The manuscripts of Pliny the Elder’s “Natural History”

Pliny the Elder, who died in the eruption of Vesuvius in 79 AD, has left us only a single work out of his vast literary activity.  This is the Historia Naturalis, a compendium of information about natural phenomena of various sort.  The work consists of a prefatory letter, addressed to his friend, the emperor Vespasian Titus, followed by 37 books.  The first book is composed entirely of a list of contents for each book from book 2 onwards.  At the end of each list is a list of the authors used to compile it.

Ancient manuscripts

Pliny’s work was read continuously and epitomised throughout antiquity; indeed the Collectanea of C. Iulius Solinus is largely derived from Pliny and can be used for the establishment of the text.   Unusually, therefore, the remains of no less than 5 ancient codices have come down to us.

  • M = St. Paul in Carinthia, Stiftsbibliothek 3.1 (25.2.36; xxv.a.3) (CLA x.1455) (=codex Moneus), 5th century.  Discovered at the Austrian monastery where it still resides in 1853 by F. Mone, hence the name.   The manuscript contains Jerome’s Commentary on Ecclesiastes, written around 700 AD in Luxeuil minuscule on reused parchment.  It also contains a 15th c. ex-libris for Reichenau.  But the book is a palimpsest.  The lower text consists of large sections of books 11-15, partly visible, in a 5th century uncial hand.  The manuscript has been treated with chemicals to try to bring out more of the ancient text. The text is of high quality. A description of the ms. and a complete transcript is available as most of vol. 6 of Sillig’s edition (Gotha, 1855).  The codex is also online here, although hard to use.
  • N = Rome, Bibl. Naz. Sessor. 55 (CLA iv.421), 5th century, uncial.  Also a palimpsest, containing patristic texts and written in the second half of the 6th century, the lower text consists of a few passages from books 23 and 25.  Both texts were written in Italy.
  • O = Vienna 1a (CLA x.1470), first half of 5th century, uncial.  Probably written in the south of Italy, it consists of fragments of 7 leaves, reused for bindings.  It contains part of books 33 and 34.
  • P = Paris lat. 9378, folio 26 (CLA v.575).  A single folio, seemingly of Italian origin, written at the end of the 6th century and containing part of book 18.  It was found, apparently, in the binding of a manuscript from St. Amand in France.
  • Pal. Chat. = Autun 24 + Paris 1629 (CLA vi.725), 5th century, uncial, containing a few sections of books 8 and 9.  Presumably from Italy, it was overwritten in the late 6th century with Cassian’s Institutions, probably in Southern France.

It is telling that 3 ancient manuscripts, M, P and Pal.Chat, found their way to France but were turned into clean parchment before they could generate a tradition in that region.

The medieval manuscripts have been divided by editors into two classes, the older or vetustiores, and the newer or recentiores.  Unfortunately the dates of the mss. have been so confused that the division is not as clean as it should be.

Medieval manuscripts – vetustiores

  • Q = Paris lat. 10318 (CLA v.593), written in central Italy ca. 800 AD, in uncial.  This contains the Latin Anthology, and includes medical excerpts from books 19-20.  The source manuscript used for this was of high quality.
  • A = Leiden, Voss. Lat. F.4 (CLA x.1578), first third of the 8th century, insular, written in the north of England.  Contains books 2-6, with large gaps.  Other books may have been known to Bede, and Pliny is listed in Alcuin’s list of books present in York.  Not as good as M or Q, but better than most continental mss.
  • B = Bamberg, Class. 42 (M.v.10), first third of the 9th century, in the palace scriptorium of Louis the Pious.  It contains books 32-37, and is the only one to preserve the ending of the work.  Of excellent quality, and clearly copied carefully from an ancient codex whose notae it carefully preserves.  Online here.

There are also a number of collections of excerpts made in this period which preserve portions of the text.  They seem to be associated with the court of Charlemagne and the scholars who communicated with it.

Medieval manuscripts – recentiores

The vetustiores do not give us anything like a complete text, unfortunately.  For most of the work we are dependent on the inferior recentiores.  These contain small lacunae, but give a more or less complete text.

