BABBAGE ON THE TEMPLE OF SERAPIS. 197 



56. Beginning at the floor, the first fact we arrive at is the dark 

 incrustation covering the walls of some of the chambers. It occurs 

 in those marked C 2, 23 C, and 29 C. 



Now this incrustation could not have arisen from water confined 

 to these chambers only, because two of them (C 2 and 29 C) are at 

 distant parts of the temple, and the height of the lines of the incrus- 

 tation are almost the same. 



57. Neither could this incrustation have arisen from the water of 

 the hot spring alone, for there are Serpulas attached to it ; unless it 

 be supposed that after its deposit the sea entered the temple, in which 

 case a larger quantity of remains of marine animals ought to have 

 been found. 



58. The incrustation could not have arisen from sea-water alone, 

 because the water of the Mediterranean does not, in that neighbour- 

 hood at least, leave any such deposit. To prove this, I examined an 

 ancient Roman house in very excellent preservation, which stands 

 partly in the sea in the immediate neighbourhood of Naples, and 

 well deserves more attention than I could bestow upon it. The 

 house consists of three stories and the basement. The walls are re- 

 markably thick, and the different stories separated by arches. 



59- The basement stands in the sea and is open to it, and may 

 have been used as a bath, a boat-house, or a reservoir for fish. I 

 entered it in a boat for the purpose of examining the wall, which 

 had probably been exposed to the action of the sea ever since the 

 building of the temple of Serapis ; but I found no calcareous deposit, 

 as the portion I have placed on the table will testify. 



60. This incrustation must have been made either before the 

 temple was ruined, or since it was cleared out. We know it has not 

 iiappened in the latter period. 



61. The reason for asserting that it must have occurred previous 

 to the destruction of the temple, is, that there is no indication of any 

 uneven termination at its lower boundary, as there is in the other 

 great incrustation, and as there would have been if the pavement 

 of the temple had been covered with the rubbish caused by its de- 

 struction. 



62. The conclusion to which these facts point, is, that at some 

 period after the temple had been built, and before it had been much 

 injured, the ground on which it stood gradually and slowly sub- 

 sided until its pavement became about 4^ feet below the level of 

 the sea. 



63. There was probably in ancient times a channel communicating 

 as there is at present, and thus the sea-water which entered became 

 diluted by the water from the hot spring within the temple. Thus 

 the hot spring supplied the calcareous matter, and the diluted sea- 

 water was still fit for the existence of the Serpulse. 



61*. It has already been remarked, that the lines at the lower edge 

 of this incrustation do not give any indications of an uneven bottom; 

 but as it does not, in the few places where it still remains, extend 

 to the floor of the temple, this fact is not conclusive as to the temple 

 not having been filled up to a certain extent previously to its deposit. 



VOL. III.~— PART I. p 



