Thucydides has given us four examples of sanctuaries outside the _polis_ which are ‘towards that part’ of it, but again, as in the first clause, he seems to feel that if he has spoken the truth it is not the whole truth, so he saves himself from misunderstanding by an additional clause, ‘_and other ancient sanctuaries are placed here_.’ It would be idle to try and give a complete list of all the sanctuaries that were situated in this particular region, still more idle to decide of what particular sanctuaries Thucydides was thinking. The precinct of Aglauros and the Anakeion on the North side, the sanctuary of the Semnae and the Amyneion on the West, the sanctuary of Aphrodite Pandemos and that of Themis on the West and South-West are all ‘towards’ the approach. Three out of these, the Amyneion, the sanctuary of the Semnae, and the sanctuary of Aphrodite Pandemos, are of such interest in themselves and so essential to the forming of a picture of the sanctities of ancient Athens that a word must be said of each. * * * * * _The Amyneion._ The Amyneion, or sanctuary of Amynos[231], is known to us only through monumental evidence, brought to light in the recent excavations. Its discovery is one of the things that make us feel suddenly how much of popular faith we, relying as we must almost wholly on literature, may have utterly lost. [Illustration: FIG. 30.] If after leaving the precinct of Dionysos in-the-Marshes we follow the main road for about 35 metres, we come on a precinct (Fig. 30) of much smaller size and of quadrangular shape, which abuts on the road and along the North side of which a narrow foot-path leads up to the Acropolis. The precinct-walls are of hard blue calcareous stone from the Acropolis and neighbouring hills, and the masonry is good polygonal. The entrance-gate (A), which has been rebuilt in Roman times, is at the North-West corner. A little to the East of the middle of the precinct, and manifestly of great importance, is a well (B). The natural supply of this well was reinforced by a conduit-pipe, which leads direct into it from the great water-course of Peisistratos, which will later (p. 119) be described. Near the well are remains of a small hero-chapel, and within this was found the lower part of a marble sacrificial table (C), decorated with two snakes. The masonry of the precinct wall, the well, and the shrine all point to a date at the time of Peisistratos. Even before the limits of this precinct were fairly made out the excavators came upon a number of fragments of votive offerings of a familiar type. Such are reliefs representing parts of the human body, breasts and the like, votive snakes, and reliefs representing worshippers approaching a god of the usual Asklepios type. Conspicuous among these was a fine well-preserved relief (Fig. 31), depicting a man holding a huge leg, very clearly marked with a varicose vein, exactly where, doctors say, a varicose vein should be.