This abstract requirement must be brought into relation with the known facts of the physical world. Now what are the known facts? Let us take it as conceded that light consists of waves traveling with the received velocity. We then know a great deal about what goes on in the parts of space-time where there is no matter; we know, that is to say, that there are periodic occurrences (light waves) obeying certain laws. These light waves start from atoms, and the modern theory of the structure of the atoms enables us to know a great deal about the circumstances under which they start, and the reasons which determine their wave lengths. We can find out not only how one light wave travels, but how its source moves relatively to ourselves. But when I say this I am assuming that we can recognise a source of light as the same at two slightly different times. This is, however, the very thing which had to be investigated. We saw, in the preceding chapter, how a group of connected events can be formed, all related to each other by a law, and all ranged about a center in space-time. Such a group of events will be the arrival, at various places, of the light waves emitted by a brief flash of light. We do not need to suppose that anything particular is happening at the center; certainly we do not need to suppose that we know _what_ is happening there. What we know is that, as a matter of geometry, the group of events in question are ranged about a center, like widening ripples on a pool when a fly has touched it. We can hypothetically invent an occurrence which is to have happened at the center, and set forth the laws by which the consequent disturbance is transmitted. This hypothetical occurrence will then appear to common sense as the “cause” of the disturbance. It will also count as one event in the biography of the particle of matter which is supposed to occupy the center of the disturbance. Now we find not only that one light wave travels outward from a center according to a certain law, but also that, in general, it is followed by other closely similar light waves. The sun, for example, does not change its appearance suddenly; even if a cloud passes across it during a high wind, the transition is gradual, though swift. In this way a group of occurrences connected with a center at one point of space-time is brought into relation with other very similar groups whose centers are at neighboring points of space-time. For each of these other groups common sense invents similar hypothetical occurrences to occupy their centers, and says that all these hypothetical occurrences are part of one history; that is to say, it invents a hypothetical “particle” to which the hypothetical occurrences are to have occurred. It is only by this double use of hypothesis, perfectly unnecessary in each case, that we arrive at anything that can be called “matter” in the old sense of the word.