12 THE PRESIDENTIAL ADDRESS. 
remarked that the theory of elastic waves, which is here involved, was 
initiated and developed in quite a different connection, in the persistent but 
vain attempts to construct a mechanical representation of the luminiferous 
ether which exercised the mathematical physicists of a generation or two 
ago. It has here at length found its natural application. One of the first 
problems of seismologists has been to construct, from observation, tables 
which should give the time an elastic wave of either of the two cardinal 
types—viz. of longitudinal and transverse vibration—takes to travel from 
any one point of the earth’s surface to any other. It has been shown by 
Herglotz and Bateman that if these data were accurately known it should 
be possible, though naturally by a very indirect process, to deduce the 
velocities of propagation of the two types throughout the interior. Such 
tables have been propounded, and are in current use for the purpose of 
fixing the locality of a distant earthquake when this is not otherwise 
known. They are however admittedly imperfect, owing to the difficulty 
of allowing for the depth of the focus, which is not always near the surface, 
and is sometimes deep-seated. This uncertainty affects, of course, the 
observational material on which the tables are based. Some partial 
corrections have been made by Professor Turner, who almost alone in this 
country, amidst many distractions, keeps the study of seismology alive, 
but the construction of accurate tables remains the most urgent problem 
in the subject. Taking however the material, such as it is, the late Pro- 
fessor Knott, a few years ago, undertook the laborious task of carrying out 
the inverse process of deducing the internal velocities of the two types of 
waves referred to. Although it is possible that his conclusions may have to 
be revised in the light of improved data, and, it may be, improved methods 
of calculation, they appear to afford a fairly accurate estimate of the wave 
velocities from the surface down to a depth of more than half the earth’s 
radius. Near the surface the two types have velocities of about 7°2 and 
4 km. per second, respectively. These velocities increase almost uniformly 
as we descend, until a depth of one-third the radius is reached, after which, 
so far as they can be traced, they have constant values of 12-7 and 6:8 km. 
per second, which, by the way, considerably exceed the corresponding veloci- 
ties in iron under ordinary conditions. The innermost core of the earth, 
i.e. a region extending from the centre to about one-fourth of the radius, 
remains somewhat mysterious. It can certainly propagate condensational 
waves, but the secondary waves are hard to identify beyond a distance of 
120° of are from the source of disturbance. Knott himself inferred that 
