THE TEXT OF THE GREEK BIBLE - A STUDENT'S HANDBOOK by Frederick G Kenyon, late Director & Principal Librarian of the British Museum. First published 1937 New edition 1949 Reprinted 1953. (Prepared for katapi by Paul Ingram 2015).
HOME | contents | << | translation : when? what? who? | arrangement | differentiation to hebrew | later translations: aquila , theodotion, symmachus | recensions: origen, others | the mss | papyri | uncials | miniscules | in-print | bibliography | >> |


The story of the first translation of the Hebrew Scriptures into Greek is told in the work known as the Letter of Aristeas, ← Edited by H. St. J. Thackeray as an appendix to Swete's Introduction to the Old Testament in Greek (Cambridge 1900), pp. which appears to have been written under one of the later Ptolemies and was known to Josephus, who paraphrases a large part of it. According to the writer (who purports to be a contemporary) Demetrius of Phalerum, director of the Alexandrian Library founded by Ptolemy I, being instructed by his successor, Ptolemy Philadelphus (285-246 B.C.), to collect all the books in the world, mentioned that the Jews possessed some books of their law which deserved to be included, but written in their own language and script, and therefore needing to be translated. The King thereupon instructed him to write to the High Priest at Jerusalem, backing his request by a gift of vessels for the Temple and the return of some thousands of Jews who had been compulsorily brought into Egypt. The High Priest sent in return seventy-two scholars (six from each tribe), who, after lavish entertainment by the king and an elaborate dialectical display of their wisdom, set to work and completed the translation of the Law in seventy-two days, to the great delight of the Jewish population of Alexandria.

The story of Aristeas, which is embellished by copious details irrelevant to the actual execution of the translation, is quite evidently a work of fiction, but there is no reason to doubt that it rests on a foundation of fact at least to this extent, that the Pentateuch was translated into Greek in Egypt before the middle of the third century B.C. There is confirmatory evidence of this in the fact that the version of Genesis was used by a writer of the name of Demetrius, who lived in the last quarter of that century. ← Swete, op. cit., pp. 17, 18. Whether it was the result of a royal command, rather than of the need of the Jews in Egypt for a version of the Scriptures in the Greek language, then more familiar to them than Hebrew, or even Aramaic, may be doubted, and is not material. What is certain is that the legend of the seventy-two translators was generally accepted, and is reflected in the name of Septuagint, by which the version is generally known. It is also certain that the style is Alexandrian, in the vernacular idiom which is known from the papyri.

Aristeas does not claim that the first translation embraced more than the Law, i.e., the Pentateuch. The exaggerations of later writers, which extended it to the whole Old Testament, and added various miraculous details, may be ignored. When and by whom the other books were added is quite unknown; but since the author of the preface to Ecclesiasticus, who states that he came into Egypt in the thirty-eighth year of Euergetes [II=132 B.C.], refers to the existence of translations of " the law itself and the prophecies and the rest of the books " (αὐτος ὁ νόμος καὶ αἱ προφητεῖαι καὶ τὰ λοιπὰ τῶν βιβλίων) and himself proceeded to translate the work of his grandfather, Jesus, son of Sirach, it may be concluded that by about this time the Greek Old Testament as we know it was complete.

Some light has been thrown on the methods of the seventy translators by the careful study of many of the books by the late H. St. J. Thackeray. The Septuagint and Jewish Worship (Schweich Lectures of the British Academy, 1920), Lecture I. He finds evidence in several of them of the co-operation of two translators. Thus in the books of Samuel and Kings (which in the Greek are denominated the four books of βασιλεῖαι, usually translated as " Kingdoms," though Thackeray would prefer " Reigns ") he finds linguistic evidence of the presence of two translators, the first of whom (whose style is of the Palestinian-Asiatic school) omitted the whole of 2 Sam. xi.1-1 Kings ii.11, the last chapter of 1 Kings and the whole of 2 Kings, perhaps as being less creditable to the nation, while the other, who appears to be of Western Asia, added these portions. In the Prophets the division of labour seems to be merely mechanical. Thus in Jeremiah one hand translated the first half, as far as the latter part of chapter xxix (the point of junction cannot be fixed to a verse), and the other the remainder. This division may well correspond to the division between two rolls, for each half about amounts to the normal contents of a Greek papyrus roll. In Ezekiel also the first hand proceeds to about the beginning of chapter xxviii; but the second translator, who there took up the task, desisted at the end of chapter xxxix, and the first translator then finished the book. There seems therefore to have been some definite organization in the translation of the Prophets, which suggests official control; while the Hagiographa seem to be the result of individual initiative, as we know was the case with regard to Ecclesiasticus.

The Greek Old Testament differs from the Hebrew canon both in contents and in arrangement. The Hebrew canon recognizes three groups:
(i) the Law, i.e., the Pentateuch;
(ii) the Prophets, subdivided into (a) the Former Prophets, i.e., Joshua, Judges, Samuel and Kings, and (b) the Latter Prophets, i.e., Isaiah, Jeremiah, Ezekiel and the twelve Minor Prophets;
(iii) the Writings or Hagiographa, i.e., Chronicles, Ezra-Nehemiah, Daniel, Esther, Ruth, Job, Psalms, Proverbs, Song of Solomon, Ecclesiastes, Lamentations; in all, twenty-four books (the Minor Prophets being reckoned as one book). ← Sometimes reduced to twenty-two, by the combination of Ruth with Judges and Lamentations with Jeremiah.
The Greek Old Testament includes along with these a number of books which apparently circulated in the Greek-speaking world (led by Alexandria) with equal acceptance, but never obtained entrance to the Hebrew canon. These are 1 Esdras (a different version of part of Chronicles and Ezra-Nehemiah, with an additional passage, 2 Esdras being the canonical Ezra-Nehemiah), Wisdom, Ecclesiasticus, Judith, Tobit, Baruch, and (in some MSS.) four books of Maccabees. These books passed also into the first Latin Bible, which was translated from the Greek; and the adverse opinion of Jerome, who accepted the Hebrew canon, did not avail in the end to keep them out of the Vulgate as finally adopted by the Roman Church. Luther, on the other hand, followed the Hebrew canon, and the English translators (headed by Coverdale) followed Luther, so that the English Bible conforms to the Hebrew canon, and relegates these books to the Apocrypha. ← The Apocrypha in the English Bible includes the books enumerated above, except 3 and 4 Maccabees, but adds to them 2 (= 4) Esdras, '' the rest of Esther '' (which are integral parts of the Greek Esther), the Song of the Three Children (which is an integral part of the Greek Daniel), Susanna and Bel and the Dragon (which form the beginning and end of the Greek Daniel), and the Prayer of Manasses (which occurs among the hymns attached to the Psalter in the Codex Alexandrinus, and also in some other MSS.). In the Latin Vulgate also only two books of Maccabees are included and the Prayer of Manasses and 3 and 4 Esdras are appended at the end of the whole Bible, as not having full canonical authority. The Vulgate 3 Esdras is our 1 Esdras, the canonical Ezra and Nehemiah being numbered as 1 and 2 Esdras.

Other works hovered on the fringe of the Greek Bible, though they did not ultimately find a permanent home in it. The most notable among these are the Psalms of Solomon, which was originally included at the end of the Codex Alexandrinus, the Book of Enoch, which is quoted by Jude, and of which a copy was included in the library of the Church from which the Chester Beatty papyri (to be described later) came, the interesting apocalyptic work known as 4 Esdras (in our Apocrypha 2 Esdras), and others. But these cannot be regarded as part of the Greek Bible.

The arrangement of the books of the Septuagint shows that the books in it which were not included in the Hebrew canon were by no means regarded as on a different level from the rest, but were an integral part of the Greek Bible, being distributed among the others according to their character. The order of the canonical books is also different from that of the Hebrew Bible, the distinction between the three classes of Law, Prophets and Hagiographa not being observed. There are variations in detail in the early MSS., but normally Chronicles-Ezra-Nehemiah are attached to the four books of Kingdoms, Lamentations and Baruch follow Jeremiah, and Daniel (with Susanna and Bel and the Dragon) is associated with the three Major Prophets. The Hagiographa also (including Wisdom, Ecclesiasticus, Judith, and Tobit) are generally placed before the Prophets, as they are in the Vulgate and in our English Bibles. In the Codex Alexandrinus, however, they follow the Prophets.

Such then are the contents of the Septuagint, which may be regarded as the Bible of Alexandria, which spread to the whole Greek-speaking world and was generally adopted by the Greek Christian Church. It was not, however, accepted by the stricter Jews, who in controversy'repudiated arguments based on Septuagint texts. Josephus follows the Hebrew canon of twenty-four books; and it is to be observed that no quotation from any of the books of the Apocrypha is to be found in the New Testament, although there are references (in Luke xi, 49-51, John vii, 38, 1 Cor. ii, 9, Eph. v, 14, James iv, 5) to words as Scriptural which do not occur either in the canonical books or in the Apocrypha, and one in Jude to the book of Enoch.

