New evidence is revealing the ages of death, birth, and menarche in Neandertals

Last week I was in Zagreb where I was beginning some new research on the skeletal remains of the Neandertals from Krapina. This site in northern Croatia has a central place in the history of paleoanthropology. More than 800 fossils of early Neandertals were excavated in 1899 and 1900 by Dragutin Gorjanović-Kramberger from a rock shelter in a sandstone cliff outside the town. The sample of skeletal remains from Krapina is still one of the largest samples of fossil hominins anywhere, more than 120 years after their discovery.

The Krapina Neandertals lived sometime between 130,000 and 120,000 years ago. This was the last interglacial, a time when the climate in Croatia was even slightly warmer than today. The skeletal remains represent more than 25 individuals, and all of them are very fragmented. Some of those individuals may be represented by only a few teeth, others by broken parts of the skeleton mixed with bones of other individuals.

On a previous visit to Krapina

Although the site was fully excavated more than a century ago, some amazing work has been coming out on the Krapina Neandertals over the last few years. Much of that work has been coordinated by the collection's curator, Dr. Davorka Radovčić. One area of this research has given new insights into the course of Neandertal growth, development, and lifespan.

Cementum evidence about adult lives

Last year, Paola Cerrito from New York University and her collaborators looked at the growth of a substance called cementum on the roots of Krapina teeth. Cementum is a hard organic substance that forms between the tooth roots and surrounding bone after a tooth erupts and the root is fully formed. The cementum grows incrementally and forms annual growth layers. These layers are very, very thin: between 18 and 33 microns, but it is possible to image them with X-rays using a synchrotron. Cerrito and coworkers used the SYRMEP beamline of the Elettra synchrotron facility in Trieste for this research.

There have been many advances in visualizing the microstructure of teeth over the last 20 years, and these have given many new insights about dental development. That work has been constrained in some ways by the realities of dental development. Permanent teeth develop and erupt during the first few years of an individual's life, so their growth and development is mostly useful to look at children's lives—even if the teeth come from adult individuals.

Cementum is different. Because it forms after a tooth has erupted, and continues to form throughout an individual's life, it provides a way of looking at the adolescent and adult portion of life history.

Part of figure 2 from Cerrito and coworkers 2022, showing the analysis for Krapina 172. The left panel shows a small portion of the tooth root in cross section, with the cementum forming a thin layer at its periphery. The high resolution scan provides evidence of peaks in the relative translucence or opacity, which reflect differences in mineralization. 

One basic insight is the age at death, obtained by counting the cementum after tooth eruption to see when an individual's clock finally ran out. More interesting, cementum is one of several dental tissues that may record significant stresses in an individual's life. Such physiological stresses can disrupt the balance of minerals in the body and the synthesis of proteins that make up the cementum. For women, this is especially significant as a view of life history, because cementum may reflect the approximate time of events like menarche, menopause, and parturition (the time that a mother gives birth).

But does it work?

Cerrito and coworkers studied a sample of teeth from research participants who had tooth extractions for medical reasons, and who were willing to provide information about life events for the study. The researchers looked at the cementum and tried to predict the timing of significant life events including menarche, parturition, significant illnesses, and menopause.

In this sample of living people's teeth, they found that the cementum provides an exceptional record of these life events. They did find a high fraction of false positives where they predicted an event that had not occurred. But very few of the real events were missed in the cementum records.

The cementum also seemed useful for assessing the sex of the research participants. The researchers found that the annual growth increments tended to be wider in women than in men in their sample, and also found other aspects of the cementum seem to differ. Although there was substantial overlap in the distributions, they suggested that cementum may be helpful in discovering the dental sex of unknown remains.

Cerrito and collaborators looked at cementum from ten Krapina Neandertal teeth. Only five of them gave reliable images; the cementum of the others was not well enough preserved for measurement. Based on the cementum characteristics, they suggested that two of the Neandertal teeth came from female individuals, while three were in the range of overlap between human male and female teeth.

A Neandertal age mystery

For me, one of the most important questions to answer with this approach is the most basic fact of life history: At what age did these Neandertals die?

