A border line of this character can obviously not have been meant to ward off attack by organized forces, but only to protect the cultivated portion of the Helmand delta against nomadic raiders. In view of the geographical facts there can be no doubt that this Limes faced to the south. The region of barren hills that extends there must already in ancient times have been occupied by nomadic tribes corresponding in character and habits, if not also in race, to the Balûch and Brahui tribes to be found there at present. The latter have maintained their reputation as very troublesome neighbours of the settled population in Sistan to the present day. Those in the Sarhad (` the border ') hills of Persian Baluchistan, due south of Sistan, have often enough defied the Shah's troops or those of his great feudatory, the chief of Birjand, acting as lord of these Marches, when attempts have been made to reduce them into effective subjection .6
In the absence of more definite evidence as to the date of construction of this defensive border line, it would serve no useful purpose to discuss here questions as to the ethnic and political conditions which are likely to have prevailed in and around Sistan in Parthian or Sasanian times and may have had their bearing on the policy indicated by this protected border. Still less should we be justified in drawing from it conjectural conclusions as to the position and extent of that portion of the Helmand delta which may then have been under cultivation. Nor is there occasion to make more than the briefest reference to the curious analogies presented to the ancient Chinese Limes which I had traced along the Kan-su border far away in the east, and to the Roman Limes systems in the west. But I may hint at least at an interesting antiquarian question. Could this protected desert border of Sistan be thought of as forming a geographical link between that ancient ` Chinese wall ', pushed out into the Tun-huang desert along the early Central-Asian high road, and the Limes lines by which Imperial Rome endeavoured in Arabia, Syria, and elsewhere in the Near East to facilitate the defence of its marches against barbarian inroads ? Future research may possibly help us to an answer.
SECTION V.—FROM SISTAN TO INDIA AND LONDON
I should have gladly faced the physical discomforts which, with the approach of spring, would necessarily attend continued work on desert ground in Sistan, if it had been possible for me to extend my survey to the Afghan portion of the areas now abandoned to the desert. Sir Henry McMahon's Mission and earlier travellers had found in that region important ruined sites, still awaiting close investigation. Permission to visit it could, however, not be secured for me, and considering the conditions created by the war I did not feel altogether surprised at this. So after completing my survey of the ancient border line on the Persian side of the old southern delta, I set out at the beginning of February on my return journey to India.
After striking the westernmost outpost of British Balûchistan at Kôh-i-Malik Siâh, I travelled by the ` Sistan Trade Route ', which the zeal of Captain (now Colonel) F. Webb Ware, of the Indian Political Department, had first pioneered through the desert some thirty years before. Well known as the route is, I found a special quasi-historical interest in this journey of close on 400 miles through desert wastes—for the ` Chagai Agency ' comprising them extends over more than five degrees of longitude but includes a population numbering only about 5,000, practically all nomads. I could not have wished for a better modern illustration of the conditions of traffic that once prevailed on that early Chinese route through the Lop Desert which two years previously I had succeeded