Back to Don's Maps

walls of china Back to Archaeological Sites

Recent additions, changes and updates to Don's Maps


Stone Lamps of the Palaeolithic

types of  lamps

Lamp designs fall into three main categories:

Open-circuit lamps (top) consist of largely unaltered slabs of rock. When the lamp is lit, melted fat runs off through natural crevices in the rock.

Closed-circuit lamps (middle) have carved depressions to contain the runoff.

Carved-handle, closed-circuit lamps (bottom) also have bowls shaped fuel chambers but are more finely finished and have formed extensions for easier handling. Burn marks indicate that the wick was placed away from the handle.

Photo and text: Beaune and White (1993)



These minimalist lamps from Lascaux Cave are of the open-circuit type. They would have been picked up, used, and then discarded, without any thought being given to them.

Photo: Don Hitchcock 2014
Source: Original on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac


This sandstone lamp from Solvieux, broken in four pieces, shows a shallow depression for the grease and wick. It is dated to the early Magdalenian, and comes from couche A in locality 3 of Solvieux.

Photo: Don Hitchcock 2014

Source: Original on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac


Back view of the lamp above.

Photo: Don Hitchcock 2014

Source: Original on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Objects described as lamps definitively appear only late in the Upper Paleolithic. The earliest uncontested lamp comes from the Gravettian at Laugerie-Haute. It is possible that two Aurignacian artefacts from La Ferrassie, one Gravettian artefact from Arcy-sur-Cure, and one from Saint-Jean-de-Verges are lamps (de Beaune 1993). Gravettian lamps carved from mammoth femur are found at Kostenki I (Hoffecker 2005).

Text above from: Smith (1979)

Grotte de la Mouthe

The lamp from La Mouthe. This was the first lamp ever identified as such.


La Mouthe La Mouthe

Photograph from Rivière (1899) showing the ibex engraved on the underside of the lamp.

Photo: Rivière (1899)

This lamp appears to have been carved from a pebble of Permian sandstone which is abundant in the basin near Brive, about 40 km as the crow flies from Les Eyzies. It may well be that the Vézère or any of its tributaries have rolled a fragment of this rock, which may have been picked up in the river near Les Eyzies, or in its vicinity and carved for use as lamp cup, while one of its ends was ground down and rounded to serve as a short, thick, triangular handle (4 cm long). The total length of the lamp is 171 mm, handle included. The cup has been carved in a regular circular shape, and , measures 106 mm longitudinal diameter and 104 mm transverse diameter, with 34 mm as its greatest depth, in the centre. Finally, its thickness is 45 mm. The edge is also thicker at the opposite end to the handle and has no grooves.

Its general color is dark gray, except in the interior of the cup there is a sooty black appearance of fat, or materials which have been burned for lighting the cave.

The outside of the lamp face on which it rests is convex, except in its central part, which is almost flat. It presents an engraving reminiscent in an astonishing way, but in much smaller than those which adorn the walls of the Grotte de la Mouthe. Indeed, this drawing represents the head seen in profile, of an ibex, a remarkable detailed head: nose, mouth, eyes, ears, horns of a considerable length (they measure up to 12 cm for one and the other 13 cm) and strongly curved in a semicircle.

The oval head measures 35mm in length and its greatest width is 23 mm. Two lines are drawn to indicate a fairly long neck. The body and legs of the animal are not drawn.

La Mouthe

A modern photo of the base of the lamp from Grotte de la Mouthe.


Lamp Lamp


Three views of a lamp from Jamblancs, early Magdalenian.

Photo: Don Hitchcock 2008, 2014
Source: Originals on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lamp Lamp

Lamp carved in limestone. It would have had fat or oil in the depression, with a twist of moss or string for a wick.

This lamps was recovered from La Madeleine. Some lamps have had a lot of care lavished on finishing them off, others are very much a case of utility before beauty.

Photo: Don Hitchcock 2008, 2014
Source: Originals on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lamp Lamp

Lamps from la Madeleine.

Photo: Don Hitchcock 2014
Source: Originals on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lamp Lamp

Lamps from la Madeleine.

Photo: Don Hitchcock 2014
Source: Originals on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lamp Lamp

Lamp from la Madeleine.

Photo: Don Hitchcock 2014
Source: Originals on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lamp Lamp

Lamp from la Madeleine.

Photo: Don Hitchcock 2014
Source: Originals on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Gabillou lamp

Lamp from Gabillou.



(left) Lamp from Laugerie Haute, Maury's excavations

(right) Lamp from Laugerie Haute East, Peyrony's excavations

Photo: Don Hitchcock 2014
Source: Originals, le Musée National de Préhistoire, Les Eyzies-de-Tayac


(left) Lamp from Laugerie Haute East, Magdalenian

(right) Lamp from Laugerie Haute, Middle Magdalenian, Peyrony's excavations

Photo: Don Hitchcock 2014
Source: Originals, le Musée National de Préhistoire, Les Eyzies-de-Tayac

Gabillou lamp

Lamp from Gabillou.

Photo: Don Hitchcock 2014
Source: Original, le Musée National de Préhistoire, Les Eyzies-de-Tayac


Lamp from la Faurélie, Upper Magdalenian.

