In each hundred, the figures from 1 to 49 are reserved for MSS. containing the book, and those from 50 to 99 for those which do not contain it. This, however, necessarily applies only to MSS. of the eleventh century or later. Further, the four-figure symbols of the a group (a1000, etc.) are reserved for MSS. which contain only one of the three groups of books included under the general title of aπóσToλos, viz., Acts (including the Catholic Epistles), Pauline Epistles, and Apocalypse; in each 100, Nos. 1-19 denote MSS. containing only the Acts and Catholic Epistles, Nos. 20-69 MSS. containing only the Pauline Epistles, and Nos. 70-99 MSS. containing only the Apocalypse. Thus the numbers a1000-1019 are reserved for MSS. of the first ten centuries which contain only the Acts and Catholic Epistles, a11201169 for MSS. of the eleventh century containing only the Pauline Epistles, and a1270-1299 for MSS. of the twelfth century containing only the Apocalypse.

So much for manuscripts which contain the plain text of the New Testament or any part of it; but no less than twenty-four additional lists are necessary in order to deal with manuscripts which also contain commentaries. In these, again, the first figure of the number indicates the century to which the MS. belongs, while a letter prefixed to it shews the author of the attached commentary. Thus K50 is a fifteenth century MS. containing the commentary of Cyril of Alexandria on St. John, A' is the uncial MS. of the eighth century formerly known as E, which contains the Antiochene commentary on the Gospels, 25 is a thirteenth century MS. containing the commentary of Theophylact on the Gospels, and so on. Lectionaries are not included in von Soden's list, nor does he take any notice of the MSS. of the various versions.

The criticisms to be made on this system of numeration are obvious. Its merits are that it is indefinitely expansible, and that (if once you can master it) it gives you some information as to the date and contents of any MS. included in it. This advantage, however, is subject to considerable deductions. It does not apply to the first millennium, which is all lumped together in a single group. Yet it is

during this millennium that all the textual families worth considering came into existence. It is only when the age of a witness ceases to be of much practical importance that information on this head is given. It must be added that the information, when given, is only approximately trustworthy. The dating of Greek minuscules is far from being an exact science, and it is quite certain that any list so long as von Soden's, compiled by students of varying palaeographical experience, must contain many errors and more uncertainties.

These, however, are not the most serious objections to von Soden's system. They are only limitations of its usefulness, supposing its general principles to be accepted; and this is the point at issue. It is a very serious drawback, perhaps the most serious drawback of all, that it involves the abandonment of all the symbols by which New Testament MSS. have been familiar to scholars for generations. The symbols ABCD and many more have acquired a definite connotation, which pervades the work of textual critics since textual criticism rose to importance. It is no light thing for a scholar to claim the right to abolish all of these, and to make the writings of his predecessors unintelligible to coming generations. It is still more. objectionable when these same MSS. are also MSS. of the Septuagint, and retain their old nomenclature in the critical texts of these books. The Cambridge edition of the Greek Old Testament is quite as important an undertaking as von Soden's edition of the New, and there is every reason to hope that the study of the character of the leading MSS. in the one will throw light on their character in the other; but this branch of comparative criticism will be impeded if the same MSS. are known by different names in the principal critical editions.

The complexity of von Soden's scheme is another objection to it. A student working with it day after day may attain to a ready comprehension of it, and may know at once where to look for any MS. that he requires to find; but those who have only to refer to it intermittently will not find it easy to carry all its complexities in their minds,

and will be bewildered by a list in which or precedes I, and 070 follows 99, 1000 follows 073, and 1100 follows 199. Only an expert can readily find a manuscript in von Soden's list, and when he has found it, the descriptions are so brief that he will often have to look further to Gregory for fuller bibliographical information. The fact is that von Soden has tried to pack more information into his symbols than they can conveniently bear. A name should be a name and not a treatise. The attempt to pack the information of a paragraph into three or four figures is incompatible with simplicity.

In these circumstances it is not surprising that an effort has been made to save the advantages of the old system, while remedying its worst defects. The second book named at the head of this article is the outcome of a correspondence which Dr. C. R. Gregory undertook in the spring of last year with half a dozen students (German, English and American) of the text of the New Testament, and which afterwards expanded into a consultation of practically allnearly 100 in number-who are in any way publicly known in connexion with the subject in Europe and America. The scheme as originally outlined was modified in several details to meet criticisms and suggestions; but it is a remarkable testimony, both to the tact and conciliatoriness of Dr. Gregory and to the reasonableness with which his colleagues in all lands abandoned their special preferences for the sake of securing a general agreement, that he was able to arrive at a result which was almost universally accepted. With hardly a single exception, the New Testament scholars of Germany, America, Belgium, Denmark, England, France, Greece, Turkey, Holland, Norway, Austria-Hungary, Palestine, Roumania, Russia, Sweden, and Switzerland have rejected von Soden's numeration, and expressed their willingness to use in future the amended scheme put forward by Gregory.

