STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. NOTES ON ATTIC INSCRIPTIONS By Allan Chester Johnson In the new edition of the Attic Inscriptions 1 the documents are grouped in divisions, and in chronological order within each division as far as possible. In Division 1, Kirchner has placed the inscrip- tions which date between 403 and 378 b.c. When the archon's name is lacking, the historical content of the document is the chief factor in determining the date. When we have neither archon nor history as guide, the forms of the letters become the chief basis of dating. The great importance which formulae have in the third century in dating inscriptions is universally acknowledged. The neglect of these for the first period has led to much confusion in the arrangement of the various documents, and the history of the financial boards which are called upon to defray the cost of recording the decree on stone becomes strangely involved. Apparently the ran'iai, ctaroSeKrai, ra/was rfjs jSokfjs, ra/uas to Sri/xo and rafdai ru>v rrjs 6eo are called upon according as the mood of the legislating council or assembly may direct. I am inclined to believe that various boards could not be so ordered within the limits of the same year, but that their activities were confined to different periods. In 405 b.c. the 'EXX^jwa/ueu pay the expenses (Nos. 1, 39), but in 403 b.c. this board is replaced by ol rafxiai. I have little doubt that these are the joint stewards of the treasures of Athena and the other gods, for no other board would be referred to as "the stewards." Apparently they took over some of the duties of the Hellenotamiai at the close of the Peloponnesian War. In No. 4 Kirchner has restored rafdai rrjs deo, but this is manifestly an error because such a board did not exist in 402 b.c. 2 In Nos. 1, 2, 3, 4, 51, and 57 these ra/uai pay the costs of the inscription. These must all date before 386 b.c, and probably belong to the first few years of the fourth century, 1 Inscriptiones Graecae, Editio Minor, Vols. II and III, Pars I, Fasc. I (cf. Class Phil., IX [1914], 455). 2 It is quite certain that the joint board existed until 390/389 b.c. (JG, II, 660; cf. Ferguson, Athenian Secretaries, 72). There is good evidence for believing that they were amalgamated until 387/386 b.c. (cf. Ferguson, loc. cit.). [Classical Philology IX, October, 1914] 417 418 Allan Chester Johnson perhaps not later than 394 b.c. At any rate this board was dis- placed at some time before 387 b.c. by the ra;uias to &7;uo. In No. 56 this steward pays the secretary of the people thirty drachmas for a decree to be set up ev iroXet (cf. No. 53). If No. 21 is correctly dated we hear for the first time of a fund provided for inscriptions [en t&v Kara if/7]4>iv ttjs 8eo at some time between 385 and 379 B.C. The latter board pays the cost out of a fund of ten talents beginning ca. 378 B.C., but before that date the source of the funds is not stated. Perhaps the earliest dateable inscription recording the stewards of Athena as paymasters is No. 245, which Kirchner for some unex- plained reason has placed ca. 352-337 B.C., although the lettering and the formulae imply an early date, as he himself says. From the contents of this inscription I should infer that it followed the capture of Thebes by the Spartans, ca. 383/382 b.c, when many from Thebes and Boeotia took refuge in Athens. If so, this inscription contains the last dateable example of kv irokei for h> aKpoirbXei,. Dittenberger has claimed 386 as the lower limit for the old formula, but the number of dateable inscriptions between 390 and 380 b.c. is so small that it is very unsafe to fix any date arbitrarily beyond which we can say absolutely that h> irb\ei must not be used, ev aapoToKei occurs first in 394 B.C. (No. 19); apparently it gradually supplanted the older phrase, and by 383/382 b.c. the latter was finally discarded. By 383/382 b.c, or possibly in the preceding year, the awodeKrai were supplanted by the rapiai, tuv ttjs deo. The latter discharged the duty of paying for the inscriptions until ca. 376 b.c. There is one apparent exception, for Kirchner dates No. 40 in 378/377 B.C., and, if so, the aTroSiicTat. and