_Cancelli_, or shields of basket work, were placed along the sides of the ships at such a height that the heads of those on board are just visible. The _cancelli_ bore a striking resemblance to the circular basket-work boats still to be found on the upper Euphrates; this supports the supposition that the _cancelli_ may have been used for other purposes, particularly if they were made comparatively watertight, as the function of a shield was not only to protect a warrior in battle, but to help to keep him dry when on shipboard by being disposed along the sides to prevent the spray from entering the ships. A forecastle was constructed upon these ships, and upon each forecastle a look-out man was stationed; and when these structures came to be built of larger dimensions they served to accommodate a number of fighting men who, from their superior position, could throw their missiles with greater effect. The forecastle had the further advantage of serving as a stronghold in the event of an attempt being made to capture the ship by boarding it. Following the Phœnicians, the Greeks are thought to have begun to build their own warships about 700 B.C., perhaps earlier, but it was about that time that the first three-banked warship was launched at Corinth. The three-banked ships were for many years the largest in existence. During the fourth century B.C. shipbuilding was practised extensively, four-banked ships being built at Chalcedon, five-banked at Salamis, and six-banked ships at Syracuse. Ships of ten banks, according to Pliny, were ordered by Alexander the Great, and about 300 B.C. ships having twelve banks are said to have been built for Ptolemy, and fifteen-banked ships for Demetrios, for a battle near Cyprus. Ptolemy Philopater, who ruled in Egypt from 222 to 204 B.C., is alleged to have had a forty-banked ship of a length of 280 cubits or, reckoning the cubit at 18 inches, of 420 feet, and a beam of 57 feet. While increasing the size and number of oars, it would, nevertheless, be impossible to augment to any appreciable extent the speed at which these ships could be rowed, and the more unwieldy would they become, and the more difficult would it be to keep steering way upon them. Again, the assertions of the historians are so contradictory that it is a thankless task to attempt to reconcile all their stories, especially as they depended much upon hearsay for their information. For that reason, therefore, a great deal that has been recorded as to the early ships and their numerous banks of oars is not to be accepted without careful inquiry and verification.