760f 
 
 L843 
 
 m 
 
LIBRARY 
 
 OF THK 
 
 UNIVERSITY OF CALIFORNIA 
 
 OF" 
 
 Class IS'OS 
 &. 
 
ON THE USAGE 
 
 OF 
 
 QUOT1ENS AND QUOTIENSCUNQUE 
 
 IN 
 
 DIFFERENT PERIODS OF LATIN. 
 
 A DISSERTATION 
 
 PRESENTED TO THE BOARD OF UNIVERSITY STUDIES OF 
 
 THE JOHNS HOPKINS UNIVERSITY FOR THE 
 
 DEGREE OF DOCTOR OF PHILOSOPHY. 
 
 BY 
 
 OMERA FLOYD LONG, 
 
 SOMETIME UNIVERSITY SCHOLAR AND FELLOW. 
 
 BALTIMORE: 
 
 JOHN MURPHY COMPANY, 
 i got. 
 
ACKNOWLEDGMENT. 
 
 During my graduate work courses were pursued under Professors 
 Warren, Gildersleeve and Bloomfield, and Associate Professor 
 Smith. To each of these I can but imperfectly express my grati- 
 tude for various kindnesses, and for lasting inspiration as well as 
 helpful instruction. 
 
 Owing to the range of authors cited in my examples the manu- 
 script of this dissertation, though complete in its inferences and 
 its main outlines, was at first submitted in provisional form. 
 Later a visit to Munich enabled me through the generous courtesy 
 of Professor Wolfflin to verify many of my references with the 
 help of the " Thesauruszettel." Professor Wolfflin himself, at 
 various points in the revision, gave helpful suggestions, such as 
 emphasis on quam saepe, of which I had at that time only a few 
 of the more striking examples, pointing out the occurrence of 
 nesoio quotiens in the Notae Tironianae, and additional examples 
 of quotienslibet, together with remarks on gemination in con- 
 nection with the compound form. Further points of indebtedness 
 will be acknowledged as far as possible in the following pages by 
 reference to his Zusdtze in an abstract of this dissertation which 
 appeared in the Archiv fur lateinische Lexikographie und Gram- 
 matik, xi, p. 395 if. an abstract due also to Professor Wolfflin's 
 generous suggestion. 
 
 O. F. LONG. 
 
 NORTHWESTERN UNIVERSITY, 
 EVANSTON, ILLINOIS. 
 
CONTENTS. 
 
 PAGE. 
 
 INTRODUCTION, 7-8 
 
 PAET I. 
 
 DERIVATION, 9 ~H 
 
 FORM, - - - H-13 
 COMPOUNDS, 
 
 MEANING, 19-23 
 
 AUTHORS EXAMINED, 23-24 
 
 PAET II. 
 
 CITATION OF EXAMPLES, 25-35 
 
 PART III. 
 
 SUMMARY, 36-48 
 
INTRODUCTION. 
 
 Little or nothing hitherto has been written with reference to 
 quotiens. The grammarians have been content with brief state- 
 ments as to its general usage, which was assumed as well under- 
 stood. Hand's Tursellinus, De Particulis Linguae Latinae was 
 interrupted before reaching quotiens; Reisig's Yorlesungen iiber 
 lateinische Sprachwissenschaft, with its various revisions, has 
 nothing; Draeger, Historische Syntax der lateinischen Sprache, 2 
 n, p. 584, dismisses the word after stating in substance that it 
 is found with the indicative in classical Latin, while in the 
 period of silver Latin and later many authors use the sub- 
 junctive mode. Only a few examples, however, are cited by 
 Draeger. Here and there, in a line, an editor has discussed 
 the form of quotiens, or perhaps has noted its substitution for 
 ubi, cum or si quando. The form has naturally been discussed 
 with greatest fullness by Neue, Formenlehre der lateinischen 
 Sprache, and by Brambach, die Neugestaltung der lateinischen 
 Orthographic. On the more difficult question of quotiens with 
 the subjunctive in iterative sentences, practically nothing has been 
 contributed beyond the brief paragraph in Draeger. Such works 
 as have been helpful in the discussion of this phase of the subject 
 have been quoted in Part III. 
 
 It is obvious, however, that for any final judgment in syntax 
 the historical perspective must be used, and each individual word, 
 with any possible alternatives before it, must give a complete 
 account of itself. The desired results may ultimately be reached 
 by investigating every point in individual authors, successively, 
 until the literature has been covered ; or by following a single 
 usage through approximately all the authors, until the various 
 moot-points have thus been historically treated. The present 
 paper offers some such treatment of quotiens. The ante-classical 
 
 7 
 
8 Introduction. 
 
 and classical literature has been thoroughly covered, and there 
 are but few omissions in the later period extending as far down 
 as Boethius and Gregory of Tours. Altogether, about one hun- 
 dred and thirty authors have been examined. 
 
 The collection of examples has of course been made with the 
 help of special lexicons and indices verborum where these were 
 available and trustworthy. In most cases, however, the instances 
 had to be collected for the first time ; -accordingly, the privilege 
 of "nifallor" is asked for in stating totals. It may be claimed 
 with confidence that the general usage for the long period con- 
 sidered has been accurately shown, since the possible omission 
 of an example here and there would make no material difference 
 in the results obtained. 
 
DERIVATION. 
 
 Quotiens is formed with quot as its basis, as Miens from tot, or 
 the post-classical multotiens from multus, plus the suffix -iens, 
 making a derivative analogous with the multiplicative numerals 
 in -ies (-iens). Various explanations for this suffix have been 
 proposed, any one of which, if satisfactory for one word, would 
 suffice for the entire group, since the principle of analogy, power- 
 ful every where, is especially strong in the case of numerals. 1 Some 
 of these may be examined. 
 
 a. Aufrecht, Die lateinischen Zahladverbien auf -iens, K. Z. I, 
 121 if., holds that -iens represents a neuter comparative suffix to be 
 compared with the neuter adjective suffix -ius (-ios). This view 
 has little probability from either the phonetic side or the meaning, 
 and apparently finds no support from later scholars. 2 
 
 6. Joh. Schmidt, K. Z. xxv, 137 anm. 2, would equate 
 triens ( = *triins) directly with rpids, from which we might sup- 
 pose other forms spread. Brugmann, 3 however, is right in reject- 
 ing this equation as unwarranted. 
 
 c. In the Archiv fur lateinische Lexikographie und Gram- 
 matik, v, 136 f., Stowasser has proposed a new explanation, based 
 upon the simple juxtaposition of the participle iens with quot, 
 tot, etc.; so, for example, he cites Verg. .ZEn. 6, 122 
 
 Itque reditque viam tot iens. 
 
 i. e., tot < itiones > iens " in soviel Gangen " ; and similarly with 
 other forms in -ies (-iens), as quinqu(e) iens, sex iens, sept(em) iens, 
 
 1 Cf. Osthoff, Morphologische Untersuchungen, I, 92 ff., and Baunack, Kuhn's 
 Zeitschrift f. vergl. Sprachforschung, xxv, 253. 
 
 8 Corssen, Aussprache, Vokalismus u. Betonung d. lat. Sprache, n, pp. 351, 552 
 anm., must be excepted. 
 
 3 Grundr. d. vergl. Gram. II, 368. 
 
 2 9 
 
10 On the Usage of Quotiens and Quotienscunque. 
 
 etc. The formation entered the language very early, before the 
 present law for accentuation obtained, and readily took on a 
 numeral connotation on account of the Latin aversion to such a 
 construction. 
 
 That this view simplifies the phonetic difficulties can hardly 
 be denied, and a similar phenomenon has been pointed out 1 in 
 the periphrastically formed multiplicatives of Anglo-Saxon, where 
 sftft "journey" is added to both ordinals and cardinals, 2 and of 
 Hebrew, which makes a similar use of paam " step, walk/' * 
 Thurneysen 4 dismisses this explanation as " kiinstlich " ; but 
 Delbriick 5 approves it, Deecke 6 also, and one must admit that it 
 is rather seductive, although the principle of juxtaposition and 
 reduction is on the whole unscientific, and accordingly to be 
 accepted in any case with great caution. 7 
 
 d. Breal, Memoires de la Societe" de Linguistique de Paris, vin, 
 474, like Stowasser, has also attempted an explanation on Latin 
 ground. From quotus, " how many," with the nominal suffix -ia t 
 -ie, which is used secondarily (cf. mater-ie-s, temper-ie-s, etc.), he 
 would make a derivative quoties, 8 i. e., to form a proportion, 
 quotus : quoties : : quantus : quantitas, or qualis : qualitas. From 
 this formation decies, centies, etc., would follow by analogy ; but 
 the persistency with which n clings in quotiens and its regular 
 occurrence early in the numerals are against Briar's view, 9 even 
 if the meaning could have been reached in this way, which may 
 be doubted. 
 
 e. Still another, and, on the whole, more satisfactory derivation, 
 is offered by Thurneysen, Zu den Zahladverbien auf -lens, ALL. 
 V, 575. Here it is rightly assumed that the suffix did not 
 
 1 See Kellner's note on Stowasser's article, 1. 1. 
 
 3 Cf. Bosworth-Tallieur, Lex. Ang.-Sax., s. v. 
 
 3 Cf. Gesenius, Hebr. u. Chald.Worterb., s. v. * ALL. v, 575 ; cf. infra. 
 
 5 Vergl. Synt. p. 628. 6 Erlauterungen z. Iat. Schulgram. \ 71, 1. 
 
 7 See Stolz, ALL. v, 285, and Mullens Handb. d. klass. Altertumswissenschaft, 
 n 8 , 2 Abteil. p. 145. 
 
 8 "Nous avons ici une expression technique venue de la langue du calcul. 
 C'est-a-dire un substantif de*signant 1'un des facteurs de la multiplication ou de 
 la division." 
 
 9 Though he generously allows n to be from an old ace. plur., preserved by 
 " purification adverbiale," or else parasitic. 
 
On the Usage of Quotiens and Quotienscunque. 1 1 
 
 originate on Italian soil. The word quotiens itself may be com- 
 pared with Skr. kiyant, 1 " how great, how many," whose neuter, 
 kiyat, employed adverbially, is taken as the starting point. From 
 -yat, with a basis of int, nasalis sonans having entered, assume 
 Ital. -lent, which appears as -iens through a change of -nt to -ns. 
 This change is warranted by what we see in the Oscan and 
 Umbrian verbs, where the ending -ns of the third plural in 
 secondary tenses stands for original -nt, while the final -nt of the 
 primary tenses stands for Ind.-Ger. -nil ; 2 but, since ordinarily 
 final -ns became -f, it must be assumed on Thurneysen's hypothesis 
 that the change of -nt to -ns is not primitive Italic. Von Planta, 
 1. L, has suggested that the Osc.-Umbr. -nt first became nd (cf. the 
 change of -t to d in the third sing, of secondary tenses), then 
 -nd > -nZj through -rr3, and -nz naturally became -ns. Here 
 again we must suppose -nd > -n% before the Osc.-Umbr. assimila- 
 tion of -nd to -nn. 3 A similar change would be that of the neuter 
 participle ferens = *ferent (cf. Skr. bharat, Ind.-Ger. *bhernt\ 
 perhaps after the analogy of adjectives like ingens, etc. 4 
 
 This view of Thurneysen's, while not wholly free from phonetic 
 difficulties, is accepted by Brugmann, 5 Stolz, 6 Lindsay, 7 and may 
 be regarded as the most satisfactory yet offered. 
 
 FORM. 
 
 In the case of the multiplicatives, the classical form, from five 
 on, is with the ending -ies, for the earlier -lens, though the earlier 
 forms were often retained by the side of the later ones. 8 This 
 elimination of n before s was of not infrequent occurrence even 
 in old inscriptions, as the familiar mesibus, cosul, cesor, etc., attest, 
 and was doubtless due to a certain nasal quality assumed by the 
 
 1 Pron. ki (Lat. gui\ plus the suffix -yant. Cf. quot with kati. 
 
 2 Cf. Bugge, K. Z. xxn, 385 ff., and von Planta, Gram. d. osk.-umbr. Dialekte, 
 Strassburg, 1892, I, p. 513. 3 See von Planta, p. 417 f. 
 
 4 Thurneysen, 1. L, and Stolz, Handb. n 3 , 2, 145. 
 
 4 Grundr.Vergl. Gram, n, 562. 
 
 ALL. v, 285; Handb. n>, 2, 145. 
 
 7 Lat. Lang. pp. 284, 409. 
 
 * Prise. Corp. Gram, in, 77, 15 ff. ; Lindsay, p. 408. 
 
12 On the Usage of Quotiens and Quotiemeunque. 
 
 preceding vowel, so that the consonantal element was more and 
 more reduced, even to the point of total disappearance. 1 
 
 Later, even the -ies in the definite numerals was often reduced 
 to -is, both in literature and in inscriptions, as decis, sexsis, GIL. 
 xii, 2087 (559 A. D.). 2 
 
 With the indefinite quotiens, aliquotiens, totiens and multotiens, 
 however, the best tradition favors the retention of -lens. The 
 statements of the grammarians show the actual practice. Compare 
 Priscian, Corp. Gram, in, 415, 18 reliqua omnia in es pro- 
 ductam desinunt .... infinitis tamen numerorum adverbiis 
 etiam n interponitur, quotiens, totiens similiter multotiens. Marius 
 Victorinus, ibid, vi, 24, 23, and Caper vii, 95, 8, are equally 
 insistent. The n seems to have fallen out entirely at last, for 
 while Albinus vji, 308, 14 at least reports that formerly the 
 n was used : quoties, toties sine n, licet veteres per n, Beda VJI, 
 287, 5 makes no reservation : quoties, toties, septies, sine n. 
 
 The inscriptions and literature show that Priscian's rule was 
 not arbitrarily formed, 3 but was based upon practice. Augustus, 
 in the Mon. Anc., has only quotienscunque (iv, 28) by the side 
 of the earlier forms deciens, viciens, quadragiens, sescentiens, etc.,* 
 and similarly quotiens is the form in both the Lex Agraria and 
 the Lex Repetundarum, CIL. I, 200, z. 25; i, 198, z. 48, Com- 
 pare further n, 1963, col. 1, z. 38-39 ; n, 3367, and O.-H. 4358, 
 z. 20; 4374, z. 16. 
 
 For the Ciceronian form, compare C. F. W. Miiller, adnot. crit. n, 
 2, p. xxxi : totiens, quotiens quae formae nusquam non sunt in rnell. 
 codd.; and, by way of further illustration, the following may be 
 cited : quotiens, Ter. Hec. 60, though milies, Eun. 422 ; 5 quotiens, 
 Sail. Hist. Frag. 2, 98, 1 ; tricies, 4, 81 ; quotiens, Propert. 1, 3, 
 
 1 So Cicero is reported by Vel. Long. Corp. Gram, vn, 79, 1, as pronouncing 
 foresia instead of forensia, Megalesia instead of Megalensia. Compare, on the 
 whole question, Job. Schmidt, Zur Gesch. d. indogerm. Vocal, i, 98 ff., Brugmann, 
 Grundriss, i 2 , p. 370 f., and Seelmann, Die Aussprache d. Latein, p. 283 ff. Bennett, 
 Appendix to Latin Grammar, Boston, 1895, p. 20, while skeptical as to this ex- 
 planation, offers nothing in its stead. 
 
 2 Cf. Lindsay, p. 409, and Schuchardt, Der Vocal, d. Vulgarlat. II, 515. 
 
 3 Yet compare Brambach, p. 269. 
 
 4 See Mommsen, Res Gestae Divi Augusti, 1883, p. 192. 
 ^Miliens, Dziatzko, against the MSS. 
 
On the Usage of Quotiens and Quotienscunque. 13 
 
 27 ; decies, 2, 4, 16 ; quotienscunque, Fronto, p. 152 (N.) ; quinquies, 
 p. 139; Hygin. Grom. ch. 36 sed habemus numerum IIII milia, 
 videmus, quotiens habeo : quod est octies ; Ael. Lamprid. Corn- 
 mod. 11, 11, quotiens, but in 11, 12, septingenties tricies quinquies. 
 Martial presents strong contrast in forty examples of quotiens, 
 while decies occurs nineteen times; centies, seven ; tricies and vicies, 
 two, etc. 1 That the definite numerals were sometimes even cor- 
 rected (sic!) to the shorter form may be seen from Verg. Ge. 4, 
 480, where noviens of Rom. and the Sched.Vat. has the n stricken 
 out in the latter. 2 
 
 Note. The frequent confusion of ti and ci before a vowel in late 
 inscriptions and MSS. may be seen in the form quociens (quociens- 
 cunque), e. g. in Cato apud Gell. 2, 28, 6 (cod. Vat.), Cic. Plane. 14 
 (Erfurtensis), Martianus Capella, 3, 302, and Victor Vitensis, 
 Corp. Eccl. p. 83, 9. No inscription or older MS. gives this 
 spelling, which obviously contradicts the derivation, and in the 
 Notae Tironianae, according to so clever an interpreter as W. 
 Schmitz, 3 there is actual representation of the t in the middle 
 stroke of the character used for quotiens, 
 
 COMPOUNDS. 
 
