Celtic population from Verona (Italy)

Journal of Archaeological Science: Reports 17 (2018) 30–38

Contents lists available at ScienceDirect

Journal of Archaeological Science: Reports journal homepage: www.elsevier.com/locate/jasrep

Infant feeding practices in a pre-Roman/Celtic population from Verona (Italy)

MARK

Zita Laffranchia,⁎, Sylvia A. Jiménez-Brobeila, Antonio Delgado-Huertasb, Arsenio Granados-Torresb, María Teresa Mirandac a Laboratorio de Antropología, Departamento de Medicina Legal, Toxicología y Antropología Física, Facultad de Medicina, Parque Tecnológico de la Salud, Universidad de Granada, Av. de la Investigación, 11, 18016 Granada, Spain b Laboratorio de Biogeoquímica de Isótopos Estables, Instituto Andaluz de Ciencias de la Tierra IACT (CSIC-UGR), Av. de las Palmeras, 4, Armilla, 18100 Granada, Spain c Departamento de Estadística e Investigación Operativa, Facultad de Medicina, Parque Tecnológico de la Salud, Universidad de Granada, Av. de la Investigación, 11, 18016 Granada, Spain

A R T I C L E I N F O

A B S T R A C T

Keywords: Isotopic analysis Feeding practices Non-adults Diet Pre-Roman Celts

We studied an osteological sample from the pre-Roman/Celtic necropolis of Seminario Vescovile of Verona (Italy), dated to the 3rd to 1st century BCE and attributed to the Cenomani Gauls population. The sample is mostly composed of well-preserved infants in their first months and years of life. In this study we combined isotopic (δ15N and δ13C) and anthropological evidence with the aim of investigating infant feeding practices in non-adult samples. 36 non-adults were selected and divided into seven age phases. The isotopic composition of their rib bone collagen was determined and related to the mean values of adults (n = 54) and animals (n = 7). δ15N values ranged between 7.1‰ and 12.9‰ (AIR), with a mean of 10.2‰ ( ± 1.5‰), while δ13C values ranged between −20.2‰ and − 9.7‰ (V-PDB), with a mean of − 15.3‰ ( ± 2.5‰). These results indicate the initiation of transitional feeding around six months. Significantly high δ15N values in some infants up to two years old suggests prolonged breast-milk consumption. In comparison with the δ15N and δ13C data from the adult females (n = 21), considered as potential mothers, these infants clearly show the effect of breastfeeding on trophic level (δ15N enriched between 1.8 and 3.3‰).

1. Introduction Quite a lot of studies have been published on the human diet during different phases of infancy in past times and its relationship with health, morbidity and mortality (e.g. Dupras and Tocheri, 2007; Herrscher, 2013; Katzenberg et al., 1996; Kaupová et al., 2014; Knipper et al., 2016; Nájera-Colino et al., 2010; Pearson et al., 2010; Redfern et al., 2012; Schurr, 1998; Waters-Rist et al., 2011). Most research on human health conditions has been based on qualitative analyses, e.g., the study of non-specific stress indicators (Larsen, 2015; Schurr, 1998). However, quantitative analysis has become possible (Katzenberg et al., 1996) through the study of isotopic composition, especially δ15N values, which are related to a greater or lesser intake of proteins of animal origin, and can also be used to estimate legumes vs non-legumes or aquatic vs terrestrial animals consumption as main food resources (DeNiro and Epstein, 1981; Hedges and Reynard, 2007). The combined study of collagen isotopic values and paleopathologic features of human bones may lead to major advances in anthropological research (Larsen,

2015). Stable isotope (δ15N and δ13C) analysis has been widely utilized to investigate dietary patterns, including infant feeding in different chronological and geographical populations (e.g. Dupras et al., 2001; Dupras and Tocheri, 2007; Fogel et al., 1989; Fuller et al., 2006a, 2006b; Herring et al., 1998; Jay et al., 2008; Katzenberg et al., 1996; Kaupová et al., 2014; Prowse et al., 2008; Schurr, 1998; Waters-Rist and Katzenberg, 2010). During the breastfeeding period, consumption of maternal tissue through breast milk ingestion by the newborns places them one trophic level above their mothers in the food chain, with an enrichment in δ15N of ≈ 2‰ and in δ13C of ≈ 1‰ in comparison to maternal values (Fogel et al., 1989; Reynard and Tuross, 2015). Infant δ15N values then decrease during weaning through the gradual substitution of breast milk with supplementary foods. Weaning is a process, not a single event, which starts when non-breast milk foods (liquids or solids) are first introduced and ends when the consumption of breast milk ceases entirely (Reynard and Tuross, 2015). When the infant is fully weaned and breastfeeding has ceased, the reduction to maternal

⁎ Corresponding author at: Departamento de Medicina Legal, Toxicología y Antropología Física, Facultad de Medicina, Parque Tecnológico de la Salud, Universidad de Granada, Av. de la Investigación, 11, 18016 Granada, Spain. E-mail addresses: zitalaff[email protected], [email protected] (Z. Laffranchi).

http://dx.doi.org/10.1016/j.jasrep.2017.10.040 Received 11 March 2017; Received in revised form 16 October 2017; Accepted 25 October 2017 2352-409X/ © 2017 Elsevier Ltd. All rights reserved.

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

levels is supposedly more rapid for δ13C values than for δ15N values (Fuller et al., 2006b). Hence, the study of δ13C values may be useful to trace the introduction of solid foods into the diet, while δ15N data are especially relevant to indicate the length of the breastfeeding period (Fuller et al., 2006a; Katzenberg et al., 1996). Finally, variations in stable isotope ratios in bone can yield information on episodes of disease or nutritional stress that cannot be macroscopically identified from the archaeological skeletal record. Thus, the study of δ15N in non-adult skeletons can contribute to revealing cases of death from nutrition-related disease (Fogel et al., 1989, 1997; Katzenberg et al., 1996; Katzenberg and Lovell, 1999). Sellen (2009) proposed a model about the weaning process comprised of 4 stages: i) exclusive breastfeeding, ii) breastfeeding plus complementary feeding (specific food for babies), iii) breastfeeding plus complementary and family foods (foods shared by older children and adults), iv) complete cessation of both breastfeeding and complementary food consumption and exclusive family food consumption. As suggested by Reynard and Tuross (2015), we also considered that the weaning process comprises three of the four stages proposed by Sellen (2009): considering stage ii as the beginning and stage iv as the end of the weaning process. It is also important to keep in mind the assumptions summarised in recent research (Beaumont et al., 2015; Reynard and Tuross, 2015), which warn to be cautious with overly definitive conclusions about weaning ages and subsequent interpretations. They recommend that the sole investigation of bone collagen is not adequate to detect the successive phases of dietary changes during infancy as it implies several general assumptions difficult to take into account for archaeological studies and propose the additional use of teeth micro-samples or other isotope ratios. Among them, the supposition that the δ15N and δ13C composition of the bone collagen of non-adults represents their diet at approximately the time of death, and that those infants who died are representative of the diet and physiology of the individuals belonging to that age group (Beaumont et al., 2015). Especially this last assumption is in conflict with the “Osteological Paradox” (Wood et al., 1992), which recommends that non-adults (or also adults) who have died may not be representative of the health condition of the entire population. Osteoarchaeological material from the pre-Roman/Celtic Seminario Vescovile necropolis (3rd to 1st century BCE) in Verona (Italy), attributable to the Cenomani Gauls population, provides an exceptional opportunity to analyze infant remains during the first days and months of life. There have been only a few anthropological studies on this particular Celtic group, mostly carried out in the 1980s (Capitanio, 1989; Corrain, 1987), with a few more recent ones (Teegen, 2014a, 2014b). This extensive new collection was subjected to a detailed anthropological and isotopic study to evaluate health status and dietary habits of the individuals (Laffranchi, 2015; Laffranchi et al., 2016). The isotopic results obtained reveal a diet mainly based on terrestrial animal protein and C4 plants, with no isotopic signals indicating the consumption of freshwater or marine foods (Laffranchi et al., 2016). These data are consistent with the archaeobotanical record from older and contemporary archaeological sites in Northeast Italy (Rottoli, 2014) and with ancient reports on the use of millet (C4 plants) in the daily diet of populations in northern Italy by Pliny the Elder (Naturalis Historiae XVIII, 83–84) and Columella (De Re Rustica, 2, 9, 14–16). Unfortunately, there is no information from contemporary historical sources (3rd-1st century BCE) on breastfeeding and weaning practices in these Celtic populations, and descriptions are only available from writers of the Roman Imperial period (e.g. Soranus of Ephesos, Galen of Pergamum), who indicate a highly flexible weaning timetable. Recommendations by ancient medical sources were to start transitional feeding after the 6th month and to stop breastfeeding from 2 years of age, introducing cow or goat milk, diluted wine, honey, and porridge, among other weaning foods; however, there is likely to have been a wide variability in weaning practices according to local customs and family circumstances (Killgrove and Tykot, 2013; Prowse et al., 2008).

