SURJ.
Anthropological Sciences
Examining the Extinction of
the Pleistocene Megafauna
During the Pleistocene, the world saw a dramatic number of extinctions of very large terrestrial
species. The losses of these megafauna have been attributed to either of two different hypotheses.
One hypothesis states that global climate changes occurring during the Pleistocene caused
environmental pressures that forced the extinction of several megafaunal species. The second
hypothesis proposes that the global spread of Homo sapiens and hunter-gatherer subsistence
practices were responsible for these deaths. Examining these two theories reveals that neither
climate changes nor human overkill were likely to be individually responsible for the Pleistocene
extinctions. Instead, a synthetic model that includes both hypotheses appears to be the most
plausible explanation for the Pleistocene losses.
Robin Gibbons
the existence of the Wisconsin ice sheet
There are two schools of thought possibility exists that both explanations
expanded (Guilday 1967). Thus, some
regarding to the underlying cause of the are simultaneously plausible. Therefore,
Pleistocene extinctions. Both attempt it follows that studies should move be- animals experienced range expansion
while others saw their habitats shrink
to explain the loss of nineteen genera yond these individual arguments and
or disappear. In addition, new compe-
and over fifty species in Australia and start formulating multi-causal models
more than seventy genera in the Ameri- that incorporate both climate and over- tition arose between formerly noncom-
petitive herbivores as a consequence of
cas almost all terrestrial animals, and
kill into their reasoning.
the changing environments (Guilday
virtually all large in size (Martin 1984).
1967). Selective evolution will favor
The North American fauna saw the ex-
Climate Changes
one species in a direct competition and
tinction of  elephants, horses, camels,
drive the other to extinction. Because
ground sloths, all but one pronghorn,
During the late Pleistocene, the re-
of their large size, competition affected
several ovibovids, peccaries, and the
treat of the Wisconsin ice sheet caused
the megafauna more than the smaller
giant beaver (Guilday 1967:125).
global climate changes and changes
existing species. Larger animals gen-
South America also lost elephants and
in local environments. Temperatures
erally require a larger primary habitat,
sloths, in addition to horses, some cam-
became less homogenous, as winters
more forage, and more water to main-
els, glyptodonts, and a rhinoceros-sized
became colder and summers became
tain themselves than a smaller species.
giant rodent species. Extinctions in Aus-
hotter. Essentially, seasonality in-
Therefore, there are usually only a few of
tralia included all herbivorous marsupi-
creased. In addition, rainfall became
them in a given area. As Grayson says,
als larger than modern day kangaroos
more variable depending on the season,
 the inverse relationship between body
(Guilday 1967). The first theory states
with distinctions between wet and dry
size and population size plays a powerful
that global climatic changes at the end
seasons (Kiltie 1984). During the de-
role in increasing the risk of extinction
of the Pleistocene were responsible for
glaciation, many streams in the glacial
faced by larger animals (Grayson 1991:
the demise of so many large terrestrial
floodplains experienced net degradation
214). Following Guilday s logic, shrink-
animals. Conversely, the overkill hy-
and incision of their channels, and the
ing habitats forced large animals into
pothesis proposes that the extinction of
water tables lowered, causing low order
cohabitating smaller areas, containing
large land animals was concordant with
streams to become sporadic and transient
insufficient resources to support them
arrival of modern Homo sapiens in the
and springs to dry up or significantly re-
all. The result was the elimination of
New World and Australia and that man-
duce discharge (Haynes 1984).
some species due to competition.
the-hunter brought about these deaths.