 A final discussion of the relation of the compound to the simple 
 form must follow the citation of examples, yet it was very early 
 suspected that examples of quotienscunque would not be compara- 
 tively numerous. In some respects the reason for this disparity is 
 obvious at once, partly in the unwieldy length of the compound, 
 especially for poetic use, and partly in the general, indefinite nature 
 which quotiens itself could so readily assume. Accordingly prose 
 examples of quotienscunque should be far in excess of its use by the 
 poets, and this is precisely what we find to be true. Indeed, with 
 the unimportant exception of two examples in Plautus, it will 
 
 l Cf. Gilbert in Friedl. Mart. I, p. Ill ; Neue n, 336 f. 
 
 3 Cf. further, Hildebrand on Apul. Met. 2, 3, citing the evidence of older editors. 
 
 3 Zeitschrift Panstenographkikon, Band i, Dresden, 1874, on quotiens in the 
 Notae Bernenses, and see Tab. 10, 28 and 29 of Schmitz's Commentarii Not. Tiron. 
 
 Prof. Wolfflin, who furnishes the first reference, makes further interpretation 
 of the character, ALL. xi, p. 395. 
 
14 On the Usage of Quotiens and Quotienscunque. 
 
 be found that the compound is wholly confined to prose. Com- 
 pare e. g. Juv. 3, 318 
 
 quotiens te 
 
 Roma tuo refici properantem reddet Aquino, 
 Me quoque ad Helvinam Cererem vestramque Dianam 
 Converte a Cumis. 
 
 and Fronto, ad Amic. 1, 3 (p. 175 N.) : quotienscunque Roruam 
 venit, in meo contubernio fuit. 
 
 Numerous examples of quotiens occur where quotienseunque 
 could have been used. Occasional attempts have been made by 
 editors to point out such instances. So e. g. Knoell, Eugippius, 
 vit. Sanct. Severini, Corp. Eccl. p. 33, 1 quam, quotiens ripas 
 excessisset, aqua superfluous occupabat, and again p. 5, 1 9 quotiens 
 sermo ortus est ; Hartel, Lucifer Calar. p. 4, 16 quotiens mur- 
 muravit populus contra deum . . . legisti deum dixisse etc. ; 
 p. 268, 22 quotiens dicitur a nobis tibi non te potuisse dei unici 
 filii negatorem existere, etc.; p. 254, 24; 293, 15. Examples 
 might be multiplied, but it is more in point to view this historically 
 and to assign a cause for what might otherwise be thought largely 
 due to individual caprice or inaccuracy. The simple quotiens must 
 have been felt to have so general a distributive force that even a 
 better writer than Eugippius or Lucifer might use it where a 
 modern editor would demand a compound ; compare Cic. ad Att. 
 11, 13, 5 velim, ut soles, facias, quotiensque habebis, cui des ad 
 me litteras, nolim praetermittas with ad Fam. 16, 11, 3 cura ut 
 valeas litterasque ad me mittas, quotienscunque habebis, cui des. 
 Here Cicero was cither uncertain which form to use, a conclusion 
 contradicted by his other examples, or quotiens alone was suffi- 
 ciently indefinite where special emphasis was not intended; just 
 as the simple totiens is used throughout the language without a 
 single example of totienscunque, whether alone or in correlation 
 with quotiens or quotienscunque. 1 Other instances of parallel usage 
 are not hard to find. Cf. Varro, R. R. 3, 10, 7 quotienscunque 
 
 1 Obviously quotienseunque . . . totienscunque would have been too clumsy. Lewis 
 and Short cite taliscunque as air. etp., Auct. Priap. 16, but quale not qualecunque is used 
 with it. Cf. Wolff., Die Gemination im Lateinischen, Sitzungsberichte d. Bayr. 
 Akad. d. Wissensch. 1882, p. 463. 
 
On the Usage of Quotiens arid Quotienseunque. 15 
 
 sumpserunt, locus solet purgari and Petron. ch. 90 quotiens 
 theatrum . . . intravi, hac me adventicia excipere frequentia 
 solet; Valer. Max. 3, 6, 4 C. Duellius . . . quotienscunque 
 epulatus erat . . . domum reverti solitus est and Epit. Par., quo- 
 tiens p. e. . . . deducebatur; Celsus 6, 6, 16 quotienscunque non 
 adicio . . . aquam intellegi volo and id. 5, 19, 24 quae quotiens 
 adiciuntur cerussae vel spumae argenti, scire licet, etc. ; 7, 12, 1 
 ad fin. quotienscunque dente exempto radix relicta est . . . exi- 
 menda est and 6, 15 ad raed. quotiens autem medicamentum 
 inicitur, . . . os eluendum est ; 8, 10, 1 ad med. quotienscunque 
 autem solvitur membrum, calida aqua fovendum est and 8, 10, 2 
 ad fin. quotiens solutae sunt, fractura manu continenda ; Cyprian, 
 Corp. Eccl. p. 430, 18 ut quotienscunque inimicus accesserit . . . 
 clausum adversum se pectus inveniat and id. p. 430, 1 9 ut quotiens 
 adire temptaverit, clausum, etc. ; Ps.-Cypr. p. 30, 20 Sic quotiens- 
 cunque . . . distenta imbribus frumenta turgescunt, fecundae messes 
 coguntur aestate and p. 31, 2 Sic quotiens ferro vitis absciditur, 
 erumpentibus pampinis melius uva vestitur; August. Civ. Dei 11, 
 31 ad rned. quotienscunque ceciderit, non peribit and Sen. Epp. 
 2, 1, 2 quotiens cecidit, contumacior resurrexit; Serv. ad Eel. 1, 
 36 vocativus Graecus est, qui brevis est, quotienscunque nominativus 
 'is ' terminatur and id. -ZEn. 1, 108 genetivus enim pluralis quotiens 
 in 'ium' exit, accusativum pluralem in Ms' mittit; JEn. 1, 9 est 
 figura hypallage, quae fit quotienscunque per contrarium verba 
 intelleguntur and .2En. 1, 23 antonomasia est, non epitheton ; quae 
 fit quotiens pro proprio nomine ponitur quod potest esse cum 
 proprio nomine; JEn. 1, 73 est tropus systole, qui fit quotiens- 
 cunque longa corripitur syllaba, etc. and 1, 208 pleonasmos est, qui 
 fit quotiens adduntur superflua ; ^En. 1, 77 figura est litotes, quae 
 fit quotienscunque minus dicimus et plus significamus and 1, 435 
 est translatio, quae fit quotiens vel deest verborum proprietas, vel 
 vitatur iteratio; cf. further ^En. 1, 209; 228; 251 ; 3, 384. In 
 the last examples may be seen the cause already referred to, for 
 the simple relative with a present indicative in definitions, general 
 statements and conditions comes to be so extensively used that in 
 the later literature quotienscunque is overshadowed and practically 
 supplanted. See Part III for statistics. 
 
 To give a pronoun or a pronominal adverb a general meaning 
 
16 On the Usage of Quotiens and Quotienscunque. 
 
 the earlier language doubled the simple form. 1 So quis gives 
 quisquis (archaic quirquir, Varro, L.L. 7, 2, 8) and similarly we 
 have ututj ubiubi, undeunde, quotquot (though not early), quantus- 
 quantus, qualisqualis. Qualiterqualiter is cited for the Jurists, 
 Digest. IV, 4, 7, 1 ; ix, 2, 7, 1 ; xxvi, 7, 5, 10, 2 but is rare, 
 if it occurs at all, elsewhere on account of aversion to doubling a 
 word of three syllables. Quotiensquotiens, therefore, was never 
 formed. "Am meistens," Prof. Wolfflin states, 1.1. p. 449, " hat sich 
 die Sprache gegen die Verdoppelung der mehrsilbigen Pronominal- 
 formen gestraubt, ... so sehen wir, dass die lateinische Sprache 
 mit Ausnahme zweier oder dreier Casus sich der geminierten 
 Formen zu erwehren oder dereu Weiterbildung zu storen gesucht 
 hat, was natiirlich nur moglich war, wenn sie dafiir eine bessere 
 und deutlichere Ersatzbildung bieten konnte. Eine solche fand sie 
 in cunque, in welchem cum=quom temporal in Sinne von 'wann, 
 jedesmal wann, immer ? zu verstehen ist ; also quicunque = wer 
 immer. In que aber erkennen wir denselben wiederholten, nur 
 abgeschwachten und unflectierten Pronoruinalstarnm." Instead 
 then of the double forms we have utcunque, ubicunque, undecunque, 
 quotcunque (as old as *Catull. 64, 280), quantuACunque, qualiscunque ; 
 cf. also quisque (= quisquis) and quandoque (= quandocunque). 
 Quandoque in this sense is found as early as the Tabb. xn, though 
 in general examples are comparatively rare. 3 A formulaic exam- 
 ple is found in Cic. Caecin. 54 quandoque te in iure conspicio, but 
 this is weakened by the parallel passage, Mur. 26 quando te in 
 iure conspicio. A clear example, however, is de Ke Pub. 6, 22, 
 24; and aside from the poetic passages Hor. C. 4, 1, 17; 2, 34 ; 
 A.P. 359, 4 the same abbreviation may be found in Livy 1, 31, 4; 
 29, 10, 5 ; 39, 24, 1 ; Col. 2, 10, 6 ; 7, 4, 7 ; 11, 3, 22 ; Juv. 2, 82 ; 
 5, 172; 14, 51 ; Lactant. Inst. p. 235, 14 (Brandt), (cf. quando- 
 cunque p. 235, 6); Orosius p. 152, 1 (Zang.) (quandocunque p. 
 366, 1). 
 
 It is evidently after the analogy of quandoque that quotiensque 
 (= quotienscunque), cited below, was formed. This compound 
 
 1 Wolfflin, Gemination, p. 422 ff., ALL. 1.1. 
 
 * Dirksen, Man. Latinitatis Fontium Juris Civilis, 797. 
 
 * Schmalz, Lat. Synt. 3 % 305. 
 
 4 Quandocunque occurs, Epp. 1, 14, 17 ; 16, 58. 
 
On the Usage of Quotiens and Quotienscunque. 17 
 
 has scant recognition in the lexicons. The citations are limited 
 to Col. 6, 17, 9 (Quidam Icgunt Forcell.) : sed quotiensque mel 
 aliusve succus remediis adhibetur. No earlier example occurs, 
 since the Augustan writers employ quotiensque invariably in the 
 sense of et quotiens. Cf. Ovid, A. A. 2, 327 
 
 Multa vove, sed cuncta palam. quotiensque libebit, 
 Quae referas illi somnia laeta vide. 
 
 Met. 3, 495 
 
 Quae tamen ut vidit, quamvis irata memorque, 
 Indoluit, quotiensque puer miserabilis ' Eheu ! ' 
 Dixerat, haec resonis iterabat vocibus, Eheu. 
 
 Met. 4, 588 ; 10, 164; Amor. 2, 8, 27; Manil. 1, 513; Cic. Cat. 
 3, 5, 11 ; ad Att. 11, 13, 5; and later in Valer. Flacc. 8, 364; 
 Sil. Ital. 12, 145; Curt. 4, 10, 6; Plin. Pan. 88; Gell. 14, 7, 4. 
 
 The apparatus criticus of Schneider's edition justifies other 
 examples of quotiensque in Columella, though neither Schneider 
 nor Gesner reads this form. 1 Cf. 2, 4, 10 Paulum tamen, quo- 
 tiensque (I. B. Br., quotiesque, ex Pol. excerpt.) iterabitur .... 
 obliquum agi sulcum oportebit ; 12, 51, 2 idque fieri debebit, 
 quotiensque (I. B. Br. R.) ; 52, 16 eamque suffitionem semper faci- 
 endam iudicaut, quotiensque (Br.) vel nova vel vetera vasa curantur. 2 
 
 Suetonius and Ulpian also furnish examples: cf. Aug. 37 
 excogitavit .... triumviratum .... recognoscendi turmas 
 equitum, quotiensque opus esset, and Calig. 7 cuius effigiem .... 
 in cubiculo positam, quotiensque introiret, exosculabatur. Two 
 examples in the Digest have been recognized by Kriegel and 
 Mommsen. In Ulp. Dig. VII, 4, 3 praef. quotiensque capite 
 minutus erit, quoties quis was read before Kriegel ; ibid. XLIII, 
 8, 2, 10 quotiensque aliquid in publico fieri permittitur is due 
 to Mommsen, since quotienscunque is read by Kriegel following 
 the corrector of the codex Florentinus. Silvia, Peregrin. 50 nam 
 consuetudo talis erat in illo tempore, ut palatia, quotiensque 
 
 1 The promised critical edition of Lundstrom has not, unfortunately, been seen. 
 
 2 Observe that Schneider in 2, 10, 16 ; 7, 4, 7; 11, 3, 22, accepts the better 
 known quandogue on the same authority. 
 
18 On the Usage of Quotiens and Quotienscunque. 
 
 fabricabantur, semper in editioribus locis fierent is a more certain 
 example, though quotienscunque is read by Cholodniak, and Geyer 
 has questioned the reading. 1 Quotiensque with good MS. authority, 
 especially in late Latin, has doubtless in many cases been quietly 
 corrected (!) by an editor unfamiliar with the abbreviated compound. 
 So in Ruricius, p. 382, 13, Engelbrecht reads quotienscunque 
 sanctos viros .... aerumnarum mole depressi coguntur expetere 
 following the lectio vulgata, while the Codex Sangallensis (Sc. ix) 
 preserves the correct quotiensque? Cf. similarly Ammianus Marc. 
 16, 10, 18 ut quotienscunque (quotiensque, Vat.) concepisset, 
 immaturum abiceret partum. 
 
 Along with the method of composition already given, -vis and 
 -libet were used to give similar effect, as in quivis, 3 quilibet, undelibet, 
 quotlibet, etc. This formation, however, is not so freely used, for 
 while quilibet is common from Plaut. on, quotlibet is cited first 
 from Hygin. Astron. 1, 6, 4 and quandolibet (= quandoque = ali- 
 quondo) occurs first in Lactant. Opific. Dei p. 15, 20 (Brandt) necesse 
 est igitur ut mortem recipiat quandolibet, quoniam corporalis est. 
 Cf. ibid, p. 16, 18 qui quandoque moriturus est and Inst. p. 151, 5. 
 Quotienslibet is also rare and is not cited by Georges earlier than 
 Boeth. Mus. 1, 4 (p. 192, 2 Fried.) where it is used as an indefi- 
 nite : cum maior numerus habet in se minorem vel bis vel ter vel 
 quotienslibet. But other examples not cited by Georges occur in 
 Boeth., e.g. Arith. 1, 14 (p. 31, 5 Fried.) Metitur numerus nu- 
 merum, quotiens vel semel vel bis vel tertio vel quotienslibet 
 numerus ad numerum comparatus .... ad comparati numeri 
 terminum usque pervenerit; 1, 29 (p. 61, 4) autduplum aut triplum 
 aut quadruplum aut quotienslibet ; and the form is not only much 
 earlier, but it occurs as a relative, potentially at least, in Celsus 4, 
 23 non quotienslibet desidere, sed quotiens necesse est. Cf. again 
 Rusticius, contra Acephalos, Patrol. LXVII, Col. 1190 (Migne) idem 
 
 1 Cf. Krit. Bemerkungen z. S. Silviae Peregrin. (St. Anna Program), Augsburg, 
 1890, p. 34: quotiensque (lies quotienscunque nach S. 67, 6: quotienscunque). 
 
 Also read in the text of Krusch, Mon. Oer. Hist, vm, Berol. 1887. 
 
 3 Even quiviscunque, Lucret. 3, 388 ; Mart. 14, 2, 1. Similarly quisquislibet of 
 late Latin, cf. Salv. adv. Avar. 3, 18 ; Avitus, Mon. Ger.p. 28, 13 (Peip.) ; 39, 10. 
 
 4 Cited by Lewis and Short as for. tip, but cf. Boeth. Arith. 1, 31 (p. 65, 5 
 Fried.) ; 2, 2 (p. 80, 9) ; 2, 43 (p. 140, 20) ; Mus. 2, 8 (p. 236, 7 ; 8) ; passim. 
 
