We must therefore have the following equations, since the state of the shell is the same at every instant,

west.

To understand the meaning of these expressions let us take a particular case.

Let the axis of the revolving shell be vertical, and let the revolution be from north to Let I be the total intensity of the terrestrial magnetism, and let the dip be 0, then

I cos is the horizontal component in the direction of magnetic north.

The result of the rotation is to produce currents in the shell about an axis inclined at a T small angle = tan-1 24πk

w to the south of magnetic west, and the external effect of these currents is the same as that of a magnet whose moment is

The moment of the couple due to terrestrial magnetism tending to stop the rotation is

R3I cos 0.

This loss of work is made up by an evolution of heat in the substance of the shell, as is proved by a recent experiment of M. Foucault, (see Comptes Rendus, XLI. p. 450).

IV. The Structure of the Athenian Trireme; considered with reference to certain difficulties of interpretation. By J. W. DONALDSON, D.D. late Fellow of Trinity College, Cambridge.

[Read November 6, 1856.]

THE formal recognition of philology, as one of the subjects for discussion at the meetings of the Cambridge Philosophical Society, seems to me to impose on those of the members, who have more especially devoted themselves to this branch of academic study, the duty of suggesting as soon as possible some discussion calculated to awaken an interest in this new or rather additional department of our transactions. And as pure linguistic investigation is a sealed book to many, and eminently uninviting to all those, who are not critical scholars by profession, I have thought it best to take an application of philological research, on which I have something new to offer, and which is, or ought to be, both intelligible and interesting to all, who care for the language or the doings of the ancient Greeks.

As the Athenians, at the time when their literature assumed its distinctive form, were pre-eminently a maritime people, it was to be expected that nautical terms would take their place among the most usual figures of speech. Many of their best writers had either, as we say, "served in the navy," or had become familiar with the language and habits of the seaports. Even if the wealthier men had not personally served as strategi or trierarchs, or had not made voyages for profit or pleasure, they had lounged in the dockyards and factories of the Piræus, and seen the triremes put to sea on some great expedition; and if the poorer citizens had not pulled the long oar on the upper benches, they had lived in familiar intercourse with many whose hands were hardened with constant rowing, and whose ears were ringing with the never ceasing drone of the pipe to which they kept stroke in the voyage or the onset of battle. It is not at all surprising then that Attic literature is full of direct allusions to the structure of the ship of war and to all the incidents of sea-life. And in point of fact nothing is more common than the occurrence of nautical metaphors. But although this has been duly noticed, and though much has been written on the subject, there are still some phrases in common use, which have not yet received an adequate explanation, and consequently some passages, which still require to be illustrated by a more complete and accurate investigation of the Athenian trireme. It is my intention, in the present paper, to submit to you some of the conclusions at which I have arrived after a renewed survey of the ancient authorities.

It is a well-known fact that ships of war in the most glorious days of the Athenian republic were mainly, if not entirely, triremes, or galleys with three banks of oars. This convenient form of the rowing-vessel, combining, as it seems, the maximum of speed and power, was invented by Ameinocles the Corinthian about 700 B. C. The elementary form, of which it

was an extension, and which kept its place by the side of the trireme, was the penteconter or single-banked galley with fifty rowers. The short flat-bottomed barges of the earliest seamen were not adapted either for rapid navigation or for warfare. And as soon as the Greek mariners put out to sea either to trade with or to plunder distant cities, they seem to have adopted the long sharp-prowed vessel with its twenty-five rowers on each side. Herodotus says expressly that the Phocæans, who navigated the Archipelago, the Adriatic, and the western Mediterranean as far as Tartessus, used for this purpose οὐ στρογγύλησι νηυσί, ἀλλὰ TEVтηKоVтÓρоιι (1. 163), and the mythical Argo, which represents the first of those voyages, half piratical, half commercial, which the Thessalians made into the Black Sea, was undoubtedly regarded as a penteconter. The tradition generally reckons fifty Argonauts, and it was not without a distinct reference to this, that Pindar describes the dragon killed by Jason as 'bigger in length and breadth than a penteconter, which blows of steel have perfected" (Pyth. iv. 255). In these galleys it is presumed that all the rowers were armed men, and Homer is careful to tell us this in speaking of the penteconters which Philoctetes took to Troy (Il. 11. 227). Whether the ships of the Boeotians, to which Homer gives a complement of 120 men (Il. 11. 16), were biremes, or large penteconters, with double crews, is a point which can hardly be decided; Pliny mentions (H. N. VII. 57), on the authority of Damastes, a contemporary of Herodotus, that the Erythræans were the first to introduce biremes, but we do not know when this form was originally adopted, and it is clear that the galley with two banks was never very common. And Thucydides seems to have understood that the penteconters only were rowed by the soldiers, who in that case were bowmen, so that the other vessels would contain, beside the rowers, who served as archers, some seventy hoplites, who only pulled on an emergency. There is a special reason for coming to this conclusion. Thucydides (1.10) speaks of the Tepivey or supernumeraries in the ships which went to Troy, and limits them to the kings and their suite. But the Scholiast says that this term included all the eπißarai or soldiers on board. Now in the nautical inscriptions published by Böckh, we have a particular class of oars called by this name, ai Tepivew K@Taι, and it is probable that these were intended to be used by the synonymous érißarai whenever additional hands were wanted, to make head against wind or tide.