As a result of this climatic shift, sev-
There are several possible reasons
However, neither theory has managed
eral changes occurred. Some primary
underlying extinction in the face of en-
to dominate the debate over the cause of
habitats were eliminated, while others
vironmental pressure brought about by
the megafaunal extinctions. Instead, the
that may have been only marginal during
22
SPRING 2004
climate change. Overspecialization (in- rather, correlate with conditions that Another climate-related argument
ability to adjust to a new environment), directly influence the survival of the was made by Gingerich (1984). He
competition with a superior species, in- mother and her offspring, such as tem- states that the late Pleistocene extinc-
ability to migrate to a more suitable hab- perature, rainfall, or photoperiod (Kiltie tions are merely a response to a high
itat, and extremely rapid environmental 1984). Larger species generally use cor- rate of species origination in the early
change (impossibility of evolutionary related cues because their increased size Pleistocene because  origination and
adaptation) will all cause extinction of a lends itself to increased gestation peri- extinction are in dynamic equilibrium
species (Guilday 1967). These problems ods. Correlated cues influence the time constrained by environmental diver-
are magnified in shrinking habitats and of mating such that the offspring will sity (Gingerich 1984:216). During
will force the extinction of larger spe- arrive during the favorable season, given the early Pleistocene, originations far
cies first. This is not to say that smaller that the seasonality does not change. outnumbered extinctions, thus sug-
species are insensitive to climatic and However, during the late Pleisto- gesting a high level of environmental
environmental change. Rather, smaller cene, seasonality increased. Unpredict- diversity. The glacial period exhibited
specimens require smaller amounts of able environment skews the correlated high environmental heterogeneity, and
resources and can exist in smaller habi- cues such that the reproductive rates of therefore, supported diverse fauna
tats. Therefore, they will not feel the larger species decreases. Therefore, (Gingerich 1984). However, during
effects of a shrinking habitat until well the increasing seasonality of the late the retreat of the Wisconsin ice sheet
after the larger species have started to Pleistocene would increase the risk of and the deglaciation of the landscape,
deal with the environmental pressures. extinction of these larger species due to environmental heterogeneity decreased,
The selective pressures brought about their decreased birth rates. Kiltie (1984) rendering a high level of faunal diversity
by environmental changes caused the also states that uncertainty in the length unsupportable. Thus, Gingerich (1984)
extinction of several large species of of gestation period and timing of birth makes the argument that the high rate
land animals. Additionally, the minor with respect to season can reduce fitness. of extinction during the late Pleistocene
halts and readvances of the Wisconsin This uncertainty reduces the reproduc- was a natural equilibration response to
ice sheet during this period of extinction tive success in seasonal environments the previous high rate of origination dur-
(between 18,000 and 11,000 years ago), because the favorable season is more ing the early Pleistocene.
were insufficient to restore the previous limited and survival of the offspring is
environmental balance and merely tem- directly related to whether or not birth
Human Overkill
porarily slowed the inexorable extinc- occurred during the favorable season.
tion of several species under selective Larger species are more affected by In addition to the climatic changes
stress (Hester 1967). uncertainty because of their increased occurring in the late Pleistocene, modern
A corollary argument to the climate gestational period and decreased favor- Homo sapiens hunter-gatherers were mi-
hypothesis concerns gestation length. able season (offspring need sufficient grating across the Berengia land bridge
Slaughter (1967) proposes that the in- time to reach a favorable size for up- into North and South America. The
ability of certain species to adjust their coming harsh weather). Additionally, dates of the extinctions of the mega-
mating habits to the changing climates of Kiltie (1984) says that reproductive rates fauna coincide with the Clovis culture
the late Pleistocene brought about their decreased during the increasing season- in the Americas. Thus, the argument has
extinctions. These species tended to be ality of the late Pleistocene because time been made that human hunting caused
larger, with longer gestation periods. was lost synchronizing reproductive the Pleistocene extinctions.
Different species mate at different times cycles to the changing environmental According to Martin (1967), the
of the year depending on either direct cycles. The  lost time should have extinctions of the megafauna are not
cues or correlated cues (Kiltie 1984). been used to produce offspring, but consistent with climatic change because
Those species that possess a gestation was instead wasted waiting for favorable the majority of losses were in North and
period shorter than the favorable season breeding cues. This problem is magni- South America and Australia. In addi-
for producing offspring determine when fied in large species with long gesta- tion, the extinctions were largely ter-
to reproduce using direct cues condi- tion periods because more time is lost. restrial animals with very few marine
tions that make birth favorable, such Reduced reproductive success in these or plant extinctions. Martin (1967)
as temperature and food availability. large species precipitated their extinc- also points out that the previous glacia-
Conversely, those species with gestation tion because their intrinsically smaller tion cycles were roughly equivalent in
periods longer than the favorable season populations are much more sensitive to magnitude to that of the Wisconsin ice
for producing offspring initiate mating decreasing numbers of offspring. If in- sheet in terms of environmental stress,
using correlated cues. Correlated cues sufficient offspring are produced to carry but did not experience nearly the same
do not affect reproductive success, but a population, it will disappear. scale of extinctions. Many of the spe-
23
SURJ.
