OSCILLATORY CHARACTER OF CONTINENTAL SEAS 337 



pressed spaces which, on account of their location were as a rule domi- 

 nated by the prevailing positive tendency of adjacent areas, were but 

 rarely overflowed. However, whether submerged or not, the essential 

 structural relations of adjoining areas remained always the same. 



The great Canadian shield — Laurentia — seems to be a region that has 

 not been entirely submerged since the beginning of the Paleozoic, and 

 submergences of interior parts of it occurred only at times of such low 

 average altitude of the continent as prevailed during the middle Ordo- 

 vician, middle Silurian, middle Devonian, and middle to late Cretaceous. 

 Small to large remnants of the deposits of these ages are now found only 

 in the gentle downwarps of the shield. Though probably still an archi- 

 pelago during even the greatest of the submergences mentioned, I do not 

 doubt that the deposits accumulated at such times, especially the Ordo- 

 vician and Silurian, were originally much more extensive than we now 

 find them. Even admitting that their erosion was exceedingly slow in 

 the downwarps, at least to the Cretaceous, we may reasonably assume it 

 to have been more active on the flanks of the upwarps. Considerable 

 parts of the Paleozoic sheets probably were removed from these more ex- 

 posed situations, and even more may have been carried away in the Pleis- 

 tocene, during which time the region was repeatedly subjected to active 

 ice and water erosion. The probable fact that every Paleozoic marine 

 formation that was ever laid down on this shield is preserved in part at 

 least in its downwarps, especially if we consider the "positive" tendency 

 of the region, argues strongly against the assumption that thick deposits 

 were ever completely removed from any large area. For this and other 

 reasons I maintain that we are justified in interpreting the total absence 

 of a given formation in any considerable structural basin as signifying 

 absence of the corresponding sea. 



However, in the making of a paleogeographic map the investigator 

 must take into account the possible channels through which faunas might 

 reach certain areas, and here we encounter the principal, if not the only 

 exceptions, to the above rule. These channels may be buried, or deposi- 

 tion may have been prohibited in them by rapid currents ; or, if they re- 

 ceived deposits, these may have been on positive areas where they would 

 be liable to rapid erosion in subsequent emergent stages. ( See discussion 

 of effects of currents in continental seas, page ,362.) Positive informa- 

 tion respecting the first condition is often acquired in boring deep wells. 

 Thus, the drill has frequently established the presence, imder cover of 

 overlapping later deposits, of considerable beds, or even whole forma- 

 tions, of which no sign is to be seen in exposed sections. Occasionally 

 the paleogeographer, reasoning from the known geographic occurrences 



