1836.] Third Memoir of the Ancient Coins. 543 



tion in Afghanistan, Turkistdn, and Badakshdn, which were alike sacred, 

 but in a less degree, which yet plainly indicate the strongholds of the 

 faith they commemorate. The distinguishing feature of these sacred 

 places is the samach, or cave, always found with them, and which 

 decides the identity in character of the honey-combed hills of Bamidn 

 in Afghanistan, and those of Tilmissus in Asia Minor. It is affirmed 

 in the Ayin Akberi, that there are 12,000 of these samaches in the 

 hills of Afghdnistdn : — the number is not overrated. There is no rea- 

 son to suppose that they were ever the residences of a multifarious 

 community, engaged in the ordinary occupations of life ; — it is obvi- 

 ous, that they were the abodes of priests and ascetics connected with 

 the temples of religion and sepulchral monuments. So plain is this 

 fact in Afghdnistdn, that, if a solitary samach or cave be discovered, 

 it is merely necessary to employ the privilege of sight to detect the 

 mound or tumulus relating to it ; and vice versa, if a tumulus be first 

 descried, the sight directed to the nearest eminence will not fail to 

 discover the cave or caves belonging to it. It is always the case, 

 that these monuments and caves are found at the skirts of hills, shew- 

 ing that they were remote from the inhabited villages, then as now, 

 and in conformity to the spirit of asceticism, enjoined by the religion 

 of the day. It need not, therefore, be deemed that the caves of 

 Afghanistan were the dwellings of a rude Trogloditic nation ; — on the 

 contrary, they are works of art, the results of vast labour and expen- 

 diture, and must have been formed under favorable circumstances of 

 national prosperity. Let no one imagine he beholds in them the 

 retreats of the Mardi. The most prominent of the sepulchral monu- 

 ments of Afghdnistdn are unquestionably its topes or royal cenotaphs 

 with their tumuli : the latter so perfectly agree in form with the 

 Buddhist dehgopa that it w T ould be difficult not to allow them to be the 

 same thing. The most ancient of the cenotaphs hitherto examined in 

 Afghdnistdn does not appear to attain the antiquity of the Christian 

 era, — most of them certainly fall much short of it : it is true that every 

 tope has its caves, but there are caves, as in the conspicuous instance 

 of Bamian, which have not topes : Bamian*, like every other spot 



* There is an error in our account of the site of Alexandria ad calcem Cau- 

 casi, contained in our memoir of 1834 relative to the river of Bamian, which 

 it is necessary should he noted. We have made that river pass by Ghorband, 

 which we supposed it did, contrary to the reports of the natives — they are cor- 

 rect, and the river flowing northerly falls into the stream of Kunduz. Ptolemy, 

 we believe, has an upper and a lower Nilabi, when noting the country about 

 Alexandria ,• and they can scarcely be other rivers than those of Ghorband, and 

 Puryshir, — May, 1836. 