the rapiax share 420 Allan Chester Johnson duties in the same year. Such a division is most unusual and could only be explained by supposing some violent change in political leadership during the year, though the history of the year does not lead us to believe that such was the case. Since Mytilene was appar- ently in alliance with Athens even after the Peace of Antalcidas (Isocrates xiv. 28; cf. Kirchner, note to No. 40), there is no reason why No. 40 1 should not be dated in the year 387/386 b.c. and con- temporary with Nos. 29 and 31. I believe therefore that the ra/ucu rrjs Bed paid for the decrees from ca. 383 until ca. 376 B.C. without a break. During the years 378-376 there is a fund of ten talents which they draw upon. Nos. 22, 43, 84, 141, 173, and 197 should be dated in these years. There is little doubt that the second decree in No. 22 is later than 390/389 b.c, and the chief argument is that the rajutat rrjs 8eo did not exist as a separate body in that year. If No. 173 is dated in the archonship of Charisander we might be able to determine the limits of the board of stewards with some exactness, but Koehler observes that the restoration of lines 8-9 is very doubtful. The termination of the connection of the rafxlai rrjs did may come ca. 377/376 B.C., when we know that the board was temporarily disgraced (AJA [1914], 6ff.), or with the reorgan- ization of 373/372 b.c. I prefer to accept the earlier date and assume that in 376/375 B.C. some other board or officer, probably the rafxias rod Srinov, resumed the duty of paying for the inscriptions, and con- tinued to discharge it until the close of the century. The position was apparently a minor one; for the steward is never mentioned by name, and Aristotle does not include him in his description of the financial organization. Apparently he acted only as paymaster and controlled only the appropriations which were yearly made by the apodektai to the assembly to cover its expenses in connection with decrees. Out of this he could be authorized to pay traveling expenses of ambassadors, the cost of the crown, and the cost of the decree. In one case, also, he pays the living expenses of a certain Peisitheides, a drachma a day (No. 222). The cost of sacrifices, boundary stones, etc., in another case (No. 204) is paid out of the inscription fund. Apparently the hiroUKrai made an appropriation to the assembly 1 It should be observed that the restoration Brilpa.luv\ is purely conjectural. It is equally possible to read Qylpaiwv] or Qn[pairlwv\. Notes on Attic Inscriptions 421 yearly, especially for decrees («c tS>v els tA /caret. ^/r}iv); II, ca. 394-387 B.C., 6 rafiias to Sifreo; III, 387-ca. 384 B.C., ol airoSiKTai,; IV, ca. 384-377 B.C., ol rajiucu rfjs Bed; V, 377-303 B.C., drafiiasTOvdrifiov, VI, 303-302 B.C., 6 lirl ry SuHKr/aei ; VII, 302-301 B.C., 6 rajuiaj rod Srjfxov. I add some notes on those inscriptions which seem to call for comment in the new edition. No. 4. The restoration in line 4, [t6s rafiias rrjs deo], should read simply [t6s rafxlas], unless this document belongs to the years before the amalgamation of the two boards. In that case it should not be published in this volume. No. 21. Koehler's objections to ca. 389 b.c. as the date of this inscription cannot be lightly disregarded. There is no evidence that Chabrias took part in the expedition of Timotheos ca. 390 B.C., and the formulae used in this decree are found nowhere else before 375 b.c There is no dateable example of [e/c rav Kara ^<^ip.t\v hi ^vyPaWecrOai is not found be- fore 378 b.c. (cf. No. 44). This decree should be dated in that year or later; if ca. 378/377 b.c. we should restore in line 17 [rds] r[ap,ias rrjs deo] instead of [tov\ T[apiav\. No. 81. In line 11 read 5[6Vai t6s airobenTa]* instead of 5[ovt>ai rovers Tafilah; for ov is not elsewhere found in the inscription, and the double sigma which had to be used to preserve the uToixn^bv arrangement makes the restoration in the text seem rather improb- able. The inscription belongs ca. 387-385 b.c. No. 82. The formulae of this inscription place it without a doubt after 376 b.c. The formula ot &i> Xdxwo't irpoedpevew does not occur before 368/367 b.c. (No. 