WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

The Robinson-Cologne Papyrus 
of Achilles T atius 

William H. Willis 

I N 1968 six larger fragments of a papyrus roll containing on 
the recto (the verso being uninscribed) parts of chapters 17 
to 21, 23 and 24 of Book III of Achilles Tatius' novel 

Leucippe and Cleitophon were identified among the Cologne 
papyri (P.Koln inv. 901) and published by Albert Henrichs. 1 

These fragments had been bought from a dealer in the mid 
1950's. From a different dealer in 1955 the noted archaeologist 
David M. Robinson acquired a group of some 200 quite small 
literary fragments in a dozen hands. Half of these were written 
in a distinct informal bookhand (P.Rob. inv. 35), which a search 
of the Thesaurus Linguae Graecae instantly revealed to derive 
from the same chapters of Achilles Tatius as the Cologne 
fragments. On seeing photographs of the Robinson fragments, 
Ludwig Koenen at once recognized the hand to be the same as 
that of the Cologne, of which no photographs had been 
published. Clearly the two sets came from the same roll. As is 
the case for most literary papyri, its provenience is unknown. 
But certain of the other fragments mixed in the same lot clearly 
came from the Thebaid, possibly Panopolis. 

Again by the aid of the machine readable TLG text, it was rela­
tively easy to align the Robinson fragments on the computer 
screen so as to determine their relation to one another, and thus 
to assemble the numerous small pieces into thirty substantial 
fragments.2 These larger groupings proved in turn to fill many 
of the gaps between the Cologne fragments, and so to provide 
an almost continuous running text of Book III chapters 17 

t .. Achilleus Tatios, aus Buch III (P.Colon. inv. 901)," ZP E 2 (1968) 211-26. 
2 The procedures are described in my report, .. Identifying and Editing a 

Papyrus of Achilles Tatius by Computer," in Aui del XVII Congresso 
Internazionale di Papirologia I (Naples 1984) 163-66. 

73 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

74 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

through 25 (I14).3 The Robinson fragments supplement all but 
ten of the 80 broken lines preserved on the Cologne fragments 
and contribute 103 additional lines for a combined total of 183. 
Few of the lines are complete, however, and of some only a 
few internal letters survive. Altogether the combined fragments 
preserve fifty per cent of the original text of these chapters, 
twenty per cent supplied by the Cologne fragments and thirty 
per cent by the Robinson. Through the good offices of 
Reinhold Merkelbach it has been arranged that the Cologne 
fragments are placed on indefinite loan to the Duke University 
Library, where they have been rejoined and photographed 
(PLATES 1-4). 

As joined together the combined Robinson and Cologne frag­
ments constitute a substantial part of the last five columns of 
Book III, and possibly also of the roll since the fifth column is 
distinctly narrower and shorter than its four predecessors. In 
depth of column, cols. i, iii and iv have 44 lines each, col. ii only 
42; but the surviving fragment of col. v preserves the middle of 
its first nine lines, and Book III could have been completed in 
five more lines for a total of fourteen, not counting a probable 
colophon. In length of line, cols. i, ii and iv average 48 letters, 
col. iii 51 letters, but col. v only 41. Possibly the scribe saw that 
he was about to run out of space at the end of his roll. It is also 
possible, however, that estimating that the remainder of Book 
III would not require a column of full width, he may have 
narrowed col. v to conserve space and may have continued 
thereafter with Book IV. 

Since the lower margin of col. i is preserved (1.6 cm.) and the 
full upper margins of cols. ii and iv (1.7 cm.; the frayed upper 
margin of col. iii measures 1.0 cm.), the height of the roll can be 
calculated as 29.1 cm.-tall but within normal limits. The 
average written width of cols. i-iv is 10.5 cm., and intercolum­
nar margins vary from 1.5 to 2.1 cm. The average width of these 
columns including an adjoining intercolumnar margin is thus 12 

3 The sigla nt, n 2, and n 3 are those assigned by E. Vilborg in his critical 
edition, Achilles T atius, Leucippe and Clitophon (Stockholm 1955). The sigla 
n4_n7 were designated by J. N. O'Sullivan in his A Lexicon to Achilles Tatius 
(= Untersuchungen zur antiken Literatur und Geschichte 18 [Berlin/New 
York 1980]). Though his n4 signifies only the Cologne fragments, I extend the 
siglum to include the Robinson fragments as well. When he compiled his 
lexicon he could not yet know that nl and his n6 derived from the same roll 
(see infra). 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 75 

cm. At an average 433 words per column one may estimate that 
the entire vulgate text of Book III would comprise thirteen 
such columns and would therefore fill a roll 1.55 meters in 
length,4 only a fraction of the usual roll-length ranging from 7 to 
10 meters (23 to 33 feet).5 At the same ratio if the roll contained 
the hrst three books together it would have measured about 5 
m. (16.7 feet), still short of the length of an average roll. The 
whole of the novel, on the other hand, would have occupied 
about 100 columns or a roll of 12 m. (40 feet), rather beyond 
ordinary size; or two such rolls of, respectively, 6.5 m. (21 feet) 
and 5.75 m. (19 feet), each somewhat shorter than average. 

The joined Robinson and Cologne papyrus constitutes the 
most extensive ancient text of Achilles Tatius yet found, and 
one of the two earliest copies. To its rapid informal quasi­
literary hand sloping slightly to the right and admitting frequent 
ligatures I have found no close parallels among either literary 
papyri or documents. The more formal features somewhat 
resemble the hand of the final columns of the Vatican 
Favorinus (P.Marm. 1 = M. Norsa, Scrittura letteraria graeea pI. 
13, or C. H. Roberts, Greek Literary Hands pI. 18b-c), dated to 
the late II or early III century, presumably before A.D. 215. The 
cursive letter-forms and ligatures are near BG U I 356 (= W. 
Schubart, Griechisehe Palaeographie Abb. 48), dated AD. 213, 
and in some cases near P. Wise. 15 pI. II, dated A.D. 186. I should 
therefore assign the papyrus to the early III century.6 Some­
what earlier still, perhaps, is P.Mil. Vogl. III 1244 (n 3), a small 
fragment of a papyrus codex having parts of lines from Book VI 
chapters 14-17, which A. Vogliano (with Schubart's approval) 
assigned to the II century;7 but the compression of its hand 
sloping to the right and its codex form suggest a later date just 
before or after A.D. 200. In any case, to the Milan codex leaf, now 

4 The calculations in this paragraph are based on a computer word-count of 
the eight books of Achilles Tatius in Vilborg's edition recorded by TLG: I, 
5175 words; II, 7287; III, 5577; IV, 4670; V, 6503; VI, 4160: VII, 3940; VIII, 
6128; total, 43440. 

S W. Schubart, Das Buch bei den Griechen und RomemJ (Heidelberg 1961) 
52f. 

6 Henrichs, citing the Favorinus but also adducing Schubart Abb. 52 (P.Ryl. 
II 117, A.D. 269), which seems to me not so close as Abb. 48, and Abb. 90 (Iliad 
14, P.Berl. inv. 11910), a literary hand of a rather different style, suggested a 
date of the end of the III century. 

7 A. Vogliano, "Un papiro di Achille Tazio,» SIFC N.s. 15 (1938) 121-30. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

76 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

reinforced by the Robinson-Cologne roll, belongs the distinc­
tion of compelling a reassignment of Achilles Tatius from the V 
or VI century to a date no later than the middle of the II. 

Four other papyri of Achilles Tatius have come to light, all 
later. The first to be recognized was P.Oxy. X 1250 (IF) from a 
papyrus roll of the III (Schubart) or IV (Grenfell and Hunt) 
century, containing part of Book II chapters 2-3 and 7-9, to 
which now may be added P.Oxy. LVI 3837 (Il6) from the same 
roll, a fragment of Book VIII chapters 6-7. P.Schub. 30 (Il2), ap­
parently lost, was a fragment of a papyrus codex dated to the III 
century, bearing lines from Book II chapters 2 and 14, now 
known only from Schubart's collation. P.Oxy. VII 1014 (fF) 
from a IV-century roll, published originally as an unattributed 
historical fragment, was subsequently recognized by M. Grone­
wald to be a fragment of Book IV chapter 14.8 Now in the most 
recent volume of the Oxyrhynchus Papyri P. J. Parsons has 
published parts of three columns from a second roll (P.Oxy. 
LVI 3836, IlS) of Book III chapters 21-23 dated to the IV cen­
tury, which though fragmentary overlaps an appreciable stretch 
of the text covered by the Robinson-Cologne paryrus. The 
papyri of Achilles Tatius now outnumber those 0 any other 
work of this genre, three from rolls, three from codices, and all 
from the late second through the fourth centuries. 

Our scribe, despite the informality of his hand, wrote with 
commendable care. In only three instances was he guilty of 
uncorrected orthographical errors: XPTl(J { t} Jlov at ii 5 (marked 
by a corrector with a slash in left margin), JlTlxav(TlJl)(l'to<; at iii 
30 and PO()o(t}<; at iv 37 (in lines of which the left margin is not 
preserved); and one probable error (1tap' for yap at i 20) which 
may confirm a lucky emendation. For dative singular endings 
of first- and second-declension forms (though never internally) 
he writes iota-adscript correctly in 25 cases, omitting it in only 
four. He regularly writes a trema over initial iota and upsilon 
and the apostrophe as divider to separate double or cognate 
consonants (i 26,30; ii 14; iii 1,16,41; iv 12,30,33,43). Only once (ii 
30) does he write -(J(J- (with the codices) for Attic - 't''t-. In the 
only two cases of short lines where the right margin is 
preserved (SO of 183 lines) he writes a space filler (ii 2,4). 

8 -Ein verkannter Papyrus des Achilleus Tatios," ZPE 22 (1976) 14-17. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 77 

Insignificant orthographic alternatives and phonetically equiva­
lent spellings are fairly frequent: there are eight instances of 
iotacism (i 19; ii 41; iii 6,9,19; iv 15,22,44), two of -E - for -Ut- (i 9; 
iv 6 with TI 5). Seven times nu-movable is followed by a 
consonant (i 15,19; ii 1,32; iii 9,26; iv 4 with TIS). Twice adjacent 
consonants are not assimilated (i 27, ii 2). Once the scribe 

writes seriptio plena where the codices show crasis (i 32), once 
he elides where the codices do not (i 44). All such deviations are 
common in ~he papyri and of course do not qualify as either 
errors or vanants. 

The text has received the attention of a corrector, but in what 
cases he is the original scribe or a diorthotes is difficult to deter­
mine. In the five instances where interlinear variants are 
written, three of the scribe's unique original readings point to 
the probably correct text (ii 4,11,27); in the other two (ii 12, iv 
42) the corrector offers a better choice as reflected in the 
mediaeval codices (see commentary ad loee.). Beside three of 
these lines there is no marginal slash; at the remaining two the 
left margin is lost. In two cases the scribe, apparently, entered a 
variant reading above the line without further marks (ii 4,11). In 
two other cases (ii 12,27) the corrector encloses the variant 
between dots, in the prior instance offering the reading of all the 
codices, in the latter recording a false variant found in no other 
source. In the fifth case (iv 42) the scribe's unique and 
conceivably right syllable has been stricken through and sur­
mounted by the vulgate reading but in the wrong case. In 
addition to the instance of error mentioned above (at ii 5), a 
corrector has marked with a marginal slash two lines containing 
a gap in the scribe's text: in the first case (ii 29) a space of about 9 
letters is crossed by a horizontal stroke extending a preceding 
final sigma, though here the codices show no loss; the second 
(iii 5), where a gap of about 12 letters is left unfilled, is discussed 
below and in the commentary. 

As usual, the text of TI4 does not align itself with anyone or 
any family of the mediaeval codices,9 for the earliest of these 
(W, Vat. Gr. 1349) is dated XII-century, and their hypothesized 
archetypes would have been copied some six centuries later 
than the papyrus. Generally the text of one or more of the 

9 For a list of these and their assigned sigla (drawn from the edition of Ebbe 
Vilborg [Stockholm 1955] xv-xxxii) see infra at the beginning of the 
commentary. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

78 THE ROBINSON-COLOGNE PAPYRUS OF TAT IUS 

codices agrees with that of the papyrus but none predominantly 
SO.10 In 63 cases not counting transpositions or obvious errors, 
however, no codex has preserved the reading of II 4, and in at 
least 41 of these the papyrus may well have the sounder text. In 
eight cases II4 confirms an editor's conjecture (Commelini in 
ed.pr. at iii 32; Castiglione at i 20; Cobet at ii 4, iv 3 and iv 22; 
Wifstrand at ii 26; Hercher at iii 28; Jacobs at iii 42). Several 
times n4 has an Attic form (ii 26, iii 1,6,7,8,16) or construction (ii 
4, iv 1,39) where the vulgate substitutes a late or trivialized 
equivalent; at i 23, iii 19,27, iv 7,18,20 the lapyrus has a rarer 
classical word, the codices a commoner an weaker. There are 
fourteen instances of transposed word order, for at least some 
of which the papyrus seems more classical. 

