Papers by Roz Gillis
Effect on milk production and calf performance of milking cross bred European/zebu cattle in the ... more Effect on milk production and calf performance of milking cross bred European/zebu cattle in the absence or presence of the calf, and of rearing their calves artificially. Tropical Animal Production 5:1-25
Animal Husbandry and Hunting in the Central and Western Balkans Through Time, 2020
Anthropological and Archaeological Sciences, 2020
The plant and animal components of Linearbandkeramik (LBK) subsistence systems were remarkably un... more The plant and animal components of Linearbandkeramik (LBK) subsistence systems were remarkably uniform with cattle, emmer and einkorn wheat providing the primary source of sustenance for Europe's earliest agricultural communities. This apparent homogeneity in plant and animal use has been implicitly understood to indicate corresponding similarity in the types of husbandry practices employed by LBK farmers across the entire distribution of the LBK culture. Here, we examine the results from the stable (δ 13 C/δ 15 N) isotope analysis of animal bone and cereal grains from the site of Vráble-Veľké Lehemby (Slovakia), providing new information about Linearbandkeramik farming practices in the western Carpathians. Moderately high carbon isotope values from animal bone collagen show that all livestock were pastured in open areas with no evidence of forest pasturing, previously associated with LBK settlements in northwestern Europe. High δ 15 N values measured from domesticated cereal grains suggest manuring took place at the site, while 15 N enrichment in bone collagen suggest livestock fed on agricultural by-products and possibly grains. An integrated plant-animal management system was in use at Vráble where livestock grazed on cultivation plots post-harvest. Use of such strategy would have helped fatten animals before the lean winter months while simultaneously fertilising agricultural plots with manure. This study contributes to our growing understanding that although the building blocks of LBK subsistence strategies were remarkably similar, diversity in management strategies existed across central and northwestern Europe.
Environmental Archaeology: The Journal of Human Palaeoecology, 2019
Archaeozoological assemblages are important sources of information on past management strategies,... more Archaeozoological assemblages are important sources of information on past management strategies, which are influenced by cultural practices as well as the physical geography and climate. Sheep, goat and cattle arrived in Europe with early Neolithic migrants. Their distribution is believed to have been mainly influenced by the geography of European regions although individual species may have held symbolic importance for specific Neolithic cultures. Domesticated animal mortality data derived from dental eruption, wear and replacement can provide insights into slaughter management and consequently animal husbandry practices. Previous studies have focused on caprines (sheep and goat) collectively as a results of their morphological similarity. Here we present a species specific study of sheep and goat mortality data from early European and Anatolian Neolithic contexts using correspondence analysis. The results show that for sheep there were significant differences in slaughter management practices between regions, cultures and site types whereas for goats there was none. This initial examination into sheep and goat husbandry during the Neolithic suggests that cultural practices as well as regional geography played an important role in shaping management practices.
Ludwinowo, site 7 Neolithic settlement in Kuyavia, Edited by Joanna Pyzel, 2019
Age-at-death profiles constructed from archaeozoological data have been used for decades to infer... more Age-at-death profiles constructed from archaeozoological data have been used for decades to infer the goals of prehistoric herd management strategies. Several 'ideal' profiles have been proposed as models for the optimal kill-off profiles that represent specific husbandry strategies, such as maximising milk or meat yields, which can then be compared to archaeological profiles. We evaluate the goodness of fit of ten caprine archaeological age-at-death profiles to five published idealised profiles, whilst properly accounting for sampling error and data where the age classes of observations are uncertain. We statistically reject all tested idealised profiles as plausible models to explain the data, and instead propose that a Gamma distribution provides a simpler and better general model to represent possible herd management strategies. Furthermore, we show that archaeological profiles can be summarised well using Gamma parameters, which allow multiple datasets (and models) to be easily compared and graphically represented together with minimal information loss, thus allowing clearer inferences to be drawn. Finally, we calculate likelihood distributions of the Gamma parameters, which provide confidence intervals that fully account for the uncertainties from small sample sizes and uncertain age classes. We have developed an R package 'GammaModel' to enable users to apply these tools to any age-at-death count data.
