APR 14 1QU. THE FUSION OF STYLISTIC ELEMENTS IN VERGIL'S GEORGICS BY META GLASS Submitted in Partial Fulfilment of the Requirements FOR the Degree of Doctor of Philosophy, in the Faculty of Philosophy, Columbia University NEW YORK 1913 THE FUSION OF STYLISTIC ELEMENTS IN VERGIL'S GEORGICS BY META GLASS Submitted in Partial Fulfilment of the Requirements FOR the Degree of Doctor of Philosophy, in the Faculty of Philosophy, Columbia University NEW YORK 1913 Press of The New Era Printing compant Lancaster. Pa 1 '' CONTENTS Introduction v I. Astronomical, Geographical and Literary Ref- erences 1 II. Word Order 26 III. Euphonic Devices 48 IV. Analysis of Special Passages 69 V. Mental Processes 78 Bibliography 92 282306 INTRODUCTION Inasmuch as the artistic beauty of Vergil's poetry has been acknowledged and enlarged upon throughout the ages, there have naturally been many attempts to reach the source of this perfection, or at least to travel along the well-marked paths in the poems themselves that give the seeker some glimpses of the phases of Vergil's fancy, the sensitiveness of his perceptions of beauty in form and color and sound, and his unique ability to pass on these perceptions to others by use of the various means of poetic expression. In the study of any poet who so avowedly followed a model as did Vergil, it is in greatest part upon these manifestations of self, the workings of mind and of imagination, that a knowledge of that poet depends. Under these conditions Vergil's work- manship and what it reveals become all the more important.^ No less in the other two works than in the Georgics did Vergil have a pattern, but for his own individuality in develop- ing that pattern the Georgics are most fruitful. The Eclogues were a more youthful work, in a style to a great extent aban- doned later, while the Georgics have sufficient of the earlier and likewise of the later development to represent his maturer powers before they had shed any of their possibilities. The Aeneid was never jfinished, but after seven years of writing and rewriting the poet was content to let the Georgics go from his hand.^ What is to be found therein must, therefore, have had his full sanction, must have approached his ideal as nearly as an author's work ever does approach his ideal, and the features of workmanship traced here we have a right to deem those that he felt most fully expressed him. 1 T. H. Wright, Style, speaks of style as an involuntary revelation of self. 2 Donatus, Vita Vergili, Chapters 22, 25. We may believe, too, that the poet was more at home in his material in this poem than in either of the others. The tone of seriousness that the Georgics have as compared with the Eclogues was surely more in accordance with the philosophical temperament of the poet who was fully conscious of the lacrimae reruvi, and in portraying Italy as a veritable biferi rosaria Paesti he found his material more congenial than he later found the wars and bloodshed that moulded the character of Aeneas, and formed the prelude to the glories of Rome. In the Georgics he spoke whereof he knew from daily sight and hearing, or from the volumes of his predecessors which he had made his own. Because the poem is classed as didactic literature, which gained an unwonted prominence in Alexandrian times, and because of Vergil's use of Hesiod,^ of Aratus, of Eratosthenes, and of Nicander,'* because of his former allegiance to Theo- critus and the general tendency of much of Roman literature to submit to Alexandrian influence, the extent to which Vergil came under this influence, and the use, in so far as his artistic expression is concerned, which he made of the knowledge gained from this source are important points in any attempt to estimate the poet in his art. It is the purpose of this study to examine the Georgics from this point of view, to investigate Vergil's use of astro- nomical, geographical and literary allusions, his elaborations of word-order, and his use of euphonic embellishments, and to discover what glimpses may be gained of the personal equation that colored his perception and imagination and, consequently, his artistic expression. » Servius on the Georgics, vol. 3, p. 128. * For the sources in general, of. Karl Brandt, De auctoribus quos in com- ponendis Georgicon libris adumbravit Vergilius. THE FUSION OF STYLISTIC ELEMENTS IN VERGIL'S GEORGICS ASTKONOMICAL, GEOGRAPHICAL AND LiTERARY REFERENCES In any study of elaboration of style among the Romans, the question of Greek influence, whether coming direct from the classical period, or exerted through Alexandrian scholars and writers, naturally thrusts itself to the fore. Certain traits do appear in Latin, in poetry rather than in prose, the counter- parts of which are not to be found to any extent in classical Greek, and the first inference is that the Alexandrians are responsible for them. Aside from the genius of the two languages, the rhetorical embellishments found in Cicero, in Livy, in