That the volumes supervised by Mr. Bancroft should contain imperfections is in the nature of the case inevitable. Perfect historical estimates of contemporaries can not as a rule be made, and history based largely on personal reminiscence must contain errors of refraction which can be corrected only in the clearer light of later years. The handling of material by a writer who did not collect it, and who is likely to find the places and conditions dealt with strange to his experience, inevitable though it be in so large an undertaking, results in the writing of faulty history. The hastening of the work and the editorial revision of manuscripts by a manager desirous of pleasing subscribers, and impelled by various other motives of his own, are not circumstances likely to increase the accuracy of the work. But after allowance has been made for all inaccuracies which have crept in through these various avenues, we still have the fact that the histories are based upon sources which may be supplemented but can never be displaced. No greater mistake could be made, therefore, than to say that because they contain errors they are worthless. All must agree with the practical argument made by a thoughtful old pioneer of the writer's acquaintance that, in spite of all criticisms which may be passed upon the Bancroft histories, they contain a great fund of information which is nowhere else to be found in print. A third result of the history plan, and one which is of importance to historical writers everywhere who have large fields to cover, was the devising of a coöperative method for organizing the vast collections in the library. Mr. Bancroft makes the claim of having been the first to resort to such a division of labor; and points out (Literary Industries, 767) that his method avoids the repetition of details and insures a more thorough working up of the field than does the coöperative method as the term is usually understood, under which the writers work independently of each other after the field is divided. Such a claim might indeed be granted had Mr. Bancroft announced himself as editor and reviser instead of author, and had he designated the part of the work written by each of his collaborators in accordance with the usual custom in coöperative works. The printing of his name as author on the title page, and his general recognition as such in accordance with press notices following those of the Native Races, have, of course, largely lost for him the credit of originating a coöperative method for the organizing of large quantities of material.