with an IE variation (k beside ) in the initial consonant. More convincing would be Hirt's comparison of vaikanetaos, gen. sg. masc., with Gr. oikos, Skt. veçá-ḥ, if we could be certain not only that ywas preserved in Messapic (valevas, later balevas: Valetium, veretahetis: Veretum, vasti: ǎoru; oikoroihi being presumably a borrowing from Greek) but also that the diphthong oi became ai. Contrast the ethnicon Poediculi and the proper names oibaliahiai[hi: Oebalia, i.e. Tarentum, Verg., G 4.125; Illyr. Oeplus, CIL 3.2891, 2900; oitinai[hi: Ital. Vtius, Vtilius, see Conway, Italic Dialects 188, 198, 257. Kretschmer apparently believes that baoçtas1o (cf. bosat, and perhaps blaozzezihi) contains an original guttural represented as a sibilant, cf. perhaps Illyr. Beucas CIL 3.7830. But, beside Illyr. Beuzas (ib. 9156), Beusas -antis (ib. x test., xi, xiv, xvii), Beuzetius -ia (ib. 9929*), Ital. Buccia and Busius, Busidius, Bussenius (Conway, ID 155, 36, 34, 307) that view is untenable. The forms with c are evidently distinct from those with s, from which there is no reason to separate the Messapic forms. The same argument applies equally to daçtas (cf. dazet, dastas) which, if I understand Kretschmer aright, he regards with baoçtas as 'wertvolle Zeugnisse für den Guttural-Charakter des Messapischen'. Beside forms with -c-[k] Dacio, Daciscus, etc. (CIL 3), Duceus, Docetius (ID 34, 276) and the frequent Decius11 we have forms with -s (8)- which there is no more reason to suppose connected with those with -c- than there is to suppose that -k- or even -ki- became -s- in Messapic, e.g. Dasius, Dassius, Dasiatus, Dasimius, Dasumius, all common both in Italy and in Illyria, where we find also Dasa, Dasas, Dazas, Dases, Dasianus, Dazanus, Dasmenus, Dasto, see Indices to ID and to CIL 3. Accordingly I connect the numerous Messapic names12 with s(ç, s, ss, z, sz) with the Illyrian ones in s (ss, z) regarding the (Illyrian) names with c as quite distinct from both. There remains Kretschmer's view of barzidihi. Here again there are Illyrian forms, as we have seen, with g; and further, quite distinct from them, other forms with s which there is no reason to suppose represents an older g or k: Barsemis, Barsimia, Barsimsus (CIL 3.10307, D lxvi), and possibly (with an 'Cf. perhaps Gr. ȧloow, root*чāžik-?

10 On ç and 3, see Whatmough Class. Quart. 19.68.

"I do not, of course, suggest that Dac-, Doc- (Duc-), Dec- are necessarily cognate; though Doc- may stand to Dec- in the same relation as docere to decère. 12 dassinar, daçta, daçtas, daoras, daszes, dazes, dazet, daze[, dašel, daševi, dazeziai, dazeh[i]as, dazes, daszes, dazihi, dazetis, dazetʊes, dazetēihi, dazihonas, dazimaihi, dazimas, dazomas, Aáσμos, dazohonnihi, dazonnes, dačos, dazohi (?) δάζου, δαζυ.

older er preserved?) Bersumno, Birziminium, Berselum, Bersula, cf. Krahe 83. Nor again should Messapic plastas (gen. sg. masc. from *plaset, plazet) be connected, as it was by Deecke, with Placens, Placentius, but rather with the names Plassarus (CIL 3.4376) or (?) with Blasa, Blazziza, Blassius (ib. 7635, 8292, 3074, 1650, 4150, 5498) with p:b as in Messapus: Mérabos, IIvous: Buxentum, Clampetra: Clambetis. 13 Deecke's account of haçtorres (= *ExTópios) is equally untenable. I connect this form with Latin hostis, cf. Hostilius (ID 32, 34-Calabria, Peucetii; represented also in the borrowed hos@ellihi with Ital. -o-?), gh being represented by h as in the pronominal hi-, in dehatan (:Lat. fingo), and perhaps in mahehe (:Skt. mahā'n?). Finally βίσβη· δρέπανον (and βισβαῖα κλαδευτηρία) which H. Petersson interprets1 as *uik-ya, comparing Skt. veç?' 'needle', Arm. gišel ‘tear, split', may just as well stand for *uis-ua 'the divider, cutter, German Messer', cf. Skt. vişva- 'on both sides, on different sides', IE *ui- 'two', as in *ui-dh- 'divide, separate', cf. OCSI věja, Skt. vaja' 'branch, bough', Lat. virga for *viz-gā; and if Messapic did not sibilise the palatal stops, then Petersson's etymology becomes untenable.

