Letter | Published:

An abstract drawing from the 73,000-year-old levels at Blombos Cave, South Africa

Nature (2018) | Download Citation


Abstract and depictive representations produced by drawing—known from Europe, Africa and Southeast Asia after 40,000 years ago—are a prime indicator of modern cognition and behaviour1. Here we report a cross-hatched pattern drawn with an ochre crayon on a ground silcrete flake recovered from approximately 73,000-year-old Middle Stone Age levels at Blombos Cave, South Africa. Our microscopic and chemical analyses of the pattern confirm that red ochre pigment was intentionally applied to the flake with an ochre crayon. The object comes from a level associated with stone tools of the Still Bay techno-complex that has previously yielded shell beads, cross-hatched engravings on ochre pieces and a variety of innovative technologies2,3,4,5. This notable discovery pre-dates the earliest previously known abstract and figurative drawings by at least 30,000 years. This drawing demonstrates the ability of early Homo sapiens in southern Africa to produce graphic designs on various media using different techniques.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Data availability

All data generated or analysed during this study are included in the published article and its Supplementary Information.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    d’Errico, F. & Stringer, C. B. Evolution, revolution or saltation scenario for the emergence of modern cultures? Phil. Trans. R. Soc. Lond. B 366, 1060–1069 (2011).

  2. 2.

    Henshilwood, C. S. et al. Emergence of modern human behavior: Middle Stone Age engravings from South Africa. Science 295, 1278–1280 (2002).

  3. 3.

    Henshilwood, C., d’Errico, F., Vanhaeren, M., van Niekerk, K. & Jacobs, Z. Middle Stone Age shell beads from South Africa. Science 304, 404 (2004).

  4. 4.

    d’Errico, F. & Henshilwood, C. S. Additional evidence for bone technology in the southern African Middle Stone Age. J. Hum. Evol. 52, 142–163 (2007).

  5. 5.

    Mourre, V., Villa, P. & Henshilwood, C. S. Early use of pressure flaking on lithic artifacts at Blombos Cave, South Africa. Science 330, 659–662 (2010).

  6. 6.

    Henshilwood, C. S. et al. A 100,000-year-old ochre-processing workshop at Blombos Cave, South Africa. Science 334, 219–222 (2011).

  7. 7.

    Jacobs, Z., Hayes, E. H., Roberts, R. G., Galbraith, R. F. & Henshilwood, C. S. An improved OSL chronology for the Still Bay layers at Blombos Cave, South Africa: further tests of single-grain dating procedures and a re-evaluation of the timing of the Still Bay industry across southern Africa. J. Archaeol. Sci. 40, 579–594 (2013).

  8. 8.

    Villa, P., Soressi, M., Henshilwood, C. S. & Mourre, V. The Still Bay points of Blombos Cave (South Africa). J. Archaeol. Sci. 36, 441–460 (2009).

  9. 9.

    d’Errico, F., Henshilwood, C. S. & Nilssen, P. An engraved bone fragment from c. 70,000-year-old Middle Stone Age levels at Blombos Cave, South Africa: implications for the origin of symbolism and language. Antiquity 75, 309–318 (2001).

  10. 10.

    Vanhaeren, M., d’Errico, F., van Niekerk, K. L., Henshilwood, C. S. & Erasmus, R. M. Thinking strings: additional evidence for personal ornament use in the Middle Stone Age at Blombos Cave, South Africa. J. Hum. Evol. 64, 500–517 (2013).

  11. 11.

    Henshilwood, C. S., d’Errico, F. & Watts, I. Engraved ochres from the Middle Stone Age levels at Blombos Cave, South Africa. J. Hum. Evol. 57, 27–47 (2009).

  12. 12.

    Wadley, L. Ochre crayons or waste products? Replications compared with MSA ‘crayons’ from Sibudu cave, South Africa. Before Farming 2005, 1–12 (2005).

  13. 13.

    Rifkin, R. F. et al. Evaluating the photoprotective effects of ochre on human skin by in vivo SPF assessment: implications for human evolution, adaptation and dispersal. PLoS ONE 10, e0136090 (2015).

  14. 14.

    Joordens, J. C. et al. Homo erectus at Trinil on Java used shells for tool production and engraving. Nature 518, 228–231 (2015).

  15. 15.

    Mania, D. & Mania, U. Deliberate engravings on bone artefacts of Homo erectus. Rock Art Res. 5, 91–107 (1988).

  16. 16.

    Hovers, E., Vandermeersch, B. & Bar-Yosef, O. A Middle Palaeolithic engraved artefact from Qafzeh Cave, Israel. Rock Art Res. 14, 79–87 (1997).

  17. 17.

    Marshack, A. A Middle Paleolithic symbolic composition from the Golan heights: the earliest known depictive image. Curr. Anthropol. 37, 357–365 (1996).