cies that became extinct during the late a greater number of species (McDonald caused some populations to shrink to
Pleistocene survived the previous glacial 1984). Overall, the climate change was numbers impossible to maintain.
cycles, necessitating another explana- favorable to terrestrial animals due to Modeling studies done by Whitting-
tion. The only unique thing about the increased primary productivity of the ton and Dyke (1984) and, more recently,
Wisconsin glacial cycle is the global land and cannot account for extinction. Alroy (2001) lend further credence to the
spread of Homo sapiens. Following this argument, Homo sapiens human overkill hypothesis. Whittington
The extinctions of the late Pleisto- were the only disruptive element in the and Dyke (1984) utilized the Mosimann
cene occurred relatively rapidly, within a existing environment of the late Pleisto- and Martin model positing that megafau-
few thousand years. Additionally, there cene. The disruption caused by this new, nal extinction in North America could
are very few kill or processing sites in highly successful predator was enough be caused by a small number of Homo
the archaeological record connecting to exterminate several species of large sapiens migrating in a  front across the
Homo sapiens to the lost fauna. Grayson land animals due to lowered population landscape, leaving little archaeological
(1984) says that this lack of associations sizes (McDonald 1984). Given the low evidence behind. A  front occurred
between extinct Pleistocene mammals reproductive rate of large animals, a when the hunting of the human popula-
and humans in North America refutes diminished population size can lead to tion caused the extinction of megafaunal
the overkill hypothesis, but Martin inadequate fitness to sustain the popula- species in the inhabited area. The model
(1984) states that this paucity supports a tion as a whole. was run using baseline parameters in-
blitzkrieg model. Blitzkrieg is a  special During migration across the Beren- cluding initial prey biomass, prey carry-
case of faunal overkill that maximizes gian land bridge, Homo sapiens brought ing capacity, prey biomass growth rate,
speed and intensity of human impact with them a meat-based diet and new initial human population size, human
and minimizes time of overlap between bacteria and parasites. A meat-based population growth rate, human carry-
the first human invader and the disap- subsistence pattern evolved during ing capacity, and minimal kill, defined
pearance of native fauna (Martin 1984: the Upper Paleolithic in Europe as a as the minimum value of prey destruc-
396). The blitzkrieg model explains the response to the frigid environment tion rate required to cause simulated
lack of kill sites by reasoning that the (Jelinek 1967). This same adaptation megafaunal extinctions (Whittington
extinction of these animals occurred allowed Homo sapiens to survive during and Dyke 1984). By assigning values
too rapidly to have left much, if any, migration across the tundra environment to these parameters based on empirical
evidence. Therefore, the uniqueness of Asia and Berengia. This predilection archaeological evidence, Whittington
of Homo sapiens in the New World and for animal protein as the mainstay of and Dyke (1984) found that overkill is
Australia, coupled with the lack of kill diet combined with the inexperience a plausible explanation for the late Pleis-
sites in the archaeological record, can be of local prey (large animals in the tocene extinctions. Alroy (2001) used
taken as evidence of blitzkrieg human Americas) with Homo sapiens hunting a computer simulation to model Homo
overkill causing the late Pleistocene practices led to profligate extermination sapiens and large herbivore population
extinctions. of these species (Jelinek 1967). Since dynamics in the North American end-
During the deglaciation process, the the indigenous species of the New World Pleistocene. He assumed slow Homo
general ambient temperature increased had not evolved in the presence of Homo sapiens population growth, random
along with the net usable moisture over sapiens, they also had not developed the hunting that was nonselective and lim-
the land. Equability of temperatures and natural wariness exhibited by similarly ited, and low maximum hunting efforts
seasonality most likely decreased in the large species in the Old World. Thus, combined with parameters culled from
northern and central portions of North the megafaunal species were easy hunt- published values. Simulations found
America, but remained stable in the ing targets. In addition, Homo sapiens that Homo sapiens growth rate and hunt-
coastal areas, southern North America, served as a transport method for foreign ing ability almost always led to mass
and the North American peninsula (Mc- bacteria and parasites during migration extinctions, with hunting ability being
Donald 1984). Essentially, areas existed across Berengia. The diffusion of new the most important of all parameters
where temperatures and seasonality re- diseases into a population of herbivores (Alroy 2001). Alroy (2001) concluded
mained stable. Therefore, animals un- unadapted to exposure decimated the in- that the overkill hypothesis was not only
able to cope with the climate changes in digenous large animal species (Edwards credible, but  an anthropogenic extinc-
their habitats might have migrated. In 1967). While disease alone is probably tion was unavoidable given the facts of
addition, deglaciation caused an increase not sufficient to cause extinction alone, ecology and the fossil record even
in continental land area due to the retreat the resultant reduction in population size assuming that human predation was
of the ice sheet. Land area positively combined with Homo sapiens hunting limited and nonselective (Alroy 2001:
correlates with species diversity, thus the practices and low reproductive rates 1896).
