When two hypotheses are possible, we provisionally choose that which our minds adjudge to be simpler, on the supposition that this is the more likely to lead in the direction of truth. It includes as a special case the principle of Occam's razor-entia non multiplicana praeter necessitatem.

Bohr's investigation had typified what had become a standard procedure in problems of theoretical physics. The first step was to discover the mathematical laws governing certain groups of phenomena; the second was to devise hypothetical models or pictures to interpret these laws in terms of motion or mechanism; the third was to examine in what way these models would behave in other respects, and this would lead to prediction of other phenomena-predictions which might or might not be confirmed when put to the test of experiment. For instance, Newton had explained the phenomena of gravitation in terms of a force of gravitation; a later age had seen the luminiferous ether introduced to explain the propagation of light and, subsequently, the general phenomena of electricity and magnetism; finally Bohr had introduced electronic jumps in an attempt to explain atomic spectra. In each case the models had fulfilled their primary purpose, but had failed to predict further phenomena with accuracy.

Heisenberg finds that facts of observation lead uniquely and inevitably to the theoretical structure known as matrix mechanics. This shows that the total radiation in any region of empty space can change only by a single complete quantum at a time. Thus not only in the photo-electric phenomenon, but in all other transfers of energy through space, energy is always transferred by complete quanta; fractions of a quantum can never occur. This brings atomicity into our picture of radiation just as definitely as the discovery of the electron and its standard charge brought atomicity into our picture of matter and of electricity.

Another conspicuous failure of classical mechanics was with one aspect of the problem of radiation. ...Imagine a crowd of steel balls rolling about on a steel floor. ...There must... be a steady leakage of energy from... causes, such as air resistance and the friction of the floor, so the balls will eventually lose energy, and, after no great length of time, will be found standing at rest on the floor. The energy of their motion seems to have been lost... most of it has been transformed into heat. The classical mechanics predicts that this must happen; it shows that all energy of motion, except possibly a minute fraction of the whole, must be transformed into heat whenever such a transformation is physically possible. It is because of this that perpetual-motion machines are a practical impossibility.

The classical mechanics had envisaged the world constructed of matter and radiation, the matter consisting of atoms and the radiation of waves. Planck's theory called for an atomicity of radiation similar to that which was so well established for matter. It supposed that radiation was not discharged from matter in a steady stream like water from a hose, but rather like lead from a machine-gun; it came off in separate chunks which Planck called quanta. This... carried tremendous philosophical consequences.

This spectrum is of the type known in spectroscopy as a line-spectrum. Its appearance is that of a group of bright lines on a dark background, indicating that the radiation divides itself between a number of clearly defined frequencies, and that there is no radiation in between. Before Bohr's explanation appeared, these frequencies had been supposed to belong to some sort of vibration taking place in the hydrogen atom - like frequencies of the musical note which is heard when a bell or piano wire is made to vibrate. It now became clear that they had an entirely different origin. The energy exhibited in the spectrum was not liberated by a vibration, or any kind of continuous motion, but by the sudden jump of an electron to an orbit of lower energy, and its frequency was determined by the compulsion put upon it to form a single quantum.

Today there is a wide measure of agreement, which on the physical side of science approaches almost to unanimity, that the stream of knowledge is heading towards a non-mechanical reality; the universe begins to look more like a great thought than like a great machine. Mind no longer appears as an accidental intruder into the realm of matter; we are beginning to suspect that we ought rather to hail it as a creator and governor of the realm of matter...

This fallacious result is not... a peculiarity of classical mechanics; it is given also by a very wide class of possible systems of mechanics. This being so, no minor modification of the classical mechanics can possibly put things right. Something far more drastic is needed; we are called upon to surrender either the continuity or the causality of classical mechanics, or else the possibility of representing changes by motions in time and space.

It can hardly be a matter for surprise that our race has not succeeded in solving any large part of its most difficult problems in the first millionth part of its existence. Perhaps life would be a duller affair if it had. For to many it is not knowledge but the quest for knowledge that gives the greater interest to thought - to travel hopefully is better than to arrive.