Photo: Don Hitchcock 2014
Source: Original, le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lamp Lamp


This early Magdalenian lamp is from Couche A, locality 4 of Solvieux.

Photo: Don Hitchcock 2008, 2014
Source: Original on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac


Lamps from la Madeleine. Since la Madeleine is an abri, these would have been the ordinary lamps people had around the place, always in the one position, perhaps in a niche on the wall - they would not have needed handles.

Photo: Don Hitchcock 2014
Source: Original on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Nature of the wick

The wick must, by definition, have the property of absorbing melted fat by capillary attraction and of conveying it to the free end without being consumed too rapidly itself. The form and structure of this wick influence its efficiency. Of the different wicks tested, lichen and moss and then juniper appeared to be the easiest to use. Palaeolithic samples sent to the wood laboratory in Zürich by the Dellucs and ourselves showed the presence of conifers, juniper, and a grass (in one case) and of non-woody residues, but it should be borne in mind that juniper is never completely consumed and so is preserved better than other plants.

Text above: Beaune (1987)

The Lamp of Lascaux - Le Brûloir de Lascaux

Lascaux Lamp

The Lascaux Lamp, found buried in the floor of the Shaft at Lascaux by l'Abbé Glory, is a superb piece of workmanship.


The Lamp of Lascaux - Le Brûloir de Lascaux, was found buried in the floor of the Shaft at Lascaux by l'Abbé Glory, and is a superb piece of workmanship. It is from the Magdalenian culture, 17 000 BP. It can be viewed in the National Prehistory Museum in Les Eyzies-de-Tayac. Shaped like a large spoon made of red sandstone, 8 3/4 inches long by 4 3/16 inches wide and 1 1/4 inches thick, the lamp is finely polished and symmetrical. Its shallow oval cup serves as a receptacle for fuel. The upper surface of the handle is decorated with two abstract signs of chevrons fitted into each other, such as are found painted or engraved in various parts of the cave. When the lamp was discovered, it still contained sooty substances grouped in a circle at the bottom of the cup on a magma of fine dust.These particles were tested and determined to be the remains of a juniper wick used for ignition.

lascaux grease lamp
An oil lamp (a deer fat lamp), found in the sediments in the floor of the Shaft at Lascaux cave in Montignac, Dordogne, Aquitaine, France. Magdalenian culture, 17 000 BP. It can be viewed in the National Prehistory Museum in Les Eyzies-de-Tayac.

The red sandstone lamp was found by Abbé André Glory at Lascaux. André Leroi-Gourhan, said in 1982 that Abbé Glory was the man who best knew Lascaux.

Photo: Don Hitchcock 2014
Source: Original on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lascaux lamp

The lamp is just as beautifully completed on the back as the front. Note the layers of sandstone symmetrically circling the bowl of the lamp.

This is a masterwork.

Photo: Don Hitchcock 2008

Source: Original on display at Le Musée National de Préhistoire, Les Eyzies-de-Tayac

Lascaux lamp Lascaux lamp

Another photo of the lamp showing the carbonaceous deposits in the bowl at the time of discovery.

Photo: © Ministry of Education, Culture and Sport

Lascaux lamp

A relatively recent photograph of the lamp.

Photo: Beaune and White (1993)

Lascaux lamp

Another photograph of the lamp.


This is by no means the only lamp found at Lascaux. Beaune and White (1993) say that many were found lying together in groups at Lascaux, 70 altogether. This is a staggering figure.

Text below from Eshleman (2003)

Glory's most spectacular find in the Shaft was a lamp (bruloir ) in a ground layer below the tail of the rhinoceros. "Shaped like a large spoon made of red sandstone, 8 3/4 inches long by 4 3/16 inches wide and 1 1/4 inches thick, the lamp is finely polished and symmetrical. Its shallow oval cup serves as a receptacle for fuel. It has a capacity of two fluid ounces. The upper surface of the handle is decorated with two abstract signs of chevrons fitted into each other, such as are found painted or engraved in various parts of the cave."

When the lamp was discovered, "it still contained sooty substances grouped in a circle at the bottom of the cup on a magma of fine dust" These particles were tested and determined to be the remains of a juniper wick used for ignition.


Alain Roussot and André Glory at Lascaux, 1953

(Note that in this photo André Glory is using a translucent medium directly on the surface of the cave, in order to trace the image - Don)


Apse, Lascaux
Abbé Glory in the apse.

Abbé Glory worked in Lascaux's engraved areas - the Passageway, the Apse, the Nave, and Chamber of Felines - from 1953 to 1962, identifying some 1433 figures. Standing, or lying on his back on scaffolding, Glory used over 140 square metres of tracing paper, recording the chaos under it, which then had to be deciphered and identified.

His work was rechecked and nearly always found to be accurate by some of the contributors to the only book that examines in detail Lascaux's engravings:

Lascaux inconnu, Leroi-Gourhan et Allain (1979).

It appears that the apse was originally painted with large animals and that to reach the ceiling (anciently formed by a whirlpool of water) saffolding had to be used.

