L334 - Varro -- On the Latin Language II: Books 8-10. Fragments

m mTHE LIBRARYofVICTORIA UNIVERSITYToronto

THE LOEB CLASSICAL LIBRARYFOUNDED BV JAMES I.OEB, I.I.. D.EDITEDBVt T. E. PAGE, C.H., LITT.D.E. CAPPS, PH.D., LL.D. W. H. D. ROUSE, litt.d.VARROON THE LATIN LANGUAGEII

VAREOON THE LATIN LANGUAGEWITH AN ENGLISH TRANSLATION BYROLAND G. KENT, Ph.D.PBOFESSOR OF COMPARATIVE PHILOLOGY IX THEUKIVEBSITY OF PENNSYLVANIAIN TWO VOLUMESIIBOOKS VIII.-X.FRAGMENTSCAMBRIDGE, MASSACHUSETTSHARVARD UNIVERSITY PRESSLONDONWILLIAM HEINEMANN LTDMCMXXXVIII

Printed in Great Britain

CONTENTSPAOBDe Lingua Latina, Text and TranslationBook Vm. 370Book IX. ....... 440BookX 534Fragments ...... 598Comparative Table of the Fragment Numbers 630IndexesIndex of Authors and Works . . .631Index of Latin Words and Phrases . . 634Index of Greek Words .... 675

VARROVOL. n B 369

M. TERENTI VARRONISDE LINGUA LATINALIBER Vll EXPLICIT ; INCIPITLIBER VIIIQUAE DICANTUR CUR NON SIT ANALOGIA LIBER II. 1. QuoM oratio natura tripartita esset, ut superioribuslibris ostendi, cuius prima pars, quemadmodumvocabula rebus essent imposita, secunda, quopacto de his declinata in discrimina iermt,^ tertia, utea inter se ratione coniuncta sententiam efFerant,prima parte exposita de secunda incipiam hinc. Utpropago omnis natura secunda, quod prius illudrectum, unde ea, sic declinata :verbis :rectum homo, obliquum hominis, quod declinatuma recto.itaque declinatur in§ 1. ^ Sciop.,/or ierunt.§ 1. " That is, bent aside and downward, from the vertical.The Greeks conceived the paradigm of the noun as the upperright quadrant of a circle : the nominative was the verticalradius, and the other cases were radii which ' declined ' tothe right, and were therefore called irraxjeis ' fallings,' whichthe Romans translated literalh' by casus. The casus rectusis therefore a contradiction in itself. The Latin verb de-370

MARCUS TERENTIUS VARRO'SON THE LATIN LANGUAGEBOOK VIIENDS HERE, AND HERE BEGINSBOOK VIIIOne Book of Arguments which are advancedAGAINST the EXISTENCE OF THEPrinciple of AnalogyI. 1. Speech is naturally di\ided into three parts,as I have shown in the previous books : its first partis how names were imposed upon things ; its second,in what way the derivatives of these names havearrived at their differences ; its third, how the words,when united with one another reasoningly, express anidea. Having set forth the first part, I shall fromhere begin upon the second. As every offshoot issecondary by nature, because that vertical trunk fromwhich it comes is primary, and it is thereforedeclined " : so there is declension in words : homo''man is the vertical, hominis man's ' ' is the obHque,because it is declined from the vertical.clinare is used in the meanings ' to decline (a noun),' ' toconjugate (a verb),' and ' to derive ' in general, as well as'to bend aside and down ' in a literal physical sense : ittherefore offers great difficulties in translating.371

VARRO2. De huiusce(modi)i multiplici natura disci'iminum?/*ae^ sunt hae, cur et quo et quemadmodumin loquendo declinata sunt verba. De quibus duoprima duabus causis percurram breviter, quod et turn,cum de copia verborum scribam, erit retractandum etquod de tribus tertium quod est habet suas permultasac magnas partes.II. 3. Declinatio inducta in sermones non solumLatinos, sed omnium hominum utili et necessaria decausa : nisi enim ita esset factum, neque di(s)cere^tantum numerum verborum possemus (infinitae enimsunt naturae in quas ea declinantur) neque quaedidicissemus, ex his, quae inter se rerum cognatioesset, appareret. At nunc ideo videmus, quod simileest, quod propagatum : legi um (de lego)^ declinatumest, duo simul apparent, quodam modoeadem dici et non eodem tempore factum ; at* siverbi gratia alterum horum diceretur Priamus, alterum//ecuba, nullam unitatem adsignificaret, quae apparetin lego et legi et in Priamus Priamo.4. Ut in hominibus quaedam sunt agnationes ac^gentilitates, sic in verbis : ut enim ab ^emilio hominesorti ^emilii ac gentiles, sic ab ^iemilii nomine declinataevoces in gentilitate nominali : ab eo enim,§ 2. ^ Added by L. Sp. * L. Sp., for orae.§3. ^ Mite., for d'lcere ; cf. § 5. '08., for legium F ;cf. declinatum est ab lego Aug. from B, and last sentence ofthis section. ^ Mue., for ut.§ 4. ^ L. Sp., for ad.§ 2. " Cf. viii. 9 in quas. * That is, the collectivevocabulary.§ 3. " The term ' inflection ' will be convenient oftentimesto express declinatio, including both declension of nouns andconjugation of verbs.372

ON THE LATIN LANGUAGE, VIIL 2-42. From the manifold nature of this sort thereare these causes of the differences : for what reason,and to what product," and in what way, in speaking,the words are dechned. The first two of these I shallpass over briefly, for two reasons : because there willhave to be a rehandling of the topics when I write ofthe stock of words,* and because the third of them hasnumerous and extensive subdi\isions of its own.IL 3. Inflection has been introduced not only

VARROquod est impositum recto casu ^emilius, orta ^emilii,Jemilium, ^emilios, ^emiliorum et sic reliquae eiusdemquae sunt stirpis.5. Duo igitur omnino verborum principia, impositio(et declinatio),^ alterum ut fons, alterum utrivus. Impositicia nomina esse voluerunt quampaucissima, quo citius ediscere possent, declinataquam plurima, quo facilius omnes quibus ad usumopus essef* dicerent.'6. Ad illud genus, quod prius, historia opus est :nisi djscendo^ enim aliter id non^ pervenit ad nos ; adreliquum genus, quod posterius, ars : ad quam opusest paucis praeceptis quae sunt brevia. Qua enimratione in uno vocabulo declinare didiceris, in infinitenumero nomiinum uti possis : itaque novis nominibusallatis^ {inY consuetudinem sine dubitatione eorumdeclinatus statim omnis dicit populus ; etiam novlciiservi empti in magna familia cito omnium conservorumom(i>na* recto casu accepto in reliquosobliquos declinant.7. Qui s* declinaretur, ab homine homines ;§ 5. ^ Added hy L. Sp., V, p. * Canal, for essent.' Ed. Veneta, for dicerentur.§ 6, ^ Stephanus, for descendendo. " For idum.' For allatius, * Added by Aug. * Aug., for omnes.§ 7. 1 Aldus, for quid. " Aldus, for ilia. » Ellis,for putant. , * -dinem H, for -dine F and other codd.§ 7. ° That is, in the singular.374

ON THE LATIN LANGUAGE, MIL 4-7was imposed in the nominative case as Aemilius weremade Aemilii, Aemilium, Aemilios, Aemiliorum, and inthis way also all the other words which are of thissame line.5. The origins of words are therefore two in number,and no more : imposition and inflection ; the oneis as it were the spring, the other the brook. Menhave wished that imposed nouns should be as few aspossible, that they might be able to learn them morequickly ; but derivative nouns they have ^\ished to beas numerous as possible, that all might the more easilysay those nouns which they needed to use.6. In connexion with the first class, a historicalnarrative is necessary, for except by outright learningsuch words do not reach us ; for the other class, thesecond, a grammatical treatment is necessary, and forthis there is need of a few brief maxims. For thescheme by which you have learned to inflect in theinstance of one noun, you can employ in a countlessnumber of nouns : therefore when new nouns havebeen brought into common use, the whole people atonce utters their declined forms without any hesitation.Moreover, those who have freshly become slavesand on purchase become members of a large household,quickly inflect the names of all their fellowslavesin the oblique cases, provided only they haveheard the nominative.7. If they sometimes make mistakes, it is notastonishing. Even those who first imposed namesupon things perhaps made some slips in some instances: for they are supposed to have desired todesignate things individually," that from these inflectionmight be made to indicate plurality, as homines'men from homo man.' They ' ' are supposed to have375

VARROsic mares liberos voluisse notari, ut ex his feminaedeclinarentur, ut est ab Terentio Terentia ; sic inrecto casu quas imponerent voces, ut illinc e sentfuturae quo declinarentur : sed haec in omnibustenere nequisse, quod et una(e> et » dicunturscopae, et mas et femina aquila, et recto et obliquovocabulo vis.8. Cur haec non tam si(n>ti in culpa quam putant,pleraque solvere non difficile, sed nunc non necesse :non enim qui potuerint adsequi sed qui voluerint, adhoc quod propositum refert,quod nihilo minus* declinaripotest ab eo quod imposuerunt' scopae scopa-. Sp., forlegis ; cf. § 3 end.^ The genitive.376

ON THE LATIN LANGUAGE, VIIL 7-9desired that male children be designated in such away that from these the females might be indicatedby inflection, as the feminine Terentia from themasculine Tereniivs ; and that similarly from thenames which they set in the nominative case, theremight be other forms to which they could arrive byinflection. But they are supposed to have beenunable to hold fast to these principles in everything,because the plural form scopae denotes eitherone or two brooms, and aquila eagle denotes both' 'the male and the female, and vis force ' ' is usedfor the nominative and for an oblique case ^ ofthe word.8. Why such words are not so much at fault asmen think, it is in most instances not hard to explain,but it is not necessarv to do so at this time ; for it isnot how they have been able to arrive at the words,but how they wished to express themselves, that is ofimport for the subject which is before us : inasmuchas genitive scoparum can be no less easily derived fromthe plural scopae which they did impose on the objectas its name, than if they had given it the name scopain the singular, and made the genitive scopae fromthis—and other words Uke^^ise.in. 9- The reason, I say, why they made theseinflected forms ° from the names which they had setupon things, is that which I have shown ; the nextpoint is for me to sketch by classes, but briefly, theforms " at which they have wished to arrive by inflection,or have not wished to arrive. For there are twoclasses of words, one fruitful, which by inflection producesfrom itself many different forms, as for examplelego ' I gather,' legi ' I have gathered,' legam ' I shall§ 9. " Understand voces with eas and with qnas.377

VARROsic alia, alterum genus sterile, quod ex se parit nihil,*ut est et iam* vix era*' magis cur.10. Quarum rerum usus erat simplex, (simplex)^ibi etiam vocabuli declinatus, ut in qua domo unusservus, uno servili op?s/,* quae imposita essentvocabula rebus, ne ab omnibus his declina^us^" puta-IV. 11. Quorum^ generum declinationes oriantur,partes orationis sunt duae,

ON THE LATIN LANGUAGE, VIIL 9-11gather,' and similarly other words ; and a secondclass which is barren,* which produces nothing fromitself, as for example et ' and,' iatn now,' ' vix ' hardly,'eras to-morrow,' magis more,' cur 'why.'' '10. In those things whose use was simple, theinflection of the name also was simple ;just as in ahouse where there is only one slave there is need ofonly one slave-name," but in a house where there aremany slaves there is need of many such names. Thereforealso in those things which are names, because thedifferentiations of the word are several, there are moreoffshoots, and in those things which are connectivesand join words, because there was no need for themto be inflected into several forms, the words generallyhave but one form : for with one and the same thongyou can fasten a man or a horse or anything else,whatever it is, which can be fastened to somethingelse. Thus, for example, we say in our talking," Tullius et'and Antonius were ' consuls " * : withthat same et we can link together any set of two consuls,or—to put it more strongly—any and all names,and even all words, while all the time that one-syllabledprop-word et remains unchanged. Therefore undernature's guidance it has come about that we shouldnot think that there are inflected forms from all thesenames which have been set upon things.''IV. 11. In the word-classes in which inflectionsmay develop, the parts of speech are two, unless,following Dion," we divide into three divisions theideas which are indicated by words : one division§11. "An Academic philosopher of Alexandria, whoheaded an embassy to Rome in 56 to seek help against theexiled king Ptolemy Auletes, and was there poisoned by theking's agents.379

VARROunam' quae adsignificat ca^us,* alteram^ quae tempera,tertia

ON THE LATIN LANGUAGE, VIIL ll-Uwhich indicates also case, a second which indicatesalso time, a third which indicates neither.* Of these,Aristotle says that there are two parts of speech '^ ;nouns, Uke homo man ' 'and equus ' horse,' and verbs,hke legit gathers and ' ' currit ' runs.'12. Of the two kinds, noun and verb, certainwords are primary and certain are secondary ° :primary Uke homo man ' 'and scribit ' MTites,' andsecondary Uke doctus learned and ' ' docte ' learnedly,'for we say homo doctus ' 'a learned man and scribitdocte ' writes learnedly.' These ideas are attendedby those of place and time, because neither hofno norscribit can be asserted without the presupposition ofplace and of time — yet in such a way that place ismore closely associated with the idea of the nounhomo, and time more closely with the act of ^\Titing.13. Since among these the noun is first—for thenoun comes ahead of the verb," and the other wordsstand later relatively to the noun and the verb—thenouns are accordingly first. Therefore I shall speakof the form-variations ^ of nouns before I take upthose of verbs.V. 14. Nouns are varied in form either to showdifferences in those things of which they are thenames, as the woman's name Terentia from the man'sname Terentius, or to denote those things outside, ofwhich they are not the names, as equiso stable-boy ''from equus ' horse.' To show differences in themselvesthey are varied in form either on account of thenature of the thing itself about which mention is' verb ' specifically, Varro here writes verbum temporale toavoid any ambiguitj'. * Declinatio denotes not only declension,but conjugation of verbs, derivation by prefixesand suffixes, and composition.381

VARROqua' dicitur aut propter illius * qui dicit.Propter ipsius rei discrimina, aut ab toto (aut a parte.Quae a toto, declinata sunt aut propter multitudinemaut propter exiguitatem. Propter exiguitatem),^ utab honiine homunculus, ab capite capitulum ; proptermultitudinem, ut ab honiine homines ; ab eo (abeo)*quod alii dicunt cervices et id Hortensius in poematiscervix.15. Quae a parte* declinata, aut a corpore, ut amamma mammosae, a manu manubria, aut ab animo,ut a prudentia prudens,* ab ingenio ingeniosi.Haec sine agitationibus ; at ubi motus maiores, itemab animo (aut a corpore),* ut ab strenuitate et nobilitatestrenui et nobiles, sic a pugnando et currendopugiles et cursores. Ut aliae declinationes ab animo,aliae a corpore, sic aliae quae extra hominem, utpecuniosi, agrarii, quod foris pecunia et ager.VI. 16. Propter eorum qui dicunt usum* dechnaticasus, uti is qui de altero diceret, distinguere posset,' Vertranius,/or quo. * Added by GS., following Reitzenstein,who added it after dicit. * Added by Reitzenstein ;aut a parte, ab toto added by L. Sp., after Aug., whoadded aut a parte, a toto, suggested to him by B aut a parteaut ab animo. a toto. • Added by Fay.§ 15. * For aperte. * L. Sp.,/or prudens. ' Addedby L. Sp.§16. ^ F

'runnersON THE LATIN LANGUAGE, VIIL 14^16made, or on account of the use to which the speakerputs the word." On account of differences in thething itself, the variation is made either \vith referenceto the whole thing, or with reference to a part of it.Those forms which concern the whole are derivedeither on account of pluraUty or on account of smallness.*"On account of smallness, homunculus manikin'' is formed from homo man,' and ' capitulum ' Uttle'head from ' caput ' head.' On account of plurality,homines men ''is made from homo man ' ' ; I pass bythe fact that others use cervices back of the neck ' ' inthe plural, and Hortensius in his poems uses oke of another383

VARROcum vocaret, cum daret, cum accusaret, sic aliaeiusdem (modi)^ discrimina, quae nos et Graecos addeclinandum duxerunt. Sine* controversia (suntobliqui, qui nascuntur a recto : unde rectus an sitcasus)* sunt qui quae(rant. Nos vero sex habemus,Graeci quinque) * : quis vocetur, ut //ercules ;quemadmodumvocetur, ut //ercule ;quo vocetur, ut ad//erculem ; a quo vocetur, ut ab //ercule ; cui vocetur,ut //erculi ;cuius vocetur, ut //erculis.VII. 17. Propter ea verba quae erant proinde accognomina, ut prudens, candidus, strenuus, quod inhis praeterea sunt discrimina propter incrementum,quod maius aut minus in his esse potest, accessitdeclinationum genus, ut a candido candidius candidissimumsic a longo, divite, id genus ahis ut fieret.18. Quae in eas res quae extrinsecus declinantur,sunt ab equo equile, ab ovibus ovile, sic alia : haeccontraria illis quae supra dicta, ut a pecunia pecunio-^ Added by Mtie. ^ For sinae. * Added by Schoellapud GS. ; cf. note b.§ 16. " Vocative, dative, accusative cases ; the accusativewas in Latin a poorly named case, through a mistranslationof its Greek name. * The only controversy was whetheror not the nominative was to be called a case, and thetext must be expanded to conform to this basic fact ; cf.Charisius, i. 154. 6-8 Keil, Priscian, ii. 185. 12-14 Keil, etc.Cf. viii. 1 note a, above." The Greeks had no ablativecase.§ 17. " Nowhere recorded as a cognomen, despite Varro.*"Recorded as a cognomen in the Claudian and the Juliangentes, and in several others. " Not recorded as a cognomen.* Namely, comparison of adjectives. ' Forsuch cognomina, cf. Fulvitis Nobilior and Fabius Maximus.' i.e., adjectives.384

ON THE LATIN LANGUAGE, VIIL 16-18might be able to make a distinction when he wascalling, when he was giving, when he was accusing,"and other differences of this same sort, which led us aswell as the Greeks to the declension of nouns. Theoblique forms which develop from the nominative arewithout dispute to be called cases ; but there arethose who question whether the nominative is properlya case.^ At any rate, we have six forms, and theGreeks five "^ : he who is called, as (nominative) Hercules; how the calling is done, as (vocative) Hercule ;whither there is a calling, as to (accusative) Herculem ;by whom the calling is done, as by (ablative) Hercule ;to or for whom there is a calling, as to or for (dative)Herculi ; of whom the calling or called object is, as of(genitive) Herculis.Vn. 17. There are certain words which are likeadded family names, such as Prudens " prudent,''Candidus ^ ' frank,' Strenuus " 'brisk,' and in themdifferences may be shown by a suffix, since the qualitymay be present in them to a greater or a smallerdegree : therefore to these words a kind of inflection **is attached, so that from candidum ' shining whitecomes the comparative candidius and the superlativecandidissimum,^ formed in the same way as similarforms from longum ' long,' dives ' rich,* and otherwords of this kind.^18. The terms which are derived for applicationto exterior objects, are for example equile ' horsestable' from equus ' horse,' ovile ' sheepfold ' fromoves ' sheep,' and others in this same way ; these arethe opposite of those which I mentioned above," such§ 18. " Here, objects named by derivation from livingbeings ; in § 15, living beings named by derivation frominanimate objects.VOL. II c 385

VARROsus, ab urbe urbanus, ab atro atratus :ut nonnunquamab homine locus, ab eo loco homo, ut ab RomuloRoma, ab Roma Romanus.19. Aliquot modis declinata ea quae foris : namaliter qui a maioribus suis, Laton(i>usi et Priamidae,aliter quae (a)^ facto, ut a praedando praeda, amerendo merces ;difficile ;omitto.sic alia sunt, quae circum ire nonsed quod genus iam videtur et alia urgent,VIII. 20. In verborum genere quae tempora adsignificant,quod ea erant tria, praeteritum, praesens,futurum, declinatio facienda fuit triplex, ut ab salutosalutabam, salutabo ; cum item personarum naturatriplex asset, qui loqueretur,

"ON THE LATIN LANGUAGE, VIIL 18-21as pecuniosiis moneyed man ' 'from pecunia money,''urbanus ' 'city man from vrbs ' city,' atratus clad ' in'booty ' from praedari ' to pillage and ' merces wages ' 'mourning ' from atrutn ' black.' Thus sometimes aplace is named from a man, and then a man from thisplace, as Rome from Romulus ^ and then Romanfrom Rome.19. The nouns which relate to exterior objects arederived in sundry ways : those like Latonius ' Latona'schild ' " and Priamidae ' Priam's sons,' ^ which arederived from the names of their progenitors, areformed in one way, and those which come from anaction are made in another way, such as praedafrom mereri to ' earn.' In the same way there arestill others, which can be enumerated \\ithout difficulty; but because this category of words is nowclear to the understanding and other matters pressfor attention, I pass them by.VIIL 20. Inasmuch as in the class of words whichindicate also time-ideas " there were these threetime-ideas, past, present, and future, there had to bethree sets of derived forms, as from the present saluto'I salute there are the past salutabam and the ' futuresalutabo. Since the persons of the verb were likewiseof three natures, the one who was speaking, the oneto whom the speaking was done, and the one aboutwhom the speaking took place, there are these derivativeforms of each and every verb ; and these forms\\\\\ be expounded in the account of the stock of verbswhich is in use.IX. 21 . Since two points have been discussed, whyderivation exists and to what products it eventuates,the remaining third point shall now be spoken of,namely, how and in what manner derivation takes387

VARROquemadmodum,nuncdic^tur.* Declinationum generasunt duo, voluntarium et naturale ; voluntarium est,quo ut cuiusque tulit voluntas declinavit. Sic trescum emerunt Ephesi singulos servos, nonnunquamalius declinat nomen ab eo qui vendit Artemidorus,atque Artemam appellat, alius a regione quod ibiemit, ab Iona* lona,* alius quod Ephesi Ephesium,sic alius ab alia aliqua re, ut visum est.22. Contra naturalem declinationem dico, quaenon a singulorum oritur voluntate, sed a com(m>uniconsensu. Itaque omnes impositis nominibus eorumitem declinant casus atque eodem modo dicunt huiusArtemidori^ et huius lonis et huius Ephesi,* sic incasibus aliis.23. Cum utrumque nonnunquam accidat, et ut involuntaria declinatione animadvertatur natura et innaturali voluntas, quae, cuiusmodi sint, aperienturinfra ;quod utraque declinatione alia fiunt similia,alia dissimilia, de eo Graeci Latinique libros feceruntmultos, partim cum alii putarent in loquendo ea verbasequi oportere, quae ab similibus similiter essentdeclinata, quas appellarunt avaAoytas,^ alii cum id* Ang., for dicitur. * Laetus, for lona. * Miie., forlonam.§ 22. ^ Apparently Varro's oion slip for Artemae.* RfioL, for Ephesis.§ 23. ^ For analogiias.§21. " This term includes both word-formation and wordinflection.* Practically equal to subjective and objective.* A common type of hypocoristic or nickname, cf. Demasfrom Democritus and similar names, Hippias from Hipparchus,etc.§ 22. " This is inflection. * Specifically, declension.§23. ' Cf. viii. 15-16, 51. " Cf. page 118 Funaioli.388

—ON THE LATIN LANGUAGE, VIIL 21-23place. There are two kinds of derivation," voluntaryand natural.* V'oluntary derivation is that which isthe product of the individual person's volition, directingitself apart from control by others. So, whenthree men have bought a slave apiece at Ephesus,sometimes one derives his slave's name from that ofthe seller Artemidorus and calls him Artemas ";another names his slave Ion, from Ionia the district,because he has bought him there ; the third calls hisslave Ephesius, because he has bought him at Ephesus.In this wav each derives the name from a differentsource, as he preferred.22. On the other hand I call that derivationnatural, which is based not on the vohtion of individualsacting singly, but on general agreement." So,when the names have been fixed, they derive thecase-forms of them in Uke fashion,* and in one and thesame way they all say in the genitive case Artemidori,lonis, Ephesi ; and so on in the other cases.23. Sometimes both are found together, and insuch a way that in the voluntary derivation the processesof nature are noted, and in the natural derivationthe effects of volition ; of what sort these are,will be recounted below.» Since in the two kinds ofderivation some things approach hkeness and othersbecome unUke, the Greeks and the Latins * have\vTitten many books on the subject : in some of themcertain writers express the idea that in speaking menought to follow those words and forms which arederived in similar fashion from like starting-pointswhich they called the products of Analogy « ; and« The regularizing principle which tends to eliminate irregularforms of less frequent occurrence, still called Analogy,by scientific linguists,389

VARROneglegendum putarent ac potius sequendam (dis)-similitudinem,* quae in consuetudine est, quamvocaruw^* d(i'>oi/xuAtai/ ,* cum, ut ego arbitror, utrumquesit nobis sequendum, quod (in) * declinationevoluntaria sit anomalia, in naturali magis analogia.24«. De quibus utriusque generis declinationibuslibros faciam bis ternos, prioris tris de earum declinationumdisciplina, posteriores de^ eius disciplinaepropaginibus. De prioribus primus erit hie, quaecontra similitudinem declinationum dicantur, secundus,quae contra dissimilitudinem, tertius de similitudinumforma ; de quibus quae experfiero^ singulistribus, turn de alteris totidem scribere ac dividere'incipiam.X. 25. Quod huiusce^ libri est dicere contra eosqui similitudinem sequuntur, quae est ut in aetatepuer ad senem, (puella)^ ad anum, in verbis ut estscribo scribam,* dicam prius contra universam analogiam,dein turn de singulis partibus. A naturasermo(nis>* incipiam.XI. 26. Omnis oratio cum debeat dirigi ad utili-est apertatatem, ad quam tum denique pervenit, si* Aug., with B, for similitudinem. ' For vocarum.* Aldus, for AtoMAeNAN. ^ Added by Aug.§ 24. ^ L. Sp.,for ex. * Mue. ; expedierint Aug. ; forexperiero. * L. Sp. deleted incipimus after dividere.§ 25. ^ For huiuscae. * Added by Aldus. ' L. Sp.deleted dico after scribam. * Aug., for sermo.''The irregularities summed up in this term are the productsof the regular working of phonetic law,' unrestrained by theoperation of Analogy ; the term Anomaly names it fromthe product rather than from the working process. * Itseems better henceforth to translate analogia by Regularityor the like, rather than to keep the word Analogy.390

ON THE LATIN LANGUAGE, VIII. 23-26others are of opinion that this should be disregardedand rather men should follow the dissimilar andirregular, which is found in ordinary habitual speech—which they called the product of Anomaly But.•*in my opinion we ought to follow both, because involuntary derivation there is Anomaly, and in thenatural derivation there is even more strikinglyRegularity.*24-. About these two kinds of derivation I shallwrite two sets of three books each : the first threeabout the principles of these derivations, and thelatter set about the products of these principles. Inthe former set the first book will contain the \-iewswhich may be offered against likeness in derivationand declension ; the second will contain the argumentsagainst unlikeness ; the third >vill be about theshape and manner of the likenesses. What I haveset in order on these topics, I shall write in the threeseparate books ; then on the second set of topics Ishall begin to write, with due division into the samenumber of books.X. 25. Inasmuch as it is the task of this book tospeak against those who follow likeness " —which islike the relation of boy to old man in the matter ofhuman Ufe, and like that of girl to old woman, and inverbs is the relation of scribo ' I wTite and scribam ' ' Ishall write '—I shall speak first against Regularity ingeneral, and then thereafter concerning its severalsubdivisions. I shall begin with the nature of humanspeech.XI. 26. All speaking ought to be aimed atpractical utiUty, and it attains this only if it is clear§ 25. " That is, regularity of paradigms resulting fromthe process of Analogy.391

VARROet brevis, quae petimus, quod obscurus' et longi(or>*orator est odio ; et cum efficiat aperta, ut intellegatur,brevis, wt' cito intellegatur, et aperta(m>* consuetudo,brevem temperantia loquentis, et utrumque fieripossit sine analogia, nihil^ ea opus est. Neque enim,utrum Herculi an Herculis clavam dici oporteat, sidoceat analogia, cum utrumque sit in consuetudine,non neglegendum,* quod aeque sunt et brevi(a> etaperta.XII. 27. Praeterea quoius^ utilitatis causa quaequeres sit inventa, si ex ea quis id sit consecutus,amplius ea(m>2 scrutari cum sit nimium otiosi, et cumutilitatis causa verba ideo sint imposita rebus utea

ON THE LATIN LANGUAGE, MIL 2&-29and brief : characteristics which we seek, becausean obscure and longish speaker is disliked. Andsince clear speaking causes the utterance to beunderstood, and brief speaking causes it to be understoodquickly, and since also habitual use makes theutterance clear and the speaker's self-restraint makesit brief, and both these can be present Avithout Regularity,there is no need of this Regularity. For ifRegularity should instruct us whether we ought tosay Herculi " or Hercidis for the genitive, as in thephrase the club of Hercules,' we must not ' fail todisregard its teaching, since both are in habitual use,and both forms are equally short and clear.XII. 27. Besides, if from a thing one has securedthat useful serxice for which it was invented, it is theact of a person with a great deal of idle time, toexamine it further ; and since the useful ser\ice forwhich names are set upon things is that the namesshould designate the things, then if we secure thisresult by habitual use alone. Regularity adds no gain.XIII. 28. There is the additional fact that inthose things which are taken into our daily Ufe foruse, it is our practice to seek utiUty and not to seekresemblance ; thus in the matter of clothing, althougha man's toga " is very unlike his tunic,* and a woman'sstola '^ is very unlike a palUum,'^ we make no objectionto the difference.XIV. 29. In the case of buildings, although we do§ 26. " This form occurs in Plautus, Persa 2, Rudens 822,and in other authors.§ 28.A shirt" Theorformalundergarment.outer garment' Theof adressRoman man.of a Roman*matron. * The long outer garment of the Greeks, properlya man's garb only, but worn also by prostitutes both inGreece and in Italy as a sign of their livelihood.

"VARRO(ad)" atrium Trepta-TvXoi'^ similitudinem et cubiculumad equile,* tamen propter utilitatem in his dissimilitudinespotius quam similitudines sequtmur^ : itaqueet hiberna triclinia et «estiva non item valvata acfenestrata facimus.XV. 30. Quare cum, wt^ in vestitu aedificiis, sic insupellectile cibo ceterisque omnibus quae usus (causa)ad vitam sunt assumpta dominetur inaequabilitas, insermone quoque, qui est usus causa constitutus, eanon repudianda.XVI. 31. Quod si quis duplicem putat esse summam,ad quas metas^ naturae sit perveniendumin usu,utilitatis et elegantiae, quod non solum vestiti essevol umus ut vitemus frigus, sed etiam ut videamur vestitiesse honeste, non domum habere ut simus in tecto ettuto solum, quo^ necessitas contruserit, sed etiam ubivoluptas retineri possit, non solum vasa ad victumhabilia,sed etiam figura bella atque ab artifice (ficta),'quod aliud homini, aliud humanitati satis est ;quodvissitienti homini poculum idoneum, humanitati(ni)si'* bellum parum ; sed cum discessum e(s)t* abutilitate ad voluptatem, tamen in eo ex dissimilitudineplus voluptatis quam ex similitudine saepe capitur.32. Quo nomine et gemina conclavia dissimiliter2 Added by L. Sp. » For nePHCThAON. * 3Iue.deleted quod after equile. * V, p, Mue., for sequamur.§ 30. ^ Stephanus, for et. " Added by L. Sp.§31. ^ For maetas. ^ Aug. {quoting a friend), forquod. ' Fay ; facta L. Sp. ; to fill a blank space in F ofabout 4 letters. * Aldus, for si. ^ Aug., with B, for et.§ 29. ° The garden in the rear part of the house, surroundedby colonnaded porticos. * The main hall in the front ofthe house, with a central opening to the sky under whichthere was a rectangular water-basin built in the floor.394.

ON THE LATIN LANGUAGE, VIIL 29-32not see the persistyle " bearing resemblance to theatrium * nor the sleeping-room bearing resemblanceto the horse-stable, still, on account of the utility inthem we seek for unlikenesses rather than likenesses ;so also we proWde >\inter dining-rooms and summerdining-rooms with a different equipment of doors and^^indows.X\\ 30. Therefore, since difference prevails notonly in clothing and in buildings, but also in furniture,in food, and in all the other things which have beentaken into our daily life for use, the principle ofdifference should not be rejected in human speecheither, which has been framed for the purpose of use.XVL 31. But if one should think that the sum ofthose natural goals to which we ought to attain inactual use consists of two items, that of utility andthat of refinement, because we wish to be clothed notonly to avoid cold but also to appear to be honourablyclothed ; and we wish to have a house not merely thatwe may be under a roof and in a safe place into whichnecessity has crowded us together, but also that wemay be where we may continue to experience thepleasures of life ; and we wish to have table-vesselsthat are not merely suitable to hold our food, but alsobeautiful in form and shaped by an artist—for onething is enough for the human animal, and quiteanother thing satisfies human refinement : any cupat all is satisfactory to a man parched with thirst, butany cup is inferior to the demands of refinement unlessit is artistically beautiful :—but as we have digressedfrom the matter of utility to that of pleasure, it is afact that in such a case greater pleasure is often gotfrom difference of appearance than from Ukeness.32. On this account, ic^entical rooms are often395

VARROpoliunt^ et lectos non omnis paris magnitudine acfigura faciunt. Quod (si)" esset* analogia petendasupellectili, omnis lectos haberemus domi ad unamformam et aut cum fulcro aut sine eo, nee cum adtricliniarem gradum, non item ad cubicularem ; nequepotius delectaremur supellectile distincta quae essetex ebore * rebus disparibus figuris quamgrabatis,* qui di'a Aoyov' ad similem formam plerumqueeadem materia fiunt. Quare aut negandumnobis disparia esse iucunda aut, quoniam necesse estconfiteri, dicendum verborum dissimilitudine(m>,quae sit in consuetudine,' non esse vitandam.XVII. 33. Quod si analogia sequenda est nobis,aut ea observanda est quae est in consuetudine autquae non est. Si ea quae est sequenda est, praeceptisnihiU opus est, quod, cum consuetudinemsequemur, ea nos sequetur ; si quae non est in consuetudine,quaeremus : ut quisque duo verba inquattuor formis finxen't^ similiter, quamvis haecnolemus, tamen erunt sequenda, ut Iuppit(r>i,'Marspitrem ? Quas si quis servet analogias, proinsano sit reprehendendus. Non ergo ea est sequenda.§ 32. ^ Koeler, for pollent. * Added by Laetus.^ Laetus, for essent. * Fay ; aliisque Laetus ; to fill ablank space of about 4 letters in F ; cf. ix. 47. * Forgrabattis. * Mue., for analogon ; cf. x. 2, ' Forconsuetudinem.§33.^ For nichil. * Vertranius, for ^nxerunt. ^ L.Sp., for luppiti.§ 33. " Namely, genitive, dative, accusative, ablative,^*from the nominative as starting-point. Such forms,retaining and inflecting the pater which forms the second396

ON THE LATIN LANGUAGE, VIIL 32-33ornamented in unlike manner, and couches are not allmade the same in size and shape. But if Regularitywere to be sought in furniture, we should have all thecouches in the house made in one fashion, and eitherwith posts or without them, and when we had a couchsuited for use beside the dining-table, we should notfail to have just the same for bedroom use ;nor shouldwe rather be delighted with furniture which wasdecorated with varying figures of ivory or othermaterials, any more than in camp-beds, which withregularity are almost always made of the samematerial and in the same shape. Therefore either wemust deny that differences give pleasure, or, since wemust admit that they do, we must say that the unlikenessin words which is found in habitual usage, isnot something to be avoided.XVn. 33. But if we must follow Regularity,either we must observe that Regularity which ispresent in ordinary usage, or we must observe alsothat which is not found there. If we must follow thatwhich is present, there is no need of rules, becausewhen we follow usage, Regularity attends us. But ifwe ought to follow the Regularity which is not presentin ordinary usage, then we shall ask. When any onehas made two words in four forms " according to thesame pattern, must we employ them just the same,even though we do not wish to—as for example adative luppitri and an accusative Marspitrem ? * Ifany one should persist in using such regular forms,''he ought to be rebuked as crazy. This kind ofRegularity, therefore, is not to be followed.part of luppiter and Marspiter, are quite abnormal, and arefound chiefly in the grammarians as examples of forms whichare not to be used.397

VARROQuod si oportet id es,^ ut a simili-XVIII. 34.bus similiter omnia declinentur verba, sequitur, ut abc?issimilibus* dissimilia debeant fingi, quod non fit :nam et ' similibus alia fiunt similia, alia dissimilia,et ab dissimilibus partim similia partim dissimilia.Ab similibus similia, ut a bono et malobonum malum ; ab similibus dissimilia, ut ab lupuslepus lupo lepori. Contra* ab dissimilibus dissimilia,ut Priamus Paris, Priamo Pari ; ab dissimilibussimilia, ut lupiter ovis, Iot;i ovi.35. Eo iam magis analogias (esse negandum,^quod non modo ab similibus) ^ dissimilia finguntur, sedetiam ab isdem* vocabulis dissimilia neque a dissimilibussimilia, sed etiam eadem. Ab isdem* vocabulisdissimilia fingi apparet, quod, cum duae sintAlfeae, ab una dieuntur Albani, ab altera Albenses ;cum trinae fuerint Athenae, ab una dicti Athenae,*ab altera Athenaiis, a tertia Athenaeopolitae.36. Sic ex diversis verbis multa facta in declinandoinveniuntur eadem, ut cum dico ab Saturni Lua Luam,§ 34. 1 id esse Canal ; ita esse Mue., for id est. * L.Sp.,for his similibus. ^ Added by L. Sp. ; a Aug.., with B.* Aug., for contraria.§ 35. ^ Added by L. Sp. * Added by Christ, who hasnon solum a, for which Groth, citing L. Sp., gives non modoab. * Mue. ; iisdem Laetus ; for hisdem. * Forhisdem. * Laetus, for Athenae.§ 34. " Or accusative masculine.§ 35. """Inhabitants of Alba Longa. Inhabitants ofAlba Fucens or Fucentia, among the Aequi on the borders ofthe Marsi." There were several cities named Athens,only that in Attica being important ; the forms of the namesare uncertain, especially that of the second, which mayhowever stand for 'Adr]vai.€LS like Aeolis v. 25 for kloXels.There were many ethnics in -evs, plural -els.398

'ON THE LATIN LANGUAGE, VIIL 34^36XVIIL 34. But if the proper thing is that all wordsthat start from similar forms should be inflectedsimilarly, it follows that from dissimilar starting formsdissimilar forrt^ should be made by inflection : andthis is not what is found. For from hke forms somelike forms are made, and other unhke forms, and fromunlike forms also come some like forms and someunlike forms. For instance, from likes cume likes, asfrom bonus good and malus bad come the neuter "' ' ' 'forms bonum and malum ; also from hkes come unUkes,as from lupus wolf and ' ' lepus hare come the unlike' 'datives lupo and lepori. On the other hand, fromunlikes there are unlikes, as from the nominativesPriamus and Paris come the datives Priamo and Pari ;also from unlikes there are likes, as nominativeslupiier Jupiter,' ' ovis sheep,' ' and datives lovi and017'.35. So much the more now must it be deniedthat Regularities exist, because not only are unlikesmade from likes, but also from identicalwords unlikes are made, and not merely likes, butidenticals are made from unlikes. From identicalnames unlikes, it is clear, are made, because whilethere are two towns named Alba, the people of theone are called Albajii " and those of the other arecalled Albenses ^ ; while there are three cities namedAthens, the people of the one are called Athenaei,those of the second are Athenaiis, those of the thirdAthenaeopolitae.36. Similarly, many words made in derivationfrom different words are found to be identical, aswhen I sav accusative Luam from Saturn's Lua,'^ and§ 36. " An old Italic goddess who expiated the blood shedher formulaic connexion with Saturn is uncertain.in battle ;

VARROet ab solvendo luo^ luam.^ Omnia' fere nostra(n)omina* vmlia* et muliebria multitudinis cum rectocasu fiunt dissimilia, ea(de>m * danc?(i>^ : dissimilia,ut mares Terentiei, feminae Terentia(e>,*eadem in dandi, vireis Terentieis et mulieribusTerentieis. Dissimile Plautus et Plautius, (Marcus etMarcius) ' ; et co(m)mune, ut huius Plauti et Marci.XIX. 37. Denique si est analogia, quod in multisverbis e

ON THE LATIN LANGUAGE, VIIL 36-39also luam as future of luo ' loosing.' ** Almost all ournames of men and women are unlike in the nominativecase of the plural, but are identical in the dative :unhke, as the men Terentii,'^ the women Terentiae, butidentical in the dative, men Terefitiis ' and womenTerentiis.^ Unlike are Plautus and Plautius, Marcusand Marciiis ; and vet there is a form common toboth, namely the genitive Plauli and Marci.^XIX. 37. Finally, if Regularity does exist for thereason that in many words there is a likeness of theword-forms, it follows that because there is unlikenessin a greater number of words the principle of Regularityought not to be followed in actual talking.XX. 38. In the last place, if Regularity doesexist in speech, it exists either in all its parts or insome one part ; but it does not exist in all, and it isnot enough that it exists in some one part, just as the. fact that an Ethiopian has white teeth is not enoughto justify us in saying that an Ethiopian is white :therefore Regularity does not exist.XXI. 39. Since those who declare that Regularitiesexist, promise that the inflected forms fromlike words ^^^\\\ be alike, and since they then say thata word is like another word only if it can be shownthat starting from the same gender and the sameinflectional form it passes in like fashion from case tocase, those who make these assertions show theirignorance both of that in which the likeness must befound and of how the presence or absence of the like-Republic, and was therefore Varro's regular orthography.In the translation the standardized Latin forms are used.**The contracted form ending in -I was practically the exclusiveform used as genitive of nouns ending in -lUS in thenominative, until the end of the Republic.VOL. II D 401

VARROsit necne. Quae cum ignorant, sequitur ut, cum

ON THE LATIN LANGUAGE, VIIL 39-42ness is wont to be recognized. Since they are ignorantof these matters, it follows that we ought not tofollow them, inasmuch as they are unable to pronounce\nth authority on the subject of Regularity.40. For I ask whether by a word ' 'they meanthe spoken word which consists of syllables, that wordwhich we hear, or that which the spoken word indicates,which we understand, or both. If the spokenword must be like another spoken word, it makes nodifference whether what it indicates is male or female,and whether it is a proper name or a common noun ;and yet the supporters of Regularity say that thesefactors do make a difference.il. But if that which is denoted by like wordsought to be like, then Dion and Theon,'* which theythemselves say are almost identical, are found to beunlike, if the one is a boy and the other an old man,or one is white and the other an Ethiopian * ; andUke\\'ise if they are unlike in some other respect. Butif the word must be like in both directions, there willnot quickly be found one that is not defective in onerespect or the other, nor will Perpenna and Alfenaprove to be alike, because the one name denotes aman and the other a woman. Therefore, since theyare unable to show wherein the likeness must exist,those who assert that Regularities exist are utterlyshameless.XXn. 42. The other matter that I have mentioned,how the likeness is to be recognized, theyclearly fail to appreciate in that they set up a preceptthat only when the passage is made from the nominaatedto the immediately following relative. ^ For the samecontrast, cf. Juvenal, 3. 23 : Loripedem rectus derideat,Aethiopem albus ; cf. also be. 4-2, below.403

VARROnominandi casibus in eos quos appellant vocandi, turndenique posse dici rectos esse similis aut dissimilis :esset enim ut si quis, Mewaechmos^ geminos cumvideat, dicat non posse iudicare similesne sint, nisiqui ex his sint nati considerarit num discrepent'inter se.43. Nihil, 1 inquam, quo magis minusve sit similequod conferas cum altero, ad iudicandum extrinsecusoportet sumi. Quare cum ignorent,* quemadmodumsimilitudo debeat sumi, de analogia dicere non possunt.Haec apertius dixissem, nisi brevius eo nunc mallem,quod infra sunt planius usurpanda. Quare quod aduniversam naturam verborum attinet, haec attigissemodo satis est.XXIII. 44. Quod ad partis singulas orationis,deinceps dicam. Quoius quoniam sunt divisionesplures, nunc ponam potissimum eam^ qua dividituroratio secundM7w^ naturam in quattuor partis : inearn' quae habet casus et quae habet (tempora etquae habet)* neutrum et in qua est utrumque. Hasvocant quidam^ appellandi, dicendi, adminiculandi,iungendi. Appellandi dicitur ut homo et Nestor,* Aldus, for Me hech mos. * Aug., for nunc discrepat.§ 43. ^ For nichil, * Rhol., for ignorarent.§ 44. ^ Lachmann, for iam. ^ Lachniann, for secundaut. ' Mue., with a, for iam. * Added from H and G.' Laetus, for quidem.§ 42. " Thus lupus and lepus, though alike in the nominative,are not aHke because their other case-forms are not alikein their endings ; c/. § 34. But bipus and campus a.re judgedto be likes, when the other case-forms are found to be alikein their endings. * In Plautus's play, the Menaechmi ; inVarro's comparison, the Menaechmus twins are the nominativeswith like endings, and the children of the Menaechmiare the derivative case-forms.§ 43. " X. §§ 3 ff., 10 ff.404

ON THE lATIN LANGUAGE, VIII. 42-44tives to the vocative forms can it be said whether thenominatives are like or unlike " ; for this would be asif a man, on seeing the Menaechmus t\\ins,'' shouldsay that he could not decide whether or not theywere alike, unless he should scrutinize their children,to see if they showed any differences from one another.43. Nothing, I say, whereby that which you arecomparing with the other may be made more like itor less hke it, ought to be brought in from outside, forthe purpose of aiding the decision. Therefore, sincethey do not know in what way the likeness ought tobe dra>\Ti, they are incompetent to speak about Regularity.I should have said this more plainly, if I werenot wishing now to speak more briefly because lateron " these matters are to be treated at greater length.Accordingly it is sufficient now to have touched uponthem as far as is connected with the general natureof words.XXIII. 44. I shall next speak of what concernsthe individual parts of speech. Since there are severalmethods of di\'ision thereof, I shall now take by preferencethat by which speech is according to its naturedivided into four parts : that which has case-forms, thatwhich has time-forms, that which has neither, that inwhich both case and time are indicated." Some grammarianscall these the parts respectively of naming,saying, supporting, joining * : the part of naming issaid to be such words as homo man ' 'and Xestor,§ 44. " Participles. * The part of'supporting ' includestfie adverbs, as indeclinable modifiers ; tfiat of ' joining' includes participles, wfiich are so called because theyjoin in the same word the indication of case and that of time,or else because they unite in themselves the syntactical functionsof adjective and verb {c/. Greek fierox^ ' sharing,' asname of the participle).405

VARROdicendi ut scribo et lego, iungendi ut* (scribens etlegens),' adminiculandi ut docte et commode.45. Appellandi partes sunt quattuor, e quis dicta aquibusdam provocabula quae sunt ut quis, quae^ ;(vocabula)^ ut scutum,' gladium ; nomina ut Romiulus,Remus ;pronomina ut hie, haec. Duo mediadicuntur nominatus ;prima et extrema articuli.Primum genus est infinitum, secundum ut infinitum,tertium ut finitum,* quartum finitum.46. Haec singulatim triplicia esse debent quodcretis^ partibus singulasperspice, quo facilius nusquam esse analogias quassequi debeamus videas. Nempe esse oportebatvocis formas ternas, ut in hoc humanus humana humanum,sed habent quaedam binas, ut cervus cerva,* Zippmann deleted que after ut. ' Added by Zippmann ;cf. X. 17.§ 45. ^ Aug., for que. * Added by Laetus, cf. viii. 52,and X. 19. * Bentinus deleted ut after scutum. * Aug.,for effinitum.§ 46. ^ quod ad L. Sp. : quoad Aug. ; for quod.^ Added by Mue. ^ L. Sp., for iouis ; cf. viii. 49, but alsoviii. 74.§ 47. ^ Aug., for decretis." The third and the fourth items are here reversed in orderfrom the previous listing.§ 45. " The neuter form of this word is quoted by NoniusMarcellus, 208. 12 M., from Lucilius (1187 Marx), though it406

'ON THE LATIN LANGUAGE, VIIL 44r^7that of saying such as scribo ' I \vrite and ' lego ' Iread,' that of joining " such as scribens writing'and legens reading,' that of supporting such 'as docte'learnedly and commode ' ' suitably.'45. The kinds of naming are four, of which thewords which are like (masc.) quis, (fern.) quae which ''have by certain grammarians been called Provocables;those like scutum ' shield and gladium '

VARROquoedam singulas, ut aper, et sic multa. Non ergoest in huiuscemodi generibus analogia.XXV. 48. Et in multitudine ut ununi significatpater, plures patres, sic omnia debuerMnt^ esse bina.Sed et singularia solum sunt multa, ut cicer, siser :nemo enim dicit cicera, sisera ; et multitudinis sunt,ut salinae (balneae)^ : non enim ab his singularispecie dicitur salina et balnea. Neque ab eo quoddicunt balneum habet multitudinis consuetude : namquod est ut praedium balneum, debuerunt esse plura,ut praedia balnea, quod non est : non est ergo in hisquoque analogia.XXVI. 49. Alia casus^ habent et rectos et obliquos,alia rectos solum, alia modo obUquos habent :utrosque ut luno, lunonis, rectos modo ut lupiter,Maspiter, obliquos solum ut lovis, lovem : non ergoin his est analogia.XXVII. 50. Nunc videamus in ilia quadripertita.Primum si esset analogia^ in infiniteis^ articulis, ut estquis' ^Moius,* sic diceretur quae quaiMS* ; et ut estquis quoi,^ sic diceretur qua quae : nam est proportionesimile : ut deae bonae quae, sic' dea bona qua*§ 48. ^ Aug., with B,for debuerint. * Added by L. Sp.§ 49. ^ Alia casus is repeated in F.§ 50. 1 L. Sp. deleted ut after analogia. " For infeineiteis.' L. Sp. deleted quem after quis. * Aug.,H, for cuius. * quaius L. Sp. ; quam quaius Aug. ; forquamuis. * quis quoi Aug., for a quiuis cui.''L. Sp.,for sit. * Sciop., for quae.§ 48. " Cf. ix. 68.§ 49. " The oblique cases lack the affixed -piter =pater ofthe nom.-vDc. form, and have the appearance therefore ofnot coming from the same word.§ 50. " This form is nowhere found in use. * Gen.408

ON THE LATIN LANGUAGE, VIII. 47-50cervus ' stag,' cerva ' hind,' and certain others have butone, like aper boar ' ' ; and so on with many others.Therefore Regularity simply does not exist in classesof this sort.XX\'. 48. In number also, as pater ' father ' denotesone and patres fathers denotes more than ' ' one,all ought in this fashion to have two forms. But manyare singulars only, Hke cicer chickpea and ' ' siser'skirret '—for nobody says cicera and ^sera in theplural ; and there are words that are plurals only,such as salinae saltworks and ' ' balneae public'baths ' : for from these there are no singulars salinaand balnea in use. Nor from the singular balneum doesgeneral usage make a plural : for because balneum'bath ' " is Vike praedium'farm estate,' there ought tobe plurals, balnea as well as praedia, but this is not thecase. Therefore in these also there is no Regularity.XX\T. 49. Some have both nominatives andoblique cases, others have nominatives only, othersoblique cases only : both, as in nom. luno, gen. lunonis; nominatives only, as in lupiter, yiaspiter °;oblique cases only, as in gen. lovis, dat. /017. Thereforein these Regularity does not exist.XXVII. 50. Now let us look into those of the fourfolddi\-ision. First, if there were Regularity in theindefinite articles, the proper forms would be femininequae, gen. quaius," Uke mascuUne quis, gen. quoius ** ;and as quoi is dative to masculine quis, so quae " wouldbe used as dative to feminine qua. For it is similar byproportion : dea bona qua a good goddess who ' ' ishke deae bonae quae ' to a good goddess to whom.'quoius and dat. quoi were the regular forms down to the end ofthe Republic. ' The writing quai for the dat, sing. fem. isfound on one inscription, Corp. Insc. Lot. ii. 89.409

VARROest ; et ut est quem quis, sic quos ques. Quare quodnunc dicitur qui homines, dici oportuit ques.XX\'III. 51. Proeterea ut est ab is^ (ei),'' sic abea eae diceretur, quod nunc dicitur ei, ' pronuntiareturut in i

ON THE LATIN LANGUAGE, VIIL 50-52Similarly, the nom. masc. ques stands in the samerelation to ace. quos, as the sing. masc. quis to the ace.quern ; therefore for the qui homines which men' 'which is now used in the nominative, we ought to sayques.'^XXVIIL 51. Besides, as from masc. is'this 'there is dat. ei, so from fem. ea there would be spokena dat. eae,'^ which is now actually spoken as ei ; andlike the dat. pi. in ieis * viris ' to these men,' therewould be pronounced a dat. pi. fem. eais mulierihus'^'to these women.' And as in the nominatives thereare masc. is, fem. ea, in the oblique forms there wouldbe masc. eius, fem. eaius ** ; but now eius is said for thegenitive not only in the mascuhne and the femininealike, but even in the neuter articles, as eius viri of'this man,' eius mulieris of ' this woman,' eius pabuli ' ofthis fodder,' although masc. is, fem. ea, neut. id aredistinguished in the nominative. I have touchedupon this classification more sparingly, because I amof opinion * that the copyists will not take proper carein transferring these quite confusing matters.XXIX. 52. From the appellations which comenearest to the indefinite nature of the articles and arecalled common nouns, such as homo man ' 'and equus'horse,' there are four kinds of derivation " : one ofname-giving, as equile horse-stable ' ' from equus'horse ' ; the second that of the cases, as accusativeequum from equus ;the third that of augmentation, as•*Nowhere found. « The condition of the manuscriptsshows that Varro was right.§ 52. " We should call these four respectively derivationby suffixes, declension, comparison of adjectives, derivationby a suffix denoting diminution ; the fourth is a division ofthe first, and so also, in the broad sense, is the third, thoughit has a more specialized function.411

VARROtertium augendi, ut ab albo albius, quartum minuendi,ut * cista cistula.53. Primum genus, ut dixi, id est, cum (ahyaliqua parte orationis declinata sunt recto casu vocabula,2ut a balneis balneator. Hoc fere tripliceshabet radices, quod et a vocabulo oritur, ut a venatorevenabulum, et a nomine, ut a Tibure* Tiburs, e^* averbo, wt* a currendo cursor. In nullo horum analogiamservan'^ videbis.XXX, 54. Primum cum dicatur ut ab ove et sueovile et suile, sic a bove bovile non dicitur ; et cumsimile sit avis et ovis, neque dicitur ut ab ave aviarium

ON THE LATIN LANGUAGE, VIIL 52-55albiiis whiter from album white ' ' ' ' ; the fourth thatof diminution, as cistula ' little box from ' cista ' box.'53. The first class, as I have said, is that in whichwords in the nominative " are derived from some partof speech, as balneator bath-keeper ' ' from balneae'public baths.' This class has in general threesources, because it develops from a common noun, as'venabulum * hunting spear from ' venator ' hunter,'and from a proper name, as Tibtirs man ' of Tibur 'from Tibur, and from a verb, as cursor runner from' 'currere ' to run.' In none of these will you see Regularitypreserved.XXX. 54. First, although from ovis sheep and' 'sus swine there ' ' are said ovile sheepfold and ' ' suite'hog-sty,' there is no bovile " from bos ox ' ' ; andalthough avis ' bird and ' ovis sheep ' ' are alike, we donot sav oviarium from ovis as we say aviarium a\iary ''from avis, nor do we say avile from avis as we say ovile'sheepfold ' from avis ; and although there ought tobe a sediculum chair ' ' from sessio ^'sitting ' likecubiculum sleeping-room ' ' from cubatio " ' reclining,'there is not.55. Since a shop where wine is sold is called vinariafrom vinum ' wine,' and cretaria from creta ' chalk,' "unguentaria from unguentum perfume,' then 'if wordswent in regular fashion a shop where caro ' meat ' is soldwould be called carnaria, one where pelles hides are' 'sold would be called /)e//«ana, one where calcei 'shoes 'are sold would be called calcearia, instead of laniena'butcher's shop,' pellesuina ' leather-shop,' sutrina336. 6 M., quotes sediculum as occurring, but without mentioningwhere. ' Proj)erly both from cubare ' to recline.'§ 55. " That is, ' Cretan (earth),' used at Rome for cleaningpurposes.41.S

VARROsicut est ab uno uni, ab tribus trini, a quattuor quadrini,sic a duobus duini, non bini diceretur ; nee nonut quadrigae trigae, sic potius duigae quam bigae.Permulta sunt huiusce generis, quae quoniam admonitusperspicere potest, omitto.XXXI. 56. Vocabula quae ab nominibus oriuntur,si ab similibus nominibus similia esse debent, dicemus,quoniam gemina sunt Parma' Roma, (ut) Parmenses

ON THE LATIN LANGUAGE, VIIL 55-57'cobbler's shop.' And just as from unus ' one 'comes the plural uni ^ ' one set of,' and from ires'three ' comes trini ' three each,' from quatiuor ' four 'quadrini four each,' so from duo two ' ' ' there shouldbe a duin't and not a hini

VARROcum dicatur lassus sum metendo ferendo, ex his vocabulanon reddunt proportionem, quo(niam)2 non fitut messor fertor. Multa sunt item in hac specie inquibus potius consuetudinem sequimur quam rationemverborum.58. Praeterea cum sint ab eadem origine verborumvocabula dissimilia superiorum, quod simulhabent casus et tempora, quo vocantur participia, etmulta sint contraria ut amo amor, lego legor,^ ab amoet eiusmodi omnibus verbis oriuntur praesens etfuturum tit^ amans et amaturus,* ab eis verbis tertiumquod debet fingi praeteriti, in lingua Latina reperirinon potest : non ergo est analogia. Sic ab araor*legor et eiusmodi verbis* vocabulum eius generispraeteriti te

ox THE LATIN LANGUAGE, VIIL 57-59though we say " I am tired with metendo reaping ''and Jerendo carrying,' " the words from these do not'represent a Uke relation, since there is no Jertor *'carrier made Uke ' messor ' reaper.' There are Ukevrisemany others of this class in which we follow usagerather than conformity to the verbs.58. Besides these there are other words whichalso originate from verbs but are unUke those of whichwe have already spoken, because they have both casesand tenses, whence they are called participles. Andas many verbs have opposite forms," such as amo 'Ilove,' amor ' I am loved,' lego ' I read,' legor ' I amread,' from amo and all verbs of this kind * theredevelop present and future participles, such as amans'loving and amatiirus about to ' ' love,' but from theseverbs the third form which ought to be made, namelythe past participle, cannot be found in the Latinlanguage : therefore there is no Regularity. So alsofrom amor ' I am loved,' legor ' I am read,' and verbsof this kind the word of this class

VARROloquor et venor,

ON THE LATIN LANGUAGE, MIL 59-61loquor and venor, not loquo and veno, whence came theforms given above. '^ The Regularities are the lesspreserved, because some of the verbs which have notboth voices, make three participles each, hke thosewhich I have named,** and other make only two each,*such as those which I shall now name : currens'running and ambulans ' ' walking,' cursurus about to'run and ' avibulaturus about to walk ' ' ; for the thirdforms, those of the past, do not exist,^ as in cursus sum'I am run,' ambulatus sum ' I am walked.'60. But Regularity is not preserved even in thosewhich indicate that something is done with greaterfrequency ; for though there is a cantitans ' repeatedlysinging ' from cantare ' to sing,' there is no amitans'repeatedly lo\ing ' from amare ' to love,' and similarlywith many others.The situation is the same inthe forms of the plural as in those of the singular :though the plural cantitantes is used, sedttantes'^'sitting ' is not.XXXIII. 61. Since there is a class of words whichthey call compositional, saying that they ought not tobe grouped in the same category with the simple wordsof which I have so far spoken, I shall deal separatelywith these compounds. Since from tibiae pipes and' 'canere to play the ' ' iibicines pipers are named, they' 'ask, If we ought to follow the Regularities, why thenfrom cithara ' lute and ' psalierium ' psaltery and'pandura Pan's ' strings should we not say ' citharicen "'lute-player ' and the rest in the same way } Iffrom aedes ' temple ' and tueri ' to guard ' the aedi-§ 60, " The singular seditans also is not used, which isimplied by Varro, but not stated.§61. " Citharista, fern, citharistria, are used, both takenfrom Greek.4.19

VARROdicatur, cur non ab atrio et tuendo>i potius atritumussit quam atriensis ; si ab avibus capiendis aucepsdicatur, debuisse aiunt a piscibus capiendis ut aucupemsic pisci

ON THE LATIN LANGUAGE, MIL 61-63tumus ' sacristan ' is named, why from atrium ' mainhall and ' tveri ' to guard ' is it not atritumus butler ''rather than atriensis ? And if from avis capere ' tocatch birds the auceps fowler ' ' ' is named, they say,from piscis capere to catch ' fish there ought ' to be apisciceps ^ fisherman named Hke the ' ' auceps.62. They remark also that establishments whereaes copper ' ' lavatur ' is refined are called aerariae''smelters ' and not aerelavinae ' copper-washery ' ;and places where argentum ' silver ' foditur ' is mined 'are called argentifodinae ' silver-mines,' but thatplaces where ferrum ' iron ' is mined are not calledferrifodinae " ; that those who caedunt ' cut ' lapides'stones are called ' lapicidae stone-cutters,' but that'those who cut ligna ' firewood ' are not called lisnicidae* : that there is no term argentifex * ' silversmith' like aurifex goldsmith ' ' ; that a person who isnot doctiis learned ' ' is called indoctus, but one who isnot salsus witty ' ' is called insulsus.^ Thus the wordswhich come from this source also, it is easy to see, donot observe Regularity.XXXR'. 63. It remains to consider the problemof the cases, on which the Aristarcheans " especiallyexert their energies. XXXV. First, if in these therewere Regularity, they * say that all names and articlesought to have the same number of cases ; but that asthings are some have one only,*' like all indi\idualletters, others have three,"* like praedium praediiAmong his pupils were important scholars of the next generation." Those who do not believe in the principle of Regularity.' These are the indeclinable nouns.''Varrocounts only different case-forms : where he finds three, thenom., ace., and voc. are identical, and the dat. and abl. areidentical ; etc421

