HISTORY OF THE TIBER 347 



of the stages of this history. We have seen that the 

 river may have flowed at first south-westward, from 

 its estuary among the Todi hills, across the site of 

 the volcanic district to the sea somewhere north of 

 Civita Vecchia. We have further found that the 

 volcanic eruptions, which gave rise to the long line 

 of heights between Aquapendente and Bracciano, 

 probably blocked up the older channel, turned the 

 stream towards the southeast, and for many miles kept 

 it from once more bending seawards until at last it 

 found its escape across the low ground of the Cam- 

 pagna. Looking from the Monti Parioli up the 

 broad strath, we see the Tiber meandering through 

 its flat alluvial plain which, from a width of a mile 

 and a half, suddenly contracts immediately below us 

 to less than a third of that space. The meaning of 

 this constriction will be understood if we remember 

 the position of the sheet of hard travertine to which 

 allusion has already been made. When the river be- 

 gan to flow across this tract of country, the general 

 level of the ground, not yet reduced by prolonged 

 denudation, was no doubt a good deal above what 

 it is now, and the bed of the stream may even have 

 lain at a higher level than the tops of the present 

 ridges on which Rome is built. After traversing 

 the volcanic plain, the Tiber reached the western mar- 

 gin of this tract near where the Monti Parioli now 

 rise. It then flowed southwards between the slopes 

 of Monte Mario and the edge of the volcanic accumu- 

 lations. It cut its way downward through the upper 

 parts of the tufF, and at length encountered the sheet 

 of travertine near the Ponte Molle. This hard stone 



