The Battle of Drepana in 249 b.c. marks the most significant defeat of Roman naval forces at the hands of their Carthaginian opponents during the First Punic War. Attempting to take the Punic fleet in the harbour of Drepana by surprise, the consul P. Claudius Pulcher sailed with his ships from Lilybaeum (a Carthaginian stronghold under Roman siege since the previous year) about midnight, and reached Drepana at dawn. Yet, owing to swift and level-headed counter-measures taken by the Punic commander, Adherbal, the unfolding fight – partly in the harbour, mostly off the shore – turned into a fiasco for the Romans.Footnote ¹ The consul got away; he returned to Rome, where the Senate instructed him to appoint a dictator.Footnote ² How soon he returned, and by what route, is the question.

Frontinus (Strat. 2.13.9) has a story about how, after the rout, Claudius escaped with his remaining ships: compelled to force his way through the enemy's positions, he ordered his remaining 20 ships to be decked out as if coming from a victory, and the Carthaginians, supposing that the Romans had won the battle, in terror allowed him to pass:

The ruse has been called ‘fantastic’,Footnote ³ and rightly so, if taken to describe a breakout through the Punic forces in the course of the battle itself. Nor will it help to postulate that, after his victory, Adherbal sent some of his ships to intercept the fleeing Romans and block their escape route: for this, the Carthaginians would first have had to overtake them as they ran south from Drepana along the coast (thus Polyb. 1.51.11: ὥρμησε πρὸς φυγήν, ἀπὸ τῶν εὐωνύμων παρὰ τὴν γῆν ἐξελίξας), then execute a 180-degree turn to face north and engage them. There is no room in Polybius’ account for such a manoeuvre, and one has trouble visualizing how the consul, in the confusion and panic of the rout, could have got all his surviving vessels – not just his flagship – to adopt that ruse in unison, and in time to be effective.

And how could the men in Adherbal's ships fail to notice that they were winning, and the Romans turning tail, or that they had just crushed the Roman fleet, whose retreating remnants they were chasing down, trying to prevent them from getting away? This might be conceivable – barely – if Claudius had led his fleet from the van and, together with the 30 ships ‘in his vicinity’ (καὶ σὺν αὐτῷ περὶ τριάκοντα νῆες, αἵπερ ἔτυχον ἐγγὺς οὖσαι, Polyb. 1.51.11) that escaped, been forced to fight his way out from north to south along the entire length of the battle zone, with the Punic vessels at its southern end not certain how things had turned out up north. But Polybius tells us explicitly (1.50.5–6) that the consul in his flagship had been bringing up the rear and had taken his position on the left (that is, southern) flank of the Roman line, bows facing out to sea and sterns towards the shore, while Adherbal manoeuvred into position exactly opposite him, on the extreme right (again, southern) flank of the Punic line.Footnote ⁴ To make good their escape, the Romans had to turn left – just as Polybius says – and run to the south, where there were no more enemy ships. However, no amount of flying victory garlands could have made the Carthaginians opposing them – including Adherbal himself! – or any from that same fleet sent after them believe that Claudius had won the battle.

As with many of the stratagems compiled by Frontinus, neither the provenance nor the historical authenticity of Claudius’ ruse is susceptible to proof. Livy and Caesar are known purveyors of exempla, and mining of an earlier collection can be assumed with near certainty.Footnote ⁵ Claudian family lore could be surmised, although, given the relentless blackening of Pulcher's reputation in virtually all surviving references to the man (save, notably, Polybius’), it is difficult to imagine how this episode would have been accepted into the wider tradition without some other authority to lend it credence. Errors of fact (mostly names and places) are not unheard of in Frontinus’ collection, and the anecdotal character of many a stratagem is obvious enough; but one is hard-pressed to find another example of a ruse that, within the apparent context of its being employed, makes no sense at all. As long as it is assumed that the praesidia through which Claudius needed to pass were Punic ships that had taken part in the Battle of Drepana, the stratagem described here is plainly unworkable: it seems fanciful, and unbelievable. Yet, before rejecting it as a mere fiction, we ought to examine whether the battle is, indeed, what Frontinus had in mind. Perhaps the context is not what it seems; and perhaps another can be found that makes the story plausible.

Frontinus organized Book 2 of the Strategemata in two sections: ea quae in ipso proelio agi solent, et deinde ea quae post proelium (2 pr. 1), corresponding, respectively, to 2.1–8 and 2.9–13. Claudius’ ruse appears in the second group, quae post proelium agenda sunt (2 pr. 3), in chapter 13, de effugiendo. Consideration of its placement alone should alert us to the possibility that it was not employed during the actual battle.

