Blümel makes out, but very much more so, and not so easily to be labelled and bottled.

There are some odd things thrown out by the way which seem to border on regions of investigation more generally understood, though their connexion with the central matter is not always clear. For example: originally there were no pure voiceless consonants-r was a product of lh or th (so r figures as the equivalent of lh in the scheme of grades of definiteness in the voicing of consonants)-ñ and ŋ do not appear before sonants and so clearly have here become l and -and so on. All these dogmas appear to apply to the Grundsprache in a far-off time before any changes had affected it: in the good old days when gonewo meant knee'. There is a 'law', too, to whose definition the writer claims that he was led by communications from Sievers on individual points, which runs: a word inherited from the Grundsprache is in its regular phonological development still represented in all its phonetic parts and all their phonetic values and qualities even after all changes have taken place. Corollary: there is no such thing as Zusammenfall, only narrowing of the acoustic difference as far as apparent identity; there is no Spaltung, since every apparent case goes back to phonetic differences, of major or minor importance, as far as you like to pursue it. Example: in present-day German we still have two different Indo-European as (Hahn, Henne), two different os (Nume, nennen), and Hahn can no more be pronounced with the a of Name than can Henne with the e of nennen. After this there is nothing of equal interest until we learn that the Germanic Rhythmusart and the Indian are Hauptarten. No people can possess at once two Hauptrhythmusarten. Since Indo-European possessed these two, the Indo-Europeans cannot have been one homogeneous people as the Germans were. The writer, it is true, shows some signs of hesitation whether to call the Germans a Volk or a Verband.

It may seem odd to refuse to give the essentials of an article and to retail isolated points. Certainly the connexion between these sayings and their context is lost (where any such was apparent), but the reliability of a witness is often to be judged not so much from his main deposition as from the things he lets out by the way. Some of these cats come out of a strange bag.

Probleme der englischen Sprache und Kultur is a slighter volume. Linguistically salutary is the article by Morsbach : Prinzipielles zur modernen Syntaxforschung. There are sound things said, which once would hardly have seemed necessary, in defence of the view of language as a tradition, and of the historical view generally. The thesis maintained is that while the ' psychological methods' have certainly opened up new avenues of approach, what is really new in modern methods as against old is the war waged on the historical view, whether in literature, linguistics, or art. 'Everything must be intuitive, and interpreted and understood as far as possible out of one's self as the witness of the Soul.... This is building in the thin air of speculation and robbing the study of its natural nourishment and foundation-a just criticism of much that is encountered on the dim borders of the linguistic field. It is a moderate and sensible article, very polite to the adversary. The attempted rearrangement of the departments of linguistic study and the reconsideration of the place and content of syntax are worthy of note. It is followed by an article (O. Funke) on the definition of the conception 'Proper name', a relatively brief contribution to a difficult subject, which considers (chiefly) the views of Marty, of Jespersen (Philosophy of Grammar), and of Noreen (Einführung in die sprachwissenschaftliche Betrachtung der Sprache). Compare the author's Innere Sprachform (Reichenberg, 1924) and his article Zum Problem von Körper- und Sprachfunktion in Luick's Festschrift. On the next seven pages W. Keller deals with some features of Scandinavian influence on the inflexions of English in the occupied areas, with its sequel in Middle English. It is held that the similarity and the points of contact between ON. and OE. inflexion has not been fully appreciated, since the phonological state of the endings in ON. during the invasion periods has not been generally understood. The writer of the article holds that final z (Y) had not become an r sound; that we have, for instance, to allow for points of contact and comparison such as ON. dagaz OE. dagas. It is true that Noreen's final opinion was that a change - to -r has not yet been proved from early inscriptional authority, but it does not so immediately follow that there is nothing in the way of assuming that the rune Y was merely a sign for z. Though R

in Danish appears not to have become r until about A.D. 900 after dentals, and not until about A.D. 1100 elsewhere, the sound we symbolize by R may have been sufficiently different from OE. 8, z sounds very seriously to diminish the supposed similarity of the inflexions of the two languages. No attempt is made to dispose of ser, serliche; helder appears to have been overlooked. The Leiden Rune-names and the Abcdarium Nordmannicum conflict in their evidence; the first shows clear -r (naudr) from the tenth century, the second neither -r nor -z (naut), possibly from the ninth. But they are rather doubtful witnesses. For the influence of Scandinavian upon the verb in the northerly areas there appears a good deal to be said. The criticism of Holmqvist's" study makes two points. The statistical fixing of the 'person where s (from p) is most frequent does not necessarily indicate the point of origin. Still more pungently: how came OE., the whole linguistic feeling of which had for generations required no inflexional distinctions in the persons of the plural, to feel the need of a different form for the 2nd person plural, if not by Scandinavian influence? The article concludes with a theory of the OE. indicative forms binde, bunde. These are polite substitutions of the subjunctive. The question at once arises-why then the crude indicative bluntness of band, bind(e)st?

