Fróðskaparrit - 01.01.1964, Síða 222
230
Norse Steinn in Hebridean Place-Names
uík ‘broad bay’ (another “Breivig” is found north of Stor-
noway”); [g'e:d'am], an islet near Bernera (Lewis), not named
on the One-Inch map, from O. N. Geit-holmr or -holmi
‘goat islet’; “Tob Leiravay” [to:b l'e:r'ovaj], a bay south of
Stornoway, from O. N. Leiru-uágr ‘bay of the river Leira’
(Leira, in its turn, means ‘the clayey one’; “Loch Leira-
vagh” [Lox k'i-L'e:r'ovay], Gaelic spelling Loch Cinn Leirea-
bhagh, a sea-loch on the Benbecula coast, likewise from
O. N. Leiru-uágr which in this case is derived directly from
leira ‘clay’; “Leiregeo” [L'e:r'og'ay], a coastal gorge in Uig
Parish (Lewis) from O. N. Leiru-giá or Leir-giá ‘clay gorge’;
“Loch Erisort” [Lox e:r'osoRd], a long sea-loch on the east
coast of Lewis, from O. N. Eiríks-fiprðr ‘Eric’s fiord’;
“Eriskay” [e:r'is'k'æj], the well-known island between South
Uist and Barra, from O. N. Eiríks-øy ‘Eric’s island’. To
these might be added a number of names in which the
correspondence between Mod. G. [e:] and O. N. ei is highly
probable but less certain than in the names already men-
tioned.
In hereditary Gaelic words, [e:] almost invariably corre-
sponds to C. G. e, as found in Old and Middle Irish.
Accordingly, O. N. ei must have had a pronunciation re-
sembling C. G. e more than any other C. G. vowel or
diphthong at the time when these names were adopted by
the Gaelic-speaking population. When C. G. e was pre-
ceded by one or more consonants, these were always pala-
tal or palatalized, hence the substitution, in the process of
adaptation, of O. N. st- by Gaelic [s't'—]. But the corre-
spondence between Gaelic [e:] and O. N. ei, although by
far the most frequent, is by no means universal. The town-
ship name “Stoneybridge” [styíN'obr'ig', stymabr'ig'] (South
Uist) corresponds to O. N. Stein(a)-bryggia ‘stone bridge’,
probably referring to stepping-stones across one of the
many fords in the neighbourhood (the element -bryggia is
extremely rare in Norse place-names, and my interpretation
will certainly be met with objections, but I think they can