BSe385 
 .H22 
 
L I B R A. R Y 
 
 Oh- Tin; 
 
 Theological Seminary 
 
 PRINCETON, N. J. 
 
 Case IBS23S>S D'V'S'on 
 
 Sliclf .H2L?- Section -- 
 
 ^^''>^' No, „..' . 
 
 A DONATION 
 
61arenir0ii Iress , Bmts 
 
 TEXTUAL CRITICISM 
 
 APPLIED TO THE 
 
 NEW TESTAMENT. 
 
 HAMMOND. 
 
Eontion 
 
 MACMILLAN AND CO. 
 
 PUBLISHERS TO THE UNIVERSITY OF 
 
Clartirbflii ||«ss Scrits 
 
 OUTLINES 
 
 OF 
 
 TEXTUAL CRITICISM 
 
 APPLIED TO THE 
 
 NEW TESTAMENT 
 
 C. E. HAMMOND, M.A. 
 
 FELLOW AND TUTOR OF EXETER COLLEGE, OXFORD 
 
 AT THE CLARENDON PRESS 
 M DCCC LXXII 
 
 \ All riehls reierved^ 
 
PREFACE 
 
 The following pages have no claim to originality. The 
 substance of them was collected for a course of College 
 Lectures; and they profess to be no more than a compi- 
 lation from other larger works. The justification of the 
 writer for publishing them, if there be any, lies in the fact 
 that there is not, so far as he is aware, any single book 
 which serves well as a first introduction to the science of 
 the Textual Criticism of the New Testament in its present 
 advanced state. Dr. Tregelles' History of the Printed Text 
 of the New Testament, and Mr. Scrivener's indispensable 
 Plain Introduction to the Criticism of the New Testament, 
 were published, the one in 1854, the other in 186 1. A new 
 edition of Home and Tregelles' Introduction to the Study 
 of the Bible was published in 1863; the fourth volume of 
 which, on the New Testament, contains a few pages of 
 addenda, with notices of collations and critical publications 
 down to that time; but in other respects it is merely a 
 reprint of the earlier edition of 1856. Since then, however, 
 a good deal has been done, with which the student should 
 be acquainted. 
 
VI PREFACE. 
 
 These three books, the articles ' New Testament/ ' Ver- 
 sions (Ancient)/ and ' Vulgate/ in Dr. Smith's Dictionary 
 of the Bible, the Prolegomena to Tischendorf 's Greek Testa- 
 ment (seventh edition), and to his editions of the Sinaitic 
 and Vatican Manuscripts, the Prolegomena to Lachmann's 
 Greek Testament, and to Kuenen and Cobet's edition of 
 the Vatican MS., as well as to Dean Alford's last edition of 
 his Greek Testament (vol. i.), and Scrivener's Collation of 
 the Sinaitic MS., are the chief sources from which informa- 
 tion has been taken. To such works as these the student 
 must have recourse, if he is led on to wish to fill up much 
 that he will here find sketched in merest outline. To collect 
 into a small compass the leading facts on which the science 
 of Textual Criticism is founded, and to present to the be- 
 ginner the principles of the science, divested of the repelling 
 mass of detail which necessarily meets him in the larger 
 works, has been the writer's aim ; in the hope that he may 
 give some little assistance to those who are entering on a 
 subject that is interesting in itself, and some knowledge of 
 which seems indispensable to an intelligent study of the 
 original Text of the New Testament. 
 
 The writer has to thank several friends for their kind help 
 and suggestions, especially the Rev. C. W. Boase of Exeter 
 College, and the Rev. G. W. Kitchin of Christ Church. 
 
CONTENTS. 
 
 Introduction ..,...••• 
 Chap. I. Sketch of the History of the Textus Receptus 
 Chap. II. On the Origin of Various Readings 
 
 Note from ]^Iad^^g's Adversaria Critica . 
 Chap. III. On the Manuscripts of the Greek Text. 
 
 § I. The number, mode of designation, &c. of MSS 
 
 § 2. On some palseographic details 
 
 § 3. On the various systems of the Divisions of the 
 ^^^ Text 
 
 § 4. An account of Codices x and B 
 Chap. IV. On Versions, and the chief Versions of the New 
 Testament. 
 
 I. 
 
 Chap. V. 
 
 On the nature and value of the e\'idence given 
 
 by Versions . 
 The Latin Versions 
 The Syriac Versions 
 The Egyptian Versions 
 The Gothic Version 
 The ^thiopic Version 
 The Armenian Version 
 On Patristic Quotations 
 
 Note on Ronsch's Das neue Testament Tertullian's 
 Chap. VI. Discussion of the Evidence derived from the foregoing 
 Sources. 
 § I. Summary of results reached so far . 
 
 MSS., though independent of each other, are 
 
 marked off by general features into groups . 
 
 Two main groups commonly recognised by 
 
 Critics 
 
 Examples of the proofs that the smaller group 
 of witnesses contains the earlier type of Text 
 An order traceable among the various docu- 
 mentary witnesses 
 
 Chap. VII. Historical Corroboration 
 
 §2. 
 
 §3- 
 
 §4- 
 
 §5. 
 
 page 
 I 
 
 44 
 46 
 
 50 
 64 
 55 
 56 
 56 
 58 
 64 
 
 65 
 
 67 
 68 
 
 70 
 
 79 
 8i 
 
Vlll 
 
 CONTENTS. 
 
 PAGE 
 
 Appendix A. On Canons of Criticism 87 
 
 Appendix B. Critical discussion of some additional disputed 
 Passages. 
 
 1. I S. John V. 7, 8 93 
 
 2. S. Mark xvi. 9-20 95 
 
 3. S. John vii. 53 — viii. 11 . . . -97 
 
 4. I Tim. iii. 16 . . . . . .99 
 
 5. S. John V, 3, 4 102 
 
 6. S. Luke xxii. 43, 44 103 
 
 7. S. Matt. xxi. 28-31 104 
 
 8. Acts XX. 28 106 
 
 9. Acts xi. 20 ...... 109 
 
 Note on S. Mark xvi. 9-20 no 
 
 C. List of Greek Uncial Codices . . . .112 
 
 D. List of the chief Latin Codices . . . .122 
 
 E. List of Patristic Writers, with Dates, &c. . .126 
 
 Appendix 
 Appendix 
 Appendix 
 Index I. 
 
 Passages of the New Testament referred to in the 
 Work 133 
 
 Index II. General 135 
 
INTRODUCTION. 
 
 Co^iPARATivE Criticism as applied to the New Testament 
 may be defined as the science zvhich determines the mutual 
 relatiojis and values of the various authorities from which the 
 original text of the Neiv Testamerit is to be ascertained. Its 
 office is to indicate the limits within which the truth is to be 
 found ; to select the witnesses most likely to speak the truth ; 
 and then, by cross-examining them and comparing their 
 testimony, to determine what is most probably the true text. 
 
 It is evidently assumed, when such a description as this 
 is given, that no value is assigned to the commonly received 
 text of the Greek Testament as such. Any claim which it 
 may be supposed to have upon our acceptance must be 
 summarily set aside while the case is being tried on its own 
 merits. That the Textus Receptus was derived from MSS. 
 transcribed at a very late date, and that there is a marked 
 difference between the text presented by such MSS. and a 
 text founded upon early authorities, are undisputed facts. To 
 account for this difference, to determine the relation of the 
 later documents to the earlier, and to decide which class of 
 documents probably most nearly represents the actual words 
 of the writers of the New Testament, are among the chief 
 problems which Comparative Criticism has to solve. 
 
 Answers directly opposed to each other have been given 
 to .these problems. We may fairly take Dr. Tregelles and 
 
% INTRODUCTION. 
 
 Mr. Scrivener, the two chief Enghsh writers on this subjectj 
 as the exponents of the opposite schools. Dr. Tregelles 
 says (Account of the Printed Text of the Greek New Testa- 
 ment, p. 138), 'The mass of recent docmnents possess no 
 determining voice in a question as to what we should receive 
 as genuine readings. We are able to take tht few docu- 
 ments whose evidence is proved to be trustworthy, and safely 
 discard from present consideration the eighty-nine ninetieths, 
 
 or whatever else their numerical proportion may be 
 
 I should feel that I did indeed put the text of the New Testa- 
 ment in peril, if I adopted the authority of the mass of MSS., 
 which is proved to be at variance with what was read by the 
 Christians of the third century at least.' The italics are Dr. 
 Tregelles' own. On the other hand, Mr. Scrivener says 
 (Introduction to the Criticism of the New Testament, p. 386), 
 ' Irenaeus and the African Fathers, and the whole Western, 
 with a portion of the Syrian, Church used far inferior manu- 
 scripts to those employed by Stunica, or Erasmus, or Stephens 
 thirteen centuries later, when moulding the Textus Receptus.' 
 And again (p. 409), ' In the far more numerous cases where 
 the most ancient documents are at variance with each other, 
 the later or cursive copies are of much importance, as the 
 surviving representatives of other codices, very probably as 
 early, perhaps even earlier, than any now extant.' This 
 hypothesis then assumes the existence of certain correct 
 codices, different from any of the early codices now extant, 
 all of which have perished, and only survive in their ' re- 
 presentatives ' ; which moreover remained in such obscu- 
 rity as to be unknown to the African Fathers, the whole 
 of the Western, and a portion''^ of the Syrian Church. 
 
 * To say ' the early Syrian Church ' would probably be more correct ; 
 but this depends upon the mutual relation of the Peshito and Curetonian 
 Syriac Versions, regarding which different estimates are formed by the 
 two schools of critics. (See below, pp. 50, 51, and reff. there.) 
 
INTRODUCTION. 3 
 
 This is however undoubtedly the view popularly held in 
 England. 
 
 The following pages are an attempt to explain the prin- 
 ciples of the other school, with the grounds on which they 
 rest. The earliest adumbration of these principles b was 
 given by Bentley in his letter to Archbishop Wake (1716), 
 and his ''Proposals" (1720); but sufficient materials were 
 not yet collected for him to bring his design to a satisfactory 
 issue. To Lachmann is due the honour of having led the way 
 to tangible results. His larger edition (vol. i. published in 
 1842, vol. ri. in 1850), in which he was assisted by Buttmann, 
 was a gigantic stride in the science of Textual Criticism 
 placing it at once on a basis of scientific accuracy. Since his 
 death (185 1) large additions have been made to our resources, 
 both by the discovery of many new ]\ISS., and by the colla- 
 tion and publication of many that were but imperfectly known 
 before. Had he been able to make use of this fresh evidence, 
 the application of his own principles must have led him to 
 modify some of his conclusions; and thus probably there 
 w^ould have been even fewer discrepancies than there are 
 between the results independently attained by him, by Tischen- 
 dorf, and by Tregelles. As it is, the constant agreement t- 
 of these three critics is a strong argument for the correctness 
 of the principles which have guided them. 
 
 ^ As it has been said very recently that these principles ' lead to con- 
 clusions little short of irrational,' it may be well to inform the reader 
 that they have at any rate satisfied such critics as Professors Westcott 
 and Lightfoot, Mr. F.J. A. Hort, and the late Dean Alford — to mention 
 only English critics. 
 
 •^ How close this agreement is, may be clearlyseen in the very con- 
 venient little edition of the Greek Testament edited by Mr. Scrivener, 
 in which he distinguishes by a difference of type all deviations from the 
 text of Stephens (1550) on the part of the editions of Beza (1565), of 
 Elzevir (1624), of Lachmann (1S42-1850), of Tischendorf (1859) ; and 
 of Tregelles, as far as that work had been published at the time of the last 
 correction: The respective readings are noted at the foot of each page. 
 
 B 2 
 
4 INTRODUCTION. 
 
 There is another question to which the two schools of 
 critics would give different answers ; namely, Wkaf weight is 
 to be assigned to subjective arguments in deciding between various 
 readings ? Those critics who profess to take the evidence 
 of actually existing early documents as the basis of their 
 conclusions are only consistent in assigning a very subordi- 
 nate place to subjective arguments. But even among them 
 peculiarities of mental constitution and training will naturally 
 dispose some individuals to attach more weight to this class 
 of arguments than others do ; and this is to a large extent 
 the cause of the differences which are found between the 
 texts of critics of the same school. 
 
 The term ' subjective ' is here taken to include what Bishop 
 Ellicott (Ep. to Gal., Preface, p. xviii, ed. 1859) distinguishes 
 into paradiplomatic and internal evidence, meaning thereby 
 respectively, ' The apparent probabilities of erroneous" tran- 
 scription, permutation of letters, itacism, and so forth ; ' and 
 
 * apparent deviations from the usus scribendi of the sacred 
 author, or the propensio^ be it critica, dogmatica, or epexegetica, 
 on the part of the copyist.' The reason why a copyist having 
 one form of words before him wrote another is after all only 
 a question of greater or less probability. Such arguments 
 cannot, and ought not to be ignored by the critic ; but it is 
 easy to magnify their weight unduly. To depend upon such 
 considerations mast be almost always precarious. For every 
 such argument on one side, it is commonly possible to bring 
 forward a corresponding one on the other d. Dean Alford's 
 expression of opinion on this subject is worth quoting ; the 
 more so that it shows a very great modification of his original 
 views. He says (G. T., vol. i. Proleg. p. 87, edd. 1863, 1868) 
 
 * Experience has brought about some change in my con- 
 victions with regard to the application of canons of subjective 
 
 '' See for instance some of the examples at the end of the book, 
 
 pp. 56, loS, &c. 
 
INTRODUCTION. 5 
 
 criticism to the consensus of ancient INISS. In proportion as 
 I have been led severely to examine how far we can safely 
 depend on such subjective considerations, I confess that the 
 limits of their applicability have become narrowed. In very 
 many cases, they may be made to tell with equal force either 
 way. One critic adopts a reading because it is in accord 
 with the usage of the sacred writer ; another holds it, for this 
 very reason, to have been a subsequent conformation of the 
 text. One believes a particle to have been inserted to give 
 completeness; another to have been omitted as appearing 
 superfluous.' 
 
 The differences exhibited by texts, as edited by critics of the 
 diplomatic school, depend almost entirely upon their views of 
 the limits of applicability of the canons referred to in this 
 statement of Dean Alford. The case may be put as follows : 
 If we decline to consider any but the diplomatic evidence, 
 we arrive at a text which was certainly current in the middle 
 of the second century, or thereabouts; that is to say, not 
 much more than fifty years after the death of the last of the 
 Apostles. So far \ve are upon ground w^hich is safe, which 
 can be easily surveyed, and which may serve as ^'-clear start- 
 ing-point and basis of operations. This basis being clearly 
 established, we may proceed further to apply various subjec- 
 tive considerations in any cases that remain still doubtful, 
 with the aim of restoring what we believe the authors of the 
 books must have actually wTitten. In this endeavour there 
 is scope for much valuable research ; at the same time, the 
 particular propensions of the critic cannot but show them- 
 selves to some extent ; and thus, the moment we leave be- 
 hind the diplomatic evidence, an element of uncertainty is 
 inevitably ffitroduced. Still, if this be borne in mind, and 
 the results stated temperately, and kept clearly distinguished 
 from those which the earliest documentary evidence seems 
 to support, much good may be done, and much help 
 
6 INTRODUCTION. 
 
 may be given towards the ultimate solution of the main 
 problem. 
 
 The principles and method of the science, as applied to 
 the text of the New Testament, are for the most part the 
 same as those required in deahng with the texts of the ancient 
 classical authors ; only the material is far more abundant 
 and various than in the fields of secular criticism. 
 
 There are three sources of evidence; viz. — 
 
 1. A large number of Manuscripts of the Greek Text, 
 some containing the whole, some containing parts only, of 
 the books which we now call collectively ' The New Testa- 
 ment ' ; written at various times from the fourth to the four- 
 teenth centuries inclusive, and in all possible states of preser- 
 vation. 
 
 2. Versions, or translations of the Books of the New 
 Testament into other languages than Greek. Those only 
 are of value for critical purposes which were made between 
 the second and seventh centuries. This class of evidence is 
 particularly valuable, as will hereafter be seen, in questions 
 concerning the early existence and prevalence of certain 
 various readings. 
 
 3. Quotations in the writings of the ecclesiastical writers 
 of the first five centuries ; which, used cautiously and under 
 conditions that will be explained afterwards (see pp. 58-64), 
 may be made to yield evidence of essential value. 
 
 Conjectural Emendation, which has been sometimes of 
 necessity exercised on the texts of secular writers, has 
 absolutely no place in the criticism of the text of the New 
 Testament. It is needless, nor does any critic seek to apply 
 it. Amidst the abundance of resources the difficulty is rather 
 to select than to invent. Whereas some of the classical texts 
 rest upon a single late MS., we have between fifteen and 
 sixteen hundred, including several of very early date, to 
 
INTRODUCTION. 7 
 
 make use of in discussing the sacred text. Translations 
 into Latin are among the most trustworthy sources of in-, 
 formation as to the text of some parts of Plato and Aristotle; 
 but we have no fewer than ten versions of the New Testa- 
 ment^, each possessing a distinct critical value. Lastly, the 
 Quotations are manifold in the case of almost all the im- 
 portant passages. Thus we have a threefold cord of evidence, 
 each strand of which is itself composed of many threads. 
 
 [It is well to bear in mind that where the phrase ' docu- 
 mentary evidence ' is used, it must be understood to include 
 ail or any of the three sources of evidence above men- 
 tioned that may bear upon the point under discussion ; and 
 must not be limited, as is too often tacitly done, to jNISS. 
 alone.] 
 
 e A sort of parallel to this is found in the sacred literature of the 
 Buddhists. The Sanscrit originals of their sacred books have been 
 translated into Thibetan, Mongolian, Chinese, and Mantschu ; and the 
 Pali (Ceylon) originals into the languages of Burmah and Siam. (Max 
 Mailer's ' Chips from a German Workshop,' vol. i. pp. 193, 195.) 
 
CHAPTER I. 
 
 SKETCH OF THE HISTORY OF THE ' TEXTUS RECEPTUS/ 
 
 The New Testament in Greek was not printed till the 
 beginning of the sixteenth century. Up to that time it was 
 circulated in manuscripts only. A few detached portions 
 had been printed earlier ; but the first complete edition was 
 that prepared at Alcala in Spain by Cardinal Ximenes, 
 forming the fifth volume of the magnificent Triglott edition 
 of the whole Bible published by him, and called, from the 
 Latin name of the place, The Complute7isiaii. The fifth 
 volume was printed in 1514, and the whole work was com- 
 pleted in 15 1 7, a few months only before Cardinal Ximenes' 
 death. Some delay occurred after this, and it was not 
 published till 1522. Only six hundred copies were printed. 
 At that period Httle was understood of Greek criticism, or 
 of the relative value of manuscripts. The Latin version was 
 thought to be the truthful standard, and held the place of 
 honour on the pages of this edition between the Hebrew and 
 the Greek. The particular manuscripts from which the 
 Complutensian text is formed have not been identified with 
 certainty, but it is clear from the character of the text that 
 none were used which do not belong to that type which we 
 shall see reason to consider of late origin. 
 
 During the preparation of this work, a printer at Basel, 
 named Froben, hearing of the Cardinal's design, and wishing 
 
THE TEXTUS RECEPTUS. 9 
 
 to anticipate it, prevailed on the well-known scholar, Erasmus, 
 to prepare an edition for the press. This was done in great 
 haste; and Erasmus' fij-st edition was published in 1516, 
 being thus the first published, though not the first printed, 
 Greek text. He had four manuscripts to work from, all of 
 which are identified : one of these is of great value, having a 
 text approaching that of B and L ; but its variations from the 
 others caused him to be suspicious of it, and he based his 
 text almost wholly on the other three, which are all of the late 
 type". In the Apocalypse he boldly retranslated i. 15-20 
 from the Latin, his manuscript being defective ; and he 
 interpolated several words elsewhere, which exist in no 
 known Greek IMS (see Scrlv. Introd. p. 296). These inter- 
 polations, as well as Acts viii. 37, for which the only ancient 
 testimony is the Latin version, have continued in the ordinary 
 Greek text to the present day, and thence hold their place 
 in our English translation. Erasmus however did not 
 insert the verse i S. John v. 7, till his third edition. His 
 second edition (1519) is of no special importance: it differs 
 from the first in having many misprints corrected, which 
 had crept in through the haste with which the work was 
 brought out. The third edition (1522) is to be remarked 
 as having for the first time a few various readings noted 
 in the margin. More important however to us is the 
 fourth edition (1527), which Erasmus corrected by the 
 Complutensian, and which became the basis of the Textus 
 Receptus. 
 
 =» The statement that the few manuscripts used for the text are of a 
 late t>-pe must be taken for granted by the reader at this stage of the 
 enquiry. They are cursives, and present the characteristics of the 
 Byzantine class (see pp. 67, 68). The only three MSS. of a different 
 type known to any of these editors, viz. D, D,, i, were looked on 
 with suspicion and little used. The relative age and value of the 
 various classes of MSS. is discussed below in Chap. vi. 
 
lO SKETCH OF THE HISTORY 
 
 The next most important edition is the third of Stephens, 
 known as the Editio Regia, pubHshed at Paris in 1550. 
 The text varies very little from that of Erasmus' fourth 
 edition. Its special value depends on the considerable and 
 systematic collection of various readings from fifteen fresh 
 manuscripts, including the valuable and ancient Codex 
 Bezse (D), which Stephens added in the margin. The 
 influence of prescription already shows itself in the fact ihat 
 Stephens often follows the text of Erasmus, in defiance of 
 the authority of his manuscripts. 
 
 Beza (15 1 9- 1 605) published various editions between 
 1556 and 1598. He added a few more various readings 
 from other manuscripts; but he still followed Stephens' 
 text closely. 
 
 Later still the Elzevirs brought out their beautifully 
 executed editions at Ley den, between 1624 and 1633. 
 The text is again little more than a reproduction of 
 Stephens' b; in fact it is asserted by them in the preface 
 to their second edition to be ab omnibus i-eceptus : and from 
 this phrase comes the designation ' Textus Receptus.' 
 
 From this sketch it will have been seen that our Textus 
 Receptus is based upon a very few manuscripts. It is true 
 that a number of various readings had been collected ; but 
 they were only placed in the margin, and were not used in 
 reconstructing the text, except occasionally, and then on no 
 fixed principles. The value of various readings was not yet 
 appreciated. 
 
 We must further bear in mind that the necessity for 
 scrupulous accuracy in the work of collation was not yet 
 understood; that the text of the Vulgate w^as corrupt; 
 
 ^ Stephens' and Elzevir's texts differ in 286 places according to 
 Scrivener (Proleg. to G. T. Cantab. 1862, p. vi). Our English 
 version appears to follow sometimes one and sometimes the other. 
 See Smith's Diet. Bib. vol. ii. p. 524. 
 
OF THE TEXTUS RECEPTUS. II 
 
 that no help was sought from Oriental Versions ; nor any 
 attention paid to Patristic Quotations. 
 
 Enough has been said to show that no critical value is to 
 be assigned to the Textus Receptus. In saying this we by 
 no means imply that blame is due to Erasmus, Stephens,, or 
 Beza, for not being on a level with the critics of the present 
 day. Principles of textual criticism could not be worked 
 out until materials had been collected : and the collection of 
 materials was the work of time and research. These men were 
 the pioneers of the advance, and did indispensable service. 
 But we must learn not to elevate the text formed from their 
 materials into an authority. The facts which we are about 
 to discuss will show us that while we are warranted in re- 
 fusing any authority to the Textus Receptus, we are led with, 
 reasonable certainty towards a new text, somewhat different 
 from the old one, and with some few points still undeter- 
 mined, but resting on the basis of an infinitely multiplied 
 stock of materials, and supported by a well-understood and 
 searching system of criticism. 
 
CHAPTER II. 
 
 ON THE ORIGIN OF VARIOUS READINGS. 
 
 It is important that we should try to reaHze the amount 
 of depreciation to which a text is Hable under the hands of 
 successive copyists. From the very nature of the case it is 
 probable that errors should creep in. We know how liable 
 printed books are to suffer from typographical errors : they 
 have however this advantage, that by due care most of the 
 errors will be corrected before the book is published ; and 
 then all the copies issued will have the same degree of 
 correctness. In the case of a manuscript, not only is the 
 difficulty of correcting the errors greater, but after all the 
 correctness of only one copy is secured. Further, when 
 this copy comes to be in its turn an exemplar to be copied, 
 its own particular errors will be reproduced ; dnd the copyist 
 will certainly be found to have made fresh errors ; and thus 
 at each stage the text will tend to recede more and more 
 from the original. The natural conclusion from this is, that 
 the text of a manuscript written in the fifteenth century would 
 probably differ from the autograph text of the Aposdes, 
 more widely than a manuscript of the fourth century. Of 
 course there is always the possibility that a recent codex m.ay 
 be a direct copy from one of great antiquity ; and thus be a 
 more trustworthy representative of the original, than one 
 made some centuries earlier than itself. Such a claim must 
 be proved for every alleged case. 
 
ON THE ORIGIN OF VARIOUS READINGS. 1 3 
 
 The case however rests on stronger grounds than mere 
 presumption. Of all the known MSS. of the Greek text, 
 amounting to just under sixteen hundred, only two pairs so 
 resemble one another as to render it probable that they are 
 in any way mutually connected one with the other. The 
 comparison of any two then would give rise to a number 
 of various readings ; and the number would of course be 
 increased as more ]MSS. were compared. The possible 
 sources of these variations are not very numerous, and can 
 be easily understood by considering the mode in which MSS. 
 were transcribed, and the chances to which they were liable 
 during the centuries which have since elapsed. 
 
 The majority of the later MSS. were doubtless executed in 
 the monasteries, of which the Scriptorium was a regular de- 
 partment. But in earlier times they must have been the 
 production of the regular professional copyists, who would 
 regard their task as a mere piece of business, and would 
 bring to it no particular religious feeling nor extraordinary 
 pains. Sometimes books were multiplied by dictation a, one 
 person reading aloud the copy, while a number of scribes 
 wrote simultaneously. i\Iore commonly, however, where 
 care was needed, the scribe would have the exemplar before 
 him and copy it singly ; or several scribes might undertake 
 different parts of the work. The copy thus taken was sub- 
 jected to a careful revision, being recompared with the exem- 
 plar, and sometimes a second time with some standard copy. 
 The technical words for these processes are respectively 
 avTi^aXkiiv and diop6ovv. The corrector was sometimes the 
 scribe himself, sometimes a different person. Such a 
 
 a This is the common statement. Professor Madvig however denies 
 that this was ever the case. He is speaking of the MSS. of secular 
 writers ; and considers that all errores scribetidi may be shown to depend 
 upon failure of attention or of memory, the hand of the scribe not 
 following his eye. (Adversaria Critica, vol. i. p. 10.) 
 
14 ON THE ORIGIN OF 
 
 comparison with a copy in repute would add value to the 
 codex, and would be noted accordingly; e.g. in the Codex 
 Friderico-Augustanus the following words occur in the 
 subscription to the Book of Esther: fxereXrjfKpdr) Km Biopdwdrj 
 
 Trpocr ra e^airXa ^ Qpiyevovcr vtt ovtov BLopOccfieva. ^KvTcovivocr ojxo- 
 XoyrjTrjcr dvTejSaXev, 'n.ap(f)t\o(T diopOcocra (sic) to revxacr iv tij 
 (f)vXaKrj, 
 
 In how merely professional a spirit this revision was some- 
 times executed is well exempHfied by some of the correc- 
 tions found in the Codex Vaticanus (B). One of the com- 
 monest errors in manuscripts is a confusion of ei with i.. 
 Now in different parts of the Cod. Vat. the same word is 
 found spelt sometimes with ft, sometimes with c; e. g. in 
 
 S. Luke XXiii. lO, S. John vii. 37, &C., eLcrTrjKeLO-av, ei(rTr,K€i 
 
 are rightly written by the original scribe; in S. Matt. xii. 46, 
 xiii. 2, &c. they stand la-TrjKeiaav, laTrjKeL. In the latter places 
 the corrector has substituted the et in the first syllable ; in the 
 former, he has wrongly substituted t for a. Manifestly this is 
 the work of no intelligent critic. The corrector must have 
 had a codex before him, in which the vowel and diphthong 
 were confused ; and with mechanical accuracy he transferred 
 the confusion to the pages which he was correcting. 
 
 Sometimes it appears as if a codex h'ad passed into the hands 
 of some learned person, who had an opportunity of recom- 
 paring it with another exemplar, and thus a further series 
 of corrections was introduced ; a process which might take 
 place more than once. When this has been the case, it is 
 easy to see what an amount of tact, patience, and judgment, 
 may be required to decipher, weigh, and arrange, all the 
 evidence that the manifold corrections may be made to give. 
 The Cod. Sinaiticus (n) has corrections by no fewer than 
 twelve hands, of dates ranging from the fourth to the twelfth 
 century. The Cod. Vaticanus (B) is corrected throughout 
 from another IMS. (See the proofs of this in Kuenen and 
 
VARIOUS READINGS. 
 
 15 
 
 Cobet's edition, Prcf. pp. xxiii-xxxviii.) As the corrections 
 in this last case are of the same age as the original writing, 
 though not by the original scribe, it is clear that within the 
 compass of one codex we have the evidence of /wo manu- 
 scripts, each perhaps much older than the codex itself, which 
 dates from the middle of the fourth centur)\ 
 
 Another fruitful source of various readings is that the 
 possessor of a MS. would write in the margin some explana- 
 tory note, w^hich a subsequent scribe, with the MS. before 
 him for a copy, looked upon as having been an accidental 
 omission, and incorporated in his new text. Instances of 
 this will be found below. 
 
 On the whole, the possible sources of various readings 
 may be classified as follows. The last head has been added 
 because certain alterations have been sometimes attributed to 
 that cause : there appears however to be no good ground for 
 the suggestion. The alterations which have been set down 
 to this source may all be as justly attributed to other sources 
 as to this. 
 
 f I . Errors of sigbf. 
 ■{ 2. Errors of hearing. 
 
 t 
 
 Possible 
 sources 
 
 of 
 various 
 
 readings. 
 
 r Unconscious, 
 or 
 unintentional. 
 
 I 3. Errors of memory. i 
 
 ' 4. Incorporation of marginal glosses, &c. 
 
 5. Corrections of harsh or unusual forms of 
 words, or expressions. 
 
 Conscious, 6. Alterations in the text to produce sup- 
 or 1 posed harmony with another passage, 
 
 intentional. to complete a quotation, or to clear up 
 
 a supposed difficulty. 
 
 7. Liturgical insertions. 
 
 [ 8. Alterations for dogmatic reasons. 
 
 A few typical instances of the various readings arising 
 from each of these sources shall now be £riven in order. 
 
1 6 ON THE ORIGIN OF 
 
 I . To this head will belong omissions arising from what is 
 called Homoioteleuton. If two consecutive lines in the 
 exemplar before the copyist ended with the same word, 
 or even sometimes with the same syllable, his eye 
 caught the second line instead of the first, and he 
 omitted the intermediate words. Occasionally this 
 happens at longer distances than single lines. This is 
 perhaps the reason of the omission in many codices of 
 
 the words 6 6ixo\oy(iiV top vlov Kat Tov 7raT€pa €)(^ei, T S. John 
 
 ii. 23, which are wanting in the Textus Receptus, but 
 which belong to the true text ; and of the words tovto 
 
 de ecTTiv to Oekrjfxa tov irejX^^avToa- fie, S. John vi. 39, in 
 
 Cod. C. In both these cases the clause preceding 
 the omission ends with the same words as the clause 
 omitted. The notes of any critical edition of the New 
 Testament will supply numerous other instances. 
 
