(παραγραφή οι παράγραφος, something written at the side), in the old Herculanean and Egyptian papyri, has already been mentioned. See, for a specimen, the beautiful papyrus of a Greek treatise on Rhetoric, written before 160 B. C., published in facsimile in "Papyrus grecs du Musée du Louvre" etc., edited after Letronne by Brunet de Presle (tom. xviii. 2e ptie of "Notices et extraits des manuscrits" etc., published by the French Institute, Paris, 1865). pl. xi., pap. No. 2. (Also in Silvestre, Paléogr. univ., pl. 55.) In the same volume, pl. xxxiv., pap. 49, in a letter of a certain Dionysius to Ptolemy, about 160 B. C., we have perhaps the earliest known example of the use of two dots like our colon for separating paragraphs, in conjunction with the marginal dash, precisely like the style which frequently occurs in both the Vatican and the Sinaitic MSS., though the Vatican more commonly omits the dots. Finally, in the curious Nativity or Thema genethliacum, dated in the first year of the Emperor Antonius (A. D. 138), of which a facsimile is given in pl. xxii., pap. 19, and also in Silvestre, pl. 58, we have numerous paragraphs indicated by the projection of the first letter, or the first two or three letters, into the lefthand margin; and for the most part, this initial is of considerably larger size than the rest of the letters. This, however, is not a book manuscript.

(5.) "Further," says Mr. Burgon, "Cod. & abounds in such contractions as ανοσ, ουνοσ (with all their cases), for ανθρωποσ, ουρανοσ &c. Not only πνα, πηρ, περ, πρα, μρα (for πνεύμα, πατηρ-τερ-τερα, μητερα), but also στη, τηλ, τηλημ, for σταυρώθη, ισραηλ, ιερουσαλημ.

"But Cod. B, though familiar with 16, and a few other of the most ordinary abbreviations, knows nothing of these compendia which certainly cannot have existed in the earliest copies of all. Once more it seems reasonable to suppose that their constant occurrence in Cod. indicates for that Codex a date subsequent to Cod. B."

א

Here Mr. Burgon, as usual, misstates the facts. The contraction for av ponoσ is found in the Vatican MS., p. 137, col. 1; 146, col. 2; 160, col. 1;-that for лvεvμа occurs twice on the first page of the New Testament (Matt. i.18,20), also Matt. iii. 11, 16, iv.1, and often elsewhere, particularly in the Old Testament (five times, for example, p. 331, col. 1, and again twice in col. 2);-рб for лаτɛроo occurs p. 69, col. 1; 190, col. 3 (marg. note); 226, col. 2;-161 for 16 panλ occurs hundreds of times: for instance, in Exod. xiv. it is contracted sixteen times out of seventeen in which it occurs, and in Josh. xi., eighteen times out of twenty. It will be hard to find "innu" as the contraction for 1ɛpovσaλnu in the Vatican MS. or in any other, but i t

occurs Josh. xii.10, and iλu, Josh. x.1,3, xv.5. Zravpwen is Σταυρώθη contracted but once in the Sinaitic MS., where we also have once (in Rev. xi.8) a unique contraction of εoτavρon, which Tischendorf has neglected to express in the text of his quarto edition, though he has spoken of it in the Prolegomena (p. xx.; compare the larger edition, vol. i., col. 8 of Prol.).

1

In this matter of contractions, much appears to have depended on the fancy of the scribe; and as a criterion of antiquity it must be used with caution. We find in the Vatican MSS. contractions for several words, as xαι, μov, aμd, δαυειδ, ισραηλ, ιερουσαλημ, which are never contracted in Cod. D (the Cambridge MS.), written two centuries later. In the papyrus MS. of Philodemus "De Deorum vivendi Ratione," published in vol. vi. of the Herculanensia Volumina, and consequently written as early as A. D. 79, we find a number of remarkable contractions not known to exist in any other Greek MS., or certainly in any of similar antiquity. In different parts of the Vatican MS. there is a marked diversity in this respect; for example, in the part of the MS. extending from 1 Kings xix. 11 to the end of Nehemiah, as compared with the preceding portion. The same is true of the Sinaitic MS., particularly in the six leaves of the New Testament which Tischendorf attributes to the scribe D, whom he now supposes to be identical with one of the scribes of the Vatican MS. For example, in fol. 15 of the Sinaitic MS., written by D, vioo (sing.) occurs five times, and is always written in full. In the contiguous leaves (14 and 16), written by A, it occurs nine times, and is always contracted. On fol. 15, aveрwnоo is written six times in full, once only contracted. In the contiguous leaves it occurs eleven times, and is always contracted. In fol. 10, written by D, ovpavoб occurs nine times, and is always written in full, as it seems to be in the Vatican MS. On the next leaf, written by A, it occurs ten times, and in six of them is contracted. (The statement in Tischendorf's Nov. Test. Vat., Prolegom. p. xxii., differs from the above in four particulars, in consequence, apparently, of oversights in counting.)

