The use of metamorphosed basic dykes as one of the most important single field criteria for subdividing gneisses in high grade areas into different chronological units is defended. The universal applicability of the "lower greenstone-granite-upper greenstone" model to all Archaean terrains is questioned using documented sequences of events in the North Atlantic craton. We prefer a chronology based on field criteria to one based on the application of theoretical crustal development models taken from one tectonic environment and used to explain the sequence of events in another.
It is shown that the average K20 and Rb content from the 3600 m.y. sialic rocks of the North Atlantic craton ranges between 2.26 (Amksoq grey gneisses, Greenland) to 2.66 (Uivak grey gneisses, Labrador). Average K/Rb ratios are respectively 200 and 177, Rb/Sr, 0.33 and 0.29 for the two areas. K and Rb values are thus markedly higher than those reported from most other Archaean gneiss suites. Secondary redistribution of K and Rb at about 3600 m.y. is demonstrated by the documentation of the massive addition of these elements to basic rocks included in the gneisses. Whole sale addition of alkalies during migmatisation to the level of crust now exposed is postulated as one explanation of the unusually high K and Rb contents. It is argued on statistical grounds that if Rb metasomatism occurred it is not possible to use low initial Sr ratios alone to preclude the possibility that part of the Archaean gneiss complexes consist of tonalitic gneiss which are much older than conventional Sr0 interpretations allow.
I. Discussion of the Depositional Environment of the Upernarik Supracrustals and Their Relative Age Compared to that of the Uivak Gneisses and the Saglek Dykes
This was outside the terms of reference of the first paper which was solely concerned with the petrology, chemistry and possible crustal history of the Uivak gneisses. We agree with Glikson that the Upernavik supracrustals could