The main mss., which all descend from a common parent, are:

  • D+G+V = Vatican lat. 3861 + Paris lat. 6796, ff. 52-3 + Leiden, Voss. Lat. F. 61 (CLA x.1580 + Suppl. p.28), written ca. 800 AD in north-east France, perhaps in the Corbie area.  This manuscript was later divided into three.  It contains most of the work.
  • Ch = New York, Pierpont Morgan Library M.871 (formerly Phillipps 8297), first half of 9th century, written apparently at Lorsch by a scribe using the style of St. Vaast.  Contains books 1-17.
  • F = Leiden, Lipsius 7, written first half of the 9th century, by a scribe from Luxeuil collaborating with one from Murbach, possibly at Murbach.  Contains books 1-37.   Possibly copied from D+G+V before it was corrected.
  • R = Florence, Bibl. Ricc. 488, second half of the 9th century, perhaps at Auxerre.  Contains books 1-34.
  • E = Paris lat. 6795, 9-10th century, France.  Contains books 1-32.

All five mss. are related; their ancestor suffered a dislocation, where leaves from book 2-3 were swapped with some from 4-5.   Attempts were made to fix this in D and E, in a botched way.

E was prone to accident; leaves were lost in the ms. from which it was copied, and then in E itself.  Unfortunately it was E that dominated all later copies.  However some of them were clearly corrected from otherwise unknown copies of the older and better tradition, in D2, F2, R2,  and E2.

Medieval manuscripts – later recentiores

  • h = Berlin (East), Hamilton 517, 11th c.
  • X = Luxembourg 138, 12th c., from the Abbaye d’Orval.
  • Leiden, Voss. Lat. Q.43, 12th c., from Orleans.
  • n = Montpellier 473, 12th c., from Clairvaux; mainly medical excerpts.
  • Co = Copenhagen Gl.Kgl.S.212 2°, ca. 1200 AD.

All these are derived from E.

  • Oxford, Bodl. Auct. T.1.27 + Paris lat. 6798, 12th c., Mosan region.
  • C = Le Mans 263, 12th c.  A beautiful book, apparently of English origin. (Image of one opening here).

These are very close to E, and may derive from it.

  • e = Paris lat. 6796A, 12th c.  A faithful copy of E.
  • a = Vienna 234, 12th c.  Not derived from E, but from its ancestor.
  • d = Paris lat. 6797, third quarter of 12th century, Northern France, probably St. Amand.  Contains a substantial amount of the older tradition.

There are many more manuscripts, many of which have not been explored for their textual value.  One which is online is Ms. British Library, Harley 2676, written in Florence in 1465-7.  The BL site adds, ” identifiable as the missing Pliny from the Badia of Fiesole (according to unpublished notes of A. C. de la Mare at the Bodleian Library, Oxford)”.

Critical editions

The text of the NHwas established by the work of German scholars in the 19th century; J. Sillig, D. Detlefsen, L. von Jan, and K. Rück.  This culminates in the second Teubner edition, that of L. Jan and C. Mayhoff (5 vols, 1892-1906).  Much of the fundamental work on the recension was done by Detlefsen, in a series of papers[1] and in his edition (5 vols, Berlin, 1866-73).  The 20th century has only produced the Budé edition, now in more than 30 volumes, containing limited and rather stale information.


I am indebted for all this information to L.D. Reynolds, Texts and Transmissions, Oxford, p.307-316.

UPDATE: My thanks to J.B. Piggin for extra links.

  1. [1]Rheinisches Museum 15 (1860), p.265-88 and 367-90; Philologus 28 (1869), p.284-337; Hermes 32 (1897), p. 321-40.

28 thoughts on “The manuscripts of Pliny the Elder’s “Natural History”

  1. Thank you for this excellent list! I would suggest two minor corrections:
    – ‘Berlin (East)’ should now be ‘Berlin, Staatsbibliothek–Preussischer Kulturbesitz’.
    – The Österreichische Nationalbibliothek in Vienna has changed its shelfmarks. What was MS 234 is now Cod. 10. Scans of its microfilm are available online.

  2. I found it somewhat circuitously, via the catalogue of the Österreichische Nationalbibliothek. I have looked for it again. The link below seems permanent and it leads to the catalogue entry on the manuscript, where one can simply click on the reproduction.

