This rule also applies at the atomic level: your body’s atoms come from your food and drink. Atoms of almost all the chemical elements (carbon, oxygen, hydrogen, etc.) have more than one possible atomic weight. These different atoms of the same element are called isotopes. For example oxygen (O) has three naturally occurring isotopes: O-16, O-17, and O-18 (16O, etc.) These isotopes never disintegrate; we call them stable isotopes to distinguish them from the better known radioactive isotopes used in medical treatment.

The relative abundances of the stable isotopes of an element are almost the same in all samples containing that element. However, we can detect tiny variations which are the result of natural processes. For example, the 18O/16O ratio in rain and snow varies by up to 3% depending on the location where it fell. We can determine 18O/16O with a precision of better than one part in 10,000, using a stable isotope ratio mass spectrometer [SIRMS] (Figure 1). These analyses give the degree to which a sample is enriched or depleted in 18O compared to the world standard (seawater); the analyses are called delta 18O ( 18O) values. In rain they always tend to be negative numbers, because rain is always depletedin 18O compared to sea-water.

Figure 1: Stable isotope ratio mass spectrometer


Most of the O atoms in our body come from the water we drink, and is usually isotopically like the precipitation where we live. Therefore we can often learn where a person lived from the isotopic composition of their teeth and bones. Fortunately we now have maps showing the distribution of 18O/16O ratios in precipitation falling over North America and Europe (Figure 2) which we can use to help us to trace the place of origin of a murder victim. Even burned remains can be analyzed this way.

Figure 2: Oxygen isotope map of North America.

Actually, by analyzing both teeth and bones of a victim, we can learn two kindsof geographic information about them. Since most teeth are formed either beforebirth or during childhood 18O/16Oof teeth records where the person was livingat that time. On the other hand, the atoms of bone are constantly being changedthroughout a person’slifetime through remodeling. As a result, their isotopes record the location(or locations) where the person was living over the last 10 to 20 years of theirlife. Remodeling slows down in later life so the bones of older people recordtheir whereabouts somewhat earlier in their life.

We can analyze these isotopic signals in our lab by drilling out a small sample of tooth or bone (a couple milligrams) and analyzing the oxygen from the sample. At the same time, we also analyze the carbon-13 (13C) content of the diet. This tells us something about what kinds of food the person ate, a subjectthat I will cover in a later article.

From the 18O value of a sample of bone we can use equations developed by geochemists to determine the most likely 18O value of the drinking water consumed by the person. If we assume that this was also close to the average 18O value of precipitation where the person lived, then we can use maps like Figure 2 to determine possible areas where the person came from. You can see, however, that the areas over which precipitation has a particular 18O value can be quite large. In general, 18O decreases from the equator to the poles, and from the west coast of a continent to the interior. Also, 18O is lower in mountains, and on the lee-side of large mountain ranges like the Rockies.

From the map you can see that a person could acquire particular 18O value while living over a wide geographic region. Nevertheless, 18O values of teeth and bones can be a powerful tool in learning something about the identity of a person, especially when combined with other data.

First of all, when human remains are found at certain locations, we’d like to know if the person was from nearby that spot. The most precise test of this is to compare the 18O of their bone with that of a tooth known to be from the same neighborhood, usually from the “Tooth Fairy.” When the 18O of the victim’s bones or teeth are clearly different from local values, we look for possible locations where the person may have lived through various times of their life (infancy, maturity, last location) by matching the calculated value for 18O of precipitation with possible map locations. Up to a couple years ago, I had only used this technique to trace the migrationhistory of people found at archaeological sites.

My first chance to apply 18O analysis in a forensic context was a case being investigated by Det. Paul Dostie of the Mammoth Lakes CA, Police Dept. The well-preserved remains of an unidentified female murder victim had been found in a shallow grave. It was hoped that by determining her identity the identity of the assailant could be found. Using the C-13 method I was able to show that the victim had a large quantity of corn in her diet, especially as a child and hence was probably a Native American. But what did the oxygen isotopes tell us?

We analyzed two samples of enamel from a premolar tooth (P4). We know that these teeth form their crowns between ages two and eight years so the analysis of this tooth will tell us something about where she lived as a child. For comparison purposes, we also analyzed a single deciduous tooth obtained from a person born and raised in Mammoth Lakes, CA. We also analyzed samples of cortical bone from a single rib, which should be a record of where she lived over the last ten years of her life.

The deciduous tooth from a Mammoth native contained less 18O than either of the samples from the victim. From its 18O value (22.0%) we could calculate the 18O values of the water that had been consumed by the donor of this tooth. Its value (-12%) agreed closely with the value for precipitation in the Mammoth Lakes region, and confirmed that we were able to evaluate 18O of the local drinking water by analyzing a person’s teeth or bones. At the same time, the higher 18O values that we got from the victim’s tooth and bone showed that she had not spent much of her life in the vicinity of Mammoth. Rather, both values indicated that she was most likely from places further south and a lower elevation.

The water which she drank as a child (represented in her tooth) could have come from a wide area of the southwestern U.S. or possibly further south into northern Mexico. This was consistent with the record of the carbon isotopes in her diet which showed that as a child she had eaten a diet very rich in maize.

At the same time as we were making these measurements, another team working on the DNA from the victim came up with a startling observation. There appeared to be a good match between the DNA of the victim and that found in a population living in Oaxaca, Mexico. The convergence of these data on a single location suggests that this was where the victim’s family originated.

Following up on this lead, we had samples of drinking water collected in the village where she might have grown up in Oax-aca. When these were analyzed, we found that they matched almost exactly to the value we had calculated from her teeth, confirming that it was possible that she came from that village.

What surprised us, however, was the 18O of the bone, which was about 3.5 % higher in 18O than the tooth, and of course even higher than that of a native of Mammoth. We calculated that at the place where she had lived most of her later life, precipitation had an 18O value of about -5.0 %. That was higher than we could expect to encounter anywhere in the coterminous United States. The most likely place to find such 18O–rich rain was at low elevations in southern Mexico or northern Guatemala, showing that she probably had moved away from her birthplace but was stillliving south of the border for most of her adult life.

Subsequently, an investigation in the area where her DNA seemed to have originated gave further tantalizing clues as to her origin, including a possible firm identity. This search is still continuing. Hopefully, using the combination of isotopic and DNA analyses, we will one day be able to give the Mammoth Lakes victim a name.

We can see from the Mammoth, CA, case that analyses of isotopes in bones and teeth can tell us something about where people may have lived at different stages in their life. Actually, the most recent record in a human body would be found in the layer of tartar that is scraped off by dentists when teeth are cleaned! These analyses allow us in a sense to track the movements of a person throughout their entire lifetime and even before birth (because the first permanent molar begins to form in the womb). Although 18O analyses don’t give us a perfectly tight fix on a person’s past whereabouts, they can be combined with other evidence (including other isotopic data)to home in on possible places of origin for a victim.

For more technical details about how these analyses are done, you can check out this paper on using O isotopes in archaeology:

Prowse, T., Schwarcz, H.P., Garnsey, P., Knyf, M, Mac-chiarelli, R. and Bondioli, L.. Isotopic Evidence for Large Scale Immigration to Imperial Rome. AmericanJournal of Physical Anthropology, in press, 2005.

Henry Schwarcz Ph.D. is Distinguished University Professor Emeritus in the School of Geography and Earth Sciences of McMaster University, Hamilton, Ontario, Canada. For over three decades he has carried out scientific research in archaeology and anthropology. Henry can be reached at