Page:American Historical Review, Vol. 23.djvu/150

140 who prefer the originals will find it convenient to use the translations side by side with them. Volume V. presents probably the best translation into English that has yet been made of the Mundus Novus. This has been made from the Latin version published at Vienna in 1504, for the original Italian manuscript has probably been lost, but Professor Northup has been able to use previous translations. This letter has been published in several languages and in many editions. The promised Winship bibliography will be a welcome addition to the tools of the historian.

But of all five volumes, the fourth, the translation of the famous Soderini Letter, with its admirable introduction, offers most interest and value. Here is a work replete with the best that scholarship can offer. In his painstaking labor, Professor Northup has done what few men would care to do, for this is work that demands not only a certain training and acquirement, but a certain temperament as well. This is, in fact, more than a translation. Professor Northup has done what Napione, Gustavo Hughes, and Uzielli dreamed over—constructed a critical text of this most perplexing Vespucci letter—and his work will not have to be done oyer again. The problem he set himself was philological not historical.

There has been no attempt before to furnish something better than the confessedly erroneous texts that have been used. Professor Northup's

The three texts selected are the Florentine print, which is labelled "P", the Magliabecchiana MS., labelled "M", and the Waldseemüller translation into Latin, the Cosmographiae Introductio, labelled "H". The first probably more nearly approaches the half-Spanish, half-Italian original, and the second is an eighteenth-century copy of a copy made in 1505 from a manuscript source differing somewhat from that from which "P" was printed. Professor Northup's attempt to construct the original text of the letter as it reached the hands of Piero Soderini in 1504 is ingenious, and his reasoning will generally be accepted as sound. He has proceeded on the theory that when two of these three versions agree as against the third the two have the correct reading. When all