Bibliotheca Arnamagnæana - 01.06.1970, Side 356

330
shows veri (llvbl6) and Selingsdals tungu (llvb40); however, it cannot
be certain that the reading here was not originally I, which is often used
for æ in this hånd, for there are several examples of e where the hook has
been almost completely wom away, e.g. aur vint, llrb21.
9.12. Oraphs for eda.
æ. is usual for eda in Hånds 1 and 2 (e.g. 3ra23, 5rbl6), as in most of
the hånds in the other two fragments, but æda appears at 2vbl3 and edr
at 3rb8 in Hånd 1, and eda twice in Hånd 2 (llra42, llrbl7, both at line-
endings). Hånd 3 always has eda (e.g. 8vb38).
9.13. The graph ai.
Hånd 1 begins with a loose observance of the use of this graph when
d forms a whole word or word-element. Including cases where influence of
single-letter forms may cause the appearance of ea elsewhere (e.g. in the
past tense mtti, as at 2ral7, where ai is perhaps caused by the present form d),
I have counted on ff.1,2 76 “correct” uses of it (e.g. 2rall), and 28 “incorrect”
ones, mostly in short words (e.g. mdr, 2rbl6), while its use is omitted in
10 cases of d as a whole word (e.g. Ivb38). Thereafter, the scribe seems to
have decided that ai or ai should be used everywhere for d, and the simple
o confined to the short vowel, although this does not quite become con-
sistent by the end of Hånd 1. That this was the process involved can be
seen by the scribe’s slowly increasing tendency to represent 6 with an
analogous double form 66, which, however, remains a minority form,
oceurring 24 times in all (e.g. gooda, 3ra42).
Hånd 2 uses m or éå as a whole word 48 times (e.g. 10vb6), and else-
where only once, in the name l‘rm\ndur at 10va38-9, where æ. is at the
line-ending, and may be only a scribal flourish. a appears as a whole word
3 times (e.g. Ilvb38).
Hånd 3 uses æ. or éå as a whole word 56 times (e.g. 8vb33), elsewhere
14 times (all in monosyllabic words except mrsalinn, 9ral0, oflæilig, 9ral2,
hmtta, 9ral7). Its use is omitted in d as a whole word 33 times (e.g. 8vb21)
but the general impression here is similar to that given by the other frag-
ments, particularly AM 445c,I. Indeed, there is a marked resemblance
between the hånd there and Hånd 3 here.
9.14. But although Hånd 3 of AM 445b is generally more conserva-
tive than the hånd in AM 445c,I, it is also more fluent and con-
sistent. Since the two are so similar, it seems likely either that
AM 445c,I is the work of a younger member of the same school,
or else, if both are the work of the same scribe, that much practice
lies between them, and that we must place AM 445b before AM
445c,I. This would imply a transition in the scribe from a
progressive stage to a more conservative one, but disregarding
this we might place AM 445b after the end of AM 564a. How-
ever, this is improbable, because Hardar saga seems to have been