VARROmel mellis melli melle, alios quinque, ut quintusquinti quinto quintum quinte, alios sex, ut unus uniusuni unum une uno : non esse ergo in casibus analogias.XXXVI. 6t. Secundo quod Crates, ^ cur quaesingulos habent casus, ut litterae Graecae, non dicanturalpha alphati alphatos, si idem mihi respondebiturquod Crateti,^ non esse' vocabula nostra, sed penitusbarbara, quaerani, cur idem nostra nomina et Persarumet ceterorum quos vocant barbaros cum casibusdicat.*65. Quare si essent in analogia, aut ut Poenicumet .-fegjyptiorum vocabula singulis casibus dicerent,aut pluribus ut Gallorum ac ceterorum ; nam dicunta\auda alaudas^ et sic alia. Sin^ quod scrib«nt' dicent,quod Poenicum sit,* singulis casibus ideo eas litterasGraecas nominari : sic Graeci nostra seniscasibus non quinis^ dicere debebant ;quod cum nonfaciunt, non est analogia.XXXV'II. 66. Quae si esset,^ negant ullum casumduobus modis debuisse dici ;quod fit contra. Namsine reprehensione vulgo alii dicunt in singulari hac§ 64. ^ Laetus, for grates. * Laetus, for grateti.' Aug., with B, for essent. * Laetus, for dicat.§ Q5. ^ Scat iff er, for alacco alaucus. " Popma, foralias in. * Popma, M, for scribent. * RhoL, for sit.* Jyaetus transposed quinis non.%66. ^ Laetus, for essent.§ 64. " Crates of Mallos, head of the Pergamene school ofscholarship, was a contemporary and opponent of Aristarchus,and championed the principle of Anomaly.* Names of letters were indeclinable both in Greek and inLatin.§ 65. " Not the Carthaginians, but the Phoenicians.^ Varro knew that neither language had a case system.422

ON THE LATIN LANGUAGE, VIIL 63-66praedio ' farm,' others four, like mel mellis melli mellehoney,' others five, Uke quintus quinti quinto quintum'quinte ' fifth,' others six, like unus unius uni unumune uno one ' ' ; therefore in cases there are noRegularities.XXX^T. 64'. Second, in reference to what Crates

VARROovi et avi, alii hac ove et ave ; in multitudinis haepuppis restis et hae puppes restes ; item quod inpatrico* casu hoc genus dispariliter dicuntur civitatumparentum et civitatium parentium, in accusandi hosmontes fontes et hos montis fontis.XXXVIII. 67. Item cum, si sit analogia, debeantab similibus verbis similiter declinatis similia fieri etid non fieri ostendi possit, despiciendam earn esserationem. Atqui ostenditur : nam qui potest similiusesse quam gens, mens,^ dens ? Cum horum casuspatricus et accusativus in multitudine sint dispariles^:nam a primo fit gentium et gentis, utrubique ut sit(I),' ab secundo mentium et mentes,* ut in priore solosit I, ab tertio dentum et dentes, ut in neutro sit.68. Sic item quoniam simile est recto casu suruslupus lepus, rogant, quor non dicatur proportionedsure lupo lepo. Sin respondeatur similia non esse,quod ea vocemus dissimiliter sure lupe lepus (sic enimrespondere voluit AristarcAus Crateti : nam cumscripsisset similia esse Philomedes Heraclides MeUcertes,dixit non esse similia : in vocando enim cum

ON THE LATIN LANGUAGE, VIIL 66-68and avi ' bird,' others say ove and ate ; in the plural,the nominative is puppis ' ship's sterns ' and restis'ropes,' also puppes and restes ; likewise there is thefact that in the genitive plural of words of this classthere are used the variant forms civitatum 'of states,'parentum ' of parents,' and civitatium, parentium, andin the accusative plural monies mountains,' ' fontes'springs,' and montis,fontis.'^XXX\TIL 67. Likewise they say that, if there isRegularity, hke forms ought to be made from hkewords declined alike, and that this can be shown notto take place ; that therefore this theory is to berejected. And yet this failure can be shown ; forhow can anything be more alike than gens ' clan,'mens ' mind,' dens ' tooth ' ? Despite which theirgenitives and accusatives in the plural are unlike ;for from the first word are made gentium and gentis,with I in both, from the second come mentium andmentes, with I in the former only, from the thirddentum and denies,'^ with I in neither form.68. So hkewise since sums ' stake,' lupus ' wolf,'lepus hare ' ' are alike in the nominative, they askwhy there is not said in like fashion " suro, lupo, lepo.But if the answer is given that they are not alike,because in the vocative we use the unhke forms sure,lupe, lepus (this to be sure is what Aristarchus wishedto say in reply to Crates ; for when Crates had\\-ritten that Pkilomedes, Heraclides, Melicertes werealike, he said that they were not aUke, because in thevocative * Pkilomedes is said %\ith a short E, Heraclidecorrect in all three ; -es was transferred to them by the influenceof consonant-stems.§ 68. " Datives or ablatives ; lepus has dat. lepori, abl.lepor^. * Greek names, with their Greek vocative forms.425

VARROcum * brevi Melicerta^), in hoc dicunt AristarcAumnon intellexisse quod quceretur se nonsolvere.*69. Sic enim, ut quicque in obliquis casibus discrepavit,dicere potuit propter earn rem rectos casusnon^ esse similis ;quom quoeratur duo inter sesimilia sint necne, non debere extrinsecus adsum

"'.ON THE LATIN LANGUAGE, VI J L 68-71with a long E, Melicerta with a short A), in this, theysay, Aristarchus did not realize that he was not givinga solution of the question that was asked.69- For in this way, whenever there was anydifference in the oblique cases, he " could say that forthis reason the nominatives were not alike ; but sincethe question is whether the two nominatives are likeeach other, or not, there should be notliing broughtin from outside, they say,'' as to why they are aUke ordifferent.70. Likewise if there were Regularity, then justas thev say in like fashion the nominatives aves *birds,'oves ' sheep,' sues ' s>\ine,' they would say in the genitivesuium " just as they do avium and ovitim. If thereis Regularity, he ^ says, why do the people say dei ''gods ' in Dei Penates ' Household Gods ' and DeiConsentes ' United Gods,' ** although in the nominativesingular deus ' god ' is just like reus ' defendant,'ferreus ' of iron,' and so the plurals should be ree?,*ferreei, deei71. Likewise they ask, if Regularity exists, whydo people all say the Temple Deum " Consentium ' ofthe United Gods ' and not Deorum Consentium ? Likewise,why do they say a thousand denarium "'ofdenarii ' and not a thousand denariorum ? For thisword denarius is in form exactly like Vaiinius andManilius, and therefore denariorum ought to be usedas genitive, like Vatiniorum and Maniliorum. Theyin council.' ' All with EI for long I ; unless we are to readdi above, and rei, ferrei, dei here. At any rate, the normalplural of deus was monosyllabic, and was not made like theplurals of the other words.§ 71. " The genitive in -UM was an older form than thatin -ORUM, and was retained in many words pertaining toreligion and law, and in weights and measures.427

VARROpuplicum mille assarium esse, sed niille assariorum :ab uno enim assario multi assarii, ab eo assariorum.72. Item secundum illorum rationem debemussecundis syllabis longis Hectorem Nestorem : est enimut quaestor praetor Nestor qu«estorem praetoremNestorem, quaestoris praetoris Nestoris ; et nondebuit dici quibus das, his^ das : est enim ut hi' quihis quis, a

ON THE LATIN LANGUAGE, VIIL 71-74say also that a cavalninan's horse * is worth not athousand assarium '^ ' of assarii,' but a thousand a*-sariorum ; for from the singular assarius comes theplural assarii, and from that should come the genitiveassariorvm.72. Likewise, according to their theory, it is withlong second syllables that we should pronounce Hectoremand Nestorem "; for the accusatives quaestorempraetarem Xestorem, and the genitives quaestoris praetorisNestoris correspond to the nominatives quaestorpraetor Nestor.And we ought not to say quibus das'to whom you give,' his das ' to these you give ' ; forit is dative his and quis, like nominative hi and qui, orelse it is hibus like quibus.^73. Although the dative patri familias " ' to thefather of the household ' is used, still, if they wishedto follow Regularities, they ought never to have saidnominative paterfamilias, because the word is genitivefamiliae, like Atiniae and Catiniae, and thereforenominative familia,^ like Atinia and Catinia. Likewise,they ought not to say patres familias '^ ' fathersof a household,' but as Sisenna ** writes, they shouldsa.y patres familiarum fathers of households.''74. Nor ought usage to fluctuate, in that somesaid herds bourn of cattle,' others 'boverum,'^ and otherssaid statues loum of 'Jupiters,' others loveru?»," sinceis a nominative ; or else the text is too corrupt for restoration.' In favour of this form, see Charisius, i. 107 Keil. *Page128 Funaioli ; L. Cornelius Sisenna, 119-67 b.c, orator andstatesman, author of a history dealing chiefly with the timesof Sulla.§ 74. " It is doubtful if these forms had any real existence ;if so, borerum was formed after lugerum, and loverum after]'enerum, to avoid the inconvenient forms bourn and loum,which \'arro would have pronounced bovom and lovom.429

VARROlovem bovem struem lovi bovi strui ; nee eum haecconvenirent in obliquis easibus, dubitare debuerunt inreetis, in quibus^ nune in consuetudine aliter dieere,pro lous' lupiter, pro bus* bos, pro strus^ struts.*XXXIX. 75. Deinceps dicam de altero generavocabulorum, in quo contentiones fiunt, ut albuTw^albius albissumum, in quo i^(em>* analogias non servariapparet : nam cum sit simile salsum caldum etdicatur ab his salsius caldius, salsissimum caldissimum,debuit dici, quoniam simile est bonum malum,ab his bonius et malius, bonissimum et malissimum.Nonne dicitur bonum melius optimum, (malum peiuspessimum) ?*76. In aliis verbis nihil^ sint* abeadem voce declinata, et ea ita ut alias desint secundumet tertium, ut in hoc mane manius manissime,alias ut duo prima absint, ut ab optimum optius* L. Sp., for propinquibus. ' L. Sp., for iouis. * L.Sp., for bus. * L. S]}., for struus. * L. Sp., for struis.§ 75. 1 RhoL, for albus. * Mue., for id. » Addedby Aug., with B.§ 76. ^ For nichil. * Aldus, for est. * Added byA. Sp. * Mue., for sint,* The reasoning demands hjT>othetical nominatives bearingthe same relation to the oblique forms, so that either loiis,bous, slrus or lovis, bovis, struis must be posited as theunused regular nominatives.' '§ 75. " Adjectives and adverbs. * Varro normally, butnot always, quotes adjectives in the neuter form. " Thesyncopated form of caltdum, common in popular speech atthe time of Varro ; cf. Quintilian, Inst. Orat. i. 6. 19. For430

ON THE LATIN LANGUAGE, VIIL 74^76like the genitive lovis bovis struts were the accusativelovem bovem struem and the dative lovi bovi strut ; andsince these agreed in the obUque cases, they oughtnot to have varied in the nominative forms, in whichit is now customary to use different formations, lupiterfor lous,^ bos for bous,^ strues heap of offering-cakes ''for strus.^XXXIX. 75. Next I shall speak of a second " classof words, in which degrees of comparison are made,like album *' 'white,' albius whiter,' albissumum'whitest,' in which likewise it is clear that the Regularitiesare not preserved. For whereas salsum 'salty'and caldum * ' hot ' are alike, and from these are madethe comparatives salsius and caldius, and the superlativessalsissimum and caldissimum, there should bemade from boniim good ' 'and malum ' bad,' sincethese are ahke, the comparatives bonius and malius,the superlatives bonissimum and malissimum. But arenot the actual forms in use botium melius optimum, andmalum peius pessimum ?76. In some words no form is lacking ; for example,dulcis sweet,' ' dulcior, dulcissimus. In othersthe first or positive degree is lacking, as peium frompeius ' -worse,' pessimum worst ' ' ; in others the secondis lacking, as caesior from caesius blue-eyed,' superlativecaesissimus ° ; in still others two are lacking'which are derived from the same word, and these insuch a way that in some instances the second and thethird are lacking, as manius and manissime in connexionwith the adverb mane early ' in the morning ' ; thatin others the first two are wanting, as opium and optiusmetrical facility the poets normally used positive calidus,comparative (nom. sing.) caldior.§ 76. " This form is not otherwise attested.431

VARROoptum, alias ut primum et tertium desit, ut a meliusmelum^ melissumum.77. Praeterea si dicerentur similiter, cum similiaassent acer^ tener et macerrimus^ tenerrimus, nondiscreparet in his macrior tenerior,^ neque alia trisyllaba(alia quadrisyllaba)* fierent ; et si in hisdominaretur similitudo,diceremus ut candidissimuscandidissima, pauperrumus pauperrima, sic candidusCandida, pauper paupera ;doctissimus doctissima, sicfrugalissima, frug

ON THE LATIN LANGUAGE, VIIL 7&-78from optimum * ' best ' ; and that in others the firstand the third are lacking, as melum and melissumumfrom melius 'better.'77. Besides, if they were spoken in like manner,then since macer ' lean ' and tener ' tender ' " arealike, and their superlatives macerrimus and tenerrimus,there would be no difference in their comparativesmacrior and tenerior, nor would some such wordshave three syllables and others four. And if Ukenessruled in such words, then just as we say candidissimus'most shining and fem. ' candidissima, pauperrumus'poorest ' and fem. pauperrima, so we should saycandidus ' shining ' and fem. Candida, pauper ' poor 'and fem. paupera.^ And as we say doctus learned ''and fem. docta, doctissimus most learned ' ' and fem.doctissima, so we should say frugaUssumus most'thrifty and fem. ' frugalissima, frugalus ' thrifty and'fem. Jrugala.'^78. And if words were in regular relation to eachother, as with one word ° we call a man and a womansapiens'wise ' and diligens'diligent,' sapientior'wiser and ' diligentior more ' diligent,' so we shouldspeak in the same way when we had come to thesuperlative—a thing which we now do quite otherwise* : for we call a man sapientissimus and diligentissimus,a woman sapientissima and diligentissima. Asfor examples of words of this class, there are manystill remaining ; but those which have been mentionedserves for all genders in the nominative, as it belongs to thethird declension. ' The form in actual use for the positiveis the dative noun frug I : older Latin used frugal is.§ 78. " These third declension adjectives use the sameforms for masc. and fem. * All superlatives are of thesecond and first declensions, distinguishing the fem. fromthe masc. forms.VOL. II F 433

VARROquae dicta, ad iudicandum satis sunt, quor' analogiasin collatione verborum sequi non debeamus.XL. 79- Magnitudinis vocabula cum possint esseterna, ut cista cistula cistella, in (aliis) media^ nonsunt, ut in his macer macricolus macellus, niger nigricolusnigellus.Item minima in quibusdam non sunt,ut avis avicula avcella,^ caput capitulum capitellum.In hoc genere vocabulorum quoniam multa desunt,dicendum est non esse in eo potius sequendam quamconsuetudinem rationem. Quod ad vocabulorumgenera quattuor pertinet, ut in hoc potius consuetudinemquam analogias dominari facile animadvertipossit, dictum est.XLI. 80. Sequitur de nominibus, quae difFerunt avocabulis ideo quod sunt finita ac significant res proprias,ut Paris Helena, cum vocabula sint infinita acres comunis designent,^ ut vir mulier ;e quibussunt alia nomina ab nominibus, ut Ihum ab Ilo et Ibaab Ik'o,^ alia a vocabulo, ut ab albo Albius, ab atroAtrius. In neutris servata est analogia : nam et cumsit a Romulo Roma, proportione non est quod debuitesse (Romula, non Roma).*' L. Sp., for quod F {corrected from quorum).§ 79. ^Auff. {quoting a friend), for in mediis. ^ Foraucella.§80. ^ Auff., with B, for designentur. ^ Aug. , with B,for ilium ab illo et ilia ab illo. ' Added by Stephanus, cf.ix. 50 ; but the lacuna is more serious, for it should show alsoirregularity in the derivation of proper names from commonnouns.§ 79. " That is diminution in size. * The non-existentforms.434

ON THE LATIN LANGUAGE, VIIL 78-80are enough for the formation of a judgement why inthe comparison of words we do not have to followRegularities.XL. 79. Whereas there can be a set of threewords to indicate size," Uke cista ' casket,' cistula,cistella, in some the middle terms do not exist, as inthese : macer ' lean,' macricolus,^ macellus, and niger'black,' nigricolus,^ nigellus. Likewise in certainwords the terms for least size do not exist, such as avis'bird,' avicula, avicella,^ and caput head,' ' capitulum,capitellum.^ Since in this class of words there aremany forms lacking, we must say that in it theorymust not be followed rather than usage.As to the four classes of common nouns, I have saidenough ; and it can easily be observed that hereusage governs rather than Regularities.XLL 80. Thematter of proper nouns now follows,which differ from common nouns in that they aredefinite and denote special things, like the namesParis and Helen, while common nouns are indefiniteand indicate general ideas, like vir ' man ' and mulier'woman.' Among these there are some propernames from proper names, Uke Ilium from Ilus," andIlia * from Ilium ; others are from a common noun,like Alhius '^ from album

VARRO81. (Item Perpenna debuit esse)^ Perpenni filia,non Perpennae (filius. Nam)^ Perpenna mulierisnomen esse debuit et nata esse a Perpenno, quod estut Art;ernus' Percelnus Perpennus, Arverna' PercelnaPerpenna. Quod si Marcus Perpenna virile estnomen et analogia sequenda, Lucius ^elia et QuintusMucia virilia nomina esse debebunt ; item quaedicunt ab Rhodo, Andro, Cyzico Rhodius, (Andrius),*Cyzicenus, similiter Cyzicius dici (debebat),* et civisunus quisque : now* ut Athenaeus dicitur rhetornomine, etsi non sit Atheniensis.82. In hoc ipso analogia non est, quod alii nominahabent ab oppidis, alii aut non habent aut non ut debenthabent.83. Habent plerique libertini a municipio manumissi,in quo, ut societatum et fanorum servi, nonservarunt proportioned rationem, et Romanorumliberti debuerunt dici ut a Faventia Faventinus, abReate Reatinus sic a Roma Romanus, ut nominentur^libertini* orti * publicis servis Romani, qui manumissiante quam sub magistratu(u)/«^ nomina, qui eosliberarunt, succedere c(o>eperunt.§ 81. ^ item added by Stephanus, Perpenna hy Kent,debuit esse by L. Sp. ^ Added by Kent. ^ Mue., forArb-. * Added by Aldus. ^ Added by Mue. ^ Canal,for nam.§ 83. ^ L. Sp., for proportionem. * Vertranius, fornominantur. * Vertranius, for a libertinis. * Addedby Mue. * Sciop., for magistratus.§81. "A well-known Roman family name of Etruscanorigin; masculine, though of the first declension. "Insteadof the actual Aelius and Mucins. ' Of the second century436

ON THE LATIN LANGUAGE, VIIL 81-8381 Likewise, Perpenna " ought to be the daughterof Perpennus, not the son of Perpenna. For Perpennaought to be the name of a woman, and to mean a childof Perpennus ; this is Uke mascuhne Arvernus, Percelnus,feminine Arverna and Percelna, so also Perpennusand Perpenna. But if Marcus Perpenna is a man'sname and Regularity is to be followed, then LuciusAelia and Quintus Mucia * ^nll have to be men's names.Likewise the names which they use derived fromRhodus Rhodes,' Andros, and ' Cyzicus, are Rhodius'Rhodian,' Andrius Andrian,' and Cyzicenus Cyzi-' 'cene ' ; but if made in like manner the last ought tobe Cyzicius, and each name ought to denote a citizenof the place : not as a certain rhetorician is calledAthenaeus," although he is not an Athenian by birth.82. In this very matter, then, there is no Regularity,because some have names from the towns,others either have names from other sources or havenames from towns from which they ought not to getthem.83. Most freedmen set free in a free to^vn get theirnames from the to\\Ti ; in which, as slaves of guildsand temples, they have not observed the theory withproper relation ; and the freedmen of the Romansought to have got the name Romanus,'^ like Faventinusfrom Faventia and Reatinus from Reate.In this waythe freedmen whose parents were state slaves wouldbe named Romanus, who had been set free beforethey began to take the names of the magistrateswho set them free.not to be confused with the more celebrated AthenaeusB.C. :of Xaucratis, of the second century a.d., who wrote theDeipnosoph Istae.§ 83. » C/. Livy, iv. 61. 10.437

VARRO84. Hinc quoque ilia nomina Lesas, Ufenas,Carrinas, Maecenas, quae cum essent ab loco utUrbinas, et tamen Urbin(i>us,i ab his debueruntdici ad nostrorum nominum (similitudinem* LesiusUfenius Carrinius Maecenius)' ... *§ 84. ^ GS, ; Urbinus older editions ; for uerbinus.^ Added hy Laetus. ^ Added by GS. * Here the rest ofthe line, and all the reverse of the folio, are left vacant in F ;hut the remaining material xchich was contained in the bookwould have occupied mvch more space. Probably an entirequaternion had been lost from the archetype of F.438

ON THE LATIN LANGUAGE, VIIL 8iSI. From this practice came also such names asLesas, Ufenas, Carrinas, Maecenas " ; since these arefrom the place of origin, like Urbinas,^ alongsideUrbinius, there should from them have been formed,after the likeness of our names, the names Lesius,Ufenius, Carrinius, Maece/iius.'^ . . .§ 84. " Regularly formed cognomina ; Lesas is not otherwiseknown, but the other three are. Maecenas was thefriend of Augustus and the patron of literary men. * FromUrbinum in Umbria. ' Inasmuch as Roman gentilenames almost always ended in -ius.439

MARCUS TERENTIUS VARRO'SON THE LATIN LANGUAGEBOOK VIIIEXDS HERE, AND HERE BEGINSBOOK IXI. 1. ... They are a prey to extraordinary error,who prefer to teach what they do not know, ratherthan to learn that of which they are ignorant. Inthis position was the famous grammarian Crates,who placed his reliance on Chrysippus, a man ofgreat acumen who left three books On Anomaly,and contended against Regularity and Aristarchus,but in such a way—as his wTitings show—that he doesnot seem to have understood thoroughly the intentof either. For Chrysippus, when he \\Tites about theInconsistency of speech, has as his object the showingthat like things are denoted by unlike words andthat unlike things are denoted by like words, as is trueand Aristarchus, when he writes about the Consistencyof the same, bids us follow a certain Ukeness ofwords in their derivation, as far as usage permits.' For Aristharcum. * Stephanas, with B, for sermones.* dissimilis similibus Wilmanns, for dissimilibus similes.* Aldus, for uerbum. ' Mue., for cum. * Mue., forconscribit. • GS., for et de. ^" Groth, for quarundam." A. Sp. ; in declinatione L. Sp. ; for inclinationes.441

VARRO2. Sed ii qui in loquendo partim sequi iube(n>t^nos consuetudinem partim rationem, non tarn discrepant,quod consuetude et analogia coniunctioressunt inter se quam iei credunt,3. quod est nata ex quadam consuetudine analogiaet ex hac (consuetudine item anomalia.^ Quarequod*) consuetudo ex dissimilibus et similibus verbiseorumque' declinationibus constat, neque anomalianeque analogia est repudianda, nisi si non est homoex anima, quod est* ex corpore et anima.4«. Sed ea quae dicam quo facilius pervideri possint,prius de trinis copulis discernendum (nam^ confusimex utraque parte pleraque dicuntur, quorum* alia adaliam referri debent summam) : primum de copulisnaturae et (u)suis' : haec enim duo sunt quo derigunt

ON THE LATIN LANGUAGE, IX. 2-52. But those who give us adWce in the matter of^peaking, some saying to follow usage and otherssaying to follow theory, are not so much at variance,because usage and regularity are more closely connectedwith each other than those advisers think.3. For Regularity is sprung from a certain usagein speech, and from this usage likewise is sprungAnomaly. Therefore, since usage consists of unlikeand like words and their derivative forms, neitherAnomaly nor Regularity is to be cast aside, unlessman is not of soul because he is of body and of soul.-i. But that what I am about to say may be moreeasily grasped, first there must be a clear distinctionof three sets of relations ; for most things are saidindiscriminately in two ways, and of them someought to be referred to one principle and others toother principles. First, the distinction of the relationsof nature and use " ; for these are two factorswhich are diverse in the goals toward which theydirect themselves, because it is one thing to saythat Regularities exist in words, and another thing tosay that we ought to follow the Regularities. Second,the distinction of the relations of extension andlimitation, whether the use of the Regularities shouldbe said to be proper in all words, or only in a majorityof them. Third, the distinction in the relations ofthe speaking persons, how * the majority of personsought to observe the Regularities.5. For some words and forms are the usage ofthe people as a whole, others belong to individualpersons ; and of these, the words of the orator andthose of the poet are not the same, because theirgenitive of the fourth declension in -UIS. * Qui is herethe ablatival adverb.443

VARROidem ius. Itaque populus universus debet in omnibusverbis uti analogia et, si perperam est consuetus,eorrigere se ipsum, cum orator non debeat in omnibusuti, quod sine ofFensione non potest facere, cum poeta^transilire lineas impune possit.6. Populus enim in sua potestate, singuli in illius :itaque ut suam quisque consuetudinem, si mala est,eorrigere debet, sic populus suam. Ego populi consuetudinisnon sum ut dominus, at ille meae est. Utrationi optemperare debet gubernator, gubernatoriunus quisque in navi, sic populus rationi, no*^ singulipopulo. Quare ad quamcumque summam indicendo referam si animadvertes, intelleges, utrumdicatur analogia esse an uti oportere (ea ; itemqueintelleges si ad analogiam usum loquendi oportea>tredigere, tu??« dici id in populum aliter ac (in singulosnee) i(de);n de omnibus dici* in eum qui sit inpopulo.II. 7. Nunc iam primum dicam pro universa analogia,cur non modo (non)^ videatur esse reprehendenda,sed etiam cur in usu quodammodo sequenda ;secundo de singulis criminibus, quibus rebus possintquae dicta sunt contra solvi, dicam ita ut generatim§ 5. ^ L, Sp., for poetae.§ 6. ^ Laetus, B, for non. * F has here uti oporteretredigeretur dici id in populum aliter ac inde omnibusdici / Aug., with B, read redigere for redigeretur / Mue.emended to uti oportere ea ; et quom poscitur ut usus ad idquod oporteret redigeretur dici, etc., deleting inde omnibusdici as a gloss ; Reiter proposed uti oportere et redigere, turndici, tcith the same deletion ; GS. proposed that which is in thetext, except that for their usus loquendi oporteret redigeretur,/ have adopted usum loquendi oporteat redigere turn, takingoporteat from Fay (ubi oporteat redigere tibi dici, with therest like Mueller^s version), and redigere tum from Reiter.§ 7. ^ Added by Stephanus.444

ON THE LATIN LANGUAGE, IX. 5-7rights and limitations are not the same. Thereforethe people as a whole ought in all words to useRegularity, and if it has a wrong practice, it oughtto correct itself ; whereas the orator ought not to useRegularity in all words, because he cannot do so withoutgi\ing offence, and on the other hand the poetcan ^yith impunity leap across all the bounds.6. For the people has power oyer itself, but theindividuals are in its power ; therefore as each oneought to correct his own usage if it is bad, so shouldthe people correct its usage. I am not the masterso to speak—of the people's usage, but it is of mine.As a helmsman ought to obey reason, and each onein the ship ought to obey the helmsman, so the peopleought to obey reason, and we indi\iduals ought toobey the people. Therefore, if you vriW take notice ofeach principle on which I shall base my argument inthe matter of speaking, you will appreciate whetherRegularity is said merely to exist, or it is said thatwe ought to follow it ; and hkewise you ^^^I1 appreciatethat if the practice of speech ought to bereduced to Regularity, then this is meant for thepeople in a different sense from that in which it ismeapt for individuals, and that that which is takenfrom the entire body of speakers is not necessarilymeant in the same form for him who is only an individualin the people.II. 7. Now I shall speak first in support of Regularityas a whole, why, as it seems, it not only shouldnot be censured, but even should in practice befollowed in a certain measure ; and secondly, concerningthe several charges against it, I shall give thearguments by which the objections can be refuted,arranging them in such a way that I shall include,4+5

VARROcomprehenda;»^ et ea quae in priore libro sunt dictaet ea quae possunt did atque* illic praeterii.III. 8. Primum quod aiunt, qui bene loqui velitconsuetudinem sequi oportere, non rationem similitudinum,quod, alteram* si neglegat, sine offensionefacere non possit, alteram* si sequatur, quod sinereprehensione non sit futurum, e

ON THE LATIN LANGUAGE, IX. 7-10item by item, those which have been narrated in theprevious book and also those which can be presentedbut were passed over by me in that place.III. 8. First, as to their alleging that he whowishes to speak well ought to observe usage and notthe theory of likenesses, because if he disregardsthe former he cannot do so without giving offence, andif he follows the latter it will not be without incurringrebuke : they are mistaken, because he who inspeaking follows the usage which he ought to employ,is following it also without disregard of the theory.IV. 9^ For we see that nouns and verbs whichwe inflect in similar ways are in general usage, andwe compare others with this usage, and if there isany error we make the correction with the help ofusage. For if those who have arranged the diningroom" have among the three couches set one that inof a different size, or among couches that matchhave brought one too far forward, or not far enough,we join in making the correction according to commonusage and to the analogies of other dining-rooms ;in the same way, if in speech any one in his utteranceshould so inflect the words as to speak irregularforms, we ought to revise his mistake according tothe model of other similar words.V. 10. Now there are two kinds of wrong forms ininflection ; one, that which has been erroneouslyaccepted into general usage ; the other, that whichisnot yet so accepted and may be called incorrect.The latter they grant ought not to be said, becauseit is not in usage, but as for the former they merelydo not admit the propriety of saying it in this way ;stood the dining-table.symmetrically placed.The couches should be identical and447

VARROut si' similiter, cum id faciant, ac, si quis puerorumper delicias pedes male ponere atque imitari vatiasc(o>eperit, hos corrigi oportere si conceda(n>t,* contrasi quis in consuetudine ambulandi iam factus sit vatiaaut conpernis, si eum corrigi non conceda(n)t.*11. Non sequitur, ut stulte faciant qui pueris ingeniculis alligent serperastra, ut eorum depravatacorrigant crura ? Cum vetuperandus^ non sit medicusqui e longinqua mala consuetudine aegrum in melioremtraducit, quare reprehendendus sit qui orationemminus valentem propter malam consuetudinem traducatin meliorem ?VI. 12. Pictores Apelles/ Protogenes, sic aliiartufices egregii non reprehendundi, quod consuetudinemMiconos, Dion's,^ Arimmae, etiam superiorumnon sunt secuti : Aristophanes improbandus, quipotius in quibusdam veritatem' quam consuetudinemsecutus ?VII. 13. Quod si viri sapientissimi, et in remilitari et in aliis rebus multa contra veterem consuetudinemcum essent (a)usi,^ laudati, despiciendisunt qui potiorem dicunt oportere esse consuetudinemratione.VIII. 14. An cum quis perperam consuerit quidfacere in civitate, non modo (non>^ patiemur, sed' Laetus, for si. * Aldus, for concedat.§11. ^ G, H, Victor ius, for detuperandus.§ 12. ^ For Appelles. ^ Aug., with B, for Dioros.Laetus, for ueteritatem.' J),§ 13. ^ Canal, for usi.§ 14. ^ Added by Aug., with B.§11. " The interrogation-mark was placed here byMueller ; the question is sarcastic.448

ON THE LATIN LANGUAGE, IX. 10-14so that when they do this it is just as if they shouldgrant that the boys ought to be corrected in caseany .of them in wilfulness begins to manage his feetawkwardly and to imitate the bowlegged, but shouldrefuse to grant that one should be corrected if he inhis habit of walking has already become bowleggedor knock-kneed.1 1 . Does it not follow that they act foolishly whofasten splints on the knees of children, to straightentheir crooked leg-bones ? " Since even that physicianis not to be censured who makes a healthier man outof one who has been ill as a result of a long-continuedbad habit, why should he be blamed who brings intobetter condition a way of speech which has been lesseflFective on account of bad usage ?VI. 12. The painters Apelles " and Protogenes,^and other famous artists are not to be blamed becausethey did not follow the ways of Micon," Diores,**Arimmas,'' and even earlier craftsmen ; then mustAristophanes * be condemned because in some thingshe followed reality rather than usage ?Vn. 13. But if the wisest men have been praisedbecause both in warfare and in other things theyhad dared do much that was against old usage, thenthey must be despised who say that usage ought tobe considered as better than good theory.VTII. 14. Or when a person has been accustomedto do something wrong in civil life, shall we not only§ 12. " Distinguished Greek painter of the time of Alexanderthe Great. * Distinguished Greek painter, contemporaryof Apelles. « Sculptor and painter at Athens,middle**of the fifth century b.c. Entirely unknownotherwise ; the names are perhaps corrupt, cf. Bergk inPhilol. XXX. 682 (1870), and Georges in Fleckeisen^s Jahrbucher,cxxxv. 768 (1887). « See v. 9 note a.VOL. II G 449

VARROetiam p

ON THE LATIN LANGUAGE, IX. 14-17not tolerate him but even visit him ^^^th punishment—and yet " if a person has the habit of saying aword wrong, shall we not correct him, when thismay be done without actual punishment ?IX. 15. And these men " who send their boys toschool to learn how to wTite words which they don'tknow—shall we not likewise instruct these men,bewhiskered adults as they are, who du not knowhow the words ought to be spoken, that they mayknow by what logical theory they may properly bepronounced ? *X. 16. But as the nurse does not with suddennesstear her nurslings away from their wontedmethod of feeding, when she changes them from theirfirst food to a better, so we ought to go graduallyand judiciously in matters of speech, in changingolder persons from less suitable words to thosewhich accord with logical theory. Since among theillogical words which are in common usage there aresome which can easily be eliminated, and others ofsuch a sort that they seem firmly fixed," it is properto correct at once in the direction of logic only thosewhich are lightly attached and can be changedwithout giving offence ; but those which are suchthat for the present you cannot make the correctionso as not to speak them thus, these you ought, ifpossible, to refrain from using. For thus they willbecome unwonted and afterward, when alreadyblurred to the memory, they can be more easilycorrected.XI. 17. Such new inflectional forms as are introducedby logical theory but are rejected by thespeech of the forum, these the good poets, especially§ 16. " Cf. § 10.451

VARROscaenici, consuetudine subigere aures populi debent,quod poetae multum possunt in hoc : propter eosquaedam verba in declinatione melius, quaedamdeterius dicuntur. Consuetude loquendi est inmotu : itaque sole(n>t* fieri et meliora' deteriora ' perperam dicuntur.XII. 18. Quare qui ad consuetudinem nos vocant,si ad rectam, sequemur : in eo quoque enim estanalogia ; si ad earn invitant quae est depravata,nihilo^ magis sequemur, nisi cum erit necesse, quam^in ceteris rebus mala exempla : nam ea quoque, cumaliqua vis urget, inviti sequemur. XIII. Neque enimL^sippus artificum priorum potius (secutus)^ estvitiosa quam artem ; sic populus facere debet, etiamsinguli, sine ofFensione quod fiat populi.19. Qui amissa

ON THE LATIN LANGUAGE, IX. 17-20the dramatists, ought to force upon the ears of thepeople and accustom them to them. For the poetshave great power in this sphere : they are responsiblefor the fact that certain words are now spoken \Wthimproved inflections, and others vrith worse. Theusage of speech is always shifting its position : this iswhy words of the better sort are wont to becomeworse, and worse words better ; words spoken wTonglyby some of the old-timers are on account of the poets'influence now spoken correctly, and on the otherhand some that were then spoken according to logicaltheory, are now spoken wTongly.XII. 18. Therefore those who summon us to obeyusage, we shall follow, if it be to a correct usage.For in this also there is the principle of Regularity :if they in\ite us to that usage which is pervertedand irregular, we shall not follow it unless it becomesnecessary, any more than we follow bad examples inother things ; for we do follow them too, thoughagainst our inclinations, when some force bears downupon us. XIII. And in fact Lysippus " did notfollow the defects of the artists who preceded him,but rather their artistry ;just so should the peopledo in their speech, and even the individuals, so faras itwhole.may be done ^^ithout offence to the people as a19- There are some persons who not only hunt forlost articles, but even of their own initiative giveany information which they may have : do the samepersons, if something has been lost from speech, notonly not exert themselves in hunting for it, buteven fight against the informers, to keep it frombeing put back into its place ?20. As for a word that is new and has been intro-453

VARROquo minus^ recipiamus, vitare non debemus. XIV.Nam ad usum in vestimentis aedificiis supellectili^novitati non impedit vetus consuetudo : quem enimamor assuetudinis potius in pannis possessoremretinet, quem ad nova vestimenta traducit ? XV. Annon saepe veteres leges abrogatae novis cedunt ?XVI. 21. Nonne inusitatis formis vasorum recentibuse Graecia arflatis^ obliteratae antiquaeconsuetudinis smorum et capularum* species ? Hisformis vocabulorum incontaminati' uti nolentqua** docu

ON THE LATIN LANGUAGE, IX. 20-24duced according to logical theory, we ought not forthis to shun giving it a hospitable welcome. XI\\For long-standing custom is not a hindrance tonovelty in garments, buildings, and utensils, whenit is a question of use ; what victim of a habit doesthe love of that habit rather keep in rags, when thelove of novelty " is leading him toward new garments? XV. Are not old laws often annulled andsucceeded by new laws ?XVI. 21. Have not the forms of the old-fashionedpots and cups been swept into oblivion by the unfamiliarshapes of the vessels recently brought fromGreece ? Shall they then, on account of old-timehabit, be un-vnlling to use these unsullied forms ofwords, which good reason has taught them ? Anddo they claim that there is such difference betweenthe two senses, that for their eyes that are alwaysseeking some new shapes of their furniture, but theywish their ears to have no share in similar novelties ?XVII. 22. Out of how many slave-owners is therenow one who has slaves bearing the ancient names ?WTiat woman calls her outfit of clothing and jewelryby the old words ? But it is not so much at the unlearnedthat anger must be felt, as at the advocatesof this perversity.23. For if there were Regularity in no place at all,then it follows that there would be none in wordseither ; not that when it is everywhere present (asit is in fact), there is none in words. XVIII. Forwhat part of the world is there which does nothave countless Regularities ? Sky or sea or land,what Regularities are there in these ?24. As in the sky there is a division from the§ 20. " Supply amor novitatis as subject of traducit.455

VARROsolstitialem et hinc ad septemtrionalem divisum, siccontra^ paribus partibus idem a bruma versum contrariaparte ? Non quantum polw^s)" superior abest'a septemtrionali cir(culo et is a solstitiali, quem solcir>cumit cum it* ad solstitium/ tantundem abestinferior ab eo quem dvTa/jKTiKov*vocant astrologi etis a brumalt ?' Non, quemadmodum quodque signumexortum hoc anno,* quotquot annis eodem modoexoritur ?25. Num aliter sol a bruma venit ad aequinoctium,ac contra cum ad solstitium venit, ad aequinoctialemcirculum et inde ad brumam ? Nonne luna, ut absole discedit ad aquilonem et inde redit in eandemviam, sin inde fertur ad austrum et regreditur inde ?Sed quid plura de astris, ubi difficilius reperitur quidsit aut fiat in motibus dissimiliter ?XIX. 26. At in mari, credo, motus non habentsimilitudines^ geminas, qui in XXIII' horis lunaribuscotidie quater se mutant, ac cum sex horisaestus creverunt, totidem decreverunt, rursus idem,itemque ab his. An hanc analogian ad diem servant,ad mensem non item, alios motus sic item cum§ 24. ^ For contra a. * Scaliger, for polo. ' Mue.deleted et abest et after abest. * Added and changed byGS. (a solstitiali Kent, for ad solstitialem GS.), for circumitcum his. ^ For solistitium. * B, for arti articon.''Kent, for ad brumalem. ^ Aid us deleted quod afteranno.§26. ^ J/wf .,/(5r dissimilitudines. ^ Aldus,for XXlll.§ 26. » Spoken in sarcasm. * Slightly longer thansolar hours. ' In relation to the month ; non-existent, butassumed for purpose of argument.456

TheON THE LATIN LANGUAGE, IX.24^26Equator to the Tropic of Cancer, and from there tothe Arctic Circle, is not also its counterpart, extendingfrom the Tropic of Capricorn in the other direction,likewise divided into equal sections ? Is it nota fact that as far as the North Pole is removedfrom the Arctic Circle and this from the Tropicof Cancer, around which the sun travels when itcomes to the summer solstice, so far the South Poleis from that Circle which the astronomers callthe Antarctic, and this from the Tropic of Capricorn ?Is it not true that in the fashion in which each constellationhas risen in the sky this year, in just thesame fashion it rises each and every year ?25. The sun does not come in one way from theTropic of Capricorn to the Equator, does it, and onthe other hand, when it comes to the Tropic ofCancer, return in a different way to the Equator and.''thence to Capricorn moon, when it goesaway from the sun to the north and returns from thereinto the same path which the sun pursues, goes onfrom there to the south and comes back again injust the same way, does it not ? But why should Ispeak further of the stars, in the case of which thereis unusual difficulty in finding any irregularity whichexists or takes place in their motions ?XIX. 26. But in the sea, I suppose, the motionsdo not have the twofold likenesses " —the motionswhich in twenty-four lunar * hours change themselvesfour times, and when the tides have risen forsix hours, and have ebbed for just as many, theylikewise rise again, and in the same fashion ebbafter this time. Or do they keep this Regularity fora day's space, and not like-wise for a month, sincesimilarly they have another set of motions " which457

VARROhabeant aliz's' inter se convenientes ? De quibus inlibro quern de v4estuariis feci scripsi.XX. 27. Non in terra in sationibus servata analogiaNee cuius modi in praeterito tempore fruc-.''tuum genera reddidit, similia in praesenti reddit, etcuius modi tritico iacto reddidit segetes, sic Aordeosato proportione reddidit parilis ? Non, ut Europahabet flumina lacus, mentis campos, sic habet Asia ?XXI. 28. Non in volucribus generatim servaturanalogia ? Non ex aquilis aquilae atque ut ex turdisqui procreantur turdi, sic ex reliquis sui^ cuiusquegeneris ? XXII. An aliter hoc fit quam in acre inaqua ? Non hie conchae inter se generatim innumerabilinumero similes ? Non pisces } An e^muraena fit lupus aut merula? Non bos ad bovemcollatus similis, et qui ex his progenerantur inter sevituli ? Etiam ubi dissimilis fetus,' ut ex (asino et)*equa mulus, tamen ibi analogia : quod ex quocumqueasino et equa nascitur id est mulus aut mula, ut exequo et asina hinnulei.XXIII. 29. Non sic ex viro et muliere omnissimilis partus, quod pueri et puellae ? Non horumita inter se^ omnia similia membra, ut separatim insuo utroque genere similitudine sint* paria ? Non,omnes cum sint ex anima et corpore, partes quoque'horum proportione similes ?' Reiter, for alios.§ 28. ^ Aug., tcith B, for suis. * an e Avg., for sane.' Aug., for faetus. * Added by L. Sp. ; Aug., with B,added et asino after equa.§ 29. ^ Sciop. deleted non after se. " Aug., for similitudinesintra. ' M, p, Laetus, for quaque.458

ox THE LATIN LANGUAGE, IX. 26-29agree with one another ? Of these I have "\ratten inthe book which I composed On Tidal Inlets.XX. 27. On the earth, is not Regularity preservedin the case of plantings ? Does it not give usto-day fruits of precisely the same kind as it hasgiven us in the past } Does it not regularly return tous a crop of barley when barley has been sown, evenas it returns a crop of wheat when wheat has beensown ? Does not Asia have rivers and lakes, mountainsand plains, even as Europe has ?XXI. 28. Is not Regularity preserved among thebirds, according to their kind ? As the progeny ofeagles are eagles and the progeny of thrushes arethrushes, are not the progeny of the other birds all oftheir own proper and special kind ? XXII. Doesthe process go on in another way in the water, thanin the air ? Are not the shell-fish here all like theirown kind, despite their countless number ? Arenot the fishes ? Is a pike or a sea-carp produced ofa moray ? Is not one head of cattle like another,when compared, and so also the calves which areprocreated by them ? Even where the offspring isunlike the parents, as the mule born of a he-ass anda mare, even there there is Regularity none the less :the offspring of any ass and mare whatsoever is amule, male or female, as the offspring of a stallionand a she-ass is a hinny.XXIII. 29. Are not in this way all the offspringof man and woman alike, in that they are boys andgirls ? Do these not have all their limbs mutuallyalike, in such a way that item by item they are pairsin likeness, in their own special kinds ? As all aremade up of soul and body, are not also the parts ofsoul and body alike with the same regularity ?459

Ecquid''VARRO30. Quid ergo cum omnes animae hominum sint^divisae in octonas partes, hae* inter se non proportionesimiles ? Quinque quibus sentimus, sexta qua cogitamus,septuma qua progeneramus, octava qua vocesmittimus ? Igitur quoniam qua loquimur voce oratioest,^ banc quoque necesse est natura habere analogias:itaque habet.XXIV. 31. An non vides, ut Graeci habeant earnquadripertitam, unam in qua si(n>t^ casus, alteramin qua tempora, tertiam in qua neutrum, quartum.''in qua utrumque, sic nos habere verbanescis ut apud illos sint alia finita, alia now,' sicutra(que>* esse apud nos ?32. Equidem non dubito, qui(n) animadvertm*'item in ea* innumerabilem similitudin?

ON THE LATIN LANGUAGE, IX. 30-3230. WTiat then of the fact that the souls of menare di\ided into eight parts—are these parts notmutually alike with regularity ? I'ive ^^-ith whichwe perceive," the sixth Anth which we think, theseventh with which we procreate, the eighth -«-ithwhich we utter articulate words ? Therefore sincethe Avord ^nth which we talk is speech, speechalso must by nature have its Regulaiities ;and itdoes.XXIV. 31. Do you not see that the Greeks havedivided speech into four parts, one in which the wordshave cases, a second in which they have indicationsof time, a third in which they have neither, a fourthin which they have both " —and that in the sameway we have all these divisions ? Do you not knowthat among them some words are definite, othersnot * —and that both kinds are present in ourlanguage also ?32. For my part I have no doubt that you haveobserved the countless number of likenesses in speech,such as those of the three tenses of the verb, or itsthree persons. XXV. Who indeed can have failedto join vou in observing that in all speech there arethe three tenses lego ' I read,' legebam ' I wasreading,'I shall read,' and similarly the'kgamthree persons lego ' I read,' legts thou readest,*'legit he ' reads,' though these same forms may bespoken in such a way that sometimes one only ismeant, at other times more ? Who is so slow-vvittedthat he has not observed also those hkenesses whichwe use in commands," those which we use in \^-ishes,those in questions, those in the case of matters notperatives and subjunctives) exhibit certain regular resemblances; and so do those used in wishes, etc.461

VARROin interrogando, quibus in infectis rebus, quibus inperfectis, sic in aliis discriminibus ?XXVI. 33. Quare qui negant esse rationem^analogiae, non vide(n>t'^ naturam non solum orationis,sed etiam mundi ;qui autem vident et sequinegant oportere, pugnant contra naturam, non contraanalogian, et pugnant volsillis, non gladio, cum paucaexcepta verba ex pelago sermonis (po)puli^ minus(usu>* trita afferant, cum dicant propterea analogiasnon esse, similiter ut, si quis viderit mutilum bovemaut luscum hominem claudicantemque equum, negetin^ bovum hominum et equorum natura similitudinesproportione constare.XXVII. 34-. Qui autem duo genera esse dicuntanalogiae, unum naturale, quod ut ex *atis^ nascuntur(lentibus)^ lentes' sic ex (lupino)* lupinum, alterumvoluntarium, ut in fabrica, cum vident sccenam utin dexteriore parte sint ostia, sic esse in sinisterioresimili ratione factam, de his duobus generibusnaturalem esse analogian, ut sit in motibus caeli,voluntariam non esse, quod ut quo

ON THE LATIN LANGUAGE, IX. 32-34completed and those for matters completed, andsimilarly in other diiferentiations ?XXVI. 33. Therefore those who say that there isno logical system of Regularity, fail to see the naturenot only of speech, but also of the world. Thosewho see it and say that it ought not to be followed,are fighting against nature, not against the principleof Regularity, and they are fighting with pincers,not with a sword, since out of the great sea of speechthey select and offer in e\idence a few worck notvery familiar in popular use, saying that for thisreason the Regularities do not exist : just as if oneshould have seen a dehorned ox or a one-eyed manand a lame horse, and should say that the Ukenessesdo not exist \\1th regularity in the nature of cattle,men, and horses.XX\ II. 34. Those moreover who say that thereare two kinds of Regularity, one natural, namelythat lentils grow from planted lentils, and so doeslupine from lupine, and the other voluntary, as in theworkshop, when they see the stage as having anentrance on the right and think that ithas for a likereason been made with an entrance on the left ; andsay further, that of these two kinds the naturalRegularity really exists, as in the motions of theheavenly bodies, but the voluntary Regularity is notreal, because each craftsman can make the parts ofthe stage as he pleases : that thus in the parts ofmen there are Regularities, because nature makesthem, but there is none in words, because men shapethem each as he wills, and therefore as names for thesame things the Greeks have one set of words, theSyrians another, the Latins still another " —I firmlythink that there are both voluntary and natural463

VARROesse puto, voluntarios quibus homines vocabulaimposwerint' rebus quaedam, ut ab Romulo Roma,ab Tihure^ Tiburtes, naturales ut ab impositis voeabulisquae inclinantur in tempora' aut in casus,ut ab Romulo Romuli Romulum et ab dico dicebamdixeram.35. Itaque in voluntariis declinationibus inconstantiaest, in naturalibus constantia ;quae utrasquequoniam iei non debeant negare esse in oratione,quom^ in mundi partibus omnibus sint, et declinationesverborum innumerabiles, dicendum est essein his analogias. Neque ideo statim ea in omnibusverbis est sequenda : nam si qua perperam declinavitverba consuetudo, ut ea aliter

ON THE LATIN LANGUAGE, IX.34^37derivations of words, voluntary for the things onwhich men have imposed certain names, as Romefrom Romulus and the Tiburtes men of Tibur from' 'Tibur, and natural as those which are inflected fortenses or for cases from the imposed names, asgenitive Romuli and accusative Romulum fromRomulus, and from dico ' I say the ' imperfect dicehamand the pluperfect dixeram.S5. Therefore in the voluntary derivations thereis inconsistency, and in the natural derivations thereis consistency. Inasmuch as they ought not to denythe presence of both of these in speech, since they arein all parts of the world, and the derivative forms ofwords are countless, we must say that in words alsothe Regularities are present. And yet Regularitydoes not for this reason have to be followed in allwords ; for if usage has inflected or derived anywords wrongly, so that they cannot be uttered withoutgiving offence to many persons, the logic of speakingshows us that because of this offence the logic of thewords must be set aside."XXVIII. SQ. As far as concerns the generalcause why likeness is present in speech and ought tobe observed, and also to what extent this should bedone, enough has now been said. Therefore in thefollowing we shall set forth its several parts item byitem, and refute the individual charges which theybring against the Regularities.37. In this matter, you should take notice that bynature there are four elements in the basic situationto which words must be adjusted in inflection : theremust be an underlying object or idea to be defavourof an irregular form if the feeling {Sprachgefiihl) ofthe speakers rebels against it.VOL. II H 465

VARROde.signetur,2 et lit sit ea res' in usii, et ut vocis naturaea sit quae significavit, ut declinari possit, et similitudofigura(e>'' verbi ut sit ea quae ex se declinat?*^genus prodere certum posst't.*38. Quo neque a terra terrus ut dicatur postulandumest, quod natura non subest, ut in hoc alterummaris, alterum feminae debeat esse ; sic nequepropter usum, ut Terentius significat unum, pluresTerentii, postulandum est, ut sic dicamus faba etfabae : non enim in simili us(u>^ utrumque ; nequeut dicimus ab Terentius Terentium, sic postulandumut inclinemus ab A et B, quod non omnis vox naturahabet declinatus.39. Neque in forma collata quaerendum solum,quid habeat in figura simile, sed etiam nonnunquamin eo quem habeat effectum. Sic enim lana Gallicanaet Apula videtur imperito similis propter speciem,cum peritus Apulam emat pluris, quod in usu firmiorsit. Haec nunc strictim dicta apertiora fient infra.Incipiam hinc.XXIX. 40. Quod rogant ex qua parte oporteatsimile esse verbum, a voce an a^ significatione, respondemusa voce ; sed tamen nonnunquam quaerimusgenere similiane sint quae significantur ac nomen* Laetus, for designentur. ' G, H, a, Laetus, for cares.*Mue.,for figura. * L. Sp.,for declinata. * Aug., forpossunt.§ 38. ^ L. Sp., for similius.§ 40. ^ After Laetus, ab voce an, for aboceana.§ 38. " The singular /«ia was used also collectively for theplural or mass idea; cf. Priscian, ii. 176 Keil. * Names ofletters.§ 39. " Cf. § 92.§ 40. " (Jf. viii. 40.466

ON THE LATIN LANGUAGE, IX. 37-40signaled ; this object or idea must be in use ; thenature of the utterance which has designated it,must be such that it can be inflected ; and the resemblanceof the word's form to other words must besuch that of itself it can reveal a definite class inrespect to inflection.38. Therefore it is not to be demanded that fromterra ' earth there should be ' also a terms, becausethere is no natural basis that in this object thereought to be one word for the male and another forthe female. Similarly, ^\ith respect to usage, whileTerentius designates one person of the name andTerentii designates several, it is not to be demandedthat in this way we should say J'aba bean and Jabae' ''beans,' for the two are not subject to the sameuse." Nor is it to be demanded that as we sayace. Terentium from nom. Terentius, we should makecase-forms from xi and B,^ because not every utteranceis naturally fitted for declensional forms.39. The likeness which the word has in its shapemust be investigated not in the comparison of thebasis merely, but also sometimes in the effect which ithas. For thus the Gallic wool and the Apulian woolseem alike to the inexperienced on account of theirappearance, though the expert buys the Apulian at ahigher price because in use it lasts better. Thesematters, which have been touched upon hastilyhere, will become clearer in a later discussion."Now I shall start.XXIX. 40. To their question in what respect aword ought to be similar, sound or meaning," weanswer that it should be so in sound. But yet sometimeswe ask whether the objects designated arelike in kind, and compare a man's name with a man's,467

VARROvirile cum virili conferimus, feminae cum muliebri :non quod id quod significant vocem commoveat, sedquod nonnunquam in re dissim(ili par)ilis^ figuraeformas in simih' imponunt dispariles,* ut calcei muliebressint an viriles dicimus ad similitudinem figurae,cum tamen sciamus nonnunquam et mulierem haberecalceos viriles et virum muliebris.41. Sic dici virum Perpennam ut AZ/enam^muliebri forma^ et contra parietem ut abietem esseforma' similem, quo(m>'' alterum vocabulum dicaturvirile, alterum muliebre et utrumque natura neutrumsi

ON THE LATIN LANGUAGE, IX. 40-43a woman's name >\ith a woman's : not because thatwhich they designate affects the word, but becausesometimes in case of an unUke thing they set uponit forms of an equivalent appearance, and on a Ukething they set unequal forms, as we call shoes women'sshoes or men's shoes by the likeness of the shape,although we know that sometimes a woman wearsmen's shoes and a man wears women's shoes.41. In like fashion, we say, a man is calledPerpenna, like Alfena, with a feminine form " ; andon the other hand paries house-wall ' ' is Uke abies'fir-tree 'in form, although the former word is used asa masculine, the latter as a feminine, and both arenaturally neuter. Therefore those which we useas masculines are not those which denote a malebeing, but those before which we employ hie and hi,and those are feminines with reference to which wecan say haec or hae.^XXX. 42. P'or this reason it amounts to nothing,that on the premise that Regularity adopts theunUkenesses of the objects as a criterion for differencein the forms " of the spoken word,*' they say thatTheon and Dion are not alike if the one is an Ethiopianand the other is a white man.

VARROinter se parentes sint, de filiis iudicare* : errant, quodnon ab eo(rum>* obliquis casibus fit, ut recti simili*facie ostendantur, sed propter eos facilius perspicisimilitudo potest eorum quam vim habeat,* utlucerna in tenebris allata non facit (ut)* quae ibi suntposita similia sint, sed ut videantur, quae suntquoius di sint.'44. Quid similius videtur quam in his est extremalittera crux Phryx^ ?Quas, qui audit voces, auribusdiscernere potest nemo, cum easdem non esse similesex a

ON THE LATIN LANGUAGE, IX.4.S-45from the children whether the parents are aUke :those who say this are mistaken, for it does not comeabout from their oblique cases that the nominativesare shown to be of like appearance, but through theoblique cases can be more easily seen what evidentialforce hes in the Ukeness of the nominatives—even asa lamp in the dark, when brought, does not cause thatthe things which are there should be alike, but thatthey should be seen in their real character.44. What seems more closely alike than the lastletter in the words crux ' cross and Phryx Phrygian' ? " No one who hears the spoken words can' 'by his ears distinguish the letters,* although weknow from the declined forms of the words thatthough alike they are not identical ; because whenthe plurals cruces and Phryges are taken and E isremoved from the last syllables, from the one thereresults crux, with X from C and S, and from the othercomes Phryx, from G and S. And the difference islikewise clear, when S is removed ; for the one becomescruce, the other Phryge^XXXII. 45. As to what they say," that sincelikeness does not exist in the greater part of speech.Regularity does not exist, they speak foolishly in twoways, because Regularity is present in the greaterpart of speech, and even if it should exist only in thesmaller part, still it is there : unless they will say thatwe do not wear any shoes, because on the greaterpart of our body we do not wear any.time, cf. ix. 75-76. ' The usual confusion of letters andsounds. ' Abl. sing. ; the manuscript has forms endingin -»', which are datives, but the removal of s from cruces andPhryges leaves forms ending in «, not in »'.§ 45. « C/ viii. 37.471

VARROXXXIII. 46. Quod dicunt nos dissimilitudinem(potius gratam acceptamque habere quam similitudinem)^:itaque in vestitu in supellectile delectarivarietate, non paribus subuculis uxoris, respondeo, sivarietas iucunditas, magis varium esse in quo aliasunt similia, alia non sunt : itaque sicut abacumargento ornari, ut alia (paria sint, alia)^ disparia, sicorationem.47. Rogant, si similitudo sit sequenda, cur malimushabere lectos alios ex ebore, alios ex testudine, sicitem genere aliquo alio. Ad quae dico non dis(similitudinessolum nos, sed)^ similitudines quoque sequisaepe. Itaque ex eadem supellectili licet videre :nam nemo facit triclinii lectos nisi paris et materia etaltitudine et figura. Qui(s)* facit mappas triclinarisnon similis inter se ? Quis pulvinos ? Quis deniquecetera, quae unius generis sint plura ?48. Cum, inquit,* utilitatis causa introductasit oratio, sequendum non quae habebit similitudinem,sed quae utilitatem. Ego utilitatis causa orationemfactam concedo, sed ut vestimenta : quare ut hiesimilitudines sequimur,* ut virilis tunica sit virilisimilis, item toga togae, sic mulierum stola ut sitstola

ON THE LATIN LANGUAGE, IX. 46-48XXXIII. 46. As to what they say," that we findunlikeness pleasing and acceptable rather thanlikeness, and therefore in clothing and in furniturewe take pleasure in variety, and not in havingour wives' undertunics all identical : I answer,that if variety is pleasure, then there is greatervariety in that in which some things are alike andothers are not ; and just as a side-table is adorned\dth silver in such a way that some ornamentsare alike and others are unlike, so also is speechadorned.47. They ask why, if likeness is to be followed,we prefer to have some couches inlaid with ivory,others with tortoise-shell, and so on with some otherkind of material. To which I say that unlikenessesare not the only thing which we follow, but often wefollow likenesses. And this may be seen from thesame piece of furniture ; for no one makes the threecouches of the dining-room other than alike in materialand in height and in shape. WTio makes the tablenapkinsnot like each other ? Or the cushions ?And finally the other things which are several innumber but of one sort ?48. Since speech, they say," was introduced for thesake of utility, we should follow not that kind ofspeech which has likeness, but that which has utility.I grant that speech has been produced for utility'ssake, but in the same way as garments have : thereforeas in the latter we follow the likenesses, so thata man's tunic is hke a man's, and a toga Uke a toga,and a woman's dress is like a dress regularly and acloak like a cloak, so also, as words that are names§46. " C/. viii. 31-32.§ 48. « C/. viii. 28-29.473

VARROcum sint nomina utilitatis causa, tamen virilia interse similia, item muliebria inter se sequi debemus.XXXIV. 49. Quod aiunt ut persedit et perstititsic (periacuit et>^ percubuit quoniam non sit,^non esse analogian, et^ in hoc e* declinata. Quare non solum in vocabulislitterarum haec non requirenda analogia, sed (ne)*in syllaba quidem ulla, quod dicimus hoc BA, huiusBA, sic alia.§ 49. ^ Added by Canal. * Kent, for sit. ' Aug.,for ut. * B, Rhol., for erant.§50. ^Aug., for OYiWa.. ^ A%ig.,forho\\\\a.. ^ Addedby Stephanas.§ 51. ^ B, G, H, a, Aug., for sequirunt. ^ L. Sp., foreo F^, ea F^. ' L. Sp. ; esset M, a, Aug. ; for esse.* Added by Aldus.§ 49. " Referring to a passage now lost. * The twoverbs are not attested in any form.§ 50. « Cf. viii. 54 and 80.474

ON THE LATIN LANGUAGE, IX. 48-51of persons exist for the purpose of utility, we oughtstill to employ men's names that are like one another,and women's names that also have mutual resemblances.XXXI\'. 49. As to the fact that they say " thatRegularity does not exist because there are noperfects periacuit remained ' lying ' and percuhuit'remained lying,' like persedit ' remained sitting 'and perstitit remained standing,' in this also they'are mistaken : for the two perfects have no presents *from which to be inflected, whereas Regularitypromises only that from two like words inflected inlike manner there \\ill be like forms.XXXV. 50. Those who say " that there are noRegularities because from Romulus there is Romaand not Romula and there is no hovilia cow-stables ''from bos cow ' ' as there is ovilia sheepfolds from' 'avis ' sheep,' are in error ; because nobody professesthat one word is derived from another word, fromnominative singular to nominative singular, but onlythat from two like words like case-forms developwhen they are inflected in like manner.XXXVI. 51. They say " that because the wordsdenoting the Latin letters are not inflected intocase-forms the Regularities do not exist. Suchpersons are demanding the declension of those wordswhich by nature cannot be inflected ;just as ifRegularity were not said ^ to belong merely to thoseforms which had already been inflected in like fashionfrom like words. Therefore not only in the names ofthe letters must this kind of Regularity not be sought,but not even in any syllable, because we say nominativeba, genitive ba, and so on.§ 51. " C/. viii. 64. " Cf. viii. 23.475

VARRO52. Quod si quis in hoc quoque velit dicere esseanalogias rerum, tenere potest :ut eni(m>^ dicuntipsi alia nomina, quod quinque habeant figuras,habere quinque casus, aha quattuor, sic minus alia,dicere poterunt esse litteras ac syllabas in voce quaesingulos habeant casus, in rebus pluris* ;quemadmoduminter se conferent ea quae quaternos habebuntvocabulis casus, item ea inter se qua(e> ternos,^sic quae* singulos habebunt, ut conferant inter sedicentes, ut sit hoc A, huic A, esse hoc E,* huie E.XXXVII. 53. Quod dicunt esse quaedam verbaquae habeant declinatus, ut caput (capitis, nihilnihili),^ quorum par reperiri quod non possit, non esseanalogias, respondendum sine dubio, si quod estsingulare verbum, id non habere analogias : minimumduo esse debent verba, in quibus sit similitudo.Quare in hoc tollunt esse analogias.54'. Sed mkilum^ vocabulum recto casu apparet inhoc :Quae dedit ipsa.^ cap/t' neque dispendi facit hilum,§ 52. ^ For eni. * GS, ; plureis Canal ; for plurimis.^ Koeler, for quaternos. * For sicque. * After hoc E,L. Sp. deleted huiusce E.§ 53. ^ Added by Reitzenstein.§ 54. ^ Lachmann ; in nihil Sciop. ; for initium.^ Sciop., for ira. ^ Seal iger, for caput.§ 52. o Cf. viii. 63. ^ That is, words indeclinable inform have only one case-form, but still have all the case-uses.§ 53. " There is no corresponding passage in Book VIII.* That is, when they select a unique word as basis for argument.476

ON THE LATIN LANGUAGE, IX. 52-5452. But if any one should wish to say that in thisalso there are Regularities in the things, he canmaintain it. For as they themselves say

VARROquod valet nee dispendii facit quicquam.Idem hocobliquo apud Plautum :Video enim* te nihili* pendere prae Philolacho* omnishomines,quod est ex ne et hili : quare dictus est nihili* qui nonhili erat. Casus tantum'' commutantur de quo dicitur,* de homine : dicimus enim hie homonihili' et huius hominis nihili et hunc hominemnihili. Si in illo commutaremus, diceremus ut hoclinum et Uium,^" sic nihilum, non hienihiU, et "huic hno et ]ibo^^, sic nihilo, non huic nihiU. Potestdici patricus casus, ut ei praeponantur*' nomina"plura, ut hie casus Terentii, hunc casum Terentii,hie miles legionis, huius militis legionis, hunc militemlegionis.XXXVIII. 55. Negant, cum omnis natura sit autmas aut femina aut neutrum, (non>i debuisse exsingulis vocibus ternas figuras vocabulorum fieri,ut albus alba album ; nunc fieri in multis rebusbinas, ut Metellus Metella,* Aemi{\i}us Aemi(\i}a,^nonnulla singula, ut tragoed?