The differences between the Septuagint and the Hebrew are not, however, confined to the difference in the books contained in them. There are also considerable divergences in the text of the books contained in both. Thus in Job the Septuagint is shorter than the Hebrew text by about one-sixth, and there are large variations in Joshua, 1 Samuel, 1 Kings, Proverbs, Esther and Jeremiah, ← In 1 Sam. the LXX omits xvii, 12-31, 41, 50, 55-58, xviii, 1-5, 10,11,17-19. In Jeremiah chapters xlvi—li follow (in a different order) xxv, 13, and the following passages are omitted: x, 6-10, xvii, 1-4, xxvii, 1, 7, 13, and much of 17-22, xxix, 16-20, xxxiii, 14-26, xxxix, 4-13. and lesser ones in other books. The cause of these divergences is one of the major problems of the Septuagint. It is maintained by some that they point to a different Hebrew text from that eventually adopted by the Jews; and since the extant manuscripts of the Septuagint are older by several centuries than the oldest Hebrew manuscripts, it is argued that the Septuagint is our best evidence for the original Hebrew Bible. On the other hand, it is maintained that the Greek translators were not always good Hebrew scholars, that they often made mistakes in translation, and in other respects took liberties with their text. There is at present no sufficient evidence to decide this question; but it is likely that there is some truth in both views. The Hebrew canon and text seem to have been finally fixed about A.D. 100, when the leaders of the Jewish community, having lost their country through the destruction of Jerusalem, and threatened by the growth of Christianity, concentrated on their Scriptures as the centre of their national existence, and fixed authoritatively the contents and the text to which thenceforward they adhered with the most minute punctiliousness. It does not follow that they were always right in the text that they accepted. Even in the Pentateuch, which was always regarded with peculiar reverence, and in which variation was less likely to occur, there are passages in which the Samaritan version agrees with the Septuagint against the Hebrew, and here there is a strong probability that they are the truer witnesses to the original. The same may be the case in other books, and throughout the Old Testament the evidence of the Septuagint should be carefully considered on its merits. On the other hand, mistakes and liberties by the translators are only too probable, and in some cases are certain; so that the evidence of the Septuagint must be accepted with caution. It remains, however, an important fact that it is the form in which the Old Testament was known to most of the Greek-speaking world, and was the basis of the original Latin Bible also. For the earliest Christian communities " the Scriptures " meant the Septuagint Old Testament.

The Septuagint did not remain, however, the only Greek Old Testament. The differences between it and the Hebrew text accepted by the Jewish community about the end of the first century after Christ (which by an anticipation of its later title we may call the Massoretic Hebrew) led in the second century to the production of three complete alternative versions and some partial ones. The first of these (said to have been produced about A.D. 128) was the work of Aquila, a proselyte from Sinope in Pontus, and was executed in the interests of the Jewish community, to whom the Septuagint was unpalatable, not only because of its divergence from the accepted Hebrew, but because it had been adopted by the Christians. It carries Hebraism to an extreme, turning the Hebrew literally into Creek in defiance of Greek idiom. It was consequently used by the Jews and condemned by the Christians. The version as a whole has perished, and is known to us only from citations in ancient authors (collected in Field's Hexapla), a few verses of 1 and 2 Kings and of the Psalter discovered in 1897 in the Genizah (receptacle for imperfect and disused manuscripts) of the synagogue at Cairo, and a fragment among the Amherst Papyri (of the late third or early fourth century) which contains the first five verses of Genesis in the versions of the LXX and of Aquila. The first words illustrate the un-Greek character of Aquila's style: ἐν κεφαλέω ἔκτισεν θεὸς σὸν τὸν οὔπανον καὶ τὴν γὴν· ἡ δὲ γῆ ἦν κένωμα καὶ οὐθέν.

The second version was that of Theodotion of Ephesus, likewise a proselyte to Judaism (or, according to Jerome, an Ebionite Christian), whose work seems to have been a revision of the Septuagint from the Hebrew, in a style far more readable than that of Aquila. Of this version we have better knowledge than of Aquila's; for not only did Origen generally use it to fill lacunae in the LXX text, but his translation of Daniel entirely superseded that of the LXX, which was universally found unsatisfactory; so much so that it is found in practically all MSS. of the Greek Bible, the LXX version surviving only in a single minuscule MS in the Chigi library at Rome, and in a Syriac translation, to which have now to be added several chapters in one of the recently discovered Chester Beatty papyri (No. 968 in the catalogue below). It has also been maintained with a good deal of probability that the ordinary version of Ezra-Nehemiah is in reality that of Theodotion, the original LXX being that which appears in the Apocrypha as 1 Esdras; it is even argued by Sir H. Howorth that Chronicles also is the version of Theodotion, the LXX text having, if so, entirely disappeared. In support of this view is the fact that Josephus, who could not have known Theodotion, certainly used the text of 1 Esdras. There is, however, reason to believe that Theodotion was using some earlier translation of which we have otherwise no knowledge; for several Theodotionic readings occur in the New Testament, notably in the quotations from Daniel in the Apocalypse and Hebrews. There must therefore have been, at least for Daniel, some other version than the LXX extant in the first century, of which Theodotion made use. Thackeray considers that the second translator of the books of Kingdoms used this version.

The third translation, made apparently towards the end of the century, was the work of Symmachus, an Ebionite Christian. His version is, in style, at the opposite extreme to that of Aquila; for he aimed at a free and literary rendering of the meaning of the Hebrew, rather than at minute word-for-word fidelity. It is known to us, like Aquila and most of Theodotion, only in isolated fragments, and it seems to have had less circulation and influence than the other two; but, as will be seen, it was utilized with them by Origen in his great Hexapla, or six-fold version, which will shortly be described, and Jerome made useof him in his Vulgate.

In addition to these three translations, Origen knew of three others, to which no names of authors were attached; but they seem to have been extant only in single copies, they had no general circulation or importance, and there is no evidence that they covered the whole Old Testament. For the present purpose, a simple mention of them will suffice.

The existence of these alternative versions, and the patent divergences between the Hebrew and the Greek Old Testaments, led, in the early years of the third century, to an enterprise which had a far-reaching effect on the text of the Greek Old Testament. This was the work of the great Christian scholar, Origen, known as the Hexapla. Origen (185-253), who from A.D. 203 was head of the catechetical school at Alexandria, was impressed, in his work on a series of commentaries on the books of the Old Testament which he produced between 220 and 250, by the discrepancies which he found in the text, and as a basis for the establishment of the true text he set himself to produce an edition in which the main extant texts were set out in six parallel columns. These columns contained (1) the Massoretic Hebrew text in Hebrew characters, (2) the same transliterated in Greek characters, (3) Aquila, (4) Symmachus, (5) the Septuagint, (6) Theodotion. These were arranged in short κῶλα, or phrases, so as to show the parallel renderings of each small group of words. The most important of these in its ultimate effects was his edition of the Septuagint; for here he set himself to incorporate the Hebrew evidence with the Greek. Words which were in the Hebrew but not in the Greek were added (generally from Theodotion), but marked by an asterisk (x or *); words which were in the Greek but not in the Hebrew were marked by an obelus (+ or —); a metobelus (: or ٪ or ) marked the conclusion of the passage to which the asterisk or obelus referred; where the order of the Greek and Hebrew differed, the Hebrew order was preferred. The general object was to bring the Greek into conformity with the Hebrew, while leaving plain the process by which this result was achieved.

The Hexapla was completed by about A.D. 245, and was followed by a shortened edition, or Tetrapla, in which the two Hebrew columns were omitted. In either form it was a colossal work, and its bulk was fatal to its survival. The originals were preserved in the library at Csesarea in Palestine, where they were seen by Jerome in the fourth century, and by other scholars or annotators as late as the seventh; but it is probable that they were never fully copied, and that they perished when Csesarea was taken by the Arabs in 638. The only extant specimens are a palimpsest leaf at Milan containing a few verses of Ps. xlv (omitting the Hebrew column), and a fragment at Cambridge from the Cairo Genizah, containing part of Ps. xxii in all six columns.

The fifth column, however, containing Origen's revised text of the LXX, was separately edited by Origen's disciples, Pamphilus and Eusebius, and it is this that has affected the subsequent history of the LXX. If Origen's critical symbols had been punctiliously preserved, no harm would have been done; for it would have been possible to re-establish from it the original LXX text on which he worked. But copyists were apt to ignore these troublesome little marks, and the result was that a text was transmitted in which the original LXX was largely contaminated by admixtures from Aquila and especially Theodotion. A considerable number of MSS. exist which give information as to Origen's Hexaplaric text and particular passages in the other columns, but these do not go far towards enabling us to recover the LXX text as it existed before Origen; and this remains the greatest problem which confronts the textual student of the Septuagint. Until we can do that, we are not in a position fully to utilize the evidence of the Greek for the recovery of the pre-Massoretic Hebrew.

