STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world byJSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.istor.org/participate-istor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. VERSUS TETRACOLOS Bt Samuel E. Bassett Two ancient writers on metric call attention to the epic verse which is made up of four words: Marius Plotius, p. 505, 11. 15 ff., [Keil] (the passage forms a sort of appendix to the detailed discus- sion of the schemata) : Huius metri, id est heroici, virtutes sunt tres: si d(n5v8«Tos versus fuerit, id est sine conjunctione . . . . ; si fuerit tetracolos, id est quattuor verbis vel quibuslibet partibus orationis fuerit divisus, cuius virtutis exemplum latinum melius lectum est quam graecum, saltantis satyros imitabitur Alphesiboeus [Verg. Eel. v. 73], graecum sic, alBfOfUu jSacriA.^ rroXv^pucroio MvKijvijs (cf. A 46), sed iroXv fecit ilium quasi pentacolon, nam iroA.uxpvcros compositum nomen est; si rhopalius fuerit, qui .... povakov .... imitatur, etc. Marius Victorinus, p. 72, 1 ff . [Keil] : Insignes autem in metris sunt aut dactylici, id est cum quinque dactyli ultimo spondee clauduntur, .... aut spondiazontes, .... aut in mono- syllabum desinentes, .... aut quattuor orationis partibus decurrentes, ut comua velatarum obvertimus antemnarum [Verg. Aen. iii. 549], aut acrvvSeroi et 8io Trcirt, ut formosum pastor Corydon ardebat Alexin [Verg. Eel. ii. 1]. Modern scholars have made special studies of the schemata and the spcmdiazontes; the versus in monosyllabum desinentes and the p6ira\ov are recognized as worthy of attention in the handbooks on metric, and even the lurhvhtros has received passing mention.' But the tetracolos, or verse composed of four words, seems to have been entirely neglected. It is the purpose of this paper to make amends for this neglect by a study of the tetracoloi of the Iliad and Odyssey, and a brief statistical comparison of these with similar verses in the later Greek epic. 'Christ, Metrik' (1879), p. 179, who, strangely enough, cites Marius Victorinus as holding the luriivSeroi. to be vitiosi (cf. Marius Victorinus, p. 71, 1. 33 [Keil]). [Classical Philoloqy XIV, July, 1919] 216 Versus Tetracolos 217 Verses which contain but four words are by no means so rare in the Homeric poems as Plotius seems to imply, for the natural inference from "exemplum latinum melius lectum est quam grae- cum" is that the Greek tetracolos is found at least no more fre- quently than its Latin equivalent, which is called "noteworthy" (insignis) by Victorinus and classed as a curiosity along with the poTTokov by Plotius. Certainly in Latin, at least in Vergil, the tetracolos is rare, occurring but four times in the Aeneid (iii. 549; iv. 542; vii. 410; viii. 158). But in the Homeric poems it is quite otherwise: 431 verses contain no more than four words each {Iliad, 262 verses; Odyssey, 169 verses) ;' not one of the 48 books of the two poems is lacking in at least one example, while B contains 25 cases and '^ 19 cases. Surely a verse which occurs on the aver- age nearly four times in every 250 verses, or about once in every two pages, cannot be called a curiosity, but should be regarded as a type, and deserves to be studied along with the other character- istic features of Homeric poetry which mark it as a separate literary genre. 