AUTOPSY AND DIDACTIC AUTHORITY: RETHINKING THE PROLOGUE OF THE PERIODOS TO NICOMEDES

Abstract All modern critics have read verses 128–36 of Pseudo-Scymnus’ iambic Periodos to Nicomedes (c.133–110/109 b.c.e.) as a description of the personal autopsies of the author. However, close analysis of both the literary dynamics of the poem and the syntax of the lacunose text that precedes this passage shows that this cannot be the case. This article proposes that Timaeus of Tauromenium (c.350–260 b.c.e.) is a superior candidate for the referent of these lines, and offers a coherent approach to emending the manifestly corrupt text. This reinterpretation makes better sense of the extant text of the Periodos, and allows these verses to be read as a second-century witness to Timaeus’ autoptic prowess.

Hellenistic iambic didactic poems. 4 The text, addressed to King Nicomedes III Euergetes (reigned c.127-c.94 B.C.E.), 5 constitutes a clockwise geographical tour around the οἰκουμένη. 6 Scholarly commentary has focussed on its extensive prologue (1-138), in which PS seeks to establish himself as a figure of intellectual authority, consistent with the conventions of both didactic and patronal poetry. 7 Recent studies of the prologue have identified personal autopsy as a key device upon which PS relied to ground his authority as a geographical poet, and which formed part of his adoption of knowledge authentication strategies more commonly associated with prose. 8 However, this interpretation should give pause for thought: for one, personal autopsy is absent from the other surviving iambic didactic poem, the first-century B.C.E. Description of Greece by Dionysius, son of Calliphon, which seems to be a conscious imitation of PS' model. 9 Moreover, we propose here that the single passage in which PS is thought to present his autoptic credentials-lines 128-36, preceded by a lacunose catalogue of PS' sources (113-27)-has been systematically misread. Rather, this passage is better understood with the prominent historian of Western Greece, Timaeus of Tauromenium (c.360-260 B.C.E.), as its referent. Our proposed identification of Timaeus makes better sense of the geographical and biographical data in the passage, and facilitates greater syntactic and literary cohesion. This reading necessitates emendation of the manifestly corrupt text preserved in the sole manuscript tradition; we propose a coherent solution that rectifies both the literary and the syntactic defects.
This revision prompts reconsideration of PS' methodology: by excluding the autoptic element, we show that, like Apollodorus of Athens (c.180-120 B.C.E.) before him, and Dionysius after, his central objective was to compress and versify information extracted from existing prose sources. Although these poets ground their credibility in the merit of their sources-explicitly invoking modes of knowledge authorization evinced in prose texts-they themselves eschew first-hand verification, focussing rather on the selection and poetic presentation of others' research. Moreover, PS' positioning of Timaeus as a reliable autoptic observer provides new insights into the latter's historiographical methodology and reception.  The damage is such that the text is essentially irretrievable. Bravo reconstructs the entire lacunose section; 14 however, his readings are highly speculative, relying on text seen by no other editor. 15 We have had the benefit of consulting a high-resolution reproduction of MS page 128, provided by the Bibliothèque nationale de France: this copy permitted digital manipulation of colour, contrast and scale, such that we were able to adapt and supplement prior visual readings. 16 Accordingly, while we also see isolated words and characters in lines 3-6, 17 we provide here a revised diplomatic edition of the text at lines 7-8, equivalent to line 126 and its surrounding text: As for the end of line 7, 22 early editors mistook ενίων for ποιῶν, owing to the faint descender of the ν. 23 The abbreviated forms after the following interpunct also caused difficulties: Höschel and Morel printed καὶ λίαν δὲ Σικελόν, which Gail amended to Καλλίαν δὲ Σικελόν, identifying the otherwise unknown 'Callias of Tauromenium'. 24 However, these readings are unmetrical, 25 and fail to account for some of the characters preceding σικελόν. The end of line 7 is admittedly difficult to decipher: while the abbreviated καί and word-initial τί are clear, the intervening mark-a left-facing lunate stroke, squeezed between two words 26 -is most likely an abbreviated δέ. 27 Although this is a novel observation made by no previous editor, the δέ introduces metrical irregularity and erroneously duplicates the δέ after ενίων; as such, it should be deleted. 28 27 Although the stroke is angled more highly than the scribe's usual abbreviation of δέ, the shape is paralleled at e.g. MS page 137, line 23 (= 495). 28 The scribe frequently misplaces δέ: see e.g. MS δὲ τούτων for τούτων δέ (146); MS δὲ κεῖται for κεῖται δέ (478). 29 In addition to those works cited below, see also the translations at Müller (n. 1), 1.199; Marcotte (n. 1), 109; Korenjak (n. 1), 27; Bravo (n. 14), 19-20.