106). No. 84. This decree belongs ca. 378-377 B.C., but not before that date. The fund of the ten talents is first drawn upon in 378 b.c. No. 96. In line 20 read [iced rds] instead of [toj nal ava]- ypa^rjaoixivos. No. 106. If we except No. 21 (see note thereto), the first date- able example of the formula indicating a regular fund for defraying the cost of inscriptions is found in this document. This fund was probably established ca. 376 B.C. No. 120. The rct/wai rrjs PovKys defray the expense of record- ing the inventory at the chalkotheke. This does not mean that the ra/xtas rod Sri/xov is temporarily displaced, but, rather, that the stewards of the council, who also have a fund for decrees at their disposal, probably pay for the cost of inscribing inventories, building accounts, etc. There were at least two of these stewards (cf. No. 223), whereas there is only one for the popular assembly, and we might infer either that their duties were more onerous or that they served as a check on each other. Their duties are never defined. Notes on Attic Inscriptions 423 Possibly they looked after the financial expenses of the irpwaveh and the irpvTavelov. No. 140. The restoration of line 1 is probably as follows: €7rl rrjs Hai>8i[oi>i8os SeK&rys irp]- vraveias ktK. In the last prytany of the archonship of Thoudemos the presiding officers instruct their successors of the following year to bring in the proper legislation. 1 Apparently the annual elections were not held before the 15th of Skirophorion, otherwise they would have given the new archon's name. No. 141. From the formula in lines 15-18, this decree belongs to the years 378-377 b.c. There is no objection to this date in the literary evidence in Athenaeus xii. 531. No. 142. This is a decree which dates in 410/409 B.C., but appar- ently it was not recorded until some time later, or else it is a later copy. No. 149. In line 24 Woodward's suggestion of tSsv olSlkuv for rwv aiddwv is very plausible. No. 173. This decree belongs to the years 378-377 b.c. because of the fund of the ten talents. No. 197. If the restoration of lines 3-4 is correct, this inscription must be dated ca. 384-378 b.c. But since the stewards of Athena never have control of such a fund, it is much more likely that we should restore 8ov[ai rbv ran'iav to Srin]o and date the inscription between 372-360. The traces of letters at the beginning of line 4, 1 0, favor the restoration [Srip]o rather than [9e]o. Wilhelm's restoration of No. 211, while attractive, is far from convincing, not only because the cttolx^ov arrangement is so poorly kept, but also for the very sound objection which Wilamowitz has to urge. The frequent use of o for ov in the inscription would imply an earlier date, but there seems to be a revival of archaisms in the years 349-340 b.c. (cf. Nos. 215, 216, 217, 229). The restoration of the name of the secretary in Nos. 216, 217 seems good evidence for the date of these inscriptions, but I am 1 Cf. Ferguson, Athenian Secretaries, 20 ff. 424 Allan Chester Johnson inclined to believe that they should be dated in the archonship of Asteios instead of Archias. The formulae of the prescript and the frequent use of o for ov are characteristic of the earlier period. The chief objection to the late date is the fact that the decree of Andro- tion seems to be very recent, and we know that his activities were centered about the seventies. It seems to me that these inscriptions deal with the reforms instituted after the burning of the Opisthodo- mos in 377/376 B.C., and relate to the reorganization of the board of stewards of Athena in the archonship of Asteios (AJA [1914], 6ff.). The decree of Androtion might be remembered after thirty years, but it is very unlikely. The restoration of the secretary's name I should therefore discard. It should be noted that the father's name and deme of the mover of the decree are not given. This alone is strong proof that these documents are earlier than 354/353 B.C. No. 222. The formula of citizenship adds the words S>v oi vbpoi. \iyov after f/s av ^ovXrjrai. This implies a restriction in the free- dom of enrolment. The first dateable example of such a restriction occurs in 331/330 