Domestication and early agriculture in the Mediterranean
Basin: Origins, diffusion, and impact
Melinda A. Zeder*
Archaeobiology Program, National Museum of Natural History, Smithsonian Institution, Washington, DC 20013
The past decade has witnessed a quantum leap in our understanding of the origins, diffusion, and impact of early agriculture in the
Mediterranean Basin. In large measure these advances are attributable to new methods for documenting domestication in plants and
animals. The initial steps toward plant and animal domestication in the Eastern Mediterranean can now be pushed back to the 12th
millennium cal B.P. Evidence for herd management and crop cultivation appears at least 1,000 years earlier than the morphological
changes traditionally used to document domestication. Different species seem to have been domesticated in different parts of the
Fertile Crescent, with genetic analyses detecting multiple domestic lineages for each species. Recent evidence suggests that the ex-
pansion of domesticates and agricultural economies across the Mediterranean was accomplished by several waves of seafaring
colonists who established coastal farming enclaves around the Mediterranean Basin. This process also involved the adoption of do-
mesticates and domestic technologies by indigenous populations and the local domestication of some endemic species. Human envi-
ronmental impacts are seen in the complete replacement of endemic island faunas by imported mainland fauna and in today’s
anthropogenic, but threatened, Mediterranean landscapes where sustainable agricultural practices have helped maintain high bio-
diversity since the Neolithic.
he transition from foraging and these developments and highlighting ronment also strongly influences body
hunting to farming and herding promising areas for future study. size, with increasing heat and aridity
is a significant threshold in hu- positively correlated with smaller size.
man history. Domesticates and Initial Animal Domestication in the What archaeozoologists had originally
the agricultural economies based on Fertile Crescent interpreted as body size reduction asso-
them are associated with radical restruc- Until the late 1990s archaeozoologists ciated with initial domestication can
turing of human societies, worldwide relied on morphological changes in tar- now be attributed to differences in the
alterations in biodiversity, and signifi- get species to identify where and when culling strategies of herders as opposed
cant changes in the Earth’s landforms wild prey animals were transformed into to hunters. In most prey species, hunters
and its atmosphere. Given the momen- herded livestock (2). A proposed sharp focus on large adult animals (particu-
tous outcomes of this transition it comes and rapid reduction in overall body size larly males) to maximize return, and the
as little surprise that the origin and among archaeological prey populations bones of these larger animals generally
spread of domesticates and the emer- was the most widely accepted morpho- dominate in prey assemblages generated
gence of agriculture remain topics of logical marker of this threshold (3, 4). by hunters. Archaeological assemblages
enduring interest to both the scholarly Based on this size reduction criterion, generated by herders, on the other
community and the general public. the established consensus was that ani- hand, are usually dominated by the
The past decade has seen remarkable mal domestication (beginning with goats bones of smaller females slaughtered
analytical advances in documenting do- and then sheep) occurred at ca. 10,000– after their prime reproductive years. Ex-
mestication (1), particularly in tracking 9,500 B.P.†, 1,000 years after the cess males not needed for herd propaga-
the domestication of four major Near domestication of crop plants in the tion were harvested at young ages and
Eastern livestock species (sheep, goats, southern Levant (3, 5). Domestication of their more friable bones are usually less
cattle, and pigs) and their subsequent these two animal species was thought to well represented in these assemblages.
dispersal throughout the Mediterranean have occurred somewhere to the north Although the linkage between domes-
Basin. New morphometric methods are and east of the heartland of plant do- tication and body size was called into
tracking changes in human prey strate- mestication (5), although a second, inde- question by this research, the marked
gies that mark the transition from hunt- pendent domestication of goats was degree of sexual dimorphism in caprines
ing to herding. Genetic analyses bring proposed for the southern Levant (6). it documented offered another approach
fresh insights into initial livestock The utility of this size reduction to tracking the transition from hunting
domestication and their dispersal. Small- marker, and indeed of all morphological to herding. Pronounced differences in
sample atomic mass spectrometry markers, has come under increasing the size of male and female skeletal
radiocarbon dating provides refined scrutiny (7). My own work on both mod- elements make it possible to separate
chronological frameworks for these de- ern skeletal collections and archaeologi- archaeological assemblages into
velopments. These recent analytical cal caprine (sheep and goat) remains sex-specific subpopulations, which, based
advances, in turn, have produced an ex- from the Near East finds little support
plosion of new information that is call- for the almost axiomatic acceptance that
This article grew out of a presentation given by M.Z. at the
ing into question prevailing hypotheses domestication results in an automatic
Calpe 2007 Symposium: People in the Mediterranean–A
about the origin and early spread of ani- overall reduction in body size in man- History of Interaction, September 27–30, 2007, the Gibraltar
mal domesticates and the Neolithic life- aged animals (ref. 8 and references Museum, Gibraltar.
ways of which they were a part. Here, I therein). Rather than domestic status, Author contributions: M.A.Z. designed research, per-
bring together these different sources of sex is the primary factor affecting body formed research, synthesized research in referenced publi-
information to consider the origins, dif- size in these ungulates, manifested by a cations and wrote the paper.
fusion, and impacts of domesticates and marked and consistent difference be- The author declares no conﬂict of interest.
agriculture in the Mediterranean Basin, tween larger males and smaller females *E-mail: email@example.com.
outlining our current understanding of in essentially all skeletal elements. Envi- †All dates are reported in calibrated years before present.
www.pnas.org cgi doi 10.1073 pnas.0801317105 PNAS August 19, 2008 vol. 105 no. 33 11597–11604
on a refined understanding of the se-
quence and timing of long bone fusion
(9), can be used to generate high-
resolution harvest profiles for male and
female animals capable of distinguishing 10,000
the prey strategies of hunters from the 10,000 CATTLE GOATS
harvest strategies of herders. 8,500
Application of this approach to ar- 9,600 9,000
chaeological assemblages from Iraq and
Iran has identified the clear signature of 10,500
a managed herd of goats (harvesting of 10,500 10,000
young males and prolonged survivorship 9,600
of females) at the site of Ganj Dareh in 9,000 8,500
highland Iran (8). Directly dated to 8,500 8,500
9,900 B.P., the goats from this site show 8,500
no evidence of size reduction or any
other domestication-induced morpholog-
ical change. Smaller body size and Fig. 1. The origin and dispersal of domestic livestock species in the Fertile Crescent. Shaded areas show
changes in the size and shape of horns the general region and the approximate dates in calibrated years B.P. in which initial domestication is
[a morphological change clearly linked thought to take place. Dates outside of the shaded areas show the approximate date when the domes-
to domestication (2)] appear 500–1,000 ticate ﬁrst appears in a region. Orange, goats (Capra hircus); blue, sheep (Ovis aries); green, cattle (Bos
years later than this demographic shift, taurus); fuscia, pigs (Sus scrofa).
when managed animals were moved
from the natural habitat of wild goats
and introduced into hotter and more from hunting to herding in this region, pigs, they arrived relatively late in more
arid lowland Iran. These follow-on mor- with managed goats arriving from out- distant parts of the Fertile Crescent
phological changes likely reflect re- side the area at ca. 10,200 B.P. (12). (Fig. 1). Morphologically altered domes-
sponses to new selective pressures, plus Similarly, demographic evidence for the tic pigs and cattle are not found in Cen-
the now more limited opportunities for management of morphologically unal- tral Anatolia until after 8,500 B.P. (16).
