AMERICAN GEOLOGY DECADE OF 1860-1869. 



529 



a previous paper, where he showed that a compound bar would, when 

 heated, bend toward the side composed of the most expansive mate- 

 rial, he compared such a bar to a portion of the earth's crust covered 

 with several thousand feet of ice and snow. The effect of this blanket 

 would be to cause the isothermal lines to move outward toward the 

 surface, causing thus an expansion of that portion of the crust imme- 

 diately beneath the ice. But the ice itself would partake very slightly, 

 if at all, of this increased temperature, and, as in the case of the com- 

 pound bar, the bending would take place in the direction of maximum 

 expansion, i. e., in this particular case, downward. In this way, he 

 suggested, the depression accompanying the period of maximum glacia- 

 tion might be accounted for. 



C. F. Hartt and Orestes St. John accompanied Agassiz in the capac- 

 ity of geologists on the Thayer Expedition to Brazil during the years 

 1865-66. In 1867 Hartt made a second 



journey, spending several ^iSBBt^^ 



C. F. Hartt's Work J , ^t ife. 



in Brazil, months on the coast, be- 



1865-1867. ' . 



tween rernambuco and Rio, 

 exploring more particularly the vicinity of 

 Bahia and the islands and coral reefs of the 

 Abrolhos. 



The results of this and the previous ex- 

 pedition were published in book form in 

 1870, under the title of Geology andPhysical 

 Geography of Brazil. In this work the 

 gneisses of the Province of Rio de Janeiro 

 are regarded as metamorphosed or sedi- 

 mentary deposits and of Azoic age. Their 

 thickness he did not even estimate, recog- 

 nizing the fact that their apparent enormous thickness was due to 

 numerous reversed folds, so that one might travel for miles over their 

 upturned edges, finding them always highly inclined and dipping in 

 the same general direction. 



Concerning the probable age of the metamorphic rocks succeeding 

 the gneisses he found no proof, though it was suggested they might 

 be Silurian or Devonian. South of Rio he found unmistakable Car- 

 boniferous rocks, including beds of bituminous coal, and in the prov 

 ince of Sergipe, underling the Cretaceous, a thick series of red 

 sandstones, referred to the Triassic. No Jurassic was recognized. 



Marine Cretaceous beds of undetermined extent were found north 

 of the Abrolhos Islands, which were conformably overlaid by clays 

 and ferruginous sandstone, referred to the Tertiary. Overlying this 

 along the whole coast he found an immense sheet of structureless 

 clays, gravels, and bowlder deposits, which he believed, with Agassiz. 

 to have been the work of glacial ice, though he noted that nowhere 

 NAT mus 1904 34 



Fig. 78.— Charles Frederick Hartt. 