Bioarchaeological analyses of the diet in ancient Roman society have provided direct evidence on the timing of breastfeeding and weaning timing. Prowse et al. (2008) studied a skeletal Roman sample from Isola Sacra, Rome (1st–3rd century CE) and found that transitional feeding commenced by the end of the first year and weaning by the age of 2–2 1/2 years. In contrast, Fuller et al. (2006b) described a gradual and prolonged period of transitional feeding from around 2 years to 3–4 years of age in a Late Roman sample from Oxfordshire, England, dated from 4th to 6th century CE. Killgrove and Tykot (2013) published isotopic data on 12 non-adults from the Imperial Roman necropolis (1st–3rd AD) of St. Callixtus, Casal Bertone, and Castellaccio Europarco and observed that children aged 2–3 years old were still being nursed, although they were probably weaned shortly afterwards. Finally, in another sample from the Roman period in Dakhleh Oasis, Egypt, dated from 250 to 450 CE, isotopic evidence indicates the introduction of complementary foods early in life, at around 6 months of age, with the completion of weaning at around 3 years of age (Dupras et al., 2001, Dupras and Tocheri, 2007). Given the lack of information on feeding practices among Celtic populations in Italy, we combined bioarchaeological and biochemical methods in order to investigate their feeding practices. The main objective of this study is to offer a reliable estimation of the age of different feeding transitions, by mostly detecting changes in stable δ15N and δ13C ratios in rib samples of a population of pre-Roman Celtic culture. 2. Archaeological and historical context The construction of an underground garage in the main courtyard of the Bishop's Seminary at Verona (Italy) (Fig. 1a–b) between 2005 and 2010 led to the discovery of a large pre-Roman necropolis (3rd to 1st century BCE) of a Roman-influenced Celtic population (Cenomani Gauls). The Seminario Vescovile necropolis is located in the eastern Veronetta district of Verona between the left bank of the river Adige (Fig. 1b) and the first hills (Thompson and Bersani, Unpublished report). During the pre-Roman period, Cenomani Gauls settled in an area corresponding to the current provinces of Brescia and Verona in Northeast Italy, as reported by ancient sources (e.g. Polybius, Titus Livius, Pliny the Elder etc.). The necropolis under study, although used by the local Celtic population, was already strongly influenced by some typical aspects of Roman culture, through trade and military alliances, as evidenced by a preliminary study of grave goods (see below). The historical context is characterized by a transition from the pre-Roman phase dominated by Cenomani Gauls and the advent of Roman domination, which was gradually established over the 1st century BCE (Late Republican times). The culmination of this long process of assimilation and integration by the Romans (“Romanisation”) was the foundation in the Cenomani Gauls territory of two Latin colonies, Brixia (now Brescia) and Verona in 89 BCE, and their subsequent acquisition of the status of Municipium in 49 BCE (Grassi, 1995, 2009; Malnati et al., 2004). The necropolis contains 163 simple burial graves that yielded a minimum (MNI) of 174 skeletons: 108 non-adults and 66 adults. They are all in a good state of preservation. The burials are currently under archaeological study and all results have not yet been published. The necropolis was tentatively attributed to the 2nd century BCE, based on the type of funeral grave goods observed in a preliminary study of the metal materials (Cavalieri Manasse, 2014). Although previously considered to have functioned from the 2nd century BCE until the subsequent construction of a Roman craft workshop in the first half of the 1st century CE, absolute dating from bone collagen analysis in some skeletons indicate its utilization starting from the 3rd century BCE (Laffranchi, 2015; Laffranchi et al., 2015). Individuals are laid out in supine position in single graves, whereas the typical funerary rituals in other pre-Roman/Celtic cemeteries include not only inhumations but also cremations, which were more 31

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

Fig. 1. a–b Map showing location of the city of Verona (a) and the location of the necropolis of Seminario Vescovile (orange area) on the left bank of the river Adige in Verona (b) (created by M. Bersani).

humic acids and most lipids, and the residue was then immersed in 10− 2 M HCl (pH 2) solution in closed Pyrex tubes at 100° for 17 h to solubilize the collagen. After centrifuging the tubes at 6300 rpm for 10 min, the supernatant (containing solubilized collagen) was lyophilized and underwent isotopic analysis. About 0.7 mg of collagen was weighed within a tin capsule per duplicate, and an elemental analyzer (Carlo Erba Model NA1500 NC) was employed for sample combustion; the N2 and CO2 obtained were separated using a chromatographic column and introduced into a mass spectrometer (Delta Plus XP) for isotope analysis. Commercial CO2 and N2 were used as the internal standards for the carbon and nitrogen isotopic analyses (see Laffranchi et al., 2016 for more information about the standards used). The analytical error for the δ15N and δ13C determinations was < 0.1‰. The measurement unit was “δ”, expressed as:

frequent. Grave goods, which are documented in most but not all of the burials, include animal bones, ceramics, decorations, and weapons, although the latter are rare (Cavalieri Manasse, 2014). Some ceramics show inscriptions in lepontic alphabet that reported onomastic forms (Solinas, 2014). 3. Materials and methods Age estimation was based on the following criteria: development and eruption of deciduous and permanent dentition (Ubelaker, 1989), development and measurement of temporal and occipital bones (Fazekas and Kósa, 1978; Redfield, 1970), development and fusion of epiphyses (Schaefer et al., 2009; Scheuer and Black, 2000; Ubelaker, 1989), and maximum length of long bone diaphyses (Schaefer et al., 2009; Scheuer and Black, 2000). The non-adult sample from Seminario Vescovile comprises 108 individuals; most of them (n = 63) are fullterm and neonates. 37 samples were selected for isotopic study and divided into 7 age phase categories: a: fullterm (37–42 weeks) neonates (n = 14); b: postneonates to 6 months (n = 4); c: > 6 months to 1 year (n = 5); d: > 1 year to 2 years (n = 3); e: > 2 years to 4 years (n = 4); f: > 4 years to 7 years (n = 3); g: > 7 years (n = 3).

