CONCLUSIONS 365 



Carboniferous ice age, and would, of course, be quite inapplicable to the 

 Pleistocene period, so that it may be left out of account. 



The most probable general cause for ice ages is to be found in changes 

 of the composition of the earth's atmosphere, as suggested long since by 

 Professor Tindall and more recently by Professor Arrhenius and elabo- 

 rated by Professor Chamberlin, since known, or possible results of the 

 increase or decrease of carbon dioxide and of aqueous vapor would go far 

 to produce the necessary changes of climate. It is highly probable, how- 

 ever, that to the general atmospheric causes should be added important 

 local changes, as, for example, of level, of ocean currents through rising 

 or sinking of sea-bottoms and land surfaces, or of the shifting of centers 

 of high and low pressure. 



It may be that ice ages can only be produced by a combination of causes, 

 local and general, and that an eclectic grouping of theories will ultimately 

 prove most satisfactory. 



Whatever theory of ancient climates is adopted must provide for long 

 periods of fairly uniform mild climates over the whole earth except on 

 lofty mountains, and for relatively brief periods of oscillation between 

 cold and warm climates during the so-called ice ages. Mr Huntington's 

 diagram of cycle climatic changes illustrates this idea very well,^® though 

 the times of rapid oscillation should probably be separated by much longer 

 stretches of comparative uniformity than he has indicated. 



Since this paper was made ready for publication the Compte Eendu of 

 the Mexican Geological Congress has reached me, containing a number of 

 important papers on related subjects by Professor David, Professor 

 Freeh, Professor Gregory, Mr Manson, and others, which should be con- 

 sulted by any one interested in glacial geology. Mr Manson's ingenious 

 theory of changes of climates as due to the cooling of the earth, followed 

 by the clearing away of clouds and the beginning of solar control of tem- 

 peratures on the earth, is of course completely at variance with the evi- 

 dence given above, showing that extensive ice-sheets reached sealevel in 

 ancient times. Professor Freeh's argument against interglacial periods 

 from the distribution of extinct mammals seems too vague to counter- 

 poise the direct evidence in favor of at least one interglacial period shown 

 in so many different parts of the world. 



The papers by Professor David strongly support several of the points 

 brought forward in the preceding pages and form an important addition 



" Bulletin of the Geological Society of America, vol. 18, p. 362. 



