in substance a duplicate of the south series. But we may assume thus much to begin with on the analogy of the two corresponding long sides of the frieze which, as we shall see, were practically duplicates. From the evidence of the existing metopes and from the Centaurs in D'Ortières' drawings, we argue further that the place of the Centaurs had been inverted in the north series, they occupying the centre, while the marriage party occupied the two ends. On the north side the stormy element would be in the middle, and the placid element at the two ends.

Let us now see how far this view is corroborated by the existing metopes of the north side-that is to say, how far they represent groups of Lapith women such as Carrey drew in the middle of the south series.1 No. 1 on the extreme left represents a biga with female charioteer (Pl. XII.), precisely as metope 15 towards the middle of the south side. As we have before said, the presence of a chariot in the Centauromachia is attested by the Phigaleian frieze. On the north side 25 (Pl. XII.), with two women beside an archaic image or xoanon, corresponds to 21 on the south with two women beside a xoanon. The grouping is not the same in both instances. Yet in each case one of the women places her hand on the head of the image. There is a similar archaic xoanon in the Centauromachia of the Phigaleian frieze, with one woman clasping it and another turning away. Apparently this had been an essential feature of the legend. Again, in the north metopes we

1 Michaelis, Parthenon, p. 138, says: "Probably in the middle were a number of Centaur scenes which had inter

rupted the order of the other representations, similarly as on the south side do the metopes 13-20 in the centre."

have three separate instances of a man and a woman, he expressing alarm and bent on protecting her, 3, 27, 28. Answering to this, we have in the south metopes a group of a young man similarly alarmed beside a young woman 14, and something nearly approaching the same subject in 13 and 17.

We cannot, of course, claim that the whole of the nine central metopes as drawn by Carrey on the south side reappear at the ends of the north side, as we might expect. But in at least two of the cases of identity which we have pointed out it will be allowed that the subjects represented are remarkably characteristic of a Centauromachia. We are not obliged to assume that the same stage of the Centauromachia was presented on the two sides. On the contrary, we can well imagine an earlier stage of the incident on the north side than on the south. That would involve a certain number of differences in the action and in the grouping. It might explain why there are more groups consisting of a young man and young woman in the north than in the south metopes. Above all, it gives a reason why the three Centaurs drawn for D'Ortières have no Lapith opponents. These Centaurs would be rushing into the fray. In a word, our argument is that the coincidences between the north and south metopes are sufficient to justify the opinion that the same subject of a Centauromachia had covered both, but that the scheme of arrangement was in the one case an inversion of the other. In these matters no one has shown greater discrimination than Professor Petersen, who says1:—“If, therefore, on both long sides (north

1 Kunst des Pheidias, p. 230.

and south) practically the same subjects were represented, only with this alteration, that greater elaboration was bestowed now on the one side, now on the other, the intention could only have been to convey to the spectator of either side the idea of the whole, and thus spare him the necessity of going round to the other side. The same intention is as clear as possible in the arrangement of the frieze."

We are bound to notice here a difficulty presented by 29 (Pl. XII.) of the north side, with a horse stumbling forward apparently the rider turned right round on his back, like one of the Amazons on the Mausoleum frieze. We confess our inability to reconcile that subject with the battle of Centaurs and Lapiths. So also in 25 (Pl. XII.), otherwise perfectly consistent with a Centauromachia, as we have pointed out, we cannot explain the presence of a diminutive winged figure above the shoulder of the woman on the left.1 So minute a figure can hardly be Eros, as Michaelis confidently supposed. It is more like a Shade or eidolon, such as we see on Athenian funeral lekythi, suggestive of death. Possibly that is its meaning here also.

The last of the series of north metopes 32 is the only one which has been well preserved. But we have already described it in some detail (Pl. XI. p. 60), and will now

1 This small figure had been overlooked till Laborde had a cast made of it and drew it in the Revue Archéologique, 1845, pp. 16, 17.

Parthenon, p. 139. He claims that metopes 24 and 25 represent a consecutive scene from the Iliupersis, the xoanon being the image of Athenè,

the two women beside it Helena and Aphroditè, accompanied by Eros, the warrior in the preceding metope 24 Menelaos. The Greek vase which he reproduces seems to him to confirm his view. But there are too many women about in these north metopes for us to admit scenes from the Trojan war.