introgression between managed and wild tered caprines (mostly sheep) is found in Genetic data from modern and ar-
animals or the restocking of herds with Central Anatolia between 10,400 and chaeological specimens both support
wild animals. 9,400 B.P. (13). These results suggest and enhance this picture of initial ani-
Looking back before Ganj Dareh, un- that both sheep and goats were brought mal domestication. Recent work has suc-
usual demographic profiles detected in under domestication (probably indepen- ceeded in definitively identifying the
sheep bone assemblages from northeast- dently of one another and possibly mul- progenitors of both domestic sheep and
ern Iraq (10) and southeastern Anatolia tiple times) in the region that stretches goat as belonging to species found in
(11) dating to 12,000 B.P. may reflect from the northern Zagros to southeast- the Fertile Crescent (Ovis orientalis and
early attempts at manipulating herd de- ern Anatolia between ca. 11,000 and Capra aegagrus, respectively) (17, 18).
mographics to maximize returns. These 10,500 B.P., and perhaps even earlier Moreover, in both of these livestock spe-
assemblages show an almost exclusive (Fig. 1). Morphologically wild, but man- cies there are at least four and, in the
focus on 2- to 3-year-old males, which is aged, goats appear to have been moved case of goats, as many as six (19), genet-
older than expected with herd manage- relatively rapidly through the region, ically distinguishable domestic lineages,
ment but younger than expected with reaching the southernmost tips of both or haplotypes. It is not entirely clear,
hunting. This pattern is argued to result the eastern and western arms of the however, whether these different lin-
from a prime male hunting strategy de- Fertile Crescent by ca. 9,500 B.P. Do- eages represent spatially and temporally
veloped under conditions of intensive mestic sheep were spread more slowly discrete ‘‘domestication events’’ in which
pressure on local wild herds during the and first appear in these regions 500– different populations of animals were
Younger Dryas climatic downturn (11). 1,000 years later than managed goats brought under domestication indepen-
The proposed scenario suggests that, (10, 12). dently of one another (20). Genetic data
rather than broadening the prey strategy Recent research has also clarified the for taurine cattle have identified five
to include both males and females, hunt- spatial and temporal context of the do- different domestic haplotypes, at least
ers conserved female breeding stock mestication of two other major livestock three and possibly four of which origi-
while at the same time relying on a species in the Near East: pigs and cattle. nated in the Fertile Crescent (21). Simi-
steady immigration of younger males Archaeological evidence now suggests larly, as many as four of the many
drawn from surrounding territories to that pigs were first domesticated some- different lineages of domestic pigs
fill the vacuum left by the kill-off of where in southeastern Anatolia by originated in the Near East (22, 23).
local prime-age males. 10,500–10,000 B.P. and that the timing Animal domestication in the Near
Lower-resolution demographic meth- of their geographical expansion as do- East can then be seen as arising from a
ods used by archaeozoologists working mesticates was similar, although perhaps period of prolonged human interaction
elsewhere in the Fertile Crescent are slower, to that of sheep (Fig. 1) (10, 14). with the ancestors of core livestock spe-
detecting parallel patterns to those doc- Morphologically altered domestic pigs cies that unfolded across much of the
umented in the Zagros. Changes in the are not found in the southern Levant or Fertile Crescent. Over time hunting
age of harvested caprines, and possibly lowland Iran until ca. 8,500–8,000 B.P. strategies aimed at maximizing local
demographically driven changes in size Recent demographic evidence suggests availability of wild ungulates developed
consistent with early herd management, that taurine cattle were initially domesti- into active management, with all four
are found in southeastern Anatolia at cated somewhere in the upper Eu- major livestock species coming under
ca. 10,500 B.P. (12). Sheep seem to be phrates Valley between ca. 11,000 and management over a period from ca.
the initial early focus of the transition 10,000 B.P. (15), but, like sheep and 11,000 to 10,000 B.P. Even species like
11598 www.pnas.org cgi doi 10.1073 pnas.0801317105 Zeder
gazelle, which are behaviorally unsuited
to domestication, may have been audi-
tioned for management in the southern
and northern Levant, where they were 000
the most abundant wild ungulate (11).
Clear-cut morphological responses to 10
domestication (i.e., changes in horns in
bovids and tooth size in pigs) are not
evident in these four livestock species
until ca. 9,500–9,000 B.P. 7,000
As is the case with animal domestica- 0
tion in the Near East, the leading edge
of plant domestication in the region is 7,000
now recognized as an extended process
(24, 25). Evidence from multiple loca-
tions point to a prolonged period of hu- Fig. 2. An integrated model of the Neolithic expansion in the Mediterranean Basin. The location of
man manipulation of morphologically colonist farming enclaves is shown in the red ellipses. Approximate dates of these enclaves are given inside
wild, but possibly cultivated, plants the ellipses in calibrated years B.P. Red dots represent areas that are proposed to have been settled by
which, in certain species, resulted in the colonist farmers; green dots indicate areas where indigenous foragers adopted elements of the Neolithic
development of morphologically altered package; and blue dots indicate areas of proposed integration of colonist farmers with indigenous
domesticated crops (26–28). This period foraging groups. Data were complied from refs. 52, 54, 56, 57, and 65 and ﬁgure 7.1 of ref. 74.
of intensified plant management dates at
least as far back as ca. 12,000 B.P., with
morphological markers of crop domesti- of incipient cultivation, if not domestica- gatherers and colonizing farming popu-
cation (i.e., nonshattering seed heads in tion, of local wild plants (35). Evidence lations from the east (41–46).
cereals) not well established until ca. for indigenous animal domestication was The shortcomings of these different
10,500 B.P. (24, 25, 29). Agricultural based on the identification of wild sheep single-agent models have become
economies reliant on a mix of domesti- in Pleistocene age deposits in southern increasingly apparent as new archaeo-
cated crops and livestock apparently do France and the presence of domestic logical data come to light and older
not fully crystallize in the region until sheep and goat remains in Mesolithic collections are reanalyzed by using new
ca. 9,500–9,000 B.P. (5, 10). contexts in France and Spain (36, 37). methods and new perspectives. Recent
Reports of domesticated pig and cattle genetic analyses of livestock species and
The Diffusion of Animal Domesticates in remains in Mesolithic (pre-8,000 B.P.) their progenitors have also contributed
the Mediterranean Basin levels from sites in southern Spain (38) important new insights into this process.
The last two decades has witnessed the were also cited as evidence for the local As a result, a much more complex, and
rise and fall of a number of models of domestication of these species. more interesting, scenario is emerging
Neolithic expansion across the Mediterra- Genetic studies have subsequently for the Neolithic transition across the
nean Basin. In the early 1980s Ammer- ruled out European ancestry for domes- Mediterranean Basin.
man and Cavalli-Sforza (30) combined tic wheat, barley, and pulses, confirming Beginning in the early 1990s a num-
archaeological and human genetic data to the Near East as the source of these ber of sites have been discovered and
frame their ‘‘wave and advance’’ model. crops (26, 39). Morphological, cytologi- excavated on Cyprus that have radically
This model attributed the westward cal, hemoglobin, and, most recently, ge- transformed our understanding of Neo-
spread of the Neolithic to Near Eastern netic studies have shown that the ‘‘wild’’ lithic emergence in the Mediterranean
colonists who, driven by agriculture-fueled sheep and goats found on Mediterra- Basin (47). Until the early 1990s Cyprus
population growth, slowly pushed aside nean islands, once argued to be the de- was thought to have been colonized ca.