The central principles of the new list are as follows:(1) the old symbols are retained for all the best-known and most important MSS. ; (2) means are provided for the future expansion of the list of uncials; (3) the minuscules are

thrown into a single list, so that each MS. in future will always have the same number, instead of a different number in different sections of the New Testament. In fuller detail, the following is the scheme. The Latin and Greek symbols (A-Z, г-N) are retained, and (as a special exception) for the Codex Sinaiticus; but all other Hebrew symbols are abolished, and the fragments which formerly were classed under the symbols I, O, T, W, and are removed, leaving those letters free to be assigned to certain substantial MSS. for which simple and handy symbols were required. Thus I goes to the recently discovered MS. of the Pauline Epistles, acquired by Mr. C. L. Freer in Egypt and now at Detroit, U.S.A.; O to the Paris fragment of St. Matthew from Sinope, written in gold letters on purple vellum; Tremains for the important Graeco-Sahidic fragment at Rome, formerly known as T, which ranks nearly with B for excellence of text; W is assigned to the Freer MS. of the Gospels, which has many claims upon our interest, besides its remarkable interpolation in the last verses of St. Mark; and ℗ to a Gospel MS. from Georgia, formerly included among the minuscules and numbered 1360. The majority of the letters are applied only to a single MS. in each case, namely, to the Gospel MS. to which it has hitherto belonged; but exceptions are made where a MS. of the Acts or Pauline Epistles is so well known that it would be inconvenient to give it a new name, as in the case of D and H of the Pauline Epistles, or E of the Acts. The letters which thus have double significations are DEFGHKLP, the several MSS. being distinguished as De and DP, E and Ea, H*, Ha, and HP, and so on. I

The uncial MSS. thus provided with letter-symbols are forty-five in all, and it is evident that there remain many which have been dispossessed of their previous symbols; these have to be provided for by some method which will allow of indefinite expansion in the future. The method proposed by Gregory, and ultimately accepted by his correspondents, is to use Arabic numerals with an 0 prefixed; and in order to minimize the risk of confusion with the numerals which denote minuscule MSS,, it is recommended

that the uncial symbols be printed in thick ('clarendon ') type. There are no less than 115 MSS. and fragments of MSS. to be thus provided for; but, in order to meet the possibility that future generations may prefer a uniform nomenclature throughout, the numbers from 01 to 045 are reserved as alternative symbols for the MSS. known by the letters to , and the remainder are numbered from 046 to 0161. Papyrus MSS. are not included in this numeration. They cannot always be classed either as uncial or minuscule; they have a peculiar history and form a class apart, which it is convenient to have separately marked. Consequently (following the example of the Oxford text of Homer) they are indicated by an antique p followed by a small number; thus, for example, the large Oxyrhynchus papyrus of Hebrews is to be known as pls.

With regard to the minuscules, the MSS. of the Gospels retain their old numbers. Those of the other books, if they also contain the Gospels, are known throughout by their number in the Gospel list. The others fill gaps in this list, or are added on at the end. Thus the MS. which formerly was Evan. 35, Act. 14, Paul. 18, Apoc. 17 will in future be 35 throughout; in Gregory's list it is printed 35eapr, in order that its contents may be clearly indicated, but in an apparatus criticus it would be simply 35. All the minuscules thus form a single list, extending from I to 2292. The lectionaries form a single list in the same manner, from 1 to 1540, with the letter 1 prefixed; if they contain the Apostolos they are marked in the list as 1; if they contain the Gospels as well, that can be indicated, if need be, by writing lea. As a rule, when it is not necessary to indicate the contents, the simple I will suffice.

The criticism which will most generally be made on Gregory's new system is, no doubt, that the symbols for the uncials which do not receive letters of their own are cumbrous, and that if a new MS. of the first importance should come to light, it would be very inconvenient to have to call it 0162, or something of the sort. No satisfactory method of meeting this difficulty has, however, been proposed, and in other respects the system seems satis