Mr. Howell, a comparatively modern writer,[395] adopts many of the views of Vossius, but differs from him in that he maintains that it would be impossible to work with effect more than five banks of oars. As his views are more in accordance with our own than those of any other writer, we shall refer to them at greater length, although differing also from him in some of his most important conclusions. And here it ought to be stated, that all modern writers, Mr. Howell included, appear to have given too little consideration to the facts, that ancient galleys varied quite as much in size as the vessels of modern times, that their power or dimensions were not, in all cases, measured by the number of their banks of oars, and that in proportion to the number of rowers the capacity of the hold would require to be increased. A war-galley would be comparatively useless if she had not ample capacity for fighting men, and for their munitions of every kind, besides stores, including water. All these points must, however, have been fully considered by the ancients, who evidently saw, when they wished to have more than thirty oars on each side of a galley, that an increase could not be obtained on the single-bank principle without constructing her of an unwieldy length in proportion to her depth and breadth, and thus sacrificing an unnecessarily large amount of space. Consequently, they invented the bireme, whereby they could, in little more than the length required for fifteen oars, place double that number without any corresponding sacrifice of space; while in the trireme, they would in nearly the same length obtain space for three times the number of oars, and secure for the use of the soldiers and stores ample accommodation and any extra length they might desire. Mr. Howell, in discussing this principle as applicable to quinqueremes, shows that, by adopting the oblique ascent, the rowers of the first and highest bank can be placed so as not to interfere with the rowers on the second, their oars having space to play free of the benches before them. “That a bank or bench of oars,” he adds, “never contained more than five oars, I think, can be proved, whatever the size of the galley was, whether a bireme or trireme, up to the galley of Philopator, which had forty banks, nine feet being the highest point from the water for the _scalmi_, from which they could pull with effect.”[396]