Of eleven stratagems listed under de effugiendo, five clearly represent manoeuvres not designed to escape an ongoing fight but ones initiated after combat has ended, and separated from the rout by a space of hours or days:

Hence, in nearly half of the instances in this section, a significant interval occurred between the stratagem and the combat operations that ultimately prompted it. What of Claudius Pulcher's?

Praesidium occurs 31 times in the collection. On about half of the occasions, it denotes a garrison or occupation force;Footnote ⁸ five times, an escort or other form of protection;Footnote ⁹ four times, a force guarding a harbour or other installation;Footnote ¹⁰ and twice each, reinforcements (2.9.2; 3.14.1) or a force holding a position as part of ongoing operations (1.4.9a, 10.3). The common denominator in all cases is one of guarding, holding, or protecting. In no instance – setting aside the one under discussion – does the word denote a force actually engaged in battle.

In over forty references to warships, Frontinus employs praesidium on just three other occasions:

The ships in (6) and (7) are acting in a defensive manner, blocking access to a port or shoreline: they pose no parallel to a hypothetical attempt to prevent the remnants of an already defeated enemy from leaving the battle zone. Praesidium here clearly represents, albeit in a particular setting, the basic protective sense of the word: there is no suggestion of battle. In (8), it simply means an escort.

Only once besides the passage in question is praesidium combined with erumpere:

The story is well known from Livy. During the night after Cannae, a few of the 6,000 or so survivors that had sought refuge in the smaller camp broke out and made their way to the larger camp, and from there with many others to Canusium and safety; although surrounded by Hannibal's army, the Roman camps were not yet actively besieged, and the only attempt to interfere with the escape of those in the smaller camp came from a Numidian cavalry patrol.Footnote ¹¹ Hence praesidia here denotes an obstacle to escape, but hardly a concerted effort to block such a move: per hos, qui inordinati atque incompositi obstrepunt portis, erumpamus, as Livy's Sempronius put it (22.50.8).

Frontinus’ wording is nearly identical to the Drepana passage (cum per hostium praesidia necesse haberet erumpere: per hostium praesidia erumperent secum); what about the context? The Romans at Cannae are not fighting their way out of a losing or lost battle, against enemy forces intent on preventing their escaping the battle field: they attempt to make their way to permanent safety a considerable time after the battle has ended, from a location of temporary safety out of the actual battle zone, as a conscious and planned effort, and against enemy patrols that present danger if encountered, but not a force specifically tasked with blocking their escape.

Besides the two passages already discussed, erumpere (or eruptio, vel sim.) is found on 21 occasions. Ten times it signifies a sortie made at the siege of a town or other fortified place;Footnote ¹² five times, an attack launched from a fortified place against an advancing enemy;Footnote ¹³ twice, a sudden manoeuvre during a battle (2.3.10, 12.1). In another two instances (1.5.10, 5.14 = 4.5.9), the word describes the breakout of a force surrounded by the enemy in difficult terrain; twice more in the same situation it specifically refers to the penetration or avoidance of enemy patrols and pickets (stationes).

As in (9), the situations in (10) and (11) involve a force not engaged in battle but prevented from moving freely by the enemy surrounding it; successful breakout requires getting past patrols and pickets, not a full-scale encounter with the enemy. Clearly, Frontinus can use praesidia as a synonym for stationes in such cases (as also in [6], above). We now have four instances of nearly identical wording (cum per hostium praesidia necesse haberet erumpere; per hostium praesidia erumperent secum; per stationes hostium perrupit/erupit; per hostium stationes eruperunt). Three of them feature nearly identical scenarios and may therefore offer the key to the fourth: the ruse that Frontinus ascribes to Claudius after Drepana.

The Punic praesidia in this scenario cannot have known the outcome of that battle: hence they were not ships in Adherbal's fleet. It follows that Frontinus was not describing the consul's escape from the actual battle; indeed, his entire usage of praesidium and erumpere militates against such an attribution. If we but allow praesidia its natural and common meaning in that author, we have the answer: naval squadrons protecting Carthaginian strongholds along the coast of Sicily (or on neighbouring islands) that could intercept Claudius and his remaining ships. There cannot have been many such locations by 249, besides Drepana itself. There were no Punic warships in the vicinity of Lilybaeum; otherwise, the attack on Drepana could never have entailed an element of surprise. The only attested naval base in Sicily itself was at Heracleia Minoa, in Carthaginian hands since before the war.Footnote ¹⁵ Some fortified positions may have existed between Drepana and Panormus – chiefly Heircte with its harbour.Footnote ¹⁶ And the Aegates Islands were still Punic.

