the classicistic sculpture of the Ara Pacis; but a glance at the Odyssey landscapes is sufficient to show that their summary treatment of landscape and figures is essentially 'impressionist'; hence it is not surprising that Wickhoff and his followers have endeavoured, though without success, to question the date assigned, on external evidence, to the house and its paintings. Further, the 'black room' of the house in the grounds of the Farnesina is decorated in its main panels with landscapes effectively sketched with the utmost economy of means; and above these runs a frieze with subjects likewise treated in the 'impressionist' manner. Here there can be no question that we are dealing with work of the Augustan period; and, if confirmation were needed, it would be found in the chapter of Vitruvius from which a quotation was given above, and in a passage of Pliny's 'Natural History,' which attributes to a painter probably named Ludius, 'diui Augusti ætate,' precisely such landscapes. Moreover, there are strong grounds for believing that Alexandria was the home of such methods of painting. Amongst the scenes from the frieze of the 'black room' is one which at first sight recalls the judgment of Solomon, but is doubtless to be interpreted in the light of a similar story told of the Egyptian king Bocchoris; the 'impressionist' technique is seen in its most extreme form at Pompeii in certain paintings representing the cult of Isis; and Petronius, whose taste in art as in literature was classicistic, laments the decay of painting 'postquam Ægyptiorum audacia tam magnæ artis compendiarium invenit.' Such a 'short cut' may be descried in the technique of the paintings just described.

It is difficult, therefore, to subscribe to Wickhoff's doctrine that illusionism is the creation of Roman artists; but he has the undoubted merit of having been the first to demonstrate that, under its influence, Roman sculpture reached the height of its destiny in the Flavian period. To this we may agree, although his almost passionate panegyric on the reliefs from the passage-way of the Arch of Titus proves to rest in part on an insecure foundation. Petersen has shown, for example, that the nice calculation by which, according to Wickhoff, the artist avoided the casting of shadows on the background of imagined sky, is an impossibility. Wickhoff's main

6

contention is that the sculptor has endeavoured to attain the impression of complete illusion, allowing natural illumination to complete the artistic effect produced by the means proper to plastic art, so that a frame is simply thrown open, and through it we look at the march past of the triumphal procession.' The relief 'has respiration like the pictures of Velasquez'; its 'marvellous effect' is unequalled except in the "Hilanderas."'

On this position Riegl has brought to bear the heavy artillery of his argument. He cannot allow that (as he translates Wickhoff into his peculiar dialect) 'the artist did not intend a composition of single figures, but one consisting of space with figures therein, a section of the universe'; nor that the background is no longer to be conceived as the ideal tactile surface of repose from which the individual forms which move in space spring, but as the optical indication of aerial space.' To admit such a possibility would be to surrender his cherished theory that the history of art may be reduced to a series of formulæ corresponding to stages in an orderly and unbroken development; for he holds that antiquity never reached the conception of space as uniting rather than dividing, of a composition which is more than a group of individual objects in 'planimetric' relation. Even in the reliefs of the Constantinian epoch he finds this to be the case with the composition as a whole, though the several figures are now disengaged from their background and isolated in individual spaces of three dimensions.

On the whole, Wickhoff is nearer the truth here than Riegl. It is impossible to deny that, at certain epochs, principles were intuitively divined by artists which their successors failed to grasp or to maintain. The ground conquered by Velasquez was not held by each and all of the painters who followed him. Flavian sculpture is in truth sui generis, and is the best which Rome has given us. Yet Wickhoff does not see the whole truth. For a few brief years the sure eye of the Italian artist was served by hands which had learnt to render with subtle selective touches the essentials of the thing seen. Wickhoff did well in bringing to due honour the processions from the triumph of Titus, arrested in full movement by the artist's magic; and we must bitterly regret the loss of many a panel and frieze from the buildings which

sprang up in all parts of Rome under the Flavian dynasty. A few fragments in the Lateran Museum remain to show how effective even the more hastily executed monuments of the time must have been; and the latest efforts of the court artists of Domitian in decorative sculpture are, as I believe, illustrated by the tondi of the Arch of Constantine.