Seneca do not differ in aim, nor so very much in means, from those used by Thucydides and the Greek orators. But the poetry of Catullus, of Vergil, of Horace, and even more that of the writers of elegy shows devices not found in the lyric poets of Greece, nor to the same extent in the drama- tists, though Euripides shows the beginning of the use of many of the later favorite devices for securing stylistic effect. When one speaks of Alexandrianism, probably the first thought which rises in the mind is that of erudition, and in looking for traits of Alexandrian influence in Latin Hterature one most readily notes references to geography, astronomy and literature in general, more or less familiar supposedly to the author's public, as examples of this influence. When Catullus introduces into so personal a song as his seventh poem, where he pleads that Lesbia's kisses shall be as many as the sands of the desert or the stars of the firmament, the restriction that the sands to be counted are those of the Libyan desert around Cyrene where silphium grows, between the oracle of sun-baked Jupiter Ammon and the sacred tomb of Battus of old, he is displaying learning pure and simple, and is in so far a follower of the Alexandrian school. So large a word as laserpiciferis may make the sands seem more than those of an ordinary desert, and they are indeed sands of distinguished company, but to the sincerity of feeling that the second simile gives such elaboration is undoubtedly hostile. If Catullus desired the almost comic touch which the verses have for a modern reader, he has succeeded most skillfully in attaining it, but the weight of learning and sound in them is all the slight poem could sustain. Such a use as this of geographical knowledge is far to seek in classical Greek poetry. The local setting of a myth or of an actual occurrence is common enough, though not often in a form at all elaborate. Even in the passages grouping rather imposing geographical names that may be quoted from the Agamemnon (281 ff.), where the course of the fire signals that bring the news of Agamemnon's approach is traced, and from the Prometheus, where the hero foretells her wanderings to lo, a fellow sufferer from the injustice of Zeus (705 ff.), the references are of the sort that would be classed as occasioned by the narrative. The mountain peaks named in the Agamemnon passage were sufficiently familiar to Aeschylus' public (all were in countries with which Athens had had intercourse for years) to give vividness rather than vagueness to the picture, and so they show no striving for the impressiveness which the mention of more distant and less known regions would effect. In the Prometheus lo's wanderings are to be largely around the region in which Prometheus is bound, involving places quite closely akin to the setting of the drama, but sufficiently vague in Aeschylus' own mind to make the course almost impossible for one to follow. Some stylistic effect is gained by each of these passages, but there is lacking in them the sense of effort and the flavor of learning that mark many such references in Latin poetry.^ In astronomical references the Alexandrian influence is even more pronounced, for the greater study devoted to the subject in these times yielded the works of Eratosthenes and Aratus, and thus, by uniting the science quite definitely with poetry, made it available for the less technical poets.^ The popularity of Aratus among the Romans before Vergil wrote is attested by Cicero's translation of his ^atvofieva, and his continuance in favor is proved by the paraphrase of Germani- cus later. Vergil's own use of him'' is evidenced in the first verse of the Georgics, where he purposes to sing quo sidere terram vertere,^ and only a few verses farther on astronomical science is drawn upon in the more elaborate description of the constellations (I. 32-4) among which the deified Caesar is to have a place. So, again, after the golden age, when man 6 In the Medea of Euripides, where the course of the Argo might entail such references, they are quite closely connected with the story and are not elaborate. There is mention of Chalcis, the Symplegades, Corinth, Athens and the Cephis- sua. In the Iphigenia in Tauris the Symplegades, Taurica, Argos, Greece, Troy and Aulis are the places named. Even in Pindar the place designated is the home of the hero, or the scene of some definite exploit described in the myth. The fourth Pjd;hian, which deals with the story of the Argo, does not mention the course in detail, but says "and they came to Phasis" (375), and on the return voyage (446-7) "they met the streams of Ocean, the Red Sea and the race of Lemnian women." « Mahafify, Greek Life and Thought, pp. 242-51. ' Both the ^at.v6/j.eva and the Aioffrifieia give matter used by Vergil in Book I of the Georgics. The general knowledge of the constellations and their order must have come from the ^aivSfieva, while the weather signs of I. 351-463 follow much of