And there are also the things which I myself have diligently examined in person, bringing the guarantee of eyewitness testimony to bear, as someone who has seen not only Greece or the towns situated in Asia, but having become knowledgeable about towns along the coast of the Adriatic and Ionian Seas, and having gone to the borders of Tyrrhenia and Sicily and towards the west and to most of Libya and Carthage.
The logic of this transition, from the list of consulted sources to the digest of autoptic observations, has never been substantially challenged. Lightfoot, following Hunter, takes it as an indication of the author's reliance on both autopsy and 'book-learning' to reify his material: 'Τhe promise of a "guarantee of eyewitness testimony" (αὐτοπτικὴν πίστιν) aligns PS's work with prose traditions of geographical and historical writing, which often stressed the importance of personal autopsy and inquiry rather than a dependence on knowledge gleaned solely from books, as a means of bolstering the authority of the work as a whole.' 30 Korenjak stresses the novelty of PS' invocation of autopsy here, noting that he, like Polybius, is one of the first authors to foreground autopsy so programmatically as a guarantee of his work's credibility. 31 He does, however, cast doubt on the reliability of PS' apparent declaration of self-description, given that these travels are not mentioned elsewhere in the poem, nor does PS again defer to autopsy to bolster his didactic credentials. 32 Bravo takes this further: though reading PS as the traveller described in lines 130-6, he deems the rehearsal of locations visited 'une fiction littéraire'. 33 This suspicion is well-founded. There are, in fact, ample reasons-both textual and literary-to doubt that the author of the Periodos is the referent of lines 128-36.

The subject of line 128
Although the Greek text opening the allegedly autoptic passage does not specify any shift to first-person self-description, the two grounds upon which prior readers seem to have based their identification provide insufficient evidence for any such transition.
First, Meineke, followed by all subsequent editors, prints an interpunct at the end of line 127, adopting the manuscript's punctuation. This would indicate a sense-break: ἐκ τῶν ὑφ' Ἡροδότου τε συντεταγμένων⋅ | ἃ δ' αὐτὸς ἰδίᾳ φιλοπόνως ἐξητακώς. However, given that the scribe regularly uses interpuncts to demarcate lines of iambic trimeter, regardless of whether a sense-break is present, 34 Meineke's mark may be disregarded. Indeed, we note that earlier editors-from Höschel to Fabricius-print a period after συντεταγμένων, perhaps following the low dot on the apographs. 35 Second, the αὐτός at line 128 fulfils no anaphoric function: its referent is left unspecified, with no textual basis for understanding ἐγώ. No first-person verb, nor any first-person pronoun, occurs in the following verses; rather, αὐτός governs several nominative masculine singular participles, unmarked for person (128 ἐξητακώς; 129 προσενηνεγμένος; 130 ὤν; 132 γεγονώς). Here αὐτός merely serves as an intensive adjective, emphasizing the understood subject; 36 in this sense, it strengthens ἰδίᾳ, to convey the independence of the investigation ('personally, by oneself'). 37 It is thus highly unlikely that αὐτός refocalizes an earlier first-person referent; indeed, αὐτός performs an anaphoric function in this period only in oblique cases, or after the article. 38 These observations suggest that the text itself does not provide sufficient justification for readings identifying a first-person subject in line 128. Τhe resumption of first-person speech at line 137-τὰ δὲ πολλὰ συνελὼν ἄρξομαι τῶν πραγμάτων-better demarcates a transition to the author's self-descriptive mode.

Autopsy and 'book-learning' in the Periodos
The substance of lines 128-36 also tends against viewing the author of the Periodos as its referent: to take these verses as a description of personal autopsies, supplementing the textual sources the author relies upon, contradicts the description of his project elsewhere in the poem.
