15
The Palaeopopulationgenetics
of Humans, Cattle and Dairying
in Neolithic Europe
Joachim Burger1
and Mark G. Thomas2,3
1
Institut f €ur Anthropologie, AG Palaeogenetik, Johannes GutenbergUniversit€at,
Germany
2
Research Department of Genetics, Evolution and Environment, University
College London, London, UK
3
Department of Evolutionary Biology, Evolutionary Biology Centre, Uppsala
University, Norbyvagen 18D, SE-752 36 Uppsala, Sweden
15.1 INTRODUCTION
The term ‘Neolithic’ most often refers to a package of features including permanent settlements,
agriculture, pottery and animal husbandry. This definition is valid for an area between
Western Anatolia and Central Europe, butapproaches the limits of its utility in many other parts
of the world. For example, pottery has been found in pre-Neolithic Africa, and signs of the
‘Neolithic package’ are found in Mesolithic cultures of the Northern Europe and the Baltic
region. The term ‘Neolithic’ is also frequently used purely chronologically and so has been
applied to parallel hunter-gatherer societies who are culturally ‘Mesolithic’, even if chronologically
they live in a Neolithic phase. Such variability in definitions and nomenclature means
that the term ‘Neolithic’ does not describe a universal and uniform process, a point that should
be taken into consideration not only in archaeology, but also in studies of population biology.
Humans first started managing goats, sheep, pigs and cattle some 11,000 to 10,000 years ago
in a regionbetween the Levant, the Zagros mountains and Central Anatolia (Zeder, 2008). After
7,000 calBC, archaeological signatures of a Neolithic package first appeared outside of the core
region (Figure 15.1), in archaeological sites in southern and western Anatolia, and seem to
have spread there from central Anatolia (O¨ zdogan and Basgelen, 1999). By around 6,400
calBC, a Neolithic package appears in the Greek Aegean (monochrome phase in Thessaly and
the Peleponnes and proto-Sesklo in other regions), and simoultaneously around the Marmara
Sea, at the north-west tip of Anatolia and in Eastern Thracia.
Human Bioarchaeology of the Transition to Agriculture Edited by Ron Pinhasi and Jay T. Stock
Ó 2011 John Wiley & Sons, Ltd.
At this point, pottery style analysis indicates two culturally distinct but geographically
overlapping trajectories for the spread of Neolithic culture in Europe: a so-called
‘Mediterranean Route’ and a ‘Danubian/Balkan Route’. The former spread via southern Italy
and the Adriatic (so-called Adriatic Impressa) to the islands of the Tuscan archipelago and later
on to the south of France and the Iberian Peninsula (the so-called Cardial Ware culture). In the
following, we focus on the so-called ‘Danubian/Balkan Route’, which is better studied from
a palaeogenetic point of view.
After an initial phase of monochrome ceramics, a multitude of archaeologically defined
cultures develop on the Balkan Peninsula and in the Carpathian Basin. These cultures can be
broadly subdivided into the Karanovo I complex and the Starcevo-K€or€os- ¸Cris complex. The
Linearbandkeramik culture (LBK) spreads rapidly from what is believed to be its Starcevo
forerunners in Central Europe from about 5,500 calBC (Pavuk, 2005). Prior to 4,100 calBC,
Neolithic culture remains confined to these limits and only later expands to the North German
lowlands and other regions of Northern Europe (Figure 15.1).
In this chapter, we compile the findings from the major palaeogenetic studies of the
European Neolithic, published in recent years. We include studies of humans as well as of
domestic animals, which can act as proxies for human population dynamics. First, we outline
the differences between modern model based approaches and classical phylogeograpic
interpretation of palaeogenetic data in population history inference. We then summarize
Figure 15.1 Chronological spread of the Neolithic. Numbers give the approximate earliest dates of the
Neolithic in years before Christ (calBC). (See Plate 15.1 for a colour version of this image)
372 Human Bioarchaeology of the Transition to Agriculture
and discuss how palaeogenetic studies have been used to examine key questions on the
demographic and evolutionary histories of Neolithic cattle and humans, including recent
studies on the spread of lactase persistence.
15.2 THE AIM OF PALAEOGENETIC STUDIES IN THE
CONTEXT OF THE NEOLITHIC TRANSITION IN EUROPE
Following the pioneering work of Luca Cavalli-Sforza in the 1970s and 1980s (Cavalli-Sforza
et al., 1986; Bowcock et al., 1987), a series of studies have been performed over the last
20 years that attempt to infer human demographic history through patterns of genetic
variability in modern populations. In the early days of research, classical genetic markers
and later mitochondrial DNA (mtDNA) were the main contributors to this field, particularly
through the interpretation of summary patterns of variation, such as Principle Components
Analysis (Cavalli-Sforza, Menozzi and Piazza, 1996) and phylogeographic (Avise, 2000)
analysis of hypervariable region I (HVR I) sequences (Richards et al., 2000).
