MANUSCRIT ACCEPTAT

MANUSCRIT ACCEPTAT
Livestock Trade during the Early Roman Period:
First Clues from the Trading Post of Empúries
(Catalonia)

L. Colominas ; C. J. Edwards

Revista

International Journal of Osteoarchaeology. Volume 27, Issue 2, March/April
2017, pages 167-179

DOI

http://dx.doi.org/10.1002/oa.2527

Disponible
27/03/2016
en línia

Data de publicació

12/04/2017

Per citar aquest document:
Colominas, L., and Edwards, C. J. (2017) Livestock Trade during the Early Roman Period:
First Clues from the Trading Post of Empúries (Catalonia). Int. J. Osteoarchaeol., 27: 167–
179. doi: 10.1002/oa.2527
Aquest arxiu PDF conté el manuscrit acceptat per a la seva publicació.

1

MANUSCRIT ACCEPTAT

Abstract
Written sources show that livestock were traded during the Roman period. However,
there is scarce information available to characterise this trade because of its invisibility in
the archaeological record. In our paper, we shed light on this issue by applying both
osteometric and genetic analyses on cattle remains from the Roman trading post of
Empúries (Catalonia) to determine how livestock contributed to Roman trade and, thus, to
the economy of the Empire. Analysis of 26 cattle metacarpals from Empúries has allowed
us to document the presence of different cattle morphotypes in this city during its Early
Roman occupation. The morphological and genetic differences seen in Empúries cattle can
be explained through trade of different cattle varieties, more appropriate for milk
production and/or traction than the local stock. Once arrived at the port of Empúries,
these imported cattle would have then been distributed to surrounding villas.

2

MANUSCRIT ACCEPTAT

1. Introduction
At its height, the Roman Empire dominated a vast area of Europe, southwest Asia
and northern Africa. Part of its success was its ability to connect efficiently the
various territories under its control because of the construction and maintenance
of large communication networks, which consisted of integrated systems
combining sea, river and land transport. In the Roman period, maritime trade was
the most lucrative and cheapest method of transport, involving long voyages over
open water, as well as coast-to-coast journeys (Tchernia, 2011; de Soto, 2013). It
was possible to move huge loads in a short time, from the straits of Gibraltar, as far
away as southwest Asia, across the entire Mediterranean basin (Arnaud, 2005).
Riverine trade was the second most economic mode of transport, with the main
rivers and many of their tributaries being the leading form of access to deep inland
areas (Parodi Álvarez, 2001; Adams, 2012). Once trade products had arrived to a
maritime and/or fluvial port, terrestrial networks of roads allowed connection
with the interior nucleus of a population. Land transport was probably the system
that was most used on a daily basis, and was essential to maintain permanent
communications between all the territories of the Empire (de Soto, 2013).
The products traded were diverse. Written sources show that merchants traded
both luxury and/or manufactured products [precious stones, beads, fabrics such as
silk and associated dyes, ivory and wooden objects, perfumes, oils and fragrances,
wine and slaves], as well as indispensable items [cereals, livestock and horses]
(Apocalypse, XVIII, 12–13 [14] and 23 [14], referenced in Tchernia, 2011).
Archaeology has investigated this trade mainly through the study of shipwrecks
(Rauh, 2003) and the non-perishable materials recovered inside them, such as
ceramics, stones or metals, which has provided information about trade routes and
the distribution of traded materials (Arnaud, 2005). However, scarce information
is available about the trade of perishable materials, such as food, or living beings,
such as slaves or livestock.
The information about livestock trade in the Roman period mainly comes from
written sources; for example, in his book De re Rustica, Lucius Junius Moderatus
Columella (AD 4 – c.70) recommends the importation of cattle from the Italian
region of Altinum for milk production (VI, 24 [5]), while also suggesting that, if the
agronomist wants to buy cattle from distant lands, he must visit to ensure that
these sites have similar natural conditions (VI, 2 [13]). Columella explains how his
paternal uncle, Marcus Columella, bought several African wild sheep at Cadiz in
order to breed them with local sheep, thereby achieving wool of a different colour
(VII, 2 [4]). However, although we know that livestock trade existed during the
Roman period, we do not have any archaeological information that allows us to
characterise this trade.
Taking into account these considerations, the aim of our paper was to explore
Roman livestock trade through the application of osteometric and genetic analyses
on bone remains. Osteometry is used to characterise animal morphology. This
information is valuable to the study of animal manipulation and improvement
(Albarella, 1997; Tekkouk & Guintard, 2007; Klein et al., 2010), and data can
inform on animal mobility (Lauwerier, 1988; Murphy et al., 2000; Vigne et al.,
3

MANUSCRIT ACCEPTAT

2009). Ancient DNA analyses allow genetic characterisation of contemporaneous
populations, providing information on issues such as migration (for example,
Edwards et al., 2003; Bollongino et al., 2006; Tresset et al., 2009; Colominas et al.,
2015). Thus, the combined use of osteometry and genetics has great potential to
give us a better understanding of the trade in animals in the past.
To carry out our investigation, we selected the Roman site of Empúries, a trading
post in the northeast of the Iberian Peninsula. The commercial activity of Empúries
in the Roman period is attested through both archaeology, with the presence of
port installations (Nieto et al., 2005; Vivar, 2012) and imported objects
(Tremoleda, 2012), and from written sources. Titus Livius emphasises the role of
the commercial port at Empúries in facilitating the trade of foreign goods inland
and the exportation of agricultural products from the surrounding areas (Ad urbe
condita XXXIV, 9). Empúries was a well-connected city where trade products from
inland arrived to the port via land transport routes, and foreign products arrived
by maritime routes. With this paper, our aim was to further analyse if livestock
contributed to this trade.

Materials and methods
The sites

In this paper, although our focus is on the early Roman occupation of Empúries
(first century BC to third century AD), we also consider zooarchaeological data
from the second century BC at this site. In addition, we compare the Empúries data
with third century BC, and first century BC to third century AD, sites located near
Empúries (Figure 1).

Figure 1.


