INTRODUCTION.

ll. 16, 17 of this volume) which will have to be taken into account in future by any one who quotes the phrase in Matt. i. 21, "she shall bear to thee a son," as evidence for the supposed heresy of the Gospel-text in the Syriac Gospels of Mount Sinai.

As I desire that the distinguishing features of the texts in this book should be their antiquity, I have given a collation only of that portion of Dr Wright's text which is founded on MSS. not later than my own; and have left Dr Budge's text entirely alone. To have included them all might have made the work more complete, but it would have greatly increased both the size of the volume, and my own labour. Those who wish to know how the story developed in the fertile soil of pious minds, as it passed down the ages, must still consult Dr Budge's book.

Relation of the later script to the earlier one.

The 9th (or 10th) century Arabic translator (or editor) of these Selections from the Christian Fathers which form the upper script of this palimpsest, folded each leaf of the ancient Transitus MS. double, and then turned it half round, and wrote his Arabic text across the Syriac one (as our grandmothers used to cross their letters); thus making one Syriac leaf into two Arabic ones. This will be readily understood by reference to Plates VI. and VII. The numbering of the folios follows, of necessity, the Arabic text. To have followed the more ancient Syriac one would have resulted in confusion; for whilst the Arabic text forms a consecutive bound volume, the Syriac leaves are mingled at random, and are interspersed with leaves from manuscripts of a wholly different character.

The quires of this book are quaternions, i.e. each consists of four pairs of conjugate leaves, except the 11th which is a quinion, the 14th which is a ternion, and the 15th which has only two pairs of leaves. A single leaf, f. 103, is inserted between quire 13 and quire 14 whilst two leaves f. 114 and f. 115 come between quire 15 and quire 16. Strangely enough, the ancient text of the Transitus Mariae which these contain, falls into sequence with that on ff. 1, 2, the compiler of the 10th century book having evidently torn two ancient leaves in twain, and used up the four halves independently of each other.