Two other recensions of the LXX text were produced during the third century, which are known by the names of Hesychius and Lucian. Hesychius is connected by Jerome with Alexandria, and. is identified by Swete with the bishop whom Eusebius (Hist. Eccl. viii, 13) mentions among the martyrs who suffered in the persecution of Diocletian; but Jerome speaks of him with less respect than one would think he would show to a martyr-bishop. In view of his association with Alexandria, his edition has been identified with the text found in the Egyptian versions and in the works of Cyril of Alexandria; and several MSS. are identified with this type, some of which are mentioned in the catalogue given below. His possible connection with the Alexandrian text of the New Testament will be dealt with in the following chapters.

Lucian, on the other hand, is associated with Antioch, and according to Jerome his edition was current in the churches of Constantinople and Antioch. He was born at Samosata, worked at Antioch, and was martyred under Maximin in 311 or 312. Several MSS. are assigned to his edition, and he has been considered by Hort and others as the originator of the Syrian revision of the New Testament which eventually established itself as the received text in the Byzantine Church. A provisional text of the Lucianic LXX for a considerable part of the Old Testament was produced by Lagarde. Its characteristics are an attempt to obtain lucidity and ease of comprehension, at a sacrifice of what seem to be characteristic features of the true text.

The result of these various attempts to revise the text of the Greek Old Testament has been a great confusion in the tradition. Although the edition of Eusebius (= Origen) circulated primarily in Palestine, that of Hesychius in Egypt, and that of Lucian in Constantinople and Antioch, the various texts reacted on one another, and it is exceedingly difficult now to disentangle them. This is indeed one of the great tasks of those who concern themselves with the textual history of the Septuagint.

The manuscripts in which the Greek Old Testament has come down to us are usually classified under three heads: Papyri, Uncial Manuscripts, and Minuscule or Cursive Manuscripts; and to these must be added the early versions in other languages, the evidence of which is very important. The papyri are the remains (often very small) which have been recovered from excavations in Egypt, almost all within the last sixty years, important because of their age and for the light which they throw on the early history of the text and the conditions under which the Scriptures circulated before the recognition of Christianity by the Empire. The uncials are the vellum manuscripts from the fourth to the ninth or tenth century, written in large capital letters (called "uncial" from a phrase of Jerome's ← Preface to Job, where he complains of the extravagant ostentatiousness of many books in his own time " uncialibus, ut aiunt, lit-teris," i.e. " inch-long," an extravagant, but not impossibly extravagant, description. of somewhat doubtful meaning, but apparently referring to their size). In the ninth century the size of the volumes necessitated by this style of writing appears to have been found cumbrous, and a new and smaller type of writing, developed from the writing in common use for documentary purposes (of which we now have many examples among the latest papyri), was invented and taken into use both for pagan and for Christian literature. It is called " minuscule " from its size or " cursive " from the fact that the letters are often linked together by ligatures so as to make a running hand. This continued in use, with variations in successive generations by which it is possible approximately to date them when other evidence is lacking, until the invention of printing. ← The dating of minuscule Greek manuscripts on purely palaeographical grounds is a hazardous business, since styles of writing were continued long or revived, so that different styles overlap. A great contribution to it has been recently made by the publication of a large number of dated specimens by Prof, and Mrs. Kirsopp Lake (Monumenta Palaeographica Vetera, Boston, 1934 and still in progress).

In the following catalogue only the more important manuscripts are included. The uncials are indicated by capital letters of the Latin, Greek, or occasionally Hebrew alphabets; the papyri and minuscules by Arabic numerals. The first list of manuscripts was that of Holmes and Parsons, in their great edition of the Septuagint, which will be described subsequently; but this has been revised and supplemented by A. Rahlfs, whose catalogue is now the accepted official list of Septuagint manuscripts. Verzeichniss der gr. Handschriften des A.T. in the Nachrichten von der konigl. Gesellschaft der Wissenschaften zu Gottingen, 1914. The maintenance of the list is now in the hands of Dr. W. Kappler, in Göttingen. All MSS. (uncials, minuscules, and papyri) are included in a single numerical sequence, which in the latest supplementary list extends to 2,055. Gaps, however, occur in it, sometimes intentionally to receive additions, sometimes because of errors in Holmes and Parsons; and the actual total is about 1,560. Although Rahlfs includes all in a single list, it seems best to group the three main categories separately, since they represent different dates, the papyri being none of them later than the seventh century and including many of the earliest extant MSS. of the Old Testament, the uncials ranging from the fourth to the tenth century, and the minuscules from the ninth to the fifteenth. In the large Cambridge critical edition, now in progress under the editorship of A. E. Brooke and N. McLean, the readings of a large selection from these MSS. is given; but the minuscules have letters assigned to them instead of numbers. These are in the following list appended to the numbers, with the initials B–M.

The papyri are mostly very small fragments, and only the larger ones are described here. By far the most important are the group known as the Chester Beatty papyri, which deserve a separate mention, both because of their textual importance and because it is to them that we owe the greater part of our knowledge of the earliest form of the Biblical books and their circulation in the first Christian centuries. They are a group of twelve (or more correctly, as is now known, eleven) papyrus codices, discovered in Egypt (apparently in a cemetery or the ruins of a church in the neighbourhood of Aphroditopolis). The greater portion of the collection was secured by Mr. A. Chester Beatty, well-known as a collector especially of illuminated manuscripts; but a considerable number of leaves were acquired by the University of Michigan, a few found their way elsewhere, and some are believed still to be in the hands of the dealers who received them from the original discoverers. The collection, which presumably represents the Biblical library of a Christian community in the fourth century, comprises portions of two MSS. of Genesis, one of Numbers and Deuteronomy, one of Isaiah, one of Jeremiah, one of Ezekiel, Daniel and Esther, and one of Ecclesiasticus; with one of the Gospels and Acts, one of the Pauline Epistles, one of Revelation, and one of the Book of Enoch and a homily by Melito of Sardis. The earliest is of the first half of the second century, the latest not later than the fourth: most are of the third. The discovery was announced in November, 1931, and the texts are in course of publication. ← The texts of all the Chester Beatty papyri have now been published, and complete facsimiles of all except the Numbers— Deuteronomy, Isaiah, Jeremiah and Ecclesiasticus. Descriptions are given below.

The following is a select list of the more important MSS.:


U. British Museum Papyrus 37. Thirty-two leaves of a papyrus codex of the Psalms, acquired in 1836, and said to have been discovered among the rubbish of an ancient convent at Thebes; the first Biblical papyrus to come to light, seventh century. Written in a rather large sloping hand. Contains Ps. x (xi in the Hebrew numbering).2—xviii (xix).6, xx (xxi).14—xxxiv (xxxv).6. The text is said by Rahlfs to belong to the Upper Egyptian family, along with the Sahidic version and 2013. Edited by Tischendorf, Monumenta Sacra Inedita, nov. coll. i, 217.

Freer Greek MS.V

X. Washington Minor Prophets (Freer Greek MS. V). Portions of 33 leaves (out of a probable total of 48) of a single-quire codex of the Minor Prophets. Late third century. Contains, besides a few scraps of Hosea i-viii, Hosea xiv.7-10, Amos i.1-5, 10—ix.15, Micah, Joel, Obadiah, Jonah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, Malachi, in that order. Its text is free from Hexaplaric influences, but shows about 30 accommodations with the Hebrew of its own. Of the greater LXX MSS. it agrees most frequently with Q and decidedly least with א, but it has many readings which are not in any of the uncials, but are found in some of the minuscules, notably 407, 198, 233, 534 and 410, in that order. All these are regarded as pre-Hexaplaric MSS. Edited by H. A. Sanders, with 911.

905 (B-M. U4) Oxyrhynchus Pap. 656, now in Bodleian. Parts of 4 leaves of a codex, early third century. Contains Gen. xiv.21-23, xv.5-9, xix.32-xx.11, xxiv.28-47, xxvii.32, 33, 40, 41. In text it does not range itself closely with any other MS., whether papyrus or vellum. Edited by Grenfell and Hunt (1904).

911. Berlin Staatsbibliothek Gr. fol. 66 I, II. A codex originally of 32 leaves, of which the first and last (the latter being blank) are lost, and all the others more or less mutilated. The first 9 leaves were in double columns, the rest in single columns with long lines. Very irregularly written., with abnormal divisions of words at the ends of lines, in a document hand probably of the early part of the fourth century. Contains (but with many mutilations) Gen. i.16-22, ii.5-9, 11—iv.7, 9—v, 13, 15-22, 24—vi.17, 19—viii.2, 4—xv.7, 10—xvi.2 (xv.4-8, 11—xvi.2 repeated), xvi.3-6, 8—xvii.10, 12-24, 26—xviii.18, 22—xix.4, 15-22, 33—xx, i, 11-18, xxi.13-17, 29—xxii.2,13-17, xxiii.6-12, xxiv.4-8, 20-23, 37—xxxv.8, ending with the title γενεσις κοσμου, showing that the rest of the book was contained in a second codex. The text shows strong affinities with the other papyri of Genesis, 961 and 962. Edited by H. A. Sanders and C. Schmidt (University of Michigan Studies, Humanistic Series, vol. xxi, New York, 1927).