A central tenant of phylogeographic inference is that the nodes of major clades on an inferred
phylogenetic tree, or a network, correlatewith demographic eventsin time and space. However,
this approach has been criticized (Goldstein and Chikhi, 2002; Nielsen and Beaumont, 2009,
Beaumont et al., 2010, but also see Templeton, 2009, 2010) and studies based on computer
simulations (Knowles and Maddison, 2002; Panchal and Beaumont, 2007) have highlighted
major problems with the more systematic incarnations of phylogeographic inference
(Templeton, 1998). While phylogeographic inference has dominated palaeogenetics over the
last 15 years, explicit model-based statistical inference approaches – particularly those
employing coalescent theory (Griffiths and Tavare, 1994) – have developed apace in the field
of population genetics. Coalescent theory is a retrospective model of gene genealogies for a
sample and permits the probabilities of lineages joining (when viewed backwards through
time) to be calculated under a range of different demographic conditions (e.g. population
constancy, growth, etc.). These approaches emphasize the stochastic nature of the genealogical
process in populations and recognize that very different ‘gene trees’ can arise from very similar
demographic histories, and vice-versa. In other words, the gene tree and the demographic
history of a population are, to a greater or lesser extent, decoupled and, in a sense, the gene tree
can be thought of as a nuisance parameter. Lineage sorting – the phenomenon whereby lineages
first coalesce not with lineages in the same population, but rather with lineages in a related
population, illustrates the dangers of misinterpretation if the demographic history of populations
is inferred from a phylogenetic tree/network. Similar problems arise when interpreting
summary patterns of genetic variation, such as Principal Components Analysis (CavalliSforza,
Menozzi and Piazza, 1996). For example, different gradients of Principle Components
in space have been interpreted as signatures of past migration events. Recent simulation studies
have shown that components may form predictable patterns in space, but that these do not
necessarily represent demographic episodes, if indeed they have any sensible interpretation at
all (Novembre and Stephens, 2008).
The challenge in population genetic inference is to understand, in a statistical framework,
what historical scenarios could have given rise to the data, and in relation to the geographical
location of samples. The solution to this problem is to explore (usually by simulation)
different historical scenarios and search for the conditions under which the data, or some
description of the data, has the highest probability of arising. This approach not only allows
The Palaeopopulationgenetics of Humans, Cattle and Dairying 373
parameters of an evolutionary model to be estimated but also allows formal statistical
comparison of different models of demographic history. Unfortunately this is not a trivial
task, because appropriate models need to be used that are sufficiently complex to reflect the
important processes that shape patterns of genetic variation, butare not so complex that they are
difficult to implement or when implemented, can explain any data. However, important
developments such as Approximate Bayesian Computation (Beaumont, Zhang and
Balding, 2002) make this task much easier as it allows estimation of parameters of a model,
and comparison of different models, without the need to compute exact probabilities of data
under specified models.
Because of problems of inference when using DNA data from modern individuals, as
outlined above, the most direct way to examine demographic and evolutionary history is to
obtain ancient DNA data directly from archaeological skeletal populations in the timeframe
under study. This does not mean that the outlined inferential challenges disappear altogether.
But for many of the questions archaeologists are interested in, such as population continuity vs.
admixture, the task becomes somewhat easier. Accordingly, by applying palaeogenetic
methods, archaeological hypotheses concerning the spread of people, raw materials, artefacts
and pottery styles (and thereby of ‘cultures’) can be tested. Although palaeodemography is
often taken as representing the biological analogue of culture, the extent of any correlation
between genetic and material culture data may actually be widely different for different
episodes in prehistory.
15.3 THE IMPORT AND MOVEMENT OF DOMESTIC ANIMALS
The early Neolithic of the Near East and Anatolia is characterized by the domestication of
animals including goats, sheep, pigs and cattle, at the latest by 8,500 calBC (Zeder, 2008).
The transformation of wild forms of these four species into domestic animals is, alongside
sedentism and crop (cereals and legumes) agriculture, a characteristic of early Neolithic
cultures. This transformation was most likely a long process and appears to have taken
place in the core regions of the Near East and Anatolia, long before the spread of agriculture
into western Anatolia and southeast Europe. As both the morphological and the populationgenetic
consequences of the management of wild animals are unlikely to differ significantly
from those offull domestication, uncertainty persists on where and when domestication in the
full zoological sense actually took place. Foreseeably, palaeogenetic analyses of gene
variants associated with the ‘domesticated’ phenotype will provide crucial information on
this question in the future. The information we have so far is mainly from non-coding
mitochondrial DNA, which does not tell us much about the phenotype of an animal but should
contain information on its demographic history. The best investigated species in Neolithic
times is cattle. Initial studies on the mitochondrial variability of present-day cattle in the Near
East, Anatolia, North Africa and Europe showed that the variability in the vicinity of the
postulated Neolithic domestication process is higher than that in Europe and/or North Africa
(Troy et al., 2001). The prevalence of T and T3 lineages in Europe and of T1 lineages in
North Africa is consistent with the idea of an export of early domestic animals from
the Neolithic core zone to Europe and North Africa. Nevertheless, this dataset can be
interpreted in a variety of ways (e.g. the reduced variability in Europe could be the result of
historical breeding practices). In order to further examine this question, skeletal
remains from the Neolithic period in central, north and south-east Europe were examined.
374 Human Bioarchaeology of the Transition to Agriculture
It was found that the genetic variability of today’s cattle is similar to that of early Neolithic
cattle (Bollongino et al., 2006). But a deeper understanding of these results requires
assessment of whether European domesticated cattle were introduced into Europe from the
Near East or were descended wholly or in part from wild European cattle.
