I seem to detect a similar cause at work. Iphigenia relates how Artemis bore her away from Aulis to the land of the Tauri ναοῖσι δ' ἐν τοῖσδ ̓ ἱερίαν τίθησί με, "Αρτεμις, ἑορτῆς, τοὔνομ ̓ ἧς καλὸν μόνον· So reads the best MS. Now we have here beginning with v. 35 a sentence which gets no further than the three words ὅθεν νόμοισιν ἑορτῆς: there is a relative clause subjoined to νόμοισι by τοῖσιν ; there is another relative clause subjoined to doprîs by hs: at the end of this clause we come to a full stop, and begin an entirely new sentence introduced by an inferential particle. The heroes who undertake the defence of the text say that we have here an aposiopesis. Aposiopesis is a comforting word; but the sphere of the figure so named is limited by conditions which here preclude it. In cases of aposiopesis it is requisite that we should be able to form a notion how the speaker was about to complete the sentence which he breaks off: this is obviously necessary to the understanding of the situation, because it is the thought of the suppressed words which causes to arise in his mind the emotion which restrains him from uttering them. But here the spoken part of the sentence consists only of three words, and we cannot tell the meaning of the first or the construction of the second: we cannot tell whether 80ev means since which time or for which reason; we cannot tell whether the inflexion of vóμoiơi means with or by or in or for or to or because of: much less then can we guess how the sentence would proceed. Hence this reading is now defended, I think, by no scholar of repute: a few have tried to mend matters by eliciting Toloid' from the rooid' of the second-best MS. in v. 35. We do indeed thus get a sentence, with detaι for its principal verb; but sense is as far off as ever. The sacrifices, as the words ὄντος τοῦ νόμου καὶ πρὶν πόλει inform us, were instituted, not upon the arrival of Iphigenia, but long before; and there is no imagin able way in which Artemis' delight in them can have Leen caused by anything recounted in the narrative which has preceded. Therefore other conjectures have been proposed, of which I mention Weil's as the most ingenious and plausible. He retains Toiσly, strikes out 'Apreμis as a gloss on eeά, and replaces it by xpμeo. This, to be sure, is good sense; yet the hypothesis is precarious. The officiousness of the scribe is surely excessive and not complimentary to our intelligence: what goddess but Artemis could we dream of Again how often eá and eós recur in this play for Artemis I do not waste time in counting, but it is very often indeed; yet nowhere else has the scribe importuned us with the explanation. I prefer then to look for light to the ductus litterarum; and I will suggest that if there once stood in the text the letters ΝΟΜΟΙΣΙΝΟΘΝΕΙΟΙΣΙΝ it required only the inversion of the two consecutive letters NE, and the infinitesimally slight change of T for 1, to produce ΝΟΜΟΙΣΙΝ ΟΘΕΝ ΤΟΙΣΙΝ ; and that the only way to inake metre of this was to arrange the words as they now stand in the MS., ὅθεν νόμοισι τοῖσιν. But the sentence νόμοις ἵν ̓ ὀθνείοισιν ἥδεται θεὰ where Artemis delights in strange rites is surely just what is wanted; and it is curious to note how closely van Herwerden has approached to this sense with his conjecture ὅπου νόμοις ὠμοῖσιν. --Α E. HoUSMAN, AISCHYLOS, Agamemnon, 1227-1230. νεῶν τ' ἔπαρχος Ιλίου τ ̓ ἀναστάτης This passage has exercised the ingenuity of scholars and many clever emendations have been proposed: nevertheless I believe that the MSS. reading (which I give with the exception of κἀκτείνασα, the inevitable correction of the irrelevant kal Kтeivaσa) gives better sense than any reading that has been suggested. It is to be observed that Kassandra utters her prophetic revelations in two forms. She first in her lyrical laments perplexes the chorus with endpyeua Oérpara (1072-1177), a series of dark pictures, drawn with a few quick strokes; but then she unveils the oracle (1178) and presents to the puzzled elders the same pictures, but painted in more detail and brought into more intelligible connexion. It is important to note that the general order of the prophecies delivered in iambics is the