919. Heidelberg Septuagint Papyrus 1. Twenty-seven leaves, all more or less mutilated, from 4 quires (two of 8 leaves and two of 10) of a codex of the Minor Prophets. Seventh century. Written in a large and coarse uncial hand. Contains Zech. iv.6—v.1, 3—vi.2, 4-15, vii.10—x.7, xi.5—xiv.21, Mal. i.1—iii.7, 10, 11, 13-16, 18—iv.1, 3-5. In text belongs to the Hesychian group headed by A and Q. Edited by A. Deissmann, Veröffentlichungen aus der Heidelberger Papyrus-Sammlung I (1905).

P957. JR 456. Deut. fragments.

957. John Rylands Library, Manchester, Pap. Gk. 458. Fragments from 4 columns of a roll of Deuteronomy from the cartonnage of a mummy. Acquired by Dr. Rendel Harris in 1917. Written in a fine book-hand of the second century B.C. Contains Deut. xxiii.24—xxiv.3, xxv.1-3, xxvi.12, 17-19, xxviii.31-33. The text agrees markedly with Θ and A rather than "With B (agreeing in this respect with 963). The earliest known fragment of a Biblical MS. Edited by C. H. Roberts, Two Biblical Papyri in the John Rylands Library, Manchester (Manchester, 1936).

P961. Genesis.

961. Chester Beatty Papyrus IV. Fifty leaves, all more or less mutilated, from a codex originally of 66 leaves of the book of Genesis. Written in double columns, in a rather large and heavy uncial hand. Fourth century. Contains Gen. ix.1-5, 10-13, 16-27, x.2-9,14-20, 22—xi.5, 8-11,13-19. 21-32, xii.3-20, xiii.2-7, 9-13, 15—xiv.18, 20—xv.14, xvii.7—xviii.8, 10-25, 27-33, xix.1—xxi.2, 5-30, 32—xxii.18, 20— xxiii.3, 6-20, xxiv.2-20, 22-27, 29-67, xxv.2-7, 9-26, 28-34, xxvi.2-16,18-22, 24-34, xxvii.1-30, 32—xxix.13, 15-20, 22-27, 29-33. 35—xxx.4, 6-12, 14-16, 18-22, 25-38, 40—xxxi.1, 3-7, 9-18, 20-25, 27-30, 32-39. 41-47, 50-54. xxxii.1-5, 7-10, 12-23, 25-29, 32— xxxiii.2, 4-7, 10-18, xxxiv.1-5, 7-9, 11-14, 16-28, 30—xxxv.1, 3-5,7-16,18-22, 26—xxxvi.1, 4-6,9-13, 15-19, 22-28, 31-35, 37-40. 43—xxxvii.2, 4-7, 9-11, 14-16, 19-21, 24-31, 33-36, xxxviii.2-7, 9-15, 17-20, 22-29, xxxix.1-7, 9-15, 18, 19, 22, 23, xl.2-5, 7-10, 13, 14, 16-18, 20—xli.1, 3-7, 9-12, 14-17, 19-21, 24-27, 30-33, 35, 36, 39-42, 44-46, 48-50, 52-55, 57— xlii.2, 11-13, 16-18, 28-30, 33, 34, xliii.6, 7, 9-11, 21, 22, 25, xliv.17,.21, 22. In text it has several readings in common with 962, with little or no other support, and is also akin to 911. Of the other MSS. it has most agreement with G and 135, the representatives of the Origenian text, and least with A. א and B being almost wholly lacking in Genesis, this group of three early papyri, covering the greater part of the book, is particularly valuable. Edited by F. G. Kenyon, Chester Beatty Biblical Papyri, fasc. iv (London, 1934), with full photographic facsimile in a separate volume (1935).

P962 CBV.

962. Chester Beatty Papyrus V. Twenty-seven leaves (17 nearly perfect) out of an original total of 84, arranged in quires of 10 leaves (except the first, which seems to have had 4 leaves, three pages of which were blank). Written in a document hand of the latter part of the third century, with one column to the page. Contains Gen. viii.13-16, 21—ix.1, xxiv.13—xxv.21, xxx.24-26, 29, 30, 33-35, 37-41, xxxi.5-9, 35-38, 40, 41, 43-45, 47—xxxii.1, 5-9, 14—xxxiii.8, 10—xxxv.16, xxxix.3—xl.4, 6-13, xli.9—xlvi.33. The text is akin to that of 961. Edited by F. G. Kenyon with 961, with full photographic facsimile in a separate volume (1936).

CB  papyrus VI. P963.

963. Chester Beatty Papyrus VI. Fifty leaves (28 substantially preserved) out of an original total of 108 of a codex containing Numbers and Deuteronomy. Written in a small and good hand, probably of the first half of the second century, with two columns to-the page. Contains Num. v.12—viii.19, xiii.4-6, 17, 18, xxv.5, 6, 10, 11, 18, xxvi.1, 2, 12, 13, 21-25, 32-35, 40-42, 47-51. 55-59. 63—xxviii.17, 19—24, 26-31, xxix.1-6, 8-18, 21-24, 27-33, 35-39, xxx.1-17, xxxi.1, 2, 6-8, 11-17, 19-22, 25-28, 30-32, 35-45. 48-50, 54, xxxii.1, 2, 5-8, 11-30, 32, 33, xxxiii.8, 9, 53-56, xxxiv.1-8,12,13, 20-23, 29, xxxv.1-3, 5-7,12, 13-15, 24, 25, 28-32, xxxvi.1, 4, 7, 8, 11-13; Deut. i.20-33. 35-46. ii.1—iii.21, 23—iv.49, v.1—vii.10, 12, 13, 15-20, ix.26, 29, x.1, 2, 5-7, xi.12, 19-21, xi.12, 13, [17, 18, 31, 32], xii.2-4, 15-17, xviii.22, xix.1, 4-6, 10, 11, 13, 14, 16, xxvii.6-8,13-15, xxviii.1-4, 7-10, 12, 13, 17-20 [22-25, 27-30, 32-35, 38-41]. 43-68, xxix.1-18, 20, 21, 23-27, xxx.1,4-6, [10,11], 12,13 16, 17, [19, 20.] xxxi.[3, 4], 8-15, 18, 21-23, 26-29, xxxii.[3-5, 10], 11-13, 17-19. 24. 25 [27-29]. xxxiii.24-27, xxxiv.11, 12. The verses between brackets are on fragments belonging to the University of Michigan, and there are a large number of very small unplaced fragments. In Numbers the text is most akin to that of B and a2, and next to G; but in Deuteronomy to G, Θ, 54, 75, and a2, and least to B. Θ, 54, and 75 in this book form a closely connected group. Edited by F. G. Kenyon, Chester Beatty Biblical Papyri, fasc. v (London, 1935). A photographic facsimile is in preparation.


964. Chester Beatty Papyrus XI. One complete leaf and part of a second of a codex of Ecclesiasticus, written in a large rough hand, probably of the fourth century. Contains Ecclus. xxxvi.28—xxxvii.22, xlvi.6-11, 16—xlvii.2. In text most akin to B, and least to א, A and C coming between. Edited by F. G. Kenyon, Chester Beatty Biblical Papyri, fasc. vi (London, 1937), with 965 and 966.


965. Chester Beatty Papyrus VII. Fragments of 33 leaves (out of an estimated total of 112, with 16 blank leaves in addition at the end) of a codex of Isaiah, arranged in a single large quire. Two of the leaves are the property of Mr. W. Merton, and several fragments originally belonged to the University of Michigan, but were transferred by it to Mr. Beatty. Written in a fine hand, apparently of the first half of the third century. Contains Isaiah viii.18, 19, ix.2, 3, xi.5-7, 10-12, 15—xii.1, 5,—xiii.2, 6-9,12-14, 18-20, xiv.2-4, 23-27, 29—xv.1, 3-5, 8—xvi.4, 7-10,12—xvii.3, [5-7, 9-12], xviii.1-4,6—xix.1, 5, 6, 11-13, xxxviii.14, xxxix.7, 8, xl.22, 23, xli.25, 28—xlii.2, 5-7, 10-13, 16-18, 22—xliii.2, 6-8, 10-13, 17-20, 25—xliv.1, 5-7, 11, 12, 15-17. 19-21, 23-25, 28, xlv.5, liv.1-5, 10, 11, [14—Iv.1, 3-7]. 9-!2, lvi.1-3, 6, 7, 11—lvii.3, 9, 10, 15, 17-20, lviii.2-4, lix.1-3, 7-9, 12-14,17-20, lx.2-5, 9,10, 14-16,19-22. The verses in brackets are those in the Merton leaves. Several short Coptic notes (but in Greek writing) are added in the margins. The text agrees most with Q and Γ  less with A, א and B, in that order. Edited by F. G. Kenyon, Chester Beatty Biblical Papyri, fasc. vi, with 964 and 966.