B.C. (No. 350). No such clause was added in any dateable inscription before 331 and therefore No. 222 should be placed in or after 331 B.C. The date of No. 245 is discussed on p. 419. No. 248. This inscription is earlier than 354 b.c. The deme name of the speaker is omitted. Probably it should be dated in 358/ 357 b.c. 1 If No. 249 is earlier than 343 b.c, we know from the secretary cycle that it is earlier than 350/349 B.C. Perhaps we may date this inscription in 352 B.C., restoring in line 1 [iw' 'ApiaroSrmov]. No argument for date can be based on the formula roiis wpoiSpovs tovs \axbvras kt\. This variant is rare, but it does not exclude an earlier date than 350 b.c. No. 265. This is probably earlier than 354 b.c. The last date- able example of 6 delva rjpx* is 349/348 B.C. (No. 206), and the deme name of the proposer of the decree is given after 354 B.C. No. 301. Koehler's restoration dovvcu is to be preferred to fiepiaai for line 5. The steward of the assembly does not make appropriations from the funds. The violation of the aroi-xn^v i Cf. ibid., 32 and 38. Notes on Attic Inscriptions 425 arrangement may be due to the fact that the mason did not want to break up the combination A A A by putting it on separate lines. In No. 336, line 33, we have, if the restoration is correct, the first appearance of the thesmothetae in citizenship decrees. In line 20 we should restore [?js av ffovKyrcu thai}. The plural 8>v is never found in this phrase, and the consistent use of the feminine singular seems to imply that only the phratry was unlimited in choice. This is, however, answered by No. 109, where we read elvai airrbv v\ijs yarivos av awoypa^>r]Tai, Kai drj/jov nal (frparpias, and hence we infer that the antecedent of the relative is all three nouns taken col- lectively. The limitation of the right to hold property does not occur in any dateable decree before 325/324 b.c. (No. 360). The restriction is not found in 330/329 B.C. (No. 351). We may assume therefore that the law which Wilhelm has shown to apply to border demes (No. 373) must have been enacted some time between 329 and 325 B.C., after the law restricting citizenship. In that case Nos. 342 and 343 must be placed after 329 instead of before 332 B.C. No. 425 is probably contemporary. In No. 350 we have the first dateable example of the restriction in citizenship Kara t6v vopov (331/330 B.C.). The date of this decree is questioned by Kirchner. The only other possible date is between 318-307 b.c. No. 366 is a summary of a decree. The deme and father's name of the proposer of the motion are not given. This omission is usual before 354/353 b.c, but not after this date, and therefore this decree probably belongs to the archonship of Kephisodoros in 366/365 b.c. No. 374. This decree should not be placed in the group dated 336-332 b.c. It belongs to the later period 318-307 B.C. Nos. 399, 400. I see no reason why these decrees should not be contemporary with No. 713, and dated ca. 288-280 b.c. This will obviate the difficulty about the i^tv kt\. Notes on Attic Inscriptions 427 This restoration has the merit of giving a uniform line which neither Larfeld nor Kirchner furnishes; for they are compelled to assume a lacuna of one space in line 4. In No. 456, lines 3-5, 1 suggest the following: 3 M-CUfxancrfa-] 4 [pi&vos h>ei Kal viae, Iftdotxei] kclI eiKooret rrjs [irpv-] 5 [raveias kt\. Kirchner's restoration involves a crowding of two letters into one space in line 4. The above assumes the regular line of 41 letters, and, other things being equal, the true v ov [kt\. No. 513. In line 3 we may restore also [daXkov artavu)]. No. 520. In line 8 the restoration Sovvai is better than ptpitrai, even though the aroixv^bv arrangement is broken by the use of the former. The steward of the assembly never makes appropriations. No. 525. In lines 7-8 the restoration rd "it.vbp.tvov av&kupa is suspicious, as it is never found with the rapias tov drjiwv. Its use here is awkward, and probably some other restoration must be sought. No. 538. In granting citizenship to this man whose father's name is Demetrios, there is no restriction in enrolment, but the thesmo- thetae scrutinize the gift. This would date this