In southeastern Romania, the Gumelnit¸a culture is characterized by the appearance of tell sites.... more In southeastern Romania, the Gumelnit¸a culture is characterized by the appearance of tell sites. Whether this phenomenon was accompanied by increasing specialization of the economy may be investigated through the zooarchaeology of pastoral systems. The scale of herding is an important element of this framework. A case study was conducted on the tell sites of H^ ars¸ova and Bordus¸ani-Popin a situated in the Danube River basin. Both sites, located respectively on a terrace of the river and on the island of Balta Ialomit¸ei, delivered occupations from the Gumelnit¸a A2 dated to the second half of 5th millennium cal BC. Their occupants subsisted on an economy focused to a large extent on aquatic resources but also heavily dependent on cattle, pig and caprines husbandry and agriculture. The role of riverbanks resources in herding strategies and the extent to which the island of Balta Ialomit¸ei may have sustained domestic animal stocks was addressed through stable isotope analysis of animal skeleton remains. At both sites, results revealed local herding for cattle and caprines, reflected in an unexpectedly high contribution of C 4 plants to their diet, most likely from ruderal C 4 plants that are more abundant around the settlements as well as in cultivated fields. Domestic pigs had a higher trophic status than their wild counterparts, highlighting a significant contribution of animal protein to their diet most likely provided by human activities, suggesting that they were maintained in the settlement. Overall the findings suggest domestic stocks were reared in close proximity to the settlements, rather than in an extensive system. This scheme complements the small-scale cultivation system highlighted from the archaeobotanical analysis. Gumelnit¸a tell sites have been previously described as being part of larger pastoral systems including locations with complementary functions, although functional complementarity in time was not made explicit in this model. In this regard, the results obtained at H^ ars¸ova and Bordus¸ani-Popin a are not in favour of large-scale seasonal mobility.
Cattle dominate archaeozoological assemblages from the north-central Europe between the sixth and... more Cattle dominate archaeozoological assemblages from the north-central Europe between the sixth and fifth millennium BC and are frequently considered as exclusively used for their meat. Dairy products may have played a greater role than previously believed. Selective pressure on the lactase persistence mutation has been modelled to have begun between 6000 and 4000 years ago in central Europe. The discovery of milk lipids in late sixth millennium ceramic sieves in Poland may reflect an isolated regional peculiarity for cheese making or may signify more generalized milk exploitation in north-central Europe during the Early Neolithic. To investigate these issues, we analysed the mortality profiles based on age-at-death analysis of cattle tooth eruption, wear and replacement from 19 archaeological sites of the Linearbandkeramik (LBK) culture (sixth to fifth millennium BC). The results indicate that cattle husbandry was similar across time and space in the LBK culture with a degree of specialization for meat exploitation in some areas. Statistical comparison with reference age-at-death profiles indicate that mixed husbandry (milk and meat) was practised, with mature animals being kept. The analysis provides a unique insight into LBK cattle husbandry and how it evolved in later cultures in central and western Europe. It also opens a new perspective on how and why the Neolithic way of life developed through continental Europe and how dairy products became a part of the human diet.
In the absence of any direct evidence, the relative importance of meat and dairy productions to N... more In the absence of any direct evidence, the relative importance of meat and dairy productions to Neolithic prehistoric Mediterranean communities has been extensively debated. Here, we combine lipid residue analysis of ceramic vessels with osteo-archaeological age-at-death analysis from 82 northern Mediterranean and Near Eastern sites dating from the seventh to fifth millennia BC to address this question. The findings show variable intensities in dairy and nondairy activities in the Mediterranean region with the slaughter profiles of domesticated ruminants mirroring the results of the organic residue analyses. The finding of milk residues in very early Neolithic pottery (seventh millennium BC) from both the east and west of the region contrasts with much lower intensities in sites of northern Greece, where pig bones are present in higher frequencies compared with other locations. In this region, the slaughter profiles of all domesticated ruminants suggest meat production predominated. Overall, it appears that milk or the by-products of milk was an important foodstuff, which may have contributed significantly to the spread of these cultural groups by providing a nourishing and sustainable product for early farming communities.
Stable isotope analysis is an essential investigative technique, complementary to more traditiona... more Stable isotope analysis is an essential investigative technique, complementary to more traditional zooarchaeological approaches to elucidating animal keeping practices. Carbon (δ13C) and nitrogen (δ15N) stable isotope values of 132 domesticates (cattle, caprines and pigs) were evaluated to investigate one aspect of animal keeping, animal forage, at the Late Chalcolithic (mid-fourth millennium BC) site of Çamlıbel Tarlası, which is located in north-central Anatolia. The analyses indicated that all of the domesticates had diets based predominantly on C3 plants. Pig and caprine δ13C and δ15N values were found to be statistically indistinguishable. However, cattle exhibited distinctive stable isotope values and, therefore, differences in diet from both pigs and caprines at Çamlıbel Tarlası. This difference may relate to the distinct patterns of foraging behaviour exhibited by the domesticates. Alternatively, this diversity may result from the use of different grazing areas or from the foddering practices of the Çamlıbel Tarlası inhabitants.