ON THE LATIN LANGUAGE, IX. 5^55which is the same as ' nor of loss does she sufferanything.' This same word is found in an obliquecase in Plautus * :I see, beside Philolaches you count all men as nothing.The word is from ne not and genitive ' ' hilt whit ' ' ;therefore he has been called nihili ' of naught who'was not hili of a whit ' 'in value. Change is madeonly in the case-forms of that about which the speakingis done, as about a man ; for we say a man nihili'of no account ' in nominative, in genitive, in accusative,changing the forms of homo but not changingthe form nihili. If we were to make changes in it,then we should say not hie nihili " but nihilum as thenominative, like linum ' flax and ' libum ' cake,' anddative not huic nihili but ** nihilo like lino and libo.The genitive case * can however be said with variousnouns set before it, like nominative casus mishap ' 'Terentii of Terence,' 'accusative casum Terentii, andnominative miles ' soldier ' legionis of the ' legion,'genitive militis legionis, accusative miliiem legionis.XXXVIII. 55. They say " that since everynature is either male or female or neuter, from theindividual spoken words there should not fail to beforms of the words in sets of three, like albus, alba,album white ' ' ; that now in many things there areonly two, like Metellus and Metella, Aemilius andAemilia, and some with only one, like tragoedus'tragic actor ' and comoedus ' comic actor ' ; thatthere are the names Marcus and Numerius, but no" Plautus, Most. 245. « The genitive nihili depending ona nominative. * The genitive nihili depending on adative. « Such as the form nihili.§ 55, ' C/. viii. 47.479

VARROnon esse ; dici con;um,^ turdum, non* dici corwam,*turdam ; contra dici pantheram, merulam, non dicipantherum, merulum ; nullius no*tn

ON THE LATIN LANGUAGE, IX. 55-57Marca and Xumeria * '; that corviis raven and'turdus thrush ' ' are said, but the feminines corva andturda are not said : that on the other hand panthera'panther and merula blackbird ' ' ' are used, but themasculines pantherus and merulus are not ; that thereis no one of us whose son and daughter are not suitablydistinguished as male and female, as Terentiusand Terentia ; that on the other hand the childrenof gods and slaves are not distinguished in the sameway,'^ as by lovis and lova for the son and the daughterof Jupiter ; that likewise a great number of commonnouns do not in this respect preserve the Regularities.5Q. To this we say that although the object isbasic " for the character of all speech, the words donot succeed in reaching the object if it has not comeinto our use ; therefore equus ' stallion and equa'mare 'are said, but not corva beside corvus, becausein that case the factor of unlike nature is without useto us. But for this reason some things were formerlynamed otherwise than they are now : for thenall doves, male and female, were called columhae,because they were not in that domestic use in whichthey are now, and now, on the other hand, because wehave come to make a distinction on account of theiruses as domestic fowl, the male is called columhusand the female columba.57. When the nature goes through the threegenders and this distinction is made in use, then finallyit is seen, as it is in doctus learned man ' 'and docta' 'learned woman and docium learned thing ' ' ; forlearning can go across through these three, and usehas taught us to differentiate a learned thing fromhuman beings, and among the latter to distinguishthe male and the female. But in a male or a femaleVOL. II I 481

VARROtransit neque feminae neque neutra, et ideo nondicitur feminus femina feminum, sic reliqua : itaquesingularibus ac secretis vocabulis appellati sunt.58. Quare in quibus rebus non subest similisnatura aut usus, in his vocabulis huiusce modi ratioquaeri non debet : ergo dicitur ut surdus vir, surdamulier, sic surdum theatrum, quod onmes tres

ON THE LATIN LANGUAGE, IX. 57-59or what is neither, the nature of the male does notshift, nor that of the female, nor the neuter nature,and for this reason there is no saying of feminns,femina, femimim, and so with the rest. Thereforethey are called by special and separate words.58. Wherefore in the names of those things inwhich there is no likeness of nature or of use as thebasis, a relation of this sort ought not to be sought.Accordingly, as a surdus deaf ' ' man is a currentterm, and a surda woman, so also is a surdum theatre,"because all three things are equally intended for theact of hearing. On the other hand, nobody says asurdum sleeping-room, because it is intended forsilence and not for hearing ; but if it has no window,it is called caecum ' blind,' as a man is called caecusand a woman caeca, because not all sleeping-roomshave the light which they ought to have.''59. The male and the female have by nature acertain association with each other ; but the neutershave no association with them, because they aredifferent from them in kind, and even of these neutersthere are very few which have any elements incommon with other neuters. As for the fact that thenames of a god and of a slave do not vary like ourfree names, there is the same reason, namely thatthe variation is connected with use, and had to beestablished with reference to free persons, but as tothe rest had no consequence, because among slavesthe clan quality has no foundation in practice, butit is necessary in the names of us who are in Latiumand are free. Therefore in that class Regularitymakes its appearance, and we say Terentius for aman, Terentia for a woman, and Terentium for thegenus 'stock.'483

VARRO60. In praenominibus ideo non fit item, quod haecinstituta ad usum singularia, quibus discernerenturnomina gentilicia, ut ab numero Secunda, Tertia,Quarta (in mulieribus),^ in viris ut Quintus, Sextus,Decimus, sic ab aliis rebus. Cum essent duoTerentii aut plures, discernendi causa, ut aliquidsingulare haberent, notabant, forsitan ab eo, quimane natus diceretur, ut is Manius esset, qui luci,Lucius,* qui post patris mortem, Postumus.61 E quibus

ON THE LATIN LANGUAGE, IX. 60-6260. In first names the situation is not the same,because these were in practice estabhshed as individualnames, by which the clan names might bedifferentiated ; from the numerals came Secunda,Tertia, Quarta for women, Quintus, Sextus, Decimusfor men, and similarly other names from other things.When there were two or more persons of the nameTerentius, then that they might have somethingindividual to distinguish them they marked themperhaps in this way," that he should be Manius whowas said to have been born mane 'in the morning,'and he who has been born luci ' 'at dawn should beLucius, and he who was born post after ' ' his father'sdeath should be Postumus.*61. When any of these things happened tofemales as well, they derived the first names ofwomen regularly in this manner—that is, in formertimes—and called them by them, for example.Mania, Lucia, Postuma : for we see that the motherof the Lares is called Mania,^ that Lucia Volumnia *is addressed in the Hymns of the Salians,' and thateven now many give the name Postuma to a daughterborn after the death of her father.62. Therefore as far as the nature and the use ofa word have jointly advanced, so far has Regularitybeen extended in like manner by a correspondingrelationship, since of the words in which there arevoluntary- inflections of male and female and neuter,those which are voluntary in inflection ought not to beinflected in similar manner, but in those in whichthere are natural inflections there are those regularnot found elsewhere ; several members of the gens Volumniaare mentioned at Rome during Varro's time. « Frag. 5,page 336 Maurenbrecher ; page 4 Morel.485

VARROclinatus hi qui esse reperiuntur.Quocirca in tribusgeneribus nominum in(i>que' tollunt analogias.XXXIX. 63. Qui autem eas reprehendunt, quodalia vocabula singularia sint solum, ut eicer, alia multitudinissolum, ut scalae, cum debuerint omnia esseduplicia, ut equus equi, analogiae fundamentum esseobliviscuntur naturam et usu(m).i Singulare estquod natura unum significat, ut equus, aut quodconiuncta quodammodo ad unum usu,'' ut bigae :itaque ' dicimus una Musa, sic dicimus unaebigae.64. Multitudinis vocabula sunt unum infinitum,ut Musae, alterum finitum, ut duae, tres, quattuor :dicimus enim ut hae Musae sic unae bigae et binaeet trinae bigae, sic deinceps. Quare tam unae et uniet una quodammodo singularia sunt quam unus et unaet unum ; hoc modo mutat, quod altera in singularibus,altera in coniunctis rebus ;multitudinis, sic bina trina.et ut duo tria sunt65. Est tertium quoque genus singulare ut inmultitudine, uter, in quo multitudinis ut utrei^ ; uter^ Aldus, for inquae.§63. ^p, Mue., for usu. ^ A. Sp., for usum.3 Added by L. Sp.§65.^ A. Sp., for utre l-.§ 62. " Crates and his followers, who uphold Anomaly.§ 63. " Cf. viii. 48. " Cf. x. 54.§ 64. " The first is the generic or collective, without specificationof the number or of the individuals ; the second isnumerical, in which the number of the individuals is given ortheir identity is clearly implied. "" A word like bigae,486

ON THE LATIN LANGUAGE, IX. 62-65inflections which are actually found to exist. Thereforein the matter of the three genders they " areunfair in setting aside the Regularities.XXXIX. 63. Moreover those who find fault "with the Regularities, because some words aresingulars only, like deer ' chickpea,' and others areplural only, like scalae ' stairs,' * although all oughtto have the two forms, like equus horse ' ' and equi'horses,' forget that the foundation of Regularityis nature and use taken in combination. That issingular which by nature denotes one thing, likeequus ' horse,' or which denotes things that by useare joined together in some way, like bigae two-horse''team.' Therefore just as we say una Musa oneMuse,' we say unae bigae one two-horse team.''64. Plural words are of two sorts," the one indefinite,like Musae Muses,' the other definite, like'duae ' two,' tres ' three,' quattuor four ' ' ; for as wesay Musae in the plural, so also we say unae bigae one'two-horse team,' and binae two ' ' and trinae * bigae'three two-horse teams,' and so on. Whereforeunae and the masc. uni and the neut. una are ina certain manner as much singulars as unus and unaand unutn : the word changes in this way becausethe one set of forms is said of individual things, theother of things joined together in sets ; and just asduo and iria are plurals, so also are bina and trina.65. There is also a third class which is singularthough expressed by a plural form, namely uter'which of two,' in which the plural form is for exalreadyplural in form, can be pluralized in meaning only bythe use of a numerical modifier ; for this purpose, distributivenumerals such as bini are used. For the singular idea,the plural form of unus is used.487

VARROpoeta singulari, utri poetae multitudinis est. Quaexplicata natura apparet non debere omnia vocabulamultitudinis habere par singulare : omnes enimnumeri ab duobus susum versus multitudinis suntneque eorum quisquam habere potest singularecompar. Iniuria igitur postulant, si qua sint singularia,oportere habere multitudinis.XL. 66. Item qui reprehendunt, quod non dicaturut unguentum unguenta vinum vina sic acetum acetagarum gara, faciunt imperite : qui ibi desiderantmultitudinis vocabulum, quae sub mensuram ae ponderapotius quam sub numerum succedunt : nam inplumbo,^ at^ plura et ea inusum venerunt, a genere multo, sic vina et unguenta,dicta : alii generis enim vinum quod Chio, aliurf^§ 66. ^ After pliimbo, L. Sp. deleted oleo. * Aug., foraceto. ^ After dicimus, Aldus deleted enim. * Aftermultum, L. Sp. deleted oleum. * After non, L. Sp. deletedmulta olea. * Aug., with B,for argentum.§ 67. ^ Laetus, for habet. ^ For aliut.§ 65. " The old spelling of the nominative plural, stillmore or less in use in Varro's time, though rarely attested inthe manuscripts.§ 66. °Cf § 67. * Derivative adjectives, ' made oflead ' and ' made of silver ' ; supply vasa ' utensils.'488

ON THE LATIN LANGUAGE, IX. 65-67ample vtrei " : titer poeta which of two poets ' ' in thesingular, utri poefae which of two ' sets of poets ' inthe plural. Now that the nature of this has beenexplained it is clear that plural nouns are not allunder obligations to have a like singular form ; forall the numerals from two upwards are plural, andno one of them can have a singular to match it.Therefore it is quite wrongly that they demand thatall singulars that there are, must have a correspondingplural form.XL. 66. LikeA\ise those who find fault becausethere are no plurals aceta and gara to acetum vinegar ''and garum fish-sauce ' ' like unguenia to unguentum'perfume and ' vina to vinum wine,' ' " act ignorantly ;they are looking for a plural name in connexionwith things which come under the categories ofquantity and weight rather than under that ofnumber. For in plumbum lead and argentum ' ' ' silver,'when there has been added an increase, we saymultum much '': thus multum plumbum or argentum,not plumba leads and ' ' argenta ' silvers,' since articlesmade of these we call plumbea and argentea * (silveris something else when it is argenteum, for that iswhat it is when it has now become a utensil ; thusargenteum if it is a small cup or the like), because inthis case we speak of many argentea ' silver ' cups,and not of much argentum 'silver.'67. But if those things which have by nature theidea of quantity rather than that of number, exist inseveral kinds and these kinds have come into use,then from the plurality of kinds they are spoken ofin the plural, as for example vina ' \\ines ' and unguenta' perfumes.' For there is wine of one kind,which comes from Chios, another wine which is from489

VARROquod Lesbo,' sic ex regionibus aliis. (Ae>que* ipsadicuntur nunc melius unguenta/ cui nunc generaaliquot. Si item discrimina magna essent olei etaceti et sic ceterarum rerum eiusmodi in usu co(m>muni,dicerentur sic olea et (aceta ut>* vina. Quarein utraque re nique' rescindere conantur analogias,et' cum in dissimili usu similia vocabula quaeront* etcum item ea quae metimur atque ea quae numeramusdici putent oportere.XLI. 68. Item reprehendunt analogias, quoddicantur multitudinis nomine publicae balneae, nonbalnea, contra quod privati dicant unum balneum,quo/7«^ plura balnea * dicant. Quibus respondert'potest non esse reprehendendum, quod scalaeet aquae caldae, pleraque* cum causa, multitudinisvocabulis sint appellata neque eorum singularia inusum venerint ; idemque item contra. Primumbalneum (nomen e(s)t* Graecum), (cum)* introiit inurbem, publice ibi consedit, ubi bina essent coniunctaaedificia lavandi causa, unum ubi viri, alterumubi mulieres lavarentur ; ab eadem ratione domisuae quisque ubi lavatur balneum dixerunt et, quodnon erant duo, balnea dicere non consuerunt, cum^ T^, p, Aldus, for Lesbio. * A. Sp., for quae. * Forunguentia. * Added by L. Sp.''Canal, for denique.* Aug., for analogiam set. * L. Sp., for querunt.§68.^ Cana/, /or quod. ^ Added by Popma. ^ Aldus,for respondere, * After pleraque, L. Sp. deleted quae.« GS., for et. * Added by GS.§ 68. " The word is a heteroclite in form, with a different490

ON THE LATIN LANGUAGE, IX. 67-68Lesbos, and so on from other localities. Likewiseunguenta perfumes themselves are now properly' 'spoken of in the plural, for of perfume there are nowa number of kinds. If in like fashion there were greatdifferences in olive-oil and vinegar and the otherarticles of this sort, in common use, then we shouldemploy the plurals olea and aceta, like vina. Thereforein both these matters their attempt to destroythe Regularities is unfair, since they expect that thewords will be alike though their uses are different,and since they think that articles which we measureand objects which we count should be spoken of in thesame way.XLI. 68. Likewise they find fault with the Regularities,because public baths are spoken of as balneae,with the form in the plural, and not as balnea, in thesingular ; and on the other hand they speak of one balneumof a private individual, though they do not usethe plural balnea.'^ To them answer can be made, thatfault ought not to be found because scalae 'stairsand aquae caldae hot springs,' mostly with good'reason, have been called by plural names and thecorresponding singulars have not come into use : andvice versa.'' The first balneum bath-room (the' 'name is Greek), when it was brought into the city ofRome, was as a public establishment set in a placewhere two connected buildings might be used forthe bathing, in one of which the men should batheand in the other the women. From the same logicalreasoning each person called the place in his ownhouse where baths were taken, a balneum ; and theywere not accustomed to speak of balnea in the plural,meaning in the two numbers. But the plural balnea began tobe used in the time of Augustus. * Cf. § 69.491

VARROhoc antiqui non balneum, sed lavatrinam' appellareconsues^ent.*69- Sic aquae caldae ab loco et aqua, quae ibiscateret, cum ut colerentur venissent in usum nostris,cum aliae ad alium morbum idoneae essent, eae cumplures essent, ut Puteolis et in Tuscis, quibus utebantur,multitudinis potius quam singular! vocabuloappellarunt. Sic scalas, quod ab scandendo dicunturet singulos gradus scanderent, magis erat quaerendum,si appellassent singular! vocabulo scalam, cumorigo nominatus ostenderet contra.XLII. 70. Item reprehendunt de casibus, quodquidam nominatus habent rectos, quidam obliquos,quod dicunt utrosque in vocibus oportere. Quibusidem responderi potest, in quibus usus aut naturanon subsit, ibi non esse analogiam. . . .71. Sed ne in his (quidem)^ vocabulis quaedeclinantur, si transeunt e recto casu in rectumcasum : quae tamen fere non discedunt ab rationesine iusta causa, ut hi qui gladiatores Faustinj^ :nam quod plerique dicuntur, ut tris extremas syllabas' Aug., with B, for lauiatrinam. * B, Ed. Veneta, forconsuescent.§ 71. ^ Added here by L. Sp. ; added after vocabulis byMadvig. ^ Mue., for faustinos." More commonly in the contracted form latrina, and inVarro's time meaning ' water-closet, privy.'§ 69. " At least nine places in Etruria bore the nameAqvae.§ 70. " Cf. viii. 49. * There seems to be a lacuna here,as examples -illustrating this point of the refutation are lacking.§71. ° That is, by derivation with suffixes, not merely by492

ON THE LATIN LANGUAGE, IX. 68-71because they did not have two in one house—thoughour forbears were accustomed to call this not abalneum, but a lavatrina wash-room.''^ '69. So also, the hot springs, on account of thelocality and the water which gushed out there, cameto be frequented for our use, since some of thesprings were beneficial to one disease and others toanother ; and because those which they used wereseveral in number, as at Puteoli and in Etruria,"they called them by a plural word rather than by asingular. So also with the scalae ' stairs ' ; becausethey are named from scandere ' to mount and there'were separate steps to be mounted, it would be amore difficult problem to answer if they had calledthem scala, in the singular, inasmuch as the origin ofthe name shows their plural nature.XLII. 70. Likewise they find fault " about thecases, because some nouns have nominative formsonly, and others have only oblique forms : whereuponthey say that all words ought to have both thenominative and the oblique forms. To them thesame answer can be given, that there is no Regularityin those instances which lack a relationship in useor in nature. . ..''71. But they should not look for complete Regularityeven in these names which are derived bypassage from one nominative form to another."Still, such words do not in general depart from thepath of logic without valid reason, such as there is forthose gladiators who are called Faustini * ; for thoughmost gladiators are spoken of in such a way that theycase-inflection. * The troops of gladiators were designatedby adjectives of this sort which were derived from the namesof the owners.493

VARROhabeant easdem, Cascelliani, (Caeciliani),^ Aquiliani,animadvertant,* unde oriuntur, nomina dissimiliaCascellius,^ Caecilius, Aquilius, (Faustus : quod siesset)' Faustius, recte dicerent Faustianos ;si(c>'a Scipione quidam male dicunt Scipioninos : nam estScipionarios. Sed, ut dixi, quod ab huiuscemodicognominibus raro declinantur cognomina neque inusum etiam perducta, natant quaedam.XLIII. 72. Item dicunt, cum sit simile stultusluscus et dicatur stultus stultior stultissimus, nondici luscus luscior luscissimus, sic in hoc generemulta. Ad quae dico ideo fieri, quod natura nemolusco magis sit luscus, cum stultior fieri videatur.XLI V, 73. Quod rogant, cur

ox THE LATIN LANGUAGE, IX. 71-73have the last three syllables alike, CascelUani, Caeciliani,Aquiliani,^ let them take note that the namesfrom which these come, Cascellius, Caecilius, Aquiliuson the one hand, and Faustus on the other, are unhke :if the name were Faustius, they would be right insaying Faustiani. In the same way, from Scipiosome make the bad formation Scipionini ; it is properlyScipionarii. But, as I have said, since appellationsare rarely derived from surnames of this kindand they are not fully at home in use, some suchformations fluctuate in form.XLIII. 72. Likewise they say," that althoughstultus stupid and ' ' luscus one-eyed ' ' are like words,and stultus is compared with stitliior and stultissimus,the forms luscior and luscissimus are not used withIttsais, and similarly with many words of this class.To which I say that this happens for the reasonthat by nature no one is more one-eyed than a oneeyedman, whereas he may seem to become morestupid.XLH'. 73. To their question why we do ** not saymane ' in the morning,* comparative manius, superlativemanissime, with a similar question aboutvesperi ' in the evening,' I reply that in matters oftime there is properly no more and ' ' ' less,' butthere can be before and after. Therefore the firsthour is earUer than the second, but not more 'hour.'But nevertheless to rise magis mane more ' in themorning ' is an expression in use ; he who rises inthe first part of the morning rises magis mane *'more in the morning ' than he who does not risein that first part. For as the day cannot be saidto be more than day, so mane cannot be said to bemore than mane.'^ Therefore that very magis ' more '495

VARROitaque ipsum hoc quod dicitur magis sibi non constat,quod magis mane significat primum mane, magisvespere novissimum vesper.XLV. 74. Item ab huiuseemodi similitudinibus^reprehenditur analogia, quod cum sit anuscadus simile et sit ab anu anicula anicilla, a cado duoreliqua quod non sint propagata, sic non dicatur apiscina piscinula piscinilla. Ad (haec respondeo)*huiuseemodi vocabulu* analogias esse, ut dixi, ubimagnitudo animadvertenda sit in unoquoque gradueaque* sit in usu comuni, ut est cista cistulacistella et canis catulus catellus, quod in pecoris usunon est. Itaque consuetude frequentius res in binasdividi partis ut maius et minus, ut lectus et lectulus,area et arcula, sic aha.XLVI. 75. Quod dicunt casus alia non habererectos, alia obliquos et ideo non esse analogias, falsumest. Negant habere rectos ut in hoc frugis frugifrugem, item cole

ON THE LATIN LANGUAGE, IX. 73-75which is commonly said is not consistent \^ith itself,because magis mane means the first part of the mane,and magis vespere the last part of the evening.XLV. 74. Similarly, Regularity is found faultwith on account of unhkenesses of this sort," thatalthough anus ' old M'oman and cadus cask ' ' ' arelike words, and from anus there are the diminutivesanicula and anicilla, the other two are not formedfrom cadus, nor from piscina fish-pond are ' ' piscinulaand piscinilla made. To this I answer that words ofthis kind have the Regularities, as I have said, onlywhen the size must be noted in each separate stage,and this is in common use, as is cista ' box,' cistula,cistella, and canis *'dog,' catulus puppy,' ' catelluslittle puppy ' ; this is not indicated in the usage'connected ^^ith flocks.*^ Therefore the usage is moreoften that things be dixided into two sets, as larger ^and smaller, like lectus couch ' ' and lectulus, area'strong-box and ' arcula, and other such words.XL\T. 75. As to their saying " that some wordslack the nominative and others lack the obliquecases, and that therefore the Regularities do notexist, this is an error. For they say that the nominativeis lacking in such words as, frugis frugi frugem *'fruit of the earth ' and colem colis cole '^ ' plantstalk,'and the oblique cases are lacking in such asDiespiter 'Jupiter,' dat. Diespitri, ace. Diespitrem, andMars,' Maspitri, Maspitrem.^Master 'way of rearranging the order of the text. * Gen., dat., ace.' Ace, gen,, abl., unless the manuscript readings are to bemore seriously altered ; the word is more properly caul-, butCato and Varro prefer the country forms, with o from au.* For Dies pater and Mars pater ; the addition of pater isfound only in nom. and voc. {luppiter, older lupiter, is avoc. form).VOL. II K 4-97

VARRO76. Ad haec respondeo et priora habere nominandiet posteriora obliques. Nam et frugi rectus estnatura frux, at secundum consuetudinem dicimus uthaec avis, haec ovis, sic haec frugis ; sic secundumnaturam nominandi est casus coZs,^ secundum consuetudinemcolis,* cum utrumque conveniat ad analogian,quod et id quod in consuetudine non est cuiusmodi debeat esse apparet, et quod est in consuetudinenunc in recto casu, eadem est analogia ac pleraque,quae ex multitudine cum transeunt in singulare,Sic cum transiretur ex eodifficulter efFeruntur ore.quod dicebatur haec oves, una non est dicta ovs sine/,' sed additum I ac factum ambiguum verbumnominandi an patrici esse* casus. Ut ovis, et avis.77. Sic in obliquis casibus cur negent esseDiespitri Diespitrem non video, nisi quod minus esttritum in consuetudine quam Diespiter ;quod innihil argumentum est : nam tam casus qui non tritusest quam qui est. Sed est(o)^ in casuum serie aliavocabula non habere nominandi, alia de obliquisaliquem: nihil enim ideo quo minus siet* ratio percellerepoterit hoc crimen.§ 76. ^ Mue.y for rois. * Mue., for rolis. ' L. Sp.,for una. * L. Sp.,for esse.§ 77. 1 L. Sp., for est. * Mue., for si et ; on the possibilityof the use of siet in Varro's tirne, cf. Cicero, Orator47. 157.§ 76. " Frux is found in Ennius, Ann. 314 (' honest man ')and 431 Yahlen^ = R.O.L. i. 116-117 and 150-151 Warmington; but nom. frugis is not quotable from a text. * Colismay be cited from Lucilius, 135 Marx, and Varro, R. R.i. 41. 6. ''Varro is speaking on the basis that therelation is nom. sing, ending in -s, nom. pi. in -es, as indux, pi. duces.•*Haec before oves is the sign of the nom.pi. fem. ; Varro appears to use hae before consonants, haec498

ON THE LATIN LANGUAGE, IX. 76-7776. To this I answer that the former have nominativesand the latter have oblique case-forms. Forthe nominative oifrugi is by natxirefrux, but by usagewe say J'ntgis," like avis bird and ' ' ovis sheep ' ' ; soalso, the nominative of the other word is by naturecols and by usage colis.^ Both of these agree AWth theprinciple of Regularity, because it is perfectly clearof what sort that form ought to be which is not in use,and in that which is now in use in the nominativethere is the same kind of Regularity as most wordshave that are hard to pronounce when they passfrom the plural to the singular.*^ So when thepassage was made from the spoken plural otes,^ theform which was pronounced was not ovs \\ithout I,but an I was added and the word became ambiguousas to whether the case was nominative or genitive.*Like the nominative oris is also the nominative avis.77. Thus I do not see why they say that in theoblique cases Diespitn and Diespitrem are lacking,except because they are less common in use thanDiespiter. But the argument amounts to nothing ;for the case-form which is uncommon is just as mucha case-form as that which is common. But let usgrant that in the list of case-forms some words lackthe nominative and others lack some one of theoblique cases ; for this charge \^-ill not for that reasonbe able in any way to destroy the existence of a logicalrelationship " among the forms.before vowels as here (and at the sentence-end, as at v. 75).' Varro is of course unaware of the fact that some nouns ofthe third declension had stems ending in i and therefore hada right to nominatives in /*, while others had stems ending inconsonants and could have the ending w only by analogywith the i-stems.§ 77. • That is. Regularity.499

VARRO78. Nam ut signa quae non habent caput^ autaliquam aliam partem, nihilo minus* in reliquis membriseorum esse possunt analogiae, sic in vocabuliscasuum possunt item fieri (iacturae. Potest etiamrefingi)' ac reponi quod aberit, ubi patietur naturaet consuetudo : quod nonnunquam apud poetasinvenimus factum, ut in hoc apud Naevium in Clastidio:Vita insepulta laetus in patriam redux.XLVII. 79. Item reprehendunt, quod dicaturhaecstrues, hie Hercules,^ hie homo : debuisset enim dici,si esset analogia, hie Hercul, haec strus, hie homon* haec ostendunt no(mi)wa' non analogian esse,sed obliquos casus non habere caput ex sua analogia.Non, ut si in Alexandri statua imposueris caputPhilippi, membra conveniant ad rationem, sic* etAlexandri membrorum simulacro* caput quod respondeatitem sit ? Non, si quis tunicam in usu itaconsuit, ut altera plagula* sit angustis clavis, alteralatis, utraque pars in suo genere caret analogia.XLVIII. 80. Item negant esse analogias, quod§ 78. ^ After caput, M and Laetus deleted et. ^ Fornihil hominus. * Added by GS. ; but the lost part may besomewhat longer.§79. ^ p,Ijaetus, for Herculis. ^ GS. ; homen Canal ;for homon. ' Kent, for noua. * G, H, Aug., for sit.* A. Sp., for simulacrum. • Aldus, for placula.§ 78. " By regular formation. ^ Trag. Rom. Frag.,Praet. II Ribbeck^. " Redux, not elsewhere found in thenom. sing.§ 79. " If the nominatives were of the usual types, whichreplace the genitive ending -IS by -S or by nothing at all,like sus, animal, nomen, genitives suls, animalis, nom,inis.* That is, the nominatives are not formed regularly from' 'the oblique cases, but from these nominatives of variant types500

ON THE LATIN LANGUAGE, IX. 78-8078. For as some statues lack the head or someother part without destroying the Regularities intheir other limbs, so in words certain losses of casescan take place, with as Uttle result. Besides, what islacking can be remade " and put back into its place,where nature and usage permit ; which we sometimesfind done by the poets, as in this verse of Naevius, inthe Clastidium * :With life unburied, glad, to fatherland restored.*XL^'II. 79. Likewise they find fault with thenominatives strues heap,' Hercules, homo man ' ''for if Regularity actually existed, they say, theseforms should have been strus, Hercul, hotnen.^ Thesenouns do not show that Regularity is non-existent,but that the oblique cases do not have a head orstarting-point according to their type of Regularity.''Is it not a fact that, if you should put a head ofPhilip on a statue of Alexander and the limbs shouldbe proportionately symmetrical, then the headwhich does correspond to the statue of Alexander'slimbs " would likewise be symmetrical ? And it isnot a fact that if one should in practice sew togethera tunic in such a way that one breadth of the clothhas narrow border-stripes and the other has broadstripes, each part lacks regular conformity within itsown class. **XLVIII. 80. Likewise they say that the Regutheoblique cases are formed regularly. " That is, theheads or nominatives may be varied, but the limbs or oblique**cases are of uniform type. For there are tunics with thebroad stripe, worn by senators, and tunics with the narrowstripe, worn by knights ; therefore, though the two halves inthe example do not belong together, each has its regularprecedent.;501

VARROalii dicunt cupressus, alii cupressi, item de ficisplatanis et plerisque arboribus, de quibus alii extremumUS, alii EI faciunt. Id est falsum : namdebent dici E et I, fici ut nummi, quod est ut nummi(s>fici(s>,^ ut nummorum ficorum. Si essentplures ficus, essent ut manus ;diceremus ut manibus,sic ficibus, et ut manuum, sic ficuum, neque has ficosdiceremus, sed ficus, ut non manos appellamus, sed(manus, nee)* consuetude diceret singularis obliquoscasus huius fici neque hac fico, ut non dici' huiusmani,* sed huius manus, ec^ hac mano, sed hacmanu.XLIX. 81. Etiam illud putant esse causae, curnon sit^ analogia, quod Lucilius scribit :Sive decusibus est.Decu(s>is,*Qui errant, quod Lucilius non debuit dubitare, quodutrumque : nam in acre usque ab asse ad centussisnumerus aes significat, et eius numero finiti casusomnes' ab dupondio sunt, quod dicitur a multisduobus modis hie dupondius et hoc dupondium, ut§ 80. ^ L. Sp., for nummi fici. * Added by Mue. ;manus neque L. Sp. ' Aug., for dici. * M, Laetus, formanui. ^ Jj. Sp., for et.§81. ^ After sit, Aldus deleted in. * Lachmanndecussi Mue. ; for decuis. ' For omnis.§ 80. " As belonging to the fourth and the second declensionsrespectively. * This shows that Varro wrote thenominative plural of the second declension with EI, and notwith I ; but it would be pedantic to substitute such spellingsthroughout his worlcs, or even merely in this section." As type of the second declension. As type of the**fourth declension.502

ON THE LATIN LANGUAGE, IX. 80-81larities do not exist, because some say cupressus'cypress-trees ' in the plural and others say cupressi,"and similarly with fig-trees, plane-trees, and mostother trees, to which some give the ending US andothers give EI. This is wrong ; for the tree-namesought to be spoken with E and \,^ Jici like nummi "sesterces,' because the ablative hjicis like nuinmis,'and the genitive is jicorum like mimmorum. If theplural were jicus, then it would be like manus ^'hand ' ; we should say ablative ^«6«* like manibus,and genitive ^CMM/« like manmim, and we should notsay accusative ^co*, but^cw*, just as we do not sayaccusative manos but manus ; nor would usage speakthe oblique cases of the singular genitive Jici andablative^co, just as it does not say genitive mani butmanus, nor ablative mano but manu.XLIX. 81. Moreover, they think that there isproof of the non-existence of Regularity, in the factthat Lucilius writes " :Priced a ten-

VARRO•hoc gladium et hie gladius ;ab tressibus viriUa multitudinishi tresses et " his tressibus confido," singulare" hoc tressis habeo " et " hoc tres(s>is* confido," sicdeinceps a

ON THE LATIN LANGUAGE, IX. 81-83gladium.^ From tressis ' three-a* ' there is a masculineplural, tresses in the nominative and tressibus inthe ablative, as in " I trust in these three asses,"singular tressis as in " I have this three-a* " and " Itrust in this three-o^y." The same usage is followedall the way to centussis hundred-o*.' ' * From here on,the numeral does not denote money any more thanother things.82. The numerals which do not signify money,from quattuor four to centum hundred,' have forms' ' 'of triple function," because quattuor is masculine,feminine, and neuter. \\Tien mille thousand ' ' isreached, it takes on a fourth function,* that of asingular neuter, because the expression in use istnille thousand of ' ' denarii, from which '^ is made aplural, tytilia thousands ' 'of denarii.83. Since therefore so far as concerns the Regularitiesit is not essential that all words that arespoken should be alike in their systems, but only thatthey should be inflected alike each in its own class,those persons are stupid who ask why as and dupondiusand tressis are not spoken according to a regularscheme ; for the as is a single unit, the dupondius is acompound term indicating that it pendebat weighed ''duo two ' ' asses, and the tressis is so called " becauseit is composed of tres three ' ' units of aes copper.''Instead of asses, the ancients used sometimes to sayaes * ; a usage which survives when we hold an as into the Swiss franc (about Is. 4d. English, or 32 cents U.S.A.,in 1936).§ 83. " From tres and as, not from tres and ass. ' Butin the genitive, if with a numeral ; just as we say " fouro'clock," = " four (hours) of the clock "; in the singular,aes might mean money ' ' collectively, like the French argent,and sometimes even a 'copper piece.'505

VARROquo dicimus assem tenentes " hoc^ aere aeneaquelibra " et " mille aeris legasse."84. Quare quod ab tressis usque ad centussis^numeri ex (partibus)* eiusdem modi sunt compositi,eiusdem modi habent similitudinem : dupondius,quod dissimilis est, ut debuit, dissimilem habetrationem. Sic as, quoniam simplex est ac principium,et unum significat et multitudinis habet suum infinitum: dicimus enim asses, quos cum finimus,dicimus dupondius et tressis et sic porro.85. Sic videtur mihi, quoniam finitum et infinitumhabeat dissimilitudinem, non debere utrumque itemdici, eo magis quod in ipsis vocabulis' ubi additurcertus numerus miliar

ON THE LATIN LANGUAGE, IX. 83-85the hand and say " with this aes copper piece and' 'aenea libra pound of copper,' " ' '^and also in the legalformula " to have bequeathed a thousand (asses) ofaes copper.' "'84-. Therefore, because the numerals from tressisto centussis are compounded of parts of the samekind, they have a likeness of the same kind ; but theword dupondkis, because it is different in formation,lias a different system of declension, as it shouldhave. So also the as, because it is a single unit andis the beginning, means one and has its own indefiniteplural, for we say asses ; but when we limitthem numericallv, we say dupondius and tressis andso on.", 85. Thus it seems to me that since the definite andthe indefinite have an inherent difference, the twoought not to be spoken in the same fashion, themore so because in the words themselves, when theyare attached to a definite number in the thousands,a form is used which is not the same as that used inother expressions. For they speak thus : mille denarium" thousand of 'denarii,' not denariorum, and twomilia denarium thousands 'of denarii,' not denariorum.If it were denarii in the nominative and it denoted anindefinite quantity, then it ought to be denariorum inthe genitive ; and the same distinction must be preserved,it seems to me, not only in denarii, victoriati,^drachmae, and nummi, but also in viri, when we sayfrom 2 to 100 asses, the compound words were used, and notasses with the numeral.§ 85. " For names of weights and measures, and for someother words, the old genitive in -um continued in use longafter the new form in -orum had been generalized. * Therictoriatus wa» a silver coin stamped with a figure of Victory,and worth half a denarius.507

VARROiudicium fuisse triumvirum, decern (virum, centum)-virum,^ non ^ decem ad LX

ON THE LATIN LANGUAGE, IX. 85-87that there has been a decision of the triumvirs,'^ thedecemvirs,'' the centumvirs,* all of which have thegenitive virum and not virorum.86. The old numbers have their Regularities,because they all have one rule, two acts, three grades,and six decades, all of which show regular internalcorrespondences. The rule is the number nine,because, when we have gone from one to nine, wereturn again to one and nine " ; hence both ninety andnine hundred are of that one and the same ninecontainingnature. So there are numbers of eightcontainingnature,* and going downwards they arriveat those which are merely ones.87. The first act " is from one to nine hundred,the second from one thousand to nine hundredthousand. Because one and thousand are alikeunities, both are called by a name in the singularfor as we say ' this one ' ;and these two,' so we say''this thousand and ' ' these two thousands,' andafter that all the other numbers in the two acts arelikewise plural. The unitary grade is found in bothacts, from one to nine ; the denary grade extendsfrom ten to ninety ; the centenary grade fromhundred to nine hundred. Thus from the threegrades, six decades are made, three in the thousands,and three in the smaller numbers. The ancients weresatisfied with these numerals.executions,**The decemviri stlitibus rudicandis, a permanentboard with jurisdiction over cases involving libertyor citizenship. * The centumviri or board of judges withjurisdiction over civil suits, especially those involving inheritances.§ 86. " As multiples of ten ; and then as multiples of onehundred. * But these do not constitute the ' rule.'§ 87. ° Technical term, taken from the drama.509

VARRO88. Ad^ hos tertium et quartum actum (addentes)^ab decie(n>s

ON THE LATIN LANGUAGE, IX. 88-8988. To these, their descendants added a third anda fourth act, imposing names which started fromdeciens " ' million ' and deciens miliens * 'thousandmillion ' ; and though the names were not formed bylogical relation with the lower numerals, still their formationis not in conflict with the Regularity aboutwhich we are writing. For inasmuch as deciens is usedas a neuter singular like iiiille, so that both words arewithout change of form for the various cases, we"^shall use deciens unchanged as nominative and asgenitive, even as we do mille ; and none the lessshall we set before viille the signs of nominative andof genitive plural, because mille is also in the othernumber—and so also shall we speak of ' these deciens 'in the same cases.L. 89. When a noun is the same in the nominativethough it has more than one meaning, in whichinstance they call it a homonymy, Regularity doesnot prevent the oblique cases from the same startingform in which the homonymy is, from being dissimilar.Therefore we say Argus in the masculine,when we mean the man, but when we mean thetown we say, in Greek or in the Greek fashion,Argos " in the neuter, though in Latin it is Argi,masculine plural. Like>\ise, if the same word denotesboth a noun and a verb, we shall cause it tobe inflected both for cases and for tenses, withdifferent inflection for noun and verb, so that fromMeto as a noun, a man's name, we form gen. Metonis,ace. Metonem, but from meto as a verb, ' I reap,' weform the future vietani and the imperfect metebam.§ 89. " The homonymy is not perfect, since the forms areArgus and Argos ; the neuter Argos is found in Latin onlyin nom. and ace.511

VARROLI. 90. Reprehendunt, cum ab eadem voce plurasunt vocabula declinata, quas awtDw/xia^^ appellant,ut^ Alc(m>aeus' et Alcaeo,' sic Ger?/on, Gen/on(e>us,*Geryones. In hoc genere quod casus perperampermutant quidam, non reprehendunt analogiam,sed qui eis utuntur imperite ;quod quisquecaput prenderit, sequi debet eius consequenti(s)*casus in declinando ac non facere, cum dixerit rectocasu Alc(m>fleus,* in obliquis' Alc(m>aeoni* etAlc(m)fleonem* ;quod si miscuerit et non secutuserit analogias, reprehendendum.LII. 91- (Reprehendunt)^ Aristarchum, quodhaec nomina Melicertes et Philomedes similia negetesse, quod vocandi casus habet alter Melicerta, alterPhilomede,* sic qui dicat lepus et lupus non essesimile, quod alterius vocandi casus sit lupe, alteriuslepus, sic socer, macer, quod in transitu fiat abaltero trisi/Uabum soceri, ab altero bisyllabum macri.92. De hoc etsi supra responsum est, cum dixide lana, hie quoque^ amplius adiciam similia non solum§90. ^ For synonimyas. ^ After ut, Aug. deletedsapho et. ' Kent, for alceus and alceo, usually correctedto Alcaeus, Alcaeo, though a variant nominative Alcaeo isunknown ; whereas Alcumeus occurs in Plautus, Capt. 6G2,and Alcmaeo in Cicero, Acad. Priora ii. 28. 89, and elsewhere.* Mue., for gerionus. ^ L. Sp., for consequent!.• Kent, for alceus, alceoni, alceonem ; cf. crit. note 3.' After obliquis, Mue. deleted dicere.§91.^ Added by L. Sp., after Atig. ^ Mue., forphilomede.§ 92. ^ For hie hie quoque.§ 90. " Son of Amphiaraus and Eriphyle, who killed hismother at the command of his father, because she tricked himinto going to a war in which he was destined to die ;cf. alsothe critical note. * The three-bodied giant whom Hercules512

ON THE LATIN LANGUAGE, IX. 90-92LI. 90. They find fault when from the sameutterance two or more word-forms are derived, whichthey call synonymns, such as Alcmaeus and Alcmaeo,''and also Geryon, Geryoneus, Geryonesy As to the factthat in this class certain speakers interchange thecase-forms \

VARROa facie dici, sed etiam ab aliqua coniuncta vi etpotestate, quae et oculis et auribus latere soleant :itaque saepe gemina facie mala negamus essesimilia, si sapore sunt alio ; sic equos eadem facienonnullos negamus esse similis, (s>i* natione s(unt>'ex procreante dissimiles.*93. Itaque in hominibus emendis, si natione alterest melior, emimus pluris. Atque in hisce omnibussimilitudines non sumimus tantum a figura, sedetiam aliu(n>de, ut in equis aetas, ut in is^cuius modi faciant pullos, ut in pomis quo sint suco.Si igitur idem sequitur in similitudine verborum quis,reprehendundus non est.94. Quare similitudinwrn^ discernendarum causanonnunquam ut pronomen assumitur, sic casumaliquem assumi,^ ut in his nemus, lepus, hielepus, hoc nemus : itaque discedunt ac dicuntur hilepores, haec nemora. Sic aliud si quid assumptumerit extrinsecus, quo similitudo penitus perspicinequepossit, non nim(is>* erit remotum ab natura :enim magnetas lapides duo inter se similes sintnecne, perspicere possis, nisi minutum extrinsecusprope apposueris ferrum, quod similes lapidessimiliter ducunt, dissimiliter dissimiles.* Sciop., for in. ' L. Sp. ; natione Vertranius ; fornationes.* For dissimilis.§ 93. ^ Lachmann, for ut inis.§ 94. ^ L. Sp.,for similitudinem. ^ L. Sp.,for assumi.' A. Sp.,for enim.§ 93. " For example, Cappadocians were notoriouslyworthless as slaves.514

ON THE LATIN LANGUAGE, IX. 92-9^spoken words rest not only upon their form, but alsoupon some attached strength and power which isusually hidden from our eyes and ears. Thereforewe often say that two apples that are identical inappearance are not alike, if they are of differentflavour ; and we say that some horses of the sameappearance are not alike, if by breed they are differenton the sire's side.And in all these matters we take the points93. Therefore in buying human beings as slaves,we pay a higher price for one that is better by nationality.''of likeness not merely from the appearance, but alsofrom other factors, as in horses their age, in assesthe kind of colts that they beget, in fruits the flavourof their juice. If therefore one proceeds in the sameway in deciding whether words are alike, he is not tobe found fault with.94^. Wherefore as the pronoun ° is sometimestaken as an aid to distinguish the resemblances, sowe take some case-form, as in nemus grove and ' ' lepus'hare,' lepus being shown by it to be masculine andnemus neuter : therefore they go in different directionsand the plurals are lepores and nemora. So also,if anything else whatsoever is taken from outside toenable a thorough examination of the problem of likenessto be made, it will not be too far from the naturalqualities : for vou cannot even see whether twomagnetic stones are aUke or not, unless you havebrought close to them from outside a particle ofsteel, which like magnets attract to a like degree,and magnets different in strength attract withdifferent powers.§ 94. " Specifically, the demonstrative hie and its forms,used as indicative of gender and of case.515

VARRO95. Quod ad nominatuom^ analogia(m)* pertinet,ita deliMatum^ arbitror, ut omnia quae dicunturcontra ad respondendum ab his fontibus sumi possit.LIII. Quod ad verborum temporalium rationemattinet, cum partes sint quattuor, tempora, personae,genera, division^*,* ex omni parte quoniam reprehendunt,ad singula i*espondebo.LI V. 96. Primum quod aiunt analogias non servariin temporibus, cum dicant legi lego legam et sicsimih'(ter>^ alia : nam quae sint ut legi rem^ perfectamsignificare, duo reliqua lego et legam' inchoatam,iniuria reprehendunt : nam ex eodemgenere et ex divisione idem verbum, quod sumptumest, per tempora traduci (infecti)* potest, ut discebamdisco discam, et eadem perfecti, ut didiceram didicididicero. LV. Ex quo licet scire verborumratione

ON THE LATIN LANGUAGE, IX. 95-9795. That which concerns the Regularity ofnouns has, I think been so cleared up that materialfor answering all objections can be drawn from thesesources. LIII. We now come to the logical systemof verbs ; this has four parts : tenses, persons,kinds," and divisions.^ As they find fault withrespect to each and every part, I shall make answerto the objections one by one.LIV. 96- First as to their saying that the Regularitiesare not preserved in the tenses, when theygive perfect legi ' I have read,' present lego ' I read,'future legam ' I shall read,' and others in just thesame way : they are wTong in finding fault ^\iththose forms like legi as denoting completed actionsand the other two, lego and legam, as denoting actiononly begun ; for the same verb which has been takenfrom the same kind and the same division, can beparaded through the tenses of non-completion, likedisceham ' I was learning,' disco ' I learn,' discam ' I'shall learn,' and the same of completion, thus didiceram'I had learned,' didici' I have learned,' didiceroI shall have learned.' LV. From this one may knowthat the logical system of verbs is consistent withitself, but that those who try to speak the verbs intheir three tenses, do this in an ignorant way ;97. that likewise those do so ignorantly who findfault because we say amor ' I am loved,' amahor ' Ishall be loved,' amatus sum ' I have been loved '; for,they say, in one and the same series there ought notto be one verb made up of two words while the othertwo verbs are each of one word. Yet if you wouldcf. § 102. * There were two divisions, one comprising thetenses of incomplete action, and the other the tenses of completedaction.517

VARROverba, discrepant inter se :nam infeeta omnia simpliciasimilia sunt, et perfecta duplicia inter se pariain omnibus verbis, ut haec amabar amor amabor,amatus (eram amatus sum amatus)* ero.98. Quare item male dicunt ferio feriam percussi,quod est ordo (ferio)^ feriam feriebam, percussipercu**ero percusseram.'^ Sic deinceps in reliquistemporibus reprehendenti responderi potest.LVI. 99- Similiter errant qui dicunt ex utraqueparte verba omnia commutare syllabas oportere autnullum, in his pungo pungam pupugi, tundo tundamtutudi : dissimilia enim conferunt, verba infecti cumperfectis. Quod si infeeta modo conferrent, omniaverbi principia incommutabilia viderentur, ut in hispungebam pungo pungam et contra ex utraque partecommutabilia, si perfecta ponerent, ut pupugerampupugi pupugero.L\'II. 100. Item male conferunt fui sum ero, quodfui est perfectum, cuius series sibi, ut debet, in omnibuspartibus^ constat, quod est fueram fui fuero ; deinfectis sum quod nunc dicitur olim dicebatur esumet in omnibus personis constabat, quod dicebatur* Added by L. Sp.§ 98. ^ Added here by Miie. ; added after feriam by G, H,Aldus. ^ Mue., for percutio percutiam / see note b.§ 100. ^ A. Sp.,for personis.§ 98. " In this section Varro changes the order in whichhe cites the tenses. * Ferio is found only in the presenttense-system ; in the perfect tense-system it is replaced bypercvssi, the present tense-system of which is relatively littleused. This justifies the emendation of the text.518

ox THE LATIN LANGUAGE, IX. 97-100put down verb-forms from a dixision of one kind,they would not differ from one another ; for all theforms denoting incomplete action are alike single,and the forms of completed action are in all verbsdouble, quite like one another : such as amabar, amor,amabor, and amatits eram, amatus sum, amatus ero.98. Wherefore likewise they do ill to cite ferioI strike,' future feriam, perfect percussi ; becausethe proper order isJ'erio,J'eriam,Jeriebani,'* and percussi,percussero, percusseram.^ And in this fashion answercan be made to the one who finds fault in the matterof the other tenses.L\T. 99- They make a similar mistake who saythat all verbs ought to change the radical syllablesin both di\isions, or no verb should—as in pungo'I prick,' future pungam, perfect pupugi, and tundoI pound,' tundam, tutudi ; for they are comparingunlikes, namely verbs of the incomplete phase x^-iththe completed. But if they were comparing onlythe incomplete, then all the stems of the verb wouldbe seen to be unchangeable, as in pungebam, pungo,pungam, and on the other hand changeable, if thevinstanced the completed, as in pupugeram, pupugi,pupugero.LVII. 100. Likewise they do ill to compare JutI was,' sum ' I am,' ero ' I shall be ' ; for Jtii is a form'of completed time, whose series is consistent withitself in all its parts, as it should be, namely Jveram,fui, fuero. Of the incomplete, that which is nowpronounced sum used to be spoken esum,° and theseries is consistent in all its persons, because they§ 100. " This form seems to have been invented by Varroto suit his argument ; all the evidence is against its everhaving existed.519

VarroVARROesum es est, eram eras erat, ero eris erit ; sic huiuscemodi cetera servare analogiam videbis.LVIII. 101. Etiam in hoc reprehendunt, quodquaedam verba neque personas habent ternas nequetempora terna : id imperite reprehendunt, ut si quisreprehendat naturam, quod non unius modi finxeritanimalis omnis.Si^ enim natura non omnes formaeverborum terna habent* tempora, ternas personas,non habent totidem verborum divisiones. Quarecum imperamus, natura quod infecta (ver>ba' solumhabe* caput rectus casus,sic hie in forma est persona eius qui loquitur ettempuspraesens, ut scribo lego.§ 101. ^ Avff., for sic. * G, H, a, for habeant.' Mue., with G,for infectaba. * Christ, with G, for habet.* L. Sp., for et. * L. Sp., for imperandi.§ 102. ^ Laetus, for discrepant. * Abl. quique Lachmann,for quisque. ' L. Sp., for nominatiuis. * Fay,for externi.§ 101. " Present imperative, future imperative, present*" "^subjunctive. The indicative mood. disregardsthe, phiral forms in this calculation.§ 102. ° Meaning mood ' ' ; c/. § 95, note a. * Cf.ix. 75-79.520

ON THE LATIN LANGUAGE, IX. 100-102used to say present esum es est, imperfect eram eraserat, future ero eris erit. In this same fashion youwill see that the other verbs of this kind preserve theprinciple of Regularity.LVIII. 101. Besid'es, they find fault \^ith Regularityin this matter, that certain verbs have not thethree persons, nor the three tenses ; but it is with lackof insight that they find this fault, as if one shouldblame Nature because she has not shaped all li\'ingcreatures after the same mould. For if by nature notall forms of the verbs have three tenses and threepersons, then the di\isions of the verbs do not all havethis same number. Therefore when we give a command,a form which only the verbs of uncompletedtime have—when we give a command to a personpresent or not actually present, three verb-forms " aremade, like lege read ' (thou),' tegito read (thou) ' ' or'let him read,' legal ' let him read ' : for nobodygives a command with a form denoting action alreadycompleted. On the other hand, in the forms whichdenote declaration,* like lego ' I read,' legis thou'readest,' legit he ' reads,' there are nine verb-formsof uncompleted action and nine of completedaction.*^LIX. 102. For this and similar reasons thequestion that should be asked is not whether onekind " disagrees with another kind, but whether thereis anything lacking in each kind. If to thesethere is added what I said above * about nouns, alldifficulties -snll be easily resolved. For as the nominativecase-form is in them the source for the derivativecases, so in verbs the source for other forms is in theform which expresses the person of the speaker andthe present tense : like scribo ' I write,' lego ' I read.'521

VARRO103. Quare ut illic fit, si^ hie item aeeiderit, informula ut aut caput non sit aut ex alieno genere sit,proportione eadem quae illic dicimus, cur nihilominus*servetur analogia. Item, sicut illic caput suumhabebit et in obliquis casibus transitio erit in aliquam' formulam, qua assumpta reliqua faciliuspossint videri verba, unde sint declinata (fit enim, utrectus casus nonnunquam sit ambiguus), ut in hocverbo volo, quod id duo significat, unum a voluntate,alterum a volando ; itaque a volo intellegimus etvolare et velle.LX. 104. Quidam reprehendunt, quod pluit etluit dicamus in praeterito et praesenti tempore, cumanalogiae sui cuiusque temporis verba debeant discriminare.Falluntur : nam est ac putant aliter,quod in praeteritis U dicimus longum pluit (luit),^in praesenti breve pluit luit : ideoque in lege venditionisfundi " ruta caesa " ita dicimus, ut U producamus.LXI. 105. Item reprehendunt quidam, quodputant idem esse sacrifico^ et sacrificor, lavat" etlavatur ;quod sit an non, nihil commovet analogian,dum sacrifice' qui dicat servet sacrificabo et sic per§ 103. ^ Mue.,for sic. ^ For nichilominus. ' 3fue.,for aliquam.§ 104. 1 Added by Aug.§ 105. ^ Aug., for sacrificio. * L. Sp. ; sacrificor etlavat Aug. ; for sacrifice relauat. ' Aug., for sacrifici.§ 103. « Cf. ix. 76.§ 104. " Found in older Latin, but seemingly shortenedby about Varro's time. * One might exempt from inclusionin the sale of a property all things dug up (sand, chalk,ete.) and ail things cut down (timber, etc.), even though theywere still unwrought materials. " The u is short in thecompounds erutus, obrutus, etc.522

ON THE LATIN LANGUAGE, IX. 103-105I103. Wherefore, if it has happened in verbs as itdoes happen in nouns, that in the pattern the startingpointis lacking or belongs to a different kind, we givethe same arguments here which we gave there," withsuitable changes in application, as to why and howRegularity is none the less preserved. And as innouns the word will have its ovm peculiar startingpointand in the oblique cases there will be a changeto some other pattern, on the assumption of which itcan be more easily seen from what the word-forms arederived (for it happens that the nominative case-formis sometimes ambiguous), so it is in verbs, as in thisverb volo, because it has two meanings, one fromwishing and the other from flying ; therefore fromvolo we appreciate that there are both volare 'to fly 'and velle 'to wish.'LX. 104. Certain critics find fault, because wesay pltiit rains ' and ' luit ' looses both ' in the pasttense and in the present, although the Regularitiesought to make a distinction between the verb-formsof the two tenses. But they are mistaken ; for it isother^vise than they think, because in the past tensewe say pluit and luit with a long U," and in the presentwith a short U ; and therefore in the law about thesale of farms we say rtda caesa things dug up and'things cut,'* with a lengthened m."LXI. 105. Likewise certain persons find fault,because they think that active sacrijico ' I sacrifice 'and passive sacrificor, active lavat'he bathes ' andpassive lavatur, are the same " : but whether this isso or not, has no effect on the principle of Regularity,provided that he Avho says sacrijico sticks to the future§ 105.