To put it concisely, in not suffering the wanderings of Odysseus, as they say in the myths, but remaining happily in his own home, he [that is, ὁ κατακούσας, Nicomedes] 39 will learn not only the foreign way of life of people but also the cities and customs of all nations.
The clear surface-level interpretation of this passage is that avoiding travel (98-9 τὴν Ὀδυσσέως … πλάνην) is a benefit incurred by the text's recipient, though not necessarily a reflection of the author's method. Lightfoot extends this, positing that the benefit is accrued because PS has-at least putatively-endured these travails on their behalf, preserving his findings in a readily consultable and memorizable text. 40 Yet PS' description of his own method does not suggest that he had travelled the globe gathering information for his patron. When commenting on his project or research process, he specifies that the text is a synthesis of others' written work, not a poetic log of his own autoptic observations. At lines 65- The project's intention-as articulated in the prologue-is to weave together disparate material, excised from existing prose geographies and histories, into a single poem. Accordingly, PS adopts comic trimeters, owing to their brevity and ease of memorization. 47 He identifies as the primary virtue of comedy (1 πάντων ἀναγκαιότατον ἡ κωμῳδία) its capacity to win over the sober-minded critic (4 ψυχαγωγεῖν πάντα τὸν ὑγιῆ κριτήν) with its concision and clarity (3 βραχέως … σαφῶς). He tells Nicomedes that he selected this metrical form after testing its persuasiveness (5 δοκιμάσας τὸ πιθανὸν τῆς λέξεως), so as to furnish his patron with a useful text for both personal study and the instruction of others (7-10). 48 He also paraphrases an analogy from Apollodorus' iambic Chronica at lines 37-44, concerning the capacity of verse to render complex material easily retainable: 49 ὥσπερ γὰρ εἴ τις ἀναλαβὼν θέλοι φέρειν ξύλων λελυμένων πλῆθος, οὐκ ἂν εὐχερῶς τούτων κρατήσαι, δεδεμένων δὲ ῥᾳδίως⋅ οὕτω λελυμένην λέξιν ἀναλαβεῖν ταχύ 40 οὐκ ἔστι, τῷ μέτρῳ δὲ περιειλημμένην ἔστιν κατασχεῖν εὐσκόπως καὶ πιστικῶς⋅ ἔχει γὰρ ἐπιτρέχουσαν ἐν ἑαυτῇ χάριν, ὅταν ἱστορία καὶ λέξις ἔμμετρος πλεκῇ.
For just as if somebody wished to pick up and carry a large quantity of loose sticks, he could not do so without difficulty, but, if they were bound together, would manage easily; so too is it impossible to retain a loose utterance quickly, but, if it is wrapped up in metre, one may 41  hold it unerringly and steadfastly. For it has a grace spreading through it, whenever historical enquiry and metrical language are woven together.
The logic of the Periodos is, therefore, clear: PS intends to collate the atomized geographical material littered throughout existing prose texts into a succinct poetic circuit of the world. Remarkably, this approach is paralleled in Dionysius' Description of Greece, in which the geographer presents his own work as a compressed metrical digest of existing prose sources: τὰ γὰρ ἐν πλείοσιν | ὑπὸ τῶν παλαιῶν συγγραφέων εἰρημένα, | ταῦτ' ἐμμέτρως ῥηθήσετ' ἐν βραχεῖ χρόνῳ. 50 Accordingly, it is extremely unlikely that PS would devote nine verses at the end of the prologue to a defence of personal autopsy absent elsewhere in the text, and lacking discernible utility for the project at hand.

THE PREFERABLE READING: TIMAEUS OF TAUROMENIUM
We have shown that the traditional reading-namely, that lines 128-36 of the Periodos refer to PS-is unjustified linguistically, and incongruous with PS' description of his own project. Rather, we propose that Timaeus, identified in line 126, is the intended subject of the autoptic section.
Of the few conceivable alternatives, Timaeus is the only plausible candidate. The names preceding Timaeus in the catalogue are too distant from line 128 to supply a natural referent. Herodotus (127) is also an unlikely candidate. Notwithstanding further syntactic difficulties (covered below), this reading depends on taking lines 127 and 128 together, rendering τῶν … συντεταγμένων the antecedent of ἅ, and thereby the object of ἐξητακώς in line 128; however, Herodotus cannot sensibly 'examine' or 'scrutinize' his own works. Moreover, the subsequent description of a historian of the Western Mediterranean and Sicily (135 καὶ τοὺς Σικελικοὺς καὶ πρὸς ἑσπέραν ὅρους) does not accurately reflect Herodotus.
