Two projecting stones, moulded, and perforated with a square opening, on the sides of the doorway, have been supposed to be intended to support an external temporary door; but one does not understand the object of such a door inclosing the inner one and all its ornaments; and as no similar instance can be produced, we must, I believe, be content to leave their purpose unexplained.
The fragment by the fountain, usually called the temple of Diana, must be of still later erection. The order is composite; the earliest ascertained example of which is, I believe, the arch of Titus, at Rome. A number of fragments are collected in it, which mostly announce the period of the decline of the art; but there is one which is completely Greek, and which probably belonged to some more ancient edifice. The principal part of this ruin is what once was a large vaulted room, perhaps a covered court; but most of the vaulting has disappeared, and its present beauty depends, not so much on its architecture, as on the beautiful colour of the stone, on the morsels of antiquity collected there, and which form a sort of museum, and on the dark green of the fig-trees which hang loosely about the walls, and give an air of freshness and coolness even in a hot summer’s day. There are three recesses at the farther end, and a dark covered passage on each side, of which I do not comprehend the object, but I know no reason to suppose it to have been a temple.
The situation of this ruin is very pleasant, in the midst of a public garden, close by a copious spring of delightful water, which supplies the town. This garden is the finest thing of the sort I have ever seen. The columns and balustrades which adorn the fountain, and the basins made for the reception of its waters, extend all through it, and there are abundance of stone seats, vases, and statues. The character of art is no where lost, but it is a beautiful character of art, and the more so, because all the parts are consistent, and there is no appearance of pretence or affectation. Every thing is part of one design; whereas, in England, where we have such ornaments, they are too detached, and seem to have dropt from the clouds, rather than to belong to the scene. Even at the Tuilleries the distribution is by no means sufficiently apparent, they want more architecture to support them. The trees here are of a good size, and uncut, principally the linden.