contemporaries and successors at the end of the eighteenth and beginning of the nineteenth century, four was a very favourite if not the most usual number.  But these volumes were usually small—­not much larger than those of the Belgian reprints of Dumas which, as one remembers, used to run into the dozen or something like it in the case of his longer books.  Three, however, has obvious advantages; the chief of them being the adjustment to “beginning, middle, and end,” though there is a corresponding disadvantage which soon developed itself—­and in fact, finally, I have no doubt helped to ruin the form—­the temptation to make the second volume a place of mere padding.  But the actual popularity of “the old three-decker” continued for quite two generations, if not more, and was unmistakable.  Library subscriptions were generally adjusted to it; and any circulating-library keeper would tell you that, putting this quite aside, even subscribers to more or fewer volumes than three would take the three-volume by preference.  More than this, still, there is a curious fact necessarily known to comparatively few people.  Although it was improper of Mr. Bludyer to sell his novel, and dine and drink of the profits before “smashing” it, there were probably not many reviewers who did not get rid of most of their books of this kind, if for no other reasons than that no house, short of a palace, would have held them all.  And, in the palmy days of circulating libraries, the price given by second-hand booksellers for novels made a very considerable addition to the reviewer’s remuneration or guerdon.  But these booksellers would not pay, in proportion, for two or one volume books—­alleging, what no doubt was true, that the libraries had a lower tariff for them.  Further, the short story, now so popular, was very unpopular in those days:  and library customers would refuse collections of them with something like indignation or disgust.  Indeed, there are reviewers living who may perhaps pride themselves on having done something to drive the dislike out and the liking in.

The circulating library itself, though not the creation of the novel, was very largely extended by it, and helped no doubt very largely to extend the circulation of the novel in turn.  Before it, to some extent, and long before so-called “public” or “free” libraries, books in general and novels in particular had been very largely diffused by clubs, “institutions,” and other forms of co-operative individual enterprise, the bookplates of which will be found in many a copy of an old novel now.  Sometimes these were purely private associations of neighbours:  sometimes they belonged to more or less extensive establishments, like that defunct “Russell Institution in Great Coram Street,” which a great author, who was its neighbour, once took for an example of desolation; or the still existing and flourishing “Philosophical” examples in Edinburgh and Bath.  In these latter