Open in figure viewer

Location of Empúries, late Iron Age and early Roman sites cited in the text, plus a plan of the Roman
city (Museu d'Arqueologia de Catalunya-Empúries). [Colour figure can be viewed at
wileyonlinelibrary.com]

4

MANUSCRIT ACCEPTAT
The site of Empúries

The city of Empúries was established by the Greeks in the northeast of the Iberian
Peninsula, now modern-day Catalonia, as a western Mediterranean trading post in
the sixth century BC (Figure 1; Aquilué et al., 1999). With the second Punic war
(218 – 201 BC), the Romans arrived in Empúries and began the conquest of the
Peninsula, and the site became a key element in the strategy of military,
administrative and economic control of the region (Santos, 2008). In the early first
century BC, the Romans built a new city (Figure 1), which, during the Augustan
period (44 BC – AD 14), assimilated the old Greek city and cemented its important
commercial role in the region (Aquilué et al., 1999). During the Roman period,
monumental and administrative buildings were constructed in the centre of the
city, emphasising its economic and political importance to the Empire (Tremoleda,
2008). Although its commercial vitality had already started to decline at the end of
the third century AD, the city was occupied into the medieval period (Tremoleda,
2008).
Zooarchaeological data show that pigs were the most abundant animals during the
Roman occupation (more than 50% of the sample), followed by cattle and
sheep/goat. Kill-off patterns reveal that pigs were killed at juvenile and subadult
ages, while cattle and sheep/goat were mainly slaughtered when adult (Buxó et al.,
2007).
The late Iron Age sites

Zooarchaeological data were sourced from three Iron Age sites – Mas Castellar de
Pontós, Sant Julià de Ramis and Sant Sebastià de la Guarda (Figure 1). These sites
were selected for their proximity to Empúries and the presence of large faunal
assemblages. These sites were oppida (large fortified settlements) with a third
century BC occupation, during which animal husbandry was focused on sheep and
goats. Cattle were the second most abundant species, followed by pigs. Kill-off
patterns show that cattle were slaughtered at adult and juvenile ages, suggesting
that meat, traction and milk were all used from these animals, with no
specialisation in any particular production (Colominas, 2013; Colominas et al.,
2014).
The early Roman sites

Zooarchaeological data were also assessed from three early Roman sites – Olivet
d'en Pujol, Vilauba and Tolegassos (Figure 1). As with the late Iron Age, these sites
were selected for their proximity to Empúries and the presence of large faunal
assemblages.
Olivet d'en Pujol was a storage place as a support facility of a farmstead. It was
occupied during the first century BC. Among the main domestic mammals, cattle
represent 59% of the faunal remains, followed by sheep/goat and pigs.

5

MANUSCRIT ACCEPTAT

Vilauba was a villa occupied from the first to the fifth century AD. Animal
husbandry here was also focused on cattle, with pig being the second most
abundant species, followed by sheep and goats.
Tolegassos was a villa occupied from the first century BC to the third century AD.
Data suggest that animal husbandry was focused on sheep/goat and cattle at this
site, with pig being less important in their relative frequency.
The analysis of cattle mortality profiles from these three sites shows that their
husbandry was more oriented towards traction rather than meat and milk
production (Colominas, 2013; Colominas et al., 2014).
Cattle

The focus is on cattle as this animal has a well-documented change in morphology
during the Roman period (Forest & Rodet-Belarbi, 2002; Lepetz & Yvinec, 2002;
MacKinnon, 2004; Oueslati, 2006; Duval et al., 2013). The possibility that this
change might be because of the presence of different (non-indigenous) cattle in the
new territories of the Roman Empire as a result of trade has been suggested by
numerous studies (Murphy et al., 2000; Schlumbaum et al., 2003; Albarella et al.,
2008; Colominas et al., 2014), although this trade has not been directly
investigated.
Another reason for focusing on cattle is that the diversification and movement of
Bos taurus has been extensively investigated using mitochondrial (mt) DNA
analyses of both modern and ancient samples (for example: Troy et al., 2001; BejaPereira et al., 2006; Bollongino et al., 2006; Ginja et al., 2010; Colominas et al.,
2015). Using data from these previous studies allows us to compare genetic
diversity with our Empúries sequences.
Our study focuses on one site that covers a wide period in time, and the combined
use of osteometric and genetic analyses will help us better understand the
movement of animals during the Roman era. Our hypothesis is that the presence of
new morphotypes and varied or increased genetic diversity will be a marker for a
greater variety of cattle types, which would indicate cattle trade when put in the
context of Roman Empire expansion.
Osteometric analysis

As it is possible to infer the presence of different morphotypes by estimating the
size and shape of metacarpals (Albarella, 1997; Tekkouk & Guintard, 2007; Klein et
al., 2010), we chose to focus the osteometric analysis on this skeletal element.
Another reason for selecting metacarpals was to avoid studying an element that
might be linked with trade of preserved meat, such as the humerus, femur or
scapula – even though several studies have shown that the commonly preserved
meat during the Roman period was that of pork, with preserved lamb and mutton
meat rarely being eaten (Maltby, 2006).

6

MANUSCRIT ACCEPTAT

A total number of 26 complete metacarpals, from 26 different individuals, were
available from Empúries: three from the Greek occupation (second century BC)
and 23 from the Roman occupation (first century BC – third century AD) (Table 1).
These 26 metacarpals constitute the only complete, unbroken and well preserved
metacarpals from the entire assemblage at Empúries. Measurements were taken
according to von den Driesch (1976) and included the greatest length (GL), the
maximum width (Bp) and depth (Dp) of the proximal epiphysis, the maximum
width (Bd) and depth (Dd) of the distal epiphysis with its condyles, and the
minimum width of the diaphysis (SD). The log-ratio technique was used to pool all
dimensional categories – this consists of calculating the difference between the
decimal logarithms for each measurement taken from the archaeological material
and the corresponding dimensions from a reference group or individual (Simpson,
1941; Meadow, 1999). The standard measurements for comparison corresponded
to a modern cow from the Camargue breed (Helmer, 1979). This standard was
considered the most appropriate in terms of the geographic area and the period
under study. All bones were mature and had no sign of pathology. Size and shape
relationships between metacarpal measurements were investigated using
principal component analysis (PCA) using the PAST Package (Hammer et al.,
2001). Where necessary, the Kruskall–Wallis test (as implemented in PAST) was
used to determine the level of statistical validity for the observed differences.
Table 1. Information about the samples studied and details of the mtDNA results
Chronology

