• C. Leonardi, Raterio e Marziano Capella
  The author draws attention to the autograph of Ratherius of Verona on the fly-leaves and in the margins of the manuscript. He studies the dedication formula of archdeacon Heliseus of Auxerre (f. 50v) and the chronology on f. 95r.
• C. Leonardi, I codici I, pp. 452, 462-463; II, pp. 67-68
  Leonardi discusses the dedication formula of Heliseus of Auxerre and briefly describes how the manuscript went through the hands of Ratherius of Verona, Petau, Christina of Sweden and Vossius
• G. Glauche, Schullektüre im Mittelalter, p. 45
  Glauche adds that this manuscript, together with 11 others, witnesses the special interest of the “Irish colony” in De nuptiis. He suggests a concentration of manuscripts and manuscript production with the commentary of the learned Irishman Martin of Laon in Northeastern France or the surroundings of Laon.
• K.A. De Meyier, Voss.Lat.F. 48
  An excellent short description of the manuscript, on which I relied especially in the description of its contents.
• J. Préaux, Les manuscrits principaux, pp. 79, 101, 123
  Préaux dates the manuscript in 847, but it is not clear why. He mentions the fact that the manuscript was given to Saint-Germain of Auxerre shortly after its origin. He suggests that this manuscript is close to the sixth-century archetype, because it carries the correction note of Securus Melior Felix. He ranges it, however, under the category of ‘second generation’ Martianus manuscripts; the many mistakes due to the scriptio continua of the archetype are corrected in Vossianus Latinus Folio 48.
• D. Shanzer, Review article
  Shanzer repeats the conclusions of the above-mentioned authors and adds that in the stemma of manuscripts, this manuscript is related to Besancon, Ms. 594, Paris, BnF lat. 8669, and Oxford, Bodleian Library, Laud.lat. 118.
• In other papers, e.g. L. Holtz, “L'école d'Auxerre”, in: L'école carolingienne d'Auxerre, pp. 131-146, esp. 138; and the unedited list of Carolingian manuscripts in Auxerre of G. Lobrichon, the same information is repeated.

The manuscript consists of 12 quires of eight leaves (4 bifolia). The first quire lacks one leaf (which was probably originally blank[1]) the rest is perfectly regular: 1⁸(-2) [1-7], 2⁸-12⁸ [8-95].[2] All quires except the twelfth carry an original quire signature: Q i - Q xi. These are written in the middle of the lower margin of the last verso of each quire, with the exception of the signature of the first quire which is written on the first (originally blank) recto-side of the quire (f 1r - i; f 15v - ii; f 23v - iii; f 31v - iiii; f 39v - v; f 47v - vi; f 55v - vii; f 63v - viii; f 71v - viiii; f 79v - x; f 87v - xi).

The foliation is modern, written in the upper right of each recto-side in pencil.

The leaves measure ca 310 x 250 mm. That this closely resembles the original size of the manuscript can be concluded from the fact that the complete text, including the glosses in the margins, has been preserved and was nowhere damaged by cutting. On f 80r in the lower margin there are drawings of two signs of the zodiac, Sagittarius and Leo, of which the body of the archer and the legs of the lion fall off the page. This, however, seems to have been the case in the original lay-out of the drawings; it is not the result of cutting.

The parchment is of medium quality. The thickness of the leaves shows some variation, but this is not extensive. There are two obvious exceptions to the general thickness of the leaves: the third quire is made of thicker and stiffer parchment, and the middle bifolium of the fifth quire is thinner. The colour of the parchment is reasonably constant, although hair- and flesh-sides are easily recognizable.[3] All quires are arranged according to the same pattern: HFHF.[4] The last quire is dampstained and shows most signs of wear.

