ii It. Mean. 
Silica, 50°29 50-51 50°40 
Alumina, 32-40 32°35 32°38 
Magnesia, 7 8-04 7:97 8-01 
ime, ... . . trace trace. trace. 
Protoxyd of iron, . 2:30 2°39 2:34 
Protoxyd of manganese, trace. trace. trace. 
ater, . . ; 6°58 6-73 
99-61 : 
The corresponding ratio of oxygen as deduced from the mean of 
those two analyses, is 
Si Al; 2 US 
26:18 : 15:12 : 3°63 : 5-98, 
proportions here obtained, we obtain expressions that are scarcely 
probable and noways in harmony with the formula of cordierite, 
though nevertheless, as will be immediately seen, both minerals, 
Owing to the circumstances of their occurring together, stand in 
op ec relation to each other. 
n comparing the composition of aspasiolite with that of the 
cordierite of Kragerée, we see that the silica and alumina occur 
i Nearly the same quantities in both substances, and that it is 
Principally in the diminished amount of magnesia, consequent 
upon the considerable amount of water, that the constitution of 
the former mineral differs from that of the latter. This circum- 
And that there is not the slightest reason to pe ee 
a 
iit is evident from the fact that the fresh and compact mass 
