80 RICHTHOFEN — NATURAL SYSTEM 
If we turn our attention from the mode of the geographical to that of the geo- 
logical occurrence of voleanic rocks, we may first consider from this point of view the 
relations just mentioned. The two modes of distribution which are most conspicuous, 
namely, on the foot of mountain ranges and along sea-coasts, are nearly identical in a 
geological aspect. For most, if not all of those ranges the flanks or vicinity of which 
are distinguished by the occurrence of volcanic rocks on a large scale, have been contig- 
uous to sea-coasts in the Tertiary period, or they are so at present, or they have been 
so in the intermediate time. This is true of active and extinct volcanoes, as well as for 
massive eruptions, in the Carpathians, on the southern foot of the Alps, on the borders 
of the plains of northern Germany, on the slopes of the central plateau of Asia, in their 
whole extent from Armenia and the Caucasus, passing Lake Issikul and Lake Baikal, 
to the vicinity of Pekin, in the coast ranges of California, in the Cascade Mountains of 
Oregon, and in numerous other countries. As regards the prominent occurrence of 
voleanic rocks on high table-lands, it is analogous to their situation on sea-coasts, imas- 
much as the salt lakes occurring on them, which were of much larger size in the Ter- 
tiary period, would be an equivalent of the vicinity of the sea. Those larger regions 
not covered by salt lakes, which are situated in the interior of continents, and have 
been so since the commencement of the Tertiary period, were generally not the theater 
of outbreaks of voleanic rocks. This circumstance renders almost certain the influence 
which the neighborhood of large quantities of salt water has had upon the commence- 
ment and main phases of the eruptive activity, though its continuation in the latter 
phases, which are those of volcanic action proper, may in numerous instances have 
been maintained by fresh water. 
The relations of the distribution of the volcanic to that of the granitic and por- 
phyritic rocks have been considered in another chapter. There is probably no place dis- 
tinguished by the accumulation of the first where the previous eruption either of granitic 
or of both granitic and porphyritic rocks may not be observed, if the conditions are such 
that observations in this direction are possible. But as there are porphyritic regions out- 
side of the volcanic belts (for instance, the quartzose porphyries of southern Norway, or 
the Triassic trap-rocks of the Connecticut Valley), and as the same is true of a great 
number of known granitic districts, and probably of a far greater number of others 
which are not accessible to observation, it is evident that the existence of ancient chan- 
nels of ejection did not necessarily imply their reopening in the Tertiary period, while 
there can scarcely be any doubt that the once shattered places where they were situ- 
ated offered less resistance to the formation of new fractures, than other portions of 
those regions in which the greatest disturbances took place in the Tertiary period. 
We have, finally, to mention the connection which apparently exists between the 
occurrence of volcanic rocks and the regions where ancient formations have been 
were laid down ona map, it would hardly show such great intermissions. The extensive fields of lava found by Capt. 
Fitzroy, in southern Patagonia, the voleanic rocks of the Desert of Atacama, and of northern Peru, are instances of their 
extensive distribution in the non-voleanic spaces of South America, while on the northern continent they are one of the 
main features in the structure of the great western mountain ranges, probably throughout their whole extent from Panama 
to the peninsula of Alaska, and are accompanied probably by thousands of extinct craters. 
(118) 