But it was in portraiture that the Flavian sculptors achieved their most signal triumph. It was not, indeed, the members of the reigning house who furnished them with their most promising subjects. The rulers of the Flavian dynasty were frankly vulgar in appearance as in origin. Nor can we feel great admiration for the female portraits of the time, recognisable at a glance by their towering toupets of artificial curls. But the best male busts of the Flavian period have rarely been equalled and never surpassed. A Polish archaeologist, M. Bienkowski, has supplied a criterion by which the date of Roman busts may be inferred from their form. In the Augustan and Julio-Claudian era the breast only was rendered. The Flavian artist took in the shoulder, but did not indicate the junction of the arm. This is found in busts of the reign of Trajan; under Hadrian and the Antonines part of the upper arm is represented. Ultimately the bust developed into a half-length figure, though a return to the earlier forms was not uncommon in the third century. We are thus able to assign a number of portraits to the Flavian era; and the practised eye soon learns to recognise their style amongst the heads set on busts of all periods in modern times.

Mr Crowfoot has published in the 'Journal of Hellenic Studies' for 1900 some fine examples of this art, together with some pages of valuable criticism on their style. He rightly notes that, though there are Italian features in these busts, such evidence as we have goes to prove that the sculptors themselves were of Greek race, and shows that their excellence consists in the happy union of Hellenistic brilliancy and artifice with the unpretending fidelity to facts native to Italy and the West. It was, no doubt, the mixture of races in the capital of the world, whence issued the extraordinary subtlety and penetration, united with the highest technical dexterity, which went to the making of such a masterpiece as the

so-called Mark Antony of the Braccio Nuovo (Pl. II, fig. 2).

Wickhoff has the further merit of having shown how ornament underwent the same transformation as sculpture in the round or in relief. He singled out a pilaster (Pl. III, fig. 1) in the Lateran Museum which once formed part of the tomb of the Haterii on the Via Labicana, and showed that its style differed essentially, not merely from that of conventional Hellenistic ornament, but even from the naturalistic work of the Augustan age. Here we have no systematic study of natural forms, but an impression of growing life seized and transmuted by the artist's fancy, and rendered with exquisite subtlety and tact.* The value of Wickhoff's criticism is not lessened by the fact that in his lyrical enthusiasm for his subject he is led into avowing a belief that art, in its highest development, is concerned with form and not with subject, rejects with disdain all sources of extraneous interest such as religion or poetry, and, sufficient to itself, becomes in its last stage an art only for artists.'

The brief but brilliant activity of the Flavian period seems to have ended almost as suddenly as it began, about the commencement of the second century of the Christian era. There is a dearth of monuments which can be assigned to the early years of Trajan; but this lack is more than compensated by the wealth of material belonging to the closing period of the reign. The inscription on the base of Trajan's Column is dated in the year 113 A.D.; the triumphal arch which spanned the entrance to his Forum was completed in 117 A.D.; and the Arch of Beneventum bears an inscription dated 114 A.D. It was doubtless from the Forum of Trajan that Constantine borrowed the great frieze which was sawn up into four portions to decorate his arch; while we must assign to the same cycle of decoration fragments in the Louvre and the Villa

* Wickhoff's sure feeling for style enabled him to pronounce unhesitatingly as to the date of the pilaster and thus to correct the view which assigned the monument to the second century. The male and female busts, as Mr Crowfoot has pointed out (J.H.S. 1900, p. 36), are certainly Flavian; and the mode of wearing the hair in the latter instance has its parallel in the case of the daughter in the portrait-group at Chatsworth (J. H. S. 1901, plate xv), where the mother is Flavian. But the sculptures from the monument of the Haterii are not all of one date, and need further discussion.