 Under this head may also be classified the variations 
 arising from the confusion of similar letters, as €, C (2), 
 O, e ; or A, A, A. This and the following kinds of 
 error chiefly occur in uncial manuscripts ; in which the 
 words are written continuously, without any break or 
 space between them. This is the origin of the well- 
 known difiiculty in i Tim. iii. 16 between the readings 
 oc (os) and ec {deos). 
 
 Similar letters or syllables are sometimes omitted and 
 sometimes inserted; e. g. for the true reading nPOC- 
 eAei2N in S. Matt. xxvi. 39, we have nPOeAeON in 
 Codd. B, M; and for eKBAAAONTAAAlMONiA in S. Luke 
 ix. 49 we find eKBAAAONTATAAAIMONIA in Cod. H. 
 
 Letters sometimes become transposed; e.g. Acts xiii. 23, 
 for CPAIN (o-coT^pa 'lrj(rodv) we find in Codd. H, L, cPlAN 
 (a-coTrjpiav). The thin horizontal lines above the words, 
 which mark a contraction, are easily misplaced or over- 
 looked, and in process of time would fade. 
 
VARIOUS READINGS. 1 7 
 
 Perhaps to errors of hearing may be assigned the frequent 
 itacisms, or confusion of letters having similar sounds, 
 which are found in manuscripts of every age : or they 
 may arise from degenerate pronunciation. One of the 
 commonest confusions is that of the letters I and 61, 
 which are interchanged continually, even in words 
 where the I is short: e.g. i Thess. i. 3 in Cod. B, 
 afitaXeiTTTcos Stands written AAGIAAIIITQC prima manu. 
 In many cases, as in this last, the variation makes no 
 difference in the sense, and can be at once corrected ; 
 but it is easy to see that such confusion might materi- 
 ally affect the sense. 
 
 The following are some of the commonest itacisms ; and 
 the instances of each are such as would involve a 
 greater or less difference in the sense. 
 
 Confusion of AI and € is very common : e. g. 
 
 VTJ-orao-o-ere for -rat, S. Luke X. 20 (Cod. B*). 
 
 eraipois for eVfpois, S. Matt. xi. 1 6 (several MSS.). 
 A — e aKovaare for -aere, S. Matt. xiii. 14 (Cod. B*). 
 7r\T]p(ocr€T€ for -aare, S. Matt, xxiii. 32 (Cod. B*). 
 
 I — H \r]vov for Xcvov, S. JMatt. xii. 20 (Cod. B^). 
 
 KafiiXov for Kafir)\ov, S. Luke xviii. 25 (Cod. S). 
 Xpio-Tos for xpw^oS) I S. Pet. ii. 3 (Codd. K, L). 
 
 I — €1 (TTpareia for arparia, ActS vii. 42 (Codd. A, B, D). 
 elarat for larai (for 'larat perf., not larat pres.), 
 
 S. Mark V. 29 (Cod. B*). 
 O — Q late and. comparatively rare : 
 
 TroiTja-ofxev foi TToirjacofxev, S. Luke iii. 1 4 (several 
 
 codices). 
 l3a6€os for IBadeoos, S. Luke xxiv. i (Cod. E, &c.\ 
 fiera hiwyp-ov for -fxu)v, S. Mark x. 30 (several 
 
 cursives). 
 6 eiVcoi/ for 6v €LTiov, S. John i. 1 5 (Cod. N^'^, B"', C*). 
 c 
 
l8 ON THE ORIGIN OF 
 
 An instance of an error of sound, slightly different in 
 kind from the foregoing, is perhaps Kmirep ia-nv for kqi 
 napea-Tai, Apoc. xvii. 8, which some of the cursives give, 
 and which has passed into the Textus Receptus. 
 
 Sometimes we find the terminations of consecutive words 
 
 assimilated, e. g. tov ayyeXov avrov Tov 8ov\ov avTov for 
 Tov ayyekov avrov tco dovXco avrov (Cod. A), ApoC. i. I ; 
 or Xeyovrcav 'lovoaLcov for Xeyovrcov 'lovdaiova (Cod. C), ApOC. 
 
 ii. 9. 
 There is one sort of error which might be placed under 
 either of these classes ; arising from a confusion be- 
 tween words spelt with a single or double consonant: 
 
 e. g. ovx on Tvepi rcov nrcoxcov e/xeXXev avrco, S. John xii. 6 
 (Cod. B), for e'fxeXev. So between yeyevvrj/jLai, iyevvqOrjcrav, 
 and yeyevrjfxai, eyevrjOrjcrav, S. John i. 1 3, &C. ; and iyevrjOrjfifv 
 
 vrjmoi for €y. rjinoi (Codd. N, B*) in I Thess. ii. 7. 
 
 3. To error of memory may probably be attributed the not 
 unfrequent substitutions of synonymous words, such 
 
 as e<^7; for elircv] iiifirjrai for ^r}\a)Tai, I Pet. iii. 1 3 (Codd. 
 
 K, L) ; interchange of Spaco and Becopeco, &c. ; while the 
 interchange, omission, or insertion of small particles like 
 Kai, de, re, give rise to numberless variations. 
 
 {. The following are probably instances of marginal glosses 
 accidentally incorporated in the text : 
 
 Kat/SXeVe inserted after epxov, Apoc. vi. i, 3, 5, 7. 
 
 Acts XV. 24, Xeyovres Trepirejxvecrdai kuI rrjpelv rov vofiov, 
 „ ,, 34, eSo^e 5e rco '2'CKq. ent/jielvai avrov, 
 
 both which passages are wanting in most of the best 
 MSS. There is a most singular instance in one cursive 
 manuscript, wher§, at 2 Cor. viii. 4, 5, the scribe has 
 
 written de^aaBai T)fias [iv TToXXoty t5>p dvTiypd(f)a>v ovrtos 
 €vpT]Tai\ Ka\ ov Kad(os rj\7riaapev. The WOrds within the 
 
 brackets, which brackets do not of course appear in 
 
VARIOUS READINGS. I9 
 
 the original, were no doubt a marginal no'.e in the codex 
 from which the scribe was copying, and have reference 
 to the words de^aadai rjfxus, which are omitted in the best 
 codices. There was a much stronger tendency to in- 
 sert than to omit ; whence springs the well-known canon 
 iec/i'o prccferatiir hrevior : that is to say, if there are 
 two readings, one longer than the other, the short 
 reading is more likely than the other to be the true 
 one. . 
 In the earlier IMSS. we find many forms of words and 
 expressions that are quite unclassical : such as reaaepa- 
 
 Kovra for T€(T aapaKovTa ; (nretprjs, ActS xxi. 31; paxaipr], 
 
 S. Matt. xxvi. 52; n^Tjfipvprjs, S. Luke vi. 48; the /x 
 constantly inserted in parts of \ap!3avco and its deriva- 
 tives, Xj7/x\//"o/iai, Xrjfxcfydeis, <fec.; the final s of ourcos and the 
 V €4>€\kv(ttik6v constantly affixed even before consonants ; 
 p not assimilated in verbs compounded with ev and a-cv, 
 e.g. evKOKeiv, avvKoKeiv; 2nd aor. forms with ist aor. 
 terminations, as el^a, rjXOa, &c.; and such harsh con- 
 structions as avro 6 coi/, Apoc. i. 4 ; with many more, of 
 which Part II. of Winer's Grammar of N. T. Greek, 
 the Proleofomena of Tischendorfs Greek Test,, or Scri- 
 vener's Introduction, will give examples. These are 
 for the most part altered in the later MSS. into classical 
 forms ; and the phrase above quoted from Apoc. i. 4 is 
 rendered less abrupt by the insertion of tov, as it is now 
 read in the Textus Receptus, ano tov 6 cov, k. t. X. Kuenen 
 and Cobet, in their edition of the Vatican MS. (Ley- 
 den, i860), make merry with the want of scholarlike 
 acumen on the part of editors who retain such forms 
 in their text; and assume that they have their origin 
 solely in the ignorance and ^ pltbeia o-wrjdeia' of the 
 scribes. But they occur with such persistent frequency 
 in the earlier MSS., that it is diflicult to believe that they 
 c 2 
 
20 ON THE ORIGIN OF 
 
 had no place in the original text. At any rate, those 
 editors whose aim is to represent the earliest form of 
 text which they believe attainable according to their 
 principles, are consistent in retaining such forms. (See 
 below, chap, vi.) 
 
 6. Alteration, either by substitution or addition, in order to 
 produce conformity in parallel passages, is a fruitful 
 source of variation. Dr. Tregelles has suggested that 
 Tatian's Diatessaron, formed in the second century, 
 probably fostered this tendency, by bringing the parallel 
 passages into juxtaposition, and thus drawing attention 
 to their differences. But the practice is not by any 
 means confined to the Gospels. Some instances are 
 
 S. Matt. xix. I7j ti- H-^ iptoras Trepi rod dyadov ', els eVrty 6 
 dyados, changed into ri [xs Xeyeis dya66v ] ouSei? dyaObs 
 
 el fiTj eh, from the parallel passages in S. Mark and 
 S. Lukeb. Again, in S. Matt. xvii. 2, for Xevm ms 
 TO cf)cos, D and other authorities have \evKa cos x''^^, from 
 S. Mark ix. 3. In the account of S. Peter's Denial 
 (S. Mark xiv.) several alterations are introduced into 
 Cod. N, apparently to produce harmony with the other 
 
 accounts : 51? is omitted in ver. 30, koi aXexrcop etpavrja-e 
 
 in ver. 68. and ex bevrepov in ver. 72. In Acts ix. 4, 
 (TKkrjpov aoL npbs Kevrpa XaKTi^eiv is added by Cod. E, from 
 the parallel passage in ch. xxvi. 14, to which the words 
 really belong. 
 
 Quotations from the Old Testament are constantly 
 amplified ; as at Rom. xiii. 9, where ov yj^evdofiapTvprjo-eis 
 is inserted in some cursives ; Heb. xii. 20, ^ /3oXtSt Kara- 
 To^evdrjo-erai is added in some after \i6o^okr]6r](jeTai. On 
 S. Matt. XV. 8, see below, p. 74. 
 
 '' A full discussion of this much-disputed reading will be found at 
 
 pp. 
 
VARIOUS READINGS. 21 
 
 Two distinct kinds of variations are assigned to this 
 head : — 
 
 a. Many of our existing MSS. are copies, not of the 
 whole New Testament, nor of consecutive portions of it, 
 but of Lectionaries ; that is to say, collections of passages 
 selected for public reading in the Church services, either 
 as Lessons, or Epistles and Gospels. In passages thus 
 taken out of their connection a word or two must often 
 be added to give a complete sense : sometimes a proper 
 name is substituted for a pronoun; and sometimes a 
 connecting particle will be dropped. All such changes 
 are noted as various readings, though of course they 
 are immaterial to the sense. Hence possibly arose the 
 readings elne fie 6 Kvpios, S.Luke vii. 31, and koI a-rpacpfh 
 irpbs Tovs padqTas ehev, S. Luke X. 2 2. Just the Same 
 sort of variation may be noticed if the Gospels for the 
 third and fourth Sundays after Easter in our Prayer-book, 
 or some of those for the Sundays after Trinity, be com- 
 pared with the same passages as they stand in their 
 original connection. 
 
 13. There are two or three insertions in the New Tes- 
 . tament which have been supposed to have their origin 
 in ecclesiastical usage. The words in question, being 
 familiarly known in a particular connection, were per- 
 haps noted in the margin of some copy, and thence 
 became incorporated by the next transcriber; or a 
 transcriber's own familiarity with the words might have 
 led to his inserting them. This is the source to which 
 Dr. Tregelles assigns the insertion of the Doxology at 
 the close of the Lord's Prayer, in S. Matt, vi., which is 
 wanting in most of the best authorities. Perhaps also 
 Acts viii. 37, containing the baptismal Profession of 
 Faith, which is entirely wanting in the best authorities, 
 found its way into the Latin text in this manner. 
 
22 ON THE ORIGIN OF 
 
 8. A charge of altering the text of the Scriptures with a 
 motive is so serious a matter, that we ought only to make 
 it if supported by very strong grounds. Now there is 
 really no evidence that the transcribers of all our known 
 manuscripts did not do their work in perfect good faith, 
 however many errors they may have allowed to creep 
 in through carelessness or ignorance. Among readings 
 for which this cause has been suggested are the altera- 
 tions in S. Matt. xix. 17 (see above, under No. 6); the 
 variant Kvplov in Acts xx. 28 for deov ; and the substitution 
 
 of ovVo) for ovK, S. John vii. 8 ; 'la)0-?)(^ for Trarrjp avTOv, 
 
 S. Luke ii. 33 ; vl6s for 6e6s, S. John i. 18; the insertion 
 of a mention oi fastijig with praying S. Matt. xvii. 21, 
 S. Mark ix. 29, Acts x. 30, i Cor. vii. 5. 
 
 We are now in a better position, after this enumeration of 
 possible sources of error, to estimate the chances against the 
 original text being preserved unaltered through a series of 
 transcriptions. One would naturally expect a divergence of 
 the text of any given MS. from the original text, proportion- 
 ate to the number of transcriptions it had undergone. Each 
 transcriber in turn would probably import some variations 
 through inadvertence. 
 
 But now another consideration must be added. So 
 long as the transcriptions are made under similar circum- 
 stances, the tendency will be to accumulate errors of the 
 same kind. Hence comes the result, paradoxical at 
 first sight, that from originals, marked by decided individual 
 characteristics, texts may be produced that converge towards, 
 and successively more nearly exhibit, another particular type. 
 ' Groups of copies spring, not from the imperfect re- 
 production of the character of one typical exemplar, but 
 from the multiplication of characteristic variations.' We 
 should expect then to find, in process of time, a number of 
 
VARIOUS READINGS. 23 
 
 ]\ISS., mutually differing from one another in small respects, 
 but tolerably unanimous in presenting a text which will differ 
 in complexion from the text presented by much earlier MSS. ; 
 and that though the former might have been derived by 
 direct descent from the latter. This is just what we do 
 find. 
 
 NOTE. 
 
 It will be instructive to compare with the foregoing list the classi- 
 fication of errors which Professor Madvig makes, in laying down the 
 principles of Textual Criticism as applied to secular writings. (Adver- 
 saria Critica, lib. i. cap. i.) 
 
 1 . Permutation of similar letters, and (in certain cases) of words. 
 
 2. Faulty division and connexion of words. (Cursive MSS. tran- 
 
 scribed from uncials which were written continuously are very 
 liable to this kind of error.) 
 
 3. Doubling of letters, syllables, or words which ought to be written 
 
 once only. 
 
 4. Omission of letters or words (by homoioteleuton or otherwise), 
 
 and transposition by carelessness. 
 
 5. Assimilation of neighbouring words to one another in respect of 
 
 the terminations. 
 The five foregoing sorts of error are unintentional, and arise from 
 failure of attention, or of memory. 
 
 6. Introduction of foreign matter (glosses, &c.). This arises from 
 
 defective knowledge, or error of judgment in the scribe. 
 
 7. Correction or interpolation by the scribes, with greater or less 
 
 degree of intentional alteration of their 'copy.' The most frequent 
 and flagrant cases of this error occur in the later MSS. It is 
 comparatively rare in early MSS. 
 
CHAPTER III. 
 
 ON THE MANUSCRIPTS OF THE GREEK TEXT. 
 
 § I. 071 the number, ?node of designation, &c., of MSS. 
 
 The gross total number of manuscripts of the Greek Text 
 whose existence is known, uncial and cursive included, is 
 stated by Mr. Scrivener to be 1583 ^. It must not however 
 
 * The exact calculation, taken from Scrivener's Introduction, p. 225, 
 corrected by the addenda et corrigenda to that volume at p. viii, is as 
 follows : — 
 
 MSS. of Gospels . . . . 
 
 Acts and Cath. Epp. 
 
 1 „ Pauline Epp. . . 
 
 Apoc 
 
 „ Evangelistaria . . 
 „ Praxap 
 
 Uncial. 
 
 Cursive. 
 
 Duplicates 
 deducted. 
 
 34 
 10 
 
 14 
 
 4 
 
 58 
 7 
 
 601 
 
 228 
 280 
 
 I03 
 
 180 
 
 65 
 
 32 
 
 15 
 
 Total 
 
 127 
 
 1456 
 
 67 
 
 The summary given in Smith's Diet, of Bible, art. 'New Testament,' vol. ii. 
 p. 516, note f, will agree with this, if the corrections from Scrivener's 
 Introduction in the two places above cited be applied. For a full 
 explanation of the last column see below, p. 26. It means that some 
 of the MSS. contain more than one of the parts of the New Testament 
 enumerated in the first column. 
 
DESCRIPTION OF FACSIMILE PLATE. 
 
 No. 1. Seven lines fioni the Codex Sinaiticus, containing 
 S. John xxi. 24, 25, Kat oiSafXfu on dK-rjOrjcr kariv rj ftapTvpia avTov' 
 (crriv de Kai aWa noWa- a ewoiTjcrtv la- ariva lav ypa(pr]Tai KaO 
 
 No. 2. Two passages from the Codex Vaticanus. 
 
 The first two lines, from the first verse of the Epistle to the 
 Ephesians, show how the words iv €<p€<Taj were omitted liy the 
 first scribe and inserted afterwards in the margin. 
 
 The other passage is from S. John xxi. 25 ; and taken with 
 No. I gives an opportunity of comparing the writing of these 
 two great manuscripts. 
 
 The accents in both are due to the scribe who inserted the 
 words (V fipfcFoj. 
 
 Both No. 1 and No. 2 are taken from the Plate in Tischen- 
 dorfs Transcription of the Codex Vaticanus. 
 
 No. 3. This facsimile, which is excellently rendered from the 
 Plate at the end of the Fifth Volume of Tischendorf's Monu- 
 menta Sacra Inedita, exhibits at once specimens of a Palimpsest, 
 an Uncial manuscript of the ninth century, and a Cursive of the 
 thirteenth century. The lower writing, which in Tischendorf's 
 facsimile is of a pale green colour, contains Acts iv. 10-15. 
 
 The passage in Cursive character is Heb. vii. 17-25. 
 
 The abbreviations for t4\os and dpxv^ marking the end and 
 beginning of lections, will be observed in the margin of the 
 Palimpsest writing, in the same ink and by the same hand as the 
 text ; showing that the book was prepared for ecclesiastical use. 
 The accents and other marks are prima maun. 
 
N9 1 . 
 
 Codex Sinaiticus 
 
 S.M.r,.,j:ir, 24-, 26. 
 
 KAI O I A AM eN OTI 
 
 AAHOHCGCTINH 
 
 M APT Yf ' AXTToy- 
 6CT-| M A6KAI AXAX 
 
 rrOAXAXenoiHC^N 
 oioAnriNAexN PfA 
 
 <4^MTAI KAOeN'Oy 
 
 K9 2 . 
 Codex Vatic anus 
 
 ^j>/,es I. 1 
 S.Jokn HI Z5 
 
 A e K A I A A. A j^^o-^A^K6^To| 
 
 H C C N 6 I C i^-TJ NN€\Nrp\'' 
 
 fHce i^>»nrA rpx<i> o mcni 
 
 <! 
 
 R-r M K AeeN o 
 
 I M N I -rb N K o c M o N X a> 
 
N" 3 . 
 I'alinipsest 
 
 7 >// v^>s-><^ cre^S-^r?^- iird^^^ 
 o ^ icCaj-*-*-o o-tV^^^^ c\<[r:>^ ^^*X" 
 
 o ::^ f^^ e «^^ -'^-^^ «^'**^ ^^^'^ 
 
 ''SZiV<^'T^ c^co;/cAj c«-rm^*^*(7y<?p 
 
 00 • 
 
MANUSCRIPTS OF THE GREEK TEXT. 25 
 
 be supposed, either that they are manuscripts of the whole 
 New Testament, or that the contents of all of them have 
 been fully examined. 
 
 The Cod. Sinaiticus is the only uncial that exhibits the 
 whole New Testament entire. Cod. A is very nearly com- 
 plete ; Cod. B is not quite so complete, but has by far the 
 larger part of the New Testament; Cod. C contains por- 
 tions of every one of the books ; and there are twenty-three 
 or twenty-four cursives which are either complete or very 
 nearly so. 
 
 The student should carefully mark the conventions com- 
 monly used in citing MSS. At the same time he must 
 remember that they are not uniformly adopted yet, but that 
 each critic has some slight peculiarities of his own. 
 a. Capital letters are used to denominate uncials ; numerals 
 
 to denominate cursives. 
 /3. The Books of the New Testament were generally divided 
 into four volumes: viz. (i) the Gospels; (2) the Acts 
 and Catholic Episdes ; (3) the Pauline Epistles ; (4) 
 the Apocalypse. If a MS. contain more than one 
 of these volumes, the books generally follow each other 
 in this order. There are a few special exceptions, 
 which are given by Scrivener (Introd. p. 61, § 24). 
 Besides these four volumes, there are the Lectionaries 
 (see p. 21), denominated (5) Evangelistaria, or (6) Prax- 
 apostoli, according as the selection of passages is made 
 from the Gospels, or Acts and Epistles. Our exist- 
 ing MSS., whether uncials or cursives, are thus distri- 
 buted into six groups. Now, with regard to the first 
 four of these groups, it is to be remarked that //te series 
 of letters and numerals commence over again for each 
 group : consequently, a IMS. which includes more than 
 one of these volumes will be counted afresh in each 
 series, and possibly not in the same place in the series. 
 
26 ON THE MANUSCRIPTS 
 
 Thus we find different MSS. denoted by the same letter 
 or numeral, in different parts of the New Testament ; 
 and the same MS. denoted by different letters, or 
 numerals, in the different parts : e. g. X, A and C, 
 whose readings run through the whole of the New 
 Testament, are quoted by the same letters everywhere ; 
 but B, the letter under which for the first three of the 
 four volumes the well-known Vatican MS. is cited, is 
 assigned to a different MS. (Cod. Basilianus) in the 
 Apocalypse, which book is wanting in the great Vatican 
 MS. So D is the designation of Cod. Bezae in the 
 Gospels and Acts, but of Cod. Claromontanus in the 
 Episdes of S. Paul ; and E means Cod. Basiliensis for 
 the Gospels, Cod. Laudianus for the Acts, and Cod. 
 Sangermanensis for S. Paul's Epistles ^ On the other 
 hand, Gosp. 33 is the same MS. as Acts 13 and Paul 
 17, &c. &c. 
 
 y. Where a MS., as is frequently the case, has been cor- 
 rected by later hands, it is customary to distinguish the 
 readings of the different correctors by small numerals 
 placed above, and to the right of, the letter denoting 
 the MS. ; like the index of an algebraical power : e. g. 
 C^, C^, C^, would denote readings introduced by first, 
 second, or third corrector respectively. An asterisk (*) 
 affixed in the same way denotes the reading of the 
 original scribe. 
 Many of the uncial MSS. are mere fragments ; some of 
 
 them contain but a few verses. The readings of all the 
 
 ^ It would seem convenient if the well-known mathematical conven- 
 tion suggested by Professor Westcott in the Article ' New Testament,' in 
 Smith's Diet, of Bible, were adopted : — viz. that where the same letter is 
 used more than once, the cases should be distinguished by a small nume- 
 ral subscribed to the letter. Thus in the examples above given we 
 should distinguish the MSS. as Bi, Bo ; E^, E2, Eg, &c. 
 
OF THE GREEK TEXT. 27 
 
 uncials may be considered to be satisfactorily determined. 
 Of the cursives, on the other hand, Scrivener marks only 
 about 1 40 in his list with an asterisk, by which he indicates 
 that a MS. has been thoroughly collated. Many more than 
 this have been inspected, and collated more or less carefully ; 
 but the work of earlier critics is often not marked by ihe 
 scrupulous accuracy that is now demanded. 
 
 § 2. On some palcEOgraphic details. 
 
 For a minute description of the materials and palseogra- 
 phic details of MSS., the reader should consult some larger 
 work, such as Mr. Scrivener's; or Home and Tregelles" 
 Introduction, vol. iv. A very few remarks will suffice for 
 the object of this book. Uncial characters were employed 
 down to the tenth or eleventh centuries ; but cursive charac- 
 ters began to come into use as early as the ninth ; therefore 
 we have some cursive MSS. older than some uncials. 
 
 It may be laid down as a general rule that the more up- 
 right, square, and simple the uncial characters are, the earlier 
 is the writing. Narrow, oblong, and leaning characters came 
 in later, together with greater elaborateness in style. Absence 
 of initial letters of larger size than the rest is a mark of anti- 
 quity. In the earlier MSS, marks of breathing, accent, and 
 punctuation are very rare, frequently absent altogether ; or, 
 if present, inserted on no apparent fixed principle, except 
 that a dot, to mark the division of sentences, became pretty 
 general about the beginning of the fifth century. In some 
 MSS., where the accents appear, they have been added by 
 a much later hand. From these and other marks of a like 
 kind it is perfectly possible for an expert to fix the date of 
 any given manuscript by inspection, to within fifty years 
 at the outside, without regard to the subject matter of the 
 book. ^ 
 
28 ' ON THE MANUSCRIPTS 
 
 § 3. 071 the various systems 0/ divismis of the text. 
 
 We must first notice the arrangement of the text in o-tLxoi 
 {line-clauses). The use of this would probably be to assist 
 the reader; for the length of the lines varies considerably, 
 answering closely enough to the intervals at which we should 
 place commas to mark the slight pauses in the sense. Not 
 many MSS. are arranged in this way; the waste of space 
 being too considerable in days when vellum was the only 
 material for books of value. 
 
 It is uncertain who divided the Gospels on this principle. 
 S. Paul's Epistles, and the Acts and Catholic Epistles, were 
 so arranged by Euthalius, a deacon of Alexandria, in the 
 latter half of the fifth century; the Pauline Epistles in 458 ; 
 the Acts and Cath. Epp. in 490 a. d. 
 
 The subscriptions however in Cod. N to some of the Epi- 
 stles, by a very ancient hand, recording the numbers of the • 
 (TTixoi in the Episdes (which numbers agree with those of 
 Euthalius in only one instance), seem to indicate the existence 
 of a still older division on this principle. 
 
 I. Of the Gospels there are several systems of division. 
 
 1, The one which seems to be the oldest of all extant is 
 
 found in Cod. B and Cod. S only. The breaks depend 
 on the sense ; a fresh section commencing wherever a 
 new subject is introduced. Hence, though valuable to 
 the reader, it would be an inconvenient division for 
 public use, since the sections are of very unequal length. 
 They are reproduced in Dr. Tregelles' edition of the 
 Greek Text. 
 
 2, Next in order both of antiquity and importance are the 
 
 so-called Ammonian Sections. It seems probable that 
 the divisions, as they stand at present, are not to be 
 attributed to Ammonius of Alexandria (third century), 
 from whom they take their name ; but that they are to 
 
OF THE GREEK TEXT. 29 
 
 be assigned to Eusebius of Ccesarea (fourth century), 
 in connexion with whose useful and ingenious system 
 of Canons they are most known. 
 Ammonius' idea was to form a harmony of the four Gos- 
 pels, taking that of S. Matthew for the basis, and Arranging 
 the others in parallel columns with it, where the accounts 
 coincided. Thus of course the thread of narrative of the 
 other three was broken. Eusebius' intention seems to have 
 been slightly different from this. He worked out a system 
 for indicating the parallel passages between the Gospels, 
 without destroying the sequence of any of them. A very 
 slight examination of the Canons will shew that by parallel 
 passages (to. TrapaTrX^a-ia) Eusebius means passages ivhich are 
 illustrative of one another^ and not passages which give ac- 
 counts of the same events : e. g. the Miraculous Draught of 
 Fishes after the Resurrection (S. John xxi. 1-6) is compared 
 with the similar miracle at the beginning of our Lord's Minis- 
 try (S. Luke v. 4-7) (see Canon IX). The object aimed at, 
 in short, is rather that of our marginal references ^ than a 
 harmony properly so called. 
 
 The length of the Ammonian Sections depends, not on 
 the sense, but upon the verbal coincidence or disagreement 
 of one Evangelist with another. Each Gospel is divided on 
 this principle, and its sections are numbered continuously from 
 the beginning throughout. S. INIatthew's Gospel.contains 355; 
 S. Mark's 233 down to c. xvi. 8 ; S. Luke's 342 ; and S. John's 
 232. Eusebius formed ten Tables (Canons) : No. i contains 
 a list of the places (seventy-one) in which all four Evangel- 
 ists agree ; Nos. 2,3, and 4 contain lists of places in which 
 three of them have something in common; Nos. 5, 6, 7, 8, 
 
 ^ The writer -was indebted for the substance of this and the last 
 paragraph to a report Ijy a pupil of a lecture of Mr. Burgon's. The 
 student may now see Mr. Burgon's views learnedly discussed in his 
 book on The Last Twelve Verses of S. Mark, Appendix G. 
 
3© ON THE MANUSCRIPTS 
 
 and 9 contain lists of places in which two combine; and 
 No. lo a list of sixty- two passages peculiar to some one 
 Evangelist. In the Greek MSS. of course the numbers are 
 given in the Greek letters which denote the numerals in ques- 
 tion. The Canons and Sections may be found in Words- 
 worth's and Tischendorfs editions of the Greek Testament, 
 transcribed into the corresponding Arabic numerals. 
 
 The method of using them is as follows. Let us suppose 
 that we want to find the parallel passages to some given 
 passage; say S. Matt. xxii. 15, &c. : we find two numbers, 
 223 and 2, prefixed to the passage, one placed above the other; 
 the upper number (223) is the number of the Section, the 
 lower one (2) denotes the Table. We refer then to Table II 
 which we find contains passages common to S. Matthew, 
 S. Mark, and S. Luke; and in a horizontal hne with the 
 No. 223 of S. Matthew's order we find the numbers of the 
 parallel Sections in S. Ma^k and S. Luke, viz. 130 and 243. 
 
 The numbers indicating the Sections are found noted in 
 the margin of by far the larger number of known MSS. ; the 
 numbers of the Canons being also added in most cases, 
 though wanting in a few examples. The earliest instance of 
 their occurrence is in Cod. ^{, written by a contemporary 
 hand with the scribe, if not by the scribe himself. Cod. B 
 does not exhibit them. 
 
 3. The tItKoi, sometimes called KecfydXaia, but improperly and 
 inconveniently, inasmuch as this designation properly 
 belongs to the Ammonian Sections, are another set of 
 divisions of the Gospels very commonly found. This 
 division is made according to the sense ; but a tiVXos 
 sometimes contains more than one subject. The name 
 is apparently derived from the fact that each section 
 has a short descriptive heading or designation, taken 
 from the first or principal subject contained in it : e. g. 
 the Sermon on the Mount, which forms the fifth of the 
 
OF THE GREEK TEXT. 3 1 
 
 t'ltXoi cf S. Matthew's Gospel, is headed irepX tcjv fxaKa- 
 piafiMv. These headings are noted sometimes in the 
 margin, sometimes at the head or foot of the page, or 
 both together ; and a list of them is generally prefixed 
 to each book. They may be seen in Dr. Tregelles' 
 edition of the Greek Testament. This division was 
 perhaps made for the sake of convenience in public 
 reading. No trace of it is found in Codd. N or B. 
 II. There are also several modes of dividing the Acts 
 and Epistles. 
 
 1. A continuation of the old system above mentioned (p. 28) 
 
 is found in Cod. B. It presents, in respect of S. Paul's 
 Epistles, two interesting peculiarities : viz. (i) The Epi- 
 stles of S. Paul are numbered continuously throughout, 
 as if they formed but one book. (2) Whoever invented 
 this division placed the Epistle to the Hebrews (as is 
 shown by the numbers of the Sections in the margin) 
 between the Episdes to the Galatians and Ephesians. 
 Though in Cod. B itself the Episde in question stands 
 next after 2 Thess., yet the numbering of its Sections 
 runs on continuously from the Epistle to the Galatians. 
 The last Section of the Epistle to the Galatians is num- 
 bered 58 ; the first of the Epistle to the Hebrews is 59 ; 
 while the first of the Epistle to the Ephesians is 70. 
 The end of the Epistle to the Hebrews is lost; but 
 there can be no reasonable doubt that the numbering 
 of the Sections from it to the Epistle to the Ephesians 
 would be consecutive. 
 
