^* cf. also II, 83. ‘In the present inadequate state of our knowledge it is hazardous to establish hard and fast decisions on bare and comparatively slight differences of style’.

^† A number of points of detail can easily be set right in a second edition; thus, the Temple of Jupiter on the Capitol, though it ‘had been burnt down by the Vitellian soldiery’ (II, 56), had been rebuilt by Vespasian and burnt down yet again in 80 A.D. before Domitian restored it. There is not a vestige of brickwork in the Mausoleum of Augustus (I, 136). The Claudian harbour was not ‘a good way ’ from the mouth of the Tiber at the time of its construction (I, 162). The reconstruction of the temple of Venus and Roma is attributed to Aurelian instead of Maxentius (II, 93) and the transference of the obelisk of Antinous to the Pincio is attributed to 1633 instead of 1822 (II, 108). In one or two cases (no doubt owing to the delays to which Mrs Strong alludes in her preface), the legends of the illustrations have been transposed (figs. 82 and 83, and figs. 180 and 181 may be cited).