Ancestry Determination from Skeletal Remains: Morphological and DNA Markers

When a forensic team recovers a skeleton with no ID, no missing-persons report, and no soft tissue left to examine, ancestry becomes one of the first questions on the table. Was this person of African, Asian, or European descent? The skull and bones carry real clues, but reading them accurately is far trickier than matching a photograph to a database. Ancestry estimation sits at the meeting point of skeletal biology, statistics, and increasingly, genetics, and getting it right (or admitting when it can’t be done reliably) is part of the job.

Why ancestry estimation is more complicated than it looks 🔗

The biggest challenge in this area of forensic anthropology is a biological reality: humans simply don’t sort into neat, separate boxes. Decades of population genetics research have shown that most human genetic variation exists within populations rather than between them. Roughly 85 percent of all human genetic variation occurs within a single population, while only a small fraction separates one broad population group from another. That means two people from the “same” ancestry category can differ from each other genetically more than they differ from someone in a different category. There are no sharp genetic borders between groups, only gradual, overlapping trends.

This is exactly why forensic anthropologists no longer talk about rigid “races” but about broad, geographically patterned trends in skeletal features. Before ancestry is even attempted, sex and age must be estimated first, because both influence how bones grow and develop, and skipping this step can distort the ancestry assessment that follows. Mixed-ancestry individuals add another layer of difficulty: their skeletons often show a blend of features rather than a textbook example of one group, so experts have to weigh multiple traits together and judge which ancestral contribution appears dominant, rather than expecting a single feature to give a clean answer.

Reading the skull for clues about ancestry 🔗

Despite these limitations, the cranium remains the single most informative bone for ancestry assessment. Most ancestry research has focused on the skull, particularly the mid-face region, because it shows the clearest population-level trends among all skeletal elements.

Features linked with European ancestry 🔗

Skulls associated with European ancestry typically show a receding frontal profile, meaning the forehead slopes back rather than sitting upright. The eye orbits tend to be more circular, and the nasal bridge is elevated with a sharp angle and a projecting nasal spine at its base. The dental arcade, the curve formed by the teeth, is usually parabolic, and the teeth themselves tend to be comparatively small.

Features linked with African ancestry 🔗

Skulls associated with African ancestry commonly show a more rounded frontal profile, often with visible vascular markings on the bone surface. The orbits tend to sit lower and appear more rounded, the nasal bridge is broader with a smaller nasal spine, and the dental arcade is hyperbolic, meaning wider and more U-shaped, with larger, more widely spaced teeth. Alveolar prognathism, where the jaw projects forward at the tooth-bearing region, is another commonly noted trait.

Features linked with Asian ancestry 🔗

Skulls associated with Asian ancestry often show a more vertical frontal profile, a tented nasal root, and a narrower nasal bridge. The zygomatic bones, or cheekbones, tend to be prominent and flare outward with a downward projection. The dental arcade is elliptic, sometimes described as horseshoe-shaped, and the upper incisors frequently show a shovel-shaped form, a distinctive scooped appearance on the back surface of the tooth.

None of these features work in isolation. Forensic anthropologists look at a combination of traits across the whole skull, because any single feature can appear in more than one population.

What the rest of the skeleton can tell us 🔗

When the skull is missing, damaged, or incomplete, the postcranial skeleton, essentially everything below the head, becomes the next best source of information. It’s less reliable than the cranium, but certain bones, especially the femur, do show measurable population trends.

Femur shape and curvature 🔗

The femur, or thigh bone, has attracted more ancestry research than almost any other postcranial bone. Anterior femoral curvature has long been studied as a possible indicator of ancestry, though early techniques were criticised for being too subjective to apply reliably. In general, femora associated with African ancestry tend to be flatter with less forward curvature, those associated with Asian ancestry show the greatest degree of anterior bowing, and those associated with European ancestry tend to fall somewhere between the two. A related feature, femoral torsion (the degree of twist along the shaft), follows a similar pattern, with Asian femora typically showing the greatest anteversion, African femora the least, and European femora landing in the middle.

Limb proportions and the intercondylar notch 🔗

The ratio between the length of the distal and proximal limb bones, essentially how the lower arm or leg segment compares to the upper one, tends to be significantly greater in individuals of African ancestry compared to other groups. At the knee end of the femur, the intercondylar notch, the groove between the two rounded condyles, is also generally greater in height among individuals of African ancestry than among those of Caucasian or Asian ancestry. Researchers caution, though, that very few postcranial methods have shown strong enough predictive power to be widely accepted, which is why the cranium still carries more weight whenever it’s available.

