In book vi. c. iv. § 2, Caesar Germanicus is spoken of as still living. He died in Syria, A. D. 20 (19). This book was therefore written before that year. In book xii. c. viii. § 11, Strabo says that Cyzicus was still a free state. It lost its liberty A. D. 25 . This book was therefore written before A. D. 25 . Whether Strabo was alive or dead at this date, we have no means of determining. The codices or manuscripts which exist of Strabo’s work appear to be copies of a single manuscript existing in the middle ages, but now lost. From the striking agreement of errors and omissions in all now extant (with such differences only as can be accounted for, arising from the want of ability or carelessness of the copyist), it appears most probable that to this single manuscript we are indebted for the preservation of the work. Strabo himself describes the carelessness of bad scribes both at Rome and Alexandria, Book xiii. c. i. § 54, vol. ii. p. 380. in the following expressive language: Some vendors of books, also, employed bad scribes and neglected to compare the copies with the originals. This happens in the case of other books, which are copied for sale both here and at Alexandria. After what Kramer has done for the text, we can hope for little improvement, unless, what is beyond all expectation, some other manuscript should be discovered which is either derived from another source, or is a more correct copy. The following is some account of those in existence:— Codices in the Imperial Library, Paris: No. 1397 of the catalogue. This is the principal codex existing in the Imperial Library, and was written in the 12th century. It was formerly in the Strozzi Palace at Rome, and was brought to Paris by Maria de Medici. Not only are parts of the leaves, but even whole leaves of the 9th book, damaged or destroyed by damp, mice, bad binding, and careless attempts at correction. This codex contains the first nine books; the second part, containing the last eight, is lost. Collated by Kramer, and partly for Falconer, by Villebrune. No. 1393 of the catalogue. On this codex Brequigny chiefly depended for his edition. Montfaucon says that it is of the 12th or 13th century; Kramer, however, judging from the character of the handwriting and contractions, maintains that it belongs to the end of the 13th or beginning of the 14th century. It contains the whole seventeen books of the Geography, and was written in the East (not, however, by the same hand throughout), and brought from Constantinople to Paris by the Abbé Servin in 1732 , to whom it had been presented by a Greek named Maurocordato. Collated by Villebrune for Falconer, and partly by Kramer.