plebeians; and fourth, Nitzsch's view that the family were protectors of the urban trading classes and advocates of commercial expansion. This paper seeks to summarize the results of an investigation which collected and analyzed all the ancient testimony bearing upon the policy of the Claudian gens from the year 495 to the year 133 B.C. As the evidence on which my conclusions rest is mainly cumulative, it will be impossible to present it here. It will appear in full, however, in an article in the next number of the Harvard Studies in Classical Philology.

In the first place, the ancient view that the Claudii were ultra-patricians is enforced largely by rhetorical speeches which are assigned to the earlier and more obscure members of the family by the ancient historians. The tone of these speeches is due, not as Mommsen supposed, to the falsification of the history of the family by some early annalist, but to two traits in the Claudian character which deceived the later Roman and Greek historians. The first of these traits is arrogant and tactless pride, the second a persistent opposition to the policy of the tribunes. We see this Claudian stubbornness, for example, in the career of Appius, the decemvir, in the admission of the sons of freedmen to the senate by Appius Claudius Caecus, in the connection of Publius Claudius, consul of 249, with his freedman Glicia, and above all in the almost contemptuous disregard for religious ritual which characterized so many members of the family from the time of Appius Claudius Caecus, who tried to do away with the carousal of the pipers, to that of Publius Claudius, who drowned the sacred chickens. Opposition to the policy of the tribunes, which may be called the first article in the creed of the Claudii, is illustrated by the attitude of the decemvirate to the tribunician families like the Icilii and Duilii, by the hostility of the tribunes to Appius Claudius Caecus, Publius Claudius, consul of 249, and Gaius Claudius, censor in 169, not to mention the minor members of the family.

The second article in the Claudian creed, and indeed the central fact in the history of the family, is the unswerving patronage of the libertini, seen especially in the improved position given them in senate and tribal ratings by Appius, the censor, in the reforms of Gaius Claudius, censor in 169, and in the close terms of intimacy existing between Claudian patron and client, as in the case of Appius Claudius Caecus and Cn. Flavius, Publius Claudius and Glicia.

As the libertini were perhaps the leading element in the plebs urbana, we find a consistent effort made to improve the condition of this class of petty shopkeepers, mechanics, and traders on the part of the great leaders of the Claudian gens. This was done by constructing great public works which improved their economic condition, by improvements in their civic position, and above all by constant efforts to expand the commercial and political sphere of Roman influence. Here Appius Claudius Caecus, Claudius Caudex, and Gaius Claudius, censor of 169, are the most important figures.

In general, then, the theory of Nitzsch comes the nearest to accounting for the facts. It errs, however, in identifying the plebs urbana with a group of transmarine merchants rather than with the larger body of petty hucksters recruited mainly from the freedman class.

Remarks were made on this paper by Dr. H. A. Sanders.

1 Geschichte der Römischen Republik, pp. 67 ff. and

103 ff.

3. The Early Greek Alphabets in the Light of Recent Discoveries in Egypt, by Professor William N. Bates, of the University of Pennsylvania.

In this paper the writer called attention to the great importance of Petrie's recent discoveries in Egypt for the study of the early Greek alphabets. He showed that the old tradition that the Phoenicians were the inventors of the characters which formed the Greek alphabet was erroneous, and that, in short, all of these characters were in use thousands of years before the Phoenicians existed as a nation. The characters discussed were engraved upon fragments of vases which were found by Petrie in great abundance at Abydos in Upper Egypt, and date from the First Dynasty and from the earlier prehistoric period. Petrie would set as a date for these prehistoric fragments about 6000 B.C.1 The writer drew up the table shown on following page, based upon that published by Petrie, showing the early characters and their later forms as they exist in Egypt.

It will be seen at a glance how important these characters are for the study of the Greek alphabet. All the theories as to the origin of the so-called supplementary signs 2 now fall to the ground, for here we find v, o, x, y, w, and § fully developed at the earliest period. What is more, peculiar forms such as the Corinthian beta, the Melian beta, or the Arcadian psi-forms which have always been a puzzle- are here found along with the more familiar signs. In fact, the number of characters is considerably in excess of those required by the normal Greek alphabet. This proves, first, that the peculiar characters found in some of the Greek alphabets are not arbitrary signs invented at a later period, but that they are as old as the more familiar signs; and second, that there were other signs which did not survive until Greek times.

