was a boy, he had been employed in the excavations made by Jagat Singh, and that he knew all about the discovery of the jewels, &c. According to his account the discovery consisted of two boxes, the outer one being a large round box of common stone, and the inner one a cylindrical box of green marble about 15 inches in height and 5 or 6 inches in diameter. The contents of the inner box were 40 to 46 pearls, 14 rubies, 8 silver and 9 gold earrings (karn phul), and three pieces of human arm bone. The marble box was taken to the Barâ Sahib (Jonathun Duncan), but the stone box was left undisturbed in its original position. As the last statement evidently afforded a ready means of testing the man's veracity, I enquired if he could point out the spot where the box was left. To this question he replied without any hesitation in the affirmative, and I at once engaged him to dig up the box. We proceeded together to the site of the present circular hole, which was then a low uneven mound in the centre of a hollow, and, after marking out a small space about 4 feet in diameter, he began to work. Before sunset he had reached the stone box at a depth of 12 feet, and at less than 2 feet from the middle of the well which he had sunk. The box was a large circular block of common Chunar sand-stone, pierced with a rough cylindrical chamber in the centre, and covered with a flat slab as a lid. I presented this box, along with about 60 statues, to the Bengal Asiatic Society, and it is now in their Museum, where I lately recognized it. In their catalogue, however, it is described as "942B, a Sarcophagus found in the tope of Manikyala (!); Donor, Lieutenant A. Cunningham."

The discovery of the stone box was the most complete and convincing proof that I could wish for of the man's veracity, and I at once felt satisfied that the relics and the inscribed figure of Buddha found by Jagat Singh's workmen had been discovered on this spot, and consequently that they could not possibly have any connexion with the great tower of Dhamek. My next object was to ascertain the nature of the building in which the box was deposited. As I had found the box standing on solid brick-work, I began to clear away the rubbish, expecting to find a square chamber similar to those which had been discoverd in the topes of Afghanistan. My excavations, however, very soon showed that, if any chamber had once existed, it must have been demolished by

Jagat Singh's workmen. Sangkar then described that the box was found in a small square hole or chamber only just large enough to hold it. I cleared out the whole of the rubbish until I reached the thick circular wall which still exists. I then found that the relic box had been deposited inside a solid brick hemispherical stupa, 49 feet in diameter at the level of the deposit, and that this had been covered by a casing wall of brick, 16 feet in thickness; the total diameter at this level was, therefore, 82 feet. The solid brick-work of the interior had only been partially excavated by Jagat Singh's workmen, nearly one-half of the mass, to a height of 6 feet above the stone box, being then untouched. I made some excavations round the outer wall to ascertain its thickness, but I left the brick-work undisturbed.

About 18 years afterwards, the excavation of this stupa was continued by Major Kittoe and Mr. Thomas until the whole of the inner mass had been removed, and the foundation of the outer casing exposed. The inner diameter is given by Mr. Thomas as 49 feet 6 inches, the slight excess over my measurement being due to the thickness of a base moulding of the original stupa. I have again carefully examined the remains of this monument, and I am quite satisfied that in its original state it was an ancient hemispherical stupa, 49 feet in diameter at base, and about 35 or 40 feet in height, including the usual pinnacle. Afterwards, when, as I suppose, the upper portion had become ruinous, it was repaired by the addition of a casing wall 16 feet in thickness. The diameter of the renewed edifice thus became 82 feet, while the height, inclusive of a pinnacle, could not have been less than 50 feet.

On a review of all the facts connected with this ruin, I incline to the opinion that the inner hemisphere was an ancient relic stupa, and that this having become ruinous, it was repaired, and an outer casing added by the brothers Sthira Pála and Vasanta Pála in A. D. 1026. In the Maháwanso we find the record of similar additions having been made to some of the stupas in Ceylon, and I know from personal inspection that many of the great Dhagopas of Barmah have been increased in size by subsequent additions.

Due south from the great tower of Dhamek, and at a distance of 2,500 feet, there is a lofty ruined mound of solid

brickwork, surmounted by an octagonal building. When I first lived at Banâras, this mound was always known by the name of Chaukandi, of which no one knew the meaning. But during my late visit I found that the old name was nearly forgotten, having been superseded by Luri-ka-kodan or "Luri's leap." Luri was an Ahir, who jumped from the top of the octagonal building some years ago, and was killed. The mound itself is 74 feet in height to the floor of the octagonal building which rises 23 feet 8 inches higher, making a total height of 97 feet and 8 inches. An inscription over one of the door-ways of the building records that it was built in the reign of Humayun, as a memorial of the emperor's ascent of the mound.

In 1835 I opened this mound by sinking a well from the floor of the building right down to the plain earth beneath the foundation. I also drove a horizontal gallery to meet the well about half way up the ascent. But as neither of these excavations resulted in any discovery, I then thought it possible that my well might not have been sunk in the axis of the building. I therefore began to widen the well from the point of junction of the gallery until it was nearly 20 feet in diameter. This work was stopped at a depth of 27 feet by my departure from Banâras. I have again examined this ruin, and I am now quite satisfied that my first well was sunk in the very centre of the mound. The absence of any relic chamber shows that this was not a relic tower, a conclusion which is fully borne out by Hwen Thsang's description of one of the most remarkable of the sacred edifices near the Deer Park at Banâras, which, I believe, may be identified with the Chaukandi mound.

At 2 or 3 li (or rather less than half a mile) to the south-west of the Deer Park Monastery, Hwen Thsang places a stupa which was no less than 300 feet in height.* This lofty monument sparkled with the rarest and most precious materials. It was not ornamented with rows of niches, neither had it the usual bell-shaped cupola, but its summit was crowned with a sort of religious vase, turned upside down, on the top of which was an arrow. This is the whole of Hwen Thsang's account of this remarkable building, which, although too meagre to gratify curiosity, is still sufficient for

* Julien's Hwen Thsang, II., p. 363.