for eÛcient foraging and
other essential tasks. Gradually, it be-
came elaborated to smooth the course
of social interactions.
I
n other words, I see the emergence
and evolution of the human line as
stemming initially from pressures to
acquire a steady and dependable supply
of very high quality foods under envi-
ronmental conditions in which new di-
etary challenges made former foraging
behaviors somehow inadequate. Spe-
cialized carnivores and herbivores that
abound in the African savannas were
evolving at the same time as early hu-
mans, perhaps forcing them to become
a new type of omnivore, one ultimately
dependent on social and technological
innovation and thus, to a great extent,
on brain power. Edward O. Wilson of
Harvard University has estimated that
SPIDER MONKEY HOWLER MONKEY for more than two million years (until
(Ateles geoffroyI) (Alouatta palliata) about 250,000 years ago), the human
TYPICAL DIET TYPICAL DIET brain grew by about a tablespoon every
Fruits: 72 percent Fruits: 42 percent 100,000 years. Apparently each table-
Leaves: 22 percent Leaves: 48 percent spoonful of brain matter added in the
Flowers: 6 percent Flowers: 10 percent genus Homo brought rewards that fa-
WEIGHT WEIGHT vored intensiÞcation of the trend toward
Six to eight kilograms Six to eight kilograms social and technological advancement.
BRAIN SIZE BRAIN SIZE Although the practice of adding some
107 grams 50.3 grams amount of meat to the regular daily in-
DAY RANGE DAY RANGE take became a pivotal force in the emer-
915 meters 443 meters gence of modern humans, this behavior
DIGESTIVE FEATURES DIGESTIVE FEATURES does not mean that people today are bio-
Small colon Large colon logically suited to the virtually Þber-free
Fast passage of food Slow passage of food diet many of us now consume. In fact,
through colon through colon in its general form, our digestive tract
does not seem to be greatly modiÞed
from that of the common ancestor of
SPIDER MONKEY (left) is a fruit specialist, whereas the howler monkey (right) eats
apes and humans, which was undoubt-
large quantities of leaves. The author proposes that diet played a major role in shap-
ing the different traits of the two like-sized species, which shared a common ances-
edly a strongly herbivorous animal.
tor. Natural selection favored a larger brain in spider monkeys, in part because en- Yet as of the mid-1980s no studies
hanced mental capacity helped them remember where ripe fruits could be found. had been done to Þnd out whether the
And spider monkeys range farther each day because in any patch of forest, ripe gut functions of modern humans were
fruits are less abundant than leaves. The digestive traits of spider and howler mon- in fact similar to those of apes. It was
keys promote eÛcient extraction of nutrition from fruits and leaves, respectively. possible that some functional diÝerenc-
92 SCIENTIFIC AMERICAN August 1993 Copyright 1993 Scientific American, Inc.

8. es existed, because anatomic evidence spond to this decrease by increasing the
had shown that despite similarity in rate at which food moves through the
the overall form of the digestive tract, tract. This response permits a greater
modern humans have a rather small quantity of food to be processed in a
tract for an animal of their size. They given unit of time; in so doing, it en-
also diÝer from apes in that the small ables the feeder to make up for reduced
intestine accounts for the greatest frac- quality by taking in a larger volume of
tion of the volume of the human diges- food each day. ( Medical research has
tive tract ; in apes the colon accounts uncovered another beneÞt of fast pas-
for the greatest volume. sage. By speeding the ßow of food
To better understand the kind of diet through the gut, Þber seems to prevent
for which the human gut was adapted, carcinogens from lurking in the colon
Demment and I decided to compare hu- so long that they cause problems.)
man digestive processes with those of If the human digestive tract is indeed
the chimpanzee, our closest living rela- adapted to a plant-rich, Þbrous diet, then
tive. We hoped to determine whether, this discovery lends added credence to
over the course of their respective evo- the commonly heard assertion that peo-
lutionary histories, humans and chim- ple in highly technological societies eat
panzees had diverged notably in their too much reÞned carbohydrate and too
abilities to deal with Þber. ( We were little Þber. My work oÝers no prescrip-
greatly encouraged in this eÝort by the tion for how much Þber we need. But
late Glynn Isaac, who was then at the certainly the small amount many of us
BURGER AND FRIES, like many popular
University of California at Berkeley.) consume is far less than was ingested
foods eaten in the U.S., bear little resem-
The feeding habits of chimpanzees by our closest human ancestors.