On the Usage of Quotiens and Qaotienscunque. 19 
 
 namque est, quotienslibet quis convertat verbositate sermones, and 
 the indefinite use, Arnmian. Marc. 21, 5, 3 vigore corumuni 
 Romanis agminibus quotienslibet Rhenum pervium feci. The 
 combination quotiensque (= et quotiens) libebit is found in Ovid, 
 A. A. 2, 327. 
 
 In a similar way a quotiensvis might be assumed from such com- 
 binations as quotiens velles, Sen. Controv. 4, praef. 7 ; quotiens 
 velis, Sen. Epp. 6, 3, 9 ; quotiens voles, ibid. 4, 6, 4 ; de Ben. 4, 
 
 7, 1 ; Vicies septies dices et quotiens volueris repetes, Marcell. 
 p. Ill, 28 (Helmr.), etc. This form, however, does not occur. 
 Quandovis in our analogy is also lacking. 
 
 An interesting combination which points to the easy language 
 of every day life is seen in nescio quotiens, after the analogy of 
 nescio quando (cf. Cic. Philipp. 2, 3), or the earlier and more com- 
 mon nescio quomodo. That nescio quotiens was of frequent use is 
 shown by its occurrence in the Notae Tironianae (Schmitz, Com- 
 ment. 10, 31) along with nescio quomodo. The extant literature 
 oddly enough shows, as far as I have noted, but one example, 
 August. Civ. Dei 12, 18 init. horainem nescio quibus circnitibus, 
 nescio quotiens revolutum. 
 
 Further comparison with quando furnishes the following parallels ; 
 quando si quando quando umquam 2 quandocunque 
 
 quotiens si quotiens 1 quotiens saepe z quotienscunque 
 
 quandoque qaandolibet nescio quando 
 
 quotiensque quotienslibet nescio quotiens. 
 
 MEANING. 
 
 The strict meaning of quotiens, " how many times," may be seen 
 e.g. in Plaut. Cure. 608 Dixi tibi . . .: quotiens dicendumst? 
 In the Nov. Test, it is accordingly used to render 7ro<ra/a? (Matth. 
 18, 21 ; 23, 27 ; Luc. 13, 24) or ocra/a?, as in the Lord's Supper, 
 
 1 After si for aliquotiens; but three instances have been noted, Ovid, Trist. 
 5, 10, 42; Sulp. Sever, p. 131, 5 (Halm) ; 204, 4. Cf. nisi quotiens Serv'. ad ^En. 
 
 8, 268; Sid. Apoll. 7, 14, 11. 
 
 'For emphasis after si, but only in Livy, e.g. 6, 42, 12; 8, 4, 6 (Weissenb. 
 adloc.); 10, 14, 11. 
 
 3 Evidently a cross between quotiens and its equivalent quam saepe (cf. infra). 
 Only one example was noted, Sid. Apoll. 1, 7, 2. 
 
20 On the Usage of Quotiens and Quotienscunque. 
 
 1 Corinth. 11, 25 quotienscunque bibetis (OCTCLKL^ eav 
 Apocal. 11, 6 quotienscunque voluerint (oo-a/a? eav 6e\r)<Tu>(Ti}. 
 Compare Prise. Ill (K.) 132, 8 adverbiura, interrogativum et 
 relativum quotiens, TTOO-CLKL^ KCLI o<ra/a?. 
 
 In the freer sense " how often " the early Latin had already 
 used. quam saepe, possibly before the formation of quotiens. Com- 
 pare for analogy quam diu. Quam saepe is found first in Ter. 
 Phorm. 757 quam saepe forte temere Eveniunt quae non audeas 
 optare, without comment by Donatus. The Auctor ad Heren. 
 3, 12, 21 follows with quam (quae codd.) saepe rerum naturae 
 gratia quaedam hire debetur. Cicero confines his few examples to 
 the Philosophical works and the Letters ; cf. Tusc. 2, 41 ; de 
 Div. 1, 97 ; 2, 62, and ad Att. 10, 9, 1 ; ad Fam. 7, 10, 3, where 
 the parallels quam insulsi and quam diligenter should be noted. 
 Similarly tarn saepe (= totiens, found as early as Plaut. Merc. 33) 
 is parallel with tarn vehementer, ad Att. 3, 10, 2, though totiens et 
 tarn longe occurs in 4, 16, 7. Compare a similar chance for sub- 
 stitution in quotiens et quam diu, Sen. Controv. 4 Praef. 7 and 
 quam diu et quotiens, Apul. Apolog. 73. In the orations Cic. 
 shifts to quam crebro, Mil. 69; Pliilipp. 2, 70, as Catull. changes 
 to ut saepe, 66, 30. Plautus had used the latter expression once, 
 Capt. 165. 
 
 Csesar, B. C. 3, 72, 4 quam parvulae saepe causae rnagna detri- 
 menta intulissent may fairly be cited since the quam saepe and 
 parvulae causae are separated by traiectio, as is the phrase in Cic. 
 de Div. 2, 62. 
 
 In the Augustan age examples are still to be cited : Tibull. 
 1, 6, 21 ; 8, 53; Ovid, Am. 1, 14, 47. Livy 34, 20, 6 memores, 
 quam saepe in agro eorum irnpune persultassent, quotiens ipsos 
 signis collatis fudissent is noteworthy especially since quam saepe 
 precedes. Cf. further Quintil. 6, 3, 77 apparet, inquit, quam 
 saepe accendatis (quoting Augustus) ; again in Juv. 9, 73, and in 
 late Latin Boeth. Mus. 1, 1 (p. 184, 7 Fried.) An interesting 
 blending of quotiens and quam saepe through the carelessness of 
 epistolary style is seen in Sid. Apoll. Epp. 1, 7, 2 O quotiens 
 saepe ipse se ad versa perpessum gloriabatur ! 
 
 A complete list of quam saepe is not offered, but a large number 
 of representative authors of various periods has been examined, 
 
On the Usage c>f Quotiens and Quotienscunque. 21 
 
 and the search shows that only sporadic examples of substitution 
 occur from Plautus on, even as late as Boethius, while their limited 
 number affords no absolute olue to their sphere. Quotiens and 
 totiens practically displaced these combinations early in the lan- 
 guage, as examples from Plautus show, and, aside from the pos- 
 sibilities in the case, the " Konkurrenz " is not striking. 
 
 Neither quotiens nor quotienscunque per se necessarily expresses 
 a large number of times. Cf. for example Alfen. Var. Digest, 
 xxxin, 1, 22 Filiae meae, quotienscunque vidua erit, in annos 
 singulos centum heres meus dato. Accordingly, it is unnecessary 
 with Wex 1 and Peerlkamp 2 to demand si quando in Tac. Agric. 
 1 clarorum virorum facta moresque posteris tradere, antiquitus 
 usitatum, ne nostris quidem teraporibus .... aetas omisit, 
 quotiens magna aliqua ac nobilis virtus vicit ac supergressa est 
 . . . . ignorantiam recti et invidiam. Peerlkamp, following 
 Wex, remarks : virtus tarn rara postulabat, ni fallor, si quando. 
 Restringe significationem adverbii quotiens, semper maiorem nu- 
 merum indicat quam in tempora praesertim Neronis et Domitiani 
 cadit. But there are in reality two apodoses in the first clause, 
 the early practice (antiquitus usitatum) and its survival (ne nostris 
 quidem temporibus, etc.), to each of which in the author's thought 
 quotiens belongs. A correct interpretation is given by Pohlmann, 3 
 though disputed by Planck. 4 
 
 Quotiens expresses only iteration, not frequency, unless the con- 
 text makes these two synonomous. Cf. Marcell. p. 376, 17 (Helmr.) 
 et die bis aut ter vel quotiens siccari coeperit mutanda erit; Plin. 
 Epp. 6, 2, 7 equidem quotiens iudico, quod vel saepius facio, quam 
 dico, quantum quis plurimum postnlat aquae do. Similarly Cic. 
 ad Fam. 6, 5, 1 quotienscunque filiurn tuum video (video autem 
 fere quotidie) ; ibid. 13, 41, 1 turn Pompeius, quotienscunque me 
 videt (videt autem saepe), while in 13, 69, 1 Apud eum ego sic 
 Ephesi fui, quotienscunque fui, tamquam domi meae even the 
 opposite of frequency seems to be denoted. 
 
 The idea of frequency may be added by the use of a frequenta- 
 
 1 Agricola, Brunsvigae, 1852, Prolegom. p. 164. 
 
 'Agricola, 1864, ad loc. 
 
 'Annotations in Corn. Taciti Agricolam, Gottingae, 1871, p. 18. 
 
 4 Zur Erklarung d. Tac. Agrik. 1874, p. 7. 
 
 
 
22 On the Usage of Quotiens and Quotienscunque. 
 
 live verb, but classical examples are wanting and in the later ones 
 the frequentative force has often faded. Cf. Tac. Ann. 6, 21, 1 
 Quotiens super tali negotio consultaret, edita domus parte . . . 
 utebatur; 13, 18, 16 quotiens ipse illuc ventitaret ; Ammian. Marc. 
 20, 11, 9; Cypr. Gall. Exod. 769; 991; Ennod. C. 2, 121, 1. 
 That quotiens should often be found where cum, quando, ubi, 
 utcunque, etc., could be used and v. v. was to be expected in a live 
 language. Compare Paul us, Digest. I, 5, 7 qui in utero est . . . 
 quotiens (=cum, quando) de commodis ipsius partus quaeritur 
 with id. L, 16, 231 tune verum est, cum de ipsius iure quaeritur. 
 Often these are parallel in a single passage to give variety. Cf. 
 Valer. Flacc. 6, 683 
 
 At quotiens vis dura ducum densique repente 
 Aesoniden pressere viri cumque omnis in unum 
 imber iit, totiens saxis pulsatur et hastis ; 
 
 Plin. N. H. 18, 323 quotiens at occidente sole cernetur . . . 
 crescens erit . . ., cum vero occidente sole orieturet adverse . . . turn 
 erit plenilunium; Ulp. Digest. L, 16, 38 duo genera autem sunt 
 ostentorum : unum, quotiens quid contra naturam nascitur; alte- 
 rum, cum quid prodigiosum videtur ; Diom. I (K.) 309, 1 quintus, 
 quotiens in o exeunt femina (cf. with tertius, cum, supra) ; Serv. 
 on JEn. 9, 192 poscere est secundum Yarronem quotiens aliquid 
 pro merito nostro deposcimus, petere vero est cum aliquid humiliter 
 et cum precibus postulamus. In Marcell. p. 311, 22 (Helmr.) 
 hoc . . . remedium ... ad alios trauslatnm, cum volueris et 
 quotiens volueris, proderit each word has its proper force. 
 
 On the other hand, compare e. g. Porph. on Hor. Epod. 17, 52 
 utcunque autem * quotienscunque ' significat; Serv. -ZEn. 1, 412 
 figura est tmesis, quae fit cum secto uno sermone aliquid interponi- 
 mus (^En. 1, 77 figura est litotes, quae fit quotienscunque, etc., 
 .ZEn. 1, 435 est translatio, quae fit quotiens, etc.) ; and compare 
 further Serv. ^ZEn. 1, 119 cum provinciam dicimus . . . et princi- 
 pale est nomen, brevis est ' Tro/ quando autem non est principale 
 et derivatio est, longa est 'Tro' with ibid. 3, 384 ut quotienscunque 
 Sicilia significatur, Trinacria totum per * r y dicatur, quotiens 
 Sicula, per unam 'r.' In 1, 41 cum and quotiens are again 
 
On the Usage of Quotiens and Quotienscunque. 23 
 
 parallel : media producitur syllaba, sed cum opus est corripitur 
 hac excusatione : nam quotiens vocalem longam vocalis sequitur, 
 ei vires detrahit. 
 
 Note. The phrase cum opus erit (fuerit) is especially common 
 in the directions of Marcell. Cf. p. 288, 27 (Helmr.) ; 362, 17; 
 364, 5 ; 1 7 ; 369, 13 ; 375, 22 ; 376, 36 ; 379, 31, though quotiens 
 opus fuerit occurs p. 149, 26; 281, 18-19 (cf. 281, 14; q. fuerit 
 necessarium). The phrase is frequent in other writers, cf. q. opus 
 est, Ovid, A. A. 1, 430 ; 3, 641 ; Luc. 6, 560 ; Sen. de Ira 1, 9, 1 ; 
 de Ben. 2, 34, 2 ; Frontin. Aq. 14 (cf. q. necesse est Sen. Phoeniss. 
 493) ; q. opus fuerit, Curt. 10, 6, 15 ; q. o. sit, Plin. N. H. 17, 129 ; 
 34, 122; q. o. esset, Sueton. Galba 10 ; quotienscunque opus erit, 
 Cato, Agri Cult. 151, 4; Cic. de Orat. 3, 123; quotienscunque 
 opus fuerit, Vitruv. 5, 1 2, 4 ; quotiensque opus esset, Sueton. Aug. 37. 
 
 In the Jurists Kalb 1 notes that for a certain period, especially 
 in Africanus, Gaius and Marcell us, who belong to the reigns of 
 Antoninus Pius and Marcus Aurelius, quotiens is given up and 
 cum and other substitutes are used. Cf. for example Gaius, Inst. 
 4, 3 in rem actio est, cum . . . rem intendimus nostram esse with 
 Ulp. Digest. II, 14, 7, 8 (pacta) in rem sunt, quotiens general iter 
 pacisoor ne petam. Gaius, however, shows two examples 1, 162; 
 4, 2, though the text is uncertain in the latter. Kalb offers no 
 explanation (deren Gruud wir uicht finden konnten), but as the 
 later Jurists make frequent use of quotiens for si and cum, it is 
 evident that they were under the influence of the same loose usage 
 cited above, while Gaius writing his own lectures made nicer 
 distinction. 2 
 
 AUTHORS EXAMINED. 3 
 
 Plautus, Ennius, Terence, Cato, Lucilius, Auct. ad Heren., Lucre- 
 tius, Catullus, Csesar, Hirtius, Nepos, Cicero, Publilius Syrus, Sal- 
 
 1 Roms Juristen nach ihrer Sprache dargestellt, Leipz. 1890, p. 31. 
 
 2 Cf. the suggestion of Prof. Wolff., abstract, p. 398. 
 
 3 Editions are regularly indicated in the citations that are made by page and 
 line, and in other cases where pertinent variants are found. The Teubner texts 
 are usually taken as the basis. The grammarians are naturally cited from Keil's 
 Corpus Grammaticorum and patristic writers from the Corpus Vindobonense, 
 unless otherwise indicated. 
 
24 On the Usage of Quotiens and Quotienscunque. 
 
 lust, Varro, Vergil, Tibullus, Propertius, Livy, Hyginus,Vitruvius, 
 Horace (Porph.), Auct. Bell. Afr., Manillas, Seneca Rhetor, Ovid, 
 Valerius Maxiruus (Epit. Par.), Veil. Paterculus, Phaedrus, Mela, 
 Columella, Curtius, Celsus, Scribonius Largus, Petronius, Calp. 
 Siculus, Nemesianus, Persius, Probus, Lucan, Seneca Phil., Valer. 
 Flaccus, Plin. Mai., Quintilian, Pseud. -Quintil., Statins, Hygiu. 
 Grom., Caper, Vel. Longus, Silius Italicus, Frontinus, Martial, 
 Plin. Min., Tacitus, Juvenal, Florus, Terent. Scaurus, Justinus, 
 Suetonius, Apuleius, Fronto, Gellius, Gaius, Min. Felix, Ampelius, 
 Licinianus, Tertullian (Vol. I Reiff.-Wiss.), Censorinus, Com- 
 ruodianus, Cyprian, Solinus, Jul. Valerius, Terent. Maurus, Arno- 
 bius, Plotius, Publil. Optatianus, Palladius, Atil. Fortunatianus, 
 Scriptores Hist. Aug., Lactantius, Juvencus, Firm. Maternus, 
 Plin. (Medicus), Mar. Victorinus, Aurel. Victor, Obsequens, 
 Charisius, Diornedes, Eutropius, Dictys Cretensis, Vegetius, Au- 
 sonius, St. Silvia, Auctor Queroli, Prudentius, Amm. Marcellinus, 
 Macrobius, Marcellus Empir., Orosius, Rufinus, Symmachus, 
 Hieronymus, Mall. Theodorus, Servius (Commentaries on Verg.), 
 Claud. Claudianus, Pelagonius, Sulpicius Severus, Mart. Capella, 
 Augustinus (C. D. only), Salvianus, Cassius Felix, Sedulius, Claud. 
 Mamertus, Faustus Reiensis, Ruricius, Lucifer Cal., Sid. Apol- 
 linaris, Agroecius, Victor Vitensis, Dares Phrygius, Priscian, 
 Engippius (Vit. S. Sever.), Boethius, Ennodius, Fulgentius, Di- 
 gesta Justiniani, Jordanis, Cyprian. Gall., Cassiodor(i)us, Gregory 
 of Tours. 
 
