Boundary in South America. 35 



lera, where it is truncated by a large fault ; to the south, 

 after gradually receding, it disappears under widespread 

 sheets of basaltic and andesitic rocks. 81 The antecedent 

 valleys of the Rio Diamante and Rio Atuel show that 

 during a great part of the Tertiary the Sierra Pintada 

 underwent a slow and gradual rising movement. The 

 unconformity at the base of the Gondwana Series and the 

 existence of Permian Glacial Conglomerates characterize 

 this range as an intermediate link, joining the Precor- 

 dillera of San Juan and Mendoza with the Sierras of 

 Buenos Aires. 82 In the great acute-angled depression 

 between it and the Andine structural element, where at 

 present the most striking hydrographic element is repre- 

 sented by the Laguna Llancanelo, ended the transgres- 

 sion of the Roca-Sea. On one side this sea of the past 

 washed the foot of the High Cordillera, approaching the 

 flanks of that great batholith, and on the other side 

 towards the east the sea ended at the southern foot of 

 the Sierra Pintada among hundreds of small islands and 

 peninsulas. From this point of view, we may perhaps 

 consider the Laguna Llancanelo as a relic of that period. 



In the region of the Cerro Payen, according to the 

 investigations of Groeber, the course of the shore was a 

 little to the west of the Rio Grande. Here too the sea 

 ended at the first folds of the Cordillera. Direct obser- 

 vations of a small portion are still lacking, but there are 

 reasons for assuming that the shore was stretched more 

 or less midway between Chos Malal and Auca Mahuida. 

 The writer has seen deposits of the Roca-Beds not far 

 from the bend of the Rio Neuquen, and from here to the 

 region of the village of General Roca, the course of the 

 Rio Neuquen and Rio Negro indicate the western shore- 

 line of the Roca-Sea. 



On the Rio Neuquen we enter into the region of the 

 great fault noted in Part II as an ancient San Jorge 

 shore-line. Although we cannot well dispense with 

 direct observations in the tableland to the south of Roca, 

 it may from the general study of tectonic and morpho- 

 logic phenomena be justifiable to combine the Neuquen 

 line with the fault-scarp at the northern border of the 



81 Stappenbeek, Apuntes hidrogeologicos sobre el Sudeste de la Provineia 

 de Mendoza, Boletin 6B, Direction Gral. de Minas etc., Buenos Aires, 1914. 



82 These characteristic features of the structure of the Sierra Pintada 

 had been observed by the writer in 1909-10, but an interpretation was 

 not possible, until Keidel discovered the relations between the Sierras of 

 Buenos Aires and the Precordillera of San Juan and Mendoza. 