δ = (Rm − Rp/Rp) ∗1000 where Rm and Rp are the isotope ratios for the test sample and international reference standard, respectively. Carbon-containing samples were referred to V-PDB (originally PDB, Pee Dee Belemnite) (Coplen, 1995). Nitrogen-containing samples were referred to the ratio in air (AIR). The isotopic (δ15N and δ13C) mean values are compared in 7 age phase groups according to the results of the Shapiro-Wilk test on data distribution (P > 0.05). The Levene's test was calculated to compare the equality of variances with the aim to apply the analysis of variance (one-way ANOVA). Because of the small sample sizes of the majority of age phase groups, which make it difficult to verify normality (although the Shapiro-Wilk normality test did not report statistical significance), the non-parametric Kruskal-Wallis test was also performed. We finally proceeded with the post-hoc Bonferroni test as Anova evidenced a significant difference among the groups in their nitrogen mean values. The significance level was set at P ≤ 0.05.

3.1. Stable isotopes Isotopic analyses were initially conducted in a sample of well-preserved ribs from 37 non-adults of different ages. Only the samples with collagen yield higher than 2% were considered in this study. Analyses were performed using the protocol described by Bocherens et al. (1991, 1997) and following the routine procedures of the Stable Isotope Biogeochemistry Laboratory of the Instituto Andaluz de Ciencias de la Tierra (CSIC, Granada, Spain). In a first stage, collagen was extracted from the bone using a chemical procedure that ensures the removal of inorganic components and other organic compounds that might contaminate the sample. In brief, samples were scraped clean and then powdered to a particle size of < 0.7 mm. Next, 300 mg of bone powder was decalcified in 1 M HCl for 20 min at room temperature, eliminating phosphates, fulvic acids and other soluble acids, and was then filtered through a 5-μm MF-Millipore filter, immersing the insoluble residue in 0.125 M NaOH for 20 h at room temperature. After rinsing with Milli-Q water, the neutralized sample was again filtered (5 μm), removing

4. Results 4.1. Bone isotopic analysis δ15N data were obtained from was excluded because insufficient between 7.1‰ and 12.9‰ (AIR), (AIR). δ13C data were obtained 32

36 non-adult samples after one rib collagen was extracted and ranged with a mean of 10.2‰ ( ± 1.5‰) from 35 non-adult samples (after

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

groups, we also included the isotopic mean values of the adult samples (21–40 years, n = 54) divided by sex (males n = 33 and females n = 21) and the mean value of a few animals buried in the necropolis, grouped in herbivore (1 ovicaprid and 2 cows, n = 3), horses (n = 2) and dogs (n = 2) (Laffranchi et al., 2016). Table 2 exhibits the isotopic values for each non-adult individual. Mean δ15N and δ13C values were compared among the seven age groups making use of a one-way ANOVA. The Snedecor's test evidenced a significant difference among the groups only for δ15N values (F = 6.934; P = 0.000) while no significant difference was observed for δ13C data (F = 1.023; P = 0.431). The post-hoc Bonferroni test showed that high statistical differences in δ15N values are present between the post-neonatal to 6 months group, the group of > 2 years to 4 years (P = 0.002) and the > 7 years group (P = 0.007). In the same way the group of > 6 months to 1 year individuals, and the follow age category (> 1 year to 2 years) also evidence significant differences with the same groups (> 2 years to 4 years group and > 7 years group; see the statistical results in Table 3). The non-parametric Kruskal-Wallis test was also performed leading to results similar to those obtained by ANOVA (χ2 = 20,642; P = 0.002). The corresponding multiple comparisons with the Bonferroni correction evidenced significant differences in δ15N composition between post-neonatal to 6 months group and the group of > 2 years to 4 years (P = 0.03), and between the group of > 6 months to 1 year and the group of > 2 years to 4 years (P = 0.05), coinciding in part with the results obtained with the ANOVA test (Table 3⁎). Table 1 also shows the isotopic data from the adult (21–40 years) sample (n = 54) and the animals (n = 7) of the studied necropolis. The adults (males + females) δ15N values ranges from 6.9‰ to 10.2‰

Table 1 Minimum (Min), maximum (Max), mean (M) and standard deviation (SD) of the isotopic values (δ15N and δ13C in per mil [‰]) of human samples by age category and faunal samples by category. n: number of individuals. δ15N‰ (AIR)

Fullterm neonates Post-neon. to 6 months > 6 months to 1 year > 1 year to 2 years > 2 years to 4 years > 4 years to 7 years > 7 years Adult males Adult females Herbivores Horses Dogs

δ13C‰(V-PDB)

n

Min

Max

M

SD

n

Min

Max

M

SD

14

8.1

11.7

10.2

1.1

14

− 18.0

− 12.8

−15.1

1.9

4

11.2

12.7

11.7

0.7

4

− 19.4

− 14.1

−16.8

2.2

5

10.2

12.3

11.4

0.9

4

− 17.7

− 13.3

−14.9

1.9

3

9.9

12.9

11.5

1.6

3

− 18.0

− 9.7

−14

4.2

4

8.0

8.4

8.3

0.2

4

− 18.0

− 14.4

−16.4

1.6

3

7.7

10.9

9.4

1.6

3

− 20.2

− 11.6

−16.2

4.3

3 33 21 3 2 2

7.1 7.8 6.9 3.8 3.9 4.9

9.8 10.4 9.9 5.6 5.9 7.1

8.4 9.0 8.4 4.5 4.9 6.0

1.3 0.6 0.7 0.9 1.4 1.6

3 33 21 3 2 2

− 16.2 − 20.2 − 20.0 − 20.3 − 20.8 − 13.7

− 11.2 − 12.1 − 11.6 − 17.2 − 19.4 − 13.2

−13.1 −15.8 −14.4 −18.8 −20.1 −13.4

2.7 2.1 1.9 1.3 0.9 0.3

another sample gave insufficient collagen) and ranged between − 20.2‰ and − 9.7‰ (V-PDB), with a mean of −15.3‰ ( ± 2.5‰) (V-PDB). The collagen yield was between 0 and 15% and the atomic C:N ratio of the samples falls within the range of 2.9–3.6 recommended by DeNiro (1985). Table 1 shows the isotopic mean values by age

Table 2 Results of stable isotope analysis of bone collagen from the non-adult sample (Laffranchi et al., 2016), ⁎a: fullterm neonates; b: post-neonates to 6 months; c: > 6 months to 1 year; d: > 1 year to 2 years; e: > 2 year to 4; f: > 4 years to 7 years; g: > 7 years. P: presence of non-specific stress markers. w: weeks; m: months; yrs: years. Sample

Age

Category⁎

δ15N‰ (AIR)

δ13C‰ (V-PDB)

C/N

%C

%N

Coll. yield

VRSV-57 VRSV-61 VRSV-64 VRSV-67 VRSV-68 VRSV-75 VRSV-76 VRSV-77 VRSV-79 VRSV-80 VRSV-81 VRSV-82 VRSV-86 VRSV-89 VRSV-55 VRSV-72 VRSV-63 VRSV-88 VRSV-70 VRSV-71 VRSV-85 VRSV-84 VRSV-58 VRSV-56 VRSV-83 VRSV-60 VRSV-59 VRSV-66 VRSV-87P VRSV-91 VRSV-54P VRSV-62 VRSV-65P VRSV-69 VRSV-74P VRSV-90P

