The oars varied in size according to the bank on which they were used, of course increasing in length as they ascended. Their length in a trireme is stated at from 9 to 9¼ cubits, but no mention is made of the part of the vessel to which they belonged. An oar, however, of only fourteen feet in length could have been of no service unless used on the lowest rank and almost on a level with the water. Those employed in the smallest wherries of the Thames are from twelve to fourteen feet long. Thucydides,[383] in describing the attack of the Peloponnesian commanders on the Piræeus, the harbour of Athens, remarks, “The plan was that each sailor should take with him his oar, his cushion and his thong, and go by land from Corinth to the sea over against Athens, and, proceeding with all speed to Megara, should put off with forty triremes which happened to be at Nisæa, their naval station, and sail immediately for Piræeus.” From these remarks it may be inferred that none of the oars belonging to a trireme were of greater weight than one man could carry for a distance of four or five miles; and that only one man was stationed at an oar, unless “his oar” might be construed as meaning the oar under his charge. But though no mention is made of different-sized oars having been used on board of a trireme, there can be no doubt that the oars of the ancients differed far more in size than those of the river barge or man-of-war sweeps as compared with the sculls of the Thames wherry of modern times. [Sidenote: Mode of rowing.] This is clear from the fact that while various ancient writers mention oars of 9½ cubits in length, Athenæus distinctly states that the oars belonging to Ptolemy Philopator’s large ship were thirty-eight cubits long.[384] Here we have a specific account of oars varying from fourteen to fifty-seven feet in length, the latter requiring to have lead embedded in their handles as a counterpoise to the weight outside the rowlock.[385] Besides, it is clear that the oars must have increased in size according to the banks on which they were employed. In the case of the oar fifty-seven feet in length, if worked from a great height a large portion of it would require to be inboard—say nineteen feet against thirty-eight; and even the one-third would not, at a line of nine feet above the water, be sufficient as a counterpoise, unless the shoulder of the oar were of unwieldy thickness or heavily weighted by lead. In all single-banked vessels the oar worked on the gunwale, and was kept in its place by means of a leather thong. In larger galleys it passed through an oar-port. Various ancient writers assert that there was only one man to each oar, and add, that he sat, when rowing, on a single bench or small stool attached to the ribs of the vessel, and within a very short distance from the _scalmus_ of his oar.