wiluams.] CONCLUSIONS. 263 



inung helped to keep up the misinterpretations. Mr. Meek's success 

 in correlations appears to have been due to his minute knowledge of 

 the characters of fossils and their relations to each other, and perhaps 

 still more to his firm faith in fossils as the one reliable guide to true 

 correlation. 



The principle of "persistent parallelism of strata" is defective in 

 several ways: (1) Although it has been often observed that a stratum 

 continues for a long distance with but slight variation in thickness and 

 character of material, the constancy of lithologic and stratigraphic 

 character can not be assumed to be the case, even for short distances, 

 unless actually so observed. From this we deduce the law that "par- 

 allelism of strata" is not a safe means of correlation, although the cor- 

 relation once being established, the parallelism of strata is a valuable 

 aid in the recognition of the correlation for detached sections. (2) 

 The errors made by this method of correlation occur at points 

 where the evidence is lacking, therefore it is impossible by merely 

 going over the field a second time to correct such errors. (3) Even 

 when there is appareut continuity of a single stratum or of a series of 

 similarly formed strata, for tens or hundreds of miles, this alone is not 

 evidence that the deposits at the two extremes were formed synchro- 

 nously. The correct interpretation of the continuity, in case the 

 material is purely clastic, is more likely to be found in a gradual shift- 

 ing of the shore line by rising or sinking of the land than in synchro- 

 nism of deposition. On the other hand, the correlation of geologic 

 formation by their fossil contents is (1) Always made upon actual 

 evidence, any errors of interpretation of which can be corrected by 

 critical review of the evidence; (2) the particular form assumed by 

 any organic structure appears to be determined almost entirely by two 

 factors, i. e., heredity and environment; hence we may deduce the law 

 that, given the locality and the conditions of environment, the fossil has 

 in itself the evidence of its geologic age. 



The precision with which correlations may be made upon paleonto- 

 logic evidence is determined by the knowledge possessed of the relations 

 of the elements of organic form to geologic age, so that a fragment of a 

 fossil in the hands of one who knows how to interpret the evidence 

 may furnish a more correct diagnosis of the age of the formation than 

 a bushel of fossils in the hands of one ignorant of the laws of organic 

 life determining the form of the structures produced. 



The lowest member of the Mississippian series in Illinois having been 

 defined as the Kinderhook group, it was a matter of simple paleonto- 

 logic correlation to fix the lower limit in Iowa at the base of the 

 "Chemung group," in Missouri at the base of the formations later called 

 Chouteau group, in Indiana at base of the Goniatite beds, in Ohio at 

 the base of the Waverly, and in Michigan at the base of the Marshall 

 group. Immediately underlying these formations or their evident equiv- 