 2. Another system, later than the last, is also found in Cod. 
 
 B. In the Acts, the Sections are shorter, and therefore 
 more numerous ; in the Epistles the opposite is the 
 case. In this system the Pauline, as well as the Catholic 
 Epistles, are divided independently. The first forty-two 
 of the Sections in the Acts of the Apostles are noted in 
 
32 ON THE MANUSCRIPTS 
 
 the margin of Cod. N by a hand ahuost as old as the 
 original scribe ; but with some want of care apparently, 
 since there are five slight omissions and variations. 
 
 3, The K€^aXaia, commonly attributed to EuthaHus, analogous 
 
 to the rtVXoi of the Gospels, and accompanied like them 
 by short headings or summaries of contents. Though 
 often attributed to Euthalius, they were probably only 
 introduced into common use by him. 
 
 4. Another division of the Acts and Pauline Epistles into avayvoi- 
 
 aeis or dvayvooo-ixara (lessons), also attributed to Euthalius. 
 
 III. The Apocalypse was divided at the end of the fifth 
 
 century by Andreas, Archbishop of Csesarea in Cappadocia, 
 
 into twenty-four \6yoi, each Xoyos being subdivided into three 
 
 KeCJ)dXaia. 
 
 Important evidence is sometimes gained from attending 
 to the presence or absence of such extra-textual marks as 
 these. For instance, the peculiarity in the Vatican num- 
 bering of the Sections of the Epistles noticed above, gives 
 proof of a valuable and unconscious kind that the Episde to 
 the Hebrews was looked upon in very early times as being 
 certainly by S. Paul. Again, the passage S. Mark xvi. 9-20 
 appears to have had no place assigned to it 5y Eusehius 
 among his (' Ammonian') Sections; which would show that 
 he at all events did not consider the passage part of the sacred 
 text. On the other hand, in two passages, viz. S. Luke xxii. 
 43, 44, where the sum of the evidence is decidedly in favour 
 of retaining the disputed clause, and S. Mark xv. 28, where 
 the evidence is for rejection, the testimony of the Sections 
 and Canons of Eusebius is in favour of both. In the first 
 case, the clause has a special number, and is placed in the 
 tenth Canon, which contains passages peculiar to the several 
 Gospels; in the second case, the clause, likewise specially 
 numbered, is assigned to the eighth Canon, which contains 
 passages common to S. Mark and S. Luke. 
 
OF THE GREEK TEXT. 33 
 
 § 4. Aft Accoimt of Codd. X andB. 
 
 The reader should consult Mr. Scrivener's Introduction, 
 or some other such authority, for an account of the chief 
 MSS. quoted in critical editions of the New Testament. We 
 will however, as an illustration of many points of criticism, 
 proceed to give a somewhat detailed account of the two 
 great manuscripts N* and B. Our authorities are chiefly 
 Mr. Scrivener's collation of the Cod. Sinait., published with 
 a Critical Introduction in 1864, and the Prolegomena to 
 Tischendorf's smaller editions of these manuscripts. 
 
 (i) Codex Sinaiticus (X). 
 
 In 1844, Tischendorf, travelling under the patronage of 
 Frederick Augustus, King of Saxony, and being at the 
 Monastery of S. Catherine, Mount Sinai, saw some vellum 
 leaves of a manuscript, apparently very ancient, in a basketful 
 of papers intended for the stove. He picked out forty-three 
 leaves, which he obtained for the asking. They contained 
 portions of the Septuagint version, viz. parts of i Chron. and 
 Jeremiah, with the whole of Nehemiah and Esther. The 
 monks however, having been informed that they belonged 
 to a MS. of probably the fourth century, concealed the 
 remainder of the MS., and Tischendorf could get nothing 
 more from them for that time. These forty-three leaves 
 he brought to Europe, and published with the title of 
 Codex Friderico-Augustanus. 
 
 He was again at S. Catherine's in 1853, but could gain 
 no further tidings of the MS. But in 1859 he went for the 
 third time to the East under the patronage of the Emperor 
 of Russia ; and one day, being once more at the Convent, 
 the steward showed him as a curiosity a MS. which he had 
 long kept in his cell. It turned out to be the missing treasure, 
 
 D 
 
34 ON THE MANUSCRIPTS 
 
 which he was now allowed to examine at leisure, and 
 which he found to contain, besides a great deal of the Old 
 Testament, the whole of the New Testament, the Epistle of 
 Barnabas entire, of which the first four and a half chapters 
 had been hitherto known in a Latin translation only, and 
 a large fragment of the Shepherd of Hermas in the original 
 Greek, which was before extant as a whole only in the Latin. 
 He soon recognized its immense value ; and, after studying 
 it for a while at Cairo, he suggested to the community that 
 they should present it to the Emperor of Russia, the great 
 patron of the Greek Church. It is now at Petersburg. 
 
 The manuscript is written on very fine vellum. The size 
 of the pages, notwithstanding mutilation by the binders, is 
 still i3|x 14^ inches. The sheets, forming only two leaves 
 each, and each requiring the skin of a single animal, are 
 arranged in sets, or quires, of four (quaferm'ones). Now this 
 fact is interesting because Eusebius was ordered by the Em- 
 peror Constantine (a.d. 331) to procure fifty copies of the 
 Scriptures, handsomely got up and well written, for the 
 churches in his new city of Constantinople. Eusebius d re- 
 lates that this was done, and that these fifty copies had their 
 sheets arranged in ternmies and quatefniones. It has been 
 suggested, with some show of plausibility, that the Cod. 
 Sinait. is one of these very fifty copies. We shall see, by 
 and bye, that Cod. B could not have been one of 
 them. 
 
 The text on each page is arranged m/our columns. This 
 is supposed to be in imitation of the papyrus rolls, and is an 
 unique arrangement so far as we know. Cod. B has three 
 columns on a page. 
 
 The writing is in plain, somewhat square, uncials ; with- 
 out spaces between the words, or breathings (except in one 
 
 d Vit. Const, lib. iv. c. 36, 37. 
 
OF THE GREEK TEXT, ^^ 
 
 place, Gal. v. 21), or accents, or iota post- or jz/<5-script : 
 there are very few marks of punctuation, but part of a line 
 is often left blank at the end of a sentence. 
 
 It must have been copied line for line from some other 
 MS., since omissions of exactly the number of letters that 
 would complete a line are found, and that in two ways : viz. 
 sometimes as if a line were dropped accidentally, and some- 
 times as if the eye of the scribe wandered from the middle 
 of one line to the middle of the next line below. Instances 
 of the error hovioioteleuion are numerous ; 115 occurring in 
 the New Testament portion alone. 
 
 Tischendorf thinks that four scribes were engaged alto- 
 gether on the manuscript, but that two only of these exe- 
 cuted any portion of the New Testament. One of these 
 uses a particular mark in the margin ( > ) to indicate quo- 
 tations from the Old Testament : this is a little fact the 
 importance of noticing which will appear by and bye. 
 
 The same critic assigns the numerous corrections, from 
 those by the original scribes themselves down to three made 
 by some hand in the twelfth century, to as many as twelve 
 correctors ; and thinks that the scribe who used the sign ( > ), 
 mentioned in the last paragraph, performed the office of 
 diop6oiTTjs. In the eighth century the ink had become so 
 faded that it was necessary to retrace the whole of the 
 writing throughout the manuscript. 
 
 It has been already mentioned that the division into rtVXot 
 is wanting ; but that the Ammonian Sections are marked ; 
 and that in the Acts there is a division which is found besides 
 only in Cod. B. 
 
 The passage S. Mark xvi. 9-20 is wanting; and the scribe 
 appears to be conscious of no omission; for, according to 
 his custom when beginning a new book, he begins S. Luke's 
 Gospel at the top of the next column. In Eph. i. i, the 
 words €v 'E^eVo) are wanting, prima manu^ being added by a 
 
 D 2 
 
3<^ ON THE MANUSCRIPTS 
 
 much later hand. The episode S. John vii. 53 — viii. 1 1 is 
 wanting, no gap or sign of omission being made by the 
 scribe. The Epistles of S. Paul precede the Acts, a peculi- 
 arity observed only in four other MSS., and those cursives. 
 The Epistle to the Hebrews has the position usual in the 
 oldest MSS., viz. after 2 Thess. and before the Pastoral 
 Epistles. 
 
 The arguments for determining its date are such as 
 follow : — 
 
 1. The beauty of the vellum. 
 
 2. The shape of the letters. 
 
 3. Absence of punctuation. 
 
 4. Absence of initial letters larger than the rest. 
 
 5. Arrangement of four columns on a page. 
 
 6. The extreme simpHcity of the titles of the books, which 
 
 exceeds that of all other known MSS. : e. g. Kara Mad- 
 6aiov, without evayyeXtov ; Trpa^cis, without dnoaToXcov ; 
 Tvpos 'Pcoixaiovs, without eTTLaToXrj. 
 
 7. The fact above mentioned of the ink having so faded by 
 
 the eighth century that the whole MS. had to be inked 
 over again. 
 All these points are arguments for great antiquity. 
 
 8. But further, the absence of the tltXoi, which came into 
 
 general use in the fifth century ; and 
 
 9. The presence of the Epistle of Barnabas and the Shep- 
 
 herd of Hermas, written exactly in the same way as the 
 rest of the book, would lead us to place it at least as 
 early as the fourth century ; for these two books belong 
 to the so-called avrikeyofxeva {disputed books), which were 
 not definitely excluded from the Canon, but were read 
 publicly, until towards the close of the fourth century. 
 
 10. Yet, on the other hand, the presence of the Eusebian 
 Canons will not allow it to be dated earlier than about 
 the middle of that century. 
 
OF THE GREEK TEXT. 37 
 
 The student should take notice that every one of these 
 arguments is independent of any t7iternal considerations of 
 the character of the text, pecuhar readings, and so forth. 
 
 (ii) Codex Vaticaiius (B). Vat. 1209. 
 
 A special interest has been attached to the great Vatican 
 manuscript, apart from its high value and antiquity, owing 
 to the difficulty which the jealousy of the Papal government 
 has always thrown in the way of strangers, however compe- 
 tent, who wished to examine it. 
 
 The MS. appears to have been in the Vatican Library 
 almost from the establishment of that library by Pope 
 Nicholas V (d. 1455); but it is first distinctly heard of in 
 the correspondence of Sepulveda with Erasmus in 1534. 
 
 The first regular collation of it was made by Bartolocci, 
 then librarian, in 1669; but was not used by anyone be- 
 fore Scholz (1820-1852), and Muralt (1844). The second 
 and third collations, known as Bentley's, were made at his 
 request by ]\Iico and Rulotta, two Roman Abbati, circ. 1720- 
 1730. The next is that of Birch of Copenhagen (1780- 
 1790). All these were more or less inaccurate. After this 
 there was no pretence of a regular collation. Hug saw and 
 commented on the MS. when it was at Paris in 18 10, but 
 did not collate it. Tischendorf in 1842, Dr. Tregelles in 
 1845-6, Dean Alford and Mr. Burgon in 1861, Mr. Cure in 
 1862, all had glimpses of it, and examined certain readings. 
 The editions of Cardinal Mai and Vercellone had appeared 
 in 1858-9; and, inaccurate as they were, added much to 
 our knowledge. It is no small benefit that they gave occa- 
 sion for the masterly preface of Professors Kuenen and 
 Cobet (of Leyden) in their transcript of the codex e. 
 
 « Novum Testamentum ad fidem Codicis Vaticani edidenint 
 A. Kuenen Theol. in Acad. Lugduno-Batava Prof, et C. G Cobct, 
 Litt, Human, in Acad. Lugduno-Batava Prof. i860. 
 
38 ON THE MANUSCRIPTS 
 
 Tischendorf had an opportunity of making a fuller exami- 
 nation of it in 1866. At first he had obtained leave to col- 
 late the codex, but not to publish a facsimile edition, as he 
 wished. However, after he had been at work on it for ten 
 days at the rate of three hours a day, which was all the time 
 allowed, his earnestness aroused jealousy, and further access 
 was refused him. Upon further application, and by the assist- 
 ance of Signor Vercellone, he was at last allowed to consult 
 the MS. again for all doubtful readings, but not thoroughly 
 to collate it: and, making the best use he could of this 
 opportunity, in forty-two hours' work, including the thirty 
 hours already nwitioned, he collated fully the first three Gos- 
 pels-, copied in facsimile about twenty pages, and collated 
 all doubtful passages through the New Testament. From 
 this examination he was able to form some conclusions on 
 various palseographic details. 
 
 Since that time a facsimile edition, worked from the types 
 which Tischendorf had had cast at Leipsic for his edition of 
 the Sinaitic MS., has issued from the Roman press. The 
 writing of the two MSS. is so nearly alike that this is a fair 
 representation. This edition, though not absolutely accu- 
 rate, supplies much additional help : and on the whole, from 
 this, together with Tischendorf 's labours and the previous 
 collations, we have a tolerably complete knowledge of all 
 the readings of this important MS., and of its history, so 
 far as a MS. can be made to tell its own history. 
 
 It is written on very fine thin vellum, in uncial characters 
 at once bold and dehcate, on the whole resembling those of 
 N very closely, but rather smaller. The size of the pages 
 too is less than in that manuscript, but they are of very 
 similar proportions. The writing is arranged in three co- 
 lumns to a page ; the initial letters are no larger than the 
 rest; the ink is of a reddish-brown colour. The accents 
 and breathings, which appear throughout the volume, have 
 
OF THE GREEK TEXT. 39 
 
 been added by a later hand than the original scribe; but 
 there are some particular marks due to him, e. g. the marks 
 of quotation ( > >), a small line interposed at the beginning 
 of a section, the apostrophus ( ' ), and a punctuation. The 
 sheets are arranged in quires of five (qmniones), not in ter- 
 niones or quateniiones ; whence it appears that Cod. B can- 
 not be one of Constantine's fifty, spoken of above (see 
 p. 34). The writing has been traced over afresh by a later 
 hand throughout the MS., except where some letters are 
 purposely passed over as erroneous. This, as in the case 
 of Cod. ^<, would only have been done when the original ink 
 had faded from age. 
 
 As to the contents of the codex, the latter part of the 
 Epistle to the Hebrews, the Pastoral Episdes, and the Apo- 
 calypse, are wanting. This however is due simply to muti- 
 lation. The MS. breaks off at Heb. ix. 14 in the middle of 
 the word Kadapu7. The passage S.John vii. 53-viii. 11 is 
 omitted without any gap or sign of omission. The words eV 
 'E^eo-o) (Eph. i. i) are wanting, just as in Cod. N. The con- 
 clusion of S. Mark's Gospel is omitted; but the scribe, con- 
 trary to his usual custom, leaves a whole column blank before 
 the commencement of the next book, as if aware of an omis- 
 sion. This may be construed to mean that a conclusion to 
 the Gospel was known at the time when the manuscript was 
 written, but that the scribe thought there was insufficient 
 authority for it. 
 
 We have already spoken of the information given by the 
 numbering of the sections in the Epistles (see p. 31); and of 
 the peculiar division of the Gospels which this MS. possesses 
 in place of the WrXoi and Ammonian Sections (see p. 28). 
 
 There appear to have been only three correctors whose 
 readings are of any importance : — 
 
 I. The original scribe made corrections of some slips in the 
 course of transcription, besides adding, probably from 
 
40 ON THE MANUSCRIPTS K 
 
 the copy before him, some various readings in the 
 margin, distinguished by a peculiar mark (s). 
 
 2. The dtopdcoTTjs introduced some readings from an appa- 
 
 rently independent exemplar. 
 
 3. A third hand, when the writing had faded from age, 
 
 inked over the whole, added the accents and breath- 
 ings, and corrected it throughout by a copy of his own 
 time. That the accents are due to this corrector is 
 evident from the fact that \yh^00ke omitted to ink 
 over the letters or syllables, as he frequently did by way 
 of correction, the accents are not inserted. He imitates 
 for the most part the writing of the original where he 
 adds anything; yet in some places, where he was 
 pressed for room, he uses forms of letters and abbre- 
 viations that belong to the tenth and eleventh centuries. 
 It is certain that the corrector who uses these abbre- 
 viations is the person who retraced the faded writing, 
 because occasionally an abbreviation occurs in a cor- 
 rection along with an omission to ink over some of the 
 letters : e. g. S. Matt. xvi. 19, for daxra aoi raa- KXfidaa 
 (the original reading) he wishes to substitute the read- 
 ing found in the Textus Receptus, km daaco aoL raa- 
 fcXeio-. He effects this by inserting the abbreviation c*. 
 before ficoo-co, omitting to ink over the syllable -dav, and 
 writing o- in the late cursive form, instead of the uncial 
 form, above it. Tischendorf considers the text from 
 which he took his corrections to be destitute of all the 
 characteristics of very ancient codices. 
 There are a few unimportant additions by other hands, 
 
 e. g.— 
 
 The subscriptions to S. Paul's Epistles are in uncial writing 
 of about the sixth century. 
 
 The coloured initial letters belong apparently to the tenth or 
 eleventh centuries. 
 
4p of the greek text. 41 
 
 There are sundry marginal notes, e.g. apxr), Ti\o<:, vnep^a..., 
 &c., which perhaps indicate that the MS. was at some 
 time used for public reading. 
 
 Many of the arguments for the age of this MS. are the 
 same, or nearly so, as those for the age of the Cod. Sinai - 
 ticus. It is assigned without hesitation to the fourth cen- 
 tury. 
 
 The next point which claims our attention however, and 
 which is extremel5l^i|g.eresting if true, is the connection 
 which Tischendorf believes he has discovered between these 
 two great MSS. It must be confessed that he makes out 
 a strong case : though the force of some of the arguments 
 will be best appreciated by those who have an oppor- 
 tunity of working carefully through them in detail; based 
 as they are in part upon a multitude of minute points, of 
 which only an instance or two can be given here by way 
 of specimens. 
 
 Certain general points of resemblance between these j\ISS. 
 have been already noticed incidentally ; but a minute inspec- 
 tion brings others to light. 
 
 It has been asserted that the first scribe of B used no 
 punctuation. This seems to be a mistake. It is true that 
 the points have often faded, so as to be visible only to prac- 
 tised eyes : but in some places within a space may be seen 
 the points of the first scribe side by side with those of the 
 restorer, proving the fact. He was however irregular in his 
 system, sometimes using a space of about one letter's 
 breadth or less, sometimes a dot without a space, some- 
 times both, sometimes neither. The use of a space in the 
 middle of a Hne without a dot is a noticeable peculiarity of 
 his ; so is the use of a double point, like our colon ( : ), at 
 the end of a book. Now here an interesting question arises. 
 It is shown that four hands were engaged in transcribing 
 Cod. ^{: of whom one, denominated D by Tischendorf, 
 
42 ON THE MANUSCRIPTS 
 
 executed six sheets of the New Testament, the Books of 
 Judith, Tobit, and part of i Maccabees ; besides adding the 
 inscriptions and subscriptions to the books, and the tides of 
 the pages ; and correcting the work of his associates. Now, 
 besides the general resemblance of Cod. B to Cod. X above 
 alluded to, we find that Cod. B bears a far more striking re- 
 semblance to those parts of Cod. N which were executed by the 
 scribe in question, than to the rest. For instance, (i) these par- 
 ticular parts of Cod. J^ have these two peculiarities in punc- 
 tuation. (2) They have also a very peculiar form of the 
 letter S. (3) There are some arbitrary signs and arabesques 
 in Cod. B in vermilion paint, which resemble one at the end 
 of S. Mark's Gospel in Cod. X written by the scribe D, and 
 one at the end of the Apocalypse, of which D wrote the 
 beginning. (4) There is great similarity in the use of cer- 
 tain contractions. (5) There are similar * itacisms,' e. g. 
 generally Cod. X has i for et, except in D's portion, where 
 the opposite is the case : Cod. B has ei for i constantly. 
 Again, Cod. ^5 has Icoawrjs, except in D's portion, where we 
 find Icoavrjs; and in one place just after D's portion is 
 finished, where his fellow- scribe writes laavrjs once, and then 
 falls back into the other spelling : Cod. B has Icoavrjs through- 
 out. These are samples of arguments which, taken together, 
 make it seem not unlikely that the Sinaitic scribe D was 
 also the transcriber of Cod. B. If this be so, a very inter- 
 esting relation would be established between the two MSS. ; 
 and one not only interesting but important. For in the first 
 place they are evidently transcribed from different originals, 
 since their texts differ in many places : if therefore it be true 
 that they both were written in the fourth century, their agree- 
 ment carries us back to a text of still higher antiquity. But 
 this is not all. Cod. {< was corrected throughout by two 
 correctors, coeval with the original scribe, and using dif- 
 ferent exemplars: it really therefore supplies us with the 
 
OF THE GREEK TEXT. 43 
 
 evidence of three MSS., all older than itself, and not impro- 
 bably considerably older ; for of course an old and standard 
 copy would be probably used, in preference to one more 
 recent, for purposes of correction. And Cod. B, as stated 
 before (p. 14), has been corrected throughout by one con- 
 temporary hand, and therefore supplies us with the evidence 
 of two older MSS. than itself. The two codices together 
 therefore supply us with the evidence of five MSS. of earlier 
 date than the middle of the fourth century; whose conver- 
 gence of course carries us back to a text of very early date. 
 
CHAPTER IV. 
 
 ON VERSIONS, AND THE CHIEF VERSIONS OF THE NEW TESTAMENT. 
 
 § I . On the nature and value of the evidejice given 
 by Versions. 
 
 By a Version is meant, as has been already said, a trans- 
 lation into some other language than the original. In the 
 case of the New Testament the Greek text has undergone 
 this process of translation sooner or later into the language 
 of almost every people that has been Christianized : but not 
 all of these versions are of critical value. A version like our 
 English version, for instance, may be very admirable, and for 
 the time when it was made a very masterpiece of rendering, 
 and yet possess no value for a critic of the Greek text. 
 
 The older versions have been transmitted to us in manu- 
 scripts, just as the Greek original has been. In some lan- 
 guages we possess large numbers, very diverse in age and 
 character and value ; in others the total number is very scanty. 
 These texts are liable to similar casualties of transmission 
 as the Greek text; but the process of deterioration could 
 scarcely ever affect documents in different languages, in the 
 same passages, in precisely the same way. Hence if an 
 ancient version accords with the early Greek MSS. in some 
 particular reading, we have at least an important proof of the 
 early prevalence of that reading. If a second version support 
 the reading in question, the weight of evidence in its favour 
 becomes enormously greater. 
 
ON VERSIONS. 
 
 45 
 
 On such points as the omission of words and clauses, 
 versions give as clear evidence as the original Greek MSS. 
 do ; and it is quite possible that even where they are not 
 precisely exact in their renderings they may be far from mis- 
 leading ; nay, they viay even indicate the true reading, since 
 it may be evident how the error arose : e. g. when in the 
 ^thiopic version. there is found (i Cor. xii. 28) 'an ear/ it 
 is clear that the translator, not very well acquainted with 
 Greek, confused ov? with ovs ; and from the very impossi- 
 bihty of his translation we infer that he must have read OYC. 
 There are other mistranslations, which would not long mis- 
 lead the critic, in the same version: e.g. in S. IMatt. iv. 13, 
 opioid seems to be supposed to be connected with opos ; and 
 in Rom. vii. 11, e^endTr^ae seems to have been read for e^rj- 
 Trd-njae. So in our English version we find (Heb. x. 23), 
 ' Let us hold fast the profession of our faith," where there is 
 not a single MS. authority for the word '■faith' ; but the 
 compositor's eye in the first edition perhaps rested upon the 
 word 'faithful' in the line immediately below; so it crept 
 in accidentally, and has never been corrected. The true 
 reading is liope! 
 
 The earhest Latin versions were so literal that they even 
 give evidence on the order of the words; the Greek order 
 being, retained even w^here it is not in accordance with the 
 genius of the vernacular. Some Greek idioms too, such as 
 a genitive absolute for the ablative, are retained. 
 
 It was long before the critical value of versions was ap- 
 preciated. The study of them has been in general too 
 subordinate to that of the Greek text, even where attention 
 has been paid to them. But in some cases, and pre-emi- 
 nendy in the case of the Latin versions, there is a grand field 
 for independent criticism, which is only now beginning to 
 be systematically explored. 
 
 In giving a short account of all the versions which 
 
46 THE CHIEF VERSIONS OF 
 
 have a critical value, it is convenient to take the Latin ver- 
 sions first ; because some points in their early history are 
 known for certain, which are matters of conjecture, though 
 of conjecture little short of certainty, in the history of the 
 Syriac, the next most important, versions. There is a special 
 interest too for us in the Latin, because the Vulgate was for 
 centuries the Bible of the West : our Reformers were trained 
 upon it; and our Prayer-book version of the Psalms is 
 founded upon S.Jerome's 'Galilean' Psalter. 
 
 § 2. The Latin Versions. 
 At the time of our Saviour Greek was the language most 
 widely spread through the world. Every educated Roman 
 spoke it freely. It was current ^ in the civilized East, at Rome 
 itself, and in Roman Europe as far west as Gaul. Greek was 
 in fact the common language of communication, the French 
 of that period. Within the range therefore of refined Roman 
 society, even if Christianity had spread more widely than it 
 appears to have done at first among the upper classes, the 
 want of a vernacular translation of the Scriptures would 
 hardly have been felt; and it was not felt in the Roman 
 Church as it actually existed, with at all events a very large 
 Greek element among its members. In Africa, not in Italy, 
 the first Latin version was formed. This is the Vetus Laiina, 
 with its strange and uncouth latinity. We know nothing of 
 its origin ; but in Tertullian's writings, and in the Latin 
 translation of Irenaeus, we see that it is in full possession 
 of the field, and therefore must be at least as old as the 
 middle of the second century. By the time of S. Augustine 
 it had thoroughly established itself. It appears to have con- 
 tained all the books included in our Canon, except the 
 
 ^ See Roberts's Discussions on the New Testament, and Westcott on 
 the Canon of the New Test., pp. 215, 216; and Milman, Hist. Latin 
 Christianity, vol. i. pp. 25-35 (^fourth ed.). 
 
Century. 
 
 Denominated by 
 Tischendorf. 
 
 iv. 
 
 a 
 
 iv. or V. 
 
 ... b 
 
 xi. 
 
 c 
 
 iv. or V. 
 
 ... h 
 
 V. or vi. 
 
 I 
 
 iv. or V. 
 
 ... k 
 
 vi. 
 
 m 
 
 THE NEW TESTAMENT. 47 
 
 Epistle to the Hebrews, with those of S. James and 2 S. Peter. 
 The Gospels are placed in the order S. Matthew, S. John, 
 S. Luke, and S. Mark. The best codices of it are — 
 
 Name of Codex. 
 
 Cod. Vercellensis 
 
 „ Veronensis 
 
 ,, Colbertinus 
 
 „ Claromontanus 
 
 ,, Vindobonensis 
 
 „ Bobbiensis 
 
 „ Mai's Speculum 
 
 The roughness of the phraseology of this African version 
 was apparently displeasing to the inhabitants of North Italy, 
 where it obtained a footing; and in the fourth century a 
 recension was made, which is clearer and more correct ; but it 
 already shows marks that characterize a later text. It pre- 
 sents the text in an intermediate state between that of the 
 earliest and that of the late Greek MSS. This is the version 
 known as Itala. The best codex is Cod. Brixianus (/) 
 (sixth century). 
 
 There seem to be grounds for believing that Britain can 
 lay claim to a recension, independent of, but not anterior to, 
 S. Jerome's revision ; of which Scotland, Ireland, North 
 England, and Wales, enjoyed the fruit^. There was also 
 perhaps a Galilean recension, nearly allied to the British in 
 its readings : but there is much yet to be done in elucidating 
 these and similar points. 
 
 By the end of the fourth century there was so much varia- 
 tion in the existing texts, that a formal revision seemed 
 necessary ; and S. Jerome was requested by Pope Damasus 
 
 ^ For some interesting notices, see Haddan and Stubbs's Councils and 
 Ecclesiastical Documents relating to Great Britain and Ireland, vol. i. 
 App. G, pp. 170-174. 
 
48 THE CHIEF VERSIONS OF 
 
 to undertake the task. The greater part of S. Jerome's 
 critical labours were spent upon the Old Testament ; it is 
 therefore beyond the scope of the present work to say much 
 about them. In the course of these labours his views on 
 several points connected with revision, and among others on 
 the amount of change necessary to be introduced, underwent 
 considerable modification. The emendation of the New 
 Testament occupied his attention first, and is not therefore 
 the result of his most mature judgment. In order to avoid 
 offending the prejudices of persons accustomed to an estab- 
 lished phraseology he made as few alterations as possible ; 
 only correcting obvious errors, and somewhat improving the 
 latinity. The traces of his work are most frequent in the 
 Gospels, which indeed, from being the most used part of the 
 New Testament, were most often transcribed, and had there- 
 fore suffered most deterioration. The rest of the New 
 Testament he only revised cursorily. 
 
 Such a work as the revision of an estabhshed Bible is sure 
 not to be popular. Two centuries elapsed before S. Jerome's 
 revision came generally into use. Meanwhile the old copies 
 of the Vetus Latina and its variations were current, the text 
 still suffering gradually in the process of transcription. The 
 new Vulgate of S. Jerome was not free from the same 
 chances ; and the consequence was again so much uncer- 
 tainty, that in the eighth century further revision was neces- 
 sary. This was attempted by Alcuin at Charlemagne's 
 desire. He seems to have used good Latin texts for his 
 work, but without having any recourse to Greek MSS. 
 During several succeeding centuries there were more 
 isolated attempts at revision; and lists of corrections 
 were drawn up at different times. The last authoritative 
 revisions were that of Sixtus V, published in 1590; and 
 the second, which was put forth two years later, rendered 
 necessary by the arbitrary corrections introduced into 
 
THE NEW TESTAMENT. 49 
 
 the former Sixtine edition by that Pope himself, and which 
 is known as the Clementine Vulgate, from having been issued 
 under Pope Clement VIII. This last is the modern * autho- 
 rized' Vulgate. It is therefore a somewhat composite work 
 in respect of its readings, but is substantially S. Jerome's 
 revision. 
 
 For our present purpose we have only to do with the 
 earlier stages of this version. From what has been said it 
 will be seen that the critical evidence of the Latin versions 
 is twofold: viz. (i) the Corrections of S. Jerome, which being 
 of the fourth century give us an independent witness of 
 nearly the same age as our oldest existing Greek j\ISS. ; 
 (2) the readings of the Vetiis Lathia, w^hich witness to a 
 still earlier text, not indeed free from corruption, but valu- 
 able from its antiquity, and because (as has been already 
 pointed out) the very corruptions follow different courses 
 from those of the Greek codices, and therefore can often be 
 made to give useful information. 
 
 From these facts such critical principles as the following 
 may be deduced. 
 
 1. If the Vetus Latina and the Vulgate^ do not verbally 
 
 accord, but support the same Greek reading, their testi- 
 mony is strongly corroborative. 
 
 2. If the Vetiis Latifia and Vulgate accord verbally, their 
 
 testimony is not necessarily that of two distinct wit- 
 nesses, and therefore is not necessarily corroborative. 
 For there are not a few places where S. Jerome left 
 errors untouched. 
 
 3. Any reading opposed to the combined testimony of our 
 
 oldest Greek MSS. and the Vulgate must have arisen 
 
 c AVhere the name ' Vulgate ' is used simply thus, without any further 
 designation, it should be understood that S. Jerome's revision of the 
 Latin text is intended. 
 