When the Magdalenians were in Lascaux, this ceiling was nine feet above the floor. Since the cave's discovery in 1940, this floor had been lowered four feet. The south, west and north walls of the Apse are circular. To the east it opens into the end of the Passageway and the beginning of the Nave.

Right above the original floor line are the remains of several large black aurochs, one after the other. Their head turn towards the enlarged opening in the west wall leading to the Shaft. Above the aurochs are traces of painted stags, including a composition of five on the south wall. On the ceiling, seeming to make use of the bulging, circular limestone formation, are two large horses, one red, the other yellow.

Photo and text: Eshleman (2003)

Le brûloir de Lascaux

Glory (1961)

Translated by Don Hitchcock

This is a very important paper which deserves wider distribution.

The stratigraphy of the basement of the Shaft at Lascaux is not significantly different from that of the upper galleries of the cave. There would have been occasional extra depositions of handfuls of soil or loose stone from time to time as people used ropes or rope ladders to climb in and out of the Shaft, dislodging dirt and stones as they did so.

These deposits fell to the bottom of this eight metres deep hole, but did not take on the appearance of layers, but simply increased the thickness of the deposits. Carbon 14 analyses, in progress at the moment, will not make much difference to the coupe of the excavations at the Shaft:

bruloir lascaux

A drawing indicating some of the layers found, and a photograph of the dig in progress, with the lamp in situ.

This sort of historical data is invaluable.

Photo: Glory (1961)

Layer 1
Dark brown clay, compacted. Soil from the discovery of the cave in 1940. Thickness: 0.05 metres. (50 mm)

Layer 2
Homogeneous sandy clay, loose and sterile. Thickness: 0.10 metres. (100 mm )

Layer 3
Complex archaeological layer, including various soil beds formed as lenses of brown clay enclosing flint, mineral colourants, abundant and voluminous charcoal fragments, some animal bones, limestone plaques blackened with burnt wood. The earth surrounding it is a sandy clay. Thickness 0.15 to 0.20 metres. (150 to 200 mm)

Layer 4
Light coloured bed of sandy clay. Thickness: 0.05 to 0.10 m. (50 to 100 mm)

Layer 5
Paleosol very thin, existing only in places, revealed by a linear brown horizon, dotted with rare carbonaceous granules. Thickness 0.005 to 0.01 m. (5 to 10 mm)

Layer 6
Sub-soil of clay platelets with the interposition of beds of white sand brought by water. Thickness about 3 metres.

The subsoil of the upper galleries offers, with more precision, the same superposition of layers which are thinner and more packed, but with some important differences:

Layer 1, at the entrance, the Hall of the Bulls, the Passage, and the Nave, is replaced by a stalagmite layer of which carbon at the base has been dated to 8 200 BP.

This rests directly on a sterile layer, Layer 2, 0.06 to 0.08 metres thick. (60 to 80 mm)

Layer 3, 0.03 to 0.05 metres thick (30 to 50 mm) is compact and laminated. Over a thickness of 10 mm, we counted with a binocular microscope a dozen varves, that is, a dozen different soils successively compacted by the passage of people in twelve periods of occupation followed by eleven periods of absence. The granules of charcoal, which were incorporated, were dated at 17 000 BP in a Dutch laboratory. While awaiting the results of ongoing analyses, we date, by extrapolation, the similar layer of the Shaft layer 3 at 17 000 years. A first analysis in Chicago in 1951 gave an age of 15 566 years BP.

This, then, is the basis of this soil formation No. 3. On the 8th of July 1960, at a depth of 300 mm and 370 mm from the left (north) rock face, appeared on the pale yellow background of the working surface, veined with brown and black streaks, a flat section of a dark red patch.

I thought that this indicated the presence of a small amount of red ochre, such as I had already found there, at the foot of the famous painted panel of a wounded bison threatening a prone man, perhaps used on this panel.

In order to avoid damaging it, I excavated it from below, and realised that it was a small baton in sandstone, and tried to follow the outline with my fingertips in order to know its length. Thus was laid gently on the palm of my hand an oblong object which I carefully extricated from its coating.

It was what is commonly called a lamp. The bottom of the bowl still contained black fragments of combustion, which apparently had not suffered displacement, despite its unexpected mode of extraction.

bruloir lascaux bruloir lascaux

(left) The lamp, seen in profile, shows the curvature of the cup and the black smudge on the outer edge of the cup.

(right) The top view of the lamp shows the stain on the margin, the disposition of the carbonaceous deposits and the three engraved signs.

Photo and text: Glory (1961)

The whole upper part of the object left a strong imprint on the roof of the pit in the ground which had sheathed it. I had a photographic record taken by a Périgueux professional photographer. We had the chance to have a fresh artefact, on which one could immediately obtain some scientific details, difficult to detect on an object that has lost its dehydrated dust layer. M. Moreau, Deputy Director of the Biological Station of Eyzies, kindly allowed us access to his laboratory to perform tests and optical micrographs. We completed the taking of photographs in colour and black and white prints.

bruloir lascaux

This is the cast of the top of the lamp.

