On the banks of the Gandaki, at Muktanath, is a precipice, from which the river is supposed to wash the Salagrams or black stones, which are considered by the Hindus as representatives of several of their deities, and which are the most common objects of worship in Bengal, where images are scarce. They are of various kinds, and accordingly represent different deities. Pilgrims, who have been at the place, say, that the stones are found partly in the precipice, and partly in the bed of the river, where it has washed down the earth. On account of its containing these stones, this branch of the river is usually called the Salagrami, and the channel every where below Muktinath, until it reaches the plain of India at Sivapur, abounds in these stones. All the Salagrams consist of carbonate of lime, and are in general quite black, but a few have white veins. Their colour is probably owing to some metallic impregnation, which also occasions their great specific weight. They rarely exceed the size of an orange, and they are rounded, I suppose, by the action of water. Most of them are what naturalists call petrifactions, and by far the most common are Ammonites, half imbedded in a ball of stone, exactly of the same nature with the petrified animal. Others, which are reckoned the most valuable, are balls containing a cavity formed by an Ammonite, that has afterwards decayed, and left only its impression, or they are what Wallerius calls _Typolithi Ammonitarum_. The Ammonites or their impressions are called the Chakras or wheels of the Salagrams, but are sometimes wanting. The stone is then a mere ball without any mark of animal exuviæ. Some balls have no external opening, and yet by rubbing away a portion of one of their sides, the hollow wheel (_chakra_) is discovered. Such Salagrams are reckoned very valuable. In many parts of these mountains, the substance called _Silajit_ exudes from rocks. I have not yet satisfied myself concerning its nature; but intend hereafter to treat the subject fully, when I describe the natural productions of Behar, where I had an opportunity of collecting it, as it came from the rock. The valley of Nepal Proper, which contains Kathmandu, or, as many call it, Kathmaro, Lalita Patan, and Bhatgang, is the largest in the dominions of Gorkha, and in this plain there is not naturally a single stone of any considerable size. The whole, so far as man has penetrated, consists of what is called alluvial matter, covered by soil. In some places the alluvial matter consists of thick beds of fine gravel and sand, much of which is micaceous. Among these beds are found concretions of the same materials, united into balls, about the size and shape of a turkey’s egg. At one end these are generally perforated with a small hole, and some, but not all of them, are hollow. The Newars call them Dungoda; but can give no account of their formation, nor did I observe any thing that could lead to an explanation.