STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. PETRONIUS, POGGIO, AND JOHN OF SALISBURY By Evan T. Sage The text of Petronius rests on peculiar foundations. The Cena is preserved entire in a single MS of relatively late date and uncertain value. Other parts are found only in late sources — editions of the sixteenth century based on MSS now lost, the composite MS of Scaliger, or the late and interpolated Monacensis. Under these circumstances, the history of the tradition is of value, the examina- tion and evaluation of the work of the sixteenth-century scholars of some importance. The editions from Goldast (1610) to Burmann (1709 and 1743) are full of real or alleged MSS readings. The task of sifting out this mass of material was well, if not completely, per- formed by Biicheler, 1 as its omission is the most serious defect in Beck's otherwise valuable work. 2 Despite the difficulties in the way, the study of these commentaries would be interesting and useful. For the present, I have had to depend on the reports of Biicheler and Beck — neither entirely free from errors — and my own meager apparatus, and therefore, in so far as my conclusions are based on MSS readings alone, I propose them with some reservations. I shall confine myself at this time to the relations of John of Salis- bury and Poggio to Petronius, considering three MSS, Bern. Lit. 357 (B), Paris. 7989 (A; the famous Traguriensis), and Laur. 47, 31 (D). Incidentally other MSS will be involved. On May 28, 1423, Poggio wrote from Rome to Niccoli in Florence as follows : Epistt., II, 3 (ed. Tonelli) : Allatus est mihi ex Colonia XV liber Petronii Arbitri, quern curavi transcribendum modo cum iliac iter feci. Mittas ad me oro bucolicam Calpurnii, et particulam Petronii quas misi tibi ex Britannia. In connection with this letter should be put corresponding facts about A. First, A is a compound MS, the two parts coming directly 1 In the editio maior of 1862. References to Biicheler are to this edition if not otherwise stated. 2 The MSS of the Satyricon of Petronius Arbiter, Described and Collated, Cambridge, 1863. [Classical Philology XI, January, 1916] 11 12 Evan T. Sage or remotely from different sources. Chap. 55 is found in both, in slightly different forms. A more conclusive proof is the fact that the Cena was not written at the same time as the preceding Vulgate part of the MS. The ink is different, and the paper from f. 209, where the first new quaternion of the Cena begins, is of slightly different quality. The hand is the same. These facts, revealed by the MS itself, have not all been noted in connection with Petronius before. I do not believe, however, that the interval between the copying of the two parts was very long. Secondly, A was written about the time of Poggio's second discovery. The date November 20, 1423, appears on fol. 179, the last page of Catullus. 1 Thirdly, A contains fragments of books xv and xvi; cf. the incipit and explicit subscriptions of the Vulgate portion. The two MSS may then have been combined under Poggio's direction between May and November, 1423, and A may be a copy of the combination. Then the con- clusion is easy that the particula from Britain is the Cena, for the Cena was known there (to John of Salisbury) and cannot be shown to have been known elsewhere in Europe during the Middle Ages. This is as a matter of fact the opinion of A. C. Clark. 2 However, a re-examination of the evidence has caused me to adopt a different interpretation. 3 It is clear that John knew the Cena and other parts of Petronius as well. Let us test first the possibility that the British particula is the Cena. It should be remembered that there is no necessary connection between John's assumed MS and Poggio's particula, but if both had the Cena, and no one else did, this is the obvious place to start. The first question, then, is whether any of the immediate ancestors of A can be the MS used by John. Only two passages will help us. The first is as follows (John, Policr. viii. 7) : " Coenam Trimalchionis 1 1 accept the suggestion of Sabbadini (Riv. di FU., XXXIX, 250), that the note containing the date is not a true subscription, but merely a memorandum of the number of verses (perc=percensui). It correct, this establishes a terminus ante quern for the MS. » Class. Rev., XXII, 178 f . This opinion is accepted by Webb in his recent edition of the Policraticus of John, and by Gaselee, Cod. Trag., pp. 9 f., though with a reserva- tion. Cf. UUman, Class. Phil., VI, 