204 HENRY S. WASHINGTON 
In fact, the more we study the igneous rocks, the more we 
are impressed with the great differences that the two minerals 
show in the two classes, as well as with the invariability of this 
behavior. In the one class of rocks we find the alteration 
d 
“noticeable by its absence,” while in the other the tendency is 
the other way, and its frequency is what most strikes us. 
I have emphasized these respective characters because their 
invariability, resting as it does on such a broad basis of observa- 
tions, shows that they are facts of prime importance for the 
solution of the problem. When we come to look at the class of 
volcanic rocks more closely, we find differences in special cases 
which are fully as constant and, while at first sight seemingly at 
variance with the above observations, are seen on further study 
not to clash but to be an almost equally important means to our 
end. 
While the statement that the alteration is almost constant in 
the volcanic rocks (except the rhyolites, SUS.) 1 (als Oi tne 
great majority of structural types, yet it has often been observed’ 
that in the more glassy modifications the alteration is less than 
in the more crystalline; so that, speaking broadly, the frequency 
and amount of alteration may be said to be in roughly inverse 
ratio to the amount of glass basis present. Hence the above 
remarks on the invariability of alteration do not hold good for 
the highly vitreous forms, the obsidians, tachylytes, etc., in 
which both the hornblende and biotite are as a rule unchanged. 
Exceptions are to be found, but they are so few that they cannot 
impair our confidence in the general law that alteration is 
roughly inversely proportional to the amount of glass basis 
present, or, coeteris paribus, to the rate of cooling. 
We must conclude, then, adopting the method of concom- 
itant variations, and the factor of composition of the magma 
being eliminated by the researches of Hague and Iddings? and 
others (except in the rhyolites, etc.), that the alteration of horn- 
blende and biotite does not take place under conditions of great 
"Cf ZiRKEL, Lehrb., I. 723. ROSENBUSCH, Mikr. Phys., I. 484. II. (1887) 650. 
210s 175 Woo Cio Say Ubstesisn 