%
cattle

Metacarpal
sample

Amplification
success

Fragment
length

Haplogroup

EM47

Positive

240 bp

T3

Negative

—

—

Negative

—

—

Positive

240 bp

T1

EM21

Positive

67 bp

T/T3

EM25

Negative

—

—

EM28
1st c. AD

EM51

EM80

1st c. BC

—

EM30

17%

—

EM53

2nd c. BC

Negative

Positive

240 bp

T3

EM29

Negative

—

—

EM46

Negative

—

—

EM50

Positive

240 bp

T3

EM52

Positive

175 bp

T/T3

15%

16%

7

MANUSCRIT ACCEPTAT

Chronology

%
cattle

Metacarpal
sample

Amplification
success

Fragment
length

Haplogroup

EM55

—

Positive

240 bp

T3

EM24

Negative

—

—

EM48

Negative

—

—

EM49

Negative

—

—

EM54

Negative

—

—

EM81

Positive

203 bp

T1

EM20

Negative

—

—

EM26

Negative

—

—

EM27

Positive

240 bp

T3

EM56

Negative

—

—

EM57

Positive

240 bp

T3

EM58

Negative

—

—

EM82

3rd c. AD

—

EM22

2nd c. AD

Negative

Positive

175 bp

T/T3

21%

14%

Molecular analysis

In order to be able to directly compare the genetic and morphometric data, the
same 26 metacarpals were used for the ancient DNA analyses. DNA was
successfully amplified and analysed from 11 samples at the University of
Cambridge. Extraction, contamination controls and amplification reactions were as
described previously (Campana, 2007), and followed standard ancient DNA
practice. For each sample, at least two independent DNA extractions were
performed, and extraction and PCR negative controls all produced negative results.
Samples were amplified with the primers detailed in Colominas et al. (2015),
which amplified a 316 base pair (bp) sequence of the control region of the
mitochondrial genome, covering the most variable 240 bp region (Troy et al.,
2001). Seven of the 11 samples that had endogenous DNA generated the full
240 bp fragment (Table 1).
A total of 294 mtDNA sequences from extant Spanish and Portuguese B. taurus
breeds (from GenBank) were used to compare archaeological cattle diversity to
that found in modern populations (Table 2). The mtDNA sequences were aligned in
8

MANUSCRIT ACCEPTAT

MEGA (ver. 6; Tamura et al., 2013) to the reference B. taurus sequence (Anderson
et al., 1982), and all were truncated to the most variable 240 bp region (Troy et al.,
2001). Median-joining networks were constructed following Bandelt et al. (1995).
Analyses of inter-population genetic distances between extant and ancient
populations were performed in ARLEQUIN (ver. 3.5; Excoffier et al., 2007) using
published modern regional data from native breeds (Table 2; Cymbron et al., 1999;
Miretti et al., 2004; Beja-Pereira et al., 2006; Ginja et al., 2010). In addition, all
those samples that gave mitochondrial results (Table 1) were screened for a
nuclear DNA SNP that characterises sex, using primers zfxyF and zfxyR as detailed
in Svensson et al. (2008).
Table 2. Information and genetic diversity values of the 1131 modern native cattle used
for comparison over the 240 bp fragment of D-loop

Country

Breed

% samples per
haplogroup

Number of
samples
T

Spain

Empúries

7

Spain

Albera
Alistana

14

Spain

Avilena

7

Spain

Berrenda

33

Spain

Cardena
andaluza

Spain

T2

6

Spain

T1

Haplotype
diversity

T3
85.7

5

0.857 ± 0.137

100

3

0.600 ± 0.215

92.9

9

0.912 ± 0.059

100

3

0.667 ± 0.160

3

97

8

0.780 ± 0.042

7

14.3 14.3

71.4

7

1.000 ± 0.076

Monchina

5

40

60

4

0.900 ± 0.161

Spain

Morucha

5

20

80

5

1.000 ± 0.127

Spain

Mostrenca

21

19

81

9

0.886 ± 0.045

Spain

Negra
serrana

5

100

2

0.600 ± 0.175

Spain

Pajuna

9

100

5

0.722 ± 0.159

Spain

Retinta

16

62.5

37.5

5

0.775 ± 0.063

Spain

Toro de lidia 19

5.3

36.8

57.9

11

0.860 ± 0.071

Spain

Tudanca

50

50

4

0.867 ± 0.129

6

14.3

Number of
haplotypes

7.1

9

MANUSCRIT ACCEPTAT

Country

% samples per
haplogroup

Number of
samples

Breed

T
Total

T1

T2

Number of
haplotypes

Haplotype
diversity

T3

153

2.6

17.6 0.7

79.1

Portugal Alentejana

48

4.2

43.7

52.1

14

0.885 ± 0.026

Portugal Arouquesa

28

14.3

85.7

15

0.889 ± 0.050

Portugal Barrosa

31

3.2

96.8

16

0.929 ± 0.030

Portugal Brava de lide 16

12.5

87.5

6

0.683 ± 0.120

Portugal Cachena

16

25

75

8

0.800 ± 0.092

Portugal Garvonesa

23

17.4

82.6

9

0.684 ± 0.106

Portugal Marinhoa

16

12.5

Portugal Maronesa

29

6.9

Portugal Mertolenga

46

4.4

Portugal Mirandesa

16

6.25

93.75 8

0.758 ± 0.110

Portugal Preta

25

12

88

0.940 ± 0.029

15.6 1.4

80.3

Total

294

2.7

18.75 68.75 10

0.917 ± 0.049

93.1
13

2.2

8

0.650 ± 0.096

80.4

19

0.864 ± 0.043

15

Results
Osteometric analysis

By ordering the osteometric log-ratio analysis of cattle metacarpals from Empúries
and other sites in chronological order, an increase in cattle size can be seen
through time (Figure 2). Although the Empúries sample from the second century
BC only constituted a few animals, these were smaller than those present at the
same site from the first century BC onwards. The cattle from the first century BC
were similar to Roman Empúries cattle from the first to third centuries AD.
Although no differences were documented between the Roman Empúries samples
through the centuries, variability was more pronounced during this time, with
considerable morphological variation. This change was clearer when we included
cattle from late Iron Age sites (third to second centuries BC) located near Empúries
(Figure 1). The variability was reduced during these earlier chronologies and cattle
from these sites were similar to Empúries cattle from the second century BC.
Similar considerations can be made when early Roman samples from sites located
near Empúries are included (Figure 1), with cattle from these Roman sites being
10

MANUSCRIT ACCEPTAT

similar to Empúries cattle from the first century BC through to the third century
AD.

Figure 2.


Open in figure viewer

Log-ratio diachronic comparison of cattle metacarpal measurements from Empúries and late Iron
Age and early Roman sites located near Empúries (Figure 1). The central line of the box represents
the median of the data, and the lower and upper limits of the box, the first and third quartiles. The
whiskers correspond to the rest of the data.

