1876.] of the Great Indian Desert. 93 



or rather to north-north-east in the direction of Bikanir. The hills in this 

 tract are not in such regular ridges as they are to the westward, but here 

 also they appear to diminish in height and to become more scattered to the 

 north. Between Jodhpur and Pokarn this eastern belt of sand-hills is only 

 about 10 miles broad. From some distance east of Pokarn to Jaysalmir, and 

 again for 50 miles west of Jaysalmir, the country is an undulating sandy 

 plain, but there are very few sand-hills. I have no personal knowledge of 

 the desert north of Jaysalmir. Stripped of the sand-hills the country would 

 be a vast plain, slightly elevated above the sea, and only broken by isolated 

 hills to the southward, by the somewhat more numerous ranges near Bal- 

 mir, by low plateaux of sandstone towards Jodhpur and Pokarn, and 

 by terraces of Jurassic sandstone and limestone around Jaysalmir. The 

 hilly regions are less sandy ; occasionally even torrent-beds are found near 

 the hills, but they are soon lost in the sand. 



§ 6. Evidence of subrecent Marine Action. Salt ' dhands.' — It is 

 impossible for any geologist to traverse this region without the suggestion 

 forcing itself upon him that this may be an example of Professor Bamsay's 

 planes of marine denudation. Such was my first impression. But I could 

 only find one circumstance, the general saltness of the ground, in confir- 

 mation of this view. Every here and there throughout the desert is a smaller 

 or larger plain of salt ground or " ran", which is said to become a shallow 

 salt lake after heavy rain. From such places salt is sometimes extracted, 

 but the quantity is small, and not more than might, very possibly, result 

 from the gradual concentration of the salt distributed in small quantities 

 throughout the soil. The water in the wells is very often brackish, but 

 this is equally the case in countries which shew no trace of having been 

 recently covered by sea water. There is, however', a very remarkable quan- 

 tity of salt in two localities which I visited, and in one of them there is, I 

 think, good evidence of the former neighbourhood of the sea. 



To take the more important and the more interesting first. North of 

 Umarkot the boundary of the Indus alluvium and of the Thar or sand- 

 hill area is formed by a river known as the Narra or the Eastern Narra,* 

 which derives its water from floods in Bahawalpur and the Bohri district 

 of Sind, and has of late years been artificially supplied by a canal cut from 

 the Indus at Bohri. On the east of the Narra rise high ridges of sand 

 with the usual NE to S W direction, and between these ridges are deep 

 valleys filled with water and known as ' dhandhs.'f Some of these ' dhandhs' 

 are said to be unfathomable ; — and doubtless they are so by an ordinary pole 



* The Sindhi form, I "believe, of the common Hindi Nala, a river channel, ravine, 

 or ditch. 



t Dhandh in Sindhi is the equivalent of jMl hi Hindi and is applied to any pool 

 of water or to a marsh. 

 13 