In regard to the story that the Phoenicians invented the alphabet, Petrie suggests that a certain number of these signs may have been used for numerals and so have been written in a fixed order, and that the Phoenicians may have taken them over for use as letters. There seems to have been a tradition to this effect in Crete, for Diodorus (v. 74) in a passage pointed out by Six says, parì (i.e. oi Κρῆτες) τοὺς Φοίνικας οὐκ ἐξ ἀρχῆς εὑρεῖν ἀλλὰ τοὺς τύπους τῶν γραμμάτων μεταθεῖναι μόνον.3

However this may be, the alphabet did not originate with the Phoenicians. Who the inventors were we cannot say. The characters are not Egyptian so far as Egyptologists can now tell. They may have originated, as Petrie seems to believe, with the prehistoric inhabitants of Egypt, but more than that we cannot say. At all events, as Petrie well points out, the history of the alphabet is as old as the history of civilization itself.

4. Some Text Emendations to the Rig-Veda, Atharva-Veda, and Kena Upanishad, by Truman Michelson, Esq., of Harvard University. I. Kena Upanishad, second Khanda: the fifth verse runs as follows: iha ced avedid atha satyam asti na ced ihī’vedin mahatī vinaṣṭiḥ: bhūteṣu bhūteṣu vicitya 1 Royal Tombs of the First Dynasty Part I, p. 31.

2 For the principal theories see Larfeld, Griech. Epigraphik, p 516 ff.
3 See Evans, Cretan Pictographs and Pre-Phoenician Script, p. 103.

dhīrāḥ pretyā'smāl lokīd amṛtī bhavanti. This is usually considered prose, but as what preceeds is metrical, we should be inclined to consider this also metrical. And by reading mah! for mahat: and considering1 iha ced as an anapaest substituted for an iambus, it is metrical. iha ced aved.d atha satyam asti na ced iḥī 'vedin mahi vinastiḥ: bhūtesu bhūtesu vicitya dhirāḥ pretyā 'smal lokād amṛti bhavanti is good tristubh metre.

II. Atharva-Veda I. 8. 4. This passage is very hard both in metre and sense. By the courtesy of my honored teacher, Professor Lanman, I read in proof-sheets what the late Professor Whitney said in his translation of the Atharva-Veda which is now rapidly being completed. "The irregular metre and broken connection of the second half-verse suggest possible corruption of the text." By reading etams for tams, and pronouncing tuám, jahi, and eṣaam, the metre of the second half-verse is restored.2

etáṁs tuáṁ bráhmaṇī vāvṛdhānó jahi eṣaam çatatárham agne.

III. Some observations on Rig-Veda I. 61.3 In this hymn there are several lines of tristubhs of ten syllables which can be emended without much violence so as to be of eleven. I do not wish it to be inferred, however, that these emendations are necessary, as the cadence and not the number of syllables is the main point in Vedic metres. As is known, the tristubh of ten syllables sprang

from a mixture of tristubhs of 4 + 7 syllables and 5 + 6 syllables.

Verse 4 a: asma id u stomaṁ sám hinomi. By reading ahám for sám we have a line of eleven syllables. In R.V. I. 184. 4, root hi occurs without a verbal prefix, so that we need have no scruples for striking out sam.

Verse 6 a: asma id u tváṣṭā takṣad vájram. Here ataksad, the augmented form, would make the line of eleven syllables, but, even then, the ninth syllable would be long where a short is wanted. Of course, in the saṁhitā-text takṣad and atakṣad would appear the same. Possibly we should also pronounce vájṛam, and then read the line as a jagati. For vájṛam cf. rudṛám R.V. I. 114. 4. Indra also is sometimes to be read indṛa.

