The appearance of this evidently very modern deposit is exactly that which would no doubt be exhibited, were the present lake close to the city of Kashmir dried up, and a section of its bed exposed. This lake contains abundance of shells, and in the neighbourhood of the town it is made the receptacle of refuse of every kind, broken pottery being particularly plentiful. In shallow places in the river, close to the town of Bijbehara, a similar deposit is accumulating, valves of a _Cyrena_ being found to some depth in the fine mud, mixed with broken pots, charcoal, bones, and other refuse.
[Sidenote: TEMPLE IMBEDDED. IN LACUSTRINE CLAY. _May, 1848._]
The most remarkable fact connected with this very recent lacustrine deposit is, that the ruins of an ancient temple exist on the plain above the Jelam, a little west of the modern village, partially buried in the clay. The upper parts of two temples, resembling in all respects the ruins on the elevated platform at Martand, near Islamabad, stand on the open plain, not far from the river, but perhaps twenty feet above its level, and certainly far below the level to which the clay containing pottery rises on the hill-sides. One of the temples is quite in ruins, the immense blocks of which it is built being piled confusedly on one another. The beautiful colonnade (exactly like that at Martand) by which it is surrounded, is evidently quite uninjured in any way; but it is entirely buried under the lacustrine clay, except a very small portion, consisting of three pillars, which were exposed by Major Cunningham in 1847. These three pillars may be seen in a cavity under the level of the present surface of the ground, and the clay in which they were imbedded contains fragments of pottery in profusion.
If these temples (the date of which I believe is approximately known to antiquarians) were contemporaneous with the ancient town, they must have been buried in the lacustrine silt at some period not very long subsequent to their erection, if I am right in supposing a lake to have existed at the same time with the town. Probably, therefore, they are anterior in age to the town, as they are imbedded in such masses of pottery as could only have been accumulated in the neighbourhood of a very dense population. Their present appearance, I think, helps to explain the nature and origin of the many lakes or marshy depressions which occur in all parts of the valley. It appears evident that at Avantipura, at some period subsequent to the building of the temples, a subsidence of the ground must have taken place during one of the many earthquakes which are well known to have convulsed the Kashmir valley. This subsidence, which must have been partial, and not co-extensive with the valley, converted the ground on which the temples stood into a lake. A fresh subsidence, or the gradual wearing away of the incoherent clay strata lower down the river, must at last have drained the little lake, and left the country round Avantipura in the state in which we now see it. Even now a marsh partly under water during the spring months extends from Avantipura for several miles up the river.