Photo: Glory (1961)

We completed all the measurements including weighing, the study of the engraving on the lamp, and putting the carbonaceous matter into a test tube in order to control the rate of its dehydration before examination in our laboratory at Bugue. We replaced the lamp in its first earth mould; then with our assistant, J.-L. Villeveygoux, we replaced the earth in the hole made in the sediments, in order to place the lamp exactly at the former level and closely examine its context in the soil.

It is curious to note that if we had proceeded at the outset in this way, we might unwittingly destroyed our research opportunities. Would we have thought about not touching the surface of the object with a brush , which would have destroyed the carbonaceous layer of the surface and scattered the ashes in the spoon? Even worse, if we had used the method of compressed air, as used in some other excavations, we would have destroyed all the fragile evidence of combustion.

The object is in the form of a racket, that is to say an oval cup extended by a handle which has a more or less semi-cylindrical cross-section. The flat of the handle is decorated with three sets of engravings: a long axial barbed arrow, with the barb near the cup, though not touching the arrow itself, is framed on either side and at the extremes by two similar stylised motifs; two pairs of slanted lines, chevrons truncated at the angle, are followed by two others on the other side at the end of the handle, with the larger chevron partly extending inside the other. In each case, one of these chevrons is twice the length of the other of the pair. L'abbé Breuil saw two horns and two ears of stylised animal heads. (presumably with the shorter chevron forming the ears, and the longer the horns - Don )

bruloir lascaux

Drawing and measurements of the artefact.

Photo: Glory (1961)

I know we can compare these drawings to five drawings of stylised heads of ibex engraved on a chisel-pusher or ciseau-poussoir, (used for pressure flaking of flint - Don ) of deer antler from Gourdan (Haute-Garonne), but they are from the final Magdalenian. One could also evoke very similar chevrons on a Lortet baguette in the Museum of National Antiquities, interpreted by l'abbé Breuil as arêtes (mountain ridges) or as the tails of fishes. If you put these drawings of Lascaux in context, I am inclined to see the characteristics of the feathers on darts used for hunting. Similar signs were engraved on horses in the Apse and also painted on horses in the axial Gallery; so far prehistorians have identified them as arrows.

Its dimensions are:

total length: 224 mm
maximum width: 106 mm
neck width: 47 mm
total thickness: 32 mm
spoon: 84 mm x 75 mm
depth: 17 mm.
weight in July, not dehydrated: 560 grams.


Ciseau-poussoir in deer antler. Magdalenian.

Schematic engraving of the head of a horse and some ibex seen from the front. Grotte de Gourdan (Haute-Garonne) 147 x 27 mm

Photo: Chollot (1962)

Lortet baguette Lortet baguette
Lortet baguette referred to above.

Engraved bone from the middle Magdalenian, discovered by Edouard Piette in 1902 at Lortet, Hautes-Pyrénées. The two sides are decorated with nested chevrons.

Length 84 mm, width 12 mm, thickness 11 mm.

Source: Saint-Germain-en-Laye; musée d'Archéologie nationale


This lamp has been fashioned from a block of fine sandstone, without visible inclusion of hematite, quartz grains and mica flakes. It seems that this sandstone was selected for its purity, delicacy, and malleability. These Permian formations are exposed at the fault of Saint-Basile-de-Meyssac near Brive, at Terrasson, not far from Montignac.

In the same outcrop, the consistency and colour of the rock come in a number of variations, as we were able to see later at the surface exposure of the rock, at Meyssac and at Gollonge-la-Rouge. One can not take the first stone which one finds, one must select from a number of possible raw pieces of stone. It is possible, too, that pebbles of a good size, worn and flattened by erosion, were collected and selected from Quaternary alluvium or in the present bed of the Vézère.


It would have been easier and more traditional to carve a circular cup, the technique of which, long known in the typical Aurignacian, and was successfully applied to batons with cups. Lamps in the same form of a racket from Scilles, Marsoulas (Haute-Garonne), Goual (Lot), Laugerie-Haute (Dordogne), have substantially circular cups or spoons. Others, from Laugerie Haute, Mouthiers, Bois du Roc (Charente), were equipped with oval shaped cups, which should not be confused with the lamps from la Mouthe (Dordogne) and Grand Moulin (Gironde) which are ovoid ( egg shaped - Don ). Is this a conventional layout, functional or artistic? The (exterior ) oval bowl of the lamp of Lascaux is an almost perfect geometrical figure, of which the carving, according to craftsmen, has been done directly into the mass of sandstone. It was only then, according to its outline, that the edge of the rim would have been created, and the cup itself excavated by pecking and scraping, of which there are still many traces. This logical method has allowed the artist to place the handle axis in the extension of the minor axis of the bowl and draw the two sides as an identical curve. When, on a transparent film, by folding, we superimpose the two sides of the handle, you realise that all points of the two curves coincide.

Some colleagues, like M. Delporte, believe that the eye of the palaeolithic artist was better than the eye of a modern technician, and that for the sake of beauty of form, he obtained an astonishing precision. According to sculptors and skilled artisans whom I interviewed, this geometric regularity on a hard sandstone block could only be obtained by using a measuring device, like a wooden ruler or notches in bones, or the use of a reversible pattern. The two theories are not opposed, as the manual skills of a true craftsman are always complemented by superior visual ability, the one controlling the other during the operation. There are several examples in Lascaux, in the engravings of signs, which might be called geometric motifs: four poles stacked at equal intervals, a series of symmetrical barbs.

