H.—ANTHROPOLOGY. 149 
bore a fragment of a stamp of about a.p. 120%, The bridge had a total 
length of about 165 metres (with a fall of 1:06 in the main portion of 
136 metres, or 1 in 129, or 7°75 per 1,000), and a maximum height of 
24°50 ; it has two tiers of arches for the most part, though the last seven 
on the left bank originally had only one. 
A little further up the valley is the Ponte 8. Pietro, belonging to the 
Aqua Marcia. The original arch over the stream in opus quadratum of 
porous travertine, with a span of 15°50 metres, and the lateral arches, 
which were probably of similar material, are completely hidden by the 
later reconstructions. The bridge was first reinforced with concrete faced 
with opus reticulatum and brick; then the whole structure was masked 
by very fine brickwork with buttresses, of the time of Septimius Severus, 
smaller brick arches taking the place of the larger ones; that over the 
stream remained of necessity fairly wide, having a span of 11:20 metres ; 
but those of the lateral openings were quite narrow. A peculiarity is the 
- disproportionate height of the portion of the bridge above the brick 
arches, due to the diminution in the height of the openings produced by 
_ these changes ; and in order to avoid a monotonous effect, pilasters were 
introduced. The whole south-east end of the bridge was reconstructed 
at a still later date, perhaps under Diocletian; a straight vertical joint 
. 
, 
——— Ls 
and a change in the character of the facing show the break. 
Higher up the same valley are the much-ruined Ponti delle Forme 
Rotte (broken aqueducts), which were the highest in the whole course of 
the aqueducts ; the top of that of the Anio Novus was at least 42 metres 
above the bed of the stream, while that of the Aqua Claudia (which is 
upstream of it) is a good deal lower, the difference between the level of 
the bottom of the two channels being no less than 10°93 metres, which is 
a good deal more than usual. Both bridges have been reconstructed, 
that of the Anio Novus in the time of Hadrian !* and that of the Claudia 
by Hadrian and Antoninus Pius," as the brick stamps show, and there is 
no trace of any later work in them. They have collapsed completely, 
owing to the giving way of the cliffs on the right bank of the stream, on 
which no remains exist. 
We now arrive at the valley of the Acqua Nera, which is crossed by the 
Ponte Lupo, the best-known and the largest of the aqueduct bridges in 
this district. 
It has hitherto been believed—and Mr. Newton and I still held that 
view when the drawings were made—that it carried all the four aqueducts. 
Accurate levelling has shown that this is not the case, and that the Anio 
Novus and Claudia both pass under the floor of the valley considerably 
higher up. This, indeed, gives them a much better line than the devious 
course which they would have taken supposing they had passed over 
Ponte Lupo. 
The upper channel is, therefore, that of the Aqua Marcia ; the fall from 
Ponte 8. Pietro is 186°79-182'27, or 4°52 metres in about a kilometre, 
or perhaps more ; for the specus, as usual, runs along the side of both 
valleys for some way both before and after the tunnel through the ridge. 
But there is a problem in regard to the Anio Vetus. At the last shaft of 
120.1.L. xv. 1345-8. 
13 C.I.L. xv. 1019, a. 7. 
14 0.1.L. xv. 223a, 1065, 2309. 