Archaeozoological mortality profiles have been used to infer site-specific subsistence strategies... more Archaeozoological mortality profiles have been used to infer site-specific subsistence strategies. There is however no common agreement on the best way to present these profiles and confidence intervals around age class proportions. In order to deal with these issues, we propose the use of the Dirichlet distribution and present a new approach to perform age-at-death multivariate graphical comparisons. We demonstrate the efficiency of this approach using domestic sheep/goat dental remains from 10 Cardial sites (Early Neolithic) located in South France and the Iberian Peninsula. We show that the Dirichlet distribution in age-at-death analysis can be used: (i) to generate Bayesian credible intervals around each age class of a mortality profile, even when not all age classes are observed; and (ii) to create 95% kernel density contours around each age-at-death frequency distribution when multiple sites are compared using correspondence analysis. The statistical procedure we present is applicable to the analysis of any categorical count data and particularly well-suited to archaeological data (e.g. potsherds, arrow heads) where sample sizes are typically small.
TheNorth-WesternMediterraneanwitnessed a rapid expansion of farmers and their livestock during th... more TheNorth-WesternMediterraneanwitnessed a rapid expansion of farmers and their livestock during the EarlyNeolithic
period. Depending on the region, cattle played a more or less important role in these communities; however
how these animalswere exploited for theirmilk is not clear. Herewe investigate calfmortality to determine indirectly
whether cattle dairying was practised by Early Neolithic stock herders. Age-at-death (AtD) frequencies for calves
from two sites: Trasano (Italy, Impressa culture: 7–6th millennium BC) and La Draga (Spain, Cardial culture: 6th
millennium BC) were estimated from dental eruption and development stages, and measurements of un-fused
post-cranial material. Adult age classes are well represented in the dental AtD frequencies and were interpreted
as the result of the slaughter of prime beef and retired lactating females. For calves aged less than 12 months,
there was no statistical difference in the AtD frequencies based on dental and post-cranial material indicating that
the data is a good representation of the mortality patterns of calves, either natural or deliberate. At both sites
there was a strong mortality peak at 3–6months in all AtD profiles. At La Draga, this peak was clearly
differentiated from a peak at 0–1month, which can be interpreted neonatal mortality possible a consequence
of the birthing season coinciding with the end of winter during more humid climatic conditions
that at present. The deliberate slaughter peak around 3–6months is discussed, and we propose that
stock herders controlled the mortality of infant classes, possibly in response to variable external environment
pressures while maintaining animal productivity.
International Journal of Osteoarchaeology
For cattle (Bos taurus), age estimations using dental criteria before the eruption of the first m... more For cattle (Bos taurus), age estimations using dental criteria before the eruption of the first molar (3–8 months)
have large error margins. This hampers archaeozoological investigation into perinatal mortality or the putative
slaughtering of very young calves for milk exploitation. Previous ageing methods for subjuveniles have focused
on the length of unfused bones, but it is rarely possible to use them because they are restricted to foetuses and
because of the fragmentation of bones. This paper presents new age prediction models based on length,
breadth and depth of post cranial bones produced from a dataset of modern calves (n=27). This reference
collection was compiled from material of known age at death, sex and breed from collections in Britain, France,
Germany and Switzerland. Linear regression models were constructed using themodern data for age prediction,
and thesemodels were then successfully tested and assessed using aMiddle Neolithic assemblage of complete
calves’ skeletons from Bourguignon-Lès-Morey, France. From the assessment, the astragalus and metapodials
were determined to be the most reliable bones, and the femur was the worst. Measurements of the epiphyseal
and distal elements and depth measurements were the most reliable. For ages before 12months, these models
can provide ±1month age estimates
World Archaeology, Sep 2013
Borduşani-Popină is a Gumelniţa tell site in south-eastern Romania. The cattle mortality profile ... more Borduşani-Popină is a Gumelniţa tell site in south-eastern Romania. The cattle mortality profile suggests a
husbandry oriented towards prime meat exploitation and dairy production highlighted by the keeping of
cattle to advanced age. Besides, the culling strategy also targeted young calves. A stable isotope ratio study
was undertaken on dental rows. Bone and dentine collagen δ15N values show that the calves within the
slaughtering peak were well-advanced in the weaning process, suggesting that the slaughter was delayed
until the end of the cows’ lactation. A consequence would be the sharing of milk production between
herders and calves. High inter- and intra-individual variability in bone collagen and enamel bioapatite δ13C
values indicated variations in the seasonal ratio of C4 and C3 plants in fodder and between age groups.
