18 C. (Save and G. Stephsins, 



But all the difficulty will disappear, if the old drawings by the 

 Celsius family and the neu- drawing by Carl Save were right P). Should 

 the stone really have SIULFIR, as I believe it has, we get a simple 

 and suitable and splendid meaning. For this SIULFIR will then be nom. 

 sing, mase., meaning SELF, HE-HDISELF, the deceast chieftain FI- 

 HIULFI. As all other words, so this for SELF has manifold dialectic- 

 shapes on our runic and other monuments. Excluding those places where 

 the word is a Proper name, I have on my runic lists — which are not 

 complete: 



n. s. m. SELFR; SIALFR; SIALVR; SYULFR or SÜULFR (Lilj. 

 739; Dybecks Runurkunder, folio, no. IOC); Lânijthora^ Upland: ULFR 

 SYULFR HAN). 



d. s. m. SIOLFUM. 



ac. s. m. SELFAN, SELFON, SIALFAN, SILFAN, SILFON. 



ac. s. f. SIALFA; SIALFUAN, this last doubtful, as the stone 

 has perisht; it is Lilj. 329. 



The ending -IR for -R, etc., is common enough in olden days 

 and on runic pieces"). 



I therfore regard the [FIHIULFI] .... SIULFIR of the Malstad 

 monolith as identical with the [ULFR] .... SYULFR HAN of the Lång- 

 thora stone. 



But we have UIHA left. Save's happy correction here, from the 

 formerly-redd UISR, makes the whole plain. This word, so famous and 

 common in Old-English, is almost unknown in all the other Scando- 

 Gothic tungs. I am not aware that it has ever before been found in 

 Sweden. In Norway and Denmark it has also probably died out. In 

 Iceland it remained, but only as the name or epithet of a Dog. Even 

 what we have of Moeso-Gothic does not show it, as little as the O.-Fris. 

 and the O.-Sax. In Ohg. it occurs, but only once and then in a com- 

 pound, in the old form UUIGO. In England we had WIGA, gen. 

 WIGAN, with a host of compounds ending in -WTGA. It lingers on in 



') It is a long time since 1851, wlien Save made his drawing, lu the course 

 of 25 years the frost and snow may have still further damaged the stone at this spot, 

 and this may have misled Prof. BuGOE, So much the more, as the différence between 

 * (S; and -^ (Bugge's K, Save's H) is not great. Ever so small a weathering or crack 

 or tlaw at the bottom of the '•, and we have -■■ at once, if we are not on our guard. 



^) As we know, the Scandinavian nom. ending in later but still old times had 

 many forms. Besides the usual- UR (commonly in the j\Iss. only -R) were also -AR, 

 -ER and -IR. Of this last we have not a few runic examples, some of them very 

 old, as here on this Malstad stone. See JoK Thorkelsson : »Um R og \^R i niÔrlagi 

 oroa og orc^stofna» (Reykjavik 18<io. 8vo). 