'Iliad: A 75, 87, 122, 189, 322, 499, 608; B 92, 113, 173, 264, 277, 288, 290, 306, 335, 403, 442, 514, 518, 541, 543, 551, 566, 624, 689, 693, 705, 706, 746, 818, 847; r 112, 122, 250, 331, 345; A 6, 33, 45, 87, 177, 228, 285, 358, 394, 402, 414, 448, 464, 511; E6, 149, 180, 415, 444, 468, 491, 526, 534, 560, 577, 649, 716, 754, 763, 779, 782, 785, 909; Z 3, 74, 204, 236, 299, 358, 395, 508, 527; H 15, 59, 166, 315, 404, 422, 453; 6 4, 42, 62, 93, 120, 187, 197, 232, 241, 264, 288, 372, 527; 1 10, 20, 51, 308, 557, 582, 624, 665, 711; K 144, 315, 547; A 18, 31, 46, 250, 279, 372, 427, 576, 694, 695, 713; M 109, 117, 128, 134, 249, 354, 379; N 24, 43, 113, 189, 203, 258, 393, 563, 635, 782, 798; S 54, 59, 355, 369, 473, 479; O 25, 77, 210, 265, 303, 378, 446, 609, 640, 678; n 125, 126, 132, 134, 174, 269, 320, 486, 496, 525, 533, 711, 797, 865; P 69, 199, 214, 259, 320, 337, 341, 369, 374, 467, 485, 675, 741, 748; S 123, 212, 260, 276, 289, 301, 315, 355, 370, 399, 418, 471, 592; T 75, 116, 123, 168, 193, 234, 269, 370; T 12, 63, 71, 85, 88, 175, 180, 212, 222, 258, 263, 295, 323, 442, 465; *28, 140, 170, 204, 278, 355, 363, 433, 469, 477, 480, 499; X 72, 132, 148, 221, 223, 413, 415; * 39, 98, 113, 124, 129, 221, 239, 264, 489, 505, 576, 584, 628, 678, 701, 747, 788, 804, 878; 395, 451, 567, 796, 798—262 verses. Odyssey: o 38, 134, 329, 429; 83, 149, 175, 189, 347, 400, 431; 76, 59, 181, 206, 364, 465; d 23, 122, 217, 234, 336, 442, 568, 631, 797, 800; e 203; f 5, 14, 17, 22, 70, 146, 213, 222, 267; v 8, 58, 93, 128, 146, 170, 241; 8, 9, 122, 191, 263, 288, 297, 327, 369, 464; i 101, 185, 503, 504, 510, 528; k 88, 106, 137, 199, 215, 331, 401, 456, 488, 492, 504, 530, 555, 565; X 16, 60, 85, 92, 166, 212, 283, 314, 349, 386, 405, 446, 473, 557, 594, 617; n 10, 100, 267, 269, 274, 453; v 87, 98, 166, 350, 373, 375; | 15, 311, 486; o 41, 52, 240, 244, 314, 399, 406, 414; v 167, 329, 426, 435, 455; p 34, 127, 220, 361, 377, 486, 562, 581, 588; 94, 148, 323, 370, 388; <^ 2, 14, 37, 71, 85, 137, 158, 164, 321; x 164, 235, 339, 441; 1^ 134, 323; a 119, 198, 240, 305, 350, 355, 378, 457, 531, 542—169 verses. 218 Samuel E. Bassett Three reasons may be given to explain the use of the long-word verse' in the Homeric poems. The first two are the same as those which are given to justify the absence of a word-end in the third foot in 335 verses^ of Homer, i.e., the presence of a compound word or a proper name within the verse.* Although Plotius might have found a tetracolos free from compound words ,in the first books of both Homeric poems (A 322; a 429), yet he is correct to a certain extent in implying a connection between the composite word and the long-word verse, for 314 verses, or nearly three-fourths, contain a compound word. Hardly less numerous are the verses in which a proper name is found (280, or nearly two- thirds). It is perhaps for this reason that the tetracoloi are slightly more common in the Iliad than in the Odyssey {Iliad, 1 to 60; Odys- sey, 1 to 72). In the Aeneid, too, three of the four examples con- tain a long proper name: Laomedonteae (iv. 542), Acrisioneis (vii. 410), Laomedontiaden (viii. 158). But neither of these reasons suffices to explain the Homeric poet's fondness for an occasional use of the verse filled with sesquipedalia verba. The real explana- tion must await a somewhat more minute examination of the verses in question. The unusual length of the words in the tetracolos* naturally leads to many metrical peculiarities. No word ends within the third foot in 32 verses, or 7 per cent, as