famelico lupo committeret; Herodot. 4, 149 тоιyapŵv en avròv καταλείψειν δῖν ἐν λύκοισι ; Diogen. 7, 62 προβάλλοντες κυσὶν ἄρνας.

LUPUS 9, p. 199. Alvar. Cordub. ep. 20 (M. 121, 513 A) atque, ut fabulae ferunt, lupum auribus retinens nec tenere potes, nec vales dimittere.

LUPUS 10, p. 199. Alcuin ep. 99 (M. 100, 310 A) passer aures habet apertas sed, ut video, proverbialis in fabula lupus gallo tulit vocem; Apost. 10, 89; see Tribukait, p. 26.

LUPUS II. For discussion, see Tribukait, p. 26; Ioh. Sar. Polycrat. 1, 13 (M. 199, 412 A) cites and explains Verg. Ecl. 9, 53. LUPUS 13. Hor. epod. 4, I lupis et agnis quanta sortito obtigit, | tecum mihi discordia est; Ovid ib. 43 pax erit haec nobis... | cum pecore infirmo quae solet esse lupis; Hom. Il. 22, 263 οὐδὲ λύκοι τε καὶ ἄρνες ὁμόφρονα θυμὸν ἔχουσιν; Diogen. 7, 63, πρίν κε λύκος δεν ποιμαίνοι.

LUPUS 14. Lact. instit. 5, 3, 23 videlicet homo subdolus voluit lupum sub ovis pelle celare, ut fallaci titulo posset inretire lectorem ; Hier. ep. 147, II sub vestitu ovium latebas lupus; ep. 22, 38 sub ovium pellibus lupos tegunt.

LUTUM I, p. 201. Leo Magn. ep. 34 (M. 54, 802 B) si vero in eodem insipientiae suae luto iacere delegerit.

LUTUM 2, p. 201. Avit. Vienn. ep. 34, p. 184, 2 (Chev.) non se studuerunt de caeno, quo . . . tenentur, evolvere; Aesch. choeph. 697 ἔξω . . . πηλοϊ πόδα ; see J. Koch, p. 33.

MORRIS C. SUTPHEN.

II.-ARISTOTLE'S DE ANIMA.'

We may repeat of French Platonists and Aristotelians what Plato said of the Athenians-when they are good they are most excellent. Mr. Rodier's laborious edition of the de Anima not only supersedes but swallows and assimilates its German and English predecessors, Trendelenburg and Wallace. On every doubtful point he reproduces the opinions of all the ancient commentators, Alexander, Themistius, Simplicius, Philoponos, Sophonias, Priscianus, and the views of all moderns accessible through Zeller or Bursian's Jahresbericht. His own judicial summing up is almost always sane and right, and, where erroneous, can always be checked by the evidence which he supplies.

The constitution of the text is conservative. Mr. Rodier reprints with some interpolations of his own to bring it down to date the critical apparatus of Biehl in the Teubner text. He discusses with inexhaustible patience the emendations of Bonitz, Torstrik, Essen, Bywater, Christ, Kampe, Susemihl, Barco, Wilson, Freudenthal and others, but whenever they involve extensive alterations of the text or venturesome theories of double recensions or interpolation, he finally waves them aside. To minor corrections that seem to restore the sense by a change of punctuation or the altering of a letter or word, he is more favorable, and contributes a few such of his own suggestion. He has made a new collation of E without gleaning much. Following are the chief points of interest in his text:

403, b 17, he retains with E and Biehl the impossible oute ws χωριστά. Cf. p. 152.

404, a 19, he deletes comma after ipnra, which he renders strangely 'on fait remarquer.'

404, b 10-11, he inserts commas before raúras and raúrny to the improvement of the sense.

407, b 28, he retains in spite of Bernays λόγους δ ̓ ὥσπερ εὐθύνας dedwxvia which he tries to justify by the translation 'qui a déjà eu à fournir ses raisons pour ainsi dire en guise de châtiment.'

1’ApιOTOTÉŽOVÇ TEρì vxns. 'Aristote Traité De L'Âme.' Traduit et Annoté par G. Rodier. Paris, Ernest Leroux, 1900.

