Occurrence at Hebron, Maine. 215 



in the case of Plattner's, with the assumption that the excess 

 was due to the presence of potash. It is the author's opinion, 

 from a consideration of one of Rammelsberg's analyses which 

 will be mentioned later and of the analyses of the new material 

 from Maine, that Pisani's excess was at least largely due to too 

 much soda, either derived from glass vessels or from some 

 other cause, hence a ratio is given under 2a above, after de- 

 ducting 1*71 per cent of soda from the analysis. Pisani de- 

 duced from his analysis the oxygen ratio, Si0 2 : Al 2 (Fe 2 )0 3 : 

 Cs/Ca^aJO : H 2 0=15 : 5 : 2 : 2. This ratio would be ex- 

 pressed by the very complicated formula, 45Si0 2 . 10Al 2 O 3 . 

 12Cs 2 . 12H 2 0. 



Pisani certainly left the question of the true composition of 

 pollucite open to doubt, and in 1878 Rammelsberg jmblished* 

 a new analysis of the mineral with the view of clearing up the 

 doubt. Rammelsberg's material was evidently not well adapted 

 to the purpose of determining the composition of the mineral, 

 for he first picked from it some pieces, " more or less translu- 

 cent," and obtained from them, A1 2 3 16*58, alkalies precipi- 

 tated by platinic chloride 23*03, Na 9 2-00, Li 2 0*83; then 

 he picked from the same material, some fragments which had 

 a specific gravity of 2*868, the lowest number which has ever 

 been given for the mineral, although Breithaupt gives the 

 same number as the lowest of a series, and he made the follow- 

 ing analysis from it : 





Rammelsberg, 









First analysis. 





Ratio. 



Si0 9 



[48-15J 





[•802 or 5-01] 



A1 2 3 



16-31 





•160 or 1-00 



Cs 2 



30*00 



•106 



) 



K 2 



0-47 



•005 



y -151 or 094 



Na 9 



2-48 



•040 



) 



H 2 



2-59 





•144 or 0-90 



100-00 



On this single analysis, where an important constituent was 

 determined by difference and where the material was of ques- 

 tionable purity, Rammelsberg obtains the formula which is 

 now generally accepted for the mineral. The analysis corres- 

 ponds to the formula H 2 R / 2 Al 2 (Si0 3 ) 5 ; Rammelsberg includes 

 the hydrogen in R 7 and writes it R' 4 Al 2 (Si0 3 ) 5 . 



It may be inferred that Rammelsberg himself was not fully 

 satisfied with his results, for about two years later, he pub- 

 lishedf an analysis of what he describes as the purest material. 



* Berlin. Akad., 9, 1878. f Berlin. Akad., 671, 1880. 