against sUghtly more than 1 per cent for all the verses of both poems. Likewise the spondia- 1 Verses which contain only three words (B 706, A 427, 678, k 137) are classed with the tetracoloi in this paper. » A. Engelbracht, Die Caesuren des homerischen Hexameters, in Serta Harteliana (1896), 299, Anm. 2, who adds to the list given by Lehrs, Aristarchus^ (1865), pp. 394-403. « Spitzner, De Versu Heroico (1816), pp. 6 £E. « The average number of words in the Homeric verse appears to be slightly more than seven, judging by samples taken at random: A, 1-330, seven and one-half words per verse: t, 1-330, seven and one-fifth words per verse. The fewest number of words in a verse is three, the largest is fourteen. In Class. Phil., XII (1917), 100, I stated that I had not noticed a verse which contained more than thirteen words. Since then Mr. A. Shewan has kindly called my attention to p466 and a 110, both of which contain fourteen words. Mr. Shewan also sends me a versus dicolos written by Mr. Drewitt some years ago, (tlXtTO S ijxov TraTpoKaanyifTjTOio SvwKauiKotjiinixv. Versus Tetbacolos 219 zontes contribute only about 5 per cent of all the verses, but nearly 15 per cent of the tetracoloi. This is a natural result of the principle that the last two feet of a spondaic verse have a tendency toward being composed of a single word, which holds true of nearly one- half of the tetracoloi. The hephthemimeral clausula consists of a single word 27 times, or 6 per cent, as against less than 2 per cent' in the entire poems. Still longer words are found twice at the end of the verse (*■ 264; 678). The tetracolos also exhibits the syntactical peculiarities which are to be expected. Particles, so common in Homer, are entirely absent. Conjunctions and other conjunctive words are extremely rare (12 cases: dXXci, N 43; S 473; arap, A 448 = 62; avrap, o 244; Kai, * 264; i 510; n^di, T 465; oweKa, N 113; tt 426; e'iws, d 800; dTTTTOTipudep, S 59). The epithet aavvSeroi may therefore be applied to the tetracoloi with almost as much fairness as to the "versus 5to Tevre," mentioned by Victorinus in the passage quoted above. This is confirmed by evidence from the punctuation. No period is found within the tetracolos in Homer, and a colon only twice (A 448 = 62, xO'^K^oduprjKUP' arap aanrlSes 6na\6e'373; o314, 399 p 486, 588; poveovTi Soocrtraro KepBiov etvcu \i 158. Oblique case modifier ( = clause of purpose, etc.) — i; 91 £f. : XP""'*'"' ^' ^KOLTtpde Kal apfipeoL kwcs ^ouv ows 'H^atoTos cTeuf ev i8viycri wpaviSicrcriv Siofta vXai^v TO i/Jiov 8oX<{> apjua ireS^cai, where ivvofflyaxov is the object of 6nvvdi in the following verse.' b) Adjective. — Less numerous, but equally characteristic, are the epexegetical tetracoloi which are introduced by an adjective (24 cases*). While these verses show a greater variety of form than those just described because the adjective is inferior to the participle in its ability to dominate the construction of a clause, the thought is complete within the verse in every case. The adjective always ' Cf. Seymour, op. cit., p. 126. « B 264, 335, 689; T 112, 122, 345; A 177, 402, 414; E 444, 491, 779, 782, 785, 909; Z 3, 508; H 404; 6 232, 241, 372; I 51, 582, 711; K547; A 46, 576; M 134; N 258, 393; S 355; 77,265,378; H 486, 711; P 199, 214, 374; 2 260,471; T 75, 193; T 180, 295; * 204, 355, 363, 433; X 72, 221, 413; * 221, 701, 878; Q 451, 796; o38; /3 400; 5 336; ri4,22,267; ij 137, 170; 8,9,288, 327; tlOl; k199, 492. 