Reviews and Notes 475 prevailing construction, the auxiliary vczre is quite commonly used in the imperfect except with durative verbs. That is vare is not only used in the quasi-passive, but also in actual passive sense: det huset va bygd same aret som utstillingen va. As Iversen says, it is interesting to find this old construction in living force to-day in the midst of a town dialect. However, Troms0 evidently stands much nearer the dialects than it does to the speech of the cities; and I am inclined to think that the above quoted construction will be found in many WNw. and NNw. dialects today, and even more generally than in Tr. at the present time, for the author notes the fact that it is now heard for the most part only among older speakers. As to the scope of the compound and the suffixal passive Tr. has gone far, perhaps farther than most dialects, in the direction of substituting entirely the analytical form for the inflected one; this is to be expected in a dialect region where the analytic ten- dency has been so strong as in Troms0. The author shows the scope that the ^/-passive still has it is used practically only immediately following the modal auxiliaries. And here again the interesting phenomenon is met with, that verbs which in ON are found with the sumxal passive have either gone over to the class of compound passives, or can no longer be used in the passive at all (page 51). That is, there has in modern Tr. been a growing tendency to avoid the ^-passive. There has also been a growing tendency to avoid the use of the passive entirely; wherever possible one uses an active construction; otherwise one uses the compound construction. But even the compound or analytic passive is avoided wherever possible, the active construction is employed. I shall point out that in varying degree both these tendencies have been in operation in all Nw. dialects of course, and, I venture to say, in an especial degree in WNw. dialects. An investigation of the whole subject of the passive in Norwegian dialects would bring interesting results, but for such an investigation existing dialect material is of course quite inadequate. Finally one more point may be mentioned: the use of at before the infinitive is omitted only after ville, skulle, kunne, maatte and faa. In the closing chapters the author deals with sentence structure, word-order, ellipsis, pleonasm, and anacoluthon. As regards printing, paper, and format the volume is uniform with Maal og Minne's issues; it is printed on excellent paper and in the neat type of this journal, producing a page that is "invit- ing" to the reader. The investigation itself was awarded the Royal Gold Medal and printed by the Nansen Fund Commit- tee. GEORGE T. FLOM.
Urbana, May 24, 1919.