There are a few instances in which it has been proposed by Ribezzo (23ff.) to see examples of the sibilisation of ĝ or ĝh. These may be easily disposed of. Thus the series of names azen, açen, azinne, azena in which Ribezzo (28) sees IE *ĝen-, is far more likely to be connected with the names Asinius, Asenius (CIL 3.7118, 10765, 8897, 8895); and the name zarres with Sarius (ib. 1204493; 5.8115108; Pais, Suppl. 1080441, 1182) cf. Sarina, Sarus, Sarronius, Sarnus (CIL 3 Index) and zairikihi with Saerius (ID 155) than with Gr. xaipw, with which Ribezzo (26) connects them both. The two forms azinnota and inzanixis according to Ribezzo (29) both contain the same root IE *ĝen(cf. azen etc. above) and mean respectively 'creavit' and 'finxit'. Taking Ribezzo's view of the meaning, though that is quite uncertain, we may quite as reasonably see in -zin- -zan- the root *sen- (son-) 'achieve, complete', cf. Gr. avvμ, Skt. sanóti.

I should not have thought it necessary to remove these examples, apparently contrary to the view which I take, namely that the palatal stops were not sibilised in Messapic, if there had not been also a con

13 Cf. Krahe 89. It would also be possible to separate plastas from the names given above and to connect it with Plarius, Plarentius (ID 155, 257, 375, cf., CIL 3.6183, D xxiii, C vi test., xiv), intervocalic -s- being preserved in Messapic (e.g. Canusium, Genusia, Galaesus, lasofihi, Busidius and the names cited above), unless in Plarius etc. -r- is original.

14 Only the summary, Glotta 15.9 (1926), is known to me.

siderable body of positive evidence in favour of that view. To this I now pass.

Apart from klaohizis, already discussed, we have, as against z from ĝ (azinnota etc.) oroagenas '(citizen, native) of Uria, Uritis' with -r- for -rr- from -ri-, and with -oa- indicating locality as in a large number of names (e.g. daran@oa, dalma@oa, kritaboa) though the form might conceivably be merely a patronymic. What cannot be doubted is the equivalence of -genas to Latin -genus, Venetic -xeneh (gen. sg.) in which -g-, -x- represent IE 9 (Skt. jánaḥ) especially when we find also genollihi gen. sg. masc. (cf. with a different suffix, Genucius CIL 3.2535, 4471, 141472) and the local name Genusia (Peucetii, ID 33). In the river name Vergellus (ib.) it is probable that we have the same root *ver (e)ĝ'bend, wind' as in Latin uergo, Skt. várjati. The meaning of Messapic argorian (cf. argora-pandes an official title) is admitted by every one to be 'silver, money', and here again there is evidence of ĝ represented by g (cf. Lat. argentum, Gr. ǎpyvpos, Skt. rajatám), for in view of the place names 'Apyv pivo (Epirus) 'Aрyúρinа (Apulia) the supposition that argorian was borrowed from Greek is quite needless. The Peucetian local name Geronium (ID 35) also probably contains IE ĝ (cf. Lat. grānum, Gr. yépas, yépwv: Skt. jaráḥ) and in y in the suffix of 'Iáπvyes (cf. 'Oprvyia) we may have either ĝ or g. It would be possible to connect Anxa (the older name of Callipolis) with either Lat. ancus or ango, referring either to the shape or to the character (narrowness) of the harbour; in the former case the word would be indecisive (IE k), but in the latter we should have g (x = g + s) for IE ĝh gs) after n as in brigannas (see above) after r. The local name trigonoxoa appears to show *ĝon- 'corner, bend' (Lat. genu, Gr. yóvu, ywvia: Skt. jā'nu) though borrowing from Greek (cf. Tpiywvos) is possible, and in konkolastis seems certain (k, gh), cf. Gr. kóyxos, kóyxn. The word agrafos seems to show ŷ (Lat. ager, Gr. ȧypós: Skt. ájraḥ) and if it is a proper name, as is likely, it will be parallel to such Latin names as Agrius, Agrestius, Agreius. There are three names in gor(gor, goro abbrev., gorrih[i, gorvaides, gorretavidihi) which are not decisive, since in these g is more plausible: cf. either Lat. gurges, root ger-, or Gr. yupós, also with g-, Lat. būra. Doubtless gronehias stands in some relation to Granius (Calabri, Daunii ID 32,36), but it is not clear whether we have gh (ĝh or gh?) represented before r by g, cf. Lat. frendo (*gheren-d- an extension of ĝher-, see Walde, s.v. and compare names like Frensidius, Fresidius: frēsus ?); or ĝ (*ĝerā2-: ĝr-, see Geronium above); or even g (*gren- with -dh-extension perhaps

in Lat. grandis). In magos it is tempting to see *maĝ- as in Latin magnus (cf. perhaps maeos: Osc. mais, Lat. maior, mazzes with -zzfrom -gi-?) beside *maĝh- in mahehe (above), while in xonedonas, xonet@es beside Xάoves, Xŵves we have presumably to do with borrowing, though x is probably equivalent to k (not to kh) cf. taimakos: Δαίμαχος.

