If Pindar used any interaction between musical and accentual melos, there should be positions displaying not only the same metrical pattern throughout all the strophes and antistrophes (or the epodes) but as well the same accentual combinations. If we apply as criterion the contour-theory (with primary and secondary accents according to the three morae-rule), these « isomelies » are to be found usually at the end of a big unit, of a stanza or a period. Partial isomely constitutes an argument for partial (but not wholesome) repetition of the melody in the stanzas. Furthermore those passages allow us to test byzantine accentuation : some words could have been pronounced in a different way by Pindar. Uncomplete isomely seems to become complete if we apply « doric » accents. Which suggests that Pindar not only used some doric sounding vocabulary, but at least more often than not doric accents. This means that the accents of the papyri and the medieval manuscrips are due to an alexandrian policy of accentuating the three older chorographers according to « doric » practice, the younger ones with attic accents. Those philologoi would not have adopted this policy if they had had any traces of Pindar’s melos. Finally, Pindar’s longer isomelies seem to interact with thematic patterns.