Now small, portable nuclear power plants could be built, with a minimum of shielding. Such enormous investments had already been made in conventional fusion that the world's electrical utilities were not – at first – affected, but the impact on space travel was immediate; it could be paralleled only by the jet revolution in air transport of a hundred years earlier.

No longer energy-limited, spacecraft could achieve far greater speeds; flight times in the Solar System could now be measured in weeks rather than months or even years. But the muon drive was still a reaction device – a sophisticated rocket, no different in principle from its chemically fuelled ancestors; it needed a working fluid to give it thrust. And the cheapest, cleanest, and most convenient of all working fluids was – plain water.

The Pacific Spaceport was not likely to run short of this useful substance. Matters were different at the next port of call – the Moon. Not a trace of water had been discovered by the Surveyor, Apollo, and Luna missions. If the Moon had ever possessed any native water, aeons of meteoric bombardment had boiled and blasted it into space.

Or so the selenologists believed; yet clues to the contrary had been visible, ever since Galileo had turned his first telescope upon the Moon. Some lunar mountains, for a few hours after dawn, glitter as brilliantly as if they are capped with snow. The most famous case is the rim of the magnificent crater Aristarchus, which William Herschel, the father of modem astronomy, once observed shining so brightly in the lunar night that he decided it must be an active volcano. He was wrong; what he saw was the Earthlight reflected from a thin and transient layer of frost, condensed during the three hundred hours of freezing darkness.

The discovery of the great ice deposits beneath Schroter's Valley, the sinuous canyon winding away from Anstarchus, was the last factor in the equation that would transform the economics of space-flight. The Moon could provide a filling station just where it was needed, high up on the outermost slopes of the Earth's gravitational field, at the beginning of the long haul to the planets.

Cosmos, first of the Tsung fleet, had been designed to carry freight and passengers on the Earth-Moon-Mars run, and as a test-vehicle, through complex deals with a dozen organizations and governments, of the still experimental muon drive. Built at the Imbriurn shipyards, she had just sufficient thrust to lift off from the Moon with zero payload; operating from orbit to orbit, she would never again touch the surface of any world. With his usual flair for publicity, Sir Lawrence arranged for her maiden flight to commence on the hundredth anniversary of Sputnik Day, 4 October 2057.

Two years later, Cosmos was joined by a sister ship. Galaxy was designed for the Earth-Jupiter run, and had enough thrust to operate directly to any of the Jovian moons, though at considerable sacrifice of payload. If necessary, she could even return to her lunar berth for refitting. She was by far the swiftest vehicle ever built by man: if she burned up her entire propellant mass in one orgasm of acceleration, she would attain a speed of a thousand kilometres a second – which would take her from Earth to Jupiter in a week, and to the nearest star in not much more than ten thousand years.

The third ship of the fleet – and Sir Lawrence's pride and joy – embodied all that had been learned in the building of her two sisters. But Universe was not intended primarily for freight. She was designed from the beginning as the first passenger liner to cruise the space lanes – right out to Saturn, the jewel of the Solar System.

Sir Lawrence had planned something even more spectacular for her maiden voyage, but construction delays caused by a dispute with the Lunar Chapter of the Reformed Teamsters' Union had upset his schedule. There would just be time for the initial flight tests and Lloyd's certification in the closing months of 2060, before Universe left Earth orbit for her rendezvous. It would be a very close thing: Halley's Comet would not wait, even for Sir Lawrence Tsung.

The survey satellite Europa VI had been in orbit for almost fifteen years, and had far exceeded its design life; whether it should be replaced was a subject of considerable debate in the small Ganymede scientific establishment.

It carried the usual collection of data-gathering instruments, as well as a now virtually useless imaging system. Though still in perfect working order, all that this normally showed of Europa was an unbroken cloudscape. The overworked science team on Ganymede scanned the recordings in 'Quick Look' mode once a week, then squirted the raw data back to Earth. On the whole, they would be rather relieved when Europa VI expired and its torrent of uninteresting gigabytes finally dried up.

Now, for the first time in years, it had produced something exciting.

'Orbit 71934,' said the Deputy Chief Astronomer, who had called van der Berg as soon as the latest data-dump had been evaluated. 'Coming in from the nightside – heading straight for Mount Zeus. You won't see anything for another ten seconds, though.'

The screen was completely black, yet van der Berg could imagine the frozen landscape rolling past beneath its blanket of clouds a thousand kilometres below. In a few hours the distant Sun would be shining there, for Europa revolved on its axis once in every seven Earth-days. 'Nightside' should really be called 'Twilight-side', for half the time it had ample light – but no heat. Yet the inaccurate name had stuck, because it had emotional validity: Europa knew Sunrise, but never Lucifer-rise.

And the Sunrise was coming now, speeded up a thousandfold by the racing probe. A faintly luminous band bisected the screen, as the horizon emerged from darkness.

The explosion of light was so sudden that van der Berg could almost imagine he was looking into the glare of an atomic bomb. In a fraction of a second, it ran through all the colours of the rainbow, then became pure white as the Sun leapt above the mountain – then vanished as the automatic filters cut into the circuit.

'That's all; pity there was no operator on duty at the time – he could have panned the camera down and had a good view of the mountain as we went over. But I knew you'd like to see it – even though it disproves your theory.'

'How?' said van der Berg, more puzzled than annoyed.

'When you go through it in slow motion, you'll see what I mean. Those beautiful rainbow effects – they're not atmospheric – they're caused by the mountain itself. Only ice could do that. Or glass – which doesn't seem very likely.'

'Not impossible – volcanoes can produce natural glass – but it's usually black... of course!'

'Yes?'

'Er – I won't commit myself until I've been through the data. But my guess would be rock crystal – transparent quartz. You can make beautiful prisms and lenses out of it. Any chance of some more observations?'

'I'm afraid not – that was pure luck – Sun, mountain, camera all lined up at the right time. It won't happen again in a thousand years.'

'Thanks, anyway – can you send me over a copy? No hurry – I'm just leaving on a field trip to Perrine, and won't be able to look at it until I get back.'

Van der Berg gave a short, rather apologetic laugh.

'You know, if that really is rock crystal, it would be worth a fortune. Might even help solve our balance of payments problem...'

But that, of course, was utter fantasy. Whatever wonders – or treasures – Europa might conceal, the human race had been forbidden access to them, by that last message from Discovery. Fifty years later, there was no sign that the interdiction would ever be lifted.