(6.) Mr. Burgon's sixth argument is founded on the following facts. The Gospel of Mark in the Vatican MS. as well as the Sinaitic ends with verse 8 of the sixteenth chapter. But in the Vatican MS., where the ending occurs near the bottom of the second column, the third column is left blank, and the Gospel of Luke begins on the next page. "This," says Mr. Burgon, "is the only vacant column in the whole manuscript" (p. 87)

* In the first 294 pages of the Vatican MS. (pp. 41-334 of the edition), vevμa occurs forty-two times, in forty of which it is contracted; in the next 290 pages it occurs forty-one times, in forty of which it is not contracted. There is a similar difference of usage in respect to the contraction of the word Iopanλ.

In the Sinaitic MS., in which there are four columns to a page, the Gospel of Mark ends on the second, and that of Luke begins on the third. The Vatican MS. has at the end of verse 8 the usual arabesque which is placed at the end of a book, and the subscription κατα Μαρκοr. But the phenomenon of the blank column is to Mr. Burgon "in the highest degree significant, and admits of only one interpretation. The older MS. from which Cod. B was copied must have infallibly contained the twelve verses in dispute. The copyist was instructed to leave them out and he obeyed: but he prudently left a blank space in memoriam rei" (p. 87). The Sinaitic, on the other hand, "was copied from a Codex which had been already mutilated" (p. 88). This difference between the MSS. seems to Mr. Burgon "a very striking indication that Cod. B is the older of the two. Cod. is evidently familiar with the phenomenon which astonishes Cod. B by its novelty and strangeness" (p. 292).

Eusebius, in the first quarter of the fourth century, expressly testifies that the last twelve verses of the Gospel of Mark were wanting "in the accurate copies," and "in almost all the copies," of that Gospel, but were found "in some copies." (Quæst. ad Marinum, c. 1. Opp. iv.937, in Migne's Patrol. Gr. tom. xxii.) Suppose, then, that the Vatican MS. was transcribed in the age of Eusebius from a copy which contained the passage, why may not the Sinaitic have been transcribed at the same time from one which did not contain it?

With Mr. Burgon, a conjecture seems to be a demonstration. There is to him but one possible explanation of that blank column. But considering the well-known tendency of copyists and possessors of manuscripts to add rather than to omit, a tendency which would be very strong in the present case in consequence of the abruptness of verse 8 as an ending, and of which the existence of two other endings, besides the disputed verses, is a proof, another conjecture may be proposed. Why may we not suppose that the exemplar from which the Vatican MS. was copied did not contain the last twelve verses, but the copyist, or owner of the MS., having at some time seen or heard of them, left on that account the blank column in question? We have a similar phenomenon in the case of Codd. L and A at John vii.52, and in Cod. G at Rom. xiv.23.

Mr. Burgon is not strictly correct in saying that the case to which he refers is "the only vacant column" in the Vatican MS. Two columns are left blank at the end of Nehemiah; but this may be accounted for by the different style (stichometric) in which the next following book, the Psalms, is written. (7.) Mr. Burgon's last argument is as follows. "The most striking feature of difference, after all, is only to be recognized by one who surveys the Codices themselves with attention.

It

is that general air of primitiveness in Cod. B which makes itself at once felt. The even symmetry of the unbroken columns ;the work of the prima manus everywhere vanishing through sheer antiquity;-the small, even, square writing, which partly reveals the style of the Herculanean rolls; partly, the papyrus fragments of the Oration against Demosthenes (published by Harris in 1848)-all these notes of superior antiquity infallibly set Cod. B before Cod. ; though it may be impossible to determine whether by 50, by 75, or by 100 years.'

