What then are the associations and qualities which give books value in the eyes of a collector? We may answer the question negatively in the first instance by reducing to their proper importance the two qualities which are popularly supposed to be the most attractive to the book-hunter--rarity and age. If a book is otherwise uninteresting, what is it the better for being rare? In passing it may be noted that unless a book is interesting for other reasons its rarity is necessarily an unknown quantity. Sir Sidney Lee's Census of the extant copies of the First Folio Shakespeare, a comparatively common book, but of supreme interest for its associations, is a striking example of the zeal with which every discoverable copy of a valuable book is now hunted down. Those whose business it is to gather such information can tell in the case of dozens of books of much less importance exactly how many copies have been discovered and in whose possession they remain. But in the case of a book of little interest the most that can be said is that it is "undescribed," and it may be "undescribed" not in the least because it is really rare, but because no bibliographer has troubled himself to make a note of it. Were some real point of interest discovered in it the chances are that the attention thus attracted would speedily bring to light other copies, as in the case of the school magazine to which Mr. Kipling was found to have contributed. Of this the first set catalogued sold for over L100, with the result that so many others were unearthed that the price speedily sank to less than as many shillings. Granted, however, that it could be proved that a dull book is not merely undescribed, but absolutely, what so few works are, unique, in what way does this make it of interest to the collector? A great library might buy it for a trifle out of compassion, or under the idea that its registration in a catalogue might help to piece out a genealogy, or that it might count as another unit in statistics (a poor reason), or justify its purchase in some other haphazard way. But considerations of this kind, such as they are, cannot affect private collectors. A really dull book is merely a nuisance, and whether only one copy of it, or many, can be proved to exist, nobody wants it. If this be so we are justified in saying that, although as soon as a book is found desirable for any other reason its rarity becomes of paramount importance in determining its price, Rarity by itself is of no interest to collectors.