The Face of Punín and What DNA Already Knew
César Paz-y-Miño. Researcher in Medical Genetics and Genomics. UTE University. Quito.
Figure 1. Digital facial reconstruction of Punín-2, a man approximately 50 years old who lived in the Punín area of Chimborazo between 3,000 and 4,000 years ago. Image: Dr. Chris Rynn, in collaboration with Héctor Flores Franco. Reported by Metcalfe (1); screenshot from El Hemisférico, October 3, 2026. See the image in the original source.
The facial reconstruction of a skull from Chimborazo is consistent with what Ecuadorian population genetics has been showing for years. It also exposes the limits of using bones to infer ancestry.
For a century, two skulls from a ravine in Chimborazo were made to tell stories they could not support. One was said to be Australian, another almost a Homo erectus, and another evidence of Pacific seafarers who supposedly reached the Andes before anyone else. An international team has now given Punín-2 a face, and the result is less exotic but far more interesting: an Andean man around fifty years old, of Indigenous American ancestry, who lived between three and four thousand years ago on the same land where his probable descendants live today (1). What digital sculpture depicts in skin and hair, Ecuadorian population genetics had already been expressing in numbers. The distinction matters. That is what this article is about.
The Ravine That Misled Anthropology
The first skull, Punín-1, was discovered in 1923 in the Chalán ravine during an expedition by the American Museum of Natural History. Its sloping forehead and prominent brow ridge were enough for Sullivan and Hellman to compare it, in 1925, with skulls of Aboriginal Australian women (2). The reasoning was typical of the period: measure, compare against a catalogue of human “types,” and infer migrations. The bones had been found near the remains of Pleistocene mastodons and giant ground sloths, and the conclusion seemed to follow naturally.
But it did not. In 1980, Brothwell and Burleigh radiocarbon-dated the skull to the Holocene, far removed in time from the fauna with which it had been associated (3). Later geological investigations explained the mistake: erosion and soil movement had mixed more recent human remains with ancient fossils (1). Punín-2, found by a farmer in the same ravine in the 1970s, inherited both the confusion and its eventual correction.
There is a lesson here worth dwelling on. Skull shape reflects hundreds of genetic variants with small effects, as well as diet, chewing, climate, and childhood growth. To read ancestry from a bony profile is to read a text written by many hands. Anthropology in the 1920s did so with the certainty of a discipline that imagined races as neatly bounded categories. They are not, as genetics demonstrated decades ago. Indeed, the very concept of “race” as a biological classification is now widely challenged.
What Facial Reconstruction Can—and Cannot—Do
Christopher Rynn, a British forensic anthropologist specializing in craniofacial anatomy, worked with Chimborazo sculptor Héctor Flores Franco. He used the so-called Manchester method: soft-tissue thickness tables, muscles modeled over the underlying bone, and a combination of digital techniques and wax (1). Rynn explains that he deliberately avoided learning about the skull’s origin so as not to bias the result. The hairstyle, which no skull can preserve, was chosen in consultation with local people. He also adjusted the depth of the eyes because, according to those consulted, earlier versions looked too East Asian (1).
I regard that as an honest decision and, at the same time, a warning. The features of the face that can be inferred from the skull have an anatomical basis; the nose, lips, ears, skin pigmentation, and hair are subject to considerable uncertainty, as the facial approximation literature acknowledges (4). A reconstructed face is a visual hypothesis. It can restore a sense of humanity to remains once treated as a museum curiosity; it cannot establish a lineage. Rynn himself acknowledges that DNA will ultimately reveal the individual’s genetic origins (1). The social-media headline claiming that the reconstruction “rules out” an Oceanian origin goes too far. The dating, the geology, and above all what we already know about the genomes of Andean peoples are what challenge that hypothesis.
Parallels with Our Own Research
This is where our genetic research in Ecuador enters the picture. I see five substantive parallels between the story of Punín-2 and the work of our group at UTE University in Quito.
