Saturn all ears in heavenly I-spy

Galileo Galilei started it all in May 1609. By January 1610, four telescopes later, this 45-year-old professor of mathematics at the University of Padua had ground his own plano-convex and plano-concave lenses to produce a telescope that magnified the celestial image 30 times. This pioneering instrument showed the moon to be mountainous, the sun to have spots, Venus to have phase changes like our moon (Galileo used this to hammer home the fact that the sun, and not the earth, was at the centre of the planetary system), Jupiter to have four orbiting moons, and Saturn to have "ears".

This latter observation underlines the perennial problem with telescopes. They are never perfect. Astronomers are always itching to observe the skies with ever-larger telescopes or wishing that there was less and less distorting atmosphere between their instruments and objects of interest.

Since the time of Galileo, telescopes have relentlessly got bigger and better. By 1659 Saturn's ears were found to be nothing more than a thin flat equatorial ring system. By 1781, William Herschel had realised that some of the nebulous objects visible through his telescope were actually self-gravitating globular clusters, each containing one million or so stars. By 1925, even bigger telescopes had shown that others of these nebulous objects were galaxies like our own. The Hubble space telescope, orbiting 600km above the earth's surface, is billed as being the single most successful scientific instrument ever built. With a design based on a United States spy satellite, it was to have a 3m main mirror but this was de-scoped to its present 2.4m objective.

Hubble was launched in April 1990. It was soon discovered that its main mirror was faulty and this had to be repaired during the first service mission. In its first ten years it has provided 120,000 images of 10,000 different objects. It is little wonder that Nasa has already extended its operational life to at least the year 2010.

In 1946 the eminent American astronomer Lyman Spitzer said that "the most important contribution of such a radically new and powerful instrument would not be to supplement our current ideas about the universe in which we live, but to uncover new, hitherto unimaginable problems". We can judge whether Hubble has achieved this goal by turning to Daniel Fischer and Hilmar Duerbeck's Hubble Revisited . This colourful, large-format book is a worthy sequel to their popular Hubble: A New Window to the Universe (1996).

Hubble Revisited abounds with stunning astronomical pictures. Jovian aurorae jostle with sequential images of the nebulous gas around the Crab pulsar. Hot spots in the expanding cloud from the 1987A supernova compete with the intricate detail of planetary nebulae and tornadoes in the Lagoon Nebula.

Evaporating gaseous globules in the Eagle Nebula are juxtaposed with the amazing third dimension of the deep field of galaxies just north of Delta Ursae Majoris. What is especially impressive is the sensible descriptions of each picture provided by Fischer and Duerbeck, and the way in which they relate the new findings revealed by each image to established knowledge of the specific astronomical area.

But, in returning to Spitzer's quest for objects revealing "new, hitherto unimaginable problems", one gets the impression that these were very few and far between. One started to worry that Hubble was mainly filling in details, crossing and dotting a few astronomical t's and i's.

David Hughes is reader in astronomy, University of Sheffield.