Iron-rich basaltic komatiites in the early Precambrian Vermilion District: Discussion

Abstract

In their paper, Green and Schulz (1977) describe and present chemical analyses of moderately magnesian, pillowed lavas and chilled margins of sills from the Vermilion District, Minnesota. They point out that the analyses are similar but not identical to those of certain komatiitic lavas, and on this basis call their rocks iron-rich basaltic komatiites. They then propose liberalization of established numerical criteria of komatiite compositions (Brooks and Hart 1974) to accomodate their samples. The paper vividly illustrates the present chaotic state of komatiite nomenclature. The rocks that Green and Schulz call komatiites are almost identical to certain rocks (aphanitic pyroxenites) in the upper part of a thick, layered flow in Munro Township, Ontario, which was identified as tholeiitic by Arndt (1976) and Arndt et al. (1977). Evidence for the similarity between the rocks from the two areas is striking. Green and Schulz suggest that their samples may represent a seawater-chilled carapace of a very thick flow; samples from both localities are texturally similar, being composed largely of prismatic augite grains mantled by green amphibole; and their representative chemical analyses are almost identical (compare analyses E-151a and E-149, Table 1, Green and Schulz (1977) with analyses 8 and 9, Table 3, Arndt (1977)). These rocks also share many features with other rocks that traditionally and consistently have been described as tholeiitic. Examples include layered sills in Archean greenstone belts (MacRae 1969; Williams and Hallberg 1973; Viljoen and Viljoen 1970), basalts in greenstone belts (Pearce and Birkett 1974; Naldrett and Goodwin 1977) and the picritic and basaltic lavas of Baffin Bay (Clarke 1970). In Munro Township the lavas were specifically distinguished from the mafic to ultramafic komatiitic lavas of that area. The komatiites were shown to be genetically related to spinifextextured ultramafic lavas that almost certainly have crystallized directly from ultrabasic liquids (a criterion that must form part of any definition of