[879] The existence of these cola forms (to us who have not the music written for Greek lyrics) one of the greatest obstacles to a clear and easy perception of periodic structure. [880] In lyrics a long syllable (if it does not end with a consonant) may be shortened—instead of disappearing by elision—before a vowel. [881] ἡ ἐπῳδός. The masculine word, ὁ ἐπῳδός, has a different meaning, with which we are familiar from the _Epodes_ of Horace—a poem which repeats from beginning to end the same period, each period being usually two cola “which either have equal length, or the second of which is catalectic or ‘falling’ or is even shortened by an entire measure” (see Schmidt’s _Introduction_, Eng. tr. by Prof. J. W. White, pp. 93 _sqq._). [882] Though my obligations to Dr. J. H. H. Schmidt’s volumes, especially _Die Eurhythmie in den Chorgesängen der Griechen_, are very great, I cannot see in his verse-pause—according to him (_Eurhythmie_, p. 89) the foundation of his system—anything but a delusion. Dr. Schmidt’s own appendices show a good minority of “verses” which end with no pause. [883] The first two syllables (⏑⏑) correspond to the first (–) of Οἰδίπου. [884] How? By examination of the whole period. If we look at the seventh line of the strophe from _Antigone_, scanned above, it may seem arbitrary to write – ˃ | αιναν ‖ rather than ⏗ – | αιν|ανꞈ‖. But the former method is suggested by the corresponding fourth line, which cannot possibly be scanned otherwise than as above, and which therefore has four feet; hence we scan -αιναν so as to give the seventh line also four, not five, feet altogether. [885] It is therefore possible to scan the ordinary iambics of dialogue as trochees:— ˃ – ⏑ – ˃ – ⏑ – ˃ – ⏑ – ειθ ⁝ ωφελ | Αργους | μη δι|απτασθ|αι σκαφ|οςꞈ (_Medea_, 1). This is the method followed by Dr. J. H. H. Schmidt, and of course changes altogether the rules given above (§ II), but will hardly perplex the student. It has the advantage of bringing “iambic” dialogue closer to lyric and to episodic trochees, but it has seemed more convenient to keep the traditional statement. [886] Printed as one line, though containing a colon which ends with the end of a word, because the corresponding line of the antistrophe contains a colon which does not:— πρῶτά σε κεκλόμενος, θύγατ‖ερ Διός, ἄμβροτ’ Ἀθάνα.... [887] Because spondaic _words_ are lacking. It is sometimes said that the only spondee in English is “amen”. The peculiar pronunciation of this word is due to the fact that it is so often sung to music where each syllable is given a whole bar. The name of Seaford in Sussex is undoubtedly pronounced by its inhabitants ∸∸; but one may perhaps therefore argue that it should be written “Sea Ford”. [888] This important sequence may be conveniently memorized—if we substitute accent for quantity—by the sentence “Attack Rome at once”.