THe Brus INscripTION AT ANNAN. 83 
{t is not so much the forms, which are not hopelessly wrong, 
as the lack of character in their cutting. There seems to 
be more character in the inscription itself. The mixture of 
‘Gothic ’ and Latin forms would be all right for the period, 
if I may judge by the coins, of which I have more experience 
than of lapidary inscriptions. I should not hesitate to take 
s as g, and the word SEN 5 YU[R| as Seigneur. There 
is a tendency, at least in the fifteenth century, to put the 
“n’ before as well as after the ‘g’ in this nasal sound. 
Thus on the medal of Charles the Bold you get the form 
AVIENGNE. Ducange, Glossarium, Tom., 7, p. 299, gives 
SEINGNIE=seing on paraphe; and I dare say that in 
Godefroy you would find other instances. From this it 
would be an easy stage for a careless cutter to drop the 
second N. Ido not feel happy about the form wv for U. It 
may be early, but on medals, etc., off-hand I can only 
remember it on sixteenth century pseudo-gothic pieces. 
‘Yours faithfully, 
Cl Sigel s Ira eben 
Subsequently he forwarded to me a letter which he had 
received from Mr Peers :— 
‘* Department of Ancient Monuments and Historic Buildings. 
H.M. Office of Works, 
Storey’s Gate, Westminster, S.W., 
7th April, 1916. 
My Dear GEORGE, 
‘““T am sure the numerals are wrong; they are just 
scratched on. It seems to me that the line beneath them is 
of the same type, scratched, and not chisel-cut like the lines 
under the rest of the inscription; but that is only to be 
decided by seeing the stone itself. The inscription as a 
whole is obviously all right, and of about the date to which 
it aspires. The U’s are curious, but I think quite possible ; 
there is just the same mixture of U and V on the 
ION LUMBARD inscription in Christ Church, Dublin, of 
which I have a squeeze at your disposal. The 5 (that’s all 
I can make out) seems to differ from the 6’s, in having a 
definitely straight upper stroke, and must be, I think, 
accepted asa G. 6 would soon get into the required form, 