In nine lines n 4 shows a longer lacuna than the vulgate text can 
fill: either a word or two had dropped out or, less likely, the 
scribe of n 4 had skipped a short space as at ii 29. The lines in 
question are i 12,24,25,37,44, iii 15 and iv 28-30. Conversely, in 
five lines (i 10,23,28 twice, iii 7 and iv 33) the lacunae appear too 
short to hold the received text. A larger omission occurs at ii 
13-14, where the codices seem to have padded the text 
unnecessarily. From the supplements in the following transcrip­
tion I have omitted as superfluous short words not required by 
the context. Otherwise, where context and lacuna space permit, 
the supplements generally retain the vulgate text of the Vilborg 
edition. Noteworthy exceptions occur at iii 24-25 (22.1) where 
editors print both the competing phrases lCaA.bc; (, lCtVOt>VOC; (aF 
and probably n 4) and 'YAt>lCUC; () 6ava'toc; (P), though no 
manuscript has more than one; and at iii 14 (21.5), where the 
agreement of II 4 and lIS in lCa't£t<Hv confirms Henrichs' 
restoration of ava.~a.{v£tV against the codices. 

But the most striking contributions of the papyrus are those 
where it has a quite different text extending through two or 
three lines. One of these (iii 4-6, at 21.3), mentioned above, is a 
passage clearly corrupt in the codices, for which editors have 

to Where family CI stands alone, there are three agreements with n 4; family 
p alone, seven agreements; F (family cp) alone, none. Where families CI and p 
agree against F, there are five agreements with n4; but where family a and F 
agree against P. there are eleven; family P . F, and n4 agree only once against 
CI. Wand F together against both a and P share three readings with n4. If 
later alignments are ignored, the family CI codices preserve 20 readings of n., 
family P 14, F 15. Once Commelini's editio princeps uniquely preserves a n4 

variant. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 79 

suggested a variety of unsatisfactory solutions. Unfortunately 
our scribe has interrupted his text with a space of some twelve 
letters, presumably because he could not read or did not trust 
his antegraph. In any case the incomplete text of TI4 is nearer the 
truth than are the codices, and from it a persuasive emendation 
may emerge. The other notable cases are at ii 10-11 (20.1) where 
the papyrus' b£01tOtVa and distribution of verbs and participles 
give the better text; at iii 27-28 (22.2) where the codices have 
abbreviated and trivialized the text; at iii 32-34 (22.3) where TI4 
agrees with neither the codices nor TI5; at iv 8-9 (23.3) where 
the first of three parallel cola was lost in the codices' archetype 
but may be confidently restored from TI4; and at iv 18-21 (24.1) 
where the text of TI4 is more graphic and doubtless right. In the 
second and third of these passages lacunae impede a restoration, 
but in all of them the papyrus promises a sounder, more precise 
and vivid text than the vulgate. 

For the first time we have a rare opportunity to collate two 
ancient copies of a significant amount of text. While TI4 and TI5 
agree with each other and with all the mediaeval manuscripts 
nineteen times, once (iii 20) they agree with a and p against F, 
twice they share insignificant orthographical features (nu­
movable before consonant, -E- for -a t -) against all codices, and 
only once do they agree in a right reading (Ka'tEtcrtV at iii 14) 
against all codices. TI4 opposes TI5 nine times: once TI4 agrees 
with the codices against TI5 (iii 22); once (at iv 6) TI4 agrees with ~ 
(though transposed) where TI5 agrees with a and F; once TI4 has 
Kat (iv 5) where TI5 with a and F have bE and Pis asyndetic; and 
once TI4 reads Oil (iv 7), TI5 reads J.1Ev O;)V, and the codices simply 
o;)v. Four times TI4 stands alone against TI 5 and the codices (iii 
15,20, iv 1, and iv 3 confirming a Cobet conjecture), in all of 
which TI4 is probably right. In only one passage (22.3) does each 
of the three sources have a markedly different text from one 
another. The codices, in some disarray, have lost a few words 
here, and TI4 (at iii 32-34) and TI5 (at col. ii lines 12-16) are both 
too lacunose to provide a solution or help for each other. In any 
case it is clear that the two papyri, while occasionally differing 
from each other, are closer to the author's original than the 
archetype of the codices. Of the two, IT4 more often has the 
sounder text. Altogether, where variants occur its style is more 
precise, clearer and more classical than that of the vulgate. 

For details of these variants the reader is referred not only to 
the line-by-line commentary following the text but also to 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

80 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

Henrichs' excellent notes on that portion of TI4 available to him 
in 1968 (n.l supra) and Parson's on TIS in The Oxyrhynchus 
Papyri LVI (1989) pp.64-66. 

TEXT 

P.Rob. inv. 35 + P.Koln inv. 901 [= TI4] 
(The text of P.Koln is underlined, that of P.Rob. is not.) 

[EYro Oe avopae; iorov] 3.17.2 

col. i 
1 [ElC 1tapaA.&you Crov'tae; lCal cp1Aoue; ou't£ 1t£Pl£]1t'tu~[aJ.L]11v [ou't£] 

[E~£1tAkrlv ucp' ;'Oovf\e;. 'toomhov;, Aim]11 tile; cruJ.Lct>Opfr[e; J.L£] 
[E~£lCcOcp11cr£. A.aJ.L~avov'tal OTt J.LOU 't1le;] O£~1.fre;· Kal acpalp£l[ cr9](Xl .3 

[E1tEXelPOuv 'to ~lq>Oe;. tyro OE. np~ 9E&V.] J.LTt J.LOl <p90VTtcr11'tE 9ava-
S ['tou lCaAou. ECP11V, J.LfrAAov Oe cpapJ.LalCOU 'trov] lCalCrov' ouo[e y]ap 

~llV En 
[ouvaJ.La1.. lCal Eav vuv J.LE ~lacr11cr9E, tile; A]EulCl1t1t11e; [ou]'tox; 

aVUP11-
[J.Liv11e;. 'tolho J.LEv yap acpalPTtcrEcreE J.L]ou 'to ~lcpo[e;. 't]o Oe tile; EJ.Llle; .4 
[AU1t11e; ~lcpoe; EVOOV lCa'ta1tE1t11YE lCal 't]EJ.LVEl lCa't' oA1Yov TItV 'lfU­
[xitv. a9ava'to)l crcpayill a1tOSvucrlCElV] J.LE ~ouA.ecr9a1. aei; ').1Y£l 

10 [M£VEA.aoe;· 'AAA' d Ola 'toU'to 9EA.ele; a1t]09av£lv. ropa crOl 'to ~lcpoe; 
[EmcrxElv. AEulCl1t1t11 OE crOl ava~1.(ocr£]'ta1.. ~AE'lfac; o[~v] 1tp[o]e; 

au't[ov]. . . 5 
[ "E "I ~" , , ] "I' ~ '? 'tl J.Lou lCa'tayEA~e;, ECP11V, Em 't 11AllCOU'too1. lCalCOOl; EU 
[ M '''1_-- -" A ' ]. ~, , , , YE, EVEAW:::, ':'EVlOU IlEllvllcra1. ulOe;. 0 oE lCpouqa~ 'tnv qopov, 
['E1td 't01vuv a1tlcr'tEl KA.el'tO<proV.] ECP[11,] LLOl. AEUlC11t1tn. LLap-tu-

15 [Pllcrov d ~iie;. (lila Oe d]1tE lCal Ole; J.LOU lCal 'tplC; E1ta'ta~EV mv .6 
[cropov. Kal Ka:troSEv cpoo]vlle; a.1Couoo Kal 1tavu A.e1t'tfic;. 'tPOl+<>C; 
[ '? , 9' " , ']' , M '''1 []' '(.l"l ' OUV £U ue; 1.crX£l J.L£ lCal 1tp oc; 'tOV EVEAao v a1tE",AE1tOV. Ilayov 
[£ival OOlCroV. (, Oe i1vO]~YEV (lJ.La 'tTtV cropov lCal n A£ulC1mtn 

lCa't 00- .7 
[9EV aVE~al]vEV, cp[O]~[Ep]OV 9EaJ.La. ir> 9£01. lC[al] CPp£llCOO-

. OEq'ta['tov.] 
20 [avEcplC'tO J.LEv] 1tap' autile; ;, yacrTItp (l1tacr~ [lCal ~v tv ]'tEPOOv 

lCEVU. [Em]-
[1t£croucra oe] 1t£p1.1t').1lCE'ta1 J.LOl lCal cruVE[cpUJ.LEV] lCal aj.lcpoo 

lC[a'tE7tE]-
[crOJ.LEV. MO]A1.e; o~v ava~C01tUPTtcrae; A~O? [1tp]oe; 'tov MEVEA.[aoV· 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

P. Rob. inv. 35 + 

P. Kaln inv. 901 col. i 

tt.i!:~ 
'f4fi..n.~ 

WILLIS PLATE 1 

P.ROB. INV. 35 + P.KOLN INV. 901, COLUMNS i AND ii 

(reduced to 55%) 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

PLATE 2 WILLIS 

'~~ ...•...••. 

~@ 
~ 

~ 

i·
·
C 

...... -•..... 
~-- ' 

cO ... 

:I~···~·.-· . ... 

i
~2 

M .~ ••..•. 
<= 

"'I ...:< ......•.. 
o _ 

~ ~, 

P. Rob. inv. 35 + 

P. Kaln inv. 901 
col. iii 

P.ROB. INY. 35 + P.K6LN INY. 901, COLUMN iii 
(reduced to 64%) 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

col. iv 

WILLIS PLATE 3 

P. Kob. my. j:, + 

P. Kaln inv. 901 

P.ROB. INV. 35 + P.KOLN INV. 901, COLUMN iv 
(reduced to 64%) 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

PLATE 4 WILLIS 

col. V P. Rob. inv. 35e 

c •• 

P.RoB. INV. 35 + P.KOLN INV. 901, COLUMN V 

(actual size) 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 81 

[Ol)1C] 18.1 
[EpEte; ~01. 't]i 'ta,ha; OUXt AEU[lCi]1t1tT\V ~Ai1tCO; [OU lCp]a'tro lCat 

. . . ... .... ... ~[lCO]Uoo 

[AaAoUOT\e;; a ot>V x9Ee; E9Eami~T\v, 'ti]va ~v; i1 yu[p ] ElCEtVa 
.; .. 
eo't1.v 

[" " 25 1'\ 'tau'ta 
, 'I 'I, '<:" "'I ''I 1\" , Y"] , " , 

. aAA lOOU Kal <plA1U1a aA1'\tJlVOV Kal 1,,00 v EKElVO 'tE 

Aru -
[lCi1t1tT\e; yAUlCU. 'AAAU vuv ~tv, 0 MEVEAaOe; E<PT\, lCat] 'tU 

01tAay'xva 
[a1tOA"\jfe'ta1. lCat 'tU O'tEpva o]uv<p[uoe'ta1., lCat au-citv ih]pCO'tov . . . " . 

o",e1.. 

2 

[E1t1.lCaAU",a1. 'to 1tpOOOO1tov' lCaAro y]up 't'h[v 'ElCa'tT\V E1tL 'to] ~~rov. 
f:yCo .3 

[bE 1t1.o'tEuoae; EvelCaAU",a~Tl]v. 0 bE u[pXE'ta1. 'tepa'tEu ] E09a1. lCat 
AO-

30 [yov 't1.VU lCa'taAEYE1.V. a]l7~ ~~r~v 1tEp[1.a1.pet 'tU E1tt 't]ill. yao'tpt 
~ay'-

[yaveu~a'ta 'tile; AEU]lCi1t1tU~ lC[at] E1t1.lCa['tEO'tT\OeV eie; 't]o apxatov. 
[AEyEl. bE ~01., 'A1tOlC]aAU'I'a1.. lCat E[yCo ~OA1.e; lCat CPO~OU~E]VOe; 

. (aATl9roe; .4 

[yap q>~T\V 't~v 'E]lC(i'tnv 1tapetv~[1.), o~roc; b' o1)v a1tEo'tTl]oa 'trov 
o<p9aA-

[~rov 'tuc; xetpac; lC]at OAOlCAnpOV AeulCi[1t1tTlv opro. E't1. ~a]AAO[V 
~]v 5 

35 [ElC1tAayetc; EbEOhmv MevEAaou Aiy[oov, ';'0 <piA'ta'tE MeVEAae, ei 
b1.alCo] -

[voe; 't1.C; d gerov, b]Eollai oou 1tOt yij~ e[i~1. lCat 'ti 1tO'tE 'tau'ta] 
[op&; Kat 1) AeUKt]1t1tU, nauqal., E<PU, MevEA[ae,] ~h6JlevoC; ]. 
[ " "I' O;::'~] '''I '" "I' ';'. au'tov. lI.E"(e uE 1tOO ~ 'tOU~ Auq'ta~ U1ta'tnqa~. t\.E)'El. ouv 0 

MEvE[Aa]- 19.1 

[oe;. otbac; roc; A i"u1t ]'t1.0~ eill1. 'to yEvO~' w8avoo yap 001. 'tou'to ei1trov 
40 [E1tt 'tile; vEroe;. ~v 01)v] 1101. 'ta 1tAEtq'ta 'trov lC'tUlla'toov 1tEpt 'tau'tT\v 

[-citv lCro~T\V lCat Ol upX]OV'tE~ au'tn~ yvroP1.1l0[1.], rndt 01)V 'till. 
vaua]- 2 

[yia1. 1tEP1.E1tEOo~ev, E'h]a IlE 1tPOOt[P1.\jf]E 'to KUbLa 'tOt[e; 'tile; Ai"u]­
[1t'tou 1tapaAio1.e;, Aa~~]avo~a1. ~E't[U ~]a'tUpou 1tp[oe; 'trov 'tau'tT\v] 
[ 1tap acpu Aauo ]v'tOOV AnO ['trov .] cO<; b' uY0l.la1. 1t[poe; 'tOY 

. . AnO'tap]-



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

82 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

col. ii 
1 [xo]v. 'taxu l1E'trov Allq'trov nVE~ yvropiGav['tEe; AUOUG]iv 110u 

'0;: '8 ~ "'I ' ~ [ , ] ~ , " '\. 'ta uEql1aappElV 'tE EKE",EUOV Kal qUV7tOVElV au't Ol~. ~ av Ol- r 
.... 'J:. .... ~" "" ~ , "'''' , " ~,. KElOV. E~al'tOUl1<1l un Kal 'tOY ~a'tUpov ~ E110V OlKE'tnV. Ol uE. 