V 
 
 On the Usage of Quotiens and Quotienscunque. 25 
 
 PART II. 
 
 CITATION OF EXAMPLES. 1 
 
 QUOTIENS. 
 
 A. EXCLAMATORY : INDICATIVE, Present Cic. ad Att. 
 7, 26, 1; TIBUM,. 2, 3, 17; PROPERT. 1, 10, 4; 18, 21; 
 OVID, A. A. 2, 447; 567; Trist. 3, 12, 25; SEN. Here. O. 
 93 ; SIL. ITAL. 13, 667 ; MART. 10, 51, 15 ; CLAUDIAN. de Rapt. 
 Proserp. 3, 126; 127; ENNOD. p. 15, 19 (Hartel). Imperfect 
 TEB. Hec. 60; SEN. Here. O. 1914; Sit. ITAI-. 6, 584; SID. 
 APOIX. Epp. 1, 7, 2. Future PBOPERT. 1, 5, 13 ; SEN. Controv. 
 9, 5, 1 ; OVID, Her. 15, 81 ; SIL. ITAL. 13, 866 ; MART. 11, 16, 5 ; 
 CLAUDIAN. de Nupt. Honor. 4, 14 ; de Consul. Stil. 2, 398 ; 404. 
 Perfect CATO, in Sulpic. p. 62, 1 (Jordan); Cic. Mil. 5, 12; 
 
 7, 20; 14, 38; 15, 41; Cat. 1, 6, 15 (bis); 16 (bis); Phil. 
 2, 18, 45 (bis); Marc. 6, 16; Dom. 23, 59; Font. 7, 16; Arch. 
 
 8, 18 (ter); Tusc. 1, 89; Div. 1, 97; Cat. Mai. 49 (bis) ; VERG. 
 Eel. 3, 72 ; Ge. 1, 471 ; Ciris 81 ; 82 ; TIBDUL. 1, 3, 19 ; 9, 17 ; 
 41 ; 2, 3, 19 ; 6, 13 ; PROPERT. 3, 31, 11 ; 4, 14, 13 ; 15 ; 5, 7, 17 ; 
 LIVY 7, 29, 2; 39, 16, 8; 40, 8, 11; SEN. Controv. 9, 6, 20; 
 OVID, Her. 5, 49; 51; 9, 79; 15, 135; 137; 16, 81 (bis); Am. 
 1, 13, 27; 29; 2, 19, 11; 13; 3, 1, 53; A. A. 1, 313; 321; 
 2, 125 ; 3, 481 ; Nux 160 ; 161 ; Met. 2, 489 ; 491 ; 3, 375 ; 427 ; 
 428; 10, 661; 14, 643; 15, 490; 492; Trist. 1, 3, 51; 53; 
 ex Pont. 1, 1, 7; 1, 9, 21; 23; 4, 1, 9; 11; VAIER. MAX. 
 
 9, 3, 5; CURT. 10, 5, 35 (bis); CAM>. Sic. Eel. 7, 69; Luc. 
 5, 615; SEN. Epp. 10, 2, 14 (bis); de Rem. Fort. 16, 5; 
 Here. F. 21; Here. O. 1912; VALER. FLACC. 1,447; 2, 53; 4, 
 
 1 Authors are arranged in an approximately chronological order, with certain 
 exceptions (e. g. the Jurists) for convenience in grouping. Convenience is fol- 
 lowed also in designating but few pseudos. 
 
 Tenses are contemporary unless otherwise indicated in the classification. 
 
26 On the Usage of Quotiens and Quotienscunque. 
 
 378; PSEFD.-QUINTIL. p. 281, 12 (Ritter); 352, 16; STAT. Silv. 
 1, 2, 93; 5, 5, 38; Theb. 5, 615; Sir,. ITAI,. 7, 152; MART. 9, 84, 7 ; 
 14, 115, 2; 119, 2; PJLIN. Pan. 15 (bis); SUETON. de Vir. 111. 
 p. 44, 6 (Reiff.) ; FI,AV. VOPISC. Prob. 12, 6 ; PRUD. Psych. 899 
 (bis); CI<AUDIAN. in Eutrop. 1, 35; Epith. de Nupt. Honor. 8 1 ; 
 Pan. Manl. Theod. 54; de Consul. Stil. 1, 122; 233; 3, 184; 
 185; de Rapt. Proserp. 1, 192; 193; AUGUST. Civ. Dei 3, 18 
 init. ; 19 init. ; SALVIAN. p. 39,11 (Pauly); SID. APOLI,. Carm. 
 23, 233 ; ENNOD. Carra. 1, 8, 39 ; 41 ; 2, 140, 1 ; BOETH. Consol. 
 
 1, 4, 29 (p. 11, 29 Peip.); 31; 32; GREG. TUR. Hist. Franc. 5 
 prol. init. 
 
 INFINITIVE, Perfect L.IVY 25, 16, 12; 30, 21, 7; TAC. Ann. 
 3, 34, 27. 
 
 B. INTERROGATIVE : INDICATIVE, Present PI.AUT. 
 Amph. 619; Cure. 609; Most. 948; SERV. Sctp. apud Cic. ad 
 Fam. 4, 5, 3; L.IVY 34, 12, 8; SEN. Controv. 10, 4, 22; OVID, 
 Trist. 5, 6, 25 ; L.UC. 8, 408 ; SEN. Dial. 10, 5, 1 ; Nat. Quaest. 3, 
 17, 2 ; STAT. Theb. 2, 337 ; HYGIN, GROM. 36 ; TERTUIX. p. 395, 
 1 (Reiff.-Wiss. 1); 3; AUSON. p. 120, 43 (Peiper). Imperfect 
 VARRO, Menipp. 186 (Buech.) ; SEN. Here. O. 212. Future- 
 Cm. Verr. 2, 59, 145; HOR. Carm. 1, 5, 5; OVID, Her. 16, 
 217. Perfect PLAUT. Asin. 522; Men. 784; 788; Mil. Glor. 
 1057 ; Cic. ad Att. 12, 6, 2 ; MANIL. 1, 509 ; SEN. Controv. 2, 7, 
 5 (bis); SEN. Epp. 14, 3, 9 (quater) ; 19, 1, 3 (bis); MART. 11, 
 43, 3; FRONTO, ad Ver. Imp. 2, 8; CYPR. p. 44, 15 (Hartel) ; 
 App. 120, 17; Carm. 6, 278; 279; ARNOB. p. 271, 16 (Reiff.); 
 AMM. MARC. 20, 5, 4 ; SYM. Epp. 10, 9, 6 ; ENNOD. p. 9, 22 ; 
 310, 8; 9; 364, 6; 7; 372, 2; 3; 22; 395, 8; 426, 26; 27; 
 470, 21; 22; 26; 479, 27; 491, 13; 17. Pluperfect ENNOD. 
 p. 480, 1. 
 
 SUBJUNCTIVE, Present Cic. Div. in Caecil. 14, 45; LIVY 
 5, 52, 9; SEN. Controv. 7, 8, 8 ; OVID, Her. 18, 31; CELS. 
 
 2, 14 fin.; AUCT. AETNAE 88; PUN. Nat. Hist. 22, 107; 
 STAT. Theb. 7, 201 ; MART. 9, 35, 7 ; Juv. 3, 270 ; APUL. Apol. 
 38 ; TERENT. MAUR. VI (K.) Met. 2873 ; MAR. VICTORIN. VI (K.) 
 72, 17 ; AUSON. p. 180, 10 (Peiper). Imperfect Jut. CAP. Opil. 
 
 1 Parallel with quam saepe ; cf. Part I, p. 20 f. 
 
On the Usage of Quotiens and Quotienscunque. 27 
 
 Macr. 3, 1. Perfect. CATO, Orig. 4, 1, (bis) ; JLIVY 26, 13, 6 ; 34, 5, 
 3; 35,18,7; 38,47, 12 (bis); MANIL. 1,513; SEN. Controv. 2, 1 , 8 ; 
 OVID, Am. 2, 8, 27 ; 3, 8, 21 ; Met. 7, 734 ; 735 ; VEM,. PATERC. 
 
 2, 42, 2 ; SEN. Dial. 4, 28, 6 ; 5, 25, 2 ; PSEUD-QUINTS p. 
 16, 12 (Ritter); STAT. Theb. 8, 510; TAC. Hist. 4, 58, 9; Ann. 
 
 3, 6, 10; APUI-. Apol. 73; TERTUUL. p. 61, 4 (Reiff.-Wiss. 1); 
 TERENT. MAUR. VI (K.) Met. 2668; Jui,. CAP. Opil. Macr. 
 1, 4; ARNOB. p. 6, 26; LACT. p. 589, 2 (Laub.-Brandt 1); 3; 
 FI.AV. VOPISC. Prob. 6, 1; MAR. VICTORIN. VI (K.) 161, 13; 
 AUCT. QUEROI,. p. 11, 4 (Peiper); CLAUDIAN. Cons. Olyb. et 
 Prob. 58; ENNOD. p. 364, 2. Pluperfect Caes. B. G. 1, 43, 7; 
 B. C. 3, 72, 4 ; Cic. Cat. 3, 5, 11 ; SAIX. Hist. Fragg. 3, 48, 1 
 (Maur.); LIVY 23, 15, 12; 34, 20, 6; 35, 33, 4; PHAED. App. 
 
 9, 27, 6; JUSTIN. 13, 1, 3; TERTUM,. p. 61, 16 (Reiff.-Wiss. 1). 
 INFINITIVE, Perfect L.IVY 33, 12, 8. 
 
 C. RELATIVE: INDICATIVE, Present PI.AUT. Epid. 175; 
 Men. 114; True. 196; Lucir. 9, 34 (M. = 911 Baehr.) ; Cic. 
 Orat. Ill; Fin. 5, 63; VERG. ^En. 4, 351; 352; VITRUT. 
 
 5, 8, 2 ; HOR. Carm. 4, 2, 26 ; Epp. 1,18, 104 ; 2, 1, 55 ; (PORPH. 
 Sat. 2, 1, 22; Epp. 1, 18, 104); MANII,. 1, 700; SEN. Controv. 
 1, 2, 15; 2, 5, 11; 3, 5 fin.; OVID, Her. 12, 190; 18, 180; 19, 
 190 ; A. A. 1, 430 ; 2, 613 ; 3, 641 ; Met. 4, 587 ; 6, 481 ; Fast. 
 
 1, 573 ; Trist. 1, 7, 9 ; 2, 1, 33 ; 4, 7, 7 ; ex Pont. 1, 2, 122 ; 2, 1, 
 13 ; 9, 16 ; 3, 1, 61 ; Cot. 4, 32, 5 ; 12, 12, 1 ; CEI.S. 1, 9 init. ; 
 
 2, 16 fin.; 2, 17 init.; 4, 26 fin.; 31 med. ; 5, 19, 24; 26, 24; 
 
 6, 15 med.; 18, 2 fin.; 7, 2 med. ; C ALP. Sic. 2, 85; 5, 53; 
 
 7, 21; Luc. 1, 256; 3, 469; 549; 6, 560; 9, 404; SEN. Epp. 
 2, 7,1; 4, 5,1; 6, 4, 8; 7, 3, 17; 13, 2,4; 18, 2, 8; 5, 24; 
 20, 5, 14; Dial 1, -4, 5 ; 3, 9, 1 ; de Clem. 1, 12, 1; de 
 Ben. 2, 10, 4; 31, 2; 34, 2; 7, 23, 1; Nat. Quaest. 1, 5, 6 ; 
 12, 1 ; 2, 58, 1 ; 3 praef. 4; 3, 30, 3; 6, 6, 4; 9, 3; VAI.ER. FLACC. 
 <8, 364; PWN. Nat. Hist. 17, 104; 18, 277; 22, 123; QUINTII,. 
 Inst. Orat. 1, 6, 29 ; 2, 16, 10 ; 3, 6, 27 ; 95 ; 4, 1, 46 ; 2, 41 ; 
 5, 26; 6,2, 14; 7, 8, 7; 8,3, 56; 82; 5,21; 6, 59; 9, 2, 7; 
 27; 34; 3, 75; 76; 4,40; 67; 10,3,27; 11,3,102; 12, 1,29; 
 
 10, 27; 29; PSEUD. -QUINT, p. 15, 18 (Ritter); 40, 7; 75, 13; 
 S9, 18; 90, 24; 26; 124, 12; 141, 16; 17; 210, 17; 228, 8; 
 259, 7; 326, 5; 358, 28; STAT. Silv. 3 praef.; 2, 37; Theb. 
 
28 On the Usage of Quotiens and Quotienscunque. 
 
 6, 11 ; 685 ; HYGIN. GROM. 2 ; 40 ; 43 ; 52 ; VEI LONG. VII (K.) 
 60, 17 ; 63, 6 ; 66, 4 ; 72, 2 ; 81,7; Sir,. ITAL. 1, 266 ; 3, 410 ; 
 12, 145; FRONTIN. Strut. 1, 11, 17; Aq. 11; 14; 15; 28; 35 
 (bis) 1 ; 112 (bis); MART. 1, 117, 1 ; 2, 14, 2 ; 5, 18, 9 ; 50, 1 ; 
 52, 3 ; 6, 36, 2 ; 93, 8 ; 7, 35, 2 ; 8, 55, 2 ; 76, 4 ; 10, 80, 1 ; 
 11, 7, 13; 99, 1; 102, 5; 12, 56, 3; 13, 70, 1; 101, 2; PLIN, 
 Epp. 1, 22, 2; 6, 2, 7; 9, 36, 2; TAC. Dial. 23, 23; 
 24; Agric. 30, 1; Germ. 15, 1; Hist. 1, 37, 10; Ann. 4, 1, 
 19; 6, 11, 16; 42, 4; Juv. 2, 2; 156; 3, 306; 5, 28; 6, 67; 
 180; 194; 535; 647; 7, 179; 9, 51; 111; 14, 21; TERENT. 
 SCAUR. VII (K,) 16, 1 ; 23, 6 ; 29, 4 ; 6 ; 15 ; SUETON. de 
 Vir. 111. p. 118, 5 (Reiff.); 140, 14; 22; 141, 3; Praturn 
 211, 11 ; GAIUS 4, 2; MIN. FEL. p. 40, 8 (Halm); TERTUIX. p. 
 66, 12 (R.-W. 1) ; 107, 8 ; 10 ; 221, 15 ; CYPR. p. 801, 18 ; App. p. 
 31,2; 33,6; 35,15; 39,1; 239,9; SOLIN. p. 221, 15 (Mommsen.); 
 16; Jut. VAI.ER. Ep. Alex, ad Dind. p. 170, 9 (Kuebler) ; 
 T E RENT.MAUR.Vi(K.)Lit.l42;172; 220; Syl. 792; 1011 ; 1128; 
 1207; 1273; Met, 1523; 1840; 1849; 1992; 2749; PALLAD. 1, 
 28,2; 3,25,22; 4,9,8; 10,10,3; LACT. p. 140, 6 (L.-B. 1); 170, 
 18 ; 207, 7; 496, 21 ; 532, 12; 693, 2 ; JUVENC. 3, 224; MAR. 
 VICTORIN. VI (K.) 10, 10; 15,8; 17,9; 19,6; 23,7; 33,8; 61, 
 6; 83,12; 107, 20; 2l (bis); 163, 11; 12; Ars 195, 21; 22; 
 23; 199, 14; 202, 14; 204, 4; 5; de Rat. Met. Com. 217, 17; 
 CHARIS, I (K.) 8, 18 (bis); 85,13; 173,9; 227,3; 10; 12; 30; 
 228, 1; 9; 16 (bis); 22; 233, 4; 10; 11; 234, 3; 235, 25; 
 236, 1; DIOM. I (K.) 308, 31; 33; 35; 309, 1; 27; 322, 11; 
 372, 7 ; 9 ; 25 (bis) ; 378, 8 ; 391, 20 ; 21 ; 332, 6 ; 28 ; 393, 8 ; 
 10; 394, 1; 3; 12; 20; 23; 27; 29; 397, 16 ; 405, 20; 406, 5; 
 11; 12; 427, 17; 21; 26; 434, 13; 446, 17; 448, 14; 478, 
 3; 480, 6; 9; 494, 32; 495, 9; 15; VEGET. 2, 22 (quater); 3, 
 5 fin. ; 4, 21 ; AUSON. p. 14, vs. 12 (Peip.); 17, 31 ; 92, 19 ; 21 ; 
 228, 43; 253, 33; 286, 24; 354, 34; 370, 433; AUCT. 
 QUEROI,. p. 37, 12 (Peip.); PRUDENT. Apoth. 838; Ham. 87; 
 Psych. 7; 713; cont. Sym. 2, 312; 1097; AMM, MARC. 17, 5, 
 14; MACROB. Sat. 1, 24, 4 ; 4, 6, 9 ; 5, 15, 6 ; Som. Scip. 1, 6, 
 66; 21, 5; SYM. Epp. 1, 37, 1 ; 84; 2, 91; 3, 32; 80; 9, 69; 
 
 1 ducitur scripsi, Buech. 
 
