NOTES TO BOOK XVI If. 685 



The great problem of the diffusion of drift and erratic 

 blocks from their parent rocks to great distances, has 

 driven geologists to the consideration of other hypothetical 

 machinery by which the effects may be accounted for : 

 especially the great northern drift and boulders; the rocks 

 from the Scandinavian chain which cover the north of 

 Europe on a vast area, having a length of 2000 and 

 breadth of from 400 to 800 miles. The diffusion of these 

 blocks has been accounted for by supposing them to be 

 imbedded in icebergs, detached from the shore, and floated 

 into oceanic spaces, where they have grounded and been 

 deposited by the melting of the ice. And this mode of 

 action may to some extent be safely admitted into geolo- 

 gical speculation. For it is a matter of fact, that our 

 navigators in arctic and antartic regions have repeatedly 

 seen icebergs and icefloes sailing along laden with such 

 materials. 



The above explanation of the phenomena of drift sup- 

 poses the land on which the travelled materials are found 

 to have been the bottom of a sea where they were 

 deposited. But it does not, even granting the conditions, 

 account for some of the facts observed ; that the drift 

 and the boulders are deposited in " trainees" 1 or streaks, 

 which, in direction, diverge from the parent rock ; 

 and that the boulders are of smaller and smaller size, as 

 they are found more remote from that center. These 

 phenomena rather suggest the notion of currents of water 

 as the cause of the distribution of the materials into their 

 present situations. And though the supposition that 

 the whole area occupied by drift and boulders was a sea- 

 bottom when they were scattered over it much reduces 

 the amount of violence which it is necessarv to assume 



