Editiones Arnamagnæanæ. Series B - 01.06.1960, Side 18

XVI
where in this manuscript (or in Brynjólfur Jónsson’s
other copies of this saga). The spelling adopted for
the expansion of an abbreviated word is used con-
sistently; for example, rnikill is spelled c. 85 times
with i, c. 30 times with y, so all the abbreviated in-
stances have been expanded with i.
The abbreviation ’ must be expanded as er in
words like sueria 36, but in endings it probably means
ir in this manuscript, and is so expanded. The statistics
for the spelling of the endings in words written in full
in this manuscript are c. 107 ir, c. 57 er. It should be
noted that this proportion is reversed in Brynjólfur
Jónsson’s other copies of the same saga: 109 has c.
52 ir, c. 112 er, and S17 has c. 41 ir, c. 262 er.
Some of the letters of the manuscript require a
comment.
A difficult minor problem is the y and the ij of the
manuscript. In their extreme forms they are clearly
distinct, but each has several shapes, which merge
into each other, so that in many instances it is difficult
to say whether y or ij was intended. The presence or
absence of two dots is not decisive: there are some
instances of a clear y with dots, and more of ij or an
indeterminate shape over which it would have been
impossible to put dots because of the use of a mark
of abbreviation. The earlier form of a word is very
little help, because the old y had lost its rounding and
fallen in with i, and seventeenth century scribes had
no guide to the use of i and y spellings in their own
pronunciation. An arbitrary decision is unavoidable,
and the rule followed here is to transcribe as ij the
clear and certain instances of ij, and to transcribe as
y or y the clear instances of these together with those
uncertain shapes which might be one thing or the
other.
The accents in the letter áí are the same in the