On the Usage of Quotiens and Quotienscunque. 29 
 
 80; SERV. Eel. 3, 28; 9, 54; Ge. 2, 477; Aen. 1, 23; 41 ; 65; 
 108 (bis); 159; 208; 209; 228; 251; 257; 332; 343; 371 
 (bis) ; 384 ; 423 ; 435 ; 436 (bis) ; 451 ; 532 ; 744 ; 2, 270 ; 
 300; 3, 205; 275; 384; 4, 418; 5,533; 7, 8; 8,12; 110; 
 168; 209; 268; 9, 81; 192; 360; 452; 10, 33; 74; 419; 
 668; 11, 284 (bis); 463; 551; 593; 659; 12, 164; 575; 
 CLAUDIAN. in Rufin. 2, 114; in Eutrop. 1, 239; 241; 2, 404; 
 446; de iv Consul. Honor. 523; 563; Pan. Manl. Theod. 100; 
 de Cons. Stil. 3, 362; de Rapt. Proserp. 1, 157; 2, 63; MART. 
 CAPELL. 3, 251; 302; 308 (bis); 311; 5, 534; 535; 7, 769; 
 781; 785 (ter); 786; SAI.VIAN. p. 151, 27; 152, 26; FAUST. 
 p. 236, 22 (Engelbr.) ; 263, 17 ; 264, 8 ; 292, 6 ; 300, 19 ; 308, 27 ; 
 LUCIF. p. 268, 22 (Hartel) ; 293, 15 ; SID. APOIX. Epp. 2, 6, 1 ; 3, 
 10, 1; 5, 5, 3; 6, 8, 1; 7, 14, 11; 15, 1; 8, 11,13; 14,4 (bis); 
 5; 6; 9, 16,3; Carrn. 2, 112; 308; 5,52; 198; 406; 22,35; 
 VICT. VIT. p. 40, 5 (Petschen.); 83, 9; PRISC. n (K.) 197, 10; 303, 
 6 ; m, 19, 9 ; 127, 12 ; 142, 17 ; 252, 4 ; 338, 15 ; BOETH. Arith. 1, 
 24 (p. 49, 16 Fried.); 29 (61, 2); 31 (65/2); 2, 29 (120, 14); 40 
 (138, 1); 51 (165, 23); 52 (167, 21); Mns. 1, 3 (190, 8); 24 
 (217, 6); 25 (218, 3); 28 (220, 3); 2, 13 (243, 5); 30 (263, 23); 
 Geom. p. 379, 23 ; Consol. 1, 2 (p. 6, vs. 4 Peip.) ; 4 (p. 10, vs. 7) ; 
 4 (p. 14, 110); 2, 6 (p. 43, vs. 16); 4, 7 (p. 118, 38); Trin. p. 
 149, 6; ENNOD. p. 19, 12; 20, 20; 22, 15; 30, 14; 40, 
 15; 61, 1; 92, 17; 97, 9; 102, 14; 107,4; 110,24; 133, 24; 
 160,7; 162,10; 176,13; 178,10; 254,23; 268, 9; 269, 18; 
 274,5; 305,13; 392,4; 414,17; 430,2; 436,4; 446,4; 454, 
 6; 18; Carm. 1, 6, praef. (521, 14) ; 6,2; 7,41; 9,72; 158; 2, 
 56, 10; 69, 2; 80, 7; 121, 1 ; 148, 1 ; DIGEST. Scaev. xxxv 2, 23; 
 Lab. xvn, 2, 84 ; xxxvn, 10, 9 ; Javol. xvni, 1, 65 ; Julian, ix, 
 4, 34 ; xn, 1,19 praef. ; vol. xxxv, 1, 24 ; Pomp. I, 2, 2, 33 (bis) ; 
 Viii, 1,15; xxxii. 85 praef. ; xxxiv, 3, 8, 3 ; L, 17, 20 ; Yenul. 
 xxi, 1, 65, 1 ; Callist. I, 18, 9 ; IV, 2, 13 ; Paul. I, 5, 7 ; n, 8, 6 ; 
 xxi, 2; L, 17, 136; Ulp. n, 14, 7, 8 (bis); iv, 6, 26, 9; XLIII, 
 8, 2, 10 ; L, 16, 38 ; Modest, xxvm, 5, 63, 1 ; XLI, 1, 52 ; Herm. 
 I,, 17, 98 ; CYPR. GAUL. Gen. 687 ; Exod. 450 ; 631 ; 769 ; 991 ; 
 Num. 105; Deut. 69; 130; Jes. Nav. 533; Jud. 330; CASSIOD. 
 Vii (K.) 161,15; 24; 162,7; 163, 18. Imperfect CURT. 10,1, 
 
30 On the Usage of Quotiens and Quotienscunque. 
 
 28 ; 4, 3 ; SEN. Epp. 18, 1, 33 ; J Apocol. 15 ; VEL. LONG, vn (K.) 
 60, 12; 70, 16; 17; FBONTTN. Aq. 95; PLIN. Epp. 1, 2, 4; 
 TEBENT. SCAUR. VII (K.) 14, 6 ; 15, 4; CHAKIS. I (K.) '217, 11 ; 
 229, 3; AMM. MABC. 20, 11, 9; SEBV. Aen. 1, 65; 3, 57; 175; 
 5, 95; 9, 624; 638; CLAUDIAN. de Bell. Poll. 377; Laus Ser. 
 134 ENNOD. p. 338, 4. 'Future CATO, Agr. Cult. 59; Cic. 
 ad Fam. 1, 7, 1; ad Alt. 11, 13, 5; PBOPEBT. 3, 8, 56; 
 12, 27; 4, 6, 73; HOB. Epp. 1, 18, 45; SEN. Controv. 2, 5, 11 ; 
 9, 5, 4 (bis); OVID, Her. 13, 69; A. A. 2, 327; 395; CELS. 5, 
 26, 22; SEN. Epp. 2, 1, 13; 4, 6, 4; 8, 2, 2 ; 12, 18, 9; 
 Dial. 4, 34, 4 ; 7, 26, 7 ; de Clem. 2, 6, 3 ; de Ben. 4, 7, 1 ; Nat. 
 Quaest. 1 prolog. 10 ; 2, 34, 3 ; de Mor. 130 ; PLIN. Nat. Hist. 18 r 
 323 (bis); QUINTIL. 1, 11, 8; 2, 13, 3 (bis); 3, 7, 24; 4, 1, 72; 
 2,26; 5, 17; 7,2,36; 4,31: 5, 3 ; 6, 9 ; 10,1,22; 7,29; 11, 
 1, 9 ; PSEUD.-QITINTIL. p. 7, 4 (Ritter) ; 70, 5 ; STAT. Theb. 4, 
 505; Achill. 1, 909; Sit. ITAL. 15, 113; MABT. 11, 7, 2; 14, 
 7, 2; PUN. Epp. 6, 12, 5; Pan. 67; TAC. Hist. 3, 51, 13; Jcv. 
 3, 318 ; APUL. Met. 2, 3 ; 26 ; TEBENT. MAIJB. VI (K.) Syl. 1234 ; 
 Met. 1566 ; 1837 ; PALLAD. 3, 24, 11 ; MAB. VICTOBIN. VI (K.) 
 10, 11 ; 18, 18 ; 19, 20 ; MABCELL. p. 298, 17 (Helmr.) ; 324, 16 ; 
 342, 15 ; SEDUL. p. 313, vs. 73 (Huemer) ; SID. APOLL. Epp. 9, 13, 
 3. Perfect, (a) Contemporary Gw. Quinct. 19, 62 ; Balb. 20, 47 ; 
 SALL. Hist. Fragg. 2, 98, 1 (Maur.) ; PBOPEBT. 1, 3, 27 ; 16, 43 ; 
 HOB. Carm. 4, 9, 40 ; SEN. Controv. 3, 8 (ter) ; OVID, Rem. Am. 
 304 ; VALEB. MAX. 8, 10, 1 ; CUBT. 6, 10, 35 ; CELS. 2, 12 fin. ; 
 SEN. Epp. 2, 1, 2 ; Dial. 2, 8, 3 ; de Ben. 2, 11, 2 ; Medea 433 ; 
 PLIN. Nat. Hist, 7, 157 ; MABT. 4, 66, 12 ; PLIN. Pan. 58 ; TAC. 
 Agric. 1, 3; Ann. 1, 56, 17; GELL. 19, 8, 1; CYPB. Carm. 6, 
 258; JUL. CAPITOL. Ver. 5, 3; MAB. VICTOBIN. VI (K.) 12, 
 9; PBFD. Apoth. 191; cont. Sym. 2, 547; Perist. 4, 81; 11, 
 178; AMM. MABC. 16, 5, 5; SYM. Epp. 1, 2, 7; CLAUDIAN. de 
 Bell. Poll. Ill ; SULP. SEVEB. p. 169, 25 (Halm) ; MABT. CAPELL. 
 1, 37; L.UCIF. p. 4, 16; 154, 19; SID. APOLL. Carm. 5, 
 113; BOETH. Consol. 4,7 (p. 118, 39 Peip.); DIGEST. Pedius 
 apud Ulp. I, 3. 13; Ulp. VII, 2, 1 ; 4, 3 praef. ; EUGIPP. p. 5, 19 
 
 1 fuerat = erat, cf. Blase, Geschichte des Plusquamperfekts im Lateinischen, 
 Giessen, 1894, p. 48. 
 
On the Usage of Quotiens and Quotienscunque. 31 
 
 (Knoell). (b) Prior PROPERT. 3, 9, 3 ; SEN. Controv. 10, 6, 2; 
 OVID, Her. 13, 51 ; CELS. 3, 16 init. ; 4, 22 init. ; 23 med.; 6, 
 18, 3; 9 fin.; 7, 29 fin.; 8, 10, 2 fin.; PETRON. 90; SEN. Epp. 
 
 5, 1, 18 ; 6, 4, 1 ; 9, 5, 4 ; 10, 1, 2 ; 14, 4, 17 ; 15, 2, 73 ; 18, 5, 
 8; 19, 1, 17; 20, 2, 1; Dial 3, 12, 1; 9, 11, 8; 12, 12, 
 4; de Clem. 1, 6, 1 ; Nat. Quaest. 2, 59, 11 ; PMN. Nat. Hist. 2, 
 106; 11, 209; 15, 134; 17, 40; 18, 146; 34, 108; 35, 44; 
 QUINTIL. 2, 14, 1; 10, 4, 3; 6, 1; 11,1,16; 3,120; PSEUD.- 
 QUINTII,. p. 18, 15 (Hitter); 46, 14; 63, 26; 166,5; 246,24; 
 307, 12; STAT. Theb. 10,924; VEI,. LONG. VII (K.) 55, 17; 
 MART. 1, 105, 2 ; 5, 7, 2 ; 47, 2 ; 7, 18, 5; 76, 3 ; 9, 25, 1; 10, 
 42, 5; 11, 102, 8; 12, 29, 18; 57, 28; 13, 18, 2; Juv. 1, 165; 
 3, 40 ; JUSTIN. 14, 4, 7 ; CYPR. App. p. 239, 10 ; 243, 9 ; Carm. 
 
 6, 282; AUSON. p. 285, 15 (Peiper) ; CI,AUDIAN. de Cons. Stil. 3, 
 60; 331; KuRic.p. 403, 16 (Engelbr.); DIGEST. Ulp. II, 14, 7, 16 ; 
 IV, 6, 26, 8; xr, 7, 22; CYPR. GAIX. Levit. 178. Pluperfect, 
 Prior CAES. B. G. 5, 34, 2 ; SEN. Suas. 3, 6 ; Controv. 1 praef. 
 15 ; 2 praef. 2 ; 2 praef. 5 (bis) ; OVID, Met. 3, 495 ; VAMSR. MAX. 
 2, 1, 6 (PAR. also, and at 3, 6, 4) ; CURT. 10, 1, 29 ; Luc. 3, 700 ; 
 SEN. Epp. 6, 3, 3 (bis); Dial. 4, 10, 5 ; 9, 15, 2 ; Nat. 
 Quaes. 2, 26, 4; SIL. ITAI 17, 339; MART. 11, 104, 14; TAC. 
 Hist. 1, 10, 8; Ann. 1, 25, 4; 15, 30, 2; Juv. 10, 29; FIX>R. 2 
 28, 18; GELT.. 9, 5, 8; PRUD. Perist. 10, 703; SutP. SEVER, p. 
 204, 4 ; MART. CAPELL. 3, 227. Future Perfect, Prior PUBMI,. 
 SYR. M. 11; HOR. Carm. 4, 10, 6; (PORPH. Epp. 1, 18, 45); 
 SEN. Controv. 1 praef. 20; OVID, Fast. 4, 861 ; Cot. 2, 17, 6; 
 de Arb. 15 ; CURT. 10, 6, 15 ; CELS. 1, 8 fin. ; 3, 18 fin. ; 21 med. ; 
 5, 19, 12; 7, 3 med.; 8, 10, 1 fin.; SC.RIB. LARG. 200; SEN. 
 Epp. 2, 1, 6; 5, 6, 11; 7, 7, 2; 9, 1, 1 ; 18, 1, 9; Dial. 4, 
 
 7, 2 ; 6, 8, 2 ; 9, 3 ; 12, 18, 9 ; Nat. Quaest. 2, 1, 4 ; 3 praef. 11 ; 
 4 praef. 21 ; 6, 23, 3 (ter); PMN. Epp. 2, 6, 6 ; QUINTII,. 4, 1, 
 70; 76; 8,3,55; 11,1, 78; HYGIN. GROM. 4 ; 32 ; MART. 5,4, 
 5; 11, 95, 1; Juv. 5, 145; TERENT. SCAUR, vn (K.) 24, 2; 
 CYPR. p. 430, 19; TERENT. MAUR. VI (K.) Lit, 199; PAIXAD. 
 3, 18, 2 ; MAR. VICTORIN. VI (K.) 13, 24 ; 23, 2 ; 33, 27; I>IOM. 
 I (K.) 473, 11 (cf. 502, 26); MACROB. Som. Scip. 1, 6, 33; 
 MARCEIX. p. 55, 6 (Helmr.); Ill, 28; 149, 26; 210, 22; 237, 
 32; 270, 22; 272, 5; 281, 14; 18; 292, 21; 311, 22; 327, 
 
32 On the Usage of Quotiens and Quotienscunque. 
 
 29; 342, 24; 345, 10; 376, 17 ; SYM. Epp. 3, 8 ; 7, 3 ; MART. 
 CAPEIX. 4, 346; CASS. FEL. 19; 33; 42; 51; 53; 59; 66; 
 FAUST, p. 112, 4 ; 266, 23 ; LUCIF. p. 254, 24 ; SID. APOIX. Epp. 8, 
 14, 5 ; BOETH. Arith. 1, 29 (p. 61, 26 Fried.) ; Geom. p. 393, 17 ; 
 ENNOD. p. 142, 18; 161,17; 164, 5; 503, 19; DIGEST. Paul, 
 xxxvir, 1,6; Ulp. vii, 4, 3 praef. ; Marcian. xxxix, 4, 16, 3 ; 
 CAssioi>.vii(K.)172, 3. 
 