50 THE CHIEF VERSIONS OF 
 
 subsequently to the fourth century; or at least have 
 been confined within a very narrow range previously. 
 (See Smith's Diet, of Bible, vol. iii. p. 17 14.) 
 4. The Ve/us Laiina and S. Jerome's Vulgate often combine 
 in a reading with other ancient witnesses against the 
 mass of later evidence ; and that, where the reading has 
 been altered in the later Latin texts to suit the later 
 Greek MSS. On the other hand, where the two com- 
 bine in giving a reading that is certainly erroneous, the 
 Eastern witnesses commonly desert them. This prin- 
 ciple is illustrated by the example given at p. 62. 
 (Smith's Diet, as before, p. 17 15, § 38.) 
 
 The two most accurate codices of S. Jerome's Vulgate 
 are Cod. Amiatinus and Cod. Fuldensis. The text of the 
 former is reprinted in Dr. Tregelles' edition of the Greek Tes- 
 tament ; the latter in Lachmann's larger edition. Both are of 
 the sixth century. 
 
 [The student should read carefully the article ' Vulgate ' in 
 Smith's Dictionary of the Bible, by Professor Westcott, the 
 foremost English scholar in this branch of learning.] 
 
 § 3. The Syriac Versions. 
 
 Our distinct knowledge of the existence of early Latin 
 versions prior to S. Jerome's revision in the fourth century, is 
 of the highest importance : for the general aspect and textual 
 characteristics of two of the Syriac versions, the Curetonian 
 and the Peshito, so closely resemble those of the Vetus Latina 
 and Vulgate respectively, that the suggestion is very obvious 
 that they bear a similar mutual relation to each other ; though 
 we de not know this as a historical fact. 
 
 All that we possess of the version called the Curetonian 
 Syriac is contained in a single manuscript of the fifth century, 
 
THE NEW TESTAMENT. jl 
 
 brought by Archdeacon Tattam in 1842 from one of the 
 Nitrian monasteries. It consists of fragments of the four 
 Gospels. It takes its name from having been brought into 
 notice by Dr. Cureton, who observed that its text differed 
 from that of the ordinary Peshito, and published it in 1858. 
 The text is ruder than that of the Peshito, and has many 
 interpolations, sometimes in common with Cod. D, some- 
 times unsupported by other authority; but in many charac- 
 teristic readings it is in remarkable agreement with the oldest 
 witnesses. 
 
 The Peshito, which, if the reasoning in a future chapter 
 (see pp. 70-78) is to be trusted, may be called the Syriac 
 Vulgate, appears, from its containing neither the disputed 
 Catholic Epistles <i nor the Apocalypse, to belong to a period 
 anterior to the fourth century, when those Epistles were 
 formally received into the Canon. The same is shown by 
 the fact that all the sects into which the Syrian Church was 
 separated in the fourth century alike use it. It also exhibits 
 readings of undoubtedly high antiquity. Seeing that,no other 
 Syriac text with any claim to antiquity was known to exist 
 until so recently, there is no wonder that the idea should 
 have gained currency that this is the Syriac version to which 
 Eusebius refers as existing in the second century (Eus. H. E. 
 iv. 22). But when a close examination shows undoubted 
 signs of assimilation to«a later type of text, analogous to 
 those displayed in the Cod. Brixianus of the Latin MSS., and 
 such as began to be current about the fourth century, the 
 suggestion is a natural one that the Peshito is a recension 
 of an older text, of which we have probably a specimen in 
 the Curetonian Syriac. As long ago as 1761 it had been 
 asserted by Dr. Gloucester Ridley, that the Peshito, as 
 now known, was the gradually formed product of several 
 successive revisions. This hypothesis was repeated by 
 ^ i. e. 2 S. Peter, 2 and 3 S. John, and S. Jude. 
 E 2 
 
52 THE CHIEF VERSIONS OF 
 
 * 
 Griesbach. And now, since the discovery of the Curetonian, 
 
 almost all critics of note have expressed their belief that in 
 
 that version we have a representative of an earlier state of 
 
 the text. 
 
 The Curetonian and the Peshito are not the only Syriac 
 versions. A version was made at Hierapolis in Eastern 
 Syria (a.d. 508) by Polycarp, a Chorepiscopus, at the in- 
 stance of Philoxenus, the Bishop of Hierapolis, from whom 
 it has received the name by which it is commonly known, 
 the Philoxenian. Some quotations in Syrian writers are 
 all that remain to us of this version in its original state, 
 unless perhaps one MS. of the Gospels at Florence (see 
 Westcott on the Canon, p. 210 n.), and one MS. in the 
 Bodleian, containing the Acts and all seven of the CathoHc 
 Epistles (see Home and Tregelles' Introd. pp. 278, 279), 
 exhibit it. 
 
 Enough is not known of the subject to speak with cer- 
 tainty ; but it is asserted that the text of these two MSS. 
 differs from that of the version of which we have next to 
 speak, just as an unrevised text would differ from a revised 
 text ; that is to say, they are said to have a close general 
 resemblance to this version, but to differ from it just in 
 those parts where their text might be thought capable of 
 improvement. 
 
 The version which has come down to us, sometimes cited 
 as the Philoxenian, but properly the Harclean, is a revision 
 of the Philoxenian properly so called, just mentioned. This 
 revision was made at Alexandria (a.d. 616) by Thomas of 
 Harkel, also Bishop of HierapoHs. There are several 
 codices of the Gospels, but only one codex containing other 
 portions of the New Testament is known to exist. This one 
 is in New College Library, Oxford. It is noticeable because it 
 includes all the seven Catholic Epistles ; but, as it is mutilated 
 at the end, it is impossible to say whether it ever contained 
 
THE NEW TESTAMENT. 53 
 
 the Apocalypse. The characteristic feature of this version is 
 its slavish adherence to the Greek : word stands over against 
 word, and particle for particle, even to the utter destruction of 
 the Syriac idiom ; so that it is difficult to conceive that it was 
 ever intended for general use. At the same time this very 
 fact gives it a special critical value ; for it becomes an ad- 
 mirable witness to the state of the current Greek text at the 
 time when it was made (seventh century) ; and it shows that 
 this text had now undergone a considerable change in its 
 character. There is another point of great value to be re- 
 marked, namely, that there are various readings from one, 
 two, and sometimes three, Greek MSS. added in the margin ; 
 the very Greek words being occasionally given. These 
 readings are clearly taken from texts of a much earlier 
 type. Hence the Harclean text and margin are always cited 
 separately. The text of the Acts has interpolations re- 
 sembling those of Cod. D. Neither the Peshito nor the 
 Harclean contain the passage S. John vii. 53 — viii. 11. 
 Ridley's Cod. Barsalibaei has the passage, but it is clearly an 
 addition by a later hand. The Curetonian MS.^is defective 
 at that part. 
 
 The so-called Jerusalem-Syriac is the only, other Syriac 
 version that is cited in critical editions. It is found in one 
 Lectionary in the Vatican Library, which, according to a 
 subscription attached to it, was written at Antioch in 1031 
 A.D. According to Tischendorf, its text resembles that of 
 the better uncials more nearly than the Peshito does: but 
 the style seems to be somewhat rude and peculiar, and Tre- 
 gelles believes it to be a mere translation of a Greek Evange- 
 listarium, and not an independent version. It contains the 
 passage S. John vii. 53-viii. 11 in a form different from that 
 of the ordinary text, and resembling that of Cod. D. 
 
 Besides these there is the Karkaphensian version, a dia- 
 lectic variety of Syriac. It is known only through one 
 
54 THE CHIEF VERSIONS OF 
 
 MS. in the Vatican, containing the same Canon as the 
 Peshito, which it is also said to resemble on the whole in 
 its text. 
 
 § 4. Tke Egyptian Versions. 
 
 There are three versions now in existence in different 
 Egyptian dialects; the Thebaic (or Sahidic), in the dialect 
 of Upper Egypt ; the Memphitic (or Coptic), in the dialect of 
 Lower Egypt; and the Basmuric, the locality of which is 
 uncertain. The two first only have an independent critical 
 value. The origin of neither of them is sufficiently well 
 known, nor are the materials in a thoroughly satisfactory state 
 as yet; a critical knowledge of these dialects being very 
 rare. Still, the value, even of our partial information, is very 
 great, and gives additional evidence as to the text current in 
 the early centuries. The period when they were made ad- 
 mits of being determined with fair probability. We know 
 that by the time of Diocletian's persecution (303-313 a.d.) 
 the native Christians in the Thebaid were very numerous 
 (see Euseb. H. E. viii. 9) ; and it is only probable they would 
 have had a version of the Scriptures in the vernacular. The 
 character of the text of the Thebaic (which is the real crite- 
 rion), judged by tests that will be explained further on, 
 agrees very well with this probability : that is to say, it pre- 
 sents features which might well belong to a version made 
 in the second century. Though often Alexandrine, it has 
 ,many of the peculiarities of the Vetus Latina. The Mem- 
 phitic on the other hand exhibits a third-century Alexan- 
 drine text, and is more polished in style. This is the ver- 
 sion which the Copts (Egyptian Christians) still use as their 
 authorized version in their public worship, notwithstanding 
 that Arabic has become the vernacular language. Here 
 again the parallel of the Vetus Latifia and the Vulgate sug- 
 
THE NEW TESTAMENT. ^^ 
 
 gests itself; and it is a fair subject for inquiry whether the 
 same process of revision of an earlier rough translation, with 
 the help of a collation of the Greek text current at the time, 
 may not be traced. On the whole, the Memphitic is assigned 
 to a period probably not later than the close of the third 
 century; while the Thebaic may date from the end of the 
 second century. We possess MSS. of the fifth century, if 
 not earlier. 
 
 The third version, the Basmuric, appears to be only a 
 dialectical version of the Thebaic. It has some use how- 
 ever, from containing a small portion of the New Testament 
 not found among the fragments of which the Thebaic, as 
 now extant, consists. 
 
 § 5. The Gothic Version. 
 
 The Gothic version was made by Ulfilas, who was bishop 
 of the Goths 348-388 a. d. It is therefore undoubt- 
 edly of the fourth century. It must have been extensively 
 circulated, since traces of its use both by Eastern and 
 Western Goths have been found in Italy and Spain. That 
 it was translated from Greek manuscripts is certain, says 
 Tregelles, from the manner in which the Greek construc- 
 tions and the forms of compound words are imitated. As 
 to the character of the text, Mr. Scrivener's judgment is that 
 it approaches nearer to the received text than the Egyptian 
 versions do ; which same fact Tregelles describes when he 
 declares it to be what he terms ' the transition text ' of the 
 fourth century, such as is found in the Cod. Brixianus of the 
 revised Itala. Seven or eight codices are known, containing 
 parts of all the books of the New Testament, except the Acts, 
 Epistle tQ the Hebrews, Catholic Epistles, and Apocalypse. 
 The order of the Gospels is the same as that which has been 
 stated above to be that of the Vetus Latina (p. 47), and which 
 
^6 THE CHIEF VERSIONS OF 
 
 is also found in Cod. D. The chief MS. of it is the cele- 
 brated Codex Argenteus at Upsala, inscribed in silver letters, 
 with gold initials, on purple vellum. 
 
 § 6. The JEthiopic Version. 
 
 The ^thiopic version has not yet been edited with critical 
 care. We do not know its date : but we do know that 
 Christianity was introduced into JSthiopia in the fourth 
 century. It might therefore date from about that time. 
 The curious mistranslations that occur in it (see p. 45) 
 shew that it was made from the Greek, but evidently not 
 by persons to whom Greek was familiar. The Gospels and 
 Epistles seem to have been the work of different handst- 
 and the idea of a revision of the text by different Greek 
 MSS. from those from which it was first translated is said to 
 be suggested in this case too by the phenomena which it pre- 
 sents : viz. by the mixture of the Byzantine with Alexandrine 
 readings. An edition ' by native editors ' was printed in 
 Rome and published as early as 1548-9. 
 
 § 7. The Armenian Version. 
 
 The history of the origin of the Armenian version is 
 known. It was made from Greek MSS. about the middle 
 of the fifth century. The materials in this department have 
 been rendered accessible in a great measure by the labours 
 of Dr. Rieu of the British Museum, who carefully collated 
 the texts and translated the various readings for Dr. Tre- 
 gelles. It has been said that this version was made from 
 the Peshito ; but it would seem that, though some readings 
 have been introduced from this source, there is no suffi- 
 ciently general characteristic resemblance to warrant this 
 inference. Dr. Tregelles' verdict on the whole is that the 
 readings of the Armenian MSS. differ so greatly, that they 
 
THE NEW TESTAMENT. 57 
 
 • 
 
 must have been derived from a comparison of Greek MSS. 
 of different ages ; and that therefore in this case too a revi- 
 sion probably took place at some time or other. 
 
 These are the only versions that possess any considerable 
 critical value. [For more detailed information the reader 
 should consult the article ' Versions (Ancient)' in Dr. Smith's 
 Dictionary of the Bible.] 
 
CHAPTER V. 
 
 ON PATRISTIC QUOTATIONS. 
 
 The materials in this, our third branch of evidence, are 
 in a far less satisfactory state than in the case either of the 
 Greek MSS., or of the Versions. This is chiefly owing to 
 the fact that so little real critical care has been yet spent in 
 editing the writings of the Fathers. Until this has been 
 done it will be impossible to place implicit confidence in the 
 alleged testimony of a Father to any particular reading, if it 
 be inferred merely from the appearance of that reading in 
 the common editions of his wTitings. It has happened not 
 seldom that transcribers, doubtless believing they were doing 
 a good work, have altered the words of a quotation in the 
 work they were transcribing, to the more familiar reading of 
 the commonly received text of their time. This will account 
 for some of the instances where an author appears to quote 
 the same passage of the New Testament with different read- 
 ings in different parts of his writings ; as is especially the 
 case with a voluminous writer like S. Chrysostom. 
 
 That transcribers did thus alter the readings is abundantly 
 proved in many instances by direct evidence ; as when ex- 
 tracts are preserved from the patristic writings in some Catena 
 or Commentary which gives the quotation in what w^e have 
 otherwise reason to believe is the older form, while the 
 
ON PATRISTIC QUOTATIONS. 59 
 
 recently transcribed ]\ISS. of their works present us with the 
 reading which we find in the Textus Receptus. Sometimes 
 again the embedded quotation, as it appears in the common 
 editions, is so manifestly inconsistent with the context, as to 
 make it clear that the author could never have given it 
 thus. 
 
 An instance of this latter case is found in a comment of 
 Eusebius on S. Matt, i, 24. The original reading of the 
 Evangelist was deiyfiariaai, and the manifest drift of the 
 comment is to account for his having used the simple verb, 
 and not the compound napaSeiynaTio-ai. But the later reading 
 is TvapabuyixaTiaai ; and the transcriber of the comment, in a 
 blundering attempt (apparently) to reconcile the comment 
 with the reading which he was familiar with, has transposed 
 the verb and its compound in such a way as to make abso- 
 lute nonsense ^. 
 
 In respect of the readings 05 and Ocos (i Tim. iii. 16) the 
 citation from S. Chrysostom preserved in Cramer's Catena 
 on the passage shows that Beo? is an interpolation, though 
 S. Chrysostom's authority has been quoted for the reading 
 deos : and that S. Cyril of Alexandria read os is proved, not 
 only by the context, but by an express marginal note in 
 several of the MSS. (see Tregelles on the Printed Text of 
 the New Testament, p. 227), viz. 6 eV aylois KvpiWos ev rw i[d' 
 
 KecPaXuLco T(ou a^oXtcov (f)r]a\v os ((pavepa^drj iv aapKL. 
 
 * The comment runs as follows : — eS yovv fioi Kal rb p.^ 9i\€iv avTTjv 
 SfLypariaai dpijaOai SokcT vtto tov ILvayyiXiarov' ov yap ecprjaev prj 6e\(iv 
 avTTjv BdyfiaTiaai, dWa ' p-q ■napa^dy/j.aTiaai OeXcuv' iToWrjs ovaijs (v 
 TovTOLS Siafpopds' dis yap ov Tavruv arjpaivei to ypdipai Kal to irapaypdxpai, 
 Kox Tu Xoy'iaaaOai kol rb irapaXoyiaaaOai, Kal xp-q^laai kol vapaxpr^cpiaai, 
 oiiTOJS ovTi TO SciypaTiaai Kal TrapahnypaTiaaf to pkv yap Trapadeiyfj-aTi- 
 oai T^v em KaK(f -npa^avTi irdyras (pavepcoaiv tc Kal SiaPoXrjv vvo^dWd 
 voHi^- u Toivvv ' 'Icoarjtp SiKaios Sjv, Kal pf) 6i\ojv avrrju irapaSeiypuiTiaai ' 
 TovTtariv els (pavtpbv rots irdaiv dyayeiv, ' e^ovX-qOr] XdOpa dnoKvaai 
 avTqv.' Cramer's Catena, vol. i. p. 12. 
 
6o ON PATRISTIC QUOTATIONS. 
 
 By way of illustrating the extent of this field of evidence, it 
 has been said that if every copy of the Greek Testament, 
 manuscript and printed, had perished, and only the Patristic 
 quotations remained, together with a copy of some one ver- 
 sion to serve as an index whereby to arrange them, we 
 should be able to reconstruct the whole. In the extant 
 works of Origen alone nearly every verse of the New Testa- 
 ment is quoted : some of them several times. 
 
 These remarks apply to verhativi citations. But it often 
 happens that the Patristic writers quote the New Testament 
 writings in a less exact way, by interweaving the words with 
 their own, and altering the structure of the sentences to suit 
 their own. In the writings of the Apostolic Fathers almost 
 all the quotations are thus introduced. Such quotations are 
 free from the chance of variation just mentioned : on the 
 other hand, they will only furnish aid to the textual critic 
 where the sense of the passage may vary with the alteration, 
 and not on such delicate questions as the insertion or omis- 
 sion of particles, choice between different tenses of the same 
 verb, and so on. Evidence might be gained from them as 
 to the existence of the passage in question in copies of the 
 author's time: on the whole however, this looser kind of 
 citation, together with the still more precarious class of viere 
 allusions, is of more value in determining the contents of 
 the Canon of Scripture, than in the problems set before the 
 textual critic. 
 
 Most important of all is the help given where the writer 
 recognises different readings of a passage, and expressly 
 states that, while many MSS. have some particular reading 
 or readings, the best and most accurate have another, which 
 he gives. The existence of various readings was recognised 
 distinctly as early as the time of Marcion '°, who was charged 
 
 ^ The date of Marcion's birth is unknown. He was settled at Rome 
 and known as a heretical teacher, before a.d. 139, the date of the publi- 
 
ON PATRISTIC QUOTATIONS. 6 1 
 
 with corrupting the text of Scripture to suit his own views. 
 The labours of Origen (186-254) and Eusebius (264-340) 
 for the Greek text, and of S.Jerome (died 430^') for the 
 Latin, were distinctly and avowedly critical. From Origen's 
 writings we gather this important information, that ' //le oldest 
 MSS. and versions which we (now) have contaiti just the saine 
 variety of text as existed in the third century! (Home and 
 Tregelles' Introd. to N. T. p. 43.) Hence it is hardly too much 
 to say that, with the exception of oral tradition, the weight of 
 which on such questions would not be great, our present 
 materials for eliciting the true text are at least as full as those 
 which were in Origen's possession. This may well serve to 
 give confidence in our labours. 
 
 The value of even the most definite Patristic citation is 
 only corroborative. Standing by itself, any such citation 
 might mean no more than that the writer found the passage 
 in his own copy, or in those examined by him, in the form 
 in which he quotes it. The moment however it is found to 
 be supported by other good evidence, the writer's authority 
 may become of immense importance. Perhaps the best 
 illustration of what is meant will be found in the discussion 
 of a reading, the determination of which turns chiefly upon 
 the statement of S. Irenaeus, a writer of the second century. 
 The passage in question is S. Matt. i. 18; the point to be 
 determined, is whether 'It^o-oO should stand in the text before 
 XpKTTov or not. It is found in the text of perhaps every 
 existing Greek manuscript ^ : on the other hand, S. Irenaeus 
 
 cation of Justin Martyr's First Apology. Nor is the time of his death 
 more certain. Tillemont only says vaguely it was after a.d. 176. Mr. 
 Clinton thinks he was alive as late as 194. See Smith's Diet. Biog., Art. 
 ' Marcion.' Robertson's Ch. Hist. vol. i. pp. 43, 44. 
 
 c The date of S. Jerome's birth is equally uncertain ; different years, 
 ranging from 329 to 345, have been assigned. 
 
 ** There seems a slight doubt about one MS., Cod. 71. Mr. Scrivener 
 
62 ON PATRISTIC QUOTATIONS. 
 
 expressly asserts that it should not be there, and gives a 
 reason for his statement. His words are, as given by his 
 Latin translator (for the Greek original does not exist), 
 ' ceterum potuerat dicere Matthceus Jesu vero generatio sic 
 erat, sed prcevidens Spiritus S. depravatores, et prcBfnumens 
 contra fraudulentiam eorum, per MatthcEimi ait Christi autem 
 generatio sic erat/ The question is not by any means to 
 be so summarily disposed of as Mr. Scrivener represents 
 (Introd. p. 420). In weighing the evidence on both sides, 
 we must anticipate some statements, the reasons for which 
 have not yet been given, but which we shall attempt to sub- 
 stantiate further on. 
 
 For the reading rov 5e 'Ij/o-oO there is no important evi- 
 dence : it may therefore be at once dismissed. 
 
 For the reading 'l^^o-oO Xpiarov there are all existing Greek 
 MSS. (except possibly No. 71, which, according to Tischen- 
 dorf, reads Xpiarov ; Cod. B, which has Xptarov 'l^o-oO ; and 
 probably D ; for though the Greek text is wanting at this 
 place, the Latin version, which is generally a slavish inter- 
 pretation of the Greek, reads Christi). With them are the 
 two Egyptian versions, the Peshito and Harclean Syriac, the 
 Armenian and the ^thiopic ; and of Patristic writers Origen, 
 Eusebius, and others of later date. 
 
 On the other side, for the reading Xpiarov are all the Latin 
 versions, including the Vetus Latina ; the Curetonian Syriac ; 
 and S. Irenseus expressly, as we have seen, with later 
 Fathers. 
 
 At first sight, no doubt, there seems an overwhelming array 
 of evidence for the reading 'It/o-oi) Xpiarov : but our estimate 
 of it will probably be modified when we take the following 
 considerations into account, viz. — 
 
 does not note that this MS. omits the word (Introd. p. 420 n.) ; but 
 Tischendorf repeats his statement that it does, in his eighth edition. 
 Mr, Scrivener is generally very accurate in these minute matters of fact. 
 
ON PATRISTIC QUOTATIONS. 6^ 
 
 I. According to S. Irenseus' express statement, Greek MSS. 
 were known to him with the reading Xpiarrov. 
 
 :. All the evidence for this reading is undoubtedly of the 
 second century; while the opposing witnesses, except 
 perhaps the Thebaic version, are all later. At any rate, 
 it is clear that Xpto-roO was the current reading through 
 so wide an area as Syria, North Africa, and Gaul, in the 
 second century, though it may have been supplanted by 
 the other in the third. 
 
 3. In no other genuine place of the New Testament is the 
 
 collocation of words 6 'It^o-ovs Xpto-ro's- found. It is found 
 in three places of the Textus Receptus, viz. Acts viii. 37, 
 I S.John iv. 3, and Apoc. xii. 17; but every one of 
 these places is undoubtedly spurious : in fact the collo- 
 cation belongs to a later time, when the distinction 
 between 'Ir/o-oC? as the personal name, and Xpto-ros- as 
 the official title had been lost, and the two were merged 
 in one common appellation. There is a peculiar force 
 in the use of 6 Xpiaros in the passage under discussion, 
 by S. Matthew, writing for Jews. 
 
 4. It is much more in accordance with the laws of the varia- 
 
 tion of MSS. that the short reading should have been 
 changed into the longer one, than vice versa. 
 
 5. Such a variation as that above remarked in B is not sel- 
 
 dom an indication of an antecedent corruption of the 
 
 text. 
 Considerations 2, 3, and 4, are of great weight. Perhaps 
 die best conclusion is to place 'It/o-oi) in the margin, as a 
 reading supported by great authority, but as having too 
 strong arguments against it to place it in the text. Tischen- 
 dorf has changed his mind since his seventh edition, and 
 now reads 'li^o-oO Xptorou in his eighth ; probably being in- 
 fluenced by finding this the reading of N. 
 
 The age at which a writer Uved must be taken into account 
 
64 ON PATRISTIC QUOTATIONS. 
 
 in weighing his evidence ; the earlier being of course cceteris 
 paribus the more valuable. 
 
 Sometimes, as for instance when a writer is commenting 
 continuously on the words of a passage, an inference of some 
 value may be drawn against words that he omits. So too, 
 though this is still more precarious, if in a discussion on 
 some particular doctrine a passage notably bearing upon it 
 is not adduced, there is some presumption that the passage 
 was unknown to that writer. 
 
 NOTE. 
 
 A very useful work has been lately accomplished for the study of the 
 Latin Versions, by a German critic named Ronsch, already known for 
 his work, Itala wid Vtilgata. He has collected all the citations from 
 the New Testament that are made in Tertullian's writings, and arranged 
 them in their proper order ; the direct citations in one column, and the 
 indirect citations, reminiscences, &c. in a parallel column ; thus giving 
 a comprehensive view of all the passages of the New Testament for 
 which Tertullian may be cited as a witness. This he calls Das ne^ie 
 Testame?it Tertullian's. It is in fact an attempt at a critical restoration 
 of the text of the Vetus Latina, so far as it is quoted by Tertullian. 
 This is the sort of work from which fruit may now be looked for. The 
 method of the science of Textual Criticism is tolerably well under- 
 stood : the materials from which conclusions are to be drawn still admit 
 of much careful sifting and purifying, especially in the particular field 
 of Patristic evidence. 
 
CHAPTER VI. 
 
 DISCUSSION OF THE EVIDENCE DERIVED FROM THE FOREGOING 
 SOURCES. 
 
 § I . Summary of results reached so far. 
 
 The results at which we have arrived may be summed up 
 as follows: — 
 
 We have seen that it is possible to assign approximately, 
 on purely external considerations, the date at which any 
 given manuscript was transcribed. We must be careful to 
 distinguish between the date of a manuscript and the date of 
 the text which it presents. A very (comparatively) recent 
 manuscript may present a very ancient text; but the first 
 presumption is against it, and the claim must be proved for 
 every separate case. A knowledge however of the date of 
 an early jNIS. is useful as gi^jpg a point of time, before which 
 any variation occurring in that IMS. must have arisen. We 
 have two MSS. of the fourth century supplying us, as has 
 been shown already, with the evidence of other MSS. of a 
 date anterior to themselves. There are other MSS. of every 
 subsequent century down to the twelfth. 
 
 Of the existing Versions, we have seen that we know 
 a great deal about the history of the Latin, including the 
 fact that one of its forms, considerable remains of which 
 are extant, belong to the second century. We know from 
 
 F 
 
66 DISCUSSION OF THE EVIDENCE 
 
 external sources that a Syriac version existed as early as the 
 second century, and that the Peshito is at least as old as the 
 early half of the fourth. The Gothic version we know 
 to belong to the middle of the fourth; the Armenian to 
 the middle of the fifth ; the Philoxenian and Harclean 
 to the sixth and seventh centuries respectively. The dates 
 assigned in Chapter IV to the two Egyptian versions, to 
 the Curetonian Syriac, and to the ^thiopic, depend in 
 part upon internal considerations, which have yet to be 
 discussed. 
 
 The dates of all the Patristic writings useful for our pur- 
 poses are known historically. Consequently, where they 
 have been edited critically, or where we meet with explicit 
 statements in them regarding any given readings, there we 
 have distinct evidence of the recognition of the reading in 
 question in the time of the writer; and in many cases his 
 opinion upon the correctness or incorrectness of it. 
 
 Some of this sort of evidence is of the second century ; 
 a great deal of it belongs to the third and early fourth. 
 
 The question next arises, Is it possible, with this amount 
 of actually dated evidence, to construct a history of the text, 
 at any rate in broad outline ? Can we gain some general 
 notions of the direction, so to speak, in which the text was 
 modified.? If we could fix but a few clear landmarks, we 
 might be able to assign to anjj^ particular text, of otherwise 
 unknown date, its historical place in the series with some 
 degree of probabihty. We must now therefore turn our 
 attention to the characteristics of the different texts pre- 
 sented by these various authorities, and see what phsenomena 
 thev exhibit. 
 
DERIVED FROM THE FOREGOING SOURCES. 67 
 
 § 2. AfSS., though independenl of each other, are marked off 
 by general features into groups. 
 
 The first and most obvious feature is, that scarcely any 
 two known MSS. show anything like verbal agreement. 
 E (S. Paul) is thought to be a transcript of D ; and F and 
 G (S. Paul) are thought to be transcripts of some third MS. 
 now lost. But the much-talked-of unanimity of the late, is 
 just as imaginary as that of the early, authorities : that is, in 
 any strict sense of the word. 
 
 On the other hand, there may be noticed certain marked 
 features which, in the judgment of all critics, are a sufficient 
 ground for separating the existing authorities into tolerably 
 w^ell-defined groups; though not so minute as to exclude 
 individual variations in the case of each separate MS. 
 
 These special features are : — 
 a. Peculiarities of spelling ; e. g. such forms as \r]jxy^oyLaL, 
 duTLK^lxyj/ns, &c. ; V preserved unassimilated in words 
 compounded with prepositions, as o-vu^rjTelv, avv^vyos, 
 &c. ; V e(p€\<v(TTLK6v, and the final s of ovrcas, &c., pre- 
 served before consonants. The aspirate substituted 
 for the tejiuis in such cases as ecfube, e(^' eXnidi; d(p€\- 
 
 TTi^ovres, K.T.X. Such forms as TeaaepaKouTa, oXeOpevcov, 
 
 ixBe^, for Tea-aapcLKovTa, oXodpevcou, xQes, and Others. 
 
 /3. Peculiar formation of inflexions ; as p.axaipr]s, <nrhpr]i, for 
 the Attic termination in -as ; the accus. of nouns of third 
 declension, and of adjectives, ending in -v, as d(r<paKriv, 
 TTobrjprjv, &c. ; ucglcct of the augment in a few verbs 
 beginning with a diphthong; second aorists with first 
 aor. terminations, as eiTra, cneaa, rjXda, &c.; a future 
 conjunctive found in a few cases, &c., &c. 
 
 y. Peculiarities of syntax : iav for liv; tva, edv, orav, &c., with 
 the indicative ; for which, as well as for abundant 
 
 F 2 
 
68 DISCUSSION OF THE EVIDENCE 
 
 examples of the previous peculiarities, the reader is 
 referred to Tischendorf s Prolegomena, and to Winer's 
 Grammar, edited by Rev. W. F. Moulton, 1870, pp. 
 37-128. 
 d. Certain characteristic readings, including variations in the 
 order of words, omissions of words, and even of clauses, 
 which must be noticed more at length presently by 
 themselves. 
 
 § 3. Two main groups commonly recognised by critics. 
 
 That manuscript texts fall into several distinct groups, 
 marked by the presence or absence of peculiarities of this 
 sort, has been recognised by all the later critics. 
 
 For just 200 years, from the time of the Complutensian 
 editors and Erasmus down to Mill and Bentley, only the 
 variations of MSS. were noted. The Textus Receptus was 
 reprinted again and again, MSS. were collated and their 
 various readings registered, but no comparison of them was 
 attempted. Nor were editors to blame for this. Sufficient 
 materials were as yet wanting. 
 
