168 THE EARLIEST MEN 



operations, unless in fact, as we put it, " it keeps 

 time." That is just what none of the "geological* 

 clocks " do. To leave the metaphor, we can never 

 feel sure that the conditions of the process in the 

 past have been identical with, or even very similar 

 to, those which we have been concerning ourselves 

 with in the present. In fact we can feel pretty 

 certain that they were anything but identical. 

 But if they were not identical, our " clock " which 

 is not, like chronometers, " compensated " for 

 all sorts of conditions under which it may find it- 

 self, must necessarily fail to " keep time," and this 

 must necessarily deceive us if we place our con- 

 fidence in it. Let us examine one or two cases in 

 order to understand this somewhat important 

 matter more fully. 



There is, first of all, the matter of erosion by 

 rivers. It is quite clear that river erosion has been 

 going on for a good many years, and it is equally 

 clear that it is going on at the present day ; can 

 we not measure the annual amount of erosion 

 now taking place, and from the amount which has 

 been eroded in the past, which can often be ascer- 

 tained without much danger of error, form some 

 conclusion as to the length of time which has 

 elapsed since the river began its work, and so of 

 the various deposits associated with it ? This 

 apparently simple calculation is vitiated by two 

 things. In the first place, it is by no manner of 

 means easy to gauge the annual amount of erosion, 

 a fact which is abundantly proved by the very 



* It is possible that an exception should be made in respect of 

 the laminated Scandinavian sands, for which see p. 217. 



