Natural History in London. 187 



20; of the gavial and alligator, about 50; of the emys, 20; and of the tri- 

 onyx, 10. 



One fragment of emys is so large, that the animal of which it formed a 

 part must have been several feet in width. 



The state of preservation of these bones is very perfect, from their being 

 penetrated with hydrate of iron, and thereby rendered strong. Not one of 

 them is silicified, though they have been erroneously stated to be so, in 

 some of the periodical journals. 



The district in which they were found is a little north of the town of 

 Wetmasut, and is composed of barren sand-hills and beds of gravel, inter- 

 sected by ravines, and cemented occasionally into a breccia by carbonate of 

 lime, and sometimes by hydrate of iron. Over the surface of these hills 

 were scattered the fragments of bones and wood, some quite naked and 

 loose, others half buried in the sand and gravel. Many fragments of wood 

 lay also at the bottom of the ravines. About one third of the bones have 

 been slightly rolled, and the rest had all been broken before they were 

 lodged in the places where Mr. Crawfurd found them, and where the>' ap- 

 pear to have been dispersed and buried by the action of the same waters 

 that produced the diluvial sand and gravel ; whence they have since been 

 washed out, and left bare by the action of rains and torrents. 



Concretions of sand and gravel adhere to many of the bones, but they 

 contain no traces of shells, and differ mineralogically from all the rock 

 specimens in this collection, which we recognize as belonging to tertiary and 

 fresh-water strata. 



Indications of fresh-water formation were found in one spot only, not far 

 from the fossil bones, and they consist of a marly blue clay, abounding with 

 shells of a large and thick species of Cyrena. 



The tertiary rocks are, Ist. a dark slaty limestone, containing many shells, 

 that have been identified by Mr. Sowerby with those of the London clay ; 

 2d. a yellow sandy limestone, containing shells, and resembling the calcaire 

 grossier ; and 3d. a soft greenish sandstone, resembling the sandy beds of 

 our plastic clay formation. 



This London clay and calcaire grossier afford an additional locality of 

 these strata to those indicated by the specimens described by Mr. Cole- 

 brooke, in vol. i. part i. second series of the Geological Transactions, which 

 had already established the existence of this formation in the north-east 

 border of Bengal. 



Mr. Crawfurd states distinctly, that it is impossible to refer the situation 

 of the bones, or the origin of the hills containing them, to any operations of 

 the existing river. These hills are 60 ft. above the level of its highest flood. 

 The effect of its actual operations, he observes also, is distinctly visible in 

 the shifting islands of mud and sand that abound along the whole course of 

 the river within this high-flood level, and in the great alluvial delta that 

 extends from a little below Prome to Rangoon and the Gulf of Martaban. 



The recent bones and recent wood which he observed to be stranded on 

 some of these islands, were not in a state of progress towards becoming 

 mineralised, but were falling rapidly to decay. 



The existence of so many animal remains analogous to those that occur 

 in the diluvium of Europe, in a matrix which so nearly resembles that dili- 

 viura, and which so decidedly differs from the alluvium, and fresh-water, 

 and tertiary strata of the adjacent country, seems to authorise us to refer 

 this matrix to a similar diluvial deposit in the valley of the Irawadi, re- 

 posing irregularly upon the tertiary and other stratified rocks, that form the 

 basis of that district. 



Besides the tertiary strata above enumerated, thei'e are specimens of 

 grauwacke and transition limestone, from several distant points in the valley 

 of the Irawadi between Prome and Ava, which render it probable that the 

 fundamental rocks of this valley belong to the transition series. 