 The next 140 years was a period in which materials were 
 more systematically amassed and classified, and various 
 theories of criticism propounded. Mill (d. 17 10) led the 
 way, pointing out the relative value of the three sources of 
 evidence, and collecting immense stores of material of each 
 kind. Bentley (d. 1742) very shortly afterwards pointed out 
 the true mode of dealing with the available evidence; but 
 ' he was in advance both of the spirit of his age and of the 
 materials at his command,' and his labours were not brought 
 to perfection. 
 
 As soon as a sufficient mass of evidence was at the dis- 
 posal of the critic to admit of comprehensive treatment, the 
 points oi similarity as well as o( divej'gence heg2in to be noticed; 
 and it was soon seen that the authorities fell into groups. 
 
DERIVED FROM THE FOREGOING SOURCES. 69 
 
 Two, three, and four groups have been distinguished by dif- 
 ferent critics; and different hypotheses propounded to ac- 
 count for their origin. All alike have recognised a broad 
 distinction between a comparatively small group, which in- 
 cludes the most ancient documents, together with some later 
 uncials and a few cursives, and the group to which the great 
 mass of more recent MSS. belong. Some critics go on to 
 subdivide one or both of these. 
 
 Bengel (d. 1752) w^ould at first have subdivided the former 
 of these and made three groups ; but finally he pronounced 
 in favour of two, which he called African and Asiatic. 
 
 Griesbach (d. 181 2) finally declared in favour of three 
 groups, which he named Alexa?idrine, Western, and Byzan- 
 tine. The two former of these would be subdivisions of 
 Bengel's 'African'; but Griesbach himself allowed that the 
 line of demarcation between them was not rigid. The 
 ' Western ' group was intended to contain D and other Grseco- 
 Latin codices, with the Latin versions. 
 
 Hug (the first edition of his Einleitimg was published in 
 1808, the second in 1826) attempted a more subtil analysis, 
 intended to exhibit the mode in which he thought the group- 
 ing had arisen. He thought he could discern four groups ; 
 one containing examples of an unrevised text, the other 
 three being derived from this by independent revisions. Two 
 of these however contain the chief part of our existing docu- 
 ments, and in the main coincide with the groups of the two- 
 fold division. Eichhorn (18 18-182 7) agreed in Hug's 
 scheme with some slight modifications. Scholz (1830- 
 1836) returned to the simpler twofold division. Tischen- 
 DORF, in his seventh edition, adopts a fourfold division in two 
 pairs, naming them Alexandrine and Latin, Asiatic and By- 
 zantine. Lachmann (d. 1851) speaks of two groups, African 
 and Byzantine. 
 
 It really seems that to go beyond the broad line of 
 
7© DISCUSSION OF THE EVIDENCE 
 
 demarcation recognised by all between the two chief groups 
 is very precarious. The gap between any subdivisions can 
 be bridged over by a number of copies exhibiting texts with 
 all intermediate degrees of resemblance. If any subordinate 
 group has a claim to be recognised, it is that called ' Western ' 
 by Griesbach, and ' Latin ' by Tischendorf. Its typical ex- 
 amples are marked by numerous interpolations. Yet the 
 existence of many of these in the Curetonian Syriac would 
 seem to show that neither the title ' Western ' nor ' Latin ' is 
 wholly appropriate. The wide acceptance of these interpo- 
 lations in the East and West has suggested the hypothesis 
 that they had their origin in the traditional oral teaching, 
 which was long the instrument alike of extending and edify- 
 ing the Church. Apart from these interpolations the docu- 
 ments of this group exhibit a text closely approaching that 
 of the Alexandrine or African group. So far then we see 
 that our documentary authorities, which all differ from each 
 other in some particulars, yet fall into at least two main 
 groups ; one characterised by the above-mentioned peculi- 
 arities, the other more conformed in diction to ordinary 
 classical Greek, and, in the passages referred to under the 
 fourth head, presenting the readings which we find in the 
 Textus Receptus. 
 
 § 4. Examples of the proofs that the earlier type of text is to he 
 found in the smaller group of witnesses. 
 
 It is next to be shown that the type of text given by the 
 first, though the smaller, of these groups is older than that 
 of the other, which is numerically so much larger. In other 
 words, the text which we should construct, if we take our 
 authorities from the first group, will be nearer to that of the 
 New Testament writers themselves than a text based upon 
 the other group. This is proved by an inductive argument, 
 depending upon a comparison of the readings of the two 
 
DERIVED FROM THE FOREGOING SOURCES. Jl 
 
 groups of MSS., in a number of passages where the true 
 reading is given by indisputably early authorities, such as 
 express citations of the early Patristic writers and versions 
 like the Vetus I.atina, whose antiquity is above dispute. 
 Here we must guard against arguing in a circle. For in- 
 stance, the Curetonian Syriac cannot be used in the proof. 
 That it can lay claim to the very highest antiquity we have 
 already once or twice implied; but this claim rests upon 
 considerations drawn from the character of its text. We 
 cannot therefore, m the first place, use it to determine what 
 type of text is the oldest. But when the characteristics of 
 the oldest type of text have been determined by other evi- 
 dence, and we find that the Curetonian Syriac, or any other 
 version, or any late Greek MSS., presents a text of this 
 type, we may assume it into the group, and henceforth 
 make use of its evidence to help in determining any doubtful 
 questions that may remain. 
 
 We proceed then to discuss a few such crucial passages 
 by way of example. [The student is strongly recommended 
 to work out the critical evidence of other passages than those 
 which follow with the help of some good critical edition of 
 the Greek Testament. A list of such passages may be found 
 in Tregelles' Account of the Printed Text, p. 133, &c.a A 
 very large number of passages are discussed, with the evi- 
 dence known at the time of the publication of the work, by 
 Mr. Green in his Course of Developed Criticism (Bagsters). 
 Apart from the power thus gained of appreciating the value 
 of the different kinds of evidence, it is only by such an exer- 
 cise that it is possible to realise the force of the argument 
 for preferring the text of the few older to that of the many 
 late witnesses.] 
 
 * Some of his instances are the various readings in S. Matt, xviii. 35 ; 
 S. Mark iii. 29; S. Luke v-iii. 9, and 20; S. John v. 16; vi. 51 ; ix. 8, 
 and 26 ; x. 33 ; Acts xv. 24 ; Rom, v. i ; xiv. 9 ; i Cor. ii. 4 ; vii. 5, &c. 
 
72 DISCUSSION OF THE EVIDENCE 
 
 a. Quite a test-passage is S. Matt. xix. 17. The Textus 
 
 ReceptUS reads tI fxe \eyeis dya66v ; ouSejy dya66s el jxr] eh, 
 
 6 Qeos, which is the unquestioned reading in the parallel 
 passages S. Mark x. 18 ; S. Luke xviii. 19. The alter- 
 native reading in S. Matthew is n' fxe eparas nepl tov 
 dyaOov ; eh eariv 6 dyados : the very existence of which, 
 backed by any good suppo7't, would be a strong primd 
 facie argument for its genuineness, on the principle 
 laid down at p. 92. Now let us see what the evidence 
 is. Not to go into extreme minutise, it will be found 
 that the reading of the Textus Receptus is supported 
 by C of the old MSS. ; by the later uncials and the 
 mass of cursives : by f and q of the Latin versions ; 
 by the Peshito and Harclean {texi) of the Syriac ver- 
 sions ; and by the Thebaic : also by Hilary, Optatus, 
 Ambrose, and Chrysostom, with the main body of the 
 later Patristic writers. For the other reading, the first 
 clause is supported by N, B, [D] b, L, i, 22: by nine 
 codices of the Vetus Latina, and the Vulgate ; by 
 the Curetonian and Jerusalem Syriac ; the Memphitic, 
 and the Armenian versions : by Eusebius, Jerome, and 
 others of the Fathers : Origen and S. Augustine men- 
 tion it expressly in these words : 6 [liv oZv Mar^aToy 
 
 a)ff Trept dyaBov epyov epcoTTjOeuros tov (rcoTrjpos iv ra, t'l 
 dyaObv Troirjo-o) ; dveypayj/^eu. 6 Se Mdp<os Koi Aovkus (pao-l 
 TOV crcoTTipa elprjKevat tl p.e Xeyeis dyaOov ; ovheXs dyados el 
 pr) els, 6 Qeos : and de illo divite .... potest videri dis- 
 tare aliquid, quod secundum Matthceum dicitur, Quid me 
 inter rogas de bono ? secundum illos autem (sc. S. Mark 
 and S. Luke) Quid me dicis bonum ? . . . . &c. The tov 
 is omitted by D. 
 
 ^ When an authority is quoted in brackets, it is implied that its 
 evidence is only partial ; as here, D, by the omission of tov, is not in 
 strict accordance with x and B. 
 
DERIVED FROM THE FOREGOING SOURCES. 73 
 
 The second clause is supported by N, B, [D], L, (i), 22 ; 
 by seven codices of the Vetus Latina, and the Vulgate ; 
 by the Curetonian and Jerusalem Syriac ; the Memphitic, 
 and Armenian. 6 is omitted by D and i. This clause 
 is not so expressly supported by any Patristic writer as 
 the first ; but it occurs very distinctly in Irenaeus, though 
 in combination with the Textus Receptus version of the 
 first clause. Several authorities give a mixed edition of 
 the passage, one clause in accordance with one form, 
 the other clause in accordance with the other form, as 
 the Harclean Syriac {inargiii), the ^thiopic, two codices 
 of the Vetus Latina ; Eusebius, Irenaeus, and Justin 
 jMartyr; while the cursive ]MS. 251 gives both the forms 
 in full, that of the Textus Receptus first, and then the 
 other. Such evidence as this points unmistakably to the 
 existence of an antecedent variation. The evidence of 
 Origen and S. Augustine is express as to a difference 
 between S. Matthew's account and those of S. Mark 
 and S. Luke. Among those authorities which present a 
 different form of the passage in S. IMatthew from that 
 in the parallel passages are included nearly all the very 
 earliest. The reading here given by ^? and B seems 
 to have been current before the time of Irenseus and 
 Justin ?klartyr, and before the formation of the Vetus 
 Latina: that is to say, we are carried back at least 
 to the beginning of the second century ; which is an 
 earlier date than can be claimed by any authority for 
 the common reading of this passage. 
 
 Further, it must be remembered that it is in accordance 
 with the observed tendency of copyists to alter one 
 passage into conformity with another parallel passage. 
 It is not their habit to introduce discrepancies. 
 
 And, once more, let us consider that on no intelligible 
 principle can it be assumed that the passage has been 
 
74 DISCUSSION OF THE EVIDENCE 
 
 tampered with on theological grbimds ; for then why- 
 were the two parallel passages left, as they are, without 
 any suspicion of a variation ? 
 On the whole, we must conclude that in this passage those 
 authorities which differ from the Textus Receptus give 
 us the earlier and truer text. 
 jS. In S. Matt. xv. 8 the Textus Receptus reads [iyy iCet ^ot] 
 
 6 Xaus ovTos [to) (TTOfjLaTL avTtov fcai] Tois p^eiXecri fie Ti[xa, which 
 
 is scarcely varied from the LXX. of Is. xxix. 13. This 
 is the reading of C and most of the later uncials and 
 of the mass of cursives; of / alone among the Latin 
 versions, and of the Harclean Syriac. On the other 
 hand, the words which are inclosed in square brackets 
 are omitted by N, B, D, L, 33, 124; by all the Latin 
 versions (except/"), the Curetonian Syriac and Peshito, 
 and by the Memphitic, Armenian, and ^Ethiopic. The 
 Patristic evidence is for the omission; Origen saying 
 expressly, after quoting the passage in full from Isaiah, 
 
 Ka\ TrpoeiTTO/xeV ye on ovk avrals Xe^eaLV dveypa-^JACv 6 Mardaios 
 
 TO 7rpO(j)T]TLK6v. 
 
 Thus here again we find the same smaller group of 
 MSS. presenting that reading for which there is 
 express authority in an early writer, and very early 
 support from the versions. Besides, it is a well- 
 known tendency of the copyists to supply defects in 
 quotations. 
 y. The case is as nearly as possible the same in S. Matt. xx. 
 
 2 2, bvvacrOe ineiv to 7roTi]pLov o eyco peWco Tviveiv, [kol to pdir- 
 
 Tio-fia b eyoy ^anTLCofxai (3a7TTi(T6rjvaL]. Here again Origen 
 expressly says that the latter clause is in S. Mark, and not 
 in S. Matthew. In S. Mark all our authorities give 
 it without variation: in S. Matthew it is omitted by 
 N, B, D, L, Z, I, 22 ; by almost all the codices of the 
 Vetus Latina, the Curetonian Syriac, Memphitic, Thebaic, 
 
DERIVED FROM THE FOREGOING SOURCES. 75 
 
 and ^thiopic. It is found in C, ^vilh the later MSS., 
 uncials and cursives ; in the Peshito and Harclean 
 Syriac, and in the Armenian, with /, h, and q of the 
 Vetus Latina. 
 The same considerations as in the previous case will 
 govern our choice of the reading, about which there is 
 no room for doubt. 
 
 fi. Even readings that are undoubtedly erroneous may help 
 to show the antiquity of the documents in which they 
 occur : e.g. after S. Matt. xx. 28, there is found in D, 
 in the Curetonian, and one codex of the Harclean 
 {margin) Syriac, and in almost all the codices of the 
 Vetus Latina, but in no other Greek MS. or early ver- 
 sion, an extensive interpolation, which may be seen in 
 Scrivener's Introduction, p. 425. There are numberless 
 variations in these authorities, and S. Jerome has re- 
 jected it. There is no doubt that it is an interpolation ; 
 but since it was certainly current in the second cen- 
 tury, and rejected in the fourth, the text exhibited by 
 any document containing it would probably be very 
 ancient. 
 
 f. A very instructive passage to examine is S. Luke xi. 2-4, 
 containing that Evangelist's" account of the Lord's 
 Prayer. As read in a modern critical edition of the 
 Greek Testament, it will be found to want three clauses, 
 which occur in the form as given by S. Matthew : viz. 
 
 riyLoav 6 iv Tois ovpavols, y(vr}6f]T(i) to 6e\rj]id aov cos iv ovpava 
 Koi eVt Trjs yrjs, and dXka pvcrai rjfxds diro tov TTOvqpov. 
 
 For the insertion of the first of these clauses entire the 
 authorities are A, C, D, with about fifteen other uncials, 
 and most of the cursives ; b, e, /, l, q of the Vetus 
 Latina; all the Syriac versions, including the Cureto- 
 nian ; the IMemphitic, and the ^thiopic. L, one 
 cursive, one early copy of the Vulgate, and the Armenian 
 
DISCUSSION OF THE EVIDENCE 
 
 version, support rjix^v only. Four copies of the Vetus 
 Latina give sancte instead of nosier. No. 33 (cursive) 
 seems to favour 6 eV roi^ ovpavo7s, but not r][xa>v. For the 
 omission entire are N, B, i, 22, 57, 130, and 346; with 
 all the chief MSS. of the Vulgate but two ; as well 
 as the express testimony of Origen, and of a scholion in 
 some of the MSS. Origen's words are, exovo-i 8e al 
 
 Xe^eis Tov {JLeu Mardaiov .... Udrep rjfxau 6 iv . . . . tov Se 
 KovKO. ovTcos, ndrep dytao-drjTco .... K.r.X. Tertullian's 
 
 testimony seems also to favour the omission. Now 
 strong as the evidence for the full form seems at first 
 sight, it is much weakened, first by the variations also 
 attested, and then by the deliberate rejection of the 
 clause from the Latin in S. Jerome's Vulgate. Against 
 this and the express assertion of Origen it cannot 
 stand ; especially when we remember that the tendency 
 of copyists to supply supposed deficiencies would 
 be likely to be stronger than ever here, where the 
 longer form was so familiar from constant public and 
 private use. 
 
 We then pass to the clause yevrjBrjTco t6 deXrjfxd o-ov COS 
 
 iv ovpava kqi eVi Ttjs y^s, which is wanting in B, L, i, 
 2 2, 130, 346; ^^ of the Vetus Latina; the Vulgate, 
 and Curetonian Syriac. There is also most express 
 testimony of Origen, Tertullian, and S. Augustine for 
 the omission in S. Luke; Origen and S. Augustine 
 drawing attention to the contrast between his form and 
 S.Matthew's. The presence of the clause is attested 
 by N, A, C, D, &c. ; by the chief codices of the Vetus 
 Latina; by the Peshito and Harclean Syriac, and the 
 Memphitic. There are slight variations here too be- 
 tween the different witnesses; and the same marked 
 disagreement between the Vetus Latina and the Vul- 
 gate of S. Jerome, In fact, on the whole the same 
 
DERIVED FROM THE FOREGOING SOURCES. 77 
 
 remarks apply here as in the previous case. It will be 
 noticed that N and the Curetonian Syriac have changed 
 sides here; giving an useful illustration of a remark 
 made above, that the true text must not be looked for 
 in any one document, but must be elicited by a careful 
 comparison of all. 
 
 For the third clause, aWa pvaai rjfias dno Tov Tzovrjpov, the 
 
 authorities are (N^*), A, C, D, &c. ; seven codices of 
 the Vetus Latina ; the three Syriac versions ; the 
 Memphitic, and the ^Ethiopic : ranged against which 
 are ^5*, B, L, i, 22, 57, and six other cursives; the 
 Vulgate, and Armenian ; with the express testimony 
 of Origen, Cyril, and S. Augustine, and apparently 
 that of Tertullian. Here again the verdict of the re- 
 cent critical editors is in favour of omitting the 
 clause. 
 It is pertinent to observe that an omission, so strongly 
 attested as this is, of three important clauses, in a 
 formulary so well known and cherished as the Lord's 
 Prayer, is utterly inexplicable on the hypothesis that 
 S. Matthew's form is the only genuine one. We can 
 easily understand the importation of the clauses, either 
 from another Gospel or from a well-known Hturgical 
 formula, into a less familiar and seemingly abridged 
 form, like that of S. Luke; but neither accident nor 
 intention can adequately account for such clear evi- 
 dence as there is in favour of so large an omission, if 
 S. Luke's Gospel had originally contained the clauses in 
 question. 
 
 These five instances are samples of a vast number*^ of 
 others, by means of which it is shown that the true text is on 
 
 '-■ Dr. Tregelles estimates that there are between two and three 
 xhousand. (On the Printed Text, p. 148, note.) 
 
78 DISCUSSION OF THE EVIDENCE 
 
 the whole to be sought for in the smaller of the two groups 
 of MSS. It must be borne in mind however that they are 
 but samples, and that the value of the induction rests upon 
 the number of instances discussed. A conclusion drawn 
 from a few might easily be erroneous. For instance, it 
 might be thought from the examples above given that C is 
 commonly opposed to N and B, and in harmony with the 
 Textus Receptus ; whereas on the whole the contrary is true. 
 
 Dr. Tregelles sums up the results of his investigation as 
 follows (Account of the Printed Text, p. 148) : — 
 
 ' Readings whose antiquity is proved apart from MSS. 
 are found in repeated instances in a _/ew of the extant 
 copies.' 
 
 'These few MSS., the text of which is thus proved to be 
 ancient, include some (and often several) of the oldest MSS. 
 extant.' 
 
 ' In some cases the attested reading is found in but one 
 or two MSS., but those of the most ancient class.' 
 
 ' And as certain MSS. are found, by a process of inductive 
 proof, to contain an ancient text, their character as witnesses 
 must be considered to be so estabHshed, that in other places 
 their testimony deserves peculiar weight.' 
 
 The same conclusions mutatis mutandis will hold of course 
 with respect to the text exhibited by those versions whose 
 dates are not known independently. 
 
 If this conclusion be not true, and if the text given by the 
 larger group of MSS. be the purer of the two, we are met by 
 a very remarkable phsenomenon. For the true text will be 
 one of which no example is found till after a lapse of several 
 centuries from its origin ; during which centuries however 
 there is tolerably abundant evidence of the (so called) cor- 
 rupted text. A development-theory of a true text is out of 
 the question. Probably no one would assert that the text 
 gradually cleared itself from errors as time advanced. So 
 
DERIVED FROM THE FOREGOING SOURCES. 79 
 
 then we have to believe that, though the original text was in 
 existence previously to and alongside of the later corrupt 
 text, the early versions were made from the corrupted form, 
 and the early writers all quote from the corrupt form ; while 
 by a singular ill-fortune no very early MS. of the true text 
 survives, though we have several of the corrupt form. It is 
 easier to accept the other hypothesis. Further confirmation 
 of our choice of alternatives is supplied by the next point 
 which we have to notice. 
 
 § 5. An order traceable among the various documentary 
 witnesses. 
 
 Amid the variations of different witnesses a certain order 
 seems traceable. The links between the normal types of 
 the two main groups may be in some measure supplied by 
 examples, in which we see Alexandrine and Byzantine read- 
 ings mixed in various degrees. The manuscripts in which 
 this phoenomenon occurs are chiefly of the fourth and sixth 
 centuries. After the eighth century only a few copies here 
 and there exhibit Alexandrine readings. (See Tisch., Prol. to 
 seventh ed., p. xlv.) That the text was undergoing a gradual 
 transition is borne out by other evidence. In quotations by 
 S. Chrysostom (fourth century) we find readings which agree 
 with the Cod. Brixianus (_/"), and with the Gothic version, 
 but which are not known to Origen, and do not agree with 
 the earliest versions. This points to the fourth century as 
 the period when the text began to be modified. We shall 
 see presently good reason for thinking -just this period to 
 have been the most important in the history of the Greek 
 text. (See Home and Tregelles' Introduction, p. 106 ; 
 Smith's Diet. Bib., art. 'New Test.,' p. 510, § 15.) 
 
 Thus then, by the mutual comparison of ecclesiastical 
 writers of various dates, with the versions, whose dates we 
 also know, and the earliest transcribed MSS., we believe that 
 
8o DISCUSSION OF THE EVIDENCE, ^'C. 
 
 we are able to trace a gradual change passing over the text. 
 And thus we arrive at a principle which enables us to judge 
 of the antiquity of the text of any version or MS. recently 
 discovered or whose history is unknown. We should infer 
 that it belongs to such an age as the admixture of Byzantine 
 with Alexandrine readings in its text seemed to indicate. 
 These are the grounds on which the Peshito is adjudged to 
 be posterior to the Curetonian Syriac, and this latter version 
 to be of the earliest possible date ; which give extreme prob- 
 ability to the assumed antiquity of the Memphitic, and the 
 still earlier date of the Thebaic ; and on which such cursives 
 as I and 33 are quoted as of higher authority than many 
 uncials. 
 
CHAPTER VII. 
 
 HISTORICAL CORROBORATION. 
 
 The foregoing conclusions have been reached by mere 
 investigation of the phsenomena of the text itself. An 
 important question still remains to be discussed, namely, how 
 far these conclusions harmonize with such historical evidence 
 as we possess. 
 
 Some of the critics whose schemes of classification were 
 just now spoken of attempted to account for them historically 
 by assuming authoritative recensions, or revisions, of the text 
 to have been made at different times, or in different places. 
 Griesbach at first propounded a theory of this kind, but 
 afterwards abandoned it. Hug's scheme was the most ela- 
 borate. He rested his hypothesis on a mistaken interpreta- 
 tion of some passages in S. Jerome », which speak of Lucian 
 and Hesychius having laboured at the text of the Scriptures, 
 and of certain copies called after their names. He thought 
 that Hesychius gave a recension in Egypt, Lucian in Asia, 
 and Origen in Palestine. If it could be shown that any 
 recension of the Greek text ever took place, there might be 
 reason in the claim made for the later MSS. to determine 
 the true character of the text ; for it might be said that they 
 are the results of an investigation and correction made by 
 
 * See the passages from S. Jerome quoted at length, and Hug's 
 deductions from them criticised, in Home and Tregelles' Introduction, 
 p. 78, seq. 
 
 G 
 
82 HISTORICAL CORROBORATION. 
 
 competent authority ; and that the few earlier witnesses are 
 merely relics of an imperfect state of things already tried and 
 found wanting. Unfortunately there is not a tittle of evidence 
 that any such recension ever took place. On the contrary, 
 such notices as we have, bearing upon the history of the 
 text in the fourth century, warrant an opposite presumption, 
 viz. that a difference in the value of MSS. was recognised, 
 and the Alexandrine text preferred. For instance, Con- 
 stantine^^ commissioned Eusebius of Caesarea to procure 
 copies of the Scriptures for the churches in Constantinople. 
 And Constansc (Emperor of the West 337-350 a.d.) gave 
 S. Athanasius of Alexandria a similar commission. Now 
 whether Eusebius procured his copies from Caesarea, where 
 he had the very MSS. of Origen, or from Alexandria direct, 
 they were pretty certain to present an Alexandrine text. 
 So would those of S. Athanasius. But this is not all : 
 S. Jerome's revision of the Latin, which we know to have 
 been less thorough-going than he would have wished, is 
 much more assimilated to the Alexandrine than to the 
 Byzantine text. But he expressly promises in his Preface to 
 revise it ad fidem Grcecorinn codicum, sed veierum ; and he 
 elsewhere speaks with respect of certain vera exemplaria, 
 and of the codices of Adamantius (Origen). These facts 
 show that he recognised a difference between the Greek 
 MSS. of his time; and they show moreover what character 
 of text he was in favour of Thus we have some evidence 
 of the variations of MSS. in the fourth century, and of a 
 
 b Euseb. Vita Const, iv. 36. 
 
 ^ S. Athan. ad Imper. Constantiiim Apologia, § 4 (ed. Bened. 
 p. 297 E) ; also see the Life of S. Athanasius prefixed to that edition, 
 p. xxxiii. This was about the year 340 a.d. Tischendorf (Proleg. 
 p. Ixvii.) says that Constans gave this commission in order to send the 
 boolvs to Byzantium {ad Byzandnos) ; but there must be some mistake 
 in this statement. 
 
HISTORICAL CORROBORATION. 83 
 
 preference being shown to the Alexandrian type by writers 
 of critical power like Eusebius, Athanasius, and Jerome. 
 There cannot therefore have been any authorized revision 
 producing any approximation to a Byzantine text. 
 
 It is always safer, as well as more philosophical, to in- 
 terpret ascertained phaenomena if possible in the light of 
 known historical facts, than to take refuge in conjectural 
 hypotheses. Will the history of the fourth century supply 
 us with any data for the solution of the problem before us? 
 will it help us to explain the change which we see already 
 gradually creeping over the text ? 
 
 Perhaps the most important event in the whole political 
 history of the Church has been the formal recognition of 
 Christianity by Constantine in the early part of this century 
 (Edict of IMilan, 313 a.d.), followed up by his favour to it, and 
 ultimate adoption of it. Now let us try to imagine the probable 
 effect upon a state of society, whose religious convictions 
 were of the weakest conceivable kind, when a form of religion 
 was placed before it, recommended with all the influence 
 that attaches to the court of an absolute Emperor ; and that, in 
 the new capital, Constantinople, which had no time-honoured 
 associations of its own, like those of pagan Rome, powerful 
 to hold men captive to the old religions. Hitherto the pro- 
 fession of Christianity had involved an almost certain risk of 
 persecution, perhaps of martyrdom. A^ozv it became fashion- 
 able to be a Christian ; and there are multitudes in every age 
 with whom such a motive is quite sufficient. The ranks of the 
 Christians would be rapidly recruited : and one consequence 
 of this, and of the legalization of public Christian worship, 
 would be a considerable and sudden demand for copies of 
 the Christian Scriptures. On the other hand, the difficulty 
 of supplying the demand was enhanced by the wholesale 
 destruction of the books during the persecutions of Dio- 
 cletian (accession 284, abdication 305 x\.d.). Now, bearing 
 
 G 2 
 
84 HISTORICAL CORROBORATION. 
 
 in mind what were the conditions of the case ; that 
 a book, marked by a certain ruggedness of style, dis- 
 figured (as it would be called) by provincialisms in spelling 
 and grammar, containing sometimes apparently discrepant 
 accounts of the same transactions, had to be suddenly 
 and rapidly multiplied for the public and private uses 
 of a fashionable capital, and that by mere professional 
 copyists ; we might reasonably expect to find just what we 
 do find to have happened from some cause or other. We 
 find a tendency to soften down and pare away those pro- 
 vincialisms and roughnesses, and to alter or supply words 
 where one passage seems at variance with another. There 
 was no sudden change. The tendency exerted itself very 
 gradually, and often no doubt quite unconsciously. A scribe 
 accustomed to a particular mode of spelling, for instance, 
 or to a particular grammatical construction, would use it 
 mechanically ; or a form of words famihar by repetition 
 might easily be suggested and transcribed quite unintention- 
 ally in a different passage, in which some similar words, or 
 perhaps only some one leading word, occurred. In later 
 times such alteration was intentional, as is shown by the 
 correction throughout, at the cost of immense trouble, of 
 such codices as B, or the Cod. Claromontanus, from the 
 Alexandrine readings to the more classical forms of the 
 later MSS. 
 
 It is no less easy to account for the existence of so many 
 more MSS. of the Byzantine than of the Alexandrine type. 
 Of course for a time the old centres of multiplication of 
 copies, Alexandria, Antioch, and Csesarea, remained in acti- 
 vity as well as Constantinople ; and thus, from the comparison 
 and correction of one copy by another, all sorts of mixed 
 readings might easily get into circulation. But after the 
 Mohammedan conquest of Egypt and Syria (633-639 a.d.), 
 Constantinople remained the centre of Eastern Christianity 
 
HISTORICAL CORROBORATION. 85 
 
 for eight hundred years, until its capture in 1453; during 
 which time the influences spoken of above would continue 
 to operate with greater or less force, until, by the continual 
 imperceptible accumulation of small changes, often without 
 any distinct conscious intention, the majority of copies in 
 circulation, though with many individual peculiarities, would 
 exhibit a family likeness of their own, gradually more and 
 more divergent from the ancient Alexandrine type. 
 
 Thus, in strict accordance with historical facts, and with- 
 out having recourse to any supposed revisions of the text, 
 we conceive that the phaenomena exhibited by the extant 
 MSS. may be fully accounted for. 
 
 The relation thus shown to exist between the early Alex- 
 andrine type and the later Byzantine type of text is the justi- 
 fication of the remark at the end of Chapter I, which at first 
 sight seems startling ; namely, that we are warranted in re- 
 fusing any authority to the Textus Receptus as such. We are 
 now more prepared to accept a text formed upon those 
 documents, IMSS., Versions, and Patristic writings, which 
 we have seen contain the earliest type of text : we do not 
 look for unanimity in the documents from which we propose 
 to elicit the true text : we do not expect to find the true text 
 complete in any single MS., or even any set of MSS. All 
 the different sources of evidence have to be laid under con- 
 tribution. Yet no one need be afraid that any uncertainty 
 is thereby introduced into the sacred text, or the slightest 
 doubt thrown upon any single doctrine whatsoever. The same 
 investigations which justify this course of proceeding indicate 
 clearly enough the proper mode of handling the materials 
 placed before us. The result being that, except in a very 
 few places, critical editors would be found to give the same 
 text ; and those few places would be of no real dogmatic sig- 
 nificance. The truth is, that no doctrine of Christianity is 
 founded on any one or two isolated passages. To argue as 
 
86 HISTORICAL CORROBORATION. 
 
 if it were so would indicate entire misapprehension of the 
 grounds of our faith. Moreover, if these principles of deal- 
 ing with the text seem to take away something with one 
 hand, they give back something at least as valuable with the 
 other. The same method which expunges the passage con- 
 cerning the Heavenly Witnesses, and denies the reputed 
 authorship of the conclusion of S. Mark's Gospel, establishes, 
 at any rate, the canonicity of this passage, and places 
 beyond all reasonable doubt the authenticity of S. Luke xxii. 
 43, 44. The often-quoted words of Bentley are as true now 
 as when he wrote them : ' Make your thirty thousand (various 
 readings) as many more, if numbers of copies can ever reach 
 that sum : all the better to a knowing and serious reader, 
 who is thereby more richly furnished to select what he sees 
 genuine. But even put them into the hands of a knave or 
 a fool, and yet, with the most sinistrous and absurd choice, 
 he shall not extinguish the light of any one chapter, nor so 
 disguise Christianity but that every feature of it will still be 
 the same.' 
 
