Precambrian Research, 22 (1983) 167--174 167 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands GLACIERS, VOLCANIC ISLANDS AND THE ORIGIN OF LIFE ALAN W. SCHWARTZ Laboratory for Exobiology, Faculty of Science, University of Nijmegen, Nijmegen 6525 ED (The Netherlands) ANN HENDERSON-SELLERS Department of Geography, University of Liverpool, Liverpool L69 3BX (Gt. Britain) (Received December 6, 1982; revision accepted March 16, 1983) ABSTRACT Schwartz, A.W. and Henderson-Sellers, A., 1983. Glaciers, volcanic islands and the origin of life. Precambrian Res., 22: 167--174. Current models for the timing and extent of early continental crust formation, as well as for the climate of the primitive Earth, lead to the suggestion that the earliest, stable, terrestrial locales may have been volcanic islands which were partially glaciated. Recent experimental results on the synthesis of biologically important molecules in ice, suggest that such glaciated, volcanic islands would have provided uniquely favorable sites for the earliest stages of chemical evolution. CONSTRAINTS ON MODELS OF CHEMICAL EVOLUTION Discussion continues concerning the composition of the Earth's primitive atmosphere. Although there are arguments (Walker, 1977) that the early at- mosphere was not strongly reducing and may even have been redox-neutral in character (essentially a contemporary atmosphere minus O2), few if any reliable data pertaining to this problem are available.Recent suggestions that the upper mantle may have been more reducing than previously thought have added to the confusion (Arculus and Delano, 1980; Methez and Delaney, 1981). A recent analysis of the current state of knowledge con- cerning the composition of the primitive atmosphere suggests that composi- tions ranging from redox-neutral to strongly reducing are allpossible (Chang et al.,1983). The latterpossibility,however, seems to be incompatible with most geological models for the early Archaean which include a substantial hydrosphere and a temperature regime not very different from that of the present.lay (Henderson-SeUers and Cogley, 1982). Although there are many uncertainties regarding the composition of the primitive atmosphere, there are a number of very definite constraints on the 0301-9268/83/$03.00 © 1983 Elsevier Science Publishers B.V.