Morphologically, European wild cattle (Bos primigenius), otherwise known as aurochs,
differ from domestic cattle (Bos taurus), mainly in terms of the size of skeletal elements. This
represents a methodical problem as individual male taurine cattle and individual female
aurochs have a significant overlap in size of all skeletal elements. Therefore, in order to be able
to study known European aurochs, Pleistocene or Mesolithic Holocene specimens must be
analysed. So far, all known aurochs identified carry mtDNA haplotypes that are classified
phylogenetically as P or E. These lineages are distinct from those found in Anatolian and
Middle Eastern taurine cattle (which are mostly T- and Q-types) (Scheu and Bollongino,
unpublished data). As all known wild cattle in central and northern Europe carry P-types, or to a
lesser extent E-types (Edwards et al., 2007), the T- and Q-types have been proposed to originate
from outside Europe, probably from early Neolithic domesticates in Anatolia or the Near East.
Using coalescent simulations conditioned on both ancient and modern DNA sequences, it
should be possible to estimate key parameters of the domestication history of these earliest
cattle, such as the effective size of the population at the time of domestication.
Admixture between imported domestic cattle and indigenous European aurochs can, to
a first-order approximation, be excluded in the case of central Europe (Bollongino et al., 2006;
Scheu et al., 2008; Edwards et al., 2007), but this remains a plausible scenario for other parts of
Europe (Beja-Pereira et al., 2006). In south-eastern Europe, most aurochs bones from Early
Neolithic sites are of relatively small-sized specimens and hence the differentiation between
Bos taurus and Bos primigenius on the basis of archaeozoological criteria is problematic.
With respect to pigs, extensive ancient and modern mtDNA data have been analysed and
interpreted in a phylogeographic framework to indicate multiple independent centres of
domestication. (Larson et al., 2005). We consider the latter scenario unlikely, and given that
wild boar is very common in Europe, it is more likely that progressive admixture between local
wild boar and imported (domesticated) Anatolian pigs is responsible for the pattern observed
(Larson et al., 2007). However, as with the other domestic species – sheep and goat – more
extensive coalescent modelling studies need to be performed to fully and explicitly examine
models of the spread of these animals in Europe. Thus, there is evidence that both pig and
cattle have been imported to south-east Europe and then spread over the continent within three
millennia, with pigs being mixed with local wild boar and cattle remaining maternally almost
unmixed with local aurochs. With regards to sheep and goat, there is little doubt that they were
imported from the Near East and/or Anatolia into Europe, as there are no wild progenitors of
these species in Europe.
The fact that four different species were disseminated over such a long time and moved into
such a large area previously free of agriculture, and also considering the difficulties of
breeding and transport in a new ecological niche, the question arises as to which cultural
and technical abilities were necessary for the spread of stock keeping. Certainly, the
rapid dissemination of all four species over the continent is unthinkable without favourable
social and economic conditions and it is tempting to speculate that the Mesolithic-Neolithic
transition in Europe involved some professional trade in domestic animals. Furthermore,
it is clear that the four domestic species did not migrate without humans. But the extent
to which humans moved into Europe continues to be debated (Barbujani, Bertorelle and
Chikhi, 1998; Torroni et al., 2000, Chikhi et al., 2002, Dupanloup et al., 2004; Belle, Landry
The Palaeopopulationgenetics of Humans, Cattle and Dairying 375
and Barbujani, 2006). As mass migration of Neolithic farmers seems at first glance unlikely,
the predominant view of continental archaeologists over the last few decades was that a few
pioneers were responsible for the dissemination of Neolithic agriculture (L€uning, 2007) and
that Neolithic economy and lifestyle were adopted by local Mesolithic hunter-gatherer
populations (Kind, 1998; Gronenborn, 1997). As we discuss in the following section, recent
palaeogenetic evidence challenges this traditional view, at least for Central and Northern
Europe.
15.4 THE DEMOGRAPHY OF LATE HUNTER-GATHERERS
AND EARLY FARMERS
Little is known about the palaeogenetics of late Glacial and early Holocene human populations
in Europe. The first molecular genetic study of early LBK skeletons from Central Europe – all
dating to between 5,500 to 5,000 calBC – did little to reveal the origin of Europe’s first
agriculturists, but did produce a surprising result. Although the LBK was the pioneer Neolithic
farming culture of central Europe, their mitochondrial lineages appear to have been severely
diluted over the subsequent millennia and some of them are rare in modern Europeans (Haak
et al., 2005). Two main explanations have been offered to account for this observation. The
first, which is consistent with some archaeological theories (L€uning, 2007), is that the early
LBK farmers were small groups of pioneers who were later replaced by surrounding huntergatherers.
The second scenario is that these pioneer LBK farmers were to a large extent
replaced by subsequent waves of farmers carrying different mtDNA lineages. However,
currently neither of these scenarios receivessupport either from archaeology or palaeogenetics.
In a more recent study and using coalescent simulations, Bramanti and colleagues (2009)
further confirmed the observation that the LBK farmers were not the sole direct ancestors of
modern Central Europeans, but to date, the missing continuity between the early farmers
of central Europe and modern Europeans remains to be explained. It is likely that a
combination of a series of demographic processes, which post-date the initial colonization
of central Europe by the pioneer LBK farmers, is responsible for this lack of genetic continuity.
Future analysis of palaeogenetic data from later periods, including spatially explicit serial
modelling of heterochronous aDNA data, will allow us to assess the effects of these
demographic processes in more detail.
