of the Brazilian Topaz. 9 
with a difference in chemical composition; and I am therefore 
disposed to ascribe the present apparent exception to the imper- 
fection of chemical analysis. 
This opinion, indeed, may be placed upon a firmer basis by 
a reference to some experiments of that excellent Chemist, the 
late Rev. William Gregor, which have recently come to my 
knowledge. In a letter to the Rev. John Rogers, dated Nov. 21, 
1811, he states that some pieces of Topaz from St. Michael’s 
Mount consisted of silica and alumina in the proportion of 3.5 
to 3.1, and a small proportion of lime. ‘‘The proportion of 
** silica and alumina,” he adds, “‘ does not agree with that in the 
*“Saxon Topaz; but still I see a great difference between the 
** Saxon and Brazilian Topaz, as to the relative proportion of 
“those ingredients. In neither of them do I see any lime re- 
““ cognized; but I must say, notwithstanding the analysis of 
* Klaproth, that I extracted a portion of Lime from the Bra- 
“‘ zilian Topaz, by means of a simple acid.” 
In another letter to the same gentleman, dated April 24, 
1812, he says, that “subsequent experiments confirmed him in. 
“the opinion, that Brazilian Topaz contains a small proportion 
“of Potash; and Dr. Paris, from whose biographical Memoir 
of Mr. Gregor these particulars are taken, states that the ex- 
periments by which he established this fact were the last he 
performed ; and that he himself was enabled to bear testimony 
to the presence of crystallized alum, which he saw Mr. Gregor 
produce by the action of sulphuric acid upon the pulverized 
‘mineral. In order to avoid every possible source of error, the 
mineral was reduced to powder in a steel mortar, and the sul- 
phurie acid employed was carefully tested, and found to be 
perfectly free from any impurity. 
Edinburgh, 
March 4, 1822. 
Vol, Il. Part I. B 
