STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world byJSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.istor.org/participate-istor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. Classical Philology Volume VIII January, IQIS Number i SOME TEMPORAL EXPRESSIONS IN SUETONIUS By John C. Rolfb I. "Post biduum," JvH. 43, Av^g. 10; "Biduo post," Jvl. 16. In Suetonius, Jul. 43.1, "diremit nuptias praetorii viri, qui digressam a marito post biduum statim duxerat," the phrase post biduum is generally translated "two days later." The enormity of the offense of the ex-praetor is, however, considerably emphasized if we take post biduum to mean "on the following day," and that this is the correct rendering is also suggested by statim. Again in Jul. 16 we are told that after Caesar had supported the tribune Metellus in his revolutionary designs, he was deposed from his praetorship, while Metellus also lost his tribunate. The biogra- pher continues: "multitudinem quoque biduo post sponte et ultro confluentem operamque sibi in adseranda dignitate tumultuosius poUicentem [Caesar] conpescuit." Here bidu^ post is parallel in meaning with post biduum, and again the situation favors the meaning "on the following day"; for the mob was a hasty and spontaneous demonstration of devotion to Caesar (sponte et ultro confluentem), and as such was likely to have collected on the next day, rather than to delay its action until two days after Caesar's deposition; although here too the translators seem to be unanimous for the latter meaning. In Aug. 10.4, speaking of the battles at Mutina, Suetonius says: "priore [proelio] Antonius fugisse eum [=Octavianum] scribit ac sine paludamento equoque post biduum demum apparuisse." Now it [Classical Philoloqy VIII, January, 1913] 1 2 John C. Rolfe seems hardly probable that the future ruler of the world ran so fast and so far (or that Antony said he did), that he did not turn up until two days after the battle, although to be strictly impartial, demum perhaps favors this interpretation as much as statim does a different one in Jul. 43. But it is a sufficiently serious charge to make, and one on the whole more likely to be believed, that Octavian was not seen until he finally turned up, minus his horse and his general's cloak, on the following day, and I should be inclined so to translate the words post biduum. In these passages, so far as I have been able to ascertain, we have only mere probability to support the view that post biduum and biduo post may sometimes mean "on the following day," and the lexicons seem to give no hint that such a meaning is possible. Yet this signification might be inferred from the peculiar usage of the Romans in expressing time relations, as well as a corresponding meaning for bidv^ ante and ante biduum. Everyone knows that while the day before the Kalends of any month was called pridie Kalendas, the preceding day was called, not ante diem secundum (or alterum), but ante diem tertium. The latter is obviously synonymous with triduo ante, and therefore biduo ante should be parallel with pridie; while the expressions biduo post and post biduum would naturally have a corresponding meaning. We may compare nundinae for a period of eight days, decem menses as the period of gestation, and similar indications of a method of reckoning time which differs from our own, although it has parallels in other modem languages. Fortunately we are not confined to this argument from proba- bility, or it would be futile to discuss the matter; for there are one or two passages in which these phrases surely have the meaning sug- gested, and a few others in which the likelihood of its correctness is very strong. In Caes. B.G. i. 47. 1, after we have been told in the preceding chapter of the treacherous and unprovoked attack which the German cavalry made on the Romans during the conference between Caesar and Ariovistus, we read: "biduo post Ariovistus ad Caesarem legates misit: velle se de his rebus quae inter eos agi coeptae neque perfectae essent agere cum eo." Two lines farther on we find: "colloquendi Caesari causa visa non est, et eo magis quod pridie eius diei Germani retineri non potuerant quin in nostras tela Some Tempokal Expbessions in Suetonius 3 conicerent." Here the commentators and critics, although they differ in their renderings of the phrase, seem for the most part strangely oblivious of the contradiction between biduo post, if it be taken in the sense of "two days later," and pridie eius did. The latter is a very definite and unmistakable expression, meaning "the day before," and the reference can hardly be to anything other than the attack referred to in the preceding chapter; for if a second attack had been made on the Romans, it seems scarcely conceivable under the circumstances that Caesar should not have mentioned it, or at least have added the word iterum to his reference in chapter 47. It is significant too, if not in itself convincing, that he uses the same words, "tela in nostros