[27] The term “Byzantine” is often applied loosely to buildings in which only the ornamental details have a Byzantine character. But the primary and distinguishing structural feature of Byzantine architecture is the dome on pendentives. The Byzantine features of the Pazzi are involved with others derived from different systems, but they are very distinct. The central vault, though of Gothic form, is supported on pendentives, and the true dome on pendentives occurs, as we have seen, in the sanctuary and the porch. [28] The entablature does, however, occur under vaulting in some provincial Roman buildings, as in the Pantheon of Baalbek, where it forms the wall cornice from which the vaulting springs. But this, though not defensible, is less objectionable than the Renaissance scheme of an entablature passing through the imposts of archivolts. [29] As in the arch of the apse of St. Paul outside the wall at Rome, and in the Baptistery of Florence. [30] The character of these details will be discussed in the chapter on the carved ornament of the Renaissance. [31] Cf. Vasari, _Opere_, vol. 2, p. 368 _et seq._, and Milanesi’s foot-note, p. 370. [32] _Op. cit._, vol. 1, p. 200. [33] _Op. cit._, vol. 2, p. 541. [34] _Op. cit._, vol. 1, p. 201. [35] _Op. cit._, vol. 1, p. 201. [36] Vasari, _op. cit._, vol. 4, p. 152, and Milizia, vol. 1, p. 214. [37] Figures 21 and 22 are taken from Serlio, _D’Architettura_, book 3, Venice, 1560, pp. 25 and 40. [38] _Op. cit._, vol. 7, p. 163. [39] Serlio, the architect (a younger contemporary of Bramante), _op. cit._, p. 33, tells us that Bramante, at his death, left no perfect model of the whole edifice, and that several ingenious persons endeavoured to carry out the design, among whom were Raphael and Peruzzi, whose plans he reproduces. That ascribed to Raphael has a long nave, while that said to be by Peruzzi has the form of the Greek cross with round apses and a square tower in each external angle. The whole question of Bramante’s scheme, and of the successive transformations to which the design for the edifice was subjected before its final completion, is fully discussed in the work of Baron H. von Geymüller, _Die ursprünglichen Entwürfe für Sanct Peter in Rom_, Wien and Paris, 1875–1880. [40] _Op. cit._, bk. 3, p. 37. [41] Serlio does not state on what authority this illustration is based, but there appears no reason to question its correctness. Its authenticity is discussed by Baron von Geymüller (_op. cit._, p. 240 _et seq._), who accepts it as genuine. [42] The alterations that have been made at different times since the original completion of this interior are of no concern here. The arrangement was practically the same in Bramante’s time as it is now.