Richard the third has long been one of the most popular of Shakespeare's plays. In its own day six quarto editions appeared before the Folio, and two more quartos before 1635. While our record of its early history in the theatre is not so clear as that of its printing, numerous contemporary references testify to its popularity. In the later seventeenth century it temporarily disappeared from the stage, but in July, 1700, it was revived with certain changes by Colley Cibber, and was later played by Garrick, Kemble, Kean, J. B. Booth, Edwin Booth, Macready, Sir Henry Irving, Richard Mansfield, and John Barrymore. No one to-day would acclaim Richard as a great tragedy rather than a melodrama, but historically it has probably led all other dramas of Shakespeare in frequency of presentation, and is still an effective stage play. Critics generally unite in dating its composition about 1593, two or three years before Romeo and Juliet. In this paper I propose to analyze anew certain aspects of its relation to its source material, and to show that its composition resembles that of Romeo. Thereby it is possible, I believe, to learn something about Shakespeare's method of writing at least two earlier plays.