More famous than either of these difficulties is one which occurs in Pausanias’s account of Athens. Here in the middle of describing the market-place, which lay to the north-west of the Acropolis, he suddenly without a word of warning transports the reader to the Enneacrunus fountain, which lay in the bed of the Ilissus, not far from the Olympieum, at the opposite extremity of the city; then, having despatched the fountain and some buildings in its neighbourhood, he whirls the reader back to the market-place, and proceeds with his description of it as if nothing had happened. Of the many attempts to clear up this mystery, as by supposing either a dislocation of the text or a confusion in the author’s notes or the existence of another fountain near the market-place which may have been shown to him as the Enneacrunus, none is free from serious difficulties. That he fell into error through copying blindly and unintelligently from a book is possible but very improbable. As it is practically certain that he visited Athens and saw both the market-place and the Olympieum, the chances that he should not have seen the Enneacrunus and should therefore have been driven to borrow his description of it from a book are so small that they may be neglected. [Sidenote: Law-courts at Athens and altars at Olympia.] Other passages which Pausanias may perhaps have taken either wholly or in part from books are his account of the Athenian law-courts and his list of the altars at Olympia. Neither of these passages, it is true, is demonstrably infected by error or confusion, though there is some ground for suspecting the existence of confusion in the enumeration of the altars. But in both of them the author departs from the topographical order of description, which is so characteristic of his method, and arranges the monuments together simply on the ground of their belonging to the same class. These departures from his usual principle of order suggest that in both cases Pausanias may have borrowed from written documents in which the monuments were grouped together according to kind rather than in topographical order. Another set of monuments which Pausanias links together by a chain other than the topographical are the buildings erected by Hadrian in Athens. It is possible that he may have taken his list of them from the inscription in the Athenian Pantheon which recorded them all. These are perhaps the most notable passages in Pausanias, which might be thought to bear traces of having been derived either wholly or in part from written documents rather than from personal observation. In none of them are the indications so clear as to amount to a proof of borrowing. At most they raise a probability of it, nothing more. [Sidenote: Predecessors of Pausanias.]