History in the Teeth

How 'reading' and studying the teeth of long-deceased people can shed light on past lives, deaths, and health conditions.

In the field of osteology, isotopes found in bones and teeth have been a valuable tool in studying a range of topics, including the diets of past populations. Over two decades ago, researchers found that an isotopic profile of infant bones and teeth could provide information about breastfeeding behaviors in the past. Knowing that the health of both mother and child during early childhood has a significant impact on the health of an individual through infancy and into adulthood, Dr. Julia Beaumont and colleagues set out to more closely examine the relationship between breastfeeding and physiology. In a study recently published in the American Journal of Physical Anthropology, they analyzed nitrogen and carbon isotope levels in the teeth of children from archaeological sites in Ireland, England, and Scotland, including two that were related to the Great Famine in Ireland between 1845 and 1852.

It has generally been accepted that nitrogen and carbon isotope levels recorded in the teeth of infants during development can show when a child was being breastfed, and when they were weaned through the introduction of local foods. By finding average nitrogen and carbon isotope levels for females from the same place and time period through the testing of bone collagen, researchers create a baseline against which to compare carbon and nitrogen levels from the teeth and bones of children. Thus, according to the commonly used weaning curve model, a picture of the child’s diet can be established: high nitrogen levels after birth indicate breastfeeding, the peak and gradual fall in nitrogen indicate the slow introduction of food, and the child’s nitrogen level approaches the adult female’s average at the time of full weaning. Carbon isotope levels should form a curve that echoes the nitrogen curve, as it provides information on the composition of an individual’s diet and changes as food replaces the mother as the source of nutrients. This weaning curve model has often been used to study the timing and duration of breastfeeding and weaning in archaeological contexts. However, Dr. Beaumont and her colleagues wondered if the relationship between nitrogen and carbon isotope levels and breastfeeding might be more complicated than previously thought.

Permanent and deciduous teeth of children from the Kilkenny Union Workhouse famine cemetery in Ireland, and the Lukin Street Catholic Mission cemetery in London were studied, representing children who died during the famine in Ireland, and those who died after their families moved from Ireland to London. The researchers also included teeth from a Bronze Age site in Scotland called High Pasture Cave, and previously published data from the Neolithic Scottish site of Sumburgh cist. Because teeth grow consistently throughout early human development, they can be cut into small sections that each represent several months of an individual’s life. Teeth were selected for use in this study based on the ages during which they were formed. Beaumont and colleagues chose teeth that would have been forming during the early stages of life, including prenatal, breastfeeding, and weaning periods.

The samples that were focused on most heavily in this study were those from the Irish workhouse and the Catholic mission house in London. Samples from these sites were especially useful because documentary and epigraphic evidence from coffin nameplates, census records, death certificates, and other sources add context to the data. The workhouse cemetery was opened during famine because a nearby cemetery was full, and people were dying at a rapid pace. During the time of famine, the Lukin Street cemetery served as a final resting place for many Irish people who had left Ireland for London, as evidenced by the Irish nameplates on the coffins. At the very least, it is known that the individuals buried in these two cemeteries died during a period of nutritional stress, the Great Famine – but in some cases, more specific information can be found. For instance, one of the people buried in the Lukin Street cemetery is Georgiana Neale. Documentary sources report that she died from Tuberculosis at the age of twenty-three while pregnant with her second child.

Each individual’s isotope profiles were created from one tooth, though the same tooth was not chosen from each individual due to the age range of the children represented in the study. Several first and second deciduous molars, which begin formation in utero, were analyzed. None of these molars had finished forming, indicating that all of the children represented by these teeth died before the age of approximately two and a half (first molars) or approximately three and a half (second molars). First permanent molars, also beginning formation before birth, were included. Some were fully formed, indicating the children lived past approximately nine and a half years, and some were incomplete, indicating death before that age. Lastly, two second permanent molars, which form between two and a half to fifteen and a half years old, were also included. Neither tooth was completely formed, indicating that the children died sometime before the age of fifteen and a half.

According to the commonly accepted model for interpreting breastfeeding behaviors with isotopes, the isotopic profiles of the children’s teeth should have followed the weaning curve. However, what Beaumont and colleagues found was far more varied and much less simple. When compared to the female average isotope values, many of the isotope profiles did not match the expected weaning curve model. Divergences were seen in the form of flat profiles, fluctuating or consistently declining profiles, and nitrogen and carbon profiles that didn’t agree. They especially noticed profile differences between those children that survived into later childhood, and those that died very young.

Many suggestions for interpreting the results and their deviation from the weaning curve were proposed by the team. Beaumont suggests that starving mothers struggling to feed starving babies in a time of famine could affect isotopic profiles, since nitrogen isotope levels are affected not only by diet, but also by periods of physiological stress. Interestingly, it is not only the famine cemeteries that exhibit evidence for periods of nutritional deficiency. Similar profiles were found in teeth from other sites included in the study, which had no connection to the famine. Still, the researchers stress that their results clearly indicate that the relationship between isotope levels and breastfeeding behaviors is not as straightforward as previously suggested by prior studies, that the weaning curve model is overly simplistic, and that future study should approach the relationship with more nuance.

This is a significant study for several reasons. Science is always seeking to better itself by testing existing hypotheses and incorporating new information, to create the best investigative methods possible at any given time. This study challenges the accepted model for examining breastfeeding behaviors in past populations and offers new information to increase the accuracy of future studies. Additionally, the new information offered by this research opens fresh pathways for investigating not only breastfeeding behaviors, but maternal and infant diet and health and their significance in studying infant mortality, and future health and mortality in individuals later in life. Lastly, such research can have interdisciplinary applications. The results of this study suggest that a lot can be learned about health and diet during the critical early stages of life – which continue to affect an individual’s health into adulthood – through the examination of naturally-shed deciduous teeth. Dr. Beaumont hopes that the information left behind by these past tragedies might prove a valuable resource in current medical efforts to improve the lives of modern children.

Cover Image, Top Left: Emailshankar, Wikimedia Commons


Allison Hummel received her Bachelor’s degree in Archaeology from Dickinson College in 2013, and her Master’s degree in Human Osteology and Palaeopathology from the University of Bradford in 2014. Her main areas of interest within archaeology and osteology are trauma and disease. She has worked on excavations in Greece and Romania.