schaller] DUMORTIERITE. 119 
* The ratios of the lust analyses are 
SiO, 5. 94 or 6X0. 99 
A1 2 3 8. 09 8X1. 01 
B 2 3 HO 1 x .80 
II,<) 1.22 1X1.22 
Approximately they agree with the formula. 
The only question of which there seems to be any doubt is whether 
the water should be 1 or 1^ molecules. As, however, the best analyses 
show but 1 molecule and their evidence is considered as of much more 
value than a larger number of inferior analyses, there is no question 
in the writer's mind but that the formula for dumortierite is 
8A1 2 3 , IRA, 1H 2 0, 6Si0 2 . 
This may be written 
(Si0 4 ),Al(A10) 7 (BO)H, 
which is the same as the formula given by Groth except that for (BO) 
he puts (AlO). His formula is (Si0 4 ) 3 Al(A10) 8 H. 
This formula may be written in the following form: 
SiO=(A10) 3 
Al^SiO=(A10) 3 
SiO — AlO 
\BO 
\H 
This formula is similar to that of andalusite, which is written 
SiO=Al 
Ah(SiO==Al 
Si0 4 =(A10) 3 
Dumortierite alters to muscovite. The change can be very well 
shown by these formula?, and seems to be entirely in- accordance with 
thc.gi oup of minerals to which it is related. 
Dumortierite. Muscovite. 
Si()=E(A10), SiO=Al 
Al^SiO=(A10) 3 =Al<(Si0 4 E=Al 
SiO.— AlO SiO— K 
\BO \1I, 
\H 
-[-alumina and boric acid set free. 
LITERATURE. 
Berthand, Emile. Sur un mineral bleu de Chaponost, ]nvs Lyon. Bull. Soc. Min., 
vol. 3, 1880, p. 171. 
Gonnard, F. Note sur l'existence d'une espece mim'rale nouvelle, la dumortie- 
rite dans le gneiss de Beaunan, audessus des anciens aqueducs galloromains de la 
vallee de l'Iseron (Rhone). Ibid., vol. 4, 1881, p. 2. 