All things considered, we may take the penteconter as the oldest and most permanent type of the Greek war-ship. Both with regard to the number of the crew, and the vessel's length and breadth of beam, it was the basis or starting-point of the trireme. The crew of the trireme consisted of about 170 rowers and 30 supernumeraries. As the length of the vessel over all from forecastle to poop was greater than that of its keel, there were more seats for rowers in the upper tier than in the two lower tiers, and the Scholiast on Aristophanes (Ran. 1074) tells us that at the stern the first thranite sat before the first zygite, and the first zygite before the first thalamite. It seems indeed that there were 62 Opavirai, or benchrowers, in the highest tier, 54 (uyital or cross-bit-rowers, on the second tier, and the same number of Oaλauiraι, or main-hold-rowers, on the lowest tier. Unless then some of the thranites were employed to work the two great oars, or πndália, at the stern, they must have had four ports on each side more than the lower tiers. Supposing that the penteconter had exactly 50 rowers, it must have been nearly as long as the trireme, for it had 25 ports or

holes for the oars, whereas the corresponding or lower part of the trireme was pierced for 27 holes on each side. And as the interscalmium, or space between the ports, was two cubits (Vitruv. 1. 2), or 3 feet 6 inches, we should require a length of 105 feet above, and 91 feet below, exclusively of the steerage and bow, or parexeiresia. That the trireme and the oldest penteconter were exactly of the same breadth of beam, I will prove directly. And of course the height was not increased more than was necessary for the accommodation of the additional tiers of rowers.

Having regard then to that permanence of numerical arrangements which is so remarkable among the ancient Greeks, we must see at once that the broad-side of the penteconter corresponded to the enomoty or triakad, a body of 25 to 30 men, sworn to act together, and constituting the basis of the Greek military system. Consequently, the whole crew of the penteconter corresponded to the pentekostys, and the crew of the trireme was a lochus, consisting, with the epibata, of four pentekostyes, which was the Lacedæmonian arrangement at the first battle of Mantineia (Thuc. v. 68), or it was two lochi of 100 men each, if we prefer Xenophon's subdivision (Rep. Lac. 11. 4).

In regard to these general features all is plain enough. Our difficulty commences, when we come to speak of the arrangements for seating the three tiers of rowers, and it is here that I hope to clear up some obscurities, and throw a little new light on the subject. Dr Arnold has called this "an indiscoverable" or "unconquerable problem" (Rom. Hist. 111. 572 on Thucyd. iv. 32), and Mr James Smith, in his elaborate and interesting Essay On the Voyage and Shipwreck of St Paul, has proposed a solution quite at variance with the meaning of the Greek words which distinguish the classes of rowers*. Even Böckh, in his Archives of the Athenian Navy, can give us no definite information, and inclines to the erroneous belief that

The following is Mr Smith's transverse section of a trireme. (Voyage and Shipwreck of St Paul, p. 194.)

Besides the objection stated in the text, that this arrangement will not explain the Greek names of the three tiers of rowers, it is impossible to conceive that the best rowers should have been placed on a platform within reach of the enemies' shot.

the rowers of all three tiers were furnished with seats of the same kind attached to the ribs of the vessels. I shall now endeavour to show, I believe for the first time, that the names of the three tiers of rowers accurately describe the manner in which they worked in the ships.

I. The Zygita.

There is a very primitive description of the structure of a Greek ship in the Odyssey v. 243 sqq., but we can infer from it that the ribs were always bound together with cross-beams before they were covered with planks. These cross-beams or cross-bits are called 'kpia in the passage to which I refer, a name elsewhere limited to the planks of the partial deck fore and aft, which till a late period was the only kaτáσтрwμa of a war-ship. As the main-yard is termed the eπipiov in this passage, and as the Christian cross was designated as an kptov, we may conclude that the word implied a transverse or cross direction of these timbers; the root is probably that of ikó-uŋv, and therefore, as we shall see, the word is synonymous with σéλua. These cross-bits are called kλnides in Homer, because, like the collar-bone, they locked together the two sides of the ship. The poets call them σéλuara, a word containing the old root sel or sal, "to go" (New Crat. § 269), and implying that they furnished the means of walking from one end to the other of the undecked vessel. The common name, retained to the last in the Athenian navy, was (uyá, "the yokes" or bridges which joined the opposite ζυγά, sides of the ship. There is a reason for these changes of designation. In a mere pinnace, like that constructed by Ulysses, there would be no occasion for a hold, and the cross-planks might be placed close together, like the foot-boards of a boat. In this case, ἴκρια would be

a.

Thranus, or long stool, placed on the alternate zygon, or supported by the selis, and extending 7 feet amidships.

b. Zygon, or cross-plank, running athwart the vessel at intervals.

e.

Platform for Epibatæ running along the trapher, and 6 feet wide, with bulwarks of 3 feet.

f. Selis, or gangway, fore-and-aft, 4 feet wide.

A-B, (Breadth of beam) = 18 feet. C-D, (Depth) = 12 feet.