1825.] 



Three new Salts of Soda, 



443 



management of the laboratory in Mr. Tennant's work. He 

 collected a considerable quantity of these crystals, and subjected 

 them to a chemical analysis ; the result of which led him to con- 

 clude, that the constituents of these crystals were 1 atom car- 

 bonic acid, 1 atom soda, and between 7 and 8 atoms water. 



To this gentleman I was obhged for about a pound weight of 

 very regular and pure crystals of this new salt, the properties of 

 which 1 shall now describe. 



The crystals are four-sided prisms terminated by four-sided 

 pyramids, some of them an inch and a half in length, and more 

 than one-fourth of an inch thick. They do not effloresce when 

 exposed to the air, even in very dry weather. But my labora- 

 tory, in which this trial was made, is a damp room ; for the 

 College of Glasgow, on a ground floor of which is my laboratory, 

 is built on a clay soil. Though I examined upwards of 100 

 crystals with care, I did not find one with faces sufficiently 

 smooth to admit of measurement with the reflective goniometer. 

 But with the common goniometer, I obtained the following 

 measurements, which I consider as tolerable approximations. 



MonP 90° 



P on M' 90 



Monb .,, 115 



M^on// 115 



Pon« 125 



a on h, 01 b' 150 



We may consider the primary form as 

 a right rectangular prism with a rectan- 



gular base 



M 



•M 



The common carbonate of soda is a 

 bipyramidal octahedron, the common base 

 of the pyramids of which is a rhomb with 

 angles of 120° and 60°. If we suppose 

 this form to be represented by the figure 

 AB CD, then an idea of the common crys- 

 tal of this salt will be obtained, if 

 we suppose the acute angles A, B, 

 of the rhomb, which forms the 

 common base of the pyramids, to 

 be truncated by a plane parallel 

 to the axis C D of the octahedron. 

 These truncations are more or less 

 deep ; but I have never met with 

 a crystal without them, though I. 

 have examined several hundred crystals of aU sizes from half an 

 inch to eight inches in length. 



It would be possible to derive the right rectangular prism from 



