The papyrus containing the odes of Bacchylides was found in Egypt by natives, and reached the British Museum in the autumn of 1896. It was then in about 200 pieces. By the skill and industry of Mr F. G. Kenyon, the editor of the editio princeps (1897), the MS. was reconstructed from these lacerated members. As now arranged, the MS. consists of three sections, (1) The first section contains 22 columns of writing. It breaks off after the 8 opening verses of Ode xii. (2) The second section contains columns 23-29. Of these, column 23 is represented only by the last letters of two words. This section comprises what remains of Odes xiii. and xiv. It breaks off before the end of xiv., which is the last of the epinikia. (3) The third section comprises columns 30-39. It begins with the mutilated opening verses of Ode xv. (Ἀντηνορίδαι, the first of the dithyrambs), and breaks off after verse 11 of the last dithyramb,Ἴδας. The number of lines in a column varies from 32 to 36, the usual number being 35, or (though less often) 34.
It is impossible to say how much has been lost between the end of column 29 and the beginning of column 30. Probably, however, Ode xiv., if not the last, was nearly the last of the epinikia. It concerns a festival of a merely local character, the Thessalian πετραῖα, and was therefore placed after the thirteen other epinikia, which are connected with the four great festivals. The same lacuna leaves it doubtful whether any collective title was prefixed to the διθύραμβοι. After the last column (39) of the MS., a good deal has probably been lost. Bacchylides seems to have written at least three other poems of this class (on Cassandra, Laocoon and Philoctetes); and these would have come, in alphabetical order, after the last of the extant six (Idas).
The writing of the MS. is a fine uncial. It presents some traits of a distinctly Ptolemaic type, though it lacks some features found in the earlier Ptolemaic MSS. (those of the 3rd or 2nd century B.C.). Among the characteristic forms of letters is the , with a shallow curve on the top of the upright; a form found in MSS. ascribed to the 1st century B.C., and different from the more fully formed upsilon of the Roman period. Another very significant letter is the ξ, written as , a form which begins to go out after c. 50 B.C., giving place to one in which the middle stroke is connected with the other two. From these and other indications it is probable that the MS. is not later than the middle of the 1st century B.C.
The scribe, though he sometimes corrected his own mistakes, was, on the whole, careless of the sense, as of the metre; he seems to have been a mechanical copyist, excellent in penmanship, but intent only on the letters. The MS. has received corrections or small supplements from at least two different persons. One of them (Kenyon's A&SUP2;) was contemporary, or nearly so, with the scribe. The other (A&SUP3;) was considerably later; he wrote a Roman cursive which might belong to the end of the 1st century A.D., or to the early part of the 2nd. The correctors seem to be generally trustworthy; though, like the scribe, they were inattentive to metre, passing over many metrical faults which could easily have been removed. They appear to have compared their MS. with another, or others; but they sometimes made a bad use of such aid, intruding a false reading where their text had the true one.
Breathings are generally added, especially rough breathings; the form is usually square, but sometimes partially rounded. Accents are added, not to all words, but only, as a rule, to those which might cause doubt or difficulty to the reader. This was the Alexandrian practice, accents being regarded as aids to correct reading, and more liberally used when the dialect was not Attic. In accordance with the older system, the accent is not written on the last syllable of a word; when the accent falls there, a grave accent is written on the preceding syllable, or on two such syllables (e.g. βλὴχρας, πὰυθὰλης).
As Kenyon observes, no MS. of equal antiquity is so well supplied with accents. The MS. which comes nearest to it in this respect is the Alcman fragment in the Louvre, which is of similar or slightly higher age, belonging perhaps to the early part of the 1st century A.D.; and in that MS. the comparatively frequent accents were doubtless designed to aid readers unfamiliar with Alcman's Laconian Doric. With regard to other grammatical or metrical signs (προσῳδίαι) used in the Bacchylides MS., there is not much that calls for special remark. The punctuation, whether by the scribe or by correctors, is very sparse, and certainly cannot always be regarded as authoritative. The signs denoting the end of a strophe or antistrophe (paragraphus), of an epode (coronis), or of an ode (asterisk), are often omitted by the scribe, and, when employed, are sometimes placed incorrectly, or employed in an irregular manner.
F. G. Kenyon, Ed. princeps (1897); F. Blass, 3rd ed. (1904); H. Jurenka (1898); N. Festa, text, translation and notes (1898). [The latest edition is by Sir Richard Jebb (1905), with introduction, notes, translation, and bibliography; text only (1906). See also T. Zanghieri, Studi su Bacchilide, Bibliografia Bacchilidea, 1897-1905 (1905)].
(R. C. J.)
 The references are given according to the numbering in Jebb's edition.
 For other explanations suggested, see Jebb's edition, Introd. p. 18.