indigenous hunter–gatherers at a pre- scendents of the progenitors of indige- 8,500 B.P. by a derived offshoot of fully
dicted average pace of 1 km per year. nous domestic caprines, are instead the established Neolithic mainland cultures
Objecting to the passive role this feral descendents of Near Eastern ca- (48). The new sites, however, date 2,000
model assigned to indigenous Mesolithic prines (for a review in chronological years earlier (10,500–9,000 B.P.) and
people, a number of researchers subse- order, see refs. 40, 41, 17, and 18). document the arrival of early pioneers
quently countered with alternative mod- Ruling out the indigenous domestica- hypothesized to have originated some-
els that awarded local populations a tion of caprines and major crop plants where in the Northern Levant (Figs. 1
starring role in Mediterranean Neolithic did not, however, lead to an embrace of and 2) (47, 49). Traveling the 60 k to
emergence. Early models within this in- colonist expansion diffusion models for Cyprus by boat, these colonists trans-
digenist perspective argued for autocho- the emergence of Neolithic lifeways in ported the full complement of economi-
nous domestication of crops and live- the Mediterranean. Instead, researchers cally important mainland fauna (50).
stock in a process parallel to, but subscribing to the indigenist perspective, including all four major livestock species
independent of, the Near East (31–33). especially those working in the western (sheep, goat, cattle, and pig). Early
The presence of wild oats, barley, and Mediterranean, argued that cultural and colonists also imported mainland game
lentils in Upper Paleolithic and Meso- not demic diffusion was the primary en- animals like fallow deer and fox that,
lithic levels at Francthi Cave on the gine driving this transition. Specifically, although perhaps kept in captivity (48),
eastern coast of Greece, followed by the proponents maintained that the selective were never truly domesticated. None of
appearance of fully domesticated barley adoption of various elements of the these animals are endemic to Cyprus.
and lentils in later Neolithic levels, was Neolithic package by indigenous popula- Although imported livestock species did
interpreted as evidence for the local tions around the Mediterranean could not show any of the morphological fea-
crop domestication (34). Legumes recov- have happened through trade and tech- tures traditionally used to mark domes-
ered from Mesolithic cave deposits in nology transfer alone without any direct tic status when they arrived on the
southern France were seen as evidence contact between indigenous hunter– island, demographic profiles of these
Zeder PNAS August 19, 2008 vol. 105 no. 33 11599
animals are consistent with human man- between interior Mesolithic settlements cultures focusing on the intensive exploi-
agement. In contrast, demographic pro- and coastal Neolithic colonies (58) Re- tation of estuary resources persist for
files of the fallow deer are indicative of cent excavation of a coastal settlement several hundred years after the estab-
hunting, suggesting that early colonists in southern France, dating to 7,700– lishment of these farming enclaves and
were engaged in game stocking and 7,600 B.P. and characterized as a beach- that the subsequent spread of agricul-
herd management (13, 48). Deep wells head colony of seafaring migrant farm- tural economies into the interior likely
constructed at one of these early sites ers from mainland Italy, has yielded proceeded through a combined process
yielded abundant evidence of domesti- pottery, domestic sheep, einkorn, and of colonist expansion, selective local
cated einkorn and emmer wheat and emmer wheat (59). adoption of Neolithic technologies, and
lentils, none of which are native to Questions have also been raised about the integration of colonist and indige-
Cyprus, and domestic barley, which in the evidence for the early occurrence nous populations. Similar patterns of
the wild is endemic to the island (26, of domestic animals and pottery in Me- development are hinted at in interior
51). Other introduced plants include solithic contexts in the western Mediter- and northern Italy, which seem to lag
pistachios and flax, as well as figs possi- ranean, which had formed a primary several hundred years behind coastal
bly domesticated in the Levant by this foundation of earlier culture diffusion areas in the appearance of plant and
time (28). Thus the initial diffusion of models. Based on a reappraisal of the animal domesticates and other markers
the nascent Neolithic package out of the complicated cave stratigraphy of Iberian of Neolithic adaptations (56, 57). In
Fertile Crescent to Cyprus involved the sites and a reanalysis of their associated southern France, the initial, essentially
transplant of all aspects of daily life ˜
radiocarbon dates, Zilhao (60–62) ar- exclusive, focus on domestic livestock
(i.e., subsistence resources, technologies, gues that the pottery and domesticated evidenced at the early coastal pioneer-
and, most likely, social networks and caprines recovered from Mesolithic lev- ing sites stands in stark contrast to sub-
belief systems) by seafaring colonists els actually derive from higher Neolithic sistence strategies of later interior sites
who, for unclear reasons, were seeking a levels. Domestic sheep reported as re- that show persistence of hunting along
fresh start in a new land (52). Far from covered from Mesolithic and earlier de- with the utilization of domesticates, a
being an isolated event, the colonization posits in southern France can also be pattern that points to the blending of
of Cyprus provides a clear and valuable argued to have been derived from over- Neolithic and Mesolithic traditions after
template for the subsequent diffusion of lying Neolithic contexts, or, in the case initial colonization (65). Farther east,
the Neolithic across the rest of the Med- of higher elevation sites, represent misi- the disjunction between later Neolithic
iterranean Basin. dentified native chamois (Rupicapra sites and their Mesolithic and early Neo-
Recent archaeological evidence from rupicapra) and European ibex (Capra lithic predecessors in the Aegean signals
the Aegean, for example, no longer sup- ibex) (41, 58, 63). Similarly, Rowley- a similar process of dispersal, adoption,
ports a model of gradual in-place transi- Conwy’s (64) reexamination of the argu- and integration (54) (Fig. 2).
tion of ancestral Mesolithic cultures into ment for the domestic status of pigs and Genetic studies of modern and an-
Neolithic cultures (53–55). Instead, cattle in Mesolithic contexts in southern cient DNA from Mediterranean Basin
there appears to have been a sharp Spain suggests that the smaller size of livestock species and their progenitors
decline in Late Mesolithic population these animals is not a sign of their do- adds further support, and nuance, to
levels, combined with the sudden ap- mestication as originally argued, but is this emerging picture. A study of ancient
pearance of radically different Neolithic instead a reflection of body size re- mtDNA has shown that two haplotypes
settlements in previously unoccupied sponses to different climatic regimes of domestic goats (the A and C lin-
locations. As on Cyprus, recent work in among native wild animals. eages) had arrived in southern France
the Aegean argues for the arrival of Having discounted evidence for piece- by 7,300 B.P., suggesting their dispersal
maritime colonists who, at ca. 9,000 to meal cultural diffusion of various ele- out of the Near East as a single package
8,000 B.P., carried many components of ments of Neolithic economy and their (66). Among modern goat breeds in
the full Neolithic package (plant and selective adoption by indigenous Meso- Portugal researchers have found both
animal domesticates, new lithic tradi- lithic populations in the western Medi- the ubiquitous A haplotype and the
tions, and, perhaps a bit later, pottery) ˜
terranean, Zilhao (61, 62) has gone on much more restricted haplotype C (67).