VARROtotam formam, ne dicat sacrificatur* aut sacrificatussum : haec enim inter se non conveniunt.106. Apud Plautum, cum dicit :Piscis ego credo qui usque dum vivunt lavantDiu minus lavari^ quam haec lavat Phronesium,ad lavant lavari non convenit, ut /* sit postremum,sed E ; ad lavantur analogia lavari reddit : quodPlauti aut librarii mendum siest, non ideo analogia,sed qui scripsit est reprehendendus. Omnino etlavat' et lavatur dicitur separatimrecte in rebus certis,quod puerum nutrix lava(t),* puer a nutrice lavatur,nos in fealneis et lavamus et lavamur.107. Sed consuetude alterum utrum cum satishaberet, in toto corpore potius utitur lavamur, inpartibus lavamus, quod dicimus lavo manus, sic pedeset cetera. Quare e balneis non recte dicunt lavi, lavimanus recte. Sed quoniam in balneis lavor lautussum, sequitur,ut contra, quoniam est soleo, oporte

ON THE LATIN LANGUAGE, IX. 105-107sacrijicabo and so on in the active, through the wholeparadigm, avoiding the passive sacrijicatur andsacrificatus sum : for these two sets do not harmonizewith each other.106. In Plautus, when he says " :The fish, I really think, that bathe through all their life.Are in the bath less time than this Phronesium,lavari are 'in the bath,' with final I instead of E, doesnot attach to lavant bathe ' ' : Regularity refers lavarito lavantur, and whether the error belongs to Plautusor to the copyist, it is not Regularity, but the writerthat is to be blamed. At any rate, lavat and lavaturare used with a difference of meaning in certainmatters, because a nurse lavat bathes a ' ' child, thechild lavatur ' is bathed by the nurse, and ' in thebathing establishments we both lavamus bathe and' 'lavamur are bathed.''107. But since usage approves both, in the caseof the whole body one uses rather lavamur we bathe'ourselves,' and in the case of portions of the bodylavamus we ' wash,' in that we say lavo ' I wash my'hands, my feet, and so on." Therefore vriih referenceto the bathing establishments they are WTong insaying lavi ' I have bathed,' but right in saying lavi'I have washed my hands. But ' since in the bathingestablishments lavor ' I bathe and ' lautus sum ' Ihave bathed,' it follows that on the other hand fromsoleo ' I am wont,' which is in the active, one oughtto say solui ' I have been wont,' as Cato * and Ennius "write, and that solitus sum, as the people in generalsay, ought not to be used. But as I have said above,"*Regularity exists none the less for these few inconsistencieswhich occur in speech.525

VARROLXII. 108. Item cur non sit analogia, a^erunt,^quod ab similibus similia non declinentur, ut ab doloet colo : ab altero enim dicitur dolavi, ab altero colui ;in quibus assumi solet aliquid, quo facilius reliquadicantur, ut i(ny Mj/rmecidis' operibus minutis soletfieri : igitur in verbis temporalibus, quo* similitudosaepe sit confusa, ut diseerni nequeat, nisi transierisin aliam personam aut in tempus, quae propositasunt nosse* similia intellegitur, cum transitumest in secundam personam, quod alterum estdolas, alterum colis.109. Itaque in reliqua forma verborum suamutr(um>que^ sequitur formam. Utrum in secunda(persona)'^ forma verborum temporal«' habeatin extrema syllaba AS

ON THE LATIN LANGUAGE, IX. 108-110LXII. 108. Likewise, they present as an argumentagainst the existence of Regularity the fact that likeforms are not derived from likes, as from dolo 'I chop 'and colo ' I till ' ; for one forms the perfect dolaviand the other forms colui. In such instances somethingadditional is wont to be taken to aid in themaking of the other forms,'' just as we do in the tinyart-works of Myrmecides ^ : therefore in verbs, sincethe likeness is often so confusing that the distinctioncannot be made unless you pass to another person ortense, you become aware that the words before youare not alike when passage is made to the secondperson, which is dolas in the one verb and colis in theother.109. Thus in the rest of the paradigm of theverbs each follows its own special type. Whetherin the second person the paradigm of verbs has in thefinal syllable AS or ES or IS or JS, is of importancefor distinguishing the likenesses. Wherefore the markof Regularity is in the second person rather than in thefirst, because in the first the unlikeness is concealed,as appears in meo ' I go,' neo ' I sew,' ruo ' I fall ; 'forfrom these there develop unlike forms by the changefrom first to second person, because they are spokenthus : meo meas, neo nes, ruo mis, each one of whichpreserves its own type of likeness.LXIII. 110. Likewise, many find fault withRegularity in connexion with the so-called participles; wrongly : for it should not be said that theset of three participles comes from each individualverb, like amaturus about to ' love,' amans ' loving,'amatus ' loved,' because amans and amaturus are fromparticiple to make up the " principal parts " which are ourguide. * Cf. vii. 1,527

VARROab amor* amatus. Illud analogia quod praestaredebet, in suo quieque genere habet, casus, ut amatusamato et amati amatis ; et sic in muliebribus amataet amatae ; item amaturus eiusdem modi habetdeclinationes, amans paulo aliter ;quod hoc genusomnia sunt in suo genere similia proportione, sicvirilia et muliebria sunt eadem.LXIV. 111. De eo quod in priore libro extremumest, ideo non es(se> analogia(m),^ quod qui de eascripserint aut inter se non conveniant aut in quibusconveniant ea cum consuetudinis discrepent^ verbis,utrumque (est leve)^ : sic enim omnis repudiandumerit artis, quod et in medicina et in musica et inaliis multis discrepant scriptores ; item in quibusconveniunt in* scriptis, si e

ON THE LATIN LANGUAGE, IX. 110-112the active amo, and amatus is from the passive amor.But that which Regularity can offer, which the participleshave, each in its own class, is case-forms, asamatus, dative amato, and plural amati, dative amatis ;and so in the feminine, amata and plural amatae.Likewise amaturus has a declension of the same kind.A mans has a somewhat different declension ; becauseall words of this kind have a regular likeness in theirown class, amans, like others of its class, uses thesame forms for masculine and for feminine.LXIV. 111. About the last argument in the precedingbook, that Regularity does not exist for thereason that those who have MTitten about it do notagree with one another, or else the points on whichthey agree are at variance with the words of actualusage, both reasons are of little weight. For in thisfashion you Mill have to reject all the arts, becausein medicine and in music and in many other arts the«Titers do not agree ;you must take the same attitudein the matters in which they agree in their WTitings,if none the less nature rejects their conclusions. Forin this way, as is often said, it is not the art but theartist that is to be found fault with, who, it must besaid, has in his ^\Titing failed to see the correct view ;we should not for this reason say that the correctview cannot be formulated in wTiting.112. As to the man who uses as ablatives montihill and ' Jbnti spring ' ' while others say vionte and'fonte,^ along with other words which are used intwo forms, one form is correct and the other is wrong,yet the person who errs is not destroying the Regularities,but the one who speaks correctly is strengtheningit ; and as he who errs in these words wherethey are used in two forms is not destroying logicalVOL. II M 529

VARROtoUit rationem cum sequitur falsum, sic etiam in his(quae)* non* duobus dicuntur, si quis aliter putatdici oportere atque oportet, non scientiam tollitorationis, sed suam inscientiam denudat.LXV. 113. Quibus rebus solvi arbitraremur possequae dicta sunt priori libro contra analogian, ut potuibrevi percucurri. Ex quibus si id confecissent^ quodvolunt, ut in lingua Latina esset anomalia, tamennihil egissent* ideo, quod in omnibus partibus mundiutraque natura inest, quod alia inter se (similia),'alia (dissimilia)* sunt, sicut in animalibus dissimiliasunt, ut equus bos ovis homo, item alia, et in unoquoque horum genere inter se similia innumerabilia.Item in piscibus dissimilis murcena lupo, is* soleae,haec TWMraenae^ et mustelae, sic aliis, ut maior illenumerus sit similitudinum earum quae sunt separatimin muraenis, separatim in asellis, sic in generibusaliis.114. Quare cum in inclinationibus verborumnumerus sit magnus a dissimilibus verbis ortus, quodetiam vel maior est in quibus similitudines reperiuntur,conftVendum^ est esse analogias. Itemque^ cumea non multo minus quam in omnibus verbis patiaturuti consuetudo comunis, fatendum illud quoquo* Added by Aug. ' After non, Auff. deleted in.§113.^ For conficissent. ^ Aug., for legissent.Added by Mue. * L. Sp.,for his. ^ G, H, Aldus, for'nerene.§114.^ Aug., for conferendum. ^ Aug., for itemquae.* That is, wrong forms not recognized as having a Hmitedcurrency, but practically individual with the speaker.§ 113. " The identification of the various kinds of fish is530

ON THE LATIN LANGUAGE, IX. 112-114system when he follows the wTong form, so even inthose words which are not spoken in two ways, aperson who thinks they ought to be spoken otherwisethan they ought,* is not destroying the science ofspeech, but exposing his o>\ti lack of knowledge.LX^^ 113. The considerations by which we mightthink that the arguments could be refuted whichwere presented against Regularity in the precedingbook, I have touched uf>on briefly, as best I could.Even if by their arguments they had achieved whatthey wish, namely that in the Latin language thereshould be Anomaly, still they would have accompUshednothing, for the reason that in all parts ofthe world both natures are present : because somethings are like, and others are unlike, just as inanimals there are unlikes such as horse, ox, sheep,man, and others, and yet in each kind there arecountless indi\iduals that are like one another. Inthe same way, among fishes, the moray is unhke thewolf-fish, the wolf-fish is unlike the sole, and this isunlike the moray and the lamprey, and others also ;though the number of those resemblances is stillgreater, which exist separately among morays,among codfish, and in other kinds of fish, class byclass."114. Now although in the derivations of wordsa great number develop from unlike words, still thenumber of those in which likenesses are found is evengreater, and therefore it must be admitted that theRegularities do exist. And likewise, since generalusage permits us to follow the principle of Regularityin almost all words, it must be admitted that we oughtin some instances uncertain, but is not important for Varro'sai^ument.531

VARROm{o)do^ analogian sequi nos debere universos,singulos autem praeterquam in quibus verbis ofFensurasit consuetude co(m>munis, quod ut dixi aliuddebet praestare populus, aliud e populo singulihomines.115. Neque id mirum est, cum singuli quoque nonsint eodem iure : nam liberius potest poeta quamorator sequi analogias. Quare cum hie liber idquod pollicitus est demonstraturum absolveri^,^faciam finem ;proxumo deinceps de declinatorumverborum forma* scribam.' Canal ; quoque modo Mue. ; quodammodo Aug. ; forquo quando.§ 11 5. ^ Aldus, for absoluerim. * Pius, for firma.532

ON THE LATIN LANGUAGE, IX.11^115as a body to follow Regularity in every way, andindividually also except in words the general use ofwhich will give offence ; because, as I have said,"the people ought to follow one standard, the individualpersons ought to follow another.115. And this is not astonishing, since not allindividuals have the same privileges and rights ;for the poet can follow the Regularities more freelythan can the orator. Therefore, since this book hascompleted the exposition of what it promised to setforth, I shall bring it to a close ; and then in the nextbook I shall write about the form of inflected words.§114, « C/. ix. 5.533

BookM. TERENTI VARRONISDE LINGUA LATINAAD CICERONEM LIBER Villi EXPLICIT ; INCIPITXI. 1 . In verborum declinationibus disciplina loquendidissimilitudinem an similitudinem sequi deberet,multi quaesierunt. Cum ab his ratio quae ab similitudineoriretur vocaretur analogia, reliqua parsappellaretur anomalia : de qua re primo libro quaedieerentur cur dissimilitudinem ducem haberi oporteret,dixi, secundo contra quae dic(er)entur,^ curpotius similitudinem^ conveniret praeponi : quarumrerum quod nee fundamenta, ut deb(u>it,' posita abullo neque ordo ac natura, ut res postulat, explicita,ipse eius rei formam exponam.2. Dicam de quattuor rebus, quae continentdeclinationes^ verborum : quid sit simile ac dissimile,quid ratio quam appellant Aoyov, quid pro portioned§1.^ Aldus, for dicentur. ^ Aldus, for dissimilitudinem.* ^M^.,/or debita.§ 2. ^ L. Sp., for declinationibus. ^ Plasberg, for proportione.§ 1. " Book VIII., which begins a fresh section of the^"entire work. IX.534

MARCUS TERENTIUS VARRO'SON THE LATIN LANGUAGEbook ixAddressed to Ciceroends, and here beginsBOOK XI. 1 . Many have raised the question whether in theinflections of words the art of speaking ought tofollow the principle of unlikeness or that of likeness.This is important, since from these develop the twosystems of relationship : that which develops fromlikeness is called Regularity, and its counterpart iscalled Anomaly. Of this, in the first book," I gavethe arguments which are advanced in favour of consideringunlikeness as the proper guide ; in thesecond,* those advanced to show that it is properrather to prefer likeness. Therefore, as their foundationshave not been laid by anyone, as should havebeen done, nor have their order and nature been setforth as the matter demands, I shall myself sketch anoutline of the subject.2. I shall speak of four factors which limit theinflections of words : what likeness and unlikenessare ; what the relationship is which they call logos ;what " by comparative likeness "is, which they call535

VARROquod^ dicunt dva Aoyov,* quid consuetude ;quaeexplicatae declarabunt analogiam et anomalia

ON THE LATIN LANGUAGE, X. 2-5" according to /ogoj " " ; what usage is. The explanationof these matters ^^ill make clear the problemsconnected with Regularity and Anomaly : whencethey come, what they are, of what sort they are.IL 3. The first topic to be discussed must be Ukenessand unlikeness, because this matter is thefoundation of all inflections and set Umits to therelationship of words. That is like which is seento have several features identical \\'ith those of thatwhich is like it, in each case : that is unhke, which isseen to be the opposite of what has just been said.Every like or unUke consists of two units at least,because nothing can be like \\ithout being like somethingelse, and nothing can be unlike without associationwith something to which it is unUke.4. Thus a human being is said to be like a humanbeing, and a horse to be like a horse, and a humanbeing to be unlike a horse ; for a human being is likea human being because they have limbs of the sameshape, which separate human beings from the categoryof the other animals.Among human beingsthemselves, for a like reason a man is more like a manthan a man is like a woman, because men have morephysical parts the same ; and so an elderly man ismore like an old man than he is like a boy. Further,they are more like who are of almost the samefeatures, the same bearing of person, the same shapeof body ; therefore those who have more points ofidentity, are said to be more like ; and those whocome nearest to having them all alike, are calledmost like, as it were, twins.5. There are those who think that things havethree natures, hke, unUke, and neutral, which lastthey sometimes call the not like, and sometimes the537

VARROnon dissimile (sed quamvis tria sint simile dissimileneutrum, tamen potest dividi etiam in duas partessic, quodcumque eonferas aut simile esse aut non esse)simile esse et dissimile, si videatur esse ut dixi, neutrum,si in neutram partem praeponderet, ut si duaeres quae conferuntur vicenas habent partes et in hisdenas habeant easdem, denas alias ad similitudinemet dissimilitudinem aeque animadvertendas : bancnaturam plerique subiciunt sub dissimilitudinisnomen.6. Quare quoniam fit^ ut potius de vocabulo quamde re controversia esse videatur, illud est potiusadvertendum, quom simile quid esse dicitur, cui^ partisimile dicatur esse (in hoc enim solet esse error), quodpotest fieri ut homo homini simih's' non sit,* ut multaspartis habeat similis et ideo dici possit similis habereoculos, manus, pedes, sic alias res separatim et unaplures.7. Itaque quod diligenter videndum est in verbis,quas partis et quot modis oporteat similis habere(quae similitudinem habere)^ dicuntur, ut infraapparebit, is locus maxime lubricus est. Quid enimsimilius potest videri indiligenti quam duo verba haecsuis et suis ? Quae non sunt, quod alterum* significatsuere, alterum suem.Itaque similia vocibus§ 6. ^Aug., for fuit. ^ quoi L. 8p., for quin cui.3 V, p, C. F. W. Mueller, for simile. * non sit Bhol.,forsit non sit.§ 7. ^ Added by GS., cf. § 12 end ; quae similia esse,added by L. Sp. ; ut similia, by Canal. ^ After alterum,p and Aug. deleted non.538

ON THE LATIN LANGUAGE, X. 5-7not unlike ; but although there are the three, Uke,unlike, neutral, there can also be a division into twoparts only, in such a way that whatever you comparewith something else either is like or is not. Theythink that a thing is like and is unlike if it is seen tobe of such a kind as I have described, and neutral, ifit does not have greater weight on one side than onthe other ; as if the two things which are being comparedhave twenty parts each, and among these shouldhave ten to be noted as identical and ten likewise tobe noted as different, in respect to likeness andunlikeness. This nature most scholars include underthe name of unlikeness.6. Therefore since it happens that the question indispute seems rather to be about the name thanabout the thing, attention must rather be directed,when something is said to be like, to the problem towhat part it is said to be like ; for it is in this that anymistake ordinarily rests. This must be noted, I say,because it can happen that a man may not be likeanother man even though he has many parts like theother's, and can be said therefore to have like eyes,hands, feet, and other physical features in considerablenumber, separately and taken together, like theother man's.7. Therefore because careful watch must be keptin words to see what parts those words which are saidto show likeness ought to have alike, and in what ways,the inquirer is on this topic especially likely to slipinto error, as will appear below. For to the carelessperson what can seem more alike than the two wordssuis and suis ? But they are not alike, because one isfrom suere to sew and means thou ' ' ' sewest,' and'the other is from sus and means of a swine.' There-539

VARROesse ac syllabis confitemur, dissimilia esse partibusorationis videmus, quod alterum habet tempora,alterum casus, quae duae res vel maxime discernuntanalogias.8. Item propinquiora genere inter se verbasimilem saepe pariunt errorem, ut in hoc, quod nemus^et lepus videtur esse simile, quom** utrumque habeateundem casum rectum ; sed non est simile, quod eis^certae similitudines opus sunt, in quo est ut in generenominum sint eodem, quod in his non est : nam invirili genere* est lepus, ex neutro nemus ; dicitur enimhie lepus et hoc nemus. Si eiusdem generis esse(n)t,*utrique praeponeretur idem ac diceretur aut hie lepuset hie nemus aut hoc nemus, hoc lepus.9. Quare quae et cuius modi sunt genera similitudinumad hanc rem, perspiciendum ei qui declinationesverborum proportione sintne quaeret. Quem^locum, quod est difficilis, qui de his rebus scripseruntaut vitaverunt aut inceperunt neque adsequi potuerunt.10. Itaque in eo dissensio neque ea unius modiapparet : nam alii de omnibus universis discriminibusposuerunt numerum, ut D/on^sius S?donius, quiscripsit ea^ esse septuaginta un^m,^ alii parti's' eiusquae habet* casus, cuius eidem hie cum dicat esse§ 8. ^ H, RhoL, for numerus. ^ Mue., for quod cum.' Aug., for eas. * After genere, A^lg. deleted nominumsint eodem, repeated from the previous line. ^ Aug., foresset.§ 9. ^ Mue., for quod.§ 10. ^ L. Sp.,for eas. ^ L. Sp.,for unam. ' Mtie.,for partes. * Mue., for habent.§ 8. " That is, so far as the termination is concerned.§ 10. " That is, schemes of inflection. * A pupil ofAristarchus.540

ON THE LATIN LANGUAGE, X. 7-10fore we admit that they are alike as spoken wordsand in their separate syllables, but we see thatthey are unlike in their parts of speech, becauseone has tenses and the other has cases ; and tensesand cases are the two features which in the highestdegree serve to distinguish the different systems ofRegularity.8. Likewise, words that are even nearer alike inkind often cause a similar mistake, as in the fact thatnemus grove and ' ' lepus hare seem ' ' to be alike sinceboth have the same nominative "; but it is not aninstance of likeness, because they stand in need ofcertain factors of likeness, among which is that theyshould be in the same noun-gender. But these twowords are not, for lepus is masculine and nemus isneuter ; for we say hie ' this with ' lepus and hoc withnemus. If they were of the same gender, the sameform would be set before both, and we should sayeither kic lepus and hie nemus, or hoc nemus and hoclepus.9. Therefore he who asks whether the inflectionsof words stand in a regular relation, must examineto see what kinds of likenesses there are and of whatsort they are, which pertain to this matter. And justbecause this topic is difficult, those who have writtenof these subjects either have avoided it or have begunit without being able to complete their treatment of it.10. Therefore in this there is seen a lack of agreement,and not merely of one kind. For some havefixed the number of all the distinctions " as a whole,as did Dionysius of Sidon,* who wrote that there wereseventy-one of them ; and others set the number ofthose distinctions which apply to the words which havecases : the same wTiter says that of these there are54.1

VARROdiscrimina quadragmta^ septem, Aristocles rertulit*in litteras XIIII, Parmeniscus VIII, sic alii paucioraaut plura.11. Quarum similitudinum si esset origo rectecapta et inde orsa ratio, minus erraret(ur>' in declinationibusv(er>borum.* Quarum ego principiaprima duum generum sola arbitror esse, ad quae'similitudines exjgi* oporteat : e quis unum positumin verborum materia, alterum ut in materiae figura,quae ex declinatione fit.12. Nam debet esse unum, ut verbum verbo, undedeclinetur, sit simile ; alterum, ut e verbo in verbumdeclinatio, ad quam conferetur, eiusdem modi sit :alias enim ab similibus verbis similiter declinantur,ut ab erus^ ferus, ero^ fero, alias dissimiliter erus^ferus, eri^ ferum. Cum utrumque et verbum verboerit simile et declinatio declinationi, turn deniquedicam esse simile* ac duplicem et perfectam similitudinemhabere, id quod postulat analogia.*13. Sed ne astutius videar posuisse duo generaesse similitudinum sola, cum utriusque inferioresspecies sint plures, si de his reticuero, ut mihi relin-* M, Laetus, for quadringenta. * Mue. ; retulit Laetus ;for rutulit.§11. ^ VertraniuSffor &cra.rei. ^ For uborum. ^Aldus,for atque. * For exegi.§ 12. 1 For herus. " For hero. ' For heri. * L.Sp., for similem. * For analogiam." Probably Aristocles of Rhodes, a contemporary of Varro.^ A pupil of Aristarchus.542

ON THE LATIN LANGUAGE, X. 10-13forty-seven, Aristocles " reduced them to fourteenheadings, Parmeniscus ** to eight, and others made thenumber smaller or larger.11. If the origin of these likenesses had beencorrectly grasped and their logical explanation hadproceeded from that as a beginning, there would beless error in regard to the inflections of words. Ofthese likenesses there are, I think, first principles oftwo kinds only, by which the likenesses ought to betested ; of which one lies in the substance of thewords," the other lies, so to speak, in the form * ofthat substance, which comes from inflection.12. For there must be one, that the word be likethe word from which it is inflected, and two, that incomparison from word to word the inflectional formwith which the comparison is made should be of thesame kind. * For sometimes there are like formsreached by inflection from like words, such as dativesero a.nd fero from erus master and ' ' Jerus ' wild,' andsometimes unlike forms, such as genitive eri andaccusative y^rMw, from eras and Jerus. When bothprinciples are fulfilled and word is like word andinflectional form like inflectional form, then and notbefore will I pronounce that the word is like, and hasa twofold and perfect likeness to the other—which iswhat Regularity demands.13. But I wish to avoid the appearance of trickinessin having declared that there are only two kindsof likenesses when both have a number of sub-forms—if I say nothing about these, you may think that Iam intentionally leaving myself a place of refuge ; I§ 11. " That is, its form and ending, in the form which isthe starting point for inflection. * The inflectional form ;c/. § 12.543

VARROquam latebras, repetam ab origine similitudinum quaein conferendis verbis et inclinandis sequendae autvitandae sint.14. Prima divisio in oratione, quod alia verbanusquam decliwantur,^ ut haec vix mox, alia declinantur,ut ab lima limae,^ a fero ferebam, et cum nisiin his verbis quae declinantur non possit esse analogia,qui dicit simile esse mox et nox errat, quod non esteiusdem generis utrumque verbum, cum nox succederedebeat sub casuum ratione(m>,* mox nequedebeat neque possit.15. Secunda divisio est de his verbis quae declinaripossunt, quod alia sunt a voluntate, alia anatura. Voluntatem appello, cum unus quivis anomine aliae (rei)^ imponit nomen, ut RomulusRomae ; naturam dico, cum universi acceptum nomenab eo qui imposuit non requirimus quemadmodumis velit declinari, sed ipsi declinamus, ut huius Romae,hanc Romam, hac Roma. De his duabus partibusvoluntaria declinatio refertur ad consuetudinem,naturalis ad rationem.*16. Quare proinde ac simile conferred non oportetac dicere, ut sit ab Roma Romanus, sic ex Capua dicioportere Capuanus, quod in consuetudine vehementernatat, quod declinantes imperite rebus nomina imponunt,a quibus cum accepit consuetudo, turbulenta§ 14. ^ For declimantur. ^ OS., for limabo. ' Lachmann,for ratione.§ 15. ^ Added by GS. ^ Aug., for orationem.§ 16. ^ StepJianus,for conferri.544

ON THE LATIN LANGUAGE, X. 13-16shall therefore go back and start from the origin ofthe likenesses which must be followed or avoided inthe comparison of words and in their inflections.14. The first division in speech is that some wordsare not changed into any other form whatsoever,like vix hardly and ' ' )7iox ' soon,' and others are inflected,like genitive limae from /n«a *file,' imperfectfereham from fero ' I bear ' ; and since Regularitycannot be present except in words which are inflected,he who says that mox and nox night ' ' are ahke, ismistaken, because the two words are not of the samekind, since fiox must come under the system of caseforms,but mox must not and cannot.15. The second division is that, of the words whichcan be changed by derivation and inflection, someare changed in accordance with will, and others inaccordance with nature. I call it will, when from aname a person sets a name on something else, asRomulus gave a name to Roma ; I call it nature,when we all accept a name but do not ask of the onewho set it how he wishes it to be inflected, but ourselvesinflect it, as genitive Romae, accusative Romam,ablative Roma. Of these two parts, voluntary derivationgoes back to usage, and natural goes back tological system.16. For this reason we ought not to compareRomanus Roman ' 'and Capuanus Capuan ' ' as aUke,and to say that Capuanus ought to be said fromCapua just as Romanus is from Roma ; for in suchthere is in actual usage an extreme fluctuation, sincethose who derive the words set the names on thethings with utter lack of skill, and when usage hasaccepted the words from them, it must of necessityspeak confused names variously derived. ThereforeVOL. II N 545

VARROnecesse est dicere. Itaque neque Aristarchei* nequealii in analogiis defendendam eius susceperunt causam,sed, ut dixi, hoc genere declinatio in co* imperito : itaque in hocgenere in loquendo* magis anomaUa quam analogia.17. Tertia divisio est : quae verba decUnatanatura ; ea dividwwtur^ in partis quattuor : in unamquae habet casus neque tempora, ut docilis et facihs ;in alteram quae tempora neque casus, ut docet facit ;in tertiam quae utraque, ut docens faciens ; inquartam quae neutra, ut docte et facete. Ex hacdivisione singulis partibus tres reliquae* dissimiles.Quare nisi in sua parte inter se collata erunt verba,si^ conveniunt, non erit ita simile, ut debeat facereidem.18. Unius cuiusque part?s^ quoniam species plures,de singulis dicam. Prima pars casualis dividitur inpartis duas, in nominatus scilicet'^ (et articulos),*quod aeque* finitum (et infinitum)^ est ut hie et quis ;de his generibus duobus utrum sumpseris, cum2 Kent, for Aristarchii / c/. viii. 63. * Added by Groth.* For ioquenda.§17. ^ L. Sp., for dividitur. ^ Mue., for reliquere.' After si. Canal deleted non.§ 18. The text of this % stands in the manuscripts between§ 20 and § i?l ; the shift ofposition teas made by Mueller, wholeft unius cuiusque partis at the end of § 20 ; A. Spengeltransferred these words also. ^ Sciop., for partes.^ Laetus,for s ( =sunt). * Added by Ahie. * L. Sp., forneque. * Added by L. Sp. ; cf. viii. 45.§ 16. "This is shown even to-day in the new technicalterminology of some near-sciences. * Varro is somewhat546

ON THE LATIN LANGUAGE, X. 16-18neither the followers of Aristarchus nor any othershave undertaken to defend the cause of voluntaryderivation as among the Regularities ; but, as I havesaid, this kind of derivation of words in commonusage is an ill thing, because it springs from thepeople, which is without uniformity and withoutskill." Therefore, in speaking, there is in this kind ofderivation rather Anomaly than Regularity.''17. There is a third division, the words whichare by their nature inflected. These are dividedinto four subdi\isions : one which has cases but nottenses," like docilis ' docile ' and facilis ' easy ' ; asecond, which has tenses but not cases,* like docetteaches,' J'acit makes ; a third which has both,*'like docens teaching,' ' faciens making ' ' ; a fourthwhich has neither,** like docte learnedly ' ' and J'acete'wittily.' The individual parts of this division areeach unlike the three remaining parts. Therefore,unless the words are compared with one another intheir own subdivision, even if they do agree the oneword will not be so like the other that it ought tomake the same inflectional scheme.18. Since there are several species in each part, Ishall speak of them one by one. The first subdivision,characterized by the possession of cases, isdivided into two parts, namely into nouns andarticles, which latter class is both definite and indefinite,as for example hie ' this ' and quis ' who.'Whichever of these two kinds you have taken, it mustnot be compared with the other, because they belongunfair here, since derivation by suflBxes, though varied, is notwithout its regular principles.§ 17. " Nouns, pronouns, adjectives (except participles).' Finite verbs. ' Participles. "* Adverbs.547

VARROreliquo non conferendum, quod inter se dissimileshabent analogias.19. In articulis vix adumbrata est analogia etmagis rerum quam voeum ; in nomin(at>ibus^ magisexpressa ac plus etiam in vocibus ac (syllabarum)"similitudinibus quam in rebus suam optinet rationem.Etiam illud accedit ut in articulis habere analogiasostendere sit difficile, quod singula sint verba, hiecontra facile, quod magna sit copia similium nominatuum.Quare non tam banc partem ab ilia' dividendumquam illud videndum, ut satis sit verecundi*etiam illam in eandem arenam vocare pugnatum.20. Ut in articulis duae partes, finitae et infinitae,sic in nominatibus^ duae, vocabulum et nomen :non enim idem oppidum et Roma, cum oppidum sitvocabulum, Roma nomen, quorum discrimen in hisreddendis rationibus alii discernunt, alii non ; nossicubi opus fuerit, quid sit et cur, ascribemus.*21. Nominatui^ ut similis sit nominatus, haberedebet ut sit eodem genere, specie eadem, sic casu,exitu eodem' : specie," ut si nomen est quod conferas,cum quo conferas sit nomen ;genere,* ut non solum(unurn sed>* utrumque sit virile ; casu,* ut si alterumsit dandi, item alterum sit dandi ; exitu, ut quas§ 19. ^ L. Sp., for nominibus. ^ Added by GS.^ After ilia, Aug. deleted ab. * Kent, for uerecundi.§ 20. ^ L. Sp., for uocabulis. * Sciop., for ascribimus.§21. ^ Mtie., for nominatus {Sciop. changed the secondnominatus to -tui). * Mue., for eius. * Liibbert, forgenere, transposing with specie (note 4). * Liibbert, forspecie (c/. preceding note) ; after this, L. Sp. deleted simile.* Added by Mve. ; sed added by Aug. * After casu, L.Sp. deleted simile.§ 21. " Here, as often in Varro, including adjective as wellas substantive.548

ON THE LATIN LANGUAGE, X. 18-21to schemes of Regularity which are different fromeach other.19. In the articles, Regularity is hardly even ashadow, and more a Regularity of things than ofspoken words ; in nouns, it comes out better, andconsummates itself rather in the spoken words andthe likeness of the syllables than in the thingsnamed. There is also the additional fact that itis difficult to show that Regularities reside in thearticles, because they are single words ; but in nounsit is easy, because there is a great abundance of likename-words. Therefore it is not so much a matterof di\-iding this part from that other part, as of seeingto it that the investigator should be too muchashamed even to call that other part into the samearena to do battle.20. As there are two groups in the articles,the definite and the indefinite, so there are in thenouns, the common nouns and the proper namesfor oppidum town and Roma ' ' ' 'Rome;are not thesame, since oppidum is a common noun, and Romais a proper name. In their account of the systems,some make this distinction, and others do not ;but we shall enter in our account, at the properplace, M'hat this difference is and why it has cometo be.21 That noun may be

VARROunum habeat extremas litteras, easdem alterumhabeat.22. Ad hunc quadruplicem fontem ordines derigunturbini, uni transversi, alteri derecti, ut in tabulasolet in qua latruncul«s* ludunt. Transversi suntqui ab recto casu obliquideclinantur, ut albus albialbo ; derecti sunt qui ab recto casu in rectosbus senis.declinantur, ut albus alba album ; utrique sunt parti-Transversorum ordinum partes appellantur*casus, derectorum genera,^ utrisque inter seimplicatis forma.*23. Dicam prius de transversis. Casuum vocabulaalius alio modo appellavit ; nos dicemus, quinominandi causa dicitur, nominandi vel nominativum.. . .^HIC DESUNT TRIA FOLIA IN EXEMPLARI^24. . . . (dicuntur una)e^ scopae, non dicitur unascopa : alia enim natura, quod priora simplicibus,§ 22. ^ Bentinus, for latrunculus. ^ Aldus, for expellantur.^ Aug., for genere. * Aug., for fonna.m.§ 23. ^ There is blank space here in F, for the rest of thepage {IS lines), all the next page {39 lines), and the first partof the following {8 lines). ^ F^, in margin.§ 24. ^ Added and altered by Kent, for et ; cf. viii. 7.§ 22. " The'men ' in a game like draughts or checkerswere called latrunculi brigands by ' ' the Romans. *Varrodid not arrange his paradigm of adjectives as we do, but setthe cases of the same number and gender in one line acrossthe page, while the other genders followed in the next twolines, and then the three genders of the plural in the succeedinglines.- " Varro counts his six genders by consideringthe genders of the plural as additional genders.''§ 23. " The cases. \'arro's names for the remaining550

ON THE LATIN LANGUAGE, X. 21-24ever last letters the one has, the other also have thesame.22. To this fourfold spring two sets of Unes aredrawn up, the ones crosswise and the others vertical,as is the regular arrangement on a board on whichthey play with movable pieces." Those are cross-\iise which are the oblique cases formed from a nominative,Uke ^* albus white,' genitive ' albi, dative albo ;those are vertical which are inflected from onenominative to other nominatives, as masculine albus,feminine alba, neuter album. Both sets of hnes areof six members. Each member of the crosswise'^Unes is called a case ; each member of the verticalUnes is a gender ; that which belongs to both in theircrossed arrangement, is a form.23. I shall speak first of the crosswise lines."Scholars have given various sets of names to thecases ; we shall call that case which is spoken for thepurpose of naming, the case of naming or nominative... *HERE THREE LEAVES ARE LACKING IN THE MODELCOPY "24. . . . To indicate one broom the ' ' plural scopaeis used, not the singular scopa.'^ For they ^ aredifferent by nature, because the names first mencases,which were listed in the lost text, are : e

VARROposteriora in coniunctis rebus vocabula ponuntur, sicbigae, sic quadrigae a coniunctu dictae. Itaque nondicitur, ut haec una lata et alba, sic una biga, sedunae bigae, neque^ dicitur ut hae duae latae, albae,sic hae duae bigae et quadrigae, munis, ut legolege.* Refert igitur ex quibus litteris quodque verbumconstet, maxime extrema, quod ea in plerisquecommutatur.*26. Quare in his quoque partibus similitudines abaliis male, ab aliis bene quod solent sumi in casibusconferendis, recte an perperam videndum ; sedubicumque commoventur litterae, non solum eaesunt animadvertendae, sed etiam quae proxumaesunt neque moventur : haec enim vicinitas aliquantumpotes^ in verborum declinationibus.27. In quis figuris non ea similia dicemus quae* After neque, p and Sciop. deleted ut. * Added by L. Sp.,cf. ix. 64.§ 25. ^ Mue., for uno. ^ Mue. added the signs ofquantity ; cf. ix. 104. ' Aug., for cursu. * Aug., forlege. * L. Sp. for commutantur.§ 26. ^ Aldus, for potes." These are all lost. ^ Scopae, as ' twigs ' done in a bundlebigae and quadrigae, because of the number of horses involved.' The distributive numeral is used to multiplyideas whose singular is denoted by a plural form: cf. ix. 64.§ 25. " I have added the signs of quantity in lego and legi,to make clear Varro's point.552

ON THE LATIN LANGUAGE, X. 24-27tioned are set upon simple objects, and those mentionedlater apply to compounded objects ^ ; thus"^higae two-horse team and quadrigae four-horse' ' 'team 'are employed in the plural because they denotea union of objects. Therefore we do not say one biga,like one lata broad and alba ' ' ' white,' but one bigae,with the numeral also in the plural ; nor do we sayduae two with ' reference to bigae and quadrigae, ' aswe say duae two with ' ' application to the pluralforms latae and albae, but we say binae two ' sets ' ofbigae and quadrigae.^25. Likewise the character of the form of a wordis important, because in the form of the spoken worda change is sometimes made in the first part ofthe word, as in suit sews and ' ' suit sewed ' ' ; sometimesin the middle, as in curso ' I run to and fro,'and cursito, of the same meaning ; sometimes at theend, as in doceo ' I teach and docui ' ' I have taught ' ;sometimes the change is common to two parts, as inligo ' I read,' legi ' I have read.'" It is importanttherefore to observe of what letters each word consists; and the last letter is especially important,because it is changed in the greatest number of instances.26. Because of this, since the likenesses in theseparts also are wont to be used in the comparison ofcase-forms, and this is done ill by some and well byothers, we must see whether this has been done rightlyor wrongly. Yet wherever the letters are altered,not only the altered letters must be noted, but alsothose which are next to them and are not affected ;for this proximity has considerable influence in theinflections of words.27. Among these forms we shall not call those553

VARROsimilis res significant, sed quae ea forma sint, uteius modi res simili*^ ex instituto significare plerumquesole

ON THE LATIN LANGUAGE, X. 27-28words like which denote like things, but those whichare of such a stamp that such forms * are in mostinstances wont by custom to denote like things, asby a man's tunic or a woman's tunic we mean not atunic that a man or a woman is wearing, but onewliich by custom a man or a woman ought to wear.^For a man can wear a woman's tunic, and a womancan wear a man's, as we see done on the stage byactors ; but we say that that is a woman's tunic,which is of the kind that women customarily useto dress themselves in. As an actor may wear awoman's dress, so Perpenna and Caecina and Spurinnaare said to have names that are feminine in form ;they are not said to have women's names. "^28. The likeness of the inflection also must bewatched, because the way which some words take isclear from the very words from which their inflectionstarts,* as how it is proper to use praetor and consul,dative praetori and consult. Others are properlyappreciated only as a result of the change seen in theinflections, as in socer ' father-in-law ' and macer'lean,' because the one becomes socerum in theaccusative, and the other macrum ; after makingthis change, each of them follows its o^\•n way in theremaining foi-ms,* both in the inflections of thesingular and in those of the plural. This method isemployed '^ because in the inflections there are twokinds of natures which can be compared with eachother, one which can be seen in the word itself, suchas homo man ' 'and equus ' horse,' but the secondcannot be seen through without bringing in someobliquecase rather than in the nominative; c/. ix. 91-94.* Varro's logical sequence is here at fault, for he brings inderivative stems, after speaking only of noun declensions.555

VARROextrinsecus perspici non possit, ut eques et equiso :uterque enim dicitur ab equo.29. Quare hominem homini similem esse aut nonesse, si contuleris, ex ipsis homini

ON THE LATIN LANGUAGE, X. 28-30thing from outside, as in eques horseman ' 'and equiso'stable-boy '—for both are derived from equus*horse.'**29. By this method, you will, on making a comparison,know that of men observed in person one is or isnot Hke the other; but you could not say that thetwo are in like fashion taller than their brothers, ifyou should not know how tall those shorter brothersare with whom they are compared. In this way theUkenesses of things broader and higher, and othersof the same kind, cannot be examined without bringingin some help from outside. So therefore, inasmuchas certain case-forms are of this kind, it is not easy tosay that they are like, if you observe the spoken wordsin one case only ; to make a correct judgement, youwill have to bring in another case-form to which thespoken word passes as it is inflected.30. These considerations are what I have thoughtenough to touch upon, for obserWng the likenesses ofnouns. It remains to speak of the articles, of whichsome are like nouns and others are different. For ofthe five classes the first two have the same properties,because they have forms for masculine, feminine, andneuter, they have some forms to denote the singularand others to denote the plural, and they have fivecases ; the vocative is not indicated by a separatespoken form. They have this of their own, thatsome are definite, Uke hie 'this,' feminine haec, andothers are indefinite, like quis which,' feminine'quae. But since their system of Regularity isshadowy and thin, it is not necessary to speakfurther of it in this book."" C/. viii. 14.§ 30. • C/. X. 19-20.557

VARRO31. Secundum genus quae verba tempora habentneque casus, sed^ habent personas. Eorum declinatuumspecies sunt sex : una quae dicitur temporalis,ut legebam gemebam, lego" gemo ; altera personarum,ut sero meto, seris metis ; tertia rogandi, utscribone legone, scribisne legisne. Quarta respondendi,ut fingo pingo, fingis pingis ;ut dicerem facerem, dicam faciam ;ut cape rape, capito rapito.quinta optandi,sexta imperandi,32. Item sunt declinatuum species quattuor quaetempora habent sine personis :in rogando, ut foditurneseriturne, et fodieturne sereturne. Ab respondendispecie eaedem figurae fiunt extremissyllabis demptis ; opandi species, ut vivaturametur, viveretur amaretur. Imperandi declinatussintne habet^ dubitationem et eorum sitne" haecratio :paretur pugnetur, parator pugnator.'33. Accedunt ad has species a copulis divisionumquadrinis :'ab infecti et perfecti, (ut)^ emo edo, emi§ 31. ^ Aug., for si. " For logo.§ 32. ^ Aug., for sum ne habent. - Aug., for sint ne.Canal, for parari pugnari.§ 33."1 Added by L. Sp.§31. " Cf. X. 17. * Respectively tense, person, interrogative(indicative), declarative indicative, subjunctive,imperative ; the technical vocabulary was not fully developedin Varro's time.§ 32. " Corresponding to the last four of the categories in§ 31 ; Varro shows a good understanding of the impersonalpassive.§33. " Cf. X. 14-17.558

ON THE LATIN LANGUAGE, X. 31-3331. The second subdivision" consists of thosewords which have tenses but not cases, and havepersons. The categories of their inflections are six * :one which is that of the tenses, as legeham ' I wasreading,' gemebam ' I was groaning,' lego ' I read,'gemo ' I groan ' ; the second is that of the persons, assero ' I sow,' 7neto ' I reap,' seris'thou sowest,'metis ' thou reapest ' ; the third is the interrogative,as scribone do ' I write ? ', legone do ' I read ? ', scribnne,legisne ; the fourth is that of the answer, as Jingo ' Iform,' pingo ' I paint,' Jingis, pingis ; the fifth that ofthe ^\^sh, as dicerem would ' I were saying,' facerem'would I were making,' dicam may ' I say,' faciam' 'may I make ; the sixth that of the command, ascape ' take,' rape ' seize,' capita, rapito.32. Like^^ise there are four categories of inflectionswhich have tenses \dthout persons " : in theinterrogative, as foditume ' is digging going on ? ',seriturne ' is sowing going on ? ' and fodieturne ' willdigging be done ? ', sereturne ' will sowing be done ? 'of the category for the answer the same forms areused, but without the last syllable ne ; the categoryfor the wish, as vivatur may there be ' living,' ametur'may there be lo\ing,' viveretur would there were'lixing,' atnaretur would there were lo\ing. ' ' Whetherthe inflections for the impersonal command exist, issomewhat doubtful ; there is also doubt about thescheme of the forms, which is given as paretur ' letthere be preparation,' pugnetur ' let there be fighting,'or parator, pugnator.33. There are added to these categories thosewhich proceed from the four sets of pairs ° consistingof the divisions : from that of the incomplete andthe completed, as emo ' I buy and ' edo ' I eat,' emi * I559

VARROedi ; ab semel et saepius, ut scribo lego, scriptitolectito^ ;(a)' faciendi et patiendi, ut uro ungo, urorungor ; a singular! et multitudinis, ut laudo culpo,laudamus culpamus. Huius generis verborum cuiusspecies exposui quam late quidque pateat et cuiusmodi efficiat figuras, in libris qui de formulis verborumerunt diligentius expedietur.34. Tertii generis, quae declinantur cum temporibusac casibus ac vocantur a multis ideo participalia,sunt hoc ge^ . . .HIC DESUNT FOLIA III IN EXEMPLARI^35. . . . quemadmodum declinemus,^ quaerimuscasus eius, etiamsi siqui^ finxit poeta aliquod vocabulumet ab eo casu(m)' ipse aliquem perperam declinavit,potius eum reprehendimus quam sequimur.Igitur ratio quam dico utrubique, et in his verbis quaeimponuntur et in his quae declinantur, neque nonetiam tertia ilia, quae ex utroque miscetur genere.36. Quarum una quaeque ratio coUata cum altera^ Added by L. Sp.§ 34. ^ Added by Rhol. ; F here leaves blank the rest of^ L. Sp.,for scriptitaui lectitaui.the page {a little more than 28 lines) and all the next page{39 lines). ^ F^, in margin.§ 35. ^ L. Sp., for declinamus. * L. Sp., for is qui.' L. Sp., for casu.^ Verbs. " Not extant.§ 34. " Adjective to the more common term participia orparticiples ; both meaning taking part ' ' in the features oftwo sets of words (nouns and verbs). For the form participalia(in F) rather than -pialia (in jd), cf. M. Niedermann,Mnemosyne, Ixiii. 267-268 (1936)." The lost text containedthe discussion of participles, that of adverbs, and the beginningof that on ratio.§ 35. " This is perhaps the simplest way of giving a meaningto the incomplete sentence. " Referring to the previousdiscussion, now almost entirely lost. " The independent560

ON THE LATIN LANGUAGE, X. 33-36have bought and ' edi ' I have eaten ' ; from that ofthe act done once and the act done more often, asscribo ' I write and ' lego ' I read,' scriptito ' I ambusy with writing,' and lectito ' I read and reread ' ;from that of active and passive, as uro ' I burn and'ungo ' I anoint,' uror ' I am burned and ' u?igor ' Iam anointed ' ; from that of singular and plural, aslaudo ' I praise and ' culpo ' I blame,' laudamus we'praise and culpamus we ' blame.' With regard ' to thewords of this class whose categories **I have described,the matter of how full an equipment of forms eachhas, and what sort of forms it makes, will be set forthwith more attention to detail in the books which"^are to be on the paradigms of verbs.34". The words of the third subdivision, whichare inflected with tenses and cases and are by manytherefore called participials," are of this kind . ..^HERE THREE LEAVES ARE LACKING IN THE MODELCOPY35. . . . When we meet a new word," we askabout its case-forms, as to how we shall inflect them ;and yet if some poet has made up some word and hashimself formed from it some case-form in an incorrectway, we blame him rather than follow his example.Therefore Ratio or Relation, of which I am speaking,is present in both ^ : in the words which are imposedupon things,*^ and in those which are formed by inflection

VARROaut similis aut dissimilis, aut saepe verba alia, ratioeadem, et nonnunquam ratio alia, verba eadem.Quae ratio in amor amori, eadem in dolor dolori,neque eadem in dolor dolorem, et cum eadem ratioquae est in amor et^ amoris sit in amores et amorum,tamen ea, quod non in ea qua oportet confertur*materia, per se solum efficere non potest analogiaspropter disparilitatem vocis figurarum, quod verbumcopulatum singulare^ cum multitudine : ita cum estpro portione, ut eandem habeat rationem, turndenique ea ratio conficit id quod postulat analogia ;de qua deinceps dicam.III. 37. Sequitur tertius locus, quae sit ratiopro portione ;^ esse in his, de utroque : siccum dicimus eandem rationem habere assem ad§ 36. ^ After et, a repeated amor et has been deleted.* After confertur, Aug. deleted a. ^ Aug., for singularem.§ 37. ^ L. Sp., for agrece. ^ Aug., for uocantur.• GS. ; analogia Mue., with G ; for analoga.§38. ^C F. W. Mueller, for simile. ^ Aug., forsimilitudine.§ 36. ° Because of the difference in number.§ 37. " As in mathematics, two ratios of equal value makea proportion.§ 38. " In the comedy of Plautus.562

ON THE LATIN LANGUAGE, X. 36-38compared with another, is either like or unhke ; andoften the words are different but the relation is thesame, and sometimes the relation is different but thewords are the same. The same relation which is inamor ' love ' and dative amori is in dolor pain ' ' anddative dolori, but not in dolor and accusative dolorem.The same relation which is in amor and genitiveamoris is in plural amores and genitive amorum ; andyet, because the subject-matter in it is not comparedas it should be," this relation cannot of itself effectRegularities, on account of the differences in theforms of the spoken word, because a singular wordhas been associated with a plural. So, when it is bya proportionate likeness that the word has the samerelation, then and not until then does this relationachieve what is demanded by Analogia or Regularity ;of which I shall speak next.in. 37. There follows the third topic : What isRatio or Relation that is pro portione 'by proportionatelikeness ' ? This is in Greek called ' according tologos ' ; and from analogue the term Analogia orRegularity is derived. If there are two things of thesame class which belong to some relation though insome respect unlike each other, and if alongsidethese two things two other things which have the samerelation are placed," then because the two sets ofwords belong to the same logos each one is saidseparately to be an analogue and the comparison ofthe four constitutes an Analogia.38. For it is as in a matter of twins : when we saythat the one Menaechmus is like the other Menaechmus,"we are speaking of one only ; but when we saythat a likeness is present in them, we are speaking ofboth. So, when we say that a copper as has the same563

VARROsemissem quam habet in argento' libella ad simbellam,*quid sit dvdXoyov ostendimus ; cum utrubique dicimuset in acre et in argento esse eandem rationem,turn dicimus de analogia.39. Ut sodalis et sodalitas, civis et civitas non estidem, sed utrumque ab eodem ac coniunctum, sicdvdXoyoi' et dvaAoyta idem non est, sed item est congeneratum.Quare si homines sustuleris, sodalissustuleris ; si sodalis, sodalitatem : sic item si sustulerisAoyov, sustuleris dvdkoyov ;si id, dvakoyiav.40. Quae cum inter se tanta sint cognatione, debebissuptilius audire quam dici expectare, id est cumdixero quid de utroque et erit comune, ^expectes, dun^ ego in scribendo transferam in reliquum,sed ut potius tu persequare animo.41. Haec fiunt in dissimilibus rebus, ut in numerissi contuleris cum uno duo, sic cum decem viginti :nam (quam)* rationem duo ad unum habent, eandemhabent viginti ad decem ; in nummis in similibus sicest ad unum victoriatum denarius, si* ad alterumvictoriatum alter denarius ; sic item in aliis rebusomnibus pro portione dicuntur ea, in quo est sicquadruplex natura, ut in progenie quom(odoy estfilius ad patrem, sic* est filia ad matrem, et ut est in' Pius, for argumento. * Pius, for singulas.§ 40. * Added hy Sciop.§41. ^ Added by Aldus. ^ Aug., for si. ^ Mue.,for cum. * After sic, Aug. deleted si.*A silver coin of the same value as the copper as.§ 41. " The quinarius, marked with a figure of Victory,and worth half a denarius.564

ON THE LATIN LANGUAGE, X. 38-tlrelation to a half-a* as a silver lihella * has to a halflihella,then we are showing what an analogue is ;when we say that both in copper and in silver thereis the same relation, then we are speaking of Analogia'Regular Relation.'39. As sodalis fellow and sodalUas ' ' ' fellowship,'civis ' citizen and ' civitas citizenship ' ' are not thesame, but both come from the same origin and areconnected, so analogue and Analogia are not thesame, but are like^^^se congenitally connected.Therefore, if you take away men, you have taken awaythe sodales ; if you take away the sodales, you havetaken away the sodalitas : just so, if you take awaythe logos or Relation, you have taken away theanalogue, and if you have taken this away, you havetaken away the Analogia.40. Since these are of such close kinship to eachother, vou must listen \n\h keen understandingrather than wait to be told, that is, when I have saidsomething about either, it A\ill be also of generalapplication to both ;you should not wait for me torepeat it in writing in a later part of my work, but youshould rather continue to follow up the line ofthought.41. These phenomena are produced in unlikethings, as in numbers, if you compare two \\ith oneand so also twenty with ten ; for twenty has to tenthe same relation which two has to one. It is foundalso in like things ; in coins, for example, one denariusis to one victoriate " as a second denarius is to a second\'ictoriate. So likewise in all other things those aresaid to be in a status of comparative hkeness, whereinthere is a fourfold nature of such a kind as amongchildren the daughter is to the mother as the son is to565

VARROte(m)poribus meridies ad diem, sic media nox adnoctem.42. Hoc poetae genere in similitudinibus utunturmultum, hoc acutissime geometrae, hoc in orationediligentius quam alii ab AristarcAo grammatici, utcum dicuntur pro portione similia esse amorem amori,dolorem dolori, cum ita dissimile^ esse videant amoremet amori,* quod est alio casu, item dolorem dolori, seddicunt, quod ab similibus.43. Nonnunquam rationes habet implicatas duas,ut sit^ una derecta, altera transversa. Quod dico,apertius sic fiet. Esto sic expositos esse numeros,ut in primo versu sit unum duo quattuor, in secundodecern viginti quadraginta, in tertio centum ducentiquadringenti. In hac formula numerorum duoinerunt quos dixi logoe, qui diversas faciant analogias: unus duplex qui est in obliquis versibus, quodest ut unus ad duo, sic duo ad quattuor ; alter decemplexin directis ordinibus, quod est ut unum ad decern,sic decem ad centum.44. Similiter in verborum declinationibus estbivium, quod et ab recto casu (declinantur in obliquoset ab recto casu)^ in rectu{m),^ ita ut formulam similiterefficiant, quod sit primo versu hie albus, huicalbo, huius albi, secundo haec alba, huic albae, huiusalbae, tertio hoc album, huic albo, huius albi. Itaque§ 42. ^ For dissimilem. * Christ ; amori Canal, omittinget ,•for et dolorem.§ 43. 1 For sic.§ 44. ^ Added by Mue. (obliquom Mue. ; obliquos L.Sp.). * Mue., /or recto.§ 42. " The rhetorical figure. * That is, proportionally.566

ON THE LATIN LANGUAGE, X. 4-1-4.4,the father, and in matters of time the midnight is tothe night as the midday is to the day.42. The poets make a great use of this kind ofrelationship in their similes," and the geometriciansuse it \s'ith greatest keenness ; in reference to speech,Aristarchus and the grammarians of his school use itwith more care than others do, as when ace. amoremand dat. amori, ace. dolorem and dat. dolori are said tobe like by comparative likeness,^ although they seethat amorem is unlike amori because it is in anothercase, and likewise dolorem is unlike dolori ; but theysay that the four are like, because they come fromlike words.43. Sometimes it has two crossed relationships,in such a way that one is vertical and the othercrosswise. What I mean will become clearer bythis : Suppose that some numerals are so set downthat in the first line there are 12 4in the second line there are 10 20 40in the third line there are 100 200 400In this scheme of numerals there will be two examplesof what I have called logos, which make differentsystems of Regularity : one is the twofold which is inthe crosswise lines, because two is to four as one is totwo ; the other is the tenfold relation in the verticallines, because ten is to one hundred as one is to ten.44. Likewise the inflections of words may go in twodirections, because from the nominative case they areinflected into the oblique cases, and from the nominativeto the nominative, so that they make a similarscheme ; which isin line 1 : masc. nom. albus, dat. albo, gen. albi ;in line 2 : fem. nom. alba, dat. albae, gen. albae ;in line 3 : neut. nom. album, dat. albo, gen. albi.567

VARROfiunt per obliquas declinationes ex his analogiae hoegenus Albius Atrius, Albio Atrio, quae scilicet eritparticula ex ilia feinaria,' per directas declinationesAlbius Atrius, Albia Atria,* quae seiHcet rfenaria*formula analogiarum, de qua supra dixi.45. Analogia quae dicitur, eius genera sunt duo :unum deiunctum sic est : ut unum ad duo sic decernad viginti ; alterum coniunctum sic : ut est unum adduo, sic duo ad quattuor. In hoc quod duo bis dicunturet turn (cum)^ conferimus ad unum et tunc cum* quattuor,46. hoc quoque natura dicitur quadrwplex^ ; sice septem cAordis^ c/tharae tamen duo dicuntur haberetetracAorda,* quod quemadmodum crepat prima adquartam c^ordam,^ sic quarta ad septumam respondet,media est alterius prima, alterius extrema. Medici'in oegroto* septumos dies qui observant, quarto dieideo dili^entius signa morbi advertunt, quod quamrationem habuit primus dies ad quartum eandempraesagit habiturum qui est futurus ab eo quartus,qui est septumus a primo.47. Quadruplices deiunctae in casibus sunt vocabulorum,ut rex regi, (lex legi>,^ coniunctae sunt triplicesin verborum tribus temporibus, ut legebam*lego legam, quod quam rationem habet legebam ad^' quae . . . binaria was transposed to this position by Mue.,from its position after albia atria ; binaria Mue., for vicenaria.*After the transposition {note 3), albia atria atriaquae is left ; the second atria is deleted. * L. Sp., forcentenaria.§ 45. 1 Added by L. Sp. * Added by Sciop.§ 46. ^ For quadriplex. * The h inserted by Aug.For midici. * L. Sp., for egrotos.^§ 47. ^ Added by Mue. * After legebam, Aug. deletedab. ' Aug., for ab.568

ON THE LATIN LANGUAGE, X. 44-47Therefore by the cross^^ise inflections there are madefrom these words systems of Regularity Uke Albiusand Atriiis, Albio and Atrio, which to be sure is only asmall part of that binary scheme ; and by the verticalinflections are made Albius and Atrius, Alhia andAtria, which is a part of the tenfold scheme of Regularitiesof which I have spoken above.45. Of that which is called Regularity, there aretwo kinds. One is disjoined, thus : as one is to two,so ten is to twenty. The other is conjoined, thus :as one is to two, so two is to four. Because in it twois said twice, both when we compare it with one, andthen when we compare it with four,46. this kind also is said to be fourfold by nature.So the cithers, though with seven strings, are none theless said to have two sets of four strings, because justas the sound of the first string stands in a certainrelation to that of the fourth, so the fourth stands inthe same relation to the seventh ; the middle stringis the first of the one set and the last of the other.The doctors who watch the seven days when a man isill," note the symptoms of the illness vnXh greater careon the fourth day, for the reason that the relationwhich the first day had to the fourth, foretells that theday which \\\\\ be fourth from it, that is, seventh fromthe first, will bear the same relation to the fourth.47. The Regularities are disjoined and fourfold inthe cases of nouns, such as rex 'king,' dative regi, andlex ' law,' dative legi ; they are conjoined and threefoldin the three tenses of verbs, such as legebam 'I wasreading,' present lego, future legam, because the relationwhich legebam has to lego, this same relation lego§ 46. « In recurrent fevers ; specifically the quartanague.569

VARROlego hanc habet lego ad legam. In hoc fere omneshomines peccant, quod perperani in tribus temporibushaec verba dicunt, cum proportione volunt pronuntiare.48. Nam cum sint verba alia infecta, ut lego etlegis, alia perfecta, ut legi et legisti, et debeant suicuius(que>i generis in coniungendo copulari, et cumrecte sit ideo lego ad'^ legebam, non recte est lego adlegi, quod legi significat quod perfectum : ut haectutudi pupugi, tundo pungo, tundam pungam, item/zecatus' sum verberatus sum, (necor verberor.*necabor^) verberabor, iniuria reprehendant,* quodet infecti inter se similia sunt et perfecti inter se, uttundebam tundo tundam et tutuderam tutudi tutudero; sic amabar amor amabor, et amatus eramamatus sum amatus ero. Itaque ' reprehenduntqui contra analogias dicunt, cur dispariliterin tribus temporibus dicantur quaedam verba, naturacum quadruplex sit analogia.49. Id nonnunquam, ut dixi, pauciores videturhabere partes, sic etiam alias pluris, ut cum est :quemadmodum ad tria unum et duo, sic ad sex duoet quattuor, quae tamen quadripertito^ comprehenditurforma, quod bina ad singula conferuntur ;quod in oratione quoque nonnunquam reperietur sic :§ 48. ^ quoiusque H,for cuius F, V, p. ^ Aug., for et.3 Aug., for haec catus. * Added by Miie. ^ Added byAug. *">J/k^., /or reprehendunt. Added by Christ.§ 49. ^ GS., for quadripertita.§ 47. " That is, past action is to present action, as presentis to future.570

ON THE LATIN LANGUAGE, X. 47-49has to legam.'^In this, almost all men make a mistake,because they cite these verbs wrongly in the threetenses, when they wish to express them in a proportion.48. For since some verbs denote incomplete action,like lego ' I read and ' legis thou ' readest,' and othersdenote completed action, like legi ' I have read and'thou hast read,' and since in the conjoinedlegisti 'form they ought to be connected vrith others of theirown kind and by this principle lego is rightly relatedto legebam—lego is not rightly related to legi, becauselegi denotes something completed ; so that they arewrong in finding fault with tutudi ' I have pounded 'and pupugi ' I have pricked,' tiindo and pungo, tundamand ptmgam, as well as necahis sum 'I have been killed 'and verberatus sum ' I have been beaten,' necor andverberor, necabor and verberabor, because the tenses ofincomplete action are like one another, and thoseof completed action are like one another. Thuswe should say tundebam tundo tundam, and tutuderamtutudi tutudero, and in the same way amabar amoramabor, and amatus eram, amatus sum, amatus ero.Therefore those who speak against the Regularitiesare unfair in finding fault on the ground thatwhereas Regularity is fourfold by nature certainwords are cited in a different way, in three tenseformsmerely.49. This seems sometimes to have fewer parts, as Ihave said ; similarly it seems, at other times, to havemore parts, as when it is thus : as one and two areto three, so two and four are to six. Yet this form isincluded in the fourfold type, because sets of two arecompared with sets of one. In speech also, this willsometimes be found, thus : As nominative Diomedes571

VARROut Diomedes confertur Diome(di et Diome>dis,* sicdicitur ab Hercules Herculi et Herculis.*50. Et ut haec^ ab uno capite ac recto casu in duoobliquos discedunt casus, sic contra multa ab duobuscapitibus recti casuis^ confluunt in obliquom unum.Nam ut ab his rectis hi biei, hae Baebiaefit his Baebieis, sic est ab his hi Caelii, hae CaeHaehis CaeUis. A duobus simiUbus similiter' dechnantur,ut fit in his nemus Aolus, nemora Aolera.AUa ab dissimiUbus similiter declinantur, ut in articulisab hie iste, hunc istura

ON THE LATIN LANGUAGE, X. 49-52compared with genitive Diomedi and Diomedis, soisfrom nominative Hercules are said the genitive formsHerculi and Herculis.'^50. And as these move away from one startingpointand nominative into two obhque case-forms,so on the other hand from two starting-points of thenominative many words unite in a single obUquecase-form. For as from the nominatives Baebiet(masc.) and Baehiae (fem.) comes the dative Baebieis,"so from the nominatives Caelii and Caeliae comesCaeliis. From two like words forms are developedin unlike fashion, as happens in nemus ' grove ' andholus ' vegetable,' plural nemora and holera. Othersfrom unlike words are developed in like fashion, as inthe articles the accusatives hunc and istunc come fromhie ' this ' and iste ' that.'51. Regularity has its foundations either in thewill of men or in the nature of the words, or in both.By will I mean the imposition of the word-names :by nature I mean the inflection of the words, through\\hich passage is made without special instruction.He who starts froni the imposition, will say that ifdolus ' guile and malus ' '' bad are alike in the nominative,there will be found in an oblique case dolo and7nalo. He who starts from the nature of the words,will say that if Marco and Quinto are alike in theoblique cases, there will be nominatives Marcus andQuintus. He who proceeds from both, will say that ifthere is a likeness, then as the change is in servus'slave ' and vocative serve, so also there will be cervus' stag ' and vocative cerve. It is a common feature ofthat the four word-forms have their inflectionalall,changes in a proportional relation.52. The first kind starts from the likeness in the573

