r language. The chief of
these are the piece coming between the death of the herd Forgemen
and the fight with Cur Mac Dalath (including Cuchulainn's meeting
with Findabair, and the 'womanfight' of Rochad), and the whole of
what follows the Healing of the Morrigan. The tale is, like others
in this MS., unfinished, the MS. being imperfect.

(2) The Yellow Book of Lecan (YBL), a late fourteenth-century MS.
The _Tain_ in this is substantially the same as in LU. The
beginning is missing, but the end is given. Some of the late
additions of LU are not found here; and YBL, late as it is, often
gives an older and better text than the earlier MS.

(3) The Book of Leinster (LL), before 1160. The _Tain_ here is
longer, fuller, and later in both style and language than in LU or
YBL. It is essentially a literary attempt to give a complete and
consistent narrative, and is much less interesting than the older
LU-YBL recension.

In the present version, I have collated LU, as far as it goes, with
YBL, adding from the latter the concluding parts of the story, from
the Fight with Fer Diad to the end. After the Fight with Fer Diad,
YBL breaks off abruptly, leaving nearly a page blank; then follow
several pages containing lists, alternative versions of some
episodes given in LU (Rochad's Woman-fight, the Warning to
Conchobar), and one or two episodes which are narrated in LL. I
omit about one page, where the narrative is broken and confused.

The pages which follow the Healing of the Morrigan in LU are
altogether different in style from the rest of the story as told in
LU, and are out of keeping with its simplicity. This whole portion
is in the later manner of LL, with which, for the most part, it is
in verbal agreement. Further, it is in part repetition of material
already given (i.e. the coming of the boy-host of Ulster, and
Cuchulainn's displaying himself to the Connaught troops).

COMPARISON OF THE VERSIONS

A German translation of the Leinster text of the _Tain Bo Cuailnge_
will soon be accessible to all in Dr. Windisch's promised edition
of the text. It is therefore unnecessary to compare the two
versions in detail. Some of the main differences may be pointed
out, however.

Of our three copies none is the direct ancestor of any other. LU
and YBL are from a common source, though the latter MS. is from an
older copy; LL is independent. The two types differ entirely in aim
and method. The writers of LU and YBL aimed at accurac