The Empty Forest
Author(s): Kent H. Redford
Source: BioScience, Vol. 42, No. 6 (Jun., 1992), pp. 412-422
Published by: University of California Press on behalf of the American Institute of Biological
Sciences
Stable URL: http://www.jstor.org/stable/1311860
Accessed: 19/08/2010 17:59
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=aibs.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
American Institute of Biological Sciences and University of California Press are collaborating with JSTOR to
digitize, preserve and extend access to BioScience.
http://www.jstor.org
The
Empty
Forest
Many large animals are already ecologically extinct in vast areas
of neotropical forest where the vegetation still appears intact
Kent H. Redford
he world conservationcommunity has focused much of
its attention on the plight of
tropical forests. Many authors have
lamentedthe loss of forest cover and
the destructionof the forestand speculated on the extent of the tropical
forest left intact. Throughoutthe discussion, tall, majestic, tropical trees
are used as a symbol for the complete
set of animaland plant speciesfound
in tropicalforests.Treesare also being
used by some conservationbiologists,
parkplanners,and othersto represent
the entiretropicalforestbiota and as a
measureof conservationworth.
The presenceof soaring,buttressed
tropical trees, however, does not
guarantee the presence of resident
fauna. Often trees remainin a forest
that humanactivitieshave emptiedof
many of its large animals. The absence of these animals has profound
implications,one of which is that a
forest can be destroyed by humans
from within as well as from without.
Until recently,human influenceon
tropical forests through such activities as burning,swidden agriculture,
and hunting was regardedby ecologists as of such low impactthat it was
negligible,as importantbut confined
to areas of human settlement,or as
confined to rapacious colonizers deKent H. Redford is the director of the
Programfor Studiesin Tropical Conservation and an associate professorin the
Center for Latin American Studies and
Departmentof Wildlife and Range Science, Universityof Florida, Gainesville,
FL 32611. ? 1992 AmericanInstituteof
BiologicalSciences.
412
With few exceptions, researchers
have concentratedon directalteration
We must not let a
of vegetation,not discussingthe ways
in which human activities have afforest full of trees fool
fected the animals of tropical forest
us into believing all
ecosystems. In this article, I expand
focus to include defaunationof
the
is well
tropical forests, concentratingon the
forests of the Amazon basin, and I
show that the long-termpreservation
stroying the forest from the outside. of tropical forest vegetation will not
In any case, ecologists looked for be possible if the forest fauna is not
study sites that would allow for ex- also preserved.
aminationof "natural"processesuncontaminated by anthropogenic ef- Indirectdefaunation
fects. Data from botany,archaeology,
and anthropologycollected in many Humans can devastate a fauna by
parts of the world are showing, how- indirector direct means. Indirectdeever, that anthropogeniceffects are faunationis the destructionof a fauna
ubiquitousand that the sought-after through human activity not aimed
virgin habitat may not exist. Flenley specificallyat animals.In tropicalfor(1979), for example,has documented ests, habitat destructionis the most
widespreadhumaneffectson tropical common of these practices-not surforests throughoutthe equatorialre- prisingly,many forest animalscannot
survive without forest. A less-oftengions.
The relativelyrecent arrivalof hu- consideredtype of habitatdestruction
mans in the western hemispherehas occurswhen animalsare absentfrom
not lessened the overall impact our an area of otherwiseexcellenthabitat
species has had on neotropical for- becausesome criticalarea elsewhere,
ests. From the forests of Mexico such as a nesting beach, was dethrough Panama, and the montane stroyed.This problemaffectsanimals
forests of Colombia to Ecuador,sci- including migratory birds, beachentists have documentedthe ways in nesting turtles,and white-lippedpecwhichpre-Columbianhumansaltered cary (Tayassupecari)herds.
There are many other types of inthe presence,extent, and structureof
forests. The forests of the Amazon direct defaunation.One of the most
basinwere also extensivelyalteredby important is probably the effect of
humanactivities.In fact, Balee(1989) forest-extraction activities by huhas recently suggested that at least mans. For example, logging can re11.8% of the terrafirmeforestsof the move fruit-bearingtrees and destroy
Brazilian Amazon, almost 400,000 nesting and other criticalareas.
Less obvious are the effects that
km2, show continuingeffects of past
stem from the much-publicizedexhuman interference.
BioScienceVol. 42 No. 6
I
tractionof forest fruits and nuts. Almost without exception, the fruits
collectedfor sale are fruits also eaten
by largebirdsand mammals(Redford
et al. in press).However,the extent to
which animalspeciescan survivewith
reducedavailabilityof fruitsand nuts
is unknown.
A recent article by Vasquez and
Gentry (1989) documentedthe huge
numberof fruits from wild trees collected by humans for sale in Amazonian markets.One of the majorfruits
in this marketwas that from Mauritia
palms-the only food of the macaw
Ara manilata (Roth 1984) and the
most importantfruitin the diet of the
tapir (Bodmer1990). The enormous
quantities of Brazil nuts removed
from a patch of tropical forest undoubtedlyaffectthe animalsthat otherwise would have fed on those nuts.
The extractive activities also remove
nutrientsfrom the ecosystem.
Indirectdefaunationcan also take
place through the effects of subsistenceor commercialhuntingand fishing that remove potential prey from
tropical forests, thereby affecting
predators, scavengers, and the animals that depend on them (Thiollay
1984). As Emmons (1987) and Jorgenson and Redford (in press) have
pointed out, every majorprey species
of the jaguaris intensivelyhunted by
humans.In these cases, it seems clear
that the removalfrom tropicalforests
of food for human consumption reduces the capacityof those forests to
supportmany animals.
Finally,manyof the by-productsof
modern human activities are important contributorsto indirectdefaunation. These byproductsinclude mercury and sediment contaminationof
fish (Martinelliet al. 1988); smoke,
which especially affects pollinators
(Lovejoyet al. 1984); and an increase
of edge habitatin the forest (Malcolm
1991).
Direct defaunation
The indirecteffectsof human activity
on defaunation, important as they
are, have arisen chiefly in recent decades. The effects of direct defaunation, the deliberate killing of animals,
has a much longer history in the Amazon forests-a history that coincides
with the presence of humans in the
area. Direct defaunation can be diJune 1992
INDIANS
COLONISTS
I
I
Cebus apella
4B
sciurids
Tayassu pecari
Agouti paca
Dasypus novemcinctus i
Tayassu tajacu
Dasyprocta &
Myoprocta
(d)
Alouatta spp.
U
Ateles spp.
Cebus spp.
I
Tamandua spp.
eE
Mazama spp.
Bradypus tridactyla
h
;CC
I
I
I
Tapirus terrestris
XmFigure1. Importanceof mammalsto contemporaryIndianand colonist hunters(only
consideredwere those species found in a minimumof five Indian studies and three
coloniststudies).Barsdenotethe numberof individualsof that taxon killedper hunter
per year.To give an idea of scale,therewere approximately2.5 individualCebusapella
monkeys killed by Indiansper consumerper year and approximately0.05 Tapirus.
(Data from Redfordand Robinson 1987.)
I
vided into two categories:subsistence have lived for decadesin tropicalforests but have recourse to domestic
huntingand commercialhunting.
animals.It is also of lesserimportance
Subsistencehunting.In manypartsof to colonists recentlyarrivedto Amathe world, wildlife serves as a major zonian forests, who frequently are
source of food for local peoples. In unfamiliarwith huntingand have acLatinAmerica,game is a vital protein cess to other sourcesof meat.
A wide varietyof wildlifeis hunted
and fat source to many groups living
outside of urban areas. As a general for food by humans. The Maraca
rule, wildlife is most important to Indians of Colombia, for example,
Indian groups that depend on game take at least 51 species of birds, inmeat for subsistence. It is of lesser cluding 10 species of hummingbirds
importance,thoughstill important,to (Ruddle 1970). Hunters generally
settlers of European descent who take more mammals than birds and
413
INDIANS
COLONISTS and has shifted from group to group
I
Penelope spp.
Ramphastos spp.
Crax spp.
th
I
Psophia spp.
4b
Crypturellus spp.
EDIBLE PRODUCTS. As
k
Amazona spp.
I
Ara spp.
