IO, _.2.; 56°190 51°703 61:157 50°043 50°12 51:36 
Biss. AG LTT 1°720 97154 2°90 2) 0:37 
Fe,0, _. 4919 OOUR eee eee jo etek a 
et £433 17°995 18°360 17°812 23°59 21°97 
MnO __. trace 0°363 0°363 07120 1:32 
CaO..... 6-996 2-873 3°81 6696 10°49 3 
MgO... 4°601 25091 24°251 91-744 11°05 21°31 
eG ea pe ee a cee 
Na,O. POG ie Pe tice 0°24 OGTR Juste 
H.O Stik s “028 HORS oy ae A ce os OPER Se WS Se RTT ae RAT eI gh Pim, Water rome pe 
P05 = ge oe Nk a RA eee Pe Hg pias eee 87 
Ores ce 0-022 
99°901 100°049 100°097 99°595 99°64 98°72 
In analyses III and IV the total amount of iron is given 
protoxide, the small quantity of available material rendering 
satisfactory results impossible. An amount corresponding to 
that given in II is to be considered as sesquioxide. The MnO 
of III is taken from I, and is undoubtedly very nearly correct, 
the portions being derived from the same rock. The micro- 
scopic investigation of the material giving results II and Il 
ties of the mineral, fully justify the application of the name 
hypersthene to the pyroxene, which, in all of the Buffalo Peaks 
*“Santorin et ses éruptions.” Plate LX, fig. 2. 