'AAA' 0- .3 
" '0;: '~ \_ /' ~ ~ 'I' ~, 4 7tro<:;, Eq>aqav. E7tloEl~ElV'Ul Ul1lV 7tpCO'tOV 'to",l1UPOV 'tl qau'tou. Kav >-

/ ' , " , o~ [] '8~ , 'tOU'trol xpuql110V lqXOUQ"l KOPUV Ka'tauuq al KalKa apal'tO 
'I' ,~'" , , 8[] [] 8 ' , 0;:' I\.Uo'tU plOY Kal 'tou l1EV U7ta't~ a7tO)'Euqacr a 1.. 't U ElqU~. 'to uE 

Mli- . 
7tOV Grol1a qOprol 7tapaoov'ta~ avaxropElV, ~ a.v 'to 'trov Evav'ti­
roy G'tpa't07tEOOV i>7tEP6aAA.ol m~ 8uqia~ 'tOY 't07tov. ').1:YE Ott 'ta 

E7ti-
'I ~..!. "" ~ 8 'l J..., , , ~..!. 'I !. l\.ol7ta. ~'tUpE' 0 qoe; yap EV'tEU EV lWfo~. Kal 0 ~'tupoe; I\,t:;YEl' 

"Afla 20.1 
0;:' 'R o;:'y , " , 0;: "'1__ " 0;:' , 1 ° uE EpaUl,.ol1EV E7tl 'to G'tpa't07tEuOV EK/l,UlOV. ro uEq7tOlVa, Kal 

ci>[O]U-
pOflTlV, 'ta 7tEpt ~ AEUKi7t7tTle; 7tU8011EV~ Kat OEOl1EVOe; MEVEA[ a]­
ou 7tav'tt 'tP07trol GOOal 'tTt[V KOp]UV, oai[flro]v OE 'tle; 'hflro\'l'/v 

.. aya8[oe;] . .2 

G'UV11P'YTlOEV. E'tUXOflEV till [7tpO'tEpal 'tile; 8]UGiac; 'hflEpal Ka8[ EeO]­
flEVOl 'till 8aAa't''tTll 7tEpt 'tOur 'troY GK07tOl>V]'tEe;. 'trov OE ATIG'trov 

15 'tlVEe; val>V ioOV'tEe; ayv[oial 7tAavTl8ElGav,] roPflTlGav E7t' 'au't11v. 
01. OE E7tt 'tile; VEcOc; GUVEv't[Ee; ofi 'tU'YX]avo[UGlV, E7tEX]dpouv 

[EAau]- " . .3 

VElV de; 'tou7tioro' roc; OE <p8a[vou]GlV au'tou[e; 01. ATIo'tat 
. " Ka'taAa~'ov]-

'tEe;, 7tpOe; afluvav 'tpE7tOV'tal. Kat yap 'tle; Ev [au'tOle; 'trov 'tal .4 

fOfl11POU 'trol o['t]oflan OEllCYuv'trov Ev 'tOLe; [8Ea'tpOle; roy E7tt] 
20 tile; VEcOc; 'tTtV 'Ofl~Plx:TtV OKEUTtV 07tAlGafl[EVOe; 'tE au'toe; Kat] 

['t]ouS a[flcp' auj'tov oihro GKEUaGae; E7tEXE[ipouv flaXEG8al. 7tpOc;] 5 
[fleV' o?>v 'toue; 1tpmou]e; E1tEA8ov'tae; Kat 'fl[aAa EPProflEVro<:;] 
av'tE'ta~a[ v't]o' 1tAElOvrov OE E7tl1tAEUGaV ['trov GKacprov ATIG'tPl]­
[Kro]V Ka~aOUOUGl 'tTtV val>V Kat 'toue; av [opae; E1C7tEGOV'taS] 

25 [a]vUpouv. AaV8aVEl Ott Kio'tTl ne; ElC'tpa7t[ElGa 'trol vauaYlrol] .6 

Kat Ka8' 'hflue; 'trol prol \·7tOPPro·/KOfllo8Eloa Ka[ ilv 0 MEVE-
AaOe; aval]-

[pEl]'tal, Kat avaxroP11Gae; 7tOl 1tapov'toe; <i[fla KaflOl> (7tpOGEOoKa 
." ... . . yap 'tt] 

[G1t ]ouoalov EVOOV dval) avoiy[ El -rTtV KiG'tTlV, Kat oproflEV 
XAafl uOa ] 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 83 

/K[al] siq>os ,TI]V IlEv K[rom,V EXOV 'to IlTlKO<; ocrov 1taAal.cr'toov] 
30 'tE[cr]craprov, 'tOY BE criBllPOV E1t~ ['tft1Kc.OOt1l1 ~paX{)'ta'tov, 

BaK'tuArov] 
OOOV 0.0 1tA£iro 'tP1OOV. ~ B[e aVEAOIlEVo<; 'to ~i<pO<; 0 MEVEAaoc; 

. n .. a] - .7 

9ev fLE"tao"tp6va[s Ka"tCx. "to "to'\) Ol.OTtpou fLEPOS, "to fLl.1CPOV EKE'i]-
vo ~t<pOs oxmEp [a1to Xll] pa[~ou 'tTl<; Krom,<; ocrov Ka'ta'tpEXEl 'tocrou]­
'tOY EtxEV ;, Kc.OOtll 't[o ~]E.yi8oc;· [~ BE aVEcr'tpE'I'EV ElS 'tou~]1taAtv, 

35 at>8~s 0 oiBllPOS d[oro K]a'tEBuE'tO. ['tOU'tro1 B' apa ~ ElKOS 0 
KaK]oBat~rov 

EKEtVO<; EV 'tOts 9[EU]'tpOl<; EXPTl'tO 1t[po<; 'taS K1~OTtAoU<;] crcpayuS. 
AEYro [ot>v] 1tpOs ['tOY] MEVEAaov, 8[EOs ;,~'iv, iiv 8EATIS vuv 

xlPT1crt<x; 21.1 
[YEVEo8a1, ouvayrovtouat. Buv ]l1o[o~E8a' yap Kat] 'tTtV KOP[l1V 

. . orocra1] . 

[Kat 'tou~ Ancr'ta<; Aa8E'iv. aKoucrov Be 1tOtrot 'tpoo ] rot' BEp~[ a 
, 1tpopa.'tou] . 2 

40 [Aaf3oV'tE<; ~ on paOtvOna'tov cruppa'l'o~EV El<; crxft~a ~]aAav'ttou, 
[~E]-

['tpov ocrov yacr'tpo<; av8pC01ttVll<;, E'ha E~1tATtcrav'tE<; 8]l1Ptrov 
cr1t[AUY]-

[xvrov Kat a'{~a'to<; 'tTtv 1tAacr'tTtV 'tau'tllv yacr'tEpa] pa'l'O~EV, [~ 
~Ttl . 

col. iii 
1 pa~?[v] 'ta cr1tAay'xva 0l.a1tt1t't01, Kat EvoKEuucraV'tE[<; 'tftl. KOPlll.] 

'tou't[OV 't]OV 'tpOOOV Kat o'toAnv E~roOEV 1tEpl.6aAo[v'tE<; Ill'tpal.<;] 
'tE x:'a[t C]~jla(nv EVOEOEjlEVnV -mv qKEunv 'tauTri[v E1tl.KpU'l'0IlEV.] 

4 n[av't]'% OE Kat 0 1pnOjl(x; TlJl'iV El6 'to Aa8E'iv 1PJlq1jlO[<;' 0 ~ev 
yap] 3 

/ !!.o'tTtV ~ vacat L1EVn q'toAn ~ha 'tau'tnv ava'tLLn[8ftva1 ~EOllV] 
'l. ' .. - §::\ '\ '\ ):.' .." ....." '\ " [ 
~qEl. opq.~ uE 'tot>'tt 'to ~1<p06 c.o<; E1El LLn1avn6' av yap Ep1q TI n~ 

E1tt nvo<;] .4 
, , \. '\ , " t 'l.' ['" 

Q:ro~a't06 <PEt>yE1 1tP06 'tnv KC01tnV coom:p E16 KOII.EOV· Ka 1 01 ~EV 
. BOKOU01 f3£]-

R", ' 8 ' , I:: ,~, • I:: ' " [ , .j.l!!1tUO at 'tOY qtunpOv Ka'ta 'tot> qro~a't06, 0 uE El6 'tOY 1npaL1 ov 
'tTl<;] 

, "8 ' I:: ' 'l ' , , , ., [ , 'l] 
Kro1t"~ aVE OllEY, Ilovnv uE Ka'tall.t1tEt mv at1LLnv. 0 oov 'tT\v 1tll.a -

10 mv yao'tEpa 'tau't[l1]~ 'tEL1E'iv Kat 'ti,v Kc.OOtnV tv 1Pij> 'tou 
qcpaC[ O~E]-



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

84 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

V?[u 't]UX£lV' lCaV [U1toO'1t(lO']'tl n~ 'tOY qiSnpov ElC 'tou 'tpauJla:to[c;, 
lCa'ta] -

Q.fl [1t]aALV ElC'tOU [X11paJlOU 't]o ~icp~ Ocrov 'tU6 lCOmn6 
gvalCo[ u<pi~£]-

'tat 't[o Jl£'tEOOpov lCat 'tOY au]'tov 'tP01tOV 'to'\>6 oprov'ta6 u1ta't[uv 
OOlC£l] 5 

yap 't[OO'OU'tov ava~aivnv E]lC 'tU6 qcpam6. Oqov lC(l'tnqLV ElC'tl1[c;] 
15 Jl11x[avilc;. 'tou'toov o.ov oU't]C% yevOl1EvOOV oUlC av lOOtev [0]1. 

ATIO'['tat] 
't"v ['tEXV11V. 'ta 't£ yap SEPJla't]a U1tOlC£lCp{HjI£'tat 'ta 't£ 

. ~ay'xv[a TItt] 
[O'<payllt 1tP01t11ST,O'£'tat, a]:rr.m [i1]11£l6 E~£A[o]v't£6 E1tt 'ton 6[OOJ.lon 

. euO'o]-' 
[ ". ~B] ., , t '\. ,~ '[ Jl£V. lCat 'to ev't£u £ v OUlC£n 1tpoqtaqLV Ot "'llq'tat 'tOOt qOOl1 an, 

. UAA'] '.6 

[ " , llt.] " 'i::' ~ '\' [ ~ nc; 't11v O'opov lCa'tav'lO'o l1ev. a1C'T]lCoa on 'tou "'llq'tapxou l1£tlC pOOt 
1tp008ev] 

20 [£i1tov'toC;. S£lV n 'tOAJl]npo'ta'tov i1tt&el~acreat 1tP06 au'to[uc;' COO't£] 
[EO'n O'Ot 1tpoO'£ABd]v aU'trot lCat \moqxEq9at 'tau't1JV mv 

t[1tiS£t~LV .] 
['tau'ta AEYOOv E]S[£O]l1UV L\ia 3tytOV lCUAroV lCatlCOLVfii 

.. [uvaJltJlvT,]-
[O'lCOOv lCat XP11O''tilc; 't]pa1tECni lCatlCOtvilc; vaua[ylac;. 0 SE XP11O''tOC; 

ofrtoc;,] 22.1 

[MEya JlEv, E<P11, 'to EPlyov UAAU {mEp <piAoU lCaV u1tOBaV£lv lCa[Aoc; 
0] . 

