It is even urged that it is harder to find a work treating of the subject in any special way among subdivisions than when there is only one alphabet, which is absurd. On the one hand one must look over a list of books embracing five or six distinct divisions of a subject and select from titles often ambiguous or provokingly uncommunicative those that seem likely to treat of the matter in the way desired. On the other plan he must run over five or six headings given by another man, and representing that man’s ideas of classification, and decide under which of them the treatise he is in search of is likely to be put. Which system gives the least trouble and demands the least brain-work? Plainly the latter. In three cases out of four he can comprehend the system at a glance. And if in the fourth there is a doubt, and he is compelled after all to look over the whole list or several of the divisions, he is no worse off than if there were no divisions; the list is not any longer. The objection then to subdivisions is not real, but fanciful. The reader at first glance is frightened by the appearance of a system to be learned, and perversely regards it as a hinderance instead of an assistance. But if anyone has such a rooted aversion to subdivisions it is very easy for him to disregard them altogether, and read the list as if they were not there, leaving them to be of service to wiser men.
As the number of titles under each heading increases in number so does the opportunity and need of division. The first and most usual groups to be made are _Bibliography_ and its companion _History_, and the “practical-form” groups _Dictionaries_ and _Periodicals_ Under countries the first grouping will be _Description and Travels_, _History and Politics_, _Language and Literature_, followed by _Natural history_, etc. For examples of further subdivisions see the longer catalogues. It is not worth while in a printed catalogue to make very minute divisions. The object aimed at,—enabling the enquirer to find quickly the book that treats of the branch of the subject which _he_ is interested in,—is attained if the mass of titles is broken up into sections containing from half a dozen to a score. Of course there are masses of titles which can not be so broken up because they all treat of the same subject in the same way, or at least show no difference of treatment that admits of classification. The general works on the Fine Arts in a library of 100,000 volumes may number 100 titles, even after _Periodicals_ and _Dictionaries_ have been set aside.