289. • Sabbadini (Riv. di FU., XXXIX, 251) has expressed his dissent from Clark's opinion, but without discussion. My disagreement from Sabbadini is only apparent. A itself may very well have been written in the neighborhood of Milan, or at least by a copyist using that script. Peteonius, Poggio, and John of Salisbuby 13 apud Petronium si potes, ingredere, et porcum sic gravidari posse miraberis, nisi forte admirationem multiplex, ignota et inaudita luxuria tollat." This seems to show that John had, or had access to, a MS, and realized that it was difficult to find others, but tells us nothing about that MS. 1 The other passage is a little more helpful. John recounts the story of the malleable glass (Policr. iv. 5 = Petron. 51) with more details than A gives, but these could be easily sup- plied. 2 There are two differences in reading. In sec. 4 (of Petronius) some of the MSS of John read marculum; others, with A, martiolum. It is therefore impossible to tell what John wrote or what he saw in his MS. Just below A reads coleum, which may be right; John reads coelum, which is easier; the editors read solium or solemn. On such slight evidence it is of course impossible to decide whether John was using a MS of the A type. It should be stated that at best A can be only a copy of the particula, and a copy of a copy of the Cologne MS, and no conclusion can be drawn as to the age of the particula if it is the archetype of A or part of A. 3 I shall turn aside at this point to discuss the relation of A, as the visible representative of its family, to John for the Vulgate, for it is possible that the particula is all or part of the Vulgate portion of the text. I shall give specimen readings from A, B, D, J(ohn), Bch. *■ 5 (Biicheler's first and fifth editions, respectively. The expo- nents are used only in case of change of readings). References are to chapter and section. a) 2, 1, haec ionium JBch. 1 haec ABDBch. 6 6) 80, 9, 2, mobile BJBch. nobik AD c) 80, 9, 5, agit BJBch. ait AD d) 111, 4, commodabat BJBch. commendabat AD e) 111, 4, quotiens B. quotienscunque Bch. -cumque J. om. AD /) 111, 5, adfulsisse BBch. 1 aff- Bch. 6 J. affluxisse AD g) 111, 10, vini cett. humi J 1 Cf. Schaarschmidt, Rh. Mus., XIV, 209, n. 5, where the same inference was drawn. 1 Isidore (Orig. xvi. 16. 6) tells the same story with no mention of Petronius. John puts the narrative into the mouth of Trimalchio, and in addition, in one impor- tant detail, follows Petronius and not Isidore where the two latter disagree. The sentence containing coleum is omitted in Isidore's briefer account. Isidore, then, is not John's source. » Keller, Rh. Mus., XVI, 532, has argued on other grounds that the archetype of A was a MS of the ninth or tenth century. Gaselee, op. til., p. 11, thinks it was "early." 14 Evan T. Sage h) 111, 12, credis sentire sepultos BJBch. 6 credis curare sepultos Bch. 1 sepultos credis sentire AD i) 111, 12, ammonere BBch. 1 adm-Bch. 6 J. commovere AD i) 111, 13, cum BJBch. dum AD A) 1 12, 6, modo B J. er^o ADBch. These are perhaps sufficient to show that A is not closely related to the MS used by John. In fact, as I shall have occasion to repeat later, a MS of the B type is closer to his text than any other existing MS. Fortunately there is other evidence on the particula. The history of Poggio's visit to Britain may serve to throw light on the question. He was in Britain from 1418 until the spring of 1423, when he returned via the Rhine, finding the Cologne MS on the way. A very important letter in this connection, the full meaning of which has not been recognized hitherto, is the following, written to Niccoli from London, June 13, 1420: Epistt. I, 7, T: De Petronio Arbitro quod scire cupis quid tractet, lege Macrobii prmcipium super somnium Scipionis, ubi enumerans genera fabularum dicit, in eis esse argumenta fictis amatorum casibus referta, quibus multum se Arbiter exercuit. Est autem homo gravis versu et constans prosa, et ut conicio, paulo post tempora Augusti. The interpretation of this letter seems to be this: An earlier letter announced to Niccoli the discovery of Petronius, an author unknown to Niccoli at the time. In answer to an inquiry from him, Poggio gave the description and comment just quoted. 