Under the name of G. A. Danziger I wrote in the year 1889 a story founded on a German tale, which I called The Monk and the Hangman's Daughter. The story was tragic but I gave it a happy ending. Submitting it to the late Ambrose Bierce, asking him to revise the story, he suggested the retention of the tragic part and so revised it. The story was published and the house failed. When in 1900 a publisher desired to bring out the story provided I gave it a happy ending, I submitted the matter to Bierce and on August 21, 1900, he wrote me a long letter on the subject of which the following is an extract: 'I have read twice and carefully, your proposed addition to The Monk, and you must permit me to speak plainly, if not altogether agreeably, of it. It will not do for these reasons and others: 'The book is almost perfect as you wrote it; the part of the work that pleases me least is my part (underscores Bierce's). I am surprised that you should yield to the schoolgirl desire for that shallowest of all literary devices, a "happy ending," by which all the pathos of the book is effaced to "make a woman holiday." It is unworthy of you. So much vii did I feel this unworthiness that I hesitated a long time before even deciding to have so much of "odious ingenuity" and "mystery" as your making Benedicta the daughter of the Saltmaster and inventing her secret love for Ambrosius instead of Rochus. '"Dramatic action," which is no less necessary in a story than in a play, requires that so far as is possible what takes place shall be seen to take place, not related as having previously taken place.... Compare Shakespeare's Cymbeline with his better plays. See how he spoiled it the same way. You need not feel ashamed to err as Shakespeare erred. Indeed, you did better than he, for his explanations were of things already known to the reader, or spectator, of the play. Your explanations are needful to an understanding of the things explained; it is they that are needless. All "explanation" is unspeakably tedious, and is to be cut as short as possible. Far better to have nothing to explain-to show everything that occurs, in the very act of occurring. We cannot always do that, but we should come as near to doing it as we can. Anyhow, the "harking back" should not be done at the end of the book, when the denouement is already known and the reader's interest in the action exhausted...."