The first is Indigenous continuity. In 240 mestizo Ecuadorians from the Coast, the Highlands, and the Amazon, analyzed using 46 ancestry-informative markers, we found a Native American contribution of up to 51%, followed by European and African contributions, with regional variation (5). When we compared these findings with nine earlier studies of autosomal, mitochondrial, and Y-chromosome DNA, the pattern was consistent with the historical record: predominantly Indigenous maternal lineages and substantially European paternal lineages (5). The ancestral legacy represented by Punín-2 did not disappear. It persists, admixed, in much of Ecuador’s population, including among those who imagine themselves disconnected from it.
The second is that some populations retain this ancestral background almost intact. Among the Shuar of the Amazon, we measured 98.7% Native American ancestry, with no appreciable signal of colonial-era admixture (6). If an ancient genome is ever obtained from Punín, populations such as this will provide reference points for distinguishing Andean from Amazonian ancestry, and ancient from more recent genetic patterns. We therefore need more locally led research in genetic and genomic anthropology and paleopathology.
The third concerns geography. A study of the Cañari people in which I participated analyzed Y-chromosome and mitochondrial DNA in communities in Cañar, the province neighboring Chimborazo, and among groups in northern and southern Peru (7). The Indigenous maternal lineages belonged to the four major founding haplogroups of the Americas—A2, B2, C1, and D1—all ultimately of Asian origin and associated with migration through Beringia. What emerged was a history of internal population movements: families relocated by the Inca as mitmakuna, returns following the Spanish conquest, and genetic exchange along the Qhapaq Ñan (7). The Ecuadorian Highlands were a corridor for human movement long before Europeans arrived. The man from Punín belongs to that broader ancestral background.
The fourth is methodological. Among both the Cañari and the Shuar, DNA corrected or qualified historical narratives and cultural assumptions. At Punín, morphological interpretations remained mistaken for half a century, until radiocarbon dating set the chronology straight. The pattern is familiar: intuitions drawn from visible form require independent molecular testing through genetic studies.
The fifth is ethical. The Cañari study was conducted with the consent of Indigenous confederations and authorities, as well as individual participants (7). The reconstruction of Punín-2 was first presented to people in Chimborazo regarded as probable descendants of the man (1). Both practices recognize the same principle: Indigenous remains and genomes are not neutral raw material. They belong to histories with living heirs.
The table summarizes these parallels while distinguishing the types of evidence on which each rests.
Aspect | Punín-2 (2026 reconstruction) | Studies by Paz-y-Miño C. et al. | Parallel |
Ancestry | Indigenous American ancestry inferred from morphology (1) | Mestizo Ecuadorians with up to 51% Native American ancestry (5); Shuar with 98.7% (6) | Continuity of Native American ancestry in Ecuador |
Method | Skull-based facial approximation, Manchester method (1) | 46 ancestry-informative markers, STRs, mitochondrial DNA, and Y chromosome (5–7) | The same question: form suggests, sequencing tests |
Geography | Punín, Chimborazo, central Ecuadorian Highlands (1) | Neighboring Cañar province; Coast, Highlands, and Amazon (5, 7) | Shared Ecuadorian Andean region |
Rejected hypothesis | Australian origin or extinct human species (2, 3) | Maternal haplogroups A2, B2, C1, and D1, with Asian origins (7) | Settlement via Beringia, without direct Oceanian contribution |
Mobility | Lived in the area 3,000–4,000 years ago (1) | Inca relocations (mitmakuna) and movement along the Qhapaq Ñan (7) | The Highlands as a corridor for population movement |
Community | Local consultation on hairstyle and eyes; presentation to probable descendants (1) | Consent from Indigenous confederations and authorities (7) | Remains and genomes with living heirs |
Outstanding work | Ancient DNA of Punín-1 and Punín-2 not yet analyzed (1) | Hispanics: 0.16% of clinical-trial participants (10) | Ecuador underrepresented in reference databases |
The Specter of Oceania, Revisited Through Genomics
The story has a twist that makes it richer. Sullivan and Hellman’s proposed mechanism was wrong, yet in 2015 genomics uncovered something unexpected. Skoglund and colleagues identified in Amazonian populations in Brazil, including the Suruí and Karitiana, a signal of affinity with populations from Australia, New Guinea, and the Andaman Islands that was absent from other American groups. They called it Population Y (8). The signal is faint—approximately 2% if the contribution came from an unadmixed source—and points to population structure among the founding groups that crossed Beringia, not to rafts crossing the Pacific.
