BABliAGE ON THE TEMPLE OF SERAPIS. 199 



66. It should be noticed, that the portion of the incrustations 

 attached to the standing columns and to the fallen fragments, although 

 the same in chemical composition, is not quite so hard, and is in a 

 different state of aggregation from that on the walls ; also that no 

 remains of Serpulse or other sea shells have been noticed on it. The 

 upper edge of this incrustation is level ; the lower edge is irregular, 

 as it would have been if the bottom of the temple had been 

 filled up. 



67. The conclusion from these facts is, that after the temple had 

 subsided to a small extent its whole area was filled, either from a 

 shower of volcanic ashes, or from some other cause, up to a certain 

 height, apparently leaving the centre most filled up — that the same 

 cause probably closed up the channel communicating with the sea — 

 that in consequence of this, the water from the hot spring, having no 

 outlet, filled the temple to such a height, that the waste from leak- 

 age and evaporation equalled the supply. 



68. Admitting this explanation, the temple could have suffered 

 little from decay previously to this event. Some of the marble 

 panelling may have fallen down, as one of the specimens seems to 

 prove. But the greater part of the columns must have been standing, 

 because their fragments are surrounded by zones of various breadths 

 arising from this incrustation. 



69. From considering this fact, we are enabled to restore many 

 of them to the positions which they must have occupied. Thus the 

 fragment described in par. 8, and marked (4) on the plan and in the 

 table, must have been a portion of one of the large columns standing 

 on its base during the period this incrustation was forming, be- 

 cause — 



ft. in. 



Height of top of incrustation on column. 6 4 



Height to the base of the shaft of the large column . . 2 4* 



8 8 



The sum of these gives very nearly the average height of the top 

 of this incrustation above the pavement of the temple, which was 

 about 8 feet 10 inches or 9 feet. 



70. Again, if we extract from the synopsis the two instances in 

 which the lower portions of smaller Cipolino columns have marks 

 of incrustation, we find from — 



ft. in. 



No. 15 3 11 



No. 23 4 2 



Average .... 4 0^ 



Add to this, height of the base, from No. 33 1 4 



Height of central part of temple, from No. 33 3 7 



W9t iS ;. 8 lly 



which is nearly the height of the great incrustation. 



p2 



