No. V. — Temple of Jupiter Serapis. 271 



the temple, and which probably gave rise to the choice of the 

 site, and was conveyed by ducts into bathing-rooms already 

 mentioned. The temperature of the water in the reservoir I 

 found on the 24th of March 1827 to be 98°.5 Fahr. The 

 mineral matter is principally muriate of soda, with a little car- 

 bonate of soda and sulphate of lime. It is still recommended 

 for medicinal purposes, taken internally, and has even been 

 imitated in Naples. 



This short description will give an idea of the general po- 

 sition of the temple and its structure. We must now notice 

 the curious phenomena connected with it, which have given 

 rise to so much controversy. 



At the height of ten feet above the base of the three stand- 

 ing pillars, and in a position exactly corresponding in all, is a 

 zone of six feet in height where the marble has been injured 

 by the action of the well known shell-fish which live in cavities 

 pierced by themselves in the rocks they inhabit. To discuss 

 these minutely, or to theorize upon the method by which they 

 form their dwellings, would be quite beyond our present pur- 

 pose ; but I am happy to refer to the excellent paper of Mr 

 Stark in the Edinburgh Transactions, vol. x. for many curi- 

 ous details on the subject. I shall not fatigue the patience of 

 the reader by giving all the conflicting opinions of travellers 

 and naturalists on the animals which once inhabited the pillars 

 of the temple of Serapis, but content myself with quoting two 

 of the ablest observers of Italy, Spallanzani * and Pini -|-. 

 They agree that the shells here found are the Mytilus litho- 

 pliagus of Linnaeus, a bivalve shell, but not dissimilar in habits 

 to that of the Pholas^ a true multivalve, with which it has often 

 been confounded:};. By the powerful action of the valves of their 

 smooth shell, they have at some former time made holes to 

 the depth of four inches in the hard limestone of which these 

 pillars are composed, avoiding, however, the nodules of quartz 

 and felspar which sometimes occur. Several fragments of the 



* Travels, i. 84. 



•\ Memorie della Societa lialiana, S^c. 



i It may save some confusion to mention, that the whole genera of 

 stone-piercing animals are known in Italy by a variety of synonimous terras^ 

 *' Mangia-pietre," " Forapietre, " Datteri del Mare" (from their resem- 

 blance to the date fruit in form), " Foladi," " Litofagi." 