conicere" (conicerent) in both references to the attack. The meaning "on the following day" is given to biduo post by Stock and Herzog in their editions of the Gallic War (and perhaps by some others), and I gave the words the same meaning independently in my own edition. So far as my examination goes, the greater number of the commentators do not translate or comment on biduo post at all, which justifies the inference that they take the phrase in the sense which it is generally assumed to have, and which is the only one given by the lexicons; nor do they refer to the contra- diction which is then made by pridie eius did. Dr. T. Rice Holmes, in the second edition of Caesar's Conquest of Gaul, 64, translates biduo post "two days later," without comment. Napoleon, however, in his Jules Cesar, II, 91, without giving any arguments to support his view, assumes a second attack of the Germans on Caesar's troops. He says : " celui-ci refusa, d'autant plus que, la veille, les Germains ^talent encore venus lancer des traites sur les Romains." He is apparently followed by Stoffel, La guerre de CSsar et d'Arioviste, 60, who speaks of Caesar as "instruit par I'acte de perfidie de la veille." Although supported by no evidence, and apparently by no argument, the assumption of a second attack is the only logical means of escape from the dilemma, if biduo post means "two days later"; yet, as has been said, it seems hardly conceivable that Caesar, who is endeavoring to emphasize the treachery of the enemy and the provocation to which he himself was submitted, should not have made at least a casual reference to a second attack, had it actually been made. Moreover, the Germans, at the time of 4 John C. Rolfe this alleged attack, were encamped twenty-four miles away from Caesar's camp {B.G. 41. 5), since it was not until the next day {eodem die, 48. 1 =eius did, 47. 2) that Ariovistus took up a position nearer to the Romans. The Germans, too, were acting strictly on the defensive, and running no risks, except when such an opportunity presented itself as they thought that they had during the conference, and might have had but for Caesar's foresight. It therefore seems certain that those are right who give biduo post in this passage the meaning "on the following day." Another passage in which this meaning seems all but certain for biduo post is Livy 36. 40. 3, where Scipio is reported to have said: "se de Gallis Boiis postulare triumphum, quos acie vicerit, castris exuerit, quorum gentem biduo post pugnam totam acceperit in dedi- tionem." In36. 38. 7 we read of this same victory: "magnamtamen victoriam fuisse apparet, quod et castra capta sunt et Boii post earn pugnam extemplo dediderunt sese." Here extemplo post pugnam certainly suggests a surrender on the following day at latest, rather than one made two days after the victory. There is happily at least one passage in which this meaning can be demonstrated mathematically for the corresponding phrase, ante biduum. This is Capitolinus, Aur. 27. 11, where we read of Marcus Aurelius: "ante biduum quam exspiraret admissis amicis dicitur ostendisse sententiam de filio eandem quam Philippus de Alexandro, cum de hoc male sentiret." Then in 28. 3 Capitolinus continues: "deinde abstinuit victu potuque mori cupiens auxitque morbum. Sexta die vocatis amicis .... dixit .... (6) et cum ab eo quaereretur cui filium commendaret, ille respondit 'Vobis, si dignus fuerit, et dis immortalibus'." Further on we read: " (8) septimo die gravatus est ... . (9) sed nocte animam efHavit." Here there seems to be no doubt that the emperor died on the seventh day after he refused food and drink, and that he called in his friends and expressed his opinion of his son on the sixth day after making the same resolution. Therefore "ante biduum quam exspiraret" clearly means "the day before he died." Perhaps we should not exclude the possibility that the emperor's death occurred after midnight, in which case he might be said to have died on the eighth day of his illness, but this seems an unwarrantable supposition, and Some Temporal Expressions in Suetonius 5 Merivale, History of the Romans under the Empire, VII, 476, puts his death, in accordance with the only natural interpretation of the words of Capitolinus, on "the seventh day of his illness." A second passage in which the context suggests this meaning for hiduo ante is Ulpian, Dig. 21. 1. 38. 11, "vendendi autem causa omatum iumentum videri Caelius ait, non si sub tempus venditionis, hoc est biduo ante venditionem omatum sit, sed si in ipsa venditione omatum sit." There are numerous other passages in which the meaning "on the following [or preceding] day" seems a natural one for biduo post or biduo ante, although no exact determination of their meaning is apparently possible; for example, Livy 31. 46. 16, "praesidium oppidanique in arcem alteram perfugere, unde bidua post deditio facta"; Livy 44. 18. 6, "senatus Cn. Servilio consuli negotium dedit, ut is in Macedoniam quos L. Aemilio videretur legaret. Legati biduo post profecti"; Livy 32. 