ORIGIN AND DIFFERENTIATION 553 



carbides. Moreover, the carbon or graphite in the Lanrentian rocks is 

 so vast in quantity — more, Sir William Logan once said, than in any 

 similar area of Carboniferous rocks in the world — that it would require 

 for its production an incalculable mass of algae, which were without woody 

 substance, and hence low in carbonaceous matter. And, further, many 

 plants, notably red and brown algae, do not require sunlight for their 

 assimilation processes; hence, if we should grant that the pre-Cambrian 

 carbon had been reduced by plant life, it would by no means follow that 

 this life process implied sunlight. It could have gone forward in 

 absence of direct sunlight. 



The last point Professor Barrell seeks to make is based on the fact that 

 glacial conditions and, as he holds, periods of widespread aridity have 

 recurred at intervals since Middle pre-Cambrian time; hence he con- 

 cludes that the composition of the atmosphere and ranges of temperature 

 have been subject to fluctuation through all geologic time. 



The deductions based on glacial conditions apparently are on the sup- 

 position that all observed glacial activities are veritable ice ages having 

 a profound effect on temperatures, life distribution, etcetera. The 

 fallacy involved in this conception has already been set forth. 



The criteria on which widespread aridity are predicated are not stated, 

 though presumably based on the conventional doctrine of red beds, de- 

 posits of salt, etcetera. The great deposits of salt are presumed to have 

 resulted from the evaporation of impounded saline waters. But has any 

 one attempted to visualize the physical setting essential for the formation 

 of a 100-400-foot bed of salt by evaporation? Is it certain that such 

 deposits are proof positive of aridity? 



The physical character of sediments and the interpretation of their 

 mode of deposition has been used to prove aridity, but when — as, for 

 example, in the case of the Old Red sandstone — interpretations of th? 

 same data range all the way from the permanent deserts postulated r)y 

 Walther and Goodchild and the semi-aridity advocated by Barrell and 

 Jukes-Brown to the contention of a marine origin by Hugh Miller and 

 recently confirmed by Macnair and Eeid, the layman may well hesitato. 



There is undoubtedly an element of truth in all of these several con- 

 tentions, but it is when they are forced to bear too great a burden that 

 they falter. As BarrelP^ has well said in another connection : 



"In general, traditional criteria are liable to lead into error because they 

 become accepted as axioms and are applied without further thought. They 

 lag behind the development of a science." 



35 Joseph Barrell : Dominantly fluviatile origin under seasonal rainfall of the Old Red 

 sandstone. Bull. Geol. Soc. Am., vol. 27, 1916, p. 354. 