document ca. 334-331 b.c. No. 558. The fund for decrees is gone and the steward is author- ized to pay out of the common monies (e/c tcov koivcov xPVt i ^ T0}v )- I suspect that during the "four years' war" the inscription fund was suspended for a time. It was revived again in 302 B.C., only to dis- appear after 301 B.C. for nearly 70 years with one or two sporadic exceptions. No. 570. This decree may be dated by its formulae in 294-288 or 279-269 B.C. — possibly as late as 256-232 B.C. Notes on Attic Inscbiptions 429 No. 582. The removal of the restriction in regard to possession of property indicates that this document is earlier than 325 or later than 301 b.c. No. 648. In lines 4-5 we may restore wapa rr\v etaoSov. After 294/293 b.c. the secretary-cycle becomes an important factor in determining the dates of the decrees. Kirchner has followed Ferguson in the arrangement of the cycle without change, except in one or two minor points. But the mechanical application of the cycle to the period 294-262 b.c. without regard to the literary evi- dence is taking far too much for granted, especially in years abound- ing in political changes and violent party strife. I have set forth elsewhere my reasons for believing that the cycle was not followed from 294-292 b.c. and that a new cycle began in 291 b.c. with a secretary from the first tribe, Antigonis. In conformity with this the dates of all the archons between 294 and 292 b.c. must be changed. 1 No. 649. From the prescript of this decree it is apparent that some attempt was made in 294/293 b.c. to establish a new system of dating somewhat similar to that in 321-319 B.C. The attempt failed and the new formula never reappears. If the stoichedon arrangement is closely followed, the deme of the secretary must have at least ten letters. Of these the second must be omicron. There are only two demes known to us which can be restored in line 2, and of these one is so obscure that it may be dis- regarded. We may therefore restore lines 2-3 as follows : [5w]po[u tov] 'E[tt]it£Xov [K]o[vdv\fjdev eirl rijs 'Iiriroduv-] [rt]Sos 8eic[aT]r}s irpv{Ta\veYt.as ' Movvix<£>vos ktA.] No. 652 is probably contemporary with No. 654, as the formulae of citizenship are identical. No. 660. The first decree on this stone must antedate 325 B.c, as the possession of land is not limited. 1 To avoid repetition the reader is referred to the papers discussing the secretary- cycle in the third century which are published in Class. Phil., IX (1914), 248, and AJP (1913), 381-417, (1914), 79-80. In the group of archons between 262-229 B.C. some changes should be made in the published list. Hagnias should be dated in or ca. 255/254, Philostratos in 249/248, Xysiades in 248/247, Phanostratos in 247/246, and Antimachos in 246/245 (all this, however, subject to note on No. 775). I am indebted to Professor Ferguson for calling my attention to Kolbe's arguments for the date of Alexandres. The correct date for this archon is ca. 170 b.c. 430 Allan Chester Johnson Nos. 662, 663. These decrees should be dated [iwl AioicXeovs apxovTos] instead of in the archonship of Menekles (cf. AJA [1914], 165; Class. Phil, IX [1914], 277). In No. 663, line 37, Kirchner is incorrect in reading r[b di'dXcojua]. There is no trace of any letter after 5iouajop[iG)vos ivy naX viq., rpiaKOcrr-] [rj]i rrjs irpv[rai>eias ktX.] In either case the restoration does not fit in with the scheme pro- posed by Kirchner in his note to No. 662. The year of Menekles is probably intercalary. No. 670. Kirchner's date for the first decree on this stone is impossible because by his cycle we have four, possibly five, ordinary years in succession. In line 1 we may restore [bri KIjucowj] and date this document in 292/291 B.C. This gives us our proof for the disturbance in the secretary-cycle between 294-292 (Class. Phil., IX [1914], 248). In lines 16-17 we must read: [nai avS>(Tai] avrbv xpwQ [oT&frhvq Kara rbv vbpov ape- rrjs eveica. nal evvoi]as kt\. No. 672. In line 1 restore hri Saxrio-rpctToi; and date this inscrip- tion in 282/281 b.c. (Class. Phil, IX [1914], 258). The political situation at Athens in 279/278 B.C. does not permit us to date this document in that year. No. 