1 Let us now see what bearing this letter has on the question of the particula, to which it must refer. There are several reasons why, in my judgment, the particula cannot be the Cena. First, and weakest, particula does not seem to be an appropriate word for even so small a work as the Cena. It would seem to fit better such a section as the Civil War or the Widow of Ephesus. It may be noted that there are MSS con- taining only short sections like these. There are, however, serious objections to taking any of these as the particula. Secondly, the comment in Macrobius is not applicable to the Cena, and therefore has no point if the particula is the Cena. The characteristic of Petronius to which Macrobius refers is his use of amatory adventures. Now if any part of the Saturae is free from them, it is the Cena. 1 Cf. Sabbadini, Le Scoperte del Codici Latini e Greet ne' Secoli XIV e XV, p. 83. Petronitjs, Poggio, and John of Salisbury 15 We shall have to say, then, either that Poggio's quotation of Macro- bius is irrelevant or that he is not describing the Cena. The second is of course the easier alternative. It is conceivable that Poggio quoted the passage merely to display the wide range of his reading, but he would hardly need to show off in that fashion to Niccoli. The other parts of the Satires give ample reason for Macrobius' words. Thirdly, Poggio expresses a judgment on the author's style. He is gravis versu et constans prosa. In the Cena there are altogether 23 verses of poetry. Of these, 16 are in one passage which is or pur- ports to be quoted from Publilius Syrus. One more is quoted from Virgil. In addition there are two passages of three lines each put into the mouth of Trimalchio (eheu nos miseros, etc., 34. 10, and quod non expectes, etc., 55. 3; the last line of this may not be a verse, but if it is not my argument is all the stronger). One may well doubt whether Poggio would have expressed any opinion on such evidence. If he had, I am inclined to think it would not have been gravis versu. It is possible, of course, that Poggio believed the ostensible quotation from Publilius Syrus to be original with Petro- nius, but there is no reason to think so. There is plenty of original verse elsewhere. The longest poem is the Civil War, of 295 verses. Fourthly, Poggio places Petronius shortly after the Augustan age, apparently on the basis of style. The language of Encolpius is rela- tively pure, and the colloquialisms which offended some of the later humanists are most numerous in the Cena. Clark's suggestion that Poggio's silence with regard to the Cena is due to his Ciceronian prejudices, and his consequent objection to Petronius' style, seems, in the light of this letter, untenable. For these reasons I am unable to admit that Poggio brought the Cena from Britain. He never refers to the Cena, and the only quotation from any part of the Satires that I have found is in a letter of 1435. 1 Unfortunately, this passage (from 94. 1) is found in nearly all the MSS, including at least two florilegia, as well as in the Speculum of Vincentius Belua- censis and the Zophilogium of Iacobus Magnus. In Epistt. IV, 2, T, dated December, 1429, Poggio complained that Niccoli had kept Petronius seven years or more. Clark takes this to refer to the Cologne MS. It seems to me more probable that 1 Quoted in Shepherd's Life of Poggio Bracciolini, p. 291. 16 Evan T. Sage it refers to the particula, and if so we might expect to find a copy of the particula in Niccoli's library. D, however, which may have been his MS, 1 has no close connection with John's MS, as the readings given above show. This chance of identifying the particula dis- appears. The relation of D and the particula cannot be determined. Where Niccoli got D, if it was his MS, I do not know. I am not yet ready to express an opinion on the exact relation of A and D. The question then remains, What was the particula ? So far as I know, there is no way to identify it. My own feeling is that the poem on the Civil War, with the introductory prose sections, would correspond in bulk to what might reasonably be considered a partic- ula, but this would not justify the characterization quoted from Macrobius. The story of the Widow of Ephesus would suggest it, at least, but contains no verse. Of the MSS now extant, the Paris florilegium (ND; Nostrad. 188 = Paris. 