anything but milites, though the omission is certainly very harsh.

XLIII. 3. de transitione. See above, ch. xl. § 4.Galliam Cisalpine Gaul. See ch. xvii. § 4.

4. movere inde in calidiora, &c. The corn was now ripening fast in the warmer plains of Apulia. "Hannibal broke up from before Geronium, descended into the Apulian plains, and whilst the Roman army was still in its winter position, he threw himself on its rear, and surprised its great magazine at Cannæ." Arnold, from Polyb. iii. 107.

7. aversa a Volturno vento, i. e. facing north-west. The Vulturnus was a wind so-called from Mount Vultur, and is probably the same as the Scirocco.

XLIV. 2. a. aditum aquatoribus, &c., "The Aufidus flowing near both the camps, afforded to the watering parties access to its banks, more or less convenient according to their respective positions."

b. ex minoribus castris. This camp was formed by Æmilius on the left bank of the Aufidus, fronting Hannibal on the western bend of the reach, within which the battle was fought.

c. trans Aufidum. On the opposite side of the Aufidus to that on which Hannibal's camp was. See ch. xxvi. § 4.

4. hic Varro.-ille: Æmilius.-usu cepisset. usu capere, is "to acquire a right and title to anything by long use or possession."

XLV. 1. trans flumen, i. e. across from the right or southern, to the left bank, where the lesser Roman camp was. The Aufidus was a small river easily fordable, in the dry season, at almost any point.

2. inconditam turbam, i. e. the aquatores.

4. a. postero die. On the second of August according to

the Roman calendar, but the Roman reckoning was probably six or seven weeks in advance of the true season.

b. copias flumen traduxit. Varro now led his forces across from the larger camp on the south side of the river, and formed a junction with the troops of the lesser camp. The general course of the Aufidus is from south-west to northeast, but near Cannæ it makes a very large reach to the south. Within the bend formed by this reach, on the northern bank, was the battle-field. The larger Roman camp was on the southern bank, about a mile higher up the river than the lesser camp, which was on the northern bank, just where the bend to the south begins. Hannibal's camp was at the south-eastern corner of the bend, a little distance from the stream. When the two armies were drawn up, Hannibal's line faced nearly north, with its flanks resting on the curve of the river. The Romans stood on the chord of the arc, facing nearly south, their flanks likewise resting on the river. Arnold takes a different view of the field of battle, and supposes that it was on the southern side of the Aufidus, (see his History, vol. iii. pp. 137-143,) between Cannæ and Canusium.

XLVI. 3. Gallis Hispanisque scuta ejusdem formæ erant. Their shields were long, but not wide.

4. alius habitus: alius reliquus. See ch. lvi. note 4. 6. Hasdrubal. His name does not occur after this time in Polybius or Livy.

7. ventus...multo pulvere in ipsa ora volvendo. In this part of Apulia the sea breeze rises pretty regularly at noon. Niebuhr thinks it quite probable that Hannibal may have so arranged his men as to take advantage of this wind blowing the dust into the faces of the enemy. There is a story that he even caused the ground to be ploughed ; but this is of course mere fable.

XLVII. 1. ad evagandum, to spread out, so as to attack in flank.

4. a. constabant, "stood firm," "kept their ground." b. impulere hostium cuneum...ab tergo hostes, "The Roman columns on the right and left, finding the Gaulish and Spanish foot advancing in a convex line or wedge, pressed forwards to assail what seemed the flanks of the enemy's column; so that being already drawn up with too narrow a front by their original formation, they now became compressed still more by their own movements, the right and left converging towards the centre, till the whole army became one dense column, which forced its way onwards by the weight of its charge, and drove back the Gauls and Spaniards into the rear of their own line. Meanwhile its victorious advance had carried it into the midst of Hannibal's army; it had passed between the African infantry on its right and left; and now whilst its head was struggling against the Gauls and Spaniards, its long flanks were fiercely assailed by the Africans, who, facing about to the right and left, charged it home, and threw it into utter disorder." Arnold, iii. pp. 142, 143.

c. æquavit frontem, "fell back so as to form a level front."

