CONDITIONS OP PALEONTOLOGIO CORRELATION. 483 



a continuous terrane may vary in its constitution, in the fineness or coarse- 

 ness of its constituent particles, or in its composition, in the course of a few 

 miles' distance. Second, stratigraphic sequence is a reliable guide in corre- 

 lation when the individual strata of two corresponding sections are certainly 

 identified. But it is known that two separate sections through correspond- 

 ing parts of a terrane may vary considerably — gaps in one may be filled 

 by important strata in another, and strata thick in one section may be thin 

 and insignificant in another. The third means of identifying individual 

 strata, as well as general terraues, is by the contained fossils. 



Fossils, as well as mineral constitution, present local variations in strata 

 known to be continuous. Geologic correlation at its best is but approximately 

 correct wherever widely extended areas or separate districts are concerned, 

 because the means of correlation are not constant. 



Sequence, or order of succession, is the fundamental principle in all geo- 

 logic classifications of sedimentary deposits, but the two groups of criteria 

 (lithologic strata and organic species) whose succession is studied are of 

 different natures, and their variations are due to different causes. 



Each geologic stratum was originally a sedimentary deposit ; hence strata 

 vary with the differences in the original conditions of sedimentation and in 

 the source of the sediments deposited. In consequence, geologic time has little 

 or nothing to do with the lithologic character of the strata. A Cambrian 

 sandstone of one region may not differ essentially from a Tertiary sandstone 

 of another region, and the representative of a Cambrian sandstone of one 

 region may be expected to be a limestone in another. 



Since, then, the nature of the deposit must depend upon the local condi- 

 tion of the source of materials and upon conditions of sedimentation, there- 

 fore close similarity in the nature of sedimentation or in the sequence must 

 necessarily be more or less local, and correlation by this means will be less 

 and less reliable the more distant the two correlated sections are from each 

 other. 



Fossil species, on the other hand, are the remains of organisms which once 

 lived, and of living organisms we know that they are more or less depend- 

 ent upon conditions of environment ; that animals or plants are adapted to 

 air, land, fresh water, or salt water ; to differences of environment due to 

 differences of temperature, moisture, height, depth, etc. Faunas and floras 

 also differ, other things being equal, coordinate with geographical distribu- 

 tion ; and, most prominently of all, differences are seen in the faunas and 

 floras of each successive stage in the geologic history of the whole world. 



Successive strata, then, may contain (a) the faunas of successive ages, or 

 (6) the faunas of varying depths of ocean, or (c) the faunas whose geographic 

 distribution has shifted; and correlation by means of fossils is liable to 

 error from confusion of these causes of difference. 