As an experiment, we drew, without points of reference, a sketch of the lamp of Lascaux, then we retouched the edges until the eye judged it to be a sufficiently symmetrical drawing, and obtained a precision equal to that of Lascaux, which indicates that the artist was able to complete the task without a measuring instrument.


The layout of the shape, the release of the shape by pecking, seeking smooth contours and a balanced keel by scraping, polishing the surfaces and the reductions of ridges, are the various operations that study of the shaping has revealed with this object.

1. The plot of the contour has disappeared during shaping, but we can still see the influence of the burin on the piece on the straight line at the end of the arrowhead near the cup and by the bold line of the shaft of the arrow. This outline could also have been marked by a sequence of pecked points.

2. The object has kept the trace of two pecking operations: a final pecking, closely spaced, on the edge of the cup, partially offset by polishing, and a last minute pecking, deeper, overriding the polishing, at the bottom end of the handle.

3. Obliquely oriented scrapings, orderly and measured, on the periphery of the cup to give a visually pleasing contour, explain the smooth profile of the keel, and the impeccable success of the semicylindrical, slightly ellipsoidal, cross section of the handle.

4. The polishing was applied to three kinds of surfaces: the flat parts, the rounded parts and the angular parts. The upper flat surface has been flattened in two stages, for the plane of the cup is not an extension of that of the handle; the discrepancy is a difference of 2 degrees. Both operations were done on a very planar grinding stone, because the crown of the rim is smooth and distortion-free. The plane of the handle shows a slight elevation of the top left edge, to the cup, of approximately 2 mm, which may indicate that the piece was ground back and forth on the grindstone, the workpiece held parallel to the body.

The inside of the bowl has a perfect polish. The internal curvature, near the handle, is steeper as a result of more abrasion. The rest of the contour has the same concave curve. The bowl is shallow (17 mm), and is fairly open. It has a capacity of approximately 60 millilitres. The circular rim has an average width of 12 mm. , The left portion ranges from 11 millimetres to 12 millimetres while the right is 13 mm.

(Note that Glory's text contains a typo which gives the bowl of the lamp a capacity of 60 centilitres, which is 600 cubic centimetres, or 600 ml, or well over two cupfuls, which is too large by a factor of 10. Simple mathematics on the dimensions of the bowl will attest to this. If we take the radius of the bowl to be an average of 4 cm, and depth on average 1.2 cm, we get a volume of 60 cc, or 60 ml.

If we look at the excellent summary of ice age lamps by Beaune and White (1993), we find a similar typo with an error of a factor of ten the other way - the text says 'The largest bowls can hold about 10 cubic centimeters of liquid' which is too small by a factor of 10 - Don

5. The end of the handle has not been broken. Its steep angle (approximately 60°) represents the natural surface of the original stone, which does not seem to have undergone the erosive action of water. The lamp would thus have been carved from a block of sandstone found on the surface. The angular portions have not been overlooked. The entire outer rim of both the cup and the handle was rounded, for ease of grip.

Finally, the projection of the inner profile shows a recess of the keel, the bottom of the handle, slightly concave towards the middle of the total length of the object (Diagram, No 2, C). Is it an artistic or functional requirement? Placed on a plane surface, the keel rests at two points of balance, one located under the end of the handle (Diagram, No 2, A), the other under the cup, B. The void formed, a geometric arc with its maximum at C is located almost next to the geometric centre of the lamp, D. We measured the centre of gravity of the object and it is located in EF, 20 mm ahead of the arrow C, although it would have been shifted to the left without the arc cut out of the keel of the handle.

The latter has had the advantage of pushing the centre of gravity towards the junction of the handle and the cup, giving the handle and the cup about the same weight. This becomes obvious when the lamp lies on the open palm of the hand. By a curious coincidence, this is also the ritual gesture of the Egyptian Pharaoh presenting an analogus object to the deity.

Mr. Dunand, director of excavations at Byblos, (Gebal, Gubla, a Mediterranean city in Lebanon now called Jubayl, the oldest continuously inhabited place in the world, founded 7 000 BP - Don ) with no hesitation told us that he saw the lamp of Lascaux as similar to the natron cups, incense burners with handles from the New Kingdom in Egypt , and in Phoenicia during the Persian period; the open hand supports the bottom of the cup while the extended handle rests on the forearm.

(Natron is a naturally occurring combination of sodium carbonate and sodium bicarbonate. Natron occurred in great quantities in antiquity on the banks of several ancient lakes, mainly those of the Wadi Natrun in Egypt, but was also found elsewhere. Natron was mainly used for purification purposes, both in daily life and in the religious sphere. Numerous rituals are known in which natron played an important role, including mummification. It was also used to make incense, glass and glazes - from )

Workshop and period

A number of models of similar design in pink sandstone have already been described in the literature in Charente (Bois du Roc, Mouthiers), Dordogne (Laugerie-Haute), in Lot (Coual), Haute-Garonne (Grotte des Scilles, Marsoulas), but none has survived in such good condition. The lamps from Charente and from Dordogne are shaped so that they suggest a centre of manufacture in the Corrèze region or in the Dordogne in the zone to the East, North East, from Châtre to Terrasson.

