596 HENRY SHALER WILLIAMS 



extinction of one and replacement by the other. ^ On such assump- 

 tion the correlation of formations by fossils becomes a complex 

 problem, for we cannot say that the last stage of the fauna of one 

 locality occurs at the same stratigraphic horizon as at another locality. 

 The range of this Nunda fauna (i. e., " Cardiola," " Portage," 

 '' Naples," "Manticoceras" fauna) has been shown to pass above the 

 Ithaca fauna in the Ithaca section; at Hornellsville it is seen in contact, 

 so to speak, with the Chemung, and traces of it also occur well up in 

 the zone dominated by Chemung fossils. The result is that to be 

 accurate in drawing formation lines on a paleontologic basis there 

 must be recognized a zone of transition of greater or less extent in 

 which either or both faunas may appear. The drawing of forma- 

 tional boundaries on lithologic basis is not more accurate; but the 

 error is more difficult to detect. The actual number of feet thickness 

 assigned to a particular formation in a local section often depends 

 more upon the positiveness of the assumption as to the actual 

 boundary stratum than upon the definiteness of the evidence of the 

 correlation. 



The passage from the Nunda fauna to the Ithaca fauna, from 

 Ithaca back to Nunda, and from Nunda to Chemung can generally 

 be detected within passage of a few feet of strata, the chief reason 

 being that the faunas thus brought into direct sequence are different 

 in their generic and, often, class composition. In the Nunda frail- 

 shelled Pelecypods, Gasterpods, Cephalopods, and Arthropods of 

 its lower orders dominate. In the faunas of both the Ithaca and 

 Chemung formations, Brachiopods, Bryozoa, and Lamellibranchs 

 of the Pectinoid type are the conspicuous forms. The assumption 

 is made that this difference between the Nunda fauna and either 

 the Ithaca or Chemung is intimately associated with environmental 

 conditions of their habitats, hence the contrasted faunas are described 

 as heterotopic in relation to each other. 



DijjicuUy in separating the Ithaca from the Chemung fauna. — 

 When the successive faunas are homeotopic, the difficulty of dis- 



I Henry S. Williams, "Shifting of Faunas as a Problem of Stratigraphic Geology, " 

 Bulletin of the Geological Society of America, Vol. XIV, pp. 177-90; "The Correla- 

 tion of Geological Faunas: A Contribution to Devonian Paleontology," and Bulletin 

 No. 210, U. S. Geological Survey, (1903). 



