a date after 462-1 is suggested by the character of the introductory formulae.

Some secretaries can be identified with prominent Athenians. Kallistratos (I. G. 12, 310) was probably the hipparch who fell in Sicily in 413. His name should be restored K[a]Mλiorparo[s 'Eμπédо 'Olev] from I. G. 12, 302. .xos in I. G. 12, 324, line 25 is certainly . . . Xuños, and should probably be restored as ['Eπi]λUKOS, son of Teisander and uncle of Andocides. Awpółeos, I. G. 12, 313, secretary in 408-7,was probably the ambassador to Persia in 409, Xen. Hell. 1, 3, 13. Mevékλns of I. G. 12, 220, 370 was probably the man mentioned by Xenophon, Hell. 1, 7, 34.

23. A Proposed Emendation of the Res Gestae Divi Augusti (Paragraph 9, First Sentence), by E. Adelaide Hahn, Hunter College.

We find the sentence printed as follows:

(1) By Theodor Mommsen, Res Gestae Divi Augusti ex Monumentis Ancyrano et Apolloniensi (Berlin, 1883): Vota pro valetudine mea suscipi per consules et sacerdotes quinto quoque anno senatus decrevit.

(2) By Ernst Diehl, Res Gestae Divi Augusti: Das Monumentum Ancyranum (Bonn, 1918): Vota pro valetudine mea suscipere per consules et sacerdotes quinto quoque anno senatus decrevit.

(3) By David M. Robinson, The Deeds of Augustus as Recorded on the Monumentum Antiochenum (Baltimore, 1926): Vota pro salute mea suscipi per consules et sacerdotes quinto quoque anno senatus decrevit.

My proposed reading is: Vota pro salute mea suscipienda per consules et sacerdotes quinto quoque anno senatus decrevit.

Concerning the alternatives valetudine-salute, on the basis of the arrangement of lines, and of the Greek translation, I conclude that either is possible; however, I incline to prefer salute.

Concerning the alternatives suscipi-suscipere, I incline to prefer the passive in view of the general sense and the presence of the per phrase, although Augustus's usage elsewhere with per would justify either voice. However, the simple passive infinitive with subject accusative does not seem possible here after decrevit. The active (complementary) infinitive is syntactically less difficult. But certainly either a subjunctive or a passive periphrastic infinitive seems called for. Moreover, the Monumentum Antiochenum fragment looks as if the reading were pi, which would force us to reject the active suscipere in any case.

I therefore propose the emendation suscipienda. This is perfectly possible so far as line arrangement goes, on both Monumentum Ancyranum and Monumentum Antiochenum. Supporting arguments may be advanced on the basis of (1) Latin syntax, as recorded by the leading grammars; (2) Latin usage, as adduced by the Thesaurus Linguae Latinae; (3) Augustus's usage, as exemplified in the Res Gestae; (4) the testimony afforded by the Greek translation of the Res Gestae.

24. Apophoreta, by Samuel Loomis Mohler, Franklin and Marshall College.

The open-handed nature of Roman hospitality is evidenced in the entertainment of clientes and umbrae, and in the custom of allowing the guests' slaves to carry home napkins filled with "samples" from the meal.

Augustus's dinner-favors at times consisted of valuable presents, while again they might be mere practical jokes. He also held raffles, and on at least one occasion supplied his guests with stakes for the gambling which was expected to furnish the evening's entertainment. At Trimalchio's dinner some of the favors were fantastic in themselves, others were distributed in an unusual way (through a hole in the ceiling), while still others owed their interest to the far fetched puns written on the cards which accompanied the various objects. Elagabalus carried the element of chance to extremes in having his guests draw lots for such things as ten camels or ten mice.

The last book of Martial's Epigrams contains our fullest source of information as to what might be given at a Saturnalian banquet. The Italian's fondness for gambling is shown in the grouping in pairs, gifts "for the rich" and "for the poor." Parallels are cited to show that some of the commonest Saturnalian presents were also typical dinner gifts, and that even the most valuable of the objects described were such as were actually given. The poems accompanying the "booby" prizes not only exhibit the Roman's sense of humor, but also show an essential kindliness of nature: many of them are toys to be given to the guest's own children or his vernae. Others are designed for women-not all of them wives or sisters.

25. The Phonology of the Messapic Dialect, by Joshua Whatmough, Harvard University.

It has been too hastily assumed on the evidence of penkeos, penkaheh[e alone, the meaning of which is unknown, that Messapic must have been a satem-speech, in spite of the k (representing I.-Eu. k) of klaohizis, klohizis, though it is admitted that this verbal form (2 sg. aor. opt.?) is, standing alone, equally unconvincing; many etymologies of Messapic words (as Ribezzo's account of azinnota, Petersson's of ẞioßn) are based on this hasty assumption, and where (as is usually the case) even the meaning has to be assumed, such etymologies are doubly uncertain. It would be easy to account for all forms supposed to show sibilised palatals, even accepting the assumed meanings. But the evidence of a few proper names, local and personal, suggests that most probably Messapic, like Venetic, did not represent original I.-Eu. palatals by sibilants. Thus we have genollohi, oroa-genas, Genusia, Genucius all probably containing the root ĝen-; similarly in Vergellus, a river name, may be seen the root uerĝ- 'turn, twist.' brigannas may contain in g either ŷ or ĝh, and this form is distinctly against Kretschmer's etymology of barzidihi. The Tarentine "IKKOS is possibly a Graecised form for Messapic *KKOS, cf. Venetic Ecco. Both Messapic and Venetic have a name-stem kelo- (Ven. kelo, n. sg. m., Mess. kelonihi gen. sg. m.), but here we may have either k or qu. But dokihi (-0- =ou or u), cf. Sicel AoUKÉT LOS, Peucetii (the -eu- being due to Greek influence?), dikoteras and korah[i (or -[iaihi?), all clearly show k for k. If therefore penkeos, penkaheh[e are to be connected with Lat. quinque (Osc.-Lat. Pompeius etc.), it must be concluded that Messapic, like Venetic again, is not clearly either a centum- or a satem-speech. The position of ancient Illyrian in general, in fact, is to be reckoned with in connexion with the entire problem of the I.-Eu. guttural series.