The longer article by Professor Horn on the OE. charm against elfshot is full of interest. Though it does not achieve the impossible by bringing any very brilliant illumination into this dark corner, it does something: more than has yet been done. Professor M. Förster's contribution is a discussion of the legend of the trinubium of St. Anne. Its only connexion, however, with the general title of the volume is that Oxford manuscripts are used beside continental ones in discussing the metrical Latin versions, while there is given the OE. version of a late Latin prose original which appears in twelfth-century transcription in MS. Cotton Vesp. D XIV. The allusions in Cursor Mundi and Myrc's Festial are recorded. But it is not on the English material that Förster's learning is chiefly expended in this long and erudite article. In yet another FestschriftAbhandlungen aus dem Gebiete der mittleren und neueren Geschichte und ihrer Hilfswissenschaften, Münster-the same See The Year's Work, 1922, p. 20.

learned writer contributes an interesting note: War Nennius ein Ire? to which we can here do no more than refer. A brief note is contributed by W. Fischer on the French of Chaucer's Prioress, returning to the older satirical view. Professor Holthausen contributes not etymology but a metrical German version of the morality, The Pride of Life. O. Ritter's article (Lauthistorisches zum Namen Don Adriano de Armado') develops into a fairly exhaustive study of the loan words from Spanish, Portuguese, Italian, Indian, and other sources which vary in the endings between o, a, and y. This exhausts the articles of direct philological bearing, but not those of interest. Of these A. Cartellieri's sketch of the history and character of Richard Coeurde-lion is good, while totally different and quite unexpectedly entertaining is J. Schick's Indische Quellen zu Longfellow's 'Kavanagh', which affords a rare mixture of Longfellow, Sanskrit, and some easy simple and quadratic equations. It is to be regretted that already at the time of preparation the late Dr. R. Jordan's ill-health robbed the volume both of his contribution and his editorship.

Except from the point of view of the reviewer struggling to keep within limits of space it is unfortunate that Luick's Festschrift is not available. Review-copies of these works are not readily obtainable, it appears. We may say that Germanica, though here treated, was in like case.

There remains the luxuriously appointed Mélanges in honour of Professor Vising, with its thirty-two articles, bibliography of Vising's work, and its portrait. Though the articles largely operate in the field of Romance they are worthy of note by English philologists. Even where the titles seem remote from the concerns of The Year's Work it is frequently found, as is natural from the interrelation of all European philology, that points, minor or major, of English vocabulary or linguistic history are touched upon. There are two place-name articles definitely concerned with England; for these see below.

The following may be mentioned (the selection has reference. solely to concern with English or Germanic): Romanisches in der ältesten isländischen Literatur (N. Beckman); Fr. chagrin, colère (C. S. R. Collin); Keltische Etymologien (E. Lidén);

Egidius > Gilles (K. Michaëlsson); Guernes de Pont-SainteMaxence et la légende de Becket' (E. Walberg). E. Wadstein contributes an article on the etymology of OE. wícing, OFris. witsing-with no reference except in jest to the name of the master honoured. His conclusion alone can be given—that the word is derived from the -wik that entered so early into Germanic and North-Sea names of commercial sea-board centres, Sliaswik, Wic (= Quentovic), Dorestad (Wijk bij Duurstede), Bardowik; that this is due to the inevitable mixture of piracy and commercial enterprise, illustrated but not monopolized by the later vikings'. In other words, it is a derivation once more from Latin vicus. But the remarks of Bohnenberger on pages 139-40 of Germanica should be noted; see below. K. F. Sundén writes On the Origin of the English affirmative particle aye "yes". The theory, which will not immediately be canonized, but is well argued, is advanced that it is ultimately the first syllable of ME. i-wisse. The difficulties that face the inquirer into the history of aye: its orthographical oddity, sudden appearance and rapid spread, are not made light of. The etymology proposed appears to meet them more successfully, at least, than any other that has ever been suggested.

The chief item to be noted under the important heading of Place-names was in 1925 the issue of the first county-volume, Buckinghamshire, by the E. P. N. S. The main part is, of course, the body of names arranged under the eighteen hundreds in geographical sequence, and alphabetically under the hundreds. Detailed criticism is here impossible, nor does it appear probable that even one who had worked with equal labour and care over the same field would have much to offer, other than praise. Among so many entries, and so many interesting notes and etymologies, it is difficult to select any for mention. we mention Ivanhoe, Fingest, Marlow, Quainton, Risborough, Linslade, and Tiddingford Hill it is a mere taste, hardly a sample. The local pronunciations which are given offer a special feature of interest. The explanation of the river-names

If

8 The Place-names of Buckinghamshire, by A. Mawer and F. M. Stenton. English Place-name Society, vol. ii. C.U.P. pp. xxxii +274. Two pocket. maps. 188.