25 [lClVSUVOC;. NOJli~oo S]E, E<P11V. ~ilv lCat KAet:[o]<prov'ta' ~ [yap lCOP11 
1tu90]- 2 

[JlEVoot JlOt lCa'taAt1t£lV au]'tov et1t£V 1tapa 'tOlC; EaAOOlCoO't 't[rov 
ATIO''trov S£]- . 

[S£JlEvOV' .... 'tOY AUO''ta]PXov 1tav'tae; Ot£1C1talO'at ~rov'ta<; 't[ ] 
[ coo't£ u]1tolCelO'£'tai O'Ot 1tap' au'trot xapte; lCat 

a [J.la] 
[EA£ilO'at lCOP11V uBAlav ElC 'tOO'OU'tOU] lCalCOU. 'tau'ta AEyOOv 1tel9oo, 

[lCat O'uv]- .3 

30 [E1tpa~ev i1 TUX11. EYro JlEV o.ov 1t£pt 't]"v 'tou Jl11xav(T,Jl)a'toc; ~v 
O'lC£[ UTtV .] 

[apn SE M£v£AaoU J.lEAAoV'tOe; 'tOlC;] ATIO''ta'ic; 1t£pt 'tile; 9UO'lac; 
.' [AEy£tV.] 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 

[0 A:nO'tapxo<; cp96:oa<; Ka'ta Oat~ova,] ~0J.;0<; [;,~]~[v], ~<{>1), ['t]ou<; 
1tpo?-

85 

['to~uo'ta<; 't11<; lEpEia<; iipXE]oOat [ ~]aAto'ta 
[<hav iivOpco1toV nva Ka'taO-6El\.;] OEn. ro[pa 'tOtVUV d<; aupt]ov [OOt 

• ~a]- .4 

35 [pa.OlCE'\><xi;EOSa.l. 1tpo<; 'tTtv SUO ]~a.v· [OeTIOEl. Oe K]a.l 'tOY o[o]~ 
o iKE't[TlV ] 

[a.~a Oat ~uTlOilv<ll. Kat ~ciAa, ot]'t[o<; ECPTl, 1tp09u~Tl]OO~E9a 
. ~TlO[EVO<;] 

[UJlOOV xEipoue; YEvEOOat. o'tE'i]A.at OrE ;'J.1il~ au'tou]e; OEftoEt 't,,[v 
. KO]-

[PTlV 0><; ap~ootm<; 1tpo<; 't"v ava'to~ftv. 'Y~oov au]'toov, Kat 0 
AnO['tap]~ 

5 

[xo<; ECPTl, 'to lEPEtOV. O'tEUO~EV 0" 't1,]v KOp[TlV 't]OV 1tpOElp[Tl~tvov] .6 

40 ['tP01tOV KaO' Eau'tou<; Kat OappEtV 1tapE]KEAEuo[a~dna, OtE~EA­
OOV['tEe;] 

[eKaO''ta, Kat ~ JlEVElV etO'CO 't11e; O'OpOU] xpft, KaV Oci't''tOV 0 U1tVOe; 
[au]- . . 

['t1,V acpu, 't1,V ;'~Epav EVOoV ~EVElV. Ei OE n ;']~tV E~1tOorov 
[ye]Vll [ 'ta.l.,] 

[O&~E oau't1,v E1tt 'to o'tpa'tom:oov. 'talha ei1tov'te]<; E~ayo~EV 
au ['tTtV ] 

[E1tt 'tOY f3co~OV' Kat 'ta AOt1ta otOa<;. 0><; o-ov T\K]ou[oa,] 1tClV'tO­
~[a1tO<;] 23.1 

col. iv 
[ ' I ] ,~ I [']" " '[ " 1 E'Ytyvo~Tlv K at oLU1tOPOUV ou K E.Xcov 0 'tt1tOtU OCO 1tpo<; 'tOY 

MEvEAa] -
[ov av'ta~lOv.] 'to 0' o-ov KOtVo'ta'tov 1tpoom:arov Ka'tU[01ta~oJlTlV 

Kat] . 
[1tPOOEKUV]OUV ffi<; OEOV, Kat 1l0U Ka'ta m~ ",uxfi~ [aOpoaKa'tEXEt'tO] 
[itoovft. ~ &~ 't]a Ka'ta AEUKt1t1tUV Etxtv 1l0t KaA[&<;, 'OOE 

. KAELVta<;] 
5 [Et1tOV, 'tt YEYO]VEV: Kat 0 MEVEAao~, auK otoa, E<on' [~E'ta yap 

't"v] 
[vauayiav etO]OIlEV EUOUC; au'tov m~ KEpEa~ Aa6oJlEVo[V, 01tOl OE] 
[KEXroPTlKEV OU]K otOa. aVEKrolCUoa on tv IlEan 'tilt Xap[uv 'taxu 

2 

yap] .3 
[Eq>OOVTlO'E Jl]m OaillCOV 'tfi~ KaOapa.~ nOOvfi~. 't[O]v Ot' [fJlE de; Ail­
[yu1t'tov a1tooE]oUIlUKo'ta, 'tOY Ot' filE qXllvollEVOV [ouoa~ou, 'tOY 

~E'ta] 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

86 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

10 ['tTtV AEUlel1t ]7ffiv ELLOV BEqmYrnV. 'tmhov be 1t[ UV'toov Ka'tEO"XEv 'h] 
[8uAa't'ta,] 'lv[a] LLn mv yruxUv [IlOVOV ci1tOAEO"TI, ci'AJ.Jx. Kal 'tTtv] 
['ta<pTtv. <1] 96:Aa't''ta a"(Yoolloov. Ecp8[6]vEO"a$ [i1J.1lv OAoKATtPOU] .4 
['tou 'tll~ <pl]Aav8pC01tlae; qou Bpulla't[o]~. a1tlIlE[V olSv d~ 'to O"'tpa]­
['tOxESoV K]OlVfi Kal me; QK1)Vue; ['tilj~ Ellll~ draco 1tapEA.aoV'tE~] 

15 ['to Aol1tOV] me; VUK'tOe; blE'tpel'l{allEv' [Kal 'to 1tP<X ]r[lla OUK £Aa8E] 
[ ' '\ '\ -] , " ~'" ~ ~'[ M ' '\ --'tOU~ 1tOI\.lW ue;. alla QE Ecp 'tOOl O"'tpa'tTlYOOl 'to v EVEI\.UOV 

ayoo] 24.1 

[Kal 'to 1t]<Xv Ka'taAEyco. 0 BE O"uvTtBE'to Kal 't[ov MEVEAaOV 
1tOld]-

['tal <plAoV, Kal] 1tuV8UVE'tal 1toO"T) 'tl~ ElT) Bu[valll~ 'tOle; Evav]­
['tlOl~. 0 BE £AEYE] 1t[&]O"av EIl1tE1tAll0"8al 'tftV E~l1[~] K[OO]-

20 llT)v [avBprov a1t]oVE[VOTJ]IlEvCOV Kal 1tOAU O"UVEppUT)KEVal 
IlUO"[~ cO]~ dval ~~p[l Il]UplOU~. AEyEl olSv 0 O"'tpa'tT)Y~, 'AAA' 'hlltV 2 
at>'ta[l dO" ]l[V] al XElAlUBE~ lK[ a]val1tpo~ ElKOO"l 'trov EKdvoov. 

. a<pl~ov-

'tat bE [00" ]ov OUOE1too E'tEpOl Olo"X~~~?~ ~1??7 :?~01.e; 'trov all<Pl 'to 
AEA't[a] 

Kal'tTtV 'HAlou 1tOAlV 'tE'taYIlE[vooV E]1tl ['to]u~ ~ap~upou~. Kal alla 
[AE]- .3 

25 [yov]'to~ au'tou 1tal~ EiO"'t~[EXEl 't]l~, AEyCOV a1to 'tou AEA'ta 
1tpOOpoJ.L[ ov ] 

[f\KElV] 'touKd8EV O"['tpa'to1tEB]ou Kal1tEv'tE £AEYEV aAArov ['hIlE]­
[prov Bl]a'tpl~ElV ['toUe; BlO"XtAlO]U~' 'tou~ IlEv yap ~apJ3<ipou[e;] 
['tou~ Ka'ta'tptxov'ta~ 1tE1tauO" ]8al, IlEAAoUO"T)~ BE f\KElV ['tlle;] 
[OUVUIlE~ 'tOV OpVlV amol~ 'to]v lEPOV E1tlOEOTJIl[l1O"al,] 

30 [<pEpov'ta'tou 1ta'tpoe; 'tTtv 'ta<pTt]v' civuy'lCTl OTt 1t<xO"[a 'tTtv] 
[£~oSov E1tlO"XElV 'toooU'toov TtIlEPro]V. Kal 'tle; OpVle; o-tYr~, [OO'tle;] 25.1 
['toO"au'tT)e;, £<pT)v, 'tllllle; ;'~lCO'tal; 1to](av BE Kal KOlll~El 'ta[<pTtv; 

~Ol]-

[Vl~ 'to IlEv ovolla, 'to BE YEvO~ Aialo]",', IlEyE9o~ Ka'ta 'taro[va· 'till] 
[XPOl&l 'tacb~ Ev KUAAEl BEU'tEPO~. K]EKEpaO"'tal IlEv 'ta 1t't[Epa 

XPUO"rol] 2 
35 [Kal 1to]~cp[Upal. auxd BE 'tOV "HAlov B]EO"1tO'tT)V, Kal 'h K[E<paATt 

Ilap'tu] -
[pEt· EO"'t]E<pUvrocr[E yap au'tftv KUKAoe; eU<pUTt]$' Tt[A]~[O]U BE EO"['tlV 0] 
['tou KUKAoU O"'tE<pavOe; dKc.OV. KU]UVEOe; EO"'tlV, pOBO(l)c; EIl<PEPTt[e;, 

EUEl]- .3 

[BTt~ 'titv 8Eav ci]K'tlo"l [KOIl<Xl, Kal] dO"w at>'tal 1t'tEProv civa'to[Aai. 
. IlE]-



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 87 

[p(~ov'tat oE a]1rcov Ai8i[01tE]S ~Ev 'tTtV ~roft[v, Aiy\m]'ttot oE ['tTtv 
. 'tE]-

40 [Ac'U't"v. E1tEt]OaV r~e [a7t09a]vu (cruv X[povrot OE 'tOtl'tO 7tacrXEt 
~aKpon) ,] .4 

[0 7taLS au'to]v E7tt 'tOY NELAov <p[ EpEt, crXEOtacrac; au'tOOt Kat ~v 
. 't<X.<PTtV .] 

[cr~uPV1lS yap] 13OOI~ovJ\Ao9' 'tTtS EUroO[Ecr'taTItS, ocrov lxavov 7tpOS 
OPVl.]-

[90S 'taq"lv, oP]U't''tEt 'tE 'tOOl. cr'to~a'tt [Kat K01.AatVEt Ka'ta ~Ecrov,] 
[Kat 'to op'Uy~a s,,]1C'Tl ydvE'tal. ·'t6)l.· [VEKpOOl.. tv9ds OE Kat 

. .. tvap~ocrro;] 5 

col. v 
1 ['tov 0PV1.V] :~~ cropOOl. ~[at ElS 'to xacr~a Yl'iivrol. xro~a]­

['tt, E7tt 'to]v NEtAoV ou't[~ 17t'ta'tal. 'to [pyov <pEproV.] 
[E7tE'tal.] oE' au'tOOl. XOPOs li[AArov opvierov c007tEp] 
[oop'U<po] prov, Kat E01.KEV <> 5[pvJ.s; a7to01'1~oov'tt 13acr1.A.t:t] 

5 [Kat 'tTtv] ·7tOA1.V ou 7tAaVii'tal 't[Ttv 'HAtou' 0pvl9os auTIt] .6 
[~E'tOtKta ~ ]eKPOU. Ecr'tT\~e~ 'ot>[v E7tt ~E'tEropO'U crK07tOOV] 
[Kat EKOEXE'tCll 'tOUS] 7tP07tOAtO'US 'tOU 9EOO. EPXE'tal. oft] 
['tl.S lEPEUS Ai yU ]7t'ttOS: i3l13AtOV [E~ aou'trov <j)EProv, 1(at] 
[OoKl~a~Et 'tOY OpVl]~ EK tfts yp~cp~[S. 

COMMENTARY 

In the following notes the text of TI4 is collated with codices 
cited in the Vilborg edition as follows: 

Family a: 
W ( Vaticanus gr. 1349), XII cent. 
M (Marcianus gr. 409), XIII cent. (or earlier?) 
A (Anglicanus 16 D.xviii), XVI cent. (copied from M) 
D (Vaticanus gr. 914), XIV cent. (excerpts) 

Family fi: 
V (Vaticanus gr. 114), XIII cent. 
R (Vaticanus gr. 1348), XVI cent. 
E (Ambrosianus gr. 394), XV(I) cent. 
E = Rand E when in agreement 
G (Marcianus gr. 607), XV cent. 

