Bibliotheca Arnamagnæana - 01.06.2007, Blaðsíða 304

274
ReLATIONSHIP BETWEEN THE TRANSCRIPTS
IIA i, 4, 7
IIB 17
IIC 22
One observation that requires comment is that there is a handful
of striking agreements between particular readings in 38 and 37 (Ji).
In several places where 35 and 18 agree against 38, the latter is sup-
ported by 37.162 In the collated text this applies to examples IA 2, IA
13, IB 4, IB 10, IC 1, IC 9, IC 12, IIA 1, IIA 4, IIC 8. We might also
add IA 9, IIB 22 and IIC 22, but in IA 9 the readings in 38 and 37
are not quite the same, and in IIC 19 the word concerned is the verb
spyrja, which is not unambiguously abbreviated in 35.163 The reading
mott in 37 (IIB 22) can be interpreted as an irregular spelling of måt,
but it could also be due to a misreading of måti. The possibility that
the lacuna in Jdfraskinna was not as extensive as has been thought is,
however, contradicted by the faet that there are considerably more sig-
nificant examples of the transcript in 38 lacking characteristic features
from the text of 37; for example, a sentence which was obviously miss-
ing in Kringla is found in37 but lacking in 38 (HkrFJ I: 276.4).
I think it most likely that commonsense corrections have been made
independently in all three transcripts. In some cases, corrections in 35
coincide with corrections in 38, and in other cases they coincide with
corrections in 18. In any event we must assume that, in some of the
cases where 38 is supported by 37, the former preserves the reading in
Kringla even though it differs from 35 (and 35 is supported by 18).
162 Page 37 here contains the text of J, so these variants are not included in the collection
of examples. They are noted in HkrFJ.
163 Finnur Jonsson has elsewhere probably mixed up the shelf-marks for the two tran-
scripts of Jdfraskinna (HkrFJ I: 3, note, line 5), but this is not the case for the
references we are dealing with here, and the apparatus in HkrFJ gives correct in-
formation at this point.