Chap. 106.] WONDEUS OF rOTJKTAINS RIYEES. 133 
Surrentum, not only twigs which are immersed in it, but 
likewise leaves are petrified; the water at the same time 
being proper for drinking. In the stream which runs from 
the marsh of Keate^ there is a rock, which continues to 
increase in size, and in the Eed Sea olive-trees and green 
shrubs are produced^. 
There are many springs which are remarkable for their 
warmth. This is the case even among the ridges of the 
Alps^, and in the sea itself, between Italy and ^naria, as in 
the bay of Baise, and in the Liris and many other rivers^. 
There are many places in which fresh water may be procured 
from the sea, as at the Chelidonian Isles, and at Arados, and 
in the ocean at Grades. Grreen plants are produced in the 
warm springs of Padua, frogs in those of Pisa, and fish in 
those of Yetulonia in Etruria, which is not far from the sea. 
In Casinas there is a cold river called Scatebra, which in sum- 
mer is more full of water^. In this, as in the river Stymphalis, 
in Arcadia, small water-mice are produced. The fountain 
of Jupiter in Dodona, although it is as cold as ice, and 
extinguishes torches that are plunged into it, yet, if they be 
brought near it, it kindles them again^. This spring always 
becomes dry at noon, from which circumstance it is called 
In a subsequent part of the work, xxxi. 8, our author remarlis, 
" Reatinis tantum paludibus ungulas jumentorum indurari." We may 
presume that the water contamed some sahne, earthy or metallic substance, 
either in solution, or in a state of minute division, which would produce 
these effects. It does not appear that any tiling of this kind has been 
observed by the moderns in this water. 
2 The coral beds with which the Hed Sea abounds may have given 
rise to this opinion : see the remarks of Alexandre in loco. Hardouin 
informs us, that this clause respecting the Ked Sea is not found in any 
of the MSS. Lemaire, i. 441. A similar observation occurs m a subse- 
quent part of the work, xiii. 48. 
^ There are thermal springs in the Alpine valleys, but not any in the 
elevated parts of the Alps themselves. 
The volcanic natiu'e of a large portion of the south of Italy and the 
neighbouring islands may be regarded as the cause of the warm springs 
which are found there. 
^ This river may be supposed to have been principally suppUed by 
melted snow ; it would appear to be colder, because its temperature would 
be less elevated than the other streams in the neighbourhood. 
^ The statement, if correct, may be referred to the discharge of a quan- 
tity of inflammable gas from the surface of the water. The fact is men- 
tioned by Lucretius, vi. 879, 880, and by Mela. 