(Fig. 2). Following a leapfrog pattern, to demonstrate that, as in other parts of Three domestic lineages were found
these seafaring pioneers established the Mediterranean Basin, the Late Me- among modern breeds of sheep in Por-
farming communities that selectively solithic of the Iberian Peninsula was a tugal, including those previously found
focused on favorable environments in period of population decline and reloca- only in the Middle East and Asia (68).
coastal Greece and on various Aegean tion. Also as elsewhere, Neolithic settle- Both Portuguese sheep and goat show a
Islands. ments with apparently fully formed much higher degree of within-breed ge-
Based on a careful reevaluation of agro-pastoral economic systems sud- netic diversity than expected at the west-
archaeological evidence, especially avail- denly appear in the Iberian Peninsula as ernmost periphery of sheep and goat
able radiocarbon dates, researchers now coastal enclaves occupying limestone- expansion. This diversity is attributed to
see major discontinuities between Meso- based soils abandoned by earlier Meso- multiple introductions of caprines into
lithic and Neolithic cultures in Italy (56, lithic peoples. The initial establishment the Iberian Peninsula, not only through
57). They argue that Neolithic lifeways of these colonies follows a familiar pat- maritime colonization from Italy and
were introduced into the Italian penin- tern, with farming enclaves appearing in France, but through subsequent intro-
sula ca. 8,000 B.P. by maritime colonists favorable coastal locals around the pe- ductions out of Africa or overland
who first established farming villages on riphery of the Iberian Peninsula at a through Europe.
the Apulian ‘‘boot heel’’ region of steady and quite rapid pace, appearing Genetic data also support a pattern of
southeastern Italy (Fig. 2). These tradi- first on the eastern and southern coasts multiple introductions of cattle into the
tions appear in northwest coastal Italy of Spain at ca. 7,700–7,600 B.P. and on region. The T3 haplotype of domestic
200–300 years later (ca. 7,800–7,600 the Atlantic coast of Portugal ca. 7,400– cattle, which dominates among modern
B.P.). In southern France, a compelling 7,300 B.P. (Fig. 2). and ancient European cattle, seems to
case can be made for a marked geo- ˜
However, Zilhao’s (61, 62) work in have followed a relatively rapid path of
graphic, ecological, and cultural break Portugal has also shown that Mesolithic expansion around the Mediterranean
11600 www.pnas.org cgi doi 10.1073 pnas.0801317105 Zeder
Table 1. The extinction of Late Pleistocene large endemic mammals and human colonization of
Gymnesic Islands/ Megaloceros Crete/ Cyprus/
Time period Myotragus balearicus cazioti Candiacervus sp. Phanourios minutus
Most recent endemic 5,000* 10,000† 9,000 11,500*†
Earliest human presence 7,000 10,000 None 11,500
Neolithic colonization 4,000 7,600 9,000 10,500
All dates in calibrated years B.P. References are as follows: Gymnesic Islands (69), Tyrrhenian Islands (57), Crete (62), and Cyprus (39, 61).
*Directly dated skeletal element.
†Radiocarbon-dated stratigraphic context.
Basin without any significant introgres- two major European clades indicative of expansion across the Mediterranean
sion with female European aurochsen two separate domestication events, and (65, 74) (Fig. 2).
(refs. 21, 69, and 70 and contra ancient an additional clade, currently restricted
DNA evidence reported in ref. 71). to Italy and Sardinia, representing an- Environmental Impacts
Modern DNA, however, indicates that T other (22, 23). The impact of Neolithic economies on
and T2 haplotype cattle were included Thus it appears that none of the ear- the biotic communities of the Mediter-
in migratory movements into the Bal- lier models for Neolithic emergence in ranean Basin is most clearly seen on the
kans and Central Europe (71). T1 cattle, the Mediterranean accurately or ade- large islands scattered across the region,
which dominate among modern North quately frame the transition. Clearly where highly endemic and disharmonic
African taurine cattle, were initially ar- there was a movement of people west- faunas were replaced by a mixture of
gued to represent a separate North Afri- ward out of the Near East all of the way domestic and wild mainland fauna (75–
can domestication event (21). This lin- to the Atlantic shores of the Iberian 77). Although humans are clearly the
eage now seems more likely to have Peninsula. But this demic expansion did agent of island introductions of main-
been brought under domestication with not follow the slow and steady, all- land faunas, their role in the extirpation
other T haplogroup cattle in the Near encompassing pace of expansion pre- of endemic island faunas is unclear.
East (72) and subsequently radiated dicted by the wave and advance model. Once again, Cyprus reflects a general
across North Africa through trade and Instead the rate of dispersal varied, with pattern for the Mediterranean Basin.
human migration. The patchy occur- Neolithic colonists taking 2,000 years to The endemic mammalian fauna of
rence of the TI haplotype among mod- move from Cyprus to the Aegean, an- Cyprus was impoverished and unbal-
ern cattle in the Iberian Peninsula, Sicily other 500 to reach Italy, and then only anced, limited to pygmy hippopotamus
and central Italy, and the Balkans sug- 500–600 years to travel the much (Phanourios minutus), a pygmy elephant
gests that T1 cattle entered southern greater distance from Italy to the Atlan- (Elephas cypriotes), a genet (Genetta
Europe out of North Africa through tic (52). Moreover, rather than entirely plesictoides), the only carnivore on the
multiple points of entry (71). It is also replacing or engulfing indigenous forag- island), and a mouse (Mus cyprinacus,
possible that T1 cattle traveled overland ing populations, these colonists seem to the only endemic to survive to the
across the Dardanelles into Eastern Eu- have been restricted to scattered coastal present day) (76, 78). None of the larger
rope. The high-diversity T haplogroup farming enclaves established around the endemics are represented among the
taurine cattle found among modern Tus- Mediterranean Basin. Although cultural imported mainland fauna associated
can cattle has been linked to a post- diffusion can no longer be argued to with the sites of colonists of the 11th
Neolithic migration of Etruscans who, provide a universal explanation for Neo- millennium B.P. (50). It is now clear
based on both historical evidence and lithic expansion into the Mediterranean, that these pioneer settlers were not the
modern human genetic data, are be- it is clear that the movement of Neo- first humans on Cyprus and that main-
lieved to have been of Eastern Mediter- lithic lifeways out of these beachhead land hunters made periodic visits to the
ranean origin (73). settlements involved selective adoption island 1,000 years earlier during the
Pigs tell a different story. Research and adaptation of elements of the Neo- Younger Dryas climatic downturn (79)
by Larson et al. (22) has shown that lithic package by indigenous peoples. (Table 1). Simmons (79) argues that a
current-day domestic pigs in Europe Moreover, although caprines, cattle, and large accumulation of pygmy hippopota-
bear no trace of Middle Eastern ances- primary crop plants were most certainly mus remains found in a collapsed sea-
try, but instead are most closely related not independently domesticated in Eu- side rock shelter is directly associated
to European wild boar. Subsequent rope, recent genetic data for pigs points with an overlying, but apparently con-
analysis by the same team of mtDNA to indigenous domestication of local temporaneous, stratum containing stone
extracted from archaeological remains wild boar, possibly occurring multiple tools and hearths dated to ca. 11,775
has found convincing evidence for the times in geographically separate sub- B.P. Other researchers, although ac-
dispersal of Near Eastern pigs into and populations. Genetic studies of rye and cepting the evidence for early visits of
across Europe between 7,500 and 5,000 oats also indicate that the modern vari- hunter–gatherers to the island, question
B.P. (23). Surprisingly, subsequent to eties of these major crop plants have a the evidence for human predation on
this initial diffusion, Near Eastern swine European and not Near Eastern ances- the endemic mammals discovered at the
are later replaced by domestic pigs of try (24). Future interpretive frameworks site (refs. 80 and 81, but see ref. 82).