Family cp: 
F (Laurentianus conv.soppr. 627), XIII cent. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

88 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

col. i [17.2-19.2] 

1. Kal del. Cobet 
2. ]11 'tfl~ o'UJ.1<popa[~ TI4 : AU1t1l J.1E 'tfl~ o'UJ.1<popa~ codd. : J.1E om. 

G. TI4 has room for J.1E at the end of the line, which would be 
short without it. 

3. E~EKW<P110E Hemsterhuis : E~EKOU<ptOE codd. : E~EKc.O<p(J)OE Sal­
maSlUS 

3-4. Kal a<patpd[oS]at [E1tEXEipO'Uv 'to ~i<poc; or 'to ~i<po~ E1tEXEi­
pO'UV TI4 : Kal 'to ~i<po~ E1tEXEipO'Uv a<patpEloSat G: Kal E1tE-
XEipo'Uv a<patpdoSat 'to ~i<po~ rell., ViI borg 

4-5. Savu'to'U KaAou rell. : 'tou KaAou F 
5. The lacuna of line 4 will not accommodate the text of the 

codices, while that of line 5 requires a short additional word. I 
have therefore transposed E<Pl1V from its position in line 4 
after SE&V in the codices to line 5 after KaAou. 

6. Kal Eav script.plen. TI 4? : KaV codd. Even so the received text 
lacks 4-5 letters of filling the lacuna; I therefore insert 'tfl~ be­
fore AE'UKi1t1t1l~. II Ptu~EOSE G : Ptu0110SE rell. 

7-8. 'to DE 'tfl~ EJ.1fl~ AU1tl1~ ~i<po~ om. M 
8. Ka'ta1tE1tAl1YE G : Ka'ta1tE1t1lYE rell. 
8-9. 't]EJ.1VEt Ka't' OAtyov 'tl)v 'Vu[xilv TI 4 recte : 'tl)v 'Vuxilv om. 

codd. omn. : 'tEJ.1El J.1E't' oAiyov Schmidt 
9. ap' ante aSaVll'tql Nabers II POUAEOSat ad TI4 : POUAEOSE 

codd. No doubt - a t for -E, phonetically equivalent; I take <lEi 
to modify a1tOSVnOKEtV. 

10. O\)V 0 MEvEAao~ codd. Since the lacuna is 3-4 letters too 
short for the received text, I omit O\)V 0 as the least essential 
words. 

11. OOt avaptrooE'tat M R : OOt vuv aVaPtroOE'tat rell. The lacuna 
will scarcely accommodate vUv. l11tpo~ TI 4a F : EU; P II The 
final letter visible in the line is tau, which would entail a 
highly unusual word division. But to its lower right a small 
triangle of surface is lost, possibly (though doubtfully) 
enough to have accommodated ov; or the scribe may have 
neglected to complete the final syllable in a line already full. 

12. The lacuna requires a space or word of about 5-6 letters not 
preserved in the codices, possibly an adjective modifying 
J.10'U: cf Cruceius' translation (1554), "tantis (conflictatum) 
malis irrides." 

13. A short line; perhaps a word of 3-5 letters in the lacuna was 
lost. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 89 

14. (17t1.(J'tE~ F: (l7tlcr'tEt rell. II Ecp[TI] flOt II4: ECPTI cru flOt codd. The 
small second lacuna has no space for aU; it might have been 
written above the line or in the first lacuna just before E<PTl. 

15. floU II4 : 7tOU codd. II 'tPl~ V£ F : 'tPEt~ aG II £7t(ita~Ev I1 4 : 
E7tcita~E codd. 

16. Ka:tro6ey cpro]Yftc; O.Kouro Kat. 1tayU I14 : Kchro6ev clKOUW 
cpwvTl~ 7t(lvU codd. 

17. EtXE G: icrXEt rell. II a1tE~AE7tOV II 4a : E1tE~AE1tOV F: E~AE1tOV 
p 

18-19. ilVOtyEV (flOU add. W) Kat aVE~atVE V £ mediis omissis: 
~VE~at]VEV II4 alone, which otherwise supports the P tradi­
tIOn. 

19. CPPEtKrooEcr'ta[ 114 : cPptKroOEcr'ta'tov codd. 
20. 7tap' au'tTl~ il yacr'tTtP a.7tacra I14 : cxu'tTl~ il ycxcr'tTtP 7tucrcx 

codd. : il yacr'tTt p cxu'tTl~ 1tacra F: (yap) cxu'tTlC; Castiglioni, 
apparently supported by the papyrus' error 7tCXp II EV'tEPro G: 
EV'tEProV rell. 

21. 1tEpt1tAEKE'tCXl flOt 114 : flOt 7tEpt7tAEKE'tCXt (Efl7tEPt7tA.EKE'tCXt G) 
codd. II Kat crUVE<PUflEV KCXt Ufl<Pro KU'tE7tEcroflEV om. W : KU t 
" 'F' M afl<pw KU'tE1tEcroflEV om. : KU'tE1tEcraflEv 

23. ~AE7tro TI4 : opro codd. II AEUKl1t1tTlV Kpa'tro; 'tuu'tTlC; OUK (?) 
opro transp. G II ante ou KPCX'tro hab. a F 'tau'tTlv, p 'tcxu'tTl~ : 
om. TI4 II OUK post Kat add. M 

24. AaAoucrTl~ a F : A.EyoucrTl~ p II oilv rell. : youv G II a oilv x8Ec; 
E8EUcraflTlV, 'tlva ~v; 11 yap EKEtVa codd. : ]va ~v 11 ya[ I14. 
Spacing of the fragments requires a word of' ca ·4 letters be­
tween yap and £KEtVCX. £VU7tVtCX here, though apparently too 
long, would provide more idiomatic word order, but its 
transposition would require a word of similar length in line 
25. 

25. 'tcxu'ta £VU7tVtCX codd. For the lacuna of 114, see the preceding 
note. 

25-26. ~ro]v EKEtVO 'tE AEU[Kl7t7tTl~ 114 : ~rov ro~ KclKEtvO 'to 'tTl~ 
AEUKl7t7tTl~ p, 'tTl~ om. a F. The 'tE of the papyrus and apparent 
omission of roc; make the awkward text of the codices even 
more difficult. One must suppose either an error or a sub­
stantially different text in the preceding and following 
lacunae. The papyrus' TEAEY may be part of 'tEAEU'taro or a 
derivative, in antithesis to ~rov. 

26. 'AAAa vuv JlEV 0 MEVEAUOC; E<PTl M: JlEV om. rell. The length 
of the lacuna requires at least flEV, preferably an even longer 
supplement. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

90 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

27. o't£pva a ~ : o't£pEa F II ]v<p[ TI 4 : O'UIl<PuoE'tat codd. II Kat 
li.'tpo)'tov O'l'Et codd. Both tlie sense and the lacuna require 
supplement, for which I suggest au'tTtv preceding cl't] POl'tOV. 

28. ava<patVOl post 0'l'n add. G II aAA' E1ttlCclA'U'I'at oo'u 'to 
1tPOOOl1tOV codd. Since the lacuna requires a shorter text, I 
omit aAA' and oo'u as least essential. 

30. (lCa t) ante alla Hercher, for which the lacuna allows no 
room. yap post (lila G II 'to. E1ti 't]11t yao'tpi llay'[yavEulla'ta 
TI4 recte : 'tCt lla"('YavEulla'ta 'tCt E7tt 'tfl yao'tpt codd. 

31. E1ttKa['t£o't110EV TI 4 : a7tOlCa't£o'tl1oEV a RG : a1tElCa't£o'tl1oEV 
VEF 

32. O£ ~ : om. a F : 'tE Hercher IIlCat E[Yro TI 4 : lCayro codd. II The 
codices' text is too long for the second lacuna; I therefore 
omi t IlEv after IlOAtC;. 

33. a1t£o'tl1oa codd. : a1t£o1tCxoa Mehler 
34. 'tltv ante AE'UlCt7t7tl1V add. codd. : om. TI4 
35. 'tou ante MEVEAclO'U VG 
36. 6Eiov G Vmg. R mg. : om. cett. II 1tOt TI 4a V EF : 7t11 G : 1tOU 

Jacobs II 'tl 1tO'tE a F : 'tl P : 'ttva 1tO'tE Cobet II At the right 
margin the papyrus preserves two elongated strokes not 
characteristic of a letter in this hand; possibly it is an irregular 
extension of a final sigma. The lacuna would accommodate an 
additional word not preserved in the codices. 

38. A£yE OE a F : aAA' 11011 AiYE ~ II AiyEt o-ov (, M£v£[AaOe; TI4 : (, 
o-ov MEv£AaOe; AiyEt codd. 

39. 'tou'to TI4 : 'tau'ta codd. 
40. o-ov IlOt TI4 rell. : IlOt o-ov F 
41. lCc1>lll1V rell. : 1tOAtV V II o-ov rell. : youv G 
41-42. va'Uayi~ rell. : va'UllaXt~ E 
42. 1tEpt£7t£OOIlEV P F : 1tEpt£1t£OaIlEV a II 1tpOO£pPt'l'E codd. : 

7tpoOE[ ... ]E TI4, presumably failing to double the rho II 'tile; 
om.G 

43. 'tou l:a'tupo'U rell. : 'tou om. TI4GF lI'tau'tl1v codd. : 'tau'tTI 
Jacobs 

44. The traditional text lacks 5-8 letters of filling the lacuna re­
quired to account for the vertical alignment of the syllables 
surviving from lines 43 and 44. Vilborg (comm. ad loc.) 
suggested 'tTtv Kc1>lll1v or'tTtv Xc1>pav only to reject them, but 
one of these probably stood here. II 1tapa<p'UAa't'tov'tOlV W F : 
1tapa<p'UAaooov'tOlV M : <p'UAaooov'tOlV ~ II 0' TI4 : OE codd. II 
clyollat TI 4, cett. : EPXOllat F 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 91 

col. ii [19.2-21.2] 

1. ]tv Jlou TI4 : A:UOUOt Jlou V £ : Jlou om. rell. 
2. OUV7tOV£tV TI4 : OUJl7tOV£tV codd. 
3. Oil om. V£ II EJlOV OtKE'tllV TI4 : OtKE'tllV om. codd. 
3-4. 01tro~ IT '\ Cabet: OJlro~ caddo The papyrus thus vindicates 

the conjecture of Cobet, who proposed also to read E7ttbd~Et<; 
in line 4, the Attic construction (independent 07tffi<; + fut. 
indic.) equivalent to an imperative (Smyth 2213), as at 6.4.2.23 
and 6.13.3.1. But the scribe of TI4 uniquely wrote E7tl,()£il;ttV, 
above the final syllable of which the same hand (apparently) 
wrote the alternative ending -at, whether as aorist active 
infinitive (so W F) or aorist middle imperative (so M). Pre­
sumably the latter interpretation gave rise to the E7tibEt~ov of 
the ~ family. With the early loss of the future tense, the now 
useless 07[(0<; was ~corrected' to OJlffi<; as all the codices attest. 

4. E<paoav aGF :E<Plloav V £ IIE7ttbd~Etv TI4 a.e., \at/p.e. : 
E7ttbd~Et<; Cobet: E7tlbEt~at M: E7ttbd~at W F : E7tlbEt~ov ~ II 
7tPro'tov 'tOAJlllPOV 'tt oa'U'tou TI4 : o£a'U'tov (Ea'U'trov F) 'tOA­
JlllPov 7tPro'tov codd. 

5. Lacunam post 'tou't(Ot conj. Cobet II xpilotJloV TI4 vitiose : 
XPll<JJlOV codd. II Ka9apat TI 4, rell. : Ka9Ttpat G F. A slanting 
stroke in the left margin marks the line as containing the 
error XPllo { t} JlOV. Cf ii 29 and iii 5 and notes ad Ioee. 

6. Kat om. W 
7. <Jopcp om. P II avaXffipdv TI4 rell. : avaxropTtoat F 
8. {)7[£p~aAA.ot TI4WF: {)7[£p~aAOt MVR: {)7[£p~aAn E Gil 'tOY 

't07tOV TI4 A : 'tOY 'tp07tOV a F : 6 'tp07to<; ~ : 'to lh07tOV Jacobs 
9. 6 <JO<; yap Ev't£u9£v AOYO<; TI 4 : <JO<; yap Ev't£u8EV 6 AOYO<; 

codd. 
10. bE TI 4, codd. : Oil Richards II E~abl~OJlEV TI4 recte : ~ta­

~OJlEVO<; codd. II EKAatOV TI4W F : (Kat) EKAatoV Koenen 
(ap. Henrichs) : EKAaov M ~ II bE<J7tOtVa TI 4 : bE<J7tO'ta codd. 
Satyros here addresses not Clitophon but Leucippe, who is 
not only present but it is she who has requested this long ac­
count for Clitophon's edification. 