same as that of the lyrical utterances, and that the correspondence is emphasised by the repetition of the same or similar words. Compare for example 1087 with 1186, 1093 with 1185, 1095sqq. with 1217sqq., &c. Thus paidpóvous (1229) may suggest λουτροῖσι φαιδρύνασα in 1109. Now in 1115-17 we find the remarkable conception of the net ; ἢ δίκτυόν τι Αΐδου; ἀλλ ̓ ἄρκυς ἡ ξύνευνος, ἡ ξυναιτία φόνου, which is plainly a covert reference to the ἄπειρον ἀμφίβληστρον which Klytaimnestra afterwards describes (1382). We might expect that Kassandra would again refer to the net in her unveiled oracle, which is intended to explain the obscurities of her former words. I would suggest that there is such a reference in the present passage and that the word din was an old word for 'net', 'casting-net', being closely connected with δίκτυον and related to δικεῖν as βόλος is related to βάλλω. The meaning then will be; He knoweth not what words the tongue of the lewd hound will speak and how having stretched out with blithe soul a net of stealthy ruin she will succeed therein (in achieving the ruin) with an ill success.' After ola it was unnecessary to express ws with exTelvaσa; and there is no difficulty in understanding κύων from the preceding κυνός. In support of this interpretation I would further remark that in 1. 1611 there may well be a suggestion of the secondary sense of δίκη : ἰδόντα τοῦτον τῆς δίκης ἐν ἕρκεσιν. Μoreover the phrase δίκην ἄτης λαθραίου has a close parallel in a passage in the Prometheus, 1077 : κοὐκ ἐξαίφνης οὐδὲ λαθραίως εἰς ἀπέραντον δίκτυον ἄτης ἐμπλεχθήσεσθ' ὑπ ̓ ἀνοίας.1 AISCHYLOS Agamemnon, 1310. τόδ ̓ ὄζει θυμάτων peoTiwv. Does not the Chorus in these words intend to suggest Ov-éσrns?—J. B. BURY, EURIPIDES, Orestês, 399. Μ. Τί χρῆμα πάσχεις ; τίς σ ̓ ἀπόλλυσιν νόσος ; Μ. δεινὴ γὰρ † ἡ θεός †, ἀλλ ̓ ὅμως ἰάσιμος. (399) The reading of the MSS. in 399 cannot be right. For (1) it is extremely unlikely that Euripidês would 1 ἐμπλεχθήσεσθ' made me entertain for a moment the idea of reading λéaσa for λétaσa in Agamemnon 1229; but there is no reason to change the MSS. reading. Nevertheless if Aischylos wrote πλέξασα the proximity of γλῶσσα would inevitably cause it to be corrupted to λέξασα. speak of λύπη as a goddess, and (2) ἡ θεός is quite incongruous with láσuos. Of these objections the first is serious, and the second is fatal to needs. Weil proposed to read on, a reading which not only introduces an unlikely word but departs altogether from the MSS. I venture to propose that the right reading is δεινὴ γὰρ ἦθος ἀλλ ̓ ὅμως ἰάσιμος. oos is used by medical writers of the character or nature of a disease, and is consequently appropriate here.-J. B. BURY. NOTES ON THE SCHOLIA OF THE PLUTUS. (1) Dübner, p. 3246. ll. 17-22. 3. τὸ ὴν ἀπόστροφον λαμβάνει, ὅ ἐστιν ἀντίστροφον. ἔστι δὲ ὥσπερ καὶ ἄλλως τοὐναντίον δρῶν· καὶ τοῦτο γὰρ ἀντίστροφον δέχεται. ἔστι γὰρ τὸ ἐναντίον δρῶν. ἡ δὲ ἀπόστροφος ἀντίστροφος καλεῖται. ̓Αρίσταρχος γὰρ τοῦτο σημειοῦται. ὅ ἐστιν παρ' ἡμῖν ἀπόστροφος, ἀντίστροφος αὐτῷ καλεῖται, This has no meaning as it stands, or when it purports to be a scholium upon they in the lines ἢν γὰρ τὰ βέλτισθ' ὁ θεράπων λέξας τύχῃ, 5. μετέχειν ἀνάγκη τὸν θεράποντα τῶν κακῶν. It contains, however, as kernel a real ancient oxónov from the school of Aristarchus. It is an explanation of a critical mark by Aristarchus, and represents the master's way of taking un dpâv in 1. 4. It has nothing to do with in the previous line. Perhaps the earliest form of the note was τὸ χ' τοῦτο γὰρ ἀντίστροφον δέχεται· ἔστι γὰρ τὸ ἐναντίον δρᾶν. The master's critical mark x is possibly concealed in the kai before TOUTо, or the Kal may be due to the compiler of the scholia in their present shape who often thus sweeps together into one note different ways of saying the same thing. In any case the first part of the scholium is also a corrupted explanation of the mark used by Aristarchus, perhaps the explanation of another pupil, viz. τὸ μή ἀντίστροφον λαμβάνει· ἔστι δὲ ὥσπερ καὶ ἄλλως τοὐναντίον δράν. (the master) takes un as reversing the sense; it it as also might be said in another way to do the reverse.' The words 'Αρίσταρχος γὰρ τοῦτο σημειοῦται make still another distinct ancient note, while the d ἀπόστροφος ἀντίστροφος καλεῖται and the closing words are later additions. ‘He (2) On the eσmideiv of 1. 