I
Mitu spp.
I
Figure 2. Importanceof birds to contemporaryIndian and colonist hunters (only
consideredwere those species found in a minimumof five Indian studies and two
coloniststudies).Barsdenotethe numberof individualsof that taxon killedper hunter
per year. To give an idea of scale, there were approximately0.9 individualPenelope
guanskilledby Indiansper consumerperyearand approximately0.09 Ara. (Datafrom
Redfordand Robinson 1987.)
morebirdsthan reptiles(Redfordand
Robinson 1987).
Throughout Amazonia and Latin
America, indigenous hunters usually
kill only a few of the many types of
mammals and birds present. Of the
mammals, monkeys, peccaries, deer,
armadillos, and large rodents like
paca and capybara are frequently
hunted; and, of the birds, the most
common prey are guans and curas-
sows, toucans, trumpeters, and macaws. The number of species hunted
by nonnative peoples is even more
restricted. Figures 1 and 2 show the
differences in the range of mammal
and bird species taken by Indians and
colonists.
The numbers of animals taken by
subsistence hunters can be large. In
less than a year, the 230 inhabitants
of three Waorani villages in Ecuador
killed 3165 mammals, birds, and reptiles (Yost and Kelley 1983). This
total included 562 woolly monkeys
(Lagothrix lagothricha), 313 Cuvier's
toucan (Ramphastos cuvieri), and
152 white-lipped peccary. Not all
subsistence hunting is of this intensity.
414
as market demand and availability
have changed (Figure3; c.f. Redford
and Robinson 1991). Even before
Europeans arrived in the Americas,
animals and their products were
traded. Among the Incas, adult
caiman and anacondas were transported from the Amazonian lowlands up to the Andeancity of Cuzco
for use in menageries (Lathrap
1975). Nonetheless, trade in wildlife
did not assume major proportions
until Europeansarrived.
I have estimated the number of
mammals killed in one year by the
rural population of AmazonianBrazil. In 1980, there were an estimated
2,847,000 people living outside of
cities in an area of 3,581,180 km2
(FIBGE1982). I multipliedthis numberof consumersby the per capitaper
annum consumption values derived
fromstudiesof colonisthunting(Redford and Robinson1987). The resulting figure,14 millionindividualmammals killed each year, suggests the
staggeringextent of subsistencehunting. Adding birds and reptiles, the
numberof game animals killed each
year in Amazonian Brazil probably
reaches19 millionanimals.The number of animals fatally wounded or
killed could reach57 million animals
a year.
Commercial hunting. The second
major cause of direct defaunationis
commercial hunting. The killing of
animals in Amazon forests by Europeans for commercial purposes has
been going on since soon after their
discoveryof the continent.This trade
has involved many different species
early as the
seventeenthcentury, the commercial
harvestingof manatees for meat began (Redford and Robinson 1991).
Until the mid-1900s, the averageannual catch of these Trichechusinunguis in the Amazon was at least several thousand animals. Two other
importantcommercializedsourcesof
meat are caiman and river turtles.
Caimanof severalspecies,but principally of the genus Caiman,have been,
and still are, an importantsource of
meat in some areas of the Amazon
basin, with the currenttrade in meat
estimated at 21,500 to 32,000 animals annually.
The first Europeanto navigatethe
Amazon River found many Indian
villageswith hundredsof pennedturtles (mostly Podocnemis expansa).
Despite ferocious exploitation for
meat and eggs, female P. expansa
continued to gather near nesting
beaches and be plentiful enough
through the 1850s to impede river
trafficon the Madeira.
Turtle eggs have been heavily exploited for industrialand nutritional
purposes. In the Amazon basin, the
eggs of P. expansa were so abundant
and in such great demand that an
industry developed to process them.
Oil from the eggs was used for cooking and lighting, and as early as the
eighteenthcentury royal decree controlled the lucrativeharvestin Brazil.
In 1719, 192,000 pounds of oil,
equaling approximately 24 million
eggs, were produced from the upper
Amazon;as late as the 1860s, at least
48 million eggs yearlywere harvested
to supply the industry. In many areas, giant riverturtles have been virtually eliminated, and there is still
heavy predation on their eggs wherever and whenever they can be
found.
BioScience Vol. 42 No. 6
,
Although no longer available on
the scale once observed,game is still
readily obtained in many local markets. Castro et al. (1975-1976) reportedthe meat of 24 speciesof wildlife, includingsix speciesof primates,
for sale in the markets of Iquitos,
Peru. They estimate that the inhabitants of the Peruvian departmentof
Loreto, which includes the city of
Iquitos, kill 370,000 monkeys annually for consumptionand sale.
The differencebetweencommercial
and subsistencehunting is becoming
increasinglyblurred.For example, in
a study of the ungulateharvestin one
watershednear Iquitos,Peru,lumbermen hunting to supply their camps
with food accounted for 51% of the
ungulate harvest, illegal commercial
hunters for 11%, and subsistence
huntersfor only 38% (Bodmeret al.
1988).
LEATHER.
The earliest European
In the 1960s, apparentlyin response
to overexploitationof jaguarand the
case with leather,large-scalecommer- concomitantdecreasein numbers,the
cialization of skins was brought to cat trade shifted to the smaller speLatin America by the Europeans. cies.
Both tradesare directedat the luxury
The period between the end of the
marketsin Europe,Japan,and North Second World War and the early
America.The skin trade has always 1970s was the golden era of the trade
focused on a relatively few species. in skins originatingfrom the Amazon
The trade in skins originatingin the Basin (McGrath 1986). In the 20
Amazon basin has been well docu- years since 1946, the Amazon River
mentedand has concentratedon giant port of Iquitos, Peru, exported
otter (Pteronurabrasiliensis),riverot- 22,644 giant otter skins, 90,574 river
ter (Lutra longicaudis),jaguar (Pan- otter skins, 12,704 jaguarskins, and
thera onca), and ocelot (Felisparda- 138,102 ocelot skins. The value of
lis). Much smaller numbers of Felis skins caused people to move into
wiedii and Felis tigrinaskins are also sparsely inhabited areas and devote
traded.
tremendous effort to commercial
The trade in cat skins began with hunting. For example, one family of
jaguarsat the end of the last century. hunters in the EcuadorianAmazon
Caiman sclerops skins.
MAMMALIANSKINS.Just as was the
Date
1500
;V
16Rn
I.IV
1700
I
,-~V
1Ivvv 00 1950 1 990
v
1R80
Ivv~
1,.v
Iv
commercial exploitation of wildlife
Product
for nonedibleproductswas the use of
that
made
from
~~~~~~~~~~~~~~~~...
leather, particularly
meat
MANATEES
ii~?;.;~.
~
hides
deerhides. Most of the recentmarket
for leatherhas been luxuryitems such
meat
;;;.,
.;:2,;;-;
s~~~~~~~~~~~~~~~.:.:-.:.....
eggs
as purses,gloves, and expensiveshoes TURTLES
oil
and overcoats.The principalanimals
skins
killed for this trade are peccaries,
meat
CAIMAN
capybara,and various species of repoil
tiles.
OTHER REPTILES
Peccary leather has always been
Caiman lizard
I...-:, ...::.>.,,....v.af.....
...-.
-::
-.-:
.: ./,:..,...:
.........
,.,..s
skins
Iguana, Anaconda,
popular, particularlyin Europe and
Boa
?a~:::~~~~~~~~~~~iiQ
Japan.Between1946 and 1966, more
' ' . :, . : ..:>
:,,.......:
than 2 million collared peccary (Tameat
CAPYBARA
and
skins
800,000
yassu tajacu)
leather
white-lipped peccary skins were exported from Iquitos, Peru, alone. DEER
skins
meat
Capybaraare anothersourceof high..
quality leather. Between 1960 and
..?..
..
->.
-::::
:::::::
....:
ts:f
skins
1969, almost 500,000 capybaraskins PECCARY
meat
wereexportedfromthe BrazilianAmazon. Tradein these speciescontinues
skins
today, although at substantiallyre- OTTERS
duced levels.
The most importantwildlife in the
skins
leather industry at present are the CATS
reptiles, principallythe crocodilians.