As part of the same study, Bramanti and colleagues (2009) also examined mitochondrial
DNA from skeletons of European hunter-gatherers from Upper Palaeolithic, Mesolithic, and
Neolithic periods. The individuals examined came from different locations in southwest
Germany, northern Germany and the Baltic region. Specimens from various archaeological
periods were considered as a single group of ‘hunter-gatherers’ because of their similar
subsistence strategies, even though it is not certain if they belong to any single population in
a narrower biological sense. As a group, they represent the descendents of the hunter-gatherers
that re-colonized central and northern Europe from southern European refugia after the end of
the last Glacial maximum, around 20,000 BP. The following major climatic/environmental
changes during deglaciation, and corresponding re-colonization of the biota in northern
Europe, meant that large parts of this region were also re-occupied by human populations.
These hunter-gatherers are probably the descendents of the first anatomically modern humans
to settle in Europe around 45,000 years ago. The authors have shown that, in genetic terms, the
LBK sample and the hunter-gatherer sample were indeed significantly different, as expressed
376 Human Bioarchaeology of the Transition to Agriculture
by a comparatively high FST value (0.163). FST is a measure of genetic difference between
populations and quantifies the proportion of genetic variation in a set of population samples that
is attributable to differences between those samples. For comparison, the observed FST (0.163)
between the two prehistoric groups was higher than the highest values found between similarly
sized samples from a set of global populations (FST ¼ 0.133). Serial coalescent simulations –
that is coalescent simulations where a simulated lineage can be sampled at different points in
time – were carried out in order to determine whether or not the difference between the
LBK sample and the hunter-gatherer sample, as measured by FST, could be explained under
the null-hypothesis of population continuity. Because genetic differences between populations
in space and time are determined by a variety of factors, including ancestral population sizes, it
was necessary to perform the simulations under a wide range of assumed combinations of
population size at the start of the Neolithic and the Upper Palaeolithic colonization of Europe
some 45,000 years ago. These simulations revealed no continuity scenarios in which
a significant proportion of the simulated FST values were the same or higher than those
actually observed. This allowed direct continuity between the hunter-gatherer sample and the
LBK sample – the null-hypothesis – to be rejected. In another palaeogenetic study on the
genetic contribution of late hunter-gatherers to modern Europeans, Malmstr€om and colleagues
(Malmstr€om et al., 2009) showed that the peoples of the last major hunter-gatherer complex in
Europe, the Pitted Ware culture of southern Scandinavia, were unlikely to be the direct
ancestors of modern Swedes, Norwegians or Saami. They did, however, show that direct
continuity between the Pitted Ware peoples and those of the eastern Baltic region was possible.
This does not mean that they actually were the ancestors of modern Eastern Baltic populations,
as it can also – and more plausibly – be explained by common ancestry between the sampled
populations. It is intriguing to note that the predominant lineage found in both the central
European and the Scandinavian hunter-gatherers, haplogroup U (Bramanti et al., 2009;
Malmstr€om et al., 2009), is the same as that observed in a single alleged Upper Palaeolithic
Russian burial from the site of Kostenkii 14 (Markina Gora), which is indirectly dated to
approximately 30,000 years ago (Krause et al., 2010), although in our opinion it is hard to
exclude contamination in such a study (P€a€abo et al., 2004).
The study by Bramanti and colleagues (2009) was not the first genetic study to address the
question of hunter-gatherer contribution to early European farming populations or to modern
Europeans (Barbujani, Sokal and Oden, 1995; Barbujani, Bertorelle and Chikhi, 1998;
Chikhi et al., 1998; Richards et al., 2000; Semino et al., 2000; Chikhi et al., 2002; Dupanloup
et al., 2004; Currat and Excoffier, 2005). However, these previous studies were all based on
modern genetic data – where inference requires many explicit, and in some cases implicit
assumptions. Also, most were aimed at obtaining estimates of admixture between huntergatherers
and incoming farmers using questionable proxies for ancestral source populations
(specific lineages in the case of those studies based on phylogeographic inference, modern
population samples in the case of those studies that used explicit modelling). Therefore,
given the uncertainties of the inference process (see above), the null-hypothesis of population
continuity remained to be formally rejected. Because Bramanti and colleagues (2009)
sampled directly from ancient populations living around the time of the transition from
hunter-gathering to farming, they – for the first time – were able to formally reject
continuity. It is likely that modern Europeans are a product of admixture between these
two groups – Central European hunter-gatherers and LBK farmers – and probably other
ancestral source populations. But those other ancestral source populations remain to be
identified and represented with ancient DNA data. Without such data, any estimates of
The Palaeopopulationgenetics of Humans, Cattle and Dairying 377
admixture proportions in early farmers, late Neolithic populations and any subsequent
European peoples up to the present day, will remain unreliable.
Since the first LBK farmers of central Europewere clearly not the direct descendents of local
hunter-gatherers, they must have migrated in from another region. There are, as yet, no
palaeogenetic data to indicate the most probable region of origin of the early LBK farmers.
From an archaeological perspective, the most plausible region of origin is around the area of
Lake Balaton in present-day Hungary (Pavuk, 2005; Figure 15.2), since this is the region (along
with south-west Slovakia) where the LBK first developed around 5,700 calBC from the
predecessor Starcevo culture. From 5,500 calBC until the end of its initial expansion phase
around 5,300 calBC, the LBK culture spread rapidly across central Europe, with settlements
quickly reaching from the Rhine valley to eastern Poland. By around 5,000 calBC, the LBK
culture had expanded further, stretching from the Paris Basin through to the Ukraine. Since the
LBK culture spread from a clearly restricted region, and because at least the early LBK farmers
investigated were such highly mobile immigrants, sampling ancient skeletal material from
the Lake Balaton core region will be essential to better understand the genetic origins of
Central Europe’s first farmers. Unfortunately, there is currently an almost complete absence
of skeletons from the oldest phase of the LBK, that is between 5,600 and 5,400 calBC
(L€uning, 2005).