Thus, the text most likely identifies Timaeus as the referent of the passage. This is corroborated by two further factors: lines 128-36 reflect the known method and scope of Timaeus' work, as well as his biography; and fundamental textual deficiencies affecting lines 125-8 may be sensibly resolved only by taking Timaeus as the subject of lines 127-36, and by emending the text accordingly.

Timaeus on Tour
We begin by considering the substance of lines 128-36.
Timaeus was clearly an important source for PS, who twice invokes the historian by name in the geography proper as a source for material on colonial foundations. At line 214, PS mentions Timaeus as an authority on the founding of the Phocaean colony at Massalia and, at line 412, he names both Timaeus and Eratosthenes as authorities on the colonies of the Hyllike Peninsula. This both justifies Miller's conjecture of Τίμαιον in line 126 and confirms Timaeus' importance to the Periodos.
Further, lines 128-36 constitute a fitting précis of the geographical reach of Timaeus' oeuvre. The western regions identified at lines 132-6 mirror the scope of Timaeus' Sicilian Histories, 51 which covered most of the Western Mediterranean, including Libya and Carthage, and regions as far west as Iberia. 52 The passage's topography might also reflect Timaeus' now-lost Wars of Pyrrhus, 53 which presumably treated the history of Pyrrhus of Epirus' campaigns against Rome and Carthage.
It is more difficult to establish a connection between Timaeus' Sicilian Histories and the travels east of the Ionian Sea described at lines 130-1: ὡς ὢν θεατὴς οὐ μόνον τῆς Ἑλλάδος | ἢ τῶν κατ' Ἀσίαν κειμένων πολισμάτων. While it is generally accepted that the Sicilian Histories did not cover regions beyond the Western Mediterranean-indeed, Polybius claims that Timaeus wrote only about Italy and Sicily 54 -there is reason to speculate about the reliability of this interpretation. For one, as Baron observes, Polybius' characterization of Timaeus' purview is influenced by polemical ambition, relegating his forerunner to the lesser status of local historian, and casting himself as the universal historian par excellence. 55 Moreover, the Suda does not credit Timaeus with a single Sicilian Histories, but rather mentions two titles, conventionally understood to represent complementary halves of his magnum opus: ἸταλικὰκαιΣ ικελικά(comprising either eight or thirty-eight books) and ἙλληνικὰκαὶΣικελικα( of unspecified length). 56 The Ἑλληνικάin the second title might therefore indicate that Timaeus' work included the geography of Greece. However, this is not determinative: as Beloch has shown, a Hellenica need not be limited by geography, but may concern Greek things wherever they occur. 57 In any case, there are three other titles attributed to Timaeus that might explain PS' reference to eastern travels: the aforementioned Wars of Pyrrhus (which may have covered Pyrrhus' campaigns in Macedon and the Peloponnese following the Battle of Beneventum); the Olympic Victors/Chronica Praxidica (which must have mentioned numerous Greek cities in its catalogue of stadion winners); 58 and the On Syria (whose alleged focus on cities accords with PS' remark about the cities of Asia at 131). 59 However, Timaeus may be identified as the subject of lines 130-1 without recourse to any lost works. At lines 130 and 132, PS deploys two participial expressions to describe the referent: ὢν θεατής (referring to his travels in Greece and Asia) and ἵστωρ δὲ γεγονώς (referring to his travels in the Western Mediterranean). Both qualify the autoptic status signalled in lines 128-9, but distinguish between the nature of the referent's observations in the east and the west. The opposition between θεατής and ἴστωρ here distinguishes two species of autopsy: while the former refers to one who sees something first-hand, the latter indicates the acquisition of specialist knowledge sufficient for making educated judgements. 60 PS thus identifies the passage's referent as somebody who has viewed both the east and the west, but attained the status of ἵστωρ only in the latter. 