380 
M. A. Stein —Topography of the Pir Pantsal Boute. [No. 4, 
direction followed by this earlier route. Opposite to ‘Aliabad Sarai 
there opens towards the south-west a high alpine valley through which 
a path, perfectly practicable for loaded animals, leads to the mountain 
lake of Nandan Sar and thence over the JDurhal Pass to the sources 
of the Tohi of Rajaurl. This route which was used with advantage 
in the years 1814 and 1819 by strong columns of the Sikh army, when 
advancing on Hor a por, finds its natural continuation on the south or 
right side of the Pir Pantsal valley, i.e ., via HasPvanj. Only by keeping 
to this side is it possible to avoid wholly the crossing of the Pir Pantsal 
stream. The latter, as personal experience showed me in the further 
course of my tour, is not easy to ford even late in the year and would 
undoubtedly in the time of the melting snows form a still more serious 
obstacle. 
The mountain-ridge of Hastivahj which in the north, where it falls 
off towards the stream, forms a precipitous wall of rock, descends to the 
west and east with grassy slopes of a comparatively easy gradient. I 
could not retain any doubt as to the practicability of this route when 
honest Pir Bakhsh confessed to me that he, in company with friends 
from Bahramgalla, had often taken over Hast 1 vanj ponies heavily laden 
with rice. On all these occasions he had successfully evaded the police 
post of ‘Aliabad Sarai and — the Kacpnir export-prohibition. Additional 
evidence for the old route here indicated is furnished by the position of 
the ancient frontier fort of Kramavarta which will be discussed below. 
The name HasPvanj contains in its first part undoubtedly the 
Ka 9 miri stem hast 1 ‘ elephant,’ derived from Skr. liastin ; for the second 
part -vanj I am unable at present to find any clear etymology. 1 In the ab¬ 
sence of all indications as to the earlier history or original meaning of the 
1 Abu-1-fazl explains according to Col. Jarrett’s translation vatnr (recte vanj) 
by ‘ injury ; ’ but the word is not found with this meaning in modern Kacjmiri. The 
above quoted Persian compilators render vauj by raftan. The inhabitants of the 
neighbouring valleys know themselves, as far as I could ascertain, of no explanation 
of the name. The derivation from Skr. bhayga, suggested in the note of the transla¬ 
tion, is based on an erroneously supposed form of the name (hastibhaiij) and is unten¬ 
able. 
[While these pages were passing through the press, the learned Editor of this 
Journal has favored me with an interesting note pointing out that a root \/vanj 
meaning ‘ to go ’ occurs in Western Panjabi. As Kafjmiri, Western Panjabi and 
Sindhi belong to one group of Indo-Aryan-Yernaculars, the North-Western, this 
root might have been used in Kac^miri too at an earlier stage of the language. 
The \/vanj is not found in modern Ka^miri, and if the information given to me 
by my friends from Bahramgalla is correct, it is unknown also to the Pahari dialects 
spoken in the valleys immediately to the south of the Pir Pantsal —For Western 
Panjabi forms of this root see Bhai Maya Singh’s Panjabi Dictionary, Lahore 1895, 
p. 1194, and O’Brien’s Glossary oj the Mnltdni Language , Lahore, 1881, p. 276.] 