674. The archon Glaukippos belongs to the year 263/262 B.C. Kirchner is incorrect in his note to this inscription because the board ol eirl ttj bioiKqaei is found during the Chremonidean war as well as between 288-280 b.c. (cf. Nos. 686, 689, 690). From this date for Glaukippos it follows also that the stemtna in PA, 4023, is correct and should not be changed. No. 675. I am not convinced that Wilhelm's combination of the four fragments in this inscription is correct unless they actually join, and this does not seem to be the case. Fragment a has 43 letters Notes on Attic Inscriptions 431 to a line, while the other fragments have only 42. In fragment c the phrase Ik t&v els r& Kara ^ij^icrjuara avaXwKOjtteVwj' t$ &finq is not found between 287 and 229 B.C., and the formula tov Tapiav tov (i.e., space for nine [?] letters) looks suspiciously like a fourth-century document. No. 682. The correct interpretation of lines 30-55 has been given by Tarn, AntigonusGonatas, 415 ff.; cf. Class. Phil., IX (1914), 248. All the evidence supports the theory that the Chremonidean war began in 267/266 B.C., and the archon Peithidemos should be placed in that year (Class. Phil., IX [1914], loc. cit). No. 688. Lines 3-4 may be restored as follows: ~2\iupo4>opiG)vos evinrri, [e/956]ju[et rrjs irpvravelas kt\.] No. 689. It is not correct to restore eiri 'AppevyiHov in line 1, because we know from the copy of the decree in Diogenes Laertios 1 that a commissioner was in charge of the administration in the archonship of Arrheneides, while in this decree the college is in power. We may be certain that the copy in Laertios is correct, because we know that Antigonos restored the single commissioner after the capture of the city. 2 I suggest that we restore in line 1 [«rl $i\ix- 7r]i8o[i/] and date the decree in 265/264 b.c. It may be noted in passing that Kirchner's note to the effect that irpoedpevuv els ttjv irp<&TT]v inKk-qaiav is not found after 262 b.c. is disproved by No. 808, line 12, if the latter is correctly dated. No. 692. In line 8 we may restore iiri Kip,a>vos or «ri Topyiov. From the tenor of the decree I think it belongs to the period 288- 280 b.c, during the struggle of the Athenians to maintain their independence. In that case we must restore eirl Topyiov. No. 696. The restoration Kara tov vbpov in line 2 is by no means 1 vii. 9. s This is clear from the fact that 6 (xl rjj SiotK^cet appears in all the decrees from 256/232 B.C., and cf. Apollodoros (Ferguson, Priests of Asklepios, 153): Kal ]et (1913), 391 and 417. 2 The date of the restoration of liberty to the Athenians by Antigonos is not pre- cisely determined. In the Armenian version of Eusebios the majority of the versions agree on 255/254 B.C., but Versio Zohrdbi reads 256/255. In Hieronymos this testi- mony is reversed, for the majority assign this event to 256/255 B.C. and only the Notes on Attic Inscriptions 433 No. 706. The value of the amount of property is never given before the last half of the century. I should date this decree ca. 232 B.C. Kirchner's note should be corrected; for the commissioner «rl 7T7 8ix>ucr)i.](To8wpov '1k- [apiebs iypapfidrevev' Bori8pofii.&vos] ij386p.ei. iir- [l 8ina, ?kt«i Kai SeKarei rrjs irpvrave]ias kt\. The archon's name must consist of twelve letters at least. The only name which can be restored is that of Pheidostratos. From the cycle we must date the document in 250/249 b.c, and since the archon Pheidostratos has been dated on prosopographical grounds ca. 250 b.c, there is little doubt that this is the correct restoration (cf. AJP [1913], 404). No. 744. In the Chremonidean war Sicyon was apparently allied with Athens for a very short time (Class. Phil., IX [1914], 274). It is possible that this document belongs to these years. In Division XV, Kirchner apparently does not place implicit confidence in the secretary-cycle, for the archons from 262-229 B.C. codex Bernensis places it in 255/254. Ferguson points out that the Sij/Mt made a dedication to Asklepios in 256/255 B.C. (Priests o/Asklep., 147 and n. 29), and it is certain that the city had its "liberty" in that year. Is it not possible that Antigonos gave them liberty in the preceding year, and that from No. 704 we may infer that the old machinery of government was in working order by the end of March in 256 B.C. ? 