1865.] LAXKESTER — ^CRAG STAMMALIA. 223 



negative certainty when the character and preservation of the Mam- 

 malian fossils is observed, and the conditions of their occurrence 

 noted. 



The Crag is, as is well known, a very loose sandy deposit, con- 

 taining a very unusual amount of oxide of iron, which stains it of a 

 deep-red or orange colour ; all the shells of MoUusca which it con- 

 tains, and which are not known to be derived from previous deposits, 

 are excessively^ fragile, disintegrated, and merely superficially stained 

 by iron, the action of which has tended to render them fragile and 

 pulverulent, rather than to mineralize or strengthen them, whilst 

 the shells and other fossUs which are acknowledged to be washed 

 from preceding beds have considerable tenacity, are very heavy, and 

 deeply impregnated with mineral matter. As examples of these 

 two classes of fossils the Tellince, Mactrce, and Fusi may be quoted 

 on the one hand, and the derived specimens of Terebratuhi spondy- 

 loides, Voluta Lamberti, and numerous Crustacean and Piscine re- 

 mains on the other. The action, then, of the Red Crag on its 

 py^oper fossils appears to be a fragilizing and destructive one ; its 

 action on derived fossils an indurating and preservative one. This 

 action may doubtless be satisfactorily explained by the laws of 

 chemical affinity. The oxide of iron, which so deeply colours the 

 Red Crag, existed in the sea as carbonate of the protoxide. It is 

 well known that with recent organic matters, which are necessarily 

 decomposing, this salt of iron forms a sulphide of the metal, whilst 

 water and carbonic anhydride are liberated. The unfossilized shells, 

 bones, <fec., therefore, in the Red Crag sea were not subjected to 

 the action of the protoxide of iron at aU, the sulphide of iron being 

 formed, which rapidly decomposes and destroys the structure it 

 invests. 



Phosphatic nodules, however, and bones and shells thoroughly 

 deprived of their organic matter by previous fossilization, do not 

 act similarly on the carbonate of the protoxide of iron. It infil- 

 trates them, and thoroughly mineralizes them, eventually becoming 

 deposited vrithin their structure as peroxide, silicate, or phosphate, 

 as the case may be, part of their constituents being dissolved and 

 removed by the carbonic acid. 



The question here involved, although a chemical one, is of very 

 great importance to the palaeontologist ; and the investigation of the 

 processes of fossilization cannot but afford the inquirer important 

 results. Much and very valuable assistance has already been fur- 

 nished to this branch of inquiry by Dr. Bischof in his work on Che- 

 mical Geology. It appears that no fossils in any loose sandy deposit 

 attain that degree of mineralization and firmness observed in most 

 bones and teeth from the Red Crag, until they have been subjected 

 for the second time to the action of water containing mineral matter, 

 either by the submergence of the Avhole deposit or by the separate 

 action on individual specimens. 



The fossils of the Middle Crag of Antwerp, of the Darmstadt 

 Miocene, and of those beds of our own Tertiaries which are sandy and 

 not argillaceous are all very friable, light, and slightly mineralized. 



VOL. XJI. — PART I. R 



