[Illustration: 576. Church of Charroux. Scale 100 ft. to 1 in.] [Illustration: 577. Plan of St. Benigne, Dijon. (From Dom Plancher’s ‘Histoire de Burgogne.’) Scale 100 ft. to 1 in.] The bodies of the confessors SS. Urban and Gregory were, it appears, originally buried in the church of St. John the Baptist, which seems to have been the name most properly applied to this circular building; they were afterwards transferred to the crypt below the high altar, in the rectangular part of the church. Above the lower storey, which retained its name as a baptistery and burial-place, was the upper church, which was dedicated to the Virgin Mary; above that was the church of the Holy Trinity; and on the top of the round towers, on one side the altar of St. Michael, on the other probably that of Gabriel. The little church of Neuvy St. Sepulchre, near Bourges, which was erected between the years 1042 and 1046, presents precisely the same arrangements as the church of Charroux, though on a smaller scale, there being only one range of ten pillars in the centre. The ancient nave having been destroyed, was replaced by a more extended one in the 12th century, but the old arrangement can easily be traced. In all these old churches—and they seem to have been very common in France before the 12th century—the circular part was the most important, but they have most of them been rebuilt; and where this has been the case, even when the outline of the circular form was retained, the lines of the nave were made tangents of the circle, and thus became parts of one design. All these arrangements were perfect before the church of Conques (Woodcut No. 574) was erected. There the architect, not being hampered by any previous building, was allowed free scope for his design. The plan so produced was never lost sight of by the French, but was developed into a vast variety of beautiful forms, which we shall shortly have to examine. When once this transformation of the round church into the chevet termination of a basilica was effected, the French adhered to it with singular constancy. I am not aware of their ever having built a circular church afterwards which was intended to stand alone; and there are very few instances of basilicas of any importance without this form of apse. Some, it is true, have been rebuilt on old foundations, with square eastern ends, but this is rare and exceptional, the chevet being the true and typical termination. The church at Conques and that of Toulouse both show it fully and beautifully developed, though externally the chapels hardly fit pleasingly into the general design, and look more as though their addition were an afterthought. This, however, was soon afterwards remedied, and the transformation made complete.