9 the Ravenna MS. has what seems to be a like ancient scholium. · τυμολογει θεσπιῳδεῖν. . . αρα τὴν θέμιν ἐκεῖ τὰς μavтelas ǎyewv. (cp. Dubner, p. 325a. ll. 44-46). This ought to be emended thus: ἐτυμολογεῖ θεσπιω δεῖν παρὰ τὸ κατὰ τὴν θέμιν ἐκεῖ τὰς μαντείας ᾄδειν. 'He (the master) gives as the derivation of teori de TV that at Delphi he (Apollo) chants (adew) his prophecies in accord with justice (Oéμis)'. (3) 1. 66 on ὦ τῶν, ἀπαλλάχθητον• ἀπ ̓ ἐμοῦ the Ravenna MS. has preserved another note of Aristarchus' school. ὅτι οὐ πρὸς ἕνα μόνον τὸ ὦ τᾶν, ἀλλὰ πρὸς δύο. The three obliterated letters are best replaced by τὸ χ. "The x because the & Tav is addressed not to one only but to two.' (4) The genesis of the unintelligible parts of the scholia is well illustrated by the Ravenna scholium on 1. 74. νὴ τοὺς θεοὺς ἡμεῖς γ', ἐὰν βούλῃ γε σύ. It is as follows:—ἐ άν) ἰδοῦ ἀπολύομεν· τοῦτο δέ φασιν ἀπολύσαντες αὐτόν. This is really an explanation of the closing words of the next line Α. μέθεσθέ νύν μου πρῶτον. Β. ήν, μεθίεμεν. The other MSS. have gone still further astray as may be seen by reading Dübner, p. 3286. ll. 45-52. (5) An editor's critical mark is lost in a scholium on μlapúтate ȧvôoŵv in 1. 78. (Dübner, 329a. 1. 27) ὅτι ἐν σχήματι ἀνδρὸς ἐφάνη ὁ Πλοῦτος. (6) We may be sure that the lost critical mark in a scholium on Ouσelev in 1. 137 was Aristarchean, 87 καὶ ἐπὶ τὸ θυμιᾶσαι τὸ θύειν. There is not the same certainly in either of the following (7) 1. 169 Κορινθίας· δηλοῖ ὡς ἀπὸ Λαΐδος Κορινθία yàp hv. He (the master) explains how it comes from Lais; for she was of Corinth;' or (8) on l. 151 οὐδὲ προσέχειν τὸν νοῦν· ὅτι οὐ μόνον οὐ προσδιαλέγονται, ἀλλ ̓ οὐδὲ ὁρῶσι σχεδόν. (mark) 'that not only do they not accost them, but almost do not even look at them.' (9) An Aristarchean note may, however, be recovered from two scholia on 1. 155, (see Dübner, p. 331b. 1. 42) τὸ χ ὅτι ἀρσενικῶς ὁ πόρνος καὶ θηλυκῶς· ξητείσθω δὲ εἰ καὶ οὐδετέρως. (10) A few lines above (p. 3316. 11. 27-30) a con vincing emendation results from reading aidotov for &TоTOV, converting the following accusative into the genitive, and the verb érioelew into the substantive ἐπίσειον. The οὐ before σφόδρα in the MSS. is then explained.-W. GUNION RUTHERFORD. CAES. B. G. iv. 17. THE BRIDGE OVER THE Rhine, (above, p. 168, 9.)-1 have always taken the fibulae to be horizontal bolts piercing the beams (trabes) at right angles, one on each side of the junction of the beams with the piles, so as to prevent the beams slipping away from the piles. This latter effect is exactly expressed by quibus disclusis et in contrariam partem revinctis, which (upper and lower) pairs of piles being (thus) kept apart and at the same time bound together, the construction was made tighter by the action of the water'. The action of the stream on the upper face of the upper piles tended to increase their slope and thus approximate the higher portions of the upper piles with the lower piles, which would perhaps not be materially affected by the stream. The inner bolts kept them apart, the outer bolts bound, as it were, the upper and lower pairs of piles together, and this latter action was in the opposite direction (in contrariam partem) to the former action. Utraque and utrimque I take to refer to both pairs, (upper and lower). Hence I require only four bolts for each two pairs. I cannot see how quibus can be separated from disclusis, which would thus be left without any subject expressed. The only point on which I feel some doubt is whether fibula would be used for a simple bolt-a straight piece of iron or wood passed through the beam. A bar of iron with the two ends bent at right angles to the centre and with one end fixed in the trabes at each side of the junction with the piles (tigna) would better answer the notion of a clasp (fibula) but would not be so simple and effective a mode of construction. And had they iron on the spot? Cross-pieces-thirty to forty feet long-as Napoleon supposes, seem to me utterly alien to the notion of fibulae. Whether a bridge so constructed would require braces between each pair of piles in the direction across the stream, is a question for engineers. The directa materia, if properly fastened to the trabes and to the banks, would I suppose act as a brace. The depth to which the piles were driven and the nature of the bed of the river would have an important bearing on the matter.