27. 6; "Sextus Aelius consul ex Gallia comitiorum causa Romam cum redisset, creavit consules .... bidvo post praetorum comitia habita"; Cic. Pis. 61, ad Att. 13. 43, ad Fam. 10. 14. 1, and numerous others. In those doubtful cases in which haste, or speedy action, is implied or called for by the context, the constant recurrence of biduo seems in itself suspicious, unless it means " on the following [or preceding] day," and in the light of the passages from Caesar and Capitolinus it seems necessary at least to consider the possibility of that meaning. Passages in which biduo post surely means "two days later" are comparatively rare, it would seem. One such perhaps occurs in Vol. Max. 1. 8. ext. 1, "quae minus admirabilia fere Eris Pamphyli casus facit, quem Plato scribit inter eos qui in acie ceciderant decem diebus iacuisse, biduoque post quam inde sublatus esset, impositum rogo revixisse." One would hardly expect the funeral of a man who had lain for ten days on the battle field to be long deferred, and arguing from probability alone, one might be inclined to suspect that here too biduo post means "on the day after." But the words of Plato, Rep. X. 614 B., seem to exclude this interpretation: 'H/ao? rov 'Apfieviov, to 'yevov TlafKpvXov 09 Trore iv -TroXe/icp Te\evTij<7a(; avacpedevTCOv SeKaraimv t&v veKpcov fjir} Bte(f}0ap/j,e'vcov iyiT)<; /lev avrjpedi], KOfiicrOel^ S^ oixaSe //.eXXxov daTrrecrOac ScoSeKaTaloi i-rrl ttj 6 John C. Rolfe irvpa Keiixevoi ave^ica. It seems evident that the funeral took place two days after the body was found, and Plato makes it clear why there was no haste (w^t^? avypedrj) . This is perhaps implied in the less circumstantial accoimt of Valerius Maximus, although I must con- fess that before I compared Plato's account, I was inclined to take hiduo post here also in the sense of "on the day after." This shows the uncertainty of arguments based on probability alone, and makes one thankful for the firm footing afforded by the passages from Caesar and Capitolinus, in which, by the way, we have evidence for the persistence of the meaning "on the following [preceding] day" for some centuries. A peculiar use of the term biduo post occurs in Cicero, Quinct. 79: "bonapostulas ut ex edicto possidere liceat. Quo die? .... Ante diem V Kalendas intercalaris De saltu deicitur Quinctius quo die? ... . Deicitur de saltu pridie Kalendas intercalaris. Bidtw post aut, ut statim de iure aliquis cucurrerit, non toto triduo DCC milia passuum conficiuntur." If biduo post here meant "on the following day," that is, the day following ante diem V Kalendas intercalaris, the reference would be to the fourth day before the Kalends, which is obviously impossible. Pridie Kalendas would be, according to our method of reckoning, the third day after the fifth before the Kalends, and according to the usual Roman method of reckoning, the fourth. Bearing in mind, however, that Cicero is trying, in the interests of his client, to make the time which elapsed between the postulatio and the eiedio appear as short as possible, and that he feels obliged to qualify biduo post by non toto triduo, we might fairly add this passage to the one from Valerius Maximus, and thus extend the use of biduo post in the sense of "two days later" over a considerable period. It is clear, however, that the most natural translation of biduo post in this passage is not "two days later," but "within two days thereafter"; in other words, this use of biduo post is not exactly parallel with those which we have been considering. We might therefore exclude this passage from consideration and class it with the ordinary uses of biduo in the sense of "within two days" or "for [during] two days"; see Caes. B.G. 5. 27. 8, "magnam manum Germanorum conductam Rhenum transisse; hanc affore biduo" (where Stock takes biduo to mean "on Some Temporal Expressions in Suetonius 7 the following day"); Nepos, Hann. 6. 3, "Mduo et duabus noctibus Hadnimetum pervenit;" Suet. Calig. 19. 2, "per hunc pontem ultro citroque commeavit Mduo continent!." The passage from Valerius Maximus cannot be so readily disposed of, unless we may be allowed to say that "two days after" is not precisely the same thing as "on the second day thereafter," a statement which does not seem wholly unreasonable, and to take the words of Valerius in the former sense. Of the numerous passages which I have examined there are no others in which the meaning "on the following day" does not seem to be a possible one for hiduo post, with corresponding meaning for post biduum, hiduo ante and ante biduum. There are two ways of interpreting this evidence. We may reject the passages from Cicero and Valerius Maximus, and maintain that biduo post, post biduum, biduo ante, and anie biduum, as concrete idiomatic (and perhaps more or less colloquial) expressions of time, always meant "on the following [preceding] day." They could be so used even in legal language, in which an effort is made to be exact and free from ambiguity, if our interpretation of the passage from Ulpian is correct. Or if this attempt to prove consistency on the part of the Romans in the use of exactly similar expressions be regarded as savoring of special pleading, we may say that these expressions sometimes meant "on the following [preceding] day," and sometimes "on the second day afterwards." In either case the former meaning should be recognized by the lexicons, and its pos- sibility should be considered in all cases in which the time is not otherwise defined. It seems the more probable one in the three passages from Suetonius. II. De die, Calig. 