11. 7tEpt \'tTt<;l AEUKl7t7tll<; TI4 : 'tTt<; om. M : 7tEpt 'tf\<; A£'UKl7t7tll<; rell. 
II 7tu8oJl£vo<; TI\ rell. : 7tn80JlEvo<; G II b£OJl£VO<; TI 4 : EbEOJlllV 
codd. 

12.11JlOOV aya8[o<;] TI4a.e., llJltV ay. p.e. : aya8o<; llJltV WV£F: 
aya8o<; EV llJltV M : pron. om. G. The corrector writes iota be-



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

92 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

tween dots immediately above the omega; cf line 26 infra and 
note. 

13. 7tpo'tEP~ a : 7tpon:pai~ PF, Vilborg. The length of lacuna 
supports the reading of the a group. 

13-14. Ka8[E~o]JlEvOt 'tilt 8aAa't''tTlt 7tEpt 'tou['trov OK07tO;)V]'tE<; 
TI4 : Ka8E~OJlEVOt 7tpo<; 'tfI 8aAa.'t'tU (8aAa.oou P) AU1tOUJlEVOt 
KClt 7tEpt 'tou'trov OK07tO;)V'tE<; a F. Here the codices appear to 
have padded the text. 

15. 1:00V'tE<; TI 4 II 7tAClVTl8EtOClV rell. : 7tAEOUOClV F II E7t' cett. : E<; F 
16. VEro<; TI4, a F : vTlOc; P 
17. <p8a.[vou]cJtv ClU'tOU[<; TI4 : au'tou<; om. codd. 
18-20. '~v post ClU'tOt~ add. codd. II 'tll 'OJlTtPou cett. : 'tll 'to;) 

'OJlllPOU M II 'trot O'tOJlClU TI 4, codd. : secl. Mitscherlich : 'tip 
0XllJlau Berper II [ca 6]1 'til<; vEro<; TI4, om. codd. Accepting 
the codices' ltv after aU'tot<; in line 18, Cobet added oilv after 
'OJlTlptKltV to rationalize the two apparently paratactic but 
asyndetic clauses. As it stands, the received text is ungram­
matical. Before 'til<; vEro<; one should expect E7t1.. To fill the 
lacuna and rectify the grammar, I suggest roy followin,g 8Ea.­
'tpot<; (without punctuation), replacing the codices' Ttv (line 
18) and obviating the need for Cobet's oily. 

21. 01hro TI 4 a G : 01)'tro<; V £ F II E7tEXElPEl P : E7tEXdpouv Cl F II 
7tpo<; cett. : 7tEpt G 

22. 7tpo)'tou<; EG : 7tpoo'tro<; rell. 
25. n<; £K'tPCl7t[ElOCl TI4 : £K'tpa7tEtoa. n<; codd. II 'trot V ClU Cly trot 

[TI4], codd., to be construed with EK'tPCl7tEtOCl, not with KOJl t-
08EtOCl as by some editors. The lacuna does not afford room 
for Wifstrand's insertion of EV before 'trot vauayirot. 

26. Kat Ka8' iwa<; TI\ confirming Wifstrand's addition of Kat, 
which all codices omit. II 'trot prot suppl. vel emend. supra 
lineam ·7t0ppro· TI4 : 'tip POt P : 'tip pEEl a : 'tip pdSpcp Wmg. : 
lacuna ca 7litt. F. TI4 alone offers the Attic contracted form 
prot for Ionic poi 'flow, tide'. As at line 12 supra, immediately 
above 'trotprot apparently the same hand writes ·7t0PPro· with­
out expunging or striking through the words below (contrast 
the 'correction' in iv line 42, where the corrector, probably a 
second hand, strikes through the rejected syllable and writes 
its replacement above without dots). It is probable, therefore, 
that the scribe uses the dots to signify an alternative reading 
drawn from another manuscript. Palaeo graphically similar, 
1tOppro might be a false emendation advanced by a scribe who 
failed to recognize 'trot prot. In any case, it is unique. II KOJlt-



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 93 

OSElOCl TI 4 , codd. : K'UAtOSE'iOCl G II KA[ om. codd. The lacuna 
affords room for a word or short phrase beginning K Cl -
which is not preserved in the received text. Appropriate 
sense would be provided by either of two phrases used by 
the author, KCl'tU 'tUXTlv (2.7.3.2, 21.4.1, 31.4.1; 5.15.1.1) and 
lCCX'ttt SCXhlOVCX (3.22.3.4 and 5.2.1.1), although either W"ould 
yield an excessively long line. Professor Henrichs persuasive­
ly suggests KCltpiro~, 'opportunely'; though not appearing else­
where in Tatius, it occurs at Chariton 1.14.8.1 and Heliodorus 
1.32.2.5. 

27. 1tClp6v'to~ aJ.1Cl KaJ.1ou cett. : 1tClp6v'to~ EJ.10U F 
28. 'tl,v cett. : 'to F 
29. A slanting stroke extends into the left margin, apparently to 

mark the vacant space of 9 letters across which the scribe has 
extended the top stroke of the sigma of ~i<p~ (el line 5 supra 
and esp. col. iii line 5 infra); no erasure or cancellation of an 
earlier text is apparent, nor do the codices record any alter­
native reading here. II EXOV 1tClACltO'tWV MVEF : EXrov 1tClA. 
W G : nvu 1tClA. nvu R: OOOV Hirschig II The length of the 
lacuna indicates the loss of up to 10 letters before 1tClACltO'troV, 
preserved in no codex. OOOV alone would seem too short : 
perhaps, e.g., 'to J.1ftKO~ OOOV (or oOllv), as I have printed. 

30. 'tEO(J(XProV TI4, codd. Only here does TIo4 show -00- instead 
of Attic -'t't-. II J3pClXU'tCl'tOV cett. : 'tPClXU'tCl'tOV M 

31. OOOV cett. : oorov W. TIo4 could have had either. II OU ll\ cett. : 
OU1tro V II OU 1tAEtro del. Hercher 

31-32. EACl]eEV TIo4 : EMSE codd. 
32. KCl'tcl W F : KClt cett. II 'tou om. ~ 
32-33. 'to J.1tKPOV EKElVO ~i<p~ cett. : om. V 
33-34. 'tooou'tov oaov ElXEV codd. : 'toaou]'tov ElXEv TIo4. Evi­

dently TIo4 inverted the order of the correlatives, since 
without oaov the lacuna of line 33 would be ea 4 letters short. 
The word order assumed for TI 4 is more Attic, that of the 
codices more colloquial. 

34. 'tOUJ.11tClAtV cett. : 'to EJ.11tClAtV F 
35. KCl'tE()UE'tO TI \ codd. : KCl'tCl()UE'tClt M II Since the received 

text is too long for the second lacuna, possibly the papyrus 
omitted either apCl or cb~ Eixo~. 

37. vuv ~ : om. cett. In the lacuna llo4 has room for VUv. 
38. OUVCl'YroVtOE'tClt cett. : auvCl'YroVtaClOeClt G : auvCl'YffiVtEl'tCXt 

Cobet 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

94 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

40. The vertical alignment of the surviving fragment suggests 
that TIo4 may have omitted the codices' on. II PClOtVcll'tO:tOV 
cett. : PClOtVcll'tEPOV G II For ouppa'VOJ,lEV see infra on line 42. 
II OXilJ,lCl cett. : XroJ,lCl F II The first lacuna appears to be 3-5 
letters too short to hold the received text. 

41. e]TlPtc.ov TIo4 : 8TtpEic.ov codd. 
42. pa'VOJ,lEV TI 4 : pa'Vc.oJ,lEV codd. Having the future indicative 

here instead of the codices' hortatory subjunctive, TIo4 prob­
ably had also ouppa'VOJ,lEV in line 40 and E1ttlCPU'VOJ,lEV at the 
end of the sentence. II The first lacuna seems to afford space 
for an additional short word. 

col. iii [21.2-23.1] 

1. pato[v] TI 4 : p~O{ClV F: p~Otox; cett. II OtCl1tt1t'tOt TIo4 a F : 
OtEKXl1t'tOt V £ : OtE1C1tl1t'tEtV G II 'tft lCOpn Cl F : 'tT]v lCOPTlV P 

2. 'tOY om. F 
3. CroJ,lClOtv TIo4p : CroOJ,lClOtV a F II EVOEOEJ,lEVTlV TIo4 EG : EVOEOU­

J,lEVTlV a VF : EVOEOEJ,lEVn Hercher II For [E1ttlCPU'VOJ,lEV] in­
stead of the codices' E1ttlCpU'Vc.oJ,lEV see supra note on col. ii 
line 42. 

4-6. 0 oioTlPoe; yap (lu'tllv EO't(lAJ,lEVTlV Ota 'tClU'tTle; (-'tTlv F) a va­
'tJ,lTleilVClt J,lEOTlV (JlEOOV F) 'tile; Eoeil'toe; AEYEt 0 XPTloJ,l6C; 
codd. : b XPTlOJ,lOC; yap - - - AEyEt anon. ap. Schaefer (8?) : 
[ ? ] I / Clu'tllv E vaeat ea 12 J,lEVTl O'tOATl Ota 'tClU'tTlV 
aVCl'tJ,lTl[ ? ] AEyEt TIo4. The text of the codices is corrupt, 
differently also that of the papyrus, for (as Henrichs noted) 
line 5 is marked in the left margin with the sign /; ef note on 
col. ii line 29 supra. In any case, nps O'tOATl obviates a need for 
the codices' 'tile; roeil'toe; (for which the lacuna affords in­
sufficient space), and the papyrus might be in part emended 
to read [0 J,lEV yap] (lu'tllv E[ O't(lA]J,lEVTl(V) O'tOAft Ota 
'tClU'tTl(C;) aVCl'tJ,lTl[eilvClt J,lEOTlV] AEyEt. Although the scribe 
usually writes iota-adscript in dat.sing., he omits the iota at 
line 10 and col. iv lines 7, 15 and 16. Thus 'tClU'tTl(C;) would 
refer to O't ° Aft , on which the codices' 'tile; Eoeil'toe; is a gloss. 
For the codices' meaningless b oiOTlPoe;. T. W. Lumb (CQ 14 
[1920] 148) proposed the palaeo graphically ingenious 1tOOllPEt 
accepted by ViI borg. Hercher proposed oAolCAll pc.oc; ac­
cepted by some editors. But Jacobs reports the variants 
(mentioned above) found in Schaefer's copy of the ed.prine. 
(now lost) which are assumed to have come from the lost 
Codex Thuaneus (8), for which see Vilborg I p. xxxi. Either 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 95 

8 may have had the same text here as II4 or the Schaefer 
marginalia are astute emendations. The sentence may have 
begun 0 XPllcrJlOe; yap, or alternatively, may have begun 
simply 0 JlEv yap (resuming the subject of the preceding sen­
tence), and criOll poe; may have been intruded by a later 
copyist -who did not perceive the referent. By this episode is 
fulfilled Panthea's dream at 2.23.5, where JlEcrllV avu'tEJlEtV 'tft 
Jluxuip~ 'tTtv YUcr'tEPU anticipates the present passage and 
may offer further clues for its restoration. 

6. op&e; OE 'to'U'tl. n 4 : op~e; tOUtO codd. II Epicr[n n4 : Ep£icrn 
codd. 

7. KOA.EOV II 4 : KO'UA.EOV codd. The papyrus has the Attic form, 
codices the Ionic. II The lacuna has no room for the codices' 
oproVtEe; preceding OOKOUcrl. 

8. pU1t'ttcr8ut II4 : PU1t'ti~Ecr8ut codd. While erroneous omission 
of -~E- is possible, the scribe of II4 (and the author) may well 
have written the Attic and more precise PEpu1tticr8at, which 
the archetype of the codices then trivialized. 

9. aVE80pEV II4 : avt80pE cett. : avtopaJlE F II Ku'taA.i1tEt II4 : 
Ku'taA.£l1tEt cett. : Ka'tUA.t1trov G II ocrov II4 P : Ee; a : Ell; F 

10. yacrtEpa tal>t[ll]v n4 : tau'tllv om. codd. II tEJlEtV n 4 ~ : 

'tEJlroV a F 
11. tt<; n\ cett. : om. G 
11-12. Katappd codd. The scribe of n4 sometimes writes single 

rho for double (ef col. i line 42). 
12. ciVUKO[ n\ ]UKO'Uq> [n s : civaKo'Uq>t~Etat codd. From this 

point to col. iv'line 's 'the fragments of P. Oxy. LVI 3836 (lIS) 
overlap those of II4; all surviving readings of lIS are cited here­
inafter. 