The leaves contain 28 or 29 lines (10 lines measure ca 63 mm), ruled with a hard point, mostly above topline.[5] The lay-out of the page differs per group of quires. The ruling for 1-6 can be described by the formula[6] 11-1:J/0/1-1:C/F, 22<32><128>68 x 37<210>63 mm — to put it in words: a narrow column for glosses (32 mm), a broader one for the text of De nuptiis (128 mm). On the first nineteen folia, the gloss column is already incidentally used for text, and from f 20 onwards the column is still present in the ruling pattern, but not used as such. The glosses are written in the inner and outer margins. In quire 7 the narrow column has disappeared and the outer margin is marked off with a double line (1-2:J/0/1-1:C/J, 22<153>5>70 x 39<209>60 mm). In quires 8-12, a single instead of a double line was used to mark off the writing space: 1-1:J/0/1-1:C/J, 25<158>68 x 45<204>60 mm. Prickings for the vertical lines are found next to the first and last text-line, for the horizontal lines, on the outer vertical line. From the fact that the prickings for the lines are not found on the edges, and that the horizontal and vertical projecting lines do not reach the edges of the leaves, one can conclude that pricking was done by use of a template. The ruling was exercised on the hair-sides and per two bifolia, folded open — as one can conclude from the irregularities and mistakes, e.g. in the last quire.[7] Thus, the general pattern of furrows and ridges is:[8] >>>>|<<<<.[9] In quires 2 and 9, however, deviations from this general pattern can be noticed: in 2 the two inner bifolia, and in 9 the two outer bifolia were (accidentally) folded the other way around: 2 >><<|>><<; 9 <<>>|<<>>.

The script is Carolingian minuscule with cc-a's (alongside closed a's), clubbed ascenders, ligatures of -nt, -rt, (-st and -ct), et-ligature within a word or even in between two words (e.g. sensificar&onis, f 84r l.26) and relatively wide line spacing. The scribe uses an open g, the insular abbreviation for est, round and straight d's. For majuscules both uncial letters and capitalis rustica are used (often in a deliberate alternation). The transmission of Greek words, written in capitals, is quite corrupt; the scribe intermingles Greek and Latin letters.[10] A clear change of hands is not recognizable, but the script slowly grows smaller (relatively wider spaced), more pointed and less round. When one leafs through the manuscript, the pages look less and less full towards the end, which is not only due to a diminishing number of glosses. The letters are largest on f 3 — measuring slightly more than 3 mm —, but from f 13 onwards their corpora are never higher than 2 mm.

In the glosses, however, three different scribes can be recognized. First, the scribe who originally glossed the text (probably simultaneously with the text of De nuptiis) and who used for this purpose the narrow column for glosses or the margins. The scribe uses a round, relatively broad letter in a slightly faded brown ink, and a quite broad pen. The glosses are usually neatly tied to the gloss with a reference symbol. Secondly, a contemporary or slightly later layer of glosses is added to the text with a thinner pen, a smaller sized Carolingian minuscule and darker ink. This scribe scribbled his glosses mostly between the lines, but also in the margins. He is less careful in the placement of his glosses and in using reference signs for tying gloss to lemma. He uses more round d's and more e-caudatae (instead of ae) than the first hand. The glosses from these two hands can be identified as belonging to the Anonymous corpus.[11] One of these two (probably the first) writes short passages in Tironian shorthand (e.g. f 2r, inner margin; f 4r, between the lines above line 21 and 23). Thirdly, the work of a later scribe can be observed, who did not actually comment upon the text, but highlighted words by repeating them in majuscules in the margin. These words are usually not keywords in the text — as one would expect —, which could serve as a kind of index, but can be characterized as `strange' or unusual words, which the student of the text would perhaps have wanted to add to his vocabulary. Only an occasional longer gloss of this hand appears in the margin (e.g. f 9r). This third hand has been identified by Leonardi as the autograph of Ratherius of Verona (ca 880-974), who also wrote a sermon on f 1r and a short letter in the lower margin of f 95v.[12]

The punctuation is original, sober and sparing. Majuscules are used for explicits/incipits, and (at random) at the beginning of verses, passages and `sentences.' Prose and verse are generally distinguished by writing the verses in two columns (for which separate verticals are sometimes drawn) and by beginning each verse in principle with a majuscule. The only rubrication in the entire manuscript is found in the explicit/incipit of book II/III — the originally red ink now rather black or silver-coloured. There are a few initials that stand out because of their tallness: the initial letter of book I is five lines tall, of book II three lines. The other books have initial letters of one or two lines high, as do some prose or verse passages.

