It is very interesting to notice how this contrivance worked. It was fitted with two cylinders, each having a piston connected by a beam. This beam raised and lowered each piston alternately, and with the help of valves--which only opened the way of the jet--propelled water to the fire, but not continuously. The method must have proved very inefficient, especially when compared with the constant stream thrown by the modern fire-engine. Indeed, it is this power to project a steady and continuous stream which chiefly differentiates the modern fire-engine from such machines as Hero's siphon. How far this siphon or any similar contrivance was used in ancient times we cannot say; but no doubt buckets in some form or other were the first appliances used for extinguishing conflagrations. Whenever mankind saw anything valuable burning, the first impulse would be to stamp it out, or quench the flame by throwing water on it; and the water would be conveyed by the readiest receptacle to hand; then when men had discovered the use of the pump, or the squirt, they would naturally endeavour to turn these appliances to account. In some places the use of water-buckets was organized. Juvenal alludes to the instructions of the opulent Licinus, who bade his "servants watch by night, the water-buckets being set ready"; the wealthy man fearing "for his amber, and his statues, and his Phrygian column, and his ivory and broad tortoise-shell." Then Pliny and Juvenal use a term--_hama_--which signifies an appliance for extinguishing fires; but the true rendering seems to be in dispute, some translators being content to describe it simply as a water-vessel. Pliny the Younger refers to _siphones_, or pipes, being employed to extinguish fires; but we do not know how they were used, or whether they resembled Hero's siphon. In fact, the earliest references to fire-engines by Roman writers are regarded by some as being merely allusions to aqueduct-pipes for bringing water to houses, rather than to a special appliance. And from Seneca's remark, "that owing to the height of the houses in Rome it was impossible to save them when they took fire," we may gather that any appliances that may have been in use were very inefficient. A curious primitive contrivance is described by Apollodorus, who was architect to Trajan. It consisted of leathern bags or bottles, having pipes attached; and when the bottles were squeezed, the water gushed through the pipes to extinguish the flames. Augustus was so enterprising as to organize seven bands of firemen, each of which protected two districts of Rome. Each band was in charge of a _tribunus_, or captain, and the whole force was under a _præfectum vigilum_, or prefect of the watch; though what apparatus they employed--whether buckets or pipe-bags, syringes or Hero's siphon--we do not know.