Lascaux lamp

Ice Age lamps have been found primarily in southwest France (left). Lamps appear in all eras of the Upper Paleolithic (40 000 to 11 000 years ago); more of them have been recovered from the later periods. Surprisingly, most larnps have been retrieved not from deep caves but from open air sites and from under rock shelters (right).

Photo and text: Beaune and White (1993)


Lamp carved in pink sandstone from Abri Lachaud, probably made from the same pink sandstone as the Lascaux lamp.

Abri Lachaud is the third of the Pouget (or Pouzet) abris, at Terrasson, Dordogne, half way between Montignac and Brive on the Vézère.

Abri Lachaud is a small Proto - Magdalenian cave behind a large collapsed shelter on the side of a steep hill overlooking the valley of the Vézère, less than thirty kilometres from Lascaux.

Cheynier (1953) suggests that the lamp was probably broken ritually.

Photo: Cheynier (1953)


The perfection of this Lascaux pink sandstone lamp's size, the choice of material, the selection of the block used, the search for a finished objet d'art, are not the work of an amateur, but a professional who had to have a rather unusual talent in working sandstone to be successful in making a second lamp of the same style as the first.

Indeed we found, in sifting through the rubble of the first excavations of the Shaft, a scrap of red sandstone 30 millimetres long, the lower left edge of another cup. The width of the rim which also widens , the curvature of the edge of the bowl, with the same thickness, juxtaposed so perfectly to the first find, that the dimensions would be very similar.

Photo and text: Glory (1961)

Objects in pink sandstone appear in the typical Aurignacian, to be followed by the Solutrean at the Grotte de Thévenard near Brive in Corrèze (fouilles Bouyssonie). However the handle cross section, and the raquette shape found here are Solutrean like those of Laugerie-Haute in the Dordogne, displayed at Musée Maury, near Les Eyzies (these displays are probably now in the Musée national de Préhistoire at Les Eyzies - Don ). This type exists in the Magdalenian (without harpoons) especially at the Magd. III of Bois du Roc, Mouthiers in Charente, at Marsoulas and at Lespugue in Haute-Garonne.

Quaternary Lamps

The principal Quaternary lamps
1 La Mouthe (Dordogne) Emile Rivière. Sandstone. The reverse carries an engraving of the head of an Ibex. Height 170 mm, width 120 mm.
2 Grotte du Coual (Lot). Félix Bergougnous. 250 mm x 150 mm. Sandstone.
3 Grotte des Scilles, at Lespugue (Haute Garonne) Comte de Saint-Perrier. Sandstone. 190 mm. On the bottom a rudimentary horse's head.
4 and 5 Anval (Puy de Dôme), Dr Baudon. Trachyte, 140 mm and 130 mm.
6 Grotte de Thévenard (Corrèze). Abbés Bardon and Bousyssonnie. Red sandstone. 130 mm. Lightly engraved ruminant.
7 Grotte des Fadets (Charente). Collection Maret. Red sandstone. 50 mm.
8 Bois du Roc (Charente). Fermont. - Engraving of a stylised fish. Sandstone. 170 mm.
9 Mouthiers (Charente). Trémeau de Roche-Brune. Sandstone. 140 mm.
10 Pair-non-Pair (Gironde). Daleau. Limestone 110 mm.
11 Grotte de la Mairie, à Teyjat (Dordogne). Bourrinet. Bottom of a lamp of sandstone with the head of a Reindeer. 130 mm.
12 Grotte des Harpons, at Lespuge. Comte de Saint-Perrier. Limestone. 60 mm.

Photo and text: Viré (1934)

Scilles Lamp

Lamp from the grotte des Scilles (Haut-Garonne) Middle Magdalenian. MAN. fine grained slightly micaceous soft red sandstone. Length 200 mm. max width 108.2 mm. bowl diameter 101.6 mm. depth 22 mm.

Photo and text: Beaune (2003)

Scilles Lamp

Lamp from the grotte de La Mairie, Teyjat (Dordogne). Upper Magdalenian. MAN. no 52456. Limestone. Diameter 102 to 108 mm, height 105 mm, diameter of bowl 82.3 mm, depth 30.8 mm, Drawings after Bourrinet (1908), (in Capitan et al. (1908)) , photos S. A. de Beaune.

Photo and text: Beaune (2003)

lamp Pilier or Codon or Domme lamp Pilier or Codon or Domme

The lamp from Grotte du Pilier, also known as the lamp from Codon, since the Grotte du Pilier is part of this cave complex. The cave is near Domme on the left bank of the Dordogne, and the lamp is also called the Domme lamp. There is a reindeer head engraved on the right margin of the lamp.

The lamp of Pilier / Codon / Domme has been made from Late Cretaceous limestone, selected from a hard grained sample.