APPENDIX A. 
 
 ON CANONS OF CRITICISM. 
 
 It remains to notice some principles of criticism which 
 have guided different critics in their task of deciding between 
 the claims of conflicting readings. With regard to their 
 value, it must be borne in mind that they are inferences 
 rather than axiomatic principles. They are the recorded 
 results of the comparison and interrogation of a large mass 
 of documents of various kinds. Further, they belong to the 
 region of probable evidence. Some of them admit of being 
 more widely applied than others, and none of them could 
 with safety be applied universally. By a well-known con- 
 vention the value of such statements may be represented by 
 a proper fraction, determined in each case according to the 
 observed facts. For instance, let us suppose that the value 
 of one of these principles is represented by the fraction ^- 
 This means that it may be expected to hold true in seven- 
 teen cases out of every twenty ; but then, if rigidly appUed, 
 it would lead to a wrong result in three cases out of every 
 twenty. Hence these canons must be applied with caution, 
 and in combination with other evidence. 
 
 The student must beware of supposing, however, that there 
 is any possibility of a mere arithmetical adjustment of the 
 claims of conflicting readings. In estimating the probability 
 of a various reading having arisen from some particular 
 
88 ON CANONS OF CRITICISM, 
 
 cause, which may vary in different MSS. according to the 
 observed idiosyncrasies of the scribe ; and in comparing the 
 evidence of different kinds, external and internal, for and 
 against conflicting readings; apart from the practical ac- 
 quaintance with the work of collating MSS., there must 
 always be ample scope for the highest critical acumen, as 
 well as for the most highly trained perception of the value 
 of evidence. 
 
 It seems almost superfluous to affirm that every element 
 of evidence must he allowed its full weight ; but it is a prin- 
 ciple that must not be forgotten. 
 
 Then, with reference to the External Evidence, such 
 canons as the following have been laid down : — 
 
 1. The combined testimony of the earliest MSS. with the 
 
 earliest versions, and quotations in the earhest writers, 
 marks a certain reading. 
 
 2. In estimating the value of conflicting evidence, great 
 
 weight must be given to the testimony of witnesses 
 from localities widely separated from each other. 
 Such testimony will outweigh that given by witnesses 
 of one class, or coming from one locality, even though 
 these may be numerically superior : and it can be 
 satisfactorily met only by a counter consensus of 
 witnesses from different localities. 
 
 3. It may be laid down generally that mere numerical 
 
 preponderance of witnesses of one kind is of very 
 httle weight. 
 
 4. The relative weight of the three classes of evidence 
 
 differs for different sorts of errors: therefore there 
 can be no mere mechanical determination of the 
 Text, by always taking the verdict of two out of the 
 three classes, or by any other similar short and easy 
 method. 
 
ON CANONS OF CRITICISM. 89 
 
 5. Disagreement of the ancient authorities often marks the 
 
 existence of a corruption anterior to them. 
 
 6. The ancient reading is generally the reading of the 
 
 more ancient manuscripts. 
 
 Of canons relating to Internal Evidence the following 
 are specimens * : — 
 
 I . Brevior lectio prceferejida verbosiori. This is Griesbach's 
 first canon. It may be found, together with his 
 others, with its various limitations and corollaries, in 
 the Prolegomena to Dean Alford's Greek Testament, 
 vol. i. It rests on the well-known tendency of tran- 
 scribers, already before alluded to, to include in the 
 text all marginal notes, glosses, &c. found in their 
 copy; nothing, if possible, being omitted. This 
 canon has additional probability in cases where the 
 shorter reading is harsher than the other, or elliptical 
 or obscure ; for then there is the possibility of the 
 longer reading being an intentional alteration; or again, 
 if there is in addition a variation between the read- 
 ings of the codices, either in the phraseology, or in 
 the order of words ; or again, at the commencement 
 of passages appointed as Church Lections. 
 On the other hand, there are considerations which may 
 sometimes cause a preference of the longer reading, 
 e. g. where a homoioteleuton may have occurred. Ex- 
 amples of cases for the application of this canon 
 
 » It must be borne in mind that this list is not intended to be 
 exhaustive. Every critical editor has laid dovv^n his own principles, 
 of which it will generally be found that some cover the same ground as 
 those of other editors, though differently worded ; others depend upon 
 the particular theories of the editor in question himself. The object of 
 these pages being to give the beginner a general notion of the subject, 
 only a few examples have been selected, of those most widely agreed 
 upon, as illustrations of the mode of dealing with the evidence. 
 
90 ON CANONS OF CRITICISM, 
 
 have been given at pp. 74, 75. See also the remarks 
 on pp. 18, 19. 
 2. Proclivi lecHoni prcBstat ardua. This was first laid down 
 by Bengel. It depends upon the tendency of tran- 
 scribers to alter (in perfect good faith, and fancying 
 that they were doing a good work) something they 
 did not understand into something which they did. 
 It is of very wide application, but requires great cir- 
 cumspection in its use, for it may easily be over- 
 pressed. Among lectiones arduce will be included 
 some cases of solecism or unusual readings, rare or 
 irregular usages of words, substitutions of less defi- 
 nite for more definite expressions (but here great 
 caution is needed), cases of want of connexion, &c. 
 This principle renders diKaioa-vvrjv for i\er)p.o(Tvur)v 
 (S. Matt. vi. i) the more probable reading. It is an 
 argument for those who would insert 6 Qeos (Rom. 
 viii. 28); though in this case the diplomatic evidence 
 on the other side is too strong. 
 Griesbach laid down a maxim which would be covered 
 by this one ; prcc/eratur aliis lectio cm subest sensus ap- 
 parenterfahus, qui vero re penitus exami^iata verus esse 
 deprehenditur . An illustration of this may be taken 
 from Tregelles' Printed Text, pp. 203, 204. In the 
 
 text, I Cor. xi. 29, 6 yap ea-dicov kol tt'lvwv (ava^ias) Kpifia 
 eavTa ia-QUi kcu TTtVet /u?) diUKpivcov to crcofxa, the WOrd 
 
 dva^lcos is wanting in the best authorities ; and its 
 absence may at first sight cause a little difficulty, as 
 long as the wrong impression remains upon one's 
 mind, caused by the mistranslation of the negative 
 fifj in the English Version, as if it were ov. Translate 
 this accurately, and the difficulty vanishes : ' He that 
 eateth and drinketh eateth and drinketh judgment to him- 
 self if he do not distinguish the Body! The clause /nij fita- 
 
ON CANONS OF CRITICISM. 9 1 
 
 KpivoiV TO (Twyia belongs tO the words 6 ia-OioiV Ka\ TTtVcoi/, 
 
 and is placed last for emphasis' sake. The roO Kvpi'ou 
 of the Textus Receptus is also wanting in the best 
 authorities, but its absence can cause no difficulty, 
 inasmuch as the word o-co/xa has occurred just before 
 in connexion with rov Kvpiov (ver. 27), and can there- 
 fore have but one meaning, ava^/co? might have crept 
 into the text from a marginal gloss intended to con- 
 nect the /x)7 8iaKpivoov TO (Tcofxa of ver. 29 with the dva^icos 
 
 of ver. 27. 
 3. T/ia^ reading is to be preferred ivhich will explain the 
 origin of the variations. (Tisch. Prol. xxxiii., xlii.) 
 A good illustration of this is given in Smith's Diet, of 
 the Bible, quoted from Tischendorf, though brought 
 forward by him to illustrate a different principle. 
 ' The common reading in ]\Iark ii. 22 is 6 olvo% eKx^l- 
 rat Koi 01 daKol diroTiovvTai, which is perfectly simple in 
 itself, and the undoubted reading in the parallel pas- 
 sage of S. Matthew. But here there are great varia- 
 tions. One important MS. (L) reads 6 olvos cKxetraL 
 KoX oi uo-Koi : another (D, with It. ^) 6 oivos kuI do-Kol 
 aTToXovvTaL : another (B) 6 oivos oTroXXurat Koi 01 da-KoL 
 Here, if we bear in mind the reading in S. Matthew, 
 it is morally certain that the text of B is correct. This 
 may have been changed into the common text, but 
 cannot have arisen out of it.' Closely connected with 
 this is another principle laid down by Tischendorf, 
 that a reading which savours of being an intentional 
 correction is to be suspected, notwithstanding that it may 
 be supported by a majority of the witnesses of one class. 
 For, in such a case, inspection of the true reading 
 will suggest the mode in which the correction was 
 
 *» This {Versio) Itala means what has been called by us the Vetus 
 Latina. Five of the best Codd. of this version agree in this variation. 
 
92 ON CANONS OF CRITICISM. 
 
 applied. Tischendorf's example is €7roirj(7€v in S. 
 Matthew xxv. 1 6, which he considers the true reading 
 for €Kepbr}(T€v. Tregcllcs, on the other hand, and 
 Westcott think that the diplomatic evidence for exep- 
 dr](Tev is too weighty to be set aside. (Treg. Gk. Test, 
 in loc. ; Diet, of Bible, vol. ii. p. 530.) 
 
 4. In parallel passages, whether quotations from the Old 
 
 Testament, or different narratives of the same event, 
 that reading is prima facie to be preferred which gives 
 a verbal dissidence, rather than a verbally concordant 
 reading. Instances of this principle have been already 
 given (pp. 20, 72, 75-7, &c.) The principle rests 
 on the well-attested tendency of the transcribers to 
 bring passages into harmony with one another. It is 
 discussed, with its cautions and limitations, in Tisch. 
 Proleg. pp. xxxix-xli. 
 
 5. Those readings are to be retained which are character- 
 
 istic either of the Hellenistic idiom, or of the style of 
 the New Testairmit writers. This principle looks to 
 the cases of unclassical idioms, unusual modes of 
 spelling, and other irregularities. Great caution is 
 needed in applying it, for it is almost as possible that 
 a scribe should alter the reading before him to a 
 form of expression characteristic of his author, as 
 that he should do the opposite. (See the remarks 
 on pp. 4, 5.) 
 
APPENDIX B. 
 
 CRITICAL DISCUSSION OF SOME DISPUTED PASSAGES. 
 
 We now propose to review the evidence for and against 
 a few readings of passages, respecting which there has been 
 some important difference of opinion. Some have been 
 already noticed incidentally. It will be convenient to arrange 
 the evidence for and against them under the four heads 
 separately, of Greek MSS., Versions, Fathers, and Subjective 
 Considerations. 
 
 (i.) The first text we will discuss shall be the famous one 
 of the Heavenly Witnesses (i S. John v. 7, 8). Are the 
 
 words iv TO) GvpavM 6 UaTTjp, 6 Aoyos, Koi to ayiov IlvevfMa' Koi ovtol 
 ol Tpels ev elai. Koi rpels elaiv ol fxaprvpovvres iv rfj yfj genuine, 
 
 or not .? 
 
 I. The evidence in favour of them is as follows : — 
 
 I. A MS. at Dublin, Codex Montfortianus (six- 
 teenth century) ; a MS. in the Vatican Library, 
 Codex Ottobonianus (fifteenth century) ; a mar- 
 ginal note by a seventeenth -century hand in 
 the MS. No. 173; and the Codex Ravianus, 
 which is copied from the printed Complutensian 
 edition. 
 
94 CRITICAL DISCUSSION OF 
 
 2. m of the Vetus Latina, and many of the Vulgate, 
 
 but not the best (see below) ; o?ie Armenian MS. 
 of the seventeenth century, and some of the 
 printed editions. 
 
 3. Some late Latin Fathers, viz. Vigilius (fifth cen- 
 
 tury), the Pseudo -Athanasius, Fulgentius, Cas- 
 siodorus, and others. 
 
 4. It is said to be incorporated in the Liturgies of 
 
 both the Greek and Roman Churches ; but it is 
 a grave question whether this is not a late inter- 
 polation. 
 
 IL The evidence against the passage is : — 
 
 1. It is omitted by every Greek MS. prior to the fif- 
 
 teenth century. 
 
 2. It is omitted in every Version, but the Latin, and 
 
 the suspicious Armenian exception mentioned 
 above. Even in the case of the Latin Version, 
 all the Codices but 7n of the Vetus Latina 
 omit it, and so do the best of S. Jerome's 
 revision. 
 
 3. No Greek Father quotes the passage, even in the 
 
 numerous arguments on the Mystery of the 
 Blessed Trinity, where its value would have 
 been immense. The passages in which it has 
 been alleged that Tertullian and S. Cyprian 
 refer to this text, are most easily explained with 
 reference to the rest of the passage, the disputed 
 words being expunged. 
 
 Our conclusion from this evidence must be, that the text 
 has not a shadow of a claim to authenticity. The scanty 
 evidence in its favour is all Latin, and even that not earlier 
 than the fourth century. 
 
SOME DISPUTED PASSAGES, 95 
 
 (2.) Our next passage shall be the disputed verses at the 
 end of S. ]\Iark's Gospel (c. xvi. 9-20) a. 
 
 I. Evidence against the verses : — 
 
 1. N omits the passage. The Gospel ends with €(f>o- 
 
 ^ovvTo yap, and S. Luke's Gospel begins at the 
 top of the next column as usual, without any 
 mark or note. 
 
 B omits the passage ; but a whole column is left 
 blank, as if the scribe were aware that something 
 was wanting. 
 
 L breaks ofif at ecjio^ovvro yap, and in the next column 
 gives two alternative endings to the Gospel, as 
 being both traditional: the first a short (and 
 certainly apocryphal) form, the second being 
 vv. 9-20, as commonly read. 
 
 [About thirty cursive J\ISS. mark the verses in 
 question as doubtful, by placing an asterisk 
 against them, or a marginal note, or a break 
 between vv. 8 and 9, with a note interposed.] 
 
 The passage appears to have no place assigned to 
 it by Eusebius among his 'Ammonian' Sections. 
 
 2. >^ of the Vetus Latina gives the same ending as 
 
 the first of L (above). Syr. H (mg) does the 
 same. ^Eth. (two old MSS.) gives nearly the 
 same. Arm. (some old MSS. omit the passage 
 altogether; others give the verses with a new 
 heading, after a break.) An Arabic Lectionary 
 (ninth century) in the Vatican Library omits it. 
 
 3. Eusebius, [Jerome, Gregory of Nyssa, Victor of 
 
 Antioch, Severus of Antioch, and Euthymius,] all 
 testify to a doubt thrown upon the verses, or to 
 their absence from many codices. 
 » See note on p. no. 
 
g6 CRITICAL DISCUSSION OF 
 
 4. (a) There are in this short passage as many as 
 twenty-one words and phrases which do not 
 occur elsewhere in the Gospel ; e. g. rropevo- 
 
 [xaL, Oedofiai, amcTTea), nera ravra, 6 Kvpios (ab- 
 solutely of Jesus Christ), Trpcor?/ aa^^drov, 
 &C. &C. 
 
 (^) The identification of S. Mary Magdalene, d.(p' 
 
 fis iKJ3e^Xr)Kei. crrTa Sai/xovta, notwithstanding 
 
 she has been mentioned already in this 
 chapter and the last, seems to favour the 
 hypothesis of an independent narrative, 
 rather than of a continuation by the same 
 writer. 
 II. Evidence for the verses : — 
 
 I. All the MSS. but those mentioned above. 
 2 Vet. Lat., Vulg.; Syrr.b C, P, H (text), J ; Memph.; 
 Goth, (to V. 12); -^Eth. (some). 
 
 3. Irenaeus, Tatian, Hippolytus, Apostolic Constitt. 
 
 4. (a) It is unlikely that S. Mark would end in such 
 
 an abrupt way as ecpojSovvTo ydp. 
 
 (0) The very difficulties in harmonizing it with the 
 rest of the accounts are an argument in its 
 favour ; for, had it not been a true account, 
 it could hardly have been so early and 
 widely accepted and transmitted as it has 
 been. 
 
 (7) Answer to (I. 4. ^). The words d(\> f/s k. t. X. 
 give the reason why our Lord appeared first 
 to her. The emphasis lies on the word 
 7rpo)Tov, not on the identifying clause. It is 
 a proof of His love that He appeared ^rsl 
 to her who had been chiefest of sinners. 
 
 ^ The letters after 'Syrr.' stand for ' Curetonian,' 'Peshito,' * Har- 
 clean/ and 'Jerusalem,' respectively. 
 
SOME DISPUTED PASSAGES. 97 
 
 Conclusion. — From this evidence we see that the passage 
 was extensively and decisively recognised in the second cen- 
 tury [and gradually worked its way to full recognition]. 
 The first positive evidence against it is of the fourth century. 
 It is ca?iomcaI, and to be received as genuine and inspired; 
 but not authentic ^^ in the sense of coming from the pen of 
 the writer to whom it is attributed. Of course an uncertainty 
 about the authorship of a book does not necessarily derogate 
 from its authority ; otherwise we should be obliged to reject 
 the Books of Judges, Ruth, Esther, Kings and Chronicles, 
 and the Episde to the Hebrews. It is possible that here we 
 have a trace of one of those many narratives which S. Luke 
 informs us were committed to writing in Apostolic times 
 (S. Luke i. i, 2). 
 
 (3.) The next passage for discussion shall be one which 
 presents several considerable difficulties (S. John vii. 53 — 
 viii. 1 1), the narrative of the Woman taken in Adultery. The 
 evidence is as follows. 
 
 I. Against the passage : — 
 
 I. N,-A, B, C, T, L, X, A, 33 omit it. (A, C are de- 
 ficient in this place, but the hiatus is not large 
 enough to have contained the passage. L leaves 
 a small gap ; as also does A, the scribe of which 
 began to write the first words of ch. viii. 1 2 con- 
 secutively after ch. vii. 52, and then erased them.) 
 
 E, M, A, S, n, &c. have the passage, but with an 
 asterisk or obelus in the margin. 
 
 Several cursives place the passage at the end of 
 the Gospel ; and one (69) after S. Luke xxi. 
 
 c For the difference between "■ genuine^ i.e. incorrupt, and *■ atithentic,' 
 see Blunt's Theological Dictionary, article 'Authenticity'; and Arch- 
 bishop Trench's Select Glossary, p. 15. 
 
 H 
 
98 CRITICAL DISCUSSION OF 
 
 2. Vet. Lat, a, h\ f, /*, q; Syrr. P, H; Theb.; 
 
 Memph. (oldest codd.) ; Goth. ; Arm. (oldest 
 codd.), omit the passage. 
 
 3. It is nearly certain, either because they do not allude 
 
 to the passage where the subject almost demands 
 it, or because their commentaries go on con- 
 secutively and yet pass over this section, that 
 Origen, Tertullian, Chrysostom, Cyril Alex., 
 Theodore Mops., Theophylact, and other writers 
 were ignorant of it. 
 
 4. (a) The authorities which give the passage present 
 
 great variations of reading ; which is gener- 
 ally suspicious. 
 
 (/3) The style is entirely unlike S. John's. There 
 are numerous words and expressions which 
 do not occur anywhere else in his writings ; 
 while on the other hand his special pecu- 
 liarities of style do not appear in this piece 
 of narrative. 
 
 (7) It gratuitously breaks into the middle of a 
 narrative, which runs on continuously but 
 for this interposition. 
 
 II. On the other hand : — 
 
 1. D has it, but in a somewhat different form. F, G, 
 
 H, K, U, r, and the mass of cursives, have it. 
 
 2. Vet. Lat, 3*, c, e,ff^, g, I (mg) ; Vulgate ; ^th. ; 
 
 Syr. J, &c. have it. 
 
 3. The earliest writing in which the passage is recog- 
 
 nised is the Apostolic Constitutions. S. Jerome 
 testifies that it was found in many Greek and 
 Latin codices ; and S. Augustine defends it. 
 
 Here the evidence against the passage is far stronger than 
 in the case of the end of S. Mark's Gospel. Scrivener says 
 
SOME DISPUTED PASSAGES. 99 
 
 (Introduction, p. 440), ' on all intelligent principles of mere 
 criticism the passage must needs be abandoned.' That is 
 to say, we cannot allow it to be S. John's writing. The 
 style and contents, indeed, in both of which it is utterly 
 different from any of the narratives of the apocryphal 
 gospels, convey an irresistible impression of genuineness; 
 and it is probable that we have a piece of apostolic narrative, 
 upon which the consent of the universal Church has set the 
 seal of canonicity. But it would be more satisfactory to 
 separate it from its present context, and place it by itself as 
 an appendix to the Gospel ; or at least print it in different 
 type from the rest, to draw attention to the peculiar footing 
 on which it stands ; or place it in brackets. Professor Light- 
 foot would adopt some such plan. (See his remarks on this 
 and the passage last discussed in his work On a fresh Revision 
 of the New Testament, pp. 27, 28.) 
 
 (4.) I Tim. iii. 16. Gfos ecpauepcodrj iu aapKi is the reading 
 of the Textus Receptus. For Qeos there are various readings, 
 OS and o. 
 
 It is convenient to summarize the evidence here, first for 
 a relative, and secondly for Geo'y ; then finally to decide 
 between 6? and o. 
 
 I. Testimony for a relative : — 
 
 I. «*, A*d, C*, F, G, 17, 73, t8i haveSj; D* reads 
 o. (B is defective here.) 
 
 ^ There is a difference of opinion as to the testimony of the original 
 scribe of A. Dr. Tregelles, in his edition of this part of the Greek Tes- 
 tament (published in 1870), cites it in favour of os without any sugges- 
 tion of doubt. Mr, Scrivener, on the other hand, thinks that 0C (©eos) was 
 the original reading (see his Introduction, p. 453, with an elaborate note). 
 We think that the impression left by a perusal of that note will proba- 
 bly be in favour of Dr. Tregelles' conclusion. It is one of those delicate 
 points which should be left to skilled collators and practical experts to 
 decide. But we hazard the suggestion that * the slight shadow of the 
 
 H 2 
 
lOO CRITICAL DISCUSSION OF 
 
 2. Vet. Lat., Vulg. ; Syrr P. and H. (text and mg.) ; 
 
 Memph., Theb.; Goth.; Arm.; and the Vatican 
 Arabic MS. 
 
 3. The testimony of the Patristic writers needs sift- 
 
 ing. The passages which have been cited from 
 S. Ignatius and Hippolytus as favouring the 
 reading Qeos are too vague to draw any con- 
 clusion from. The words of S. Ignatius are 
 
 (Ad EpheS. 19) Geou dvdpcoTrivais ^^avepovjievov, 
 those of Hippolytus are Qebs iv a-anari i<^avepa>6rj : 
 
 but it is evident that these may be only statements 
 of the doctrine of the Incarnation, which is in- 
 volved in the verse under discussion, without being 
 intended for express allusions to the verse. We 
 must further set on one side those citations, 
 which have been made, some in support of one, 
 some of the other reading, but which, though 
 they manifestly refer to the passage in question, 
 are paraphrastic, and might follow naturally 
 enough from either reading; since the mystery 
 of God manifest in the flesh in the Person of 
 Jesus Christ is unmistakeably expressed in both 
 readings, though more clearly in one than in 
 the other. Such quotations as Barnabas (Ep. 
 
 12), ^Irja-ovs ovx o vlos dv6pa>7rov aXX' 6 vlos tov Qeov 
 rvTrco Koi cv o-apKL (jiavepaBeis. TheodotUS, Ep. ad 
 
 Diog., and Origen (the passages out of whose 
 writings are given in Alford's Greek Testament 
 at length), are thought to favour o? ; on the other 
 
 real ancient diameter' (of the 0), which Mr. Scrivener says he saw 'just 
 above the recent one,' after he had been 'gazing at it with and without 
 a lens,' 'one singularly bright hour, February 7, 1861,' was really the 
 impression of the recent diameter retained for an instant upon the retina 
 of the eye. 
 
SOME DISPUTED PASSAGES. 10 1 
 
 hand, Theodoret, Dionysius Alex., and Gregory 
 Nyss. have been quoted as supporting Gedy. 
 The authorities which certainly favour the relative 
 are Chrysostom^^ Cyril of Alexandria^, Epi- 
 phanius, Theodore Mops., the Latin translator 
 of Origen, Jerome, Hilary, and Augustine, with 
 all the Latin Fathers. It may be added, as 
 contributing a certain weight to the evidence on 
 this side, that the text is not quoted by writers, 
 as S. Cyprian for instance, in arguments where 
 the word Geoy, had it been the acknowledged 
 reading, would have supplied a weapon too 
 powerful to be left unused. 
 
 IL Testimony for Geo? : — 
 
 1. All MSS. in which the passage is contained, ex- 
 
 cept those above mentioned. 
 
 2. No version of any critical value. It is the reading 
 
 of the Slavonic. 
 
 3. The later Greek Fathers, as John of Damascus, 
 
 CEcumenius, and Theophylact. 
 Thus then for Geos there is no certain testimony prior 
 to the ninth century — nothing before K, L, P of 
 the later uncials, and the Slavonic version ; while 
 there is an im^mense mass of early testimony for 
 a relative. 
 
 III. It remains then to decide between U and o. This is a 
 point on which most of the versions can give no help. 
 The Latin favours the neuter; but its weight is 
 diminished by S. Jerome's opinion ; the Gothic sup- 
 ports the mascuHne. The testimony of the early 
 Greek witnesses, both I\ISS. and writers, with very 
 few exceptions, is for 09. 
 
 « See the remarks on p. 59. 
 
102 CRITICAL DISCUSSION OF 
 
 We may take further into consideration : — 
 
 (a) That OS is the harder reading, owing to the 
 
 want of a clearly-expressed antecedent. 
 (j3) That 6 would be more likely to arise out of o?, 
 
 than the converse, because of the foregoing 
 
 neuter word fivo-Trjpiov. 
 (y) That the other reading, ec", would more easily 
 
 arise out of OC than out of O ; so that the 
 
 reading 6s best accounts for the existence 
 
 of both the other readings. 
 Hence, finally, we conclude that 6s is the true reading. 
 
 ayyeXos yap Kara Kaipov Kare^aivev iu rfj KoXv/x^rjBpa Koi irdpaao-e 
 TO vd(op' 6 ovv TTpcoTos ip^as pera Tr]v Tapa)(r]V rod vbaros, vyirjs 
 eyivero, a brjTroTe KareixeTO voarjpaTi. The question is whether 
 
 this passage is genuine or not. 
 
 I. Testimony against it : — 
 
 1. N, B, C*, 157, 314 omit the whole passage. 
 A*, L, 18 omit the clause cK^exopevcop. . Kivrjo-iv. 
 D, 33 omit the whole of verse 4. 
 
 S, n, A, and about fourteen cursives, mark verse 4 
 with either asterisks or obeli. 
 
 2. q omits the whole ; /, I omit verse 4. 
 Syr. C omits the whole ; H obeHzes. 
 Theb., and Memph. (Schwartze) omit. 
 Arm. (many of the codices) also omit. 
 
 3. No writers, but those mentioned below, allude to 
 
 the narrative. 
 
 II. Testimony for it : — 
 
 1. (A), C^ E, F, G, (L), &c. &c. give the passage, but 
 
 with many variations, 
 
 2. All the other I^atin codices, but those mentioned 
 
SOME DISPUTED PASSAGES. IO3 
 
 above, both of the Vet. Lat. and the Vulgate ; 
 
 Syrr. P. and J. ; and Memph. (Wilkins). 
 3. Tertullian, Chrysostom, Cyril Alex., Ambrose, 
 
 Theophylact, and Euthymius recognise the 
 
 narrative. 
 In reviewing this evidence, we find that the further back 
 we go the weaker becomes the support ; and the numerous 
 variations with which the passage is given cause still further 
 suspicion. It is a Uttle singular that the earliest evidence in 
 its favour seems to point to Africa as its origin ; as if there 
 were perhaps some tradition afloat there, which took the 
 fonn of a marginal gloss, and thence crept into the text. 
 The weight of the earliest evidence is too strongly adverse 
 to warrant our retaining the passage m the text. 
 
 (6.) S. Luke xxii. 43, 44. (li<^6r] 8e avra ayyeXos an ovpavov 
 ivKTXVoiV avTov. Koi yevoyuevos iv ayoavla eKTevicTTcpov TrpoarjvxeTo . 
 Koi eyevfTO 6 iSpcoj avTov cocrel dpo/x^oi mfiaTos Kara^aivovTOs eVi Tr]v 
 
 yrjv. These two verses have been called in question; but 
 without sufficient reason, as will be seen from the following 
 statement of the evidence. 
 
 I. Evidence against the passage : — 
 
 1. x^ A, B, R, T, 124. 
 
 13 has axjiOr] 8e {prima manu); the remainder added 
 
 sec. man. 
 C^ 69, and all known Evangelistaria, have the 
 
 passage inserted after S. Matt. xxvi. 39. 
 E, S, V, A, and others, including nine cursives, 
 
 place an obelus or asterisk against it. 
 
 2. /; Memph. (one codex), Thebaic (ed. Woide), and 
 
 some Armenian, omit. 
 Syr. H (mg) marks with an obelus. 
 
 3. Cyril Alex, does not notice the verses in his 
 
 Homilies on the Gospel of S. Luke ; nor does 
 
104 CRITICAL DISCUSSION OF 
 
 S. Athanasius quote them, where it would have 
 been natural for him to do so. Hilary testifies 
 I that the passage is wanting in very many Greek 
 
 and Latin codices ; S. Jerome, that it is found 
 in some. 
 11. Testimony for the passage : — 
 
 1. t{* and 3^ D, F, G, H, K, L, &c., and nearly all 
 
 cursives. A has the Ammonian section which 
 belongs to the passage marked in the margin; 
 though the verses are wanting in the text. 
 
 2. All the codices of the Vet. Lat., but/"; Syrr. C, P, 
 
 H, and J ; Thebaic (one codex) ; Memph. (edd. 
 Wilkins and Schwartze) ; Arm. (some). 
 
 3. Justin Martyr, Irenseus, Hippolytus, and Dionysius 
 
 Alex, very clearly refer to it; as do Hilary, 
 
 Jerome, and Augustine. 
 
 Thus there is very full and early evidence in favour of the 
 
 passage ; in fact, the only very strong argument against it is 
 
 its omission by B ; and with this may be contrasted its 
 
 presence in x. 
 
 On the whole, there is no reasonable doubt upon the 
 passage. 
 
 (7.) S. Matt. xxi. 28-31. The difficulties in connexion 
 with this passage do not admit of being stated very shortly. 
 There is a question of words in verse 31, viz. whether vVrepoy, 
 or '^axo-Tos, the meaning of which would be nearly the same, 
 is to be substituted for irpatTos in the answer of the Chief 
 Priests. But this is complicated by a question of the order 
 of the narrative ; for some of the authorities transpose the 
 answers of the two sons in the parable, placing first the 
 answer of the son who professed to do his father's bidding 
 but went not, and the answer of the other son second. Thus 
 we really have three questions to consider : — 
 
SOME DISPUTED PASSAGES. 105 
 
 (a) The order in which the sons are mentioned. 
 
 (/3) Which of the two sons did the Chief Priests intend 
 
 to assert had done his father's bidding ? 
 (y) The choice between the three words, Trpwroy, va-repos, 
 
 or ea^aros. 
 
 And we must take the evidence in the order here indicated, 
 (a) To decide, then, the order in which the two sons are 
 mentioned we have the following data : — , 
 
 I. For the order of the Textus Receptus — 
 
 1. N, C, D, L, X, Z, &c., and most of the cursives. 
 
 2. Vet. Lat. ; Vulg. ; Syrr. C, P, and H. 
 
 3. Origen, Eusebius, Cyril, Chrysostom, Irenseus {mf.), 
 
 Hilary. 
 
