488 R. H. RASTALL 
we have abundant evidence in the steam that accompanies volcanic 
eruptions, and the geysers and hot springs belonging to the later 
stages of igneous activity in many parts of the world, such as the 
Yellowstone Park, Iceland, and New Zealand. At Steamboat 
Springs, Nevada, ores are now visibly being formed by hot springs 
which must derive their heat from intratelluric sources. It is 
natural to suppose, therefore, that many of the ancient ore deposits 
have been formed in a similar way. The common association of 
ores with propylitization and certain other types of rock-alteration 
points in the same direction. 
Moreover, ore minerals occur in rocks undoubtedly igneous as 
well as in lodes and veins. Cassiterite is found as an original 
mineral in granites, in pegmatites, and in quartz veins in undoubted 
and visible connection with granites. The same applies to its 
constant companions wolframite and molybdenite. Gold is known 
in almost every conceivable geological situation of almost every 
age. Chromium and platinum are found beyond doubt as original 
constituents of ultrabasic rocks such as peridotites and serpentines. 
Magnetite and ilmenite have segregated in enormous masses from 
plutonic intrusions, and so on indefinitely. A great part of the 
rich mineralization of Mexico and South America is in obvious and 
visible connection with the great volcanic outbursts ranging from 
late Cretaceous to modern times. It is clear that a vast number of 
the most valuable ore deposits are of direct igneous origin. In 
many other instances, though the connection is not so clear, it is 
still highly probable or indeed certain. 
Nevertheless, although the main fact in its broadest outlines 
is established, considerable doubt still remains as to the mechanism 
of the processes by which the concentration and deposition of the 
ores has been effected, and the underlying and fundamental reasons 
for these processes. Another highly important aspect of the subject 
is the distribution in space and time of the different types of 
mineralization and their relation to crust disturbances and various 
petrographic types. This in the broadest view is the scope of the 
study of metallogenesis. . 
Hitherto it has not been found possible to devise a really 
satisfactory and logical classification of ore deposits on a genetic 
