of their use, are remarkable in their isolation in Germanic, and still more remarkable in their similarity (and even identity in use) to the b- forms of Welsh. The closest point of contact is, of course, OE. bið, used as a consuetudinal, a future, and sometimes indistinguishably from the present, as compared with Welsh byd, with the same uses (which are proper to the whole tense to which it belongs). The peculiar OE. subjunctive béo also recalls the Welsh subjunctive bo, &c. It does not seem probable (nor is it expressly suggested by Keller) that we have to assume direct borrowing of any single form-even in the case of bið, byd, this is not phonologically likely; the forms of either language permit of explanation from their own native material. But at least worthy of consideration is the possibility that the peculiar development in OE. of two different 'present' paradigms, partly differentiated in use is due to British influence-due, that is, to the transference of British habits of speech to English in the mouths of Britons, accustomed to associate differences of function with the b- forms of the verb. Bid, which offers difficulties of etymology from the purely Germanic point of view, would then, though constructed from native material (b-the-ip of the third person), be due to imitation rather than mere adoption of byd (or its antecedents), where actually -8 has nothing to do with the third person at all in origin. This is an interesting theory, and in keeping with what can be observed in the development of invading languages elsewhere; unfortunately we know all too little of the relations of Briton and Saxon. Keller's theory of the influence of Scandinavian speechhabits at a later period on the English verb (below, p. 45) should be compared.

The second part of the article (Englisches und keltisches Gerundium) is not so striking. The development studied—the difficult question of the English uses of the verbal noun in -ing-belongs admittedly to a later time when the sort of influence envisaged is no longer likely. The verbal noun and its uses in Celtic and English are not lightly to be compared in a few pages. The elaborate study of the English gerund noted below was not, of course, known at the time of writing.

Professor Hoops contributes Werder, Rasen und Wiese, eine Untersuchung zur germanischen Wortgeschichte-not Weltge

schichte as the table of contents (faulty in other particulars, see p. 39) has it, which would attribute to the author a theory of a wet green world for primitive Germans that is no part of his article. This actually is an interesting etymological study, and an attempt to unravel the tangled connexions of a group of words represented in the title and in OE. war, waroð, wāse, wōs, wār, waru, weord, wer (weir), not to detail other German and Scandinavian words connected with these. The words are not, of course, all etymologically related. The article will interest both lexicographers and place-namers. The author says that he has not permitted himself to follow up many of the interesting side-issues. Knowing how these little lexicographical chases open vista after vista and one complication after another, we can well believe that much self-denial was practised to keep the notes down to thirteen pages. Space here does not allow us to describe the hunt afresh.

Professor Luick's article we reserve for consideration with his article in Germanica. Zum deutsch-englischen Wörterbuch, contributed by Liebermann, is of interest primarily to Germans and secondarily to all lexicographers. We are reminded what a struggle these patient fellows have trying to keep pace with the world, with their labour ever at the mercy of the reckless. We learn that, after the great enlargement of der deutsche Geist in the nineteenth century, the principal cause of the deficiencies in the three chief German-English dictionaries is the 'general haste and fever of modern life grasping at sensation, tearing ideas to pieces, turning ethical values upside down, and introducing finer nuances into the language of poetry, art, and science'.

Sympathy will be felt with the plea that linguistic history belongs to the dictionary of an individual language and should not cumber a work that is a bridge between two modern languages; and with the plea for a wider and less puristic range of German words to be glossed (from all sources, colloquialisms, slang, dialect, technical language, and journalese), and to be glossed more generously and idiomatically. Even in the comparatively narrow range of philological writings there are many words belonging to the youngest Schicht of the vocabulary of that mystery that the ordinary Englishman can only exactly

gloss after a close study of different contexts (if then), however well known the component parts may be. It is arguable, none the less-a point of view not represented in the article-that ultimately no dictionary will, or should, be able to get rid of the necessity for this sort of attention, or dispense with wide reading, painful reading at the outset, as the inevitable road to exact understanding of writings in a foreign language. It is impossible to avoid the feeling that the scheme outlined for organizing and canalizing the observations of all German students of modern English to one centre for the enrichment of a revised German-English dictionary somewhat over-values the functions of a bridge-dictionary. But a perfect dictionary is an attractive mirage, and its nearest possible realization an aesthetic joyappreciated most by those least in need of it.

A curious list is given at the end of the article. It contains several hundreds of words drawn from the Rev. A. J. Carlyle's History of Mediaeval Political Theory (entire) with the addition of one number each of The New Statesman and the Manchester Guardian. The most exact German gloss (a gloss not to be found in the chief dictionaries) precedes; the English follows. With some of the omissions it is possible to sympathize, with others to be surprised-such as grappling with problems, earned income, to motor, a trawler, to whetten (sic) the ardour of. It is perhaps disappointing to learn that German has nothing nearer than viel zuwege bringen for to cut much ice'; but the English is a pointless expression, at any rate in England where ice-cutting is not a familiar pastime.

