368 HENRY S. WASHINGTON 
my comparative study of the norms and modes of the Italian leucitic 
rocks' showed that it was more frequently the normative nephelite 
rather than the normative olivine which became silicated, though the 
latter may also take up silica. If follows then in these cases that, not 
only did the potash show a less affinity for silica than did the soda, at 
least in part, but that the soda had a greater affinity for siliea than had 
the magnesia and ferrous oxide. This latter fact, it will be observed, 
is in perfect harmony with the order of affinity for silica assumed 
as fundamental in the calculation of the norm and mentioned 
above. 
One more fact is to be noted in this connection—namely, that, while 
the amount of leucite formed varies much, even attaining the maxi- 
mum possible in some cases, the silica liberated is never more than 
what may be completely used up in the formation of either albite or 
pyroxene, so that we never find an excess of it which would be crystal- 
lized as quartz. This is in accordance with the conclusion reached 
from the study of the diagrams, that quartz should not occur with 
leucite; and this applies as well to nephelite. 
These exceptional cases, represented by colored squares, fall toward 
the left side of the diagram, extended over a rather broad oblique 
zone, which is approximately the space between the lines O°N® and 
O4N‘, only a few being to the right of this on the less siliceous side, 
while rather more fall to the left of O4N*. As regards potash they 
vary widely, from about .o25 to .125. In connection with their 
distribution it must be remembered that, were all known rock analyses 
plotted, this space would also contain a much greater number of 
uncolored squares, representing rocks free from both normative and 
modal leucite, only a few of which are presented here. 
This zone, occupied by rocks which are normatively free from 
leucite, but which may or may not carry it in the mode, is that of the 
magmas which have been referred to as “critical” ones, as regards 
the formation of leucite, this implying that they are in a nicely balanced 
condition chemically, so that a comparatively slight change in the 
1 The Roman Comagmatic Region (Carnegie Publication No. 57, 1906), pp- 155; 
187. Cf. also A. Lacroix, Comptes Rendus, Vol. CXLI (1905), p. 1191, who shows 
that in some cases much of the olivine may take up.silica, and so induce the for- 
mation of leucite, when the soda is so low that the entrance of all of it into albite may 
not do this. 