40 w 39–42 w 40 w 38–40 w ± 39 w 40 w 38–40 w 38–40 w 38–40 w 40 w 38–40 w 40 w 40 w 38–40 w 1–3 m 2–3 m 3–6 m 3–6 m 7–9 m 7–9 m 7–11 m 8–16 m ± 1 yr 1 yr ± 4 m 18 m ± 6 m 2–3 yrs 3–4 yrs 3–4 yrs 3–4 yrs 3–5 yrs 4–5 yrs 5–6 yrs 6–7 yrs 7–9 yrs 10–12 yrs 17–19 yrs

a a a a a a a a a a a a a a b b b b c c c c c d d d e e e e f f f g g g

11.1 8.8 9.4 10.9 11 9.8 10.1 10.2 9.1 11.3 11.2 10 11.7 8.1 11.2 11.2 11.6 12.7 10.2 10.7 12.3 12.1 11.5 12.9 11.6 9.8 8.3 8 8.3 8.4 7.7 9.7 10.9 8.3 7.1 9.8

− 16.5 − 12.8 − 12.9 − 16.8 − 13.9 − 14.2 − 13.3 − 17.5 − 14.5 − 18 − 17 − 14.3 − 16.8 − 12.9 − 16.0 − 14.1 − 17.7 − 19.4 – − 14.3 − 17.7 − 14.4 − 13.3 − 18.0 − 9.7 − 14.3 − 18 − 17.4 − 15.7 − 14.4 − 16.9 − 20.2 − 11.6 − 11.2 − 11.9 − 16.2

3.1 2.9 2.9 3.2 2.9 3.3 3.2 3.1 2.9 3.2 3.3 3.1 3.1 3.3 3.2 3.2 3.2 3.3 – 2.9 3.1 3.2 3.1 3.1 3.3 3.2 3.0 2.9 3.2 3.2 3.1 3.2 2.9 3.2 3.2 3.3

41.8 40.3 40.1 42.0 39.8 42.6 42.1 41.7 40.4 41.8 42.1 41.9 41.6 42.6 42.1 42.6 42.2 42.2 – 40.7 41.7 42.1 41.9 41.6 42.4 42.3 41.8 39.0 42.2 42.2 41.6 42.0 40.5 42.6 41.8 42.3

15.8 16.6 17.2 15.5 16.1 14.8 15.2 15.5 16.8 15.4 15.0 15.6 15.6 15.3 15.5 15.4 15.2 15.0 – 16.8 15.4 15.2 15.5 15.7 15.1 15.6 16.3 17.0 15.6 15.3 15.5 15.5 17.4 15.6 15.2 14.9

7.5 8.3 5.2 9.1 7.3 5.4 5.2 8.6 9.7 6.7 6.5 10.2 8.9 4.8 13.6 11.6 14.1 8.7 2.2 6.7 5.5 13.7 15.3 12.8 9.5 15.5 8.8 9.1 7.7 14.3 6.8 8.7 8.0 13.4 13.6 9.4

33

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

Table 3 Results of post-hoc Bonferroni test. We only report the comparisons between the age groups that evidenced significant differences for δ15N values. (M): mean value. ⁎Age group comparison which shows significant differences after the use of non-parametric Kruskal-Wallis and the Bonferroni correction. Dependent variable (M) δ15N‰

Bonferroni

(I) Age category

(J) Age category

Mean difference (I–J)

P

Post-neonates–6 months

> 2 years–4 years > 7 years > 2 years–4 years > 7 years > 2 years–4 years > 7 years

3.43000 3.30667 3.12400 3.00067 3.23000 3.10667

0.002 (0.03⁎) 0.007 0.003 (0.05⁎) 0.013 0.009 0.027

> 6 months–1 year > 1 year–2 years

(AIR) with a mean of 8.8‰ ( ± 0.7‰), while δ13C values range from − 20.2‰ to −11.6‰ (V-PDB) with a mean of − 15.3‰ ( ± 2.1‰). The isotopic values of the animals are grouped into three categories: herbivores, horses and dogs. The herbivores show a δ15N mean value of 4.5 ± 0.9‰ (AIR) and a δ13C mean value of −18.8 ± 1.3‰ (VPDB), the horses show a δ15N mean value of 4.9 ± 1.4‰ (AIR) and a δ13C mean value of −20.1 ± 0.4‰ (V-PDB), while the dog group presents a δ15N mean value of 6.0 ± 1.6‰ (AIR) and a δ13C mean value of − 13.4 ± 1.3‰ (V-PDB).

statistical results of the post-hoc test, there are significant differences even between group d (> 1 year to 2 years) and group e (> 2 years to 4 years). This fact could help us to speculate about possible prolonged breast-milk consumption (at least up to two years old) integrated with complementary food. These observations are completely in line with the conclusions of some authors mentioned before (Fuller et al., 2006b; Killgrove and Tykot, 2013) and the reported recommendations from ancient medical sources. Analysis of mean non-adult isotopic values by age phase category (a–g) shows that mean δ15N values are more enriched in post-neonatal infants to 6 months (Fig. 3), while the mean δ13C values are slightly more negative at this age in comparison to the adult females (21–40 years). If we assume that many of the adult females in this necropolis possibly were the mothers of the children buried there, isotopic variations in breastfed infants can reasonably be expected to reflect those in their mothers (Fuller et al., 2006a; Katzenberg et al., 1996). As depicted in Fig. 3, mean δ15N values are 1.8‰ higher in the fullterm/neonates and 3.3‰ higher in the post-neonatal to 6 months individuals than in the adult females (8.4‰ ± 0.7 AIR; Laffranchi, 2015). Comparison with the mean δ15N value of 10.2‰ ± 1.5 (AIR) in the present nonadults indicates an enrichment of 1.8‰ with respect to the women. Among the adult females, δ15N values range between 6.9‰ and 9.9‰ (AIR), while δ13C values range between − 20‰ and − 11.6‰ (V-PDB). This wide variation in intra-site δ15N and δ13C values limits our ability to accurately determine the feeding status of the infants. For instance, the δ15N value in the breastfeeding child of a woman with lower values

5. Discussion 5.1. Isotopic evidence for infant feeding practices in the Verona pre-Roman sample Fig. 2 shows the trend of δ15N values across the seven age phases (a–g) proposed in this study. We can observe that there is a drop of the δ15N values between the age phases c (> 6 months to 1 year), d (> 1 year to 2 years) with nearly equal δ15N mean values (≈11.5‰ AIR) and the age phase e (> 2 years to 4 years). This could coincide with the Seller's stage ii: introduction of non-breast milk complementary foods or the start of the weaning process as suggested in Reynard and Tuross (2015). We also point out that the nitrogen values returned to the adult baseline during age phase g (> 7 years old) (δ15N mean value of 8.4‰ AIR). This would coincide with the Seller's stage iv: cessation of breastfeeding and the exclusive consumption of an adultlike diet or the end of the weaning process. If we observe Fig. 2 and the

Fig. 2. Graph showing trend in δ15N values in the 7 age phases proposed above. Note that δ15N values increase between age phase a and b (stage i: exclusive breastfeeding) and then start to decline first between the age phase b and c (stage ii) and then, there is a second major decline coinciding with the age phase e (stage iii).