During the peak of the trade in the PRIMATES
biomedical
trade
1950s and 1960s, five to ten million
,,,,,. .,
crocodilian skins were reported as
.
,a''as''
..,
^ ,,
X>-?:?:?~?~?~?:?~;?~?~,:?:
......
traded annuallyworldwide, with the BIRDS
plumes
=
_
W
i
X __~::~~:
actual figure probably much higher.
The extent of the trade is staggering:
pets
for example, in Venezuela during PARROTS/MACAWS
1930 and 1931, 3000-4000 caiman
skins were being sold daily, and beof faunain theAmazonsincethetimeof European
tween 1951 and 1980 Colombia le- Figure3. Commercial
exploitation
million
12
almost
discovery.
gally exported
**
............
. . ........
,r
:'.,
','.' .' . .....' -
*
June 1992
fS:::,y.........:S
.
* f8">2'
.....
.. .. ............ ....
\,,,.,,.*-Z
..*ss*.*s
+v
s*
.................
,
,....
????1?.1;.,...r?~~
. .........?????
..- ' -. * : , ?i:-::.. ......... .......
?:?.?:?......................
????????
.;l
?????
.......
.''=
.-.'
.........'.
'
;.
'
'
'
-
"'-*
'v. ;_s_g
-''
...............
. ...-...
415
100
I ... . .*..... . . .**..*@@@@|@@...............
......
.
.
.....
?.................. .....
-.......................
................................... ........
...........................
.
.
.
.
.
.
..
.
..
.
..
.
............
..........
............
..
.
..
.
I...................................
..
.
ous ports of entry. Fromthe perspective of an ecologist, however,what is
of interest is the total number of
individualanimalsremovedfromecosystems in the course of commercial
trade. For each animal that makes it
into the import/export trade, many
others are killed.
The number of animals actually
i:::::iiiii Nogae i:i:i:i:*i:i:i:i:*
i:i*:i*
Zkilled is unknown, but from the
limit
data
ited
seems
reasonable
to
esti...........
.... Non-game.
. .. . . . . ...birds.....
.
mate that for each animal enteredas
....
::::::::::*:::*::-::*::-::*::-::-::|::@::@::::::::::|:-::|::@::-:::::::
an export statistic approximately
~~~~~~~................-.-.....
':':':':':':':':':':':':':':':':':':':':':':-:':':'::-:':-:-:':O:':':-:
three additional individuals die.
These animals may be wounded by
...........
...........
..........
...
... ....
... .......
... ...
....
....
.... .......... .......
the hunter or the hunter's dogs or
trapsand escape to die later;the skin
.. . . .............
.
??...........
??
???
....................................
....
....
....
....
....
....
....
may be damagedeither in the course
???
?????
....................................?
....................................
???
...................................??
..............................................
.................
of
the hunt or in processingor stor... .... . .. . .. .. . .. . .. .
....................................
.... .... . .. . .. . ..??
.???
.
.. ??
the animalmay be too small and
....................................
age;
....................................
::::::::::::::::::::::::::::::::::::::::::::::
::::::::::.::::.:::::
???? ??
...................................?
...................................
...........
...........
...........
is
discarded
either by the hunter or
..:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:-:.:.:.:.:.:.:.:.:.:.:.:-:.:.:-:.:.:-:
.
.
.
.
.
.
.
.
.
..
.
..
.
.
.
.
.
.
.
.
.
.
..
.
..
.
..
.
....
-....-..... ... ...
... ... .........
......
... .... ......
......
... ... ......
......
.....................................
. ....
. . . . . . . . . . . .
. .
....................................
the buyer; or a lactatingfemale may
... *................................
.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-:-.-.-:
....................................
...................................???
....................................~?
be killedand her infantdies as a result
..................................
.?? ?????
....................................
...................................
....................................
...........
.. ???
....??
...
??
...............::::::::::::::::::::
...................................
.:..................
........
of
her death. In the case of the live
.....................
.. .........
...?????~?
....
....................
?
?
....................................??
...................................
. ...................................
. . . . . . . . . . . . . . . . . . . . . . . . .?
. .I
. .?
. . ...............
.....**.v@.@@.**.@
ww.
..............................
...
??. ????
?. ?. ??
?????
....................................?
....................................
:::-::::::::::::::::W:::w:::-:::-:::-:::w::::::::::::::::::::
animal
....................................
I
...................................?
trade, the mortalityin trans....................................
...................................
..................
. . . ...... . . ...
. .. . ...
.. . ......................
. .. . ...
. .. ...
.. . . ...
. . . . ..
. . .........
....................................
??
?
??
?~?~~?~
....................................
to
the point of final sale is even
....................................
port
....................................
..................
..
..
...
..
..
........
. ..
. ..
. . ..
. ..
. .
(Grimwood
1968).
higher
....................................
???
???????
??I
..
...................................
.......
*.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.-.:.:.:.:.:.:.:.:.:.:.:.-:.
.....................@
9.
...................................
....................................
....................................
To
illustrate
the
number
of animals
..........................
. .......
. .....
........
........
........
?
??
?
???
....................................
::-::-::-::-::-::-::-::-::-::-::-::-::-::-:-:-:-:-:::::::::::::::::::::
....................................
....................................
............................. ......?
? ? ?I?
~??
....................................
....................................
in
involved
the
commercial
wildlife
....................................
...................................
....................................
...............:::::::::::::::::::::::::::
....................................
....................................@
1
exTable
lists
the
number
trade,
? ??~???
???
....................................?I
??
...................................?
????? ?
....................................?
between
1962
and
1967
from
?
??
ported
::::::::::::::::::::::::::::::::@:::::::::::::::::::::::
...................................?
?..........
?? ???~
....................................?
...........
the Amazonianport of Iquitos.Using
....... .................
...... ......
......
the
correction factor of three, the
................................... ???? ?
.:.:.:.:.:.:.:.:,:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:,:.:.
total numberof animalskilled would
be almost5 million,or approximately
800,000 animalsper year or one ani---------I.........................
mal for each square kilometerof the
PeruvianAmazon each year.
.................................
................
........
........
......
.
....
...............
........
.....
-.-,............... ....
90
-
...................................
...................................
...................................
.......................,.....................
~ .:,~...
.o........,.....-. ....-..- .. ... .-.......,.:%.......!
.. . . . . . . .
...................................
iiiiiiilll
.....-.............
...................................
l ..........
............
.
...................................
:::::::::...
80 -
.
.
.
.
.
..
. ..........
..
.
.
70 -
.
...
e::::::::::::::::::::
.....-.-........
**................
..
.......
^
...
. ......
*@@.
.
.
. .
.
.
...........
..
.
.
...........
.
....................................
....................................
.....
....
....
.....
.....
.....
...
....................................
...................................
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Psittacidae(6/18)
6.2%
Psophidae(1)
4.7%
Phasianidae
(1)
5.2%
Cracidae(4)
12.6%
20 -
10 -
@.. *....
.....
.....
....
..
...................................
....................................
................................
....................................
...................................
....................................
...........................................
50 -
30 -
...,
.............
....................................
....................................
.........
.
. ...........
.............
.............
40 -
..- ..
-......
................-.-.-.-.-.-..-.....
No-ampaide()
60 -
...-.....-..
....-,,.,
*
.............
. ....................................
....................................
..
::::::::::;:::::::No-a
:::::::::::::::::
...-..
....'
...........
..........
...................................
............-..-.-.--...................................
........
...................................
....................................
,...................................
...................................
...................................
...................................
...................................
.........................................
...................................
...................................
...................................
...................................
...................................
...................................
...................................
....................................
.................................
,...................................
...................................
...................................
. .....................................
.. . ... -..-.-.-*.*..-.........
. . . . . .
l. *.. @...
l .. *..@ . . . . . . . .
.....-
...
......................................
...........
.............
. ..... ... . . .. ,....
@*.. ..
..
..
..
..
..
..
. . . . ..
...................................
.....................
.........................
,.-,.......,...
.. ... .............
. ...
.....
. ...
....,,
.....
.........
.......
.....
***...
..
..
..
..
..
..
..
..
.. .
...
.....................
'.-..,.-.,.........,.....,
...-...-...
-..
...
.........
..........
..................
