This is, in a very compressed form, the meaning of the special theory of relativity.

As there is still some space left on our quarto page, we may add a remark that, it is hoped, will make a little clearer the above-mentioned discrepancy.

Let us choose as our system of reference an express train 18 miles long. There are two passengers—Mr. Front, right at the front of the train, and Mr. Back, at the extreme end of the train, so that a rigid distance of 18 miles separates the two passengers. The carriages are transparent, so that the two passengers can signal to one another. They are, moreover, furnished with ideal clocks that run at exactly the same rate.

First, suppose that the train is at rest. Back is just opposite milestone 100, whilst Front is opposite milestone 118. By means of a flash, Back signals to Front his time, exactly 12 o'clock. It takes light very nearly ¹⁄₁₀₀₀₀ second to traverse the length of the train—18 miles; hence the flash will reach Front at 12 o'clock ¹⁄₁₀₀₀₀ second. Exactly the same result would have come about if Front had signalled his time to Back. Light makes no difference in travelling forwards and backwards. If the train moves at a great speed, the two travellers can conduct the same experiment as when the train was at rest. They will then set the time that light takes to travel from Back to Front equal to the time that it takes to traverse the same way in the reverse direction. But this phenomenon will assume a different aspect if viewed from the railway embankment. An observer on the latter would affirm that light does not take the same time in travelling the length of the train in one direction as it does when travelling in the opposite direction.

For the ray of light moving in the forward direction has to traverse not only the distance between Back and Front, but also the very short distance that Front has moved forward during the interval that the light has been moving; whereas, inversely, the flash sent out by Front to Back will traverse a distance that is correspondingly less than that between the passengers, since Back is moving towards the signal.

Thus the duration of the two phenomena of light propagation is the same or different, respectively, according as it is judged from the train or from the embankment. In other words, the judgment of the length of time depends on the state of motion of the observer.

All further pronouncements of the special theory of relativity are based on the preceding arguments of the relativity of time.

* * * * * * * *

Would Man be able to construct a Science if he possessed one sense less than at present—for example, if he were deprived of sight? Let us apply this to a definite case. In the new physics the velocity of light plays a decisive part as a world-constant. At first sight it would appear impossible for us to determine it and recognize its importance, if we had not at our disposal some organ which enabled us to become aware of optical phenomena.

But, as Einstein explained to me, even under such difficult circumstances, it would be possible to build up a science, for the reason that phenomena, as far as they are perceptible, may be transformed so that they become manifest to other senses if one sense should be absent. For example, the electrical conductivity of selenium is strongly influenced by the amount of illumination that falls on it. Thus light acts on a selenium cell, causing changes of current intensity, which in their turn may be perceived by feeling, or by chemical action on the mucous fluid of the tongue. Ultimately we are concerned only with a differentiation that enables us to refer identical experiences to identical events. We should certainly encounter enormous difficulties in endeavouring to form a physical picture of our surrounding world if the number of our senses should become less than the organs with which we actually operate. Yet, in principle, we should be able to overcome all difficulties by means of much lengthened and complicated lines of research, even if we should have only a single sense left, or if we had only one at the very outset. The construction of a Science would then be possible, and would give the same results, although it might be propounded only after a delay of perhaps millions of years.