the technical side, because of the patience and manipulative skill required, and difficult on the intellectual side because the comprehension of many measuring instruments is dependent upon advanced theory and considerable analytical power, it has not been forgotten that the laboratory may serve certain other though perhaps minor purposes. Just, for instance, as aid to distinct thinking maybe rendered by the use of mathematical symbols and models, so also it will probably assist the unimaginative student to comprehend the laws under discussion if he can examine under his own manipulation the behavior of the apparatus which embodies those laws. Again there are certain phenomena which should be given the student for personal examination in the laboratory. Thermal phenomena involving the reading of thermometers; the passage of a liquid through the critical state; the study of compound tones; the observation of spectra, diffraction and polarization of light waves are examples of the kind of phenomena which require laboratory instruction. The number, however, of such exercises which appear even in the manuals of a college course is insignificant.
The pedagogists who, either with or without any definite knowledge of exact science, are perfectly sure how it ought to be taught assert that the first step in all good teaching is an appeal to the observing powers. "It is a cardinal principle in modern pedagogy that real and adequate knowledge of things can be obtained only in the presence of the things themselves," says one. Assuming that this is as true as the author thought it to be, it is but a half truth, the other half being that the presence merely of the things can not impart any really adequate knowledge. A boy, for instance, might watch the motion of the planets till he was gray without ever learning the first thing about gravitation or the solar system. Facts are but the raw materials of knowledge upon which the reasoning faculties must be exerted in order to extract the hidden principles of nature.
A writer of a well-known series of text-books has adopted as a sort of motto for his pupils, 'Read nature in the language of experiment.' One can not crtiticize an oracular utterance of this sort for the reason that it is not possible to say just what its author had in mind. If it is meant that empirical knowledge derived from the observation of detached facts and not brought into accordance with other facts by means of a hypothesis concerning their relation is sufficient for one to divine the laws of nature, we must certainly dissent. The language of experiment is in general a most difficult one to read, since, as we have been insisting, the measurements are in the great majority of cases indirect and to be interpreted only by tracing through a train of complex relations the consequences of the things observed upon some hypothesis whose truth it is desired to test.