THE CLASSICAL QUARTERLY OCTOBER, 1914. ON CERTAIN FRAGMENTS OF PINDAR. THE Fragments of Pindar have not, so far as I know, been adequately discussed since the great editions of Boeckh and Dissen, and as many of them are extremely important, the following remarks are offered in the hope of turning the attention of scholars in their direction. Fr. 1721 interests us, first, since it mentions Peleus and not Telamon as the comrade of Heracles in his attack on Troy. But a second and more interesting feature of this Fragment is the fact, which has, I believe, not been noticed by other students, that this is probably from the same poem as Fr. 227 and seems to follow it directly. Putting the two together, we have one of the longest Fragments of Pindar. The rhythm, it will be noticed, is the same—the dactylo-epitritic. The continuous line of thought thus developed is quite in the manner of Pindar, who often states a general principle and then follows it with a well-known example, which fixes the principle in mind. So O. 1, 36 sqq., the poet says, ‘ In truth it is meet for a man to speak (only) what is honourable concerning the gods, for less is the blame. And touching thee, O son of Tantalus, I will speak at variance with earlier bards. How, when thy father had bidden them to that most lawful feast, offering to the gods a banquet in requital of theirs.’ O.2,23sqq., ‘ When the fate of the gods sends high prosperity to be uppermost again. This saying suits the fair-throned daughters of Cadmus,’ etc. SoO. 7, 26 sqq.; O.9, 28 sqq.; P. 3, 54 sSqq-; and passim. Our own case is much like these. ‘The thoughts of young men, if revolved with toil, find glory. And their deeds, in time, are bathed in light, shining up into the midst of the aether. ... Did not the youth of god-like Peleus shine, by reason of his countless toils? First with the son of Alcmena, etc.’ And the perfect harmony of the rhythm of the two Fragments will be apparent on reading them consecutively. It will be observed that a trochee is lacking at the beginning of Fr. 227. I would suggest inserting the word aiwa, which suits admirably both meter and sense. Compare P. 4, 133, aiiva 8 amo 1 The Fragments are cited according to Christ in his large edition. NO. IV. VOL. VIII. 226 E. B. CLAPP K\clav wpto. It is noticeable that Pindar makes Peleus the companion of Heracles on that famous expedition, as mentioned before, instead of Telamon, as in the more common version of the story. The usual form also is found in Pindar, N. 4, 25; J.6, 27; Fr.50, 51, and often in Greek literature. But Peleus is a favourite hero with Pindar, who never tires of singing of the glories of his august marriage with Harmonia, daughter of Ares, when ‘ The gods did join in the feast, and the happy pair beheld the sons of Kronos as kings on golden thrones, and received their bridal gifts,’ P. 3, 93 sqq. The compound word évréXauwev in 172, I, is an &raké Aeyouevoy in Pindar, though the simple word Adu occurs no less than five or six times in the poems which are preserved, in accordance with Pindar’s predilection for ideas of brilliancy and splendour. In Fr. 177, 4, we notice a metrical peculiarity which calls for some remark. The verse reads aiviypa tapQévov é€& ayprav yrabwv. The diphthong -ov in mapOévov is thus scanned as long, in spite of the initial vowel in the following word. This is exceptional, since it occurs but five times in the whole range of Greek Melic poetry. And four out of these five cases are genitives of the second declension where the termination can easily be emended to -ou’, the elided form of the older genitive ending in -ovo. If, therefore, we read here mapQévor’ instead of mapévov, we secure metrical regularity. The ending -ovo is well attested in O. 13, 35 and P. 1, 39, these two cases proving that Pindar, unlike Homer, does not hesitate to elide the final vowel of the termination -ovo. The elided -oco has been restored by conjecture by Ahrens in N. 9, 55, and by Mommsen in J. 1, 16, and both emendations are generally accepted. The same emendation is easily made in Fr. 84 of Simonides, for the same reason as here. Let me make it clear why this emendation seems to be called for. A long vowel at the end of a word before a word beginning with a vowel is not usually shortened in cases where elision has already taken place, the sacrifice of a short vowel apparently satisfying the poet’s ear. It is worth remarking that if we accept this slight emendation there remains but one abnormal case in Greek Melic poetry, in Pratinas I, 15, jv idov: ade co Sefid, and in this case the sense pause after idov clearly makes the hiatus unobjectionable. Fr. 169 is preserved to us by Plato. In the brilliant speech of Callicles, in the Gorgias 484 B, in which that sophist argues that might makes right, he says, ‘In my opinion Pindar sustains what I assert, in the poem in which he says, ‘‘ Law, king of all, both mortal and immortal, makes use of the utmost violence with overweening hand, and calls it just. I infer this from the deeds of Heracles, since he drove off the oxen of Geryon to the Cyclopean porticoes of Eurystheus, without buying them or asking for them.”’ (It is worthy of note that Plato recurs again to Geryon, in Book VII. of the Laws, 795 C, where he couples him with the other monster, Briareus, as illustrating the principle that the warriors in his commonwealth must be able to use both hands alike.) This Geryon, or Geryoneus, is a little known mythological figure. Most of us learned of him first in the well-known passage in the 2 * i. ae 7 ne OC iti tnt ¢ A game welcor many § and sa W Geryor where before with h sit at o struggl W logical speaks silver s of the Andalu as aS Geryor ‘ Roare Hyperi depths This sk glowins Geryon monste orderec island 1 by Ger same g bodies three a were he Echidn journey arriving ceived | nion of lamon, yund in Peleus s of his id join golden 1 word e word served, dour. emark. -ov in lowing inge of of the 2’, the d here ng -o1o indar, Nn -oOt0. and by The reason called vowel ce, the worth it one d€ col hiatus licles, ‘ht, he ich he tmost deeds ticoes thy of 95 C, g the both ogical n the I A en el nme a ~— TT Te = ee — ON CERTAIN FRAGMENTS OF PINDAR 227 Agamemnon of Aeschylus, where Clytemnestra, in her hypocritical speech of welcome to the returning husband whom she is plotting to kill, refers to the many griefs and anxieties which she has had to endure during his long absence, and says (857-862) in Headlam’s version : ‘Wounds! if my lord had got as many wounds As rumour channeling to us homeward gave him, He had been more riddied than a net with holes. Or had his deaths but tallied with all tales, He might have been a second Geryon, Three-bodied, with a triple coverture Of earth above to boast him.’ We learn at once from this quotation of the most striking peculiarity of Geryon, viz., that he was triple-bodied. Pindar mentions him again in J. 1, 13, where he says of Heracles that ‘Even the fierce hounds of Geryon trembled before him.’ And in Fy. 81 we read, ‘ Thee, O Geryon, I praise in comparison with him. For it is not fitting, when one’s possessions are being robbed, to sit at one’s hearth and be a coward,’ which seems to refer once more to the struggle with Heracles. We are indebted to Stesichorus for further information about this mytho- logical personage. In the third book of Strabo, 148, 11, we read, ‘ Stesichorus speaks thus about the herdsman Geryoneus, that he was born beside the silver spring of the river Tartessus, close opposite famous Erytheia, in a vault of the rock.’ The Tartessus river is the river Baetis, or Guadalquivir, in Andalusia, so that Geryoneus is thus located in the far west, and Stesichorus, as a Sicilian poet, would be specially familiar with his history. The name, Geryoneus, is evidently derived from ynpv@, so that it would signify the ‘Roarer.’ Stesichorus, in the poem quoted, goes on to the effect that Helios Hyperion embarked upon his golden cup that he might cross the ocean to the depths of the sacred dark night, and to his wedded wife and his dear children. This shows that the island of Erytheia (reddish) is situated in the far west, glowing in the rays of the setting sun. But the real locus classicus for Geryoneus is the Bibliotheca of Apollodorus 2, 5, 10, where the story of the monster is told clearly and intelligibly. As his tenth labour, Heracles was ordered to bring the cattle of Geryoneus from Erytheia. The latter was an island near the ocean, which is now called Gadeira. This island was occupied by Geryoneus, the son of Chrysaor and Callirhoe, daughter of Oceanus (the same genealogy is given by Hesiod, Theog. 287 sqq.), and Geryoneus had the bodies of three men, grown together into one at the belly, but divided into three at the flanks and thighs. Geryoneus had a herd of dun cattle, which were herded by Eurytion, and guarded by the two-headed dog Orthros, son of Echidne and Typhon (and therefore brother of the delectable Cerberus). So journeying after the cattle of Geryoneus, and crossing many wild regions and arriving in Erytheia, he camped on Mt. Abas. The dog of Geryoneus per- ceived him and attacked him. Heracles struck him with his club, and when 228 E. B. CLAPP the herdsman Eurytion came to support the dog, Heracles slew him. And Geryoneus finding Heracles at the river Anthemus, driving away his cattle, joined battle with him, and was slain by an arrow-shot. And Heracles put the cattle into the cup of Helios and sailed across to Tartessus, after which he gave the cup back to Helios. And so, after many adventures, he brought the cattle safely to Eurystheus. The character of Geryon seems to have been taken up by the comic poets as typical of the stupid giant who is the predestined victim of the clever giant-killer. His size and limitless voracity must have hugely amused the gallery-gods of the Dionysiac Theatre, as well as the provincial stage. Athenaeus, in the eighth book of his Detpnosophists, 346, cites a fragment of the Geryoneus of the comic poet Ephippus, where the monster is a gigantic king to whom the world is subject. His subjects catch a fish larger than the island of Crete, for which the monarch possesses a kettle of corresponding size, the rim of which is inhabited by Sindians, Lycians, Macedonians, and Paphians, whose duty it is to provide wood, as well as water and salt, for the cooking of their catch. Five boats ply around the edge of the kettle, and exhort the labourers, ‘ Why don't you kindle the fire? It is cold’; and to the Celts and Macedonians in their turn, ‘ Stop blowing your bellows, Macedonian. Put out the fire, you Celt, lest you burn the fish.’ The awkward figure of the monster no doubt made him a favourite character upon the comic stage. In the fifth century of our era the Homeric imitator, Quintus Smyrnaeus, does not forget Geryon. In emulation of Homer’s description of the shield of Achilles, in Jliad 18, Smyrnaeus (6, 196 b) depicts the shield of Eurypylus as ornamented with representations of the labours of Heracles, beginning with his throttling the two serpents sent by Hera to destroy the hero in his infancy, with traces of the scene as described by Pindar in N. 1. Then follows the Nemean lion, the Lernean hydra, the Erymanthian boar, the Stymphalian birds, the Augean stables, and the cattle of Geryon, following in general the order of Apollodorus. Quintus, imitating the Homeric style, tells us that thereon (on the shield of Eurypylus) was the form of Geryon, dead among his cattle, his bloody head subdued by the mighty club. In front of Geryon was slain his dog, Orthros, resembling his brother Cerberus, as well as the body of Geryon’s herdsman, Eurytion. I have dwelt a little longer upon the history of Geryon because he is a little-known mythological figure, a representative, apparently, of the rude forces of nature, which it seems to have been the special mission of the hero Heracles to subdue, and thus to tame the earth and render it habitable by man —a mission which is expressed by Pindar, N. 1, 62 sqq., where the seer Teiresias prophesies to the hero’s father and mother, after their infant son had slain the two serpents, ‘how many lawless beasts he should slay on land, and how many in the deep. And many a man, too, who walked with crooked insolence, he should give over to hateful death.” The encounter with Geryon is a typical case of this activity of Heracles, behind which the special features ee ee enlarge sink in F? épebito entirely the dol the fan strictly vTroKpt’ beginn: épebilo, yap 2 govern of the ; we hav hymn is In aj N.1, 4 somew ‘ Than! apposit can st govern perhap mB. 8%, €TLET SO will li ETLETO the mz widely UTOKpL There tives 11 ae Bow: eTEX, means some ] sition | tion in Unr And cattle, les put nich he xht the C poets clever ed the igment igantic ian the onding is, and for the le, and to the lonian. of the mnaeus, uield of ylus as g with ifancy, ws the yhalian ral the s that yng his on was ody of le is a > rude e hero yy man e seer on had d, and rooked reryon atures (| ON CERTAIN FRAGMENTS OF PINDAR 229 enlarged upon by Callicles in the Gorgias of the violence done to Geryon sink into the background. Fr. 235 contains a usage that may be unfamiliar to some. épeBilouar adiouv Serdivos vmoxpiowv followed by a relative, leaving iméxproew entirely without an apparent connection. The rest of the Fragment states how the dolphin of the sea is excited by the sound of the flute, in accordance with the familiar conception of the dolphin as a musical animal. wd7ro«piow means strictly an answer, as we remember that the actor in tragedy was called uTroxpitns, ‘the man who answers’ the leader of the chorus, and so made the beginning of dialogue. When we endeavour to ascertain the meaning of épeOifopar SeXdivos broxpictv, we are reminded of the familiar construction of api and certain other accusatives in an adverbial or prepositional sense and This construction probably developed from a free use of the accusative as inner object, in apposition with a clause. Thus, in J. 3,7, we have evxrAéwv 8 Epywy arowa ypn péev buvjcar Tov éodov. ‘We ought to hymn the brave man in recompense for his noble deeds.’ Here épywv drova is in apposition with the clause dyvjcar tov éorov. Sowith yapey in P. 11, 12; N.1, 4sqq. InP. 3, 95 Acos d€ yap, the transition to an adverb in ydpu is somewhat further advanced, and we are inclined to translate with Gildersleeve, ‘Thanks to Zeus.’ But the nucleus of the clause with which yapuyr is in apposition is still an accusative xapdiav, and therefore the appositional force Yet ydpw has practically become a preposition This usage is It begins governing the genitive. can still be vaguely felt. governing the genitive, Avos d€ xdpuv, ‘Thanks be to Zeus.’ perhaps more frequent with ydpw than with any other accusative, but in N. 11, 16 we have teAevrdy used in the same way, TeAevTay amdyTwy yap ‘As the end of all things he will be clothed in earth ’—i.e. he Here teXevray is in apposition with the clause yay So in P. 2, 84, Aveo Sixay, ‘ after ET LET OOMEVOS. will lie in the grave. émuecaouevos, but takes its case from yap. the manner of a wolf.’ The prepositional force of ydpu and dieny is found widely in Greek literature, nor is it confined to poetry. But the case of uToxptow in the present Fragment is, so far as I know, unparalleled in Greek. There is a like example of this tendency to extend the adverbialization of accusa- tives in the Wasps of Aristophanes, 338, rod 8 ébeEw @ pata, tadta dpay ae BovXetat; where Graves says from the scholiast ‘with what pretext, from éméxyw, “to put forward,” as in émicyeoiny in Od. 21, 71, where the word means pretext.’ This use of an accusative in apposition with the sentence or some part of it in a sense which closely approximates to an adverb or prepo- sition is probably not sufficiently frequent to call for more conspicuous men- tion in our grammars, but it is at least a noticeable syntactical freedom. E. B. CLaApp. UNIVERSITY OF CALIFORNIA. VINDICIAE PLATONICAE I. As the plan of the Oxford Bibliotheca does not allow an editor to justify his treatment of a text in detail, it is inevitable that misunderstandings should arise, and I see that even a competent Platonic scholar like C. Ritter has missed the point more than once in his recent Bericht (Jahresber. fiir Altertumsw., Vol. CLVII., 1912, 1). I make no complaint of that—it was inevitable—but I feel bound to put on record my views on at least the more important pro- blems suggested by the text. First, however, I must say something about the MSS. and the use which I have made of them. I suppose we have all by this time got over the superstition of ‘the best MS.,’ which played such havoc with K. F. Hermann’s text of Plato by leading him to prefer the most arbitrary conjectures based on the blunders and omis- sions of B (the Clarke MS.) to perfectly sound readings found in most other MSS. Even Schanz’s original text of the first tetralogy is seriously marred by the unique position he at one time assigned to this MS., but it must not be forgotten that he changed his opinion at a later date, and that he has since then re-edited the Euthyphro, Apology, and Crito, on a much sounder basis. Even if there were nothing else, the lacuna at Theaet. 208d, 4 Ildvu pév oty ... 209a, 2 Aoyov would prove the point. There nineteen lines of printed text are omitted by B, and have to be supplied from other sources, and it is obvious that what has happened here on a large scale may have happened elsewhere on a smaller one. It is now admitted, accordingly, by everyone that the archetype of the second family, the oldest of the Plato MSS. at Venice (T), must be regarded as co-ordinate with B. Following a hint by Jordan (De cod. plat. auctoritate, p. 629), M. Henri Alline has recently insisted! on the fact that T really represents a form of the text older than B, and I think I can add something to the proofs he has given. No one would ever guess from the printed editions that there are two quite distinct sets of scholia to Plato, but an inspection of the Leyden photographic reproduction of B, and a comparison of its margin with Schanz’s collation of the scholia contained in T (Platocodex, pp. 6 sqq.) will enable anyone to control my statement that the original scholia of B are quite different from those of T,? and a study of their contents t L’histoire et la critique du texte platonicien (Rev. the scholia which he found in B as O, to distin- de philologie, 1910, pp. 251 sqq.). I can recom- guish them from those he found in Ruhnken mend this article as the clearest statement of the (R). The fact is, however, that the scholia problem to be found anywhere. common to B and other MSS. are regularly 2 In his edition, K. F. Hermann designates all added by a later hand. © Allie TT TERR mn rm will sho that th logical mentat of Plat: scholia, and inc were 1n from tl the ren not so. (Rep. 1. para o T is no T and. reasons for Phe If Calligr: recensi writers and T . mistak out reg for the respect the ord letters. self. I follows dt0p0w« held to his ins’ things, scholia correct with tl MS. as an eve 1 This Introduct duction. however author o justify should r has uNSW., — but t pro- ut the e best ading omis- other ed by 10t be since basis. Wits kt are vious vhere t the >» (T), e cod. that add 1 the , but rison odex, yinal tents listin- hnken cholia ularly -~_ ee iain — VINDICIAE PLATONICAE I. 231 will show further that they are of quite another character. I have no doubt that they are the work of Arethas, who wished to replace the mainly lexico- logical scholia of the archetype by biographical notes and extracts from com- mentators like Proclus and Hermeias. Now the chief MS. of the second part of Plato (Par. A.) contains the old scholia and not what I may call Arethas- scholia, from which it follows that TA give us a text quite independent of B, and incidentally that Cobet and his followers, who rejected all MSS. but BA, were in fact basing the text of the first six tetralogies on a different foundation from that of the eighth and ninth It has been said, and M. Alline repeats the remark (p. 272), that T is a copy of the lost first volume of A, but that is not so. It is true that in the portion of Plato which is common to A and T (Rep. i.-iii., 389d) we have a practically identical text, but the long scholium on para capdavioy (337a) is much more fully given in T than in A. Accordingly, T is not derived from A, but is co-ordinate with it (though later in date), and T and A together give us a recension which is older than the text of B. For reasons which we shall come to later, they probably represent a dvop@wo1s made for Photios. If we confine ourselves to the readings of the first hand, that of John the Calligrapher, I cannot see any ground for holding that B represents a further recension, and here I feel bound to differ from M. Alline and other recent writers. It appears to me that everything can be explained if we regard B and T as two copies of the same text, the writers of which are apt to make mistakes of a different kind. B, as a calligrapher, copies letter by letter with- out regard to the sense. That accounts for his frequent omissions, and also for the fact that he often writes nonsense. His superior accuracy in certain respects is due to the same cause. T writes more hastily, and he often alters the order of the words because he is thinking of the sense rather than of the letters. We know that, because he often adds the signs of transposition him- self. I hold, therefore, that B is not a recensed text except in the sense that it follows the same recension that T does. There is, however, a recension or diopOwars of the text in the MS. itself by another hand (B?), which is generally held to be that of Arethas,' and is at any rate that of someone acting under his instructions. Of course, Arethas must have known all about those puzzling things, the duépA@wors of Leon and ‘the Patriarch’s book,’ mentioned in certain scholia to the Laws,? and he must have had good reasons for taking his corrections from the source he did. Now, as a rule, these corrections agree with the text of W, so that Arethas must have regarded the original of that MS. as in some sense the standard text, and in that case W represents for us an even older tradition than BT. I cannot agree, however, with O. Immisch, the archetype. In that case, he may well have written them in the margin himself. 2 The importance of these references has been justly emphasized by O. Immisch, Philol, Stud., I. p. 49 sq. 1 This is doubted by Mr. T. W. Allen in his Introduction to the Leyden photographic repro- duction. One of his points falls to the ground, however, if we regard Arethas as himself the author of the scholia, which are not derived from 232 JOHN BURNET from whom we have all learnt so much about these matters, in identifying this tradition with some branch of the memoria non recensita, and in classing W along with Vind. F, which does represent something of the sort. On the contrary, it seems to me that the original of W deserves the name of a recen- sion far more than that of Band T. It certainly had a far larger number of marginal variants and it had the old scholia. I should not be surprised, indeed, if W should prove to be a direct copy of ‘the Patriarch’s book,’ nor even if that should prove to have come from the Academy. When Justinian closed the school in 529 A.D., it would be natural that some of its MSS. should find their way to the Patriarchate. I shall return to this point later; but I may suggest now that this assumption would at once explain the fact that the recently discovered Commentary on the Theaetetus (second century A.D.) has so many points of agreement with W. A full discussion of this question must be postponed, however, till we have dealt with the details which have a bearing upon it. I propose to take the dialogues in order, and to deal chiefly with those passages which have been unnecessarily tampered with by editors. Euthyphro. 2a, 5. w Evdvdpor. It is clear that this form stood in the archetype of BT, though B? (Arethas?) has corrected it to ® Ev@v¢pov in accordance with grammatical precept, and W fluctuates. Just so, in Soph. Ant. 572 the Laurentian MS. has @ diAtaé’ Aiwwv, and in Ai. 903 @ Ttadraidpwv yuva. In such cases, an editor's duty is to reproduce the best tradition, not to follow what may be only an arbitrary canon. 3a, I. Tovs Tav véwy Tas Braotas S:adGeipovtas. Schanz follows Gomperz in deleting tov véwy, but Adam rightly declines to follow him. I do not think, however, that we should understand the words to mean ‘ the development or “sprouting ” of the young.’ The young are the crop, and their BAdora: are the ‘sprouts.’ The phrase will be seen to be correct if we remember that we could say, e.g., rovs Trav tup@v tas BrAdoTas diag Geipovtas. 3e, 2. et d€ ormovddcorvTa. .. . The orovéafovras of B is a characteristic blunder, which B? has corrected. Its recurrence in CD only proves the dependence of those MSS. on B. There can be no doubt that TW give the genuine tradition. 4a, 12. ov yap olpai ye Tov émituyovtos [ opOas| avTo mpaéat. My deletion of op@@s> (which I assume to have been erroneously repeated from 67n toté opOas éyes above) occurred a little later to Gomperz (cf. Hellentka, vol. i. p. 348), apparently before he had seen my text. I understand the passage to mean ‘ Good gracious! People must be mostly in the dark as to what you sp Madvig unnece: here, a: infiniti can als we can must n have n eMLTUXC omissio 4b, Io. Ad homick If you « to you therefo Tetr. I. KaATENO: O“oT par for atro 4e, 2. Sc: note To we rem below ¢ 5b, 4. Th text are Konalor Of cou thing is 5c, 6. Oo In see yol formida eyes to most ed scious ° 1 B? tical MS. » an only ines ords the » be rT AS ted. 1ere ted (cf. and VINDICIAE PLATONICAE I. 233 to what the right thing is. It isn’t everyone, I should say, who would do what you speak of (avro), but only a man of advanced wisdom like yourself.’ Madvig and Schanz assume a lacuna after dan moré opOdas éyet, but this is quite unnecessary. The phrase dpOds éye, ‘it is the right thing’ is impersonal here, as in opO0ds exer émeEvévar (ga, 7), and it does not necessarily require an infinitive to define its meaning, though it may have one. Of course the phrase can also be used personally as in dp0ds éxyeww tavtny Thv mpaéw (gb, 2). Here we can ‘ understand’ mpdéau, if we like to put the thing in that way, but we must not say that mpd&ac is subject to éyex. Impersonal verbs and phrases have no subject. I think now that I was wrong in omitting eiva: after émetuxovTos with Schanz. It is in W as well as T, and B is prone to sins of omission. The meaning is clearer with eZvaz, but not different. 4b, 10. éme&sévas (sc. Seiv) edvrrep 0 KTeivas auvéortios cor Kal opotpatrefos 7. Adam gives the right interpretation here. The point is that, if the homicide shares your hearth and your board, he infects you with the piacpa. If you did not come in contact with him in your daily life, it would not matter to you at all. Schanz’s édvrep yo xreivas ..., ‘even if the slayer... therefore destroys the point of the sentence. He himself quotes Antiphon Tetr. I. i. 1 which settles the matter, and Adam appositely adds Laws ix. 868e katedOav S& 6 te ToLwodTov Spdcas Tois avTodD TraLciv lepav wy KoweveiTw pwNde opotpatretos yuyvécOw mroré. It may be well to add that Plato only uses «reivw for avoxteivw in referring to bloodguiltiness. It is the legal term. 4e, 2. TO Oeiov ws exer. Schanz suspects these words, and suggests that they come from a marginal note 70 dovov ws yet. They will be seen to be correct and even necessary, if we remember that Euthyphron is appealing from human to divine law. See below on 5e, 2. vouGetovvTe . . . KoAaCovTt. 5b, 4. Suddonovtt... This instance shows how old the comparatively few corruptions in Plato’s text are; for BTW all have the impossible dsddcxovta . . . vovOerodvTa .. . xoddfovtra. Schanz quotes Vat. 225 (fifteenth century) for the correct reading. Of course that does not prove it to have an independent tradition. The strange thing is rather that the mistake was not corrected earlier. 5c, 6. oé pév ovd€ Soxet opav, eve 5é ows oFéws [atexvas] Kal padiws xaTeider wore ageBeias éypawaro. In the first place, cé pév ovd€ Soxe? opav means ‘he pretends not even to see you’ (cf. Eur. Hipp. 119 pn Soxee tovrov xdAvew). Euthyphron is too formidable a person to attack, as he himself has just said, so Meletos shuts his eyes to his proceedings. It is important to observe that Euthyphron is not, as most editors say, a representative of traditional orthodoxy. He is quite con- scious that his views are not those of the Athenian people, and he sym- 234 JOHN BURNET pathizes strongly with Sokrates as a fellow-heretic. That is very important historically. As to atexvas, it is hard to believe that it is a mere interpolation in T, and Schanz’s view that it is intended to explain the xat- in cateiéev is most improbable. So is the view I once held that it represents dtevas (it was suppressed in the reimpression of my text). The suggestion of Mr. Richards (Platonica p. 1) that it should be transposed and that we should read ateyvas aé pev ovde Soxet opav (cf. Ar. Clouds 425 ov8’ dv diarexOeinv atexvas) is most attractive, though I should prefer cé pév ateyvas ovdé Sone? opav. It would mean that dteyy@s was written in the margin of the archetype and that B left it out, while T inserted it in the wrong place. In both cases the pro- cedure would be characteristic. It is odd, however, that we find no trace of atexvas in W. 5d, 4. Kata Thy avoo.wrnta. I cannot believe that cata tH oovwTyta is right, though it is doubtless an ancient variant. I wish to point out, at any rate, that Adam’s argument from the scholium dyti rod ouoiws, raparAnciws TH opowornte has no weight. It is not an old scholium but an Arethas-scholium, and Arethas is explaining the text before him as well as he can. He could not correct it; for W too has oo.oTntra, and we have seen that his corrections come from the original of W. The correction avoo.rnta is apparently by a later hand in that MS. (It is a minor matter, but it is as well to note that there is no lemma «ata tnv oooTnTa in the scholium, as Adam says there is. The lemmata are due to the editors.) A la) / 54 5e, 2. Oéacas ws péya cou €p@ TEeKuNpLov TOD vopov OTL OVTWS EXEL. Schanz rightly objects to the interpretation of tod voyov as ‘ public opinion.’ Public opinion was against Euthyphron in this matter, as he says himself, and such a use of vdsuos would be hard to parallel. But there is no necessity for reading tod voyivov with Baumann as he does. The law on which Euthyphron relies is one of the aypadot voor, and therefore he cannot cite any precise text in his justification. He can only appeal to the actions of the gods as a texunptov of what the divine law really is. As a matter of fact, there is reason to believe that the merely human law of Athens would not have allowed Euthyphron to bring his indictment (see Adam’s note), and we are probably to imagine him as appealing to a higher law (like Antigone) before an unsympathetic Baowrevs. If that is right, there is no need to suspect the words 671 tavta dpOws av ein oTw yuyvopueva, since 6rs depends on Texpnpiov represented by the relative. After the parenthesis, the words pa émitpére “.T.r. give the purport of the alleged unwritten law and explain obtws éxev. 7e,2. dAdo addra dixata nryovvTat K.T.X. C. Ritter (p. 5) thinks I should have mentioned that Hirschig and Schanz insert cal aduxa after dixava. I have, if anything, mentioned such things too —— ~—— often. special aoika a silence. gd, 2. Th of 6 in’ eTravope ap’ out thread « él KAN@ IIb, 7. Mr seems t says SO (Bekke1 howeve which \ synony! therefo1 the ordi Ile, 3. Ob therefo. saw tha 12a, g. It | and ove spite o! The péé No words « difficult ordinati verb in ELKELVY O VELKEELD) the firs 1 Schar yp. orépéa anz too — ~— VINDICIAE PLATONICAE I. 235 often. Editors insist upon forcing a formal symmetry on Plato which he took special pains to avoid. In the present case some interpolated MSS. have xa! adica after nyotvvrac or after Aoyov, but that too is better passed over in silence. gd, 2. Tovto 6 viv érravoplovpeba ev TH AOYO. . . . The evidence for -ovjpe@a is overwhelming (-wue0a ZT), and the omission of 6 in T seems to be accidental. Sokrates was about to say dpa tovdto 6 viv éravop0ovpe0a €v TO AOYH KAaABS A€yerac; but he interrupts himself at d, 4 dp’ ovtw «.7.r. to make sure that Euthyphron accepts the éravop@wors. The thread of the sentence is picked up at e, 4 Ovxotv émicxoTapev av ToUTO.. . el KANWS A€YETAL. I1b, 7. 06 dv mpodapeba. Mr. Richards desiderates ‘the old trofmpeOa,’ though he adds that ‘it seems to have no MS. authority.’ That is not quite the case, though Fischer says so in order to discredit the veracity of Stephanus. It is found in Vat. 225 (Bekker’s A), for instance, and in the margin of some other MSS. _ It seems, however, to be nothing more than a correction of the blunder wpoduvpwpela which we find in B. The point really is that mrpoti@ec@ar and t7rotiPecGau are synonyms. The former is the Attic word, while the latter is the Ionic (and therefore the scientific) term. It is only natural that Euthyphron should use the ordinary word in the first instance. Ile, 3. avtos cor cuprrpoOvupnoopar | detEar| drws av pe dudaEns. Observe that the interpolation de7€a (BT) is absent from W. It appeared, therefore, for the first time in the archetype of BT. K. F. Hermann already saw that it must go, though he knew nothing of W. 12a, g. ZLiva 5é tov épEavta Kai Os Tabe TavT’ epuTEevoeV ovn eOérer vetxeiv* iva yap déos évOa Kai aidas. It is clear that the archetype had the alternative readings 0’ ép£avta (BW) and orépEavra (T and yp. BW).' The reading & ép£avta is naught; for, in spite of Herodian and others, ép£avta can never have had a rough breathing. The pé€avra of Stobaeus, etc., points plainly to ép£avra. Now Tov ép£avta nal os . . . épvtevoev would be very odd Greek if the words are taken together, though Adam is the only editor who has felt the difficulty (see his note im loc.). Placed as it is, mai Os cannot well be co- ordinate with the participle, but must mean ‘even he who.’ That implies a verb in the third person, and we get é€Oédev vecxety at once from the é0érew elxecv Of the scholium in Cramer’s Anecdota and the veixeowr (i.e., I believe, veixéecv) in the margin of T. In that case, the €0éXeu evzretvy of W represents the first stage in the corruption. If we remember that ¢utevew is used in 1 Schanz says ‘oréptayra in marg, yp. b,’ but Calligrapher and not that of Arethas or any later yp. orépéavra is certainly in the hand of John the corrector, so ‘b’ should be B. 236 VINDICIAE PLATONICAE I. Homer of the author of troubles, we shall render ‘Even he who was the author of all these things will not revile Zeus.’ So much appears to me certain, but I cannot refrain from two rather hazardous speculations. In the first place, we know from the Venice scholium on I/tad A, 5 that, in the Kypria, Zeus brought about the Trojan War on the advice of Momos, who may therefore be fitly called the author of all these troubles. It is his special function to find fault with the works of the gods, and later writers tell us that he burst from chagrin because he could find nothing to censure in Aphrodite. Aphrodite, of course, played a great part in the Kypria. In the second place, is it conceivable that Plato read the ‘digamma’ in his text of Stasinos and reproduced it? We must remember that this letter was quite familiar to educated and travelled men in his day, and it is written once in the papyrus fragment of Alkman. (=6). pletely ; for the numeral 6 and the ligature of or were quite indistinguishable at a later date. Is there a reference here to that or a similar story? It was also in use as a numeral If Plato wrote it, that would explain the old reading orép£avta com- I4C, 3. So B. Though TW have épwravtra, T has épwpévm and W épopéevo. It seems clear, then, that ép@vra is the original reading, though Schanz reads Apart from the MS. evidence, the chief objection oo 2 / 4 \ > a 3 / > a vov O€ avayKn yap TOV épavTAa TO Epwpév@ axonroveiv. €pWTO@VTA . . . EPWTWLEVY. to this is that, according to the rules of dialectic, it is the €pwtwpevos who must axorovbciv tH Epwravte and not vice versa. This is a case where B displays his characteristic excellence. 16a, 3. e ¥ ; [OTe] apecvov Biwcoipnv. I now regret having followed Schanz in bracketing étz. I have often repented of leaving a bracket, very seldom of removing one. JOHN BURNET. St. ANDREWS. ~- TTT ON TI THI in the ye discusse: tradition are inde 1906-19¢ We selves 0} and, if < that he follows | tions wi Stromate The injury t thrown in any © that we which t homoeog the scril lessness Me should | that (as with wh Th in the . I. 67. 2 1 That | mony of ] 2 Encycel Criticism,’ 16 (1902), 3 E.g, | ten ON THE TEXT OF THE STROMATETS OF CLEMENT OF ALEXANDRIA. THE germ of the following paper is as old as the previous century, when in the year 1898 my attention was accidentally drawn to one of the passages discussed below. But progress was impossible until the facts of the MS. tradition of the Stromateis' were properly presented—a service for which we are indebted to the excellent edition of Clement by Dr. O. Stahlin (Leipzig 1906-1909). We need do no more than turn the pages of this book to convince our- selves of the large part which omissions have played in deforming the text, and, if a textual critic is to do anything to repair the mischief, it is obvious that he must first inquire into their character. The object then of what follows is not to maintain a thesis but to submit certain facts and considera- tions which may be of use to systematic students of the traditional text of the Stromateis as presented in the above-named edition of Stahlin. The causes of omission in MSS are three. For the first of these—external injury to the exemplar—the copyist is in no way responsible, nor can light be thrown upon such omissions by any study of a scribe’s personality as disclosed in any other part of his work. This is not the case with the other two causes that we may distinguish—oversight for which there is excuse and oversight for which there is none. The first includes all the errors covered by the term homoeographon.* The second those in respect to which there is no evidence that the scribe’s eye has been ensnared by similarities of words or letters, and care- lessness is therefore alone to blame. Method in textual criticism requires that the two last kinds of omission should be discriminated, as the effects they produce are often dissimilar, and that (as I have said elsewhere*) they should both be established for the text with which we are concerned. That homoeographon has produced omissions of letters, syllables and words in the Stromateis the following selection of examples may suffice to show: 1.67.2 cal C&XAOLD> AAAOGD:L TOrAAAYOD, 79. 3 0 Kupaios mp@tros tod KptTLKOD EionynoaTo TovVoNa, 140. I Tovds Bapovus. IV. 125. 3 evtTuyetv te xpy. Vz. 92. 6 wuyny evotxotvaav , tb. piav } tTrEious; . VI. 24. 1 tédos émibcivat tO tmpdahopov <"Edopor>, 1b. 149. 5 ov yap wa Sofwperv civar X prot ov ric Tt evoper, 1b. 162. 5 e& Pirtocogdnréov . VII. 4. 2 arorenré- ouata TO yryvwokey, Cf. ib. 76. 5 Tas ders avTOD mpds TA dpat a xetpaywyav, 29. 3 deb ac> dar pévor, ib. 46. 2 TOL YYwWoTLKaL éxacta arodidotat. That omission occurred without the lure of homoeographon may be shown by the following: I. 150. 5 fin. 67c padsota . II. 106. 1 évredOev ofwar nai Tas TereTAds, ib. I15. 2 IV. 30. 3 ws To Mavons. Grr’ ovK ex hvcews THY owrTnpiar. TETUPWLEVOY Kal TO TEPWTLOMEVOY <_TOU Trupos Kal Tod PwTds> Has odv, ib. g2. 4 Xalpwv Kai cvyxaipwv pos ods 6 AOxyos THL aTrogTOAw@. VII. 45. 1 Ta wept Tod Oeod Siecrndpas >Tpos avTis Tis anrneias. The causes which have produced the two kinds of losses that we have been considering are not to have their sphere artificially restricted to letters, syllables or words: they must be allowed to affect larger blocks of writing and to cause the omission of whole lines, single or several. Nor shall we be surprised to find such omissions in the text of the Stromateis. Here, however, it is necessary to discriminate; for the two causes do not produce quite similar effects. The omissions of negligence in such cases are in units, we may say; that is, they are (approximately) losses of whole lines or of multiples of these. But omissions through the straying of the eyes in homoeographa from one line to another are not necessarily to the same extent in units. We have always to allow for the possibility of some lateral deflexion. In amount therefore an omission of this kind may from the outset be somewhat more or less than a single line or a multiple of lines. This must be borne in mind when we come to calculations. In the present connexion Loss cannot be separated from Repetition and Transposition. Meaningless (that is, mechanical) repetition of a unit is obviously correlated to its mechanical omission; and, as I have said elsewhere,’ transposi- tion of lines and passages is really ‘ arrested loss.’ There is one element of uncertainty which an inquiry into the constitu- tion of the unit-line, if I may call it so, of the exemplar of the Stromatets cannot hope to eliminate, and which therefore it is obliged to neglect. This is the question to what extent abbreviations occurred in that exemplar. That numerals, ordinal as well as cardinal, were denoted frequently, if not invariably, by letters is not only probable in itself, but supported by the also is sometimes due to ‘arrested loss,’ VI. 65.6 yew perper kal yewpyet (L, corr. St.) is an in- teresting case. 1 Encyclopaedia Britannica |.c., p. 712, col. 2. The matter is dealt with at length in the Classical Review, 16(1902!, p 308. Transposition of words indicati mevte (Vv abbreviz ourselve as one. The perhaps line cont Il. | Tepipane from Ez » KOTtOoOd oTeias > letters, ; Eusebius IV. Sorta’ KUpLOS > ' restored IV. apvouvTa $4. 69) V. 4 Kat cat they are caused tl VI. ha.vopéeve TOV TEVT inserted appear af VII. THY AKO euppocvvr one of the certain. ~~ ON THE TEXT OF THE STROMATEIS 239 indications of our MS. as recorded in Stahlin’s apparatus; thus é represents mévte (vol. 1., p. 139. 26) and wéumros (ib. p. 474. 12). But how far other abbreviations were employed we can only conjecture. We must then content ourselves with simply counting the letters, the « adscript of course reckoning as one. The evidence which I would submit to readers of the Classical Quarterly as perhaps adequate to support a definite conclusion, viz., that the average unit- line contained slightly more than 30 letters, consists of the following passages : OMISSIONS DUE TO ‘ HOMOEOGRAPHA.’ II. 135. 1 (I. p. 187, St.) tov aptovy adtovd mewavtT,. Seoet Kal yupvor TeptBarel TO apyvplov avTov émi toxwt SWaeu>. Loss of 46 letters, supplied from Ezekiel. I. 102. 4 (I. p. 66, St.) qgavyceras Tat Sevtépw. Kal TpiaxooTat 1) Mawvoews x.7.€. Loss of 80 or 52 letters, according as words or numeral letters were used, supplied from Eusebius. IV. 33. 1 (I. p. 262, St.) 82a rovto adara yevnOntw TavTa Ta Yeti AH TA SOXLAa Kal yAoooar xk.T.€. The omitted words, restored from ‘Clem. Rom. Syr.,’ total 71 letters, 68 if «vpsos was abbreviated. MECHANICAL REPETITIONS. I. OF BLocKs oF WorbDs. IV. 72. 4 (I. p. 281, St.) dsdvep apyncacbar avtov ovdérore Svvavrat ° apvoovrat S€ avTov oi uy ovTEs Ev AUT@L, an exact repetition of the last words of $ 4. 69 letters. V. 4.2 (I. p. 328, St.) «al mpo tis Tov Kupiov Tapovaias éxXoyAs Kai $17 Kai cailopév ns, 52 letters, a duplication of the same words in § 1, where they are preceded by etipicoxopévns. The homoeographon seems to have caused their repetition in the wrong place. VI. 35. 4 (I. p. 449, St.) av TO pév éorte Trepi Tov dvaKocpov TaY aTAaVaV paivopévwv dotpwv TO 5€ TrEpl T@V GUVddwv. The omitted words were inserted by Gruppe from 36. 1, where with the exception of ro 6€ mepi they appear after yewypadias. They contain 63 letters, 59 if wévre was written é. II. OF SINGLE WORDS OVER A CONSIDERABLE SPACE. VII. 13. (II. p. 10, St.) évapyt dé as Ge padtota Kai axpiBa@s eidixpiv7) THY GKOpEaTOYV UTEeppuas ayaTwcas Yuxais eatiwpévas Béav, ardiws aidiov eUppoctyny GKOPETTOV KapTroupévas Eis TOS aTENEUVTHTOUS al@vas. That one of the dxdpeorov’s (St. brackets the second) has come from the other seems certain. There are 59 letters between them. 1 The notes on pp. 446, 447, 449, 489 afford other examples, 240 J. P. POSTGATE Similarly in 15. I (II. p. 11, St.) to Oeiov is repeated at a distance of 100 letters from its proper place. Again in VIII. 14. 3 (II. p. 88, St.) awdvrwy is repeated 65 letters, or 61 if mpwtn was abbreviated, away from its proper place. Similarly in V. 94. 4 (I. p. 388, St.) éretaosdov has produced eicodoy in the following line 26 letters off. In VIII. 30. 2 (II. p. gg, St.) after aitiot rod the following words, TpoKoTTE KaTnyopnmaTos NéyeTat SE GAANXOLS aiTLA TrOTE PeV TOV AUTOV @S O éumopos Kal 0 Kadmnros aAAHAOLS Eloly aiTLoL TOV, 102 letters, were repeated by the first hand in L. PASSAGES TRANSPOSED. An interesting example of homoeographon appears in a verse quotation in V. 109. 3 (I. p. 400, St.) : ard’ el Tor Yeipas xy’ elyov Boes Hé AéovTeES ws yparrat xeipecor Kai Epya Tereiv Atrep avdpes imroe pév 0 immoror Boes 5é Te Bovaly opoias kai Ke Oeady id€as &ypadov Kal cwpat’ étroiovy rotavd’ olov rep Kal avtol déuas eyov opoiov. So the passage is given in Stahlin. But his order for the last three lines and the ouoias which he takes from the quotation in Theodoret are obviously wrong. The original form of the verses was cai Ke Oedy idéas Eypahov Kai cwpat’ éetrotovy to.av@ olov wep xavTol déuas eiyov Ex aatot (rmrot wev 0 immorot Boes Sé Te Bovaoiv opota, with the écacrou of Hiller and Diels and the transposition of Karsten. The first corruption was the miscopying of opuota for éeaoro from the following line. After that, homoeographon at the end of the two verses caused the omission of the latter which was subsequently reinserted but copied into a wrong place. Theodoret’s oyoias is worthless in itself, but interesting as pointing to ouota which in Clement’s text, as offered by L, and in Eusebius has been changed to ouoiov. The lines may have been written continuously, and the distance from the beginning of one ooita to that of the other would be 35 letters. 3 I deviate for a moment to consider a textual question arising on the first line of the fragment. Stiahlin’s critical note is as follows: éyov Eus.... aA’ ef xelpas exov Boes é déovTes . . . HE réov tes | 4 EX€havres Theod. MSCV 7é Kérntes Schultess 7 xedéovres Diels Berl. Sitz. B. 1891 S. 578. The motive underlying these changes is that Aéovres and immo are ‘incompatible’ (Diels l.c.). If real ‘lions’ were meant, this would be true; but these animals are only illustrative. For the writer’s purpose ‘lions’ are as good as ‘horses,’ and the substitution of the one for the other is merely legitir Persit docult artife: poetri talkin; This s Hor. . calce | both ° diplon has sil might (Crusi I. KWAUVO} auTiKa HV, TO €TLTEN aAUTOUS KATA ° Kabat KTilect evepryet in the inevita In amo Ta ) oppen Tpos Td Kal 7) wv TaQeav ov Kpiv casual « QUuTiKa In the Chi reed of NEyovTE These y €E ay He 1 It is vy MSS. of designed ice of - 61 if ‘cOOoV vords, 1 @S O ed by on in » lines iously The owing d the into a ng as us has y, and ild be e first 9\ Diels o are true ; s’ are nerely ON THE TEXT OF THE STROMATEIS 241 legitimate poetic variation. One might as reasonably attack the text of Persius Choliamb. 8 sqq. ‘ quis expediuit pstttaco suum ‘“chaere” | picasque docuit uerba nostra conari? magister artis ingenique largitor | uenter, negatas artifex sequi uoces. | quod si dolosi spes refulserit nummi, | coruwos poetas et poetridas picas | cantare credas Pegaseium nectar,’ where psittacus means ‘a talking bird such as a parrot,’ and corwus ‘a talking bird such as a raven.’? This superficial inconsistency in the details of an illustration occurs elsewhere, Hor. S. II. 1. 52 sqq. ‘dente lupus, corvnu taurus petit — mirum | ut neque calce lupus quemquam neque dente petit bos.’ The ox, or bull, is formidable both with hoofs and horns: the reader may take his choice. There is no diplomatic reason for doubting our text; % €AXédavrTes (in itself quite possible) has simply come from a wrong division of words 7 éAéovtes. And that lions might be imagined in the réle of artists we can see from Babrius Mythiamb. 194 (Crusius) ef NéovTtes Hideccav yrAvHev ToAXOS av avdpas cides UTTOKaTW A€oVTOS. I. 82 (I. p. 53, St.) discusses the question whether non-interference (To pu xwAdov) is the same as responsibility (To aitiov). §§ 2-5 read as follows autixa KoXdlovtay pos Tov voyxov of TOUTwY aiTLoL* w@L yap KwrADaaL Sivas hv, ToUTw@L Kal 1) altTla Tov aoupBaivovtos tpocdmTeTat. hayev 57 TpOS avTovs TO aitioy év TaL ToLety Kal évepyeiy Kai Spav voeicOar, TO S€ wn KwWAVOY KaTa ye TOUTO avevépynTtoy elvar* ETL TO peV aiTLOY TpOS THe evepyeiat eal, Kabatrep 0 pev vauTrnyos TPs TAL yiyverBat TO oKdgos oO Sé oiKoddpmos Tpos TAL Ti yap évepye’ 0 un KwAVwYV ; Prof. J. B. Mayor’s transposition of the words which stand in the MS. after yeyvouévov, and break and stultify the argument there, seems inevitable. They contain 64 letters. In II. 59. 6—6o. 1 (I. p. 145, St.) opyr pev odv dopa Stavoias emi te 7 , \ > , \ \ \ a , nw J xTitecOar THY oikiav: TO S€ Hn KwWAVOY KEYwpPLTTAaL TOD YyiyVOpmévou. ’ , / \ / e \ nN e , \ \ , / amo tous m7a0os b€ TAEovdlovca opun 7) vIEpTeivovca KaTa TOV Aoyov péTpa, FY t 4 ; / \ > \ , \ A 9 / 7 4 4 i) open exhepomevyn Kal amerOns ANoywr* Tapa pvow ovv Kivnots Wuyns Kata THY \ \ / > / \ 4 e » 22 / \ ’ / 949 ¢ A e/ Tpos Tov Noyor aTeiOerav Ta TAON* 4 8 atrootacts Kai aTreiOea éf’ piv, woTrEp kal 1) UTraKon éd’ Hpiv: S10 Kal Ta Exovata Kpiverac: avTixa Kal’ &v ExacTov Tov a 7 > , > / ’ / e/ bal > 4 \ An > / waQeav ei tis émeEior, adoyous opé-ets eUpor Gv avTta. (60) TO yovr axKovcrov ov Kpivetat* Surtov d€ TOUTO TO peEV YyiyVvomEVOY peET’ ayvolas TO S€ avayKne even a casual observer may see, what Prof. ]. B. Mayor has pointed out, that the words avtixa to avra at the end of 59 are out of place. They include 63 letters. In III. 52, 53 (1. pp. 220 sq. St.) Clement cites the example of pillars of the Church against the contemporaries whom he Is attacking. In 52.1 we read of 8€ cal rovrous (as generally understood, Elijah and Samuel) i7repdépery A€yovTes TroAuTetas Kal Biwe ovde auyxpltOjvar Tais éxeivwv mpakeor SuvyncovTac. These words follow on cai Yapounr dé o mpodyrns [iv] dv KatadeXoite: Kw@XAEOVY €E wv Hobe hépwv Edwxe Tar Laovr hayeiv, and they precede o un écOiwy roivur 1 It is worth noting that some of the inferior MSS. of Persius give before or after v. 9 a line designed to get rid of the ‘incompatibility,’ ‘coruos quis olim concauum (or Caesarem) salu- tare,’ which bears a strong family resemblance to Prof. Diels’s first correction here R 242 ]. P. POSTGATE Tov écbiovta px e€ov0eveitw with more on the practice and teaching of Christ and his apostles as regards eating, drinking and marriage; this ends in 54. § 4 GAXa pHV 6 AUTOS OvTOS Kéxparyey @S OUK EoTLV 4» Bactreia TOU Beod Bpwars Kal Tools Ovde pV aTroY?) Oivou Kal Kpewv ‘adda SitxaLocvrN Kai EipHYN Kal yapa év Next comes a section relating to ot paxdpior mpodjrar : Elijah, Isaiah, Jeremiah, John the Baptist are mentioned. It begins tis airav unrwrny cai Covnv Sepyativny éxwv trepiépxetar ws "Hrias; With the present order, there is nothing for this avt@y to refer to, except the New Testament worthies, which is absurd. It must obviously mean the same people as the of in 52. 1. Recognising this, Professor Mayor proposed to move the words to follow dvyncovta: above. They contain 251 letters, so that a block of 8 lines would have been displaced (8 x 31= 248). But there is another alternative. The first passage contains gI letters, i.e. about 3 lines; and so it might be moved down in front of the second one. This will do away with the break in the exposition of the right doctrine of eating and drinking which is the effect of its present place, while it will also give more pertinence to its expression. For it will now mean ‘ Our friends who actually claim to be superior to the saints of the New Testament (rovrovs) cannot stand a com- parison even with those of the Old (éxetvwy).” A small and easily cured displacement has occurred at IV. 138. 4 (I. p. 309, St.) od Set 5é apPévtas petateOjvar adrra Badifovtas adgixécbar of Sei [rovro yap éote TO EXxvaOjvat Wo TOU TaTpos| bia TWaagns THS TTEVHS SeeNOovtras 05030, TO akiov yevécOar thy Siva, THS KadpiTOs Tapa Tov @cod AaBeiv! axwrUTws avadpapetv. The words 6sa—odod, 31 letters in all, were omitted after of de@ and reinserted when the words tovto—trarpos had been written, leaving the latter in a place where they furnish neither proper sense nor construction. The correction is due to Potter. In VI. 80. 1 (I. p. 471, St.) again Professor Mayor has detected a dislocation. The passage should run: xat’ érraxodovOnpa Toivuy Kai Tots els yv@ow yupva- \ A / Covow avTov TpocavaKettat, Tap’ Exadotouv paOnpatos TO Tpdahopov THe adrnOeiat e , TVEVPAaTL wyiol.’ LapBavev, THs wéev ovV povaoltKHs THY év TOIs Hppwoopévols avadoyiav Si@xwv , \ eO.Lopevos auvexés Tt SidoTynpa voeiv> ev 6€ The apiOuntixhe tas av&joes Kai pevooets TOV aplOuav TapacnpelovpEevos Ka TAS TpOS AAANXOUS oYEéTELS Kal as TA TKEtoTA avanroyiat Tl apLOuav broTréTTMKeY, év SE THE yewpeTpLKTL OUTLAY > \ + . e ” a X > / > 4 e 4 7 rn / autny ep’ éavtns Oewpov Kai ovciay ayetadBXnTOP, érépav Tavde THY THpdTwV odcay* €K TE AUTHS AoTpovopias K.T.é., except that the second ovciay appears to be another case of the intrusion of a word from an adjacent line. The words xat to voeiv (L voet), which obviously refer to povaotxn, follow Gewpav in L and contain 34 letters as preserved in that MS. Homoeographon ypyoopévors, €0.£6 we vos may have been at work here. VII. 83. 1 (II. p. 59, St.) appears thus in the tradition (except that L has amoxexabappévous) ém’ ovdevi TolvuY ELKOTWS TapdooETaL TOV TUpBaLVOYTwY OSE A \ \ > , > \ “ / > 4 STOTTEVEL TOV KATA THV OLKOVOMLaV ETL TAL TUpPEpoVTL yuyvomévwr OVSE aicxXv- 1 xal dxwdvrws Stahlin. —e VETAL ¢ TOUS T AUT aL Wilam 53 or ( after a a sense gloss o Our pe will he after wu MEVOV, ovee aic In pirobead yvoaiwl Kkabapo words 7 But Lo 28 lette In that of | In tov IIx TEXVAS f Kai Over (p- 137 HL TOUTE KuTrralo élvat Tod Aoytos ‘J maintail proposex duplicat ‘ duplica conceive ov ovr: the text similar e though ; TEXVAS 1 1 uh ovk cepted by ¢ 2 rhéyer S 8. 4 at ot >VHS | TOU 1 all, - had roper ation. upva- Peta 4 ALT U- ON THE TEXT OF THE STROMATEIS 243 verar aTroVaveiv, evovveidntos dv tais éEovciats obOAvat, mavtas ws eros eitreiv Tous THS Wuxns atoxexaOappévos otrirous, 6 ye ev para émictdpevos apervov auTa. peta TtHv Eé€odov yevécBar (yevnoecOac Mayor, Stahlin). Professor Wilamowitz has rightly noted that 6 ye and the words following. totalling 53 or (taking yevyoecOar) 55 letters, are out of place; and he would put them after advro@aveiv. But this verb is itself an interloper. aioyvveras arroOaveir is a senseless phrase; goBetra: amrofaveiy would be intelligible. The word is a gloss on tats é£ovciais (i.e. the angelic hierarchy, a well-attested use) 6d0jva. Our perfect Gnostic will not be afraid to die, nor, as he has a good conscience, will he be ashamed to meet the angel hosts. And accordingly, inserting te after i7romrever with Heyse or, perhaps better with Mayor, ovdév after yiyvo- pévwv, We May appropriately place the words in question immediately before ovoe alayuveETat. In VII. 10g. 1 (II. pp. 76, 77, St.) omy ody tiva oriynv brodei~avtes Tots pirobedpoot THs adnOeias Ex TOU KaTa Tas Ovaias vouou Tepi Te lovdaiwy TaY yudaiwv Trepi Te TOY aipécewy pvaoTiKas Siaxpivopévav, ws dxadaptwr, amd THs [ Trepi xkabapav kal axabaptwov Carwy] Oeias éxxrAnoias KataTavowpev Tov AOyov. The words 7repi— faiwv are senseless where they stand and Stahlin brackets them. But Lowth’s remedy, to put them after voyov, is clearly better. They contain 28 letters, and homoeographon zrepi(re)— epi will have caused their omission. I now take a passage where the text of Clement has been used to emend that of an earlier author. In I. 93. 1 (I. p. 59, St.) the manuscript reads cay tat Anpodoxar et 57 tov IIkadtwvos TO ovyypaypa ‘wn S€ ayod! TobTo didrogodeiv’ réyerv? ‘mepl Tas Téyvas kuTTalovtas * SHv ovdé ToAvpad ot VT a AAA GAXO TI, ErrEl Eywye wLNV Kai dvevdos eivat.. The quotation is from the Ervastae attributed to Plato (p. 137 B), where the tradition gives dAXa pi) ov>y ovTAS, @ Hire, Exwor pnde* ne TovTO didocodpety epi Tas Téxvas éotrovdaKévat OVSE TrOAUTTpaypoV ODY Ta kutTavovta Chv ovde TroAvpabodvta GAN’ GAXO TL errel eyw wiv Kai dveLdos elvat TOUTO Kal Bavavcous Karelabat Tovs Trepl Tas Téyvas éotrovdaKdtas. Cobet, Aoytos ‘Epuijs (p. 531) and Professor Jackson subsequently (J. Phil. l.c. p. 139) maintained, and this with justice, that the text here was unsatisfactory; and proposed to strike out the words éorovdaxcévar ovde troXvTpaypovovyta as ‘a duplicate of cumtdafovtra fmv ovd€ trodvpabodyTa.’ But anything less like a ‘duplicate’ of this than éozrovéaxévas ovd€ TroAvTpaypovodrTa it is not easy to conceive. And it will be cbserved that the homoeographon in zoAv7pay- pov o0v Ta—roAvpal o dv Ta was quite capable of causing the omission from the text of Clement of a total of 32 letters. The same cause produced a similar effect in the text of the Evastae; but here the words lost were restored, though not to their proper place. The passage then should run zepi tas Téeyvas KxuTTalovta fv, ovdé troAvpaboiyta éotrovéaxévat ovdé TodAUTTPAypO- 1 uh obx qu Todro Professor H. Jackson, ac- 27. 140. cepted by Stahlin. 3 xumrdgovra Dindorf. 2 vAéyee Stahlin, Professor Jackson, J. Phil. * od’ % Professor Jackson. 244 J. P. POSTGATE vouvta @AXr’ aAdo Te x.T.€. There is now nothing faulty in the Greek. xuTrT ate Trepi requires no illustration, and o7ovédtw with a participle (ddacKcav as here moAvya0ovrta) is attested in Xen. Oecon. g. 1.! A very natural question now suggests itself. To what extent should acceptable supplements of undeniable gaps in the text of the Stromatets be expected to recognise the average unit-line suggested by this inquiry. From the nature of the case the precise form of such supplements can never be free from doubt, and with our present knowledge of the textual conditions of the Stromateis a systematic examination of the lacunae might, I fear, be both a lengthy and a sterile undertaking. But I may be permitted to give examples of supplements already proposed which, if the hypothesis we have been con- sidering is correct, will be seen to have unconsciously conformed. I cite then by way of instance the following: I. 17. 4 (I. p. 12. 26, St.) «a@azep 5’ of yewpyot mpoapdevcartes thy yHhv <« ottw 67 Kai npeis Tat ToTipwr THY Tap’ “EAL Oyo mMpoapdevouev TO yew@bes avToV, ws Tapadé~acBar TO KatTaBaddopevov orrépya TVEVPATLKOY Kal TOUTO Eevpapas ExOpérar SvvacOa. IT. 68. 3 (I. p. 149. 17, St.) Soxei S€ Kal GdAXAwS TPL@Y aTOXTs apapTtias TpbTeV SiddoKe O vopobéTns, TOY pev ev NOyot Sia TOV ixXOvwv TOV avavdwy* EoTL yap TAL VTL OY aLyN AoYoU Siadéper* eats kai cuyns axivdvvov yépas* tav dé év Epywr ia TOV apTraKTiKav Kal capxoBopwv opvéwy: yotpos BopBopar foerar. V. 59. I (I. p. 365. 26, St.) vai pv % Ilv@ayopov cuvovoia Kal % mpos Tovs opmiANnTas OiTTH KoLvovia GKOVGMATLKOUS TOUS TOAXNOUS Kai Tas paOnpaTiKoUs ETEpoUS Kadovaa, TOUS yunoiws avOatrTopéevous THS pirocotias, ovyi” macav Tact THY adnGeLay cuve- yopnoev (35 letters, Mayor)> dAda To pév ddobar To S€ Kai Kexpuppévor eivat mMpos TovS ToANOvs HuwiooetTo. IV. 152. I (I. p. 315. 25) xaOorov yap To maQntikov TavTi yever errOupias, eis dé THY aTaOerav Oeovpevos avOpwiros aXpavTws povad«.Kos yiverat. Lastly in II. 129. 1 (I. p. 183. 1, St.) madkw 8 ad Zyvov per o Srwixds TédAOS Hyelta, TO Kat’ aperny Chv KredvOns S5é€ TO oporoyoupévas Tit hice Civ... év Tat evAOyoTEiy 6 ev TH. TaV KaTAa vow ExroyHe KeicOat SuchdBavev. The supplements of Arnim Avoyévns 5€ 0 BaBudA@vios ro Tédos (29 letters) and of Stahlin Avoyévns d€ 76 TéXos KetcOat Hnryetro (or nryetTat, 30 OF 31 letters) will both conform, though I should prefer Stahlin’s, as 0 BaBuAwvri0s is not a necessary addition in a context where it is clear the Stoic Diogenes is referred to. A question intimately affecting the text of the Stromateis in more than one place is that of Clement’s accuracy in quotation. On this Professor Mayor says fairly enough, ‘Clement is never very exact in his quotations’ (C/. Rev. 8. 2370). 1 The quotation stops abruptly in Clement, éowovdaxéras. so I note that there are 55 letters from roiro to 2 Or, better perhaps, ov racav (33 letters). Hypsi Cleme TAaUTAa Ir KaATAX Kal o@ Kndopé ‘ou pei The re are as: diroco dwpnbe But to lambic inexact of Il. 1 couple I sugge Apropc Mr. Ar met wi TE €Tr’ a are an | the pro The me Str exalted OUTOS 1) Koo mov, Els yuo yeyupvd Buodv rreek, Lo KW hould 1s be From e free »f the oth a mples - con- I cite Jatrep LOUoLY Noyov TE ppa 7, ot.) S, TOV Noryou CTLEOV etters, 9, St.) tvovia , TOUS ouve- y elvat ap TO vitz) > A OLKOS pev 0 OS THL eto Oat . TEXOS , 30 or A@VLOS enes 1S an one Mayor 1. Rev. ers). ON THE TEXT OF THE STROMATEIS 245 So at IV. 53. 4 (I. p. 273, St.) where, continuing a quotation of Euripides Hypsipyle (fr. 757 Nauck), he gives according to the MS. elta émipépes ° TavuTa et aoréyetv amep bet xata pivow Siexrepav' ov dewvov ovdév TAY avayKaiwy BpoTois, Clement may have written orépyecv, though, that Euripides wrote ri Tavta et | or évertv, the evidence of Plutarch and Stobaeus leaves no doubt. In I. 67. 4 (I. p. 42, St.) we read Wuyas yap ayabas xata Idatova KaTaditToveas TOY UTrEpouvpavioy ToTrov UTopeivar éAOeiv eis Tévde TOY TapTapoV Kal c@pa avaraBovoas Tov év yevéces KAKOY aATadVTwY pEeTAaTKeElW VroNapBavovet Knodomevas TOD TOV avOpwTrwV yévous al Vopous Te EDecay Kal dirocodiay éexnpvEav ‘ov petlov ayaboy Tat TOV avOpweTrwy yéver OUT’ HAOEV Tote ex Dewy odT adikerat.’ The reference to Plato appears not to have been identified; but the last words are assumed to be a quotation from T7maeus, p. 47 A, B é& ov érropiocducba dirocodias yévos ot peifov ayabov ovr’ HrAOev ovTE HEEL TroTé Tat OvnTau yéver dwpnbev ex Oedv. If so, the quoting is something worse than ‘ not very exact.’ But to me the words od petfov dyafor irresistibly suggest the beginning of an iambic line, and I think that both Plato and Clement may have been quoting inexactly, the former no doubt for reasons of his own; compare his paraphrase of Il, 1. 25-28 at Rep. 393 E. Both forms of the quotation fall easily into a couple of trimeters; but all restoration must be tentative. In this spirit I suggest ov peilov ayabov ove Tar Ovntav yévet HAO’ éx« Oewv Swpntov ovO’ HEEev troté or ovT adiéerat. Apropos of the Timaeus I wonder if a suggestion which I made to the late Mr. Archer-Hind when his edition was preparing and which he accepted has met with any favour. It was that in 76 E the words dépya tpixyas dvuyas Te €m”’ axpows TOs KMAOLS Epvoav (with his necessary addition of 7’ after tpiyvas) are an hexameter taken from ‘some old physical poet’ and accommodated to the prose rhythm. , / 9 Ww / ; 2. & / 4 dépya Tpiyas T’ dvuxas T’ em’ Akpols Kw@XOLOLWW Efucar. The metre and prosody of the verse have nothing alien to Empedocles. Strom. VII. 45. 4 with the following chapter (II. p. 34, 18, St.) is an exalted description of the perfect Gnostic in the genuine Stoic vein. Beginning OUTOS HUY O YYwWOTLKOS O TLOTOS O TeTELopévos aptota SLotKeicOaL TA KaTAa TOV xoopov, in 46. 7 it pursues the theme in the following strain rau dé évOévde , / > / \ \ > \ e >? \ > a els yoMoews AKpOTNHTa Kal TO ETavaBReRnKOS wos avdpos EvTEXRODS yeyupvacpuéevwr po odov Ta KaTa Ypovoy Kal TéTOY GravTa GmETATTwWTWS Buobv éXopévwat kal aoxovvte. Sia THY THS Yyvouns wavToWey povoTtovorv 1 dcexmepav Stob. Plut., det 5’ éxmepay L. 246 J. P. POSTGATE ESpatotnta* Goos 5 BpiPovod tis Ett wroreimeTar yovla KaTw pétmovaa, kal KaTacTaTae TO Sia THS TicTews GVAYOMEVOY. The word in black type has been attacked by scholars who have failed to see its relation to the words which are put in spaced type. Yet no part of the Stoic doctrine should be better known than the comparison of the sage, the perfect man, to the perfect figure, the sphere of the divine and perfect universe and the perfect soul. I subjoin, chiefly from v. Arnim’s Stotcorum Veterum fragmenta, a selection of passages sufficient to make this clear. Aétius Plac. I. 6 opifovras S¢ thv tod Oeod ovciay oi Xtwikol ovTws .. . éxyov 5é Evvotay TovTOV TPWTOV meV ATO TOU KaAXOUS TOV Euhawopevwv Tpocrap- Bavovtes ...Kkanros b€ 0 koopos: Shrov bé€ Ex TOU oXNHpaTOS Kal TOD Ypwmpatos Kal TOU peyéBous Kal THs Tepi TOV. KOTpOY THY aaTépwY TroLKiNias. sharpoetdns yap 0 KOT MOS 6 TAYVTMY TCXHPATOV TPwTEVEL’ wovov yap TOUTO TOS EaUTOD pEpETgLV OpotodTaL’ Tepihepys Sé av éye Ta wépn Trepihepn: dia TovTO yap Kata Tov IIAdtwva év The Kepariu TO iepwratoy aouvéotnxev Oo voids, cf. Diog. Laert. VII. 148 ovciay dé Oeod Znvev pév pno. Tov GAov Kocpov Kai TOV ovpavov opoiws Sé Kal Xpvovrtos x.7.€. Aét. Plac. Il. 2. 1 of pév Xtwixoit cdhaipoedH tov Kocpov. Seneca A pocolocyntosis 8 (in ridicule of the deified Claudius) ‘’Emixovpevos @Oeos non potest esse: ov’ avTos mpayya exer OUTE AdXOLs Trapéyer. Stoicus? quomodo potest votundus esse ut ait Varro sine capite sine praeputio. est aliquid in illo Stoici dei, iam uideo; nec cor nec caput habet.’ Cf. the Herculanean fragm. Arnim II. no. 1059 as yap idiav Eyer poppnv TO oPatpoersés. Schol. in Hom. I/. VW 65 Xpvovrmos dé [Tas yuyas| pera Tov ywpicpory Tod cwpatos chatpoerdeis yevécOar doypatife. So Lucan g. g ‘ aeternos animam collegit in orbes’ (cf. ‘ orbem’ Ov. Fast. 6. 271 below). Hieronymus Ep. 108. 23 ‘ ossa audis et carnem et pedes et manus et g/obos mihi Stoicorum atque aeria quaedam deliramenta, confingis.’ Marcus Aurelius, 12. 3 édy... Touons Te TeauTOV olos Oo ’Epredoxretos ofat pos KuKrorepns povin. trepiynbéi yaiwv. Cf. Id. 8. 