the Aioa-ij/j-eia. 8 ^aivdneva 10 ff. Ai5t6s yap rdSe c^/tiar' ip ovpavifi iffr-^pi^ev 'AffTpa SiaKpivas iffKhj/aTo S' eh iviavrbv ''karipas, o'l kc fidXiffra T€Tvyp.4va p' f/xireda irdvTa (jtiuvrai. was with difficulty forging his way in the world, the sailor had to give name and number to the stars (I. 138), Pleiadaa, Hyadas, claramque Lycaonis Arcton. With verse 204 comes the more definite division of the seasons by the position of constellations, where Arcturus and the Haedi and Anguis, Libra, the Pleiades, and the Crown and Bootes all serve as signs for different phases of the farmer's activity. Thus the zones are marked out,^ and the sun's path across the zodiac and the moon, too, indicate days to be chosen or shunned for varying duties. Again, after the description of the spring storm, Vergil returns to the planets (I. 336-7), pauses to prescribe the worship of Ceres in the fields, and then pursues further the signs the Father has willed the moon to give of wind and rain and fair weather. With this description are mingled the prophecies of birds and beasts, and the climax is reached with the sun (solem certissima signa sequuntur, I. 439) whose pity for Rome at Caesar's death made him hide his shining face in murky darkness. A comparison of the truly poetic and beautiful way in which Vergil handles the science that he borrows with the technical and unimaginative work of Aratus^° gives but another instance of his fine perception of the poetic possibilities of his material, his readiness to use his learning always, but never to display it to the detriment of artistic harmonj\ Let us note now to what extent the Latin writers before Vergil advised the farmer to depend on knowledge of the heavens. As one would expect, Cato gives no such guide. He designates his seasons by primo vere,^^ per sementim,^'^ per ver,^^ once by piro florente,^^ as also by cum uva floret,^^ per » Eratosthenes is the source of the description of the zones and the path of the sun. Compare K. Brandt, op. cit., p. 4. "> Couat, La pofeie alexandrine, pp. 445 ff. " De Agri Cultura, 50. « lb. 61. »Ib. 121. » lb. 41, 161. solstitium,^^ per vindemiam,^^ and declares : "ficos, oleas, mala, pira, mtes inseri oportet luna silenti post meridiem sine vento austro."^^ There is not a single star to guide in the De Agri Cultura. Varro shows the man of learning in his designations. He too uses piro fiorente (I. 37.5) to date a season, but his real divisions are much more scientific. Scrofa, his mouthpiece, notes the four seasons according to the position of the sun; yet this is not sufiicient, and the year is divided into eight intervals designated thus: "a favonio ad aequinoctium vernum, hinc ad vergiliarum exortum, ab hoc ad solstitium, inde ad caniculae signum, dein ad aequinoctium autumnale, exin ad vergiliarum occasum, ab hoc ad brumam, inde ad favonium."^^ The proper activities for each division are set forth in order, and not content with this, Varro adds the divisions of the month into waxing and waning moon, and specifies things proper to one phase or another. In Vergil's guidance of the farmer first by the rising and setting of the constellations, then by the moon, and again by the sun, one is reminded of Varro's treatment, but the poet has said nothing of eight divisions, nor does he even run the risk of dryness by uniform designations of the time-divisions he does make. In the other three books of the Georgics astronomy plays no part at all, yet there is a similar enriching but restrained use of geographical and literary allusions with a kindred artistic effect. The purposes served by localizing epithets may roughly be classed as four: (1) to state a fact the mention of which is occasioned by the narrative, (2) to designate a deity by the place of birth or worship, (3) to define a common object by the place of its origin or excellence, (4) to make specific some abstract idea. Instances of the first usage should not be classed as means to a stylistic effect, except »Ib. 40. 1. »6 De Re Rustica, I. 28. in so far as an author shows a fondness for choosing his subject to afford an opportunity for their use. Despite the instance of Alexandrianism cited above (page 2) from Catullus, and his more happy use in the eleventh poem of far-off India beaten by the eastern wave, the Hyrcanians, Arabians, Parthians and Scythians, the region of the seven- mouthed Nile, the land across the Alps beside the Rhine and off to Britain's farthest edge to designate the distances to which the affection of his comrades would bid them follow him, Catullus' geographical allusions are oftener occasioned by his narrative than deliberately sought. The choice of subject matter in poems sixty-four and sixty-six is largely responsible for the definite places mentioned, though even in these a more general notion is made specific by this means. ^'^ To references occasioned by t'he narrative belong the places named in the fourth poem, though, of course, there is some choice of what will be mentioned and what passed by. The references to Spain in the poems touching Veranius and Fabullus (9. 