40 PHILOSOPHICAL SOCIETY OF WASHINGTON. 
prolongation to the northward and southwestward ; the narrow inter- 
vals where they run along the western flanks of, the Appalachians 
and across the Florida peninsula; and the wide intervals in the 
Ohio and upper Mississippi valleys. The following considerations 
were offered as helping to explain these apparently anomalous fea- 
tures: The crystalline Archean rocks and parallel ridges of the 
mountain chain favor a rapid axial transmission of the vibrations, 
The first tremors spreading to the northwestward, however, are cut. 
off and deadened by the mountains, so that it isa later phase of the 
wave which is felt and recorded beyond; having passed the range 
it then spreads with little obstruction and high velocity through 
comparatively level strata. Similarly in the littoral and alluvial 
deposits to the northeast, south, and southwest the earlier tremors 
are lost and later phases of the wave are successively recorded. 
It is especially to be remembered that all these times are from non- 
instrumental observations ; an exact instrumental observation made 
at Toronto, Ontario, by Prof. Chas. Carpmael (9: 54: 50 P. m.), 
could not be used here because so early as to be wholly inconsistent 
with all other reliable but non-instrumental observations. 
The epicentrum, or point on the surface directly above that part 
of the deep-lying fissure where the earliest vibrations originated, is 
placed by these coseismals about a hundred miles north of Charles- 
ton, which is not at all inconsistent with the fact that the greatest 
damage was caused in that city. In fact, it is to be expected that 
the destruction of buildings should be greater at a distance, where 
the angle of emergence is less. Moreover,'most of the evidence 
seems to point to a very deep-lying origin, in which case one can- 
not but attribute much of the local damage, as well as the continu- 
ance of shocks of considerable intensity but small area, to the 
character of the recent geologic formations in that region. Borings 
for artesian wells at Charleston indicate that the Tertiary and 
Cretaceous strata are very heterogeneous in character—sands, 
clays, limestone-marls, and imperfectly consolidated beds of con- 
glomerate, with occasional cavities containing running water—and 
much of the city is built upon made land. Such considerations 
may explain the extremely local character of many of the shocks 
which have been felt at various points in the State of South 
Carolina since August 31st, as well as the great intensity of the 
shock at Charleston on that date. Local sinks in the ground are re- 
ported even in northern Florida, far from the origin of the disturb- 
