Bibliotheca Arnamagnæana - 01.06.1960, Blaðsíða 322
304
and 50:a:23 oldb), after n (50:b:4 hundrap and 50:a:36 hundrap),
before the abbreviation for -er (after l in 50: a: 24 ald? and 50: a: 25
ald?; after n in 50:b:5 stundP and 50:b:l stund?), and alone as a
numeral or an abbreviation (50:a: 12 .d. and 50:b:25 .d.). Examples
of geminate d are quite rare, but type I is found on page 50 (50: b: 37
hlyddi) and type II on page 51 (51 :a:25 odda). On page 50 there are
36 occurrences of type I and 22 of type II12.
4.3. Page 51 shows quite a different situation. Here there are
only 5 examples of type I, as opposed to 20 of type II (plus a possible
Capital in 51: b: 4; capitals which may be equated to type I are
found on pages 48, 54, 57, and 60). Type II occurs initially (51 :a:23
daga), medially between vowels (51:b:27 bxgidar), after l (51:a:3
halda), after n (51:b:29 Siaunde), before the abbreviation for -er
(51:b:4 stund''), in gemination—as mentioned above—(51:a:25
odda), and in a ligature for Nd (51 :a:30 staNda). Type I is found
only after l, and indeed only in the forms of one word, gid (51 :a: 18,
21, 24, 31, and 51:b: 10). It may be noted that on this page this
particular word happens never to oceur with a type II d.
4.4. In this case too, as with g and p, pages 48 and 49 agree with
page 50, whereas pages 52 and 53 seem to agree with page 51. Page
48 has 60 occurrences of type I and 26 of type II; page 49 has in
the edition 36 occurrences of type I and 18 of type II. But for the
36 occurrences of type I on page 49, 11 are in the form aldar which
is abbreviated al in the manuscript and expanded by Larsson with
the -dar in italics. There is, nevertheless, a predominance of type I
on both of these pages. Page 52 has 8 occurrences of type II, but only
one certain occurrence of type I, here in the word heldr (written
hldr). There are two examples of the word hende where Larsson
prints the -end- in italics, but the abbreviation of the manuscript
is hé. Page 53 has 10 occurrences of type II and none of type I;
hende occurs in the edition 5 times with type II but always with
the -end- abbreviated in the manuscript. On both page 52 and page
53 forms of gid occur with type II d.
4.5 I will briefly summarize the investigation of the two types
of d. When one analyzes the distributions of both types, two distinet
12 Perhaps it would be well to note here and to keep in mind during the following
discussion that in Old Icelandic d will naturally be statistically more frequent in
initial position and after l and n than elsewhere.