34

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

Fig. 3. Mean δ15N (AIR) and δ13C (V-PDB) values (in ‰) for the different age categories (a–g) of non-adults and for adult females (potential mothers). Blue points represent δ15N mean values: a higher trophic level is indicated in the group of post‑neonatal to 6 months age individuals, whereas the results in the older infant group (> 7 years old) point to a similar diet to that of the female adults. The red points correspond to δ13C mean values: an increase to less negative carbon values is recorded in the g phase category, indicating that their diet was probably characterized by the consumption of cereals from C4 plants or of animal protein from herbivores consuming C4 plants. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

than those of other female adults (e.g., VRSV-1 with δ15N of 6.9‰ AIR) could be expected to be lower than those of other breastfed children. It would therefore not be correct to attribute this infant's reduced δ15N value to the absence of breastfeeding or a fully-weaned state. For this reason, it is essential to take into account the variation in δ15N values among associated adult females (Beaumont et al., 2015; Waters-Rist and Katzenberg, 2010). Most neonates from Verona display δ15N values above or around 10‰ (AIR) and are generally higher than the maternal values, suggesting that they were likely to have been breastfeeding at the time of their death. However, three of the neonates (VRSV-61, VRSV-64, and VRSV-79) show similar nitrogen values to those in the adult females (< ≈10‰ AIR), suggesting that they may possibly have died at birth, or shortly after birth, and they still reflect the intrauterine values. Prowse et al. (2008) proposed that low δ15N values in neonates, similar to those of adults in the sample, indicate their death at birth before breastfeeding could be recorded in the bone collagen (Katzenberg and Pfeiffer, 1995). Eight of the non-adult individuals (estimated age between postneonatal and 1 year) (Table 2 and Fig. 5) show among the highest δ15N values (> 10‰ AIR), around 2–3‰ above those of the adult women, which may suggest that they were being breastfed at the time of death. This is in line with observations made by Fuller et al. (2006a, 2006b). Four individuals (VRSV-54, VRSV-59, VRSV-66 and VRSV-87) have not only depleted δ15N values (7.7–8.3‰ AIR) but also more negative δ13C values (− 18‰ and − 15.7‰ V-PDB), suggesting that they may have been weaned for some time, consistent with their age (3 to 5 years old). It is conceivable that the 2- to 3-year-old infant VRSV-60, who shows a slightly higher δ15N value (9.8‰ AIR), may have just started Seller's stage iii (introduction of complementary and family foods) at the time of death. Finally, we highlight that individual VRSV-62 (5–6 years old) has an enriched δ15N value for his/her age (9.7‰ AIR) and the most negative δ13C value (−20.2‰ V-PDB) among all of the non-adults studied. Osteological analysis yielded no pathological evidence, and it is not possible to verify the presence/absence of enamel hypoplasia in his/her teeth because only part of the parietal of the cranium is preserved. One possible hypothesis is that VRSV-62 might have died from a fast-acting infection that left no traces in his/her bones and that the enriched δ15N value might result from food supplementation given to improve his/her health condition. Alternatively, as noted above, this child's growth and

the contribution of new collagen may have been slowed by disease, resulting in a preponderance of this signal prior to weaning (Fogel et al., 1989, 1997). Moreover δ15N values can also increase when an individual enters a state of negative nitrogen balance, when the amount of nitrogen excreted from the body is larger than the ingested amount of nitrogen. This can happen with generalized starvation, protein malnutrition, and even severe disease bouts if the individual is not able to eat or keep down food (Fuller et al., 2005; Long et al., 1979). Considering that we are analysing the individuals who died, not those who survived, it is possible that some of them were very sick and consequently their bodies began breaking down their own tissues to obtain the needed nitrogen causing duplication of the processes that cause the trophic level shift and hence enriched δ15N values (D'Ortenzio et al., 2015; Fuller et al., 2005). It is difficult to explain the negative δ13C value obtained, which is very close to that of the adult female VRSV-19 (− 20‰ V-PDB), whose negative δ13C value is also highly similar to those observed in some male adults (≈− 20‰ V-PDB). Given that the mean δ13C value of herbivores from the necropolis is − 18.8‰ ± 1.3 (V-PDB) the negative δ13C values of these individuals may be related to the consumption of meat or by-products from herbivores fed on C3 photosynthetic plants, or alternatively they may proceed from other places. In order to shed light on the origin of these individuals other isotopic analysis (δ18O in bone phosphate and δD in bone collagen) are currently underway. Individuals VRSV-74 and VRSV-54 show the lowest δ15N values and VRSV-83, VRSV-69 and VRSV-74 the least negative δ13C values (see Fig. 4). Individual VRSV-65 (6–7 years old) shows a δ15N value (10.9‰ AIR) (see Table 2) that is elevated for his/her age and higher than in the other two individuals of the same age phase (7.7‰ and 9.7‰ AIR).The high δ15N value may reflect a change in nitrogen metabolism produced by the effects of chronic stress or possibly disease (D'Ortenzio et al., 2015). Indeed, the multiple enamel hypoplasia bands observed in this child are compatible with stress episodes between the age of 1 and nearly 5 years. It is also possible that the presence of some disease might have slowed the growth of the child and the intake of new collagen, which may have extended the period of pre-weaning δ15N enrichment (Fogel et al., 1989, 1997; Katzenberg and Lovell, 1999). On the other hand, given that this individual (Fig. 5) was buried with grave goods that may indicate a certain social status, it is also possible that his/her poor health condition may have led the infant to be given food rich in animal protein (Cheung et al., 2012; Le Huray and Schutkowski,

35

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

Fig. 4. Scatter plot of δ15N and δ13C data for non-adults from Verona. Individuals affected by non-specific stress markers (e.g. cribra orbitalia and enamel hypoplasia) are indicated in the chart. Mean values of adult and animal's groups are also reported.

adults have very low δ15N levels; VRSV-69 (7–9 years old) displays growth delay, while VRSV-74 (10–12 years old) has enamel hypoplasia bands in his permanent teeth. Furthermore, two younger individuals also show non-specific stress markers, they are VRSV-54 (4–5 years old) and VRSV-87 (3–4 years old), who both evidence enamel hypoplasia and active cribra orbitalia. In both cases, hypoplastic defects indicate a stress episode very close to the time of their death. The sixth non-adult with non-specific stress marker (enamel hypoplasia) is VRSV-90 (17–19 years old), whose isotopic values are similar to those of adult females at the site, although with a slightly higher δ15N value. Cribra orbitalia has been associated in the anthropological literature with anaemic disorders (iron malabsorption by the intestine), malnutrition, infectious conditions, and vitamin deficiencies (Ortner et al., 2001). The main causes of iron malabsorption include the presence of intestinal parasites, infectious disease (e.g. diarrhoea), or phytates in food (cereals) (Jiménez-Brobeil et al., 2008; Ortner et al., 2001). Walker et al. (2009) proposed that porotic hyperostosis and cribra orbitalia lesions can result from megaloblastic anaemia in nursing infants due to a combination of depleted maternal vitamin B12 reserves and unsanitary living conditions, which can increase nutrient losses from gastrointestinal infections around the time of weaning (Walker et al., 2009). Radiographic and anatomical studies associate cribra crania and cribra orbitalia with severe forms of hereditary haemolytic anaemia