Its color is sandy yellow tinged with pink on the sides of the upper face, in three places. On the underside, the two colours are divided into two irregular zones of substantially equal area. The general shape is hexagonal, with one side removed; the bowl is slightly eccentric in the longitudinal direction; the upper and lower surfaces are not absolutely flat, especially the lower, which is rather curved.

Dimensions. - The greatest length, nearly axial, reached 120 mm and the width varies from 85 to 115 mm, the thickness, variable, does not exceed 48 mm. The cup, almost circular, has diameters between 55 mm and 58 mm, while the depth at the centre reaches 14mm. The weight is 1035 grams.

Photo and text: Bastin et al. (1940)


Without much discrimination, European inventories grouped, whole and broken, 70 scoops, stemmed cups, grooves in shell, cups with spouts, plates with a depressed centre, etc.., under the generic name and misnomer of 'lamps'. The responsibility lies with Alphonse Trémeau Rochebrune who in 1865 discovered in the Grotte de Mouthiers (Charente) the first cut red sandstone, for which the edge of the curve opposite the handle bore traces of charcoal.

Since then, D. Peyrony noted black residue and traces of fire in the bottom of limestone cups from La Madeleine, R. Saint-Perier saw black marks on the edge of a curved sandstone fragment from Isturitz, Dr. Cheynier, on the edge of a large shell fossil at the site of Badegoule (Dordogne). There were found to be areas reddened by fire on hollowed limestone blocks at Pair-non-Pair. These indications are few in number, whereas the other fifty articles classified as 'lamps' have no signs of the action of fire.

As for the 140 pieces of limestone covered with charcoal ash found by MM. Breuil, White, and ourselves in the earth at Lascaux, are these lamps? The dark matter of the la Mouthe lamp was analysed by the chemist Berthelot and published in the Academy of Sciences in Paris on October 28, 1901, three years after the discovery by E. Rivière. Berthelot recognised burnt carbonaceous material which he interpreted as residues of fat of animal origin. This is a very real possibility, but difficult to analyse in the laboratory, because fats or liquids compounded usually of carbon, hydrogen and oxygen oxidise quickly in the soil and disappear over time.

bruloir lascaux bruloir lascaux
Wood, resins, herbs, plant or animal fats, when consumed, leave black carbonaceous remains, a thin skin or film of carbon deposits, and small clusters of amorphous carbonaceous materials such that a microscopic examination may be able to interpret their condition and conformation. The top left edge of the rim of the Lascaux lamp was covered by a flat, black film, hard and very thin, not overflowing either on one side or the other. However there was a dark-stained smudge extending for a few millimetres on the outer rim of the cup (visible on the photos shown here).

Photo and text: Glory (1961)

bruloir lascaux deposits of carbon
Microphotograph of the carbon deposits on the left margin, x40.

The deposits on the lamp were examined under a binocular microscope on 8th August, 1960, work which was attended by M. Moreau, Deputy Director, M. Gouillard, technician, M. and Mme. Lacorre, and M. J.-L. Villeveygoux. We extract from the laboratory notes the following essentials: 'At 20x magnification, we found a polymorphous blackened material, with flat blisters and a few granules, cracked in every direction in a tightly polygonal network. At 40x magnification, the deposited film surface was dotted with small white crystals (calcite?), and could be seen to be a thin film superimposed on the pink sandstone.'

Photo and text: Glory (1961)

A former fluid, now a solidified stain, was deposited on the sandstone. It either came from seepage from a wick, contact from a resinous stick, or perhaps from dirty fuel. At the bottom of the bowl, the carbonaceous materials occurred in the form of small isolated pellets of filamentous vegetable fibres, and were grouped around the centre of the cup on a mass of fine soot. The entire loose mass, which apparently had not been displaced since the last time it was lit, broke away with a feather-like touch, and was placed in a test tube ready for analysis at the Technical Institute for Wood.

The inside of the cup did not have any trace of the direct action of fire, but was sprinkled with grains of charcoal, and the peripheral ridge was bordered by a strong black mark. After the removal of the film, the margin of the sandstone remained impregnated with carbon. Beneath the lamp, the ancient deposit, sprinkled with granules of charcoal, had no evidence of heated earth.

Based on this data, three interpretations are possible: lamp, incense or smoke burner, smoke or incense burner and light:

a) A horizontal wick made ​​of plant fibres, placed on the left edge of the rim, could shed light if the other end were immersed in some fat soaked plant remains at the centre of the bowl. An axial wick could also rest on its support of a clump of vegetable matter, fuelled by the dripping of fat suspended above the burner (a method used by some Eskimos, according to M. Leroi-Gourhan). The marginal stain is the result of the cleanup of a spill, exacerbated by the holder of the lamp hitting the lamp against the wall, which would explain the crack in the cup and the occurrence of fragmentation of all the cups at Lascaux.

b) Aromatic twigs bundled in groups may have been dropped onto a foundation of hot coals in the bowl. The resins and tars which would emanate from this combustion stained the rim, this staining occurring while draining the hot residues from the cup.

c) Short sticks of aromatic plant fibres could both shed light and emit fragrant vapours for incense or smoke, either placed on the edge of the rim or placed in the centre of the bowl.


