Abstract

The paper is based on the study of a series of specimens extending from the core of a residual kaolinized boulder-like body of nepheline syenite into the surrounding concretionary (so-called pisolitic and oolitic) bauxite. Alteration of feldspars to finely crystalline gibbsite, of this gibbsite to more amorphous-looking concretionary bauxite (cliachite), and subsequent kaolinization of these alteration products by silica from the unbauxitized inner part, is described and illustrated. Iron liberated during kaolinization forms an envelope of limonite in the outer part of the kaolinized bauxite. The probable beginnings of kaolinization are seen in a specimen from the core of the pseudo-boulder of syenite in which there are remnants of feldspars. In the more kaolinitic inner gibbsitic parts of the bauxite envelope there is evidence of flowage, here attributed to expansion resulting from kaolinization of the gibbsite. Carbonates, mainly dark brown and ferruginous and including ankerite, are present in all bauxite-free kaolin specimens even in kaolin believed to be derived from silication of bauxite, but they are absent in kaolin associated with bauxite. Sphene altered to leucoxene is present throughout, but tends to be corroded and to disappear in the more external, more mature bauxite. Sulfides were found only in the kaolinized core.It is suggested that the anomalous process of the reduction of crystalline gibbsite to unresolvable or amorphous cliachite may be due to dehydration of part of the gibbsite to one of the monohydrates, diaspore or boehmite.An envelope of bauxite could not form around a weathering core if silica escaping from the core recombined with the bauxite to form kaolin. It seems probable, therefore, that when bauxitization was the prevailing weathering process, kaolin was not an intermediate product. The silica escaping during bauxitization may have been in a form in which it could not react with the bauxite; perhaps in an organic compound. This may have been due to the action of microorganisms as suggested by Holland (3).

First Page Preview

First page PDF preview
You do not currently have access to this article.