 SUBJUNCTIVE, Present SEN. Controv. 2, 5, 11 (ter) ; CURT. 
 4, 10, 6; SEN. Epp. 6, 3, 9 ; Nat. Quaest. 1, 13, 1 ; PJLIN. 
 Nat. Hist. 17, 129; 18.350; 24,158*; 27,7; 32,15; 133; 34, 
 122; QUINTIL. 1, 7, 10; PSEUD. -QUINTII,. p. 377, 21 (Ritter) ; 
 PJLIN. Epp. 8, 20, 1; TAC. Ann. 1, 26, 13; 3, 69, 16; 12, 
 19, 8; 47, 6; TERENT. SCAUR. VII (K.) 11, 9; 12, 17; 21, 17; 
 AMM. MARC. 17, 10, 4 ; SID. APOIX. Epp. 1, 7, 7 ; PRISC. n (K.) 
 37, 21 ; DIGEST. Scaevol. XLVI, 3, 93, 2; Ulp. L, 17, 161. Im- 
 perfect LIVY 27, 6, 7; 39, 33, 8; Perioch. 59, 20; PROPERT. 
 2, 1, 27; SEN. Controv. 7, 4, 4; 7, 19; CURT. 10, 4, 3; PmN. 
 Nat. Hist. 7, 141 ; 8, 14; 27, 69; QUINTIL. 1, 7, 20; 2, 4, 13; 
 27; 9, 4, 38; STAT. Silv. 5, 3, 215; FRONTIN. Strat. 3, 3, 4; 
 Aq. 12; 125; MART. 7, 37. 3; PUN. Epp. 3, 16, 4; Pan. 13; 
 TAC. Hist. 1, 24, 5; 66, 19; Ann. 2, 2, 12; 34, 22; 3, 33,14; 
 65, 10; 4, 31,9; 6,10,13; 15,13; 21,1; 29, 8; 13,3,16; 18, 
 16; 27, 14; 14, 20, 9; 52, 11; 15, 13, 10; TERENT. SCAUR. 
 Vii (K.) 25, 3; 4; 18, 18; SUETON. Jul. 26; 28; Aug. 21; 
 37; 38; 41; 44; 45; 56; 57; 77; 82; Tib. 19; Calig. 4 ; 33; 
 35; Claud. 8; Nero 15; 27; 28; Galba 10; Otho 4; Dorait. 
 2; 21; de Vir. 111. p. 24, 3 (R.); 61, 4; 138, 17; CYPR. 
 Carm. 2, 5 ; Jui CAPITOL. M. Anton. 10, 6 ; AEL. LAM- 
 PRID. Commod. 11, 11 ; MAR. VICTORIN. VI (K.) 12, 16 ; 20, 7 ; 
 OBSEQ. 21; DIOM. I (K.) 476, 2; AMM. MARC. 16, 12; 67; 
 CLAUDIAN. Carm. Min. 30, 97 ; DIGEST. Javol. vni, 1, 20 ; Papin. 
 xix, 5, 8. Perfect, Prior OVID, Trist 5, 10, 42 ; SEN. Epp. 2, 
 1, 6; 19, 1, 17; Dial. 7, 24, 3; PMN. Nat. Hist. 18, 259; 23, 
 103; 28, 11 (bis); 36, 134; QUINTII,. 12, 10, 33; PSEUD.- 
 QUINTII,. p. 3, 1 (Ritter) ; 210, 2 ; VEL. L.ONG. VII (K.) 63, 9 ; 
 11 ; TERENT. SCAUR. VII (K.) 25, 19 ; AMFEL. 8, 17 ; CYPR. App. 
 p. 33, 7 ; DIOM. I (K.) 427, 28 ; AMM. MARC. 14, 3, 1 ; SYM. 
 Epp. 3, 65; 4, 61, 3; SERV. ^En. 12, 151 ; CASS. FEL. 38; 60; 
 
UNIVERSITY )/ v 
 
 ^^~-:..u.J=L^ay 
 
 On the Usage of Quotiens and Quotienscunque. 33 
 
 61 ; 72 ; FAUST, p. 330, 24 ; SID. APOLI,. Epp. 9, 1 1, 8 ; Carm. 22, 
 5; ENNOD. p. 69, 4 ; 171,8; 437, 23; 471,16; Carm. 1, 8 
 praef. 1 ; 9, 62 ; DIGEST. Terent. Clemens xxxv, 2, 67 ; Ulp. vn, 
 1, 25, 7; BOJETH. Consol. 1, 6 (p. 22, 54 Peip.) ; CYPR. GAIX. 
 Exod. 980. Pluperfect, Prior LIVY 23, 32, 3 ; SEN. Controv. 
 10 praef. 13 ; CURT. 6, 11, 2; 9, 6, 4; *EN. de Ben. 3, 32, 5; 
 PLIN. Nat. Hist. 23, 56; QCINTII,, 6, 1, 40; FRONTIN. Strat. 4, 
 7, 7; Aq. 87; TAC. Hist. 3, 83, 3 ; Ann. 3, 71, 11 ; 4, 16, 19; 
 6,2,10; 13,49, 8; SUETON. Aug. 99 ; Calig. 36; Claud. 42 ; 
 MAR. VICTORIN. VI (K.) 24, 4 ; PRUDENT. Perist. 10, 758 ; SERV. 
 ^En. 11, 517 ; CLAUDIAN. in Eufin. 2, 75 ; de in Consul. Honor. 
 25 ; de vi Consul. Honor. 82 ; EUGIPP. p. 33, 1. 
 
 D. CORRELATIVE : INDICATIVE, Present CATUIX. 71,5; 
 Cic. Vatin. 12, 29 ; de Orat. 1, 125 ; 251 ; 2, 130 (bis); PUBLIL. 
 SYR. H. 13 ; SEN. Controv. 1,1,6; OVID, Her. 19, 113 ; Met. 10, 
 164 ; ex Pont. 4, 9, 112; Coi,. 11, 3, 32 ; CAUP. Sic. 2, 88 ; SEN. 
 Epp. 19, 4, 7; Dial. 9, 17, 1; 11, 12, 3; Here. O. 426*; 614; 
 VEL. LONG. VII (K.) 58, 10; 66, 7; SUETON. Prat. p. 275, 9 
 (Reiff.); CYPR. p. 576, 23; MAR. VICTORIN. VI (K.) 72, 15; 
 SERV. Aen. 1,407; MART. CAPEIX. 7, 791; 792*; SAI.VIAN. 
 p. 302, 10; FAUST, p. 39, 1; 199, 23; DIGEST. Scaevol. XLV, 3, 
 19 ; Javol. vm, 6, 15 ; xvin, 1, 65 ; CASSIOD. vm (K.) 161, 11 ; 
 BOETH. Trin. p. 151, 17 (Peip.). Imperfect OVID, Her. 10, 
 23; SEN. Dial. 5, 18, 1; APUI. Met. 10, 22. Future CATO, 
 Agr. Cult. 1,1; Cic. Marc. 6, 19; Rab. Post. 9, 25; SEN. 
 Dial. 4, 7, 2; PHAED. 177; Here. O. 426; PUN. Pan. 88; 
 MART. CAPEILL. 7, 792. Perfect, (a) Contemporary VERG. Aen. 
 12, 483; PORPH. ad HOR. Carm. 4, 9, 40; SEN. Controv. 7, 
 1,27; TAC. Ann 1 . 14,64,14; MIN. FEL. p. 36, 8 (Halm); CYPR. p. 
 580, 7 ; Jui,. VALER. Ort. Alex. 3, 60 ; CI.AUDIAN. de IV Consul. 
 Honor. 621 ; FAUST, p. 68, 24 ; L.UCIF. p. 286, 4 ; ENNOD. p. 348, 
 9. (b) Prior. Cic. Sull. 30, 83 ; SEN. Suas. 6, 21 ; OVID, Met. 3, 
 451 ; VAUER. MAX. 4, 1, 7 ; CEI,S. 4, 18 med. ; VAI/ER. FI.ACC. 
 6, 683; GEIX. 14, 2, 19; SERV. Aen. 9, 138; DIGEST. Ulp. 
 Ill, 2, 2, 2 ; vii, 2, 3 praef. Future Perfect, Prior SCRIB. LARG. 
 232; MAR. VICTORIN. VI (K.) Met. 211, 16; 1 MACROS. 7, 8, 12; 
 
 1 Cf. Teuff. Hist. Rom. Lit. $ 408, 4 ; no especial fondness seems to be shown. 
 
34 On the Usage of Quotiens and Quotienscunque. 
 
 MARCELL. p. 385, 10 (Helmr.); SALVIAN. p. 302, 14; CLAUD* 
 MAMERT. p. 93, 6 (Engelbr.) ; DIGEST. Callist. L, 16, 220 praef. ; 
 BOETH. Mus. 1, 3 (p. 190, 10 Fried.). 
 
 SUBJUNCTIVE, Present Cic. ad Fam. 7, 7, 1 ; LIVY 6, 
 15, 7; SEN. Epp. 14, 2, 15; de Ben. 1, 11, 6; GAI. 1, 
 162. Imperfect Cic. ad Att. 1, 14, 3 (ter); SEN. Controv. 4 r 
 praef. 7; PEIN. Pan. 42; 58; FRONTIN. Strat. 2, 13, 7; AEL. 
 SPART. Anton. Geta 7, 5 (bis). Perfect, Prior PSEUD.-QUINTIL* 
 p. 16, 18 (Hitter); 24; TERTUM,. p. 360, 13 (R.-W. 1); LACT. 
 p. 576, 17 (L.-B. 1); MACROB. 7, 14, 9 ; SERV. Eel. 1, 8 ; AUGUST, 
 Civ. Dei 2, 1 fin. ; LUCIF. p. 190, 9; SID. AFOLI,. Epp. 7, 14, 11. 
 
 QUOTIENSCUNQUE. 
 
 A. RELATIVE : INDICATIVE, Present PLAUT. Capt. 97 ; 
 True. 282; Cic. ad Fam. 6, 5, 1 ; 13, 41, 1 ; PORPH. ad HOR. 
 Epod. 17, 52; HYGIN. Astron. 2, 14 raed. ; VALER. MAX* 
 8, 15, 1; CELS. 6, 6, 16; 7, 1 fin.; 8, 10, 1 med. ; PETRON. 
 36; PSEUD.-SEN. Epp. ad Paul. 4; 10; CYPR. p. 331, 24; 
 706, 3; 714, 24; App. 30, 20; CHARTS. I (K.) 217, 11 ; 229, 3 ; 
 OROS, p. 638, 1 7 (Zang.) ; MART. CAPEIX. 7, 794 ; SERV. Eel. 1, 37 ; 
 Ge. 1,397; Mu.l, 9; 17; 73; 77; 3, 16; 241; 384; 4, 179; 
 FAUST, p. 110, 19 ; 226, 3 ; PRISC. n (K.) 25, 16 ; 33, 19 ; FUL- 
 GENT. Myth. 2, 16 med.; BOETH. Arith. 2, 4 (p. 87, 6 Fried.) ; 46 
 (p. 150, 19); GREG. TUR. Vit. Patr. 20 praef. Imperfect VALER. 
 MAX. 8, 12, 1 (PAR. also) ; CHARIS. I (K.) 195, 9. Future CATO, 
 Agr. Cult. 151, 4 ; AUCT. AD HER. 3, 20, 34 ; Cic. Tusc. 3, 84 ; 
 Fin. 5, 29 ; ad Fam. 16, 11, 3 ; de Orat. 3, 123 ; LIVY 43, 14, 6 ; 
 VITRUV. 10, 14, 7; DIGEST. Alf. Var. xxxm, 1, 22; Ulp. xxv, 
 4, 1, 10; GREG. TUR. Hist. Franc. 4, 16 med. Perfect, (a) Contem- 
 porary NEPOS 17, 3, 6 ; 18, 5, 7 ; 23, 1, 2 ; Cic. Prov. Cons. 1,2; 
 Verr. 4, 26, 57 ; Cat. 1, 5, 11 ; ad Fam. 5, 2, 9 ; 6, 1 ; 13, 69, 1 ; 
 CURT. 5, 2, 22; PETRON. 85; FRONTO, ad Am. 1, 3 (p. 175 N.); 
 TERTUIX. p. 152, 11 (R.-W. 1); ST. SILVIA, Peregrin. 50; 
 OROS. p. 189, 19; SERV. JEn. 4, 627; SALVIAN. p. 20, 10; 
 GREG. TUR. de Virt. S. Jul. 43. (b) Prior VARRO, R. R. 3, 10, 
 7; LIVY 7, 26, 5; SEN. Controv. 1, 8, 6; CELS. 7, 12, 1 fin.; 
 DIGEST. Ulp. IV, 6, 26, 9. Pluperfect, (a) Contemporary AUCT* 
 
On the Usage of Quotiens and Quotienscunque. 35 
 
 BELL. AFR. 72, 1 . (b) Prior VALER. MAX. 3, 6, 4 ; PETRON. 111. 
 Future Perfect, Prior Cic. de Orat. 2, 244 ; Sest. 69, 146 ; 
 VITRUV. 5, 12,4; CYPR. p. 430, 18; 708, 23; 24; 709, 3; 
 AUGUST. Civ. Dei 11, 31 med.; FAUST, p. 38, 24; BURIC. p. 425, 
 23 ; DIGEST. Ulp. L, 5, 1, 1 ; BOETH. Mus. 2, 8 (p. 236, 7 Fried.). 
 
 SUBJUNCTIVE, Present Cic. Brut. 4. Imperfect Cic. Phil. 5, 
 8, 21 ; Dom. 29, 78 ; de Orat. 1, 123 ; LIVY 5, 54, 3. Perfect, (a) 
 Contemporary CAES. B. C. 1,7, 5. (b)- Prior FRONTO, ad Anton. 
 3 (p. 1 52 N.). Pluperfect, (a) Contemporary LIVY 39, 39, 1 1 . (b) 
 Prior NEPOS 7, 3, 5 ; LIVY 8, 35, 10; AMM. MARC. 16, 10, 18. 
 
 B. CORRELATIVE : INDICATIVE, Present VARRO, R. R. 2, 
 
 4, 15 ; Cic. Cluent. 18, 51. Future Cic. de Orat, 2, 249; Plane. 
 6, 14. Perfect Cic. Rose. Com. 6, 18 ; Dom. 26, 69 ; 27, 71. 
 Future Perfect, Prior Cic. de Orat. 2, 137. 
 
 SUBJUNCTIVE, Imperfect I/IVY 31, 44, 6. Perfect Cic. Yerr. 
 
 5, 8, 21. 
 
 QTJOTIENSQUE. 
 
 RELATIVE: INDICATIVE, Preset COL. 6, 17, 9; 12, 52, 
 16; DIGEST. Ulp. XLin, 8, 2, 10; RURIC. p. 382, 13. Imper- 
 fect ST. SILVIA, Peregrin. 50. Future COL. 2, 4, 10; 12, 51, 
 2. Future Perfect, Prior DIGEST. Ulp. vii, 4, 3 praef. 
 
 SUBJUNCTIVE, Imperfect SUETON. Aug. 37 ; Calig. 7. 
 
36 On the Usage of Quotiens and Quotienscunque. 
 
 PART III. 
 
 SUMMARY. 
 
 Many authors show a distinct fondness for quotiens as suited to 
 a given sphere, or appropriate for some particular feature of style, 
 while others evidently avoid its use. In Orosius and Augustine, 
 for example, but few occurrences were found in works of consider- 
 able extent. Of the authors in whom no examples were noted, the 
 works are in most cases brief. Lucretius, Pomponius Mela and 
 Probus are the most notable exceptions, the last especially, in con- 
 trast with the striking usage of other grammarians. No examples 
 were noted in the following list : 
 
 Lucretius, Hirtius, Mela, Nemesianus, Persius, Probus, Caper, 
 Granius Licinianus, Censorinus, Commodianus, Plotius, Publilius 
 Optatianus, Atilius Fortunatianus, Jul. Firm. Maternus, Aurel. 
 Victor (de Caess.), Eutropius, Dictys Cretensis, Rufinus (de 
 Comp.), Hieronymus (deVir. 111.), Mall. Theodorus, Pelagonius, 
 Agroecius, Dares Phrygius, Jordanis. 
 
 On the other hand, but two authors, Cicero and Livy, show 
 examples in all the main divisions given, followed next in range 
 by Cato and the Senecas, though the latter do not use the com- 
 pound correlative; while Vergil, Ovid, Pseudo-Quiutilian, Taci- 
 tus, Claudius Claudianus and Ennodius have quotiens in all its 
 divisions, with intentional omission of the compound. 
 
 Quotiens interrogative (119 examples) is less frequent than quo- 
 tiens in exclamation (148 examples), and these two classes combined 
 are but little more than one-fourth of the simple relatives (1053 
 examples). The proportion given between interrogatives and 
 exclamations is in reality less exact because the rhetorical question 
 and the exclamation are so readily interchanged ; cf. for example 
 Serv. Sulpic. apud Cic. ad Fam. 4, 5, 3 quotiens in earn cogita- 
 tionem necesse est et tu veneris et nos saepe incidimus . . .? 
 Ovid, Her. 16, 217 ipse mihi quotiens irattis " ad ultera" dices ? 
 
