These pairs can be resolved by supposing a dialect variation in the primitive period, by which & was labialized to w. A vastly greater quantity of evidence than is here put together would be needed, however, before any such supposition might be proved. IV VEMENS AND CLEMENS. In agreeing (p. 64) with Niedermann's explanation of vemens, as the allegro form of vehemens, I would not be understood as accepting the derivation of vehemens from a participle *vehemenos. Terence, if we may trust the Westerhov index, uses vehemens and vehementer only of the emotions or what rouses the emotions. Plautus, who uses the adjective but twice (Rud. 70-71: nam Arcturus signum sum omnium acerrimum; vehemens sum exoriens, cum occidor vehementior) uses it as explanatory of acerrimus, and the sense is certainly not more definite than our 'violent'. He uses more frequently the adverb vehementer, and I herewith submit the examples in the order of semantic extension as I conceive it: Merc. 923: mater irata est patri vehementer. Curc. 724: ego te vehementer perire cupio. Curc. 568: vapulare ego te vehementer iubeo. Mil. 205: ita vehementer icit (sc. femur). Rud. 903: ita fluctuare video vehementer mare. In these passages the sense is 'violently', 'mightily', not 'speedily, quickly'. The evidence submitted to explain -mens as syncopated from -menos is extremely weak, viz.: the two adjectives vehemens and clemens, which have gone entirely over to the inflexional type of mens and must undoubtedly have been, to the Romans, compounds of mens. Not to find compounds of mens in Latin would be surprising. In the Rig-Veda twelve compounds of mati-s 'mind' are indexed by Grassmann and eleven of mánas 'mind, thought'. To the latter correspond perfectly the Greek compounds εὐμενής, δυσμενής. Generally cognate with the Sanskrit -mati- compounds is avróμATOS, and perhaps πολύμητις. Other synonyms of 'mind' occur in Taλaí-opov, with the contrasting meanings 'patient of mind' and 'daring', Tλn-Oupos 'stout-hearted'. Similar compounds I find in Latin vehemens and clemens. The formative type that occurs in vehemens is the type of Greek μeve-dŋtos 'awaiting the enemy', 'brave'; μeve-тóλepos 'persistent in battle'; μενεχάρμης 'staunch in battle'; δακέθυμος eating the heart'; raλa-Tevens 'enduring woe'. So I would define vehemens as 'moving (rousing) the mind', though mens may have a force in this compound like Greek μévos 'might, rage', compared with Skr. mánas 'mind, will'. This type of compound is occasionally found in Latin, e. g. in Verti-cordia, flexanimus, posci-nummius, etc., but the type is rare: it is the more noteworthy that two of the examples contain synonyms of mens in their final member. As to clemens, Osthoff's derivation from Skr. çrayamānas 'leaning', supporting Bréal's derivation from clivus 'slope', inclinat 'leans', seems to me utterly without cogency on the semantic side, to say nothing of the general objections already made to throwing over the apparent cognation of clemens with mens. There can hardly be any question but that clemens means 'gentle', and Bréal's argument that Tacitus writes collis clementer assurgens and iuga clementius adirentur can mean nothing to an American who talks of 'gentle' slopes and 'easy' approaches. Το explain clemens, I resort to Greek ταλαί-φρων and τλή-θυμος, cited already (cf. also ταλα-πενθής, ταλα-εργός ταλα-κάρδιος, ταλαTeípios, etc.), and derive clemens from *tle-mens 'patient of mind, gentle'; tlè- is a reduced form to a 'base' TELE 'bear': at least this, and not TELA, is the 'base' recognized by Hirt (Indogerm. Ablaut, 279). It may well be that tle- would normally yield lein Latin (cf. latus from *tlatus), but in the compound, inclementer, -cle is normal. Query: In view of the well-known Romance adverbial suffix derived from Latin mente, and foreshadowed by furiata mente 'full of fury' (Aeneid, 2. 407, 588), memori mente 'mindful' (Horace, Serm. 2. 6. 31); in view of the Greek compound avróμaTos 'self-willed'-or should we explain as avтó-paт-os 'of self will', gen.-ablv., raised to a nominative, as Latin penitus 'from within' becomes penitus 'inner'?-; may we not seek for the Sanskrit possessive compounds in -mat- a derivation from the stem man-t-? Cf. manyu-mát- 'furious-minded, full of fury' mantu-mat- 'counsel-minded, full of counsel'. The development of the sense of numerousness in the Skr. suffix -mant- 'rich in' may be illustrated by the signification of Gr. μévos 'force' (:Skr. mánas 'mind, will '), for the notion 'force' suggests that of number very naturally, as we may see from Latin vis = multitudo, and English force' a troop of soldiers'. V QUINTUS: QUINCTUS. The issue whether -nct- is the normal or an analogical form in these words has long since been joined, and the handbooks adequately represent the discussion. The whole issue seems to me to have been created by the old over-insistence on the blindness of the phonetic laws. Now that the blindness is a little less of a shibboleth, methodologically considered, it may well be that the phonetic questions touching the greater or less compression of awkward consonant groups may be left to settlement by the principle of relative speed of utterance in the allegro and lento dialects, which hardly differs in this problem from saying the relative distinctness of utterance in individuals. My present purpose, however, is