The customs of the old New Year feasts have frequently wandered from their original November date, and to this fact we owe whatever elements of northern paganism are to be found in Christmas. Some practices seem to have been put forward to Michaelmas; one side of the festivals, the cult of the dead, is represented especially by All Saints' and All Souls' days (November 1 and 2). St. Martin's Day (November 11) probably marks as nearly as possible the old Teutonic date, and is still in Germany an important folk-feast attended by many customs derived from the beginning-of-winter festival. Other practices are found strewn over various holy-days between Martinmas and Epiphany, and concentrated above all on the Church's feast of the Nativity and the Roman New Year's Day, January 1, both of which had naturally great power of attraction.{30} The progress of agriculture, as Dr. Tille points out,{31} tended to destroy the mid-November celebration. In the Carolingian period an improvement took place in the cultivation of meadows, and the increased quantity of hay made it possible to keep the animals fattening in stall, instead of slaughtering them as soon as the pastures were closed. Thus the killing-time, with its festivities, became later and later. St. Andrew's Day (November 30) and St. Nicholas's (December 6) may mark stages in its progress into the winter. In St. Nicholas's Day, indeed, we find a feast that closely resembles Martinmas, and seems to be the same folk-festival transferred to a later date. Again, as regards England we |174| must remember the difference between its climate and that of Central Europe. Mid-November would here not be a date beyond which pasturing was impossible, and thus the slaughter and feast held then by Angles and Saxons in their old German home would tend to be delayed.{32} Christmas, as will be gathered from the foregoing, cannot on its pagan side be separated from the folk-feasts of November and December. The meaning of the term will therefore here be so extended as to cover the whole period between All Saints' Day and Epiphany. That this is not too violent a proceeding will be seen later on. For the purposes of this book it seems best to treat the winter festivals calendarially, so to speak: to start at the beginning of November, and show them in procession, suggesting, as far as may be, the probable origins of the customs observed. Thus we may avoid the dismemberment caused by taking out certain practices from various festivals and grouping them under their probable origins, a method which would, moreover, be perilous in view of the very conjectural nature of the theories offered. * * * * * Before we pass to our procession of festivals, something must be said about the general nature and _rationale_ of the customs associated with them. For convenience these customs may be divided into three groups:--