European maternal ancestry, even will have to take a more integrated ap- Large endemic mammals on Crete
within the Near East. Indigenous do- proach, which recognizes colonization, included pygmy hippopotamus (Hippo-
mestication of European boar also ap- diffusion, and independent domestica- potamus creutzburgi), elephants (Elephas
parently happened several times, with tion as all playing a role in Neolithic creutzburgi), and several species of
Zeder PNAS August 19, 2008 vol. 105 no. 33 11601
megalocerine deer (Candiacervus sp.) and the last documented Myotragus characterized by highly anthropogenic
(76). Dating the last appearance of specimens on the island dated at ca. environments shaped by thousands of
these species has proven problematic 5,000 B.P.. A case was even mounted years of human landscape management,
(83), although it appears that the Cre- for domestication of this species before species introductions, and associated
tian hippopotamuses and elephants be- its extinction (90), which, if true, made responses by indigenous faunas and flo-
came extinct before the endemic deer. Myotragus the only domestic animal to ras, all dating to the Neolithic emer-
There is as yet no evidence for a tempo- undergo extinction. Recent reevaluation gence. They also note that the various
ral overlap between humans and larger of the primary arguments for Myotragus forms of traditional human landscape
endemic mammals on Crete. As on domestication has, however, overturned engineering in the Mediterranean have
Cyprus, Cretian endemics do not occur this hypothesis (89, 91). A reexamina- created viable, sustainable ecosystems,
among the faunal remains recovered tion of the dates for earliest human oc- which have, in fact, been highly benefi-
from the earliest known Neolithic settle- cupation of these islands and the most cial to Mediterranean biodiversity. This
ment on the island, which dates to ca. recent evidence for Myotragus, more- balanced system is undergoing increas-
9,000 B.P. (84) (Table 1). It is possible over, has all but eliminated any overlap ing threat from urban growth and agri-
that the larger endemic mammals of between humans and Myotragus (89). cultural intensification, however, threats
Crete became extinct long before Neo- The earliest evidence of substantial hu- that can only be met with a clear under-
lithic farmers and herders colonized the man settlement of Mallorca dates to ca. standing of the long-term role of human
island, thus limiting the island’s appeal 4,000 B.P., whereas the most recent management in shaping current-day
to earlier mainland hunting parties. It is solid evidence for Myotragus on the biodiversity in the Mediterranean Basin.
also possible, however, that the ephem- island dates to ca. 5,000 B.P., with Myo-
eral camps of these hunters have yet to tragus extinction and initial human colo- Future Research
be discovered. nization happening sometime between Recent research on the initial develop-
A closer connection between humans these two dates (Table 1). ment and subsequent expansion of domes-
and now extinct endemic island faunas has Even lacking definitive evidence for tication and agricultural economies in the
been proposed for the islands of Sardinia humans involvement in the extirpation of Mediterranean Basin provides a clear
and Corsica. The case for an overlap be- larger endemic mammals on Mediterra- roadmap for future research. This is espe-
tween humans and indigenous megalocer- nean islands, the progressive east-to-west cially true for the Fertile Crescent where
ine deer (Megaloceros cazioti) on Sardinia disappearance of larger endemics coinci- recent advances are transforming our un-
in the Upper Paleolithic (ca. 20,000 B.P.) dent with human settlement of Mediterra- derstanding of the origins of plant and
(85) is contested (86). But there is firmer nean islands (Table 1) clearly suggests animal domestication in this key heartland
evidence that Mesolithic hunter–gatherers that humans played some role in their region. Traditional approaches to docu-
at least visited (if not colonized) Sardinia extinction. Having evolved in the absence menting domestication relied on the
by ca. 10,000 B.P. and that they encoun- of major predators, these larger herbivores appearance of genetically driven morpho-
tered endemic deer and, possibly, the is- probably lacked protective behaviors, logical change (i.e., the development of
land’s sole carnivore, a small fox-size making them especially vulnerable to sus- nonshattering seed heads in cereals and
canid (Cynotherium sardous) (refs. 76 and tained human predation. Moreover, all of body size reduction in animals). The de-
86 and Table 1). Neither of these species the species that became extinct around the velopment of new analytical approaches
survived much beyond this initial contact, time of initial human settlement likely had has, however, provided a window into the
however, and there is no evidence of over- low reproductive rates, a trait commonly preceding processes of human interaction
lap between Mesolithic hunter–gatherers found among island endemics, which fur- with target plant and animal species and
and these two endemics on Corsica (76). ther limited their ability to withstand any the genetic responses to this interaction
A number of sites on both islands attest to degree of hunting pressure (76). Extermi- that eventually resulted in morphological
heavy utilization of smaller endemic mam- nation of these island endemics by hu- change. Researchers in the Fertile Cres-
mals, especially a rat-size endemic lago- mans could have happened within a cent are detecting early signs of human
morph (Prolagus sardus), which seems to generation. Smaller endemic mammals ecosystem engineering aimed at encourag-
be the primary source of animal protein seem to have survived initial human colo- ing plant production (24, 25); they are
until Neolithic colonizers brought domes- nization somewhat better, perhaps as a able to document the manipulation of
tic livestock to these islands at ca. 7,600 result of their higher reproductive rates herd structure to promote a secure and
B.P. (87). All of these smaller endemics, and because they were less attractive prey predictable yield of animal products (7, 8,
however, were extripated sometime be- species. However, they, too, eventually 10, 11). In both plants and animals these
tween the Late Roman period and the could not withstand the combination of new indicators precede the manifestation
early Middle Ages, probably through com- overhunting, loss of habitat, and competi- of traditional morphological markers of
bined pressures associated with the intro- tion with invasive mainland imports. The domestication by hundreds, if not thou-
duction of Rattus rattus during the Early almost complete turnover of island faunas sands, of years. Estimating exactly when
Roman colonization of the island and throughout the region [involving essen- during this extended coevolutionary pro-
subsequent episodes of deforestation and tially 100% of mammal species, plus many cess a plant or animal species crossed the
agricultural intensification (75, 88). avian and herpetological species (77)] and domestic threshold is now more a seman-
The apparent exception to this pat- their replacement with domestic livestock, tic issue than a substantive research ques-
tern of rapid extinction of endemic is- game species, and an array of anthropolo- tion (20, 93). Although some researchers
land fauna just before, or slightly after, philous small vertebrate species (76) may require the appearance of specific
substantial human colonization was, clearly implicates humans in these island morphological traits before conferring
until recently, thought to be a small extinctions. domestic status, others may be more will-
derived caprine species (Mytotragus The impact of the Neolithic transition ing to consider a managed animal or a
balearicus), found only on the Gymnesic on mainland environments throughout cultivated plant as having achieved this
islands off the eastern coast of Spain the Mediterranean, although perhaps status. Regardless of where one chooses
(89). Here initial dating argued for a less dramatic, is no less pronounced. to draw the line between wild and domes-
3,000-year overlap between initial hu- Blondel and Aronson (92), in fact, argue tic, recent advances have provided re-
man presence on Mallorca at 8,000 B.P. that the entire Mediterranean Basin is searchers with powerful new tools capable
11602 www.pnas.org cgi doi 10.1073 pnas.0801317105 Zeder
of examining the entire process of domes- all call out for additional investigation. pigs, however, still cannot be ruled out.
tication, not just its morphological impacts The southern margin of the Mediterra- Application of enhanced archaeological
which, if they occur at all, only appear nean Basis along coastal North Africa is and genetic techniques for detecting and
after the process is well underway. essentially terra incognita for understand- dating domestication to both extant and
Just how far back this process of active ing the course of Neolithic emergence and yet-to-be-recovered assemblages is key to
human resource management goes or how seems an especially promising region for understanding patterns of indigenous do-
widespread it was in the Fertile Crescent future research (60). mestication around the Mediterranean
is, at this point, an open question. The The subsequent inland transfer of do- Basin.
