STRATIGRAPHY 737 



a fluvio-glacial accumulation, composed of debris spread broadcast over 

 a lowland area by streams issuing from melting ice. In part it may be 

 lacustrine or even marine, but its materials have in the main been derived 

 directly or indirectly from glaciers or ice-sheets. Doubtless it is genetic- 

 ally related to the tillite in the Bermejo beds, observed along the trail 

 from Cochabamba to Santa Cruz, 8 and to the glacial formations in the 

 eastern Andes in northern Argentina, recently described by Keidel 9 as 

 being overlain by strata carrying the Gondwana flora. 



The Mandiyuti conglomerate varies in thickness from about 1,500 feet 

 in the north part of the Sierra de Aguarague to about 3,200 feet at the 

 type locality in the Sierra de Mandiyuti. In most places it is between 

 2,000 and 3,000 feet thick. Individual beds as much as 250 feet in thick- 

 ness occur in the midst of the formation at several localities. These dis- 

 play closely spaced, irregular laminations. 



The Mandiyuti conglomerate is well exposed in the three canyons in 

 the Sierra de Charagua, through which reconnaissance traverses were run, 

 and evidently it forms the central massif of that mountain range through- 

 out its entire length. No unusual features were noted in that sierra, al- 

 though in general the formation is possibly somewhat more irregularly 

 bedded there than elsewhere. The lower part of the Mandiyuti conglom- 

 erate is exposed in only one of the three gorges examined; in that one, 

 the Charagua canyon, the formation is at least 2,000 feet thick. Approxi- 

 mately 1,500 feet of Mandiyuti beds are exposed in the gorge of Rio 

 Saipuru, while Rio Tacuru has cut only 500 feet below the top of this 

 formation. 



Again, in the Sierra de Aguarague the Mandiyuti conglomerate forms 

 the central portion of the range from end to end. Among the pebbles 

 contained in the conglomerate members, cobblestones four or five inches 

 in diameter, with undoubtedly glaciated surfaces, were noted at several 

 places. In the Vitiacua gorge the Mandiyuti conglomerate is about 1,600 

 feet thick, while in the Quebrada de Los Monos its thickness was esti- 

 mated at 2,500 feet. At the latter locality three distinct conglomerate 

 horizons, separated by sandstone and shales, were noted. 



In all probability, the Mandiyuti conglomerate is the source of the 

 "few scattering fragments" of igneous rocks reported by Herold 10 "to be 

 found in the creek beds" and interpreted by him as having "been trans- 

 ported from the higher mountains of the Andes ranges . . . previous 



8 Heald and Mather, 1922. 



9 Juan Keidel : Observaciones Geologicas en la Precordillera de San Juan y Mendoza. 

 Anales del Ministerio de Agricultura, Sec. Geol., etc., Tomo XV, no. 2. Buenos Aires, 

 1921, pp. 57-63. 



10 S. C. Herold, 1920, p. 556. 