-H. J. ROBY. NOTE ON HOR. Epod. xvii, 32, nec Sicana feruida Uirens in Aetna flamma. Uirens is the reading of the best MSS.; others have furens or urens. Editors generally retain uirens, though they do not agree as to the exact meaning of the word, of which they offer the three following interpretations: (1) That it is a synonym for uigens, (2) That it refers to the green colour of the sulphurous flames, (3) That it is equivalent to fulgens. In support of this rendering they cite Plautus, Menaechmi 818 (Vahlen), Uiden tu illi oculos uirere? Here however Ritschl proposes to substitute lurere for wirere, and I would ask if a similar change should not be made in the present passage of the Epodes. Lurens seems precisely the word required to describe the lurid fires of the volcano, and Ovid (Met. XIV. 791) actually uses lurida as an epithet of sulfura. J. H. ONIONS. MACROB. Sat. vii 4 § 7 nec tertium defuisset exemplum, ita esse vitandam ciborum varietatem ut varia solent vina mutari. So it runs in Eyssenhardt's edition; the context convinced me that the last word should be vitari; cf. c. 5 § 15 et ideo varia vina vitantur. On turning to Jan's edition I find that MS. P has mutari, but Jan with the margin of MS. H reads vitari. This is not the only place in which Eyssenhardt has forsaken his predecessor for the worse.-J. E. B. M. APSYRTUS. In the preface (§ 3) to his Mulomedicina Vegetius speaks contemptuously of the style of his predecessor Apsyrtus. From Suidas (s. v. "AupToS) we learn that Apsyrtus wrote in Greek. The Latin version found by Prof. Wilh. Meyer (cod. Monac. 243) at München, confirms the censure of Vegetius. Prof. Meyer promises critical editions of Apsyrtus, Vegetius and Pelagonius (Archiv für lat. Lexik. III 588, citing Sitzungsber. d. bayr. Akad. d. Wiss. Philol.-philos. -histor. Klasse, 1885, p. 395).—J. E. B. M. DIFFICULTIES IN JUVENAL.-It is with extreme diffidence that I venture to re-suggest some of the old difficulties connected with Juvenal's Satires, and I do so really in the hope of eliciting a solution of them from English authorities on the subject. The first I will notice is the custom of the sportula as represented by Juvenal, which in two or three important particulars is quite unknown to Martial, who beyond a doubt was an eye-witness of what he described. Juvenal for example represents it as distributed in the morning. Professor Friedlander calls this an unimportant difference. I cannot agree with him. Considering the rigid uniformity of Roman society life under the Empire, such a difference in two writers describing Roman manners at about the same period seems to me most puzzling. Suppose two writers describing Cambridge University life should one of them represent the time for boating practice as regularly beginning at 9 A. M., and the other at 2 P. M., it would surely raise doubts as to the accuracy of the information of one of them. I leave out of the question the probabilities of the case, that is, the improbability (1) that a dole, which appears to have been a substitute for a dinner, should be given in the morning; and (2) that a practical people should pay wages at the beginning and not at the end of the day. I confine myself to the difficulty raised by the difference of representation between the two writers. It is clear, I think, from a comparison of Martial X. 70, 5-14, with Juvenal I. 127 fol., that the one regards the distribution of the sportula in the evening, the other in the morning as a matter of course; that Martial knew nothing of a morning, Juvenal nothing of an evening distribution. Now surely such a discrepancy as this cannot be considered unimportant. But there are other discrepancies. Juvenal represents the sportula as received by women. Martial knows absolutely nothing of such a custom. There is only one line (quoted by Professor Mayor in his note on Juv. 1. 