565; X 165, 594; it 100; o 41; x 329, 455; p 127, 361, 581; t 187; (t> 137, 164; x 441; ^323; 0)119. » Also p 361. TTOiKiXov auTtpoevra 7ro8o)K«os AtaKiSao. "% 369 f . : 'J3.ai 14, 37. 226 Samuel E. Bassett while others (like the second example) are little more than the Homeric equivalent of the modern "full name." It is to be noted that tetracoloi which are similar in form, in fact the same tetracolos, may be used with different functions. Thus the tetracolos which owes its unity to the presence of a sub- stantive'may be used as purely epexegetical, as deferred subject, as object, direct or indirect, or as a vocative at the beginning of a speech. The following examples will make this clear: Epexegetical — V 387 f . : ^ 8c kot' avrrjcrriv Oe/xivr) nepiKaWw. Bipov, Kovpr] 'iKapioio iT€pLv UrjveXoTraa- Deferred subject — a 328 f . : rov 8' virtpwi66fv p€(tI arvvOcTO Otairiv &oiSr]v Kovpr) iKapioio TTipl.pll>V YlijVfXoTTua.- Indirect object — p 561 f. : "Ev/*ai', (uifia K iyib vij/j^epTfa iravr' ivejroipj. Kovpr) 'iKapioio iripL 85. Although the tetracoloi of the type which we have just been considering amount to more than one-fifth of the entire number there is only one in which any word except the proper name, or word denoting relationship, is governed syntactically by a word outside of the tetracolos: \ 385 f . : a«Tap iwii ilrvxa.'S /Jiiv dirctTKeSatr' oXAvSis aXkrj dyvrj Tlepcretftoi'ua yvvaiKibv 6r)\vrtpaiov. Here yvvauuv depends for its construction upon i^uxas in the previ- ous verse. In reality, however, it is added epexegetically to pre- Versus Tetbacolos 227 vent the hearer from forming the idea that the interview of Odysseus with all the souls is at an end. There remains to be considered a small group (20 tetracoloi) in which a common noun binds together the syntactical construction, the tetracolos standing either in apposition to an idea in the pre- ceding verse, or as "deferred" subject, object, etc. In five of these (B 514; K 315; 7 6; w 198, 378) the last part of the tetracolos is epexegetical of an idea in the previous verse; in the remaining sixteen (A 75, 499; A 6; E 649; H422; 3; M 117; 25; -!> 28; 5 442; V 98; ^ 311; p 220, 377; r 434) the thought of the verse is a unit. One of these verses deserves special attention: 8 441 f . : evOa K€v atvoTttTos \6)(0i iwXiTO' rtipt yap alvloi i,v rjv /xev yap koi cixfxuvoTepoK ebrovra cSuTai to /xfrpov TpAcs 8' ippiyija'av oirojs oiv CU0A.0V tlSov oAA' ovt' av 6 Xiyiov htivoi ovtms (So$€V, ovrt o<^is ovTos. May we not say that in addition to the order of words in S 442, the absence of a word-end in the third foot and the long words of which the verse is composed also contribute to deivirrjs, and that the 'Cf. Diomedes, p. 499, 11. 30 ff. [Keil]: "vocales (so. versus) sunt qui alte pro- ducta elocutione sonantibus litteris universam dictionem inlustrant, ut illud Pasoni- anum, Eoo Oceano Hyperion fulgurat Euro, Aroturo plaustro Boreas bacohatur aheno, Hesperio zephyro Orion volvitur* [austro], fulva Paraetonio vaga Cynthia promit Austro." 228 Samuel E. Bassett Homeric poet deliberately neglected the normal caesura and used the tetracolos occasionally for the sake of the effect ? The verse which we have just been considering (5 442) may serve to illustrate by way of summary some of the most important char- acteristics of the tetracolos in Homer. 