6; 12. 14) and those to Bithynia just after his return from that province are of this direct character (10. 7; 31. 5; 46. 4, 5, 6).^^ The use of names of places to designate the divinity who is worshiped there occurs with similar frequency in Catullus and the Georgics (some ten times in Catullus and twelve times in the Georgics), but it is a greater favorite apparently with Horace, who in Book I of the Odes so designates the gods some sixteen times. The place-names so used vary little in their probable familiarity to the public; in general they are fairly well-known, and, though there is inherent in them the power of a vivid picture to one who has seen the place, their use is so frequent throughout literature, Greek as well as Roman, that they tend to become colorless synonyms for the deity's name. To Catullus Hymen is a dweller on Mount Helicon »" Catullus, 64. 35-7, 75, 105, 156, 178. »» Compare also Catullus, 29. 3, 4, 18, 19, 20; 34. 7; 35. 3, 4; 63. 2. 70, 91; 64. 5, 52, 74. 121, 172; 66. 12, 36. because he is the son of a Muse (61. 1), for which reason also he is bidden to leave the Aonian caves of the Thespian cliff (61. 27-8); Venus is colens Idalium (61. 17). The deified Arsinoe is Zephyritis because her temple was on the prom- ontory of Zephyrion (66. 57); Trivia when not seen in the sky is on Mount Latmos with Endymion, to whom there was a sanctuary there." So in 36. 11-17 Venus unnamed is identified by seven places of worship before she is called on to mark the vow as paid : Nunc o caeruleo creata ponto quae sanctum Idalium Vriosque apertos, quaeque Ancona, Cnidumque harundinosam colis, quaeque Amathunta, quaeque Golgos, quaeque Durrachium Hadriae tabernam, acceptum face redditumque votum, si non inlepidum neque invenustum est. In Horace Venus is diva potens Cypri (I. 3. 1), Cytherea (I. 4. 5); Clio's name iocosa imago is to reproduce aut in umbrosis Heliconis oris aut super Pindo gelidove in Haemo (I. 12. 5-6); Cybele is Dindymene (I. 16. 5); Dionysus, son of Semele of Thebes (I. 19. 2); Diana joys in the groves of Algidus or Erymanthus or Cragos (I. 21. 6, 7, 8). Again, Venus is regina Cnidi Paphique (I. 30. 1), and Fortuna, diva gratum quae regis Antium (I. 35. 1). In the Georgics Pan is the only deity in the opening invo- cation so localized: reference is made to his haunts on Lycaeus, at Tegea, or on Maenala (I. 16, 17, 18), while elsewhere even the planet Mercury is called Cyllenius (I. 337), and later the Indigites, Romulus and Vesta are given a home at Rome by the Tiber (I. 499). Jove is the Dictaean king (II. 536). The Muses the poet will bring from their Aonian mount (III. 11); the master of Cyllarus is Pollux of Amyclae (III. 89) ; the love of Parnassus of the Muses leads Vergil in their worship up its 19 Paus. V. 1. 5. steeps and down the gently sloping path to Castalia (III. 291 fF.). The guardian of the meadow for the bees is to be Priapus from the Hellespont (IV. Ill) ; Jove is the king in the Dictaean cave (IV. 152), and Aristaeus' father is Thymbraeus Apollo (IV. 323). Both Vergil and Horace show a great fondness for identi- fying objects with the place of their origin and for making general ideas specific by the names of certain localities. Sellar-" draws a contrast between Horace and Propertius along this line: "Though Horace, for the form of his art, for some- thing of his thought and much of his diction, goes back to Homer and the Greek lyric and tragic poets, yet in his use of Greek mythology as a kind of storehouse of romantic adven- ture, and in his numerous geographical allusions, we see that he is yielding to tastes formed and fostered by Alexandrian learning. But if we compare Horace in these respects with so thorough an Alexandrian as Propertius, we find that it is the personages and tales of mythology familiar from Homer and Pindar and the Greek tragedians, not the obscurer beings and more artificial fancies of later creation, that live for us again in the Odes, and that his geographical allusions are not introduced as so much dead learning, but give new life to his subject by names which stirred the imagination in his own day with the thought of distant lands, or wild and wandering tribes on the confines of the Empire, or seas, suggestive of the enterprise of the present time and the memories of a more adventurous past." Professor Shorey^^ classes as one of the chief compensations that relieve Horace's plainness or par- simony of vocabulary and imagery the use of proper names charged with associations of mythology, history, literature and travel. "More than seven hundred distinct proper names or adjectives," he says, "are employed in the Odes, a sixth of *" Horace and the Elegiac Poets, p. 147. 