 II. For the converse order, which would make the elder 
 
 son promise to go and then fail : — 
 
 1. B, and seven cursives. 
 
 2. One MS. of the Vulg. {sec. man.); Memph.; Syr. J ; 
 
 Arm. ; JEih. (two codices). 
 
 3. Isidore, John of Damascus, the Pseudo-Athanasius. 
 (/3) As to the second question, which of the two sons the 
 
 Chief Priests meant to say had done the father's 
 bidding, we have to notice that all the MSS. and 
 versions enumerated above, which reverse the order 
 in which the sons are mentioned, also substitute 
 varepos or bevrepos, or some equivalent word, for the 
 TTpcoTos of the Textus Receptus : thus the reply of 
 the Chief Priests to our Lord is represented as 
 virtually the same in either case. But D, and a 
 good many codices both of the Vetus Latina and 
 the Vulgate, which agree with the Textus Receptus 
 in the order of the sons, have respectively eo-xaros 
 and novissimum for TrpcaTos; thus transposing the 
 connexion. S. Jerome interprets this answer on 
 
Io6 CRITICAL DISCUSSION OF 
 
 the hypothesis that the Chief Priests knew what 
 answer our Lord intended them to give, but 
 purposely gave a wrong one : at the same time, 
 however, he asserts that ' vera exemplar ia ' had 
 primum and not novissimum for their reading. 
 There is only then the witness of D, backed by 
 the partial testimony of the Latin versions, in 
 favour of this answer of the Chief Priests. 
 On the whole, then, the evidence for the order of the 
 Textus Receptus is conclusive ; and the evidence 
 for making the Chief Priests recognise the obedi- 
 ence of the son, who at first refused but afterwards 
 repented, is overwhelming. 
 (7) And thus we are helped to an easy solution of the 
 third question : namely, that we must adopt the 
 reading TtpoaTos of the Textus Receptus. 
 There are one or two subordinate variations, but not of 
 sufficient consequence to demand separate treatment. 
 
 It may be remarked that Dr. Tregelles adopts the reading 
 6 v(TTipos without the previous transposition of the two sons, 
 and explains it as equivalent to 6 va-repou fieTa[X€\T]deis ; the 
 grammatical possibility of which may well be questioned. 
 
 (8.) Acts XX. 28. There are six readings here to decide 
 
 between, viz. (1) tov Qeov. (2) rov Kvplov. (3) Tov Kvplov Koi 
 Geov. (4) TOV Kvpiov Qeov. (5) rov Qeov kol Kvplov. (6) tov 
 
 XptOTOV. 
 
 It will be most convenient to consider them in the reverse 
 order to that in which they are here enumerated. 
 In favour of (6) there is — 
 
 1. No MS. authority. 
 
 2. Syr. P; and Vet. Lat. m {/esu Christi). 
 
 3. Athanasius (some codices), Origen, Theodoret. 
 This therefore may be at once dismissed as a gloss. 
 
SOME DISPUTED PASSAGES. lO; 
 
 In favour of (5), only cursive No. 47 is quoted. 
 In favour of (4), only No. 3, and 95 {sec. ?nan.). 
 In favour of (3) — 
 
 1. C^, H, L, P, and more than one hundred and ten 
 
 cursives. 
 
 2. The Slavonic (Tregelles' Printed Text), but no 
 
 version of critical value. 
 
 3. Theophylact (in one place). 
 
 These three variants then may be dismissed as conflate 
 readings, which really only testify to the existence of a doubt 
 in early times between the claims of the two remaining 
 important readings, eeoG and KvptW. Between these the 
 evidence is so nearly balanced, that the decision cannot be 
 absolutely final. 
 
 In favour of (2) we find — 
 
 1. A, C*, D, E, and about fifteen examined cursives. 
 
 2. Theb.; Memph.; Syr. H (mg); Arm.; and (accord- 
 
 ing to Tischendorf) the Roman ^thiopic. 
 
 3. Irenaeus [int.), Apostolic Constitutions, Athana- 
 
 sius (one codex), Didymus, Chrysostom (in a 
 catena). 
 But some of the quotations adduced, as that of Eusebius, 
 
 (TvvTjyiu.€Voi 8ia Kvpiov ovs avTos ekvTpiJXTaTO rw lhi(0 alfiaTi, are not 
 
 close enough to the text to warrant us in asserting that one 
 and not the other reading was intended to be quoted. There 
 is a reminiscence of the passage, doubtless, but not a verbal 
 quotation. 
 
 On the other hand, in favour of (i) are ranged — 
 
 1. N, B, about ten cursives, and twelve Lectionaries. 
 
 2. Vulg. ; Syr. H (text). 
 
 3. Chrysostom (three times), Cyril Alex, (twice), 
 
 Epiphanius, and others. This is the only pas- 
 sage that would give Scriptural warrant for the 
 
108 CRITICAL DISCUSSION OF 
 
 remarkable expression of S. Ignatius, eV anian 
 Qeov (Ad Ephes. i) ; but in opposition to this 
 the strong assertion of S. Athanasius is alleged, 
 
 ovSa/xov aljxa Qeov hi)(a crapKos TrapabchooKacnv al 
 ypacpai. 
 
 This is just one of the cases to which the remark of 
 Dean Alford, quoted at pp. 4, 5, applies with its full force. 
 Whichever of the two readings we suppose to have been the 
 original, some reason may be supposed for the substitution 
 of the other. 'H eKKki^a-ia (al e/cK.) tov Qeov is a common ex- 
 pression of S. Paul ; 'H eKK. TOV Kvpiov occurs nowhere else 
 in the New Testament. Again, it is a small argument, per- 
 haps, but not to be wholly passed over, that while 6 Kvpios 
 occurs three times in this speech of S. Paul to the Ephesian 
 Elders, it is always with some addition : in two places (verses 
 24j 35) it is 6 Kvpios 'irja-ovs; in the third (verse 21) there is 
 some little doubt, but the reading is perhaps t6v Kvpiov f)p.wv 
 'irjo-ovv [Xpiarop). Now when a person is speaking under the 
 influence of strong emotion, he commonly uses his own 
 natural, that is, his characteristic, style ; and moreover, he is 
 very apt to repeat without variation the expressions in which 
 the idea which he desires to impress upon his audience 
 first suggested itself. There seems a peculiar tenderness in 
 S. Paul's dwelling thus upon the name of his Lord. These 
 considerations would rather lead us to look for the familiar 
 Tr}v cKKkijaUv rov Qeov, and to expect that if Kvpiov were 
 S. Paul's word he would have added 'It^o-oC or '1770-0^ Xpio-rov. 
 
 It may be said on the other side with much force, that it 
 is more likely that the unusual Kvpiov should be altered into 
 the familiar Qeov, than the reverse, which could only be done 
 for theological reasons ; a charge we are always unwilling to 
 bring. 
 
 There is weight too in what Tischendorf says ; that, if we 
 assume Kvpiov to be the original reading, it is much easier to 
 
SOME DISPUTED PASSAGES. icg 
 
 understand the addition of GfoG, and thus get at the origin 
 of those mixed readings, than to understand the addition of 
 Kvpiovy if Geou had stood originally in the text. 
 
 Tregelles and Tischendorf both place Kvplov in the text ; 
 Tregelles places Qeov in the margin, as an alternative reading 
 strongly supported. Perhaps this is the best conclusion that 
 the evidence admits of. 
 
 (9.) Acts xi. 20. We will discuss one more passage, which 
 records an interesting fact in the history of the infant Church. 
 The question here is between the reading of the Textus Re- 
 
 CeptUS 'EXXtjvkttcls, and "EWrjvas. 
 
 I. For 'EX\r]viaTas I — 
 
 1. B, D^ E, H, L, P, 13, 61, and almost all 
 
 cursives. N*, which has the strange reading 
 EvayyeXiVray, seems from the termination of that 
 word to favour this reading. 
 
 2. No version can be quoted in its support ; but no 
 
 great stress can be laid on this fact, since the 
 versions in general appear not to recognise the 
 distinction. 
 
 3. S. Chrysostom, with CEcumenius and Theophylact, 
 
 in quoting the passage favour this reading ; but as 
 their commentaries clearly imply the other read- 
 ing, it may be that the text has here been altered 
 by the transcribers. 
 
 II. For "EXkrjvas I — 
 
 1. ^<^ A, D*, c (of Mr. Scrivener's MSS.). 
 
 2. Armenian; and apparently the ^thiopic (Tre- 
 
 gelles). 
 
 3. Eusebius and Chrysostom, followed by CEcu- 
 
 menius and Theophylact as indicated above, in 
 his commentary, e. g. opa^ "EXXrja-iv evayyeXl^ovT ai. 
 
no CRITICAL DISCUSSION OF 
 
 Taken simply by itself the evidence might seem to be 
 pretty evenly balanced : but we must throw into the scale the 
 important consideration that the reading 'EXXrjvKTTas makes 
 nonsense of the passage. There is evidently a contrast 
 intended by the writer between the 'lovSalot, to whom the 
 other preachers of the gospel spoke, and the persons 
 addressed by these men of Cyprus and Cyrene at Antioch. 
 This contrast is heightened by the kqI, which is undoubtedly 
 to be inserted after iXaXow. But the 'eXXtjvkttoI were Jews ; 
 and the proper antithesis to 'EXXrjvia-Trjs is not 'lovBalos but 
 'E/3paToy. We are constrained therefore to adopt "EXXrjvas as 
 the true reading. 
 
 NOTE ON S. MARK xvi. 9-20. 
 
 Mr. Burgon's book on The Last Twelve Verses of the Gospel ac- 
 cording to S. Mark has been published since these pages were put into 
 the printer's hands. We have thought it best to leave the original 
 statement of the case unaltered, only placing in square brackets those 
 portions of the evidence which Mr. Burgon's researches have shown to 
 be untrustworthy. The reader will thus see what was supposed to be 
 the evidence until now, and what we must now accept. He will also see 
 how little real difference there is after all between the conclusion stated 
 above, which is substantially that of Dean Alford and Dr. Tregelles, and 
 the conclusion to be drawn from Mr. Burgon's corrected premisses. 
 
 We believe that the results of Mr. Burgon's work, so far as it bears 
 directly on the evidence, may be not unfairly summarised as follows : — 
 
 I. The evidence of the 'about 30 cursives' is really in favour of, and 
 not adverse to, these verses forming a part of the Gospel. They 
 all have a scholion recognising the absence of them from some 
 codices ; at the same time in various words they testify to their 
 being found 'in others,' 'in many,' 'in the ancient copies,' 'in 
 the true Palestinian copy,' or ' in the approved copies preserved 
 at Jerusalem.' 
 
 3. The -evidence of the Fathers commonly quoted as adverse to the 
 authenticity of the verses is really to be reduced to that of 
 Eusebius. He does seem to have had some doubt about them, 
 but the others only quote his words. 
 
SOME DISPUTED PASSAGES, III 
 
 3. The force of the argument drawn from the alleged sudden change 
 
 of style and pliraseology is shown to be much less than it is 
 commonly represented to be. 
 
 4. Most striking of all is Mr. Burgon's explanation of the undoubted 
 
 omission of these verses from so many codices. He shows that 
 the word reXos, whose occurrence at the 8th verse has misled so 
 many critics, is really only the mark of the conclusion of an 
 important ecclesiastical Lection ; and reminds us, in addition, that 
 S. Mark's Gospel often, and in the West usually, stood last in 
 order of the four ; whence it might easily happen that the last 
 verses of S. Mark were written on the last leaf of the codex, and 
 so might be in danger of being damaged or torn away. 
 
 The adverse testimony then is reduced to — 
 
 1. The fact of the absence of these verses from a certain number of 
 
 codices. 
 
 2. The deliberate opinion of Eusebius, which would be implied by 
 
 his not ' canonizing ' further than verse 8; assuming the statement 
 ecus ov EvaiPios u IIafX(pl\ov kfcavouiaev. 
 
 3. The somewhat marked difference of phraseology. 
 
 But this is balanced by such strong external evidence, that we are 
 driven to the conclusion that these verses have formed part of the 
 canonical Gospel from the earliest times of which we have knowledge. 
 The question only remains, Are they from S. Mark's own pen ? 
 Now, inasmuch as the claim of any part of Scripture to be received 
 by us depends, not upon our knowledge of the writer, but upon the 
 authority of the universal Church which has pronounced it canonical, it 
 appears to us that a question of doubtful authorship is to be treated as 
 a purely literary question, to be solved by the proper use of the critical 
 and judicial faculties; and that such considerations as these may be 
 allowed their full weight. There is certainly a difference between the 
 first eight and the last twelve verses of this i6th chapter, which is not 
 likely to occur in the composition of an author writing continuously. 
 Whether S. Mark wrote them at an interval of some time from the 
 rest, or whether he incorporated an account by a different hand from 
 his own, as S. Luke certainly did both in his Gospel and in the Acts, or 
 whether his Gospel, being for some reason incomplete, was completed 
 in Apostolic times by the addition of an already existing narrative, need 
 make no difference in our acceptance of the passage as inspired. 
 
APPENDIX C. 
 
 LIST OF THE CHIEF UNCIAL MSS. 
 
 In the following list are given, the letter by which each 
 MS. is usually cited, the common name of the MS., the 
 century when it was transcribed, and its present locality; 
 and in some instances its contents and condition are indi- 
 cated. 
 
 The designation-letters are of course those now commonly 
 assigned ; and will be found to agree with the list of Tis- 
 chendorf, so far as his latest (eighth) edition has been 
 published. But if the student should compare it with any 
 old list, as that prefixed to Bruder's Concordance, or that of 
 any old critical edition of the New Testament, he will find 
 discrepancies. For some of the older known MSS. have been 
 dropped out for critical reasons, as O (Montefalconii) and 
 R (Tubingensis) in Bruder's list; and others, as J (Cotto- 
 nianus) and T (Vaticanus), have been proved to be parts of 
 N (Codex Purpureus), and are now quoted under that same 
 letter N : and the letters thus set free have been assigned to 
 other MSS. more recently discovered. The names of pri- 
 mary uncials are in capitals, the names of secondary uncials 
 in black type. 
 
 X. Cod. SINAITICUS [IV]. Imperial Library at Peters- 
 burg. A great part of Old Testament, and the New 
 
CHIEF UNCIAL MSS, II3 
 
 Testament entire. The Cod. Friderico-Augustanus 
 at Leipsic is really a part of this IMS. (See further, 
 p. 33> &c.) 
 
 A. Cod. ALEXANDRINUS [V]. Library of the British 
 INIuseum in London. The whole of the Old and 
 New Testaments, except a few leaves which have 
 been lost. It contains also the only extant copy of 
 the first Epistle of Clement of Rome, and a frag- 
 ment of the second, placed as if they belonged to 
 the Canonical books. The writing is continuous, 
 in uncial characters of very elegant and clear form, 
 with capital letters larger than the rest, and pro- 
 jecting beyond the line, at the beginning of books 
 and sections. A very simple punctuation is intro- 
 duced, consisting of a single point at the end of a 
 sentence, followed by a break in the writing. 
 There are no accents or breathings, except at the 
 beginning of the book of Genesis, where the first 
 four lines of each column are written in vermillion. 
 Each page has two columns. The tItXol, the 
 ' Ammonian Sections,' and the Eusebian Canons 
 are found complete in the Gospels ; but there are 
 no marginal marks of division throughout the rest 
 of the New Testament, though the text is divided 
 as the sense requires by paragraphs and capitals. 
 The titles and subscriptions of the books are still 
 very short and simple, though a little longer than 
 those found in N and B : e. g. for Kara UadSaiou we 
 
 here find evayyikiov Kara MadBmov, &C. 
 
 To determine the date of the Codex we have such 
 arguments as these : — The presence of the Epistles 
 of Clement, the shortness of the subscriptions, and 
 the absence of the Euthalian divisions of the Acts 
 
 I 
 
Tf4 LIST OF THE 
 
 and Epistles, would all point to a date not later 
 than the middle of the fifth century ; while the in- 
 sertion of the Eusebian Canons, and of the Epistle 
 of Athanasius to Marcellinus, would prevent our 
 assigning a date earlier than the latter half of the 
 fourth. But the style of the writing is somewhat 
 later than that of N and B, and would point to the 
 early part of the fifth century. 
 
 B. I. Cod. VATIC ANUS [IV]. Vadcan Library in Rome. 
 
 The Old and New Testaments, except the Epi- 
 stles to Timothy, Titus, and Philemon, and a part 
 of the Episde to the Hebrews. The Apocalypse 
 and the missing part of the Epistle to the He- 
 brews have been added by a more recent hand. 
 (See further, p. 37, &c.) 
 
 2. Cod. Basilianus [VIII]. Vatican Library in Rome. 
 Apocalypse. 
 
 C. Cod. EPHRAEMI [V]. Imperial Library in Paris. 
 
 Fragments of the LXX, and of all the books 
 of the New Testament but 2 Thessalonians and 
 2 S. John. It is a palimpsest MS. {Codex rescripiiis). 
 In many palseographical details there is great simi- 
 larity between this MS. and Cod. A. The writing 
 is somewhat smaller and a little more elaborate 
 than that of A, and there is but one column of long 
 lines on a page ; but there is the same absence of 
 accents and breathings, the same simple punc- 
 tuation, the same sort of inidal capital letters, and 
 the same simple subscriptions to the books. More- 
 over the Ammonian Sections are marked, and the 
 lists of TiVXot are given at the beginning of each 
 Gospel ; while there are no marks of the division 
 
CHIEF UNCIAL MSS. 115 
 
 into chapters in the other books. These charac- 
 teristics point to the fifth century as the date of its 
 transcription. Three correctors have left their 
 traces on the MS., which is one of first-rate im- 
 portance. 
 
 Cod. BEZ.E [VI]. Cambridge University Library. 
 This MS. contains portions of the Gospels in the 
 Weshr?i order (viz. SS. Matthew, John, Luke, 
 Mark), and the Acts; between which stood for- 
 merly the Catholic Epistles, now represented by 
 only a few verses of 3 S. John. Out of 534 leaves, 
 which it must once have possessed, 128 are gone. 
 It is a Graeco-Latin MS., written stichometrically, 
 the Greek being placed on the left-hand page of 
 the opening, the Latin on the right, and on the 
 whole corresponding line for line. The Latin is 
 thought (see Scrivener's Edition of the Cod. Bezos, 
 Introd. pp. xxxiv. n. i ; Ixiv) not to be an inde- 
 pendent version, but a translation from another 
 Greek text almost identical with that of the codex 
 itself. 
 
 The initial letters are not larger than the rest, but 
 stand out a little from the line, as in cod. N ; and 
 there are no marks of divisions inserted by the 
 original scribe. 
 
 A great deal of the interest of this MS. depends upon 
 the interpolations with which it abounds, especially 
 in the Acts ; some of which are unsupported by 
 any other authority, some are countenanced by the 
 Vetus Latina and Curetonian Syriac versions. 
 These are so characteristic that, as stated above 
 (p. 69), some critics have formed a separate group 
 of the authorities in which they occur. They are 
 I 2 
 
Il6 LIST OF THE 
 
 probably due to the influence of tradition still 
 lingering on, and are at all events a proof of the 
 extreme antiquity of any such text. 
 Apart from these interpolations D presents a very 
 valuable text, akin in its readings to that of the 
 Alexandrine type. 
 
 D. 2. Cod. CLAROMONTANUS [VI]. Imperial Library 
 
 at Paris. The Epistles of S. Paul, with one small 
 hiatus, Romans i. 1-7. A Graeco-Latin MS., sticho- 
 metrically written. The Latin version represents 
 the Vetus Latina. 
 
 E. I. Cod. Basileensis [VIII]. Public Libraiy at Basel. 
 
 The Gospels entire, except a few verses of S. 
 Luke. 
 
 2. Cod. LAUDIANUS [VI]. Bodleian Library at Ox- 
 
 ford. The Acts, with one hiatus (xxvi. 29-xxviii. 
 26). A Greeco-Latin MS., written in very short 
 ari'xot. The Latin follows the Greek closely, and 
 is therefore not an independent authority. 
 
 3. Cod. Sangermanensis [X]. Imperial Library at 
 
 Petersburg. The Epistles of S. Paul, but muti- 
 lated in two or three places. A Graeco-Latin MS., 
 and a transcript of D2. 
 
 F. I. Cod. Boreeli [IX]. Public Library at Utrecht. The 
 
 four Gospels, but mutilated. The MS. appears to 
 have suffered further injury since its first collation 
 by Wetstein (Tischendorf ). 
 
 2. Cod. AUGIENSIS [IX]. Library of Trinity College, 
 Cambridge. The Epistles of S. Paul. A Graeco- 
 Latin MS. ; the Latin being an example of the 
 best Vulgate, ' somewhat tampered with in parts 
 
CHIEF UNCIAL MSS. 11/ 
 
 to make it suit the Greek text.' Rom. i. i-iii. 19 
 is wanting: and //le Greek of 1 Cor. iii. 8-16, 
 vi. 7-14, Col. ii. 1-8, and Philem. 21-25, ^vith the 
 entire Epistle to the Hebrews, is wanting; the 
 Latin however remains. 
 
 F^ Cod. Coislianus i (marg.) [VII]. Paris. By this letter 
 are designated some fragments of the Gospels, 
 Acts, and Epistles of S. Paul, found in marginal 
 notes to the great Septuagint Octateuch known as 
 Codex Coislianus i. 
 
 G. I. Cod. Harleianus (formerly known as Seidelii I, or 
 Wolfii A) [IX or X]. Library of British Museum 
 in London. The Gospels, much mutilated. 
 
 2. Cod. Angelicus (or Passwnei) [IX]. Library of the 
 
 Augustinian monks at Rome. The Acts and Catho- 
 lic Epistles entire, except that it only commences 
 at Acts viii. 10 (/^is tov Qeov). The same JMS. con- 
 tains the PauHne Epistles; but is cited for them 
 under the designation L (see below, Lg). 
 
 3. Cod. BOERNERIANUS [IX]. Royal Library at 
 
 Dresden. The Episdes of S. Paul, but mutilated 
 in places. A Grseco-Latin MS. The Latin is 
 interlinear, and in a cursive character ; a specimen 
 of the Vetus Latina altered to suit the Greek. As 
 to the Greek text, this MS. is a sister MS. to F2 ; 
 the two MSS. having been clearly copied from the 
 same archetype: not so the Latin. Moreover it 
 once formed part of the same volume as A (see 
 below) - 
 
 H. T. Cod. Seidelii (formerly Seidelii II, or Wolfii B) [IX 
 or X]. Public Library at Hamburg. The Gos- 
 pels, a good deal mutilated. 
 
Il8 LIST OF THE 
 
 H. 2. Cod. Mutinensis [IX]. Grand Ducal Library at 
 Modena. The Acts, mutilated. 
 
 3. Cod. COISLIANUS 202 [VI]. Fragments of the 
 Episdes of S. Paul, stichometrically written, of 
 which twelve leaves are at Paris, and two at 
 Petersburg. 
 
 I. FRAGMENTA PALIMPSESTA TISCHENDORFIA- 
 
 NA (or Cod. Tischendorfianus II). Under this 
 designation are cited (severally as I^, I^, &c.) seven 
 fragments of the Gospels, Acts, and Pauline Epi- 
 stles, now at Petersburg, ranging from the ftfth to 
 the seventh century.' 
 
 [J. This letter is not now used. In older critical editions 
 three different MSS. might be found cited under 
 it, viz. 
 
 1. For the Gospels, the MS. here described under N. 
 
 2. For the Acts, the MS. described under G^. 
 
 3. For the Cathohc Epistles, the MS. described under 
 
 K. I. Cod. Cyprius [IX]. Imperial Library in Paris. 
 The four Gospels complete. 
 
 2. Cod. Mosquensis [IX]. Library of the Holy Synod 
 at Moscow. The Catholic Episdes entire; and 
 S. Paul's Epistles, with two hiatus, one of which 
 extends to five verses only. 
 
 L. I. Cod. REGIUS [VIII or IX]. Imperial Library in 
 Paris. The four Gospels, with four small hiatus. 
 
 2. Cod. Angelicus [IX]. Rome. That portion of 
 Cod. G2 (see above) which contains the Pauline 
 Epistles down to Heb. xiii. lo. 
 
CHIEF UNCIAL MSS. IT 9 
 
 M. I. Cod. Campianus [IX or Xj. Imperial Library in 
 Paris. The four Gospels complete. 
 2. Cod. RUBER [X]. Fragments of the two Epistles to 
 the Corinthians and of the Epistle to the Hebrews, 
 amounting to 196 verses in all. Two folio leaves 
 are at Hamburg, in the Johanneum ; and parts of 
 two more in London, at the Library of the British 
 Museum. 
 
 N. Cod. PURPUREUS [VI or VII]. Three fragments of 
 the Gospels of S. Matthew, S. Luke, and S. John 
 are cited under this designation. Four leaves are 
 in the British Museum, six at the Vatican, and two 
 at Vienna. These fragments used to be cited 
 separately as J, N, and F respectively. 
 
 N^\ [IV or V]. Some palimpsest fragments of S. John 
 in the British Museum, brought from a Nitrian 
 monastery. 
 
 Oa,...Of. Copies of the Evangelic Hymns (Magnificat, &c.) 
 found in Psalters at different places. There are 
 seven such, varying between the sixth and ninth 
 centuries. 
 
 P. Cod. GUELPHERBYTANUS I [VI]. The Ducal Li- 
 brary at Wolfenbiittel. A palimpsest containing 
 fragments of the Gospels. 
 
 Q. Cod. GUELPHERBYTANUS II [VI]. A MS. of the 
 same place, date, and character as P, but containing 
 fragments only of S. Luke and S. John. 
 
 R. Cod. NITRIENSIS [VI]. British Museum in London. 
 Large fragments of S. Luke. 
 
 S. Cod. Vaticanus 354 [X]. Vatican Library in Rome. 
 The four Gospels entire. 
 
3 20 LIST OF THE 
 
 T. Cod. BORGIANUS I [V]. Library of the Propaganda 
 in Rome. Fragments of S. Luke and S. John. 
 A Grseco-Sahidic MS. 
 
 U. Cod. Nanianus [IX or X]. Library of S. Mark's, 
 Venice. The four Gospels entire. 
 
 V. Cod. Mosquensis [VIII or IX]. Library of the Holy 
 Synod, Moscow. The four Gospels, but mutilated. 
 It is written stichometrically. 
 
 X. Cod. MONACENSIS [IX or X]. University Library in 
 Munich. The four Gospels, but much mutilated. 
 
 Y. Cod. BARBERINI 225 [VIII]. Barberini Library in 
 Rome. A fragment containing 137 verses of 
 S. John. 
 
 Z. Cod. DUBLINENSIS RESCRIPTUS [VI]. Library of 
 Trinity College, Dublin. A palimpsest fragment, 
 with 290 verses of S. Matthew's Gospel. 
 
 r. Cod. Tischendorfianus IV [IX]. A codex of the 
 four Gospels, complete except two passages of 
 S. Matthew and S. Mark : but part of it is in the 
 Bodleian Library at Oxford, part at Petersburg. 
 
 A. Cod. SANGALLENSIS [IX]. Library of the monastery 
 at S. Gall in Switzerland. A Grseco-Latin MS., 
 containing the four Gospels entire, except S. John 
 xix. 17-35, with an interlinear Latin translation. 
 (See above under G^.) 
 
 e. Cod. TISCHENDORFIANUS I [VII]. University 
 Library at Leipsic. A few fragments of S. Mat- 
 thew. 
 
CHIEF UNCIAL MSS. J 21 
 
 A. Cod. Tischendorfianus III [VIII or IX]. Bodleian 
 I/.brary at Oxford. The Gospels of S. Luke and 
 S. John enUre. 
 
 r. Cod. ZACYNTHIUS [VIII]. Library of the British 
 and Foreign Bible Society in London. A palimp- 
 sest, containing considerable portions of S. Luke's 
 Gospel, with a catena. 
 
 n. Cod. Petropolitanus [IX]. Petersburg. Contains the 
 Gospels nearly entire. 
 
 There are besides a number of small fragments referred 
 to by Tischendorf. It has been thought needless to insert 
 them in the above list. Any one using his last edition, in 
 which they are cited, will find there all the necessary informa- 
 tion about them. 
 
APPENDIX D. 
 
 A LIST OF THE LATIN CODICES MOST COMMONLY CITED IN 
 CRITICAL EDITIONS. 
 
 The following nomenclature will be found to differ con- 
 siderably from that given by Professor Westcott in his article 
 ' Vulgate ' in Smith's Dictionary of the Bible. Without pre- 
 suming to give an opinion on the merits of one or the other, 
 we have chosen this ; because, as it is the nomenclature used 
 by Tregelles, Tischendorf, and Alford in their editions of the 
 Greek Testament, there is at all events a necessity for the 
 student to be acquainted with it. 
 
 I. Of the Text before S.ferome's Revision. 
 
 a. Cod. Vercellensis [IV]. At Vercelli. The four Gospels, 
 
 but much mutilated. Probably the best example of 
 #> the Vetus Latina. 
 
 b. Cod. Veronensis [IV or V]. At Verona. The four 
 
 Gospels, with several hiatus. A good example of 
 the Vetus Latina. 
 
 c. Cod. Colberthius [XI]. At Paris. In the four Gospels 
 
 it is a very pure specimen of the Vetus Latina : the 
 rest of it is by a different hand, and gives S. Je- 
 rome's text. 
 
A LIST OF THE LATIN CODICES, ^r. 1 23 
 
 Is the Latin version of I)^ (see p. 115). Of little 
 critical importance, except where the Greek is 
 
 2. Is the Latin version of D.^ (p. 116). This is of more 
 critical value than d^, and appears to be a specimen 
 of the Vetus Latina. 
 
 . I. Cod. Palatums [V]. At Vienna. A MS. much muti- 
 lated, containing fragments only of SS. Matthew 
 and Mark, and very nearly the whole of SS. Luke 
 and John. The Gospels stand in the order SS. 
 Matthew, John, Luke, Mark. An example of the 
 Vetus Latina slightly altered. 
 
 2. The Latin version of E_, (see p. 116). 
 
 3. The Latin version of Eg (see p. 116). 
 
 f. Cod. Brixianus [VI]. At Brescia. The four Gospels, with 
 only two hiatus in S. Mark. Supposed to be an 
 example of the Versio Itala, or North Italian re- 
 cension of the Vetus Latina. 
 
 ff^iff'. Codd. Corbeienses. Described as 'very ancient;" 
 but no exact date given. They take their name 
 from the Abbey of Corbey in Picardy, to which 
 they once belonged, ff^ is now at Petersburg, 
 and contains S. Matthew's Gospel and the Epistle 
 of S. James. There is some doubt apparently 
 whether or not it contains the first five chapters of 
 S. Mark, ff- contains the four Gospels almost 
 entire. The text is mixed ; i. e. the Vetus Latina 
 altered by some independent corrector. 
 
 §^, g'- Cpdd. Sanger }?iafienses. Very ancient. They contain 
 
124 ^ LIST OF THE LATIN CODICES 
 
 the four Gospels (perhaps a little mutilated). 
 A mixed text. 
 
 £■. The Latin version of G3. (See p. 117). The Epistles 
 of S. Paul. 
 
 k. Cod. Claromontajius [IV or V] . Vatican Library at 
 Rome. The Gospel of S. Matthew in the Vetus 
 Latina ; the other three in S. Jerome's Revision. 
 
 i. Cod. Vindoboftensis [V or VI]. Vienna. Portions 
 of S. Mark and S. Luke. A very valuable example 
 of the Vetus Latina. 
 
 k. Cod. Bobbiensis [IV or V]. Turin. Fragments of 
 S. Matthew, and one of S. Mark. An example of 
 the Vetus Latina revised. 
 