61 This is clearly applicable to Timaeus, the so-called 'Herodotus of the west' in PS' day, 62 who authored a text probably titled Σικελικαὶ Ἱστορίαι. Moreover, Timaeus resided in Athens after his exile from Sicily in c.315 B.C.E. 63 and, according to Polybius, travelled to meet the mainland Greek Locrians during this period: ἀλλ᾿ ἀληθινῶς αὐτὸς ἐπιβαλὼν εἰς τοὺς κατὰ τὴν Ἑλλάδα Λοκρούς … 64 We may then speculate that the source underpinning lines 130-1 was a reference in the Sicilian Histories (or another text) to time spent in Athens or travels further east. 65 One obstacle to the identification is Polybius' censure of Timaeus' autoptic prowess. In assailing his βιβλιακὴ ἕξις, 66 Polybius claims that Timaeus subverted the Heraclitan proverb, pursuing research by hearing rather than by seeing. 67 This criticism appears to have gained currency in subsequent centuries: indeed, the author of the Suda tells us that Timaeus was a known γραοσυλλέκτρια ('gossip-monger'). 68 However, following Baron's revisionist approach, 69 there is reason to suppose that this critique is distorted. No other ancient author who mentions Timaeus criticizes him for refraining from autopsy. Moreover, as others have suggested, 70 the exaggerated tone of Polybius' critique calls into question its objectivity. Thus Baron cautions against reliance on Polybius, 'given the polemical context in which most of his judgments are found, Polybius' rivalry with his predecessor as an authority on Rome, and the fact that no other source corroborates Timaeus' supposed lack of political experience or his bibliakē hexis'. 71 The surviving fragments of the Sicilian Histories show that travel and personal observation played a significant role in Timaeus' project. Several fragments dealing with specific locations show evidence of autoptic observation, especially those concerning Locris (F 12 = Polyb. 12.9.2), Acragas (F 26a = Diod. Sic. 13.80.5) and Lavinium (F 59 = Dion. Hal. Ant. Rom. 1.67.4). The last is especially telling: σχήματος δὲ καὶ μορφῆς αὐτῶν πέρι Τίμαιος μὲν ὁ συγγραφεὺς ὧδε ἀποφαίνεται⋅ κηρύκεια σιδηρᾶ καὶ χαλκᾶ καὶ κέραμον Τρῳκὸν εἶναι τὰ ἐν τοῖς ἀδύτοις τοῖς ἐν Λαουινίῳ κείμενα ἱερά, πυθέσθαι δὲ αὐτὸς ταῦτα παρὰ τῶν ἐπιχωρίων.
Concerning their [the Penates'] form and shape, the writer Timaeus states that the sacred objects lying in the innermost sanctuary at Lavinium are iron and bronze herald's wands and a Trojan ceramic vessel, and that he himself learned these things from the locals.
Timaeus thus claims that he has visited Lavinium, seen the place with his own eyes, and interviewed locals about the form and shape of the Penates. Moreover, Polybius indicates that Timaeus made the effort to see things for himself: at the close of Book 12, he paraphrases Timaeus' claim that he took great pains and incurred significant expense to collect information about the Ligurians, the Celts and the Iberians. 72 While Polybius deems this indicative of the fact that Timaeus merely compiled notes and made cursory enquiries (συνάγειν ὑπομνήματα καὶ πολυπραγμονεῖν), this censure is probably a product of his 'distorting lens' 73 rather than reflective of Timaeus' actual method: as Baron points out, 'the tenor of Timaeus' comment … would seem to indicate some sort of travel or at least attempts to obtain information from various sources'. 74 There is thus no reason to assume that Timaeus eschewed autopsy in favour of purely bookish research.
One final factor to consider is that, if Timaeus is the referent of lines 128-36, he is over-represented in the catalogue as compared to PS' other sources, who, in the extant text, are never afforded more than a single verse. However, this emphasis is perhaps expected, given the order of PS' itinerary. The geography proper begins at line 139 at the Pillars of Heracles, and from there moves eastward through the Western Mediterranean. PS thus begins the Periodos in precisely the region for which Timaeus would have been his primary source. Placing the extended description of Timaeus at the prologue's conclusion was likely a literary strategy, functioning as a pivot between the general list of authors and the technical geographical material of the poem proper.