434 Allan Chester Johnson are dated approximately only. Since the tribe Ptolemais was created in 232 b.c. (AJP [1913], 381 ff.), the cycle as constructed by Fer- guson is one year out between 262-232 B.C., and the archons of this period should be changed accordingly. 1 Nos. 765, 766. Philoneos must be dated in 270/269 b.c. There is no evidence that the ephebic system existed from 262-240 B.C. I do not know why Kirchner says that No. 766 is not ffroixySbv. Only lines 13-14 are irregular, being crowded in the first part of the line. Nos. 768, 769. Kirchner is correct in assigning Antimachos to the period of the twelve tribes instead of ca. 206/205 b.c. Since the secretary is from Pandionis we may choose either 258/257 or 246/245 b.c. as the date of this archon. From an inscription found at Rhamnous and published in IIpa/criKa (1891), 16 we learn that Philostratos and Phanostratos preceded Antimachos, not neces- sarily immediately, and that an army officer was elected by popu- lar vote (x€ipoTOt>r]deis) in all three archonships. This would be impossible in 262-256 b.c; so we must conclude that Antimachos is to be dated in 246/245 b.c. and Philostratos and Phanostratos probably in 249/248 and 248/247 b.c. It is possible that Philos- tratos may be dated in 251/250 b.c. or even earlier. 2 The importance of this inscription from Rhamnous has not yet been pointed out. When Alexander Krateros revolted and set up an independent kingdom in Euboea and Corinth, he maintained himself by piracy, and from his island harbors the coast of Attica was particularly vulnerable. It was an easy matter for the pirates to run down from Eretria, plunder the country, and sail back to safety before the natives could muster their forces to repel the invaders. Hence Antigonos had to keep a standing army along the coast from Rhamnous to Sunion. I think that we may be justified in claiming that Euboea had not yet come into the power of Antig- onos in the archonship of Antimachos. If it had, there would be no necessity for troops at Rhamnous whatever, and it is quite certain that Antigonos would not have kept them there unless they were necessary; for he needed every available man elsewhere. We may now hazard a guess about the date of the recovery of the island by i Cf. note, p. 429. 2 Cf. No. 775. Notes on Attic Inscriptions 435 Antigonos. Apparently in 246/245 B.C. Eretria was still the home of pirates, and Rhamnous had to be defended. After the battles of Andros and Kos (probably 246 and 245 B.C., Tarn, Antigonus Gonatas, 378, and appendix 12) Antigonos was free to turn his attention to the reduction of the island. There is no direct evi- dence that he did so, but Chalkis was too important a post in the line of communication between Macedon and the south to be neglected. No. 775. From this decree we learn that the formula for sac- rifices in honor of Antigonos contains a maximum of about 42 letters. We must assume that no mention was made either of the queen or the offspring. This could only be explained if Phila were dead when this decree was passed, and that, 1 think, is the true solution. We may therefore restore lines 14-15 as follows: [nai rod (3a tw[i 'Ehevmviui] were accepted, we should date this document in the second year of an Olympiad; Van der Loeff suspects the phrase; but I do not believe that his objections are necessarily valid, since we know practically nothing about the length of the festival of the Greater Eleusinia in the third century. That the assembly was held in the Eleusinion only after the celebration of the mysteries is a theory which must be examined with care. How are we to explain IG, II, 431, -which seems to belong to the year following the Mys- teries ? No. 795. The archon Theophemos belongs to an ordinary year. There is no other clue to the date of this document, whose limits are 256-254, 249, 247, 244-243 b.c. Nos. 796, 797. These decrees belong to the same year, and the vacancies for an archon whose name has 9 or 10 letters, and whose secretary belongs to Antiochis, are exceedingly rare. In fact, the only available place for these documents is 213/212 b.c. The archon for that year is Herakleitos, and we can restore his name in both decrees. Since there were thirteen tribes at that time, we must restore No. 797 as follows: [iij>' 'HpafcXetrov S.pxovro]s «ri rfjs A?7/M7Tpict[5os Tptrijs] [■jrpvraveias j; ]os Avkov 'A\uireKrjd[ev irypanjia-] \revtv' Bovdpofiiuvos ev]ei /cal viai, TpiaKoltrios in line 2, since this is the only deme belonging to Oineis which has the requisite number of letters. 