Bibliotheca Arnamagnæana - 01.06.1970, Page 347

321
AM 564a, 4to and Addit. B.U.H. 20, fol. can be seen to come from
the same codex, because the positions of their binding-holes
correspond34. (Addit. B.U.H. 20, fol. then contained what are now
ff. 1—2 and 5-7 of AM 564a). But Gubbrandur mentions eight leaves,
whereas AM 564a has only seven, and his list of the contents
of those eight is the same as that of the present AM 564a, so that
the extra leaf can only have contained something of which we
already have a fragment in AM 564a; and the only single-leaf
fragment of a saga represented in AM 564a which is known to
have been in Addit. B.U.H. 20, fol. at any time is f.l of AM 445c,I
—the other fragment of Viga-Glums saga35.
6.2. Unfortunately, AM 564a, ff.5,6 and AM 445c,I, f.l have
all lost their inner margins, and it is therefore not possible to
say what binding-holes they originally had; but if the binding-
holes of the Gisla saga leaves (ff.2-5) can be shown to correspond
with those of AM 564a, it will still seem almost certain that the
fragments are from the same codex. Some leaves have damaged
inner edges, so that it is not always clear whether or not a particular
indentation was a binding-hole, but within each fragment the
certain cases can to some extent be used as a check on the un-
certain ones, and when this is done it becomes apparent that the
binding-holes are arranged in pairs in both fragments (see fig. 1,
where rows of binding-holes represent the inner edges of leaves in the
34 It cannot, of course, be proved that fragments which show the same binding-
holes have not been bound together at a later date, but when they have other
features in common this must be regarded as improbable.
36 In Islenzkar fornsogur I (see note 7, above), Gudmundur t>orlåksson prints
the fragments of Viga-Gliims saga in AM 445c,I and AM 564a as Vidauki (I and II)
(pp. 88—106), but his heading to the fragment which is now AM 445c,I, f.l refers
to it as “Skinnblad i A.M. 564.A” (p. 88). However, it is clear from his Introduc-
tion (p. xiv) that he does not regard it as part of the same Ms. as the rest of AM
564a, and he calls it “brot af sama handriti, sem er lir Gisla sogu Surssonar og enn
hefir ekki verid prentad”. It seems probable, therefore, that f.l of AM 445c,I
was moved from its previous place in Add. B.U.H. 20, fol. to AM 564a along with
what are now ff. 1-2 and 5-7 of AM 564a, following Gudbrandur Vigfiisson’s con-
elusions of 1860 (see Stefån Karlsson, above, 1.0 and 1.1.2), that it was still there
when Gudmundur Borlaksson published his edition in 1880, and that, perhaps
as a result of his remark, quoted above, it was moved to its present place at some
time between 1880 and the publication of the first volume of the Arnamagnæan
Catalogue in 1889.
Opuscula IV. — 21