For Krapina, that question goes back to Milford Wolpoff, who published the major assessment of the ages of Krapina individuals in 1979. Milford was my PhD advisor and remains a close colleague and friend. The Krapina skeletal remains include parts of seven upper jaws and 10 mandibles with teeth, in addition to nearly 200 teeth that are isolated from their jaws. It's an enormous sample, and when Milford began his work, it wasn't clear which teeth might represent the same or different individuals. He set out to discover how many individuals there had been, and what ages they had reached at the time of their deaths.

Gorjanović-Kramberger was a brilliant excavator by the standards of 1900. He recognized and collected very small fragments of bone, including in some cases the tiny developing germs of teeth. He recognized the position of hominin and faunal fossils within geological strata in the site. Most of the Neandertal bones came from a thick layer of sandy sediment, together with the bones of animals that had inhabited Europe during the interglacial.

So we have some confidence today that there was not an obvious spatial ordering of the bones and teeth. These did not seem to be a series of burials of whole bodies.

But it is not a random assortment of bones, either. There are clear biases in the skeletal remains with some parts of the skeleton overrepresented, other parts missing. Some of the Neandertal bones have cutmarks from sharp stone blades. A few have tooth marks on them. One skull has a series of incisions, carefully made parallel to each other on its forehead. Yet there is relatively little evidence of fragmentation or breakage at the time of death. All the bones are heavily broken and fragmented, but most of the fragmentation occurred after the bones lost their organic strength, long after death.

Cutmarks on the frontal bone of the Krapina 3 skull. Photo from research by David Frayer and coworkers, 2006

Anthropologists have debated the role of cultural and natural processes in what happened to the Krapina individuals. Some have favored the idea that other Neandertals ate the bodies and discarded the remains. Others have suggested that the individuals had died elsewhere, and their family or group later relocated some of their bones to this rockshelter, a practice known as secondary burial. If the Krapina skeletal remains had been the part of some kind of social event, it might be written in the sex and age of the individuals.

Milford began his work by following the methods of earlier anthropologists who had studied age and tooth wear in ancient populations. Today most of us don't think about tooth wear very much. Humans eating a diet of processed foods in industrialized countries don't tend to wear our teeth down. But our ancestors ate foods that were tough and sometimes coated with particles that damaged and wore down their enamel. Teeth erupt in a sequence, each tooth at a characteristic age. The first molars have several years of time chewing before the second molars erupt, and the second several years before the third molars erupt. Relying on these data may not give a very impressive accuracy for the age of any one individual. People vary, and their individual life events matter to the way their teeth wear. But across a sample of many individuals, if the tooth wear occurs at a fairly consistent rate, it is possible to work out which teeth came from older or younger individuals.

One of the sets of Neandertal teeth from Krapina matched by interproximal facets, wear, and occlusal characteristics. From Wolpoff (1979)

Going further, the teeth also wear directly against each other. If you find it difficult to floss, that's because teeth that are next to each other are actually wearing a flat surface between them, increasing their area of contact as you age. It gets very hard to squeeze a thread of floss between them, and eating popcorn can lead to hours of frustration. Those flat surfaces are called interproximal facets, and they are mirror images in size and curvature. Additionally, the upper and lower teeth that bite against each other form flat wear surfaces that match. With all these different elements taken together, Milford could start to find the teeth that belonged together, teeth from the same individual.

What he quickly noticed was that the Krapina Neandertals died young. There were many children in the sample, and the adults all seemed to have been less than thirty years of age. Indeed, only three or four individuals were clearly older than twenty.

Histogram from Wolpoff (1979) showing the percentage of individuals by age at death from the Krapina sample.

That was a remarkable contrast to skeletal samples of more recent humans from archaeological contexts. Even those populations of recent people where adults tend to die at younger ages, there is a large fraction of individuals who lived to be much older than thirty years. From today's perspective, with demographic data from various hunting and gathering populations around the world, the Krapina results stand out even more. This sample of individuals died very young.

Wolpoff reflected on the work that Gorjanović-Kramberger had done excavating the remains. The lack of older adults could not be explained by anything that Gorjanović-Kramberger had missed, not by excavation technique or preservation of the bones. Whatever determined this age distribution was in the time of the Neandertals themselves.

But the skeletal remains showed some clear biases. As young as the age of death was, the sample had few if any infants younger than around 3 years old. Maybe a similar kind of bias had kept the remains of older adults out of the rock shelter.

