The Antiquity paper counts 280 motifs, 55 of them animals, on the walls of a Catalan cave that tourists walked through for more than six decades, but the often-quoted 26,000-year age belongs to a dated layer nearby, not to any engraving.
Peer reviewed · Descriptive site study · Engravings not directly dated
Archaeologists Josep Maria Vergès Bosch and Míriam Salas-Altès of IPHES-CERCA and Rovira i Virgili University report 280 Palaeolithic engraved motifs, including 55 animal figures, in Cova de la Font Major at l’Espluga de Francolí, Catalonia. Their paper appeared in Antiquity on 5 November 2026. Before the 2019 discovery, Catalonia had only two cave images attributed to the Palaeolithic. The open question is age: no engraving has been dated directly, and the authors place many of them before the Magdalenian period, which began about 19,000 years ago, without ruling out younger ones.
What you need to know
- Finding: 280 engraved motifs in five sectors of one gallery system, including 55 animals (horses, deer, goats or ibex, wild cattle and a rare bear) plus hundreds of abstract marks.
- Method: Wall-by-wall recording with high-resolution 3D models, then comparison of drawing conventions with other Mediterranean sites and radiocarbon dates from charcoal on the gallery floors.
- Source: Vergès Bosch and Salas-Altès, Antiquity, published online 5 November 2026, open access.
- Why it matters: By its count of animal figures, Font Major is the third-largest Palaeolithic cave art site on Spain’s Mediterranean side, in a region once thought almost empty of it.
- Limitation: The age is inferred from style and from dated traces of people in the same galleries (about 41,000 to 26,000 years ago); uranium-series dating of one calcite-covered animal is still to come.
Evidence at a glance
| Item | Assessment |
|---|---|
| Evidence status | Peer reviewed; descriptive archaeological study of one site; age is a relative (stylistic and contextual) attribution. |
| Primary source | Vergès Bosch J.M. and Salas-Altès M., “Filling gaps: the Palaeolithic rock art of the Cova de la Font Major on the north-east Iberian Peninsula”, Antiquity, First View, published online 5 November 2026, DOI 10.15184/aqy.2026.10447. Institutional release: IPHES-CERCA, 5 November 2026. |
| Evidence strength | Strong for the count and description of the figures, which are on the walls and recorded in 3D. Moderate to limited for the age, because no engraving has a measured date and the authors say later engravings cannot yet be excluded. |
| Principal limitation | No direct dating. The 41,000 to 26,000-year dates come from charcoal and a flint tool in the galleries’ floor layers, not from the engraved grooves. |
| Coverage at a glance | 7 distinct credible publishers as of 7 November 2026, 08:10 UTC (Nature, Scientific American, Gizmodo, Archaeology Magazine, The Olive Press, Infobae, VRT NWS), found by SciQuest web searches in English, Spanish and Catalan. Syndicated copies (such as Yahoo’s reprint of Scientific American) and aggregator sites were excluded. Coverage is not evidence the finding is correct. |
Sources: the Antiquity paper, the IPHES-CERCA release and SciQuest’s coverage search on the observation date.
Is this a big deal?
For the map of Ice Age art in Spain, yes. A region that had two uncertain images now has a site with 55 animal figures, and a survey prompted by it at Cova Simanya has found more. For the age, it is a careful first estimate. Treat “older than about 19,000 years” as the authors’ best reading, and treat any single figure such as “26,000 years” as a headline shortcut.
What scientists found
The engravings sit in the Gatoneres de Cal Palletes, a set of low passages about 250 m from the cave’s present entrance, roughly two and a half football pitches in. Ceilings there range from under 1 m to 3 m. The authors split the decorated stretch into five sectors and count 280 graphics, a tally that groups marks into what they call graphic units and graphic sets.
Fifty-five of those are animals. Sector 5 alone holds 33 of them, about 60%. The species include horses, deer (mostly females), caprines such as goats or ibex, wild cattle including an aurochs, and one bear drawn with its ears facing the viewer and its mouth slightly open. The paper calls the bear a singular subject in Palaeolithic art. Some animals appear to have swollen, pregnant-looking bellies. About 30% of the 55 cannot be assigned to any animal group, and another 15% could be one of two, such as goat or deer.