966. Chester Beatty Papyrus VIII. Small portions of two leaves of a codex, containing Jer. iv.30— v.1, 9-14, 23, 24. Written in a clear and rather large, but not calligraphic, hand, probably of the late second or early third century. The text shows slightly most agreement with Q, and least with K: there are several singular readings, of no great importance. Edited by F. G. Kenyon, Chester Beatty Biblical Papyri, fasc. vi, with 964 and 965.

P968. CB pap.X.

967, 968. Chester Beatty Papyri IX, X. Portions of a codex (originally described as two separate codices) containing Ezekiel, Daniel and Esther, the Ezekiel being in a different hand from the other books. Eight leaves of Ezekiel, 13 of Daniel, and 8 of Esther (conjoint with the Ezekiel) belong to Mr. Beatty, and recently the University of Princeton has acquired 21 leaves of Ezekiel. The complete codex may be estimated at 118 leaves, in a single quire. Probably first half of third century. Contains Ezek. xi.25—xii.6, 12-17, 23—xiii.6,11-17, 20—xiv.3, 6-10,15-20, 23— xv.7, xvi.5-11, 16-22, 28-34, 39-45, 48-53. 57—xvii.1, 6-10, 15-21, [xix.12—xxxix.29, with gaps of 5 leaves, Princeton]; Daniel iii.72-78, 81-88, 92-94, 96—iv.9 (omitting iii.98-100 and iv.3-6), 11-14a 16-19. 22-24, 28-30c, 34-34, vii.1-6, 8-11, 14-19, 22-25, 28—viii.4, 7-11, 15-20, 24—v.5, 7-12,17-26 (omitting 18-22, 24, 25), vi.1-8, 12, 13, 16-18; Esther ii.20-23, 4-8, 13—xiii.3, 5—iii.14, iv.3-7, 11-16, xiii.12-17, xiv.3-8, 13-17, xv.5-10, 16—v.4, 9-14, vi.3-6, 11-14, vii.6-9, viii.2-6. The text of Ezekiel (in the Chester Beatty leaves) agrees markedly with B and Γ (where extant), and next with Q, but not with A. In Daniel the text is the original LXX (not, as in all other Greek MSS. but one, that of Theodotion) without the Hexaplar additions. In Esther it agrees with B and א, but markedly not with A. (Edited by F. G. Kenyon, Chester Beatty, Biblical Papyri, fasc. vii (London, 1937), the Princeton leaves by A. C. Johnson, H. S. Gehman, and E. H. Kase (Princeton, 1938). Pap. Fouad 1. Deut. xxxi.28-xxxii.7. Second Century B.C. Published in Biblical Archaeologist ix. 2, May 1946.)

2013. Leipzig Papyrus 39. Portions of a papyrus roll, about 13 feet 6 inches long, containing Ps. xxx— lv, written on the back of a document dated A.D. 338. Late fourth century. The first five Psalms are much mutilated. The text is said by Rahlfs to contain the Upper Egyptian text, which is found also in the Sahidic version and the British Museum Papyrus U (see above, p. 40). Edited by C. F. Heinrici, Beiträge zur Geschichte und Erklärung des A. T. pt. iv. (1903).

2055. Papyrus Società Italiana 980. Two leaves of a codex of the Psalms, containing Ps. cxliii, 14— cxlviii, 3. Late third or fourth century. Its text appears to be mixed; it agrees least with א, but about equally with A B R T. In several instances it is found supporting the same readings as אca R T. (Edited by G. Vitelli, Papiri Societa Italiana, vol. viii, fasc. 2, 1927.)


The four great codices which contain both Testaments are more fully described in the New Testament list.

Codex Sinaiticus

א or S: Codex Sinaiticus, fourth century. See p. 75. The Old Testament portion consists of 43 leaves at Leipzig, fragments of 3 at Leningrad, and 199 in the British Museum. Originally it was a magnificent copy of the entire Greek Bible, written in a beautiful hand on about 720 leaves of fine vellum, measuring 15 by 13½ inches, with four columns to the page (two in the poetical books); but the Old Testament portion had suffered severely during its residence at Sinai, and not much more than one-third of it has survived. The extant portions are Gen. xxiii.19—xxiv.14, 17-19, 25-27, 30-33, 36-46 (all with mutilations), Num. v.26-30, vi.5, 6, 11, 12, 17, 18, 22, 23, vii.4, 5, 12, 13, 15-20, 1 Chr. ix.27—xix.17, 2 Esdr. ix.9—end, Esther, Tobit, Judith, 1 and 4 Maccabees, Isaiah, Jeremiah, Lamentations i.1—ii.20, Joel, Obadiah, Jonah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, Malachi, Psalms, Proverbs, Ecclesiastes, Song of Solomon, Wisdom, Ecclesiasticus, Job. According to both Tischendorf, who discovered and first edited the MS., and Lake, who edited a complete facsimile from his own photographs, four scribes were employed on it. Several correctors have also been at work on it, some contemporary (or identical) with the original scribes, and others later. By far the most important of the corrections are those made by a group of scholars (denoted by the symbol אc.a. or אc.b.), one of whom wrote notes at the end of Esdras and Esther, stating that the MS. was collated with an exceedingly early copy corrected by the hand of the martyr Pamphilus, with an autograph note by him saying that he corrected it in prison from Origen's own copy of the Hexapla. This note is probably of the sixth or early seventh century, and makes it extremely probable that the MS. was then at Caesarea, where the library of Pamphilus was, and also that the corrections in these hands were taken from a very early MS. which itself was only by one step removed from Origen. This gives exceptional value to this group of corrections. In general א has the same type of text as B, but according to Ropes it is superior in the Prophets and in Chronicles and 2 Esdras. In Tobit it has a different recension.

Codex Alexandrinus

A. Codex Alexandrinus, fifth century. See p. 82. The Old Testament portion occupies 630 leaves, measuring 12⅝ by 10⅜ inches, written with two columns to the page. Ten leaves have been lost, containing 1 Sam. xii.18—xiv.9, Ps. xlix (1).20—lxxix (lxxx).11. A few verses in Genesis are lost by mutilation. Otherwise the whole Old Testament is present, together with 3 and 4 Maccabees, while a table of contents shows that the Psalms of Solomon were originally appended at the end of the whole MS. Two scribes were employed on it, of whom one wrote the Octateuch, Prophets, Maccabees, and the poetical books Job, Proverbs, Ecclesiastes, Song of Solomon, Wisdom and Ecclesiasticus, while the other wrote the historical books from 1 Kingdoms [= 1 Sam.] to the end of Chronicles, Esther, Tobit, Judith, 1 and 2 Esdras, and Psalms. The MS. has been extensively corrected, especially in the Pentateuch, often by the original scribes, but oftener by a corrector (Aa) who seems to be about contemporary.

Codex Vaticanus.

B. Codex Vaticanus, fourth century. See p. 85. The Old Testament portion occupies 617 leaves, measuring 10½ by 10 inches, written with three columns to the page. Thirty-one leaves are lost from the beginning, containing Gen. i. 1—xlvi, 28, and ten leaves from the Psalter, containing Ps. cv (cvi).27— cxxxvii (cxxxviii).6; while 2 Kingdoms [2 Sam.] ii.5-7, 10-13 are lost by mutilation. The books of Maccabees were never included in it; otherwise all books are present. The poetical books precede the prophets, whereas in א and A they are at the end of the Old Testament. The hand is smaller than in א and A, but is graceful in formation. There are corrections by various hands, those indicated as B2 being the best. B is generally regarded as having the best text of any single MS. of the LXX, and has therefore been made the basis of the two Cambridge editions. This is perhaps partly due to its pre-eminence in the New Testament, and it must be recognized that in the Old Testament it is often found in opposition to the other earliest uncials, and the question of superiority must be regarded as still uncertain. It also seems clear that its character is not uniform throughout; as is only natural, since it must have descended from a number of different rolls, which might well be of different characters. Burkitt observes that it has a good text in Ezekiel, but a bad one in Isaiah ; and this is confirmed by the papyri. In Judges it has a text differing substantially from that of A and the majority of authorities, ← Both texts are printed in full in Rahlfs' edition of the LXX. In the Cambridge editions B is printed, with the variants of A in the apparatus. According to G. F. Moore (in his edition of Judges) the B text is the later. but found in the Sahidic version and in Cyril of Alexandria. In Job it has the additional 400 half-verses from Theodotion, which are not in the Old Latin and Sahidic versions; i.e., it is post-Hexaplaric. The early papyri support B rather than A in Numbers, Ezekiel, Esther, and Ecclesiasticus, but A rather than B in Deuteronomy and Isaiah. According to Ropes B is distinctly inferior to A also in Chronicles and 1 and 2 Esdras. He considers that in Joshua, Ruth, 1-4 Kingdoms, Chronicles, 1 and 2 Esdras, Psalms, Ezekiel, Esther its text is pre-Hexaplar, but in Isaiah, Jeremiah, the Minor Prophets and Job it is Hexaplar; and Rahlfs agrees in respect of the Psalter.