Turning bones into numbers: metric analysis and software 🔗

Alongside visual assessment, forensic anthropologists use metric analysis, taking precise measurements between defined landmarks on the skull and skeleton, to make ancestry estimation more objective. These measurements are compared statistically against reference populations to calculate the probability that a given skeleton belongs to a particular ancestry group.

Several software programs support this process, including FORDISC, 3D-ID, OSSA, 3SKULL, CRANID, and Ancestree. Each relies on a geography-specific reference database of known individuals to classify unknown remains. This is also where the method’s biggest weakness shows up: these tools are only as reliable as the populations represented in their databases. If the individual’s actual geographic ancestry isn’t well represented in the reference sample, the software’s classification can be inaccurate or misleading. Many populations remain poorly represented in published skeletal databases, which is a real limitation for regions, including South Asia, where reference data is comparatively sparse. Measurements from an unidentified skeleton are typically compared against data collected from living individuals of known ancestry, adding another layer of assumption to the analysis.

What DNA adds to the picture 🔗

When skeletal features alone can’t resolve ancestry, or when researchers want to cross-check a morphological estimate, molecular analysis of ancient or degraded DNA extracted from bone or tooth material offers another line of evidence.

Ancestry-informative genetic markers 🔗

Several types of DNA markers are used for this purpose: single nucleotide polymorphisms (SNPs), short tandem repeats (STRs), variable number tandem repeats (VNTRs), insertion-deletions (InDels), and transposable elements such as Alu SINE sequences. These are collectively known as ancestry-informative markers, selected because their frequency varies meaningfully between geographic populations.

Autosomal, mitochondrial, and Y-chromosomal markers 🔗

Autosomal markers are inherited from both parents, so they reflect the combined ancestral background of an individual and work well for both sexes. A complete picture of biogeographic ancestry typically draws on autosomal, Y-chromosomal, and mitochondrial markers together, since each traces a different line of inheritance. Mitochondrial DNA (mtDNA) is passed down only through the maternal line, so it reveals maternal lineage specifically. Y-chromosomal markers, both SNPs and STRs, are passed from father to son and therefore reveal paternal population background. When all three marker types point to the same geographic origin, confidence in the ancestry estimate is high. When they diverge, it often signals mixed ancestry across recent generations, since maternal and paternal lines can trace back to different regions.

Combining these molecular tools with the statistical models used in SNP-based ancestry and phenotype prediction research has considerably improved the accuracy of biogeographic ancestry assignment in forensic casework, especially for skeletal remains too degraded for standard identification methods.

Bringing morphology and molecular data together 🔗

No single method settles the question of ancestry on its own. Cranial morphology offers the richest visual information but can be damaged or absent. Postcranial features add supporting evidence but carry weaker predictive power. Metric software brings statistical rigor but depends entirely on how representative its reference database is. DNA analysis provides a molecular check, especially valuable for degraded remains, but requires suitable lab access and viable genetic material. In real casework, forensic anthropologists usually combine several of these approaches, treating ancestry not as a fixed label but as a probability statement built from multiple independent lines of evidence.

What do you think? If cranial and DNA evidence pointed to two different ancestry backgrounds for the same skeleton, which line of evidence do you think should carry more weight, and why? How might the shortage of skeletal reference data for South Asian populations affect the accuracy of software-based ancestry estimation in Indian forensic cases?

References

1. https://www.ncbi.nlm.nih.gov/books/NBK20363/
2. https://www.dovepress.com/ancestry-estimation-advances-and-limitations-in-forensic-applications-peer-reviewed-fulltext-article-RRFMS
3. https://pubmed.ncbi.nlm.nih.gov/10432602/
4. https://www.sciencedirect.com/science/article/abs/pii/S0379073813005501
5. https://pressbooks.ccconline.org/ppscant2315introtoforensicanthropology/chapter/chapter-10-estimating-ancestry-in-human-skeletal-remains/
6. https://pmc.ncbi.nlm.nih.gov/articles/PMC6976916/
7. https://www.ojp.gov/pdffiles1/nij/grants/244250.pdf