increased land area could have supported common to megafauna most certainly
24
SPRING 2004
Utilizing newer dating methods, North America. Hildebrandt and Mc- Neither model takes climate change into
Roberts (2001) looked at extinctions Guire (2002) confirm that the relation- account, which is negligent considering
in Australia in attempt to determine if ship between hunter and hunting weap- the widespread acknowledgement that
Homo sapiens or climate changes were onry is fundamental to the high-risk, yet climate was changing during this time
the underlying cause. He rejected dates high-reward, behavior of hunting large period. Given ethnographic data col-
calculated by radiocarbon (14C) as unre- animals. Thus, it is possible to argue lected from modern hunter-gatherer
liable. Using optical dating, which mea- that human overkill of megafauna was groups, hunting behavior varies depend-
sures the time elapsed since sediments motivated by the pursuit of fitness ben- ing on environment. The Whittington
were last exposed to light (luminescence efits associated with hunting large game and Dyke (1984) model assumed a rapid
230
method), and Th/234U dating, which animals and resulted in extinction. rate of migration as a  front that left
measures crystallization age of flow- extinction in its wake. It ascribes the
stone, he determined the burial ages of lack of archaeological data for this front
Discussion
28 sites. From these measurements, he to the rapidity with which Homo sapiens
found that extinctions occurred around In looking at these arguments, it be- migrated across the continent, following
46,400 years ago across Australia, thus comes apparent that they are not mutual- Martin s (1984) blitzkrieg model. At-
ruling out any climatic impacts from the ly exclusive. It is quite possible that cli- tributing the lack of associated sites to
late Pleistocene as the underlying cause mate worked cooperatively with Homo being caused by a short time frame of
(Roberts 2001). Since Homo sapiens sapiens to produce the extinction effect extinction and migration is somewhat
arrived in Australia at least 40,000 years that claimed so many terrestrial animals questionable. Sites might be sparse
ago (Horton 1984), Roberts (2001) data in the New World and Australia. The and lack high densities of accumulated
concurs with the human overkill hypoth- changing climate, complete with rising material because of the short time frame,
esis of extinction. temperatures, desiccation of the environ- but that is not to say they do not exist.
While hunting large game seems to ment, and retreat of the Wisconsin ice Simply, these sites have yet to be uncov-
counter expectations generated by opti- sheet could have pushed large terrestrial ered. If they were uncovered, this would
mal-foraging and diet-breadth models, animal populations into dangerously low not refute the blitzkrieg model unless
Hildebrandt and McGuire (2002) see it numbers due to the environmental pres- they showed evidence of long inhabit-
as an important social aspect in hunter- sures. Then, the hunting practices of ance periods (i.e.: high densities of ac-
gatherer society apart from group pro- Homo sapiens groups, unaware of the cumulated debris). In addition, Alroy s
visioning. Large-game hunting served imminent dangers of small population model (2001) sets the minimum num-
as an activity that had the added bonus sizes for megafauna, could have pushed ber of people entering a given region at
of increased fitness due to increased these animals over the brink, onto the 100. Modern hunter-gatherer group size
mating opportunities and favorable inexorable path towards extinction. In tends to hover around 25 individuals,
treatment for offspring, in addition to this model of megafaunal extinction, especially if they exhibit high mobility
greater access to social networks, po- both arguments play a role, yet neither like the overkill hypothesis assumes. It
litical alliances, and political authority is dominant enough to assume sole is unknown from Alroy s (2001) data
(Hildebrandt and McGuire 2002). This responsibility for the late-Pleistocene what effect reducing group size to 25
view is corroborated by Kaplan and Hill extinctions. If Homo sapiens groups individuals would have on the outcome
(1985), who contend that sharing behav- were not present, population numbers of his model.