17903), a MS of the thirteenth century, which contains the Widow of Ephesus and one poem of 22 verses in addition to other shorter ones, would more nearly meet the requirements than any other. Without additional evidence I should not be willing to suggest that it was the particula. It does not con- tain the malleable-glass story, and omits other passages quoted by John, but these may have dropped out after his time. On the other hand, the selection of Petronius passages in John is not that which the compilers of florilegia would choose. The stuffed-pig passage, for example, which may refer to Petronius, chap. 41 or chap. 49, one would hardly find among the moralia excerpta so dear to the heart of the maker of the anthologies. The relation between Calpurnius and Petronius remains to be considered. It will be recalled that in the first letter of Poggio quoted above he asks for the return of bucolicam Calpurnii et particu- lam, quas, etc. Just how closely this passage associated Calpurnius and Petronius it is hard to say. At least they were associated for the moment in Poggio's thoughts. Now ND contains extracts from Calpurnius (so does Paris. 7647 for that matter) 2 as well as Petronius. 1 Cf. Wotke, Wiener Studien, XI, 301, note. * Cf., e.g., Manitius, Gesch. d. lat. Lit. d. MA, I, 255, n. 1, and Teuffel, f 306, 1. Protzen, however (De Excerptis Tib., Greifswald, 1869), gives the contents of ND, but does not mention Calpurnius. Apparently ND has the name in the form Scal- purius (Manitius, Philol. aus alien Bibliothekskatalogen, 57, n. 2). Petronius, Poggio, and John of Salisbury 17 As we cannot define what Poggio meant by particula, so it is not cer- tain what he meant by bucolica. Whether the extracts contained in a florilegium would justify that title is open to doubt. Calpurnius and Petronius are found together in another MS, Paris. 8049 (P). The whole matter of the associates of Petronius in the MSS needs further investigation, and, in some cases at least, perhaps no positive results can be obtained until this is done. Let us return to a consideration of the MS from Cologne. No MS now in existence except A preserves any indication of the book division, though there is some external evidence. Fulgentius 1 cites a passage from chap. 20 as from book xiv. The attempt to rule this out as an interpolation merely shifts the responsibility for the numeral to the scribe. The only real reason to doubt its genuineness is the conflict with the subscriptions in A. Biicheler inclines to accept it, and puts the division between books xiv and xv at chap. 26, the beginning of the Cena. Modern editors generally assume that the Cena was book xv or part of book xv. An item in a Fleury lexicon assigns chap. 89 to book xv 2 and Bucheler may then be right in putting the division between books xv and xvi at chap. 99. 3 The remainder of the work may then belong to book xvi. If we begin to reject numerals as corrupt there is no sure place to stop. It is sufficient here to regard the subscriptions in A as genuine tradition. Whether the Fulgentius passage is corrupt or not makes no differ- ence for our present purpose, for even if we have no part of book xiv the Cena might still belong to book xv. It may well be true that in the original form of the work the Cena was book xv (its relative position is fixed with sufficient accuracy by the fragments common to it and the Vulgate), but we need not concern ourselves with that point. The Cena, when removed from its proper place, virtually ceased to be book xv. If, as is conceivable, the Cologne MS had the Cena in its true position, it is hard to say what circumstances would have caused its transposition to the place it occupies in A. 4 It seems 1 Biicheler on Frag. VII. 2 Quoted by Bucheler, ad loc, and often by older editors (from Pierre Daniel, for whom see below) . » Bucheler, p. vii. * My own suspicion is that the Cena was removed bodily from the Saturae com- paratively early, and that someone (in the tradition now represented by A) later 18 Evan T. Sage more natural that, if there is any connection at all, A represents the contents and external form of the Cologne MS. Then for some reason A and its immediate ancestors vanished. A reappeared about 1650, while the others have never been found. It is useless to specu- late on their fate. Though the family ties were strong, this branch alone retained any traces of the book division. Let us review the arguments that favor the establishment of a relation between A and Poggio's Cologne MS. They are (1) the dates of the two MSS; 1 (2) the similarity in book numbers; (3) the fact that of all the MSS known to exist, only these and the source of the Fleury lexicon have preserved indications of the book division. These may be mere coincidences, but the accumulation of three such coincidences is striking. For other reasons, too, it would be desirable to keep A in the circle of Niccoli and Poggio if it were possible. I should not venture the assertion that A was Poggio's MS, but I should not be surprised if it should be some day proved to be at least a near relative. It may be that in the unpublished correspond- ence of the fifteenth century the whole story will be found. Few things are more needed by classical scholars than complete and accessible collections of the letters of the Renaissance scholars. Our particular problem would be simplified if the scribe of A could be identified. 