1. If the tetracolos forms a part of the normal sentence of narra- tive poetry it stands at the end, and frequently the sentence begins at the preceding bucolic dieresis (Group I). 2. There is a strong tendency toward unity of thought in the tetracolos itself (Groups II and III). 3. The tetracolos often takes the syntactical form of a substantive (Group IIIc). The verse under discussion is used as the "deferred" subject of reipe in vs. 441. 4. The function of a majority of the tetracoloi in the Iliad and Odyssey is epexegetical. In the verse under discussion the burden of the thought, the oppressiveness of the situation, is first expressed in the bucolic clausula Tttpt yap] aivus, which is followed by the vivid and poetic epexegesis of the tetracolos, u>Kd.(mv oXowraros oS/xi;. "It was sorely oppressive — the ocean-nurtured seals' most loathe- some odor." 5. The use of the verse which is filled with three or four long words was not accidental nor due primarily to the presence of a proper name or a compound word; it was used intentionally to give a pleasing variety by contrast with the normal verse which usually contains many short particles, and sometimes for the sake of the effect. If we now review the results of our examination of the tetracoloi in the Homeric poems, we find that 85 per cent of the verses take one or other of a few typical forms r^ 1 For convenience of reference an example of each type of verse is added: Type 1, A 607f.: ^x* iiciurT(^ Sufia xtpU\vTos &ft1p^yvr|t^s ■H^atffToj TToltjctv Uvtijiruf irpairtSevfftv, Type 2, e 121 f . : otS' tfta wivres KapiraXtfXti)S iriTolrTO KOviovTts mSloio' Type 3, B 276 f.: ot diiv tuv tAXic aCru &pi]trei 6vtt&s dy^fup iffiKftnv 0affi\rias 6veiSttpiia\ot rplTaros kUv, iff60tos 0<«», MijKiarrjos vtis ToKaiovLSao Acoxros. Versus Tetraoolos 229 1. Distich tetracolos 44 verses (10 per cent) 2. Octameter tetracolos 49 verses (11 per cent) 3. Infinitive tetracolos 45 verses (10 per cent) 4. Participle tetracolos 88 verses (20 per cent) 5. Adjective tetracolos 24 verses ( 6 per cent) 6. Proper name tetracolos 116 verses (27 per cent) Total for the six types 366 verses (85 per cent) It would be interesting to know how frequently these types occur elsewhere in Homer, but this must wait until an analysis of the relation between the sentence or clause and the verse shall have been made. There are indications, however, that the six verse-types just mentioned find frequent illustration in Homeric verses which contain more than four words. For example, in a we find the following cases: Type 1, a 11, 69, 75, 153, 219, 222, 232, 443; Type 2, a 26, 33, 78, 86, 92, 106, 123, 128, 144, 200, 210, 296, 348, 363, 393; Type 3, a 39, 83, 385; Type 4, a 25, 73, 94, 105, 157, 183, 193, 202, 259, 324, 375, 415; Type 5, a 49, 54, 97, 199, 278, 312, 327; Type 6, a 72 (the first book of the Iliad gives more examples of Type 6, i.e., vss. 69, 102, 122, 489, 538, 556). It remains to compare the Iliad and Odyssey with later Greek epic poetry in respect to the use of the tetracolos. For the sake of brevity the results are given in tabular form (see p. 230 f.). Table I gives the number of verses in each poem or author; the number of tetracoloi; the ratio of tetracoloi to the whole number of verses; the number of tetracoloi in each of the three groups which have been described (narrative, intermediate, epexegetical), and the percentage, shown in parenthesis, of all the tetracoloi in each group. Table II shows the number and percentage, shown in parenthesis, of tetracoloi which take any one of the six typical forms just described. Of course too much weight is not to be given to these figures,^ especially in those poems which are comparatively short. But some generalizations may be made with safety. 