37; Domit. 21. In Suet. Calig. 37. 2 we read: "fabricavit et deceris Liburnicas. .... quibus discumbens de die inter choros ac symphonias litora Campaniae peragraret," and in Domit. 21, "lavabat de die prande- batque ad satietatem." In both of these passages de die has its usual meaning of "in the daytime" (see PAPA, XXXIV, Iv ff.), but in both instances, as well as in numerous other passages, that translation without further qualification is obviously misleading. Dies may refer to the complete civil day, from midnight to the next midnight 8 John C. Rolfe (Paul. Dig. 2. 12. 8, "more Romanorum dies a media nocte incipit et sequentis noctis media parte finitur"); to the day in distinction from night, from sunrise to sunset (Quint. 5. 8. 7, "sol est super terram, dies est") ; and finally to the "working day," the solidus dies of Horace, Odes, 1.1. 20.^ The second of these two meanings is certainly the most obvious, and it is probably the most usual, so that dies would naturally be taken in that sense when there is nothing in the context to suggest a different interpretation, and in English "in the daytime" would ordinarily be understood in that way. In the two passages from Suetonius, however, the reference is clearly to the solidus dies, more obviously perhaps in the second, since the practice of bathing only at night could hardly have been common enough to make Domitian's practice worthy of note, if he merely bathed "in the daytime," and not before the usual hour for the bath. Bagge, De elocutione C. Suetonii Tranguilli, Upsala, 1875, 70, referring to Calig. 37, but not citing the obvious parallel in Domit. 21, says: "de die=medio die. Ita Ruhnk. ad h. 1. Ter. Ad. 5. 9. 7 ( =965). Liv. 25. 23." It appears from this note that Bagge recog- nized that de die refers to the working day, but although medio die is in a very general way synonymous with solido die, it is misleading as an equivalent for it and for de die. There seems to be a difference of opinion as to the exact meaning of solidus dies, which is apparently not of very frequent occurrence, in spite of the fact that it is perhaps the only expression which designates the "working day," except the ambiguous dies. The commentators as a rule say that the solidus dies extended to the hour of the cena, citing Martial 4. 8. 6. But Martial's words are: In quintam varies extendit Roma labores; Sexto quies lassis, septima finis erit. Sufficit in nonam nitidis octava palaestris; Imperat exstructos f rangere nona toros.^ 1 Sen. Episi. 83. 3, which is quoted by many editors of Horace as a parallel, may be seen not to be one, if the next sentence be added to the quotation: "hodiernus dies solidus est. Nemo ex illo quicquam mihi eripuit. Totus inter stratum lectionemque divisus est." ! Mau's interpretation in Marquardt-Mau, Privatleben, 263, n. 6, seems the correct one. Some Temporal Expressions in Stjetoniits 9 The editors of Horace express various views, saying that the solidus dies extended "to the end of the ninth hour," "to the middle of the afternoon," "to about three p.m.," etc., etc. Perhaps the greater number confine themselves to citing Sen. Epist. 83. 3, which is not particularly illuminating, as has been shown, but is perhaps the safest course, for there can be no doubt that solidus dies was, like most general temporal expressions of the kind, an elastic term, not to be defined exactly for all cases in terms of the clock. Its duration doubtless varied not a little according to individual habits and tastes. Some men probably worked steadily, or with a very brief intermission, until shortly before the hour of the cena, which all seem to agree in assigning to the ninth hour; while others allowed more time between the end of their solidus dies and the principal meal of the day for the siesta, which, as its derivation shows, came at the sixth hour, for exercise of some kind, and for the bath. Probably, too, the habits of the same individual varied somewhat according to the season of the year and other circumstances. That the solidus dies was the part of the day spent in actual work of some kind, and that it did not include the time devoted by some to the siesta, to exercise, and to the bath, is a priori probable, and is indicated by the fact that Horace's reference is probably to a man who broke into his solidus dies by "a long siesta at noon" (Page) ; cf ., however, OreUi, ad. I. As to the duration of the solidus dies, we see that even a vir strenuus like Philippus (Hor. Ejrist. 