Most of the repertoire is abstract: parallel and converging lines, circles, rectangles and, on one panel, hundreds of small loops. Most figures are 15 to 45 cm long, so the largest are about one and a half times the long side of an A4 sheet, and the grooves are mostly 3 to 7 mm wide, no wider than a pencil.
How they found it
Nobody was looking for art. On 30 October 2019, during the first excavation season of a new project, one of the authors was walking through the Gatoneres and noticed that some of the scratches among the modern names and graffiti had Palaeolithic features. The passage had been part of the cave’s “Adventure Route” tourist trail since 1993, and the entrance had been open since 1957. Researchers had also assumed the walls, mostly irregular conglomerate rock, were a poor surface for engraving.
The figures are cut into thin bands of soft marl between the conglomerate, a surface so soft that hands and fingers leave marks. That made them easy to make and easy to damage. Salas-Altès told Nature the engravings are “really difficult to see, you have to play with light from different directions.” It works the way a torch held low across a notepad picks out the dents left by the page above. The comparison has limits: these are carved grooves, not pressure dents, and centuries of graffiti cross them.
To avoid touching the walls, the team built high-resolution 3D models and measured the grooves from them. Most were made by simple incision and some by scraping.
How the age was estimated
This is the part most coverage compressed. The paper reports no direct date for any engraving. Instead it combines two lines of evidence.
First, drawing conventions. Some animals have hind legs drawn as a double “Y”, and some heads are outlined with three lines. Both habits appear at other Mediterranean sites that researchers link to pre-Magdalenian times, which the paper defines as roughly 43,000 or 40,000 to 19,000 calibrated years before present.
Second, who was in the cave and when. Radiocarbon dates on charcoal from the inner galleries show people there between 40,810 and 25,850 calibrated years before present, with an apparent gap from 35,400 to 26,300. The youngest Palaeolithic layer, dated to about 26,000 years ago, also held a small flint bladelet. After that, the inner galleries show no trace of people until the eighth to second centuries BCE, when an Iron Age sanctuary was in use in the Sala del Llac. That 26,000-year layer is where Nature’s headline figure comes from. It dates the floor, not the walls.
The authors say these data are “consistent with” a pre-Magdalenian date, and that “it is still too early to rule out” Magdalenian or later engravings. The hundreds of loops are a separate puzzle: charcoal from inside one groove shows some incisions are no later than the 17th century, and an inscription dated 1199 cut over one of them shows that one is older still. Whether they are Palaeolithic is unresolved.
Go deeper: every dated clue, oldest first
| Period | Evidence | What it dates |
|---|---|---|
| 41,000 to 35,200 cal BP | Radiocarbon dates on charcoal in floor layers | People in the galleries; not the engravings |
| 26,400 to 25,800 cal BP | Radiocarbon dates; one layer also held a flint backed bladelet | The youngest Palaeolithic visits; source of the “26,000 years” headline |
| 8th to 6th and late 3rd to 2nd centuries BCE | Iron Age finds in the Sala del Llac | A sanctuary in use; no link to the animal figures |
| 1199 CE | Inscribed date cut over a loop engraving | A minimum age for that one loop |
| No later than the 17th century | Radiocarbon date on charcoal inside an engraved groove | Some loop incisions; their origin is unresolved |
Source: Vergès Bosch and Salas-Altès, Antiquity (2026), text and Table 2 synthesis (rounded ranges; the paper’s text gives 40,810 to 25,850 cal BP overall, with a gap from 35,400 to 26,300). cal BP = calibrated radiocarbon years before 1950. No row is a date for an animal figure.
Why it matters to you
Nothing here changes daily life. It matters for the wonder lens: where people lived and made images during the last Ice Age, and how much of that record may still be hiding in places we already visit. The paper notes that thousands of visitors passed these walls, some scratching names over the figures. If 280 motifs went unrecognised for 62 years in a show cave, other well-trodden caves may hold more.
What the evidence does not show
It does not give a measured age for any figure, and it does not show that all 280 motifs were made at the same time. The original set was probably larger, because touching and graffiti have damaged every sector except the end of sector 5. The study also does not support claims of “animation”. The paper describes artists using natural features, such as a rock boundary as the ground under a horse and a crack as a possible ground or water line, but makes no claim about depicted movement. Nature’s subheading used the word without explaining it.