blance to the fruits and leaves most pri-
are well known. Despite their skill in mates have emphasized since the incep- More recently, my colleagues and I
capturing live prey (particularly mon- tion of our order. Early humans, too, are have analyzed plant parts routinely
keys), these apes actually obtain an es- thought to have consumed large quan- eaten by wild primates for their con-
timated 94 percent of their annual diet tities of plant foods. Hence, modern di- tent of various constituents, including
from plants, primarily ripe fruits. Even ets often diverge greatly from those to vitamin C and pectin. Pectin, a high-
though the fruits chimpanzees eat tend which the human body may be adapted. ly fermentable component of cell walls,
to be rich in sugar, they contain far less is thought to have health beneÞts for
pulp and considerably more Þber and humans. Our results suggest that diets
seeds than do the domesticated fruits pull my hair, throw fecal matter and eaten by early humans were extremely
sold in our supermarkets. Hence, I cal- generally let me know they were under- rich in vitamin C and contained nota-
culated that wild chimpanzees take in whelmed by our experimental cuisine. ble pectin. Again, I do not know wheth-
hundreds of grams of Þber each day, er we need to take in the same pro-
O
much more than the 10 grams or less ur results showed that the chim- portions of these substances as wild
the average American is estimated to panzee gut is strikingly similar primates do, but these discoveries are
consume. to the human gut in the eÛcien- provocative.
Various excellent studies, including a cy with which it processes Þber. More- To a major extent, the emergence of
Þber project at Cornell University, had over, as the fraction of Þber in the diet modern humans occurred because nat-
already provided much information rises (as would occur in the wild during ural selection favored adaptations in
about Þber digestion by humans. At seasonal lulls in the production of fruits our order that permitted primates to
one time, it was believed that the hu- or immature leaves), chimpanzees and focus their feeding on the most energy-
man digestive tract did not possess mi- humans speed the rate at which they dense, low-Þber diets they could Þnd.
crobes capable of degrading Þber. Yet pass food through the digestive tract. It seems ironic that our lineage, which
bacteria in the colons of 24 male col- These similarities indicate that as in the past beneÞted from assiduously
lege students at Cornell proved quite quality begins to decline in the natural avoiding eating too much food high in
eÛcient at fermenting Þber found in environment, humans and chimpanzees Þber, may now be suÝering because we
a variety of fruits and vegetables. At are evolutionarily programmed to re- do not eat enough of it.
their most eÝective, the microbial pop-
ulations broke down as much as three
quarters of the cell-wall material that FURTHER READING
the subjects ingested; about 90 percent ECOLOGY OF ARBOREAL FOLIVORES. Edited Theory of Human Food Habits. Edited by
of the volatile fatty acids that resulted by G. Gene Montgomery. Smithsonian In- Marvin Harris and Eric B. Ross. Temple
were delivered to the bloodstream. stitution Press, 1978. University Press, 1987.
Following the example of the Cornell DISTRIBUTION PATTERNS OF TROPICAL DIGESTION AND PASSAGE KINETICS OF
study, Demment and I assessed the ef- PLANT FOODS AS AN EVOLUTIONARY STIM- CHIMPANZEES FED HIGH AND LOW-FIBER
Þciency of Þber breakdown in chim- ULUS TO PRIMATE MENTAL DEVELOPMENT. DIETS AND COMPARISON WITH HUMAN
K. Milton in American Anthropologist, Vol. DATA. K. Milton and M. W. Demment in
panzees fed nutritious diets contain-
83, No. 3, pages 534Ð548; September Journal of Nutrition, Vol. 118, No. 9, pag-
ing varying amounts of Þber. Demment 1981. es 1082Ð1088; September 1988.
handled the statistical analyses, and I FOOD CHOICE AND DIGESTIVE STRATEGIES FORAGING BEHAVIOUR AND THE EVOLU-
collected raw data. How dry that sounds OF TWO SYMPATRIC PRIMATE SPECIES. TION OF PRIMATE INTELLIGENCE. K. Milton
in comparison to the reality of the ex- K. Milton in American Naturalist, Vol. in Machiavellian Intelligence: Social Exper-
perience! At the Yerkes Primate Center 117, No. 4, pages 496Ð505; April 1981. tise and the Evolution of Intellect in Mon-
in Atlanta, I whiled away the summer PRIMATE DIETS AND GUT MORPHOLOGY: IM- keys, Apes, and Humans. Edited by Rich-
with six extremely cross chimpanzees PLICATIONS FOR HOMINID EVOLUTION. K. ard Byrne and Andrew Whiten. Oxford
Milton in Food and Evolution: Toward a University Press, 1988.
that never missed an opportunity to
Copyright 1993 Scientific American, Inc. SCIENTIFIC AMERICAN August 1993 93

9. 94 SCIENTIFIC AMERICAN Month 1993 Copyright 1993 Scientific American, Inc.