§ 3. As to the date when this idea of forming a continuous series of tales was first entertained, we can hardly be wrong in dating it from 1386 or 1387 onwards. As it was left in an incomplete state, it was most likely in hand up to the time of his death, though he probably neglected it towards the last. The year 1385 is, almost certainly, the date of his Prologue to the Legend of Good Women, and of his first attempt to write in heroic couplets[77]. He was then full of the idea of writing a series of stories concerning 'Good Women,' and himself tells us that he intended to write stories of nineteen Women, to be followed by the Legend of Alcestis; but we find him suddenly desisting from his task without completing his ninth Legend, that of Hypermnestra. For this we may reasonably assign two causes; he was probably already somewhat weary of his self-imposed task, and he also began to see his way to a still grander collection, on a larger scale. It is important to observe that Chaucer was, throughout life, haunted by great ideas; and especially, by the desire to leave behind him at least some one great work which would attract general attention. Thus it was that he attempted a translation of the huge French poem of Le Roman de La Rose, which he probably never finished, though we do not know how far he proceeded. He planned the poem of Troilus and Criseyde, which terminates rather suddenly, but not until it had extended to the great length of more than eight thousand lines. Next he planned the House of Fame, which was to be largely a work of imagination; but here once more he was dissatisfied, and abandoned it whilst still incomplete. Almost at once he took up the Legend of Good Women, with its Prologue and twenty stories, but again abandoned it for a larger scheme. It is also tolerably clear that the Monkes Tale originally took its rise from a similar desire to write a succession of lives of illustrious men; and that the first conception of this idea preceded that of the Canterbury Tales. We thus see our author constantly striving after the endeavour to produce some great original work; and the Canterbury Tales was, in fact, the result of the latest and greatest of these endeavours. To assign any exact date for the Man of Lawes Prologue, which mentions April 18, is difficult. Yet we must exclude 1389, when that day was Easter Sunday, a day unsuitable for travelling and telling tales; as well as 1390, when April 17 was Sunday, which would have prevented the pilgrims, at any rate, from making an early start (Prol. 822-5). The year 1391 is certainly too late; so that only 1386, 1387, and 1388 are left for consideration. But in 1386, Easter-day fell on April 22, and Good Friday on April 20; and we cannot suppose that the pilgrimage could have taken place in Passion-week, when the Parson and others would be much in request for the duties which the season imposed upon them.