...................................
....................................
.,-.,.........
....................................
....................................
.....:::::.:
. . ...........
..
..
.
..........
, . ..........
. ..........
..
. .
........
...........
.....
............... .................
.......... ......
?.-..-..-.-...--,....-.......-.................-.-.....
. . .....
.
.. ..............
.. .. .. ...........
...........
.........
.........................
..?....................-...-.-.-.
. . .
...........
...........
.. .. ..-........
......-....
.....
. . . . .. ..........
. ..... . . . . . . .,. . .. . ,.. . .. . . ..
...................................
...................................
...................................
...................................
...
...
Tinamidae
(9)
18.8%
..
.
..
....
...
.
..
.
.
...
.
.
..........
...........
..........
............
Game birds (29)
9.1%
0
Biomass
(190kg/km
NumberSpecies
(319 total)
2)
Figure4. Avian diversityin AmazonianPeru.Numbersin parenthesesare numbersof
et al. 1990.)
speciesin taxon.(DatafromTerborgh
III
killedan estimated10,000 smallspotted cats in a 15-year period (Paz y
Mifio C. 1988).
FEATHERS.Much less well-studied
is the exploitation of birds for their
feathers. Although currently out of
fashion,featherswere once important
elements in women's fashion. The
tradeconcentratedon egretsand herons. At the height of the feather
frenzy,premiumfeatherswent for $5
per plume or $28 per ounce in New
York.Between1899 and 1920, South
416
America (principallyArgentina,Brazil, and Venezuela)exported 15,000
kg of egret and heron feathers,representing an estimated 12-15 million
individualsof smallerspeciesof birds
and 3-4.5 million largerones.
BODY COUNTS. In interpretingthe
numbers of individual animals in-
volved in the commercialtrade emanating from the Amazon, previous
authors used either the number reportedlyexported at various ports of
exit or the numberimportedat vari-
Table 1. Animals and animal skins exported
fromIquitos,Peru,1962-1967. (FromRedford
and Robinson1991.)
Numberof
individuals
Animals
exported
Live monkeys
183,664
Skins
Caiman
Melanosuchus
47,616
Caiman
101,641
Mammals
67,575
Capybara(Hydrochaeris)
Otter (Lutra)
47,851
Giantotter (Pteronura)
2529
Ocelot (Felispardalis)
61,499
9565
Margay(Feliswiedii)
5345
Jaguar(Panthera)
Collaredpeccary(Tayassu
690,210
tajacu)
White-lippedpeccary
239,472
(Tayassupecari)
Deer (Mazama)
169,775
Total
1,626,751
BioScienceVol. 42 No. 6
It is importantto realize that subsistence and commercialhunting occur simultaneously.As a means of
providinga firstapproximationof the
combinedeffectsof these two typesof
hunting, at least in the 1960s and
1970s, I have extrapolatedthe above
estimate to the 3,581,180 km2 Brazilian Amazonian states. This calculation yields an estimate of 4 million
animals killed for commercial purposes per year, which may be combined with an estimated 19 million
killed each year for subsistence.
Therefore, approximately 23 million
animals are killed per year in the
Brazilian Amazonian states. If the
correction factor for fatal wounding
is applied to subsistence hunting as
well, the total reaches 60 million.
Given the increase in human population in the Brazilian Amazon since
1980, this total undoubtedly is an
underestimate of current kill rates.
Which animals are killed
by hunters?
Before attempting to assess the impact of the removal of such large
numbers of animals from Amazonian
forests, it is necessary to consider
several factors. First, the most commonly taken game animals are almost
always the largest members of their
group and usually the largest species
in the forest. Hunters prefer birds and
mammals of large body size. The only
large mammals not commonly hunted
for food (felids and otters) are hunted
for their pelts. Large wading birds
and raptors are the only large birds
not commonly hunted for food.
Second, large animals, although
represented by relatively few species,
are major contributors to the overall
biomass. The best data available to
illustrate this point comes from the
work done at Cocha Cashu Biological
Station in Manu National Park, Peru
(anson and Emmons 1990, Terborgh
et al. 1990). Of the 319 bird species
recorded at this site, 9% are commonly hunted (Redford in press, Redford and Robinson 1987). However,
these 29 species make up 52% of the
total avian biomass (Figure 4).
This pattern is even more striking
for the mammals, for which data are
available only for nonvolant species.
Of the 67 total species, 18% of the
species are commonly hunted. These
June 1992
100
*i!i!i
*
i!!iii!iiii!i
iiii!i
[!iii!!iii
....................................
....
...
......
:lill:ml-:llelil.l.l.'l......I.
b-11'llllll1111111lll
o. o%o.
. . .
. . .
.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:-:.:.:.:.:.:.:.:.:.:.:.:.:.:.:l
oo.o.. o.o. ooo.~OoO.oOO.o....o...
.oo~.o..
,.......
..,.,...,,.........
,.... ,... ,.....,..,..,.,...,
. . . o .o o. oo.
.
.
... . . .
.
. ..o.o..o...
.
.
.
....
.
o..o.o.oooo...
Rrl
..........-......
. .
.o.%
..o.
. .,.....,
..........-.
..,.-.,...,
..............
utd.-.-.....-..-
. .... .
.o
.....o
,..,....,..,.
.-.....,...,.....,.. o....~o
,........,.
......-................,...,.-..,,.,......
.
,.-.-.-.-.-.-...
. .oo.oo..oo....o~o
.
..
. o.o.o.....
.
.
.
.........
.O...o.......
.....ooo..oo.
...................................
.........................-.-.-.-.-.-.-.-.....................................
.-:.:.:.:-:.:.:-:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:-:.:.:-:.:.:.:.:.:.:.:.::...................................
....................................?:
....................................?
90 - ...................................???????
...................................???????????????
...
...
......................................
... .... ...
... .... ...
... .... ...
... ....
......................................
.....................................~
... . . . .. . . .. . . .. . .
...........
..
..........
. . . . . .............
.
............................
...................................???
..............................
,...................................
. .................................
...................................????????????
... ..... ......... ...
...
....
...... ....
..........................
.,..........
...
..
..
...
...
..
....................................??
80 - ...................................??????????
....................................??
...................................????????
. .. . . . . . ......-............---------.
. . . . . . ...
..........
. . . . . .
..
. . . . . .......---------
.... ... .... ... ... .... ... ...
LargePrimates
(4)
35.6%
60 -
Primates (4
Large
Alouatta=1
11.8%
Ateles=l
1.5%
Cebus(2)=1
2.3%
..............
....
...................
...............
....................
.
.........................
. . . . . . .......................................
............................l
. . . . . . . . . . . . . . ..
:.-- - :.;:
:-:.:.-::::::::::-:.::-- :-..
...................
...................................
....................................
....
...-... -..-...-.-.-.-.-.-.-.-.-.-.-.-.-.-.............
..o..o-...-......
o..o.-....-..
-;..-:.-.--:.:.:.:.-.:-:
..
..
....
......
. ...
....
..
. ...
...
...
....
..
. ..
...
...
...
...
...
...
...
..
. ...
...
...
...
...
..
..........................................
..................
. .
.
.
.
.
.
.
. ..........................................
.
.
.
.
.
.
.
.......................................
....................................
. .
.
.
.
.
.
.
. ..........................................
.
.
.
.
.
...................................
..............-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-...............
.........
...
.-.-..
.......
.....
.
.
.
..
..
.
.
..
..
.
.
.
..
..
..
.
.
-
.
-
.
-
..
-
e
.
.........................
..........
....................................
....................................
.................-.-.-.-.-.-.-.-.-.-.-.-.-...
...
..
. ..
. . ..
...
...
...
.....
....................................
..
...
. ..
. ..
. ....
. . . ...
. ..
....................................
...............
...............-.-.-.-.-.-.
.
.
-
..
..
..-
..
.??
....................................
..................
...
..
....
.....
..
...
.........l
.
.
............ .... .... .... ....
....................................
........
........
... ..... .. ........
... .........
...
..
..
... .. .....
..... ....
.... ... ....
....................................
....
...................-..
......
................................
.........................
..............
1..7. 9?.
*e..-.. -.. -.-... .. ............