  18. 18.

    Texier, P.-J. et al. The context, form and significance of the MSA engraved ostrich eggshell collection from Diepkloof Rock Shelter, Western Cape, South Africa. J. Archaeol. Sci. 40, 3412–3431 (2013).

  19. 19.

    Henshilwood, C. S. et al. Klipdrift Shelter, southern Cape, South Africa: preliminary report on the Howiesons Poort layers. J. Archaeol. Sci. 45, 284–303 (2014).

  20. 20.

    Rodríguez-Vidal, J. et al. A rock engraving made by Neanderthals in Gibraltar. Proc. Natl Acad. Sci. USA 111, 13301–13306 (2014).

  21. 21.

    Hoffmann, D. L. et al. U-Th dating of carbonate crusts reveals Neandertal origin of Iberian cave art. Science 359, 912–915 (2018).

  22. 22.

    Pearce, D. G. & Bonneau, A. Trouble on the dating scene. Nat. Ecol. Evol. 2, 925–926 (2018).

  23. 23.

    Quiles, A. et al. A high-precision chronological model for the decorated Upper Paleolithic cave of Chauvet-Pont d’Arc, Ardèche, France. Proc. Natl Acad. Sci. USA 113, 4670–4675 (2016).

  24. 24.

    Sadier, B. et al. Further constraints on the Chauvet cave artwork elaboration. Proc. Natl Acad. Sci. USA 109, 8002–8006 (2012).

  25. 25.

    Pike, A. W. et al. U-series dating of Paleolithic art in 11 caves in Spain. Science 336, 1409–1413 (2012).

  26. 26.

    Wendt, W. E. ‘Art mobilier’ from the Apollo 11 Cave, South West Africa: Africa’s oldest dated works of art. S. Afr. Archaeol. Bull. 31, 5–11 (1976).

  27. 27.

    Aubert, M. et al. Pleistocene cave art from Sulawesi, Indonesia. Nature 514, 223–227 (2014).

  28. 28.

    d’Errico, F. et al. The technology of the earliest European cave paintings: El Castillo Cave, Spain. J. Archaeol. Sci. 70, 48–65 (2016).

  29. 29.

    Siegel, S. & Castellan, N. J. Jr. Nonparametric Statistics for the Behavioral Sciences (McGraw-Hill, New York, 1988).

  30. 30.

    Brown, C. A. in Characterisation of Areal Surface Texture (ed. Leach, R.) 129–153 (Springer, Berlin, 2013).

Download references


Partial funding for this research was provided to C.S.H., K.L.v.N. and F.d’E. by the Research Council of Norway through its Centres of Excellence funding scheme, Centre for Early Sapiens Behaviour (SapienCE), project number 262618; to C.S.H. by a South African National Research Foundation Research Chair (SARChI) at the University of the Witwatersrand and the Evolutionary Studies Institute at the University of the Witwatersrand, and the University of Bergen, Norway; F.d’E., L.D. and A.Q. by the LaScArBx, a research programme supported by the ANR (ANR-10-LABX-52). We thank C. Foster for the image in Fig. 2; P. Keene for assistance in the Cape Town laboratory, I. Svahn for assistance with electron microscopy in Bordeaux, G. Devilder for his input on Fig. 3 and M. Haaland for his stratigraphy image on Fig. 1.

Reviewer information

Nature thanks J. C. A. Joordens, G. van den Bergh and the anonymous reviewer(s) for their contribution to the peer review of this work.

Author information


  1. SFF Centre for Early Sapiens Behaviour (SapienCE), University of Bergen, Bergen, Norway

    • Christopher S. Henshilwood
    • , Francesco d’Errico
    •  & Karen L. van Niekerk
  2. Evolutionary Studies Institute, University of the Witwatersrand, Johannesburg, South Africa

    • Christopher S. Henshilwood
  3. CNRS UMR 5199, University of Bordeaux, Bordeaux, France

    • Francesco d’Errico
    • , Laure Dayet
    •  & Alain Queffelec
  4. Laboratoire TRACES UMR 5608, Université Toulouse Jean Jaures, Toulouse, France

    • Laure Dayet
  5. Unité d’Anthropologie/Laboratoire Archéologie et Peuplement de l’Afrique, Geneva, Switzerland

    • Luca Pollarolo
  6. School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Johannesburg, South Africa

    • Luca Pollarolo


  1. Search for Christopher S. Henshilwood in:

  2. Search for Francesco d’Errico in:

  3. Search for Karen L. van Niekerk in:

  4. Search for Laure Dayet in:

  5. Search for Alain Queffelec in:

  6. Search for Luca Pollarolo in:


C.S.H. and K.L.v.N. directed the excavations at Blombos Cave. C.S.H., F.d’E. and K.L.v.N. planned the methodology for examination of L13, and conceived and carried out the experimental replication tests. F.d’E. and L.D. carried out the microscopic analysis of L13 and experimental lines. L.D. carried out the chemical analyses of L13. A.Q. carried out the tribological analysis of the surfaces of L13 and produced the MP4 video (Supplementary Video) and three-dimensional PDF (Supplementary Data) of L13. L.P. recovered L13 during lithic analysis and recognized its importance. C.S.H., F.d’E., K.L.v.N., L.D. and A.Q. co-wrote the paper. L.P. contributed to editing the final paper.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Christopher S. Henshilwood.