Syntactic issues and a proposed emendation
Beyond the vagueness of line 128, the verses following the illegible text are clearly problematic. For ease of reference, we print again the relevant passage: The first is the τε in line 127: with what does the phrase ἐκ τῶν ὑφ' Ἡροδότου … συντεταγμένων coordinate? Even if taken as a single connective τε, 76 the conjunction must connect clauses of equal grammatical 'status': the posited conjunct must, at least, refer to another source of information, and likely form a prepositional phrase. No such conjunct is identifiable. Were Τίμαιον at line 126 the conjunct, the passage would give rise to an awkward syllepsis, with the putative preceding verb functioning transitively to govern the accusative noun, and intransitively to introduce the prepositional phrase. Any conjunct supplied in line 125 would be not only distant from line 127 but also incompatible with the known gaps, particularly given the identifiable instances of καί in MS line 7, themselves coordinating conjunctions. Rather, lines 125-6 must be read together, with the instances of καί joining Timaeus to preceding sources in the accusative (likely Καλλισθένη and another), all governed by the same verb. This is paralleled in lines 114-18, where the sources appear in the dative, relying on συμπεπεισμένος (114), and are separated by τε, καί or both. The required conjunct for the τε in line 127 is simply absent.
One dissimilar case of opposition poses no difficulty. 79 The relative pronoun (used demonstratively) + δέ at lines 545-6 lacks a preceding μέν: [πόλεις] ἃς δὲ καὶ συνοικίσαι | αὐτῶν. However, the sense ('the Cretans held the island-cities, and some of those they settled') justifies the omission: the implicit opposition (that is, that some cities were not settled) is trivial enough for an adversative δέ to convey the intended meaning, which arises by way of afterthought. 80 The ἃ δέ at line 128 does not warrant comparable analysis: the emphatic, verse-initial shift from a textual source in line 127 to new material treated by a distinctive autoptic methodology (128-36) indicates a substantive opposition, which-in line with the aforementioned parallelswould normally be balanced by μέν. We would thus expect τὰ μέν to precede the ἃ δέ in line 128. While supplying τὰ μέν in line 124 or earlier may be syntactically feasible, the collocation's placement in the catalogue is questionable: if the ἃ δέ in line 128 conveys an autoptic source, distinct from the preceding textual sources, there is no clear disjunct in the list from line 114 onwards to justify the oppositional μέν. Finally, there is no finite verb in the clause that extends from line 128 to line 136. Such extended omission is rare in Greek, and none of the typical justifications is applicable here. 81 The lengthy section is not parenthetical, and no finite verb may be readily supplied: the participles are sufficiently substantive, and the preceding list of sources sufficiently distinct, to leave a true semantic gap. Indeed, while PS does frequently deploy sequences of conjoined participial phrases, these always circumstantially qualify a simple finite verb, or else an εἰμί + participle periphrastic construction. 82 Neither is present in the extant text. As before, it is possible that the finite verb appears in the illegible section; Meineke, for example, proposes this as a solution. 83 However, this hypothesis is textually unrealistic: any supplement would need to sensibly govern both the putative μέν-clause in the catalogue of sources and, at a great distance, the autoptic method described at lines 128-36.
