upon fresco work for St Peter's is a point on which it is needless to insist. The developed powers of design which he displays in the ciborium are sufficient to prove that his experimental days were over. Moreover, the editor must have forgotten that, according to his own chronology, there is an interval between Giotto's work in Rome and in Assisi, and that in this interval the artist was occupied with the famous frescoes in the chapel of the Podestà. We have the right to expect a strong manifestation of Roman influence in this Bargello work, but nothing of the kind has as yet been detected there. Every reader will remember that the Bargello frescoes are a subject of keen controversy, and that many connoisseurs believe them to have been executed in 1337 by one of Giotto's pupils. This fact would seem in itself a conclusive evidence of their purely Florentine character; nor can this purity of origin be doubted for a moment by any one who has examined them or given but a cursory glance to the lovely relics of the great fresco of the Paradise. The simple rows of its standing figures, rank after rank, the natural severity of the falling lines of drapery-as near as nature to monotony, and as free from it-the intensity of calm devotion which pervades the whole, testify to an artist whose style is his own, who knows the effect he wishes to produce and the means by which he can produce it.

A further difficulty of a serious kind connects itself with the new date suggested by the editor for the famous allegorical series of the Lower Church at Assisi. It was recognised by the authors that, in the absence of external evidence, Giotto's works could only be classed reasonably by adhering to the general law of progress in style. It was this consideration-so, in the first edition, they openly state-which led them to place the Allegories immediately before the Roman altarpiece. The editor, it will be remembered, regards the Allegories as a later work than the Arena Chapel. Nevertheless, he considers that the allegorical series at Padua stand on a higher plane. Whereas, in the Allegories at Assisi, the symbolism is crudely, obviously, even vulgarly, expressed, the figure of Injustice at Padua is 'symbolically as well as artistically one of the most remarkable achievements of its kind that the world has seen.' As a master of alleVol. 200.-No. 399.

D

gorical composition,' the editor explains, Giotto was very fitful' (ii, 115).

In finally estimating the editor's contribution to the history of the subject, we are compelled to admit that his taste is by no means faultless, and that he has failed to show a due sense of the dignity of his position. The inclusion of a theory in his notes is no guarantee that he has examined it critically; and he is not content merely to state what he believes. Far from feeling that divergence of opinion between himself and the authors is at best a misfortune which nothing but the claims of truth must lead him to disclose, he is careful to emphasise divergence wherever it occurs, and to question the value of the authors' observations where they lead to conclusions other than his own. If this method was dictated by the belief that his authority carries equal weight with that of Crowe and Cavalcaselle, it is our duty to state definitely that such a belief is groundless. Moreover, his writing is often lacking both in dignity and in reserve, and it is not always impartial. He thinks it necessary to remark that the authors never allowed their æsthetic judgment to be warped by personal or pecuniary considerations'; that neither of them belonged to that parasitic cosmopolitan class from which the writers of little art-books are frequently drawn'; that Cavalcaselle was 'neither a place-hunter nor a picture-dealer in masquerade,' Had the remotest suspicion existed that these things were true, it might be possible to understand the value of the editor's denials. Whether his own æsthetic judgment is in the same degree unwarped, it is only fair to question. Throughout these two first volumes we search in vain for mention of one of our most distinguished connoisseurs; we find an attribution, of which the merit is his, referred to another writer, who was careful to preface the series of articles in which he made use of it by the statement that the theories he promulgated were not necessarily original. The error must be as disconcerting to Mr Fry as it is discourteous to Mr Berenson.

At this point it will be well to leave the work of the editor and devote an undivided attention to that of Sir Joseph Crowe himself. Here, as already suggested, the reader needs to exercise a certain leniency, for

nothing is more obvious than that the author was prevented from giving it his final revision. This appears unmistakably from the number of ungrammatical sentences which are to be found in the text. Unimportant in itself, the presence of such sentences does much to disarm criticism. It is impossible to say what the author might not have changed if he had had a full opportunity of revising what he wrote. But, in spite of this disadvantage, there can be no question as to the great increase of value in the text as he has now presented it. To begin with the more superficial consideration, it will be noted that a certain exuberance of style the quality of which was peculiar to the early Victorian period of our literature, and which was more than usually prominent in the first edition of the book -has been successfully curtailed in the second edition. 'On the whole, however, he [Gaddi] was inferior to Orcagna; and the unity of talent which characterised the son of Cione was not conspicuous in the last scion of the Gaddi.' So wrote Sir Joseph Crowe in 1864. In the new edition he has substituted: On the whole, however, Agnolo Gaddi must be allowed to rank as a painter below Orcagna, who shows more unity of power and more depth of intellect than any of his contemporaries.'

