But I do not think that the mediocre quality of most book-reviewing is due to low payment. It is a result, I believe, of a wrong conception of what a book-review should be. My own opinion is that a review should be, from one point of view, a portrait of a book. It should present the book instead of merely presenting remarks about the book. In reviewing, portraiture is more important than opinion. One has to get the reflexion of the book, and not a mere comment on it, down on paper. Obviously, one must not press this theory of portraiture too far. It is useful chiefly as a protest against the curse of comment. Many clever writers, when they come to write book-reviews, instead of portraying the book, waste their time in remarks to the effect that the book should never have been written, and so forth. That, in fact, is the usual attitude of clever reviewers when they begin. They are so horrified to find that Mr. William Le Queux does not write like Dostoevsky and that Mrs. Florence Barclay lacks the grandeur of AEschylus that they run amok among their contemporaries with something of the furious destructiveness of Don Quixote on his adventures. It is the noble intolerance of youth; but how unreasonable it is! Suppose a portrait-painter were suddenly to take his sitter by the throat on the ground that he had no right to exist. One would say to him that that was not his business: his business is to take the man’s existence for granted, and to paint him until he becomes in a new sense alive. If he is worthless, paint his worthlessness, but do not merely comment on it. There is no reason why a portrait should be flattering, but it should be a portrait. It may be a portrait in the grand matter, or a portrait in caricature: if it expresses its subject honestly and delightfully, that is all we can ask of it. A critical portrait of a book by Mr. Le Queux may be amazingly alive: a censorious