Introduction

The use of ochre has been an important component of Northwest Coast ceremonial activities, and these practices were essential in the enactment and maintenance of socio-political relationships (Ames et al. 1999).

Ochre is composed of iron oxide mixed with clays, silicates and other minerals (Popelka-Filcoff et al. 2007), and most commonly occurs as sedimentary infill or as precipitate of iron-rich rock. Ochre sources are discrete deposits of iron oxides that are geochemically distinct from each other (Erlandson et al. 1999). The ochres used in this study are nodules of oxides that occur as cultural deposits within shell middens at four village and camp sites in traditional Heiltsuk and Wuikinuxv First Nation territories on the central coast of British Columbia, dated to c. 2800 BP and the contact period (Reference CannonCannon 2000) (Figure 1; Table 1). In the analysis presented here, we have used neutron activation analysis (INAA) to determine the geochemical composition of ochre and to characterise different source groups. The insights gained into patterns of ochre procurement, distribution and use have enabled us to further our understanding of the role and long-term use of ochre among culture groups on the central coast of British Columbia.

Table 1.Archaeological sites included in this study.

Analytical procedure

53 samples were chosen for instrumental neutron activation analysis (INAA). Two irradiations were performed to acquire data on short and long-lived isotopes (Table 1). Standard reference materials (1632b Coal; 1633b Fly Ash; 688 Basalt; 278 Obsidian Rock issued by NIST, and Ohio Red Clay) and control samples were run with each batch of materials.

To account for the variable content of Fe in ochre (5-70 per cent), all elemental concentrations are expressed as a ratio to Fe. Concentrations are then converted to log10 format (Figures 2-4).

Figure 2.

Plot showing log10 correlation between Sc/Fe and La/Fe (ppm).

Figure 3.

Log10 plot of V/Fe and Mn/Fe (ppm) showing distribution of sources by site. Ellipses are at 80 per cent confidence.

Figure 4.

Log10 plot of Ti/Fe and Eu/Fe (ppm) showing characteristics of multiple source groups.

Elements that produce short-lived isotopes: Al, Ba, Br, Ca, Co, Cl, Dy, Ho, K, Mg, Na, Ni, Ti and V; Eu, Ga, K, La, Mn, Na and Sm.

Elements that produce long-lived isotopes: As, Au, Ba, Br, Ce, Co, Cr, Cs, Eu, Fe, Hf, La, Lu, Ni, Na, Nd, Pr, Rb, Sb, Sc, Sm, Ta, Tb, Th, Yb and Zn.

Principal components analysis (PCA) and bivariate plotting demonstrates the geochemical variability in the materials. PCA indicates that elements driving the variance within this set are Hf, Ti, V and rare earth elements.

Results

Geochemical analysis indicates moderate tendencies towards different source groups of ochre, each corresponding with a different village site, suggesting each village consistently accessed a single source of ochre, which was not traded. Samples from five short-term campsites were comparable to ochre from the village sites in their immediate vicinity. One exception noted was that a small percentage (12 per cent) of the ochre from Namu matched the source represented at Katit.

Interpretations and conclusions

Our results indicate that ochre was procured by each village group from a unique local source. Residents of Namu may have acquired ochre from the same source represented at Katit, possibly suggesting limited exchange.

The unexpected differences among villages in relatively close regional proximity suggests that ochre was not a widely traded commodity and that individual sources may have had specific meaning for local residential groups.

The results of this analysis contribute to our understanding of the role of ochre among these culture groups, and provide an archaeological history of its long-term use.