........ .. ... .. ... .. ... ..
!i!i!*!i!i!i!i!i!e!e!i!i!i!i!*!i!i!*!i!ii*!iii
.. . ....
.. .
.. . ......... ......
.......
.. . . .....
.....
. ..
..
. ..
...
...
...
. ..
..
..
...
.......
....................................
.. .. .. .. . .. .. .. . .. .. .. .
......................-----
..................
.....................-.----
...............
... ........
...
........ ........ ....
.....
...................................
.. .... ...
...
.. ....... ...
..
.. .... ...
.................
..............
.
.. ......
.... ....
.
.
.
..
....
....
.
.
.
.
---------..........-
.
.
.
.
.
.
............
. ...
.
.
.
..
. ..
.. .
... .. ...... ....... ........ ....
...... ........
....................................
...................................
.-.-.-.-.-.-......
...--.-.
...................................
.............
.. .......
. . . . . .
............ :.:.:.2
.......... ...:.:n-am.:.:.:.:.:.:.:.:.:....:.-............
..
...
:.:.--:.:.:.:.:.:.:.:.;:.:.:.-:.::-:.:.:-.:.:.2:.:.:.:.:-:.:.:-:-:.:-:.-:.:.:.:.:.:.:.:.:.:.:.:.-:.:.:.:.:.:.:.:-:-.:.:.:.:.2.:.:-;.:.:--.:.
...........
...
.......
. ....
LargeRodents(4)
11.3%
.
.. . . .. . . . .. . . . .. . . .
...................................
....
...
..
.......
40 -
..... ............................
............................
??-. ??o.-.o??..........
??.- ??.- ?-.........?????
.. ?
...
..
.....
...... ?o... .o .. .-o.o.
...........-.-.-.-.-.-.-.-.-............
..........
...................
50 -
....
-.-.- ... -...-.-.-. -. --..
.-.......-.,...........,......,...-...............................
.................
......
.............
...................................
. . . . . . . . . . . ..
...................................
.
70 -
...
...
....................................
...............
.....................
....................................................................
...
.
..... . ....
........
. . . . .
...
. ..
...
............
..................................
... .........
... .. .........
......... ....
....
.....................
.................
...... ......
...... .... ......
......
... ....
........(2
.... ...
.................
mammals
Game:--::
mammals
:..::.:::.
Game...
...................................
.....
.....
.....
.... .....
....
....
............. .............
.....
.....
.....
.....
......
......
......
...................................
.... ......
....
.... ....
....
.... .... ......
...... ... ... ... ... ... ... ...
..... .. . .. . . .. . . .. . .
30 -
Perissodactyla(1)
7.3%
20 -
10 -
Artiodactyla(3)
21.0%
U
Biomass
(1526.6 kg/km2)
Number Non-Volant Species
(67 total)
Figure5. Nonvolantmammaliandiversityin AmazonianPeru.Numbersin parentheses
are numbersof species in taxon. (Data from Terborghet al. 1986 and Janson and
Emmons1990.)
12 species make up 75% of the mammalian biomass (Figure 5). To show
that the Manu site is not atypical of
neotropical forest sites, Figure 6 compares mammalian biomass data from
three other nonhunted sites. Clearly,
species that are preferred game make
up a large proportion of the biomass
in unhunted sites.
Third, and as a result, if hunting
affects the abundance of game species, areas that have been hunted
should show decreases in density and
therefore biomass of these species. A
review of the available data of effects
of hunting shows that under condi-
tions of moderate hunting, densities
of nonprimatemammaliangame species decreased80.7% when compared
with similar, unhunted sites. Under
hunting conditions describedby the
authors as heavy, nonprimatemammal densities decreased93.7% compared with similar, unhunted sites
(for details, see Redfordin press).
Due to the greateramount of data
on primates,a similarcomparisonof
the effect of huntingon primatescan
include both density and biomass.
Data from many differentAmazonian
sites show that in hunted areas large
primate biomass drops 93.5% when
417
100-
non-hunted
75 -
1990) shows that 69.5% of the game
bird biomass is composed of fruiteating species. This total would increaseto 96.2% with the inclusionof
the curassow (Mitu),which some experts believe is also primarilyfrugivorous.1
Edentata
50
-
Large Primates
LargeRodents
25 -
Perissodactyla
0
__?__
Manu, Peru
Terborghet al. 1986
. . . . . . .
_.
Guatopo,Venezuela
Eisenberg1980
.
. . . i......
. . .
Artiodactyla
..
HatoMasaguaral, BarroColorado,
Venezuela
Panama
Eisenberg1980
Eisenberg1980
Figure 6. Biomass of game mammals as a proportion of total mammal biomass at four
neotropical sites. Only the animals indicated in black are not hunted.
-
comparedwith similar unhunted areas, and large primate density drops
80.7% (Table2). In fact, Freeseet al.
(1982) have statedthat "predationby
humans is clearly one of the most
important factors affecting monkey
densitiesin most of the Peruvianand
BolivianAmazon, and probablyelsewhere in South America"(p. 82).
A similar comparison made for
game birdsshows a drop of 73.5% of
originaldensityundermoderatehunting and of 94.6% underheavy hunting. Similarconclusionswere reached
by Terborghet al. (1990) when comparing Panamanian and Peruvian
avian biomass. Finally, Thiollay
(1986) showed that in comparing a
hunted forest site with an unhunted
forest site in French Guiana, avian
game speciesbiomassdecreasedfrom
50.9% to 26.8% of the total avian
biomass.
A fourth factor that must be consideredbefore assessingthe effectsof
the removalof large numbersof animals from Amazonianforests is that
in tropicalforestsmany of the largest
animals,both terrestrialand arboreal,
are frugivores.Turning again to the
Cocha Cashudata, 84% of the gamemammal biomass is composed of
fruit-eating species, according to a
calculationtaking the biomass of the
most commonly hunted mammalian
taxa and considering frugivoreand
omnivores,frugivore-granivores,
frugivore-herbivores as frugivores
(data from Robinson and Redford
1986).
A similaranalysisof the gamebirds
(food habit data from Terborghet al.
Table 2. Impact of hunting on large primates in Amazonia. Means (standard deviations) are given.
a
Unhunted
Hunted
b
Primate biomass*
(kg/km2)
Large primate percentage of
total primate biomass*
Number of sites
363.8(314.3)
23.8 (38.1)
64.5(11.4)
19.6(29.6)
9
19
Primate densityt
(individuals/km2)
Large primate percentage of
total primate densityt
Number of sites
7
34.1 (10.8)
33.8(16.3)
8
6.6 (5.8)
8.1(10.0)
*Data from Peres 1990, Freese et al. 1982, and C. Mitchell and E. Raez Luna (unpublished results,
1991, Wildlife Conservation International).
tData from Peres 1990 and C. Mitchell and E. Raez Luna (unpublished results, 1991, Wildlife
Conservation International).
Unhunted
Hunted
418
Many of the most importantcommercial food fish in the tropical forest are also large fruit-eatingspecies.
Many of these frugivores are involved in seed dispersal,seed predation, and the structuringof tropical
forests.
The ecology of game animals
and of their absence
Althoughmany ecologistshave documentedthe importantroles playedby
large animals in seed dispersal,seed
predation,herbivory,pollination,and
predation, until recently few have
consideredthe role of largeanimalsin
tropicalforests and what would happen if they were removed from the
system (Emmons 1989, Janson and
Emmons 1990, Terborgh 1988).
Some ecologists have stated that removal of some individual species
from an ecosystem would probably
not have any substantialeffectson the
remaining species (c.f. Feinsinger
1983). Some studies, however, have
not only documentedthe existenceof
keystone species but also have demonstrated what happens when such
species are extirpated. Recently,
Brownand Heske (1990) have shown
that the removalof a guild of kangaroo rats from an experimentalplot in
the southwestern United States resultedin majorchangesin the vegetation structure.
Suchclear-cutcases are not known
from neotropicalareas, but there is a
growingbody of work suggestingthat
in this ecosystem large vertebrates
may perform important ecological
roles and that theirabsencewill result
in a changed forest (Janzen 1988).