Germanica is a tree of altogether larger girth and bigger branches-a not unworthy reminder of the honour and affection, the great tale of years, and the great work achieved by Sievers. In addition to its 727 pages it has several illustrations, and two portraits of Sievers. The Tabula Gratulatoria (pp. iii-viii) contains nearly 200 names, a list so large and drawn from so many countries that it is not possible to avoid the comment in passing that, except for the impersonal Taylor Institution, Oxford, it contains no name from the British Empire or from France.

The mere list of contents would take too much space to detail. All the contributions come more or less justly under the heading

Germanica (the connexions of Porzig's Das Rätsel im Rigveda are not close) and so all have an interest, more or less direct, for English philology. Those dealing most directly with German. or Germany have this connexion least. The article by V. Michels on the history of German accentuation (pp. 39–90) is an exception: the developments in German of the modern and recent periods show remarkable similarity to English and are well worth noting even by one with little direct interest in German; moreover, in this relatively long and elaborate article many points of common Germanic philology with direct bearing on English are dealt with. Such are the discussion of the accentuation of nominal and verbal compounds, and the reopening (or the continuation) of the debate concerning the history of the perfectreduplication in Germanic. The author, in a Sievers Festschrift, feels the necessity for apology; more than once he disclaims that finer ear which is necessary for the application of the more recent theories to these problems. What he has to say is none the less worth reading.

L. Bloomfield's article (Einiges vom germanischen Wortschatz, pp. 90-106) is concerned chiefly with German, and largely with German colloquial and dialectal oddities. There is, however, a not uninteresting discussion, and long lists, of the words with geminated consonants (hoppian) that cause so much etymological stumbling, even when the Kluge-Bezzenberger 'sound-law' is admitted, from the earliest Germanic onwards. The article is valuable, too, in emphasizing, if that is necessary, the fact that Urindogermanisch, even when literally incredible ingenuity is revealed in getting back to it, fails to account for the larger part of our vocabulary; and in insisting that the linguistic principles so far established are still the tools by which many of the remaining problems are to be solved. Lautsymbolik comes in for scorn; but by it is apparently meant creation in the void (without pre-existing models developed regularly), and 'spontaneous gemination' and the like. Lautsymbolik of the sort that attaches significance to sound-groups developed at first mechanically, and extends their use, is of the essence of the article. Why it should more than once be called 'naïve' is not made clear. Once you admit even naïve feeling, however vague, for the significance of certain groups of sounds you have Laut

symbolik of a sort, and it requires attention. The grouping of words which is here offered both in rhyming and in alliterative series (flame, flare, flash, &c.; flash, splash, &c.) brings out many interesting points of word-formation.

It was not to be expected that all the difficulties could even be touched on. One misses notably the -gg- words (dog, wag) that from an OE. point of view present especial difficulty for Mr. Bloomfield's thesis, since it is far from clear in what cases they could have been developed with phonological regularity and so available for extension. Looking at wag alone-and it is part of the difficulty of the field on whose borders this article touches that we need as a preliminary much clearer documentation of each individual word in each language than we are likely ever to get mere juxtaposition in a list without dates with, say, drag, sag, flag, rag, would not lead very far. Here we find wagian, wazien, at a fairly definite moment in English, on the one hand continuing its lawful history until it gives the doomed and inexpressive wawe, on the other being ousted by (or transformed into) wagge-n. What are the models? Drag appears to be much later. It looks like 'spontaneous gemination', or, if you will, deliberate and significant alteration.

There follows an interesting article on one group of related names in Germanic and Romance for the pole-cat, wherein H. Suolahti pursues this animal (less easily trailed in etymology than in nature) over most of North-West Europe. English fitchew, fitchet are incidentally dealt with. Many other articles of equal or greater interest must be passed over, as being too remote from our immediate concerns. It is impossible to exclude bare mention, all the same, of H. Lindroth's article on the Röstein inscription in Bohuslän (a by-product of which is an illumination of the question of the survival of I-E. adjectival -uent in Germanic, and the matter of OE. -wende, wynde); of Falk's note on the Old Norse names of the hawk and falcon, which naturally has things to say about several English words; of Mogk's contribution Nordgermanische Götterverehrung nach den Kultquellen, which offers a rapid and remarkably clear sketch of the evidence of sagas and place-names on the question of Odin's position, and concludes that nowhere in the Germanic world, and least of all Scandinavia, was there ever a Woden-Odin that