1. Certain kinds of hexameter verse, e.g., the bucolic and mimetic idyls, in which there is much dialogue, or the Works and Days, ' No claim is made for more than relative accuracy in the figures for post-Homeric poetry; in most cases the poems have been read but once in making the collection, and undoubtedly some tetracoloi have escaped observation. But this only makes the contrast between Homer and the later epic more striking. 230 Samuel E. Bassett where the gnomic character of the thought frequently isolates single verses or groups of verses, use the long-word verse rather sparingly. In Hymns, however, it is used quite as frequently as in the true epic. TABLE I Name of Poet Total Number Number of Tetra- of Verses coloi 27,803 431 2,326 56 2,330 53 1,022 23 828 13 480 17 303 5 3,344 49 2,377 27 876 8 443 2 1,058 17 1,154 43 610 26 941 15 5,835 200 1,588 66 3,506 134 2,144 102 2,992 102 1,384 36 768 34 1,096 134 8,770 229 6,615 437 341 19 691 41 392 24 Ratio to Number of Verses Group I (Narra- tive) Group II (Inter- mediate) Group III (Epexegetl- cal) Homer, II, and Od. Homeric Hymns^ . . . Hesiod, Theog., Works and Days, Shield Theog Works and Days. . . Shield Batrachomyomachia . . Bucolic Poets TheocritusJ Buc. Idyls Mim. Idyls Epic Idyls AratusJ Maximus Philos Callimachus ApoUonlus Rhodius. . Nicander, Theriaca and Alexiphar-mica . Oppian, Halieutica . . . lOppian] Cynegetica,, . Manetho Orphica Argonautica Lithica Hymns Qulntus Smyrnaeus . . Nonnus, Dionysiaca, Books i-xivl | Musaeus Tryphlodorus Colluthus 1:65 1:42 1:44 1:44 1:64 1:28 1:61 1:68 1:88 1:110 1:222 1:62 1:27 1:23 1:63 1:29 1:24 1:26 1:21 1:29 1:38 23 8 38 15 18 17 1:16 138(32) 25(45) 26(49) 10(43) 9(70) 7(41) 1(20) 30(61) 22(81) 7(88) 1(50) 14(82) 36(84) 13(50) 11(73) 156(78) 49(74) 91(68) 46(45) 56(55) 18(50) 26(76) 8( 6) 142(62) 247(57) 9(47) 27(66) 14(58) 67(16) 5( 9) 4( 8) 2(15) 2(12) 5(10) 3(11) 1(50) 2(12) 4( 9) 4(15) 2(13) 18( 9) 5( 8) 14(10) 15(15) 3( 3) 7(19) 4(12) 29(13) 40( 9) K 2) K 4) 226(52)* 26(46) 23(43) 13(57) 2(15) 8(47) 4(80) 14(29) 2( 7) 1(12) K 6) 3( 7) 9(35) 2(13) 26(13) 12(18) 29(22) 41 (40) 43(42) 11(31) 4(12) 126(94) 58(25) 150(34) 10(53) 13(32) 9(38) * The twelve verses mentioned on p. 223 . which exhibit no enjambement, have been transferred to Group I because they are narrative rather than epexegetical in function. The same principle has been followed in the statistics for the later poetry. t The figures for the shorter Hymns (vi-xxxiv), some of which are late and show a similarity to the Orphic Hymns in the use of the tetracolos (see pp. 231, 233), are as follows: total number of verses, 412; number of tetracoloi, 16; ratio to number of verses, 1 :36; Group I, 6(38 per cent); Group II, 1(6 per cent); Group III, 9(56 per cent). J Kunst, De Theocriti versu heroico (Leipzig, 1886), p. 9, has been followed in the classification of the Idyls, § Phenomena and Diosemeia. II The collection of tetracoloi stopped with Book xiv because it was not thought necessary to read the entire poem; the first fourteen books contain approximately the same number of tetracoloi as both the Iliad and Odyssey. 