Codex Ephraemi

C. Codex Ephraemi, fifth century. See p. 88. [katapi ed: The Codex Ephraemi page index is HERE.] Portions of a fine MS. of the whole Bible which in the twelfth century was converted into a palimpsest, the original writing being sponged away and some works of Ephraim Syrus written over it. A large part of the MS. has disappeared altogether, and the original writing of what is left cannot always be discerned. Of the Old Testament only 64 leaves remain, containing portions of the poetical books, viz. Prov. i.2—ii.8, xv.29—xvii.1, xviii.11—xix.23, xxii.17—xxiii.25, xxiv.22-56, xxvi.23—xxviii.2, xxix.48—end ; Eccl. i.2-14, ii.18—end. Cant. i.3—iii.9; Job ii.12—iv.12, v.27—vii.7, x.9—xii.2, xiii.18—xviii.9, xix.27—xxii.14, xxiv.7—xxx.1, xxxi.6—xxxv.15, xxxvii.5—xxxviii.17, xl.20—end; Wisd. viii.5—xii.10, xiv.19—xvii.18, xviii.24—end; Ecclus. prol. 1—vii.14, viii.15—xi.17, xii.16—xvi.1, xvii.12—xx.5, xxi.12—xxii.19, xxvii.19—xxviii.25, xxx.8—xxxiv.22 ← In all Greek MSS. xxx, 25—xxxiii 16a is transposed to follow xxxiii, 16b—xxxvi, 10., xxx.25—xxxi.5, xxxii.22—xxxiii.13, xxxvii.11—xxxviii.15, xxxix.7—xliii.27, xiv.24—xlvii.23, xlviii.11—xlix.12. It is written with a single column to the page, which is unusual in uncials.

D. Cotton Genesis, in the British Museum, fifth century. Written in a fine hand, with 250 illustrations, but hopelessly ruined by fire in 1731. Fortunately it had previously been collated by Grabe, and for most of its readings it is necessary to depend on his evidence.

E. Bodleian Genesis, at Oxford, tenth century. Somewhat seriously mutilated, causing the loss of Gen. xiv.7—xviii.24, xx.14—xxiv.54. The Bodleian MS. ends with the first words of xlii.18, but a leaf containing xlii.18—xliv.13 is at Cambridge. Of this leaf the first page is written in uncials and the second in minuscules; and this revealed the fact that the MS. originally contained much more than Genesis, and that all except Gen, i.1—xlii.31 was written in minuscules. A MS. at Leningrad (cod. lxii) carries on the text from the Cambridge leaf to 3 Kingdoms xvi.28, with some lacunae, of which the largest (Jos. xxiv.27—end of Ruth) is in the British Museum (Add. MS. 20002). Written in sloping uncials of a very late type, and in good minuscule hands by three different scribes. In spite of its late date, its text is valuable. The minuscule portion has the symbol a2 in B—M.

F. Codex Ambrosianus, at Milan, fifth century. Fragments of the Hexateuch, written (like B) with three columns to the page. Contains Gen. xxxi.15-37, xlii.14-21, 28—xlvi.6, xlvii.16—xlviii.3, 21—1, 14 ; Exod. i.10—viii.19, xii.31—xxx.29, xxxi.18— xxxii.6, 13—xxxvi.3, xxxvii.10—end ; Lev. i.1— ix.18, x.14—end ; Num. complete ; Deut. i.1—-xxviii.63, xxix.14—end ; Jos. i.1—ii.9, 16—iv.5, 10—v.1, 7—vi.23, vii.1—ix.33, x.37—xii.12. Many various readings and notes in the margins.

Codex Sarravianus

G. Codex Sarravianus, mainly at Leiden (130 leaves), but 22 leaves in Paris and one in Leningrad, fifth century. It contains (in its present state) portions of the Heptateuch, viz. Gen. xxxi.54—xxxvi.18; Exod. xxxvi.35—xxxvii.21, xxxviii.24— xxxix.21, xxxix.37—end ; Lev. i.1—xiii.17, 49—xiv.6, 33-49, xv.24—xvii.10, xviii.28—xix.36, xxiv.9—xxvii.16 ; Num. i.1—vii.85, xi.18—xviii.2, 30—xx.22, xxv.2—xxvi.3, xxix.12—end ; Deut. iv.11-26, vii.13—xvii.14, xviii.8—xix.4, xxviii.12— xxxi.11; Jos. ix.33—xix.23; Judg. ix.48—x.6, xv.3—xviii.16, xix.25—xxi.12. The special importance of this MS. lies in the fact that it is a Hexaplar MS. having the critical marks of Origen, but unfortunately not completely. It is thus of much value as a guide to the pre-Hexaplar text on which Origen worked.

I. Codex Bodleianus of the Psalms, in the Bodleian, ninth century. Valuable because in the margin many readings are given from Aquila, Theodotion, and Symmachus, and from the " fifth " and " seventh" versions used by Origen. Numbered by Holmes and Parsons as 13, as though it were a minuscule.

L. Vienna Genesis, fifth or sixth century. Written on purple vellum, with illustrations in classical style, like those of D. Twenty-four leaves of Genesis, containing Gen. iii.4-24, vii.19—viii.1, 4, 13-17, 19, 20, ix.8-10, 12-15, 20-27, xiv.I7~20, xv.1-5, xix.12-17, 24, 26, 29-35, xxii.15-19, xxiv.1-4, 9-11, 15-20, 22-25, 28, 29, 31, xxv.27-34, xxvi.6-11, xxx.30-37, xxxi.25-34, xxxii.6-8, 13-18, 22-32, xxxv.1-4, 8, 16-20, 28, 29, xxxvii.1-19, xxxix.9-18, xl, 14—xli.2, 21-32, xiii.21-25, 27—xliii.22, xlviii.16—xlix.3, 28—I.4.

Codex Coislianus

M. Codex Coislinianus, at Paris, seventh century. 227 leaves, containing Gen. i.1—xxxiv.2, xxxviii.24—Num. xxix.23, xxxi.4—Jos. x.6, xxii.34—Ruth iv.19,1 Kingd. i.1—iv.19, x.19—xiv.26, xxv.33—3 Kingd. viii.40. Has Hexaplaric signs in the margin and belongs to the same class as G.

N-V. Codex Basiliano-Vaticanus, eighth or ninth century. In two volumes (mutilated), one at Rome and one at Venice, formerly regarded as two distinct MSS., and therefore numbered separately, as N and V. Contain Lev. xiii.59—end of Chron. (with lacunae), 1 Esdras i.1—ix.1, 2 Esdras v.10—xvii.3, Esther, Job xxx.8—end, Proverbs, Ecclesiastes, Song of Solomon, Wisdom, Ecclesiasticus, Prophets, Tobit, Judith, 1-4 Maccabees. Numbered by H—P as 23, as if it were a cursive. Its text is regarded as Lucianic.

O. Codex Dublinensis Rescriptus, at Dublin, sixth century. Eight palimpsest leaves, containing Isaiah xxx, 2—xxxi, 7, xxxvi, 19—xxxviii, 2. Apparently Hesychian in text, and therefore useful.

Codex Marchalianus

Q. Codex Marchalianus, at Rome, sixth century. A very handsome MS. of the Prophets, written in Egypt. Its editor, Dr. Ceriani, has shown that the text is that of the edition of Hesychius, and where Hesychius has himself introduced words which were not in the original LXX, such words are marked by the original scribe by an asterisk in the text itself. Further, an almost contemporary scribe has added both in the text and in the margins a great number of critical signs indicating the modifications made by Origen which Hesychius had adopted, together with many readings from Aquila, Symmachus, and Theodotion, as well as from Origen's own text. It is therefore a principal authority for the text of the Hexapla in the Prophets.

R. Codex Veronensis, at Verona, sixth century. A Graeco-Latin MS. of the Psalter and canticles, the Greek text being written in Roman characters. In Rahlfs' classification it represents the Western text of the Psalter.

T. Codex Turicensis, at Zurich, seventh century. Contains the Psalter and canticles, but Ps. i-xxv are lost, also xxx.2—xxxvi.20, xii.6—xliii.3, Iviii.24— lix.3, lix.9, 10, 13—lx.1, lxiv.12—lxxi.4, xcii.3— xciii.7, xcvi.12—xcvii.8, and the first five canticles and part of the sixth. Written in silver letters, with gold headings, and with readings of Jerome's Gallican version in red in the margins. Generally agrees with A, and still more with אc.a..