Frontinus reports another stratagem, similar to the one he ascribes to Claudius, that allowed an inferior force to pass unmolested in front of an enemy fleet guarding the shore. Fearing a Punic fleet lined up along the beach near Syracuse, the Spartans dressed up ten captured Carthaginian ships as if victorious, and took their own vessels in tow behind them; the enemy fell for the deception:

The terms praesidia and stationes do not appear here, but given the parallels noted above ([6], [7], [9]–[11]), either could without difficulty be applied to the Punic force that the Spartans were trying to evade. As with Claudius’, the ruse relies on the enemy's ignorance of recent events; we have no grounds for dismissing it as fiction without concrete indication to that effect.

One more thing follows: the ruse must have been employed before word of Drepana could spread among the various Carthaginian naval stations – no more than a few days after the battle. A measure of clarity may be obtained by closely examining the situation that Claudius faced after escaping the cauldron off Drepana. The consul had got away by running south from Drepana along the coast (Polyb. 1.51.11). Conceivably, he could have turned west once out of sight of the enemy, then north, and passed Drepana at a safe distance, to continue along the north coast to Panormus and Messana. But, in a state of headlong flight, Lilybaeum appears the most likely destination, and Orosius says outright that this is where he went (Lilybaeum in castra confugit, 4.10.3). How many ships in all did Claudius have left by now?

He had sailed for Drepana παντὶ τῷ στόλῳ (Polyb. 1.49.3), with the entire fleet. Polybius eventually notes that ‘about thirty ships’ escaped, while the rest (τῶν δὲ λοιπῶν), 93 in number, were captured (1.51.11–12). Both figures are frequently taken to add up to the Roman total,Footnote ¹⁸ despite Polybius’ reference to numerous ships being rammed and sunk during the battle (1.51.7). The excerptor of Diodorus gives an overall strength of 210, of which 117 were lost (24.1.5); by implication, 93 would have survived. That figure is common to all sources stating the Roman losses; either Diodorus or, more likely, his excerptor somehow garbled its significance, mistaking it for the number escaped instead of captured.Footnote ¹⁹ Evidently Diodorus’ source (whether Philinus or some other) had, like everyone else, 93 ships captured – hence, presumably, about thirty escaped, and about ninety sunk. The sources likewise agree, it would seem, on the number of ships that escaped – about thirty.Footnote ²⁰

The Romans had launched the siege of Lilybaeum the previous year, with a fleet given as numbering 200 (Polyb. 1.41.3; Oros. 4.10.2) or 240 (Diod. Sic. 24.1.1) warships. As Polybius mentions no reduction in strength prior to Drepana, it would be rational to assume that he still took this to be the size when Claudius sailed ‘with the entire fleet’; if so, his numbers sunk (by implication), captured and escaped would be in essential agreement with Diodorus. However, one of the consuls of 250 had returned home early with his army (Zonar. 8.15); if the troops were transported by sea, as appears to have been the prevailing practice, this would require a substantial part of the fleet at Lilybaeum: hence Claudius may indeed have had only 120 ships available.Footnote ²¹ (Polybius omits the consul's departure, and thus might easily have missed an occasion to note the reduction in fleet strength.) The Diodorus excerpt has Claudius pick the 210 ‘best’ ships, which may imply that up to 30 were left behind. If so, the reason is right there (‘best’): those ships were not seaworthy or combat-ready (and, quite possibly, Claudius lacked crews for themFootnote ²²). There can be no question of his leaving a strong force to cover Lilybaeum: the strike against Drepana was to be a swift affair (Polyb. 1.49.5), and to ensure success he must concentrate his forces.Footnote ²³ In any case, the only Punic fleet that could move on Lilybaeum was Adherbal's, at Drepana.

