Writers differ with regard to the number of men placed at an oar; but herein there is, practically also, no difference of opinion, for the number of men at an oar must depend on the size of the vessel. A jolly-boat has never more than one man at an oar, but a launch has two, and, in the phraseology of our own time, launches thus rowed are called “double-banked,” although this expression was not applicable to the galleys of the ancients. Nor is there much difference of opinion with reference to what was meant by a bireme. Apart, therefore, from direct testimony, there can be no doubt that a bireme (as a rule) meant a vessel, not with two rowers to an oar, but with two banks or tiers of oars. Some writers, as already explained, have maintained that the oar-ports were placed directly one above the other; but this could only be possible in vessels such as the bireme. On the other hand, a careful examination of Trajan’s column, and of other remains of ancient sculpture, proves that the oars were placed _obliquely_ over each other: thus, on coins of the Emperor Gordian, two tiers of oars so placed are very conspicuous. If, however, any doubts still exist on this point, they are entirely set at rest by the recent discoveries of Layard and others. The illustration on the next page, a copy of an alabaster slab found at Kouyunjik, the site of ancient Nineveh, clearly shows the oar-ports arranged obliquely on the broad side of the galley. The only further question is, to what extent this principle could be applied, so as to place at work the vast number of rowers some galleys are said to have carried, and also to afford accommodation for the troops and stores. [Sidenote: Mr. Howell’s plan.] [Illustration: GALLEY FROM KOUYUNJIK (NINEVEH).]