2 Let us return to John of Salisbury. The readings quoted above have shown that AD belong to a branch of the family different from that of the MS John used. I have pointed out the closer resemblance between B and John's MS. Omissions in B make it very doubtful if John used that MS, though it was in existence in his time, and, I believe, accessible to him. (The fact that B does not contain the Cena points in the same direction.) On the other hand, John takes such liberties with his texts, changing the order of words, interpolating words and phrases, etc., that it is somewhat dangerous to insist. brought the parts together in the artificial union seen in A. I do not forget the signs that the two parts of A were not written continuously. Were the two parts first brought together by the scribe of A ? If so, possibly the Cologne MS wa3 the Cena, but the book-number argument is against this conjecture. ■ A letter of Poggio to Niccoli written November 6, 1423 (Epistt., II, 7), shows that Petronius had been copied iam dudum then. 2 For efforts to identify him see Sabbadini, Riv. di Fil., XXXIX, 249 ft. Petronius, Poggio, and John of Salisbury 19 The external history of B points to Fleury, where I suspect that John's acquaintance with Petronius was formed. We have already seen that in this important school Petronius was known. John's quota- tions from Petronius seem confined to the Policraticus, which was written about 1160; in other words, when John was in middle life and still comparatively poor. 1 It is not likely that under these cir- cumstances he would have owned many MSS, especially of rare authors like Petronius. Schaarschmidt 2 has noted that John worked mainly with excerpts, and so I doubt whether he had a complete MS at all. If he had merely a collection of passages taken from a MS he had seen at Fleury, this might even be Poggio's particula, though, as I have shown, this cannot be proved. 3 Various interesting questions concerning this hypothetical Fleury MS now suggest themselves. There were indications of the book division there: did this MS have them? What is its relation to B, to vetus Pithoei, and to vetus Benedictinum (Floriacense) used by Pithoeus ? If John's knowledge of the Cena came from this MS, can any connection between it and the Cologne MS of Poggio be traced ? If it had the Cena and the book numbers, why does B, which seems its best representative, not have them ? I cannot now attempt the solution of the problems thus briefly stated. I am not sure that all can be solved, but their importance for the history of the Petronius tradition is obvious. I stated above my belief that B came from Fleury, but before taking up that point I wish to clear up another matter regard- ing the same MS. B is at present incomplete, lacking, through accident, half (leaves 3-6 inclusive) of the first quaternion of Petronius. Petronius begins on fol. 2v. Mommsen thought he found in Leyden, bound with a volume of Plautus (Voss. xxx), the third and fourth of the missing leaves. Bticheler accepted the identification, and in his editio maior gave the Leyden leaves, which I shall call for convenience Va, all the authority of B. Beck, however, questioned this identification. 4 Among his various grounds was his 1 Diet. Nat. Biog., art. "John of Salisbury." The exact date is debated. » Rh. Mus., XIV, 221. 3 Except for John there is no sure sign of a knowledge of Petronius during the Middle Ages except in France. < Philol., XX, 293 ff. 20 Evan T. Sage belief that the hands were not the same, and he was not even sure that the two leaves of Va were in the same hand. Biicheler replied with some vigor. 