The only inscription at Wan that I could not copy was the trilingual tablet of Xerxes. It is on the most inaccessible part of the rock, about seventy or eighty feet above the plain. Not having a glass of sufficient power, I was unable to distinguish the characters from below. As it had been accurately transcribed by Schulz, and resembles those of the same king at Persepolis and Hamadan, I did not think it necessary to incur any risk or expense in reaching it by means of ropes or scaffolding.[166] In the rock there are numerous excavated chambers, some even exceeding in dimensions those I have described; but, with the exception of a simple seat or bench of stone, about two and a half feet high on one side of them, they are perfectly plain and unornamented. They appear to have been used as tombs. Some are approached by flights of steps cut in the precipice; others are altogether inaccessible except by ropes from above. It is yet doubtful to what family of languages the Wan inscriptions must be assigned. Some believe it to be a Tatar dialect; Dr. Hincks, on the contrary, is of opinion that it is Indo-Germanic. Two of the inscriptions, and the earliest in date, as I have already observed, are in pure Assyrian. With regard to the date of the monuments there appears to be a clue which may enable us to fix it with some degree of certainty. In an inscription from Khorsabad, amongst the kings conquered by Sargon one is mentioned whose name corresponds with Arghistis, the fifth in the Wan dynasty. Supposing the two, therefore, to be the same, and there is no reason to doubt their being so, we may assume that the monarchs of the Wan records reigned from about the middle of the eighth century before Christ to the end of the seventh; and the evidence afforded by the forms of the characters leads to this conjecture.