I have treated the mimetic element of budding drama in the agricultural festivals as being primarily a manifestation of the activities of play determined in its direction by the dominant interests of the occasion, and finding its material in the débris of ritual custom left over from forgotten stages of religious thought. It is possible also to hold that the _mimesis_ is more closely interwoven with the religious and practical side of the festivals, and is in fact yet another example of that primitive magical notion of causation by the production of the similar, which is at the root of the rain-and sun-charms. Certainly the village dramas, like the other ceremonies which they accompany, are often regarded as influencing the luck of the farmer’s year; just as the hunting-and war-dances of savages are often regarded not merely as amusement or as practice for actual war and hunting, but as charms to secure success in these pursuits[658]. But it does not seem clear to me that in this case the magical efficacy belongs to the drama from the beginning, and I incline to look upon it as merely part of the sanctity of the feast as a whole, which has attached itself in the course of time even to that side of it which began as play. The evolution of folk-drama out of folk-dance may be most completely studied through a comparison of the various types of European sword-dance with the so-called ‘mummers’,’ ‘guisers’,’ or ‘Pace-eggers’’ play of Saint George. The history of the sword-dance has received a good deal of attention from German archaeologists, who, however, perhaps from imperfect acquaintance with the English data, have stopped short of the affiliation to it of the play[659]. The dance itself can boast a hoar antiquity. Tacitus describes it as the one form of _spectaculum_ to be seen at the gatherings of the Germans with whom he was conversant. The dancers were young men who leapt with much agility amongst menacing spear-points and sword-blades[660]. Some centuries later the use of _sweorda-gelac_ as a metaphor for battle in _Beowulf_ shows that the term was known to the continental ancestors of the Anglo-Saxons[661]. Then follows a long gap in the record, bridged only by a doubtful reference in an eighth-century Frankish homily[662], and a possible representation in a ninth-century Latin and Anglo-Saxon manuscript[663]. The minstrels seem to have adopted the sword-dance into their repertory[664], but the earliest mediaeval notice of it as a popular _ludus_ is at Nuremberg in 1350. From that date onwards until quite recent years it crops up frequently, alike at Shrovetide, Christmas and other folk festivals, and as an element in the revels at weddings, royal entries, and the like[665]. It is fairly widespread throughout Germany. It is found in Italy, where it is called the _mattaccino_[666], and in Spain (_matachin_), and under this name or that of the _danse des bouffons_ it was known both in France and England at the Renaissance[667].