Ancient life is not usually considered to have been very cleanly, but it is to the credit of the Romans that as much as 2,200 years ago they made up their minds to reject the water of the Tiber as unfit to drink. They hunted for springs in the mountains, and in the course of a few centuries so many aqueducts were built that Rome had theoretically a better supply of water than any modern city enjoys. Practically, however, the Romans suffered from a peculiar kind of water pilfering. Instead of 400,000,000 gallons daily which the springs furnished, the city received only 208,000,000 gallons. This immense loss, says a careful paper by the Austrian engineer, E. H. d'Avidor, arose partly through neglect of the necessary repairs in the aqueducts, but still more through the water being positively _stolen_. For one of the principal favors by which the State and the emperors were in the habit of rewarding minor services was by granting concessions for the _lost_ water; that is, for the water which escaped through the overflow of the reservoirs, cisterns, and public fountains, or through the defects in the aqueducts and mains. The consequence, of course, was that every landed proprietor who had obtained a concession for the waste water escaping from an aqueduct passing through his grounds was anxious to increase this waste as much as possible--and from this wish to intentional injury was but a step. The overseers and slaves in charge were constantly bribed to abstain from repairing damages which had arisen, or to cause new ones to arise, and these abuses reached such a pitch that one aqueduct (Tepula) brought _no_ water whatever to Rome during several years, the whole having been wasted, or rather abstracted on its way. The irregularities of the water supply were still further increased by the nature of the mains and distributing pipes, which, as I have mentioned, were mere lead plates soldered into a pear-shaped section, incapable of resisting even the most moderate pressure and liable to injury by a common knife, so that any evil-disposed person could tap the main almost wherever he pleased. At a later period, indeed, the Romans appear to have used short clay pipes; lengths of such mains have been discovered, consisting of two-feet spigot and socket pipes carefully laid in and covered with a bed of concrete. These have outlasted all the lead pipes, and are still frequently found in good condition.