But in the middle, or perhaps we might say upon the threshold of our inquiry, we must pause to examine into the amount of confidence which can be given to those under whose guidance we are invited to place ourselves. Such investigations are too frequently omitted. Those who have faith in the Bible usually decline to search into the grounds of their belief, and, in like manner, those who have always heard the author of the Zend Avesta quoted as trustworthy are apt to take everything which it may say as correct. To avoid this error, I have consulted all the volumes of the transactions of the Royal Asiatic Society of London, and have found therein sufficient to throw the gravest doubts upon the great antiquity of the Parsee religion. It will be an useful task if I attempt to classify the evidence on each side, and to draw an inference therefrom. Our knowledge respecting the Magian religion which the Bactrian* prophet founded, is built, with the exception of the notices in Greek and Latin authors, already quoted, upon the work known as the Avesta. This is written in a language called Zand,** and there are within it parts, which are written in another tongue, to which the name of Pahlavi has been given, and from these the sacred books of the Parsees have been translated into French by Anquetil du Perron, into German by Spiegel, and into English by Haug. All these writers assume that the language referred to is Ancient Persian, and closely allied to the Sanscrit, and Haug especially endeavours to demonstrate that the Avesta, and the origin of the religion of the Parsees, must be as old as the time of the Vedas, inasmuch as the same sort of legends, the same names, and, to a certain extent, the same genii, are to be found in both. There is not absolute identity, however, for those which are spoken of as good by the Vedas are treated as bad in the Avesta. Viewed from this point, Haug assigns to the Zand volumes an age of about four thousand years, and he supports his belief by a reference to the length of time which would be required to make up the two million verses attributed to Zoroaster by some Greek author. In the conclusion that both the Zand and the Pahlavi are very ancient Persian tongues, it is stated that the majority of German and French critics agree. * Zoroaster is said by many early writers to have been a king in Bactria.--Smith's Dictionary, s.v. ** The word "Zend" is more familiar to many than the form "Zand;" but I have adopted the latter, as also the spelling of Pahlavi, from an essay by Mr Romer, with an introduction by Professor Wilson, in Vol. IV., Royal Asiatic Society's Journal.