XLVIII. 2. Quingenti ferme Numidæ. Polybius knows nothing of this, and the whole story is a fiction.

4. Hasdrubal, qui ea jam parte præerat. Hasdrubal's post was on the left wing with the Gauls and Spaniards; but after having almost destroyed the Roman right wing, he had ridden across to his own right to aid the Numidians who had been skirmishing indecisively with the Italian cavalry. Having effected this, he next wheeled round and completed the ruin of the Romans by breaking in upon their rear, while they were still engaged with the Gaulish and Spanish foot in front.

XLIX. 2. a. Equitum pedestre prælium...poterant. These horsemen fought on foot, the victory of the enemy being no longer doubtful, in such sort as might have been expected; for being vanquished, they chose to die where they stood rather than flee, and the victors, angry at their delaying the completion of the victory, slew those whom they could not put to flight.

b. mori in vestigio. See Bk. xxI. ch. xxxv. § 10.

4. a. prætervehens equo, "riding by." Prætervehens is from prætervehor, a deponent verb. The more usual expression is prætervectus equo.

b. reus iterum e consulatu. See note, ch. xxxv, 2. a.

7. a. nulli fugientium insertus agmini, i. e. neither of the three bodies of men just mentioned.-insertus: the text is very uncertain; editions vary between infestus, which gives no sense, and immistus.

b. quadraginta quinque millia, &c. Livy's estimate of the loss of the Romans is the lowest. Polybius reckons it at 70,000 slain. In ch. lix. it is estimated in round numbers at 50,000.

c. tanta prope civium sociorumque pars. The proportion of Roman citizens and allies was nearly the same. d. aliquot annis ante consul. In the year B.C. 221.

L. 1. a. Aliensi cladi. See Bk. v. ch. xxxviii.

b. alterius morientis prope totus exercitus fuit, "almost all the army followed the example of the other consul, and perished.” — morientis...fuit: see a similar construction, Bk. xxi. ch. ii. 1, "adeo prope omnis senatus Hannibalis erat."

2. nuntium...mittunt, i. e. to those in the lesser camp on the other side of the river.

5. a. P. Sempronius Tuditanus. Two years after this, Sempronius was curule ædile (Bk, xxiv, ch. xliii,), and the

year after (B.c. 218) prætor, at Ariminum (Bk. xxIV. ch. xliv.) He captured the town of Aternum, or Atrinum (Bk. xxiv. ch. xlvii.), and was continued in his office for the two following years. In B.c. 209 he was censor, and in B.C. 204 consul, with the conduct of the war in South Italy against Hannibal. He was repulsed with some loss before Croton, but shortly after gained a complete victory over Hannibal, who was obliged to seek shelter within the walls of the town. (Bk. xxix. ch. xxxvi.)

b. ut ex tua contumelia, &c. Hannibal had always dismissed without ransom his Latin captives, while he retained the Romans as prisoners of war until exchanged or ransomed. See below, ch. lviii.

c. non tu, sc. ita facies.

d. qui se bene mori maluit: a pleonasm for qui bene mori maluit. Cf. Cic. in Cat. i. 2, "Cupio me esse clementem."

8. a. ad sexcenti, "about six hundred." Ad is here used as an adverb, without having any influence on the case. See Bk. Iv. ch. lix., and inf., ch. lii.; Madv. Lat. Gram., § 172, obs. 2.

b. Canusium (Canosa), about nine miles south-west of Cannæ, near the right bank of the Aufidus. It bears a good name for its fidelity to Rome, and the kindness with which it received the remnants of the defeated army.

There can

LI. 3. Hannibali nimis læta, res est visa, &c. be very little doubt that Hannibal estimated the chances rightly, in not yielding to Maharbal's wish. The distance was quite two hundred miles, and no part of the army but the cavalry could possibly have reached Rome in a week. Unless the surprise could have been perfectly complete, which is very improbable, the Romans would only have had to shut their gates, and they would have been safe from any immediate danger,