2. With the exceptions just stated, the use of the tetracolos steadily increases after Homer. It is about twice as frequent in Apollonius Rhodius as in Homer, and in Nonnus and his followers nearly four times as frequent. Versus Tetracolos 231 3. The comparative numbers of tetracoloi in the three groups in the later epic show a decided change. The narrative group gains largely, and at the expense of the epexegetical group; the inter- mediate group shows only minor fluctuations. 4. The six types of tetracoloi continue to suffice for a majority of the later tetracoloi, although naturally the percentage is not so large, but the popularity of some of the types varies considerably. TABLE II Name of Poet Number Type 1 (Distich) Type 2 Type 3 Type 4 Type 5 Type 6 of Tetra- (Octam- (Infini- (Parti- (Adjec- (Proper coloi eter) tive) ciple) tive) Name) 431* 44(10) 49(11) 45(10) 88(20) 24 ( 6) 116(26) 56 9(16) 6(11) 2( 4) 8(14) 5( 9) 6(11) 53 11(20) 3( 6) 7(13) 6(11) 8(15) 23 5(22) 3(13) 4(17) 5(22) 13 3(23) 3(23) K 8) 17 3(18) 3(18) 2(12) 3(18) 5 1(20) 1(20) 2(40) 49 11(22) 7(14) 1( 2) 5(10) K 2) 5(10) 43 9(21) 5(12) h\i\ 2( 5) 26 5(19) 3(12) 4(15) 3(12) 15 K 7) 25(13) 5(33) 1( 7) 200 86(43) 7( 4) 15( 8) 2( 1) 5( 3) 66 13(20) 14(21) 9(14) 2( 3) 1( 2) 134 23(17) 30(22) 5( 4) 22(16) 5( 4) 102 16(16) 12(12) 7( 7) 21(21) 16(15) 102 10(10) 11(11) 12(12) 14(14) 36 4(11) 4(11) 1( 3) 6(17) 3( 8) 34 11(32) 4(12) 2( 6) 4(12) K 3) 134 1( 1) 14(10) 84(63) 10( 7) 229 33(14) 64(28) 11( 5) 54(24) 3( 1) 437 62(14) 44(10) 115(26) 5( 1) 9( 2) 19 3(16) K 5) 7(37) 41 9(22) 6(15) 11(27) 24 9(38) 1( 4) 6(25) 1( 4) 1( 4) Number of Tetra- coloi in the Six Types Homer Homeric Hymns Hesiod Theogony Works and Days. . . Shield Batrachomyomachia . Bucolic Poets Aratus Maximus Pliilos Callimachus Apollonlus Rhodius. . Nicander Oppian Halieut [Oppian] Cyn Manetho Orphica Argonautica Lithica Hymns Quintus Smyrnaeus. , Nonnus, Dionysiaca i— xiv Musaeus Tryphiodorus Colluthus 366(85) 36(64) 35(65) 17(74) 7(54) 11(65) 4(80) 30(61) 17(40) 17(65) 7(47) 147(78) 39(59) 85(63) 71(70) 47(47) 18(50) 22(65) 109(81) 165(72) 235(54) 11(58) 26(63) 18(75) * One tetracolos (17 137) was overlooked when this article was written. The whole number should be 432. Apollonius and Quintus are decidedly partial to the octameter; the Nonnians and Quintus are almost equally fond of the participial type, which, except in Aratus and Apollonius, is generally well represented; in the Orphic Hymns more than half of the tetracoloi are of the adjec- tive type. On the other hand the proper name type, which leads the others in Homer, shows a decHne in popularity beginning with Hesiodic poetry, and with the beginning of the Alexandrian period becomes practically negligible. The infinitive type is not represented by a single verse in the Hesiodic poetry or in the first fourteen books 232 Samuel E. Bassett of Nonnus; more tetracoloi of this type are found in Homer than in all the later epic which was examined in making up the tables. Other features of the individual use of the tetracoloi by later poets will be noted below. BEMAEKS ON TABLE II 1. The fondness of the epic poets generally for the octameter type is due to the prevailing tendency to make a decided pause in the thought at the bucolic dieresis. This is indicated in many ways: (1) in Hesiodic poetry the pause at the bucolic dieresis is less frequent than in the Iliad and Odyssey, and likewise the percentage of octameters is small. (2) In the bucolic poets tetracoloi are rarely found in the bucolic and mimetic idyls, and in the epic idyls the bucolic pause is avoided to a considerable extent; hence the distich is more frequent than the octameter. The same is true of Aratus (and Maximus), who alone of Alexandrian poets make a pause in the thought at the bucolic dieresis less frequently than the Homeric poet.