1. 7. 47) left the forum about an hour before dinner time ("octavam circiter horam"), in order to have time to return to his home on the Carinae without undue haste and perhaps to bathe or otherwise curare corpus before dining. It seems probable that the solidus dies must in all cases have ended somewhat before the ninth hour, to allow at least time enough to return home from the place of one's business or profession. For those who indulged in the siesta, exercise, and bathing it must have ended earlier still, and probably at the time mentioned by Martial; cf. Juvenal, 11. 204 f.: "iam nunc in balnea salva Fronte licet vadas, quamquam solida hora supersit Ad sextam." Here, of course, solida is used in its literal sense of "whole"; cf. Stat. Silv. 4. 3. 36 "At nunc quae solidum diem terebat, Horarum via facta vix duarum"; Livy 1. 19. 6, "desuntque dies solido anno." A "working day" 10 John C. Rolfb which ended at so early an hour is less surprising when we recall the well known fact that many Romans of all classes began their day's work before dawn; see for example, Suet. Vesp. 21. The translation of de die in the sense of solido die is not easy, for "during the working day," although accurate, is rather clumsy. In the first of the passages from Suetonius the translators for the most part say simply "in the daytime," which has been shown to be misleading. Holland, however, has "in the very daytime," an indication that he felt the meaning of the phrase, although he is not so happy as usual in bringing out its exact force. In the second passage, although de die obviously has the same meaning as in the first, it is rendered "early," "by noon," and by Holland, less success- fully even than in the former case, "by daytime." Since in both these passages the dies solidus may fairly be assumed to have its shortest duration, de die might be translated "in the forenoon," a sense which seems certain for the second passage and not improbable for the first. It also suits Plant. Asin. 825, Catull. 47. 6, Livy 23. 8. 6, Curt. 8. 3, and Eutr. 7. 18. We must distinguish from the above mentioned use of de die those cases in which the phrase, usually if not always in connection with diem, means " [day] after day." This use of de in the sense of " after " seems to be confined to verbs implying motion (literal or figurative, expressed or understood; see PAPA, XXXIV, Ivi); for example, Livy 5. 48. 6, "diem de die prospectans" ; Livy 25. 25. 4, "cum is diem de die diflferret." With this in dies is sjTionymous, and is used with verbs not implying motion; less frequently we have the singular in diem, and very rarely ad diem: XII Tab. ap. Gell. 20. 1. 45, "libras farris endo dies dato"; Caes. B.G. 3. 23. 7; Livy 22. 39. 13, "[Hannibal] in diem rapto vivit"; Cic. Phil. 2. 34. 87, "cum per- ditissimis latronibus non solum de die, sed etiam in diem bibere" (var. lect. vivere; here the singular is perhaps used for the sake of the word-play with de die, "in the daytime," i.e., "in broad daylight"); Treb. Poll. Gallien. 17. 4, "lavit ad diem septimo aestate vel sexto, hieme secundo vel tertio." III. De media nocte, Calig. 26. In Suet. Calig. 26. 4, "inquietatus fremitu gratuita in circo loca de media nocte occupantium, omnis fustibus abegit," two things seem Some Temporal Expressions in Suetonius 11 to be clear about the expression de media node: first, that the mean- ing of de is in general " at," not " after" ; and secondly, that it is used with comparatively extensive periods of time, which are regarded as not yet completed; see PAPA, XXXIV, Ivi. Bearing in mind these two points, we may translate the phrase "in the middle of the night." This is more accurate than "at midnight," and it enables us to exclude the possibility that de with the ablative ever means "from .... on," which is expressed by the ablative with 06; see PAPA, XXXIV, Ivii. It is rather surprising to find Bliimner in Die romischen PrivataUer- tumer, MuUer's Handbuch, IV, 2, 3d ed., 374, taking de media node as an actual division of the night, equivalent to "mediae noctis inclinatio." This makes it necessary to take de in the sense of "after," which it apparently never has in such a connection, and Dissen, whose De partibus nodis et diei ex divisionibus veterum Blum- ner cites, showed clearly enough that the nine divisions of Censorinus (De Die Nat. 24) and the seven of Macrobius {Sat. 1. 3. 