Last week, Dr. Bethany Turner of Georgia State University gave a talk at Vanderbilt called, “Diet versus locale: isotopic support for causal influences in pathological conditions at Machu Picchu, Peru.” Bethany’s work centers on analysis of skeletal remains for multiple isotopes – Sr, O, Pb, C, and N – to investigate the heterogeneity of the population, which was composed of slaves, short-term (non-local) laborers, and locals. I greatly enjoyed the talk because, even though Machu Picchu is far removed in time and place, Bethany and I are using similar methods to answer similar questions about physical mobility in the past. Imperial Rome also, of course, had millions of slaves, as well as free immigrants who came looking for work and locals who were born there.
One of Bethany’s research questions was whether the immigrants were less healthy than the locals. In my dissertation research (Killgrove 2010), I investigated the frequencies of common diseases – osteoarthritis, dental caries, abscesses, linear enamel hypoplasias, and porotic hyperostosis – and found that immigrants to Rome were not significantly less healthy than locals, although they did seem to die at an earlier age (possibly of new diseases they were not immune to, possibly because the immigrant population had a different demographic profile than the locals did). Bethany took a slightly different approach to this question: she looked at porotic hyperostosis, which is a bony reaction to anemia that develops in childhood, and found that it was significantly correlated with oxygen isotopes.
Backing up a bit, anemia has many causes, but it often results from diet or from parasites, although it can also be the result of a genetic condition (such as sickle-cell anemia or thalassemia). If a person eats too much maize, for example, that individual is at greater risk of developing a dietary anemia because maize is low in iron. This also holds for millet, which is much lower in iron than its C3 cousins, wheat and barley. So people with high carbon isotope values indicative of C4 (maize/millet) consumption may be expected to have higher frequencies of porotic hyperostosis if diet was the primary contributing factor to anemia. But people who grew up in an area without clean water, particularly an area with a large parasite load like hookworms, may also be at great risk of developing anemia when the parasite attaches to the intestinal lining and robs its host of needed nutrients like iron.
To distinguish between dietary and parasitic anemias as a cause of porotic hyperostosis, Bethany graphed her Machu Picchu individuals on a carbon/oxygen scatterplot. She found two fairly distinct groups of people along the oxygen axis: those with porotic hyperostosis and those without. This clustering she interpreted along the lines of Blom et al. 2005, who argued that a latitudinal patterning of porotic hyperostosis along the coast of Peru and a tendency for childhood anemia to be present in populations from more humid environments may be related to high parasite loads in certain locations rather than to differences in diet. In fact, Bethany’s data did not vary much on the carbon axis, further suggesting a parasitic origin for anemia rather than a dietary one.
Since I have all the same data from my two Roman populations, I created a similar graph to see what patterns there were in the carbon, oxygen, and porotic hyperostosis data. In the scatterplot below, individuals with and without porotic hyperostosis are plotted, and the yellow box represents the “local” oxygen isotope range of Rome:
|C and O isotope data from the first molars of two|
Imperial Roman (1st-3rd c AD) populations
Unfortunately, my Roman data were not as clear-cut as Bethany’s Peruvian data. Except for the one individual who consumed a C4-heavy diet and suffered from porotic hyperostosis, the rest of the diseased individuals are distributed within -13 to -11 permil on the carbon axis, which represents the average Roman diet of mostly C3 resources like wheat and barley. The people with porotic hyperostosis are spread out on the oxygen axis; however, there are none with oxygen isotope values lower than that of Rome. You may recall from older blog posts (like this one) that oxygen isotope values are more negative in cool, dry climates and more positive in hot, humid climates. It’s actually not a surprise, then, that the non-local people with porotic hyperostosis are on the right side of the graph: they were likely from places warmer and more humid than Rome, which means places along the sea and to the south – places that historically had more malaria, for example, than even Rome did. There are few data points on the left side of the graph, but again, I would expect there to be less malaria and fewer parasites in general in cooler, drier climates like the Apennines that were the source of freshwater springs.
This Roman sample size is small, and the data are not perfectly correlated. A simple t-test, though, actually indicated a statistically significant difference between the oxygen isotope means of the group with porotic hyperostosis and the group without it (t=3.06, p<.005), so with more data, I may find a more robust result. Graphing carbon versus oxygen isotope data has been done for years, but I’d never thought to add porotic hyperostosis as a variable until I heard Bethany’s wonderful talk. This technique has great potential for investigating parasitic disease in ancient Italy, and additional bioarchaeological research - specifically, isotopic analysis - on this front could yield a much stronger argument for the disease ecology of malaria and other parasitic diseases in the peninsula, adding a new dimension to previous osteological studies (e.g., Facchini et al. 2004).
Blom, D., Buikstra, J., Keng, L., Tomczak, P., Shoreman, E., & Stevens-Tuttle, D. (2005). Anemia and childhood mortality: Latitudinal patterning along the coast of pre-Columbian Peru American Journal of Physical Anthropology, 127 (2), 152-169 DOI: 10.1002/ajpa.10431
Facchini, F., Rastelli, E., & Brasili, P. (2004). Cribra orbitalia and cribra cranii in Roman skeletal remains from the Ravenna area and Rimini(I–IV century AD) International Journal of Osteoarchaeology, 14 (2), 126-136 DOI: 10.1002/oa.717
Killgrove, K. 2010. Migration and mobility in Imperial Rome. PhD dissertation, UNC Chapel Hill. [PDF]
Turner, B., Kamenov, G., Kingston, J., & Armelagos, G. (2009). Insights into immigration and social class at Machu Picchu, Peru based on oxygen, strontium, and lead isotopic analysis Journal of Archaeological Science, 36 (2), 317-332 DOI: 10.1016/j.jas.2008.09.018