2' Shorey and Laing, Horace, Odes and Epodes, Introd., pp. xxiv-xxv. the total vocabulary. The fourth book of the Golden Trea- sury contains less than two hundred, and an equal amount of Greek lyric presents at the most three or four hundred, mostly persons known to the poet or gods directly invoked. In the learned rhetoric of Lucan and Statius mythological and geo- graphical allusion passes into the conundrum. The tact of Horace selects just those names which will arouse pleasant associations in the mind of the average educated man, and which will adorn without overloading his style." Vergil in the Georgics uses some 602 proper names,^^ ^ gj-gat many of which involve naming an object by the name of the deity in whose province it lies, as Ceres for grain, Bacchus for wine, and the winds by their specific names, Auster, Zephyrus, Boreas, etc. In Book I of the Odes Horace in 40 cases^^ describes an object by means of the place where it is indigenous or well known. Vergil in the whole of the Georgics has only 82 cases.24 Thus, cranes are Strymoniae grues (I. 120); lentils, Pelusicae lentis (I. 228); the sling is Balearis fundae (I. 309). Dodona comes to denote the acorn because of Jove's sacred oak groves there (I. 149). The laurel is Parnasia laurus (11. 18) though it is planted on an Italian farm, and the myrtle is Paphian (II. 64). Cypresses are Idaean (II. 84), and the olive is the Sicyonia baca (II. 519). So the palm is Olympic because it is won there (III. 49) ; the gad-fly is made to swarm about the groves of Silarus; the Greek games are represented by the Alpheus or by Pisa (III. 19, 180) ; to the 22 In Book I there are 145 examples; in II, 145; in III, 144; in IV, 168. « See I. 1. 3, 10, 13, 19, 28, 34; 8. 6; 14. 11; 15. 17; 16. 9; 18. 9; 19. 6; 20. 1, 9; 22. 2; 23. 10; 24. 13; 26. 9, 11; 27. 2, 5, 21; 29. 9, 15; 31. 4, 5, 6, 7, 9, 12; 33. 7; 35. 7; 36. 10, 14; 37. 5, 6, 14, 20, 30; 38. 1. 2^ I. 120, 149, 228, 265, 309; II. 18, 64, 67, 84, 88, 90, 91, 96, 97, 98, 116, 117, 120, 121, 122, 126, 143, 176, 193, 197, 198, 224, 225, 437, 438, 448, 506. 519; III.' 12, 17, 19, 33, 43-4, 49, 115, 121, 146, 147, 151, I8O2, 202, 204, 249, 255, 268. 306, 307, 312. 345, 405, 425, 450, 626, 551; IV. 41, 119, 177, 270, 283, 287, 334, 379, 380, 545. 10 tiger is given a home in Libya (III. 249), to the boar one in SabelHan territory (III. 256); while fleece is Milesian (III. 306). The dog is Amyclean and the quiver Cretan even when the African herdsman has them in the long stretches of the desert (III. 345). The epithets are hardly fitting in the last citation, but the verse that contains them both, armaque Amyclaeumque canem Cressamque pharetram, by its alliteration, its repeated -que, the way in which the vowels vary round the a and e sounds must have made Vergil deem the verse justified by its own perfection, despite its artificial touch.^^ This is the only case of so unfitting an epithet; in geographical allusions, as elsewhere, there is evi- dence in Vergil of fine discrimination in the aptness and frequency of use of such attributes. Most arbitrary of all geographical references are those used to make specific some abstract idea or more general concept. They touch the preceding classification in that there must be a certain fitness in them to their purpose, but neither the origin of the object nor the fame of the place for it deter- mines their choice. They are borrowings, conscious or un- conscious, from literary sources, or they are due to associations in the author's own experience, and their effect upon his style is the same in either case. We find Horace thus making specific any sea for traffic by the Myrtoan and again by the Icarian (C. I. 1. 14, 15), and the sea is frequently so treated. Now it is the Adriatic, on which Horace must have suffered in view of his stormy memory of it, now the Tyrrhenian.^^ Again, the Empire's foes are made vivid as the Persians and Medes (I. 2. 22, 51), or the Marsi (I. 2. 39) are used to denote a bitter foe, or the Parthians, Scythians and Indians (I. 12. 53, 56).-^ So as " See below pp. 48-63. « For the general idea of the harshness of the Adriatic, cf. C. III. 9. 23. I. 33. 15. For specific seas, cf. I. 3. 15; 11. 6; 14. 20; 16. 4; 26. 2. " Compare I. 19. 10, 12; 21. 15; 35. 40. 11 dangerous places are named Acroceraunia (I. 3. 20), the Syrtes, Caucasus, the shores of the Hydaspes (I. 22. 5, 7, 8), aequor Atlanticum (I. 31. 14). In the seventh Ode of the first book famous places of the Greek world are marshalled to give way before the delights of Tibur and the Anio. Book I of the Odes shows in all 44 instances of such arbitrary locali- zation.