6c: vṛtrásya cid vidád yéna márma. To be sure, we can pronounce vṛtrasia, but this will scarcely recommend itself, for genitive singulars in -sia are of exceedingly rare occurrence. If ávidad be read, we have a good line of eleven syllables.

10 a: asyéd eva çávasă çuṣántam. We can resolve asyéd into asyá id, and then the line will have eleven syllables, but observe that in verse 9 a, asyéd can not be so resolved.

10 b: vi vṛecad vájreņa vṛtrám indraḥ. By reading avṛçcad, the augmented form, we have eleven syllables. Furthermore, it has this to recommend itself, viz. that this avoids the pause after the third syllable. As is known, the pause usually is after the fourth or fifth syllable (A. Bergaigne and V. Henry, Manuel, § 122). Of course, it is also possible to pronounce vájṛena, but this latter does not commend itself to me, as we should still have the objectionable pause after the third syllable.

1 Or we can read satyam asti with the elision of -m and crasis.

2 I am aware that Professor Bloomfield, SBE xlvii., p. 239, suggests resolving both tains and tvám or inserting jahi after tvåm. I cannot accept the proposed resolution of tams; and reading etains is easier than inserting jahi

3 This hymn occurs without variants A.V. XX. 35

Or shall we read takṣata? This would make the ninth syllable short.

11 b: pári yád vájreņa sīm dyachat. To make this line of eleven syllables, we must pronounce vájṛena.

12 c: gór ná párva vi radā tiraçcă. If we substitute the fuller ending, -āni in párva, we have a line of eleven syllables.

13 a: asyid u prá brūhi pūrvyani. We may here resolve asyéd into asya id. In any case, we must pronounce pūrviāņi.

13 b: turásya kármāni návya ukthāiḥ. We can here substitute návīya for navya to make the line have eleven syllables.

14 d: sadys bhuvad viryàya nodhaḥ. Here the augmented form abhuvad will give the verse eleven syllables (but see Oldenberg Prol. Rig-Veda, 174). Two other lines are a little more difficult:

4 c: giraç ca girvāhase suvṛkti. Unless we pronounce girvaahase, we must insert te before suvykti to make the verse have eleven syllables.

15 d: săúvaçvye súṣvim āvad indraḥ. It is possible to insert sd before āvad or sușvim. Săuvaçvye must be pronounced -vie.

5. Brief Notes on Thucydides, by Professor W. S. Scarborough, of Wilberforce University.

II. 4. 2: Toû μǹ èкþeúyeɩv denotes purpose, not result. But it is better to read, with Dobree, rò for ToÛ. - II. 5. 7: d' ovv means nevertheless.' Cp. I. 3. 5.-II. 18. 3: Evvaywyń is to be taken metaphorically ('war-clouds'). – III. 16. 3: ös eμeλλev K.T.λ. is not ambiguous or superfluous. As admiral Alcidas would naturally conduct the expedition. III. 31. 1: σplot is to be construed with γίγνηται. III. 82. 1: the sense demands either two finite verbs or two participles.

6. The Harpalus Case, by Professor Charles D. Adams, of Dartmouth College.

This paper appears in full in the TRANSACTIONS.

7. Notes on Tacitus and Vergil, by Professor H. W. Magoun, of Redfield College.

In a well-known passage of the Agricola (x. 6), Tacitus writes: Unum addiderim, nusquam latius dominari mare, multum fluminum huc atque illuc ferre, nec litore tenus adcrescere aut resorberi, sed influere penitus atque ambire, et iugis etiam ac montibus inseri velut in suo. What does he mean by the word fluminum? Church and Brodribb render the passage: I would simply add, that nowhere has the sea a wider dominion, that it has many currents running in every direction, that it does not merely flow and ebb within the limits of the shore, but penetrates and winds far inland, and finds a home among hills and mountains as though in its own domain.' Others translate the word 'currents in the sea.' But are currents in the sea flumina?

So far as I have been able to discover, all seem to have this idea. Draeger to be sure, refers the passage, and refers it correctly, I believe, to the narrow bays and inlets of the coast; but he changes the reading to fluctuum. Is the