VARROrectis casibus, secundum ab similitudine quae est inobliquis, tertium ab similitudine quae est intransitibusde casu in casum. Primo genere ab impositoad naturam profieiscimur, in secundo contra, in tertioab utroque, Quocirca etiam hoc tertium potestbifariam divisum tertium et quartum dici, quod in eovel prosus et n/sus* potest dici.53. Qui initia faciet analogiae impositiones, abhis obUquas figuras declinare debebit ;qui naturam,contra ;qui ab utraque, reliquas declinationes abeiusmodi transitibus. Impositio est in nostro dominatu,nos in natura^ : quemadmodum enimquisque volt, imponit nomen, at declinat, quemadmodumvolt natura.54. Sed quoniam duobus modis imponitur vocabulumaut re singulari aut multitudine, singulari, utcicer, multitudinis, ut scalae, nee dubium est, quinordo declinatuum, in quo res singulares declinabuntursolae, ab singulari aliquo casu proficiscatur,^ ut cicerciceri ciceris, item contra in eo ordine, qui multitudiniserit solum, quin a multitudinis a(li>quo casuordiri conveniat, ut scalae scalis scalas :aliud videndumest, cum duplex natura copulata acdeclinatu(u)mbini fiant ordines, ut est Mars Martes, unde turn ratioanalogiae debeat ordiri, utrum ab singulari re inmultitudinem an contra.55. Neque enim si natura ab uno ad duo pervenit,§ 52. ^ For rosus.§ 53. * Stephanus, for natura.§ 54. ^ Stepha7iu,i, for proficiscantur.574

ON THE LATIN LANGUAGE, X. 52-55nominatives, the second from a likeness which is inobhque cases, the third from a Hkeness which is inthe changes from case to case. In the first kind weset out from the imposed name to the nature, in thesecond we go in the other direction, in the third wego in both directions. Therefore in fact this thirdcan be divided into two parts and called the thirdand the fourth, because in it the argument can actuallygo both forward and backward.53. He who makes the imposed forms the startingpointfor the Regularity, will have to develop theoblique forms from these ; he who makes the naturethe starting-point, will have to work in the otherdirection ; he who starts from both, will have to makethe rest of the inflections from the changes of thesame kind. The imposition is in our power, but we areunder the control of the nature of the words : for eachone imposes the name as he wishes, but he inflects itas its nature requires.54. But since a noun is imposed in two wavs, eitheron a singular thing or on a plural—singular like cicer'chickpea,' plural like scalae ' stairs '—and there isno doubt that the hne of the inflections whereinthings which are singular only ^\ill be declined,proceeds from some case of the singular, as cicerciceri ciceris ; and likewise that in the line of inflectionswhich is in the plural only, it is proper to beginfrom some case of the plural, as scalae scalis scalas :another point must be examined, since their connectednature is twofold and two lines of inflectionsare made, like Mars and Martes, namely from whatplace the relation of Regularity ought to start,whether from the singular to the plural or vice versa.55. For not even if nature does proceed from one575

VARROi\)ais'nature ' as an originating ormoving power. * Properly, of sounds.§ 56. " Prineipia are the singular forms, in whicheverdirection the argument is carried ; but perhaps quam insingulari should be inserted between ordiri and quod.* Because the B and the C ending the stems can be seen in the576

ox THE LATIN LANGUAGE, X. 55-57to two, should the conclusion be drawn that in teachingthe later thing cannot be the clearer, for thepurpose of beginning from it, to show what the priorthing is. Therefore even those who deal with thenature of the universe and are on this account calledphysici " *natural philosophers,' proceed from natureas a whole and show by backward reasoning from thelater things, what the beginnings of the world were.Though speech consists of letters,* it is neverthelessfrom speech that the grammarians start in order toshow the nature of the letters.56. Therefore in the explanation, since one oughtrather to set out from that which is clearer thanfrom that which is prior, and rather from the uncorruptedthan from a corrupt original, from thenature of things rather than from the fancy of men,and since these three factors which are more to befollowed are less present in the singulars than in theplurals, one can more easily commence from theplural than from the singular, because in the latteras starting-points " there is less of a basis for relationshipin the forming of words. That the singularforms of words can be more easily interpreted fromplural forms than plural forms from the singular, isshown by these words ** : plural trahes ' beams,' singulartrabs ;plural duces ' leaders,' singular dux.57. For we see that from the plural nominativestrahes and duces the letter E of the last syllable hasbeen eUminated and thereby in the singular have beenplural, but cannot be inferred with certainty from the nominativesingular, especially if we read not trabs but traps(Roth, Philol. xvii. 176, and Mueller's note to § 57), whichrepresents the actual pronunciation. Yet \'arro wrote trabsand not traps, according to Cassiodorus, Gram. Lat. vii.159. 23 Keil.VOL. 11 p 577

1VARROlari factum esse trabs dux. Contra ex singularibusnon tam videmus quemadmodum facta sint ex B et Strabs^ et ex C et S dua'.*58. Si mMl(t>itudinis^ rectus casus forte figuracorrupta erit, id quod accidit raro, prius id corrigemusquam inde ordiemur ;(ab>* obliquis adsumereoporte^' figuras eas quae non erunt ambiguae, sivesingulares sive multitudims,* ex quibus id, cuius modidebent esse, perspici possit.^59. Nam nonnunquam alterum ex altero videtur,ut Chr^sippus scribit, quemadmodum pater ex filioet filius ex patre, neque minus in fornicibus proptersinistram dextra stat quam propter dextraw^ sinistra-Quapropter et ex rectis casibus obliqui et ex obliquisrecti et ex singularibus multitudim's^ et ex multitudinissingulares nonnunquam recuperari possunt.60. Principium id potissimum sequi debemus, utin eo fundamentum sit^ natura, quod in declinationibusibi facilior ratio. Facile est enim animadvertere,peccatum magis cadere posse in impositioneseas quae fiunt plerumque in rectis casibus singularibus,quod homines imperiti et dispersi vocabula rebusimponunt, quocumque eos libido invitavit : natura§57. ^ Aug., for trahes. ^ Aug., for duces,§58. ^ '^si multitudinis 3/m^., /or similitudinis. Addedby Canal. ' L. Sp., for oportere. * Aug., for multitudines.^ Sciop., for possint.§59. ^ La^^w*, /or dextras. ^ Vertranius, for mu\titadines.§ 60. 1 After sit, L. Sp. deleted in.§ 59. " Frag. 155 von Arnim.578

ON THE LATIN LANGUAGE, X. 57-60made the nominatives trabs and dux. But on theother hand, if we start from the singulars we do notso easily see how they have become trabs, from Band S, and dux, from C and S.58. If the nominative plural is by any chance acorrupted form, which rarely occurs, we shall correctthis before we make it our starting-point ; it is properto take from the oblique cases, either singular orplural, some forms which are not ambiguous, fromwhich can be seen the make-up which the other formsought to have.59- For sometimes the one is seen from the otherand at other times the other is seen from the one, asChrysippus writes," as the father's qualities may beseen from the son, and the son's from the father, andin arches the right-hand side stands on account of theleft-hand side, no less than the left on account oftheright. Therefore the oblique forms can sometimesbe regained from the nominatives, and sometimes thenominatives from the oblique forms ; sometimes theplural from the singular forms, and sometimes thesingular forms from the plural.60. The principle that we should most of all follow,is that in this the foundation be nature, because innature " there is the easier relationship in inflections.For it is easy to note that error can more easily makeits way into those impositions * which are mostlymade in the nominative singular, because men, beingunskilled and scattered,*^ set names on things just astheir fancy has impelled them ; but nature ** is of§ 60. " Rather than in voluntas. * Or imposed wordnames,characterized by voluntas. ' For this point of theStoic philosophy, cf. Cicero, de Inventione, i. 2.**ThequaHty underlying the paradigms.579

VARROincorrupta plerumque est suapte sponte, nisi quiearn usu inscio depravaftit.61. Quare si quis principium analogiae potiusposuerit in naturalibus casibus quam in positiciis,^non multa' (inconcinna)' in consuetudine occurrentet a natura libido humana corrigetur, non a libidinenatura, quod qui impositionem sequi voluerintfacient contra.*62. Sin ab singular! quis potius proficisci volet,ini^ium^ facere oportebit ab sexto casu, qui est propriusLatinus : nam eius casuis' litterarum discriminibusfaeilius reliquorum varietate(m>' discernerepoterit, quod ei habent exitus aut in A, ut hacterra, aut in E, ut hac lance, aut in I, ut hac

ON THE LATIN LANGUAGE, X. 60-64itself for the most part uncorrupted, unless somebodyperverts it by ignorant use.*61. Therefore, if one has founded the principleof Regularity on the natural cases rather than on theimposed case-forms, not many awkwardnesses >\ill behis to face in usage ; human fancifulness will be correctedby nature, and not nature by fancy, becausethose who have \\ished to follow imposition will inreaUty act in the opposite wav.62. But if one should prefer to start from thesingular, he ought to start from the sixth case," whichis a case peculiar to Latin ; for by the differences in theletters * of this case-form he will be more easily able todiscern the variation in the remaining cases, becausethe ablative forms end either in A, like terra 'earth,'or in E,'' like lance ' platter,' or in I, Uke clavi ' key,'or in O, like caelo ' sky,' or in U, like versu ' verse.'Therefore, for the explaining of the declensions, thereis this way, which may proceed from either of twostarting-points.63. But where there is Regularity, there are threefactors, one which is in the things, a second which isin the spoken words, a third which is in both ; the firsttwo are simple, the third is twofold. In \-iew of this,attention must be given to the relation which theyhave to one another.64. First, of the differences which exist in thethings, there are some which have no bearing onspeech, others which are connected with it. Thosewhich are not connected with it are like those whichthe artificers observe in making buildings and statues\'arro's time. * That is, the endings. ' Varro does notlist separately the ablative of the fifth declension, ending inlong E.581

VARROartifices, e quis vocantur aliae ^armonicae, sic itemaliae nominibus aliis : sed nulla harum fit ^loqiiendo pars.^65. Ad orationem quae pertinent, res eae suntquae verbis dicuntur pro portione neque a similitudinequoque vocum declinatus habent, ut lupiter Marspiter,lovi Marti. Haec enim genere^ nominum etnumero et casibus similia sunt inter se, quod utraqueet nomina sunt et virilia sunt et singularia et casunominandi et dandi.66. Alterum genus vocale est, in quo voces modosunt pro portione similes, non res, ut biga bigae,nuptia nuptiae : neque enim in his res singularissubest una, cum dicitur biga quadriga, neque ab hisvocibus quae declinata sunt, multitudinis significantquicquam, id^ quod omnia multitudinis quae declinanturab uno, ut a merula merulae : sunt (enim)^eius modi, ut singulari subiungatur, sic merulae duae,catulae tres, faculae quattuor.67. Quare cum idem non possit subiungi, qxiod^(non>* dicimus biga una,' quadrigae duae, nuptiaetres, sed pro eo unae bigae, binae quadrigae, trinaenuptiae, apparet non esse a biga et quadriga* bigaeet quadrigae, sed ut est huius ordinis una* duae tres§ 64. ^ Added by L. Sp. ^ Sentence division of Boot.§ 65. ^ Mue., for genera.§66. 1 Fay, /or ideo. ^ Added by Fay.§ 67. ^ Sciop., for cum. ^ Added by Sciop. ' L.Sp. ; una biga Sciop. ; for bigae unae. * After quadriga,L. Sp. deleted et. ® Aug., for unae.§ Q5. " The unlikeness is in the forms of the nominative :but both words denote male deities.§ Q6. " The two words belong to the same declension andboth lack the singular forms ; but the objects denoted areentirely unlike.582

ON THE LATIN LANGUAGE, X. 64-67and other things, of which some are called harmonic,and others are called by other names ; but no one ofthese becomes an element in speaking.65. The differences which pertain to speech,consist of those things which are expressed by thewords in a proportionate way, and yet do not have alikeness of the spoken words also to help in formingthe inflections : such as nominative lupiier andMarspiter, dative lovi and Marti.'* For these are Ukeone another in the gender of the nouns, and in thenumber, and in the cases ; because both are nouns,and are masculine, and singular, and nominative anddative in case.66. The second kind has to do with the sounds,in which the spoken words only are similar ina proportionate way—and not the things—as inbiga and bigae, nuptia and nuptiae.'* For in thesethere is no underlying unit thing expressed bythe singular when we say biga or quadriga, norhave the plural forms which are derived from thesewords any plural meaning. Yet all plurals whichare derived from a unit singular, like merulae frommerula blackbird,' do have such plural meaning ' ;for they are of such a sort that there is subordinationto a singular form : thus two merulae blackbirds,'three catulae female puppies, four faculae'''torches.'67. Therefore since there cannot be the same subordinatingrelation because we do not say una biga,duae quadrigae, tres nuptiae, but instead unae bigae'one two-horse team,' binae quadrigae two teams of'four horses,' trinae nuptiae three sets of 'nuptials,' it isclear that bigae and quadrigae are not from biga andqtiadriga, but belong to another series : the usual583

VARROprincipum una, sic in hoc ordine altero unae binaetrinae principium est unae.68. Tertium genus est illud duplex quod dixi, inquo et res et voces similiter pro portione dicuntur utbonus malus, boni mali, de quorum analogia et Aristophaneset alii scripserunt. Etenim haec deniqueperfecta ut in oratione, illae duae simplices inchoataeanalogiae, de quibus tamen separatim dicam, quodhis quoque utimur in loquendo.69- Sed prius de perfecta, in qua et res et vocesquadam similitudine continentur, cuius genera sunttria : unum vernaculum ac domi natum, alterumadventicium, tertium nothum ex peregrino hie natum.Vernaculum est ut sutor et pistor, sutori pistori ;adventicium est ut Hectores Nestores, HectorasNestoras ; tertium ilium nothum ut Achilles et Peles.70. De (his primo)^ genere multi utuntur nonmodo poetae, sed etiam plerique omnes qui solutaoratione loquuntur. Haec primo^ dicebant ut quaestorempraetorem, sic Hectorem Nestorem ; itaqueEnnius ait :Hectoris natum de muro tactari.'§ 70. ^ Added by A. Sp. " Lachmann transferred haecprimo from its position after plerique. * So Scaliger(moero) for Hectoris natum de Troiano muro lactari F ; themisquoted line must be made metrical to establish the longo in Hectoris.§ 68. ° Page 267 Nauck; the grammarian of Byzantium,see V. 9, note a.§ 69. " The type is marked by the o in the stem. * Thetype is marked by the o in the stem, and the short vowels inthe case endings -es and -as, for both of which native Latinhad -es. "With -es replacing the normal -evs of the Greeknominatives (though Arcadian and some other dialects had' -j]s) ; the standardized Latin had Achilles, but Peleus.584

ON THE LATIN LANGUAGE, X. 67-70series una, duae, ires, has una as its beginning, but inthis second series unae, bifiae, irinae, the beginning isunae.68. The third kind of Regularity is that which hastwo elements, which I mentioned, in which both thethings and the spoken words are uttered with asimilarity in a proportionate way, like bonus good ''and malus bad,' plural boni and mali ' ; Aristophanes" and others have written about the Regularityin such words. And indeed this is a perfected Regularityin speech, but those two simple forms of Regularityare only incomplete beginnings ;yet I shallspeak of them separately, because we use them alsoin speaking.69- But first I shall speak of the perfected Regularity,in which both the things and the spoken wordsare held together by a certain Ukeness ; of this thereare three kinds : one native, bom here among us ;the second coming from abroad ; the third hybrid,born here of foreign paternity. The native type issuch as sutor ' cobbler ' and pisior ' baker,' dativesutori and pisfori " ; the foreign tj'pe is such asHectares ' men like Hector ' and Xestores ' men likeNestor,' accusative Hectoras and XestorSs * ; thatthird type, the hybrid, consists of such words asAchilles and Peles.'70. Of these, many use the first type, not merelypoets, but also almost all who speak in prose. At firstthey used to say Hectorem and Xestorem like quaestoremand praetdrem ; so Ennius says ° :That Hector's * son be hurled from the Trojan wall.§ 70. " Andromache Aechmalotis, Traa. Rom. Frag. 93Ribbeck» ; Scsn. 82 Vahlen» ; R.O.L. i. 248-249 Warmington.* Hectoru, as the scansion shows.585

VARROAccius haec in tragoediis largius a prisca consuetudinemovere coepit* et ad formas Graecas verborum magisrevocare,^ a quo Valerius ait :Accius He harum parte(m)^worfque^ sit inferend2* et cuius modi, videndum.Nam, ut dixi, neque rerum neque vocis similitudo adhas duplicis qua** in loquendo quaerimus analogias* For caepit. * After revocare, L. Sp. deleted et. * Forhaetorem. ' V, for nominatos. ^ A. Sp., with a, forplurimus,§ 71. ^ The quantities of the voioels were added by Christ ;cf. Charisius, G. L. i. 148. 38 Keil, and Diomedes, G. L.i. 305. 13 Keil. * Sciop., for minores. * Aldus, forregentes.§ 72. ^ Kent, for in qua harum parte cumque. * GS.,for inferendo. * Mue., for quae."Trag. Rom. Frag., inc. fab. XXXVI Ribbeck» ; R.O.L.ii. xxiii, 364-365, 556-557, 599

ON THE LATIN LANGUAGE, X. 70-72Accius in his tragedies " began to take these wordsaway from the early usage and rather to restore themto their Greek forms ; hence Valerius says ** :Accius would not use Hectorem, but Hectora rather.Because most of our foreign words are Greek, it hasfollowed that the greatest number of the hybridnouns which we have are also Greek in origin. Therefore,as in these types some words are Greek andothers are Greek in origin, so also are the systems ofRegularity.71. Of the hybrid inflectional forms which aremade from these materials in our country, some areearly, like Bacchides and Chrysides,'^ others areyounger, like Chrysides and Bacchides,^ and still othersare recent, like Chrysidas and Bacchidds " ; our fellowcountrvmenuse all three, but those who follow themiddle forms in speaking give the least offence,because those of the first set seem insufficiently likethe Greek forms from which they are taken, andthose of the third seem insufficiently like our ownforms.IV. 72. The basis of all Regularity is a certainlikeness, that, as I have said, which is wont to be inthings and in spoken words and in both ; we must seein which one of these sections each word should beentered," and of what sort it is. For, as I have said,neither the likeness of the things nor that of thespoken words is separately sufficient to express thesedouble Regularities of the words, which we seek in§71. " Nom.-acc, with Latin ending. "Norn., withGreek ending, but diflFering from Latin only in the quantityof the vowel. * Ace, with Greek ending.§ 72. "A bookkeeping idiom.587

VARROverborum exprimendas* separatim satis est, quodutraque parte opus est simili.ut perducas aceedere debet usus :Quas ad loquendumalia enim ratio quafacias vestimentum, alia quemadmodum utare vestimento.73. Usui(s)^ species videntur esse tres : una consuetudinisveteris,altera consuetudinis huius, tertianeutra(e>.'' Vetera, ut cascus casci, *urus *uri*;huius consuetudinis, ut albus caldus, albo caldo ;neutrae, ut scala scalam, phalera joAaleram.* Adquas aceedere potest quarta mixta, ut amicitia inimicitia,amicitiam inimicitiam.Prima est qua usi antiquiet nos reliquimus, secunda qua nunc utimur,tertia qua utuntur poetae.74. Analogia^ non item ea definienda quae derigiturad naturam verborum atque ilia quae ad usumloquendi. Nam prior definienda sic : analogia estverborum similium declinatio similis, posterior sic :analogia est verborum similium declinatio similisnon repugnante consuetudine co(m>muni. At quom^harum duarum ad extremum additum erit hoc " ex* analogias verborum exprimendas, after simili in F, was sethere by A. Sp. (after duplicis, by Mue.).§ 73. ^ L. Sp. ; usus Aug. ; for usui. ^ Canal, forneutra. ' Scaliger, for furus furl. * For falera faleram.§ 74. ^ Aug., for analogiae. ^ Sciop., for ad quam.* That is, the ' regular ' form may be constructed, but it mustalso be found in use to have any value.588

ON THE LATIN LANGUAGE, X. 72-74speaking, because there must be a likeness in bothrespects. To introduce them into speech there mustbe also actual use ; for the method by which youmake a garment is quite different from that in whichyou wear it.*73. The categories of use appear to be three :one that of old usage, the secondthat of to-day's usage,the third that of neither. Old words are such ascascus casci ' old,' sums suri ' stake ' ; words of to-day'susage, such as albus ' white,' caldus " ' hot,' datives alboand caldo ; words of neither usage, such as scala andace. scalam 'stair,' phalera and phaleram 'trapping.'''To these there can be added a fourth kind which doesnot belong exclusively to one category, like amicitiaenmity,' accusatives' friendship ' and inimicitia*amicitiam and itiimicitiam ." The first is that whichthe ancients used and we have abandoned ; the secondis that which we now use ; the third is that which thepoets use.74-. That Analogia or Regularity which is directedtoward the nature of the words

VARROHarumquadam parte," poetica analogia erit definita.primam sequi debet populus, secundam omne**singuli e populo, tertiam poetae.75. Haec diligentius quam apertius dicta essearbitror, sed non obscurius quam de re simili definitionesgrammaticorum sunt, ut Aristeae, Aristodemi,Aristocli, item aliorum, quorum obscuritates eo minusreprehendendae, quod pleraeque definitiones re incognitapropter summam brevitatem non facileperspiciuntur, nisi articulatim sunt explicatamuni, quid exquadam parte.77. Verbum dico orationis vocalis* partem, quaeSi declinationem naturalemsit indivisa et^ minima.habeat, simile^ verbum verbo tum quom* et re* quamsignificat et voce^ qua significat et' in figura e transitudeclinationis parile. Declinatio est, cum ex verboin verbum aut ex verbi discrimine, ut transeat mens,vocis commutatio fit aliqua.^ Aug., for omnem.§ 75. ^ Sciop., for explicata.Similitudo declinationis,§ 77. ^ Aug., for vocabulis. ^ For eo. ^ Sciop., forsimilem. * Aug., for quoniani. * Mue., for rem.* Mue., for vocem. ' Aug., for est.* Cf, § 78 ; the poets, while having certain privileges, are yetunder some restrictions.§ 75. " Apparently followers of Aristarchus. Varroseems to have had a book containing a collection of definitionsof analogia, arranged according to an alphabetical listof their authors.590

ON THE LATIN LANGUAGE, X. 74-77Analogia will be defined.'' The first of these is thatwhich the people ought to follow ; the second isthat which all the individuals in the people ought tofollow ; and the third is that which the poets oughtto follow.75. I think that these things have been said withmore care than clarity, but not more obscurely thanare the definitions of the same subject given by thegrammarians, such as Aristeas, Aristodemus, Aristocles,and others," whose obscurities are the less to befound fault with, because most definitions, being onan unknown theme and being expressed -with, extremebrevity, are not easily understood unless they areexpounded point by point.76. Therefore the matter will be more apparent ifthere is a clear exposition of the parts one by one, asto what is meant by a word, what is meant by thelikeness of the word, by inflection, by likeness ofinflection not inconsistent with common usage, andby " within a certain range."77. By word I mean that part of spoken speechwhich is the smallest indivisible unit. If a word hasnatural inflection," then a word is like another wordwhen it is similar to the other word in the thingwhich it denotes and in the spoken word by whichit denotes the thing and in the form which it has afteran inflectional change has taken place. Inflection *•is that which takes place when some change of thespoken word is made from word-form to word-formor to a new word-stem by derivation, in order toexpress a change of the thought. Likeness of in-§ 77. " That is, inflection rather than stem-derivation ;both these ideas are included under declinatio. * Includingderivation.591

VARROcum item ex aliqua figura in figuram transit, ut idtransit, cum quo confertur.78. Adiectum est " non repugnante consuetudineco(m>muni," quod quaedam verba contra usumveterem inclinata patietur, ut passa Hortensiumdicere pro hae^ cervices cervix, quaedam non, ut sidicas pro fauces faux. Ubi additur " ex quadamparte," significat non esse in consuetudine in hisverbis omnis partis, ut declinatum ab amo* ^ivo amor

ON THE LATIN LANGUAGE, X. 77-80flection exists, when it passes from some form toanother form in the same way in which that otherword passes with which it is being compared.78. There is the addition " not inconsistent withcommon usage," because usage tolerates some wordsinflected contrary to the old practice, as it sufferedHortensius " to say cervix neck ' ' instead of the pluralcervices, but does not tolerate certain others, as whenyou should say faux throat instead of the plural' 'fauces. When the addition " ^\-ithin a certain range "is made, it means that in the relevant words not all theforms are in use, as, for example, there is derived fromamo ' I love and ' vivo ' I Uve the ' passive amor butnot the passive vivor.V. 79- What Analogia or Regularity in speech isseen to be and what categories it has, and which ofthese seem essential to follow, I have set forth asbriefly as I could. Now I shall speak of the categoriesin which it ought not to exist and yet it is usuallylooked for just as if it ought to be there ; these arein general of four kinds. First, Regularity oughtnot to be looked for in such words as are not inflected,for example nequam 'worthless,' viox 'soon,'vix 'hardly.'"80. Among these, a greater error is made inone word than in another. For they grant thatmox and vix have no cases, but assert that nequamhas, because we use it with nominative hie ' this,'with genitive huius, with dative huic. For when wesay hie nequam and huius nequam, then we are utteringthe cases of this man whom we wish to showas worthless, and before the word we then set hieto represent the name of him whose worthlessnesswe are considering.VOL. II Q 593

VARRO81. Quod vocabulum factum ut ex non et volonolo sic ex ne et quicquam item media extrita syllabacoactum est nequam. Itaque ut eum quern putamusesse non hili dicimus nhili,* sic in quo putamus essene quicquam dicimus nequam.82. Secundo, si unum solum habent casum invoce, quod non decline(n>tur,^ ut litterae omnes.Tertio, si singularis est vocabuli series neque habetcum qua comparari possit, ut esse putant caput capiticapitis capite. Quartum, si ea vocabula quattuorquae conferuntur inter se rationem (non>* habentquam oportet, ut socer socrus, soceros socru*.'VI. 83. Contra in quibus debeat quaeri analogia,fere totidem gradus debent esse coniuncti :primumut sint res/ secundum ut earum sit usus, tertium utihae res vocabula habeant, quartum ut habeant declinatusnaturalis. De primo gradu, quod naturasubest et multitudinis et singularis, dicimus hiasseshosce as*es,* hie as hunc assem ; contra quod innumeris finitis multitudinis natura singularis non est,dicitur hi duo et hi tres, his duobus et his tribus.84. Secundo gradu si est natura neque est usus,§ 81, 1 For anhili.§ 82. ^ Laetus, for declinetur. ^ Added by Mue.* Mue., for socerum.§ 83. ^ Aug., for tres. ^ For asces.§ 81. " From n? and volo. * From ne and quam." From 71^ and hili ; cf. ix. 54.§ 82. » Cf. ix. 51-52. > Cf. ix. 53.594

ON THE LATIN LANGUAGE, X. 81-8481. This word is made like nolo'* 'I do notwish from non not ' ' ' and volo ' I wish '; thus fromne 'not' and quicquam ' anything,' \\ith loss of themiddle syllable, is likewise compounded nequam.^So as him whom we think to be non hili worth'not a whit we ' call nihili,'^ him in whom we thinkthat there is ne quicquam not anything we ' ' callnequam.82. Second , Regularity is not to be looked for ifthewords have only one case in their spoken form, becausethey are not inflected, like all names of letters."Third, it is not to be looked for if the series of formswhich the noun has is unique and has nothing withwhich it can be compared, as they consider true ofcaput head,' ' dat. capiti, gen. capitis, abl. capiteJ*Fourth, it is not to be sought if those four noun-formswhich are compared with one another fail to have themutual relation which they should have, as in socer'father-in-law and ' socrus mother-in-law,' accusa-'tive plural soceros and socrus.VL 83. On the other hand, in words in whichRegularity ought to be looked for, in general thesame number of stages should be found in conjunction: first, the things should exist ; second, thethings should be in use ; third, these things shouldhave names ; fourth, they should have natural inflection.As for the first stage, because the nature ofplural and singular is basic, we say plural nom. asses,ace. asses, singular nom. as, ace. assern ; on the otherhand, because in definite plural numerals the singularnature does not exist, only plural forms are used,such as nominative duo 'two' and tres 'three,' dativeduobus and tribus.84. In the second stage, if the nature exists but595

VARROid genus ut sit discriminandum, ut fit in faba et idgenus, quae item et ex parte et universa nominamus :non enim opu

ON THE LATIN LANGUAGE, X. 84-there is no practice of making this kind of distinction,as happens in faba ° ' bean ' and in that class of wordswhich we use for one and for all collectively, withoutchange of form : for there was no need, as in thematter of slaves ... *collective meaning. * The thought seems to have been,that as one name was sufficient for a slave since he had nofamily name (c/. ix. 59), so one form was adequate for suchwords as faba, whether denoting an individual thing or manyas a collective unit.597

M. TERENTI VARRONISDE LINGUA LATINALiBRORUM II-IV FrAGMENTAIIIFr. 1.^ Deus autem vel dea generale nomen est omnibus.. . . Varro ad Ciceronem tertio ": Ita respondeantcur dicant deos, cum ^ omnibusantiqui dixerint divos."Fr. 2.^ Figor ambigue declinatur apud veterestempore perfecto. Reperimus enim fictus et fixus :. . . Varro ad Ciceronem tertio " fixum."Fr. 3.^ Hoc nomen licet veteres Latinum negent,auctoritate tamen valet. Dicebant enim leonemmasculum et feminam. Leam vero Varro ad. . .Ciceronem dicit libro III :" Sicut now est" pantheraet lea."Fr. 1. 1 Serv. Dan. in Aen. xii. 139. ^ Added by Thilo.Fr. 2. ^ Diomedes, Ars Grammatica, i. 377. 11-13 Keil.Fr. 3. ^ Philarg. in Buc. ii. 63. " For nocest.II.-IV. » Cf. also V. 1 ; V. 6 ; vii. 109-110 ; viii. 1 ; Frag.21.Fr. 1. " The word split into two paradigms, as a result ofextensions from deus, divi, divo, deum, etc., which has comeabout by regular phonetic development.598

M. TERENTIUS VARRO'SON THE LATIN LANGUAGEFragments of Books II-IV "IIIFr. 1 . Deus god ' ' or dea goddess ' ' is in fact ageneral name for all. . . . \'arro, in the third bookof the treatise addressed to Cicero, says " : So letthem give answer why they say del gods,' when in'reference to all of them the old-time Romans usedto say divi.'^"Fr. 2. Figor ' I am transfixed ' is by the old «Titersinflected in two ways in the perfect tense. For wefind both Jictus and Jixus ; . . . \'arro in the thirdbook of the treatise addressed to Cicero has "Jlxum."Though the old \^Titers say that the nameFr. 3.leaena ' lioness ' is not good Latin, still it has theforce of authority. For they used to say leo Uon ' 'both as masculine and as feminine. . . . But lea'lioness ' ^'arro has, in the third book of the treatiseaddressed to Cicero : " Just as panthera ' panther 'and lea ' lioness ' are not." "Fr. 3. " That is, are not similarly formed from a correspondingmasculine ; c/. v. 100, ix. 56.599

VARROIVFr. 4.^ Varro de Lingua Latina lib. IIII : " ProluftiMm*et lubidinem' dici ab eo quod lubeat : undeetiam lucus Veneris Lubentina(e> dicatur."LiBRi VIII FragmentumFr. 5.^ (5) M. Varronis liber ad Ciceronem deLingua Latina octavus nullam esse observationemsimilium docet inque omnibus paene verbis consuetudinemdominari ostendit :(6) " Sicuti cum dicimus,"inquit, " lupus lupi, probus probi et lepus leporis, itemparo paravi et lavo lavi, pungo pupugi, tundo tutudiet pingo pinxi. (7) Cumque," inquit, " a ceno etprandeo et poto et cenatus sum et pransus sum etpotus sum dicamus, a destringor tamen et extergeoret lavor destrinxi et extersi et lavi dicimus.(8) " Item cum dicamus ab Osco Tusco GraecoFr. 4. 1 Nonius Marcellus, 64. 15-17 M. * Guiet, forproluuiem. ' Quicherat, for prolubidinem.Fr. 5. ^ Aldus Gellius, Noct. Att. ii. 25. 5-10; the textfollows Rolfe's, in the Loeb Classical Library.Fr. 4. < Cf vi, 47.VIII. " See also vii. 90, ix. 53, 66, 79, 80, 81, 89, 90, 96,97, 100, 101, 104, 105, 108, 111, which refer to passages inVIII. that are not extant.Fr. 5. " The translation is essentially Rolfe's, in the LoebClassical Library. 'Cf. especially viii. 25. "Cf. viii.34, ix. 91.

FRAGMENTS, 4-5IVFr, 4. Varro in the fourth book of the treatise Onthe Latin Language ": Prolubiiim and lubido'desire *are derived from lubet ' it is pleasing ' ; whence alsothe grove of Venus Luhentina gets its name." "Fragment of Book VIII

VARROOsce Tusce Graece, a Gallo tamen et Mauro Galliceet Maurice dicimus ; item a probus probe, a doctusdocte, sed a rarus non dicitur rare, sed alii raro dicunt,alii rarenter."(9) Idem M. Varro in eodem libro ": Sentior,"inquit, " nemo dicit et id per se nihil est, adsentiortamen fere omnes dicunt. Sisenna unus adsentioin senatu dicebat et eum postea multi secuti, nequetamen vincere consuetudinem potuerunt."(10) Sed idem Varro in aliis libris multa pro dva-Aoyi'ci tuenda scribit.LiBRORUM XI-XXIV FragmentaXIFr. 6.^ Et ubi auctoritas maiorum genus tibi non demonstraverit,quid ibi faciendum est ? Scripsit Varroad Ciceronem ": Potestatis nostrae est illis rebusdare genera, quae ex natura genus non habent."Fr. 7a. ^ Nunc de generibus dicamus. Varro dicit" genera dicta a generando. Quicquid enim gignitaut gignitur, hoc potest genus dici et genus facere."Fr. 6. ^ Julianus Toletanus, Commentarius in Donatum,V. 318. 31-34 Keil.Fr. 7. ^ [Serffii] Explanat. in Donatum, iv. 492. 37-493. 3Keil.* Charisius, i. 217. 8 Keil, cites rare as used by Cicero,Cato, and Plautus {Rudens 995) ; but editors usually replace'it by raro. That is, not a deponent unless compoundedeven in a passive meaning, the passive form of the uncompoundedverb is rare, though occasionally found, as inCaesar, Bellum Civile i. 67 {sentiretur), where it is howeverimpersonal. ' Notably in ix.602

FRAGMENTS, 5-7aand Graece ' in Greek,' yet from G alius Gaul ' ' andMaurus Moor ' ' we have Gallice ' in Gallic ' andMaurice ' in Moorish ' ; also from prohus honest'comes probe honestly,' from doctus ' ' learned ' docte' 'learnedly,' but from rarus rare there ' is noadverb rare, but some say raro, others rarenter." *(9) In the same book A'arro goes on to say : " Noone uses the passive sentior,^ and that form by itself isnaught, but almost every one says adsentior 'I agree.'Sisenna alone used to say adsentio in the senate, andlater many followed his example, yet could notprevail over usage."(10) But this same Varro in other books' wrote agreat deal in defence of Regularity.Fragments of Books XI-XXIV"XIWhere the authority of our ancestors has notFr. 6.sho-wn you the gender of a word, what in this instancemust be done ? \'arro \\Tote, in the treatise addressedto Cicero ": We men have the right and power togive genders to the names of those things which bynature have no gender." "Fr. 7a. Now let us speak of genders, ^'arro says :" Genera ' genders ' are named from generare ' togenerate.' For whatever gignit begets or gignitur' ''is begotten,' that can be called a genus and canXI.-XXIV. « On Books XI.-XIII., see also vii. 1 10, viii. 2,20, 34, X. 33 ; and on Books XIV.-XXV., see vii. 1 10.Fr. 6. " Varro uses genus both for grammatical genderand for natural sex ; each is a ' kind ' or ' cleiss,' c/. Frag. 7,note a.603

VARROQuod si verum est, nulla potest res integrum genushabere nisi masculinum et femininum.Fr. 7b.* Tractat de generibus. Varro ait "generatantum ilia esse quae generant : ilia proprie dicunturgenera." Quodsi sequemur auctoritatem ipsius, nonerunt genera nisi duo, masculinum et femininum.Nulla enim genera creare possunt nisi haec duo.Fr. 8.^ Ostrea* si primae declinationis fuerit, sicutMusa, feminino genere declinabitur, ut ad anima/^referamus ; si* ad testam, ostreum* dieendum estneutro genere et ad secundam declinationem, ut sithuius ostrei, huic ostreo,* quia dicit^ Varro " nuUamrem animalem neutro genere declinari."Fr. 9-^ Ait Plinius Secundus secutus Varronem :" Quando dubitamus principale genus, redeamus addiminutionem, et ex diminutivo cognoscimus principalegenus. Puta arbor ignoro cuius generis sit :fac diminutivum arbuscula, ecce hinc intellegis etprincipale genus quale sit. Item si dicas columna,* Pompeius, Commentum Artis Donati, v. 159. 23-26 Keil.Fr. 8. ^ Cledonius, Ars Gramviatica, v. 41. 24-28 Keil.* For ostria. * Keil, for animam. * For sic. * Forostrium. * Keil, for sicui ostri. ' For dicitur.Fr. 9. ^ Pompeius, Commentum Artis Donati, v. 164. 13-18 Keil.Fr. 7. " The root gen- lies at the basis of all these words ;but genus has the weakened meaning ' kind, class,' fromwhich the idea of ' begetting ' has faded out. * Donatus,the eminent grammarian who flourished about 350 a.d.* That is,'kinds ' : cf. Frag. 6, note a.Fr. 8. " This distinction is not borne out by the use of thewords in the Latin authors. * Almost precisely true forLatin, though there are many exceptions in Greek and in theGermanic languages {cf. rixvov, German das Kind, and theneuter diminutives in -lov, -chen, -lein).604.

FRAGMENTS, 7a-9produce a genus." " If this is true, then the genusthat a thing has is not perfect unless it is masculineor feminine.Fr. 7b. He* treats of genders. Varro says :" Onlythose are genera genders which ' generant generate ' ' 'those are properly called genera." But if we followhis authority, there will be only two genders, masculineand feminine. For no genders can procreate^"except these two.Fr. 8. If osirea 'oyster' is of the first declension,'like Musa Muse,' it ^\^ll be declined in the femininegender, so that we refer the word to the U\ing being ;if we use it for the shell, then the word must beostreum, inflected in the neuter and according tothe second declension, so that it is genitive ostrei,dative ostreo " : because Varro says :" No Uvingcreature has a name which is inflected in the neutergender." *Fr. 9- Plinius Secundus " says, following Varro :" When we are in doubt about the gender of a mainword, let us turn to the diminutive form, and fromthe diminutive we learn the gender of the main word.*Suppose that I do not know the gender of arbor'tree ' ; form the diminutive arhuscula, and lo !from this you observe as well the gender of the wordfrom which it comes. Again, if you say, WTiat is theFr. 9. " This and subsequent citations from Pliny aretaken from the Elder Pliny's I>ubius Sermo, a work in eightbooks, mentioned by the Younger Pliny, Epist. iii. 5. 5.* Diminutives have in Latin the gender of the words fromwhich they are derived ; the exceptions are very few. InGreek and in the Germanic languages, however, diminutivesare commonly neuter without regard to their primitives ; ef.Frag. 8, note b.605

VARROcuius generis est ? facis inde diminutivum, id estcolumella, et inde intellegis quoniam principalefeminini generis est."Fr. 10.^ " //ypocorismata semper generibus suisundis haec ungula,h hoc pastillum," ut Varro dixit " : haec beta hiebetace(us, haec malva hie malvaceus), hoc pistrinumhaec pistrilla, ut Terentius in Ad(elphis, hie ensishaec ensicula et hie ensiculus) : sic in RudentePlautus."die intelligitur.'Fr. 11.^ Dies communis generis est. Qui masculinogenere dicendum putaverunt, has causas reddiderunt,quod dies festos auctores dixerunt, non festas,et' quartum et quintum Kalendas, non quartam neequintam, et cum hodie dicimus, nihil aliud quam hocQui vero feminino, eatholico utuntur,quod ablativo casu E non nisi producta finiatur,Fr. 10. 1 Charisius, Instit. Gram. i. 37. 13-18 Keil. Theright-hand edge of the mantiscript is destroyed, but the restorationsare made with certaintyfrom almost verbatim repetitionsCharisius i. 90. 10-12, 155. 14-17, 535. 21-25, 551. 36-38 Keil,in which Varro is not mentioned as the source. Hie pastillus,required by the spcu;e, was added by Keil from i. 90. 11, i.94. 4.Fr. 11. ^ Charisius, Instit. Gram. 1. 110. 8-16 Keil.* For ut. * For intellegatur.Fr. 10. " As substantive, for pes betaceus : but betaceus isan adjective, not a diminutive.''Also an adjective ; itsapplication as substantive is not known. ' Adelphoe 584.'Rudens 1156-U57.Fr. 11. " Dies was by origin a masculine ; in Latin,because it was declined like the feminines of the fifth declension,possiblj' also because its counterpart nox was606

FRAGMENTS, g-11gender of columna ' column ' ?, make from it thediminutive, that is, columella, and therefrom youunderstand that the word from which it comes is of thefeminine gender."Fr. 10. " Diminutives always agree in genderwith the words from which they come : a few differ,such as fem. rana 'frog,' diminutive masc. ranunculus'tadpole '; masc. unguis 'nail (of finger or toe),' fem.ungula ' hoof, talon ' ; neut. glandium ' kernel ofpork ', fem. glandula ' tonsil ' ; masc. pants ' loafof bread,' masc. pastillus and neut. pastillum "' roll,'as Varro said ; " fem. beta ' beet,' masc. betaceus "'beet-root ' ; fem. malva ' mallow,' masc. malvaceus ^'mallow-like vegetable ' ; neut. pistrinum ' pounding-mill,'fem. pistrilla ' small mill,' as Terence saysin The Brothers '^ ; masc. ensis ' sword,' fem. ensiculaand masc. ensiculus ' toy-sword ' : so Plautus in TheRope.'^ "Fr. 1 1 . Dies ' day ' is of common gender." Thosewho thought that it must be used as a masculine,offered these reasons : that their authorities said diesfesti 'hoHdays,' with the mascuHne adjective, not thefem. festae ; that they said the fourth and the fifth daybefore the Kalends, with the masculine and not the**feminine form of the adjective ;and that when wesay kodie ' to-day,' it is understood as hoc die ' on thisday,' %\-ith the masculine article," and nothing else.On the other hand, those who regard dies as feminine,use the general argument, that in the ablative thefeminine, it acquired use as a feminine in some meanings.* Full phrase : ante diem quartum (quintum) Kalendas.' The demonstrative was an ' article,' in the grammaticalterminology of the Romans ; cf. viii. 4a.607

VARROet quod deminutio eius diecula sit,non dieculus, utait Terentius :Quod tibi addo dieculam.Varro autem distinxit, ut* masculino genere uniusdiei cursum significare(t>, feminino autem temporisspatium ;quod nemo servavit.Fr. 12.^ Catinus masculino genere dicitur . . . ethinc deminutive catillus fit. . . . Sed Varro adCiceronem XI " catinuli " dixit, non catilli.Fr. 13.^ Ncevus generis neutri, sed Varro adCiceronem " hie naevus."Fr. 14a.^ Antiquissimi tamen et hie gausapes ethaec gausapa et hoc gausape et plurale neutri haecgausapa quasi a nominativo hoc gausapum protulisseinveniuntur, . . . Varro vero de Lingua Latina ait," talia ex Graeco sumpta ex masculino in femininumtransire et A litera finiri : o KoxAias haec cochlea,o x«P'''^5haec charta, 6 yava-aTrrj^ haec gausapa."Fr. 14b.* Varro autem ait " vocabula ex Graecosumpta, si suum genus non retineant, ex masculinoin femininum Latine transire et A littera terminari* For et,Fr. 12. 1 Charisms, Instit. Gram. i. 79. 23-80. 4 Keil.Fr. 13. 1 De Dubiis Nominibus, v. 584. 27 Keil.Fr. 14. ^ Priscian, Inst. Gram. ii. 333. 9-14 Keil.* Charisius, Inst. Gram. i. 104. 13-16 Keil.•*But this masculine diminutive was used in Oscan.' Andria 710. ^ That is, a short space of time, as arespite.608

FRAGMENTS, 11-Ubword ends in a long E, never in a short E ; and thatits diminutive is the feminine diecula (not the masculinedieculus **), as Terence has it * :That to you I give a daytime.'But Varro made the distinction, that in the masculineit means the course of one day, in the feminine aspace of time : a distinction to which nobody hasconformed in practice.Fr. 12. Catinus ' bowl ' is used in the masculinegender . . . and from it is made the diminutive catillus.. . . But Varro, in the eleventh book of his treatiseaddressed to Cicero, sp>onsored the form catinuU, andnot catilU.Fr. 13. Naevus ' mole, wart,' is of the neutergender " ; but ^'arro in the treatise addressed toCicero uses it as a masculine.Fr. 14a. Yet the oldest writers are found to haveemployed masculine gausapes ' cloth,' and femininegausapa, and neuter gausape, and a neuter pluralgausapa as if from a neuter nominative singular ga«*apum.. . . But Varro in his treatise On the LatinLanguage says " : Such words, when taken fromGreek, pass from the masculine to the feminine, andend in the letter A : fem. cochlea ' snail from masc.'Ko\Xtas, fem. charta paper ' ' from masc. X'^P^V'^'fem. gausapa from masc. yaiia-dirqs."Fr. 14b. But \'arro says ": Words taken fromGreek, if they do not keep their own gender, passfrom the masculine to the feminine in Latin and endin the letter A, like cochlea from Kox\ia

VARROvelut KoxAtas cochlea, ^^p/xijs herma, X^P''"'!'^ charta,ergo yavcraTrrys gausapa."Fr. 14c.' Margarita feminini generis est, quiaGraeca nomina -)/s terminata in A transeunt et fiuntfeminina, ut 6 x^P'''*/^ haec charta, /xapyapiVv/s margarita,aut communia, ut u^Av/riys athleta. Ergoneutrahter hoc margaritum dicere vitiosum est ; ettamen multi dixerunt, ut \ algius . . . et VarroEpistularum* VIII " margaritum unum, margaritaplura." Sed idem Varro saepe et ahi plures margaritafeminine dixerunt ; in genetivo tamen pluralinon nisi feminino genere margaritarum.Fr. 15.^ VAS terminata et SIS faciunt genitivo etDIS, hoc vas huius vasis ; utrumque Varro ait deLingua Latina : hie vas huius vadis.Fr. 16.^ VIS et ipsa tertiae sunt declinationis etsimilem nominativo faciunt genetivum, (hic)^ civishuius civis, haec vis huius vis et pluraU hae vis, sicutretius* et Varro : nam hae vires numero semperplurali declinantur.Fr. 17.^ M. Varronem et P. Nigidium, virosRomani generis doctissimos, comperimus non aliter' Charisuts, Inst. Gram. i. 108. 4-12 Keil. * Or Epistulicarum;c/. Charisius, i. 104. 21 Keil.Fr. 15. 1 Probus, Catholica iv. 30. 26-27 Keil.Fr. 16. 1 Probus, Catholica iv. 30. 30-31 . 2 Keil. « Additions/ramPriscian, ii. 24.9. 9-10 Keil.Fr. 17. 1 Auhis Gellius, Noct. Att. iv. 16. 1 ; Rolfe's text,in the Loeb Classical Library.Fr. 14. "A sculptured head, originally of a beardedHermes, at the top of a rectangular pillar. * On the title610

FRAGMENTS, 14b-17bust " from ' 'Epfirj^, charia from \dpTrj

VARROelocutos esse et scripsisse quam senatuis et domuis etfluctuis, qui est patrius casus ab eo quod est senatus,domus, fluctus ; huic senatui, domui, fluctui, ceteraqueis eonsimilia pariter dixisse.Fr. 18.^ Amni Maro,Secundo defluit amni.Ubi Plinius eodem libro " Ab antiquis " inquit " quosVarro reprehendit, observatio omnis ilia damnata est,non quidem in totum. Dicimus enim," inquit, " abhoc canali siti tussi febri. Maiore tamen ex parteforma mutata est. Ab hoc enim cane orbe carboneturre falce igne veste fine monte fonte ponte strigletegete ave asse axe nave classe dicimus."Fr. 19-* Quern Plinius ad eundem XI " rureordinatum arbustum "* dixisse laudat.Fr. 20.^ Fonteis* :" Quorum nominum genetivipluralis ante UM syllabam I litteram merebuntur,Fr. 18. ^ Charisius, Inst. Gram. i. 122. 23-29 Keil.Fr. 19. ^ Charisius, Inst. Gram. 1. 142. 20-21 Keil.* For ambustum.Fr. 20. ^ Charisius, Inst. Gram. i. 129. 19-24 Keil.* For fontis.Fr. 18. " Georg. iii. 447. * The rule, stated at i. 120.19-24 Keil, is that nouns with genitives ending in is have theablative in e, unless the genitive is identical with the nominative,when the ablative ends in i ; an adjective also has theablative in i if it stands before a noun which it modifies. Thescientific formulation is that consonant-stems should haveshort e in the ablative, and «-stems should have long / : astatus much disturbed by the encroachment of the «-endingon the t-ending.' Not all these should, by the'rule,'end in i ; for carbo, falx, mons, fans, pons, teges do not haveidentical nom. and gen. ; and the nom. of asse is as, veryrarely assis. As to the actual forms of the ablative, igni isorbi, turri,fini, strigili, avi, axi, navi,commoner than igne ;612

FRAGMENTS, 17-20said and wTOte senatuis, domuis, and Jiuctuis as thegenitive case of the words senatus ' senate,' domus'house,' and Jluctus wave,' and used ' senatui, domui,jiuctui as the dative ; and that they used other similarwords •with the corresponding endings.Fr. 18. Avini was used by \'ergil " as ablative ofamnis 'river,' as inHe drifts with the stream of the river.On this point, PUny in the same book says : "By theold wTiters, whom \'arro criticizes adversely, allobservance of the rule * is disregarded, yet notutterly. For we still say," says he, " canali ' canal,'siti ' thirst,' tussi cough,' fehri fever ' ' ' as the ablativeforms. But in most words the form has beenchanged, and uses the ablative which ends in E :cane ' dog,' orbe ' circle,' carbone charcoal,' ' turre'tower,' falce ' sickle,' igne ' fire,' veste garment,''Jine ' Umit,' monte mountain,' fonte ' ' spring,' ponte'bridge,' strigile ' scraper,' tegete ' mat,' ave ' bird,'asse ' as,' axe ' axle,' nave ' ship,' classe ' fleet.' " "Fr. 19. Varro, whom PUny mentions as havingsaid, in the eleventh book of his treatise addressed toCicero " a plantation of trees set in rows rure'^ ' inthe country.' "Fr. 20. Fonteis ' springs,' accusative plural spelledi»ith EIS ": The nouns which gain an I in the genitiveplural before the ending UM," says PUny, " have theclassi are found in authors of the first century b.c, but areless common than the forms with «, or are used to satisfymetrical requirements ; ponti is found once in older Latin ;monti and /on^t are cited by Varro, ix. 112.Fr. 19. " Instead of the usual locative form ruri.613

VARROaccusativus," inquit Plinius, " per EIS loquetur,montium monteis ; licet Varro," inquit, " exemplishanc regulam confutare temptarit istius modi, falciumfalces, non faleeis facit, nee has merceis, nee hos axeisItntreis ventreis stirpeis urbeis corfeeis' vecteis menteis.*Et tamen manus dat praemissae regulaeridicule, ut exceptis his nominibus valeat regula."Fr. 21.^Poematorum et in II et in III idem Varroadsidue dicit et his poematis, tam quam nominativohoc poematum sit et non hoc poema. Nam et adCiceronem XI, horum poematorum et his poematisoportere dici.Fr. 22.^ Git : Varro ad Ciceronem XI per omnescasus id nomen ire debere conmeminit ; vulgo autemhoc gitti dicunt.XIIIXIII dixit.Fr. 23.^ Palpetras per T Varro ad CiceronemSed Fabianus de Animalibus primo palpebrasper B. Alii dicunt palpetras genas, palpebrasautem ipsos pilos.' For curueis. * GS., for inepteis, c/. viii. 67.Fr. 21. 1 Charisius, Inst. Gram. i. 141. 29-31 Keil.Fr. 22. ^ Charisius, Inst. Gram. i. 131. 7-8 Keil.Fr. 23. ^ Charisius, Inst. Gram. i. 105. 14-16 Keil.Fr. 20. " This EI does not represent an earlier diphthong,but was often written for a long i after the original diphthonghad becomte identical in sound with the long i. There arescattered examples of the ending EIS in the accusative, foundin inscriptions and manuscripts.614

FRAGMENTS, 20-23accusative in EIS," like genitive montium mountains,''accusative monteis ; although \'arro," he continues," tried to refute this rule by examples of the follo^Ningsort : to the genitive ya/««/« ' sickles the accusative'is falces and not falceis, nor is the proper spelUngmerceis ' wares,' nor axeis ' axles,' lintreis'skiffs,'venireis'belUes,' stirpeis'stocks,' urbeis'cities,'corbels ' baskets,' vecteis'levers,' vienteis ' minds.'And yet he gives up the fight against the aforesaidrule in a ridiculous fashion, saying that apart fromthese nouns the rule holds."Fr. 21. In the second and the third books Varroconstantly uses the genitive poematorum ' poems ' andthe dative poematis, as though the word were poematumin the nominative and not poema. For in theeleventh book of the treatise addressed to Cicero hesays that genitive poematorum and dative poematis arethe proper forms to be used.Fr. 22. Git ' fennel ' " : \arro in the eleventhbook of the treatise addressed to Cicero states thatthis form ought to be used in all the cases ; butpeople quite coHunonly say gitii in the ablative,XIIIFr. 23. \'arro in the thirteenth book of the treatiseaddressed to Cicero used palpetrae, with T. ButFabianus," in the first book On Animals, wrote palpebraewith B. Others say that palpetrae means theeyeUds, and palpebrae the eyelashes.Fr. 22. " yigella sativa.Fr. 23. " Papirius Fabianus, who wrote on philosophyand on natural history in the time of Augustus.615

VARROFr. 24.^ Oxo " : Varro ad Ciceronem XIII olivoet oxo putat fieri," inquit Plinius Sermonis Dubiilibro VI.XVIIIFr. 25.^ Indiscriminatim, indifferenter. Varro deLingua Latina lib. XVIII " : Quibus nos in hoc libro,proinde ut nihil intersit, utemur indiscriminatim,promisee."XXIIFr. 26.^ Rure Terentius in Eunucho :Ex meo propinquo rure hoc capio commodi.Itaque et Varro ad Ciceronem XXII " rure veni."XXIIIFr. 27.^ Varro ad Ciceronem in libro XXIII :" ingluvies tori," inquit, " sunt circa gulam, quipropter pinguedinem fiunt atque interiectas habentrugas." Sed nunc pro gula positum.Fr. 24. 1 Charisnis, Inst. Gram. i. 139. 15-16 Keil.Fr. 25. ^ Nonius Marcellus, de Compendiosa Doctrina,127. 24-26 M.Fr. 26. 1 Charisius, Inst. Gram. i. 142. 18-20 Keil.Fr. 27. ^ Serv. Dan. in Georg. iii. 431.Fr. 24. " Antecedent unknown. * Greek o^o^ (neuter,third decl.), denoting sour wine, and vinegar made therefrom.Fr. 25. " Antecedent unknown.Fr. 26. " 971. * These are examples of rure as a pureablative. The continuation is our Fragment 19, in whichexamples of rure as a locative are discussed.Fr. 27. " That is, double chins.616

FRAGMENTS, 24^27Fr. 24. 0x0, ablative : " Varro, in the thirteenthbook of the treatise addressed to Cicero, expressesthe opinion that it " is composed of olive-oil and oxos *'vinegar,' " says Pliny in the sixth book of the treatiseentitled Variations in Speech.XVIIIFr. 25. Indiscriminatim means without ' difference.'Varro in the eighteenth book of the treatiseOn the Latin Language says :" Which " in this bookwe shall use indiscriminatim without ' distinction,'promiscuously, just as if there were no differencebetween them."XXIIFr. 26. The ablative rare is used by Terence inthe Eunuchus " :I get this comfort from my near-by country-seat.So also Varro, in the twenty-second book of thetreatise addressed to Cicero, says :" I have comerure ' from the country.' " *XXIIIFr. 27. Varro, in the twenty-third book of thetreatise addressed to Cicero, says :" The ingluvies isthe bulging muscles around the throat, which areproduced by fatness and have creases betweenthem." " But now the word is used merely for thethroat.617

VARROFr. 28.^ (1) Cum in disciplinas dialecticas induciatque imbui vellemus, necessus fuit adire atquecognoscere quas vocant dialectic! «lo-ayojyas. (2)Turn, quia in primo Trepl d^tw/vtaToji' discendum, quaeM. Varro alias profata, alias proloquia appellat, Commentariumde Proloquiis L. Aclii, docti hominis, quimagister ^"arronis fuit, studiose quaesivimus eumquein Pacis Bibliotheca repertum legimus. (3) Sed ineo nihil edocenter neque ad instituendum explanatescriptuni est, fecisseque videtur eum librum Aeliussui magis admonendi quam aliorum docendi gratia.(4) Redimus igitur necessario ad Graecos libros.Ex quibus accepimus a^i'w/ia esse his verbis (definitum): XeKTOv avTOTekk

FRAGMENTS, 28Fr. 28." (1) When I wished to be introduced tothe science of logic and instructed in it, it was necessaryto take up and learn what the logicians callewraywyat, or introductory ' exercises.' (2) Thenbecause at first I had to learn about axioms, whichMarcus \'arro calls, now prqfata or propositions,' and'now proloquia or forthright statements,' ' I soughtdiligently for the Commentary on Proloquia of LuciusAelius, a learned man, who was the teacher of \'arro ;and finding it in the Library of Peace,* I read it.(3) But I found in it nothing that was written toinstruct or to make the matter clear ; Aelius '^ seems tohave made that book rather as suggestions for his o>vnuse than for the purpose of teaching others.(4) I therefore of necessity returned to my Greekbooks. From these I obtained this definition of anaxiom : "a proposition complete in itself, declaredwith reference to itself only." (5) This I have forborneto turn into Latin, since it would have beennecessary to use new and as yet uncoined words, suchas, from their strangeness, the ear could hardlyendure. (6) But Marcus Varro, in the twenty-fourthbook of his treatise On the Latin Language, dedicatedto Cicero, thus defines the word very briefly ": Aproloquium is a statement in which nothing is lacking."(7) But his definition will be clearer if I give anexample. An axiom, then, or a forthright statement,if you prefer, is of this kind " : Hannibal wasa Carthaginian " "; Scipio destroyed Numantia "" ;Milo was found guilty of murder " " ; Pleasure isneither a good nor an evil " ; (8) and in general anysaying which is a full and perfect thought, so expressedin words that it is necessarily either true or false, iscalled by the logicians an axiom ;by Marcus Varro,619

VARROappellatum est, a M. Varrone, sicuti dixi, proloquium,a M. autem Cicerone pronuntiatum, quo ille tamenvocabulo tantisper uti se adtestatus est, " quoadmelius," inquit, " invenero."(9) Sed quod Graeci a-vvrjufifvov a^ioj/xa dicunt, idalii nostrorum adiunctum, alii conexum dixerunt.Id conexum tale est : Si Plato ambulat, Plato movetur; Si dies est, sol super terras est. (10) Item quodilli (TVfXTreTrXtyfieioy, nos vel coniunctum vel copulatumdicimus, quod est eiusdemmodi : P. Scipio,Pauli filius, et bis consul fuit et triumphavit et censurafunctus est et coUega in censura L. Mummi fuit.(11) In omni autem coniuncto si unum est mendacium,etiamsi cetera vera sunt, totum esse mendaciumdicitur.Nam si ad ea omnia quae de Scipione illo veradixi addidero Et Hannibalem in Africa superavit,quod est falsum, universa quoque ilia quae coniunctedicta sunt, propter hoc unum quod falsum accesserit,quia simul dicentur, vera non erunt.(12) Est item aliud quod Graeci Su^fvy/itvov d^tw/xa,nos disiunctum dicimus. Id huiuscemodi est : Autmalum est voluptas aut bonum, aut neque bonumneque malum est. (13) Omnia autem quae disiungunturpugnantia esse inter sese oportet, eorumqueopposita, quae avriKiifieva Graeci dicunt, ea quoqueipsa inter se adversa esse. Ex omnibus quae dis-^ Tusc. Disp. i. 7. 14. * Two connected statements, ofwhich the second follows as the result of the first. ^ Thisis the younger Africanus, who destroyed Carthage in 146 B.C.;it was the older Africanus who defeated Hannibal at Zamain 202 B.C.620

FRAGMENTS, 28as I have said, a proloquium or forthright ' statement; but by Marcus Cicero ** a pronuntiatum'or pronouncement,' ' a word however which hedeclared that he used " only until I can find a betterone,"(9) But what the Greeks call a a-wrjiiyikvov d^iwfxaor connected ' axiom,'* some of our countrymencall adiunctum adjoined,' others ' call conexum ' connected.'The following are examples of this : "IfPlato is walking, Plato is moving " "; If it is day,the sun is above the earth." (10) Also what theycall crv/iTTc-Aey/xfiov or a compound 'axiom,' we callconiunctum conjoined ' ' or copulatum coupled ' ' ; forexample " : Publius Scipio, son of Paulus, was twiceconsul and celebrated a triumph, and held thecensorship, and was the colleague of Lucius Mummiusin his consulship." (11) But if in the whole of a propositionof this kind one member is false, even if theothers are true, the whole is said to be false. For ifto all those true statements which I have made aboutthat Scipio ^ I add " and he worsted Hannibal inAfrica," which is false, all those other statementswhich have been made in conjunction will not betrue, on account of this one false statement whichhas been added to them, because they will now allbe spoken of together as one statement of fact.(12) There is also another form, which the Greekscall a Su^ii'yfj.ei'oi' d^tw/xa or 'disjunctive proposition,'and we call disiunctum separated.' For'example " : Pleasure is either good or evil, or it isneither good nor evil." (13) Now all statementswhich are contrasted ought to be opposed to eachother, and their opposites, which the Greeks callavTiKtt'/xtia, ought also to be opposed. Of all state-621