The manuscript is not decorated. There are two drawings on f 80r of two signs of the zodiac, Sagittarius and Leo, but these function, to my mind, more as commentary to book VIII (De astronomia) than as decoration.

The history of the manuscript is full of mysteries. Although there seem to be many clues for origin and date, they are in fact no more than riddles, as, for example, in the case of the chronology found on f 95r:

In this enigmatic chronology the numbers do not add up: 892 plus 1056 is not 1848, but 1948; the destruction of the temple by Titus in the year 70 A.D. is neither 776 years away from the year 815 (corrected version), nor from 915 (original version). In short: it actually gives us no hold for dating the manuscript. It is highly unlikely that the manuscript was written in 815 — all paleographical evidence points to a later date. Possibly the chronology was copied out of an exemplar from the year 815, but that does not prove anything for the text of De nuptiis, let alone for the glosses added to it.

There is, however, another clue for the dating of this manuscript, hidden in the margin of f 50v. Along the edge of this folio the text HUNC LIBRUM DEDIT HELISEUS ARCHIDIACONUS SANCTO GERMANO PRO VITA ETERNA (archdeacon Heliseus gave this book to St. Germain for eternal life) is written, from bottom to top and left to right, in rustic capitals. Heliseus can be traced back to the late 10^th-, early 11^th-century obituary list of the St. Stephanus cathedral of Auxerre. He died on 20 December,[15] but the list unfortunately does not mention of which year. He is also found, however, as one of the signatories to a decree at the council of Douzy in 871.[16] In modern scholarship it was agreed upon that he must have given this book to the abbey of Saint-Germain in Auxerre around the middle of the ninth century. It remains unclear where the manuscript was actually written, but shortly after its origin it was in the possession of the monks of Saint-Germain of Auxerre. It has been suggested that archdeacon Heliseus not only donated the manuscript to the abbey, but that he had it written in their scriptorium. The paleographical evidence could support this theory, since the manuscript shows a good similarity to manuscripts which have been placed in Auxerre with more certainty.

There is, however, also evidence which points at Tours as the place of origin. In two glosses, the name of Hildebertus is mentioned. The name Hildebertus can be traced back to Tours in the 820's, when he was listed as a monk of Tours in a liber fraternitatum from St. Gall, and when he appears as the scribe of a Tours Bible manuscript, of which only a single bifolium survives. The place of origin thus still remains unclear.

In the tenth century, the manuscript came into the hands of Ratherius of Verona (890-974), whose autograph is found in the manuscript. In his childhood Ratherius was oblated to the monastery of Lobbes, which — at that moment — was closely tied to the famous school of Liege,[17] possessed a rich collection of books and was therefore able to educate its pupils well. Ratherius, however, led a turbulent life, in which he undertook many travels. Three times he became bishop of Verona, but every time he was sabotaged, expelled, imprisoned or exiled by the local aristocracy. From 953 to 955 he held the episcopal see of Liege and, hence, was abbot of Lobbes, supported by Bruno, the brother of Otto I. In 955 he became a victim of a conspiracy of Balderic, bishop of Utrecht, and Rotbert, the archbishop of Trier. The nephew of the bishop of Utrecht, also called Balderic, became the new bishop of Liege and Ratherius was again exiled. With the support of Bruno he became abbot of a little monastery in Aulne, in the vicinity of Lobbes, from 955 to 960. He became bishop of Verona for the third time in 961, only to be expelled again after 6 years; with his sharp tongue and pen he made too many enemies and was forced to leave Verona in 967. For some time he drifted from monastery to monastery and he died in Aulne in 974.

