nsible. Yet all the manifestations that he
observes are found to repeat themselves in certain unvarying sequences.
Certain applications of energy will produce certain movements of matter.
We may not know the nature of the so-called cause, but we learn to
measure the result, and in other allied cases we learn to reason back
or infer the cause from observation of results. The latter indeed is
the essence of scientific inquiry. When certain series of phenomena have
been classified together as obviously occurring under the domination
of the same or similar causes, we speak of having determined a law of
nature. For example, the fact that any body in motion tends to go on at
the same rate of speed in a direct line forever, expresses such a law.
The fact that the gravitation pull is directly as the mass and inversely
as the square of the distance of the bodies it involves, expresses
another such law. The fact that the planetary bodies of the solar system
revolve in elliptical orbits under the joint influence of the two laws
just named, expresses yet another law. In a word, then, these so-called
"laws" are nothing more than convenient formulae to express the
classification of observed facts.

INDUCTIVE VERSUS DEDUCTIVE REASONING

The ancient thinkers indulged constantly in what we now speak of
as deductive reasoning. They gave heed to what we term metaphysical
preconceptions as to laws governing natural phenomena. The Greeks, for
example, conceived that the circle is the perfect body, and that the
universe is perfect; therefore, sun and moon must be perfect spheres
or disks, and all the orbits of the heavenly bodies must be exactly
circular. We have seen that this metaphysical conception, dominating the
world for many centuries, exerted a constantly hampering influence upon
the progress of science. There were numerous other instances of the same
retarding influence of deductive reasoning. Modern science tries to cast
aside all such preconceptions. It does not always quite succeed, but
it makes a strenuous effort to draw conclusions logically from observed
phenomena instead of trying to force observations into harmony with
a preconeived idea. Herein lies the essential difference between the
primitive method and the perfected modern method. Neither the one nor
the other is intended to transcend the bounds of the natural. That is
to say, both are concerned with the sequence of actual events, with the
observation of actual pheno