VARROiunguntur unum esse verum debet, falsa cetera.(14) Quod si aut nihil omnium verum aut omniaplurave quam unum vera erunt, aut quae disiunctasunt non pugnabunt, aut quae opposita eorum sunt eontrariainter sese non erunt, tunc id disiunctum mendaciumest et appellatur TrapaSte^evyfieifov, sicuti hocest, in quo quae opposita non sunt contraria : Autcurris aut ambulas aut stas. Nam ipsa quideminter se adversa sunt, sed opposita eorum non pugnant: non ambulare enim et non stare et non currerecontraria inter sese non sunt, quoniam contraria eadicuntur quae simul vera esse non queunt ;possissimul eodemque tempore neque ambulare nequestare neque currere.Fr. 29.^Excipiuntur haurio hausi (invenitur tamenetiam haurivi vel haurii ; Varro in XXI III adCiceronem :" Cum indidem haurierint "), saepiosaepsi, . . .Incertae SedisFragm-entaFr. 30a. ^ Proceres : Varro ad Ciceronem dixit" proceres qui processerunt ante alios, unde et procerestigna quae alia tigna porro excesserunt."Fr. 30b.* Proceres autem ideo secundum Varronemprincipes civitatis dicuntur, quia eminent inFr. 29. ^ Priscian, Inst. Gram. ii. 540. 3-5 Keil.Fr. 30. ^ Schol. cod. Ambr. in Aen. iii. 58. * Serv. inAen. i. 740.Fr. 29. ," To the rule that in the fourth conjugation theperfect is formed by changing the is of the second personsingular of the present, to ivi or ii; cf. Priscian, ii. 539. 8-11Keil.622

FRAGMENTS, 28-30bments which are contrasted, one ought to be true andthe rest false. (14.) But if none at all of them is true,or if all, or more than one, are true, or if the contrastedthings are not at odds, or if those which areopposed to each other are not contrary, then thatis a false contrast, and is called TrapaSu^evyfihov or'WTong-disjunctive.' For instance, this case, inwhich the things which are opposed are not contraries" : Either you run or you walk or you stand."These acts are indeed contrasted, but their oppositesare not contrary ; for not ' to walk and ' ' not tostand 'and not ' to run ' are not contrary to oneanother, since those things are called contraries ' 'which cannot be true at the same time. But youmay together and at the same time neither walk norstand nor run.Fr. 29. Exceptions " are kaurio'I draw off,'perfect kausi (yet haurivi or haurii also is found ;\'arro, in the twenty-fourth book of the treatiseaddressed to Cicero, says " when they haurierint'have drained ' from the same "), saepio 'I fence in,'perfect saepsi, , . .Fragments of Undeterminable PositionFr. 30a. Proceres : Varro in the treatise addressedto Cicero said that proceres are those who processerunt'have advanced ahead of ' others," whence also thosebeams are called proceres which project beyond theother beams.*Fr. 30b. Proceres however, according to Varro,is a name applied to leaders of the state, because theyFr. 30. ° Wrong etymology. » The ends of the raftersproject beyond the Une of the wall on which they rest.623

VARROea, sicut in aedificiis mutuli quidam, hoc est capitatrabium, quae proceres nominantur.Fr. 31 a. ^ Senior : secundum Varronem senior etiunior comparativi sunt per inminutionem. . . . Ergosenior non satis senex, sicut iunior non satis iuvenis,intra iuvenem, sicut pauperior intra pauperem. Dicitautem hoc Varro in libris ad Ciceronem.Fr. 31b.^ lam senior : aut pro positive posuit,id est senex, aut, ut diximus, senior est virens senex,ut iunior intra iuvenem est : quam rem a Varronetractatam confirmat et Plinius.In summo vero constructionis eius, quamFr. 32.^similem navaU carinae diximus, caput conlocavit, inquo esset regimen totius animantis, datumque illi hocnomen est, ut quidem Varro ad Ciceronem scribit,quod hinc capiant initium sensus ac nervi.Fr. 33.^ Non nulli proprie calamos lupinorumalas dici putant, ut Aelius : alae ex lupino, s(urculisine fo)liis ; Cato in Originibus : alae ex lup(ino)leg

FRAGMENTS, 30b-34stand out in it, just as in buildings certain mutules,that is, heads of beams, stand out, which are calledproceres.Fr. 31a. Senior : according to Varro, senior andtuntor are comparatives of diminution.". . , Thereforesenior is not a man who is completely old, just as aiunior is not an entirely youthful person, but one notquite as much as a iuvenis 'young man,' as a pauperior'poorer person ' is not even as much as pauper poor.'Varro says this in the books addressed to Cicero.Fr. 31b. lam senior now older ' ' : either he setthis as a positive, namely senex, or, as we have said, asenior is a vigorous old man,*" as a iunior is one not yeta youth. The matter is mentioned by Varro, and isconfirmed by Pliny.Fr. 32. But at the top of this edifice," which wehave said was like a ship's keel, he set the caput'head,' in which there was to be the guidance of thewhole creature, and the name caput was given to it,as indeed \'arro A\Tites in the treatise addressed toCicero, because from it the senses and the nervescapiunt ' take ' their start.Fr. 33. Some think that the steins of lupines areproperly called alae " ' wings ' ; thus Aelius ^ :" alae oflupine, shoots without leaves " ; Cato in the Origins '^ :" alae of lupine pulse "; \'arro in the treatise On theLatin Language instructs us that the stalk of thebean is called an ala.Fr. 34. With regard to the four conj ugations , theseFr. 33. " Properly, the places where the branches orshoots start from the stem. * Page 59 Funaioli. « Frag.122 Peter.VOL II. s 625

VARROpertinent ad verba quae analogiae parent, quarumexempla passim perscripta sunt et sunt nota. Quaesiquis conceperit animo, non facile labetur. Suntenim evidenter exposita et Varroni Menippeo.Fr. 35.^ Sapio tarn sapui vel sapii quam sapiviprotulisse auctores inveniuntur ;Probo tamen sapuiplacet dici, Charisio sapui vel sapivi, Aspro sapivi etsapii secundum Varronem, quod Diomedes etiamapprobat.Fr. 36.^ Sunt alia verba, quibus desunt diversatempora, usu deficiente, non ratione significationis,Et quibusdam deest praeteritum perfectum et omnia,quae ex eo nascuntur, ut ferio sisto tollo fero aio furo,quod Varro ponit. Ergo aliorum verborum perfectis,quae videntur eandem significationem habere, prohis utimur.Fr. 37.^Puer et in feminino sexu antiqui dicebant,ut Graeci 6 Trais Kal rj Trais ; ut in Odyssia vetere,quod est antiquissimum carmen :Mea puer,* quid verbi ex tuo ore audio ?Fr. 35, 1 Priscian, Inst. Gram. ii. 499. 17-19 Keil.Fr. 36. ^ Priscian, Inst. Gram. ii. 418. 27-419. 2 Keil.Fr. 37. 1 Charisius, Inst. Gram. i. 84. 5-1 1 Keil. " Theverse is quoted with mea puera and supra fugit {for audio)by Priscian, ii. 231. 10 Keil.Fr. 34." The rules for the formation of the perfect active.* Varro, as author of the Menippean Satires, is sometimescalled Varro Menippeus, as here ; but the present passagemay be textually corrupt.626

FRAGMENTS, 34^37are the principles applying **to the verbs which obeyRegularity ; examples are set down everywhere andare well known.easily avoid error.Anyone who has mastered them willby Varro of the Menippean Satire.*P'or they are set forth clearly alsoFr. 35. To sapio ' I am wise,' authors are found tohave employed as perfect sapui or sapii as well assapivi ;yet Probus " prefers sapui, Charisius * sapuior sapivi, Asper sapivi or sapii in reliance on Varro, aview which Diomedes " also approves.Fr. 36. There are other verbs, which lack differenttenses ; it is the use of a form that is lacking, thereason for the meaning exists. Certain verbs lack theperfect and all the forms made from it ; such areferio ' I strike,' sisto ' I put,' tollo ' I pick up,' ^ero ' Ibear,' aio ' I say,' Juro ' I am mad'—a list whichVarro sets down. Therefore for the perfects of theseverbs we use the perfects of other verbs which seemto have the same meaning."Fr. 37. Puer boy the ancients used to use ' ' alsoas a feminine, like the Greek irals, masc. meaning'boy and ' fem. meaning ' girl ' ; as in the oldOdyssey,'^ a very ancient poem :My child, what word is this I hear fall from your lips ?Fr. 35. « C/. iv. 36. 12 Keil. » i. 246. 11 Keil. « i.369. 25 Keil. All these four grammarians belong to thefourth century a.d.Fr. 36. " The text then cites ferio percussi, sisto statui,tollo sustuli, /era tuliffuro insanivi. After Varro 's time, aperfect /«r Mi was occasionally used.Fr. 37. " Livius Andronicus, frag. 3 Baehrens ; R.O.L.24-25 Warmington.ii.627

VARROEt in Nelei Carmine, aeque' prisco :Saucia puer filia sumam ;ubi tamen Varro cum A puera putat dictum, sedAelius rilo,* magister eius, et Asinius contra.Fr. 38.^ Leontion et Chrysion et PAanion* exneutris Graecis feminina nostri^ fecere, et Plautusquod dixit haec P^ronesium et Caecilius (haec>*Leontium. Varroni autem placet talia nominadativo tantum casu et ablativo declinari, m* ceterisvero sic efferri ut nominativo.' Keil, for eoque. * For aedilius cilo.Fr. 38. ^ Charisius, Inst. Gram. i. 104. 1-4 Keil. * Forthyrusion et faunion. * For neutra. * Added byFabricius. ^ For de." Traff. Rom. Frag., page 271 Ribbeck»; R.O.L. ii. 628-629Warmington." The verse is incomplete in sense; perhaps"*the text is corrupt. Page 69 Funaioli. « Page 499Funaioli.628

FRAGMENTS, 37-38And in the Song ofXeleus,'' which is equally old :A wounded child, a daughter, I'll take * . . .In this however Varro thinks that puera, with A, wasused, but his teacher AeUus Stilo '* and Asinius * takethe opposite view.Fr. 38. Leontion and Chrysion and Phanion " areneuters in Greek, but when we took them into Latinwe made feminines of them ; Plautus ^ used Phronesiunias a feminine, and Caecilius ^ used Leoniium

Comparative Table of the Fragment NumbersK =the present edition.GS = Goetz and Schoell, edition of 1910.F = Funaioli, Grammaticae Romanae Fragmenta, 1907.W = Wilmanns, de M. Terenti Varronls lihris grammaticis,1864.K

INDEX OF AUTHORS ANDWORKSReferences are to Book (Roman numeral) and Section (Arabic number),and to Fragment (F.) and serial number (Arabic), with subdivisioiis.Accius, V. 21, 80 (Brutus), OS ; vi.80; vU. 11 ter (Philo:tetes), 14,15, 19, 50, (H, 6i) (Melanippus), 80(PhilocUtes), 83, 85, 88, 96; x.70 ; V. CassiusAelius, V. 18, 21, 25, 66, 101 ; vL 7,59 ; vii. 2; viiu 81Afraniu-s, v. 25 (togata)AnnaUi, v. 74, 101Antipater, vi. 2ApoUodorus, v. 105 ; vi. 2Aprissius (?), vi. 68Aquiliu.s, vi. 89 (BoeotUi)Argei, see Sacra ArgeorumAnstarchei, viii. 63 ; x. 16, 42 (abAristarcho granmiatici)Aristarchus, viii. 68 ; ix. 1, 43, 91Aristea.s, x. 75Aristocles, X. 10, 75Aristodemus, x. 75Aristophanes (Byzantius), v. 9 ; vi.2 ; ix. 12 ; X. 68Aristoteles, vii. 70 (cd^t.ua /3appapiKd);viii. 11Atellanae, vii. 29, 84, 95Athenis in librit «acrorm», v. 97Atilius, \ii. 90, 106Augures, Anguria, Auspicia, v. 21,33, 47, 58, 85 ; vL 42, 53 bU, 64,76, 82 bu, 86, 91 ; vii. 6, 7, 8,31, 51Aurelius Opillus, vii. 50, 65, 67, 70,79, 106Auspicia, v. AuguresBrutus, V. luniugCaecilius, vii. 103CaUimachu-s, v. 113; vii. 34Canius ( = Cassius?), vi. 81Carmen Ndei, F. 37Carmen Priami, vii. 28Carmen ScUiomm, see SaliiCassius, vi. 7 (Brutus), 81 ; vii. 72Cato, vii. 5S ; ix. 107Catullus, ^ii. 50 noteCatulus ( = Pacuvius), vi. 6Censoriae Tabulae, vi. S6-87Chorus Proserpinae, vi. 60, 94Chrysippus, vi. 2, 11, 56; ix. 1X. 59Claudius, viL 66, 70, 106Cleanthes, v. 9Comici, V. 62, 73 ; vi. 71, 73Commsnlaria Consularia, vi. 88-89Commentarium cetus anqnititionit,M. Sergii Mani t, vi. 90-92Cornelius, v. 148, 150 ; vii. 39 («xmwentariumNaevii)Cosconius, vi. 36, 89 {Actionet)Crates, viii. 64, 6J8 ; ix. 1Democritus, vi. 39Dion, viii. 11Dionysius Sidonius, x. 10Duodecim Tabulae, v. 22, 140; vi.5 ; vii- 15, 51Ennius, v. 9, 163Annaleg, v. 22, 42, 55, 59, 60, 65,111, 182; vi. 82; vii. 6, 7, 12,20, 21, 26, 28, 32, 33, 36, 37, 41681

INDEXbis, 42, 43, 44, 45, 46 ter, 48,100, 103, 104 ter ; ix. 54Trag. v. 14, Ifi Ur, 23 ; vi. 6(i4jax),81 {Ajax), 81 (Medea), 83 (Alexander); vii. 6 (Hecuba), 6 (Andronuwha),9 (Medea), 12, 13, 16,19, 33, 49, 73, 76 (Ajax), 82 bis(Andrormcha), 87, 89, 93 Us,101 ; X. 70Mi'cellan^mis, v. 50 (Epicharmiis),60, 62 (.S'ota), 01 bis, 65, 68 (Spicharmus),86 ; vi. 61 ; vii. 35,71, 101, 104 ; ix. 107Epicurus, vi. 39Euripides, vii. 82Fasti, V. 84Flaccus flainen Martialis, vi. 21Fulvius, vi. 33Glossae, Glossemata, vii. 10, 34, 107Grammatici et similes, sine nominecitati, V. 30, 34, 43, 49, 51, 53,85, 120, 146, 147, 154, 157 ; vi.7, 34, 96 ; vii. 10, 17, 34, 36, 46,107 ; viii. 23, 44Hesiodus, v. 20 (Theogonia)Homenis, vii. 74, 85Hortensius, viii. 14 ; x. 78Hypsicrates, v. 88Junius Brutus, v. 5, 42, 48, 55 ; vi.33 bis, 95 (commentaria)luventius, vi. 50 ; vii. 65, 104 noteLeges, vi. 60 ; v. Duodecim TabulaeLeges privatae aedificiorwn, v. 42Lex niancipioriim, v. 163 ; vi. 74Lex Plaetoria, vi. 5Lex praedioi-um urbanorum, v. 27Lex venditionis fundi, ix. 104Litterae antiquae, v. 143 ; vi. 33Livius (poeta), v. 9 ; vii. 3Lucilins, v. 17, 24, 44, 63, 80, 13S(Urbs); vi. 69; vii. 30, 32, 47t«r, 94, 96, 103 bis ; ix. 81Lutatius, V. 150Maccius, vii. 104 ; see PlautusManilius, vii. 16 bis, 17, 28, 105Manlius, v. 31Matius, vii. 95, 96Mimu$, vi. 61632Mucius Scaevola pontifex, v. 5, 83vi. 30 ; vii. 105Naevius, v. 43, 53, 153; vi. 70;vii. 7, 23, 39, 51, 53 bis, 54 bis(Cemetria, Komuliis), 60 (Corollaria),70 (Fretum), 92, 107novies (Aesiona, Clastidiiim,Ihlus, Demetrius, Lampadio,Nagido, Rom,ulus, Stigmatias,Technicus), 108 ter (Tarentilla,Tunicularia, Bellum Punicum) ;ix. 78 (Clastidium)Nelei Carmen, v. Carmen NeleiOpillus, V. AureliusPacuvius, V. 7 ter, 17 bis, 24, 60;vi. 6 bi$, 60 (Medus), 94 (Htrmiona); vii. 6 (Periboea), 18, 22,34 (Medus), 59, 76, 87, 88, 91,102Papinius ('0, vii. 28 (Epigrammation)Parmeniscus, x. 10Physici, v. 69 ; x. 55Piso, V. 148, 149 (Annales), 165(Annates)Plato, vii. 37Plautus :Amph. vi. 6 ; vii. 50Asin. vi. 7 ; vii. 79Aid. V. 14, 108, 181 ; vii. 103Bac. vii. 16Cos. vii. 104, 106Cist. V. 72 ; vii. 64 bis, 98, 99 bisCure. V. 146 ; vii. 60, 71Bpid. V. 131Men. vii. 12, 54, 56, 93Merc. vii. 60Miles, V. 108 ; vii. 52, 86Most. ix. 54Peraa, vi. 95 ; vii. 55Poen. V. 68 ; vii. 52, 69, 88 notePseud. V. 108 ; vii. 81Rttd. F. 10Stich. V. 68Trin. vii. 57, 78True vi. 11 ; vii. 70 ; ix. 106Lost playsAstraba, vi. 73 ; vii. 66Boeotia, vi. 89Cesistio, vii. 67Colax, vii. 105Condalium, vii. 77

INDEXCumicularia, v. 153 ; vii. 52Faeneratrix, vii. 96Frivolaria, v. 89 ; \'ii. 58FngUivi, vii. (53Xervohiria, vii. 65, 68Pagon, vii. 61Parasitus piger, viL 62, 77SiteUUergus, vii. 66Unnamed : vii. 38, 91, 103Poetae sine nomine citati, v. 1, 88 ;vL 11, 60, 67, 83; vii. 52; v.Comici, Mimus, Scaenici, TragiciPolybius, V. 113Pompilius, vii. 93Pontifices, v. 23. 98Porcius, v. 163 ; vii. 104Priami Carmen, v. Carmen PriamiProcilius, v. 148, 154Pythagora.s, v. 11 ; vii. 17Sacra vel Saerificia Argeoram, v.47-54 ; in aliquot sacris et sactllisseriptum, vii. 84 ; v. AthenisSaliorum camiina, v. 110 ; vu 14,49 ; ^^i. 2, 3 bit, 26, 27 ; ix. 61Satumii versus, vii. 36Scaenici, vi. 76Scaevola, i'. MucinsScenici, v. ScaeniciScriptores antiqui Graeci, v. 123Sergius, v. CommentariumSibyllini libri, vi. 15Sisenna, viii. 73, F. 5. 9Sophron, v. 179Sueius, viL 104 bisSulpicius, v. 40Terentius: Adel. vi. 69; vii. 84,F. 10Tragici, vi. 67 M», 72; vii. 23, 24,25Valerius SoraniLS, vii. 31, 65 ; x. 70VarroAntiquitatiim libri, vi. 13, 18De Aestuariis, ix. 26De Poematis, viL 36; De Poetis,vi 52Epistulae, F. 14 cTribuum liber, v. 56Vergilii commentarium Naevi, vii. 39Volnios, V. 55Zenon Citieus, v. 59633

INDEX OF LATIN WORDS ANDPHRASESReferences are to Book (Roman numeral) and Section (Arabic number),and to Fragment (F.) and serial number (Arabic), with subdivisions.A, viii. 68 ; ix. 38, 52 ; A additum,V. 97 ; A exitus, x. (52 ; A litterafinite, F. 14 a, F.^14 b ; A : E, vii.94; AS: ES: IS: IS, ix. 109;c/. Babacus, ix. 40abies, ix. 41aborigines, v. 53aboriuntur, v. 66abrogatae, v. vetusabscessit, vi. 3SAcca Larentia, vi. 23 ; sepidcrnmAccae, vi. 24accanit, vi. 75accensus, v. 82 ; vi. 88, 89, 95 ; vii.58accessit, vi. 38accipe, vii. 90Acculeia, v. Curiaaccusandi casus, viii. 66 ; accusativus,viii. 67 ; v. casusacetum (non aceta), ix. 66, 67Acherusia templa, vii. 6Achilles, x. 69acquirere, vi. 79acsitiosae, t'. axitiosaeactio, v. 11, 12 ; vi. 41 ; actionestres, vi. 42 ; in actionibus, vi. 89,vii. 93actor, vi. 77 ; actores, v. 178, vi. 58,X. 27actus, V. 22, ,34, 35 ; actus numerorum,ix. 86-88adagio, vii. 31addici numo, vi. 61addico, vi. 30634>addictus, vi. 61additio litteraruni, v. 6 ; t». Iaddixit iudicium, vi. 61adicere, v. litteraeadiectio (syllabarum), v. 6adiunctum, F. 28. 9adlocutum ire, vi. 57adlucet, vi. 79adminiculandi pars, viii. 44administra, vii. 34 ; administros,v. 91 ; c/. amministerad Murciae, v. Circusadsentior adsentio, F. 5. 9adseque, vi. 73adserere manu, vi. 64adsiet, vi. 92adventicium (genus similitudinis),X. 69 ; adventicia (verba), x. 70advocare, v. contioadytum, v. 8aedificia, v. 42, 141 ; viii. 29, 30 ; ix.20 ; X. 64aedilis, v. 81 ; v. Publiciusaedis aedes, v. 80, 160 ; vi. 61vii.*10, 12 ; v. Aesculapii, cavumaedium, Concordia, deus, DiusFidius, luno, lupiter, Minerva,Portunus, Quirinus, Romulus,.sacrae, Salus, Saturnus, Venus,Vestaaeditumus, v. 50, 52 ; viii. 61 ; aedituumnon aeditumum, vii. 12aedus, v. haedusaeges, vii. 21Aegeum fretum, vii. 22aegrotus, v. 71 ; x. 46

INDEXAegyptionim vocabuU, viii. 65Aegj-ptus, V. 57, 79Aelia, \iii. 81Aelius Sextus, vii. 46Aemilius -lii, etc., viii. 4 ; Aemiliiis-ia, ix. 55 ; v. Basilicaaenea, v. vasAeneas, v. 144 ; ^-i. 60Aeolis, V. 25, 175 ; Aeolis Graeci,V. 101, 102aequabilitas, ix. 1, etcAequimaeliiun, v. 157aequinoctium, vi. 8 ; %-ii. 14 ; ix. 25 ;r. circulnsaequor, vii. 23aequutn, vL 71 ; r. pilaaer, v. 65 ; c/. animaliaaerariae (non aerelavinae), viii. 62aerarii, v. milites, tribuniaerarium, v. 180, 183aes, V. 169-171, 173, 180-183 ; ix. 81-83 ; X. 38 ; aes et libra, vii. 105,ix. 83 ; v. militare, mille, raudusAesculapii aedes vetus, viL 57aesculetum, v. 152aestas, v. 61 ; vi. 9aestivum, vL 9 ; aestiva triclinia,viii. 29aestus, vii. 22 ; ix. 26aetas, vi. 11 ; ix. 93aetemum, vi. 11Aethiops, viii. 38, 41 ; ix. 42Aetolia, vii. 18aevitemum aetemum, vi, 11aevum, vi. 11Africa, V. 159Africae bestiae, vii. 40Africus vicus, v. 159Agamemno, v. 19Agenor, v. 31ager, v. 13, 34, 37 ; ciiltus, incultiis,V. 36 ; Romaniis, v. 33, 55 ; agrorumgenera quinque, v. 33 ; r.Arictnus, Calydonius, Gabinus,hosticus, incertiis, Latius, novalis,peregrinus, Praenestinus,Reatinus, restibilis, Romanus,Sabinus, Tusculanus, uliginosusagger, v. 141agitantur quadrigae, vi. 41, 42agitatus, v. 11, 12 ; vi. 41, 78 ; mentis,vi. 42agnus, V. 99ago, V. 34 ; vi. 41, 42, 77, 78 ; agitgestum tragoedus, vi. 41 ; agiturpecHS pastum, vi. 41 ; agitnrfabula, vi. 77 ; agere causam,augurium, vi. 42 ; agere ex sponsu,vi. 72 ; V. facio, gerit, gradusagonales, v. diesAgonenses, vi. 14Agonia, vi. 14agrarius, v. 13 ; agrarii, viii. 15agrestis liostias, viL 24 ; c. Iocsagricola, v. 13agrosius, v. agntrinsaio, F. 36ala, V. 33alauda alaudas (Gall.), viii. 65Alba, V. 144 ; \-iii. 35 ; Alba Longa,V. 144Albani Albenses, viii. 35 ; v. AventinusAlbanus mons, vi. 25 ; rex, v. 43albatus, v. 82Albius, viii. 80 ; x. 44 ; Albia, x. 44AlbuU, V. 30albus -a -um, etc., viii. 38, 41, 80;ix. 42, 55; x. 22, 24, 44, 73;album albius albissimum, viii.52, 75alcedo, v. 79 ; vii. 88Alcmaeus Alcmaeo, ix. 90alcyonia, vii. 88Alcyonis ritu, vii. 88Alexander (Magnus), ix. 79 ; eiusstatua, ix. 79Alexander (Paris), vii. 82Alexandrea, v. 100Alfena, viii. 41 ; ix. 41aliena verba, v. 10alienigenae, v. 90alites, V. 75allecti, v\. 66AUia, Alliensis dies, vi. 32alpha, viii. 64altiores, x. 29altisono caeli clipeo, v. 19 ; vii. 73altitonantis Io^^s, vii. 7amator, Wiu 57ambages, ^ii. 30ambagio, vii. 31ambe, vii. 30ambecisus, viu 43ambegna bos, vii. 31ambiectum, v. 132ambiguus rectus ca.sus, ix. 103ambit, v. 28635

INDEXambitiosus, vii. 30ambitus, v. 22, 28 ; vii. 30 ; v. indagabilisAmbivius, vii. 30ambulatur, vi. 1 ; ambulans, ambulaturus,viii. 59amburvom, v. 127ambustum, vii. 31amens, vi. 44amia, vii. 47amicitia -am, x. 73amictui, v. 131, 132amiectum, v. ambiectiimamitans (non est), viii. 60Amiternini, v. 28Amiternum, vi. 5amminister, vii. 34 ; c/. administraamnis, v. 28amo amor, etc., viii. 58, 60 ; ix. 97,110; X. 32, 48, 78; amans amaturusamatus, viii. 58, ix. 110amor amorem, etc., x. 36, 42amphimallum, v. 167analogia, viii. 23, 25-27, etc. ; ix. 1,2, 7, 74, etc. ; X. 1, 36-38, 43, 44,61, 52, 63, 70, 72, 74, 79, 83, etc.,F. 34; perfecta, inchoata analogia,X. 68, 69 ; index analogiae, ix.109 ; analogiae genus deiunctum,coniunctum, x. 45-47 ;poeticaanalogia, x. 74 ; v. genus, poetica,principium, proportione, ratio,similitudoanas, v. 78ancilia, vi. 22 ; vii. 43Andrius ab Andro, viii. 81Andromacha, vii. 82anfractum, vii. 15Angerona, Angeronalia, vi. 23angiportum, v. 145 ; vi. 41anguilla, v. 77angulus, vi. 41ani, vi. 8anicula anicilla, v. anusanima, v. 59, 60 ; animae hominum,ix. 30animalia, v. 75, 102 ; ix. 113 ; aquatilia,v. 77 ; in aere, v. 75 ; inaqua, V. 78 ; in locis terrestribus,V. 80 ; animalium semen, v. 59species, x. 4"; voces, vii. 103animalis res, F. 8animantium (animalium) voces, v.75, 78, 96, 100 ; vii. 103636animum, i'. despondisseAnio, V. 28annales, v. 74, 101 ;(feriae), vi. 25,26annus, vi. 8 ; v. novusanomalia, viii. 23; ix. 1, 3, 113; x.1, 2, 16 ; V, dissimilitudoanquisitio, vi. 90, 92anser, v. 75Antemnae, v. 28antiqua, vi. 61, c/. vi. 82 ; antiqui, v.34, 71, 79, 96(Graeci), 131, vi. 19,33, 58 (nostri), 63, vii. 26, 36, 73(rustici), 84, ix. 17, 68, 83, 87,X. 73, F. 1 ; antiquissimum, v. 133 ;antiquissimi, v. 132 ; antiquiGraeci, v. 103, 166 ; antiquaemulieres, v. 69 ; antiquum oppidumPalatinum, v. 164, vi. 34 ;antiquum Graecum, vi. 84 ; antiquumnomen, v. 50 ; antiquislitteris, v. 143, vi. 33 ; v. Graecus,gramma tica, lupiter, numerus,urbs, verbumAntonius, i'. Tulliusanuli, vi. 8anus, viii. 25 ; anicula anicilla, ix.74 ; V. Libera parte totum, v. 155 ; vii. 18, 75Apelles, ix. 12aper, v. 101 ; viii. 47apexabo, v. IllAphrodite, vi. 33Apollinar, v. 52ApoUinares ludi, vi. 18Apollo, V. 68 ; vii. 16, 17 ; cortinaApollinis, vii. 48 ; v. Solappellandi pars, viii. 44 ;partesquattuor, viii. 45Aprilis, vi. 33aprunum (Sab.), v. 97Apula, V. lanaApulia, V. 32aqua, v. 61, 122, 123 ; v. animalia,ignisaquae caldae, v. 25, 156 ; ix. 68,69aquae frigidae, v. 25aqualis, v. 119aquarium, v. vasaquatilia, v. animaliaaquila, viii. 7 ; ix. 28Aquiliani gladiatores ab Aquilio,ix. 71

INDEXAquilo, ix. 25ara, T. 38 ; arae, v. 74 ; r. Census,deiis, Elicii, Hercules, lupiter,Lavemae, Tatiusarationes, v. 39aratrum, v. 135Arbemus -na, v. Airemusarbitrium ( = censio), vii. 58arbor arbuscula, F. 9 ; arbores, vii.8, 9, ix. 80arborariae felces, v. 137area, v. 128 ; ix. 74Areas, v. 21arcera, v. 140arcs, V. ArxareuJa, ix. 74ardor, v. 38, 61area ar«ae, v. 38arefacit, r. 38arena, r. a.senaAreopagus, Areopagitae, \ii. 19Argei, v. 45 ; viL 44 ; Argeorumsacriticia, v. 52 ; sacra, v. 50saceUum quartum, v. 47 ; sextum,V. 48 ; sacraria septem etviginti, V. 45, c/. 47argentarii, xi. 91argenteom argentea, ix. 66argentifex non dicitur, viiL 62argentifodinae, v. 7 ; ^^iL 62argentum, v. 169, 173, 174 ; ix. 66(non argenta) ; x. 38.\i~gi, vii. 44 ; ix. 69Ar^letum, v. 157Argivi, vii. 38 ; v. HerculesAi^os, ix. 89Argus, ix. 89 ; Argus lArisaens, v.157Arieia, v. 32, 143Arieinus ager, v. 34aries, v. 98, 117ariga, r. ariugaArimmas, ix. 12arista, vi. 49Aristarchum, vL 2ariuga ariugus, v. 98anna, v. 115 ; t». sonantarmamentarium, v. 128armarium, v. 128Armenia (linguaX v. 100armenta, v. 96Armilustrium, v. 153; vi. 22arrabo, v. 175arruit, v. 135ars, V. 93 ; \iiL 6 ; v. medicina,musica, sutrinaArtemas, viiL 21Artemidorus, viii. 21, 22articuli, ^-iii. 45, 51, 52, 63 ; x. 18-20, 30, 50 ; r. genus, inflnitiartifex artufex, v. 93; ix. 12, 18,111; X. 64; artificum vocabula,V. 93aruspex, r. baruspexAr\-ales (FratresX v. 85Arvernus -na, » iii. 81ar\iga, r. ariugaar\Tis, V. 39Arx, V. 47, 151 ; vi. 28, 91, 92 ; vii.8, 44 (arcs)as, V. 169, 171, 174 ; ix. 81, 83, 84X. 38 ; asses, v. 170, 182 ; as assemasses, X. 83 ; asse, F. 18asbestinon, v. 131ascripti^^, viL 56a.sellus, v. 77 ; ix. 113a.sena (= arena), vii. 27Asia, V. 16, 31 ; vii. 21 ; viii. 56ix. 27Asiatici, viiL 56asinus -a, ix. 28 ; asini, ix. 93asparagi, v. 104aspicio, %i. 82assarius as.sarium, viii. 71asserere, r. adserereasseres, vii. 23assiduus, \iL 99assipondium, v. 169assnetudo, ix. 20assum, V. 109, c/. 120a-sta (Osc.), viL 54asta, V. hastaastrolc^, ix. 24atavus, vii. 3Atemus, v. 28Athenae, v. 97 ; vii. 19 ; viii.35Athenaeopolitae, viiL 35Athenaeus, viii. 81 ; Athenaei, viiL35Athenaeus rhetor, viiL 82Athenaiis, viiL 35Atheniensis, \iiL 81Atinia -ae, ^nii. 73atratus, viiL 18atri (diesX vi. 29Atriates (Tusci), v. 161atriensis (non atritumus), viiL 61637

INDEXatrium, v. 161 ; viii. 29 ; atriumsutorium, vi. 14Atrius ab atro, viii. 80 ; Atrius -a-o, X. 44Attalus rex, vi. 15Attica, V. 76Attici, V. 108, 175 ; vi. 10attributum, v. 181attuiinur, vii. 7aucella, v. avisauceps, viii. 61audio, vi. 83augendi genus, viii. 52augmentum, v. 112augur, vi. SI5 ; augures, v. 47, ^i.42, 53, 64, 76 ; v. Index of AiUhm-s,s.v. Auguresaugurium, vi. 42 ; vii. 8, 51 ; auguria,V. 85 ; v. agoauraculum, v. auguraculumaures, vi. 83 ; aurium aviditas, vi.83 ; auribus lupum teiieo, vii. 31aurescit, vii. 83aurifex, viii. 62aurora, v. 24 ; vii. 83ausculto, vi. 83auspicato, v. diesauspicium, vi. 76, 82, 86, 91 ; vii. 8 ;auspicium j>etitum, vi. 91 ; auspicia,V. 33, 143 (urbana), vi. 53(caelestia), vii. 8, 97 (sinistra)V. Index of A uthors, s.v. AuguresAuster, ix. 25autumnus, vi. 9auxilium, v. 90Aventinus (raons), v. 43, 152 ; vi.94 ; rex Albanus, v. 43averrvuicassint, vii. 102Averruncus, vii. 102aviarium (non avile), viii. 54avicula, v. avisaviditas, v. auresavis, viii. 54 ; ix. 76 ; avi et ave,viii. 66 ; ave, F. 18 ; aves avium,viii. 70 ; avis avicula aucella,viii. 79 avem specere, vi. 82axis, vii.;74 ; axe, F. 18 ; axes nonaxeis, F. 20axitiosae, acsitiosae, vii. 66B, ix. 38 ; BA~, ix. 51 ; BS, x. 57bacca in Hispania vinum, vii. 87Bacehae, vii. 87Bacchides Bacchidas, x. 71638Bacchus, vii. 6 ; Bacchi sacra, vii.87 ; Bacchi templa, vii. 6Baebii -iae -lis, x. 50Balatium ( = Palatium), v. 53balneae (non balnea), viii. 48, 53 ;ix. 68, 106, 107 ; balneum, viii.48, ix. 68balneator, viii. 53balteum, v. 116barbara (vocabula), barbari, viii. 64barbatus, v. 119 ; ix. 15Basilica Aemilia et Fulvia, vi. 4 ;Opimia, v. 156beatus, v. 92Bellona, v. Duellonabellum, 1'. Carthaginiense, duelluni,indicit, Pimicum, Pyrrhi, Sabinumbes olim des, v. 172bestiae, v. Africaebeta betaceus, F. 10 ; v. pesbibo, vi. 84bicessis, v. vicessisbigae, viii. 55 (non duigae), ix. 6364 ; X. 24, 66 (non biga), 67binaria, v. formulabini (non duini), viii. 55 ; binaebina, ix. 64. x. 24, 67 ; v. unusbiselliuni, v. 128bonus boni, x. 68 ; bonum malum,V. 11, viii. 34; melius optimum(non bonius bonissimum), viii.75, 76 (optum optius, melummelissimum desunt) ; v. Copia,dea, duonus, meliosem, quod bonum,scaevabos boves, etc., v. 96 ; vii. 74 ; viii.54, 74 (bos non bous ; bourn etbovenim) ; ix. 28, 113 ; bovis vox,vii. 104 ; V. ambegna, Lucabovantes, vii. 104Bovarium Forum, v. 146bovile (non dicitur), viii. 54 ; ix.50brassica, v. 104breviores, x. 29bruma, vi. 8 ; ix. 24, 25Bruti, 1». Mucinsbubo, V. 75buceo, vi. 68bucinator, vi. 75bulbum, V. 112bura, V. 135Busta Gallica, v. 157

INDEXC : G, V. 64, 101, 116 ; vi. 95 ; CS :X, ii. 44, X. 57Cabirum delubra, viu 11caccabus, v. 127cadus, ix. 74Caeciliani gladiatores a Caecilio,ix. 71Caecilius Cecilius, vU. 96Caecina, x. 27caecus -a -tun, ix. 58 ; r. cubicuhuncaelare, v. 18Caeles Vibenna, v. 46caelestia, v. auspiciumCaeliani, v. 40caeligeua, v. 6i ; v. VenusCaelii -iae -iis, x. 50Caeliohun, v. 46caelites, viL 5, 34Caelius mons, v. 46, 47Caelum, v. 57-60, 63, 65, 67 ; caelum,V. 16-18, 20, 31. viu 20; hoccaelo, X. 62 ; caeli loca supera,V. 16 ; caelum principium, v. 64 ;caeli regiones, v. 31 ; v. signumCaeriolensi-s (locusX v. 47caesa, r. exta, rutacaesins (caesior non diciturj caesissimus,viii. 76Calabra, v. ciuriacaiamLstrum, v. 129calatio, v. 13calcearia tabema non dicitur, riiu55calcei, viiu 55 ; li. 40caldor, v. 59caldus caldo, x. 73 ; caldum calditiscaldissimum, riii. 75 ; v. aquaeCalendae, c. lanuariae, Kalendaecalix, V. 127calo (kaloX vi. 16, 27calor, V. 60Calpumius, C, vi. 88Calydon, vii. 18Calydonius ager, non terra, vii. 18camelopardalis, v. 100camelus, v. 100Camena, vi. 75 ; vii. 27 ; Camenarumpriscum vocabulum, vii. 26 ;r. Casmenacamillu.s Camilla, Aii. 34Campania, v. 137campus, v. 36 ; \-i. 92 (Martins) ; «.Flaminius, Martinscanali, F. 18cancer, viL 81candelabrum, v. 119candens, r. signumcandidus -um candidiu.s candidissimum,^-iiL 17 ; candidus -a candidissimus-a, ^iiL 77canes, viL 32 (canes laniorumX 33(caninam non est) ; canis, v. 99,vii 32 ; canis catulus catellus,ix. 74 ; cane, F. 18canicula (piscis), v. 77canistia, v. 120canit canere, vi. 75 ; canite cant«,vii. 27cantatio, vi. 75cantator non dicitur, viiL 57cantitat, vi. 75 ; cantitans, viii. CO(cantitantes non dicitur)canto cautat, vi. 75cape, vii. 90 ; cape capito, x. 31caperrata fronte, viL 107capides, v. 121capilli (gen. sing.), vii. 44capiUl, V. 130capitales, v. trium\Tricapitellum, r. caputcapitium, v. 131Capitoliuus, v. 41 ; Capitolinuscli\'us, vL 32Capitolium, v. 149, 158 ; ^n. 27, 68Capitolium vetus, v. 158capitulum, v. caputcapra, v. 97caprea, v. 101caprilicus, vL 18Caprotina (Iimo), vi- 18 ; CaprotinaeXonae, vi. 18Capua Capimnus, x. 16capulae, v. 121 ; ix. 21caput capitis, etc, ix. 53 ; x. 82,F. 32 ; caput capitulum, viii. 14capitellum (deest), ^iiL 79 ; caputSacrae viae, v. 47 ; caput undedeclinatur, x- 50, ef. ix. 102, 103,X. 50carbone, F. 18career, v. 151 ; carceres, v. 153carere (lanam), viL 54Carinae, v. 47, 48cariosas, vii. 28Carmena -ae, vii. 26, 17Cannentalia, vi. 12Carmentis feriae, vi. 12639

INDEXearminari, vii. 54carnaria taberna non dicitiir, viii.55caro, viii. 55 ; carnem petere (exAlbano monte ex sacris), \'i. 25 ;V. pecusCarrinas (non Carrinius), viii. 84Carthaginiense bellum, v. 165cartibuliim, v. 125Cascelliani gladiatores a Cascellio,ix. 71cascus -i, X. 73 ; cascws -a, vii. 28 ;Casca, vii. 28caseus, v. 10(>, 108 ; vi. 43Casinuin, vii. 29Casmena -ae, vii. 26-28Casmilus, vii. 34casnar (Osc), vii. 29cassabundus, vii. 53Castor, V. 58, 66, 73castra, v. 121, 162, 166casuale (genus declinationis), viii.52 ; onitionis prima pars easualis,X. 18casus, V. 4 ; vi. 36 ; viii. 11, 16, 22,42, 44, 46, 58, 63-67 ; ix. 31, 34,50-52, 54, 70, 77, 81, 88-90, 94,110 ; X. 7, 10, 17, 21, 22, 26, 29-31,34, 35, 42, 47, 54, 65, 80, 82 ; decassu in cassum, viii. 39 ; casuumvocabula, x. 23 ; casuum iacturae,ix. 78 ;(casus) quis, quemadmodum,quo, a quo, cui, cuiusvocetur, viii. 16 ; cum vocaret,cum daret, cum accusaret, viii.16 ; secundum naturam nominandiest casus, ix. 76 ; casus communis,viii. 46 ; casus singuli,terni, etc., ix. 52 ; casus naturaleset impositicii, x. 61 ; v. accusandi,dandi, declinatio, exitus, nominandi,obliqui, patricus, patrius,ratio, rectus, series, sextus, transitus,vocandicatellus, V. canesCatinia -ae, viii. 73catinus, v. 120 ; catinuli, F. 12Cato Catulus, v. 99catulae, x. 66catulus, V. 99, ; v. canescatus -a, vii. 46caulis, V. 103caullae, v. 20causam orare, vii. 41 ; causae ver-640borum, vi, 37 ; v. ago, dicis,nascendicava, V. 19 ; cava cortina, vii. 48cavatio, v. 19, 20cavea, v. 20cavernae, v. 20ca\Tim, V. 19, 20, 135 ; cavumcaelum, v. 19, 20 ; ca\'umclipeum,V. 19 ; V. cava, chaos, coumcavimi aedium, v. 161, 162Cecilius, v. Caeciliuscelare, v. 18cella, V. 162cenaculum, v. 162ceno cenatus sum, F. 5. 7censio ( = arbitrium), v. 81 ; vii. 58censor, v. 81 ; vi. 86, 93 ; censores,vi. 11, 87, 90, 92censorium iudicium, vi. 71 ; censoriaetabulae, vi. 86centenarius, v. gradus, numeruscentum, ix. 82, 87 ; x. 43centumvirum (non -viiorum), ix. 85centuria, v. 35, 88 ; v. olluscenturiato constituit, vi. 93 ; v.comitiumcenturio, v. 88centussis, v. 169, 170 ; ix. 81, 84cerei, v. 64cereo, vi. 81Ceres, v.' 64; vi. 15; templumCereris, vii. 9Cerialia, vi. 15Cermalus, Germalus, Germalense,V. 54cerno, cernito, cemere vitam, vi.81 ; cernere crevi, vii. 98Ceroliensis, v. Caeriolensiscerus, vii. 26cervices cervix, viii. 14 ; x. 78cervus cerva, viii. 47 ; cer\'us cerve,x. 51 ; cervi, v. 101, 117Cespius Mons, v. 50chaos, V. 19, 20charta, F. 14 a, F. 14 bChersonesice, v. 137Chio vinum, ix. t)7chlamydes, v. 133 ; clamide, v. 7chorda citharae, x. 46clioum, V. 19Chrysides Chrysidas, x. 71Chrysion, F. 38cibaria, v. 64, 90cibus, viii. 30

INDEXciccum, vii. 91cicer, viii. 48, 63 ; cicer cicericiceris, x. 54cicur cicurare, vii. 91Cicurini, v. Veturiicilibantiun, v. cillibacilUba, V. 118, 121cinctus, V. 114cinerarius, v. 129cingillum, v. 114cippi pomeri, v. 143Ciprius, r. Cypriusciprum, r. cj'prumcirculus aequinoctialis, solstitialis,septemtrionalis, brumalis, ix. 24,25 ; circuli, v. 106circumiectui, v. 132circum muros, vi. 90, 92, 93circumtextiim, v. 132Circus, V. 153 ; vi. 20 ; Flaminius,V. 154 ; Maximus, v. 153 ; adMurciae, v. 154 ; v. oppidumcista cistula, viii. 52 ; cista ci.stulacistella, viii. 79, ix. 74cis Tiberim, v. 83cistula, V. cistacithara, viii. 61 ; x. 46civilia vocabula dierum, \i. 12civis, X. 39civita.s, X. 39 ; civitatum -ium, viii.66clam, vii. 94clamare, vi, 67clamide, v. chlamydesclasses, v. 91 ; classe, F. 18classicus, V. 91 ; vi. 92claustra, vii. 21clavi, X. 62clepere clepsere, vii. 94clipeus, V. 19Clivos, r. Capitolinus, Cosconius,proximus, Publicius, Pullinscloacae, v. 149Cluaca Maxuma, v. 157clucidatus, vii. 107clupeat, V. 7cobius, vii. 47cochlea, F. 14 a, F. 14 bCodes, vii. 71cocus, vii. 38coemptio, vi. 43Coeus Titan, ^-ii. 16cogitare, vi. 42, 43cogitatio, vi. 42VOL. IIcc^natio verbomm, v. verbumcognomina, viii. 17; ix. 71,cohors, V. 88colem colis cole, ix. 75 ; colis noncols, ix. 76collatio verborum, viii. 78collecta, vi. 66collega collegae, vi. 66, 91colles (Romae), v. 36, 51, 52 ; v.Latiaris, Mucialis, Quirinalis,Salutaris, ViminalisCollina tribus, v. 56 ; regio, v. 45coUocatum, v. 14colloquium, vi. 57colo colis colui, ix. 108colonia nostra, v. 29 ; coloniae nostrae,V. 143columba, v. 75 ; ix. 56 ; coliunbus,ix. 56columna columella, F. 9coma, V. frondenticomissatio, vii. 89comiter, vii. 89comitiales (dies), vi. 29comitiatum (ad c. vocare), v. 91xl 93 ; c/. vL 91comitia\-it, r. quando rexcomitium, v. 155 ; vi. 5, 29, 31comitia, v. 85, 91, 155, vL 91, 92,vii. 42, 97 ; comitia centuriata,vi. 88, 92, 93 ; comitia curiata, v.155commentum, comminisci, vi. 44commode, viiL 44Commotiles Lymphae, v. 71commimis, r. casus, consensus, consuetudo,nomencommutatio (syllabarum, litterarum),V. 3, 6, 79, 103, 137 ; vi. 2,62, 83 ; vii. 31 ; ix. 99 ; x. 25commutatio vocis, x. 77comoedia comoediae, vL 55, 71, 73comoedus, ix. 55comparativi, F. 31 acompendium, v. 183competa, v. compitumcompitalia, vi. 25, 29compitum, vi. 43 ; competa, vi. 25compluium, v. 161 ; compluvium,V. 125composita, ^-iiL 61 ; compositinumeri, ix. 84compositicium genus, viii. 61 ; compositicia(verba), vi. 55641

INDEXcomputatio, vi. 63conceptis verbis, vii. 8conceptivus, v. dies, feriaeconcessit, vi. 38conchae, ix. 28conchylia, v. 77conciliari, vi. 43concilium, vi. 43concinne loqui, vi. 67conclavia, viii. 32Concordia, v. 73; ( = templum), v.148 ; aedis Concordiae, v. 156concubitus, vii. 78concubium, vi. 7 ; vii. 78condere, v. lustrum, oppidum, urbsconexum, F. 28. 9confessi, vi. 55conflctant, vii. 107confingere, v. 7coiigerro, vii. 55coniugationes quattuor, F. 34coniunctae res, x. 24 ; j;. analogiaconiunctio (ignis et humoris), v.63 ;(verborum), vii. 110, c/.viii. 1coniunctum, F. 28. 10, 11conpernis, ix. 10conquaestor, v. q\iaestor, vi. 79conregio, vii. 8consensus communis, viii. 22Consentes, v. deusconserere manum, vi. 04 ; consertummanum, vi. 64consilium, vi. 43Consiva, v. Opsconsortes, vi. 65conspicare, vii. 9conspicio, vi. 82 ; vii. 9conspicio -nis, vii. 8, 9consponsus, vi. 69 ; vii. 107 ; consponsi,vi. 70constantia, ix. 35Consualia, vi. 20consuetudo (comnnmis), v. 1, 6, 8vi. 78, 82 ; vii. 32 ; viii. 6, 23, 26,27, 32, 74, etc. ; ix. 1, 2, 8, 74, 76,78, etc., 114; x. 2, 15, 16, 73,etc. ; F. 5. 9 ; non repngnanteconsuetudine comnuini, x. 74, 76,78 ; consuetudo nostra, vetenim,vi. 2 ; c6ns\ietudo vetus, liaec,x. 73 ; V. priscaconsul, V. 80, 82 ; vi. 61, 88, 91, 93,95; X. 28; consules, vi. 91, 99,642viii. 10 ; v. Curtius, Manlius,TulliusCensus (et eius ara), vi. 20contemplare contempla, vii. 9contentiones, viii. 75conticinium, vi. 7 ; vii. 79contio, vi. 43, 90 ; contionem advocare,vi. 91, 93contraria (verba), viii. 58, 59 ; v.deusconvallis cavata vallis, v. 20convivium, v. 124, 168 ; conviviumpublicum, v. 122conum, V. 115Copia Bona, vii. 105 ; copia verborum,viii. 2, 20copis copiosus, v. 92copulae, viii. 10 ; trinae copulae,naturae et usuis (c/. viii. 14), personarummultitudinis ac finis, ix.4 ; divisionis quadrinae copulae,X. 33 ; V. faciendicopulatum, F. 28. 10cor, vii. 9, 48corbes corbulae, v. 139 ; corbesnon corbeis, F. 20corda, v. chordaCorduba, v. 162Cornelius, vi. 4Corneta, v. 146, 152cornicen, vi. 91 ; cornicines, vi. 75cornices, vi. 56cornua, v. 117 ; vii. 25cornutus, vii. 25, 39corolla Veneria, v. 62 ; corollae inscaena datae, v. 178corollarium, v. 175, 178corona, v. 62 ; coronas iaciunt infontes, puteos coronant, vi. 22corpus, V. 11, 12, 59-61 ; a corporedeclinata, viii. 15correptio (syllabarum), v. 6cortina, v. Apollo, cavacortumio, vii. 8, 9corvus, v. 75 ; ix. 55, 56 (noncorva) ; corvi, vi. 56Cosconius (Clivus), v. 158 ; viocurus,V. 158coum a cavo, v. 135 ; v. choumCovella, V. lunocoxendices, vii. 67Cozevi, vii. 26eras, viii. 9cratis, vii. 55I

INDEXcreatus, v. vitiocrepare, vi. 67creperum, creperae res, vi. 5 ; vii.77Crepusci, vi. 5crepusculum, ^-i. 5 ; vii. 77Cretaea, vi. 69cretaria tabema, viii. 55cretio, vi. 81crocodilos, v. 78cruda holera, v. 108erusta, v. 107crustulmn, v. 107Crustumerina secessio, v. 81crux cruce cruces, ix. 44ciibicularis gradus, viiL 32cubiculum, v. 162 ; viii. 29, 54cubiculum caecum, ix. 58cuculus, v. 75cuciimeres, v. 104culcita, V. 167culmen, v. 37culmi, V. 37cnlpo c\ilpam\is, x. 33cultus, V. agercumerus, vii. 34ciun muliere fuisse, vi. 80Cupidini.s Forum, Forum Ciippedinis,v. 146cuppedium, v. 146cupre.ssi cupres-sus, is. 80cur, viii. 9cura, vi. 46curare, vL 46curatores omnium trihnum, ^-i. 86Curenses, vi. 86Cures, V. 51Curia Acculeia, vi. 23 ; Calabra, v.13, vi. 27; Hostilia, v. 155, vii.10 ; curiae, v. 83, 155, vL 15, 46curiae veteres, v. 155curiata, v. comitiumcuriones, v. 83 ; vi. 46curiosus, vi. 46currit, viii. 11, 53; cuiren-s cursurus,viii. 59cursio, V. 11curso cursito, x. 25cursor, v. 11, 94 ; viiL 15, 53cursus, vi. 35Curtius, v. 148 ; Curtius lacu.s, v.148-150 ; Mettius Curtius Sabinus,V. 149 ; Curtius, consul, v. 150eurvor, viL 25 ; c/. v. 104CutUiensis lacus, v. 71cyathus, v. 124cybium, v. 77Cyprius Vicus, v. 159cyprum (Sab.) bonum, v. 159Cyzicenus (non Cyzicius) a Cyzico,ViiL 81D : R, vi. 4, ef. vi. 83damnum, v. 176dandi casus, viii. 36 ; x. 21, 65 ; v.casus ; ef. viii. 16dea bona, r. quisdecern, x. 41, 43, 45December, \\. 34decemplex, v. logoedecemvirum (non -virorum) iudicium,ix. 85decemimt de vita, vi. 81decessit, \\. 38decessus, r. Gallideciens, hoc deciens, huius deciens,ix. 88Decimus, ix. 60 ; r. decumadeclinata verba vel vocabula, v. 7vi. 37; viii. 1, 2, 9; ix. 115; declinatanomina, viii. 5declinatio -ones, viii. 3, 5, 11, 13,15, 20, 21, 24 ; ix. 10, 17 (novae),110; X. 3, 11, 12, 16, 28, 44, 51,53, 60, 62, 74, 76 (verbi), 77;declinatio in casus, vii. 110 ; declinationaturalis et voluntaria,viii. 21-23, ix. 35, i. 15 (voluntas),17, 51, 77, 83; declinationumgenus, viiL 17, 21 ; declinationumgenera quattuor, viiL 52 ;declinationes verborum, vi. 2, 36,38, ix. 3, X. 1, 2, 9, 11, 26, 44;V. declinatus, derectae, iuniores,nothus, priscum, recentes, similitiido,tran.situsdeclinatvLs, viii. 6, 10; ix. 37, 38,51, 53 ; X. 51, 76 ; declinatusvoluntarius, naturalis, ix. 34, 62,X. 77, 83 ; vociun declinatus, x.65; verborum declinatuum generaquattuor, vi. 36 ; declinatuumspecies quattuor, x. 32 ; sex, x. 31imperandi declinatus, x. 32, c/.ix. 32, 101 V. ordo declinatuum;decuma, r. Herculesdecuriae numerorum, ix. 86, 87 ; cf.v. 34, 91 ,643

INDEXdecuriones, v. 91decussis, v. 170 ; ix. 81dedicat dedicatur, vi. 61definitiones grammaticonim, x. 75deierare sub tecto, v. 66Dei Penates, v. densdeiunctum, v. analogiaDeli, vii. 16Deliadae, vii. 16delicuum, deliquare, vii. 106Delphi, vii. 17delubra, v. CabirumDemetrius rex, vii. 52demptio litterarum, v. 6 ; vii. 1denarius denarii, v. 170, 173, 174 ;viii. 71 ; ix. 85 ; x. 41 ; denarium(non -orum), viii. 71, ix. 82, 85 ;V. formula, gradus, numeridenasci, v. 70dens, v. 135 ; viii. 67 ; dentumdenies, viii. 67densum, v. 113deorsum, v. 161 ; deorsum versus,ix. 86depsere, vi. 96derectae declinationes, x. 44derectus, v. ordo, ratiodes, V. besdespicio, vi. 82despondet, vi. 69 ; desponsa, vi. 70 ;despondisse, vi. 71 ; despondisseanimum, filiam, vi. 71desponsor, vi. 69destringor destrinxi, F. 5. 7detrectio (syllabanim), v. 6detrimentum, v. 176detritum, v. E, Sdeunx, v. 172deus deei, viii. 70 ; dei, v. 57, 65,66, 71 ; decs, F. 1 ; dei contrarii,v. 71 ; deo principe, vi. 34 ; deiprincipes, v. 57 ; Dei Consentes,viii. 70 ; Deum (non Deonim)Consentium aedem, viii. 71 ; deimagni, v. 58, vii. 34 ; diis inferis,vi. 34 Dii Penates nostri, v. 144 ; ;Dei Penates, viii. 70 ; aedes DeumPenatium, v. 54 ; Di Manes serviles,vi. 24 ; ara deum, v. 38arae deorum, v. 52 ; liberorumdei nomina, ix. 55, 59 ; v. Samothraces; c/. Novensidesdextans, v. 172dextra, v. propter644diabathra, vii. 53Dialis flamen, v. 84 ; vi. 16Diana, v. 68 (Diviana), 74 ; vii. 16 ;Dianae templum, V. 43; i'. Titanis,Triviadil«lare, vii. 103dicare, vi. 61dicendi pars, viii. 44dicis causa, vi. 61, 95dice, vi. 30, 61, 62; dicit, vi. 78;dicere, vi. 42 ; dico dicebam dixeram,ix. 34 ; dicerem dicam, x.31 ; V. dodictata in ludo, vi. 61dictator, v. 82 ; vi. 61, 93 ; v. Poeteliusdictiosus, vi. 61dictum in mimo, vi. 61 ; dicta inmanipulis castrensibus, vi. 61dies, v. 68 ; vi. 4 ; ix. 73 ; x. 41 ; F.11 ; Dies Agonales, vi. 12 ; dieauspicate, v. 143 ; dies concepti-\iis, vi. 25 ; dies fasti, vi. 29, 53 ;die.s Fortis Fortunae, vi. 17 ; diesnefasti, vi. 30, 53 ; dies sacriSabini, v. 123 ; dies et nox, v. 11diemm nomina, vi. 10-32 ; dierumsingiilonim vocabula, vi. 33 ; v.AUiensis, atri, civilia, comitiales,februatus, intercisi, lupiter,Larentinae, prodixit, quando,quartus, septumus, statuti, VenusDiespiter, v. 66 ; Diespiter DiespitriDiespitrem, ix. 75, 77dilectus, vi. 65diligens, vi. 65 ; diligentior diligentissimus-ma, viii. 78Di Manes, v. deus. Manesdiminutio, F. 9diminutixTim, F. 9diobolares, vii. 64Diomedes -di -dis, x. 49Dionem, vi. 2 ; Diona, viii. 41, ix. 42Diores, ix. 12Diovis, V. 66, 84directus, v. derectusdiscere, vi. 62 ; discebam disco discam,didiceram didici didicero,ix. 96discerniculum, v. 129discessit, vi. 38disciplina, vi. 62 ; loquendi, x. 1discordia verborum novorum acveterum, v. 6

INDEXdiscrimen, vi. 12 (naturaleX 81 ; ix.56 ; X. 20, 77 (verbi) ; discriminaverborum, rerum, vi. 36, 38, viii.1, 2, 10, 14, 16, 17 (c/. 51), ix. 32,X. 64 (in rebus) ; discrimina litterarum,x. 62 ; discriminumniimerus, x. 10disertus, vi. 64disparilita.s vocis flgurarum, x. 36Dis pater, v. 66dispendium, v. 183 ; ix. 54dispensator, v. 183dispntare, disputatio, vi. 63disserit, vL 64dissimilia, viii. 34, etc ; v. similedi.ssimiIitudo, viii. 23, 24, 29, 31, 32,etc ; ix. 46, etc. ; x. 1,3, etc.distractio doloris, vii. 60distrahuntur, vii. 60dius, V. 66 ; vii. 34Dius Fidius, v. 66 ; aedes Dei Fidi,V. 52diva, V. Palatnadives, V. 92 ; viii. 17Diviana, v. Dianadividia, vii. 60divisio, vii. 60 ; ix. 97 ; x. 14, 15,17, 33 ; divisiones, ^-iii. 44, ix.95, 101 ; ex eodem genere et exdivisione, ix. 96, 97divum, v. 66 ; vii. 27, 50 ; divos,F. 1 ; sub divo, v. 66 ; divi pote.s,V. 58 ; V. deusdo dico addico, vi. 30do, r. ollusdoceo, \i. 62 ; docet, x. 17 ; doceodocui, X. 25 ; docentiir induciintur,^^. 62 ; docens, x. 17 ; doctus,F. 5. 8docilis, X. 17docte, viii. 12, 44 ; x. 17 ; F. 5. 8doctiioqui, vii. 41doctor, vi. 62doctus -a -um, viii. 46 ; ix. 67doctus -a doctissimus -a, viiL 77 ;doctus docte, viii. 12documenta, vi 62dodrans, v. 172dolia, r. slrpatadoliola, V. 157dolo dolas dolavi, ix. 108dolor dolori dolorem, x. 36, 42 ; r.distractiodolus malus, dolo malo, x. 51domare, vL 96domus, v. 160 ; domus domuisdomui, F. 17 ; v. hibemum, Maelius,videodonum, v. 175dos, V. 175Dossennus, vii. 95drachmae, ix. 85ducenti, v. 170 ; x. 43ducere ductor, vL 62DueUona Bellona, v. 73, vii. 49duellum, v. 73 ; vii. 49duigae, r. bigaeduini, V. binidulcis didcior dulcissimus, viii. 76duo duae, ix. 64, 65, 87 ; x. 24, 41,43, 45, 49, 67, 83duodenarius numerus, v. 34duonus, vii. 26duplex verbum, ix. 97 ; dnpliciavocabula, ix. 63 ; ». logoeduplicarii, v. 90dupondium dupondius, v. 169, 173 ;ix. 81 (-um), 83, 84dux, \i. 62 ; duces dux, x. 56, 57E, viii. 68; ix. 52; E:AE. v. 97,viL 96 ; EI :U8, ix. 80 ; E : A, v.114, vii. 94 ; E : I, \-i. 95 ; E : U,V. 91 ; E detritum, vii. 74 ; E exclusum,X. 57 ; E exemptum, ix.44 ; E exitus, x. 62ecbolicas aulas, v. 108echinus, v. 77ecurria, vi. 13edictiim, vi. 92edo, vi. 84 ; edo edi, x. 33edulium, WL 61 ; edulia, vi. 84edus, r. heduseffari, templa efEsntor, fines effiuitur,vi. 53eflata, vi. 53efiFutitum, ^ii. 93Egeria, vii. 42elegantia, ^-iiL 31elephans, elephantos, viL 39Elicii lovis ara, vi. 94eliquatum, viL 106elixum, v. 109eloquens, vi. 57eloqui, vi. 57eminisci, vi. 44emo emi, x. 33 ; r. homoem pa, viL 27645