On this we may remark (a) that "the even symmetry of the unbroken columns" has been shown to exist, so far as a large part of the MS. is concerned, only in Mr. Burgon's imagination; and that, where it does exist, it has a parallel in parts of the Sinaitic. (b) The work of the prima manus is rarely to be seen in the Vatican MS., a scribe of the tenth or eleventh century having retraced all the letters with fresh ink, adding accents and breathings, except in those places where he wished to indicate that something should be omitted (e. g. the accidental repetition of a word or sentence). In the passages where the work of the first hand remains untouched, of which we have facsimiles (e. g. John xiii.14, Rom. iv.4, 2 Cor. iii. 15,16), the original writing appears to have been well preserved. We may add that a scribe of the eighth or ninth century has retouched with fresh ink many pages of the Sinaitic MS.; and this had already been done to a considerable extent by a still earlier scribe (Tischendorf, N. T. ex Sin. Cod., p. xxxviii. f.). As to the appearance of the Sinaitic MS., we have the testimony of Dr. Tregelles that, "though the general semblance of the whole work is somewhat less worn than that of Cod. Vaticanus (whose extensive hiatus prove how carelessly it has been kept), when it comes to be contrasted with such a manuscript as the illustrated Dioscorides at Vienna (whose age is fixed by internal evidence at about A. D. 500), that interesting and valuable manuscript looks comparatively quite fresh and modern" (Scrivener's Coll. of Cod. Sin., p. xxxi.). (c) The writing in the Sinaitic is just as "even and square" as that of the Vatican. In the form of the letters Tischendorf expressly says that there is not the least difference, ne minimam quidem discrepantiam (Nov. Test. Vat., p. xix.). Mr. Burgon's argument, then, must rest wholly on the difference in size, the letters in the Vatican MS. being perhaps one-third smaller than those in the Sinaitic. (There is a difference in size in different parts of the two MSS. themselves, as is shown by the facsimiles, and by Tischendorf's express testimony.) It is difficult to deal seriously with such an argument. But if any explanation is needed, it may be suggested that the extraordinary size of the skins on which the Sinaitic MS. is written, allowing four columns to a page, of 48

lines each (the Vatican has three columns of 42 lines), would naturally lead a calligrapher to make letters somewhat larger than usual. And if Mr. Burgon will look again at a few of the Herculanensia Volumina, say the one last published (vol. v. of the second series), he will find that in some of the papyri there represented we have letters of the size of those in the Codex Sinaiticus, while in others they are less than half that size.

Such are "the notes of superior antiquity" which "infallibly " prove that the Vatican MS. is 50 or 100 years older than the Sinaitic.

A few words may be added in respect to Mr. Burgon's treatment of the principal subject of his work. The specimen which has been given illustrates some of his prominent characteristics as a writer; but judging from this alone, we might do him injustice. His book is not entirely worthless, or merely serviceable as showing how a critical question ought not to be treated; though it is often instructive in this respect. It is really to be welcomed as giving the results of earnest original research on the subject to which it relates. It brings to light some interesting facts, and corrects some errors of preceding scholars. It is written, however, with great warmth of feeling, in the spirit of a passionate advocate rather than that of a calm inquirer. The author appears to have been especially stimulated to the defense of the last twelve verses of the Gospel of Mark by his zeal for the damnatory part of the Athanasian Creed, which he not only regards as justified by Mark xvi. 16, but actually iden tifies with that verse. He says: "The precious warning clause,

...

(miscalled 'damnatory'), which an impertinent officiousness is for glossing with a rubric, and weakening with an apology, proceeded from Divine lips-at least if these concluding verses be genuine" (p. 3). This is only one of many examples which might be cited of the tendency of Mr. Burgon to confound the certainty of a fact with the certainty of a very dubious or even preposterous inference from it. For the new critical material which he has amassed every student will thank him, and also for the clear and satisfactory discussion of some special topics, as the so-called Ammonian sections; but there is much in his book which cannot fail to mislead an unwary or ill-informed reader. His conclusions are often strangely remote from his premises, but his confidence in them is boundless. He not only claims to have shown that the genuineness of the disputed passage "must needs be reckoned among the things that are absolutely certain," but appears to expect that in consequence of his labors "it will become necessary for Editors of the Text of the New Testament to reconsider their