26. Indic Speech and Religion in Western Asia, by Edgar Howard Sturtevant, Yale University.

The Aryan names of men, gods, and a city, and the Aryan technical terms of horsemanship which are quotable from Assyrian and Hittite documents of the fourteenth century B.C. are all Indic rather than Iranian. Kretschmer's etymology of Varuna (from Hittite arunas "sea") proves that the main body of the Indic nation passed through Asia Minor or northern Mesopotamia. They were in the Zagros mountains as early as the eighteenth century.

Since the art of cuneiform writing on clay tablets was not carried to India, direct contact with Babylonian civilization must not be assumed. But Indic civilization was left behind in Mitanni, and influenced both Hittites and Assyrians. The Assyrian god Ašur is Indic Asura. There are some reasons for thinking that Indic religion is responsible for Ihnaton's worship of the sun-god Aton; and Indic influence on Hebrew religion is possible.

27. The "Ab Urbe Condita" Type of Expression in Greek and English, by E. Adelaide Hahn, Hunter College.

Two signs on a garage-"Battery Charging" and "Batteries Charged"—exemplify two distinct linguistic tendencies. The second tendency underlies several common Latin constructions: ablative absolute, gerundive, ab urbe condita. But Latin does not use this type, as do Greek and English, after verbs of sensual perception.

This tendency, very common in sixteenth and seventeenth century English and French, is gaining ground anew in English. Thus frequently even good writers use after a preposition an objective noun plus a participle instead of a genitive noun plus a gerund— e.g. "about Mrs. Templeton being in love." The usage does not as yet seem extended to pronouns, except in illiterate speech.

Analogous cases, perhaps justifying this usage, may be cited from Greek, as ἤκουσα περὶ Δαρείου τεθνεῶτος, and even from English, as "the rumor of plundered caravans" (= "the rumor of caravans plundered" "the rumor of caravans having been plundered"). Comparable instances with adjectives instead of participles are also found.

=

Similar turns are especially common when brevity is desired, as in newspaper heads (probably, however, having a different origin, the omission of the copulative verb), and proverbial expressions (as "Too many cooks spoil the broth," cited by Jespersen).

The usage seems particularly common, in both English and Greek, in prepositional phrases; but we find instances of ab urbe condita phrases used otherwise, e.g. as subject of a verb.

When the participle simply repeats or reinforces the noun in sense, and could be omitted without appreciable change in meaning, instances are not particularly striking. But when the modifier has a negative or privative force, the passage may become ambiguous, as "[Despatches tell] of the delayed arrival of Lieutenant Locatelli" (which may mean either that Locatelli did or that he did not arrive); or amusing, as "all traces of his wrenched foot will disappear."

28. The Accusative and Ablative of Degree of Difference, by Bernard Melzar Allen, Roxbury School, Cheshire, Conn.

Extent of space or time is expressed by the accusative, a use probably developed from the notion of limit or goal. It easily passes on to the general notion of extent or degree, where no special notion of space or time is involved. One particular kind of degree, the degree of difference, is also expressed by the ablative, but by no means exclusively. Verbs of difference regularly express degree by the accusative, comparative adjectives and adverbs, by the ablative. Distance away in space or time may logically take either case, for this notion is not very different from that of distance covered and also indicates a difference.

Constructions of space and time are notably parallel in all languages, but distance away in space takes the accusative to express degree some two or three times as often as the ablative, and distance away in time regularly takes the ablative, mostly with post and ante.

One reason for this difference may be that distance away in time involves a comparative notion of earlier or later (not usually present in distance away in space), and so would be likely to take the same case to express degree as do comparative adjectives, in connection with which the ablative of degree of difference probably originated, and surely has its most notable development. Another possible reason for the difference is the confusion likely to follow between an accusative of degree used to modify the adverbs post and ante, and the accusative object of these words when used as prepositions. Phrases like ante multos annos, now usually considered prepositional, may originally have been like abhinc multos annos, an adverb with an accusative of extent.

In phrases like ante diem tertium, the accusative may have been originally one of goal or limit "before, as far as the third day." In the phrase post diem tertium quam, the quam shows still more clearly the adverbial feeling of post. Both these phrases, considered as prepositional, are illogical or inaccurate.

29. The Nature of the Latin Passive in the Light of Recent Discoveries, by Edith Frances Claflin, Rosemary Hall.

The presence of the element r as a prominent characteristic of the Latin deponent and passive conjugation has led to the theory that the Latin passive is composite in its nature, being made up in part of middle endings and in part of the impersonal suffix -r. In a paper read at the Third Annual Meeting of the Linguistic