13. ]pov Kat'tov [ II S II tP01tOV 'toile; oprov'tae; II4, codd., ]v 'toile; 
opc.ov[ ns ' 

13-14. oOKd yap tocroUtOV codd. : ] yap t[ n4, ]OKEl yap tOr ns 
14. KU'tUPUtVEtV codd. : Ka'tuPilvut (tv) Cobet: ]atvEtv 'tile; II S 

II t]K 'tile; crq>ayfte; II 4 : 'tile; crq>uyfte; P : 'tft crq>uyft a'F II KU't£tcrtV 
n4 ns : W£tOtV codd. Since the envisioned motion of the blade 
is reversing, II4 and ns doubtless had civapaivEtv in the pre­
ceding clause, as Henrichs proposed (supra n.l : 225.). 

15. JlllX[ II '*, ]vTle; 'tou[ lIS, JlTlxuvile; 'tou'trov codd. II ouv o{h]roe; 
n4 : ou'tc.oe; om. codd. : the lacuna of lIS has not room for both 
oilv and Outc.oe;, so presumably agrees with the codices. II 
tOo1£\' n 4 M, in lac. n S : £l001£\' W : EWotrV F : d&'i£v ~ II An­
cr'tUt codd., A.ncr[ II'*, ]ncr'tUt lIS 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

96 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

16. a1tOKEKptnVE'tat TI~ : a1tOKEKpU1t'tat codd. 
17. 1tP01tT\5lloE'tat p (1tPOO- R ) : a1t01tT\5lloE'tat a F II E1tt om. 

M 
17-18. 8UOOJlEV A G : 8llooJlEV M V e F: 8lloc.oJlEV W II 't 0 

Jacobs: 'to'tE codd. : desunt MG 
18. all' TtJlEtc; codd. : TtJlEtc; om. TI~. In the second lacuna there is 

no room for TtJlEtC;. 
18-19. Kat --- Ka'ta8ll00JlEV om. M G, a clear case of 

haplography. 
19. Ka'ta8llOOJlEV cett. : Ka'ta8lloc.oJlEV W II ax:ilKoa 5it TI4: 

aKllKoa codd. II OUK ante aKllKoac; add. G II JlEtK[p&t TI4 : 
JltKPCP codd. 

20. 5EtV cett. : 5d G II nom. M II 'to All ]T\ po'ta'tov TI 4 : 'tOAJl T\ POV 
codd. : ]OA[ TIS, with lacuna apparently too small for the su­
perlative II E1tt5Ei~a08at TI 4, cett., ]~a08at TI 5: E1tt5Ei~aV'tEC; F 

20-21. COO'tE Eon codd., ]O'tE Eon [ TI 5 

21. ]v au't&[] I [ e a 11 Jat tau'tT\''" TI5 
22. 't·au'ta TIS, codd. Ir ·EOEOJlT\V i\ia codd., ]JlT\v i\ia TI4 : ]T\v 

ME[VEAUOU i\ia EE]VtoV Ka[ TI5 
22-23. [ ea 8-12 I ea 10-14 't]pa1tE~T\C; Kat KOtVllC; TI4 : avaJltJl­

VllOKc.ov --- 'KOtVllC; om. M : 'tpa1tE~T1C; om. F : avaJltJlVllOKc.oV 
'tpa1tE~T\C; Kat XPT\o'tilc; Kat KOtVllC; cett. The first lacuna of line 
23 no doubt held Kat XPT\o'tllC;, which the codices have 
transposed to follow 'tpa1tE~T\C;, giving rise to confusion and 
conjecture. I should reconstruct: KOtVllC; [avaJltJlvllloKc.ov 
Kat XPT\o'tilc; 't] pa1tE~ T\C; Kat KOtVllC; vaua[ 'Yia~. 

24-25. a1to8avdv TI 4 : a1t08aVEtV 5EllOn (5EllOEt M) codd. : ef 
KaV a1t08aVEtV OEllon vel oEn Xen. Eph. 2.7.5, 4.5.3 II Ka[ TI~ : 
KaAl>c; 0 KivouvoC; a F : 'YAUKU~ 0 8uva'toc; P : ambo retinent 
edd. II Here in the lacunae of lines 24-25 there was clearly a 
different text. There seems to be no proper place for the 
OEllon omitted after a1to8avEiv : either it (or OEn) stood at the 
end of line 24 after Ka[Aoc; or it was omitted altogether either 
by error or by the rare construction (ef Soph. El. 1483) 
wherein an obvious subjunctive introduced by KaV is 
omitted (Smyth 1766.b). If the following Ka[ began the 
phrase KaAl>c; 0 KivouvoC; as attested by the a codices and F, 
the line remains ea 4-7 letters short at the right margin and 
line 25 3-4 letters short at the left, but insufficient room is 
given for the rival phrase 'YAUKUC; 0 8uva'toc;. If Ka[ intro­
duced some different word, there would be space for either 
phrase but not for both. For KaAo~ with Ki.v5uvo~ there are 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 97 

many parallels, from the locus classicus at Plato Phaedo 
114D6 onward to Heliod. 7.1.1 and Charit. 1.1.4, 1.10.6,5.3.1; 
for YA:U1(1)e; with 8uv(X'toe; only Ach.Tat. 2.26.3 and Anth. Gr. 
14.36.2, so far as I can ascertain. If, however, an original 
1(aAOe; (, 8uva'toe; (which one would expect) gave rise to an 
amplification, there are even more parallels, including Ach. 
Tat. 3.17.3 and 6.2.10. I find no occurrence of YAU1(1)e; with 
1(tVOUvoe;. After 1(aAOe; there would be room for OEn or the 
missing ikflon, but its position there would be awkward and 
unparalleled. 

25. Eq>llV ~ : Eq>ll a F II 't£ post 'h add. codd. : del. Hercher, Vil­
borg. The surface of I14 is broken away here above a kollesis. 
II 1(0 Pll cett.: om. M 

25-26. 1tU80JlEVql cett. : 1t£l80JlEVCP F 
26. d1t£ ante 1(a'taAt7tdv transp. F l11tapa. I14, cett. : 1t£pt V 
27-28. Here I14 had a radically different text from the codices' 

amplification of the context, 01 OE 'troy AnO'tOOV 1tPOC; 'tOY 
AnO'tapxov E1(qmyoV't£e; EA£YOV (1tpOe; F) (JlEV E) 1tuv'tae; (JlEV 
VRG) 'toue; im' au'toov dAllJlJlEVOUC; 'tflv dC; 'to o'tpa'to1t£OOV 
lluXllv E1(1t£q>£uy£val-a loose style not characteristic of 
Tatius, and not consistent with 3.13.7. To fill the lacunae one 
might consider, e.g., [ .... 'tOY AnO'ta]pxov, 1tuv'tae; Ot£1(1tatOal 
srov'tae; 't[ oue; o£oJlcO'tae; I de; 'to o'tpa'to1t£Oov. wa'tE &]1t01(dOE­
'tat 001, which would satisfy the required sense and fill the 
lacunae, "the bandit-chief, that all the captives without loss of 
life had broken through to (the safety of) the army camp"; I 
cannot guess what governed AnO'tapxov, and it is of course 
fruitless to supplement lacunae so large. I take OtE1(1tatoat as 
infinitive continuing the oratio obliqua of the preceding 
clause; in Josephus, Pausanias, and Philostratus the verb is 
intransitive, in Appian and Plutarch transitive. It is a relatively 
rare word (el its synonym Ola1C01t'tOJl£V at 3.13.5) not occur­
ring elsewhere in the romances and apparently unfamiliar to 
the scribe of the codices' archetype. 

28. au'trol I1 4, cett. : au'toov G II XUple; II 4 : 'h XUple; codd. : del. 
Hercher 

29. EAr110al codd. : Ep1>On Jacobs 
29-30. The few letters surviving at the margins of II 5 accord 

with the vulgate text (ed. n. ad loe.). 
30. oilv cett. : om. V II 1t£pt ~ : 1tapa. a F II JlTlxavTtJla'toe; codd. : 

JlTlxava'toe; II4. Anticipatory omission of - TlJl- is one of the 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

98 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

few certain errors of the scribe of TI4. II cnn:ul1v cett. : 1ta.pa.cr­
KEUl1V G 

31. J1EAAoV'tO<; cett. : AEYOV'tO<; F 
32-34. The vulgate text of these lines condenses that docu­

mented in the fragments of TI4 and TI s, which in turn 
evidently differed from each other. 

32. Ka.'tft oa.iJ1ova. codd., in lac. TI4: ]a.1J10Va. o[ TIS II NOJ1o<; TtJ1tV 
Ecrnv a F : TtJ1tV V E : NOJ1o<; [Tt J1]t [v ] TI 4 : N'0J10<; ~v G: in lac. 
TIS . . . 

32-33. eCPl1 'tou<; 1tpro'tOJ1ucr'ta.<; Commelini ed.pr. (from C?), eCPl1 
['t]ou<; 1tPro[ TI4 : ]CPl1 'tou<; 't[ TIs, where the editor thinks the 
crossbar of 't extends too far left for 1t to be read: 1tOU vice 
'tou<; a : ECPl1: 1tPOl'tOJ1ucr'ta.<; cett. : transp. G 

33. 't11<; tEpda.<; (tEpoupyia.<; FWmg.) elPXEcr9cll, J1a.A1cr'ta. a ~ : 
]cr9a.1 [ ca 15 J1]a.A1cr'ta. TI4 : 't[ ca 11 ]11<; i£lp[ ca 12 ]0.<; a1tO I 't[ 
ca 10 elp]XEcr[9]a.1 I J1a.Atcr'ta.·TIs, for which the editor suggests 
as possi ble restorations 't [11<; 't01a.U't ]11<; tEl p [da.<; 1tpro'tO­
J1ucr't]a.<; or 't[O 1tpiihov 'tOU'tO 't]11<; tElp[da.<; J1E'tEXOV't]a.<; a1tO 
I 't[11<; 9ucria<; elp ]XEcr[9] 0.1. Neither TI4 nor TI S offers help for 
the restoration of the other. 

34. <ha.v elv9pro1toV Ka.'ta.9uE1V OEn codd.: ] OEn TI4 : elv]9pro1toV 
n[va. Ka.'ta.9U£lV olEn TI Sed. II ~ cett. : E<; G II a.UP10V 001 (cru 
W) codd. : a.up]tov I crO[l TIS, ]ov [ TI4 

34-35. 1ta.pa.crKEUa.~Ecr9(li 1tpo<; codd. : ]Ecr9a.t 1t[ TIs, in lac. TI4 
35. 9ucria.v· OcitcrE1 codd.: ]ta.v oEillcrE[l TIs, ha.v [ TI411 Ka.t 'tOY 

crov OtKE't1lV codd., K]a.t 'tOY cr[o]v OtKE't[l1V 1 TI4, OtKE]I'tl1[v TIS 
36. J1a.Aa. cett. : J1a.A.tcr'ta. G. By lacuna space TI4 and TIS 'support 

J1a.Aa.. II o-o'to<; a F : 01hoo<; p l11tp09uJ1l1croJ1E9a. cett. : 1t p09u­
J111911cr0J1E9a. M F : ]croJ1E9a. TI4, ]19u[ TIS 

37. UJ1rov cett. : TtJ1rov F : TtJ1tV G II TtJ1ii<; cett. : uJ1ii<; VR II a.u'tou<; 
om. F II OE post a.u'tou<; add. W 

38. apJ1ooiro<; a F : apJ1ooiou<; P II UJ1rov p : om. a F. Vertical 
alignment with the preceding line leaves a space for ca 5-7 
letters additional in the lacuna to the left: apparently n 4 had 
[ ... UJ1rov a.U hrov, reflecting TtJ1ii<; cxu'tou<; of the preceding 
sentence. 

39. <J'tEAAOJ1EV cett. : J.1EAAOJ.1EV F II1tpo£lPl1J1EVOV TI\ cett. : EiPl1-
J.1EVOV G 

40. ecxu'tou<; cett. : cxu'tou<; W F 
41-42. eii't''tov b U1tVo<; [a.u'tllv TI4, p : eii't'tov a.u'tllv b U1tvo<; 

aF 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 99 

42. d codd.: llv Hercher II 'tt Cobet : n; codd. II it1J.iLV EJ.11toOffiv 
II4, Jacobs: llJ.1rov flC7tOOOOV codd. 