not to go into deep questions of method in phonetics, but humbly to call attention to the Plautine passages: and quincto quoque sulco (Trin. 524, A) quincto anno quoque (Merc. 66, B. quicto, alii quinto) in contrast with quinto- in all the other Plautine passages, so far as Leo's apparatus and the Lemaire index furnish a means of control. Is it not likely that quincto quoque owes its phonetic peculiarity to an alliterative or rhythmic impulse? The common occurrence of the name-forms of which Quinctius may be taken as the sample may suggest another, and now a social, impulse to preserve the group -nct-, viz., the conservative spelling of proper names; cf. the names Johnston and Campbell, sometimes pronounced (chiefly by persons enjoying these names) in accordance with the orthography. VI CULPA, CULTER. I propose to connect culpa with the verbs scalpit, sculpat 'scratches, marks, cuts', and to define by 'blemish', 'stigma'; cf. nota 'mark', 'reproach', 'disgrace'. The explanation of culter from a stem *certro- (:keipei 'shears') is an excellent suggestion of Skutsch's. Another admissible explanation may, however, be presented, viz.: from a stem *sculptro-, with loss of p in the heavy consonant group. It can hardly be decided from Capt. 266, ad cultros adtinet 'he's reaching for the shears', that 'shears' was the original sense of culter; rather is it a special sense of the plural of culter 'knife' (six times in Plautus), cf. habenae' reins': habena 'strap'. VIII POPULUS, POPULARI. The following suggestion as to the etymology of populus and populari will perhaps satisfy those who do not feel attracted by the 'dialectic' explanation that derives populus from *quoqlo-: Sk. cakrá- 'wheel', a wheel-shaped formation of the army: cf., for the signification only, corona 'audience', but also a circular military formation. Altogether the safest definition to adopt for populus seems to me to be 'army' (cf. magister populi), but 'army' as a 'fighting division', a 'detachment'. So the German word Schar, originally a division of an army, has come to mean in general 'multitudo'. I would therefore derive populus from po- (cf. pono, po-lio) and pello 'drive', whence populus = 'driving off',-a raiding party: cf. populari 'to raid'. For populus as a subdivision of gens cf. Livy, 6. 12. 4: simile veri est . . . aut non ex iisdem semper populis exercitus scriptos, quanquam eadem semper gens bellum intulerit, aut etc.; cf. also Aen. 10. 200 sq. ... POSTSCRIPT: The Editor of this Journal calls my attention to the British giant pair, Gog and Magog, which seems to have geminated from Gogmagog. Statues in Guildhall, known as Gog and Magog, are now understood to be statues of Gogmagog and Corineus. The name, Gogmagog, in turn, seems to owe its origin to "Gog of the land of Magog" (Ezekiel, 38. 2). UNIVERSITY OF TEXAS, AUSTIN, TEXAS. EDWIN W. FAY. REVIEWS AND BOOK NOTICES. Religion und Kultus der Römer, von GEORG WISSOWA, Ord. Professor an der Universität Halle. Pp. xii, 534. München, 1902. Beck. In a field of study in which there has been manifested for the past century a constantly increasing interest, and which has been, especially of late years (for it is only within comparatively recent years that the spirit of scholarship has with any measure of success searched the deep things of the Roman gods) the subject of so rapid a development, it is only natural that no work can long remain authority. Marquardt's Religion der Römer, first published in 1878, had been in use but a scant half dozen years when a revision became necessary; and the second edition of that work had hardly appeared when it became manifest that a new statement of the whole subject would soon be demanded. August Reifferscheid was looked to for the performance of the task, but his death occurring before the work had been begun, Georg Wissowa, who had been the reviser of Marquardt's work, and who had since 1882 shown a brilliant and productive activity in the field of Roman religion, entered in 1887 upon the preparation of the desired work, which has at length appeared, after a space of fifteen years, as Vol. V 4 of Müllers Handbuch der klassischen Altertumswissenschaft, under the title of Die Religion und Kultus der Römer. The knowledge that so long a period of time has been devoted to one work by a scholar who, even at the beginning of his undertaking, was already spectatissimus, justly arouses great expectations. It is sufficient praise for the author to say that these expectations are met by his work and that the conviction that the work was needed has been fully justified. Wissowa has brought to the performance of his task a thorough command of ancient sources, both literary, epigraphic, and monumental, as thorough a familiarity with the results of modern scholarship, and, above all, a measure of enthusiasm and devotion which has made it impossible for him to slight any part of his arduous undertaking; and the result is a work which is clear, comprehensive, and conservative, which is as nearly up-to-date as it is possible for a work to be in this field, and which shows a degree of scholarly ability of which even a nation of scholars may be proud. In the arrangement of his matter, Wissowa is not only logical and orderly, but original. After an introduction, in which he touches on (1) the difficulties of investigation in the field of |