Extended data figures and tables

  1. Extended Data Fig. 1 Microscopic examination and chemical analyses of the juxtaposed patches of red deposit that form the drawn lines on L13 and the smoothed surface of the silcrete flake.

    The lines consist mainly of fine-grained iron oxide (Fe), that were applied to the surface, as no haematite occur naturally in the silcrete raw material of L13. A, Photographs (left) and SEM–EDS images (right) of the red lines of the surface of L13. In the subpanels of A, images in a, b show lines 2 and 3; c, d show lines 5, 8 and 9; and e, f show lines 6 and 9 and red spots on a flake scar. The white rectangles in a, c and e indicate the areas that are enlarged in subpanels b, d and f, respectively. Notice the white appearance of the lines in the back-scattered electrons SEM–EDS images due to the presence of iron rich deposits. B, SEM–EDS images (back-scattered electrons) of line 2 and 5. Subpanels ac show line 2; df show line 5. White squares indicate areas that are enlarged in the image with the corresponding letter. The rectangles and black/white circles in subpanels c, f show differences in elemental composition between the drawn lines (light areas) and the silcrete surface (dark areas). C, Raman analysis of line 4. Subpanel a shows a photograph with the location of the analysed area (white rectangle). Subpanel b shows the analysed spots and identified minerals. Subpanel c shows Raman spectra and micrographs of the analysed areas with peaks identifying haematite (red numbers) and quartz (black numbers).

  2. Extended Data Fig. 2 Microscopic examination and chemical analyses of microresidues.

    The microresidues on the smoothed silcrete surface outside of the lines differ in Fe content from the red lines, which—along with the presence of microstriations—supports the theory that the silcrete flake was part of an ochre grindstone before the drawing was made. A, Photographs and micrographs of the lines drawn on L13. Black squares in subpanels a, c, e indicate the areas enlarged in the adjacent subpanels b, d, f. Red residues are clearly visible on the matrix and on quartz grains. f, Black lines highlight superficial randomly oriented striations. B, SEM–EDS analysis of the silcrete outside the drawn lines. In subpanels a, c, d, black and white squares indicate the areas enlarged in the adjacent photograph. In subpanels e, f, the analysed spots (black squares and circles) identify the presence of isolated iron-rich particles on the surface of the matrix and the quartz grains. C, Raman analysis of microresidues preserved in quartz grain pits. Subpanel a shows a photograph with the location of the analysed area (white rectangle). Subpanel b shows analysed spots (white squares) and identified minerals. Subpanel c shows Raman spectra and micrographs of the analysed areas, with peaks identifying haematite (red numbers) and quartz (black numbers). The area shown in subpanel b of panel is the same as the area shown in subpanel b of panel B.

  3. Extended Data Fig. 3 Results from experimental marking of silcrete surfaces with ochre paint of varying viscosities and with an ochre crayon, and subsequent rinsing.

    A, Micrographs of experimentally painted lines before and after rinsing. Subpanels ac show lines produced by applying a liquid (a), viscous (b) and very viscous (c) paint with a thin wooden brush on a silcrete surface. Subpanels df show the three-dimensional rendering of the same lines showing the surface topography. Subpanels gi show the same lines after gently rinsing the surface of the silcrete under running tap water. B, Lines produced experimentally on a silcrete flake with an ochre crayon. Subpanel a shows a single-stroke line drawn from the top to the bottom. Subpanels b, c show close-up views and three-dimensional renderings (subpanels d, e) of selected areas of subpanel a. Subpanels h, i show a photograph (h) and three-dimensional rendering (i) of a single-stroke line produced from the top to the bottom after gently rinsing the silcrete flake under running tap water. Subpanels f, g and j are depth maps and sections of b, c and a selected area of h, respectively. The locations of the sections are indicated on the depth maps by white bars. White arrows indicate deposits of powdery ochre preserved in recesses, and larger yellow arrows indicate prominent areas with compacted ochre deposits covered by striations. Compacted patches of ochre covered by striations and small deposits of ochre powder in recesses are preserved after the rinsing; these features and lines are similar to those on L13 (Fig. 4).