At the moment of Claudius’ return, then, the Roman ships at Lilybaeum numbered at least 30 and at most 60; but of the latter, barely half – the ones that had escaped the rout – were fit for service. Immediately after the battle, the trierarch Hannibal son of Hamilcar (who had led the previous year's daring mission to run reinforcements into Lilybaeum, Polyb. 1.44) with 30 ships raided the Roman supply depot at Panormus and delivered much of its contents to the Punic garrison at Lilybaeum (Diod. Sic. 24.1.6): the Roman naval forces there apparently made no attempt to stop him. A short time later, the new Carthaginian commander, Carthalo, attacked the Roman vessels at Lilybaeum with 100 (or 120) ships, sank or otherwise destroyed a few, and dragged off five.Footnote ²⁴ From both Polybius’ and Diodorus’ description of the action it is evident that none of the Roman ships were manned and fighting back; the sole resistance came from the land forces rushing to the shore in an attempt to save the vessels drawn up there. No hint here that the consul was present and directing the Roman response, in contrast to Himilco's quick-witted sortie that kept the enemy from concentrating on the events along the shore. Clearly, Carthalo's raid was a complete success, and, just as clearly, he found not many Roman ships at Lilybaeum. Were the 30 survivors of Drepana among them? So it has often been assumed,Footnote ²⁵ even though plainly the evidence implies nothing of the sort. And nothing suggests that the consul was still there.

A plausible context for the stratagem now emerges. If Claudius had need to get past Punic praesidia within days of his defeat, he must have been attempting to leave Lilybaeum. His destination is not hard to guess: the ruse gives us the opening movements in the consul's journey back to Rome.

Why would Claudius go to Rome by sea at all, and so soon after the battle? We have little direct information on the preferred means of transportation for Roman commanders between Rome and Sicily during the First War. For those in command of fleets, enough evidence exists to conclude that they usually went by sea, and a good case has been made for the view that, after 260, Roman land forces were sent in like manner.Footnote ²⁶ Polybius notes that the 10,000 fresh rowers sent to Lilybaeum in 249 were marched there overland after being ferried across the straits (1.49.3); the unique detail suggests that he did not consider it a routine operation. As for Claudius’ return, this in itself proves nothing; but it militates against the assumption that he must have travelled by land.

From a cryptic note in Polybius it can be inferred that the ships that had survived the Battle of Drepana had reached Messana before the consul's colleague, L. Iunius Pullus, arrived there on his way to escort a huge convoy – some 800 cargo vessels – carrying supplies to Lilybaeum, via Syracuse: having set out from Rome with 60 warships, Iunius ‘took under his command those of the ships that had gathered there from the camp and from the rest of Sicily’ (ἀφικόμενος εἰς τὴν Μεσσήνην καὶ προσλαβὼν τὰ συνηντηκότα τῶν πλοίων ἀπό τε τοῦ στρατοπέδου καὶ τῆς ἄλλης Σικελίας, 1.52.5–6).Footnote ²⁷ The ‘ships from the camp’ have caused undue headache. Everywhere else in this part of Polybius’ narrative (1.42–55), τὸ στρατόπεδον invariably refers to the Roman camp and army besieging Lilybaeum: it cannot possibly mean something different – and entirely unknown – here.Footnote ²⁸ Yet it causes difficulty only when approached from a preconceived notion that, if Lilybaeum is meant, the ships must have left it before Drepana, as no Roman ships could have made it from Lilybaeum to Messana after the battle, with Adherbal's (and soon, Carthalo's) fleet blocking the way.Footnote ²⁹ Of course, nothing prevented them from simply sailing along the southern coast of Sicily, and past Syracuse to Messana. The ‘ships from the camp’, then, can only be the survivors of Drepana, transferred to Messana before Iunius’ arrival. There are no grounds for doubting, and plenty for believing, that Claudius had sailed with them.

Conditions at Lilybaeum now were unenviable, to say the least. The Roman siege works had been so thoroughly destroyed during the previous winter that all effort at an active siege by assault, with engines, towers, ramps, and so forth, had been abandoned long before Claudius took command, and was never fully resumed: from now on, the siege proceeded by a simple circumvallation designed to cut off Lilybaeum from all landward communications. Polybius is quite clear about that (1.48.9–11). At the same time, there was no threat of a Carthaginian offensive by land: Adherbal at Drepana lacked an army, and, while Himilco's garrison had shown that it could take care of itself, it was not equipped to take on the Romans in a pitched battle. On that count, the consul's continued presence was not required; his military tribunes could handle what remained of the siege.

