History of the Peloponnesian War
Thucydides
Thucydides, Vol. 1-4. Smith, Charles Foster, translator. London and Cambridge, MA: Heinemann and Harvard University Press, 1919-1923.
And it was determined that each sailor, taking his oar and cushion and oar-loop, should go on foot from Corinth to the sea on the Athenian side and hastening to Megara should launch from the docks at Nisaea forty slips of theirs which chanced to be there, and then sail straight for the Peiraeus.
For there was no fleet on guard in the harbour, nor was there any expectation that the enemy would ever suddenly attack it in this way, since they would not dare such a thing openly, and if they should plan it secretly they would not fail to be detected in time.
But once they had determined upon the scheme they set to work immediately. Reaching Nisaea at night they launched the ships and sailed, not now to the Peiraeus as they had intended, since they were appalled by the risk—and a wind, too, is said to have prevented them—but to the promontory of Salamis that looks towards Megara. There was a fort here and a guard of three ships to prevent anything from entering or leaving the harbour of the Megarians. This fort they assaulted, towed away the triremes without their crews, and ravaged the rest of Salamis, falling on the inhabitants unawares.
Meanwhile fire-signals indicating a hostile attack were flashed to Athens, where a panic was caused as great as any in this war.[*](This must refer to the so-called Decelean War (or last ten years of the Peloponnesian War), for in Thuc. 8.96.1 we read that a panic occurred greater than any before (τοῖς Ἀθηναίοις ... ἔκπληξις μεγίστη δὴ τῶν πρὶν παρέστη).)For the inhabitants of the city thought that the enemy had already entered the Peiraeus, and those of the Peiraeus that they had taken Salamis and were all but sailing into their own harbour-as indeed might easily have happened if the enemy had resolved that there should be no flinching;