68 REPORT—1858. 
coloured seismic bands of Japan, the Kuriles, and Kamtschatka ; and, pass- 
ing to the opposite shore of the Pacific, we are presented with the deeply- 
coloured seismic bands of Mexico and the South American Andes, whose 
influence reaches far out into the ocean, but eastward or landward is 
checked by the great chain. ‘The reason of this fact, whick has been before 
alluded to, is not hard to find. The general section of the South American 
continent, from west to east, consists of a comparatively low-lying narrow 
littoral border-country on the Pacific; then the immense chain of the 
Andes rising in successive ranges to the axial peaks, and beyond these a 
vast plateau—the elevated land of the great continent—reaching over to 
near the western coast, where some lower ranges of mountains terminate 
the Atlantic shore and bound its basin. This is rudely shown in the accom- 
panying figure (1). 
Fig. 1. 
Now if a shock be transmitted from any origin within the great chain, 
and below the level of the great tableland, a6, as from a point 2, the 
transmitted elastic wave in the direction as, reaching the surface after a 
very short transit, will, in accordance with the well-known law of elastic 
bodies, have its amplitude increased (just as the last billiard-ball of a line 
of touching balls, is that which is projected when the first of the line is struck 
by the blow of a propelled ball), and more powerfully shake all surface- 
objects at s than others situated at a, although at an equal radial distance 
from the centre of effort,—the free movement of the elastic wave being 
here reacted upon by the elastic mass of the tableland which blocks its way 
until compressed. Objects on the tableland, at an equal distance from 
the origin, may (dependent upon its depth) receive the shock (even if 
of only equal amplitude) at such an angle of emergence as will give a less 
power of overthrow to the horizontal component of the wave’s transit. 
There will in every case be a reflected wave back from the mass of the 
tableland—an earthquake echo—producing at s, or along the littoral border, 
a second shock, with a line of direction nearly the same, but with a direction 
of motion reverse to the first, one shock only being felt on the tableland. 
To return, the seismic band of the Andes, at the extreme north of the 
continent, and at Trinidad, inosculates with that of the West India Islands, 
which sweeps round the Caribbean Sea, and appears, so far as records go, 
to transmit its movements further into the Atlantic, than into the former 
sea; if so, that probably arises from causes quite analogous to those already 
explained for South America—a shallower sea-bottom to the westward, on 
the Caribbean Sea, thus playing the part towards the deeper bottom of the 
Atlantic that the tableland plays towards the littoral slope of South America. 
The North American records have been too few and ill-defined as to boundary 
to produce as yet any very distinct conclusions from the tints, which prove, 
however, that its western and southern seaboard are by no means free from 
earthquake. This has in great part arisen from the great want of orographic 
delineation on nearly all (even the largest and best) maps of the United States, 
which omit all heights and natural features. The Californian system west of 
the Rocky Mountains, that of Upper Missouri, of the Mississippi, and that of 
the northern lakes and basin of the St. Lawrence, form the chief and separate 
regions in which earthquakes haye been so far observed most frequently. 