Things looked no better on the naval side. The thirty or so ships left in combat-ready condition could not possibly hope to survive, let alone ward off a Carthaginian strike at Lilybaeum by sea, the enemy fleet now numbering upwards of 170 ships.Footnote ³⁰ That strike – Carthalo's – came soon enough; what matters here is that it was entirely predictable, as was the fact that any Roman vessel still at Lilybaeum at that moment would probably have been lost for good. On that count, too, the consul was not needed to preside over the pending elimination of the remnants of his fleet. There was, however, an alternative. The ships would be of infinitely greater use in Iunius’ fleet than here at Lilybaeum. To try to take as many of them to safety as was feasible, before the Carthaginians attacked, made eminently good sense, as did the consul's decision to sail with them: there was nothing left for him to do at Lilybaeum.Footnote ³¹ Nor could he indefinitely postpone reporting on the disaster, painful though it was, to the Senate. With luck, he might persuade them to build more ships; and he could carry home at least one prize: the magnificent fast quinquereme of Hannibal ‘the Rhodian’. That storied vessel later served as the model for the new fleet built in 242 (Polyb. 1.59.8) that would, on 10 March 241, destroy Carthage's last fleet at the Aegates Islands, and thus win the war.Footnote ³² At some point, it must have been brought to Rome. When, and by whom?

Polybius tells the story of the Punic blockade-runner known as Hannibal ‘the Rhodian’ (1.46.4–47.10) in a long excursus inserted among his narrative of the siege of Lilybaeum, between Himilco's first assault (1.45) and the eventual firing of the siege works during a gale (1.48). The latter clearly occurred in the consular year 250; hence one would naturally conclude that the commander who put an end to the Rhodian's enterprising voyages was one of the consuls of that year. Zonaras (8.15), however, not only mentions the Rhodian (as ‘Hanno’) after introducing the consuls of 249, but expressly attributes his capture to Claudius Pulcher. Was he mistaken? (He does, after all, get Hannibal's name wrong.)Footnote ³³

In Polybius, the – unnamed – Roman consul who first encountered the Rhodian on his initial run took an active part in the failed attempt to prevent the latter from leaving Lilybaeum again (1.46.8–12); it is safe to conclude that a name was attached to the consul's person in Polybius’ (and Dio's/Zonaras’) source, be it directly in this context, or on an earlier occasion, when that Roman commander was introduced. Three scenarios thus remain: (a) the name attached in our surviving authorities’ source(s) was that of Claudius’ predecessor, either C. Atilius Regulus or L. Manlius Vulso, and the Rhodian's eventual capture took place under the same commander; (b) the name attached was that of Claudius’ predecessor, Atilius or Manlius, but the Rhodian's capture occurred under his successor, P. Claudius Pulcher; and (c) the name attached was that of Claudius throughout.

Now the Romans, in an effort to stop the blockade-running, had earlier tried to bar the entrance to the harbour of Lilybaeum: in vain, as most of the material used to block it was soon scattered by the current (1.47.4). In one area, however, a small sandbar eventually accumulated: a Punic quadrireme ran aground on it, was seized by the Romans, and was used to overtake and capture the Rhodian (1.47.5–9). Polybius notes these efforts at obstruction only here, in connection with individual blockade-runners; but it is evident from Diodorus (24.1.1–2) that an initial attempt at shutting off the access to the harbour had been undertaken right at the beginning of the siege, and another after the arrival of the first Carthaginian relief expedition, led, it seems, by Adherbal.Footnote ³⁴ In both instances, the current quickly defeated the effort.

Claudius’ first action on assuming command was to attempt the operation for a third time, with the same lack of success as had been encountered before (καὶ παραλαβὼν τὰς δυνάμεις πάλιν τὸν λιμένα ἔχωσεν, ὥσπερ καὶ οἱ πρὸ αὐτοῦ, καὶ πάλιν ἡ θάλασσα κατεσκόρπισε, Diod. Sic. 24.1.5). Polybius states clearly enough that, with the Rhodian's capture, the Romans faced no further problems from Punic blockade-runners (οὕτως ἐκώλυσαν τοὺς κατατολμῶντας καὶ πλέοντας εἰς τὸ Λιλύβαιον, 1.47.10) in the period up to the Battle of Drepana, since they now had in their possession two ships capable of intercepting anything afloat: the Rhodian's quinquereme and the even faster quadrireme that had caught him. In consequence, the information from Diodorus confirms scenario (b), for Claudius’ attempt at barring the harbour entrance can only be the one reported, again without a name attached, by Polybius – in other words, the one that resulted in the capture of the quadrireme, and in turn of the Rhodian.