early dispersal of an integrated economy mesticates, agriculture, and associated Finally, although we may never be able
based on crop plants and managed ani- Neolithic lifeways from newly arrived ˆ
to detect the final coup de grace for en-
mals to Cyprus at least 2,000 years before colonists to indigenous populations demic island faunas, the future holds con-
the apparent crystallization of agricultural around the Mediterranean Basin is an- siderable promise for a much fuller
economies on the mainland, however, sug- other intriguing research area that will understanding of the human impact on
gests that our understanding of plant and benefit from recent advances in our Mediterranean biodiversity. As it has in
animal domestication and agricultural ability to detect and date domesticates the Gymnesic islands and on Cyprus,
‘‘carpet-dating’’ large numbers of archaeo-
emergence on the mainland is, at best, in the archaeological record. Careful
logical materials by the small-sample
incomplete. Researchers working in this analysis of increasingly more precise ra-
atomic mass spectrometry radiocarbon
area are only just beginning to realize the diocarbon dates will continue to be criti-
method will certainly help refine the chro-
potential of these newly available analytic cal in discriminating between demic nology of the disappearance of endemic
tools. Additional even more effective ana- diffusion and selective adoption of Neo- island faunas and the arrival of human
lytical approaches will almost certainly be lithic components in different parts of colonists. Application of demographic
developed in the future as researchers the Mediterranean (e.g., ref. 94). New profiling techniques to the remains of
embrace this broader concept of domesti- demographic techniques for profiling these animals may make it possible to dis-
cation and begin to exploit the many op- prey strategies, morphometric tech- tinguish between natural-death accumula-
portunities available in the Fertile Cres- niques capable of tracking genetic and tions and prey assemblages resulting from
cent region for documenting it. plastic responses to human management, human predation. Similarly, recognition of
Recent research has also shown that the isotopic analysis, and the increasing suc- the broader role of humans in shaping
dispersal of domesticates and the Neo- cess of ancient DNA studies of domesti- post-Neolithic environments is central to
lithic way of life west across the Mediter- cates will enhance our understanding of understanding how Mediterranean biodi-
ranean Basin was much more complex the ways in which both colonists and versity evolved and how we might best
and multifaceted than previous prime local populations adapted management work to conserve it. The archaeobiological
mover models could accommodate. To strategies to these new environments. sciences have a valuable role to play in
varying degrees, in different areas, this There are also obvious opportunities providing greater time depth to biodiver-
process involved elements of demic diffu- for those interested in understanding the sity studies by monitoring the creation of
sion, local adoption, and independent independent domestication of European anthropogenic ecosystems and tracing the
domestication. But the outlines of this wild species. Larson et al. (23), for exam- development and impacts of both environ-
complex process are just beginning to ple, suggest that European wild boar were mentally sustainable and destructive agri-
come into focus. Maritime colonization of domesticated only after the introduction cultural economies over thousands of
the Mediterranean clearly involved not of Near Eastern domestic swine, repre- years of human occupation.
one, but multiple unrelated seaborne mi- senting a case of apparent technology
grations (52). The cultural context of transfer rather than truly independent ACKNOWLEDGMENTS. This work has beneﬁted
from the comments of Greger Larson, Bruce
these migratory movements, their causes, domestication. Local, culturally indepen- Smith, and Jean-Denis Vigne and from the art-
their routes, their timing, and their tempo dent domestication of indigenous wild istry of Marcia Bakry, who drew the maps.
1. Zeder MA, Emshwiller E, Smith BD, Bradley DG, eds 8. Zeder MA (2006) A critical examination of markers of ˜
15. Helmer D, Gourichon L, Monchot H, Peters J, Sana Segui
(2006) Documenting Domestication (Univ of California initial domestication in goats (Capra hircus). Docu- M (2005) Identifying early domestic cattle from prepot-
Press, Berkeley). menting Domestication, eds Zeder MA, Emshwiller E, tery Neolithic sites on the middle Euphrates using sex-
2. Zeder MA (2006) Archaeological approaches to docu- Smith BD, Bradley DG (Univ of California Press, Berke- ual dimorphism. The First Steps of Animal Domestica-
menting animal domestication. Documenting Domes- ley), pp 181–208. tion, eds Vigne J-D, Peters J, Helmer D (Oxbow Books,
tication, eds Zeder MA, Emshwiller E, Smith BD, Bradley 9. Zeder MA (2006) Reconciling rates of long bone fusion Oxford), pp 86 –95.
DG (Univ of California Press, Berkeley), pp 171–180. and tooth eruption and wear in sheep (Ovis) and goat 16. Martin L, Russell N, Carruthers D (2002) Animal remains
3. Uerpmann H-P (1979) Problems of the Neolithization of (Capra). Ageing and Sexing Animals from Archaeolog- from the central Anatolian Neolithic. The Neolithic of
the Mediterranean Region (original in German). Supple- ical Sites, ed Ruscillo D (Oxbow Press, Oxford), pp 87– ´
Central Anatolia, eds Gerard F, Thissen L (Ege Yayınları,
ments to the Tuebingen Atlas of the Near Orient, series B. 118. ˙
Istanbul, Turkey), pp 193–206.
(Ludwig Reichert, Weisbaden). 10. Zeder MA (2009) The Neolithic (macro)-revolution. J 17. Bruford M, Townsend SJ (2006) Mitochondrial DNA
4. Meadow RH (1989) Osteological evidence for the pro-
Archaeol Res, in press. diversity in modern sheep: Implications for domestica-
cess of animal domestication. The Walking Larder, ed
11. Redding RW (2005) Breaking the mold, a consideration tion. Documenting Domestication, eds Zeder MA, Em-
Clutton-Brock J (Unwin Hyman, London), pp 80 –90.
of variation in the evolution of animal domestication. shwiller E, Smith BD, Bradley DG (Univ California Press,
5. Bar-Yosef O, Meadow RH (1995) The origins of agricul-
The First Steps of Animal Domestication, eds Vigne J-D, Berkeley), pp 306 –316.
ture in the Near East. Last Hunters, First Farmers, eds
Peters J, Helmer D (Oxbow Books, Oxford), pp 41– 48. ´
18. Luikart G, Fernandez H, Mashkour M, England PR,
Price TD, Gebauer, A-B (School of American Research
12. Peters J, von den Driesch A, Helmer D (2005) The Taberlet P (2006) Origins and diffusion of domestic
Press, Santa Fe, NM), pp 39 –94.
upper Euphrates-Tigris Basin, cradle of agro-pasto- goats inferred from DNA markers. Documenting Do-
6. Horwitz LK (1993) The development of ovicaprine do-
mestication during the PPNB of the southern Levant. ralism? The First Steps of Animal Domestication, eds mestication, eds Zeder MA, Emshwiller E, Smith BD,
Archaeozoology of the Near East I, eds Buitenhuis H, Vigne J-D, Peters J, Helmer D (Oxbow Books, Oxford), Bradley DG (Univ of California Press, Berkeley), pp 294 –
Clason AT (Universal Book Service, Leiden, The Nether- pp 96 –124. 305.
lands), pp 27–36. 13. Vigne J-D, Buitenhuis H, Davis S (1999) The ﬁrst steps in 19. Naderi S, et al. (2007) Large-scale mitochondrial DNA
7. Vigne J-D, Peters J, Helmer D (2005) New archaeozoologi- animal domestication into the east of the Euphrates: analysis of the domestic goat reveals six haplogroups
cal approaches to trace the ﬁrst steps of animal domesti- Cypres and Central Anatolia (original in French). Pale-´ with high diversity. PLos One 10:1–23.
cation. The First Steps of Animal Domestication, eds orient 25/2:49 – 62. 20. Dobney K, Larson G (2006) Genetics and animal domes-
Vigne J-D, Peters J, Helmer D (Oxbow Books, Oxford), pp 14. Ervynck A, Dobney K, Hongo H, Meadow RH (2001) tication. J Zool 286:261–271.
