Imperfect volumes are always a source of great inconvenience to the collector. First-class bookselling firms will not allow an imperfect book to leave their hands without notice to the purchaser, and, as a consequence, they charge a higher price than would be the case if the latter exercised his own judgment. There are mutual advantages to be gained in dealing with first-class people, for, if a mistake is made on one side or the other, there is usually no difficulty in rectifying it afterwards. Fine old crusted book-worms of the John Hill Burton type prefer, however, to exercise their own discretion in these matters, looking upon that as no inconsiderable part of the pleasure to be derived from the pursuit of their favourite occupation. They do not care to pay for being taught, at least not directly, and make it part of their business to find out for themselves whether a copy offered for sale is perfect or the reverse. As each page is usually numbered, there is no difficulty in ascertaining whether any are missing; not so with the plates, for, unless there is an index to these, the loss of one or two may hardly be noticed until the book comes to be collated with another copy known to be complete. This is a risk which the book buyer has to run, though, as a matter of practice, he protects himself when the purchase is an expensive one, and the dealer a man of credit. In buying books at a cheap rate, or, in other words, when making a bargain either at a shop or an ordinary street-stall, the purchaser will have to observe the maxim, "Caveat emptor," and it will probably not be until he arrives home with his treasure under his arm that he will have the satisfaction of ascertaining that his bargain is a real one, or the mortification of adding another imperfect book to the long row already on his shelf. Imperfect books are frequently what is called "made up," that is, completed from other copies, themselves imperfect in other respects. One complete book is worth more than two incomplete ones, and many desirable specimens, in the public libraries and elsewhere, are made up so well that it is frequently impossible to detect the hand of the renovator. So long as all the leaves of a made-up book are of the same measurement, there would not seem to be much objection to this practice, but there certainly is when the paper of the interpolated leaves is different from the rest, or smaller in size, which it will be if cut down by the binder. Great care must be taken to see that neither of these defects is present, especially when, from the value of a book offered for sale, it may have been worth anyone's while to perfect it.