THESEUM, ERECHTHEUM, AND OTHER WORKS. 173 Pausanias' remark might very well refer to the tiling only, for here at Bassae the marble tiles were, Cockerell says, finer even than those at the Parthenon. At ^gina, again, marble tiles were only used at the eaves and up the verges, and Pau- sanias himself, when describing Olympia, remarks that the tiles were not clay, but marble. Moreover, the marble ceiling of the portico at Bassae, with a bearing of 1 3 feet, was especially fine and it may have been this also, which, as Cockerell says, excited the admiration of Pausanias. We are thus thrown back on the evidence of the structure itself. Choisy, who accepts the open cella, says that an internal cymatium was found by the French expedition to the Morea, but the cymatium in question seems to be that of the external pediments, and the writer of the text appears to be rather opposed to the idea of an open cella. Cockerell was in favour of a roof with a sky-light opening in it, and brought into evidence a small fragment of a roof tile with a margin which showed that it edged an opening. The size of this, however, is uncertain, and Papworth brought forward some evidence to show that the opening was only in the middle of a tile of the ordinary size, or rather that there was niore than one.* Such openings might have been mere smoke vents in the adytum, serving a similar purpose to the palm-tree chimney at the Erechtheum, or there may have been a light in the slope of the roof opposite to the statue for its illumination. From Haller's account it appears that only two small fragments of this kind were found, and they were discovered in the adytum. Papworth further suggested that the rafters might have been marble, but their size and the manner of fixing hardly allows of this interpretation, for the outer parts of the roof over the recesses at least, and Cockerell speaks of " the ascertained structure of the wooden roof" of the portico. Much has been made of Cockerell's showing an interior view Fig. 171. — From Athens. R.I.B.A. Joicrjial, 1865, and Anderson and Spiers.