2005; Moghaddam et al., 2016; Oelze et al., 2012; Pearson et al., 2013). In Fig. 4 the isotopic values from animal bones (herbivore, horses and dogs) are plotted with the human values. Domesticate herbivores and horses generally show a diet mainly based on C3 plants (δ13C mean value respectively of − 18.8 ± 1.3‰ and − 20.1 ± 0.9 V-PDB), while the dogs show relatively high δ13C values (≈−13‰ V-PDB) and present low δ15N values (respectively 4.9‰ and 7.1‰ AIR). Considering that dogs usually eat men's discards, this might imply a significant proportion of human food scraps based on C4 plants in their diet. These results indicate a mixed terrestrial diet (meat or dairy products and vegetal foodstuffs) for humans and, in particular the δ13C delta values point out that the vast majority of individuals (adults and non-adults) seem to include in their diet C4 plants in a direct way (or indirectly through the consumption of herbivores that fed on them) (Laffranchi et al., 2016). 5.2. Stable isotopes and non-specific stress markers Enamel hypoplasia and/or cribra orbitalia was detected in six nonadults of the sample analyzed for stable isotopes. One of these individuals (VRSV-65, see above) shows several bands of enamel hypoplasia and a possible growth delay (dental age of 7 years ± 24 months vs. 4.5–5 years old estimated from long bone length). Others two non-

Fig. 5. Tomb of individual VRSV-65 with associated grave goods (Photo S. Thompson).

36

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

advice in the interpretation of one case and to the anonymous reviewers for their valuable insights and suggestions.

such as thalassemia major and sicklemia (Hershkovitz et al., 1997; Walker et al., 2009). Several zones of Italy (e.g. Sardinia's island, the area around Roma and the Po river delta) show a great incidence of thalassemic genes, which are considered as an adaptive response to malarial environments (Silvestroni and Bianco, 1975; Ricci et al., 1997; Manzon and Gualdi-Russo, 2016). We cannot surely affirm that this was relevant for the Verona area, because the spread of malaria vectors seems to have extended later in Northern Italy respect to the South (Manzon and Gualdi-Russo, 2016; Sallares et al., 2004). In the present study, the low δ15N values in the non-adults with any of non-specific stress markers may be attributable to a low amount of animal protein in their cereal-based diet, but may also indicate a severely malnourished state and increased risk of infectious diseases. In fact, the frequency of cribra orbitalia is slightly higher in the non-adults than in the adults, which may be related to an increased risk of infection in children during vulnerable periods such as weaning (Fogel et al., 1989; Katzenberg et al., 1996). The frequency of cribra orbitalia is also slightly, but significantly, higher in the females (12.5%) than in the males (11.43%) (Laffranchi, 2015).

Funding The isotopic analysis were supported by the Projects RNM 8011 (Junta de Andalucía, Spain), HAR 2016-75788 (Ministerio de Economía y Competitividad, Spain) and the research group RNM-309 (Junta de Andalucía, Spain). References Beaumont, J., Montgomery, J., Buckberry, J., Jay, M., 2015. Infant mortality and isotopic complexity: new approaches to stress, maternal health, and weaning. Am. J. Phys. Anthropol. 157, 441–457. Bocherens, H., Fizet, M., Mariotti, A., Lange-Badre, B., Vandermersch, B., Borel, J.P., Bellon, G., 1991. Isotopic biogeochemistry (13C, 15N) of fossil vertebrate collagen: application to the study of a past food web including Neanderthal man. J. Hum. Evol. 20, 481–492. Bocherens, H., Biliou, D., Patou-Mathis, M., Bonjean, D., Otte, M., Mariotti, A., 1997. Paleobiological implications of the isotopic signatures (13C, 15N) of fossil mammal collagen in Scladina cave (Sclayn, Belgium). Quat. Res. 48, 370–380. Burt, N.M., 2015. Individual dietary patterns during childhood: an archaeological application of a stable isotope microsampling method for tooth dentin. J. Archaeol. Sci. 53, 277–290. Burt, N.M., Garvie-Lok, S., 2013. A new method of dentin microsampling of deciduous teeth for stable isotope ratio analysis. J. Archaeol. Sci. 40, 3854–3864. Capitanio, M.A., 1989. Esame antropologico degli inumati di Valeggio sul Mincio (Verona) d'epoca Romana (I sec a.C.- I sec. d.C.). In: Atti e Memorie dell'Accademia di Agricoltura Scienze e Lettere di Verona. Serie VI XXXVIII. pp. 159–198. Cavalieri Manasse, G., 2014. La necropoli del Seminario Maggiore di Verona. In: Marinetti, A., Prosdocimi, A.L. (Eds.), Rivista di epigrafia italica, Studi Etruschi LXXVII. Bretschneider Editore, pp. 373–375. Cheung, C., Schroeder, H., Hedges, R.E.M., 2012. Diet, social differentiation and cultural change in Roman Britain: new isotopic evidence from Gloucestershire. Archaeol. Anthropol. Sci. 4, 61–73. Coplen, T.B., 1995. New IUPAC guidelines for the reporting of stable hydrogen, carbon, and oxygen isotope-ratio data. J. Res. Natl. Inst. Stand. Technol. 100, 285. Corrain, C., 1987. I resti scheletrici umani della Necropoli gallo-romana (II-I sec a.c.), scavati in località Casalandri (Isola Rizza, Verona). In: Quaderni di scienze antropologiche. 13. pp. 21–64. DeNiro, M.J., 1985. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317, 806–809. DeNiro, M.J., Epstein, S., 1981. Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta 45, 341–351. D'Ortenzio, L., Brickley, M., Schwarcz, H., Prowse, T., 2015. You are not what you eat during physiological stress: isotopic evaluation of human air. Am. J. Phys. Anthropol. 157, 374–388. Dupras, T.L., Tocheri, M.W., 2007. Reconstructing infant weaning histories at Roman period Kellis, Egypt using stable isotope analysis of dentition. Am. J. Phys. Anthropol. 134, 63–74. Dupras, T.L., Schwarcz, H.P., Fairgrieve, S.I., 2001. Infant feeding and weaning practices in Roman Egypt. Am. J. Phys. Anthropol. 115, 204–211. Fazekas, I.G.Y., Kósa, F., 1978. Forensic Fetal Osteology. Akadémiai Kiadó, Budapest. Fogel, M.L., Tuross, N., Owsley, D.W., 1989. Nitrogen isotope tracers of human lactation in modern and archaeological populations. In: Annual Report of Geophysical Laboratory Carnegie Institution of Washington, pp. 111–117. Fogel, M.L., Tuross, N., Johnson, B.J., Miller, G.H., 1997. Biogeochemical record of ancient humans. Org. Geochem. 27, 275–287. Fuller, B.T., Fuller, J.L., Sage, N.E., 2005. Nitrogen balance and δ15N: why you're not what you eat during nutritional stress. Rapid Commun. Mass Spectrom. 19, 2497–2506. Fuller, B.T., Fuller, J.L., Harris, D.A., Hedges, R.E.M., 2006a. Detection of breastfeeding and weaning in modern human infants with carbon and nitrogen stable isotope ratios. Am. J. Phys. Anthropol. 129, 279–293. Fuller, B.T., Molleson, T.I., Harris, D.A., Gilmour, L.T., Hedges, R.E.M., 2006b. Isotopic evidence for breastfeeding and possible adult dietary differences from Late/SubRoman Britain. Am. J. Phys. Anthropol. 129, 45–54. Grassi, M.T., 1995. La romanizzazione degli Insubri: Celti e Romani in Traspadana attraverso la documentazione storica ed archeologica. Collana di Studi di Archeologia Lombarda, Milano. Grassi, M.T., 2009. I Celti in Italia. Biblioteca di Archeologia. 16 Longanesi, Milano. Hedges, R.E.M., Reynard, L.M., 2007. Nitrogen isotopes and the trophic level of humans in archeology. J. Archaeol. Sci. 34, 1240–1251. Herring, D.A., Saunders, S.R., Katzenberg, M.A., 1998. Investigating the weaning process in past populations. Am. J. Phys. Anthropol. 105, 425–440. Herrscher, E., 2013. Détection isotopique des modalités d'allaitement et de sevrage à partir des ossements archéologiques. Cah. Nutr. Diét. 48 (2), 75–85. Hershkovitz, I., Rothschild, B.M., Latimer, B., Dutour, O., Leonetti, G., Greenwald, C.M., Rothschild, C., Jellema, L.M., 1997. Recognition of sickle cell anaemia in skeletal remains of children. Am. J. Phys. Anthropol. 104, 213–226. Jay, M., Fuller, B.T., Richards, M.P., Knüsel, C.J., King, S.S., 2008. Iron Age breastfeeding

