Vital statistics of heavenly bodies

In carrying out a laboratory experiment on a sample of material or an arrangement of apparatus, a scientist has some measure of control. The angle of view, the energy input, the temperature, the magnetic field - it might be possible to choose these parameters so that the combination is instructive, or it might even be possible to vary the parameters over a critical range. If the expected effect does not show up, then try something different: "Change the polarity!" as Dr Who would cry, in an early version of the television series.

For the sciences of natural phenomena there are more limited choices - particularly in medicine and astronomy. Ethical considerations restrict the experiments that can be carried out on people. Practical considerations rule out any experiments in astronomy. The information that comes across space from a natural astronomical "experiment" is all that is available. There is no different angle from which to view the experiment, and you cannot turn up the power so as to get a brighter source and more data. What you see has all to be interpreted by imagination and conjecture, guided by theory, and tested by further observations, perhaps made through new and larger telescopes.

Astronomy is thus a science of the possible. Sometimes the "possible" is mind-numbingly banal, although the interpretation is stunning.

For example, galaxies are not distributed randomly on the sky. The techniques to study this phenomenon consist of looking at statistical distributions of how close in the sky to one another the galaxies appear.

What this means in practice is looking at lots of photographs and pointing to the images of galaxies. This is a "ho! hum!" task, suitable only for a PhD student or machine.

To make sure what star I missed I should have to check on my list Every star in sight.

It might take me all night.

Robert Frost (1947) The payoff, however, is that this statistical exercise means something about the origin and growth of structure in the early Universe, from the "ripples" in the cosmic microwave background, discovered by the satellite Cobe, to the present-day clusters and "filaments" of galaxies surrounded by large "voids".

From the observational material, and restricted by these unpromising constraints, astronomers have built up the exciting modern picture of the universe, a dazzling array of stars, star systems, galaxies and cosmic structures. ("Stars teach, as well as shine" - Edward Young, 1744).

Robert Smith's comprehensive book for undergraduates is based on a successful series of lectures at the University of Sussex. It is divided in two: first Smith goes clearly and methodically through the observational techniques, then he tells us what the application of the techniques has found. A young physicist reading through it will end up with a good idea of what modern astrophysics is about.

Paul Murdin is head of astronomy, Particle Physics and Astronomy Research Council.