108 WRITINGS OF JAMES BMITHSON. 



their aid on the occasion, would have disseminated the 

 clustered animals, and dispersed the powdery stratum. 



That the bodies should in every case have descended into 

 the calcareous pulp, in one unbroken group ; that in none 

 a fragment, even a lock of hair, should have parted from 

 the putrid mass, and stopped bj' the way, cannot certainly 

 plead probability in its favour. Yet what cabinet shows 

 even the slenderest bone of a water-rat bedded in the solid 

 stone ? What limestone stratum has astonished the learned, 

 by presenting them, in its substance, with an antediluvian 

 hyaena's bristles, or lion's mane ? 



Formation of the Gave. 



If the limestone pulp was too thin, the gas would pass 

 through it and escape, and the pulp full back upon the 

 bodies ; if too thick, the elastic force of the gas would be 

 insufficient to repel it from them. A precise point of indu- 

 ration, at which it would at once yield and resist, was indis- 

 pensable. This exact condition would but rarely occur; 

 would, at least, often not do it, and consequently bodies 

 buried in the solid rock must be frequent, if not most so. 



It is incredible that in every case the gas should have 

 driven away from the bodies the whole of the mud in con- 

 tact with them. Some of the mud must have insinuated 

 itself between the several individuals of the cluster, some 

 have penetrated by the mouth, by lacerations, into the cavity 

 of the bodies, and isolated pieces of rock must now occur 

 among the bones, bearing the impression of the parts with 

 which thoy had been in contact ; as at Pompeii, indurated 

 ashes presented the cast of a woman's breasts. 



As the parts receded from the bodies, it would carry with 

 it some adhering fragments of them — bones, teeth, hair, 

 feathers ; and which would now be fixed to the sides and 

 roofs of the caves. 



Bodies which had been previously putrefying for twelve 

 months in a tropical temperature, would not probably have 



