Page:Folk-lore - A Quarterly Review Volumes 32 and 33.djvu/659

 Museums and Rcwce Shows in A^itiqiiity. 349

chopped off his head with an axe. Therefore the image is headless.

" I saw another Triton among the marvels of Rome, but it was not so big as the one at Tanagra. The appearance of the Tritons is this. On their heads they have hair which resembles the hair of marsh frogs both in hue and in this, that you cannot separate one hair from another. The rest of their body bristles with fine scales like those of a shark. They have gills under their ears and a human nose, but their mouth is wider and their teeth are those of a beast. Their eyes, I think, are blue, and they have hands, fingers and nails like the shells of mussels. Under their breast and belly, instead of feet, they have a tail like a dolphin's."

Sir James Frazer comments on these passages : "It seems it was not an image, but a real Triton or what was exhibited as such. For in the next chapter Pausanias says that he saw another Triton at Rome, describes the appear- ance of the supposed creature, and then gives a list of other strange animals. That the Triton at Tanagra was pro. fessedly a real animal embalmed or stuffed, appears from a statement of Demostratus, reported by Aelian,^ that he had seen at Tanagra an embalmed or pickled Triton ; the creature resembles the pictures and images of Tritons, except that the head was decayed with time and no longer distinct or recognisable ; and when he touched it, some hard rough scales fell off. A Roman senator, in the presence of Demo- stratus, took a piece of the beast's skin and burned it, as an experiment, it emitted a fetid odour ; but the spectators could not decide from the smell whether the creature was a sea or land animal. From Demostratus' description and Pausanias' story it would seem that the Triton of Tanagra was headless. What was shown as a Triton may have been either a real sea beast of some sort or an effigy made up by the priests. Had it been merely an effigy, it would probably have been complete, since it is just as easy to make a false

1 TTfpt s'wwi', xiii. 21.