STRATIGRAPHIC CLASSIFICATION PALEONTOLOGIC CRITERIA 499 



returned only after long intervals of local or general absence in a basin 

 or province. As for the new occurrences, which constitute the principal 

 element of difference by which the paleontologist distinguishes the suc- 

 cessive faunules: these have very often no genetic relationship to their 

 predecessors in the same locality or basin, even when there is no appre- 

 ciable change in sea-bottom or other environmental features. Evidently 

 the new forms came in "ready made." 



That the faunas of the continental seas were not subject to stimulated 

 modification, either local or general, within the seas themselves, as they 

 should have been if expansion of favorable habitat had been an important 

 factor in their evolution, seems readily demonstrable. If it had been then 

 the successive faunules in a single or in contiguous formations must have 

 been developed out of each other. In fact, however, this is true of only a 

 small percentage of the species, and those of which this may be said are 

 usually derivatives of the vigorous, long-lived species that are dominant 

 through considerable vertical and geographic ranges and which therefore 

 are correspondingly inexact in their time relations. On account of their 

 wide range in time and space the successive appearances of mutations of 

 such species is as readily accounted for under the view of frequent, dis- 

 tinct invasions or replenishments as under the supposition of continuous 

 existence within the continental basins. 



The determination of the direction and extent of marine currents 

 within the continental basins has an important bearing on this problem. 

 Convincing evidence is presented by the distribution of certain bottom- 

 dwelling organisms which depend on currents for the expansion of their 

 geographic range. The corals and bryozoa probably offer the most com- 

 petent data. As most of these assume sessile habits in maturity but are 

 free and subject to transportation by shore currents in their larval 

 stages it is at once suggested that their range in the continental basins 

 must be limited by the extent of the marine current to which they owe 

 their occasional migrations from the oceanic to the continental seas. 

 Many bryozoan and coral faunas are known in the Ordovician, Silurian, 

 and Devonian rocks of southern North America. Most of these invaded 

 the continental seas from the south through the Mississippi embayment, 

 and in a number of instances the path and extent of the several in- 

 vasions has been satisfactorily determined. In all cases the numerical 

 representation of the two classes of fossils, both as to species and in- 

 dividuals, is greatest at the localities nearest the point of ingress. Be- 

 yond these localities their number decreases sometimes rapidly, at other 

 times slowly, the layers in which they occur become fewer, and the per- 

 centage of species found at the more inland stations that are unknown 