43. 'to cett. : om. F II au'tTtv II4, cett. : 'tTtv lCOPTlV M 
44. ~C.t>Jlov· lCat codd., ]JlOV' Kat lIS II otoae; a F : dOEe; ~,]tOEe; lIs 

II oilv cett. : ,,(OtlV G II iiKouoa 7tav'tooa7tOe; codd., ]o'U[oa] 
1tav'to~[ II4, ],,?oa 1tav'toS~[ II 5 

col. iv [23.1-25.5] 

1. f"(tyvOJlTlv (f"(tV- G, fYEV- R) lCat OtTl7tOpo'Uv aVE F : ]tv0JlTlV 
Kat Ot[ lIS II [OU]lC EXC.t>v II4 : OUlC EXrov om. codd. et IT sIn lac. II 
o 'tt iotTtoC.t> codd., 0 n 7tOtTl[ II 4, ] ~~ 7tOtTtoC.t> lIS 

1-2. 1tpOe; 'tOY MEvEAaov codd., 1t[ ca 7 M]EvEAal[ IT s, in lac. IT4 
2. 'to 0' o-ov IT 4, codd. : 'to ~/oiJv IT S (n. ad IDe.) II KOtVO'ta'tov 

7tPOO7tEOOOV II 4, codd. : hoy 7tP07tEOo)[ lIS II lCatTlO7taCoJ.1Tlv 
lCat codd., lCatTl[ II4, ]o7taCoJlTlv Kat ['ITs 

3. 1tPOOEKUVO'UV roe; eEOV codd., ] llC'Uvo'Uv roe; e [ IT S, ]o'Uv roe; eEOV 
II4 II Kat J.10'U Ka'ta 'tile; 'V'Uxile; IT 4, conj. Cobet : lCat J.1O'U Ka'ta 
'tTtv 'V'UXTtv codd., ] JlO'U lCa'ta 'tTtv 'V'Ux[ II S 

4. llOOVTt. roe; OE to. (to V £) lCata AE'UlCt7t7tTlV codd., llo[ovTt]1 roe; 
OE 'to. Ka'ta AE'U[lCl7t]I7tTlv ITS, t]a lCa'ta AE'UKt7t7tTlV II 411 dXE 
JlOt lCaAroe; codd. : ElxEv J.10t lCaAffi[ lIS, ElxEv JlOt lCaA[ II4 II b OE 
KAElVtae; lIs, codd. . 

5. d1toV ti "(E"(OVEV codd., [d]1to[v] 'tl YE"(OVEV ITs, ]VEV IT 4 IIlCat b 
MEVEAaOe; IT4 : 0 MEVEAaO~ .~ : b OE MEVEAaOe; ITS cett. II OUK 
otoa E<I>TI (£<I>Tlv G) codd., IT4, ITS II J.1E'ta "(ap Thv lIS, codd. 

6. [va'UaYlav £LO]OJ.1EV EUeUe; II 4 : va'UaYlav EUeUe; ElOOJ.1EV ~: 
va'Uaytav EUeUe; doov J.1EV lIsa F II au'tov 'tile; II 4, codd., 
a[u]tov 'tile; ITS IIlCEpEae; IT4ITS : lCEpatae; codd. II Aa~OJ.1EVOV 
codd., Aa~OJlEvo[ v II 4, Aa~[ 0 ] JlEVOV II S II 01tOt oE II s, cett. : 
07to'U oE F : 07tTl oE G 

7. lCEXroPTllCEV OUlC otoa codd., lCEX[ro]PTllC[E]V OUlC otoa' lIS, 
OU]lC oioa II4 II aVElCoolC'Uoa oTt II 4 : aV[E]lCro[lC]'Uoa J.1EV ouv 
lIs: aVElCwlC'Uoa (ov Fsupraser.) o-ov (,,(OtlV G) codd. II EV 
J.1E011 'tn xap ~ codd., f[ v J.1Eo]11 'tn xap~ II S, fV J.1E011 'til t 
Xap[iXt II4 

7 -8. taxu yap fcpeOVTlOE Jlot codd. : 'tax[ U I yap fcpeOVTl]OEV Jl[ Ot 
ITs, which breaks off here: JOt IT 4 

8. 'ttc;; post OatJ.1C.t>v add. codd. 
8-9. 't[o]v Ot' [fJl£ ea 5 lea 5 a1tOOE]OTlJ.1TllCo'ta II4, om. codd. 

The lacuna doubtless began with fJ.1E, making explicit the 
reason for its repetition in line 9 and the codices' omission of 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

100 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

the phrase by haplography; Henrichs' supplement appears to 
be confirmed by the added Robinson fragment. 

9 • 'TI4 ,~ V . EIlE , cett. : EIlOU 
10. OEil'tEPOV post AEUKt7t1tTlV add. F II EK 1t[uv't(J)v TI4, rell. : om. 

p 
11. SaAaaaa codd. II 1l0VOV a : 1l0v1'\v P F 
12. ib SUAaaaa clypia ante ib add. G II SaAa't''ta TI\ SUAa't'ta 

a F : SUAaaaa P (cf. col. ii lin. 14) II ayvwJlwv II4 : liyvwJ,1ov 
Jacobs: ayvroJ,1ov codd. II E<pS[o]vEaac; TI 4 : E<pSOVTlaac; codd. 
Here TI4 writes the later orthography: cf. LSJ S.v. 

13. aou TI\ cett. : 'tou G : om. F 
14. a1CT\vilc; ['til]c; Ellilc; E[ TI 4 : a1CT\vilc; daco 1tapEASov'tEC; 'tilc; EJ,1ftC; 

transp. codd. To fill the lacuna at the right margin, TI4 must 
have read a1C'Jlvf\c; 'tf\c; EJ,1f\c; daco 1tapEABov'tEC;. 

15. OtE'tpEiWaJ,1EV TI4 : otE'tpiwaJ,1EV codd. 
16. rep 'trot a'tpa'tTlyrot 'to[v TI 4 : rep ayco (1tpoaayco Jacobs) 'tOY 

MEVEAaov 'tip a'tpa't1'\yip transp. codd. The papyrus must have 
had'tov MEVEAaov liyco in the right-hand lacuna. 

17. [Kal'to ? 1t]<iv Ka'taAEYco TI4 : Kal1tav'ta AEYCO a F : Kal (om. 
G) O:1tav'ta AEYCO P 

18. Kal] 1tUVSavE'tat 1toaTl nc; dTl ou[valltC; TI 4 : 1tUVSaVE'tat OE: 
1toa1'\ (1troc; it W) ouvalliC; Ean codd. In TI4 1tUVSavE'tat must 
have been introduced by Kal or £1tEt'ta OE: vel sim., and dTl 
here (optative after historical present) was replaced by the 
codices' Ean: "And (then) he enquired how large a force the 
enemy had." 

20-21. auvEppuTlKEVat J,1ua[oc; ro]C; TI4 : auvTlSpo'iaSat Ana'tllPtoV 
codd. For the vivid and uncommon phrase "a great abomina­
tion has streamed together" the codices substitute the weak 
"a great robber-camp has collected." 

22. au'ta[ .... ]t[.]at TI4 : ai>'tat 1tEV'tE codd. : (ai) ante 1tEV'tE 
suppl. Cobet. Apparently ail'ta[i da]t[v] ai without a 
number: "we have these thousands adequate against those 
twen ty." II XEtAtUOEC; TI 4 : XtAtaOEC; codd. 

23. OUOE1tCO TI 4, cett. : OU1tCO M II r'tEpOt OtaXlAtot 1tPOC; 'tou'tOtC; 
TI4 : 1tPOc; 'tou'tOtC; r'tEpOt OtaxiAtot transp. codd. 

24. 'HAlou xOAtV G : 'HAtoUXOAtV cett. After 'HAiou TI 4 shows a 
slight space, like the space this scribe often leaves between 
words. 

26. 'tOUKE'iSEv TI 4WV : 'tou KE'iSEV cett. II £A.eYEV TI4: AEYEtV M G, 
edd. : AEYCOV cett. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

WILLIAM H. WILLIS 101 

28. 'to\><; del. Cobet II KU'tU'tPEXOV't(X<; cett. : KU'tEXOV'tU<; G 
28-30. Vertical alignment indicates either that 5-7 letters are lost 

from the first lacunae of each of these lines, possibly leaving 
an open square of ca 2 cm. inset from the left margin or (so as 
not to interrupt phrases) within the column, as shown. It is 
hard to imagine that each of the lines lost some word of the 
same length at the same relative position. Perhaps the pa­
pyrus had a fault which the scribe had to avoid; or perhaps he 
left space for a small drawing of the Phoenix, for ancient 
copies of romances were sometimes illustrated (cf K. Weitz­
mann, Ancient Book Illumination [Cambridge (Mass.) 1972] 
52 and fig. 38a), though I know no parallels either for an 
original hole in a literary roll or for a single small vignette 
inset in such a text. 

30. 'tllv 'tuCPTtV cett.: 'tacpov V II ANAr'KHL\HIIAL[ II 4 : uvaYKll 
(aut uvaYKUt?) 0' ~cruv codd. : uvaYKUt 0' ~cruv ed.pr., 
Hirschig : ,;vuYKacr81lcrav Goettling: uvaYKll(v) OE 1tacra(v) 
Jacobs: avaYKllv 0' rivat Hercher : avaYKU OE1l0av Lumb : 
avaYKll 0' ~v Vilborg. Among these only Jacobs comes close 
to the original reading of II4 : did he find it in Schaefer's notes 
from cod. 8 (see note on col. iii line 2)? The L\H of II 4 might 
be interpreted Oll, 0' 1), or o'~, though a verb is not required 
here. 

30-31. 'tllv E~OOOV E1ttcrXE'iv 'tocrou't(ov 1)JJ.EPWV cett. : 'toO'ou't(ov 
E1ttcrXE'iv 'tllv E~OOOV 1)JJ.EPWV M 

31. 'tic; opvt<; II\ W F : 'tie;; 0 OPVt<; cett. II oo'tt<; cett. : 0<; G 
32. ECPllV ante 'tOOU1J'tlle;; F, post 'ttJJ.lle;; G IIKoJli~Et II 4, cett. : 

KOAa~Et G 
33. <l>o'iVt~ JlEV (, opVte;; ovoJla codd. For this the first lacuna is 

too short; the suggested restoration better balances the fol­
lowing 'to OE YEVOe;; Ai8imjl and preserves the chiasmus. The 
codices' (, OpVte;; is probably a gloss. 

33-34. 'tV XpOt~ cett. : 'tllV xpotav MG 
34. EV KaAAEt del. Hercher. The length of the lacuna requires 

these words. II JJ.Ev 'ta II\ cett. : JlEV yap 'ta D 
36. EO''tEcpav<OOE II 4, codd. : Ecr'tEcpav(oKE Jacobs II EUCPUTt<; codd. 

(Eustath.) : EucpaTt<; Wifstrand : EUcpEme;; Cobet 
37. PO~OL TI4 vitiose : pOOote;; codd. 
37-38. pOOOte;; --- KOJl~ om. M 
39. a]U'tov II4 : au'to\> codd. The scribe of II4 evidently under­

stood uu'tov as dir. obj. of JJ.Epi~ov'tat, ~(Oftv and 'tEAEU'tTtV as 
acc. of respect, while the codices take these nouns as dir. obj. 



WILLIS, WILLIAM H., The Robinson-Cologne Papyrus of Achilles Tatius , Greek, Roman and 
Byzantine Studies, 31:1 (1990:Spring) p.73 

102 THE ROBINSON-COLOGNE PAPYRUS OF TATIUS 

limited by the gen. airto-u : "The Ethiopians have possession 
of him in his life, the Egyptians in his death," cf LSJ s.v. 
JH:pt~c.o II.l. 

41. au'tep P : oU'tc.o a F 
42. ~roAov codd. : ~roAo~ corr. e ~c.oJ.lOV n4. While ~roA.oV is more 

graphic and doubtless the right reading, ~c.oJ.lOV in the sense 
ctomb' is at least possible. Unlike the variants recorded be­
tween dots at col. ii lines 12 and 26, the syllable -J.lov is 
cancelled, and above it an apparently different hand has 
written -A.o~, nom. for acc. 

44. y£ivE'tat n 4 : ytVE'tat codd. II 'tep VEKPep cett.: 'trov VEKProV F II 
oE cett. : ya.p V 

col. v [25.5-.6] 

1. Kat d~ codd.: Kat KAEiCfa~ Hercher: KAEiCfa~ 'tE Jacobs : 
KAEiCfa~ Castiglioni : Kat ~UCfa~ Hirschig II 'Y'ltVCP codd. : 
CfJ.lUPVtVCP Jacobs. Averaging 39 letters, the lines of col. v are 
some 20% shorter than those of the preceding columns. The 
probable length of the second lacuna would accommodate 
the text of the codices better than that of the emenders. 

3. The received text gives a line 3-4 letters shorter than average. 
Perhaps flAA.otc.oV for CiAAc.oV? 

4. oopucpopc.ov G : oopucpoprov cett. 110 opVU; n 4, cett. : 0 om. V 
5.1tAavcl'tat a F, prob. ll4 : 1tAaV~ p II (~'Tl'trov) vel (1tpOCft1t'ta­

J.lEVO~) post 1tAaV(l'tat Richards. The papyrus has no room 
for either. 

6. CfK01troV V, edd. : CfK01tOU cett. 
7. 1tp01tOAOU~ cett.'rrob. ll4 : 1tpOCf1tOA.oU~ G 
[10-17]. The rest 0 the received text of Book 3 would have oc­

cupied five additional lines, not including a colophon. 

DUKE UNIVERSITY 
October, 1990