CHAPTER II THE WRITTINGS OF EUSEBIUS 43 served in the Vita Const. IV. 35. From this epistle we learn, moreover, that the work had been translated into Latin. It is no longer extant in its entirety, but a considerable fragment of it was discovered by Mai in Nicetas’ Catena on Luke, and published by him in his Bibl. Nova Patrum, IV. p. 208 sq. The extant portion of it contains twelve chapters, devoted partly to a discussion of the nature of the Passover and its typical significance, partly to an account of the settlement of the paschal question at the Council of Nicæa, and partly to an argument against the necessity of celebrating the paschal feast at the time of the Jewish Passover, based on the ground that Christ himself did not keep the Passover on the same day as the Jews. Jerome, although he does not mention this work in his catalogue of Eusebius’ writings (de vir. ill. 81), elsewhere (ib. 61) states that Eusebius composed a paschal canon with a cycle of nineteen years. This cycle may have been published (as Lightfoot remarks) as a part of the writing under discussion. The date of the work cannot be determined with exactness. It was written after the Council of Nicæa, and, as would seem from the connection in which it is mentioned in the Vita Constantini, before the Emperor’s tricennalia (335 a.d.), but not very long before. The extant fragment, as published by Mai, is reprinted by Migne in the Opera, VI. 693–706. V. CRITICAL AND EXEGETICAL WORKS. Biblical Texts. We learn from Jerome (Præf. in librum Paralip.) that Eusebius and Pamphilus published a number of copies of Origen’s edition of the LXX., that is, of the fifth column of the Hexapla. A colophon found in a Vatican ms., and given in facsimile in Migne’s Opera, IV. 875, contains the following account of their labors (the translation is Lightfoot’s): “It was transcribed from the editions of the Hexapla, and was corrected from the Tetrapla of Origen himself, which also had been corrected and furnished with scholia in his own handwriting; whence I, Eusebius, added the scholia, Pamphilus and Eusebius corrected [this copy].” Compare also Field’s Hexapla, I. p. xcix. Taylor, in the Dictionary of Christian Biography, III. p. 21, says: “The whole work [i.e. the Hexapla] was too massive for multiplication; but many copies of its fifth column alone were issued from Cæsarea under the direction of Pamphilus the martyr and Eusebius, and this recension of the LXX. came into common use. Some of the copies issued contained also marginal scholia, which gave inter alia a selection of readings from the remaining versions in the Hexapla. The oldest extant ms. of this recension is the Leiden Codex Sarravianus of the fourth or fifth century.” These editions of the LXX. must have been issued before the year 309, when Pamphilus suffered martyrdom, and in all probability before 307, when he was imprisoned (see Lardner’s Credibility, Part II. chap. 72. In later years we find Eusebius again engaged in the publication of copies of the Scriptures. According to the Vita Const. IV. 36, 37, the Emperor wrote to Eusebius, asking him to prepare fifty sumptuous copies of the Scriptures for use in his new Constantinopolitan churches. The commission was carefully executed, and the mss. prepared at great cost. It has been thought that among our extant mss. may be some of these copies which were produced under Eusebius’ supervision, but this is extremely improbable (see Lightfoot, p. 334). Ten Evangelical Canons, with the Letter to Carpianus prefixed (κανόνες δέκα; Canones decem harmoniæ evangeliorum præmissa ad Carpianum epistola). Ammonius of Alexandria early in the third century had constructed a harmony of the Gospels, in which, taking Matthew as the standard, he placed alongside of that Gospel the parallel passages from the three others. Eusebius’ work was suggested by this Harmony, as he tells us in his epistle to Carpianus. An inconvenient feature of Ammonius’ work was that only the Gospel of Matthew could be read continuously, the sequence of the other Gospels being broken in order to bring their parallel sections into the order followed by Matthew. Eusebius, desiring to remedy this defect, constructed his work on a different principle. He made a table of ten canons, each containing a list of passages as follows: Canon I. passages common to all four Gospels; II. those common to Matthew, Mark, and Luke; III. those common to Matt., Luke, and John; IV. those common to Matt., Mark, and John; V. those common to Matthew and Luke; VI. those common to Matt. and Mark; VII. those common to Matt. and John; VIII. those common to Luke and Mark; IX. those common to Luke and John; X. those peculiar to each Gospel: first to Matthew, second to Mark, third to Luke, and fourth to John. Each Gospel was then divided into sections, which were numbered continuously. The length of the section was determined, not by the sense, but by the table of canons, each section comprising a passage common to four, to three, to two Gospels, or peculiar to itself, as the case might be. A single section therefore might comprise even less than a verse, or it might cover more than a chapter. The sections were numbered in black, and below each number was placed a second figure in red, indicating the canon to which the section belonged. Upon glancing at that canon