There is nothing which implies that it was passed round the ship externally; and it is not clear how a ship could be bound together in the mode supposed: the "naves longæ," from the weight of the rostra and towers at the extremities, and from their great length, must have been extremely apt to "hog," or fall down at each end; but as the stem and stern posts rose above the rest of the vessel, a simple way of preventing this would be to pass a rope round them, and heave a strain upon it by twisting the parts together, as was done in the military engines called tormenta; and Isidore's etymology of the name 66 tormenta, a tortu dicta," seems to confirm this. Böckh also notices the hypozomata of the great ship of Ptolemy Philopater. I agree with him that the word eλaubave ελαμβανε (took) shews that they were not fixed to the ship; but I do not see anything in the account of Athenæus to prove that they were meant to be applied lengthways, and still less that this was the only mode.

The last proof which he adduces in favour of this hypothesis, is taken from a bronze relief in the public museum at Berlin. It is figured in the Thesaurus Brandenburgicus of Beger (vol. iii. p. 406.), and in Montfauçon.* I have not seen the original bronze, but the figures do not warrant the inference. The rope mouldings are evidently ornamental, and three out of the four do not go round the vessel, but are interrupted by the stem-post. The Victoria and Albert royal yacht has also a rope moulding exactly where the antique has it; it would scarcely be a fair inference to suppose that it was meant to hold a crazy

* Antiquité expliquée, tom. iv. pt. 2. p. 214. pl. cxxxiv.

vessel from falling to pieces. I cannot, therefore, see any reason for supposing that ships were undergirded lengthways, a mode which must have been as impracticable as it would have been unavailing for the purpose of strengthening the ship. It would appear from the Attic Tables that the hypozomata formed a regular part of the gear of every ship, and that they were laid up in the magazines.

In the account of St. Paul's Voyage, I have adduced examples to show that the practice of undergirding ships is still occasionally resorted to.

I have only a few remarks to offer on the capabilities of the antient ships in working to windward. Paul Hoste has observed that no person could infer à priori that a vessel impelled by the wind could sail to a place which, in respect to that from which it started, was directly to windward. This may be true, but on the other hand no person who tried to impel a vessel by sails could avoid making the discovery for on the most unfavourable supposition, that of a sail set at right angles to the keel, it would be discoveredthat even though the wind did not blow directly upon it, so long as the sail was full, the vessel would go a-head, and of course, if the yard could be braced, that she could go nearer to the wind than at right angles to it, or within eight points. We have no information as to the exact angle with the wind which an antient ship could sail. It must, however, have been less than eight points, but more than six, the usual allowance for a modern merchantship in moderate weather. I have, therefore, in calculations taken seven as the mean between these

my

extremes; and I cannot suppose it would be much greater or less.

Notwithstanding the imperfect manner in which the ships were rigged, they appear to have made excellent passages. Pliny has enumerated several which would be considered respectable in modern times. Thus he tells us that the prefects Galerius and Babilius made quick passages from the Straits of Messina to Alexandria; the former arrived on the seventh, the latter on the sixth day; that in the following summer Valerius Marianus made the passage from Puteoli, on the ninth day, "lenissimo flatu ": he also mentions passages from the Straits of Hercules to Ostia, in seven days; from the nearest port of Spain, in four; from the province of Narbonne, in three; and from Africa, in two.* Upon these passages Admiral Beechey offers the following remarks:

"It does not appear that there is any mistake in the numbers here mentioned by Pliny; for the instances are all of them consistent with each other, one only being below 140 M. P. per day, and another 143; two examples afford 160; two 175 and 185. The lowest of these rates of sailing may be reckoned at between six and seven M. P. per hour, and the highest at something less than eight; giving a mean of seven M. P. per hour, which would be reckoned a good one for ships of the present day."— Appendix to Trav. in Africa, p. xxxviii.

The most rapid run which I have met with is mentioned by Arrian, in his Periplus of the Euxine (p. 5.), who stated that "they got under weigh about daybreak," apavτες μεν ύπο την έω; and that by mid-day they had come more than 500 stadia, xa noμED TO THE μeonßpias

*Hist. Nat. Procem, ad lib. xix.

σταδίους πλείονας η πεντακοσιους, that is, more than ffty geographical miles, which is at least eight miles an hour. Major Rennel, in his Observations on the Geography of Herodotus (p. 678.), supposes that the

average rate of a day's sail was only thirty-seven miles. Vessels navigating unknown coasts, such as those of Africa when Herodotus wrote, and putting into port at night, may not have made more in a day; and it would be no objection to the credibility of a narrative, were this stated to be the case: but it is absolutely impossible that ships four times as long as they were broad, with a large square sail, could make so little progress with a fair wind; and the foregoing examples prove that they did not. When St. Luke states that the ship sailed from Rhegium on one day and arrived at Puteoli on the following, he tells us that the wind was south (xxviii. 13.). Now, as the course is nearly due north, the vessel was running right before the wind, which to a single-masted vessel is the most favourable point of sailing. The distance is about 182 miles. If we suppose she sailed at the rate of seven miles an hour-the mean of the foregoing examples, the time consumed would be about twenty-six hours, which agrees perfectly with St. Luke's account.

The passage, therefore, from Rhegium to Puteoli, which terminated on the day following that upon which they left it, although a quick one, was by no means unprecedented.

We are apt to consider the antients as timid and unskilful sailors, afraid to venture out of sight of land, or to make long voyages in winter. I can see no evidence that this was the case. The cause of their not making voyages after the end of summer arose, in a great measure, from

P

the comparative obscurity of the sky during the winter, and not from the gales which prevail at that season. With no means of directing their course, except by observing the heavenly bodies, they were necessarily prevented from putting to sea when they could not depend upon their being visible.

In what manner they calculated the distance, as well as the direction of their course, is uncertain. Vitruvius describes what may be termed a perpetual log; that is, a mode of estimating the distance passed over by the revolutions of a wheel (x. 14.). From the manner in which he speaks of it, it appears rather to be a scheme which might be adopted, or the tradition of one which had been in use, than the description of an instrument actually in use. The wheels were, or supposed to be, fixed to the sides of the ships. It appears to be one of those plans that look well in theory; but which the disturbing causes, arising from the inclination of the vessel or the violence of the sea, would render of little value.

The internal arrangement of the rowers in the war galleys of the antients is a problem of great difficulty, as to the true solution of which much difference of opinion exists. No work expressly describing the arrangement is extant, and it is one not well fitted for graphic representation. The incidental notices of antient writers, and the representations on coins, marbles, bronzes, and pictures, however, in a great degree limit the problem, and, as appears to me, when combined with the essential condition of practicability, lead us to the true solution.

I shall, in the first place, notice the solutions which have been proposed by other writers.