Page:Essays and Addresses.djvu/470

 to the dramatic concentration which marked his art; the form which he was the first to make popular by excellent illustrations, and which he continued to prefer. As to the comparative prevalence of the two forms in the whole tragedy-literature of the fifth century, the evidence is too scanty to warrant any precise estimate. Towards the end of the century we meet with two certain instances of the Aeschylean form, the Pandionis of Philocles in 429, and the Oedipodeia of Meletus in 405. On the other hand, the relatively small number of such fable-trilogies which can safely be inferred from the extant documents, and the fact that in 340 B.C. the trilogical form itself had ceased to be imperative—as is shown by an inscription published in the Transactions of the German Institute at Athens for 1878—would lead us to believe that, after Aeschylus, the general tendency was in favour of the trilogy with unconnected plays. (3) Thirdly, we may observe that there seems no ground for an assumption which has been made, or implied, in some writings on this question—viz. that a trilogy would have appeared defective as a work of art if the three tragedies had not been in some way related to each other. We saw that Günther, while giving up the theme-trilogy, suggests that the author of three plays not linked by fable may still have studied some general effect of harmony or contrast between his pieces. The poet may, of course, have sometimes done so, and with good result; but it seems unlikely that either poet or