On the Usage of Quotiens and Quotienscunque. 37 
 
 Lucan 8, 408 Parthorum dominus quotiens sic sanguine mixto 
 Nascitur Arsacides ? and further, Sen. Epp. 14, 3, 9 ; Stat. Theb. 
 5, 615; Fronto, ad Ver. Imp. 2, 8; Cypr. p. 44, 15; Carm. 6, 
 278; 279 ; Arnob. p. 271, 16 ; Ennod. p. 9, 22 ; 310, 8 ; 9 ; 372, 
 2 ; 3 ; 426, 26 ; 27 ; 479, 27, etc. 
 
 In the exclamatory usage most of the examples, as was to be 
 expected, refer to the past. The perfect indicative is so used in 
 more than three-fourths of the total number of examples. Here, 
 too, the emotional character of the elegiac in Tibullus, Proper- 
 tins and Ovid adds largely to the total, though the strongly- 
 marked rhetorical tendency 1 in Ovid must also be taken into 
 account. 
 
 The most noteworthy features in connection with the simple 
 relative usage are the departmental characteristics and the marked 
 tendency toward the present indicative, contemporary, type. Epic 
 poets, on the whole, do not use quotiens so freely as do the Elegiac 
 and the Lyric poets. In prose, wherever the oratorical or rhetor- 
 ical element is strong, quotiens is naturally in frequent use ; com- 
 pare the numerous examples in Cicero, the speeches in Livy, 2 the 
 Senecas, Pliny the Elder, etc. No examples are found in Sallust 
 except in fragments of speeches. Again, among medical writers 
 where definite direction is given : As often as the fever comes on, 
 whenever certain symptoms appear, etc., the very large percentage 
 of future and future perfect examples is to be noted, inasmuch as 
 quotiens in these authors is largely confined to the one class of 
 examples. Compare especially Marcellus and Cassius Felix, the 
 latter having, in indicative examples, only the future perfect with 
 quotiens. 
 
 The grammarians show a characteristic no less marked in the 
 frequency of quotiens in definitions and rules. Here the terms 
 are general and the present indicative prevails. Yet this type is 
 in the line of a general development. Observe its growth : Cicero, 
 out of six examples of the relative, uses the present indicative but 
 twice (not at all in the orations), Vergil out of three examples has 
 
 1 Cf. Schanz, Gesch. d. rom. Lit. Miinch. 1899, u, p. 190. 
 
 2 More frequently as interrogative, however, than as relative. 
 
38 On the Usage of Quotiens and Quotienseunque. 
 
 it twice, but Propertius has no occurrence in seven examples, nor 
 Livy in four. Seneca Rhetor in 24 examples has three cases, 
 while Seneca the Younger has only 24 instances out of 71 ex- 
 amples. But Ovid has 15 out of 23, and Lucan five out of six. 
 Curtius again has no occurrence in nine examples, Pliny the 
 Elder but three out of 30, Quintilian, forerunner of the gram- 
 marians, 25 out of 55, Tacitus eight out of 40. Then the pro- 
 portion becomes larger : Tertullian has four out of five, Maurus 
 13 out of 17, Lactantius has only the single usage in six examples, 
 as has also Charisius in 19, Diomedes has 41 oat of 45, Flavius 
 Vegetius six of the one class, Ausonius nine out of ten, Macrobius 
 five out of six, Servius 55 out of 63, Claudius Claudianus 11 out 
 of 20, Martianus Capella 13 out of 17, Sid. Apoll. 18 out of 26, 
 Priscian seven out of eight, Ennodius 37 out of 49 and Oyprianus 
 Gall us 10 out of 12. Or by periods, until the time of Tertullian 
 out of 414 indicative examples only 196 were of the present tense; 
 but from Tertullian to the end of the period examined out of 
 471 examples, 370 are of this classification. This shows a decided 
 falling off in the exactness obtained at an earlier period by the use 
 of tenses of priority. Our list would appear still larger if we 
 counted here the 39 examples out of 91 indicatives with the full 
 compound, and four out of eight with quot\ensque. In fact this 
 usage became so identified with the simple form that we may find 
 here an explanation of the comparative infrequency of the com- 
 pound except in the sphere of vulgar Latin, where fullness of 
 expression is a characteristic. 
 
 The list of examples has been carefully examined to see whether 
 any possible conditional force in quotiens 1 has affected the choice 
 of quis and aliquis in collocation, as in the ordinary use after si. 
 But in both the earlier authors and the better late ones aliquis is 
 the favorite, whether as nominative, or accusative or other oblique 
 cases, in both pronominal and adjectival use. 
 
 The greater exactness of the Latin in expressing priority is 
 shown in the 88 examples of future perfect against 66 examples 
 
 1 Cf. the combination si quotiens, Ovid, Trist. 5, 10, 42 ; Sulp. Sever, p. 131, 5 ; 
 p. 204, 4. 
 
On the Usage of Quotiens and Quotienscunque. 39 
 
 of the future. Note also 70 examples of the perfect prior against 
 43 of the perfect contemporary. 1 
 
 The total number (97) of correlatives is strikingly small in com- 
 parison with the number of times quotiens is used without deter- 
 mination. Similarly, quotienscunque (quotiensque) is used 112 times 
 without determination, and but ten times with totiens. The expla- 
 nation for this lies apparently in the formality or stiffness of this 
 particular correlative. Cicero, the Senecas and Ovid are about the 
 only classical authors with more than a few sporadic examples. 
 Only something like the stiffness of the expression must have kept 
 the Roman, fond as he was of antitheses and parallelisms in both 
 prose and poetry, from making more extensive use of it here. 
 Other correlatives, especially quantus . . tantus, quails . . tails, con- 
 stantly recur. 2 
 
 One instance of quotiens . . totiens was found in an inscription, 
 CIL. II, 3367 quotiens poto totiens propino. 
 
 Interesting .variations in the order of correlatives occur. So for 
 example Ovid uses tot . . quot 14 times, quot . . tot six, totidem . . quot 
 six; quot . . totidem once. In the poets metrical considerations nat- 
 urally affect the order. In general, however, the relative might be 
 expected to precede. The examples show the order quotiens . . totiens 
 53 times, totiens . . quotiens 44 times. With quotienscunque , totiens 
 invariably follows. 
 
 An emphatic tune is sometimes substituted for totiens, especially 
 where quotiens has rather the force of cum. Cf. for quotiens . . tune, 
 Yaler. Flacc. 8, 364; Gell. 14, 7, 4; Mar. Victorin. vi (K.) 217, 
 17 ; Charis. n (K.) 85, 13 ; Diom. I (K.) 322, 11 ; Claud. Claudian. 
 de Rapt. Proserp. 1, 157; Sid. Apoll. Epp. 8, 14, 4; Boeth. Mus. 
 1, 28 (p. 220, 3 Fried.) ; 2, 13 (243, 5) nine examples. For tune . . 
 quotiens, Vel. Long. VII (K.) 63, 6 ; Sil. Ital. 15, 113 ; Juv. 14, 21 
 
 1 The aoristic perfect has not been treated separately ; but few examples occur, 
 cf. Juv. 3, 40; Mart. 5, 72; and associated in Claudian. in Eutrop. 1, 239. 
 
 In another class, the present (= fut.) associated with a future has been treated 
 as contemporary ; cf. Col. 6, 17, 9 ; Sen. de Ben. 4, 7, 1 ; Stat. Achill. 1, 909 ; 
 Hygin. Grom. 43; Mar. Victorin. vi(K.) 10, 10; 199, 14; Mart. Capell. 7, 792; 
 Prise, ii (K.) 303, 6. See also, Werth, de Terentiani Sermone, Jahrb. Spbd. 
 xxn, pp. 345, 369. 
 
 8 Cf. Draeger, n, 520 ff.; and Weise, Char. d. lat. Sprache, Leipz. 1891, p. 19. 
 
40 On the Usage of Quotiens and Quotienscunque. 
 
 Terent. Maur. vi (K.) Lit. 199; Mar. Victoria, vi (K.) 83, 12; 
 Charis. I (K.) 173, 9 ; Ennod. p. 454, 18 ; Carm. 1, 9, 62 ; 72 ; 
 Digest. Callist. IV, 2, 13; Ulp. VII, 1, 25, 7; Cassiod. p. 161, 
 24 12 examples. The balance in favor of the latter class, 
 though not large, would seem to indicate that when quotiens pre- 
 cedes, tune is less likely to be substituted for totiens. 
 
 Sometimes, as in the case of other correlatives, the forms do not 
 correspond ; compare for illustration tanta . . quam, Ter. Hec. 
 416; Verg. Aen. 6, 352; frequent in Livy, 26, 1, 3; 37, 51, 9, 
 etc., Tac. Dial 6 (see Gudeman's note) ; quanta . . tarn, Lucret. 5, 
 463. Such examples, however, are not frequent. 1 The following 
 instances for quotiens have been noted : tot . . quotiens, Sail. Hist. 
 2, 98, 1 (Maur.) ; totiens . . quot, Livy 2, 13, 2 ; Plin. Nat. Hist. 
 26, 64; 36, 101; Apul. p. 324 ad fin. (Hild.); quot . . totiens, 
 Yaler. Flacc. 1, 580; Boeth. Arith. 2, 4 (p. 87, 4 Fried.). 
 
 While quotienscunque belongs to prose, with the unimportant 
 exception of two examples in Plautus, it cannot be limited to any 
 one sphere. The 122 examples found (counting here the ten ex- 
 amples of quotiensque) are scattered through a wide range of 
 authors. "With Plautus, Cato, Nepos, Vitruv., Auct. Bell. Afric., 
 Col., CeJs., Petron., Fronto, Tertull., Silvia, Ammianus, Gre- 
 gory, and other similar names in the list, there would seem at first 
 glance to be a distinct tendency toward vulgar usage. Nepos 2 has 
 only the compound, four examples, while Petronius 3 has five ex- 
 amples against a single quotiens; Vitruvius 4 and Fronto have each 
 a proportion of two to one, but the number of examples is small. 
 In many of the other authors quotiens outnumbers the compound, 
 while in such an author as Apuleius only quotiens is found. Cicero 
 even seems to prefer quotienscunque in the orations and rhetorical 
 works, though in a large number of these examples it is evident 
 that he aimed purposely at the additional emphasis of the compound ; 
 cf. Verr. 4, 26, 57; Cat. 1, 5, 4; Phil. 5, 8, 21 ; Dom. 78, etc. 
 On the other hand, in the letters quotiens is used nine times (three 
 times, however, in anaphora, ad Att. 1, 14, 13), quotienscunque six 
 times ; cf. again the special emphasis in such an example as ad Fam. 
 
 1 Cf. Draeger, n, 521. 2 Teuff. g 198. 
 
 3 Durch und durch vulgar, Wolff, Philol. xxxiv, 145. 
 4 Miodonski, ALL. vm, 146. 
 
On the Usage oj Quotiens and Quotienscunque. 41 
 
 16, 11, 3 tu etiam atque etiam cura, ut valeas litterasque ad me 
 mittas, quotienscunque habebis, cui des. It is safe to conclude that 
 Cicero merely used the compound with greater exactness than is 
 found in most of his successors. That the simple and the com- 
 pound relative were not always carefully distinguished has been 
 shown in Part I, but the comparatively small number of com- 
 pounds in the language is due not so much to this inaccuracy as to 
 the increasing number of present indicative examples with the 
 simple relative having a general force. 
 
 The summary thus far has dealt chiefly with indicative examples, 
 and their classification has been relatively simple. But in turning 
 to the subjunctive usage much greater difficulty is encountered. 
 Omitting the interrogative subjunctives, as indirect and therefore 
 outside of the present discussion, we find in the various relative 
 classes approximately 200 subjunctive examples against five times 
 as many of the indicative. Of these approximately 200 examples, 
 considerably over 50 per cent, represent uses of the mode common 
 in all periods, such as by attraction, or in the subordinate clause 
 of indirect discourse, while about 86 examples are of the so-called 
 iterative usage. Yet with the exception of fewer than a half dozen 
 examples, as listed below, these 86 examples are found only with 
 the simple relative and the proportion then becomes practically 
 one-half of all the subjunctives in this class. In thus narrowing 
 the count, however, to a single category it will be observed that 
 the proportion given is deceptive ; and after all, the vital question 
 is why other possible classes do not show numerous examples of 
 the iterative, and why with the simple relative the proportion of 
 indicative examples is so overwhelmingly large that reasonable 
 expectations, aroused by the natural force of quotiens and encour- 
 aged by Draeger and other grammarians, have been disappointed. 
 Does the meaning of quotiens naturally invite the iterative subjunc- 
 tive, or is the point of view merely a summing up in retrospect l 
 with the indicative ? 
 
 An unqualified assent cannot be given to either alternative. The 
 usage of the language in a large measure apparently contradicts the 
 
 x For the phraseology, cf. Gildersleeve-Lodge. Lat. Gram. 567, note, in 
 remarks on the indicative vs. the subjunctive in general for iterative sentences. 
 
42 On the Usage of Quotiens and Quotienscunque. 
 
 first, while on the other hand the retention of the indicative so 
 generally in iterative tenses after the iterative subjunctive had 
 become well established may be due in part to the conservative 
 influence of an overwhelming growth in the use of the present 
 indicative with general conditional force. This growth, as already 
 shown above, served materially to restrict the use of quotienscum- 
 que in later Latin, and in the present connection it is perfectly con- 
 ceivable that it helped even in other tenses to hold the indicative 
 in line. Especially is this true where there could be little or no 
 idea of description, characterization or other such connotation to 
 facilitate the extension of the subjunctive. 1 
 
 1 Prof. Hale has even maintained, in connection with cum, that the subjunctive 
 for repeated action in Latin is simply an extension of the ordinary use of the 
 cum-clause of situation. " The subjunctive of repeated action is generally believed 
 to appear first in the imperfect and pluperfect tenses, and for a considerable time 
 to be confined to these tenses. This is precisely what would be expected if the 
 change is due to the influence of the narrative guom-clause, itself always in the 
 pluperfect or imperfect. But if the change were due to the idea of potentiality, 
 there would be no reason why that idea should not take effect in the present or 
 perfect as well ; and, similarly, if the change were due to Greek influence, the 
 use of the subjunctive in secondary tenses under the model of the Greek general- 
 izing optative would be inseparable from the use of the subjunctive in primary 
 tenses under the model of the Greek generalizing subjunctive." The Cum 
 Constructions, n (1889), p. 237. Cf. also Brenous, Etude sur les Hellenismes dans 
 la Syntaxe latine, Paris, 1895, p. 362 ff. The initial idea of doppelte Modalitdt is 
 pushed to its extreme by J. B. Sturm, Ueber iterative Satzgef iige im Lateinischen 
 (Program) Speier, 1890/91, in which examples are classed as iterative-causal, 
 iterative-concessive, iterative-conditional (potential), etc. 
 
 M. Bonnet, Le Subjonctif de Re'pe'tition, Revue de Philologie, vm (1884), p. 
 75, had already advanced the idea that the so-called iterative subjunctive instead 
 of marking the idea of repetition, as claimed in some quarters (though expressly 
 denied by Kiihner, Ausf. Gram. d. lat. Sprache, n, 182, 9), was even used in 
 spite of that idea : On doit trouver pour le moins bizarre, s'il est dans la nature 
 du subjonctif de marquer l'ide"e de re'pe'tition, que les meilleurs e*crivains semblent 
 e*viter ce mode quand ils veulent exprimer cette ide"e .... Les subjonctifs de cette 
 fespece peuvent et doivent s'expliquer de meme que s'il s'agissait de faits isole"s et 
 non re'pe'te's. Le subjonctif est employe", dans ces exemples comme ailleurs, parce 
 que 1'action de la proposition subordone"e et celle de la principale, au lieu 
 d'etre simplement juxtaposes, sont pre'sente'es dans leur relation logique .... 
 L'indicatif s'est maintenu plus longtemps, comme le constate la rSgle des 
 anciennes grammairs, parce qu'il etait naturel de conside'rer surtout, dans 
 une action re"pe*tee, le simple paralle'lisme entre cette action et un autre fait, 
 egalement re'pe'te'. L'e*crivain, bien souvent, restait libre d'e'tablier une con- 
 nexite" plus e"troite entre les deux series de faits .... Mais, en somme, Fide*e de 
 
* 
 On the Usage of Quotiens and Quotienscunque. 43 
 
 The subjunctive in quotiens clauses was, therefore, not taken up 
 so early as in the cum clauses, nor was its use at any period ever 
 so free. The earliest example for cum cited by Dittmar is Enn. 
 Ann. 294 (M.); examples are then found with some frequency 
 in Caesar and Cicero. Lupus l cites thirteen examples in Nepos ; 
 but it is in the silver and later Latinity of Valerius Maximus, 
 Pliny the Elder, Tacitus, Florus, Suetonius, and the Scriptores 
 Historiae Augustae that the construction is most at home; cf. 
 Schmalz, 3 311. Si in the iterative clause appears at least as early 
 as Catullus 84, 1. 
 