Such a conclusion has been reached
for paleotropicalsettingsand in comparisonsbetween the Pleistoceneand
the present,but it has not been made
explicit for neotropicalsettings.Studies that shed light on the ecological
functions of large neotropicalverte'A. Grajal, 1991, personal communication.
Wildlife Conservation International.
BioScienceVol. 42 No. 6
brates fall into three categories:her- in Panama,Howe (1984) stated that
bivory and seed predation, seed dis- "animal-mediateddispersalis certain
to be critical for the demographic
persal, and predation.
recruitmentof many or most tropical
Herbivory and seed predation. In forest species"(p. 266).
Mexico, Dirzo and Miranda (1990)
Large birds, particularlythe touand cracids,are amongthe most
two
one
cans
forests,
compared
tropical
with its full complement of large important seed dispersers.Many of
mammals(peccaries,deer, and tapir) the speciesof cracids,particularlythe
and another in which these species curassows, are among the species
had beenextirpatedby hunters.There whose local populations are most
were strikingdifferencesbetween the rapidlydepletedby hunting.They are
two forests. The hunted forest was also slow to reproduce,with the avtypified by seedling carpets, piles of erage cracid requiring at least six
uneatenrotting fruits and seeds, and years to replace itself in the populaherbs and seedlings undamaged by tion (Silvaand Strahl1991). Because
mammalian herbivores-phenomena of the cracids' importance as seed
much less evident in the unhunted dispersersand susceptibilityto huntforest.
ing, Silva and Strahl have suggested
A second example comes from that "humanimpact on the Cracidae
work done comparingseed and seed- may have irreversiblelong-term efling predation, tree recruitment,and fects on the biology of neotropical
rodent populations on Barro Colo- forest ecosystems"(p. 51).
The other groupof importantlarge
rado Island, Panama,with adjoining
mainlandareas (De Steven and Putz seed dispersersare the primates,par1984, Glanz 1990). On the island, ticularlythe woolly and spider monocelots and other large mammalian keys. In a study in Surinam,spider
predators are absent, with agouti monkeys were shown to disperse
(Dasyprocta) and squirrel popula- seeds 93.5% of the times they fed on
tions high comparedwith the main- fruit and apparently served as the
land, where not only felids, but also only dispersal agent for several tree
humans,have reducedthe population species(von Roosmalen1985). Defler
of these rodents. Predationof seeds (1989) has shown a similar pattern
and seedlingsof severalcanopy trees for woolly monkeys.Like curassows,
on Barro Colorado Island is much spiderand woolly monkeysare highly
higher than on the mainland due to prized game animals and are rapidly
high populationsof seed-eatingmam- huntedout of a forest(Peres1990). In
malson the island.In an elegantstudy the absence of such large primates,
of small (1-4 ha) islands in an artifi- many species of plants may expericial lake in Panama,Putzet al. (1990) ence severely altered seed dispersal
showed that, in the absence of seed- patterns.
Interestingly,agoutis in additionto
eating mammals, trees with large
seeds had a distinct advantage over being majorseed predatorsalso serve
those with small seeds and came to as seed dispersalagents. It seems that
dominate the small forest patches in at lower densitiesthey are important
less than 75 years.
dispersalagentsfor large-seededtrees
Otherimportantseed predatorsare (c.f. Forget 1991, Hallwachs 1986);
peccaries, deer, and tapirs (Bodmer in their absence, at least some trees
1989); these species consume an would become locally extinct.
Another important group of seed
enormous number of seeds, particuis tropicalfish. Fish apparThese
from
those
dispersers
ungupalms.
larly
lates, especiallythe white-lippedpec- ently disperse seeds of at least nine
cary with its large group size, were plant families(Gouldinget al. 1988).
probably important elements in af- Many of the fish seed-dispersersare
fecting forest composition and struc- important food fish and are heavily
ture but now are very rare.
pursuedby both subsistenceand commercialfishers.
Seed dispersal. Many authors have
Gentry (1983) has shown, in the
documented the important role neotropics,that areas of higher rainplayed by large birds and terrestrial fall have highertree diversityand that
mammalsin the dispersalof the seeds much of the overall increasein plant
of tropicalplants. Based on his work diversity accompanying this higher
June 1992
rainfallis due to the additionof birdtree
dispersedand mammal-dispersed
In
that
he
shows
addition,
species.
wind-dispersed species tend to be
wide-ranging, whereas mammaldispersedspecies tend to be localized
in distribution.For example,43% of
the tree speciesfound at Rio Palenque
were presumablydispersedby large,
nonvolant mammals, and 50% of
these are endemicto coastal Ecuador
or to Ecuador and adjacentColombia, whereas only one mammaldispersed species ranges throughout
tropical America. Given these facts,
the loss of largevertebratedispersers,
which has undoubtedly occurred in
much of this area, would affect not
only local populationsof trees, but in
many cases it might result in the extinction of tree species with only localized distributions, resulting in a
loss of diversityof plant speciesmuch
greaterthan expected.
Predation.The role played by predators in structuringcommunitieshas
been well studiedin marineand intertidal systems. This work has shown
that predatorscan increasethe overall
species diversity in a communityby
decreasingthe abundanceof smaller
predators and competing herbivores
and by reducingdominanceof plant
prey species.
Such research has not been conducted in neotropicalforests, but biologists working in various locations
have observed that a decrease in
abundanceof large predatorymammals is correlatedwith the increasein
abundance of medium-sized terrestrial mammals, particularlyagoutis
(Glanz 1990, Janson and Emmons
1990). Absence of large predators
such as jaguars, pumas, and ocelots
also seems to result in more uneven
densities of prey species (Emmons
1987). Da Fonseca and Robinson
(1990) suggestedthat the absenceof
ocelots in forest patcheswas the reason for large numbers of opossums
and consequent lower small rodent
densities.
Intactcommunitiesof largecats are
rare in neotropical forests. Even
where there has been no hunting of
theseanimalsfor theirskins,therehas
usuallybeengamehuntingfocusedon
the speciesthat are primaryprey species of the large cats. Large raptors
are likewise affectedas they eat pri419
and mammals, an indication of the
ecological roles played by these species is possible.These examplesmake
clear the importance of fruit-eating
vertebratesin structuringand maintaining tropicalforests. What exactly
will happen as a result of the loss of
the game animals is not clear. Some
cases, as on Barro Colorado Island
and in the Mexican forests, provide
hints as to these consequences-the
resultingforest will be determinedby
a complicatedmix of more predation
on some species, less on others, and
the rarificationor extinction of still
others. What furthercomplicatesour
abilityto predictoutcomesis the fact
that humanshave been alteringtropical forests all along. What has
changedis the intensityand scale.
Figure7. A group of Braziliansliving along the AmazonRiverdisplayan ocelot skin.
Photo: Kent H. Redford.
marily animals that are major game
speciesfor humans.
There has been much less work
done on the effects of large mammalian and reptilian predatorson neotropical aquatic ecosystems. Yet this
guild has probablybeen the one most
heavily affected by human hunting.
For example, an estimated3 million
caiman, river otters, and giant otters
were removedfrom from Brazil,Colombia, and Peru between 1962 and
1969.2 Grimwood (1968) calculated
that, based on the home-range of
river otters, the mid-1960s commercial harvestingof river otters in Peru
2K.H. Redford,1992, unpublishedresults.
420
resulted in the elimination of adult
otters from more than 20,000 km of
waterway per year. Both black
caimanand giant otters, the two largest carnivoresin neotropicalfreshwater ecosystems,have also been eliminated from many areas.The effectsof
the loss of these two predators is
unknown, although Fittkau (1970,
1973 in Best 1984) has speculated
that caimanserve an importantfunction in nutrienttransferfrom terrestrial to aquatic ecosystems and that
their absenceis correlatedwith a decreasein the diversityand biomassof
fishes.
By examiningstudiesof herbivory,
seed predation, seed dispersal, and
predation by large neotropical birds
Ecological extinction. During the
height of the skin trade, many animals with valuableskins were killed.
Since the collapse of the skin trade,
tropicalforestpeopleshave continued
to hunt many of these animals because their meat is appreciated.The
result of this widescale human activity has been the reductionor extinction of local game populationsin virtually all areas of Amazonia.
Conservationbiologists are by definition concerned about extinction.
