Here too is a very significant fact, which ought to speak trumpet-tongued to the Bodleian Curators. In 1827 Mr. Kerrich, the Public Librarian at Cambridge, possessed an Arabic Manuscript (a history of the Berbers), which was in the strictest sense of the word unique. In one sense all manuscripts are unique, for no two are or can be exactly alike, but Mr. Kerrich’s book was the only known copy of the work in existence anywhere. He was strongly urged to give or sell it to the University Library over which he presided, but he utterly declined to do either the one or the other, because the Cambridge Library is a lending library. Few men, he said, know the value of manuscripts; and he declared that there were only two libraries in England where his book would be open to the use of scholars and at the same time safe, the British Museum and the Bodleian. This manuscript now reposes on our shelves, and we got it simply and solely because in 1827 (and for many years after) we still possessed common sense. Kerrich would never have let us have this unique volume, had he supposed it possible that we should ever have been so forgetful of our duty as to lend Bodleian books. We might learn something from the Persians, who, as I was informed the other day, on what seemed to be very good authority, have a saying which runs thus:--‘The man who lends a book is a fool, but that man is a greater fool who returns a book that has been lent to him’--a fearful mixture of true with false doctrine. Now for the letters, and as Dr. Rost is a librarian he shall have precedence. His epistle will be found in the _Academy_ (March 5, 1887), and it is a real contribution to the facts of the case. It is reducible to two statements:-- 1. During nearly eighteen years there have been from the India Office ‘thousands of loans’ and ‘there has not been a single loss to record.’ In February, 1887, there were ‘337 Oriental MSS. out on loan, 47 of which are in the hands of scholars in India.’ 2. ‘Numerous editions of texts and other works based on our collections of MSS. would either have been impossible, or at least not possible, to their actual extent except for the existing arrangement.’