254 



SCIENCE 



[N. S. Vol. XXXI. No. 790 



Given any species (or kind) in any region, the 

 nearest related species (or kind) is not to be 

 found in the same region, nor in a remote region, 

 but in a neighboring district separated from the 

 first by a barrier of some sort. 



This law, worked out by observation of 

 existing faunas and based on their distri- 

 bution in our highly diversified lands, owes 

 its recognition to the fact that topography 

 and climate have undergone great changes, 

 and provincial environments have been 

 individualized during the latest geologic 

 periods. 



The latest period to which we can as- 

 sign fairly uniform conditions of climate 

 and moderate relief in North America is 

 the Miocene, and the diversity of environ- 

 ments developed since then is so great that 

 there is reason for surprise at the persist- 

 ence of geminate species. One might ex- 

 pect differentiation to a degree which 

 would have obscured or obliterated twin- 

 ship. But it appears that there are prov- 

 inces in which variations of some ancestral 

 species have not diverged greatly, presum- 

 ably because conditions within these par- 

 ticular provinces have not undergone any 

 very stimulating or very restrictive change, 

 as regards those species. Such surviving 

 varieties must indeed have existed to a 

 greater or less extent during any such 

 period of changing environments, and the 

 persistence of geminate species must have 

 been a feature of many epochs of dias- 

 trophic activity in the past. How long 

 they may have persisted, how slowly or 

 rapidly or impulsively they may have 

 varied, we do not know. The time rela- 

 tions of geminate species are therefore in- 

 determinate. 



CORRELATION 



Definition. — By correlation in paleogeog- 

 raphy or geology, I understand that proc- 

 ess of reasoning which seeks to demonstrate 

 that certain events of past history occurred 

 simultaneously. 



Contemporaneity. —In dealing with the 

 enormous time intervals of the earth 's his- 

 tory the concept of simultaneous or con- 

 temporary events must be liberally grasped. 

 A fair statement is that the phenomena de- 

 scribed as contemporaneous shall have 

 existed at the same time within limits of 

 error which do not equal a large fraction 

 of the life of either. Thus we call two men 

 contemporaries when the periods of their 

 active lives coincide, though one may have 

 been born notably later and live longer 

 than the other. But we do not so term a 

 youth and a graybeard, whose living occu- 

 pies but a few years in common. 



It is evident that two long-lived events 

 may differ from near coincidence in time 

 by a larger margin than two short-lived 

 events, and yet be reasonably regarded as 

 contemporaneous. The marine transgres- 

 sion which submerged most of North Amer- 

 ica during the Cambrian was in a broad 

 sense contemporaneously paralleled by sub- 

 mergence of much of Eurasia; but the 

 moment of arrival of the earliest Cambrian 

 fauna, the Olenellus, which followed the 

 spreading shores over each continent, can 

 not be regarded as contemporaneous at 

 points reached earlier and later in course 

 of the submergence. 



The evidences of contemporaneity are 

 both inorganic and organic, but, though we 

 are wont to classify them thus in two dis- 

 tinct categories, they are most intimately 

 related through that principle of periodic- 

 ity, which is at the bottom of all terrestrial 

 phenomena. Diastrophism is periodic, all 

 changes in the inorganic as in the organic 

 are conditioned by that periodicity, and all 

 such changes are therefore themselves 

 periodic. Moreover, the physical and bio- 

 logical phenomena are linked in a continu- 

 ous chain of cause and effect, which 

 stretches from gravity and internal heat 

 at one end to life at the other, and which 