If the scenario of a migration from the north-western part of the Balkans westwards to central
Europe during the Neolithic could be examined using ancient DNA data, then the question that
immediately arises is whether it was a single event (perhaps of a mass migration) or only one of
Figure 15.2 Earliest known LBK sites (5,700–5,500 calBC; white squares) north of Lake Balaton after
Pavuk (2005) and the inferred geographical origin of selection of lactase persistence, after Itan et al., 2009
(about 4,310–6,730 calBC, concentric ellipses). (See Plate 15.2 for a colour version of this image)
378 Human Bioarchaeology of the Transition to Agriculture
a series of Neolithic expansions. It is possible that populations of the Starcevo-K€or€os-Cris
complex were themselves the direct descendents of the earliest farmers to enter Europe,
presumably from Western Anatolia or even from the western foothills of the secondary
Neolithic core zone in Central Anatolia. If we consider the new picture that is now emerging
for Central Europe on the basis of palaeogenetic data from domestic animals and humans,
then an ultimate Anatolian (or even Near-Eastern) origin no longer seems as implausible as it
once was.
We also need to consider the vexing question of hunter-farmer contacts in Europe. It is.
however, necessary to point out that cultural and economic contacts may or may not lead to
biological admixture. There is clear archaeological evidence of cultural exchange between
hunter-gatherers and central Europe’s first farmers, as witnessed, for example, in the
Mesolithic silex artefacts found in LBK settlements (Gronenborn, 1999). Because the
exchange of underlying silex technologies (e.g. sourcing, tool production methods) mainly
involved hunting weapons, it is not unreasonable to assume that contact was primarily
mediated by men. So far, there has been no palaeogenetic evidence of any admixture at all, if
only because ancient Y-chromosome data has yet to be generated. The analysis of ancient Ychromosome
data is more challenging than for mtDNA, because of the reduced survival of
nuclear DNA over time, but recent developments in sequencing technologies (Mardis, 2008)
do offer a solution to this problem. However, where data does exist, there appears to be little if
any exchange of mitochondria in the early centuries following the Neolithic transition in
central Europe. Perhaps the most plausible scenario is that two distinct societies existed during
this early Neolithic period, which interacted with each other culturally, but maintained female
marriage restrictions and fundamentally differed in their economic and cultural attributes.
There are numerous present-day parallels of this kind, where ethnic groups live side by side but
have different economic systems. To give one example, in Mali the semi-nomadic Fulbe are
cattle breeders and pastoralists, the Dogon are crop farmers but also keep some goats and
sheep, and the Bozos are fishers of the Niger river. These three linguistically separated groups
exchange the products of their work, for example the Fulbe exchange milk for millet with
the Dogon but keep a strict marriage ban (Huysecom, personal communication). Although the
precise nature of co-existence in Neolithic Europe will probably remain unknown, to infer the
progressive degree of admixture between the two groups during the fourth and third millennia
BC is an upcoming task of palaeogenetic research. Besides purely demographic phenomena
and culturally defined barriers, selection of certain genetic traits is likely to have played an
important role in the central European Neolithic transition. One such example is the rise of
lactase persistence in Europe, a process on which we focus in the following section.
15.5 THE ROLE OF, AND ADAPTATIONS TO DAIRYING
Lactose is the main carbohydrate in milk and is a major energy source for most young
mammals. The enzyme responsible for hydrolysis of lactose into glucose and galactose is
lactase. Without this enzyme, mammals are unable to break down and thus utilize lactose. After
the weaning period is over, lactase production usually declines. However, some humans
continue to produce lactase throughout adult life, and are thus able to digest the lactose found in
fresh milk; a trait which is called lactase persistence (LP). Genetic studies on modern
The Palaeopopulationgenetics of Humans, Cattle and Dairying 379
populations have shown that a common variant of the lactase gene (LCT) known as
À13 910Ã
T, is strongly associated with, and probably causative of LP in Europe (Enattah
et al., 2002) and appears to be one of the most strongly selected alleles in the human
genome in Europeans in the last 30,000 years (Tishkoff et al., 2007; Bersaglieri et al., 2004;
Coelho et al., 2005). In Europe, clines are seen with LP found at frequencies of 10 to 30% in
the south-eastern part of the continent, 50 to 60% in Central Europe, rising to 70 to 95% in
north-western continental Europe and the British Isles. A similar trend is seen for the
À13,910Ã
T allele (Ingram et al., 2009). Palaeogenetic typing of this variant site in 9 Mesoand
Neolithic skeletons revealed an absence of the À13,910Ã
T allele in all cases (Burger
et al., 2007). Although this is a relatively small sample, statistical analysis has demonstrated
significantly lower frequencies of LP when compared to modern Europeans from the same
region. This is consistent with previous studies, inferring high selection coefficients acting
on this trait. It is unlikely that natural selection would have driven the À13,910Ã
T allele to
high frequencies without a supply of fresh milk and this ties the biological evolution of LP
to the culture of dairying through a gene-culture co-evolution process. Clear evidence of
milk production can be seen in South-east Europe as early as 6,200 calBC, using lipid
analysis on pot sherds (Evershed et al., 2008). In order to better understand the co-evolution
of LP and dairying in Europe, Itan and colleagues (Itan et al., 2009) developed a demic
computer simulation model in order to examine how demographic and evolutionary
parameters could have shaped both the modern distribution of LP in Europe and the
timing of the arrival of farming at different locations throughout Europe. The study
modelled the spread of dairying and non-dairying farmers into a Europe that was previously
occupied by hunter-gatherers, under the plausible assumption that an LP-associated allele
would only be selected in dairying farmers. Values for a number of different parameters
must have shaped this process, including:
1. the extent of sporadic unidirectional migration;
2. the extent of gene flow between different cultural groups;
3. the extent of gene flow between neighbouring demes;
4. the extent to which people take up the culture of their neighbours;
5. the strength of selection favouring LP; and
6. the origin time and location of LP-dairying co-evolution.