 Chommoda dicebat, si quando commoda vellet 
 
 Dicere, et insidias Arrius hinsidias, 
 
 where, however, si quando is practically equivalent to cwm, or even 
 quotiens might have been used. By the time of Caesar, Nepos and 
 Cicero other particles and relatives have become associated. At 
 this period, for the sake of safety, all subjunctives with quotiens 
 
 re*pe*tition, loin de motiver Pemploi du subjonctif, a e*te* cause pendant longtemps 
 du maintien de 1'indicatif. Au lieu done d'affirmer que le subjonctif .... sert a 
 marquer Pidee de repetition, on devrait montrer dans quelles conditions et pour 
 quelles raison le subjonctif s'est introduit, dans les exemples dont nous parlons, 
 malgre Videe de repetition. 
 
 On the other hand, Draeger, Hist. Synt. u, p. 574, Schmalz, Lat. Synt. 3 \\ 274, 
 311 and Riemann, Synt. lat., pp. 331 and 353-355, hold the widely accepted view 
 that this use of the subjunctive is due to the influence of the Greek optative. 
 Dittmar, Stud. z. lat. Moduslehre, Leipz. 1897, p. 146, is less positive, though 
 he seems to share die Meinung . . . . , der Konjunktiv sei in iterativen Satzen 
 eigentlich etwas Unlateinisches oder wenigstens etwas Unklassisches. Dittmar's 
 list, however, is only illustrative and some of his earlier examples cannot be 
 viewed without reasonable doubt. 
 
 The whole usage while due to Greek influence originally, and even directly so 
 at times later on, seems to have been modified in its possibilities by developments 
 taking place in the Latin itself. It thus became early identified with cum, and 
 by consequence with the narrative tenses, though it was restricted to neither the 
 one nor the other. The transition, however, to ut, ubi and other general rela- 
 tives with the imperfect and the pluperfect was more readily effected than was 
 the extension of the present and the perfect tenses, where the indefinite second 
 person and the potential construction did partial service. That another force is 
 often present, especially in earlier examples, where the subjunctive is associated 
 with the idea of repetition is true (cf. Mdvg. Klein. Phil. Schriften, Leipz. 1875, 
 p. 233), but M. Bonnet has carried his statement too far in saying that each sub- 
 junctive should be explained in the same manner as for a single action ; cf. Rie- 
 mann, Etudes sur Tite-Live, 2 Paris, 1885, p. 294, n. 1. 
 
 1 Sprachgebr. d. Corn. Nepos, Berl. 1876, p. 153. 
 
44 On the Usage of Quotiens and Quotienscunque. 
 
 have been accounted for by every other means possible, even when 
 the idea of repeated action was clearly in the sentence ; as in Nepos 
 
 7, 3, 5 quare fiebat ut omnium oculos, quotienscunque in publicum 
 prodisset, 1 ad se converteret ; cf. similar cases of attraction in Livy 
 
 8, 35, 10; 23, 32, 3; Curt. 10, 4, 3; Plin. Nat. Hist. 7, 141; 
 Quint. 2, 4, 27; Frontin. Aq. 87; Tac. Hist. 1, 24, 5; and even 
 in Sueton. Aug. 44; 57; de Vir. 111. p. 138, 17 (Reiff.), although 
 the strong tendency toward the so-called iterative subjunctive in 
 Sueton. is pointed out by Draeger n, p> 584, and the subjunctive 
 in the passages cited is not necessarily due wholly to attraction ; 
 cf. Sen. Controv. 1, 2, 15 iubemur ut, quotiens possurnus. de omni- 
 bus legis verbis controversiam faciamus; Quintil 1, 11, 8 curabit 
 . . . ut, quotiens exclamaudum erit, lateris conatus sit ille non cap- 
 itis ; Plin. Epp. 6, 12, 5 ; Jul. Capitol. Yer. 5, 3 ; Serv. Aen. 3, 
 241 ; 384. 2 
 
 Clearer cases of subjunctives due to indirect discourse, though 
 the idea of repeated action is present, may be seen in Livy 31, 44, 
 6 sacerdotes publicos, quotienscunque pro populo Atheniensi soci- 
 isque, exercitibus et classibus eorurn precarentur, totiens detestari 
 atque exsecrari Philippum ; 39, 39, 11 gratias populo Romano egit, 
 quod tanto studio, quotienscunque declarandae voluntatis potestas 
 facta esset, praetorem se voluisset facere; and, similarly, Petron. 
 117; Plin. Nat. Hist. 8, 14 ; Tac. Ann. 2, 34, 23; 3, 65, 10; 71, 
 11 ; Sueton. Calig. 4 ; Ner. 28 ; Galba 10. 
 
 In Curtius 9, 6, 4 mos erat principibus amicorum et custodibus 
 corporis excubare ante praetorium, quotiens ad versa regi valetudo 
 incidisset, the subjunctive depends upon the infinitive construction, 
 as stated by Draeger, though the example is classed by him as an 
 iterative. With this may be compared Pseudo-Quintil. p. 377, 
 21 (Ritter) consuetudine iudiciorum consequens est, q. aliqua propria 
 actio in rem non detur, uti proxima et simili ; Tac. Ann. 12, 47, 6 
 
 1 Lupus, p. 155, classes this as iterative. 
 
 2 Many later examples of non-attraction are omitted, especially after the lan- 
 guage has reached the point where it often chooses to fare without the " luxury 
 of a subjunctive." Cf. conspicuous examples in Amm. Marc. 20, 5, 4 retexere 
 superfl uum puto, q. hieme cruda . . . indomitos antea . . . reppulimus Alamannos ; 
 Cassiod. vii (K.) 161, 15 Donatus dicit, q. u et i seu sibi seu aliis vocalibus prae- 
 ponuntur, loco habendas esse consonantium. 
 
On the Usage of Quotiens and Quotienscunque. 45 
 
 mos est regibus, q. in societatem coeant, implicare dextras. The 
 last example is followed by a clear case of the perfect subjunctive 
 in the same usage with ubi: mox ubi sanguis in artus extremos 
 suffuderit, levi ictu cruorem eliciunt. 1 
 
 One-third, or more, of the examples to be quoted are found in 
 Tacitus and Suetonius, with by far the larger number in the latter. 
 Prior to Tacitus the earliest example is Livy 5, 54, 3, fatebor vobis 
 . . . cum abessem, quotienscunque patria in mentem veniret, haec 
 omnia occurrebant, colles campique et Tiberis, etc. Here the com- 
 pound relative emphasizes the iteration, although the subjunctive 
 in the clause might be due to Livy's having had in mind haec omnia 
 occurrisse to follow fatebor. Certainly at the end of the sentence, 
 which continues, et hoc coelum sub quo natus educatusque essem, 
 the subjunctive is due to the thought in the mind of the speaker 
 Camillus. Other instances of the imperfect are Sen. Controv. 4 
 praef. 7 q. velles eandem rem et quamdiu velles 2 diceret, aliis totiens 
 figuris (sc. dicebat); Quintil. 1, 7, 20, q. S littera media vocalium 
 longarum . . . esset, geminabatur; 2, 4, 13 q. eandem materiam 
 rursus a me tractatam scribere de integro iuberem ; 9, 4, 38 q. ul- 
 tima esset (sc. littera S) : Stat. Silv. 5, 3, 215 Latios q. ego carmine 
 patres Mulcerem ; Frontin. Aq. 12 q. siccitates egerent auxilio ; 
 Strat. 2, 13, 7 q. prima luce moveret; ibid. 3, 3, 4 q. acie decerta- 
 retur ; and it is not surprising to find that the younger Pliny has 
 the construction already allowed by his master Quintilian ; 3 cf. 
 Epp. 3, 16, 4 q. cubiculum eius intraret, vivere filium . . . simu- 
 labat; Pan. 13 q. ... clipeo gravior ictus incideret ; ibid. 42 q. 
 quisque similes principi servos haberet. 
 
 For the pluperfect up to the time of Tacitus, but three examples 
 are to be cited: Sen. Controv. 10 praef. 13 hi (sc. rhetores) q. 
 conflixissent, penes Gallionem palma; Pliu. Nat. Hist. 23, 56 ab 
 aspide percussus utrem aceti ferens, q. deposuisset, sentiebat ictum ; 
 Quintil. 6, 1, 40 q. respexisset patron us. 
 
 Examples of the present and the perfect subjunctives of itera- 
 
 1 Dittmar, p. 149, cites both the examples from Tacitus, Prof. Hale, p. 239, 
 only the latter. 
 
 *The post potential force is of course still felt. 
 3 Cf. Plin. Epp. 2, 14, 10, in another connection. 
 
46 On the Usage of Quotiens and Quotienscunque. 
 
 tion l are neither so numerous as the other two tenses, nor so early. 
 For the period selected above, the following instances may be cited : 
 Plin. Nat. Hist. 17, 129 atque ita alia arbore ex eadem iuvenescente 
 iterumque et q. opus sit ut aevis eadem oliveta constent; 18, 350 
 sunt et ipsius lunae vin articuli, q. in angulos solis incidat ; 34, 
 122 q. opus sit molliri vim eius, mel adspergitur; Plin. Epp. 8, 
 20, 1 quod differimus tamquam saepe visuri quod datur videre, q. 
 velis cernere. Sen. Epp. 19, 1, 17 hoc itaque ipse mihi dico, q. 
 tale aliquid praestrinxerit oculos meos; Plin. Nat. Hist. 18, 259 
 senescunt prata restituique debent . . . , praeterea q. secta sint 
 siciliri ; 23, 103 fit et oleum ex his quod melinum vocavimus, q. 
 non fuerint in umidis nata; 36, 134 palmati circa Mundam in 
 Hispania . . . reperiuntur, idque q. fregeris, etc. 
 
 Later writers with the exception of Suetonius have held rather 
 to the classical usage with the indicative. In Tacitus an interest- 
 ing development presents itself. The early Dialogus, Agricola 
 and Germania, which show most strongly the influence of classical 
 models, contain only indicative examples ; the Historiae show but 
 two examples 2 1, 66, 19 q. pecuniae materia deesset, stupris et 
 adulteriis exorabatur, and 3, 83, 3 q. pars altera inclinasset, while 
 in the Annales, where the author is most under the influence of 
 contemporary usage, six instances are found : 2, 2, 12 q. per urbes 
 
 1 The use of these tenses in this manner has not always been granted, especially 
 since Madvig's note denying the usage in the perfect in Cic. de Fin. 5, 41 : Libra- 
 rii tamen, quibus post cum coniunctivus modus magis placebat, non raro aut u 
 litteram in mutarunt aut notam illam qua er significatur, addiderunt . . . edito- 
 resque habuerunt obsequentes. Madvig, therefore, in his second edition of Livy r 
 adopted Wesenberg's fuit for fuerit of the MSS. in 1, 32, 8 quicunque ei primus 
 vir obvius fuerit. The MS. reading, however, is stoutly defended by Frizell, 
 Epilegg. ad T. Liv. lib. I, Upsala, 1881, p. 49, who adds, with reference to Livy's 
 whole usage, iam per se credibile atque ad fidem pronum est etiam in perfectis et 
 praesentibus eum ab usu coniunctivi non abhoruisse. In his review of Frizell, 
 Kevue critique, 1881, u, p. 89, Riemann denies this usage for the particular ex- 
 ample and in general. Later, however, he withdraws his general objection : J'ai 
 combattu cette opinion, peut-etre d'une maniere trop absolue . . . je crois aujourd' 
 hui que cet emploi du subjonctif dont parle Frizell a pu en effet exister en latin. 
 Etudes sur Tite-Live, Paris, 1885, p. 297 ; cf. also Draeger n, p. 573, and Ditt- 
 mar, p. 149, Anm. 
 
 2 Cf. Hist. 1; 10, 8 nimiae voluptates, cum vacaret; q. se expedierat, magnae 
 virtutes, where the indicative is retained even in violation of symmetry. See 
 Wolff., abstract, 1. 1. p. 399. 
 
On the Usage of Quotiens and Quotienseunque. 47 
 
 incederet; 4, 3, 19 promptus eloquebatur, q. subveniret; 6, 10, 13 
 q. necessitas ingrueret; 15, 13 q. curiam ingrederetur : 21, 1 q. 
 super tali negotio consultaret, edita domus parte .... utebatur ; 
 13, 3, 16 q. meditata dissereret. 
 
 Further examples of the two common tenses may be cited by 
 reference alone. 
 
 Imperfect Terent. Scaur, vn (K.) 25, 18 ; Sueton. Jul. 26 ; 28 ; 
 Aug. 21 ; 38 ; 41 ; 45 ; 56 ; 77 ; 82 ; Tib. 19 ; Calig. 7 (quotiens- 
 que) ; 33 ; 35 ; Claud. 8 ; Nero 15 ; 27 ; Otho 4 ; Domit. 2 ; 21 ; de 
 Vir. 111. p. 24, 3 (Reiff.) ; 61, 4; Ael. Lamprid. Commod. 11, 
 11 ; Mar. Victorin. vi (K.) 12, 17 ; Jul. Paris ad Valer. Max. 8, 
 10, 1 ; Claudian. Carm. Min. 30, 97. 
 
 Pluperfect Sueton. Aug. 99; Calig. 36; Claud. 42; Plin. 
 (Med.) 3, 30 ad fin ; Prudent. Perist. 10, 758 ; Serv. ad Aen. 11, 
 517; Claudian. de in. Cons. Honor. 25; de vn. C. H. 82; in 
 Kufin. 2, 75; Eugipp. Vit. Sanct. Sever, p. 33, 1. 
 
 In this second period but three examples of the present sub- 
 junctive in iterative clauses have been noted : Terent. Scaur, vn 
 (K.) 11, 9 per immutationem, ut cum at coniunctionem per d scri- 
 bunt et eandem, q. praepositio sit, per t ; Sid. Apoll. Epp. 1, 7, 
 7 q. desperatum furor arripiat ; Prise, n (K.) 37, 21 in Graecis 
 vero, q. huiuscemodi fiat apud nos diaeresis paenultimae syllabae, 
 i pro duplici consonante accipitur. 
 
 Examples of the perfect, however, are somewhat more numerous, 
 though sometimes, as in Cass. Felix, for example, it is difficult to 
 distinguish the future perfect from examples belonging properly in 
 this category. With the exception of Cassius Felix and Enno- 
 dius, the examples are sporadic. Terent. Scaurus again has one 
 instance, vn (K.) 25, 19 q. ergo a liquida littera sequens coeperit 
 verbum .... primam litteram geminant ; cf. further Ampel. 8, 
 17 sed q. ventus aut pluvia fuerit non movet; Cypr. App. p. 33, 7 
 sic q. tempestatibus diu nauta fatigatus litora tuta cpntigerit, felici- 
 tatem suam de perpesso periculo ducit; Lactant. p. 576, 17 (L.-B. 1) 
 totiens sacrificat, q. bonum aliquid ac pium fecerit ; Diom. I (K.) 
 427, 28 q. inter duas vel plures consonantes posita sit ; Serv. Eel. 
 1, 8 unde necesse est, pastores totiens aras imbuere, q. mutaverint 
 pascua; Cass. Felix 38 contingit frequentur . . . , q. plus frigoris 
 quam possit natura corporis ferre sustinuerint laborantes ; 60 con- 
 
48 On the Usage of Quotiens and Quolienscunque. 
 
 tingunt frequenter . . . , q. glandulosa fuerint ipsa inguina ; 61 et 
 fit q. tumor epatis et splenis apparuerit ; 72 contingit vero hoc, q, 
 ipsa violentia insurrexerint aniinalia; Ennod. p. 69, 4 condicio, 
 quae q. desideriis aliquo sapore respondent, mox et in foribus con- 
 cessa permutat; 171, 8 q. votiva res repente contigerit, pretiuni de 
 ipsa temporis brevitate sortitur; 437, 23 blandimentum est, q. 
 laborem gloria ingesta praecesserit ; 471, 16 reducitur puritas, q. 
 generalitatis oculos expiaveris cruce noxiorum; Carm. 1, 8 praef. 
 1 non ponit fistulam q. rusticura pecus agrestis pastor iuspexerit ; 
 Cypr. Gall. Exod. 980 q. sementem sparseris agro. 
 

 MAR 4 , 948 
 
 1 1954 
 
YC 00523