^^ In the Georgics the number is greater, but the proportion about the same, there being 122 instances in the 2186 verses of the four books. In the first book of the Georgics man's food before he had learned from Ceres was the acorn of Chaonia (I. 8), his drink the Achelous (I. 9). As the farmer must observe where on his farm to plant grain, grapes, fruit- trees and grass, so over the wide earth India bears one thing, Tmolus another, while from the land of the Chalybes and Pontus (I. 519) and Epirus come different products. So the fruitful land is Mysia, and more specifically Gargara (I. 102). The sailor is he who tempts the Pontus and the narrows at oyster-bearing Abydus (I. 207). Frequently the points of the compass are designated by the countries lying in the desired quarter. The north is Scythia (I. 240), the land of the Hyperboreans (III. 196), Scythia, the region of Lake Maeotis, the Hister and Rhodope (III. 349-51), or again the region lying beneath the Hyperborean Wain exposed to Eurus as he comes over the Riphaean mountains (III. 381). The south is Libya (I. 241); the east is the country of the Arabs and the Geloni (II. 115), or the land along the Euphrates (IV. 561). The birds that foretell rain are those that dive into the crevices of the Caystrian meadows (I. 384); the Italian farmer's pleasant valley is Tempe (II. 469); the poet's own chosen haunts lie along the Spercheus and on the 'revel- ground' of Taygetus (II. 487). The bees come forth thick as "C.I. 1.14, 15; 2. 13, 14, 22, 39, 51; 3. 15, 20; 7. 1-11 (12 names); 8.16; 11. 6; 12. 53, 56; 14. 20; 16. 4; 19. 10, 12; 21. 15; 22. 5, 7, 8, 14, 15; 26. 2, 3; 31. 14; 33. 15; 34. 10, 11; 35. 7, 9, 40. 12 the arrows the Parthians shoot (IV. 314) ; the farmer with a good dog fears no hostile Iberian (III. 408). In two longer passages something of the same freedom of localization is seen — in II. 136-76, in the praise of Italy and of the rich foreign lands with which it is contrasted, and in IV. 365-73, in the list of the rivers Aristaeus sees when he goes down into the halls of his ocean mother. The choice of these rivers seems most arbitrary, for, with the exception of the Tiber, the Anio and the Po, named for their nearness and dearness, they are neither great nor storied enough to justify the wonder Aristaeus feels before them. When Cicero mentions the Hypanis (Tusc. I. 39. 94) in an illustration, he takes pains to say where it is, and to point out that Aristotle is responsible for what he knows about it; references to this river elsewhere are confined largely to geographical writers. The use of the Lycus is equally unaccounted for. Some six or more rivers bear the name; Lewis and Short decide in favor of the one in Paphla- gonia for Vergil's reference here, though Servius puts it in Syria. So unidentified a stream would serve only for impres- sive strangeness. The Caicus and the Enipeus are better known, but when one thinks of all the larger and more famous rivers that the poet has passed by to name these less impressive streams, which boast not even especially musical names to justify their choice, it would seem that he is taking some combination of names that had elsewhere fallen under his eye, or that he is yielding to a bit of vanity in his erudition. The combination eludes our eye before Vergil's time, but it is easy enough to see what must have been before Ovid when he wrote Metamorphoses XV. 273-9 where the Lycus, the Hypanis and the Caicus all appear serving in another guise. The eight rivers that Vergil names are separated widely enough to designate north, east and west, but the order of their mention, which brings one from Chalcis to Paphlagonia, or elsewhere, to Thessaly, to Italy, to Scythia, to Mysia and 13 back to Italy, and the fact that no southern stream is men- tioned, do not favor the supposition that the author meant to represent the waters from all quarters of the earth meeting in Gyrene's halls. In the second book (136 ff.) the east is again called on to typify the rich lands that yet can not equal Italy. Not the forests of the Medes, not the Ganges and the Hermus, nor Bactria, nor India, nor Panchaia deserves such measure of praise. Here again the vague and far-off lands are chosen, but at the same time their fame is such that they readily give the desired effect, and the richness of sound in the closing line, totaque turiferis Panchaia pinguis harenis, sums up a mass of wealth to set against Italy's charms. In Italy itself places of fertility or beauty, or where man's work, or where the men themselves are great, are enumerated, and its superiority is sealed in the last verse, where Italy's son, victorious, is turning the Indians from Rome's citadel (II. 172). In somewhat the same fashion Vergil arbitrarily identifies the features of the lower world just as he does the regions above. In Lucretius we find the under-world designated by Orcus twice (I. 115. VI. 762), by Tartarus three times (III. 42, 966, 1012), by Acheron or its adjective Acherusia eight times (ill. 37, 86, 628, 978, 984, 1023. IV. 37, 170); there is no mention of the Styx, of Gocytus, or of Avernus, except to explain the physical nature of the lake and to disprove its con- nection with the lower world. Vergil calls the region Tartarus when he sets aside the possibility of Caesar's becoming king there (I. 36) ; next it is the dark Styx that sees the pole of heaven that is opposite our own (I. 243); then it is into Tartarus that the oak sends its roots (II. 