Based on known ethnographic and archaeological materials, the defence of the thesis of the object being an incense/smoke burner seems more logical than a lamp, but it is not necessary to go into that fully here. The carbodendrology (identification of the fibres used) does not seem to settle the debate, because we can identify the fibre as a fuel for lighting. The fibres can be considered as the source of coals, on which resinous particles were burnt for smoke. However, the case for an incense/smoke censer includes the following considerations:

1) The number and perfection of engravings and paintings at Lascaux required fairly extensive lighting in all the rooms above the Shaft. Excavations there have uncovered no sandstone lamps, but many limestone flakes covered with cinders of charcoal, which have an artificial cup. Beside this sandstone brûloir, was found an apparatus specially designed to illuminate the bottom of the Shaft, made with these limestone flakes, which we might call 'candlesticks'. The description is irrelevant here, as well as another type of lighting used in the Middle East and found at Lascaux.

2) The 70 'lamps' catalogued in the inventory of Lascaux, if they were all lamps, would not suffice to explain the problem of lighting in the painted caves, the inhabited caves, the rock shelters, or the archaeological sites, whose number exceeds several thousand throughout Europe. This object, if it had been customary, should have much more diffusion.

3) The perfection of the carving and the rarity of these objects, of the type raquette and navette (racket and shuttle (la Mouthe, le Grand Moulin) shaped lamps) barely numbering ten across the whole of Europe, establishes them as a ritual vessel.

lamp Grand Moulin

The lamp from Grand Moulin.

Curiously, its length and width are exactly those of the lamp from Grotte du Pilier, being 120 mm x 85 mm.

The thickness is variable, between 5 and 7 mm.

The almost circular cup measures 85 mm in its large diameter, and 68 mm in its small diameter, and has a depth of 18 mm.

Photo and text: Ferrier (1942)

4) None of these objects have a notch or groove in the edge, to stabilise the wick.

5) The location of discovery of the two 'lamps' at the foot of the famous disemboweled Bison panel, hidden in a hard to reach Shaft, looks like a secret spot reserved for initiates.

6) In the absence of any remains of meals, nor of tool making or the production of material for painting, the deposition of this brûloir or burner laid flat on the floor does not seem to have a utilitarian function.

Instead the quality of the piece, its freshness from a lack of constant use, the orientation of the 'lamp' near the painting of Rhinoceros, the burnt vegetable fibre, and its abandonment in a secluded place, militate in favour of a ceremonial use.

The burner is covered with earth fallen from the top of the Shaft, and has thus escaped the notice of those who frequented these places thereafter.

Abbé Glory.


  1. Bastin, A., Chassaing, J., 1940: Découverte d'une lampe paléolithique à Domme (Dordogne) Bulletin de la Société préhistorique française, 1940, tome 37, N. 10-12. pp. 219-229.
  2. Beaune, S., 1987: Palaeolithic Lamps and Their Specialization: A Hypothesis Current Anthropology, Vol. 28, No. 4. (Aug. - Oct., 1987), pp. 569-577.
  3. Beaune, S., 2003: De la domestication du feu aux premières lampes, Nouveautés Lychnologiques, 2003.
  4. Beaune, S., White R., 1993: Ice Age Lamps Scientific American, March 1993.
  5. Capitan L., Breuil H., Bourrinet P., Peyrony D., 1908: La grotte de la mairie à Teyjat (Dordogne) Fouilles d'un gisement magdalénien., Paris, F. Alcan, 1908.
  6. Cheynier, A., 1953: Stratigraphie de I'abri Lachaud et les cultures des bords abattus Archivo de Prehistoria Levantina , IV, 1953
  7. Chollot, M., 1962: Arts et techniques de la préhistoire, Paris: Ed. Albert Morancé, 1962.
  8. Eshleman, C., 2003: Juniper fuse: upper paleolithic imagination & the construction of the underworld Wesleyan University Press, paperback, 356pp, ISBN-13: 9780819566058, ISBN: 0819566055
  9. Ferrier, J., 1942: Les Lampes du Paléolithique en GirondeBulletin de la Société préhistorique française, 1942, tome 39, N. 3-4. pp. 124-128.
  10. Glory, A., 1961: Le brûloir de Lascaux Gallia préhistoire, Tome 4, 1961. pp. 174-183.
  11. Leroi-Gourhan, A., Allain J., 1979: Lascaux inconnu, Centre National de la Recherche Scientifique, 1979, 381 p.
  12. Rivière E., 1897: La grotte de la Mouthe (Dordogne), Bulletins de la Société d'anthropologie de Paris, IV Série. Tome 8, 1897. pp. 302-329
  13. Rivière E., 1899: La lampe en grès de la grotte de La Mouthe (Dordogne), Bulletins de la Société d'anthropologie de Paris, IV Série. Tome 10, 1899. pp. 554-563.
  14. Smith, R., 1979: An individual-based comparative advantage model: Did economic specialization …, Rutgers The State University of New Jersey - New Brunswick
  15. Viré, A., 1934: Les Lampes du Quaternaire moyen et leur bibliographie, Bulletin de la Société préhistorique française, 1934, tome 31, N. 11. pp. 517-520.

Back to Don's Maps

walls of china Back to Archaeological Sites