100 W. T. Blanford— On the Physical Geography [Xo. 2, 



are usually produced at right angles to the wind's direction.* Parallel rows 

 of sand-dunes along a coast are frequently due to the regular sea-breeze, and, 

 as may he seen on the east coast of India, there are often several such rows 

 one behind the other, but they exhibit the usual evidence of their origin by 

 having a long slope towards the sea and a short steep slope landwards. I 

 think it quite possible that the sand-hills of Umarkot and Eastei'n Sind 

 generally may be of such antiquity as to date from a period when the rela- 

 tive distribution of sea and land in the region was different from what it 

 now is, and that to so great an extent as to completely modify the pre- 

 vailing winds, and I have even been induced to speculate on the possibility 

 of the existent parallel ridges of sand-hills marking successive coast-lines as 

 the sea receded from the face of the country. This hypothesis, however, 

 would render it necessary to suppose that the Indus valley was a land area 

 whilst the present desert was part of the sea, and that the western coast- 

 line of the sea with a general north-east to south-west direction gradually 

 receded towards the south-east ; or, vice versa, that the Indus valley was sea, 

 and the country to the south-east dry land. But I can hardly conceive 

 that such gigantic changes as this would involve could have taken place 

 without completely changing the original form of the sand-hills, and it is 

 evident that the ridges in the region of the salt ' dhandhs' must be posterior 

 in date to the time when their present site was part of an inlet of the sea, 

 and not anterior to it. Moreover, had the sand-hills been formed along a 

 coast -line, or even inland at right angles to the prevailing wind, they would, 

 here and there at all events, have preserved some traces of their original 

 slopes shewing the direction of the wind which produced them. But there 

 is nothing of the kind to be found. I looked most carefully for some evi- 

 dence of a steeper slope on one side than on the other, but without success, 

 and I found double ridges having a trough-like hollow along the crest, 

 with the slopes on both sides of the hollow, as well as those on both sides of 

 the main ridge, equally steep. For such ridges I am quite unable to 

 account by the effect of a wind blowing at right angles to their direc- 

 tion. If they were formed by one great sand-wave overtaking another, 

 one side of the depression between the crests of the two waves must 

 be much steeper than the other, and although this would be slightly modi- 

 fied by time, it could not be entirely obliterated and yet leave the general 

 form of the waves so little altered as they now appear. 



I am obliged therefore to reject the theory that these parallel ridges 

 are due to a wind acting at right angles to their direction. I cannot accept 

 Sir Bartle Frere's view that they are due to earthquake-action. The ridges 



* Naumann, however, in his ' Geognosie' (edition of 18-54, Vol. 11, p. 1171), says — 

 " The sand-hills themselves are in every country extended in length in one direction 

 which agrees with the direction of the prevailing wind." 



