The alloying and distribution pattern of early Indian copper and copper-based artefacts and the occurrence of ingots in some archaeological contexts have been variously discussed in Indian archaeological literature. First, the presence of pure copper artefacts in early Indian archaeological contexts is considered to be symptomatic of the scarcity of alloying agents. For instance, D. P. Agrawal has argued that the presence of 70% of pure copper objects among the analysed Harappan artefacts in this metal is reflective of a problem in the procurement of tin; since this metallic ore was not easily available, evidence of copper alloying is scarce. Secondly, the variations in element composition of alloyed artefacts have generally been regarded as evidence of incomplete knowledge of mixing – E. J. H. Mackay's argument that “mixing of the two metals was performed in a most perfunctory manner ” to explain the variations in the average proportions of tin to copper in Mohenjodaro specimens is evocative of this general approach. Thirdly, not many sites which have yielded copper objects have yielded any positive evidence of manufacture – in some cases this may be due to the limited nature of the excavated data, but on the whole one would argue that the sites which can be described as centres of copper metallurgy are comparatively few and far between in the Indian context. Fourthly, the occurrence of copper ingots at different sites has been generally taken to be indicative of trade in this metal.