From the very first, the Roman fleet as a whole, even when still at full strength, had proved singularly ineffectual at preventing the supply of the Punic garrison by sea (witness its embarrassing performance during Hannibal the trierarch's run in 250: Polyb. 1.44). The capture of Hannibal ‘the Rhodian’ and the two fast ships it netted had put an end at least to individual Punic blockade-running, and Claudius Pulcher could fairly claim the credit for this feat. But within a few months the catastrophe at Drepana rendered moot his one success: henceforth, the shattered Roman naval forces in western Sicily were in no shape, with fast quinqueremes and faster quadriremes or without, to stop the Carthaginians from supplying, by sea and openly – and in fleet strength – their garrison at Lilybaeum. Had the Romans still been able to interdict such efforts after Drepana, Lilybaeum could never have outlasted the siege of more than seven years that was to come. As Polybius makes abundantly clear, Carthage now (and even more so after Iunius’ shipwreck off Cape Pachynus) held complete control of the sea, and used it to maintain her forces in Sicily – at Lilybaeum, Drepana, Eryx and Heircte – even after most of her fleet was laid up by 244.Footnote ³⁵ When Lutatius Catulus arrived in 242, he first seized the harbour at Drepana and the anchorages at Lilybaeum (Polyb. 1.59.9): evidently, no Roman vessels had been keeping watch on them.

As with the other Roman warships left at Lilybaeum, so with the Rhodian's quinquereme: it was of no further use on this station. Against a supply in force, such as Hannibal's a few days after Drepana, a single ship was powerless, and Lilybaeum henceforth need not rely on individual blockade-runners, the Romans having lost the means to maintain a blockade. To leave a ship as valuable as this behind and exposed to enemy action would make no sense. If the quinquereme had still been there during Carthalo's raid, it surely would have shared the fate of those dragged off, or been destroyed if recapture proved impossible: it was too well known in Carthage to be overlooked. It did not spend the next seven years at Lilybaeum chasing blockade-runners, to be brought to Rome only when the new fleet was being built. No doubt the consul took it with him when he left for Rome.

Frontinus (Strat. 2.13.9) gives Claudius 20 ships for the journey past hostium praesidia. The number may be just an error for the 30 he led back from Drepana;Footnote ³⁶ but once it is understood that the stratagem does not describe Claudius’ escape from battle but his movements in the aftermath, other explanations of the discrepancy become possible. A handful of usable ships – perhaps 10 or so – may have been desired for dispatch duties; more likely, if Frontinus’ number is in fact correct, Claudius now wanted his ships fully manned, and could assemble complete crews for no more than 20.

Thus Claudius sailed from Lilybaeum with some 20 or 30 ships. The safest route would have been along the south coast. One wonders, though: why take the ships all the way to Messana, seeing that Iunius was going to put in at Syracuse anyway? Why the stratagem in Frontinus? It strongly suggests that Claudius had to pass a superior Punic force; but if any enemy ships were stationed at Heracleia at that moment, they could hardly have outnumbered his. Valerius Maximus apparently placed his defeat at the Aegates: this may be mere confusion for Drepana, but it is also possible that he had read that name somewhere in connection with Claudius – we have no way of telling which.Footnote ³⁷

If time was of the essence, the northern route became the obvious choice: once – if – he reached Panormus, the consul in his flagship could sail directly to Rome, via the Aeolian Isles, and send the rest of his force on to Messana; or he could order a vessel to carry dispatches from Panormus to Rome, and wait with the others at Messana for instructions from the Senate. It appears that Claudius took the northern route, and the risk that came with it. (Whatever his faults, timidity was not one of them.)

The ruse described in Frontinus could not, of course, get him safely past Adherbal's fleet at Drepana: those Carthaginians knew who had won the battle. He must keep well clear of the coast – which meant a course through the Aegates Islands, or even west around them. How many Punic ships were stationed there we do not know, but more than 20 would be unsurprising; in fact, Carthalo with his 70 may already have arrived, waiting for a fair wind to cross to Drepana. Having 70 transports to protect,Footnote ³⁸ the sight of Claudius’ ships decked out as if in victory could well have made him hesitate.

Nor would the ruse deceive the Punic squadrons at the Aegates (or at Heircte or Heracleia, for that matter) once word of the battle had got to them: two days or three at most. We may assume that Claudius left Lilybaeum the day after his defeat at Drepana. The news of what had happened would have been in Rome within a week.Footnote ³⁹ According to Livy Per. 19, the Senate recalled Claudius, which would imply that word of the disaster (whether through the consul's report or some other means) had arrived in Rome before he left Sicily; if so, he probably sent a dispatch from Panormus and proceeded with the other ships to Messana. The Rhodian's quinquereme, in fact, would have been the perfect choice to carry the news to Rome: the sight of the great prize might have softened the blow.