THE ORIGIN OP IGNEOUS ROCKS. 
145 
presses a natural order of eruption of frequent occurrence, 
which his own theory of the origin of igneous rocks failed to 
explain satisfactorily. 
In a number of localities studied by the writer, especially • 
in the region of the Yellowstone Park, the order of eruption, 
which may be repeated more than once in some localities, 
is: andesite of mean composition, including hornblende- 
andesite and hornblende-mica-andesite, followed by eruptions 
of more basic andesite and basalt, and more silicious andesite 
and dacite, and by basalt and rhyolite and basalt, which 
may be modified in many ways in particular localities. Thus 
in various parts of the west, especially in the Sierra Nevada, 
we are told that rhyolite is usually succeeded by andesite.* 
The general succession is from a rock of average composi¬ 
tion through less silicious and more silicious ones to rocks 
extremely low in silica and others extremely high in silica— 
that is, the series commences with a mean and ends with 
extremes. 
This law of succession, expressed in its most general terms, 
is of very wfide application and holds true for all the local¬ 
ities studied by the present writer. 
It has been shown by Juddt to be the order of eruption 
of the lavas of the Lipari islands, which began with rocks of 
intermediate composition, and has reached the stage where 
rhyolite and basalt are being thrown out. 
It is to be remarked that constant exceptions to the gen¬ 
eral law of succession are observed. Many and possibly all 
of them arise from the fact that only a part of the whole 
series of events may take place in one limited locality or 
center of eruption, as at Electric peak and at the volcano of 
Crandall basin. The whole series of eruptions in this region 
embrace all of the igneous rocks of the region belonging to 
the period of volcanic activity, and the volcanoes mentioned 
* Turner (H. W.) Moliawk Lake Beds. Bull. Phil. Soc. Washington. 
8°. Washington, 1892, vol. 11, p. 389. 
f Judd (J. W.) Contributions to the Study of Volcanoes. Geol. Maga¬ 
zine. 8°. London, 1875, new series, vol. 2, p. 60. 
22-Bull. Phil. Soc., Wash., Vol. 12. 