6

The change here introduced is typical. Nor is it only in the style that evidences of increased control and abler handling may be found. Already, in the first edition, the authors displayed, in the treatment of earlier authorities and of the various sources from which they drew, a reticence and discrimination which gave peculiar value to their work. A balanced judgment, rarely losing sight of the scope and purpose of the history, directed the choice of references and quotations. The authors conferred distinction upon every writer whose name appeared upon their pages. Yet passages occurred in their writing which showed that the subject at times overpowered them, and, particularly where their appreciation was highest, they were apt to allow criticism to disappear in rhetoric. Their account of Giotto, for instance, was disturbed by repeated comparisons of his achievement with those of Ghirlandaio and Raphael; while their numerous allusions to the antique and their appreciations of its superior excellence were hardly re

quired of historians of the art of Italy. Nothing is more remarkable in the second edition than its increased directness and reserve. The work now possesses an accuracy of focus and a sustained concentration which can only be described as masterly.

It is of the greatest interest to trace from the authors' pages the methods of study which they seem to have pursued, and to deduce the quality of their genius and its limitations. Much light is thrown upon this question by the biographies which the editor has supplied; but nothing can be more suggestive than actually to follow the authors at their work. Their treatment of the great St Matthew, now in the Uffizi, provides us with a typical instance. The picture was formerly attributed to Lorenzo di Bicci, but was claimed without hesitation by Crowe and Cavalcaselle for the school of Orcagna; and, as Milanesi showed that Mariotto di Nardo di Cione-believed to be of the same family-had been commissioned to paint a San Matteo for the hospital in which this picture was thought originally to have hung, it was concluded that Mariotto was its author. Evidence, however, has since been brought to light showing that Orcagna, while painting a St Matthew for the Consuls of the Arte del Cambio, fell ill, and that the completion of the work was entrusted to his brother Jacopo. The authors remark that the work bears the device of the exchange-golden coins in a medallion—and justly claim that their original observations are fully endorsed.

The picture divides itself into three equal parts, of which the central is occupied by the life-size figure of St Matthew, while of the lateral portions each depicts two scenes from his life. To one of these the authors refer in terms of the highest praise; it is a grand composition of four figures of tall proportions, full of life and character, and in the pure Giottesque style'; praise is given to one of the figures in another panel; of the other two, the subjects only are given, and in one case incorrectly. The error deserves notice, for the importance of the picture, and of the problem connected with it, would naturally act as an incentive to the greatest vigilance and care; and to find St Matthew's martyrdom described as a decapitation, when, in fact, he is stabbed in the back while standing before the altar, shows con

clusively that the scene made little or no impression upon the authors. They originally connected the work with Orcagna, because they discerned in certain passages a quality beyond the reach of a painter of the second rank. They passed over, with insufficient attention, the passages in which it was not attained. Yet the inequality at once of conception and of execution to which Crowe and Cavalcaselle thus negatively testify, not only becomes increasingly obvious when the picture is carefully reviewed, but is clearly evidence of the first order in favour of their own theory with regard to its authorship.

The inaccuracy here noted is surprising, chiefly because it is connected, as we have seen, with a picture to which problems of a peculiar interest attach. The mistake, however, is characteristic of the authors; several instances, indeed, might be quoted, in which they admire the presentation of one subject where the artist clearly intended to depict another. It was undoubtedly in technique and execution that their chief interest lay; and, though they rightly discerned that it belonged to them as historians to consider also the mind of the artist in its relation to the subjects he was required to present, and the changes and developments to which, as time went by, these subjects themselves became liable, they were unable to bring the same spontaneity of insight to bear upon this aspect of their work; and it will always be felt that they have treated it with a somewhat heavy hand. This again may best be judged from a quotation. In their description of the paintings at Assisi they write:

'But the frescoes of the Upper Church do not merely tell the story of art, they were intended to declare the abstinence, the piety, and the miracles of St Francis. And a sketch of these from the legend may be welcome to the reader.'

A condensed life of St Francis here follows, by no means always correct in its relation to the frescoes it is meant to illustrate. The quotation of its opening sentence will suffice :

'Son of Pier Bernardone, a rich citizen of Assisi, St Francis was born to affluence, but preferred, even in those years in which the passions prompt youth to the pursuit of pleasure, the exercise of charity.'

Possibly the authors shrank from the task of recreating