because by entering one description of anonymous books in one way and another in another, there would be an end to the uniformity of rule which is necessary to prevent a very extensive catalogue from getting into confusion. Another instance is the cataloguing of academical transactions and periodicals under the respective heads of Academies and Periodical Publications, which has been much criticised. It is quite true that the Quarterly Review can be found more easily under that head than under "Periodical Publications, London," but it is also true that the grouping of all academical and all periodical publications under these two great heads is invaluable to the bibliographer, the literary historian, and the statistician, who must be exceedingly thankful that the information of which they are in quest is presented to them in a concentrated form, instead of having to be sought for through an enormous catalogue. These observations do not in any way apply to libraries of an essentially popular character, and I merely make them by way of enforcing the proposition that the works of such libraries and those of national or university libraries are different, and that we must beware of a cast-iron uniformity of rule. There is yet another intermediate class of library, the comparatively small but highly select, such as college and club libraries, which will probably find it more advantageous to pursue an intermediate course, as I imagine they do, judging from the very excellent specimens of cataloguing for which we are indebted to some of them. And there is yet another class, the libraries
