FORMATION OF LEUCITE IN IGNEOUS ROCKS 383 
and who leaves the discordance between their chemical and modal 
characters unexplained. In this condition we also are perforce com- 
pelled to leave them still, as study of the relations of norm and mode, 
and consideration of the presence of the phlogopite, but add to the 
difficulties of explanation. ‘These rocks may well be considered as, 
chemically, the most remarkable and the least susceptible of corre- 
lation of any igneous rocks so far discovered. 
GENERAL CONCLUSIONS 
From the above discussions and comparisons we may draw certain 
general conclusions, both as to the affinity of the various bases in 
igneous rocks for silica and as to the formation of leucite, these con- 
clusions being stated under the two separate heads. 
Affinity jor silica.—The assumption with which we started, that 
potash has a greater affinity for silica than has any of the other oxides, 
is amply substantiated, and may be considered as fully established, 
subject to certain modifications mentioned later. 
Next to potash, soda has the most affinity for silica, the affinities 
of magnesia and ferrous oxide being considerably less, so that norma- 
tive nephelite will take up silica before normative olivine. 
While these relative affinities apply in the great majority of cases, 
and under most conditions, yet, in accordance with the general laws 
of chemical affinity and of mass action, they are susceptible of notable 
change under certain conditions, especially of temperature and 
pressure, and the presence of mineralizers, and may be in part 
superseded by the influence of the mass action of abundant complex 
molecules and the relative crystallizability of the minerals involved, 
the influence of the ready crystallization of soda-lime feldspars con- 
stituting the most notable case in point. 
Soda and lime, as alumino-silicates, have a strong affinity for each 
other, as have soda and potash alumino-silicates, though to a somewhat 
less extent; and if lime and soda are present in sufficient amount in 
the magma, these affinities may supersede, at least partially, the usual 
greater affinity of potash for silica. 
In the presence of water in the magma, and under certain conditions, 
the affinity of soda for silica may supersede that of potash, resulting 
in the formation of analcite and leucite. 
