488 E. O. ULRICH CORRELATION OF THE STRAND-LINE 



This instance is discussed in considerable detail because it affords i. 

 good illustration of the incompetence of faunal correlations that take into 

 account only the described species, and at 'that without pretending to 

 identify them with absolute certainty. It also illustrates the general 

 similarity of successive faunal invasions from the same oceanic realm in 

 contrast to the obvious distinctness of directly superposed fossil faunas 

 that invaded from different oceanic realms. 



Proper use of Fossils in Correlation 



In many — but why not confess the fact and say in most— instances of 

 revised opinion regarding the range of fossil species and genera the sug- 

 gestion, and often the proof, came first through detailed stratigraphic 

 investigations. These put the paleontologist on the defensive and often 

 forced the admission that fossils previously identified as well known 

 species are in fact easily distinguishable. I am glad to say such humili- 

 ation is less likely now than heretofore, because, as the burnt child shuns 

 the flr.e, so may we escape by closer discrimination of fossils at the outset. 



And therein lies the salvation and the final justification of correlation 

 by means of fossils. But remember, the function of the fossils is to 

 identify horizons and not to decide how the division of geologic time into 

 epochs and periods is to be carried out. 



The only kind of fossil evidence that has a definite and unimpeachable 

 correlation value is that afforded by the identification of relatively trivial 

 biological differentiations. In the very nature of things these, be they 

 called varieties or mutations, must be trustworthy indices of contem- 

 poraneity. If the basic principle of the idea is apparent, then the reader 

 will see at once also the truth of the following qualifying or at least 

 kindred principle, namely, the inherent con-elation value of a fossil species 

 is in proportion to its cotiiplexity of structure, being greatest when the 

 complexity is in parts which are structurally unessential. 



On the contrary, simply constructed organisms, with few characters 

 that might be preserved in the fossil state — as, for instance, a shell with 

 rounded, low conical form and smooth or but simply marked surface — 

 these are of little value, because such types usually persisted through long 

 ages, and their numerous successive occurrences are often so much alike 

 that the human perceptions are incapable of distinguishing them. 



The differentiations regarded as particularly valuable are such as per- 

 tain to the surface marking or ornament of shells of mollusks or brachio- 

 pods; granulation and peculiarities of venation of the test of trilobites 