95) which wears even a semblance of supporting such a supposition. But in this line it seems pretty clear on careful consideration of the passage that sportula there signifies not the distribution, but the amount of a dole. The former signification would spoil the progress of the epigram. Again, Juvenal represents the sportula as received by men of high position and wealth, nobles, consuls, and rich freedmen. Of this there is no hint in Martial. Yet it is alike incredible that, if he knew of such a practice, he should not utilise it in his epigrams, and that he should not know of it, if it existed. For if he laughs at the rich and noble, as he does, for attending the morning levées of court favorites and generally of men more influential than themselves, he would surely laugh louder at them for receiving a dole. This fact, one may remark in passing, viz. that great men performed the salutatio, might suggest to one not intimately acquainted with the life of the period the inference that they received the sportula, which in the case of clients was so intimately connected with the salutatio. Now Professor Friedlander apparently considers that these discrepancies are sufficiently accounted for by the fact that Juvenal belonged to a rather later time than Martial. But this allows an extremely small margin for a remarkable change to take place in the habits of Roman society. If the date of any of the satires may be fixed by allusions contained in them, that of the first satire cannot be later than about A.D. 100, the date of the banishment of Marius (cf. I. 49, VIII. 120). Professor Friedlander says rather vaguely that the changes alluded to probably began first after the death of Domitian. But Martial can hardly have left Rome earlier than A.D. 98, and was in constant communication with the city until he died some few years after. This leaves at most a space of five or six years for these changes to have developed in. But such changes must surely have developed gradually. If therefore the difference between the two poets in their allusions to this subject is due to difference of periods, it seems necessary to assign a date to Juvenal considerably lower than the one usually assumed. And this leads on to the whole question of the date of the author of these satires, and what evidence we have for fixing it. However, I confine myself to one difficulty here, content for the present if I can provoke an authority to solve it.-H. M. STEPHENSON. MOMMSEN (Röm. Gesch. v. p. 145) says that Vespasian disbanded five of the German legions after the Batavian revolt, viz. I. Germ., IV. Maced., XVI. Gall., V. Alaud., and XV. Primig. That he disbanded the first three is almost certain. I. was probably joined with VII. Galbiana, afterwards called Gemina: IV. and XVI. were replaced by IV. Flavia Felix and XVI. Flav. Firma. But there seem to be several reasons against the disbanding of the two last. (1) Vespasian only created three new legions, the two already mentioned, and II. Adjutrix. If therefore he disbanded five, the frontiers would be left with two legions less than before. This, in view of the state of Britain, the East, Germany an the lower Danube, is very improbable. (2) The V. and XV. were the two legions besieged by Civilis in Vetera, and it was not till Vocula's final abandonment of them and the surrender of I. and XVI. that they at last gave in. Compared with the mutinous conduct of the other legions, theirs was praiseworthy. (3) We know that one legion was annihilated by the Sarmatae under Domitian, Suet. Dom. 6., but with the exception of the five in question no legions are missed up to the middle of the second century (C. I. L. vi. 3492) except XXII. Deiot., which was never in the west, XXI. Rapax almost certainly disbanded by Domitian after the rising of Saturninus (Marqdt Staatsw. ii. p. 450 n. 8) and IX. Hispana in Britain. It seems therefore necessary to infer that out of these two legions, probably V. Alaud. was the one destroyed under Domitian, and replaced by I. Minervia. Grotefend (Real-Encycl. IV. p. 896) with much probability assigns the disbanding of XV. Prim. to Trajan. E. G. HARDY. He ON THE WORD γλυφίδες. -The word γλυφίδας in Il. iv. 122 and Od. xxi. 419 