Taken cumulatively, these linguistic grounds-the syntactic non sequitur of the hanging τε in line 127; the absent first element of the opposition completed at line 128; and the atypical omission of a finite verb for more than ten verses-indicate a fundamental deficiency in the extant text. Therefore, we propose that the text be emended by inserting a verse between lines 126 and 127 (here denoted 126a). On this hypothesis, the verse, though integral to PS' own text, was lost before the thirteenth century, or in the production of Parisinus suppl. gr. 443. Bravo ([n. 14], 16) already recognizes that multiple verses may have been omitted here; however, he presumes that the reference to Timaeus marks the end of the catalogue, and that the omitted text concerns PS' other literary sources. 84 We suggest that, by the principle of parsimony, a single omitted verse is more likely to constitute the scribal error. Given the accurate representation of Timaeus in the following verses, and his presence in line 126, such a verse would make Timaeus the referent of what follows, and resolve the aforementioned deficiencies. This verse must contain the following five elements: 1) a nominative relative pronoun with antecedent Τίμαιον, introducing a relative clause encompassing lines 126a-36, with subject Timaeus; 2) a finite verb, with a meaning similar to 'to write material on [a topic]', governing all the following circumstantial participles (up to line 134); 3) a masculine singular participle, describing the act of providing information based on sources, and balancing the participles in the second half of the μὲν … δὲ … opposition; 4) the collocation τὰ μέν, in opposition to ἃ δέ (128), acting as the object or other complement of the balancing participle and perhaps of the finite verb; as ἃ δέ, being the object of ἐξητακώς, refers to other geographical topics/material, τὰ μέν must bear a similar referent; and 5) an adverbial complement of the verb, with the meaning 'based on [some author or source]', serving as the conjunct of line 127. To show that a single verse of Greek iambic trimeter can satisfy these restrictions, we tentatively compose the following verse: This verse aligns with the metre and language of the Periodos. PS commonly uses both ἱστορέω ('report on one's enquiries [into]') 85 and συγγράφω ('describe, write a work about') 86 in these senses. The collocation of ἱστορέω and συγγράφω, the use of the present tense to describe a past act of composition, 87 and even the verse-final συγγράφων are all paralleled in line 565 Ἡρόδοτος ἱστορεῖ δὲ ταῦτα συγγράφων. Syntactically, both verbs occur in the Periodos either with definite objects or absolutely; 88 thus, while τὰ μέν in line 126a serves as the direct object of συγγράφων, ἱστορεῖ may either be intransitive or transitively govern a null object, the definite referent of which is understood from τὰ μὲν … ἃ δὲ … This latter approach to argument sharing is regular throughout Classical authors, 89 and common in PS. 90 Finally, the preposition κατά + accusative bears the sense 'in accordance with', or 'following', in relation to a named source at lines 472-3: τοὺς … τόπους | ἐθνικῶς ἅπαντας κατ' Ἔφορον δηλώσομεν. 91 The collocation between transitive συγγράφω and the ἐκ of source (as in line 127) occurs in Dionysius of Halicarnassus. 92 Although this verse contains a medial caesura after the third princeps-contrary to the convention in tragic trimeters-this feature accords with the comic form explicitly adopted by PS. 93 To specify Timaeus' source here is speculation. However, the fourth-century historian Ephorus both satisfies the metre and aligns with the available data. Timaeus appears to have engaged with Ephorus' work, both substantively and methodologically, 94 and other texts routinely associate the two authors. 95 Indeed, Ephorus may have been the model for Timaeus' combination of chronological history and geographical excursuses. 96 Given this close relationship, it is feasible that PS deemed Ephorus a core intellectual predecessor to Timaeus, comparable to Herodotus, whose importance as a model for the Sicilian Histories is well established. 97 We thus propose the following text and translation for lines 125-9: … δὲ καί 125 Τίμαιον, ἄνδρα Σικελὸν ἐκ Ταυρομενίου, 126 <ὃς ἱστορεῖ τὰ μὲν κατ' Ἔφορον συγγράφων> 126a ἐκ τῶν ὑφ' Ἡροδότου τε συντεταγμένων, ἃ δ' αὐτὸς ἰδίᾳ φιλοπόνως ἐξητακώς αὐτοπτικὴν πίστιν τε προσενηνεγμένος … … and Timaeus, a Sicilian man from Tauromenium, <who composed a historical account, describing some matters in accordance with Ephorus> and based on those things composed by Herodotus, but diligently examining other matters personally and supplying the guarantee of eyewitness testimony … CONCLUSION A difficulty with conventional readings of the prologue of the Periodos is the absence of autopsy, from both the poem itself and the iambic didactic tradition. Reading lines 128-36 with PS as their referent ignores the bookish nature of the rest of the poem, and renders the author an autoptic observer, in a departure from the genre's methodological conventions. Our proposal resolves this dissonance, demonstrating that PS cannot be describing himself in these verses, and that this difficult text is most coherently read as a portrayal of the autoptic activities of Timaeus. This has two major implications. First, it removes any uncertainty about the nature of PS' project: like Apollodorus and Dionysius, PS was a poetic compiler, who selected and versified information excised from existing prose sources. Second, it renders PS a crucial witness to Timaeus' autoptic competence, and perhaps provides new evidence for his eastern travels. Ultimately, autopsy in the Periodos serves as a criterion by which PS judged-and called upon his readers to judge-the reliability of his prose sources.