“I do not believe that the sample represents the age distribution at death for any real population that lived in the cave.”—Milford Wolpoff

What was less clear at the time was whether this young age at death might be unique to Krapina or whether it was characteristic of Neandertals more generally. Some Neandertals clearly had been aged at the time of their deaths—the so-called “Old Man” of La Chapelle-aux-Saints was a clear example. But there were so few of these older individuals, and how old they really had been was unclear.

Today we can say a little more about this. Neandertals did sometimes live to older than thirty years. But not nearly as many as there are people over thirty today. Rachel Caspari and Sang-Hee Lee looked at 113 Neandertals from across sites and time periods who lived long enough to erupt their third molars. They found that 37 of those individuals lived to ages older than 30 years or so. That's around a third of the total.

It's a smaller fraction than found in archaeological sites and cemeteries of recent people. But it's a larger fraction than for any earlier hominin populations, like those attributed to Homo erectus. The survival of Neandertal adults was not as high as later people associated with Upper Paleolithic toolkits in Europe, but it was just the same as early humans from Skhūl and Qafzeh.

Still this result leaves many unanswered questions. The jaws of Neandertals wound up in sites investigated by archaeologists for many kinds of reasons. Some were buried by their families, others scavenged by hyenas, some may have died alone. Certainly it is not an unbiased sample, and the direction of the bias across these different causes of death and deposition is not obvious. Also not obvious is whether tooth wear is really giving an accurate and reliable picture of age.

The cementum count

That's where the results from Cerrito and coworkers come in. The five teeth that yielded cementum data all gave results similar to Wolpoff's estimates from tooth wear.

They were not exactly the same. One tooth that Wolpoff surmised had belonged to a 16-year-old actually was older, estimated around 25 years by cementum. Another estimated to be 19 by Wolpoff instead was 26. But one tooth estimated by Wolpoff at around 25 by wear was actually only around 21, and the others were within a year of Wolpoff's estimate.

What seems clear from this small sample is that old-school tooth wear estimates in this sample did not have a large systematic bias toward younger ages. Tooth wear may not reflect age at death with great precision, but teeth from 40-year-old Neandertals are not hiding in the sample. The Krapina Neandertals really were young.

This result is kind of breathtaking to me. The Neandertal lifespan is one of the thorniest mysteries about these ancient people. The high level of mortality that we are seeing for younger adults must have had enormous effects on Neandertal cultural and social systems. Finding a new area of evidence that has a chance to put better numbers on their ages may help us understand how they lived.

New evidence about maturation

Meanwhile, the really fascinating part of the cementum work is the potential of understanding those big physiological stressors like menarche, parturition, and significant illnesses. Cerrito and colleagues were surprised to find only one event recorded in the cementum of each of the Neandertal teeth. That is a contrast with their sample of extracted teeth from living people, which typically had several events over their lifespan.

However, if we look only at the proportion of the humans' lifespans under 30 years, the picture is not so different. Five of the living individuals had a signature of menarche, two birth, and one of illness in that span. It's not that the Krapina Neandertals were very different, it's that they didn't live as long as the modern research participants.

Considering the evidence for sex attribution of the teeth, it seems plausible that the cementum from two of the individuals may be providing evidence of menarche.

“[T]he physiologically impactful events reported for A2309 (15.5 years) and KR172 (16.6 years) could reasonably relate to menarche, which in contemporary hunter–gatherer populations has a mean of 16.6 and a median of 17.1 years….”—Paola Cerrito and coworkers

Two of the Krapina individuals with indeterminate sex had significant stress events within two years of their deaths. As Cerrito and coworkers point out, it is not possible from the cementum data to tell which kind of event is being recorded. If these were female individuals, possibly the event was the birth of a child. Or for individuals of either sex, some major illness might be the stressor. I think it would be an amazing discovery to find that major social stressors might also leave such physiological traces. Conceivably, in the future we may find chemical or biomolecular signs of these different kinds of events.

If there is a criticism of the work, it is the relatively small sample of living human participants for whom life events were recorded. When a new technical advance enables a new kind of data collection, there's often a time of great enthusiasm when it is applied to small samples. I find the small sample in this paper quite compelling, but I've seen instances where surveying more human populations reveals a lot of variation in the biology. I'd love to see this research repeated on recent people from many parts of the world, to understand that variation better.

