ARRANGEMENT OF STRATA. 113 



Again, a stratum of one age may rest upon any stratum in 

 the whole of the series below it, — the Coal measures on either 

 the Azoic, Silurian, or Devonian strata ; and the Jurassic, Creta- 

 ceous, or Tertiary on any one of the earlier rocks, the intermediate 

 being wanting. 



In addition, denudation and uplifts have thrown confusion among 

 the beds, by disjoining, disarranging, and making complex what 

 once was simple. In the United States, many a sandstone in New 

 York and Pennsylvania is represented by a limestone in the Ohio 

 and Mississippi valleys, — that is, the two were of cotemporaneous 

 origin ; some rocks in eastern New York are not found in the 

 western part of that State, and some in the central and western 

 not in the eastern. The Post-Tertiary in America in some places 

 rests on Azoic, in others on Silurian or Devonian, in others on 

 Cretaceous or Tertiary. And, if so great diversity of condition 

 exists in one country, far greater may be expected between dis- 

 tant continents. 



Amidst all these sources of difficulty, how is the true order ascer- 

 tained ? 



123. Means of determination. — It is plain from the preceding 

 remarks that the true method cannot consist in grouping rocks of 

 a kind together, as limestones, shales, or sandstones. It is irre- 

 spective of kinds, and is founded on a higher principle, — the 

 same which is at the basis of all history, — successiveness in events. 

 The following are the means employed. 



(1.) Order of superposition. — "When strata are little disturbed, ver- 

 tical sections give the true order in those sections and afford 

 valuable information. Or where the strata outcrop over the surface 

 of a country, the succession of outcropping layers affords a section, 

 and often one of great range. The vertical extent of such a section 

 may be ascertained as explained in \ 119. In using this method by 

 superposition, several precautions are necessary. 



Precaution 1st. — Proof should be obtained that the strata have not 

 been folded upon one another, so as to make an upper layer in any 

 case a lower one in actual position (see p. 107), — a condition to be 

 .suspected in regions where the rocks are much tilted, but not where 

 the tilting is small. 



Precaution 2d. — It should be seen that the strata under examina- 

 tion are actually continuous. 



A fault in the rocks may deceive ; for it makes layers seemingly 

 continuous which are not so. In some cases, beds forming the 

 upper part of a bluff (as a b, fig. 114) have settled down bodily 

 (c) to the bottom, so as to seem to be continuous with the older 



