Carlsbad Waters and Uranium. 185 



ter formations, and partlj by plants brought from a distance by 

 floods, and deposited. Hence the more dehcate parts are 

 generally destroyed in the more recent formations, while 

 in the coal beds the finest leaves and nerves are exactly 

 preserved. As the cryptogamic monocotyledons which 

 constitute nine tenths of the coal fossils are not articulated, 

 but continuous throughout their whole structure, they could 

 not have been detached from their roots without a destruc- 

 tion of their organization. He therefore supposes they 

 grew where they are now found. 



The volume is illustrated by six lithographic plates, il- 

 lustrating the Genera and Species described in the textj 

 and exhibiting some of their analogies with existing veget- 

 ablesr 



2. Carlsbad Waters and Uranium, 



Extract of a letter from M. Berzelius, Stockholm March 

 20, 1822. 



'^r have nothing of importance to communicate from my 

 own laboratory, except an analysis of the waters of Carlsbad 

 in Bohemia, which I visited last summer. I have found in 

 them many substances, which had not hitherto been found 

 in mineral waters, viz.Jiuate af lime, Carbonate of strontian, 

 phosphate oj lime, and phosphate of Alumine. These sub- 

 stances are found there dissolved in carbonic acid uncom- 

 bined. The tufas, deposited by these waters, are arrag- 

 onitic, which corroborates the idea of M. Stromeyer that 

 it is the carbonate of strontian, which determines the aTra-^ 

 goniticform of this species ofcarb. of lime. 



M. Arfredson, who has been engaged in researches on 

 Uranium, has just found, that this metal is very reducible- 

 by means of Hydrogen gas at a temperature scarcely equal 

 to redness. The combinations of the yellow oxide of Ura- 

 nium, with barytes, lead, and iron, are reducible in the 

 same way and give metallic Uranures, which take fire when 

 they come in contact with the air, and burn like pyrophori* 



The experiments are so easy that they can be made dur- 

 ing a lecture." 



VoL.VIL— No. 1. 24 