INDEXensis ensiciilus ensicula, F. 10eo, V. i, iteEpeus, vii. 38Ephesi (loc.), viii. 21Bphesius, viii. 21, 22epichysis, v. 124epicrocum, vii. 52epigrammation, vii. 28Epimenides, vii. 3epityrum, vii. 85Epnlo, vi. 82equa, v. equiiseques equites, vii. 4 ; x. 28 ; v.ferentariu.s, magistereqiiile, viii. 18, 29, 52eqwiiria, v. eciirriaeqiiiso, viii. 14 ; x. 28equitatum, vii. 4, 103equus, vii. 4 ; viii. 11, 14, 52 ; ix.113 ; X. 4, 28 ; equus equi, ix. 63 ;equoeqimm, viii. 52 ; equus equa,ix. 28, 56 ; equus publicus, viii.71 ; equi dissimiles eadem facie,ix. 92, 93 ; V. Troianuserrare, vi. 96eras eri ero, x. 12esca, vi. 84escaria mensa, v. 118, 120esculentum, vi. 84esculetum, v. aesculetumEsquiliae, v. 25 (Exq-), 49, 50, 159(Exq-)Esquilina (regio), v. 45 ;(tribus), v.56 ; Esquilinus lucus, v. 50esum es est, v. sumet, viii. 9, 10Etruria, v. 30, 32, 46 ; vii. 35Etrusco ritu, v. 143etymologia, vii. 109etymologice, vii. 3etymologiis, vi. 39eu, vii. 93Euander, v. 21, 53euax, vii. 93eum {gen. pi.), vii. 26Buropa, v. 16, 31, 32 ; vii. 21 ; ix. 27exbolas, v. ecbolicasexceptum, v. osexcessit, vi. 38exercitus, v. 87 ; v. urbanusexiguitas, viii. 14exitium, v. 60exitus v. 60 ; exitus nominatuum,X. 21 ; exitus casus sexti, x. 62646ex iure (coctum), v. 109ex iure manum consertum vocare,vi. 64exorat, vi. 76ex parte, x. 84explanandi, v. gradusexpecto, vi. 82expensum, v. 183ex quadam parte, x. 74, 76, 78exquaeras, vi. 91Exquiliae, v. 25, 159 ; v. Esquiliaeexta oUicoqua, v. 104, c/. v. 98 ; extacaesa et porrecta, vi. 16, 31extemplo, vii. 13extergeor extersi, F. 5. 7extcrmentarium, v. 21externa, ix. 102extremum, vi. 59 ; v. littera, syllabaextrita, v. syllaba, I, R, SF : H, V. 97faba, ix. 38 ; x. 84fabri, vi. 78fabulae, vi. 55 ; nova fabula, vi. 58 ;V. agofacete, x. 17faciendi et patiendi copulae, x. 33facies, vi. 78 ; ix. 92facilis, X. 17facio facere, vi. 42, 77, 78 ; facit, x.17 ;poeta facit fabulam (nonagit), vi. 77 (v. ago, gerit)facerem faciam, x. 31 ; faciens, x.17 ; facere verba, vi. 78 ; v. lumen,lustrum, velaturafactiosae, vii. 66faculam, vi. 79 ; faculae, v. 137,X. 66facundi, vi. 52Facutalis lucus, v. 49, 50 ; v.FagutalFaeneratricem Feneratricem, vii. 96faenisicia fenisicia, vii. 96faenus, vi. 65Fagutal, lovis Fagutalis, v. 152 ; v.FacutalisFalacer flamen, pater, v. 84 ; vii. 45falces, V. 137 ; falce, F. 18 ;falciumfalces, non falceis, F. 20 ; v.arborariae, fenariae, lumariae,sirpiculaefalera, v. phaleraFalerii, v. Ill, 162Faliscus venter, v. Ill

INDEXfallacia, vi. 55falli, vi. 55falsum, vi. 55fiilx, V. falcesfama, \i. 55famigerabile, vi. 55familia, v. fiinesta, mater, pater,purgarefamosi, vi. 55fana, v. 51 ; vi. 54 ; fanorum servi,viii. 23 ; v. Fortis Fortunae, Liber,magmentaria, Quirinus, Sabinus,Saturnu-sfanatur, vi. 54far, v. 106 ; v. molaforcimina, v. Illfari fatur, vi. 52, 56 ; vii. 36larina, v. 106, 107lariolus, vi. 52ferticulum, v. Illfertum, V. Illfartura, v. Illfas, vu 31 ; v. quandofasces, V. 137fasciola, v. textafassi, vi. 55fasti, V. diesfastidium, v. 146fatales res, vi. 52fatidici, vi. 52Fatuae, vi. 55fatum, vi. 52fatuus, vi. 52 ; Fatuus, vi. 55fauces, V. 42 ; vii. 21 ;(non faux),X. 78Fauni (Faunus, Fauna), vii. 36Faustiani gladiatores a Faustio, ix.71Faustini gladiatores a Fausto, ix.71Faventinus a Faventia, viii. 83febri, F. 18Februarius, vi. 13 (Xonae), 34februatio, vi. 13febniatur, vi. 34februatus dies, vi. 13, 34februm ( = extremum), v. 79februm (Sabini, purgamentum), vi.13fedus, V. 97felix, V. quod bonumfemina, ix. 57 ; feminae, v. 130, ix.67 ; feminae nomen, ix. 40 ; v,masfemininum, F. 14 a, F. 14 b; femininumgenus, F. 9, F. 11fenariae falces, v. 137Feneratricem, v. Faeneratricemfenestra ta, viii. 29fenisicia, v. faenisiciaferae, v. 80 ; ferarum vocabula, v.100feralia, vi. 13fere, \ni. 92ferentarius, ferentarii equites, vii.57feretnim, v. 166feriae «conceptivae, vi. 26 ; conceptae,vi. 29 ; menstruae, vi.13 ; V. annates, Carmentis, Furinales,Latinae, paganicae, rex,sementivaeferio, F. 36 ; ferio feriam percussi,feriam ferio feriebam, ix. 98ferme, vii. 92fero, F. 36 ; fero ferebam, x. 14 ;ferte, vL 96 ; ferendo, viii. 57ferundo, v. 104Feronia, v. 74ferreus ferreei, viii. 70ferrifodinae non dicitvu-, viii. 62fertor non dicitur, viii. 57ferus fero ferum, x. 12fervere, \i. 84fetiales, v. 86fetus, V. 61fiber, V. 79fibra, V. 79ficedulae, v. 76ficta (verba), v. 9flctor, vi. 78 ; flctores, vi. 78, vii.44Ficuleates, vi. 18ficus, V. 76 ; ficus fici, ix. 80 ; v.ruminalisFidenates, vi. 18Fides, V. 74Fidius, V. Diusfidus, t'. foedusfiglinae, v. 50figuli : inter flgulos, v. 154figura figurae, \-i. 78 ; viii. 39, 71 ;ix. 39, 40, 42, 52, 93 ; x. 4, 11, 27,32, 33, 58, 77 ; tigura vocis, i. 25,36, 51 ; figura verbi, viii. 39, ix.37, X. 11, 25 ; figurae vocabulorum,ix. 55 ; v. Graecus, obliqui,singularis6i7

INDEXfilius, X. 59 ; Alius -a, ix. 55, x. 41filum, V. 113fimbriae, v. 79fingo, V. 7 ; vi. 78 ; flngo flngis, x.31finis fine, F. 18 ; v. copulae, eflfariflnitum et infinitum, v. 11 ; viii.45 ; ix. 31, 64, 85 ; x. 18, 20, 30V. infinitei. Humerus ; cf. templumfircus, V. 97flscina, v. 139fistula, V. 123fixum, F. 2Flaccus flamen Martialis, vi. 21flamen vinum legit, vi. 16 ; flamines,V. 84, vii. 45 ; v. Dialis, Falacer,Flaccus, Floralis, Furinalis,Martialis, Palatualis, Pomonalis,Quirinalis, Volcanalis, VolturnalisFlaminius circus, campus, v. 154flexura, x. 28Flora, V. 74, 158 ; vii. 45Floralis flamen, vii. 45fluctus fluctuis fluctui, F. 17flumen, v. 27, 28fluvius, V. 27fodari, vii. 100foditurne fodieturne, x. 32foedus, V. 86 ; fidus, v. 86 ; foedesum,vii. 27Fens, v. 74; vi. 22; fons, v. 123;fonts, F. 18 ; fonti et fonte, ix.112 ; fontis fontes, viii. 66 ; v.coronaFontanalia, vi. 22forda (quae fert in ventre), vi. 15Fordicidia, vi. 15forma, vi. 78 ; viii. 9, 47 ; Ix. 21,39-41, 82 ; x. 1 (vocabulorum), 22,27, 49 ; forma etymologiae, vii.109 ; forma in declinando, ix. 37 ;formae verborum, ix. 101, 102,109, 115, X. 56 ; formae Graecaeverborum, x. 70 ; v. similitudinumformido, vi. 48formo, vi. 78formula, ix. 103 ; x. 44 ; formulanumerorum, x. 43 ; formulae verborum,X. 33 ; analogiarum formulabinaria, denaria, x. 44fornices, v. 19 (caeli) ; x. 59Fortis Fortunae fanum, vi. 17 ; v.dies648Fort\ina, v. 74 ; vii. 93 ; v. vocabulafortunatum, v. quod bonumforum, V. 47, 145, 148, 149 ; vi. 59 ;vii. 94 ; ix. 17 ; v. Bovarium,Cupidinis, Holitorium, Piscarium,vetusfossa, v, 143 ; vii. 100Fratres, v. Arvalesfratria, v. 85fremere, vi. 67 ; vii. 104fremor oritur, vi. 67frendit, vii. 104frequens, vii. 99frequentare, vii. 99fretum fretu, vii. 22frigidum, v. 59 ; v. aquaefrigus, V. 60fringuillae vox, vii. 104fringvittis, vii. 104fritinnit, vii. 104frondenti coma, vii. 24fructus, V. 37, 40, 104liaiges, V. 37, 104 ; frugis frux(haec), ix. 76 ; frugis -i -em, ix.75frugi (non frugalus -a) frugalissumus-ima, viii. 77frumentum, v. 104fnior, v. 37, 104fugitiva, V. 5fulgur, V. 70fulguritum, v. 70, 150fuUo, vi. 43fulmen, v. 70fulmentum, viii. 10Fulvia, V. Basilicafumificus, vii. 38fundolus, V. Illfundula, v. 145fundus, V. 37funesta familia, v. 23funus, V. indicit, indicti\'umFurinalis flamen, v. 84 ; vi. 19vii. 45 ; Furinales feriae, v. 84Furnacalia, vi. 13furo, F. 36Furrina, v. 84 ; vi. 19 ; \'il. 45Furrinalia, vi. 19fustes, V. 137futis, V. 119futurum, viii. 20, 58G, i;. C ; GL, v. 134 ; GS : X, ix. 44Gabii, V. 33

INDEXGabinas ager, v. 33galea, v. 116galeritus, v. 76Galli obsederunt Romam, vi. 32decessus Gallorum, vi. 18 ; Galloruinos.

INDEXgrammatica antiqua, v. 7grammatici, x. 55, 75granarium, v. 105grandis olea, v. 108granum, v. lOtigreges, v. 76giibernator, ix. 6giistat, vi. 84guttus, V. 124liaedus, v. 97 ; haedi vox, vii. 104hahae, vii. 93harmonicae res, x. 64Harpocrates, v. 57haruspex, vii. 88 ; haruspice.s, v. 148liasta, V. 115hastati, v. 89haiirierint, F. 29Hectorem, viii. 72 ; Hectorem -is -a,X. 71 ; Hectores Hectoras, x. 69Hecuba, viii. 3hedus, v. 97heliae, v. hahaeheiulitabit, vii. 103Helena, viii. 80Heliconides, vii. 20Hellespontns, vii. 21hemisphaerium, vii. 7Heraclides Heraclide, viii. 68Hercules, v. 66 ; vii. 82 ; viii. 16(-les -lis, etx;.), 26 (-li an -lis) ; ix.79 (non Hercul) ; Hercules -li -lis,X. 49 ; Herculi immolata iuvenca,vi. 54 ; Herculi in aram, vi. 54 ;Herculi decuma, vi. 54 ; HerculesArgiv\is, V. 45herma, F. 14 bherois tritavus atavus, vii. 3heu, vii. 93hibernacula, vi. 9hibernum, vi. 9 ; hibernum domus,V. 162 ; hiberna triclinia, viii. 29hie, viii. 22 ; x. 18 ; hi his hibus,viii. 72 ; hie haec, viii. 45, x. 30hie hi haec hae, viii. 46 ; hiehunc, X. 50hiems, v. 61 ; vi. 9hilum hila, v. Ill ; hilum hili, ix.54hinnitus, vii. 103hinnulei, ix. 28hippos potamios, v. 77hireus, v. 97hirpices, v. irpicea650hirundo, v. 75Hispania, vii. 87historia (verborum), viii. 6holitor, vi. 64 ; holitores, vi. 20Holitorium Forum, v. 146holus holera, v. 104, 108, 146 ; x. 50homo, viii. 11, 12, 14, 44, 52, 79(non homen) ; ix. 113 ; x. 4, 6, 28,29 ; hominis, viii. 1 ; homines,; viii. 7, 14 homines imperiti etdispersi vocabiila rebus imponunt,X. 60 ; homines eniendi,ix. 93 ; hominum voeabula, v. 80 ;V. nihil, sene.scendihomunculus, viii. 14honestum, v. 73honor publicus, v. 80 ; honos, v. 73hora, v. 11 ; hora prima, secunda,ix. 73 ; horae lunares, ix. 26hordeum, v. 106 ; vi. 45 ; ix. 27horologium ex aqua, vi. 4horrent, vi. 45horti, vi. 20 ; hortorum, vi. 146quae in hortis nascuntur, v. 103hostia, vii. 31 ; v. agrestis, infulatae,piacularishosticus ager, v. 33Hostilia, v. OuriaHostilius rex, v. 155hostis, V. 3humanitas, viii. 31humanus -a -um, viii. 47humatus, v. 23humectus, v. 24humidus, v. 24 ; humidum, v. 59 ;humidissimus, v. 24humilior humillimus, v. 23humor humores, v. 24, 59-61, 63humus, V. 23, 59hypocorismata, F. 10I, viii. 67 ; I littera extrita, v. 96I:E, ix. 106; I additum, ix. 76;I exitus, X. 62 ; «;. Bi {imperaiive), vi. 96 ; v. itelapetus, V. 31Idus, vi. 14, 28, 29 ; v. Itus, luniaeignis, V. 59, 61, 63, 70 ; igne, F. 18 ;ignis et aqua in nuptiis, v. 61Ilium, viii. 56, 80Ilius (non Ilienus), viii. 56, 80 ; abIlo, viii. 80 ; Ilia, viii. 56, 80illex inlex, vi. 95 ; inlicis, vi. 94illiei, vi. 94

INDEXillicit, vi. 95 ; v. inlicereimmortales, v. 75 ; immortalia inlocis, V. 57impendium, v. 183 ; vi. 65imperandi (fecies), x. 31, 32 ; in imperando,x. 32 ; cum imperamus,is. 101 ; r. declinatusimperator, v. 87 ; vi. 77 ; viL 37impius, vi. 30impluium, v. 161imponenda (vocabula), vi. 3impos, r. inposimpositicia nomina, ^iii. 5 ; impositieiicasus, x. 61impositio verborum (vocabulonun),V. 1,3; vi. 3 ; vii. 32, 109, 110viii. 5, 7; X. 15, 16, 34, 51, 53,60, 61impositor, v. 18 ; vii. 1, 2impositum (verbxim, vocabulum,nomen), v. 1-3 ; viii. 1, 9, 10, 22,27 ; ix. 34, 52impurro, v. amburvomina«quabilitas, viii. 28, 30 ; ix. 1incertus ager, v. 33incessit, vi. 38incfaoata (=infecta) re.s, ix. 96; r.analc^iaincirctim, v. 25inclinanda verba, x. 13inclinatio inclinationes, ix. 1, 113e/. X. 13incommutabilia, ix. 99incrementiim, viii. 17increpitare, vi. 67incnltus, r. agerincnrvicenicum, v. 7indagabilis ambitus, r. 28indagare, v. 5indeclinabilia, x. 14, 79, 80, 82;viiL 9indicandi, ix. 101indicit belliim, vi. 61 ; indixit funns,vL 61indicium, vi. 61indictivum funus, v. 160 ; viL 42indiscriminatim, F. 25indoctus, ^iiL 62 ; indocti, ix. 22indusiatam, v. 131indusium, v. 131indutui, v. 131 ; s. 27infantes, vi. 52Infecta (verba), ix. 97, 100, 101 ; x.48 ; infecti verba, ix. 99, 101, x.33, 48; infecti tempora, ix. 96;infectae res, ix. 32 ; c/. inchoatainfeineitei, r. intiniteiinferi, viL 37, v. deus ; infera loca,V. terrainficientem esse, ^i. 78infima Nova Via, v. 43infinitei articuli, viiL 50 ; inflnitanatiua articulonun, viiL 52 ; inflnitaenaturae verborum, ^iiL 3 ;infinitum, v. 11, viii. 45, ix. 84V. finituminfrequens, vii. 99infulae, %iL 24infulatae hostiae, vii. 24ingeniosi, viiL 15ingluvies, F. 27inhumatus, v. 23inimicitia -am, x. 73initia, v. 60 rerum ; initia, v. 11 ;initiorum quadrigae, v. 12 ; initiaregis, V. 8 ; initia analogiae, x.53 ; V. nascendi. Samothracesinlex inlicis, r. illexinlicere populum, \i. 90 ; inliciatorad magistratus conspectum, vi,94inlicium vocare, >nsere, vi. 86-88,93-95 ; V. illicitinlocabilem, v. 14inminutio, F. 31 ainops, v. 92inpos, v. 4insane, vii. 86insicia, v. 110insidiae, v. 90insignia militaria, vii. 37insipitur, v. 105Insteianus Vicus, v. 52institutum, x. 27instrumentum, v. 105 ; instrumentummuUeris, ix. 22 ; r. rusticainsulsus, viii. 62intemi)esta nox, vi. 7 ; viL 72Interamna, v. 28intercisi dies, vL 31interduo, vii. 91intermestris, vi. 10interpolata (verba), v. 3interrex, vi. 93interrogando, ix. 32intertrigo, v. 176intertrimentom, v. 176651

INDEXintervallum miindi motiis, vi. 3, c/.V. 12intuiti, vii. 7intusium, v. indusiuminumbravit, vi. 4invident, vi. 80 ; invidit inyidendum,vi. 80Ion, viii. 21, 22lones, v. 146 ; vi. 9Ionia, V. 16 ; viii. 21irons, V. hircusirpices, v. 136irundo, v. hinindois ea id, ea eae, eins eaius, ei eae,ieis eais, viii. 51 ; eins viri, eiusmulieris, eius pabuli, viii. 51Isis, V. 57iste istunc, x. 50Italia, vii. 86ite (imperat.), vi. 96iter, V. 22, 35Itus (Tn.scornm), vi. 28iactarier, x. 70iaculum, v. 115iam, viii. 9ianens, vii. 26ianitor ianitos, vii. 27lanualis Porta, v. 165lanuarius, vi. 34 ; Kalendae lanuariae,vi. 28lanus, V. 165 ; lanus geminus, v.156, c/. vii. 26 ; lani signnm, v.165loum lovenim, v. lupiteriuba, vii. 76, c/. vi. 6iubar, vi. 6, 7 ; vii. 76iiibilare, vi. 68iucunditas, ix. 46index, vi. 61 ; indices, vi. 88indicare, vi. 61iudicinm, v. addixit, censorinm,decemvirnmiugernm, v. 35iuglans, v. 102Ingula, vii. 50iugnm, V. 135iumentnm, v. 135inngendi pars, viii. 44luniae Idus, vi. 17 ; Iwnius mensis,vi. 17, 33inniores, vi. 33, (declinationes) x.71 ; V. iuvenisInno, v. 65, 67 ; Innonis, viii. 49 ;652luno Covella, vi. 27 ; aedes lunonisLannvi, v. 162; InnoLncina, v.69, 74 ; aedes Innonis Lncinae, v.50 ; Incus lunonis Lucinae, v. 49 ;luno Regina, v. 67 ; v. Caprotina,lupiter, terralupiter, V. 65, 67, 84 ; vi. 4 ; vii.12, 16, 85 ; lupiter non Ions,viii. 74 ; lupiter lovi, viii. 34, x.65 ; luppitri, viii. 33 ; lupiterlovis lovem, viii. 49 ; lovis lovemlovi, viii. 74 ; loum lovemm, viii.74 ; aedes lovis, v. 41 ; aedes(lovis) in Capitolio, v. 158 ; aralovis Vimini, v. 51 ; sacellumlovis lunonis Minervae, v. 158dies lovis non Veneris, vi. 16lovis luno coniunx, v. 67 ; lovisfilium et filiam (non) lovem etlovam, ix. 55 ; antiquius lovisnomen, v. 66 ; v. Elicii, Fagutalinrgare, iurgium, vii. 93ins, V. ex iure, praetoriumluturna lympha, v. 71iuvencus, v. 96 ; v. Herculesiuvenis iunior, F, 31 a, F. 31 bKalendae, vi. 20, 27-29 ; v. lanuariaekalo, vi. 16, 27L, V. G ; L : 8, V. 79Ijacedaenionii, v. 146lact, v. 104lactuca, V. 104lacus, V. 26 ; v. Curtius, Cutiliensis,Velinilaena, v. 133laeta, vi. 50laetari, vi. 50laetitia, vi. 50lana, v. 113, 130, c/. 133; vii. 24;ix. 92 ; lana Gallicana et Apula,ix. 39 ; V. carere, vellerelanea, v. 130langula, v. 120laniena, viii. 55Lanuvium, v. 162lanx lance, x. 62lapathium, v. 103lapicidae, viii. 62lapidicinae, v. 151Larentalia, vi. 23Larentia, v. Acca

INDEXLarentiime dies, vL 23Lares, v. 74 ; Lares Lasibus, vi. 2 ;Lares ^iales, vL 25 ; Lamm Qiierquetulanumsacellum, v. 49 ; v.ManiaLarisaeus, v. ArgiisLarunda, v. 74Lasibus, v. Lareslata latae, x. 24Latiaris CoLlis, v. 52Latinae feriae, vi. 25, 29I^tine, vi. 6, 84 ; vii. 89 ; ix. 89Latinus (rex), v. 9, 32, 53, 144;Latinus (adj.), v. 29, passimLatinus casus sextus, x. 62Latinum vocabulum, Latina vocabula, v. '29, 68, 78, 79, 103, 167,vi. 35 ; Latini populi, vL 25Latini, v. 30, 43, 69, vi. 25,vii. 28, 33Lavemalis Porta, v. 163I^vinia, v. 144Lavinium, v. 144lavo lavor lavat lavatur lavarelavari lautvis sum, ix. 105-107lavo lavi, F. 5. 6 ; lavor lavi,F. 5. 7lea, F. 3leaena Leaena, v. 100lecte lectissime, vi. 36lectica, v. 166lectio, vL 36lectito, X. 33lector, vi. 36 ; viii. 57lectus lectulus, ix. 74 ; lecti, viiL32 ; ix. 47 ; lectulorum vocabula,V. 166 ; lectus mortui, v. 166 ; v.peslegasse mille aeris, ix. 83legati, V. 87 ; vi. 66legio, V. 87, 89; \i. 66; mileslegioniB, militis militem legionis,ix. 54legitima, vi. 66lego, vi. 36, 37 ; viii. 44 (lego legeas),ix. 102 ; X. 33 ; lego legis, x. 48lego legis legit, ix. 32, 101 ; legolegis legit legam, vi. 37 ; legolegam, ix. 96 ; lego legi, x. 25, 48 ;legi lego legam, viii. 3, 9, ix. 96 ;legebam lego legam, ix. 32, x. 31,47, 48 ; legit, viii. 11 ; leges lege,vi. 36 ; lege legito legat, ix. 101 ;legere, vi. 6tj ; legisti, vi. 35legi legisti, x. 48 ; legor, viii. 58legone legisne, x. 31 ;legem lecturus,vi. 36 ; v. flamenlegtilus, V. 94 ; leguli, vi. 66leguniina, vi. 66Lemnia litora, vii. 11 ; LemniusPhiloctetes, viu 11lentes, ix. 34leo, V. 100 ; vii. 76 ; leones, .vii.M) ; leonis vox, vii. 104Leontion, F. 38lepestae, v. 123lepus, v. 101 ; viii. 68 ; ix. 91, 94X. 8 ; lepus leporis, F. 5. 6 ; lepuslepori, viii. 34 ; lepores, ix. 94Lesas non Lesius, viiL 84Lesbo vinum, ix. 67letum, vii. 42 ; v. olioslex, vi. 71 ; lex legi, x. 47 ; leges,vi. 60, 66, viu 15, ix. 20; v.poetica, vetuslibella, v. 174 ; x. 38Libentina, v. VenusLiber, vii. 87 ; Liberi cognomentiunGraecum, vi. 68 ; Liberi lanum,v. 14 ; sacerdotes Liberi anus, vi.14 ; 1^. LoebesoLiberalia, vi. 14liberi, ix. 59 ; r. deus, servusRomanusliberti, v.libertini a municipio manumissi,653

INDEXviii. 83 ; orti a publicis servisRomaiii, viii. 83Libethrides, vii. 20libidinosus, vi. 47libido, vi. 47 ; x. 60, 61 ; v. lubidoLibitina, v. VenusLibo, V. Poeteliuslibra, v. 169, 174, 182 ; vii. 14 ; v. aeslibrarii, viii. 51 ; ix. 106libuin, V. 106 ; libum libo, ix. 54liba, vii. 44Libya, vii. 40 ; viii. 56Libyatici non dicitur, viii. 56Libyci, vii. 39lictores, vii. 37ligna, vi. 66ligiiicidae non dicitur, viii. 62ligo, v. 134lilium, V. 103lima, vii. 68 ; limae, x. 14limax, vii. 64lingers, vi. 96lingua Latina, v. 1, 29 ; vii. 55, 110 ;viii. 58 ; ix. 113 ; lingua nostra,V. 3, 29 ; V. Armenia, Graecus,Osca, Sabinuslingula, vii. 107lingulaca, v. 77lintres non lintreis, F. 20linum lino, ix. 64liquidum, vii. 106liquitur, vii. 106lis, vii. 93liticines, v. 91 ; vi. 75litora, V. Lemnialitterae, v. 30 ; vi. 2, 66 ; vii. 2 ;viii. 63 ; ix. 52 ; x. 25, 26, 55, 82antiquae litterae, v. 143, vi. 33litterae Latinae, v. 73, vii. 2, ix.51 ; litterae Graecae, viii. 64, 65 ;interpretationem exili littera expeditam,vii. 2 ; littera praeterita,vii. 2 ; littera extrema, ix. 44, x.21, 25 ; littera extrita, v. 96litterani adicere, vii. 1 ; litterasassumere, vi. 2 ; litteras mittere,vi. 2 ; litterarum vocabula, ix.51 ; V. additio, commutatio,demptio, discrimina, productio,traiectiolixulae, v. 107locare, v. 14, 15locarium, v. 15locatum, v. 14654locus, V. 11-15, 57 ; viii. 12 ; loca,vi. 97, vii. 5 ; loca agrestia, vii.10 ; loca Europae, v. 32 ; locimuliebres, v. 15 ; loca naturae,V. 16 ; loca urbis, v. 45 ; origineslocorum, vii. 110 ; vocabula velverba locorum, v. 10, 184, vi. 1 ;V. animalia, caelum, Caeriolensis,terra, Tutilinae, urbsLoebeso ( = Libero), vi. 2logoe, x. 43 (duplex, decemplex),c/. X. 2, 37, 39lolligo, V. 79longavo, V. Illlongus, viii. 17 ; longiores, x. 29loquax, vi. 57loquela, vi. 57loquor, viii. 59 ; loquontur, vi. 1 ;loqui, vi. 56 ; loquens locuturuslocutus, viii. 59 ; v. concinne,disciplinalorica, v. 116Lua Saturni, viii. 36Lubentina, v. Venuslubere, vi. 47lubido hominum, x. 56 ; lubidinem,F. 4 ; V. libidoLuca bos, vii. 39, 40 ; Lucana bos,vii. 39Lucani, v. 32, 111 ; vii. 39, 40Ijucana origo, v. 100Lucanica, v. IllUlcere, vi. 79Luceres, v. 55, 81, 89, 91hicerna, v. 9, 119Lucia, ix. 61 ; Lucia Volumnia, ix.61Lucienus, vi. 2Lucina, v. lunolucifer (stella), vii. 76Lucius, ix. 60 ; Lucii, vi. 5 ; v.Aelius ; c/. LuciaLucretia, vi. 7lucrum, V. 176L\icumo, V. 55lucus, V. Esquilina, Facutalis, luno,Mefitis, Poetelius, Venusludens, vi. 35ludus, ix. 15 ; ludi quibus virginesSabinae raptae, vi. 20; v. Apollinares,dictata, Ta\u'iilumariae falces, v. 137lumecta, v. 137lumen facere, vi. 79

INDEXLuna, luna, v. 68, 69, 74 ; vi. 10vii. 16 ; ix. 25 ; v. hora, novalunaris, v. horaluo luam, viii. 36; luit, ix. 104;luendo (id est solvendo), vi. 11V. solvuntLupe (voc.), vii. 47Lupercal, v. 85 ; vi. 13 ; Lupercalia,vi. 13 ; Lupercalibus, v. 85Luperci, v. 85 ; vi. 13 ; Lupercinudi, vi. 34lupinum, ix. 34lupus, v. 77 (piscis) ; ix. 28 ; lupuslupi, F. 5. 6 ; lupus lupo lupe,viii. 34, 68, ix. 91, 113 (piscis);lupi vox, viu 104lusciniola, v. 76luscus (non luscior luscissimus), Ix.72lustrare, vi. 93lustrum, vi. 11, 22 ; lustrum facere,condere, vi. 87lutra, V. 79lux, vii. 40 ; v. primoLyde, vii. 90lympha, v. 71 ; vii. 87 (a Nympha) ;V. Commotiles, lutiu'iialymphata, lymphatos, vii. 87Lysippus, ix. 18M :N, vi. 75Macedonia, vii. 20Macellum, v. 146, 147, 152Macellus, v. 147macer macri macrum, ix. 91 ; x. 28 ;macer macricolus macellus, viii.79 ; macri, vi. 50 ; macrior macerrimus,viii. 77Maecenas (non Maecenius), viiL 84Maelius (et eius domus), v. 157maerere, vi. 50Maesium, v. Mesiummagida, v. 120magis, viii. 9 ; ix. 73magister equitum, populi, v. 82vl 61magistratus, v. 82 ; vi. 87, 91 ; viii.83 ; V. vitiomagmentaria fana, v. 112niagmentum, v. 112magnetae lapides, ix. 94magnitudo, ix. 74 ; magnitudinisvocabula, viii. 79magnus, v. deus, pesmaiores, v. 5 ; vL 17, 33 ; ix. 16Mains (mensis), vi. 33malaxare, vi. 96malum, v. 102 ; mala, ix. 92 ; v.Punicummalus mali, x. 68 ; malum peiuspessimum (non malius malissimum),viii. 75, 76 (peium nondicitur) ; v. bonum, dolusmalva, v. 103 ; malva malvaceus,F. 10Mamers (Sab.), v 73maramosae, viii. 15Mamuri Veturi, vi. 49mancipium, vi. 85 ; vii. 105mandier, vii. 95manducari, vii. 95Manducus, vii. 95mane, vi. 4 ;(manius manissimenon dicuntur) viii. 76, ix. 73magis mane surgere, ix. 73 ;primomane, ix. 73Manes, v. 148 ; v. deusMania mater Lamm, ix. 61manica, vi. 85manicula, v. 135Manilius Maniliorum, viii. 71manipularis, vi. 85manipulus, v. 88; vi. 61, 85Manius, ix. 60Manlius, T., consul, v. 165mantelium, vi. 85manubriinn, vi. 85 ; manubria, viii.15, c/. V. 118manum ( = bonum), vi. 4manumissi, viii. 83 ; v. vitiomanupretium, v. 178; vi. 85manus, vi. 85 ; ix. 80 ;quae manufacta, V. 105 ; v. adserere, consereremappae tricliniares, ix. 47marcescere, vi. 50Marcius Marci, viii. 36Marcus Marci, viii. 36 ; MarcusMarco, viii. 46, x. 51 ; Marcusnon Marca, ix. 55 ; v. Perpennamargaritum margarita margaritanim,F. 14 c.Maro, V. 14Mars, V. 73 ; vi. 33 ; Mars Martes,X. 54 ; Marspiter Marti, x. 65Maspiter, viii. 40 ; Maspiter sednon Maspitri Maspitrem, ix. 75 ;Marspitrem, viii. 33655

INDEXMartialis (flamen), v. 84 ; vii. 45V. FlaccusMartins (mensis), vi. 33 ; Martiuscampus, V, 28, vi. 13, 92mas femina, v. 58, 61 ; viii. 7, 40Ix. 38 ; mas femina neutrum, ix.55, 57, 59, 62, cf. viii. .36, 47, 78,X. 22 ; V. genus ; cf. viriliamasculinum, F. 14 a, F. 14 b ; masculinogenere, F. 11Maspiter, v. Marsmatellio, v. 119mater, x. 41 ; matres familias, vii.44 ; V. Mania, Ops, terramateria, x. 11, 36Matralia, v. 106mattea, v. 112ma tula, v. 119Maurus Maurice, F. 5. 8maximus, v. Circus, CluacaMecinus, v, maximusmedia, v. 118; media nox, x. 41;media vocabula, viii. 79medicina ars, v. 93 ; vii. 4 ; ix. Illmedicus, v. 8, 93 ; ix. 11 ; x. 40Meditrinalia, vi. 21Mefitis lucus, v. 49Megalesion, vi. 15 ; Megalesia, vi. 15mel mellis melli melle, viii. 63melander, v. 76Melicertes Melieerta, viii. 68 ; ix. 91melios, vii. 20 ; meliosem, vii. 27V. bonusmelius, v. bonusMelius, V. Maeliusmeminisse, vi. 44, 49memoria, vi. 44, 49Menaechmum -mo, x. 38 ; Menaechmigemini, viii. 43mendicus, v. 92mens, v. 59 ; vi. 43-45, 48, 49 ; mensmentium mantes, viii. 67 ; mentesnon menteis, F. 20 ; v. agitatusmensa, v. 118 ; vii. 43 ; v. escaria,urnarium, vasaria, vinariamensis, v. 69 ; vi. 10, 33 ; mensiumnomina, vi. 33 ; v. novus, lanuarius,Februarius, Martius,Aprilis, Maius, lunius, Quintilis,Septembres, October, Decembermenstruae, v. feriaemensura, ix. 67 ; mensura ae pondera,ix. 66menta, v. 103656meo meas, ix. 109mera ( = sola), v. 76merces, v. 44, 175, 178 ; .vii. 52viii. 19 ; merces non merceis,F. 20mergus, v. 78meridies, vi. 4 ; vii. 7 ; x. 41menila, v. 76 ; ix. 28, 55 (non memlus); merula merulae, x. 66merulae vox, vii. 104Mesium rustici, non Maesium, vii.96messor, viii. 57Metellus Metella, ix. 55meto metis, x. 31 ; meto raetammetebam, ix. 89 ; metendo, viii.57Meto Metonis Metonem, ix. 89Mettius, V. Curtiusmetuere, vi. 48 ; metuit (nonsperat), vi. 73 ; metuisti, vi. 45metus, vi. 45Mico, ix. 12miliariae (aves), v. 76miliariae (decuriae numerorum), ix.87 ; miliaria (vocabula), ix. 85miliens, ix. 88militare aes, v. 181 ; v. raudus ;militiiria, v. insignia, ornamentamilites, v. 89 ; milites aerarii, v.181 ; militis stipendia, v. 182 ; v.legio, tribunimilium, v. 76, 106mille milia, ix. 82, 85, 88 ; milleaeris, ix. 83 ; hi, hoc, huius,liorum mille, ix. 87, 88 ; haec duomilia, ix. 87Minervae, v. 74 ; aedes Minervae,vi. 17 ; 1'. lupiterMinervium, v. 47minima vocabula, viii. 79rainores, ix. 87minuendi (genus declinationis), vii.52minusculae, v. Quinquatrusminuta opera, v. Myrmecidismiraculae, vii. 64miriones, vii. 64miser, v. 92mitra, v. 130moenere, v. 141moenia, v. 141moerus, v. 141 ; moeri, vi. 87 ; v.murus

INDEXmola (sale et farre), v. 104 ; molae,V. 138monere, vi. 49 ; monerint, vii. 102mouimenta, vi. 49monitor, v. 94montes (Romae), v. 41 ; vi. 24monte, F. 18 ; monti monte, ix.112 ; montes montis, viii. 66 ; t>.Albanus, Caelius, Cespius, Oppius,Ripaei, Saturnius, Tarpeiusmorbus, v. quartus, septumusmors, V. vitamortales, v. 75 ; mortalia in locis,V. 57morticinum, vii. 84mortui lectus, v. lectusmotacilla, v. 76motus, v. 11, 12 ; vi. 3, 4, 8 ; ix. 34(caeli) ; motus in mari, ix. 25 ; i'.solmox, X. 14, 79, 80Mucialis collis, v. 52Mucionis porta, v. 164Hucius, Q., vi. 30; viii. 81; Muciet Bruti sedulitas, v. 5 Mucia,;viii. 81Mugionis, v. Mucionismugit, vii. 104mulgere, ^-i. 96muliebre, viii. 46, 51 ; muliebria,ix. 41, 48, 110, X. 30; nominamuliebria, viii. 36 ; v. locus, mundus,stola, tunicamulier, viii. 80 ; x. 4 ; mulieris,mulieribus, viii. 51 ;praenominamulierum antiqua, ix. 61 ; v.antiqua, cum mulieremulta, V. 95, 177multitude multitudinis, viii. 7, 14,36, 46, 48, 60, 66, 67; ix. 64-66,68, 69, 76, 81, 82, 84, 85, 87; x.28, 33, 36, 54, 56, 58, 59, 66, 83multitudinis solum, ix. 63, x. 54,66 ; multitudinis vocabula, ix.64, 65 ; multitude verborum, vi.35, 40 ; V. copulaemulus mula, ix. 28 ; v. mutuummundus, vi. 3; (=omatus muliebris),v. 129 ; v. intervallum, terramunicipes, v. 179municipium, viii. 83munus, V. 141, 179muraena, v. 77 ; ix. 28, 113Murciae, v. CircusVOL. IIMurmecidis, v. Myrmecidismurmurantia litora, vi. 67murmurari, vi. 67murtatimi, v. 110Murteae Veneris sacellum, v. 154murtetum, v. 154murus, V. 143; v. ctrcum, moerus,postici, Saturnii, terreusMusa, ix. 63 ; Musae, vii. 20, 26,ix. 64musica, ix. Illmussare, vii. 101mustela, ix. 113muti, vii. 101Muti, V. Mucinsmutuum, V. 179 ; mutua mull, vii.28Myrmecidis opera minuta, ix. 108obscuram operam Myrmecidis,vii. 1mysteria, vii. 11, 19, 34mystica vada, vii. 19N, cf. MXaevia Porta, v. 163 ; Naevianemora, v. 163naevus, F. 13nanus, v. 119jnarratio, vi. 51narro, vi. 51narus, \i. 51nascendi initia, v. 15 ; causa, v. 61cf. V. 60, 70natare, viii. 74 ; ix. 71natator, v. 94natura, ix. 37, 38, 58, 62, 63, 70, 72,76, 78, 94, 101 ; X. 15, 17, 24, 41(quadruplex), 51-53, 55, 56, 60,61, 83, 84, F. 6 ; natura novenaria,octonaria, ix. 86 ; natiu^ sermonis,viiL 25 ; natura verborum,viii. 43, X. 51, 74; naturae verbi,v. 2 ; naturarum genera, x. 28V. copulae, gentilicia, infinitei,locusuaturalis, v. casus, declinatio, discrimen,genusnaviculae ratariae, vii. 23navis longa, vii. 23 ; nave, F. 18Xeapolis (Xovapolis), v. 85 ; vi. 58necatus sum necor necabor, x. 48necessitas, viii. 31nefas, vi. 30, 31ne£asti, v. dies657

INDEXneraus nemora, v. 36 ; ix. 94 ; x. 8,60 ; V. Naevianeo nes, ix. 109Neptunalia, vi. 19Neptunus, v. 72 ; vi. 19 ; v. Salacianequam, x. 79-81Nestor, viii. 44; Nestorem Xestoris,viii. 72, X. 70 ; Nestores Nestoras,X. 69neutnim, viii. 46, 51 ; ix. 41 ; x. 8,31 (neutra) ; F. 8 (genus) ; v. mas,similenexus, nexum, vii. 105niger nigricolus nigellus, viii. 79nihil niliili, ix. 53 ; nihili nihilum,homo nihili (non hili), ix. 54,X. 81nobiles nobilitas, viii. 15Noctiluca, V. 68 (et eius templum)vi. 79noctua, V. 76noctulucus, V. 99Nola, Nolani, viii. 56nolo, X. 81nomen nomina, viii. 13, 14, 40, 45,53, 56, 80 ; ix. 40, 43, 52, 54, 89, 91 ;X. 20, 21, 27, 53, 54, 65, 80 ; nomenan vocabulum, viii. 40 ; nomencommune, ix. 89 ; nomen Latinum,V. 30, 119 ; Latina nomina,vii. 109 ; nomina nostra, vi. 2,viii. 64, 84 ; v. deus, dies, Graecus,impositicia, lupiter, mensis,muliebre, Persarum, pisces, proprio,servile, servus, Syriacum,tralaticio, translaticium, virilenomenclator, v. 94nominandi casus, viii. 42 ; ix. 76,77 ; x. 23, 65 ; nominandi genusdeclinationis, viii. 52 ; v. casusnominare, vi. 60nominativus, x. 23nominatus, viii. 45, 52, 63 ; ix. 69,70, 95, 102; x. 18, 20, 21, 30;V. exitus, Graecanica, speciesNonae, vi. 27-29; v. Caprotina,Februariusnonaginta, ix. 86, 87Nonalia .sacra, vi. 28nongenta, ix. 86, 87non hili, v. ilihilnonussis, v. 169nostri, v. 36, 100, 166 ; vi. 2, 6 ; vii.39, 87, 88 ; ix. 69 ; x. 71 ; nostra658memoria, vi. 40 ; v. antiqua,colonia, consuetudo, deus, nomina,provincia, sacerdotes,sacra, verba, vetus, vocabulumnothum (genus similitudinis), x.69 ; notlia (verba), x. 70 ; nothaedeclinationes, x. 71novalis (ager), v. 39 ; vi. 59nova luna, vi. 28Nova Via, v. 43, 164; vi. 24, 59;V. infimaNovapolis, V. Neapolisnovem, ix. 86, 87novenarius, v. natura, numerusNovendiales, vi. 26Novensides, v. 74novicius, vi. 59 ; novicii servi, viii. 6novitas, vi. 59novus annus, mensis, sol, vi. 28sub Novis, vi. 59 ; novius novissimum,vi. 59 ; novissimum vesper,ix. 73 ; V. fabulae, nova luna.Nova Via, senex, verbumnox, vi. 6 ; x. 14, 41 ; v. dies, intempesta,silentiumNuma, V. Pompiliusnumen, vii. 85Numerius (non Numeria), ix. 55numerus, ix. 66, 67, 81, 85 ; x. 65numeri, ix. 65, 84, 87, x. 41, 43numerus novenarius, ix. 86 ; denarius,V. 170 ; duodenarius, v.34 ; centenarius, v. 88 ; numerussingularis, v. 169 ; numerus verborum,vi. 38, 39, viii. 3 ; numeriantiqui, ix. 86 ; flniti, x. 83 ; v.actus, decuriae, formula, gradus,regulanummi, v. 173, 174 ; ix. 80, 85 ; x.41 ; V. addicinuncupare, vi. 60nuncupatae pecuniae, vi. 60nuntium, vi. 86nuntius, vi. 58nuptiae, v. 72; vi. 70; vii. 28, 34;X. 66, 67 (non nuptia) ; v. ignisnuptu ( = opertione), v. 72nuptus, V. 72nutus, vii. 85nux, V. 102Nympha, v. lymphaO exitus, X. 62obaeratus, viL 105

INDEXobiurgat, vii. 93obliqui casus, viii. 1, 2, 6, 7, 16, 46,49, 51, 69, 74 ; ix. 43, 54, 70, 75-77, 79, 80, 89, 90, 103 ; x. 22, 44,50-52, 58, 59 ; obliquae figiirae, x.53 ; obliqiiae declinationes, x. 44 ;obliqui versus, x. 43oblivia verba, v. 10oblivio, V. 5 ; vii. 42obscaenum obscenum, vii. 96, 97obscuritas verborum, vi. 35, 40obscuritates grammaticonini, x.75obsidium, v. 90occasus (solis), vi. 4, 5 ; vii. 7, 51ocimum, v. 103ocrea, v. 116October mensis, vi. 21octonaria, v. natiiraodor olor, vi. 83odora res, vi. 83odorari, vi. 83odoratus, vi. 83offula, V. 110olea, V. 108olet, vi. 83oleum (non olea), ix. 67olitores, v. holitoresoUtx)rum, v. Holitoriumollaner, v. olla veraoUa vera arbos, vii. 8ollicoqua, v. extaollus olla, vii. 42 ; olla centuria,vii. 42 ; ollus leto datus est,vii. 42olor, V. odorolus olera, v. holusOlympiades, vii. 20Olympus, vii. 20omen, vi. 76 ; vii. 97omnicarpa, v. 97Opalia, vi. 22Opeconsiva, vi. 21opercula, v. 167operimenta, v. 167Opimia, v. Basilicaopinio, V. 8oppidum, V. 8, 141 ; x. 20 ; oppiduniin circo, v. 153 ; oppida condere,V. 143 ; V. antiqua, GraecusOppius Mons, v. 50Ops, V. 57, 64 (mater), 74 ; vi. 22Ops Consiva (et eius sacrariiim),vi. 21 ; V. terraoptandi species, x. 31, 32 ; in optando,ix. 32optimum, v. bonusoptiones, v. 91opulentus, v. 92opus, v. 64oratio, vi. 64, 76 ; vii. 41 ; viii. 1(tripertita), 38, 44; ix. 9, 11, 30,32, 33, 35, 36, 45, 46, 48, 56, 112X. 14, 42, 49, 55, 64, 65, 68, 77(vocalis), 79 ; oratio poetica, vi.97 ; oratio soluta, vi. 97, vii. 2,110, X. 70; V. partes, scientiaorator, vi. 42, 76 ; vii. 41 ; viii. 26ix. 5, 115orbis, V. 143 ; orbe, F. 18orchitis, v. 108Orcus, V. 66 ; vii. 6ordo, X. 67 ; ordo declinatuum, x.54 ; ordines transversi et derecti(pel directi), x. 22, 23, 43oriens, vi. 4 ; vii. 7origo, origines verborum, v. 3, 4, 6,7, 92, 166; vi. 1, 37, 97; vii. 4,47, 107, 109 ; viii. 58 ; origo duplex,vii. 15 ; origo nominatus, ix. 69 v, ;Graecus, locus, Lucana, poetica,Sabin\is, similitudoOrion, vii. 50ornamentum, vi. 76 ; ornamentamilitaria, vii. 37ornatus (muliebris), v. 129, c/. v.167oro, vi. 76 ; v. causamortus, V. hortusortus (solis), vi. 6 ; vii. 83 ;(Luciferi),vii. 76OS exceptum, v. 23 ; ossa, v. GalliOsce, V. 131, F. 5. 8Osci, vii. 29, 54 ; Osca lingua, vii.28 ; Oscus Osce, F. 5. 8oscines, vi. 76o.sculura, vi. 76osmen ( = omen), vi. 76 ; vii. 97ostrea, v. 77ovile, viii. 18, (non ovarium) 54 ;ovilia, ix. 50ovillum pecus, v. 99ovis, V. 96 ; viii. 46, 54 ; ix. 76 (nonovs), 113; ovis ovi, %iii. 34; oviove, viii. 66 ; oves ovium, viii.70, ix. 26 ; v. peculiariaeovum, V. 1120X0, F. 24659

INDEXpa ( = patrem), vii. 27pabulum, viii. 51Paganalia, vi. 24paganicae (feriae), vi. 26pagus, vi. 26 ; v. Succusanuspala, V. 134Palanto, v. 53Palatina tribus, v. 56 ; regio, v. 45 ;Palatini, v. 53, 54 ; Palatinum, v.antiquaPalatium, v. 21, 53, 68 (Bal-), 164Palatua diva, vii. 45Palatuali.s flamen, vii. 45Pales, V. 74 ; vi. 15Palilia, vi. 15palla, v. 131Pallantes, v. 53pallium, v. 133, 167 ; viii. 28 ; ix. 48palma, v. 62palpetras, non palpebras, F. 23Paluda, vii. 37paludamenta, vii. 87paludatus, vii. 37palus, V. 26panarium, v. 105Pandana Porta, v. 42pandura, viii. 61panificium, v. 105panis, V. 105 ;pani.s pastillus pastillum,F. 10pannus, v. 114panther, v. 100 ;panthera, v. 100,(non pantherus) ix. 55, F. 3pantherae, vii. 40Pantheris, v. 100panuvellium, v. 114Pappu.s, vii. 29, 96parapechia, v. 133Parcae, vi. 52Parentalia, vi. 23parentant, vi. 23 ;parentare, vi.13, 34parentum parentium, viii. 66paries, ix. 41Paris, vii. 82 ; viii. 80 ; Paris Pari,viii. 34parma, v. 115Parma (urbs), viii. 56Parmenses (non Parmani), viii. 56jjaro paretur parator, x. 32 ;pareparavi, F.-5. 6partes animae, ix. 30 ;partes orationis,viii. 11, 38, 44, 53, x. 7,c/. vi. 36, viii. 48, ix. 31, x. 17660V. casuale, ex quadam, scaena,templum, urbsparticipalia, x. 34participia, viii. 58; ix. 110patella, v. 120patena, v. 120pater, v. 65 ; x. 41, 59 ;paterpatres, viii. 48 ;pater familias,patres familias familianim, viii.73; patres ( = sena tores), vi. 91;V. Dis, Falacerpaterae, v. 122patiendi, v. faciendipatricus (casus), viii. 66, 67 ; ix. 54,76, 85 ; cf. viii. 16patrius casus, F. 17Patulcium, vii. 26pauper, v. 92 ;pauper (sed nonpaupera) pauperrumus pauperrima,viii. 77 ;pauper pauperior,F. 31 apavet, vi. 48pavo, V. 75pavor, vi. 48pecten, v. 129pectere, vi. 96pectunculi, v. 77peciilatns, v. 95peculiariae oves, v. 95peculium, v. 95pecunia, v. 92, 95, 175, 177, 180,181 ; vi. 65, 70 ;pecunia debita,vii. 105 ;pecuniae signatae vocabula,V. 169 ; v. nuncupataepecimio.'^us, v. 92 ; viii. 15, 18pecus (pecoris), v. 80, 95, 110 ; vii.14 ; ix. 74 ;pecudem, v. 95pecudis caro, v. 109 ;pecus ovillum,V. 99 ; v. agopedem posuisse, v. 96pedica, V. 96pedisequus, v. 96peius, V, malumpelagus sermonis, ix. 33Peles, X. 69Pelium, vii. 33pellesuina, viii. 55pellexit, vi. 94pelliaria taberna non dicitur, viii. 55pellicula, vii. 84peloris, V. 77pelvis, V. 119penaria, v. 162Penates, v. deus

INDEXpensio prima, secunda, etc., v. 183Percelnus Percelna, viii. 81perciU)uit, is. 49percutio percussi percutiam, is.08perduelli-s, v. 3 ; ^ii. 49peregrinus, v. 3 ;peregrinus ager,V. 33 ;perpgrina vocabiila, v. 77,100, 103, 167Ijerfectum, ix. 100, 101 ; x. 48perfect! (verba), ix. 96, 101, x.33, 48; perfecta, ix. 97, 99, x.48 ;i)erfectae res, ix. 32, 96 ; v.analogia, similitiidoPergama, vi. 15 ; Pergamum, viii.56Pergamenus (non Pergamus -a),viii. 56pergendo ( = progrediendo), v. 33periacuit, ix. 49peripetasmata, v. 168peristromata, v. 168pema, v. 110perorat, vi. 76Perpenna, viii. 41, SI (non Perpennus),ix. 41 ; x. 27 ; MarcusPerpenna, viii. 81Persarum nomina, viii. 64persedit, ix. 49perseverantia, v. 2persibus, ^ni. 107I^ersonae, verbi, viii. 20 (quiloqueretur, ad quern, de quo) ; ix.32, 95, 100-102, 108, 109; x. 31,32 ; V. copulae, secundaperstitit, ix. 49pertinacia, v. 2pervade, v. polumpes, V. 95 ;pes lecti ac betae, vi. 55 ;pes magnus, v. 95 ; v. pedempessimum, v. malusphalera -am, x. 74phanclas, v. zanclasPhanion, F. 38Philippi caput, ix. 79Philolacho, ix. 54Philomedes -dfs, viii. 68 ; ix. 91philosophia, v. 8Phoenice, v. 31Phoenicum, v. PoenicumPhryx Phryge Phryges, ix. 44physici, v. 69 ; x. 55piacularis hostia, vi. 30piaculum, vi. 29, 53pila terrae, vii. 17 ;pila aequa, vii.19pilani, v. 89pili in corpore, vi. 45pilum, v. 116, 138pingo pingis, x. 31 ;pingo plnxi,F. 5. 6pinnae, v. 142pinus, V. 102pipatus pullorum, vii. 103Pipleides, vii. 20pipulo, vii. 10aPiscarium Fonim, v. 146pisces, viii. 61 ; ix. 28, 113 ;pisciumnomina, vii. 47 ;piscium vocabula,V. 77pisciceps non dicitiu-, viii. 61piscina (non dicuntur piscinula piscinilla),ix. 74pistor pistori, x. 69pistrinum pistrina, v. 138 ;pistrinumpistrilla, F. 10pistrix, v. 138placenta, v. 107platanus platani, ix. 80plaustrum, v. 140Plautius Plauti, viii. 36Plautus Plauti, viii. 36plebs, V. tribuniphut, ix. 104plumbea, ix. loioplumbum (non plumba), ix. 66plura, ix. 32 ; x. 31plusima, vii. 27pocillum, ix. 66poculum, vi. 84 ; viii. 31 ;pocula,V. 122poema non poematum, F. 21 ;poemata,vii. 2 ;poematorum, F. 21poematis, vii. 2, 36, viii. 14, F. 21poematibus, vii. 34poena, v. 177Poeni, V. 113, 182 ,Poenicum, v. 113 ; Poenicum vocabula,viii. 65poeta poetae, v. 22, 88 ; vi. 52, 58,67, 77, 83 ; vii. 36, 110 ; ix. 5, 17,65, 78, 115 ; X. 35, 42, 70, 73, 74vocabula apud poetas, v. 1vocabula a poetis comprehensa,V. 10 ; vocabula ^jetarum, vii. 1poetanim verba, v. 7, 9 ; verba apoetis posita, viL 5 ; verba apudpoetas, vii. 107 ; cf. poetica, vetus661

INDEXPoetelius lucus, v. 50 ; C. PoeteliusLibo Visolus dictator, vii. 105poetica verba, v. 9 ; vii. 3 ;poeticaanalogia, x. 74 ; de pocticis verborninoriginibus, vi. 97 ; legepoetica, vii. ISpoetice, vii. 2poUuctiim, vi. .54Pollux, v. 58, 73 ; Polhices, v. 73polus, vii. 14 ; ix. 24 ;pervadepolum, vii. 14Polybadisce, vi. 73polypus, V. 78pom ( = potissimuni), vii. 26poma, ix. 93, c/. v. 108pomerium, v. 143Pomona, vii. 45Pomonalis flamen, vii. 45Pompilius (Xuiiia), v. 157 ; vii. 4.5 ;Pompili regntim, vii. 3 ;Pompiliorege, v. 165pondera, v. mensurapons, V. 4, S3 ;ponte, F. 18 ; PonsSublicius, V. 83, vii. 44pontifex, v. 180 ; vi. 61 ;pontiflces,vi. 26, 27, 54 ;pontufices, v. 83,vi. 61Poplifugia, vi. 18populus, v. 1, 35 ; viii. 6 ; ix. 5, 6,18, 114; X. 16, 74; v. inlicere,Latinus, magister, rex, Romanusporea, v. 39porcus, V. 97porrecta, v. extaporta, V. 142 ; v. lanualis, Lavernalis,Mucionis, Xaevia, Pandana,Rauduscula, Romanula, Saturnia,TusculanusPortiinalia, vi. 19Portunium, v. 146Portumis (et eius aedes), vi. 19portus, V. Tiberinuspo8, V. 4 ;potes, v. 58 ; v. ponsposca, V. 122positi\iis, F. 31 bposteriora (vocabula et verba), viii.12postici muri, v. 42postilioneni postulare, v. 148postmoerium, v. 143Postumus, ix. bo' ; Postuma, ix. 61potatio, V. 122 ; vi. 84potens, V. 4potio, V. 122; vi. 84662poto, vi. 84 ;poto potus sum, F. 5. 7Potoni filia, vii. 28praebia, vii. 107praeco, v. 15, 160; vi. 86, 87, 89,91, 95; vii. 42praeda, v. 178 ; viii. 19praedium, viii. 48 ;praedium -ii -io,viii. 63 ;praedia, v. 40, vi. 74,viii. 48praefica, vii. 70praelueidum, vii. lOSpraemium, v. 178Praeneste, v. 32 ; vi. 4Praenestinus (ager), v. 32praenomina, ix. 60praes, vi. 74 ;praedes, v. 40praesens, viii. 20, 58 ; ix. 102, 104praesidium, v. 90praestigiator, v. 94praeteritum, viii. 20, 58 ; ix. 104praetor, v. 80, S7 ; vi. 5, 30, 89, 91c/. 93 ; viii. 72 (-toris -torem) ; x70 (-torem) ;praetores, vi. 29, 53,87, 91 ; praetor -tori, x. 2S ;praetorin Comitio supremam pronuntiat,vi. 5 ;praetor urbauusvi. 54 ; cf. pretorpraetorium ius, vi. 71praeverbia, vi. 38, 82prandeo pi-ansus sum, F. 5. 7prata, V. 40pretium, v. 177pretor (rusticus), vii. 96Priamidae, viii. 19Priamus Priamo, viii. 3, 34prima pars casualis, v. casualeprimigenia verba, vL 36, 37primo luci, vi. 92primo mane, v. maneprincipale genus, F. 9principes, v. 89 ; v. deusprincipium, x. 56, 60, 67 ;principia,vi. 38 ; X. 56 ;principium analogiae,X. 61 ;principia verborum,vi. 37, 39, viii. 5, ix. 99 ;principia(declinatiomim), x. 11principia mundi, x. 55 ; v. caelumcf. initiapriora (vocabula et verba), viii. 12priscum vocabulum, vii. 26 ; priscaconsuetudo, x. 70 ; Prisci Latini,vii. 28 ;priscae declinationes, x.71 ; prisca nomina, ix. 22 ;prisca,vii. 2