438 Allan Chester Johnson Since month and prytany coincide in 796 and 797, we must assume that 213/212 B.C. is an intercalary year. These inscriptions should be republished in the second fascicle in their proper place. No. 798. This decree probably belongs to the archonship of Charikles, 240/239 B.C., and in line 10 we should restore kir' 'Adyv- oS&pov (cf. No. 784). No. 800. Kirchner observes that the symproedri are not recorded in the official order, since Anaphlya (XII) comes before Sphettos (VII). Perhaps the true explanation is that this decree belongs in the period of the thirteen tribes, and that Anaphlya was a divided deme, part being given to Ptolemais, and part remaining in the old tribe. I am inclined to think that the constitution of Ptolemais between 232-200 B.C. was different from that which prevailed in the second century. When Antigonis and Demetrias were disbanded, and Attalis created, there must have been some changes made in Ptolemais as well. No. 802. h> rrj wp&rri enKkriaiq.. There seems little doubt that this formula is admissible after 262 b.c Nos. 804, 805. Both these decrees belong to the first part of the century rather than the middle, and probably earlier than 268 b.c. No. 808. In lines 27-28 we must restore top iirl t§ Sioudioei. The college existed at no time between 260-230 b.c. Wilhelm's assignment of this decree to ca. 234 b.c. may be seriously doubted. The restricted enrolment for citizenship and the omission of scrutiny cannot be proved for this period. I am inclined to think that Koehler may be right in dating this inscription ca. 303/302 B.C. I give a brief summary of the historical limits of certain formulae used in the Attic \[/riicriJ.aTa. In the prescript 6 delva rjpx*v is rare after 387/386 b.c and disappears in 349/348 b.c. (No. 206). The number of the prytany is seldom given in the first quarter of the century. The phrase eiri ttjs detvos tKTrjs irpmavevdaris occurs twice (Nos. 18, 34). ewl rrjs Setvos TrpvraveLas first occurs in 378 b.c. (No. 43). With the latter comes in 77 6 delva iypap-p-arevev. Ca. 354 B.C. there was a revival of the phrase 57 beiva brpvraveve, pos- sibly as a conscious archaism, but it disappeared finally in 340/339 b.c. (No. 233). The day of the prytany is first given in 368/367 b.c. (No. 105). It is never added when 57 detva iirpvrdveve is used. Notes on Attic Inscriptions 439 The name and day of the month first appear in 338/337 b.c. (No. 237). iKKkrjaia first appears in the prescript in 335/334 b.c. (No. 330). t&p irpoihpav iirexf/ri^ev 6 delva first appears in 378/377 B.C. (No. 44) and the last dateable example of 6 8eiva eVeoTdrei is 342/341 B.C. (No. 227). The avuirpoeSpoi. first appears in 318/317 B.C. (No. 448). The custom of giving the father's name and deme of the pro- poser of the motion appears first in 354/353 b.c. (No. 136) and seems to be invariable thereafter. The only exception if it is an exception appears to be No. 214 (347/346 B.C.). In matters of procedure the following points may be noted: yvu>fir)v 8e l-vufiaWeadai rijs PovXrjs kt~K. first appears in 378/377 B.C. (No. 44), rds Se Trpvraveis dovvai rr\v yj/r\4>° v "" € P' abrov rcj? 8rin

xoXei appears as early as 394/ 393 b.c. (No. 19) and displaces it entirely after 383/382 b.c. The phrase eV rS>v ets tA /card ^ij^ianara apakiiXorijuia is praised but once before 350 b.c. (No. 169). Piety becomes praise- worthy in 276 b.c. (No. 678). A selfish motive appears in this type of document in the latter half of the fourth century; for they add the phrase oVws &v eidaxn iravTv; kt\. (343/342 B.C. [No. 223], but Kirchner dates No. 183 before 353/352 b.c). It is interesting to observe that there is no example of otl avrip ayados 1