41, II. 12 odaipa wWeuyijs avtoedys, Hor. S. II. 7. 56 ‘(sapiens) in se ipso totus, teres atque votundus.’ The metaphorical language of Clement will however remain obscure until it is noted that it has a special reference to the central sphere of the universe, to wit the earth, of which, to begin with, we read in Aét. Plac. III. 9. 3 of Sroixol Kal of am’ aitav cparpoerdH tHv yav. Next may be cited from Achilles Isagoge 4 (II. 555 Arnim) part of the description of the xdcpos as a series of concentric spherical shells tov d€ ai@épa Kai ovpavoy (ete 0 adTos elre Suaddpos) EwOev elvar oparpixov oxjpa EyovTa. peta O€ TOUTOV EvTOS avTOU Tov aépa elvat, Kal avTov odhaipixas Trepixeipevov EEwOev Thu yi evdotépw O€ avTov Tpitny elvar opaipay, THv Tov KdaToOs, TEpi avTHY THY yHV peTaEd Tod dépos Kal Ths yns ev b€ Tat peocaitdtwL THY YynV eival, KévTpov Tak Kai péyeOos 1 The doctrine of the perfectness of the sphere _ the xéoyos) and 73 C sqq. (of the brain and head) is of course pre-Stoic. In the Timaeus of Platoit are not without their instructiveness for parts of appears in more than one application; 33 B(on the present inquiry. éméyouo TEPLOLVE édparoTy Taxo0ev jadw « iooppomre feVOV OT 5é date elmrev 0 péoov* « KaL CUYD avappipe Katagpep apparen use of ‘1 of the 1 This is which I but on \ fulcimin position évravaBe of light premat f keeps th parts — (angle) s down wl disclosec which is bility of op. cit. ci should h OTL TA ge tecOar § The not: M. Aure ewt Te June 22 U: 1 Stoic ir been tracec T@ | to the the ect um ON THE TEXT OF THE STROMATEIS 247 éméyovoav ws &v ohaipar> Kai Tas pev addras Tpeis chaipas 7 Téccapas meproivetcOar + THhv SE THS yYns povnv éEatdvat. This povorovos édpaorns, ‘isolated stability,’ is described further on: «ai thy ynv 6€é, Trapv- Taxobev b7ro Tov dépos WOovpEevnY icoppoTras ev THL péow Eival Kai EoTaVAaL* 7 Taw womep eb Tis AaBwv copa Synoee TravTaxoPev cyowioas Kai Soin Tioiv isoppoTws édxew ém’ axptBeias+ cupBycetar yap Tavtayobev errions TeEpiEedxo- pevov oTHvat Kal atpeunoa. Plutarch de facte in orbe lunae cap. 6 p. 923 E yhv Sé hate tyes (Stoici) dvev Badcews cai pitns Siapévew: avy pev ovr, elev 0 Dapvaxns, Tov oixetov Kai Kata dvow ToTov Exovear, doTrEp avTHL TOV pécov* ovTos ydp éoTt Tepi ov avTepeider TdvTa Ta Bapn pérovTa, Kai hépetat Kal cuvvever TavTaxXdOev * 1) 5€ dvw ywpa Taoa, Kav Ti déEnTat yewdes WTO Bias avappipév, evOds exOriBer Sevpo padrov O€ adinow He mépuKev oiKeiar poTri xcatagepopevov. I have quoted this passage in full because it contains an apparent but very natural and insignificant discordance with Clement in the use of ‘up’ and ‘down.’ Toa spectator on the surface of the earth the rest of the universe is ‘up’; to one at the centre of the same it would be ‘ down.’ This is the point of view in a noteworthy parallel in Ovid, Fasti 6. 169 sqq., which I have already discussed and interpreted in this Journal (IV. pp. 196 sqq.) but on which I have now something further to say: ‘terra pilae similis, nullo fulcimine nixa, | aere subtecto tam graue pendet onus.’ It is thus that the position of the earth presents a parallel to the yya@oews axporns and the érmravaBeBnxos tirwvos of the ‘perfect man.’ The next pentameter throws a flood of light on Clement’s ywvia ‘ipsa uolubilitas libratum sustinet orbem | quique premat partes angulus omnis abest,’ or, in my translation, ‘ its very rotundity keeps the globe in equilibrium; it has no corners to depress any one of its parts "—an exact parallel to our author’s expression ‘ All that have any corner (angle) still left to depress them by its downward moment find that dragged down which faith would elevate.’ The comparison of the passages has now disclosed that Ovid was drawing from a Stoic source,! another indication of which is his identification of Terra and Vesta and his insistence on the immo- bility of the earth as the centre of the universe, as we learn from Plutarch op. cit. cap. 6, p. g23 A that Cleanthes said Aristarchus of Samos’ countrymen should have indicted him for impiety as xiv odvta Tov Kbopou THY EgTiav Ott Ta havopeva owrtery avnp emepato, pévery TOV Ovpavoy UToTépevos, e&ediT- tecOar S€ Kata AoEoD KiKAOV THY YHV, Gpa Kai Tept TOY avTHs AEova Swovpéevny. The notion of absence of angularity in the perfect man’s soul appears also in M. Aurelius 11. 12 (l.c) odhaipa Wuyis atvtoecdys, dTavy pHnte EMLTELYHT aL ETL TL PHT ETW TUYTPEYX NE. J. P. PosTGATE. June 22, 1914. UNIVERSITY OF LIVERPOOL, Stoic influence, it may be added, hasrecently Mr. F. E. Robbins, Classical Philology, 8. (1913) been traced in the cosmogony of Ovid Met. i. by pp. 401 sqq. THE CODEX BAMBERGENSIS OF THE FIRST DECADE OF LIVY (I.-VII. 17). I EXAMINED this MS. at Bamberg in the Christmas vacation of 1912-13 for the purposes of the forthcoming critical edition of Livy by Professor Conway and Professor Walters. In 1856 H. W. Heerwagen described it, and published a collation of the first book.1 Apart from that it has been until now neglected. Bekker had it sent to him at Leipzig about 1830 for the purposes of his text: on its way back it fell into Kreyssig’s hands (see the latter’s preface to his edition of Book XXXIII.), but neither of them seems to have made any use of it. It was collated by J. G. Seebode in 1835-8 (as is recorded on the flyleaf); but no results were published. It is, however, curious that Weissenborn in his second edition (1859),? though quoting Heerwagen as his sole source of information about the MS., which he calls B, cites its readings regularly not only in Book I., but in Book II. also, up to c. 22; after this only four or five scattered readings are cited. The source from which he ascertained these are not known; in any case, they are few and unimportant.® Conway and Walters’ apparatus will contain the first account of the MS. ever given which approaches completeness. It has now been examined at every point where any of the other MSS. show any variation of the least importance from one another, or where the text has been for any other reason the subject of discussion. § 2. The MS. (Conway and Walters’ B) may be assigned from its writing to the late tenth or early eleventh century. It came, like most other Bamberg MSS., from the Dombibliothek, as the inscription on the flyleaf testifies, Ad Aibliothecam Reverendissimi Capitult Bambergensis. Some account of the history of this and the other Bamberg MSS. of Livy will be found in Traube, Palaeographische Forschungen, 1V. Teil (Miinchen, 1904). It used to be assumed that the ‘duos libros Titi Liuii,’ or, as it originally stood, ‘ Titi Liuii non minimam partem,’ which are mentioned in an eleventh-century catalogue of the books of Otto III., ‘quos placentiae inuenit sibi seruatos,’ found on the blank page of a medical MS. (L. III. 8 in the Bamberg Library) 1 Sollemnia Anniversaria in Gymnasio Regio Baruthino, Baruthi, 1856. 4 Preface, pp. Ixxvi, xcii. 3 I have noted the following errors: ii, 5. 1, he cites B as reading 7bi, whereas it omits all from 2. 8 to 5. 3 (v. inf.): ii. 5. 10 he give’s B’s reading as uindicias, where it is really uindiciae. 4 Traube gives further details, partly con- jectural, of the romantic history of these works, which seem to have been collected by John, Bishop of Placentia (Anti-Pope in 997, and then put to death by Otto) ; Traube supposes John to have been formerly Otto’s tutor. They de- scend to (St.) Heinrich II., whose name is great in the history of Bamberg: he bequeathed his library to the chapter on his death in 1024. V435" 44 ‘ - tty " m yn st } « Asta vv , ‘i.e > “ a, J tiv oo rpaqgect 7 jeawly ¥ NTE DICE vyisgeasteey’ ctehers , : = f ‘ J ‘ne?! SPAS ted cuter pe "| TIS mMosf.-199- ~F we: Seas ad arreeg™ - a | Tos ‘ ttirn Che irs $+" 2.200-Te0 . P $ ¢ ’ ee OS nh oe OS oe hd a ee ee ee ed ee oo } 1 , - ’ P&M oO ANT! ater i f FULT LA i alate ofist ee | hothe Sera (tuts. all | _ > getter tte ceeds feet 1overpaet ogee tery PiciT Stcenry be Topsy .& Dagmr cha he * dol : MMe ditrAinougeculs mistufYaronrtnss Miley it 7° LTR NT “ ’ ' a — f i= oo MV” cami Lin elOeFR aia Coss gua pougeerryy Pros te I. gyi." Piwrwevyettyl: Vite " - : f oi . ’ e c ame soles vet Are an) TeV NVM HETIN angele TP! j50143 AST TTYY” SPF mOTI DAI PeMerritet cry 7 crus sd ‘ : “Y t ‘ { ; ~* . ‘ - “ . - $ ’ ’ . ; a —— be og - rete are Bed ’ * MPS re We veqeon ott fy 7 prert . DEITT SHV de? SIV TTT M Vie Ts CMT POLIT P pias t - , 4 . _ ‘ . ° ‘ ‘ sae ‘ qs! Mr ai a , So PTD Rie PHP MI EOE yas ie? RAPE NNTB pomelris Els . ; nn oy? s* 8 . ° ? , 4 , ~ a 22gfRl CORP AN PITT Mage MU TIA Spat ies? tat tater conringiy Comise BET ' evista ‘ , : — or che Poets 1? ~ 4 [> 1799 74 ee t i+ ’ ./y Fee “7; +5 * aes P ra, ~ 4 7 1A F eu & o ' = “uw ’ “~w34t7 “f { y Lp or Ween e eee) pe & 21) ‘SoR LETS GR Ds iat” Tarre te +9 99 We ind‘ eyrt % PT het? j Ms —* : pyeehe Ara DiTreg7r easy be, Mw ry ar $0 Y poet” ge97 : EST “! $y 203 COM ree” peti ned! pet lags 4a he 7 ." ‘ 4 . F ’ o 010° . t+ - 7 af . att. a 8 * berth Be OL) ee. ek ee Ll“ rigger tere MAN I" yo Nery Mesnoe? plebe ascie Weer Aer % 2 ' (tt 4 rw" Tr Pe wera Treas 14° R2ee ay? Lreet Poe’ 226° ou aheoe~ e218 Te? yt : 2. 5 . Hy Ts oe Le = ws | 4<"I MeL] PAS TLV IT EVENT. FON peg tive CRP Ay Tiere 867) Fe ryeeise oh in +" j f : { 4 -_ - ae y * ve i ° a? > o ; era 9" —e yen * bodes .* 14 eh — Lire?® Pees Lae tmS te TATION! CUT TAUACUTY Sverre ae ttl if S| TUatoeck « P ~~ ’ oleniuy? ef - f lowe) rempore ~ » 9 ! . t wd oe 7 ar ~ oo ot 4 4) Tee & bt Pie 3 Pore wry THIRTIES c*s -€ 4 TT a ag? MAT?) ‘2 tiger 3 24x erer perenne easy: Arpt . vhs cto errr RQ aye Orr wet ers port 12> } 2 $ i? s —_ * ae ae Te 71+ a. %i ~ a ‘ ‘ t ’ . _— . er 5 ark th eee tae A ue" ad eNT -e oFTVI Ate tt Lor TUN VirAcepresnst impem eres Sal ayTD copes ) 4 ‘ee ru oe) or wy Picr WL’ < a &-% f19 4¢ ty } gare he te fon ee © hab wed ve 0 tn * 13 « } tT "a" P . ~-oIT 2 FI ee » snrre™ Fad | pes » - le. et. Tris — bwee? ia di . PAT RCTS PROT! Parris Cu adit Ve wens 7 f f + f ? Bix sata — Pa Ya bit sf + Fawn Sct . a ie 7" = 31+ —. Se lanbeaes igi * Lag i ype ne i« . , ? eas ‘ ——— ° sabteweF Ti tltc sT -tUl Tis 9) ‘ ¢» ~ce disc 1% Sra Soir. p 1 oS. CMP eU : e. ‘Aameembit \ - ~ s-= 6 7% 4% 4 j Tine -wt Yeu ress slay 17 ue bet nk’ Ime rv eet: or +e) . " af Fri APPEL Pee av. ONInion , + a i y ~ ’ v7») al cw tilr Riisgii.® ~ rs 5 > ‘ ee 4 4 ‘ Se ss Se 4a oe . 7 gait? / " a4 ‘ a ‘5 +27 7 * uTT i ct ee \ y~ FYI s¢ ; ' 2G YT OA COLIS ovetyirs >.<” sPpae™Te &)- . i S .~1y Rtyermof lo ‘ , te é al AA auise Yy** M414? 734 by + . ah aes ka” > + ¢- itt Ser ist <*.% sretes Sie & j . i 3 . - : j ee %~, Til . PU 8ee TUTE wis Urey: ry 7 he Tce ‘ ’ a3,° 94.8 o£ ae 4 atu set 3 *F TU rt syet ’ . ‘ 4 . 4 > “TTL $7 ae: .&* ~ — ~ a. - _ ' " +,% ¢ eeu otic é4 . oe % ’ }“L? a See Se OM 1%e7, wwe Berasisyc” poe , , } : o oe c® , . F saute: ttn tag 4 wmTts ° TS /9 ’ ° > 7 . 7 4 sy ? " HT iti a? aT .2t te $ + os - e ' ih _ ee +4 7 SV3crt j ; ‘ 4S - SA RR A MN THE were Ou But Tre the sixt ments Possibl\ as we fragme S 5th qua I to X the last pages a at leas though ab urbe At leas any ma is that III. 25. All the for inst his ‘I ‘ feliter part of a hand as I17 is ofte fol. 60° noticed in the earlier interes Minusc only us His ha often ¢ § appare or heac is vent of the fewer, summe in the THE CODEX BAMBERGENSIS OF LIVY’S FIRST DECADE 249 were our own & and the more famous Bamberg MS. of the fourth decade. But Traube shows (op. cit. pp. 11 sqq.) that one of them at any rate would be the sixth-century uncial archetype of the fourth-decade MS., of which frag- ments were discovered in 1904 at Bamberg by Oberbibliothekar Fischer. Possibly the other may have been the lost archetype of our 4, from which, as we shall see, the Codex Floriacensis (F), which, save the Veronese fragment, is our oldest MS. of the first Decade, was also copied. § 3. Our MS. is written in single column, 33 lines to a page (in the 5th quaternion only 31), and contains 112 folios in 14 quaternions, numbered I to XI1V—the last three also lettered F, G, H—breaking off at the foot of the last page with the words trepido agmine inciderunt, VII. 17. 3. The last pages are very badly worm-eaten, quite as much as the first pages; and it is at least possible that no more of the MS. than we now possess ever existed ; though the colophon which stands at the foot of the last page, Tituls (sic!) ab urbe condita, is no argument for this, being in a later hand than the text. At least three hands were employed upon 4. I have not, however, noticed any marked difference among the scribes in accuracy. The best and neatest is that which wrote the sixth quaternion, foll. 41-48, containing II. 56. 2 to III. 25.5. In this part the words are hardly divided from one another at all. All the scribes of B seem to have been anxious to save parchment. This one, for instance, finishes Book II. two lines from the foot of the page, but crowds his ‘Incipit feliciter’ into the end of the line (cutting down ‘feliciter’ into ‘feliter’ to make it fit), and begins Book III. on the same page. In the third part of the MS., fol. 49, III. 25. 5 to end, written in another and not so neat a hand, the overcrowding is at times still worse. Some lines contain as many as 117 letters: in this part (not in the others) an extra half or quarter line is often found at the foot of the page, outside the regular alignment. On fol. 60", lines 4, 5, there are ten examples of the open a in two lines: I have noticed two other examples on foll. 60’, 108, but have not seen it elsewhere in the MS. In F, on the other hand (which is, of course, assigned to an earlier date), the open a is very frequent. On fol. 62’, 1. 28, there is an interesting siglum 7X for respondit (IV. 6. 2): Lindsay (Contractions in Latin Minuscule MSS., pp. 3-4, 10) says that in the eleventh century this siglum was only used in legal formulae. Yet another scribe wrote foll. 1-40 to II. 56. 2. His hand is the coarsest ; he divides his words more than the others do, and often divides them wrong. § 4. A peculiarity of this first part is the existence of marginalia all apparently in the same hand and ink as the text. They are mostly summaries or headings of the ordinary type: nothing in the way of comment or criticisra is ventured upon, except ‘ mira descriptio,’ opposite the account of the battle of the Horatii and Curiatii, I. 25 (so again, II. 5-6). They get fewer and fewer, and cease altogether after fol. 26 (II. 15). Among the marginal summaries are other marginalia of a different character—words which occur in the text are written in its margin opposite their place, usually in capitals 250 A. H. KYD and in an abbreviated form; e.g. its first is fol. rz", 1.1, MERC referring evidently to mercatu on the last line of the previous page. Then on the same page we have LAP for lapidibus, fol. 12° R for regum (?). On fol. 1o9¥ INCESSIT, fol. 22y INCESSUNT are written in full (in the latter place the word in the text is zmcessit II. 7. 1). And in some cases its words are written in full in minuscule: fol. 17% Inuectus, 18 eniuef, 20 Inopinato, 22” parma, 23” conparant (text conparandii). The words (apart from those already mentioned) represented by abbreviations in capitals in the margin are: fol. 13, augur: 21, lapicidas, abrogaret, exactique: 22, inuectis, deligatt: 26 (the last marginale of any kind), distineat (DISTIN). On 24v RITAS (all that can be read) seems to refer to caritate. — Urlichs (Eos, I. 1, pp. 84 sqq.), who noticed these marks, suggests that they refer to words which stood at the top of a page or column of 4’s archetype. But the intervals at which they occur are so completely irregular that this does not seem likely: and on fol. 21" there are three within thirteen lines. On the other hand, they are not all rare words of the kind that we might expect to have been specially noted for a glossary. They might, however, be simply lemmata for a commentary. A note like that in P (v. inf.), fol. 7, FASTI ET NEF. D, opposite nefastos dies fastosque (N.B. neglect of order and case) strongly favours this view. § 5. Turning over the leaves of the tenth-century Paris MS. (P, Bibl. Nat. 5725 Lat.) on a recent visit to that city, I discovered that the marginalia of & seem to correspond, as far as they go, exactly with those of P. (F, on the other hand, though #’s own twin, seems to have no correspondence with it in this respect: it has few marginalia of any kind.) The marginalia of P may be divided into three classes: (1) Corrections of the text by P? or later correctors. Of course & has nothing corresponding to this. (2) Marginal summaries, almost always in the same hand and ink as the text, though one or two have been added by later hands. Wherever I have notes of the marginal summaries of B, I find the same to exist in P. Photographs of three pages of JS, foll. 6%, 7°, and 21", corresponding to within a few lines with the folia similarly numbered in P, show how close the correspondence is. The Unde Interregnum of P, fol. 6¥ (top), does not show in & (though it might conceivably be at the bottom of the previous page); otherwise, except for trifling errors such as 4’s Laus mammae (sic) for the Laus numae of P, the headings or summaries are identical. It seems fair therefore to presume that no heading occurs in B which is not also to be found in P. Such summaries in P are very infrequent (though they do occur here or there) in Book II. after the point at which those of 8 cease; and seem hardly to be found in later books, except in Book IV., where, for some reason, they are regular. (3) Repetitions in the margin of words in the text, usually, though not always, abbreviated and in capitals, answering to those noted above in &. Every note of this kind which occurs in B (except the R of fol. 12%, which is in any case a doubtful example) is to be found in the corresponding place in P. But in P there are a great many more of them; not only are there many ee ae - — ——— OT ee Rote ee OTH oe THE addition to be fo frequenc is that nearly < characte and the! where s observes with th F was c § 6. text. I obvious F, was | wofully of F. 7 Lucretia VII. 12 close re Conway the rea reipublic confirm $7 which 1 to B is B conté dominis § 8 the orig or less stitute in I. 19 time th overloo! MSS., | an eras the nu infensus 1 They 6. 4, 61. 18. 3, 18. 35- 7: 49. referring the same fol. 197 place the e written Vv parma, - already +: fol. 13, (the last t can be > noticed top of a »ccur are 21° there re words glossary, note like Jastosque P, Bibl. arginalia (F, on nce with lia of P ‘ or later Marginal ugh one ; of the raphs of nes with lence is. it might cept for f P, the me that mmaries 300k II. ound in regular. always, Every ch is in ice in P. ‘e many ee THE CODEX BAMBERGENSIS OF LIVY’S FIRST DECADE 251 additional examples within that part of the first two books in which they are to be found in B; but they are continued, though perhaps with diminishing frequency, throughout the whole MS. What is especially interesting, perhaps, is that these notes, as distinct from the ordinary marginal summaries, are nearly all in an ink perceptibly paler than that of the text. This is the characteristic mark of the corrector of P called P? by Conway and Walters: and their Notes show that wherever P differs from F# the corrections of P*, where such exist, tend to bring it into line with them. Professor Walters observes that the discovery about the marginalia would seem to fit very well with the view that the corrector P? had before him the archetype from which F was copied, and from which, later, 6 was drawn. § 6. Our MS. represents the ‘northern tradition’ of the Nicomachean text. Its similarity to the Parisinus (P), the chief northern MS., was at once obvious to Heerwagen. In his time the other Paris MS. of the first decade, F, was unexplored. F has the distinction of seniority, but its value has been wofully lessened by its correctors. Now # has been found to be the twin of F. That it is not a copy of F is certain, since in III. 44. 1 (urbe nefas.. . Lucretiae), in III. 64. 3 (Quid ... adorti essent), in V. 31. 4 (im fuga), in VII. 12. 8 (ves), words and phrases omitted in F are found in &. But their close relationship will be evident on the most cursory inspection of Professor Conway’s critical apparatus. To take one instance out of hundreds: II. 47. 12, the reading of all other MSS. is salubri reip. parte: but Gruter’s conjecture, veipublicae arte, made before either F or B had been heard of, has now been confirmed by reiparte having been found in both. § 7. B has in common with F the large lacuna in Book IV. 21. 6—50. 4, which in F occurs between two pages, in B in the middle of a line. Peculiar to B is a lacuna in Book II. 2. 8, eadem multis—5. 3, ets hominum. Like F, B contains only one Nicomachean subscription, the ‘ Victorianus emendabam dominis symmachis’ at the end of the first book. § 8. In the part of the text contained in B there are 49 places in which the original reading of F is unknown,’ and at least 123 in which it is more or less hypothetical. The great value of 4 is that it enables us to recon- stitute in these places the text of the common archetype of BF. E.g., in I. 19. 6 Professor Conway has been enabled by B to restore for the first time the true reading, desuntque sex dies (written in B des qui dies). This was overlooked by Heerwagen. F's corrector, like the scribes of all other extant MSS., has not understood what he took for the relative, and so in F there is an erasure of two letters before dies. In no other MS. is there any trace of the numeral at all. A few other instances are: III. 28. 9 (is tgnomimiam infensus addidit) infensus HRn? DAld: insensus ML : incensus PUPBOD*—in F %. 18, G&S J Bo He WT hm 1.7 De MR OO, 59. 6, 61. 4, V. 3. 4, 4. 8, 6. 7, 36. 9, 40. 4, 44. I, 53. 2, VI. 1. 3, 13. 7, 18. 1, 42. 2, 42. 14. 1 They are: I. 5. 6, 18. 9, 19. 6, 23. 9, II. 6. 2, 6. 4, 61. 3, III. 11. 11, 12. 4 (bis), 14.6, 16. 4, 18. 3, 18. 11, 28.9, 29. 5, 34. 8, 35. 2 (bis), 35. 5, 35- 7, 40. 12, 48. I, 49. 3, 51. 8, 52. 2, IV. 2. 7, 252 A. H. KYD the word has been entirely lost in the erasure. III. 40. 12, decemuirum (gen. pl.) apse scil. unus (cf. IX. 34.1; XXVII. 8.4) MPBRuDL : decemuir ipse UpOHRn?D*—in F the noun is completely erased. V. 44. 1, eguit (rightly) is given by Upsaliensis: ego id P?B: egit MF7OHDLA—in F all but the letters eg .. has perished in the erasure and correction made by F°. § g. Of the 122 places where the corrector has only partially succeeded in obliterating what stood in F and where the symbol F ? is used in Conway and Walters’ critical notes, B confirms the reading suggested for F in 86 cases and differs from it in 36.1 I have myself examined F at some of the most important of these points, specially for the purpose of comparison with B. E.g., Il. 34. 12, Haud tam facile dictu est is the reading of UpRn. Apart from F, all other MSS. and B have dictum est, which is kept by Alschefski. In F the state of the parchment-surface makes it just possible (though very doubtful) that a contraction-stroke may have originally stood above the u and been erased. This hypothesis is favoured by the report of B. In IV. 2. 3, maximum Romae praemium sedttionum esse; ideo singulis uniuersisque semper honort fuisse, MPUPBOHDLA?’ have id et singulis. Ideo was the con- jecture of Weissenborn, eas is the insertion of Conway and Walters.? In F, the three letters which follow td- are joined together in such a way as to make it doubtful whether they really are et s, and it is suggested that F has ideo. The centre letter of the three, however, seems too triangular to be an 0, and B’s reading is an additional argument for Weissenborn’s originality. In § 11 (according to Conway and Walters’ arrangement; otherwise § 4) of the same chapter guem ad modum plebs gloriari posset, it is thought that F may possibly have had gloriaretur, as in the gloriart possent of F* the last word and the latter part of the first are written somewhat closely over an erasure. But B agrees with F*, gloriari possent. V. 27. 11, Fides Romana, wustitia imperatoris . . celebrantur is the reading of MPFxUJOEHDLA, and, almost certainly, of F himself, although it has been overwritten in blacker ink by Fx: B here exhibits (with Ver.) a genuine difference, celebratur. V. 47. 2, saxo in adscensum aequo is the emendation of Heraeus. In F the -u of ascensu has been over- written by its corrector, and there seems to be an erasure after (not above) it, but perhaps hardly large enough te have held a normal-sized m. B&B, with MPH?, has ascensu. §$ 10. Turning to those passages in which there is no doubt of F's reading, we find the comparison between F and & no longer so favourable to the latter: B is clearly a less accurate copy of the archetype, and acquires merit beyond its twin solely through having escaped correction. Of 290 places in which FB differ (of course the number of cases in which they agree is enormously larger) there are only 85 in which the difference is in B’s favour. 17. rr. 8,17. 2, 28. 4, 56. 1, 56. 11, II. 10.5, 13.13, 19. 2, 58.°6, V. 18.1, 27. 11, 47.2, 47. 7, 18. 3, 18. I1, 24. 5, 27. 2, 34. 12, 39. 3, 52. 5, 47.9, VI. 2. 10. III. 9. 1, 11. 12, 12. 4, 12. 5, 19. I, 45. 2, 56.9, 2 See their article, C. Q. v. (1911), p. 5. IV. 2. 3, 2. 11 (bis), 2. 13, 3. 10, 7. 9-10, 12. 8, THE Apart fr may quo other M perdita ¢ blunder III. 19. B with ‘ correct forte, F O, By a administ § 11 amend t non PU: oe different § I: biblioth Bamber and aft make tk Conway former, manusc: THE nurrum uty ipse ightly) letters ceeded onway H cases > most ith B, Apart hefski. h very /u and 7 7" -Y SUSQue € con- In F, ) make S 1deo. 0, and n § 11 > same »ssibly - latter agrees rer , of F > here rensSUuM over- ve) it, , with of F's urable quires If 290 agree avour. 9 47+ 7) THE CODEX BAMBERGENSIS OF LIVY’S FIRST DECADE 253 Apart from the places mentioned above, in which B supplies lacunae in F, we may quote as instances: II. 26.1, B with MP im urbem (nuntiatum est), F and other MSS. in urbe. III. 19. 4 (the tribunes conducted themselves), ut in perdita domo. B alone with Ver. and D* has domo, all other MSS. the blunder modo (perdita being taken to agree with lingua, the word following). III. 19. 12, Nescto quo fato magis bellantes quam pacatt propitios habemus deos. B with Ver. P* UpfO has fato, F and all other MSS. facto (which F* has ‘corrected’ to pacto). VI.Q. 10, porta... quae una forte non obsidebatur, B has forte, F fronte. In Il. 56. 2, permissurum HRnDL administraturum P?F ? O, B, along with Vorm.MPUf}, has preserved both readings side by side, administraturum permissurum. § 11. There is seldom evidence of any conscious attempt on B’s part to amend the archetype, though we may quote: II. 7. 10, adeone} adeon F.: adeo non PU4B ; III. 67. 11, submouit] subuit F: subcubuit B; VI. 14. 13, differenti . . dicenti F'al (differenti ... dicendi F ?); differendi . . . dicendi B}, differenti . . . dicendi B. [F'=the first hand in F correcting himself. ] § 12. I have to express my heartiest thanks to Dr. Fischer, Kgl. Ober- bibliothekar, and Drs. Plébst and Miiller, Kustodes of the Royal Library at Bamberg, for their kindness and help, both when I was working in the library and afterwards in ascertaining for me some further readings of B required to make the account complete. I need hardly say how grateful I am to Professor Conway and Professor Walters for their help at every point, especially to the former, to whom 1 owe the opportunity of making acquaintance with the manuscript. A. H. Kyo. THE UNIVERSITY OF MANCHESTER. IN MEMORIAM. The writer of this paper, Alastair Hope Kyd, Assistant Lecturer in the University of Manchester, and formerly scholar of Wadham College, Oxford, was engaged in making a few final additions to it, when he was seized by an illness which ended fatally only five days later (April 26, 1914). This has been his only complete contribution to scholarship, and he was able to send it to the Editors in its present form ; but the duty of seeing it through the press has fallento me. At one point (p. 250, line 12 from foot) I sought to supplement Mr. Kyd’s notes by direct information from Bamberg, but by the lamentable out- break of war in Europe we are now cut off from the kind help of Herr Fischer. Even the readers of this paper, to whom Mr. Kyd was a stranger, will realise how bright a promise has been cut short. His colleagues at Manchester and his many friends mourn the loss of a frank, un- selfish, and singularly lovable comrade. oe R. S.C. NOTES | Here adventur Ehwald | das folge: NOTES ON OVID’S JB/S, EX PONTO LIBRI, AND | ss seine avd HALIEVTICA. who was | the meat IBIS 131-132 mask * Cs quam dolor hic umquam spatio euanescere possit, I pr leniat aut odium tempus et hora meum. | dumitis te Here ‘spatio’ means ‘ lapse of time’: it is illustrated by A. A. II. 113 | Auds: tup forma bonum fragile est, quantumque accedit ad annos, pepe fit minor et spatio carpitur ipsa suo. | Zeus we As regards the whole couplet, besides at this place, it is found also after ae line 40 in all the MSS. except the Galeanus Vaticanus and Phillipps MS. ny * There, though it fits in with the context, it is not required: here (after 130) | _ _ it is indispensable. It should therefore be omitted from the text after line 40, | soll oats where its presence is due to that species of interpolation which consisted in the insertion of other portions of a writer’s work kindred in meaning, on which the ‘in’ see Mr. Hall’s Companion to Classical Texts, p. 198. | good ins IBIS 137-140 tecum bella geram: nec mors mihi finiet iras, | Here. fo saeua sed in Manis Manibus arma dabit. EY tum quoque, cum fuero uacuas dilapsus in auras, —_— found in exanguis mores oderit umbra tuos. ssumeres These two couplets are closely connected in sense, and should not be othtinns| separated from one another by such transposition as Mr. Housman has intro- , duced into the text in the Corpus Poetarum Latinorum. In the first couplet | a Ovid says that death will not end his wrath, but will furnish his spirit with | parum | merciless arms against the spirit of his enemy. This idea is repeated in the XIII. 7: next couplet with redundancy characteristic of Ovid. Heinsius’ emendation ‘ dabit ’ is preferable to ‘dabo,’ the reading of the MSS., because Death is the subject and efficient factor in the sentence. ‘ Dabo’ originated through TBI: assimilation to the termination of ‘fuero’ in the next line, and possibly of ‘ero’ in line 144. This class of error is treated by Mr. Hall, l.c., p. 174, The forms of this verb are often confused in MSS. Thus at ex P. I. 2. go Thi meque minus, uitam cui dabat ipse, capi, and hin ‘dabat,’ restored by conjecture by Merkel, is right because it accounts for the Minns, | reading of the three best MSS., ‘ dabit,’ for which the rest have ‘ dedit.’ \.ND 113 50 after ps MS. ter 130) line 40, sted in . which not be ; Intro- -ouplet it with in the dation is the rough bly of The NOTES ON OVID’S JBIS, EX PONTO LIBRI, AND HALIEVTICA 255 Isis 289-90 utque parum mitis, sed non inpune, Prometheus aerias uolucres sanguine fixus alas. Here ‘parum mitis’ has caused trouble. Ellis translates, ‘ whose ill- adventured philanthropy’: but this meaning cannot be extracted from ‘parum.’ Ehwald (Jahresbericht XXXI. 190) says: ‘ich glaube das parum mitis ist wie das folgende sed non inpune zeigt, auf die Schuld des Pr., zu beziehen und auf seine avGadia’: that is to say, it refers to the defiant character vo: Prometheus, who was not sufficiently disposed to yield to Zeus. This seems to me to force the meaning of mitis, which means ‘gentle’ (evvous) rather than ‘ yielding, meek ’ (razreivos). I propose to read ‘parum inmitis.” Prometheus was not, like Zeus, inmitis to mankind, but was considerate (Aesch. P. V. 10, ws dv dday0 thv Avs tupavvida | otépyerv, piravOpwrov dé mwavecOat tporov. 445 réEw 4é, peru ovtiw’ avOpwras Exwv, | arr’ wv dédwx’ edvorav éEnyovpevos) ; whereas Zeus was an insolent monarch, cruel towards mankind. ‘ Miserable and brutish as their state was, he would take no heed of them; but was even ready to sweep away their whole race’ (Prickard, Aesch. P. V. p. vi). The error is due to haplography : the ‘in’ of ‘inmitis’ disappeared beside the ‘ um’ of ‘parum,’ as in Iuu. XI. 106 ac nudam effigiem in clipeo uenientis et hasta the ‘in’ (inserted by Valesius) disappeared beside the ‘m’ of ‘effigiem.’ A good instance from Ovid is quoted by Mr. Hall, /.c., p. 190, ex P. I. 4. 35 nos fragili ligno uastum sulcauimus aequor: quae tulit Aesoniden, densa carina fuit. Here, for ‘densa’ the Hamburgensis, our best MS., has ‘sa,’ which was mis- interpreted as a contraction, and gave rise to the pointless reading ‘sacra’ found in many MSS., and interpolated from I. 3. 76. It is clear that ‘ densa,’ preserved by the excerpta Scaligeri, is the right word, for it supplies the antithesis required to ‘fragili.” ‘ Densa carina,’ a ‘solid ship,’ is like Lucan’s ‘densus murus’ (Phars. III. 491) and like Homer’s trucuvos Sopos. For ‘parum’ cp. 521, ‘parum stabili carmine’; ex P. II. 10. 15, ‘ Naso parum prudens’; III. 2. 80, ‘parum fausta puppe’; for ‘inmitis’ cp. Met. XIII. 739, ‘ genus haud inmite uirorum.’ IBIS 307-308 aut pia te caeso dicatur adultera, sicut qua cecidit Leucon uindice dicta pia est. This Leucon appears to have been a son of Paerisades, king of Bosporus, and himself to have been king of Bosporus in the third century B.c. See Minns, Scythians and Greeks (i913), p. 581. 256 S. G. OWEN IBIS 413-416 qualis Achaemenidae, Sicula desertus in Aetna Troica cum uidit uela uenire, fuit, qualis erat nec non fortuna binominis Iri, quique tenent pontem, t que tibi t maior erit. So the last pentameter stands in the MSS., except that for ‘que’ the Galeanus has ‘qui,’ the Bodleian Can. Lat. 20 ‘qua’; the last MS. also has the extraordinary reading ‘ maphor’ for ‘ maior.’ Most editors mark the line as corrupt, and no satisfactory solution has been proposed. I believe that ‘maior’ is an interpolation which arose from line 412 euenient aut his non leuiora malis. The error is due to what Havet calls the suggestion of the context (Manuel | de Critique Verbal, ch. XXV). It was felt that, as there, so in this line, a comparative was wanted. But what is wanted is the correlative to balance ‘qualis.’ I read quiaue tenent pontem, uae tibi talis erit. The construction is: ‘qualis fortuna Achaemenidae fuit, et qualis erat fortuna Iri qualisque est eorum fortuna qui tenent pontem [beggars], talis uae! tibi | fortuna erit.’ For the omission of the antecedent ‘eorum’ before ‘ qui’ cp. 295 nec tibi fida magis misceri pocula possint quam (ei) qui cornigero de Ioue natus erat. 259 } nec plus aspicias quam (is) quem sua filia rexit. | Iuu. I. 161 accusator erit (ei) qui uerbum dixerit ‘ hic est.’ For the use of ‘ uae’ absolute, cp. 203 | tot tibi uae misero uenient talesque ruinae. Am. III. 6. 101 huic ego uae demens narrabam fluminum amores. IBIs 445-446 et quae Pitthides fecit fraterque Medusae eueniant capiti uota sinistra tuo. So I think the couplet should be read. ‘ Pitthides’ is the conjecture of Salvagnius for ‘ Penthides’ and other corruptions in the MSS. ‘ Fraterque Medusae’ is the reading of the Galeanus, Phillipps and Paris MSS., and is preferable to the obscure ‘de fratre Medusae’ of the rest, accepted by Ellis. I may note in passing that this is one of the passages in which Ellis deserted the reading of the Galeanus in spite of the importance which he assigned to_ | it. In the readings 137, ‘tecum bella geram,’ and 468, ‘ Dexionesque,’ its ii ER ee excellence is undoubted. Pitthides is Theseus, the grandson of Pittheus, the father of his mother NOTES Aethra. dé xai N éyévETO, cated by and the dispatch For the II. 5. 6, IBIS Her and othe is Simon that I ar ‘ Sa in Thess vention « in the r Crannon nobilem ornandi | illum sor carmine Phaedru: Simonidi opere eu spatio cc cognatis ‘ Ste So Mic; p. 287. tion wot IT. 629, | habent f distress ( eripiunt persons { other a Ouintiliz e’ the lso has on has e from M anuel line, a valance ortuna e! tibi p- 295 ture of iterque and is y Ellis. asserted med to ue,’ its nother NOTES ON OVID’S IBIS, EX PONTO LIBRI, AND HALIEVTICA 257 Aethra. The ‘frater Medusae’ is Eurystheus: Apollodorus II. 4. 5, Oevérov Sé nal Neximaans tis TléXomos ’AXxwvdn nai Médovea, totepov Se cai Evpucbevs éyéveto, 65 Kai Muxnvadv é€Bacirevoev. The reference is to the curses impre- cated by Theseus against Hippolytus (Eur. H7pp. 887, foll. Cic. Off. III. 25, 94), and the evil wishes expressed against Hercules by Eurystheus when he dispatched him to perform his labours. Cf. Heroid. IX. 7 hoc uelit Eurystheus, uelit hoc germana Tonantis. For the expression ‘ uota facere’ cp. ex P. I. 6. 38, atque aliquis pendens in cruce uota facit. II. 5. 6, ‘ut faciam talia uota.’ IBIS 509-510 | lapsuramque domum subeas, ut sanguis Aleuae, stella Leoprepidae cum fuit aequa uiro. Here ‘ Leoprepidae’ is the certain emendation of Scaliger for ‘ leotepide’ and other corruptions in the MSS. It is universally accepted. Leoprepides is Simonides (Simonid. fr. 146 and 147). It is with the rest of the pentameter that I am concerned. ‘Sanguis Aleuae’ is Scopas, one of the Aleuadae, who lived at Crannon in Thessaly. The story of the escape of Simonides from death by the inter- vention of the Dioscuri, when the house of Scopas collapsed, burying its owner in the ruins, is told by Cicero, de Or. II. 352: ‘Dicunt enim, cum cenaret Crannone in Thessalia Simonides apud Scopam fortunatum hominem et nobilem cecinissetque id carmen, quod in eum scripsisset, in quo multa ornandi causa poetarum more in Castorem scripta et Pollucem fuissent, nimis illum sordide Simonidi dixisse se dimidium eius ei, quod pactus esset, pro illo carmine daturum; reliquum a suis Tyndaridis, quos aeque laudasset [cp. Phaedrus IV. 25, 9], peteret, siei uideretur. paulo post esse ferunt nuntiatum Simonidi, ut prodiret; iuuenis stare ad ianuam duo quosdam, qui eum magno opere euocarent ; surrexisse illum, prodisse, uidisse neminem: hoc interim spatio conclaue illud, ubi epularetur Scop 3, concidisse; ea ruina ipsum cum cognatis oppressum suis interisse.’ ‘Stella’ is usually explained as the constellation of the Dioscuri, Gemini. So Micyllus, followed by Salvagnius, Ellis, Ehwald, Jahresb. CIX. (1902), p. 287. This explanation is unsatisfactory, for the function of that constella- tion would appear to have been to promote unity among friends (Manil. II. 629, ‘quosque dabunt Chelae Geminique et Aquarius ortus, | unum pectus habent fideique immobile uinclum’), or to save from shipwreck mariners in distress (Hor. Carm. IV. 8. 31, ‘clarum Tyndaridae sidus ab infimis | quassas eripiunt aequoribus rates’). I do not find that its function was to rescue persons from collapsing houses. Moreover, in the story as told by Cicero and other authorities, Valerius Maximus (I. 8, extern. 7), Phaedrus (IV. 25), Quintilian (Inst. Or. XI. 2. 11), and as referred to by Callimachus in lines put S 258 S. G. OWEN into the mouth of Simonides (fr. 71, Schneider, p. 224), we are told nothing about the constellation of the Dioscuri. It is the Dioscuri themselves who appear miraculously in human form, as ‘iuuenes duo’ in Cicero, as ‘duo iuuenes’ in Valerius, as ‘duo iuuenes sparsi puluere’ in Phaedrus, as ‘duo inuenes equis aduecti’ in Quintilian. Equally unsatisfactory is the conjecture of Mr. Housman, which he has introduced into the text of the Corpus Poetarum Latinorum—‘ stella Iouis.’ He quotes in support of it Hor. Carm. II. 17. 22, ‘te Iouis impio | tutela Saturno refulgens | eripuit.’ There Horace speaks of the benign star of Jupiter in conflict with the malign star of Saturn, intervening to save Maecenas from a dangerous illness. But what has the star of Jupiter here to do with Simonides ? The text requires not emendation, but elucidation. ‘Stella’ I understand to mean Simonides’ own natal star, the star of his good fortune, which rose above the horizon at the moment of his birth. Some were born under an unlucky, some, as presumably Simonides, under a lucky star. I[uu. vii. 194 distat enim quae sidera te excipiant modo primos incipientem edere uagitus et adhuc a matre rubentem. si Fortuna uolet, fies de rhetore consul ; si uolet haec eadem, fiet de consule rhetor. Ventidius quid enim? quid Tullius? anne aliud quam sidus et occulti miranda potentia fati ? Cp. Manil. III. 58 foll. so of [bis Ovid says, 208 The influence of the natal star is often referred to: natus es infelix, ita di uoluere, nec ulla commoda nascenti stella leuisue fuit. Other passages are Heroid. viii. 88, ‘quodue mihi miserae sidus obesse querar ?’ Trist. V. 10. 45,‘0 duram Lachesin, quae tam graue sidus habenti { fila dedit uitae non breuiora meae!’ Prop. I. 6. 36, ‘ uiuere me duro sidere certus eris.’ II. 27. 4, ‘quae sit stella homini commoda quaeque mala.’ On the influence of opposing stars see Ovid Am. I. 8. 27 foll.; Persius V. 50. As regards ‘uiro’ this word is emphatic, ‘the illustrious man,’ as in Lucret. III. 371, ‘ Democriti quod sancta uiri sententia ponit’; as to which expression Mr. Housman appropriately says that this Lucretius can write, ‘because uirt means a right worthy man, like avdpos in Soph. Atax 817 and elsewhere (Classical Review XV. 264). I concur, and think that to alter ‘ uiro’ in this line of the bts is superfluous. For the substantive in apposition to the patronymic cp. [bis 543, ‘ puer Harpagides.’ mors faciet certe, ne sim, cum uenerit, exul: ne non peccarim mors quoque non faciet. So the MSS. except that for ‘ne non’ the Hamburgensis has ‘ nec non.’ a ae ~ Sy eres meee but tl amato: ness oO ex Pon been t Pontic I. 8. 40 (see E} Tl what | ‘Even destroy langua; But O negativ Co and ex. I belie by the is writt ex P., p some eé: the ‘ne ‘ sugges ‘ut’ th exile: e Ex So. thing ; who ‘duo ‘duo e has | He turno ter in rom a ides ? ‘stand h rose ler an 94 ad to: obesse abenti sidere ’ On as in which write, 17 and ‘uiro’ to the + al ee Ry corres seers NOTES ON OVID’S JBIS, EX PONTO LIBRI, AND HALIEVTICA 259 Since Heinsius and Bentley the couplet has been generally condemned. It is true that it repeats the thought of the preceding line poena potest demi, culpa perennis erit, and that the same thought is resumed 75-76 nec prius hi mentem stimuli quam uita relinquet : quique dolet, citius quam dolor | «+, cadet: but that it is superfluous is no argument against it. Ovid being ‘nimium amator ingenii sui’ (Quintil. I. 10. 88) is full of redundance, and the genuine- ness of the couplet is rightly asserted by Schreuders (Obseruationes in Outdu ex Ponto, Libros 1.-III. Lugd. Bat. 1895, p. 11). Groundless exception has been taken to the trisyllabic verb at the end of the pentameter, for in the Pontic Epistles Ovid does permit himself to end with trisyllabic verbs: I. 8. 40, ‘liceat’; III. 5. 40, ‘ recitent’; 6. 46, ‘uideor’; IV. 9. 26, ‘ tegeret ’ (see Ehwald, Knit. Bettrage, p. 11). The real crux is ‘ne non,’ which makes Ovid say the direct opposite of what he means: ‘ Even death will not cause me not to have not sinned ’= ‘Even death will not cause me to have sinned,’ for the doubled negatives destroy one another. Schreuders defends the expression as a negligence of language, and quotes what Ovid says ex P. III. 9. 17 saepe aliquod uerbum cupiens mutare reliqui iudicium uires destituuntque meum. But Ovid cannot have been so negligent as to have failed to distinguish a negative from an affirmative. Comparing A. A. II. 101 non facient, ut uiuat amor, Medeides herbae, and ex P. II. 2. 24 non tamen efficies ut timeare mihi, I believe that the genuine reading is ‘ut non peccarim.’ This is supported by the fact that in the Munich MS. 19476, one of the three best MSS., ‘ut’ is written by the second hand above ‘ne’ (Korn, De codd. duobus carm. Ouid. ex P., p. 6). The variant may have found its way in through collation with some early MS. The substitution of ‘ne’ for ‘ut’ in our MSS. was due to the ‘ne’ in the preceding line. This is what Havet calls an error due to ‘suggestion d’un mot antérieur’ (Manuel de Critique Verbale, § 496). Reading ‘ut’ the meaning is: ‘ Assuredly death, when it comes, will set an end to my exile: even death will not obliterate my offence.’ Ex PonTo I. 7. 21-22 quis se Caesaribus notus non fingit amicum ? da ueniam fasso: tu mihi Cesar eras. So the Hamburgensis, and some MSS., while the Munich MSS. and others, have ‘eris.’ ‘Eras’ is preferable, meaning ‘at the time when I wrote the Ss. G. OWEN 260 compliments that gave you offence, you were my Caesar.’ This offence is mentioned line 17 me miserum, si tu uerbis offenderis istis, nosque negas ulla parte fuisse tuos! The flattery by which Ovid speaks of Messalinus as held by him in equal reverence with his sovereign is paralleled by I. 9. 49, where, addressing Celsus, he says that his patron Maximus Cotta was revered by him as if he were a god (‘ quem tu pro numine uiuus habebas’). Ex Ponto I. 7. 65-66 idque facis, quantumque licet meminisse, solebas officii + causa pluribus esse tf dari. So the Hamburgensis and Munich MS. 384. For ‘ dari’ the other MSS. have ‘ dati.’ The pentameter is so obviously corrupt that Korn rejected it altogether from the text. Various unsuccessful attempts at emendation have been made. The point is that Messalinus is advised to beseech the gods that he may confer more benefits than he has received, which in fact is what he does. To confer rather than to return a favour is what matters. For the unmetrical ‘causa’ a few MSS. and some editions have ‘ causam,’ which cannot grammatically be construed as predicate after ‘solebas.’ And yet such a predicate is wanted. This difficulty is removed by Dr. Purser’s neat emendation, ‘ causae’ (he reads ‘ causae—dati’; see Keene, ad luc.). Ehwald (Jahresbertcht LXXX. p. 113) objects to ‘ causae,’ thus used as predicative dative as unovidian. It certainly cannot be defended by Am. II. 6. 31 nux erat esca tibi causaeque papauera somni, doubtfully quoted by Roby, L. Gr. II. p. xlii, for in that line ‘causae’ must be the same case as ‘esca,’ viz. nominative plural. Parallel uses of ‘ causae’ as predicative dative from other writers are quoted by Roby. But as Ovid can say ex P. II. 1. 4 iam minus hic odio est, quam fuit ante, locus it is clear that he could have used ‘ causae’ as predicative dative. Accepting therefore Dr. Purser’s ‘ causae,’ I propose to read officil causae pluribus esse datis, ‘You are accustomed to be the occasion for devotion by reason of your conferring more gifts than you receive.’ The thought goes back to line 64 ut des quam reddas plura precare deos. As Cicero says, Off. I. § 48, there are ‘duo genera liberalitatis, unum dandi beneficii, alterum reddendi.’ I understand ‘datis’ as a substantive, like Met. VI. 462, ‘nec non ingentibus ipsam sollicitare datis.’ ll NO’ ‘ triflir saxa, conta It ise illegit read, MSS. I MS. 1 Kr. B free fi evil is three hunte clutch hound N illustr: I] TI attemp if left strugg se is qual Isus, . god MSS. ether nade. onfer onfer usa ’ lly be nted. reads II3) ainly must usae ’ d can ea NOTES ON OVID’S IBIS, EX PONTO LIBRI, AND HALIEVTICA 261 Ex Ponto II. 2. 31-38 tuta petant alii. fortuna miserrima tuta est: nam timor euentu deterioris abest. qui rapitur porrigit spinas dura quae saxa accipitremque timens pennis trepidantibus ales audet ad humanos fessa uenire sinus, nec se uicino dubitat committere tecto, quae fugit infestos territa cerua canes. So the passage stands in our best MS., the Hamburgensis, except for the trifling error, ‘infesto’ for ‘infestos.’ The mutilated couplet, ‘qui rapitur— saxa,’ is omitted altogether in the two Munich MSS., though Munich 384 contains an obviously wrong supplement, added by a later hand qui rapitur fatis quid praeter fata requirat ? saepe creat molles aspera spina rosas. It is evident that the archetype of these three MSS. was here defaced, and partly illegible. The scribe of the Hamburgensis copied faithfully what he could read, leaving the couplet without metre or sense: the scribes of the Munich MSS. omitted altogether what they could not understand. In line 32, ‘euentu,’ the reading of the Hamburgensis and one Munich MS. must be restored in place of the vulgate reading ‘euentus’ (Ehwald, Kr. B., p. 16). Ovid says: ‘The most miserable condition is secure, for it is free from the fear of one worse in its result.” The proposition that the lesser evil is preferable in the face of a still worse contingency is next enforced by three illustrations: (1) That contained in the mutilated couplet ; (2) the hunted bird, which prefers to fall into human hands rather than into the clutches of the hawk; (3) the stag which, rather than be caught by the hounds, flees into a house. Now it is characteristic of Ovid to exemplify a proposition by three illustrations. Thus, ex P. II. 5. 61 rusticus agricolam, miles fera bella gerentem, rectorem dubiae nauita puppis amat. IT. 7. 9 qui semel est laesus fallaci piscis ab hamo, omnibus unca cibis aera subesse putat. saepe canem longe uisum fugit agna lupumque credit, et ipsa suam nescia uitat opem. membra reformidant mollem quoque saucia tactum, uanaque sollicitis incutit umbra metum. Thus in Am. I. 2.9 the statement that it is best for one in love not to attempt resistance is exemplified by three illustrations: fire, which dies out if left undisturbed ; oxen, which when tamed are beaten less than those which struggle before they have been tamed; and the unbroken horse, whose mouth 262 Ss. G. OWEN is galled more by the bit than when he has been broken. Thus in A. A. II. 177 the statement that woman’s heart must be won by gentleness, not by com- pulsion, is exemplified by three illustrations: the crooked branch, which may be straightened by patience, but is broken by the use of force; the swimmer, who succeeds by swimming with the stream; and wild animals, which are tamed by patience, not by force. Thus in the Medicamina faciei 37 the statement that simples are useless for the complexion is exemplified by three illustrations: spells cannot split snakes, nor make streams flow backward, nor bring down the moon. I conclude therefore that the mutilated couplet cannot be dealt with by mere omission. It clearly contained the first in a series of three illustrations. Its missing portions are excellently restored in the version of it found in the Leipzig and Erfurt MSS. qui rapitur spumante salo, sua brachia tendens porrigit ad spinas duraque saxa manus. For ‘salo’ the Erfurt MS. has ‘freto.’ If this be accepted, the first of Ovid’s three illustrations is a man carried away by the surge, who prefers to clutch at thorns and rough rocks rather than be drowned. I suspect that this couplet was preserved in some very ancient MS., which was collated by the scribes of the Leipzig and Erfurt MSS., who thence supplied it into their text. It is known that such collation of ancient MSS. did take place in the scriptoria. The preservation of the supplementary lines at Juvenal vi. 365 is due no doubt to this factor. The lines are preserved in the Oxford MS. alone, which was written in the Beneventan script, probably at Monte Cassino. It is probable that in that library some very early MS. of Juvenal was preserved containing these lines, which Leo plausibly argues formed part of the poet’s first edition, and were suppressed in the second. The Heroides of Ovid offer further confirmation of this hypothesis. It has been pointed out by Vahlen (Ueber die Anfange der Heroitden des Ovid) that whereas Heroides I, II, III, IV, XIII, XIV all begin with a couplet in which the name of the writer of the epistle is given, V, VI, VII, VIII, IX, X, XI, XII begin abruptly without mention of the writer. There are three chief MSS. of the Heroides; Parisinus, sec. xi, Guelferbytanus, sec. xii, both Frankish MSS., written in Caroline minuscules, and the Eton MS., sec. xi, written in the Beneventan script, probably at Monte Cassino. It is an extraordinary fact that the desiderated couplets, containing the writer’s name, are supplied at the beginning of V, VI, VII in the Eton MS. I suggest that the scribe of the Eton MS. had access to some very early MS., such as may well have been preserved in a South Italian library, by collation with which he supplied these couplets.!' It should also be mentioned that the missing couplets in VIII, IX, X, XI, XII are supplied in certain later MSS., which may have got them from a similar source. If these supplementary couplets be accepted, we 1 I base this suggestion on the valuable essay p. 895. of P. von Winterfeld, Gétt. gel. Anzeige, 1899, rrr se ee - FE has ‘ 5 that c theref furthe Mart. staine in the infest: Senec Thou; the er if the It less, a epithe have tortuo was f and | by M V. 10 1 Th Sed1lm.; - 177 com- may mer, h are 7 the three 1, nor th by tions. n the Yvid’s ‘ch at which hence MSS. lines ed in bably {S. of rgues The been ereas name begin f the MSS., n the y fact it the f the been yplied ts in e got d, we NOTES ON OVID’S JBIS, EX PONTO LIBRI, AND HALIEVTICA 263 have in the Heroides complete uniformity of structure, which is what we should expect to find in these elaborate poetical ‘ suasoriae,’ which are com- posed methodically according to the strict rules of ancient rhetoric. It was easy for a couplet to fall out at the beginning of a poem: it may have been omitted, for instance, with a view to rubrication, and then have never been filled in: it may have been omitted at the beginning of V because the last line of IV and the first line of V both begin with ‘Perlegis,’ and at the beginning of XI, because the last line of X begins with ‘si prius’ and the first line of XI with ‘siqua’: it may have been omitted at the beginning of VI because the supplementary couplet begins with ‘ Lemnias,’ and VI. 1 with ‘ Litora.’? Ex Ponto II. 8. 53 Caesaris aduentu tuta gladiator harena exit, et auxilium non leue uulnus habet. For ‘aduentu’ the Hamburgensis has ‘aduentum,’ and for ‘ tuta’ it has ‘ tota.’ Schreuders objects to ‘aduentu’ because he says that it is not recorded that on the arrival of Augustus at the podium gladiators were spared. He therefore accepts Riese’s conjecture, ‘ad nutum,’ followed by Ehwald, who further proposes to alter ‘tuta’ to ‘ tincta,’ ‘the blood-stained sand,’ comparing Mart. II. 75. 6, ‘sanguineam humum.’ But ‘tincta’ could not mean ‘blood- stained ’’ unless ‘sanguine’ were added. I think the passage is to be explained in the light of the statement of Suetonius, Iul. 26, ‘ gladiatores notos, sicubi infestis spectatoribus dimicarent, ni rapiendos reseruandosque mandabat,’ and Seneca in Anth. Lat. P.L.M. (Baehrens) IV. 25, 27 (p. 66) sperat et in saeua uictus gladiator harena, sit licet infesto pollice turba minax. Though the fate of the defeated gladiator depended generally on the people, the emperor might specially intervene to save him. This would only happen if the emperor was personally present. It remains to consider the readings ‘tuta’ and ‘tota.’ ‘Tota’ is point- less, and will not do. ‘Tuta’ can only be explained as a case of transferred epithet, as Fabricius explained it, ‘ pro ipse gladiator tutus.’ But as it would have been simpler to say ‘tutus gladiator’ outright, this seems peculiarly tortuous. Therefore I propose to read ‘tuto,’ the termination of which word was falsely assimilated to ‘harena’ by a species of error which is frequent, and has been treated by Havet, Manuel de Critique Verbale, ch. xxiv, and by Mr. Hall, Companion, p. 174. Ovid uses ‘tuto’ in Am, II. 14. 43; Tmst. V. 10. 39. 1 The supplementary couplets are given in and in Ehwald’s Teubner text. Sedlmayer’s critical edition, in Palmer’s edition, 264 S. G. OWEN Ex Ponto III. 3. 41 at non Chionides Eumolpus in Orphea talis. Eumolpus was son of Posidon and Chione (Apollodor. III. 15. 4). Though in Chione the first syllable is short (Met. XI. 301), in Chionides it is length- ened through stress, according to the principle which I have illustrated, Classical Quarterly, VIII. 29. Ex Ponto III. 4. 89 inrita uotorum non sunt praesagia uatum. Ehwald (Kr. Bettr. p. 48) justly defends ‘ uotorum’ against alterations such as ‘motorum’ (Heinsius) and others which will be found recorded in my note in the Corpus Poetarum Latinorum. ‘ The prophecies,’ says Ovid, ‘of poets are not vain in respect of their prayers,’ the things for which they pray. It should be pointed out that just as the words ‘ praesagia’ and ‘uatum’ are picked up lower down in ‘ praedico’ and ‘ uaticinor,’ line 94 haec duce praedico uaticinorque deo, so ‘uotorum’ is picked up in ‘ precor,’ line 114: uerba, precor, celeri nostra probate fide. Ex Ponto III. 4. 107-108 squalidus inmissos fracta sub harundine crines Rhenus et infectas sanguine portet aquas. So the text stands in some MSS. The Hamburgensis, which ends at III. 2. 67, is here wanting. For ‘portet’ the Munich MS. 384 has ‘ potat,’ the Munich MS. 19476 ‘portat’; a few have ‘portet,’ accepted by Korn. Ehwald (Kr. B. p. 82), after successfully refuting the absurd conjecture of Merkel, ‘ porgat,’ objects to ‘ portet’ because he says that the verb required must fit both ‘crines’ and ‘aquas,’ which this does not. He proposes ‘ mon- stret.’ A subjunctive is of course required. But I think that ‘ portet’ is right. The image of the river-god exhibited in the triumph (Plin. N. H. V. 37; Tac. A. II. 41, ‘uecta spolia, captiui, simulacra montium, fluminum, proeliorum’), must have carried a vessel containing blood-stained water. Thus he is said to carry both his hair and this vessel. Ex Ponto III. 5. 49 hac ubi perueni nulli cernendus in urbem. The Vulgate reading ‘nulli’ is preferable to that of the Munich MSS. ‘nullis,’ since the plural of ‘nullus’ as a substantive is very rare (Reisig, Vorlesung. lat. Sprachw. 111", p. 73), though Ovid has Met. VIII. 172, ‘ nullis iterata priorum ianua.’ NOTI Se ship. by the couplet I prop providi like ‘u There in Ovi Fast. 1 ‘ Certa miss w to miss In ‘tum ¢ favour Jupiter the tim to all t! explanz wrong Jupiter joins to In to the singula: cum pl Liv. X2 ‘Thia cv Ex Fo hough ength- trated, rations in my id, ‘of y pray. n’ are nds at potat,’ Korn. ‘ure of quired ‘mon- s right. . ‘Tac. rum’), said to MSS. Reisig, ‘ nullis NOTES ON OVID’S IBIS, EX PONTO LIBRI, AND HALIEVTICA 265 Ex Ponto IV. 4. 27-34 cernere iam uideor rumpi paene atria turba, et populum laedi deficiente loco, templaque Tarpeiae primum tibi sedis adiri, et fieri faciles in tua uota deos, colla boues niueos certae praebere securi, quos aluit campis herba Falisca suis, cumque deos omnes, tunc hos inpensius, aequos esse tibi cupias, cum loue Caesar erunt. Sextus Pompeius is addressed, who is entering at Rome on his consul- ship. In line 31 the asyndeton at ‘colla,’ for which ‘ collaque’ is demanded by the context, is so awkward that Ehwald (Kr. B. p. 64) suggests that a couplet has fallen out after line 32, and that a new period begins at line 31. I propose as a less drastic remedy to read ‘cerno’ for ‘ certae’ in line 31, thus providing a verb to introduce ‘praebere.’ ‘Cerno,’ ‘I see in imagination,’ is like ‘ uideo,’ ‘I see in imagination,’ III. 5. 47 te nisi momentis uideo paene omnibus absens. There is little point in the epithet ‘ certae,’ which adjective is not elsewhere in Ovid applied to ‘securis.. In Met. XV. 126, ‘percussit colla securi.’ Fast. 1V. 415, ‘apta iugo ceruix non est ferienda securi’ there is no epithet. ‘Certa’ is common as applied to ‘hasta’ and ‘sagitta’ because one might miss with those weapons, but the sacrificial slaughterer would not be expected to miss his victim. In line 33 I accept ‘tunc hos’ from Korn’s Paris MS. 7993 in place of ‘tum quos,’ the reading of the rest, translating, ‘And since you desire the favour of all the gods, and at that moment the favour of these two especially, Jupiter and Caesar will be present.’ ‘ Tunc’ refers to the prayers offered at the time of the sacrifice of the oxen (line 31), as opposed to the prayers offered to all the gods (‘ deos omnes’) in line 30 (‘ faciles in tua uota deos’). In this explanation put forward by Ehwald (Kr. B. p. 64) I concur, but I think he is wrong in understanding ‘hos’ of the Capitoline deities: it refers rather to Jupiter and Caesar, the two chief gods in Ovid’s parlance, whom he often joins together, e.g. Trist. II. 39. In the last line, ‘ erunt,’ the reading of the Munich MSS,., is preferable to the Vulgate ‘erit,’ because it is idiomatic to use a plural verb with a singular nominative joined to a substantive with ‘cum.’ Met. IV. 734, ‘ litora cum plausu clamor superasque deorum inpleuere domos.’ So in_ prose, Liv. XXI. 60, ‘ipse dux cum aliquot principibus capiuntur.’ Cp. Fast. IV. 54, ‘Ilia cum Lauso de Numitore sati.’ Ex Ponto IV. 8. 85 clausaque si misero patria est, ut ponar in ullo, qui minus Ausonia distet ab urbe, loco. For ‘distet’ the Munich and some other MSS. have ‘distat,’ which 266 S. G. OWEN Ehwald (Kr. B. p: 64) proposes to restore. But the subjunctive is required by attraction after ‘ ponar’: cp. I. 7. 13: ‘nos habeat regio nec pomo feta nec uuis, et cuius nullum cesset ab hoste latus.’ So IV. 10.71. Indicatives and subjunctives are so constantly confused in Ovid’s MSS. that judgement in each case must decide what is right. See my Tristia, Prolegg. p. Ixxvii. Ex Ponto IV. 16. This interesting poem, in which Ovid enumerates the contemporary poets, is fully commented on by Schanz, Rém. lit. II. 1, p. 360 foll. The poem consists of two propositions: (1) ‘ Before my exile, while | was a poet recognised by society, I had reputation’ (lines 1-4). (2) ‘And at the time when there was Marsus and all the rest of the galaxy of poets enumerated, my poetry was celebrated and read, though so many distinguished authors were writing besides myself. Therefore, now that I am fallen, I should not be defamed ’ (5-52). The construction of 45-46 dicere si fas est, claro mea nomine Musa atque inter tantos quae legeretur, erat. is ‘ Mea Musa erat claro nomine atque erat Musa quae inter tantos legeretur.’ The punctuation of the poem is unsatisfactory in the editions. At line 4 there should be a full stop at ‘erat.’ At line 5 begins a long series of protases, each contained in a couplet, or sometimes in a single line. There should be a semicolon at the end of each couplet, till lines 41-44, ‘ te tamen—dedit,’ which should be placed in a parenthesis, since the verb changes from past, ‘cum foret,’ etc., to present, ‘non ausim silere.’ ‘Among the many poets,’ says Ovid, ‘I cannot forget to mention you, Cotta.’ The parenthesis is suggested by ‘adpellandorum’ in line 4o. At line 45 the apodosis is introduced, and the full stop comes at ‘ erat.’ There is much difficulty as to 33-34 Tityron antiquas f passerque rediret ad herbas, aptaque uenanti Grattius arma daret. Line 33 is corrupt: some of the solutions proposed, none in my opinion satisfactory, are given by Schanz, Rém. Lit., l.c. p. 355. The whole couplet appears to allude to the poet Grattius, the pentameter to his Cynegetica, in direct reference to the words of Grattius, line 22 lusus carmine et arma dabo et uenandi persequar artis. The corrupt hexameter is by many, as by Schanz, supposed to refer to a lost Bucolica by Grattius. This seems to me to be the most probable hypothesis. Accepting it, I propose to read Tityron antiquas pastorem exciret ad herbas. ‘Cum’ is understood from line 31 two lines above. Grattius summoned NOTES Tityrus t herbas’), This inscribed its title 1 ‘Versus ‘Halieuti Halteuttc the exilec to and cit XXXII. The opin of a late) made up incredible Ovid, an poem. ~ irrelevant and has A bhandlu I que ‘In « es ist da Ovid das Tod an d der vomisc ‘Le a discute: Zingerle | It m that, as it nature of the perfec probably In t attention ‘Those \ must eitl feature w 1 ¢ Mihi 1 didit pisciu pinion ouplet ca, in a lost thesis. noned NOTES ON OVID’S IBIS, EX PONTO LIBRI, AND HALIEVTICA 267 Tityrus to take his place once more on the old world pastures (‘antiquas herbas’), pastures sung by poets of long ago. For ‘antiquas’ cp. A. A. III. 405 cura deum fuerunt olim regumque poetae, praemiaque antiqui magna tulere chori. HALIEVTICA. This short didactic poem on fishes, cited by Pliny as Halieuticon,' is inscribed ‘ Versus Ouidi de piscibus et feris’ in the Vienna MS. _ Probably its title was Halteuticon Liber, as that of Grattius is Cynegeticon Liber, and ‘Versus Ouidi de piscibus et feris’ may conceal an alternative title, thus, ‘Halieuticon liber siue de piscibus et feris.’ It cannot be questioned that the Halteutica is the genuine work of Ovid, a trifle with the composition of which the exiled poet amused himself at the close of his life. The precise references to and citations from the poem as we have it which are found in Pliny (N. H. XXXII. § 11-13 and § 152-153) preclude any doubt as to its authenticity. The opinions that have been advanced to the contrary, that it is the work of a later poet which Pliny mistook for that of Ovid, or that it is a forgery made up from Pliny’s quotations, cannot be seriously maintained. It is incredible that Pliny could have mistaken the work of another for that of Ovid, and no forger could have been ingenious enough to have forged this poem. The sceptical view has been supported with much minute, often irrelevant learning, by Birt, De Halteuticis Ouidio poetae falso adscriptis (1878), and has been refuted, amongst others, by A. Zingerle, Kleine philologische A bhandlungen II. pp. 1-44, and F. Vollmer, Rhein. Mus. LV. pp. 528-530. I quote two recent judgements on the matter by two consummate critics : ‘In dem Zustand, in dem wir die Schrift haben, lag sie auch Plinius vor; es ist daher ein Zweifel an ihrer Echtheit vollstandig ausgeschlossen. Da Ovid das Gedicht in der letzten Zeit seines Exils verfasste, wird er durch den Tod an der Vollendung desselben verhindert worden sein.’ Schanz, Geschichte der romischen Litteratur, 11. 1, p. 338. ‘Le reste n’ajoute rien aux titres d’Ovide et ne vaut pas que l’on s’attarde a discuter l’opinion de Birt contre l’authenticité, opinion réfutée d’ailleurs par Zingerle et par Bahrens.’ Plessis, La Poésie Latine, p. 461. It must be assumed that we have in this poem the genuine work of Ovid, but that, as it was written at the end of his life and not finally completed, it is of the nature of a rough draught. It is therefore likely to contain things at variance with the perfection found in the best works of the great artist, and and which he would probably have removed, if he had lived. And this in fact is what we do find. In the Classical Quarterly, I. pp. 275 sqq.. Mr. Housman has drawn attention to certain irregularities in the prosody of the poem. He says: ‘Those who believe that Ovid wrote the halieutica, or that Pliny read it, must either renounce their beliefs or else find means to rid the poem of a feature which renders both alike incredible: its false quantities.’ ‘*Mihi uidentur mira et quae Ouidius pro- Halieuticon inscribitur.’ Plin. N. H. XXXII. didit piscium ingenia in eo uolumine quod § 11. 268 S. G. OWEN As I believe that Ovid wrote the Halteutica, I propose to consider these irregularities. First, there is the scansion of ‘ milui’ as a disyllable, line 95 hippuri ceteres et nigro tergore milui, of which, says Mr. Housman, ‘the first certain example is some forty years later than Pliny’s death and a whole century later than Ovid’s: Iuu. ix. 55, ‘tot miluos intra tua pascua lassas.’ Synizesis is indeed used seldom by Ovid, and is generally of a normal character as ‘conibid’ (Met. VI. 428), ‘aired’ (H. vi. 49). More unusual are ‘Antiim’ (Met. XV. 718), ‘arcuatd’ (Met. XI. 590). It is probable that this scansion of ‘ milui’ is one of those rough- nesses which Ovid would have removed if he had lived. His scansion of ‘arcuato’ shows that such synizesis was possible to him: and Ovid’s scansion ‘milui’ may have influenced Juvenal. The fact that the word ‘ larua,’ trisyllabic in Plautus, became a disyllable in classical Latin, and that ‘ tenuis’ is some- times three and sometimes two syllables shows that Ovid would have found no difficulty in scanning ‘ miluus’ as a disyllable (see Lindsay, Lat. Language, p. 46). He may well have started it. Things must begin somewhere. Secondly, there is the scansion of ‘ anthias’ as a dactyl in line 46 anthias his, tergo quae non uidet, utitur armis. I agree with Mr. Housman that the word cannot be scanned as a spondee, since such synizesis in dealing with a Greek word was for a Roman poet impossible. I disagree with him when he denies that it can be a dactyl. Everybody knows that in Greek it is acretic. But Ovid, located in exile on the Black Sea, when writing of fishes, may well have drawn his knowledge from local fishermen, Greek settlers who spoke the local vulgar Greek dialect, by no means the same as classical Greek. The importance of the fish trade of the Black Sea made its fishermen not negligible authorities.1 The poet may well have preserved the pronunciation of his informants. The shortening of long vowels is an obvious indication of the vulgar dialect (Grandgent, Vulgar Latin, p. 75). Even in classical Latin the long -as and -es of Greek substantives tended to become &@; thus ‘ Perdiccas’ and ‘sycophantes’ became ‘ Perdicca’ and ‘sycophanta’ (Roby, L. Gr., i. p. 163). It is probable enough that av@ias became shortened in its final syllable in the vulgar dialect. In fact, even Quicherat’s suggestion is not impossible: ‘ Velim esse Anthias, ados, nisi obsit graeca uox’ (Thesaurus Poeticus, p.77). Ovid does complain that he lives in an environment of alien tongues, Trist. III. 14. 47 Threicio Scythicoque fere circumsonor ore ; and from ex Ponto III. 9. 17 saepe aliquod uerbum cupiens mutare reliqui, iudicium uires destituuntque meum. saepe piget (quid enim dubitem tibi uera fateri ?) corrigere et longi ferre laboris onus, 1 Minns, Scythians and Greeks, p. 440. Plin., N. H. XXXII. § 152, ‘his adiciemus ab Ouidio posita animalia, quae apud neminem alium reperi- untur, sed fortassis in Ponto nascentia, ubi id uolumen supremis suis temporibus inchoauit.’ eee NOTES ‘ it is evidel ex P. I. 5. Thirc in line 10: the name fr.z.1;3 € ‘ mormyr the name woppupon - of the ab: Black Se I am sail different ° The spell I cor Halieutica of these : them ; fo1 find in tk neglected came to s BaXavetov (Strab.) < to the vu origin has Tpiyes in as a trib “three”’ pronunciz HAL Birt fluous,’ a anaphora (Rh. Mus by readin paring P: 1 Lindsay etymology ¢ er these ty years . 1X. 55, y Ovid, *‘aitired’ »’ (Met, - rough- ision of cansion syllabic 3 some- e found nguage, pondee, in poet dactyl. xile on »wledge dialect, h trade ne poet vulgar ne long as’ and p. 163). 2 in the ‘Velim id does - 47 a, ubi id pauit.’ NOTES ON OVID’S IBIS, EX PONTO LIBRI, AND HALIEVTICA 269 it is evident that towards the end of his life he wrote less carefully. See also ex P. 1. 5. 3 foll., and Zingerle, K/. ph. A bhandl. I1., p. 41. Thirdly, there is the strange false quantity ‘ pompile’ as an antibacchius in line 101 qui semper spumas sequeris, pompile, nitentes, the name of which fish, woumidos, is a dactyl in Greek poets (Erinna, fr.1.1; Oppian, Hal. I. 186), and there is the equally strange antibacchius ‘mormyres’ in II0 et rarus faber et pictae mormyres et auri, the name of which fish in Greek is popptpos, a dactyl, Anth. Pal. VI. 304. 4, poppupov 7 KixAnv 7 odpov } cpapida. In both these cases I see the influence of the abnormal pronunciation of the semi-barbarous Greek fishermen of the Black Sea. Such influences are always at work in language. If I say that I am sailing for the port of Genoa I pronounce the name of that city in a different way to that which I use in ordering Genoa cake of a pastrycook. The spell of the pastrycook’s tongue is then upon me. I conclude therefore that the evidence of Pliny for the authenticity of the Halieutica is overwhelming. It cannot be declared to be unovidian on account of these metrical peculiarities. They are there; there is more than one of them; for this there must be a common cause. This cause I am content to find in the influence of vulgar Greek pronunciation, a factor which cannot be neglected in dealing with Latin literature. It came upon Latin early, and came to stay. Thus a@yxipa appears in Latin as anchéra, Bpayiov as brachium, Baraveiov as balinéum, balnéum, wrateia as platéa, Xnveves (Polyb.) Otdoxwves (Strab.) as Séndnes, Vascénes. Thus the Plautine Accheruns, Acchilles are due to the vulgar Greek pronunciation of y as k-kh; and ‘the word tricae, whose origin has been traced to South Italy . . . seems to be nothing but the Greek tpiyes in a Latinized form.’* Thus the coin Philippus ‘is always scanned as a tribrach in Plautus, just as we tolerate a mispronunciation of the word “three”? in “threepence.” But the name Philippus ... has the “literary ”’ pronunciation.’ # HALIEVT. 2 uitulus sic namque minatur qui nondum gerit in tenera iam cornua fronte. sic dammae fugiunt. Birt (Halieut., p. 13) objects to ‘ sic namque’ because ‘namque’ is super- fluous,’ and ‘sic’ without it would be more natural, as producing the figure anaphora, ‘ uitulus sic—sic dammae—sic scorpius—leuis sic ales,’ and Vollmer (Rh. Mus. LV. 528), following the same line of criticism, removes the difficulty by reading ‘manca’ for ‘namque’ (he prints ‘ manca’ in his edition), com- paring Prudent. Peristeph. II. 231, ‘errorque mancum claudicat,’ and expres- 1 Lindsay, Lat. Language, p. 58. Another given in Walde, Latein. etymol. Worterbuch, p. 791. etymology of ‘ tricae’ besides that of Lindsay is 2 Lindsay, Captivi of Plautus (1900), p. 24. 270 S. G. OWEN sions like Copa 3, ‘saltat lasciua.’ It is a far cry from Ovid to Prudentius, Ovid uses the uncommon word ‘ mancus’ once only, Fast. III. 826 et licet antiquo manibus conlatus Epeo sit prior, irata Pallade mancus erit. Though a carpenter be more skilful than Epeus, if Pallas be unfavourable, he will lose the use of his hands, be powerless. The word is there used in its strict sense, with relation to the loss of a limb, the hands. But the calf that has not yet grown any horns could hardly be said to have lost them. Therefore I think ‘manca minatur’ is inappropriate, and could be used only of a calf that had been divested of its horns. ‘Namque’ on the other hand, is appropriate as introducing the series of illustrations of the proposition that the world has received its law. It has distributed arms among all creatures, and made them careful of self-preserva- tion. For the calf which as yet has grown no horns makes believe to strike. ‘Namque’ is placed third word as Met. II. 474, ‘adimam tibi namque figuram.’ XIV. 312, ‘cum duce namque meo Circe dum sola moratur.’ Verg. Aen. I. 444, Kiihner, L. Gv., Il. 715. ‘Namque’ is the certain conjecture of Sannazarius, which is usually accepted, for ‘manuq’; the reading of the Vienna MS. The corruption ‘manuque’ is an instance of multiplication of letters, a common source of error in MSS., possibly due to the mispronuncia- tion of the scribe. The same error appears in the Vienna MS. at 26, ‘ atque atque’ for ‘atque’; 49, ‘et cetera’ for ‘cetera’; 60, ‘denuntiate’ for ‘denuntiat’; 70, ‘aurate’ for ‘aurae’; 114, ‘merolateque’ for ‘ merulaeque’; 124, ‘soleate’ for ‘soleae,’ which examples I take from Haupt’s apparatus Criticus (Halieut., Lips., 1838). The same error appears often in the MSS. of Catullus, as I have pointed out in my note on Catull. 116. 7, as 63. 5, ‘ iletas’ for ‘ili’; 64. 3, ‘ Fasidicos’ for ‘ Phasidos’; 66. 6, ‘ guioclero’ for ‘ guro.’ The calf attempting to butt with hornless brow is a commonplace. Ov. Am. III. 13. 15 et uituli nondum metuenda fronte minaces. Iuuen. xii. 7 quippe ferox uitulus qui uexat nascenti robora cornu. Mart. III. 58. 11 uitulusque inermi fronte prurit in pugnam. VI. 38. 8 sic uitulus molli proelia fronte cupit. Claud. IV., Cons. Honor. 383 sic pascua paruus uindicat et necdum firmatis cornibus audax iam regit armentum uitulus. But though the calf is said to butt, and to butt having no horns, he is not said to butt ineffectively. For Ovid’s ‘nondum metuenda fronte’ is only another NOTES way of Vollmer Bir ‘iam’ \ amplius, to ‘ tene and ‘m« ‘ jamque Hal In | *quomq tion as | destruct Hist. I. : und Stil HA! Mr. conjectu ‘Capto’ timorer gaudia ’ eagernes which r which m Ha) Ovi because ‘cupiunt means ‘ laetiore, dentius. ourable, used in the calf t them. ed only e series It has reserva- trike. namque Verg. cture of of the ation of ynuncia- , ‘ atque te’ for aeque’; yparatus MSS. of ‘ iletas’ 0.” onplace. not said another NOTES ON OVID'S IBIS, EX PONTO LIBRI, AND HALIEVTICA 271 way of putting Martial’s ‘inermi fronte.’ For these reasons I cannot accept Vollmer’s ‘ manca.’ Birt further objects to ‘iam,’ which he takes with ‘nondum’ because ‘iam’ when compounded with an adverb of negation is equivalent to ‘ non amplius,’ which is ‘contrarium uoci nondum.’ This is futile, for ‘iam’ belongs to ‘tenera,’ ‘his brow now tender’; it means the same as Martial’s ‘ inermi’ and ‘molli fronte.’ ‘lam’ is often joined with an adjective, as Iuu. iil. 206, ‘iamque uetus cista’; xi. 127, ‘iam nimios capitique graues.’ HALIEVT. 53-57 inpiger ecce leo uenantum sternere pergit agmina et aduersis infert sua pectora telis, quoque uenit fidens magis et sublatior ardet concussitque toros et uiribus addidit iram, procidit atque suo properat sibi robore letum. In line 55 ‘quo’ means ‘in proportion as.’ It should not be altered into ‘quomque,’ as by Schenkl, followed by Vollmer. The meaning is ‘in propor- tion as he advances with more confidence and rages with more spirit, he meets destruction sooner.’ For the omission of ‘eo’ answering to ‘ quo,’ cp. Tac. Hist. I. 14, ‘quo suspectior sollicitis, adoptanti placebat’ (see Drager, Syntax und Stil des Tacitus, § 181). HALIEVT. 65 et capto fugiens ceruus sine fine timore. Mr. Edwards, in the Corpus Poetarum Latinorum, prints ‘cauto,’ the conjecture of Heinsius, without remark, as though it were the MS. reading. ‘Capto’ is the reading of the Vienna MS., and should not be altered. ‘ Capere timorem ’ means ‘to conceive fear,’ like ‘ capere spem ’ (Fast. II. 334), ‘ capere gaudia’ (Trist. V. 8. 21). It is strange that so learned a scholar as Birt, in his eagerness to defend it, should cite in its support ‘capere iram’ (Met. VI. 610), which means ‘to restrain one’s rage,’ and ‘capere animos’ (A. A. IJI. 557), which means ‘ to captivate the affections,’ phrases which are quite irrelevant. HALIEVT. 67 nam capiunt animis palmam gaudentque triumpho. Ovid is speaking of the noble instinct of horses. Birt objects to ‘ capiunt ’ because ‘animis nemo facile coronam prehenderit.’ He therefore conjectures ‘cupiunt’ (Halieut., p. 23), which Githling accepts. But ‘capiunt animis’ means ‘ understand in their minds,’ viz. appreciate, as Liv. IX. 9. 14, ‘somnio laetiore, quam quod mentes eorum capere possent.’ S. G. OWEN. NOTES ON THE NATVRALES QVAESTIONES OF SENECA A NEw text of the Naturales Quaestiones, prepared by A. Gercke, was issued by the Teubner Press in 1907. Thirteen years earlier Gercke had, in his Seneca-Studien, prepared the way for a new recension by an elaborate classification of the MSS. An Appendix to the Seneca-Studien is the Studia Annaeana (1900), now unfortunately out of print. These works are an indis- pensable introduction to the study of the text; and they have cost their author an amount of labour which it would be ungracious to utilize, as I propose to utilize it, without very ample recognition. In what follows I shall be largely occupied in criticizing Gercke. I wish, therefore, to begin by allowing the immense debt under which he has put all students of the Naturales Quaestiones. His classification of the MSS. is in itself a fine achievement, and not, I con- ceive, likely to be challenged. Gercke has, moreover, a very fair perception of the relative merits of his numerous MSS. He has, for example, done excellent service in exposing the inferior character of what used to be regarded as our primary authority for the text, the Berlin MS. E. He cherishes, again, no delusions about the rather perplexing Geneva MS. Z. He has rightly called attention to the paramount importance of the Paris MS. H. I fancy, indeed, that if he and I were to place the codices which he enumerates in an order of merit there would not be any great disagreement between us. I differ from Gercke, in fact, not upon the question which MSS. are the best, but upon the hardly less important question which are essential for criticism. I should be inclined to say of Gercke, in a word, that he was a good critic but not a good editor. He cannot persuade himself to leave on one side what has cost labour, or what has been achieved by ingenuity. He crowds his apparatus with the variants of MSS. which he knows to be interpolated at all points—exactly as he crowds it with emendations of which more than half are at the least very improbable. Gercke discovered early—as any one must—that Seneca has always been remissly read, and that he has been left to second-rate critics. Having himself an alert, and even captious, mind, he has been able by a close attention to discover not only a great many hitherto unregarded faults in the text, but also the true remedies for some of these. But he sows with the sack and not with the hand, so that together with not a few brilliant corrections, he has introduced into both text and notes, numberless conjectures NOT. which ¢ that he without nearly e and eve: that it 1 are ofte had rea had sup) more th perhaps time [ 11 poor cri Ger Both cla 20-30 of ‘einen si the strer true, the with the them bel ever In a plain, th the A c revision exempt f upon str authority the other by Gerck offers thi fulgt atio est ¢€ The as thoug ‘2. @ reddant ®, I, 55 7 reli quo ®, ib. quid eas uc a uero ®, transilit A, A, re (-de co A, populis « sentence re. )F 2, was ad, in borate Studta indis- author ose to argely ig the stiOnes. I con- tion of cellent as our in, no called ndeed, rder of sr from on the uld be a good labour, ith the ictly as e least Seneca nd-rate on able >garded ie sOWS yrilliant lectures NOTES ON THE NATVRALES QVAESTIONES OF SENECA 273 which are unnecessary, confusing, and irritating. I would say, in passing, that he has no notion at all of framing an apparatus that can be followed without a close attention of mind. The form of his notes is so cramped that nearly every important note has to be read twice before it can be understood, and even then the report of readings (and of emendations) is given so obscurely that it is sometimes impossible to discover what is meant. The symbols used are often not explained, nor are they even used consistently throughout. I had read two books before I discovered that the symbol Ge. stood, not, as I had supposed, for a late MS., but for Gercke himself—even this remains no more than a brilliant guess! Who, or what, Wim. and Bong. are, I shall perhaps never know—nor yet the distinction between Fort. (which for a long time I interpreted as foriasse) and Fortun., Schult., and Schultess (both of them poor critics in general). Gercke divides the MSS. into two main classes, which he calls A and ®, Both classes he regards as essential for the reconstitution of the text. On pp. 20-30 of his Seneca-Studten he raises the question whether the A class can claim ‘einen selbstandigen Werth.’ This question is answered in the affirmative on the strength of thirteen passages of the first book in which the A class offer a true, the ® class a false, reading. It might be thought that I ought to deal with these passages one by one; but it will, I think, be sufficient if I give them below in a footnote,' and I merely ask the reader to consider whether he ever in any MS. saw thirteen more obvious examples of ‘emendation.’ To be plain, the matter is not worth discussing. For Gercke himself admits that the A class has been everywhere subjected to a thorough-going process of revision and interpolation How, then, he can think these thirteen passages exempt from the suspicion of interpolation passes belief.2 He would have been upon stronger ground if he had based the claim of these MSS. to independent authority on two passages where they supply, or seem to supply, a lacuna in the other class of codices. These passages are cited, for a different purpose, by Gercke, on p. 35 of the Studien. The first is 2, 12, 1, where the ® class offers this : fulguratio ostendit ignem, fulminatio emittit; illa, ut ita dicam, commin- atio est et conatio sine ictu. The A class add the words tsta taculatto stne tctu. At first glance it looks as though these words had been lost from ® by haplography (tctu .. . ictu) 11, I, 15 coire A, ire; 1, 5, 5 reddunt A, in A!). ¢b. exciperentur A, erip(recip- curr.) reddant . ib, unaquaque A, unaquaeque %. erentur %, Praef. 8 and 1, 15, 4 I omit since I, 5, 7 reliqua A, aliqua ®. 1, 6, 1 quod A, there the advantage lies wholly with #. (1, 15, quo &, ib. poterunt A, potuerunt >. 1, 11,1 4, indeed, Gercke must be supposed to have in- quid eas uocem A, quid eas ®. 1,13, 1ueraA, cluded by an accident.) a uero 1,13, 3 autem A,enim ®. I, 14, 4 2 He cites nothing from Books ii.-vii. ; but it transilit A, transtulit @. 1, 15,2 diminutionem would not, I think, help him much if he did, A, re (-de cory.) minutionem #, 1,15, 7speculis though he could easily have found more con. A, populis ® (not unimpressive: the rest of the vincing passages, sentence re-written by an interpolator, however, 274 H. W. GARROD and faithfully preserved in A; and it is certain that something has been lost. Yet I feel sure that the supplement in A is merely conjectural. Not only is the combination tila . . . ista—for illa . . . haec—so far as I am aware, with- out example ; not only is zaculatio a word not elsewhere employed by Seneca; but—and this is what seems to me decisive—the supplement offered by A is, in itself, inadequate, for the reason that Seneca must, in the words that have been lost, have said something about tonitrua. For the sentence which Gercke quotes is immediately preceded by this sentence (which he has, accidentally, suppressed) : tria sunt quae accidunt: fulgurationes, fulmina, tonitrua quae una facta serius audiuntur. Something, therefore, must have been added about tonttrua: and A does not supply us with what has been lost, but at best only with a part of it ; and it is most natural, in view of the extensive interpolations in other parts of these MSS. (interpolations which Gercke is as ready to recognize as I am), to regard this part as a purely conjectural supplement. The other passage where A supply a lacuna is 2, 33: nunc ad fulmina reuertamur, quorum ars in haec tria diuiditur: quemad- modum interpretemur, quemadmodum exploremus. The casual reader will say at once, ‘ Where is the third of the three divisions mentioned?’ It is lost in ®. But A adds quemadmodum exploremus before gquemadmodum interpretemur: and at the end of the sentence has exoremus for the exploremus of ®. This is very likely what Seneca wrote; but I think it was a correction not beyond the conjectural powers of the scholar to whom we owe elsewhere in A numberless conjectures of immense ingenuity. His text, after all, told him plainly that a threefold division was required. These two passages would, in fact, be impressive’ save that the A MSS. come to us with so bad a character on other grounds. There are one or two other passages—not dealt with by Gercke—where A fills in plausibly a presumed lacunae—e.g. 2, 12, 6 , 2, 18 ictus inaequalis, 6, 32,9 <_feres>. Lam not able here to deal in detail either with these or with a number of variants in A which certainly merit attention. It must suffice here to have dealt briefly with those passages on which Gercke himself mainly relies as indicating the independence of A. I am satisfied if I can set up in a few readers of Gercke's text an active suspicion of this family of MSS. I feel, at any rate, that the whole question of their independence needs reconsidering—and by some one, let me add frankly, who has analyzed all the minutiae of the MS. data in a manner more thorough than has been possible for me. Meanwhile, if I might put dogmatically a view of the textual problem which seems to me truer than that of Gercke, I would say this. 1 Mainly because the omissions in @look as _ haplography. though they might easily be explained as due to NOT Thi the Nat will be t to cast himself. tively e: criticism which s conjectu to be lez but the : our natu in perioc seriously I no mo! continue interpola Berlin E Let The ® cl and p. sented b thinking than any duty of t my Own to be full accurate. JKLMO I have ne derived f slight val of omissi« rather L} family. ® group i claim to interpolat and AX (or from this readings. We principles cannot g 1 lost. nly is with- neca ; » Ais, - have rercke ntally, facta \ does 3 and | these regard emad- - three Oremus oremus hink it om we s text, MSS. or two ‘sumed lam ariants briefly ng the ercke’s vat the 1e one, fa in a might er than —"“* NOTES ON THE NATVRALES QVAESTIONES OF SENECA 275 The first duty of the next editor of Seneca will be to rid the Apparatus of the Naturales Quaestiones of the endless lumber of A readings. His next duty will be to do what Gercke, on his own principles, ought long ago to have done to cast aside the MS, known as Z (and the readings of L*, which Gercke himself rightly recognizes to have been introduced from Z). Z is a compara- tively early MS. (saec. XII.), and highly interesting. But its interest for criticism is only the interest which attaches to a pre-Renaissance MS., in which some scholar of real accomplishment has occasionally restored by conjecture a true reading. Vuixere fortes ante Poggium—that is the only lesson to be learnt from Z. It is rarely that this lesson is put to us so convincingly ; but the study of other MSS., in other authors, is gradually helping us, against our natural prejudices, to recognize the existence of a refined Latin scholarship in periods hitherto held innocent of such a thing. I do not think that Gercke seriously dissents from this view of Z. But why he persists in citing this MS. I no more know than I know why, having said that L* is derived from Z, he continues to cite it as though it were not. Two other MSS. so hopelessly interpolated as to be valueless are the Paris T and the once much vaunted Berlin E. Let us see what we are now left with. We have dismissed the A class. The ® class is split up by Gercke into three sub-classes, which he calls @, X, and p. Of these the best is undoubtedly, as Gercke holds, @. It is repre- sented by EFGH. Of these MSS., as I have said, I agree with Gercke in thinking 7 the best MS. we have. G appears to be more closely allied to it than any other; but G has been examined only very imperfectly, and a primary duty of the next editor will be to collate this MS. in full. F, which belongs to my own college, Merton 250, is inferior to either H or G, but possibly deserves to be fully cited. Gercke’s collation of it is endlessly defective and not always accurate. Of E 1 have already spoken. The sub-class X is represented by JKLMO (why editors allow themselves such barbarisms in their notation as J, I have never understood). Of these / is an inferior twin of K, and, as perhaps derived from K, may be dispensed with. slight value. M consists merely of excerpts of O is marred by endless carelessnesses, particularly in the matter of omissions. The best representative of this sub-class is undoubtedly L (or rather L*). This MS. by itself is, perhaps, a sufficient representative of its family. If we need another it will have to be K. The third sub-class of the ® group is p._ In this sub-class I can discover no independent merit. Its only claim to note is that it has preserved entire 2, 48, 1; but so has the deeply interpolated Z. This entirety is pretty certainly achieved by conflating @ and X (or possibly A and A). Gercke adduces no examples of independence from this sub-class, and I think we might well relieve the Apparatus of its readings. We need, then, a new Apparatus to the Naturales Quaestiones, in which the principles adumbrated by Gercke shall be carried—as, for reasons which I cannot guess at, Gercke does not carry them—to their logical conclusion; an 276 H. W. GARROD Apparatus, that is to say, in which four MSS., or at most five, shall figure, viz.. HGFL(K). In what follows I shall take account only of these codices— where I refer to any others, it will merely be for the purpose of exposing their unreliable character. In what I have said I have not, of course, attempted to prove in detail the scheme adumbrated. The proof of it is to be sought in a study of Gercke’s Apparatus, and in the further information furnished by his two volumes of Studien. In the criticisms and suggestions here offered I depend entirely for the report of readings upon Gercke. I have not done more myself than examine the two Merton College MSS., of one of which Gercke entertains a high opinion. (The later of the two I have only glanced at.) The numbers attached to my Notes refer to page and line in Gercke’s edition. This is not an ideal method of citation, but any other would be necessarily very cumbrous, and Gercke’s edition is likely to remain for some time the standard text. 2, 1-3 cum disco.. aliquando respiciat intendatur Madvig: secedat? . quid sit deus, totus in se tendat an et ad nos 3, 22 sqq. o quam ridiculi sunt mortalium termini, ué ultra Istrum Dacus non exerceat imperium, Haemo Thrace se includat, Parthis obstet Euphrates ut ante ultra addidi exerceat scripst: exeat codd. Thrace se_ scripsi: Thraces codd. 5» 3-4 paucissimorum dierum spatium, si nauem suus ferat uentus. For ferat uentus there are variants from A, uentus impleuit and impleat uentus. These could perhaps be explained by an original st sinus impleat uentus—nauem being a mere gloss upon sinus. But, from what has been said of the character of A, it seems far more likely that zmpleat is merely interpolated. 7, 27-8, 3 quemadmodum nubes collisae mediocriter fulgurationes effciunt, maiore impetu impulsae fulmina, sic quanto illa minus press? aeres minoresue, tanto leuiora /umina emittunt. illa minus pressi aeres scvipfst: illas minus presseris codd. lumina scrtpst: fulmina codd. (tlla=1bt; fulmina is clearly a blunder :—for lumina cf. 7, 22 minora lumina excutiuntur). 8, 9-10 acrior sit Caniculae rubor, Martis remissior. Martis would seem to be a blunder for Venerts—the symbol! ? being mistaken for d. At any rate I know of no writer who attributes to Mars a vemissus color, and he is usually represented as of a peculiarly fiery hue. 18, 24-25 nec dubium cuiquam relinquitur quin arcus solis imago sit male expressi ob uitium figuramque specull. For expressi (which is of course quite possible) we should perhaps, in view of 26, 24, solis imago cuius exprimendae capax non est, write expressa. 1 At 258, 24 the symbol ©=Sol has been lost before the first letter of obumbratuy (Skutsch). NOT IQ, uidemu: que quomod 20, tudinun sim 2i, et varil: Ge! ‘not on 22, radiorul Per A, arose 23; teneas, 1 alti not be t! ostendat was easl 26, aliquid | est in qu For everythi it to be (t:eeiam ° 26, : et ad ho foret, tot mat fabricata inspector In t altering : and the 2 has not f 29, 2 imag figure, \dices-— g their ted to it ina by his for the Kamine pinion. to my nethod ercke’s ad nos Dacus hrates scripst : impleat wmpleat said of olated. ationes 4 aeres scrips : 9, oe being Mars a ago sit in view ch). NOTES ON THE NATVRALES QVAESTIONES OF SENECA 277 Ig, 23-20, I nunc nihil ad rem pertinet quomodo uideamus quodcumque uidemus: sed quo¢mod? imago similis reddi debet e speculo. quoimodi (uel quoiquoi modi) scrifst (=cuimodi, for cuiuscuiusmodi): quomodo codd. 20, 19-20 in unam imagines non coibunt, sed unaquaeque in se simili- tudinum seriem claudet. similitudinum seriem scripst: similitudinem in se rei codd. 21, 13-16 alioquin, ut ait Nero Caesar disertissime, colla Cytheriacae splendent agitata columbae et variis coloribus pauonum ceruix, quotiens aliquo deflectitur, nitet. Gercke interprets alioquin by ‘uebrigens.’ No: alioguin ... et means ‘not only is it true that ... but also’—the usage is illustrated in the lexicons. 22, 13-15 non enim idem facit si undique effulsit, et ad hoc opus est radiorum idoneus ictus. Perhaps si undecumque effulsit, set. The variant zdonea locatio, found in A, arose from ictus read as locus. 23, 9-II purpuram Tyriam, quo melior est saturiorque, eo altius oportet teneas, ut fulgorem suum teneat. altius is surely a blunder for /atius—the cloth must be spread out, it would not be the brighter for being held high. teneat gives a wholly false sense: the ostendat of other MSS. might do, but Seneca probably wrote intendat (the in was easily lost after suwm). 26, 5 sqq. in nube est aliquid uitro simile, quod potest perlucere, est aliquid et aquae quam, etsi nondum habet, iam parat, id est iam eius natura est in quam ex sua uertatur. For id est iam etus natura write td est, essenttam etus nactura, and excise everything from id est... to uertatur: the words only need to be emended for it to be seen at once that they are merely a gloss on the clause preceding (i:eelam was read as 1.¢.1am). 26, 24-27,4 ita conatur quidem reddere imaginem, quia leuis est materiae et ad hoc habilis, sed non potest quia enormiter facta est. si aperta fabrica foret, totidem redderet soles quot aperuisset inspectut toros. materiae scripsi: materia codd. aperta (=apta) fabrica scripst: apta fabricata codd. aperuisset (=apuisset) inspectui toros scrifsi: habuisset inspectores codd. In the sentence that follows qui quia discernuntur etc. nothing requires altering : gui, which is the reading of all the good MSS., refers back to toros, and the uix which Gercke inserts after guia he has merely inserted because he has not perceived that discernuntur =non continuantur (not utdentur). 29, 24-25 quid uocem ? imagines solis? imagines? soles? Leo. Malim: solesne an imagines solis ? 278 H. W. GARROD 30, Q-II sunt autem imagines solis in nube spissa et uicina in modum speculi. uicina seems nonsensical. It is perhaps a blunder for tucurua (i.e. uicina for inc’ua). 31, 24 debent autem hae nubes quae hoc praestant densae esse, leues, splendidae, planae, imaginaturae soles. imaginaturae scripst: naturae codd. (nacturae would perhaps be just possible). For solis I have written soles (naturae solis similes, Leo: if this be what Seneca wished to say a simpler correction would be natura est (@) solts. 32, 13 sqq. horum plura genera conspiciuntur: sunt ut ei cum uelut corona cingente introrsus ingens caeli recessus est similis effossae in orbem specus: sunt pithiaecum .. . etc. For sunt ut et Gercke gives sunt puter. But Pliny is giving Greek names throughout. I would suggest sunt cute (or perhaps ceuthe). pithiae would be better written pitheae. In the parallel passage in Pliny, N.H. 2, 90, why do editors insist on retaining the false form pztheus, which is a mere blunder for pitheas ? 33, II sqq. quia cito spatium sursum transilit et oculis nostris occurrit uniuersum per quod deiectus est. sursum wel suum uel sursum suum codd. Perhaps s(e)orsum. 40,17 sqq. tunc quoque cum antiqui illi uiri incondite uluerent, satis nitidi si squalorem opere collectum aduerso Aumine eluerant, cura comere capillum fuit ac prominentem barbam depectere, et in hac re sibi quisque alteri in uicem operam dabat. Between eluerant and cuva Gercke wishes to insert (and, indeed, has inserted) the word parua. Nothing could be more false, since, from tunc quogue, it is clear that Seneca would then be implying that in his time also the ‘ cura’ for these adornments was ‘ parua ’—which is precisely the opposite of what he Is trying to say. Some deep corruption lurks in the end of the sentence. I would suggest that alteri has taken the place of the genitive of some noun meaning ‘hair- dresser.’ cinerarii, which would give the required sense, is too far from the ductus litterarum. Can the right word be comatori(t): (? atori=alteri) ? Petronius has acu comatoria (21), of a hairpin, and (seruus) comatorius seems just possible in the sense of cinerartus. 41, II1-13_ 0 felix paupertas quae tanto titulo locum fecit, non fecisset illa dotem si habuissent. So the best MSS. (I have put a comma after fecit). Both Madvig and Leo have suggested violent changes. I see no reason for any change at all, and would render: ‘O blessed Poverty that made such glory possible but would ns NO’ not hz begin which tion, b not be illogic 4. parat | corpus al writte severa contra MSS. tions ' The s N.Q. | glorial which and c been : some hand, fusion Revues draw — MSS. differe minus 4! terrog: illa ut usquar stellis in (uniue? scvipst the m; editor modum Bb, uUlcina e, leues, ossible). ; Seneca m uelut 1 orbem k names ‘ould be why do nder for occurrit nt, satis comere quisque eed, has ‘om func ime also opposite suggest ig ‘ hair- from the ’etronius possible isset illa lvig and t all, and ut would NOTES ON THE NATVRALES QVAESTIONES OF SENECA 279 not have left room for it if they (the daughters of Scipio) had had a dowry to begin with (if their family had possessed wealth).’ «lla refers back to paupertas, which is practically personified. Poverty was the cause of this great distinc- tion, but she would not have been the cause of it if the daughters of Scipio had not been too poor to receive the ordinary marriage portion. There is a slight illogicality, but it is that of an epigram missing fire. 45, 7-IO rursus quasi pars est sanguis nostri, et tamen est materia: prae- parat enim idem alimentum et nihilominus in numero eorum est quibus totum corpus efficitur. ; t alimentum scrips: alia codd. (alt was taken for alia). I have further written idem for the zd or et of the MSS., and have accepted the et tamen of several of the interpolated MSS. as against the qui et tamen of the others. The t contraction ali for alimentum is given as fourteenth century in Cappelli, but the MSS. of Seneca bear witness to an archetype which contained many contrac- tions which lack attestation (so far as our knowledge goes) from early MSS. The subject is one on which we are still very insufficiently informed. At N.Q. p. 136, 26 generabitur is accepted as a certain correction of the MSS.’ gloriabitur. But this implies a confusion of glabitur and gnabitur—both of which contractions, therefore, must be of considerable antiquity. The date and character, I may say in passing, of the Archetype of our MSS. has not been adequately investigated. A proper investigation would yield, I fancy, some notable problems. The Archetype was clearly in a minuscule hand: a hand, incidentally, in which autem and enim easily lent themselves to con- fusion (see e.g. 20, 1: 61, 3: 146, 8; but Lindsay has recently pointed out, Revues des Bibliothéques, 1912, that we must be cautious what inference we draw from this confusion in MSS.). But the remarkable thing about the MSS. of Seneca is the thoroughgoing and repeated recensions to which the different groups have been subjected since their separation from the common minuscule stock. (The Archetype is scarcely older than saec. ix.) 45, 26-46, 6 pars quare sit non puto te interrogaturum, ut neque in- terrogas quare caelum pars sit: quia scilicet non magis sine hoc quam sine illa uniuersum potest esse; quod cum im his universi summa est ex quibus usquam alitur, ex hac alimenta omnibus animalibus, omnibus satis, omnibus stellis diuiduntur ... in ante his addidi: uniuersi summa scripsi : uniuersis codd. (uniuersum dett.) a (untuersis is a misunderstanding of uniuersi s-= uniuersi summa) quibus usquam scvipst: quibus quam codd. alitur addid: (‘the sum of the universe consists in the materials everywhere of its nourishment’). These proposed changes are not, I think, more drastic than those of most editors—of whom Gercke is the least satisfactory. ‘Why Gercke should refuse 280 H. W. GARROD to accept Kroll’s certain correction ut neque (aeque codd.) interrogas I am not able to divine. 47, 9-10 nonne intenti spiritus opera sunt ? Perhaps opera fiunt (opera abl. sing.). 47, 17 monne aeris intentione partes suas explicant ? Gercke’s comment I find unintelligible. explicant appears to be put for explent (unless explent be the true reading). 51,16 hae tot partes. The best MSS. have haec. Why Gercke prefers hae I cannot surmise, haec, as the rarer form, is clearly preferable, and the c might easily have been lost before fot. 52, 2-3 nec tamen eodem modo totus afficitur sed aliter alibi et partibus inquietus ac turbidus est. The text can hardly be sound. Perhaps for e¢ partibus we should read nec paribus uicibus>. 52, 9-12 sed ceterae quoque stellae non minus terrena quam incumbentem terris spiritum afficiunt et ciycuttu suo occursuue contrario modo frigora modo imbres aliasque terris turbide iniurias mouent. circuitu scripst: cultu (uel ortu dett.) codd. 54, 20-21 ita non descendit ignis sed praeciptatur et deducitur. decutitur or deicitur seems more likely than deducitur. 60, 15-17. nec hoc tantum in his debere credi ac siderum quorum ingens in confesso potentia est. For ac siderum read consideramus (=csideram’). This is as near as Madvig’s suggestion, and does not, as does his, necessitate our receiving the debet of the inferior MSS. (I find it difficult, moreover, to believe that accidere (Madvig) could easily have passed into ac stderum). 61, 13 quomodo, inquit, cum dicatis ignis hanc esse naturam ut petat superiora, fulmen terram petit? aut falsum est hoc aut falsum est quod de igne dixistis. The words italicized are my own supplement. 62, 10-12 quomodo ergo possunt gignere ignem, quem non ueri simile est ex nube quam ex aqua nascitur ? ET have fieri for nascitur. This is, no doubt, the merest interpolation. But it seems to point to masci as the true correction. Yet Gercke has actually printed in the text his own conjecture nasciturum! Adeo grammatici nostri ea quae quiuis puer Romanus sciebat neglegunt, nos autem senes ea operose quaerere cogimur quae magistri nostri olim tradere debebant! (Lachmann, ee ee NO Lucr. itself 6 cum i fereba ex int \ correc we sh For a ( Prese 7 ista et T sugge: emenc 7 relictt A would 7 comb est: 1 nescit S punct either = bur 7 dissen tres il 2 corun gener nouen to Tou ME im not put for urmise, e been artibus ead nec bentem 1 modo ingens advig’s of the ladvig) t petat Je igne nile est lation. ctually ystri ea yperose mann, ere NOTES ON THE NATVRALES QVAESTIONES OF SENECA 281 Lucr. p. 207), mascttur arose from ex urbe nascitur following (which, no doubt, itself needs explanation). 63, 5.sqq- quidni? maiorum nostrorum memoria, ut Poseidonius tradidit, cum insula in Aegaeo mari surgeret, spumabat interdiu mare et fumus ex alto ferebatur : nam demum (deinde /KLMO) prodebat ignem, non continuum sed ex interuallis emicantem fulminum more .. . etc. Madvig altered am to nox, with the approval of Gercke and the world—a correction palaeographically singularly improbable. I suspect that for interdiu we should read interdum (‘for a time’), and then, for nam, una (with detnde). For a while there was just smoke, then smoke accompanied by (una) flame. (Presently there was a discharge of rocks and stones). 72, 10sqq. permitte mihi illam rigidam sectam tueri eorum qui excipiunt ista et nihil esse aliud quam aegrae mentis solatia existimant. The corruption perhaps goes deeper than the various remedies proposed suggest. Possibly qui extispicum ista nihil is not more remote than most of the emendations recorded by Gercke. 74, 8-13 ista nobis opponi solent ut probetur nihil uoluntati nostrae relictum et omne ius fati faciendi traditum. A much-vexed passage where none of the proposed corrections satisfy. I would hazard relictum, et non nist tus fatt patiendt traditum. 76, 3 sqq. omnia ista urunt sed genere et modo differunt : quodcumque combustum est utique et ustum est, at quod ustum est non utique combustum est: item quod accensum est (potest autem illud ipso transitu ignis ussisse) quis nescit uri quidem nec ardere, nihil autem ardere quod non et uratur ? Save autem for enim (see on 45, 7-10) I have altered nothing but the punctuation—following here a suggestion of Gercke. I see nothing the matter either with the sense or with the Latinity. Seneca is distinguishing uro = burn, comburo= burn up, accendo= fire, afflare=scorch, ardere = blaze. 76,17 sqq- haec adhuc Etruscis philosophisque communia sunt: in illo dissentiunt quod fulmina a Ioue (fulmina nouem, the best MSS.) dicunt mitti et tres illi manubias dant. So Gercke’s text (a Ioue AE). But compare Pliny, N.H. 2, 138: Tus- corum litterae nouem deos emittere fulmina existimant, eaque esse undecim generum: Iouem enim tria iaculari. This passage would seem to guarantee nouem to Seneca: in that case we shall have to correct il/1 (which A omits) to lout. H. W. GARROD. MERTON COLLEGE, OXFORD. [To be continued. ] CLAVDIVS AND THE PARIMORES GALLIAE: A REPLY AND A RESTATEMENT. In the April number of the Classical Quarterly, Mr. H. J. Cunningham calls in question my interpretation of a passage in the speech of Claudius on the claims of the Gallic chieftains, which I will follow his good example by quoting in full. It forms the first legible sentence of Column II. ‘Sane nouo mfore] et diuus Aug{ustus aujunciulus mjeus et patruus Ti. Czsar omnem florem ubique coloniarum et municipiorum, bonorum scilicet hominum et locupletium, in hac curia esse uoluit. Quid ergo? Non Italicus senator prouinciali potior est? lam uobis, cum hanc partem censurae meae adprobare coepero, quid de ea re sentiam, rebus ostendam. Sed ne prouinciales quidem, si modo ornare curiam poterunt, reiciendos puto.’ That this statement represents not only the practice, but the deliberate policy (uoluit) probably of Augustus, certainly of Tiberius, as to the admission of senators from outside Rome, will, I imagine, be admitted by all. The only question is whether any geographical limitation is to be placed upon the phrase ‘coloniae et municipia,’ or in other words, whether a ‘municipalis origo’ generally, joined, of course, to respectability and wealth, was by these emperors made the condition of eligibility to the senate, or whether, in spite of the generality of the phrase, and the addition of the word ‘ubique,’ we are justified in limiting this eligibility to a particular portion of the empire. In consideration of the fact that there were a large number of colonies and municipia in provinces like Narbonensis, Baetica, Tarraconensis and Africa, and that ‘ubique’ could not have been inserted without some reason, I have, as Mr. Cunning- ham complains, given to the latter word its full force, and translated it ‘ throughout the empire.’ As the logical result of this interpretation, 1 have assumed that, under the conditions of eligibility thus laid down by the first two emperors, Roman citizens belonging to the provinces named above might be, and were, admitted to the senate, but that the Gallic chieftains, about whose want of a ‘ municipalis origo’ there can be no doubt, would be excluded. I was, of course, aware that Mr. Furneaux had given a different and a narrower meaning to the phrase ‘coloniae et municipia,’ taking it as equivalent to ‘the towns of Italy,’ an interpretation which would limit ‘ubique’ to mean ‘ throughout Italy.’ In his note to Ann. III. 55, he infers from this passage of the Speech of Claudius that Tiberius admitted many senators from Italian towns, and refers back to his note on I. 79, 10. On the latter passage he says that ‘these terms (coloniae et municipia) are constantly used together by Tacitus (cf. III. 55,4; IV. 67, 1; XV. 33, 3; Hs#. II. 20, 1; 56,1; 62, 4; etc.) to designate the towns of Italy.’ If Mr. Furneaux had expressed any strong view as to the policy of either Tiberius or Claudius in this matter, I certainly should not have ignored it, but, as these notes seemed more or less _ incidental, and I thought his references far from convincing, I did not notice his sug- gestion. Since, however, Mr. Cunningham has restated Furneaux’s interpretation of the pl altoge ‘the o no Su] adduc \ ently the cc they 2 of col 55» ca only | which valent prefix De bel A questi releva conte: passa mean: As th woulc decid groun indicz dent « ing Oo! woulc will s whickt I the ex twelv equiv the cc | presu of it. I cluiul are, © Italy, colon of al majo! to Ita 2 equite were It is m calls aims of ull. It uus Ti. scilicet [talicus e meae inciales : policy enators stion is niae et ned, of tion of ind the rticular ies and id that inning- ughout , under citizens senate, can be rrower - towns Italy.’ us that note on bia) are ast. II. ux had in this or less is sug- tion of CLAVDIVS AND THE PRIMORES GALLIAE 283 the phrase ‘coloniae et municipia,’ adding to the references cited above some not altogether happily selected ones of his own, and claims by their means to show that ‘the one piece of evidence which I can produce in support of my explanation gives it no support whatever,’ I propose to examine the imposing array of passages which are adduced to establish this statement. Mr. Cunningham hardly does justice to my position when he says that I ‘ appar- ently think that the colonies and municipia alluded to are those in the provinces.’ On the contrary, my whole argument, as well as my translation of ‘ ubique’ implies that they are the towns of that description both in Italy and in the provinces. I do not, of course, deny that the great majority of the ‘noui homines’ spoken of in Amn., III. 55, came from Italian towns. Mr. Cunningham simplifies the issue by asserting not only that the phrase ‘coloniae et municipia’ is constantly used of Italian towns, which in its distributive sense nobody will deny, but that it ‘so comes to be an equi- valent for Italy.’ To prove this last point, on which his whole argument depends, he prefixes to Furneaux’s references the following: Lex agr. 31; Sall. Cat. 17 and 58; De bell. Gall., VIII., 50; and Mon. Anc. 21. As all these passages are cited with the sole object of proving that the phrase in question is ‘equivalent to Italy,’ it is curious that Mr. Cunningham should think it relevant to remark that ‘in most of these passages’ (he might have said ‘in all’) ‘ the context shows that Italian towns are meant.’ The whole question involved in this passage of the speech is whether the words ‘ coloniae et municipia’ are to decide the meaning of ‘ubique,’ or whether that word is to decide the extension of the phrase. As there were notoriously a large number of colonies and municipia outside Italy, it would seem natural to let the context in each case (and here it is the word ‘ ubique’) decide the range of the expression. If this is objected to, it can only be on the ground that ‘ coloniae et municipia’ was not only a description of Italian towns, as indicated by the context, but such a recognized synonym for ‘ Italy,’ and so indepen- dent of any context, as actually to limit, as in the present case, the usual wide mean- ing of a word like‘ ubique.’ Considering the point he has to prove, Mr. Cunningham would have been wiser to keep out of sight, what an examination of his references will soon enough reveal, that it is in every case the context and not the phrase itself which fixes the reference to Italy. But what is even more surprising after all this marshalling of evidence to prove the equivalence of the phrase to ‘ Italy,’ is the fact that in only two passages out of twelve is it used collectively at all, and that in the other ten, so far from being equivalent to Italy, it is merely applied to certain Italian towns strictly limited by the context as to number or locality or both. I will not apologise for taking Mr. Cunningham’s references one by one. He presumably depends upon their cumulative weight, and he shall have the full benefit of it. 1. Lex Agr. 31. Mention is made here of certain ‘coloniae seiue mounicipia ciuium Rom. nominisue Latini’ to which public land ‘ fruendus datus est.’ They are, of course, Italian towns, because this part of the law is dealing only with land in Italy, and because in 111 B.c. there were, with a solitary exception or two, no colonies or municipia outside Italy. As the expression is not only not used collectively of al] towns within this class, but excludes all ciuttates foedevatae, i.e., the great majority of Italian towns, it seems preposterous to cite the phrase here as ‘ equivalent to Italy.’ 2. Sall. Cat. 17. After naming Catiline’s associates among the senators and equites, Sallust adds: ‘ad hoc multi ex coloniis et municipiis domi nobiles.’ There were still no such towns outside Italy, and again the phrase is not used collectively. It is not equivalent to ‘Italy,’ but to ‘some of the towns in Italy.’ The collective 284 E. G. HARDY description of Italian towns at this time would have been ‘coloniae municipia et praefecturae.’ 3. Id. 2b. 58. Catiline, addressing his army, points out that their case is desper- ate unless they gain a victory. But if they do, ‘omnia nobis tuta erunt, commeatus abunde, municipia atque coloniae patebunt.’ He does not mean that ‘Italy’ would be won by a single victory, but that the neigbouring towns and those along the road to Rome would open their gates and furnish supplies. 4. Hirtius, De bell. Gall, VIII., 50. Caesar, after spending the winter in Gaul, hastens into Italy, in order to recommend Antony’s candidature for an augurship to the ‘municipia et coloniae.’ The passage is quite irrelevant to Mr. Cunningham’s point. Caesar’s objective was ‘Italy,’ because he had business with some of the Italian towns, obviously those in the proximity of his own province. 5. Tac. Ann. I. 79. The question was raised in the senate as to preventing Tiber floods by diverting some of its tributaries and by damming some of the lakes which fed its waters. Many towns besides the three named by Tacitus, might be injuriously affected by these operations, and accordingly ‘auditae municipiorum et coloniarum legationes.’ The towns were obviously all within the area directly or indirectly affected by the suggested works. It is difficult to believe that Mr. Cun- ningham had looked this reference up before including it among passages supposed to prove that ‘coloniae et municipia,’ is a phrase ‘ equivalent to Italy.’ 6. Id. +b. iv., 67. Tiberius, passing through Campania, though he had notified the people of the towns not to disturb his privacy, ‘ perosus tamen municipia et colonias omniaque in continenti sita, Capreas in insulam se abdidit.’ The towns are so obviously those in Campania along the emperor’s route, that we need waste no more time over this irrelevant passage. 7. Id. tb. xv., 33. Nero, wishing for a wider audience, chooses Naples as the scene of his first public performance. Accordingly, the theatre at Naples is filled with ‘ contractum oppidanorum uulgus, et quos e proximis municipiis et coloniis eius rei fama acciuerat.’ Again I am compelled to ask, had Mr. Cunningham verified this reference, which alludes only to the colonies and municifia in the immediate proximity of Naples? 8. Id. Hist. II. 20. Caecina, leaving his cruelty and licence on the other side of the Alps, ‘modesto agmine per Italiam incessit. Ornatum ipsius municipia et coloniae in superbiam trahebant.* The local limitation of the phrase is not less evident here than in the previous passages. The words ‘per Italiam’ are in contrast with ‘post Alpes,’ and merely mark the fact that the towns in question are south of the Alps. g. Id. 1b. II. 56. ‘Ceterum Italia grauius et atrocius quam bello adflictabatur. Dispersi per municipia et colonias Vitelliani spoliare rapere.’ I suppose it might be argued here that ‘ municipia et coloniae’ is a mere repetition of ‘Italia.’ I do not think so. Italy, as in the last passage, is used in strong contrast with the provinces. Unlike them, Italian soil was supposed to be free from war and its concomitants, and the violation of this theory is marked by the first sentence. The second gives particular instances. The towns, 7.¢., those which lay along their line of march, were brutally treated by the soldiers. Possibly more towns are included under the phrase than in the other passages, but it is no more used collectively than in them. 10. Id. 2b. II. 62. Among other acts of moderation and reform, Vitellius forbade Roman knights to disgrace themselves in the arena. Previous emperors had often enforced this, ‘ac pleraque municipia et coloniae aemulabantur corruptissimum quemque ... pretio inlicere.’ The qualifying ‘ pleraque’ is of course fatal to the phrase being taken in a collective sense. It means not ‘Italy,’ but ‘a good many Italian towns.’ ee In passag reserve undoul suspec 12 in sens quotes, It occ accessi though statem: essentl: ‘coloni has bre what il with re was in also th: ditione not ne or Tibe the alr the rule describ But as law, he really 1 passage accordi! non sequ ‘honest argume interpre and imy Hi: misunds speech. ‘senato: ‘coloniz with ar these we thing.’ propose: pia et esper- neatus would e road Gaul, ship to ham’s some enting » lakes tht be um et tly or _ Cun- sed to otified pia et ns are ) more as the , filled is eius erified ediate ir side pia et ot less yntrast uth of batur. ght be do not yinces. itants, gives 1, were phrase itellius rs had simum to the many CLAVDIVS AND THE PRIMORES GALLIAE 285 In the light of this examination, it is, I think, not too much to say that these ten passages are utterly irrelevant to the point which they are cited toestablish. I have reserved to the last two passages in which the phrase ‘municipia et coloniae’ is undoubtedly used collectively, and is applicable to all or any Italian towns. I suspect that they will not do much to help Mr. Cunningham’s cause. 11. Mon. Anc. IV. 21. In this passage Augustus makes a statement which is equivalent to saying that he remitted the auvum coronartum to the whole of Italy, and he uses the words ‘municipiis et coloniis.’ But those who have followed Mr. Cun- ningham’s argument will be surprised to find that he does not use them alone, and that the actual statement is ‘Italiae municipiis et coloniis ... remisi.’ Does Mr. Cunningham really think that this is an instance which he has a right to cite of ‘municipia et coloniae’ being ‘equivalent to Italy?’ If he does, he is following some canon of logic unknown to me. 12. Tac. Ann. III. 55. ‘Noui homines e municipiis et coloniis ac etiam prouinciis in senatum crebro adsumpti.’ This is the only passage which Mr. Cunningham quotes, and I wish that I had remembered it when I was dealing with the subject. It occurs among some general reflections on the period between Actium and the accession of Galba, but I believe that it refers to the time of Augustus or Tiberius, though it was no doubt true uuder Claudius, if not under Nero. It is undoubtedly a statement that senators were adlected from Italy and the provinces. But it is essential to remember that this is not one instance among many others of the phrase ‘coloniae et municipia’ standing for ‘ Italy,’ for every one of the supposed instances has broken down. I claim, therefore, that the phrase with an open context means, what it ought to mean, all towns constituted on the Italian model. That the policy with regard to admission to the senate, certainly of Tiberius, probably of Augustus, was in some way formulated, is clear from the words of Claudius, quoted above, and also that it turned on the phrase ‘ coloniae et municipia.’ I understand that it con- ditione’ eligibility by an ovigo in one of these communities. The word ‘ubique’ was not necessary, and was probably not contained in the rule as laid down by Augustus or Tiberius, but was not unnaturally inserted by Claudius, who wished to emphasise the already established eligibility of provincials. Tacitus was presumably aware of the rule as formulated by the earlier emperors, and knew that technically it could be described by saying ‘Noui homines e municipiis et coloniis in senatum adsumpti.’ But as his readers were not likely to be always thinking in terms of constitutional law, he added, to make an important point explicit, ‘ac etiam prouinciis,’ which is really no more than the ‘ubique’ of Claudius. Mr. Cunningham comes to the passage with the preconceived opinion that ‘ municipia et coloniae’ mean ‘ Italy,’ and accordingly finds in the phrase a sharp contrast to ‘ prouinciae,’ which by a curious non sequituy he thinks is made practically certain by the words of Suetonius (Vesp. IX), ‘honestissimo quoque Italicorum et prouincialium adlecto.’ Mr. Cunningham’s first argument has depended on an appeal to unquoted and irrelevant references, and a mis- interpretation on the strength of their supposed cumulative weight of the one relevant and important passage. His second objection to my view will be found to rest upon a still more serious misunderstanding of the argument and order of thought in this part of Claudius’ speech. He thinks, as apparently did Mr. Furneaux, that the introduction of the ‘senator Italicus’ into the next sentence confirms the restricted application of ‘coloniae et municipia.’ Ciaudius is here, according to Mr. Cunningham, dealing with an anticipated objection, which ‘freely paraphrased,’ amounts to this: ‘ But these were Italians, whereas you are proposing to bring in provincials, a very difterent thing.’ But this way of putting things inevitably suggests that what Claudius proposed in 46 a.D., what the senators ob 2cted to, and what was ‘a very different 286 E. G. HARDY thing.’ from anything done by Augustus or Tiberius, was the admission of provincials to the senate. In that case, to what period does Mr. Cunningham assign the statement in Amn. III. 55 that new men were admitted to the senate from the provinces? When were the senators of Narbonensis admitted, to whom in 49 a.p. Claudius gave the right of visiting their estates at will? (Ann. XII. 23) How does he explain Claudius’ own statement about Vienna, that it had sent senators to Rome ‘longo iam tempore’? or the fact that Valerius Asiaticus was a consular under Caligula? or the presence of the worthy senators from Lugudunum? It is abundantly clear that provincials had been admitted in some numbers by Tiberius and perhaps by Augustus, and this fact is entirely irreconcilable with Mr. Cunning- ham’s explanation of the objection which Claudius is answering. Besides, how can there have been any objection to Claudius’ proposal at this point of the speech, when the most superficial reading of it shows that no proposal had yet been made at all? It is not until after he has dealt with Vienna and Vestinus and Asiaticus, that the emperor reminds himself that it is time ‘detegere patribus conscriptis quo tendat oratio tua.’ Even then, the actual proposal is postponed for half a dozen lines, and, when it comes at last, it is not a proposal to admit provincials in contrast with Italians, but to admit senators from a particular region, Gallia Comata, lying outside ‘adsuetos familiaresque uobis prouinciarum terminos,’ from which senators had hitherto been taken. Then, having made his proposal, but naturally not before, Claudius proceeds to deal with anticipated objections to it. It would have been well if Mr. Cunningham had revived his memory of the speech before presenting such a travesty of its argument as that involved in his interpretation. He asks why Italy and Italian senators should ‘be dragged into a discussion which only concerns two classes of residents in the provinces.” ‘They are of course dragged into no such discussion for the simple reason that Claudius has not yet reached the point at which these two classes of provincials are distinguished. That there is a sufficient reason on my interpretation for the introduction of the Italian senator will, I think, be obvious if I too may be allowed the liberty of a ‘free paraphrase.’ I will venture to put the whole passage in this way: ‘It was an innovation for Augustus and Tiberius to lay it down that the senate- house should be open to men of property and respectability from any colonies or municipia throughout the Empire’ (cries of disapproval from benches on the right). ‘Ah! you think that policy puts Italians and provincials too much on the same footing, and that Italians make the better senators. Well, when I come to my actual adlection, you will see that to some extent I share that feeling, but not the prejudice which would exclude provincials altogether. But to return to my predecessors’ policy, look, for instance, at the colony of Vienna, etc.’ As a matter of fact, there are two points about the passage which on Mr. Cun- ningham’s view of it are absolutely inexplicable. (1) Howcan the policy of Augustus and Tiberius be described as ‘nouo more,’ if it merely opened the senate to the towns of Italy? No one, I imagine, will deny that in Cicero’s time a considerable number of senators were, like himself, new men from the Italian towns. Tacitus, indeed, makes Claudius himself speak of men, ‘ Etruria Lucaniaque et omni Italia in senatum accitos,’ at a date before Italy was extended to the Alps (Amn. XI. 24). (2) Why in the very next sentence of the set speech—for Mr. Cunningham himself admits that ‘quid ergo . . . puto’ is a digression answering an objection—immediately following upon the phrase ‘coloniae et municipia,’ does Claudius take as an instance the colony of Vienna? Surely the only reason can be that it fell under the colonies just alluded to. Mr. Cunningham concludes with exhorting any with whom my theory may have found ¢ of Pro! mine u relingu chiefs c the zus With b that all ineligib might | quaesto compet agree t had alr these cl It venture When | oo the disz stripe, a strikes in Profe Gauls, z chiefs w provinci to this 1 was cle: and eve Asiaticu But and Tibs these ch attitude Sure to grounds, policy of would h: believe t the situa such fror by reaso1 those wh and esse with its organisat us, is it 1 responsik to be a Mr. Cun diplomat Augustus incials n the m the 19 A.D. v does Rome under It is berius nning- at this oposal a and tegere »sal is sal to ticular slarum de his ‘ipated ed his is that ild ‘ be in the simple sses of of the a ‘free senate- nies or right). e same actual ejudice sessors: r. Cun- igustus > towns nber of _ makes enatum Why in its that llowing ice the ies just ay have CLAVDIVS AND THE PRIMORES GALLIAE 287 found acceptance, to return to the view of Professor Pelham. Fortunately the value of Professor Pelham’s view does not depend on the success of his supporter to prove mine untenable. Nor is there any occasion to return to what no one need have relinquished. Professor Pelham showed that the citizenship granted to the Gallic chiefs could not have been anything analogous to the old ciuitas sine suffragio, and that the ius honoru» could not have been separated from the other rights of the ciuttas. With both these ,oints I am in entire agreement. I also agree with his contention that all provincials and almost all Italians were on exactly the same footing as being ineligible for the senate, unless the emperor did for them one of two things. He might give them the /atus clauus with which they could become candidates for the quaestorship, and so, if elected, pass into the senate; or, he might by his censorial competence adlect them at once into the senate ‘inter quaestorios.’ We further agree that by one or other of these methods provincials from Narbonensis and Spain had already got into the senate, but that so far no emperor had come to the relief of these chieftains from Gallia Comata or their ancestors. It is in the explanation why the Gallic chiefs were treated differently that I venture to think Professor Pelham’s statement ambiguous and in need of supplement. When he says ‘the real obstacle in the path of these noble Gauls was the fact that . . . they had not the broad senatorial stripe,’ it is obvious that he is merely stating the disability and not explaining it, for why did not the emperors give the broad stripe, as they did to other provincials? But though the sentence I have quoted strikes one as insufficient, I think, from reading his paper again, that the real obstacle in Professor Pelham’s opinion was the deep seated Roman prejudice against these Gauls, and the fact that, through their wealth and the number of their retainers, the chiefs were more powerful and possibly more dangerous personages than the ordinary provincial aspirant to the senate. Like Mr. Cunningham, I ‘see no serious objection to this view,’ though I would note that the prejudice, if stronger against the Gauls, was Clearly felt against all provincial candidates, and that the possession of wealth and even of retainers (gentsles) had not prevented the advancement of Valerius Asiaticus. But even if we assume, as I think we may, with Professor Pelham, that Augustus and Tiberius identified themselves with this prejudice, and regarded the admission of these chiefs to senatorial dignity as an experiment too hazardous to be tried, what attitude were they to take up towards the chiefs themselves, some of whom would be sure to ask for what other provincials received? To put a refusal on the real grounds, if these were the real grounds, would have been too inconsistent with the policy of conciliation towards Gaul, of which Augustus gave so many proofs, and would have turned these powerful chiefs into open malcontents. ‘This is where I believe that my theory may usefully supplement Professor Pelham’s explanation of the situation. How was the new imperial principle of not excluding provincials as such from an official or senatorial career to be reconciled with the necessity, dictated by reasons of state, for at least marking time in the case of Gallia Comata? Even those who do not accept my hypothesis will not deny that there was a very marked and essential difference between the administrative organisation of the Tres Galliae with its practical absence of colonies and municifia and the thoroughly urban organisation of Italy, Spain, and Narbonensis. But if this distinction is so patent to us, is it too far-fetched to suppose that Augustus, who more than anyone else was responsible for it, should have used it as a solution for what can hardly have failed to be a delicate question? To me, at any rate, the principle, as set forth, pace Mr. Cunningham, by Claudius, appears in the light of a solution not unworthy of the diplomatic finesse, which from first to last was the most conspicuous characteristic of Augustus. There was not a word about exclusion or disqualification ; it was a broad 288 CLAVDIVS AND THE PRIMORES GALLIAE and generous manifesto of inclusion. More than this; it was obviously and logically the next step forward in advance of the course taken in this matter under the late Republic, when senators had been freely admitted from the municipalities of Italy. Suitable men were now to be eligible from colonies and municipia, wherever they were found. There was no indication therefore of any intention to bar the way against anyone. But when from time to time Gallic chiefs demanded what Augustus was not prepared to give, it was possible to point out with tact and politeness that their ciuitates, having been expressly organised in such a way as not to stifle national life or national aspirations, did not exactly fit in with the municipal system usually adopted, and on which unfortunately admission to a career in Rome had been made to depend. This may not always have satisfied ambitious chiefs, but at least they were made to feel that it was the imperial system which blocked their way, and not considerations which would have wounded their individual and national susceptibilities, It was very easy, too, as Claudius proved, in spite of senatorial prejudice, to remove the obstacle whenever it seemed advisable, and it can hardly be doubted that the first step taken in 46 a.D. towards its removal was wise and statesmanlike. E. G. Harpy. Jesus College, Oxford. THE PAEAN OF PHILODAMOS OF SCARPHEIA. LinEs 53 to 56 stand thus in Weil, Bulletin de Corr. Hell. xix. 393 sqq., and Weir Smyth’s Greek Meltc Poets, p. 525: ["E]v[ Gev €]m’ oABias xGovds 5? OeA[ Evvdas] exeAoas, a Urioe pevos t(e) "OAvprifas] 55 |’Efop|iay re kAacrav. Weil supposes this to mean ‘the land which ended the anger of the Olympian goddess, and Dionysos’ famous exile’: (the context speaks of his wanderings); and the goddess he takes to be Hera. But in 1. 55 the stone has 2THTE, and orjee is Weil’s correction of what he supposes to be an error: the letters in biackets are also his restoration. All this is far-fetched, and dependent on this alteration in the text: efopia also is a late and rare word. ‘The stone has preserved the reading correctly EKEASASA | STHTEMENOSTE, that is, éxeAcas Gorn Téuevds te, which Weil has divided wrongly. We may then restore the lines thus : "E|v[Gev €|m’ orASias xPovds 53 Gea[caXias| ExeAcas a- orn, Teyevos T ’Odvprifov, 55 [ ILcep |tav Te KAELTaV. OeooaXias, "OAvpriov and Ilepiav suit the context. There appears to be some uncertainty how the third letter in 1. 54 should be read: Weil, p. 405. Tepevos ‘OAvpriov I take to mean ‘the sacred region of Olympos.’ For Dionysos and Pieria cf. Eur. Bacch. 565, paxap & IIvepia, weBerai o’ Evuos. J. U. Powe t. St. John’s College, Oxford. Lr 453 (t sustained made for Iliad sinc speeches omitted wv to have | papyri. § syntactic Partictple to the pa Berli June (Eberharc (Ammon). Nasonis W highly prz June The revie Grundriss (Lenschau (down to a June : vomischen F highly pra: par la Di (Oehler). June 15 of whic Most of tk R. Delbrii geschichtli 119 Textal under Gree July 4 Textgeschi L. Dalma rically e late Italy. y were gainst 1S was t their al life sually - made t they 1d not ilities, emove at the RDY, q-, and 5? 55 ympian ); and rT Hoe iS re also e text: actly y then 53 55 > some Tépevos | Pieria VELL. SUMMARIES OF PERIODICALS. LITERATURE AND GENERAL. American Journal of Philology. Vol.35. No.2. 1914. G. M. Bolling, The Archetype of Our Iliad and the Papynt. It is probable that all the MSS. of Homer come from a single archetype, though Cauer’s argument from I 453 (based on the necessity of accepting the correction éxevfov dv) cannot be sustained. The study of the vulgate papyri shows that (when allowance has been made for their accidental omissions) a number of lines have been inserted in the Iliad since 150 B,c. About one-fifth of these lines omitted by the papyri introduce speeches after a verb of speaking has already been employed. All the rest can be omitted without detriment, and are omitted in many MSS., and none can be proved to have been in Aristarchus’s text. Lines only athetized by him do appear in the papyri. E. W. Fay, Varvoniana de Lingua Latina. Pavi J. A sheaf of notes critical, syntactical and etymological on book V. of Varro’s treatise. R. B. Steele, The Participle in Livy. A miscellany of details of usage and statistics relating mainly to the past and present participles. Berliner philologische Wochenschrift. 1914. June 6. Adolf Clausing, Kvittk und Exegese dey homerischen Gleichnisse im Altertum (Eberhard). H. Mutschmann, Tendenz, Aufbau und Quellen dey Schrift vom Erhabenen (Ammon). O. Hartlich, De Galent “Yyvevav libvo quinto (Kind). H. Magnus, P. Ovidi Nasonis Metamorphoseon libyi XV, rec. app. crit. instr. H. M. (Ehwald). Very highly praised. June 13, O. Kramer, C. Valent Flacct Argonauticon libs VIII, ed. O. K. (Helm). The reviewer discusses the Praefatio at some length. W. Strehl und W. Soltau, Grundriss dey alten Geschichte und Quellenkunde. 2. verm. u. verb. Aufl. Bd. I (Lenschau). A good guide to controversial questions and the literature about them (down to about the middle of 1913). June 20. E, Taubler, Imperium Romanum. Studien zuy Entwickhlungsgeschichte des vomischen Reiches. I. Band: Die Staatsvertvige und Vertvagsverhaltnisse (Soltau). Very highly praised. A. Merlin, Forum et églises de Sufetula. Notes et Documents publiés par la Direction des Antiquités et Arts. Mit 5 Tafeln und 11 Abbildg. im Text (Oehler). Shows the results of the work done by the French from 1906 to 1911. June 27. O. Hirschfeld, Kleine Schriften (Wissowa). Contains 74 papers, 15 of which were read to the Berlin Academy, but have not been printed before. Most of these latter deal with the Republican period, some with the early Empire. R. Delbriick, Hellenistische Bauten in Latium. Bd. II. Baubeschreibungen und geschichtliche Erlauterung (Késter). An expensive book (58 M.) with 22 Tafeln, 119 Textabb. ; it shows how Roman building developed in the last two centuries B.c. under Greek and Oriental influences. July 4. K. B. Bitterauf, Dev Schlusstetl dey Aristotelischen Biologie. Beitrage zur Textgeschichte und Textkritik der Schrift ‘de generatione animalium’ (Stadler). L. Dalmasso, (1) La formazione delle pavole in Palladio; (2) Il vocabolario tecnico U 290 SUMMARIES OF PERIODICALS di un tavdo scrittore georgico. Contributo alla semasiologia palladiana (Schmaiz), W. Riezler, Weissgrundige attische Lekythen. Nach A. Furtwanglers Auswahl. Text- band: XI, 143 S. 56 Abb. Tafelband: 98 Tafeln. 300 M. (Pagenstecher). V. Macchioro, Le Teyvme romane di Agnano (Behn). K. Brugmann, Gmechische Gram. matik. Vierte verm. Aufl. bearb. von A. Thumb (Meltzer). Recommended to all teachers of Greek, not only to the specialist in the history of language. July 11. J. T. Allardice and E. A. Junks, An Index of the Adverbs of Plautus (Schmalz). It is a pity that only the Oxford text has been used. W. M. Lindsay, S. Pompei Festi De verborum significatu quae supersunt cum Pauli epitome. Thewrew- kianis copiis usus ed. W. M. L. (Goetz). Praised. F. Délger, Konstantin dev Grosse und seine Zeit. Gesammelte Studien (Bauer). Nineteen papers by various writers. E. Buschor, Griechische Vasenmalevet (Behn). Very highly praised. July 18. E. Sittig, De Graecorum nominibus theophoris (K. F. W. Schmidt). Important for the study of Greek religion. Seeks to trace by such names as Avoyeizwv the original site and the extent of the worship of various gods. The author thinks that Athena came to Athens relatively late from Boeotia; the reviewer says that the name of the goddess and of the town date, like the names of the rivers and mountains of Attica, from a time before the coming of the Greeks. E. Babelon, Moneta (Kiithmann). Discusses the origin of the word and the development of its meaning. J. Wackernagel, Uber die Geschichte dev griechischen Sprache (Meltzer). The reviewer gives a summary of this interesting ‘ Festrede.’ July 25. R. von Péhlmann, Die Weltanschauung des Tacttus (Bardt). Second edition, considerably enlarged. T. Sauciuc, Andvos. Untersuchungen zur Geschichte und Topographie der Insel (H. v. Gaertringen). Valuable. Classical Philology. Vol. 8. No. 3. 1913. R. C. Flickinger, Tragedy and the Satyric Dvama. Defends the ‘goat-song’ etymology and rejects the ‘goat-men’ speculations. A.Shewan, The ‘Continuation’ of the Odyssey. Detailed defence of the genuineness of Od. 23. 296 to end so far as relates to A ‘ Language and Verse.’ E. W. Fay, Catullus carmen 2. In favour of the tradition on the whole. E. B. Clapp, The Pindanc Poems of Theocritus. On Theocr. xvi, xvii. R.G. Kent, The Vedic Path of the Gods and the Roman Pontifex. The pontifex made or kept in order the fontes (‘ paths’) between the worlds of the living and that of the gods and the dead. J. P. Postgate, Propertiana. On five passages of books III and IV. E. H. Sturtevant, Studses in Greek Noun-Formation. On nouns in-y. J. A. J. Drewitt, A Note on the Augment. Max Radin, The Wife of Gaius Gracchus and her Dowry. P. Shorey, On Aristotle’s Nicomachean Ethics vii. 7, II49B, 31 Sqq. Vol. 8 No.4. 1913. E. T. Merrill on Cic. Fam. xv. 20, Verg. (?) Catal. 10 and Ventidius. ‘There is no allusion to Ventidius. F. E. Robbins, The Creation Story in Ovid Met.i. His account is essentially Stoic. Lynn Thorndike, A Roman Astrologer as a Htstorial Source. On Julius Firmicus Maternus. C.C. Mierow, Adverbial Usage in Eugippius. J. A. Scott, The Assumed Duration of the War of the Hiad. R. B. Steele, The Future Peviphvastic in Latin. P. Shorey, Note on Aristotle, Ethics ii. 3. 5 (1104 5 21); S. E. Bassett, ravradwGeis in Soph. Ant. 134, means ‘treated like Tavrados’ ‘ hurled down.’ J. A. Scott, A Note on Herodotus i.86. The expression cxoiv diaperpijcacba is explained from II Sam. 8. 2 (LXX). Vol.g. No.1. 1914. B. L. Ullman, Dramatic ‘Satura.’ A discussion and interpretation of the well- known passage of Livy (VII. 2 §§ 1-12), which gives an account of the origin of both tragedy and comedy. Livy in his wish to produce a logical scheme of the develop b (the 2 closely | from th Fabius other e A. Shev case ag with Od 534 535 the Mace the dep (2) with binding A full t example rence in 2nd cent Vol. Mar nine illus open col Tuscantcu Examine and Ecm of origin: and Latis Odyssey. tion of H to Hades A. R. An for interr is a devel ne (nt). , T. Frank of the Eq Vol. Kend of the ins with a ne’ to patrici:; employme A. C. Joh with a sk On its sim The insta of the ver! feeling. \ Egypt. A ship to th Under Ex explanatio. hmaiz), Text- techer), e Gram- d to all Plautus _indsay, 1eWrew- oy Grosse writers. =hmidt). mes as e author eviewer le rivers 3abelon, nt of its r). The Second schichte at-song’ tation’ of 0 far as avour of tus. On Pontifex. ls of the On five ovmation. e Wife of CS Vil. 7, There .i. His Tistorical ugip prus. he Future 4b 21); > « hurled pioar Ba the well- origin of 1e of the SUMMARIES OF PERIODICALS 291 development of the drama has brought together two pairs: a (the Etruscan dances), b (the zocularta, § 5); and c (the satuvae, § 7), d (the fabulae), §§ 8 sqq.), which are not closely related, by linking c tod. The burlesques mentioned by Livy gain support from the account of the pompa of the ludt Romam in Dionysius Hal. VII. 72 (from Fabius Pictor). Allen B. West, Formation of the Chalcidic League. Numismatic and other evidence shows that there was a real Chalcidic league formed about 432. A. Shewan, The ‘ Continuation ’ of the Odyssey. B. Imitation. Summarizes the general case against the view that repetitions are marks of spuriousness ; and deals in detail with Odyssey 24. 39 Sq., 156-8, 165 sq., 235-40, 248, 368 sqq., 397 Sqq., 410, 479 Sqq., 534, 535 (Agar’s omi, Homertca 152 sq.). Tenney Frank, Representative Government in the Macedonian Republics. On the reconstitution of Macedonia by the Romans after the deposition of Perseus (167). There was (1) a representative body (the boule), (2) with legislative power, (3) whose ordinances, like those of modern States, were binding over the territory. A. R. Anderson, Studies in the Exclamatory Infinitive. A full treatment of the construction from Naevius to Horace, supplemented by examples from prose writers of that period and the principal instances of its occur- rence in Greek. Tracy Peck points out that the centre of the book-trade in the 2nd century A.D. was in the Vicus Sandaliarius. Vol.g. No.2. 1914. Margaret C. Waites, The Form of the Early Etruscan and Roman House (with nine illustrations). The Etruscan house, after the hut-stage, was built round a large open court. Some time between the First and Second Punic Wars the Atrium Tuscanicum was invented. G. M. Calhoun, Documentary Frauds in Litigation at Athens. Examined and exemplified under the heads Wills, Other Private Documents, Depositions, and Ecmartyriae, Official Documents, Miscellaneous. In actual practice the production of original documents was not regarded as very important. Francis A. Wood, Greek and Latin Etymologies. On fifty-nine words. A. Shewan, The ‘Continuation’ of the Odyssey. C. The NEKYIA. Defence of this against objections based on the descrip- tion of Hermes as KvAAjjvios and as Yvyxoropmres, and his conduct of the Suitors’ souls to Hades before burial, the Land of Dreams of v. 11, the nine Muses of v. 60, etc. A. R. Anderson, The Unity of the Enclitic ne. M. Warren’s theory of a dual origin for interrogative aud affirmative (confirmative) mé cannot be maintained. The latter is a development of the former, which is to be connected with the negative particle ne (nt). A. Shewan discusses (against J. J. Drewitt) ‘the Homeric Augment again.’ T. Frank has a note on Pompey’s ‘ expansionist’ policy and support of the interests of the Equites in B.c. 70-66. Vol.g. No.3. 1914. Kendall K. Smith, A New Fragment of the Decree of Cremonides. Reconstruction of the inscriptions in Kirchner’s revised volume of Attic inscriptions Nos, 686, 687 with a new text. Max Radin, Gens, Familia, Stivps. In Livy’s use gens is restricted to patrician families; familia is applied to both patrician and plebeian. Cicero’s employment is not quite so strict. Stivps is never used for a division of a ‘ gens.’ A. C. Johnson, Attic Avchons from 294 to 262 B.c. A new chronological arrangement with a sketch of the history of the period. J. J. Schlicher, The Htstovical Infinitive. On its simple form (denominated the infinitivus tmpotentiae) as it occurs before Sallust. The instances are collected and arranged under six heads, according to the meaning of the verbs. Its primary function is to express the effec: of a strong impulse or feeling. W.L. Westermann, Apprentice Contracts and the Apprentice System in Roman Egypt. A connected presentation of the usages and regulations affecting apprentice- ship to the various réxvace (i.e. trades or professions) as deduced from the papyri. Under Execution of a Vestal and Ritual Marriage E. T. M. criticizes Miss Harrison’s explanation of the ceremonial attending an erring Vestal’s immurement. 292 SUMMARIES OF PERIODICALS Deutsche Literaturzeitung. 1914. June 6. H. Bohlig, Die Getsteskultur von Tarsos (Stahlin). The carefully collected material is insufficient to enlighten us as to St. Paul’s upbringing. The author's scholarship is very defective. G. Lippold, Griechische Portvatstatuen (Hekler). The portraits of Epicurus, Metrodorus and Hermarchus are now established. June 13. J. Scham, Der Optativgebrauch ber Clemens von Alexandrien (J. H. Moulton). The frequency of the mood shows that Clemens wrote for the learned world. F. Krohn, Vitruvit De Architectura libri decem, ed. F. K. (Kraemer). Includes an interesting introduction. June 20. P. Rabbow, Antike Schriften tiber Seelenhetlung und Scelenleitung (Ringel- taube). This first part deals with the treatment of Anger, and shows that Seneca’s treatment in Bk. 2 of the De Iva depends on Posidonius, and Bk. 3 is a rhetorical development of Bk. 2. W. Brewitz, Scipio Africanus Mator in Spanien (Grosse). A careful dissertation. June 27. O. Hense, L. Annaei Senecae Epistulae, ed. O. H., and J. Buck, Seneca De Beneficiis und De Clementia in dey Ueberlieferung (Bickel). Hense’s second edition has the advantage of deducing the text of older MSS. from their later transcribers. Buck endeavours unsuccessfully to maintain the authority of the Nazarianus. July 4. L. Laurand, A propos d’Homéve (Max Wundt). A clear statement of the modern suspicion of Homeric criticism. W. Helbig, Fuhrer durch die offentlichen Sammlungen klassischer Altertiimer in Rom. 3 Aufl. (Pollak). This indispensable work is now much enlarged. A. Steiner, Dey Fiskus dey Ptoleméer (Zucker). Untrustworthy. July 11. V. G. Jaeger, Avistotelis De Animalium motione, De Animalium incessu; Ps.-Ay. De spiritu; ed. V.G. J. (Mutschmann). An excellent edition by the scholar who has proved the genuineness of the rept xivyjcews. July 18. J. Charpentier, Die Desiderativ-bildungen dev indo-tvanischen Sprachen (Gintert). Not helpful. C. Halm, P. Cornelit Taciti libri qui supersunt, rec. C. H., ed. quintam cur. G. Andresen. I. (Wackermann). Has collected all valuable con- tributions to the literature of the Ab Excessu of the last thirty years; includes an excellent reasoned Index Historicus. July 25. H. Mutschmann, Sexti Empivict opera I., rec. H. M. (Max Wundt). This now becomes the standard edition, Aug. 1. J. Schroeter, Plutarchs Stellung zuv Skepsis (Falter). Shows that P.'s criticism of the senses is influenced by the physicians and the Pyrrhonists. A. Riese, Das rheinische Germanien in den antiken Inschriften (Drexel). Includes paraphrases of some 5,000 inscriptions, either found in Germany or referring to it. Hermes. Vol. 49. Part 3. 1914. J. Geffcken, Die Hirten auf dem Felde. Christian and Mithraic legends of the Adoration. Virgil was influenced by Poseidonios. W. Soltau, Dey Uvsprung der Dtktator. He was originally the com- mander-in-chief of the Romano-Latin league. Cl. Lindskog, Zur Uberlieferungs- geschichte dev Biographien Plutarchs. A discussion of the tradition of the text of the first section of the Lives. Contains a good analysis of typical interpolations, which are shown to be often unconscious. Chr. Jensen, Zu dem Menanderpapyrus in Kavwro. C. Robert, Das Oxyrhyncosblatt dey Epitrepontes. U. von Wilamowitz-Méllendorf, Der Feldzugsbericht des Ptolemaios Euergetes. F. Jacoby, Eine vergessene Hovazemendatin. Revives the emendation ‘ Othone contento’ in Epod. iv. 16 from the editio Veneta of 1478. W. Hoffa, Texthritische Untersuchungen zu Senecas Tragédien. A discussion of the MSS. of the interpolated A-group with notes on a number of passages. MisceLten. Th. Néldeke, ATBATANA. A defence of the Greek spelling with the spiritus lenis. FF. W. Hall, Ad Ovid. Met. XV. 411. Restoration of a medieval An examination of the Virgilian, interpc read av G. Wi Hymn: not Hi Mi J. Fa Ses Hartm An uit conside postion a poet Patrocl part of facta s later v NZO is greater of thes books a Inseript the Arc Leuctra Cicevoni: who mz propera. Ne) R. | of ancie later w« psycholc Menippi and soci Askalon. XX A. i forms of logograp drine po Apuleius held to i ‘clever j enthusias in itself j de Rome. Les Prét¢ ollected author’s ‘). The (J. H. learned Includes ( Ringel- Seneca's hetorical sse). A k, Seneca ; second eir later y of the ement of ffentlichen ble work stworthy. 1imalium lition by Sprachen c. C. H., ible con- sludes an Wundt). that P.’s A. Riese, hrases of Virgilian, nced by the com- lieferungs- xt of the ns, which > im Kasro. sllendorff, mendation. o Veneta liscussion passages. r with the medieval SUMMARIES OF PERIODICALS 293 interpolation in Cod. Augustanus, M. Wellmann, Zu Philumenos 4.14. For adpos read av@pwros (avos). F. Bechtel on Xen. Hell. 11.3. 10. For ’Ayyevidas read ’AyyeAiéas. G. Wissowa, Zu Hor. Sat. 1. 10. 5%. Read multum puerum, O. Kern on Orphic Hymns 48 and 49. The conjecture that the name of the nurse of Bacchus was Hipta not Hippa, is confirmed by cod. Laurentianus 32. 45. Mnemosyne. 42. 3. J. L. V. Hartmann, Ad Platonis Rempublicam. Notes on text of Bks. I.-V. P, H. Damsté, Lectiones Valevianae (cont.). Notes on text of Bks. VII.-XI. J. J. Hartmann, Ad Plutavchi Moralia annotatwnes criticae. Notes on the De Amore Prolis, An ustisitas ad infelicitatem sufficiat, Animine an corporis affectiones sint petoves. H. considers authorship and points of text (to be cont.). M. Valeton, De Iliadts com- positione (cont.). Valeton now turns to consider the Patrocleta, which is composed by a poet who accepts the first part of the Achilleis, but then brings in the motive of Patroclus’ death in place of the pavidos aroppnors. To the Patrocleta belongs the great part of Bks. II.-X. Then he considers ‘Qua maxime ratione ex Achilleide Ilias facta sit.’ There is first the working together of the Achtllets and Patrocleta, with later writing conditioned by this joining, as the koAds payxy of O. So most of NO is later than both the Tet yooxomwia and Tet yopayia and these than the greater part of TAKE, and all this writing is later than the Patvocleia. Later than any of these are the Nux«tepyeoia and AoAwveia, And scattered over the other books are passages worked into the older material (to be cont.). W. Vollgraff, Novae Inscriptiones Argivae. Discusses an inscription dealing with boundaries of members of the Arcadian federation and fixes many points of topography. The date is soon after Leuctra, and the delimitation is made by the Argives as referees. J. J. H., Ad Ciucevonis ad Attic. XIV. 18.§$1. For the unknown Fabertit read fabri, the craftsman who made duplicate keys for Dolabella. Ad Tacit. H. 11. 25; for prospeva read propeva. Neue Jahrbucher fiir das klassische Altertum, ete. XXXIII. 6. 1914, R. Petsch, Die Lehve von den gemischten Gefiihlen im Altertum. A critical account of ancient views on mixed emotions, chiefly Plato’s and Aristotle’s, on whom all later work depends. Their faults are due to the pre-scientific roots of Greek psychology. K. Mras, Vavvos Mentppeische Sativen und die Philosophie. Varro follows Menippus’ form, but does not translate him. Unlike Menippus, he respects science and society, and is tolerant in philosophy, following his eclectic master Antiochus of Askalon. XXXIII. 7. ror. A. Hausrath, Die tonische Novellistik. A study of the development of the shorter forms of ancient prose fiction: the early lonic type found in Herodotus and the logographoi, akin to the folk-tale: a type reflected in the New Comedy and Alexan- drine poetry: and the Aristides type, the connection of which with Petronius and Apuleius is discussed. Stress is laid on the Farnesina house paintings, which are held to illustrate a cycle of picaresque stories, mostly unknown, grouped round the ‘clever judge’ motive. H.F. Miller, Plotinos tibey Notwendigheit und Fretheit. An enthusiastic exposition of Plotinus’ religious and philosophic views. Revue de Philologie. Vol. 38. Part 1. 1914. G. Ramain, Sur l'emploi de Vinfinitif historique. ‘The historic infinitive does not in itself imply any idea of time.’ D. Serruys, Un fragment de la Chronique d’Hippolyte de Rome. Pap. Oxyrhynchus 870=Hippolytus ed. Bauer, pp. 104-107. P. Fabia, Les Pvétoviens de Vitellius, exegetical notes on a number of passages from the Histories 294 SUMMARIES OF PERIODICALS of Tacitus. E. Cavaignac, Polybe VI. 19. 2, et la durée du service militaire a Rome. Supports Steinwender’s view that <£ should be retained. L. Havet, Notes cratiques suv les Bucoliques de Virgile. H. Grégoire, Aesch. Eum. 506 sqg. Supports orevoera for wevoerar. J. E. Harry, Eur. lon 1424. Read idov, rod’ eo? tgacp’—a 7’ eordbas evpioxopev. R. Waltz, Notes critiques, on Seneca, Tacitus, Petronius. Vol. 38. Part 2. 1914. H. de la Ville de Mirmont, Les personnages épisodiques dans l’ Andrienne de Tévence. H. Grégoire, Notes sur l’Héléene d’Eunpide. Reads for 390-1, €i8’ apeAres ror’ Hes Aureiv Biov; 619, Keprouetv= ‘to mystify.” Corrections in 1321, 1353, and in Bacchae 983, 1002-3. R. Cahen, Ad Bucol. i. 70. L. Havet, Bucoliques iii. too. Reads ‘Eheu, guom....’ L. Maries, Diodore, ’Ard gwvis ’Avacraciov. “he Commentary of Diodorus of Tarsus on the Psalms was provided with marginal extracts by Anastasios. G. Lafaye, L’dne et la vigne. In Hyginus Fab. 274. 1 read suauitatem inuenisse. G. Mathieu, Survivances des luttes politiques du Vé stécle chez les ovateurs attiques du IV¢. L. Bayard, Notes sur la Vita Cypriani e¢ suv Lucianus, W.H. Buckler, La déesse Julie, CIG. 2815 et 3642. These refer to Livia, and not to Julia Domna. D. Serruys, La notation ascendante des nombres dans la Chronique d’Euscbe. ‘ Eusétbe avait adopté l’usage palestinien de la notation ascendante, mais, dés avant le VI¢ siécle les copistes avaient substitué la notation descendante.’ Rheinisches Museum. 69. 3. F. Jacoby, Dvet Gedichte des Properz (cont.). Elaborate discussion of II. 24. A. (Phillimore, II. 24. 1-16), with particular reference to v. 4, where the sense requires aut pudory ingenuus aut vetinendus amor, a true Propertian sentiment: also of III. 8, which is to be interpreted by itself and not taken closely with III. 6. E. Lattes, A che punto siamo coll’ interpretazione dei testi etruscht ? Ingenious interpretations based on closing formulae in dedicatory inscriptions. A. Schulten, Das Pilum des Polybios. Comparison of Polybius’ account with finds in the recent excavations of Numantia. W. Bannier, Wiederholungen bet dlteren gniechtschen und lateinischen Autoren. Repeti- tions in authors collected and classified; defence of text in some passages where repetitions occur, and warning against cutting out words which offend our taste more than the ancients’. J. M. Stahl, Zum Aias des Sophokles. Defends vv. 961-970 (reading ov for 7 in g16) and 1105-1117; in v. 1312 read avrovd for 7 Tov, in 799 éeXaiferv Opoet for eAmifer pepe; in 1081 wapy is called for by construction of meceiv. J. Dietze, Zur kyklischen Theogome, Argues (against Aly) that the Apollodorus version reflects a later stage of development than Hesiod. The Titanomachy is also later than Hesiod, and is a working over of the 7 heogony and the basis of later Orphic speculation. O. Kénnecke, Zu den griechischen Bukolixern. Notes on Theocritus, Epigr. 22, Id. III. 28-30, VII. 5, XV. 7 (read Exagrépw aiév dmorxeits ‘each time you move you go further away’); XV. 15-7 (read with Ahrens Gavra... adyopaadecv); XXVII. 20-1; Mosch, 11. 37 sqq., on the golden basket of Europa. Notes on text of Mosch. I1., III. and IV., of Simmias ’Quév 9, 14, and of Dosiades, Bwpds 15, 18. O. Viedebantt, Antike Messungen dev Landenge von Suez: all these go back to old Egyptian measuring (in oyxoivot), very scientifically made. MuiszELvLen: O. Seeck, Das Epigramm des Germanus und seine Ueberschrift : is not good evidence for Julian’s birthday. A. Ludwich, Zu Tryphiodorus, Zu Musaios. Notes on text. H. Heimannsfeld, Zum Text des Helladius bet Photius (cod. 279). H.Schenkl, Dey Dichter dey Ilias Latina. To show real existence of Baebius Italicus. A. Klotz, Zu Cic. pro Milone 2. To refute Birt’s assumption of a considerable lacuna. FE. Hohl, Reste einer Handschrift des Kollektaneums des Sedulius Scottus in Paris. A. Brinkmann, Liickentisser. R \ of Ori polatic contai the lo adduc Heges Episod the nz similal incide! using Tevence the Az Pevint] the ch: and th U. Mo view t] of Luc W Ju hensive Stand | (H. G municep Blinke at Linc Ju A. E. | defence ae 1 Ju introdu ad Luc unlimit (A. Ros Ju Les écy useful.’ Jul exegetict Lateinis to Nied Jul ‘A co von M. a Rome. orstiques TITEVTET UL S| y 4 > eorrabas e Tévence, oT 1 0e0s 1 Bacchae Reads imentary racts by uarum> stecle chez Lucianus, and not Chronique ite, mais, _? fe LI. 24. A. - requires f ILI. 8, ‘. Lattes, ons based ; Polybios. Jumantia. Repeti- es where aste more . 961-970 »v, IN 799 ‘uction of ollodorus hy is also er Orphic heocritus, és ‘each VTGMA eas ' Europa. Dosiades, | these go SZELLEN: dence for text. H. ey Dichter u Ctc. pro Reste einer chentnisser. SUMMARIES OF PERIODICALS 295 Rivista di Filologia e d’Istruzione Classica. Vol. 42. No. 3. 1914. V. Ussani, The Earliest History of Josephus’ Text (paper read before the Congress of Orientalists at Athens in 1912). In addition to Christian and ‘ apologetic’ inter- polations already recognized in parts, Josephus’ works, especially the Bellum Iudaicum, contain alien importations of anti-Semitic tendency, in many cases probably from the lost Hzstory of Pliny, used by Tacitus in his Histories. Interesting evidence is adduced from comparisons of the Greek text with the Latin version of the so-called Hegesippus and the Slav translation of Josephus. Giuseppe Procacci, On an Episode in Stlius Italicus (VII. 162-211). The passage, an abrupt digression from the narrative, describes a manifestation of the divine power of Bacchus in very similar language to that used by Nonnos, Dionys. XLVII. 34 sqq. in regard to the incident of Icarius and Erigone; and it is suggested that both writers have been using the (lost) Evigone of Eratosthenes. P. Fossataro, Menander’s Perinthia in Tevence’s Andria. F. Scholl’s views on the contaminatio of the Andria of Terence from the Andria and Perinthia of Menander should be accepted. Terence took from the Pevinthia Andr. Act 1. Sc. i., the midwife scene (cf. 228, 332= Fragm. Per. 397 Kock), the character of Byrria, the punishment of Davus, Fragm. Per. Oxyrhynchus Pap. 855 and the character of Charinus. Also probably the doypa ’Emixotpeov of Andr. 959-60. U. Moricca, On Lucian’s rXoiov 7) evxat (conclusion). Further development of the view that it is a satire on human desires in general, with comparison of other works of Lucian and citation of parallels from Greek and Latin literature in general. Wochenschrift fur klassische Philologie. 1914. June 8. G. Lippold, Grtechische Portratstatuen (H. Lamer). ‘Shows compre- hensive knowledge and independent judgment.’ O. Schroeder, Uber den gegenwiirtigen Stand dey gviechischen Verswissenschaft (J. Sitzler). O. Kraus, Platons Hippias Minor (H. Gillischewski). ‘Rich in contents.’ F. Geiger, De sacerdotibus Augustorum munictpalibus (M. Gelzer). June 15. W. H. Roscher, Die hippokvatische Schrift von dev Siebenzahl (W. Nestle). W. H. Roscher, Omphalos (W. Nestle). ‘A well-thought-out exposition of the evidence.’ E. Obst, Dey Feldzug des Xerxes (H. Philipp). ‘ Valuable.’ C, Blinkenberg, La chronique du temple Lindien (W. Larfeld). ‘The most important find at Lindos.” G. Nieschmidt, Quatenus in scriptuva Romani littens Graecis ust sint (N.). June 22. A. Reinach, L’autel rustique du Mont Phylakas, Crete (H. Lamer). A. E. Kalischek, De Ephoro et Theopompo Isocratis discipults (H. Mutschmann). A defence of the tradition. T.G. Tucker, Plutarch, Selected Essays, translated by T. G. T. (J. Tolkiehn). ‘ May well fulfil its object.’ June 29. K. F. Smith, The Elegies of Albius Tibullus (G. Friedrich). ‘ The introduction is suitable for school use, but not the commentary.’ O. Hense, Senecae ad Lucilium epistularum moralium quae supersunt ed. O. H.(W.Gemoll). ‘ Deserves unlimited praise.’ Th. Mommsen, Epigvaphische und numismatische Schriften. I. (A. Rosenberg). July 6. J. Schaefer, De Jove apud Cretas culto (H. Steuding). P. Marestaing, Les écrituves égyptiennes et l’antiquité classique (A. Wiedemann). ‘ Very learned and useful.’ A. Gercke, Dey Christenname ein Scheltname (O. Stahlin). July 13. J. van Leeuwen J.F., Ilias cum prolegoments, notts criticts, commentaris exegeticis, Libri XIII.-XXIV. (F. Stiirmer). E. Ernout, Historische Formenlehre des Lateinischen. Deutsche Ubersetzung von H. Meltzer (R. Wagner). ‘ A continuation to Niedermann’s Historische Lautlehre des Lateinischen.’ July 20. A.Elter, Ein athenisches Gesetz tiber die eleusintsche A parche (W. Larfeld). ‘A convincing exposition.’ W. Kopp, Geschichte der vomischen Literatur. g. Aufl. von M. Niemeyer (H. Gillischewski). 296 SUMMARIES OF PERIODICALS July 27 (double number). W. Siiss, Aristophanes, Die Frésche, mit ausgewahlten antiken Scholien, herausg. von W. S. (R. Wagner). B. Schmidt, De Cornuti theolo- gicae graecae compendio capita duo (H. Steuding). P. Sternkopf, De M. Tullu Ciceronis partitionibus ovatorus (C. Atzert). ‘Both learned and rich in results.’ J. Nye, Sentence connection, tllustvated chiefly from Livy (Ed. Wolff). ‘Not without value.’ G, Norlin, The conventions of the pastoral elegy (Ed. Wolff). LANGUAGE. Glotta. V. Band, 4 Heft. 1914. This number, which was published (May 27) after the appearance of the first instalment of the sixth Band (April 14), contains nothing but indexes and a very valuable and complete Report on the linguistic books and articles of 1911. One is pleased to note the ever-increasing attention paid to medieval and modern Greek and vulgar Latin, especially the numerous monographs on the language and style of individual authors. Lofstedt’s Philological Commentary on the Peregrinatio Aetheriae and Harmon’s work on The Clausula in Ammianus Mavrcellinus come in for special praise. We have the accustomed crop of etymologies, most of them unconvincing as usual. There is a striking instance of ‘ subjective treatment’ in a German work on the criticism of Horace, which the reviewer illustrates by the following restoration of the First Epode: ‘Ibis Liburnae creditus sub nauium | turrita propugnacula, | curas paratus Caesaris periculo | lenire, Maecenas, tuo. Zeitschrift fur vergleichende Sprachforschung. XLVI. 1. 2. Hans Hessen, Bettrage zur altirischen Grammattkh. J]. Charpentier, Zur arischen Wortkunde. Herman Lommel, Kleim- und Grossuieh. Mainly the history of Gk. aposarov. H. Jacobsohn, Lat. canés. The relation of the fem. nom. endings -és and -is, Canés was preserved through being, at an early date, attached to the conson. stem. A. Fick, Aelteste griechische Stimmverbande. ‘. .. ein Versuch die Urgeschichte der altesten Griechenstamme allein aus den Angaben der Alten herzu- stellen.’ Herbert Petersson, Eznige Tter- und Pflanzennamen aus itdg. Sprachen. J. Pokorny, Keltische Etymologten; Ivische Personennamen auf -le und -re. P. Maas, Koptagos und xoptados. These forms are to be read in Plat. com. fr. 84 and Eurip. Hypsipyle fr. 1 respectively. F. Bechtel, Paverga. E.Schwyzer, Zu den homerischen und avestischen v- n- Stimme. In Gk. and Av. there has been a parallel extension of this type. W. Prellwitz: Homerisch aydeyvjes ‘der Kinstler.’ From apduyvy ‘dexterity ’ abstract to aydiyvos. F. Edgerton, Vedic sabha is sm-bha =col-loguium. F. Holthausen, Lateinische Etymologien. KR. Trautman, Got. mag und slav. mogo. The latter is not a Germanic loanword. W. Schulze, Ueber einige Collectiva. Vangio in C.1.L. vi, 31149 refers to the country of origin (Vangwnes = Wortmatienses). AOHNA transp -a.os anc accusati'’ admissio adoratio; Aeneid, acque ut Agamem 259 4 Agamer Aithiopi Alcibiad altars of ancient | Andersoa Antenor Antoniu: aorists re Apolloni Arethas, Aristoge ‘ ascendi Asconius Atellana Athena, Atheniar Athens < 290 Atticism authentic authorsh lime, } wablten tt theolo- Ciceronis J. Nye, t value.’ the first la very One is reek and style of etheriae r special 1vincing an work toration nacula, | avischen of Gk, ings -és to the uch die 1 herzu- prachen. . Maas, 1 Eurip. nevischen nsion of aupry vy logutium. . mogo. ollectiva. tenses ). INDICES 1L—GENERAL INDEX. A. AOHNAIQN IIOAITEIA, xxx. 3-4, a suggested transposition, 13 ff. -acos and -ecos in Latin poetry, 66 accusative, adverbialized, 229 of exclamation, 61 admission of provincials to the senate, 132 f., 282 ff. policy of Augustus or Tiberius regarding, 285 f. speech of Claudius on, 282 ff. adoration of the shepherds, the, 292 Aeneid, authenticity of first seven lines of, 134 acgue ut, not in Golden Latin, 147 f. Agamemnon, notes on the, 1 ff. 259 sgqg., 1 f. ; 427, 536, 2f. ; 831, 3 f. ; 1028, 4f. ; 1321, 5; 1536, 5 f.; 1651, 6 Agamemnon does not suspect Clytemnestra, 3 f. Aithiopis, the, 139 Alcibiades, 77 altars of dyvwaro Geol, 222 ancient handwriting, 68 Anderson (W. B.), notes on Lucan VIII., 105 ff. Antenor, 17 Antonius Julianus and Tacitus, 67 aorists referring to future, 74 Apollonius Rhodius, notes on, 66 Tyrius, 220 Arethas, 231 Aristogeiton, 78 ‘ ascending notation’ in Eusebius, 294 Asconius, MS. of, 68 Atellanae, the, 140 Athena, provenance of, 290 Athenian archers, the, 67 Athens and other Attic place-names pre-Hellenic, 290 Atticisms in Ionic inscriptions, 217 authenticity of Aeneid I. 1-7, 134 the Halieutica, 267 the NEKTIA, 291 the Pseudo-Vergiliana, 123 ff. the Rhesus, 211 authorship of the anonymous treatise On the Sub- lime, 137 ‘Plato's’ first letter, 64 B. Baebius Italicus, 294 Baehrens on Latin Syntax, 72 battle of Oenoe, date of, 141 Bellum Gallicum, the text of, and the work of I. Meusel, 156 ff. Bennett (Charles E.), notes on Horace, 145 ff. birth-star, influence of the, 258 Boethus of Sidon, 88 ff. Sidon had two Boethi, (2) a Peripatetic, (4) a Stoic, 89 heresies of the Stoic Boethus, 89 their respective dates, 89 f. Booker (8. C.), note on Thucydides vii. 28. 3, 104 Boulé, functions of the, 14 Brugmann’s Grundriss, radical change of plan, 72 Burnett (John), Vindiciae Platonicae I., 230 ff. Butcher (W. G. D.), the caesura in Virgil, and its bearing on the authenticity of the Pseudo-Ver- giliana, 123 ff. butting calf, the, a poetical commonplace, 270 C. Caesar uses passive to describe duration of battle, 161 caesura, the, in Virgil, and its bearing on the authenticity of the Pseudo-Virgiliana, 123 ff. treatment of the, by Ennius, 141 Callimachus and Apollonius Rhodius, 65 Catullus and Mallius, 216 dates of birth and death, 222 Celsus, 219 Chalcidic League, 291 chronology of the Fronto correspondence, 112 ff. Ovid's works, 63 Polybius’ History, 139 Cicero, MSS. of, 68 Clapp (E. B.), on certain fragments of Pindar, 225 ff. Clark (R. T.), notes on Dares and Dictys, 17 ff. Claudian’s ‘ prefaces,’ 217 Claudius and the primores Galliae, 132 f., 282 ff. 298 Clemeut’s Stromateis, on the text of, 237 ff. omissions of two kinds prime cause of errors in, 237, 239 passages transposed in, 240 quotations in, 244 repetitions in, 239 f. colony of Mouum Comum, the, 66 Colson (F. H.), the grammatical chapters in Quin- tilian I. (4-8), 33 ff. composition of the //iad, 138, 221, 293 Consolatio ad Liuiam, date of, 63 Cunningham (H. J.), Claudius and the frimores Galliae, 132 f. ‘Cupid and Psyche,’ story not a folk-tale, 220 Curculio, MS. of the, 216 D. Dares and Dictys, notes on, 17 ff. date of battle of Oenoe, 141 birth and death of Catullus, 222 Consolatio ad Liutam and Elegiae ad Mae- cenatem, 63 Epigrammi Cyzici, 61 Eurip. Cyclops, Andromeda, Helena, 142 Soph. /chneutae, 141 ‘ datiuus sympatheticus,’ 62 defeated gladiators saved by Emperor’s interven- tion, 263 Democritus, a gnomic work by, 65 dictator, origin of the, 292 Dictys, notes on, 1g ff. digamma, did Plato write the ? 236 Dionysius Thrax on relation of ‘grammar’ to YPAMparcKy, 34 Dionysus, religion of, 134 Dobson (J. F.), Boethus of Sidon, 88 ff. Domitia Lucilla, 114, 117 Donatius’ letter to L. Munatius, 68 Doric a in Attic Tragedy, 141 E. Flegiae ad Maecenatem, date of, 63 ‘enantiosemia,’ 223 Ennius in the Middle Ages, 61 his treatment of the caesura, 141 Epterammi Cyzicz, their date and relation to Greek tragedy, 61 ‘epistolary tenses * in Latin, 135 Erzgone of Erastosthenes, the, 295 Etruscan house, the, 291 etymologies and derivations (E. W. Fay), 50 ff. Eumolpus, 264 Euripides’ Cyclops, Anrdromeda, Helena, date of, 142 Eurystheus, brother to Medusa, 257 Euthyphron a heretic, 233 F. Farnesina house-paintings, 293 Faustina, 113 f., 116 Fauorinus rep! raidwv rpopijs, 64 GENERAL INDEX Fay (Edwin W.), etymologies and derivations, 50 ff. Fortuna in Lucan, 107 ff. Fronto’s correspondence, 112 ff. date of his consulship, 112 G. Gallic chieftains and the /atus clauus, 287 Garrod (H. W.), ‘salapantium disertum,’ 48 f. notes on the Waturales Quaestiones of Seneca, 272 ff. Gellius (xxi. 1) derives from Chrysippus, 64 Gemini, the constellation, 257 genitive after AdBe, dpxw etc., 224 Gercke’s text of Seneca’s Naturales Quaestiones, 272 ff. Geryon, myth of, 226 ff. Goodell (Thomas D.), XPH and AEI, 91 ff. Gow (A. 8. F.), notes on the Agamemnon, 1 fi. ‘Grammar,’ ancient definitions and divisions of, 33 ff. grammarians of the Empire, their influence on the Roman codes and Catholic Creeds, 47 grammatical chapters in Quintilian, the (i. 4-8), 33 ff. history of the two-fold division of gvam- matica, 33 ff. Q.’s treatment of grammar proper, 36 ff. of literary branch of grammatica, 41 ff. work and influence of the ygrammarians of the Empire, 47 grammatical classifications of Quintilian, Cicero, Philo, Dionysius Thrax, Varro, 33 ff. Greek heroic sagas, 222 romances, 293 H. Haines (C. R.), on the chronology of the Fron‘o correspondence, 112 ff. Halieutica, authenticity of the, 267 ff. metrical peculiarities due to vulgar Greek pronunciation, 268 f. Hall (F. W.), a false quotation from Plautus, 205 Hardy (E. G.), Claudius and the primores Galliae ; a reply and a restatement, 282 ff. (see also 132 ff.) Harrison (E.), verse-weight, 206 ff. table of percentages of iambs and spondees in melic, tragic and comic trimeters, 207 four rules for proportion of iambs to spondees in the tragedians, 208 bearing of results on (a) authenticity of the Rhesus, (6) Porson’s Law, 209 ff. Headlam on the Agamemnon, criticized, 3, 5 Hebraisms in N. T. Greek, 203 f. Hellenistic art in Syria, 138 f. Greek, reconstructed by study of Modern, 187 f. Hellenotamiae, functions of, 14 Herodes Atticus, 113, 115 Hesiod’s 7heogony, the prooemium to, 138 Hipta, nurse of Bacchus, 293 historic infinitive, 291, 293 ee ‘ Homeri Horace, | Housma! 151 ff. iambs, pt trimet In Mem initiation inscripti interpolz Isidore, Josephu Julia, th Knight palim Kyd (A decad Laceda: land sys LANGU 144, Leopre| Leucon Liciniu LITERS PERI furth literary Livy o1 Livy’s palin AOTO: ions, 50 ff. 7 48 f. sttones of 64 uaestiones : fi. s% ! isions of, ice on the (i. 4-8), of 2ram- r, 36 ff. tca, 4t ff. nmarians » Cicero, e Fronto ar Greek $5 205 Galliae ; see also pondees fers, 207 imbs to y of the yr, 5 Aodern, eRe Holmes (T. Rice), the text of the Bel/lum Gallicum and the work of H. Meusel, 156 ff. Homer : liiad, composition of, 138, 221, 293 interpolations in, 289 meteorology of, 212 ff. Odyssey, indifference to natural phenomena in, 20. ‘ Homeric hymn,’ traces of, in Hesiod, 68 Horace, notes on, 145 ff. Odes, i. 3. I sqq., 145 f.; 1. 12. 45 5qq. 146 f.; i. 16. § sqq., 147 f. ; il. 18. 7, 148 ff. i. 14 a propempticum for Octavian, Housman (A. E), a transposition in Propertius, 151 ff. I. iambs, percentages of, in melic, tragic and comic trimeters, 207 ff. In Memoriam Alastair Hope Kyd, 253 initiation, the Spartan Cryptia the final stage of, 67 inscription, a new Argive, 293 at Lindus, 63, 295 interpolations in the //zad, 289 in Josephus, 295 Isidore, MSS. of, 137 J. Josephus, interpolations in, 295 Julia, the elder and younger, 25 K. Knight (Clara M.), the importance of the Veronese palimpsest in the first decade of Livy, 166 ff. Kyd (A. H.), the codex Bambergensis of the first decade of Livy, 248 ff. i memoriam notice, 253 L. Lacedaemonian Cryptia, the, 67 land system in Thessaly, 135 LANGUAGE: SUMMARIES OF PERIODICALS, 71 f., 144, 224, 296. See further under ‘ Summaries.’ Leoprepides, 257 Leucon, 255 Licinius, the Emperor, 67 LITERATURE AND GENERAL: SUMMARIES OF PERIODICALS, 61 ff., 134 ff., 216 ff., 289 ff. See further under ‘ Summaries.’ literary dialect in Modern Greek, 184 f. Livy on origins of tragedy and comedy, 290 f. epitome of, 65 Livy’s first decade, importance of the Veronese palimpsest in, 166 ff. codex Bambergensis of, 248 ff. AOTOS, on the Aristotelian use of, a reply, 9 ff. objections to Cook Wilson’s view (C.2. June, 1914) that Aéyos means ‘reason’ in certain passages of Nic. Zth., 9 f. another interpretation stated, ro ff. GENERAL INDEX 299 Lucan VIII., notes on, 105 ff. Lucan's fondness for rhetorical questions without introductory particles, 106 repetitions, intentional and unconscious, 110 f. reproaches of the gods, 109 f. use of a ‘ pregnant ’ construction, 107 Lucian’s rdotov 7 evxal, 295 speeches on Phalaris, 65 Lucretius, MSS. of, 147 /udi, corrupting influence of the, 25 Luxemburg MS. of the Moretum, 103 ff. M. Macedonian republics, 291 trophies, 221 Macurdy (G. H.), rainbow, sky and stars in the Lhiad and Odyssey, 212 ff. Maecenas’ relation to Caesar, 153 f. MSS. of Asconius, 68 Bellum Gallicum, 1§7 ff. Cicero, 68 Curculio, 216 Fronto, 112, 119 f. Isidore, 137 Livy, 166 ff., 248 ff. Lucretius, 141 Moretum, 103 f. Ovid's 7ristia, 21 Plato, 230 ff. Seneca, 68, 140, 142, 272, 292 Marcus Aurelius, 112 ff. date of his children’s births, 113 f. marriage, 113 Massiliots discover Britain, 64 Menaechmus, 68 Menander’s Andria and Ferinthia the sources of Terence’s Andria, 295 vocabulary, 219 mensae paniceae, 66 meteorological phenomena, their prominence in the lliad, 212 ff. Meusel’s work on the text of the Be//um Gallicum, 156 ff. miasma of homicide, 233 Milan MS. of Seneca’s ‘ Dialogues,’ 140 Modern Greek, its value for the study of Ancient Greek, 181 ff. erroneous theory of development, 185 key to written and spoken xow7, 187 ft. new literary idiom, 184 f. sound-changes in, 182 ff. Momos in the Ayfria, 236 moon and stars in //iad and Odyssey, 214 Moretum,‘a neglected MS. of the, 103 Munro (J. A. R.), "AOnvalwy moditela (xxx. 3. 4), 13 ff. music in the Roman Army, 221 mythical names, 71 N. names of barbarian tribes in Ovid, 24 Neaera as 2 commun name, 121 f. 300 GENERAL INDEX Nekuia, authenticity of the, 291 Nero’s literary circle, 216 Nettleship on Quintilian’s borrowings from Palae- mon and Pliny, 36 ff. new codex of Seneca’s Epistles, 68, 142 O. Odyssey, the, differs from the //éad by indifference to celestial phenomena, 212 ff. omission of ¢# and ego in poetry, 155 Orphic anthropogony, an, 66 orthography, 69 -ov scanned long before vowel, 226 Ovid, echoes Menander, 141 his fondness for three illustrations, 261 influenced by Stoic physics, 247 notes on the /éts, Ex Ponto Libri, and Halieutica, 254 ff. some passages of the 77éstia, 21 ff. Owen (S. G.), notes on Ovid’s /dis, Ex Ponto Libri, and Halieutica, 254 ff. on some passages of Ovid's 77réstza, 21 ff. P. Paean of Philodamos of Scarpheia, the, 288 palimpsest, the Fronto, 112, 119 f. the Veronese, of Livy, 166 ff. Palladius, word-formation in, 68 Peleus at Troy with Heracles, 225 f. Perilla, 22, 26 f. Phaidros, emendations in Plato’s, 7 f. Phalaris, Lucian’s speeches on, 65 Philetas, a poem of, 223 Phoenix of Colophon, 140 Pinarus, the river, 218 Pindar, emendations in, 225, 226 on certain fragments of, 225 ff. Platt (Arthur), Theocritea, 36 f. Plautus, a false quotation from, 205 Polybius, chronology of, 139 Porson’s Law, 208, 211, 219 Postgate (J. P.), Neaera as a common name, 121 f. on the text of the Stromateis of Clement of Alexandria, 237 ff. Powell (J. U.), the Paean of Philodamos of Scar pheia, 288 ‘pregnant’ construction, 107 primores Galliae, the, 132 f., 282 ff. Procopius, interpolations in, 222 Prometheus Vinctus, 217 Propertius, a transposition in, 151 ff. — Jewish and Essene sources of, Ptolemais tribe at Athens, the, 134 Q. quantities of Chione, Chionides, 264 of proper names, poetic license with regard to, 29 f. Quintilian, the grammatical chapters in (i. 4-8), 33 ff. R. rainbow, sky and stars in the Ziad and Odyssey, 212 ff. Rhegium, derivation of, 142 Rhesus, authenticity of the, 65, 211 peculiar rhythm in, 206 Rhine, Caesar’s description of the, 158 f. Rhinthon’s iNaporpaywola, 26 Richards (H.), Thucydidea (continued), 73 ff. river-gods represented in triumphs, 264 Roberts (W. Rhys), note on Thucydides ii. 48. 3, 16 Rufinus, 221 S. St. Paul’s speech on the Areopagus, 222 ‘ salapantium disertum,’ 48 f. Samnites, 142 Sappho, the new ode, 221 scholia to Plato, 230 f. Scopas the Aleuad, 257 Seneca, MSS. of, 68, 140, 142, 272, 292 notes on the Naturales Quaestiones of, 272 ff. Seneca’s de Jra, 292 influence on Neronian poets, 216 shortening of diphthongs in Greek and Latin poetry, 27 f. of long vowels in vulgar atin, 268 Sicyonian chronology, 68 Simonides and the Dioscuri, 257 f. snow in the //zad, 214 Sophocles’ /chneutae, date of, 141 notes on, OI sound-changes in Modern Greek, 182 f. Souter (A.), a neglected manuscript of the More- tum, 103 spondees, percentages of, in melic, tragic and comic trimeters, 206 ff. Stocks (J. L.), on the Aristotelian use of AOTO2: a reply, 9 ff. Stoic comparison of Wise Man to sphere, 246 doctrine of the central sphere, 246 f. SUMMARIES OF PERIODICALS, 61 ff, 134 ff., 216 ff., 289 ff. American Journal of Philology, 61, 134, 216, 289 Athenaeum (Pavia), 61, 134, 216 Berliner philologische Wochenschrift, 62 f., 134 ff., 217 f., 289 f. Classical Philology, 290 ff. Classical Weekly, 64, 136 f., 218 f. Deutsche Literaturzeitung, 64 f., 137 [., 219, 292 Glotta, 71 f., 144, 224, 206 Hermathena, 65 Hermes, 65, 138, 219 f., 292 f. Indogermanische Forschungen, 72, 144, 224 Journal of Philology, 66 Mnemosyne, 66, 138, 220, 293 Musée Belge, 220 SUMMAI I ] I ] ] ] 4 cipryg Te Tasgetius tautology teaching Theocrite Terence’: Theogon Thesaurs three ill 261 Thucydie (contin Thucydi 104 Thumb | the sti thundert trahere, transpos trimeter in, 20 Tsaconi Odyssey, ll. 48. 3, tones of, 1 Latin 8 e More- ric and LOTOS : 16 f. 61 ft, I, 134, ft, 62f., SUMMARIES OF PERIODICALS—continued. Neue Jahrbiicher fiir das klassische Alter- tum, etc., 66 f., 138 f., 220 f., 293 f. Philologus, 67, 139, 221 f. Revue de Philologie, 140 f., 294 Revue des Etudes grecques, 67, 221 Rheinisches Museum, 68, 141 f., 222, 294 Rivista di Filologia e d’ Istruzione Classica, 68 f., 142, 222 f., 295 Wiener Studien, 69 Wochenschrift fiir klassische Philologie, 70 f., 142 f., 223 f.. 295 f. Zeitschrift fiir vergleichende Sprachfor- schung, 144, 296 cipryé Technopaegnium, the, 66 T. Tasgetius, assassination of, 161 tautology in Latin poetry, I10 f. teaching of literature in Roman schools, 46 f. Theocritea, 86 f. Terence’s Andria, sources of, 295 Theogony, the Cyclic, 294 prooemium to Hesiod’s, 138 Thesaurus Dialecti Lonicae, a proposed, 62 three illustrations of a maxim, an Ovidian trait, 261 Thucydidea, Part II. Miscellaneous emendations (continued), 73 ff. Thucydides, note on ii. 48. 3, 16; on vii. 28. 3, 104 Thumb (Albert), on the value of Modern Greek for the study of Ancient Greek, 181 ff. thunderbolts wielded by nine gods, 281 trahere, different meanings of, 25, 148 ff. transposition, examples of, 155 in "AQ, IIo. xxx. 3-4, 13 ff. in Propertius iii. 9, 151 ff. trimeter verse, percentages of iambs and spondees in, 207 ff. Tsaconian dialect, 186 GENERAL INDEX V. Varro and Menippus, 293 Varro’s divisions of exarratio poetarum, 41 of grammatica, 34 Velleius, 140 verbals in -réos, 93 Veronese palimpsest of first decade of Livy, charac- teristic errors of, 167-172 importance of, 166 ff. true readings preserved by, 177 ff. variant readings and glosses in, 172-177 ‘ verse-weight,’ 206 ff. Vicus Sandaliarius, 291 Viljoen (H. G.), emendations in Plato’s Phaidros, 7 f. Vindiciae Platonicae I., 230 ff. Virgil, mosaic picture of, 134 Virgil’s treatment of the caesura, 123 ff. vulgar Greek pronunciation reflected in Latin literature, 269 X. XPH and AEI, 91 ff. historical development of two synonyms, 9I-3 their use’ in Aeschylus, Euripides, Sopho- cles, 95 f. Aristotle, 102 Herodotus, 97 f. Homer, 93 f. orators, Xenophon, 100 Pindar and Bacchylides, 94 Plato, ror f, Pre-Socratics, 96 f. Thucydides, 98 ff. Z. Zamolxis an epithet, 67 Zeus, physical aspect of, preponderates in the //ad, 213 Thaulios, 61, 134 A. Aeschylus :— Ag. (259 syg. coll. 187, 228 sgg., 435 sgg-) 1 f.; (427 sg. coll. Ach. Tat. vi. 6, Eumath. iii. 7) 23 (536 sgg. coll. Ch. 301) 2 f. ; (831) 3 f.; (1028 sgg. coll. Theogn. 543, 945; Sopb. /v. 421; Eur. Jom 1514; Plaut. 27. 348) 4 f.; (1321 57.) 53 (1536 sg. coll. Ch. 643, 909; Zum. 169=174) 5; (1651 Sg. coll. Suidas s. v. mpéxwaos; Eur. Or. 14773 Eum. 595: Eur. Phoen. 363) 6; Zum. (506 594.) 294 Aétius :— Plac. (I. 6 coll. Diog. Laert. vii. 148), 246; (ii. 2. coll. Herc. Frag. Arnim II. no. 1059 ; schol. Hom. //. ¥Y. 65; Lucan 9. 9. Hiero- nym. Zp. 108. 23; Marc. Aurel. 12. 3; Id. 8-41, 11. 12; Hor. S. ii. 7. 56) 2d. 5 (ili. 9. 3) 26. Apollonius Rhodius :— Argonaut. (i. 936; ii. 368; iii. 1585; iv. 786) 66 Aristotle :-— "AO. Ilon. (xxx. 3-4. coll. xxix. 3, xxxi. 2, lvi.1) 13 ff. Nic. Eth. (1106 6 36; coll. 1139 a 18; 1139 6 16; de An. 435 616; Fol. 1235 a 23; N.E. 1115 6 22; 26. 1166 a 10; tb. 1170 @ 16; de Caclo 293 6 13; #6. 297 6 34; Phys. 219 a 22) 9f. de Part. An. (639 5 11-19; coll. 1140 6 5; 640 a 31; 1144 6 27; 1107 a 1; N.EZ. Il. iv.; 20. VI. iv.: W.Z. vi. 1139 a2 31; MZ. ili. 1103 4 31; Aspasius [ed. Heylbut, 48, 20]) 10 f. C. C JG. (2815, 3642) 294 Caesar : — Bell. Gail. iv. 10 (coll. iii. 28, 15 vi. 243 vii. 77, 143 il. 20, 4; 7b. 30. 4) 158 ff. Callimachus :— Hymn. ad Apoll. (105-113 coll. Anth, Pal. x1. 275) 65 Catullus :— (liii. 5 coll. Plaut. Curc. 463) 48 f. Cicero :— ad Att. xiv. (18. 1) 293 ad Fam, vii. (12. 1) 140; xv. (20) 290 I]l.—INDEX LOCORVM. Clement of Alexandria :— Strom. vii. 45 sq. (coll. Aét. Plac. i. 6; ii. 2; ili. 9. 3; Herc. Frag. Arnim ii. no. 1059; schol. Hom. //. ¥. 65; Lucan 9. 9; Hiero- nym. £/. 108. 23; Marc. Aurel. 12. 3; id. 8 41; 11, 12; Hor. S. II. 7. 56; Achilles /sagoge 4; Plut. de fac. in orbe lunae c. 6. p. 923 E; Ov. Fast. 6. 169 sqq.) 245 ff. D. Dares :— Kips, 2;Cana 2 a; C.ao2¢ 12; 13; C. iv. p. 5, 24; C. viii. p. Ic, 7, 18; C. xi. p. 14, 5; C. xviii. p. 23, 3) 17; (C. xx. p. 25, 26; C. xxi. p. 26, 19, 23; C. xxvi, p. 22, 1, 18; C. xxvii. p. 33, 15; C. xxviii. p. 34, 19; C. xxxvi. p. 44, 3; C. XXXVIl. p. 44, 21; C. xli. p. 49, 9, 20) 18; (C. xlii, p. 51, 8) 19 Dictys :— I. (C. ii. p. 4, 9) 19: (C. viii. p. 7, 19) 19: (C. xxiii. p.17, 31) 19; II. (C. xxviii. p. 36, 15) 19: (C. xlii. p.45, 17) 19; III. (C. iii. p. 53, 33) 19: (C. xxiii. p. 59, 26, p. 66, 13) 19 f.; V.(C. vii. p. 92, 18) 20: (C. xvi. p. 100, 10) 20; VI. (C. ii. p. 103, 15) 20 Diogenes Laertius :— vil. (54) 90 Euripides :— Cyclops (694) 191 Helena (390 sg., 619) 294 Jon (1424) 294 H. Herodorus : — F. H. G. (ii. 34) 138 Herodotus :— i. (86) 290 Hesiod :— Theogony (1-115) 138 Horace :— Epod. iv. (16) 292 Odes i. (3, 1 sgg. coll. Cato de Agr. 70,2; Mart. ix. 99, I sgg.) 145 f. (12, 45 sy. coll. Lucret. i. 314, Hor. Eff. i. 1, 80) 146f, ; (16, 5 sgqg. coll. Varro Sent. 26; Cic. Phil. Hyginus :. Fab. Index Hei Manilius :- i, (28: Marius Vi K. vo Xv. a 2 Halie y Con His 271 Lois ( 140 70, 2; sy. coll. 146 f, ; c. Phil. ii. 94: Sen. Rhetor, Cont. Exc. 4,8; Apul. Apol. 14) 147 f.; ii. (18, 7 coll. Varro, /r. 190, B: Juv. ii. 54 sgg.: Ov. Met. xiv. 264; ib. xiii. 511; 20. ii. 4115 Id. Her. ii, 75 | Hor. A.P. 215) 148 ff. Sat. i. (10. 5#) 293 Hyginus :— Fab. (274. 1) 294 I. Index Herculanensis (Col. LI. ll. 7-9) 89 Livy :— vii. (2. I-12) 290 Lucan :— vill. (5 599.5 102 syg.) 105 f. ; (306-10 coll. i. 299 Sgq., ili. 137 SG., iv. 215-219, Vi. 153- 156, 326 sg., ix. 1073-1076, vill. 793-796) 106 ; (701-708 coll. Liv. ix. 20. 5; Verg. Georg. ii. 505) 107 f. ; (663-667 coll. 622- 635, 642, 861) 108 f. ; (858-861 coll. i. 121- 124; iv. 799 sgg.; Vili. 548 sg., 713 5q., 793 sgg.) 109 f. ; (717-720 coll. Verg. Aex. vi. 152, 460; 2b. ix. 337 sg.; Catull. Ixvi. 39; Liv. ix. 2. 8 sg.; Luc. i. 82, 93; il. fin. ; iv. 264, 292, 426, 448 sgq., vill. 570, 625, 631, 793, 796) 110 f.; ix. (980-986) 216 M. Manilius :— i, (285) 222 Marius Victorinus :— K. vol. vi. (p. 4) 343 (#.) 35 N. Nonius :— (p. 172, Lindsay), 48 O. Ovid :— ex Ponto I. (i. 65-66 coll. ii. 2. 24, A. A. ii. 101) 258 f. : (7. 21-22 coll. i. 9. 49) 259 f.: (7. 65-66 coll. ii. 1. 4, Cic. OF i. § 48) 260 ; II. (2. 31-38 coll. ii. 5. 61; Am. i. 2. 9; A. A. ii. 177; Med. faciet 37) 261 ff. ; (8. 53 coll. Suet. Ju/. 26, Seneca ap. Antzh. Lat. P. L. M. iv. 25, 27) 2633 III. (3. 41 coll, Met. xi. 301) 264: (4. 89, coll. 94, 114) 264: (4. 107 sq. coll. Plin. MV. Z. v. 37; Tac. Amn. il. 41) 264: (5. 49 coll. eZ. viii. 172) 264; IV. (4. 27-34 coll. lili. 5. 47; Fast. iv. $4, 415; Met. iv. 734, xv. 126; Liv. xxi. 60) 265: (8. 85 sg. coll. i. 7. 133 iv. 10. 71) 265 f. : (16) 266 f. Halieut, (2 coll. Am. iii. 13. 15; Juv. xii. 7: Mart. iii. 58. 11, vi. 38. 8; Claud. iv. Cons. Honor. 383) 269 f. ; (53-57 coll. Tac. Flist. i. 14) 271 ; (65, 67 coll. Liv. ix. 9. 14) 271 /bis (131-132 coll. A. A. ii. 113) 254; (137- 140 coll. ex P. i. 2. QO) 254; (289 5g. coll. INDEX LOCORVM 303 Aesch. P. V. 10) 255 ; (307 5g.) 255; (413- 416 coll. 203; Am. iii. 6. 101) 256; (445 sg. coll. Herotd. ix. 7; ex P. i. 6. 38, il. 5. 6) 25Af. ; (509 sg. coll. Juv. vii. 194 ; Manil. ili. 58 sgq.) 257 f. Tristia 1. (i. 47, 48: coll. Verg. Georg. iii. 390; Livy i. 30; Juv. viii. 95; Seneca Dial. i. 3. 3) 21 £. II. (77-80: coll. ii. 379; iii. 4. 71; Pont. i. 76; bis 639) 22; (83-86) 22 f.; (ii. 89-92: coll. Tac. Ann. i. 74; Cic. pro Mur. § 87) 23; (il. 109-114) 23 f. ; (191-192 coll. Plin. N. H. vi. 19) 24; (253-254 coll. Font. ini. 2.143; Ars Am. ili. 421) 24 f.; (ii. 277-278, 279-282) 25; (il. 409-410) 25 f. III. (i. 47) 26 ; (3. 21) 26; (7. 27-28 coll. Cic. c. Rull, i. 91; Verg. Aen. i. 382) 26 f. ; (10. 11-12 coll. Pont. i. §. 513; i. 3. 35; Her. 16, 32) 27; (12. 1-2. coll. Plaut. Asin. 333 ; Eur. Herc. 348-450; Soph. Antig. 1131 27 f. IV. (10. 79-80 coll. Stat. Si/u. v. 1. 221, Thebd, xii. 246; Lucan viii. 849) 30 V. (1. 69-72 coll. Fomt.i. 1.19; Ars Am. il. 108 ; Her. 9. 129; Metam. i. 585; Hor. Carm. ill. 5. 42) 30 f.; (10. 41-42 coll. Metam. ix. 768, xii. 57) 31; (12. 11 coll. v. 13.8; Lucret. v. 909) 31 f. ; (12. 49-50) 32 Persius :— 1. (99-102) 216 Philo :— de Conegressu (4) 343 (26) 2/. de Ab. et Caint (22) 34 Philodainos of Scarpheia :— Paean (ll. 53-56) 288 Philumenos :— 4. (14) 293 Pindar :— Fr. 172 (coll. Fr. 227; O. i. 36 sgg.3 O. 2, 23 sqq. ; O. 7, 26 sgg. 3 O. 9, 28 sqq. 3 P. 3,54 ogg.3 F. ap 1333 VW. & 283; G29; &. 50, 51; P. 3, 93 $qq.) 225 f.; fr. 177, 4 (coll. O. 13, 35; P. 1, 39) 226; Fr. 169 (coll. Plat. Zaws, vii. 795 C; Aesch. Ag. 857-862; /. i. 13; Fr. 813; Strabo iii. 148, 11; Apollod. 2, 5, 10; Hes. Zheog. 287 sqq.; Athen. 8, 346; 2 Smyrn, 6, 1096 2; Pind. WV. 1, 62 sqq.) 226 ff. ; Fr. 235 (coll. 1.%7 3 F 08, 83; &H. 1,4 OH; F. 3 N.11, 16; P. 2,84; Ar. Wasps 338; tlom. Od. 21, 71) 229 Plato :— Euthyphro (2a,5: 3a, 1: 3¢, 2: 4a, 12) 232f.; (44, 10: 4¢, 2: 56, 4: 5c, 6) 233 f. (54, 4: 5¢, 2: Je, 2) 234 f.; (9d, 2: 118, 7: 1l¢, 3: 12a, 9) 235 f.; (14¢, 3: 16a, 3) 236 Phaedr.. (230d, coll. Gorg. 511¢) 7 ; (236¢, coll. Soph. 0. 2. 47) 7 f. 304 INDEX LOCORVM Plautus :-— Crist. (iv. 12. 2) 205 (false) Curc. (463) 48 Polybius :— vi. (Ig. 2) 294 Propertius :— ! II. i. 17-42, III. ix. 33 sg. (coll. Luc. iii. 122 ; Prop. III. vii. 21: 26. xiii. 51; Pind. Mem. ix. 39 sqg.; Hor. Serm. II. ili. 211-3; Id. Epp. I. ii. 91; Ov. Her. vi. 135; 2d. xii. 187 ; 2b. xvii. 99; 26. xix. 185; Luc. ii. 292-5; Mart. V. xiii.9; Ov. 7rzs¢. IIT. xi. 21 sg.; German. Phaen. 387-393) 151 ff. ; ii. (24, 1-16 Phillimore) 294 ; iii. (8) 7d. Prudentius :— Pevist. x. (239 sq.) 121 f. contra Symmachum I (135 sqq.) #6. Q. Quintilian :-— (i. 4-8) 33 ff. Sextus Empiricus :— Adu. Gramm. (92 sq.) 46 ; (249) 35,42 n. ; (250 $99.) 42, 45 Silius Italicus :— vii. (162-211 coll. Nonn. Diuonys. xlvii. 34 59g.) 294 Sophocles :— Az. (961-70: 1081: 1105-1117: 1312) 294 Ant. (134) 290 Strabo :— xvi. (p. 757) 90 T. Tacitus :— Hist. v. (13) 67; xi. (25) 293 Terence :— Andria (Act. I. Sc. i.) 294; (959 5g). #6 Theocritus :— (i. 109, 139: ii. 140 coll. xiv. 23: xii, 24) 86 - (xv. 7) 294; (xxil. 66) 86; (xxiii. 41 coll. xll. 25: xxvii. 60) 87 Thucydides :— li. (48. 3 coll. i. 22. 1-3, Hdt. i. 197) 16 v. (I: 2.2: 4.6: 8.9: 16.2: 26. 3: 17.2: 18. 4) 73; (#. 5: 20.1: 26. §: 27.2: 40, 3: 6o. §: 63. 1: 76. 3: 66. 1: 2b. 2: 71, 3: 72.2: 76. 3: 84. 4) 74; (88: 89: 95: 105. I: 105. 3: 106: Il. 2: 26. 3) 75 vi. (1. 1: 2. 2: 3. 1: 6 2: 97. 1: 8. 3: 10. 4: 11. 2) 76; (11. 4: 12. 1: #.2: 13.1: 16. 2: 26.5: 18. 2: 26. 3: 2. 6: 21. 2 22: 31. 4: 33-6: 34-5) 77; (34.9: 35.13 36. 1: 2b. 2: 46. 3: 49.4: 53. 3: 54: 3: 55. 2: 64. 1: 72. 4: 76. 4: 78.2: 2.3; 79. 2:) 78; (80. 5: 82. 4: 83. 1: 85. 3: 87. 3: 89. 2: 91. 4: 2. 7) 79 vii. (1. 3: 3 3: 13. 2:) 79; (25. 2B: 8.9: 27. 4: 28. 2:) 80; (28. 3) 80. 104; (29. 5: 30. 2: 31.1: #6. 4) 803 (48. 3: 49. 3: 51. I: 56.4: 61. 1: 63. 4: 66. 3: 75. 4: wb. 6) 81; (77.3: 78. 1: 80. 4: 83.4: 86. 3: 1b. 5: 87. 4) 82 viii. (1. 2) 82; (7. 1: 8. 1: 26. 4: 27. 2: 46. 2: 16.5: 48, 4: 16.5: 16.7: 54.2: §6. 3) 83; (57-2: 58. 6; 60. 3: 62. 2: 63.4: 64. t: 66.2: 69.1: 71. 1: 73. 2: 7§. 3: 86. 3: tb. 4: 89. 3) 84; (90.4: 92.9: 10,11: 94. 3: 96.4: IOI. 2: 104. 5) 85 Tibullus :— li. (1-58) 220 V~. Virgil :— Georg. iii. (100) 294 X. Xenophon :— Hellen. ii. (3. 10) 293 ATBA dvyaor dyvwo dypou alyAn alfip, Axoal des, | augey hofos, &paxos domde aruny, avrouc "Axire Bpvwy1 Busiv, YPGMe ypttros det, QI 51d pw dbyua doa, dodXos, Awptijs ‘EXAnv €urrecpl éfemNd, émrel, ‘ émelok) ernrhs Epvos (¢ eva 87 etvh, 5 éxwomr *>\cér Arop, § li, 24) 86 ; i. 41 coll. ATBATANA, 292 dyhaoparis, 192 &yvworto Geol, 222 dypoawrns (Hom.), 52 aly\n (Hom.), 213 aidhp, odpavds, dnp (Hom.), 213 ’"Axoal (I. G. iv. 955, 1. 10), 138 ddes, ‘wit,’ 139 duduyujers, 296 hofos, 50 &paxos, 192 domdusrns, ‘in scuto iens,’ 51 arunv, 60 avrouoNla, 79 "Axireds, 71 Bpywyn—Bpvwvia, 192 Buglvy, 190 YPAMMaATLKH, 33 yptros, 194 det, gt ff. didp0wors, ‘textual criticism,’ 41, 44 déyua ’Emixovperov, 295 dox®, ‘think,’ 75 dodXos, 224 Awptijs, Awpls, 72 ‘EdAnvcuss, 35 éurecpla (Thuc.), 74 A. éterdaynoav (Thuc.), 74 érei, ‘though,’ 86 émeloxAnro, 14 érnrhs (Hom.), 60 Eovos (cf. Spvusu), 144 ev\aBHOnr., absolute use of, 7 etvt, 57 éxwdtrous, 192 *n\ubmupos, 194 Rrop, 59 Ii1.—INDEX VERBORVM. A.— GREEK. 8. *Papperés, 194 *favXdos, 61 Ovperpa (Hom.), 54 I. immérns, 52 ton ppovpd, 80 K. kaxws, ‘in vain,’ 191 Képxoy, cf. Ilé\oy, 71 keprouetv, *‘ to mystify,’ 294 xolpavos (Hom.), 57 Kévuga, 192 Kkdpragos, xdprados, 296 xrelvw (Plato), 233 KWAWTNS, 53 *\aBbw, 194 Aawdvn, 192 Aéyos=‘ plan’ (Arist.), 10 ff. M. *uéra=comes, 50 “wou Bos, 189 N. velatpa, véaipa, 122 vijcos, ‘ peninsula,’ 66 voulfw, ‘ practise,’ 82 dapiorhs, dapio vs, 60 duparnrépos, 139 dEvpuyxos Xapaxrhp, 68 oplferOal re Tivl, 9 dptcceayv, 86 Il, *rapadadapuldiov (?), 194 mepolkiov, 192 wrdouara, ‘ comedies,’ 45 whevpd =costa, ‘ wife,’ 67 modayés (Soph.), 52 mwédemos, 85 mpoaipeors (Arist.), II mporidecGar (Attic) = bworl@ec Oa (Ionic), 235 mpdoxerpos, 42 N. IIpwreal-daos, 59 306 INDEX VERBORVM pbov = xara pbov, 86 céXas, 213 oTdOun, 4 oTpaTrwrns, orparaydés, 52 *ciuTdnyos, 194 suvaypls, 193 TravTahw@eis, 290 réyva, ‘ trades,’ 291 Tpaywodés, ‘singer,’ 192 A, anthias, 268 B. barbarismus) (soloecismus, 38 busta, ‘ burnt remains,’ 30 C. canés, 296 capere timovem, 271 capiunt animis, 271 civcumspicere, ‘ provide for,’ 21 f. Cocles, 52 f. coloniae et municipia= ‘Italy,’ 132 comatorius =cinerarius (?), 278 comes, 50 comites, 105 crimen adipisci, 23 densa carina, 255 Domina, 114 familia (Livy), 291 fenestra, 142 fictum, 31 fricticulae, ‘ baked ’ bread or fish, 222 fultus, with in instead of simple abl., 31 G. gens (Livy), 291 gloria, 142 grammatica, 33 ff. H, halophanta, halapanta, salap(h)anta, 48 f. historia, 45 honestus, ‘ having held honores,’ 23 I. lazyges, read Ciziges, 24 L. lacerta, 53 lappa, ‘ burdock,’ 140 ludt, 25 luperct, 67 mancus, 270 mensae paniceae, 66 methodice)(historice, 33 mitis, ‘gentle,’ 255 T. vyijs, ‘valuable,’ 46 n, brbxpiow, 229 pépeTpov, 54 gurixyn, 192 dvoxia (?), 86 X. xem, 9o ff. xphua—xphuara (Hdt.), 138 B.—LATIN. miluus, 268 mormyres, 269 mulo inscitior, 205 ne enclitic, 291 neaeva=meretrix, 121 f, nitor, with in instead of simple abl., 32 nuntiave, used absolutely, 169 O. occulte, 147 oculis inunctis, ‘unsalved eyes,’ 142 e. paeonius, 134 patet, impersonal, 27 pellex =mddd\ak = Tladdds ? 67 penes=‘ nach jemandes Ansicht,’ 144 Perilla, 22, 26 f. petit, in pregnant constr., 107 plura uerba =‘ formations,’ 38 pompilus, 269 pontes, ‘ paths,’ 290 poples, 53 prohibere, followed by infin., 27 purpurae, 150 QO. queruntuy, ‘ sing,’ 61 R, vecognitio equitum, 23 vuvicola (Ovid), 54 S. salapantium, 48 f. sermo, ‘ usage,’ 36=pluva uerba, 38 solutus, 26 spatium, ‘ lapse of time,’ 254 stirps, 291 Suauulua, 122 _Z Tecusa =‘ mother,’ 67 Teretae (Toretae), 24 testari aliquem, 152 f. tityrus =satyrus, * lecher,’ 216 trahere, ‘card,’ 149; (medical), 25; ‘trail,’ 150 V. uiy, emphatic, 258 uota facere, 257 trail,’ 150