The Kruskall–Wallis test corroborated these results (Table 3). There were
statistically highly significant differences between the samples from the
third/second century BC and each of the time periods of the first to third century
AD. At the same time, no statistic differences were seen between the samples from
the first, second or third century AD. Interestingly, first century BC samples were

11

MANUSCRIT ACCEPTAT

significantly different from third to second century BC samples, but not to samples
from later periods.
Table 3. Descriptive statistical parameters and Kruskal–Wallis p values for the osteometric
data used in Figures 2 and 4 by time periods. Significant values, with a P value of ≤0.025,
are shaded in grey
Metacarpal
measurements

Summary
n min. max. mean

Kruskall–Wallis test results
SD

V

1st c. BC

1st c. AD 2nd c. AD 3rd c. AD

3rd–2nd c. BC

54 −0.06 0.05 −0.01 0.026 0.0006 4.00E − 05 2.59E − 09 4.16E − 09 8.03E − 10

1st c. BC

26 −0.05 0.07 0.02

0.031 0.0009 —

0.9335

0.6515

0.6213

1st c. AD

57 −0.04 0.11 0.03

0.030 0.0009

—

0.6016

0.4898

2nd c. AD

77 −0.06 0.12 0.03

0.038 0.0015

—

0.9044

3rd c. AD

64 −0.05 0.13 0.03

0.037 0.0013

—

In order to have an immediate visual comparison of shapes and sizes of Roman
Empúries metacarpals, a log-ratio diagram was computed (Figure 3). The most
striking information obtained from this diagram was the variability in size and
shape of the Roman cattle metacarpals. It is possible to observe three subpopulations: (1) very robust and large specimens; (2) smaller more gracile
specimens (the majority of the sample); and (3) variably shaped but very small
individuals. These three Empúries sub-populations can be seen more clearly in
Figure 4, which shows several bivariate plots drawn using raw cattle metacarpal
measurements. Data from late Iron Age and early Roman sites located near
Empúries were again included for comparison. Figure 4A compared size variation,
with the use of GL against Bd measurements. While the first and second subpopulations differed in the distal width, the second and third sub-populations were
different from each other in both length and distal width. When we compared the
size of the Roman metacarpals and the third to second century BC metacarpals, we
noticed that both the length and the distal width clearly increased over time.

12

MANUSCRIT ACCEPTAT

Figure 3.


Open in figure viewer

Log-ratio diagram of dimensions of Roman cattle metacarpals from Empúries.

13

MANUSCRIT ACCEPTAT

Figure 4.


Open in figure viewer

Scatter plots of: A) length (GL) against distal width (Bd); B) length (GL) against minimum width of
the diaphysis (SD); and C) length (GL) against proximal width (Bp) measurements for the cattle
metacarpals from Empúries and later Iron Age and early Roman sites located near Empúries
(Figure 1).
14

MANUSCRIT ACCEPTAT

Figure 4B compares differences in gracility, with the plot of GL against SD
measurements. This figure showed similar results to Figure 4A, with an increase of
length and robustness across time. The three Empúries Roman sub-populations
were also, once again, documented – the first sub-population was long and robust,
the second long and gracile and the third short and gracile. Figure 4C aimed to look
at differences because of sex by plotting GL against Bp, with males more likely to
be both longer in length and have a greater proximal width. This plot again showed
high differences in length and width between the third and second century BC
samples and the Roman samples. It also clearly showed the presence of the three
Roman Empúries sub-populations, also documented in the other Roman sites. It
might be assumed that the first sub-population were bulls, the second oxen and the
third cows. However, as bulls have more robust bones than cows, and oxen
(castrated bulls) have longer bones than both bulls and cows (Albarella, 1997;
Guintard, 1998; MacKinnon, 2010), the distribution of Figure 4C does not match
very well with sex differences.
Molecular analysis

Amplification of a 240 bp section of the control region was successful in seven of
our 26 samples, with a further four samples yielding sequence data of less than
240 bp in length (Table 1). The sequences were aligned against a highly
polymorphic region of the control region as characterised by Troy et al. (2001)
(Table 4). All archaeological samples gave sequences similar to those encountered
in modern native cattle. The dominant feature among the archaeological samples
was the predominance of the T3 haplotype (six samples). Three individuals had
identical sequences to the T3 reference sequence (EM27, EM50 and EM57). In
addition, a further three individuals grouped within T3 haplotypes, but with one or
more nucleotide substitutions (EM22, EM28 and EM51). Three samples, where the
full sequence could not be generated (EM21, EM51 and EM82), could only be
classified as T/T3, because of the non-amplification of the diagnostic position at
16,255.
Table 4. Variation in mitochondrial control region sequences. The variable positions in
control region sequences of archaeological cattle samples aligned to the European
consensus haplotype (T3). Differences are indicated, with a period ( . ) denoting identity.
Sequence codes from Table 1 are given in the first column, and only variable sites are
shown. The sequence positions from the BOVMT GenBank sequence are given above each
column (accession number V00654; Anderson et al., 1982). The common and putatively
ancestral Middle Eastern and European consensus sequences are denoted T and T3, with
T1 and T2 denoting the consensus sequences from Africa and the Middle East/Europe,
respectively. All ancient sequences generated as part of this study are shown in bold type,
and each sample has been assigned to one of the four main haplogroups by means of its
relative position in the median-joining networks (Figure 5)

15

MANUSCRIT ACCEPTAT

Seq
ue
nce
cod
e

1
6
0
2
8

1
6
0
3
2

1
6
0
3
3

1
6
0
4
2

1
6
0
4
3

1
6
0
5
0

1
6
0
5
1

1
6
0
5
7

1
6
0
5
8

1
6
0
6
3

1
6
0
6
8

1
6
0
7
4

1
6
0
7
7

1
6
0
8
2

1
6
0
8
5

1
6
0
8
6

1
6
0
9
3

1
6
0
9
6

1
6
1
0
4

1
6
1
0
8

1
6
1
1
3

1
6
1
1
9

1
6
1
2
2

1
6
1
3
9

1
6
1
4
2

1
6
1
5
8

1
6
1
8
5

1
6
1
9
3

1
6
1
9
6

1
6
2
0
1

1
6
2
1
5

1
6
2
2
2

1
6
2
3
1

1
6
2
3
3

1
6
2
3
5

1
6
2
3
7

1
6
2
3
8

1
6
2
3
9

1
6
2
4
7

1
6
2
5
5

Assi
gnm
ent
to
hapl
ogr
oup

T T A T A C T G C C T T T G T G G C C T T T T C T G G C G G G G C G G G G G C T T3
.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

C T

.

.

.

.

.

T .

.

.

.

.

.

.

.

.

.

.

.

.

.

C .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

C T1

.

.

.

.

.

.

.

C .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

A .

.

.

.

.

.

.

.

.

.

.

.

C T2

EM
52,
.
EM
82

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

?

?

?

?

?

?

?

?

?

?

?

T/T
3

EM
.
21

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

?