  4. Extended Data Fig. 4 Lines produced experimentally on silcrete flakes.

    These images show that, as it is difficult to exactly superimpose a new line on a previous one, superimposing a line on a previous line generally results in a wider line. Unidirectional, superimposed lines retain the same features observed on a single-stroke line. Multiple lines produced by a to-and-fro movement of the ochre edge show microscopic evidence that the crayon was moved in both directions. a, Straight single- (left) and five-stroke line (right) produced from the top to the bottom. b, Curved single- (left) and five-stroke line (right) produced from top left to the bottom right. c, Straight single- (left) and five-stroke line (right) produced by a to-and-fro motion. The lines in this figure were not rinsed with water.

  5. Extended Data Fig. 5 Experimental marking of silcrete flakes with a variety of ochre crayons.

    The morphology of lines will depend on the properties and composition of the ochre, the roughness of the silcrete surface, the pressure exerted and the morphology of the ochre area in contact with the silcrete. In general, soft, plastic, clayish ochre will produce thicker and more continuous lines than silty or sand-rich ochre. Lines on fine-grained silcrete will be better defined than those on coarse silcrete. Stronger pressure will produce comparatively wider, thicker and better defined lines. Six lines made with each of eight unmodified ochre crayons had a maximum width ranging from about 0.9 to 3.3 mm. Lines produced with a pointed ochre crayon tend to be wider and more variable in width than those made with a linear edge. The width of lines made with a linear edge is strongly correlated with the maximum width of the facets on the ochre piece. By contrast, no correlation is observed between the lines made with pointed crayons and the maximum width of the facets on the crayon. The width of the lines on the drawn cross-hatching present on L13 is comparable with that of the experimental lines. The range (1.8–2.9 mm) of this width best fits the width variability observed when marking the silcrete with a pointed crayon rather than an edge. This indicates that a pointed ochre crayon was used to produce the cross-hatching and that the facet of the crayon in contact with the silcrete was about 1.3–2.9 mm wide. A, Correlation between the width of lines and the width of the resulting facets on eight experimental ochre crayons. Subpanel a shows results from crayons with pointed active areas. Subpanel b shows results from crayons with linear active areas. The grey bars indicate the width of lines on L13. B, Wear facet appearing on the natural surface of an ochre crayon after a single stroke (subpanel a) and five strokes (subpanel b).

  6. Extended Data Fig. 6 Three-dimensional rendering of microscopic areas of L13 and silcrete flakes from BBC.

    Three-dimensional rendering shows flattening of the surface of L13 with the drawing, dissolution of the matrix between quartz grains on the cortex of the BBC silcrete flakes and an unworn appearance of the other surfaces of L13 and the ventral aspect of the BBC silcrete flakes. A, Roughness analysis of L13 and MSA silcrete flakes from BBC. Subpanels ad show three-dimensional renderings of a selected area of the surface of L13 with the drawing (a), other surfaces of L13 (b), knapped (c) and cortical (d) surfaces of BBC silcrete flakes. Subpanel e shows box plots of the variation of roughness variables Sq and Sdr, and a bi-plot correlating these two variables. Notice the high degree of smoothness of the surface of L13 with the drawing relative to the other surfaces. A Kruskall–Wallis multiple comparison test demonstrates that Sq, Sdr and Spc on the surface with the drawing are significantly lower (P < 0.01) than those measured on the remainder of the analysed surfaces of L13. B, Areal fractal analysis confirms a clear difference in roughness between the surface with the drawing and other surfaces of L13. This is consistent with the interpretation of the wear on the surface with the drawing as being produced by grinding activities before the drawing occurred. C, Analysis of L13 with confocal microscopy. Rectangles indicate the locations of the analysed areas on the surface with the drawing (red) and on the other surfaces (green). Letters distinguish analyses conducted with areal fractal analysis on the surface with the drawing (a), a flake scar on the surface with the drawing (b), a flake scar on the dorsal surface (c) from analyses made on the dorsal (d) and ventral surfaces (e).

  7. Extended Data Fig. 7 Ochre and silcrete used in the replication experiments.

    a, Pieces of ochre used experimentally to produce lines on silcrete flakes. b, Silcrete flakes used during the experiments.

Supplementary information

  1. Supplementary Information

    This file contains microscopic examination and chemical analyses; experimental marking of silcrete flakes; ochre deposits from hammer-stone contact; tribological analysis; Tables 1-3: Data on experimental ochre crayons and resulting lines; SEM-EDS analyses on lines and other surfaces of L13; Kruskall-Wallis test of surfaces of L13 and other silcrete flakes.

  2. Reporting Summary

  3. Supplementary Data

    A 3D .pdf model of L13

  4. Supplementary Video

    Video showing a 3D view of L13 and tracing of the drawn lines.

About this article

Publication history







By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.