lob BULLETIN UNITED STATES GEOLOGICAL SURVEY. [vm v. 



contain very few fossils. There are some exceptions to the latter rule, 

 but in all cases, so far as I am aware, when the rock approaches a true 

 dolomite, the contained fossils are all and of all kinds in the form of 

 moulds, the inner surfaces of which usually have a drusy lining. 



Casts of the more perfect kinds are formed by the infiltration of for- 

 eign mineral matter into complete moulds which were previously formed 

 in the manner already explained, and by its precipitation there until 

 the cavities became completely filled, ^uch casts are sometimes com- 

 posed of iron pyrite, some of which are often ve]::^- perfect and beautiful.- 

 More frequently, however, the casts are such as have been formed by 

 a filling of the spaces which the fossils formerly occupied by the sub- 

 stance of the rocky matrix in which they were imbedded, the fossils 

 themselves having gradually passed away by solution, while the sub- 

 stance of the adjacent matrix followed closely upon it in consequence of 

 the pressure of the superincumbent strata. It is in the form of such 

 casts as these alone that many fossils, especially the shells of several 

 families of Couchifers, have been preserved 5 and single valves of such, 

 shells are often very well preserved in that manner. 



The manner of preservation of invertebrate fossils wlich is here des- 

 ignated as fossilization proper consists of the minimum of change in 

 the mineral constituents of the skeletal parts. As already stated, some 

 change is always effected in the mineral composition of the object fos- 

 silized, but almost all molluscan as well as certain other shells, and also 

 corals, being composed almost wholly of lime-carbonate, when x)reserved 

 in limestone strata, or in such strata as contain any considerable amount 

 of lime-carbonate, still retain in the fossil state much the same compo- 

 sition which they had while living. In most of these cases, the change 

 has been so slight that even the microscopic texture of the shells and 

 tiie minute and intricate details of the corals are as perfectly preserved 

 as they were in the living state. Thus, the fibrous shell-structure of 

 certain families of Brachiopods, and the punctate structure of others, 

 the pearly shell-structure of certain Cephalopods, Gasteropods, and 

 Conchifers, and the prismatic shell-structure peculiar to certain families 

 of Couchifers, are often, indeed generally, as easily studied in the fossils 

 as they are in their living representatives. 



The siliceous skeletal parts of fossil invertebrates, when such were 

 produced, seem to have been so purely siliceous that they have appar- 

 ently not changed at all by fossilization. Chitinous substances, how- 

 ever, seem always to have undergone more or less alteration, although 

 they are among the more permanent of fossil forms. For the purposes 

 of this discussion, under tlie head of chitinous fossils are included the 

 sheUs of lyopomatous brachiopods (such as those of Lingula, Discina, &g.), 

 the crusts of trilobites and other crustaceans, and stipes of graptolites. 

 In-hmestones, chitinous fossils seem to have become quite calcareous in 

 some instances, and in all cases they are more brittle and mineral-like 

 than living chitine is, and yet in all cases they are readily recognized as 



