infer that this city was Kot Kamália, and I would explain the discrepancy in the two narratives by a reference to the details of this campaign which are given by Arrian. Alexander divided his army into three great bodies, of which the advanced division, commanded by Hephaestion, marched five days ahead; the centre was commanded by himself, and the rear division, which was commanded by Ptolemy, followed three days behind. As the campaign was directed against the Malli, I conclude that the army marched by the direct route, via Shorkot towards Multân, which was certainly the capital of the Malli. Shorkot would thus have fallen to Hephæstion, who commanded the advanced division of the army. Alexander's own route will be described presently, when I come to speak of Kot Kamâlia.

The antiquity of Shorkot may be ascertained approximately by the coins which are found in its ruins. These consist chiefly of Indo-Scythian copper pieces of all ages, with a few Hindu specimens, and a large number of Muhammadan coins. A single copper piece of Apollodotus was obtained by Burnes. From these data I would infer that the town was certainly occupied as early as the time of the Greek kings of Ariana and the Panjab, and that it was in a flourishing state during the sway of the Indo-Scythians, or from B.C. 126 down to A.D. 250, or perhaps later. But as the Hindu coins which I obtained from Shorkot were entirely confined to the Brahman kings of the Kabul valley and the Panjab, I conclude that the place was either deserted, or, at least, in a very decayed state, during the middle ages; and that it was either re-occupied or restored in the tenth century by one of these Brahman kings.

Kot Kamália.

Kot Kamâlia is a small but ancient town situated on an isolated mound on the right or northern bank of the Râvi, which marks the extreme limit of the river's fluctuations on that side.* It is 44 miles to the south-east of the junction of the Hydaspes and Akesines, and 35 miles to the east-south-east of Shorkot. It possesses an ancient mound of burnt-brick ruins, and is said to have been overthrown by a king from the West at the same time as Shorkot and Harapa. Its present name, according to some people, was derived from a Muhammadan governor, named Kamâlud-din. But this is not certain; and I think it is quite possible that it may owe its origin to the Malli tribe, which still exists in this part of the country; but whether the name be old or not, it is quite certain that the site is very ancient; and I am, therefore, led to believe that it may be identified with the first city captured by Alexander in his campaign against the Malli.

Arrian's account of the capture is so clear and concise that I will quote it in his own words.† Qn leaving the junction of the rivers Alexander "marched through a desert country against the Malii, and the first day pitched his tents on the banks of a small rivulet, about one hundred stadia distant from the river Akesines. Having there allowed his troops a little time for refreshment and rest, he ordered every one to fill all his vessels with water, which done, he continued his march the remaining part of that day and all night, and early the next morning arrived at Anabasis,' vi. 7.

*See Maps Nos. V. and VI.

infer that this city was Kot Kamália, and I would explain the discrepancy in the two narratives by a reference to the details of this campaign which are given by Arrian. Alexander divided his army into three great bodies, of which the advanced division, commanded by Hephæstion, marched five days ahead; the centre was commanded by himself, and the rear division, which was commanded by Ptolemy, followed three days behind. As the campaign was directed against the Malli, I conclude that the army marched by the direct route, via Shorkot towards Multân, which was certainly the capital of the Malli. Shorkot would thus have fallen to Hephæstion, who commanded the advanced division of the army. Alexander's own route will be described presently, when I come to speak of Kot Kamâlia.

The antiquity of Shorkot may be ascertained approximately by the coins which are found in its ruins. These consist chiefly of Indo-Scythian copper pieces of all ages, with a few Hindu specimens, and a large number of Muhammadan coins. A single copper piece of Apollodotus was obtained by Burnes. From these data I would infer that the town was certainly occupied as early as the time of the Greek kings of Ariana and the Panjâb, and that it was in a flourishing state during the sway of the Indo-Scythians, or from B.c. 126 down to A.D. 250, or perhaps later. But as the Hindu coins which I obtained from Shorkot were entirely confined to the Brahman kings of the Kabul valley and the Panjâb, I conclude that the place was either deserted, or, at least, in a very decayed state, during the middle ages; and that it was either re-occupied or restored in the tenth century by one of these Brahman kings.

fixing the site of the fortress assaulted by Alexander with Kot Kamalia.

Arrian describes the place as a walled city with a castle seated on an eminence of difficult access, which the Indians held for a long time. At last it was carried by storm, and the whole of the garrison, to the number of 2000, were put to the sword.

Harapa.

Whilst Alexander was engaged in the assault of the city just described, Arrian relates that he had dispatched Perdikkas with the cavalry against "another city of the Malii, into which a great body of Indians had fled for safety."* His instructions were to blockade the city until Alexander arrived; but the inhabitants deserted the place on his approach, and took refuge in the neighbouring marshes. This city I believe to be Harapa. The mention of marshes shows that it must have been near the Râvi, and as Perdikkas was sent in advance of Alexander, it must also have been beyond Kot Kamâlia, that is, to the east or south-east of it. Now this is exactly the position of situated 16 miles to the east-south

Harapa, which is

east of Kot Kamâlia, and on the opposite high bank of the Râvi.† There are also several marshes in the low ground in its immediate vicinity.

Harapa has been described by two well-known travellers, Burnes and Masson, and to their descriptions I am not able to add much, although I have been encamped at the place on three different occasions. Burnes estimated the extent of the ruins as "about

* Anabasis,' vi. 6.

See Maps Nos. V. and VI. 'Bokhara,' i. 117.

three miles in circumference, which is one-half too much, as the actual ruined mound forms an irregular square of only half a mile on each side, or two miles in circuit. But this comprises only the remains of the walled town, to which we may fairly add the suburbs, or fields now covered with broken bricks and other remains, which would bring the size of the old town quite up to Burnes's estimate. Masson* notices a tradition that Harapa once extended on the west as far as Chichawatni, a distance of 12 miles, which serves, at least, to show the belief of the people as to the former size and importance of their town.

The great mass of ruins is on the western side, where the mound rises to 60 feet in height in the centre. At this point there are several massive walls built of large bricks, which are, no doubt, the remains of some extensive building. The other portions of the mound vary from 30 to 50 feet in height, the mass being formed almost entirely of broken bricks. Tradition assigns its foundation to Raja Harapa, of unknown date, and its destruction to the same western king, of the sixth century, who overthrew Shorkot, and whom I believe to have been the leader of the White Huns. The crimes of its ruler, who claimed the husband's privilege on every marriage, are said to have drawn down the vengeance of Heaven, and Harapa remained uninhabited for several centuries. As the coins that are found in its ruins are similar to those discovered at Shorkot, I think that the two places must have experienced the same fortunes; I would, therefore, assign its downfall to the Arabs, who overran the whole of the lower Panjâb imme

*Travels,' i. 453, and Journ. Asiat. Soc. Bengal, vi. 57.