Yet, as Estes et al. (1989) point out,
there are several types of extinction:
global extinction, local extinction,
and ecological extinction. Ecological
extinctionis definedas "thereduction
of a species to such low abundance
that although it is still presentin the
communityit no longer interactssignificantly with other species" (p.
253). Althoughof tremendousimportance, conservationists ignore the
widespreadnature of ecological extinction in neotropicalforests, focusing instead on demographicextinction (extinction of a population or
deme) and calculations of minimal
viable populationsizes.
Even if jaguars, woolly monkeys,
or large curassows have not gone
extinct in the wild, their populations
may have been reduced to such an
extent that they no longer perform
their ecological functions. What is
needed is movement beyond the genetically based concern with demographic size to a new emphasis on
minimum ecologically operational
population size that incorporatesinBioScienceVol. 42 No. 6
teractionsbetween plant and animal
species.
The animalsthat are the most populargamespecies,and the ones whose
populationshave most likely become
ecologicallyextinct, include the most
important predators, the large-seed
dispersers,and the seed predatorsin
neotropical forests (anson and Emmons 1990). These largeanimalsprovide what Terborgh (1988) has referredto as a "stabilizingfunction."
Black caiman, jaguars,and harpy eagles maintainthe incrediblediversity
of tropical forests through indirect
effects,"the propagationof perturbations through one or more trophic
levels in an ecosystem,so that consequences are felt in organisms that
may seem far removed,both ecologically and taxonomically, from the
subjects of the perturbation" (Terborgh 1988, p. 402).
Humans are among the most severely affectedspecies. The effects of
huntingon large animalsare not just
of concernto those interestedin jaguars and bird-watching.Hunting is a
tremendouslyimportantsourceof nutritionfor millionsof neotropicalforest-dwelling humans-a "subsidy
from nature" (c.f. Hecht et al. 1988)
without which many other so-called
sustainableactivities, such as rubber
tapping, would not take place. As
Peres (1990) has shown, in one year
and a half, one family of rubbertappers in the BrazilianAmazon killed
more than 200 woolly monkeys, 100
spider monkeys, and 80 howlers. As
manyare beginningto realize,gameis
becoming rare in many areas inhabited by rubbertappers, affectingnot
only the game animals, but also the
ability of the people to live in the
forest. Animals are important not
only as food for humans, but also as
pollinatorsand dispersersof economically importantplant species, as regulators of pest populations, and as
providersof myriad other ecological
services.
Huntingaffectsnot only game species, but also local densities of nongame species. In a study of the bird
communitiesin hunted and unhunted
forests of French Guiana, Thiollay
(1986) showed that hunting significantly reduced the species richness,
diversity, density, and biomass of
game birds. Hunting was also correlated with decreaseddensitiesof rapJune 1992
tors and insectivores, as well as a
threefold decreasein the biomass of
frugivores.
humans,hunting,
Integrating
and conservation
game does not even appear (Myers
1988).
Conclusions
In tropical forests, large animals are
importantnot only as food for people
but also as integralecologicalcomponents of forestedecosystems.If these
ecosystemsare to continueas forests,
providingall of the financial,ecological and aesthetic benefits currently
desired, then animals must not be
ignored.Many large animalshave alreadygone ecologicallyextinctin vast
areas of neotropical forest-areas
with large, towering tropical trees,
lush ferns, and beautifulorchids.
We mustnot let a forestfull of trees
fool us into believingthat all is well.
Many of these forests are "living
dead" (Janzen1988), and, although
satellitespassing overheadmay reassuringlyregisterthem as forest, they
are emptyof much of the faunalrichness valued by humans. An empty
forest is a doomed forest.
Almost everyremainingpiece of neotropical forest has been affected by
humans during pre-Columbian
times, during the rubber-boomera,
during the golden era of the skin
trade, or more recentlyby gold miners, timber extractors,ranchers,and
farmers.Even where no sign of human habitation is to be found, rubber trees show the unmistakable
signs of having once been tapped,
piles of open Brazil nut shells show
the telltale mark of a machete, and
bones of caimankilled for their skins
whiten on the beaches.
Today, hunting is an integralpart
of all forest-basedactivities,be it lumbering,fishing,or medicinalplantcollection. For those living in the forest,
hunting is an essentialcomponentof
feeding a family; for those venturing
into the forest to collect forest products, it is also a necessarysubsistence Acknowledgments
I would like to thankAllynStearman,
activity.
In virtuallyall areas of neotropical FrancisE. Putz, Marianne Schmink,
forest,game animalpopulationshave and PeterFeinsingerfor theirvaluable
alreadybeen affectedby humanhunt- comments on this paper. Carol
ing. This patterncontinuesas human Mitchell and ErnestoRaez Lunagrapopulationsgrow and forestexploita- ciously agreed to make available
tion increases. The trend in recent some of their unpublisheddata. This
yearshas beento increasethe amount is a contribution from the Program
of land allocated to multiple-usear- for Studiesin TropicalConservation,
eas-including Indian lands, Man Universityof Florida.
and the Biosphere Programme reserves,extractivereserves,faunalpro- Referencescited
duction areas, and national forestsand to keep at a much lower level the Balee, W. 1989. The culture of Amazonian
forests. Pages 1-21 in D. A. Posey and W.
amountof land dedicatedto national
Balee,eds. ResourceManagementin Amazoparks and other traditionalconserva- nia: Indigenousand Folk Strategies.New
York BotanicalGardens,Bronx,NY.
tion units. This patternin the neotro1984. The aquatic mammalsand
pics is illustratedby Brazil,which has Best, R. C.
of
the Amazon.Pages371-412 in H.
reptiles
74 millionhectaresin all categoriesof
Sioli, ed. The Amazon: Limnology and
Indianlands, comparedwith 13 milLandscapeEcology of a Mighty Tropical
lion hectaresin all categoriesof conRiverand its Basin.Dr. W. Junk,Boston.
servation units, and by Colombia, Bodmer,R. E. 1989. Ungulatebiomassin relation to feeding strategywithin Amazonian
which has 18 million hectaresin InOecologia81: 547-550.
dian reservesand 2.5 millionhectares forests.
. 1990. Fruitpatchsize and frugivoryin
in NaturalNational Parks(sourcesin
the lowland tapir (Tapirus terrestris).J.
Redfordin press).
Zool. 222: 121-128.
As with other subsidies from na- Bodmer,R. E., T. G. Fang,and L. M. Ibafiez.
1988. Ungulatemanagementand conservature, game has been undervalued,unin the PeruvianAmazon.Biol. Conserv.
tion
derstudied,and ignoredby the conser- 45: 303-310.
vation and developmentcommunities. Brown,J. H., andE.J. Heske.1990. Controlof
For example, in the World Bank's a desert-grasslandtransitionby a keystone
rodentguild. Science250: 1705-1707.
1978 list of benefitsfrom the forest,
421
Castro, N., J. Revilla, and M. Neville 19751976. Carnede monte como una fuente de
proteinasen Iquitos,con referenciaespeciala
monos. RevistaForestaldel Peru 6: 19-32.
da Fonseca, G. A. B., and J. G. Robinson.
1990. Forestsize and structure:competitive
andpredatoryeffectson smallmammalcommunities.Biol. Conserv.53: 265-294.
Defler,T. 1989. Recorridoy uso del espacioen
un grupo de Lagothrix lagothricha (Primates:Cebidae)mono lanudochurucoen la
AmazoniaColombiana.Trianea(Colombia)
3: 183-205.
De Steven,D., and F. E. Putz. 1984. Impactof
mammalson early recruitmentof a tropical
canopy tree, Dipteryx panamensis,in Panama. Oikos 43: 207-216.
Dirzo, R., and A. Miranda.1990. Contemporary neotropical defaunation and forest
structure,function, and diversity-a sequel
to John Terborgh.Conserv. Biol. 4: 444447.
Emmons, L. H. 1987. Comparativefeeding
ecology of felids in a neotropicalrainforest.
Behav.Ecol. Sociobiol.20: 271-283.
. 1989. Tropicalrainforests:why they
have so many speciesand how we may lose
this biodiversitywithoutcuttinga singletree.
