3.2 Some general properties of galaxies
Firstly, we note that galaxies tend to occur in clusters rather than singly. The mutual gravitational attraction of galaxies naturally tends to hold them on paths that remain close to each other. Typically a cluster contains tens or hundreds of galaxies. There are, however, large clusters with thousands of galaxies, and there are some solitary galaxies. Our own Galaxy is a member of a smallish cluster of about 36 galaxies called the Local Group (see Figure 7). A typical cluster of moderate size is shown in Figure 8.
How are the galaxy clusters distributed? Are they close enough to affect each other? If they are, then the motion of a given cluster would depend mostly on the distortion of spacetime caused by its nearest neighbours. Because the distances between clusters vary, this would correspond to a large spread in relative speeds. At the other extreme, if the clusters of galaxies were very far apart, then the attraction of neighbouring clusters would be negligible, and the motion would be dominated by the overall spacetime curvature due to the matter of the whole Universe – a much simpler situation.
In fact, galaxy clusters are loosely associated in superclusters. However, they are far enough apart for one to regard clusters as essentially independent (see Figure 9). The fact that clusters are, in effect, independent of one another is of central importance to cosmology. It means that a cluster of galaxies can be taken as the basic ‘particle’ of cosmological dynamics, and the motion of individual galaxies within a cluster can be ignored on the grand scale of cosmology. Thus, we conclude:
Galaxy clusters are the basic test particles of cosmology, their motion following geodesic paths through spacetime.
In other words, a galaxy cluster plays a similar role on the grand scale to that of a planet (e.g. Mercury) mapping out the local region of spacetime in the Solar System, or Newton's falling apple doing the same thing closer to the Earth's surface.
The next important property of galaxies is that there is a statistical predictability about their masses and luminosities. Figure 10 shows the proportion of galaxies having a given luminosity. It varies over quite a wide range, the distribution falling off steeply on the high-luminosity side. (Note that the luminosity is expressed in terms of (absolute) ‘magnitudes’, this being a parameter such that the smaller or more negative the value of the magnitude, the more luminous the object. This rather odd choice arises for historical reasons.) There are large numbers of the least luminous galaxies (shown by the rising curve on the left-hand side), but in practice these tend to be invisible in the more distant clusters which are of greater interest to cosmologists. No doubt it would have been easier for cosmology if galaxies had been more similar. However, because galaxies occur in clusters, one can at least use statistical methods. For instance, the distribution of luminosities is more or less the same for all clusters, resulting in the average luminosity of galaxies in a cluster being fairly standard even though the individual galaxies vary greatly. This statement assumes, of course, that we are dealing with galaxy clusters of the same age. As we have already pointed out, this might not be the case when observing galaxies at a great distance, and hence as they were some time in the distant past.