By choosing random values for these parameters (within reasonable ranges), performing the
simulation, and then comparing outcomes to observed data using Approximate Bayesian
Computation (Beaumont, Zhang and Balding, 2002), the authors were able to identify
parameter values that best explained the modern distribution of LP in Europe and the timing
of the arrival of farming at different locations throughout Europe. Although the LP allele is
most frequent in Northern Europe today, the simulations that best explained the observed data
(on the distribution of LP and the arrival time of farming at different locations) required LPdairying
co-evolution to start in an area between the Carpathian Basin and Central Europe
(Figure 15.2) between 6,260 and 8,680 years ago. The LP selection coefficient had inferred
values between 0.0518 and 0.159 (in dairying farmers only). The inferred location and dates
for the co-evolution of LP and dairying correspond well with the origins of the LBK culture
in the Lake Balaton region (see Figure 15.2). This again is in accordance with Bramanti
et al. (2009) who inferred a massive immigration of the LBK farmers from this region.
380 Human Bioarchaeology of the Transition to Agriculture
15.6 CONCLUSIONS
By integrating evidence from ancient and modern DNA, archaeozoology and archaeology,
we now envisage the following plausible scenario: whereas in the early Neolithic there
are already archaeometric indications of milk usage in West Anatolia (Evershed
et al., 2008) and the Balkans (Craig et al., 2005), these early farmers (who imported
their cattle ultimately from the Neolithic core zone in South-east Anatolia and the Northern
Levant), were not able to consume significant quantities of fresh milk. Instead, they would
have extracted the nutritional benefits of dairying by processing milk to make yoghurt,
cheese and other products. In this process much or most of the lactose in the milk is
converted to fatty acids, rendering the product consumable by lactase non-persistent
individuals.
Starting in early Neolithic south-eastern Central Europe, LBK people began to settle in
Central Europe from 5,500 calBC, without mixing significantly with the local hunters and
gatherers. Although there are attested cultural contacts between the two groups, exchange
of females is likely to have been limited, at least in the early stages. An initially rare
–13,910Ã
Tallele began to rise in frequency as fresh milk became more readily available and
the ability to drink it became more advantageous. We are not yet sure why LP provided
such a big advantage at the time. The fact that it takes longer to ferment milk in cold
climates may be a factor, but some buffering of the food supply, when compared to the
boom-and-bust of seasonal crops, is most likely to have been involved, especially amongst
pioneering farmers in uncharted territories. Significantly, the reduced mortality of postweaning
period lactase persistent children would have had a major demographic effect over
the centuries. Already by the Middle Neolithic this co-evolutionary process had resulted in
a specialized dairying economy something like we see in sites of the R€ossen culture
(Benecke, 1994). From 4,100 calBC, the Neolithic spread across the central low mountain regions
to the North German lowlands and Northern Europe, would have been aided by the constancy of
milk supply. Accordingly, a further increase in the frequency of the LP occurred, driven in part by
selection and in part by the process of allele surfing (Edmonds, Lillie and Cavalli-Sforza, 2004;
Klopfstein,CurratandExcoffier,2006).Followingfurtherdemographicexpansion,thefarmersthat
reached the north-western reaches of Europe were now predominately LP dairyers. We still don’t
know to which extent the regionally varying admixture with local hunter-gatherers influenced the
LP frequency in this region.
Such scenarios are easy to envisage and eminently plausible. But to develop any real
confidence in them, it is necessary to show by quantitative simulation modelling that the
relevant processes fit well with ancient and modern DNA data, and also with the archaeological
data. Only by this route, we believe, can the different data sources satisfactorily be integrated
into a true understanding of the human past.
ACKNOWLEDGEMENT
We thank Jens L€uning, Norbert Benecke, Pascale Gerbault, Adam Powell and Yuval Itan for
useful comments, Bernhard Weninger for critical reading of the manuscript. MTwas supported
by the AHRC Centre for the Evolution of Cultural Diversity.
The Palaeopopulationgenetics of Humans, Cattle and Dairying 381
REFERENCES
Avise, J.C. (2000) Phylogeography. The History and Formation of Species, Harvard University Press,
Cambridge, London.
Barbujani, G., Bertorelle, G. and Chikhi, L. (1998) Evidence for Paleolithic and Neolithic gene flow in
Europe. Am. J. Hum. Genet., 62, 488–492.
Barbujani, G., Sokal, R.R. and Oden, N.L. (1995) Indo-European origins: a computer-simulation test of
five hypotheses. Am. J. Phys. Anthropol., 96, 109–132.
Beaumont, M.A., Nielsen, R., Robert, C. et al. (2010) In defence of model-based inference in
phylogeography. Mol. Ecol., 19, 436–446.
Beaumont, M.A., Zhang, W. and Balding, D.J. (2002) Approximate Bayesian computation in population
genetics. Genetics, 162, 2025–2035.