292), and fittingly it is the noise of greedy Acheron that Nature's great poet has put beneath his feet forever, where that poet is plainly Lucre- tius (II. 492). It is the cruel stream of Gocytus that the 14 figure of Envy on Vergil's great temple to Caesar shall fear (III, 38), but from the darkness of Styx Tisiphone is sent up for her cruel proj^ress among the flocks (III. 551). It is the jaws of Hell at Taenarus that Orpheus enters (IV. 467), after which the shades of Erebus (IV. 471) are stirred by his song, and he goes on to see the streams of Cocytus and Styx and Death's secret Tartarus. But it is the waters of Avernus that echo thrice after Eurydice's recall, nor will the harbor official in Orcus suffer Orpheus to return, though Eurydice is moving off across the Styx (IV. 502-6). Vergil seems to have ex- hausted all the names his language knew for the under-world, and yet there is such skill in their use that no casual reader would ever stop to think how the poet was varying his designa- tions. The contrast with Lucretius' usage, as seen above, shows the Alexandrian touch, but there is ample restraint in the elaboration. The frequent references to the East in both Vergil and Horace would naturally be most effective at this time, when its intercourse with Rome had been so greatly increased, and even the transfer of the Empire from Rome to the East by Antony was feared. To express the ideas of distance and wealth no other countries had so good a claim. Even the poets had learned since Catullus' day that Britain had not the wealth Mamurra squandered. Where the east and north and, in general, where distant lands are named, the references savor of literature, sometimes even of technical literature, but naturally, where the places are those of Italy, the flavor is of acquaintance and experience. It is to Mantua that Vergil would bring the palms of Palestine which he will win by his triumph in poetry (III. 12), and he will raise his temple to Caesar beside the Mincius whose green fields, slow- moving bends and reed-covered banks he takes pains to mention as a fit scene for his glorious structure. A few verses farther on Greece, the Alpheus, and the groves of Molorchus, 15 in which are hidden the Nemean games, are but names with no description. The Po, king of rivers, has been known to overflow all the fields, tearing up whole forests and carrying off herds and their stalls as well (I. 481-3). Ameria (I. 265) yields staves, as Massicus and the Falernian fields and Amin- nea produce wine, and Crustumeria pears (II. 88-97). Cli- tumnus has along its banks white flocks and the great sacri- ficial bull, who bathe in its stream and later join in the Roman triumph to the temples of the gods (II. 146-8). So Larius, Benacus, the Lucrine and Avernian lakes are pictured as by one who had both seen and heard them: II. 159-64 te, Lari Maxime, teque, fluctibus et fremitu adsurgens, Benace, marino? an memorem portus Lucrinoque addita claustra atque indignatum magnis stridoribus aequor, lulia qua ponto longe sonat unda refuse Tyrrhenusque fretis immittitur aestus Averms?^' Mantua has the familiar touch as described in II. 198-9 et qualem infelix amisit Mantua campum pascentem niveos herboso flumine cycnos, and with the stretches around Tarentum it shares the following verses II. 200-202, non liquid! gregibus fontes, non gramina deerunt, et quantum longis carpent armenta diebus exigua tantum gelidus ros nocte reponet. The gad-fly about the groves of Silarus and Albernum, green with its oaks, torments the herds, which flee in fright from the forests and the woods, and the air and the banks of dwindling Tanager resound to their lowing (III. 146-51). So we know that an eye-witness is describing the old Corycian's small farm beneath the towers of Oebalia's citadel where the dark Galaesus waters the yellowing fields (IV. 125-6). 29 And in my head, for half the day, The rich Vergilian rustic measure Of Lari Maxime, all the way. Like ballad-burthen music, kept. 16 We find, then, the mystery that stimulates the imagination, the association to the cultured reader, often the pleasing sound in the names of distant places that Vergil uses knowingly to heighten his effects. Sometimes the association seems vague and the learning a little heavy, to the modern reader at least, but the real and vivid pictures of things nearer home and experience relieve the mere book learning and make one less inclined to call Vergil's use of geography Alexandrian, at least to the point of abuse. He has more of geographical reference than Lucretius, but not much more than Catullus, and no more than Horace, while they all fall short of the extent to which Propertius was imbued with love of it. The account of the constant and wide travel over the Empire described by Friedlaender^° makes one realize how familiar the names of many places even far-off would be to the Roman public, through oflBcial happenings and the common talk of the street without any acquaintance through literature proper. Real literature, too, in one of its most popular forms, the comedies of Plautus and Terence, had long before taken the privilege of wide and frequent reference to places remote from Italy in translations and adaptations of Greek originals.