Those who died very young-Inferences from δ15 N and δ13 C in bone collagen and the absence of a neonatal line in enamel related to the possible onset of breastfeeding

Am J Phys Anthropol. 2019 Aug;169(4):664-677. doi: 10.1002/ajpa.23847. Epub 2019 May 3.


Objectives: Stable isotope analysis has often been used in neonatal remains from archeological contexts to investigate the presence of a signal of breastfeeding and weaning in past populations. Tooth histology on the other hand might be used as an indicator of birth survival. This pilot study aimed to investigate the feasibility of using stable nitrogen (δ15 N) and carbon (δ13 C) isotope values from neonatal bone collagen to elucidate if values deviating from the adult female average could indicate breastfeeding and co-occur with the presence of a neonatal line (NNL). The combination of these independent indicators might be useful in clarifying the fate of individuals who died around birth.

Materials and methods: Bone collagen from 21 archeological human and animal specimens was extracted and analyzed via mass-spectrometry for δ15 N and δ13 C. A verification of the stable isotope results was undertaken using tooth histology on three individuals who were investigated for the presence of a NNL as an indicator of live birth and short survival.

Results: The biological age of the human samples varied between 8.5 lunar months (Lm) and 2 postnatal months (Pm) of age. All except one individual exhibited elevated δ15 N values compared to the female average. The histological analyses revealed no NNL for this and two further individuals (n = 3).

Discussion: The results indicate that elevated nitrogen values of very young infants relative to a female average in archeological contexts are not necessarily associated with a breastfeeding onset signal, and therefore cannot be used exclusively as a proxy of birth survival. The elevation might be possible due to various reasons; one could be nutritional, in particular maternal stress during pregnancy or a metabolic disorder of mother and/or her child. In those cases, the evaluation of a NNL might reveal a false breastfeeding signal as seen for two individuals in our sample who have elevated nitrogen values despite the fact no NNL could be observed. Overall, our data support the growing awareness that bone collagen δ15 N values of neonates/infants should not be used as a proxy for breastfeeding or birth survival on its own.

Keywords: breastfeeding; neonatal line; postnatal survival; stable isotopes; stillbirth.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Animals
  • Anthropology, Physical
  • Breast Feeding*
  • Carbon Isotopes / analysis*
  • Collagen* / analysis
  • Collagen* / chemistry
  • Dental Enamel / anatomy & histology*
  • Female
  • Femur / chemistry
  • Humans
  • Infant
  • Infant, Newborn
  • Models, Statistical
  • Nitrogen Isotopes / analysis*
  • Pregnancy
  • Stillbirth
  • Tooth / chemistry


  • Carbon Isotopes
  • Nitrogen Isotopes
  • Collagen