is usually explained as the 'notch' of the arrow into which the bow-string fits. It is so explained by Hesychius s. v. in Et. Magn. and by Eustath. on the two places in Homer, and translated 'notch' merely by all translators. Moreover, in Apoll. Rhod. iii. 282, we find yλupídas μέσσῃ ἐνικάτθετο νευρῇ, where it can have no other meaning. In Homer this meaning makes fairly good sense, but that it has not been thought to be entirely satisfactory is clear from the fact that Hayman, ad. Od. 1.c. understands it to mean the extremities on either side of the arrow nick'; and Merry says, it is not impossible that by yλupídas are meant some slight projections at the nock of the arrow, for the fingers of the right hand to press against in drawing the bow.' Both these interpretations have the merit of making excellent sense and explaining the plural, but the etymology of the word appears to me entirely to preclude their adoption. The word occurs again in Hdt. viii. 128, τοξεύματος παρὰ (or περὶ τὰς γλυφίδας περιειλίξαντες καὶ πτερώσαντες τὸ βιβλίον ἐτόξευον ἐς συγκείμενον χωρίον. Here the meaning assigned to the word in Homer does not make sense, and Schweighäuser consequently explains it to mean 'the grooves into which the feathers fitted.' adopts rep for Tapà, after Valcknär, from Aen. Tact. 31, where the story from Hdt. is quoted, and he translates (after Casaub.) 'circa sagittae crenas epistolam volventes et alas ei applicantes.' I may add that crena seems to be a word of uncertain meaning, and here merely represents yλupídas without explaining it further than as 'notch' of some sort (see L. and S. sub xnλ). Liddell and Scott, in their last edition, adopt Schweigh.'s explanation of γλυφίδες, but by a strange blunder translate πτερ. Tò B. having made it (ie. the letter) serve feathers to the arrow,' which is not only a mistake, but an absurdity. Paley, ad. Il. 1.c. and Eur. Or. 274, also adopts this explanation. It does not however suit Homer well, for there is little point in saying he drew the grooves for the feathers.' In Abicht's Herodotus I find a third explanation. He keeps rapà, the MSS. reading, and says in his note, near the notches at the butt end of the arrow, which served as a holding place for the fingers while drawing the string.' We should rather expect the dat. after apà in this sense, but the explanation is very plausible, and is supported by Suid., who has under γλυφίδας, τὰς χηλὰς τῆς ἀκίδος αἷς τὴν νευρὴν προσάYouer, and by a second explanation of Et. Magn. Tàs παρὰ τοῖς πτεροῖς ἐντομὰς τοῦ βέλους. It makes very as good sense too in Homer, and as to Eur. Or. 274, πτερωτὰς γλυφίδας, either of the last two meanings, or even the first (though not so well) will suit, because feathered notches' is clearly put by synec. doche for feathered arrows.' If then yλupidas is confined to mean some particular notch, I would 6 suggest this explanation, and consider that Apoll. Rhod. is here an ignorant imitator of Homer,' a character often assigned to him by Buttmann, and sometimes unjustly, I think. I am however informed by Mr. Cecil Smith that, as far as one can judge from sculpture and paintings, no arrow-Assyrian, Greek, or Roman-has these notches assumed by Abicht, and that certainly none of the Egyptian reed arrows at the British Museum have it. This is strong negative evidence, therefore I am rather inclined to believe that yλupides, which merely means 'a carving,' may have more than one meaning, that e.g. it may mean notch at the end of the arrow in Homer, and grooves for the feathers (or perhaps notches at the side) in Herodotus. This is hinted at by the double explanation in Et. Magn. and by Schweigh. himself, who says that not the crena in infima parte sagittae' is meant, but the crena into which the feathers were inserted. I may add that in Apoll. Rhod. iii. 218 the word occurs as a term of architecture. R. C. SEATON. = NOTE ON Aikaiрov.—In a pamphlet by Mr. V. Ball of Dublin (Animals and Plants of India. 