1 The most obvious procedure under the circum- stances would be to compare the two hands, but hitherto this simple test has not been applied. I shall not repeat all the arguments in this interesting controversy. I shall merely suggest some additional points that Biicheler might have made. Beck's first point is the omission of a word at the end of the first page of B or the beginning of the first page of Va. At this place Biicheler in his critical apparatus and Beck (The MSS, etc., p. 33; no report on p. 43) both state that the word meditantur (3. 3) has dropped out, prius being the last word in B, quam the first word in Va. Furthermore, between fol. 48p and fol. 47r of Va, the leaves being bound in this order, the word aetate (17. 6) had dropped out. To this Biicheler replied that there was a lacuna in Va in which the word meditantur had stood, though there was no mention of this in his apparatus. My photographs seem to make the situation clear. The corners of both leaves are torn off. Nine lines on 48r, eleven on 48», six on 47r, and five on 47w have suffered more or less loss. The reports of both Biicheler and Beck are so confused here that I shall give later the full readings of Va at the points of damage. It is sufficient for the present to say that ntur preceded by a single stroke that might well be the last part of a may still be read. Beck's second text was the measurement, by the standard of A, of the missing four pages. As a matter of fact, this test is practically worthless: in the first place, he should have chosen a MS more closely related to B, as P; in the second place, we do not know what individual omissions B may have allowed himself. The third point was that the relative frequency of certain varia- tions of spelling was not the same. It should be noted that B is far from being consistent. On the first page we find adulescentulos, adolescentuli, and adhulescens, though the last may have been worked over. I come now to the question of the hands. I have no serious doubt that the hand is the same throughout. Other circumstances i Philol., XX, 723 ff. Petronius, Poggio, and John of Salisbury 21 confirm my belief. I have already shown the weakness of Beck's arguments, while Biicheler's position was stronger than he realized. The external history of the two is similar, as I shall show below. If Beck is right in seeing two hands in Va, or if there is only one, we have three (or two) MSS of about equal age so exactly alike in form that the joints are perfect, of parchment, quarto, in two columns to the page and thirty lines to the column on the average, and of the same history. Though the MSS cited by sixteenth-century editors were much alike, I cannot believe that such a situation as I have indicated can exist. I believe, therefore, that Va and B are parts of the same MS, and worthy of the same confidence, as Biicheler also thought. I have referred before to the external history of B. The Plautus volume passed through the hands of Vossius and Paul Petau, 1 pre- sumably then also of Alexander Petau, Paul's son. B belonged to Bongars, 2 and came to Berne through Graviset, we may suppose. 3 Now P. Petau and Bongars were joint heirs of the library of Pierre Daniel of Orleans, who secured the greater part of the Fleury library when that monastery was sacked in 1562. B is part of a once larger MS, now Bern. 347+357+330+Paris. 7665. 4 In 347 Daniel's hand appears. 5 There seems no doubt, then, that B belonged to Daniel, and therefore probably came from Fleury. It is doubtless true, as Traube says, 6 that it is uncritical to regard all Daniel-Bongars MSS as from Fleury. Other sources of Bongars' library are known, and the MSS of Daniel may have come from other places than Fleury. 7 Yet the hypothesis that B came from Fleury fits all the facts (at least all that I have been able to learn) better than any other. Bucheler believed that B came from Auxerre, 8 but erroneously, as I shall try to show later. This opinion seems to have been accepted 1 Bucheler, p. xviii; Hagen, Der Jurist u. Philolog Peter Daniel cms Orleans, Berne, 1873, pp. Iff., traces Va to Daniel, but does not mention B. 1 Bucheler, loc. tit.; Hagen, loc. tit. » Cf ., e.g., Sandys, History of Classical Scholarship, II, 192. * Manitius, Gesch., etc., p. 502. » TJsener, Rh. Mus., XXII, 421. • Munch. Sitsmngsb., 1891, p. 400. ' Hagen, Cat. Codd. Bern. {Bib. Bongars) , p. xvii. » P. xviii. 22 Evan T. Sage without question or investigation by Usener, 1 Traube, 2 and Manitius. 