Beja-Pereira, A., Caramelli, D., Lalueza-Fox, C. et al. (2006) The origin of European cattle: evidence
from modern and ancient DNA. Proc. Natl. Acad. Sci. U. S. A., 103, 8113–8118.
Belle, E.M., Landry, P.A. and Barbujani, G. (2006) Origins and evolution of the Europeans’ genome:
evidence from multiple microsatellite loci. Proc. Biol. Sci., 273, 1595–1602.
Benecke, N. (1994) Der Mensch und seine Haustiere. Die Geschichte Einer Jahrtausendealten
Beziehung, Konrad Theiss Verlag, Stuttgart.
Bersaglieri, T., Sabeti, P.C., Patterson, N. et al. (2004) Genetic signatures of strong recent positive
selection at the lactase gene. Am. J. Hum. Genet., 74, 1111–1120.
Bollongino, R., Edwards, C.J., Alt, K.W. et al. (2006) Early history of European domestic cattle as
revealed by ancient DNA. Biol. Lett., 2, 155–159.
Bowcock, A.M., Bucci, C., Hebert, J.M. et al. (1987) Study of 47 DNA markers in five populations from
four continents. Gene Geogr., 1, 47–64.
Bramanti, B., Thomas, M.G., Haak, W. et al. (2009) Genetic discontinuity between local hunter-gatherers
and central Europe’s first farmers. Science, 326, 137–140.
Burger, J., Kirchner, M., Bramanti, B. et al. (2007) Absence of the lactase-persistence-associated allele in
early Neolithic Europeans. Proc. Natl. Acad. Sci. U. S. A., 104, 3736–3741.
Cavalli-Sforza, L.L., Kidd, J.R., Kidd, K.K. et al. (1986) DNA markers and genetic variation in the human
species. Cold Spring Harb. Symp. Quant. Biol., 51 (Pt 1), 411–417.
Cavalli-Sforza, L.L., Menozzi, P. and Piazza, A. (1996) The History and Geography of Human Genes,
Princeton University Press, Princeton.
Chikhi, L., Destro-Bisol, G., Bertorelle, G. et al. (1998) Clines of nuclear DNA markers suggest a largely
Neolithic ancestry of the European gene pool. Proc. Natl. Acad. Sci. U. S. A., 95, 9053–9058.
Chikhi, L., Nichols, R.A., Barbujani, G. and Beaumont, M.A. (2002) Y genetic data support the Neolithic
demic diffusion model. Proc. Natl. Acad. Sci. U. S. A., 99, 11008–11013.
Coelho, M., Luiselli, D., Bertorelle, G. et al. (2005) Microsatellite variation and evolution of human
lactase persistence. Hum. Genet., 117, 329–339.
Craig, O.E., Chapman, J., Heron, C. et al. (2005) Did the first farmers of central and eastern Europe
produce dairy foods? Antiquity, 79, 882–894.
Currat, M.and Excoffier,L. (2005) The effect of the Neolithic expansion on Europeanmolecular diversity.
Proc. Biol. Sci., 272, 679–688.
Dupanloup, I., Bertorelle, G., Chikhi, L. and Barbujani, G. (2004) Estimating the impact of prehistoric
admixture on the genome of Europeans. Mol. Biol. Evol., 21, 1361–1372.
Edmonds, C.A., Lillie, A.S. and Cavalli-Sforza, L.L. (2004) Mutations arising in the wave front of an
expanding population. Proc. Natl. Acad. Sci. U.S.A., 101, 975–979.
Edwards, C.J., Bollongino, R., Scheu, A. et al. (2007) Mitochondrial DNA analysis shows a Near Eastern
Neolithic origin for domestic cattle and no indication of domestication of European aurochs. Proc.
Biol. Sci., 274, 1377–1385.
382 Human Bioarchaeology of the Transition to Agriculture
Enattah, N.S., Sahi, T., Savilahti, E. et al. (2002) Identification of a variant associated with adult-type
hypolactasia. Nat. Genet., 30, 233–237.
Evershed, R.P., Payne, S., Sherratt, A.G. et al. (2008) Earliest date for milk use in the Near East and
southeastern Europe linked to cattle herding. Nature, 455, 528–531.
Goldstein, D.B. and Chikhi, L. (2002) Human migrations and population structure: what we know and
why it matters. Annu. Rev. Genomics Hum. Genet., 3, 129–152.
Griffiths, R.C. and Tavare, S. (1994) Sampling theory for neutral alleles in a varying environment. Philos.
Trans. R. Soc. Lond. B Biol. Sci., 344, 403–410.
Gronenborn, D. (1999) Avariation on a basic theme: the transition to farming in southern central Europe.
J. World Prehistory, 13, 123–210.
Gronenborn, D. (1997) Silexartefakte der €altestbandkeramischen Kultur. Mit einem Beitrag von JeanPaul
Caspar, Universit€atsforschungen zur pr€ahistorischen Arch€aologie, 37, Habelt Bonn.
Haak, W., Forster, P., Bramanti, B. et al. (2005) Ancient DNA from the first European farmers in 7500year-old
Neolithic sites. Science, 310, 1016–1018.
Ingram, C.J., Mulcare, C.A., Itan, Y. et al. (2009) Lactose digestion and the evolutionary genetics of
lactase persistence. Hum. Genet., 124, 579–591.
Itan, Y., Powell, A., Beaumont, M.A. et al. (2009) The origins of lactase persistence in Europe. PLoS
Comput. Biol., 5, e1000491.
