The city water pipes in Rome were usually of baked clay or lead; copper was sometimes used and also hollowed stone. For the large supply conduits leading to the city the Romans used covered channels with free water surfaces, rather than pipes. Perhaps this choice was a matter of economics, for apparently they could make lead pipes up to 15 inches in diameter. While pipes can follow the profile of undulating ground, with the pressure increasing in the lower areas, channels cannot. They must slope continuously downwards, because water in channels does not normally flow uphill; and the grade must be flat, from 1 in 60 in small channels to perhaps 1 in 3,000 in large ones, to keep the water speed down to a few feet per second. Thus the main supply channels or aqueducts had long lengths of flat grade and where they crossed depressions or valleys they were carried on elevated stone bridges in the form of tiered arches. At the beginning of the Christian era there were over 30 miles of these raised aqueducts in the 250 miles of channels and tunnels bringing water to Rome. The channels were up to 6 feet wide and 5 to 8 feet high. Sometimes channels were later added on the tops of existing ones. The remains of some of these aqueducts still grace the skyline on the outskirts of Rome and elsewhere in Europe similar ruins are found.
Brick and stone drains were constructed in various parts of Rome. The oldest existing one is the Cloaca Maxima which follows the course of an old stream. It dates back at least to the third century B.C. Later the drains were used for sewage, flushed by water from the public baths and fountains, as well as street storm run-off.
The truly surprising aspect of the achievements of all the ancient hydraulic artisans is the lack of theoretical knowledge behind their designs. Apart from the hydrostatics of Archimedes, there was no sound understanding of the most elementary principles of fluid behaviour. Sextus Frontinus, Rome’s water commissioner around A.D. 100, did not fully realize that in order to calculate the volume rate of flow in a channel it is necessary to allow for the speed of the flow as well as the area of cross-section. The Romans’ flow standard was the rate at which water would flow through a bronze pipe roughly 4/3 inch in diameter and 9 inches long. When this pipe was connected to the side of a water-supply pipe or channel as a delivery outlet, it was assumed that the outflow was at the standard rate. In fact, the amount of water delivered depended not only on the cross-sectional area of the outlet pipe but also on the speed of water flowing through it and this speed depended on the pressure in the supply pipe.
The Romansused all of the following to make water pipes EXCEPT（）
A．earth
B．wood
C．copper
D．stone