with longer oars to the system, and so introduced the trireme. If the number of the zygites in the penteconter was twenty-five a side, and the first bireme was a converted vessel of that class, the number of thalamites, owing to the contraction of the bow and the stern, would necessarily be two or three a side less. Thus we may consider a converted penteconter to have been capable of carrying a rowing crew of between go and 100 men. Similarly a triaconter would have been capable of adding nearly twenty pairs of arms to her propelling power. When, in consequence of the new invention, vessels were expressly built as triremes, we may imagine that for convenience’ sake the benches or zyga would be a little raised, so as to give more room for the raised seat of the thalamites that was fastened on to the floor of the vessel.
The narrowness of the vessels affected the disposition of the rowers in the Greek galleys in a peculiar way. It is evident from the testimony of the ancients that they adhered strictly to the principle of ‘one man to each oar.’ The arrangement seen in mediæval galleys was absolutely unknown to them, and would not have suited them. It belongs to a different epoch and a different order of things, when the invention of the ‘apostis’ had made the use of large sweeps rowed by two or three men possible, and a vessel with sets of three rowing upon the same horizontal plane might be called a trireme, though utterly unlike the ancient vessel of that name.
In the ancient vessel the tiers of oarsmen must have sat in nearly the same vertical plane, obliquely arranged, one behind and below the other. Thus in the bireme the zygite, as he sat on his bench, had behind him and below him his thalamite whose head was about 18 inches behind the zygite thwart and a little above it. Moreover, as his seat was now a little raised, the zygite required an appui for his feet, which was formed for him on the bench on which the thalamite next below and in front of him was sitting; on either side of him his feet found a resting-place. As the zygite fell back during the stroke and straightened his knees, there was plenty of room for the thala-
