apsidal chapels is one of the most striking and most successful features of Mr. Richardson’s design. As will be seen from the ground-plan, however, these are features that do not proceed from the interior arrangement so much as features to which the interior arrangement is conformed. Even when viewed from the outside the undeniable power and picturesqueness of the group is marred by the suggestion of something forced and arbitrary in their arrangement. There are precedents in Romanesque architecture for such a disposition, among them “the great triapsal swing” of the twelfth-century churches of Cologne, though evidently the example that inspired Mr. Richardson was the chevet of Clermont in Auvergne, which he has followed even to the introduction of the mosaic above the springing of the arches. All these, however, are much simpler than the apse designed for Albany. What Mr. Richardson doubtless had in mind was to reproduce the effect of the ring of chapels that forms the chevet of a French Gothic cathedral, without reproducing Gothic forms. But the flying buttresses that radiate from the apse of a French Gothic cathedral determine and bound the chapels that fill the spaces between them, and, by making these appear integral parts of the main structure, save them from the look they would otherwise have of extraneous appendages. III It seems, then, that the question of style in a modern cathedral is not to be determined according to the individual preference of a designer for round arches or pointed, for openings traceried or plain. If the problem he is working at has been successfully solved heretofore, he is not at liberty to ignore this solution because it falls without the limits of the historical period he has proposed to himself, and to content himself with an incomplete solution. Of course this remark does not apply as a criticism to Mr. Richardson’s design for Albany, prepared under limitations that he was compelled to observe, but which the competitors for the Cathedral of New York were at liberty to disregard. Whether he was right in so far sacrificing the monumental character of his interior to the monumental features of his exterior, is not a practical question for designers of whom no sacrifice in either direction is demanded. There are very noble examples of vaulted architecture in the Romanesque period--examples which it will be glory enough for the architect of the Cathedral of New York if he succeeds in equalling without slavishly imitating. But in all these there is a lack of that complete correspondence between the interior and the exterior structure that makes the organic unity of a true cathedral, and that was attained for the first time in the thirteenth century, after a series of tentative experiments embodied in these very Romanesque buildings.[F] It is by no means necessary for an architect to revert to these experiments because he does not sympathize with the expression of strained intensity