559
ancient texts are at least more obscure than elsewhere. Whether B ever contained the Apocalypse or not, it is now defective from Hebrews ix 14 onward. The loss is the greater because in the Apocalypse א has a text conspicuously inferior to its text of the other books, partly inherited from earlier more or less corrupted texts, partly due to increased licence of transcription; and, though A, more especially when it is supported by C, here proves itself entitled to considerable authority, it does but imperfectly supply the deficiency, and moreover the want of early and good versions other than the Latin is sensibly felt. Yet even here the number of variations in which it is difficult to come to a trustworthy conclusion is much smaller than might have been anticipated.
The sketch contained in the preceding pages may suffice to indicate the principal lines of criticism which have been followed in this edition. The aim of sound textual criticism must always be to take account of every class of textual facts, and to assign to the evidence supplied by each class its proper use and rank. When once it is clearly understood that, by the very nature of textual transmission, all existing documents are more or less closely related to each other, and that these relations of descent and affinity have been the determining causes of nearly all their readings, the historical investigation of general and partial genealogy becomes the necessary starting-point of criticism. Genealogical results, taken in combination with the internal character of the chief ancient texts or of the texts of extant documentary groups, supply the presumptions, stronger or weaker as the case may be, which constitute the primary and often the virtually decisive evidence for one reading as against another. Before however the decision as to any variation is finally made, it is always prudent, and often necessary, to take into consideration the internal evidence specially affecting it, both intrinsic and transcriptional. If it points to a result different from that which the documentary evidence suggested, a second and closer inspection will usually detect some hitherto overlooked characteristic of the best attested reading which might naturally lead to its alteration; while sometimes on the other hand reexamination brings to light an ambiguity in the attestation. No definite rule can be given in the comparatively few cases in which the apparent conflict remains, more especially where the documentary evidence
