AFTER I had pretty well explored the neighbourhood of Tivoli on that side towards the Campagna, I was desirous to look at the aqueducts remaining on the other side. About two miles from my inn, at a place where a brook called Fosso degli Arci crosses the road to Siciliano, are several fragments. _Fosso_, as perhaps you know, is the Roman name for a brook. The Roman aqueducts which derive their origin from the valley of the Anio above Tivoli are, the Anio vetus, the Marcia, the Anio novus, and the Claudia. The first we meet with at this place is the Marcia, which is here subterraneous, running along the side of the hill. It passes under that of the Anio vetus where that crosses the road, and is there composed of a mass of rubble, intermixed with some brickwork. We then lose it; but some piers of the same nature, on the steep banks of the brook, probably belong to it. The second is the Anio vetus, which passes here on a higher level than the Marcia, contrary to what takes place at Rome, where the latter is the highest; but it is supposed to be known by its thick crust of deposit, the water of this arriving turbid at Rome, while the others were clear. Yet the crust of the Claudia is of immense thickness, and indeed it seems too abundant in all of them, to be depended upon to distinguish any one. This tartar is deposited at the top as well as at the sides, proving the channel to have been completely filled. We have the evidence of Frontinus to prove, that the water of the Anio vetus might have been brought on a much higher level to the city, had it been supported in its course on numerous arches, like the Claudia and Anio novus. It here crosses a depressed part of the road, over an arch constructed of square stones. The core of the work is of rubble, but there is no appearance of brick. Frontinus says that it is taken from the river at the twentieth mile above Tibur, but antiquaries dispute whether this is twenty miles from Rome, the words _above Tibur_ being added to show its position; or whether its origin be twenty miles from Tibur or Tivoli. The length of its course being forty-three miles, seems too much for one supposition, and too little for the other, considering how much it winds. It is considerably elevated, and Cabral and Del Re assure us, that its level at Tivoli is one hundred and fifty palms, or above a hundred feet above the river; a difference far too great to be obtained in the course of a mile or two; which is all that can be allowed, if the distance be measured from Rome. A few paces beyond the channel of the Anio vetus, the Claudian aqueduct crosses the road on a lofty arch of rubble, faced with brick. There are some blocks of tufo on both sides of the stream, which look as if they had supported an aqueduct lower than any of these, but they are perhaps the remains of an ancient bridge, which is more unequivocally the case with some other fragments of rubble-work. This aqueduct may again be seen among the