180 AFGHAN BOUNDARY COMMISSION. 



The whole valley of the Upper Murghab is a deep network of 

 such defiles ; there are no roads crossing the lines of lateral drainage, 

 and not any continuous road along the river, the Murghab being 

 on the whole a marked exception to the general rule in .Afghanistan 

 of the coincidence of direction between main roads and main rivers. 

 The following roads and passes connecting Afghan Turkestan with 

 Herat through the Firuzkhui districts were explored, viz. : — (I) the 

 Kharajangal pass from Chaharshamba to Herat ; (2) the group 

 of tracks and passes between the Ao Barik and the Baragan ; 

 (3) the Chahardara crossing, which carries the Tailan route over 

 the Murghab into Gharjistan and Chakcharan ; (4) the direct 

 route from Maimana and Saripul through Chaharsada, and eastern- 

 most of all (5) the Saripul, Ohiras, and Daulatyar connexion, 

 followed probably by Ferrier, and surveyed as far south as Chiras 

 by Imam Sharif. 



The roads and tracks between Kadis and Bala Murghab, and 

 those from Kadis eastward on the northern slopes of the Safed 

 Koh, have all been explored and reported on. High up on these 

 slopes, at least 5,000 feet above the sea, and about 20 miles east 

 of Kadis, Hira Sing found the remains of a city,* which must once 

 have been large and have possessed considerable local importance ; 

 the ruins of an old fort, of stone walls surrounding orchards, and 

 traces of irrigation are still visible. At the head of the Murghab 

 valley the surveying was carried out by Imam Sharif, who found 

 the position of Chiras to be approximately correct, but was unable 

 to identify the Dev Hissar of Ferrier. He found, however, a city,f 

 called the Shahr-i-Wairan (or deserted city), not far from Chiras, 

 situated in a green plain, bounded by glittering white cliffs on the 

 north and by the long straight slopes from the crest of the cliffs 

 overhanging the Askarab on the south. The city must have been 

 of some magnitude, and such buildings as remain are of sun-dried 

 brick. There are no indications of a wall or moat, nothing even 

 suggestive of a canal or Icarez, nothing, in fact, but scattered ruins 

 covering an area of about 1\ square miles. The graveyard was 

 easily recognisable, and its immense size furnished some clue to 



* Major Raverty tells me lie tliiuks this may be identical with Ashizar of Grhar- 

 jistan, mentioned ;it pp. 391 and 1071-1072 of his translation of the Tabakat-i-Nasiri. 



f This appears to be the same as that mentioned at page 334 of the same work as 

 being in the district of Wajiristan. 



