walcott. J CORRELATION. 433 



correlation. He considers that " the method involves a yet broader 

 conspectus of phenomena and principles than the paleontologic method; 

 for in its application it is necessary mentally to restore the various 

 physical and biotic conditions of the past, just as paleontology vivifies 

 the fossils of past ages." Again: 4 



In discriminating the general and local genetic conditions it is necessary to ascer- 

 tain the relations between each formation and its newer and older neighbors, and to 

 interpret the record of each unconformity in terms of continent growth. By this 

 means the different parts of a formation may be found to represent not only general 

 community of genesis but community of beginning and ending ; in short, entire com- 

 munity of structural relation. Each part of the formation then records in similar 

 terms the same episode in continent building and world growth. 



So, when a coastal plain formation is found to represent general community of 

 genesis and structural relation in its various parts it is considered homogenic, and 

 accepted as a record of an episode in geologic history. The parts may or may not be 

 homotaxial ; one part may be slightly older than another part ; but in a general way 

 it is contemporaneous throughout. Iloruogeny implies not only equivalence but 

 synchrony. 



The principles of homogenic correlation have long been used in the 

 correlations of formations by geologists. An illustration of this is the 

 correlations that have been made of the Potsdam sandstones of New 

 York with those of the Mississippi Valley, the eastern slope of the 

 Rocky Mountains, the islands of Texas and Missouri, and the valley of 

 the St. Lawrence. In these correlations to subjacent and superjacent 

 formations, stratigraphic continuity, lithologic or petrographic resem- 

 blance, and similarity of organic remains have all been used. 



Under the definition of homogeny, that is, correlation by community of 

 genesis, decided errors must necessarily occur unless checked by paleon- 

 tologic data. An illustration of this is the correlation of the Potsdam 

 sandstone of the Adirondack Mountains of New York with the " Granu- 

 lar Quartz" of western New England and southeastern Pennsylvania, 

 and the Ohilhowee sandstone of Tennessee. These sandstones are litho- 

 logically alike, they were accumulated along a coast line, their petro- 

 graphic characters are the same, and the action of the waves upon and 

 the distribution by tidal currents of the sediments originating from 

 similar crystalline rocks all point to a community of origin ; iu fact, 

 they did originate in the same manner. They have been correlated 

 upon the characters mentioned for upwards of 40 years, but the study 

 of the fossil remains shows they are not equivalent or synchronous. 

 One represents the basal beds of the Cambrian system and the other 

 the Potsdam sandstone or the closing deposits of Cambrian time in the 

 same area. I think that after we have fully proven to our satisfaction 

 the similarity of origin or the community of genesis we must still rely 

 upon paleontologic data for the final determination of geologic equiva- 

 lence. For the determination of synchrony, except in a limited area, 

 there is little hope for satisfactory conclusions by any method yet de- 

 vised. 



4 Op. cit., pp. 38, 39. 



Ball. 81 28 