U. See under Papyri.

V. Codex Venetus. See N-V, above.

Γ. Codex Cryptoferratensis, at Grotta Ferrata, eighth-ninth century. Palimpsest fragments of a MS. of the Prophets, in a text which seems to be of good quality and akin to that of Q and 965; but so much has been completely obliterated that its readings are only incompletely available.

Washington MS1

Θ. Codex Washingtonianus I (Rahlfs, W.) in the Freer Library in Washington, sixth century. A codex of Deuteronomy and Joshua, but originally (as the quire-numeration shows) containing the previous books of the Hexateuch, while Judges and Ruth may have been added at the end. Two double leaves are missing, containing Deut. v.16—vi.18, Jos. iii.3—iv.10. In Deuteronomy its text agrees more with A than with B, but its closest association seems to be with the minuscules 54 and 75, and often with G and the Beatty papyrus 963. It is probable therefore that it is to be assigned to the Origenian group. In Joshua the grouping is different; for though Θ still agrees with A more than with B, it parts company with 54 and 75.

Washington MS2

1219. Codex Washingtonianus II in the Freer Library at Washington, sixth-seventh century. A much mutilated MS. of the Psalms, of which 107 fragmentary leaves are preserved, extending from Ps. i.1 to cxlii.8, but very incomplete, especially in the earlier psalms. The final quire must have been lost early, and has been replaced by a quire from another MS. probably of the ninth century, containing Ps. cxlii.5—cli.6. According to Rahlfs, its text is akin to A, forming with it and minuscule 55 a group which is distinct both from the Lower Egyptian group (B א Boh.) and the Upper Egyptian group (U, 2013, Sah.), and also from groups which he classifies as Western, Origenian, and Lucianic. Both Washington MSS. have been edited by H. A. Sanders.


The first list of minuscule MSS. was compiled by Holmes and Parsons (see p. 63), who, after beginning with thirteen Roman numerals for uncial MSS., continued with Arabic numerals from 14 to 311 (with two numbers duplicated) for the minuscules (of. which six were actually uncials). This list has been extended by Rahlfs, whose figures now run up to 2,055 (including the papyri), with however some gaps deliberately left, which are being gradually filled. Most of these MSS. do not contain the whole Old Testament, but some particular division of it. Brooke and McLean have made selections from these, and utilized them for their large critical edition, from which the groupings into which they fall are emerging. It is on this that their value principally depends; for their importance lies in the help they give in elucidating the several editions of Origen, Hesychius and Lucian. This investigation is by no means complete, and the attributions of various editors must be accepted with some reservation. ← Lagarde, in editing the Lucianic text, used 19, 82, 93 and 108. Hautsch (in Mittheilungen des LXX-Unternehmens, Göttingen, 1909) questions this in the Hexateuch since the readings in the Antiochian Fathers do not usually agree with those of 19 and 108. In the historical books Rahlfs (Septuagintastudien, 1911) considers 82 and 93 superior to 19 and 108. He believes the basis of Lucian's edition was a pre-Hexaplar text akin to B and the Aethiopic version.
Among the more important are the following:

13 = I above

19 (B–M. b). At Rome, 10th century. Contains Octateuch and historical books. Closely akin to 108. Lucianic in text.

23 = N–V above.

26 At Rome, 11th century. Contains Prophets, in a Hesychian text.

36 At Rome, 13th century. Contains Prophets, in a Lucianic text.

54 (B–M. g.) At Paris, 13th-14th century. Contains Octateuch, in an Origenian text.

82 At Oxford (New Coll.), 13th century. Contains Prophets in a Lucianic text.

75 (B–M.n). At Oxford (Univ. Coll.), A.D. 1126. Contains Octateuch; said to be Lucianic in Genesis, but closely allied with 54 in Deuteronomy.

82 (B–M. o). At Paris, 12th century. Contains Octateuch and historical books in a Lucianic text.

85 At Rome, nth century. Contains Heptateuch (imperfect). Said by Field to be of high class, but not used in B–M.

87 At Rome, 9th or 11th century. Contains Major Prophets, Daniel being in the original LXX, of which this is the only complete Greek MS.

90 At Florence, 11th century. Contains Major Prophets (including Daniel). Classed as Lucianic by Field, but said by Cornill to be Hesychian in Ezekiel.

91 At Rome, 11th century. Contains Prophets, in a Hesychian text.

93 (B–M. e2). In British Museum, 13th century. Contains Ruth, 1-4 Kingdoms, 1 and 2 Chronicles, x and 2 Esdras, Esther (in two different texts), 1-3 Maccabees, and Isaiah; but the quire numeration shows that it formerly contained the whole of the Octateuch, and it may have contained the remaining prophets and the poetical books. It has three columns to the page, an old-fashioned arrangement, which suggests that it may have been copied from an early uncial archetype. The text is Lucianic.

108 At Rome, 14th century. Contains Octateuch and historical books. See 19 above. This was the MS. mainly used for the Complutensian Polyglot (see below).

118. At Paris, 13th century. Contains Octateuch (imperfect) in a Lucianic text.

135 (B–M. e2). At Basle, nth century. Contains Genesis and Exodus i.1—xii.4, in an Origenian text.
198 In Paris, 9th century. Contains Prophets, in a Hesychian text. This MS. also contains the New Testament in a very good text (33 in N.T. list).

838 At Rome, 13th century. Contains poetical books and prophets, and in the latter at least is said to be Hesychian.

238 At Rome, 10th century. Contains Ezekiel, in a Hesychian text.

308 At Rome, 13th century. Contains Isaiah, in a Lucianic text.

This list is very imperfect. Many of the minuscules, especially those more recently added to the list, have not been fully examined, and less work has been done on the historical and poetical books than on the Octateuch and the Prophets. But the majority of the minuscules contain mixed texts, and are not of much use in arriving at the original form of the several editions, still less of the primaeval Septuagint.

Besides the Greek MSS., valuable evidence can be derived from the early translations that were made not from the Hebrew but from the Septuagint. These include the Old Latin, Coptic and Syriac versions. The Old Latin was so entirely superseded by Jerome's Vulgate that it has not come down to us complete, and can only be recovered from imperfect MSS. and quotations in the early Latin fathers, especially Cyprian. So far as these go, they give us a pre-Hexapla text. The apocryphal books have fared better than the canonical, since (except Tobit and Judith) they were not revised or re-translated by Jerome, but were incorporated direct in the later Latin Bible. Of the other books, only Numbers (with most of Leviticus) Deuteronomy, Joshua, Judges, Ruth, i Esdras, Esther, Judith, Tobit, i and 2 Maccabees, Baruch and the Psalms have come down complete. The largest single MS. is the Codex Lugdunensis, of the fifth century, which contains portions of Genesis, Exodus, and Leviticus, and the whole of Numbers-Judges.

When Jerome was invited in 382 by Pope Damasus, in view of the great diversities of text then in existence, to prepare a revised text of the Latin Bible, he began with the New Testament, which was completed by about 391. Proceeding to the Old Testament, he first made a very slight revision of the Old Latin Psalter, with reference only to the LXX; this version is still in use at St. Peter's, and is known as the Roman Psalter. Next (between 387 and 390) he made a more thorough revision, still only from the'LXX, which is known as the Gallican Psalter, and is the Psalter generally in use in the Roman Church (just as in our English Prayer-Book we still use the Psalter from the Great Bible of Coverdale, rather than the improved translation of the Authorized Version). Finally, after producing a version of Job from the Hexaplaric text, having by this time satisfied himself of the superiority of the Massoretic Hebrew over the Greek, he abandoned the idea of revising the Old Latin from the Septuagint, and undertook a fresh translation of the whole (including the Psalter, which in this form is known as the Hebrew Psalter) from the Hebrew. With this, as it has no bearing on the text of the LXX, we are not concerned.

The earliest Egyptian version is that produced in Upper Egypt in the Sahidic dialect of Coptic. This was probably made before the end of the second century, and was therefore pre-Hexapla. Thus Job in this version is in the shorter text which we know to have been characteristic of the original LXX. Subsequently, however, the Sahidic version was revised, and many MSS. show Hexaplaric additions. Being Egyptian in origin, it would naturally be influenced mainly by the edition of Hesychius, and where it is not pre-Hexapla it may be reckoned as Hesychian. Until recently it was known only from rather numerous fragments (collected by Ciasca, Revillout and others), but discoveries in Egypt of late years have added greatly to the material. They include a complete MS. of Deuteronomy and Jonah (with Acts) of the fourth century, in the British Museum, edited by Budge (1912), with revision by Sir H. Thompson (1913); a palimpsest of the seventh century in the British Museum, containing Joshua, Judges, Ruth, Judith and Esther, nearly complete, edited by Sir H. Thompson (1912); 62 leaves of a MS. of the Sapiential Books in the British Museum, of about the seventh century, containing considerable fragments of Proverbs, Eccle-siastes, Song of Solomon, Wisdom and Ecclesiasticus, with a small fragment of Job, edited by Sir H. Thompson (1908); three MSS. in the collection of Mr. Pierpont Morgan, one (dated 893) containing 1 and 2 Kingdoms, another (8th-9th century) Leviticus, Numbers and Deuteronomy, and the third (9th century) Isaiah (none of these is yet edited); a complete Psalter, 7th century, in the British Museum, edited by Budge (1898); a fragmentary Psalter of the 6th century in the Freer Library, containing (with mutilations) Ps. vi.5—liii.3, edited by Worrell (1916 and 1923); a Psalter divided between the Chester Beatty collection and the University of Michigan, not yet edited, and a fragmentary Psalter at Berlin, edited by Rahlfs and assigned by him to about 400. There is thus now a considerable amount of material for the Sahidic version, which, according to Rahlfs, is in the Psalter marked by a number of additions of a definitely Christian character.