It will be clear that Ratherius' notes in the margins of Vossianus Latinus Folio 48 hold no unequivocal clues to the whereabouts of the manuscript in his time. It is unlikely that the manuscript was present in the library of Lobbes, for De nuptiis is mentioned in neither of the two catalogues of the library that have survived.[18]

From the tenth century onwards the history of the manuscript eludes us completely. On folio 20v a note in a thirteenth-century script says Guido de Lidols and Petrus. On folio 53r again Petrus et qui con illo — perhaps in the same hand. The meaning of these notes, however, is unclear. Only at the end of the sixteenth century the codex reappears as part of the collection of Paul Petau (1558-1614): his signature D.32 is written on one of the paper fly-leaves. At his death the collection came into the possession of his son, Alexander Petau, who sold it to the Swedish queen Christina in 1650. The manuscript then came into the hands of Isaac Vossius and via Gerardus Vossius it became part of the collection of the University Library of Leiden, where it was given its present signature: Vossianus Latinus Folio 48. A schidula with his name and a stamp of the University Library can be found in the lower margin of f 2r.

Vossianus Latinus Folio 48 once had a parchment cover with a French charter on the inside. This cover is now separated from the manuscript and preserved in a box with signature BPL 2513. Its present binding is a 19^th-century parchment back with cardboard boards. In the gutters of the quires six sewing stations can be seen, spaced (measured down from the head of the spine) 21-40-66-66-64-33 mm.[19] There are six paper fly-leaves at the beginning and three at the end of the book.

The manuscript contains:

• f 1r Originally blank, in the tenth century filled in with a sermon of Ratherius of Verona (autograph): CUM POTENTUM ALIQUIS AD PENITENTIAM ACCEDIT DE QUO IGNORATUR UTRUM CORDE AN PRETEXTU TANTUM UT ASSOLET HOC AGAT HUMILITATIS. HEC ORATIO SUPER EUM POST PSALMOS ET LETANIAM FUNDATUR.
• f 1v On rasura an index in 15^th-century French cursive script:

  liber satiricorum libri viii

  item de dyaletica liber i

  item de rethorica liber i

  item de geometria liber i

  item de arismetica liber i

  item de astrologia liber i

  item de armonia liber i[20]

• f 2r Martianus Capella, De nuptiis Philologiae et Mercurii, books I-IX

  -91vI 2r-11r; II 11r-19r; III 19r-30v; IV 30v-40r; V 40r-52v; VI 52v-66r; VII 66r-74v; VIII 74v-82r; IX 82r-91v.

• f 92r Figures to book VI (On geometry) and VIII (On astronomy)
• f 92v Pseudo-Cyprianus, De xii abusivis saeculi,[21] prologue: Duodecim abusiva sunt seculi. hoc est sapiens sine operibus ... dei iudicium rotatur.
• f 93rPseudo-Hieronymus, Commentarius in canticum Deborae:[22]
• f 94v INCIPIT CANTICUM DEBORAE. Barach vir Deborae prophetidis fuisse traditur ... Et si queritur cur vinum non dederit. respondendum est quia domus recabh vinum non bibebant. sicut habes in iheremia propheta.
• f 94v(l.28) Anonymous commentary on days of fasting:
• f 95r haec dicit dominus exercituum. Ieiunium quarti. Ieiunium quinti. Ieiunium septimi. Ieiunium decimi erit domui iuda gaudium et letitia ... et vasa domus domini in babilone translata fuerunt.
• f 95r, l.22 List of the names of the months in Hebrew:

  Nomina mensuum hebreorum. Primus mensis martius aebraicae nisan vocatur ... Ianuarius sebbath. febroarius adar.

  (l. 25) Chronology:

  Ab origine mundi usque ad nativitatem noe ... Ab origine mundi usque in presentem annum III milia CLXXV.

• f 95v Figures to book IX (On harmony), enumeration of note names; note to book 4 (On dialectic): predicativus. UTRUM UTILIS SIT DIALECTICA. Omnis disciplina utilis est ... MIXTUS. UTRUM UTILIS EST DISPUTANDI SCIENTIA. Si bene disputare utile est ... omne iustum honestum ...

  Lower margin: letter of Ratherius of Verona (autograph):[23]

  Obsecro karissimam dominationis vestrae paternitatem ... et ego propterea X missas vobis faciam cantare.