INDEXpro (=ant«X vL 58probiis probi, F. 5. 6 ;probusprobe, F. 5. 8procare, viL 80procedere, vii. SI ; processit, vi. 38proceres, F. 30 a, F. 30 bprodire, viL 81prodixit diem, vi. 61IM-oductio syllabarum {eel llttenrumXV. 6 ; ix. 104profenatnm, vL 54prolaiumi, vi. 54profata, F. 28. 2professi, vi. 55Progne, v. 76proiecta, r. porrectaproUbitur, vi. 47prolocutiis, vi. 56jHWloquium proloquia, F. 28. 2, 6,7, 8prolubium, F. 4proludit, vi. 58Prometheus, v. 31promisee, F. 25pronomen, viii. 45 ; ix. 94pronuntiare, vi. 42, 58Propontis, vii. 81proportione vel pro portione, v.170, 181 ; viiL 50, 68, 78, 80, 83ix. 27, 29, 30. 33, 48, 61, 62, 83,103, 110 ; X. 2, 9, 36, 37, 41, 42,47, 51, 65, 66, 68 ;proportionem,viii. 57 ; c/. ratiopropc«itio, vi. 63, 76 ; v. {mtariproprio nomine, vi. 55, 78propter deitram sinistra, proptersinistram dextra, x. 59prosapia, -^-ii. 71proscaenium, \i. 58prosectum, v. 110proserpere, v. 68Proserpina, v. 68prosicium, v. 110prosns et rusns, x. 52protinam, viL 107Protc^nes, ix. 12proversus, viL 81providere, vi. 96provincia nostra, v. 16provocabula, viii. 45proximus a Flora clivus, v. 158prudens, viii. 15, 17psalterimn, viiL 61publici servi, v. libertiiiiPublicius Clivos, v."-_158; Publiciiaediles, v. 158pubUcus, V. honorPublius, V. Scipiopner, vii. 28 ; viiL 41 ; x. 4 ; puerpuella, viii. 25, ix. 29 ;paeri, vL56, ix. 10, 11, 15, 16pnera, F. 37pugil, V. 94 ;pugiles, viiL 15pngnetur pugnator, x. 32 ; r. volsillLspolli, ix. 93 ; r. pipatusPullius Clivtis, V. 158; Pulliusviocnrus, v. 158pulmentarium, v. 108pulmentum, v. 108puis, V. 105, 107, 108, 127pulvinar, v. 167pulvini, ix. 48pongo pupugi pimgam, ix. 99; x.48 ; pongo pupugi, F. 5. 6 ;pnngebampongo pungam, popogerampapogi pupugero, is. 99Punicom bellum, v. 159 ; Panicummalum, viL 91poppis puppes, viiL 66porgamentuin, v. februmporgare (Euniliam), v. 23purpura, V. 113pusus pusa, vii. 28putari, propositio potandi, vi. 63putator, vi. 63Puteoli, v. 25 ; PuteoUs, ix. 69putere, vL 96poteus, V. 25 ; vL 84 ; v. coronapoticuli, V. 25pntidus, V. 25putiloci, V. 25putor, V. 25patmn, vL 63Pyrrhi bellum, viL 39Pythagoras (artifex), v. 31^iJUMH» tumulus, viL 17qoadraginta, x. 43quadrans, r. 44, 171, 172, 174quadrigae, vilL 56 ; x. 24, 67 ; quadriga,X. 66 ; V. agitantur, initiaqnadrbigenti, x. 43quadrini, viiL 55qoadripertitio, v. 11 ; vii. 5 ; e/. v.6, 12, viii. 50, ix. 31, x. 49quadruplex fons, x. 22 ; natura, x.41 ; analogia, x. 47, 48663

INDEXquadrupes, v. 34, 79 ;quadripedem-des, vii. 39quaerere, vi. 79quaesitores, v. 81quaestio, vi. 79quaestor, vi. 79, 90 ; viii. 72 (-tori.s-torem), x. 70 (-torem) ;quaestores,V. 81, vi. 90 ; v. Septumius,Sergiusquando rex comitiavit fas, dies, vi.31quando'stercum delatum fas, dies,vi. 32Quarta, ix. 60quarta chorda citharae, x. 40quarticeps, v. 50, 62quartus dies morbi, x. 46quattuor, ix. 64, 82 ; x. 43, 45, 49, 66querquedula, v. 79Querquetulanum, v. Laresqui, V. quisquinarii, v. 173Quinctius, vi. 2 ; v. Quintiusquindecimviri, vii. 88Quinquatrus, vi. 14 ; minusculae,vi. 17quintanae (Nonae), vi. 27quinticeps, v. 50, 52, 54Quintilis, vi. 34Quintius Trogus, T., vi. 90, 92Quintus, ix. 60 ;Quintus -to, x. 51V. Muciusquintus -ti -to -turn -te, viii. 63Quirinalia, vi. 13Quirinalis collis, v. 51, 52 ; flamen,vii. 45Quirinus, v. 73, 74 ; vi. 13 ;Quiriniaedes, v. 52 ; Quirini fanum, v. 51quiritare, vi. 68Quirites, v. 51, 73; vi. 68, 86;omnes Quirites, vi. 88quirquir, vii. 8quis quae, viii. 45 ; x. 18, 30 ;quisquoius quae quaius, quis quoiqua quae, quern quis quos ques,viii. 50 ;qui quis quibus, viii. 72 ;qui homines, oportuit ques, viii.50 ; deae bonae quae, dea bonaqua, viii. 50quod bonum fortunatum felix salutarequesiet, vi. 86extrito, vii.27 ; R et D, vi. 4 ; c/. 8R exclusum, v. 133 ; R664radix, v. 103; radices (nominum etverborum), v. 74, 93, 123, vi. 37,vii. 4, 28, 35, viii. 53 ; c/. v. 13Ramnenses, v. 55 ; Ramnes, v. 55,81, 89, 91rana, v. 78 ; rana ranuncuhis, F. 10rapa, v. 108rape rapito, x. 31rams raro rarenter, sed non rare,F. 5. 8rastelli, rastri, v. 136ratariae naviculae, vii. 23ratio, vi. 39, 63 ; viii. 57, 67, 72, 79,83 ; ix. 2, 6, 8, 9, 13, 15, 16, etc. ;X. 1-3, 15, 36, 37, 41, 43, 82, etc. ;ratio analogiae, x. 54 ; ratiocasuum, x. 14 ; ratio derecta,transversa, x. 43ratis, vii. 23ratiti quadrantis, v. 44raudus, aes, v. 163Rauduscula (Porta), v. 163rauduseuhim, v. 163Reatinus ab Reate, viii. 83 ; agerReatinus, v. 53 ; Reatinum, vi. 5recentes (declinationes), x. 71recessit, vi. 38reciproca, vii. 80reciprocare, vii. SOrecordari, vi. 46rectus casus, v. 4 ; vii. 33 ; viii. 1,4, 6, 7, 16, 36, 42, 46, 49, 51, 53,68, 69, 74 ; ix. 43, 50, 54, 70, 71,75, 76, 85, 90, 102, 103 ; x. 8, 22,44, 50-52, 58-60; v. casus, nominandirecum, vii. 26redux, ix. 78regia, vi. 12, 21Regina, v. lunoregio regiones (Romae), v. 45-54 ; v.caelum, CoUina, Esquilina, Palatina,Suburanaregula, F. 20 ; regula numerorum,ix. 86reliquum, v. 175reloqui, vi. 57reminisci, vi. 44Remus, v. 54 ; viii. 45 ; v. Romulusreno (Gall.), v. 167repotia, vi. 84res, V. animalis, creperae, discrimen,fatales, genus, homo, initiarespicio, vi. 82

INDEXrespondendi, x. 31, 32 ;(species), x.31, 32rcspondet, vi. 69 ; respondere, vi.72 ; respondere ad spontem, vi."2 ; v. speciesrestibilis ager, v. 39restipulari, v. 182restis restes, viii. 66rete, v. 130reticulum, v. 130reus, vi. 90 ; reus reei, viii. 70rex regi, vi. 12, 13, 28, 31 ; x. 47ad regem conveniebat populus,vi. 28 ; ferias rex edicit populo,vi. 28 ; V. Attalus, Aventinus,Demetrius, Hostilius, initia,Latinus, Pompilius, quando, recum,Romulus, Tatius, Tiberinus,TulUusRhea, v. 144Rhodius ab Rhodo, viiL 81rica, V. 130ricinium, v. 132, 133Ripaei montes, vii. 71rite, vii. 88ritu, vii. 88 ; v. Alcyonis, Etrusco,Graecus, RomanusRobigalia, xi. 16Robigo, vi. 16rogandi (species), x. 31, 32Roma, V. 33, 41, 45, 51, 56, 74, 101,143, 144, 157, 164 ; vi. 15-17, 32vii. 10 ; viii. 18, 56, 83 ; ix. 34 ; x.15, 16, 20 ; Roma non Romula,viii. 80, ix. 50 ; Romae -am -a, x. 15Romanula Porta, v. 164 ; ^i. 24Romanus, viii. 18, 83 ; x. 16Romanus ager, v. 33, 55, 123Romanus populus, vL 86 ; Romanoritu, V. 130, vii. 88 ; Romani, vi.25, vii. 3, viii. 56 (non Romenses),83 ; Romanorum liberti,viii. 83 ; Romana stirps, v. 144Romilia tribus, v. 56Romulus, V. 9, 33, 46, 54, 55, 144149; viii. 18, 45, 80; ix. 34, 50X. 15 ; Romulo -i -um, ix. 34Romulus et Remus, v. 54 ; aedesRomuli, V. 54Romus, V. Romulus, v. 33rorarii, vii. 58rosa, v. 103Rostra, v. 155 ; vi. 91rosus, V. rususnidentum sibilus, v. 7rudet, vii. 103nifae (mulieres), vii. 83ruminalis ficus, v. 54runcina, vi. 96runcinare, vi. 96ruo ruis, ix. 109mm, v. 40 ; nire, (Joe.) F. 19, (cM.)F. 26rustici, V. 177 ; \i. 68 ; vii. 73, 84,1*6 ; rustica instnimenta, v. 134 ;V. pnjtor, Virailiarusus, V. prosusnita, V. 103ruta caesa, ix. 104rutilare, vii. 83nitili rutilae, vii. 83rutrum, v. 134rutimda stagna, v. 26S : R, \ii. 26 ; 8 demptum, ix. 44 ;8 detritum, v. 136 ; 8 extritum,vii. 97 ; r. C, G, LSabine, v. 159Sabinus (ager), v. 123 ; Sabinalingua, V. 66, c/. 74 ; origo 8abina,vii. 28 ; Sabinum belliun, v. 149 ;Sabinum vocabulum, v. 107Sabini cives, v. 159 ; Sabini, v.32, 41, 68, 73, 74, 97, 107. vi. 5,13, 28, vii. 29, 46, 77; Sabinaevirgines, vi. 20 ; fana ^bina, vi.57 ; V. Ciirtius, diessacellum, v. 152 ; r. Argei, lupiter.Lares, Murtea, Strenia, Velabrum,Volupia ; aliquot sacra etsacella, vii. 84sacer, vii. 10 ; v. dies, sacra, vassacerdos, sacerdotes, v. 83 ; vi. 16,20, 21, 23 (nostri), 24 ; vii. 44 ; v.Libersacerdotiilae, v. 130sacra nostra, vi. 13 ; v. Argei,Bacchus, camem, Xonalia, sacellum,tuhae, tubicinesSacra Via, v. 47, 152 ; r. caputsacrae aedes, vii. 10sacramentum, v. 180sacraria, r. Argei, Ops Consivasacrificia, v. 98, 124 ; v. Argeisacrifieo sacrificor, sacrificabo, sacrificaturusaut sacrificatus sum,ix. 105 ; in sacrificando deis v.122665

INDEXsaepius, v. semelsagum (Gall.), v. 167sal, V. niolaSalaeia Neptuni, v. 72Salii, V. 85 ; vi. 49sallnae, viii. 48sallere, v. 110salsiim salsius salsissimum, viii. 75.saltus, V. 3(5Salus, V. 74 ; aedes Salutis, v. 52Balutaris coUis, v. 62 ; salutare, v.quofl bonumsalutator, viii. 57saluto salutabiini salutabo, viii. 20Samnites, v. 142 ; vii. 29Samnium, v. 29Samothraces, v. 58 ; dii, v. 58Samothracum initia, v. 58Samotliracia, v. 58 ; Samothrece,vii. 34.sanctum sancta, vii. 10, 11Sancu.s, v. 6(5saperda, vii. 47sapien.s .sapientior sapientis.simus•ma, viii. 78sapio sapivi et sapii, F. 35sarcuhim, v. 134sardare, vii. 108sartum, vi. 64satio, vi. 26 ; sationes, ix. 27Saturnalia, v. 64 ; vi. 22Saturnia (antiqtuim oppidum), v.42 ; Saturnia Porta, y. 42 ; Saturniaterra, v. 42, 45Saturnii rauri, v. 42 ; Saturnii versus,vii. 36Saturnius mons, v. 42Saturnus, v. 57, 64, 74 ; vi. 22Saturni aedes, v. 42, 183 ; Saturnifanum, v. 42 ; v. Luasatus, V. 37saxum, V. Tarpeiussciabellum, v. 168.scaena scena, vii. 96 ; x. 27 ;partesscaenae, ix. 34 ; v. coroUaescaenici, vi. 76 ; scaenioi poetae, ix.17scaeptrum, v. sceptrumscaeva, vii. 97 ; scaeva avi, vii. 97 ;bonae scaevae causa, vii. 97Scaevola, vif. 97scalae, ix. 63, 68, 69 (non scala), x.54 ; scalae -is -as, x. 54 ; scala-am, X. 73666s(3alpere, vi. 96.scamnum, v. 168scauripeda, vii. 65Sceleratus Vicus, v. 159scena, v. s(3aena.scenici, v. s(jaenici.sceptrum, vii. 96.schoenicolae, vii. 64scientia, v. 8 ; scientiam orationis,ix. 112Scipio, P., vii. 31 ; ix. 71Scipionarii gladiatores a Scipione(potius quam Scipionini), ix. 72.scirpeis, vii. 44scobina, vii. 68scopae, viii. 7, 8 ; x. 24scortari, vii. 84scorteum scortea, Ail. 84scortum, vii. 84.scratiae, vii. 65scribae, vi. 87scribo, vi. 37 ; viii. 12, 25, 44(.scriboscribens) ; ix. 102 ; x. 33 ; scribonescribisne, x. 31scriptito, X. 33scriptor, viii. 57 ; scriptores, ix. Illscrupea, vii. 6, 65scrupipedae, vii. 65scutum, v. 115 ; viii. 45se ( = dimidium), v. 171secessio, v. Cnistumerinaseclum, vi. 11 ; seculum, v. 5seculae, v. 137Secunda, ix. 60secunda persona, ix. 108sedeo, vi. 37 ; sedetur, vi. 1sedes, v. 128sedile, v. 128 ;(non sediculum),viii. 54seditantes (non dicitur), viii. 60sedulitas Muci et Brnti, v. 5seges, V. 37 ; vi. 16 ; ix. 28segestria, v. 106selibra, v. 171sellae, v. 128semel et saepius, vi. 75 ; x. 33semen, v. 37sementis, vi. 26 ; sementes, v. 37sementivae feriae, vi. 26seminaria, v. 37semis, v. 171, 173, 174 ; x. 38semis tertius, etc., v. 173.semita, v. 35.semo

INDEXsemuncia, v. 171senaculum, v. 15«)senattis, F. 5. 9 ; senatus senatuissenatui, F. 17senecta, v. 5senescendi homines, vi. 11senex, \-iii. 25, 41 ; x. 4 ; F. 31 a,F. 31 b ; senes niniium noTtimverbxim vitabant, vi. 59senior, x. 4, F. 31 a, F. 31 bsentior nemo dicit, F. 5. 9septem chordae citliarae, x. 46septem monies, vi. 24septem stellae triones, vii. 74 ; c/.circulusSeptem bres Kalendae, vi. 20septemtrio, vii. 7 ; v. circulusseptimanae (Xonae), vl. 27Septimatnis, \i. 14Septimontiuni, v. 41 ; vL 14Septumius quaestor, v. 1 ; viL 109septumus dies morbi, x. 46septunx, V. 171sepulcrum, v. Acca, gleba, Tiberinus; ad sepulcrum ferunt frondemet flores, vii. 24sera, vii. 108Serapis, v. 57serare, vii. 108Sergius, M'., M'. f. quaestor, vi. 90series, ix. 97, 100 ; series casuum,ix. 77 ; -series perfecti, ix. 100series vocabuli, x. 82sermo, ^i. 63 ; viii. 25, 37 ; ix. 1, 19,107 ; .sermones Jjatini, viii. 3, 30V. natura, pelagussero sens, x. 31 ; seritume sereturne,x. 32serpens, v. 68serperastra, ix. 11serpere, v. 68serpyllum, v. 103serta, vi. 64ser^-ile nomen, viii. 10 ; v. deiisSer\nus, v. Tulliusservus, viii. 10 ; servu-s serve, x. 51servonmi nomina, viii. 21, 83,ix. 22, c/. x. 84 ; liberorum servinomina, ix. 22, 55, 59 ; v. fitna,novicius, publici, societmssesquisenex, vii. 28sessio, viii. 54sestertius, v. 173.sex, X. 49Sexatrus, vi. 14sextans, v. 171, 172 •sexticeps, v. 50, 52, 54sextula, V. 171Sextos, ix. 60 ; v. Aeliussextus casus, qui est propriusLatinus, x. 62 ; c/. viii. 16sexus, viii. 46sibilus, V. mdentumSicilia, vii. 86Siculi, V. 101, 120, 173, 175, 179sidera, vii. 14signiticatio, ix. 40 ; cf. vii. 1signum candens, vii. 14 ; signa, vii.14, 50, 73 (in caelo), 74, ix. 24,78, X. 46 (morbi), 64 ; v. lanussilentium noctis, vi. 7siliquastrum, v. 128silurus, vii. 47simbella, v. 174 ; x. 38simile similia, viii. 34, etc. ; ix. 92,etc. ; X. 1, etc. ; simile dissimileneutrum, x. 5simillimi, v. geminisimilitudo, viii. 25, 28, 29, 31, 37,39, etc. ; ix. 1, 26, 46, 53, etc. ;X. 1, 72, etc. ; similitudo perfecta,X. 12 ; .similitudo decltnationis,X. 76, 77 ; declinationum, viii. 24 ;verbi, x. 76 ; vocis rel sonitus,vi. 45, 52, 67, 75, 84, etc. ; similitudoconfiisa in verbis temporalibus,ix.108 ; similitudo verbomm,ix. 1 ; similitudiniuu forma, viii.24 ;genera, x. 9, 13, 69 ; origo, x.11, 13 ; ratio, ix. 8, x. 11 : species,X. 13 ; V. adventicium, animantiumvoces, nothum, vemaeulumsimilia, ix. 92, etc.similixulae, v. 107simplicia (verte), vi. 37 ; viii. 61ix. 97 ; simplices analogiae, x. 68res, X. 24simpuium, v. 124sine sponte sua, vi. 72singularis -re -res -ria, vii. 33 ; viii.60, 66; ix. 50, 60, 63-65, 6S, 69,80-82, 86, 87; x. 28, 33, 36, 54,56, 57-60, 62, 65, 83; singularisnatura, i. 83 ; res, x. 66 ; vocabuliseries, x. 82 ; singularesfigurae, x. 58 ; casus, x. 59, 60sing\ilaria solum, viii. 48, ix. 63singulare verbum, ix. 53 ; singu-667

INDEXlare vocabiilum, ix. 57, 69 ; v.gradussingiili (homines), ix. 5, 6, 18, 114,115; X. 74; singula, ix. 32sinistra, v. anspicinni, proptersinum, v. 123 ; ix. 21sirpando (=alliganclo), v. 137 ; sirpatur,V. 139sirpata dolia, v. 137sirpea, v. 139sirpices, v. 136sirpiculae, v. 137siser, viii. 48sisto, F. 36sisymbrium, v. 103siti, F. 18socer soceri, ix. 91 ; socer socenim,X. 28 ; socer socnis soceros socrus,X. 82societas verbonim, v. 13 ; vi. 40societatum servi, viii. 83sodalis et sodalitas, x. 39 ; Sodales,V. TitiiSol, V. 68( = Apollo), 74 ; ix. 24, 25 ;de sole, v. 59 ; solis motus, vi. 4,8 ; V. novus, occasussola terrae, v. 22solarium, vi. 4solea, ix. 113soleo solitus sum solui, ix. 107solium, v. 128solstitium, vi. 8 ; ix. 24, 25 ; c/. circulussolu solum, vi. 2soluta, V. oratiosolvunt ( = luunt), v. 137; v. luo,trutinasonant (arma), vi. 67sonitus, vi. 84 ; v. similitudosonus vocis, vi. 84sorbeo, vi. 84sors, v. 183 ; vi. 65 ; sortes, vi. 65,vii. 48sortilegi, vi. 65species, vi. 36 ; viii. 57 ; x. 13, 18,79 ; species aninialium, x. 4nominatus, x. 21 ; usuis, x. 73declinatuum (imperandi, optandi,personarum, respondendi, rogandi,temporalis), x. 31-33, e/.ix. 32 ; V. dieclinatusspecillum, vi. 82specio, vi. 82 ; x, 18, 21, 79 ; specere,V. 129, vi. 82, V. avis668spectare, vi. 82spectio, vi. 82specula, vi. 82speculator, vi. 82speculor, vi. 82sjjeculum, v. 129 ; vi. 82sperat, vi. 73 ; sperata, vi. 73spes, vi. 73 ; c/. v. 37spica, vi. 45 ; spicae, v. 37spiceret, vii. 12spondere, vi. 69-72sponsa, vi. 69, 70sponsalis, vi. 70sponsio, vi. 70sponsor, vi. 69, 74sponsu, vi. 69, 70, 73 ; v. agosponsu alligatus, vi. 71sponsus, vi. 70 ; vii. 107sponte, vi. 69, 71-73 ; v. respondere,sine spontespumae, v. 63Spurinna, x. 27stadium, v. 11stagnum, v. 26stamen, v. 113status, V. 11statuti dies, vi. 25stercum stercus, vi. 32sternere, vi. 96stillicidium, v. 27stipare, v. 182stipatores, vii. 52stipendium, v. 182 ; v. militesstips, V. 182stipulari, v. 182stirps, V. Romanus ; stirpes nonstirpeis, F. 20stiva, V. 135sto, vi. 37stola muliebris, viii. 28 ; x. 27 ; cf.ix. 48stragulum, v. 167strangiilare, vi. 96Streniae sacellum, v. 47strenuitas, viii. 15strenuus, viii. 17 ; strenui, viii. 15strettillare, vii. 65stribula, vii. 67strigile, F. 18stringere, vi. 96strittabillae, vii. 65strittare, vii. 65strues (non strus), struis -em -i,viii. 74 ; ix. 79

INDEXstultus stultior stultissimiis, ix. 72sub divo, V. divumsublecti, vi. 66subliciiis, V. ponssub Xovis, V. no\'ussubsellium, v. 12Ssubsidium, v. 89subsipere, v. 128sub tect», V. 66subtemen, v. 113subucula, V. 131 ; ix, 46subulo, vii. 35Subura, v. 48Suburana regio, v. 45, 46 ; tribus,v. 56succanit, vL 75successit, vi. 38succidia, v. 110Succusa, V. 48Succusanus pagus, v. 48sucus, V. 102, 109 ; ix. 93sudis, V. 77sudor, V. 24sueris, v. 110sufflbulum, vi. 21suile, viii. 54suilla, V. 109sulcus, V. 39sum fui ero, ix. 100 ; esum es esteram eras erat ero eris erit, ix.100 ; fiieram fui fuero, ix. 100siet, vi. 86, vii. 66, ix. 77, c/.adsiet ; v. cum muliereSummanus, v. 74summum (contentionis), viii. 78suo suis, X. 7 ; suit suit, x. 25supellex, viii. 30, 32 ; ix. 20, 21, 46,47supera loca, v. caelumsupercilia, v. 69suppanis, v. 131suprema, vi. 5 ; supremum, vi. 5,vii. 51surdus, -a, ix. 58; surdum theatrum,ix. 58surenae, v. 77surgere, r. manesurus -o -e, \-iiL 68 ; surus -i, x. 73sus, v. 96 ; suis, x. 7 ; sue, viii. 54 ;sues sed non suium, viii. 70suspicio, vi. 82susum versus, ix. 65 ; susus versus,V. 158sutor, v. 93 ; sutor sutori, x. 69sutorium, v. atriumsutrina (ars), v. 93 ;(taberna), viii.55syllaba, viii. 72 ; ix. 51, 52, 71 ; x.7, 19, 57, 81 (extrita); syllaba extrema,ix. 44, 109, x. 32, 57; v.adiectio, commutatio, correptio,detrectio, productiosyncerastum, vii. 61Syracusis, v. 151SjTi, ix. 34Syria, v. 16Syriacum nomen, v. 100tabema, viii. 55tabemola, v. 47, 50tabulae, v. censoriumTarenti, v. 31Tarpeius mons, v. 41 ; Tarpeia(\Trgo), v. 41 ; Tarpeium saxum,V. 41Tarquinius Sxiperbus, v. 159Tartarinus, viL 37Tartarus, vii. 37Tatius rex, v. 46, 51, 55, 74, 149,152 ; \-i. 68 ; arae Tati regis votededicatae, v. 74Taurii ludi, v. 154taunis, v. 96tectum, V. deieraretegete, F. 18t^ius, V. 110Tellus, V. 62, 67temo, ^'ii. 73-75tempestas, vii. 51 (suprema), 72tempestiva, vL 3tempestutem, vii. 51templum, vii. 6-10, 12, 13; templipartes, vii. 7 ; v. Acherusia, Ceres,Diana, effari, XoctUuca, Volcaniatempliun tescumque, templa tescaque,vii. 8temporalis (species), x. 31 ; «. verbumtempus, V. 11, 12; vi. 3; vii. 80;viiL 12 ; ix. 73, 108 ; tempora, v.184, vL 1, 3, 36, 40, 51, 52, 65, 97,^ii. 5, 72, 110, viii. 11, 20, 44,58, ix. 31, 32, 34, 73, 89, 95, 96,98, 101, 104, X. 7, 17, 31, 32, 34,41, 47, 48 ; temporum vocabula,V. 10, vi. 1, 35 ; t;. futurum,praesens, praeteritumtener tenerior, tenerrimus, viii. 77669

INDEXtera, v. 21, 22 ; v. terraTerentius Terentii (pJ.). ix. 38, 60;Terentius Terentium, ix. 38Terentius Terentia, viii. 7, 14, ix.55, 59 ; Terentium genus, ix. 59 ;Terentii casus, ix. 54 ; TerentieiTerentiae Terentieis, viii. 36tergiis, V. tegusterinien, v. 21teritorium, v. 21termen, v. terimenTerminalia, vi. 13Terminus, v. 74 ; termini, v. 21terra, v. 16, 17, 21-25, 31, 34, 36, 39,61) ; vil. 17 ; ix. 38 ; terra, x. 62 ;Terra, v. 57-60, 64 ( = Ops), 65( = Iuno), 67, 69 ; terra mater, v.64 ; terrae loca infera, v. 16terra mundi media, vii. 17 ; v.Calydonius, pila, Saturnia,Thraeca ; c/. teraterrestris, v. animaliaterreus murus, v. 48temmcius, v. 174Tertia, ix. 60terticeps, v. 50, 62tesca, vii. 8, 10-12testuaciuni, v. 106testudo, V. 79, 117, 161tetrachorda, x. 46Teucer, vii. 3texta fasciola, v. 130theatra, vi. 83 ; v. surdusThebris( = Tiberis), v. 30Thelis(=Thetis), vii. 87Tlieona, viii. 41 ; ix. 42thesaurus, vii. 17Thespiades, vii. 20Thraces, vii. 43Tliraeca terra, v. 14thrion, v. 107thynnus, v. 77 ; vii. 47Tiberinus (deus), v. 29, 71 ; TiberinusrexLatinorum, v. 30 ; Tiberinisepulcrum, v. 30 ; Tiberinusportus, vi. 19Tiberis, v. 28-30, 43, 54, 71, 83, 146 ;vi. 17 ; vii. 44 ; v. Tliebris, cisTiberim, uls Tiberimtibiae, viii. 61tibicines, vi.-17, 75 ; vii. 35 (Tusci) ;viii. 61Tibur, viii. 53 ; ix. 34Tiburs, viii. 53 ; Tiburtes, ix. 34670tigris, V. 100Tigris (flumen), v. 100^mor, vi. 45tinguere, vi. 96Titanis, vii. 16Titan, v. CoeusTitienses, v. 55, 89, 91Titii sodales, v. 85Titium tribus, v. 81toga, V. 114; viii. 28; ix. 48; togapraetexta, vi. 18toral, V. 167torpedo (piscis), v. 77torus tonilus, v. 167trabes trabs, vii. 33 ; x. 56, 57tragoediae, vi. 55 ; x. 70 ; i'. Tuscustragoedus, ix. 55 ; v. agotragula, v. 115, 139traiectio litterarum, v. 6tralaticio nomine, vi. 55 ; c/. translaticiumtralatum, vi. 77 ; vii. 23, 103 ; x. 71trama, v. 113trames, vii. 61transitus de casu in casum, viii. 39 ;X. 28,' 51-53, 77, cf. x. 29; transitusdeclinationis, x. 77translaticium nomen, v. 32 ; translaticiumverbum, vi. 64 ; translaticiaverba, vi. 78 ; c/. tralaticiotransversus, vii. 81 ; v. ordotrapetes, v. 138tremo, vi. 45 ; tremuisti timuisti,vi. 45tres tria, ix. 64 ; x. 49, 67, 83tressis, v. 169; ix. 81, 83, 84; hitresses, hoc tressis, ix. 81triarii, v. 89tribulum, v. 21tribuni aerarii, v. 181 ; militum, v.81, 91 ;plebei, v. 81, vi. 87plebis, vi. 91tribus, v. 35, 55, 56, 81, 91 ; i'. Collina,curatores, Esquilina, Palatina,Romilia, Suburana, Titiumtributum, v. 181tricessis, v. 170tricliniaris gradus, viii. 32 ; tricliniaresmappae, ix. 47triclinium, ix. 9, 47 ; v. aestixiim,hibernumtriens, v. 171trigae, viii. 55trigona, vii. 75

INDEXtrini, viiL 55 ; trinae trina, is. 64trinae, x. 67 ; ». copulae; iones, vii. 74, 75iripertita, v. 10 ; tripartite, v. 35V. oratiotriplicia, viii. 46triporteuta, viL 59tritavus, vii. 3triticum, v. 106 ; ix. 27trittiles, vii. 104tritura, v. 21triumpliare, \i. 68triumviri capitales, v. 81 ; tudiciumtriumvirum (non -viromniX ix. 85Trivia, viL 16trivium, vii. 16trivolum, p. tribulumTrogus, f. QuintiusTroia, vii. 38Troianus equus, vii. 38trua, V. 118truleum, v. 118trulla, V. 118trutina (per tnitinam sohi), v. 183tryblia, v. 120tubae tubi, r. 117 ; tubae sacroruin,vi. 14tubicines, v. 91 ; vi. 75 ; tiibicincs.sacrorum, v. 117Tubulustrium, vL 14tueri, tueri villain, vii. 12Tullia Tarqnini Superbi uxor, v.159Tullianum, v. 151TuUius rex, v. 49, 151 ; SerrinsTullius rex, vi. 17Tullius et Antouius consules, viii.10tumulus, V. Pythonostundo tundam tutudi, ix. 99 ; tundebamtundo tundam, tutuderamtutudi tutudero, tutudi tundotundam, x. 48 ; timdo tutudi,F. 5. 6tunica, v. 114; viii. 28; ix. 79;tunica ferrea ex anulis, v. 116tunica virilis et muliebris, ix. 48,X. 27turdarivim, \i. 2turdelix, vi. 2turdus, v. 77 ; vi. 2 ; ix. 28, 55 (nonturda)turma, v. 91turres, v. 142 ; turre, F. 18Tuscanicum, v. 161Tusce, F. 5. 8TuscuJanus ager, vii. 18 ; Tuscnlani,^-L 14 ; Tusculanae portae,vi. 16Tuscus dux, V. 46 ; Tusci, v. 32,161, vi. 28, 35, ix. 69; TuscusTusce, F. 5. 8 ; tragoediae Tiiscae,V. 55 ; vocabula Tusca, v. 55Vicus Tuscus, V. 46tussi, F. 18Tutiliiiae loca, , 163tutulati, ^ii. 44tutulus, vii. 44U producere, ix. 104 ; Ulongum,breve, ix. 104 ; U exitus, x. 62US : EI, ix. 80 ; r. Eudor, V. 24udus uvidus, v. 24Ufenas (non Ufenius), viii. 84uliginosus (ager), v. 24ullaber, r. olla verauls Tiberim, v. 83ulula, V. 75ululantis (luporum vox), viL 104umbilicus, viL 17umbones, v. 115umbra (piscis), v. 77Umeri, vii. 50imcia, V. 171, 172, 174ungo ungor, x. 33unguentaria tabema, viii. 55unguentum -ta, ix. 66, 67unguis ungula, F. 10 ; ungues, v. 77universa, x. 84unoculi, vii. 71unus -ius -i -um -e -o, viii. 63 ; unus-a -um, ix. 64, X. 24 ; uni -ae -a,ix. 64 ; una -ae, x. 24, 67 ; imiun,ix. 87, X. 30, 41, 43, 45, 49; uni(pJ.), \iiL 55 ; unae et binae, viiL 7upupa, v. 75uraeon, v. 76urbanus, viii. 18; urbanus exercitus,vL 93 ; urbani, vi. 68 ; r.auspicium, praetorUrbinas, viii. 84Urbinius, \-iii. 84urbs, V. 28, 41, 43, 97, 151, 158 ; vL17, 18, 24, 28, 68, 93 ; vii. 44 ; ix.68 ; urbes non urbeis, F. 20antiqua urbs, v. 48, vL 24 ; urbisloca, v. 45 ; urbis partes, v. 56671

INDEXurbes, v. 143 ; urbes condere, v.143 ; in Urbe Lucili, v. 138urinare, v. 126urinator, v. 126urnae, v. 126urnarium (genus mensae), v. 126uro uror, x. 33ursi, V. 100 ; vii. 40urvuni, V. 127, 135usura, V. 1S3usus (communis), viii. 28, 30, 31ix. 7, 20, 37, 38, 56-60, 62, 63, 67-71, 74; X. 72, 73, 83, 84; ususloquendi, ix. 6, x. 74 ; usus vetus,X. 78 ; V. copulae, speciesuter utrei, ix. 65utilitas, viii. 26-29, 31 ; ix. 48uvae, V. 104uvidus, V. 24, 109uvor, V. 104V, V. 117vagit (haedi vox), vii. 104valentes glebarii, vii. 74vallum ( = murus), v. 117; ( = vannus),V. 138valvata, viii. 29varietas, ix. 46 ;(casuuni), x. 62vas vadis, vi. 74, F. 15vas vasis, F. 15 ; vas aquarium, v.119; vas argenteum, ix. 66; vasvinarium, v. 123 ; vasa, viii. 31,ix. 21 ; vasa aenea, v. 125 ; vasain mensa escaria, v. 120 ; vasasacra, v. 121va.saria mensa, v. 125vates, vii. 36vatia, ix. 10vaticinari, vi. 52Vatinius Vatiniorum, viii. 71vectes non vecteis, F. 20Vediovis, v. 74vehiculum, v. 140Veientes, v. 30Velabrum, v. 43, 44 ; vi. 24 ; minuset maius, v. 156 ; Velabrum sacellum,V. 43velaturam facere, v. 44Velia, Veliae, Veliense, v. 54Velinia, v. 71Velini lacuS, v. 71vellere lanam, v. 54velli( = villi), v. 130vellus, v. 130 ; vellera, v. 54672venabulum, viii. 53venator, v. 94 ; viii. 53Veneria, v. corollaVenilia, v. 72venor venans venaturus venatiis,viii. 59venter, v. Faliscus ; ventres nonventreis, F. 20ventilabrum, v. 138Venus caeligena, v. 62 ; Libentina,Libitina, vi. 47 ; victrix, v. 62 ;Veneris vis, v. 61 , 63 ; dies lovisnon Veneris, vi. 16 ; e spumisVenus, V. 63 ; Aprilis a Venere,vi. 33 ; Veneri dedicata aedes, vi.20 ; lucus Veneris Lubentinae,F. 4 ; V, Murteaever, V. 61 ; vi. 9verberatus sum verberor verberabor,X. 48verbex, v. 98verbum, x. 77, etc. ; verba, viii. 11,12, 53, 57, etc., ix. 56, 89, etc.,F. 34 ; verbum temporale, viii.13, 20, 53, ix. 95, 108, 109 ; verbaquae tempore adsigniflcant, vi.40 ; verba aliena, v. 10 ; antiqua,v. 9 ; concepta, vii. 8 ; declinata,vi. 37, ix. 115 ; verba facere, vi.78 ; verba Acta, v. 9 ; inclinanda,X. 13 ; interpolata, v. 3 ; Latina,V. 120, vi. 96, vii. 3; verbanostra, v. 10, x. 71 ; verba oblivia,V. 10 primigenia, vi. 36,37 ;;verborum novorum et veterumdiscordia, v. 6 ; verborumcognatio, v. 13, vi. 1 ; coUatio,viii. 78 ; copia, viii. ; formulae,2X. 33 ; materia, x. 11 ; societas,V. 13 V. discrimen, duplex,;figura, forma, genus, Graecus,impositio, infecta, multitude,natura, numerus, obscuritas,origines, perfectum, personae,poeta, poetica, principium, radix,similitudo, simplicia, translaticium,vernaculaVergiliae, vi. 6 ; vii. 50vernacula verba vel vocabula, v. 3,77, 104 ; vi. 40 ; similitudinisgenus vernaculum, x. 69versu, X. 62 ; versus obliqui, x. 43 ;V. ^turnii, vieriveru, v. 127 ; cf. v. 98

INDEXvesper, vi. 6 ; vii. 50 ; vesperi, ix.73 ; vespere, ix. 73 ; v. novusVesperugo, vi. 6, 7 ; vii. 50Vesta, V. 74 ; vi. 17 ; Vestae aedes,vi. 32Vestales virgines, vi. 17, 21 ; virgoVestalis Tarpeia, v. 41Vestalia, vi. 17vestibulum, vii. 81vestigator, v. 94vestimentiim -ta, ix. 20, 48 ; x. 72vestis, V. 130 ; veste, F. 18vestispica, \\\. 12vestitus, V. 105 ; viii. 28, 30 ; c/.viii. 31, ix. 45Veturi, ». MamuriVetiirii Cicurini, vii. 91vetus vetiistius veterrimum, vi. 59 ;vetus consuetiido, v. 2, ix. 13,20, 21, X. 73 ; Forum Vetus, vii.29 ; veteres leges abrc^tae, ix.20 ; veteres, v. 14, 52 (poetaeX 98(nostri), vii. 32 ; vet«ra, x. 73 ;Vetera vocabula, ix. 20 ; v. Aesculapii,Capitolium, Curia, verbum•tustas, V. 3, 5 ; vi. 21 exillum, vi. 93via, V. 8, 22, 35 ; vii. 15 ; v. no\T»,sacraviales, v. LaresVibenna, v. Caelesvibices, vii. 63vicessis, v. 170Victoria, v. 62 ; caeligena, v. 62victoriatus, ix. 85 ; x. 41victrix Venus, v. 62victus, V. 105vicus, V. 8, 160 ; vici, v. 145 ; r.Africus, Cyprius, Insteianus,Sceleratus, Tuscusvideo, vi. 80 ; tu domi vldebis, viu12 ; vide, vii. 12vieri (=vinciri), v. 62; versibusviendis, vii. 36vigilant, vi. 80vigiliuni, \\. SOviginti, X. 41, 43, 45villae, v. 35villi, V. velliV^iminalis Collis, v. 51Viminius lupiter, v. 51Vinalia, v. 13 ; \\. 16 ; nistica,vi. 20vinaria mensa, v. 121 ; tabema,VOL. IIviiL 55 ; rasa, v. 123 ; vinariumtmleum, v. 118vinciri, v. 62vinclum, v. 62vinctio, V. 61vincttira, v. 62vindemia, v. 37 ; vi. 16vindemiator, v. 94vineae, v. 37, 117vineta, v. 37vinum, v. 13, 37 ; vi. 16 ; vinumvina, ix. 66, 67 ; v. Chio, flamen,Lesboviocurus, v. 7, 158\iola\it virginem, vi. 80violentia, v. 70vir, viii. 80 ; ix. 85 ; x. 4 ; vireis,viii. 36 ; r. centiunvirum, decem-Tirum, quindecimviri, triumvirivirago, vii. 37virgo virgines, v. Sabimis, Tarpeius,Vestales, violavitvirgultum, v. 102viride, V. 102virile virilia, viii. 46, 51 ; ix. 41, 48,81, 110; X. 8, 21, 30; nominavirilia, viii. 36, x. 65 ; nomenTirile, viii. 81, ix. 40, i. 65V. tunica ; c/. genusvirtus, T. 73vis, v. 37, 61, 63, 70, 102; vi. 80;viii. 7 ; haec vis, huius vis, baevis, F. 16 ; V. Venus, vitavisenda, vi. 82visere, v. inliciumVisolus, V. Poeteliusvisus, vi. 80vita a vi, v. 63 ; vita et mors, v11 ; r. decemuntvitio manumissus, creatus magistratus,vi. 30vitis, V. 37, 102Vitula, vii. 107vitulantes, vii. 107vitulus, V. 96 ; vituli, ix. 28 ; vitulivox, vii. 104vivices, v. vibicesvivo non vivor, x. 78 ; vivatur viveretur,x. 32vix, viii. 9 ; x. 14, 79, 80vocabulum vocabula, vi. 56 ; viii.11, 12, 40, 45, 52, 53, 56-59, 61, 64,71, 75, 78, 79, 80 ; ix. 1, 9, 21, 34,41, 5

INDEX66-68, 71, 74, 77, 78, 85, 88, 90 ; x6, 20, 23, 24, 3u, 47, 51, 54, 81-83vocabulum Latiniim, v. 29, 68priscum, vii. 26 ; vocabula a Fortuna,V. 91 ; Aegyptiorum, viii65 ; aquatilium animalium, v. 77artificum, v. 93 ; barbara, viii64 ; dierum, vi. 12 (civilia), 33vocabula ex Graeco sumptaF. 14 a, F. 14 b ; vocabula feramm, v. 100; Gallica, v. 167Gallorum, viii. 65 ; vocabiila imponenda, vi. 3 ; lectulonim, v166 ; litterarum Latinarum, ix51 ; locorum, v. 10 ; magnitudinis, viii. 79 ; mensium, vi. 33miliaria, ix. 85 ; multitudinis, ix64-66, 68, 69 ; vocabula nostra,viii. 65 pecimiae, v. 169 pis; ;cium, V. 77 ; Poenicum, viii. 65temporum, v. 10, vi. 1, 35 ; vocabula Tusca, v. 55 ; Vetera, ix. 22V. casus, figura, Graecus, homo,impositio, peregrinus, poeta;Sabinus, series, singularis, vernaculavocalis, V. oratiovocandi casus, viii. 42, 68 ; ix. 43,91 ; X. 31 ; c/. viii. 16vocare, v. eomitiatum, inliciumVolaminia, v. VolumniaVolcanalia, vi. 20Volcanalis flamen, v. 84Volcania templa, vii. 11Volcanus, v. 70, 74 ; vi. 20volgus, v. 58volo (vis et volas), vi. 47 ; ix, 103 ;X. 81volpes, v. 101volsillis pugnare, non gladio, ix. 33Volturnalia, vi. 21Volturnalis flamen, vii. 45Volturnum (oppidum), v. 29Volturnus (amnis), v. 29 ; vi. 21vii. 45volucres, v. 75 ; ix. 28 ; volucrumvox, vii. 104Volumnia, v. Luciavoluntarius -a -um, v. declinatio,declinatus, genusvoluntas hominum, ix. 34 ; x. 15,51 ; cf. voluntariusVolupiae sacellum, v. 164voluptas, viii. 31vomer, v. 135Vortumnus, v. 46, 74vox voces, \iii. 40, etc. ; ix. 38, 40,42, 52, 55, 70, 88 ; x. 7, 19, 29, 30,36, 63, 66, 68, 69, 72, 77, 82 ; v.animantium, declinatus, tigura,lupus, similitudo, sonusX, cf. 08, OSXerxes, vii. 21zanclas, v. 137674

INDEX OF GREEK WORDSReferences are to Book (Roman numeral) and Section (Arabic number),and to Fragment (F.) and serial number (Arabic), with subdivisions.aya06v, v. ^;aypov, V. 34aywf, vi. 12act Of, vi. 11aleta^ai, \i. 9«uwKo, vi. 11oAfiUoKov, vii. 82oAjcvwi', V. 79 ; vii. 88afia^av, viL 74ififiuyt^, V. 115ififXytiv, vi. 96ofi^'^to, V. 78a(>a'yapyapi^«r0

INDEXKaC, vi. 10 ; V. eiTji»Kavovv, V. 120Kanpoi, V. 101KOLTIVOV, V. 120KepKrjSrii, v. 79KXinreiv, vii. tHKoti'Ttoi', vi. 2Ko^iavSpov, V. 103KoyAi'as, F. 14 a, F. 14 bKpavov, V. 105Kv\lKeLO}f V. 121KVIXIVOV, V. 103KiufxuSt'a, vii. 89Ka>^ov, vii. 89AaTpov, vii. 52Aetpi'w, v. 103\eKTpov, V. 166AtTTopi»', v. 101AevKirfvov, vi. 2A^flrj, vii. 42Ai;i(/naokrJTrTovi, viii, 87Kiif, vi. 6ois, V. 96bp.a\6u, vii. 17bp.uii>vp.ta, ix. 80OTrio-edSo/iioi', V. 160opxiv nopicLv, v. 108napaSie^evypevov, F. 28. 14irapoip^iav, vii. 31TreVceii', vi. 96Trept tti'tojixaAtaSj ix, 1Trepl TToAoi', vii. 4Trepl (rr]p.aLvop,iviav, v. 2nepiiTTv\ov, viii. 29TTJjyai'Oi', V. 103TToAos, vii. 14 ; V. nepCIloAuSevKT)?, v. 73TTOpicr), v. 97TTopicos, v. 977rdro9 7r6ra>, v. 122 ; Troror, vi. 84npoParov, V. 96Trpofiop.oi', v, 160irpo'iSelv, vi. 96llvOayopa, I'. ai'Tixflcoi'TTv0e

THE LOEB CLASSICALLIBRARYVOLUMES ALREADY PUBLISHEDLATIN AUTHORSAMMIANUS MARCELLINUS. J. C. Rolfe. 3 Vols.Vols. I. and II.APULEIUS. THE GOLDEN ASS (METAMORPHOSES). W. Adlington (1566). Revised by S. Gaselee.{6th Imp.)AULUS GELLIUS. J. C. Rolfe. 3 Vols.AUSONIUS. H. G. Evelyn White. 2 Vols.BEDE. J. E. King. 2 Vols.BOETHIUS: TRACTS and DE CONSOLATIONEPHILOSOPHIAE. Rev. H. F. Stewart and E. K.Rand.(3rd Imp.)CAESAR : CIVIL WARS. A. G. Peskett (4(h Imp.)CAESAR : GALLIC WAR. H. J. Edwards (Ith Imp.)CATO AND VARRO : DE RE RUSTICA. H. B. Ash andW. D. Hooper. {2nd Imp.)CATULLUS. F. W. Cornish; TIBULLUS. J. B.Postdate; and PERVIGILIUM VENERIS. J. W.Mackail.CELSUS : DE{10th Imp.)MEDICINA. W. G. Spencer 3 Vols.CICERO: DE FINIBUS. H. Rackham. (3rd Imp. revised.)CICERO: DE NATURA DEORUM and ACADEMICA.H. Rackham.CICERO : DE OFFICIIS. Walter Miller. (3rd Imp.)IlEPUBLICA and DE LEGIBUS. Clin-CICERO : DEton W. Keyes.CICERO: DE SENECTUTE, DE AMICITL\, DEDIVINATIONE. W. A. Falconer. (4M Imp.)CICERO: IN CATILINAM. PRO xMURENA, PROSULLA, PRO FLACCO. Louis E. Lord.CICERO: LETTERS TO ATTICUS. E. O. Winstedt.3 Vols. (Vol. I. ith Imp., Vol. II. 3rd Imp. and Vol.III. 2nd Imp.)CICERO: LETTERS TO HIS FRIENDS. W. GlynnWilliams. 3 Vols.CICERO : PHILIPPICS. W. C. A. Ker. (2nd Imp.)CICERO: PRO ARCHIA. POST KEDITUM. DEDOMO, DE HARUSPICUxM RESPONSIS. PROPLANCIO. N. H. Watts. (2nd Imp.)CICERO: PRO CAECINA, PRO LEGE MANILIA,PRO CLUENTIO, PRO RABIRIO.1H. Grose Hodge.

THE LOEB CLASSICAL LIBRARYCICERO: PRO MILONE, IN PISONEM, PROSCAURO, PRO FONTEIO, PRO RABIRIO POS-TUMO, PRO MARCELLO, PRO LIGARIO, PROREGE DEIOTARO. N. H. Watts.CICERO : PRO QUINCTIO, PRO ROSCIO AMERINO,PRO ROSCIO COMOEDO, CONTRA RULLUM.T IT Frpf*S6*CICERO : TUSCULAN DISPUTATIONS. J. E. King.CICERO: VERRINE ORATIONS. L. H. G. Greenwood.2 Vols.CLAUDIAN. M. Platnauer. 2 Vols.FLORUS: E. S. Forster; and CORNELIUS NEPOS :J. C. Rolfe.FRONTINUS: STRATAGEMS and AQUEDUCTS.C. E. Bennett and M. B. McElwain.FRONTO : CORRESPONDENCE. C. R. Haines. 2 Vols.HORACE : ODES and EPODES. C. E. Bennett. (10^/jImp. revised.)HORACE: SATIRES, EPISTLES, ARS POETICA.H. R. Faircloiigli. (4

THE LOEB CLASSICAL LIBRARYPLINY: LETTERS. Melmoth's Translation revised byW. M. L. Hutchinson. 2 Vols. {4dh Imp.)PLINY : NATURAL HISTORY. H. Rackham andW. H. S. Jones. 10 Vols. Vol. I.PROPERTIUS. H. E. Butler. (4

THE LOEB CLASSICAL LIBRARYGREEK AUTHORSACHILLES TATIUS. S. Gaselee.AENEAS TACTICUS: ASCLEPIODOTUS and ONA-SANDER. The Illinois Greek Club.AESCHINES. C. D. Adams.AESCHYLUS. H. Weir Smyth. 2 Vols. {Vol I. 4th hnp..Vol. II. Srd Imjj.)APOLLODORUS. Sir James G. Frazer. 2 Vols.APOLLONIUS RHODIUS. R. C. Seaton. (4th Imp.)IHE APOSTOLIC FATHERS. Kirsopp Lake. 2 Vols.(Vol. I. 5th Imp., Vol. II. 4th Imp.)APPIAN'S ROMAN HISTORY. Horace White. 4 Vols.(Vol. I. 3rd Imp., Vols. II., III. and IV. 2nd Imp.)ARATUS. C/. CALLIMACHUS.ARISTOPHANES. Benjamin Bickley Rogers. 3 Vols.(Vols. 1. and II. 4th Imp., Vol. III. 2rd Imp!)Verse trans.ARISTOTLE : ART OF RHETORIC. J. H. Freese.ARISTOTLE : ATHENIAN CONSTITUTION, EUDE-MIAN ETHICS, VIRTUES and VICES. H. Rackham.{2nd Imp.)ARISTOTLE: METAPHYSICS. H. Tredennick. 2 Vols.{2nd Imp.)ARISTOTLE: MINOR WORKS. W. S. Hett. OnColours, On Things Heard, Physiognomics, OnPlants, On Marvellous Things Heard, MechanicalProblems, On Indivisible Lines, Situations and NamesOF Winds, On Melissus, Xenophanes, and Gorgias.ARISTOTLE: NICOMACHEAN ETHICS. H. Rackham.{2nd Imp. revised.)ARISTOTLE: OECONOMICA and MAGNA MOR-ALIA. G. C. Armstrong ; with Vol. II. Metaphysics.{2nd Imp.)ARISTOTLE: ORGANON. H. P. Cooke and H. Tredennick.3 Vols. Vol. I.ARISTOTLE : ON THE SOUL, PARVA NATURALIA,, ON BREATH. W. S. Hett.ARISTOTLE: PARTS OF ANIMALS. A. L. Peck;MOTION AND PROGRESSION OF ANIMALS.E. S. Forster.ARISTOTLE : PHYSICS. Rev. P. Wicksteed and F. M.Cornford. 2 Vols. (Vol. II. 2nd Imp.)ARISTOTLE : POETICS and LONGINUS. W. HamiltonFyfe; DEMETRIUS ON STYLE. W. RhysRoberts.{2nd Imp. revised.)ARISTOTLE : POLITICS. H. Rackham.4

THE LOEB CLASSICAL LIBRARYARISTOTLE: PROBLEMS. W. S. Rett. 3 Vob.ARISTOTLE: RHETORICA AD ALEXANDRUM.H. Rackham. (With Problems, Vol. II.)ARRIAN : HISTORY OF ALEXANDER and INDICA.Rev. E. IlifFe Robson. 2 Vols.ATHENAEUS: DEIPNOSOPHISTAE. C. B. Gulick.7 Vols. Vols. I.-VI.ST. BASIL : LETTERS. R. J. Deferrari. 4 Vols.CALLIMACHUS and LYCOPHRON A. W. MairARATUS. G. R. Mair.CLEMENT OF ALEXANDRIA. Rev. G. W. Butterworth.COLLLTHUS. Cf. OPPIAN.DAPHNIS AND CHLOE. Thornley's Translation revisedby J. M. Edmonds; and PARTHENIUS. S. Gaselee{2rd Imp.)DEMOSTHENES: DE CORONA and DE FALSALEGATIONE. C. a. Vince and J. H. Vince.DEMOSTHENES : MEIDIAS, ANDROTION, ARISTO-CRATES, TIMOCRATES. ARISTOGEITON. J. H.Vince.DEMOSTHENES: OLYNTHIACS, PHILIPPICS andMINOR ORATIONS : I-XVII and XX. J. H. Vince.DEMOSTHENES : PRIVATE ORATIONS. A. T. Murray.4 Vols. Vol. I.DIOCASSIUS: ROMAN HISTORY. E. Gary. 9 Vols.(Vol. II. 2nd Imp.)DIO CHRYSOSTOM. J. W. Cohoon. 6 Vols. Vol. I.DIODORUS SICULUS. C. H. Oldfather. 13 Vols.Vols. I. and II.DIOGENES LAERTIUS. R. D. Hicks. 2 Vols. (Vol.I. 3rf/ Imp.)DIONYSIUS OF HALICARNASSUS : ROMAN ANTI-QUITIES. Spelman's translation revised by E. Cary.7 Vols. Vol. I.EPICTETUS. W. A. Oldfather. 2 Vols.EURIPIDES. A. S. Way. 4 Vols. (Vols. I., II., IV.6th Imp., Vol. III. 3rd Imp.) Verse trans.EUSEBIUS: ECCLESIASTICAL HISTORY. KirsoppLake and J. E. L. Oulton. 2 Vols. (Vol. H. '2nd Imp.)GALEN: ON THE NATURAL FACULTIES. A. J.Brock.(2nd Imp.)THE GREEK ANTHOLOGY. W. R. Paton. 5 Vols.(Vol. I. 3rd Imp., Vols. II. and III. 2nd Imp.)GREEK ELEGY AND IAMBUS with the ANACRE-ONTEA. J. M. Edmonds. 2 Vols.

THE LOEB CLASSICAL LIBRARYTHE GREEK BUCOLIC POETS (THEOCRITUS,BION, MOSCHUS). J. M. Edmonds. (6th Imp. revised.)GREEK MATHEMATICAL WORKS. Ivor Thomas.2 Vols. Vol. I.HERODES. C/. THEOPHRASTUS : CHARACTERS.HERODOTUS. A. D. Godley. 4 Vols. (Vol. I. 3rdImp., Vols. II.-IV. 2nd Imp.)HESIOD AND THE HOMERIC HYMNS. H. G. EvelynWhite. {5th Imp. revised and enlarged.)HIPPOCRATES AND THE FRAGMExNTS OF HERA-CLEITUS. W. H.S.Jones and E.T.Wlthington. 4 Vols.HOMER :HOMER :ILIAD. A. T. Murray. 2 Vols. (Vol. I. 4,thIhip., Vol. II. Srd Imp.)ODYSSEY. A. T. Murray. 2 Vols. {4th Imp.)ISAEUS. E. S. Forster.ISOCRATES. George Norlin. 3 Vols. Vols. I. and II.ST. JOHN DAMASCENE: BARLAAM AND lOASAPH.Rev. G. R. Woodward and Harold Mattingly. {2nd Imp.revised.)JOSEPHUS. H. St. J. Thackeray and Ralph Marcus.9 Vols. Vols. I.-VI. (Vol. V. 2nd Imp.)JULIAN. Wilmer Cave Wright. 3 Vols. (Vols. I. and II.2nd Imp.)LUCIAN. A. M. Harmon. 8 Vols. Vols. I.-V. (Vols.I. and II. 3rd Imp.)LYCOPHRON. C/. CALLIMACHUS.LYRA GRAECA. J. M. Edmonds. 3 Vols. (Vol. I.3rd Imp., Vol. II. 2nd Ed. revised and enlarged.)W. R. M. Lamb.{3rd Imp. revised.)LYSIAS.MARCUS AURELIUS. C.R.Haines.MENANDER. F. G. Allinson. (2nd Imp. revised.)MINOR ATTIC ORATORS (ANTIPHON, ANDOCIDES,DEMADES, DEINARCHUS, HYPEREIDES). K. J.Maidment. 2 Vols. Vol. I.OPPIAN, COLLUTHUS, TRYPHIODORUS. A. W. MainPAPYRI (SELECTIONS). A. S. Hunt and C. C. Edgar.4 Vols. Vols. I. and II.PARTHENIUS. Cf. DAPHNIS and CHLOE.PAUSANIAS: DESCRIPl ION OF GREECE. W. H. S.Jones. 5 Vols, and Companion Vol. (Vol. I. 2nd Imp.)PHILO. 10 Vols. Vols. I.-.V. F. H. Colson and Rev. G.H. Whitaker; Vols. VI. and VII. F. H. Colson.PHILOSTRATUS : THE LIFE OF APOLLONIUS OFTYANA. F. C. Conybeare. 2 Vols. (Vol. I. 3rd Imp.,Vol. II. 2nd Imp.)6

THE LOEB CLASSICAL LIBRARYPHILOSTRATUS : IMAGINES ; CALLISTRATUS :DESCRIPTIONS. A. Fairbanks.PHILOSTRATUS and EUNAPIUS : LIVES OF THESOPHISTS. Wilmer Cave Wright.PINDAR. Sir J. E. Sandys. {6th Imp. revised.)PLATO : CHARMIDES, ALCIBIADES, HIPPARCHUS.THE LOVERS, THEAGES, MINOS and EPINOMIS.W. R. M. Lamb.PLATO: CRATYLUS. PARMENIDES, GREATERHIPPIAS. LESSER HIPPIAS. H. N. Fowler.PLATO: EUTHYPHRO, APOLOGY, CRITO, PHAE-DO, PHAEDRUS. H. N. Fowler. (7«A Imp.)PLATO : LACHES, PROTAGORAS MENO, EUTHY-DEMUS. W. R. M. Lamb. (2nd Imp. revised.)PLATO : LAM'S. Rev. R. G. Bury. 2 Vols.PLATO : LYSIS, SYMPOSIUM, GORGIAS. W. R. M.Lamb. (2nd Imp. revised.)PLATO: REPUBLIC. Paul Shorey. 2 Vols. (Vol.1.2nd Imp. revised.)PLATO: STATESMAN, PHILEBUS. H. N Fowler;ION. W. R. M. Lamb.PLATO : THEAETETUS and SOPHIST. H. N. Fowbr.(2nd Imp.)PLATO : TIM ALUS, CRITIAS, CLITOPHO, MENEXE-NUS, EPISTULAE. Rev. R. G. Bury.PLUTARCH: MORALIA. 1 4 Vols. Vols. I.-V. F. C.Babbitt ; Vol. X. H. N. Fowler.PLUTARCH: THE PARALLEL LIVES. B. Perrin.II Vols. (Vols. I., II., III. and VII. 2nd Imp.)POLYBIUS. W. R. Paton. 6 Vols.PROCOPIUS: HISTORY OF THE WARS. H. B.Dewing. 7 Vols. Vols. I.-VI. (Vol. I. 2nd Imp.)QUINTUS SMYRNAEUS. A. S. Way. Verse trans.SEXTUS EMPIRICUS. Rev. R. G. Bury. 3 VoU.SOPHOCLES. F. Storr. 2 Vols. (Vol. I. 6th Imp., VoLII. 4th Imp.)Verse trans.STRABO: GEOGRAPHY. Horace L. Jones. 8 Vols-(Vols. I and VIII. 2nd Imp.)THEOPHRASTUS : CHARACTERS. J. M. Edmonds ;HERODES, etc. A. D. Knox.THEOPHRASTUS: ENQUIRY INTO PLANTS. SirArthur Hort, Bart2 Vols.THUCYDIDES. C. F. Smith. 4 Vols. (Vol. I. 3rd Imp..Vols. II., HI. and IV. 2nd Imp. revised.)C/. OPPIAN.7TRYPHIODORUS.

THE LOEB CLASSICAL LIBRARYXENOPHON : CYROPAEDIA. Walter Miller. 2 Vols.(2nd Imp.)XENOPHON : HELLENICA, ANABASIS, APOLOGY,AND SYMPOSIUM. C. L. Brownson and O. J. Todd.3 Vols. (2nd Imp.)XENOPHON : MEMORABILIA and OECONOMICUS.E. C. Marchant. {2nd Imp.)XENOPHON : SCRIPTA MINORA. E. C. Marchant.VOLUMES IN PREPARATIONGREEK AUTHORSARISTOTLE: DE CAELO. W. K. C. Guthrie.ARISTOTLE: HISTORY AND GENERATION OFANIMALS. A. L. Peck.ARISTOTLE: METEOROLOGICA. H.P.Lee.MANETHO. W. G. Waddell.NONNUS. W. H. D. Rouse.PAPYRI: LITERARY PAPYRI. Selected and translatedby C. H. Roberts.PTOLEMY : TETRABIBLUS. F. E. Robbins.LATIN AUTHORSS. AUGUSTINE : CITY OF GOD. J. H. Baxter.CICERO : AD HERENNIUM. H. Caplan.CICERO : DE ORATORE. Charles Stuttaford and \V. E.Sutton.CICERO : BRUTUS, ORATOR. G. L. Hendrickson andH. M. Hubbell.CICERO: PRO SESTIO, IN VATINIUM, PROCAELIO. DE PROVINCIIS CONSULARIBUS, PROBALBO. J, H. Freese.COLUMELLA : DE RE RUSTICA. H. B. Ash.PRUDENTIUS. J, H. Baxter.QUINTUS CURTIUS: HISTORY OF ALEXANDER.J, C. Rolfe.DESCRIPTIVE PROSPECTUS ON APPLICATION^Cambridge, Mass. HARVARD .London . . WILLIAMUNIVERSITY PRESSHEINEMANN LTD