6. Conclusions The pre-Roman/Celtic sample from Verona provides an exceptional opportunity for in-depth studies of bioarchaeological aspects in a nonadult population. Thus, in the present study, examination of δ15N and δ13C stable isotope values from rib bone collagen offered an effective approach for exploring the different phases of the infant feeding practices and interpreting the health status of this ancient population during childhood. The trend of δ15N data and the statistical results show that the nitrogen values increase between the birth and the age of six months (stage i: exclusive breastfeeding). After that they slightly decrease to the age of one year, and then they remain stable until the age of two years (stage ii: breastfeeding plus specific food for babies). Finally, after the age of two years, they experiment a bigger decline (stage iii: breastfeeding plus complementary and family foods).The significant δ15N value difference found between the non-adult group with age of 1 to 2 years old and the following group with older age (> 2 years to 4 years) may suggest a possible prolonged breast-milk consumption surely integrated with other complementary food till the age of two years old. The Verona data clearly show a δ15N and a δ13C trophic level effect of breastfeeding, with an enrichment of 1.8‰ (for fullterm/neonatal) and of 3.3‰ (for post-neonatal to 6 months) in mean δ15N value versus adult females considered as potential mothers in the necropolis. In the present study, stable isotope values were analyzed in rib samples, and further investigation is warranted in tooth dentin samples in order to verify our results. Application of the stable isotope microsampling method for dentin (Burt, 2015; Burt and Garvie-Lok, 2013), as also recommended by Reynard and Tuross (2015) and by Beaumont et al. (2015), could help to further elucidate the relationship between neonatal and maternal isotopic values and between diet and health. Acknowledgements The authors thank Dr. Giuliana Cavalieri Manasse of the Soprintendenza Archeologica of Verona for the opportunity to study this skeletal collection and the archaeologists Simon Thompson and Marzia Bersani of Verona (Thompson Simon scavi e rilevamenti archeologici, Verona, Italy), who excavated the necropolis of Seminario Vescovile, for use of their archaeological data. They wish to express their gratitude to Prof. Jose Antonio Riquelme Cantal (University of Cordova, Spain) for help with the collagen extraction process and Richard Davies and Dr. Alexander E.S. Van Driessche (ISTerre Grenoble, France) for assistance with the English version of the manuscript. The authors are also grateful to Dr. Emma Lightfoot (University of Cambridge, UK) for her 37

Journal of Archaeological Science: Reports 17 (2018) 30–38

Z. Laffranchi et al.

diet according to burial practice and sex in the early Neolithic. J. Anthropol. Archaeol. 32, 180–189. Prowse, T.L., Saunders, S.L., Schwarcz, H.P., Garnsey, P., Macchiarelli, R., Bondioli, L., 2008. Isotopic and dental evidence for infant and young child feeding practices in an Imperial Roman skeletal sample. Am. J. Phys. Anthropol. 137, 294–308. Redfern, R.C., Millard, A.R., Hamlin, C., 2012. A regional investigation of subadult dietary patterns and health in late Iron age and Roman Dorset, England. J. Archaeol. Sci. 39, 1249–1259. Redfield, A., 1970. A new aid to aging immature skeletons: development of the occipital bone. Am. J. Phys. Anthropol. 33, 207–220. Reynard, L.M., Tuross, N., 2015. The known, the unknown and the unknowable: weaning times from archaeological bones using nitrogen isotope ratios. J. Archaeol. Sci. 53, 618–625. Ricci, R., Mancinelli, D., Vargiu, R., Cucina, A., Santandrea, E., Capelli, A., Catalano, P., 1997. Pattern of porotic hyperostosis and quality of life in a II century AD farm near Rome. J. Anthropol. Sci. 75, 117–128. Rottoli, M., 2014. Crop diversity between Central Europe and the Mediterranean: aspects of northern Italian agriculture. In: Chevalier, A. (Ed.), Plants and People: Choices and Diversity Through Time. Oxbox Book, Oxford, pp. 75–81. Sallares, R., Bouwman, A., Anderung, C., 2004. The spread of malaria to Southern Europe in antiquity: new approaches to old problems. Med. Hist. 48, 311–328. Schaefer, M., Black, S., Scheuer, L., 2009. Juvenile Osteology. A Laboratory and Field Manual. Academic Press, London. Scheuer, L., Black, S., 2000. Developmental Juvenile Osteology. Academic Press, London. Schurr, M.R., 1998. Using stable nitrogen isotopes to study weaning behavior in past populations. World Archaeol. 30, 327–342. Sellen, D.W., 2009. Evolution of human lactation and complementary feeding: implications for understanding contemporary cross-cultural variation. In: Goldberg, G. (Ed.), Breastfeeding: Early Influences on Later Health. Springer, pp. 253–282. Silvestroni, E., Bianco, I., 1975. Screening for microcytemia in Italy: analysis of data collected in the past 30 years. Am. J. Hum. Genet. 27 (2), 198–212. Solinas, P., 2014. Iscrizioni in alfabeto leponzio dalla necropolis del Seminario Maggiore di Verona. In: Marinetti, A., Prosdocimi, A.L. (Eds.), Rivista di epigrafia italica, Studi Etruschi vol. LXXVII. Giorgio Bretschneider Editore, pp. 375–379. Teegen, W.R., 2014a. rime notizie sulla paleopatologia degli inumati della necropoli tardo celtica di Povegliano Veronese, loc. Ortaia (prov. Verona, Italia). In: Les Celtes et le Nord de l'Italie (Premier et Second Âges du fer). Actes du XXXVIe colloque international de l'AFEAF, Verona, 17–20 May 2012, 36e Supplément à la R.A.E., S.A.E. et A.F.E.A.F., Dijon, pp. 523–530. Teegen, W.R., 2014b. Mago o giocatore- L'individuo della tomba 4 della necropoli tardo celtica di Povegliano Veronese (Verona, Italia). In: Les Celtes et le Nord de l'Italie (Premier et Second Âges du fer). Actes du XXXVIe colloque international de l'AFEAF, Verona, 17–20 May 2012, 36e supplément à la R.A.E., S.A.E. et A.F.E.A.F., Dijon, pp. 531–534. Thompson S.M. and Bersani M. Relazione archeologica del Seminario Vescovile 20052009. Verona, unpublished report. Ubelaker, D.H., 1989. Human Skeletal Remains, 2nd ed. Taraxacum Press, Washington DC. Walker, P.L., Bathurst, R.R., Richman, R., Gjerdrum, T., Andrushko, V.A., 2009. Perspectives. The causes of porotic hyperostosis and cribra orbitalia: a reappraisal of the iron-deficiency-anemia hypothesis. Am. J. Phys. Anthropol. 139, 109–125. Waters-Rist, A.L., Katzenberg, M.A., 2010. The effect of growth on stable nitrogen isotope ratios in subadult bone collagen. Int. J. Osteoarchaeol. 20, 172–191. Waters-Rist, A.L., Bazaliiskii, V.I., Weber, A.W., Katzenberg, M.A., 2011. Infant and child diet in Neolithic hunter-fisher-gatherers from Cis-Baikal, Siberia: intra-long bone stable nitrogen and carbon isotope ratios. Am. J. Phys. Anthropol. 146 (2), 225–241. Wood, J.W., Milner, G.R., Harpending, H.C., Weiss, K.M., Cohen, M.N., Eisenberg, L.E., Hutchinson, D.L., Jankauskas, R., Cesnys, G., Česnys, G., Katzenberg, M.A., Lukacs, J.R., McGrath, J.W., Roth, E.A., Ubelaker, D.H., Wilkinson, R.G., 1992. The osteological paradox: problems of inferring prehistoric health from skeletal samples [and comments and reply]. Curr. Anthropol. 33 (4), 343–370.