T/T
3

EM
50, .
EM

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

T3

16

MANUSCRIT ACCEPTAT

Seq
ue
nce
cod
e

1
6
0
2
8

1
6
0
3
2

1
6
0
3
3

1
6
0
4
2

1
6
0
4
3

1
6
0
5
0

1
6
0
5
1

1
6
0
5
7

1
6
0
5
8

1
6
0
6
3

1
6
0
6
8

1
6
0
7
4

1
6
0
7
7

1
6
0
8
2

1
6
0
8
5

1
6
0
8
6

1
6
0
9
3

1
6
0
9
6

1
6
1
0
4

1
6
1
0
8

1
6
1
1
3

1
6
1
1
9

1
6
1
2
2

1
6
1
3
9

1
6
1
4
2

1
6
1
5
8

1
6
1
8
5

1
6
1
9
3

1
6
1
9
6

1
6
2
0
1

1
6
2
1
5

1
6
2
2
2

1
6
2
3
1

1
6
2
3
3

1
6
2
3
5

1
6
2
3
7

1
6
2
3
8

1
6
2
3
9

1
6
2
4
7

1
6
2
5
5

Assi
gnm
ent
to
hapl
ogr
oup

27,
EM
57
EM
.
51

.

.

.

.

.

.

.

.

T .

C .

.

EM
.
28

.

.

.

.

.

.

.

.

.

.

.

.

A .

A A .

EM
.
22

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

EM
.
80

.

.

.

.

T .

.

.

.

.

.

.

.

.

EM
?
81

?

?

?

?

?

?

?

.

C C C .

.

?

.

.

.

T .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

T3

.

.

.

.

.

.

.

.

.

G .

.

.

.

.

.

.

.

.

.

.

.

T3

A .

.

.

.

.

.

.

G .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

T3

.

.

.

.

.

C .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

C T1

.

.

.

.

.

C .

.

.

.

A .

.

.

.

.

.

.

.

.

.

.

.

.

C T1

17

MANUSCRIT ACCEPTAT

Two individuals belonged to haplogroup T1. Of these, one individual had an identical sequence to the T1 reference sequence (EM80),
while the other exhibited a T1 haplotype (EM81). The Empúries samples were compared with mtDNA sequences from extant Iberian
native B. taurus sequences from GenBank (Table 2) to compare modern and ancient cattle diversity. As can be seen from the medianjoining networks (Figure 5), the Roman archaeological data show a large diversity, similar to the living cattle diversity.

18

MANUSCRIT ACCEPTAT

Figure 5.


Open in figure viewer

Median-joining networks comparing Empúries samples with extant native Iberian breeds. The top
network shows all modern T3 sequences with the six T3 Roman samples where 240 bp could be
recovered (EM22, EM27, EM28, EM50, EM51 and EM57) while the bottom network shows the
location of EM80, one of the T1 sequences found at Empúries (only 203 bp of EM81 could be
amplified).

19

MANUSCRIT ACCEPTAT

Of the samples screened for sex typing, only a single individual (EM50) returned
sufficient quality sequence to be characterised. This individual was a female,
because the sequence traces displayed a T at the 243 SNP site.
Integrating osteometric and genetic data

PCA was carried out in order to link the morphological variability of the Empúries
samples documented through osteometric analyses with the genetic diversity
documented through molecular analyses (Figure 6). The description of the results
was based on the first two components, which explained 97.4% of the variation.
The first component exhibited the greatest positive correlation with the length
measurement (GL: 0.84%), followed by width of the proximal epiphysis (Bd:
0.32%) and width of the distal epiphysis (Bp: 0.32%). Positive values were
associated with the first and second Roman Empúries sub-populations (longer
metacarpals), while negative values were associated with the second century BC
Empúries metacarpals and the Roman Empúries sub-population 3 (shorter bones).
Despite the predominance of these three measurement variables, all of the abovementioned six variables participated positively in the correlation, reflecting the
overall size of the bones (Dd load with 0.18%; Dp with 0.16% and SD with 0.16%).
The second component contributed only 8.5% of the total explained variation and,
within this, the GL measurement contributed in a negative way (−0.53%). In
contrast, there was a high positive correlation with Bp (0.57%) and also a
significant correlation with Bd (0.38%), mainly reflecting stockiness of the bones.
The other variables also participated positively in this correlation (Dp load with
0.33%; SD with 0.29% and Dd with 0.21%). Therefore, the positive values of this
component originated from the first Roman Empúries sub-population, but also
from the second century BC Empúries metacarpals (robust metacarpals), while
negative values originated with the second and third Roman Empúries subpopulation (gracile bones).

20

MANUSCRIT ACCEPTAT

Figure 6.


Open in figure viewer

Principal component analysis of cattle metacarpal measurements from Empúries. The two
metacarpals typed to the taurine haplogroup T1 are shown in grey, while those typed to the taurine
haplogroup T3 or T/T3 (n = 8) are shown in black. The larger black circle corresponds to the
sample characterised as female.

Those metacarpals where a genetic haplotype was generated were marked with
filled circles. The metacarpals belonging to the T3 or T/T3 haplogroup (eight
metacarpals) were grouped with the first and second Roman sub-populations
(Figure 6; black filled figures). Conversely, the two metacarpals belonging to the T1
haplogroup were grouped with the third Roman sub-population (Figure 6; grey
filled figures). The individual characterised as female corresponds to the larger
black figure from the second Roman sub-population (Figure 6).

