Bibliotheca Arnamagnæana - 01.10.1967, Blaðsíða 226

200
other instances. Since Miss Loth’s article will be ihe.locus classicus
for this kind of work, it may be an advantage to have as an appen-
dage to it the present examination of this other evidence, namely
the testimony of the leaves and gatherings of the manuscripts.
When the texts in an anthology were separated from eaeh other,
it often happened that the division came somewhere within a
gathering. Then it was necessary to cut some of the bifolia in two,
so that the first part of the divided gathering could go in one
manuscript and the second part in another. By examining the
gatherings in the separate manuscripts, and in particular observing
whether there are any single leaves at either the beginning or the
end, it is often possible to decide whether a given manuscript could
or could not have been next to another particular one, or at the
beginning or the end of the whole anthology.
An example may be taken from Jon Torfason’s manuscripts. It
is known from the contents that 540 was followed by 779c III, and
this is confirmed by the faet that 540 ends with three single leaves
and 779c III begins with the conjugate ff. 1+2 and then has three
single leaves, ff. 3, 4 and 5. Clearly the last three leaves of 540 and
the first five of 779c III were once a normal gathering of eight leaves
(four bifolia).
An examination of all the manuscripts in question shows that
the gatherings were uniformly regular with 8 leaves. There is only
one exception, and the exception is significant.
Fifteen gatherings are still in their original condition, and fourteen
of these have 8 leaves each. There are five complete gatherings
which can be reconstructed with certainty on the basis of their
contents, each having 8 leaves. Finally there is one gathering of
which the reconstruction is a matter for debate, but which also
probably has 8 leaves. It is 180e, a manuscript consisting of 9 leaves:
ff. 1 and 2 are single, 3+8, 4+7 and 5+6 are conjugate, and 9 is
single. Ff. 1 and 9 are joined together by a strip of paper stuck
down their spine on the outside (i.e. stuck on the inner margin of
ff. Ir and 9v). Ff. 2 and 9 were once stuck together by a strip down
the spine on the inside. It looks as if ff. 2 and 9 were originally
conjugate, as if they were strengthened down the spine, and as if
f. 1 was stuck to them. Some time later f. 2 was torn out, but not
removed or lost from the manuscript. This method of strengthening