iors had their origins in increased fitness might have recovered after an unstable It is possible to combine the ar-
benefits accrued from certain activities. period of adjusting to the new climate guments presented by some of these
Given that large game is most conducive and habitat. Conversely, if climate had studies, such that they would support a
to sharing outside the immediate fam- not already reduced the population sizes multifaceted model of extinction. For
ily, it stands that Homo sapiens would of large animals, the relatively small example, Guilday (1967) proposes that
pursue these animals, even after their number of migrating Homo sapiens ar- the changing climate in the late Pleis-
numbers had dwindled, because of the riving in the New World and Australia tocene caused many habitats to shrink.
perceived benefits. This is an example may not have been sufficient to decimate This reduction in suitable habitat size
of  show off behavior. These aspects entire species of megafauna. forced a higher number of large ter-
of late Pleistocene Homo sapiens hunter- The models proposed by Whitting- restrial animals to inhabit a given area
gatherer societies would be evident in ton and Dyke (1984) and Alroy (2001) than normally would. While Guilday
the archaeological record in the form as are used to further the hypothesis that (1967) argues that the concentration of
specialized tools for hunting big game. human overkill caused the late-Pleis- megafauna in these habitats caused com-
An example of these tools includes the tocene extinctions; however, there are petition such that several species became
fluted points found in Clovis sites in some aspects that they do not address. extinct, this accumulation also served
25
SURJ.
to concentrate prey for hunter-gatherer porting opposing hypotheses, works frames of these two events, it follows
groups. A high number of prey in a small very well to support combining the that the effects of climate and Homo
area increases the chance for success- effects of Homo sapiens with global sapiens on the megafaunal populations
ful hunting. Hildebrandt and McGuire climate change to explain megafaunal should not be differentiated either. Both
(2002) argue that hunter-gatherer groups extinction in a multifaceted model. hypotheses seem to have holes that are
pursue large game for the perceived fit- Whereas a resolution to the debate most logically filled with data from the
ness benefits gained by successful hunt- between the climate and overkill hypoth- opposing argument. Therefore, a multi-
ers. Thus, it makes sense that a higher eses may not be possible to reach, a com- faceted model, incorporating arguments
hunting success rate in a given area due promise might serve as a better conclu- and data from both sides of the debate,
to concentration of large game animals sion. The opposing arguments agree on makes the most sense in explaining the
into climate-shrunk habitats would en- the fact that the late-Pleistocene climate late-Pleistocene megafaunal extinc-
courage increased hunting. Combined changes occurred at approximately the tions.
with competition between species, this same time as the arrival of Homo sapiens
increase in hunting activities could very in the New World. There is simply no
well push a species to extinction. This separating the two events. Since there
blending of arguments, originally sup- is no differentiating between the time
References
Alroy, John. (2001).  A Multispecies Overkill Simulation of the End-Pleistocene Megafaunal Mass Extinction. Science 292:1893-
1896.
Edwards, William Ellis. (1967).  The Late-Pleistocene Extinction and Diminution in Size of Many Mammalian Species. In
Pleistocene Extinctions: The Search for a Cause. Paul S. Martin and H.E. Wright, Jr., eds. Pp. 141-154. New Haven: Yale
University Press.
Gingerich, Philip G. (1984).  Pleistocene Extinctions in the Context of Origination-Extinction Equilibria in Cenozoic Mammals. In
Quaternary Extinctions: A Prehistoric Revolution. Paul S. Martin and Richard G. Klein, eds. Pp. 211-222. Tucson: University of
Arizona Press.
Grayson, Donald K. (1984).  Archaeological Associations with Extinct Pleistocene Mammals in North America. Journal of
Archaeological Science 11(3):213-221.