Kind, C.-J. (1998) Komplexe Wildbeuter und fr€uhe Ackerbauern. Bemerkungen zur Ausbreitung der
Linearbandkeramik im s€udlichen Mitteleuropa. Germania, 76/1, 1–24.
Klopfstein, S., Currat, M. and Excoffier, L. (2006) The fate of mutations surfing on the wave of a range
expansion. Mol. Biol. Evol., 23, 482–490.
Knowles, L.L. and Maddison, W.P. (2002) Statistical phylogeography. Mol. Ecol., 11, 2623–2635.
Krause, J., Briggs, A.W., Kircher, M. et al. (2010) A complete mtDNA genome of an early modern human
from Kostenki. Russia. Curr. Biol., 20(3), 231–236.
Larson, G., Dobney, K., Albarella, U. et al. (2005) Worldwide phylogeography of wild boar reveals
multiple centers of pig domestication. Science, 307, 1618–1621.
Larson, G., Albarella, U., Dobney, K. et al. (2007) Ancient DNA, pig domestication, and the spread of the
Neolithic into Europe. Proc. Natl. Acad. Sci. U. S. A., 104, 15276–15281.
L€uning, J. (2005) Bandkeramische Hofpl€atze und absolute Chronologie der Bandkeramik. (eds J. L€uning,
C. Friedrich and A. Zimmermann), Die Bandkeramik im 21. Jahrhundert: Symposium in der Abteil
Brauweiler bei K€oln, pp. 49–74.
L€uning, J. (2007) Bandkeramiker und Vor-Bandkeramiker – Die Entstehung des Neolithikums in
Mitteleuropa, in Vor 12 000 Jahren in Anatolien – Die €altesten Monumente der Menschheit (ed.
Badisches Landesmuseum Karlsruhe), Konrad Theiss Verlag, Stuttgart, pp. 177–189.
Malmstr€om, H., Gilbert, M.T., Thomas, M.G. et al. (2009) Ancient DNA reveals lack of
continuity between Neolithic hunter-gatherers and contemporary Scandinavians. Curr. Biol.,
19, 1758–1762.
Mardis, E.R. (2008) The impact of next-generation sequencing technology on genetics. Trends Genet.,
24, 133–145.
Nielsen, R. and Beaumont, M.A. (2009) Statistical inferences in phylogeography. Mol. Ecol., 18,
1034–1047.
Novembre, J. and Stephens, M. (2008) Interpreting principal component analyses of spatial population
genetic variation. Nat. Genet., 40, 646–649.
O¨ zdogan, M. and Basgelen, N. (1999) Neolithic in Turkey. The Cradle of Civilization, Ancient Anatolians
Civilizations Series 3, Istanbul.
P€a€abo, S., Poinar, H., Serre, D. et al. (2004) Genetic analyses from ancient DNA. Annu. Rev. Genet., 38,
645–679.
The Palaeopopulationgenetics of Humans, Cattle and Dairying 383
Panchal, M. and Beaumont, M.A. (2007) The automation and evaluation of nested clade phylogeographic
analysis. Evolution, 61, 1466–1480.
Pavuk, J. (2005) Typologische Geschichte der Linearbandkeramik. (eds J. L€uning, C. Friedrich, A.
Zimmermann), Die Bandkeramik im 21. Jahrhundert: Symposium in der Abteil Brauweiler bei K€oln,
pp. 17–39.
Richards, M., Macaulay, V., Hickey, E. et al. (2000) Tracing European founder lineages in the Near
Eastern mtDNA pool. Am. J. Hum. Genet., 67, 1251–1276.
Scheu, A., Hartz, S., Schm€olcke, U. et al. (2008) Ancient DNA provides no evidence for independent
domestication of cattle in Mesolithic Rosenhof, Northern Germany. J. Arch. Sci., 35, 1257–1264.
Semino, O., Passarino, G., Quintana-Murci, L. et al. (2000) MtDNA and Y-chromosome polymorphisms
in Hungary: inferences from the Palaeolithic, Neolithic and Uralic influences on the modern Hungarian
gene pool. Eur. J. Hum. Genet., 8, 339–346.
Templeton, A.R. (2009) Why does a method that fails continue to be used: the answer. Evolution, 63,
807–812.
Templeton, A.R. (2010) Coalescent-based, maximum likelihood inference in phylogeography. Mol. Ecol.
Early View. doi:10.1111/j.1365-294X.2009.04514.x.
Templeton, A.R. (1998) Nested clade analyses of phylogeographic data: testing hypotheses about gene
flow and population history. Mol. Ecol., 7, 381–397.
Tishkoff, S.A., Reed, F.A., Ranciaro, A. et al. (2007) Convergent adaptation of human lactase persistence
in Africa and Europe. Nat. Genet., 39, 31–40.
Torroni, A., Richards, M., Macaulay, V. et al. (2000) mtDNA Haplogroups and Frequency Patterns in
Europe. Am. J. Hum. Genet., 66, 1173–1177.
Troy, C.S., MacHugh, D.E., Bailey, J.F. et al. (2001) Genetic evidence for Near-Eastern origins of
European cattle. Nature, 410, 1088–1091.
Zeder, M. (2008) Domestication and early agriculture in the Mediterranean basin: origins, diffusion, and
impact. Proc. Natl. Acad. Sci. USA, 105, 11597–11604.
384 Human Bioarchaeology of the Transition to Agriculture