This piece was first printed by Stowe in 1561. Stowe happened to have access to a MS. which was really a miscellaneous collection of Middle-English pieces of various dates; and he proceeded to print them as being 'certaine workes of Geffray Chauser,' without paying any regard to their contents or style. In vol. i. pp. 33, 34, I give a list of his additions, numbered 42-60[34]. By good fortune, the very MS. in question is now in Trinity College Library, marked R. 3. 19. We can thus tell that he was indebted to it for the pieces numbered 46, 47, 48, 49, 50, 51, 53, 54, 55, 56, and 59. These eleven pieces are all alike remarkable for being non-Chaucerian; indeed, no. 56 is certainly Lydgate's. But it has so happened that no. 59, or The Court of Love, being the best of these pieces, was on that account 'attributed' to Chaucer, whilst the others were unhesitatingly rejected. And it happened on this wise. § 67. After Tyrwhitt had edited the Canterbury Tales afresh, it occurred to him to compile a Glossary. He rightly reasoned that the Glossary would be strengthened and made more correct if he included in it all the harder words found in the _whole_ of Chaucer's Works, instead of limiting the vocabulary to words which occur in the Canterbury Tales only. For this purpose, he proceeded to draw up a List of what he conceived to be Chaucer's _genuine_ works; and we must remember that the only process open to him was to consider all the old editions, and _reject_ such as he conceived to be spurious. Hence his List is not really a list of genuine works, but one made by striking out from all previous lists the works which he _knew_ to be spurious. A moment's reflection will show that this is a very different thing. Considering that he had only his own acumen to guide him, and had no access to linguistic or grammatical tests, still less to tests derived from an examination of rimes or phonology, it is wonderful how well he did his work. In the matter of rejection, he did not make a single mistake. His first revision was made by considering only the pieces numbered 1-41, in the _first_ part of Stowe's print (see vol. i. pp. 31-33); and he struck out the following, on the express ground that they were _known to have been written by other authors_; viz. nos. 4, 11, 13, 25, 26, 27, 28, 29, 33, and 40[35]. Then he went over the list again, and struck out, on internal evidence, nos. 15, 18, 21, 22, and 32[36].