The Sahidic version was eventually superseded by the Bohairic, or Lower Egyptian, version, which became the accepted Bible of Egypt, and consequently survives in a number of MSS., mostly of relatively late date. No complete edition of this version exists, but several books (Pentateuch, Psalter, Prophets, Job) have been edited, though not recently. In general it belongs (as in the New Testament) to the same group as B א. In the Psalter Rahlfs regards it as pre-Hexapla.

There are fragments of other Coptic versions in Middle Egyptian dialects (Akhmimic, Fayyumic, etc.), but they are not extensive, and their character has not been determined, though they appear to lean rather to the Sahidic than the Bohairic.

Of the Syriac versions, the Peshitta, which became the accepted Bible of the Syrian Church, was mainly translated from the Hebrew, probably about the end of the second century. It is said, however, to show considerable traces of the influence of the LXX in the Psalter, Isaiah, and the Minor Prophets, and it included the non-canonical books of the Greek Old Testament, except i Esdras and Tobit. It was completely edited by Lee (1823 and 1826). Much more important for the LXX is the Syro-Hexaplar version produced by Paul, bishop of Telia, in 616-17, at Alexandria, which was a close translation of Origen's Hexapla text, retaining his critical symbols. A MS. containing the poetical and prophetical books is in the Ambrosian Library at Milan; another, now lost, but used by A. du Maes (d. 1573) when he first published this version, contained part of Deuteronomy, Joshua, Judges, 3 and 4 Kingdoms, Chronicles, Ezra, Esther, Judith and part of Tobit: and Genesis, most of Exodus, and Ruth have been edited from other MSS., together with fragments of other books. The translation is fortunately extremely literal, so that we have in it one of the principal authorities for the text of Origen, and through it, by observation of the critical symbols, of the pre-Hexapla Septuagint. It also contains the original LXX of Daniel, which otherwise exists only in the Chigi MS. (no. 87 above) and partially in the Chester Beatty papyrus (968).

Considerable fragments, covering most of the books of the Old Testament, also exist of a Syriac version in the Palestinian dialect, made from the LXX. In the Major Prophets the text appears to be Origenic, and Job contains the additions from Theodotion, which also points to a Hexaplar origin; but the small fragments of 2 and 3 Kingdoms are said to be Lucianic. Other Syriac versions may be ignored here.

The other versions are of comparatively small value. Some use is made by scholars of the Armenian and Ethiopic versions, and the Gothic is of some interest, being known to have been made in the fourth century from a Lucianic text; but details of these would be superfluous here.


The Greek Old Testament was first printed in the great Complutensian Polyglot, produced at Alcala (Complutum) under the direction of Cardinal Ximenes, completed in 1517, but not published until 1522. The four volumes which include the Old Testament give it in Hebrew, Latin, and Greek (with an interlinear Latin translation of the latter). The editors (of whom Stunica appears to have been the chief) used mainly the minuscules 108 and 118, sent from Rome, with other MSS. in Madrid and perhaps one from Venice, but none of very early date. But as in the case of the New Testament, Ximenes' great work was anticipated by a publisher's enterprise; for in February 1518/19 Aldus produced a complete Greek Bible, edited by his father-in-law, Andreas Asolanus. For the OldTestament he seems to have depended on MSS. at Venice, and the edition had no lasting importance, being superseded by the edition published at Rome in 1587 under the auspices of Pope Sixtus V, and editorially directed by Cardinal Carafa. This is a scholarly edition, based on a considerable number of MSS., but principally on the great Codex Vaticanus (B). The uncial N–V was also used. This edition rightly won general acceptance, with the result that the Septuagint (more fortunate than the New Testament) has almost from the first been available in a text largely based on the MS. which is generally regarded as the best extant. The Sixtine text was reprinted in Walton's Polyglot (1657) and in the next great critical edition, that of Holmes and Parsons (1798-1827), which requires special mention. In this, as mentioned above, a list of MSS. is given, 313 in number, and readings from them are quoted in a critical apparatus, which has remained until recently (and for some books still remains) the principal repertory for the text of the LXX. Versions and patristic quotations were also used, so far as available. Unfortunately the collations made for it were not always complete or accurate, so that much of the work requires revision, but it remains a colossal performance, to which scholars will always be deeply indebted. It was begun by R. Holmes, Canon of Christ Church, Oxford, and afterwards Dean of Winchester, in 1788, but he died in 1805, when only one volume had appeared, and the remaining four volumes were edited by J. Parsons.

Constantin Tischendorf produced four editions of the Septuagint, consisting of a revised Sixtine text, with a select apparatus criticus derived from the four principal uncials (א A C and in part B). But the next great advance was made when Dr. H. B. Swete in 1887-94 produced a manual edition in three convenient volumes for the Cambridge University Press, in which the main text is that of B when available, elsewhere א or A, with a select apparatus from א A C, supplemented to some extent from other uncials (principally D E F in the Pentateuch, R T U in the Psalter, Q V Γ  in the Prophets). This edition has been and is extremely serviceable for students. Quite recently its pre-eminence as a handy edition has been challenged by one produced by A. Rahlfs in two larger volumes at Stuttgart (1935), with a revised text based upon א A B, and a short apparatus giving the readings of these MSS. and occasionally of others.

Meanwhile a larger critical edition has been in progress at Cambridge under the editorship of A. E. Brooke (late Provost of King's) and N. McLean (late Master of Christ's). This gives the same text as Swete, with slight modifications, but has a full textual apparatus, giving the readings, not indeed of all extant MSS. (as was aimed at by Holmes and Parsons), but of all uncials and papyri, a considerable number of selected minuscules, all the principal versions, and patristic quotations. This gives, so far as it has gone, all that the scholar can require in the way of a critical edition (without revised text), and its progress to completion is greatly to be desired. At present eight parts have appeared (1906-1935), comprising the Octateuch and the Later Historical Books (1-4 Kingdoms, Chronicles, 1 and 2 Esdras); Progress with the Cambridge Septuagint has been delayed by the death of Dr. Brooke in 1939 and the ill-health and death (in 1947) of Dr. McLean. Part 1 of vol. iii, containing Esther, Judith and Tobit, appeared 1940.

A large critical edition has also been in contemplation by the Septuaginta-Kommission of Gottingen, but material conditions have seriously delayed its progress. The Psalter was published by Rahlfs in 1930-1, and Genesis in a form intermediate between the large edition and the small in 1926. x Maccabees has just appeared under the editorship of Kappler (1936). The German committee is wisely devoting its attention first to the later books, which are not likely to be reached by Brooke and McLean in the near future. 2 Maccabees and Isaiah are now in preparation. Meanwhile the smaller edition of Rahlfs, just mentioned, which is issued by the Wiirttembergische Bibelanstalt at a very moderate price, will be very serviceable.


H. B. Swete, Introduction to the Old Testament in Greek (Cambridge, 1900, 2nd ed. 1914), by far the best and fullest repertory of information up to 1914; F. Buhl, Canon and Text of the Old Testament (Eng. tr. 1892) pp. 108-193; the Mittheilungen of the Göttingen Septuaginta-Kommission; H. Rahlfs, Septuaginta-studien; H. St. John Thackeray, The Septuagint and Jewish Worship, 1920. For the Hexapla, F. Field, Origenis Hexaplorum quae supersunt (Oxford, 1875). For the Chester Beatty papyri, F. G. Kenyon, Recent Developments in the Textual Criticism of the Greek Bible (1933) and editions of the several texts in The Chester Beatty Biblical Papyri (1933-6 and in progress). References are given above to the chief editions of the LXX, Holmes and Parsons, Swete, Brooke and McLean, Rahlfs.

[katapi ed: In recent years, much attention has been given to digitizing MSS., and a good place to search on-line is the CSNTM website. The source MSS websites can accessed through their Manuscripts list. They deal in the main with NT mss., but they are worth checking out none-the-less. (Your might also try the Wikipedia website for more background on individual MSS.)