Page:Popular Science Monthly Volume 15.djvu/111

This page has been validated.


IN his "Preliminary Discourse on the Study of Natural Philosophy," Sir John Herschel remarks upon the importance of examining those phenomena of nature which are not wholly explicable in terms of any well-established theory. Instances of such residual phenomena, as Sir John Herschel terms them, are given in the discourse.

Newton's theory of comets, viz., that these bodies obey the law of gravitation while revolving in oblique orbits round the sun, appeared to account for the facts which had been noticed concerning the comet of Halley; but the period calculated for Encke's comet, on this hypothesis, was found to be rather longer than the actual, observed period, and, moreover, the duration of the observed period showed a small but regular diminution. Hence, Newton's theory, taken alone, was not sufficient to account for the facts. But, inasmuch as Newton's law of gravitation rested upon a sure and well-established foundation, the fact observed concerning Encke's comet could not be regarded as disproving the law; hence these facts were to be explained by tracing them to the action of some agent either of known or of, as yet, unknown nature.

The regularly diminishing period of Encke's comet remained a residual phenomenon, not contradicting the law of gravitation, but awaiting full explanation.

A residual phenomenon is, then, a phenomenon which is not fully explained by any established theory; but at the same time it is not a phenomenon which is absolutely contradictory to any such theory, for, if this were the case, the theory in question must perforce be abandoned.

Advances are made in natural science by a judicious use of hypotheses. Facts are accurately observed, or are gained by exact experiment, and are compared with facts; inferences are drawn, and are compared with other inferences, until a good working hypothesis is attained. From this hypothesis deductions are made which must necessarily prove true if the hypothesis be correct; the truth or falsity of the alleged facts is tested by an appeal to Nature; and so wider hypotheses are gained, each in turn being tested and tried by an appeal to facts, until, finally, that generalization is reached which includes in its expression so many and so varied phenomena that to it is given the name of a "law of Nature."

But notwithstanding the sure and tried foundations upon which each law of nature rests, phenomena ever and anon become apparent which refuse to be completely explained by any of these laws. Upon more careful examination, it may be found that such phenomena have been erroneously observed, and they may be brought under the application