12 ff.; the numbers are significant, although Dissen does not seem to note this) consist of general terms, which can no more be considered actual formal divisions of the night (like the four vigiliae, for example), than solidus dies can be regarded as a formal division of the day. Cen- sorinus would have us believe that there was a period before midnight called "ad mediam noctem" and one after midnight called "de media nocte." But if these were in any sense actual divisions of the night, we should have to consider the possibility of using such expressions as "post de media nocte," "ante ad mediam noctem," and the like. Bliimner avoids such a possibility by excluding prepositional phrases from his list, with the exception of "de media nocte," which he gives in parentheses as an alternative for "mediae noctis inclinatio." He also avoids the suspicious nine and seven by making eight divisions. Even so, it seems inconceivable that there should have been eight divisions of the kind which Bliimner gives, in any real sense of the term; and even Dissen's four divisions, which he believed to correspond with the four vigiliae ("prima nox, nox concubia, media nox, and gallicinium") can hardly have been fixed and definite, or used with any consistency and regularity. Whether or not we accept any or all of these divisions, it seems 12 John C. Rolfe certain that media nox meant in the phrase de media node, not "midnight," but "the middle of the night;" in other words, that it did not designate a point of time, but a period of some duration, although its limits were not strictly defined either in hours or with reference to other divisions, real or assumed. We may note Juvenal 5. 54: "et cui per mediam nolis occurrere noctem." The same thing is true of de msdio die, and apparently medius dies as a rule designated an indefinite period before and after noon, while m£ridies was used to designate the point of time called midday. See Cic. Tusc. Disp. 2. 3. 9, "cum ante meridiem dictioni operam dedissemus, post meridiem in Academiam descendimus" ; Horace, Odes 3. 28. 5. It is true that this distinction, convenient as it is, is not always observed, since examples like "aestivom diem si non diflSnderem meo insiticio somno meridie, vivere non possem (Varro, B.R. 1. 2. 5) and the existence of the verb mmdio (m-eridior) for taking a noonday nap (CatuU. 32. 3; Suet. Calig. 38. 3) show that meridies was sometimes used of a period of some duration; that is, that it was synonymous in such cases with medius dies. Censorinus tells us (24. 3) that the day was divided into four parts, corresponding to the watches of the night, of which two were called "ad meridiem" and "de meridie," and Bliimner (373) seems to accept this division. The obvious disadvantage of prepositional phrases as actual divisions of time has already been spoken of, and the improbability of their existence indicated. While ad meridiem occurs in the sense of "until [up to] midday," I have looked in vain for an example of "de meridie" outside of the passage from Cen- sorinus. When meridies designates a point of time, "de meridie" is a very doubtful expression, if not an impossible one; see PAPA, XXXIV, Ivi. In other cases, if such exist, it must be synonymous with de medio die and mean "in the middle of the day." If media nox and medius dies designate indefinite periods of some duration, all diSiculty connected with the expressions de m^dio die and de media node disappears. The usage corresponds exactly with such expressions as de tertia vigilia, which surely means "in the third watch," and not "after the third watch," to correspond with the hypothetical "de meridie" and with "de media nocte" as a divison of the night; still less does it mean "about the third watch," as it Some Temporal Expressions in Suetonius 13 is often mistranslated. Just as de tertia vigilia is a more or less indefi- nite expression, referring to any part of a period of approximately three hours, but varying with the time of year, so de media node refers to any part of an indefinite period in the vicinity of midnight. The former expression, however, is the less indefinite, since it always consisted of a fourth part of the period from sunset to sunrise, while the latter had no fixed duration, so far as we know. Hence vigilia (prima, secunda, etc.,) is a division of the night in the true sense of the word, while de media node is not. Media node seems usually to be sjmonjTnous with de media node; that is, both expressions commonly mean "in the middle of the night," and not "at midnight"; cf. Suet. Aug. 94. 4, "cum ad soUemne ApoUinis sacrum' media nocte [Atia] venisset" ; Claud. 13. 1 "e plebe homo nocte media iuxta cubiculum eius [=Claudii] cum pugione deprehensus est." In the absence of a term corresponding with m^ridies, media nox perhaps sometimes referred to the point of time called midnight; Suet. Nero, 47. 3, "ad mediam fere noctem excitatus;" Claud. 2. 2, "togae virilis die circa mediam noctem sine soUemni officio lectica in Capitolium latus"; Domit. 16. 1, "circa mediam noctem est exterritus." This does not, however, appear to be very common, as is perhaps indicated by the occasional use of a more definite expression, for example in Juv. 7. 222, "mediae quod noctis ab hora Sedisti" and Varro, Sat. Menip. 269 Biich., "repente noctis circiter meridie." At any rate, cases in which the meaning "in the middle of the night" is absolutely excluded are not frequent, if they exist at all. Univeesitt of Pennsylvania