1885.) treating of the fauna and flora described by Herodotus, Ktesias, and others, mention is made (p. 22) of the Sikapov, which the author with great probability identifies with the scarabaeus sacer: deriving the name from the Arabic zikairon concealer. Ktesias describes it as a bird, the size of a partridge's egg, which buried its dung in the earth. The Greeks commonly called it Sixalov, an obvious corruption found in Aelian De. Nat. Anim. iv. 41. Is this mysterious animal the original of the still more mysterious being haleout pidikaios = the bird Dikaios (avis cognomine Justa' Zoëga) which appears in the fragmentary Sahidic Acts of Andrew and Paul (printed in part by Zoëga Catal. Codd. Coptt. Vatt.)? The Apostles send it into a city from which the Jews have excluded them, and commission it to raise a dead youth. In other respects it seems to be treated as an ordinary bird.-M. R. JAMES. CAVILLOR. In discussing this word in Wölfflin's Archiv iv. 78f, Funck says that its first sense is 'to jest,' its later meaning 'to quibble.' I venture to think that the second is the older, as it is the more special, sense. The word and its connexions are rare in early Latin all pre-Ciceronian literature seems to show only four exx, all from Plautus, viz. Aul. 639, Truc. 685, Stich. 226, Miles 641. In the first, aufer cavillam non ego nunc nugas ago,' the sense of quibble' is by far most appropriate; the second and third passages will admit of either sense; the fourth is bracketed by Ribbeck, though not by most editors. I think that the idea of 'quibbling in jest' suits all four contexts. Festus (Verrius) explains it exactly thus c. est iocosa calumniatio. The derivation is, of course, uncertain, but if the word be connected with caveo (as is stated in Pareus' lexicon and as Prof. Nettleship has lately conjectured anew), the sense of quibble' will come very naturally. The objection to the derivations usually given is that they do not suit the meaning. Cavillor always means rather mild witticism: καύαξ (Fick) and κόβαλος (de Saussure Voyelles p. 106) denote violence, while calvo presents phonetic difficulties.-F. HAVERFIeld. JOHN CLEMENT.-If we are ever to see an adequate history of classical philology in England, many hands must combine to glean from the field of letters every scattered ear. Few would look for an elaborate account of Cleinent, once page of Sir Thomas More, and afterwards Greek Professor at Oxford, in Teubner's Bibliotheca. Yet there it stands, in the Preface to the Third Edition (1868) of the Anacreontica, by Valentine Rose.-JOHN E. B. MAYOR. [Since the above was in type two writers have spoken of Clement, Mr. Thompson Cowper in the Dict. of National Biography, and Prof. Rendel Harris in his book on the Leicester Codex. Neither of them is acquainted with Rose's researches.] So TO POUR OIL ON THE TROUBLED WATERS.-This proverb points to a practice which has, I believe, been lately tried with life-boats in tempestuous weather, and which was not unknown to the ancients, as we read in Pliny, N.H. ii. 106 (Quem fallit) omne (mare) olco tranquillari, et ob id urinantes ore spargere quoniam mitiget naturam asperam lucemque deportet ? Plutarch (Nat. Qu. p. 914 F) makes it a subject of inquiry, Διὰ τί τῆς θαλάττης ἐλαίῳ καταῤῥαινομένης γίνεται καταφάνεια καὶ γαλήνη; compare also his treatise, De Primo Frigido, p. 960. In both passages he combats Aristotle's explanation of the fact, and gives another which he considers more satisfactory. Bede must have forgotten this in his account of the miraculous preservation of Eanfleda, bride of King Oswy, from the danger of shipwreck (H. E. iii. 15.) St. Aidan, giving his blessing to the priest who was sent to escort her from her home in Kent; entrusted to him a flask of holy oil, which he told him would be required to still the storm on their return. So it : turned out the vessel was on the point of sinking, when the priest bethought him of Aidan's remedy; the oil was poured out, and the sea was immediately smooth. Sicque factum est, is Bede's comment, ut vir Dei et per prophetiae spiritum tempestatem praedixerit futuram, et per virtutem ejusdem spiritus hanc exortam, quamvis corporaliter absens, sopivcrit. Bede is careful to cite his authority, and evidently regards the incident as entirely supernatural, no less so than the miracle of Germanus (H.E. I. 17), qui adsumpto in nomine sanctae Trinitatis levi aquae spargine fluctus sacvientes opprimit. May we not look upon this as a typical case? One cause of the casy acceptance of miraculous stories in the Middle Ages, was that Roman civilization was itself miraculous to the barbarian mind. NOTE ON C. R. I p. 192 col. i.