 /. Cod. Rhedigerianus [VII]. Breslau. The four 
 Gospels, mutilated. A mixed text. 
 
 ;//. Cardinal Mai's Speculum [VI]. Monastery of S. Croce 
 at Rome. Contains extracts from almost all the 
 books of the New Testament. The text accords 
 with the Vetus Latina. 
 
 n. Cod. Sangallensis [IV or V]. St. Gall. Fragments 
 of SS. Matthew and Mark. Vetus Latina. 
 
 0. [VII]. A fragment of S. Mark, and 
 
 p. [VIII]. A fragment of S. John. Both at St. Gall. 
 
 q. Cod. Monacensis [VI]. Munich. Fragments of each 
 of the Gospels. According to Professor Westcott 
 an example of the Versio Itala. Smith's Diet, of 
 Bible, art. 'Vulgate,' vol. iii. p. 1694. 
 
COMMONLY CITED IN CRITICAL EDITIONS. 1 25 
 
 r. Cod. Fn'si?igensis [V or VI]. Munich. Fragments 
 of S. Paul's Epistles. 
 
 s. Another Cod. Bobbiensis [V ?]. Vienna. Frag- 
 ments of the Acts and Catholic Epistles. 
 
 gue. Cod. Giielpherhytaniis [VI]. Wolfenbiittel. A palimp- 
 sest, containing a fragment of about thirty-three 
 verses of the Epistle to the Romans. 
 
 h. The interlinear Latin version of A. (See above, 
 p. 120). 
 
 2. Of S. Jerome s Revision. 
 
 Only a few of the best known are here mentioned. 
 
 am. Cod. Amiatinus [VI]. Laurentian Library at Florence. 
 Old and New Testament nearly perfect. 
 
 for. Cod. Forojuliensis [VI]. At Friuli. Gospels of SS. 
 Matthew and Luke, and nearly the whole of S. 
 John. Part of S. Mark's Gospel is at Venice, and 
 part at Prague. 
 
 ful. Cod. Fuldensis [VI]. Abbey of Fulda in Hesse 
 Cassel. The whole of the New Testament. 
 
 harl. Cod. Harleianus [VII]. The Gospels. 
 
 pe. or per. Fragmenta Periisina. Very ancient. At Perugia. 
 Fragments of S. Luke. 
 
 prag. Under this designation Tischendorf cites the portion 
 of Cod. Forojuliensis said above to be at Prague. 
 
 tol. Cod. Toletaniis [VIII]. Cathedral Library at Toledo. 
 Old and New Testament written in Gothic cha- 
 racters. 
 
APPENDIX E. 
 
 A LIST OF FATHERS WHOSE WRITINGS ARE OF IMPORTANCE IN 
 THE CRITICISM OF THE TEXT OF THE NEW TESTAMENT. 
 
 Those whose works are in Latin are printed in italics : 
 those which are of primary importance are in capitals. With 
 the exception of a very few writers, who are frequently cited 
 in critical editions, none are inserted of later date than the 
 fourth century. In each case the century is given to which 
 the working-life of the writer belonged. This must be re- 
 membered in comparing these notices with some lists in 
 which the year of the birth or death only is given. A general 
 description of the works of each is added. 
 
 Ambrose, Bishop of Milan [IV]. Commentaries, Ser- 
 mons, Epistles, and Treatises on various ecclesiastical 
 subjects. 
 
 AMBROSIASTER : perhaps Hilary the Deacon [IV]. 
 So called because his Commentaries on S. Paul's 
 Epistles were frequently published among the works 
 of S. Ambrose. 
 
 ANDREAS OF CAPPADOCIA, Bishop of Csesarea in 
 Cappadocia [VI]. A Commentary on the Apoca- 
 lypse. (Not to be confounded with Andreas of 
 Crete, a writer of the next century.) 
 
A LIST OF THE FATHERS COMMONLY CITED. 12 
 
 ATHANASIUS, Archbishop of Alexandria [IV]. Ora- 
 tions, Epistles, and Treatises, chiefly on subjects 
 connected with the Arian controversy. 
 
 AUGUSTINE, Bishop of Hippo [IV]. His works are 
 very numerous. The most important are, his great 
 work de Civitate Dei, his Confessions and Retracta- 
 tions, and his Commentary on the Psalms. There are 
 besides many Letters and Sermons, as well as Con- 
 troversial and Philosophical Treatises. He appears 
 to use the Versio Itala in his quotations. 
 
 BASIL THE GREAT, Bishop of Csesarea in Cappa- 
 docia [IV]. Homilies, Ascetic writings, Letters, and 
 some Treatises on special subjects. 
 
 Cassiodorus [VI]. At first a statesman, then a monk, of 
 Italy. His works are various; Historical, Literary, 
 and Scientific Treatises, as well as others expository, 
 or illustrative of the Scriptures. 
 
 CHRYSOSTOINI of Antioch, Archbishop of Constantinople 
 [IV]. Homilies, Commentaries, Letters, and Trea- 
 tises on special subjects. 
 
 CLEMENT OF ALEXANDRIA [III]. His three prin- 
 cipal works are the Aoyoy TrpoTpenTiKos npos "EXXtjucis (a 
 Hortatory Address to the Gentiles), naiSayco-yo'?, and 
 ^TpoipuTfU (Miscellanies). There is also a short prac- 
 tical treatise, tis 6 aco^opevos ttXovo-ios ; 
 
 CYPRIAN, Bishop of Carthage [III]. A number of 
 short treatises on various subjects, apologetic, expo- 
 sitory, and controversial ; and a valuable collection 
 of Letters. 
 
 CYRIL OF ALEXANDRIA, Bishop [V]. Commentaries, 
 
128 A LIST OF THE FATHERS 
 
 Homilies, Letters, and Dialogues on some of the 
 chief Mysteries of the Faith. 
 
 Cyril of Jerusalem, Bishop [IV]. KaTTjxqa-eis, or Lectures 
 on the Faith and Doctrines of the Church to Cate- 
 chumens and Newly-baptized Persons. 
 
 Damascenus (Joannes) [VIII]. Numerous short trea- 
 tises on controversial, theological, and ecclesiastical 
 subjects. 
 
 DIDYMUS, of Alexandria [IV]. Likr de Spiritu Sancto, 
 de Trinifate, and adversus MmiichcEos, 
 
 Dionysius, Bishop of Alexandria [III]. Treatises, chiefly 
 controversial ; and Epistles. Only extracts and frag- 
 ments remain. 
 
 Ephraem Syrus [IV]. Treatises, theological and moral, 
 Homilies and Commentaries ; they are in Syriac, and 
 of use in connection with the Syriac versions. 
 
 EPIPHANIUS, Bishop of Constantia in Cyprus [IV]. 
 Ancoratus, on the doctrine of the Trinity ; Panarium, 
 a treatise against Heresies ; De Ponder thus et Men- 
 suris Liber. 
 
 EUSEBIUS OF C^SAREA, Bishop [IV]. His chief 
 works are the Chrom'con, Prceparatio Evangelica, 
 Demonstratio Evangelica, Historia Ecclesiastica, De 
 Martyribiis PalcBsUncz, De Vita Constantini, Onomas- 
 iicon, and several controversial treatises. 
 
 Euthymius Zigabenus [XII]. A Greek monk of Con- 
 stantinople. His chief work for our purpose is a 
 Commentary on the Four Gospels, compiled from 
 the writings of S. Chrysostom and other early 
 Fathers. 
 
COMMONLY CITED. IH) 
 
 Gregory of Nazianzus, in Cappadocii, Bishop [IV]. Ser- 
 mons, Letters, and Poems. 
 
 Gregory of Nyssa, in Cappadocia, Bishop [IV]. Trea- 
 tises, doctrinal and practical ; Discourses, Letters, 
 Biographies. 
 
 Gregory Thaumaturgus, Bishop of Neocaesarea [III]. A 
 Paraphrase of Eccksiasies, an Explanation of the 
 Creed, an Epistola Canonica, and a Panegyrical Ad- 
 dress to Origefi, are his extant works. 
 
 Hilary of Poictiers {Pictavensis), Bishop [IV]. His chief 
 work is De Trinitate Lihri XII. He wrote Com- 
 mentaries on the Psalms and on S. Matthew's Gos- 
 pel. . Several smaller treatises are extant. 
 
 Plippolytus, Bishop of Portus [III]. Fragments only of 
 his ^vorks remain, which are partly controversial, 
 partly expository. 
 
 IRENiEUS, Bishop of Lyons [II]. Only one work of his 
 remains, Adversus Hcereses ; and of this only frag- 
 ments of the original Greek are extant. But there 
 is an old Latin translation, apparently contempora- 
 neous with the original. The translator gives the 
 quotations from Scripture in the Vetus Latina : hence 
 the authority of S. Irenseus is of service in the criticism 
 both of the Greek and Latin texts. The original 
 and the translation are always cited separately, thus : 
 Iren(/^jv/), and Iren(m/.). 
 
 JEROME [IV]. Epistles, which are chiefly disquisi- 
 tions on various Theological or INIoral questions ; 
 Tracts, biographical or polemical ; Commentaries ; 
 the Chronica Eusehii, translated and extended; the 
 Bibliotheca Divina, which is the result of his critical 
 labours on the Text of the Old and New Testaments. 
 
130 A LIST OF THE FATHERS 
 
 Justin Martyr [II]. Two Apologies for the Christians, 
 addressed to Antoninus Pius and Aurelius respectively ; 
 and a Dialogue with Trypho, a Jew. 
 
 Lactantius [IV]. Divince InstituHones, a philosophical 
 introduction to Christianity, against the pagan system. 
 An Epitome of the same, and two or three other 
 smaller pieces. His works are useful in the study of 
 the Vetus Latina version. 
 
 LUCIFER OF CAGLIARI (Calaritanus), Bishop [IV]. 
 Several treatises on questions of dogma and disci- 
 pline arising out of the Arian controversy. Useful 
 in consequence of the numerous quotations from the 
 Vetus Latina version of the Scriptures. 
 
 [Marcion of Pontus, the Heretic [II]. None of his works 
 survive entire, but there are many quotations in the 
 writings of Tertullian and Epiphanius, which are cited 
 as Marcion-tert.^ Marcion-^pip^., respectively.] 
 
 Methodius Patarensis, Bishop [III]. Treatises on Free- 
 will, the Resurrection, and Virginity. 
 
 CEcumenius, Bishop of Tricca in Thrace [X]. Commen- 
 taries on all the books of the New Testament but 
 the Gospels. 
 
 ORIGEN [III]. The Tetrapla and Hexapla editions of 
 the Old Testament; exegetical works, in the forms 
 of Commentaries, Scholia, and Homilies. Of the rest 
 of his voluminous writings only a few letters and 
 extracts remain. 
 
 Rufinus of Aquileia [IV]. An Exposition of the Apostles' 
 Creed. An Ecclesiastical History. A collection of 
 Biographies ; and several other original works, as 
 well as numerous translations of Greek works, among 
 
COMMONLY CITED. I31 
 
 \vhich are the Homilies of Origen, the works of 
 Gregory of Nazianzus, and the Recognitions of 
 Clement of Rome. He was a contemporary of 
 S. Jerome. 
 
 TERTULLIAN, of Carthage [H and III]. Numerous 
 treatises on various points of order and discipline : 
 some also controversial. His quotations of Scripture 
 are from the Vetus Latina. 
 
 Theodoret, Bishop of Cyrus or Cyrrhus in Syria [V]. 
 His works were partly exegetical, including a Com- 
 mentary on S. Paul's Episdes, partly historical, and 
 partly controversial. 
 
 Theodore of Mopsuestia (in Cilicia), Bishop [V]. His 
 chief works were exegetical. His Commentaries 
 on the Twelve INIinor Prophets are extant entire. 
 Fragments only of his Commentaries on the Books 
 of the New Testament remain, in catenae. 
 
 Theophylact, Archbishop of Bulgaria [XI]. Commenta- 
 ries, founded on those of S. Chrysostom. 
 
 Victor Antiochenus [V]. Commentaries, of which frag- 
 ments remain extant in catenae. 
 
 Victorinus [IV]. Commentaries on the Epistles to the 
 Galatians, Philippians, and Ephesians. His quota- 
 tions are from the Latin before S. Jerome's Re- 
 vision. 
 
 K 2 
 
INDEX I. 
 
 OF TEXTS OF THE NEW TESTAMENT REFERRED TO 
 OR DISCUSSED IN THE COURSE OF THIS WORK. 
 
 S. Matthew 
 
 
 
 
 PAGE. 
 
 S. Mark. 
 
 page 
 
 i. iS 
 
 ... 61 
 
 x. 18 . 
 
 . 72 
 
 i. 24 
 
 
 
 
 59 
 
 x. 30 . 
 
 • 17 
 
 iv. 13 
 
 
 
 
 45 
 
 xiv. 30, 68, 72 
 
 20 
 
 vi. I 
 
 
 
 
 90 
 
 XV. 28 . 
 
 • 32 
 
 vi. 13 
 
 
 
 
 21 
 
 xvi. 8 . 
 
 . 29 
 
 xi. 16 
 
 
 
 
 17 
 
 xvi. 9-20 32,35,39,95,110,111 
 
 xii. 20 
 
 
 
 
 17 
 
 
 
 xii. 46 
 
 
 
 
 14 
 
 S. Luke. 
 
 
 xiii. 2 
 
 
 
 
 14 
 
 i. I, 2 
 
 . 97 
 
 xiii. 14 
 
 
 
 
 17 
 
 ii. 33 • 
 
 22 
 
 XV. 8 
 
 
 
 
 74 
 
 iii. 14 
 
 17 
 
 xvi. 19 
 
 
 
 
 40 
 
 V. 4-7 . 
 
 . 29 
 
 xvii. 2 
 
 
 
 
 20 
 
 vi. 48 . 
 
 . 19 
 
 xvii. 21 
 
 
 
 
 22 
 
 vii. 31 . 
 
 21 
 
 xix. 17 
 
 
 
 20, 
 
 22,73 
 
 ix. 49 . 
 
 . 16 
 
 XX. 22 
 
 
 
 
 74 
 
 X. 20 . 
 
 • 17 
 
 XX. 28 
 
 
 
 
 75 
 
 x. 22 
 
 21 
 
 xxi. 28-3 
 
 I 
 
 
 
 104 
 
 xi. 2-4 . 
 
 • 75 
 
 xxiii. 32 
 
 
 
 
 17 
 
 xviii. 19 . 
 
 . 72 
 
 XXV. 16 
 
 
 
 
 92 
 
 xviii. 25 . 
 
 . 17 
 
 xxvi. 39 
 
 
 
 
 16 
 
 xxii. 43, 44 . 
 
 3^, 103 
 
 xxvi. 52 
 
 
 
 
 ^9 
 
 xxiii. 10 . 
 
 • 14 
 
 
 xxiv. I . 
 
 . 17 
 
 S. Mark. 
 
 
 
 ii. 22 . . . . 91 
 
 S. John. 
 
 
 V. 29 . . . .17 
 
 i. 13 
 
 . 18 
 
 ix. 3 . . . .20 
 
 i. 15 
 
 • 17 
 
 ix. 29 
 
 
 
 
 22 
 
 i. 18 . 
 
 22 
 
134 
 
 
 
 INDEX I. 
 
 S. John. page. 
 
 Galatians. 
 
 V. 3, 4 . . . .102 
 
 V. 21 
 
 vi. 39 . . . . i6 
 vii, 8 . . . .22 
 
 Ephesians. 
 
 vii. 37 .... 14 
 
 i. I 
 
 vii. 53— viii. II 36, 39' 53. 97 
 xii. 6 . . . .18 
 xxi. 1-6 .... 29 
 
 1 Thessalonians 
 
 i-3 
 ii. 7 
 
 Acts. 
 
 I Timothy, 
 
 vii. 42 
 
 
 
 • 17 
 
 iii. 16 
 
 viii. 37 
 
 
 
 . 9,21,63 
 
 
 ix. 4 
 
 
 
 . 26 
 
 Hebrews, 
 
 X. 30 
 
 
 
 22 
 
 ix. 14 
 
 xi. 20 
 
 
 
 . 109 
 
 X. 23 . 
 
 xiii. 23 
 
 
 
 . 16 
 
 xii. 20 . 
 
 XV. 24 
 
 
 
 . 18 
 
 
 XV. 34 
 
 
 
 . 18 
 
 I S, Peter. 
 
 XX. 28 
 
 
 
 22, 106 
 
 ii. 3 
 
 xxi. 31 
 
 
 
 . 19 
 
 iii, 13 . 
 
 XX vi. 14 
 
 
 
 20 
 
 
 
 I S. John. 
 
 Romans. 
 
 ii. 23 . 
 
 vii. II . . . .45 
 
 iv. 3 
 
 viii. 28 . . . .90 
 
 V. 7 
 
 xiii. 9 . . . .20 
 
 
 
 Apocalypse. 
 
 I Corinthians. 
 
 i. 1 
 
 vii. 5 . . . .22 
 
 i. 4 
 
 xi. 29 . . . . 90 
 
 i, 15-20 . 
 
 xii. 28 . . . -45 
 
 ii,9 . 
 
 
 vi. I, 3> 5, 7 
 
 2 Corinthians. 
 
 xii. 17 
 
 viii. 4, 5 
 
 
 
 . 18 
 
 xvii. 8 . 
 
 PAGE. 
 
 35 
 
 35.39 
 
 16, 59' 99 
 
 39 
 45 
 
 20 
 
 16 
 
 63 
 9,92 
 
 18 
 
 19 
 
 9 
 
 18 
 18 
 63 
 
INDEX II. 
 
 GENERAL. 
 
 Acts of the Apostles, modes of 
 
 dividing, 31. 
 Adamantius (Origen\ 82. 
 ^thiopic Version, 56. 
 Africa, 46, 103. 
 African group of MSS., 69. 
 Alcuin's revision, 48. 
 Alexandrine group of MSS., 69. 
 — readings, 79, 84. 
 Alford, Dean, 3 «., 4, 89, loS, 
 Amialinus Codex, 50. 
 ' Ammonian' Sections, 28-30. 
 Ammonius, 28. 
 dvayvuiaeis (. . crfxaTa), 32. 
 Andreas of Cappadocia, 32, 126. 
 dvTipaWiiv, 14. 
 avTiXcyu/xeva, 36, 97. 
 Antioch, 84. 
 
 Antiquity of a Text, test of the, 80. 
 Apocalypse, divisions of, 32. 
 Apostrophus, the, 39. 
 ArdiicB lediones, 90. 
 Argenteus Codex, 56. 
 Armenian Version, 56. 
 Asiatic group of MSS., 69. 
 Assimilation of terminations, 23 n. 
 Athanasius, 82, 127. 
 ' — Epistle of, to Marcellinus, 114. 
 Augustine, 72, 127. 
 ' Authentic,' meaning of, 97. 
 
 Baptismal profession of faith, 21. 
 Barnabas, Epistle of, 34, 100, 
 Barsalibfei Codex, 53. 
 Basmuric Version, 54, 55. 
 Bengel, 69, 90. 
 Bentley, 3, 68, 86. 
 Beza, 10, 
 
 I Bezae Codex, 10, 115. 
 
 I Books of the New Testament, order 
 
 of, 25. 
 Britain, revision of Latin Versions 
 
 in, 47. 
 Brixianus Codex, 51, 55, 79. 
 Buddhists, Sacred Books of the, 
 
 7«. 
 Burgon, Mr., 29, no, in. 
 Buttmann, 3. 
 Byzantine group of MSS., 9 «., 69. 
 
 — readings, 79, 84, 85. 
 
 Csesarea, 84. 
 
 Canonical, S. Mark xvi. 9-20, 97. 
 
 Canonicity, 86, 99. 
 
 Canons, Eusebian, 29, 32, 114. 
 
 — of criticism, value of, 87. 
 Catena, Cramer's, 59. 
 Characteristic expressions, 92. 
 Chrysostom, 59, 127. 
 Claromontanus Codex, 84. 
 Clement of Rome, Epistles of, 113. 
 Clementine Vulgate, 40. 
 Collations of Codex B, 3 7. 
 Comparative criticism of New Tes- 
 tament defined, I. 
 
 — problems of, i. 
 
 — opposite views of, 2. 
 
 — appHed to secular writings, 23. 
 Complutensian Edition, 8. 
 'Conflate' readings, 107. 
 Conformity, alterations to produce, 
 
 19. 
 Confusion of letters, 16, 17. 
 Conjectural emendation, 6. 
 Conquest, Mohammedan, of Egypt 
 
 and Syria, 84. 
 
135 
 
 INDEX II. 
 
 Constans, the Emperor, 32. 
 Constantine, 34, 39, 82, 83. 
 Constantinople, 83, 84. 
 Coptic Version, 54. 
 Copyists, tendency of, to assimilate 
 passages, 73, 92. 
 
 — to supply supposed defects, 74, 76. 
 
 — to include everything in their 
 
 copy, 89. 
 Corrections, evidence derived from, 
 
 43- 
 
 — of unclassical forms, 1 9. 
 
 — S. Jerome's, 49. 
 
 — often unintelligent, 14. 
 
 — a source of error, 15, 23 n. 
 Correctors, 14. 
 
 — mode of designating, 26. 
 
 — of Codex «, 35. 
 
 — of Codex B, 39, 40. 
 Cureton, Dr., 51. 
 
 Curetonian Syriac Version, 50, 70. 
 Cursive MSS., how denominated, 25. 
 
 — number of, fully collated, 27. 
 Cyril of Alexandria, 59, 127. 
 
 Date of a MS., arguments for fixing, 
 
 36, 113. 
 Dated evidence, amount of known, 
 
 65, 66. 
 Diatessaron, Tatian's, 20. 
 Diocletian's persecutions, 54, 83. 
 SiopOovu, biOpOoor-qs, 1 4. 
 'Diplomatic' evidence, 5. 
 ' Documentary ' evidence, 7. 
 Dogmatic alterations of the Text, 
 
 15, 22. 
 Doubling of letters, &c., 18, 23 n. 
 Doxology of the Lord's Prayer, 
 
 Editio Regia, 10. 
 
 Eichhorn, 69. 
 
 Einleitung, Hug's, 69. 
 
 Ellicott, Bishop, 4. 
 
 Elzevir, 10. 
 
 English Version, readings adopted 
 
 in the, 10 n. 
 Epistles, modes of dividing the, 31. 
 Erasmus, his editions, 9-1 1. 
 
 — interpolations by, 9. 
 
 — MSS. used by, 9. 
 
 Errors, of sight, 15, 16. 
 
 — of sound or hearing, 15, 17, 18, 
 
 — of memory, 15, 18. 
 
 — tendency to accumulate, 22. 
 
 — Prof. Madvig's classification of, 
 
 23 ti. 
 Eusebius of Ceesarea, 29, 32, 34, 51. 
 
 59, 6r, 82, 107, 128. 
 Euthalius, 28, 32. 
 Evangelistaria, 25. 
 Evidence, canons of external, 88. 
 
 — canons of internal, 89, &c. 
 
 — dated, amount of, 65, 66. 
 
 — sources of, 6. 
 
 Friderico-Augustanus Codex, 14, 
 
 33. 113- 
 Froben, 9. 
 Fuldensis Codex, 50. 
 
 Galilean Psalter, the, 46. 
 
 ' Genuine,' 97. 
 
 Glosses, a source of error, 15, 18, 
 
 23 n. 
 Gospels the, systems of divisions of, 
 
 28. 
 
 — Western order of, 115. 
 Gothic Version, the, 55. 
 
 — canon of, 55. 
 Grseco-Latin codices, 69. 
 
 Greek spoken everywhere at the 
 
 time of our Saviour, 46. 
 Greek Testament, first pi'inted, 8. 
 
 — first published, 9. 
 
 Green, Mr., Developed Course of 
 
 Criticism of, 71. 
 Griesbach, „ „ 69, 89, 90. 
 Groups of copies, characteristics of, 
 
 67. 
 
 — how formed, 22. 
 
 — number of, 68, 69. 
 
 — relation of to the true Text, 70, &c. 
 
 Haddan and Stubbs' Councils of 
 
 Great Britain, 47 n. 
 Harclean Version, 52, 53. 
 
 — how quoted, 53. 
 Harkel, Thomas of, 52. 
 Hebrews, the Epistle to the, sections 
 
 of, 3i> 32. 
 
 — position of, 36. 
 
INDEX II. 
 
 ^^7 
 
 Hermas, the Shepherd of, .^4. 
 Hesychius, 81. 
 Hierapolis, 52. 
 Homoiotelettton, 16, 23 n. 
 Hort, Rev. F. J. A., 3. 
 Hug, 69, 8r. 
 
 'Irjaovs XpiffTus 6, collocation no- 
 where found in the New Testa- 
 ment, 63. 
 
 Ignatius, 100. 
 
 Inflexion, peculiarities of, 67. 
 
 Initial letters, 36, 38, 113, 114. 
 
 Insertion of similar letters, 16. 
 
 Interpolations, an occasional source 
 of evidence, 75. 
 
 — in Codex Bez*, &c., 115. 
 Irenaeus, how quoted, 129, 
 Itala Versio, 47. 
 
 Itala unci Vulgata (Ronsch's), 64 n. 
 Itacisms, 17. 
 
 — of Codd. N and B, 43. 
 
 Jerome, his revision, 48, 49, 61. 
 Jerusalem-Syriac Version, 53. 
 Justin Martyr, 61 «., 130. 
 
 Karkaphensian Version, 53. 
 
 — canon of, 54. 
 K€((>dKaia, of the Gospels, 30. 
 
 — of the Acts and Epistles, 32. 
 
 — of the Apocalypse, 32. 
 Kuenen and Cobet's edition of 
 
 Cod. B, 15, 19, 37. 
 
 Lachmann, 3, 69. 
 
 Latin group of MSS., 69, 70. 
 
 Lectio prce/eratur brevior, 19, 89. 
 
 Lectionaries, 21, 25. 
 
 Lectioni proclivi prcestat nrdua, 90. 
 
 Letters, similar, confused, 16. 
 
 — „ omitted or inserted, 16. 
 
 — transposed, 16. 
 
 Lightfoot, Professor, on revision of 
 
 New Testament, 3 «., 99. 
 Liturgical insertions, 15, 21, 
 \6yoi of the Apocalypse, 32. 
 Lord's Prayer, the, 75. 
 
 — the doxology df, 21. 
 Lucian, 81. 
 
 Madvig, Professor, 13 «., 23. 
 Manuscripts, a source of evidence, 6. 
 
 — dates of, 65. 
 
 — different, denoted by the same 
 
 letter, 26. 
 
 — gross total number of, 24. 
 
 — groups of, 67. 
 
 — mode of copying, 13. 
 
 — mode of determining the dates of, 
 
 — palaeographic characteristics of, 
 
 27. 
 
 — the same denoted by different 
 
 letters or numerals, 26. 
 Marcion, 60 n., 130, 
 Mark, S., Gospel according to, 80. 
 Memory, errors of, 18. 
 Memphitic Version, 54. 
 Milan, Edict of, 83. 
 Mill, 68. 
 Mohammedan conquest of Egypt 
 
 and Syria, 84. 
 
 New College Library, 52. 
 
 Octateuch Septuagint, 117. 
 Omission of similar letters, 16. 
 Origen, 60, 61, 72, 74, 76, 130. 
 
 Papyrus rolls, 34. 
 Paradiplomatic evidence, 4. 
 Parallel passages, 20. 
 ■jrapaTrXi^aia ra, 29. 
 Particles accidentally varied, 18. 
 Paul, S., speech of to the Ephesian 
 
 elders, 108. 
 Permutation of letters, 23 n. 
 Peshito Syriac Version, the, 51. 
 
 — canon of, 51. 
 Philoxenian Version, the, 52. 
 Philoxenus, 52. 
 
 Polycarp, Chorepiscopus of Hiera- 
 polis, 52. 
 
 Praxapostoli, 25. 
 
 Proclivi lectioni prcestat ardua, 90. 
 
 Profession of Faith, the Baptismal, 
 21. 
 
 Punctuation, rare in early MSS., 27. 
 
 — of Cod. A, 113. 
 
 — of Cod. B, 41. 
 
138 
 
 INDEX 11. 
 
 Qttaterniones, 34, 39. 
 Qniniones, 39. 
 
 Quotations, a source of evidence, 6, 
 58 seq. 
 
 — altered by transcribers, 58, 59. 
 
 — value of, 61. 
 
 Recensions, 81. 
 
 Ridley, Dr. Gloucester, 51. 
 
 Ricu, Dr., 56. 
 
 Ronsch, 64 n. 
 
 Sahidic Version, the, 54. 
 
 Scholz, 69. 
 
 Scriptorium, 13. 
 
 Scrivener. Mr., 2, 24, 27, 62, 98, 99 n. 
 
 — his Greek Testament, 3 n. 
 Sections, ' Ammonian.' 28-30. 
 Similar letters confused, omitted, or 
 
 inserted, 16. 
 Sinaiticus. Cod., 14, 25. 33 seq. 
 
 — its connexion with Cod. B, 41, 42. 
 Sixtine Vulgate, 48, 49. 
 
 Sources of evidence for the true 
 
 text, 6. 
 Spelling, peculiarities of, 67. 
 Stephens, 10. 
 OTixoi, 28. 
 
 Subjective arguments, 4. 
 Subscriptions of the Books of the 
 
 New Testament, 113. 
 Synonymous words substituted, 18. 
 Syntax, peculiarities of, 67. 
 
 Tatian's Diatessaron, 19. 
 
 Tattam, Archdeacon, 51. 
 
 Tc'Aos, marking the end of an ecclesi- 
 astical Lection, iii. 
 
 Terfiiones, 34, 39. 
 
 Tertullian's, Das neue Testament 
 (Ronsch), 64 n. 
 
 Text, current in the second century, 5. 
 
 — liability of, to depreciation, 12, 
 
 22. 
 
 j Text, test of the antiquity of a, 80. 
 : Textus Receptus, critical value of, 
 
 10, II. 
 I — origin of the name, 10. 
 I Thebaic Version, the, 54, 63. 
 j Tischendorf, 33, 38, 69, 91, 108. 
 Titles of the Books of the New Tes- 
 tament, simplicity of, in early 
 MSS., 36, 1 13. 
 titKol, 30, 36. 
 
 Transcriptions, successive, 22. 
 Transposition of letters, 1 6. 
 Tregelles, Dr., 2, 20, 21, 71, jS, 90, 
 99 n., 106. 
 
 Ulfilas, 55. 
 ! Unanimity of MSS. imaginary, 67. 
 i Uncial MSS., how denominated, 25. 
 
 Upsala, 56. 
 
 Various readings, the sources of, 
 
 classified, 15. 
 Vaticanus Codex, 14, 31, 37 seq. 
 
 — connexion of with Cod. ^,41,42, 
 
 — facsimile edition of, 38. 
 Vera exemplaria, 106. 
 Verbal dissidences, 92. 
 Vercellone, Signor, 37, 38. 
 Versions, a source of evidence, 6, 
 
 44. 45. 
 Vetus Latina Version, codices of, 47. 
 
 — critical use of, 49, 50. 
 
 — origin of. 46. 
 
 — Ronsch 's labours upon, 64 n. 
 Vulgate, the, critical use of, 49, 50. 
 
 — of S. Jerome, 48, 49. 
 
 — Sixtine and Clementine, 49. 
 
 Westcott, Professor, 3 «., 26 «., 92. 
 Western group of MSS., 69, 70. 
 
 — order of the Gospels, 115. 
 
 Ximenes, Cardinal, 8. 
 
DATE DUE 
 
 
 ^1 S 
 
 
 
 t>iA.A>ti£i 
 
 ft^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CAYLORO 
 
 
 
 PRtNTCOtNU.S.A