Overall, the complementary study of mortality profiles and stable isotopes provide evidence of sophisticated
husbandry during the fifth millennium cal. BC.
Journal of Archaeological Science, Jan 1, 2011
This paper presents an assessment of all known dental and mandibular morphological criteria for d... more This paper presents an assessment of all known dental and mandibular morphological criteria for differentiating sheep (Ovis aries) and goats (Capra hircus) using for the first time an archaeological sample of complete caprine skeletons (90 sheep and 13 goats) from burials at Kerma (Sudan, 3rd and 2nd millennia BC). The species determinations were assessed using cranial and post-cranial morphological criteria. Consequently, the reliability (percentage of correct determination) and efficiency (complementary percentage of intermediate scores i.e. neither sheep nor goat) of 38 individual dental criteria could be fully assessed using a prehistoric homogeneous domesticate population. We demonstrate that, for this sample, individual criteria for lacteal teeth are more reliable for sheep (sheep: 95 ± 3%) than the adult premolars (85 ± 5%) and molars (sheep: 88 ± 2%), whereas for goats premolar criteria were more reliable (83 ± 12%). For efficiency, lacteal dental criteria are better (goat: 97 ± 5%; sheep: 95 ± 3%) than those for premolar (goat: 85 ± 10%; sheep: 79 ± 5%) and molar (goat: 82 ± 6%; sheep: 83 ± 2). We also demonstrated that most isolated teeth can be determined with less than 10% error. However, on average, within specific age classes (0–1 year, 1–4 years and more than 4 years), isolated teeth increased in reliability and decreased in efficiency. The average reliability of the criteria for complete mandibles for each age class for goats and sheep was 100%, when the efficiency was 67, 40 and 50% for goats and over 90% for sheep. This is due to the effect of age on the efficiency of isolated criteria and the poor performance of specific criteria mainly those P3, M1 and M2. We conclude that separate species kill-off profiles are possible. The effect of age on dental criteria would not significantly change the interpretation for specific subsistence strategies focused on one particularly species.
Thesis Chapters by Roz Gillis
PhD thesis, 2012
The review of historic, ethnographic and modern references of dairy husbandry The review of histo... more The review of historic, ethnographic and modern references of dairy husbandry The review of historical, ethnographic and modern dairy systems will provide a reference for the lactation process mechanism for sheep, goats and cattle. This is important to consider particularly for cattle as the presence of a calf to stimulate milk production is a much debated subject (
PhD thesis, 2012
The review of historic, ethnographic and modern references of dairy husbandry The review of histo... more The review of historic, ethnographic and modern references of dairy husbandry The review of historical, ethnographic and modern dairy systems will provide a reference for the lactation process mechanism for sheep, goats and cattle. This is important to consider particularly for cattle as the presence of a calf to stimulate milk production is a much debated subject (
Uploads
Papers by Roz Gillis
period. Depending on the region, cattle played a more or less important role in these communities; however
how these animalswere exploited for theirmilk is not clear. Herewe investigate calfmortality to determine indirectly
whether cattle dairying was practised by Early Neolithic stock herders. Age-at-death (AtD) frequencies for calves
from two sites: Trasano (Italy, Impressa culture: 7–6th millennium BC) and La Draga (Spain, Cardial culture: 6th
millennium BC) were estimated from dental eruption and development stages, and measurements of un-fused
post-cranial material. Adult age classes are well represented in the dental AtD frequencies and were interpreted
as the result of the slaughter of prime beef and retired lactating females. For calves aged less than 12 months,
there was no statistical difference in the AtD frequencies based on dental and post-cranial material indicating that
the data is a good representation of the mortality patterns of calves, either natural or deliberate. At both sites
there was a strong mortality peak at 3–6months in all AtD profiles. At La Draga, this peak was clearly
differentiated from a peak at 0–1month, which can be interpreted neonatal mortality possible a consequence
of the birthing season coinciding with the end of winter during more humid climatic conditions
that at present. The deliberate slaughter peak around 3–6months is discussed, and we propose that
stock herders controlled the mortality of infant classes, possibly in response to variable external environment
pressures while maintaining animal productivity.