Reasonably clear instances of IE & represented by k, both initially and medially, occur apart from the form klaohizis (see above), in the following words: korah[i (or -[aihi?) with ō-grade; or, less likely, with epenthesis (as in Greek), ou being written o; or, conceivably with loss of after r and 'compensatory' lengthening, Gr. Koūpos Att. kópos, Lat. Cerus, creo: Skt. çardhaḥ 'herd', cf., if a proper name, 'ETI-KOU POS (?). Peucetii: Greek Teún, Lith. puszis 'pine', either with the diphthong preserved as in @eotoras, or more probably with the (Greek?) spelling eu. kordomaos: Cordus, Cordius (Calabri), Lat. cor, OCSI srudice 'heart'; with -om- (-um-) as in dazomas beside dazimas (cf. Lat. -imus, -umus) and ō in the initial syllable. Greek 'I for 'E?), cf. Venetic Ecco, ekupe@aris. ter'?) :dico (k, cf. Skt. diç-).

Tarentine "IKKOS (with

dikoteras (quasi 'Dex

vaikaneataos (?) see above.

calare. kri@onas, kritaboa. balakrahiaihi. kra@eihi.

IE k (or k alternating with k and therefore equally indecisive for our present purpose) may occur in the following instances, which it will be sufficient merely to enumerate. kelonihi15 (cf. Venetic kelo: Lat. Celsus, but with either k or k if connected with Lat. Celer, Cillius, see Walde s.v. Luceria, Leuса, λEνкaν. Lacinium promonturium. nerikiden, cf. Ven. nerikah. Canusium: KÓVIS ? kalatoras Lat. saihikas. inkermali : κρεμάννυμι ? koileihi. kavasbo : καίω ? kraapati carpo? hipakali: scalpo ? For IE k I find no other certain instances than the one already cited (penkahel) except dokihi : Sicel AOUKÉTLOS, Lat. Docetius, Duceus, dūco. It would be easy to multiply conjectures, or to write at length on the examples discussed in this note. But nothing is gained by that method, which has too often marred the earlier stages of investigation of little known or ill preserved dialects. All I am concerned with here is to point out that the view that Messapic was a satem-speech is based upon quite inadequate evidence, and that it has been too hastily accepted by Kretschmer and others.16

15 Cf. kilahialhi? But why -i-?

16 Cf. Jokl in Ebert's Reallexicon, s.v. Illyrier 6.41 (1925).

LOSS OF FINAL n IN INFLECTIONAL SYLLABLES OF MIDDLE

ENGLISH

SAMUEL MOORE

UNIVERSITY OF MICHIGAN

In no respect perhaps do the Southern and Midland Middle English texts of the thirteenth and fourteenth centuries show greater differences than in the loss or retention of the final n of unstressed inflectional syllables. Disregarding all but the gross differences, we can recognise three types of distribution to one or the other of which the various texts tend, in the roughest way and in varying degrees, to conform. These three types of distribution may be illustrated by the Owl and the Nightingale, the Cotton Nero A 14 text of the Ancren Riwle, and the London English of Chaucer.

In the Owl and the Nightingale we find an approximation to the complete loss of final n in all the unstressed inflectional syllables that developed from the Old English endings -an, -um, -on, and -en: i.e. in both the singular and plural of weak nouns, in the weak adjective inflection, in the dative singular and plural of the strong adjective inflection, in the dative plural of strong nouns, and in the present subjunctive plural, the preterit indicative and subjunctive plural, the infinitive, and the past participle of strong verbs. The loss of final n is approximately complete in the singular of weak nouns and in the strong and weak adjective inflection. There are a few examples, however, of -en as the plural ending of nouns that were weak in Old English and a very small number of analogical -en plurals of nouns that were not weak in Old English, such as children; one or two plural forms also may be interpreted as the Middle English development of the Old English strong dative plural in -um, e.g. of heore sunnen, 858. In verbs the final n is lost in at least 80 per cent of all the plurisyllabic forms that were entitled historically to the ending -en.

In the Cotton Nero A 14 text of the Ancren Riwle there is an approximately complete loss of final n in the singular of weak nouns and in the strong and weak adjective inflection. But final n is nearly always retained in the plural form of weak nouns and there are numerous ex