A second clue did emerge in Ecuador. In 2011, Geppert and colleagues identified among Amazonian Kichwa and Waorani a rare paternal Y-chromosome lineage, haplogroup C3*, nearly absent from other Indigenous peoples of the Americas (11). Later studies found it in 26% of the Kichwa and 7.5% of the Waorani individuals examined (12). The broader Y-chromosome haplogroup C has branches in Australia, Melanesia, and Polynesia, while the C3 branch is common in Mongolia and East Asia. The temptation to invoke trans-Pacific contact returned. Contact with Jōmon Japan had even been proposed on the basis of Valdivia pottery. The genomic evidence did not support it: carriers showed no recent or ancient Asian admixture in their autosomal DNA, and the simplest explanation is a founding lineage that reached the Americas through Beringia 15,000–20,000 years ago, persisted in Ecuador through genetic drift, and disappeared from most of the continent (12).
In 2019, in a study I coauthored, we sequenced the entire Y chromosome of a newly identified Ecuadorian carrier and compared it with that of 21 Kichwa and Waorani carriers of C3. All shared the same mutations, indicating a single South American branch, which we named C3-MPB373 (13). Its closest relative is in Siberia, not Australia, and the lineages diverged approximately 21,600 years ago (range 19,100–24,400), during the last glacial period and before entry into the Americas (13). It is one of the three or four founding paternal lineages of the continent, and its confinement to part of Ecuador is consistent with the geographical structure already evident in South American lineages some 12,000 years ago (13). What looked like Oceania was Beringia, its legacy preserved in the Ecuadorian Amazon.
No skull can reveal 2% ancestry. That requires a genome. Ancient-DNA studies in the Andes have documented early regional differentiation and continuity over thousands of years (9), but Ecuadorian samples remain scarce in those datasets. Ancient-DNA analysis of Punín-1 and Punín-2 could help fill that gap. The question is not whether they were “Australian.” It is how populations of the northern Andes were structured four thousand years ago, how closely they resemble the present-day Puruhá and Kichwa peoples of Chimborazo, and whether they retain any echo of that founding diversity.
A technical caution is necessary, because the excitement of headlines tends to overlook it. Remains exposed in a damp ravine and handled over the course of a century are poor candidates for the preservation of endogenous DNA. The petrous portion of the temporal bone often offers the best chance of recovery. If DNA can be obtained, the resulting genome will need to be compared with reference datasets in which Ecuadorians are poorly represented.
Who Studies, and Who Is Studied?
This issue matters more than it may seem. In 2018, we reviewed the composition of samples used in basic cancer research and clinical trials. Among more than 89,000 trial participants, Hispanics accounted for just 0.16%, and two-thirds had no recorded information on ancestry or origin (10, 11). Population genomics carries a similar bias. Much of South America’s ancient DNA is sequenced, analyzed, and published outside the region. An Ecuadorian skull now has a face modeled in the United Kingdom, and its DNA, if studied, will probably travel to a foreign laboratory. We must ensure that any such research takes place within an equitable collaboration that safeguards local scientific independence.
The researchers involved in this project are working with colleagues in Chimborazo and seeking funding to continue. The broader problem lies at home. Ecuador has technical expertise in human genetics. What it lacks is sustained funding, a dedicated ancient-DNA laboratory with clean-room facilities, and a policy that treats biological heritage as heritage worth protecting.