How this fits what we already knew
Ice Age cave art in western Europe is best known from the Franco-Cantabrian region of southern France and northern Spain. On Spain’s Mediterranean side, decorated caves are usually small: about 75% have five animal figures or fewer, and more than 90% have 20 or fewer. With 55, Font Major ranks behind only Ardales and La Pileta in Andalusia. Catalonia’s earlier record was thin. One painted bovid from Cova del Tendo was stolen soon after its discovery, and a possible deer at Cova de la Taverna may not be human-made.
“Contrary to previous assumptions, the north-east Iberian Peninsula preserves Palaeolithic rock-art sites of the highest importance,” Salas-Altès said in the IPHES-CERCA release. The same galleries also matter for the early Upper Palaeolithic, since the 40,000-year dates show people deep in the cave in a period when southern Catalonia was thought marginal. For another Catalan window on deep time, see our report on Neanderthal wooden tools preserved as moulds at Abric Romaní.
Independent perspective
SciQuest found no on-record comment from a researcher outside the team in the coverage it read, and has not yet obtained one. The only quoted voices so far are the two authors. Research at the cave is funded by the Government of Catalonia’s Department of Culture, Catalan and Spanish research agencies, Rovira i Virgili University and the l’Espluga de Francolí town council, according to the release.
How other headlines framed it
- Nature put a single age in its headline, art hidden “for 26,000 years”, and promised signs of animation; the paper gives no measured age and makes no animation claim.
- Gizmodo and The Olive Press used “at least 19,000 years”, which matches the pre-Magdalenian reading but drops the authors’ caveat.
- Scientific American and Archaeology Magazine led on the hidden-in-plain-sight angle and gave no specific age.
- Back in 2020, Catalan News reported “more than 100” engravings “around 15,000 years old”, a first estimate the paper has since replaced.
What happens next
At least one animal figure lies under a calcite crust, and the team plans uranium-series dating of it. That would date the crust on top, giving a minimum age for that one figure; the authors stress that one point date cannot fix the age of a whole assemblage. A full study of the site is under way, and a flooded passage beyond the Sala del Llac has not yet been explored. We will update this story if direct dates are published.
In one sentence
A Catalan show cave holds one of the largest sets of Ice Age engravings on Spain’s Mediterranean side, and its age, probably older than 19,000 years, rests on style and nearby dated layers rather than on any measured engraving.
Quick check: did it click?
- Where does the “26,000 years” figure come from? From a dated floor layer with a flint bladelet in the galleries, the youngest Palaeolithic trace of people there. No engraving has been dated.
- How many of the 280 motifs are animals, and which one is unusual? 55; the bear is rare in Palaeolithic art.
- What test could give the first direct date? Uranium-series dating of the calcite crust over one animal figure, which would give a minimum age for that figure.
Read next
- Neanderthal wooden spears and a possible club survive as moulds in Spain: another Iberian site where the trace, not the object, survives.
- Eland collagen found in 60,000-year-old ochre: older pigment, and an equally open question of purpose.
- What tools were made in the Stone Age?: background on stone blades like the one found in the 26,000-year layer.
Sources and disclosure
- Vergès Bosch J.M., Salas-Altès M. “Filling gaps: the Palaeolithic rock art of the Cova de la Font Major on the north-east Iberian Peninsula.” Antiquity, First View (2026). doi:10.15184/aqy.2026.10447. Published online 5 November 2026; open access (CC BY-NC-SA 4.0).
- IPHES-CERCA, 5 November 2026: Cova de la Font Major challenges the traditional map of European Palaeolithic art.
- Simms C. “The palaeo cave art that hid in plain sight for 26,000 years.” Nature news, 6 November 2026.
- Coverage read for framing: Scientific American, Gizmodo, Archaeology Magazine, The Olive Press, Catalan News (2020).
Disclosure: This article was drafted with AI assistance and checked against the sources above, including the paper’s open-access full text (the radiocarbon table itself is an image SciQuest could not read; dates are taken from the paper’s text). Calculations by SciQuest: 33 ÷ 55 = 60%; 2019 − 1957 = 62 years; 450 mm ÷ 297 mm (A4 long side) = 1.5. The timeline graphic is a SciQuest original drawn from dates in the paper and release; no third-party images are used. Spotted an error? See our corrections policy.
Discover more from SciQuest
Subscribe to get the latest posts sent to your email.

Leave a comment