THE PLEISTOCENE OR GLACIAL PERIOD. 445 



rium between the carbon dioxide of the air and that of the ocean would 

 require the whole oceanic content to be reduced proportionally with 

 the reduction in ' the atmosphere. But this view seems to neglect 

 (1) the very slight efficiency of diffusion; (2) the limitation of agita- 

 tion to a comparatively shallow surface layer; (3) the effects of life 

 in this surface layer; (4) the interference of uncarbonated waters 

 arising from ice melting; (5) the long period of circulation necessary 

 to bring about an interchange between the body of the ocean and 

 the atmosphere; (6) the part played by temperature in this inter- 

 change; (7) the part played by ice-formation, and (8) fundamentally, 

 the change in the basis of equilibrium itself. 



II. Variations in supply the working factor. — As already noted, the 

 foregoing hypothesis makes the depletion of carbon dioxide by chem- 

 ical union or by oceanic absorption, the working feature, while varia- 

 tions in the supply are regarded as modifying elements not easily dis- 

 cussed at present, because the distribution of volcanic action, regarded 

 as the chief variable, is not well determined. It is possible, how- 

 ever, to reverse the point of view, and regard the variation in the sup- 

 ply of carbon dioxide as the working factor and variations in con- 

 sumption the modifying ones. This latter, if we have not misappre- 

 hended, is essentially the view of Arrhenius * and Hogbom. 2 



The working application of this form of the hypothesis would be 

 rather markedly different from that sketched above, but it has not 

 been worked out into detail, so far as we are aware. 



III. Proximate hypotheses. — In the atmospheric class of hypotheses 

 are to be reckoned two that are proximate but not ultimate hypoth- 

 eses : namely, the cloud hypothesis, 3 and the wind hypothesis. 4 Without 

 doubt clouds and wind are important factors in the development of 

 glaciation; but if clouds are made the essential factor, the problem 

 is only shifted to the cause of such persistent clouds covering such 

 large areas for tens of thousands of years consecutively, with a cool- 

 ing potency competent to develop the great ice-sheets. The solution 

 of this seems as formidable as the problem in its usual form. Much 



' Loc. cit. 



2 Svensk Kemisk Tedskrift, Bd. VI, 1894. 



'Manson, Am. Geol, Vol. XIV, 1894, pp. 192-194; Vol. XXIII, 1899, pp. 44-57 ; 

 •ind Vol XXIV, 1899, pp. 93-120, 157-180, 205-209. 



' Harmer, Geol Soc. London, 1901; Abstract in Geol. Mag., 1901, p. 327. 



