Studia Islandica - 01.08.1937, Síða 44

42
heterogeneous chapter was composed, on the contrary, it gradu-
ally draws closer to A in similarity. This is evidence of scribal
relationship.
6. Similarity of words in the two versions covers about 10 %
of the total in chs. 13—18 of A, and this must be reckoned as
indisputable evidence of scribal relationship.
V. The chief results. The author considers that enough
evidence has been brought forward to show that chs. 13—18 in
C have been constructed from a written version of A. From
this it follows that Ljósv. is, indeed, “an instructive exampie”
as Liestöl says, but the instruction which it gives is the opposite
to that which he asserts. It is, moreover, improvident ío dis-
regard such instruction in discussions of the Buchprosa and
Freiprosa theories.,
VI. Treatment of the Saga in C. Finally the scribal reia-
tions of the two versions are investigated further in order to
discuss the method employed by the writer of C. In addition
to A, he made use of genealogies and general historical know-
ledge, though it is not necessary to suppose that he reiied on
oral stoi'ies independant of A. Much more than an historian, he
is an author, who rewrites chapters of Ljósvetninga Saga as an
historical novel.