Identifying and Dating Pigment Deposits on the Blombos Cave Calcite Roof

Project Objectives and Background
In 2009 a red-coloured substance was detected on a portion of the calcite roof at Blombos Cave.
Small samples of the pigment were collected (Fig. 1) and samples of the calcite were removed from
the roof for testing at the University of Bergen. In preliminary tests in 2011 on microscopic
samples the red substance appears to be pigment (ochre - Iron oxide). In-situ examination and 3D
photography of this section of the roof seem to show that the pigment was applied to the roof
perhaps during a period when the cave was occupied during the LSA or the MSA. In 2011 we attempted
to date when the pigment was applied to the roof using U- Th. The results were inconclusive but
favoured a younger date post 2ka. We were also not able to determine whether there was a layer of
calcite overlying the pigment or vice versa.The project at this stage was halted.

Several novel dating methods for rock art have since been developed and tested including AMS dating
of carbonates (see Bonneau et al below) and accurate U-Th dating using laser ablation. We now wish
to thin section and date the roof samples that we have in our Cape Town laboratory in collaboration
with Dr Warren Sharp, Berkeley Geochronology Center, California, with Prof. David Pearce, Wits
University and Dr Adelphine Bonneau, Oxford University.

We wish to therefore apply for an export permit for these Blombos Cave roof samples (Table
1) to be sent to Prof. Henshilwood at the University of Bergen (UiB), Norway

Date Samples
28.12.2010 Roof sample 1 (E4a)
28.12.2010 Roof sample 2 (D4a)
14.12.2011 Samples 1, (PH5), session 5
14.12.2011 Sample 2, (PH1), session 1
14.12.2011 Sample 3, (PH6), session 6
25.03.2009 Sample 4, Crumbs to SEL.

Table 1. Blombos Cave roof samples requested for export

1. Methodology
i. Thin sectioning of roof samples at UiB and analysis of pigment. Conducted by Prof.
Henshilwood at University of Bergen

The roof samples will initially be thin sectioned in our geoscience’s laboratory at UiB to
determine whether there is a clear stratigraphy relating to the deposits of calcite and pigment.
From these thin sections we will analyse the red stains using FTIR spectroscopy and SEM-EDS
microanalysis to ensure the ‘pigment’ is not naturally growing iron-based minerals/substances. A
detailed recording of the individual layers followed by a detailed characterization of the paint layer at different points will be carried out. This detailed characterization will also be conducted on the roof samples that we have without red stains. Once this examination is completed then some of the thin sections will be sent
to Prof. Sharp at the Berkeley Geochronology Center, California and to Dr Adelphine Bonneau, Oxford University

ii. U-Th dating- Conducted by Dr Warren Sharp – Berkeley Geochronology Center,
California.

The thin sections will be sampled using laser ablation and then U-Th dated. This will bracket the
age of the Blombos pigments. With modern ICP mass spectrometry, Berkeley will obtain U-Th dates
with percent-level precision on circa 1 mg of typical cave carbonate, making sampling from such
micro-stratigraphic contexts feasible.
Each visually distinct carbonate layer above and below the ochre will be sampled. This will test
whether U-Th ages preserve stratigraphic order, assessing their reliability, and bracketing the ochre’s age as closely as possible.

iii. 14C ages of thin sections – Conducted by Dr Adelphine Bonneau, University of Oxford, Dr
Ruth Ann Armitage at Eastern Michigan University

Comparing the U-Th ages with 14C ages derived from the pigments themselves will be highly
informative since the assumptions and limitations of the two techniques are quite distinct. Since
the effective range of 14C is limited to c. 50 ka, however, only U- Th dating could resolve ochre
ages contemporaneous with the cave’s MSA occupation.
Although 14C dating may be applied on calcite crusts, it may also be applied to other components in
and/or surrounding the paint layer. The detailed characterization is the only way to know what is
present and what can be used or not.
Samples of these same thin sections will be sent from Oxford to Dr Ruth Ann Armitage at Eastern
Michigan University. Using a DART-MS she will determine if organic substances are still present in
the paint and whether they can be dated. A plasma oxidation system will be used to extract organic
compounds for 14C dating, on top of a conventional extraction for AMS 14C dating.