3 It is true that no Petronius is found in the Fleury catalogue of 1552,* but their MS may have gone (to Daniel or possibly Cuias ?) before that time. 5 On the next point I cannot agree with Biicheler. He states that he has relied mainly on B (including Va) for the true text, and, when these have failed, on Pithoeus' reports of a MS he called Altissiodurensis. This Biicheler identifies with B. 6 The resem- blance is indeed striking. I quote a few representative examples: Z) 2, 1, coria Alt.B, rest scattered m) 119, vs. 30, hosterik Alt.B, rest scattered n) 120, vs. 210, pinnis Alt.B, pennis cett. o) 135, 8, at Alt.B, et cett. These cases seem to show identity, or at least very close relation. However, note the following: p) 16, 3, sacrum Alt., om. B g) 112, 5, crucariae Alt., cruciaria B r) 113, 3, e le Alt., hedile B s) 119, vs. 30, sensum trahat Alt., s. trahant B t) 126, 17, planum Alt., pharium B u) 127, 7, Belienon Alt., polyenon B 1 Loc. cit. * Op. cit., p. 401. » Loc. cit. * Delisle, Notices et extraits, XXX, 1, 426 ff. * Since first conjecturing that B came from Fleury, I have found myself antici- pated, or the same conclusion independently reached, by others. Thus Hagen, Cat., p. xvii, states that the MSS of s. IX-X came from Fleury via Daniel, and in his classi- fication by age puts B in that group. Lindsay, Nonius, I, xxiv, says of Paris. 7665 (see above) :" Hie codex .... bibliothecae Petri Danielis erat ex bibliotheca nimirum Floriacensi ad eum delatus." Both Hagen and Lindsay, however, may be liable to Traube's charge. An interesting note is to be found in Memmius' Coniecturae (see, e.g., the Lotichius edition of 1629, p. 78). Memmius has commented on scattered passages, the last being 79. 7. His final note is: "Caetera deerant in exemplari manuscripto, quod ex bibliotheca Petri Danielis comparatum cum Dn. Goldasto com- municavit Nobiliss. & Ampliss. Dominus Bongarsius." B now breaks off at the end of chap. 80, resuming at 109. 9, vs. 8. While not exact, Memmius' description fits well enough. Apparently Va was still part of B. Hauler, Wiener Studien, XVII, 133, n. 16, thinks B may have come from Fleury, but his argument as a whole is not convincing. Professor E. K. Rand, who kindly had the photograph of B made for me, tells me that when he saw the MS he suspected from its general style that the MS had some connection with Fleury. He was then studying the script of the school of Fleury, but did not know the history of this MS nor the arguments I have advanced. A thorough study of Fleury will be very welcome. ■ P. xviiii. Petronius, Poggio, and John op Salisbury 23 These are sufficient to show, I think, that B is not the Altissio- durensis of Pithoeus. 1 This is not to say that there was no such MS. It is likely enough that there was. Petronius was known at Auxerre, e.g., to Heiric. On the extent of Heiric's knowledge of Petronius, and his relation to him in general, I shall have something to say at another time, and shall pass over those matters now. It is not clear that Biicheler had any other reason for thinking that B came from Auxerre than the identification with Pithoeus' Altissiodurensis, which I have just shown to be incorrect. At the same time it should be said that Alt. had few readings that B might not also have had. In 89, vs. 14, it is doubtful if B had to semper for mendacium, but I have no other sure case. A new study of the work of sixteenth- century editors would be desirable, with fuller knowledge of their methods. I must content myself now with a statement of my skepti- cism with regard to more than one sixteenth-century scholar. The results of our inquiry I believe to be briefly these: While John of Salisbury knew all parts of Petronius, it is not certain that he had a MS. His acquaintance might have been made at Fleury, with a MS of the type of B, but containing also the Cena, unless this was in a separate MS. Poggio did not bring the Cena from Britain; it may have come from Cologne in a MS similar to A in contents and form. The British particula cannot be identified. It probably remained in Niccoli's hands. There is some reason, not perfectly convincing, for associating A with Poggio's circle. Biicheler was right in regarding the Leyden leaves (Va) as part of B; wrong in identifying B with Pithoeus' Altissiodurensis. The study of Petro- nius at Auxerre is left for later investigation. ADDENDUM Following are the readings of Va at the damaged corners. First is given the reading of Bch. 5 , then the reports of Biicheler and Beck, and finally what actually appears in Va. f . 48r, 3. 3 : mediantur] For reports, see above, p. 20. t ntur visible auditoribus] no report Bch. Beck, nibus visible -bunt quod] no report Bch. Beck.