409, b 20-24, he incloses oi... axedóv in parentheses and inserts a colon after ἄλλων.

410, a 29, he separates xai #pòs by commas, translating ' en outre.' Cf. infra p. 153. The inserted footnote calling attention to this has got mixed with Biehl's note on Torstrik's emendation so as to make it appear that this punctuation and not Torstrik's reading rests on Sophonias.

412, a 16, he retains the perhaps unnecessarily explicit reading ἐπεὶ δ ̓ ἐστὶ σῶμα καὶ τοιονδὶ τοῦτο.

sense.

417, b 6, he keeps eis avrò where els avrò is better suited to the In actualization the thing moves, if it can be said to move, to its (real) self. Mr. Rodier's 'en lui' can hardly be got out of his text.

426, a 27, he reads with Simplicius and Plutarch ei dǹ ovμ¢wvia pwvý Tís éori for ei d' and renders strangely 'comme une certaine espèce de voix est accord.' Cf. infra, p. 159.

427, a 10, he keeps with Biehl μía dúo, suggesting, however, μía dúo which, though harsh, gives the required sense.

428, a 24, φανερὸν τοίνυν ὅτι οὐδὲ δόξα μετ ̓ αἰσθήσεως . . . . φαντασία ἂν εἴη διά τε ταῦτα καὶ δῆλον ὅτι οὐκ ἄλλου τινός ἐστιν ἡ δόξα ἀλλ ̓ ἐκείνου ἐστὶν οὗ καὶ ἡ αἴσθησις. So Mr. Rodier prints, connecting διά τε ταῦτα with what precedes. The lack of any construction for re seems to give him no concern, though he lightly remarks that we might read ye. Something is wrong. If one cared to emend, the whole could be smoothed out either by dropping 8λov or reading or diλov or, and, though this is not indispensable, changing or to oral. Two reasons will then be alleged against the identification of dóga and αἴσθησις, the foregoing διά τε ταῦτα, and also the fact that it involves the (intolerable) supposition that the object of doga and alooŋois is the same, which he proceeds to refute. Below, 428, b 8, Mr. Rodier retains the vexatious parenthesis ἀλλὰ ψευδὴς ἐγίνετο, ὅτε λάθοι μεταπεσὸν τὸ πρᾶγμα of which he gives precisely the explanation tentatively proposed at the end of Wallace's note, remarking at the same time that Wallace's corrections are unnecessary. 429, b 7, he accepts Bywater's excellent suggestion d' auroù. 429, b 13, cf. infra, p. 155.

430, b 17, in place of ἀλλ ̓ ᾗ αδιαίρετα he proposes and reads ἄλλῃ adiaípera, which makes the sentence read smoothly, but leaves the connection with the following hopelessly obscure, a fact which he tries to disguise by a long explanatory parenthesis in the translation. The general meaning of Aristotle is plain enough, but the

wording is desperate and can be cured only by rewriting the passage.

430, b 25, he retains Tv airíor which Zeller (Aristotle, Trans. vol. II, p. 105) plausibly explains as a blundering dittography of ἐναντίον.

The not infrequent anacolutha of the de Anima and the hopeless passages which could be cured only by extensive changes, Mr. Rodier generally leaves, after discussion, translating them defiantly according to his final judgment of the general meaning.

The translation which accompanies the text is almost always right, and in precision and definiteness is, barring a few slips, a great improvement on Wallace. An extensive use of the bracket disfigures the page, but distinguishes most helpfully the literal version from the additions demanded by French idiom, or inserted to bring out the sequence of thought as conceived by Mr. Rodier. The following are the chief passages where he seems to have erred, or where at least difference of opinion is permissible: 402, b 8, óμoiws dè kåv eï tɩ kolvòv ädλo katŋyopoîro—' et de même tout autre attribut commun que l'on pourrait en affirmer.' En is misleading. The question, as Alexander rightly takes it, relates to any predicate that is used as a general term, not merely to any other general predicate of Cov.