21

MANUSCRIT ACCEPTAT

Discussion
The morphological differences seen between the samples, not only in size but also
in shape, cannot solely be explained by differences between males, females and
castrates, or by an improvement of local cattle, as the change seen is not
progressive through time. The larger individuals appeared during the first century
BC and cohabited with the smaller individuals. The sex typing of one individual
from the second sub-population as female, and the genetic characterisation of
different haplotypes linked to different morphotypes (T/T3 associated to the first
and second sub-populations and T1 associated to the third sub-population),
corroborate the osteometric data. The results appear to reflect the simultaneous
presence of novel cattle morphotypes in the commercial city of Empúries during its
early Roman occupation. The large genetic diversity seen is suggestive of these
cattle morphotypes arriving to Empúries from different origins during the first
century BC, while the Roman city was still under construction.
Today, cattle milk breeds have thin and large bones in comparison to cattle meat
breeds, which are characterised by robust bones and short extremities. Cattle used
for traction are characterised by larger, more massive bones (Sañudo, 2011). This
being the case, the morphological and genetic differences of Empúries cattle may
reveal the importation of cattle from different regions for different purposes. Killoff patterns from this area show that during the third and second centuries BC,
cattle were slaughtered at both adult and juvenile ages, showing that meat, traction
and milk were probably all used from these animals with no specialisation to any
particular production. In contrast, during the Roman period, cattle were mainly
slaughtered at adult ages, most likely to exploit traction and perhaps to obtain milk
(Colominas, 2013). During the late Iron Age, the conformation of cattle was short
and robust; a feature that today is mainly linked to cattle meat breeds. With the
arrival of the Romans, not only do we see the continuation of this local stock, but
we have also documented two other types of cattle, one with thin large bones, and
another larger, more robust population. As mentioned above, these morphological
characteristics are linked today with those cattle oriented towards milk production
and traction respectively. Consequently, the diversity seen in the Roman period
could reflect the trade of specialised cattle varieties, more appropriate for milk
production and traction than the local stock. Roman written sources show the
existence of different varieties of cattle (Columella VI, 1 [1–3]; Varro II, 5 [10]) and
there is also archaeological evidence of different types during that period in Italy
(Mackinnon, 2010).
The trade of specialised products has already been documented at Empúries, with
the arrival of perfumes, textiles, jewels, tableware and wine from the
manufacturing centres of Italy, North Africa and the Eastern Mediterranean
(Ramon, 2008; Tremoleda, 2012). The wreck Culip IV, a trading vessel of about
10 m long, which sunk near Empúries during the Vespasian period (AD 69–79), is
an example of the maritime traffic in these coasts during the Roman Empire. Its
excavation produced evidence for a varied cargo, including red-slip wares from
southern Gaul, lamps from Italy and oil from the south of the Iberian Peninsula
(Nieto, 1989). This variety in goods from many regions led researchers to the
hypothesis that the entire cargo was acquired from the port warehouses in
22

MANUSCRIT ACCEPTAT

Narbonne (Nieto, 1989). It is reasonable to assume that cattle would also have
been acquired in Narbonne or in any other port of the Mediterranean basin.
Written sources show the existence of different routes from east to west, such as
the route between Liguria and Gaddes, following the coast and returning across
Balearic Islands and Sardinia or Sicily (Arnaud, 2005). There were also north–
south routes, such as between Narbonne and Carthago Nova following the coast
(Arnaud, 2005).
Once all these products arrived to Empúries, they would have been redistributed
in the surrounding territory via the road network. The surrounding area was
occupied by several villas devoted to arable farming and livestock activities, such
as Vilauba and Tolegassos, where imported products resulting from this inland
distribution have been documented (Castanyer & Tremoleda, 1999; Casas & Soler,
2003). In these villas, cattle of large dimensions have been attested, showing that
once cattle arrived to Empúries, they too were distributed between the villas to the
surrounding territory along with the other imported goods.

Conclusions
The current available data do not allow us to say where the imported cattle
originated from, and we do not know if this livestock trade was constant, as with
ceramics, or sporadic, with perhaps only a few bulls brought in to breed with local
cows. However, our study has shown that, as literary sources imply, a trade in
cattle for productive purposes was practiced during the early Roman period. The
implicit characteristics of livestock trade made it more complex than the trade of
raw materials or manufactured products, as it would have involved the
consideration of the journey time, the housing conditions of the animals and the
necessity to feed them en route. Despite these difficulties, and if we take into
account that the presence of larger cattle is documented in several new territories
of the Roman Empire (Gallia, Britannia, Germania, Pannonia, Dacia and Hispania;
Murphy et al., 2000; Forest & Rodet-Belarbi, 2002; Lepetz & Yvinec, 2002;
Schlumbaum et al., 2003; MacKinnon, 2004; Oueslati, 2006; Gudea, 2007; Albarella
et al., 2008; Boschin & Toskan, 2012; Colominas et al., 2014), we can venture that
cattle trade was vital during the early Roman period but has been undocumented
because of its invisibility in the archaeological record.
We consider that our new approach, focused on the combination of osteometry
and genetics, has delivered new and valuable data about Roman trade, a key
element of the economy of the Roman Empire. We have demonstrated that
livestock was another traded commodity during the early Roman period at the
trading post of Empúries. Our approach could be used at other sites and other time
periods to study further the impact of trade through time. In addition, in
assemblages where teeth are available, strontium isotope analysis could be used to
uncover direct information about animal mobility. We hope that this study
encourages other scholars to investigate the trade of Roman livestock as more
comparative data from other Mediterranean coastal sites in the Roman Empire is
needed.

23

MANUSCRIT ACCEPTAT

Acknowledgements
For providing access to the archaeological samples, we thank P. Castanyer, J.
Tremoleda and M. Santos. L. Colominas is currently supported by a postdoctoral
grant (no. FPDI-2013-18324), but would like to recognise a previous postdoctoral
fellowship (no. HA2010-0293) and support from the McDonald Grants and Awards
Fund 2013 (McDonald Institute for Archaeological Research, University of
Cambridge, UK).

References

















Adams C. 2012. Transport and the economy. The Cambridge Companion to the
Roman Economy, W Scheidel (ed.). Cambridge University Press: Cambridge; 218–
240. CrossRef
Albarella U. 1997. Shape variation of cattle metapodials: age, sex or breed? Some
examples from medieval and postmedieval sites. Anthropozoologica 25–26: 37–47.
Albarella U, Johnstone C, Vickers K. 2008. The development of animal husbandry
from the Late Iron Age to the end of the Roman period: a case study from SouthEast Britain. Journal of Archaeological Science 35: 1828–1848.
CrossRef | Web of Science® Times Cited: 20
Anderson S, De Bruijn MH, Coulson AR, Eperon IC, Sanger F, Young IG. 1982.
Complete sequence of bovine mitochondrial DNA. Journal of Molecular Biology
156: 683–717.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 1314
Aquilué X, Burés L, Buxó R, Castanyer P, Esteba J, Fernández E, de la Reguera J,
Pons E, Santos M, Solé J, Tremoleda J. 1999. Intervencions Arqueològiques a Sant
Martí d'Empúries (1994–1996). De l'Assentament Precolonial a l'Empúries Actual.
Monografies Emporitanes 9: Girona.
Arnaud P. 2005. Les Routes de la Navigation Antique. Itinéraires en Méditerranée.
Editions Errance: Paris.
Bandelt H-J, Forster P, Sykes BC, Richards MB. 1995. Mitochondrial portraits of
human populations using median networks. Genetics 141: 743–753.
PubMed | Web of Science® Times Cited: 749
Beja-Pereira A, Caramelli D, Lalueza-Fox C, Vernesi C, Ferrand N, Casoli A, Goyache
F, Royo LJ, Conti S, Lari M, Martini A, Ouragh L, Magid A, Atash A, Zsolnai A, Boscato
P, Triantaphylidis C, Ploumi K, Sineo L, Mallegni F, Taberlet P, Erhardt G, Sampietro
L, Bertranpetit J, Barbujani G, Luikart G, Bertorelle G. 2006. The origin of European
cattle: evidence from modern and ancient DNA. Proceedings of the National
Academy of Sciences of the United States of America 103: 8113–8118.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 165 | ADS
Bollongino R, Edwards CJ, Alt KW, Burger J, Bradley DG. 2006. Early history of
European domestic cattle as revealed by ancient DNA. Biology Letters 2: 155–159.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 64
Boschin F, Toskan B. 2012. Changes in cattle body size in Slovenja from the Iron
Age to the Early Middle Age. Atti del 6° Convengo Nazionale di Archeozoologia, J De
Grossi, D Saccà, C Tozzi (eds.). Associazione Italiana di Archeozoologia: Italy; 393–
395.
Buxó R, Colominas L, Saña M. 2007. Les espècies de fauna d'Empúries
documentades a través de l'arqueozoologia: Palaiapolis, Neapolis i ciutat romana.
Animals d'Empúries. La fauna i l'Home a l'Antiguitat, J Tremoleda (ed.). Museu
d'Arqueologia de Catalunya-Empúries: Empúries; 38–42.
24