^ (3) Nonnus, whose fondness for the third trochaic pause is well known, uses the narrative tetracolos 69 times in a sentence which begins at the feminine caesura of the preceding verse, a greater number than of any other of the typical tetracoloi in his poetry, and more than half of all the verses of this kind in the epic poetry which we have studied. This is a striking indication of the bearing of the pause in the thought upon the types of tetracoloi, and, conversely, of the importance of the study of the tetracoloi if we are to understand the relation of the thought to the metrical pauses. 2. Rarely does the sentence containing a narrative tetracolos begin at the preceding hephthemimeral pause. There is a single instance in Homer (A 250), none in Hesiod, Batrachomyomachia, Callimachus, Maximus, Oppian, Quintus Smyrnaeus, Nonnus, CoUuthus, and Musaeus, and only eleven instances in the 75,000 lines of poetry which we have examined ([Theoc] XXV. 208; Ap. Rhod. iii. 758; iv. 705; Meander Ther. 289, 742, 780; Manetho, y 134; Orph. Argon. 533, 1312; Lith. 239; Tryph. 10). This has an important bearing on the relation of the hephthemimeral pause to the bucohc dieresis, which cannot be overlooked in discussing the general question of caesura. 3. Both the triemimeral pause and the pause after the first trochee are preferred to the hephthemimeral for the beginning of the sentence containing a narrative tetracolos; other pauses are very rarely used in this way: (o) triemimeral (28 cases): Hesiod Works 386; Aratus 734, 1136; Ap. Rhod. i. 555, 1134; ii. 878; iii. 1019; iv. 1257, 1371, 1688; Oppian Halieut. ii. 483; iii, 25, 226; iv. 682; [Oppian] Cyn. iv. 329; Manetho, o- 428; Orph. Argon. 1304; Quint. Smyrn. i. 544; ii. 483; iv. 149, 382; viii. 12; Nonnus ii. 256, 368, 429; xi, 158, 399; xiii. 308; (6) dieresis after the first foot (13 cases): I Cf. Trana. Amer. Phil. Asao., XXXVI (1905), 111. Versus Tetracolos 233 Aratus 333; ii. 996; iv. 1000, 1605; Maximus 51, 318; Nicander Akxiph. 606; Orph. Lith. 585; Quint. Smyrn. v. 176; xii. 475; xiii. 381; Nonnus i. 360; (c) after the first trochee: Ap. Rhod. i. 959; Quint. Smyrn. vi. 123; X. 303; (d) in the middle of the fifth foot: Manetho, £ 24; (e) after the fifth trochee: [Theoc] xxi. 5. The following scheme shows the comparative frequency with which the sentence containing the narrative tetracolos begins after the various pauses of the preceding verse : (a) in Homer: 44 4 5 5 1 49 18 (6) in post-Homeric epic poetry: 276 213 28 44126' 11 2919 1 102 A 4. The tetracolos consisting of four adjectives deserves special attention. It is found once in Homer (O 406), Hesiod (Theog. 925), Batr. (295); twice in the Horn. Hymns (viii. 2; xix. 37); the Alexandrians avoid it entirely. It occurs more frequently in [Oppian] Cyn. (ii. 102, 103, 104, 177, 178, 423, 607; iv. 235) and in Manetho (8 58, 283, 307, 563, e 199), and finally is the most common type of verse in the Orphic Hymns, being found 57 times, or once in every 20 verses. Univehsitt op Vermont ' Of which 69 are in Nonnua Dionysiaca i-xiv.