402, b 22, ἐπειδὰν γὰρ ἔχωμεν ἀποδιδόναι κατὰ τὴν φαντασίαν περὶ τῶν συμβεβηκότων, ἢ πάντων ἢ τῶν πλείστων, τότε καὶ περὶ τῆς οὐσίας ἕξομέν τι λέγειν κάλλιστα. Here κατὰ τὴν φαντασίαν does not mean d'une façon conforme à ce que l'expérience manifeste,' but simply 'in sensuous presentation.' Wallace's 'to the mind's eye' is substantially right, though it errs in implying that the presentation must be always representation. Karà is probably used somewhat as in καθ' ἑαυτὸν (apud animum) ζητεῖν; or as in 427, b 23, κατὰ δὲ τὴν φαντασίαν ὡσαύτως ἔχομεν ὥσπερ ἂν εἰ θεώμενοι. Mr. Rodier argues that the logic requires his rendering. Aristotle, he thinks, could not mean to say that the essence can be inferred from the ovμßeßnκότα. He means that the possibility of explaining (amodidóvai) the συμβεβηκότα καθ ̓ αὑτὰ from the essence is an a posteriori confirmation that the essence has been correctly defined. Otherwise, too, the following yàp is pointless. This is hypercritical. The passage is one of many in which Aristotle states that the definition is often best approached through a survey of particulars. (Zeller, Eng. Trans. I. 172). This process is virtually if not strictly induction (Zeller, 1. 269). The Kaì of TÓTE Kaì and the future goμev are inex

plicable on Mr. Rodier's interpretation. The yàp that follows in πάσης γὰρ ἀποδείξεως ἀρχὴ τὸ τί ἐστιν did not trouble Simplicius (15.9) and need not us. It loosely assigns the reason for the emphasis laid on κátora. The sequence is: (and it is important to define ovoía well) for the what is it is the starting point of all proofs and (here we have Mr. Rodier's idea) definitions that are not accompanied by concrete knowledge of the accidents, are empty and verbal. There is no real difficulty in the unprecise use of ȧrodidóvai (cf. 406, a 27), and we need not introduce the distinction between συμβεβηκότα and καθ ̓ αὑτὰ συμβεβηκότα.

403, b 17, ἐλέγομεν δ ̓ ὅτι τὰ πάθη τῆς ψυχῆς οὔτε ὡς χωριστὰ τῆς φυσικῆς ὕλης τῶν ζῴων, ᾗ δὴ τοιαῦθ ̓ ὑπάρχει, θυμὸς καὶ φόβος, καὶ οὐχ ὥσπερ γραμμὴ καὶ ἐπίπεδον. It is a pity that Mr. Rodier follows Biehl's text here which drives him to a forced unnatural translation inconsistent with his punctuation. Oure ws and kai oux are impossible correlates here. Obviously we must read with the majority of MSS and editors οὐ χωριστά or αχώριστα. The meaning is that the man, qua such; i. e. qua, e. g., Ovpós and póßos, are ȧxplora, inseparable, even in thought from their material embodiment, and not like the line which qua line is separable in thought from physical matter. This is the interpretation of Simplicius (whose reference of ToLaura Mr. Rodier misunderstands), and of Themistius. It is easy, though not necessary, to read ye, instead of 8, with U and Simplicius.

404, b 21, eri de kai aλλws: 'Platon dit aussi.' The name of Plato should not be mentioned in connection with these fooleries of Xenocrates except where Aristotle explicitly attributes them to him.-405, a 16 yoûv: 'en conséquence'; rather: at any rate.405, b 26, διὸ καὶ τοῖς ὀνόμασιν ἀκολουθοῦσιν. May not this mean not that they 'raisonnent d'après les noms,' but that they etymologize to suit their respective theories? The phrasing of Cratylus, 436, b, εἴ τις . . . ἀκολουθοῖ τοῖς ὀνόμασι seems against it, but the general tenor of the discussion in the Cratylus favors it, and diò kaì is certainly clearer so. Their physical theories are no reason for their etymologizing, but do explain the particular etymologies in which they seek support for the respective doctrines.

406, b 2, ὥστε καὶ ἡ ψυχὴ μεταβάλλοι ἂν κατὰ τὸ σῶμα : in spite of the Greek commentators may this not mean 'within the body' rather than 'comme le corps'? This gives point to the following antithesis: (if it can move in the body) it would follow that it can also, kai, go forth from the body and return. The same thought