302 EXPERIMENTS ON METAMORPHISM AND 



carbonaceous combinations which we meet with in the more ancient 

 formations,* (graphite of St. Marie-aux-Mines, anthracite of Konigs- 

 berg. in Norway, or of Dannemora, Avhere it is in a gray limestone 

 hardly crystalline, bitumen of the veins of granite of Finbo, near 

 Fahlun, and the numerous iron formations of Sweden.) 



Other geologists, on the contrary, struck with the intimate con- 

 nexion which exists between the granite and this gneiss, have con- 

 sidered the latter as a granite which has become schistose by having 

 been flattened out. If this was so, it would be necessary to conclude 

 that certain masses of limestone, quartzite, magnetic iron, and other 

 metallic ores, pre-existed in the granite, and that they have been 

 softened at the same time with it, so as to be susceptible of being 

 extended simultaneously, and thus take the form of plates parallel 

 to the foliations of gneiss, imitating stratification in a very striking 

 manner. This is a supposition that can hardly be admitted. 



There are still two important remarks to be made on this subject : 

 1st. The absence of transition in the azoic schistose rocks of the 

 Silurian formation shows that the first rocks had already acquired their 

 crystalline state, before the deposit of the oldest known fossiliferous 

 rocks. This fact is confirmed by the pebbles of well characterized 

 gneiss, which the formations of transition sometimes contain. 2d. 

 There is no indication that these same ancient rocks have in certain 

 countries ever been covered by any considerable thickness of other 

 rocks, otherwise we should have to admit, and we have no ground for 

 doing so, that extended and scarcely undulated countries like Scandi- 

 navia or Canada, have undergone enormous denudations. Besides, 

 formations like those of Scandinavia and North America, which we 

 have just cited as examples, are found in all parts of the globe with 

 analogous characters ; they form almost universally a kind of cover- 

 ing for granite. 



Supposing that the mass of all our seas were diffused as vapor in 

 the atmosphere, the pressure on the surface of the globe would 

 amount to at least two hundred and fifty times what it now is ,t it 

 would even be greater by reason of the intervention of gases and of 

 other vapors. There could, then, not have existed liquid waters on 

 the earth, until the temperature of its surface was reduced below that 

 degree of heat which can give to the vapor of water a tension of 

 two hundred and fifty atmospheres. The surface of the globe, then, 

 .was at that epoch at a very high temperature, and if there existed 

 any silicates on the surface of the earth they must have been formed 

 in the dry way. Afterwards, when water had commenced forming in 

 the liquid state, it would have reacted on these previously existing 

 silicates, and would have given rise to an entire series of new pro- 



• So long as synthesis, which in the hands of Berthelot had led to much remarkahle 

 results, has not imitated the anthracites of Sweden without the aid of plants, we must 

 believe that these combustibles are of vegetable origin, and consequently that the plants 

 existed when the gneiss which these deposits contain was formed. 



•j- Admitting, with Baron Humboldt, a mean depth of 1,900 fathoms for the ocean, we 

 should have 1,400 fathoms as the depth of a bed of water uniformly spread over the globe, 

 which corresponds to a pressure of 2-48, or in round numbers 250 atmospheres. 