Born free! Paleorient 27/2:47–73.
Zeder PNAS August 19, 2008 vol. 105 no. 33 11603
21. Bradley DG (2006) Genetics and the origins of domestic 45. Dennell R (1992) The origin of crop agriculture in Eu- 68. Pereira F, et al. (2006) Genetic signatures of a Mediter-
cattle. Documenting Domestication, eds Zeder MA, rope. The Origins of Agriculture, eds Cowan CW, ranean inﬂuence in Iberian Peninsula sheep hus-
Emshwiller E, Smith BD, Bradley DG (Univ of California Watson PJ (Smithsonian Institution Press, Washington, bandry. Mol Biol Evol 23:1420 –1426.
Press, Berkeley), pp 317–328. DC), pp 71–100. 69. Edwards C, et al. (2007) Mitochondrial DNA analysis
22. Larson G, et al. (2005) Worldwide pylogeography of 46. Lewthwaite J (1986) The transition to food production. shows a Near Eastern origin for domestic cattle and no
wild boar reveals multiple centers of pig domestica- Hunters in Transition: Mesolithic Societies in Temper- indication of domestication of European aurochs. Proc
tion. Science 307:1618 –1621. ate Eurasia and Their Transition to Farming, ed Zvelebil R Soc London Ser B 274:1377–1385.
23. Larson G, et al. (2007) Ancient DNA, pig domestication, M (Cambridge Univ Press, Cambridge, UK), pp 53– 66. 70. Cymbron T, et al. (2005) Microsatellite diversity sug-
and the spread of the Neolithic into Europe. Proc Natl 47. Guilaine J, LeBrun eds (2003) The Neolithic of Cyprus gests different histories for Mediterranean and
Acad Sci USA 104:15276 –15281. (original in French) École Français d’Athèns, Athens, Norther Eurpean cattle populations. Proc R Soc London
24. Weiss E, Kislev ME, Hartmann A (2006) Autonomous Bull Corr Hellenique Suppl 43.
´ Ser B 272:1837–1843.
cultivation before domestication. Science 312:1608 – 48. LeBrun A, Cluzan S, Davis SJM, Hansen J, Renault- 71. Beja-Pereira A, et al. (2006) The origin of European
1610. Mislkovsky (1987) The pre-ceramic Neolithic of Cyprus cattle. Proc Natl Acad Sci USA 103:8113– 8118.
25. Wilcox G, Fornite S, Herveux L (2007) Early Holocene (original in French). L’Anthropologie 91:283–316. 72. Achilli A, et al. (2008) Mitochrondrial genomes of ex-
cultivation before domestication in northern Syria. 49. Peltenburg E (2004) Introduction. Neolithic Revolution, tinct aurochs survive in domestic cattle. Curr Biol
Vegetative History Archaeobot, 10.1007/s00334 – 007- eds Peltenburg E, Wasse A (Anthony Row, Chippen- 18:157–158.
0121-y. ham, UK), pp. xi–xx. 73. Pellecchia M, et al. (2007) The mystery of Etruscan
26. Wilcox G (2002) Geographical variation in major cereal 50. Vigne J-D, Carrere I, Guilaine J (2003) Unstable status of
´ origins. Proc R Soc London Ser B 274:1175–1179.
components and evidence for independent domestica- early domestic ungulates in the Near East. The Neo- 74. Zvelebil M (2000) The social context of the agricultural tran-
tion events in Western Asia. The Dawn of Farming in lithic of Cyprus (original in French), eds Guilaine J, sition to Europe. Archaeogenetics: DNA and the Population
the Near East, eds Cappers RTJ, Bottema S (Ex Oriente, LeBrun A École Français d’Athèns, Athens, Bull Corr Prehistory of Europe, eds Renfrew C, Boyle K (McDonald
Berlin), pp 133–140, Studies in Near Eastern Production, Hellenique, Suppl 43, pp. 239 –251.
´ Institute Monographs, Oxbow, Oxford), pp 57–80.
Subsistence, and Environment no. 6. 51. Colledge S (2004) Reappraisal of the archaeobotanical 75. Blondel J, Vigne J-D (1993) Space, time, and man as
27. Wilcox G (2005) The distribution, natural habitats, and evidence for the emergence and dispersal of the determinants of diversity of birds and mammals in the
availability of wild cereals in relation to their domesti- ‘‘founder crops.’’ Neolithic Revolution, eds Peltenburg Mediterranean Region. Species Diversity in Ecological
cation in the Near East. Vegetative History Archaeobot E, Wasse A (Anthony Row, Chippenham, UK), pp 49 – Communities, eds Ricklefs, RE, Schluter D (Univ Chicago
14:534 –541. 60. Press, Chicago), pp 135–146.
28. Kislev ME, Hartmann A, Bar-Yosef O (2006) Early do- 52. Guilaine J (2003) The Neolithization in the Mediterra- 76. Vigne J-D (1999) The large ‘‘true’’ Mediterranean is-
mesticated ﬁg in the Jordan Valley. Science 312:1375– nean and France. The Widening Harvest, eds Ammer- lands as a model for the Holocene human impact on the
1374. man A, Biagi P (Archaeological Institute of America, European vertebrate fauna? The Holocene History of
the European Vertebrate Fauna, ed Benneke N (Deut-
29. Nesbitt M (2002) When and where did domesticated Boston), pp 189 –206.
sches Archaologisches Institut, Eurasien-Abteilung,
cereals ﬁrst occur in Southwest Asia? The Dawn of 53. Broodbank C (1999) Colonization and conﬁguration in
Berlin), pp 295–322.
Farming in the Near East, eds Cappers RTJ, Bottema S the insular Neolithic of the Aegean. Neolithic Society in
77. Alcover JA, Seguı B, Bover P (1999) Extinctions and local
(Ex Oriente, Berlin), pp 113–132, Studies in Near Eastern Greece, ed Halstead P (Shefﬁeld Academic, Shefﬁeld,
disappearances of vertebrates in the western Mediter-
Production, Subsistence, and Environment no. 6. UK), pp 15– 41.
ranean islands. Extinctions in Near Time, ed McPhee
30. Ammerman AJ, Cavalli-Sforza LL (1984) The Neolithic ´
54. Perles C (2001) The Early Neolithic in Greece (Cam-
RDE (Kluwer, New York), pp 165–188.
Transition and the Genetics of Populations in Europe bridge Univ Press, Cambridge, UK).
78. Cucchi T, et al. A new endemic species of the subgeneus
(Princeton Univ Press, Princeton). 55. Runnels C (2003) The origins of the Greek Neolithic. The
Mus (Rodentia Mammalia) on the island of Cyprus.
31. Higgs E, Jarman MR (1969) The origins of agriculture. Widening Harvest, eds Ammerman A, Biagi P (Archae-
Antiquity 43:31– 41. ological Institute of America, Boston), pp 121–132.