^^ The Scythians, Hyperboreans, Dacians and Iberians were beyond the geography of Plautus and Terence, but the East and Africa served the same purpose even then, when the touch was more purely Alexandrian than it became after the activi- ties of Lucullus, Pompey, Caesar and Augustus. In the Curculio 438-48 Curculio explains why the miles has not come in person for the meretrix: "It is only three days since the soldier and I arrived in Caria from India; he stayed there to have a statue of himself set up to commemorate his exploits, because Persas, Paphlygonas, Sinopas, Arabas, Caras, Cretanos, Syros, Rhodiam, atque Lyciam, Perediam et »" Friedlaender, Sittengeschichte, 8te Auflage, pp. 97-292, 2ter Toil. " C. Knapp, Travel in Ancient Times as Seen in Plautus and Terence, CI. PhU. II. pp. 1-24. 281-304. 17 Perbibesiam, Centauromachiam et Classiam Unomammiam, Libyamque oram omnem, omnem Conterebromniam, dimidiam partem nationum usque omnium subegit solus intra viginti dies." ^2 There is the same vagueness and greatness in the list of real places, heightened and turned broadly to ridicule in the fictitious names. Turning to literary allusions regardless of localization, one finds the case quite similar. Some book-gathered infor- mation is involved in telling the story the author has made the subject of his poem, and here Catullus again, as in the case of geographical references occasioned by the narrative, most shows his Alexandrianism, if such references are to be called Alexandrian. The subject matter of poems 64, 65 and 66 is responsible for most of the references to literature in them, though 64 and 66 show a few of different kinds. Lucretius naturally deems it necessary to the exposition of his doctrine to name and refute his predecessors in the realm of natural philosophy, so we find references to Heraclitus (I. 638), Empedocles (I. 716), Anaxagoras (I. 830), one to Ennius prized as the first of * us ' who sang, who brought from lovely Helicon a wreath of never-dying leaves, to whom the spirit of Homer spoke (I. 117). So Democritus (HI, 371, V. 622) is quoted to be refuted, and by Bahylonica Chaldaeum doc- trina (V. 727) Berosus^^ is meant. Otherwise the nature of Lucretius' subject calls for the fruits of his own observation, in- vestigation and reasoning, or that of his master, which he makes his own, and one finds no more learned allusions, except those that are introduced for illustration and stylistic purposes. In Horace, Book I of the Odes, there are two poems with themes so chosen that the underlying story is presupposed (L 10, 15). With the theme once chosen the references are like the greater number in Catullus 64, 65, 66 — details of the story the poet is telling. 32 Quoted by Knapp, op. cit., p. 281. " Merrill, Lucretius, ad loc. 18 The Georgics have the same sort of references in the story of Aristaeus in the fourth book, with all its ramifications touching Cyrene and the Ocean nymphs, Orpheus and Eury- dice. Such references are not stylistic, except in so far as an author's taste runs to such themes. Again, as with localizing allusions, the story of some deity's activities is used to name or describe that deity, as Lucretius uses the phrase Aeneadum genetrix in his opening line, or as Catullus names the sun in progenies Thiae (66. 44), or describes the Eumenides (64. 193) as Eumenides, quibus anguino redimita capillo frons expirantis praeportat pectoris iras.^* So Horace frequently designates a person: in I. 17. 22-3 where Bacchus is Semeleius Thyoneus; in I. 19. 1-2 Venus is Mater saeva Cupidinum, and Bacchus is Thebanae Semeles puer. Melpomene in I. 24. 4 is the daughter of Zeus, to whom he gave the lyre and song (Compare Hesiod, Theog. 52 ff.). Vergil uses this means of identification in Georgics 1. 19, where uncique puer monstrator aratri denotes Triptolemus and refers to the story of his invention of the plough and instruction of men in agriculture. So in the naming of the stars in I. 138 (claramque Lycaonis Arcton) the story is recalled in Lycaonis, as in the Atlantides of Eoae Atlantides (I. 221), the Gnosia of Gnosiaque ardentis decedat stella coronae (I. 222), the Inoo of Inoo Melicertae (I. 437) and in Tithoni of Tithoni croceum linquens Aurora cubile (I. 447). Again, in the phrase pastor ah Amphryso (III. 2) a story of mythology is recalled by a localizing epithet, but these are the only cases in the Georgics. In a poem so full of the thought of the imminence of God, in which the country itself is divine (I. 168), and man by toil is to thrive under divine guidance, and dare not go about his ** Compare Aesch., Coeph., 1049. Paley quotes Paus. I. 28, 6 Trpwroj 5^ ffUTiv AJirxiJXos Sp6.Kovras iirol-rfffev ofwO Ta?s iv ry Ke