  3. What use, if any, has been made of the Bodleian MS? Sillig did not mention it.

  4. Thank you. Yes, it is obvious. Yet, Carl Ritter suggested in September 1828 to the Board of the Gesellschaft deutscher Naturforscher und Aertze, which supported the preparation of a critical edition, that [his former student] Friedrich Rosen, who had just been appointed professor of oriental languages at the new University of London, be asked to collate the Bodleian manuscript. His colleague Lichtenstein reported a year later that Rosen had taken up that task. There my trail ends. Is there anywhere further information about the outcome of that attempt?

  5. If you mean Auct. T 1 27, there’s a lot of bibliog. on it. Richard Hunt of the Bodleian spotted that it’s the first vol. of Paris B. N. Lat. 6798; the beginning was already damaged by the end of the 15th c., and since about 1760 it has contained only Books 7-15. There’s no point in collating it, because its ultimate sources survive, namely E and d (see Roger above); not that anyone c. 1830 would have known that. In the 15th c., when it was first in Florence and then in Naples, it was used by Politian amongst others. I haven’t come across Friedrich Rosen, if he’s the main reason for your interest, but a collation made c. 1830, whether on a printed ed. or in a separate notebook, is quite likely to survive if one knew where to look. Try the Verzeichnis deutscher Nachlaesse or ask the librarian of the institution where he held his chair in London.

  6. The expedition to Oxford was launched in the hope of finding V, a Vossianus that Gronovius had extolled. By the time Sillig found that it was in Leiden, Rosen had made what Lichtenstein in Sept. 1829 described as a successful visit to Oxford, but in Sept. 1830 Lichtenstein reported that the ms. in England consisted of ‘just some leaves’ and so nothing could be expected from there. The three mss. now in the Bodleian were all there at the time; Canon. Class. Lat. 295 has all 37 books, but Rawl. G 144 gives out in the capitula for Book 9, and Auct. T 1 27 has only books 7-15. Which of these two Rosen saw is anyone’s guess, but probably he didn’t collate it. By the way, Roger, Pliny’s letter to ‘Vespasian’ is actually addressed to Titus.

  7. Thank you for sharing this, unfortunately too late to be included in Rosen’s biography, which is about to appear as a volume in the series of Abhandlungen für die Kunde des Morgenlandes: Rosane Rocher, with Agnes Stache-Weiske, For the Sake of the Vedas. The Anglo-German Life of Friedrich Rosen 1805–1837, Wiesbaden: Harrassowitz, 2020. I wonder now if the manuscript Rosen sought to find in Oxford might have been that which his father had collated in part in Leiden. Pliny on pp. 6 and 242 of Rosen’s bio

  8. Thank you for your excellent post! I am also interested in the critical editions that you briefly mentioned. I am wondering if it is possible to let me know more about this part, for example, how to find more information about critical editions? Thank you very much!

  9. Many thanks for this precise, very useful synthesis. I already worked with the online Ms. British Library, Harley 2676 you cite (, and, for plant names in latin; Harley seems to me more “edited” than the latter, as some names take a more modern form, with similarity to the latin text in Loeb Classical Library’s Rackham (1960) text (, seems to me. There’s a good part of Libro XII in the extraordinary “M” palympsest ms. i discovered thanks to you and just now looked into, but not all the sections on the plants i’m interested in, unfortunately.
    I’ll look for online access to the other old mss. you cite, although i guess that there’s not so much hope as you don’t give internet links. So, most of all that stays from N.H. comes from “E”; is there a scan or a transcription of it with online access ?
    Thanks very much for your very clear synthesis.

  10. E is available on Gallica: feed in ‘latin 6795’. Other important mss. are also available on line: d = Paris B. N. Lat. 6797 on Gallica, D = Vat. Lat. 3861 at digivatlib, P = Prague Univ. XIV A 12 on Manuscriptorium. You should look first in the Teubner edition of Jan-Mayhoff, because they use E, d, and either D or its copy F. The biggest gap in the ed. is the absence of P, late (15th c.) but indispensable. Harl. 2676 descends from d by way of L = Florence Laur. 82.1-2 (on line: feed in ‘plut.82.01’ with no spaces), which has corrections from a ms. related to P. Alternatively, tell me what plants (in Latin) you’re looking for.

Leave a Reply