316 JOURNAL, SOMSAY NATURAL HISTORY SOCIETY, Vol. XII. 



(luce ii torrent sufficient to carry large stones over a nearly level surface for 

 ^% mile. If we take the longer oblique line and greater height, the difficulty 

 is found to be not lessened but increased. 



In the 29th volume (1873) of the Journal of the Geological Society, p. 493, 

 Dr. W. T. Blanford describes similar deposits en a far larger scale in Persia. 

 Here there are boulder- ridges extending for 5 to 10 miles from the foot of 

 the hills, with a fall of their upper surface in that distance of 1 ,000 to 2,000 

 feet. He says that the large fragments are commonest near places where 

 small streams issue from the higher ranges, but the mounds increase in 

 quantity towards the north and east, where the rainfall is less. I thought that 

 this last fact would have led the author to enunciate the theory which I am 

 about to expound, but he argues only that there must have been a greater 

 rainfall in past times, and that lakes were thus produced— without saying 

 how even then these boulders could have been transported for 5 to 10 miles 

 with so little fall. 



In Cutch these boulder-beds occur only where there are deposits of a^olian 

 origin ; and in Persia they are most abundant where there is less rain and 

 therefore presumably more dry sand to be blown about, so that some con- 

 nection between the two is suggested. It appears to me that if we suppose 

 that at one time there was more blown sand present so as to make a greater 

 slope, the weathered blocks which fell on it from the hills would, under the 

 influence of the rains saturating the sand below, slip gently forward along the 

 slope, supported by the underlying sand, till they reached their farthest 

 destination without sinking to the bottom. Thus the aeolian deposits have 

 served as the carrier (see fig. 2 ). 



Fig. 2. — Boulder-beds in the north of Patcham. 



A^Boulder-beds, 

 B:=Subrecent concrete. 



C^Jurassic rocks, 



D=Hypothetical former extension of 

 concrete with boulders. 



This explanation is analogous to that made use of by Sir Wyville Thomson to 

 account for the forward motion of the stone-river in the Falkland Islands*; 

 and, if it be a true one, it is possible that it may in some cases account for 

 deposits of loose blocks which have been referred to glacial action. There 

 will always be antagonism between this process and the running away of the 

 water in definite channels, and at last, when the slope of the ajolian deposit 

 became too low, the growth of the mounds would cease and the streams would 

 begin to sensibly denude the deposits, and even cut channels in the bed-rock 

 It might be thought that all along the rain would wash the sand away and 

 let the boulders drop, but we see that as a matter of fact it does not ; besides 

 which, the boulder itself protects the sand below it, as in the case of earth- 

 pillars, and what is washed away above or below will be replaced by the next 

 dust-storm. 



Nature," vol xv (1876), p, 359, 