Bottom line

We're forming new ideas today about growth, development, and life history in human origins. Some of those new ideas have come from new techniques for gathering data from skeletal and dental remains, like the cementum study profiled here.

Krapina is such a remarkable example of how collections can give rise to new research and insights long after they are discovered. These Neandertals played an important part in the revolution of paleoanthropology during the 1970s, and again are yielding data that cause us to revise old ideas and build a foundation for a new generation of science.

Clearly something must have happened to the Neandertal individuals whose remains were at Krapina to explain why their age at death distribution is so young. The sample is no longer alone: there are now other large samples of Neandertal individuals with their own unique depositional scenarios. El Sidrón, Spain, is a site where at least 12 Neandertal individuals were subject to cannibalism. The six adults at El Sidrón are also relatively young, with no individuals assesssed as very old, but with such a small sample it is unclear how surprising this underrepresentation may be. I have a lot of hope that our better understanding of juvenile individuals at sites like El Sidrón may open new doors to look back at the Krapina sample.

Diarama of individuals reconstructed at the Krapina rockshelter by Elisabeth Daynès. Photo: John Hawks

Our record of human evolution is built out of a series of instances and events, each of which is unique in important ways. As we build more and more lines of evidence, we start to notice the ways in which the lives of these individuals really matter. Right now, there may be two Neandertal individuals were we can say something concrete about the age that they reached menarche. It's an area of biology where many researchers have assumed Neandertals might be very different from recent humans, and it turns out they may have been much the same.

We have a long way to go before the mystery of Krapina is solved. There may be no single solution; it may be that the remains came from different processes and events involving varied groups over a long time. But I have a hunch that in the long run, we'll be able to piece much of this story together.


Notes: I appreciate very much the assistance of Davorka Radovčić during my visit to the Krapina collection, as well as the insights of the many researchers who have contributed to the study of the remains and the questions discussed in this post. My special appreciation goes to Milford Wolpoff, David Frayer, Rachel Caspari, Sang-Hee Lee, Mary Russell, Fred Smith, and the late Jakov Radovčić.

The El Sidrón Neandertals have been described in a continuing series of publications. I've pointed below to the 2011 paper from Carles Lalueza-Fox and collaborators, which provides the basic demography of adult and juvenile individuals, together with evidence of possible kinship among the male individuals in the assemblage.

References

Caspari, R., & Lee, S.-H. (2004). Older age becomes common late in human evolution. Proceedings of the National Academy of Sciences, 101(30), 10895–10900. https://doi.org/10.1073/pnas.0402857101

Cerrito, P., Nava, A., Radovčić, D., Borić, D., Cerrito, L., Basdeo, T., Ruggiero, G., Frayer, D. W., Kao, A. P., Bondioli, L., Mancini, L., & Bromage, T. G. (2022). Dental cementum virtual histology of Neanderthal teeth from Krapina (Croatia, 130–120 kyr): An informed estimate of age, sex and adult stressors. Journal of The Royal Society Interface, 19(187), 20210820. https://doi.org/10.1098/rsif.2021.0820

Frayer, D. W., Orschiedt, J., Cook, J., Russell, M. D., & Radovčić, J. (2006). Krapina 3: Cut Marks and Ritual Behavior? Periodicum Biologorum, 108(4), 519–524. https://kuscholarworks.ku.edu/handle/1808/7044

Lalueza-Fox, C., Rosas, A., Estalrrich, A., Gigli, E., Campos, P. F., García-Tabernero, A., García-Vargas, S., Sánchez-Quinto, F., Ramírez, O., Civit, S., Bastir, M., Huguet, R., Santamaría, D., Gilbert, M. T. P., Willerslev, E., & de la Rasilla, M. (2011). Genetic evidence for patrilocal mating behavior among Neandertal groups. Proceedings of the National Academy of Sciences, 108(1), 250–253. https://doi.org/10.1073/pnas.1011553108

Wolpoff, M. H. (1979). The Krapina dental remains. American Journal of Physical Anthropology, 50(1), 67–113. https://doi.org/10.1002/ajpa.1330500110