We must also be vigilant about the ethnic discrimination and anti-Indigenous prejudice that so often resurface when stories like this make the news. In genetic terms, the face of Punín-2 represents an ancestor of a substantial proportion of those now viewing it on their phones. In Ecuador, racism against Indigenous peoples also reflects ignorance of our own genomes.
A Face Is Not a Genome
I am left with a tension I do not intend to resolve. The reconstruction of Punín-2 has value because it humanizes the remains and returns to a community an image of someone who may have been its ancestor. Yet it also risks repeating, with digital tools, the old mistake of reading identity from physical form. A hairstyle chosen through consultation and eyes adjusted in response to local perceptions are, in part, contemporary cultural constructions projected onto a man from the past. If DNA confirms his genetic relationship to the peoples of Chimborazo, the image will acquire deeper meaning. If it reveals something unexpected, we will have to accept that the face now circulating on social media was more ours than his. Are we willing to fund the science in Ecuador that would allow us to find out?
References
Metcalfe T. 'It's a portrait of their actual ancestor': Facial reconstruction of Indigenous man who lived in Ecuador 3,000 years ago shared with his descendants. Live Science [Internet]. 2026 Oct 2 [cited 2026 Oct 4]. Available from: https://www.livescience.com/archaeology/americas/its-a-portrait-of-their-actual-ancestor-facial-reconstruction-of-indigenous-man-who-lived-in-ecuador-3-000-years-ago-shared-with-his-descendants
Sullivan LR, Hellman M. The Punin calvarium. Anthropol Pap Am Mus Nat Hist. 1925;23(7).
Brothwell D, Burleigh R. The human cranium from Punin, Ecuador, with particular reference to morphology and dating. J Archaeol Sci. 1980;7(1):97-9.
Wilkinson C. Facial reconstruction: anatomical art or artistic anatomy? J Anat. 2010;216(2):235-50.
Zambrano AK, Gaviria A, Cobos-Navarrete S, Gruezo C, Rodríguez-Pollit C, Armendáriz-Castillo I, et al. The three-hybrid genetic composition of an Ecuadorian population using AIMs-InDels compared with autosomes, mitochondrial DNA and Y chromosome data. Sci Rep. 2019;9. doi:10.1038/s41598-019-45723-w
Paz-y-Miño C, Astudillo-González O, Maldonado-Oyervide D, López-Cortés A, Pérez-Villa A, Armendáriz-Castillo I, et al. Allele frequency data for 15 autosomal STRs and ancestral proportions using AIMs-InDels in the Shuar ethnic group from Ecuador. Forensic Sci Int Genet Suppl Ser. 2019;7(1):65-7. doi:10.1016/j.fsigss.2019.09.026
Sandoval JR, Lacerda DR, Jota MMS, Robles-Ruiz P, Danos P, Paz-y-Miño C, et al. Tracing the genetic history of the 'Cañaris' from Ecuador and Peru using uniparental DNA markers. BMC Genomics. 2020;21(Suppl 7):413. doi:10.1186/s12864-020-06834-1
Skoglund P, Mallick S, Bortolini MC, Chennagiri N, Hünemeier T, Petzl-Erler ML, et al. Genetic evidence for two founding populations of the Americas. Nature. 2015;525(7567):104-8.
Nakatsuka N, Lazaridis I, Barbieri C, Skoglund P, Rohland N, Mallick S, et al. A paleogenomic reconstruction of the deep population history of the Andes. Cell. 2020;181(5):1131-45.e21.
Guerrero S, López-Cortés A, Indacochea A, García-Cárdenas JM, Zambrano AK, Cabrera-Andrade A, et al. Analysis of racial/ethnic representation in select basic and applied cancer research studies. Sci Rep. 2018;8(1):13978. doi:10.1038/s41598-018-32264-x
11. Geppert M, Baeta M, Núñez C, Martínez-Jarreta B, Zweynert S, Vacas Cruz OW, et al. Hierarchical Y-SNP assay to study the hidden diversity and phylogenetic relationship of native populations in South America. Forensic Sci Int Genet. 2011;5(2):100-4.
Comentarios