79. Simmons A (1999) Faunal Extinction in an Island Society
32. Dennell R (1983) European Economic Prehistory (Aca- 56. Biagi P (2003) A review of the late Mesolithic in Italy
(Kluwer, New York).
demic, London). and its implication for the Neolithic transition. The
80. Bunimovitz S, Barkai R (1996) Ancient bones and mod-
33. Barker G (1985) Prehistoric Farming in Europe (Cam- Widening Harvest, eds Ammerman A, Biagi P (Archae-
ern myths. J Med Arch 9:85–96.
bridge Univ Press, Cambridge, UK). ological Institute of America, Boston), pp 133–156.
81. Ammerman A, Noller J (2005) New light on Aeotokrem-
34. Hansen JM, Renfrew JM (1978) Paleolithic-Neolithic 57. Skeates R (2003) Radiocarbon dating and interpreta-
nos. World Archaeol 37:533–543.
seed remains at Francthi Cave, Greece. Nature tions of the Mesolithic-Neolithic transition in Italy. The
82. Simmons A, Mandel R (2007) Not such a new light.
271:349 –352. Widening Harvest, eds Ammerman A, Biagi P (Archae-
World Archaeol 39:475– 482.
35. Vaquer J, Geddes D, Barbaza M, Erroux J (1986) Meso- ological Institute of America, Boston), pp 157–188.
83. Reese DS, Belluomini G, Ikeya M (1996) Absolute dates
lithic plant exploitation at the Balma Abeurador 58. Binder D (2000) Mesolithic and Neolithic interaction in
for the Pleistocene fauna of Crete. Pleistocene and
(France). Oxford J Arch 5:1–18. southern France and northern Italy. Europe’s First
Holocene Fauna of Crete and its First Settlers, ed Reese
36. Ducos P (1976) Some evidence for the origin of domes- Farmers, ed Price TD (Cambridge Univ Press, Cam- DS (Prehistory Press, Madison, WI), pp 47–51.
tication in France (original in French). The Prehistory of bridge, UK), pp 117–145. 84. Broodbank C, Strasser TF (1991) Migrant farmers and the
France (original in French), ed Guilaine J (Centre Na- 59. Guilaine J, Manen C, Vigne J-D (2008) New Insights into Neolithic colonization of Crete. Antiquity 65:233–245.
tional de la Recherche Scientiﬁque, Paris), pp 165–177. the Neolithization of Mediterranean France (original 85. Sondaar PY, Martini F, Ulzega A, Klein G, Hofmeijer
37. Estevez J (1988) Study of the faunal remains (original in in French) (Centre d’Anthropologie, Toulouse, France), (1991) Pleistocene Man in Sardinia (original in French).
Spanish). Cova Fosca, ed Olaria C (Servico de Publica- in press. L’Anthropologie 95:181–200.
´ ´ ´
ciones Diputacion de Castellon, Castellon, Spain), pp ˜
60. Zilhao J (1993) The spread of agro-pastoral economies 86. Vigne J-D (2005) The ﬁrst evidence of modern humans in
281–337. across Mediterranean Europe. J Med Archaeol 6:5– 63. Corsica and Sardinia (original in French). Human Migra-
38. Boessneck J, von den Driesch A (1980) Finds of animal ˜
61. Zilhao J (2000) From the Mesolithic to the Neolithic in tion and Environmental Variation in the Quaternary
bones from four caves in southern Spain (original in the Iberian Peninsula. Europe’s First Farmers, ed Price (original in French), eds Hedges J, Hedges E (Archaopress,
German). In Studies on ﬁnds of early animal bones TD (Cambridge Univ Press, Cambridge, UK), pp 144 – Oxford), pp 139 –145, BAR International Series 1352.
from the Iberian Peninsula 7, eds Boessneck J, von den 183. 87. Vigne J-D, Desse-Berset N (1995) The exploitation of
Driesch A (Institut fur Palaeoanatomir, Domestika- ˜
62. Zilhao J (2001) Radiocarbon evidence for maritime pi- animal resources in the Mediterranean Islands during
tionsforschung und Geschichte de Tiermedizin der Uni- oneer colonization at the origins of farming in west the Preneolithic, the example of Corsica. Man and Sea
versitat Munchen, Munich), pp 1– 83. Mediterranean Europe. Proc Natl Acad Sci USA in the Mesolithic, ed Fisher A (Oxbow Books, Oxford),
39. Zohary D (1996) The mode of domestication of the 98:14180 –14185. pp 309 –318.
founder crops of Southwest Asian agriculture. The Or- 63. Guilaine J, et al. (1993) Dourgne (Centre 88. Vigne J-D (1996) Small mammal fossil assemblages as
igins and Spread of Agriculture and Pastoralism in d’Anthropologie de Societes Rurales, Archeologie en indicators of environment change in northern Corsica
Eurasia, ed Harris D (Smithsonian Institution Press, Terre d’Aude, Toulouse/Carcassonne, France). during the last 2,500 years. J Archaeol Sci 23:199 –215.
Washington DC), pp 142–158. 64. Rowley-Conwy P (2003) Early domestic animals in Eu- 89. Bover P, Alcover JA (2003) Understanding Late Quater-
40. Geddes DS (1985) Mesolithic domestic sheep in west rope. The Widening Harvest, eds Ammerman A, Biagi P nary extinctions, J Biogeog 30:771–781.
Mediterranean Europe. J Archaeol Sci 12:25– 48. (Archaeological Institute of America, Boston), pp 99 – 90. Waldren WH (1982) Balearic Prehistoric Ecology and
41. Poplin F, et al. (1986) The beginning of herding in 120. Culture (Archaeopress, Oxford), British Archaeological
France (original in French). The French Neolithic (orig- 65. Tresset A, Vigne J-D (2007) Substitution of species, Reports IS 149.
inal in French), edd Demoule J-P, Guilaine J (Picard, techniques and symbols at the Mesolithic-Neolithic 91. Ramis D, Bover P (2001) A review of the evidence for
Paris), pp 37–51. transition in western Europe. Proc Br Acad 144:189 – domestication of Myotragus balearicus Bate 1909 (Ar-
42. Guilaine J (1976) First Herders and Farmers in the 210. tiodactyla, Caprinae) in the Balearic Islands. J Archaeol
Occidental Mediterranean (original in French) (Mou- ´
66. Fernandez H, et al. (2006) Divergent mtDNA of goats in Sci 28:265–282.
ton, Paris). an Early Neolithic site, far from the initial domestica- 92. Blondel J, Aronson J (1999) Biology and Wildlife of the
43. Price TD (1987) The Mesolithic of Western Europe. J tion areas. Proc Natl Acad Sci USA 103:15375–15379. Mediterranean Region (Oxford Univ Press, Oxford).
World Prehist 1:225–305. 67. Pereira F, Pereira L, Van Asch B, Bradley DG, Amorim A 93. Zeder MA (2006) Central questions in the domestica-
44. Zvelebil M (1989) On the transition to farming in Eu- (2005) The mtDNA catalogue of all Portuguese auto- tion of plants and animals. Evol Anthropol 15:105–117.
rope, or what was spreading with the Neolithic. Antiq- chothonous goat (Capra hircus) breeds. Mol Ecol 94. Gkiastra M, Russell T, Shennan, Steele J (2003) Neolithic
uity 63:379 –383. 14:2313–2318. transition in Europe. Antiquity 77:45– 62.
11604 www.pnas.org cgi doi 10.1073 pnas.0801317105 Zeder