-Mr. Margoliouth writes: A story very similar to that quoted by you is to be found on p. 154 of the Vulgar Arabic Grammar of Caussin de Perceval ed. 5, with some variation in the numbers and the mise-en-scène. Caussin does not state from what book he obtained the Arabic text, but the story is Eastern enough.' Mr. Stout, Fellow of St. John's College, Cambridge, refers me to Grimm's Märchen No. 7 with the notes. The exact story will probably be found in some Italian or French collection of tales.-J. E. B. M. II. CLASSICAL EDUCATION IN FRANCE. (Letter from a French University Professor.) Il existe dans les lycées des examens de passage d'une classe à l'autre. Ces examens deviennent plus sérieux d'année en année, et d'après les circulaires ministérielles doivent continuer à le devenir de plus en plus; ce qui prouve qu'ils ne le sont pas encore au plus haut degré. Il y a pourtant chaque année un certain nombre d'élèves refusés à l'entrée de diffé rentes classes; il y en a même qu'on fait redescendre dans le courant du premier trimestre. Un examen plus important que les autres a lieu au sortir de la 4o. C'est ce qu'on appelle l'examen de grammaire. Il faut l'avoir subi avec succès pour entrer dans la division supérieure du lycée et aussi dans certaines écoles. Entre la classe de Rhétorique et celle de Philosophie, le lycée ne fait pas subir d'examen à ses élèves. Ils vont se présenter devant la Faculté des lettres, et alors commencent les examens vraiment importants et qu'on n'affronte pas sans émotion. Essayons de les caractériser rapidement. Le premier est un examen qui se passe en deux fois, à un an d'intervalle. Au sortir de la Rhétorique, on subit une première série d'épreuves, après la Philosophie, une autre série. Le candidat qui a subi avec succès les unes et les autres obtient le titre de bachelier ès lettres. Ce titre est la condition de l'admission dans les Facultés, dans certaines écoles du gouvernement, et enfin dans une multitude de carrières diverses, sans parler du volontariat dans l'armée, pour lequel le baccalauréat dispense de l'examen spécial. Aussi se fait-il chaque année des milliers de bacheliers dans toute la France. Il y a des Facultés qui voient en une année se présenter devant elles plus de mille candidats pour la première partie, un peu moins pour la seconde. Deux tiers environ des candidats sont ajournés à la première partie, la moitié à la seconde. La plupart reviennent à la charge jusqu'à ce qu'ils réussissent. La première partie de l'examen du baccalauréat consiste en trois épreuves écrites et plusieurs orales; composition française, version latine, et thème allemand ou anglais; interrogations sur le latin, le grec, le français, l'allemand ou l'anglais, l'histoire et la géographie, l'histoire littéraire. Le sujet de la composition française est tantôt historique, tantôt littéraire. La version latine est tirée de Cicéron, de Quintilien, quelquefois de Tacite ou de Lucrèce, etc., mais autant que possible d'un des ouvrages qu'on ne lit pas communément dans les classes. On n'est admis à se présenter aux épreuves orales que si les trois compositions sont passables. La plupart des échecs sont attribuables aux épreuves écrites. Les épreuves orales de latin et de gree portent sur les auteurs qu'on étudie dans la division supérieure du lycée. Chaque candidat indique six ouvrages de chaque langue, parmi lesquels l'examinateur choisit. Comme on est pressé, à cause du grand nombre des candidats, ces interrogations se bornent le plus souvent à faire traduire quelques lignes; tout au plus y ajoutera-t-on deux ou trois questions sur des temps de verbes en grec, sur un détail de grammaire ou d'antiquités en latin. Les bons élèves des lycées s'en tirent fort bien en général, quelques-uns d'une manière brillante ; mais la différence entre les dix premiers d'une classe et les dix derniers est très forte. Cela vient de ce que les élèves faibles n'ont pas été éliminés avec assez de fermeté des classes où leur insuffisance a commencé |