have large error margins. This hampers archaeozoological investigation into perinatal mortality or the putative
slaughtering of very young calves for milk exploitation. Previous ageing methods for subjuveniles have focused
on the length of unfused bones, but it is rarely possible to use them because they are restricted to foetuses and
because of the fragmentation of bones. This paper presents new age prediction models based on length,
breadth and depth of post cranial bones produced from a dataset of modern calves (n=27). This reference
collection was compiled from material of known age at death, sex and breed from collections in Britain, France,
Germany and Switzerland. Linear regression models were constructed using themodern data for age prediction,
and thesemodels were then successfully tested and assessed using aMiddle Neolithic assemblage of complete
calves’ skeletons from Bourguignon-Lès-Morey, France. From the assessment, the astragalus and metapodials
were determined to be the most reliable bones, and the femur was the worst. Measurements of the epiphyseal
and distal elements and depth measurements were the most reliable. For ages before 12months, these models
can provide ±1month age estimates
husbandry oriented towards prime meat exploitation and dairy production highlighted by the keeping of
cattle to advanced age. Besides, the culling strategy also targeted young calves. A stable isotope ratio study
was undertaken on dental rows. Bone and dentine collagen δ15N values show that the calves within the
slaughtering peak were well-advanced in the weaning process, suggesting that the slaughter was delayed
until the end of the cows’ lactation. A consequence would be the sharing of milk production between
herders and calves. High inter- and intra-individual variability in bone collagen and enamel bioapatite δ13C
values indicated variations in the seasonal ratio of C4 and C3 plants in fodder and between age groups.
Overall, the complementary study of mortality profiles and stable isotopes provide evidence of sophisticated
husbandry during the fifth millennium cal. BC.
Thesis Chapters by Roz Gillis
period. Depending on the region, cattle played a more or less important role in these communities; however
how these animalswere exploited for theirmilk is not clear. Herewe investigate calfmortality to determine indirectly
whether cattle dairying was practised by Early Neolithic stock herders. Age-at-death (AtD) frequencies for calves
from two sites: Trasano (Italy, Impressa culture: 7–6th millennium BC) and La Draga (Spain, Cardial culture: 6th
millennium BC) were estimated from dental eruption and development stages, and measurements of un-fused
post-cranial material. Adult age classes are well represented in the dental AtD frequencies and were interpreted
as the result of the slaughter of prime beef and retired lactating females. For calves aged less than 12 months,
there was no statistical difference in the AtD frequencies based on dental and post-cranial material indicating that
the data is a good representation of the mortality patterns of calves, either natural or deliberate. At both sites
there was a strong mortality peak at 3–6months in all AtD profiles. At La Draga, this peak was clearly
differentiated from a peak at 0–1month, which can be interpreted neonatal mortality possible a consequence
of the birthing season coinciding with the end of winter during more humid climatic conditions
that at present. The deliberate slaughter peak around 3–6months is discussed, and we propose that
stock herders controlled the mortality of infant classes, possibly in response to variable external environment
pressures while maintaining animal productivity.
have large error margins. This hampers archaeozoological investigation into perinatal mortality or the putative
slaughtering of very young calves for milk exploitation. Previous ageing methods for subjuveniles have focused
on the length of unfused bones, but it is rarely possible to use them because they are restricted to foetuses and
because of the fragmentation of bones. This paper presents new age prediction models based on length,
breadth and depth of post cranial bones produced from a dataset of modern calves (n=27). This reference
collection was compiled from material of known age at death, sex and breed from collections in Britain, France,
Germany and Switzerland. Linear regression models were constructed using themodern data for age prediction,
and thesemodels were then successfully tested and assessed using aMiddle Neolithic assemblage of complete
calves’ skeletons from Bourguignon-Lès-Morey, France. From the assessment, the astragalus and metapodials
were determined to be the most reliable bones, and the femur was the worst. Measurements of the epiphyseal
and distal elements and depth measurements were the most reliable. For ages before 12months, these models
can provide ±1month age estimates
husbandry oriented towards prime meat exploitation and dairy production highlighted by the keeping of
cattle to advanced age. Besides, the culling strategy also targeted young calves. A stable isotope ratio study
was undertaken on dental rows. Bone and dentine collagen δ15N values show that the calves within the
slaughtering peak were well-advanced in the weaning process, suggesting that the slaughter was delayed
until the end of the cows’ lactation. A consequence would be the sharing of milk production between
herders and calves. High inter- and intra-individual variability in bone collagen and enamel bioapatite δ13C
values indicated variations in the seasonal ratio of C4 and C3 plants in fodder and between age groups.
Overall, the complementary study of mortality profiles and stable isotopes provide evidence of sophisticated
husbandry during the fifth millennium cal. BC.