MANUSCRIT ACCEPTAT






















Campana MG. 2007. Changes in cattle breeds after the bubonic plague: evidence
from ancient microsatellites. Unpublished MPhil dissertation. University of
Cambridge: Cambridge.
Casas J, Soler V. 2003. La Villa de Tolegassos: Una Explotación Arícola de Época
Romana en el Territorio de Ampurias. BAR International Series 1101: Oxford.
Castanyer P, Tremoleda J. 1999. La vil·la Romana de Vilauba. Un Exemple de
l'Ocupació i Explotació del territori a la Comarca del Pla de l'Estany. MAC-Girona,
Ajuntaments de Camós, Banyoles i Porqueres, Centre d'Estudis Comarcals de
Banyoles: Girona.
Colominas L. 2013. Arqueozoología y Romanización: Producción, Distribución y
Consumo de Animales en el Nordeste de la Península Ibérica entre los Siglos V aneV dne. BAR International Series 2480: Oxford.
Colominas L, Schlumbaum A, Saña M. 2014. The impact of the Roman Empire on
animal husbandry practices: study of the changes in cattle morphology in the
north-east of the Iberian Peninsula through osteometric and ancient DNA analyses.
Archaeology and Anthropological Science 6: 1–16.
CrossRef | Web of Science® Times Cited: 3
Colominas L, Edwards CJ, Beja-Pereira A, Vigne J-D, Silva RM, Castanyer P,
Tremoleda J, Saña M, Pérez-Ripoll M, Goyache F, Howe CJ, Barker G, Bower MA.
2015. Detecting the T1 cattle haplogroup in the Iberian Peninsula from Neolithic to
Medieval times: new clues to continuous cattle migration through time. Journal of
Archaeological Science 59: 110–117.
CrossRef | Web of Science® Times Cited: 3
Cymbron T, Loftus RT, Malheiro MI, Bradley DG. 1999. Mitochondrial sequence
variation suggests an African influence in Portuguese cattle. Proceedings of the
Royal Society B: Biological Sciences 266: 597–603.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 82
de Soto J. 2013. Los sistemas de transporte romanos y la configuración territorial
en el noroestepeninsular. O Irado Mar Atlântico. O Naufragio Bético Augustano de
Esposende (Norte de Portugal), R Morais, H Graja, A Morillo (eds.). Braga: Portugal;
1–15.
Duval C, Horard-Herbin MP, Lepetz S. 2013. Morphological changes in domestic
cattle in Gaul, from the second century BC to the fifth century AD: diversity of
herds in the Seine valley (France) and northern Gaul. Journal of Archaeological
Science 40: 3977–3990.
CrossRef | Web of Science® Times Cited: 3
Edwards CJ, Connellan J, Wallace PF, Park SDE, McCormick FM, Olsaker I,
Eythórsdóttir E, MacHugh DE, Bailey JF, Bradley DG. 2003. Feasibility and utility of
microsatellite markers in archaeological cattle remains from a Viking Age
settlement in Dublin. Animal Genetics 34: 410–416.
Wiley Online Library | PubMed | CAS | Web of Science® Times Cited: 13
Excoffier L, Laval G, Schneider S. 2007. Arlequin (version 3.0): an integrated
software package for population genetics data analysis. Evolutionary
Bioinformatics Online 1: 47–50.
PubMed | Web of Science® Times Cited: 9377
Forest V, Rodet-Belarbi I. 2002. A propos de la corpulence des bovins en France
durant les periodes historiques. Gallia 59: 273–306.
CrossRef
Ginja C, Penedo MC, Melucci L, Quiroz J, Martínez-Lopez OR, Revidatti MA,
Martínez-Martínez A, Delgado JV, Gama LT. 2010. Origins and genetic diversity of
New World Creole cattle: inferences from mitochondrial and Y chromosome
polymorphisms. Animal Genetics 41: 128–141.
Wiley Online Library | PubMed | CAS | Web of Science® Times Cited: 38

