Newly formulated models of the earth's core have been described recently as ‘simultaneously provocative and vexing’ (R. Jeanloz, Nature, 299, 108, 1982). Part of the reason for this is that concepts about the earth's core are being rapidly updated.
That there is knowledge at all of a core results from the analysis of seismic data; such analysis is model-dependent. Further analysis of more extensive data results in new models; alternative approaches to the problems from theoretical geochemical considerations result in another type of model. The conjunction of such models may seem incongruous, but not so. A great deal is being learned about the geochemistry of processes that have a bearing on core formation. Significantly, there has been a simultaneous surge of improved seismic data and analysis.