1921 REVIEWS OF BOOKS 297 called patronymics, and moreover it shows that when in a local name we find a derivative in -ing of obscure meaning, we must not assume that it is necessarily from a personal name. The use of the suffix to denote place of origin is paralleled in Old Norse, e. g. in Islendingar, ' the Icelanders ', Iomsvikingar, ' the men of Jomsvik '. The name Skerningham (c. 1090 Skirningheim) is, I believe, a Scandinavian formation of this type, meaning the 'home' of the settlers by the river Skerne. Mr. Mawer would identify the middle syllable of this name with the word eng, a meadow ; but the phonetic change of eng to ing is surely much later than the eleventh century. When a derivation in -ing denotes a person's parentage or ancestry, it may allowably be called a patronymic, though the term is not strictly accurate ; a corresponding term is needed to express the function of the suffix when it denotes local habitation or origin. Perhaps it is owing to the lack of such a technical term that the importance of this function has been so generally overlooked. To supply the deficiency satisfactorily is perhaps impossible ; provisionally I venture to suggest topophyletic, 1 hoping that somebody will invent a better substitute. An OE. place-name containing as its first element a derivative in -ing (whether ' patronymic ' or ' topophyletic ') may belong to either of two types. It may be a syntactic combination of the genitive plural, as Mgelbyrhtingahyrst, Stdnmceringa gemcere, or a thematic or proper com- pound of the stem, as JElfredingtun. The formal difference is much the same as that which exists in modern English between ' the Johnsons' estate ' and ' the Johnson estate '. What determined the choice of the one type or the other in particular cases it is difficult to say. 2 It may be suggested that the syntactic type would be more naturally chosen when a place or a district was in the possession of a group of persons of common ancestry or the same local origin, and the thematic type when there was a succession of single owners known by the same patronymic or ' topophyletic '. However this may be, I cannot accept the author's view that the -ing in such a name as JElfredingtun is not patronymic but simply possessive, i. e. virtually equivalent to the geni- tive ending. One of the facts adduced in support of this proposition seems to me to prove exactly the opposite. A charter of the seventh century relates to a place called therein Wighelmestun ; an endorsement of the tenth century adds ' nunc Wighelmingtun '. This means that the farm was called Wighelmes tun when Wighelm was alive, but that when it passed into the hands of his descendants it came to be known as Wighelm- ingtun. As i§ well known, a place-name containing the genitive of the original owner's name often survived for centuries ; but this example shows that sometimes people did feel that the ' patronymic ' was more appropriate than the possessive in the designation of a place when the man after whom it was named had died and left it to his descendants. This is so natural that if we did not remember the conservative tendency 1 Without any great deviation from classic usage, we might say that the Cat- mseringas or the Eastuningas were a <f>v^ roinich. 2 I formerly suggested that the choice between -ing and -inga might depend on the length of the word, but Mr. Alexander (Essays and Studies of the English Association, vol. ii) has shown that the explanation does not accord with the facts.