Grayson, Donald K. (1991).  Late Pleistocene Mammalian Extinctions in North America: Taxonomy, Chronology, and Explanations.
Journal of World Prehistory 5(3):193-231.
Guilday, John E. (1967).  Differential Extinction During Late-Pleistocene and Recent Times. In Pleistocene Extinctions: The Search
for a Cause. Paul S. Martin and H.E. Wright, Jr., eds. Pp. 121-140. New Haven: Yale University Press.
Haynes, C. Vance. (1984).  Stratigraphy and Late Pleistocene Extinction in the United States. In Quaternary Extinctions: A
Prehistoric Revolution. Paul S. Martin and Richard G. Klein, eds. Pp. 345-353. Tucson: University of Arizona Press.
Hester, James J. (1967).  The Agency of Man in Animal Extinctions. In Pleistocene Extinctions: The Search for a Cause. Paul S.
Martin and H.E. Wright, Jr., eds. Pp. 169-192. New Haven: Yale University Press.
Hildebrandt, William R. and Kelly R. McGuire. (2002).  The Ascendance of Hunting During the California Middle Archaic: An
Evolutionary Perspective. American Antiquity 67(2):231-257.
Horton, David R. (1984).  Red Kangaroos: Last of the Australian Megafauna. In Quaternary Extinctions: A Prehistoric Revolution.
Paul S. Martin and Richard G. Klein, eds. Pp. 639-680. Tucson: University of Arizona Press.
Jelinek, Arthur J. (1967).  Man's Role in the Extinction of Pleistocene Faunas. In Pleistocene Extinctions: The Search for a Cause.
Paul S. Martin and H.E. Wright, Jr., eds. Pp. 193-200. New Haven: Yale University Press.
Kaplan, H. and K. Hill. (1985).  Food Sharing Among Ache Foragers: Tests of Explanatory Hypotheses. Current Anthropology 26:
223-245.
Kiltie, Richard A. (1984).  Seasonality, Gestation Time, and Large Mammal Extinctions. In Quaternary Extinctions: A Prehistoric
Revolution. Paul S. Martin and Richard G. Klein, eds. Pp. 299-314. Tucson: University of Arizona Press.
Martin, Paul S. (1967).  Prehistoric Overkill. In Pleistocene Extinctions: The Search for a Cause. Paul S. Martin and H.E. Wright,
Jr., eds. Pp. 75-120. New Haven: Yale University Press.
Martin, Paul S. (1984).  Prehistoric Overkill: The Global Model. In Quaternary Extinctions: A Prehistoric Revolution. Paul S.
Martin and Richard G. Klein, eds. Pp. 354-403. Tucson: University of Arizona Press.
McDonald, Jerry N. (1984).  The Reordered North American Selection Regime and Late Quaternary Megafaunal Extinctions. In
Quaternary Extinctions: A Prehistoric Revolution. Paul S. Martin and Richard G. Klein, eds. Pp. 404-439. Tucson: University
of Arizona Press.
Roberts, Richard G. et al. (2001).  New Ages for the Last Australian Megafauna: Continent-Wide Extinction About 46,000 Years
Ago. Science 292:1888-1892.
26
SPRING 2004
Slaughter, Bob H. (1967).  Animal Ranges as a Clue to Late-Pleistocene Extinction. In Pleistocene Extinctions: The Search for a
Cause. Paul S. Martin and H.E. Wright, Jr., eds. Pp. 155-168. New Haven: Yale University Press.
Whittington, Stephen L. and Bennett Dyke. (1984).  Simulating Overkill: Experiments with the Mosimann and Martin Model. In
Quaternary Extinctions: A Prehistoric Revolution. Paul S. Martin and Richard G. Klein, eds. Pp. 451-465. Tucson: University
of Arizona Press.
Robin Gibbons
Robin Gibbons is a senior majoring in Human Biology with an area of concentration in Evolutionary
Studies. At Stanford, she has been a member of the Leland Stanford Junior University Marching Band
and has served as a College Assistant at Freshman-Sophomore College. This is her first submission to
a Stanford publication. After graduating in 2004, she hopes to attend medical school. She would like
to thank Dr. Teresa Steele for all of her insightful comments and helpful suggestions in the writing of
this paper.
27