practices in Britain: isotopic evidence from Wetwang Slack, East Yorkshire. Am. J. Phys. Anthropol. 136, 327–337. Jiménez-Brobeil, S.A., Al Oumaoui, I., Nájera, T., Molina, F., 2008. Salud y enfermedad en la Motilla del Azuer. Una población de la edad del bronce de la Mancha. R.E.A.F. 28. pp. 57–70. Katzenberg, M.A., Lovell, N.C., 1999. Stable isotope variation in pathological bone. Int. J. Osteoarchaeol. 9, 316–324. Katzenberg, M.A., Pfeiffer, S., 1995. Nitrogen isotope evidence for weaning. In: Grauer, A.L. (Ed.), Bodies of Evidence: Reconstructing History Through Skeletal Analysis. Wiley-Liss, New York, pp. 221–235. Katzenberg, M.A., Herring, D.A., Saunders, S.R., 1996. Weaning and infant mortality: evaluating the skeletal evidence. Yrbk. Phys. Anthropol. 39, 177–199. Kaupová, S., Herrscher, E., Velemínskỳ, P., Cabut, S., Poláček, L., Brůžek, J., 2014. Urban and rural infant-feeding practices and health in early medieval central Europe (9th–10th century, Czech Republic). Am. J. Phys. Anthropol. 155, 635–651. Killgrove, K., Tykot, R.H., 2013. Food for Rome: a stable isotope investigation of diet in the Imperial period (1st–3rd centuries AD). J. Anthropol. Archaeol. 32, 28–38. Knipper, C., Pichler, S.L., Rissanen, H., Stopp, B., Kühn, M., Spichtig, N., Röder, B., Schibler, J., Lassau, G., Alt, K.W., 2016. What is on the menu in a Celtic town? Iron Age diet reconstructed at Basel-Gasfabrik, Switzerland. Archaeol. Anthropol. Sci. http://dx.doi.org/10.1007/s12520-016-0362-8. Laffranchi, Z., 2015. Antropología de la población Pre-Romana de Verona (Italia). PhD Dissertation. Universidad de Granada. Laffranchi, Z., Martín-Flórez, J.S., Jiménez-Brobeil, S.A., Castellani, V., 2015. Foot polydactyly and bipartite medial cuneiform: a case of co-occurrence in a Celtic skeleton from Verona (Italy). Homo 66, 216–228. Laffranchi, Z., Delgado-Huertas, A., Jiménez-Brobeil, S.A., Granados-Torres, A., Riquelme-Cantal, J.A., 2016. Stable C & N isotopes in 2100 Year-B.P. human bone collagen indicate rare dietary dominance of C4 plants in NE-Italy. Sci Rep 6, 38817. http://dx.doi.org/10.1038/srep38817. Larsen, C.S., 2015. Bioarchaeology. Interpreting Behavior From the Human Skeleton, Second edition. Cambridge University Press, Cambridge. Le Huray, J.D., Schutkowski, H., 2005. Diet and social status during the La Tène period in Bohemia: carbon and nitrogen stable isotope analysis of bone collagen from Kutná Hora-Karlov and Radovesice. J. Anthropol. Archaeol. 24, 135–147. Long, C.L., Schaffel, N., Geiger, J.W., Schiller, W.R., Blakemore, W.S., 1979. Metabolic response to injury, illness: estimation of energy and protein needs from indirect calorimetry and nitrogen balance. J. Parenter. Enteral Nutr. 3, 452–456. Malnati, L., Salzani, L., Cavalieri Manasse, G., 2004. Verona la formazione della cittá. In: Agusta-Boularot, S., Lafon, X. (Eds.), Des Ibères aux Venetes. Collection de l'école française, Roma, pp. 347–377. Manzon, V., Gualdi-Russo, E., 2016. Health patterns of the Etruscan population (6th–3rd centuries BC) in Northern Italy: the case of Spina. Int. J. Osteoarchaeol. 26, 490–501. Moghaddam, N., Müller, F., Hafner, A., Lösch, S., 2016. Social stratigraphy in Late Iron Age Switzerland: stable carbon, nitrogen and sulphur isotope analysis of human remains from Münsingen. Archaeol. Anthropol. Sci. 8, 149–160. Nájera-Colino, T., Molina-Gonzalez, F., Jiménez-Brobeil, S.A., Sánchez-Romero, M., Al Oumaoui, I., Aranda-Jiménez, G., Delgado-Huertas, A., Laffranchi, Z., 2010. La población infantil de la Motilla del Azuer: Un estudio bioarqueológico. Complutum 21 (2), 69–102. Oelze, V.M., Koch, J.K., Kupke, K., Nehlich, O., Zäuner, S., Wahl, J., Weise, S.M., Rieckhoff, S., Richards, M.P., 2012. Multi-isotopic analysis reveals individual mobility and diet at the Early Iron Age monumental tumulus of Magdalenenberg, Germany. Am. J. Phys. Anthropol. 148, 406–421. Ortner, D.J., Butler, W., Cafarella, J., Milligan, L., 2001. Evidence of probable scurvy in subadults from archeological sites in North America. Am. J. Phys. Anthropol. 114, 343–351. Pearson, J.A., Hedges, R.E.M., Molleson, T.I., Özbek, M., 2010. Exploring the relationship between weaning and infant mortality: an isotope case study from Aşikli Höyuk and Çayönü Tepesi. Am. J. Phys. Anthropol. 143, 448–457. Pearson, J., Grove, M., Özbek, M., Hongo, H., 2013. Food and social complexity at Çayönü Tepesi, southeastern Anatolia: stable isotope evidence of differentiation in

38