Orion Nature Quarterly8(3): 8-14.
Estes,J. A., D. 0. Duggins,and G. B. Rathbun.
1989. The ecology of extinctions in kelp
forest communities.Conserv.Biol. 3: 252264.
Feinsinger,P. 1983. Coevolutionand pollination. Pages 282-310 in D. J. Futuymaand
M. Slatkin,eds. Coevolution.SinauerAssociates,Sunderland,MA.
Flenley,J. R. 1979. The EquatorialRainforest:
A Geological History. Butterworths,London.
Forget,P. M. 1991. Seed-dispersalof Vouacapoua americana(Caesalpiniaceae)by caviomorph rodents in FrenchGuiana.J. Trop.
Ecol. 6: 459-468.
Freese,C. H., P. G. Heltne, N. CastroR., and
G. Whitesides.1982. Patternsand determinants of monkey densitiesin Peru and Bolivia, with notes on distributions.Int. J.
Primatol.3: 53-90.
Fund.Inst. Bras.Geogr.Estat. (FIBGE).1982.
AnuarioEstatisticodo Brasil.FIBGE,Rio de
Janeiro,Brazil.
Gentry, A. H. 1983. Dispersal ecology and
diversityin neotropicalforest communities.
Sonderbd.Naturwiss.Ver.Hambg.7: 303314.
Glanz,W. E. 1990. Neotropicalmammaldensities: how unusual is the Barro Colorado
Island,Panama,community?Pages287-313
in A. H. Gentry,ed. FourNeotropicalRainforests. Yale UniversityPress, New Haven,
CT.
Goulding,M., M. L. Carvalho,and E.G. Ferreira. 1988. Rio Negro: Rich Life in Poor
Water.SPBAcademicPubl.,The Hague,The
Netherlands.
Grimwood,I. R. 1968. Notes on the Distribution and Statusof SomePeruvianMammals.
422
AmericanCommitteefor InternationalWild
Life Protection,Bronx,NY.
Hallwachs, W. 1986. Agoutis (Dasyprocta
punctata):the inheritorsof guapinol(Hymenaea courbaril:Leguminosae).Pages285304 in A. Estradaand T. H. Fleming,eds.
Frugivoresand SeedDispersal.Dr. W. Junk
Publ.,Dordrecht,The Netherlands.
Hecht, S. B., A. Anderson,and P. May. 1988.
The subsidy from nature: shifting cultivation, successionalpalm forestsand ruraldevelopment.Hum. Org. 47: 25-35.
Howe, H. F. 1984. Implicationsof seed dispersal by animalsfor tropicalreservemanagement.Biol. Conserv.30: 261-281.
Janson, C. H., and L. H. Emmons. 1990.
Ecologicalstructureof the nonflyingmammal communityat Cocha Cashu Biological
Station, Manu National Park, Peru. Pages
314-338 in A. H. Gentry,ed. FourNeotropical Forests. Yale UniversityPress, New
Haven, CT.
Janzen, D. H. 1988. Managementof habitat
fragmentsin a tropicaldry forest: growth.
Ann. Mo. Bot. Gard.75: 105-116.
Jorgenson,J., and K. H. Redford. In press.
Humans and big cats as predatorsin the
neotropics. Symposium of the Zoological
Societyof London No. 65. Oxford University Press,New York.
Lathrap,D. W. 1975. The antiquityand importanceof long-distancetraderelationships
in the moist tropicsof Pre-Columbian
South
America.WorldArchaeology5: 170-186.
Lovejoy,T. E.,J. M. Rankin,R. 0. Bierregaard
Jr., K. S. BrownJr., L. H. Emmons,and M.
E. Van derVoort. 1984. Ecosystemdecayof
Amazonforest remnants.Pages295-325 in
M. H. Nitecki,ed. Extinctions.Universityof
ChicagoPress,Chicago.
Malcolm,J. R. 1991. The small mammalsof
Amazonianforest fragments:patterns and
process. Ph.D. dissertation, University of
Florida.
Martinelli,L. A., J. R. Ferreira,B. R. Forsberg,
and R. L. Victoria.1988. Mercurycontaminationin the Amazon.Ambio 17: 252-254.
McGrath,D. G. 1986. The animal products
tradein the BrazilianAmazon.Unpublished
reportto WorldWildlifeFund,Washington,
DC.
Myers,N. 1988. Tropicalforests:much more
than stocksof wood. J. Trop.Ecol. 4: 209221.
Pazy MifioC., G. 1988. Notas sobrela caceria
y la conservaci6nde los felidos en la Amazonia Ecuatoriana.Fundacion"SimonBolivar" (Quito, Ecuador) Boletin Cientifico
Afio 11(3).
Peres,C. A. 1990. Effectsof huntingon western Amazonianprimatecommunities.Biol.
Conserv.54: 47-59.
Putz, F. E., E. G. Leighjr., and S. J. Wright.
1990. Solitaryconfinementin Panama.Garden March/April1990: 18-23.
Redford,K. H. Inpress.Huntingin neotropical
forests:a subsidyfromnature.In M. Hladik,
ed. Nutritionin TropicalForests.UNESCO,
Paris.
Redford, K.H., B. Klein, and C. Murcia. In
press. The incorporationof game animals
into small scale agroforestrysystemsin the
neotropics.In K. H. Redfordand C. Padoch,
eds. Conservationof Neotropical Forests:
Buildingfrom TraditionalResourceUse.ColumbiaUniversityPress,New York.
Redford,K. H., andJ. G. Robinson.1987. The
gameof choice:patternsof Indianand colonist huntingin the neotropics.Am. Anthropol. 89: 650-667.
. 1991. Subsistenceand commercial
usesof wildlifein LatinAmerica.Pages6-23
in J. G. Robinsonand K. H. Redford,eds.
NeotropicalWildlifeUse and Conservation.
Universityof ChicagoPress,Chicago.
Robinson, J. G., and K. H. Redford. 1986.
Body size, diet and population density of
neotropicalforest mammals.Am. Nat. 128:
665-680.
Roth, P. 1984. Reparticaodo habitat entre
Psitacideossimpatricosno sul da Amazonia.
Acta Amazonica14: 175-221.
Ruddle, K. 1970. The hunting technologyof
the MaracaIndians.Anthropologica25: 2163.
Silva, J. L., and S. D. Strahl. 1991. Human
impact on populations of chachalacas,
guans, and curassows (Galliformes:
Cracidae)in Venezuela.Pages37-52 in J. G.
Robinsonand K. H. Redford,eds. Neotropical WildlifeUseand Conservation.University of ChicagoPress,Chicago.
Terborgh,J. 1988. The big thingsthat run the
world: a sequel to E. 0. Wilson. Conserv.
Biol. 2: 402-403.
Terborgh,J., L. H. Emmons,and C. Freese.
1986. La fauna silvestrede la Amazonia:el
despilfarrode un recursorenovable.Boletin
de Lima (Peru)46: 77-85.
Terborgh,J., S. Robinson,T. A. ParkerIII,C.
A. Munn, and N. Pierpont.1990. Structure
and organizationof an Amazonianforest
bird community.Ecol. Monogr. 60: 213238.
Thiollay,J. 1984. Raptorcommunitystructure
of a primaryrain forest in FrenchGuiana
and effectof humanhuntingpressure.Raptor Res. 18: 117-122.
Thiollay,J. M. 1986. Structurecompareedu
peuplementavien dans trois sites de foret
primaireen Guyane.Rev. Ecol. 41: 59-105.
Vasquez,R., and A. H. Gentry.1989. Use and
fruitsin the Iquimisuseof forest-harvested
tos area. Conserv.Biol. 3: 350-361.
von Roosmalen,M. 1985. Habitatpreferences,
diet, feedingstrategyand socialorganization
of the black spidermonkey(Atelespaniscus
paniscus Linnaeus1758) in Surinam.Acta
Amazonica15 (3/4) suppl.
Yost, J., and P. Kelley.1983. Shotguns,blowguns, and spears:the analysisof technological efficiency.Pages189-224 in R. B. Hames
and W. T. Vickers,eds. AdaptiveResponses
of NativeAmazonians.AcademicPress,New
York.
BioScienceVol. 42 No. 6