25

MANUSCRIT ACCEPTAT























Gudea A. 2007. Contributii la Istoria Economica a Daciei Romane. Cluj-napoca,
Mega: Studiu Archeozoologic.
Guintard C. 1998. Identifier et Mesurer l'Évolution de l'Élevage Bovin: un Problème
de Méthode en Archéozoologie (Comment Appréhender la Variabilité Biologique?).
Actes du VIe Congrès International d'Archéologie Médiévale 6: Dijon; 21–29.
Hammer O, Harper D, Ryan PD. 2001. PAST: Paleontological Statistics Software
Package for Education and Data Analysis. Palaeontologia Electronica 4: 9.
Web of Science®
Helmer D. 1979. Recherches sur l'Économie Alimentaire et l'Origine des Animaux
Domestiques d'après l'Étude des Mammifères Postpaléolithiques en Provence.
Unpublished PhD dissertation. Université de Montpellier: Montpellier.
Klein R, Franciscus R, Steele T. 2010. Morphometric identification of bovid
metapodials to genus and implications for taxon-free habitat reconstruction.
Journal of Archaeological Science 37: 389–401.
CrossRef | Web of Science® Times Cited: 19
Lauwerier M. 1988. Animals in Roman Times in the Dutch Eastern River Area.
Amersfoort, ROB: Amsterdam.
Lepetz S, Yvinec JH. 2002. Présence d'espèces animales d'origen méditerranéenne
en France du nord aux périodes romaine et médiévale: actions anthropiques et
mouvements naturels. Mouvements ou Déplacements de Populations Animals en
Méditerranée au Cours de l'Holocène, A Gardeisen (ed.). BAR International Series
1017: Oxford; 33–42.
MacKinnon M. 2004. Production and Consumption of Animals in Roman Italy:
Integrating the Zooarchaeological and Textual Evidence. Journal of Roman
Archaeology Supplementary Series 54: Portsmouth.
MacKinnon M. 2010. Cattle ‘breed’ variation and improvement in Roman Italy:
connecting the zooarchaeological and ancient textual evidence. World Archaeology
42: 55–73.
CrossRef | Web of Science® Times Cited: 13
Maltby M. 2006. Salt and animal products: linking production and use in Iron Age
Britain. Integrating Zooarchaeology. Proceedings of the 9th ICAZ Conference, M
Maltby (ed.). Oxbow Books: Oxford; 117–122.
Meadow R. 1999. The use of size index scaling techniques for research on
archaeozoological collections from the Middle East. Animalium Ex Ossibus, G
Becker (ed.). Verlag Marie Leidorf GmbH: Rahden; 285–300.
Miretti MM, Dunner S, Naves M, Contel EP, Ferro JA. 2004. Predominant Africanderived mtDNA in Caribbean and Brazilian Creole cattle is also found in Spanish
cattle (Bos taurus). Journal of Heredity 95: 450–453.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 32
Murphy P, Albarella U, Germany M, Locker A. 2000. Production, imports and
status: biological remains from a Late Roman farm at Great Holts Farm, Boreham,
Essex, UK. Environmental Archaeology 5: 35–48.
CrossRef
Nieto X. 1989. La ruta de la nau i la ruta del carregament: una hipòtesi de treball
sobre l'organització del comerç naval en el segle I d.c. Excavacions Arqueològiques
Subaquàtiques a Cala, I Culip, J Nieto, A Jover, P Izquierdo, AM Puig, A Alaminos, A
Martin, M Pujol, H Palou, S Colomer (eds.). Sèrie Monogràfica 9: Girona; 239–246.
Nieto X, Revil A, Morphange C, Vivar G, Rizzo E, Aguelo X. 2005. La fachada
marítima de Ampurias: estudios geofísicos y datos arqueológicos. Empúries 54:
71–100.
Oueslati T. 2006. Approche Archéozoologique des Modes d'Acquisition, de
Transformation et de Consommation des Ressources Animales dans le Contexte
Urbain Gallo-Romain de Lutèce (Paris, France). BAR International Series 1479:
Oxford.
26

MANUSCRIT ACCEPTAT
























Parodi Álvarez MJ. 2001. Ríos y Lagunas de Hispania Como vías de Comunicación.
La Navegación Interior en la Hispania Romana. Editorial Graficas Sol: Spain.
Ramon J. 2008. El comercio púnico en occidente en época tardorepublicana (siglos
II-I aC). Una perspectiva actual según el tráfico envasado en ánforas. Los Fenicios y
el Atlántico, R González, F López, V Peña (eds.). Cuarto Coloquio del CEFYP,
Universidad Complutense, Centro de Estudios Fenicios y Púnicos: Spain; 233–258.
Rauh N. 2003. Merchants, Sailors and Pirates in the Roman World. Tempus
Publishing: UK.
Santos M. 2008. L'arqueologia grega a Empúries. Un discurs en construcció. Annals
de l'Institut d'Estudis Empordanesos 39: 49–79.
Sañudo C. 2011. Atlas Mundial de Etnología Zootécnica. Editorial Servet: Spain.
Schlumbaum A, Stopp B, Breuer G, Rehazek A, Turgay M, Blatter R, Schibler J. 2003.
Combining archaeozoology and molecular genetics: the reason behind the changes
in cattle size between 150BC and 700 AD in Switzerland. Antiquity 77: 298.
Simpson G. 1941. Explanation of ratio diagrams. American Museum Novitates
1136: 23–25.
Svensson EM, Götherström A, Vretemark M. 2008. A DNA test for sex identification
in cattle confirms osteometric results. Journal of Archaeological Science 35: 942–
946.
CrossRef | Web of Science® Times Cited: 17
Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. 2013. MEGA6: molecular
evolutionary genetics analysis version 6.0. Molecular Biology and Evolution 30:
2725–2729.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 10685
Tchernia A. 2011. Les Romains et le Commerce. Centre Jean Bérard and Camille
Jullian: Naples.
Tekkouk F, Guintard C. 2007. Approche osteometrique de la variabilite des
metacarpus de bovins et recherché de modeles applicable pour l'archeozoologie:
cas de races rustiques Francaises, Algerienne et Espagnole. Revue de Médecine
Vétérinaire 158: 388–396.
Web of Science® Times Cited: 6
Tremoleda J. 2008. Cent anys d'excavacions arqueològiques a Empúries.
L'arqueologia Romana. Un camí obert. Annals de l'Institut d'Estudis Empordanesos
39: 81–100.
Tremoleda J. 2012. El comerç marítim en època Imperial. Del declivi urbà
d'Emporiae a l'Empúries Tardoromana. Dels Mercaders d'Empúries als Pescadors
de l'Escala, Museu d'Arqueologia de Catalunya-Empúries (ed.). Ajuntament de
l'Escala: Girona; 34–36.
Tresset A, Bollongino R, Edwards CJ, Hughes S, Vigne J-D. 2009. Early diffusion of
domestic bovids in Europe: an indicator for human contacts, exchanges and
migrations? Becoming Eloquent. Advances in the Emergence of Language, Human
Cognition, and Modern Cultures, F d'Errico, JM Hombert (eds.). John Benjamins
Publishing Company: The Netherlands; 73–92.
Troy CS, MacHugh DE, Bailey JF, Magee DA, Loftus RT, Cunningham P, Chamberlain
AT, Sykes BC, Bradley DG. 2001. Genetic evidence for Near-Eastern origins of
European cattle. Nature 410: 1088–1091.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 316
Vigne J-D, Zazzo A, Saliège JF, Poplin F, Guilane J, Simmons A. 2009. Pre-Neolithic
wild boar management and introduction to Cyprus more than 11,400 years ago.
Proceedings of the National Academy of Sciences of the United States of America
106: 16135–16138.
CrossRef | PubMed | CAS | Web of Science® Times Cited: 42 | ADS

27

MANUSCRIT ACCEPTAT



Vivar G. 2012. Les noves infraestructures portuàries en el litoral. Dels Mercaders
d'Empúries als Pescadors de l'Escala, Museu d'Arqueologia de Catalunya-Empúries
(ed.). Ajuntament de l'Escala: Empúries; 31–34.
Von den Driesch A. 1976. A Guide to the Measurement of Animal Bones from
Archaeological Sites. Peabody Museum: Harvard.

28