part 1 ] THROUGH THE ANDES OP PERU AND BOLIVIA. 57 



statement of the observations made in the area now described may, 

 however, not be out of place bere. 



A journey through the Pre-Alps of Chablais, made some years 

 ago, under the leadership of Prof. Lugeon, enabled me to draw a 

 comparison between the structure of a mountain-chain of Alpine- 

 type and that presented by the Cordilleras of the Andes. After 

 becoming acquainted with the latter, I could not fail to be impressed 

 by the complete dissimilarity of the two types of structure. 



Preconceived notions of vast sheets or recumbent folds, trans- 

 ported over wide areas and separated from their roots to expose an 

 underlying ' yoke,' were at once swept away. Throughout the 

 Peruvian Cordilleras, inverted folds are the exception rather than 

 the rule, and great zones of overthrusting appeal' to be entirel}' 

 wanting. Any directional movement of the folding, moreover, 

 is hard to determine, and the relationship between that part of 

 Grondwanaland represented by the Brazilian platform and the 

 folded chains of the Andes is, in the light of recorded facts, still 

 obscure. 



(Suess, in a summary of the structure of South America, makes 

 the two following contradictory statements : ' In South America 

 the Brazilian mass occupies the place of the baekland within the 

 arc, and the foreland lies beneath the ocean'; and later ' In about 

 the latitude of the Bay of Arica the western promontory of Brazilia 

 was overwhelmed by the folding movement directed towards the 

 east.') 



Belying on my own observations, I am led to believe that the 

 folded chains of the Andes are the result of intermittent com- 

 pression of a series of transgressive deposits, laid down in a geo- 

 syncline, between two ancient resistant masses, represented on the 

 east by the metamorphic and plutonic rocks of the Amazon region, 

 and on the west by the crystalline rocks of the coastal Cordillera. 

 The trend-lines of the system, according to this theory, were deter- 

 mined at a comparatively-early date, and the areas covered by later 

 transgressions were therefore limited by the continued rise of the 

 chains on the east. 



In the Alpine type of folding vertical uplift has been over- 

 shadowed by movement in a horizontal direction, whereas in the 

 Andean Cordilleras the reverse is the case, and the terms 'baekland' 

 and ' foreland ' as applied to the direction of movement have no 

 longer the same significance. 



Attention may once more be drawn to the fact that in the two 

 marginal areas the igneous rocks are of ' alkaline ' facies, whereas 

 the later plutonic and extrusive rocks of the chains themselves are 

 of ' calcic ' type ; and, if we assume that relief from increasing 

 compression was eventually afforded by deep-seated dislocation, 

 possibly in the form of submarginal thrust-planes, an explanation 

 is offered for the great batholitic invasion of the chain by a 

 granodioritic magma in post-Jurassic times. 



